Investigation of detoxification capacity of rat testicular germ cells and Sertoli cells

Isolated pachytene spermatocytes liver longer than round spermatids in vitro. Indigenous formation of oxygen-derived free radicals and hydrogen peroxide can cause damage to germ cells. The germ cell antioxidant capacity may play an important role in this respect. In view of this, we have examined the activity and cellular localization of superoxide dismutase (SOD) and glutathione S-transferases (GST) in rat testicular cells. We have found significant differences in the distribution of these enzymatic activities in the germ cells. In addition, this study shows that alpha-tocopherol is found in various amounts in rat testicular cells in the order of: Sertoli cells greater than pachytene spermatocytes greater than round spermatids, with a factor of 4 in the alpha-tocopherol content between Sertoli cells and round spermatids.     

Immunohistochemical study of calmodulin in developing mouse testis

The purpose of this study was to determine the localization of calmodulin in the developing mouse testis by the indirect immunoperoxidase method. In addition, the amount of calmodulin in pachytene spermatocytes, spermatids, and residual bodies isolated from the mouse testis and epididymal spermatozoa was quantitated by the adenylate cyclase activation assay and by enzyme immunoassay. The relative levels of calmodulin in the developing mouse testis and in the isolated testicular germ cells were confirmed by western transfer staining. The level of immunoreactive calmodulin was very low in the testes from immature animals. In testes from the mature mouse, calmodulin was found to be localized in spermatocytes and spermatids, but was not found in spermatogonia, Sertoli cells, and interstitial cells. By contrast, immunochemical staining of tubulin was extremely intense in Sertoli cells. Biochemical determinations also showed that pachytene spermatocytes, round spermatids, spermatozoa, and residual bodies contained 14.9 micrograms, 15.8 micrograms, 2.3 micrograms and 5.2 micrograms of calmodulin per mg of protein, respectively. Both the immunochemical and the biochemical studies revealed that levels of calmodulin were high in the spermatocytes and in the round spermatids, as compared to the level in spermatozoa. This fact strongly suggests that the large amount of calmodulin in mammalian testes may be associated primarily with meiotic divisions and/or spermatogenesis.     

Determination of daily sperm production in stallion testes by enumeration of germ cells in homogenates

Thirty adult stallion testes were selected with high (n = 15) and low (n = 15) Daily Sperm Production (DSP)/testis. Parenchymal samples were prepared for morphometric analysis, and the numbers of germ cells and Sertoli cells were determined. Testicular samples were homogenized, and germ cells and Sertoli cells were enumerated using phase contrast microscopy. Numbers of germ cells and Sertoli cells and potential DSP during spermatogenesis were determined. Significant correlations existed between morphometric and homogenate determinations of number per testis of preleptotene, leptotene plus zygotene primary spermatocytes (r = 0.58; P < 0.001), pachytene plus diplotene primary spermatocytes (r = 0.67; P < 0.0001), all primary spermatocytes (r = 0.67; P < 0.0001), round spermatids (r = 0.72; P < 0.0001), and Sertoli cells (r = 0.70; P < 0.0001). Significant correlations (P < 0.0001) existed between morphometric and homogenate determination of DSP/testis based on preleptotene, leptotene plus zygotene primary spermatocytes (r = 0.78), pachytene plus diplotene primary spermatocytes (r = 0.88), and round spermatids (r = 0.85). Using morphometric determination as the standard, the sensitivity (i.e., ability to detect low DSP/testis) and specificity (i.e., ability to detect high DSP/testis) by homogenate enumeration of germ cells was 81 and 93% for round spermatids, 100 and 24% for pachytene plus diplotene primary spermatocytes, and 67 and 87% for preleptotene, leptotene plus zygotene primary spermatocytes, respectively. Enumeration of primary spermatocytes in homogenates was less accurate than enumeration of round or elongated spermatids. Enumeration of round and elongated spermatids in homogenates was a rapid and useful method for determining DSP in horses, and it may prove to be a useful technique for quantitating potential DSP from testicular biopsies.     

Activité aromatase dans les cellules germinales mâles: quelle signification fonctionnelle?

