|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||









* Center for Animal Transgenesis and Germ Cell Research, Department of Clinical Sciences, New Bolton Center, and
Department of Animal Biology, School of Veterinary Medicine, University of Pennsylvania, Kennett Square, Pennsylvania, USA;
GTC Biotherapeutics, Framingham, Massachusetts; and
Tufts Cummings School of Veterinary Medicine, North Grafton, Massachusetts, USA
3 Correspondence: Center for Animal Transgenesis and Germ Cell Research, University of Pennsylvania, 382 W. State Road, Kennett Square, PA 19348, USA. E-mail: dobrinsk{at}vet.upenn.edu
| ABSTRACT |
|---|
|
|
|---|
Key Words: transgenesis testis mouse goat
| INTRODUCTION |
|---|
|
|
|---|
The male germ line stem cell is the only cell in the adult body that both mitotically divides and passes its genetic material on to the next generation, making it an attractive target for genetic manipulation. In 1994, Brinster and colleagues reported that transplantation of mouse male germ line stem cells to the testis of a recipient animal results in donor derived spermatogenesis and transmission of the donor haplotype to the offspring of the recipient (11
, 12)
. More recently, we demonstrated that germ cell transplantation is also successful in goats (13
, 14)
. Virus-mediated transduction and subsequent transplantation of male germ line stem cells to recipient animals has been employed to generate transgenic mice and rats (15
16
17
18
19)
. Recently, a combination of germ cell culture and transplantation has been reported that allowed gene knockout and targeting in the mouse (20)
. Introduction of transgenesis through manipulation of male germ cells is a particularly valuable approach in species like rats and domestic animals in which embryonic stem cell technology is not available. Currently, somatic cell nuclear transfer or pronuclear injection is used to generate transgenic domestic animals. These methods are inefficient and in some species carry a high risk of developmental abnormalities in the few resulting offspring (21)
. Sperm-mediated transgenesis has been demonstrated for pigs (22)
but has not been easily reproducible across species.
Establishment of long-term cultures of mouse and rat germ line stem cells has been reported (23
, 24)
. However, primary male germ line stem cells are quiescent or at best proliferate very slowly in culture, making the use of standard transfection methods difficult, if not impossible. Transgenesis via transplantation of genetically altered male germ cells in mice and rats has used retroviral or lentiviral transduction of germ cells prior to transplantation (15
16
17
18
19)
. These strategies have resulted in germ line transgenesis, but efficiency was moderate and the use of retroviral or lentiviral vectors requires that animals are handled and maintained under higher biosafety precautions [biosafety level (BSL) 2]that render this approach less practical for transgenesis in large animal species. In contrast, animals exposed to AAV can be maintained under standard husbandry conditions (BSL 1).
The objective of the current study was to explore whether AAV could stably transduce male germ line stem cells and lead to transgene transmission through the male germ line. Initially, we used the established germ cell transplantation model in the mouse to investigate whether AAV-mediated transduction of germ cells was possible and could result in germ line transgene transmission. To broaden the applicability of the results for different mammalian species, the approach was then applied to a large animal species in which germ cell transplantation-mediated transgenesis would provide an important alternate approach to the generation of transgenic animal models for biomedical research.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Experiments in the mouse model
Experiment 1: AAV-mediated mouse germ cell transduction in situ
In experiment 1, 16 6-wk-old B6C3 male mice were used as recipients. In 15 animals, 1 to 25 x 108 genomic count (gc) of AAV2-CMV-eGFP in 20 µl was injected into the seminiferous tubules via the efferent ducts as described previously (25)
. One animal served as an untreated control. Animals were sacrificed between 1 wk and 8 months after injection, and the testes were analyzed by fluorescence microscopy. Nine animals were mated to wild-type females beginning 1 month after injection, and the resulting pups were examined with UV goggles for green fluorescence.
