By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
All germ cells throughout the animal kingdom contain cytoplasmic cloud-like accumulations of material called nuage. Polar bodies in Drosophila oocytes are probably the best known forms of nuage. In spermatogenic cells, the nuage is called the chromatoid body (CB; Parvinen,2005). The CB is a typical cytoplasmic organelle of spermatogenic haploid cells and has a function related to RNA and protein accumulation and/or reserves for later germ-cell differentiation (Söderström and Parvinen,1976a; Saunders et al.,1992). Some recent studies indicate that the CB is a highly specialized structure that may function as an intracellular focal domain that organizes and controls RNA processing in male germ cells. These reports suggest a model in which the CB functions as a subcellular coordinator of different RNA-processing pathways, centralizing post-transcriptional mRNA control in the cytoplasm of haploid male germ cells (Kotaja et al.,2006; Kotaja and Sassone-Corsi,2007). Ultrastructural analysis showed that the CB has a sponge-like structure, with regions that have different electron-density levels (Figueroa and Burzio,1998). The origin of this structure remains, at this point, unclear. Some authors suggest that the origin of the CB is nuclear (Parvinen and Parvinen,1979; Parvinen et al.,1997). Others authors have proposed that the CB precursor is dense interstitial material occurring between mitochondrial clusters (Fawcett et al.,1970) or mitochondrial product that is released in the cell cytoplasm (Reunov et al.,2000). However, some authors have speculated that the CB originates from nuages, a ribonucleoproteic complex derived from the nucleolus, and that it migrates to the cytoplasm during the initial spermatogenesis (Comings and Okada,1972; Andersen,1978; Andonov,1990; Peruquetti et al.,2008, in press).
The nucleolus is a distinct nuclear territory related to the compartmentalization of nuclear functions (Hernandez-Verdun,1991). It is a large, highly organized, nonmembrane-bound subcompartment of the nucleus and is the site of biogenesis of ribosomal subunits (Gerbi et al.,2003; Boisvert et al.,2007; Sirri et al.,2008). The size of the nucleolus increases in growing cells and decreases in resting cells, and it forms and disperses once every mitotic or meiotic cell-division cycle (Pikaard,2002; Teruel et al.,2007). Most nucleolar proteins have several functions related to RNAr synthesis and processing; however, some of them are related to other functions as well: nucleotide modifications of several small RNAs, biosynthesis of the signal recognition particle and the phased sequestration and release of proteins involved in gene silencing, senescence and cell division (Pederson,2002).
Many studies had been carried out to describe CB localization and function during the spermatogenic process in some species of invertebrates and vertebrates (Parvinen,2005; Yokota,2008), but few of them focused on fish reproduction (i.e., Braat et al.,1999; Knaut et al.,2000). Moreover, no study to obtain some relation between the fragmentation of the nucleolar material during the spermatogenesis and CB formation had been performed previously.
In this study, cytochemical and ultrastructural analyses were performed to follow the nucleolar cycle during spermatogenesis of Tilapia rendalli and to find some relationship between this phenomenon and CB formation. T. rendalli was introduced in Brazil in 1956 (Gurgel and Fernando,1994; Soares et al.,2004). This species has a great economic importance and is widely raised in captivity.
MATERIAL AND METHODS
Five adult male specimens of Tilapia rendalli (Teleostei, Cichlidae) taken from fish farm tanks at Sao Paulo State University (CAUNESP-UNESP/IBILCE) in Sao Jose do Rio Preto, Sao Paulo, Brazil, were kept in asbestos tanks (500 L, 1 fish, 5L-1) for 15 days before experimentation. Water was kept at 27°C, light was kept at 12D:12L (from 7:00 to 19:00 hr), and constant aeration was supplied. Fish were fed pellets for tropical fish (28% protein) offered ad libitum 1 hr after starting the light period and 1 hr before ending the light period. The animals were killed using an excess of anesthesia (benzocaine: 25,6 mg/L) and their gonads were removed.
