Cancer Research Meeting Calendar Advances in Breast Cancer Research
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Cancer Research 67, 11226, December 1, 2007. doi: 10.1158/0008-5472.CAN-07-2487
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wang, H.
Right arrow Articles by Moser, A. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wang, H.
Right arrow Articles by Moser, A. R.

Molecular Biology, Pathobiology, and Genetics

Identification of Novel Modifier Loci of ApcMin Affecting Mammary Tumor Development

Hua Wang1,3, Douglas Teske1, Alyssa Tess1, Rebecca Kohlhepp1, YounJeong Choi2, Christina Kendziorski2 and Amy Rapaich Moser1

Departments of 1 Human Oncology and 2 Biostatistics and Medical Informatics, School of Medicine and Public Health and 3 Department of Genetics, University of Wisconsin-Madison, Madison, Wisconsin

Requests for reprints: Amy R. Moser, Department of Human Oncology, University of Wisconsin-Madison, K4/310 CSC Box 3684, 600 Highland Avenue, Madison, WI 53792. Phone: 608-265-6520; Fax: 608-263-9947; E-mail: armoser{at}wisc.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Genetic background affects the susceptibility to mammary tumor development in ApcMin/+ mice. Here we report the identification of four novel modifier loci that influence different aspects of mammary tumor development in ApcMin/+ mice. Analysis of tumor development in a backcross of (FVBB6 ApcMin/+) x B6 ApcMin/+ mice has identified a modifier on chromosome 9 that significantly affects tumor multiplicity, and a modifier on chromosome 4 that significantly affects tumor latency and affects tumor number with suggestive significance. This modifier was also identified in a backcross involving 129X1/SvJ and B6 ApcMin/+ mice. A modifier on chromosome 18 specifically affects tumor latency but not tumor number. Kaplan-Meier analysis suggests there is at least an additive interaction affecting tumor latency between the loci on chromosomes 4 and 18. We also identified a modifier locus on chromosome 6 that interacts with the loci on chromosome 4 and chromosome 9 to affect tumor number. These results suggest that multiple genetic loci control different aspects of mammary tumor development. None of these modifiers is associated with intestinal tumor susceptibility, which indicates that these modifiers act on tumor development in a tissue-specific manner. [Cancer Res 2007;67(23):11226–33]


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Breast cancer develops through multiple steps that are rigorously controlled by genetic factors (1). It is essential to identify and characterize genes controlling the development of breast cancer to better understand factors affecting tumor susceptibility and contribute to better diagnosis and treatment. Advances have been made toward identifying common alleles of genes that affect breast cancer susceptibility in humans, using single nucleotide polymorphisms (SNP) and very large sample sizes (2). The risk associated with each of these common variants is small, and such studies have the little power to detect low frequency variants. Mouse models provide useful tools to identify modifiers of cancer with major or minor effect (3). As many of the genes known to be involved in human breast cancer also affect mammary tumor development in mice, modifier genes identified in mice are likely to also play a role in regulating the development of breast cancer in humans.

The transition from hyperplastic lesions to tumors is critical for the development of cancers (4). In the human, ductal and lobular hyperplasias are known risk factors for cancer development. For example, in one study, about 20% of women diagnosed with atypical lobular hyperplasia or lobular carcinoma in situ later developed cancer (5). Interestingly, this risk for cancer development applies to both the involved and contralateral breast in these women (5). This transition is poorly understood due to both the difficulty of identifying and tracking early hyperplastic lesions and the lack of good mouse models.

The ApcMin/+ mouse is a well-characterized mouse model that is predisposed to both intestinal and mammary tumors (6). Upon exposure to the alkylating agent ethylnitrosourea (ENU), >90% of B6-ApcMin/+ female mice develop mammary squamous cell carcinomas (SCC) within 65 days, with an average 3.3 tumors per mouse and an average latency of about 56 days. FVBB6 F1 ApcMin/+ mice are resistant to mammary tumors but susceptible to focal alveolar hyperplasias under the same treatment (7). This suggests that FVB carries alleles at modifier loci of ApcMin that act dominantly to affect the progression of hyperplasias to tumors. The identification of such modifiers could provide biomarkers of hyperplasias that could be helpful in the early diagnosis of breast cancer.

The inheritance of a mutation in APC gives predisposition to the autosomal dominant disorder familial adenomatous polyposis in humans (8, 9). Mouse APC shares 90% amino acid identity with human APC, and germline mutations result in similar phenotypes. Min, a dominant allele of Apc in the mouse, was generated by ENU germline mutagenesis and results in a truncation mutation at codon 850 (10). The insight that genetic background affects intestinal tumor development led to the mapping and molecular identification of Mom1 (modifier of Min 1; refs. 11, 12).

