Skip Navigation


Carcinogenesis Advance Access originally published online on December 13, 2006
Carcinogenesis 2007 28(6):1356-1363; doi:10.1093/carcin/bgl239
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Material
Right arrow All Versions of this Article:
28/6/1356    most recent
bgl239v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Wang, L.
Right arrow Articles by Banerjee, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wang, L.
Right arrow Articles by Banerjee, S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2006. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Mammary carcinogenesis in transgenic mice expressing a dominant-negative mutant of DNA polymerase ß in their mammary glands

Liming Wang{dagger}, Nandan Bhattacharyya1,{dagger}, Thangaiyan Rabi, Li Wang and Sipra Banerjee*

Department of Cancer Biology, Lerner Research Institute, Cleveland Clinic, Cleveland, OH 44195, USA
1 Present address: Department of Biotechnology, Haldia Institute of Technology, Pin 721657, West Bengal, India

* To whom correspondence should be addressed. Tel: +1 216 444 0631; Fax: +1 216 445 6269; Email: banerjs{at}ccf.org


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
DNA polymerase ß (polß) is a major contributor to mammalian DNA damage repair through its gap-filling DNA synthesis and 5'-deoxyribose phosphate lyase activities. In this way, polß plays pivotal roles in the repair of oxidative DNA damage, replication, embryonic survival, neuronal development, meiosis, apoptosis and telomere function. A 36 kDa truncated polß{Delta} protein is expressed in human colorectal, breast, lung and renal carcinomas, but not in normal matched tissues. Interestingly, a binary protein–protein complex of polß{Delta} and X-ray cross-complementing group 1 acts as dominant-negative mutant. In this study, the potential tumorigenic activity of polß{Delta} was examined in nude and transgenic mouse models. Mouse embryonic fibroblasts (MEFs) expressing polß{Delta} in the absence of endogenous polß exhibited increased susceptibility to N-methyl-N-nitrosourea (MNU)-induced morphological transformation as compared with cells expressing wild-type (WT) polß. This was accompanied by reduced gap-filling DNA synthesis activity. Anchorage-independent transformed cells derived from polß{Delta}-expressing MEFs induced 100% tumor occurrence in nude mice. To support these data, we established transgenic mice expressing polß{Delta} specifically in the mammary glands from a whey acidic protein promoter-driven transgene. This is the first report of transgenic mice with tissue-specific expression of polß{Delta}. MNU-induced tumor formation was analyzed in transgenic mice expressing polß{Delta} together with endogenous WT polß in their mammary glands and in normal control mice expressing only WT polß. The latent period of tumor appearance was markedly shorter and tumor incidence was significantly higher in transgenic animals than in control animals treated under the same conditions. These results indicate that cells expressing the mutant polß{Delta} display an enhanced sensitivity to MNU that probably underlies an increased susceptibility to tumorigenesis.

Abbreviations: BER, base excision repair; MEF, mouse embryonic fibroblast; MNU, N-methyl-N-nitrosourea; PBS, phosphate-buffered saline; PCR, polymerase chain reaction; polß, polymerase ß; WAP, whey acidic protein; WT, wild type; XRCC1, X-ray cross-complementing group 1


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
DNA polymerase ß (polß) is a major contributor to the repair of DNA damage via single-nucleotide gap-filling DNA synthesis in the mammalian short-patch base excision repair (BER) pathway (1,2). Polß has been shown to play a pivotal role in repairing oxidative damage in DNA (3–5) and Sobol et al. (6) reported that BER activity was defective in a mouse embryonic fibroblast (MEF) cell line derived from a polß knockout mouse. A DNA repair function for the 39 kDa polß enzyme is consistent with its domain structure. An 8 kDa N-terminal domain has 5'-deoxyribose phosphate lyase and single-strand DNA-binding activities and a 31 kDa C-terminal domain has catalytic activity (1,7–9). Following from its repair function, polß plays key roles in a number of cellular processes such as DNA replication, embryonic survival, neuronal development, meiosis, apoptosis, maintenance of chromosomal integrity and establishment of drug-resistant phenotype (10–15). In addition, interaction between polß and telomere repeat-binding factor 2 has been shown to cause telomere dysfunction (16). Additionally, a high mutation frequency with lower fidelity of DNA synthesis has been reported (17), mediated by a variant of the polß gene expressed in a colorectal adenocarcinoma (18).

We identified a specific 87 bp deletion encoding amino acid residues 208–236 in the polß cDNA in primary human colorectal, breast and lung tumors and in a primary culture of renal cell carcinoma (18,19–22). Both wild-type (WT) polß protein and the mutant polß protein (polß{Delta}) lacking 29 amino acids within the catalytic domain were expressed in colorectal and breast tumors and renal cell carcinoma (20,22,23). Other polß variants have been shown to be expressed in prostate, bladder and gastric adenocarcinomas (19,24–26). Mutations in the polß genomic sequence have also been reported in the early stages of serous ovarian cancer (27). Taken together with the known DNA repair function of polß, the observation of polß variants in tumors strongly suggests that polß plays an important role in determining susceptibility to neoplasia.

Further results suggest that polß{Delta} inhibits functions of the WTpolß protein in a dominant-negative manner (23,28). Interestingly, a binary protein–protein complex of polß{Delta} and X-ray cross-complementing group 1 (XRCC1), a nuclear BER protein, mediates such dominant-negative activity (29). The survival and growth of cells expressing polß{Delta} were markedly reduced upon exposure to N-methyl-N-nitro-N-nitrosoguanidine and N-methyl-N-nitrosourea (MNU), two known DNA alkylating agents compared with cells expressing WTpolß protein (23,28,30). These results suggest that the expression of polß{Delta}, a dominant-negative mutant, may contribute to hypersensitivity of these cells to N-methyl-N-nitro-N-nitrosoguanidine or MNU.

