UNC-55 and GABAergic motor neuron subtype differentiation

Author’s Accepted Manuscript Meis/UNC-62 isoform dependent regulation of CoupTF-II/UNC-55 and GABAergic motor neuron subtype differentiation Richard F. Campbell, Walter W. Walthall www.elsevier.com/locate/developmentalbiology

PII: DOI: Reference:

S0012-1606(16)30438-9 http://dx.doi.org/10.1016/j.ydbio.2016.09.009 YDBIO7252

To appear in: Developmental Biology Received date: 13 July 2016 Revised date: 24 August 2016 Accepted date: 9 September 2016 Cite this article as: Richard F. Campbell and Walter W. Walthall, Meis/UNC-62 isoform dependent regulation of CoupTF-II/UNC-55 and GABAergic motor neuron subtype differentiation, Developmental Biology, http://dx.doi.org/10.1016/j.ydbio.2016.09.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Meis/UNC-62 isoform dependent regulation of CoupTF-II/UNC-55 and GABAergic motor neuron subtype differentiation Richard F. Campbell, Walter W. Walthall* Department of Biology, Georgia State University, Atlanta, GA 30303 *

Corresponding author. Georgia State University, Department of Biology, 161 Jesse Hill Jr.

Dr., SE, Atlanta, GA 30303. Tel.: 404 413 5391. [email protected] ABSTRACT Gene regulatory networks orchestrate the assembly of functionally related cells within a cellular network. Subtle differences often exist among functionally related cells within such networks. How differences are created among cells with similar functions has been difficult to determine due to the complexity of both the gene and the cellular networks. In Caenorhabditis elegans, the DD and VD motor neurons compose a cross-inhibitory, GABAergic network that coordinates dorsal and ventral muscle contractions during locomotion. The Pitx2 homologue, UNC-30, acts as a terminal selector gene to create similarities and the Coup-TFII homologue, UNC-55, is necessary for creating differences between the two motor neuron classes. What is the organizing gene regulatory network responsible for initiating the expression of UNC-55 and thus creating differences between the DD and VD motor neurons? We show that the unc-55 promoter has modules that contain Meis/UNC-62 binding sites. These sites can be subdivided into regions that are capable of activating or repressing UNC-55 expression in different motor neurons. Interestingly, different isoforms of UNC-62 are responsible for the activation and the stabilization of unc-55 transcription. Furthermore, specific isoforms of UNC-62 are required for proper synaptic patterning of the VD motor neurons. Isoform specific regulation of

differentiating neurons is a relatively unexplored area of research and presents a mechanism for creating differences among functionally related cells within a network. Keywords Gene regulatory networks, Subtype determining transcription factors, Isoform specific regulation, Motor neurons INTRODUCTION Cellular networks are typically composed of different classes of functionally related cells. These related cells often perform similar functions, for example, motor neurons (MNs) in the vertebrate spinal cord are organized into MN pools that innervate specific muscle targets. How individual MNs within a pool differentiate and establish distinguishing characteristics such as unique synaptic profiles is an important question. In the developing chick spinal cord, MNs differentiate via a combinatorial code of Hox factors and cofactors that confer a unique identity to each MN pool (Dasen et al., 2005). This pool can be subdivided via the expression of subclass specific transcription factors such as PEA, Lim-HD and Islet, thus creating over 50 different MN subtypes (Dasen et al., 2005; Vrieseling and Arber, 2006). Interestingly, the development of the ventral nerve cord (VNC) in Drosophila melanogaster utilizes many of the same gene regulatory networks (GRNs) as vertebrates, suggesting conservation from an ancient ancestor (Certel and Thor, 2004). Although advances have been made in understanding the molecular logic for generating subtype differentiation of MNs, how each subtype is distinguished from its neighbors is not well understood. This is due, in part, to the complex interactions among post mitotic terminal differentiation factors and their upstream regulatory components.

In the nematode, Caenorhabditis elegans, the organization of the MNs is much simpler than insects or vertebrates. The C. elegans MNs involved in locomotion are located in the ventral nerve cord (VNC). The adult VNC is composed of 75 MNs organized into eight classes. Of these classes, six are cholinergic (White et al., 1976). Most, but not all, of the cholinergic MNs require the COE transcription factor UNC-3 for terminal differentiation (Kratsios et al., 2012). These six classes are further sub-divided by the subtype determining transcription factors: paired domain/UNC-4 (VA) and even-skipped/VAB-7 (VB) (Miller et al., 1992; Winnier et al., 1999; Esmaeili et al., 2002). There are two GABAergic inhibitory classes of MNs, the DD and VDs. Together they form a cross-inhibitory network in which the DD MNs receive ventral inputs from excitatory MNs and innervate dorsal muscle whereas the VD MNs receive dorsal inputs from excitatory MNs and innervate ventral muscle. Their GABAergic identity is determined by the expression of the conserved Pitx transcription factor, UNC-30, which activates directly the transcription of GABA synthesis and transport genes (Jin et al., 1994; Eastman et al., 1999; Westmoreland et al., 2001). In newly hatched animals the DD MNs innervate ventral muscle and then reverse their signaling polarity to achieve the adult pattern described above (White et al., 1978; Park et al., 2011). The initial synaptic profile of the DD MNs is maintained during the L1 (first postembryonic stage) until the heterochronic gene, LIN-14 activates the synaptic re-arrangement (Hallam and Jin, 1998; Park et al., 2011). At about the same developmental stage of the DD MN synaptic reorganization (late L1) the VD MNs are born and immediately establish pre- synaptic sites on ventral muscle (White et al., 1976). A sub-type determining transcription factor, UNC55, is expressed in the VD MNs. UNC-55 represses a subset of genes activated in the DD MNs by UNC-30 and stabilizes the innervation of ventral muscle in the VD MNs (Zhou and Walthall,

1998; Shan et al., 2005; Petersen et al., 2011). UNC-55 suppresses the expression of the Irquois like transcription factor IRX-1 in the VD MNs, while IRX-1 expression persists in the DD MNs (Petersen et al., 2011). UNC-30 and LIN-14 are required for the expression of the single immunoglobin domain, OIG-1 in the DD MNs before the L1 molt (Howell et al., 2015). IRX-1 represses OIG-1 expression in DD MNs of L2 animals and allows the DD MNs to undergo a synaptic rearrangement (He et al., 2015; Howell et al., 2015). Since IRX-1 is not expressed in the VD MNs, OIG-1 expression is present. In unc-55 mutants, VD MN expression of OIG-1 decreases due to de-repression of IRX-1 and the VD MNs adopt a DD synaptic profile (Zhou and Walthall, 1998; Shan et al., 2005; Petersen et al., 2011; He et al., 2015; Howell et al., 2015). The transcription of unc-55 in the VD MNs is not regulated by UNC-30 (Shan et al., 2005), indicating that multiple GRNs converge to generate the synaptic profiles that distinguish the DD and VD MNs from one another. Insight into the converging networks required for DD and VD synaptic profiles could be gained by understanding how UNC-55 expression is regulated in the VD MNs. To address this issue, we examined the transcriptional regulation of unc-55 in the VD MNs, reasoning that this would reveal the relationship between a subtype determining GRN and the terminal selector gene unc-30. To accomplish this we dissected the unc-55 promoter and tested mutant candidates that may be responsible for UNC-55 expression and synaptic organization of the VD MNs. Materials and Methods Strains

