BRCA1–BARD1 Regulates Axon Regeneration in Concert with the Gqα–DAG Signaling Network

The breast cancer susceptibility protein BRCA1 and its partner BRCA1-associated RING domain protein 1 (BARD1) form an E3-ubiquitin (Ub) ligase complex that acts as a tumor suppressor in mitotic cells. However, the roles of BRCA1–BARD1 in postmitotic cells, such as neurons, remain poorly defined. Here, we report that BRC-1 and BRD-1, the Caenorhabditis elegans orthologs of BRCA1 and BARD1, are required for adult-specific axon regeneration, which is positively regulated by the EGL-30 Gqα–diacylglycerol (DAG) signaling pathway. This pathway is downregulated by DAG kinase (DGK), which converts DAG to phosphatidic acid (PA). We demonstrate that inactivation of DGK-3 suppresses the brc-1 brd-1 defect in axon regeneration, suggesting that BRC-1–BRD-1 inhibits DGK-3 function. Indeed, we show that BRC-1–BRD-1 poly-ubiquitylates DGK-3 in a manner dependent on its E3 ligase activity, causing DGK-3 degradation. Furthermore, we find that axon injury causes the translocation of BRC-1 from the nucleus to the cytoplasm, where DGK-3 is localized. These results suggest that the BRC-1–BRD-1 complex regulates axon regeneration in concert with the Gqα–DAG signaling network. Thus, this study describes a new role for breast cancer proteins in fully differentiated neurons and the molecular mechanism underlying the regulation of axon regeneration in response to nerve injury. SIGNIFICANCE STATEMENT BRCA1–BRCA1-associated RING domain protein 1 (BARD1) is an E3-ubiquitin (Ub) ligase complex acting as a tumor suppressor in mitotic cells. The roles of BRCA1–BARD1 in postmitotic cells, such as neurons, remain poorly defined. We show here that Caenorhabditis elegans BRC-1/BRCA1 and BRD-1/BARD1 are required for adult-specific axon regeneration, a process that requires high diacylglycerol (DAG) levels in injured neurons. The DAG kinase (DGK)-3 inhibits axon regeneration by reducing DAG levels. We find that BRC-1–BRD-1 poly-ubiquitylates and degrades DGK-3, thereby keeping DAG levels elevated and promoting axon regeneration. Furthermore, we demonstrate that axon injury causes the translocation of BRC-1 from the nucleus to the cytoplasm, where DGK-3 is localized. Thus, this study describes a new role for BRCA1–BARD1 in fully-differentiated neurons.


Introduction
Genetic susceptibility to breast cancer is caused largely by mutations in the BRCA1 and BRCA2 genes (Fackenthal and Olopade, 2007). These regulate a wide range of biological processes, including DNA damage repair by homologous recombination, gene silencing, cell cycle checkpoint, and centrosome duplication, all of which are relevant to the regulation of cell proliferation (Scully and Livingston, 2000;Moynahan and Jasin, 2010;Zhu et al., 2011;Li and Greenberg, 2012). BRCA1 exists primarily in a heterodimeric complex with the BRCA1-associated RING domain protein 1 (BARD1; Wu et al., 1996). It has been shown that this BRCA1-BARD1 complex possesses E3-ubiquitin (Ub) ligase activity, and this activity can be disrupted by cancerderived mutations, underscoring the critical role of this enzymatic function in suppressing tumorigenesis (Baer and Ludwig, 2002). To date, intensive efforts have been devoted to understanding the tumor-suppressive functions of BRCA1-BARD1 and BRCA2 in mitotic cells. However, their roles in postmitotic cells, such as neurons, remain poorly understood at the molecular level.
Neurons are one type of postmitotic cell, specialized for transmitting information over long distances through axons. Although axons can be damaged by various internal and external stresses, neurons have a conserved system of regenerating axons postinjury, and failure of this system can cause sensory and motor paralysis. This axon's regenerative capacity is controlled by intrinsic neuronal signaling pathways (He and Jin, 2016). Upon axon injury, Ca 21 and cAMP levels rise in severed neurons, which drives various signaling pathways (Ghosh-Roy et al., 2010;Mar et al., 2014). For instance, cAMP elevation activates cAMP-dependent protein kinase A (PKA), which promotes axonal regeneration through phosphorylation of various downstream targets (Neumann et al., 2002;Bhatt et al., 2004;Gao et al., 2004). However, the intrinsic signaling pathways that regulate regeneration in the adult nervous system have yet to be fully elucidated.
