Abstract
Successful meiotic progression of germ cells is crucial for gametogenesis. Defects in this process affect proper genetic transmission and sometimes lead to tumor formation in the germline. In Caenorhabditis elegans, the RNA-binding protein GLD-1 is essential for the meiotic development of oocytes. However, its role during spermatogenesis has not been understood. Here, we show that GLD-1 functions redundantly with the PUF family protein PUF-8 to ensure proper meiotic development of spermatocytes. When grown at 20°—the standard laboratory temperature for C. elegans growth—primary spermatocytes in both gld-1 and puf-8 single-mutant males and hermaphrodites complete the meiotic divisions normally. By contrast, some of the gld-1; puf-8 double-mutant spermatocytes exit meiosis and form germ cell tumors in both sexes. During larval development, gld-1; puf-8 double-mutant germ cells begin to express the meiotic marker HIM-3, lose P granules, and form the sperm-specific membranous organelle, which are characteristics of developing spermatocytes. However, some of these cells quickly lose HIM-3 and form germ cell tumors that lack membranous organelle but contain P granules. Mutations that block meiotic progression at late pachytene or diakinetic stage fail to arrest the tumorigenesis, suggesting that the gld-1; puf-8 double-mutant spermatocytes exit meiosis prior to the completion of pachytene. Together, results presented here uncover a novel function for gld-1 in the meiotic development of spermatocytes in both hermaphrodites and males.
Germ cells have the unique ability to switch from the mitotic to the meiotic mode of cell division. Surprisingly, in Caenorhabditis elegans, germ cells that have progressed significantly through the meiotic program are still capable of mitotic proliferation (Francis et al. 1995a; Subramaniam and Seydoux 2003). For example, C. elegans male germ cells missing the RNA-binding protein PUF-8 enter the meiotic program but fail to progress beyond the diakinetic stage. Instead, they dedifferentiate into germ cell tumors (Subramaniam and Seydoux 2003). Presently, it is not clear whether the dedifferentiation of developing gametes is a consequence of certain unknown defects in meiotic progression or to the result of inappropriate expression of mitosis-promoting factors. Regardless of the cause, the dedifferentiation phenotype clearly indicates that meiotic germ cells can return to the mitotic mode at least during prophase I. The significance of why these cells retain mitotic potential and the mechanism(s) that suppress this potential are largely unknown.
In puf-8 mutant worms, spermatocytes dedifferentiate in to germ cell tumors only at elevated temperatures. When grown at 20°—the standard laboratory temperature for C. elegans growth—these worms do not develop germ cell tumors. In contrast, their germlines become tumorous when grown at 25°, a temperature at which wild-type germlines are still normal (Subramaniam and Seydoux 2003). Even worms homozygous for alleles that delete most of the PUF-8−coding region do not develop germ cell tumors at 20°. Thus, PUF-8 is crucial for spermatogenesis only at greater temperatures.
PUF proteins control the translation of their target mRNAs. A large number of mRNAs have been identified as potential PUF targets in yeast, worms, flies, and human (Gerber et al. 2004, 2006; Galgano et al. 2008; Morris et al. 2008; Kershner and Kimble 2010; Mainpal et al. 2011). However, PUF-mediated translational regulation in the actual biological context has not been demonstrated for many of the potential PUF targets. Results from several studies indicate that puf-8 functions redundantly with many other genes (Bachorik and Kimble 2005; Ariz et al. 2009; Racher and Hansen 2012; Vaid et al. 2013). Thus, in addition to identifying the mRNA targets of PUF-8, identification of its redundant genetic interactors is equally important for a comprehensive understanding of how PUF-8 controls germ cell development.
Current evidences indicate that PUF-8 promotes meiotic progression only in male germ cells. Germlines of feminized hermaphrodites—these form oocytes but no sperm—missing PUF-8 are not tumorous even at 25° and form functional oocytes (Subramaniam and Seydoux 2003). Previous studies have shown that the female germ cells missing GLD-1, another translational regulator, exit meiosis prematurely and form germ cell tumors (Francis et al. 1995a,b). Because genetically masculinized hermaphrodites and males grown at 20° do not form germ cell tumors, the role of GLD-1 during the meiotic progression of male germ cells has not been recognized previously. We examined the gld-1 mutant males grown at 25° and investigated the genetic interaction between gld-1and puf-8. We found that some of the gld-1(-) males do develop germ cell tumors when raised at 25°, which is strikingly similar to the puf-8 mutant. In addition, in the absence of both GLD-1 and PUF-8, germ cells exit meiosis and form tumors in all males and masculinized hermaphrodites even when they are grown at 20°. These results clearly show that GLD-1 and PUF-8 function redundantly to promote the meiotic progression of male germ cells in both males as well as hermaphrodites.
