* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Multiple genetic loci modify risk for retinoblastoma in
Survey
Document related concepts
Polycomb Group Proteins and Cancer wikipedia , lookup
Designer baby wikipedia , lookup
Genomic imprinting wikipedia , lookup
Epigenetics of neurodegenerative diseases wikipedia , lookup
Genome (book) wikipedia , lookup
Oncogenomics wikipedia , lookup
Genetic engineering wikipedia , lookup
Microevolution wikipedia , lookup
Site-specific recombinase technology wikipedia , lookup
Nutriepigenomics wikipedia , lookup
Public health genomics wikipedia , lookup
Epigenetics in learning and memory wikipedia , lookup
Gene therapy of the human retina wikipedia , lookup
Transcript
Multiple Genetic Loci Modify Risk for Retinoblastoma in Transgenic Mice Anne E. Griep,1 Jeff Krawcek,2 Denis Lee,2 Amy Liem,2 DanielM. Albert? Rey Carabeo2 Norman Drinkwater,2 Maureen McCall,4'5 Carol Sattler,2 Jacques G. H. Lasudry,5 and Paul F. Lambert2 PURPOSE. Forty percent of cases of retinoblastoma, a childhood malignancy of the retina, are linked to the inheritance of a mutant allele of the retinoblastoma susceptibility gene Rbl. Tumor penetrance varies among earners in different family pedigrees, indicating that other genetic factors may modify risk for occurrence of retinoblastoma. This study was undertaken to determine whether multiple genetic loci modify the risk for retinoblastoma in mice. A line of aAcry-HPVl6E6/E7 transgenic mice expressing the human papillomavirus type 16 E6 and E7 oncogenes (HPV-16 E6 and ET) ectopically in the retina was characterized. E6 and E7 proteins bind to and inactivate the cellular tumor suppressor proteins p53 and Rb, respectively. METHODS. Retinoblastomas developed rarely when the aAcry-HPVl6E6/E7 transgene was maintained on the FVB background, but tumors arose with high frequency on C57BL/6 X FVB and C3H X FVB F, hybrid backgrounds. The incidence of retinoblastoma in the LHfi-TAG transgenic mice, which express simian virus 40 large tumor antigen (SV40 T-ag), was also influenced by the FVB and C57BL/6 backgrounds. Resistance of the aAcry-HPVl 6E6/E7 FVB mice to retinoblastoma mapped in part to the retinal degeneration (rd) locus. However, multiple genetic experiments indicate that resistance to retinoblastoma depends on additional loci in FVB mice. RESULTS. Multiple cellular genes can modify risk for retinoblastoma in mice. (Invest Ophthalmol Vis Sci. 1998;39:2723-2732) CONCLUSIONS. R From the 'Department of Anatomy, the 2McArdle Laboratory for Cancer Research and the Department of Oncology, the 'Department of Ophthalmology and Visual Sciences, and the "Waisman Center, University of Wisconsin Medical School, Madison. 5 Present address: Departments of Psychology and Ophthalmology and Visual Sciences, University of Louisville, 301 East Mohammed Ali Boulevard, Louisville, KY 40202. Supported in part by Grants EY09091, EY01917, EY01003, CA22443, and CA07175 from the National Institutes of Health, Bethesda, Maryland; and Grants JFRA-393 and RPG-96-043-03-MGO from the American Cancer Society, Atlanta, Georgia. Submitted for publication February 26, 1998; revised June 12, 1998; accepted July 6, 1998. Proprietary interest category: N. Reprint requests: Anne E. Griep, Department of Anatomy, University of Wisconsin Medical School, 1300 University Avenue, Madison, WI 53706. generally estimated to be 80% to 95%, but in some pedigrees it is as low as 20%.6'7 The absence of complete penetrance may relate to differences in the severity of the defect in mutant Rbl alleles.8 Alternatively, it may indicate that other cellular genes can affect the risk in a person who carries a mutant Rbl allele for development of retinoblastoma. Through the study of inbred strains of mice, genetic loci that affect the sensitivity or resistance of an animal (that is, those that modify risk) to specific cancers have been identified. For example, the mouse mom locus modifies risk to intestinal tumors in mice that inherit a defect in the mouse homologue to the human adenomatous polyposis coli gene,9 and multiple mouse loci modify risk to liver tumors (for review, see Ref. 10). In this study, we provide experimental evidence that multiple genes modify risk to retinoblastoma in the mouse. Genes in addition to Rbl are likely to contribute to the development of retinoblastoma in mice. Mice carrying a germline mutation in one Rbl allele are not predisposed to retinoblastoma '' as are humans; nor are mouse chimeras that harbor Rbl~'~ cells in their retinas predisposed to retinoblastoma. 1213 Yet transgenic mice expressing simian virus 40 (SV40) large-tumor antigen (T-ag) gene in the retina show development of retinoblastoma.14"16 SV40 T-ag is a potent oncoprotein that binds to and inactivates the Rb protein.17 In addition to these direct effects on Rb, SV40 T-ag possesses other activities that contribute to its transforming potential in tissue culture18 and tumorigenic phenotype in vivo.l9'20 These activities include its abilities to bind and inactivate Rb-like proteins plO7 and pl30, 21 ' 22 to modulate function of the p300 protein,23 and to inactivate the tumor-suppressor protein Investigative Ophthalmology & Visual Science, December 1998, Vol. 39, No. 13 Copyright © Association for Research in Vision and Ophthalmology 2723 etinoblastoma is a malignancy of the retina that occurs in young children and is associated with the loss of function of the retinoblastoma susceptibility gene Rbl.1'3 Children who inherit one mutant allele of Rbl are predisposed to bilateral, multifocal retinoblastoma and secondary tumors in numerous nonocular tissues. In these tumors, loss of function of the remaining wild-type allele is found at the Rbl locus. Nonhereditary, or sporadic, retinoblastoma is associated with somatic mutations in both alleles of Rbl and is usually focal. As such, Rbl is representative of tumor-suppressor genes, the disruption of which is associated with cancer development.4 Loss of function of both alleles of Rbl is the cornerstone of the two-hit hypothesis of Knudsen5 for retinoblastoma development in humans. However, the penetrance of retinoblastoma among human pedigrees is incomplete; it is Downloaded From: http://iovs.arvojournals.org/ on 08/11/2017 2724 Griep et al. p53 2 4 Of these activities, the inactivation of plO7 by T-ag may contribute to the induction of retinoblastoma. Mice that are plO7~/~;Rbl+/~ show retinal dysplasia, a potential precursor to retinoblastoma.25 In additional, inactivation of p53 may contribute to