NHR-23 and SPE-44 regulate distinct sets of genes during C. elegans spermatogenesis

Spermatogenesis is the process through which mature male gametes are formed and is necessary for transmission of genetic information. While much work has established how sperm fate is promoted and maintained, less is known about how the sperm morphogenesis program is executed. We previously identified a novel role for the nuclear hormone receptor transcription factor, NHR-23, in promoting C. elegans spermatogenesis. Depletion of NHR-23 along with SPE-44, another transcription factor that promotes spermatogenesis, caused additive phenotypes. Through RNA-seq, we determined that NHR-23 and SPE-44 regulate distinct sets of genes. Depletion of both NHR-23 and SPE-44 produced yet another set of differentially regulated genes. NHR-23- regulated genes are enriched in phosphatases, consistent with the switch in spermatids to post-translational regulation following genome quiescence. In the parasitic nematode Ascaris suum, MFP1 and MFP2 control the polymerization of Major Sperm Protein, the molecule that drives sperm motility and serves as a signal to promote ovulation. NHR-23 and SPE-44 regulate a number of MFP2 paralogs, and NHR-23 depletion caused defective localization of MSD/MFP1 and NSPH-2/MFP2. Although NHR-23 and SPE-44 do not transcriptionally regulate the casein kinase gene spe-6, a key regulator of sperm development, SPE-6 protein is lost following NHR-23+SPE-44 depletion. Together, these experiments provide the first mechanistic insight into how NHR-23 promotes spermatogenesis and an entry point to understanding the synthetic genetic interaction between nhr-23 and spe-44.


ABSTRACT INTRODUCTION
Spermatogenesis involves a cascade of morphogenetic events to produce the highly 46 specialized, haploid, motile gametes essential for sexual reproduction. Universal features 47 of spermatogenesis include the biosynthesis of sperm-specific proteins and assembly of 48 sperm-specific complexes, preparing for and progressing through the meiotic divisions, 49 and a streamlining step during which cells discard cytoplasmic components that are no   The transcription and translational programs that govern C. elegans spermatogenesis 86 must not only drive differentiation of spermatocytes and oocytes, but they must also 87 create the materials that spermatocytes subsequently need for sperm differentiation and   (Kulkarni et al. 2012). 96 Consistent with the MSP loading defects, these genes include MSP genes and members 97 of the small sperm-specific protein family, which promote MSP polymerization ( injected progeny. Where possible, we selected "GGNGG" crRNA targets as these have 187 been the most robust in our hand and support efficient editing (Farboud and Meyer 2015).

188
All STOP-IN cassettes were designed to be inserted directly at the DNA DSB site.

189
Sequences and descriptions of the crRNAs and repair oligos are provided in Table S2 190 and S3, respectively. We picked L4 animals, aged them one day at 20ºC and injected 6-  Table S4. 198 Knock-in sequences are provided in File S1.  Table S1 contains the processed data with gene name 232 and location, mean expression, log2 fold-change compared to control, and p-value.

233
Sequences and descriptions of the crRNAs and repair oligos are provided in Table S2 234 and S3, respectively. Genotyping primers are provided in Table S4. All supporting files 235 will be deposited in Figshare.

NHR-23 and SPE-44 regulate distinct sets of target genes 239
Given that NHR-23 is a transcription factor, we hypothesized that loss of NHR-23-  Table S6). For all three datasets, we set the cutoff for differentially regulated genes We were surprised to find so few differentially-regulated genes following NHR-23  (Table 1). smz-2 was also NHR-23 regulated (Table 1). as spe-7 in that paper) (Table 1). Two msp genes (msp-49, msp-63) were common to the 289 microarray and RNA-seq datasets (Table 1). There were five spe genes that were down-290 regulated in our RNA-seq data, but not in the spe-44 microarray (Table 1). Three genes   Fifteen differentially-regulated genes were common to NHR-23-, SPE-44-, and NHR-314 23+SPE-44-depleted animals (Fig. 2) not found in any of the datasets. We note that nsph-4.3 has a large deletion that removes   348 We focused on the NHR-23-regulated genes as they were few in number so therefore 349 amenable to a candidate-based knock-out approach. In addition to the aforementioned    were not observed in either NHR-23 or the double depleted strains (Fig. 3A). Due to this 388 unexpected phenotype, we also investigated spe-44 null (ok1400) mutants (Kulkarni et  in spe-44(ok1400) animals (Fig. 3A'). As predicted from our RNA-seq data (Table 1)  Relatively few genes were differentially regulated following NHR-23-depletion (Table S5; 441 70 genes) compared to SPE-44-depletion (Table S6; (Table S7; (Table 1). No phenotype has been previously 475 reported for the C. elegans MFP1 or MFP2 paralogs, though redundancy may be 476 responsible since these paralogs appear to have arisen through duplication. We have 477 inactivated six of the MFP2 paralogs either singly or in double mutant combinations and 478 have seen no difference in fertility. Three possible explanations for the lack of phenotype 479 are: i) that these genes do not play a role in C. elegans spermatogenesis; ii) extensive 480 genetic redundancy exists and we will have to inactivate more paralogs, if not all, to 481 produce a phenotype; or iii) that our knock-out approach caused genetic compensation. 482 We favor the second and third possibilities. Our initial approach to inactivating candidate 483 NHR-23-and SPE-44-regulated genes was to insert premature termination codons in all 484 three reading frames, as this approach was reported to produce null phenotypes and 485 requires a single crRNA/sgRNA and repair oligo (Wang et al. 2018). However, given the 486 homology between the MFP2 paralogs, they may be subject to genetic compensation.            Figure S3. MFP2 genomic DNA alignment. An alignment produced with Clustal