Studies suggest that the maternal immune response
to paternal antigens is induced prior to conception and
possibly during insemination. After sexual intercourse, the
infiltration of neutrophils, macrophages and lymphocytes
in the female reproductive tract (FRT) due to the presence
of immune stimulating factors in semen has been clearly
shown. The consequence of this inflammation is the
adaptation of maternal-innate and adaptive immune
responses for the occurrence of pregnancy (1-3). In
immunity against semen, the immune effector response
and the regulatory response (regulatory T cells or Tregs)
are induced. Evidence suggests a delicate balance between
the effector and regulatory responses, and this may be
required for pregnancy to occur. One of the roles of the
immune effector response could be preparation of tissue
destruction factors such as metalloproteinase enzymes and
inflammatory cytokines needed for embryo implantation.
A role of the regulatory response is prevention of an
excessive effector response (4). Disruptions in balance
between the effector and regulatory response results in
pregnancy aberrations, such as recurrent spontaneous
abortion (RSA) (5). These findings suggest that induction
of both effector and regulatory responses against semen
antigens is essential for pregnancy occurrence.Despite many studies on the induction of the maternal
immune response to semen, we could not find any
study that has addressed beneficial humoral immunity
(antibody production) against semen or spermatozoa. We
believe that pregnancy-protective alloantibodies against
the paternal antigen may be produced in the context of
stimulated immunity at insemination and they increase in
pregnancy due to the increase of paternally-derived foetal
antigens in the maternal blood circulation. Therefore,
any destruction of semen or spermatozoa antigenicity
may result in disturbed production of alloantibodies
and pregnancy complications. This supposition, the
production of pregnancy-protective alloantibody at
insemination, was presented when we observed that
absence of pregnancy-protective alloantibodies resulted in RSA (6, 7). Thus, pregnancy-protective alloantibodies
must be present before pregnancy and are necessary for
pregnancy occurrence; otherwise, RSA occurs.Anti-paternal cytotoxic antibody (APCA) is one of the
pregnancy alloantibodies which belongs to the IgG class and
is directed to paternal human leucocyte antigens (HLAs) (8).
There is scant information about its function and mechanism
of action. The absence of APCA is related to RSA. There is
a relationship between APCA development after lymphocyte
therapy and the success of this treatment in improving live
birth rates in RSA women (6, 9). This suggests that APCA
may be produced upon contact with HLAs in semen and
spermatozoa. Although there is considerable controversy
surrounding HLA class I and II expression by spermatozoa
(10-19), we recently demonstrated that these antigens were
expressed by spermatozoa (20). Given this description,
it is very likely that APCA would be produced in contact
with spermatozoa in the context of stimulated immunity at
insemination and help pregnancy to occur. We hypothesize
that APCA production in contact with spermatozoa may
benefit humoral immunity following insemination.As mentioned, we have not found any study that assessed
beneficial humoral immunity against spermatozoa. Thus,
to commence the study about maternal humoral immunity
against spermatozoa, we aim to determine: i. Whether
antibody is produced by the female’s peripheral blood
mononuclear cells (PBMCs) in the presence of the husband’s
spermatozoa, ii. Whether APCA is produced by the female’s
PBMCs in contact with the husband’s spermatozoa, and iii.
The correlation of APCA and antibody production with HLA
class I and II expressions by spermatozoa. To the best of our
knowledge, no study has addressed these topics.
Materials and Methods
Subjects
In this cross-sectional study, we included 30 fertile
couples aged 28-41 years who had at least one child. The
anti-sperm antibody (ASA) test was negative in these
couples. The maternal participants denied any history of
pregnancy complications (e.g., ectopic pregnancy, preterm
and post-term labour or preeclampsia), blood transfusion
or organ transplantation. The husband of each woman had
normal semen status according to the criteria from 2010
guidelines of the World Health Organization. None of the
male partners had any history of genital tract disorders
such as infections, undescended testis, inguinoscrotal
surgery, genital trauma or testicular torsion.Five women, aged 28-42 years, who were virgins
and had no history of semen contact, blood transfusion
or organ transplantation were recruited to this study as
the uncontacted negative control. Because they had no
history of alloantigen contact, we expected that PBMCs
obtained from these women would not produce antibodies
in the presence of spermatozoa. We chose an age and sex
matched control to remove the effect of age and sex related
confounding factors such as hormonal factors, variation of
microbiota with age and other factors that have important effects on the immune system. More reliable results could
be acquired when the only difference between virgin
women (uncontacted control) and the maternal women
(with partners) was the lack of semen contact.Informed consent was obtained from all subjects. The
Ethics Committee of Isfahan University of Medical
Sciences (Isfahan, Iran) approved this protocol (approval
letter: IR.MUI.REC.1395.3.480).