Wnt signaling has been observed to be abnormally activated in many cancers including breast cancer in both mice and humans. APC is critically important in the canonical Wnt signaling pathway and, together with GSK-3β and Axin, can participate in the phosphorylation of β-catenin, which is then targeted for degradation (13, 14). Upon Wnt activation or mutations of other components, β-catenin can translocate into the nucleus and bind with transcription factors, such as TCF4, to activate target genes including cylin D1 and c-myc (13, 14). Transgenic expression of β-catenin, cyclin D1, and c-myc in mammary tissue leads to mammary tumors in mice (15–17). Amplification of cyclin D1 was found in >50% human breast cancer, and somatic APC mutations were found in 18% breast cancer (18–20). The aberrant regulation of β-catenin is a prognostic marker in human breast cancer (21, 22). Overall, these data provide strong evidence that activation of canonic Wnt signaling is a common feature in human breast cancer.

In B6 ApcMin/+ female mice, SCC develop rapidly after ENU treatment with nearly all of the mice developing tumors by 65 days after treatment. In contrast, when treated with ENU, FVBxB6 ApcMin/+ female mice develop multiple alveolar hyperplasias and few SCC with long latency (7). Because the initiation step is most likely the same in both of these strains of mice, loss of the wild-type allele of Apc due to ENU-induced mutation, the difference in mammary tumor susceptibility must be due to genes that affect the process of tumor development. The resistance to mammary tumor development in FVBB6 F1 ApcMin/+ females could be due to the effect of modifier loci on any of the steps of tumor development. However, the susceptibility of FVBB6 F1 ApcMin/+ females to hyperplasias suggests the modifier loci do not affect the initiation of hyperplasia. Thus, the modifier loci most likely affect the transition of hyperplasias to mammary tumors or the progression and growth of mammary tumors. As a first step toward understanding the genetic control of mammary tumor development in these mice, we performed backcross analysis, mapping the traits of tumor number and tumor latency. Further molecular identification of these modifier genes would provide more insights into the regulation of the Wnt signaling pathway and mechanisms of breast cancer development.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Animals and phenotypes. All mice were maintained in the Laboratory Animal Research Facility of University of Wisconsin School of Medicine and Public Health. All procedures were approved by the Institutional Animal Care and Use Committee. Female C57BL/6J (B6), 129X1/SvJ (129X1), and FVB/NTac (FVB) mice were purchased from The Jackson Laboratory or Taconic Labs. C57BL/6J ApcMin/+ (B6 ApcMin/+) mice are maintained by crossing wild-type B6 females from The Jackson Laboratory with B6 ApcMin/+ males. The males used in these experiments were from generations N35-N66.

For the FVBB6 backcross, wild-type FVB females were mated to B6 ApcMin/+ males (N59-66) to produce F1 females. The F1 Apc+/+ females (n = 72) were backcrossed with B6 ApcMin/+ males (n = 48) to produce N2 females. A total of 307 ApcMin/+ N2 females were produced and treated with 50 mg/kg ENU by i.p. injection when between 35 to 40 days of age and were euthanized ~75 days after ENU treatment as described (7). Mice were palpated weekly by the same observer (D. Teske), and palpable tumors were recorded when first noted and then confirmed at necropsy. The total tumor number was determined at necropsy by noting all visible discrete masses. Tumors were collected and fixed in formalin for histologic analysis. We also collected 4-cm regions from the proximal, middle, and distal small intestine and the whole colon for counting intestinal tumors.

For the 129X1B6 backcross, wild-type 129X1 female mice (N = 42) were crossed to B6 ApcMin/+ males (N35-36; n = 7) to produce (129X1B6) F1 mice. Backcross mice were produced by crossing (129X1B6) F1 females to B6 ApcMin/+ males (n = 7), (129X1B6) F1 ApcMin/+ females to B6 male mice (n = 18), and B6 females to (129X1B6) F1 ApcMin/+ male mice (n = 55). A total of 80 129X1B6 backcross ApcMin/+ females were produced. The mice were ENU-treated and followed as described above for the FVB backcross.

Genotyping. For ApcMin/+ genotyping, genomic DNA was obtained after incubating tissues in 100 µL of 50 mmol/L NaOH at 95°C for 1 h and then adding 10 µL of 1 mol/L Tris (pH = 5), modified from the previous report (23). Animals were genotyped for ApcMin/+ by PCR using an allele-specific PCR assay (11).

For both backcross analyses, DNA was isolated from the spleens of backcross females using the DNeasy kit (Qiagen). Simple sequence length polymorphism (SSLP) primers were purchased from Research Genetics or Integrated DNA Technologies. For the FVBxB6 BC, a total of 102 SSLP markers with an average spacing of about 20 cM were genotyped for 180 mice selected from the ends of the tumor number distribution (90 mice with 1 or fewer tumors and 90 with 3 or more). Chromosomes with putative modifier loci were then genotyped with higher density markers for the remaining 127 mice. For the 129 backcross, all 80 mice were genotyped with 79 SSLP markers. Markers are available upon request. The PCR reactions were set up with 20 µL containing 50 to 100 ng genomic DNA, 2.0 µL 10 x buffer, 1.5 µL 25 nmol/L MgCl2, 30 nmol/L forward and reverse primers, and 0.5 µL 10 nmol/L deoxynucleotide triphosphates. The PCR conditions were 2 min at 94°C for 1 cycle; 15 s at 94°C, 45 s at 55°C, and 45 s at 72°C for 35 cycles; and finally, 7 min at 72° for 1 cycle. PCR products were resolved on 4% agarose gels and visualized by ethidium bromide staining.