retinoblastoma formation in mice, because expression in the retina of the E7 oncogene from human papillomavirus (HPV) type 16 causes retinoblastoma only on a p53~/~ background.26 Human papillomavirus E6 and E7 oncoproteins belong to the same class of viral oncoproteins as does SV40 T-ag. E6 and E7 are expressed in most HPV-associated cervical cancers27"29 and together are necessary and sufficient to transform cells in tissue culture.30 Similar to T-ag, E6 inactivates p53, although through a different mechanism, by targeting p53 for degradation by a ubiquitin-mediated proteolytic pathway.31'32 Similar to T-ag, E7 binds to and sequesters Rb and Rb-like proteins33'34 leading to the deregulation of the E2F transcription factor family. We have shown that expression of HPV-16 E6 and E7 oncoproteins in the ocular lens could induce proliferation and inhibit differentiation in the lens in three independent lines of transgenic mice.35 These lines of transgenic mice, referred to as aAcry-HPVl6E6/E7 mice, were generated and maintained on the FVB inbred mouse genetic background, which is homozygous for a recessive mutation at the retinal degeneration (ret) locus. In the line of aAcry-HPVl6E6/E7 mice expressing the highest levels of E6/E7 message, line 19, lens tumors developed in approximately 40% of mice by 1 year of age. Transgene expression in this line of mice was found in multiple ectopic sites including the skin and extralenticular portions of the eye. Expression of E6 and E7 in the skin correlated with the high incidence of skin tumors.36 However, expression in other tissues such as the extralenticular portions of the eye did not correlate with tumor formation. In this study we report that, when crossed with other mouse genetic backgrounds such as C57BL/6 and C3H, the line 19 aAcry-HPVl6E6/E7 transgene confers high susceptibility to retinoblastoma formation. Whereas C57BL/6 is wild type at the ret locus, C3H is rd/rd. The retinoblastoma tumors expressed the transgene, appeared to arise from the inner nuclear layer of the retina, and sometimes developed into exophytic tumors that invaded the brain through the optic nerve and metastasized to the cervical lymph nodes. Crosses of line 19 aAcryHPVl 6E6/E7 transgenic mice with transgenic mice of the C57BL/6 genetic background in which photoreceptors have been ablated (rdta [rhodopsin promoter driven diphtheria toxin] transgenic mice37) also showed development of retinoblastoma. These results suggest that the FVB genetic background in which the transgene originally resides confers protection against retinoblastoma formation associated with expression of E6 and E7. Backcross experiments have shown that the FVB background can also confer resistance to retinoblastoma formation associated with expression of SV40 T-ag. Taken together, these studies indicate that the protective effect of the FVB genetic background is associated not only with the rd/rd allele but also with additional loci. METHODS Mouse Strains The generation, screening, and characterization of all transgenic mouse lines have been described. Line 19 aAcry- Downloaded From: http://iovs.arvojournals.org/ on 08/11/2017 IOVS, December 1998, Vol. 39, No. 13 HPVl 6E6/E7 transgenic mice were generated and maintained on the FVB genetic background.35 In these transgenic mice, expression of the HPV-16 E6 and E7 genes is directed to the lens using the aA crystallin promoter.35'38 The LH&-TAG transgenic mice express SV40 T-ag ectopically in the retina from the luteinizing hormone promoter.15 These transgenic mice were originally generated on the C57BL/6 X BALB/c hybrid background and subsequently maintained on the C57BL/6 inbred background. The rdta transgenic mice express an attenuated form of the diphtheria toxin A gene in rod photoreceptors by virtue of the cell type specificity of the rhodopsin promoter that drives its expression.37 These transgenic mice were generated originally on the FVB background, and subsequently, the transgene was moved onto the C57BL/6 background through backcrossing F, hybrids. All stock C3H/HeJ and C57BL/6J mice were purchased from Jackson Laboratory (Bar Harbor, ME), and FVB mice were purchased from Taconic Farms (Tarrytown, NY). All experiments with animals were conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Genotyping The aAcry-HPVl6E6/E7 transgenic mice were screened for the presence of transgene by Southern blot or polymerase chain reaction (PCR) analysis of DNA prepared from tail biopsy specimens, as described previously.35 The LH(3-TAG transgenic mice were screened for the presence of the transgene using PCR analysis (Jolene J. Windle, personal communication). The rdta transgenic mice were screened for the presence of the transgene by PCR analysis as described.37 The rd status of F, hybrid mice backcrossed with the FVB background was performed by PCR analysis. The murine-endogenous retrovirus Xmv is integrated into the first intron of the rd allele of the /3-7 subunit of the phosphodiesterase gene, ((5-PDE^9). This retroviral integration event Xmv-28 is not present in the wild-type allele of fi-PDE. The following oligonucleotides were generated to amplify the retrovirus-disrupted intron in the rd allele or the intact intron of the wild-type allele of (i-PDE in ([FVB line 19H X C57BL/6] X FVB) B, mice: rdl, 5'-ACCTGCATGTGAACCCAGTATTC-3'; rd2, 5'-GGGGAACCTGAAACTGAGGTGGG-3'; and rd3, 5'-CTCCTTTCTATTGCCCTGATCCACACC-3'. Oligo rdl is complementary to (i-PDE sequences in intron 1 lying upstream of the Xmv-28 integration site, and rd2 is complementary to sequences within Xmv. The combination of oligos rdl and rd2 selectively amplifies a region in the rd allele, across the junction of die first intron and the integrated retroviral element, generating a 470-bp fragment. Oligo rd3 is complementary tofi-PDEsequences in intron 1 downstream of rdl in the region disrupted by Xmv-28 in the rd allele. The combination of rdl and rd3 selectively amplifies the intact first intron of the wild-type allele resulting in a 370-bp fragment. The PCR products were resolved on a 2% agarose gel. Animals that were rd/rd at the locus would give rise only to the larger of the two fragments, whereas animals that were rd/+ would contain both fragment sizes in their products. The rdta and rd status was verified phenotypically, when possible, by examining the retina microscopically for the presence or absence of an outer nuclear layer. IOVS, December 1998, Vol. 39, No. 13 Mapping aAcry-HPVl6E6/E7 Transgene Integration Site Genomic DNA samples were isolated