Materials and Methods
C. elegans strains
All strains used in this study were maintained as described (Brenner 1974). However, transgenic lines were maintained at 25° to prevent silencing of the germline expression (Strome et al. 2001). Transgenes were introduced into the puf-8(-) background with the use of standard genetic techniques. To avoid the potential influence of the marker alleles on the phenotype, the marker genetic backgrounds were kept identical among the different strains. For example, to compare the phenotypes of puf-8(-) and gld-1(-); puf-8(-), we examined the germlines of puf-8(zh17) unc-4(e120); dpy-5(e61) and puf-8(zh17) unc-4(e120); gld-1(q485) dpy-5(e61). To rule out allele-dependency of phenotypes, we repeated all experiments using two additional alleles of puf-8, namely ok302 and q725, and did not see any significant variations among the three alleles (Table 1 and data not shown). To rule out marker-effects [dpy-5(e61)] on the gld-1 phenotype, we repeated the experiments using RNA interference (RNAi)-based depletion of GLD-1 (Table 1 and Table 2, and data not shown). Strains used in this study are listed in Supporting Information, Table S1. Details on the generation of males of various genotypes are given in File S1.
Immunostaining and fluorescence microscopy
Gonads were extruded and stained with antibodies against HIM-3, membranous organelle (MO), PH3, and P granules and the DNA-binding dye DAPI as described previously (Ariz et al. 2009). For the Hoechst staining reported in Figure S2, dissected gonads were incubated in 5 µM Hoechst dye (Hoechst 33343; Invitrogen) in dark for 2 hr, followed by washing with M9 buffer. Gonads were mounted on agar pad for imaging. Immunostaining using anti-RME-2 antibodies was performed as described previously (Grant and Hirsh 1999; Hansen et al. 2004b). Fluorescence images were acquired using a Carl Zeiss M2 microscope and the Axiovision software package following manufacturer’s protocol (Carl Zeiss). The optimum exposure conditions were determined using the wild-type genotype and the same conditions were used to image the mutants. All images presented are representative of at least 30 gonads per experiment, and each experiment was repeated at least four times.
RNAi
Target sequences were amplified by polymerase chain reaction from cDNA from total RNA and ligated to the pSV2 RNAi feeding vector using the TA cloning method (Mainpal et al. 2011). The resulting RNAi constructs were introduced into the Escherichia coli strain HT115(DE3) and lawns prepared using the transformed bacteria were fed to worms for inducing RNAi response (Timmons et al. 2001).
Results
gld-1 and puf-8 function redundantly to prevent tumorigenesis in the male germline
To test whether gld-1 and puf-8 interact genetically, we generated a strain carrying gld-1(q485) and puf-8(zh17) alleles and examined the meiotic progression in animals homozygous for both these alleles (see the section Materials and Methods and Table S1). For convenience, we refer to this genotype as gld-1(-); puf-8(-) or gld-1; puf-8 in this paper. Both these alleles have been previously shown to be null alleles (Francis et al. 1995a; Ariz et al. 2009). To rule out allele-specific effects, we repeated the experiments by RNA-mediated depletion of GLD-1 in worms homozygous for the puf-8(q725) or puf-8(ok302) allele. These two also have been shown to be null alleles of puf-8 (Subramaniam and Seydoux 2003; Bachorik and Kimble 2005).
