Effects of Cyclosporin A on Erythropoietin Production by the

From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
Effects of Cyclosporin A on Erythropoietin Production by the Human
Hep3B Hepatoma Cell Line
By Alessandro M. Vannucchi, Alberto Grossi, Alberto Bosi, Daniela Rafanelli, Marinella Statello, Stefan0 Guidi,
Riccardo Saccardi, and Pierluigi Rossi-Ferrini
There is evidence that the inadequate erythropoietin (Epo)
production observed in patients undergoing allogeneic
bone marrow transplantation (BMT) might be ascribed to
an inhibitory effect caused by the immunosuppressive
drug cyclosporin A (CsA). In this in vitro study, we have
evaluated the effects of CsA on the release of Epo in the
culture medium by the human Hep3B hepatoma cell line.
In cultures incubated with both CsA and the nonimmunosuppressive CsA analog MeAla-6, but not with the CsAunrelated immunosuppressive agent FK-506, the levels of
Epo in the medium were significantly reduced in comparison with controls, at concentrations (0.01 to 1.6 pmol/L)
not affecting total protein synthetic rate nor the constitutive secretion of a-fetoprotein. Hep3B cells were found to
contain a CsA-binding molecule, with an M,of 18 Kd, as
assessed by high performance liquid chromatography
(HPLC) and ligand-blotting analysis. CsA did not affect the
expression of the Epo gene, as judged by Northern blot
analysis, but caused a significant amount of Epo t o remain
unsecreted within the cells; almost all (97%of total) of the
intracellular Epo was associated with the plasma membrane subcellular fraction. We conclude that: (1) CsA is
able to inhibit Epo release in vitro by Hep3B cells, further
supporting the hypothesis that the drug might have a role
in the inappropriately low Epo levels observed in BMT patients; (2) the inhibitory effect appears t o be specific and
not caused by a general impairment of protein synthesis
and/or secretion; and (3) the reduced Epo levels found in
the medium of CsA-treated Hep3B cultures are supposed
to be the consequence of an inability of the cells to
correctly process Epo molecules for the secretory pathway.
0 1993 by The American Society of Hematology.
planted with a T-cell-depleted BM graft, exhibited higher
serum Epo levels than those treated with cyclosporin-A
(CsA) for the prevention of graft-versus-host disease
(GVHD). Moreover, in a recent study” the Epo response to
anemia was found to be impaired in autologous transplant
patients, most of which had been treated with CsA for the
induction of autologous GVHD, although the influence of
CsA on Epo production could not be definitely clarified in
comparison with patients not receiving CsA. Finally, we
have recently shown that mice treated with therapeutical
doses of CsA have an impaired capacity to produce adequate amounts of Epo in response to anemia.13Because this
effect occurred in the animal model in the absence of any
biochemical or histologic evidence of kidney toxicity, we
hypothesized that the inhibitory effects of CsA might be
somehow specific, and different from the classical CsA-induced nephrot~xicity.’~
Further support to this idea derived
from the observation that both captopril14 and pentoxyfillineI5 failed to prevent the inhibition on Epo production in
mice treated with CsA (June 199 1, personal observations).
In this paper, we provide experimental evidence supporting the hypothesis that the inadequate Epo production characteristic of allogeneic BMT patients might be ascribed, at
least in part, to an inhibitory effect of CsA on the mechanisms regulating Epo production. To this end, we took advantage of the human hepatoma Hep3B cell line, which is a
well-characterized in vitro system for studies of the mechanisms regulating Epo production.16 In fact, these cells release high amounts of Epo in the culture medium in response to hypoxia or to certain transition metals,” and this
regulation has been shown to occur at the Epo mRNA
IRCULATING levels of erythropoietin (Epo), the
main regulator of red blood cell (RBC) homeostasis in
humans,’ are finely adjusted according to the oxygen tension of the blood. In fact, Epo levels are maintained constant within a relatively narrow range ( I O to 25 mU/mL)
under steady-state conditions, but they increase several fold
when the RBC mass is reduced, as in acute or chronic anemias. An inverse, exponential correlation exists between
serum Epo concentrations and hematocrit levels in normal
subjects and in the majority of anemic patients.’ However,
this correlation may be lost in some types of nonrenal anemias, such as in rheumatoid a r t h r i t i ~~, ~a n c e rand
, ~ the acquired immunodeficiency syndrome (AIDS);’ therefore,
these patients are considered to have an “inappropriate”
Epo production.
In studies of patients subjected to bone marrow transplantation (BMT), an inappropriate Epo production has
generally been observed in patients receiving an allogeneic
graft,”” while there is some disagreement about the behavior of autologous rescue patient^.'-^^''^'^ In this regard, Abedi
et aI6 interestingly observed that allogeneic BMT patients
not receiving immunosuppressive therapy, being trans-
From the Bone Marrow Transplant Unit, the Division ofHematology, Careggi Hospital and University of Florence, Italy.
