Dependence of the sperm/oocyte decision on the nucleosome remodeling factor complex was acquired during recent Caenorhabditis briggsae evolution

Abstract

The major families of chromatin remodelers have been conserved throughout eukaryotic evolution. Because they play broad, pleiotropic roles in gene regulation, it was not known if their functions could change rapidly. Here, we show that major alterations in the use of chromatin remodelers are possible, because the nucleosome remodeling factor (NURF) complex has acquired a unique role in the sperm/oocyte decision of the nematode Caenorhabditis briggsae. First, lowering the activity of C. briggsae NURF-1 or ISW-1, the core components of the NURF complex, causes germ cells to become oocytes rather than sperm. This observation is based on the analysis of weak alleles and null mutations that were induced with TALENs and on RNA interference. Second, qRT-polymerase chain reaction data show that the C. briggsae NURF complex promotes the expression of Cbr-fog-1 and Cbr-fog-3, two genes that control the sperm/oocyte decision. This regulation occurs in the third larval stage and affects the expression of later spermatogenesis genes. Third, double mutants reveal that the NURF complex and the transcription factor TRA-1 act independently on Cbr-fog-1 and Cbr-fog-3. TRA-1 binds both promoters, and computer analyses predict that these binding sites are buried in nucleosomes, so we suggest that the NURF complex alters chromatin structure to allow TRA-1 access to Cbr-fog-1 and Cbr-fog-3. Finally, lowering NURF activity by mutation or RNA interference does not affect this trait in other nematodes, including the sister species C. nigoni, so it must have evolved recently. We conclude that altered chromatin remodeling could play an important role in evolutionary change.

Keywords: Caenorhabditis briggsae; NURF complex; chromatin remodelers; evolution.

Fig. 1.

The Caenorhabditis briggsae NURF complex regulates the sperm/oocyte decision. (A) Location of mutations affecting Cbr-isw-1. Caenorhabditis briggsae genomic DNA is shown as a double line, with exons as boxes. Nearby transcripts are above the line in gray, and isw-1 is below the line, in blue. The position of each new mutation is marked in red, and critical domains of the protein are coded by color. Regions of the transcript targeted by RNAi are underlined in red. (B) Nomarski photomicrograph of a Cbr-isw-1(v196) XX adult, showing a protruding vulva (red *) and small germ line (blue arrow). (C) Molecular lesions and phenotypes for each isw-1 mutation (for details, see table 1). (D) Location of mutations affecting Cbr-nurf-1. Conventions are like those for isw-1. (E) Molecular lesions and phenotypes for each nurf-1 mutation (for details, see table 1).

Fig. 2.

Caenorhabditis briggsae NURF-1 controls the sperm/oocyte decision in both sexes. (A) Caenorhabditis briggsae wild-type XX young adult. (B) Caenorhabditis briggsae nurf-1(RNAi) XX adult. This animal is older than the control in panel A, because the nurf-1 germ line develops more slowly. (C) Caenorhabditis briggsae wild-type male. (D) Caenorhabditis briggsae nurf-1(RNAi) Fog male. In all panels, anterior is left and ventral is down. The size of each inset is shown by a box on the adjacent animal; the green bar in the inset is 50 µm. Finally, “o” indicates oocytes, “e” embryos, “V” the vulva, a solid blue arrow marks a spermatheca filled with sperm, and a hollow blue arrow marks an empty spermatheca.

Fig. 3.

Caenorhabditis briggsae NURF-1A and ISW-1 control the expression of fog-1 and fog-3. Transcript levels were measured by real-time RT-PCR and calculated using 2exp(−ΔΔC_T), with rpb-1 transcripts as a reference; fog-1 and fog-3 were normalized to wild-type animals at the XX L3/L4 molt and spe-4 to the XX L4. Error bars represent standard error of the mean. (A) Wild-type animals subjected to RNAi. (B) tra-1(nm2) animals subjected to RNAi. P values were determined with a Student’s t-test, one-tailed, with unequal variance. (C) The phenotypes of siblings of animals that were collected for transcript analysis.

Fig. 4.

The role of NURF-1 and ISW-1 in germ cells changed during evolution. (A) At 20 °C, Caenorhabditis elegans NURF-1A or ISW-1 were knocked down by RNAi in fem-1(hc17ts) mutants. The double mutants were not more feminized than fem-1. Error bars represent 95% confidence intervals for a proportion. (B) Phylogeny of Caenorahbditis (Kiontke et al. 2011; Felix et al. 2014) showing the results of RNAi experiments (supplementary table S2, Supplementary Material online). (C) Caenorhabditis nigoni dsRNA causes a Fog phenotype in C. briggsae males (top) but not in C. nigoni (bottom). Gonads are outlined with dotted yellow lines, and two oocytes are shaded pink.

Fig. 5.

Model for the initiation of spermatogenesis in Caenorhabditis. (A) In most species, the NURF complex is not required. For spermatogenesis, an unknown activator promotes the expression of fog-3. Currently, the only candidate for this activator is full-length TRA-1, perhaps in association with the Tip60 HAT complex (Guo et al. 2013). (B) In C. briggsae, the NURF complex is required for fog-3 to promote spermatogenesis. Because NURF is known to remodel chromatin, the simplest possibility is that it opens up the fog-3 promoter, so that TRA-1 binding sites (shown in red) are accessible. TRA-1 then works with the Tip60 HAT complex to promote expression of fog-3, which works with fog-1 to cause germ cells to initiate spermatogenesis.