Purification of spermatozoa
Semen samples were collected by masturbation after 2-3
days of sexual abstinence. Sampling was performed under
sterile conditions to prevent false results due to a change of
spermatozoa antigenicity induced by toll-like receptors on
the spermatozoa in response to microbial antigens. After
liquefaction, semen quality was assessed according to WHO
standard guidelines (WHO, 2010) and couples were excluded
if the husband had an abnormal semen quality. We added 2
ml of AllGrad Wash (LifeGlobal® Group, Canada) to the
liquefied semen sample, and centrifuged the sample at 350
g for 10 minutes. The pellet was re-suspended in 1 ml of
AllGrad Wash® and 2 gradient solutions (95% and 45%)
prepared from AllGrad 100% (LifeGlobal® Group, Canada).
In each tube, 1 ml of AllGrad 90% gradient, followed by 1 ml
of AllGrad 45% gradient and then 1 ml of the spermatozoa
suspension were carefully layered. The tubes were centrifuged
at 400 g for 18 minutes. The spermatozoa pellet at the bottom
of the centrifuge tubes was washed with AllGrad Wash® and
then re-suspended in Ham’s F-10 medium (Dacell, Iran) with
1% bovine serum albumin (CMG, Iran). Ham’s F-10 medium
was used because of its antioxidant properties, which makes
it a suitable medium for spermatozoa (21).
Flow cytometry assay
We added 1×106 spermatozoa per 100 µl medium was added to two tubes of each
purified spermatozoa sample. One tube was incubated with phycoerythrin (PE) mouse
anti-human HLA-ABC (clone: G46-2.6, BD Pharmingen, USA) and the other was incubated with
PE mouse anti-human HLA-DR (clone: G46-6, BD Pharmingen, USA) at room temperature for 30
minutes. After two washes with AllGrad Wash® (400 g for 5 minutes), the spermatozoa were
run through a flow cytometer (BD FACSCalibur, USA). Data from at least 100000 events were
collected using forward scatter (FSC) and side angle of light scatter (SSC), a logarithmic
amplifier. Fluorescence data were obtained with the logarithmic amplifier. To determine
background fluorescence (auto-fluorescence and non-specific binding of antibodies), we
used two isotype controls (mouse IgG1, κ [clone: G46-2.6, BD Pharmingen, USA] and mouse
IgG2a, κ [clone: G46-6, BD Pharmingen, USA]). An antibody titration was performed and we
selected the optimal titre that displayed the minimum background to eliminate any
background fluorescence. FSC versus SSC gating was used to identify spermatozoa and remove
debris. To determine the percentage of HLA class I and II positive spermatozoa, SSC versus
logarithmic FL2 (PE) gating was used. The cell viability test was not performed because
abnormal and dead spermatozoa were removed by AllGrad solution before staining. The data
were analysed using FlowJo version10 software.
Isolation of peripheral blood mononuclear cells and
performing co-culture
After taking 5 ml of heparinized venous blood, the PBMCs were separated by centrifugation
on a Ficoll-Hypaque (Lymphoprep, Sigma, USA) density gradient. Cells at the interface were
harvested, washed twice and suspended in complete RPMI 1640 medium supplemented with
HEPES, L-glutamine, penicillin (100 U/ml), streptomycin (10 mg/ ml), 2-mercaptoethanol
(2×10-5 M) and 20% autologous serum. In this suspension, we adjusted the cell
concentration to 1×106 cells/ml. A 2 ml suspension (2×106 PBMCs) was
transferred to 24-well plates and cultured in the presence of 5×106
spermatozoa. As the negative control, 2×106 maternal PBMCs without spermatozoa
were cultured in parallel to the co-cultures. Cells were then incubated at 37˚C in a
humidified 5% CO2 atmosphere. After four days, cells were washed three times
and the pellet was re-suspended in 2 ml complete RPMI 1640 medium in which autologous
serum was replaced by 20% foetal calf serum (FCS). After incubation for eight days at 37˚C
in a humidified 5% CO2 atmosphere, the supernatants were harvested, aliquoted
and kept at -80°C for future use.For the uncontacted negative control, blood was taken from
the women volunteers who were virgins. The PBMCs were
separated and these cells were co-cultured with the pooled
spermatozoa obtained from five normal donors. The culture
procedure was similar to that mentioned above. There were
two controls: i. Negative control (culture of PBMC alone
from the maternal women) and ii. Uncontacted negative
control (co-culture PMBC from virgin women with pooled
spermatozoa). The first control was run to ensure that any
unknown factors did not lead to the production of antibodies,
and that the production of antibodies in the co-cultures
was because of the presence of spermatozoa. The second
control was performed to ensure that antibodies produced
in the presence of spermatozoa indicated that sensitization
had previously occurred. In other words, the uncontacted
negative control was run to confirm that the female humoral
immune response (antibody production) was induced by
spermatozoa at the time of insemination.
