Renal Association October 2003 Meeting – Commonwealth Centre, London ABSTRACTS

Renal Association
October 2003 Meeting – Commonwealth Centre, London
Renal Association October 2003
Plenary Session
Papers O1-O10
Tuesday 7 October
Main Auditorium
Renal Association October 2003
Impact of dialysis modality on outcomes in end stage renal disease
A Trehan, J Winterbottom, B Lane, R Foley, M Venning, R Coward, A MacLeod and R
SIRS group, Manchester Royal Infirmary, Manchester, M13 9WL, United Kingdom
Different types of dialysis modalities are suitable for different groups of people with end
stage renal disease (ESRD), and have been shown to give comparable outcomes.
However, as the pressure on renal services increases, choice of dialysis modality is
limited by resources. We believe that this will adversely impact on patient outcomes.
SIRS is an inception, cohort study of people with ESRD who started renal replacement
therapy (RRT) between April 2000 and April 2003 in the Northwest of England. Of
1109 people entered (mean age 57.6 years, 61% male, 12.8% from ethnic minority
groups, 26% diabetics, 31.1% with vascular disease and 21% requiring dialysis within 1
month of referral to renal services) 16.1% started outpatient haemodialysis (HD) with a
fistula or graft, 24.6% HD with a line and 59.6% started peritoneal dialysis (PD). 292
(26.3%) patients died and 144 (13%) patients had 185 major septic events (47 deaths
and 138 hospital admissions). The tables below show the impact of mode of dialysis on
outcomes (with HD with a fistula as the reference category).
In the Northwest there is a high dependence on peritoneal dialysis to provide RRT,
which is largely resource driven. The inability to switch patients freely from different
modalities leads to poor outcomes in terms of mortality and morbidity.
Renal Association October 2003
YB-1 Controls Translation of TGF-_ in Proximal Tubular Cells: Implications for
the Pathogenesis of Diabetic Nephropathy
D J Fraser1, C Van Roeyen2, P Muhlenberg2, P R Mertens2 and A O Phillips1
Institute of Nephrology, University of Wales College of Medicine, Heath Park
Campus, Cardiff, CF14 4XN, United Kingdom and 2Department of Nephrology,
Universitaetsklinikum Aachen, Aachen, 52074, Germany
TGF-_ is an important mediator of progressive renal tubulointerstitial fibrosis in
diabetic nephropathy. We have shown previously that in proximal tubular cells, TGF-_
synthesis is controlled independently at the level of translation by stimuli of relevance
to the pathogenesis of diabetic nephropathy (glucose, insulin and PDGF). Other in vitro
work supports the importance of translational control in TGF-_ synthesis, and suggests
that this is mediated by interaction of an unidentified RNA binding protein with a key
area within the 5' untranslated region of TGF-_ mRNA (5'UTR). Our aims in this study
were to identify this protein, and to characterise its role in control of translation of TGF_ in proximal tubular cells. RNA Electrophoretic mobility shift analysis (RNA-EMSA)
and ultraviolet crosslinking experiments showed that proximal tubular cell extracts
contain two protein complexes that bind to the key region of the TGF-_ 5'UTR, with
molecular weights 50 and 100 kDa. Supershift experiments suggested that both
complexes are composed of YB-1 protein, and RNA-EMSA using recombinant YB-1
protein confirmed this. By comparison with known YB-1 binding sites a 20-nucleotide
putative YB-1 binding site was identified within the TGF-_ 5'UTR, competitor
experiments confirmed that this sequence has a high affinity for YB-1. To determine the
role of YB-1 in control of TGF-_ translation we used small interfering RNA to knock
down YB-1 expression in proximal tubular cells. This inhibited their TGF-_ generation,
suggesting that YB-1 binding to TGF-_ mRNA directs TGF-_ synthesis in proximal
tubular cells. In conclusion, we have identified the protein that binds to the key
regulatory region of the TGF-_ 5'UTR as YB-1. Our results also suggest that reduction
in YB-1 expression is a potential therapeutic mechanism to reduce TGF-_ synthesis in
the kidney.
Renal Association October 2003
The Epidemiology of Acute Renal Failure in Scotland: A prospective, population
based study.
Department of Medicine and Therapeutics, Polwarth Building, Foresterhill, Aberdeen,
AB25 2ZD, United Kingdom, 2department Of Public Health Medicine, Polwarth
Building, Foresterhill, Aberdeen, AB25 2ZD, United Kingdom And 3Scottish Renal
Registry, Walton Building, Glasgow Royal Infirmary, Glasgow, G44, United Kingdom
There are few population based data available on the incidence of Acute Renal Failure
(ARF) requiring treatment with Renal Replacemnt therapy (RRT). We recruited all adult
patients (age≥15 years) treated with RRT for ARF and Acute on chronic renal failure
(ACRF) in Scotland (population 5 064 200) over a 36 week period and recorded patient
and renal outcome at 90 days.
Over the data collection period, 842 patients (61% males, median age 62.1 years) started
RRT for either ARF or ACRF.
Table 1 shows the number of patients registered in each region of Scotland and the
incidence of ARF and ACRF requiring RRT in these regions.
The overall incidence was high at 240 pmp/year (292 pmp/year for ages ≥15 years).
Table 2 shows renal and patient survival for patients with ARF and ACRF.
Age group at start of RRT (P<0.001), requirment for first RRT within an ICU (p<0.001),
region of Scotland where first RRT took place (p=0.009) and Khan comorbidity index at
start of RRT (p=0.007) were all associated with a higher risk of death within the first 90
days of treatment in univariate analysis. Multivariate analysis is underway.
The incidence of this condition in Scotland is in excess of 200 pmp/year, higher than
that previously shown and survival is low. This is the first, prospective, population
based study of ARF and ACRF requiring treatment with RRT and provides the basis for
future national service planning and audit.
Renal Association October 2003
Urinary Levels of Monocyte Chemoattractant Protein-1 (MCP-1) as a Marker of
Disease Activity in ANCA Associated Vasculitis
TM Hammad, FWK Tam, CD Pusey and JB Levy
Renal Section, Faculty of Medicine, Imperial College, Hammersmith Hospital, Ducane
Road, London, W12 ONN, United Kingdom
The vasculitides are relapsing and remitting diseases. Relapses can however be difficult
to distinguish from infective complications or drug toxicity, and urinary abnormalities
may not distinguish renal scarring from active inflammation. We hypothesised that
levels of urinary MCP-1 (uMCP-1), a potent T cell and macrophage chemoattractant,
might be a useful non-invasive marker for disease activity in primary vasculitis. We
have previously shown in a small cross sectional study that patients with active renal
disease have significantly higher uMCP-1 levels than those with inactive renal disease
(Tam et al).
In this study we have measured serial uMCP-1 levels by sandwich ELISA in 134
samples from 32 patients with ANCA associated vasculitis over 6 months. Disease
activity was assessed by Birmingham Vasculitis Activity Score (BVAS), ANCA titre and
inflammatory markers. 14 patients had acute disease and 18 either inactive or chronic
grumbling disease. Mean age of patients was 60 years and 10 were female.
uMCP-1 levels were significantly higher in active patients; median 157ng/mmol urinary
creatinine for patients with acute active disease, 56ng/mmol for those with chronic
grumbling disease and 12ng/mmol for those with inactive disease (p<0.0001). Nonvasculitic control patients or normal individuals had median uMCP-1 of 13 ng/mmol
creatinine. Serial uMCP-1 levels did not fluctuate significantly in patients with stable
inactive disease, even if they had persistent or fluctuating ANCA titres (anti-MPO or
anti-PR3). uMCP-1 declined rapidly in patients with active disease after initiation of
treatment, usually reaching baseline levels by one month. There was no correlation
between uMCP-1 and serum CRP or ANCA titre, but a strong association with BVAS
(r=0.52; p<0.0001). uMCP-1 increased rapidly and significantly in patients with chronic
disease developing a renal relapse. Patients with extra-renal disease did not have
significantly raised uMCP-1.
These results suggest that urinary MCP-1 is a very useful marker for renal inflammation
in ANCA associated vasculitis, and may identify active nephritis. Ongoing work will
extend this data to determine whether uMCP-1 can be useful in predicting renal relapses
and response to treatment.
Renal Association October 2003
Molecular dissection of putative targeting motifs in AE1
MAJ Devonald1, N Rungroj1, AW Cuthbert1, F Reimann1, PT Yenchitsomanus2, WM
Bennett3 and FE Karet1
Departments of Medical Genetics, Medicine and Clinical Biochemistry, Cambridge
Institute for Medical Research, Box 139 Addenbrooke's Hospital, Cambridge, CB2
2XY, United Kingdom, 2Institute of Molecular Biology and , Genetics, Mahidol
University, Salaya, TH, Thailand and 3Northwest Renal Clinic, Portland, OR, USA
Mutations in SLC4A1, encoding the anion exchanger AE1, cause distal renal tubular
acidosis (dRTA), a disease of defective urinary acidification. Reported missense
mutations, and one that truncates the last 11 residues of AE1 (RDEYDEVAMPV,
designated AE1_11) have been shown to retain anion transport function. AE1_11
exhibits non-polarised targeting in epithelial cells in vitro, the Y904 residue being
critical for normal targeting, while the destination of missense mutants is not clear.
To dissect further the role of the C-terminal tail of AE1, we investigated the surrounding
residues. V907, the component of a putative YXXØ targeting motif was replaced by A
(VA) or S (VS), to determine whether its hydrophobicity is important. The region
contains 2 diacidic patches (DE) which were each replaced by AA (named AA1 for the
pair preceding Y904, and AA2 for the pair following) since acid patches are implicated
in some protein-protein interactions. The final construct deleted the C-terminal 4
residues (AE1_4); this region has been proposed as a PDZ-binding domain.
By confocal microscopy, localization of these mutants was compared with wild type
(WT) AE1 and AE1_11 in MDCK and IMCD cells grown to polarity. Steady-state
distributions of VA and VS were basolateral, similar to that of WT AE1. In contrast,
AA1, AA2 and AE1_4 appeared in a non-polarized distribution including apical and
considerable intracellular components.
We also asked whether a missense mutant, G609R, exhibited aberrant targeting. We
have identified this mutation in a large Caucasian pedigree who all exhibited classical
features of dRTA. Expression in Xenopus oocytes and measurement of Cl- and HCO3fluxes revealed normal DIDS-inhibitable anion exchange, suggesting that as with
AE1_11, loss-of-function of AE1 cannot explain the severe disease phenotype in this
kindred. Epitope-tagged AE1-G609R expressed in polarized cells exhibited non-specific
targeting. Of note, the degree of nephrocalcinosis and renal impairment in affected
individuals with either this or AE1_11 is striking.
These findings suggest that the 7th transmembrane domain containing G609, together
with various motifs in the C-terminus (in addition to Y904), play important roles in
targeting AE1 to the correct cell surface compartment. They confirm that dominant
dRTA is associated with non-polarized trafficking of AE1 with no significant effect on
anion transport function in vitro.
Renal Association October 2003
Erythropoietin Protects the Kidney against the Injury and Dysfunction caused by
Ischaemia-Reperfusion Injury by the inhibition of the Caspase Cascade and
EJ Sharples1, N Patel1, P Brown2, K Stewart2, H Mota-Philipe3, M Sheaff1, DA Allen1,
SM Harwood1, J Kieszkiewicz1, MJ Raftery1, C Thiemermann1 and MM Yaqoob1
Anthony Raine Laboratory, Suite 22, Dominion House, Bartholomew close, West
Smithfield, London, EC1A 7BE, United Kingdom, 2Department of Pathology,
University of Aberdeen, Aberdeen, EC1A 7BE, United Kingdom and 3Laboratory of
Pharmacology, University of Lisbon, Lisbon, EC1A 7BE, Portugal
Recombinant human erythropoietin(EPO)has been shown to protect neuronal cells in
vitro and in vivo from apoptotic cell death induced by ischaemia. The EPO receptor is
present on proximal tubular epithelial cells, mesangial cells and the glomerulus. We
studied the direct effects of EPO on human proximal tubular epithelial cells (HK-2
cells) and cell death induced by serum starvation, hydrogen peroxide and ATP depletion.
EPO attenuated cell death (LDH release, DNA fragmentation) in response to oxidative
stress, ATP depletion and serum starvation (p<0.001, p<0.05, p<0.05 respectively).
Inhibitor studies with LY294002(PI-3 kinase inhibitor) and AG490 (JAK2 inhibitor)
abrogated the protection observed with EPO in all experimental models, indicating the
anti-apoptotic effects of EPO were dependent on EPO receptor/ JAK2 signalling and the
phosphorylation of AKT by phosphoinositide-3 kinase. After serum starvation, EPO
reduced cytochrome c release and associated caspase-3 activation, with upregulation of
BCL-XL and XIAP(Western blotting).
In an established animal model of severe renal ischaemia reperfusion injury (45 minutes
bilateral ischaemia, 6 hours reperfusion), a single intravenous bolus of EPO (300 U/Kg),
administered either 30 minutes pre-ischaemia, or 5 minutes before the onset of
reperfusion, significantly reduced renal dysfunction (serum creatinine I/R 217±7.2, EPO
pre-ischaemia 144±5.9, EPO pre-reperfusion 143±7, p<0.01,p<0.01), and tubular
injury(urinary NAG activity: I/R 22.7±5.6, EPO pre ischaemia 9.69 ±1.6, EPO prereperfusion 7.96±0.7, p<0.01). EPO prevented caspase-3, -8 and -9 activity in vivo
(activity assay and Western blotting) and reduced apoptotic cell death(p<0.05) and
histological changes of tubular injury, luminal congestion and necrosis(p<0.05). EPO
administration as late as 30 minutes after the onset of reperfusion was still associated
with a significant reduction in renal dysfunction (p<0.05). In summary, erythropoietin
significantly reduces the injury caused by renal ischaemia-reperfusion through direct
inhibition of proximal tubular epithelial cell death. These findings have major
implications to the management of ischaemic acute renal failure.
Renal Association October 2003
Anti-inflammatory effect of NF-_B-inhibited macrophages in experimental
HM Wilson, D Walbaum, AJ Rees and DC Kluth
Department of Medicine and Therapeutics, University of Aberdeen, Institute of Medical
Sciences, Foresterhill, Aberdeen, AB25 2ZD, United Kingdom
Macrophages infiltrating kidney can either cause injury or facilitate its resolution and
repair depending on how they are activated. We have previously shown that glomerular
macrophages in acute nephritis have fixed proinflammatory properties even when
stimulated with anti-inflammatory cytokines. The present experiments examine the
effect of inhibiting the major proinflammatory signalling pathway NF_B, on
macrophage activation and were designed specifically to answer the question whether
NF_B-inhibited macrophages develop anti-inflammatory rather than proinflammatory
properties when infiltrating inflamed glomeruli.
Primary cultures of macrophages transduced with adenovirus containing a dominant
negative form of I_B (Ad-I_B) or Ad-null (control vector with no insert) were labelled
with PKH-26L and injected into the left renal artery of rats, 24 hours after the induction
of nephrotoxic nephritis. They localised efficiently and in similar numbers within
glomeruli. We developed a unique method whereby injected, modified macrophages
were recovered from inflamed kidney and their properties analysed by flow cytometry
after in vivo incubation. Using this method, Ad-I_B transduced macrophages (recovered
24h after injection) showed a significant decrease in expression of the inflammatory
markers MHC class II (17% decrease) and inducible nitric oxide synthase (iNOS) (27%
decrease) when compared to Ad-null transduced control macrophages. In terms of
disease pathogenesis, Ad-I_B transduced macrophages reduced albuminuria (24 hours
after injection) compared to unmodified disease or injection of Ad-null transduced
macrophages (mg/24h: 189±32-I_B, 297±30-control, 278±35-null: p< 0.05). In
addition, Ad-I_B transduced macrophages significantly decreased the number of
infiltrating macrophages and MHC class II expressing cells within tissue sections of
injected kidneys compared to the contralateral kidney, unmodified disease or kidney
with Ad-null transduced cells.
This is the first demonstration that blocking the NF-_B pathway prevents the
development of inflammatory responses by macrophages in a complex inflammatory
environment in vivo and that these "anti inflammatory" macrophages decrease
glomerular inflammation. Diverting macrophage properties in this way may provide a
powerful mechanism to downregulate renal inflammation and restore normal function.
Renal Association October 2003
The novel vascular endothelial growth factor (VEGF) isoform VEGF165b is antiangiogenic in vivo
W-Y Wang, SJ Harper and DO Bates
Microvascular research laboratories, Department of Physiology, Pre-clinical Veterinary
school, Southwell St, Bristol, BS2 8EJ, United Kingdom
Vascular endothelial growth factor (dominant form VEGF165) is regarded as a propermeability pro-angiogenic vasodilator. However healthy human podocytes produce
VEGF in the absence of angiogenesis. While investigating this paradox we have
identified a novel VEGF isoform – VEGF165b –in kidney and other tissues which
differs in 6 crucial amino acids. VEGF165b inhibits VEGF165-mediated endothelial
cell proliferation and migration in vitro and vasodilatation ex vivo. We have presviously
presented siRNA data indicating that VEGF165b is the predominant isoform produced
by differentiated human podocytes
To determine whether VEGF165b is anti-angiogenic in vivo we developed an
angiogenesis model in a system amenable o functional characterisation. Male wistar rats
(300g) were anaesthetised with halothane (5%), the gut exposed after laparotomy, and
the mesentery superfused with Ringer’s solution at 37°C. A mesenteric panel with few
vessels and no overt angiogenesis was imaged using a digital camera and light
microscope. 50ml of adenovirus (Adv)(1-3.3x108 TCID50/ml) expressing VEGF165
(Adv-165), VEGF165b (Adv-165b), both or Adv-GFP and Monastral blue (0.6%)
diluted in rat Ringer were injected into the mesenteric fat pad adjacent to the panel (n=6
each group). The mesentery was then imaged using intra-vital microscopy and videorecorded, replaced in the animal, incision sutured and the animal allowed to recover. Six
days later the same panel, identified by the Monastral blue depot, was identified as
above. The mesentery was imaged and before. The mesentery was fixed and stained for
endothelial cells (isolectin-1B4), dividing cells (Ki-67), and actin (phalloidin).
Fractional vessel density (FVD) was calculated with confocal microscopy. Data
analysed by ANOVA
Adv-165 injection significantly increased Ki-67 staining, sprouting of blood vessels,
number of vessels. This data is summarised by an increase in FVD (Adv-GFP
44.9±7.6% vs Adv-165 421±55%, p<0.001). In contrast, Adv-165b did not stimulate
angiogenesis (FVD for Adv-165b alone = 26±17.8) and significantly inhibited Adv-165mediated changes when injected together (Adv-165 PLUS Adv-165b 185±47%, p<0.04
vs Adv-165 alone).
In summary, this data demonstrates that VEGF165b is ant-angiogenic in vivo and
explains the paradox of high VEGF expression in the absence of new vessel formation
in the glomerulus.
Renal Association October 2003
The renal histopathology of patients with renal impairment and atherosclerotic
renovascular disease (ARVD) is indistinguishable from hypertensive injury
AE Shurrab1, JR Wright1, R Reeve2 and PA Kalra1
Department of Renal Medicine, Hope Hospital, Stott Lane, Salford, M6 8HD, United
Kingdom and 2Department of Histopathology, Hope Hospital, Stott Lane, Salford, M6
8HD, United Kingdom
Many patients present with ARVD and chronic renal failure (CRF). That renal function
improves in the minority of patients after revascularization of renal artery stenosis
(RAS) lesions, and that baseline proteinuria predicts renal functional prognosis in
ARVD, highlight the importance of renal parenchymal injury in CRF pathogenesis. We
undertook a histopathological analysis of renal biopsies obtained in hypertensive
patients with CRF who were all investigated for ARVD.
Thirty patients underwent renal angiography and 14 were found to have RAS. Renal
biopsies were obtained from the RAS kidney in 11, and from the contralateral (normal
vessel) kidney in 3 of the latter patients, and from the left kidney of the 16 non-RAS
patients. Semi-quantitative histopathological analysis (maximum score 3 for each of
glomerulosclerosis, tubular atrophy, interstitial fibrosis and vascular damage) was
performed by a pathologist blinded to the clinical diagnosis. Pathological scores were
compared to clinical parameters at baseline and latest follow-up.
The mean±SD age of the RAS patients (10 males, 4 F) was 70.1±7.2 and the non-RAS
patients (11 M, 5 F) 65.0±11.2 (NS) years. Follow-up periods (24.1±10.8 vs. 29.0 ±14.4
months, NS) were comparable, and no difference in blood pressure control was evident.
Baseline creatinine clearance (RAS 22.8±13.9 vs. 31.8±17.1 ml/min) and proteinuria
(RAS 1.1±0.9 vs. 1.0±1.3 g/24-hr) were similar. Rate of renal functional deterioration
over time (_GFR = -2.4 RAS vs. –3.8 ml/min/year), was also NS. The histopathological
‘chronic damage scores’ were 5.6±3.1 for patients with RAS and 7.8±2.8 (NS) for nonRAS. Chronic damage score correlated with _GFR, P<0.05, but with no other
parameter. The pathologist was unable to differentiate the histology of the RAS patients
from those without RAS, and all were classed as having varying severity of
hypertensive injury. Twelve of the RAS patients were also investigated with isoSKGFR; the mean SK-GFR of the RAS kidneys (6 had RAS > 70%, 3 RAS 50-70% and
3<50%) was 13.5±6.8 and that of the contralateral (normal vessel) kidneys 10.5±6.6
ml/min (NS).
This study suggests that in most patients with ARVD, CRF is due to hypertensive renal
injury, not only in the contralateral kidney, but also in the kidney supplied by a vessel
with RAS.
Renal Association October 2003
Population based estimates of the prevalence and referral of diabetes-related renal
failure in a UK region.
DG Fogarty1, EA Reaney2, G Savage2, F Kee2 and AP Maxwell1
Nephrology Research Group, Queen's University Belfast, Belfast City Hospital,
Belfast, BT9 7AB, United Kingdom and 2Department of Epidemiology & Public
Health, Queen's University Belfast, Belfast, BT12 6BA, United Kingdom
Background: Despite published guidelines it is recognised that many patients are
referred late to nephrologists. In Scotland, in the early 1990s, almost two-thirds of
patients with serum creatinine > 300mmol/L were not referred for nephrological
assessment (Khan IH, 1994). In England, the reported prevalence of unreferred CRF is
4700 pmp (~0.5%) (Stevens P, 2002).
Aims and Methods: We used routine clinical biochemistry systems in Northern Ireland
to download all renal- and diabetes- related tests performed over a 2-year period. Files
were cross-referenced with renal databases to identify those patients already known to
renal services.
Results: Northern Ireland has a population of 1.7 million. In 2001, 399,511 individuals
had a serum creatinine test, (23.7% of the total population). 5.3% of patients tested had
a creatinine > 150 umol/l. We grouped the tested individuals according to 3 levels of
renal function based on serum creatinine values used in previous surveys (serum
creatinine of 150, 300 and 500 umol/l). Patients likely to have diabetes were identified
on the basis of having glycosylated HbA1c results in either 2001 or 2002. 50,652
patients are likely to have diabetes, a prevalence of 3%. The predicted diabetes patients
have a prevalence of CRF of 10.5%. The degree of renal failure in the non-diabetic and
diabetic populations and whether they attend nephrologists is shown in the Table.
Nephrologists have seen 29%, 59% and 76% of patients with the 3 grades of renal
failure and labelled as having diabetes. The projected figures for the whole of the UK
would represent 14,000 diabetes patients with creatinine > 300 umol/l and over 4000
patients with a level > 500 umol/l not attending nephrologists. Northern Ireland has a
predominantly Caucasian population so it is likely that this projection is an
Conclusion: The prevalence of measured renal failure is much higher than previous UK
studies suggested. Delaying progression of renal failure by earlier referral of patients
with an elevated creatinine is desirable but could not be achieved with existing
nephrology resources. Although we have yet to define how many of these patients may
have had terminal illnesses and/or acute renal failure the diabetes data is alarming. New
approaches to screening and management of CRF patients are clearly needed employing
greater primary care input.
Renal Association October 2003
Parallel Session – Clinical Nephrology
Papers O11-O16
Wednesday 8 October
Main Auditorium
Renal Association October 2003
Effect of ACE Inhibition on Haemoglobin in Chronic Renal Failure: A
Randomised Controlled Trial.
M.S. MacGregor1, C.J. Deighan1, R.S.C. Rodger2 and J.M. Boulton-Jones1
Renal Unit, Walton Building, Royal Infirmary, Glasgow, G4 0SF, Sc, United Kingdom
and 2Renal Unit, Western Infirmary, Glasgow, G11 6NT, S, United Kingdom
ACE inhibitors may cause anaemia in renal transplant recipients, and indeed are used as
a treatment for post transplant erythrocytosis. However, it remains controversial
whether they can cause anaemia or erythropoietin resistance in chronic renal failure
patients. We previously reported a randomised controlled trial of ACE inhibition in
progressive renal failure. Here we present an analysis of the effect of quinapril on
haemoglobin in these patients.
Seventy three patients with progressive non-diabetic chronic renal failure were
randomised to quinapril (Q; n=28), amlodipine (A; n=28) or both drugs (QA; n=17) and
followed for four years. Haemoglobin was measured quarterly, and erythropoietin use
recorded. Follow-up was censored if patients were put on renal replacement therapy,
died or if an ACE inhibitor was stopped (Q, QA) or started (A).
Renal function was poor (GFR 20.3±10.5 ml/min/1.73m2), but equal between the three
groups. Eighteen patients had polycystic kidney disease. Excluding them from the
analysis did not change the results. Haemoglobin was equal in all three groups at
baseline (Q 11.7±1.9 g/dl, A 12.0±2.2 g/dl, QA 12.1±2.0 g/dl). Baseline use of
erythropoietin was similar (Q n=1, dose 4000 units/week; A n=1, dose 2000 units/week;
QA n=2, mean dose 1500 units/week). The quinapril groups had a consistently lower
mean haemoglobin than group A from months 3 to 27 (difference between groups 0.271.06 g/dl), but this was not significant. Erythropoietin use was generally higher in the
quinapril groups, but only significantly at month 6. Group Q had a lower mean
haemoglobin than group A from months 3 to 36 (difference between groups 0.19-1.46
g/dl), which approached significance in months 3, 15 and 21. Erythropoietin use was
significantly higher in group QA compared to group A in months 6, 9 and 12. There was
no correlation between baseline GFR and change in haemoglobin at 3, 6 or 12 months in
any of the groups.
Quinapril appeared to worsen anaemia or increase erythropietin use during this four
year trial. However the effect of ACE inhibition on haemoglobin in chronic renal failure
is likely to be too small to be of clinical significance, even in patients with relatively
advanced renal failure. Although we show no major effect in our population, ACE
inhibitors should nevertheless be considered as a cause of anaemia in individual
Renal Association October 2003
On-line Conductivity Monitoring Demonstrates Alterations of Sodium Flux: A
Randomised Controlled Trial of Reduced Dialysate Sodium Concentration
S H Lambie and C W McIntyre
Renal Unit, Derby City General Hospital, Uttoxeter Rd, Derby, DE22 3NE, United
Relatively low dialysate conductivity (D cond.) is desirable as it is associated with
lower interdialytic weight gains, and improved BP control. Excess sodium removal can,
however, lead to hemodynamic instability. We performed a randomised controlled trial
of empirical reduction of dialysate conductivity. Sodium removal as ionic mass balance
(IMB), as well as plasma conductivity, was measured by conductivity monitoring, and
routine clinical measurements were used to assess the clinical impact.
28 patients were recruited, and randomised to either maintenance of D cond at 13.6
mS/cm (equivalent to 140 mmol/l of Na+), or serial reduction of D cond in steps of 0.2
mS/cm. Reduction was guided by symptoms and BP.
Of 16 patients in Group 2, 6 achieved D cond 13.4 mS/cm, 6 achieved 13.2 mS/cm, and
4 achieved 13.0 mS/cm (13.0 mS/cm was pre-specified as the lowest acceptable D
cond). No episodes of dysequilibrium occurred. Results are expressed as those achieved
at minimum D cond once this was established, compared with baseline at 13.6 mS/cm
(shown as mean ± SEM). Interdialytic weight gain was reduced from 2.34 ± 0.10 kg to
1.57 ± 0.11 kg p<0.0001. Both pre and post dialysis BPs were significantly reduced
(pre-dialysis systolic BP fell from 144 ± 3 mmHg to 137 ± 4 mmHg p<0.05). The
reduction in convective sodium removal due to reduced weight gains was matched by
an increase in the amount of sodium removed by diffusion (91 ± 12 mmol cf. 158 ± 12
mmol). Finally pre-dialysis plasma conductivity also fell, from 14.23 ± 0.04 mS/cm to
14.02 ± 0.05 mS/cm, suggesting a reduction in total body sodium levels.
In summary, we have demonstrated that reduction in D cond monitored by IMB is safe
and practical, and leads to improvement in interdialytic weight gains, and BP control
while maintaining total sodium removal at the same level.
Renal Association October 2003
Locking of tunnelled haemodialysis catheters with gentamicin and heparin
significantly reduces line sepsis rates and epoetin requirements
CW McIntyre, LJ Hulme, MW Taal and RJ Fluck
Department of Renal Medicine, Derby City General Hospital, Uttoxeter Road, Derby,
DE223NE, United Kingdom
Line related sepsis is a major cause of morbidity and mortality in patients receiving
haemodialysis. Antibiotic locking combined with citrate as a bacteriocidal anticoagulant
has been shown to increase both the success of systemic antibiotic treatment in line
sepsis, and to reduce the incidence of sepsis. We have studied the use of gentamicin
locking with standard heparin, to reduce line infection rates. Furthermore, we have
investigated the effects of this strategy on epoeitin requirements and vascular access
50 patients were recruited over a one year period. Patients were randomised at the time
of insertion of catheter to receive locking with either standard heparin (5000 iu/ml)
alone, or gentamicin and heparin (5 mg/ml). Line spsis was diagnosed using CDC
criteria. Access function was studied using on-line measurement of Kt/V by ionic
dialysance and intermittent QA measurements using an ionic dialysance/ cross-lined
based method. Epoeitin requirements and haemoglobin response were monitored over
the study period.
The gentamicin locked group suffered only one infective episode (0.3/1000 catheter
days) as compared to 10 episodes in six patients in the heparin alone group (4/1000
catheter days, p=0.02). Time from first insertion to first episode of line sepsis was also
significantly different (130±16.4 days for gentamicin locked group c.f 85±14.2 days for
the heparin group, p=0.03). 11 of the 25 patients in the antibiotic locked group had had a
catheter in situ in the preceding six months. This had resulted in 10 episodes of line
related sepsis in 5 patients (rate of 8/1000 catheter days). There were no difference in
access function. Use of antibiotic locking was associated with both a higher mean
haemoglobin (10.1±0.14 g/dl vs. 9.2±0.17 g/dl in the heparin group, p=0.003), and a
lower mean epoetin dose (9000±734 iu/week vs. 10790±615 iu/week in the heparin
group, p=0.04).
In conclusion, the practise of locking newly inserted tunnelled central venous catheters
with gentamicin and heparin is a cost effective strategy to reduce line sepsis rates, with
no adverse consequences in terms of gentamicin toxicity or poorer access function. The
reduced levels of overt, and possibly sub-clinical, sepsis are associated with beneficial
effects on epoetin requirements.
Renal Association October 2003
META-ANALYSIS of randomised trials
A Findlay1, D Adu1, N Ives2 and K Wheatley2
Deapartment of Nephrology, Queen Elizabeth Hospital, Edgbaston, Birmingham,
B152TH, United Kingdom and 2Birmingham Clinical Trials Unit, Park Grange, 1
Somerset Road, Edgbaston, Birmingham, B15 2RR, United Kingdom
The use of intravenous radiographic contrast media may lead to the development of
significant renal impairment or renal failure. Risk factors for the development of
contrast nephropathy include renal impairment, congestive cardiac failure, dehydration,
diabetes mellitus, a high dose of contrast medium and old age. The possible role of
contrast media induced reactive oxygen products in nephrotoxicity has led to the use of
acetyl cysteine a scavenger of reactive oxygen in prophylaxis.