The following mutant alleles were crossed into jdIs101 (1.6kb of the unc-55 promoter fused with GFP) unc-62 (e917, e644, jd1002 and mu232) and ceh-20 (ay9). The transgene juIs1 was used for puncta counting. The N2 Bristol strain was used for all microinjections. Microscopy For transcriptional reporters: animals were mounted on a 0.01M Sodium Azide 2% agar pad and fluorescence imaging was captured by an Applied Precision Deconvolution microscope using a PlanApo 40 × 1.35 numerical aperture objective lens (Olympus America, Center Valley, PA). All images were deconvolved (5 iterations) with softWoRx 5. MNs were counted regardless of intensity. Z projections (max intensity) were obtained using Image J. Puncta counting was performed between VD cell bodies as previously described with the exception that puncta were counted using Image J cell counter tool for each z-stack (Petersen et al., 2011). UNC-55::GFP animals were imaged on sodium azide pads(as described above) on a Zeiss LSM780 confocal microscope with a 40x objective. Data collected for smFISH was collected using a Nikon eclipse 90i fluorescent microscope with a PlanApo 60 × 1.4 numerical aperture objective lens. Quantitative PCR Four replicates were performed for each allele examined. Animals were synchronized to late L1 via bleaching. Eggs were allowed to hatch in M9 overnight and animals were plated at the outer edges of a seeded plate. Since mutations in unc-62 cause embryonic and larval arrest, the lawn portion of the plate was chunked when isolating animals for mRNA to prevent capturing arrested embryos and L1s. RNA was isolated using a RNeasy plus mini kit (Qiagen) and cDNA was synthesized using High Capacity RNA-to-cDNA™ Kit (Applied Biosystems). Taqman® probes

(Applied Biosystems) for unc-55 were used in conjunction with Taqman® universal mastermix II (Applied Biosystems). Taqman® probes for pmp-3 and pha-4 were used for normalization. smFISH Probes for unc-55 smFISH were designed via the Stellaris® FISH Probe Designer (Biosearch Technologies, Inc., Petaluma, CA) available online at www.biosearchtech.com/stellarisdesigner. Animals were hybridized with the unc-55 Stellaris FISH Probe labeled with Quasar® 570 dye (Biosearch Technologies, Inc.), following a modified protocol from the manufacturer www.biosearchtech.com/stellarisprotocols. Animals were fixed in ice cold acetone for 10 minutes at room temperature followed by three washes with nuclease free PBS (Lonza). All steps were taken in a 1.5 mL centrifuge tube instead of an inverted chambered coverslip. Once hybridizations (four hours) and DAPI (30 minutes)(Sigma Aldrich) counterstaining were completed, animals were pipetted onto a lysine coated microscope slide (Fisher Scientific) and Prolong Gold Antifade Mounting medium (Thermo Fisher Scientific) was applied before sealing the coverslip with nail polish and curing for 3 hours. Image analysis and automated smFISH quantification was performed using the FISHquant module in MatLab(Mueller et al., 2013). Nuclear Labeling Animals were fixed using 4% paraformaldehyde and suspended in 0.2M sodium cacodolate solution for 45 min at room temperature. Hoescht staining was performed for 30 minutes at room temperature in the dark. If staining penetrance was not robust, reapplication of Hoechst was used followed by 3 washes. The reduced fixation time was used to prevent GFP quenching and fixation artifacts. Bioinformatics

To identify potential transcription factor binding sites (cis elements) that regulate unc-55 we compared the 2 kb region upstream of the first unc-55 exon of C. elegans with C. briggsae and C. remanei using MUSSAGL. TESS (http://cbil.upenn.edu/tess/) and Matinspector (www.genomatix.com) were used to analyze regions of shared homology to identify conserved transcription factor binding sites in the unc-55 promoter (Schugn, 2003). Each candidate’s DNA binding domain was submitted to BLAST and the top five gene alignments were selected. We then selected candidates that are expressed during L1 or L2 stages in VNC. To assess the Meis binding sites for site directed mutagenesis, the region between -376 bp and -493 bp upstream of the first unc-55 exon was resubmitted to Matinspector. A single Meis binding site was identified via position weight matrix with a core sequence of TGTC. Transgenic Strains and Injections The unc-55 promoter fusions with GFP were constructed using the PCR fusion technique described previously (Hobert, 2002). The unc-55 promoter constructs were injected with a concentration of 2.5-5 ng/µL in conjunction with the pRF4 plasmid containing rol-6 (su1006) DNA at a concentration of 100 ng/µL. The site directed mutagenesis of the Meis binding site in the unc-55 promoter was performed using PCR fusion with primers designed with overlapping 3’ and 5’ overhangs that contained the TGTC core sequence changed to AAAA. These constructs were blunt end cloned using Thermo Scientific ClonJet (#K1231) and sequenced for confirmation that the core sequence was changed. The resulting plasmid containing the 493 bp promoter of unc-55 with the altered Meis core sequence was used as a template for PCR based injections. CRISPR/Cas9

All guide RNAs were designed using the Zhang lab’s online gRNA selection tool (Hsu et al., 2013). The top two hits (having a score greater than 98) were selected. Guide RNA’s were then inserted into pDD162 via New England Biolab’s Q5 site directed mutagenesis kit as previously described (Dickinson et al., 2013). All plasmids were sequenced for verification of guide RNA insertion. Young adult animals were injected with 50 ng/µL of each plasmid (pDD162 and pJA58) and 5 ng/ µL of pCFJ104 (myo-3::mcherry) and 600 nM of cn64 repair ssDNA. Injected animals were screened for either myo-3::mcherry or rol/dpy phenotypes. F2 animals were screened for unc-62 phenotypes including but not limited to: uncoordination, morphological defects, lethality/reduced brood size and egg-laying defects. F2 hermaphrodites exhibiting any of these phenotypes were isolated and allowed to reproduce. RESULTS The unc-55 promoter contains multiple cis-regulatory elements required for repression and activation. To investigate the transcriptional regulation of UNC-55, we generated multiple fluorescent reporter lines using different lengths of the predicted unc-55 promoter fused with GFP (Fig. 1A.) The 1.6 kb punc-55::GFP integrated line, jdIs101, reliably produced expression in the AS and VD motor neurons (Fig. 2B) (Shan et al., 2005). However, recent chromatin immunoprecipitation (ChIP) data from ModENCODE suggests that multiple transcription factors bind to the 1.5-2 kb region of the unc-55 promoter (Gerstein et al., 2010). Since jdIs101 encompasses 1.6 kb 5’ to the first exon and a portion of the first intron of unc-55, it could be missing important elements of the unc-55 promoter. Thus we created punc-55::GFP extra chromosomal lines that encompassed 1932 bp (punc-55-1932) and 1472 bp (punc-55-1472)

