Bst302.dvi

Biochemical Society Transactions (2003) Volume 31, part 2 Transgenic mouse models for studies of the role of
polyamines in normal, hypertrophic and
neoplastic growth
A.E. Pegg*1, D.J. Feith*, L.Y.Y. Fong†, C.S. Coleman*, T.G. O’Brien‡ and L.M. Shantz*
*Department of Cellular and Molecular Physiology, The Milton S. Hershey Medical Center, Pennsylvania State University College of Medicine, P.O. Box 850,
Hershey, PA 17033, U.S.A., †Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA 19107, U.S.A., and ‡Lankenau Institute for Medical Research, 100 Lancaster Avenue, Wynnewood, PA 19096, U.S.A.
Abstract
Transgenic mice expressing proteins altering polyamine levels in a tissue-specific manner have considerable
promise for evaluation of the roles of polyamines in normal, hypertrophic and neoplastic growth. This short
review summarizes the available transgenic models. Mice with large increases in ornithine decarboxylase
(ODC), S-adenosylmethionine decarboxylase or antizyme, a protein regulating polyamine synthesis by
reducing polyamine transport and ODC in the heart, have been produced using constructs in which the
protein is expressed from the α-myosin heavy-chain promoter. These mice are useful in studies of the
role of polyamines in hypertrophic growth. Expression from keratin promoters has been used to target
increased synthesis of ODC, spermidine/spermine-N1-acetyltransferase (SSAT) and antizyme in the skin.
Such expression of ODC leads to an increased sensitivity to chemical and UV carcinogenesis. Expression of
antizyme inhibits carcinogenesis in skin and forestomach. Expression of SSAT increases the incidence of skin
papillomas and their progression to carcinomas in response to a two-stage carcinogenesis protocol. These
results establish the importance of polyamines in carcinogenesis and neoplastic growth and these transgenic
mice will be valuable experimental tools to evaluate the importance of polyamines in mediating responses
to oncogenes and studies of cancer chemoprevention.
Mammalian polyamine biosynthesis and
donor, decarboxylated S-adenosylmethionine, is formed interconversion
from S-adenosylmethionine by S-adenosylmethionine de- Polyamines are small basic molecules that have many roles carboxylase (AdoMetDC). The sequential synthesis of in cellular physiology. There is a large body of evidence spermidine and spermine is then accomplished by two linking elevated polyamine levels to cancer and inhibitors of aminopropyltransferase enzymes, spermidine synthase and polyamine synthesis have activity as both anti-tumour agents and in cancer chemoprevention in high-risk populations.
These aminopropyltransferase reactions are essentially References to these studies can be found in numerous reviews irreversible, but reversal of the pathway can occur via the of the polyamine field (such as [1–5]). This article summarizes activity of FAD-dependent polyamine oxidases (PAOs) that studies with transgenic mice in which polyamine metabolism remove the aminopropyl groups [6]. The first such oxidase has been altered in specific tissues.
to be cloned, PAO1 (also called spermine oxidase, SMO), Polyamines are formed from methionine and arginine acts on spermine to generate spermidine [7]. However, a (Figure 1) [6]. Putrescine is formed by ornithine de- more widespread peroxisomal enzyme, usually referred to carboxylase (ODC) acting on ornithine generated by as PAO, acts on the N1-acetyl derivatives of spermine and arginase. The aminopropyl groups needed to convert spermidine, forming spermidine and putrescine respectively putrescine into the higher polyamines are provided by [6,8]. These acetylated polyamines are formed by the action of methionine after its conversion into S-adenosylmethionine spermidine/spermine-N1-acetyltransferase (SSAT) and this by methionine adenosyltransferase. The aminopropyl enzyme appears to be the rate-limiting step in the reversalpathway, which may play a major role in maintainingpolyamine homoeostasis [8].
Key words: acetyltransferase, antizyme, decarboxylase, putrescine, spermidine, spermine.
Although the polyamine biosynthetic enzymes are univer- Abbreviations used: ODC, ornithine decarboxylase; sally present in mammalian tissues, there are active uptake and SSAT, spermidine/spermine-N1-acetyltransferase; PAO, polyamine oxidase; AZ, antizyme; ES, embryonic stem; K5, K6 and K14, keratins 5, 6 and 14; DMBA, efflux systems for polyamines. Polyamines (and agmatine, 7,12-dimethylbenz(a)anthracene; DFMO, α-difluoromethylornithine; ERK, extracellular-signal- which can be converted into putrescine by agmatinase) regulated protein kinase; MEK, mitogen-activated protein kinase/ERK kinase; NMBA, are present in the diet and intestinal microflora so uptake N-nitrosomethylbenzylamine.