We performed a meta-analysis of published randomised controlled trials (RCTs) data to
study the efficacy of acetyl cysteine in preventing contrast nephropathy using an
assumption free method. A search was performed of the Medline, Embase, and
Cochrane databases for years 1996-2003 inclusive, plus search of medical editors’ trial
amnesty for RCTs comparing acetyl cysteine with placebo or no additional treatment in
patients with renal impairment who received saline prior to radiographic contrast media.
Seven RCTs involving 805 patients met the selection criteria. Acetyl cysteine
significantly reduced the risk of contrast nephropathy (odds ratio: 0.41, 95% confidence
interval: 0.27-0.62, P=0.00002). There was significant heterogeneity of treatment effect
between trials (chi-square=15.4, df=6, p=0.02), but this could not be explained by
patient age or trial design features such as renal function, contrast dose, prevalence of
diabetes, use of angiotensin converting enzyme inhibitors or the incidence of
nephrotoxicity in controls.
Overall, adding acetyl cysteine to normal saline substantially reduces the risk of contrast
nephropathy in patients given radiographic contrast media, but because of significant
heterogeneity between trials we conclude that the effectiveness of acetyl cysteine cannot
be definitively established from our meta-analysis and that new appropriately sized
studies are required to confirm its role in preventing contrast nephropathy.
Renal Association October 2003
Nurse led protocol for achieving better blood pressure control in peritoneal
dialysis patients, without increasing antihypertensive medication.
W Harman, J Cox, A Collinson and S Holt
Renal Unit, Royal Sussex County Hospital, Brighton and Sussex University Hospitals,
Brighton, BN2 5BE, United Kingdom
The commonest causes of death in dialysis patients are related to cardiovascular
complications. Despite the fact that blood pressure control is a major and modifiable
risk factor for cardiovascular disease, blood pressure (BP) targets are infrequently
achieved. We set out to improve BP control by using a nurse-led multidisciplinary
protocol to achieve good fluid salt and water balance. Anecdotally we had found that
medical input in the process often resulted in an increase in anti-hypertensive
medication.56 patients were on peritoneal dialysis in both June and December 2002 and
were audited for BP control (mean last 3 readings), antihypertensive medication, weight
and target weight. Hypertensive patients were identified as those in whom BP
>140/85mmHg, and a flow-chart guided patient care emphasising dietetic and nonpharmacological interventions aimed at salt reduction. Following intervention the audit
was repeated and the results were analysed by using a paired t-test.
[table 1 here]
We have shown that it is possible to significantly improve BP control by simple
attention to salt and water balance and consequent reduction in weight. This has been
achieved, without additional antihypertensive medication, by a structured
multidisciplinary approach. This method merely encourages a logical approach to blood
pressure control in dialysis patients, and discourages the use of antihypertensive
medication before non-pharmacological intervention has been optimised. Apart from
trying to minimise pill count, this approach is also potentially cost saving over the life
of the peritoneal dialysis patient. We will continue to use refine this method and repeat
audit to optimise fluid balance, concentrating further on those patients who did not
achieve target blood pressures, in order to improve the algorithm. We now have revised
Renal Association BP targets (130/80) which are going to be even more challenging to
acheive, so this approach will assume even more importance in the future.
Renal Association October 2003
Parathyroid Hormone (PTH) in Children with Chronic Renal Failure; relationship
between the 1-84PTH:C-PTH Ratio and Growth
S Waller1, D Ridout2, T Cantor3 and L Rees1
Nephro-Urology Unit, Institute of Child Health, 30 Guilford Street, London, WC1N
1EH, United Kingdom, 2Paediatric Epidemiology and Biostatistics, Insitute of Child
Health, 30 Guilford Street, London, WC1N 1EH, United Kingdom and 3Scantibodies
Laboratories Inc., 9336 Abraham Way, Santee, California, Ca 92071, USA
Effects of PTH and long carboxyl-terminal fragments (C-PTH) on growth are not
understood. Optimal PTH levels in childhood are unknown; we aim for normal range
PTH levels. Current 'intact' immunoradiometric (iIRMA) PTH assays cross-react with
C-PTH, which antagonises the biological actions of 1-84PTH. The introduction of an
assay (CAP-PTH) that is specific for 1-84PTH, enables estimation of the agonist (184PTH) to antagonist (C-PTH) ratio. In adults this ratio may be predictive of bone
turnover; it may, therefore, also be relevant to growth. The aim of this study was to
investigate the relationship between growth and the 1-84PTH:C-PTH ratio in children
with chronic renal failure (CRF).
195 patients, median (range) age 9.9 years (0.3-17.1) were recruited: 168 with a GFR
<60 mls/min/1.73m2 (including 42 transplanted patients), 19 on peritoneal- and 8 on
haemo-dialysis. Over a median (range) period of 1.1 (0.5 to 1.7) years, children attended
5 (3-15) clinics at which iIRMA PTH and CAP-PTH were measured and height
standard deviation score (Ht SDS) was calculated. Mean PTH levels were within the
normal range (NR) for both assays; CAP-PTH 28.5 pg/mL (NR 5-39), iIRMA 45.1
pg/mL (NR 14-66). The patients grew normally (change in Ht SDS per year (_Ht SDS) 0.01). There was a correlation between _Ht SDS and the 1-84PTH;C-PTH ratio (r=0.2,
p=0.01). Furthermore those with normal range PTH levels had a significantly higher
ratio than those with PTH levels outside NR and grew significantly better than those
with PTH levels outside NR (see table).
Normal PTH levels in children with CRF are associated with a higher 1-84PTH:C-PTH
ratio and normal growth. Normal range PTH levels are appropriate in children with
CRF. C-PTH may be of clinical significance.
Renal Association October 2003
Parallel Session – Renal Laboratory Science
Papers O17-O22
Wednesday 8 October
Jehangir Room
Renal Association October 2003
Leptin Deficient Mice are Protected from Nephrotoxic Nephritis
RM Tarzi1, HT Cook2, I Jackson3, CD Pusey1 and GM Lord3
Division of Medicine, Imperial College, Hammersmith Hospital, London, W12 0NN,
United Kingdom, 2Department of Histopathology, Hammersmith Hospital, London,
W12 0NN, United Kingdom and 3Department of Immunology, Hammersmith Hospital,
London, W12 0NN, United Kingdom
Leptin is an adipose tissue-derived peptide hormone, and is a central mediator of the
neuroendocrine pathways involved in the control of food intake, basal metabolism and
reproductive function. Recent evidence indicates that leptin modifies proinflammatory
immune responses and may provide a key link between nutritional deficiency and
immune dysfunction. To study the influence of leptin deficiency on glomerulonephritis,
susceptibility to accelerated nephrotoxic nephritis was examined in genetically leptin
deficient C57BL/6-ob/ob mice and wild-type C57BL/6 controls. The C57BL/6-ob/ob
mice display a number of metabolic abnormalities including obesity, diabetes, reduced
activity, reduced metabolic rate and infertility. The glomerulonephritis model was
induced with sheep anti-mouse glomerular basement membrane globulin (nephrotoxic
globulin) injected to mice pre-immunized against sheep IgG. The dose of nephrotoxic
globulin was adjusted according to the weight of the animals, and mice were sacrificed
eight days after induction of disease. The leptin-deficient ob/ob mice were strongly
protected from glomerular crescent formation, macrophage infiltration, glomerular
thrombosis and albuminuria. Wild-type mice had crescents in 11 ± 2% of glomeruli,
compared with none in the leptin-deficient mice (p[<]0.01). There were 3 ± 0.2
macrophages/glomerular cross section in the wild-type mice, but only 0.6 ± 0.1 in the
leptin-deficient mice (p[<]0.01). Albuminuria, measured at day 3 after induction of
disease, showed 9 ± 2mg/12h in the wild-type, and 3mg ± 1mg/12h in the leptindeficient group (p[<]0.05). In two of three experiments performed, the leptin-deficient
mice had a diminished humoral immune response to the sheep nephrotoxic globulin,
reflected by lower glomerular mouse IgG deposition and lower circulating levels of IgG
against sheep IgG. However, in one of the three experiments, the immune response of
the leptin-deficient mice was as strong as wild-type, but the mice were still protected
from disease. This suggests additional defects in the innate immune responses of the
leptin-deficient mice, in addition to the defects in the humoral immune response. These
findings demonstrate that leptin is required for the induction and maintenance of
immune-mediated glomerulonephritis, and that blockade of the leptin axis might
provide an attractive therapeutic possibility in human autoimmune disease.
Renal Association October 2003
Targeting Rho GTPases to inhibit human mesangial cell proliferation
A Khwaja and B.M. Hendry
Department of Renal Medicine, Kings College London, Bessemer Road, London, SE5
9PJ, United Kingdom
Rho GTPases play a key role in the transduction of proliferative cytokines from the
plasma membrane to the cell nucleus and in maintaining cytoskeletal structure. The
post-translational modification of these GTPases by the addition of a prenyl group is
thought to be essential for their plasma membrane localisation and signaling function.
We present a study of Rho isoform expression in human mesangial cells (HMC) and the
actions of prenylation inhibitors (lovastatin and geranylgeranyltransferase inhibitor,
GGTI 298) and a specific Rho inhibitor (C3 exoenzyme) on HMC proliferation,
apoptosis and Rho activation.
Proliferation was assessed using a tetrazolium salt based assay of viable cell number
and BrdU uptake. Growth was stimulated using PDGF and FCS. Apoptosis was
measured using Hoechst 33342 staining. Activated Rho was detected with selective
affinity precipitation of Rho-GTP with the Rho binding domain of Rhotekin. The actin
cytoskeleton was stained using TRITC-conjugated phalloidin.
RT-PCR demonstrated that HMC express mRNA for the isoforms RhoA, RhoB and
RhoC. Quantitative Western blotting indicated that RhoA was the predominantly
expressed protein with little expression of RhoB or RhoC. GGTI 298 (20uM) reduced
cell number by over 90% at 48hours, reduced BrdU uptake by over 60% and induced
apoptosis in 25% of cells. Lovastatin (10uM) reduced cell number by over 80%, BrdU
uptake by 60% and induced apoptosis in 10% of HMC. C3 exoenzyme reduced BrdU
uptake by over 50% and induced apoptosis in 25% of HMC. Both GGTI 298 and
lovastatin reduced expression of RhoA in the plasma membrane and were also
associated with a reduction in Rho-GTP activity. The effects of lovastatin on
proliferation, apoptosis and Rho-GTP activity were attenuated by the coincubation of
geranylgeranylpyrophosphate. GGTI 298 and lovastatin inhibited actin stress fibre
formation in a manner similiar to that seen with C3 exoenzyme.
We conclude that i) HMC express mRNA for RhoA, RhoB and Rho C, with RhoA being
the predominantly expressed protein. ii) RhoA needs to be geranylgeranylated to
localise to the plasma membrane and to be functionally active. iii) the effects of GGTI
298 and lovastatin on proliferation, apoptosis and the actin cytoskeleton appear to be
mediated by inhibition of RhoA activity.iv)Rho inhibitors, such as lovastatin may be
beneficial in the treatment of mesangioproliferative renal disease.
Renal Association October 2003
Constitutive over-expression of polycystin-1 is associated with decreased cell
proliferation and defective tubulogenesis in PKD1 transgenic cells
AJ Streets, E Parker and ACM Ong
Academic Nephrology Group, Sheffield Kidney Institute, University of Sheffield,
Sheffield, S5 7AU, United Kingdom
We have recently shown that polycystin-1, the gene product of the PKD1 gene, can
function as a cell adhesion molecule, primarily mediating intercellular adhesion in renal
epithelial cells (JASN 2003; 14:1804-1815). Cyst epithelial cells are characterised by
changes in proliferation, apoptosis and tubulogenesis. Since all these properties could be
directly related to alterations in cell-cell adhesion, we have investigated whether
polycystin-1 over-expression in conditionally immortalised murine PKD1 transgenic
collecting duct cells (M7) leads to an alteration in these key cellular functions compared
to genetically identical control cells (M8). Under basal conditions, the growth rate of
M8 cells was consistently 50% faster than M7 cells from d2-d5 after plating suggesting
that polycystin-1 inhibits cell proliferation. By contrast, no difference in the rate of
spontaneous or induced (serum starvation) apoptosis could be detected between the two
lines. We next compared the ability of M7 and M8 cells to undergo spontaneous
tubulogenesis in 3D collagen gels. Unexpectedly, while M8 cells consistently formed
branching tubules, M7 cells predominantly formed cysts in this assay. Since M7 cells
have been shown to have a higher surface expression of functional polycystin-1, we
hypothesized that the defect in tubulogenesis might be due to a dysregulation of
polycystin-1 mediated cell adhesion. Preincubation with staurosporine, a broad
spectrum protein serine/threonine kinase inhibitor resulted in a 30-50% increase in
resistance to the disruptive effects of the IgPKD antibody on cell-cell adhesion in M8
cells. In contrast, M7 cells remained resistant to IgPKD antibody treatment and this was
not altered by preincubation with staurosporine. These observations suggest that
constitutive over-expression of polycystin-1 leads to defective tubulogenesis in PKD1
transgenic cells, possibly due to a dysregulation of polycystin-1 trafficking or function
despite a reduction in cell proliferation. These findings may explain why PKD1
transgenic animals develop age-dependent renal cysts. Future work will also seek to
define the precise mechanism by which phosphorylation events regulate polycystin-1
mediated cell adhesion in non-transgenic cells.
Renal Association October 2003
Clearance of Apoptotic Cells: A Novel Function of the Greater Omentum
S Watson, J-F Cailhier, S Clay, A Lacy-Hulbert, J Savill and J Hughes
Phagocyte Lab, MRC Centre for Inflammation Research, Teviot Place, Edinburgh, EH9
8AG, United Kingdom
Apoptotic cell (AC) clearance by macrophages (M_) is crucially important in the
resolution of peritonitis with previous work indicating an important role for C1q. In this
study we have investigated the role of the greater omental lymphoid organ (GOLO), a
unique interface between the innate and adaptive immune systems, in AC clearance.
Apoptosis (>85%) was induced in syngeneic thymocytes by incubation with
dexamethasone. AC were labeled with either blue (Hoescht 33342) or green (CM green)
fluorochromes. 5x106 labeled AC were injected IP with peritoneal lavage after 30 mins
indicating >95% AC clearance. Fluorescent microscopy of the excised GOLOs
demonstrated marked AC localisation to specialised M_ -rich areas termed milky spots .
IP administration of a 50/50 mix of differentially labeled AC and viable, non-apoptotic
cells resulted in a 48 fold greater localisation of AC to the GOLO compared to viable
cells (p=0.00006) indicating that localisation was specifically dependent upon
We then studied the involvement of C1q in this system. C1q knockout (KO) AC were
opsonised with either C1q KO or C1q wild-type (WT) serum. A 50/50 mix of
differentially labeled opsonised and non-opsonised AC was injected IP into C1q KO
mice and AC localisation to the GOLO quantified. C1q opsonised AC were 4-fold more
abundant in the GOLO compared to non-opsonised AC (p=0.000005).
We previously noted a profound and consistent reduction in recoverable M_ numbers
following IP AC administration suggesting M_ emigration. Adoptive transfer of 2x106
fluorescently labeled peritoneal M_ together with 5x106 AC resulted in preferential M_
localisation to the GOLO compared to transfer of M_ in isolation (12.24±1.41 M_ /field
vs 2.06±0.50, M_ AC vs M_ alone, p=0.003). In addition, M_ adherent to the GOLO
exhibited evidence of AC ingestion.
In conclusion, this is the first demonstration of specific AC clearance by the GOLO; a
C1q-dependent process. Additionally, M_ exhibit enhanced localisation to the GOLO
following AC ingestion suggesting that AC clearance imparts a migration signal to M_,
a concept with major implications for the resolution of inflammation.
Renal Association October 2003
OFD1 Codes for a Centrosomal Protein Mutated in Human Polycystic Kidney
L Romio1, A S Woolf1, A Fry2, S Malcolm1 and S A Feather3
Institute of Child Health, UCL , 30 Guilford St, London, WC1N 1EH, United
Kingdom, 2Department of Biochemistry, University of Leicester, Leicester, LE1 7RH,
United Kingdom and 3Department of Paediatric Nephrology, St. James's University
Hospital, Leeds, LS9 7TF, United Kingdom
We recently identified OFD1, the gene responsible for the oral-facial-digital syndrome
type 1, a cause of dominant polycystic kidney disease (PKD). The protein colocalised
with gamma-tubulin in human renal mesenchymal cells, consistent with a centrosomal
localisation. In the current experiments,OFD1 was found in a similar location in adult
human proximal epithelia (RPTEC and HK2), and in an established human embryonic
kidney line (HEK293). The subcellular localisation was confirmed by western blot of
centrosomes isolated from HEK293 cells. The OFD1 protein is 1011 aa long, and is
characterised by a lissenchephaly 1 homology (LisH) motif in its N-terminus, with
putative role in microtubule dynamics, followed by five coiled coil (CC) domains.
When OFD1 fused to GFP was transiently expressed in HEK293 cells, GFP
fluorescence localized mainly with gamma-tubulin, confirming that the construct was
targeted to the centrosome. Using deletion constructs, we determined that the Nterminus containing the LisH domain was redundant for centrosomal targeting;
conversely, the number of CC domains was critical for this process, with a single
domain producing partial targeting, and at least two domains needed for complete
targeting. Interestingly, most of the OFD1 mutations reported are predicted to cause
premature truncation of OFD1, with loss of a variable number of CC domains. In OFD1
patients with full clinical phenotype, missense mutations have been reported in the LisH
domain, yet a construct harbouring one such mutation still localised to the centrosome,
suggesting a critical role for this domain in centrosome biology. Recently, the protein
products of several PKD genes have been detected in primary cilia, mechanosensors
which may transduce differentiation signals into renal epithelia. With this in mind, we
sought OFD1 protein in primary cilia of RPTEC and HK2 cells. We conclude that
OFD1 was not a major ciliary protein in these cells, although it was present in the basal
body, consistent with a possible role in biogenesis of the primary cilium.
Renal Association October 2003
DNA Vaccination Against Specific Pathogenic T Cell Receptors Reduces
Proteinuria In Active Heymann Nephritis by Inducing Specific Autoantibodies
H Wu1, G Walters2, JF Knight1 and SI Alexander1
Centre for Kidney Research, Children's Hospital at Westmead, Hawkesbury Road,
Westmead, Sydney, 2145, Australia and 2Department of Nephrology, Leicester General
Hospital, Gwendolen Road, Leicester, LE5 4PW, United Kingdom
We have previously identified potential pathogenic T cells within glomeruli that use T
cell receptors (TCR) encoding V_5, V_7 and V_13 in combination with J_2.6 in
Heymann nephritis (HN), a rat autoimmune disease model of human membranous
nephritis. We have now tested whether DNA vaccination, specifically targeting these
three TCR V_ chains, alters the disease course.
Rats were divided into four groups (n = 6): Group 1: DNA vaccination with TCR V_5,
7, &13 in HN; Group 2: DNA vaccination with TCR V_8 used as a control vaccine in
HN; Group 3: HN; Group 4: Complete Freund’s Adjuvant (CFA) without antigen as a
HN control. The rats were pretreated with 0.75% bupivacaine (1ml/g bw) by injection
into the tibialis anterior muscle 1 week before vaccination. Subsequently 300ug of DNA
was injected into the same site three times at weekly intervals. Two weeks after the last
vaccination, rats were challenged with Fx1A in CFA to induce HN.
DNA vaccination targeting V_5, V_7 and V_13 with Jb2.6 significantly protected
against HN compared with control groups. Proteinuria was reduced at 6, 8, 10 and 12
weeks after immunisation of Fx1A (P < 0.001). Glomerular infiltrates of macrophages
and CD8+ T cells (P<0.005) and glomerular IFN-_ mRNA expression (P<0.01) were
also decreased significantly. DNA vaccination also induces loss of clonality of T cells in
the glomeruli. We examined T lymphocytes in DNA vaccinated rats and identified a
subset with surface binding of antibodies. These CD3+/IgG+ T cells expressed V_5 and
V_13 suggesting the antibodies were induced by the DNA vaccination. Furthermore,
flow cytometry shows that these CD3+/IgG+ cells were CD8+ cells. Analysis of
cytokine mRNA expression showed that IL-10 and IFN-_ mRNA were not detected in
these CD3+/IgG+ cells. These results suggest that TCR DNA vaccination produces
specific antibodies that bind to the TCRs that the DNA vaccination encoded, resulting in
blocking of activation of pathogenic T cells and amelioration of disease.
We have shown that treatment with TCR based-DNA vaccination protects against HN
based on our previous identification of pathogenic V_ TCRs by TCR CDR3
spectratyping analysis and that the mechanism may involve inducing V_ specific
Renal Association October 2003
Moderated Poster Session - One
Group (A)
Anaemia and Access
Tuesday 7 October
Studio Suite
Renal Association October 2003
Non-invasive Assessment Of Central Venous Vascular Access For Haemodialysis
By Ionic Dialysance Measurements
L.J. Hulme, S.H. Lambie and C.W. McIntyre
Renal Unit, Derby City General Hospital, Uttoxeter Rd, Derby, DE22 3NE, United
The haemodialysis (HD) population is facing an increasing reliance on tunnelled semipermanent catheters to provide dialysis access. The problems with semi-permanent
catheters are manifold and include sepsis, stenosis and occlusion of vessels, and poor
flow rates. Satisfactory flow rates are critical for the delivery of adequate HD. QA
represents the maximum flow achievable within venous vascular access and is
assessable by a number of methods. Significant falls in QA are predictive of access
QA can be calculated using ionic dialysance measures. Ionic dialysance represents the
value of the dialysance of electrolytes corrected for ultrafiltration and recirculation, and
can be measured using the Diascan® (Hospal, Dasco, Italy) feature on suitably
equipped dialysis monitors. Measuring ionic dialysance first in the standard
arrangement, and then after reversal of the blood lines to deliberately induce
recirculation, allows QA to be derived using the equation below.
QA = ((D - QF ) . Drev)/ (D -Drev)
Where D = Ionic dialysance, Drev = Ionic dialysance with lines reversed, and QF =
Ultrafiltration rate.
QA can thus be calculated in a reproducible, non-invasive manner without the need for
doppler or ultrasonic thermodilution studies. In fistulae this technique compares well to
the current gold standard of ultrasonic thermodilution (r2 = 0.86- Mercadal et al 1999).
QA derived from ionic dialysance was measured in patients established on HD using
tunnelled catheters to assess its predictive value for access failure in this setting.Results
are presented as mean ± SEM (range).
12 patients (Age 39 – 78) were studied over a period of 30 weeks.
· At inclusion mean QA was 1471 ± 126 ml/min (766 - 2020)
· At 1 month mean QA was 1482 ± 171 ml/min (663 - 2090)
· At 2 months mean QA was 1423 ± 226 ml/min (778 - 2084)
In one case, there was a reduction in QA of 34%, this catheter subsequently clotted. All
of the other catheters studied, which demonstrated no such reduction, remained patent.
This data suggests that assessment of QA in semi-permanent catheters by repeated ionic
dialysance measurement is practicable, and may be useful in the prediction of vascular
access failure. As this is a reproducible, non-invasive, non-operator dependant method
of vascular access assessment, it could be used more widely than other currently
available techniques.
Renal Association October 2003
Experience with Dialock: Subcutaneous implantation of tunnelled vascular access.
L J Chesterton, S H Lambie, M W Taal, R J Fluck and C W McIntyre
Renal Unit, Derby City General Hospital, Uttoxeter Rd, Derby, DE22 3NE, United
The haemodialysis population is facing an increasing reliance on tunnelled semipermanent catheters to provide dialysis access. This is due to many factors, including
the increasing age and frailty of haemodialysis patients and lack of sufficient surgical
time for creation of fistulas. The problems with semi-permanent catheters are manifold
and include sepsis, stenosis and occlusion of vessels, and poor flow rates leading to
inadequate dialyisis. One potential solution to some of these problems is subcutaneous
implantation of the access device.
The Dialock System consists of a titanium body and two central venous catheters made
from coil reinforced silicone that are fully implanted beneath the skin. Specific Dialock
needles can then be inserted percutaneously to open valves in the device, allowing
access to the venous circulation. The valves close automatically when the needles are
withdrawn, sealing off access to the circulation. In the intra-dialytic period the whole
system is entirely subcutaneous, thus reducing the risk of sepsis.
We have prospectively recorded our experience of implantation of this device in terms
of sepsis rates, blood flow, dialysis adequacy, catheter days and site of implantation.
Patients were selected on the basis of their clinical need, most commonly because of
lack of further possibilities for peripheral access and increasingly difficult central
venous access, as well as recurrent line sepsis.
We have inserted 14 Dialock devices, for a mean of 345 ± 72 (34-650) days. The overall
sepsis rate was 3.8 /1000 catheter days, compared with a mean of 5.6/1000 catheter days
within the unit for permcaths over the same time period. There was a mean blood flow
of 310 ± 18.5 (200-400) ml/min, and a mean Kt/V of 1.1 ± 0.12 (0.6-1.4). 8 catheters
were inserted into internal jugular veins (2 left, 6 right), 5 into femoral veins, and one
into a left subclavian vein. Complications have been limited to one catheter which
clotted, two which eroded through the overlying skin requiring surgical flap
transposition, and one device was removed after exposure secondary to haematoma
We conclude that Dialock provides a useful further alternative for vascular access in a
group of patients in whom formation of further access with arteriovenous fistulae or
grafts is precluded.
Renal Association October 2003
On-line measurement of haemoglobin concentration for the management of
anaemia in chronic haemodialysis patients
LJ Chesterton, SH Lambie, LJ Hulme and CW McIntyre
Department of Renal Medicine, Derby City General Hospital, Uttoxeter Road, Derby,
DE223NE, United Kingdom
Regular monitoring of haemoglobin in chronic haemodialysis patients is essential in
ensuring that targets for anaemia management are consistently achieved. However,
repeated and multiple attempts at blood sampling can be both time consuming and
invasive. On line optical, continuous monitoring of haemoglobin concentration would
allow non-invasive assessment of haemoglobin and allow instantaneous changes be
introduced accordingly. This study aimed to evaluate the clinical application of on-line
haemoglobin concentration measurement.
Twelve dialysis monitors (Hospal Integra) were calibrated using a minimum of five
haemoglobin samples spread over at least 4g/dL. Optical measurement of haemoglobin
concentration is already incorporated into the dialysis monitor to allow the study of
relative blood volume (Hemoscan). Paired haemoglobin measurements (dialysis
monitor value and conventional laboratory assessment) were taken at regular intervals
during the following seven months (mean 11.0±0.28 range 7.5-14.8 g/dL).
The linear regression plot (fig.1) documents the accuracy of this technique (R2=0.85,
p<0.0001), which indicates that the machine values are comparable to laboratory
figures. The Bland-Altman plot (fig.2) illustrates the precision of the data with a mean
overestimate of 0.09%. There was no significant deterioration in the quality of this
correlation over the study period (R2=0.90 at the start and 0.97 at the end). On-line
measurement accurately tracked changes in haemoglobin concentration over the study
In conclusion, utilising the ability of the dialysis monitor to measure optical
concentration of haemoglobin is both precise and accurate. Regular on-line assessment
of haemoglobin would allow proactive micromanagement of renal anaemia, give early
warning of sub-optimal response to treatment with a reduction in the time taken to
achieve clinically important targets.
Renal Association October 2003
Conversion of sub-cutaneous to once weekly intravenous erythropoietin alpha
(Eprex) in haemodialysis patients: Stable haemoglobin in most patients
RJ Fluck, MW Taal and CW McIntyre
Department of Renal Medicine, Derby City General Hospital, Uttoxeter Road, Derby,
DE223NE, United Kingdom
Reports of pure red cell aplasia have prompted a change in the recommended route for
Eprex administration. Previous reports have suggested that substantially higher doses
are required when epoetins are administered intravenously (iv) vs. subcutaneously (sc).
In this study Eprex administration was converted from sc to iv in 111 haemodialysis
patients without dose adjustment and haemoglobin (Hb) was monitored over 6 weeks.
After 6 weeks, Eprex doses were increased in those patients with an initial fall in Hb or
in those still below the target Hb of 11g/dl. After conversion to iv dosing, a subset of
patients receiving multiple doses of Eprex per week was converted to once a week
dosing without dose adjustment for the first 6 weeks. During the first 6 weeks Hb
decreased by ≤1g/dl in 20 (18%) patients, two of whom required blood transfusions;
increased by _1g/dl in 31 (28%) patients and changed by <1g/dl in 60 (54%). There was
no significant difference between these groups with respect to BMI, Kt/V, PTH or CRP.
A small but significant increase in serum ferritin and decrease in %HRC was observed.
After 6 weeks Eprex dose was decreased in 5%, left unchanged in 67% and increased in
23% of patients. Mean Hb and Eprex dose over time are shown in the table. There were
small but significant increases in both Hb and Eprex dose over 18 weeks (P<0.01 for
both). Among 26 patients switched from multiple to single weekly iv doses, Hb
decreased by ≤1g/dl in 5 (19%); increased by ≤1g/dl in 7 (27%) and changed by <1g/dl
in 14 (54%). We conclude that Eprex administration may be switched from sc to iv
without initial dosage adjustment in the majority of patients. A small Eprex dose
adjustment is required to maintain Hb in a minority of patients.
Renal Association October 2003
Use of venography at insertion of tunnelled internal jugular vein dialysis catheters
reveals significant occult stenosis
MW Taal, LJ Chesterton and CW McIntyre
Department of Renal Medicine, Derby City General Hospital, Uttoxeter Road, Derby,
DE223NE, United Kingdom
Percutaneous insertion of tunnelled internal jugular dialysis catheters requires the use of
large calibre, rigid tissue dilators that have the potential to cause trauma to the central
veins, particularly if anatomical abnormalities are present. In this study we evaluated the
use of direct venography performed at the time of catheter insertion to minimize the risk
of central vein trauma in 58 consecutive patients (mean age 60.2±15.3y). The internal
jugular vein was entered under ultrasound guidance and venography was performed by
injection of 10ml of contrast material, prior to insertion of a guide-wire. Images were
evaluated on-screen by the operator and a decision made regarding the need for
additional x-ray screening. Hard copies of the venography were assessed after the
procedure. In 25 cases (43%), venography showed evidence of unexpected stenosis
and/or angulation of the central veins of sufficient severity to warrant additional x-ray
screening during insertion of the dilators, or abandonment of the procedure. Patients
who had previously had tunnelled internal jugular catheters had more than double the
incidence of significant central vein anatomical abnormalities than those who had not
(13/19 (68%) vs. 12/39 (31%); P=0.02). In two patients the procedure was abandoned
due to severe stenosis. No patient suffered central vein trauma or pneumothorax. There
were no adverse reactions to contrast injection. We conclude that direct central
venography performed immediately prior to tunnelled internal jugular dialysis catheter
insertion detects unexpected, significant anatomical abnormalities of the central veins in
a substantial proportion of patients and may be useful in minimizing the risk of central
vein trauma.
Renal Association October 2003
Conversion from epoietin to darbopoietin in stable peritoneal dialysis patients
B Camilleri1, A Heaton2, J Hollis1, H Moss3, J Palmer2 and PF Williams3
Dialysis Unit, Addenbrookes Hospital NHS Trust, Cambridge, CB2 2QQ, United
Kingdom, 2Norfolk and Norwich University Hospital, Norwich, NR4 7UZ, United
Kingdom and 3Dialysis Unit, Ipswich Hospital NHS Trust, Ipswich, IP4 5PD, United
Darbopoietin has a prolonged half life compared to both epoietin _ or _ and this may
allow a prolonged administration interval. We switched stable peritoneal dialysis
patients in the maintenance phase of epoietin treatment to darbopoietin to assess the
dose equivalence, patient acceptability and stability of haemoglobin over a six month
study period.