The testes of each animal were removed and fixed by immersion in a Karnovsky fixative solution for 24 hr. The material was embedded in glycol–metacrylate historesin. Sections (1–3 μm thick) were obtained in Leica RM 2,155 microtome. Tissue sections were submitted to various ordinary cytological and cytochemical procedures, including: hematoxylin–eosin (HE), following Ribeiro and Lima (2000), toluidine blue (TB), modified Critical Electrolyte Concentration for detecting RNA (CEC; Mello et al.,1993), silver-ion impregnation (AgNOR; Howell and Black,1980) and Feulgen reaction (Mello and Vidal,1980). The sections of germ epithelium were evaluated under an Olympus BX 60 photomicroscope and documented using Image Pro-Plus; Media Cybernetics, version 6.1 for Windows computer software for image analysis.
In addition to the qualitative analysis of the nucleolar material distribution, the tissue sections that were subjected to silver ion impregnation were used for quantitative analysis: to determine the number of nucleoli in the spermatogonia and initial spermatids; and to measure the nuclear and nucleolar areas of the spermatogonia and initial spermatids.
Determination of the number of nucleoli in the spermatogonia and initial spermatids.
The number of nucleoli was determined in the all spermatogonia and initial spermatids used in this study. Spermatogonia (130.8 ± 48.1) and initial spermatids (145.0 ± 5.05) were used from each fish studied (N = 5). Because of the different number of the cell analyzed in each cell type, after determining the absolute values, the calculation of the percentage of the number of nucleolus in each cell type was performed.
Measuring the nuclear and nucleolar areas of the spermatogonia and initial spermatids.
The cells analyzed in advance were photo-documented using an Olympus BX 40 photomicroscope and an Image Pro-Plus; Media Cybernetics, version 4.5 for Windows computer software for image analysis. Next, the nuclear and nucleolar area of these cells were measured using Image J – Image Processing and Analysis in Java, Version 1.40 (http://rsb.info.nih.gov/ij/) software for image analysis. The cells that had one single nucleolus were measured immediately; in cells that presented two or more nucleoli, each individual nucleolus was measured before the calculating the total value.
Normal distribution of data sets was tested using Skewness and Kurtosis analysis (Ha and Ha,2007), and variance homogenity was tested using the F max test (Zar,1999). The number of nucleoli was compared between spermatogonia and initial spermatids, and also within the same cell type using the two-factor analysis of variance, complemented by the LSD multiple comparisons test (Zar,1999). Nuclear and nucleolar areas of spermatogonias were compared with nuclear and nucleolar area of initial spermatids using an independent t test (Zar,1999). Statistical significance was considered when P ≤ 0.05.
Standard meiotic cells of T. rendalli were obtained using the method introduced by Kligerman and Bloom (1977) and adapted by Bertollo and Mestriner (1998). Soon after, the testes were sectioned into small fragments and submitted to hypotonic treatment with 0.075 M KCl for 20–30 min. The material was then treated with a fresh fixative solution (methanol:glacial acetic acid, 3:1) for 30 min, and this procedure was performed a minimum of two times per fragment. After fixation, some fragments were placed in the well of a depression slide with some drops of a 50% glacial acetic acid solution and mixed until a homogeneous cell suspension was obtained. One drop of this suspension was then deposited onto a clean slide, and heated to about 38°C with the aid of a fine-tipped Pasteur pipette. The drop was then sucked back into the pipette, forming a cell ring measuring about 1 cm in diameter on the slide, with the cells preferentially deposited on the border of the ring. These last two steps were repeated on 2–3 additional fields of the same slide. Then, the slides were submitted to silver ion impregnation (AgNOR) (Howell and Black,1980) to follow the nucleolar material distribution during the meiotic division in the germ cells. The preparations were evaluated under an Olympus BX 60 photomicroscope and documented using Image Pro-Plus; Media Cybernetics, version 6.1 for Windows computer software for image analysis.