1 To whom correspondence should be addressed (e-mail [email protected]).
provides an alternative supply. This may be insignificant Polymines and Their Role in Human Disease The enzymes are shown in italics with grey background shading. PAO, polyamine oxidase; AdoMet, S-adenosylmethionine;dcAdoMet, decarboxylated AdoMet; MAT, methionine adenosyltransferase; MTA, 5 -methylthioadenosine; other abbreviationsare defined in the text.
under normal conditions. However, the up-regulation of this of ODC [12] and AdoMetDC [13] are lethal at the earliest transport system in response to polyamine depletion may stages of embryonic development. A line of mice termed provide a major method of resistance to treatments limiting Gy contains a deletion of part of the X chromosome that includes the spermine synthase gene. These mice can Antizyme (AZ) is a unique protein that regulates only be propagated on the B6C3H background. Males polyamine content (reviewed in [9,10]). AZ binds non- lacking spermine synthase are sterile, have a wide variety covalently to the ODC monomer. This binding inactivates of physical and neurological abnormalities and a very short ODC but, more importantly, targets it for degradation by life span (summarized in [14]). Fibroblast cell lines from the 26 S proteasome with release of free AZ, which can Gy mice [14] and ES cells with a targeted disruption of function again. The AZ mRNA contains an internal stop the spermine synthase gene [15] have been described and codon and a +1 frameshifting event is required for translation grow normally but differ in response to certain drugs. ES of the active protein. This frameshifting is increased by cells with gene deletions in SSAT (which is also located on high levels of polyamines, which therefore inhibit their the X chromosome) also grow normally and have normal own synthesis by increasing AZ and reducing ODC. The polyamine content [16] but their responses to stresses that polyamine transport system is also blocked by AZ through an may alter polyamines have not been reported.
unknown mechanism. It has also been suggested that AZ maystimulate polyamine excretion [11]. Therefore, AZ suppressespolyamine accumulation through multiple mechanisms. The Rodents with increased polyamine
rapid turnover of ODC requires a C-terminal domain that is metabolism in multiple tissues
not needed for activity. Deletion of this domain has therefore Transgenic mice and rats in which ODC, AdoMetDC, been used in some experiments to increase the steady-state spermidine synthase or SSAT has been increased generally level of ODC protein derived from transgenic constructs.
in multiple tissues have been described (see [17–21] andreferences therein). In these studies, either a metallothioneinpromoter was used or the inserted transgene was a fragment Mice with deletions of polyamine
of DNA containing the entire gene with its own associated biosynthetic enzymes
promoter. This work has been described extensively and Genetically altered mice or embryonic stem (ES) cells that will be not be covered here. The general overproduction of have been described at present include gene deletions of ODC led to an increase in putrescine but had much less ODC, AdoMetDC, spermine synthase and SSAT. Deletions effect on higher polyamines. This is due partly to the limited Biochemical Society Transactions (2003) Volume 31, part 2 AdoMetDC activity but mainly to the extensive homoeostatic as PMA was not needed for tumour development [28,29].
control of polyamine metabolism, which can still occur via Treatment with α-difluoromethylornithine (DFMO), an either post-translational mechanisms or regulatory influences ODC inhibitor, blocked the appearance of papillomas and mediated through the sequences in the inserted transgene.
caused a rapid regression of existing tumours in K6/ODC Mice overexpressing SSAT had a more striking phenotype with a substantial alteration in polyamine pools and a wide Further evidence for the importance of ODC in the variety of changes including hair loss, female infertility, development of neoplastic growth was obtained by studies in weight loss and altered lipid metabolism. The general nature which ODC expression was regulated by using a tetracycline- of the alteration in polyamine metabolism, the presence of responsive K6-driven construct. Down-regulation of the multiple secondary phenomena and the fact that these ODC expression by including doxycycline in the drink- animals are maintained on a mixed genetic background makes ing water reduced papilloma incidence in response to them unsuitable for studies on carcinogenesis and tumour The sensitivity of K6/ODC mice to carcinogens is not Although it was reported that there was no increase limited to the polycyclic aromatic hydrocarbon DMBA.