Patients and methods:
29 stable PD patients were enrolled in a multicenter open label single arm study. There
were 19 males, 10 females, mean age 57.3 years, mean duration of PD 37 months and
they were well dialysed (mean Kt /V 2.34, Cr Cl 66.8 l / week) and free from recent
infections. The mean Hb pre trial was 11.7 g / dl, ferritin 446, transferrin saturation 35
% and 13 were receiving epoeitin therapy twice weekly and 16 weekly with a mean
dose of 4931 u / week (65.5 u / kg / week).
Patients were switched from epoietin therapy to darbopoietin with a conversion factor of
200 u : 1 ì, darbopoietin was administered weekly to the 13 patients previously
receiving epoietin twice weekly and fortnightly to the 16 receiving weekly therapy.
Haemoglobin levels and iron stores were measured over the next 26 weeks and
darbopoietin doses adjusted to maintain a target haemoglobin of 10 - 13 g / dl.
27 patients completed the 26 week study - 1 patient received a transplant and 1
withdrew because of an exacerbation of pre existing headaches. Haemoglobin levels
were unchanged at 0, 12 and 24 weeks (mean 11.6, 11.9 and 11.3 g / dl) with no
significant changes in ferritin levels or transferrin saturation (mean 403, 32.5%).
Dialysis adequacy and BP control was unchanged and the need for darbopoietin dose
adjustment was infrequent. At the end of the study period 10 patients were receiving
darbopoietin once weekly and 17 fortnightly. The initial weekly dose was 24.2 u and
was unchanged at the end of the study period (25.5u).
Conversion from epoietin _ or _ to darbopoietin therapy was well tolerated and 25 of
the 27 patients completing the 6 month study elected to stay on darbopoietin.
Haemoglobin levels were maintained within target range without the need for frequent
dose changes and the dose equivalence at the end of the study was unchanged.
Renal Association October 2003
A discordant reticulocyte response to anaemia in EPO treated patients reveals
PRCA in renal databases.
C Tolman1, EJ Will1, C Bartlett1 and G Woodrow2
Department of Renal Medicine, St James's University Hospital, Leeds, LS9 7TF,
United Kingdom and 2Department of Renal Medicine, Leeds General Infirmary, Leeds,
LS1 3EX, United Kingdom
Pure red cell aplasia associated with the production of neutralising antibodies to
erythropoietin (EPO-PRCA) is characterised by reticulocytopaenia, transfusion
dependence and characteristic bone marrow findings, with a history of epoetin (EPO)
treatment. In light of two recent and two historical cases of EPO_-PRCA within the St
James's unit, we decided to establish local reticulocyte ranges from the PROTON
database, in the hope of identifying those with inappropriately low reticulocyte counts
as unrecognised cases.
A total of 3735 paired Haemoglobin (Hb) and reticulocyte counts were identified
between 1998 and 2003, with a mean (±SD) reticulocyte count of 69±37 x 109/l
(2.4±1.1 %).
The only patients in our records with relative reticulocyte counts lower than two
standard deviations below the mean, Hb <10g/dl and a history of EPO treatment were
the 4 known cases of PRCA and a patient with acute myeloblastic leukaemia.
We subsequently screened three other regional Renal Unit databases (over the same
period), using the same criteria, with examination of individual case records where
Of the 8,090 result pairs screened, only two other patients fulfilled the criteria. One is a
male HD patient with hypoplasia of all cell lines (requiring only intermittent
transfusion); the second is confirmed as a previously undiagnosed case of PRCA: a 74
year old man, with advanced chronic renal failure due to renovascular disease.
Investigation for loss of EPO_ efficacy in 1999 demonstrated reduced erythroid series
within the marrow, yet no diagnosis was made. Despite continuing EPO_ therapy he
recovered spontaneously and maintains an Hb of 11.2g/dl on a weekly dose of 10mcg
(sc) darbepoetin_. Anti-Glycophorin staining of marrow sections has since been carried
out, confirming the absence of erythroid precursors and a diagnosis of PRCA.
Unfortunately no historical sera are available for examination for neutralising
antibodies, nor was his Parvovirus status ascertained.
The concealed case confirms the value of examining historical databases for discordant
low reticulocyte counts, which may indicate cases in whom EPO-PRCA has been
previously missed. Such patients are likely to have been given further EPO, review of
which should enlarge understanding of the syndrome. It is likely that other renal units
harbour previously undiagnosed cases of EPO-PRCA in small numbers.
Renal Association October 2003
Successful treatment of pure red cell aplasia with mycophenolate mofetil and
R C Preston1 and A Tan2
Kettering General Hospital, Rothwell road, Kettering, NN16 8UZ, United Kingdom
and 2Leicester General Hospital, gwendolen road, Leicester, LE5 4PW, United
Pure red cell aplasia [PRCA]is a rare complication of recombinant human erythropoetin
use. We describe a case of PRCA associated with the subcutaneous use of epoetin alfa
[Eprex] and its successful treatment with mycophenolate mofetil [MMF] and
A 78 year-old man presented with nephrotic syndrome and renal impairment. Renal
biopsy was consistent with FSGS. He was treated with oral prednisolone [0.5mg/kg],
complicated two months later by the development of spontaneous bacterial peritonitis.
Prednisolone therapy was discontinued. Hospital based haemodialysis therapy was
commenced 8 months later due to progressive renal failure and fluid overload.
Haemoglobin [Hb] was 10g/dl and he was treated with subcutaneous Eprex, 4000
units/week. There was a good initial response to Eprex therapy with Hb rising to 14g/dl.
Eight months later the patient presented with a Hb of 7.9g/dl. MCV, platelet and
leucocyte count, B12, folate, ferritin, CRP and PTH levels were normal. There was no
evidence of gastrointestinal blood loss, haemolysis, sepsis or mitosis. Reticulocyte
count was <5000/mm3.
The dose of Eprex was increased incrementally to a maximum of 20000 units/week with
no improvement in Hb. A bone marrow aspirate demonstrated markedly reduced
erythroid precursors with <1% erythroblasts in an otherwise normal aspirate, consistent
with PRCA. Radioimmunoprecipitation assay confirmed the presence of erythropoetin
antibodies. Eprex therapy was discontinued.
Despite spontaneous recovery of renal function and remission of nephrotic syndrome
the patient remained blood transfusion dependent with an average requirement of 1
unit/week to maintain a Hb>8g/dl. The patient developed severe iron overload
necessitating iron chelating treatment with desferrioxamine.Six months later, due to the
lack of spontaneous recovery of PRCA and the risk of further morbidity and mortality
due to iron overload, the patient was commenced on a combination of MMF [1g/day]
and prednisolone [0.5mg/kg]. Within two months, reticulocyte count had increased
dramatically to 180000/mm3 and haemoglobin to above 10g/dl. MMF was discontinued
after four months. Prednisolone was reduced and withdrawn over 6 months.The patient
has not required any further blood transfusion or desferrioxamine therapy.
We believe that this is the first documented report of the successful treatment of PRCA
with MMF and prednisolone.
Renal Association October 2003
Intravenous Aranesp® (Darbepoetin Alfa) in the management of anaemia of
haemodialysis patients previously receiving subcutaneous erythropoietin
K J Jenkins1, L I Bennett2 and A Monk3
Department Renal Medicine, Kent& Canterbury Hospital, Ethelbert Road, Canterbury,
CT1 3NG, United Kingdom, 2Oxford Kidney Unit, Churchill Hospital, Headington,
Oxford, OX3 7LJ, United Kingdom and 3Kings College Hopsital, Dulwich Renal Unit,
East Dulwich Grove, London, SE22 8PT, United Kingdom
This study looks at the effect on haemoglobin (Hb) in a group of 30 stable
haemodialysis patients from 3 different units when switching them from
subcutaneous(SC) rHuEPO administered up to three times a week to once weekly
equivalent dose of intravenous (IV) Aranesp® for a period of 6 months.(200iu
rHuEPO=1_g Aranesp®)
Aranesp® was administered intravenously (IV) rather than SC in all patients for reasons
of obesity, poor skin, diabetes and non-compliance. There were no adverse events
reported during the study
Doses of Aranesp® were adjusted to maintain Hb levels between 11 g/dl and 13 g/dl. IV
iron supplementation as per unit protocol was given to sustain serum ferritin levels >
150 µg/l and %Tsats > 20%.
The primary end-point was the change in Hb level, and the secondary end-point was the
change in dose of Aranesp® administered during the study.
A rise in Hb level from baseline to the end of the study period was shown. The starting
Hb level was 10.4 ± 1.64 g/dl which increased to 11.8 ± 1.58 g/dl. This gave a mean
change in Hb level of 1.4 ±1.63 g/dl which was found to be significantly different from
zero by performing a Student’s t-test (p < 0.001).
Mean starting dose of Aranesp® was 74.7 ± 35.60 _g/wk and the mean dose at the end
of the 6 months was 68.3 ± 35.63 _g/wk indicating a reduction in mean dose of 6.3 ±
25.26 _g/wk. The median change in dose was not found to be significantly different
from zero by carrying out a Wilcoxon signed rank test (p = 0.231).
The percentage of patients achieving the European Best Practice Guidelines (EBPG)
standard (Hb ≥ 11 g/dl) rose from 40.0% at baseline to 80.0% at the end of the study
In conclusion, this study reveals that haemodialysis patients receiving SC rHuEPO can
be effectively switched to once weekly doses of IV Aranesp® with no significant
change in dose level but the advantage of a significant increase in Hb level. In addition,
the number of patients achieving target Hb is significantly increased.
Renal Association October 2003
Moderated Poster Session - One
Group (B)
Cell and Molecular Biology
Tuesday 7 October
Studio Suite
Renal Association October 2003
P 10
Calcium oxalate adhesion and agglomeration is altered by CLC-5 expression in
mouse inner medullary collecting duct cells
JA Sayer1, G Carr2 and NL Simmons2
School of Clinical Medical Sciences, Medical School, University of Newcastle,
Newcastle upon Tyne, NE2 4HH, United Kingdom and 2School of Cell and Molecular
Biosciences, Medical School, University of Newcastle, Newcastle upon Tyne, NE2
4HH, United Kingdom
Dent’s disease, characterised by proteinuria, hypercalciuria, nephrocalcinosis and renal
stone formation, results from mutations affecting the chloride channel CLC-5. We have
previously demonstrated that mCLC-5 is expressed in inner medullary collecting duct
cells, localised to acidic endosomes. Here we report the consequence of antisense
mCLC-5 expression upon endocytosis of wheat germ agglutinin (WGA) and binding of
calcium oxalate crystals.
mIMCD-3 cells were either transiently transfected with GFP expression vector
(pcDNA3.1/GFP) as control or together with vector expressing sense mCLC-5 or
antisense mCLC-5. Positive transfectants were identified 24-48h post transfection, by
their GFP fluorescence, using a confocal microscope. Binding and endocytosis of
TRITC-wheat germ agglutinin (WGA)in these cells was followed over a 1h time course.
Optical sections of positive transfectants were collected to allow identification of
membrane bound and internalised WGA. After 1h, internalisation of WGA occurred in
81% of control transfectants and 76% of sense mCLC-5 transfectants. With antisense
mCLC-5 transfectants endocytosis was disrupted with the majority of cells showing
only membrane bound WGA; only 8% of cells showed WGA internalisation (p<0.001
antisense vs. control/sense).
Calcium oxalate monohydrate crystals were grown in a high calcium (50mM CaCl2)
medium exposed to diethyloxalate vapour. Crystals were harvested and added to
mIMCD-3 cultures for 30 minutes before imaging. For the majority of control or sense
transfectants (34/44 and 38/44 cells respectively), no crystal adhesion was observed.
The remaining cells (5/44, or 3/44) showed adhesion of single crystals (<10 _M size) or
agglomerates (5/44 and 3/44 >10 _M size). For antisense mCLC-5 transfectants the
majority of cells were associated with crystal agglomerates (33/50, p<0.01 antisense vs.
control/sense cells), whilst 1/50 showed adhesion of single crystals and 16/50 showed
no crystal adhesion.
Altered CLC-5 expression by using transient transfection of antisense mCLC-5 is
associated with disruption of endocytosis and adhesion of agglomerates of calcium
oxalate crystals on inner medullary collecting duct cells. Crystal retention and
agglomeration at the point of maximal urinary concentration are likely to be key factors
in renal stone formation.
Renal Association October 2003
P 11
Molecular diagnosis of a renal dysmorphology in a novel transgenic rat model
MM El-Kasti, T Wells and DA Carter
School of Biosciences, [Biomedical Sciences Buildings], Cardiff University, P.O.Box
911, Museum Avenue, CARDIFF, CF10 3US, United Kingdom
We have generated a novel transgenic rat line (presumed insertional-mutant) that
exhibits a pleiotropic phenotype associated with neonatal male lethality. Females exhibit
juvenile-onset growth retardation accompanied by skin abnormalities and kidney
dysmorphologies. Macroscopically, the female mutant kidney exhibits abnormal colour
and texture. Microscopically, it exhibits an alteration of the normal cortico-medullary
architecture: there is a decrease in the number of glomeruli (37.0 ± 3.0%), tubular
ectasia, tubular atrophy, fibrosis with mononuclear cell infiltration, and colloid-filled
tubules. These dysmorphologies are more pronounced in the mutant adult females
compared with the mutant neonatal males.
To identify gene(s) associated with this phenotype, DNA microarray analysis of kidney
gene expression in the neonatal mutant males was conducted, and a candidate
differentially expressed gene list of 30 ‘up-regulated’, and 17 ‘down-regulated’ genes
was produced. To date, northern blot analysis has confirmed down-regulation of two
distinct transcripts, kidney androgen-regulated protein (KAP) and Nedd4-WW domain
binding protein4 (N4WBP4) in both neonatal male, and adult female mutants. KAP is
known to be the second-most abundant kidney-specific transcript as determined by
serial analysis of gene expression. KAP expression in the epithelial cells of the proximal
tubules is regulated by steroid hormones. Androgens regulate expression in all
segments, whereas oestrogen and thyroid hormone control expression primarily in the
S3 segment. However, the functional role of KAP is unknown. N4WBP4 has been
identified recently as one of multiple novel N4WBP’s. The functional significance of
the interaction between the ubiquitin-protein ligase Nedd4 and this protein remains to be
established. However, since N4WBP4 is a putative membrane-associated/spanning
protein, it is believed that it is a strong candidate for Nedd4-mediated regulation.
Interestingly, there is evidence to suggest that the N4WBP4 transcript is also regulated
by androgens. Further investigation is underway in an attempt to reveal the disrupted
gene(s) causing the mutant phenotype. This novel transgenic rat may prove to be a
useful model of human kidney disease, in addition to providing a model for elucidating
KAP and N4WBP4 function.
Renal Association October 2003
P 12
Gremlin, a Bone Morphogenetic Protein antagonist, and the development of the
Xenopus pronephros.
V Dolan, P Alarcon, M Murphy, F Martin, HR Brady and C Hensey
Departments of Pharmacology and Medicine, Centre for Integrative Biology , The
Conway Institute for Biomedical and Biomolecular Research, University College
Dublin and The Dublin Molecular Medicine Centre, Dublin, Dublin 4, Irish Republic
Developmental regulators are known to contribute to disease processes and diabetic
nephropathy illustrates this paradigm in renal disease. Diabetic nephropathy associated
genes with important roles in development include TGF-beta, EGF, VEGF, CTGF and
gremlin (Murphy et al., J. Biol. Chem. 1999; 274(9):5830-4, Dolan et al., Pediatr
Nephrol 2003 Feb;18(2):75-84.).The finding of increased gremlin expression is
particularly noteworthy given its role as a putative antagonist of bone morphogenetic
proteins and regulator of cell turnover in non-renal development (e.g. mouse lung and
chick limb bud). Using the Xenopus embryo as a model for nephrogenesis we are
investigating the function of gremlin in pronephros development. Gremlin is expressed
in the developing pronephros and overexpression and knockdown experiments suggest a
function for this protein in pronephric development. Overexpression of gremlin by
mRNA injection was associated with increased size and complexity of the pronephros as
determined by wholemount immunohistochemistry using tubule and duct specific
antibodies (3G8/4A6). A similar phenotype was observed upon injection of murine
gremlin protein into the blastocoele of late blastula embryos. In addition to the observed
increase in size, development of ectopic pronephric structures was observed in some
embryos. This was observed in both mRNA and protein injected embryos and suggests
that in addition to impacting on the size of the pronephros gremlin may also have the
capacity to direct cells to form pronephric structures. Gremlin depletion was achieved
by injection of a morpholino antisense oligonucleotide designed to inhibit gremlin
translation and this resulted in loss of kidney structures in a dose dependant fashion. In
aggregate these data suggest a key role for gremlin in renal development in this system
and add to the emerging paradigm whereby important developmental genes re-emerge
in the context of disease.
Renal Association October 2003
P 13
Ca2+-activated cation conductances in control (M8) and PKD1 transgenic (M7)
mouse collecting duct cell lines
C Haigh1, SK Laycock1, T Brier1, ACM Ong2 and L Robson1
Department of Biomedical Science, Alfred Denny Building, Western Bank, University
of Sheffield, Sheffield, S10 2TN, United Kingdom and 2Sheffield Kidney Institute
Division of Clinical Sciences (North), University of Sheffield, Sheffield, S5 7AU,
United Kingdom
Autosomal dominant polycystic kidney disease is caused by mutations in the genes that
code for two proteins, polycystin-1 and polycystin-2. Previous studies have suggested
that these proteins form a non-selective conductance that is important in regulating
cellular Ca2+ permeability. The aim of the following study was to determine whether
there were differences in the non-selective cation conductances of a conditionally
immortalised collecting duct cell line transgenic for PKD1 (M7) and a control collecting
duct cell line (M8).
Whole cell currents were examined in single M7 and M8 cells. Clamp potential was
stepped to between +100 and –100 mV in –20 mV steps. The bath contained NaCl,
while the pipette contained CsCl with either 1_M free Ca2+ or no added Ca2+. Whole
cell currents were recorded with a non-flowing bath on achieving the whole cell
configuration and then at steady state. Bath flow was then started and cells were
exposed to 100 _M gadolinium (Gd3+), an inhibitor of cation channels. The cation to
anion selectivity of the Gd3+-sensitive current was examined by decreasing bath CsCl
five-fold (osmolality maintained with mannitol).
The initial current in the M7 cells at +100 mV with 1_M Ca2+was 169.7 ± 52.7 pA/pF
(n=15). This increased to 743.8 ± 172.9 pA/pF over time. A similar response was
observed in the M8 cells, 109.8 ± 51.4 pA/pF versus 807.6 ± 192.8 pA/pF (n=18). On
starting the bath flow, current increased in both the M7 (n=15) and M8 (n=13) cells, and
there was no significant difference between the responses. Currents in both cells were
also reversibly inhibited by addition of Gd3+ to the bath (M7 n=8 and M8 n=10).
Dropping bath CsCl shifted the reversal potential (Vrev) of the Gd3+-sensitive current in
the M7 cells by –7.45 ± 1.86 mV. This was not significantly different to the shift
observed in the M8 cells, –7.36 ± 2.05 mV (n=8). In the absence of pipette Ca2+ the
Gd3+ sensitive current was absent.
These data show that both control and polycystin-1 over-expressing lines possess a
Ca2+-activated, flow sensitive conductance that is selective for cations over anions. The
lack of a difference between the cell types suggests that over-expression of polycystin-1
alone does not alter the cation conductance in this collecting duct cell line.
This work was supported by the National Kidney Research Fund (UK).
Renal Association October 2003
P 14
The d2 Subunit of the H+-ATPase has a Limited Tissue Distribution and Forms
Part of the Apical Proton Pump in Intercalated Cells of the Distal Nephron.
A N Smith1, R S Al-Lamki1, K J Borthwick1, C A Wagner2 and F E Karet1
Dept of Medical Genetics/Nephrology, University of Cambridge, Box 139,
Addenbrooke's Hospital, Cambridge, CB2 2XY, United Kingdom and 2Institute of
Physiology, University of Zurich, Winterthurerstrasse 190, Zurich, 8057, Switzerland
Vacuolar-type H+-ATPases are multi-subunit proton pumps essential for the acidification
of intracellular compartments in eukaryotic cells. Specialized versions are also present
at the plasma membrane of certain cell types, for example the renal intercalated cell (IC)
where they are critically involved in urinary acidification. H+-ATPases consist of a V1
domain catalysing ATP hydrolysis and a V0 domain responsible for H+ translocation.
The precise function of many of the pump's subunits, and the interactions between them,
are not yet clear. The B, C, E, G, a, d and e subunits also have multiple isoforms
encoded by different genes with differing expression patterns, and it is likely that these
play an important role in the cellular localization and activity of H+-ATPases. We have
previously shown that in man, the d2 subunit isoform is expressed predominantly in
kidney, bone and lung, whereas d1 is ubiquitously expressed. Here we report an
investigation of the distribution of d2 in the kidney.
We first confirmed that murine d2 has a similar tissue distribution to that seen in man.
Mouse nephron segment-specific RT-PCR then showed that d2 is chiefly expressed in
the collecting duct. To determine whether d2 forms part of the IC apical proton pump
that contains the mainly kidney-expressed a4 and B1 subunits (defects of which cause
recessive distal renal tubular acidosis), we have raised a novel polyclonal antibody
(SK20) against human d2. SK20 recognizes a protein of appropriate size in membrane
extracts from human, rat and mouse kidney, and does not cross react with the d1
isoform. Immunolocalization in unfixed frozen human kidney cortex revealed high
intensity d2 staining at the apical surface of ICs, which colocalizes with a4 staining, but
not with AQP2, a principal cell marker.
These data demonstrate that the IC apical proton pump contains a third tissue-restricted
subunit isoform, and support the hypothesis that subunit differences play a key role in
H+-ATPase localization and function. SK20 also represents a useful new tool for
studying d2 expression in other tissues, notably bone and lung.
Renal Association October 2003
P 15
Identification of Proteins that Interact with WSC Domain of polycystin 1
H S Sonbol, S W Knight and R G Price
King's College, Biochemistry, Department of Life Sciences, 150 Stamford Street,
London, SE1 9NN, United Kingdom
Autosomal Polycystic Kidney Disease (ADPKD) is characterised by the aberrant
proliferation of tubular epithelia cells, the mis-localisation of proteins in the polar
epithelia cells and changes in the extracellular matrix (ECM). ADPKD is caused by
mutation in either the PKD1 or PKD2 gene resulting in dysfunction of either polycystin
1 or polycystin 2. Polycystin 1 is a 460kDa transmembrane protein that contains a large
extracellular region. Polycystin 1 has many features of non-canonical G-protein coupled
receptors. A series of different domains have been identified within the extracellular
region, many of which are predicted to interact with other proteins including
components of the ECM. The purpose of this study was to identify the proteins that
interact with the WSC domain (residues 177 to 240) of polycystin 1. WSC domains are
shared with the cell wall integrity proteins of Saccharomyces cerevisiae (WSC1-3) and
the kremen family of receptor proteins. The WSC domain of Polycystin 1 was amplified
from cDNA and cloned into pGEX2T expression vector. The WSC domain was
expressed and purified as a Glutathione-S-transferase (GST) fusion protein. The WSC
fusion protein was used in GST pulldown experiments using HEK293 cell lysate. Of
those proteins shown to specifically interact with WSC one was identified by Western
Blotting to be Laminin. The other proteins are currently being analysed by MALDITOF mass spectrometry and Western blotting. This data indicates that the WSC domain
of polycystin 1 plays an important role in ECM cell interactions.
Renal Association October 2003
P 16
Low Ouabain Concentrations Stimulate Calpain Activity in Multiple Cell Lines
SM Harwood, DA Allen, MJ Raftery and MM Yaqoob
Experimental Medicine and Nephrology, Queen Mary, University of London, Renal
Research Laboratories, Suite 22, 59 Dominion House, Bartholomew Close, West
Smithfield, London, EC1A 7BE, United Kingdom
Endogenous ouabain is elevated in uraemia but currently little is understood of its action
at pathophysiological concentrations (~nmol/L), as supraphysiological amounts are
required (_mol/L) to inhibit NaK-ATPase function in vitro. We previously showed that
l0 nmol/L of ouabain stimulates calpain activity in human derived myoblasts (Girardi),
and that calcium influx was essential for this process. Here an in situ assay was used to
determine the calpain activity in four different cell lines treated with 5 and 10 nmol/L
ouabain. The calpain substrate Suc-LLVY-AMC (5 _mol/L) was added to 6 well plates
containing monolayers of cells, half of which wells were pre-treated with the selective
calpain inhibitor calpeptin (45 _mol/L) or vehicle (DMSO). Blanks (no cells) were
prepared for each experiment to ensure that only cellular calpain activity was calculated.
After 3 h incubation at 370C fluorescence was measured (380 nm ex, 460 nm em) and
the results are shown in the table below.
We also studied the effect of inhibiting different calcium channels to see if we could
determine the route by which calcium enters the cells to effect calpain activation. The Ltype channel inhibitor nimodopine (0. 1 _mol/L) and the T-channel inhibitor flunarizine
(15 _mol/L) were both able to abolish the rise in calpain activity in Girardi cells due to
ouabain (rise in calpain activity 14% and 0% that of those untreated with inhibitor),
whilst the P-channel inhibitor _-agatoxin lVa (0. 1 _mol/L) was unable to inhibit
ouabain induced calpain activation. With all inhibitors baseline calpain activity was
unaffected. We conclude that the pathophysiological concentrations of ouabain found in
uraemia might stimulate calpain universally and that calcium entry via L and T-type
channels are possible mechanisms by which calpain activation occurs.
Renal Association October 2003
P 17
S Steddon, S Harwood, J Burrin, J Cunningham, MJ Raftery and MM Yaqoob
Renal Research Laboratories, Suite 22, 59 Dominion House, Bartholomew Close, West
Smithfield, London, EC1A 7BE, United Kingdom
Osteoblast longevity is an important determinant of bone formation and skeletal mass.
IGF-I in the bone microenvironment may act as a survival factor, though the mechanism
of its anti-apoptotic effect is poorly understood. Renal osteodystrophy is associated with
a lack of IGF-I in bone and local resistance to its action, possibly through a post
receptor signalling defect. The SOCS proteins have been shown to undertake a
regulatory role in both pro- and anti-apoptotic pathways in other systems and could
potentially be operative in bone. The binding of IGF-I to its receptor activates several
intracellular messengers, including the JAK/STAT system - a target for SOCS. In the
present studies MG-63 human osteoblastic cells were incubated in the presence or
absence of IGF-I. In situ assays were performed utilising DEVD-AMC and DEVDCHO as substrate and inhibitor for caspase-3 and LEHD-AFC and LEHD-CHO for
caspase-9. At 24 h caspase activity was measured fluorometrically and cell lysates
assessed for DNA fragmentation by ELISA. IGF-I (10 ng/mL) reduced the activity of
caspase-3 compared with untreated cells (n=6) (p<0.002). A dose dependent reduction
in caspase-3 activity was observed with escalating concentrations (0, 2.5, 5, and 10
ng/mL) of IGF-I (n=6 for each concentration). Caspase-9 activity was also significantly
reduced (p<0.005) following IGF-I exposure (n=6). IGF-I treatment consistently
attenuated DNA fragmentation (n=18). In separate experiments, IGF-I significantly
induced the expression of SOCS1 (p<0.05) and CIS (p<0.05), but not SOCS2 or 3 (all
n=3), as determined by real-time quantitative PCR. All PCRs were performed in
triplicate. The pattern of SOCS expression was compatible with their emerging negative
feedback function. These studies confirm the anti-apoptotic effect of IGF-I in human
osteoblastic cells and imply mediation through mitochondrial pathways involving
caspase-9. The induction of SOCS suggests that JAK/STAT activation is relevant to
IGF-I signalling in osteoblasts and hints at a potential role for SOCS1 and CIS in the
regulation of the anti-apoptotic effect of this important osteotropic growth factor.
Renal Association October 2003
P 18
S Steddon, S Harwood, J Burrin, J Cunningham, MJ Raftery and MM Yaqoob
Renal Research Laboratories, Suite 22, 59 Dominion House, Bartholomew Close, West
Smithfield, London, EC1A 7BE, United Kingdom
Although osteoblast apoptosis is an essential component of bone turnover, the effectors
responsible for its initiation and regulation are not well understood. Disruption of local
survival signals may be relevant to the disordered remodelling of many metabolic bone
diseases. Adynamic uraemic bone exhibits a relative lack of IGF-I, with possible
consequences for cell longevity. This study aimed to assess the relative roles of calpain
and caspase-3, known mediators of cell death, in the survival of osteoblasts and evaluate
their involvement in the protective effect of IGF-I. To this end, MG-63 human
osteoblastic cells were incubated in the presence or absence of IGF-I for 24 h. In situ
assays were performed utilising DEVD-AMC and DEVD-CHO as substrate and
inhibitor for caspase-3 and Suc-LLVY-AMC and calpeptin for calpain. The activity of
the relevant protease was measured fluorometrically after 24 h. Cell lysates were
assessed for DNA fragmentation by ELISA. IGF-I (10 ng/mL) reduced the activity of
caspase-3 compared with untreated cells (n=6) (p<0.002). A dose dependent reduction
in caspase-3 activity was observed with escalating concentrations (0, 2.5, 5, and 10
ng/mL) of IGF-I (n=6). Calpain activity increased significantly after 24 hours when
incubated with 5ng/mL IGF-I (n=12, p<0.01), whilst 10ng/mL caused an increase, but to
a lesser extent. A consistent reduction in DNA fragmentation was seen with both
5ng/mL and 10ng/mL IGF-I (n=12, p<0.01). To assess necrosis, supernatant LDH was
measured. No differences were seen between IGF-I treated and untreated cells. There is
emerging evidence that calpain plays a role in the proliferation and differentiation of
osteoblastic cells and it is possible that this activity dominates the response to IGF-I.
Alternatively, there is evidence that calpain and caspase cascades interact. Since
caspase-3 and calpain share many substrates it is possible that the suppression of
caspase-3 by IGF-I is compensated by an increase in calpain activity. This would allow
the essential processing of cytoplasmic proteins to continue without altering the antiapoptotic influence of IGF-I.
Renal Association October 2003
P 19
Polycystin-1 Targeting to the Lateral Membrane of Renal Tubular Epithelial Cells
is Mediated by a Novel Motif in the Cytoplasmic Tail.
J Poon1, C Boucher1, M Devonald1, J Luzio2 and R Sandford1
Department of Medical Genetics, University of Cambridge, CIMR, Addenbrookes
Hospital, Cambridge, CB2 2XY, United Kingdom and 2Department of Clinical ,
Biochemistry, CIMR, Addenbrookes Hospital, Cambridge, CB2 2XY, United Kingdom
Mutations in PKD1 cause 85% of cases of autosomal dominant polycystic kidney
disease (ADPKD). Polycystin-1, the protein product of the PKD1 gene, is localised to
adhesion complexes, including the desmosome, and to the renal primary cilium.
Polycystin-2 is also localised to desmosomes and the primary cilum but a major site of
expression is the ER. Signals in the C-terminus of polycystin-2 regulate its localisation
to the ER, but the signals that target polycystin-1 to its sites of expression are unknown.
In particular, signals that target polycystin-1 to the lateral membrane of polarised
epithelial cells have not been defined. The ectodomain of CD44 contains no basolateral
targeting signals and is expressed on the apical membrane of polarised epithelial cells. A
stably transfected MDCK Flp-In cell line expressing a fusion protein consisting of the
ectodomain of CD44 and the intracellular C-terminus of polycystin-1 was generated by
homologous recombination. The CD44-polycystin-1 chimera (CP1) was expressed at
the lateral membrane with some points of colocalisation with endogenous polycystin-1;
no cilial localisation was observed. As the C-terminus of polycystin-1 contains no
classical tyrosine or dileucine targeting motifs, deletion and mutagenesis studies of CP1
were carried out. A novel phosphorylated lateral targeting motif was identified Cterminal to a coiled-coil domain. Constructs containing point mutations in this motif
were localised to the ER suggesting this motif is important in directing ER export and
protein localisation. Additional infomation is therefore required for normal complex
assembly and localisation to the primary cilium which may be contained in any of the
other polycystin domains or cytoplasmic loops.
