Calcineurin tax-6 regulates male ray development and counteracts with kin-29 kinase in Caenorhabditis elegans

ABSTRACT Phosphorylation is one of the critical protein modifications, which can lead to changing the activity of the proteins and regulating a variety of biological processes. Therefore, it is essential to properly maintain the phosphorylation level on proteins by balancing the activity of kinases and phosphatases. In this study, we report that calcineurin, a serine/threonine phosphatase, counteracts with a salt inducible kinase (SIK) to control male tail development in Caenorhabditis elegans. The counteracting regulation is cell lineage-dependent; the number of defective rays from T lineage in animals lacking calcineurin tax-6 is decreased by knock-down of SIK kin-29. This result is in contrast with the knock-down of bone marrow protein (BMP) receptor kinase sma-6, which slightly aggravates the T lineage defect. Also, sma-6 knock-down results in modest defect in ray 1 of V5 lineage in the absence of tax-6 activity. Finally, knock-down of a tyrosine phosphatase cdc-25.3 does not affect the defective ray phenotype of calcineurin tax-6 loss-of-function(lf) mutants. Altogether, these results suggest that balanced phosphorylation mediated by tax-6 and kin-29 is required for proper development of T lineage rays, and tax-6 and sma-6 may function in a parallel pathway in the developmental process of V5 lineage ray 1. This study emphasizes the elaborated developmental process of male ray formation, in which carefully coordinated expression of various genes is essential.


Introduction
Caenorhabditis elegans are hermaphroditic nematodes and reproduce primarily by self-fertilization. However, C. elegans males also infrequently occur at a rate of 0.1∼0.2%, as a result of fertilization of gametes, which spontaneously lost X chromosome due to non-disjunction of X chromosome homologs during meiosis (Emmons 2005). The existence of male allow outcrossing, which significantly enhances the species genetic diversity (Barrière and Félix 2005). Male worms develop the distinctive tail structure, which is essential in male mating. The fan-like male tail has 9 pairs of rays, numbered rays 1-9 from anterior to posterior, covered within the fan cuticle, which play a pivotal role in male mating by sensing the contact of hermaphrodite mates. Although the rays share a similar structure, each has specific properties in development and function. For example, rays 1, 5, and 7 senses dorsal contact, whereas rays 2, 4, and 8 detect ventral contact (Liu and Sternberg 1995). Also, rays 7, 8, and 9 are essential for turning around hermaphrodites after contact, while rays 5, 7, and 9 are required for the timing of sharp ventral turns (Lints and Emmons 1999). Development and function of the rays are crucial in male mating behavior (Barr 2006). Therefore, what determines C. elegans ray structure and how it is regulated have mainly been examined to help understand genetic interactions underlying the ray developmental procedures.
The rays arise from three different lateral hypodermal blast cells of V5, V6, and T, which are differently programmed to develop. V5 gives rise to the most anterior ray 1, expressing the Hox gene mab-5 in a repetitive on-and-off cycling mode (Salser and Kenyon 1996). mab-5 is also activated in V6 lineages by the caudal homolog pal-1, activating egl-5 and the bHLH transcription factor lin-32, which are necessary to establish the rays 2, 3, 4, 5 and 6 (Kenyon 1986;Zhao and Emmons 1995;Ferreira et al. 1999;Hunter et al. 1999). Cells in T lineage asymmetrically divide, and some of them finally become rays 7, 8, and 9. The process is dependent on Wnt ligand LIN-44, which plays a role in the proper orientation of cell divisions, while uncharacterized gene mab-19 is essential in the development of T rays, but neither of V5 nor V6 (Sutherlin and Emmons 1994;Herman et al. 1995). These studies indicate that specific preprogrammed gene activation defines the identity of each ray. However, it is still mostly unknown what genes participate in the ray determining genetic program.
Calcineurin is a calcium-dependent serine/threonine phosphatase, which plays a critical role in various calcium-dependent signaling pathways (Liu et al. 2005). It consists of catalytic subunit A and a regulatory subunit B. Upon binding of calcium-bound calmodulin to subunit A, subunit B is released, and the autoinhibitory domain of subunit A frees from the active site. Subunit A becomes active (Rusnak and Mertz 2000). In C. elegans, tax-6, formerly known as cna-1(calcineurin a-1), encodes calcineurin subunit A and cnb-1(calcineurin b-1) encodes subunit B. tax-6 loss-of-function(lf) mutants are small, and defective in chemotaxis and thermotaxis (Kuhara et al. 2002). The cnb-1null mutants have thinner cuticle surface, smaller sperm, smaller brood size, and defects in movement and egg-laying (Bandyopadhyay 2002). On the other hand, tax-6 gain-of-function(gf) mutants exhibit abnormal enteric muscle contraction and high sensitivity to serotonin-induced egg-laying (Lee et al. 2004;Lee et al. 2005). In males, tax-6 is highly expressed in tail neurons and rays (Hu 2006). Both tax-6(lf) and (gf) mutants show low efficiency in male mating behaviors, including clumsy responding to hermaphrodites. Although these reports strongly suggest tax-6 may function in structural integrity of ray, it has not been studied whether tax-6 is required explicitly for ray development.
