Expression of Avian -Defensins in the Oviduct and Effects
of Lipopolysaccharide on Their Expression in the Vagina of Hens
A. M. Abdel Mageed, N. Isobe, and Y. Yoshimura1
Graduate School of Biosphere Science, Hiroshima University, Higashi-Hiroshima 739-8528, Japan
ABSTRACT The aims of this study were to (i) determine
the types of avian -defensin genes (AvD) expressed in
the hen oviduct and (ii) to examine the effects of lipopoly-
saccharide (LPS) treatment in vivo on their expression in
the vagina. Birds were i.v. treated with LPS (1 mg/kg of
BW), and subsequently the oviducts were analyzed 0,
3, 6, 12, or 24 h after LPS administration. The mRNA
expression for AvD was examined by reverse transcrip-
tion-PCR using RNA preparations from the mucosal tis-
sues of all the oviductal segments. Furthermore, changes
Key words: avian -defensin, lipopolysaccharide, hen oviduct, vagina
INTRODUCTION
The hen oviduct consists of the infundibulum, mag-
num, isthmus, uterus, and vagina. Egg formation, includ-
ing the secretion of albumen, egg-shell membrane, and
egg-shell components, is completed during the passage
of the yolk through the oviduct. Immune functions in the
oviduct play essential roles in protection of the oviductal
tissues from infections and production of pathogen-free
eggs. The oviduct itself may be infected by various micro-
organisms such as Salmonella enteritidis (Barnhart et al.,
1993) and Mycoplasma meleagridis (Yamamoto and Herrad,
1966). An adaptive immune response through immuno-
competent cells in the oviduct including antigen pres-
enting cells expressing MHC class II as well as T and B
cells has been well described (Yoshimura et al., 1997;
Zheng et al., 1997, 1998, 2001; Zheng and Yoshimura,
1999). Although the innate immunity plays an essential
role in the first line of defense against infection, there is
only limited information available as what extend on the
hen oviduct.
Defensins are antimicrobial peptides that may trigger
an innate immune response and are divided into 3 groups,
namely -, -, and -defensins. The -defensins are char-
2008 Poultry Science Association Inc.
Received July 12, 2007.
Accepted February 10, 2008.
1
Corresponding author: yyosimu@hiroshima-u.ac.jp
979
in their mRNA expression profiles in the vagina were
analyzed by semiquantitative reverse transcription-PCR.
The AvD-1, -2, -3, -4, -5, -7, -8, -9, -10, -11, and -12 were
identified in each oviductal segment from infundibulum
to vagina. Among these AvD, the expression of AvD-
3, -5, -10, -11, and -12 in the vagina were significantly
increased in response to LPS treatment, whereas the oth-
ers did not show significant changes. These results sug-
gest that all 11 types of AvD are expressed in the hen
oviduct and at least 5 of them in the vagina show in-
Downloaded from http://ps.oxfordjournals.org/ by guest on March 29, 2015
creased expression in response to LPS.
2008 Poultry Science 87:979984
doi:10.3382/ps.2007-00283
acterized by 6 cysteine residues, and have been found in
many animal species such as bovine (Luenser and Lud-
wig, 2005), ovine (Luenser et al., 2005), and porcine and
human (Yongming et al., 2006). These peptides are poten-
tially kill a wide range of microorganisms including
gram-positive and gram-negative bacteria, fungi, and
yeast (Haryadi and Pak, 2004). Avian antimicrobial pep-
tides classified as -defensins had been previously called
gallinacins. However, it has now been agreed to use their
gene names [i.e., avian -defensin (Lynn et al., 2007)]. We
therefore decided to use the new terminology for this
study rather than previous reports. So far, 13 avian -
defensin genes (AvD) have been identified. Zhao et al.
(2001) reported that gallinacin gene (Gal)-1 and -2 were
expressed in bone marrow and lung, whereas Gal-3 was
more preferentially expressed in bone marrow, tongue,
trachea, and the bursa of Fabricii. Xiao et al. (2004) de-
scribed that Gal-1 to -7 are predominantly expressed in
the bone marrow and the respiratory tract, whereas Gal-
8 to -13 were restricted to the liver and urogenital tract.