Ultrastructural Analysis—Transmission Electron Microscopy
Testes fragments of each animal were removed, sliced into small pieces and samples of the germ epithelium were cut and fixed through immersion with 3% glutaraldehyde plus 0.25% tannic acid solution in Millonig's buffer (pH 7.3) containing 0.54% glucose for 24 hr at room temperature (Cotta-Pereira et al.,1976). After being washed with the same buffer, samples were post-fixed with 1% osmium tetroxide for 1 hr at 4°C, washed in Millonig's buffer, dehydrated in a graded acetone series and embedded in Araldite resin. Ultrathin silver sections (50–75 nm) were cut using a diamond knife and stained with 2% alcoholic uranyl acetate for 30 min (Watson,1958), followed by 2% lead citrate in sodium hydroxide for 10 min (Venable and Coggeshall,1965). Samples were evaluated using a Leo-Zeiss 906 (Cambridge, UK) transmission electron microscope and documented using ITEM (Soft Image System–Csmera Veleta 2K x 2K TEM CCD Camera) software for image analysis.
This study was approved by the Ethical Committee for Animal Research (CEEA), of Sao Paulo State University (UNESP) in Botucatu, Sao Paulo, Brazil, under protocol no057/06.
The hematoxylin-eosin (HE) technique was used to analyze the male germ epithelium structure of Tilapia rendalli, and it showed that T. rendalli has a pattern of cystic spermatogenesis. It also showed that the spermatogonia occur along the entire length of the tubule (Fig. 1A). Testes were classified as a continuous germinal epithelium with Sertoli and germ cells that border an anatomizing tubular or tubular lumen supported by a basement membrane, following Grier (2000), Brown-Peterson et al. (2002), and Lo Nostro et al. (2003). The toluidine blue (TB) reaction allowed us to observe the cells of the male germ epithelium with an intense metachromasy in all of the nuclear domains (Fig. 1C–E), as well as some cytoplasmic basophilia. The metachromasy degree varied according to the genetic material compactation, the ploidy of the cell nucleus and the complexation of the nucleic acids with ribonucleoproteic (RNPs) corpuscles. The TB reaction was employed as a control for the critical electrolyte concentration (CEC) variant method for RNA detection (Fig. 1B). Tissue dyed using the CEC method showed spermatogonia with organized nucleoli (Fig. 1F), primary spermatocytes with fragmented nucleolar material randomly distributed inside the nucleus (Fig. 1G), and initial spermatids showing no nucleolus or a nucleolus smaller than the nucleolus of the spermatogonia (Fig. 1H). “Residual bodies” or “chromatophilic spheres” that originated from the accumulation of cytoplasm material residue in the germinal loci lumen presented intense metachromasy (Fig. 1B). Feulgen reaction is a DNA-specific method in which all germ cells nucleus are dyed purple, and are categorized according to ploidy degree, functional stage and chromatin compaction. Consequently, the Sertoli cells, the spermatogonia (Fig. 1I) and the initial spermatids (Fig. 1K) presented weakly dyed nuclei, whereas in the other germ cells, such as the primary spermatocytes (Fig. 1J) and later spermatids, the nuclei were more intensely dyed. Nucleolar material was observed as light regions connected with heterochromatin (Fig. 1I–K). In the spermatogonia (Fig. 1I), the nucleolar area was larger than in initial spermatids (Fig. 1K), indicating the occurrence of fragmentation and reduction of the nucleolar area in the last cell type. The nucleolar region of the germ cells was strongly impregnated using the AgNOR technique (Fig. 1L–N). In primary spermatocytes, the nucleolar region was fragmented (Fig. 1M), and in initial spermatids (Fig. 1N), the nucleolus was reorganized, although it presented a lesser volume than the spermatogonia nucleolus (Fig. 1L).
Determination of the number of nucleoli in the spermatogonia and initial spermatids.
There were no interaction between the cell type and the number of nucleoli (F = 0.5788; P = 0.633). Nevertheless, there were differences in the number of nucleoli in each cell type (F = 826.10; P < 0.05). All the spermatogonia and the initial spermatids analyzed presented 1–4 nucleoli. In both cell types, most of the cells presented single nucleoli (LSD, P < 0.05). The number of cells with 2 nucleoli was higher than the number of cells with 3 or 4 nucleoli (LSD, P < 0.05). The percentage of spermatogonia and initial spermatids showing 3 or 4 nucleoli was similar (LSD, P = 0.355). There were no differences in the number of nucleoli between the analyzed cell types (F = 0.01902; P = 0.891; Fig. 2).