In this study, we investigated the ray structure in tax-6(lf) mutant animals. We found that the lack of tax-6 affect the development of all rays, while the defect in T lineage rays 7, 8, and 9 is most severe. Although a serine/threonine kinase kin-29(lf) mutants exhibited little ray defect, RNAi treatment against kin-29, which is previously reported to counteract the tax-6 activity (Singaravelu et al. 2007), suppressed the defect rate in T lineage of tax-6(lf) mutant. In contrast, RNAi to a protein tyrosine phosphatase cdc25.3, which is critical in the development of hermaphrodite germline, did not. On the other hand, feeding RNAi against BMP receptor type 1B sma-6, which is crucial in ray development (Krishna et al. 1999), slightly aggravated the defect in T lineage.
Interestingly, both tax-6(lf)and sma-6(lf) mutant animals were little defective in ray 1, but knock-down of sma-6 on the lack of tax-6 resulted in approximately 20% defect rate. These results indicate that calcineurin phosphatase tax-6 is required for normal development of rays, differently interacting with two different types of kinases, sma-6, and kin-29 in different cell lineage. This study provides insight as to how phosphorylation mediated by kinases and phosphatases regulate ray development in blast cell lineage-dependent mode.

Strains and worm maintenance
All worms investigated in this study were obtained from CGC (Caenorhabditis Genetics Center) and maintained according to the standard protocols (Brenner 1974).

Feeding RNAi
Bacteria containing each RNAi plasmid clone were cultured overnight). Each RNAi bacterial culture was seeded in an NGM plate containing 100 μg/ml of ampicillin and 1mM isopropyl-β-D-1-thiogalactopyranoside (IPTG). The seeded plates were incubated at room temperature overnight to induce the expression of RNAi clone. Synchronized worms were allowed to feed on RNAi bacteria, and F3 generation progenies were examined to study.

Observation of male reproductive phenotypes
Worms used in this study were in him-10(e1511) background. Males are collected from the plate by washing with M9 buffer and then placed on a 3% agar pad. 2% levamisole was added on the agar pad to paralyze worms. After putting a cover glass on the agar pad, worms were observed 400× magnification using a Zeiss Axio Imager A1 microscope, and images were captured using an Axiocam camera and AxioVision software. The number of fused, lost, or malformed rays were counted as a defect, and defect rate was calculated by the number of defective rays divided by the total number of rays.

Results and discussion
Calcineurin tax-6 is previously reported to be expressed in male rays and required for male mating behaviors in C. elegans (Hu 2006). However, it has been unknown how tax-6 affect ray function. To see whether tax-6 directly affect ray development, we set out to examine ray structure in tax-6(lf) mutant animals. We crossed tax-6(ok2065) with him-10(e1511), to obtain tax-6(lf) in him-10 mutant background to produce many male progenies. Approximately 40-50% of 72 observed tax-6 (lf)worms showed defective rays at the position of 7, 8, and 9, which are T lineage (Figures 1 and 4). Also, about 15% of ray 6 was defective, although tax-6 is reported not to be expressed in ray 6 (Hu 2006). Modest defect in ray 6 where tax-6 expression is seemingly low in adult stage suggest that tax-6 may be expressed and function in the early developmental stage or development of ray 6 may be influenced by neighboring rays. In contrast, less than 10% was defective in rays 1-5. Especially, rays 1 and 2 showed little defect in tax-6(lf). Altogether, these data indicate that tax-6 is necessary to develop normal rays, and its effect is predominant in T lineage rays.