in spontaneous tumours in mice overexpressing ODC by Several other carcinogens from different chemical classes were increasing the gene copy number [22], a more recent study also shown to induce tumours in these mice within a short shows an increase in spontaneous tumour formation in mice period, suggesting that they may be a useful model system overexpressing ODC from the mouse mammary tumour for the rapid and sensitive identification of potential human virus long-terminal-repeat promoter [23].
carcinogens [32]. The use of chemical carcinogens to initiatetumorigenesis in K6/ODC mice is not absolutely required,as double transgenic mice combining the K6/ODC and Mice with altered cardiac polyamines
v-Ha-ras transgenes rapidly develop spontaneous squamoustumours of the skin [33]. These tumours regress rapidly on Transgenic mice have been generated in which either a C-terminally truncated ODC [24] or AdoMetDC [25] ODC also is implicated in photocarcinogenesis. K5/ODC is expressed in the heart via the α-myosin heavy-chain mice developed a high incidence of papillomas and squamous promoter. These animals have a massive increase in enzymic cell carcinomas and a 100% incidence of pigmented cysts activity but are viable. However, crosses in which both within 30 weeks of treatment with UVB radiation [35].
enzymes are expressed at this level are lethal in utero [25].
Shaved non-transgenic littermates and SKH1 mice (which The increase in ODC (>1000-fold) leads to a slight cardiac are also hairless) did not develop any tumours or cysts until hypertrophy but this is greatly increased upon treatment with 50 weeks of treatment. DFMO prevented UVB-induced isoproterenol and arginine, providing a useful model system tumour development in K5/ODC mice [35].
to study the role of polyamines in hypertrophic growth.
Transgenic ODC expression in K6/ODC mice leads to A constitutively active form of AZ (see below) has also increased tumour incidence in response to DMBA in other been expressed from this promoter. In these animals the rise strains including C3H/HeJ and FVB, although these were less in ODC in response to isoproterenol is blocked but the sensitive than C57BL/6J. This may allow the identification cardiac hypertrophy induced by this drug is not diminished of genes that influence skin tumour development whose expression or function is regulated by polyamines [36].
Notably, a high frequency of squamous cell carcinomas wasseen in K6/ODC mice on the FVB background in these Mice with increased ODC expression from
the bovine keratin 5 (K5) or keratin 6
(K6) promoters
Bovine K5 and K6 promoter elements have been used to di-
Mice with increased AZ or dominant-
rect the expression of a C-terminally truncated ODC to negative ODC expression from the K5 or
specific skin cell populations. This led to a large increase in K6 promoters
ODC activity with elevations in putrescine and spermidine.
The K5 and K6 promoters were also used to direct the The mice exhibited hair loss, dermal follicular cysts, increased expression of AZ to specific skin cell populations [37]. The nail growth and skin wrinkling. In old mice, there was a AZ cDNA construct used had a single nucleotide deletion high incidence of spontaneous squamous neoplasms with (T-205) to remove the requirement for polyamine-stimulated keratoacanthomas and well-differentiated papillomas [27].
frameshifting in the translation of AZ mRNA. The cDNA Breeding of these mice to the C57BL/6J background contained both potential start codons although Western blots abolished the development of spontaneous tumours but these of epidermal extracts indicated that the second site was mice were much more susceptible than controls to tumour utilized preferentially in vivo. Both K5/AZ and K6/AZ development after treatment with an initiating dose of with transgenic mice developed normally and were phenotypically 7,12-dimethylbenz(a)anthracene (DMBA) in adults or new- indistinguishable from wild-type littermates. The transgenic borns. Furthermore, treatment with a tumour promoter such AZ expression blocked the increase in skin ODC induced by Polymines and Their Role in Human Disease PMA and reduced epidermal and dermal polyamine content, of Ras activation pathways [40]. These conclusions are particularly spermidine. In DMBA/PMA carcinogenesis consistent with the response to DFMO. When K14/MEK studies on a mixed B6D2 genetic background, two founder mice were given DFMO in the drinking water from birth, lines of K6/AZ mice had a delay in tumour onset and a there was a dramatic delay in the onset of tumour growth reduction in tumour multiplicity. K5/AZ mice also developed (approx. 6 weeks), and only 25% of DFMO-treated mice fewer papillomas than littermate controls and combination developed tumours by 15 weeks of age compared with 100% of these lines to produce K5/K6 double-transgenic animals yielded an additive decrease in tumour multiplicity [37].