Submitted October 26, 1992; accepted March 25, 1993.
Supported by grants,from Boehringer Mannheim (Milano), Associazione ltaliana per le Leucemie (Firenze), Associazione Italiana
per la Ricerca SUI
Cancro, and MURST 40% -60%.
Address reprint requests to Alessandro M . Vannucchi, MD, Bone
Marrow Transplant Unit, Division of Hematology, Ospedale di
Careggi, 50134 Florence, Italy.
The publication costs of this article were defayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 solely to
indicate this ,fact.
0 1993 by The American Society of Hematology.
0006-49 71/93/8Z03-0035$3.00/0
Cell cultures. Hep3B cells were obtained from the American
Type Culture Collection (ATCC; Rockville, MD), and cultured in
Iscove’s modified minimal essential medium (IMEM; GIBCO,
Grand Island, NY), supplemented with penicillin (100 U/mL),
Blood, Vol 82, No 3 (August 1). 1993: pp 978-984
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
streptomycin (100 pLg/mL), and 10% heat-inactivated fetal calf
serum (FCS; Boehringer Mannheim GmbH, Germany). Cells were
maintained in a fully humidified, 5%COzatmosphere at 37"C, and
the culture medium was changed every other day during the logarithmic growth. Stock solutions of CsA and MeAla-6 (a synthetic
analog of CsA, modified at the 6 position) at 1 mg/mL were prepared by dissolvingthe drugs in 10%vol/vol ethanol, 0.2% vol/vol
Tween 80 (Boehringer Mannheim) in IMEM. FK-506 was dissolved at 1 mg/mL in 10% vol/vol ethanol in IMEM. From these
stock solutions, maintained frozen in aliquots at -7O"C, working
dilutions in IMEM were freshly prepared. In preliminary experiments, we observed that the addition of the buffers used for dissolving the drugs to mock cultures had no effects on the levels of Epo,
when compared with control cultures. Starting from the day before
experiment, nearly confluent cultures were fed with a serum-free
medium, consisting of IMEM, 1% Nutridoma-SP (Boehringer
Mannheim), L-glutamine (0.29 mg/mL), plus penicillin-streptomycin. Epo production by Hep3B cells was routinely induced by
incubating cultures with 50 pmol/L cobalt chloride (CoCl,) for 24
hours; in some experiments, cells were also stimulated with 300
pmol/L nickel chloride (NiCl,) or with a low (1%)oxygen tension
atmosphere. At the end of the incubation period, supernatants were
harvested, clarified by centrifugation, and stored frozen at -20°C
until assayed.
Epo levels in supernatants and cell lysates were determined with
an enzyme-linked immunoassay kit (Clinigen; Amgen Diagn,
Thousand Oaks, CA); each sample was assayed in duplicated using
at least two dilutions. Epo levels were normalized to the amount of
total cellular protein in cell pellets (Bradford's dye test''), and expressed as mU/mg cell protein. a-fetoprotein (a-FP) levels were
measured with a radioimmunoassay kit (Cis Diagnostici, Vercelli,
Italy), and expressed as ng/mg cell protein.
Characterization of Hep3B cytosolic CsA-binding protein($.
To prepare cytosolic proteins from Hep3B cells, about 5 X 10'
unstimulated cells were collected by mild trypsinization, washed
three times with calcium- and magnesium-free phosphate buffered
saline (PBS), and either used immediately or stored frozen as a cell
pellet at -70°C. Cytosolic protein extraction was performed by
homogenizing cells with a ground-glasshomogenizer in ice-cold 10
mmol/L Tris-HCI, pH 7.2, containing 150 mmol/L KCl, 0. I % sodium azide, and l mmol/L phenylmethylsulphonyl fluoride (TrisKCl-PMSF buffer), at a 1:4 cell/buffer ratio. 2-mercaptoethanol
(2-ME) was then added to give a final concentration of 5 mmol/L,
and supernatants from crude homogenates spun at lO0,OOOg for 1
hour (Beckman TL- 100 Ultracentrifuge; Beckman Instruments,
Palo Alto, CA) were saved. Protein concentration in the supernatant was then determined, and the samples either used immediately
for binding assay or stored at -70°C in aliquots.