by phenol-chloroform extraction from skin tissues of 49 (FVB[line 19] X C57BL/6) X C57BL/6 backcross animals. Each animal was genotyped for simple sequence length polymorphism (SSLP) markers by PCR. Briefly, 100 ng genomic DNA was used as a template in the PCR reaction containing 0.13 mM forward and reverse SSLP primers (Research Genetics, Birmingham, AL), 50 mM deoxyribonucleoside triphosphates, and Tag polymerase in IX PCR buffer (10 mM Tris-HCl, 1.5 mM MgC12 and 50 mM KC1 [pH 8.3])- Amplification was performed in a GeneAmp PCR System 9600 (Perkin Elmer, Norwalk, CT) in an initial denaturing cycle of 2 minutes at 94°C (1 cycle), 30 seconds at 94°C, 30 seconds at 55°C, and 1 minute at 72°C (50 cycles), 7 minutes at 72°C (1 cycle), and a 4°C soak. PCR prodvicts were run on a 7% polyacrylamide gel and visualized under UV illumination. Linkage analysis for the transgene was performed using the MAPMAKER program (Whitehead Institute Genome Center, Cambridge, MA). Light Microscopy Eyes were enucleated from mice, fixed in 10% buffered formalin overnight, dehydrated through a graded ethanol series, and embedded in paraffin using an automatic tissue processor (Tissue-Tek, Miles Scientific, Naperville, IL). Sections (5 fxm) were cut, mounted on glass slides, stained with hematoxylin and eosin, protected with a coverslip, and examined. The intact skull and brain were fixed overnight in 10% buffered formalin and then decalcified in 5% nitric acid. Brain tissue was processed as described for ocular tissue and also examined by light microscope. Electron Microscopy Eyes were fixed for 8 hours in Karnovsky's paraformaldehydeglutaraldehyde fixative40 prepared in cacodylate buffer immediately before use. Eyes were washed in 0.1 M cacodylate buffer (pH 7.4) on ice for 1.5 hours and then washed in cold ddH2O. After dehydration in a graded series of ethanol, the eyes were infiltrated using a rotator with propylene oxide and Eponate (Ted Pella, Redding, CA), beginning with a 2:1 ratio and gradually increasing the proportion of Eponate over several hours until the eyes were in 100% Eponate. After overnight infiltration with Eponate, each eye was embedded in freshly made Eponate in individual flat-bottomed plastic vials. The sections were cut on a microtome equipped with a diamond knife (Ultracut E3; Reichert, Buffalo, NY), stained with uranyl acetate and lead citrate, and examined by electron microscope (H-7000; Hitachi, San Jose, CA) at 75 kV. In Situ Hybridization Eyes for in situ hybridization analysis were enucleated from line 19 aAcry-HPVl6E6/E7 F, hybrid transgenic mice and fixed in 4% paraformaldehyde overnight at 4°C, transferred to phosphate-buffered saline (PBS), dehydrated through a graded ethanol series, embedded in paraffin, and sectioned at 5-ju-m thickness. For in situ hybridization, 5-ju,m paraffin sections were deparaffinized in xylene, rehydrated through a graded series of ethanol, and subjected to in situ hybridization, as described previously,36'43 with the riboprobes described later. Antisense and sense cRNA probes were synthesized in vitro Downloaded From: http://iovs.arvojournals.org/ on 08/11/2017 Risk Factors for Retinoblastoma in Mice 2725 using a-35S-uridine triphosphate and T3, T7, or SP6 polymerase and the following templates: For generation of the ah crystallin-specific probes, the clone pGEMaCrl containing a 900-bp fragment of the mouse aA crystallin gene was used (obtained from Kathleen Mahon, Baylor University, Houston, TX). For generating the HPV-16 E6/E7-specific probes, the plasmids pAB1.5 and pGEM/E6 containing the E7 and E6 open reading frames of HPV-16, respectively, were used.36 Sections were hybridized with either sense or antisense 35S-labeled riboprobes, washed to remove nonspecifically bound probe, dipped in autoradiographic emulsion (NTB-2; Eastman Kodak, Rochester, NY), and exposed in the dark at 4°C for 2 weeks. After developing, dehydrating, and mounting, sections were examined by dark-field microscopy. Three to four sections from three tumors from three mice were examined. RESULTS Effect Of C57BL/6 Genetic Background on Tumor Incidence in aAcry-HPVl6E6/E7 Mice We have reported the generation and phenotypic characterization associated with expression of the HPV-16 E6 and E7 oncogenes in the mouse lens.35 In that study, three independent lines of aAcry-HPVl6E6/E7 mice, lines 4, 18, and 19, were created on the FVB inbred genetic background. In these mice, expression of the viral oncogenes was directed to the lens by the murine aA crystallin promoter. Approximately 40% of line 19 transgenic mice, which had the highest level of transgene expression, showed development of lens tumors by 13 months of age. The transgenic mice on the FVB inbred background carry a recessive mutation at the rd locus, which in the homozygous condition leads to a complete loss of rod photoreceptor cells during the first month of postnatal life.42 Because it was possible that the rc//rc/-induced defects in the retinas of FVB/FVB mice could influence the fate of E6/E7expressing lens cells, homozygous aAcry-HPVl6E6/E7 mice from line 1935 were crossed with the C57BL/6 inbred strain of mice, which is homozygous for the wild-type rd allele, and the F, transgenic progeny were characterized. No significant difference in the lens phenotypes was observed in perinatal and young adult aAcry-HPVl 6E6/E7 FVB X C57BL/6 F, mice compared with that of their inbred FVB transgenic parents (data not shown). However, the eye tumor incidence in line 19 aAcry-HPVl 6E6/E7 mice increased substantially in these F, animals (Fig. 1). Less than 10% of line 19 mice on the inbred FVB background had tumors by 10 months. In comparison, by 10 months of age 90% of the F, line 19 aAcry-HPVl6E6/E7 mice showed tumor development. These tumors began to arise in mice as early as 3 months of age. The high incidence of eye tumor formation seen in the F, mice was retained when the line 19 aAcry-HPVl6E6/E7 transgene was backcrossed onto the C57BL/6 background to generation N10 (data not shown). A significant proportion of the line 19 aAcry-HPVl 6E6/E7 F, mice had tumors develop in both eyes. Thus, genetic background influences the incidence of tumor formation in aAcryHPVl6E6/E7 mice. Development of Retinoblastomas Rather than Lens Tumors in Line 19 aAcry-HPVl6E6/E7 Fx Transgenic Mice To determine the cellular origins of the eye tumors in the line 19 aAcry-HPVl6E6/E7 FVB X C57BL/6 F, mice, microscopic 2726 Griep et al. IOVS, December 1998, Vol. 39, No. 13 100- line 19 x C57B1/6 F, 90" SO 