Experiment 2: AAV-mediated mouse germ cell transduction and germ cell transplantation
Experiment 2 consisted of three different experiments exploring different cell numbers and vector concentrations (Table 1
). In all experiments, donor cells were obtained from male mice in which experimental cryptorchidism was induced at 6 wk of age (26)
, and cells were collected 2 months later by enzymatic digestion as described previously (25)
. Collection of germ cells from cryptorchid testes results in a relative enrichment of germ line stem cells (27)
. In all experiments, strain-matched recipient animals were treated with 50 mg/kg busulfan intraperitoneally at 6–8 wk of age to deplete endogenous germ cells and were used as recipients at least 4 wk after treatment (11
, 12)
.
|
Isolated testis cells were incubated with viral vectors for 2–4 h, washed 3 times, and transplanted to recipient testes by microinjection into the seminiferous tubules through the efferent ducts as described previously (25)
. On average, 10 µl of cell suspension can be introduced into each testis. Some recipient males were mated to wild-type females. Pups were analyzed with UV goggles for overt fluorescence, and fibroblasts were obtained from biopsies of 6 animals positive for GFP by PCR, grown in culture and analyzed for GFP expression by fluorescence microscopy and Western blot analysis. Genomic DNA from pups was analyzed by PCR and Southern blot analysis, and protein was analyzed by Western blot analysis. Three male and three female transgenic F1 pups were mated, and the resulting F2 pups genotyped by PCR. Recipient testes were analyzed at 6–12 mo of age by fluorescence microscopy.
In experiment 2.3, recipient animals were sacrificed at 10 days (n=2) or at 6 months (n=3) after transplantation, and tissue was collected from liver, lung, kidney, brain, testis, diaphragm, spleen, and heart for DNA isolation and PCR analysis.
Experiments in the goat model
Experiment 3: AAV-mediated goat germ cell transduction and transplantation
Four male dairy goat kids were subjected to fractionated testicular irradiation of 3 x 2 Gy at 4 wk of age, as described previously (28)
. Donor cells were collected from testes obtained from 8- to 11-wk-old dairy goats (n=4) by two-step enzymatic digestion as described previously (13, 14). Cells were exposed to AAV2-CMV-eGFP (1 recipient testis), AAV2-PGK2-eGFP (2 recipient testes), or AAV2-ACR3-eGFP (4 recipient testes) at 2000–12,000 gc/cell and 100–500 x 106 cells/testis were transplanted to recipient testes by ultrasound-guided cannulation of the rete testis, as described previously (13)
, when the goats were 4 months of age. Testes from one goat were analyzed by fluorescence microscopy 3 months after transplantation. Starting 5 months after transplantation, the remaining goats were trained for semen collection, and semen was collected once a week for 18 months using an artificial vagina in the presence of a female goat. Sperm were processed for DNA extraction, and sperm DNA was analyzed by PCR. One goat was sacrificed at 13 months after transplantation, and the testes were analyzed for transgene expression by fluorescence microscopy. Two ejaculates from each of the remaining goats were extended in Tris-based extender containing 20% egg yolk and 7% glycerol, shipped cooled to GTC Biotherapeutics (Framingham, MA, USA), and used for IVF. IVF was performed as described previously (29)
, using a total of 687 in vivo or in vitro matured goat oocytes. The resulting embryos were cultured for 4 to 7 days and analyzed by single-embryo PCR.
All experiments were approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania and performed in accordance with relevant guidelines and regulations.
Analysis of tissue and sperm samples for the presence and expression of the transgene
Fluorescence microscopy
Testes were recovered from recipient mice and goats and analyzed for the presence of fluorescent cells in the seminiferous tubules under a dissecting microscope equipped for epifluorescence analysis. Seminiferous tubules were subsequently isolated by mechanical dispersion of the testes and mounted onto microscope slides for analysis at higher magnification. Images were captured using Image Pro software (Media Cybernetics, Santa Barbara, CA, USA).
Fibroblast culture
Fibroblasts were grown from skin samples of 6 transgenic F1 generation pups. Cells were incubated at 37°C with 5% CO2 in Dulbeccos Modified Eagle Medium (DMEM) with 10% FBS (Hyclone, Logan, UT), 100 U/ml penicillin and 100 µg/ml streptomycin, harvested by digestion with trypsin/EDTA (Invitrogen, Carlsbad, CA, USA), passaged, and maintained in the above media until further analysis. Fluorescence microscopy images were captured as above.
