e tribut Metastasizing and Non-Metastasizing Tumors Likely Evolve from DNA

[Cell Cycle 3:10, e45-e46, EPUB Ahead of Print: http://www.landesbioscience.com/journals/cc/abstract.php?id=1169; October 2004]; ©2004 Landes Bioscience
Metastasizing and Non-Metastasizing Tumors Likely Evolve from DNA
Phenotypes via Independent Pathways
This manuscript has been published online, prior to printing for Cell Cycle, Volume
3, Issue 10. Definitive page numbers have not been assigned. The current citation
is: Cell Cycle 2004; 3(10):
Once the issue is complete and page numbers have been assigned, the citation will
change accordingly.
metastasis, tumorigenesis, DNA structure, prostate
Fourier transform-infrared
The classic concept that a primary tumor gives rise to metastasis via clonal selection of
a few rare cells1 persists to this day and has profoundly influenced the direction and interpretation of cancer research.2 However, this concept has been challenged on the basis of
gene expression profiles of human breast tumors.3-6 These studies suggest that the proclivity
for metastasis might develop in progenitor cells early in multistep tumorigenesis, as put
forth by Bernards and Weinberg.7
We have now discovered a metastatic cancer DNA phenotype in histologically normal
tissues from prostates with tumors having confirmed distant metastases.8 These matched
normal and tumor tissues from the same individual were found to have virtually identical
DNA base and backbone structures. This surprising discovery was made using statistical
models of Fourier transform-infrared (FT-IR) spectra of DNA having the proven ability to
identify subtle modifications in DNA structure.8-11 The presence of a metastatic cancer
DNA phenotype in normal tissues of prostates with metastasizing tumors implies that this
phenotype likely arises early in tumor evolution and signals a significant risk for metastasis.
Using the FT-IR models, evidence for the early development of an analogous phenotype for non-metastasizing tumors (a primary cancer DNA phenotype) was obtained with
BALB/c mice injected in the hind leg with the carcinogen 3-methylcholanthrene
(MCA).10 Strikingly, this primary phenotype appeared in histologically normal leg tissues
about eight weeks before the mean occurrence of palpable sarcomas and was indistinguishable
from the DNA structure of these tumors. We also found that the prodrug cyclophosphamide
delayed tumor formation by about 30%. Concomitantly, the MCA-injected mice receiving
the cyclophosphamide showed almost complete suppression of the primary phenotype.10
This finding firmly established that a relationship exists between the appearance of this
phenotype and the development of a tumor.
We reported in 1995 that the DNA structure of histologically normal tissue adjacent
to breast tumors (invasive ductal carcinomas) was spectrally indistinguishable from that of
the tumor.12 This evidence appears to be the first demonstration that a cancer DNA
phenotype occurs in normal tissues surrounding a tumor, and is consistent with data
showing that benign tissue in a cancerous breast has an elevated risk for developing a
tumor.13 In a recent study,9 we showed that 42% of cancer-free men (ages 55–80) had a
primary cancer DNA phenotype in histologically normal prostates, thus implying a high
risk for prostate cancer. The prostates of healthy younger men (ages 16–36) did not exhibit
this phenotype.
The discovery of the primary8-10 and metastatic8 cancer DNA phenotypes clearly calls
into question the widely-held concept that primary tumors somehow acquire mutations
that yield a few rare cells with the ability to metastasize. If metastasis were indeed to arise
in this way from a primary tumor, then the DNA structure of the metastasizing tumor (as
evinced by FT-IR spectroscopy) would be essentially identical to that of the primary
tumor. However, the DNA base and backbone structures of the non-metastasizing
(primary) and metastasizing tumors are, in fact, substantially different.8 Most significantly,
the DNA structures of the corresponding histologically normal tissues surrounding these
tumors exhibited virtually the same differences.8 These data challenge the classic theory
Received 08/12/04; Accepted 08/13/04
We have recently discovered, using statistical models of Fourier transform-infrared
spectra, two distinct cancer DNA phenotypes in histologically normal tissues surrounding
non-metastasizing primary tumors and tumors with evidence of distant metastases.
