Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Apr 20.
Published in final edited form as: Cell Rep. 2016 Apr 7;15(3):550–562. doi: 10.1016/j.celrep.2016.03.051

PROS-1/Prospero Is a Major Regulator of the Glia-Specific Secretome Controlling Sensory-Neuron Shape and Function in C. elegans

Sean W Wallace 1, Aakanksha Singhvi 1, Yupu Liang 2, Yun Lu 1, Shai Shaham 1,*
PMCID: PMC4838487  NIHMSID: NIHMS771566  PMID: 27068465

SUMMARY

Sensory neurons are an animal’s gateway to the world, and their receptive endings, the sites of sensory signal transduction, are often associated with glia. While glia are known to promote sensory-neuron functions, the molecular bases of these interactions are poorly explored. Here we describe a post-developmental glial role for the PROS-1/Prospero/PROX1 homeodomain protein in sensory-neuron function in C. elegans. Using glia expression profiling, we demonstrate that, unlike previously characterized cell fate roles, PROS-1 functions post-embryonically to control sense-organ glia-specific secretome expression. PROS-1 functions cell autonomously to regulate glial secretion and membrane structure, and non-cell autonomously to control the shape and function of the receptive endings of sensory neurons. Known glial genes controlling sensory-neuron function are PROS-1 targets, and we identify additional PROS-1-dependent genes required for neuron attributes. Drosophila Prospero and vertebrate PROX1 are expressed in post-mitotic sense-organ glia and in astrocytes, suggesting conserved roles for this class of transcription factors.

graphic file with name nihms-771566-f0001.jpg

INTRODUCTION

Animals use specialized sense organs to obtain information from their environment. Sense organs, in general, consist of sensory neurons or receptor cells, the primary mediators of sensory signal transduction, and glia-like support cells, which have been defined anatomically for decades but have been relatively poorly studied (Cuschieri and Bannister, 1975; Engstrom, 1967; Murray, 1993). Glia-like cells in the mammalian taste bud, olfactory epithelium, cochlea and retina, classified as type I cells, sustentacular cells, Deiters’ cells and RPE (retinal pigment epithelial) cells, respectively, show a number of functions typically associated with glia. These include regulating neurotransmitter signaling (Bartel et al., 2006; Dooley et al., 2011; Hegg et al., 2009; Lawton et al., 2000; Matsunobu et al., 2001), buffering extracellular potassium levels (Boettger et al., 2002; Dvoryanchikov et al., 2009; Trotier, 1998) and releasing neurotrophic factors and neuromodulators (Breunig et al., 2010; Hansel et al., 2001; Strauss, 2005). Thus, functional similarities between sense-organ and CNS glia, and similarities between sensory-neuron receptive endings and synapses (Shaham, 2010), suggest that understanding sensory neuron-glia interactions may have broad relevance.

C. elegans is an attractive model system for studying glia-neuron interactions (Oikonomou and Shaham, 2011; Shaham, 2006, 2015). Its simple anatomy, facile genetics, optical transparency, and well-characterized behavioral responses allow for single-cell-resolution in vivo analyses of the contributions that glial cells make to nervous system function. The C. elegans amphids are a pair of bilaterally symmetric head sense organs that mediate responses to multiple sensory inputs (Bargmann et al., 1993; Bargmann and Horvitz, 1991; Kimura et al., 2004; Ward, 1973). Each amphid consists of 12 sensory neurons, possessing specialized receptive endings that respond to distinct cues. These neurons associate with two glial cells, the AMsh and AMso glia, that together form a matrix-filled channel through which sensory cilia of eight neurons are exposed to the external environment (Perkins et al., 1986). The remaining four neurons have elaborately-structured receptive endings that are ensheathed by the AMsh glial cell. These glia-ensheathed receptive endings include the branched wing-like cilia of the odor-sensing neurons AWA, AWB and AWC, as well as the microvilli-like fingers of the AFD thermosensory neuron (Doroquez et al., 2014; Perkins et al., 1986) (Figure 1A)

Figure 1. AMsh glia-enriched genes.

Figure 1

Open in a new tab

(A) Amphid structure. (B) AMsh glia transcripts at >3.5-fold or >10-fold over all other cells. (C) Enrichment levels of known and new genes. (D) AMsh glia-specific expression of indicated genes. Scale bar,10 μm. See also Table S1.

We previously showed that post-embryonic ablation of AMsh glia, after the amphid had formed, results in defects in the associated sensory neurons (Bacaj et al., 2008). AMsh glia are required for the proper morphology of the AWA, AWC and AFD sensory neurons, and for dye-uptake by cilia of channel neurons, which serves to report on neuronal integrity and access to the environment. Glial ablation also results in defects in behaviors mediated by these neurons. However, the molecular mechanisms controlling these interactions were not well understood.

Here, we characterize the molecular basis for the post-embryonic regulation of sensory-neuron morphology and function by AMsh glia. We developed a larval glial-cell-isolation method using fluorescence-activated cell sorting (FACS), followed by RNA sequencing, to identify 598 AMsh glia-enriched transcripts. About 90% of highly-enriched AMsh genes encode secreted or transmembrane proteins, consistent with a role in regulating neuronal properties. A post-embryonic RNAi screen of a subset of these genes led to the identification of the homeodomain transcription factor pros-1 as a glial regulator of neuronal morphology and function. Expression studies, cell-specific RNAi experiments, and mosaic rescue experiments confirm that pros-1 acts in glia to control sensory-neuron shape and function. To characterize the mechanisms mediating pros-1 functions, we identified pros-1 target genes using differential gene-expression analyses of AMsh glia isolated from pros-1 RNAi-treated and control animals. About a third of AMsh-enriched genes encoding secreted or transmembrane proteins require PROS-1 for their expression, revealing a key function for this gene in regulating expression of the glia-specific secretome. PROS-1 targets include genes with known roles in amphid morphology and function, and RNAi screens we performed identify new genes that contribute to pros-1 function. Thus, PROS-1 is a key transcriptional regulator governing the expression of secreted and membrane proteins in AMsh glia.

Prospero, the Drosophila homolog of pros-1, plays well-known roles in the establishment of cell fate during nervous system development (Doe et al., 1991; Freeman and Doe, 2001). Importantly, we found that while postembryonic inactivation of pros-1 by RNAi promotes profound defects in sensory-neuron function, general aspects of glial cell fate are unperturbed, thus revealing functions for this classical cell fate gene in differentiated glia. Prospero homologs are expressed in glia associated with sense organs in many species, including sheath cells in Drosophila external sense organs (Manning and Doe, 1999), mechanosensory support cells in fish neuromasts (Pistocchi et al., 2009), and Deiters’ cells in the mammalian ear (Bermingham-McDonogh et al., 2006), and are also expressed in astrocytes in Drosophila and mammals (Peco et al., 2016) (Zhang et al., 2014). Thus, our studies of pros-1 function in C. elegans provide general insights into the regulation of neuronal function by glia.