We hypothesize that cells expressing polß{Delta} protein will display diminished DNA repair activity. Consequently, due to the persistence of lesions in DNA, error-prone replication will occur that may initiate a chain of events leading to expression of a tumorigenic phenotype. To test this hypothesis, we took two approaches. First, we examined survival, morphological transformation and gap-filling DNA synthesis activity of MNU-treated cells expressing polß{Delta} in the absence of WTpolß. The ability of these cells to form tumors in nude mice was also investigated. In a second approach, we established a transgenic mouse line expressing the polß{Delta} specifically in the mammary glands. In this model, endogenous WTpolß was expressed in the same cells as the transgene-encoded polß{Delta} protein, allowing the dominant-negative function of polß{Delta} to be expressed. Tumor formation in transgenic and control mice treated with MNU was evaluated. The data from both of these experimental approaches provide evidence that polß{Delta} expression potentiates MNU-induced cellular transformation and tumor formation. Moreover, the increased susceptibility of polß{Delta} transgenic mice to MNU-induced mammary carcinogenesis supports a dominant-negative role for the polß{Delta} mutant.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Cell lines
The stable MEF 19.4{Delta}P cell line was established in this laboratory by transfecting 19.4 MEF derived from a polß knockout mouse (6) with a polß{Delta}208–236 plasmid (23). G418-resistant colonies were selected by growth in Dulbecco's modified Eagle's medium–glutamax. The 19.4{Delta}P cells were grown in Dulbecco's modified Eagle's medium supplemented with 80 µg/ml hygromycin, 700 µg/ml G418, 10% fetal bovine serum and penicillin and streptomycin. To obtain average activity, a pooled cell line named 19.4{Delta}P was made by mixing an equal number of cells from 30 individual expanded colonies. The 16.3 MEF cell line (6) expressing WT polß was used as a control (23,29).

Western blot analysis
To determine expression of polß in cells or tumor tissues, cell extracts were prepared (23) and separated by 12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis. The proteins were transferred to membrane and detected by a purified anti-polß antibody.

Gap-filling DNA synthesis activity assay
The gap-filling DNA synthesis activity in nuclear extracts was determined essentially as described in (23,28). A 51 bp double-stranded DNA with a G:U mismatch at position 22 was used as the substrate. The reaction mixture consisted of 10 U creatine phosphokinase, 2 µg bovine serum albumin, 500 mM 4-(2-Hydroxyethylpiperazine-1-ethanesulfonic acid, pH 7.9, 100 mM MgCl2, 600 mM KCl, 20 mM dithiothreitol, nuclear extract and 32P-dCTP (3000 Ci/mmol) in a volume of 50 µl. It was incubated at 37°C for 30 min and cooled on ice to terminate the reaction. The product was separated by 6% polyacrylamide gel electrophoresis and the radioactivity was detected on the X-ray film at –80°C.

Cell survival assay
A colony-forming assay (23) was used to quantitate the survival of MNU-treated cells. A freshly made stock solution of MNU, >99% pure (NCI Chemical Carcinogen Reference Standard Repository, St Louis, MO), in dimethyl sulfoxide was serially diluted with medium and used immediately. Prior to the treatment, 400 cells were grown in a 60 mm dish for 18 h. Cells were exposed to MNU for 1 h at 37°C, washed with phosphate-buffered saline (PBS) buffer and allowed to grow for 7–10 days. Cells were then fixed with 70% methanol and stained with 10% Giemsa. Colonies containing >30 cells were counted.

Morphological transformation assay
Two thousand cells were treated with MNU at various doses for 1 h (31,32). Cells were also incubated in the medium containing dimethyl sulfoxide (final concentration: 0.005%), which served as a solvent control. The MNU–dimethyl sulfoxide was removed through multiple washings with PBS and cells were allowed to grow for 6 weeks with a change of medium twice a week. Transformed colonies were then fixed with 70% methanol, stained with Giemsa and scored as either type II or type III (31,32). The foci index was calculated as the number of foci per dish. For transfer into nude mice (see below), type III foci were isolated from unstained dishes using cloning cylinders.

Tumorigenic potential of transformed cells
Type III transformed cells were grown in mass culture for three passages. To determine whether growth was anchorage independent or dependent in soft agar, cells were seeded in 0.33% agar on a 5% agar base layer (31,32). Untreated cells were used as a control. Colonies of transformed cells that formed in the soft agar layer were isolated using Pasteur pipettes and grown in mass culture. Cells (5 x 106) from these cultures were suspended in PBS and were injected subcutaneously into the backs of male athymic Nu/Nu mice (32). Animals were examined for tumor appearance every third day. Resulting tumors were excised, fixed in 4% formalin, paraffin blocked and stained with hematoxylene and eosin for histopathological evaluation. Injected mice were observed closely for body weight loss, abnormal behavior such as poor physical activity, heavy breathing, poor eating or drinking, serious scratching or for any signs of pain. Mice approaching one of these states were humanely euthanized in a CO2 chamber.

Construction of the polß{Delta} transgene
To establish a transgenic human polß{Delta} mouse line, we generated a 2.9 kb transgene construct consisting of a 5' whey acidic protein (WAP) promoter followed by the polß{Delta}208–236 coding sequence, WAP exon 1 (partial, 33 bp), WAP exon 3 (partial), WAP intron C, WAP exon 4 and 70 bp of the WAP 3' untranslated region including its polyadenylation signal. The polß{Delta} coding sequence was amplified from cDNA (23) with SpeI/polß forward (5'-ATCTAGACTAGTCCGGAGCTGGGTTGCTCCTG-3') and HindIII/polß reverse (5'-ATACCCAAGCTTTCATTCGCTCCGGTCCTTGG-3') primers. The polymerase chain reaction (PCR)-amplified product was cloned into a pBluescript vector that has WAP 3' sequence at SpeI–HindIII sites. The polß{Delta} and +WAP 3' sequence was further amplified from the resulting plasmid pPolß{Delta}/WAP 3' using the forward primer (as above) and a polß reverse primer containing a SacII cleavage site at the 3' end (5'-ATATCCCCGCGGTATAGGGCGAATTGGGTACC-3'). The amplified product was then cloned into the SpeI–SacII sites of the plasmid pBL103, 3' to the WAP 5' sequence. The orientation of the Polß{Delta}/WAP insert and the cloning junctions were confirmed by sequencing. The transgene was excised from the plasmid for microinjection using BssHII.