Statistical analysis. The number of QTL controlling mammary tumor number was estimated as described in Dietrich et al. (11) with the classic formula of Wright, n = (uF1-uP)2/4{sigma}2G. Usually this formula underestimates the number of QTL because the assumptions are not exactly satisfied. MSTAT, a program provided by Dr. Norman Drinkwater of the McArdle Laboratory at the University of Wisconsin-Madison (Madison, WI), was used for the Kaplan-Meier analysis of tumor latency.

Linkage analysis was carried out using the parametric and nonparametric methods implemented in R/qtl (ref. 24; referred to hereinafter as R/qtl-P and R/qtl-NP, respectively) and the nonparametric methods implemented in the Q-link program, provided by Dr. Drinkwater. R/qtl-P implements a hidden Markov model to deal with missing genotypes; the expectation-maximization algorithm was used to perform single QTL genome scans (24). For Q-link, the statistic Zw is determined for each of the markers using the Wilcoxon rank-sum test and a logarithm of odds (LOD) score was calculated by the function LODW = 0.5 (log10 e) (ZW) (ref. 25); R/qtl-NP also uses a Wilcoxon rank-sum test but allows for LOD score evaluation between markers (24). For the linkage analyses of tumor latency, nonparametric methods were used because the raw and transformed phenotypes were not Gaussian. Permutation tests were performed (10,000 permutations per phenotype) to determine thresholds for suggestive (P < 0.05) and significant linkage (P < 0.01) at the genome level.

Two methods were used to assess the presence of interactions among QTL. We first considered a model selection procedure as detailed in Lan et al. (26). That procedure identifies potential interactions among QTL and other genome regions that perhaps do not show significant main effects. In short, the procedure first identifies putative QTL using a LOD score profile. In the initial step, we do not require that putative QTL be statistically significant; they are defined as those with LODs >1, where LOD scores are calculated using R/qtl-P. We then consider all possible models allowing for additive effects among the putative QTL and pairwise interactions. The Bayes Information Criterion (BIC; ref. 27) is used to score each model. The BIC balances goodness-of-model fit with the number of model variables. The model with the best (lowest) BIC is then identified. We also considered the two dimensional scans provided by R/qtl. There, each pair of loci is evaluated allowing for main effects and the pairwise interaction between those effects. The likelihood of the full model is then compared with alternatives to test for additive and interaction effects (24).

To combine the FVBB6 and 129X1B6 backcrosses (28), the FVB and 129X1 alleles were assigned the same genotype and the B6 allele was treated as the other genotype. Therefore, a binary allelic pattern was used, and only loci where both FVB and 129X1 carry alleles that affect tumor development in the same direction will be identified as modifiers.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Tumor number and tumor latency in the FVB backcross. To map modifier loci affecting the development of mammary tumors, 307 backcross ApcMin/+ females were treated with ENU (7). Approximately 90% of the backcross ApcMin/+ female mice developed visible mammary tumors with an average of 2.54 tumors per mouse at necropsy. More than 400 mammary tumors from the first 200 mice were examined histologically, and all were classified as SCC (data not shown). With the classic Wrights formula, it was estimated that 1.3 genetic loci control tumor number in this backcross.

By the end of the experiment, 57% of backcross females had developed at least one palpable tumor (meaning that the tumor was detected before necropsy). The classification of mammary tumors as palpable tumors is dependent on both tumor size and location. However, because some tumors did develop in each of the mammary glands in the set of backcross mice, it is expected that tumors in all mice had approximately equal chances of detection. Among a total of 780 visible mammary tumors in the backcross females, about one third (259 of 780) were palpable and two thirds (521 of 780) were detected only at dissection. On average, the backcross ApcMin/+ females developed 0.84 palpable tumors and 1.70 nonpalpable tumors.

Identification of modifiers affecting tumor number in the FVBB6 backcross. A whole genome scan with 102 SSLP markers spaced an average of 20 cM apart was performed with mice at the two extremes of the tumor number distribution: 90 mice with 0 or 1 tumor and 90 mice with <3 tumors (Fig. 1A ). The rest of the mice were then genotyped only on chromosomes with a LOD score of >1. Three loci, on chromosomes 4, 6, and 9, were associated with mammary tumor number. These QTL are designated Mmom1 (mammary modifier of Min; chromosome 9), Mmom2 (chromosome 4), and Mmom3 (chromosome 6) in order of effect. The summary of the QTL analysis is shown in Table 1 , and the LOD profile for each chromosome is shown in Fig. 1B.