The theca and granulosa layers of ovarian follicles ex-
pressed 6 types (theca) or 4 types (granulosa layer) of
Gals, respectively, whereas the ovarian stroma expressed
12 types including Gal-1 to -12 (Subedi et al., 2007). Milona
et al. (2007), who studied the antimicrobial activity of Gal-
4, -7, and -9 in the intestine of the chicken, suggested
that gallinacins act as antimicrobial agent constituting an
integral part of the avian host innate defense system.
Ohashi et al. (2005) identified the expression of Gal-
1, -2, and -3 in all the oviductal segments with greater
980 ABDEL MAGEED ET AL.
Figure 1. Pattern of reverse transcription-PCR products for avian -defensins mRNA (AvD) in the mucosal tissues of hen oviduct treated with
lipopolysaccharide (LPS). The hens were i.v. treated with 1 mg of LPS/kg of BW before 3 h of examination. The PCR products were electrophoresed
on 2% agarose gel containing ethidium bromide. M = marker.
expression in the infundibulum and vagina than in the
other segments. The vagina is one of the sites that may
be contaminated by microorganisms because it opens to
the cloaca. Recently, Yoshimura et al. (2006) reported that
the expression of Gal-1, -2 and -3 was enhanced in re-
sponse to Salmonella enteritidis infection or in response to
purified lipopolysaccharide (LPS) in the cultured vaginal
cells. The LPS is a gram-negative bacterial cell wall com-
ponent (Sunwoo et al., 1996), which mimics the effects
of a bacterial infection (Leshchinsky and Klasing, 2003).
Although expression of 3 types of AvD have been shown
in the oviduct, that of the other types of AvD in the
different segments of oviduct have not yet been reported.
If the synthesized avian -defensins play roles in the host
immunity to eliminate microorganisms, their expressions
are expected to be enhanced in response to bacterial com-
ponents.
The aim of this study was to determine the types of
AvD expressed in the oviduct and the effects of in vivo
treatment with LPS on the expression of AvD in the
vagina. The types of AvD expressed in the oviduct were
examined using birds treated with or without LPS.
MATERIALS AND METHODS
Experimental Birds
White Leghorn hens (approximately 400 d old) regu-
larly laying 5 or more eggs in a sequence were used. Hens
Downloaded from http://ps.oxfordjournals.org/ by guest on March 29, 2015
were kept in individual cages under a daily light regimen
of 14L:10D and provided with feed and water ad libitum.
They were i.v. treated with LPS preparations from E. coli
(Wako Pure Chem., Osaka, Japan) dissolved in PBS at a
dose of 1 mg/kg of BW. This dose of LPS had been
confirmed to increase AvD expression in the ovarian
theca tissue in our previous report (Subedi et al., 2007).
Birds were euthanized under anesthesia with sodium
pentobarbital (Abbott Lab., Chicago, IL) after 0, 3, 6, 12,
or 24 h of LPS treatment to collect the oviduct. Handling
of birds was done in accordance with the regulations of
Hiroshima University for animal experiments.
RNA Extraction
Total RNA was extracted from the mucosal tissues of
the infundibulum, magnum, isthmus, uterus and vagina,
and liver tissue using Sepazol RNA I super (Nacalai
Tesque, Kyoto, Japan) according to the manufacturers
directions. The obtained RNA pellet was then dissolved
in Tris-EDTA buffer and kept at 80C until use. A mix-
ture (10 L) of the RNA sample, 2 U DNase I (RNase
free; TaKaRa Shuzo, Shiga, Japan), and 1 DNase buffer
was subjected to 37C for 1 h, at 80C for 10 min, then
at 4C using Programmable Thermal Controller PTC-100
(MJ Research, Waltham, MA). The concentration of the
purified total RNA was then determined using Gene
Quant Pro (Amersham Pharmacia Biotec, Cambridge,
UK).
 -DEFENSINS IN HEN OVIDUCT 981

Downloaded from http://ps.oxfordjournals.org/ by guest on March 29, 2015

Figure 2. Changes in the expression of avian -defensins mRNA (AvD) in the vagina and AvD-3 in the liver after i.v. treatment with
lipopolysaccharide (LPS). Birds were i.v. treated with 1 mg of LPS/kg of BW, and AvD expression in the vaginal mucosa was examined by
semiquantitative reverse transcription-PCR at different periods after LPS treatment. (a) (k): AvD expression in the vagina. (l) AvD-3 expression
in the liver. Values are mean SEM of the AvD/-actin ratio (n = 4). acValues with different letters are significantly different (P < 0.05).