Microinjection and generation of transgenic mice
The purified 2.9 kb WAP-polß{Delta} construct was microinjected into the pronuclei of B6CBA mice by our Transgenic Core Facility. Sixteen pups were born. Tail clips were taken from the pups at 7–10 days of age. DNA was isolated from the tail clips and analyzed by PCR and Southern blot for the presence of the transgene.

DNA PCR
DNA was initially characterized by PCR using a WAP+1 forward (5'-ATCAGTCATCACTTG CCTGCCGCCG-3') and a polß reverse primer (5'-TTCTTCTCAAAGTTTGCGAG-3').

Southern blot assay
DNA samples from all 16 mice were digested with SpeI and HindIII and analyzed by Southern blot using an ~960 bp SpeI + HindIII-digested polß{Delta} cDNA transgene probe. The probe was labeled with {alpha}-32P-dCTP by the random priming method.

Treatment of polß{Delta} transgenic mice with MNU
To induce WAP-polß{Delta} transgene expression in the mammary glands, timed 7-day pregnant female transgenic mice (8 weeks old) were used in these experiments. Mice were injected intra-peritoneally with 50 mg/kg MNU in PBS once a week for 7 weeks. A control group of pregnant transgenic mice received weekly injections of PBS only. An additional control group of pregnant, female non-transgenic Balb/c animals of similar genetic background and age were treated with MNU in a manner identical to the transgenic animals. This group represented induction of tumors by MNU in normal animals. Another group of normal control mice was given injections of PBS only. This group controlled for spontaneous occurrence of tumors. Animals were examined for tumor appearance by palpation once a week for 3 months and then every third day. The latency period before the first tumor appearance and the tumor index (the number of tumor bearing mice divided by the total number of mice given MNU) were recorded. Animals were watched closely for weight loss, abnormal behavior such as slow mobility, heavy breathing, poor eating or drinking or serious scratching and for any signs of pain. Animals showing any of these signs were euthanized humanely and were subjected to a complete necropsy. Tumors or organs were fixed in Histochoice solution, paraffin embedded, sectioned and stained with hematoxylin and eosin.

All studies involving mice were performed in accordance with National Institutes of Health and Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC) guidelines for the humane care and use of research animals.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
Characterization of polß knockout MEFs with stable expression of the polß{Delta} mutant
In order to examine the biological relevancy of the expression of polß{Delta}208–236 variant found in a number of cancers, we established a stable MEF cell line that lacks WTpolß expression, but expresses the polß{Delta}208–236 mutant. This cell line, 19.4{Delta}P, was generated by transfection of the 19.4 MEF cell line that was originated from a polß knockout mouse (6). The well-characterized 19.4 cells do not express WT polß and are lacking in DNA repair activity (6). The expression and function of polß in the newly generated 19.4{Delta}P cells were characterized by western blot analysis and gap-filling DNA synthesis activity. As expected, the 19.4{Delta}P cells expressed only a 34 kDa form of polß protein, corresponding to the deletion mutant (Figure 1A, lane 2). A 39 kDa polypeptide corresponding to WTpolß was expressed in 16.3 cells used as a positive control (Figure 1A, lane 3). No polß protein was detected in extracts from untransfected 19.4 cells. A 41 kDa actin expression indicates equal loading of all samples.


Figure 1
View larger version (31K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. (A) The upper panel shows western blot analysis of expression of the polß protein in cell extracts of 19.4 (lane 1), 19.4{Delta}P (lane 2) and 16.3 (lane 3) cells. Protein (25 µg) was loaded in each lane. The bottom panel shows the same membrane reprobed with an anti-ß-actin antibody to confirm that each lane contained equivalent amounts of protein. (B) Gap-filling DNA synthesis activity (51 bp repair product) in nuclear extracts of 19.4 (lane 1), 19.4{Delta}P (lane 2) and 16.3 (lane 3) cells.

 
To characterize the activity of the polß mutant expressed in the 19.4{Delta}P cells, DNA repair activity was measured in nuclear extracts using gap-filling DNA synthesis assay. The repair activity was reduced substantially in the 19.4{Delta}P cells (Figure 1B, lane 2) as compared with the 16.3 cells (Figure 1B, lane 3). These results indicate that the deletion of 29 amino acid residues in the catalytic domain of polß results in an enzyme (polß{Delta}) that has impaired DNA repair capability.

Expression of polß{Delta} adversely affects survival of MEFs exposed to the mutagen, MNU
To determine the effect of polß{Delta} expression on the survival of cells exposed to a DNA damage-inducing mutagen, 19.4{Delta}P cells and 16.3 (WTpolß) cells were treated with increasing concentrations of MNU from 1 to 100 nM. Cell survival, expressed as relative cloning efficiency decreased from 53 to 6% for 19.4{Delta}P cells treated with 1–100 nM of MNU (Table I). At a concentration of 300 nM, MNU completely shut off growth of these cells. The survival of 16.3 cells used as a positive control also decreased in a dose-dependent manner, although the survival rates of 19.4{Delta}P cells were markedly reduced compared with the survival rates of equivalently treated 16.3 cells. For example, following treatment with 1 nM MNU, 100% of the 16.3 cells survived, but only 53% of the 19.4{Delta}P cells survived. Although none of the 19.4{Delta}P cells treated with 300 nM MNU survived, similarly treated 16.3 cells still had a relative cloning efficiency of 37%. The 19.4 cells devoid of any polß expression were used as a negative control and these cells were found to be the most sensitive to MNU. These results indicate that the lack of efficient DNA repair activity due to expression of a polß deletion mutant causes 19.4{Delta}P cells to be more sensitive to MNU. Thus, WTpolß protein plays an important role in the survival of cells following exposure to DNA-damaging agent.