Figure 1
View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. The LOD score profile of mammary modifiers of ApcMin/+. The tumor numbers were analyzed by the parametric method, and the tumor latency was analyzed by the nonparametric method in R/qtl. Solid lines, the result of tumor number analyses; dashed lines, the result of tumor latency analyses. B and D, the suggestive or significant levels; C, the suggestive level. A, the LOD profile of tumor number and tumor latency with 180 selectively genotyped mice from the FVBB6 backcross. B, the LOD profile on the selected chromosomes for tumor number and tumor latency from the entire FVBB6 backcross. C, the LOD profile on the selected chromosomes for tumor number and tumor latency from 129X1B6 backcross. D, the LOD profile of the modifier loci for tumor number and tumor latency analysis from the combined FVBB6 and 129X1B6 crosses.

 

View this table:
[in this window]
[in a new window]

  		Table 1. Summary of the effect of modifier loci on tumor multiplicity and tumor latency using Q-link

 
As shown in Tables 1 and 2 , both the Q-link and R/qtl-P analysis methods gave comparable results, although they adopt different models for analysis. Mmom1, associated with D9Mit182 (55 cM), reaches statistical significance for an effect on tumor multiplicity. Mmom1 heterozygous females show about a 30% decrease in tumor number compared with homozygous mice. Mmom2, on chromosome 4 at D4Mit82 (32.5 cM), shows suggestive significance for an effect on tumor number. Mmom2 heterozygous females developed 24% fewer mammary tumors than homozygous mice.


View this table:
[in this window]
[in a new window]

  		Table 2. The combined analysis of modifiers of ApcMin/+ from FVBBC and 129B6 backcross using R/qtl

 
The modifier on chromosome 6 at D6Mit31 (38.5 cM) does not reach significance, but the approach of Lan (26) applied to the FVBB6 backcross identified an additive model with Mmom1, Mmom2, and Mmom3. Accounting for interactions did not improve the BIC (data not shown). These results are consistent with the joint scans provided by R/qtl. Using that approach, the joint LOD score between Mmom1 and Mmom3 reaches significance (LOD = 4.46; P < 0.01), providing further evidence of the additive effects of these two loci. The effects are shown explicitly in Fig. 2A . When mice are heterozygous at both Mmom1 and Mmom3, tumor number is reduced ~50% compared with mice homozygous at both loci (Fig. 2A; 1.7 ± 1.5 versus 3.2 ± 2.0). Mmom3 also has an additive effect in concert with Mmom2 (Fig. 2B; 3.3 ± 2.2 versus 2.0 ± 1.6). Taken together, these analyses indicate that although the locus on chromosome 6 fails to achieve significance in the one-dimensional scans, this locus does influence tumor number in this backcross.


Figure 2
View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Effect of Mmom1, Mmom2, and Mmom3 on mammary tumor number in the FVBB6 Backcross. A, additive effects of Mmom1 and Mmom3. B, additive effects of Mmom2 and Mmom3.

 
Identification of modifiers associated with tumor latency in the FVBB6 backcross. The number of mammary tumors observed at necropsy reflects the combination of multiple events including lesion initiation, hyperplasia growth, transition of hyperplasias, tumor progression, and tumor growth. Using nonparametric methods, we analyzed the FVBB6 backcross to search for genetic loci controlling the time to tumor (tumor latency). Time to tumor was defined as the time of appearance of the first palpable tumor after ENU treatment. All mice without palpable tumors at necropsy were assigned the highest rank order.

A whole genome scan was carried out as with tumor multiplicity (using the mice on the ends of the tumor number distribution) and chromosomes with LOD > 1 were selected for further genotyping and analysis using all of the mice. Modifier loci on chromosomes 4 and 18 were found to significantly affect tumor latency (Fig. 1B and Table 1). The modifiers on chromosomes 6 and 9 did not show a significant effect on tumor latency. The peak position of the QTL affecting latency is the same as the QTL affecting tumor number on chromosome 4 (D4mit82 and Mmom2). This suggests that Mmom2 could affect both tumor number and tumor latency. Less than 50% of mice carrying an FVB allele at Mmom2 developed a palpable mammary tumor by the end of experiment (75 days after ENU), whereas 50% of mice homozygous for B6 alleles had palpable tumors by 57 days after ENU treatment. A locus on chromosome 18 at D18Mit24 (25 cM) specifically affects tumor latency and was designated Mmom4. This locus shows no effect on the tumor multiplicity. This locus maps about 10 cM distal of the Apc locus, which also maps to chromosome 18.

To determine the effect of each modifier on tumor development, we performed Kaplan-Meier analysis stratifying the mice by genotype at D4Mit82 or D18Mit24. The tumor latency for the mice homozygous at each marker is significantly different from the tumor latency for the mice heterozygous at each marker (Fig. 3A and C ). We then stratified for both markers and found additive effects between Mmom2 and Mmom4 (Fig. 3D). When mice were homozygous for B6 alleles at both loci, around 80% of females developed a palpable tumor. When the mice were heterozygous at both loci, only 40% of the mice developed a palpable tumor (P < 0.001). Heterozygosity for either one of the loci has approximately the same effect on tumor latency.


