ACI/Seg (ACI) rats develop prostate cancers spontaneously with aging, similar to humans. Here, to identify genes involved in prostate cancer susceptibility, we did linkage analysis and oligonucleotide microarray analysis. Linkage analysis was done using 118 effective rats, and prostate cancer susceptibility 1 (Pcs1), whose ACI allele dominantly induced prostate cancers, was mapped on chromosome 19 [logarithm of odds (LOD) score of 5.0]. PC resistance 1 (Pcr1), whose ACI allele dominantly and paradoxically suppressed the size of prostate cancers, was mapped on chromosome 2 (LOD score of 5.0). When linkage analysis was done in 51 rats with single or no macroscopic testicular tumors, which had larger prostates and higher testosterone levels than those with bilateral testicular tumors, Pcs2 and Pcr2 were mapped on chromosomes 20 and 1, respectively. By oligonucleotide microarray analysis with 8,800 probe sets and confirmation by quantitative reverse transcription-PCR, only two genes within these four loci were found to be differentially expressed >1.8-fold. Membrane metalloendopeptidase (Mme), known to inhibit androgen-independent growth of prostate cancers, on Pcr1 was expressed 2.0- to 5.5-fold higher in the ACI prostate, in accordance with its paradoxical effect. Cdkn1a on Pcs2 was expressed 1.5- to 4.5-fold lower in the ACI prostate. Additionally, genes responsible for testicular tumors and unilateral renal agenesis were mapped on chromosomes 11 and 14, respectively. These results showed that prostate cancer susceptibility of ACI rats involves at least four loci, and suggested Mme and Cdkn1a as candidates for Pcr1 and Pcs2.

Prostate cancer is one of the leading causes of cancer-related death in men in most developed countries, and its incidence is increasing in Japan (1). A variety of genetic and environmental factors are considered involved in the initiation and progression of human prostate cancers (2, 3), and clarification of these factors is urgently required. Genes that are likely to confer dominant susceptibility to prostate cancers have been mapped by linkage analysis of familial prostate cancers at 1q24-25 (4), 1q42-43 (5), 1p36 (6), Xq27-28 (7), 20q13 (8), 17p (9), and 8p22-23 (10). Three putative prostate cancer susceptibility genes, RNASEL/HPC1 at 1q25 (11), MSR1 at 8p22 (10), and ELAC2/HPC2 at 17p11 (9), have been recently identified. In addition, polymorphisms of the prostate-specific antigen, isoforms of cytochrome P450, androgen receptor(12), vitamin D receptor(13), and steroid 5α-reductase 2 (SRD5A2; ref. 14) have been proposed to be related to prostate cancer risks. In spite of the efforts and expertise in the field, identification of a responsible gene in each locus and clarification of the interaction among the genes are still facing difficulties due to the complex interactions among these genetic factors in individuals who have been exposed to different levels of various environmental factors.

Experimental animal models are useful because a population of animals can be exposed to the same homogeneous environmental factors and the number of genes involved is limited. As for prostate cancers, a number of rodent models are reported (15). Among these, ACI/Seg (ACI) rats are unique in that they spontaneously develop a high incidence of microscopic cancers of the ventral prostate along with aging (16–18). By 33 months of age, 95% to 100% of the rats develop intra-alveolar dysplasia in the ventral lobe, and 35% to 40% develop invasive carcinomas (17). The high incidence of microscopic cancers and the lower incidence of macroscopic cancers are considered to mimic the natural history of human prostate cancers, where the incidence of microscopic cancers is very high and that of clinical diseases is much lower (19, 20). The ACI model was used to show that a high-fat diet had a promotional effect on the early stage of prostate carcinogenesis (21), that 5α-reductase inhibitor suppresses prostate carcinogenesis (22), and that long-term feeding of a 1% cholesterol diet promotes prostate carcinogenesis, possibly by increasing tissue oxidative stress (23). However, no information is available for the genetic loci or genes responsible for the predisposition of ACI rats to prostate cancers.

In this study, we did linkage and oligonucleotide microarray analyses to identify candidate gene(s) involved in the genetic predisposition of ACI rats to prostate cancers. Linkage analysis was done in F2 intercross rats produced by crossing ACI rats with F344 rats, which were less susceptible to prostate cancers (24) but more susceptible to spontaneous testicular tumors (25). Oligonucleotide microarray analysis was done considering that if genes differentially expressed in the prostates of ACI and F344 rats are present in the mapped loci, they are good candidates for the responsible genes. Additionally, genes responsible for testicular tumors and unilateral renal agenesis (URA) were mapped.

Animals and carcinogenicity test. ACI/SegHgd (ACI) and F344/Jcl (F344) rats were purchased from Harlan (Indianapolis, IN) and CLEA Japan, Inc. (Tokyo, Japan), respectively. Male ACI and female F344 rats were mated to produce F1 progeny. F1 rats were intercrossed to produce F2 intercross rats. A total of 21 male ACI rats, 22 male F344 rats, 21 male F1 rats, and 219 male F2 intercross rats were fed an Oriental MF (Oriental Yeast Co., Tokyo, Japan) diet and housed three per plastic cage on woodchip bedding in an air-conditioned animal room at a temperature of 21°C to 25°C and a humidity of 40% to 60% with a 12-hour light 12-hour dark cycle. Moribund rats were sacrificed to clarify the cause of death, and the remaining rats were sacrificed at the age of 30 months. The animal experiments followed the Guidelines for the Care and Use of Laboratory Animals of Nagoya City University Medical School and were approved by the Institutional Animal Care and Use Committee.

Histologic and serum analysis. The male sex organs and bladder were resected en bloc and were examined for gross abnormalities and fixed in 10% buffered formalin. After fixation, the ventral prostate was removed from the base of the bladder neck and the entire organ weighed. One sagittal slice through the ventral prostate was embedded in paraffin, and three 4-μm-thick sections were stained with H&E. The areas of ventral lobe and lesions were quantitatively measured by an Image Processor for Analytical Pathology (IPAP-WIN, Sumika Technoservice, Takarazuka, Japan). The area of lacuna with secretions was included as the area of the prostate. Hyperplastic and tumorous lesions were classified according to their structural and nuclear atypism as described in Results. The area ratio was calculated as the percentage of the area of a lesion in the total area of the ventral lobe.