Semiquantitative Reverse facturers instructions. The reaction mixture (10 L) con-

Transcription-PCR
Semiquantitative reverse transcription-PCR was per-
formed as described by Subedi et al. (2007). The RNA
samples were reverse-transcribed using ReverTra Ace
(Toyobo Co. Ltd., Osaka, Japan) according to the manu-
sisted of 1 g of the total RNA, 1 RT buffer, 1 mM dNTP
mixture, 20 U of RNase inhibitor, 0.5 g of oligo (dT)20,
and 50 U of Rever Tra Ace. The reverse transcription was
performed at 42C for 30 min, followed by heat inactiva-
tion for 10 min at 99C using the PTC-100 programmable
thermal controller (MJ Research, Waltham, MA). The PCR
982 ABDEL MAGEED ET AL.
Table 1. The PCR primers and their accession numbers used for profiling of avian -defensins (AvD) and -actin
Target
AvD-1 F-AAACCATGCGGATCGTGTACCTGC
AvD-2 F-GTTCTGTAAAGGAGGGTCCTGCCAC
AvD-3 F-CTGCCGCTTCCCACACATAG
AvD-4 F-ATCGTGCTCCTCTTTGTGGCAGTTCA
AvD-5 F-ATGCAGATCCTGCCTGCCTCTCTCTTTGCT
AvD-6 F-GATCCTTTACCTGCTGCTGTCT
AvD-7 F-CTGCTGTCTGTCCTCTTTGTGG
AvD-8 F-ACAGTGTGAGCAGGCAGGAGGGA
AvD-9 F-ATGAGAATCCTTTTCTTCCTTGTTGC
AvD-10 F-CTGTTCTCCTCTTCCTCTTCCAG
AvD-11 F-ACTGCATCCGTTCCAAAGTCTG
AvD-12 F-CCCAGCAGGACCAAAGCAATG
AvD-13 F-CATCGTTGTCATTCTCCTCCTC
-actin F-TTCCAGCCATCTTTCTTG
was performed in a reaction mixture of 25 L containing
0.5 L of cDNA, 1 PCR buffer, 0.2 mM dNTP mixture,
0.4 M each primer, and 0.625 U of Takara Taq (Takara
Bio. Inc., Shiga, Japan). Table 1 shows the primers used
for PCR, which were the similar primers as have been
used in a previous study (Subedi et al., 2007). To deter-
mine the types of AvD expressed in the oviduct, the
oviducts collected pre or at 3 h after LPS treatment were
used. The cDNA samples from each oviductal segment
were amplified using the primers of all types of AvD at
different annealing temperature ranging from 52 to 60C
and 40 PCR cycles. For semiquantitative reverse transcrip-
tion-PCR analysis of AvD expression in the vagina, dif-
ferent PCR cycles (30, 35, 40, and 45 cycles) were examined
to optimize the amplification. A linear response for the
30 through 45 cycles was observed, and 35 cycles for
AvD-12 and 40 cycles for the other AvD and -actin
were considered optimal. Each observed AvD was am-
plified using an optimal number of cycles to examine the
changes in their expression in response to LPS in the
vaginal mucosa and liver tissue. The cycle parameters
were denaturation at 94C for 30 s, 35 cycles (for AvD-
12), or 40 cycles (for AvD-1 to -11 and -actin), annealing
at 58C (for AvD-1, -2, -4, -5, -7, -9, and -10, and -
actin) or 60C (for AvD-3, -8, -11, and -12) for 30 s, and
extension at 72C for 1 min, followed by the final exten-
sion at 72C for 6 min. The PCR products were separated
by electrophoresis on 2% (wt/vol) agarose gels containing
0.5 mg/mL of ethidium bromide and photographed un-
der UV illumination. Densitometry for the PCR product
bands was performed using UN-SCAN-IT gel TM, ver.
6.1, (Silk Scientific Corporation, Orem, UT), and then the
ratio of AvD to -actin densities was obtained.
Statistical Analysis
The values of each AvD expression were obtained
as the mean SEM of the ratio of AvD/-actin. The
significance of differences in the expression of AvD after
LPS treatment was examined by 1-way ANOVA, fol-
lowed by Duncans multiple range test. Differences with
P value of < 0.05 were considered statistically significant.