The nematode Caenorhabditis elegans has recently emerged as an attractive model to dissect the mechanisms of axon regeneration in the mature nervous system (Yanik et al., 2004). Recent studies in C. elegans have identified many signaling molecules that promote or inhibit axon regeneration (Chen et al., 2011;Nix et al., 2014;Kim et al., 2018). We have previously demonstrated that the evolutionarily conserved JNK MAP kinase (MAPK) pathway, consisting of MLK-1 MAPKKK-MEK-1 MAPKK-KGB-1 JNK, drives the initiation of axon regeneration (Nix et al., 2011). Two different protein kinases act as MAP4Ks for MLK-1 in a manner specific for different life stages. The Ste20-related kinase MAX-2 phosphorylates and activates MLK-1 mainly at the L4 stage to promote axon regeneration . On the other hand, the protein kinase C (PKC) ortholog TPA-1 can activate MLK-1 at the young adult stage, but not at the L4 stage (Pastuhov et al., 2012). The Gqa protein EGL-30 acts as a component upstream of TPA-1. EGL-30 activates the phospholipase Cb (PLCb ) EGL-8, which in turn generates diacylglycerol (DAG), an activator of TPA-1, from phosphatidylinositol bisphosphate (Lackner et al., 1999). DAG kinases (DGKs) antagonize the EGL-30 pathway by converting DAG to phosphatidic acid (PA; Miller et al., 1999).
We have recently found that BRC-2, the C. elegans ortholog of BRCA2, acts as a regulator of axon regeneration (Shimizu et al., 2018). C. elegans also has two genes, brc-1 and brd-1, which encode orthologs of mammalian BRCA1 and BARD1, respectively ( Fig. 1A; Boulton et al., 2004). BRC-1 and BRD-1 share extensive sequence and domain conservation with their mammalian counterparts, including RING and BRCT domains. Similar to mammalian BRCA1-BARD1, BRC-1 heterodimerizes with BRD-1 to form a complex having E3-Ub ligase activity (Polanowska et al., 2006). BRC-1-BRD-1 is involved in DNA repair at sites damaged by ionizing radiation. Our finding that BRC-2 is implicated in axon regeneration prompted us to Figure 1. C. elegans BRC-1 and BRD-1. A, Structures of BRC-1 and BRD-1. Schematic diagrams of BRC-1, BRD-1, and their mammalian counterparts, BRCA1 and BARD1, are shown. RING finger domain is shown in red and BRCT domains in yellow. The bold lines underneath denote the extent of the deleted regions in the tm1145, dw1, and gk297 mutants. An asterisk indicates a premature stop codon caused by the km88 mutation. B, Isolation of brc-1 mutants. Genomic structure of the brc-1 gene is shown. Exons are indicated by boxes, and introns and untranslated regions are indicated by bars. Small and capital letters indicate nucleotides and the corresponding amino acids, respectively. The brc-1(km88) mutation is a 2-bp deletion, causing a frameshift (bold amino acids) and premature stop codon (p) in exon 2. C, Ring finger domain. Sequence alignment in the RING finger domain between BRCA1 and BRC-1 is shown. Identical and similar residues are highlighted with black and gray shading, respectively. The black arrow indicates the conserved isoleucine residue required for E3-Ub ligase activity. explore the possibility that BRC-1 and BRD-1 also participate in this process.
In this study, we investigated the roles of BRC-1 and BRD-1 in axon regeneration. We found that the BRC-1-BRD-1 complex is required for axon regeneration after injury, specifically in the adult stage. We demonstrate that BRC-1-BRD-1 poly-ubiquitylates DGK-3, resulting in its degradation. Thus, BRC-1-BRD-1 enhances the EGL-30 signaling pathway by downregulating DGK-3 to promote axon regeneration. Furthermore, we show that PKA phosphorylates BRC-1, which causes the translocation of BRC-1 from the nucleus to the cytoplasm, where DGK-3 is localized. These results suggest that the BRC-1-BRD-1 complex regulates axon regeneration in concert with the Gqa-DAG signaling network. Thus, this study uncovers an unexpected role of BRC-1-BRD-1 in postmitotic neurons and suggests a molecular mechanism by which BRC-1-BRD-1 regulates axon regeneration in response to nerve injury.

