Somatic mutations are frequent and increase with age in human kidney epithelial cells
Somatic mutations are frequent and increase with age in human kidney epithelial cellsGeorge M. Martin*, Charles E. Ogburn, Lorel M. Colgin, Allen M. Gown, Steven D. Edland1 and Raymond J. Monnat, Jr
Departments of Pathology and 1Biostatistics, University of Washington, Seattle, WA 98195, USA
Received September 20, 1995;Revised and Accepted December 4, 1995
We have used a primary cloning assay to determine the frequency of 6-thioguanine (TG)-resistant tubular epithelial cells in kidney tissue from 72 human donors ranging in age from 2 to 94 years. The frequency of TG-resistant mutants ranged from ~5 * 10-5 for donors in the first decade of life to ~2.5 * 10-4 for donors in the eighth and later decades of life. Two different statistical analyses indicated that this increase in mutant frequency is exponential with age. We also observed a 2-fold higher TG-resistant mutant frequency in nephrectomy kidneys containing a coincident renal carcinoma. DNA sequence analyses revealed HPRT gene mutations in each of 14 TG-resistant mutants from seven unrelated donors. Thirteen of these 14 mutants resulted from independent mutational events. These results suggest that somatic mutations are common in renal-and perhaps in other human-epithelia, and thus could play an important role in the genesis of age-associated disease.
Somatic mutations were first suggested, nearly 40 years ago, to play a role in the pathogenesis of age-associated disease (1 ,2 ). However, comparatively few data exist as yet on somatic mutation frequency as a function of age, or on the role of specific somatic mutations in the genesis of age-associated disease apart from cancer. Particularly surprising is the lack of data on somatic mutations in epithelial cells. Epithelial cells play critical roles in tissue structure and function, and are the target tissue for development of most human age-associated neoplasms (3 ).
We have used the HPRT locus to quantify and characterize somatic mutations occurring in vivo in human kidney cortical epithelial cells. The mutational target in these analyses, the X-linked HPRT locus, encodes the purine salvage enzyme hypoxanthine phosphoribosyl transferase (EC 2.4.2.8). The HPRT locus is widely used for mammalian mutation analyses due to a combination of facile somatic cell and molecular genetic methods, and dispensibility of the HPRT protein in most somatic cells (4 -6 ).
The frequency of renal cortical cells resistant to the purine analogue 6-thioguanine (TG) was quantified using a primary epithelial cell cloning assay. Mutations were characterized by DNA sequence analysis of HPRT cDNA or genomic DNA segments recovered from independent, TG-resistant mutants. These data indicate that HPRT-deficient kidney tubular epithelial cells are common, arise from a wide range of mutational events, and show an exponential increase in frequency over the human lifespan. These results, to the best of our knowledge, represent the first analysis of the frequency and molecular basis for somatic mutation in a human epithelial cell lineage as a function of donor age.
Human kidneys from three sources were analyzed: unused transplant kidneys, nephrectomy specimens and autopsy kidneys. Virtually all (~95-98%) of the primary cell clones we grew from collagenase-dissociated kidney cortex displayed an epithelioid morphology. There was considerable heterogeneity in the proliferative capacity of both primary TG-resistant and control clones, with only a subset (5-10%, depending on the donor) reaching sufficient size (~106 cells or 20 cumulative population doublings) for initial molecular genetic analysis.
Immunocytochemical analyses of several independent TG-resistant and control clones from different donors confirmed their epithelial character: cytokeratin 8 and stratum corneum-associated cytokeratin expression were observed in an intermediate filament pattern with variable vimentin co-expression (Fig. 1 A-C), whereas no desmin, muscle actin or factor VIII expression was observed (Fig. 1 D). The remaining few percent of colonies we recovered had a fibroblastoid morphology and fibroblast-associated immunocytochemical markers (data not shown).
The quantitative and molecular data described above provide an interesting glimpse of somatic mutation in a human epithelial cell lineage. Morphologic and immunocytochemical features of TG-resistant and control clones grown from human kidney cortex suggests that a majority ( >= 95%) were epithelial-derived. The most probable origin for these clones is from convoluted tubule epithelial cells, as these are the most abundant of the cortical epithelial cell types (8 ,9 ). It should be possible to more precisely identify the specific tubular region(s) from which these cells are derived by further studies that use a combination of immunocytochemical, in situ hybridization or PCR and functional analyses (9 ).