The binding of ['HI-CsA ([mebmt-B-3H]cyclosporin A, specific
activity 10.0 Ci/mmol; Amersham Int plc, Buckinghamshire, England) to cytosolicproteins was performed as described." The binding buffer consisted of 20 mmol/L Tris-HCI, pH 7.2, containing 5
mmol/L 2-ME, 7.5% FCS, and 0.1% sodium azide. One hundred
pL-volumes of cytosolicproteins, of different protein content, were
mixed with 50 pL of ['HI-CsA to give 0.5 nmol/L of labeled ligand,
in the presence (nonspecificbinding) or absence (total binding) of a
200-fold molar excess of unlabeled CsA. The mixture was incubated for 30 minutes at room temperature with occasional agitation, then kept in ice. Free-labeled ligand was separated from the
one bound to receptofls) by loading the sample on minicolumns
(total volume, 3 mL) of LH-20 resin (Pharmacia P-L Biochem Inc,
Milwaukee, WI); in these experimental conditions, free ['HI-CsA is
significantly retarded in the column, and separated from the protein-containing void volume. The fraction corresponding to the
void volume was collected and radioactivity counted in a liquid
scintillator. Specific binding was calculated by the difference between total and nonspecific binding.
Size-exclusion high performance liquid chromatography (HPLC)
was performed with an LKB apparatus (Pharmacia LKB Biotechnology, Uppsala, Sweden) equipped with a TSK-G 3000 SW column. About 1 mL of cytosolic protein extract was concentrated to
0.2 mL with a Centricon 10 microconcentrator (Amicon-Grace,
Beverly, MA), 0.45 pm filtered, and injected; the column was developed with 20 mmol/L KH2P0,, pH 7.2, at 1.0 mL/min. Fractions
of 0.25 mL were collected, and immediately additioned with 2-ME
to give 5 mmol/L final concentration; each fraction was then processed for CsA binding with the LH-20 assay,
The identification of CsA-binding protein(s) by ligand blotting
was performed by running a concentrated cytosolic protein sample
on a 12%sodium dodecyl sulfate-polyacrylamidegel electrophoresis (SDS-PAGE); the sample was simply diluted with unreducing
SDS-sample buffer, but not boiled. Separatedproteins were electrophoretically transferred to nitrocellulose filters (Bio-Rad, Richmond, CA), and the blot incubated in a blocking buffer, consisting
of 25 mmol/L Tris-HC1, pH 7.4, 140 mmol/L NaC1, 4% bovine
serum albumin (BSA), and 1% nonfat dried milk. Thereafter, the
nitrocellulose sheet was incubated overnight at room temperature
with 5 nmol/L ['HI-CsA, in the presence or absence of a 500-fold
excess of unlabeled CsA, and then extensively washed. Fluorography-treated gels were exposed at -70°C for 7 to 10 days.
Northern blot analysis. The probe for human Epo was a 40-mer
oligonucleotide, whose sequence is derived from the exon 2 of the
Epo gene (Oncogene Science,Manhasset, NY). A 7-mer oligonucleotide probe for human 0-actin was purchased from Clontech Lab
(Palo Alto, CA). Both oligonucleotides were 5'4abeled with [-y3'P]
adenosine triphosphate (ATP) (specific activity 6,000 Ci/mmol;
NEN-DuPont, GmbH, Dreieich,Germany) and T4-polynucleotide
kinase to a specific activity of 3.5 to 5.0 X lo6 cpm/pmol.
Total RNA was prepared by cultured cells as described." The
RNA was denatured by formaldehyde-formamide,and subjected to
Fig 1 . Dose-dependent reduction of Epo levels in the medium of
Hep3B cells incubated with CsA ( 0 )or MeAla-6 (A).The effects of
FK-506 (m) are also shown. Cells were fed with serum-free media
containing 5 0 rmol/L CoCI,, in the absence or in the presence of
the drugs. Epo levels were determined after a 24-hour incubation,
and normalized to the amount of cell proteins. Data are derived
from five experiments for CsA and MeAla-6, and two experiments
for FK-506. *P < .05; **P < .01, Student's t-test.
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
Table 1. Effects of CsA on Total Cellular Protein Synthetic Rate
Experimental Condition
CsA (pmol/L)
[3H]-Aminoacids Incorporated
(cpm x
77 f 16
75 f 19
81 2 2 5
87 f 27
75 f 23
Hep3B cells were incubated for 16 hours in media containing 50
pmol/L cobalt chloride, in the presence or absence of CsA, and then
pulsed for 2 hours with 15 pCi of a mixture of [3H]-aminoacids.Cellular
proteins were precipitated with TCA and counted in a beta counter.