PCR of sperm and tissue samples
DNA was extracted from mouse tissue and goat sperm using a commercially available kit (QiAmp DNA Mini kit, Qiagen Science, Valencia, CA, USA). For the isolation of sperm DNA, 107 sperm/ejaculate were incubated for 1 h at 37°C in the presence of 100 mM dithiothreitol (DTT) in lysis buffer prior to DNA extraction. All samples were analyzed in triplicate. For PCR detection of the eGFP gene, the following primers were used: 5'-TCA CCT TGA TGC CGT TCT TCT-3' and 5'-GCA AGC TGA CCC TGA AGT TCA-3', resulting in a 372-bp fragment, or 5'-CGG CCA CAA GTT CAG CGT GTC CGG CG-3' and 5'-CCA TGT GAT CGC GCT TCT CGT TGG GG-3', resulting in a 575-bp fragment. As control, primers were used for the endogenous goat β-casein exon 7: 5'-CCA GGC ACA GTC TCT AGT CTA-3' and 5'GGA CAG GAC CAA GTA CAG CT-3', resulting in a 440-bp fragment; a 983-bp fragment of glycerol-3-phosphate dehydrogenase (GAPDH) was amplified as a control for all mouse samples using the primers provided in RT-for-PCR Kit (Clontech Laboratories, Mountain View, CA, USA). Positive control DNA was from GFP-transgenic mice (Tg(ACTB-EGFP)D4Nagy/J; Jackson Laboratory, Bar Harbor, ME, USA). Negative control DNA was from obtained from wild-type mouse tissue or wild-type goat sperm.
Southern blot analysis
Eight micrograms of genomic DNA from liver from each sample was digested with BamHI and NotI at 37°C overnight. This releases a 700-bp internal fragment of the GFP coding region. Digested DNA was separated by electrophoresis in a 1% agarose gel and blotted to a GeneScreen Plus nylon membrane (PerkinElmer, San Diego, CA, USA) in 20 x SSC. A 575-bp fragment from the GFP coding region was amplified by PCR, as described above and used as a probe. Probes were labeled by using Ready-To-Go DNA Labeling Beads kit (Amersham Biosciences, Piscataway, NJ, USA). The membrane was hybridized with 32P-labeled GFP DNA probe and the hybridized blot was exposed to HyBlot CLTM film (Denville Scientific, Plainfield, NJ, USA).
Western blot analysis
Protein was extracted from liver tissue. Negative control tissues were from wild-type mice, and positive control tissues were from a GFP-transgenic mouse (Tg(ACTB-EGFP)D4Nagy/J; Jackson Laboratory). Proteins were separated on 12% SDS-PAGE and transferred to PVDF membranes (Millipore, Billerica, MA, USA). Membranes were blocked with 5% cold water fish gelatin (Sigma, St. Louis, MO, USA) in 0.01% Tween Tris-buffered saline (TTBS) and incubated for 1 h in monoclonal antieGFP antibody (1:20,000; BD Biosciences, Franklin Lakes, NJ, USA), washed thoroughly, and incubated in goat anti-mouse IgG peroxidase conjugated secondary antibody (CalBiochem, Burlingame, CA, USA) at 1:5,000 for 30 min. Immunodetection of the eGFP protein was completed with the GE Amersham Enhanced Chemiluminescence Detection (ECL) Kit.
PCR analysis of goat embryos
Goat embryos were generated by IVF using cryopreserved semen from goats 07 and 17. Embryos were analyzed at days 4 to 7, with the majority analyzed at 7 days, and contained from 1 to
132 cells. Negative control samples were 7-day-old ionomycin-activated parthenogenic embryos and nontransgenic genomic goat DNA. The positive GFP control was genomic DNA from a GFP-transgenic mouse (57°C; Dr. Yuko Fujiwara, Childrens Hospital, Boston, MA, USA).
Groups of 5 to 11 embryos were placed in acid Tyrodes solution (30)
at 37°C for
1 min followed by treatment with 5 µg/ml pronase (Sigma) and 1.65 µg/ml polyvinylpyrrolidone-10 (PVP) (Sigma) in PBS until the zonae had visibly dissolved. Embryos were washed individually 3 times in 0.66 mg/ml PVP in HBSS, and single embryos were placed with 3–5 µl of the wash media into GeneAmp® PCR tubes containing 25 µl dH2O (Ultra Pure, Life Technologies, Carlsbad, CA, USA). The borosilicate glass pipette was rinsed in fresh HBSS + PVP between embryos and changed between groups. Tubes were placed on dry ice and then stored at –80°C.