Testicular tumors were diagnosed by histologic examination of H&E-stained samples. Leukemia was diagnosed by hypertrophy of the spleen and the liver with infiltration of leukemic cells. Unilateral renal agenesis was diagnosed by the lack of one kidney. All histologic diagnoses were made by experienced pathologists (To.S and S.S.) and a urologist (Y.H.).

Serum testosterone levels are known to be at their peaks between 9:00 am and 1:00 pm (26), and blood samples were collected between 10:00 am and 12:00 noon. Concentrations were measured by RIA at SRL, Inc. (Tokyo, Japan).

Genotyping and linkage analysis. Genotyping was done using 201 markers (157 microsatellite markers and 44 AP-RDA markers; Supplementary Table S1) and genomic DNA extracted from the tails (27–30). Linkage maps were constructed by MAPMAKER/EXP (version 3.0b) software (31), and quantitative trait locus (QTL) analysis was done using MAPMAKER/QTL (version 1.1b) software. The presence of a lesion(s) and absence of any lesions were scored as “1” and “0,” respectively, and treated like a QTL. The number of lesions was square root transformed to obtain improved normality. The area and area ratio of lesions were treated as actual values. Association of a quantitative trait with an allele was evaluated by a logarithm of odds (LOD) score. Threshold LOD scores at 2.8 and 4.3 were considered as “suggestive” and “significant” linkages, respectively (32).

Oligonucleotide microarray analysis. Equal amounts of RNA were pooled from the whole prostates of three F344 and three ACI rats at two ages 8 and 48 weeks. Oligonucleotide microarray analysis was done using GeneChip Rat Genome U34A (Affymetrix, Santa Clara, CA) with 8,800 probe sets as in our previous studies (33–35). The signal intensities were normalized so that the average of all the genes on a GeneChip would be equal, and the data were processed using Affymetrix Microarray Suite version 5.0 at the default variable. The annotation was according to RG_U34A Annotations (June 2004; Affymetrix).

Quantitative reverse transcription-PCR. cDNA was synthesized from 2 μg of total RNA with oligo (dT)12-18 primer. Real-time PCR analysis was done separately for the three rats in each group using an iCycler iQ detection system (Bio-Rad Laboratories, Hercules, CA) with SYBR Green PCR Core Reagents (Applied Biosystems, Foster City, CA). Primers with the following sequences were used, Mme upper 5′- CCTACCGGCCAGAGTATGCAG -3′ and lower 5′- TATGAGTTCTTGCGGCAATGA -3′, Cdkn1a upper 5′- CCTTCTTCTGCTGTGGGTCA -3′ and lower 5′- AAGACACACTGAATGAAGGCTA -3′. To quantify the number of molecules of a specific gene in a sample, a standard curve was generated using templates that contained 101 to 106 copies of the gene and was normalized to that of β-actin.

Incidences and sizes of prostate lesions in inbred and crossed rats. Twelve of 21 ACI, 8 of 22 F344, 14 of 21 F1, and 118 of 219 F2 intercross rats survived to 30 months of age, premature deaths being caused by renal failure due to URA and leukemia. Hyperplastic and/or tumorous lesions in the ventral prostate were observed in 100%, 50%, 100%, and 89% of ACI, F344, F1 and F2 intercross rats, respectively (Table 1). The lesions in ACI and F344 rats had distinct morphologies (Fig. 1). Lesions in ACI rats showed a solid structure, and round, irregular (ACI type) nuclei whereas those in F344 had a papillary or cribriform structure, and angled nuclei with regular sizes. The lesions in ACI rats were much larger (1.8 versus 0.2 mm2, P = 0.001), often associated with inflammatory infiltrates, and occasionally invaded into the adjacent alveoli. In contrast, lesions in F344 rats never developed into macroscopic cancers. Based on the structural and nuclear atypisms and sizes of tumors, ACI (n = 12) and F344 (n = 8) rats could be distinguished with 95% accuracy in a blinded manner.

Table 1.