C. elegans strains
The C. elegans strains used in this study are listed in Table 1. All strains were maintained on nematode growth medium plates and fed with bacteria of the OP50 strain by the standard method (Brenner, 1974).
1(km88) and brc-1(S266A)] and KU1448 [for dgk-3(km89, km90)] strains. Each of the F1 animals carrying the transgene was transferred onto a new dish and used for single-worm PCR, followed by DNA sequencing to detect the mutations. The brc-1(km88) mutation is a 2bp deletion in the brc-1 gene, causing a frameshift and premature stop codon in exon 2. The dgk-3(km89) mutation is a 20-bp insertion that contains an in-frame stop codon, thus terminating translation in the middle of exon 1. The dgk-3(km90) mutation is a 5-bp deletion, causing a frameshift and premature stop codon in exon 1.

Microscopy
Standard fluorescent images of transgenic animals were observed under an 100Â objective of a Nikon ECLIPSE E800 fluorescent microscope and photographed with a Zyla CCD camera. Confocal fluorescent images were taken on a Zeiss LSM-800 confocal laser-scanning microscope with a 63Â objective.

Axotomy
Axotomy and microscopy were performed as described previously . Animals were subjected to axotomy at the young adult or L4 stage. The young adult stage was defined as a state in which the vulva is well developed and no eggs have formed yet. Imaged commissures that had growth cones or small branches present on the proximal fragment were counted as "regenerated." Proximal fragments that showed no change after 24 h were counted as "no regeneration." A minimum of 20 individuals with one to three axotomized commissures were observed for most experiments.

Measurements of regenerating axons
The length of regenerating axons for either D-type motor neurons or touch sensory posterior lateral microtubule (PLM) neurons was measured using the segmented line tool of ImageJ. Measurements were made from the site of injury to the tip of the longest branch of the regenerating axon. Axons that did not regenerate were excluded. Data were plotted using R (ver. 4.0.1) and R studio (ver. 1.3.959).
In vitro kinase assays GFP-BRC-1 proteins were immunopurified from transfected COS-7 cells using anti-GFP antibody (mouse M048-3; MBL Forskolin treatment Treatment of animals with forskolin was performed as described previously (Ghosh-Roy et al., 2010). Forskolin (ab120058; Abcam) dissolved in DMSO was diluted in M9 media (500 mM). L4 stage worms were incubated in the forskolin solution (containing heat-killed OP50) for 12 h followed by fluorescent microscopic observation.

Quantification of DGK-3 poly-ubiquitylation
To compare differences in DGK-3 poly-ubiquitylation, band intensity minus background of HA (Ub) and T7 (DGK-3) was quantified in lanes 4 and 5 using the FUSION system (VILBER). The HA (Ub) value was divided by the corresponding T7 (DGK-3) value to determine a normalized HA (Ub) value for lanes 4 and 5. To compare DGK-3 poly-ubiquitylation levels between lanes 4 and 5, normalized HA (Ub) values in lane 5 were divided by the values in lane 4, and the derived ratios were plotted on a bar graph.
Quantitative measures of fluorescence intensity for DGK-3 degradation Animals expressing mCherry and DGK-3::GFP in D-type motor neurons were imaged immediately after axotomy (0 h) and 8 h after axotomy of selected motor neuron axons. A LSM800 confocal microscope (Zeiss) was used to obtain a z-stack of fluorescent images for mCherry and DGK-3::GFP. Mean intensity of DGK-3::GFP and mCherry in cytoplasm of neurons with severed axons was measured by drawing a circular region of interest in the center of the cell and using the measure function of ImageJ. Background intensity was   determined near analyzed cells. Relative DGK-3::GFP intensity (RI DGK-3 ) was obtained by dividing the background-subtracted value of GFP by the corresponding background-corrected value of mCherry, followed by dividing the value 8 h after axotomy by the corresponding value 0 h after axotomy. The RI DGK-3 values for wild-type and brc-1 brd-1 mutants were plotted and checked for significant differences by Wilcoxon ranksum test using R (ver. 4.0.1) and R studio (ver. 1.3.959).