Structural comparisons of the DNA from these tumor types imply that each evolves via a
separate pathway from DNA phenotypes originating in progenitor cells. These findings
challenge the widely-held concept of metastasis.
Correspondence to: Donald C. Malins; Biochemical Oncology Program; Pacific
Northwest Research Institute; 720 Broadway; Seattle, Washington 98122 USA; Tel.:
206.726.1240; Fax: 206.726.1235; Email: [email protected]
Donald C. Malins
I am thankful to Katie M. Anderson, Edward A.
Barker, Naomi K. Gilman, Virginia M. Green, Karl
Erik Hellström, Thomas R. Sutter and Edward L.
Weber for their helpful comments.
Cell Cycle
2004; Vol. 3 Issue 10
of metastasis and support the hypothesis that primary and metastasizing
tumors evolve along separate pathways
established early in tumorigenesis (Fig.
1). How then might the primary and
metastatic cancer DNA phenotypes
We found that the DNA base and
backbone structures from the prostates
of cancer-free older men were markedly
different from those of younger men.9
In addition, we have shown that free
radical-induced base lesion concentrations (e.g., mutagenic 8-hydroxyadenine)
were considerably higher in the prostate Figure 1. Two-pathway model for non-metastasizing and metastasizing tumor development. (A) Changes in
DNA of older men.9 The introduction DNA as a function of genetic and epigenetic modifications produce distinctly different structures (1) that result
of just a single oxygen (a known radical- in the emergence of two independent tumor pathways. (B) Non-metastasizing tumor pathway. Progenitor cells
induced lesion) in an adenine or guanine expressing the primary cancer DNA phenotype (2) develop into a non-metastasizing primary tumor. (C)
of a 25-base DNA strand has been Metastasizing tumor pathway. Progenitor cells expressing the metastatic cancer DNA phenotype (3) develop
reported to significantly disrupt vertical into a metastasizing tumor. (See text for details)
base stacking interactions, as well as to
induce perceptible conformational changes in the backbone.14
10. Malins DC, Anderson KM, Gilman NK, Green VM, Barker EA, Hellström KE.
Development of a DNA cancer phenotype prior to tumor formation. Proc Natl Acad Sci
Other factors (e.g., hypermethylation)15 can also contribute to modUSA 2004; 10721-5.
ifications of DNA structure.
Collectively, these and possibly
11. Malins DC, Polissar NL, Su Y, Gardner HS, Gunselman SJ. A new structural analysis of
other structural perturbations arising largely as a function of increasDNA using statistical models of infrared spectra. Nat Med 1997; 3:927-30.
12. Malins DC, Polissar NL, Nishikida K, Holmes EH, Gardner HS, Gunselman SJ. The etiing age would be expected to create substantial genetic instability.
ology and prediction of breast cancer. Fourier transform-infrared spectroscopy reveals proThis instability may well provide a fertile environment for generatgressive alterations in breast DNA leading to a cancer-like phenotype in a high proportion
ing the primary and metastatic cancer DNA phenotypes. The genetof normal women. Cancer 1995; 75:503-17.
13. Henderson IC. Risk factors for breast cancer development. Cancer 1993; 71:2127-40.
ic and other structural features unique to each of these DNA phe14. Malins DC, Polissar NL, Ostrander GK, Vinson MA. Single 8-oxo-guanine and 8-oxo-adenotypes would then determine the biological properties of the tumor
nine lesions induce marked changes in the backbone structure of a 25-base DNA strand.
formed. These properties may be regulated via gene de-repression
Proc Natl Acad Sci USA 2000; 97:12442-5.
and include, for example, aggressiveness in primary tumors, or dis15. Yegnasubramanian S, Kowalski J, Gonzalgo ML, Zahurak M, Piantadosi S, Walsh PC, et
al. Hypermethylation of CpG islands in primary and metastatic human prostate cancer.
tant site preference18 and cellular heterogeneity19 in metastasizing
Cancer Res 2004; 64:1975-86.