RESULTS

Identification of transcripts enriched in post-embryonic AMsh glia

To identify glial genes controlling sensory neuron function, and not development, we isolated amphid sheath (AMsh) glia from late-stage C. elegans larvae and performed gene expression analysis (see Experimental Procedures). Cells expressing the AMsh glia-specific reporter F16F9.3::dsRed were sorted by FACS and subjected to RNA extraction and purification. mRNA was amplified and used to generate a cDNA library for sequencing. Quantitative gene expression profiling was carried out to identify genes enriched in AMsh glia, by comparing AMsh glia to all other cell types not expressing dsRed. This approach led to the identification of 598 candidate AMsh glia-enriched genes (fold enrichment > 3.5, p-adj <0.05, see Experimental Procedures for statistical analysis and cutoff rationale). Of these 598 genes, 383 (64%) encode proteins with signal peptides or transmembrane domains (Table S1) (Figure 1B). This is significantly higher than the prevalence of such proteins in the C. elegans genome (45%) (Suh and Hutter, 2012). Strikingly, of 177 highly enriched AMsh genes (fold enrichment > 10), 155 genes (88%) encode proteins with signal peptides or transmembrane domains (Figure 1B). To validate this gene list, we checked the expression levels of six known AMsh glia-specific genes (Bacaj et al., 2008). All six genes showed strong enrichment in AMsh glia in our analysis (Figure 1C), confirming that we successfully isolated larval AMsh glial transcripts. We also made transcriptional reporters with upstream regulatory regions for 2 genes, K02E11.4 and R11D1.3, both of which are highly enriched in AMsh glia in our analysis, but not previously studied. Transcriptional reporters for both genes showed specific expression in AMsh glia (Figure 1D), as well as in the analogous phasmid sheath glia in the tail, confirming that we successfully identified AMsh glial genes.

Glia-enriched pros-1/Prospero is required for sensory-neuron function

To uncover the molecular pathways through which AMsh glia regulate functional properties of associated sensory neurons, we screened animals for sensory deficits following post-embryonic RNAi against 140 of the 598 AMsh glia-enriched genes. Following 3 days of RNAi treatment, first-day adults were assessed for defects in chemosensation, using standard chemotaxis and avoidance assays. RNAi targeting mRNAs encoding the Prospero-related transcription factor, PROS-1, results in defects in all chemosensory behaviors tested, including chemotaxis to benzaldehyde, methylpryazine, and sodium chloride, attractants sensed by AWC, AWA and ASE sensory neurons, respectively. Avoidance of 1-octanol, a repellant sensed by ADL and ASH sensory neurons is also perturbed (Bargmann, 2006; Bargmann et al., 1993; Bargmann and Horvitz, 1991; Chao et al., 2004) (Figure 2A-D). pros-1 RNAi animals also exhibit defective thermotaxis behavior (Figure 2E), suggestive of functional defects in the AFD thermosensory neuron. Postembryonic loss of pros-1, therefore, phenocopies ablation of AMsh glia (Bacaj et al., 2008), which also results in functional defects of associated sensory neurons.

Figure 2. Post-embryonic pros-1 RNAi results in chemosensory and thermosensory defects.

Figure 2

Open in a new tab

(A-C) Chemotaxis to indicated odors. Relevant neurons are indicated. pL4440, vector control. No AMsh glia, genetic ablation of AMsh glia by expression of diphtheria toxin (D) Avoidance of 1-octanol, sensed by ADL and ASH. (A-D) n=3, mean +/− SD, *p=0.016, **p < 0.01, ***p < 0.001, **** p <0.0001 (t-test). (E) pros-1 RNAi thermotaxis defects. Cultivation temperatures and no. animals scored are indicated.

PROS-1/Prospero functions post-embryonically and does not affect glial cell fate

In Drosophila, Prospero regulates the decision between glial and neuronal fates during external sense organ formation. Prospero mutants fail to produce glia, and Prospero overexpression results in overproduction of glia and of cells inappropriately expressing both glial and neuronal markers (Manning and Doe, 1999). C. elegans pros-1 is also expressed in glial cells and their precursors during embryogenesis (EPIC, Expression Patterns In Caenorhabditis); however, we specifically carried out our RNAi studies post-embryonically, after glial and neuronal fate had been established. Thus, the pros-1(RNAi) defects we observe may not result from a loss of glial cell fate. Indeed, we found that AMsh glia continue to express glial markers following post-embryonic pros-1 RNAi, including the AMsh-specific marker F16F9.3 and the pan-glial marker mir-228 (Figure 3). Furthermore pros-1(RNAi) AMsh glia do not show inappropriate expression of markers specific to their sister neurons. The cholinergic neuron URB, the sister cell of the AMsh glial cell, expresses unc-17/vAChT, a specific marker for cholinergic neurons, as well as rab-3, a pan-neuronal marker (Alfonso et al., 1993; Duerr et al., 2008). Neither is expressed in AMsh glia of pros-1(RNAi) animals, nor are AMsh reporters expressed in URB (Figure 3). Finally, AMsh glia of pros-1(RNAi) animals are grossly normal in morphology (see below). These results indicate that the defects we observed following pros-1 RNAi are not a result of general cell-fate abnormalities. Thus, pros-1 controls specific post-developmental aspects of glial function.

Figure 3. Post-embryonic pros-1 RNAi does not affect AMsh glia fate.

Figure 3

Open in a new tab

(A) Animals expressing F16F9.3::dsRed, an AMsh glia reporter, and unc-17::GFP, a cholinergic neuron reporter. (B) Animals expressing the pan-glial marker mir-228::GFP and the pan-neuronal marker rab-3::mCherry. White arrowheads, AMsh glia cell body. Scale bar, 10 μm.

Animals lacking PROS-1/Prospero exhibit defects in sensory neuron receptive-ending structure and microenvironment

To uncover the cellular basis for the behavioral deficits exhibited by pros-1(RNAi) animals, we assessed the morphology of sensory-neuron receptive endings and the structure of the AMsh glial membranes surrounding these receptive endings using fluorescence and electron microscopy. The AWC odor-sensing neuron and the AFD thermosensory neuron both have elaborate receptive endings that are ensheathed by AMsh glia (Figure 1A). Following pros-1 RNAi, both neurons extend dendrites normally to the tip of the nose, but their receptive endings are largely absent. AWC neurons lack the wing-like cilium found in wild-type adults (Figure 4A,F), while AFD neurons lack the microvilli-like fingers found in wild-type adults (Figure 4B,C,F).

Figure 4. pros-1 regulates the morphology of glia-ensheathed neuronal receptive endings (NRE) and the amphid channel structure.

Figure 4

Open in a new tab

Scale bar, 10 μm unless stated otherwise. (A) AWC NRE, visualized with odr-1::YFP. (B) AFD NRE, visualized with srtx-1::GFP. Scale bar, 5 μm. (C) Electron micrographs confirm loss of AFD microvilli after pros-1 RNAi (asterisk, AFD dendrite, arrowheads, AFD microvilli). Scale bar, 0.5 μm. (D) ASE NRE, visualized with gcy-5::GFP. (E) AMsh glia, visualized with F16F9.3::dsRed. (F) Quantification of morphologies assessed by fluorescence microscopy in A (lack of wing-like cilium), B (lack of microvilli-like fingers), D (normal channel cilium), E (normal sheath membrane). Error, SEM. AWC n>100, AFD n>70, ASE n>20, AMsh n>50, ****p<0.0001 (z-test). (G) Left, amphid channel schematic: green, AMsh glia, red, channel neuron cilia, blue, glial-secreted extracellular matrix, grey, AMso glia. Right, corresponding electron micrographs at 2 sections through the amphid. Scale bar, 0.5 μm (H). Top - pros-1 RNAi amphid neuron dye-filling defect. Bottom – quantification of dye-filling defect, error, SEM, n=150, ****p <0.0001 (z-test).