Calcineurin tax-6 is a serine/threonine protein phosphatase. Thus high phosphorylation level on its substrate proteins may be responsible for ray defect observed in tax-6(lf) mutant animals. Previously, SIK kin-29 was reported to physically and genetically interact with tax-6, acting downstream of tax-6 to regulate various phenotypes such as body size, egg-laying, and sensory neuronal activity (Singaravelu et al. 2007). Therefore, we asked the question of whether kin-29 interacts with tax-6 in male ray development. To knock-down the expression of kin-29, we fed tax-6(lf) in him-10 mutant background worms with RNAi bacteria against kin-29. The RNAi treatment to kin-29 suppressed the profound defect of T lineage rays exhibited in tax-6(lf), approximately half, but slightly enhanced the defect rate in rays 3, 4, and 5 (Figures 1 and 4). To ask a question of whether kin-29 itself functions in ray development, we examined rays in kin-29(oy38lf). Interestingly, kin-29(lf) mutant animals showed little ray defect in overall ( Figure 1B). This result is consistent with the previous report (Maduzia et al. 2005). These data suggest that kin-29 activity would counteract with tax-6 in ray development, possibly sharing common substrates, then downstream of tax-6.
BMP signaling pathway is one of the crucial genetic networks regulating male ray development in C. elegans. Worms lacking regular activity of BMP receptor type 1B sma-6, a membrane serine/threonine kinase, develop malformed male tail structure resulting in abnormal mating behavior (Krishna et al. 1999). We examined the ray structure of sma-6(wk7lf) mutant animals. As previously reported, many of the mutant animals showed severe malformation in their male tails. We scored individual ray defects and found that about 80% of sma-6(lf) worms had defects in rays 6 and 7 ( Figure 2B). They also showed more than 30% defect rate in rays 4, 5, 8, and 9. These data indicate that sma-6 functions in development mainly of V6 and T lineage rays. To investigate genetic interactions between tax-6 and sma-6, we treated tax-6(lf) in him-10 mutant background worms with RNAi bacteria against sma-6. In Figure 1. tax-6 and kin-29 in ray development. Representative phase-contrast images of male rays in him-10 control, tax-6(ok2065lf) kin-29(oy38lf), and tax-6(lf);kin-29 RNAi are shown (A). The defect rate was calculated by dividing the number of worms having fused, missing, mislocated, and/or malformed rays by the total number of observed worms, and is plotted (B). The defect in ray 6-9 in tax-6(ok2065) was suppressed by kin-29 RNAi, whereas that in ray 3-5 was enhanced. Figure 2. tax-6 and sma-6 in ray development. Representative phase-contrast images of male rays in him-10 control, tax-6(ok2065lf), sma-6(wk7lf), and tax-6(lf);sma-6 RNAi are shown (A). The defect rate was calculated by dividing the number of worms having fused, missing, mislocated, and/or malformed rays by the total number of observed worms, and is plotted (B). The high defect rate in ray 7-9 in tax-6(ok2065) was not suppressed by sma-6 RNAi, but the defect rate in ray 1-5 in tax-6(ok2065) was enhanced by sma-6 RNAi.  The defect rate of V5, V6, and T lineages in tax-6(ok2056), sma-6(wk7), and tax-6(ok2065);sma-6 RNAi. sma-6 RNAi enhances the ray defects of all lineages in tax-6(ok2065), while sma-6 lf mutants almost exclusive shows ray defects in only V6 and T lineages. (C) The defect rate of V5, V6, and T lineages in tax-6(ok2056), kin-29(oy38), and tax-6(ok2065);kin-29 RNAi. kin-29 RNAi suppresses the ray defect of V6 and T lineages, but not V5. kin-29 lf mutants have normal rays. (D) The defect rate of V5, V6, and T lineages in tax-6(ok2056), and tax-6(ok2065); cdc-25.3 RNAi. cdc-25.3 knockdown aggravates the ray defects of V6 and T. these worms, the overall pattern of rays defect was closer to tax-6(lf) than to sma-6(lf), and the defect rate in rays 4-9 from V6 and T lineage was similar to that of tax-6(lf), not to sma-6(lf). Interestingly, in the ray 1 of V5 lineage, tax-6 (lf) treated with sma-6 RNAi showed approximately 20% of defect rate, in contrast to tax-6(lf) or sma-6(lf) single mutants which were almost normal in ray 1 (Figures 2  and 4). These data suggest that tax-6 and sma-6 may interact differently in different cell lineage rays, and tax-6 may be downstream of sma-6 in T and V6 lineage, but they are partially in parallel in V5 lineage. sma-6 is reported to likely phosphorylate SMAD transcription factors, which are significant substrates for BMP receptor kinase (Wang 2005). In mammals, BMP signaling is dampened by protein phosphatase 2A(PP2A), directly dephosphorylating SMAD (Bengtsson et al. 2009). It is unknown that PP2A functions counteracting sma-6, as well as tax-6, the PP2B in C. elegans, which does not seem to counteract with sma-6.