The ability of the male gonad to convert androgens into estrogens is well known. According to age, aromatase activity has been already measured in immature and mature rat Leydig cells as well as in Sertoli cells. Recently, in different studies, a cytochrome P450_arom has even been immunolocalized not only in Leydig cells but also in germ cells of mouse, brown bear and rooster whereas in pig, ram and human the aromatase is mainly present in Leydig cells. Our purpose was to investigate the testicular cell distribution of cytochrome P450_arom mRNA in adult rat using RT-PCR. With 2 highly specific primers located on exons 8 and 9, we have been able to amplify a 289 bp aromatase fragment not only in Leydig cells and Sertoli cells but more importantly in highlyenriched preparations of pachytene spermatocytes, round spermatids and testicular spermatozoa. These amplified products showed 100% homology with the corresponding fragment of the rat ovary cDNA. In parallel, using an anti-human cytochrome P450_arom antibody we have demonstrated the presence of a 55 kDa protein in seminiferous tubules and crude germ cell (pachytene spermatocytes and round spermatids) preparation of the mature rat. After incubation with tritiated androstenedione, the aromatase activities in the microsomal fractions were 3.12±0.19 pmoles/mg/h in the testis, 1.25±0.13 in the seminiferous tubules and 1.53±0.15 in the crude germ cells. In purified testicular spermatozoa the aromatase activity was 2.96±0.69 pmoles/mg/h and found to be 5-fold higher when compared to that of either purified pachytene spermatocytes or round spermatids. Using a quantitative RT-PCR method with a standard cDNA 29 bp shorter, we have compared the amount of cytochrome P450_arom mRNA in mature rat Leydig cells and Sertoli cells. In purified Leydig cells from 90 day-old rats the P450_arom mRNA level was: 36.2±3.4×10^?3 amoles/μg RNA whereas in Sertoli cells the mRNA level was 10 fold lower. In pachytene spermatocytes, round spermatids and testicular spermatozoa the P450_arom mRNA levels were re pectively 367.2±76.6, 117.6±22.0 and <1×10^?3 amole/μg RNA. In conclusion we have demonstrated that the P450 aromatase is present not only in Sertoli cells and Leydig cells from mature rat testis but a biologically active aromatase exists also in germ cells (pachytene spermatocytes, round spermatids and spermatozoa). The existence of an additional source of estrogens within the genital tract of the male is now well documented and that suggests a putative role for these hormones during the male germ cell development.     

Stimulation of rat Sertoli cell secretory activity in vitro by germ cells and residual bodies

The direct influence of germ cells and residual bodies on Sertoli cell basal and FSH-stimulated secretion of androgen-binding protein (ABP) was studied using Sertoli cells, recovered from 20-day-old rats, cultured alone or cocultured with a crude germ cell preparation from adult rats or with pachytene spermatocytes, round spermatids or populations of residual bodies enriched by centrifugal elutriation. The effect of a rat liver epithelial cell line (LEC) on Sertoli cell function was also tested. Addition of a crude germ cell preparation increased basal and FSH-stimulated ABP secretion. Pachytene spermatocytes and residual bodies adhered to the Sertoli cell monolayer to a much greater extent than did round spermatids. Addition of pachytene spermatocytes markedly enhanced basal and FSH-stimulated ABP secretion over 12 days of culture. Round spermatids and residual bodies stimulated ABP secretion although to a lesser extent than did spermatocytes. Furthermore, the increase of FSH-stimulated ABP levels was not maintained after 4 or 8 days of culture. LEC also enhanced basal and FSH-induced ABP levels but the increase of FSH-induced ABP production was only observed until Day 8 of culture. The influence of LEC on Sertoli cell secretion could be mediated through the production of an extracellular matrix. It is concluded that germ cells, particularly pachytene spermatocytes, can directly stimulate Sertoli cell secretory activity in vitro.     

Immunolocalization of inhibin and activin alpha and betaB subunits and expression of corresponding messenger RNAs in the human adult testis