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

  		Table I. Survival of cells treated with MNU

 
Morphological transformation of 19.4{Delta}P cells treated with MNU
We next examined how susceptible 19.4{Delta}P cells were to transformation as compared with 16.3 cells. Having established the relationship between cell growth and doses of MNU, we used MNU as the transforming agent in these experiments. Transformed colonies or foci when unstained were easily visible in Petri dishes by the naked eye. Foci were scored as type II and type III (31,32). Type II foci cells were characterized as a piled cluster of markedly basophilic cells, moderately polar, in which criss-crossing was not pronounced. Type III foci cells were intensely basophilic and highly polar, showing marked piling with disorientation, criss-crossing, occassional chording and invasiveness to neighboring normal cells' monolayers. The morphology of a typical type III transformed focus, generated in MNU-treated 19.4{Delta}P cells, is shown in Figure 2. Type III foci appeared in MNU-treated 19.4{Delta}P cultures earlier than in similarly treated 16.3 cultures. Type III foci were observed 2 weeks after treatment of 19.4{Delta}P with 50–100 nM MNU. The foci index for both 19.4{Delta}P and 16.3 cells increased with increasing the MNU concentrations (Table II). However, the foci index was consistently significant in 19.4{Delta}P than in 16.3 cells. For example, following treatment with 50 nM MNU, the foci index for the 19 4{Delta}P cells was 15 times higher than that of the 16.3 cells. These results indicate that expression of the mutant polß protein in the absence of WTpolß makes cells more vulnerable to morphological transformation mediated by a DNA-damaging agent such as MNU.


Figure 2
View larger version (146K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. Morphology of a representative type III colony of 19.4{Delta}P cells treated with MNU using x10 magnification.

 


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

  		Table II. Morphological transformation of 19.4{Delta}P and 16.3 cells treated with MNU

 
Tumor formation by transformed 19.4{Delta}P cells in nude mice
To test whether transformed 19.4{Delta}P cells are tumorigenic, we selected cells displaying anchorage-independent growth. Although the growth of many normal eukaryotic cells is dependent upon attachment to a matrix, transformed cells are often able to grow in suspension in semi-solid media such as agar. Hence, anchorage-independent growth is considered a marker of transformed fibroblast and epithelial cells (31,32). We randomly selected four 19.4{Delta}P MNU-transformed cell lines and two 16.3 MNU-transformed cell lines to test in an anchorage-independent growth assay. All of these cell lines had formed type III foci following MNU exposure as described in the preceding section. Large distinct colonies were formed in agar dishes with all four 19.4{Delta}P transformed cell lines. In contrast, although few small colonies were detected in agar dishes with the 16.3 tranformants, these colonies disappeared after 3–5 days. After 7 days of growth in agar, 19.4{Delta}P colonies were picked from agar dishes and grown in mass culture for injection into nude mice. To minimize the numbers of mice required, we injected cells from just one of the lines, 19.4{Delta}P#53 (from #53 colony) into nude mice. Tumors appeared at multiple sites in these mice within 4 weeks (Figure 3); tumors were observed in 100% of the mice injected with 19.4{Delta}P#53 cells (Table III). Histopathological examination revealed that the tumors that developed were fibrosarcomas (cells were fibroblasts). These results demonstrate that MNU-treated 19.4{Delta}P cells that display anchorage-independent growth in culture have neoplastic activity in vivo. As expected, no tumors formed in nude mice inoculated with 16.3 cells expressing WTpolß protein even after 4 months. Since MNU-transformed 16.3 cells did not survive in agar, these cells were not injected into mice. Interestingly, tumors did not develop in nude mice injected with parental 19.4{Delta}P cells that were not exposed to MNU. These data suggest that polß{Delta} expression alone is not sufficient to cause tumor induction in this model system; another factor, e.g. MNU or another DNA-damaging agent, is required. This is consistent with the findings of Bergoglio et al. (33) who suggested a single polß over-expression event was not sufficient to initiate tumorigenesis in vivo.


Figure 3
View larger version (86K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. Representative appearance of multiple tumors in nude mice injected with transformed 19.4{Delta}P cells.

 


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

  		Table III. Tumorigenic activity of transformed 19.4{Delta}P cells in nude mice

 
Polß expression and DNA repair activity in transformed 19.4{Delta}P cells and the tumors that they form in nude mice
To evaluate whether the transformation or tumor formation processes modified the expression or function of polß, we performed western blot analysis using extracts from 19.4{Delta}P anchorage-independent MNU-transformed cells (clone #53) and two tumors that grew in nude mice inoculated with 19.4{Delta}P#53 cells (#29 and #53). Cell extracts from 16.3 and 19.4{Delta}P (non-MNU treated) cells were used as controls. As shown in Figure 4A, a 36 kDa protein corresponding to the polß deletion mutant was observed in the 19.4{Delta}P#53 cell extract (lane 2) and there was no expression of WTpolß protein. This was expected since these cells originated from untransformed 19.4{Delta}P cells that do not express WTpolß protein (lane 1). The 36 kDa mutant protein was also expressed in extracts from tumor tissues #29 and #53 (lanes 3 and 4) along with the 39 kDa endogenous polß protein. As expected, the 39 kDa WTpolß protein and the 36 kDa mutant polß protein were expressed in 16.3 (lane 5) and 19.4{Delta}P cells (lane 1), respectively. The bottom panel demonstrated equal loading by 41 kDa ß-actin protein expression. These data indicate that expression of the deleted form of polß is maintained in the 19.4{Delta}P cells throughout their transformation and growth as tumors in nude mice.


Figure 4
View larger version (28K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. (A) Western blot analysis of polß and polß{Delta} proteins in cell extracts (100 µg per lane) from 19.4{Delta}P and 19.4{Delta}P#53 cells and nude mouse tumors. Lane 1, untransformed 19.4{Delta}P cells; lane 2, #53 clone of transformed 19.4{Delta}P cells selected in soft agar assay; lanes 3 and 4, tumor#29 and tumor#53 from nude mice and lane 5, 16.3 positive control. Upper panel shows 36 and 39 kDa polß proteins. Lower panel shows equal loading by expression of 41 kDa ß-actin. (B) Gap-filling DNA synthesis activity assay. Nuclear extracts (50 µg) from 19.4{Delta}P cells, 19.4{Delta}P#53 cells, tumors #29 and #53 and 16.3 cells were tested for their ability to generate a 51 bp repaired DNA product. Lane MCF-7, nuclear extract of MCF-7 cells used as marker.