3 H (n=58) 80" 7060 5040" 3020- 4 6 8 10 12 14 16 18 20 22 24 Mouse Age at Tumor Detection (months) FIGURE 1. Eye tumor incidence as a function of age in aAcryHPV16E6/E7 transgenic mice on FVB ( • ) inbred versus with C57BL/6 X FVB F, hybrid genetic backgrounds ( • ) . Line 19 aAcry-HPVl6E6/E7 FVB transgenic mice were crossed with C57BL/6 nontransgenic mice. Transgenic mice maintained on the FVB inbred background and Ft progeny were monitored for the presence of overt eye tumors at the indicated ages and killed when tumors were observed. Mice were genotyped by polymerase chain reaction analysis (see the Materials and Methods section). Indicated are the percentages of mice showing eye tumors at various ages up to 13 months. The number of mice in each group is indicated in parentheses. analyses were performed. Examples of the histology seen in intermediate and advanced line 19 aAcry-HPVl6E6/E7 Fj eye tumors are shown in Figures 2A and 2B, respectively. These histologic features are representative of what was observed in most of the eye tumors in the line 19 F, aAcry-HPVl6E6/E7 mice. The tumor in Figure 2A seemed to arise in the retina, rather than the lens. A small cataractous lens located within the anterior portion of the eye remained intact, that is, the lens capsule was not broken. The lens is similarly intact in the advanced tumor in Figure 2B. These features suggest that the retina rather than the lens was the source of the tumors. Homer-Wright rosettes, a hallmark of retinoblastoma, were evident in the tumor masses by light and electron microscopic analyses (Figs. 2A, 2B, 3Q. Ultrastructural analyses showed that these tumors also possessed basal bodies and trilaminar nuclear membrane (Figs. 3A, 3B, respectively), both characteristics of retinoblastoma in humans. Further histologic analyses indicated that a minority of the retinoblastomas were exophytic, invading the optic nerve (Fig. 2C). Of these, a significant number established metastases within the brain and cervical lymph nodes (Fig. 2C). These metastases also displayed histopathologic characteristics of retinoblastoma including Homer-Wright rosettes (Fig. 2D). The tumors that developed in line 19 aAcry-HPVl6E6/E7 F, transgenic mice resemble those observed previously in a line of SV40 TAG transgenic mice, LHfi-TAG, that expressed SV40 T-ag ectopically in the retina15 and in IRBP-TAG transgenic mice expressing SV40 T-ag specifically in photoreceptors.16 In each of these trans- Downloaded From: http://iovs.arvojournals.org/ on 08/11/2017 genie mouse models the retinal tumor bears significant resemblance to human retinoblastoma. To assess further the cellular origin of the tumors in line 19 aAcry-HPVl 6E6/E7 F, transgenic mice, eyes with earlystage tumors were examined microscopically. In multiple examined eyes, dysplastic and hyperplastic regions were evident in the retina, whereas the cataractous lens remained intact (Figs. 4A, 4B, 4C). Early lesions arose within the inner nuclear layer of the retina (Fig. 4B), similar to the apparent origin of tumors in the LHfi-TAG transgenic mice15'43 Crystallin-specific immunohistochemical and in situ hybridization analyses indicated that these lesions failed to express lens-specific crystallins (data not shown). This further suggests that the tumors arising in line 19 aAcry-HPVl6E6/E7 F, mice are not lenticular in origin. Characterization of more than 150 tumors arising in line 19 aAcry-HPVl6E6/E7 F, mice indicate that more than 90% of the eye tumors were retinal in origin. Thesefindingsled us to examine additional eye tumors from line 19 aAcryHPV16E6/E7 FVB mice. From more than 100 tumors examined, less than 1% of the tumors arising in the inbred FVB background were retinoblastomas. Thus, the genetic makeup of the mouse influences not only the frequency of tumor development but also the type of eye tumor in the line 19 aAcry-HPVl6E6/E7 mice. Heightened Expression of the E6 and E7 Genes in the Retinoblastomas in Line 19 otAcry-HPVl 6E6/ El Fx Mice Previously, we reported that mRNA specific for the HPV-16 E6 and E7 genes was detected in extralenticular portions of the eye of line 19 aAcry-HPVl6E6/E7 mice35 using the sensitive method of reverse transcription (RT)-PCR analysis. We performed in situ hybridization to assess whether viral transgenes were expressed in the retina and in the retinal tumors. Highlevel expression of the viral transgenes was detected not only in the lens but also within the retinal tumor mass of this and other aAcry-HPVl6E6/E7 F, mouse eyes (Fig. 4D). Lower-level expression of the transgene was noted within the microscopic, dysplastic lesions arising within the inner nuclear layer of the retina. Expression in the unaffected regions of the retina was at or below the limit of detection by in situ hybridization, as was the case in the retina of aAcry-HPVl6E6/E7 FVB mice (data not shown). This finding suggests that the basal level of transgene expression was not influenced by the mouse's genetic background. Thus, the level of expression of the viral transgenes seems to be augmented specifically in the retinal tumors of line 19 aAcry-HPVl 6E6/E7 F, mice. Similar increased levels of E6/E7 expression have been noted in squamous cell carcinomas arising in line 19 transgenic mice.36 Chromosomal Site of Integration of the aAcryHPVl6E6/E7 Transgene in Line 19 Mice Retinoblastoma in humans is associated with mutations at the Rbl locus. We therefore determined whether the high incidence of retinoblastoma in line 19 Ft aAcry-HPVl 6E6/E7 mice was a consequence of the transgene insertion into and disruption of the Rbl locus in the FVB genome. An insertion of this kind would lead to an Rbl+/~ genotype in the FVB X C57BL/6 F, animals. The chromosomal location of the transgene was mapped in the (FVB X C57BL/6) X C57BL/6 backcross (B,) progeny using SSLP analysis.44 Briefly, linkage of 17 SSLP mark- IOVS, December 1998, Vol. 39, No. 13 Risk Factors for Retinoblastoma in Mice 2727 \ FIGURE 2. Eye tumor histology in aAcry-HPVl6E6/E7 line 19 C57BL/6 X FVB F, hybrid mice. (A, B) Hematoxylin and eosin-stained 5-/m.m paraffin sections from representative intermediate (A) and advanced (B) eye tumors in the line 19 F, hybrid mice. Note the small cataractoiis lens characteristic of line 19 mice and large retinoblastoma with characteristic rosettes. (C) Low magnification of head and neck from a line 19 F, hybrid mouse shows the large retinoblastoma in the eye (E), the invasion of the tumor into the brain (arrow), and the metastasis to the cervical