Inhibin B is a testicular peptide hormone that regulates FSH secretion in a negative feedback loop. Inhibin B is a dimer of an alpha and a beta(B) subunit. In adult testes, the cellular site of production is still controversial, and it was hypothesized that germ cells contribute to inhibin B production. To determine which cell types in the testes may produce inhibin B, the immunohistochemical localization of the two subunits of inhibin B were examined in adult testicular biopsies with normal spermatogenesis, spermatogenic arrest, or Sertoli cell only (SCO) tubules. Moreover, using in situ hybridization with mRNA probes, the mRNA expression patterns of inhibin alpha and inhibin/activin beta(B) subunits have been investigated. In all testes, Sertoli cells and Leydig cells showed positive immunostaining for inhibin alpha subunit and expressed inhibin alpha subunit mRNA. Using inhibin beta(B) subunit immunoserum on testes with normal spermatogenesis and with spermatogenic arrest, intense labeling was located in germ cells from pachytene spermatocytes to round spermatids but not in Sertoli cells. Inhibin beta(B) subunit mRNA expression was intense in germ cells from spermatogonia to round spermatids and in Sertoli cells in these testes. In testes with SCO, high inhibin beta(B) subunit mRNA labeling density was observed in both Sertoli cells and Leydig cells, whereas beta(B) subunit immunostaining was negative for Sertoli cells and faintly positive for Leydig cells. These results agree with the recent opinion that inhibin B in adult men is possibly a joint product of Sertoli cells and germ cells.     

DNMT1 and HDAC1 gene expression in impaired spermatogenesis and testicular cancer

DNA methylation catalyzed by DNA methyltransferases (DNMTs) and histone deacetylation catalyzed by histone deacetylases (HDACs) play an important role for the regulation of gene expression during carcinogenesis and spermatogenesis. We therefore studied the cell-specific expression of DNMT1 and HDAC1 for the first time in human testicular cancer and impaired human spermatogenesis. During normal spermatogenesis, DNMT1 and HDAC1 were colocalized in nuclei of spermatogonia. While HDAC1 was additionally present in nuclei of Sertoli cells, DNMT1 was restricted to germ cells exhibiting a different expression pattern of mRNA (in pachytene spermatocytes and round spermatids) and protein (in round spermatids). Interestingly, in infertile patients revealing round spermatid maturation arrest, round spermatids lack DNMT1 protein, while pachytene spermatocytes became immunopositive for DNMT1. In contrast, no changes in the expression pattern could be observed for HDAC1. This holds true also in testicular tumors, where HDAC1 has been demonstrated in embryonal carcinoma, seminoma and teratoma. Interestingly, DNMT1 was not expressed in seminoma, but upregulated in embryonal carcinoma. Olufunmilade A. Omisanjo is a scholarship holder of the German Academic Exchange Service (DAAD). Sonja Hartmann is a member of the German Research Foundation (DFG) Research Training Group 533 Cell–cell-Interaction in Reproduction.     

Pachytene spermatocyte proteins influence Sertoli cell function

Isolated Sertoli cells were cultured on MatrigelTM-coated Millipore filters in bicameral chambers. The Sertoli cells form confluent epithelial sheets that, by virtue of the Sertoli cell tight junctions, form transepithelial permeability barriers between the apical and basal domains of the cells. These Sertoli cells secrete metabolically labeled proteins in a polarized manner. Three peptides, P1 (pI = 4.5-5.0, MW = 70,000), P2 (pI = 4.5-5.0, MW = 50,000), and P3 (pI = 4.0-4.7, MW = 34,000) are secreted apically from the epithelial sheets of Sertoli cells and are not found in basal secretions from the same Sertoli cells. Pachytene spermatocyte-conditioned medium contains proteins released from the germ cells that are uniquely different from the Sertoli cell-secreted proteins. Addition of the pachytene spermatocyte-conditioned medium to the apical reservoir of the bicameral chambers over an epithelial sheet of Sertoli cells stimulated the synthesis and secretion of total protein, transferrin, and specifically induced peptides S1 and S2 from Sertoli cells. As controls, conditioned medium from 3T3 fibroblasts and round spermatids did not stimulate the Sertoli cells. Hence, the ability of pachytene spermatocyte proteins to induce specific Sertoli cell secretion indicates that the pachytene spermatocytes are able to influence their surrounding milieu, and provides further support to the concept of a paracrine interaction between germ cells and Sertoli cells during spermatogenesis.     

beta-Nerve growth factor participates in an auto/paracrine pathway of regulation of the meiotic differentiation of rat spermatocytes