Measuring the nuclear and nucleolar areas of the spermatogonia and initial spermatids.
There were significant differences between the nuclear areas in the spermatogonia and in the initial spermatids (t = 7.15; P < 0.05) and between the nucleolar area of the spermatogonia and initial spermatids (t = 12.40; P < 0.05). In the both findings, the areas of the spermagotonia were bigger than the areas of initial spermatids (Fig. 3).
Standard meiotic preparations impregnated using the silver ion method (AgNOR) showed spermatogonia mitotic metaphase with 2N = 44 chromosomes (Fig. 4A,B). Some chromosome regions presented a stronger impregnation with silver ion, indicating the possible localization of the nucleolar organizer regions (NORs) in these sites. In the nucleus of spermatogonia in interphase, the central nucleolus could be discerned (Fig. 4C). The nucleolus was fragmented in the leptotene stage. In primary spermatocytes in zigotene stage (Fig. 4D) and in pachytene stage (Fig. 4E), no nucleolar corpuscles were observed, neither during metaphase I (Fig. 4F) nor during metaphase II (Fig. 4G), demonstrating the complete disorganization of the nucleolus during these stages. Second spermatocytes in telophase II were observed, and showed the nucleus strongly impregnated with the silver ion (Fig. 4H). Initial and later spermatids presented central nucleolar corpuscles (Fig. 4I,J); however, these nucleolar areas were smaller than the nucleolar areas of the spermatogonia.
Ultrastructural Analysis—Transmission Electron Microscopy
Testis fragments of T. rendalli were ultrastructurally analyzed (Figs. 5 and 6). Spermatogonia presented a nucleolus with perfectly organized concentric layers, and the formation of some ribonucleoproteic nuages in the cytoplasm could be discerned (Fig. 5A–C). The origin of these agglomerations of the ribonucleoproteic material was probably related to some nuclear products that crossed the nuclear-pore complex and accumulated in the adjacent regions of the nucleus. In the nucleus of the primary spermatocyte the nucleolus was observed during the nucleolar fragmentation process, and in the cytoplasm of this cell type, the CB with a large area and in association with mitochondria aggregates was detected (Fig. 5C,D). Furthermore, in the nucleus of the primary spermatocytes, the presence of a ribonucleoproteic material similar to the products that crossed the nuclear-pore complex was verified, in association with the synaptonemal complex (SC; Fig. 5F). The CB is rarely found isolated in the cytoplasm of primary spermatocytes (Fig. 5G). It was typically observed in association with mitochondria aggregates (Fig. 5C–F), but it was also surrounded by them (Fig. 5E). The CB presented granular aspect (Fig. 6A) and a smaller area in the cytoplasm of later spermatids. However, the CB remains associated with mitochondria that are directed to the nucleus posterior region where the mitochondrial sheath will be formed (Fig. 6C) and in association with the centrioles in the region where the spermatozoa tail will originate (Fig. 6B,D). No CB material was observed in the final spermatids or in the mature spermatozoa (Fig. 6E,F).
By means of various cytochemical procedures used in histological sections and cytogenetic preparations of T. rendalli germ epithelium, spermatogonia showing 1–4 organized nucleolar corpuscles and primary spermatocytes in the leptotene (prophase I) was detected, and presented completely disorganized nucleoli. After the second meiotic division, initial spermatids showed 1–4 reorganized nucleolar corpuscles again. The phenomenon of nucleolar fragmentation during prophase I and its posterior reorganization in the initial spermatids had been widely described in some studies (i.e., Takeuchi and Takeuchi,1990; Peruquetti et al.,2008, in press). It has been described that, in most chordates, there is strong nucleolar activity during prophase I, and the nucleolar activity reaches its maximum at pachytene (Hofgärtner et al.,1979; Schmid et al.,1982; Wachtler and Stahl,1993; Teruel et al.,2007). However, T. rendalli showed nucleolar activity only until the leptotene stage, because during the zigotene stage, the presence of an organized nucleolus was not observed. The process of nucleolar fragmentation in T. rendalli is very similar to the human process of nucleolar fragmentation; in humans, it happens in the type B spermatogonia, showing the interruption of DNAr transcription during the initial leptotene stage (Hartung et al.,1990). Organized nucleoli were not observed in the nucleus of later spermatids or mature spermatozoa. The presence of a nucleolus is not necessary for the spermatozoa, because after fecundation, the nucleoli of the male and female pronuclei in the zygote are both of maternal origin. Recent studies have suggested that the maternal nucleolus associated with other nucleoplasmatic products is essential for the embryonary development (Lefrève,2008).