 
To test polß function, we measured gap-filling DNA synthesis activity in nuclear extracts from the 19.4{Delta}P#53 cell lines and tumor#29. Nuclear extracts from 16.3 cells were used as a positive control. A 51 bp repaired product was identified in all tested nuclear extracts (Figure 4, panel B). The intensity of the repaired product in tumor#29 was comparable with that generated by the 16.3 nuclear extract. This result suggests that the endogenous WTpolß protein expressed in tumor#29 provides sufficient catalytic activity to generate normal levels of DNA repair activity. In contrast, decreased gap-filling DNA synthesis activity was observed in the 19.4{Delta}P and 19.4{Delta}P#53 nuclear extracts. This is expected since these cells express only the mutant polß protein that has a deletion of 29 amino acid residues within the catalytic domain. Since we reported previously that MCF-7 breast cancer cells expressing 39 kDa WT polß protein exhibit efficient repair activity (34), we used a nuclear extract from MCF-7 cells as an additional control in this experiment. The data indicate that polß-mediated DNA repair is deficient in 19.4{Delta}P cells due to their expression of the deleted form of polß. This deficiency is maintained through the transformation process as demonstrated by the 19.4{Delta}P#53 cell line. Tumors derived from these cells, however, express endogenous as well as mutant polß and therefore demonstrate WT gap-filling DNA synthesis activity.

Generation of transgenic mice expressing mutant polß in the mammary glands
To direct expression of polß{Delta}208–236 in transgenic mice, we chose the WAP promoter (35–37). The advantage of using the WAP promoter is that it is highly expressed in the mammary glands of pregnant mice (35–37). Our strategy was to establish transgenic mice expressing both endogenous polß and polß{Delta} that would mimic expression of polß identified in human tumors (18–23). The transgenic construct is illustrated in Figure 5A. Digestion of the plasmid polß{Delta}/WAP3'/WAP5' by BssHII separated the 2.9 kb transgene from the 3 kb vector (Figure 5B) prior to microinjection. Sixteen founder pups were born and DNA prepared from tail clips was analyzed for the presence of the transgene using DNA PCR with a WAP+1 forward primer and a polß reverse primer. A 245 bp PCR product of the expected size was identified in two founder mice (Figure 5C, lanes 4 and 5). The PCR product was not detected in the remaining 14 animals (represented by two mice, lanes 2 and 3), indicating that the transgene was not transmitted to the DNA of these animals. DNA samples from all 16 mice were further analyzed by Southern blot by digestion with SpeI and HindIII and hybridization to a ~960 bp SpeI- and HindIII-digested polß{Delta} cDNA probe. Figure 5D shows a hybridizing fragment of ~960 bp in two transgenic mice shown in lanes 5 and 6 (shown by an arrow). DNA samples shown in lanes 4 and 5 (PCR) and lanes 5 and 6 (Southern) were from same positive mice, respectively. These results confirm that the polß{Delta} transgene was transmitted to two founder mice whereas the other 14 animals were non-transgenic. The Southern blot also shows fragments >2 kb that hybridized to the transgene probe. These fragments might be due to incomplete digestion of the genomic DNA. Lane 8 shows the ~960 bp probe used as a marker.


Figure 5
View larger version (74K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. (A) Schematic diagram of the 2.9 kb WAP-polß{Delta} transgene construct. (B) Gel electrophoresis of the polß{Delta}/WAP3'/WAP5' digested by BssHII. Lane T shows separation of the 2.9 kb transgene and 3 kb vector. Lane M shows the molecular size markers in kilobase. (C) PCR amplification of tail DNA to identify transgene sequences in founder mice. The arrow indicates the ~245 bp PCR product. Lane 1, molecular size marker. Lanes 2–5, DNA PCR products from tail clips of four pups. (D) Southern blot analysis of tail DNA from the founder mice. Lanes 1–7, DNA from animal #1–7. Lane 8 shows the ~960 bp transgene probe indicated by an arrow. (E) Southern blot analysis of tail DNA from F1 progeny of matings between transgenic founders and Balb/c mice. Lanes 1–18, DNA from animal# 1–18. The arrow indicates the 960 bp fragment hybridized to the transgene probe.

 
Balb/c mice have been reported to be susceptible to MNU-induced mammary tumorigenesis (38,39). Hence, the two positive founder animals were mated with normal Balb/c mice to establish a transgenic mouse line. Eighteen pups were born from these matings. Tail DNA from the F1 pups was examined by Southern blot to confirm that the transgene had been transmitted. As shown in Figure 5E, the 960 bp band indicating the presence of the transgene was observed in samples from pups #1–3, 5, 8, 10, 11 and 16–18. F1 animals #4, 6, 7, 9 and 12–15 were transgene negative. These results demonstrate that the WAP-polß{Delta} construct is not lethal during mouse embryogenesis.

To evaluate whether the WAP-polß{Delta} transgene was expressed in the mammary glands of transgenic mice, cell extracts were made from the mammary glands of two F1 mice (#1 and #2), 1 week after delivery of littermates as WAP is expressed when prolactin and progesterone are at high levels in serum. Figure 6A shows expression of both the transgene 36 kDa polß{Delta} protein and the endogenous 39 kDa polß protein in the mammary glands of both F1 mice (lanes 1 and 2). Positive controls included 39 kDa WTpolß protein expressed in the 16.3 cell line (lane 3) and a 39 kDa polß recombinant protein (lane 4) purified in our laboratory (29). Figure 6B shows that only WTpolß (39 kDa) and not the transgene-encoded deleted form is expressed in lung, liver, stomach, heart, kidney and spleen (lanes 1–6, respectively). These results strongly suggest that expression of the transgene is limited to the mammary glands, as expected based upon the known tissue specificity of the WAP promoter. Breeding to Balb/c mice was repeated six times to generate F6 transgenic mice that were at 98.2% of Balb/c background. Animals were maintained in our institution's animal facility strictly according to National Institutes of Health–AAALAC guidelines.