lymph node (LN)- (D) High magnification of a brain tumor from another line 19 FT hybrid mouse shows the characteristic rosette appearance {arrow). L, lens. ers to the transgene was analyzed using the Fisher's exact f-test. A marker on chromosome 12, D12MU34, yielded the best linkage significance, with P = 4.1 X 10~13 (Table 1). SSLP markers that flank the tumor suppressor genes, Rbl (Dl4Mit7, P = 0.07) and Trp53, (D11MU19, P = 1.0; D11MU14, P= 1.0) did not show linkage to the transgene. Based on our genotype data from the backcross, the transgene was positioned 2.1 centimorgans (cM) proximal to D12MU34, using the MAPMAKER program (data not shown). The location of line 19 transgene on chromosome 12 discounts the possibility diat its position causes the disruption of the Rbl or Trp53 loci, which are located on chromosomes 14 and 11, respectively. Thus, a direct effect of the transgene's insertion does not account for the high incidence of retinoblastoma in line 19 F, aAcryHPVJ6E6/E7 mice. Resistance to Retinoblastoma in an SV40 T Antigen Transgenic Mouse line on the FVB Background Alternatively, the high incidence of retinoblastoma in line 19 F, aAcry-HPVl 6E6/E7 transgenic mice may reflect an influence of genetic background on transgene expression, a consequence of the location of the transgene integration in the mouse genome. To investigate this possibility, studies were performed on another line of transgenic mice in which retinoblastomas develop15. This particular strain of mice carries an LHji-TAG transgene whose integration site is on chromosome 4. In these mice SV40 large-tumor antigen (SV40 TAG), which Downloaded From: http://iovs.arvojournals.org/ on 08/11/2017 is under the transcriptional control of the luteinizing hormone (LH£) promoter, also is expressed ectopically in the retina. These transgenic mice were originally generated on a C57BI7 6 X BALB/c hybrid background15 and subsequently maintained on the C57BL/6 background in our laboratories. To examine the influence of genetic background on SV40 T-ag-induced retinoblastoma, we crossed the C57BL76 mice harboring the LHfi-TAG transgene with FVB mice, and the F, hybrids were then backcrossed for two generations onto the FVB background. Tumor incidence in the LH(5-TAG transgenic B2 generation decreased to approximately 75% of that seen on the inbred C57BL/6 background (Fig. 5). That the FVB genetic background seems to confer resistance to retinoblastoma induced by a second viral oncoprotein, SV40 T-ag, indicates that the genetic influence does not result from specific effects on the HPV transgene in line 19 aAcry-HPVl6E6/E7 mice or from the site of transgene integration. Involvement of the rd Locus in Retinoblastoma Formation The C57BL/6 and FVB inbred strains are highly polymorphic. Therefore, the difference in susceptibility to retinoblastoma between these mouse strains may be influenced by many genes. We suspected diat one allele in particular may influence retinoblastoma incidence-that is, the rd allele that is present on the FVB genetic background but not on the C57BL/6 genetic background. The rd allele present in FVB 42 is a recessive mutant allele of the gene that encodes the /3 subunit of cGMP 2728 Grlep et al. IOVS, December 1998, Vol. 39, No. 13 FIGURE 3. Ultrastructural analysis of retinoblastoma in line 19 aAcry-HPVl6E6/E7 C57BI/6 X FVB F, hybrid mice. Electron micrographs show characteristic ultrastmctural features of retinoblastoma in an eye tumor from a line 19 F, hybrid mouse. (A) Basal body located in center of field; (B) trilaminar nuclear membrane on nucleus in upper left; and (C) HomerWright rosette. Magnification: (A) X 19,500; (B) X31,200; (C) X3,200. phosphodiesterase gene 0-PDE?9*). In rd/rd mice (such as inbred FVB), the rod photoreceptors in the outer nuclear layer of the retina degenerate during the first month of life, and subsequently, the cone photoreceptors degenerate as well. We reasoned that the absence of the photoreceptor cells in the retina may directly or indirectly influence the incidence of retinoblastomas. The degeneration of the photoreceptor cells Downloaded From: http://iovs.arvojournals.org/ on 08/11/2017 observed in rd/rd mice can be experimentally recapitulated by directing expression of a diphtheria toxin transgene to the rod photoreceptor cells using the rhodopsin promoter.37 We therefore generated FVB X C57BL/6 F, animals that were cotransgenic for line 19 aAcry-HPVl6E6/E7 and rdta transgenes. In these mice, the percentage of eyes in which retinoblastomas developed was less than half that seen in F, litter- Risk Factors for Retinoblastoma in Mice IOVS, December 1998, Vol. 39, No. 13 2729 FIGURE 4. Microscopic and in situ hybridization analyses of early retinoblastomas in line 19 aAcry-HPVJ 6E6/E7 C57BL/6 X FVB F, hybrid mice. (A, B, C) Hematoxylin arid eosin-stained paraffin section from a representative early retinoblastoma in a line 19 aAcry-HPVJ6E6/E7 F, hybrid mouse. (A, C) Low and high magnification shows tumor in peripheral retina and (B) high magnification of central retina. Note the clysplastic foci (arrows) in the inner nuclear layer in (B). (D) In situ hybridization analysis for E6/E7 transgene expression. A sequential slide was hybridized to an 35S-labeled antisense E6 and E7 probe. The peripheral retina, corresponding to (C), is shown. Hybridization was also observed in the central retina corresponding to the region shown in (B). No hybridization was observed in the retina when sense E6/E7 or antisense ctA crystallin probes were used (data not shown). INL, inner nuclear layer; ONL, outer nuclear layer; L, lens; R, retina. mates transgenic only for the line 19 aAcry-HPVJ6E6/E7 transgene (Fig. 6). This result indicates that the degeneration of photoreceptor cells in line 19 aAcry-HPVl6E6/E7 FVB mice that resulted from the rcl/rd genotype probably contributed in part to the low incidence of retinoblastomas. tomas by 10 months). Thus, there must be recessive alleles in the FVB mice that, in addition to the rd allele, confer resistance to retinoblastoma in the FVB background. To test this hypothesis, we scored the incidence of retinoblastomas in F, hybrid mice generated by crossing the ctAcry-HPVl6E6/E7 mice (inbred FVB background) to the C3H inbred mouse strain (Fig. 7). Involvement of FVB Alleles in Addition to rd in C3H and FVB inbred strains are also highly polymorphic, but Resistance to Retinoblastoma in Line 19 otAcryboth strains carry the rd/rd allele. The F, animals, all of which HPV16E6/E7 Transgenic Mice retain the rd/rd