Both C. elegans male and hermaphrodite germlines display distal-proximal polarity with respect to germ cell development. Undifferentiated proliferating germ cells are found at the distalmost part. As cells migrate proximally, they enter meiosis and progress through the different stages of meiosis, and gametogenesis, and form mature gametes at the proximal end of the gonad. In puf-8(-) males grown at 25°, germ cells enter meiosis at the normal time and space; however, some meiotic cells return to the proliferative mode without completing the meiotic divisions and form a population of proliferating cells in the proximal gonad. The other meiotic cells complete the meiotic program and form mature sperm (Subramaniam and Seydoux 2003). These mature sperm accumulate on the distal side of the proliferating-population, which expands proximally. A few mature sperm are often interspersed among the proliferating cells, and found on the proximal side of the tumor as well (Figure 1).
gld-1 and puf-8 function redundantly to suppress tumorigenesis in the male germline. Dissected male gonads of the indicated genotypes stained with the DNA-binding dye DAPI. Only the germline nuclei are visible. These males have been raised at the indicated temperatures and were 1-d-old adults at the time of dissection. Primary spermatocytes and spermatids are seen at the proximal part in the wild-type, puf-8(-), and gld-1(-) males grown at 20°. However, all the gld-1; puf-8 double-mutant germlines contain a large population of proliferating germ cells proximal to the region containing a few sperm. Similar tumors are present in approximately 51% of the gld-1(-) and 47% of the puf-8(-) males grown at 25° (Table 1). Arrows point to representative primary spermatocytes. Full description of the genotypes: gld-1(-) = dpy-5(e61) gld-1(q485) I; unc-4(e120) II; puf-8(-) = dpy-5(e61) I; puf-8(zh17) unc-4(e120) II; and gld-1(-); puf-8(-) = dpy-5(e61) gld-1(q485) I; puf-8(zh17) unc-4(e120) II. Unless otherwise mentioned, these genotypic descriptions apply to the images shown in the subsequent figures as well.
Consistent with the earlier observations, neither puf-8 nor gld-1 single-mutant males developed germ cell tumors when grown at 20°. By contrast, all gld-1(-); puf-8(-) males developed tumors in the proximal germline (Figure 1 and Table 1). As seen in puf-8(-) males raised at 25°, a few mature sperm did accumulate on the distal side of the tumor in gld-1(-); puf-8(-) male germlines, indicating that some meiotic cells complete spermatogenesis in these germlines as well. However, perhaps because fewer cells successfully complete meiosis when both GLD-1 and PUF-8 are absent, we could not readily observe mature sperm interspersed among the tumor cells, or on the proximal side of the tumor in these germlines (Figure 1). Immunostaining for the germ cell-specific P granules confirmed that the proliferating cells in the proximal germline were indeed germ cells (Figure S1).
Because some of the puf-8(-) males form germ cell tumors when grown at 25° (Subramaniam and Seydoux 2003), we grew the gld-1(-) males at 25° to check whether they also formed similar tumors at this temperature. Approximately 51% of gld-1(-) males grown at 25° developed germ cell tumors, which was strikingly similar to that observed in puf-8(-) males (Figure 1 and Table 1). Although no specific data have been published, tumor formation in gld-1 mutant males at 25° has been observed by others as well (see the legend to Figure 3 in Ciosk et al. 2004).
One possibility is that the germ cells of gld-1(-) males grown at 25° and gld-1(-); puf-8(-) males grown at 20° fail to enter spermatogenesis and instead switch to the oogenic mode. Female germ cells thus formed in the male germline probably exit meiosis and form tumors, like their counterparts in gld-1(-) hermaphrodites (Francis et al. 1995b). To test this possibility, we examined the germlines of these males for the presence of RME-2, an oocyte-specific yolk receptor, and the sperm-specific MO by immunostaining with specific antibodies (Ward et al. 1986; Grant and Hirsh 1999). Both gld-1(-) and gld-1(-); puf-8(-) male germlines did not express RME-2, but stained positively for MO (Figure 2), which shows that the germ cells did not switch to oogenic mode in these germlines. rme-2 mRNA is a well-known target of GLD-1, and RME-2 has been shown to be misexpressed in the pachytene region of gld-1(-) hermaphrodite germlines (Lee and Schedl 2001). However, the absence of RME-2 in male germlines missing GLD-1 is not surprising because the suppression of RME-2 expression in male germlines has been shown to be independent of GLD-1 (Ciosk et al. 2004). Together, the aforementioned data show that puf-8 and gld-1 function in a redundant fashion to suppress tumorigenesis in the male germline.
Specification of the male sexual fate of germ cells does not depend on gld-1 or puf-8. (A) Germlines of males, of the indicated genotypes and growth temperatures, stained with anti-membranous organelle (MO) antibodies (green) and DAPI (red). Cells positive for the sperm marker MO—outlined and marked sperm—are present on the distal side of the tumor in both gld-1(-) and gld-1(-); puf-8(-) germlines. (B) Male germlines stained for the oocyte marker RME-2 (green) and DAPI (red). No RME-2−positive cells are seen in any of the three genotypes shown. For a positive control of RME-2 immunostaining, see Figure 4B.