In contrast to neurons with glia-ensheathed receptive endings, the ASE channel neuron does not show morphological defects following pros-1 RNAi, with an apparently normal cilium at the tip of the dendrite (Figure 4D,F). However, unlike wild-type animals, ASE and other channel neuron cilia are not exposed to the environment following pros-1 RNAi treatment. While the overall morphology of the AMsh glial cell appears normal when assessed by fluorescence microscopy (Figure 4E,F), electron microscopy reveals striking defects in the glial channel housing sensory-neuron receptive endings (Figure 4G). In wild-type animals, neurons with simple cilia penetrate through the AMsh glial cell and enter an extracellular channel, generated by these glia. This channel is filled with an electron-dense extracellular matrix, secreted by AMsh glia, and thought to be important for sensory function (Perkins et al., 1986) (Bacaj et al., 2008). The channel formed by the AMsh glial cell is continuous with an opening through the cuticle formed by the anteriorly-positioned AMso glial cell (Figures 1A,4G). Animals treated with pros-1 RNAi show a striking defect in amphid channel architecture. At the posterior position where cilia normally enter the glial channel, individual cilia of pros-1(RNAi) animals are surrounded by the AMsh cell, but the channel is indistinct and glia-secreted matrix material is absent (Figure 4G, lower panels). More anteriorly, cilia that normally access the environment are blocked by an electron-dense deposit (Figure 4G, upper panels). To confirm these structural defects, we assessed the accessibility of channel-neuron cilia to the environment by carrying out dye-filling assays. In wild-type animals, a subset of exposed channel neurons can take up dye. Consistent with the channel defects we observed by electron microscopy, pros-1(RNAi) animals are unable to take up dye (Figure 4H).

In summary, pros-1 is required post-embryonically for maintenance of the glial channel through which channel neurons are normally exposed to the environment, for secretion of the extracellular matrix surrounding channel neurons, and for the proper morphology of the receptive endings of other glia-ensheathed neurons. Defects in these functions are consistent with the behavioral deficits we observed in pros-1(RNAi) animals, and suggest a role for pros-1 in regulating aspects of glial membrane structure and secretion.

PROS-1/Prospero functions in glia to control neuron structure and function

We identified pros-1 as a candidate glial regulator of neuronal function based on AMsh glia expression profiling. To confirm that pros-1 expression is indeed prominent in these glia, we examined animals carrying a recombineered fosmid expressing a C-terminally tagged PROS-1::GFP reporter under the control of pros-1 genomic regulatory elements. PROS-1::GFP is detected in the nuclei of AMsh glia (Figure 5A), confirming our expression profiling data. Importantly, PROS-1::GFP is not detected in amphid sensory neurons, identified by specific markers for AWC and AFD glia-ensheathed neurons and by DiI-staining for channel neurons (Figure 5B-D). PROS-1::GFP is also not detected in the AMso glia (Figure 5E). PROS-1::GFP is detected in many other sense organ-associated glia in the head, including the astrocyte-like CEP sheath glia, and the glia of the inner and outer labial sensilla (Figure S1), suggesting that pros-1 may play conserved roles in regulating sensory function across sense organs.

Figure 5. pros-1 is expressed and functions in AMsh glia to regulate neuronal properties.

Figure 5

Open in a new tab

(A-E) PROS-1::GFP translational reporter under the control of pros-1 genomic elements. Scale bar, 10 μm. (A-C,E) AMsh glia, AWC, AFD, AMso marked with F16F9.3::dsRed, odr-1::RFP, srtx-1::mCherry, itr-1::NLS-RFP, respectively. (D) Amphid channel neurons, identified by DiI staining. 2 sections from the same animal are shown. (F-H) AMsh glia-specific RNAi using rde-1(ne219) RNAi deficient strain expressing rde-1 in AMsh glia. (F) Amphid neuron dye-filling, n>200, error, SEM, ****p<0.0001 (z-test). (G, H) Benzaldehyde and sodium chloride chemotaxis, n=7, mean +/− SD, ****p<0.0001 (t-test). (I,J) pros-1(tm258) deletion mutant. (I) Dye-filling defects of indicated genotypes. WRM0617bG01, fosmid covering pros-1 locus. +/− glia, fosmid present/absent in AMsh glia. Error, SEM, larvae n=300, adults n>200, ****p<0.0001 (z-test). (J) Same as (I) but benzaldehyde chemotaxis assay. n=5, mean +/− SD, ****p<0.0001 (t-test). See also Figure S1.

The expression of pros-1 in AMsh glia but not in amphid sensory neurons is consistent with the hypothesis that pros-1 functions within glia, controlling glial channel morphology and extracellular matrix secretion cell autonomously, and sensory-neuron receptive-ending shapes and functions through a non-cell autonomous mechanism. To test this idea, we used two approaches. First, we generated a transgenic strain for glia-specific RNAi, by restoring expression of the Argonaute protein RDE-1 specifically in AMsh glia in an rde-1(ne219) mutant background. rde-1 is essential for RNAi induced by exogenous dsRNA, and is required for incorporation of processed siRNAs into silencing complexes, but is not required for the intercellular transport of dsRNA precursors (Parrish and Fire, 2001; Tabara et al., 1999). rde-1(ne219) mutant animals therefore take up dsRNA, but cannot carry out RNAi. Expression of rde-1 under an AMsh glia-specific promoter thus results in AMsh glia-specific gene silencing after dsRNA feeding. Importantly, AMsh glia-specific pros-1 RNAi results in the same defects observed using systemic pros-1 RNAi, including amphid channel neuron dye-filling, sodium chloride chemotaxis and benzaldehyde chemotaxis defects (Figure 5F-H), albeit with reduced penetrance. This can be explained, at least in part, by our finding that AMsh glia RNAi efficiency is reduced in this transgenic strain (S.W. and S.S., unpublished observations).

A genetic lesion in pros-1 resembles pros-1(RNAi)

To further confirm that pros-1 expression is required in AMsh glia for sensory neuron function, we characterized a pros-1 deletion mutant, tm258, provided by the NBP (National Biosciences Project) as a pros-1(tm258)/hT2 balanced strain. tm258 is a large deletion and is predicted to be a null allele. pros-1(tm258) mutant animals show greatly-reduced larval viability, characterized by slow larval development and larval lethality. On rare occasions, pros-1(tm258) mutants survive to adulthood, but usually fail to give progeny. pros-1(tm258) larvae show highly penetrant amphid neuron dye-filling defects (Figure 5I). Larval viability and dye-filling defects are not seen in pros-1(tm258)/hT2 heterozygotes, indicating that a single copy of pros-1 is sufficient for pros-1 function. Larval viability and amphid neuron dye-filling defects in pros-1(tm258) mutants can be rescued with a fosmid covering the pros-1 locus, indicating that these defects are a specific consequence of pros-1 loss (Figure 5I).

The larval growth defects precluded direct analysis of pros-1(tm258) adults. To overcome this, we took advantage of the mosaic expression pattern of extrachromosomal arrays, and analyzed pros-1(tm258) adults that contain a rescuing pros-1 fosmid as an unstable array, but which lack expression of the array in AMsh glia, based on a glia-specific GFP co-injection marker. Such mosaic transgenic pros-1(tm258) animals showed normal larval development, indicating that the larval-development defects are not a result of pros-1 loss in AMsh glia. Importantly, mosaic adults lacking pros-1 expression in AMsh glia showed the same amphid neuron dye-filling defects and benzaldehyde chemotaxis defects observed following pros-1 RNAi, while control animals with the pros-1 rescuing fosmid present in AMsh glia showed normal responses (Figure 5I,J). These results confirm the specificity of the RNAi phenotype observed, and support our conclusion that pros-1 expression is required in AMsh glia for sensory function.