Embryos were lysed using modifications of a method reported by Xu et al. (31)
, whereby embryos were thawed, refrozen on dry ice, rethawed and heated at 95°C for 10 min. Samples were then cooled to 4°C and placed on ice. Negative goat and positive mouse control genomic DNA samples were serially diluted in dH2O to 10, 5, and 2 cell equivalents of DNA per microliter. To increase the amount of DNA available for testing, the genomic DNA present in the lysed embryos and controls was subjected to whole genome amplification by multiple displacement amplification (MDA) (32)
using Phi29 DNA Polymerase (New England BioLabs, Ipswitch, MA, USA), and an exonuclease resistant thiophosphate modified random hexamer, 5'-NNNN*N*[N
Q]-3' (Operon, Alameda, CA, USA).
PCR for endogenous goat β-casein exon 7 (GEX7) and endogenous mouse β-globin served to confirm that amplifiable DNA was present in the post-MDA samples. Material from the first GFP PCR was then used in a nested PCR to obtain a more visible band on the gel. The primers are listed in Table 2
, and amplification conditions in Table 3
. All embryos in this study were tested both for the presence of the GFP transgene, as well as for an endogenous goat β-casein gene. Embryos that did not produce any bands, including the endogenous GEX7 sequences (10.5%), after two attempts were considered devoid of amplifiable DNA. Embryos were scored as inconclusive (12%) if the PCR gave at least one positive result for the GFP transgene, but the results could not be confirmed. Embryos were scored positive (9.8%) if at least two separate attempts yielded a positive GFP result.
|
|
| RESULTS |
|---|
|
|
|---|
Experiment 2: AAV-mediated transduction of germ cells prior to transplantation in the mouse
As direct introduction of the vector into the seminiferous tubules did not lead to efficient germ cell transduction in experiment 1, experiment 2 tested direct exposure of germ cells to different concentrations of viral vectors in vitro prior to transplantation. Germ cells were isolated from experimentally induced cryptorchid donor testes to increase the relative percentage of undifferentiated germ cells (27)
, exposed to the AAV2-CMV-eGFP vector in vitro and transplanted to recipient animal testes depleted of endogenous germ cells by busulfan treatment (see Table 1
for experimental design). Testes were analyzed for fluorescent germ cells at different time points. Analysis of recipient animal testes showed extensive colonization by germ cells expressing the GFP transgene (Fig. 1
a, b).
|
Males were mated to wild-type females, and the resulting pups were analyzed for transgene transmission by PCR and Southern blot analysis. Transgene expression in pups was monitored by Western blot analysis and by observation of fluorescence in whole animals and isolated fibroblasts. Four of 10 animals that received donor cells at a concentration of at least 100 x 106 cells/ml (experiment 2.2) subsequently became fertile. Pups did not show overt green fluorescence, but 26 of 260 pups analyzed carried the GFP transgene as detected by PCR genotyping. In addition, the GFP transgene was detected by Southern blot analysis in 4 of 11 pups analyzed (Fig. 2
). Fibroblast cultures obtained from 6 transgenic pups all showed green fluorescence in vitro, whereas fibroblasts from wild-type control animals did not exhibit fluorescence (Fig. 3
). Expression of GFP protein was also documented in 6 of 7 additional pups analyzed by Western blot (Fig. 4
).
|
|
|
Six transgenic pups were reared to adulthood (designated F1 animals), mated to each other in 3 breeding pairs, and resulting pups were analyzed by PCR and Western blot. One pair produced only one litter of 5 pups that were negative for the GFP transgene by PCR. One pair produced 4 litters with 10 of 33 pups positive for GFP by PCR, and one pair had 6 litters where 17 of 48 pups were GFP positive by PCR. While overt fluorescence was not observed, Western blot analysis of F2 pups demonstrated transgene expression (Fig. 5
).
|
Finally, analysis by PCR of several tissues from additional recipient animals at 10 days and 6 months after transplantation of transfected germ cells (experiment 2.3) revealed the presence of the transgene only in testis tissue (Fig. 6
).