Results are expressed as the mean ? SD from two experiments.
electrophoresis in I % agarose gels in 2.2 mol/L formaldehyde gel
running buffer. The transfer of denatured RNA to Hybond-N filters
(Amersham) was performed by standard procedures in 1OX SSC
(saline-sodium citrate buffer).22Prehybridization was performed at
42°C for 6 hr with prehybridization buffer (6X SSPE [saline-sodium phosphate-EDTA], 50%formamide, 0. I % SDS, 10% Denhart
solution, 100 pg/mL denatured salmon sperm DNA). Hybridization was performed at 42°C for 16 hours with hybridization buffer
(6X SSPE, 50% formamide, 0.1% SDS). The final washing was in
1X SSC at 60°C. Autoradiography was performed with two intensifying screens at -70°C with Hypefilm-MP (Amersham) for 1 to 3
Intracellular Epo total content and distribution. For the determination of total intracellular Epo content, trypsinized cells were
washed twice with PBS, and then dissolved in Tris-KC1-PMSF
buffer containing0.2%Triton X-100 (BDH Chem, Poole, England)
for 30 minutes on ice. The sample was centrifuged at 1.OOOg for 10
minutes at 4°C to precipitate insoluble material, and the clear supernatant was used for Epo determination. In experiments where
the intracellular distribution of Epo was determined, trypsinized
cells were washed twice, and finally resuspended in Tris-KCI-PMSF
buffer containing 0.25 mol/L sucrose. Cell lysis was obtained by
forcing the cells through a 27-gauge needle for six times, on ice; the
loss of cellular integrity was confirmed by phase-contrast microscopy. Crude nuclei were prepared at 1,OOOg for 10 minutes, the
Table 2. Effects of CsA and MeAla-6 on the Level of a-FP
in the Culture Medium of Hep3B Cells
CsA (pmol/L)
MeAla-6 (pmol/L)
a-FP (ng/mg protein)
445 f 56
f 45
f 27
f 33
445 f 35
438 f 18
4 1 6 ? 69
445 f 3 0
Hep3B cells were incubated in media containing5 0 pmol/L cobalt chloride, in the presence or absence of CsA and MeAla-6, and the amount of
a-FP was measured 2 4 hours later. Results are the mean f SD from two
p 30.0
0Q: 15.0
Fig 2. Identification of CsA-binding protein(s) in Hep3B cells.
The specific binding of r3H]-CsA by increasing amounts of cellular
cytosolic proteins, and the displacement after incubation with an
excess of unlabeled CsA, are shown in the figure. The line represents specific binding, which was calculated by subtracting from
total binding the unspecific binding measured in the presence of a
200-fold molar excess of unlabeled CsA.
supernatant collected and spun down at 100,OOOg for 6 0 minutes at
4°C. The final supernatant (soluble) was saved, while the membrane pellet was dissolved on ice in Tris-KC1-PMSF buffer containing 0.2% Triton X- 100 to yield the membrane-associated Epo fraction.
In nearly confluent, unstimulated Hep3B cells maintained for 16 to 24 hours under serum-free conditions, Epo
was barely detected in the medium (4.0 ? 3.5 mU/mg protein). On the other hand, cultures stimulated with CoCI, (50
pmol/L) released high amounts of Epo in the medium, up to
255 t- 40 mU/mg protein (Fig 1). Comparable results were
obtained when cells were incubated in an atmosphere containing 1% 0, (305 f 95 mU/mg), while stimulation with
300 pmol/L NiCI, proved to be less efficient ( 1 90 f 75 mU/
mg). Therefore, owing to its simplicity and reproducibility,
we have routinely used CoCI, for inducing Epo production
by Hep3B cells in all subsequent experiments. However, in
preliminary experiments not reported in detail, similar effects of CsA on Epo production were observed when Hep3B
cells were stimulated with either hypoxia or NiCI,.
Cultures incubated with CsA (in the range of 0.0 1 to I .6
pmol/L) showed a dose-dependent reduction in the amount
of Epo released into the culture medium after CoC1, stimulation (Fig 1). At CsA concentrations of 0.4 pmol/L, about
half of the controls’ Epo was found in the medium, and
decreased further up to about 30% of controls (78 k 15
mU/mg) at 1.6 pmol/L. The inhibitory effect was not
caused by a reduction in the total rate of protein synthesis
by Hep3B cells; in fact, the amount of tritiated amino acids
incorporated in trichloroacetic acid (TCA) precipitates was
virtually comparable in control and CsA-treated cultures
(Table 1). Moreover, concentrations of CsA up to 2.0 pmol/
L were found not to be toxic to the cells, because no significant modification of total cell number per plate, nor of the
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
Fig 3. Size-exclusion HPLC characterization of CsA receptor(s) in cytosolic extracts from Hep3B cells. The fractions containing functional CsA receptor(s) were identified by the LH-20 assay. Insert, ligand blotting identification of CsA-binding protein. Unseparated cytosolic
proteins were resolved on a 12% SDS-PAGE, transferred to nitrocellulose filters, and incubated with [3H]-CsA in the presence (lane A) or
absence (lane B) of an excess of unlabeled CsA.
percentage of Trypan-blue’ cells, was observed (data not
shown in detail). Finally, to evaluate the specificity of the
inhibitory effects of CsA on Epo, the amount of a-FP, a
constitutively secreted protein by Hep3B cells, was also
measured in culture media. As shown in Table 2, the levels
of this protein were not significantly modified by CsA.