Figure 6
View larger version (32K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6. (A) Western blot analysis of cell extracts from the mammary glands of #1 and #2 transgenic F1 mice. Lanes 1 and 2 show that the 39 and 36 kDa proteins of polß are identified in both #1 and #2 transgenic mice. Lane 3 shows the 39 kDa polß protein expressed in the 16.3 cell line. Lane 4 shows purified recombinant human polß that was loaded as a positive control. (B) Western blot analysis of cell extracts of other tissues of F1 mouse #2. Lanes 1–6 show the 39 kDa polß protein expressed in the lung, liver, stomach, heart, kidney and spleen, respectively.

 
Mammary carcinogenesis in transgenic mice expressing the polß deletion mutant
In order to examine tumorigenesis in transgenic mice expressing polß{Delta} in the mammary glands, we first established the LD50 of MNU in these mice. Transgenic animals were given MNU at a dose of 25–200 mg/kg body wt intra-peritoneally once a week for 3 weeks (at 7 and 14 days of pregnancy and 1 week after litters were born). MNU at a dose of 100 mg/kg body wt was found to be lethal, establishing 50 mg/kg body wt as the LD50 (data not shown).

To compare the susceptibility of transgenic and WT mice to MNU-induced tumorigenesis, pregnant female transgenic OCF6 generation (~98% Balb/c background) and normal pregnant female Balb/c mice (of similar age and genetic background as transgenic animals) were treated on identical schedules with MNU at 50 mg/kg body wt. Intra-peritoneal injections were given for 7 weeks, starting at day 7 of pregnancy and continuing to the fourth week of the lactation period. Pups were kept with their mothers so that nursing would maintain high levels of prolactin in the serum important for expression of the WAP-driven polß{Delta} transgene. Mammary tumors were identified by palpation and were first detected in MNU-treated transgenic mice at 12 weeks after MNU injection. In contrast, tumors did not appear in similarly treated control mice until 28 weeks (Figure 7). Thus, transgene expression is correlated with an earlier appearance of MNU-induced mammary tumors.


Figure 7
View larger version (27K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7. Comparison of mammary tumorigenesis in control Balb/c and transgenic mice treated with MNU as described in Results. Data are expressed as the percentage of tumor-free mice. Tested groups included Balb/c control mice injected with PBS (diamond symbol, n = 9); Balb/c normal control mice treated with MNU (triangle symbol, n = 9); transgenic mice injected with PBS (square symbol, n = 7) and transgenic mice treated with MNU (circle symbol, n = 24).

 
Moreover, the percentage of transgenic mice that developed MNU-induced mammary tumors (70%) was significantly higher than for control mice (22%) (Table IV). Mammary tumors were identified at both upper right and left sides of upper and lower body of transgenic mice treated with MNU. All mammary tumors were identified by the Mouse Phenotyping Service, Ohio State University, Columbus, OH, as adenosquamous carcinomas (Figure 8, lower panel). The histology of a normal mammary gland from an untreated Balb/c mouse is shown in Figure 8, upper panel, used as a control. In addition to mammary tumors, MNU-treated transgenic mice also had enlarged spleens and bronchioalveolar carcinomas with intra-pulmonary metastases and adenomas in their lungs. Hemangiosarcoma in spleens and carcinomas in ovaries were found in transgenic mice treated with MNU. Interestingly, Balb/c control mice treated similarly did not develop any tumor in the mammary glands or any other organ until 28 months after treatment. Strikingly, on average, two to four adenomas were found in lungs of each of control mouse, whereas 10–20 adenomas were identified in lungs of each transgenic mouse. Neither transgenic nor control mice treated with PBS developed tumors as of 16 months after treatment. Taken together, these results indicate that expression of the polß{Delta} transgene in the mammary glands potentiates MNU-induced tumorigenesis as evidenced by a shortened latency period and increased tumor incidence.


Figure 8
View larger version (124K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 8. Hematoxylin and eosin-stained sections, x10 magnification. Upper panel, histopathology of a normal mammary gland from an untreated Balb/c control mouse. Lower panel, histopathology of a mammary adenosquamous carcinoma from a polß{Delta} transgenic mouse treated with 50 mg/kg MNU as described in Results.

 


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

  		Table IV. Occurrence of mammary adenosquamous carcinomas in transgenic mice

 

   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
DNA polß is essential for filling the one-nucleotide gap at apurinic–apyrimidine sites in damaged DNA via the short-patch BER pathway. In contrast, the long-patch BER pathway is largely dependent on polß in cell extracts, but can be reconstituted with pol{delta}, pol{epsilon}, proliferating cell nuclear antigen and flap endonuclease1 (1,2). A deletion variant of polß encoding a 36 kDa protein (polß{Delta}) with diminished DNA repair activity is expressed in various human cancers. The molecular details of how the deleted form of polß impacts DNA repair are not completely understood. The polß{Delta} mutant like WTpolß has been shown to interact with XRCC1, poly(ADP-ribose) polymerase, apurinic endonuclease and MGC5306 in vitro and in vivo (29,40). MGC5306 is a recently identified novel protein that binds not only to polß{Delta} but also to WTpolß (40). Although the purified polß{Delta} protein is capable of efficiently repairing apurinic–apyrimidine sites, a presumed binary complex of polß{Delta} and XRCC1 is incapable of filling these gaps. This complex binds more strongly to a gapped DNA strand than a complex of WTpolß and XRCC1. Thus, polß{Delta} bound to XRCC1 inhibits functions of the WT polß in a dominant-negative fashion (29).

To investigate the biological consequences of expression of the dominant-negative deletion mutant of polß, we used two animal models: nude and transgenic mice. Results from both studies demonstrated that cells expressing the polß mutant protein displayed increased sensitivity to MNU, an environmentally related DNA-damaging agent, which resulted in enhanced tumorigenesis in vivo. When exposed to MNU, MEF cells expressing only the mutant form of polß protein exhibited altered morphology. Cells further selected for anchorage-independent growth in soft agar induced multiple tumors in 100% of injected nude mice. Two other polß variants identified in colorectal and prostate tumors (18,24) were expressed in murine mammary tumor cells and found to induce focus-forming transformation and anchorage-independent growth (41).