NGF appears to be involved in spermatogenesis. However, mice lacking NGF or TrkA genes do not survive more than a few days whereas p75(NTR) knockout mice are viable and fertile. Therefore, we addressed the effect of betaNGF on spermatogenesis by using the systems of rat germ cell culture we established previously. betaNGF did not modify the number of Sertoli cells, pachytene spermatocytes, secondary spermatocytes nor the half-life of round spermatids, but increased the number of secondary meiotic metaphases and decreased the number of round spermatids formed in vitro. These effects of betaNGF were reversible and maximal at about 4 x 10(-11) M. Conversely, K252a, a Trk-specific kinase inhibitor, enhanced the number of round spermatids above that of control cultures. The presence of betaNGF and its receptors TrkA and p75(NTR) was investigated in testis sections, in Sertoli cell and germ cell fractions, and in germ cell and Sertoli cell co-cultures. betaNGF was detected only in germ cells from pachytene spermatocytes of stages VII up to spermatids of stages IX-X. TrkA and p75(NTR) were detected in Sertoli cells and in these germ cells. Taken together, these results indicate that betaNGF should participate in an auto/paracrine pathway of regulation of the second meiotic division of rat spermatocytes in vivo.     

Human spermatogenic cell marker proteins detected by two-dimensional electrophoresis

Highly homogeneous populations of human pachytene spetmatocytes and round spermatids have been obtained from normal adult testis using unit gravity (STA-PUT) sedimentation. Contaminating Leydig cells have been removed by density centrifugation in discontinuous Percoll gradients to yield resultant germ cell purities of 90–95% for pachytene spermatocytes and 89–96% for round spermatids. The total cellular polypeptide composition of separated human germ cells has been analyzed by two-dimensional polyacrylamide gel electrophoresis to compare 1) human and mouse pachytene spermatocytes (species specificity), 2) samples of human spermatocytes obtained from different individuals (allo specificity), and 3) pachytene spermatocytes and round spermatids from the same patients (stage specificity). Mouse and human germ cells have been found to exhibit extensive homology, but identified marker proteins limited to human spermatocytes include a group of polypeptides at p45/5.9 as well as a protein at p67/5.2. Proteins unique to mouse germ cells include component p65/5.5. Comparisons between different preparations of human pachytene spermatocytes have revealed about 90% electrophoretic homology, but some striking allotypic variations have been noted including the proteins at p45/5.9. Finally, presumptive stage-specific spermatogenic cell markers have been identified including the p45/5.9 polypeptides that are present only in human spermatocytes. Although the physiological roles of particular marker proteins have not yet been determined, the present findings indicate that purified spermatogenic cell populations may be analyzed biochemically to identify constituents important in the regulation of sperm development in man.     

Indirect Sertoli cell-mediated ablation of germ cells in mice expressing the inhibin-alpha promoter/herpes simplex virus thymidine kinase transgene

In the present study, we describe a novel mouse model for inducible germ cell ablation. The mice express herpes simplex virus thymidine kinase (HSV-TK) under the inhibin-alpha subunit promoter (Inhalpha). When adult transgenic (TG) mice were treated with famciclovir (FCV) for 4 wk, their spermatogenesis was totally abolished, with only Sertoli cells and few spermatids remaining in the seminiferous tubules. However, testicular steroidogenesis was not affected. Shorter treatment periods allowed us to follow up the progression of germ cell death: After 3 days, spermatogonia and preleptotene spermatocytes were no longer present. After a 1-wk treatment, spermatogonia, preleptotene, and zygotene spermatocytes were missing and the amount of pachytene spermatocytes was decreased. After a 2-wk treatment, round and elongating spermatids were present. During the third week, round spermatids were lost and, finally, after a 4-wk treatment, only Sertoli cells and few spermatids were present. Interestingly, the transgene is detected in Leydig and Sertoli cells but not in spermatogonia. This suggests that FCV is phosphorylated in Sertoli cells, and thereafter, leaks to neighboring spermatogonia, apparently through cell-cell junctions present, enabling trafficking of phosphorylated FCV. Because of the many mitotic divisions they pass through, the spermatogonia are very sensitive to toxins interfering with DNA replication, while nondividing Sertoli cells are protected. Using transillumination-assisted microdissection of the seminiferous tubules, the gene-expression patterns analyzed corresponded closely to the histologically observed progression of cell death. Thus, the model offers a new tool for studies on germ cell-Sertoli cell interactions by accurate alteration of the germ cell composition in seminiferous tubules.     