Renal tubular epithelial cells are an interesting target tissue for somatic mutation analyses. They are metabolically active and consume large amounts of oxygen; they perform a majority of the secretory and resorptive work needed to process glomerular filtrate; they are mitotically competent, though largely quiescent, in the absence of a growth stimulus or injury; and they are the cell type from which the predominant primary human renal tumor, renal adenocarcinoma, arises (8 ,10 -13 ). The ability to obtain proliferating colonies of renal tubular epithelial cells from routine human autopsy kidney is also of interest as it expands the investigative potential of the autopsy (see, e.g., 14 ).
Virtually all of the TG-resistant mutants we recovered are likely to have arisen in vivo. We applied TG selection within 24 hours of plating cell suspensions, and grew presumptive TG-resistant mutants continuously in TG until mutation analysis. A delay or phenotypic `lag' is commonly seen in the expression of induced HPRT mutations, and low levels of HPRT enzyme activity (<2% of control activity) have been found in most human and animal cells resistant to 10 µg/ml (60 µM) TG (4 -6 ,15 -17 ). These observations suggest that continuous TG selection should select strongly for the growth of preexisting HPRT-deficient mutants that arose in vivo, and against new TG-resistant mutants that arose during cell isolation or rare, TG-sensitive cells that can survive, though not grow, in the presence of TG (18 and unpublished results).
Our initial molecular genetic analyses of TG-resistant and control clones focused on the 5-10% of clones from each donor that had sufficient proliferative potential to grow to ~106 cells. The considerable heterogeneity we observed in the proliferative capacity of both TG resistant and control clones is most likely the result of clonal attenuation, in which proliferating primary somatic cells continually segregate progeny with variable, though declining, growth potential (19 ). All of the TG-resistant clones we examined contained HPRT mutations that would eliminate or severely attenuate HPRT enzyme activity (4 -7 ), and virtually all appear to have arisen as independent mutational events (Table 1 ). In order to determine a representative spectrum of HPRT mutations occurring in human kidney tubular epithelial cells, we are characterizing a larger number of mutations in clones with a range of different growth potentials.
The TG-resistant mutant frequencies we observed in primary renal epithelial cells is approximately 10-fold higher than have been described in human peripheral blood T lymphocytes (20 ). One explanation for this could be differences in the TG concentration at which the mutant frequency reaches a plateau. This possibility is unlikely for at least two reasons. The mutant frequency for kidney tubular epithelial cells falls to a minimum at 60 µM TG, and mutants recovered at this TG concentration contain HPRT mutations similar to those recovered from human T-lymphocytes or continuous cell lines at comparable TG concentrations (7 ,20 ,21 ). The higher mutant frequency in renal epithelial cells is also unlikely to be explained by an age-dependent decrease in CFE, as has been described for T-cells (20 ); the CFE of renal epithelial cells does not appear to decline with age.
One unexpected finding in mutant frequency analyses was a consistently higher frequency of TG-resistant mutants in donor kidneys containing a coincident renal carcinoma. This increase cannot be explained simply by the greater average age of nephrectomy versus other kidney donors. An elevated mutant frequency in tumor nephrectomy kidneys could have several different and mechanistically interesting explanations: patients with renal carcinomas could have had unusual or excessive exogenous mutagen exposure; carcinoma patients could have a heritable or acquired DNA repair deficit or genomic instability; or they could have relaxed somatic selection against, and thus a lower probability of eliminating, TG-resistant HPRT-deficient mutant cells as they arise. Additional data on the spectrum of HPRT mutations recovered from nephrectomy kidneys may begin to shed mechanistic light on these intriguing possibilities. The observation of a higher mutant frequency in kidneys having a coincident renal carcinoma may be clinically relevant, as there is renewed interest in nephron sparing therapy in which a renal carcinoma, though not adjacent normal kidney, is removed at the time of surgery (21 ).