genotype, displayed a moderately high inciRetinoblastomas developed in line 19 aAcry-HPVJ'6E6/E7, dence of retinoblastoma (30%). This result supports the premise that FVB mice possess recessive alleles in addition to rdta cotransgenic FVB X C57BL/6 F, animals (Fig. 6) at a rd that confer resistance to retinoblastoma. frequency (35% had at least one retinoblastoma by 7.5 months or age) far above that seen in FVB inbred mice carrying only the line 19 aAcry-HPVl6E6/E7 transgene (<1% had retinoblasDISCUSSION 1. Significance of Linkage to the Line 19 aAcryHPV16E6/E7 Transgene of Different Mouse Markers on Chromosome 12 TABLE Markers Position (cM) Linkage Significance (P) 17.0 19.0 29.0 38.0 2.3 X 10~6 1.3 X 10~ri 4.1 X 10" 13 1.4 x 10" 9 DJ2MU46 DJ2MU2 DJ2MU34 DJ2MU5 cM, centimorgan. Downloaded From: http://iovs.arvojournals.org/ on 08/11/2017 In this study we report that retinoblastomas arose in mice transgenic for the HPV type 16 E6 and E7 oncogenes and that the animal's genetic background strongly influenced the penetrance of this tumorigenic phenotype. Our data suggest that recessive alleles at multiple loci in the FVB inbred mouse contributed to the resistance of these mice to development of retinoblastomas and that these loci also influenced the induction of retinoblastoma in mice transgenic for SV40 T-ag. Role of Viral Oncogenes in Retinoblastoma The induction of retinoblastoma in the aAcry-HPVJ6E6/E7 mice specifically occurred in the line that ectopically ex- 2730 Griep et al. IOVS, December 1998, Vol. 39, No. 13 100" LHp-TAG C57B1/6 Mice (n=30) 90" inactivation of the Trp53 gene.26 E7 induces apoptosis through p53-dependent and p53-independent pathways. Of interest, E6 can inhibit both.41 80- o LHP-TAG B 2 Mice 70- (n=18) 60OeS 5 0 40302010" 1 2 I 3 I 4 I 5 I 6 I 7 I 8 Mouse Age at Retinoblastoma Detection(months) FIGURE 5. Retinoblastoma incidence in LHfi-TAG C57BL/6 inbred (A) or FVB B2 transgenic (A) mice. LH&-TAG transgenic mice maintained on the C57BL/6 background were crossed to FVB mice, and the F, hybrids then were backcrossed to FVB mice for two generations. Mice were genotyped by polymerase chain reaction analysis (see the Materials and Methods section). Transgenic mice on the C57BL/6 and B2 backgrounds were monitored for the presence of overt eye tumors at the indicated ages and killed when tumors were observed. Tumors were confirmed to be retinoblastomas by microscopic analysis. Indicated are the percentages of mice in each group showing retinoblastomas at ages up to 5 months for LHfi-TAG C57BL/6 mice and 7 months for LH(5-TAG B2 mice. The number of mice in each group is indicated in parentheses. Differential Sensitivity of Site-Specific Tumor Induction in Inbred Strains of Mice By crossing line 19 aAcry-HPVl6E6/E7 FVB transgenic mice with C57BL/6, we observed that the ocular phenotype in this transgenic line can be modified by genetic background, indicating FVB genotype confers a resistance to tumor formation in the retina. However, FVB mice are not highly resistant to tumor formation in all tissues. For example, FVB mice are highly sensitive to squamous cell carcinoma associated with expression of HPV-16 E6 and E7 together,36 HPV-16 E7 alone,51, or resulting from treatment with chemical carcinogens.52 Conversely, C57BL/6 mice are relatively resistant to skin tumor induction.53 Such differential susceptibility of tissues in a given inbred mouse strain to tumor induction is commonly observed.53 This property of inbred mouse strains makes them a O 3 o a & pressed its transgenes in the retina (Fig. 2), implying that expression of E6 and E7 in the retinal tissue is necessary for tumor development. In HPV-positive cervical cancers, the Rbl and p53 genes have been found primarily to be intact, whereas they are mutant in the HPV-negative cervical cancers.45'46 This finding supports the premise that the functional inactivation of p53 and Rb proteins by E6 and E7, respectively, substitutes for the selective advantage conferred by the mutational inactivation of these tumor suppressors. Mutational inactivation of Rbl is a hallmark of human retinoblastoma. To assess whether inactivation of Rbl and Trp53 had occurred in the retinoblastomas that arose in the FVB X C57BL/6 F, aAcry-HPVl 6E6/E7 mice, we performed SSLP analysis on the tumors. The Rbl and Trp53 loci, located on chromosome 14 or 11, were found to be intact (data not shown). Thus, as is true in HPV-positive cervical cancers, the presence of the dominantly acting viral transgenes obviates the need for mutational inactivation of the genes encoding Rb or p53- That SV40 T-ag and E6/E7 each can induce retinoblastoma in mice yet retinoblastoma does not develop in mice carrying germline mutations in Rbl, Trp53, or both47"50 indicates that additional activities shared by T-ag and E6/E7 must be involved. E7 can modulate the activities of the Rb-like proteins, plO7 and or pl30 34 ; inactivation of plO7, in addition to Rb, may be required for tumorigenesis in the retina, because retinal dysplasia is noted in Rb+/~;pl07~/~ mice.25 However, expression of E7 alone in the retina is insufficient to induce development of retinoblastoma; rather, E7 efficiently induces cell proliferation but also apoptosis of the photoreceptor cells. This apoptosis may be partially relieved by mutational Downloaded From: http://iovs.arvojournals.org/ on 08/11/2017 10 12 Mouse Age at Retinoblastoma Detection (months) FIGURE 6. Retinoblastoma incidence in line 19 aAcryHPV16E6/E7, rdta C57BL/6 X FVB F, cotransgenic mice. Homozygous line 19 aAcry-HPVl 6E6/E7 FVB transgenic mice were crossed with heterozygous rdta C57BL/6 transgenic mice, and F, hybrid progeny were monitored for the presence of overt eye tumors at the indicated ages. Mice were genotyped for the presence of the aAcry-HPVl6E6/E7 and rdta transgenes by polymerase chain reaction analysis (see the Materials and Methods section). ( • ) Mice with rdta transgene; (•) mice without rdta transgene. Tumors were confirmed to be retinoblastomas by microscopic analysis. Indicated are the percentages of mice in each group showing retinoblastomas at ages up to 12 months. The number of eyes in each group is indicated in parentheses. Risk Factors for Retinoblastoma in Mice 2731 IOVS, December 1998, Vol. 39, No. 13 present in the inner nuclear layer to a tumor cell displaying characteristics of photoreceptor cells. An alternative explanation is that retinoblastoma arises from an undifferentiated precursor cell, that is, a retinoblast, that is retained within the inner