The role of tumor necrosis factor-alpha and interleukin-1 in the mammalian testis and their involvement in testicular torsion and autoimmune orchitis

This review will focus the roles of TNF-alpha, IL-1 alpha, and IL-1 beta in the mammalian testis and in two testicular pathologies, testicular torsion and orchitis. TNF alpha in the testis is produced by round spermatids, pachytene spermatocytes, and testicular macrophages. The type 1 TNF receptor has been found on Sertoli and Leydig cells and numerous studies suggest a paracrine mode of action for TNF alpha in the normal testis. IL-1 alpha has been reported to be produced by Sertoli cells, testicular macrophages, and possibly postmeiotic germ cells. IL-1 receptors have been reported on Sertoli cells, Leydig cells, testicular macrophages, and germ cells suggesting both autocrine and paracrine functions. While these proinflammatory cytokines have important roles in normal testicular homeostasis, an elevation of their expression can lead to testicular dysfunctions. Testicular torsion is a clinical pathology with results in testicular ischemia and surgical intervention is often required for reperfusion. A pivotal role for IL-1beta in the pathology of testicular torsion has been recently described whereby an increase in IL-1beta production after reperfusion of the testis is correlated with the activation of the stress-related kinase, c-jun N-terminal kinase, and ultimately resulting in neutrophil recruitment to the testis and germ cell apoptosis. In autoimmune orchitis, on the other hand, TNF alpha produced by T-lymphocytes and macrophages of the testis has been implicated in the development and progression of the disease. Thus, both proinflammatory cytokines, TNF alpha and IL-1, have significant roles in normal testicular functions as well as in certain testicular pathologies.     

Developmental changes in cyclin B1 and cyclin-dependent kinase 1 (CDK1) levels in the different populations of spermatogenic cells of the post-natal rat testis

Spermatogenesis is a highly ordered process which requires mitotic and meiotic divisions. In this work, we studied the relative changes in the levels of the two components of the M-phase promoting factor (MPF): the regulatory subunit cyclin B1 (CycB1) and its catalytic subunit cdk1, in spermatogenic cells of rats between 16 and 90 days of life. A multivariate flow cytometry analysis of forward scatter (FSC), side scatter (SSC) and DNA content was used to identify six populations of rat germ cells: spermatogonia with preleptotene spermatocytes, young pachytene spermatocytes, middle to late pachytene spermatocytes, secondary spermatocytes with doublets of round spermatids, round spermatids, and elongated spermatids. For any population studied no significant difference in the relative cellular content of CycB1 or cdk1 proteins between animals of different ages was observed. By contrast, CycB1 and cdk1 levels were different between the different populations of germ cells. CycB1 and cdk1 were rather high in young pachytene spermatocytes and culminated in late spermatocytes, i.e. just before the first meiotic division. The relative levels of the two proteins remained high in secondary spermatocytes then decreased in round spermatids at the exit of meiosis. Similar results were obtained by Western-blot analysis of total proteins obtained from lysates of elutriated fractions of spermatocytes and spermatids. MPF activity was assessed in lysates of germ cells from 32-day-old rats or adult animals using p13suc1 agarose and histone H1 as an exogenous substrate. H1 kinase activity was higher in pachytene spermatocytes than in round spermatid fractions from both adult and young rats. These results indicate that the meiotic G2/M transition is associated to high levels of CycB1 and cdk1 leading to high MPF activity irrespective of the age of the animals.     

Distribution and levels of cellular retinol- and cellular retinoic acid-binding protein in various types of rat testis cells

The distribution and levels of cellular retinol-binding protein (CRBP) and cellular retinoic acid-binding protein (CRABP) were measured in rat testicular peritubular and Sertoli cells and in isolated rat pachytene spermatocytes and spermatids. Two Sertoli cell preparations, one containing some germ cells and another that had been osmotically shocked to destroy germ cells, were examined. CRBP and CRABP levels were measured by specific and sensitive radioimmunoassays. Testicular peritubular cell cytosol preparations were found to contain high levels of CRBP (1.48 +/- 0.87 microgram CRBP/mg protein) but CRABP could not be detected. The mean CRBP level in Sertoli cell preparations that contained some germ cells was 0.93 +/- 0.24 microgram CRBP/mg protein; this value was similar to the level of 1.11 +/- 0.20 microgram CRBP/mg protein measured for Sertoli cells free of germ cells. The level of CRABP found in Sertoli cell preparations containing germ cells (0.81 +/- 0.32 microgram CRABP/mg protein) was approximately five times greater than was observed in Sertoli cells free of germ cells (0.16 +/- 0.03 microgram CRABP/mg protein). CRBP and CRABP levels in cultured Sertoli cells were not affected by time in culture for up to five days of culture. Pachytene spermatocytes and spermatids were very enriched in CRABP (0.72 +/- 0.26 microgram CRABP/mg protein for spermatocytes and 0.65 +/- 0.21 microgram CRABP/ml protein for spermatids). A search for a high molecular weight retinol-binding protein did not demonstrate the existence of such a protein in Sertoli cell-conditioned medium. In summary, these studies provide quantitative information about the distribution of the cellular retinoid-binding proteins in the cell types that compose the rat testis.(ABSTRACT TRUNCATED AT 250 WORDS)     