Quantitative measures of fluorescence intensity for BRC-1 localization Animals expressing mCherry and GFP-BRC-1 in Dtype motor neurons with or without forskolin treatment were imaged using a Nikon ECLIPSE E800 fluorescent microscope and Zyla CCD camera. Mean intensities of GFP-BRC-1 and mCherry were measured in the cytoplasm and nucleus of D-type neurons, respectively. Background intensity was determined by measuring the mean GFP (or mCherry) intensity of adjacent regions of the same size. Normalized cytoplasmic and nuclear GFP-BRC-1 values were calculated by dividing background-subtracted cytoplasmic or nuclear GFP-BRC-1 by the corresponding background-corrected mCherry intensity. To compare cytoplasmic and nuclear GFP-BRC-1, a cytoplasmic-to-nuclear ratio was calculated and plotted using R (ver. 4.0.1) and R studio (ver. 1.3.959).

Experimental design and statistical analyses
All experiments were not randomized and the investigators were not blinded to the group allocation during experiments and outcome assessment. No statistical methods were used to predetermine sample size. Data visualization was performed using Microsoft Excel 2016, R (ver. 4.0.1), and R studio (ver. 1.3.959). Statistical analysis was conducted as described previously (Pastuhov et al., 2012). Briefly, 95% confidence intervals were calculated using the modified Wald method, and the twotailed p values were calculated using Fisher's exact test on GraphPad QuickCalcs (http://www.graphpad.com/ quickcalcs/contingency1/). The Wilcoxon rank-sum test (two-tailed) was performed using R (ver. 4.0.1), R studio (ver. 1.3.959), and the R exactRankTests package.

Results
BRC-1 and BRD-1 are required for axon regeneration To assess whether the BRCA1 ortholog BRC-1 is involved in axon regeneration, we used the CRISPR-Cas9 system to generate the null mutant brc-1(km88), which harbors a 2-bp deletion generating a premature stop codon in the second exon of the brc-1 gene (Fig. 1A,B). We first assayed regrowth following laser axotomy in GABAreleasing D-type motor neurons ( Fig. 2A). In young adult wild-type animals, ;70% of the axons initiated regeneration within 24 h after axon injury ( Fig. 2A,B; Table 2). However, in brc-1 (km88) mutants the frequency of axon regeneration was significantly reduced ( Fig. 2B; Table 2). This indicates that BRC-1 is required for efficient axon regeneration following laser axotomy. To test whether BRC-1 can act in a cell-autonomous manner, we expressed the brc-1 cDNA from the unc-25 promoter in brc-1 mutants. We found that  the axon regeneration defect of brc-1(km88) mutants was rescued by expression of brc-1 in D-type motor neurons ( Fig. 2B; Table 2). These results demonstrate that BRC-1 functions cell autonomously in injured neurons. We next asked whether the BARD1 ortholog BRD-1 also participates in axon regeneration. We found that the brd-1 (gk297) deletion (Fig. 1A) markedly reduced axon regrowth following laser injury ( Fig.  2B; Table 2). Furthermore, we observed that the regeneration defect observed in brc-1 (tm1145) brd-1(dw1) double mutants (Fig.  1A) was no greater than that seen in the single brd-1(gk297) mutant ( Fig. 2B; Table 2), suggesting that BRC-1 and BRD-1 act in the same pathway. This suggests that BRC-1 and BRD-1 function as a complex to regulate axon regeneration.
We investigated the effects of brc-1 and brd-1 on growth cone behavior, and found that the length of regenerated axons in brc-1 (tm1145) brd-1(dw1) mutants was shorter than observed in wild-type animals (Fig. 2C). In contrast, when both brc-1 and brd-1 were overexpressed using the unc-25 promoter, regenerated axons were longer than those in wild-type animals (Fig. 2C). In fact, 28% (25/ 90) of regenerated axons reached the dorsal nerve cord of animals overexpressing brc-1 and brd-1 compared with 11% (7/62) in wild-type adult animals. Overexpression of brc-1/brd-1 appeared to increase the frequency of axon regeneration, but the difference was not statistically significant ( Fig. 2B; Table 2). Thus, BRC-1-BRD-1 is required to initiate axon regeneration and control growth cone behavior.