Figure 3
View larger version (17K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Kaplan-Meier analyses of the effect of modifiers of ApcMin/+ on tumor latency. The P values indicated were determined by the log-rank test. BB, B6 homozygous; BF, B6 and FVB heterozygous; B129 means B6 and 129X1 heterozygous. A, the effect of Mmom2 on tumor latency in the FVBB6 backcross. B, the effect of Mmom2 on tumor latency in the 129X1B6 backcross. C, the effect of Mmom4 on tumor latency in the FVBB6 backcross. D, the interaction of Mmom2 and Mmom4 on tumor latency in the FVBB6 backcross.

 
Identification of Mmom2 in the 129X1B6 backcross. The identification of Mmom2 is supported by an independent backcross involving 129X1 and B6 ApcMin/+ mice. A locus that reaches suggestive significance for an effect on tumor number maps to the same region of chromosome 4 as Mmom2 (Fig. 1C and Table 1). We then tested for an effect on tumor latency. A marker, D4Mit26, which maps about 10 cM distal to D4Mit82, reaches significance for an effect on tumor latency (Fig. 1C and Table 1). The Kaplan-Meier analysis also shows that tumor latency of the D4Mit26 heterozygous mice is significantly different from the homozygous mice (Fig. 2B). Because the modifiers identified in the 129B6 backcross and the FVBB6 backcross affect the same phenotype, and the two peak markers, D4Mit82 and D4Mit26, map close to each other, they most likely represent the same modifier QTL, Mmom2. In this backcross, we also found that a potential modifier affecting tumor number mapped to the region of Mmom3 on chromosome 6 and a potential modifier affecting both tumor latency and tumor number mapped to the same position as Mmom4 on chromosome 18, but neither locus reached significance (Fig. 1C and Table 2).

As the data from both the FVBB6 and the 129B6 backcrosses were collected in a similar way, we combined these two backcrosses to improve the power of modifier mapping (28). Because different markers are used in these two backcrosses, 1-cM interval mapping was used to calculate the LOD score for each modifier with the software R/qtl. The hypothesis is that if a modifier is shared between different crosses, the combined cross can improve the LOD score and would dilute the effect of cross-specific modifier QTL. Figure 1D shows the LOD profile in the combined analysis and the results are summarized in Table 2. On chromosome 4, the LOD score in the region of Mmom2 was highly significant (P < 0.001) for both tumor number and tumor latency. This suggests that the resistant allele of Mmom2 is shared by the FVB and 129X1 strains. On chromosome 6, the LOD score of Mmom3 reached suggestive significance (P < 0.05) for tumor number with no effect on tumor latency. This provides more evidence that there is a shared modifier on chromosome 6 affecting tumor number. On chromosome 9, the LOD score of Mmom1 decreased compared with FVBB6 backcross and there was no effect on tumor latency. This suggests that the 129X1 and B6 strains share alleles with similar effect at Mmom1. On chromosome 18, the LOD score of Mmom4 was highly significant (P < 0.001) for tumor latency, indicating that FVB and 129X1 strains share alleles at Mmom4. In summary, the combined cross analysis confirmed the results of individual backcross analyses and provided more support for the effect of Mmom2, Mmom3, and Mmom4 on tumor development.

Tissue specificity of modifiers. Because ApcMin/+ mice develop both mammary and intestinal tumors, we could test for the effect of any modifiers on both tumor types. None of the Mmom loci affected intestinal tumor development (Table 3 ). D4Mit13 (71 cM), the most distal marker on chromosome 4 in the FVBB6 backcross, is significantly associated with intestinal tumor number but has no effect on mammary tumor number and latency. D4Mit13 maps to the same position as Mom1 (11). This suggests that the FVB strain carries a resistant allele of Mom1 (11). A database search identified a SNP (rs27560348, T/C, and dbSNP126) between B6 and FVB in the phospholipase gene Pla2g2a. This SNP causes a frameshift mutation and generates a mutant allele for Pla2g2a in B6 mice (12). Taken together, these data provide strong evidence that the modifiers of ApcMin show tissue specificity for mammary tumors and intestinal tumors.


View this table:
[in this window]
[in a new window]

  		Table 3. Analysis of modifiers if ApcMin/+ affecting intestinal tumor multiplicity in the FVBB6 backcross using Q-link

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The development of mammary tumors is a complex process that involves a series of events including initiation, the growth of hyperplasias, transition from hyperplasia to tumor, and tumor progression and growth (29, 30). To test for an effect on each stage of tumor development, we recorded the number of tumors and hyperplasias and classified each tumor and hyperplastic lesion histologically. No loci showed an association with the total number of mammary lesions, which included all tumors and hyperplastic lesions (data not shown). These results suggest that modifiers are not involved in DNA repair after ENU treatment, the initiating event in this model. This was not necessarily a surprising result as the number of total lesions in the B6 and FVBxB6 F1 mice is similar (7); thus, we did not expect to see an effect on lesion initiation.