Our data indicate that TG-resistant renal cortical epithelial cells are frequent, and that the frequency of mutant cells appears to increase exponentially with donor age. This apparent exponential increase, and comparable age-dependent increases in TG-resistant mutant frequencies in human and mouse T-lymphocytes (20 ,22 -25 ), are compatible with `error catastrophe' and other somatic mutation-based theories of aging (26 ). One as yet unexplained finding is that we have seen comparable age-dependent increases in TG-resistant mutant frequency in human and dog kidney, though not in mouse kidney or mouse skeletal muscle fibroblasts (18 ,27 ). This difference is unlikely to reflect strong, species-specific selection against HPRT-deficient cells: the frequency of TG-resistant T-lymphocytes increases in aged mice in tissue where the selection against HPRT-deficient variants appears to be more stringent than in slowly dividing tissues such as kidney (25 ,28 ,29 ).
These results provide, to the best of our knowledge, the first analysis of mutant frequency and of the molecular characteristics of somatic mutations arising in a human epithelial cell lineage. Somatic mutations in kidney, and perhaps in other human epithelia, are common and appear to increase exponentially with age. The presence of independent mutational events in different TG-resistant mutants from individual kidneys indicates that a simple enumeration of TG-resistant colonies can, in most instances, be used to estimate in vivo mutation frequencies. It should also be possible to recover and analyze large numbers of independent mutations from single donor kidneys. This will allow us to better determine the spectrum, or frequency and distribution of different molecular types, of HPRT mutations occurring in kidney epithelial cells in vivo. These data will also allow us to determine whether the kidney HPRT mutation spectrum differs in important ways from HPRT mutational spectra observed in T lymphocytes or other human cells (7 ,20 ). These approaches should allow us to develop a better understanding of the role of somatic mutation accumulation and of specific types of mutations in the loss of renal function and elevated risk of malignancy that accompany human aging (26 ,30 -34 ).
A total of 72 human kidneys, from three sources, were used in this study: (i) kidneys obtained for renal transplantation (n = 27) by the Northwest Kidney Center (Seattle, WA) that could not be used for technical reasons; (ii) nephrectomy specimens (n = 40) from the University of Washington and Swedish Hospital Medical Centers (Seattle, WA). A majority (36 of 40) consisted of normal tissue from kidneys removed for histopathologically verified renal adenocarcinoma or, in one case, invasive non-papillary, poorly differentiated transitional cell carcinoma. None of these renal carcinoma patients had received radiation or chemotherapy prior to surgery. The remaining four nephrectomy specimens came from horseshoe or hypoplastic kidneys (one each), or from nephrectomy kidneys containing hematoma or ischemia-associated tubular necrosis; and (iii) autopsy kidneys (n = 5) from the rapid autopsy program of the University of Washington Alzheimer's Disease Research Center. All five autopsy kidneys were from patients with pathologically confirmed Alzheimer's disease. Transplant kidneys were perfused with Belzer's solution (Dupont) supplemented with 200 000 units/l penicillin G sodium, 4 mg/l dexamethasone and 0.4 ml/l insulin, after which vessels were clamped and the kidneys placed on ice for transportation. The range of times on ice for the specimens used for these studies was 24 to 96 h. Nephrectomy and autopsy specimens were aseptically dissected, and tissue samples transported in cell culture medium (DMEM; Dulbecco-Vogt modified Eagle's medium supplemented with 100 units/ml of penicillin G sodium and 100 µg/ml of streptomycin) supplemented with 30 mM HEPES buffer (pH 7.4). Grossly normal cortical tissue from the pole opposite tumor was used from nephrectomy specimens containing a carcinoma. Tissue samples from a subset of the kidneys (n = 25) were paraffin embedded, sectioned and stained with hematoxylin and eosin for morphologic analysis. Independent tissue samples from opposite poles or from the anterior and posterior cortex at each pole were obtained from a subset of kidneys to examine intra-organ variability in mutant frequencies.
Primary clones were grown from cell suspensions prepared by collagenase dissociation of kidney cortical tissue. Cortical tissue samples (~3 g/kidney) were aseptically dissected to remove renal capsule and medullary tissue, then minced to obtain ~1 mm3 tissue fragments. These were washed 2-3 times in Ca+2 and Mg+2-free phosphate-buffered normal saline (pH 7.1) (PBS; Grand Island/BRL), then digested with type 1 collagenase (Sigma) to generate cell suspensions. Washed tissue fragments were resuspended in 50 ml of a 1:1 (v/v) mixture of type 1 collagenase (1 mg/ml in PBS) and DMEM. Digestion was performed in a 100 ml homogenizing flask with gentle stirring at 37oC. Supernatant was collected after 1 h and remaining tissue was digested with an additional 50 ml of the collagenase:DMEM mixture for 1 h at 37oC. Supernatants were combined and centrifuged at 2600 * g for 5 min. The resulting cell pellet was resuspended in 50 ml of growth medium (DMEM supplemented with 16% (v/v) fetal bovine serum) for plating.