Next, to determine whether the effect of BRC-1-BRD-1 complex on axon regeneration is specific to D-type motor neurons, we examined the effect of brc-1 and brd-1 on axon regeneration in glutaminergic touch sensory PLM neurons (Fig. 3A). Chen et al. (2011) previously performed a systematic mutant screen looking for defects in axon regeneration, and identified brd-1 as a positive regulator of axon regeneration in PLM neurons. Consistent with their finding, we found that brc-1(tm1145) brd-1 (dw1) mutants were defective in axon regeneration in PLM neurons (Fig. 3A,B). These results suggest that BRC-1-BRD-1 is generally required by neurons for axon regeneration.
BRCA1 contains a RING finger domain that functions as an E3-Ub ligase in vitro. This activity is greatly increased when complexed with BARD1, which also harbors a RING domain ( Fig.  1A; Baer and Ludwig, 2002). The Ile-26 residue in the BRCA1 RING domain is essential for its interaction with the E2-Ub conjugating enzyme but not for its interaction with BARD1, suggesting that BRCA1 is the critical subunit required for E3-Ub ligase activity. Accordingly, the I26A mutant, in which Ile-26 was replaced with alanine, is defective in E3-Ub ligase activity (Brzovic et al., 2003). Similar to mammalian BRCA1, BRC-1 possesses a RING domain with a conserved site, Ile-23, corresponding to the mammalian Ile-26 (Fig. 1C). To determine the importance of BRC-1 E3-Ub ligase activity in axon regeneration, we generated a mutant form of BRC-1 [BRC-1(I23A)] with Ile-23 mutated to alanine. We found that the I23A point mutation could not rescue the brc-1(km88) phenotype ( Fig. 2B; Table 2). Taken together, these results suggest that the BRC-1-BRD-1 complex is required for axon regeneration in a manner dependent on its E3-Ub ligase activity.

BRC-1-BRD-1 functions in the EGL-30 Gqa signaling pathway to regulate axon regeneration
We have previously demonstrated that the CED-10 Rac type GTPase-MAX-2 and EGL-30 Gqa-TPA-1 PKC pathways regulate axon regeneration mainly at the L4 and young adult developmental stages, respectively (Pastuhov et al., 2012. It has been shown that max-2 is expressed in ventral cord neurons during early development, but not at the young adult stage (Lucanic et al., 2006). This suggests that TPA-1 takes the place of MAX-2 to activate MLK-1 in axon regeneration at the adult stage. Therefore, we examined the relationship between life stage and axon regeneration in brc-1(tm1145) brd-1(dw1) double mutants. We found that axon regeneration in brc-1(tm1145) brd-1(dw1) mutants was reduced only in young adult animals and not in L4 larvae, a phenotype similar to that observed in egl-30(ad805) loss-of-function and tpa-1(k501) mutants ( Fig. 4A; Table 2; Pastuhov et al., 2012Pastuhov et al., , 2016. Thus, the BRC-1-BRD-1 complex participates in axon regeneration specifically at the adult stage. This result raised the possibility that BRC-1-BRD-1 functions in the EGL-30 signaling pathway. To investigate this possibility, we examined the genetic interactions of brc-1 and brd-1 with egl-30. We found that the defect in axon regeneration caused by the egl-30(ad805) mutation was not enhanced by introduction of the brc-1(tm1145) brd-1(dw1) mutations ( Fig. 4B; Table 2). This result supports the possibility that BRC-1-BRD-1 and EGL-30 act in the same pathway. Moreover, a gain-of-function egl-30(tg26) mutation was able to suppress the brc-1 brd-1 phenotype ( Fig. 4B; Table 2). These results suggest that BRC-1-BRD-1 promotes axon regeneration upstream of EGL-30. Alternatively, it is possible that BRC-1-BRD-1 enhances the EGL-30 pathway by inhibiting the action of a negative regulator of this signaling pathway.