Enzyme-linked immunosorbent assay analysis
The concentration of IgG in the supernatant was
measured using an enzyme-linked immunosorbent assay
(ELISA) kit in accordance with the manufacturer’s
protocol [IgG (Total) Human Uncoated ELISA Kit,
Invitrogen, USA]. Normal serum samples that contained
IgG were used as the positive control.
Complement-dependent cytotoxicity assay for
determining the anti-paternal cytotoxic antibody titre
in the supernatants
We assessed the APCA percentage by cross-matching between supernatants (1:64 dilution)
and freshly prepared paternal PBMCs. The test was performed in triplicate in Terasaki
plates covered with light paraffin oil. We mixed one µl of paternal PBMCs suspension
(density: 2×103 cells/ml) with 1 µl supernatant. For quality control of the
complement-dependent cytotoxicity (CDC) assay and to ensure accuracy of the assay, two
control samples – a negative control (antibody without cytotoxic antibody) and positive
control (antibody with cytotoxic activity) were used. The positive control was the serum
of the women at the third trimester of pregnancy because it contained a high level of
APCA. APCA increases to detectable levels from the 28th week of pregnancy until
four weeks after delivery, after which it decreases to an undetectable level (22). The
serum of these women was mixed with their male partner’s PBMCs. The negative control
consisted of supernatants obtained from co-culturing PBMCs from virgin women with pooled
spermatozoa, which lacked any antibodies. After one hour at room temperature, we added 5
µl of rabbit complement (Inno-Train, Germany). One µl eosin dye was added to the wells
after one hour of incubation at room temperature, followed by 5 µl of formalin (37%). The
test plates were left overnight to allow the cells to settle. The plates were read using a
phase contrast microscope (Olympus, Japan). The number of dead cells among 1000 PBMCs was
determined and reported as the percentage of APCA.
Statistical analysis
Descriptive analysis of the percentage of HLA class I
and II expression by spermatozoa, IgG concentration and
percentage of APCA in supernatant included the mean
and standard deviation (SD). A Pearson product-moment
correlation coefficient was used to assess the relationship
between variables. All data analysis was performed using
IBM SPSS statistics 25 software. A P≤0.05 was considered
statistically significant.
Results
Evaluation of human leucocyte antigen class I and II
on the surface of spermatozoa
Flow cytometric results showed that 25.77 ± 9.9% of
spermatozoa expressed HLA class I on their surface and
that 29.95 ± 10.10% of spermatozoa expressed HLA class
II. Figure 1 presents a representative flow cytometry dot
plot and histogram overlay.
IgG concentration in the supernatant
Supernatant IgG levels were measured by ELISA. IgG
was detected in all of the supernatants. The mean ± SD
of the IgG concentration in supernatants was 669 ± 132
ng/ml. The IgG concentration in the negative control was
3.20 ± 2.31 ng/ml and in the uncontacted negative control,
it was 3.2 ± 2.90 ng/ml.
Percentage of anti-paternal cytotoxic antibody in the
supernatant
CDC results showed that all supernatants were positive
for APCA. The percentage of APCA in the supernatant was
73.93 ± 16.01%. The percentage of dead cells in the negative
controls was 3.12 ± 2.96%. Figure 2 shows phase contrast
microscope images of PBMCs after the CDC assay.Representative of the flow cytometry dot plot and histogram overlay. Purified spermatozoa were assessed for expression of human leucocyte
antigen (HLA) class I and II by flow cytometry. The dot plot is used to show the percentage of HLA class I and HLA class II positive spermatozoa. C is a
representative stagger histogram overlay of the isotype control with the experimental sets to show the background level after minimizing background
auto-fluorescence.Phase contrast microscope images of the husbands’ peripheral blood mononuclear cells (PBMCs)
after the complement-dependent cytotoxicity (CDC) assay. Eosin dye was taken up by the
dead cells and they became a dark colour. A. Positive control,
B. Test (cytotoxic activity of supernatant on paternal PBMCs, and
C. Negative control. Black arrow; Dead cells and Gray arrow; Live
cells.
Correlation assessments
A Pearson product-moment correlation coefficient was
computed to assess the relationship between APCA production
and HLA class I and II expression by spermatozoa. There
was a positive correlation between these factors [P=0.011
(HLA-I) and P=0.013 (HLA-II), Table 1].Correlation results*; Correlation is significant at the 0.05 level (2-tailed), **; Correlation is
significant at the 0.01 level (2-tailed), n; Number; HLA; Human leucocyte
antigens, APCA; Anti-paternal cytotoxic antibody, and r; Correlation
coefficient.