On the basis of the expression pattern of a PUF-8::GFP fusion, which rescues the germline defects of puf-8 mutations, PUF-8 is thought to be expressed predominantly in the distal germline of hermaphrodites (Ariz et al. 2009; Racher and Hansen 2012). We observed a similar expression pattern in males as well with the puf-8::gfp transgene described by Ariz et al. (Ariz et al. 2009) (Figure S2). To determine whether the expression of PUF-8 in germ cells is essential for normal spermatogenesis, we depleted PUF-8 by RNAi in the rrf-1 mutant [rrf-1(ok589)], which is known to restrict the RNAi effect to germ cells (Smardon et al. 2000; Sijen et al. 2001). Similar to puf-8(zh17) males, rrf-1(ok589); puf-8(RNAi) males grown at 25° also developed tumors in the germline (Table 1), indicating that the activity of PUF-8 is indeed required in the germ cells to prevent tumorigenesis.
gld-1 and puf-8 function redundantly to suppress the spermatogenesis-dependent tumor formation in hermaphrodites
During larval development, the first few germ cells that enter meiosis in hermaphrodites differentiate as sperm. In approximately 50% of puf-8(-) hermaphrodites raised at 25°, spermatocytes dedifferentiate into germ cell tumors (Subramaniam and Seydoux 2003). The gld-1 and puf-8 functional redundancy that we saw in males prompted us to ask whether gld-1 functioned redundantly with puf-8 in hermaphrodites as well during spermatogenesis. As mentioned previously, in gld-1(-) hermaphrodites, tumors originate from female germ cells that fail to progress through meiosis (Francis et al. 1995a,b). Therefore, to address the above question, it is necessary to distinguish between tumors arising from male (puf-8-type) and female (gld-1-type) germ cells. Similar to puf-8(-) males, a cluster of sperm is present on the distal side of the tumor in puf-8(-) hermaphrodites. In contrast, because spermatogenesis is completed in gld-1 mutant hermaphrodites before the onset of tumorigenesis, sperm are present only on the proximal side of the tumor (Figure 3). Staining with DAPI revealed that all gld-1; puf-8 double-mutant hermaphrodites developed the puf-8-type tumor even when grown at 20° (Figure 3 and Table 2). None of these worms had sperm on the proximal side of the tumor. At this temperature, consistent with the earlier observations, none of the puf-8(-) hermaphrodite germlines became tumorous, and all gld-1(-) hermaphrodites formed only the gld-1-type tumor (Figure 3 and Table 2). These results suggest that the removal of gld-1 activity enhances the puf-8 mutant phenotype in hermaphrodites.
Comparison of germ cell tumors between gld-1(-) and gld-1(-);puf-8(-) hermaphrodite germlines. Hermaphrodite gonads of the indicated genotypes stained with DAPI. Mitotically cycling population of germline stem cells is present at the distal end of the gonad. Successive temporal stages of differentiating germ cells are present consecutively in the distal-to-proximal axis with the fully-developed gametes at the proximal end. Sperm nuclei appear as tiny dots within the marked area. Whereas the tumor is on the distal side of sperm in the gld-1 single mutant, it is on the proximal side of sperm in the gld-1; puf-8 double mutant.
To further confirm the aforementioned observations, we compared gld-1(-) and gld-1(-); puf-8(-) hermaphrodite germlines after immunostaining with anti-MO and anti-RME-2 antibodies. As shown in Figure 4, gld-1 single mutant germlines stained positively for RME-2, but not for MO. By contrast, in gld-1(-); puf-8(-) germlines, the sperm-like nuclei revealed by DAPI-staining were all MO-positive, and no RME-2-positive cells were observed. Additionally, P granules were present continuously throughout the gld-1(-) germline, whereas they were not detected in the sperm-rich region of the gld-1(-); puf-8(-) germline (Figure S3). Nevertheless, these results do not establish that the tumor in the double-mutant hermaphrodite germline was actually dependent on spermatogenesis. Loss of puf-8 function has been shown to masculinize the hermaphrodite germline (Bachorik and Kimble 2005; Ariz et al. 2009). Therefore, absence of PUF-8 could have led to the accumulation of a few sperm, and the tumor actually originated from female germ cells that did not express RME-2 for unknown reasons.