The number of nucleoli in the spermatogonia and initial spermatids was registered and compared in order to observe if this number decreases after meiosis cell division. There was not a significant difference among the number of nucleoli between the analyzed cell types, whereas the majority of the spermatogonia and initial spermatids presented single nucleoli. Cells with two nucleoli were observed less frequently, and cells containing 3 or 4 nucleoli were observed the least often. It had been previously shown that the number of nucleoli of the germ cells is related to its number of RONs (Guo et al.,1996; Teruel et al.,2007). Thus, diploid spermatogonias are expected to have twice as many nucleoli as the haploid initial spermatids. However, because the majority of spermatogonia had a single nucleolus, such a difference was not observed.
The nuclear and nucleolar areas were measured in the spermatogonia and initial spermatids to establish if the reorganized nucleolus of the initial spermatids has a similar or smaller area than that of the nucleoli of spermatogonia. We observed that the nuclear and nucleolar areas of spermatogonia were statistically larger when compared to the nuclear and nucleolar areas of initial spermatids, showing a reduction of the nucleolar area after meiotic division. The reduction of the nuclear area was expected because this measure follows the cell volume, which is significantly smaller in the spermatids, but it has been shown that the size of a nucleolus is proportional to the amount of rRNA synthesized (Caspersson,1950), that NOR size (the number of rRNA cistrons) is, in general, correlated with its expression level (Shubert and Künzel,1990), that hypertrophy of the nucleolus is a state in which rRNA and ribosome synthesis has increased (Nakamoto et al.,2001), and that large nucleoli may correlate with cell-division activity and with cellular stages having high protein demand (Mosgoeller,2004). These processes may be related to the decrease of the nucleolus size in the initial spermatids, but the nucleolus reduction can also be explained by the migration of the nucleolar fragments to the cytoplasm of germ cells in prophase I, where these fragments may participate in chromatoid body (CB) formation.
The CB is a typical cytoplasmic organelle of spermatogenic haploid cells and has a function related to RNA and protein accumulation and/or reserves for later germ-cell differentiation (Söderström and Parvinen,1976a; Saunders et al.,1992). This structure seems to be very important to the spermatogenesis process, because mutations in some components of the CB, such as TDR1/MTR-1 protein and OX3 histocompatibility antigen, induce male sterility in mice (Head and Kresge,1985; Chuma et al.,2006). Although we recognize the importance of this macromolecular complex, its origins are still unclear.
In this study, transmission electron microscopy (TEM) was used to verify the probable relationship between nucleolar fragmentation at the beginning of prophase I during spermatogenesis of T. rendalli and CB formation. Ultrastructural analysis showed the origins of CB ribonucleoproteic materials as they were migrating to the cytoplasm of spermatogonia across the nuclear pore complex and then as they accumulated in an adjacent region of the nucleus. The nucleolus of the spermatogonia is still organized; thus, it does not participate in initial CB formation. However, after the nucleolus fragmentation in the primary spermatocytes, a fraction of this fragmented material seems to move to the cytoplasm, and this ribonucleoproteic material joins with other material that has already agglomerated in the primary spermatocyte cytoplasm to form a single structure called the CB. The CB in the primary spermatocyte cytoplasm has a larger area, and it is here that it starts to perform its functions. CB formation in T. rendalli is different from CB formation in the freshwater turtle (Phrynops geoffroanus; Peruquetti,2009, unpublished data) and in some mammals, such as rats (Rattus novergicus), mice (Mus musculus; Peruquetti et al.,2008), and Mongolian gerbils (Meriones unguiculatus; Peruquetti et al.,in press), because in these species, the process of CB formation starts only in the primary spermatocytes phase rather than in the spermatogonia phase.