The CFE of cell suspensions was determined by inoculating 0.1 ml of a 1:100 dilution of cell suspensions into 15 ml of growth medium in each of twenty 100 mm culture plates. Thioguanine (TG)-resistant clones were isolated by inoculating 0.5 ml of undiluted cell suspension in 10 ml of growth medium into each of an additional eighty 100 mm plates. After 24 h, these plates were refed with 15 ml of growth medium supplemented with 10 µg/ml (60 µM) 6-thioguanine (TG). Plates under TG selection contained 5.2-10.6 × 104 viable cells/plate. All plates were grown in a water vapor-saturated 37oC, 7% CO2 incubator, and were refed with fresh growth medium twice each week. TG-resistant colonies were grown continuously in the presence of TG-containing medium. Colonies on 10 CFE and at least 20 TG selection plates were counted after staining plates with 0.5% (w/v) crystal violet in 20% (v/v) ethanol. Mutant frequencies were calculated by dividing the total number of TG-resistant colonies on selection plates by the mean number of colonies on CFE plates corrected to take into account dilution differences at the time of inoculation. TG resistant and control clones were isolated for each kidney using 10 mm glass cloning cylinders. We isolated multiple TG-resistant clones from a randomly chosen subset of donors from whom we received large amounts of kidney tissue. These mutants were analyzed to determine whether different TG-resistant mutants contained independent mutational events. TG-resistant and control clones for these initial mutation analyses were grown to 5-10 × 106 cells/clone, then PBS washed, pelleted and stored at -80oC for molecular analyses.
Early passage cells, from colonies that had undergone 6 to 8 population doublings, were plated in growth medium on multiwell teflon-coated slides (Meloy Laboratories, Springfield, VA) and allowed to attach overnight, then PBS washed and fixed in 95% ethanol or acetone (-20oC) for 10 min before being air-dried. Indirect immunofluorescence was performed as previously described using monoclonal antibodies to vimentin (43[beta]E8); cytokeratin 8 (35[beta]H11); stratum corneum keratins (34[beta]E12), muscle actins (HHF35); and commercially available monoclonal antibodies to desmin and factor VIII (35 ,36 ).
DNA and RNA for mutation analyses were obtained from TG-resistant and control clones using a modification of the method developed by Henikoff that omitted the DNase step (37 ). Aliquots of the final preparation were resuspended in TE buffer (10 mM Tris/1 mM EDTA, pH 8.0) for RT-PCR or PCR amplification as previously described (38 ). The primers used for cDNA amplification were 1676 (-) and 41502 (+) (see primer listing below). The resulting product was a 710 bp DNA containing the complete 657 bp human HPRT protein coding region. Genomic DNA amplifications were performed with primers flanking the intron/exon boundaries of HPRT exon 8 [39848 (-) and 40195 (+)]. The resulting 387 bp DNA fragment contained both the 3' acceptor splice site of intron 7 and the 5' donor splice site of intron 8. Direct DNA sequencing of amplified fragments was used to identify HPRT mutations. DNA fragments were purified by electrophoresis through 1% low melting point agarose gels, excised and melted at 70oC for 10 min. Aliquots were then added directly to sequencing reactions. Sequencing was performed using the Cycle Sequencing System kit and primers end-labelled with [gamma]32P-ATP and polynucleotide kinase according to the manufacturer's protocol (Life Technologies). Each strand was sequenced with four different primers. Control clones isolated from the same donor were amplified, sequenced and run in parallel. Putative mutations were verified by repeating amplification(s) and DNA sequencing in all instances. The primers used for amplification and sequence analyses are listed below. Primer position refers to the position of the 3' end of the primer in the human HPRT genomic DNA sequence (39 ). Primers followed by a `(+)' hybridize to the HPRT mRNA or to the sense strand of genomic DNA, while primers followed by a `(-)' hybridize to the mRNA complementary strand of HPRT cDNAs or to the antisense strand of genomic DNA. Position Sequence 1676(-) 5'-CCGCGCGCTGGCCGGATCCGTT-3'14873(-) 5'-GGGTGTTTATTCCTCATGGAC-3'16639(+)5'-CACAGAGGGCTACAATGTGATGGCCTCCC-3'16760(-) 5'-GTGATGGATCCATTCCTATGACTGTAG-3'27891(+) 5'-TTAGGTACCCTGTTGACTGGTCATT-3'34946(+) 5'-CAAAGCTTGCATTGTTTTGCCAGTGTC-3'35016(-) 5'-CAGTAGGATCCAAAGATGGTCAAGGTCGC-3'39848(-) 5'-GACCCCACGAAGTGTTGGA-3'40195(-) 5'-GGTGCTGGAAGGAGAAAACA-3'41502(+) 5'-GTTTCCAAACTCAACTTGAATTCTCATC-3'