The effect of polß{Delta} on tumor formation was also evaluated in a transgenic mouse model of MNU-induced mammary carcinogenesis. For this purpose, we created a transgene construct in which polß{Delta}208–236 expression is controlled by the strong mammalian tissue-specific promoter, the WAP promoter (35–37). The construct also uses an 843 bp of 3' WAP sequence that contains a 70 bp untranslated region including a poly A region. These sequences have been shown to determine tissue specificity and integration site, independent of transgene expression (35,42). The advantage of using the WAP promoter for our study is that it is highly expressed in mammary glands of pregnant transgenic mice (35–37). The prolactin hormone secreted during pregnancy stimulates the ovaries to secrete progesterone and estradiol. This combination of hormones in pregnant mice causes proliferation of mammary epithelial cells that enhances chemical- or oncogene-mediated mammary tumorigenesis (35–37,39,42). In addition, a number of transgene constructs driven by the WAP promoter have been successfully used to study mammary carcinogenesis (36,37). The data presented in this study demonstrate that the polß{Delta}208–236 transgene was effectively expressed in the target organ of the WAP-driven polß{Delta} transgene, the mammary glands. Since the transgenic mice express endogenous WTpolß in addition to the transgene-encoded mutant polß, this transgenic model mimics the observed expression of polß (WT and mutant) in many human tumors when normal control mouse expressing WTpolß mimics normal matched tissues (20,22,23). Our analysis of control and transgenic pregnant mice exposed to MNU showed that transgene expression resulted in a higher incidence of tumor formation, with tumor forming in multiple mammary glands of each animal. Furthermore, the latent period prior to tumor appearance was significantly shorter in transgenic animals than in control mice.

Taken together, these accumulated results unequivocally demonstrate that expression of the polß{Delta} protein enhances acquisition of neoplastic phenotype following exposure of cells to a DNA-damaging agent. Furthermore, since the polß{Delta} protein has this effect despite expression of WT endogenous polß in the same cells, these data support the hypothesis that polß{Delta} acts as a dominant-negative mutant. This transgenic model expressing a mutant repair gene will be useful for future studies at improving our understanding of the mechanisms underlying human oncogenesis.

Based upon the impaired DNA repair activity of the polß{Delta} mutant protein, it is tempting to speculate that expression of polß{Delta} might lead to persistence of unrepaired MNU-induced DNA lesions, resulting in mutagenesis that might lead to cellular transformation. In this way, polß{Delta} expression might have significant effects on the development of human cancers.