Some ribonucleoproteic material similar to the material that moves from the nucleus to the cytoplasm and plays a part in CB formation was observed near the synaptonemal complex of the primary spermatocytes. This observation suggests that this ribonucleoproteic material may be related to the RNA that will be translated during later spermatogenesis reserve. This is a very important function proposed for the CB, because RNA transcription occurs in the initial spermatids, but protein translation is required until the last steps of spermiogenesis (Monesi,1965). Using different procedures and identification of different CB components, many authors have suggested that the CB functions as an RNA/protein reserve (Söderström and Parvinen,1976b; Söderström,1977; Morales and Hecht,1994; Moussa et al.,1994; Oko et al.,1996; Figueroa and Burzio,1998). Parvinen et al. (1986) suggested that some presynthetized proteins that were stored in the CB will become structural components of the spermatids. In addition to mRNA, rRNA 5S and 5,8S (Figueroa and Burzio,1998), actin (Walt and Armbruster,1984; Aumüller and Seitz,1988) and calcium ion (Andonov and Chaldakov,1989) were also found in the chemical composition of the CB. The presence of actin and calcium may be related to the high mobility of this organelle, since the CB was observed in several positions of the primary spermatocytes and spermatids cytplasm in this study. The high mobility of this organelle also was detected by Parvinen et al. (1997), who traced the path of the CB and analyzed its rapidly changing positions in relation to the nuclear envelope, Golgi complex and nuclear pale chromatin areas in living early spermatids of the rat.
The results of this study also demonstrated a strong connection between the CB and mithocondrial clusters. This connection had been previously described, and some authors proposed that the origins of the CB could be from intermithocondrial material (Fawcett et al.,1970). Other authors suggested that the CB has an origin from nuclear genome products, and it is then supplemented by mitochondrial genome products (Reunov et al.,2000). The connection between the CB and mitochondria may also be explained by the participation of the CB in the synthesis and transport of the apocytochrome c, a type of cytochrome c isoform, which is expressed in the testis tissue (Hess et al.,1993). Therefore, in this study, we suggest that the relationship between the CB and mitochondria can be related to the migration of these structures to the caudal nuclear region where the mithocondrial sheath and spermatozoa tail are formed. This has also been suggested by other authors (Soley,1994; Peruquetti et al.,2008, in press). CB material was not observed in the later spermatids, suggesting that this material is dispersed and, possibly, dissolved in the residual bodies (Sud,1961; Yokota,2008).
In conclusion, the nucleolus seems to be related to CB formation during spermatogenesis of T. rendalli, because, at the moment of nucleolus fragmentation in the primary spermatocytes, the CB reaches its largest area and is able to complete its important functions during spermatogenesis. The reorganized nucleolus of the initial spermatids has a lower area due to several factors, among them the probable migration of nucleolar fragments from the nucleus to the cytoplasm, playing a part in the CB formation.
The authors thank Dr. Thaís Billalba Carvalho (Laboratory of Fish Behavior–UNESP/IBILCE), Roselene Carvalho (Laboratory of Ichthyology–UNESP/IBILCE), and Mr. Carlos Eduardo de Souza (Laboratories of Animal Biology–UNESP/IBILCE) for their help with the collection of the specimens used in this study. They also thank Dr. Tiago da Silveira Vasconcelos (Laboratory of Herpetology–UNESP/IB) and Dr. Thaís Billalba Carvalho (Laboratory of Fish Behavior–UNESP/IBILCE) for their help with statistical analysis. They also thank Mr. Luis Roberto Faleiros, Jr. (Laboratory of Microscopy and Microanalysis–UNESP/IBILCE), and Rosana Silistino de Souza (Laboratory of Cell Biology–UNESP/IBILCE) for their help with laboratory techniques.