16. Forde G, Gorb L, Shiskin O, Flood A, Hubbard C, Hill G, et al. Molecular structure and
In conclusion, we see no inconsistency between our proposed
properties of protonated and methylated derivatives of cytosine. J Biomol Struct Dyn 2003;
two-pathway model (Fig. 1) and recent findings based on gene
17. Banyay M, Gräslund A. Structural effects of cytosine methylation on DNA sugar pucker
expression profiles,3-5,20 or Slaughter’s 1953 concept of field cancerstudied by FTIR. J Mol Biol 2002; 324:667-76.
ization21 for which supportive genetic evidence has been recently
18. Kang Y, Siegel PM, Shu W, Drobnjak M, Kakonen SM, Cordón-Cardo C, et al. A multigenic program mediating breast cancer metastasis to bone. Cancer Cell 2003; 3:537-49.
provided.22,23 It appears that sufficient data now exist to justify that
19. Fidler IJ. The pathogenesis of cancer metastasis: The ‘seed and soil’ hypothesis revisited.
more attention be paid to mechanisms leading to the formation and
Nat Rev Cancer 2003; 3:453-8.
development of the DNA phenotypes described. Overall, we believe
20. Ramaswamy S, Ross KN, Lander ES, Golub TR. A molecular signature of metastasis in prithat the novel perspective presented here for the evolution of metasmary solid tumors. Nat Genet 2003; 33:49-54.
21. Slaughter DP, Southwick HW, Smejkal W. Field cancerization in oral stratified squamous
tasizing and non-metastasizing tumors from DNA phenotypes in
epithelium; clinical implications of multicentric origin. Cancer 1953; 6:963-8.
histologically normal cells has potentially profound implications for
22. Braakhuis BJ, Tabor MP, Kummer JA, Leemans CR, Brakenhoff RH. A genetic explanafuture cancer research and clinical practice.
tion of Slaughter’s concept of field cancerization: Evidence and clinical implications.
1. Nowell PC. The clonal evolution of tumor cell populations. Science 1976; 194:23-8.
2. Fidler IJ. Critical determinants of metastasis. Semin Cancer Biol 2002; 12:89-96.
3. Weigelt B, Glas AM, Wessels LFA, Witteveen AT, Peterse JL, van’t Veer LJ. Gene expression profiles of primary breast tumors maintained in distant metastases. Proc Natl Acad Sci
USA 2003; 100:15901-5.
4. Ma XJ, Salunga R, Tuggle JT, Gaudet J, Enright E, McQuary P, et al. Gene expression profiles of human breast cancer progression. Proc Natl Acad Sci USA 2003; 24:5974-9.
5. van’t Veer LJ, Dai H, van de Vijver MJ, He YD, Hart AAM, Mao M, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002; 415:530-36.
6. Weigelt B, van’t Veer LJ. Hard-wired genotype in metastatic breast cancer. Cell Cycle 2004;
7. Bernards R, Weinberg RA. A progression puzzle. Nature 2002; 418:823.
8. Malins DC, Gilman NK, Green VM, Wheeler TM, Barker EA, Vinson MA, et al.
Metastatic cancer DNA phenotype identified in normal tissues surrounding metastasizing
prostate carcinomas. Proc Natl Acad Sci USA 2004; 101:11428-31.
9. Malins DC, Johnson PM, Barker EA, Polissar NL, Wheeler TM, Anderson KM.
Cancer-related changes in prostate DNA as men age and early identification of metastasis
in primary prostate tumors. Proc Natl Acad Sci USA 2003; 100:5401-6.
Cancer Res 2003; 63:1727-30.
23. Yu YP, Landsittel D, Jing L, Nelson J, Ren B, Liu L, et al. Gene expression alterations in
prostate cancer predicting tumor aggression and preceding development of malignancy. J
Clin Oncol 2004; 22:2790-9.
Cell Cycle