nuclear layer. We know from cell fate experiments that many cell types arise from a common progenitor cell. Therefore, differentiation of a retinoblast residing in the inner nuclear layer could result in a tumor displaying characteristics of a photoreceptor cell. 90- o 80" bias mas 100-1 70- o 60" 5040" line 19 x C3H Fj (n=51) 3020- 10" line 19 (n=50) Mouse Age at Retinoblastoma Detection (months) Role of Alleles in Addition to rd in Modifying Risk to Retinoblastoma We observed a high incidence of retinoblastoma in line 19 aAcry-HPVl6E6/E7 FVB X C3H F, mice that are rd/rd. Thus, recessive alleles on the FVB genetic background other than rd must contribute to resistance to retinoblastoma. We have begun backcross experiments to determine whether homozygosity at other loci, in conjunction with homozygosity at rd, correlates with resistance to retinoblastoma in the aActyHPV16E6/E7 (FVB X C57BL/6) X FVB B, mice. Preliminary results indicate the existence of at least two alleles; homozygosity at either one decreases risk of development of retinoblastoma in rd/rd mice (data not shown). Confirmation of these preliminary results and more refined mapping of these loci are under way. FIGURE 7. Retinoblastoma incidence in line 19 aAcryHPV16E6/E7 C3H X FVB F, mice. Homozygous line 19 aAcryHPV16E6/E7 FVB transgenic mice were crossed with nontransgenic C3H mice, and F, hybrid progeny were monitored for the presence of overt eye tumors at the indicated ages. Mice Acknowledgments were genotyped by polymerase chain reaction analysis (see the Materials and Methods section). Mice were killed when tumors were observed. Tumors in the F, mice were confirmed to be The authors thank Angie Buehl for expert histotechnology, Henry Pi tot for preliminary histopathology, Kathleen Mahon for the pGEMcvCrl retinoblastomas by microscopic analysis. Indicated are the perplasmid, Nancy Robinson for editorial suggestions on the manuscript, centages of mice showing retinoblastomas at various ages up to and Jolene Windle and David Papermaster for sharing unpublished 14 months for F, mice and 13 months for FVB inbred mice. The results. number of mice in each group is indicated in parentheses. References useful for identifying strain-specific modifiers of tumor induction in a given tissue and may provide insight into the variable penetrance of tumor development in humans. Role of the Retinal Degeneration Phenotype in Modifying Risk to Retinoblastoma We think that the phenotype resulting from the rd allele in FVB mice contributes in part to their resistance to retinoblastoma. This is based primarily on our observations that ablation of the photoreceptor cell layer by the directed expression of a diphtheria toxin transgene caused a twofold decrease in the incidence of retinoblastoma in line 19 cxAcry-HPVl6E6/E7 FVB X C57BL/6 F, transgenic mice. How the physical ablation of the photoreceptor cell layer in rd/rd mutant and rdta transgenic mice contributed to the reduced incidence of retinoblastoma remains unclear. The retinoblastomas that arose in the HPV-16 transgenic mice clearly possessed histopathologic and ultrastructural characteristics of photoreceptor cells. The early dysplastic lesions that arose in the retinas of these mice and are the presumed precursors of the retinoblastomas were located within the inner nuclear layer of the retina, where bipolar, amacrine, and horizontal cells reside (Fig. 4). Similarly, the retinoblastomas that develop in the LHfi-TAG mice arise within the inner nuclear layer.15 However, photoreceptor cells normally reside in the outer nuclear layer. One possible explanation is that retinoblastoma in these animals results from metaplasia, that is, the conversion of a differentiated cell type Downloaded From: http://iovs.arvojournals.org/ on 08/11/2017 1. Friend SH, Bernards R, Rogelj S, et al. A human DNA segment with properties of the gene that predisposes to retinoblastoma and osteosarcoma. Nature. 1986;323:643-646. 2. Fung Y-KT, Murphree AL, T'Ang A, et al. Structural evidence for the authenticity of the human retinoblastoma gene. Science. 1987; 236:1657-1661. 3. Lee W-H, Bookstein R, Hong F, et al. Human retinoblastoma susceptibility gene: cloning, identification and sequence. Science. 1987;235:1394-1399. 4. Cavenee WK, Hansen MF, Nordenskjold M, et al. Genetic origins of mutations predisposing to retinoblastoma. Science. 1985:228:501503. 5. Knudson AG Jr. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA. 1971;68:820-823. 6. Matsunaga E. Hereditary retinoblastoma: delayed mutation of host resistance? Am JHum Genet. 1978;30:406-424. 7. Bonaiti-Pellie C, Clerget-Darpoux F, Babron M-C. Hereditary retinoblastoma: can balanced insertion entirely explain the differences of expressivity among families? Hum Genet. 1990;86:203208. 8. Otterson GA, Chen W, Coxon AB, Khleif SN, Kaye FJ. Incomplete penetrance of familial retinoblastoma linked to germ-line mutations that result in partial loss of RB function. Proc Natl Acad Sci USA. 1997;94:12036-12040. 9. Dietrich WF, Lander ES, Smith JS, et al. Genetic identification of Mom-1, a major modifier locus affecting Min-induced intestinal neoplasia in the mouse. Cell. 1993;75:631-639. 1.0. Poole TM, Chiaverotti TA, Carabeo RA, Drinkwater NR. Genetic analysis of multistage hepatocarcinogenesis. In: Walker C, Groopman J, Slaga T, Klein-Szanto A, eds. Genetics and Cancer Susceptibility: Implications for Risk Assessment. New York: Wiley-Liss; 1996:33-45. 2732 Griep et al. 11. Williams BO, Remington L, Albert DM, et al. Cooperative tumorigenic effects of germline mutations in Rb and p53. Nat Genet. 1994;7:480-484. 12. Maandag EC, van der Valk M, Vlaar M, et al. Developmental rescue of an embryonic-lethal mutation in the retinoblastoma gene in chimeric mice. EMBOJ. 1994; 13:4260-4268. 13- Williams BO, Schmitt EM, Remington L, et al. Extensive contribution of Rb-deficient cells to adult chimeric mice with limited histopathological consequences. EMBOJ. 1994;13:4251-4259. 14. Hammang JP, Baetge EE, Behringer RR, et al. Immortalized retinal neurons derived from SV40 T-antigen-induced tumors in transgenic mice. Neuron. 1990;4:775-782. 15- WindleJJ, Albert DM, O'Brien JM, et al. Retinoblastoma in transgenic mice. Nature. 199O;343:665-669. 16. Howes KA, LasudryJGH, Albert DM, WindleJJ. Photoreceptor cell tumors in transgenic mice. Invest Ophtbalmol Vis Sci. 1994;35: 342-351. 17. DeCaprio JA, Ludlow JW, Figge J, et al. SV40 large tumor antigen forms a specific complex with the product of the retinoblastoma susceptibility gene. Cell. 1988;54:275-283. 