Because we found that a cytoplasmic SIK kin-29 and a plasma membrane kinase sma-6 seem to interact with calcineurin tax-6 in slightly different modes, we decided to test a phosphatase, which is unlikely to counteract with tax-6. Therefore, we treated tax-6(lf) in him-10 background with cdc-25.3, a tyrosine phosphatase, which is required for normal germline development. As expected, RNAi treatment to cdc-25.3 did not alter much of ray defect in tax-6(lf), although cdc-25.3 (ok358lf) mutant animals showed severe embryonic lethality in hermaphrodites. However, cdc-25.3 RNAi slightly aggravated defects only in rays 6, 7, 8, and 9 (Figures 3  and 4). This result indicates that cdc-25.3 might be in parallel with tax-6 in T lineage, but is unlikely to be much effective in male tail development in overall.
tax-6 has been reported to be expressed in neurons and supporting cells in male rays and is essential for normal male mating behavior (Hu 2006). In this study, we investigated the function of tax-6 in ray development. We found that tax-6 is critical for normal ray development, including ray 6, in which tax-6 is hardly expressed (Hu 2006). Especially, rays 7, 8, and 9 in T lineage are mainly affected by the lack of tax-6 activity. However, rays from V6 lineage seem to also need tax-6 for normal development because of two reasons; First, all V6 rays exhibited low but similar defect rates, and second, those defect rates increased by RNAi treatment to the counteracting kinase kin-29. Also, there is a possibility that tax-6 may function in Ray 1 from V5 in parallel with other genes such as sma-6, a membrane serine/ threonine kinase with high substrate specificity (Vogt et al. 2014). In our research, cdc-25.3 did not affect much ray defect phenotype in tax-6(lf) but increased the defect rate a bit in rays of T lineage. Both cdc-25.3 (lf) and RNAi treatment to the tyrosine phosphatase result in the germline phenotype with embryonic lethality, so it was not feasible to obtain viable males to observe ray development in these worms (Figure 3).
Notably, the rays at the border between two different cell lineages tend to show defect together. For example, ray 6 is from V6 lineage, but its defect rate is relatively higher than other V6 rays despite no presumable expression of tax-6, in the worms in which the defect rate in T lineage is high. Also, ray 2 neighboring ray 1 from V5 tends to follow the fate of ray 1. These results indicate that the development of individual rays is mostly affected by neighboring developmental environments, as well as genetic deposition determined by cell lineages.
In this study, we report the ray development phenotype of tax-6(lf) mutant animals and genetic interaction between some kinase and phosphatase genes of kin-29, sma-6, and cdc-25.3 (Figures 5 and 6). According to our data, tax-6 regulates ray development mainly of T lineage, also affecting V5 and V6 lineages. kin-29/SIK counteracts with tax-6 activity in T lineage development, because knock-down of kin-29 in tax-6 lf mutants suppresses developmental defects in T lineage, although the loss of kin-29 shows little effect in ray development. Therefore, we propose a novel regulatory pathway of tax-6/calcineurin and kin-29/SIK, which functions in T lineage ray development. sma-6/BMP receptor kinase also regulates T lineage as well as V6 lineage. The loss of tax-6 Figure 5. Effects of tax-6, sma-6, kin-29, and cdc-25.3 on ray development. The loss of tax-6 activity affects the development of rays from all of V5, V6 and T lineages. sma-6 regulates V6 and T lineage rays, and tax-6 may inhibit sma-6 regulation on V6 and T lineage development. Counteracting with tax-6, kin-29 regulates the development of rays 2-9, which are V6 and T lineages. cdc-25.3 may be in parallel to tax-6 in the development of rays 7-9 of T lineages.
in the knock-down of sma-6 partially restores impaired rays in T and V6 lineage, suggesting tax-6 may suppress the sma-6 pathway in ray development. sma-6 also seems to be in parallel with tax-6 in regulation of V5 lineage. cdc-25.3/tyrosine phosphatase regulates T lineage at least in parallel with tax-6, while it is unknown whether its activity is required for ray development. This study provides helpful information to understand the genetic interaction between kinases and phosphatases in tissue development at the organismal level and reinforce the intriguing features of male ray development regarding differential gene regulation of different blast cell lineages in C. elegans. Figure 6. Genetic network model regulating T lineage. S is an unknown phosphorylation substrate for tax-6 and kin-29. Phospho-S (S-P) positively regulates ray development in T lineage, in the pathway tax-6 phosphatase and kin-29 kinase balance phosphorylation level on the substrate S. sma-6 is a direct regulator on T lineage, and tax-6 may suppress the downstream of sma-6.