upstream of the first exon. The punc-55-1932and punc-55-1472lines both recapitulated the expression pattern reported for jdIs101, suggesting that transcription factors occupying the 1.5-2 kb region upstream of unc-55 may be involved in unc-55 transcriptional regulation in other tissues (Fig. 1C and 1H) (Shan et al., 2005). We also observed expression of punc-55::GFP in the mother cell of AS and VD, which is consistent with previous observations (Thompson-Peer et al., 2012). We next truncated the promoter fusion to 1077 bp (punc-55-1077) and observed a significant increase in the number of GFP positive MNs in the VNC as compared to the punc-551472

constructs (Fig 1E and 1H). We observed that the intensities of the additional MNs were

much fainter than the presumed AS/VD MNs. These data are indicative that a motif between 1077 and 1472 bp upstream of the first unc-55 exon is required for the repression of unc-55 transcription in VNC MNs other than AS and VD. To find the minimal promoter required for AS/VD UNC-55 expression, we further truncated the promoter to 493 bp fused with GFP (punc-55493). We observed that punc-55::GFP expression reverted to the expression pattern observed in jdIs101, punc-55-1932and punc-55-1472 bp constructs (Fig. 1F). These data suggested that a motif resides between -472 to-1077 bp that is responsible for the activation of unc-55 transcription in VNC motor neurons other than the AS and VD (Fig. 1A). A 366bp punc-55::GFP (punc-55-366) construct was generated to determine the motif in the unc-55 promoter responsible for the unc-55 transcription in the AS and VD MNs. A complete loss of expression was observed for punc-55-366 constructs in the VNC, however, expression persisted in head neurons (Fig. 1G). This suggested that motifs necessary for the expression of punc-55::GFP in the AS/VD MNs resides between 366-472bp upstream of the first exon (Fig. 1A). In summary, the data from the transcriptional reporters suggest that at

least two separate regions of the unc-55 promoter activate unc-55 transcription in the AS/VD and other MNs. Additionally; at least one region represses unc-55 transcription in MNs other than the AS/VD. Identification of candidate genes for unc-55 transcriptional regulation The promoter analysis identified regions of interest for the expression and repression of UNC-55 in the VNC. We identified candidates for unc-55 transcriptional regulation by searching for conserved candidate binding sites in the unc-55 promoter of C. elegans and in related nematode species, C. briggsae and C. remanei via TESS and MatInspector (Schugn, 2003; Cartharius et al., 2005). Pbx/CEH-20 and Meis/UNC-62 emerged as promising candidates as both are expressed in the VNC (Jiang et al., 2009; Potts et al., 2009). Using the jdIs101 reporter strain we assayed the spatial and temporal pattern of VNC expression in candidate mutants. We observed an increase in the number of GFP positive cells in the VNC in ceh-20 and unc-62 mutant backgrounds. We observed additional punc-55::GFP expression in the posterior portion of the VNC in ceh-20 mutants; however, expression in the VNC was highly variable. In addition to expression in the AS/VD MNs, we observed a consistent increase in the number of MNs expressing of punc-55::GFP in the unc-62 allele, e644 (Fig. 2F). Thus we investigated the regulatory role of unc-62 in the transcription of unc-55. Five putative UNC-62 binding sites were found in the unc-55 promoter with at least one putative binding site located within each of the regions that had altered punc-55::GFP expression from the promoter analysis (Fig. 1A). Different alleles of unc-62 produce different expression patterns The unc-62 locus has at least 13 alternative transcripts. The source of four of the alternative transcripts is the result of multiple transcriptional start sites (exon 1a and 1b) and two

alternatively spliced exons (exon 7a and 7b). Additional transcriptional start sites have been reported for exon 2 along with multiple truncated transcripts identified by RNA-seq which account for the rest of the alternative transcripts (Spencer et al., 2011; Craig et al., 2013). Previous reports indicate that isoforms containing exon 1a and 1b may be expressed in the VNC (Craig et al., 2013). However, UNC-62 isoforms containing 7b are expressed in the VNC while conflicting reports indicate that isoforms containing exon 7a may or may not be present in the VNC (Spencer et al., 2011; Craig et al., 2013; Van Nostrand et al., 2013). In regards to class specific expression, RNA-seq data indicates that UNC-62 exon1a is expressed in the GABAergic and A class MNs in L1 but not L2. Furthermore, exon 7b is expressed in GABAergic and A class MNs during L1 and L2 (Spencer et al., 2011). Predicted null mutations for unc-62 such as s472, result in 100% embryonic or L1 lethality and no attempt was made to analyze these alleles (Van Auken et al., 2002). The viable allele, e917, is a chromosomal inversion that disrupts a predicted enhancer site for exon 1a, thus it is a regulatory mutant (Van Auken et al., 2002). Similar to other alleles that disrupt or knock out exon 1a, e917 has severe embryonic and larval lethality and a maternal phenotype (Van Auken et al., 2002). Other non-viable alleles that disrupt exon 1a include: a deletion allele that spans the region of the break point (ct344), an exon 1a single nucleotide polymorphism (SNP) allele that disrupts the initiation codon (t2012) and a deletion allele that spans exon 1a, exon 1b, exon 2 and exon 3 (s472) (Fig. 2A) (Van Auken et al., 2002). All alleles that are predicted to disrupt exon 1a are a 100% embryonic and L1 lethal with the exception of e917 (Van Auken et al., 2002). The mu232 allele is an SNP that disrupts the initiation codon (Met) in exon 1b and alters Q cell migration (Fig. 2A) (Yang et al., 2005). The e644 allele is an SNP that introduces a premature stop codon in exon 7b (Fig. 2A) (Van Auken et al., 2002). Since UNC-55 expression

begins at the onset of late L1, the alleles that caused embryonic or L1 lethality were not examined. We crossed each of the above viable alleles into a punc-55::GFP (jdIs101) background and scored the number of motor neurons present. To test the regulatory role of unc-62 isoforms containing exon 1a in the expression of unc-55, we utilized the e917 mutation. We observed a decrease in the number of AS and VD GFP positive neurons in e917;jdIs101 animals (Fig. 2C). To determine whether apoptotic events or lineage defects could explain the loss of punc-55::GFP expression in the e917 allele, we stained VNC nuclei with Hoechst. This revealed that some MNs were missing primarily in the anterior portion of the VNC (Supplemental Fig. 1B).However, the AS and VD MNs appeared to be present based upon positioning. We cannot completely rule out that the AS/VD MNs cell bodies were missing in the anterior portion of the VNC, however, in the mid body and posterior regions we noted that no nuclei were missing and punc-55::GFP was absent in the predicted nuclei for AS/VD MNs (Supplemental Fig 1A-C). This suggested that UNC-62 is required for UNC-55 expression in the AS and VD MNs. We next examined the unc-62 exon 1b mutant allele, mu232, in the jdIs101 background. The punc-55::GFP expression pattern in mu232;jdIs101 animals was identical to wild-type expression in the VNC (Fig. 2D). This suggested that unc-62 isoforms containing exon 1b had no observable role in unc-55 transcription. To examine the potential role of unc-62 transcripts containing exon 7b, we crossed e644, into the jdIs101 background. We observed a consistent increase in the number of GFP positive MNs in addition to the expected expression in the AS and VDs in e644;jdIs101 animals (Fig. 2F and 2G). However, the intensity of expression of punc-55::GFP in the ectopically expressing

MNs was noticeably fainter than that of the AS/VD MNs. This expression pattern was observed prior to the L1 molt, which correlated with the onset of UNC-55 expression in the VNC, however, expression in L2 and L3 was much more difficult to detect. To determine if the ectopic expression was associated with additional MNs in the VNC, we stained nuclei in e644;jdIs101 animals (Supplemental Fig. 1D-F). We observed that additional nuclei co-localized with the punc-55::GFP signal in addition to AS/VD MNs in e644 animals (Supplemental Fig. 1D-F). UNC-62 is required for VC survival in the mid body region in L3 animals (Potts et al., 2009). However, from our analysis we concluded that there was no gain or loss of MN nuclei during late L1 in e644 animals (Supplemental Fig. 1E). These results suggested that UNC-62 isoforms containing exon 7b are primarily required for the repression of unc-55 transcription in MN classes other than the AS/VD MNs.