Results of the carcinogenicity test

ACI/Seg (n = 12)F344/Jcl (n = 8)(F344 × ACI)F1 (n = 14)(F344 × ACI)F2 (n = 118)
Body and organ weights     
    Body weight (g) 419.6 ± 65.8 (309.2−506.9) 350.7 ± 29.7 (309.3−391.6) 509.5 ± 67.5 (315.2−583.3) 418.9 ± 77.5 (195.9−652.9) 
    Prostate weight (g) 1.54 ± 0.83 (0.61−3.00) 0.92 ± 0.19 (0.67−1.26) 1.61 ± 0.75 (0.86−2.91) 1.64 ± 1.58 (0.38−11.75) 
    Ventral prostate weight (g) 0.32 ± 0.14 (0.13−0.58) 0.27 ± 0.07 (0.19−0.42) 0.36 ± 0.12 (0.22−0.59) 0.35 ± 0.32 (0.07−2.20) 
    Testes weight (g) 2.81 ± 0.50 (2.04−3.59) 6.62 ± 2.51 (2.48−9.84) 7.73 ± 9.10 (2.57−33.93) 4.20 ± 2.50 (0.67−21.16) 
    Liver weight (g) 11.3 ± 1.4 (8.5−12.7) 12.7 ± 3.4 (9.6−19.0) 15.0 ± 4.8 (1.5−20.3) 13.2 ± 2.9 (4.7−31.3) 
    Spleen weight (g) 0.80 ± 0.14 (0.57−1.06) 2.34 ± 1.74 (1.11−5.68) 1.10 ± 0.23 (0.82−1.63) 1.31 ± 0.79 (0.34−5.57) 
Any lesions in prostate     
    Presence (%) 100 50 100 89 
    Number 5.3 ± 2.1 1.5 ± 1.9 4.5 ± 1.9 3.8 ± 3.3 
    Area (mm21.78 ± 1.22 0.22 ± 0.30 1.03 ± 0.80 1.11 ± 1.50 
    Area ratio (%) 1.61 ± 1.29 0.25 ± 0.33 0.84 ± 0.59 1.31 ± 1.75 
Lesions with a solid structure     
    Presence (%) 92 25 93 59 
    Number 3.0 ± 2.6 0.4 ± 0.7 2.0 ± 1.3 1.3 ± 1.6 
    Area (mm21.17 ± 1.25 0.06 ± 0.14 0.40 ± 0.35 0.40 ± 1.07 
    Area ratio (%) 1.09 ± 1.31 0.07 ± 0.15 0.38 ± 0.37 0.48 ± 1.27 
Lesions with ACI-type nuclei     
    Presence (%) 100 38 100 81 
    Number 5.1 ± 2.4 1.3 ± 1.8 3.3 ± 1.4 3.1 ± 3.0 
    Area (mm21.73 ± 1.24 0.22 ± 0.30 0.89 ± 0.74 0.90 ± 1.41 
    Area ratio (%) 1.57 ± 1.31 0.24 ± 0.33 0.73 ± 0.56 1.08 ± 1.68 
ACI/Seg (n = 12)F344/Jcl (n = 8)(F344 × ACI)F1 (n = 14)(F344 × ACI)F2 (n = 118)
Body and organ weights     
    Body weight (g) 419.6 ± 65.8 (309.2−506.9) 350.7 ± 29.7 (309.3−391.6) 509.5 ± 67.5 (315.2−583.3) 418.9 ± 77.5 (195.9−652.9) 
    Prostate weight (g) 1.54 ± 0.83 (0.61−3.00) 0.92 ± 0.19 (0.67−1.26) 1.61 ± 0.75 (0.86−2.91) 1.64 ± 1.58 (0.38−11.75) 
    Ventral prostate weight (g) 0.32 ± 0.14 (0.13−0.58) 0.27 ± 0.07 (0.19−0.42) 0.36 ± 0.12 (0.22−0.59) 0.35 ± 0.32 (0.07−2.20) 
    Testes weight (g) 2.81 ± 0.50 (2.04−3.59) 6.62 ± 2.51 (2.48−9.84) 7.73 ± 9.10 (2.57−33.93) 4.20 ± 2.50 (0.67−21.16) 
    Liver weight (g) 11.3 ± 1.4 (8.5−12.7) 12.7 ± 3.4 (9.6−19.0) 15.0 ± 4.8 (1.5−20.3) 13.2 ± 2.9 (4.7−31.3) 
    Spleen weight (g) 0.80 ± 0.14 (0.57−1.06) 2.34 ± 1.74 (1.11−5.68) 1.10 ± 0.23 (0.82−1.63) 1.31 ± 0.79 (0.34−5.57) 
Any lesions in prostate     
    Presence (%) 100 50 100 89 
    Number 5.3 ± 2.1 1.5 ± 1.9 4.5 ± 1.9 3.8 ± 3.3 
    Area (mm21.78 ± 1.22 0.22 ± 0.30 1.03 ± 0.80 1.11 ± 1.50 
    Area ratio (%) 1.61 ± 1.29 0.25 ± 0.33 0.84 ± 0.59 1.31 ± 1.75 
Lesions with a solid structure     
    Presence (%) 92 25 93 59 
    Number 3.0 ± 2.6 0.4 ± 0.7 2.0 ± 1.3 1.3 ± 1.6 
    Area (mm21.17 ± 1.25 0.06 ± 0.14 0.40 ± 0.35 0.40 ± 1.07 
    Area ratio (%) 1.09 ± 1.31 0.07 ± 0.15 0.38 ± 0.37 0.48 ± 1.27 
Lesions with ACI-type nuclei     
    Presence (%) 100 38 100 81 
    Number 5.1 ± 2.4 1.3 ± 1.8 3.3 ± 1.4 3.1 ± 3.0 
    Area (mm21.73 ± 1.24 0.22 ± 0.30 0.89 ± 0.74 0.90 ± 1.41 
    Area ratio (%) 1.57 ± 1.31 0.24 ± 0.33 0.73 ± 0.56 1.08 ± 1.68 
Figure 1.

Distinct histological features of prostate cancers in ACI and F344 rats. Lesions of ACI rats (A and B) were characterized by high incidences of solid structures (A) and round, irregular nuclei (B). Lesions of F344 rats (C and D) were characterized by a papillary or cribriform pattern (C) and nuclei with angles and relatively regular sizes (D). Lesions in ACI rats were larger and invaded into flanking lobules showing their more malignant characteristics.

Figure 1.

Distinct histological features of prostate cancers in ACI and F344 rats. Lesions of ACI rats (A and B) were characterized by high incidences of solid structures (A) and round, irregular nuclei (B). Lesions of F344 rats (C and D) were characterized by a papillary or cribriform pattern (C) and nuclei with angles and relatively regular sizes (D). Lesions in ACI rats were larger and invaded into flanking lobules showing their more malignant characteristics.

Close modal

Therefore, all the lesions were classified according to their structural and nuclear atypisms (Table 1). A solid structure was present in 92%, 25%, 93%, and 59% of ACI, F344, F1 and F2 intercross rats, respectively. ACI-type nuclei were present in 100%, 38%, 100%, and 81%, respectively. In ACI and F2 intercross rats, 100% and 99%, respectively, of lesions with a solid structure had ACI-type nuclei.

Testicular tumors, unilateral renal agenesis, leukemia, and subcutaneous tumors. Testicular tumors, histologically diagnosed as Leydig cell tumors, were observed in 92%, 100%, 100%, and 92% of ACI, F344, F1 and F2 intercross rats, respectively. Bilateral macroscopic testicular tumors were observed in 25%, 88%, 100%, and 57%, respectively. URA was observed in 24%, 0%, 0%, and 11%, respectively, at 30 months. Leukemia was observed in 0%, 25%, 0%, and 9%, respectively, at 30 months. S.c. tumors were observed in 0%, 38%, 0%, and 10%, respectively.