Expression patterns of sperm- and oocyte-specific markers in the gld-1(-) and gld-1(-); puf-8(-) hermaphrodite germlines. The gonads of hermaphrodites of the indicated genotypes were dissected out and stained with (A) anti-membranous organelle (MO) antibodies (green) and DAPI (red), or (B) anti-RME-2 (green) and DAPI (red). Whereas the gld-1(-) germlines have stained positively for RME-2, MO-specific staining is undetectable in them. In contrast, the gld-1(-); puf-8(-) germlines show an opposite expression pattern.
If tumors in gld-1(-); puf-8(-) hermaphrodite germlines originate only from female germ cells, then masculinization should prevent them from forming. So, we masculinized the hermaphrodite germlines using a temperature-sensitive, gain-of-function mutation in fem-3 [fem-3(q20)], a sex-determination pathway gene that promotes spermatogenesis (Barton et al. 1987). gld-1(-); fem-3(gf) hermaphrodites developed germ cell tumors at the permissible temperature of 15°, at which fem-3(q20) mutation does not affect oogenesis. However, gld-1(-); fem-3(gf) hermaphrodites did not develop tumors at 20°, at which fem-3(q20) prevents the switch from spermatogenesis to oogenesis (Table 2). This finding is consistent with the earlier conclusion that tumors in gld-1(-) germlines arise only from female germ cells (Francis et al. 1995b). puf-8(-); fem-3(gf) hermaphrodites did not develop tumors at either temperature, which is again consistent with the earlier observation that the puf-8(-) worms develop germ cell tumors only when grown at 25° (Subramaniam and Seydoux 2003). By contrast, all gld-1(-); puf-8(-); fem-3(gf) hermaphrodites developed germ cell tumors at 20° (Table 2), indicating that the tumorigenesis in germlines missing both GLD-1 and PUF-8 is not dependent on oogenesis.
On the basis of the results presented in the above two sections, we conclude GLD-1 functions redundantly with PUF-8 to suppress the formation of tumors arising from male germ cells in both males as well as hermaphrodites.
gld-1; puf-8 double-mutant germ cells exit meiosis and form germ cell tumors
In C. elegans, germ cell tumors arise from defects at two distinct stages of development – one, mitosis-to-meiosis transition, and two, meiotic progression (Francis et al. 1995a,b; Berry et al. 1997; Kadyk and Kimble 1998; Pepper et al. 2003; Subramaniam and Seydoux 2003; Hansen et al. 2004a). Tumors arising from puf-8(-) male germ cells and gld-1(-) female germ cells, as mentioned in the Introduction, are of the latter type (Francis et al. 1995a; Subramaniam and Seydoux 2003). In gld-1(-); puf-8(-) adult germlines, chromatin condensation characteristic of pachytene-stage spermatocytes began at the normal location. In addition, the sperm-specific structure MO started to form and the germ cell-specific P granules, which are normally absent in sperm, began to disappear at the same location. We saw similar expression patterns in the germlines of gld-1(-) males grown at 25° (Figure 2A, Figure 4A, Figure S1, and Figure S3). These observations clearly indicate that the double-mutant germ cells enter meiosis and spermatogenesis normally in adults. Nonetheless, it is possible that a few germ cells never enter meiosis, and remain quiescently in the proximal germline, and resume proliferation later.
To test the aforementioned possibility, we performed two sets of experiments. First, we blocked the meiotic progression using mutations such as mek-2(q425) and spe-6(hc49), which prevent tumor formation from cells that prematurely exit meiosis (Subramaniam and Seydoux 2003; Vaid et al. 2013). Whereas 51% of gld-1(-) males grown at 25° developed germ cell tumors, only 10% of gld-1(-); spe-6(-) males developed similar tumors (Table 1). Thus, germ cell tumors in gld-1(-) males arise largely as the result of premature meiotic exit, which is similar to what has been reported in puf-8(-) males (Subramaniam and Seydoux 2003). Surprisingly, all gld-1(-); puf-8(-); spe-6(-) males developed germ cell tumors even when grown at 20° (Table 1). Similarly, the removal of MEK-2 also did not stop tumor development in germlines missing both GLD-1 and PUF-8 (Table 2).