Using a novel method to rapidly generate stable cell lines we have shown that the
cytoplasmic tail of polycystin-1 contains a novel ER export motif that is required for
normal polarised expression to the lateral membrane. This is distinct from other
functional domains such as the G protein activation site and the coiled-coil domain that
mediates the interaction with polycystin-2. As no mutations 3' to the region encoding
the targeting motif have been identified protein mislocalisation may play an important
role in the pathogenesis of ADPKD.
Renal Association October 2003
P 20
Inhibition of sodium reabsorption in rat collecting ducts by luminal nucleotides:
evidence against P2Y2 receptor mediation
D.G. Shirley and R.J. Unwin
Centre for Nephrology and Department of Physiology, Royal Free & University College
Medical School, London, NW3 2PF, United Kingdom
There is increasing evidence for P2 receptor expression in apical, as well as basolateral,
membranes of collecting duct cells, and we have shown that luminal application of a
relatively unspecific P2 receptor agonist, ATP_S, can inhibit collecting duct sodium
reabsorption in vivo (Shirley et al. 2001, JASN 12: A3001). In vitro studies using mouse
collecting duct have provided pharmacological evidence that this effect results from
inhibition of amiloride-sensitive sodium channels (ENaC) and is mediated by the P2Y2
receptor sub-type (Lehrmann et al. 2002, JASN 13: 10-18). The present study has tested
the effects of two P2Y2 agonists in our in vivo rat model.
Sprague-Dawley rats were maintained on a low-sodium diet (Na content 13 mmol kg-1
dry weight) for one week (in order to stimulate ENaC activity), then anaesthetised
(sodium thiopentone, 100 mg kg-1, I.P.) and prepared surgically for micropuncture of the
left kidney. A brisk urine flow rate was maintained by I.V. infusion of 2 % glucose
solution. Late distal tubules were perfused orthogradely at 3 nl min-1 with artificial
tubular fluid containing [14C]inulin and 22Na. Each tubule was perfused twice, first with
a control perfusate, then with one containing either P1, P4-di [adenosine-5’]
tetraphosphate (Ap4A; Sigma; 1mmol l-1) or P1-[cytidine-5’]P4-[uridine-5’]
tetraphosphate (Cp4U; Inspire Pharmaceuticals; 1 mmol l-1). Only those perfusions for
which the urinary recovery of [14C]inulin exceeded 85% of the amount perfused were
accepted; 22Na recovery was then calculated as a percentage of [14C]inulin recovery
(Na/inulin recovery ratio).
For Ap4A, the Na/inulin recovery ratio (mean ± S.E.M.) was 13 ± 3 % with the control
perfusate vs.14 ± 3 % with the agonist (n = 16 pairs; NS, paired t test). Corresponding
values for Cp4U were 14 ± 3 % vs.13 ± 4 % (n = 14 pairs; NS). When the control
perfusate was used for both the first and second perfusions (time controls), there was
also no consistent change in the Na/inulin recovery ratio (13 ± 3 % vs.14 ± 4 %; n = 15
pairs; NS).
The lack of effect of Ap4A or Cp4U, whose P2Y activity is restricted to P2Y2 and P2Y4
receptor sub-types, in a preparation previously shown to be responsive to ATP_S, argues
against apical P2Y2 receptor mediation in nucleotide-induced inhibition of sodium
reabsorption in the rat collecting duct in vivo.
Renal Association October 2003
Moderated Poster Session - One
Group (C)
Haemo-and Peritoneal Dialysis
Tuesday 7 October
Garner Room
Renal Association October 2003
P 21
Improvement of the Haemodynamic Derangement in Hepatorenal Syndrome with
Albumin Dialysis
S H Lambie1, J G Freeman2, T Bennett3 and C W McIntyre1
Renal Unit, Derby City General Hospital, Uttoxeter Rd, Derby, DE22 3NE, United
Kingdom, 2Department of Gastroenterology, Derby City General Hospital, Uttoxeter
Rd, Derby, DE22 3NE, United Kingdom and 3Centre for Integrated Systems Biology
and Medicine, Queens Medical Centre, Nottingham, NG1 6RA, United Kingdom
The hemodynamic changes in hepatorenal syndrome are dominated by arteriolar
dilatation, resulting in low peripheral resistance (TPR), high cardiac output(C.O.) and
systemic hypotension.
Therapy for hepatorenal syndrome using the Molecular Adsorbents Recirculating
System (MARS) is becoming increasingly well recognised. In this system blood is
circulated through a hollow fibre dialyser that allows water-soluble solutes to diffuse
across the membrane, while lipophilic and protein bound solutes are adsorbed onto it. A
washing solution of 20% albumin provides the driving force for the removal of ‘hepatic
toxins’. The albumin is then regenerated on-line using two adsorption columns and a
standard kidney dialyser.
We used the Finometer to study the hemodynamic response to MARS therapy. The
Finometer detects changes in digital arterial diameter by means of an infrared
photoplethysmograph. These changes are opposed by a pressure servo controller that
alters pressure in an inflatable bladder mounted in a finger cuff. Pulse wave analysis of
the resultant arterial waveform allows calculation of a wide variety of hemodynamic
In three MARS sessions C.O. fell to a nadir of 78% of the initial value, while TPR rose
to maximum of 146 % of the initial value. By the end of the first 30 minutes of therapy
C.O. had fallen by 12% with the remainder of the fall in C.O. occurring more steadily
over the subsequent 450 minutes of therapy. Systolic and diastolic BP improved slightly
with therapy (by 8% and 13% respectively).
In conclusion MARS therapy for acute hepatorenal syndrome is associated with a
significant and sustained improvement in the systemic hemodynamic disturbance
associated with this condition, and the Finometer is a useful and practical tool with
which to study these changes.
Renal Association October 2003
P 22
Bi-modal dialysis (BMD): an integrated approach to renal replacement therapy
CW McIntyre
Department of Renal Medicine, Derby City General Hospital, Uttoxeter Road, Derby,
DE22 3NE, United Kingdom
PD and HD are both widely used as sole therapies for ESRD. There is still controversy
over which (if either) is superior in terms of patient outcomes. PD offers the advantages
of long slow continuous ultrafiltration and potentially enhanced protection of residual
renal function (RRF). In contrast, HD offers superior solute removal at the cost of
undesirable cardiovascular tolerance of high rates of sodium and water removal. The
aim of this study was to investigate the clinical feasibility of offering a combined
treatment of both modalities (bimodal dialysis, BMD) to incident patients reaching
We prospectively studied eight patients. An arterio-venous fistula was formed and
peritoneal catheter were inserted. BMD consisted of two, three hour, high efficiency
euvolaemic HD sessions per week, in combination with two PD exchanges per day to
provide a degree of solute clearance, and all of the ultrafiltration. Adequacy was
measured both independently for each modality and as an equivalent urea clearance
(EKR). Patients were followed using the standard range of evaluations, in addition to Cr
EDTA GFR, echocardiography, treatment outcomes and complications.
Mean time on BMD was 346±74.9 (245-431) days. Peritonitis rate was 21 months per
episode (mean 0.6±0.9 0-2 episodes per patient). Mean peritoneal ultrafiltration volume
was 1.58±0.32 (1.3-2.1) l/per day. Delivered EKR did not change statistically over the
treatment period (2.3±0.13 (2.2-2.5) at 12 months). RRF was maintained over the study
period. LVMI reduced over the treatment period from a mean of 194±31.2 (161-265) to
156±21.2 (138-189) g/m2 (p=0.05). Ventricular performance remained unchanged over
the study (ejection fraction 50.4± 11.1 (38-67) to 48±8.0 (48-67) %). Mean time during
BMD spent on HD alone was 4.2±6.9 (0-16) days and on PD alone 9.2±10.6 (0-25)
This study suggests that BMD is a feasible treatment for ESRD. It is associated with
adequate solute removal, good haemodynamic/volume control and allows increased
treatment flexibility for coping with complications normally requiring recourse to
unplanned HD with temporary central venous access.
Renal Association October 2003
P 23
Kt/V Underestimates Hemodialysis Dose In Women And Patients With Small Body
EM Spalding, S Chandna and K Farrington
Renal Unit, Lister Hospital, Coreys Mill Lane, Stevenage, SG1 4AB, United Kingdom
Background: Kt/V is the standard index of hemodialysis adequacy, though body size
(reflected in V) independently determines dialysis outcome. We wished to examine the
use of alternative normalising factors more reflective of metabolic size in the dialysis
prescription e.g. body surface area (BSA) and weight0.67 (W0.67).
Methods: A retrospective study of 655 hemodialysis patients in a single centre was
conducted. Dialysis dose was prescribed with a target 2-pool Kt/V of 1.0-1.2. The
effects on mortality of age, co-morbidity, Watson volume, delivered Kt/V, delivered
Kt/BSA, delivered Kt/W0.67 and gender were studied.
Results: In Kaplan-Meier analysis, patients with a Kt/V greater than the group mean
fared no better than those with lower mean Kt/V. However, a Kt/BSA or Kt/W0.67
greater than the group mean conferred a survival advantage (p <0.0001). Age, comorbidity, Kt/BSA and V were independent predictors of survival. The relationship
between Kt/V and Kt/BSA was different in small and large patients, and in males and
females. A Kt/V of 1.14 in a large male, 1.2 in a small male and 1.37 in a female all
delivered the same dialysis dose with respect to Kt/BSA. (figure 1)
Conclusions: The use of V in dialysis prescription results in under-dialysis of small
patients, particularly females in whom the relationship between V and BSA is different.
The use of a normalising factor representative of metabolic size such as BSA or W0.67 to
prescribe dialysis deserves further investigation.
Key words: dialysis; mortality; metabolism; body surface area; body water; survival
Renal Association October 2003
P 24
Survival after reinsertion of Peritoneal Dialysis (PD) catheter following severe
peritonitis: a case controlled study.
SB Walsh, S Cox, S Fan and MM Yaqoob
Renal Unit, Bart's and The London NHS Trust, Whitechapel, London, E1 1BB, United
Published data suggests technique survival 3 months after severe peritonitis (PDP) is
30.8% (JASN 2002) after reinsertion in all patients. We retrospectively analysed the
outcome of peritonitis in our cohort of 450 patients (Jan 2000 to Dec 2001) to identify
predictors of both technique survival and mortality. There were 457 PDP episodes with
106 PD catheter removals due to non resolution despite 72 hours of antibiotic therapy.
Four week mortality was 2% vs 12% in resolving vs non-resolving (p<0.01). Of the
remaining 93 non resolving episodes, Group 1 (n=42) underwent catheter reinsertion,
Group 2 (n=16) elected to stay on haemodialysis (HD) despite being judged suitable to
return to PD and Group 3 (n=35) were unsuitable to return to PD due to membrane
failure (n=5), loculated collections (n=21) and inability to manage exchanges at home
There was no significant difference in age, gender, diabetic status, timing of catheter
removal or type of organism. Dialysis vintage was significantly greater in Group 3 than
Groups 1 and 2 (p<0.03). Mortality at follow up (mean 22 months) was significantly
greater in Group 3 (43%) than Group 2 (0%) (p<0.01). There was marginal statistical
significance in the difference between group 1 (27%) and 2 (p=0.057), and no difference
between groups 1 and 3.
In Group 1 at the end of follow up, 23 patients were either still on PD (18), transplanted
whilst on PD (3) or died on PD (2), i.e technique survival was 55% (vs 69% at 3
months). In contrast 19 of the original Group 1 cohort were transferred to HD, 8 of these
are still on HD at the end of follow up, 2 were transplanted and 9 died whilst on HD.
There were no independent predictors of successful outcome for Group 1 patients.
Only dialysis vintage predicted failure to return to PD after severe peritonitis. Despite
our clinical criteria for reinsertion of PD catheter there was a trend to improved survival
if patients remained on HD (Group 1 vs Group 2). PD technique survival was good
(69% at 3 months) if patients did return to PD if preselection is adopted prior to
reinsertion. This study was unable to identify any other predictors that might improve
successful return to PD after severe peritonitis.
Renal Association October 2003
P 25
Factors affecting dialysate MCP-1, a marker for peritoneal inflammation, in
patients on PD
A Malik, F Tam, M Henriksson, J Levy and E Brown
Imperial College London, Charing Cross Hospital, Fulham Palace Road, London, W6
8RF, United Kingdom
Increasing membrane permeability (D/P), peritoneal thickening and development of a
hyalinizing vasculopathy have been demonstrated with increased time on PD and have
been shown to correlate with systemic and peritoneal markers of inflammation (plasma
CRP, plasma and dialysate IL6). Monocyte chemoattractant protein 1 (MCP-1) is a
chemotactic cytokine that is important in chronic leukocyte recruitment. MCP-1 levels
have not previously been measured in PD patients, but in-vitro and ex-vivo studies have
shown that high dialysate glucose concentration up-regulates MCP-1 synthesis in a time
and dose-dependent manner. We present here the results of a cross-sectional study
assessing peritoneal MCP-1 production in PD patients.
51 stable PD patients (22 CAPD, 29 APD) were studied. Mean age was 57.2 (20-82)
years, 27% were diabetic, mean PD duration was 29 (1-114) months. No patient used
3.86% dextrose and most a combination of 1.36% and 2.27% dextrose. 41% of patients
used icodextrin. Mean dialysate volume was 10.33±0.62 l/day and mean glucose
concentration was 90.0±2.0 mmol/day. Membrane permeability (D/P) was measured by
standardised peritoneal equilibration test (PET). Dialysate effluent at the end of a 4 hour
PET was stored at -20°C for MCP-1 measurement by ELISA. Mean MCP-1
concentration was 239.9±24.5 pg/ml. MCP-1 levels were significantly greater with
higher D/P (r=0.43, p<0.001), longer duration of PD (r=0.31, p<0.02), and total
dialysate volume (r=0.43, p<0.001). In addition, MCP-1 levels were higher in patients
with a previous episode of peritonitis (n=21, p<0.001). There was no relationship
between MCP-1 levels and age, mean dialysate glucose concentration, plasma CRP or
serum albumin. 10 patients were using bicarbonate/lactate dialysate (Physioneal®);
MCP-1 levels were not different compared to 10 patients using conventional dialysate
and matched for duration of PD.
This study suggests that the changes in peritoneal membrane structure and function with
time on PD are related to local peritoneal inflammation. The relationship to total
dialysate volume can be partly explained by the increase in dialysis needed as residual
renal function declines with time on PD. There was no effect of dialysate glucose
concentration on peritoneal inflammation, at least within the limited range used by this
group of PD patients.
Renal Association October 2003
P 26
Effect of high flux dialysis on B-type natriuretic peptide levels
M Suresh and K Farrington
Renal Unit, Lister Hospital, Coreys Mill Lane, Stevenage, SG1 4AB, United Kingdom
The effects of high flux dialysis on B-type natriuretic ]peptide (BNP) levels have not
been previously studied. Twelve patients were studied, 6 on high flux haemodialysis and
6 on haemodiafiltration. None had clinical evidence of heart disease. All had a clinically
defined dry weight. The mean duration of dialysis was 180± 25 minutes. All used
polysulfone dialysers. BNP was measured at the start, end and 15 minutes after the end
of dialysis. In addition levels were measured at intervals during dialysis after a first
phase of dialysis with no ultrafiltration and successively after 2 subsequent phases of
ultrafiltration and dialysis. The carboxy-terminal BNP was measured by an
immunoassay using a portable kit.(Triage meter, Biosite Incorporated, San Diego,
CA).The assay had a detection range between 5-5000 pg/ml with the serum
concentration in the normal population being less than 100 pg/ml. Initial BNP levels
ranged between 60.7 and 2520 pg/ml. Eleven out of twelve patients had elevated BNP
levels at the start of dialysis. The BNP levels fell significantly through dialysis from
500.89± 691.16 pg/ml at the start to 274.95 ± 389.77 pg/ml at the end of dialysis
(p<0.01). There was no significant change in levels between dialysis end and 15
minutes after. After the first 30 minutes of dialysis with no ultrafiltration there was a
significant fall in BNP from 500.89 ± 691.16 pg/ml to 253.45 ± 377 pg/ml (p<0.011).
There were no significant changes during the subsequent ultrafiltration phases. All
patients remained haemodynamically stable throughout. BNP levels vary dramatically
in haemodialysis patients with no overt cardiac disease even when they are at their
clinically defined dry weight. BNP levels fall significantly during high flux dialysis and
haemodiafiltration. As a consequence it may be difficult to interpret changes in BNP
levels related to changes in the volume state.
Renal Association October 2003
P 27
Calcium Flux During Haemodialysis
M Sigrist and CW McIntyre
Department of Renal Medicine, Derby City General Hospital, Uttoxeter Road, Derby,
DE22 3NE, United Kingdom
There is an increasing appreciation of the risks associated with excessive calcium
exposure for chronic HD patients. Forthcoming K-DOQI guidelines recommend a
maximum dialysate calcium concentration of 1.25 mmol/l, to avoid intra-dialytic
calcium loading. The aim of this study was to investigate calcium flux during dialysis
against a fixed dialysate calcium concentration of 1.25 mmol/l.
We studied 30 patients. All patients underwent HD using Hospal Integra dialysis
monitors, bicarbonate buffering and dialysate sodium and calcium concentrations of 140
mmol/l and 1.25 mmol/l respectively. Blood was sampled pre and post HD for total
calcium, albumin, bicarbonate and phosphate measurement, in addition to ionised
calcium levels measured by the patient using a portable analyser. Calcium flux was
determined from measurements of ionised calcium levels in dialyser inlet samples and
those in continuous partial waste dialysis collection, with reference to total waste
dialysate and ultrafiltration volumes.
There was marked variability of delivered dialyser inlet calcium concentration between
dialysis monitors (range 1.24-1.36mmol/l). Although calcium concentration delivered
during each session was constant with a mean coefficient of variance of 2.2±1.2% when
inlet dialyser calcium concentration was measured at 15 minute intervals. All but one of
the patients experienced calcium removal during HD against a dialysate calcium
concentration of 1.25 mmol/l. Mean calcium flux was 245±28 (range –9.6-667) mg.
Calcium flux and phosphate removal were correlated with pre-dialysis plasma
concentrations (R2=0.25, p=0.006 and R2=0.31, p+0.0001 for calcium and phosphate
In conclusion these data suggest that dialysate concentration should be prescribed with
reference to plasma calcium levels, to avoid excessive removal or loading. Such an
individualised approach may be increasingly important given the propensity for calcium
loading that lower serum calcium concentrations resulting from novel therapeutic
strategies utilising non-calcium containing phosphate binders and calcimimetic agents
are likely to result in.
Renal Association October 2003
P 28
Statins have anti-inflammatory effects in continuous ambulatory peritoneal
dialysis (CAPD) patients: A case control study.
S Kumar, R Thuraisingham, M Raftery, S Fan and M Yaqoob
Renal office, Royal London Hospital, Barts and The London NHS Trust, Whitechapel,
London., London, E1 1BB, United Kingdom
Elevated plasma concentrations of C-reactive protein (CRP) is an independent risk
factor for cardiovascular (CV) disease and predictor of mortality in the general
population and in uremic patients. Even single point CRP levels are strongly predictive
of excessive CV mortality in CAPD patients. 3-hydroxy-3-methylglutaryl coenzyme A
reductase inhibitors (statins), can reduce CRP levels through mechanisms independent
of their effects on lipid levels in general population. Few available data on
haemodialysis patients have failed to demonstrate this effect. However, to our
knowledge no such data is available in CAPD patients.
We evaluated a total of 257 patients on our CAPD programme. Group 1 comprised 137
patients who are taking statins (atorvastatin, simvastatin and pravastatin) for a mean
period of 12 months. Group 2 comprised 120 patients who are not taking any
cholesterol lowering drugs. Two groups were similar in age, gender distribution,
duration of dialysis, dialysis regimes, and racial distribution. However, there were more
diabetic patients in group 1 (n=52) as compared to group 2 (n=25), (p< 0.01).
Cumulative cardiovascular co-morbidity burden was also significantly higher in group 1
compared to group 2 (p < 0.01).
Haemoglobin and serum albumin levels were similar in two groups. As expected group
1 had significantly lower total cholesterol levels as compared to group 2 (4.74+ 1.05 vs
5.02+1.17; p=0.04) and interestingly CRP levels were also significantly lower in group
1 compared to group 2 (14.2+13.8 vs 19.4+19.5; p=0.02) despite increased CV comorbidity and higher number of diabetic patients. The finding of this observational case
control study confirms the anti-inflammatory properties of statins in CAPD independent
of cholesterol lowering effects like general population. These data have implications for
considering statin therapy in CAPD for their anti-inflammatory properties.
Renal Association October 2003
P 29
Peripheral blood mononuclear cell telomere length is unaltered in haemodialysis
despite increased oxidative stress
M C Boxall, T von Zglinicki and T H J Goodship
School of Clinical Medical Sciences, University of Newcastle upon Tyne, Newcastle
upon Tyne, NE1 4LP, United Kingdom
Increase oxidative stress is a well described feature of haemodialysis (HD). This is
secondary to both an increase in free radical generation and impaired antioxidant
mechanisms. Oxidative stress is also known to cause telomere shortening. Telomeres in
somatic human cells shorten with each cell division and it is well recognised that
telomere length in peripheral blood mononuclear cells (PBMCs) decreases with age. In
this study we tested the hypothesis that PBMC telomere length would be shortened in
haemodialysis patients with increased oxidative stress. 20 control subjects, 21 nondiabetic and 18 diabetic HD patients were studied. They were of a similar age and sex
distribution. Serum malondialdehdye concentration (a marker of oxidative stress) was
significantly higher in the HD patients (control, all < 0.5; non-diabetic HD, 2.3 ± 0.3;
diabetic HD, 2.0 ± 0.3 _mol/l). However, PBMC telomere length was not substantially
different in the three group (control, 8283 ± 179; non-diabetic HD, 7983 ± 153; diabetic
HD, 8033 ± 197 base pairs). Neither was there any difference between the slopes of
telomere length vs. age in the three groups (figure). This study shows that PBMC
telomere length in HD patients is not affected by oxidative stress.
Renal Association October 2003
P 30
Single centre experience of encapsulating peritoneal sclerosis (EPS) in CAPD
M J Clancy, F Syed, A Summers, N Harwood, P E C Brenchley, R C Pearson and R
Manchester Royal Infirmary, Oxford rd, Manchester, M13 9WL, United Kingdom
Encapsulating peritoneal sclerosis is a rare but extremely serious complication of
peritoneal dialysis. This report on 21 cases (9 males and 12 females) of EPS diagnosed
between August 1996 and July 2001at this single centre, represents an incidence of
0.3%. The mean age was 41.1 years (SD 15.19 years) with a mean duration of CAPD of
71.9 months (SD 38.27 months). These patients had 51 episodes of peritonitis
(peritonitis rate - 1 episode/30 patient months). Thirteen patients developed severe
manifestations (defined by the clinical necessity for surgery), while 7 had mild to
moderate symptoms. The severe group required a total of 14 laparotomies with 4
perioperative deaths. In total there were 6 deaths – an overall mortality of 28.6%.
Thirteen patients required TPN, 11 perioperatively, 2 for nutritional support without
surgery. Most patients also received tamoxifen long-term with good effect.
Patient groups for severe and mild to moderate EPS were compared using the t-test for
independent samples and the Mann Whitney test. There was no significant difference in
patient age. There was, however, a significantly longer mean duration of PD in the
severe group, 82.7 months as compared with 50.3 months in the mild to moderate group
(p=0.029 Mann Whitney, p=0.033 t-test). There was also a greater mean number of
peritonitis episodes in the severe group, 2.79 compared with 1.29 (p=0.051).
We propose a summary of recommendations for all capd patients which include an early
CT scan of the abdomen and pelvis in presence of suspicious symptoms and peritoneal
biopsy at any operative surgery in which the peritoneum is opened. Early diagnosis,
where possible, followed by cessation of peritoneal dialysis and therapy with tamoxifen
are advocated. If surgery is indicated, then nutrition is a critical factor and must be
addressed at the earliest possible time. Meticulous surgical technique is recommended
to avoid enterotomies and their consequences.
Renal Association October 2003
P 31
The Expression of the Suppressors of Cytokine Signalling (SOCS) is Reduced
Following Haemodialysis
JE Kieszkiewicz, DA Allen, MJ Raftery and MM Yaqoob
Dept. Experimental Medicine & Nephrology, William Harvey Research Institute, Queen
Mary, University of London, London, EC1A 7BE, United Kingdom
The response of haemodialysis (HD) patients to erythropoietin (EPO) is impaired by
cytokines produced by the dialysis process but the optimal time for administration of IV
EPO is unknown. SOCS are induced by cytokines and negatively regulate signaling. In
this study we have examined the effect of sera from haemodialysis patients on the
expression of SOCS-1, 2, 3 and CIS in a human leukaemic cell line. UT-7 cells were
cultured in RPMI 1640 containing 10% v/v fetal calf serum. Sera from HD patients
(10% v/v) were added to the media and cells. After 0, 1, 2 and 4 h the cells were
collected and RNA extracted. cDNA was synthesised and SOCS expression analysed by
real-time RT-PCR using the Taqman assay. Primers and probes specific for human
SOCS were designed using Primer Express software and DNA templates for standard
curve generation were purified plasmids containing the coding region of each of the
SOCS genes. A ribosomal RNA template was used as a control. The relative expression
of each SOCS gene was analysed in triplicate at each time point for all patients. The
ratio of SOCS:ribosomal RNA was then determined and results expressed as a fold of
control (expression at time 0). Uraemic sera increased the expression of all 4 SOCS
genes. Pre-dialysis sera induced SOCS to a greater extent than did post-dialysis sera
although this was significant only for CIS (n--3, P<0.02). In addition, the expression of
SOCS genes was lower after 4 h exposure to post-dialysis sera when compared with
pre-dialysis sera (% reduction: SOCS-2 42%; SOCS-3 38%; CIS 55%, n=4). Overall,
pre-dialysis sera were more effective at inducing the expression of SOCS when
compared with post-dialysis sera. Moreover, in UT-7 cells exposed to pre-dialysis sera
SOCS expression was prolonged when compared with post-dialysis sera. We conclude
from this study that, in HD patients, pre-dialysis sera causes induction of SOCS genes
and that this effect is reduced by dialysis. The SOCS proteins are important inhibitors of
EPO signaling and they are at the lowest level of expression in the post-dialysis phase.
The optimal time for administration of IV EPO is in the post-dialysis phase.
Renal Association October 2003
P 32
Mechanism of Leukocyte Recruitment in Experimental Peritonitis.
JF Cailhier, S Watson, T Kipari, MF Clay, R Lang, J Savill and J Hughes
Center for Inflammation Research,Medical School University of Edinburgh, Teviot
Place, Edinburgh, EH9 8AG, United Kingdom
We previously utilised conditional macrophage (Mø) ablation mice to demonstrate that
macrophage depletion significantly attenuated both neutrophil (PMN) and Mø
infiltration in thioglycollate (TG) peritonitis. In this study we have further investigated
the factors orchestrating leukocyte recruitment in this model.
Conditional Mø ablation mice are transgenic for the human diphtheria toxin receptor
(DTR) under the CD11b promoter and exhibit 98% depletion of resident peritoneal Mø
following intraperitoneal (IP) administration of diphtheria toxin (DT, 25ng/gBW). DT
exerts no effects upon PMNs but does diminish peritoneal B lymphocyte numbers. We
adopted a Mø repletion strategy to further assess the role of Mø. Mø-depleted DTR
mice were reconstituted by IP administration of either (i) Mø-rich peritoneal cells or (ii)
peritoneal cells depleted of Mø by adhesion to tissue culture plastic for 2 hrs. Mice were
injected with 1ml 3% TG 4 hrs later and underwent peritoneal lavage at various time
points. Peritoneal cells were counted and labelled for Mø (F4/80) and PMN (GR1)
markers and analyzed by flow cytometry.
Administration of Mø-rich peritoneal cells completely restored PMN to levels
comparable to wild-type mice and increased Mø infiltration. In contrast, mice
administered peritoneal cells depleted of Mø exhibited markedly diminished leukocyte
influx comparable to control Mø-depleted mice.
ELISA analysis of PMN chemokines showed that Mø-depleted mice had significantly
reduced levels of peritoneal KC and MIP-2 at the 1hr time point compared to nondepleted mice (KC: 2467±264 vs 1408±322 pg/ml, p<0.05; MIP-2: 1762±153 vs
148±35 pg/ml, Mø-depleted vs control, p<0.0001). We also found reduced monocyte
chemokine concentrations in the Mø-depleted mice, which had lower levels of MIP-1_
and MCP-1 at 8h (MIP-1_: 26.7±3.4 pg/ml vs 153±16.4 pg/ml, p<0.005; MCP-1:
444±122 pg/ml vs 997±96, Mø-depleted vs control, p<0.05).
In summary, Mø-directed production of KC and MIP-2 effect PMN influx whilst MIP1_ and MCP-1 regulate Mø traffic. These data demonstrate the critical importance of
resident peritoneal Mø in leukocyte recruitment to the inflamed peritoneum.
Renal Association October 2003
Moderated Poster Session - One
Group (D)
Tuesday 7 October
Garner Room
Renal Association October 2003
P 33
Single Nucleotide Polymorphisms(SNPs)in the Cytokines Interferon _, Interleukin6, Tumour Necrosis factor _ and Transforming Growth Factor _1 and Peritoneal
Membrane Function: A longitundinal observational Study
EJ Sharples1, M Varagunam2, PJ Sinnott2, D McCloskey2, SL Fan1, MJ Raftery1 and
MM Yaqoob1
Department of Renal Medicine and Transplantation, Royal London Hospital,
Whitechapel, London, E1 IBB, United Kingdom and 2Tissue Typing Laboratory, Royal
London Hospital, Whitechapel, London, E1 1BB, United Kingdom
Peritoneal solute transport progressively increases with time on dialysis and is
associated with reduced ultrafiltration capacity. Polymorphisms in the promotor regions
of several pro-inflammatory cytokines have been described. We studied 112 consecutive
patients attending a single CAPD clinic for at least 3 months. Genomic DNA was
extracted from peripheral blood leukocytes and genotyping performed with ARMS-PCR
methodology. We examined gene polymorphisms for the cytokines IL-6(-174 G/C),
TNF_(-308 G/A), TGF_1(+869 T/C and +915 G/C) and interferon_(-869 T/A). Clinical
data for CRP, peritonitis events, time on dialysis, albumin and nPCR was collected
longitundinally, and PET performed annually. There was a mean follow up for
25±1.2(range 12-60) months. Initial membrane transport was affected by age and
diabetes (p<0.05) but there was no effect of other clinical variables or gene
polymorphisms. There was a significant increase in D/P creatinine ratio over the
study(p<0.01). Initial transport status predicted glucose transport, ultrafiltration and
change in transport over the study period(p<0.01, P<0.001, P<0.01). Peritonitis had a
significant effect on the change in transport(p<0.01), which was proportional to the
number of peritonitis events. There was a significantly higher incidence of peritonitis in
over-producers of interferon_(p<0.05). G/A polymorphism in the TNF_ promotor was
associated with a significant difference in small solute transport over time(p<0.05).
There was no effect of the other cytokine gene polymorphisms studied on transport or
ultrafiltration. In summary, changes in membrane function are dependent on initial
membrane function and the incidence of peritonitis, with a small but significant effect of
TNF_ polymorphism. However, interferon _ genotype was strongly associated with
increased peritonitis. This study highlights the importance of SNPs in a subgroup of
CAPD patients.