Linkage mapping of a gene that affects the development and the size of prostate cancers. Linkage analysis with the presence of (i) any lesions, (ii) lesions with a solid structure, and (iii) lesions with ACI-type nuclei was done in the 118 F2 intercross rats, and one QTL was identified in the vicinity of D19Rat75 (LOD scores of 2.9, 3.0 and 5.0, respectively; Table 2). The same QTL was mapped using the numbers of these three kinds of lesions. When F2 intercross rats were classified by the genotypes at D19Rat75, lesions with ACI-type nuclei were observed in 85% (22 of 26), 92% (60 of 65), and 52% (14 of 27) of rats with the ACI/ACI, ACI/F344, and F344/F344 genotypes, respectively. This locus, prostate cancer susceptibility 1 (Pcs1), therefore conferred ACI-dominant prostate cancer susceptibility (Fig. 2A). This region (D19Rat75-D19Rat46) corresponded to rat 19q11-12.

Table 2.

Results of QTL analysis in F2 intercross rats

Lesions used for linkage analysisPcs1 (chromosome 19)
Pcr1 (chromosome 2)
LOD scores% Variance explainedLOD scores% Variance explained
Any lesions in prostate     
    Presence 2.9 10.6 0.4 1.7 
    Number 3.2 12.2 1.9 7.6 
    Area (mm20.4 1.5 3.7 16.2 
    Area ratio (%) 0.2 0.8 5.0 20.3 
Lesions with a solid structure     
    Presence 3.0 11.4 0.7 2.7 
    Number 2.8 10.7 1.1 4.2 
    Area (mm20.1 0.4 2.3 9.0 
    Area ratio (%) 0.2 0.7 2.9 11.4 
Lesions with ACI-type nuclei     
    Presence 5.0 17.6 0.5 2.0 
    Number 3.7 13.5 1.3 5.2 
    Area (mm20.4 1.7 2.7 11.4 
    Area ratio (%) 0.1 0.5 4.0 15.6 
Lesions used for linkage analysisPcs1 (chromosome 19)
Pcr1 (chromosome 2)
LOD scores% Variance explainedLOD scores% Variance explained
Any lesions in prostate     
    Presence 2.9 10.6 0.4 1.7 
    Number 3.2 12.2 1.9 7.6 
    Area (mm20.4 1.5 3.7 16.2 
    Area ratio (%) 0.2 0.8 5.0 20.3 
Lesions with a solid structure     
    Presence 3.0 11.4 0.7 2.7 
    Number 2.8 10.7 1.1 4.2 
    Area (mm20.1 0.4 2.3 9.0 
    Area ratio (%) 0.2 0.7 2.9 11.4 
Lesions with ACI-type nuclei     
    Presence 5.0 17.6 0.5 2.0 
    Number 3.7 13.5 1.3 5.2 
    Area (mm20.4 1.7 2.7 11.4 
    Area ratio (%) 0.1 0.5 4.0 15.6 
Figure 2.

LOD score obtained for linkage with prostate lesions and testicular tumors. A, LOD score obtained by the presence of lesions with ACI-type nuclei in the 118 F2 intercross rats. ACI allele in this locus dominantly promoted induction of the lesions. B, LOD score obtained by the area ratio of any lesions in the 118 F2 intercross rats. ACI allele in this locus dominantly and paradoxically suppressed the growth of the lesions. C, LOD score obtained by the presence of lesions with ACI-type nuclei in the F2 intercross rats with single or no testicular tumors. ACI allele in this locus dominantly promoted the growth of the lesions in this subgroup of rats. D, LOD score obtained by the area ratio of any lesions in the F2 intercross rats with single or no testicular tumors. ACI allele in this locus dominantly and paradoxically suppressed the growth of the lesions in this subgroup. E, LOD score obtained by development of testicular tumors. The F344 allele in this locus promoted development of testicular tumors. Dashed line, Lander-Kruglark threshold for “significance” (LOD score of 4.3 in intercross), and “suggestive” (LOD score of 2.8 in intercross) linkage (32).

Figure 2.

LOD score obtained for linkage with prostate lesions and testicular tumors. A, LOD score obtained by the presence of lesions with ACI-type nuclei in the 118 F2 intercross rats. ACI allele in this locus dominantly promoted induction of the lesions. B, LOD score obtained by the area ratio of any lesions in the 118 F2 intercross rats. ACI allele in this locus dominantly and paradoxically suppressed the growth of the lesions. C, LOD score obtained by the presence of lesions with ACI-type nuclei in the F2 intercross rats with single or no testicular tumors. ACI allele in this locus dominantly promoted the growth of the lesions in this subgroup of rats. D, LOD score obtained by the area ratio of any lesions in the F2 intercross rats with single or no testicular tumors. ACI allele in this locus dominantly and paradoxically suppressed the growth of the lesions in this subgroup. E, LOD score obtained by development of testicular tumors. The F344 allele in this locus promoted development of testicular tumors. Dashed line, Lander-Kruglark threshold for “significance” (LOD score of 4.3 in intercross), and “suggestive” (LOD score of 2.8 in intercross) linkage (32).

Close modal

Linkage analysis was done using areas containing the three kinds of lesions, and a QTL was identified in the vicinity of D2Rat161 (LOD scores of 3.6, 2.3, and 2.7, respectively; Table 2). The same QTL was mapped using the area ratio. Notably, this locus conferred ACI-dominant resistance to an increase in size, and the effect was clearer using areas with any lesion than with using areas with lesions characteristic to ACI rats. F2 intercross rats with the ACI/ACI, ACI/F344, and F344/F344 genotypes at D2Rat161 had area ratios of 0.8 ± 1.2% (mean ± SD), 0.9 ± 1.0%, and 2.5 ± 2.6%, respectively (Fig. 3). This locus, prostate cancer resistance 1 (Pcr1), had a paradoxical effect on the size of prostate cancers (Fig. 2B). This region (D2Rat26-D2Rat40) corresponded to rat 2q23-q32.

Figure 3.