Novel UNC-62 7a mutants generated with CRISPR/Cas9 We attempted to generate novel mutations in unc-62 exons 1a, 1b, 7a and 7b utilizing CRISPR/Cas9 via nonhomologous end joining. Unfortunately the guides targeting exon 1a, 1b and 7b did not produce viable animals. However we were able to recover two novel alleles in exon 7a of unc-62. The resulting mutants were isolated based upon reduced brood size and a disorganized gut, the latter, a phenotype that had been described previously using RNAi specific to exon 7A (Van Nostrand et al., 2013). Two mutants were isolated from this screen, jd1001 and jd1002, both of which were 4 bp deletions that caused a frame shift and premature stop codons in exon 7a. Since there were conflicting reports about the expression of UNC-62 exon 7a in the VNC (Spencer et al., 2011; Craig et al., 2013; Van Nostrand et al., 2013), we tested whether

UNC-62 isoforms carrying exon 7a were regulating punc-55::GFP transcription. We crossed the jd1002 allele into jdIs101 and the number of GFP neurons was scored. The resulting cross showed no significant decrease/increase in MNs expressing punc-55::GFP when compared to jdIs101 alone (Fig 2E and G). Additionally, jd1001 and jd1002 animals did not exhibit an uncoordinated phenotype further supporting the conclusion that unc-62 transcripts containing exon 7a do not regulate unc-55 transcription in the VNC.

Different UNC-62 isoforms repress and activate UNC-55 expression through distinct regions of the unc-55 promoter Based upon the promoter analysis of unc-55 and mutant analysis of unc-62, we reasoned that the differential regulation of unc-55 by UNC-62 may occur through a singular DNA binding motif or through different DNA binding motifs in the unc-55 promoter. Since the observations from punc-55-493 constructs suggested that it contains the minimal promoter for AS/VD expression, we sought to find UNC-62 bindings sites in this region. A potential UNC-62 binding site was identified 375 bp upstream of the 1st unc-55 exon. This site was predicted based upon the homology of unc-62 to the Meis1 Hox cofactor protein (Van Auken et al., 2002). The core motif of TGTC of the consensus sequence was changed to AAAA within the same 493 bp punc55::GFP (punc-55ΔMEIS -375) construct previously characterized (Fig. 3A). Variability in GFP expression was observed in the punc-55ΔMEIS -375 animals and expression was difficult to detect, even with long exposure times (Fig. 3B). Quantification of GFP positive MNs revealed a significant decrease in the number of GFP expressing MNs (Fig. 3C). This suggested that the

Meis motif located at 375 bp upstream of the 1st unc-55 exon was required for most but not all of the expression of UNC-55 in the AS and VD MNs. To determine if unc-62 acts through multiple regions of the unc-55 promoter or through a single region, we crossed the unc-62 alleles, e644 and e917, into the punc-55-1472, punc-55-1077, punc-55-493 and punc-55-366constructs. In performing this experiment, we sought to determine which regions of the unc-55 promoter were interacting with different isoforms. We observed a significant increase in GFP positive MNs in e644;punc-55-1472 constructs when compared to the punc-55-1472 alone (Fig. 3C). We also observed that e917; punc-55-1472 mirrored e917;jdIs101 animals in that there was a significant reduction in MN expression (Fig 2C and Fig 3D). These data correspond with the jdIs101 crosses with e644 and e917, validating our previous results with a different punc-55::GFP reporter construct. We predicted that if isoforms of UNC-62 disrupted by e644 were interacting with the region of the unc-55 promoter between -1077 and -1472, then we would expect to see no gain in expression in the number of GFP positive MNs. GFP positive MNs in e644;punc-55-1077 animals were not statistically different when compared to punc-55-1077animals suggesting that UNC-62 isoforms in which e644 disrupts are repressing unc-55 transcription through the -1077 and -1472 region (Fig. 3D). Surprisingly e917;punc-55-1077 exhibited similar levels of MN expression to that of e917; punc-55-1472animals (Fig. 3C), suggesting that the gain of GFP positive neurons in punc-55-1077animals was dependent upon a mechanism disrupted by the e917 mutation. To determine if the additional expression of punc-55::GFP in the e644 allele was due to a motif in between the -493 and -1077 bp region upstream of the first unc-55 exon, we crossed e644 and e917 into punc-55-493. Although there was a trend suggesting an increase in GFP

positive neurons in e644; punc-55-493, the increase was not statistically significant as compared to punc-55-493 alone, suggesting that removal of the region between 493-1077 bp was sufficient to partially rescue the ectopic GFP phenotype (Fig 3D). Interestingly, e917; punc-55-493animals showed no change in the number of GFP positive neurons in the VNC as compared to e917; punc-55-1077and e917; punc-55-1472 (Fig 3D). This result provided further evidence that isoforms disrupted by e917 activate unc-55 transcription through a region within 493 bp upstream of the first exon and potentially between the -493 and -1077 bp region of the unc-55 promoter. Lastly, we crossed e644 and e917 into the punc-55-366construct. Since punc-55-366had a complete loss of expression in the VNC, we expected this phenotype to be epistatic to the mutations in e644 and e917. With the exception of 3 animals out of 20 e917; punc-55-366 animals, all animals in both strains e917; punc-55-366 and e644; punc-55-366 had a complete loss of GFP expression in the VNC. Taken together these results suggest that UNC-62 isoforms that are disrupted by e917 most likely activate unc-55 transcription through a region within 366-493 bp (the same region in which site directed mutagenesis of a putative Meis site was performed) and 493-1077 bp upstream of the first exon of unc-55 in the AS/VD and other MNs respectively (Fig 3E). UNC-62 isoforms disrupted by e644 likely repress unc-55 transcription through a region 1077-1472bp 5’ to the first exon (Fig 3E). UNC-62 is required for full UNC-55 mRNA and protein expression in AS/VD motor neurons Transcriptional reporters are known to only partially represent the natural expression of the mRNA and proteins they represent and may produce artefactual expression. Indeed, the lack of expression or ectopic expression of punc-55::GFP in the VNC could be due to transcriptional