Renal Association October 2003
P 34
Pre transplant soluble CD30 levels are predictive of type and severity of rejections
R Rajakariar1, N Jivanji2, M Varagunam1, A Gupta2, M Sheaff3, P Sinnott2 and MM
Suite 22, Dominion House, Bartholomew Close, London, EC1A 7BE, United
Kingdom, 2Department of Tissue Typing, The Royal London Hospital, Whitechapel,
London, E1 1BB, United Kingdom and 3Department of Histopathology, The Royal
London Hospital, Whitechapel, London, EC1A 7BE, United Kingdom
CD30 molecule is a member of the tumour necrosis factor/nerve growth factor receptor
superfamily and is preferentially expressed in CD4+ and CD8+ cells that secrete TH2
type cytokines. In renal transplantation higher soluble CD30[sCD30] levels are
associated with increased rates of rejection and graft loss. We investigated whether the
levels of sCD30 correlated with the severity of acute rejection. In our unit 279 renal
transplants were performed between January 1996 and June 2001. Pre transplant sera
were available in 51 patients with cellular rejection, 18 patients with vascular rejection
and 31 patients with no rejection. sCD30 levels were measured using a commercially
available ELISA assay. The groups were comparable in age, gender, dialysis modality,
PRA and immunosuppression regime. sCD30 levles were lower in cellular
rejection[119.3+/-8.8, p<0.001], higher in vascular rejection[260.6+/- 37.6, p=0.03]
compared to no rejection group[180.5+/-10.5]. Also the difference in sCD30 levels
between the cellular and vascular groups was significant[p<0/001]. In cellular and
vascular groups, sCD30 did not determine the responsiveness to high dose steroids and
ATG respectively or graft outcome at last follow-up. The data suggests sCD30 levels are
associated with type and severity of rejection. Higher sCD30 levels may be associated
with TH2 cytokine mediated humoral immunity leading to vascular rejection where as
depressed levels lead to a TH1 type response hence cellular rejection. Monitoring pre
transplant sCD30 may be useful to risk stratify patients and indivdualise
Renal Association October 2003
P 35
Lack of Differential Effect of Tacrolimus and Cyclosporin on Platelet Activation
and Circulating Platelet-Monocyte Aggregates Post Renal Transplantation
J Templar1, C Whittaker1, MG Macey2, MJ Raftery1 and MM Yaqoob1
Department of Renal Medicine and Transplantation, Royal London Hospital,
Whitechapel, London, E1 1BB, United Kingdom and 2Department of Haematology,
Royal London Hospital, Whitechapel, London, E1 1BB, United Kingdom
Cyclosporin (CyA) may contribute to the high cardiovascular morbidity and mortality
observed in transplant patients. In contrast, Tacrolimus (Tac) is associated with an
improved cardiovascular risk profile in terms of blood pressure and cholesterol control.
Recent studies suggest that activated platelets and in particular platelet-monocyte
aggregates reflect plaque instability and ongoing thrombosis in acute coronary
syndromes in non-uraemic patients. We have investigated the effect of two different
calcineurin inhibitors on this novel cardiovascular risk factor and other platelet
activation parameters. Age, gender, duration of transplantation and renal function
matched patients, who were either on CyA microemulsion (n=12) or Tac (n=12) as part
of their triple immunosuppression therapy, were analysed for platelet P-selectin
expression, monocyte and neutrophil P-selectin glycoprotein ligand-1 (PSGL-1)
expression, platelet-leukocyte aggregates and platelet microparticles using whole blood
flow cytometry. Samples were also obtained from haemodialysis patients (HD (n=30),
who were on the transplant waiting list.
Results (mean±SEM) revealed that the total transplant group (n=30) had significantly
higher levels of P-selectin (p=0.009), platelet-monocyte aggregates (p=0.001), plateletneutrophil aggregates (p=0.044) and PSGL1-monocytes (p<0.001) than HD patients. In
contrast, there was no significant difference in all platelet activation parameters between
the CyA and Tac transplantation sub-groups.
Here, we have shown for the first time that both calcineurin inhibitors worsen the
rheological cardiovascular risk factor following transplantation and that Cyclosporin
and Tacrolimus have similar effects on platelet activation and platelet-monocyte
aggregate formation.
Renal Association October 2003
P 36
A Polymorphism of the NAD(P)H oxidase p22 phox Component is Associated with
Reduced Susceptibility to Acute Rejection in Renal Allograft Recipients
UV Bhandary1, W Tse2, AD Hodgkinson1, PA Rowe2, RJ McGonigle2 and AG
Molecular Medicine Research Group, Institute of Biomedical Sciences, Peninsula
Medical School, ITTC Building, Tamar Science Park, Plymouth, PL6 8BX, United
Kingdom and 2Department of Renal Medicine, Derriford Hospital, Derriford Road,
Plymouth, PL6 8DH, United Kingdom
Reactive oxygen species (ROS) such as superoxide (O2-) play an important role in the
inflammatory process. By altering the redox environment, ROS modulate the activation
of transcription factors and expression of cytokine genes involved in acute cellular
rejection. The NAD(P)H oxidase is a multi-subunit enzyme present in a variety of cells,
including leukocytes and endothelial cells, and is a key source of O2-. A C242T
polymorphism of the p22 phox subunit, encoding a histidine-72 to tyrosine substitution
at the putative haem binding site, has been shown to be associated with reduced
NAD(P)H oxidase activity. The aim was to determine the effect of the C242T
polymorphism on susceptibility to acute rejection in renal allograft recipients.
The C242T polymorphism was studied in 134 Caucasian renal allograft recipients using
PCR. Acute rejection was diagnosed by renal biopsy in 63 patients (47%). Controls
were DNA samples extracted from umbilical cord blood following uncomplicated
obstetric delivery (n=131).
A highly significant increase in the frequency of the T242 allele in allograft recipients
compared to normal controls was demonstrated (95/268 (35.4%) versus 43/262 (16.4%),
_2=24.9, p<0.0001). There was also a highly significant increase in the frequency of the
T242 allele in allograft recipients without acute rejection, compared to patients with
acute rejection and normal controls (58/142 (40.8%), 37/126 (29.4%) and 43/262
(16.4%) respectively, _2=29.5, p<0.0001). A highly significant increase was also
demonstrated in the frequency of the TT242 genotype in patients without acute rejection
compared to those with rejection and normal controls (15/71 (21.1%), 4/63 (6.3%) and
0/131 (0%) respectively, _2=38.1, p<0.0001).
In conclusion, these results suggest that possession of the T242 allele in renal allograft
recipients may predispose to the development of end stage renal failure, but
paradoxically reduce susceptibility to acute rejection through reduced NAD(P)H
oxidase activity.
Renal Association October 2003
P 37
Paraoxonase 1 polymorphisms in chronic allograft rejection following renal
S Harris1, CD Short1, ISD Roberts2 and PEC Brenchley1
Renal Research, MINT, Manchester, M13 9WL, United Kingdom and 2Cellular
Pathology, John Radcliffe Hospital, Oxford, OXU 9DU, United Kingdom
The pathology of chronic renal allograft nephropathy is dominated by vascular damage
similar to that seen in atherosclerosis and cardiovascular (CV) complications account
for ~50% of patient deaths on renal replacement therapy (RRT), a much higher
proportion than in the general population.
Human paraoxonase 1 (PON1) an esterase associated with HDL in the serum is reported
to prevent oxidative modification of LDL, a key step in the formation of atherosclerotic
plaques. Prevention of oxidation or hydrolyses of oxidised lipids could, therefore, retard
the development of atherosclerosis.
PON1 activity is partly genetically controlled although environmental factors also have
an impact on enzyme activity. Seven polymorphisms have been described within the
gene coding for PON1. Five lie in the promoter region of the gene at positions -108, 126, -162, -832 and –909 and 2 in the coding sequence, L55M and Q192R.
We aimed to investigate the possible association between PON1 polymorphisms and the
development of chronic allograft nephropathy in a cohort of renal transplant recipients.
90 renal transplant recipients were recruited prospectively between 1994-95. DNA
samples from this cohort and 125 normal healthy volunteers were genotyped for 5
polymorphisms 3 in the promoter region (-909, -162, -108) and 2 in the coding sequence
(55 and 192) using PCR-RFLP.
The genotype frequencies of the total patient group were comparable to frequencies
found in volunteer controls and also to the published frequencies for all 5
polymorphisms. We found no significant differences between the genotypic or allellic
frequencies between the patients who developed CAN and those with stable renal
function at 4 years.
The PON1 genotypes do not appear to be involved in the development of CAN in this
cohort of renal transplant recipients.
Renal Association October 2003
P 38
Living donor transplantation at 20 pmp/yr in the UK – achievable, sustainable and
R Higgins, FT Lam, H Kashi, A Stein, S Fletcher, F Price, P Hart and N West
Renal Unit, Walsgrave Hospital, Coventry, CV2 2DX, United Kingdom
Introduction. In the first quarter of this year, the living donor transplant (LDT) rate in
our unit was 20pmp/yr, compared to 15 in 2002 and 13.8 in 2001. The rest of the UK
achieved 6.5, 6.2 and 5.9pmp/yr in 2002, 2001 and 2000 respectively. The optimum rate
for LDT is not defined. International practice varies (all figures pmp/yr):- Spain, 0.8;
France, 1.8; USA, 21.5; Norway, 21.6. There is variation between UK units, from
1.7–15pmp in 2002.
Aim. Is a LDT rate of 20 pmp/yr is likely to be sustainable, and is it desirable in terms
of case mix and outcomes?
Methods. The last 50 LDT in our unit were analysed, and divided equally into early and
recent cohorts. National figures from Renal Registry and UKT.
Results. Comparison of the 50 LDT to our current transplant list showed non-white
European ethnicity was 20% of LDT vs 16% of list; diabetic nephropathy was 6% of
LDT vs 9% of list. Early and recent cohorts were similar; respectively:- recipient age
40.4 (SEM 2.7) and 38.8 (2.0) years; donor age 45.9 (2.3) and 46.0 (2.6) years; mean
DR mismatch 0.88 (0.12) and 0.88 (0.16); acute rejection rate 40% and 38%; rescue for
resistant rejection 4% and 12%; estimated CrCl in surviving grafts at 3 months 60.3
(3.5) and 59.0 (4.2) ml/min. One graft from each cohort was lost in the first 3 months,
from vascular thrombosis. One recipient died at 22 months. There were differences
between the cohorts; proportion of pre-emptive transplants, 8% and 44%; mean time on
dialysis 33.8 (7.0) and 12.3 (3.2) months, p = 0.008; female recipients 48% and 28%;
unrelated donors 28% and 36%.
A LDT rate of 20 pmp/yr seems achievable and sustainable in our unit. However,
compared to the rest of the UK, our area has a high prevalence of renal failure (655 vs
565 pmp), and a higher rate on the transplant list (122.5 vs 84.4pmp). It is possible that
a rate of 20 pmp/yr in our area would equate to about 15 pmp/yr across the UK.
Summary. Our unit has achieved an increased rate of LDT. Many donors were
unrelated, usually spouses, and pre-emptive transplantation was frequebntly performed.
Quality of life research further to examine the outcomes of unrelated donation and preemptive transplantation are indicated. Our results should inform a debate about the
future capacity for LDT in the UK, as the current target of 10 pmp may be modest.
Renal Association October 2003
P 39
Efficiency of immunoglobulin and anti-HLA antibody removal by plasma
exchange before and after renal transplantation
R Higgins1, M Hathaway2, D Briggs2, FT Lam2, H Kashi2 and L Hunns2
Renal Unit, Walsgrave Hospital, Coventry, CV2 2DX, United Kingdom and 2Blood
Transfusion Service, Vincent Drive, Edgbaston, Birmingham, B15 2SG, United
Introduction. Transplantation across antibody barriers can be facilitated by antibody
removal. Recently, repeated plasma exchange (PEx) has been successfully used, and
pooled immunoglobulins (ivig) given to reduce antibody resynthesis. A patient was
treated who was unable to tolerate ivig. Therefore the efficiency of PEx and the tempo
of antibody resynthesis under treatment with tacrolimus, mycophenolate, prednisolone
and basiliximab was measured.
Patient and method. A 23 year old female received 5 sessions of PEx, each of approx 1
plasma volume, on alternate days before a renal transplant from her brother. PEx was
performed daily from days 7-10 because of an anti-DR15 response associated with
vascular rejection.
Modelling was performed, assuming a session of PEx would reduce serum IgG level by
60%. As 50% of IgG is extravascular and has a slow redistribution rate, rebound in the
serum level then occurs after PEx.
Results. Predicted versus measured serum IgG levels, respectively and % of initial
levels, were
70% and 66% pre-2nd PEx; 49% and 40% pre-3rd PEx; 34% and 35% pre-4th PEx;
24% and 24% pre-5th PEx; and 17% and 16% at transplantation. Serum IgG level fell
from 9.8 g/l to 1.2g/l at the time of transplantation. Dilution of pre-treatment serum
showed reduction in the titres of anti-HLA DR4, DR7, and DR15 antibodies followed
the percentage reductions in total IgG level after each PEx. IgM, which has no
significant extra-vascular distribution fell to 8% of starting level at transplantation.
After transplantation, the proportion of anti-HLA DR antibody (donor and third party)
rose sharply as a % of total IgG, and also rose faster than the rate of removal achieved
by daily PEx.
Summary. PEx may be appropriate technology to achieve antibody removal pretransplant, and a simple prediction of the rate of removal of serum IgG pre-transplant
was mirrored by the total IgG level and anti-HLA antibody levels. These data may help
plan PEx schedules pre-transplant. However, post-transplant, anti-donor antibodies were
resynthesised faster than removal by daily PEx. Successful engraftment in this context
requires either a more effective antibody removal than by PEx; and/or other strategies to
achieve modulation or deletion of cells producing antibody and/or accommodation to
anti-donor antibodies.
Renal Association October 2003
P 40
The potential impact of introducing the recent guidelines for the management of
glucocorticoid-induced osteoporosis in a renal transplant programme.
SD Roe1, CJ Porter1, R Sims1, C Parker3, DJ Hosking2 and MJD Cassidy1
Renal Unit, Nottingham City Hospital, Hucknall Road, Nottingham, NG5 1PB, United
Kingdom, 2Department of Bone Mineral Metabolism, Nottingham City Hospital,
Hucknall Road, Nottingham, NG5 1PB, United Kingdom and 3Department of Diabetes
and Endrocrinology, Airedale NHS Trust, Keighley, BD20 6TD, United Kingdom
The management of bone disease post renal transplantation varies between centres.
Many patients may not have been evaluated for osteoporosis risk or commenced on
treatment. Guidelines for the management of glucocorticoid-induced osteoporosis have
recently been published 1.
Since 1999 we have been monitoring bone mineral density(BMD) in newly transplanted
patients. We also have data in established female renal transplants from 1996 and male
transplants from 2000 (mean time post transplant 11 and 9 yrs respectively). The
distribution of BMD in patients using either the WHO classification or the newly
published guidelines is tabled below.
The majority of renal transplant patients are likely to require regular BMD monitoring
or treatment with bone protective agents. The cost of a two site BMD scan in our centre
is £86 and the cost of alendronate treatment is £277/yr. In the UK 1775 transplants were
performed in 2002-03 and the size of the established renal transplant population may be
as large as 14000 (UKTS figures). Using these numbers the annual cost of introducing
BMD monitoring for new transplant patients would be £152,650. Additional resources
would also be required for extra monitoring of patients with a T score between 0 to
–1.5. The potential cost of bisphosphonate treatment for established transplant patients
could be as high as £2.3M/year.
In the future the development of steroid sparing/free immunosuppressive regimes may
reduce the burden of transplant osteoporosis. Today, for most renal transplant centres
the additional cost of introducing bone protective treatment for newly transplanted
patients will be fairly small. Of greater concern are the logistics and additional cost of
investigating and treating the established kidney transplant population, especially when
epidemiological data of fracture incidence in this population is sparse and no studies
have established the efficacy of bone protective treatment in fracture prevention.
1) Royal College of Physicians. Glucocorticoid-induced osteoporosis. A guide to
prevention and treatment. London: RCP; 2002.
Renal Association October 2003
P 41
Long-Term Cost-Effectiveness of Mycophenolate Mofetil (CellCept®) in Renal
Transplantation in the UK
G Lewis1, L Garrison2 and A Ready3
Roche Products, 40 Broadwater road, Welwyn Garden City, AL7 3AY, United
Kingdom, 2PBSE, Hoffman La Roche, Basel, CH4070, Switzerland and 3University
Hospital Birmingham NHS trust, Birmingham, B15 2TH, United Kingdom
Background: Three major Phase 3 randomized clinical trials (n>1500) with 3 years of
follow-up showed that immunosuppressive regimens with Mycophenolate Mofetil
(MMF) improve acute rejection rates and both graft and patient survival (GS and PS).
Better clinical outcomes also mean cost reductions for acute rejection, graft failure, and
hemodialysis. The long-term cost-effectiveness (CE) of immunosuppressive regimens,
including those with MMF, has not been assessed in clinical trials. Such assessments
require economic models, combining trial results with registry data and information on
evolving transplant protocols.
Methods: To assess the potential CE of current MMF protocols, a Markov model was
constructed to compare cadaveric donor recipients receiving MMF or Azathioprine
(Aza) triple therapy over a 10 year period following transplantation. UK transplant audit
data were used to project patient, graft, and dialysis survival with Aza. The relative
impact of MMF compared to Azathioprine in survival outcomes was based on a large
US registry study (n> 60,000), showing a 14% relative risk reduction in PS [p<0.0001]
and GS [p=0.005] for MMF compared to Aza. MMF dosing was based on trial data and
cyclosporine (CsA) dosage was reduced by 30% in the MMF arm based on a recent UK
audit. Quality of life and costs for each health state were based on published literature.
Results: The graft survival advantage of MMF compared to Azathioprine was projected
to grow from 3.8 per 100 initial recipients at the end of year 1 to 6.3 per 100 after 10
years. The patient survival advantage grew from 0.5 per 100 recipients in year 1 to 3.1
per 100 after 10 years. MMF was estimated to improve quality-adjusted life years
(QALYs) by 0.24 years compared to Aza. The total additional discounted cost (drug +
treatment costs) per patient was £5,534. The cost-utility ratio was therefore £23,060 per
QALY. Sensitivity analyses confirmed this to be a robust estimate.
Conclusion: In a long-term economic model, an MMF regimen is projected to generate
a growing, favorable differential in graft and patient survival, resulting in substantial
savings in costs for dialysis, treatment of acute rejection, and graft failure. Patient
quality of life is better by avoiding dialysis. Thus, based on current CsA-sparing
protocols, an MMF regimen is projected to be cost-effective in the long term.
Renal Association October 2003
P 42
Predictors of Falls and Fractures in Elderly Transplant Patients
MC Casey1, M Healy1, C Walsh1, G Mellotte2, N Maher1, C Kirby1, B Keogh2, O
Geraghty1, E Thornton1, D Coakley1, C Cunningham1 and JB Walsh1
Osteoporosis and Falls Unit, Mercers Institute for Research into Ageing, St James
Hospital and 2Department of Nephrology, AMNCH Hospital, Dublin 8
Falls and fractures increase with age. Post transplantation, glucocorticoid therapy
compounds the abnormal bone histology due to secondary hyperparathyroidism,
osteomalacia/adynamic bone disease. This study examines biochemical, densitometric
and clinical risk factors for falls and fractures in older transplant patients.
Forty consecutive older (>60y) transplant recipients underwent DEXA with lateral
vertebral spine and total body mineral (BMC) assessment, body mass index, PTH (1565 pmol/l), 25 (OH)D (20-40 ng/ml), estimated Cr/Cl, osteocalcin (OC, 10-50 ng/ml)
and c-telopeptide (CTX normal <1.0 mcg/l), markers of bone formation and resorption.
Measurements of falls risk performed most patients were: Tandem Stand and Walk,
Visual acuity, history of <4 hrs/day spent on feet and postural hypotension (systolic BP
drop>20mmHg, or diastolic drop>10 mmHg on going from lying to standing).
Their mean (SD) age was 66(5.2), 18m and 22fm, 27% had osteoporosis (T< -2.5 on
DEXA). 25% of all patients had fractured vertebrae. Mean (+/- SD)of results were
osteocalcin was 54(41)ng/ml, mean CTX was 0.56(0.39) mcg/l, mean Cr/Cl was 50+14
mls/min, mean PTH was 126(30)pmol/l and mean 25(OH)D levels were 17(7)ng/ml.
Age, BMI and Osteocalcin were significantly related to T scores (p<0.001, p<0.01,
p<0.05 respectively). BMC was significantly determined by female gender, 25(OH)D
level and age (p<0.001, 0.001 and 0.05). OC and CTX were highly positively correlated
(p<0.001), both correlated positively with PTH (p<0.05 and p<0.001 respectively). A
negative relationship was noted when OC and CTX were correlated with 25(OH)D (p
<0.05 and p<0.07(NS) respectively). Both markers correlated comparably with femoral
T score (p<0.05).
45% failed the tandem walk, 30% failed the tandem stand with eyes open, 75% with
eyes closed. 50% had postural hypotension, 65% had impaired vision despite aids of <
6/12 using a snellen chart at 5 meters. 25% spent <4 hours a day on their feet. 20%
reported a previous fall.
Multiple variables including age, parathyroid hormone 25(OH) vitamin D deficiency
have adversely effected the densitometric and biochemical status of these patients
bones. The high prevalence of vertebral fractures, as well as fall related risk factors
underline the urgent need for multidisciplinary intervention to prevent further adverse
events in this ageing population
Renal Association October 2003
P 43
RJA Sims1, CJ Porter1, SD Roe1, J Monaghan2, DJ Hosking3 and MJD Cassidy1
Renal Unit, Nottingham City Hospital, Hucknall Road, Nottingham, NG5 1PB, United
Kingdom, 2Department of Clinical Diagnostics, Nottingham City Hospital, Hucknall
Road, Nottingham, NG5 1PB, United Kingdom and 3Department Of Bone Mineral
Metabolism, Nottingham City Hospital, Hucknall Road, Nottingham, NG5 1PB, United
Bone loss is a well-recognised complication of renal transplantation. Glucocorticoid use
is implicated as a risk factor and guidelines for the management and treatment of
glucocorticoid-induced osteoporosis have recently been published 1. The causes of
reduced bone mass in renal transplant recipients are more complex encompassing both
high and low turnover states. Patients also have impaired renal function and in some
glomerular filtration is below that recommended for bisphosphonate use. In order to
study the natural history of loss of bone mass following transplantation we have
measured bone mass (BMD) at the femoral neck (FN) and lumbar spine (LS) shortly
after transplant and at 1 year while monitoring bone biochemistry. Eighty-one patients
had received a renal allograft between September 1999 and May 2002 and 70 completed
a one-year follow up (36 males, median age 43 years, range 21 –71).
Using baseline BMD results we divided patients into 3 groups using WHO criteria for
osteopaenia and osteoporosis(either FN or LS). There was a significant reduction in
mean BMD at the FN at 1 yr compared to baseline 0.91±0.18 g/cm2 vs. 0.89±0.17
g/cm2 (p=0.005). LS BMD was unchanged. The table below shows demographics
according to BMD group.
Over the course of the year our best practice guidelines for maintaining healthy bones
included: reviews at a multi-disciplinary clinic, correction of vitamin D insufficiency,
hormone replacement therapy where indicated and patient education. Bisphosphonates
were not prescribed.
At one year patients generally remained in the same BMD group. Within the
osteopaenic group 2 patients became osteoporotic and 4 moved into the normal range
and in the normal group 7 patients became osteopaenic. All patients who were
osteoporotic remained so.
In view of these findings we now treat all patients who have osteoporosis at their first
DEXA post-transplant with a bisphosphonate provided there are no contra-indications
and monitor other patients with serial DEXA scans in addition to our best practice
1) Royal College of Physicians. Glucocorticoid-induced osteoporosis. A guide to
prevention and treatment. London: RCP; 2002.
Renal Association October 2003
P 44
North West Kidney Transplant Audit Project – Audit against Standards
HJ Moore, PA Dyer, F Qasim and HN Riad
Transplantation Laboratory, Manchester Royal Infirmary, Oxford Road, Manchester,
M13 9WL, United Kingdom
The North West Kidney Transplant Audit Project aims to develop a process for audit of
renal transplantation in adults and children, of equity of access to the transplant waiting
list and of surgery related to renal failure in the former NW RHA. The clinical audit
process involves measurement of performance against nationally recognised standards
from the British Transplant Society
BTS Standard 4.2.3: “The kidney cold storage time whenever possible should be kept
below 24 hours.”
•In 1993 to The number of cases not meeting this standard has decreased since 1999. In
2002 79.5% of transplants met the standard (range 2 – 40hours)
Prospective audit started January 2003 to identify where delays are occuring, 6 month
data is near completion.
BTS Standard 4.2.4: “Each unit should aim to transplant at least 26 patients per million
population per year with cadaver kidneys.”
Manchester Royal Infirmary serves a population of approximately 4.25 million. To
achieve the standard for one year 111 people should receive a cadaveric kidney
transplant. In 2002 108 patients received a cadaveric renal transplant (including those
receiving a simultaneous pancreas and renal transplant).
BTS Standard 4.5.1: “45% of recipients should receive a ‘favourably’ HLA matched
In 2002 69.7% of all recipients receiving a renal alone transplant, had a “favourably”
matched kidney. By definition that is those with 000 or 100 / 010 / 110 HLA mismatches.
BTS Standard 4.7.2: “Patient survivals should exceed 90% at one year, 80% at 5 years
and 60% at ten years.”
Patient survivals have consistently met the one year target since 1992 peaking in 2001 at
99.1%. 5 year survivals met the standard except in 1995 with a low of 77.6%. 10 year
survivals surpassed the 60% target
BTS Standard 4.7.3: “Graft survivals should exceed 80% at one year, 60% at 5 years
and 45% at ten years.”
Graft survivals have consistently met the one year survival target. Indeed the trend has
shown a steady upwards trend. For 2002 the 1 year survival was 92.3%. At 5 years
survivals have surpassed the standard. A steady improvement has been observed since
1996 The 10 year survival for transplants carried out in 1993 shown was 47.6%
The NWKTA Project is carrying out ongoing audit and re-audit of a selection of
standards. This is novel within renal transplantation in the UK. Data is available at
Renal Association October 2003
Moderated Poster Session - Two
Group (A)
Cardiovascular System in Kidney Disease
Wednesday 8 October
Studio Suite
Renal Association October 2003
P 45
Carotid artery intima-media thickness is influenced by the aetiology of the
underlying renal disease.
E C Preston1, D V Pavitt2, D Jackson2 and E A Brown1
Dept. of Nephrology, Imperial School of Medicine, Charing Cross Hospital, London,
W6, United Kingdom and 2Dept. of Clinical Chemistry, Imperial School of Medicine,
Charing Cross Hospital, London, W6, United Kingdom
Carotid artery intima-media thickness (CA-IMT) is an established measure of
atherosclerosis that correlates well with the prevalence and incidence of coronary heart
disease and stroke in the normal population. Renal patients are said to have an early or
‘accelerated’ atherosclerosis and one could therefore hypothesise that we should find an
increased CA-IMT in these patients.
CA-IMT was measured in 115 patients (age range 30-82yrs) with mild to moderate
renal impairment (EDTA clearance 10-109mls/min/1.73m2) and 13 controls. This was
compared with the patients cardiovascular risk factors including homocysteine and
Lipoprotein (a). The patients were divided into what we defined as those with primary
renal disease (i.e. glomerulonephritis and urological disease) and secondary disease (i.e.
diabetes, hypertension and renovascular disease) and those with an unknown diagnosis.
CA-IMT significantly increased with increasing age (mean CA-IMT: age<55: 0.053cm,
ages 56-70: 0.057cm, age>70: 0.068cm, p=0.0085). We found increasing levels of
lipoprotein (a) and homocysteine as GFR decreases suggesting increasing
cardiovascular risk with declining GFR. No correlation was found though between CAIMT and GFR. Those patients with secondary renal disease had a significantly increased
mean CA-IMT (0.0622cm) over controls (0.0438cm) as compared to those with primary
renal disease (0.0568cm, p=0.0435). This evidence of increased cardiovascular risk in
the secondary disease group is also illustrated by the fact that 29.5%(n=13) of this group
has previously documented ischaemic heart disease as compared to only 4%(n=2) in the
primary renal disease group. There was no difference in the amount of peripheral
vascular disease between the groups.
We conclude that when measuring CA-IMT in patients with renal failure the underlying
renal disease must be taken into account.
Renal Association October 2003
P 46
The progression of cardiac dysfunction in atherosclerotic renovascular disease
JR Wright1, AE Shurrab1, A Cooper2, RN Foley1 and PA Kalra1
Department of Renal Medicine, Hope Hospital, Stott Lane, Salford, M6 8HD, United
Kingdom and 2Department of Cardiology, Hope Hospital, Stott Lane, Salford, M6 8HD,
United Kingdom
Baseline proteinuria predicts deterioration in the renal function of ARVD patients;
baseline creatinine clearance predicts renal and patient survival. Here, the natural
history of cardiac dysfunction and factors which predict declining function in patients
with ARVD are delineated.
Forty-three patients (27M and 16F, age at study entry [mean ± SD] 69.7 ± 8.0 years)
with ARVD underwent echocardiography and 24-hour blood pressure investigations
within three months of diagnosis and 12 months thereafter. None of the patients
underwent renal revascularization procedures. The two data sets were interrogated to
determine changes in blood pressure and cardiac status (morphological and functional);
baseline factors which predicted such changes were ascertained.
At 12 months creatinine clearance had fallen (38.6 ± 18.3 vs 35.0 ± 18.5 ml/min; NS)
and proteinuria had increased (0.31 ± 0.37 vs 0.60 ± 0.77g/24hours; P<0.03). Despite no
increase in the number of blood pressure medications (2.44 vs 2.46; NS), there was a
striking fall in blood pressure between baseline and follow-up investigations (140.0 ±
16.5 / 75.3 ± 11.8, MAP 98.6 ± 12.3mmHg vs 135.5 ± 16.1 / 69.6 ± 9.1, MAP 92.5 ±
10.2mmHg; P<0.02 for diastolic BP and MAP).
At 12 months, there were increases in the number of patients with LVH (72.9% vs
81.4%; NS) and both systolic (76.7% vs 83.3%; NS) and diastolic (60.5% vs 72.1%;
NS) left ventricular functional abnormalities. At 12 months an increase in left
ventricular dilatation was observed, evidenced by an increase in left ventricular end
diastolic diameter (LVEDD, 5.1 ± 0.8 vs 5.5 ± 0.8cm; P<0.05) and left ventricular end
diastolic volume (LVEDV, 140.9 ± 39.5 vs 163.3 ± 61.0 ml; P <0.05). Progressive LV
dilatation was predicted by the presence of LVH at baseline and low baseline CrCl. No
changes in cardiac parameters were observed across renal artery patency score groups
(severe ARVD, n=6, moderate ARVD, n=24, mild ARVD, n=13).
Despite an improvement in blood pressure control, which may in part be the effect of
participation in the study, there was an increase in left ventricular dilatation in the year
between investigations. This progression of cardiac disease is associated with baseline
LVH and predicted by low initial CrCl. Treatment strategies to prevent such progression
must now be formed to break the deadly synergy of cardiac and renal dysfunction in
patients with ARVD.
Renal Association October 2003
P 47
Blood Pressure Control in Renal Replacement Therapies: the achieved, the
achievable and the aspirational.
EJ Will2, C Bartlett2, D Ansell1 and K Thomas1
UK Renal Registry, Southmead Hospital, Westbury-on-Trym, Bristol, BS10 5NB,
United Kingdom and 2Dept of Renal Medicine, St James's University Hospital, Beckett
Street, Leeds, LS9 7TF, United Kingdom
The blood pressures (BP) associated with desirable clinical outcomes are defined in
current Guidelines and Standards (G&S), derived from the clinical efficacy literature.
They are presented typically as recommended values less than a certain maximum
(e.g.<140/90 mmHg). These values are underachieved in both practice and research, the
upper limit of desired BP often becoming the mean of BP achieved. The spread of BP
readings to be expected in a successfully treated population is not usually in the mind of
clinicians during consultation or the development of treatment policies.
To define these distributions we examined the BP records in the UK Renal Registry
2001 database (23 Renal Units, n =4272 haemodialysis (HD)-predialysis values, 1371
peritoneal dialysis (PD) and 3313 transplant (TP) patients) for digit (0-9) bias and
compliance with modern G&S.
Digit bias (rounding to zero values, for example) was occasional in HD cohorts but
usual for PD and TP patient groups, the latter managed largely as outpatients. We could
show that automated BP results directly downloaded to a clinical database had no digit
bias (10.00 +/- 0.75% per digit, n = 2802 TP clinic records). ‘Rounding down’ could be
demonstrated in some Units. Plotting Renal Unit median BP against the percentage
compliance with standard values (say <140 or <160 systolic)(Rose/Day plots*) gave
linear relationships that confirmed ‘normal’ distributions.