Next, we investigated whether BRC-1-BRD-1 regulates DGK-3 levels in animals by expressing GFP-fused DGK-3 in D-type motor neurons using the unc-25 promoter. In wild-type animals, DGK-3::GFP was uniformly distributed in the cytoplasm of D-type neuron cell bodies (Fig.  8A). Following axon laser ablation, fluorescence intensity of DGK-3::GFP in the cytoplasm of D-type neurons was significantly decreased (Fig. 8A,B). In contrast, we found that the brc-1 (tm1145) brd-1(dw1) mutations resulted in significant stabilization of cytosolic DGK-3::GFP levels (Fig. 8A,B). Thus, BRC-1-BRD-1 is involved in axon injury-induced destabilization of DGK-3 in animals. These results suggest that increases in DGK-3 protein levels in brc-1 brd-1 mutants lead to a decrease in DAG levels, which eventually results in the inhibition of the TPA-1 PKC signaling pathway.
PKA phosphorylation induces cytoplasmic localization of BRC-1 How is BRC-1 function regulated in axon regeneration? Upon axon severance, intracellular levels of cAMP increase and PKA is Figure 9. PKA phosphorylates BRC-1. A, A schematic diagram of BRC-1. RING finger domain is shown in red, and BRCT domains in yellow. The amino acid sequences around a PKA phosphorylation consensus site (underline) and a putative nuclear localization signal (red characters) are shown below. The Ser-266 residue is indicated by an asterisk. B, PKA phosphorylates BRC-1 at Ser-266 in vitro. In vitro phosphorylation of BRC-1 by PKA is shown. COS-7 cells were transfected with GFP-BRC-1 (WT) or GFP-BRC-1(S266A), and cell lysates were immunoprecipitated (IP) with anti-GFP antibody. The immunoprecipitates were subjected to in vitro kinase assay using active recombinant PKA. Phosphorylated BRC-1 was detected by immunoblotting (IB) with anti-phospho-PKA substrate rabbit monoclonal antibody. C, Percentages of axons that initiated regeneration 24 h after laser surgery at the young adult stage. The number of axons examined is shown. Error bar indicates 95% confidence interval; pppp , 0.001, as determined by Fisher's exact test. Figure 10. PKA phosphorylation induces cytoplasmic localization of BRC-1. A, Localization of BRC-1 in response to PKA activation. Fluorescent images of wild-type animals expressing Punc-47::mcherry (D-type motor neuron, top) and Punc-25::GFP::brc-1 or Punc-25::GFP::brc-1(S266A) (bottom) with or without forskolin treatment are shown. Red and yellow arrowheads indicate cell nucleus. Scale bar: 10 mm. B, Quantification of GFP::BRC-1 fluorescence levels in D-type neurons with or without forskolin treatment. The cytoplasmic-to-nuclear ratio of GFP::BRC-1 signal was calculated as a fraction of the relative GFP::BRC-1 intensity in the cytoplasm divided by the corresponding value in the nucleus. Data are presented as a box-plot representing median (thick line within the box) and interquartile range (edge of box) with individual data points. The number (n) of cell bodies examined is shown. Statistical significance was determined by Wilcoxon rank-sum test.
activated (Neumann et al., 2002;Bhatt et al., 2004). Interestingly, BRC-1 contains a PKA phosphorylation consensus motif (Arg-Arg-Xxx-Ser) at Ser-266 (Fig. 9A). We therefore asked whether PKA phosphorylates BRC-1 at this residue. We performed in vitro kinase assays with active PKA and immuno-purified GFP-BRC-1 from COS-7 cells. Western blot analysis using an antibody recognizing phosphorylated PKA substrates revealed PKA phosphorylation of GFP-BRC-1 (Fig. 9B, lanes 1 and 2). To determine whether PKA can phosphorylate BRC-1 at Ser-266, we generated a mutant form of BRC-1 [BRC-1(S266A)], in which Ser-266 was replaced with alanine. In vitro kinase assays revealed that the S266A mutation abolished the phosphorylation of BRC-1 by PKA (Fig. 9B, lane 3). These results demonstrate that PKA phosphorylates Ser-266 of BRC-1 in vitro. In order to address the physiological significance of this phosphorylation, we used the CRISPR-Cas9 system to engineer a non-phosphorylatable brc-1(S266A) mutant, replacing the codon encoding the Ser-266 residue with an alanine codon in the endogenous brc-1 locus. We found that axon regeneration was significantly reduced in brc-1(S266A) mutants ( Fig. 9C; Table 2). This result indicates that Ser-266 phosphorylation is important for activation of the regeneration pathway by BRC-1.