Two different statistical analyses, least squares regression analysis and the Box-Cox [lambda] power transformation, were performed to investigate the relationship of mutant frequency to age. For least squares regression analysis, linear and exponential models of mutant frequency as a function of age were fit and compared using p values and R2 percent of variance explained. The influence of kidney source on mutant frequency was tested using regression analysis of the log-transformed data after controlling for donor age. Mutant frequency data were also analyzed using the Box-Cox [lambda] power transformation (40 ) as implemented by Venables and Ripley (41 ). The Box-Cox [lambda] statistic estimates, by maximum likelihood, the power transformation of mutant frequency as a function of age that best fits the data. A Box-Cox [lambda] of 0 indicates a log transformation and exponential relationship between mutant frequency and donor age, while a [lambda] of 1 indicates no transformation and a linear relationship between mutant frequency and age. If 95% confidence intervals around the estimated [lambda] ([lambda]) do not contain 1, the null hypothesis of a linear relationship can be rejected in favor of a greater than linear increase in mutant frequency with age. The increasing variability seen in mutant frequency with age may itself favor the log transformation in Box-Cox [lambda] analyses. Thus it is important to confirm the results of Box-Cox [lambda] analyses with a second independent method such as least squares regression analysis.
We thank the physicians and staff of the Northwest Kidney Center, of the Surgical Pathology services of the University of Washington (Dr Rodger C. Haggitt and colleagues) and Swedish Medical Centers (Dr Bruce G. Kulander and colleagues), and of the Rapid Autopsy Program of the University of Washington Alzheimer Disease Research Center for their assistance in obtaining kidneys for this study. All kidneys were obtained and studied under a protocol approved by the University of Washington Human Subjects Review Committee. We also thank Drs Earl Benditt, Colin Manoil, Stanley Gartler and Adonis Skandalis for thoughtful comments on a draft of the manuscript. This work was supported by grants AG01751, AG00057 and P50 AG05136 from the National Institute on Aging to GMM, and by NCI grant CA48022 to RJM, Jr.
2 Failla, G. (1958) The aging process and carcinogenesis. Ann. N.Y Acad. Sci. 74, 1124-1135.
3 Szilard, L. (1959) On the nature of the aging process. Proc. Natl. Acad. Sci. USA45, 30-45.
4 Wright, N. and Alison, M. (1984) The Biology of Epithelial Cell Populations, Clarendon Press, Oxford.
5 Jacobs, L. and DeMars, R. (1984) In Kilbey, B. J., Legator, M., Nichols, W., and Ramel, C. (eds.), In Handbook of Mutagenicity Test Procedures, Elsevier Science Publishers, Amsterdam, pp. 321-356.
6 Fenwick, R. G. (1985) In Gottesman, M. (ed.), Molecular Cell Genetics, Wiley, New York, pp. 333-373.
7 Rossiter, B. J. F. and Caskey, C. T. (1995) In Scriver, C. R., Beaudet, A. L., Sly, W. S., and Valle, D. (eds.), The Metabolic and Molecular Bases of Inherited Disease, 7th Ed., McGraw-Hill,Inc. Ney York, pp. 1679-1706.
8 Cariello, N. F. (1994) Database and software for the analysis of mutations in the human hprt gene. Nucleic Acids Res.22, 3547-3548.MEDLINE Abstract
9 Wright, N. and Alison, M. (1984) The Biology of Epithelial Cell Populations, Clarendon Press, Oxford, pp. 981-1003.
Hebert, S. C. (1992) In Windhager, E. E. (ed.), Handbook of Physiology Section 8: Renal Physiology, Oxford University Press, New York, pp. 876-926.