We next tested whether the nonimmunosuppressive,
stable CsA analog MeAla-6 had similar effects as the parental molecule. As reported in Fig 1, dosedependently reduced Epo levels were found in cultures incubated with
MeAla-6 (0.01 to 1.6 pmol/L), although on a molar basis
the analog proved to be less efficient than CsA, especially at
the lower concentrations. No inhibitory effect on a-FPlevels was observed with MeAla-6 (Table 2). On the other
hand, the CsA-unrelated,immunosuppressiveagent FK506
failed to reduce Epo levels in the medium, in comparison
with control cultures (Fig 1).
Because it is known that the effects of CsA on target cells
are mediated by specific cytoplasmic receptors, we searched
for the presence of (a) [email protected])in Hep3B cells. As
shown in Fig 2, cytosolic proteins prepared from Hep3B
cultures were found to bind [’HI-CsA in a dose-related fashion, and the bound labeled molecule was efficiently displaced by the cold substrate. By hypothesizinga 1:1 stoichio-
metric interaction between [3H]-CsA and the receptor($,
the calculated concentration of the CsA receptor(s)was 0.12
f 0.04 pglmg cytosolic proteins. To further characterizethe
CsA receptor(s),a molecular size-exclusion HPLC analysis
of cytosolic extracts was performed, which allowed for the
identification of a major peak of [3H]-CsAbinding protein
correspondingto a calculated M, of about 18.0 Kd (Fig 3).
The CsA-receptorprotein was also identified by ligand-blotting experiments,in which gels were deliberatelyoverloaded
to allow evidentiation of even less represented CsA-binding
species. However, only a single band was observed in these
experimental conditions, with a M, of 17.5 Kd (Fig 3, insert).
To verify the hypothesis that the reduced levels of Epo
found in the medium of cultures incubated with CsA were
caused by an effect on the expression of the Epo gene, total
RNA was prepared from CoCI,-stimulated Hep3B cultures
that had been incubated for 16 hours in the presence or
absence of CsA. However, in four separateexperiments,one
of which is reported in Fig 4, we were unable to document
any apparent reduction in the amount of Epo mRNA in
cultures incubated with up to 1.6 pmol/L CsA. Then, we
reasoned that alterations of posttranslational processes
might be involved in causing the reduced Epo levels found
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
Fig 4. Effects of CsA on the expression of Epo
mRNA in Hep3B cells stimulated with CoCI,. Total
RNA (20 pg) was resolved on a 1%fonnaldehydeagarose gel, and hybridized with a [3ZP]-labeledoligonucleotide probe for human Epo. Filters were
then stripped, and incubated with a probe for human ,!?-actin,to provide a control of the amount of
RNA loaded in the gel. Epo levels in the medium,
normalizedto the amount of proteins in cell pellets,
are also shown.
in the medium ofcultures incubated with CsA. To this end,
the amount of Epo remaining within the cells at different
times after stimulation with CoCI, was measured. While no
effect was noticed within the first 4 hours of incubation, a
progressive increase in the intracellular Epo content was
observed in the presence of CsA (1.2 pmol/L) at 8 to 24
hours after CoCI, stimulation (Fig 5): at 24 hours, cells incubated with CsA contained about IO-fold the amount ofcontrols’ Epo. The bulk (97% oftotal) ofthe intracellular Epo in
1 cl w/o CsA 0 with CsA]
cells incubated with CsA was found in the fraction precipitated at lO0,OOOg and dissolved with Triton X- 100, suggesting that it was membrane-associated (Table 3).
Results presented in this paper indicate that in Hep3B
cultures incubated with CsA at nanomolar concentrations,
the amount of Epo released in the medium is significantly
reduced when compared with control cultures. The following considerations support the idea that the inhibitory effect
of CsA was specific: (1 ) at the doses used, CsA did not cause
a general reduction in total cellular protein synthetic rate,
nor did it affect the secretion of a-FP, a constitutively secreted protein by Hep3B cells; (2) the nonimmunosuppressive CsA analog MeAla-6 proved to be inhibitory, although
Table 3. Intracellular Distribution of Epo in Hep3B Cells
24 hr
Fig 5. Total intracellular Epo content as a function of incubation
time. Hep3B cells were stimulated with CoCI,, in the presence or
absence of CsA (1.2 pmol/L), and the amount of Epo in detergentlysed cells was measured at different times. Epo levels were normalized to the amount of proteins in cell lysates. Data from two
experiments. **P < .01, Student’s t-test.