Many PROS-1 target genes encode secreted proteins

pros-1 encodes a homeodomain transcription factor. To identify target genes that mediate pros-1 activity, we first looked at expression of known AMsh glia genes. Animals treated with pros-1 RNAi showed a strong reduction in expression of the glial genes vap-1, F11C7.2, and fig-1 as determined by analysis of transcriptional reporters (Figure 6A). vap-1 encodes a secreted protein with conserved CAP domains, found in cysteine-rich secretory proteins that play a role in regulating extracellular matrix remodeling, while F11C7.2 and fig-1 encode secreted proteins containing conserved TSP-1 (thrombospondin type 1) domains, found in extracellular matrix proteins including thrombospondins, known to be involved in glia-neuron interactions (Christopherson et al., 2005; Gibbs et al., 2008).

Figure 6. pros-1 target genes in AMsh glia.

Figure 6

Open in a new tab

(A) AMsh glia-specific gene expression with pros-1 RNAi. (B) AMsh glia secreted/transmembrane gene enrichment correlates with pros-1-dependency. Genes with indicated value or higher enrichment are scored for being PROS-1 targets. Red numbers, no. genes in each data point. For very high glial enrichment levels, gene number are low. (C) pros-1-dependent genes discussed in text. (D) fig-1(ns306) dye-filling defects (Error, SEM, n>80, ****p<0.0001 (z-test)) and 1-octanol avoidance defects (n=5, mean +/− SD, ****p<0.0001 (t-test)). (E) LIT-1C localization in AMsh glia. Scale bar, 10 μm. Right, error, SEM, n>300, ****p<0.0001 (z-test). (F) pros-1-dependent genes have dye-filling defects (p<0.0001 z-test; see text). (G) Model for PROS-1 function in controlling glial membrane structure and secretion. DAF-6 and LIT-1 control channel size, VAP-1 and FIG-1 are extracellular matrix components. See also Figure S2, S3, Table S2, S3.

To identify additional pros-1-dependent genes, we isolated AMsh glia using FACS and carried out gene expression profiling from pros-1 RNAi-treated and control animals. Cells were isolated from late stage larvae (L3 and L4) grown post-embryonically for 36-42 hours in liquid cultures of E. coli HT115 for RNAi by feeding. This time point was chosen because time-course experiments showed that pros-1(RNAi) defects of animals treated from the L1 stage first become apparent at the L3/4 stage (Figure S2), likely enriching for primary transcriptional targets. Effectiveness of RNAi by feeding when animals are grown in liquid culture was verified with dye-filing assays (Figure S2).

Homeodomain transcription factors, including the pros-1 homolog Prospero, can act as both positive and negative regulators of transcription (Choksi et al., 2006), and, indeed, we found glial genes with either decreased or increased expression levels following pros-1 RNAi. 439 genes show a significant reduction in expression (Table S2), while 634 genes show a significant increase (Table S3) (see Experimental Procedures for statistical analysis). Comparison of AMsh-enriched genes encoding secreted or transmembrane proteins with pros-1-dependent genes revealed a positive correlation between the degree of enrichment in AMsh glia and likelihood of being a pros-1 target (Figure 6B). Strikingly, of the 155 highly-enriched (>10×) AMsh glia genes that encode proteins with signal peptides or transmembrane domains (Figure 1), 49 genes (32%) showed a significant reduction in expression following pros-1 RNAi. Thus, pros-1 controls the expression of a large portion of the AMsh glia-specific secretome (along with other genes, Figure S3), consistent with the defects in extracellular matrix secretion and membrane abnormalities we observed in pros-1 mutants.

To validate the candidate pros-1 target genes we identified, we assessed the expression levels of transcriptional reporters for two genes, R11D1.3, encoding a secreted protein of unknown function, and K02E11.4, encoding a secreted protein containing a conserved UDP-glucoronosyltransferase domain, found in enzymes that act in the olfactory epithelium to modulate chemosensory cues (Leclerc et al., 2002). These were identified in our RNAseq studies as having 18-fold and 4.3-fold decreased expression in pros-1 RNAi animals (Figure 6C). Importantly, both reporters show clear reductions in AMsh glia GFP expression in animals treated with pros-1 RNAi (Figure 6A). The reduction in expression is particularly strong for R11D1.3::GFP, which correlates with the increased pros-1-dependence of this gene seen in our expression profiling data. These results confirm that we identified pros-1-dependent genes in our expression profiling experiments.

Of the genes we identified as pros-1-dependent based on transcriptional reporters (vap-1, fig-1, F11C7.2; Figure 6A), only F11C7.2 also showed pros-1-dependence in our expression profiling (Figure 6C). One explanation for this difference may be in the timing of the experiments. Reduction in transcriptional reporter expression was observed in adults after 3 days of RNAi treatment, while we isolated AMsh glia from late stage larvae at 36-42 hours of RNAi treatment. Thus, fig-1 and vap-1 may represent secondary targets for PROS-1.

Taken together, our expression profiling studies, coupled with phenotypic analysis of pros-1 mutants, reveal that PROS-1 is a major regulator of the AMsh glia secretome.

PROS-1 target genes encoding membrane and secreted proteins account for neuronal dye-filling defects in pros-1 mutants

To identify functionally important genes whose expression changes contribute to the defects observed in pros-1 mutants, we searched for known regulators of AMsh glial function amongst our list of candidate pros-1-dependent genes, and carried out additional RNAi screens.

As described above, expression of the fig-1 gene is strongly dependent on pros-1. fig-1 encodes a secreted thrombospondin-domain-containing extracellular matrix protein necessary for amphid neuron dye-filling and avoidance of 1-octanol (Figure 6D) (Bacaj et al., 2008). Thus, pros-1 acts in glia to control neuronal function and to regulate the amphid channel extracellular microenvironment partly by promoting expression of FIG-1 protein.

We also found that expression of the Patched-related gene daf-6, a negative regulator of amphid channel expansion, increases significantly following pros-1 RNAi (4.2-fold increase), a result we confirmed by analysis of a daf-6::GFP expression reporter (Fig. 6A). DAF-6 is a membrane protein that restricts amphid channel size, and daf-6 mutants have abnormally expanded channels. Thus, overexpression of DAF-6 may contribute to the loss of amphid channel membrane following pros-1 RNAi. Supporting these observations, DAF-6 functions by antagonizing the activity of lit-1, a Nemo-like kinase that localizes to the amphid channel membrane and promotes channel expansion (Oikonomou et al., 2011). While lit-1 expression level was not affected in pros-1(RNAi) animals, LIT-1 protein, which normally lines the amphid channel, showed only diffuse localization (Figure 6E). Together, these results suggest that pros-1 acts, at least in part, to regulate amphid channel size by inhibiting expression of daf-6 to support lit-1 function.