Effect of Pcr1 locus on the area ratio of lesions. When F2 intercross rats were classified by their genotype at D2Rat161, rats with the ACI allele(s) showed smaller area ratios.

Figure 3.

Effect of Pcr1 locus on the area ratio of lesions. When F2 intercross rats were classified by their genotype at D2Rat161, rats with the ACI allele(s) showed smaller area ratios.

Close modal

Effect of bilateral testicular tumors on prostate cancers. The size of the ventral prostate was significantly smaller in F2 intercross rats with bilateral macroscopic testicular tumors (0.24 ± 0.13 g, n = 67) than in those with single or no tumors (0.49 ± 0.42 g, n = 51, P < 0.005). However, the incidences and mean areas of tumorous lesions in the ventral prostate were not different (60 of 67 and 1.0 ± 1.4 mm2 in the former rats, and 45 of 51 and 1.3 ± 1.6 mm2 in the latter rats). The mean serum testosterone level at 30 months was 0.1 ± 0.2 ng/mL in rats with bilateral tumors and 1.3 ± 5.6 ng/mL (0.3 ± 0.4 ng/mL when two extremely high values were excluded) in rats without bilateral tumors. Although the presence of bilateral testicular tumors did not affect the incidence or area of tumorous/hyperplastic lesions in the prostate, two new QTLs were mapped when linkage analysis was done in rats without bilateral testicular tumors (Table 3). One QTL in the vicinity of D20Rat3, Pcs2, conferred an ACI-dominant increase in the incidence of lesions (LOD scores of 1.9-3.3; Fig. 2C). Another QTL in the vicinity of D1Rat211, Pcr2, conferred a paradoxical effect, ACI-dominant suppression of the area ratio (LOD scores of 3.9-4.3; Fig. 2D). Pcs2 (D20Rat41-D20Rat5) and Pcr2 (D1Rat121-D15Rat45) corresponded to rat 20p12-11 and 1p12-q11, respectively.

Table 3.

QTLs for prostate lesions in subgroups classified by testicular tumors and testosterone levels

PopulationVariable for linkage analysisPcs1 (chromosome 19)
Pcr1 (chromosome 2)
Pcs2 (chromosome 20)
Pcr2 (chromosome 1)
LOD scores% Variance explainedLOD scores% Variance explainedLOD scores% Variance explainedLOD scores% Variance explained
All F2 intercross rats (n = 118) Presence of any lesions 2.9 10.6 0.4 1.7 0.4 1.9 0.9 3.6 
 Presence of lesions with ACI-type nucleus 5.0 17.6 0.5 2.0 1.9 8.6 0.9 3.4 
 Area ratio of any lesions 0.2 0.8 5.0 20.3 0.5 2.0 0.6 2.3 
 Area ratio of lesions with ACI-type nucleus 0.1 0.5 4.0 15.6 1.5 6.7 0.8 3.1 
Rats with no or unilateral testicular tumors (n = 51) Presence of any lesions 0.8 7.1 0.7 6.3 1.9 17.2 1.6 17.5 
 Presence of lesions with ACI-type nucleus 1.6 13.7 0.6 5.7 3.4 32.4 1.2 12.0 
 Area ratio of any lesions 1.3 11.1 2.0 22.9 1.1 10.0 3.9 33.0 
 Area ratio of lesions with ACI-type nucleus 0.8 6.8 1.3 11.1 1.2 10.6 4.3 35.4 
Rats with bilateral testicular tumors (n = 67) Presence of any lesions 2.7 17.2 0.6 6.2 1.6 14.0 0.5 3.2 
 Presence of lesions with ACI-type nucleus 3.9 24.5 1.3 11.0 1.2 10.5 1.0 6.5 
 Area ratio of any lesions 0.6 3.9 3.4 22.9 1.4 9.1 0.7 6.0 
 Area ratio of lesions with ACI-type nucleus 0.5 3.1 2.8 18.3 1.2 8.0 0.5 3.7 
PopulationVariable for linkage analysisPcs1 (chromosome 19)
Pcr1 (chromosome 2)
Pcs2 (chromosome 20)
Pcr2 (chromosome 1)
LOD scores% Variance explainedLOD scores% Variance explainedLOD scores% Variance explainedLOD scores% Variance explained
All F2 intercross rats (n = 118) Presence of any lesions 2.9 10.6 0.4 1.7 0.4 1.9 0.9 3.6 
 Presence of lesions with ACI-type nucleus 5.0 17.6 0.5 2.0 1.9 8.6 0.9 3.4 
 Area ratio of any lesions 0.2 0.8 5.0 20.3 0.5 2.0 0.6 2.3 
 Area ratio of lesions with ACI-type nucleus 0.1 0.5 4.0 15.6 1.5 6.7 0.8 3.1 
Rats with no or unilateral testicular tumors (n = 51) Presence of any lesions 0.8 7.1 0.7 6.3 1.9 17.2 1.6 17.5 
 Presence of lesions with ACI-type nucleus 1.6 13.7 0.6 5.7 3.4 32.4 1.2 12.0 
 Area ratio of any lesions 1.3 11.1 2.0 22.9 1.1 10.0 3.9 33.0 
 Area ratio of lesions with ACI-type nucleus 0.8 6.8 1.3 11.1 1.2 10.6 4.3 35.4 
Rats with bilateral testicular tumors (n = 67) Presence of any lesions 2.7 17.2 0.6 6.2 1.6 14.0 0.5 3.2 
 Presence of lesions with ACI-type nucleus 3.9 24.5 1.3 11.0 1.2 10.5 1.0 6.5 
 Area ratio of any lesions 0.6 3.9 3.4 22.9 1.4 9.1 0.7 6.0 
 Area ratio of lesions with ACI-type nucleus 0.5 3.1 2.8 18.3 1.2 8.0 0.5 3.7 

Oligonucleotide microarray analysis of genes within the loci. By oligonucleotide microarray analysis using the prostates of ACI and F344 rats at 8 weeks, 219 and 143 probes were found to be expressed higher and lower, respectively, at ≥1.8-fold in the ACI prostates. At 48 weeks, 371 and 154 probes were expressed higher and lower, respectively, in the ACI prostates. Forty-eight and 34 genes were expressed higher and lower both at 8 and 48 weeks (Supplementary Table S2). Among these genes, those located within the Pcs1, Pcr1, Pcs2, and Pcr2 regions were RT1 class II (locus DMa; 2.1- to 3.0-fold), membrane metalloendopeptidase (Mme; 1.9- to 3.7- fold), and Gstt2 (3.0- to 6.1-fold).