In this analysis, we have identified modifier loci that affect either mammary tumor number or/and tumor latency. Mmom1 on chromosome 9 affects only mammary tumor number, suggesting that it may affect the probability of transition from hyperplasia to tumor. Mmom2 on chromosome 4 affects both tumor number and tumor latency, suggesting that it may affect both the probability of transition from hyperplasia to tumor and the rate of tumor growth. This effect is similar in both backcrosses investigated. Mmom3 on chromosome 6 seems to cooperate with Mmom1 or Mmom2 to affect tumor number. Mmom4 on chromosome 18 affects mammary tumor latency, suggesting that it may affect mammary tumor growth. The additive interaction between Mmom2 and Mmom4 suggests the genes underlying these two loci do not act in the same pathway.

Humans carrying germ-line mutations on APC usually develop colon cancer but very few other cancer types, implying that there are tissue-specific effects of loss of APC function. Our previous data suggest that there is no association between susceptibility to mammary tumors and intestinal tumors in F1 ApcMin/+ mice and raises a question whether there are tissue-specific modifiers of ApcMin (7). Our current data show that the susceptibility to intestinal tumors in this backcross is mostly controlled by Mom1 on chromosome 4, which has no effect on mammary tumor number or tumor latency. Conversely, the mammary modifiers show no association with intestinal tumor multiplicity. Alina et al. (31) reported that recombinant congenic strains (RCS) carrying the ApcMin allele show opposite effects on intestinal and mammary tumors. These opposite effects could be due to one modifier locus with opposite effects in each tissue or the effects of tissue-specific modifiers. Our data favor the hypothesis that the observed phenotype in RCS is caused by different modifiers. The identification of tissue or tumor type–specific modifiers may aid in the identification of the genes underlying the modifier loci based on function. This also indicates the complexity of the phenotype conferred by loss of Apc function on different tissues.

Interaction between modifiers is frequent in colon cancer, lung cancer, and skin cancer (32–35). However, to identify low-penetrance disease-related genes presents a challenge in human and mouse genetics. The powerful genetic manipulation possible in mice makes it feasible to identify these low-penetrance genes. These low-penetrance genes might only be identified when they interact with major QTL or other low-penetrance loci to achieve a statistically significant effect. In our experiments, we used backcrosses to reduce the genetic complexity (36) because of the narrow mammary tumor number range in B6-ApcMin/+, FVBB6 F1-ApcMin/+, and 129X1B6 F1 ApcMin/+ mice. Our results reveal that additive effects between modifiers not only affect mammary tumor number but also regulate mammary tumor latency. These results suggest that multiple loci control each stage of the development of mammary tumors and distinct stages are controlled by different modifier loci.

Only a few QTL affecting mammary tumor susceptibility have been mapped in mice, all of which are in the mouse mammary tumor virus (MMTV)-PyVT transgenic mouse model (37, 38). Mmom1 maps near Apmt2, which interacts with Apmt1 (on chromosome 15) to affect tumor latency but not tumor number (37) in MMTV-PyVT mice. Mmom2 maps near Mmtg1, which affects total mammary tumor weight but not tumor number or tumor latency (38). However, the FVB alleles confer sensitivity in the MMTV-PyVT model, whereas the FVB alleles confer resistance in the ApcMin/+ mouse model. As they affect different phenotypes with different susceptibility/resistant status, these modifier loci most likely represent different genes, although they map to a similar location. Several QTL affecting various aspects of mammary tumor development in the rat have been mapped to the orthologous regions to the loci we have mapped on chromosomes 4. 9, and 18 (39, 40). Some of the rat and mouse loci that map to orthologous regions seem to have similar effects on tumor development. However, given the imprecision of the mapping and the differences in the models, it is premature to conclude that we have mapped the same genes. However, with more precise mapping and further phenotypic analysis, it will be possible to determine whether the same genes have indeed been identified.

Mmom1 maps to the distal part of chromosome 9, which is orthologous to human 3p21-23. This region undergoes frequent loss of heterozygosity (LOH) and hypermethylation in lung cancer and breast cancer (41, 42). In addition, the Mmom2 region is orthologous to human chromosome 9q, which has been shown to undergo LOH in about 40% of breast cancers (43). Therefore, it is possible that the loci we have identified may have a role in the development of human cancer.

B6 and FVB are two strains that are frequently used for mouse models of cancer research. As such, information about loci that alter the course of tumor development or progression can possibly be applied to other models. The identification of QTL is the first important step toward understanding the development of mammary tumors, and the characterization of modifier genes could provide valuable insights into the regulation of mammary tumors. These modifier genes could be biomarkers for early stages of breast cancer and improve the diagnosis of breast cancer.


   Acknowledgments
 
Grant support: Department of Defense Breast Cancer Program DAMD17-02-1-0627 (A.R. Moser) and W81XWH-06-1-0413 (H. Wang).

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.

We thank Laura Hegge and Jared Finger for assistance with the mouse colony and backcross experiments; Dr. Norman Drinkwater for the Q-Link program and for the advice; Dr. Karl Broman for the advice on using R/qtl; Dr. Ruth Sullivan for the assistance with pathology; Meng Chen for the help with statistical analyses; and the University of Wisconsin Comprehensive Cancer Center Histology Core for assistance with processing and sectioning of tumor samples.