reporter artifacts. To address this issue we tested unc-55 mRNA levels via whole animal qPCR, UNC-55 protein expression via a CRSIPR/Cas9 based reporter and smFISH for localized unc-55 mRNA expression patterns. To determine if the global mRNA levels were disrupted in different unc-62 alleles, qPCR was performed for unc-55 transcripts. Based upon the unc-62 mutant and promoter analysis, we predicted that e917 animals would have reduced relative quantities (RQ) of unc-55 mRNA. Likewise, we expected that e644 animals would have elevated levels of expression of unc-55 mRNA while the mu232 and jd1002 mutants would have no change in mRNA quantities as compared to wildtype animals. The resulting relative quantities of unc-55 from four replicates indicated no change in relative expression level between N2 and mu232 (Fig. 4). Surprisingly we saw a 1.7 fold reduction of unc-55 mRNA in e644 and a 5.3 fold reduction in jd1002 (Fig. 4). This is not consistent with the observation of increased punc-55::GFP expression in the VNC and no change in expression in e644 and jd1002 animals respectively. The RQ of the e917 allele was similar to the observations of e917;punc-55::gfp animals showing a 2.1 fold reduction compared to N2 (Fig. 4). To address the discrepancies between the unc-55 qPCR result and the unc-55 transcriptional reporter data, we tested UNC-55 protein expression. This was achieved via an in genome CRISPR/Cas9 homologous recombined reporter with GFP fused to the C-terminal end of UNC-55 (a kind gift from Ge Shan). The localization of UNC-55::GFP expression in the VNC was consistent with transcriptional reporters in a N2 background. Additionally, in an e917 background, UNC-55::GFP mimicked the reduction of expression in the number of cells and overall transcripts seen in the transcriptional reporter and qPCR (Fig 4C and E). However, UNC55::GFP expression in the e644 background was very faint and difficult to detect (Fig. 4D). The

number of neurons expressing UNC-55::GFP was variable and the mean number of cells expressing UNC-55::GFP was reduced in the e644 background, indicating that it may correlate with the qPCR expression observed (Fig 4E). However, the localization of unc-55 transcripts may be different than that of the UNC-55 protein. We performed smFISH for unc-55 in e644 and e917 backgrounds in late L1 animals in order to determine if the unc-55 transcript expression pattern in MNs was similar to the promoter constructs or the UNC-55 translation reporter (Fig 5). We measured two sets of data: 1) the number of nuclei in the VNC associated with smFISH signal (Fig 5J) and 2) the overall amount of smFISH signal in the VNC (Fig 5K). We determined that N2 animals had more nuclei associated with smFISH signal than was expected. The AS/VD MNs total 25 cells, whereas our analysis yielded a median of 33 nuclei associated with smFISH signal. This may be confounded by the general organization of the VNC at late L1, since many postembryonic MNs have just finished dividing and are crowded (Fig 5B). Similar to the qPCR and transcriptional reporters for unc-55, all e917 animals showed a reduced number of nuclei associated with unc-55 smFISH signal and overall smFISH signal in the VNC as compared to N2 (Fig J and K). However, nuclei counts in e644 had more variation than that of N2 and e917. This variation ranged from a reduction to similar or increased number of nuclei counted. As with the cell counts, the quantity of smFISH signal in the VNC in e644 animals was variable and ranged from higher to lower than that of the total smFISH signals counted in N2 (Fig J and K). The variability in mRNA and protein expression of unc-55 suggested that the e644 allele may result in instability of the unc-55 transcription between MNs rather than disrupting activation or repression. Role of unc-62 alleles in VD motor neuron synaptic differentiation

UNC-55 is required for the proper synaptic targeting of the VD MNs and in the absence of UNC-55; the VD MNs innervate dorsal muscle instead of ventral muscle (Zhou and Walthall, 1998). If unc-55 transcriptional activation is disrupted in an e917 unc-62 mutant background, then we would predict a reduction in the number of detectable synapses on ventral muscle and relocation of synapses to the dorsal muscle. Similarly, we sought to address the variability observed in UNC-55 expression in the e644 allele as well. If there was a loss in UNC-55 expression in e644 mutants, then we would expect a reduction in ventral presynaptic varicosities (puncta) associated with VD MNs. However, it has been previously reported and confirmed by our own observations that unc-62 mutants display defects in VD/DD commissures, thus we quantified ventral puncta from the VD MNs in e644 and e917 animals (Siddiqui, 1990) (Fig. 6). The unc-62 alleles, e644 and e917 were crossed into the transgenic strain punc-25::snb1::gfp(juIs1) to determine if presynaptic varicosities and cellular morphology of GABAergic MNs were disrupted (Hallam and Jin, 1998) (Fig 6). The puncta were counted between VD MNs in L4 animals as previously described (Petersen et al., 2011). A mean of 23 puncta was observed in between VD regions in wild-type animals (Fig 6A and 6D). We observed no significant change in VD puncta in the e644;punc-25::snb-1::gfp strain as compared to wild type (Fig 6B and 6D). Since we observed no change in ventral puncta in an e644 mutant background, this suggested that the variability in UNC-55 expression was not sufficient to alter the synaptic targeting of the VD MNs. However, the number of puncta observed in e917;punc-25::snb-1::gfp animals was significantly reduced as compared to both wild-type and e644 animals (ANOVA with Sheffe’s post hoc n=20 VD regions p<0.001) (Fig. 6C and 6D). This suggested that the unc62 transcripts disrupted by the e917 allele altered the synaptic specificity of the VD MNs. We also observed cases in which the dorsal nerve cord (DNC) appeared to have increased puncta in

the regions where the VD puncta were reduced (data not shown). However, we did not quantify the DNC puncta due to observations of commissural defects in the VD and DD MNs. We also observed cases in which some of the DD MNs failed to express GFP suggesting that UNC-62 may also contribute to establishing the GABAergic identity. DISCUSSION How subtype determining transcription factors are regulated represents an important question in developmental biology. This study demonstrates that the VD subtype determining transcription factor, UNC-55, is both transcriptionally activated and stabilized by different isoforms of the Meis/Tale transcription factor homologue UNC-62. This conclusion is supported by the four molecular lesions associated with different unc-62 alternative transcribed/spliced exons (Fig. 2A) and the resulting three different punc-55::GFP expression patterns observed (Fig. 2B-F). We also found that UNC-62 transcriptional regulation of unc-55 acts through distinct regions of the unc-55 promoter, suggesting that the unc-55 promoter contains multiple cis elements responsible for expression in different MNs and at least one repressive region. The consistency of phenotypes between the promoter dissection analysis and the unc-62 mutant analysis supports the hypothesis that UNC-62 differentially regulates UNC-55 expression through distinct regions of the unc-55 promoter. Additionally, the UNC-55 transcriptional reporter, mRNA and protein expression was consistently reduced in e917 mutant backgrounds. Given that UNC-62 is not required for the complete expression of unc-55 in the VNC, we assume that additional cofactors or other transcription factors work in conjunction with UNC-62 to regulate unc-55 expression. However, the disruption of the putative Meis binding site in the unc-55 promoter resulted in a larger loss of punc-55::GFP positive cells as compared to e917 animals. This discrepancy could be the result of lower expression levels of unc-62 transcripts in