The BP distributions displayed Standard Deviations (s.d.s) of up to twice the values
expected of hypertensive populations without ESRD (even after excluding centres with
large digit bias). The average systolic/diastolic s.d.s for HD, PD, TP were 27/15 25/12.5,
20/11 respectively, compared with 18/10, say, in non-ESRD hypertensive populations.
We could confirm the wider dispersions in ESRD populations from other literature
If these greater ranges are inherent to the ESRD population, compliance with BP G&S
in ESRD will be more challenging than in the community and translates into a
requirement to achieve lower median BP results (e.g. for HD median BP 115 mmHg for
85% < 140 mmHg at age <60, TP 142 for 85% <160 at age >60 years).
These data allow the differences between currently achieved, achievable and
aspirational BP distributions in ESRD to be quantified. They demonstrate the necessity
for calibration of G&S assertions with practical outcome data, if management policies
and clinical expectations are to be realistic.
* Rose G, Day S. BMJ 1990:301:1031-4
Renal Association October 2003
P 48
V Reddy1, A-M. L Seymour1 and S Bhandari2
Department of Biological Sciences, University of Hull, Cottingham Road, Hull, HU6
7RX, United Kingdom and 2Department of Nephrology, Hull Royal Infirmary, Anlaby
Road, Hull, HU3 2EZ, United Kingdom
Currently over thirty thousand patients in the United Kingdom receive treatment for
end-stage renal failure. The leading cause of mortality in this population is from cardiac
related causes. Secondary carnitine deficiency frequently occurs in haemodialysis
patients and this deficiency can lead to impaired fatty acid oxidation in the myocardium.
In addition, carnitine deficiency is associated with cardiac hypertrophy and heart failure.
A specific uraemic cardiomyopathy has been proposed on the basis of clinical and
experimental studies, which have demonstrated impaired contractile function and a
deficit in the myocardial energy reserve. However the mechanisms underlying this
disorder still require definition.
The aim of this study was to determine the effects of chronic uraemia with and without
an acute carnitine load on cardiac function and energy provision. Chronic renal failure
was induced in male Sprague-Dawley rats via a two-stage 5/6 nephrectomy. Three or six
weeks post surgery, hearts were removed and perfused in the isovolumic mode, with
oxygenated Krebs - Henseleit buffer containing 3% albumin, 5mM [1-__C] glucose,
0.3mM [U- __C] palmitate, 1mM lactate, 0.1mM pyruvate, 100_U/ml insulin ± 5mM
carnitine. Cardiac function and myocardial oxygen consumption were monitored
simultaneously. Subsequently, extracts of heart tissue were analysed by __C NMR
spectroscopy to determine the relative contributions of glucose, fatty acid and
unlabelled substrates to oxidative metabolism (Figure 1)
No significant difference in myocardial function or substrate utilisation was observed at
three weeks of uraemia. However, by six weeks there was a reduction in fatty acid
oxidation (33.0 ± 11.5% vs 25.8 ± 6.7%) and increase in glucose oxidation (19.0 ± 3.7%
vs 23.4 ± 6.9%) in the uraemic group, with evidence of cardiac hypertrophy and mild
cardiac dysfunction. At three weeks, acute carnitine load resulted in an 18.5% and
39.1% increase in fatty acid oxidation in both uraemic and control groups respectively.
In conclusion, prolonged uraemia leads to altered myocardial substrate selection, similar
to that observed in left ventricular hypertrophy. These changes can be ameliorated by
provision of carnitine.
Renal Association October 2003
P 49
Anti-atherosclerotic Effects of Sirolimus:Dual Actions on Both Cholesterol
Homeostasis and Anti-inflammation
Z Varghese, R Fernando, J Moorhead, S Powis, D Wheeler and X Ruan
Centre for Nephrology, Royal Free and UC Medical School, Rowland Hill Street,
London, NW3 2PF, United Kingdom
The immunosuppressive agent Sirolimus reduces atherosclerotic lesions in apo E
knockout mice and has recently found clinical application in the context of drug eluting
stents. To determine the possible mechanisms underlying the anti-atherosclerotic
properties of this compound, we investigated its effects on cellular cholesterol
homeostasis in vitro. We have previously shown that the inflammatory cytokines TNF_
and IL-1_ cause lipid accumulation in human mesangial and vascular smooth muscle
cells both by increasing lipid uptake through LDL, VLDL and scavenger (CD 36)
receptors, and by reducing cholesterol efflux via the ATP binding cassette transporter
A1(ABCA1) pathway. Addition of Sirolimus (10-100 ng/ml) reduced lipid accumulation
in human mesangial cells (HMCs) incubated with IL-1 _ (5ng/ml)as demonstrated by a
reduction in Oil red O staining of intracellular lipid droplets. Using real-time PCR, we
screened for mRNA expression of lipoprotein receptors and transporters. Sirolimus
significantly suppressed LDL receptor (32% of control), VLDL receptor (56%), and
CD36 (56%) gene expression. Furthermore, sirolimus increasing expression of ABCA1
(420% of control) and enhanced efflux of radiolabelled cholesterol from HMCs. The
drug also up regulated peroxisome proliferator-activated receptor (PPAR)_ , PPAR_ and
liver X receptor_, intracellular ligand-activated transcription factors involved in the
regulation of genes that control of lipid homeostasis. Interestingly, Sirolimus overrode
the suppression of cholesterol efflux and ABCA1 gene expression induced by the
inflammatory cytokine IL-1_. In a separate study, we demonstrated that Sirolimus
inhibited production of the pro-inflammatory cytokine IL-6 by cells of the human THP1 monocytic cell line.
These results suggest that in addition to its anti-proliferative effect, Sirolimus may
inhibit intracellular cholesterol accumulation both by reducing uptake and increasing
export. The drug also appears to counteract some of the adverse effects of inflammatory
cytokines on cellular cholesterol homeostasis.
Renal Association October 2003
P 50
Anti-Atherosclerotic Effects of Sirolimus: Dual Actions on both Cholesterol
Homeostasis and Inflammation
Z Varghese, R Fernando, J Moorhead, S Powis, D Wheeler and X Ruan
Centre for Nephrology, Royal Free and UC Medical School, Rowland Hill Street,
London, NW3 2PF, United Kingdom
_The immunosuppressive agent Sirolimus reduces atherosclerotic lesions in apo E
knockout mice and has recently found clinical application in the context of drug eluting
stents. To determine the possible mechanisms underlying the anti-atherosclerotic
properties of this compound, we investigated its effects on cellular cholesterol
homeostasis in vitro. We have previously shown that the inflammatory cytokines TNF_
and IL-1_ cause lipid accumulation in human mesangial and vascular smooth muscle
cells both by increasing lipid uptake through LDL, VLDL and scavenger (CD 36)
receptors, and by reducing cholesterol efflux via the ATP binding cassette transporter
A1(ABCA1) pathway. Addition of Sirolimus (10-100 ng/ml) reduced lipid accumulation
in human mesangial cells (HMCs) incubated with IL-1_ (5ng/ml)as demonstrated by a
reduction in Oil red O staining of intracellular lipid droplets. Using real-time PCR, we
screened for mRNA expression of lipoprotein receptors and transporters. Sirolimus
significantly suppressed LDL receptor (32% of control), VLDL receptor (56%), and
CD36 (56%) gene expression. Furthermore, sirolimus increasing expression of ABCA1
(420% of control) and enhanced efflux of radiolabelled cholesterol from HMCs. The
drug also up regulated peroxisome proliferator-activated receptor (PPAR)_ , PPAR_ and
liver X receptor_, intracellular ligand-activated transcription factors involved in the
regulation of genes that control of lipid homeostasis. Interestingly, Sirolimus overrode
the suppression of cholesterol efflux and ABCA1 gene expression induced by the
inflammatory cytokine IL-1_. In a separate study, we demonstrated that Sirolimus
inhibited production of the pro-inflammatory cytokine IL-6 by cells of the human THP1 monocytic cell line.
These results suggest that in addition to its anti-proliferative effect, Sirolimus may
inhibit intracellular cholesterol accumulation both by reducing uptake and increasing
export. The drug also appears to counteract some of the adverse effects of inflammatory
cytokines on cellular cholesterol homeostasis.
Renal Association October 2003
P 51
Dysregulation of LDL Receptor by an Inflammatory Cytokine: A New Pathway
for Foam Cell Formation in Vascular Smooth Muscle Cells
X Ruan, J Moorhead, S Powis, R Fernando, D Wheeler and Z Varghese
Centre for Nephrology, Royal Free and UC Medical School, Rowland Hill Street,
London, NW3 2PF, United Kingdom
Inflammation within the arterial intima and foam cell formation involving both
macrophages and vascular smooth muscle cells (VSMCs) are characteristic early
features of atherosclerosis. Incubation of VSMCs with low-density lipoprotein (LDL)
does not normally lead to intracellular cholesterol accumulation because of feedback
down-regulation of the LDL receptor by raised intracellular cholesterol concentrations.
Using oil red O staining to visualise intracellular lipid and butylated hydroxytoluene to
prevent lipid peroxidation, we observed that VSMCs in primary culture formed foam
cells when incubated with interleukin-1_ (IL-1_ , 5 ng/ml) and unmodified LDL (200
_g/ml). This process was not inhibited by polyinosinic acid, which blocks scavenger
receptors, but was prevented by heparin, which displaces LDL bound to its cell surface
receptors. The electrophoretic mobility of LDL recovered from the culture medium was
the same as that of freshly isolated LDL, excluding the involvment of oxidised LDL via
scavenger receptor. To study the mechanisms, we used Real-Time RT-PCR to
demonstrate that IL-1_ overrode the suppression of LDL receptor mRNA induced by a
high concentration of LDL. In addition, exposure of cells to IL-1_ induced overexpression of the sterol responsive element binding protein (SREBP) cleavage
activation protein (SCAP), which is thought to mediate cholesterol-induced changes in
the expression of the LDL receptor. By co-staining cells with anti-human SCAP and
anti-human golgi antibodies, we demonstrated using confocal microscopy that exposure
to IL-1_ caused translocation of SCAP to the Golgi, where SCAP cleaves SREBP, and
that this occurred even in the presence of a high concentration of LDL (which would
normally inhibit this process).
Thus inflammatory cytokines modify cholesterol-mediated LDL receptor regulation in
VSMCs, permitting unregulated intracellular accumulation of LDL and causing foam
cell formation. This process appears to result from dysregulation of the SCAP-SREBP
pathway. These results provide a further possible mechanism whereby inflammation
may contribute to vascular disease.
Renal Association October 2003
P 52
A novel approach to estimate aortic distensibility in patients with Renal Failure
using a combination of Magnetic Resonance Imaging and Applanation Tonometry.
A Doyle1, N Johnstone2, S Cleland1, B Groening2, J Connell1, H Dargie2, A Jardine1 and
N Padmanabhan1
Division of Cardiovascular Sciences, Gardner Institute, Western Infirmary, Glasgow,
G11 6NT, United Kingdom and 2Research Initiative in Heart Failure, Department of
Cardiology, Western Infirmary, Glasgow, G11 6NT, United Kingdom
We describe a novel technique to estimate aortic elastic properties using cardiac
magnetic resonance imaging (CMRI) in combination with an estimation of the central
arterial pressure waveform by applanation tonometry.
10 Patients with End Stage Renal Disease(ESRD) and 8 control subjects underwent
CMRI. Transverse Images of the ascending aorta were obtained every 22ms after the
ECG R-wave. Contemporaneous assessments were made using Sphygmocor apparatus
to estimate the central aortic pressure waveform from the radial pulse waveform. Aortic
Distensibility(D) was calculated from both peripheral and estimated central pulse
pressure(PP) using the equation D= (Vmax-Vmin) / Vmin(Pmax-Pmin)mmHg-1 where
Vmax and Vmin are the maximum and minimum volumes of the aortic images
respectively and Pmax and Pmin are the systolic and diastolic pressures respectively
Using brachial PP, patients with ESRD had reduced D (0.001965mmHg-1) compared to
the control (0.004381mmHg-1). p = 0.0088. However, using estimated central pulse
pressure, greater definition between the two groups was achieved. D was
0.002679mmHg-1 in ESRD and 0.00688mmHg-1 in normal controls. (p = 0.00055).
The combination of CMRI and applanation tonometry provides a sensitive method of
directly estimating aortic distensibility. Using this method we have demonstrated a
marked reduction in aortic distensibility in patients with ESRD. The use of brachial
pressure underestimated the aortic distensibility in those with normal phenomenon of
pulse pressure amplification in the brachial artery. This technique may be applied to
study cardiovascular remodelling in a variety of pathological states.
Furthermore, by comparing the central pressure waveform with volume, velocity and
flow data from the aorta, further insight into the functional abnormalities of the aorta in
ESRF can be gained. (See figure1)
Figure 1
The plot shows the derived aortic pressure wave in pale blue and the mean velocity of
aortic blood in dark blue. In the ESRF subject, the second rise in pressure in the systolic
upstroke is the reflected pressure wave returning from the distal arterial tree. It can be
seen to have a negative effect on forward velocity in the aorta.
Renal Association October 2003
Renal Association October 2003
P 53
Long-term outcome after cardiac investigation in patients with end-stage renal
failure (ESRF) screened pre-transplantation
DYY Chiu1, P Atkinson2, C Ward2, RN Foley1, M Venning3, S Waldek1, DJ
O'Donoghue1 and PA Kalra1
Department of Renal Medicine, Hope Hospital, Stott Lane, Salford, M6 8HD, United
Kingdom, 2Department of Cardiology, Wythenshawe Hospital, Manchester, M23 9LT,
United Kingdom and 3Department of Renal Medicine, South Manchester University
Hospital, Manchester, M20 2LR, United Kingdom
Coronary artery disease (CAD) is a major cause of morbidity and mortality in patients
with ESRF, including those who receive renal transplants. Detection of CAD prior to
transplantation is important, but controversy surrounds the optimal approach to
investigation.The aims of this study were to determine the diagnostic power of
myocardial perfusion imaging (MPI) in detection of CAD in potential renal transplant
candidates and to assess the accuracy of both MPI and coronary angiography (CA) in
predicting cardiovascular outcome (cardiac death, myocardial infarction and angina)
and overall mortality.
All new dialysis patients presenting to one centre between 1995 and 1999, and who
were being considered for possible renal transplantation, were invited to participate in
the study. 70 patients consented and 47 patients (median age 51, range 20-71years, 37
males) underwent both MPI (dipyridamole-technetium myoview scan) and CA as part
of pre-transplant assessment. They were unselected, and 10 were diabetics. Prospective
follow-up investigated the relationship of the test results to later cardiovascular events
and mortality.
22(46.8%) patients had >50% stenosis of at least one major coronary artery (CAD), but
only 10 patients had abnormal MPI. The sensitivity of MPI to detect CAD was 41%,
specificity 96%, and positive (PPV) and negative (NPV) predictive values, 90% and
65%, respectively. Mean follow-up was 49±27 (range 3-92) months, during which 14 of
the patients received a transplant and 20 patients died (two transplanted patients). There
were 2 myocardial infarctions, 5 new onset angina and 16 patients had proven or
suspected cardiac deaths. 4 patients died from non-cardiac causes. CAD proven by CA
at screening had an identical PPV and NPV (68%) for the combined outcome of death
and cardiovascular events, whereas the PPV was 80% and NPV 60% for positive MPI.
Mean survival was 68 (CI 57, 78) vs. 41 (28, 54) months in patients with negative and
positive CA, respectively (P<0.05), and 63 (54, 72) vs. 25 (12, 38) months for negative
and positive MPI (P<0.05).
In conclusion, MPI proved to be an insensitive screening test for significant CAD.
However, the presence of either a positive MPI or CA at baseline both proved to be of
similar value as significant predictors of future mortality or cardiac events, which were
common in this high-risk group.
Renal Association October 2003
P 54
Electrocardiographic Findings In The Chronic Renal Insufficiency Population:
Baseline Results From CRISIS
CA Hogan1, RJ Middleton2, P MacFarlane3 and DJ O'Donoghue2
Medical School, Stopford Building, University of Manchester, Manchester, M13 9PT,
United Kingdom, 2Department of Renal Medicine, Hope Hospital, Stott Lane, Salford,
M6 8HD, United Kingdom and 3Department of Cardiovascular and Medical Sciences,
Glasgow Royal Infirmary, Glasgow, G32 2ER, United Kingdom
There is limited observational data on the association of cardiovascular disease (CVD)
and chronic renal insufficiency (CRI). The National Kidney Foundation states that CVD
mortality is more likely than developing end-stage renal failure (ESRF) in non-diabetics
with CRI. In particular, there is scant data with regards to the prevalence of
electrocardiographic (ECG) abnormalities within this population. The objective of this
study was to establish whether CRI is associated with an increased prevalence of ECG
abnormalities, independent of traditional cardiovascular risk factors.
The study was conducted in conjunction with the Chronic Renal Insufficiency Standards
Implementation Study (CRISIS), a larger prospective cohort study ongoing at a single
centre in the UK. 12 lead ECG’s were performed within 3 months of study recruitment
on all subjects. The presence of major and minor Minnesota code ECG abnormalities
was assessed in 189 subjects with an estimated MDRD-GFR between 10 and 60
The mean age was 65 years, with an MDRD-GFR of 28.1ml/min/1.73m_, 61.4% were
male. 95.8% white, 27% had Diabetes Mellitus and 71.4% were past or present
smokers. 29.1% had a history of ischaemic heart disease. 32% had a major Minnesota
anomaly and 27% a minor anomaly. 12.3% were found to have a major ventricular
conducting defect. 4.2% had atrial fibrillation. Major ST depression and T wave
anomalies were found in 8.5% and 12.7% respectively and minor ST and T wave
anomalies in 15.9% and 21.7%. Patients who proved positive for either a major or
minor anomaly were significantly more likely to have a history of a myocardial
infarction (MI) (25.7% vs. 9.2% p=0.012) or angina (23% vs. 11.8%, p=0.028). Patients
aged over 55 years were more likely to have an ECG abnormality (OR 9.14, p<0.05)
than their younger counterparts. Neither gender, ethnicity, smoking history, peripheral
vascular disease nor diabetes were associated with ECG documentation of an anomaly.
Similarly there was no significant association between ECG findings and GFR, BP,
cholesterol or statin use however there was a trend towards a positive association
between ECG anomalies and C-reactive protein (p=0.067).
The prevalence of ECG abnormalities is high in the setting of CRI. An older age, a
history of angina and a previous MI all proved significantly associated with an
abnormal ECG.
Renal Association October 2003
P 55
High serum calcium phosphate product is predictive of major vascular events in
patients with end stage renal disease.
A Trehan, J Winterbottom, B Lane, R Foley, M Venning, R Coward, A MacLeod and R
SIRS Group, Manchester Royall Infirmary, Manchester, M13 9WL, United Kingdom
Vascular disease is the leading cause of mortality and morbidity in people with end
stage renal disease (ESRD). Here we report the incidence, associations and predictors of
vascular morbidity and mortality. SIRS is an inception, cohort study of people with end
stage renal disease who started renal replacement therapy between April 2000 and April
2003 in the Northwest of England. Of 1109 people entered (mean age 57.6 years, 61%
male, 12.8% from ethnic minority groups, 26% diabetics, 31.1% with vascular disease
and 21% requiring dialysis within 1 month of referral to renal services) 126 (11%) had a
major vascular event. The events were 56 deaths and 92 hospital admissions (63 for
ischaemic heart disease, 15 for peripheral vascular disease and 14 for cerebrovascular
disease). The incidence of vascular events was 3% in first 90 days after inception of
dialysis, 5.4% in first year and 9.1% by 2 years. Vascular events were associated with
older age, previous history of vascular disease at inception of dialysis and diabetic
nephropathy as the cause of ESRD. Independent predictors of vascular events were high
serum phosphate (adjusted odds ratio (AOR) 1.2, p=0.06), high calcium phosphate
product (AOR 1.14, p=0.02), low haemoglobin (AOR 0.99, p=0.05) and high serum
cholesterol (AOR 1.2, p=0.06). Serum albumin, blood pressure and body mass index
were not independent predictors of vascular events. In conclusion, people with ESRD
are at high risk of vascular morbidity and mortality, largely due to their co-morbidity at
the inception of dialysis, but measures to control anaemia, serum phosphate and
cholesterol may help to reduce vascular events.
Renal Association October 2003
Moderated Poster Session - Two
Group (B)
Wednesday 8 October
Studio Suite
Renal Association October 2003
P 56
Identification of the source of plasma complement factor H by isoelectric focusing
of factor H polymorphisms in hepatic, renal and bone marrow allograft recipients.
J Stratton1, D Talbot2, N Moghal3 and P Warwicker1
Dept of Nephrology , Lister Hospital, Stevenage, SG1 4AB, United Kingdom, 2Dept of
Surgery, Freeman Hospital, Newcastle upon Tyne, NE1 4LP, United Kingdom and
Dept of Paediatric Nephrology, Royal Victoria Infirmary, Newcastle upon Tyne, NE1
4LP, United Kingdom
Factor H is the most important plasma bound modulator of the alternative pathway of
complement activation. Deficiency and dysfunction of complement factor H has been
implicated in the aetiology of non-diarrhoeal hemolytic uraemic syndrome. These,
predominantly young patients, are destined to a life of dialysis, with little prospect of
renal transplantation (because of a high chance of recurrence, often within the first two
weeks). A pioneering approach offers a potential solution. Two recent case reports have
described the use of liver transplantation in the treatment of refractory H-HUS.
In plasma, factor H circulates as a glycoprotein, and it has been postulated that it is
produced predominantly in the liver. We have developed an isoelectric focusing
technique to differentiate polymorphic forms of factor H. This was used this to establish
if the liver is indeed the predominant source of plasma factor H, by studying the pattern
of polymorphic forms of the protein before and after allogenic hepatic, renal and bone
marrow transplantation.
We tested pre and post transplant blood samples of 4 bone marrow, 6 renal transplants
and 12 liver transplants recipients. There was no evidence of a switch in the
polymorphic form of factor H in any of the bone marrow or renal transplant recipients
but there was a change in plasma factor H polymorphic form in 4 of the 12 liver
transplant recipients. Our data suggests that that the liver is the predominant source of
plasma factor H.
Renal Association October 2003
P 57
Immunodominant B and T cell epitope in experimental autoimmune
J Reynolds, J Haxby, J Juggapah and CD Pusey
Renal Section, Division of Medicine, Imperial College London, Hammersmith Hospital,
Du Cane Road, London, W12 0NN, United Kingdom
Previous studies have shown that experimental autoimmune glomerulonephritis (EAG),
an animal model of Goodpasture’s disease, can be induced in Wistar Kyoto (WKY) rats
by immunisation with collagenase-solubilised (cs) rat glomerular basement membrane
(GBM), or recombinant rat alpha 3 chain of type IV collagen (_3(IV)NC1). This results
in the development of circulating and deposited anti-GBM antibodies, together with
focal necrotising glomerulonephritis with crescent formation. In patients with
Goodpasture’s disease, the major B cell epitope is located at the N-terminus of
_3(IV)NC1, but the T cell epitope has not been defined. In order to investigate whether
B and T cell responses in EAG are directed towards immunodominant peptides within
the same region of rat _3(IV)NC1, we immunised WKY rats with csGBM (positive
control) and five 15-mer overlapping synthetic peptides from the N-terminus of rat
_3(IV)NC1. Positive control animals produced an antibody response directed towards
csGBM, recombinant rat _3(IV)NC1 and peptide 2. Splenic T cells from these animals
proliferated in response to csGBM, _3(IV)NC1 and peptide 2. No significant antibody
or T cell responses were observed to the other peptides examined. Animals immunised
with peptide 2 (n=6) developed circulating and deposited antibodies to the GBM,
albuminuria, and focal necrotising glomerulonephritis with crescent formation, by week
6 after immunisation. Circulating antibodies from these animals recognised csGBM,
_3(IV)NC1 and peptide 2, while their T cells proliferated in response to peptide 2.
Animals immunised with the other peptides (n=3) developed no significant immune
response to _3(IV)NC1 or GBM, and no disease. In conclusion, these results
demonstrate that a 15-mer peptide (peptide 2) from the N-terminus of _3(IV)NC1 is
recognised by B and T cells from rats immunised with csGBM, and that the same
peptide is capable of inducing crescentic glomerulonephritis. Identification of this
immunodominant peptide should be value in designing therapeutic strategies.
Renal Association October 2003
P 58
Operational Tolerance to Class I mismatched skin grafts following inhalation of
the immunodominant epitope.
M Harber1, C Burkhart2, M Day3 and D Wraith4
Ashford General Hospital, London Road., Ashford, TW15 3AA, United Kingdom,
Unit of Allergic and inflammatory Disease, Novartis Forschungs Institut, Wien, A1235, Austria, 3Langford Veterinary College, University of Bristol, Bristol, BS8 1TD,
United Kingdom and 4Department of Pathology and Microbiology, University Walk,
School of Medical Sciences, Bristol University., Bristol, BS8 1TD, United Kingdom
Mucosal administration of antigens has been shown to be a powerful method of
inducing antigen-specific tolerance. The mechanism of nasal tolerance has recently been
shown to be IL-10 dependent and occur via the generation of regulatory T cells.
Using a single Class I (Kb) mismatched murine model, pools of overlapping 15mer
peptides derived from the Kb molecule were screened for their immunogenicity. A
single epitope (peptide p164) was found to be immundominant and mice primed with
p164 demonstrated accelerated graft rejection.
Nasal inhalation of p164 reduced in vitro proliferation and cytokine production but
failed to prolong graft survival. However, in CD8+ depleted mice, intranasal p164
abrogated the accelerated rejection of p164 primed mice and surprisingly, resulted in
50% long term graft survival.
Inhalation of a single peptide prevented rejection by CD4+ cells in a stringent allograft
model. This approach offers the prospect of a non-invasive, antigen-specific method of
preventing allograft rejection.
Renal Association October 2003
Tr1 Regulatory T cells induced in vitro, suppress naive CD4+ cells via an IL-10
dependent mechanism involving suppression APC activation.
M Harber1, A Sundstedt2 and D Wraith2
Ashford General Hospital, London Road, Ashford, TW15 3AA, United Kingdom and
Department of Pathology and Microbiology, School of Medical Sciences, University
Walk., Bristol, BS8 1TD, United Kingdom
There is considerable data supporting the natural occurrence of innate and inducible
regulatory T cells. Tr1 regulatory T cells are a recently described novel inducible
regulatory cell which, unlike CD4+CD25+ (innate regulatory T cells) suppress in an
antigen-specific cytokine-dependent mechanism.
Using CD4+ T cells from the Tg4+ transgenic mouse Tr1 cells were generated in vitro
using repetitive peptide stimulation in the presence of either IL-10 or dexamethasone
and vitamin D3.
Tr1 cells generated this way were profoundly anergic, secreted negligible IL-2 but did
produce IL-10. Suppression of naïve cells was both IL-10 and Antigen Presenting Cell
(APC) dependent (suppression did not occur in the presence of CD3 and CD28 coated
latex beads).
Analysis of the effect of Tr1 cells on Dendritic cells showed a profound suppression of
Class II, CD80 and CD86 with down regulation of CD40. In addition, IL-12 production
by APCs was completely abrogated in an IL-10 dependent manner.
Tr1 cells have profoundly suppressive effects on naïve CD4+ T cells, this regulation is
IL-10 dependent and appears to occur by preventing the maturation of ‘professional’
Renal Association October 2003
P 60
Regulation of the innate immune response to ascending urinary tract infection by
renal tubular epithelial cells
P Chowdhury, SH Sacks and NS Sheerin
Department of Nephrology and Transplantation, 5th Floor Thomas Guy House, Guy's
Hospital, London, SE1 9RT, United Kingdom
As well as pathogen associated virulence factors,the host inflammatory response is a
major determinant of clinical outcome in pyelonephritis. Renal epithelial cells play a
key role in orchestrating this response at a local level.
Initiation of a response against pathogens occurs via innate immunity,understanding of
which has advanced with the discovery of toll-like receptors (TLRs). TLR4 has been
identified as the receptor for lipopolysaccharide (LPS),a component of the cell wall of
Gram negative bacteria. In archetypal cells of innate immunity,TLR4 forms a receptor
cluster with other molecules e.g. MD2, CD14 leading to recruitment of intracellular
adaptor molecules e.g. MyD88 and stimulation of intracellular signalling pathways
resulting in the production of pro-inflammatory cytokines.
We aim to show that renal epithelial cells,rather than merely providing a physical barrier
to infection,respond in a similar manner to other cells of innate immunity and play a
role in the development of an inflammatory response. Specifically,we are examining the
role of TLRs in this process.
Presence of the relevent molecular machinery,namely TLR4, CD14, MD2 and MyD88
was demonstrated by RT-PCR on RNA extracted from proximal tubular epithelial cells
(PTECs) of mice.
Monolayers of PTECs were grown and stimulated with LPS. This led to a 6 fold
increase in C3 and 3 fold increase in TNF-_ production compared with unstimulated
controls. Increases were seen in a time and dose dependent manner,with TNF-_
preceeding C3. Stimulation of PTECs from C3H/HeJ mice,with a spontaneous mutation
of the TLR4 gene,demonstrated a 50% reduction in C3 production and even greater
reduction in TNF-_ compared with C3H/HeN controls. Complete loss of response in the
HeJ mice was not seen,possibly due to contamination of LPS with other bacterial
products,or may suggest TLR4 is not an absolute requirement for LPS mediated
PTECs were grown on transwells and confluency of monolayers determined by
measuring trans-epithelial cell resistance. Stimulation via the upper apical cell surface
led to greater C3 and TNF-_ production compared with stimulation via the lower
basolateral surface,suggesting greater apical expression of TLR4. Apical stimulation
also demonstrated a polarised response,not seen with basolateral stimulation. This
would support a role for PTECs sampling luminal contents.
Renal Association October 2003
P 61
The role of p21ras in anti-neutrophil cytoplasm antibody activation of neutrophils.
JM Williams and COS Savage
Renal Immunobiology, The Medical School, University of Birmingham, Birmingham,
B15 2TT, United Kingdom
Anti-neutrophil cytoplasm antibodies (ANCA) are implicated in the pathogenesis of
systemic vasculitis. These antibodies are directed against antigens expressed on the
surface of cytokine-primed neutrophils, including proteinase 3 (PR3) and
myeloperoxidase (MPO). Intact ANCA IgG, but not its F(ab’)2 fragment, is able to
induce neutrophils to generate superoxide. This signifies that both antigen binding and
Fc_ receptor engagement are necessary for functional responses. Binding of ANCA to
neutrophils induces dual intracellular signalling pathways involving G proteins and
receptor tyrosine kinases. We have previously shown that whole IgG ANCA and its
F(ab’)2 fragment can activate the small GTPase p21ras. This molecule is the molecular
switch mediating cross talk between different intracellular events and could link the G
protein-coupled pathway to the Fc_ receptor-induced tyrosine kinase pathway. We now
show that p21ras activation by ANCA is functionally linked to the production of
superoxide. Additionally we investigated differential activation of isoforms of p21ras and
found that ANCA induced activation of both K and N ras.
Neutrophils were primed (2ng/ml TNF_ and 5_g/ml cytochalasin B, 15min 37oC) then
treated with 200_g/ml ANCA IgG. Superoxide production was assessed by the
superoxide dismutase inhibitable reduction of ferricytochrome C. After 120min
unstimulated cells had a basal production of 1.59±0.26, normal IgG gave 1.34±0.67
whereas ANCA gave 8.09±1.1 (nmols superoxide/105 cells). Neutrophils were next
pretreated with farnesylthiosalicylic acid (FTS, specific p21ras inhibitor) at various
concentrations for 2h at 37oC prior to priming and stimulation with ANCA. A dosedependent decrease in superoxide production was observed with 100_M giving
0.55±0.35, 50_M 0.61±0.28, 25_M 2.55±1.2, 12.5_M 4.42±1.6 and 6.25_M 6.58±2.3
nmols superoxide/105 cells. Using the ras activation assay it could be seen that FTS
inhibited p21ras binding to its substrate at comparable concentrations to those seen for
superoxide inhibition. Additionally, using the ras activation assay, ANCA was able to
activate both isoforms of p21ras present in neutrophils (K and N).