   Footnotes
 
{dagger} These authors contribute equally to this work. Back


   Acknowledgments
 
This study was supported by the National Institutes of Health grant RO1 CA83768 (S.B.). The authors gratefully acknowledge Dr Jeffrey M.Rosen, Baylor College of Medicine, Houston, TX, USA, for the 5' WAP promoter and 3' WAP promoter and for many helpful suggestions. We thank Dr Samuel H.Wilson, National Institute of Environmental Health Sciences, National Institutes of Health for the kind gift of 19.4 cell line and Dr Patricia Stanhope-Baker, Department of Cancer Biology, Cleveland Clinic for excellent editorial assistance.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Wilson SH. Mammalian base excision repair and DNA polymerase beta. Mutat. Res. (1998) 407:203–215.[ISI][Medline]
  2. Memisogmson L. Base excision repair in yeast and mammals. Mutat. Res. (2000) 451:39–51.[ISI][Medline]
  3. Feig DI, et al. Mechanisms of mutation by oxidative DNA damage: reduced fidelity of mammalian DNA polymerase ß. Biochemistry (1993) 32:4466–4473.[CrossRef][Medline]
  4. Dianov GL, et al. Single nucleotide patch base excision repair is the major pathway for removal of thymine glycol from DNA in human cell extracts. J. Biol. Chem. (2000) 275:11809–11813.[Abstract/Free Full Text]
  5. Pascucci B, et al. Reconstitution of the base excision repair pathway for 7, 8-dihydro-8-oxoguanine with purified human proteins. Nucleic Acids Res. (2002) 30:2124–2130.[Abstract/Free Full Text]
  6. Sobol RW, et al. Requirement of mammalian DNA polymerase ß in base-excision repair. Nature (1996) 379:183–186.[CrossRef][Medline]
  7. Matsumoto Y, et al. Excision of deoxyribose phosphate residues by DNA polymerase ß during DNA repair. Science (1995) 269:699–702.[Abstract/Free Full Text]
  8. Prasad R, et al. Human DNA polymerase ß deoxyribose phosphate lyase. Substrate specificity and catalytic mechanism. J. Biol. Chem. (1998) 273:15263–15270.[Abstract/Free Full Text]
  9. Allison SL, et al. DNA polymerase ß is the major dRP lyase involved in repair of oxidative base lesions in DNA by mammalian cell extract. EMBO J (2001) 20:6990–16926.[CrossRef][ISI][Medline]
  10. Sweasy JB, et al. Detection and characterization of mammalian DNA polymerase ß mutants by functional complementation in Escherichia coli. Proc. Natl Acad. Sci. USA (1993) 90:4626–4630.[Abstract/Free Full Text]
  11. Gu H, et al. Deletion of a DNA polymerase ß gene segment in T cells using cell type-specific gene targeting. Science (1994) 265:103–106.[Abstract/Free Full Text]
  12. Sugo N, et al. Neonatal lethality with abnormal neurogenesis in mice deficient in DNA polymerase beta. EMBO J (2000) 19:1397–1404.[CrossRef][ISI][Medline]
  13. Plug AW, et al. Evidence for a role for DNA polymerase ß in mammalian meiosis. Proc. Natl Acad. Sci. USA (1970) 94:1327–1331.[CrossRef]
  14. Ochs K, et al. Cells deficient in DNA polymerase beta are hypersensitive to alkylating agent-induced apoptosis and chromosomal breakage. Cancer Res. (1999) 59:1544–1551.[Abstract/Free Full Text]
  15. Ali-Osman F, et al. Enhanced repair of a cisplatin-damaged reporter chloramphenicol-O-acetyltransferase gene and altered activities of DNA polymerases {alpha} and ß, and DNA ligase in cells of a human malignant glioma following in vivo cisplatin therapy. J. Cell. Biochem. (1994) 54:11–19.[CrossRef][ISI][Medline]
  16. Fotiadou P, et al. DNA polymerase beta interacts with TRF2 and induces telomere dysfunction in a murine mammary cell line. Cancer Res. (2004) 64:3830–3837.[Abstract/Free Full Text]
  17. Lang T, et al. A DNA polymerase beta mutant from colon cancer cells induces mutations. Proc. Natl Acad. Sci. USA (2004) 101:6074–6079.[Abstract/Free Full Text]
  18. Wang L, et al. DNA polymerase ß mutations in human colorectal cancers. Cancer Res. (1992) 52:4824–4827.[Abstract/Free Full Text]
  19. Wang L, et al. Mutations in DNA polymerase ß occur in breast, prostate and colorectal cancers. Int. J. Oncol. (1995) 6:459–463.[ISI]
  20. Bhattacharyya N, et al. Alteration of hMSH2 and DNA polymerase ß genes in breast carcinomas and fibroadenomas. Biochem. Biphys. Res. Commun. (1999) 259:429–435.[CrossRef][ISI][Medline]
  21. Bhattacharyya N, et al. Variant forms of DNA polß in primary lung carcinomas. DNA Cell Biol. (1999) 18:549–554.[CrossRef][ISI][Medline]
  22. Chen HC, et al. Defective repair genes in a primary culture of human renal cell carcinoma. J. Cancer Res. Clin. Oncol. (2000) 126:185–190.[CrossRef][ISI][Medline]
  23. Bhattacharyya N, et al. A variant of DNA polymerase ß acts as a dominant negative mutant. Proc. Natl Acad. Sci. USA (1997) 94:10324–10329.[Abstract/Free Full Text]
  24. Dobashi Y, et al. DNA polymerase ß gene mutation in human prostate cancer. Cancer Res. (1994) 54:2827–2829.[Abstract/Free Full Text]
  25. Matsuzaki J, et al. DNA polymerase ß gene mutations in human bladder cancer. Mol. Carcinog (1996) 15:38–43.[CrossRef][ISI][Medline]
  26. Iwanaga A, et al. Functional mutation of DNA polymerase beta found in human gastric cancer--inability of the base excision repair in vitro. Mutat. Res. (1999) 435:121–128.[ISI][Medline]
  27. Bhattacharyya N, et al. Mutations occur in the genomic sequence of the DNA polymerase ß of ovarian cancer. Proc. Am. Assoc. Cancer Res. (2003) 44:261.
  28. Bhattacharyya N, et al. Impaired repair activity of a truncated DNA polymerase ß protein. Life Sci. (2001) 69:271–280.[CrossRef][ISI][Medline]
  29. Bhattacharyya N, et al. A novel role of XRCC1 in the functions of a DNA polymerase ß variant. Biochemistry (2001) 40:9005–9013.[CrossRef][Medline]
  30. Banerjee S, et al. Neoplastic activity and chromosomal alterations of cells expressing a DNA polymerase beta mutant. Proc. Am. Assoc. Cancer Res. (2005) 46. (abstract no.1779).
  31. Banerjee S, et al. In vitro transformation of C3H/10T1/2 and NIH/3T3 cells by acrylonitrile and acrylamide. Cancer Lett. (1986) 32:293–304.[CrossRef][ISI][Medline]
  32. Schultz K, et al. Neoplastic expression in murine cells induced by halogenated hydrocarbons. In Vitro Cell. Dev. Biol. (1992) 28A:267–272.[CrossRef]
  33. Bergoglio V, et al. Evidence of finely tuned expression of DNA polymerase ß in vivo using transgenic mice. FEBS Lett (2004) 566:147–150.[CrossRef][ISI][Medline]
  34. Bhattacharyya N, et al. Heterogeneity in expression of DNA polymerase ß and DNA repair activity in human tumor cell lines. Gene Expr. (2002) 10:115–123.[ISI][Medline]
  35. Bayna E, et al. Tissue-specific, high-level expression of the rat whey acidic protein gene in transgenic mice. Nucleic Acid Res. (1990) 18:2977–2985.[Abstract/Free Full Text]
  36. Li B, et al. A transgenic mouse model for mammary carcinogenesis. Oncogene (1998) 16:997–1007.[CrossRef][ISI][Medline]
  37. Murphy KL, et al. Bcl-2 expression delays mammary tumor development in dimethylbenz(a)anthracene-treated transgenic mice. Oncogene (1999) 18:6597–6604.[CrossRef][ISI][Medline]
  38. Medina D. Mammary tumorigenesis in chemical carcinogen-treated mice. I. Incidence in BALB/c and C57Bl mice. J. Natl Cancer Inst. (1974) 53:213–221.[ISI][Medline]
  39. Swanson SM, et al. Pituitary-isografted mice are highly susceptible to MNU-induced mammary carcinogenesis irrespective of the level of alveolar differentiation. Carcinogenesis (1994) 15:1341–1346.[Abstract/Free Full Text]
  40. Wang L, et al. A novel nuclear protein, MGC5306 interacts with DNA polymerase ß and has a potential role in cellular phenotype. Cancer Res. (2004) 64:7673–7677.[Abstract/Free Full Text]
  41. Sweasy JB, et al. Expression of DNA polymerase ß cancer-associated variants in mouse cells results in cellular transformation. Proc. Natl Acad. Sci. USA (2005) 102:14350–14355.[Abstract/Free Full Text]
  42. Hennighausen L, et al. Whey acidic protein extrinsically expressed from the mouse mammary tumor virus long terminal repeat results in hyperplasia of the coagulation gland epithelium and impaired mammary development. Cell Growth Differ (1994) 5:607–613.[Abstract]
Received April 26, 2006; revised November 10, 2006; accepted November 25, 2006.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Material
Right arrow All Versions of this Article:
28/6/1356    most recent
bgl239v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Wang, L.