AZ expression from the K5 promoter is not limited to These transgenic lines were backcrossed on to the the skin but may occur in other epithelial cells. The K5/AZ carcinogenesis-resistant C57BL/6J inbred strain as well C57BL/6J mice were strikingly resistant to forestomach as the sensitive DBA/2J strain and again subjected to a carcinogenesis by N-nitrosomethylbenzylamine (NMBA) DMBA/PMA treatment protocol [38]. On the C57BL/6J [41]. Tumour formation in this model is increased by feeding background, tumour incidence and multiplicity were reduced a zinc-deficient diet, which induces forestomach epithelial in both K6/AZ lines. In K5/AZ transgenic mice, there was cell proliferation. This proliferative response was blocked in also a reduction in tumour multiplicity but the effect was K5/AZ mice and AZ was effective in reducing tumours in smaller and there was no difference in tumour incidence.
both zinc-sufficient and zinc-deficient mice. AZ expression This result is consistent with the more complete inhibition reduced the proliferating cell nuclear antigen (‘PCNA’)- of ODC activity in K6/AZ compared with K5/AZ mice in labelling index and the content of cyclin D1 and its catalytic response to PMA application. On the DBA/2J background, partner Cdk4, key regulatory proteins controlling G1-to-S both K5/AZ and K6/AZ developed fewer tumours with a progression. AZ increased expression of Bax and increased the apoptopic index in the forestomach. These results The studies with ODC- and AZ-expressing mice show demonstrate that overexpression of AZ stimulates apoptosis clearly that increased polyamine content plays a critical role and restrains cell proliferation. Very similar changes were also in tumour development. A dominant-negative form of ODC produced in the NMBA/zinc-deficient mice by treatment was not effective in blocking carcinogenesis but did not fully with DFMO, confirming the importance of polyamines in lower endogenous ODC, which was stabilized due to the binding and sequestration of AZ by the stable dominant In summary, the K5/AZ and K6/AZ mice demonstrate that negative ODC, supporting the key physiological role of AZ AZ suppresses tumour growth in at least two animal cancer models and provide a valuable model system to evaluate the The hypothesis that ODC induction by pathways down- role of ODC and polyamines in tumorigenesis.
stream of oncogenic ras is a necessary step in carcinogenesiswas tested using a transgenic mouse line overexpressinga constitutively active mutant of mitogen-activated pro-tein kinase/extracellular-signal-regulated protein kinase Mice with increased SSAT expression from
(ERK) kinase (MEK) in the skin under the control of the K6 promoter
the keratin 14 promoter (K14/MEK mice). Activation of K6/SSAT mice appeared to be phenotypically normal the Raf/ERK pathway in these mice leads to moderate and were indistinguishable from normal littermates. A hyperplasia, with spontaneous skin tumour development preliminary experiment using hybrid B6D2 mice showed within 5 weeks of birth. The tumours had high levels of an increase in skin tumour incidence and multiplicity in ODC protein and activity, indicating that Raf/ERK activation response to a two-stage tumorigenesis protocol [42]. A is a sufficient stimulus for ODC induction. The K14/MEK more detailed study using K6/SSAT transgenic mice on the mice on the ICR background were crossed with K5/AZ or C57BL/6J background showed a 10-fold increase in the K6/AZ mice on both the carcinogenesis-resistant C57BL/6J number of epidermal tumours that developed in response background and the sensitive DBA/2J background. Ex- to a single application of DMBA followed by 19 weeks pression of AZ driven by either promoter significantly of promotion with PMA [43]. Tumours from transgenic delayed tumour incidence and reduced tumour multiplicity animals showed marked elevations in SSAT enzyme activity on both backgrounds [40]. Since MEK overexpression causes and SSAT protein levels compared with tumours from non- hyperplasia and the K6 promoter requires hyperproliferation transgenic littermates, and the accompanying changes in for maximal expression, K14/MEK-K6/AZ mice test whether putrescine and N1-acetylspermidine pools indicated activa- ODC inhibition is effective in preventing tumour formation tion of SSAT-mediated polyamine catabolism in transgenic after a carcinogenic environment is established. The K5 animals. Of particular interest was that an unusually high promoter is constitutive, and K14/MEK-K5/AZ mice express number of tumours rapidly progressed to carcinomas in AZ from birth in the same cells as the activated MEK protein.