18. Fanning E. Modulation of cellular growth control by SV40 large T antigen. In: Doerfler W, Bohrn P, eds. Malignant Transformation by DNA Viruses. Weinheim, Germany: VCH; 1992:1-19. 19. Chen J, Tobin G, Pipas JM, Van Dyke TA. T-antigen mutant activities in transgenic mice: roles of p53 and pRb binding in tumorigenesis of the choroid plexus. Oncogene. 1992;7:1167-1175. 20. Symonds HS, McCarthy SA, Chen J, Pipas JM, Van Dyke T. Use of transgenic mice reveals cell specific transformation by a simian virus T-antigen amino terminal mutant. Mol Cell Biol. 1993; 13: 3255-3265. 21. Zalvide J, DeCaprio JA. Role of pRb-related proteins in simian virus large T-antigen-mediated transformation. Mol Cell Biol. 1995;15: 5800-5810. 22. Stubdal H, Zalvide J, DeCaprio JA. Simian vims 40 large T antigen alters the phosphorylation state of Rb-related proteins pi30 and p107./ Virol. 1996;70:2781-2788. 23. Yaciuk P, Carter MC, Pipas JM, Moran E. Simian virus large Tantigen expresses a biological activity complementary to the p300associated transforming function of the adenovirus El A gene products. Mol Cell Biol. 1991;ll:21l6-2124. 24. Perry ME, Levine AJ. Tumor suppressor p53 and the cell cycle. CurrOpin Genet Dev. 1993;3:5O-54. 25. Lee M-H, Williams BO, Mulligan G, et al. Targeted disruption of pi 07: functional overlap between plO7 and Rb. Genes Dev. 1996; 10:1621-1632. 26. Howes KA, Ransom N, Papermaster DS, et al. Apoptosis or retinoblastoma: alternative fates of photoreceptors expressing the HPV-16 E7 gene in the presence or absence of p53. Genes Dev. 1994;8:1300-!310. 27. Schwarz E, Freese UK, Gissman L, et al. Structure and transcription of human papillomavirus sequences in cervical carcinoma cells. Nature. 1985;3l4:ll 1-114. 28. Yee C, Krishnan HI, Baker CC, Schlegel R, Howley PM. Presence and expression of human papillomavirus sequences in human cervical carcinomas cell lines. Am J Pathol. 1985;! 19:361-366. 29- Smotkin D, Wettstein FO. Transcription of human papillomavirus type 16 early genes in a cervical cancer and a cancer-derived cell line and identification of the E7 protein. Proc Natl Acad Sci USA. 1986;83:4680-4684. 30. Munger K, Phelps WC, Bubb V, Howley PM, Schlegel R. The E6 and E7 genes of the human papillomavirus type 16 together are necessary and sufficient for transformation of primary human keratinocytes./ Virol. 1989;63:44l7-4421. 31. Huibregtse JM, Scheffner M, Howley PM. A cellular protein mediates association of p53 with the E6 oncoprotein of human papillomavirus types 16 and 18. EMBOJ. 1991;10:4l29-4l35. 32. Huibregste JM, Scheffner M, Howley PM. Cloning and expression of the cDNA for E6-AP, a protein that mediates the interaction of the human papillomavirus E6 oncoprotein with p53- Mol Cell Biol. 1993;! 3:775-784. Downloaded From: http://iovs.arvojournals.org/ on 08/11/2017 IOVS, December 1998, Vol. 39, No. 13 33. Dyson N, Howley P, Munger K, Harlow E. The human papilloma virus-16 E7 oncoprotein is able to bind to the retinoblastoma gene product. Science. 1989;243:934-937. 34. Dyson N, Guida P, Munger K, Harlow E. Homologous sequences in adenovirus E1A and human papillomavirus E7 proteins mediate interaction with the same set of cellular proteins./ Virol. 1992; 66:6893-6902. 35. Griep AE, Herber R, Jeon S, et al. Tumorigenicity by human papillomavirus type 16 E6 and E7 in transgenic mice correlates with alterations in epithelial cell growth and differentiation. / Virol. 1993;67:1373-1384. 36. Lambert PF, Pan H, Pitot HC, et al. Epidermal cancer associated with expression of human papillomavirus type 16 E6 and E7 oncogenes in the skin of transgenic mice. Proc Natl Acad Sci USA. 1993:90:5583-5587. 37. McCall MA, Gregg RG, Merriman K, et al. Morphological and physiological consequences of the selective elimination of rod photoreceptors in transgenic mice. Exp Eye Res. 1996;63:35-49. 38. Overbeek PA, Chepelinsky AB, Khillan JS, Piatigorsky J, Westphal H. Lens-specific expression and developmental regulation of the bacterial chloramphenicol acetyltransferase gene driven by the murine alpha A-crystallin promoter in transgenic mice. Proc Natl Acad Sci USA. 1985;82:7815-7819. 39- Bowes C, Li T, Danciger M, et al. Retinal degeneration in the rd mouse is caused by a defect in the |3-subunit of rod cGMP-phosphodiesterase. Nature. 1990;347:677-680. 40. Hayat MA. Principles and Techniques of Electron Microscopy: Biological Applications. New York: Van Nostrand Reinhold; 1970. 41. Pan H, Griep AE. Temporally distinct patterns of p53-dependent and p53-independent apoptosis during mouse lens development. Genes Dev. 1995;9:2157-2l69. 42. Taketo M, Schroeder AC, Mobraaten LE, et al. FVB/N: an inbred mouse strain preferable for transgenic analyses. Proc Natl Acad Sci USA. 1991;88:2065-2069. 43. Kivela T, Virtanen I, Marcus DM, et al. Neuronal and glial properties of a murine transgenic retinoblastoma model. Am J Pathol. 1991;138:1135-1148. 44. Dietrich WF, Miller JC, Steen RG, et al. A genetic map of the mouse with 4006 simple sequence length polymorphisms. Nat Genet. 1994;7:220-245. 45. Scheffner M, Munger K, Byrne JC, Howley PM. The state of the p53 and retinoblastoma genes in human cervical carcinoma cell lines. Proc Natl Acad Sci USA. 1991;88:5523-5527. 46. Crook T, Wrede D, Tidy J, et al. Status of c-myc, p53 and retinoblastoma genes in human papillomavirus positive and negative squamous cell carcinomas of the anus. Oncogene. 1991 ;6:1251 1257. 47. Clarke AR, Maandag ER, van RM, et al. Requirement for a functional Rb-1 gene in murine development. Nature. 1992;359:328330. 48. Donehower LA, Harvey M, Slagle BL, et al. Mice deficient for p53 are developmentally normal but susceptible to spontaneous tumours. Nature. 1992;356:215-221. 49. Jacks T, Fazeli A, Schmitt EM, et al. Effects of an Rb mutation in the mouse. Nature. 1992;359:295-3OO. 50. Lee EY, Chang CY, Hu N, et al. Mice deficient for Rb are nonviable and show defects in neurogenesis and haematopoiesis. Nature. 1992;359:288-294. 51. Herber R, Liem A, Pitot H, Lambert PF. Squamous epithelial hyperplasia and carcinoma in mice transgenic for the human papillomavirus type 16 E7 oncogene./ Virol. 1996;70:1873-1881. 52. Hennings H, Click AB, Lowry DT, et al. FVB/n mice: an inbred strain sensitive to the chemical induction of squamous cell carcinomas in the skin. Carcinogenesis. 1993;l4:2353-2358. 53. Drinkwater NR, Bennett LM. Genetic control of carcinogenesis. In: Ito N, Sugano H, eds. Modification of Tumor Development in Rodents. Basel, Switzerland: S. Karger; 1991 ;33:1-20.