Cytological and expression studies and quantitative analysis of the temporal and stage-specific effects of follicle-stimulating hormone and testosterone during cocultures of the normal human seminiferous epithelium

In vitro culturing of normal human seminiferous epithelium remains largely unexplored. To study normal human spermatogenesis in vitro, we used a micromethod for the purification and culture of Sertoli cells, spermatogonia A, spermatocytes, and early round spermatids. Cytological quantitative data for Sertoli and premeiotic germ cell cocultures isolated from normal testicular biopsies demonstrated that cells were able to proliferate (4%), complete meiosis (6.7%), and differentiate into late round (54%), elongating (49%), and elongated (17%) spermatids at similar in vivo time delays (up to 16 days) in response to FSH + testosterone stimulation. Cells maintained normal meiotic segregation, chromosome complements, and specific gene expression profiles. Follicle-stimulating hormone + testosterone stimulated spermatogonia proliferation and Sertoli cell survival. Follicle-stimulating hormone and especially FSH + testosterone increased diploid germ cell survival during the first week, whereas only FSH + testosterone was able to inhibit cell death during the second week of culture. Follicle-stimulating hormone and especially FSH + testosterone also stimulated meiosis resumption, although this was restricted to late pachytene and secondary spermatocytes. In contrast, spermiogenesis was only stimulated by FSH + testosterone. Expression studies showed that apoptosis was induced in the nucleus of diploid cells, and in nuclear and cytoplasmic compartments of spermatids, mainly triggered by the Fas pathway. Although junctional complexes between Sertoli and premeiotic germ cells were partially reacquired, the same did not apply to spermatids, suggesting that FSH potentiated by testosterone was unable to render Sertoli cells competent to bind round spermatids.     

The immunolocalization of small nuclear ribonucleoprotein particles in testicular cells during the cycle of the seminiferous epithelium of the adult rat

The objective of this study was to determine the cellular and subcellular distribution of small nuclear ribonucleoprotein particles (snRNPs) in the adult rat testis in relation to the different cell types at the various stages of the cycle of the seminiferous epithelium. The distribution of snRNPs in the nucleus and cytoplasm of germ cells was quantitated in an attempt to correlate RNA processing with morphological and functional changes occurring during the development of these cells. Light-microscopic immunoperoxidase staining of rat testes with polyclonal anti-Sm and monoclonal anti-Y12 antibodies localized spliceosome snRNPs in the nuclei and cytoplasm of germ cells up to step 10 spermatids. Nuclear staining was intense in Sertoli cells, spermatogonia, spermatocytes, and in the early steps of round spermatid development. Although comparatively weaker, cytoplasmic staining for snRNPs was strongest in mid and late pachytene spermatocytes and early round spermatids. Quantitative electron-microscopic immunogold labeling of Lowicryl embedded testicular sections confirmed the light-microscopic observations but additionally showed that the snRNP content peaked in the cytoplasm of midpachytene spermatocytes and in the nuclei of late pachytene spermatocytes. The immunogold label tended to aggregate into distinct loci over the nuclear chromatin. The chromatoid body of spermatids and spermatocytes and the finely granular material in the interstices of mitochondrial aggregates of spermatocytes were found to be additional sites of snRNP localization and were intensely labeled. This colocalization suggests that these dense cytoplasmic structures may be functionally related. Anti-U1 snRNP antibodies applied to frozen sections showed the same LM localization pattern as spliceosome snRNPs. Anti-U3 snRNP antibodies applied to frozen sections stained nucleoli of germ cells where pre-rRNA is spliced.