e917 animals, suggesting that e917 is a hypomorphic allele. Indeed, e917 is 97% lethal, whereas other alleles that disrupt unc-62 exon 1 or the putative enhancer site that contains the e917 breakpoint are not viable (Van Auken et al., 2002). Additionally, quantitative analysis of unc-55 transcripts and UNC-55 localization only partially matched the results of the unc-55 promoter constructs observed in the e644 mutant allele. Thus, the transcriptional reporters for unc-55 may not be capturing all of the regulatory motifs. This could explain why there was a consistent increase in punc-55::GFP in an e644 background but more variable expression was observed when analyzing mRNA and UNC-55::GFP localization. This variability suggests that isoforms of UNC-62 that contain exon 7b play a key role in stabilizing unc-55 transcription. This stabilization could be the result of a non-randomizing action by UNC-62 isoforms. UNC-62 isoforms containing exon 7b could be required for maintaining wild type levels of unc-55 transcription in the VD/AS MNs, while repressing UNC-55 expression in other MNs. However, our evidence for this mechanism is circumstantial. This mechanism is similar to the stabilizing role played by ALR-1 in mechanosensory neuron gene expression, however, the stabilization ALR-1 target genes was involved in enhancement of expression and not repression (Topalidou et al., 2011). Furthermore, proper VD MN synaptic patterning and activation of unc-55 transcription is dependent on UNC-62 exon 1a isoforms, suggesting that UNC-55 expression and VD MN differentiation is dependent upon UNC-62 exon 1a (Walthall and Plunkett, 1995; Shan et al., 2005). Our data indicate that a subset of UNC-62 isoforms activate unc-55 transcription in the VD MNs; whereas a second subset both repress and activate unc-55 expression in AS/VD and non-AS/VD MNs. Isoform specific regulation of UNC-62 on transcriptional targets is already known. Van Nostrand and colleagues in 2013 demonstrated that isoforms containing exon 7a of

UNC-62 were necessary for longevity and transcriptional activation of yolk proteins in the gut, whereas isoforms containing exon 7b were not. In the same study it was shown that the alternatively spliced exons 7a and 7b are differentially expressed in the nervous system and gut, further suggesting different modes and sites of action for UNC-62 isoforms (Craig et al., 2013; Van Nostrand et al., 2013). UNC-62 physically interacts with CEH-20 and other Hox cofactors via the HM domain (Jiang et al., 2009). Interestingly, the HM domain is differentially transcribed by exons 1a and 1b, suggesting that different isoforms of UNC-62 may have different protein to protein interactions with cofactors. In terms of overall regulation of MN subtype differentiation, UNC-62 has already been identified as establishing VC and CP cellular identities in hermaphrodites and males respectively (Kalis et al., 2014). Furthermore, evidence that UNC-62 is required for correct mechanosensory subtype differentiation and the expression of pan neuronal genes provides interesting clues into how upstream GRNs are orchestrating differentiation of neurons (Stefanakis et al., 2015; Zheng et al., 2015). Since UNC-62 is a Hox cofactor, these findings provide evidence that MN subtype differentiation is linked to a potential Hox regulatory network in the VNC of C. elegans. This suggests conservation in at least part of the regulatory programs required for MN subtype differentiation between nematodes, insects and vertebrate species. An even more intriguing prospect is that of isoform specific mechanisms to generate diversity in MN subtypes. It is clear that isoforms play a role in the development of the nervous system. In silico analyses of predicted and verified splice variants via RNA sequencing suggest that there is a significant increase in splice variations in the human brain as compared to other species and organs (Barbosa-Morais et al., 2012; Merkin et al., 2012). The combination of conserved GRNs and isoform specific regulation of downstream targets provides a mechanism for not only creating

diverse subtypes among functionally related cellular networks, but also could provide a conserved mechanism for generating subtle variation among functionally related cells within a cellular network. AUTHOR CONTRIBUTIONS R.F.C. performed all experiments and data analysis under the supervision of W.W.W. R.F.C. and W.W.W. wrote and edited the manuscript. ACKNOWLEDGEMENTS We would like to thank: Dr. Casonya Johnson for her guidance, advice and editing of the manuscript; Han Ting Chou for advice and guidance; Dr. Anne Murphy for her expert advice with statistical analyses; Walthall lab members for comments on the manuscript. We are very grateful to the C. elegans community as a whole. This work has been funded in part by the Neurogenomics Center, Brains and Behavior Area of Focus Grant and Biology Department of Georgia State University. We are very thankful for the Brains and Behavior Program that provided financial support to R.F.C. REFERENCES Barbosa-Morais, N. L., Irimia, M., Pan, Q., Xiong, H. Y., Gueroussov, S., Lee, L. J., Slobodeniuc, V., Kutter, C., Watt, S., Colak, R. et al. (2012). The evolutionary landscape of alternative splicing in vertebrate species. Science 338, 1587-1593. Cartharius, K., Frech, K., Grote, K., Klocke, B., Haltmeier, M., Klingenhoff, A., Frisch, M., Bayerlein, M. and Werner, T. (2005). MatInspector and beyond: promoter analysis based on transcription factor binding sites. Bioinformatics 21, 2933-2942.

Certel, S. J. and Thor, S. (2004). Specification of Drosophila motoneuron identity by the combinatorial action of POU and LIM-HD factors. Development 131, 5429-5439. Craig, H. L., Wirtz, J., Bamps, S., Dolphin, C. T. and Hope, I. A. (2013). The significance of alternative transcripts for Caenorhabditis elegans transcription factor genes, based on expression pattern analysis. BMC genomics 14, 249. Dasen, J. S., Tice, B. C., Brenner-Morton, S. and Jessell, T. M. (2005). A Hox regulatory network establishes motor neuron pool identity and target-muscle connectivity. Cell 123, 477-491. Dickinson, D. J., Ward, J. D., Reiner, D. J. and Goldstein, B. (2013). Engineering the Caenorhabditis elegans genome using Cas9-triggered homologous recombination. Nat Methods 10, 1028-1034. Eastman, C., Horvitz, H. R. and Jin, Y. (1999). Coordinated transcriptional regulation of the unc-25 glutamic acid decarboxylase and the unc-47 GABA vesicular transporter by the Caenorhabditis elegans UNC-30 homeodomain protein. J Neurosci 19, 62256234. Esmaeili, B., Ross, J. M., Neades, C., Miller, D. M., 3rd and Ahringer, J. (2002). The C. elegans even-skipped homologue, vab-7, specifies DB motoneurone identity and axon trajectory. Development 129, 853-862. Gerstein, M. B. Lu, Z. J. Van Nostrand, E. L. Cheng, C. Arshinoff, B. I. Liu, T. Yip, K. Y. Robilotto, R. Rechtsteiner, A. Ikegami, K. et al. (2010). Integrative analysis of