10 Schoolwerth, A. C. and Drewnowska, K. (1995) In Massry, S. G. and Glassock, R. J. (eds.), Massry and Glassock's Textbook of Nephrology, Williams & Wilkins, Baltimore, pp. 123-137.
11 Berry, C. A. and Rector, Jr. (1991) In Brenner, B. M. and Rector, Jr. (eds.), The Kidney, W.B. Saunders Company, Philadelphia, pp. 245-282.
12 Tannenbaum, M. (1971) Ultrastructural pathology of human renal cell tumors. Pathology Ann. 6, 249-277. MEDLINE Abstract
13 Bennington, J. L. and Beckwith, J. B. (1975) Tumors of the Kidney, Renal Pelvis, and Ureter, Armed Forces Institute of Pathology, Washington,D.C.
14 Pellett, O. L., Smith, M. L., Thoene, J. G., Schneider, J. A., and Jonas, A. J. (1984) Renal cell culture using autopsy material from children with cystinosis. In Vitro 20, 53-58.MEDLINE Abstract
15 Monnat, R. J.,Jr. (1989) Molecular analysis of spontaneous hypoxanthine phosphoribosyltransferase mutations in thioguanine-resistant HL-60 human leukemia cells. Cancer Research 49, 81-87.MEDLINE Abstract
16 Fukuchi, K., Martin, G. M., and Monnat, R. J.,Jr. (1989) Mutator phenotype of Werner syndrome is characterized by extensive deletions. Proc. Natl. Acad. Sci. USA86, 5893-5897.MEDLINE Abstract
17 Thilly, W. G., DeLuca, J. G., Furth, E. E., Hoppe IV, H., Kaden, D. A., Krolewski, J. J., Liber, H. L., Skopek, T. R., Slapikoff, S. A., Tizard, R. J., and Penman, B. W. (1980) In de Serres, F. J. and Hollaender, A. (eds.), Chemical Mutagens 6, Plenum Press, New York, pp. 331-364.
18 Turker, M. S., Monnat, R. J.,Jr., Fukuchi, K.-i., Johnston, P. A., Ogburn, C. E., Weller, R. E., Park, J. F., and Martin, G. M. (1988) A novel class of unstable 6-thioguanine-resistant cells from dog and human kidneys. Cell. Biol. Toxicol. 4, 211-223.MEDLINE Abstract
19 Martin, G.M., Sprague,C.A., Norwood, T.H., and Pendergrass, W.R. (1974) Clonal selection, attenuation and differentiation in an in vitro model of hyperplasia. Am. J: Pathol.74, 137-154.MEDLINE Abstract
20 Cole, J. and Skopek, T. R. (1994) Somatic mutant frequency, mutation rates and mutational spectra in the human population in vivo. Mutat. Res. 304, 33-105.MEDLINE Abstract
21 Licht, M. R. and Novick, A. C. (1993) Nephron sparing surgery for renal cell carcinoma. Urol. 149, 1-7.
22 Morley, A. A., Cox, S., and Holliday, R. (1982) Human lymphocytes resistant to 6-thioguanine increase with age. Mech. Ageing Dev. 19, 21-26. MEDLINE Abstract
23 Trainor, K. J., Wigmore, D. J., Chrysostomou, A., Dempsey, J. L., Seshadri, R., and Morley, A. A. (1984) Mutation frequency in human lymphocytes increases with age. Mech. Ageing Dev. 27, 83-86.MEDLINE Abstract
24 Morley, A. A., Trainor, K. J., Seshadri, R., and Ryall, R. G. (1983) Measurement of in vivo mutations in human lymphocytes. Nature 302, 155-156.MEDLINE Abstract
25 Dempsey, J. L., Pfeiffer, M., and Morley, A. A. (1993) Effect of dietary restriction on in vivo somatic mutation in mice. Mutat. Res. 291, 141-145.MEDLINE Abstract