Finally, to identify additional genes mediating pros-1 function, we screened candidate pros-1 target genes by RNAi for defects in amphid neuron dye-filling. To compensate for possible redundancy amongst pros-1 target genes, we carried out these experiments in pros-1(tm258)/hT2 heterozygous animals, which show normal dye-filling (Figure 5I), but which we predict to have reduced levels of pros-1 expression, and may in turn have reduced expression of redundantly-acting target genes and therefore provide a sensitized background for detecting functional defects. Indeed, pros-1(tm258)/hT2 heterozygotes treated with RNAi against the gene rdy-2, previously identified as required for dye filling (Liegeois et al., 2007), show increased defects compared to rdy-2(RNAi) animals alone (56% defective vs. 16%). This screen identified 5 candidate pros-1 target genes that exhibit low, but statistically significant, dye-filling defects, and thus may represent functional targets of pros-1 in AMsh glia (Figure 6F). Notably, all are predicted to encode proteins with transmembrane domains.

DISCUSSION

Differences between embryonic and post-embryonic glial transcriptomes

The amphid sense organ of C. elegans is an attractive model system for studying interactions between glia and neurons because of its relatively simple anatomy and because the functional properties of amphid sensory neurons have been extensively studied (Shaham, 2006, 2015). To characterize the molecular mechanisms through which amphid sheath (AMsh) glia regulate neuronal function, and not development, we took advantage of recently developed larval cell isolation techniques and adapted these to identify AMsh glia post-embryonic transcripts by RNA sequencing (Spencer et al., 2014; Zhang and Kuhn, 2013). This approach led to the identification of 598 candidate AMsh glial genes. We previously identified 298 transcripts enriched in embryonic AMsh glia (Bacaj et al., 2008). Of the 598 post-embryonic transcripts identified in this study, only 72 were also detected in our previous study. While the large number of newly identified transcripts can be partly explained by differences in transcript analysis and statistical methods used (RNAseq vs. microarray), this difference highlights the advantage of using larval cell isolation for identification of genes with post-embryonic functions, and reveals distinct transcriptomes for these stages.

PROS-1 is a regulator of the glial secretome

We demonstrate here that RNAi against pros-1 results in similar defects to AMsh glia ablation, including defective morphology of glia-ensheathed sensory-neuron receptive endings and defective functions of channel neurons. pros-1 was previously noted to affect larval survival and amphid neuron dye-filling (Kolotuev et al., 2013), and its role in excretory canal formation was characterized in detail, but the nature of the amphid neuron dye-filling defect was not previously explored. Importantly, in this study we show that pros-1 is expressed in AMsh glia, as well as glia of other sense organs, but not in amphid neurons, and is required for maintenance of the matrix-filled amphid channel, through which channel neurons are exposed to the external environment, and for sensory behaviors controlled by these neurons. Additionally, we have shown that pros-1 is required for proper morphology of the glia-ensheathed receptive endings of odor- and thermosensory neurons, and that pros-1 loss of function results in defects in the corresponding chemotaxis and thermotaxis behaviors. pros-1 thus acts cell autonomously to maintain amphid channel structure, and non-cell autonomously to maintain the receptive-endings of ensheathed sensory neurons, perhaps through glia-to-neuron signaling. Both glia-neuron communication and amphid channel shape and content are predicted to require secreted and/or transmembrane proteins. Consistent with this idea, AMsh glia show a striking over-representation of genes encoding secreted or transmembrane proteins, of which about a third are pros-1-dependent. PROS-1 is, therefore, a key regulator of the AMsh glia secretome. Of note, our expression profiling experiments also reveal that pros-1 regulates expression of non-secreted genes, suggesting pros-1 has additional functions.

Some pros-1 targets we identified, such as fig-1 and daf-6, may partly explain the defects we observe in pros-1 mutants (Figure 6G), as these components have known independent roles in AMsh glia. The large number of pros-1 target genes we identified suggests that other targets may act redundantly, making the identification of functionally important target genes challenging. Remarkably, the robustness of amphid neuron dye-filling in wild-type animals allowed us to identify weak but statistically significant defects in five candidate pros-1 target genes. Of these, F41H10.5 encodes a transmembrane protein with a conserved TLC (TRAM, LAG1, CLN8) domain, which have general roles in lipid metabolism, and in mammals regulate calcium channel activity (Papanayotou et al., 2013; Winter and Ponting, 2002), and F31D5.2 encodes a transmembrane protein with a major facilitator superfamily domain, a conserved domain found in solute transporters (Yan, 2015). While the remaining 3 genes encode proteins of unknown function, it is notable that all 3 encode predicted transmembrane proteins. While further work is required to assess the roles of these genes in pros-1-mediated functions, these observations are consistent with roles in glia-neuron interactions.

PROS-1/Prospero has cell-fate-independent functions

pros-1 encodes a Prospero-related homeodomain transcription factor. Prospero has well known roles in determining cell fate decisions during nervous system development in Drosophila (Doe et al., 1991). When neuroblasts divide, Prospero is distributed asymmetrically between daughter cells and promotes differentiation of ganglion mother cells into neurons and glia by inhibiting expression of cell cycle genes and promoting expression of genes for terminal differentiation (Choksi et al., 2006; Freeman and Doe, 2001). During the final round of cell divisions generating external sense organs in Drosophila, a precursor cell divides to produce a glia-like sheath cell and a neuron. Prospero mutants produce neither, while Prospero overexpression results in overproduction of glia and causes a partial neuron-to-glia fate conversion (Manning and Doe, 1999). Prospero expression is downregulated in neurons after the final cell division, but is retained in glia. Of note, in Prospero mutants that do generate neurons, these exhibit defective axon and dendrite morphology. While this defect might reflect a role for the transient expression of Prospero in neurons, our results raise the possibility that the defects may originate from the glia-like sheath cell.

In addition to sheath cells in the sense organs of the Drosophila wing, Prospero is also expressed in the glia-like support cells of the antenna-maxillary complex (AMC), the larval chemosensory organ (Balakireva et al., 2000). Expression profiling to identify Prospero-dependent genes in whole AMC extracts was performed (Guenin et al., 2010). However, Prospero is expressed in neurons and glia-like cells of the AMC, and reported targets are mainly involved in neuronal axon growth and synaptic function, and therefore likely to represent Prospero targets in neurons and not support cells.

Vertebrate homologs of Prospero are expressed throughout the nervous system. In the mammalian brain, Prox1 is expressed in the hippocampus, is involved in neurogenesis, and likely has a role in establishing cell fate (Galeeva et al., 2007) (Elsir et al., 2012). Prox1 is also expressed in differentiated glia-like support cells in sense organs, including support cells of the mammalian cochlea and the zebrafish neuromast (Bermingham-McDonogh et al., 2006; Pistocchi et al., 2009). Prox1 is expressed in progenitor cells during cochlea development, but post-embryonically its expression is restricted to glia-like cells including Deiters’ cells and pillar cells (Bermingham-McDonogh et al., 2006). Prox1 loss in neuromast support cells results in defective uptake of dye by associated mechanosensory hair cells, suggestive of functional defects (Pistocchi et al., 2009). These observations show that expression of Prospero homologs is a conserved feature of sense organ-associated glia. Importantly, our results suggest that Prospero is likely to be a key factor maintaining the functions of these organs well past embryonic development.

EXPERIMENTAL PROCEDURES

Strains and Plasmids

C. elegans strains were maintained using standard methods (Stiernagle, 2006). Wild-type strain is Bristol N2. See Supplemental Experimental Procedures for details of alleles and transgenic strains.