the Caenorhabditis elegans genome by the modENCODE project. Science 330, 17751787. Hallam, S. J. and Jin, Y. (1998). lin-14 regulates the timing of synaptic remodelling in Caenorhabditis elegans. Nature 395, 78-82. He, S., Philbrook, A., McWhirter, R., Gabel, C. V., Taub, D. G., Carter, M. H., Hanna, I. M., Francis, M. M. and Miller, D. M., 3rd. (2015). Transcriptional Control of Synaptic Remodeling through Regulated Expression of an Immunoglobulin Superfamily Protein. Curr Biol 25, 2541-2548. Hobert, O. (2002). PCR fusion-based approach to create reporter gene constructs for expression analysis in transgenic C. elegans. Biotechniques 32, 728-730. Howell, K., White, J. G. and Hobert, O. (2015). Spatiotemporal control of a novel synaptic organizer molecule. Nature 523, 83-87. Hsu, P. D., Scott, D. A., Weinstein, J. A., Ran, F. A., Konermann, S., Agarwala, V., Li, Y., Fine, E. J., Wu, X., Shalem, O. et al. (2013). DNA targeting specificity of RNAguided Cas9 nucleases. Nature biotechnology 31, 827-832. Jiang, Y., Shi, H. and Liu, J. (2009). Two Hox cofactors, the Meis/Hth homolog UNC62 and the Pbx/Exd homolog CEH-20, function together during C. elegans postembryonic mesodermal development. Developmental biology 334, 535-546. Jin, Y., Hoskins, R. and Horvitz, H. R. (1994). Control of type-D GABAergic neuron differentiation by C. elegans UNC-30 homeodomain protein. Nature 372, 780-783.

Kalis, A. K., Kissiov, D. U., Kolenbrander, E. S., Palchick, Z., Raghavan, S., Tetreault, B. J., Williams, E., Loer, C. M. and Wolff, J. R. (2014). Patterning of sexually dimorphic neurogenesis in the caenorhabditis elegans ventral cord by Hox and TALE homeodomain transcription factors. Dev Dyn 243, 159-171. Kratsios, P., Stolfi, A., Levine, M. and Hobert, O. (2012). Coordinated regulation of cholinergic motor neuron traits through a conserved terminal selector gene. Nat Neurosci 15, 205-214. Merkin, J., Russell, C., Chen, P. and Burge, C. B. (2012). Evolutionary dynamics of gene and isoform regulation in Mammalian tissues. Science 338, 1593-1599. Miller, D. M., Shen, M. M., Shamu, C. E., Burglin, T. R., Ruvkun, G., Dubois, M. L., Ghee, M. and Wilson, L. (1992). C. elegans unc-4 gene encodes a homeodomain protein that determines the pattern of synaptic input to specific motor neurons. Nature 355, 841-845. Mueller, F., Senecal, A., Tantale, K., Marie-Nelly, H., Ly, N., Collin, O., Basyuk, E., Bertrand, E., Darzacq, X. and Zimmer, C. (2013). FISH-quant: automatic counting of transcripts in 3D FISH images. Nat Methods 10, 277-278. Park, M., Watanabe, S., Poon, V. Y., Ou, C. Y., Jorgensen, E. M. and Shen, K. (2011). CYY-1/cyclin Y and CDK-5 differentially regulate synapse elimination and formation for rewiring neural circuits. Neuron 70, 742-757.

Petersen, S. C., Watson, J. D., Richmond, J. E., Sarov, M., Walthall, W. W. and Miller, D. M., 3rd. (2011). A transcriptional program promotes remodeling of GABAergic synapses in Caenorhabditis elegans. J Neurosci 31, 15362-15375. Potts, M. B., Wang, D. P. and Cameron, S. (2009). Trithorax, Hox, and TALE-class homeodomain proteins ensure cell survival through repression of the BH3-only gene egl-1. Developmental biology 329, 374-385. Schugn, J. (2003). Using TESS to Predict Transcription Factor Binding Sites in DNA Sequence. Current Protocols in Bioinformatics. Shan, G., Kim, K., Li, C. and Walthall, W. W. (2005). Convergent genetic programs regulate similarities and differences between related motor neuron classes in Caenorhabditis elegans. Developmental biology 280, 494-503. Siddiqui, S. S. (1990). Mutations affecting axonal growth and guidance of motor neurons and mechanosensory neurons in the nematode Caenorhabditis elegans. Neuroscience research. Supplement : the official journal of the Japan Neuroscience Society 13, S171-190. Spencer, W. C., Zeller, G., Watson, J. D., Henz, S. R., Watkins, K. L., McWhirter, R. D., Petersen, S., Sreedharan, V. T., Widmer, C., Jo, J. et al. (2011). A spatial and temporal map of C. elegans gene expression. Genome Res 21, 325-341. Stefanakis, N., Carrera, I. and Hobert, O. (2015). Regulatory Logic of Pan-Neuronal Gene Expression in C. elegans. Neuron 87, 733-750.

Thompson-Peer, K. L., Bai, J., Hu, Z. and Kaplan, J. M. (2012). HBL-1 patterns synaptic remodeling in C. elegans. Neuron 73, 453-465. Topalidou, I., van Oudenaarden, A. and Chalfie, M. (2011). Caenorhabditis elegans aristaless/Arx gene alr-1 restricts variable gene expression. Proc Natl Acad Sci U S A 108, 4063-4068. Van Auken, K., Weaver, D., Robertson, B., Sundaram, M., Saldi, T., Edgar, L., Elling, U., Lee, M., Boese, Q. and Wood, W. B. (2002). Roles of the Homothorax/Meis/Prep homolog UNC-62 and the Exd/Pbx homologs CEH-20 and CEH40 in C. elegans embryogenesis. Development 129, 5255-5268. Van Nostrand, E. L., Sanchez-Blanco, A., Wu, B., Nguyen, A. and Kim, S. K. (2013). Roles of the developmental regulator unc-62/Homothorax in limiting longevity in Caenorhabditis elegans. PLoS Genet 9, e1003325. Vrieseling, E. and Arber, S. (2006). Target-induced transcriptional control of dendritic patterning and connectivity in motor neurons by the ETS gene Pea3. Cell 127, 14391452. Walthall, W. W. and Plunkett, J. A. (1995). Genetic transformation of the synaptic pattern of a motoneuron class in Caenorhabditis elegans. J Neurosci 15, 1035-1043. Westmoreland, J. J., McEwen, J., Moore, B. A., Jin, Y. and Condie, B. G. (2001). Conserved function of Caenorhabditis elegans UNC-30 and mouse Pitx2 in controlling GABAergic neuron differentiation. J Neurosci 21, 6810-6819.

White, J. G., Albertson, D. G. and Anness, M. A. (1978). Connectivity changes in a class of motoneurone during the development of a nematode. Nature 271, 764-766. White, J. G., Southgate, E., Thomson, J. N. and Brenner, S. (1976). The structure of the ventral nerve cord of Caenorhabditis elegans. Philos Trans R Soc Lond B Biol Sci 275, 327-348. Winnier, A. R., Meir, J. Y., Ross, J. M., Tavernarakis, N., Driscoll, M., Ishihara, T., Katsura, I. and Miller, D. M., 3rd. (1999). UNC-4/UNC-37-dependent repression of motor neuron-specific genes controls synaptic choice in Caenorhabditis elegans. Genes Dev 13, 2774-2786. Yang, L., Sym, M. and Kenyon, C. (2005). The roles of two C. elegans HOX co-factor orthologs in cell migration and vulva development. Development 132, 1413-1428. Zheng, C., Diaz-Cuadros, M. and Chalfie, M. (2015). Hox Genes Promote Neuronal Subtype Diversification through Posterior Induction in Caenorhabditis elegans. Neuron 88, 514-527. Zhou, H. M. and Walthall, W. W. (1998). UNC-55, an orphan nuclear hormone receptor, orchestrates synaptic specificity among two classes of motor neurons in Caenorhabditis elegans. J Neurosci 18, 10438-10444.