RT1 and Gstt2 in Pcs2 were found to have sequence polymorphisms in their oligonucleotide probes, which caused apparent expression difference in the microarray analysis (35). In contrast, Mme in Pcr1 was confirmed to be expressed at higher levels in the ACI prostate (2.0- to 5.5-fold) by quantitative reverse transcription-PCR (RT-PCR; Table 4). The difference and its location in the Pcr1 region suggested Mme as a candidate for Pcr1. In addition, we previously found that Cdkn1a (p21) on Pcs2 was differentially expressed in the prostates of various rat strains,4

4

Yamashita et al., submitted for publication.

and its high expression in F344 rats was confirmed by quantitative RT-PCR (1.5- to 4.5-fold, Table 4). Although the Cdkn1a difference could not be identified by the oligonucleotide microarray analysis, the difference and its location in the Pcs2 region suggested Cdkn1a as a candidate for Pcs2.

Table 4.

Quantitative RT-PCR analyses for genes within the QTLs

GeneLocationGene expression levels copy number/β-actin (×104, mean, n = 3)
8 wks
48 wks
ACIF344(F344 × ACI) F1ACIF344
Mme Pcr1 chromosome 2q31 1,413 255 976 2,608 1,328 
Cdkn1a Pcr2 chromosome 20p12 9.9 15.1 16.6 8.3 37.7 
GeneLocationGene expression levels copy number/β-actin (×104, mean, n = 3)
8 wks
48 wks
ACIF344(F344 × ACI) F1ACIF344
Mme Pcr1 chromosome 2q31 1,413 255 976 2,608 1,328 
Cdkn1a Pcr2 chromosome 20p12 9.9 15.1 16.6 8.3 37.7 

Mapping for the other QTLs. Linkage analysis with the presence of testicular tumors was done in the 118 F2 intercross rats, and one QTL was identified in the vicinity of D11Ncc20. An F344 allele of this locus dominantly promoted the development of testicular tumors (LOD score of 3.9; Fig. 2E). The incidences were 71% (17 of 24) in rats with the ACI/ACI genotype and 94% (91 of 94) in rats with the ACI/F344 and F344/F344 genotypes. The QTL, testicular tumor resistance 1 (Ttr1), explained 14% of the total variance.

One QTL, whose ACI allele semidominantly induced URA, was mapped in the vicinity of D14Rat65 (LOD score of 2.9). URA was observed in 20% (10 of 49), 9% (7 of 77), 0% (0 of 56) of rats with ACI/ACI, ACI/F344 and F344/F344 genotypes, respectively. The QTL, Ura1, explained 13% of the total variance. No QTLs were mapped for the serum testosterone levels, the development of leukemia and that of s.c. tumors at 30 months.

Prostate cancers in ACI rats occasionally progressed into invasive cancers, which were considered to model human clinically significant prostate cancers. In contrast, prostate cancers in F344 rats stayed as small lesions, if present, which was considered to model human indolent prostate cancers. By using these strains, two prostate cancer susceptibility and two resistance genes were mapped using F2 intercross progeny of ACI and F344 rats. Dominant ACI alleles of Pcs1 (chromosome 19) and Pcs2 (chromosome 20) increased the incidence and the number of prostate lesions, and these genes were considered to be involved in the development of prostate cancers. In contrast, dominant ACI alleles of Pcr1 (chromosome 2) and Pcr2 (chromosome 1) paradoxically decreased the area and area ratio of lesions, and these genes were considered to suppress the growth of prostate cancers. A similar example that an allele of a susceptible strain confers a resistance has been reported (27). Even when these four genes were combined, these did not fully explain the genetic predisposition of ACI rats. It was therefore shown that susceptibility of ACI rats to prostate cancers is determined by the four genes mapped here and many other weak genes.

The effect of Pcs2 and Pcr2 was clearly observed in rats without bilateral macroscopic testicular tumors and masked in rats with such tumors. Testicular tumors of rats are known to be associated with a tendency towards feminization (37). Although serum testosterone levels are known to be highly variable (26), 30-month-old rats with bilateral macroscopic testicular tumors had lower testosterone levels in this study. Therefore, as a possible mechanism for the differential effects of Pcs2 and Pcr2, involvement of different molecular pathways in the development of prostate cancers depending upon testosterone levels was considered. Because Pcs2 promotes and Pcr2 suppresses prostate cancers and many other unmapped genes are potentially involved, the incidence and area of prostate cancers were not different between rats with and without bilateral macroscopic testicular tumors. Temporal analysis on the development of prostate cancers, that of testicular tumors, and testosterone levels is necessary to clarify possible involvement.

By combining positional information obtained by linkage analysis and expression information obtained by oligonucleotide microarray analysis (38), two candidate genes were identified. GeneChip “Rat Genome U34A” covered 39 of 158 genes within the region of Pcs1, 77 of 308 in Pcr1, 86 of 191 in Pcs2, and 39 of 115 in Pcr2, accounting for 31% of the genes within the mapped loci. In addition, the responsible genes in the loci are not necessarily differentially expressed. Therefore, the combination strategy is effective only when candidate genes are obtained, and no good candidates were found for Pcs1 and Pcs2 in this study. Nevertheless, Mme (also called Neutral endopeptidase, Cd10), which showed higher expression in ACI rats, was considered as a good candidate for Pcr1 that had a paradoxical effect. Mme expression was known to suppress growth of androgen-independent prostate cancer cells and was decreased in androgen-independent prostate cancers (39). Therefore, its higher expression in ACI was in accordance with its paradoxical effect. Cdkn1a, which showed lower expression in ACI rats, was a good candidate for Pcs2. Cdkn1a is a well-established negative regulator of the cell cycle, and a structural abnormality in its 5′ of promoter region has been identified.4