€PO (mU/mg protein)
Crude nuclei
5.5 2 3.0
0.2 f 0.2
57.4? 10.0
1.7 f 0.5
Hep3E cells were incubatedin media containing 50 pmol/L cobalt chloride, in the presence or absence of CsA (1.2 pmol/L), and cells were
collected 16 hours later. The subcellular fractionation was performed as
described in Materials and Methods. Results are expressed as the mean
2 SD from two experiments.
Abbreviation: NA, not measureable.
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
about 40% less efficiently on a molar basis than the parental
drug. This correlates well with the previous demonstration
that MeAla-6 binds to the CsA receptor, the protein cyclophilin, with an efficiency of about 50% in comparison with
CSA;’~and (3)the CsA-unrelated immunosuppressive agent
FK506 had no effect on the release of Epo by Hep3B cells.
We have also found that Hep3B cells contain a protein capable of binding t3H]-CsA, with a molecular weight comparable with the main cyclophilin species of T-lymphocytes
( 18.0 Kd).24Finally, evidence has been provided that the
reduced levels of Epo found in the culture medium of
Hep3B cells are not caused by an effect of CsA on the expression of the Epo mRNA, at least at the level of sensitivity
of Northern analysis; rather, they appeared to be dependent
on (an) alteration(s) in the mechanisms leading to hormone
secretion in the culture medium. In fact, the amount of Epo
remaining inside the cells was significantly higher in cultures incubated with CsA than in controls, and subcellular
fractionation studies indicated that the most part of the unsecreted, intracellular Epo was associated with the plasmamembrane fraction.
The immunosuppressive activity of CsA results from the
inhibition of some receptor-mediated signal transduction
pathways in T lymphocyte^.^^^^^ CsA becomes active after
the binding to a specific cytoplasmicreceptor, named cyclophilin, which is part of a family of ubiquitous immunosuppressant receptor proteins, the imm~nophilins.~’
These are
all rotamases (peptidyl-prolylcis-trans isomerases; PPIase),
enzymes that catalize the cis-trans isomerization of peptidyl-prolyl amide bonds of peptide and protein substrates,
and which may act by accelerating the correct folding of
proteins during protein synthesis. Although the rotamase
activity is strongly inhibited by C S A , ~ *there
~ * ~ is evidence
that the immunosuppressive activity of CsA derives from
the interaction of the active drug-receptor complex with the
protein phosphatase c a l c i n e ~ r i nmore
, ~ ~ than from the inhibition of PPIase activity. It has also been suggested that the
physiologic role of the ubiquitous PPIases in the cells may
be that of interacting with proline-containing proteins, perhaps stabilizing their structure and/or allowing their translocation within different cellular compartments.”
The results of our experiments seem to suggest that the
formation of an active complex between CsA and the cyclophilin-like protein found in Hep3B cells may elicit some
alterations in the posttranslational processing of Epo, with
the consequent accumulation of the molecule within the
endocellular membranes and a net reduction of the amount
of hormone secreted. Theoretically, this might be caused by
(1) a defect in the correct folding of the Epo molecule, because of the inhibition of PPIase activity, and/or ( 2 )a defect
in the intracellular translocation processes of the Epo molecules required for the secretion in the medium. About the
first mechanism, it may be worthy to recall that Lodish and
Kong’’ observed that CsA inhibited the secretion of transferrin from another human hepatoma cell line (HepG2) at
concentrations not affecting the release of other proteins,
and suggested this effect might be caused by a lag in the
folding of transferrin molecules and to a retarded maturation from the endoplasmic reticulum (ER). However, ex-
periments performed up to now failed to document significant structural alterations of the Epo molecule produced by
Hep3B cells incubated with CsA (unpublished data, September 1992), possibly indicating that the first mechanism
is less likely. Therefore, although the unequivocal intracellular localization of Epo in CsA-treated cultures will require
the use of specific antibodies and further subcellular fractionation studies, at the moment we favor the second hypothesis, which has analogies with studies of a cyclophilinlike protein encoded by the n i n d gene of Drosophila
m e l a n ~ g a s t e rThe
. ~ ~ n i n d gene product, highly homologous to the B isoform of mammalian cyclophilin, is found
only in the retina of Drosophila, and it has been shown to be
essential for the posttranslational processing of rhodopsin
Rhl molecules, which require n i n d product for the exit
from the ER and their traveling through the cytoplasm to
the rhabdomeres. In fact, n i n d mutant flies showed a dramatic accumulation of Rh 1 opsin molecules in the cisternae
of ER.33A similar mechanism of action has also been proposed by Hultsch et
who observed that both CsA and
FK-506 can inhibit receptor-mediated exocytosis of secretory granules from a rat basophilic leukemia cell line. They
suggested that the two drugs might act by inhibiting the
intracellular translocation of the secretory granules, in the
same way as the translocation of nuclear factors required for
the activation of the early genes involved in the immune
response is affected in T lymphocytes.