26 Holliday, R. (1995) Understanding Ageing, Cambridge University Press, Cambridge.
27 Horn, P. L., Turker, M. S., Ogburn, C. E., Disteche, C. M., and Martin, G. M. (1984) A cloning assay for 6-thioguanine resistance provides evidence against certain somatic mutational theories of aging. J. Cell. Physiol. 121, 309-315.MEDLINE Abstract
28 Dewey, M. J., Martin, Jr., Martin, G. R., arid Mintz, B. (1977) Mosaic mice with teratocarcinoma-derived mutant cells deficient in hypoxanthine phosphoribosyltransferase. Proc. Natl. Acad. Sci. USA 74, 5564-5568.MEDLINE Abstract
29 Ansell, J. D., Sarnuel, K., Whittingham, D. G., Patek, C. E., Hardy, K., Handyside, A. H., Jones, K. W., Muggleton-Harris, A. L., Taylor, A. H., and Hooper, M. L. (1991) Hypoxanthine phosphoribosyl transferase deficiency, haematopoiesis and fertility in the mouse. Development 112, 489-498.MEDLINE Abstract
30 Levi, M. and Rowe, J. W. (1992) In Seldin, D. W. and Giebisch, G. (eds.), The Kidney: Physiology and Pathophysiology, Raven Press, New York, pp. 3433-3456.
31 Zawada, Jr., Boice, J. L., and Santella, R. N. (1995) In Massry, S. G. and Glassock, R. J. (eds.), Massry and Glassock's Textbook of Nephrology, Williams&Wilkins, Baltimore,MD, pp. 1138-1147.
32 Stein, W. D. and Stein, A. D. (1995) Familial and sporadic human renal cell carcinoma: evidence against a double-loss mechanism of carcinogenesis. J. Clin. Epidemiol. 48, 767-777.MEDLINE Abstract
33 McCredie, M. (1994) Bladder and kidney cancers. Cancer Surv. 19, 343-368.
34 Lindeman, R. D. (1995) In Masoro, E. J. (ed.), Handbook of Physiology Section 11:Aging, Oxford University Press, New York, pp. 485-503.
35 Gown, A. M. and Vogel, A. M. (1982) Monoclonal antibodies to intermediate filament proteins of human cells: unique and cross-reacting antibodies. J. Cell Biol. 95, 414-421.MEDLINE Abstract
36 Tsukada, T., Tippens, D., Gordon, D., Ross, R., and Gown, A. M. (1987) HHF35: a muscle-actin-specific monoclonal antibody. I. Immunocytochemical and biochemical characterization. Am. J. Pathol. 126, 51-60.MEDLINE Abstract
37 Henikoff, S. (1983) Cloning exons for mapping of transcription: characterization of the Drosophila melanogaster alcohol dehydrogenase gene. Nucleic Acids Res.11, 4735-4752.MEDLINE Abstract
38 Monnat, R. J.,Jr., Chiaverotti, T. A., Hackmann, A. F. M., and Maresh, G. A. (1992) Molecular structure and genetic stability of human hypoxanthine phosphoribosyltransferase gene duplications. Genomics13, 788-796.MEDLINE Abstract
39 Edwards, A., Voss, H., Rice, P., Civitello, A., Stegemann, J., Schwager, C., Zimmermann, J., Erfle, H., Caskey, C. T., and Ansorge, W. (1990) Automated DNA sequencing of the human HPRT locus. Genomics 6, 593-608.MEDLINE Abstract
40 Box, G. E. P. and Cox, D. R. (1964) The analysis of transformations. J. Royal Stat. Soc. B 26, 211-252.
41 Venables, W. N. and Ripley, B. D. (1994) Modern Applied Statistics with S-Plus, Springer-Verlag, New York.
42 Jolly, D. J., Okayama, H., Berg, P., Esty, A. C., Filpula, D., Bohlen, P., Johnson, G. G., Shively, J., Hunkapillar, T., and Friedmann, T. (1983) Isolation and characterization of a full-length expressible cDNA for human hypoxanthine phosphoribosyltransferase. Proc. Natl. Acad. Sci. USA 80, 477-481.MEDLINE Abstract
43 Wilson, J.M., Tarr, G.E., Mahoney, W.C., and Kelly, W.N. (1982) Human hypoxanthine guanine phosphoribosyltransferase: complete amino acid sequence of the erythrocyte enzyme. J. Biol. Chem. 257, 10978-10985.MEDLINE Abstract
*To whom correspondence should be addressed
This page is maintained by OUP admin. Last updated Thu Oct 31 15:21:55 GMT 1996. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996