Chromosomal regions around Pcs1, Pcr1, Pcs2, and Pcr2 corresponded to human 19p13, 4q28-31, and 16q21-22; 3q26-27, 13q12-14, 3q21-26, and 4q32; 6p21, 21q22, 22q11, and 10q21; and 6q23-25 and 5p15, respectively. As for Pcs1, linkage of Swedish familial prostate cancers to 19p13 (40) and frequent loss of heterozygosity in human prostate cancers (16q22; refs. 41, 42) are reported. Introduction of human chromosome 19p13 into both rat and human prostate cancer cells suppressed tumorigenicity in nude mice (43). As for Pcr1, its candidate, Mme, is located on human 3q25. Frequent deletion of human 13q in prostate cancer is reported (44). As for Pcs2, in addition to Cdkn1a on human 6p21, polymorphisms of Gstt1 on human 22q11 were reported to give differential risks to prostate cancers (3). As for Pcr2, Srd5a1, whose polymorphism is associated with the metastatic potential of prostate cancers (36), was located on Pcr2.

The genes involved in the development of testicular tumors (Ttr1) and URA (Ura1) were successfully mapped on chromosomes 11 and 14, respectively. These regions correspond to human chromosome 21q21-22 and human chromosome 4q11-12, respectively. Linkage analysis is almost impossible in humans for diseases with low penetrance, such as URA, and positional information obtained here would be useful if candidate genes are obtained from their functions.

In summary, we were able to map four susceptibility and resistance genes involved in the prostate cancer susceptibility of ACI rats. A complex interaction of multiple susceptibility genes and effects of testicular tumors in prostate cancer susceptibility was shown. Combining the data with those obtained by oligonucleotide microarray, Mme and Cdkn1a were identified as candidates.

Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org).

Grant support: 3rd-term Comprehensive Cancer Control Strategy; Ministry of Health, Labor and Welfare Cancer Research; Organization for Pharmaceutical Safety and Research of Japan program for promotion of fundamental studies in health sciences; and Foundation for Promotion of Cancer Research, Japan grant-in-aid.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The authors are grateful to Dr. Tadao Kakizoe (National Cancer Center) for his critical discussion.