Received 7/ 3/07. Revised 9/27/07. Accepted 10/ 8/07.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100:57–70.[CrossRef][Medline]
  2. Easton DF, Pooley KA, Dunning AM, et al. Genome-wide association study identifies novel breast cancer susceptibility loci. Nature 2007;447:1087–93.[CrossRef][Medline]
  3. Balmain A. Cancer as a complex genetic trait: tumor susceptibility in humans and mouse models. Cell 2002;108:145–52.[CrossRef][Medline]
  4. Cardiff RD, Aguilar-Cordova E. Proto-neoplasia revisited: the molecular biology of mouse mammary hyperplasia. Anticancer Res 1988;8:925–33.[Medline]
  5. Page DL, Schuyler PA, Dupont WD, Jensen RA, Plummer WD, Jr., Simpson JF. Atypical lobular hyperplasia as a unilateral predictor of breast cancer risk: a retrospective cohort study. Lancet 2003;361:125–9.[CrossRef][Medline]
  6. Moser A, Mattes E, Dove W, Lindstrom M, Haag J, Gould M. ApcMin, a mutation in the murine Apc gene, predisposes to mammary carcinomas and focal alveolar hyperplasias. PNAS 1993;90:8977–81.[Abstract/Free Full Text]
  7. Moser AR, Hegge LF, Cardiff RD. Genetic background affects susceptibility to mammary hyperplasias and carcinomas in ApcMin/+ mice. Cancer Res 2001;61:3480–5.[Abstract/Free Full Text]
  8. Groden J, Thliveris A, Samowitz W, et al. Identification and characterization of the familial adenomatous polyposis coli gene. Cell 1991;66:589–600.[CrossRef][Medline]
  9. Nishisho I, Nakamura Y, Miyoshi Y, et al. Mutations of chromosome 5q21 genes in FAP and colorectal cancer patients. Science 1991;253:665–9.[Abstract/Free Full Text]
  10. Su LK, Kinzler KW, Vogelstein B, et al. Multiple intestinal neoplasia caused by a mutation in the murine homolog of the APC gene. Science 1992;256:668–70.[Abstract/Free Full Text]
  11. 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;75:631–9.[CrossRef][Medline]
  12. MacPhee M, Chepenik KP, Liddell RA, Nelson KK, Siracusa LD, Buchberg AM. The secretory phospholipase A2 gene is a candidate for the Mom1 locus, a major modifier of ApcMin-induced intestinal neoplasia. Cell 1995;81:957–66.[CrossRef][Medline]
  13. Rubinfeld B, Souza B, Albert I, et al. Association of the APC gene product with β-catenin. Science 1993;262:1731–4.[Abstract/Free Full Text]
  14. Polakis P. Wnt signaling and cancer. Genes Dev 2000;14:1837–51.[Free Full Text]
  15. Michaelson JS, Leder P. β-catenin is a downstream effector of Wnt-mediated tumorigenesis in the mammary gland. Oncogene 2001;20:5093–9.[CrossRef][Medline]
  16. Schoenenberger CA, Andres AC, Groner B, van der Valk M, LeMeur M, Gerlinger P. Targeted c-myc gene expression in mammary glands of transgenic mice induces mammary tumours with constitutive milk protein gene transcription. EMBO J 1988;7:169–75.[Medline]
  17. Wang TC, Cardiff RD, Zukerberg L, Lees E, Arnold A, Schmidt EV. Mammary hyperplasia and carcinoma in MMTV-cyclin D1 transgenic mice. Nature 1994;369:669–71.[CrossRef][Medline]
  18. Gillett C, Smith P, Gregory W, et al. Cyclin D1 and prognosis in human breast cancer. Int J Cancer 1996;69:92–9.[CrossRef][Medline]
  19. Furuuchi K, Tada M, Yamada H, et al. Somatic mutations of the APC gene in primary breast cancers. Am J Pathol 2000;156:1997–2005.[Abstract/Free Full Text]
  20. Zhang SY, Caamano J, Cooper F, Guo X, Klein-Szanto AJ. Immunohistochemistry of cyclin D1 in human breast cancer. Am J Clin Pathol 1994;102:695–8.[Medline]
  21. Dolled-Filhart M, McCabe A, Giltnane J, Cregger M, Camp RL, Rimm DL. Quantitative in situ analysis of β-catenin expression in breast cancer shows decreased expression is associated with poor outcome. Cancer Res 2006;66:5487–94.[Abstract/Free Full Text]
  22. Lin S-Y, Xia W, Wang JC, et al. β-Catenin, a novel prognostic marker for breast cancer: its roles in cyclin D1 expression and cancer progression. PNAS 2000;97:4262–6.[Abstract/Free Full Text]
  23. Truett GE, Heeger P, Mynatt RL, Truett AA, Walker JA, Warman ML. Preparation of PCR-quality mouse genomic DNA with hot sodium hydroxide and tris (HotSHOT). Biotechniques 2000;29:52–4.[Medline]
  24. Broman KW, Wu H, Sen S, Churchill GA. R/qtl: QTL mapping in experimental crosses. Bioinformatics 2003;19:889–90.[Abstract/Free Full Text]