Our proposed model is that the immunophilin found in
Hep3B cells is physiologically involved in some steps of the
intracellular processing and/or secretion of the Epo molecules, and that these processes are hindered in the presence
of CsA. In this regard, it has been shown that a cyclophilin
identified from a murine cDNA library (cyp C) binds specifically to a 77-Kd protein in the absence of the drug, while in
the presence of CsA it binds to a 55-Kd protein.35The authors suggested that the formation of the cyp C-p77 complex may reflect the physiologic function of this immunophilin in the cells, while the cyp C-p55 complex is involved
in signal transduction events elicited by CsA binding. Furthermore, an immunophilin specific for FK-506 has been
recently shown to associate, in the absence of the drug, with
two heat-shock proteins and with the glucocorticoid receptor to form the inactive glucocorticoid receptor complex,36
thus suggesting a physiologic role for the F’K506-binding
protein different from that involved in immunosuppression.
In conclusion, we have observed that CsA is able to significantly affect the release of Epo in vitro by the Hep3B hepatoma cell line, thus supporting previous suggestions that the
reduced Epo levels characteristic of allogeneic BMT patients might be ascribed, at least in part, to the drug. Although it might be questioned whether Hep3B cells are a
physiologic model of Epo production, the previous demonstration that CsA impairs Epo production in vivo’3seems to
indicate that results obtained with the hepatoma cell model
are possibly applicable also to kidney cells, which represent
the major source of the hormone in the intact animal. Finally, our data add some further insights into the roles of
immunophilins in the regulation of protein metabolism in
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
cells other than T lymphocytes, and suggest the existence of
a complex network of target molecules for CsA, whose
knowledge might also help in understanding the molecular
mechanisms of the toxicity associated with the drug.
Authors are indebted to Prof F. Paoletti for helpful suggestions.
Dr G. Corbetta (Sandoz, Milano, Italy) kindly provided CsA and
MeAla-6; FK-506 was a gift of Dr M.P. Piccinni (University of
1. Krantz SB: Erythropoietin. Blood 77:419, 199 I
2. Erslev AJ: Erythropoietin titers in health and disease. Semin
Hematol 28:2, 1991
3. Hochberg MC, Arnold CM, Hogans BB, Spivak JL: Serum
immunoreactive erythropoietin in rheumatoid arthritis: Impaired
response to anemia. Arthritis Rheum 31:1318, 1988
4. Miller CB, Jones RJ, Piantadosi S, Abeloff MD, Spivak JL:
Decreased erythropoietin response in patients with the anemia of
cancer. New Engl J Med 322:1689, 1990
5. Spivak JL, BarnesDC, Fuchs E, Quinn TC: Serum immunoreactive erythropoietin in HIV-infected patients. JAMA 26 1:3 104,
6. Abedi MR, Backman L, Bostrom L, Lindback B, Ringden 0:
Markedly increased serum erythropoietin levels following conditioning for allogeneic bone marrow transplantation. Bone Marrow
Transpl 6:121, 1990
7. Ireland RM, Atkinson K, Concannon A, Dodds A, Downs K,
Bigs JC: Serum erythropoietin changes in autologous and allogeneic bone marrow transplant patients. Br J Haematol76: 128, 1990
8. Schapira L, Antin JH, R a n d BJ, Antman KH, Eder PJ,
McGarigle CJ, Goldberg MA: Serum erythropoietin levels in patients receiving intensive chemotherapy and radiotherapy. Blood
76:2354, 1990
9. Beguin Y ,Clemons GK, Oris R, Fillet G: Circulating erythropoietin levels after bone marrow transplantation: inappropriate response to anemia in allogeneic transplants. Blood 77:868, 1991
IO. Bosi A, Vannucchi AM, Grossi A, Guidi S, Saccardi R, Rafanelli D, Longo G, Rossi Femni P Inadequate erythropoietin production in allogeneic bone marrow transplant patients. Haematologica 76:280, 1991
1 1. Miller CB, Jones RJ, Zahurak ML, Piantadosi S, Burns WH,
Santos GW, Spivak J L Impaired erythropoietin response to anemia after bone marrow transplantation. Blood 80:2677, 1992
12. Bosi A, Vannucchi AM, Grossi A, Guidi S, Vannucchi L,
Saccardi R, Bernabei PA, Longo G, Rafanelli D, Rossi Fenini P:
Serum erythropoietin levels in patients undergoing autologous
bone marrow transplantation. Bone Marrow Transpl 7:42 1, 1991