1
Franceschi S, La Vecchia C. Cancer epidemiology in the elderly.
Crit Rev Oncol Hematol
2001
;
39
:
219
–26.
2
Deutsch E, Maggiorella L, Eschwege P, Bourhis J, Soria JC, Abdulkarim B. Environmental, genetic, and molecular features of prostate cancer.
Lancet Oncol
2004
;
5
:
303
–13.
3
Peters ME, Ostrander EA. Prostate cancer: simplicity to complexity.
Nat Genet
2001
;
27
:
134
–5.
4
Smith JR, Freije D, Carpten JD, et al. Major susceptibility locus for prostate cancer on chromosome 1 suggested by a genome-wide search.
Science
1996
;
274
:
1371
–4.
5
Berthon P, Valeri A, Cohen-Akenine A, et al. Predisposing gene for early-onset prostate cancer, localized on chromosome 1q42.2-43.
Am J Hum Genet
1998
;
62
:
1416
–24.
6
Gibbs M, Stanford JL, McIndoe RA, et al. Evidence for a rare prostate cancer-susceptibility locus at chromosome 1p36.
Am J Hum Genet
1999
;
64
:
776
–87.
7
Xu J, Meyers D, Freije D, et al. Evidence for a prostate cancer susceptibility locus on the X chromosome.
Nat Genet
1998
;
20
:
175
–9.
8
Berry R, Schroeder JJ, French AJ, et al. Evidence for a prostate cancer-susceptibility locus on chromosome 20.
Am J Hum Genet
2000
;
67
:
82
–91.
9
Tavtigian SV, Simard J, Teng DH, et al. A candidate prostate cancer susceptibility gene at chromosome 17p.
Nat Genet
2001
;
27
:
172
–80.
10
Xu J, Zheng SL, Komiya A, et al. Germline mutations and sequence variants of the macrophage scavenger receptor 1 gene are associated with prostate cancer risk.
Nat Genet
2002
;
32
:
321
–5.
11
Carpten J, Nupponen N, Isaacs S, et al. Germline mutations in the ribonuclease L gene in families showing linkage with HPC1.
Nat Genet
2002
;
30
:
181
–4.
12
Giovannucci E, Stampfer MJ, Krithivas K, et al. The CAG repeat within the androgen receptor gene and its relationship to prostate cancer.
Proc Natl Acad Sci U S A
1997
;
94
:
3320
–3.
13
Ingles SA, Ross RK, Yu MC, et al. Association of prostate cancer risk with genetic polymorphisms in vitamin D receptor and androgen receptor.
J Natl Cancer Inst
1997
;
89
:
166
–70.
14
Makridakis NM, Ross RK, Pike MC, et al. Association of mis-sense substitution in SRD5A2 gene with prostate cancer in African-American and Hispanic men in Los Angeles, USA.
Lancet
1999
;
354
:
975
–8.
15
Shirai T, Takahashi S, Cui L, et al. Experimental prostate carcinogenesis: rodent models.
Mutat Res
2000
;
462
:
219
–26.
16
Shain SA, McCullough B, Segaloff A. Spontaneous adenocarcinomas of the ventral prostate of aged AXC rats.
J Natl Cancer Inst
1975
;
55
:
177
–80.
17
Ward JM, Reznik G, Stinson SF, Lattuada CP, Longfellow DG, Cameron TP. Histogenesis and morphology of naturally occurring prostatic carcinoma in the ACI/segHapBR rat.
Lab Invest
1980
;
43
:
517
–22.
18
Isaacs JT. The aging ACI/Seg versus Copenhagen male rat as a model system for the study of prostatic carcinogenesis.
Cancer Res
1984
;
44
:
5785
–96.
19
Bosland MC, Dreef-Van Der Meulen HC, Sukumar S, et al. Multistage prostate carcinogenesis: the role of hormones.
Princess Takamatsu Symp
1991
;
22
:
109
–23.
20
Rinker-Schaeffer CW, Partin AW, Isaacs WB, Coffey DS, Isaacs JT. Molecular and cellular changes associated with the acquisition of metastatic ability by prostatic cancer cells.
Prostate
1994
;
25
:
249
–65.
21
Kondo Y, Homma Y, Aso Y, Kakizoe T. Promotional effect of two-generation exposure to a high-fat diet on prostate carcinogenesis in ACI/Seg rats.
Cancer Res
1994
;
54
:
6129
–32.
22
Homma Y, Kaneko M, Kondo Y, Kawabe K, Kakizoe T. Inhibition of rat prostate carcinogenesis by a 5α-reductase inhibitor, FK143.
J Natl Cancer Inst
1997
;
89
:
803
–7.
23
Homma Y, Kondo Y, Kaneko M, et al. Promotion of carcinogenesis and oxidative stress by dietary cholesterol in rat prostate.
Carcinogenesis
2004
;
25
:
1011
–4.
24
Reznik G, Hamlin MH II, Ward JM, Stinson SF. Prostatic hyperplasia and neoplasia in aging F344 rats.
Prostate
1981
;
2
:
261
–8.
25
Weisburger JH, Rivenson A, Reinhardt J, Braley J, Pittman B, Zang E. On the occurrence of Leydig cell tumors in the F344 rat.
Cancer Lett
2002
;
182
:
213
–6.
26
Heywood LH. Testosterone levels in the male laboratory rat: variation under experimental conditions.
Int J Androl
1980
;
3
:
519
–29.
27
Ushijima T, Yamamoto M, Suzui M, et al. Chromosomal mapping of genes controlling development, histological grade, depth of invasion, and size of rat stomach carcinomas.
Cancer Res
2000
;
60
:
1092
–6.
28
Yoshida Y, Ushijima T, Yamashita S, Imai K, Sugimura T, Nagao M. Development of the arbitrarily primed-representational difference analysis method and chromosomal mapping of isolated high throughput rat genetic markers.
Proc Natl Acad Sci U S A
1999
;
96
:
610
–5.
29
Yamashita S, Yoshida Y, Kurahashi A, Sugimura T, Ushijima T. Construction of a high-throughput rat genetic mapping system with 466 arbitrarily primed-representational difference analysis markers.
Mamm Genome
2000
;
11
:
982
–8.
30
Yamashita S, Furumoto K, Nobukiyo A, Kamohara M, Ushijima T, Furukawa T. Mapping of A gene responsible for cataract formation and its modifier in the UPL rat.
Invest Ophthalmol Vis Sci
2002
;
43
:
3153
-9.
31
Lander ES, Green P, Abrahamson J, et al. MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations.
Genomics
1987
;
1
:
174
–81.
32
Lander E, Kruglyak L. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results.
Nat Genet
1995
;
11
:
241
–7.
33
Kuramoto T, Morimura K, Yamashita S, et al. Etiology-specific gene expression profiles in rat mammary carcinomas.
Cancer Res
2002
;
62
:
3592
–7.
34
Abe M, Yamashita S, Kuramoto T, et al. Global expression analysis of N-methyl-N′-nitro-N-nitrosoguanidine-induced rat stomach carcinomas using oligonucleotide microarrays.
Carcinogenesis
2003
;
24
:
861
–7.
35
Yamashita S, Nomoto T, Ohta T, Ohki M, Sugimura T, Ushijima T. Differential expression of genes related to levels of mucosal cell proliferation among multiple rat strains by using oligonucleotide microarrays.
Mamm Genome
2003
;
14
:
845
–52.
36
Habib FK, Ross M, Bayne CW, Bollina P, Grigor K, Chapman K. The loss of 5α-reductase type I and type II mRNA expression in metastatic prostate cancer to bone and lymph node metastasis.
Clin Cancer Res
2003
;
9
:
1815
–9.
37
Konishi H, Okajima H, Okada Y, Yamamoto H, Fukai K, Watanabe H. High levels of cholesteryl esters, progesterone and estradiol in the testis of aging male Fischer 344 rats: feminizing Leydig cell tumors.
Chem Pharm Bull (Tokyo)
1991
;
39
:
501
–4.
38
Wayne ML, McIntyre LM. Combining mapping and arraying: an approach to candidate gene identification.
Proc Natl Acad Sci U S A
2002
;
99
:
14903
–6.
39
Papandreou CN, Usmani B, Geng Y, et al. Neutral endopeptidase 24.11 loss in metastatic human prostate cancer contributes to androgen-independent progression.
Nat Med
1998
;
4
:
50
–7.
40
Wiklund F, Gillanders EM, Albertus JA, et al. Genome-wide scan of Swedish families with hereditary prostate cancer: suggestive evidence of linkage at 5q11.2 and 19p13.3.
Prostate
2003
;
57
:
290
–7.
41
Suzuki H, Komiya A, Emi M, et al. Three distinct commonly deleted regions of chromosome arm 16q in human primary and metastatic prostate cancers.
Genes Chromosomes Cancer
1996
;
17
:
225
–33.
42
Osman I, Scher H, Dalbagni G, Reuter V, Zhang ZF, Cordon-Cardo C. Chromosome 16 in primary prostate cancer: a microsatellite analysis.
Int J Cancer
1997
;
71
:
580
–4.
43
Gao AC, Lou W, Ichikawa T, Denmeade SR, Barrett JC, Isaacs JT. Suppression of the tumorigenicity of prostatic cancer cells by gene(s) located on human chromosome 19p13.1-13.2.
Prostate
1999
;
38
:
46
–54.
44
van Dekken H, Paris PL, Albertson DG, et al. Evaluation of genetic patterns in different tumor areas of intermediate-grade prostatic adenocarcinomas by high-resolution genomic array analysis.
Genes Chromosomes Cancer
2004
;
39
:
249
–56.

Supplementary data