  25. Bilger A, Bennett LM, Carabeo RA, et al. A potent modifier of liver cancer risk on distal mouse chromosome 1: linkage analysis and characterization of congenic lines. Genetics 2004;167:859–66.[Abstract/Free Full Text]
  26. Lan H, Kendziorski CM, Haag JD, Shepel LA, Newton MA, Gould MN. Genetic loci controlling breast cancer susceptibility in the Wistar-Kyoto rat. Genetics 2001;157:331–9.[Abstract/Free Full Text]
  27. Schwarz G. Estimating the dimension of a model. Ann Stat 1978;6:461–4.[CrossRef]
  28. Li R, Lyons MA, Wittenburg H, Paigen B, Churchill GA. Combining data from multiple inbred line crosses improves the power and resolution of quantitative trait loci mapping. Genetics 2005;169:1699–709.[Abstract/Free Full Text]
  29. Going JJ. Stages on the way to breast cancer. J Pathol 2003;199:1–3.[CrossRef][Medline]
  30. Zhu K, Bernard LJ, Levine RS, Williams SM. Estrogen receptor status of breast cancer: a marker of different stages of tumor or different entities of the disease? Med Hypotheses 1997;49:69–75.[CrossRef][Medline]
  31. Czarnomska A, Krysiak E, Piskorowska J, et al. Opposite effects of modifiers of the ApcMin mutation in intestine and mammary gland. Cancer Res 2003;63:4533–7.[Abstract/Free Full Text]
  32. Fijneman RJ, Jansen RC, van der Valk MA, Demant P. High frequency of interactions between lung cancer susceptibility genes in the mouse: mapping of Sluc5 to Sluc14. Cancer Res 1998;58:4794–8.[Abstract/Free Full Text]
  33. Nagase H, Mao JH, de Koning JP, Minami T, Balmain A. Epistatic interactions between skin tumor modifier loci in interspecific (spretus/musculus) backcross mice. Cancer Res 2001;61:1305–8.[Abstract/Free Full Text]
  34. Fijneman RJ, de Vries SS, Jansen RC, Demant P. Complex interactions of new quantitative trait loci, Sluc1, Sluc2, Sluc3, and Sluc4, that influence the susceptibility to lung cancer in the mouse. Nat Genet 1996;14:465–7.[CrossRef][Medline]
  35. Ruivenkamp CA, Csikos T, Klous AM, van Wezel T, Demant P. Five new mouse susceptibility to colon cancer loci, Scc11–15. Oncogene 2003;22:7258–60.[CrossRef][Medline]
  36. Darvasi A. Experimental strategies for the genetic dissection of complex traits in animal models. Nat Genet 1998;18:19–24.[CrossRef][Medline]
  37. Le Voyer T, Lu Z, Babb J, Lifsted T, Williams M, Hunter KW. An epistatic interaction controls the latency of a transgene-induced mammary tumor. Mamm Genome 2000;11:883–9.[CrossRef][Medline]
  38. Le Voyer T, Rouse J, Lu Z, Lifsted T, Williams M, Hunter KW. Three loci modify growth of a transgene-induced mammary tumor: suppression of proliferation associated with decreased microvessel density. Genomics 2001;74:253–61.[CrossRef][Medline]
  39. Gould KA, Tochacek M, Schaffer BS, et al. Genetic determination of susceptibility to estrogen-induced mammary cancer in the ACI rat: mapping of Emca1 and Emca2 to chromosomes 5 and 18. Genetics 2004;168:2113–25.[Abstract/Free Full Text]
  40. Schaffer BS, Lachel CM, Pennington KL, et al. Genetic bases of estrogen-induced tumorigenesis in the rat: mapping of loci controlling susceptibility to mammary cancer in a Brown Norway x ACI intercross. Cancer Res 2006;66:7793–800.[Abstract/Free Full Text]
  41. Maitra A, Wistuba II, Washington C, et al. High-resolution chromosome 3p allelotyping of breast carcinomas and precursor lesions demonstrates frequent loss of heterozygosity and a discontinuous pattern of allele loss. Am J Pathol 2001;159:119–30.[Abstract/Free Full Text]
  42. Wistuba II, Behrens C, Virmani AK, et al. High resolution chromosome 3p allelotyping of human lung cancer and peneoplastic/preinvasive bronchial epithelium reveals multiple, discontinuous sites of 3p allele loss and three regions of frequent breakpoints. Cancer Res 2000;60:1949–60.[Abstract/Free Full Text]
  43. An H-X, Claas A, Savelyeva L, et al. Two regions of deletion in 9p23–24 in sporadic breast cancer. Cancer Res 1999;59:3941–3.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wang, H.
Right arrow Articles by Moser, A. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wang, H.
Right arrow Articles by Moser, A. R.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online