Second, we dissected gonads at different stages of larval development and immunostained them for the lateral-element component HIM-3, which is commonly used as a marker for meiotic cells (Zetka et al. 1999). In gld-1(-); puf-8(-) hermaphrodite germlines, all proximal-most germ cells became positive for HIM-3 at approximately 36 hr after hatching (mid L3 larva) (Figure 5A). No HIM-3-negative germ cells were found in the proximal region at this time point. Interestingly, a group of HIM-3-negative cells arose among the HIM-3-positive cells at approximately 48 hr after hatching (Figure 5B). No such HIM-3-negative cells were observed in the control [puf-8(-)] germlines; instead, at this time point, primary spermatocytes and spermatids were present in the proximal part of these germlines. At approximately 60 hr after hatching, the population of HIM-3-negative cells expanded at the proximal end of gld-1; puf-8 double-mutant germlines. However, several sperm were present in the region between the HIM-3-positive cells and the proliferating cells at the proximal end, indicating that some germ cells successfully complete meiosis and form sperm (Figure 5C). Thus, in gld-1(-); puf-8(-) germlines, the onset of meiosis during larval development is unaffected and no mitotic germ cell remains quiescently in the proximal region. Consistently, proliferating cells in the proximal germline did not appear before the start of HIM-3 expression. On the basis of the results of the aforementioned two sets of experiments, we conclude germ cell tumors in the gld-1; puf-8 double-mutant germline originate from cells that exit meiotic development prior to the stage requiring MEK-2 or SPE-6 activity.
Primary spermatocytes lacking GLD-1 and PUF-8 prematurely exit meiosis. Hermaphrodites of the indicated genotypes grown at 20° and were dissected at 36 (A), 48 (B) and 60 (C) hr after hatching. The extruded germlines were stained with anti-HIM-3 antibodies (red) and DAPI (green). Germ cells at the proximal end of the germline are positive for HIM-3 in both genotypes at 36 hr. However, at 48 hr, a population of HIM-3-negative cells is readily observable amid HIM-3-positive population in all the gld-1(-); puf-8(-) germlines (n = 50). At 60 hr, in the gld-1(-); puf-8(-) germlines, some of the meiotic cells complete meiosis and form a few sperm, whereas the HIM-3-negative population expands at the proximal end (C, right panel). Two gonads, lying side-by-side are shown for the puf-8(-) single mutant. See Figure S4 as well.
Cells positive for HIM-3 expression were present in gld-1(-) males grown at 25° and gld-1(-); puf-8(-) males grown at 20° (Figure S4), indicating that the meiotic entry is unaffected in these worms as well. In gld-1(-) males, HIM-3 expression could be detected in a few cells in the proximally proliferating population, perhaps due to meiotic re-entry. Presence of HIM-3-positive cells has been previously observed in puf-8(-) males as well (Subramaniam and Seydoux 2003).
Germ cells in which the onset of meiosis is delayed have been shown to come under the influence of proliferation-promoting signals from somatic gonadal cells called the sheath cells, which act as a “latent niche” (McGovern et al. 2009). Similar to the earlier observations in gld-1(-) hermaphrodites (McGovern et al. 2009), depletion of APX-1, the primary latent niche signal, did not affect tumor formation in gld-1(-); puf-8(-) hermaphrodites, suggesting that the tumor development in these worms is not dependent on the latent signal (Figure S5).
Discussion
The work presented here uncovers a novel function for gld-1 and a functional relationship between gld-1 and puf-8. At the standard C. elegans growth temperature of 20°, formation of germ cell tumor in worms homozygous for gld-1 null alleles is dependent on the female sex of germ cells. Based on this, previous studies had concluded that gld-1 is essential only for the meiotic development of oocytes (Francis et al. 1995b). The discovery of gld-1’s role in the meiotic development of spermatocytes has been hampered by the strong functional redundancy between gld-1 and puf-8, which the current study has uncovered. Results presented here provide unequivocal evidence for the involvement of GLD-1 in the meiotic development of spermatocytes in both hermaphrodites and males.