We conclude that the activation of p21ras is functionally linked to the ANCA-induced
production of superoxide by neutrophils and that inhibition of p21ras may be a valuable
therapeutic approach for active ANCA-associated vasculitis.
Renal Association October 2003
P 62
Neutrophil activation by anti-neutrophil cytoplasm autoantibodies (ANCA): the
role of tyrosine kinases
MRC Centre for Immune Regulation, University of Birmingham, Birmingham, B15
2TT, United Kingdom
Neutrophils are primary mediators of early damage in vasculitic glomerulonephritis,
making the mechanisms underlying neutrophil activation by ANCA relevant to our
understanding of disease pathogenesis. ANCA activate the neutrophil respiratory burst
via Fc gamma receptors(FcgR). Moreover the respiratory burst can be attenuated by
tyrosine kinase inhibitors(TKI). We hypothesized that ANCA recruit Syk tyrosine kinase
since it is intimately involved in FcgR signaling. By western blotting(WB), intact
ANCA-IgG stimulate tyrosine phosphorylation of neutrophil proteins whereas ANCAF(ab’)2 do not. By immunoprecipitation and WB, ANCA-IgG but not ANCA-F(ab’)2 or
normal-IgG are shown to tyrosine phosphorylate Syk, indicative of its activation.
ANCA induced Syk phosphorylation is reduced by blocking either FcgRIIa or RIIIb.
Furthermore, blocking CD18 diminishes Syk phosphorylation, indicating beta 2 integrin
involvement in ANCA signaling. ANCA also induce tyrosine phosphorylation of the
adapter protein Cbl. The Src TKI, PP2 (10 _mol/L) inhibits ANCA stimulated
superoxide release (8.6 ± 0.35nmol SO uninhibited vs 0.65 ± 0.36nmol SO with PP2)
and abolishes ANCA induced Syk and Cbl phosphorylation. Moreover, ANCA induced
Syk phosphorylation is attenuated by piceatannol (40 _mol/L), which is relative specific
inhibitor of Syk at this concentration. Piceatannol also mediates a dose dependent
decrease in ANCA induced respiratory burst. In contrast, Syk phosphorylation induced
by cross-linking anti-FcgRIIa antibody is unaffected by piceatannol, suggesting that
ANCA signaling differs from conventional FcgR signaling. ANCA also stimulate
intracellular calcium ([Ca]i) fluxes. In FURA-2/AM loaded neutrophils, ANCA induced
[Ca]i fluxes are sensitive to PP2, suggesting that they require tyrosine kinase dependent
phospholipase C gamma. However, ANCA induced calcium fluxes are resistant to
pertussis toxin despite the fact that ANCA induced SO release involves Gi/0 GTPases.
In conclusion, Src and Syk kinases have obligate roles in ANCA activation of
neutrophils, which occurs via FcgRIIa, FcgRIIIb and CD18. Simultaneous ligation of
FcgR and ANCA target antigens may account for the differential sensitivity of ANCA
signaling to specific tyrosine kinase inhibitors, compared to conventional Fc gamma
receptor clustering.
Renal Association October 2003
P 63
IL-18 is important in important in driving neutrophil responses in ANCA
associated vasculitic glomerulonephritis
Divisional Medical Sciences, Medical School, University of Birmingham, Birmingham,
B15 2TT, United Kingdom
Anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis is associated with
primed neutrophils (PMN) which have increased expression of the target antigens
proteinase 3 (PR3) and myeloperoxidase (MPO). Interleukin (IL)-18 has potent proinflammatory effects in rheumatoid arthritis promoting PMN chemotaxis and activation.
We aimed to investigate the role of IL-18 in ANCA associated vasculitis. Biopsies
isolated from patients with active vasculitis or normal controls were stained for IL-18
by immunohistochemistry. Serum levels of IL-18 were measured by ELISA. Isolated
PMN were primed with IL-18 or TNF and superoxide production by PMN following
activation by ANCA was measured by ferricytochrome c reduction. Surface expression
of PR3 and MPO was assessed by FACS. The role of p38 MAP kinase was investigated
using the inhibitor SB202190. Patients with ANCA-vasculitis showed strong staining of
IL-18 within the glomerulus and interstitium by both resident and infiltrating cells
however serum IL-18 levels were not elevated in those patients with vasculitis
compared with normal controls (p=0.25). IL-18 amplified the superoxide response to
both PR3-ANCA (control 1.63+/- 0.5, IL-18 primed 8 +/- 1.7) and MPO-ANCA
(control 1.8 +/- 0.65, IL-18 primed 3.9 +/- 1.02) after 5 mins. There was no difference in
the response by PMN to ANCA using IL-18 or TNF as priming agents (p=0.54).
Prolonged exposure of PMN to IL-18 resulted in blunting of the response. IL-18
increased surface expression of both PR3 (control 51% +/- 7.59, IL-18 primed 63% +/4.7) and MPO (control 12.1% +/- 2.75; IL-18 primed 28.2% +/- 3.8) on PMN. P38
MAP kinase inhibition reduced PR3 and MPO-ANCA induced superoxide production
by 81.2% +/- 3.69 and 78.2% +/- 12.03 respectively. We conclude that IL-18 has a
pivotal role in ANCA associated vasculitis through PMN recruitment and priming the
inflammatory response to ANCA.
Renal Association October 2003
P 64
An active role for complement inhibitors in the development of an inflammatory
PS Tsang, SH Sacks and NS Sheerin
Department of Nephrology and Trasnplantation, 5th Floor, Thomas Guy House, Guy's
Hospital, London, SE1 9RT, United Kingdom
Excessive activation of the complement system can lead to tissue injury as seen in
several autoimmune diseases. A series of fluid phase and cell surface inhibitors
therefore exist to prevent complement activation. In human several cell surface
complement inhibitors, including Membrane Cofactor Protein (CD46), are expressed
extensively on most cell types. Many pathogens, including Dr fimbriae expressing
uropathogenic E coli, exploit these complement inhibitors in order to increase
pathogenicity. In contrast these inhibitors are able to augment immune responses from T
cells and macrophages. We have investigated the potential of CD46 to modify renal
epithelial cell function in response to infection.
CD46 was demonstrated on the epithelium in normal human kidney. Confocal
microscopy was used to localise CD46 on cultured tubular epithelial cells. Both apical
and basolateral expression were demonstrated. Cross-linking of CD46 with either
monoclonal or polyclonal antibody resulted in the rapid phosphorylation of several
intracellular proteins of approximately 80, 130 and 150kDa, including the adaptor
molecule cortactin. The cytoplasmic tail of CD46 has no intrinsic kinase activity.
However dimerisation of CD46 leads to activation of Src family kinases, as PP2 a Src
family inhibitor blocks the cytoplasmic phosphorylation events. Cross linking of CD46
leads to an increase in the epithelial cell production of neutrophil chemoattractant IL8.
Up-regulation of this cytokine is inhibited by PP2 confirming the importance of Src
kinases in epithelial cell signalling.
The interaction between the renal epithelium and pathogenic bacteria is vital in
determining the outcome after exposure to pathogens. These pathogens are readily
opsonised by complement proteins. Although complement receptors with the capability
to bind these opsonised pathogens are well described on leucocytes they are not present
on renal epithelial cells. Here we present data demonstrating that the complement
inhibitor CD46 is able to signal to the epithelium and in doing so promote a proinflammatory response. It therefore may serve as an epithelial C3 receptor.
Renal Association October 2003
Moderated Poster Session - Two
Group (C)
General Nephrology
Wednesday 8 October
Garner Room
Renal Association October 2003
P 65
The induction and maintenance of remission of the nephrotic syndrome and
preservation of renal function by Tacrolimus
NDC Duncan, A Dhaygude, T Cairns, M Griffith, A McLean, A Palmer and D Taube
Renal and Transplant Unit, St Mary's Hospital, Praed St, London., W2 1NY, United
There are a limited number of uncontrolled and randomized control studies reporting the
successful use of Cyclosporine (CyA) in the treatment of the nephrotic syndrome (NS).
Tacrolimus (FK) is a potent immunosuppressive agent which at therapeutic dose has a
neutral effect on renal haemodynamics and systemic blood pressure in comparison to
CyA, which limit its nephrotoxicity.
We have therefore used FK to treat 10 patients with NS (defined as serum albumin
(SAlb) <35g/dl with 24 hour urinary protein (24UP) excretion >3g). 6 had idiopathic
membranous nephropathy (IMN), 1 with CyA-resistance; 2 focal segmental
glomerulosclerosis (FSGS) and 2 minimal change nephropathy, both with steroidresistance. Mean age 55.2±14.9 years, mean 25.7±3.8 SAlb g/dl, mean 24UP 6.3±2.0 g.
Mean FK dose 0.06±0.01 mg/kg/d, mean 12 hour trough level 7.0±2.9 ng/ml. 5/10
achieved remission defined by normalisation of their SAlb with reduction in 24UP in
9/10. Mean time to remission 4.5±2.3 months with a short period of follow-up so far
(mean 6.6 months, range 1.3-12.7). None have relapsed.
5 patients (4FSGS and 1 membranous) were CyA-dependent but had declining renal
function and were converted to FK. Mean FK dose 0.1±0.05 mg/kg/d, mean 12 hour
trough 6.7±2.7. Mean follow-upwas 8.7 months (range3.1-20.2). Renal function
improved in all 5 conversions from CyA with mean change in estimated creatinine
clearance of +3.8±3.2 ml/min/month. There was only one relapse associated with low
FK levels, responding to increased FK dose and short course oral steroid.
A further 2 patients with steroid-dependent MCN were in established remission and
treatment with FK allowed steroid withdrawal without relapse.
Preliminary results therefore suggest FK alone is effective at inducing remission with no
significant side-effects encountered to date. It maintains remission and is not associated
with the nephrotoxicity seen with CyA therapy, and allows steroid withdrawal.
Renal Association October 2003
P 66
Patients with early renal dysfunction – health gain achieved from timely referral.
DN Bennett-Jones and T Rose
Cumberland Infirmary, Newtown Road, Carlisle, CA2 7HY, United Kingdom
It has been suggested that patients with urinary protein excretion (UPE) > 1 g/24hr or
creatinine (SeCr) >150_mol/l should be referred for a specialist opinion. We identified
all patients referred to a renal department with isolated proteinuria (SeCr 0-149 _mol/l,
UPE >= 1 g/24hr -Gp 1, n=214) or mild/moderate renal impairment (SeCr 150-299
_mol/l -Gp 2, n=527). For comparison, we also identified patients referred to the renal
department who did not meet the specified criteria for referral (SeCr <= 149 _mol/l,
UPE < 1g/24 hr -Gp C, n=600). Clinical and laboratory parameters were measured
during routine outpatient follow-up. Table 1 shows patient characteristics at entry to the
study and the actuarial probability of RRT or death within 5 yrs. Estimates of GFR were
made using the MDRD formula, and the rate of decline of renal function was
determined for those patient who had a minimum of 4 measurements in years 1-2 and
years 3-4 following referral.
In 129 patients with initial SeCr <200 _mol/l, there was a significant fall in Hb (12.64 to
12.11 g/dl, p<0.001), and increase in serum phosphate (1.17 to 1.28 mmol/l, p<0.001) as
the SeCr rose from a mean value of 176 to 214 _mol/l (student paired t test). 10/129
patients with SeCr 150-199 _mol/l and 22/129 with SeCr 200-249 _mol/l had
Hb<10g/dl. Progression to RRT or death within 5 years was noted even in patients with
mild renal dysfunction at referral. After 2 years treatment a significant improvement
was achieved in SBP (159.5 v 145.4 mm Hg, n=65, p<0.001), DBP (85.8 v 80.3 mm
Hg, n=65, p=0.001), UPE (3.2 v 2.1 g/24hr, n=64, p<0.00) and rate of decline of GFR (0.38 v –0.21 ml/min/mth, n=67, p<0.01). These findings suggest that there is much to be
gained from early referral to a renal department for specialist assessment and follow-up
of selected patients.
Renal Association October 2003
P 67
Prospective experience of 5-Aminosalicylate associated nephrotoxicity in the UK
P E Stevens1, R F Logan3, A S McIntyre4 and A F Muller2
Dept of Renal Medicine, Kent and Canterbury Hospital, Ethelbert Road, Canterbury,
CT1 3NG, United Kingdom, 2Dept of Gastroenterology, Kent and Canterbury Hospital,
Ethelbert Road, Canterbury, CT1 3NG, United Kingdom, 3Department of
Epidemiology, University Hospital of Nottingham, Nottingham, NG7 2UH, United
Kingdom and 4Department of Gastroenterology, Wycombe Hospital, High Wycombe,
HP11 2TT, United Kingdom
The purpose of this study was to describe the epidemiology of 5-ASA associated
nephrotoxicity. UK gastroenterologists and nephrologists were prospectively surveyed
by postal questionnaire for 2 years to establish their collective experience of this
complication of inflammatory bowel disease (IBD). Basic demographic information was
collected together with all treatment (Rx) details and type of IBD. Duration of 5-ASA
therapy prior to diagnosis of nephrotoxicity and details of renal function (pre-5-ASA
Rx, at diagnosis and after recovery) were sought. Where renal biopsy was performed the
histology was recorded. Cases reported by gastroenterologists and nephrologists were
cross referenced to avoid duplication.
60 cases were reported, median age 47 yrs, 50 male, 31 UC, 16 Crohns and 7
indeterminate IBD. Mesalazine was the commonest agent (40 cases), mirroring sales
data. Median time to diagnosis from starting drug Rx was 41 mths (1-96). Median
creatinine pre-Rx was 96 _mol/L (range 66-139). Median peak creatinine was 227
_mol/L (range 92-1361) and median recovery creatinine 142 _mol/L (range 78-ESRF).
In 19 cases renal biopsy was undertaken, 16 showed appearances of tubulointerstitial
nephritis (TIN), 1 end stage kidney, 1 FSGS and 1 MCGN + TIN. Recovery of renal
function was more likely in those diagnosed ≤12 mths after starting 5-ASA (n=10,
median recovery SCr 107 _mol/L) vs. those diagnosed > 12 mths after starting 5-ASA
(n=35, median recovery SCr 165 _mol/L. 4 patients did not recover renal function (1
died on dialysis), 24 were left with significant renal impairment (SCr ≥150 _mol/L).
Nephrotoxicity is a significant complication of 5-ASA therapy for IBD. It may present
after several months of treatment and is manifest by an indolent TIN that can result in
marked residual renal damage. This complication is readily reversible if diagnosed early
(within 12 months of start of treatment) and dictates the need for regluar monitoring of
renal function during therapy with these agents. Educate your gastroenterologists!
Renal Association October 2003
The Asian Population Is Under Represented In UK Chronic Kidney Disease
Clinics: Baseline Results From The Chronic Renal Insufficiency Standards
Implementation Study (CRISIS)
RJ Middleton1, RN Foley2, PA Kalra1, J New3, S Waldek1, GN Wood1 and DJ
Department of Renal Medicine, Hope Hospital, Stott Lane, Salford, M6 8HD, United
Kingdom, 2Nephrology Analytical Services Minneapolis Medical Research Foundation,
Minneapolis, 55404, USA and 3Department of Diabetes and Endocrinology, Hope
Hospital, Stott Lane, Salford, M6 8HD, United Kingdom
The Chronic Renal Insufficiency Standards Implementation Study (CRISIS) is a
prospective study initiated to investigate the influence of modifiable and non-modifiable
risk factors on progression of renal and cardiovascular disease in incident and prevalent
patients with chronic kidney disease (CKD) attending renal services at a single centre in
the UK. The Asian population have a 3 times higher acceptance rate onto renal
replacement therapy (RRT) programmes in the UK. There is a lack of data on the
epidemiology of CKD in ethnic minorities.
Data is presented on the first 354 subjects randomly recruited with an estimated GFR
between 10 and 60ml/min.
4.8 % are Asian and 1.7% Black. Asians are significantly younger than non-Asians
(53yrs vs 64.5yrs p=0.001). Comparison of the whole group revealed no difference in
the prevalence of diabetes or vascular disease. Results after sub-dividing the group
based on age greater or less than 55 yrs are shown in table 1.
Asians >55yrs old were more likely to be diabetic (OR 14.7, p<0.05), on EPO (OR 8.8,
p<0.01). There was no difference in haemoglobin or blood pressure. The younger age
echoes the dialysis population. However within this region of the UK 11% of dialysis
patients are of Asian origin compared to 4.8% in our study of chronic kidney disease.
Prospective data from CRISIS will identify whether this difference is due to inequalities
in referral to renal services, late referral, more rapid progression of CKD or a survival
advantage of the Asian population with CKD progressing to ESRF.
Renal Association October 2003
P 69
JB Eastwood1, L Emmett2, J Plange-Rhule3, SM Kerry2, FB Micah3 and FP Cappuccio2
Department of Renal Medicine, St George's Hospital, Blackshaw Road, London, SW17
0QT, United Kingdom, 2Department of Community Health Sciences, St George's
Hospital Medical School, London, SW17 0RE, United Kingdom and 3Department of
Medicine, Komfo Anokye Teaching Hospital, P.O.Box l934, Kumasi, Ghana
Background. Hypertension and renal failure are important threats to the health of adults
in sub-Saharan Africa. In a Teaching Hospital population in the Ashanti Region of
Ghana there is a close correlation between blood pressure (BP) and serum creatinine
(SCr). Whether or not this relationship holds in an unselected population in the same
region is not known.
Objective. To assess the relationship between BP and SCr in adult men and women in
12 village populations in Ashanti, Ghana, and to explore male:female differences as
well as differences between semi-urban and rural villages.
Design and Methods. 1,013 adult men and women aged between 40 and 75 were
screened as part of a community-based trial of health promotion in villages near
Kumasi. The individuals were identified through a household survey and population
census of 16,965 inhabitants of 12 villages. The participants, who were studied in their
villages, underwent measurements of height, weight, body mass index (BMI), BP (by
OMRON HEM705CP) and in 1007 a sample for SCr was obtained.
Results. The mean age was 54.8 (SD 11.3) yr. In men BP was 126/76 (25/14) mmHg,
BMI 20.2 (3.1) kg/m2, SCr 89 (18) umol/l. In women BP was 125/73 (27/13), BMI 21.6
(4.6) and SCr 73 (13). Individuals in semi-urban villages were heavier (BMI 22.3[4.6] v
19.8[3.2]; p<0.001), and had higher BP (129/76 [26/14] v 121/72[25/13]; p<0.001 for
SBP and DBP) than those from rural villages. The Table shows associations between BP
and SCr after adjustments for age, BMI, gender and village type.
Table 1
Conclusions. In an unselected sample of villagers in the Ashanti region of Ghana higher
BP levels were associated with higher SCr levels. The association was independent of
age and BMI, and appeared weaker than that found in a Hospital population. In view of
the linear association and the relative lack of high SCr, the results are likely to reflect a
deleterious effect of elevated BP on the kidney rather than primary renal cause.
Renal Association October 2003
P 70
Population estimates of prevalent renal impairment in ambulant older women in
England and Wales
D de Takats1 and EV McCloskey2
Department of Nephrology, University Hospital of North Staffordshire NHS Trust,
Princes Road, STOKE-ON-TRENT, ST4 7LN, United Kingdom and 2University of
Sheffield Metabolic Bone Centre, Sorby Wing, Northern General Hospital, Herries
Road, SHEFFIELD, S5 7AU, United Kingdom
We applied the Cockcroft & Gault (C&G) and MDRD formulas to 5,203 Caucasian
women aged ≥75yrs recruited in central England (1996-99). Since the sample comprised
those subjects remaining after exclusions for malignancy, leucopoenia, nonhyperparathyroid hypercalcaemia, increased bone turnover and serum creatinine
>300_mol/l, who were able to travel for assessment, it represents a fitter group than the
general population. The sample ranged 75-100 years (mean 80yrs, median 78.9yrs). The
age distribution differed from the general population in having more younger and less
older women, reflecting a bias towards healthier individuals. The sample represents
16.7% of those approached, and 0.2% of the female population ≥ 75yrs of England and
These formulas, thought to be superior to creatinine clearances, have previously had
their relationship with 51Cr-EDTA GFR evaluated by others in similarly elderly people1.
Applying them to our sample confirmed that, though correlated (r2=0.34), C&G gives a
value on average 11 ml/min lower than the MDRD formula. Differences between the
two methods are less marked at lower calculated function.
We then extrapolated the prevalence of different levels of renal function given by each
formula within age bands in our sample to adjusted Office of National Statistics 2001
Census population figures for England and Wales to derive conservative maximum and
minimum estimates for the numbers within age bands likely to have renal impairment of
various degrees. (See table below.)
These results may raise questions concerning the appropriateness of current reference
ranges for renal function in older women and whether to replace serum creatinine with
calculated renal function routinely in elderly care  they may also be of use in service
Lamb EJ, Webb MC, Simpson DE, Coakley AJ, Newman DJ and O’Riordan SE
Estimating kidney function in older patients with renal insufficiency: Is the MDRD
formula an improvement? Presented at the Spring 2003 Renal Association, Keele
Renal Association October 2003
P 71
Two-year survival among patients with chronic kidney disease not on dialysis: The
Chronic Renal Impairment in Birmingham (CRIB) study
SL Nuttall1, MJ Landray2, JN Townend1, S Ball1 and DC Wheeler3
Division of Medical Sciences, University of Birmingham, Birmingham, B15 2TH,
United Kingdom, 2Clinical Trial Service Unit, University of Oxford, Oxford, OX2 6HE,
United Kingdom and 3Centre for Nephrology, Royal Free and University College
Medical School, London, NW3 2PF, United Kingdom
There is little prospective outcome data among patients with chronic kidney disease not
requiring renal replacement therapy. Starting in December 1997, a cohort of 386
patients attending a nephrology programme with an elevated serum creatinine >130
_mol/L were recruited into the CRIB study. Baseline demographic data were recorded
and plasma samples stored for detailed laboratory analyses. 2-year data collection is
now complete for all patients and all-cause mortality is reported.
At baseline, the mean age was 61 years and 65% were male. 36% had a history of overt
vascular disease (including 15% with a previous myocardial infarction) and 16% had a
history of diabetes mellitus.
2-year mortality among this group of patients was 12.8%. Patients who died were
significantly older than those who survived (mean age 68.4 vs 59.8 years, p<0.001) and
were more likely to have a history of previous vascular disease (49% vs 34%, p=0.04)
but there were no significant differences in sex, history of diabetes, serum creatinine at
screening, or serum lipid levels. However, plasma homocysteine was somewhat higher
among those who died (mean 22.4 vs 20.0 _mol/L, p=0.08).
The observed all-cause mortality rate is substantially higher than that which would be
expected among individuals without chronic kidney disease. In this study, over onethird of patients with chronic kidney disease had a history of vascular disease, and this
was more common at baseline among those who died during the following 2 years.
Renal Association October 2003
Moderated Poster Session - Two
Group (D)
Cell Biology of the Gromerulus
Wednesday 8 October
Garner Room
Renal Association October 2003
P 72
Dysregulation of Heparanase in Steroid Sensitive Nephrotic Syndrome
RCL Holt1, SA Ralph2, J Davies2, WR Lamb2, NJA Webb1 and PEC Brenchley2
Dept of Nephrology, Royal Manchester Children's Hospital, Pendlebury, Manchester,
M27 4HA, United Kingdom and 2Manchester Institute of Nephrology &
Transplantation, Manchester Royal Infirmary, Manchester, M13 9WL, United Kingdom
Heparanase (HPSE) is the only known endoglycosidase capable of degrading
extracellular heparan sulphate glycosaminoglycan (HSGAG). The enzyme has a
physiological role in leukocyte migration and is expressed by peripheral blood
mononuclear cells (PBMC). Glomerular HSGAG forms the major charge barrier
restricting albumin filtration and reduced HSGAG quantity has been implicated in
development of proteinuria. We have tested the hypothesis that dysregulated HPSE
expression underlies the pathophysiology of childhood-onset steroid sensitive nephrotic
syndrome (SSNS).
28 SSNS patients were studied in remission and 14 in the proteinuric phase of disease;
none had received corticosteroids in the preceding two weeks or other
immunosuppressive therapy in the six months prior to sampling. 24 healthy subjects
served as age- & gender-matched controls. HPSE activity (HA) was quantified in
plasma and urine using a published solid-phase HSGAG degradation assay. HPSE
mRNA expression by PBMC was quantified by real-time RT-PCR.
Data are given as median values with interquartile range (IR). Plasma and urine HA
levels differed significantly between proteinuric phase, remission and controls (p=0.000,
Kruskal-Wallis). Proteinuric patients exhibited significantly lower plasma HA (811
u/ml, IR 479-1092 u/ml) than remission patients (1148 u/ml, IR 1059-1215 u/ml;
p=0.003, Mann-Whitney); remission plasma HA was also significantly lower than
control values (1391 u/ml, IR 1071-1628 u/ml; p=0.012, MW). In contrast, urine HA
was significantly higher in proteinuric patients (14.26 u/mg creatinine, IR 8.37-17.60
u/mg) than remission patients (7.43 u/mg, IR 5.06-11.48 u/mg; p=0.016, MW);
remission urine HA was also significantly higher than control values (2.29 u/mg, IR
1.67-3.32 u/mg; p=0.000, MW). There was no correlation between plasma and urine
HA levels in proteinuric patients (r = 0.000, p = 1.000, Spearman, 2-tailed). No
significant differences were observed between the study groups in HPSE mRNA
expression by PBMC (proteinuric phase: 2.49 arbitrary units, IR 1.41-3.45 au;
remission: 2.04 au, IR 1.60-2.36 au; control: 2.40 au, IR 1.81-3.93 au; p=0.156, KW).
In summary, plasma and urine HA were abnormal in the remission group and markedly
abnormal in the proteinuric group. The functional significance of reduced plasma HA is
unknown. Elevated urine HA, persisting in confirmed proteinuria-free remission, may
suggest an intra-renal source of dysregulated HPSE in SSNS.
Renal Association October 2003
P 73
Targeting T-type Calcium Channels in Human Mesangial Cell Proliferation
C J Mulgrew1, M J Shattock2, G Brooks3 and B M Hendry1
Department of Renal Medicine, GKT School of Medicine, King's College London,
London, SE5 9PJ, United Kingdom, 2Centre for Cardiovascular Biology & Medicine,
The Rayne Institute, St Thomas' Hospital, King's College London, London, SE1 7EH,
United Kingdom and 3School of Animal and Microbial Sciences, University of Reading,
Whiteknights, Reading, RG1 6AJ, United Kingdom
Mesangial cell proliferation is a characteristic feature of many glomerular diseases.
Novel therapies directed at this process would be a significant advance in the
management of renal disease. A role for low voltage-activated, T-type calcium channels
(T-CaCN) has been proposed in the growth and cell-cycle regulation of a number of
excitable and non-excitable cells, including vascular smooth muscle cells. Our study
investigates the effects of inhibitors of T-type (Mibefradil and TH-1177) and L-type
(Verapamil) calcium channels on the proliferation of human mesangial cells (MC) in
culture, and the expression of T-CaCN isoforms in quiescent and proliferating MC. MC
were cultured in RPMI-1640 supplemented with 10% fetal calf serum (FCS). Cell
proliferation was measured using the microculture tetrazolium (MTS) assay. After
serum-starving in 0.2% FCS for 48 hours, cells were seeded into 96-well plates at a
density of 5000 cells/well and were incubated with:(i) Mibefradil (0-15_M), (ii) TH1177 (0-15_M), or (iii) Verapamil (0-20_M). MC proliferation was stimulated by 10%
FCS. Cell number was measured at 24, 48 and 72 hr. RT-PCR was used to determine
which isoform(s) of the T-CaCN _1 subunit was predominant in proliferating MC. Cells
were cultured as above to 70-80% confluence, and total RNA extracted. The FirstStrand RT reaction was performed using 2_g of total RNA. For PCR amplification,
primers directed towards sequences of _1-G, -H and -I and the _1-H _25 splice variant
were used. Finally, in order to demonstrate an effect of serum-stimulation on T-CaCN
expression in MC, cells were serum-starved in 0.2% FCS for 72hr. RNA was then
extracted after the cells were exposed to 20% FCS for 0min / 30min / 4hr. T-CaCN
isoform expression was demonstrated by RT-PCR. Serum-stimulated MC proliferation
was significantly reduced by Mibefradil and TH-1177 by over 80% at 72hr, while
Verapamil had no effect on cell number. MC T-CaCN expression was predominantly the
_1-H _25 variant; no other T-CaCN isoforms were identified. T-CaCN expression in
MC significantly increased in a time-dependent manner upon serum-stimulation, as
demonstrated by RT-PCR. We conclude that expression of T-CaCN mRNA is upregulated in serum-stimulated MC, and that T-CaCN inhibitors significantly reduce MC
proliferation in vitro and thus may be of therapeutic use in proliferative renal disease.
Renal Association October 2003
P 74
Generation and Characterisation of Conditionally Immortalised Human
Glomerular Endothelial Cell Lines.
A Singh1, SC Satchell1, L Ni1, TJ Van der Velden2, MA Saleem1, L Van den Heuvel2
and PW Mathieson1
Academic Renal Unit, University of Bristol, Southmead Hospital, BRISTOL, BS10
5NB, United Kingdom and 2Department of Paediatrics, University Medical Centre
Nijmegan, Nijmegan, Netherlands
Glomerular endothelial cells (GEnC), podocytes and the GBM they produce form the
glomerular filtration barrier and hence have a crucial role in glomerular function.
Human GEnC have been little studied not least because of difficulties in maintenance of
primary cultures. Our group has addressed similar problems with podocytes by
generation of conditionally immortalised (ci) cell lines (JASN 2002; 13: 630-638) and
we have now applied the same technology to GEnC.
Primary culture GEnC were derived from glomeruli isolated from human renal cortex
using medium optimised for EnC culture. Proliferating cells were exposed to separate
retroviral vectors transducing a temperature sensitive mutant of SV40 large T antigen
(tsSV40LT) and the catalytic subunit of human telomerase (hTERT). Both these
elements are required for successful immortalisation of EnC. At the permissive
temperature of 33_C the tsSV40LT transgene is activated causing cell proliferation
(without telomere shortening) while at 37_C the transgene is inactive. Successfully
transfected cells were selected by sequential antibiotics and cloned by dilution
techniques. Clones were selected for further study by homogeneity and morphological
similarity to primary culture cells and were characterised by expression of EnC-specific
markers and SV40LT by immunofluorescence and western blotting (at both 33_C and
37_C) and by typical responses to TNF_.
CiGEnC at 33_C retained morphological features of early passage (p) cells at least to
p25 while primary culture cells senesced after p8. At 37_C the cells stopped dividing by
day 4 but maintained healthy appearances with almost complete loss of SV40LT antigen
expression by day 10. At both temperatures and at late passages expression of
endothelial markers remained comparable with primary culture GEnC and there was
upregulation of e-selectin in response to TNF_.
These novel ciGEnC represent a unique resource to generate large numbers of cells for
study and will enable more detailed understanding of the role of GEnC in the
Renal Association October 2003
P 75
A new urinary space within the glomerulus? 3-D reconstruction from Electron
micrograph reveals a significant restrictive sub-podocyte space.
CR Neal, DO Bates and SJ Harper
Microvascular Research Laboratories, Department of Physiology, Pre-clinical
Veterinary School, Southwell St, Bristol, BS2 8EJ, United Kingdom
The current understanding of glomerular ultra-filtration suggests there is little resistance
to flow after the filtrate has traversed the glomerular filtration barrier (GFB) since it has
direct access to Bowman’s space (BS).
We have reconstructed podocytes and underlying basement membrane (GBM) in threedimensions (3D) using electron micrographs of ultra-thin (100nm) serial sections of rat
and human kidney. Micrographs of serial sections were aligned and analysed using
Adobe photoshop/NIH image and 3D models made. An initial survey of the GFB in rats
revealed that while the majority of filtration slits between foot processes open directly
into the BS (66±7%, range 20–80%), the remainder open into the space between the
podocyte cell body and GBM. Full reconstruction of podocyte and sub-podocyte space
(SPS) showed narrow pores (0.6±0.2_m diameter, n=5) or channels (0.16±0.03_m wide,
n=6) connecting the SPS to BS. These pores/channels lie between neighbouring
podocytes. A preliminary examination of human podocytes reveals similar results.