Discussion
Despite numerous studies about the cellular immunity
that is induced in FRT at insemination, the humoral
immunity and the production of useful antibodies against
spermatozoa (ASA) has not been assessed. In this study, we
show, for the first time, that IgG and APCA are produced
by the wife’s PBMCs in the presence of the husband’s
spermatozoa. Furthermore, we show a positive correlation
between HLA expression by spermatozoa and antibody
production. To the best of our knowledge, this is the first
study that has investigated the ability of spermatozoa to
induce the production of IgG and APCA. We could not
find any study that investigated the correlation between
HLA expression on spermatozoa and the induction of
immunity (e.g., antibody production) by spermatozoa.The production of antibodies (IgG) against an antigen
in vitro by PBMCs is indicative of the stimulation of the
immune response against the antigen and development of
memory B lymphocytes that respond to the presence of
the antigen in the body. In this study, we have shown that
the antibody (IgG) was produced ASA when the wife’s
PBMCs were co-cultured with the husband’s spermatozoa.
This result suggested development of memory humoral
immunity ASA in the wife (or maternal) body.Regarding these produced antibodies, the question arises
as to which antigens of spermatozoa have the capability to
cross-react with paternal lymphocyte antigens. One of the
most likely antigens are the HLA molecules. This study
showed a positive correlation between HLA class I and
II and APCA production. It is possible that APCA, which
is originally developed through exposure to paternally
derived foetal antigens during pregnancy, can cross-react with the same antigens on spermatozoa. However,
it should be noted that the immune system is exposed to
semen antigen before foetal antigen.Our results show that fertile women produce
antibodies against spermatozoa (ASA). In contrast
to our results, many studies have shown that ASA in
serum or in FRT secretions results in infertility. Also,
numerous studies have sought to determine the antigen
specificity of ASA. Despite the large number of studies
on the pathogenicity of ASA for pregnancy, the presence
of this antibody was reported in a very small percentage
of fertile women. Accordingly, the results of our study
showed that fertile women have immunological memory
of antibody production in the presence of the husband’s
spermatozoa (23-25). However, the question is raised
as to why APCA, a type of ASA, is not detectable in
most normal parous women. Two possible reasons can
be suggested. First, under normal conditions, APCA
producing plasma cells migrate in the FRT and locally
produce a small amount of antibody so that the release
of antibody into the blood is very limited, and often not
identifiable by current methods. Second, APCA may be
produced when first encountering spermatozoa, after
which it may bind to spermatozoa and be eliminated by
cells in the FRT. Consequently, the amount of unbound
antibody is very low and cannot be detected in serum
and cervical secretion or blood.The question arises as to why APCA with cytotoxic
activity does not lead to infertility. We believe that, in
addition to not leading to infertility, APCA has some
major roles for supporting pregnancy. These possible
roles for APCA before pregnancy include:i. Opsonizing senescent and damaged spermatozoa
for phagocytosis, ii. Assisting in promoting a proper
immune response required for implantation and tolerance
induction, iii. Because the cytotoxic activity of immune
cells (Natural killer cells and cytotoxic T lymphocytes)
is essential for implantation (26-28), it is probable that
APCA also has a role in this cytotoxicity process. This
role of APCA may occur through the antibody-dependent
cell-mediated cytotoxicity (ADCC) that is intermediated
by NK cells.Therefore, it seems logical that the APCA in normal
women (which we named natural ASA) has a different
function and features compared to the pathogenic ASA in
infertile women. We believe that natural and pathogenic
ASA may have different properties, such as antigenic
specificity, glycosylation type and isotype. Further studies
should be performed to assess this supposition.Determination of the level of APCA in the serum by
the CDC assay is a diagnostic test for RSA. However, the
sensitivity of this method is low because of the decreased
level of APCA in the serum. We intend to use APCA
production in the presence of spermatozoa as a diagnostic
test for RSA. To achieve this purpose, a study should be
performed on a large number of fertile and RSA couples
to determine a cut-off value and the normality of APCA.
Conclusion
Based on the results of this study it can be concluded that
APCA can be a natural ASA, which might be produced
during exposure to spermatozoa and might have some
influence before pregnancy. Further research is required
to determine the role of APCA before pregnancy.
Table 1
Correlation results
Pearson correlation
IgG (ng/ml)
APCA (%)
HLA-I %
r
0.459*
0.509**
P value
0.011
0.004
n
30
30
HLA-II %
r
0.449*
0.499**
P value
0.013
0.005
n
30
30
*; Correlation is significant at the 0.05 level (2-tailed), **; Correlation is
significant at the 0.01 level (2-tailed), n; Number; HLA; Human leucocyte
antigens, APCA; Anti-paternal cytotoxic antibody, and r; Correlation
coefficient.