We next examined how PKA-mediated phosphorylation might regulate BRC-1 in axon regeneration. Interestingly, the PKA phosphorylation site of BRC-1 is in a putative nuclear localization signal (NLS) sequence (Fig. 9A), raising the possibility that phosphorylation of BRC-1 might impact its localization. We investigated this possibility by monitoring GFP::BRC-1 localization during activation of PKA. Under normal conditions, GFP:: BRC-1 was predominantly localized in the nucleus (Fig. 10A,B). Treatment of animals with forskolin is expected to cause an increase in cAMP levels by activating adenylyl cyclase and concomitantly PKA (Ghosh-Roy et al., 2010). We found that forskolin treatment strongly induced cytoplasmic localization of GFP:: BRC-1 (Fig. 10A,B). In contrast, forskolin was unable to induce cytoplasmic accumulation of the GFP::BRC-1(S266A) mutant (Fig. 10A,B), suggesting that PKA phosphorylation of the Ser-266 site is required for the translocation of BRC-1 from the nucleus to the cytoplasm. Thus, by altering its subcellular localization, the phosphorylation of BRC-1 at Ser-266 can regulate axon regeneration.

Discussion
BRCA1 and BRCA2 genes were identified as causative genes for early-onset hereditary breast cancer (Fackenthal and Olopade, 2007). BRCA-deficient cells use error-prone DNA-repair pathways, which cause increased genomic instability (Scully and Livingston, 2000;Moynahan and Jasin, 2010;Ceccaldi et al., 2016). However, recent studies have identified new functions of BRCA1 and BRCA2 in the regulation of transcription and RNA processing relevant to their tumor-suppressive activity (Kleiman et al., 2005). Previous studies have established that the C. elegans orthologs, BRC-1 (for BRCA1) and BRC-2 (for BRCA2), possess many functional similarities with their human counterparts, including DNA damage repair, homologous recombination, and meiosis (Martin et al., 2005;Polanowska et al., 2006;Adamo et al., 2008;Janisiw et al., 2018;Li et al., 2018). Therefore, C. elegans has proven to be a very useful model system for studying the function and signaling pathways of BRCA1 and BRCA2.
We have recently found that BRC-2 regulates axon regeneration of postdifferentiated GABAergic D-type motor neurons after injury through the Rho GTPase signaling pathway (Shimizu et al., 2018). In the present study, we find that BRC-1 is also involved in axon regeneration. In humans, BRCA1 exists mostly in a heterodimeric complex with its binding partner BARD1 (Wu et al., 1996). Similarly, BRC-1 forms a complex with the C. elegans BARD1 ortholog BRD-1, and BRD-1 is also required for the regeneration of severed axons. However, the site of action of BRC-1-BRD-1 in the regulation of axon regeneration is different from that of BRC-2. BRC-1-BRD-1 participates in adult-specific axon regeneration regulated by the EGL-30 Gqa signaling pathway. Activated EGL-30 signaling induces increased production of DAG, which in turn activates TPA-1 PKC (Lackner et al., 1999). TPA-1 phosphorylates and activates MLK-1 MAPKKK to promote axon regeneration (Pastuhov et al., 2012). DGK converts DAG to PA (Miller et al., 1999); thus, inactivation of DGK activity results in elevated DAG levels. The BRC-1-BRD-1 complex enhances the EGL-30 pathway by poly-ubiquitylating DGK-3, which results in its degradation through the 26S proteasome pathway (Fig. 11). Based on this possibility, the recovery of axon regeneration in brc-1 brd-1 mutants by gain-of-function egl-30 or dgk-3 deletion mutations could be a compensatory effect. The brc-1 brd-1 mutant is defective in DGK-3 degradation, resulting in reduced DAG levels. The egl-30 or dgk-3 mutation can suppress the brc-1 brd-1 deficiency by increasing DAG levels.