Accession
Primers 5-3 number
R-CAATGCTAAACTGCACACCTTTA AF033335
R-ACTCTACAACACAAAACATATTGC AF033336
R-GCAATGCCAAACTGCACGCCTTTA NM_204650
R-CTACAACCATCTACAGCAAGAATACT NM_001001610
R-TCAGGAATACCATCGGCTCCGGCAGCAGAA NM_001001608
R-TCCTCACACAGCAAGATTTTAGTC NM_00100193
R-CATTTGGTAGATGCAGGAAGGA NM_00100194
R-CTCTTCTGTTCAGCCTTTGGTG NM_001001781
R-TTAGGAGCTAGGTGCCCATTTGCAGC NM_001001611
R-AATCTTGGCACAGCAGTTTAACA NM_001001609
R-TCGGGCAGCTTCTCTACAAC NM_001001779
R-GTGAATCCACAGCCAATGAGAG NM_001001607
R-ACTTGCAGCGTGTGGGAGTTG NM_001001780
R-TCCTTCTGCATCCTGTCA X00182
RESULTS
Figure 1 shows the pattern of PCR products of AvD
Downloaded from http://ps.oxfordjournals.org/ by guest on March 29, 2015
obtained from the oviduct after 3 h of LPS treatment. The
PCR products of AvD-1, -2, -3, -4, -5, -8, -9, -10, -11,
and -12 were observed in all the oviductal segments, and
AvD-7 products showed only faint bands in the isthmus,
uterus, and vagina. No AvD-6 and -13 expression was
observed in any segments of the oviduct. In the oviduct
of nontreated birds, some of the bands that could be
identified in the treated birds were faint or undetectable
(data not shown).
Effects of LPS treatment on the expression of 11 types
of AvD in the vagina are shown in Figure 2. The expres-
sion of AvD-3, -5, -10, and -11 increased by 3 to 6 h of
treatment. Then, the expression of AvD-5 and -10 were
kept higher until 24 h of treatment, whereas those of
AvD-3 and -11 showed a tendency to decline after the
peak at 3 h and 6 h of posttreatment. The AvD-12 expres-
sion was also increased by LPS treatment although the
significance was found only at 24 h of posttreatment. The
average values at the peak of AvD-1, -2, -4, -7, -8, and -
9 showed approximately 2- to 5-fold in the LPS-treated
groups compared with pretreatment; however, the differ-
ences were not statistically significant.
The effect of LPS treatment on the expression of AvD
in the liver was also examined. The liver also expressed
11 types of AvD (except for AvD-6 and -13). The expres-
sion of AvD-3 was decreased by 3 h of posttreatment
and kept lower until 24 h (Figure 2-l), whereas the other
AvD did not show significant changes in their expression
(data not shown).
DISCUSSION
We are reporting the types of AvD expressed in hen
oviduct and the changes in their expression in response
to i.v. treatment with LPS. Significant findings were (1)
11 types of AvD were expressed in all segments of ovi-
duct, (2) expression of 5 types of AvD was increased in
the vagina in response to LPS treatment.
Ohashi et al. (2005) reported that all the oviductal seg-
ments expressed Gal-1, -2, and -3 with a higher expression
 -DEFENSINS IN HEN OVIDUCT
in the infundibulum and vagina. The current study pro-
vided further findings that 11 types out of 13 reported
AvD were expressed, but the expression of AvD-6 and -
13 was not observed in any segments of the oviduct.
Previous reports suggested that avian -defensins play
significant role in the host innate immunity of mucosal
tissues; namely, the expression of Gal-4, -6, -7, and -9 were
expressed in the intestine and Gal-3, -4, -5, -6, and -7 were
expressed in the trachea (Albert et al., 2007; Milona et al.,
2007). Although the healthy mucosal tissues of oviduct
and other organs express AvD, the types of the AvD
are likely to be different among the organs. However, the
current results suggest that the types of AvD are same
among each oviductal segment; namely, 11 types of AvD
could be expressed in all the segments.