Functional redundancy between gld-1 and puf-8
gld-1 has two distinct functions during gametogenesis in C. elegans. One, it promotes meiotic entry in both hermaphrodites and males by functioning redundantly with gld-2 (Kadyk and Kimble 1998). Two, during meiotic progression, gld-1 has an essential role in female germ cells (Francis et al. 1995a) and, as shown in the current study, a redundant function with puf-8 in male germ cells. Previous epistasis analyses revealed that the redundant gld-1 and gld-2 pathways that promote meiotic entry are not dependent on puf-8 (Racher and Hansen 2012). Thus, the functional redundancy between gld-1 and puf-8 is most likely limited to the promotion of spermatocyte meiotic progression.
Currently we are not sure about how gld-1 and puf-8 function redundantly. GLD-1 has been shown to interact with the puf-8 3′UTR (Beadell et al. 2011; Wright et al. 2011). Therefore, one possibility is that GLD-1 suppresses puf-8 mRNA translation, and the overexpression of PUF-8 contributes to the premature meiotic exit in gld-1(-) males grown at 25°. However, this is inconsistent with the observation that the loss of PUF-8 leads to tumor formation at 25° (Subramaniam and Seydoux 2003). On the other hand, the observed expression patterns of PUF-8 and GLD-1do not support the possibility that gld-1 promotes puf-8 expression: PUF-8 is most abundant in the distal germline, where GLD-1 expression is low, and less abundant in the early pachytene region, where GLD-1 is most abundant (Jones et al. 1996; Ariz et al. 2009; Racher and Hansen 2012).
Because both GLD-1 and PUF-8 are known translational regulators, an alternative possibility is that their independent downstream targets function redundantly to promote spermatogenesis. As per this model, expression of either one of a GLD-1 or a PUF-8 target is sufficient to maintain the meiotic commitment in male germ cells. However, GLD-1 has not yet been shown to promote the expression of any of its targets; even a large-scale analysis did not find any mRNAs positively regulated by GLD-1 (Wright et al. 2011). Nevertheless, the existence of a factor positively regulated by GLD-1 cannot be excluded; therefore we do not formally rule out this model. We propose a third model in which the misexpression of a common target(s), which is controlled redundantly by GLD-1 and PUF-8, interferes in spermatogenesis. A large number of potential targets have been identified for GLD-1 and PUF-8, and several of these potential targets are common to both proteins (Mainpal et al. 2011; Wright et al. 2011). Therefore, it is probable that some of the potential common targets are either redundantly or synergistically controlled by GLD-1 and PUF-8. The product of a synergistically suppressed mRNA is likely to accumulate faster in the gld-1; puf-8 double mutant than in the gld-1 and puf-8 single mutants, which might explain why germ cells exit meiosis at an earlier time point in the double mutant.
Control of meiotic progression of male vs. female germ cells
From the results of earlier work, control of meiotic progression appeared to depend on distinct regulators in male and female germ cells (Francis et al. 1995b; Subramaniam and Seydoux 2003). Our current results, together with the previous findings, however, reveal that a single regulator, namely GLD-1, is sufficient to maintain meiotic commitment in both sexes. Thus, the downstream effectors and the underlying mechanism(s) are possibly shared between the two sexes. However, the meiotic progression of female germ cells requires GLD-1, but not PUF-8, which is somewhat inconsistent with the above suggestion. Differences in the expression patterns of GLD-1 and PUF-8 may, at least partly, explain this apparent inconsistency. GLD-1 is strongly expressed in the extended pachytene region of hermaphrodites, whereas the expression of PUF-8 in this region is weak (Jones et al. 1996; Ariz et al. 2009).
Acknowledgments
We thank David Greenstein and two anonymous reviewers for several useful suggestions; Tim Schedl for anti-RME-2 antibodies; and Steven L’Hernault for 1CB4 (anti-MO) antibody. Several of the C. elegans strains used in this study were provided by the Caenorhabditis Genetics Center, which is funded by National Institutes of Health Office of Research Infrastructure Programs (P40 OD010440). Research in the KS laboratory is supported by grants from the Indian Council of Agricultural Research through the National Agricultural Innovation Project and National Fund schemes, and by the Department of Biotechnology, Govt. of India.
Footnotes
Supporting information is available online at www.g3journal.org/lookup/suppl/doi:10.1534/g3.115.019521/-/DC1
Communicating editor: D. S. Fay
- Received April 15, 2015.
- Accepted June 5, 2015.
- Copyright © 2015 Priti and Subramaniam
This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