RNA Interference

Standard RNAi was carried out as described (Kamath and Ahringer, 2003). Overnight cultures of E. coli HT115 carrying pL4440 expression vectors targeting specific genes (Ahringer library, Vidal Library) were seeded on to NGM plates supplemented with carbenicillin (25 μg/ml) and IPTG (1 mM). After 1-2 days induction at room temperature, embryos obtained from bleaching gravid adults were added to the RNAi plates. Unless stated otherwise, animals were scored 3 days later, as first-day adults.

Behavioral Assays

All behavior assays were performed on first-day adults. Chemotaxis assays, avoidance assays and thermotaxis assays were carried out using standard procedures (Bargmann et al., 1993; Chao et al., 2004; Ryu and Samuel, 2002; Ward, 1973). See Supplemental Experimental Procedures for details. Statistical analysis was performed with unpaired t-test (GraphPad).

Dye-filling Assays

DiI (1,1’-dioctadecyl-3,3,3’3’-tetramethylindocarbocyanine perchlorate) (Sigma) stock solution was made at 5 mg/ml in N,N-dimethylformamide. Animals were with DiI (stock solution diluted 1:1000 in M9 buffer) for 30-60 minutes, followed by 3 × M9 washes to remove excess dye. Dye-filling was scored as % animals that show amphid neuron dye-uptake, with statistical anaylsis performed with two-proportion z-test (in-silico).

Fluorescence and Electron Microscopy

Fluorescence and electron micrographs were obtained using standard procedures. See Supplemental Experimental Procedures for details. Quantification of morphological defects assessed by fluorescence microscopy was carried out by scoring individual animals for wild-type vs defective morphology, with statistical analysis performed with two-proportion z-test (in-silico).

Cell Isolation and FACS Analysis

Briefly, synchronized larvae were grown from L1 in liquid culture of E. coli HT115 for RNAi treatment. After growth for 36-42 hours, cells were isolated from late-stage larvae (L3 and L4) by protease treatment and mechanical disruption, as described previously (Zhang and Kuhn, 2013), with modifications (Menachem Katz, personal communications). AMsh glia expressing a dsRed transgene (nsIs143) were sorted using a BD FACS Aria sorter (Rockefeller University Flow Cytometry Resource Center). See Supplemental Experimental Procedures for details.

RNA Isolation and Sequencing

RNA was extracted from Trizol LS (Ambion)-treated cells by phase separation, following the manufacturers guidelines. RNA was purified using PicoPure RNA isolation kit (Arcturus). 1-5 ng purified total RNA was obtained per sample. All subsequent steps were performed by the Rockefeller University Genomics Resource Center. RNA quality was verified to make sure sample degradation had not occurred (Agilent Bioanalyzer). mRNA amplification and cDNA preparation were performed using the SMARTer mRNA amplification kit (Clontech). Labelled samples were sequenced using an Illumina HiSeq 2000 sequencer.

RNA-Seq Quality Assessment and Differential Gene Expression (DE) Analysis

Sequence quality was assessed and gene alignment was carried out as described in Supplemental Experimental Procedures. 3 independent replicates were obtained from control animals, and 4 independent replicates were obtained for pros-1 RNAi animals. Two different statistical methods were used for differential gene expression analysis, DESeq and voom. For DEseq analysis, DESeq2 was applied to normalize count matrix and to perform differential gene expression on the counts using negative binomial distribution (Love et al., 2014); for voom analysis, edgeR (Robinson et al., 2010) was applied to normalize count matrix, and voom (Law et al., 2014) was applied for gene differentiation analysis. Significant genes from both analyses were combined.

To identify transcripts enriched in AMsh glia compared to other cells (control AMsh vs control nonAMsh) we used a fold-change of 3.5 and an adjusted p-value threshold of <0.05. This fold-change cutoff was determined as follows. We sought genes that are > 4-fold enriched in AMsh glia, as these changes can be easily assayed in vivo using GFP reporters. For genes showing pan-glial expression, a true enrichment of R, should correspond to an RNAseq-detected enrichment of 959/(46+(909/R)), assuming each animal has 959 total cells and 50 glial cells. For R=4, RNAseq enrichment is 3.5×. To identify genes whose expression is regulated by pros-1 (pros-1 RNAi AMsh vs control AMsh), we used a less stringent threshold to take into account possible variation in RNAi efficiency between replicates (fold-change > 2, adjusted p-value < 0.1). To identify putative secreted or transmembrane genes, we used Phobius, a combined membrane topology and signal peptide predictor (Stockholm Bioinformatics Center, phobius.sbc.edu).

Supplementary Material

1
2
3
4
5

ACKNOWLEDGEMENTS

We thank Menachem Katz for help with cell isolation and gene expression analysis, Anupriya Singhal for help with data analysis, and Shaham lab members for helpful discussions. We thank The Rockefeller University Bio-Imaging, Flow Cytometry and Genomics Resource Centers for outstanding technical support. S.W. was a William N. and Bernice E. Bumpus Foundation postdoctoral fellow. A.S. was an American Cancer Society postdoctoral fellow (PF1308301DDC), an NIH-T32 postdoctoral fellow (5T32CA967334) and a Murray Foundation fellow. Y. Liang is supported by the NIH Center for Clinical and Translational Science Award (CTSA) and National Center for Advancing Sciences (NCATS). This work was supported in part by NIH grant R01NS064273 to S.S.

Footnotes

Publisher's Disclaimer: 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 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.

ACCESSION NUMBERS

RNA sequence data generated in this study has been submitted to the European Nucleotide Archive (ENA) (study accession number PRJEB12954).

AUTHOR CONTRIBUTIONS

S.W. and S.S. conceived of the studies, designed experiments, interpreted data, and wrote the paper. A.S. performed the studies on AFD neuron (thermotaxis and morphology). Y. Liang performed most of the RNAseq analysis. Y. Lu performed the EM studies. S.W. performed all the other studies.