The patent area of the pores/channels connecting SPS to BS is 0.25-0.5% of the GBM
area draining into the SPS. The narrow exits from the SPS therefore form a downstream restriction to ultra-filtrate formation over a significant area of GFB and increase
the path length that filtrate is forced to traverse before BS is reached.
The functional significance of this newly discovered urinary space has yet to be
determined, however some intriguing possibilities present themselves. We speculate that
the podocytes themselves may be able to influence single nephron GFR since an
assessment of the relevant Starling forces shows that re-absorption into the blood
capillaries is possible if the pressure within the SPS is only modestly increased above
that in BS. Such changes in SPS pressure would produce backwashing of the GFB
(cleaning the filter) and enable podocyte derived proteins (eg Ang I or VEGF) to reach
neighbouring receptor bearing endothelial cells.
We submit that the current understanding of glomerular micro-anatomy/physiology
derived from 50 years of EM and physiological study requires re-evaluation.
Renal Association October 2003
P 76
Human Podocytes Influence the Barrier Properties of Co-Cultured Human
Glomerular Endothelial Cells Via Mediators Including VEGF.
SC Satchell, MA Saleem and PW Mathieson
Academic Renal Unit, University of Bristol, Southmead Hospital, Bristol, BS10 5NB,
United Kingdom
We have previously described the detection of the endothelium-specific factors
angiopoietin-1 (ang1) and vascular endothelial growth factor (VEGF) in podocyte foot
processes in human renal glomeruli while the adjacent glomerular endothelial cells
(GEnC) express the cognate receptors Tie2 and VEGFR2 (JASN 2002;13:544-50). In
other vascular beds ang1 and VEGF work in concert and we proposed that in the
glomerulus these mediators direct the unique phenotype of GEnC. We have shown that
GEnC monolayers respond to both ang1 and VEGF by alteration in their barrier
characteristics (JASN 2002;13:496A). We have now examined the effect of co-culture
with podocytes on GEnC barrier function and the role of soluble mediators including
VEGF in this interaction.
We used ELISA and Western blotting respectively to confirm secretion of VEGF and
ang1 into the culture media by conditionally immortalised human podocytes. Human
GEnC monolayers in tissue-culture inserts were placed in wells of culture plates
containing either podocytes, control cells (GEnC or HK2) or no cells. Effects on GEnC
were examined morphologically by immunofluorescence staining for VE-cadherin and
by assessment of barrier function by measurement of trans-endothelial electrical
resistance (TEER, a measure of permeability to ions and small molecules) over time and
by passage of FITC-labelled albumin across the monolayer.
Co-culture with podocytes produced a change in growth pattern of GEnC such that they
tended to form whorls (whilst still remaining in a monolayer) rather than the usual
polygonal pattern. There was no loss of cells from the monolayer. Monolayers cocultured with podocytes developed a lower TEER compared with controls (e.g. 10+/-0.6
vs 26.8+/-1.4 _/cm2 at 11 days, p<0.0005) and showed less resistance to passage of
albumin. Blocking VEGF by addition of VEGFR2-Fc chimera abrogated these effects.
These data confirm that podocytes have specific effects on GEnC morphology and
monolayer permeability acting via soluble mediators including VEGF. These
experiments demonstrate in vitro the type of cell-cell communication that is likely to be
important in vivo in normal glomerular physiology and in the response to glomerular
Renal Association October 2003
P 77
Mesangial matrix-activated mononuclear cells express functional scavenger
receptors and accumulate intracellular lipid
EU Rahman1, RS Chana2, XZ Ruan1, SH Powis1, Z Varghese1 and DC Wheeler1
Centre for Nephrology, Department of Medicine, Royal Free and University College
Medical School, London, NW3 2PF, United Kingdom and 2Department of Cell
Physiology and Pharmacology, Faculty of Medicine and Biological Sciences, University
of Leicester, Leicester, LE1 9HN, United Kingdom
Monocyte recruitment into the mesangium and foam cell formation are recognised
features of glomerular injury. External signals encountered by these infiltrating cells
may determine their behaviour and thereby potentially influence disease outcomes. Our
previous studies indicate that activation of monocytes by mesangial matrix stimulates
the production of a variety of mediators including inflammatory cytokines and matrix
degrading enzymes. Using expression of peroxisome proliferator activator receptor _
(PPAR_) and scavenger receptor (ScR) as differentiation markers, we examined whether
matrix activation was associated with the expression of monocyte characteristics usually
associated with a macrophage phenotype. THP-1 mononuclear cells were incubated for
7 days with 500_g/ml solublised matrix extracted from cultured human mesangial cells.
Using phorbol methyl ester (PMA) (125nM) and albumin (500_g/ml) as positive and
negative controls respectively, we demonstrated that matrix activation of monocytes led
to intracellular lipid accumulation as demonstrated by oil red O staining. Matrix
activation was also associated with a concentration-dependent increase in the expression
of both ScR and PPAR_ mRNA and a corresponding increase in PPAR_ protein
expression on Western blotting. The presence of functional ScR was confirmed using
FACS analysis in which incubation of matrix-activated monocytes with Dil-labelled
acetylated low density lipoprotein (LDL) led to an increase in mean fluorescent
intensity of 373% as compared to albumin (100%, p<0.001) and PMA (423%). This
could be inhibited by addition of excess unlabelled ligand, suggesting specific binding
to the ScR. Furthermore, incubation of LDL with mesangial matrix in the absence of
cells led to enhanced electrophoretic mobility of recovered lipoprotein on agarose gel. A
similar shift was seen when LDL was incubated with Cu2+, a powerful lipoprotein
These results demonstrate that interactions with mesangial matrix induce expression of
monocyte characteristics associated with a macrophage phenotype, including ScR
expression and promote oxidation of LDL, thereby converting it to a ScR ligand. Such
observations may help to explain foam cell formation in the mesangium in the context
of glomerular disease.
Renal Association October 2003
P 78
Identification of common patterns of matrix gene expression in glomerular injury
dependent and independent models of tubulo-interstitial fibrosis in vivo
DM Sadlier, DF Higgins, D Brazil, P Doran and HR Brady
Human Genomics and Bioinformatics Research Unit, Department of Medicine and
Therapeutics, University College Dublin, MaterMisericordiae Hospital, Eccles St,
Dublin 7, Ireland., Dublin, 0000, Irish Republic
Transcriptomic analysis using microarray technology represents a powerful unbiased
approach for delineating pathogenic mechanisms in renal disease. Here, we used
Affymetrix gene chip (mu74av2) to explore the changes in matrix protein gene
expression underpinning renal injury in glomerular disease-dependent (adriamycin
nephropathy) and glomerular disease-independent (UUO) models of murine tubulointerstitial fibrosis (TIF).
Gene expression patterns were monitored temporally during progressive renal injury in
both models with microarrays being performed in triplicate and duplicate respectively.
Hierarchical cluster analysis was performed for 148 matrix proteins or modulators of
matrix turnover as determined by Onto-Compare.
Intriguingly, among the range of clusters identified for each model, five prominent
patterns were common to both. These represented clusters of matrix genes whose
mRNA levels were (a) increased most prominently either early or late in disease (one
cluster), (b) decreased prominently late in disease (one cluster) and (c) increased
progressively following an initial marked fall in mRNA levels. In addition,
transcriptomic analyses suggest an inter-relationship of specific pro-fibrotic factors
(egTGFBi, CTGF) and related modulating signalling molecules (eg Sparc).
The further exploration of these complex and interrelated networks should shed light on
the core molecular pathways that underpin renal TIF in human disease.
Renal Association October 2003
P 79
Normal human podocytes rapidly utilize glucose by both GLUT1 and GLUT4 in
response to insulin, dynamically alter their actin cytoskeleton, and show significant
differences in diabetic nephropathy.
RJM Coward1, GI Welsh2, GD Holman3, D Kerjaschki4, JM Tavaré2, PW Mathieson1
and MA Saleem1
Academic renal unit, Southmead Hospital, Westbury-on-Trym, Bristol, BS10 5NB,
United Kingdom, 2Department of Biochemistry, University of Bristol, Bristol, BS,
United Kingdom, 3Department of biochemistry, University of Bath, Bath, BA, United
Kingdom and 4Department of Pathology, University of Vienna-Allgemeines
Krakenhaus, Vienna, V, Austria
Using a conditionally immortalised human podocyte cell line we have previously
described that mature podocytes are capable of rapid glucose uptake (by a factor of
2.05), and that the glucose transporter GLUT 4 is present in these cells. We have now
investigated this further by utilizing a bis-glucose photo-labelling technique, which
“tags” functional cell surface glucose transporters, and have shown that both GLUT4
(factor 13.5 - p<0.05) and GLUT1 (factor 3.5 - p=0.05) are increased at the cell surface
of podocytes after 15 minutes stimulation with 220nM insulin. GLUT1 is classically
thought to be a constitutional basal glucose transporter that resides in the plasma
membrane (PM) of the cell, but in podocytes is capable of being rapidly up-regulated at
the PM which is further demonstrated by real time green fluorescent protein tagged
GLUT1 imaging. Furthermore both GLUT1 and GLUT4 translocation are dependent on
the actin cytoskeleton as its disruption with 10 _M Cytochalasin D results in only basal
glucose uptake into the cell over 15 minutes (tritiated glucose uptake increased by factor
of 1.03 compared with 2.05 in intact cells).
The actin cytoskeleton is dynamically effected by insulin as demonstrated by live GFP
actin tagged imaging showing significant changes in its structure in response to 220nM
insulin, which must be acting as an actin remodelling ligand.
Immunogold electron microscopic analysis of both normal and type 2 human diabetic
kidney sections suggests that the podocyte glucose transporter level and distribution
change in disease, with a significant up regulation of GLUT 4 occurring at the base of
the foot processes adjacent to the glomerular basement membrane (factor 2.6 - p<0.005)
Diabetic nephropathy is dominated by progressive proteinuria making the podocyte a
highly appealing target cell. This data suggests the podocyte is a dynamic insulin
responsive cell and demonstrates that the glucose transport mechanism changes in
diabetic nephropathy.
Renal Association October 2003
P 80
The differentiated podocyte displays key features of a contractile smooth muscle
IR Witherden, PW Mathieson and MA Saleem
Academic and Children's Renal Unit, Medical School Building, University of Bristol,
Southmead Hospital, Bristol, BS 10 5NB, United Kingdom
The podocyte is traditionally described as an epithelial cell, though little is known about
its true differentiated phenotype. We studied a conditionally immortalised human
podocyte cell line, with the ability to mature from proliferating undifferentiated
cobblestoned epithelial cells to growth-arrested differentiated podocytes over a period
of 14 days. cDNA microarray studies indicated that the gene expression profile of
differentiated podocytes included genes involved in smooth muscle differentiation.
Differentiated podocytes upregulated transcripts for sarcosin, a sarcomeric muscle
protein, RYR2, the gene coding for a Ca release channel in muscle contraction, FHL2
and MBNL, coding for proteins involved in terminal differentiation of muscle cells.
Additionally there was robust upregulation of transcripts for smoothelin and calponin,
proteins specifically associated with differentiated, contractile smooth muscle. In
differentiated cells in culture these proteins were expressed in a filamentous pattern on
immunofluorescence with predominantly the visceral isoform of smoothelin expressed
on Western blotting. In human glomerular sections, only podocytes expressed calponin
and smoothelin, whereas smooth muscle actin and myosin-heavy chain, which are more
indicative of a myofibroblastic phenotype, were expressed only in mesangial cells.
The contractile properties of undifferentiated and differentiated podocytes were assessed
using a single cell contractility assay, in which cells were cultured on collagen coated
flexible silicon elastomer (Fray et al 1998: Tissue Eng.4:273-283). Podocyte
contractility was assayed by blinded counting of the number of wrinkles generated per
cell. Differentiated podocytes produced 3.71+/-0.3 wrinkles/cell after 4 hours, compared
with 1.4+/-0.2 wrinkles/cell for undifferentiated podocytes (n=9 fields), and both were
significantly greater than undifferentiated 3T3 fibroblasts which generated 0.04+/-0.01
wrinkles/cell (p<0.001). Contraction was reversibly inhibited by the addition of the
actin depolymeriser, Latrunculin B.
These data imply that mature podocytes have key features of contractile smooth muscle
cells, which has implications for their ability to regulate glomerular blood flow and
Renal Association October 2003
Moderated Poster Session - Two
Group (E)
Proximal Tubule Biology
Wednesday 8 October
Garner Room
Renal Association October 2003
P 81
BMP-7 modulates HA mediated proximal tubular cell-monocyte interaction
WD Selbi1, AO Phillips1, V Hascall2 and CA De La Motte3
Institute of Nephrology, University of Wales College of Medicine, Heath Park, Cardiff,
CF14 4XN, United Kingdom, 2Department of Colorectal Surgery, The Cleveland Clinic
Foundation., Cleveland, 44195, USA and 3Department of Biomedical Engineering, The
Cleveland Clinic Foundation., Cleveland, 44195, USA
The aim of our work was to examine the role of Hyaluronan (HA) in monocyte binding
by proximal tubular cells.
Confocal imaging was used to examine the organisation of HA on the cell surface of the
proximal tubular cell line HK-2. Unstimulated cells formed peri-cellular HA cable-like
structures composed of coalescing bundles of thinner HA strands originating from
neighbouring cells. We used U937 monocytic cells to examine the potential binding
capacity of inflammatory cells by HK-2 cells. The relationship between bound U937
cells and HA cable structures was examined by confocal imaging of HA using the
biotinolayted HA-binding protein, and U937 cells, using an anti-CD68 antibody. These
images showed co-localisation of the HA cables and the chains of U937 cells. Pretreatment of HK-2 cells with hyaluronidase prior to addition of U937 cells prevented
U937 binding in chain-like configurations, although binding of scattered individual cells
was not affected.
Stimulation of cells with BMP-7 induced HAS2 mRNA expression and decreased
expression of Hyal1 and Hyal2 mRNA. The functional significance of the increase in
HA cables was assessed by addition of U937 cells to BMP-7 stimulated HK-2 cells.
U937 cell binding quantified by determination of bound radioactivity following addition
of 51Cr labelled U937 cells. BMP-7 stimulated a significant increase in bound U937
cells at all doses of BMP-7 added. In contrast to BMP-7, IL-1_ did not influence Hyal
expression, nor HA cable formation. As a consequence IL-1_ did not increase HA cablemediated CD44-dependent binding of U937 cells.
The data presented provides insight into how alterations in HA synthesis in the renal
cortex may be involved in modulation of the interaction between infiltrating
inflammatory cells and resident cells. In addition we have demonstrated that regulation
of HA-mediated monocyte binding by BMP-7. The specificity of the effects was
confirmed by data demonstrating that IL-1_ , despite being a potent stimulus of HA
synthesis, was unable to alter HA cable formation. This suggests that not only the
quantity, but also the structure of hyaluronan is crucial for its function.
Renal Association October 2003
P 82
Hyaluronan and proximal tubular epithelial cell migration
T ITO1, JD Williams1, S Al-Assaf2, GO Phillips2 and AO Phillips1
Institute of Nephrology, University of Wales College of Medicine, Heath Park, Cardiff,
CF14 4XN, United Kingdom and 2The North East Wales Institute, Centre for Water
Soluble Polymers, Wrexham, LL11 2AW, United Kingdom
The ubiquitous polysaccharide hyaluronan (HA) has been associated with both acute
renal injury and progressive renal disease. The aim of this study was to examine the
effect of HA on proximal tubular cell migration. The proximal tubular cell line, HK-2
cells were grown in monolayer culture, and cell migration following addition of HA
characterised in an in vitro model of injury that we have previously developed and
characterised. Addition of well-defined preparations of exogenous HA increased cell
migration, however optimum enhancement of migration was seen with HA of high
molecular weight. Activation of the MAP kinase signalling cascade, as assessed by
increased expression of the dually phosphorylated active form of MAPK, could be
demonstrated following addition of HA, This was blocked by the addition of a specific
antibody to the HA receptor, CD44. HA dependent enhanced migration was abrogated
by addition the CD44 blocking antibody, and by inhibition of MEK activity. Generation
of a denuded area also led to increased synthesis of endogenous HA and activation of
MAP kinase, and blockage of either CD44 or MAP kinase activation inhibited cell
migration and re-epithelialisation under non-stimulated conditions. We have
demonstrated that HA activation of the MAP-kinase pathway through binding to its
receptor CD44, enhances PTC migration. In addition the results suggest that mechanical
injury of PTC stimulated HA generation. These observations may have implications for
both recovery from acute tubular injury and progressive renal fibrosis.
Renal Association October 2003
P 83
Hyaluronan modulates TGF-_1 signalling in proximal tubular cells (PTC)
T ITO and AO Phillips
Institute of Nephrology, University of Wales College of Medicine, Heath Park, Cardiff,
CF14 4XN, United Kingdom
Increased expression of hyaluronan (HA) has been associated with both acute renal
injury and progressive renal disease, although the functional significance of this remains
unclear. Recent studies suggest an interaction between HA and TGF-_ signalling in
cancer cell biology. The aim of this study was to examine the relationship between
TGF-_1, HA and PTC function.
Under resting conditions, co-localisation of the principal receptor for HA, CD44, and
TGF-_1 type I receptor was demonstrated by immunoprecipitation and western analysis
and further confirmed by immunocytochemistry and confocal microscopy. Stimulation
of PTC with TGF-_1 led to increased synthesis of type IV collagen assessed by western
analysis. Addition of HA did not alter type IV collagen synthesis, but abrogated TGF-_1
mediated increase in type IV collagen. This effect was blocked by the addition of a
blocking antibody to CD44. Furthermore HA decreased TGF-_1 mediated SMAD
signalling as determined by decreased activity of a luciferase-SMAD responsive
construct, and decreased translocation of SMAD 4 into the cell nucleus, determined by
immunocytochemistry and confocal microscopy. We have previously demonstrated an
anti-migratory effect of TGF-_1 in a scratch wounding model. As with HA antagonism
of TGF-_1 extracellular matrix generation, HA reduced the anti-migratory effect of
TGF-_1, in a CD44 dependent manner. In contrast to the effect of TGF-_1 on collagen
synthesis, which is SMAD dependent, the anti-migratory effect of TGF-_1 in this model
is known to be dependent of activation of RhoA. In the presence of HA, TGF-_1
mediated activation of RhoA was also abrogated in a CD44 dependent manner.
The results suggest that co-localisation of CD44 and TGF-_1 receptors facilitate
modulation of both SMAD and non-SMAD dependent TGF-_1 mediated events by HA.
There is overwhelming evidence that TGF-_1 is critical to the development of
progressive renal disease. Our results therefore suggest that alteration of HA synthesis
may represent an endogenous mechanism to limit renal injury.
Renal Association October 2003
P 84
Calpain: A Mediator of High Glucose-Induced Apoptosis and Necrosis in Proximal
Tubular Epithelial Cells
DA Allen, SM Harwood, MJ Raftery and MM Yaqoob
Dept. Experimental Medicine & Nephrology, William Harvey Research Institute, Queen
Mary, University of London, London, EC1A 7BE, United Kingdom
We have previously demonstrated that high glucose-induced oxidative stress caused
apoptosis in proximal tubular epithelial cells (PTEC) that was mediated by a caspase
cascade (FASEBJ, 2003). However, there is little information about the processes that
initiate the cascade within the first 24 h, immediately following the increased oxidative
stress. The aim of this study was to determine if other proteases contributed to high
glucose-induced cell death in the first 24 hours. The calcium dependent cysteine
protease calpain is a mediator of both renal and cardiac cell death and is a candidate
mediator for early cell death in PTECs exposed to high glucose concentrations. A PTEC
cell line (LLC-PK1) was grown to sub-confluence and exposed to 5mM or 25mM Dglucose for up to 48 hours. To assess the role of cysteine proteases in high glucoseinduced PTEC necrosis and apoptosis, the specific calpain inhibitor calpeptin and the
pan-caspase inhibitor Z-DCB were used. Calpain and caspase activities were measured
by a fluorescent microplate assay and by an in-situ assay. DNA fragmentation and LDH
release were also determined as a measure of apoptosis and necrosis respectively.
25mM D-glucose caused a significant increase in calpain activity at 24 h when
compared with controls (5mM 1.0±0.31 vs 25mM 1.87±0.25, n=6, P<0.01).
Importantly, the increase in calpain activity could be detected as early as 4 h following
exposure to 25mM D-glucose (n=6, P<0.05) with unchanged caspase activity. There
was more necrosis in cultures exposed to 25mM D-glucose for 24 h compared with
controls (n=6, P<0.02) that was reduced by calpeptin (n=6, P<0.001) in parallel with
reduced calpain activity. There was increased apoptosis in cells exposed to 25mM Dglucose for 48 h when compared with controls (n=6, P<0.02) that was partly attentuated
by calpeptin and completely prevented by Z-DCB (both P<0.001). We conclude from
this study that calpain is activated by high glucose in PTECs and causes early necrotic
cell death. Calpain activation occurred much earlier than caspase activation and may
contribute to subsequent caspase-dependent apoptosis in PTECs.
Renal Association October 2003
P 85
Impaired mitochondrial function and increased apoptosis rate in human cystinotic
proximal tubular cells
GF Laube, V Shah, VC Stewart, IP Hargreaves, MR Haq, SJ Heales and WG
van’t Hoff, Nephro-Urology Unit, Institute of Child Health, UCLMS, London
WG Van't Hoff, V Shah, VC Stewart, IP Hargreaves, MR Haq, SJ Heales and GF Laube
Nephrourology Unit, Institute of Child Health, 30 Guilford Street, London, WC1N 1EH,
United Kingdom
Cystinosis is the commonest genetic cause of the renal Fanconi Syndrome (FS) but the
mechanisms of proximal tubular cell (PTC) dysfunction are unknown. We hypothesised
mitochondrial dysfunction and increased apoptosis might occur in the FS in cystinosis.
We cultured cells exfoliated from childrens’ urine (9 patients, 9 controls), confirmed
proximal tubular and cystinotic phenotype. Mitochondrial complex activity, ATP,
glutathione (GSH) and apoptosis rate were determined under standard and after culture
with hypoxic stress (3% for 48h). Cystinotic PTC exhibited depletion of GSH
(cystinotic: 6.8 ± 2.8 nmol GSH/meg protein, normal: 11.8 ± 2.8) but no significant
differences in mitochondrial complex activities or ATP level under basal conditions
(data not shown). After hypoxic stress, there were significant reductions (p < 0.001) in
complex I (cystinotic: 0.034 ± 0.011, normal: 0.067 ± 0.017) and IV (cystinotic: 0.0024
± 0.0008, normal: 0.0074 ± 0.003) activities, ATP (cystinotic: 12.2 ± 3.9 nmol/mg
protein, normal 26.9 ± 8.2) and further depletion of GSH (cystinotic: 2.4 ± 0.9 nmol
GSH/meg protein, normal: 7.2 ± 2.5) compared to controls. Hypoxia also led to a
reduction of GSH in normal PTC indicating the sensitivity of this marker. After
triggering with TNF-_ and anti-Fas antibody, cystinotic PTC showed significantly (p <
0.001) increased apoptosis. Using the TUNEL method, apoptosis rate in cystinotic PTC
was as twice as high as in control PTC: 15.5 ± 3.1 % without triggering, 18.1 ± 5.3 %
triggered with TNF-_ and 20.5 ± 5.2 % triggered with Anti-Fas antibody.
Mitochondrial dysfunction leading to increased apoptosis rate and thereby cell death,
has been considered a mechanism of cellular dysfunction in many tissues. We speculate
that these mechanisms may be implicated in the renal FS in cystinosis.
Renal Association October 2003
P 86
Recombinant Human Albumin (rHSA) stimulates Cell Growth, L-arginine
Transport and Metabolism to Polyamines in Human Proximal Tubular Cells
independent of PI-3 kinase inhibition
N Ashman, SM Harwood, J Kieskiewicz, MJ Raftery, AC Mendes-Ribeiro and MM
Department of Experimental Medicine and Nephrology, St Bartholomew's and the
Royal London Hospitals, London, EC1A 7BE, United Kingdom
Albumin stimulates proximal tubular epithelial cell (PTEC) growth via a PI-3 kinase
dependent process, suggesting a mechanism of disrupted PTEC turnover in proteinuric
nephropathies. We investigated a role for arginine and its’ metabolites the polyamines in
this process, given polyamines ability to trigger proliferation in other mammalian cells.
In HK-2 PTEC exposed to 20mg/ml rHSA for 24 hours, cell proliferation as [3H]thymidine incorporation is increased (1.41 + 0.13 fold control, p<0.01). In parallel, Larginine transport capacity is increased (control vmax 401 + 17, rHSA vmax 628 + 56
pmol/min/106 cells, p<0.01, n=6) in a dose- and time-dependent manner specific to
rHSA – transferrin and IgG do not have this effect.
The intracellular L-arginine metabolising enzyme arginase shows increased expression
by RT-PCR, and enhanced activity (control 920 + 310, rHSA 1331 + 376 pmol urea
generated/h/106 cells, p<0.05, n=6), with unchanged nitric oxide synthase activity after
rHSA incubation. Ornithine decarboxylase (ODC, converting ornithine to polyamines)
expression is increased by RT-PCR, with increased intracellular polyamine synthesis as
measured by HPLC (eg spermine, control 308 + 70, rHSA 524 + 14 pmol/106 cells,
p<0.05, n=6).
10mM _-difluoromethylornithine (DFMO), an ODC inhibitor, prevents increased
polyamine synthesis, and reduces cell proliferation to control levels - an effect in part
reversed by co-incubation with polyamines (p<0.05). Wortmannin and LY294002,
inhibitors of PI-3 kinase, have no effect on increased L-arginine transport or polyamine
synthesis, despite inhibiting cell growth. This suggests an effect upstream of this
enzyme, or a complementary pathway effecting PTEC proliferation.
The arginine-ornithine-polyamine pathway appears enhanced in PTEC incubated with
rHSA, and is involved in cellular proliferation: this may offer novel therapeutic targets
for the treatment of progressive proteinuric nephropathies.
Renal Association October 2003
P 87
Phosphorylation of Megalin in Human Proximal Tubule Cells Indicates a Novel
Role in Cell Signalling and Regulation of Tubular Function.
RJ Baines1, J Brown1, AB Tobin2, H Gallagher3, DJ Newman3 and NJ Brunskill1
Department of Nephrology, Leicester General Hospital, Gwendolen Road,
LEICESTER, LE5 4PW, United Kingdom, 2Dept of Cell Physiology and
Pharmacology, University Of Leicester, Leicester, LE1 9HN, United Kingdom and
South West Thames Institute for Renal Research,St. Helier Hospital, Carshalton.,
London, SW, United Kingdom
In proteinuria bioactive molecules entering the proximal tubule (PT) are thought to
contribute to progressive renal disease. Megalin (meg), the PTC receptor for albumin
and other proteins, is a member of the low-density lipoprotein receptor (LDLR) family
and is expressed in the PT brush border. Meg is predominantly extracellular and has a
defined role as a multiligand endocytic cargo receptor. Unlike other members of the
LDLR family meg has a large intracellular cytoplasmic tail (meg-CT) with a variety of
potential phosphorylation (phosph) sites suggesting a novel role in cell signalling. We
investigated the ability of meg to act as a signal transducing receptor in PTC.
The size of meg hinders the study of the intact receptor. We studied the ability of lysates
from human PTC (HPTC) to phosphorylate a meg-CT-GST fusion protein purified from
E coli. HPTC were lysed under control conditions or after stimulation with agonists
predicted to stimulate meg-CT phosph. Lysates were then mixed with meg-CT-GSTgluathione sepharose in the presence of [32P]-ATP and incubated for 20 mins at 37°C.
Reactions were stopped, sepharose beads washed, subjected to PAGE and gels
autoradiographed. Modest phosph of meg-CT was observed under non-stimulated
conditions. This increased 1.4 fold by 1mg/ml albumin, 1.6 fold by 10mM bradykinin,
1.9 fold by 1ng/ml EGF, and 2.4 fold by 1mM phorbol dibutyrate. Pretreatment of
HPTC with protein kinase C inhibitor abolished PDBU evoked phosph, and
pretreatment with EGF receptor inhibitor blocked EGF mediated phosph.
In order to examine phosph of native meg-CT in HPTC we generated rabbit polyclonal
antibodies directed at meg-CT peptide sequences. HPTC were loaded with [32P]orthophosphate and then stimulated with various agonists of meg-CT phosph informed
by the fusion protein studies +/- kinase inhibitors. After stimulation HPTC were lightly
trypsinised to remove extracellular meg and meg-CT immunopreciptitated from cell
lysates. Immunoprecipitates were subjected to PAGE and gels autoradiographed.
Endogenous HPTC meg-CT demonstrated a similar pattern of phosph in response to
agonists as did meg-CT-GST. Kinase inhibitor sensitivity was identical.
These results confirm phosph of meg-CT by pathophysiologically relevant agonists in
PTC and suggest a key role for meg signalling in PT function in health and disease.
Renal Association October 2003
P 88
Protein overload-induces oxidative stress in human proximal tubular (HK-2) cells:
effects on superoxide generation
L. Shalamanova, F. McArdle, J.L. Alexander, J.M. Bone, M.J. Jackson and R. Rustom
University of Liverpool, Department of Medicine, Duncan Building, Daulby Street,
Liverpool, L69 3GA, United Kingdom
In diseased native kidneys proteinuria and progression to renal failure are often linked.
Proteinuria can itself be directly damaging to the proximal tubules provoking interstitial
inflammation and progressive renal fibrosis. The mechanisms remain complex and
multifactorial and include increased oxidative metabolism. This study was undertaken to
investigate a potential involvement of oxidative stress in an in vitro model of protein
overload in human proximal tubular (HK-2) cells.
HK-2 cells were grown to confluency, growth arrested in serum-free medium (SFM) for
48 h, and then exposed to 30 mg/ml human serum albumin (HSA) for 24 h. Various
preparations of HSA were used including globulin-free/fatty acid-free (GF/FAF),
globulin free (GF), Fraction V (FrV), or fraction V/fatty acid-free (FrV/FAF) HSA to
determine the potential additional effects of the associated fatty acids or globulin.
Markers of oxidative stress were measured following addition of HSAs.
Trypan blue exclusion assay showed that the viability of the control cells (Co) and cells
exposed to different HSAs was >95%. All HSAs led to a significant decrease in cellular
total glutathione (Co 41.9±1 _mol/mg; Fr V/FAF 25.6±0.8 _mol/mg; p<0.001) together
with an increased activity of the antioxidant enzymes catalase (Co 4.7±0.3 IU/mg;
FrV/FAF 7.2±0.5 IU/mg; p<0.05) and glutathione peroxidase (Co 20.2±1.7 mU/mg; Fr
V/FAF 32.5±3.9 mU/mg; p<0.05). Malondialdehyde was also significantly increased in
a dose-dependent fashion in cells exposed to GF and FrV HSA, reaching maximal levels
at 30 mg/ml (Co 204±30 pmol/mg; GF 706±34 pmol/mg; p<0.0001).
Superoxide generation was measured in Co cells and cells incubated with 15 mg/ml GF
HSA over a 48 h period. Surprisingly, there was a progressive increase in superoxide
generation in Co cells reaching a maximum by 36 h (Co 34.4±0.2 pmol/20 min/mg; GF
6.5±1 pmol/20 min/mg; p<0.00001). By contrast, generation in cells exposed to GF
HSA was reduced, reaching a plateau within the first 3 h of the experiment (6.4±1
pmol/20 min/mg).
We conclude that protein overload induces oxidative stress with all preparations of
HSA, but maximal MDA levels were achieved with those containing fatty acids.
Superoxide generation increased significantly in control cells with time and was
apparently reduced by albumin. Further studies are required to define this relationship.
Renal Association October 2003
Renal Association October 2003