We found that the expression of AvD-3, -5, -10, -11,
and -12 in the vagina was increased by the i.v. treatment
with LPS. Yoshimura et al. (2006) reported that the expres-
sion of Gal-1, -2, and -3 was increased in the cultured
vaginal cells by stimulation with Salmonella enteritidis or
LPS. The current study confirmed that AvD-3 expression
is enhanced by LPS even in vivo. Furthermore, the current
results suggested that expression of AvD-5, 10, -11, and -
12 was also increased by LPS stimulation. However, there
were temporal differences for the increase of expression;
namely, that of AvD-3 and -11 was increased only from
3 to 6 h of posttreatment with LPS, AvD-5 and -10 were
elevated from 6 to 24 h, and AvD-12 was increased only
after 24 h. These temporal differences in the increase may
enable the oviductal mucosa to express AvD for a longer
time to attack the microorganisms by the combination of
different types of AvD. Although the reason why such
differences in the increase of expression occur still un-
known, we assume that the types of cells expressing AvD
may not be single and the time to recognize circulating
LPS, the time of intracellular response to express AvD,
or both, may be different among these cells. Ohashi et al.
(2005) showed AvD-1, -2, and -3 expression in the basal
cells of the vaginal mucosal epithelium. Leukocytes that
distribute in the vaginal mucosa may also contribute to
synthesize AvD (Evans et al., 1994; Harwig et al., 1994;
Brockus et al., 1998).
The expression of Gal-1 and -2 of cultured vaginal cells
was significantly enhanced by the LPS stimuli (Yoshi-
mura et al., 2006), whereas their expression did not show
statistically significant changes under in vivo treatment
with LPS in the current study. Although the reason why
these differences exist between the 2 studies is not known,
we assume that the intensity of the LPS stimulation to
the vaginal cells was different between the studies. Fur-
thermore, the expression level of not only AvD-1 and -
2 but also AvD-4, -7, -8, -9 did not change significantly
in response to LPS, although the average value differed
2- to 5-fold between pre- and posttreatment. It is assumed
that a proper stimulation by LPS may also cause a signifi-
cant increase of those AvD.
Unlike in the oviduct, AvD-3 expression in the liver
was reduced by LPS treatment, whereas no change was
observed in the expression of the other AvD. Thus, the
983
effect of LPS on the AvD expression may be different
among the different tissues and cells.
In conclusion, we suggest that 11 types of AvD were
expressed in all the segments of oviduct and at least 5
types were enhanced in the vagina in response to LPS.
ACKNOWLEDGMENTS
We wish to thank W. Schwaeble, Department of Infec-
tion, Immunity and Infection, University of Leicester, UK,
and Animesh Barua, Rush University Medical Center,
Rush University, Chicago, IL, for critical reviewing of this
manuscript. This work was supported by Grant-in-Aid
for Scientific Research from the Japan Society for the Pro-
motion of Science.
REFERENCES
Albert, V. D., J. A. Edwin, I. C. Stefanie, L. M. Johanna, A. R.
Roland, and P. H. Henk. 2007. The -defensin gallinacin-6 is
Downloaded from http://ps.oxfordjournals.org/ by guest on March 29, 2015
expressed in the chicken digestive tract and has antimicrobial
activity against food-borne pathogens. Antimicrob. Agents
Chemother. 51:912922.
Barnhart, H. M., D. W. Dreesen, and J. L. Burke. 1993. Isolation
of Salmonella from ovaries and oviducts from whole carcasses
of spent hens. Avian Dis. 37:377380.
Brockus, C. W., M. W. Jackwood, and B. G. Harmon. 1998.
Characterization of beta-defensin prepropeptide mRNA
from chicken and turkey bone marrow. Anim. Genet.
29:283289.
Evans, E. W., G. G. Beach, J. Wunderlich, and B. G. Harmon.
1994. Isolation of antimicrobial peptides from avian hetero-
phils. J. Leukoc. Biol. 56:661665.
Harwig, S. S., K. M. Swiderek, V. N. Kokryakov, L. Tan, T. D.
Lee, E. A. Panyutich, G. M. Aleshina, O. V. Shamova, and
R. I. Lehrer. 1994. Gallinacins: Cysteine-rich antimicrobial
peptides of chicken leukocytes. FEBS Lett. 342:281285.
Haryadi, S., and L. Pak. 2004. Avian antimicrobial peptides: The
defense role of -defensins. Biochem. Biophys. Res. Com-
mun. 323:721727.