REFERENCES

  1. Alfonso A, Grundahl K, Duerr JS, Han HP, Rand JB. The Caenorhabditis elegans unc-17 gene: a putative vesicular acetylcholine transporter. Science. 1993;261:617–619. doi: 10.1126/science.8342028. [DOI] [PubMed] [Google Scholar]
  2. Bacaj T, Tevlin M, Lu Y, Shaham S. Glia are essential for sensory organ function in C. elegans. Science. 2008;322:744–747. doi: 10.1126/science.1163074. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Balakireva M, Gendre N, Stocker RF, Ferveur JF. The genetic variant Voila causes gustatory defects during Drosophila development. J Neurosci. 2000;20:3425–3433. doi: 10.1523/JNEUROSCI.20-09-03425.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bargmann CI. Chemosensation in C. elegans. WormBook; 2006. pp. 1–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bargmann CI, Hartwieg E, Horvitz HR. Odorant-selective genes and neurons mediate olfaction in C. elegans. Cell. 1993;74:515–527. doi: 10.1016/0092-8674(93)80053-h. [DOI] [PubMed] [Google Scholar]
  6. Bargmann CI, Horvitz HR. Chemosensory neurons with overlapping functions direct chemotaxis to multiple chemicals in C. elegans. Neuron. 1991;7:729–742. doi: 10.1016/0896-6273(91)90276-6. [DOI] [PubMed] [Google Scholar]
  7. Bartel DL, Sullivan SL, Lavoie EG, Sevigny J, Finger TE. Nucleoside triphosphate diphosphohydrolase-2 is the ecto-ATPase of type I cells in taste buds. J Comp Neurol. 2006;497:1–12. doi: 10.1002/cne.20954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bermingham-McDonogh O, Oesterle EC, Stone JS, Hume CR, Huynh HM, Hayashi T. Expression of Prox1 during mouse cochlear development. J Comp Neurol. 2006;496:172–186. doi: 10.1002/cne.20944. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Boettger T, Hubner CA, Maier H, Rust MB, Beck FX, Jentsch TJ. Deafness and renal tubular acidosis in mice lacking the K-Cl co-transporter Kcc4. Nature. 2002;416:874–878. doi: 10.1038/416874a. [DOI] [PubMed] [Google Scholar]
  10. Breunig E, Czesnik D, Piscitelli F, Di Marzo V, Manzini I, Schild D. Endocannabinoid modulation in the olfactory epithelium. Results and problems in cell differentiation. 2010;52:139–145. doi: 10.1007/978-3-642-14426-4_11. [DOI] [PubMed] [Google Scholar]
  11. Chao MY, Komatsu H, Fukuto HS, Dionne HM, Hart AC. Feeding status and serotonin rapidly and reversibly modulate a Caenorhabditis elegans chemosensory circuit. Proc Natl Acad Sci U S A. 2004;101:15512–15517. doi: 10.1073/pnas.0403369101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Choksi SP, Southall TD, Bossing T, Edoff K, de Wit E, Fischer BE, van Steensel B, Micklem G, Brand AH. Prospero acts as a binary switch between self-renewal and differentiation in Drosophila neural stem cells. Dev Cell. 2006;11:775–789. doi: 10.1016/j.devcel.2006.09.015. [DOI] [PubMed] [Google Scholar]
  13. Christopherson KS, Ullian EM, Stokes CC, Mullowney CE, Hell JW, Agah A, Lawler J, Mosher DF, Bornstein P, Barres BA. Thrombospondins are astrocyte-secreted proteins that promote CNS synaptogenesis. Cell. 2005;120:421–433. doi: 10.1016/j.cell.2004.12.020. [DOI] [PubMed] [Google Scholar]
  14. Cuschieri A, Bannister LH. The development of the olfactory mucosa in the mouse: electron microscopy. Journal of anatomy. 1975;119:471–498. [PMC free article] [PubMed] [Google Scholar]
  15. Doe CQ, Chu-LaGraff Q, Wright DM, Scott MP. The prospero gene specifies cell fates in the Drosophila central nervous system. Cell. 1991;65:451–464. doi: 10.1016/0092-8674(91)90463-9. [DOI] [PubMed] [Google Scholar]
  16. Dooley R, Mashukova A, Toetter B, Hatt H, Neuhaus EM. Purinergic receptor antagonists inhibit odorant-mediated CREB phosphorylation in sustentacular cells of mouse olfactory epithelium. BMC neuroscience. 2011;12:86. doi: 10.1186/1471-2202-12-86. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Doroquez DB, Berciu C, Anderson JR, Sengupta P, Nicastro D. A high-resolution morphological and ultrastructural map of anterior sensory cilia and glia in Caenorhabditis elegans. eLife. 2014;3:e01948. doi: 10.7554/eLife.01948. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Duerr JS, Han HP, Fields SD, Rand JB. Identification of major classes of cholinergic neurons in the nematode Caenorhabditis elegans. J Comp Neurol. 2008;506:398–408. doi: 10.1002/cne.21551. [DOI] [PubMed] [Google Scholar]
  19. Dvoryanchikov G, Sinclair MS, Perea-Martinez I, Wang T, Chaudhari N. Inward rectifier channel, ROMK, is localized to the apical tips of glial-like cells in mouse taste buds. J Comp Neurol. 2009;517:1–14. doi: 10.1002/cne.22152. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Elsir T, Smits A, Lindstrom MS, Nister M. Transcription factor PROX1: its role in development and cancer. Cancer metastasis reviews. 2012;31:793–805. doi: 10.1007/s10555-012-9390-8. [DOI] [PubMed] [Google Scholar]
  21. Engstrom H. The ultrastructure of the sensory cells of the cochlea. The Journal of laryngology and otology. 1967;81:687–715. [PubMed] [Google Scholar]
  22. Freeman MR, Doe CQ. Asymmetric Prospero localization is required to generate mixed neuronal/glial lineages in the Drosophila CNS. Development. 2001;128:4103–4112. doi: 10.1242/dev.128.20.4103. [DOI] [PubMed] [Google Scholar]
  23. Galeeva A, Treuter E, Tomarev S, Pelto-Huikko M. A prospero-related homeobox gene Prox-1 is expressed during postnatal brain development as well as in the adult rodent brain. Neuroscience. 2007;146:604–616. doi: 10.1016/j.neuroscience.2007.02.002. [DOI] [PubMed] [Google Scholar]
  24. Gibbs GM, Roelants K, O’Bryan MK. The CAP superfamily: cysteine-rich secretory proteins, antigen 5, and pathogenesis-related 1 proteins--roles in reproduction, cancer, and immune defense. Endocrine reviews. 2008;29:865–897. doi: 10.1210/er.2008-0032. [DOI] [PubMed] [Google Scholar]
  25. Guenin L, Raharijaona M, Houlgatte R, Baba-Aissa F. Expression profiling of prospero in the Drosophila larval chemosensory organ: Between growth and outgrowth. BMC genomics. 2010;11:47. doi: 10.1186/1471-2164-11-47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hansel DE, Eipper BA, Ronnett GV. Neuropeptide Y functions as a neuroproliferative factor. Nature. 2001;410:940–944. doi: 10.1038/35073601. [DOI] [PubMed] [Google Scholar]
  27. Hegg CC, Irwin M, Lucero MT. Calcium store-mediated signaling in sustentacular cells of the mouse olfactory epithelium. Glia. 2009;57:634–644. doi: 10.1002/glia.20792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kamath RS, Ahringer J. Genome-wide RNAi screening in Caenorhabditis elegans. Methods. 2003;30:313–321. doi: 10.1016/s1046-2023(03)00050-1. [DOI] [PubMed] [Google Scholar]
  29. Kimura KD, Miyawaki A, Matsumoto K, Mori I. The C. elegans thermosensory neuron AFD responds to warming. Curr Biol. 2004;14:1291–1295. doi: 10.1016/j.cub.2004.06.060. [DOI] [PubMed] [Google Scholar]
  30. Kolotuev I, Hyenne V, Schwab Y, Rodriguez D, Labouesse M. A pathway for unicellular tube extension depending on the lymphatic vessel determinant Prox1 and on osmoregulation. Nature cell biology. 2013;15:157–168. doi: 10.1038/ncb2662. [DOI] [PubMed] [Google Scholar]