the K6/SSAT mice. These findings are not due to the These mice represent a model in which ODC is inhibited at integration site of the SSAT transgene since they have now the time of the initiating stimulus.
been confirmed using a second K6/SSAT transgenic line These results indicate that increased ODC activity is (C.S. Coleman, T.G. O’Brien and A.E. Pegg, unpublished central to tumour development in response to overexpression Biochemical Society Transactions (2003) Volume 31, part 2 Although total polyamines were not reduced by the 12 Pendeville, H., Carpino, N., Marine, J.C., Takahashi, Y., Muller, M., Martial, induction of the SSAT/PAO pathway, the relative content J.A. and Cleveland, J.L. (2001) Mol. Cell Biol. 21, 6459–6558
13 Nishimura, K., Nakatsu, F., Kashiwagi, K., Ohno, H., Saito, H., Saito, T. and of polyamines was strikingly affected and the increases in Igarashi, K. (2002) Genes Cells 7, 41–47
putrescine and N1-acetylspermidine may indicate a key role 14 Mackintosh, C.A. and Pegg, A.E. (2000) Biochem. J. 351, 439–447
for these polyamines in chemically induced mouse skin 15 Korhonen, V.-P., Niranen, K., Halmekyto, M., Pietil ¨a, M., Diegelman, P., Parkkinen, J.J., Eloranta, T., Porter, C.W., Alhonen, L. and J ¨anne, J. (2001) neoplasia. Another possibility is that the increased oxidative Mol. Pharmacol. 59, 231–238
damage as a result of the elevated SSAT/PAO pathway is 16 Niiranen, K., Pietil ¨a, M., Pirttil ¨a, T.J., J ¨arvinen, A., Halmekyto, M., responsible for the advanced tumour phenotype. The extent Korhonen, V.-P., Kein ¨anen, T., Alhonen, L. and J ¨anne, J. (2002) J. Biol.
Chem. 277, 25323–25328
to which high levels of SSAT may influence carcinogenesis in 17 Kauppinen, R.A. and Alhonen, L.I. (1995) Progr. Neurobiol. 47, 545–563
18 Heljasvaara, R., Veress, I., Halmekrt ¨o, M., Alhonen, L., J ¨anne, J., Laakala, P. and Pajunen, A. (1997) Biochem. J. 323, 457–462
19 Suppola, S., Heikkinen, S., Parkkinen, J.J., Uusi-Oukari, M., Korhonen, V.-P., Kein ¨anen, T., Alhonen, L. and J ¨anne, J. (2001) Biochem. J. 358, 343–348
Conclusions
20 Alhonen, L., R ¨as ¨anen, T.L., Sinervirta, R., Parkkinen, J.J., Korhonen, V.-P., Mouse models in which polyamine levels in particular cell Pietil ¨a, M. and J ¨anne, J. (2002) Biochem. J. 362, 149–153
21 R ¨as ¨anen, T.-L., Alhonen, L., Sinervirta, R., Kein ¨anen, T., Herzig, K.-H., types are perturbed by transgenic expression of proteins Suppola, S., Khomutov, A.R., Veps ¨al ¨ainen, J. and J ¨anne, J. (2002) J. Biol.
influencing polyamine metabolism are currently available and Chem. 277, 39867–39872
others could be developed. Modification of the cDNA for 22 Alhonen, L., Halmekyt ¨o, M., Kosma, V.M., Wahlfors, J., Kauppinen, R. and J ¨anne, J. (1995) Int. J. Cancer 63, 402–404
the expressed protein to avoid post-transcriptional regulation 23 Kilpel ¨ainen, P., Saarimies, J., Kontusaari, S.I., J ¨arvinen, M.J. and Soler, A.P.