Figure 1: Serial promoter deletions of the unc-55 promoter result in increased and decreased expression of punc-55::GFP. A) The unc-55 locus (top). A schematic of promoter constructs injected into N2 animals. Yellow ticks indicate predicted Meis/Tale binding sites (below). B) A cartoon depicting a late L1 animal. The dotted rectangle indicates the region represented in the micrographs. C-G) Expression of GFP from each of the promoter constructs injected. Arrows indicate expected position of AS/VD. Each image was cropped from original image to emphasize the MNs (scale = 20µm). H) Quantification of GFP positive MNs from promoter constructs. The mean+/-s.e.m. of two independent transgenic lines for each construct was examined (Brown-Forsythe ANOVA with Sheffe’s Post Hoc * = p<0.05 n≥20 for each line). (**= MN expression was 0) Figure 2: Mutations in alternatively spliced/transcribed unc-62 exons caused increases and decreases in punc-55::gfp expression. A) Schematic of unc-62 alleles disrupting alternatively spliced/transcribed exons. B-F) punc-55::gfp in different unc-62 alleles. Arrows indicate expected positions of AS/VD neurons. All animals were synchronized to late L1 stage (scale = 20µm). G) Quantification of the mean +/- s.e.m. of motor neurons expressing punc-55::GFP in VNC in wildtype (WT) and unc-62 alleles (Brown-Forsythe ANOVA with Games Howell Post Hoc * = p<0.01 n≥20). Figure 3: UNC-62 works through a Meis binding site and multiple other regions of the unc55 promoter. A) The unc-55 promoter construct in which the putative Meis site was altered. B) A micrograph of an animal containing the ΔMEIS 493 bp punc-55::GFP construct. Arrows

indicate expected positions of AS/VD motor neurons. The dotted circle indicates the expression of punc-55::GFP in a single cell. C) The mean+/- s.e.m. of GFP positive motor neurons in the VNC of late L1 animals in a 1472 bp and ΔMEIS at 375 bp constructs (Student’s t-test p<0.05). D) Quantification of GFP positive MNs in each promoter construct crossed with e644 or e917 (Brown-Forsythe ANOVA with Games-Howell Post hoc n=20 p<0.05) (* = p<0.05). E) Predicted model for UNC-62 isoform interactions with the promoter. E) A model of UNC-62 transcriptional regulation of unc-55 based upon the results of the promoter analysis and unc-62 mutant analysis of punc-55::GFP expression. Figure 4: UNC-55 RNA and protein expression in unc-62 mutant alleles. A) Whole animal qPCR for unc-55 of late L1 animals in unc-62 alleles (four replicates). B) Expression of CRISPR/Cas9 homologous recombined UNC-55::GFP in late L1 animals. C) Late L1 UNC55::GFP;e917. D) Late L1 UNC-55::GFP;e644 animal (scale =50 µm) E) Quantification of the total number of GFP positive neurons in all alleles. Each black circle represents the count of GFP positive neurons for a single animal. Figure 5: smFISH for unc-55 in unc-62 mutant alleles. A) smFISH for unc-55 in a N2 background. The image was filtered via FISHquant using a kernel Gaussian method for removal of background and enhancement of smFISH signal. B) DAPI counterstain of N2 animal used to determine the number of nuclei associated with smFISH. C) Merge of smFISH and DAPI. D-E) e917 mutant animals stained for unc-55 smFISH and DAPI. G-I) e644 mutant animals stained for unc-55 smFISH and DAPI. (scale =50 µm) J) Quantification of the number of nuclei associated with smFISH signal in N2 and unc-62 mutants. Each black circle represents a single animal. K) Total mRNA in the VNC of in N2 and unc-62 mutants.

Figure 6: UNC-62 isoforms are responsible for synaptic localization of the VD MNs in the VNC. A-C) e644 and e917 in a punc-25::gfp::snb-1 reporter background. Puncta were counted between VD MNs in the VNC. Dashed circles indicate VD or DD cell bodies. Arrows indicate the puncta on the VNC used for quantification. (scale =20 µm) D) Quantification of the mean+/s.e.m. puncta per VD anterior region (ANOVA with Sheffe’s post hoc n=20 VD regions p<0.001)

Highlights   

Sought to understand the transcriptional regulation of CoupTFII unc-55 UNC-62 isoform specific regulation of unc-55 transcription was proposed based upon alternatively transcribed/spliced exon specific mutations UNC-62 isoforms differentially regulate the mRNA and protein expression of UNC-55

M E TA IS LE 375 M 75 bp E M IS 6 b EI 9 p S 24 1 TA 03 bp LE 1 b 13 p 19 bp

Figure 1

A unc-55 (F55D12.4)

gfp

1932 bp

gfp

1472 bp

gfp gfp gfp

B

punc-55::gfp expression in the VNC

H 35

GFP Expression in the VNC of unc-55 Promoter Dissection Constructs

*

Average GFP MNs

30 25 20 15 10 5 0

** 1932 bp

1472 bp

1077 bp

493 bp

unc -55 Promoter Constructs

366 bp

1077 bp 493 bp 366 bp

Figure 2

A

e917

//

1b

2

34

5

6

mu232

7a 7b 8

9

10

jd1002 e644

G

35

*

30

Average GFP MNs

unc-62 (T28F12.2) 1a

25 20

*

15 10 5 0

WT

e917

mu232 Alleles

jd1002

e644

Figure 3

C

EI S

37

5

bp

A ΔM

x 493 bp

25 Average GFP MNs

gfp

30

B

20 15

*

10 5 0

unc-55 Promoter Constructs

D

Promoter Constructs of unc-55 in unc-62 mutant backgrounds

Average Motor Neurons in VNC

40

n.s .

35

UNC-62 exon 1a UNC-62 exon 7b

n.s .

30

MEIS Binding Site

N2

25

e644

20

e917

15

unc-55 (F55D12.4)

-493 bp

10 5 0

E

1472 bp

1077 bp

493 bp

366 bp

-1077 bp

-1472 bp

N

U

:G

-5 5:

FP

;e 64 4

;e 91 7

FP

::G

-5 5

C

C

N

jd1002

U

FP

:G

E

-5 5:

mu232

C

e917

N

N2

U

0

GFP Positive MNs

RQ of unc-55 Compared to N2

Figure 4 A

1.2 1

0.8

0.6

0.4

0.2

e644

Alleles

35

30

25

20

15

10

5

0

Figure 5

J

Nuclei

80 60 40 20 0

N2

e917

e644

Total mRNA in VNC 500

smFISH counts

Nuclei localized with smFISH

K

400 300 200 100 0

N2

e917

e644

Figure 6

D

Puncta per VD Anterior Region

mean Puncta

30 25 20

*

15 10 5 0

N2

e644 Allele

e917