  31. Law CW, Chen Y, Shi W, Smyth GK. voom: Precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol. 2014;15:R29. doi: 10.1186/gb-2014-15-2-r29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Lawton DM, Furness DN, Lindemann B, Hackney CM. Localization of the glutamate-aspartate transporter, GLAST, in rat taste buds. Eur J Neurosci. 2000;12:3163–3171. doi: 10.1046/j.1460-9568.2000.00207.x. [DOI] [PubMed] [Google Scholar]
  33. Leclerc S, Heydel JM, Amosse V, Gradinaru D, Cattarelli M, Artur Y, Goudonnet H, Magdalou J, Netter P, Pelczar H, et al. Glucuronidation of odorant molecules in the rat olfactory system: activity, expression and age-linked modifications of UDP-glucuronosyltransferase isoforms, UGT1A6 and UGT2A1, and relation to mitral cell activity. Brain research Molecular brain research. 2002;107:201–213. doi: 10.1016/s0169-328x(02)00455-2. [DOI] [PubMed] [Google Scholar]
  34. Liegeois S, Benedetto A, Michaux G, Belliard G, Labouesse M. Genes required for osmoregulation and apical secretion in Caenorhabditis elegans. Genetics. 2007;175:709–724. doi: 10.1534/genetics.106.066035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:550. doi: 10.1186/s13059-014-0550-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Manning L, Doe CQ. Prospero distinguishes sibling cell fate without asymmetric localization in the Drosophila adult external sense organ lineage. Development. 1999;126:2063–2071. doi: 10.1242/dev.126.10.2063. [DOI] [PubMed] [Google Scholar]
  37. Matsunobu T, Chung JW, Schacht J. Acetylcholine-evoked calcium increases in Deiters’ cells of the guinea pig cochlea suggest alpha9-like receptors. Journal of neuroscience research. 2001;63:252–256. doi: 10.1002/1097-4547(20010201)63:3<252::AID-JNR1018>3.0.CO;2-0. [DOI] [PubMed] [Google Scholar]
  38. Murray RG. Cellular relations in mouse circumvallate taste buds. Microsc Res Tech. 1993;26:209–224. doi: 10.1002/jemt.1070260304. [DOI] [PubMed] [Google Scholar]
  39. Oikonomou G, Perens EA, Lu Y, Watanabe S, Jorgensen EM, Shaham S. Opposing activities of LIT-1/NLK and DAF-6/patched-related direct sensory compartment morphogenesis in C. elegans. PLoS Biol. 2011;9:e1001121. doi: 10.1371/journal.pbio.1001121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Oikonomou G, Shaham S. The glia of Caenorhabditis elegans. Glia. 2011;59:1253–1263. doi: 10.1002/glia.21084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Papanayotou C, De Almeida I, Liao P, Oliveira NM, Lu SQ, Kougioumtzidou E, Zhu L, Shaw A, Sheng G, Streit A, et al. Calfacilitin is a calcium channel modulator essential for initiation of neural plate development. Nature communications. 2013;4:1837. doi: 10.1038/ncomms2864. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Parrish S, Fire A. Distinct roles for RDE-1 and RDE-4 during RNA interference in Caenorhabditis elegans. Rna. 2001;7:1397–1402. [PMC free article] [PubMed] [Google Scholar]
  43. Peco E, Davla S, Camp D, Stacey S, Landgraf M, van Meyel D. Drosophila astrocytes cover specific territories of CNS neuropil and are instructed to differentiate by Prospero, a key effector of Notch. Development. 2016 doi: 10.1242/dev.133165. In press. DOI 10.1242. [DOI] [PubMed] [Google Scholar]
  44. Perkins LA, Hedgecock EM, Thomson JN, Culotti JG. Mutant sensory cilia in the nematode Caenorhabditis elegans. Dev Biol. 1986;117:456–487. doi: 10.1016/0012-1606(86)90314-3. [DOI] [PubMed] [Google Scholar]
  45. Pistocchi A, Feijoo CG, Cabrera P, Villablanca EJ, Allende ML, Cotelli F. The zebrafish prospero homolog prox1 is required for mechanosensory hair cell differentiation and functionality in the lateral line. BMC developmental biology. 2009;9:58. doi: 10.1186/1471-213X-9-58. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26:139–140. doi: 10.1093/bioinformatics/btp616. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Ryu WS, Samuel AD. Thermotaxis in Caenorhabditis elegans analyzed by measuring responses to defined Thermal stimuli. J Neurosci. 2002;22:5727–5733. doi: 10.1523/JNEUROSCI.22-13-05727.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Shaham S. Glia-neuron interactions in the nervous system of Caenorhabditis elegans. Curr Opin Neurobiol. 2006;16:522–528. doi: 10.1016/j.conb.2006.08.001. [DOI] [PubMed] [Google Scholar]
  49. Shaham S. Chemosensory organs as models of neuronal synapses. Nat Rev Neurosci. 2010;11:212–217. doi: 10.1038/nrn2740. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Shaham S. Glial development and function in the nervous system of Caenorhabditis elegans. Cold Spring Harbor perspectives in biology. 2015;7:a020578. doi: 10.1101/cshperspect.a020578. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Spencer WC, McWhirter R, Miller T, Strasbourger P, Thompson O, Hillier LW, Waterston RH, Miller DM., 3rd Isolation of specific neurons from C. elegans larvae for gene expression profiling. PloS one. 2014;9:e112102. doi: 10.1371/journal.pone.0112102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Stiernagle T. Maintenance of C. elegans. WormBook; 2006. pp. 1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Strauss O. The retinal pigment epithelium in visual function. Physiol Rev. 2005;85:845–881. doi: 10.1152/physrev.00021.2004. [DOI] [PubMed] [Google Scholar]
  54. Suh J, Hutter H. A survey of putative secreted and transmembrane proteins encoded in the C. elegans genome. BMC genomics. 2012;13:333. doi: 10.1186/1471-2164-13-333. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Tabara H, Sarkissian M, Kelly WG, Fleenor J, Grishok A, Timmons L, Fire A, Mello CC. The rde-1 gene, RNA interference, and transposon silencing in C. elegans. Cell. 1999;99:123–132. doi: 10.1016/s0092-8674(00)81644-x. [DOI] [PubMed] [Google Scholar]
  56. Trotier D. Electrophysiological properties of frog olfactory supporting cells. Chem Senses. 1998;23:363–369. doi: 10.1093/chemse/23.3.363. [DOI] [PubMed] [Google Scholar]
  57. Ward S. Chemotaxis by the nematode Caenorhabditis elegans: identification of attractants and analysis of the response by use of mutants. Proc Natl Acad Sci U S A. 1973;70:817–821. doi: 10.1073/pnas.70.3.817. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Winter E, Ponting CP. TRAM, LAG1 and CLN8: members of a novel family of lipid-sensing domains? Trends Biochem Sci. 2002;27:381–383. doi: 10.1016/s0968-0004(02)02154-0. [DOI] [PubMed] [Google Scholar]
  59. Yan N. Structural Biology of the Major Facilitator Superfamily Transporters. Annual review of biophysics. 2015;44:257–283. doi: 10.1146/annurev-biophys-060414-033901. [DOI] [PubMed] [Google Scholar]
  60. Zhang S, Kuhn JR. Cell isolation and culture. WormBook; 2013. pp. 1–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Zhang Y, Chen K, Sloan SA, Bennett ML, Scholze AR, O’Keeffe S, Phatnani HP, Guarnieri P, Caneda C, Ruderisch N, et al. An RNA-sequencing transcriptome and splicing database of glia, neurons, and vascular cells of the cerebral cortex. J Neurosci. 2014;34:11929–11947. doi: 10.1523/JNEUROSCI.1860-14.2014. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS771566-supplement-1.docx (54.8KB, docx)
2
NIHMS771566-supplement-2.xlsx (87.3KB, xlsx)
3
NIHMS771566-supplement-3.xlsx (81.5KB, xlsx)
4
NIHMS771566-supplement-4.xlsx (97.6KB, xlsx)
5
NIHMS771566-supplement-5.pdf (3.4MB, pdf)

RESOURCES