of the protein expression may be needed to maximize the (2001) Int. J. Biochem. Cell Biol. 33, 507–520
alterations in polyamine content. These models provide 24 Shantz, L.M., Feith, D.J. and Pegg, A.E. (2001) Biochem. J. 358, 25–32
25 Nisenberg, O., Shantz, L.M. and Pegg, A.E. (2002) FASEB J. 16, A1115
opportunities for a greater understanding of the importance 26 Mackintosh, C.A., Feith, D.J., Shantz, L.M. and Pegg, A.E. (2000) of polyamines in normal, hypertrophic and neoplastic growth Biochem. J. 350, 645–653
and may in the future provide useful model systems to guide 27 Megosh, L., Gilmour, S.K., Rosson, D., Peralta Soler, A., Blessing, M., the optimal use of the polyamine pathway for the treatment Cancer Res. 55, 4205–4209
28 O’Brien, T.G., Megosh, L.C., Gilliard, G. and Peralta Soler, A. (1997) Cancer Res. 57, 2630–2637
29 Megosh, L., Halpern, M., Farkash, E. and O’Brien, T.G. (1998) Mol. Carcinogen. 22, 145–149
Research on this topic in the authors’ laboratories is supported by 30 Soler, A.P., Gilliard, G., Megosh, L., George, K. and O’Brien, T.G. (1998) grants CA-18138 (A.E.P.) and CA-82768 (L.M.S.) from the National Cancer Res. 58, 1654–1659
31 Guo, Y., Zhao, J., Sawicki, J., Peralta-Soler, A. and O’Brien, T.G. (1999) Cancer Institute, National Institutes of Health (NIH), U.S.A.; grant ES- Mol. Carcinogen. 26, 32–36
01664 (T.G.O.) from the National Institute of Environmental Health 32 Chen, Y., Megosh, L.C., Gilmour, S.K., Sawicki, J.A. and O’Brien, T.G.
Sciences, NIH, U.S.A.; and grant AHA 0040140N (L.M.S.) from the (2000) Toxicol. Lett. 116, 27–35
33 Smith, M.K., Trempus, C.S. and Gilmour, S.K. (1998) Carcinogenesis 19,
34 Lan, L., Trempus, C. and Gilmour, S.K. (2000) Cancer Res. 60, 5696–5703
35 Ahmad, N., Gilliam, A.C., Katiyar, S.K., O’Brien, T.G. and Mukhtar, H.
References
(2001) Am. J. Pathol. 159, 885–892
36 Megosh, L.C., Hu, J., George, K. and O’Brien, T.G. (2002) Genomics 79,
1 Pegg, A.E. (1988) Cancer Res. 48, 759–774
2 Marton, L.J. and Pegg, A.E. (1995) Annu. Rev. Pharm. 35,
37 Feith, D.J., Shantz, L.M. and Pegg, A.E. (2001) Cancer Res. 61,
3 Cohen, S.S. (1998) A Guide to the Polyamines, Oxford University Press, 38 Feith, D.J. and Pegg, A.E. (2002) Proc. Am. Assoc. Cancer Res. 43, 5012
39 Guo, Y., Shantz, L.M., Pegg, A.E., Sawicki, J.A. and O’Brien, T.G. (2002) 4 Casero, Jr, R.A. and Woster, P.M. (2001) J. Med. Chem. 44, 1–26
Carcinogenesis 23, 657–664
5 Bachrach, U., Wang, Y.C. and Tabib, A. (2001) News Physiol. Sci. 16,
40 Feith, D.J., Duong-Ly, K.C. and Shantz, L.M. (2002) Proc. Am. Assoc.
Cancer Res. 43, 5011
6 Pegg, A.E. (1986) Biochem. J. 234, 249–262
41 Fong, L.Y.Y., Feith, D.J., Nguyen, V.T. and Pegg, A.E. (2002) Proc. Am.
7 Wang, Y., Devereux, W., Woster, P.M., Stewart, T.M., Hacker, A. and Assoc. Cancer Res. 43, 851
Casero, R.A. (2001) Cancer Res. 61, 5370–5373
42 Coleman, C.S., O’Brien, T.G., Megosh, L.C. and Pegg, A.E. (2000) 8 Casero, R.A. and Pegg, A.E. (1993) FASEB J. 7, 653–661
Proc. Am. Assoc. Cancer Res. 41, 542
9 Hayashi, S., Murakami, Y. and Matsufuji, S. (1996) Trends Biochem. Sci.
43 Coleman, C.S., Pegg, A.E., Megosh, L.C., Guo, Y., Sawicki, J.A. and O’Brien, 21, 27–30
T.G. (2002) Carcinogenesis 23, 359–364
10 Coffino, P. (2001) Nat. Rev. Mol. Cell. Biol. 2, 188–194
11 Sakata, K., Kashiwagi, K. and Igarashi, K. (2000) Biochem. J. 347,

Source: http://www.asp.uma.es:8080/asp_vieja/pdf/pdf-2.pdf