13. Vannucchi AM, Grossi A, Bosi A, Rafanelli D, Guidi S, Saccardi R, Alterini R, Rossi Ferrini P: Impaired erythropoietin production in mice treated with cyclosporin A. Blood 78:1615, 1991
14. Bantle JP, Paller MS, Boudreau RJ, Olivari MT, Ferris TS:
Long-term effects of cyclosporine on renal function in organ transplant recipients. J Lab Clin Med I15:233, 1990
15. Brunner LJ, Vadiei K, lyer LV, Luke DR: Prevention of
cyclosporine-induced nephrotoxicity with pentoxifylline. Renal
Fail 11:97, 1989
16. Goldberg MA, Glass GA, Cunningham JM, Bunn H F The
regulated expression of erythropoietin by two human hepatoma cell
lines. Proc Natl Acad Sci USA 84:7972, 1987
17. Goldberg MA, Dunning SP, Bunn HF: Regulation of the
erythropoietin gene: Evidence that the oxygen sensor is a heme
protein. Science 242: 14 12, 1988
18. Goldberg MA, Imagawa S, Strair RK, Bunn H F Regulation
of the erythropoietin gene in Hep 3B cells. Semin Hematol 28:35,
19. Bradford MM: A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of
protein-dye binding. Anal Biochem 72:248, 1976
20. Handschumaker RE, Harding MW, Rice J, Drugge RJ,
Speicher DW: Cyclophilin: A specific cytosolic binding protein for
cyclosporin A. Science 226544, 1984
2 1. Chomczynski P, Sacchi N: Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chlorophormextraction. Anal Biochem 162:156, 1987
22. Sambrook J, Fritsch EF, Maniatis T: Molecular cloning: A
laboratory manual. Cold Spring Harbor, NY, Cold Spring Harbor
Laboratories, 1989
23. Sigal NH, Dumont F, Durette P, Siekerka JJ, Peterson L,
Rich DH, Dunlap BE, Staruch MJ, Melino MR, Koprak SL, Williams D, Witzel B, Pisano JM: Is cyclophilin involved in the immunosuppressive and nephrotoxic mechanism of action of cyclosporin
A? J Exp Med 173:619, 199I
24. Harding MW, Handschumaker RE, Speicher DW: Isolation
and amino acid sequence of cyclophilin. J Biol Chem 261:8547,
25. Kahan BD: Cyclosporine. New Engl J Med 321:1725, 1989
26. Kronke MW, Leonard WJ, Depper JM, Arya SK, WongStall F, Gallo RC, Waldman TA, Greene W C Cyclosporin A inhibits T-cell growth factor gene expression at the level of mRNA
transcription. Proc Natl Acad Sci USA 8 15214, 1984
27. Schreiber SL: Chemistry and biology of the immunophilins
and their immunosuppressive ligands. Science 25 1:283, 1991
28. Takahashi N, Hayano T, Suzuki M: Peptidyl-prolyl cis-truns
isomerase is the cyclosporine A-binding protein cyclophilin. Nature
337:473, 1989
29. Fisher G, Wittmann-Liebold B, Lang K, Kiefhaber T,
Schmid F: Cyclophilin and peptidyl-prolyl cis-trans isomerase are
probably identical proteins. Nature 337:476, 1989
30. Liu J, Farmer JD, Lane WS, Friedman J, Weisman I,
Schreiber S L Calcineurin is a common target of cyclophilin-cyclosporin A and FKBP-FK506 complexes. Cell 66:807, 1991
3 I. Schreiber SL, Crabtree GR: The mechanism of action of cyclosporin A and FK506. Immunol Today 13:136, I99 2
32. Lodish HF, Kong N: Cyclosporin A inhibits an initial step in
folding of transferrin within the endoplasmic reticulum. J Biol
Chem 266:14835, 1991
33. Colley NJ, Baker EK, Stamnes MA, Zucker CS: The cyclophilin homolog ninaA is required in the secretory pathway. Cell
67:255, 1991
34. Hultsch T, Albers MW, Schreiber SL, Hohman RJ: Immunophilin ligands demonstrate common features of signal transduction
leading to exocytosis or transcription. Proc Natl Acad Sci USA
88:6229, 1991
35. Friedman J, Weissman I: Two cytoplasmic candidates for
immunophilin action are revealed by affinity for a new cyclophilin:
One in the presence and one in the absence of CsA. Cell 66:799,
36. Tai P-KK, Albers MW, Chang H, Faber LE, Schreiber SL:
Association of a 59-kilodalton immunophilin with the glucocorticoid receptor complex. Science 256: 1315, 1992
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
1993 82: 978-984
Effects of cyclosporin A on erythropoietin production by the human
Hep3B hepatoma cell line
AM Vannucchi, A Grossi, A Bosi, D Rafanelli, M Statello, S Guidi, R Saccardi and P Rossi-Ferrini
Updated information and services can be found at:
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
Information about ordering reprints may be found online at:
Information about subscriptions and ASH membership may be found online at:
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.