|
Experiments in the goat model
Experiment 3: AAV-mediated goat germ cell transduction prior to transplantation
Germ cells isolated from prepubertal goats were exposed to AAV vectors carrying the eGFP reporter gene under the control of the CMV, PGK2 (33)
or ACR3 (34)
promoter and were transplanted to prepubertal recipient goats in which endogenous spermatogenesis had been depleted by testicular irradiation. Semen was collected from three recipient goats allowed to reach sexual maturity and analyzed by PCR for the presence of the transgene in sperm. Sperm carrying the GFP transgene were detected by PCR (Fig. 7
) in 37% and 35% of ejaculates collected over an 18-month period from two of the goats (n=59 and 54 ejaculates, respectively). One goat did not adapt well to semen collection and produced only two ejaculates that were both positive for the GFP transgene on PCR; it was sacrificed 13 months after transplantation. Colonies of green fluorescent cells (Fig. 1c, d
) could be detected in all goat testes analyzed at 3, 13, 22, and 24 months after transplantation of germ cells exposed to AAV2 vectors carrying the ACR3-eGFP, CMV-eGFP, or PGK2-eGFP constructs.
|
Semen collected from 2 recipient goats (goats 17 and 07) was used for IVF of wild-type goat oocytes, and transmission of the transgene was monitored by single embryo PCR analysis. A total of 155 embryos (8-cell stage to blastocyst) from recipient goat 17 and 121 embryos from recipient goat 07 were analyzed by single embryo PCR (Fig. 8
). Fifteen of 155 embryos from goat 17 were transgenic, 100 embryos were negative for GFP by PCR, the results for 20 embryos were inconclusive, and 20 embryos did not contain sufficient DNA for analysis. From goat 07, 12 of 121 embryos were transgenic, 87 were negative, 13 were inconclusive, and 9 had insufficient DNA.
|
| DISCUSSION |
|---|
|
|
|---|
In the current study, direct infusion of the viral vector into mouse seminiferous tubules resulted in transgene expression largely in Sertoli cells and was therefore not further pursued. The same phenomenon was also observed after direct injection of retroviral or adenoviral vectors into mouse seminiferous tubules (15
, 35)
. Therefore, in an alternate approach, isolated testis cells were exposed in vitro to the viral vector followed directly by transplantation to recipient testes. The efficiency of recipient testis colonization by donor germ cells in the mouse has been estimated at
10% (36
, 37)
. Although the recipients had been treated to deplete endogenous germ cells prior to transplantation, spermatogenesis does recover from both the remaining endogenous germ cells and the donor germ cells. Therefore, even if all transplanted germ cells had been successfully transduced, not all offspring resulting from mating between recipient males and wild-type females can be expected to be transgenic. The number of transgenic pups detected in the current experiments indicated that roughly 10% of sperm resulted from transduced donor cells while the remaining pups were sired by sperm resulting either from endogenous germ cells or from nontransduced donor cells.
When heterozygous transgenic F1 mice were mated to each other, 33% of pups analyzed were transgenic, providing further evidence for germ line transmission. However, this transmission rate is lower than expected for mating of heterozygous animals. From the current experiment, it cannot be ruled out that in some F1 animals the transgene was episomal rather than stably integrated (38)
. Dilution of episomal DNA during transmission to the F2 generation could account for the lower than expected rate of F2 transgenic pups.
While AAV-mediated transduction has the potential for episomal transmission as well as stable integration into the genome (5
, 38
, 39)
, transmission of the transgene through many cell divisions required to form sperm from a self-renewing population of germ line stem cells, for longer than a year, and to subsequent generations make it unlikely that the transgene remained episomal in the present study but rather indicates that integration into the genome had occurred.
Similar to the results obtained in F1 mice resulting from mating recipient males to wild-type females, the percentage of F1 transgenic goat embryos resulting from fertilization of wild-type oocytes with recipient sperm also indicated that at least 10% of sperm were derived from transduced donor germ cells. This rate of transgenesis is similar to that observed after transplantation of germ cells from transgenic donor goats (14)
, suggesting that the majority of the donor germ line stem cells had been successfully transfected. The results also compare favorably with previous reports of
5% transgenic offspring after transplantation of retrovirus- or lentivirus-transduced germ cells in rodents (16
, 17
, 40)
but are lower than the 30% transgenic pups recently reported from lentiviral transduction of rat germ cells (24)
.
