Bruce R. Gilbert, M.D., Ph.D.
Steven S. Witkin, Ph.D.
Marc Goldstein, M.D.
The James Buchanan Brady Foundation, The New York Hospital-Cornell
Medical Center and the Population Council, New York, New York
Received August 15, 1988; revised and accepted
Department of Surgery, Division of Urology, The James Buchanan
Brady Foundation, The New-York Hospital-Cornell Medical Center
Immunology Division, Department of Obstetrics and Gynecology,
The New York Hospital-Cornell Medical Center
The Population Council, New York, New York
CAPSULE ABSTRACT: 32% of infertile men with palpable varicoceles
were found to have immunoglobulins associated with their ejaculated
sperm. This subset of patients had significantly poorer semen
quality.
ABSTRACT
Sperm-bound immunoglobulins were found in 27 (32%) of 84 infertile
men with palpable varicoceles. These men were divided into two
groups based upon the presence (group I; 32%) or absence (group
II; 68%) of sperm-bound immunoglobulins, as measured by an enzyme
linked immunosorbant assay. Circulating antisperm antibodies were
identified in 52% of patients with sperm-bound immunoglobulins
and 14% of group II patients. The presence of sperm-bound immunoglobulins
was associated with a small but significant decrease in both sperm
concentration and motility. Sperm-bound immunoglobulins are present
in a greater percentage of infertile men with varicoceles then
infertile men without varicoceles. Their presence may be a marker
for damage to the seminiferous epithelia in men with varicoceles
and may also contribute to varicocele associated infertility.
INTRODUCTION:
A palpable varicocele is present in 30% to 40% of infertile
men1,2. This is 2 to 3 times the incidence of varicocele in the
general male population 3,4. MacLeod 5,6 described an abnormal
semen analysis pattern associated with varicocele consisting of
increased tapering forms, immature germinal cells and severe oligospermia.
In addition, 85% of these patients exhibited impaired sperm motility.
Varicocelectomy results in improved semen parameters in two thirds
of patients and pregnancy rates ranging from 24% to 53% 7. Although
this data clearly establishes a link between varicocele and male
infertility the pathophysiology of varicocele remains incompletely
understood.
Recent studies8,9 have associated varicocele related infertility
with antisperm antibody formation. Golomb et al 8 compared a group
of infertile men with varicoceles to a group of infertile men
without a palpable varicocele. They found a greater percent of
total antisperm antibody (i.e., serum, semen and sperm-bound)
in the infertile men with varicoceles (91%) as compared to infertile
men without varicoceles (41%), 38% of their varicocele patients
had antibody bound to sperm. Also, sperm-bound antibody was present
in 25% of men in the study of Ozen et al9. However, in their study
the use of methanol fixation of sperm is now known to expose antigens
other then those bound to the sperm surface. Therefore the clinical
significance of their results is not easily evaluated.
In our prospective study, infertile men with palpable varicoceles
were found to have a high incidence of sperm-bound immunoglobulins.
In addition, the presence of sperm-bound immunoglobulins was associated
with a poorer semen quality.
MATERIALS AND METHODS
Patients Eighty-four consecutive infertile men with palpable
varicoceles were evaluated. A detailed reproductive history was
obtained 10. All men studied had been attempting to conceive for
at least one year with partners of either previous proven fertility
or having had a negative prior infertility work-up. A complete
physical examination was performed in a warm room. A Test-Size
orchidometer (REMCAT Trade AB, Vallingby, Sweden) was used to
estimate testicular volume. Urine cultures were obtained for all
men entered into this study. Varicoceles were categorized by a
single examiner with the patient standing, as follows: grade I
varicoceles were small with a distinct impulse felt with Valsalva
maneuver, grade II varicoceles were moderate in size with a palpable
dilation detected with the Valsalva maneuver, grade III varicoceles
were large and easily palpable without the Valsalva maneuver,
grade IV varicoceles were easily visualized thru scrotal skin
without Valsalva maneuver. Semen Collection Semen was collected
by masturbation into sterile wide mouthed containers, after a
three day abstinence period. After liquefaction, the sample was
thoroughly mixed and divided into aliquots for semen analysis
and sperm-bound antibody assays. Semen Analysis Specimens were
examined within one hour of collection. Volume, color, pH and
time to liquefaction were recorded. Sperm counts were performed
with a Makler chamber (Sefi-Medical Instruments, Haifa, Israel).
Motility was evaluated by videomicrography. The presence of agglutination
was recorded. Morphology was assessed after preparation with the
Papanicolaou stain. The mean values of samples collected at least
one month apart was used for data analysis. Antisperm Antibody
Evaluation An enzyme-linked immunosorbent assay (ELISA) was performed
using sperm that were isolated from patients' ejaculate by a swim-up
technique11. Fresh ejaculates were allowed to liquefy at room
temperature for 20 minutes. An equal volume of phosphate-buffered
saline (PBS), warmed to 370 C, was carefully layered over the
semen, and the sample was incubated at 370 C for 60 minutes. Motile
sperm migrated into the PBS layer, and a visibly pure population
of viable sperm was obtained by removing the PBS with a Pasture
pipette. In patients with impaired sperm motility, spermatozoa
were collected by low-speed centrifugation. Purified spermatozoa
from patients and from fertile donors were diluted to 1x106/ml
with PBS, and 0.1 ml of each sample was added to six wells of
a microtiter plate. The sperm were pelleted by centrifugation,
the supernatants were removed, and the sperm irreversibly bound
to the wells by addition of 0.2 ml 0.25% glutaraldehyde in PBS.
After 10 minutes at room temperature, the glutaraldehyde solution
was removed, and the wells washed three times with 0.2-ml aliquots
of PBS plus 0.05% Tween 20 (PBS Tween). Alkaline phosphatase-conjugated
goat antibody to human IgG, IgA, or IGM was diluted 1:200 in PBS-Tween,
and 0.1 ml of each was added to two wells and to blank wells.
The plates were covered with parafilm and floated in a 370C water
bath. After 60-120 minutes, the wells were again washed three
times with PBS-Tween, and 0.2 ml of the alkaline phosphatase substrate,p-nitrophenyl
phosphate (1 mg/ml in diethanolamine buffer, pH 9.8) was added
to each well. After 60 minutes (IgG), 120 minutes (IgM), or 180
minutes (IgA), the optical densities at 405 nm of the solutions
in the wells were determined with an ELISA plate reader. A positive
reading was defined as a value at least twice the absorbance measured
in the wells with control sperm. This method has been shown to
give data relevant to infertility11 and comparable to the Immunobead
binding assay 12. The presence of any of the three immunoglobulin
isotypes placed a patient in group I: Sperm-bound immunoglobulins
present. Absence of all three isotypes placed the patient in group
II: Sperm-bound immunoglobulins absent. Serum antisperm antibody
was assayed utilizing the ELISA assay as follows. Aliquots of
sperm (5x105/0.05 ml) were pipetted into glass tubes containing
an equal volume of sera from the patient and a positive or negative
control, diluted 1:8, 1:16, 1:32, 1:64 in PBS. The samples were
incubated in a 370 C water bath for 60 min to allow sperm-reactive
antibodies to bind to the sperm surface. The sperm were then concentrated
by centrifugation, washed three times with PBS, and resuspended
in 0.5 ml PBS. Aliquots (0.1 ml) were then added to six wells
of a microtiter plate and fixed with glutaraldehyde. Alkaline
phosphatase-conjugated antibody to IgG, IgA, or IgM were each
added to two of the wells, and bound antibody was quantitated,
as described above. Hormone Analysis Serum levels of follicle
stimulating hormone (FSH), luteinizing hormone (LH) and prolactin
(PRL) were measured by a double antibody radioimmunoassay (RIA)
method (Diagnostic Products Corporation, Los Angeles, CA). The
lower limit of detection for FSH, LH and PRL ranged from 1 to
2 IU/ml. Testosterone (T) was measured by a similar RIA method
(Radioassay System Laboratories,Inc.). The lower limit of detection
using this method was approximately 50 ng/dl. Normal values in
our laboratory were as follows: T 300 to 1100 ng/dl, FSH 4 to
19 IU/ml, LH 8 to 18 IU/ml and PRL 7 to 18 IU/ml. Statistical
Methods: The mean, standard error of the mean were calculated.
Unpaired t tests and Chi-Square analysis were used where applicable.
RESULTS
Sperm-bound immunoglobulins were found in 27 of 84 (32%) infertile
men with palpable varicoceles. In these men, IgA was identified
in 85% of the samples, while bound IgM and IgG were present in
74% and 67% of the patients respectively. All three immunoglobulin
classes were present 56% (15/27) of the time. Two immunoglobulin
types were present 11% (3/27) of the time and only one class was
present 33% (9/27) of the time. When only one Immunoglobulin class
was present no specific Immunoglobulin type was most prevalent.
Men with sperm-bound immunoglobulins had a significantly decreased
sperm concentration, percent motile sperm, percent of sperm with
normal morphology and percent of sperm with tapering forms, when
compared with the group not having sperm-bound immunoglobulins
(Table 1). The presence of sperm-bound immunoglobulins was also
associated with a greater percent of head and tail defects. Agglutination
was observed in only 3 of the group I patients and 4 of the group
II patients. Fourteen of 27 men (52%) with sperm-bound immunoglobulins
and 13 of 57 men (23%) without sperm-bound immunoglobulins had
anti-sperm antibodies present in their sera. Serum antisperm antibodies
were thus present in 27 (32%) and absent in 57 (68%) of these
84 patients. The measured immunoglobulin classes were proportionally
the same as those bound to sperm. However, no statistical correlation
between circulating antisperm antibody and semen analysis was
found in this population. A progressive decline in semen quality
was noted when antisperm antibodies were present in serum, on
sperm or in both serum and on sperm respectively (Fig. 1). The
group with sperm-bound immunoglobulins had a mean age of 36.2
years not significantly different from a mean age of 35.7 years
for those without sperm-bound immunoglobulins. Testicular volume
was not significantly different between the groups. Right testicular
volume averaged 21.5 ± 0.8 cc while left testicular volume
averaged 21.3 ± 0.7 cc. There was no significant difference
in the size of the varicoceles between the two groups. A significant
decrease in FSH (p0.01), but not serum T, LH or PRL was found
in men with sperm bound immunoglobulins present. No urine cultures
were positive. DISCUSSION In the present study, immunoglobulins
bound to sperm-surface antigens were found in 32% of infertile
males with varicoceles. This compares to an incidence of 15% in
all infertile men referred to us for sperm-antibody testing 13.
In the infertile male population as a whole an immunological etiology
accounts for less than 10% of the total 14. Rumke and Hekman found
an incidence of antisperm antibodies of 3.3% in infertile men
and an incidence of 0% in their control group of fertile men 15.
Haas has found the prevalence of antisperm antibody to be 10%
in a normal male population 16. Our data suggests that men with
varicoceles may have an increased incidence of antisperm antibodies
bound to their ejaculated sperm. Although the absence of associated
pathology ( e.g., infection or obstruction) cannot be completely
excluded, the lack of leukospermia, positive urine cultures or
variation of physical exam between the groups makes this posibility
less likely. Of interest is the distribution of immunoglobulin
isotypes bound to sperm. In our study as well as in Golomb et
al8, IgA and IgM were the most commonly sperm-associated immunoglobulins.
IgM is a large immunoglobulin (900,000 daltons) and when formed
in the vascular space tends to stay there. It crosses capillary
boundaries with difficulty. It is thus tempting to speculate that
the measured IgM is produced locally and together with the locally
secreted IgA are formed in response to a local damage produced
by the varicocele. Some support for this comes from the recent
work by El-Demiry and James17 in which alterations in T-cell subsets
lining the genital tract are found with direct injury (e.g., obstruction
of the vas deferens). The occurrence of sperm-bound immunoglobulins
was associated with poor semen quality. Although the differences
were small they were statistically significant. Indeed, this relationship
was not found when only circulating antisperm antibodies was considered.
However, the semen quality was quantifiably similar to that of
the group I patients when both sperm-bound and circulating antibodies
were present. These findings suggest that men with varicoceles
who also have sperm-bound immunoglobulins, have more extensive
damage to the seminiferous epithelia than do men with varicoceles
who lack this finding. As noted previously, a progressive decline
in semen quality was noted when antisperm antibodies were also
present in serum, on sperm or in both serum and on sperm respectively
(Fig. 1). Thus the presence of circulating and sperm-bound antibodies
together might reflect continuing damage to the seminiferous tubular
epithelia. A direct comparison between our results and those of
Golomb et al8 cannot be made since their study was designed to
compare infertile men with and without varicoceles, and not as
in ours the the subgroup of infertile men with varicoceles with
or without sperm-bound antibody. Experimentally produced varicoceles
in rats and monkeys have been shown to damage both the Sertoli
cells and basal lamina of seminiferous tubules 18,19 These lesions
would allow immunocompetent cells to gain access to spermatozoal
antigens, resulting in activation of a sperm-specific immune response.
The slightly higher FSH level found in men negative for sperm-bound
immunoglobulins is surprising in view of the better semen analysis
in these patients. However, FSH is still well within the normal
range for both groups of patients. Serum antisperm antibodies
did not always predict the presence of sperm-bound immunoglobulins.
Only 54% of men with sperm-bound immunoglobulins had antisperm
antibodies detected in the serum. The predictive value of serum
antisperm antibody measurement, that is, the percent of the time
that sperm-bound immunoglobulins will be present when serum antibody
is detected, is 30% in Golomb et al's study 8 and 52% in data
from our patient population. Only the occurrence of sperm-bound
immunoglobulins and not circulating antisperm antibodies, correlated
with a poor semen quality. This data supports other recent studies
indicating that antibodies bound to sperm may be more relevant
to infertility than are circulating antisperm antibodies11,19,20
We recognize that the sperm-bound immunoglobulin identified in
these men may not always be due to the presence of antisperm antibodies.
Immunoglobulins21, or antigen-antibody complexes22, may non-specifically
adhere to sperm, especially sperm with impaired motility or altered
surface properties. The relation between sperm-bound immunoglobulins
and poorer semen quality and the high level of sperm-associated
IgM in some ejaculates, suggest that antigen nonspecific immunoglobulin
binding to sperm may have occurred in some of the samples. In
most of the men however, the data are consistent with specific
antibody binding to the surface of ejaculated sperm. IgA, the
isotype detected on sperm at the highest frequency in this study,
can be produced locally in the genital tract23, is not the immune
globulin present in the highest quantity in seminal fluid and
is also the most prevalent sperm-associated antibody isotype in
normospermic men with immune infertility11. In addition, the men
with sperm-bound immunoglobulins also had a higher incidence of
circulating antisperm antibodies then did men without immunoglobulins
on their sperm. This strongly indicates that at least in these
cases, antibodies to the patients sperm were being produced. Our
data suggests that varicocele-associated injury to the seminiferous
epithelium promotes induction of sperm-bound immunoglobulins and
a further impairment of fertility. Although we cannot formally
exclude an underlying defect in spermatogenesis that would contribute
to autoimmunity, decreased sperm quality and non-specific immunoglobulin
binding to the the sperm surface, the data also suggests that
sperm-bound immunoglobulins may play an etiologic role in varicocele
related infertility. Figure 1:
REFERENCES
1. Dubin L, Amelar RD: Etiologic factors in 1294 consecutive cases of male infertility. Fertil Steril 22:469, 1971 2. Stewart BH: Varicocele in infertility:Incidence and results of surgical therapy. J Urol 112:222, 1974 3. Saypol DC: Varicocele. J Androl 2:61,1981 4. Stewart BH: Varicocele in infertility: incidence and results of surgical therapy . J Urol 112:222, 1974 5. MacLeod J: Seminal Cytology in the presence of varicocele. Fertil Steril 16:735, 1965 6. MacLeod J: Further observations on the role of varicocele in human male infertility. Fertil Steril 20:545 , 1969 7. Pryor JL, Howards SS: Varicocele: Urol Clin North Am, 14:499, 1987 8. Golomb J, Vardinon N, Homonnai ZT, Braf Z, Yust I: Demonstration of antispermatozoal antibodies in varicocele related infertility with an enzyme linked immunosorbent assay (ELISA). Fertil Steril, 45:397, 1986 9. Ozen H, Asar G, Gungor S: Varicocele and antisperm antibodies. Int Urol Nephrol 17:97, 1985 10. Goldstein M: Evaluation of the Infertile Male. AUA Update Series, 3:(lesson 22), 1984 11. Witkin SS, David SS: Effect of sperm antibodies on pregnancy outcome in a subfertile population. Am J Obstet Gynecol 158:59, 1988 12. Bronson R, Cooper G, Rosenfeld D, Witkin SS: Detection of Spontaneously Occurring Sperm-directed Antibodies in Infertile Couples by Immunobead Binding and Enzyme-linked Immunosorbent Assay. Ann NY Acad Sci 438:504,1984 13. Witkin SS : Sperm-reactive antibodies as measured by enzyme-linked immunosorbent assay. In Perspectives in Immunoreproduction: Conception and Contraception, Edited by S Mathart and CM Fredricks, Hemisphere Publishing Corp.,Washington D.C., 1988 14. Hudson RW, Perez-Marrero RA, Crawford VA, McKay DE: Hormonal parameters in incidental varicoceles and those causing infertility. Fertil Steril 45:692 , 1986 15. Rumke P, Hekman A: Sterility: an immunologic disorder? Clin Obstet Gynecol 20:691,1977 16. Haas GG, Jr, Schreiber AD, Blasco L: The incidence of sperm-associated immunoglobulin and C3, the third component of complement, in infertile men. Fertil Steril 39:542,1983 17. El-Demiry, James K: Lymphocyte subsets and macrophages in the male genital tract in health and disease. Eur Urol 14:226, 1988 18. Carmignani G, Teddle G, DeStefani S: Experimental varicocele in the rat: A new experimental model. Urol Res 11:195, 1983 19. Haas GG Jr: The inhibitory effect of sperm-associated immunoglobulins on cervical mucus penetration. Fertil Steril 46:334, 1986 20. Haas GG Jr, Ausmanus M, Culp L: The effect of immunoglobulin occurring on human sperm in vivo on the human sperm/hamster penetration assay. Am J Reprod Immunol Microbiol 7:109, 1985 21. Richards JM , Witkin SS: Non-immune IgG binding to the surface of spermatozoa by disulfide rearrangement. Clin Exp Immunol 58:497, 1984 22. Witkin SS, Burgiovanni AM, Berkley A,Ledger WJ, Toth A: Detection and characterization of immune complexes in the circulation of infertile women. Fertil Steril 42:384, 1984 23. Witkin SS, Zelikovsky G, Good RA, Day NK: Demonstration of 11S IgA antibody to sperm in seminal fluid. Clin Exp Immunol 44:368,1981