Leshchinsky, T. V., and K. C. Klasing. 2003. Profile of chicken
cytokines induced by lipopolysaccharide is modulated by
dietary alpha-tocopheryl acetate. Poult. Sci. 82:12661273.
Luenser, K., F. Jorns, and A. Ludwig. 2005. Evolution of caprine
and ovine -defensin genes. Immunogenetics 7:487498.
Luenser, K., and A. Ludwig. 2005. Variability and evolution of
bovine -defensin genes. Genes Immun. 6:115122.
Lynn, D. J., R. Higgs, A. T. Lloyd, V. Heroe-Groepinet, Y. Nys,
F. S. L. Brinkman, P. L. Yu, A. Soulier, P. Kaiser, G. Zhang,
C. OFarrelly, and R. I. Lehrer. 2007. Avian-beta defensin
nomenclature: A community proposed update. Immunol.
Lett. 110:8689.
Milona, P., C. Townes, R. Bevan, and J. Hall. 2007. The chicken
host peptides, gallinacins 4, 7, and 9 have antimicrobial activ-
ity against Salmonella serovars. Biochem. Biophys. Res. Com-
mun. 356:169174.
Ohashi, H., K. Subedi, M. Nishibori, N. Isobe, and Y. Yoshimura.
2005. Expressions of antimicrobial peptide gallinacin-1,-2
and-3 mRNAs in the oviduct of laying hens. J. Poult. Sci.
42:337345.
Subedi, K., N. Isobe, M. Nishibori, and Y. Yoshimura. 2007.
Changes in the expression of gallinacins, antimicrobial pep-
tides, in ovarian follicles during follicular growth and in
response to lipopolysaccharide in laying hens (Gallus domes-
ticus). Reproduction 133:127133.
Sunwoo, H., T. Nakano, W. Dixon, and J. S. Sim. 1996. Immune
responses in chickens against lipopolysaccharide of Esche-
richia coli and Salmonella typhimurium. Poult. Sci. 75:342345.
984 ABDEL MAGEED ET AL.
Xiao, Y., A. L. Hughes, J. Ando, Y. Matsuda, J. F. Cheng, O.
Skinner-Noble, and G. Zhang. 2004. A genome-wide screen
identifies a single beta-defensin gene cluster in the chicken:
Implications for the origin and evolution of mammalian de-
fensins. BMC Genomics 5:56.
Yamamoto, R., and B. Herrad. 1966. Pathogenicity of Mycoplasma
meleagridis for turkey and chicken embryos. Avian Dis.
10:268272.
Yongming, S., A. P. Amar, Z. Guolong, R. Chris, and B. Frank.
2006. Bioinformatic and expression analysis of novel porcine
-defensins. Mamm. Genome 17:332339.
Yoshimura, Y., H. Ohashi, K. Subedi, M. Nishibori, and N.
Isobe. 2006. Effects of age, egg-laying activity, and Salmonella-
inoculation on the expressions of gallinacin mRNA in the
vagina of the hen oviduct. J. Reprod. Dev. 52:211218.
Yoshimura, Y., T. Okamoto, and T. Tamura. 1997. Localization
of MHC class II, lymphocytes and immunoglobulins in the
oviduct of laying hens. Br. Poult. Sci. 38:590596.
Zhao, C., N. Tung, L. Lide, E. Randy, K. Sacco, and I. L. Robert.
2001. Gallinacin-3, an inducible epithelial -defensin in the
chicken. Infect. Immun. 69:26842691.
Zheng, W., M. Nishibori, N. Isobe, and Y. Yoshimura. 2001. An
in situ hybridization study of the effects of artificial insemina-
tion on the localization of cells expressing MHC class II
mRNA in the chicken oviduct. Reproduction 122:581586.
Zheng, W., and Y. Yoshimura. 1999. Localization of macro-
phages in the chicken oviduct: effect of age and gonadal
steroids. Poult. Sci. 78:10141018.
Zheng, W., Y. Yoshimura, and T. Tamura. 1997. Effect of sexual
maturation and gonadal steroids on the localization of IgG,
IgM and IgA positive cells in the chicken oviduct. J. Reprod.
Fertil. 111:277284.
Zheng, W., Y. Yoshimura, and T. Tamura. 1998. Effects of age
and gonadal steroids on the localization of antigen presenting
cells, and T and B cells in the chicken oviduct. J. Reprod.
Fertil. 114:4554.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 29, 2015