The ability of the AAV vector to be incorporated into the male germ line might be a cause for concern for its use in human gene therapy (41)
; however, the viral load that the isolated germ cells were exposed to in the current study was certainly higher than what could be expected from somatic cell gene therapy applications. We demonstrated the presence of the transgene only in the testis after transplantation of germ cells exposed in vitro. Therefore, this approach is unlikely to generate unwanted ectopic transgene expression in recipient animals.
Similar to reports of GFP expression driven by a CMV promoter in germ cells of transgenic mice and rats (19
, 42)
, we observed colonies of fluorescent germ cells in recipient mouse and goat testes. However, although the presence of the transgene in mouse offspring and goat embryos produced using sperm from recipient males was conclusively demonstrated, we did not observe overt green fluorescence in mouse pups as an indication of GFP transgene expression. Expression of transgenes from the CMV promoter is widely variable between tissues (43)
, and silencing of CMV promoter-driven transgenes has been reported (19
, 44)
and could have contributed to the observed results. Gene silencing has also been identified as a problem of AAV-mediated transduction (45)
.
Although overt fluorescence was not detected, transgene expression was demonstrated by the observation of fluorescence in fibroblasts collected from F1 pups and the detection of GFP protein by Western blot analysis. This indicates that gene expression indeed occurred, albeit perhaps at a low level. To address potential problems with expression of the CMV promoter, GFP transgenes under the control of the mouse acrosin promoter, previously shown to drive gene expression in the acrosome, (34)
or the mouse PGK2 promoter (testis-specific isoform of PGK, ref. 33
), were employed in experiment 3. Faint green fluorescence could be detected in a small number of sperm from 2 recipient goats that received transplantations of germ cells exposed to the AAV2-ACR3-eGFP vector. Similar fluorescence was never observed in wild-type goat sperm or in sperm from recipients of germ cells transduced with different vectors.
Compared to lentiviral vectors, AAV vectors can only accommodate a smaller insert size (1)
, which might limit their utility to transgenes that are within this size range. However, strategies have been reported to overcome this limitation (46)
. In addition, AAV-mediated transgenesis has also been used successfully for gene targeting in human and bovine fibroblasts (47)
. Therefore, the promiscuity (4)
and the superior biosafety rating of the AAV vector make it a more appropriate choice than, for example, lentiviral vectors for animals that cannot be easily maintained under tight biosafety constraints.
Although germ cell transfection and transplantation to recipient testes have been shown to result in transgenic offspring in rodents, this is the first report of successful germ line transduction in a nonrodent animal. Introduction of a genetic modification prior to spermatogenesis will circumvent problems associated with manipulation of gametes and early embryos and developmental abnormalities associated with nuclear reprogramming. Even if embryonic stem cell technology becomes available for domestic animals, the time required before transgenic sperm can be harvested will be significantly shorter using germ cell transplantation (48)
. In addition, introduction of a genetic modification before meiosis allows recombination to occur. A recipient animal therefore can produce sperm with different transgene integrations that may permit screening of offspring to select the most desirable genotype.
In summary, introduction of genetic modifications through the male germ line by transduction of germ cells followed by transplantation to a recipient testis provides a promising strategy for the generation of transgenic animals, especially in species in which embryonic stem cell technology is not available. The use of an AAV2-mediated transduction strategy resulted in efficient transduction of mammalian germ cells and transgene transmission by recipient sperm. This study demonstrates for the first time that this approach is feasible both in mice and goats, a domestic animal species used for the efficient production of recombinant biopharmaceutical proteins in the milk of transgenic animals (49)
.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Current address: Department of Veterinary Biomedical Sciences, Western College of Veterinary Medicine, University of Saskatchewan, Saskatoon, SK, Canada S7N 5B4. ![]()
Received for publication May 10, 2007. Accepted for publication August 16, 2007.
| REFERENCES |
|---|
|
|
|---|
)2 antibody. Nature Biotech. 17,181-186[CrossRef][Medline]This article has been cited by other articles:
![]() |
Y. Kim, D. Turner, J. Nelson, I. Dobrinski, M. McEntee, and A. J Travis Production of donor-derived sperm after spermatogonial stem cell transplantation in the dog Reproduction, December 1, 2008; 136(6): 823 - 831. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |