THE ANATOMICAL RECORD 227:37-46 (1990)

Early Testicular Changes After Vasectomy and Vasovasostomy in Lewis Rats CHARLES J. FLICKINGER, JOHN C. HERR, STUART S. HOWARDS, JOHN R. SISAK, JANICE M. GLEAVY, TOD J. FUSIA, LISA D. VAILES, AND HAROLD H. HANDLEY Department of Anatomy and Cell Biology (C.J.F., J.C.H., J.R.S., L.D.V., H.H.H.), Department of Urology (S.S.H., J.M.G., T.J.F.), and the Cancer Center (J.C.H.), School of Medicine, University of Virginia Health Sciences Center, Charlottesville, Virginia 22908

ABSTRACT The testes of Lewis rats were studied a t intervals from 2 weeks to 3 months after bilateral vasectomy, vasectomy followed 1 month later by vasovasostomy, or sham operations. Aims were to determine the nature of early alterations after vasectomy, and to determine whether vasovasostomy after 1 month would result in reversal of vasectomy-induced changes. Approximately one-fourth of the testes in the vasectomy and vasovasostomy groups displayed histological changes, which consisted mainly of depletion of germ cells. The extent of the depletion varied greatly in different seminiferous tubules. In testes altered in this way, no abnormal infiltrations of lymphocytes, macrophages, or other cells were observed in the seminiferous epithelium or in the interstitium. The rete testis and straight tubules were normal in testes with altered seminiferous epithelium. A few testes in the vasectomy and vasovasostomy groups had necrotic centers. The results suggest that depletion of germ cells occurred as a result of shedding from the seminiferous epithelium into the lumen of the tubules. A cellular immune response, such as occurs in experimental allergic orchitis in other species, did not appear to be responsible for the observed loss of germ cells. This suggests a possible role for humoral antibody in this model, since there is a n association between testicular changes and serum antisperm antibodies at longer intervals after vasectomy. Testicular alterations were not reversed by performance of a vasovasostomy 1 month after vasectomy. Vasectomy continues to be a popular means of male contraception, and it is estimated that a s many a s 500,000 men undergo the procedure annually in the United States (Lepow and Crozier, 1979). Such widespread application has stimulated interest in the effects of the operation and in its potential reversibility. Studies in animals raised concern about possible systemic vascular effects of the antisperm antibodies that are induced by vasectomy (Bigazzi et al., 1976; Anderson et al., 1983; Alexander and Clarkson, 1978; Clarkson and Alexander, 19801, but these reports were balanced by controlled epidemiological studies which found no increased incidence of serious diseases in vasectomized men (Goldacre e t al., 1979; Walker et al., 1981; Petitti et al., 1982; Massey et al., 1984). In addition, a more recent study in monkeys showed no effect of vasectomy on development of atherosclerosis (Clarkson et al., 1988). Prominent among the remaining questions is the practicality of reversing a vasectomy in the event that a n individual wishes to regain his fertility at a later time. Refinements in surgical procedures for vasovasostomy (Howards, 1980) now permit urologists to restore anatomic patency of the vas deferens in 90% or more of men, but the proportion who become fertile is generally less (Silber, 1978; Lee and McLoughlin, 1980; Martin, 1981; Thomas et al., 1981). It has become 0 1990 WILEY-LISS, INC.

of interest, therefore, to investigate the extent to which changes that occur after vasectomy are reversed by vasovasostomy. We have been studying the effects of vasectomy and their reversal by vasovasostomy in a rat model system. The Lewis rat was chosen because these animals develop antisperm antibodies and display testicular alterations after vasectomy (Brannen et al., 1974; Bigazzi et al., 1977; Neaves, 1978). In a previous experiment (Flickinger et al., 1987), when testicular damage was observed 3-7 months after vasectomy, it was generally so extensive that only Sertoli cells and a few spermatogonia remained in the seminiferous epithelium. Other changes that might have indicated the basis for the depletion of germ cells were not observed. However, the presence of testicular alterations in Lewis rats was subsequently found to be correlated with serum antisperm antibody levels (Herr et al., 1987), and autoimmune testicular disease is known to occur after vasectomy in other species including guinea pigs (Tung, 1978), rabbits (Bigazzi et al., 19761, and monkeys (Tung and Alexander, 1980). Vasovasostomy performed 3 months after vasectomy did not appear to reverse the testicular changes, but vasovasostomy did prevent the progression of testicular alterations which Received June 1, 1989; accepted August 7, 1989.

38

C J . FLICKINGEK ET AL.

occurred in animals with a persisting vasectomy (Flickinger et al., 1987). In the present investigation, animals were studied at intervals between 2 weeks and 3 months after vasectomy, and vasovasostomies were performed 1 month after vasectomy. The first objective was to determine the nature of the testicular changes at early intervals, in an effort to achieve a better understanding of their pathogenesis and to obtain clues as to the mechanism of the alterations. The second objective was to determine whether shortening the interval between vasectomy and vasovasostomy from 3 months to 1 month would result in reversal of the vasectomy-induced alterations. MATERIALS AND METHODS Animals and Procedures

Male Lewis rats weighing 200-250 g were obtained from Harlan Sprague-Dawley (Indianapolis, IN). The animals were caged individually under a 14 h r 1ight:lO h r dark cycle and were allowed to acclimate to their surroundings for 2 weeks. Rats were then assigned to one of 3 groups. Those in the vasectomy group received a bilateral vasectomy. A bilateral vasectomy was followed 1month later by bilateral vasovasostomy for animals in the uasouasostomy group. A sham vasectomy and, 1 month later, a sham vasovasostomy were performed on rats in the sham group. Animals in the vasectomy and sham groups were killed 1, 2, and 3 months after the start of the experiment. Rats in the vasovasostomy group were killed at the 2 and 3 month intervals (1and 2 months after vasovasostomy). There were 25 rats in the vasectomy group, 15 in the vasovasostomy category, and 21 sham-operated animals, constituting 7-8 rats per interval per group. In a separate experiment to check for the possibility of very early alterations, 11 additional animals that received a bilateral vasectomy and 5 that received a sham vasectomy were killed 2 weeks after operation. For vasectomy, rats were anesthetized by inhalation of halothane. The operation was performed under aseptic conditions through a vertical midline abdominal incision. Each vas deferens was divided between two 4-0 silk ligatures approximately 1.5 cm proximal to the end of the vas a t the base of the bladder. The deferential vessels were ligated and divided along with the vas deferens. However, the scrotal contents were not disturbed during the operation, and special attention was paid to avoiding the spermatic vessels. The incision was closed in two layers, using 3-0 chromic catgut for the peritoneum and muscle and 4-0 silk sutures for the skin. The scrotal sacs were checked on alternate days for 2 weeks and then a t weekly intervals to ascertain that the animals did not become cryptorchid postoperatively. In the case of the sham vasectomies, a n identical procedure was employed, except that the vas deferens was not ligated or divided. A microscopic vasovasostomy was performed on rats in the vasovasostomy group one month after vasectomy. A transabdominal approach similar to that for vasectomy was utilized, under halothane anesthesia. The proximal and distal ends of the divided vas deferens were mobilized on each side, and if a spermatic granuloma was present it was excised. Bleeding from the cut ends of the proximal and distal vas deferens

was controlled with bipolar diathermy. With the aid of a Zeiss operating microscope (Eastern Microscope CO., Raleigh, NC) to visualize the procedure, the two vas segments were approximated with a Weck vasovasostomy clamp (Edward Weck, Inc., Raleigh, NC) and the vasal anastomosis was performed with Nylon sutures (10-0, Ethicon, Inc., Sommerville, NJ) and a TG 175-6 spatulated needle (Howards, 1980). Initially four transmural sutures were placed at 90" intervals around the circumference of the vas, and then 4-5 seromuscular sutures were placed between them. The abdominal wound was closed in two layers a s before. A sham vasovasostomy was performed on rats in the sham group, which had received a sham vasectomy 1 month previously. The sham operation was identical with that for vasovasostomy except that the intact vas deferens was isolated, and then the wound was closed. When animals were killed a t intervals of 1, 2, and 3 months, the flow of fluid through each vas deferens was tested in vitro (Carey e t al., 1988). A segment of vas about 1 cm long with the anastomosis a t its center was removed from vasovasostomized rats, and a similarly located portion of each vas deferens was obtained from sham-operated animals. The heat-drawn end of a piece of PE 90 tubing was inserted into the proximal end of each vas segment, which was suspended with a vas clamp (Weck, Inc.). The opposite end of the tubing was connected to a saline bag, which was elevated 90 cm or 40 cm above the vas, and the volume of saline that flowed through the vas in 2 min was recorded. Three determinations were made for each vas deferens. Histology

The animals were anesthetized with urethane, and tissue was fixed by vascular perfusion of fixative through the abdominal aorta (Forssmann et al., 1977). Blood vessels were rinsed by perfusion for 6-7 min with 1% procaine-HC1, for vasodilatation, and heparin 25 units/ml, for anticoagulation. The fixative solution, containing 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M cacodylate buffer pH 7.3, was subsequently perfused a t a rate of 12 ml/min for 20-25 min. The testes, epididymides, and spermatic granulomas were removed and the organs were weighed. Slices 2 mm thick, perpendicular to the long axis of the testis, were obtained from cephalic, middle, and caudal thirds of the organ. A slice was also obtained from the margin of the testis in the region of the rete testis. After rinsing in cacodylate buffer, the samples were dehydrated in a graded series of ethanols and embedded in glycol methacrylate (Historesin, LKB). Blocks were sectioned at 1-2 km using a Sorvall MT-2B microtome (Ivan Sorvall, Inc., Norwalk, CN) adapted for the use of glass knives of the Ralph type. Sections were mounted on slides and stained with toluidine blue. Qualitative evaluations of histological sections were supplemented by use of the semiquantitative testicular biopsy score count (TBSC) of Johnsen (1970) to estimate the extent of testicular alterations. For each of two blocks per testis, 50 consecutive seminiferous tubule profiles were scored on a scale of 1-10, depending on which cell types were present (Table l),and the individual tubular values were averaged to determine a score for each testis. It is to be emphasized that the testicular biopsy score count is not a quantitative as-

TESTICULAR CHANGES

39

Fig. 1. Light micrograph of a normal testis from a sham-operated animal. Cross sections of several seminiferous tubules display normal seminiferous epithelium in different stages of the spermatogenic cycle. x 130.

Fig. 2. Part of a normal testis from a rat 3 months after vasectomy. The appearance is similar to that of testes from sham-operated animals (Fig. 1). The seminiferous tubules contain stages of germ cells from spermatogonia through late spermatids. x 130.

TABLE 1. Testicular biopsy score count (TBSC)and criteria for scoring seminiferous tubule profiles (criteria modified from Johnsen, 1970)

TABLE 2. Distributions of testicular biopsy score counts (TBSC)'

Score 10:

Score 9: Score 8: Score 7 :

Score 6: Score 5: Score 4: Score 3: Score 2: Score 1:

Complete spermatogenesis with many late spermatids (elongate cells in the acrosome and maturation phases, i.e., step 8 and older spermatids); epithelium is of normal thickness with an open lumen Many late spermatids, but some germ cells sloughed and the lumen may be obliterated Only a few late spermatids present No late spermatids but many early spermatids (round spermatids in the Golgi and cap phases, i.e., steps 1-7 of spermiogenesis) No late spermatids and only a few early spermatids present No spermatids but many spermatocytes present Only a few spermatocytes and no spermatids present Spermatogonia are the only germ cells No germ cells, but Sertoli cells are present No cells in tubular section

TRSC

Period 1 month 2month

GrouD Sham VX Sham VX

3 month

vv

Sham VX vv

1-2 0 1 0 1 3 0 1 2

3-4 0 1 1 0 0 0 0 0

5-6 0 1 0 0 2 0 1 2

7-8 2 1 2 0 3 0 1

9-10 10 13 13 13 8 14 15

2

8

'Numbers of testes are shown. Vx, vasectomy group; Vv, vasovasostomy group.

sessment of spermatogenesis. It is a semi-quantitative scoring technique which represents the histological patterns that are present. The a i m of its use was to facilitate comparison between different specimens a n d to supplement subjective evaluations of morphology. Correlations between testis weight a n d testicular bi-

40

C.J. FLICKINGER ET AL.

Figs. 3-5.

TESTICULAR CHANGES

opsy score counts were carried out with a computer using a statistical program (STATPAK, Northwest Analytical, Portland, OR). Differences in the distributions of TBSC in the different groups were tested for significance by chi-squared analysis, using the same statistical package. RESULTS

Testicular Morphology-the Nature of Early Testicular Changes

Testes in most of the sham-operated animals had a normal morphology (Fig. 1). The seminiferous epithelium consisted of Sertoli cells accompanied by spermatogonia, spermatocytes, and early and late stages of spermatids. The specific combinations of germ cells varied according to the stage of the cycle of the seminiferous epithelium represented in a given section of a tubule. The interstitial tissue contained Leydig cells surrounding small blood vessels, macrophages, fibroblasts, and lymphatic spaces. The weight of the testes in the sham-operated animals was .31 .06 g per 100 g body weight (mean S.D.; n = 38 [body weight n.a. for 2 rats]). In accord with the predominantly normal appearance on qualitative evaluation, 37 of 42 testes from sham-operated animals had testicular biopsy score counts (TBSCs) of 9-10, and only one testis had a TBSC less than 7 (Table 2). Many testes from rats in the vasectomy and vasovasostomy groups also had a normal histological appearance (Fig. 2) and a TBSC of 9 or better. For descriptive purposes, small testes can be defined a s those weighing less than .20 gil00 g. (The rationale is that this value is 1.96 S.D. below the mean for the sham-operated sample, and the mean t 1.96 S.D. includes 95% of the sham-operated sample.) By this criterion, there were no abnormally small testes among those with a normal TBSC, since each weighed at least .20 g/100 g. Furthermore, considering all of the testes studied, testis weight was correlated with TBSC (r = .67, P < .001). In the vasectomy and the vasovasostomy groups, 72% testes were histologically normal (57 of 79, one specimen lost). When abnormal seminiferous tubules were observed, the alterations consisted of varying degrees of depletion of germ cells (Fig. 3), ranging from depletion or absence of late spermatids to very severely altered sem-

*

Fig. 3. An altered testis 1 month after vasectomy shows varying degrees of alterations in the seminiferous tubules. Some apparently normal profiles contain various types of germ cells, including many late spermatids (C). Others possess spermatocytes but lack spermatids (B). Very few germ cells remain in some tubules (A). x 150.

Fig. 4.Portion of a severely altered testis 2 months after vasectomy. Germ cells are greatly depleted, so that the seminiferous epithelium is populated mainly by Sertoli cells and some spermatogonia. The interstitial tissue in this region is prominent. Study of comparable fields at higher magnification revealed normal cellular constituents of the interstitium (Leydig cells, macrophages, fibroblasts). x 270. Fia. 5. Part of a moderately altered testis from a rat in the vasovasost&ny group a t the 3 month interval. The micrograph emphasizes the focal nature of the testicular changes. A band of obviously altered seminiferous tubules (A) is surrounded by tubules that are either normal or show only moderate changes. x 120.

41

iniferous tubules that contained only Sertoli cells and spermatogonia (Figs. 3, 4).The changes usually had a focal or regional character (Fig. 5 ) . Thus, altered testes usually displayed a mixture of normal areas, intermediate regions, and some severely altered tubules. In some instances, multinucleated (or “fused”) spermatids were present in the seminiferous tubules (Fig. 81, usually in a region that displayed a n intermediate degree of alteration. It should be stressed, however, that apart from Sertoli cells, various germ cells, and multinucleated spermatids, other cells were not present in the seminiferous epithelium. In particular, lymphocytes, macrophages, monocytes or neutrophils were not observed within the seminiferous tubules. Overall, 22 of 79 (28%) testes from animals in the vasectomy and vasovasostomy groups showed some alterations (TBSC less than 9) (Table 2). When altered testes a t sequential monthly intervals were compared, there was no obvious temporal progression in the nature of the changes. The interstitial tissue in sections of severely altered seminiferous tubules frequently appeared prominent because it contained many closely apposed cells (Fig. 4). This appearance may be due largely to shrinkage of seminiferous tubules, since we previously showed that the diameter of seminiferous tubules decreases and the relative volume of the interstitium increases in altered testes after vasectomy (Flickinger et al., 1987). While i t was difficult to identify every cell in altered testes, i t was clear that the great majority represented the normal cellular components of the interstitium, including Leydig cells, macrophages, and fibroblasts. Leydig cells had deeply staining cytoplasm and round or oval nuclei with small nucleoli and many small clumps of heterochromatin; macrophages were distinctive with their pale-staining cytoplasm and many cytoplasmic granules and vacuoles; fibroblasts were elongated cells with similarly shaped nuclei. We searched for but did not identify accumulations of mononuclear cells (e.g., lymphocytes, monocytes, plasma cells) surrounding either blood vessels or seminiferous tubules. Except for those testes described below, which appeared necrotic, neutrophils were not observed in the interstitium, and they were never seen in the seminiferous epithelium. Specimens from the rete region of altered testes contained normal rete testis (Fig. 6), lined by a cuboidal epithelium, and terminal segments of seminiferous tubules, in which the apical prolongations of tall epithelial cells projected toward the rete. The lumen of the rete was for the most part empty, containing only scattered sperm and a few immature germ cells. There was no influx of phagocytes or blood cells into the epithelium or the lumen of either the rete or the terminal segments of the seminiferous tubules, and no accumulations of blood cells were observed in the interstitial tissue near the rete or the terminal segments. A few testes from the animals in the vasectomy and vasovasostomy groups deserve special comment because of their unusual appearance. Histologically, their camules resembled the walls of spermatic aranulomas,&which are foci of chronic inflammation-that occur frequently around the proximal end of the vas deferens or near the epididymis after vasectomy (Bedford, 1976; Kennedy and Heidger, 1980; Flickinger et al., 1986). Proceeding inward, the margin of the testis ~~

42

C.J. FLICKINGEK ET AL.

Figs. 6-8

43

TESTICULAR CHANGES

TABLE 3. Fluid flow through segments of vas deferens Fluid flow (ml/minl 40 cm' 90 cm 1.8 O.Z3 3.7 0.3 1.6 0.9 3.1 I 1.6

Group Sham

*

*

VV

*

% patent

100 81

N2 31 26

'Elevation of the saline reservoir. 'Number of vasa succesfully tested; for technical reasons measurements were not obtained from a few vasa in each group. 3Mean 2 S.D. The groups were not significantly different by t-test (two tailed), using the Welch method for samples with unequal variances (Remington and Schork, 1985).

large number of multinucleated spermatids (Fig. 8). The remaining specimens were either completely normal or showed moderate changes that involved only a small fraction of the seminiferous tubules (TBSC 910). No abnormal infiltrations of cells in the interstitium or in the seminiferous tubules were observed. The region of the rete testis was studied in 14 of these specimens, and the rete, straight tubules, and terminal segments appeared normal. All 10 testes from the accompanying five sham-vasectomized rats were normal. Comparison of Vasectomy and Vasovasostomy Groups

Approximately 16% of the testes in the vasectomy group 1 to 3 months after operation had histological (Fig. 7) was composed of a layer of connective tissue, alterations. This is similar to the previous observation epithelioid cells, and debris mixed with mononuclear that about 20% testes showed alterations by 3 months cells and neutrophils. The interior of the testis ap- after vasectomy (Flickinger et al., 1987). peared necrotic. Seminiferous tubules were repreIt was evident from comparing the distribution of sented only by circular or oval profiles that contained TBSC (or testis weight/100 g) in the vasectomy and poorly defined remnants of cells. The interstitium con- vasovasostomy groups (Table 2) that performance of a tained debris and, in some instances, irregularly vasovasostomy a t 1month did not result in a n improveshaped globules of deeply staining material. Because of ment in testicular structure. Although there appeared the lack of cells, these testes received a score of 1in the to be a greater proportion of altered testes in the vasTBSC; since their weights were variable they repre- ovasostomy group at the 2 and at the 3 month intervals sented the few instances in which the testis weight did than in the vasectomy group, the numbers of testes not correlate with the TBSC. When these animals were involved in each instance were small and the distribukilled, it was uncertain from gross view whether there tions of TBSC between the groups were not signifiwas a spermatic granuloma adherent to the testis or cantly different, whether the analysis for the vasovathe entire organ was involved in a spermatic granu- sostomized animals considered all testes (n = 30) or loma. Testes of this description were found unilaterally only those connected to patent vasa ( n = 2 1 ; 5 not in one vasovasostomized animal at 2 months, in one patent, 4 n.a.1. vasovasostomized r a t at 3 months, and in one vasectoMean fluid flow through the vasa deferentia of vasmized r a t a t 3 months. In addition, a single testis in a ovasostomized rats did not differ significantly from vasovasostomized r a t at 2 months contained seminifer- that for sham-operated rats (Table 3). In vasovasosous tubules that uniformly lacked cells (TBSC of 1, tomized animals, 81% of vasa were patent (21126, 4 very low testis weight), although cells were present in n.a.1. The mean TBSC for testes from the five tracts in the interstitium and no reaction was visible beneath which fluid did not flow in vitro was less than that for the testicular capsule. testes with patent vasa (5.0 versus 8.4, P < .05). Testes 2 Weeks After Vasectomy

DISCUSSION

The most striking feature of the testicular alterations was that depletion of germ cells occurred with few associated histological changes. In general, germ cells simply seemed to disappear. Migration of phagocytic cells or any type of blood cells into the seminiferous epithelium was not observed; unusual numbers of degenerating germ cells were not detected in the seminiferous epithelium; and conspicuous accumulations of debris or lipid in Sertoli cells, which might indicate uptake of parts of germ cells in increased numbers by Fig. 6. The region of the rete testis 1 month after vasectomy. PorSertoli cells, were not encountered. These considertions of the rete (R) are lined by a regular cuboidal epithelium. Terminal segments of seminiferous tubules are visible, including one (TI ations suggest that the depletion of germ cells occurred as a result of their shedding into the lumen of the semwhich appears to be emptying into the rete within the plane of the section. The cells between the seminiferous tubules in regions such as iniferous tubules and transport into the excurrent duct this comprised normal constituents of the interstitium. No inflamma- system. Their subsequent fate is unknown, but, as for tory cells are present in the lumen of the rete or in the seminiferous spermatozoa after vasectomy (Barratt and Cohen, tubules. x 150. 1987; Flickinger, 1982), it could include any of the folFig. 7. Part of the margin of a necrotic testis from a vasovasoslowing; disposal in spermatic granulomas (Kwart and tomized rat a t 2 months. The organ is surrounded by connective tissue Coffey, 1973; Bedford, 1976), phagocytosis (Alexander, (C), epithelioid cells (E), and a layer (D) of mixed debris and cells, including neutrophils and macrophages. The interior of the testis (I) 1972; Flickinger, 1982) or dissolution (Galle and contains virtually no intact cells. Poorly defined circular outlines ap- Friend, 1977; Barratt and Cohen, 1987) in the lumen of parently represent the former locations of seminiferous tubules. the ducts, passage to the lymphatic system (Ball and x 125. Setchell, 1983; Barratt and Cohen, 1986), or release to the outside (in the case of vasovasostomized animals). Fig. 8. A seminiferous tubule in an altered testis contains large Multinucleate spermatids were present in some tesmultinucleated spermatids (arrows). x 280.

The earliest interval examined in the main part of the study was 1 month after vasectomy. To search for even earlier stages of alterations, testes were obtained from 11 additional animals 2 weeks after vasectomy. Of the 22 testes, one showed severe depletion of germ cells (TBSC 2.6), and a second (TBSC 5.9) contained a

~~

~

44

C.J. FLICKINGEK ET AL.

tes, and they were most often encountered when tubules were altered to a n intermediate degree. They may form as a result of fusion of spermatids or expansion of cytoplasmic bridges between cohorts of germ cells. In any case, they have been observed in testes subjected to a variety of deleterious conditions (Holstein and Eckmann, 1986). Thus, the presence of multinucleated spermatids is neither specific for effects of vasectomy nor is it, by itself, helpful in ascertaining the mechanism of the testicular lesions. Antisperm antibodies were correlated with the presence of testicular alterations in Lewis rats studied at intervals up to 7 months after vasectomy (Herr et al., 1987). This association and the occurrence of autoimmune testicular disease after vasectomy in other species (Bigazzi et al., 1976; Tung, 1978; Tung and Alexander, 1980) led us to inquire whether features of autoimmune (allergic) orchitis could be detected a t early intervals after operation. Although this condition has not been as extensively studied in the rat as in other species, experimental allergic orchitis (EAO) has been induced in experimental animals by immunization with testicular components (Tung et al., 1970; Tung, 1983; Tung and Menge, 1985). In the guinea pig, the histopathological characteristics of EAO in the testis include 1)degenerative changes of spermatids and exfoliation of germinal epithelium; 2) mononuclear infiltrative lesions in the testis; and 3) later, complete aspermatogenesis with severe atrophy of seminiferous tubules and interstitial fibrosis (Tung et al., 1970; Tung and Menge, 1985). In the mouse, early testicular lesions are characterized by peritubular andlor intratubular accumulation of eosinophils, neutrophils, lymphocytes, and macrophages; this is followed by aspermatogenesis, and late changes include widespread necrosis and extensive fibrosis of seminiferous tubules (Kohno et al., 1983; Sato et al., 1981). The majority of the altered testes in the present study and in our previous longer-term experiment (Flickinger et al., 1987) had some of these features, most notably depletion and exfoliation of germ cells, and in some cases severe atrophy of the seminiferous tubules. However, depletion of germ cells occurs in a variety of conditions and by itself is not diagnostic of allergic orchitis (Tung and Menge, 19851, and the more specific features of EAO such a s infiltrates of various types of blood cells were lacking in the altered testes in the present study (with the qualification discussed below). These observations of the present study are in agreement with those of Neaves (19781, who studied Lewis rat testes 3 months after vasectomy and also found no indication of a cellmediated immune response. However, it might be thought that the specimens described in this and in previous studies on the Lewis rat (Flickinger et al., 1987; Neaves, 1978) represented late changes and that transient cellular infiltrates a t a n earlier time (e.g., Tung et al., 1970) were missed. This possibility cannot be excluded, but i t is notable that none of the testes examined 2 weeks after vasectomy showed such changes, which do occur within about 10-25 days when allergic orchitis is induced in other species by immunization (Kohno et al., 1983; Tung e t al., 1970). It was of particular interest to examine the region of the rete testis, straight tubules, and terminal segments of seminiferous tubules, because early lesions in adop-

tive transfer of EAO in mice consist of infiltrates of lymphocytes and macrophages in this area (MahiBrown et al., 1987). However, the region of the rete was normal in the present study. In actively induced EAO in mice, on the other hand, inflammation involving seminiferous tubules under the testicular capsule was common (Tung et al., 1987). While altered seminiferous tubules were observed in the periphery of the testis in the present study, numerous altered tubules were also present in the interior. The apparent testicular necrosis in three testes in the present study is difficult to interpret in relation to the depletion of germ cells observed in the majority of the altered testes. The extent of the changes seen in the former suggests a catastrophic event, which may be distinct from the other graded changes observed. There was no apparent temporal progression from one to the other, and it seems unlikely that the extreme lesions could progress to the other images a t a later time because virtually all the cells appeared to be destroyed. Conversely, in most cases even when testes were greatly depleted of germ cells, Sertoli cells and a few spermatogonia survived and there was no granulomalike reaction a t the testicular periphery. One possibility is that rupture of the epididymis occurred in the animals with unusual testes, and that the resulting inflammatory reaction and granuloma formation came to involve not only the epididymis but also the nearby testis. This interpretation is in accord with the gross observation that these testes appeared to be involved in a large granuloma. However, the possibility of a severe inflammatory reaction directly involving these testes cannot be excluded. It is unknown whether these lesions are related to those that involve the testicular periphery in actively induced EAO in the mouse (Tung et al., 1987). In summary, morphological evidence of autoimmune orchitis depends largely on detection of a cellular response. The lack of a cellular reaction in most altered testes, coupled with the association between testicular lesions and elsvated antisperm antibodies (Herr et al., 19871, may be interpreted as points in favor of a role for humoral antibody in production of testicular changes in this model. The three necrotic testes constitute a possible exception, but their relation to the more prevalent type of alteration is problematic. Clearly, more immunopathological information is needed. Studies of the epididymides of rats a t both short and long intervals after vasectomy and vasovasostomy are in progress and may aid in interpreting the overall nature of the changes, because lesions in the duct system are associated with autoimmune orchitis in some species (Tung and Menge, 1985). The Efficacy of Vasovasostomy in Reversing Vasectomy-Induced Changes

In a previous study (Flickinger et al., 1987) when vasovasostomies were performed 3 months after vasectomy, approximately 20% testes were altered in both vasectomy and vasovasostomy groups a t 3 -4 months. By 7 months, however, the fraction of altered testes had increased to 70% in rats with a persisting vasectomy, while the proportion of altered testes remained at about 20% when a vasovasostomy was performed a t 3 months. These data suggested that vasovasostomy

TESTICULAR CHANGES

did not result in reversal of testicular alterations induced by vasectomy, but that a vasovasostomy did prevent the progression of testicular changes. This raised the question of whether vasectomy-induced alterations might be reversed by performing a vasovasostomy at an earlier interval. The results of the present study indicate that vasovasostomy 1 month after vasectomy did not reverse vasectomy-induced testicular changes, because the proportion of altered testes in the vasovasostomy group was at least as high as in the vasectomy group at the 2 and 3 month intervals. Indeed, the appearance of the testes was poorer in the vasovasostomized animals than in those with a persisting vasectomy, although it is uncertain as to how much significance should be attached to this observation because the numbers of testes involved were small. Nevertheless, this observation is interesting in the light of suggestions that under certain circumstances vasovasostomy may serve as an immunologic stimulus (Parslow et al., 1983; Herr et al., 1989). For example, we found that vasovasostomy failed t o lead to a decline in serum antisperm antibodies in Lewis rats (Herr et al., 1987) and even resulted in an increase in antisperm antibodies in SpragueDawley rats (Herr et al., 1989). ACKNOWLEDGMENTS

This research was supported by grant No. HD18825 from the NIH. LITERATURE CITED Alexander, N.J. 1972 Vasectomy: Long term effects in the rhesus monkey. J . Reprod. Fertil., 31t399-406. Alexander, N.J., and T.B. Clarkson 1978 Vasectomy increases the severity of diet-induced atherosclerosis in Macaca fascicularis. Science, 201 538-541. Anderson, D.J., N.J. Alexander, D.L. Fulgham, and J.L. Palotay 1983 Spontaneous tumors in long-term vasectomized mice: Increased incidence and association with antisperm immunity. Am. J. Pathol., 111t129-139. Ball, R.Y., and B.P. Setchell 1983 The passage of spermatozoa to regional lymph nodes in testicular lymph following vasectomy in rams and boars. J . Reprod. Fertil., 68t145-153. Barratt, C.L.R., and J . Cohen 1986 Fate of superfluous sperm products after vasectomy and in the normal male tract of the mouse. J. Reprod. Fertil., 78:l-10. Barratt, C.L.R., and J. Cohen 1987 Quantitation of sperm disposal and phagocytic cells in the tract of short- and long-term vasectomized mice. J . Reprod. Fertil., 81t377-384. Bedford, J.M. 1976 Adaptations of the male reproductive tract and the fate of spermatozoa following vasectomy in the rabbit, rhesus monkey, hamster and rat. Biol. Reprod., 14t118-142. Bigazzi, P.E., L.L. Kosuda, K.C. Hsu, and G.A. Andres 1976 Immune complex orchitis in vasectomized rabbits. J . Exp. Med., 143t382404. Bigazzi, P.E., L.L. Kosuda, and L.L. Harnick 1977 Sperm autoantibodies in vasectomized rats of different inbred strains. Science, 197:1282-1283. Brannen, G.E., A.M. Kwart, and D.S. Coffey 1974 Immunological implications of vasectomy: I. Cell mediated immunity. Fertil. Steril., 25508-514. Carey, P.O., S.S. Howards, C.J. Flickinger, J.C. Herr, T.N. Gallien, D. Caloras. and D.R. Sue11 1988 Effects of granuloma formation a t site of vasovasostomy. J. Urol., 139t8531856. Clarkson, T.B., and N.J. Alexander 1980 Vasectomy: Effects on the occurrence and extent of atherosclerosis in rhesus monkeys. J. Clin. Invest., 65t15-25. Clarkson, T.B., N.J. Alexander, and T.M. Morgan 1988 Atherosclerosis of cynomolgus monkeys hyper- and hyporesponsive to dietary cholesterol. Lack of effect of vasectomy. Arteriosclerosis, 8t488498. Flickinger, C.J. 1982 The fate of sperm after vasectomy in the hamster. Anat. Rec., 202t231-239.

45

Flickinger, C.J., E.S. Yarbro, S.S. Howards, J.C. Herr, D. Caloras, T.N. Gallien, and D.R. Spell 1986 The incidence of spermatic granulomas and their relation to testis weight after vasectomy and vasovasostomy in Lewis rats. J . Androl., 7t285-291. Flickinger, C.J., J.C. Herr, S.S. Howards, D. Caloras, E.S. Yarbro, D.R. Spell, and T.N. Gallien 1987 The influence of vasovasostomy on testicular alterations after vasectomy in Lewis rats. Anat. Rec., 217:137-145. Forssmann, W.G., S.Ito, E. Weihe, A. Aoki, M. Dym, and D.W. Fawcett 1977 An improved perfusion fixation method for the testis. Anat. Rec., 188t307-314. Galle, J., and D.S. Friend 1977 The fate of sperm after vasectomy in the guinea pig: Fine structure, freeze-fracture, and cytochemistry. Lab. Invest., 37r79-95. Goldacre, M., M. Vessey, J. Clarke, and M. Heasman 1979 Record linkage study of morbidity following vasectomy. In: Vasectomy: Immunologic and Pathophysiologic Effects in Animals and Man. I.H. Lepow and R. Crozier, eds. Academic Press, New York, pp. 567-575. Herr, J.C., C.J. Flickinger, S.S. Howards, D. Caloras, E.S. Yarbro, D.R. Spell, and T.N. Gallien 1987 The relation between antisperm antibodies and testicular alterations after vasectomy and vasovasostomy in Lewis rats. Biol. Reprod., 37t1297-1305. Herr, J.C., S.S. Howards, D.R. Spell, P.O. Carey, S.J. Kendrick, T.N. Gallien, H.H. Handley, and C.J. Flickinger 1989 The influence of vasovasostomy on antisperm antibodies in rats. Biol. Reprod., 40t353-360. Holstein, A.F., and C. Eckmann 1986 Multinucleated spermatocytes and spermatids in human seminiferous tubules. Andrologia, 18: 5-16. &wards, S.S. 1980 Vasovasostomy. In: The Craft of Urologic Surgery. D.G. Skinner and G.M. Gloege, eds. Urol. Clin. North Am., Vol. 7, pp. 165-169. Johnsen, S.G. 1970 Testicular biopsy score count-A method for registration of spermatogenesis in human testes: Normal values and results in 335 hypogonadal males. Hormones, 1t2-25. Kennedy, S.W., and P.M. Heidger, J r . 1980 Fine structure of the spermatic granuloma of the rat vas deferens following vasectomy. Anat. Rec., 198t461-474. Kohno, S., J.A. Munoz, T.M. Williams, C. Teuscher, C.C.A. Bernard, and K.S.K. Tung 1983 Immunopathology of murine experimental allergic orchitis. J. Immunol., 130:2675-2682. Kwart, A.M., and D.S. Coffey 1973 Sperm granulomas: An adverse effect of vasectomy. J . Urol., 11Ot416-422. Lee, L., and M.G. McLoughlin 1980 Vasovasostomy-Comparison of macroscopic and microscopic techniques at one institution. Fertil. Steril., 33t54-55. Lepow, I.H., and R. Crozier, eds. 1979 Vasectomy: Immunologic and Pathophysiologic Effects in Animals and in Man. Academic Press,-New York. Mahi-Brown. C.A.. T.D. Yule. and K.S.K. Tuna 1987 Adoptive transfer of murine ‘autoimmune orchitis to n a h e recipients with immune lymphocytes. Cell. Immunol., 106;408-419. Martin, D.C. 1981 Microsurgical reversal of vasectomy. Am. J . Surg., 142:48-50. Massey, F.J., G.S. Bernstein, W.M. O’Fallon, L.M. Schuman, et al. (total of 21 authors) 1984 Vasectomy and health. Results from a large cohort study. J A M A, 252t1023-1029. Neaves, W.B. 1978 The effect of vasectomy on the testis of inbred Lewis rats. J . Reprod. Fertil., 54t405-411. Parslow, J.M., M.G. Royle, M.M.B. Kingscott, D.M.A. Wallace, and W.F. Hendry 1983 The effects of sperm antibodies on fertility after vasectomy reversal. Am. J . Reprod. lmmunol., 3:28-31. Petitti, D.B., R. Klein, H. Kipp, W. Kahn, A.B. Siegelaub, and G.D. Friedman 1982 A survey of personal habits, symptoms of illness, and histories of disease in men with and without vasectomies. Am. J. Public Health, 72:476-480. Remington, R.D., and M.A. Schork 1985 Statistics With Applications to the Biological and Health Sciences. Prentice-Hall, Englewood Cliffs, N.J. Sato, K., K. Hirokawa, and S. Hatakeyama 1981 Experimental allergic orchitis in mice. Virchows Arch. LA], 392t147-158. Silber, S.J. 1978 Vasectomy and vasectomy reversal. Fertil. Steril., 29t125-140. Thomas, A.J., Jr., J.E. Pontes, N.R. Rose, S. Segal, and J.M. Pierce, J r . 1981 Microsurgical vasovasostomy: Immunologic consequences and subsequent fertility. Fertil. Steril., 35t447-450. Tung, K.S.K. 1978 Allergic orchitis lesions are adoptively transferred from vasoligated guinea pigs to syngeneic recipients. Science, 201t833-835. Tung, K.S.K. 1983 Models of autoimmunity to spermatozoa and testis.

46

C.J. FLICKINGEI; ET AL

In: Immunology of Reproduction. T.G. Wegmann and T.J. Gill, eds. Oxford University Press, New York, pp. 387-423. Tung, K.S.K., and N.J. Alexander 1980 Monocytic orchitis and aspermatogenesis in normal and vasectomized rhesus macaques (Macaca mulatta). Am. J. Pathol., 101:17-27. Tung, K.S.K., and A.C. Menge 1985 Sperm and testicular autoimmunity. In: The Autoimmune Diseases. N.R. Rose and I.R. MacKay, eds. Academic Press, New York, pp. 537-590. Tung, K.S.K., E.R. Unanue, and F.J. Dixon 1970 The immunopathol-

ogy of experimental allergic orchitis. Am. J. Pathol., 60:313328. Tung, K.S.K., T.D. Yule, C.A. Mahi-Brown, and M.B. Listrom 1987 Distribution of histopathology and Ia positive cells in actively induced and passively transferred experimental autoimmune orchitis. J. Immunol., 138t752-759. Walker, A.M., H. Jick, J.R. Hunter, A. Danford, and K.J. Rothman 1981 Hospitalization rates in vasectomized men. J A M A, 245: 2315 -23 17.