HTR6 and SSTR3 ciliary targeting relies on both IC3 loops and C-terminal tails

Abstract

G protein-coupled receptors (GPCRs) are the most common pharmacological target in human clinical practice. To perform their functions, many GPCRs must accumulate inside primary cilia, microtubule-based plasma membrane protrusions working as cellular antennae. Nevertheless, the molecular mechanisms underlying GPCR ciliary targeting remain poorly understood. Serotonin receptor 6 (HTR6) and somatostatin receptor 3 (SSTR3) are two brain-enriched ciliary GPCRs involved in cognition and pathologies such as Alzheimer's disease and cancer. Although the third intracellular loops (IC3) of HTR6 and SSTR3 suffice to target non-ciliary GPCRs to cilia, these IC3s are dispensable for ciliary targeting of HTR6 and SSTR3 themselves, suggesting these GPCRs contain additional ciliary targeting sequences (CTSs). Herein, we discover and characterize novel CTSs in HTR6 and SSTR3 C-terminal tails (CT). These CT-CTSs (CTS2) act redundantly with IC3-CTSs (CTS1), each being sufficient for ciliary targeting. In HTR6, RKQ and LPG motifs are critical for CTS1 and CTS2 function, respectively, whereas in SSTR3 these roles are mostly fulfilled by AP[AS]CQ motifs in IC3 and juxtamembrane residues in CT. Furthermore, we shed light on how these CTSs promote ciliary targeting by modulating binding to ciliary trafficking adapters TULP3 and RABL2.

Figure 1.. HTR6 ciliary targeting requires the ciliary trafficking adapter TULP3.

(A) HTR6-IMCD3 cells were transiently transfected with siRNAs targeting mouse TULP3 (siTULP3 #1 or siTULP3 #2) or firefly luciferase (siLUC) as negative control, serum-starved to promote ciliogenesis and immunostained with anti-ARL13B (green) and anti-HTR6 (red) antibodies. DNA was stained with Hoechst (blue). Scale bar, 5 μm. (A, B) HTR6 ciliary intensity was quantified from (A). Data are mean ± SEM of n = 23,32,29 cells for siLUC, siTULP3 #1, and siTULP3 #2, respectively. (C) Mouse Tulp3 mRNA levels were analyzed by qRT-PCR and expressed relative to Gapdh mRNA. Data are mean ± SEM of n = 3 independent experiments. Significance in (B, C) is shown as P < 0.01(**) in unpaired two-tailed t tests. (A, D) HTR6-IMCD3 cells transfected as in (A) were analyzed by Western blot with antibodies against HTR6, TULP3 and Actin, as indicated. Despite the siRNA-induced reduction in TULP3 protein, HTR6 protein levels remained unaltered. Molecular weight markers shown on the right. Source data are available for this figure.

Figure 2.. The ATAGQ motif in HTR6-IC3 is dispensable for ciliary targeting of wild-type HTR6.

(A) Schematic representation of HTR7 (green), HTR6 (purple), Chimera N, and the mutant versions of the latter two, carrying the A230F+Q234F double mutation (mut1) in the first half of HTR6’s IC3 loop, whose sequence is shown below. (A, B) The G protein–coupled receptors (GPCRs) from (A), with EGFP fused to their C-termini, were expressed in IMCD3 cells and their cilia localization was analyzed by immunofluorescence with antibodies against EGFP (green), ARL13B (red) and gamma-tubulin (γTub, blue). Scale bar, 5 μm. (B, C) Percentage of GPCR-positive cilia in GPCR-transfected cells was quantitated from (B). Data are mean ± SEM of n = 3 to 5 independent experiments per construct, in each of which at least 50 transfected-cell cilia were counted for each GPCR. Data were analyzed by one-way ANOVA followed by Tukey’s multiple comparisons tests. Significance is indicated as P < 0.0001 (****). (D) Immunofluorescence pictures of Chimera N and Chimera N (mut1) showing the latter’s intracellular retention. Scale bar, 5 μm. (E) Percentage of transfected cells where indicated chimera was retained intracellularly with no observable plasma membrane staining was quantitated from immunofluorescence experiments. Data are mean ± SEM of n = 3 independent experiments, each with at least 150 transfected cells counted per chimera. Significance in unpaired two-tailed t test shown as P < 0.001 (***).

Figure S1.. HTR6 labels cilia not only more frequently but also more intensely than HTR7.

(A) IMCD3 cells expressing HTR6-EGFP or HTR7-EGFP were analyzed by immunofluorescence with antibodies against EGFP (green) and ARL13B (red). DNA was stained with DAPI (blue). Scale bar, 10 μm. (A, B) Quantitation of ciliary intensity of G protein-coupled receptor-positive cilia from (A). Cilia showing no detectable G protein-coupled receptor staining were not included in the analysis (as shown in Figs 2 and 3, HTR7 only labels about 20% of cilia, whereas HTR6 labels virtually 100%). Data are mean ± SEM of n = 42 (HTR6) and n = 31 (HTR7) cilia. Significance is shown as P < 0.0001 (****) for an unpaired two-tailed t test.

Figure 3.. HTR6 ciliary targeting relies on redundancy between IC3 loop (CTS1) and C-terminal tail (CTS2) sequences.

(A) Schematic representation of HTR6 (purple), HTR7 (green) and their chimeras, wherein purple segments come from HTR6 and green ones from HTR7. Ciliary targeting sequences in IC3 loop (CTS1) and C-terminal tail (CTS2) are labeled where present. (A, B) The G protein-coupled receptors (GPCRs) from (A), with EGFP fused to their C-termini, were expressed in IMCD3 cells and their cilia localization was analyzed by immunofluorescence with antibodies against EGFP (green), acetylated tubulin (AcTub, red) and gamma-tubulin (γTub, blue). Scale bar, 5 μm. (B, C) Percentage of GPCR-positive cilia in GPCR-transfected cells was quantitated from (B) as described in the Materials and Methods section. Data are mean ± SEM of n = 5, 3, 5, 5, 8 (from left to right) independent experiments, in each of which at least 50 transfected-cell cilia were counted for each GPCR. One-way ANOVA followed by Tukey’s multiple comparison tests shows all samples are significantly different from each other with P < 0.0001 (****) or P < 0.05 (*), as indicated.

Figure S2.. Plasma membrane targeting of HTR6 mutants.

The indicated HTR6 constructs, all containing C-terminal EGFP, were expressed in IMCD3 cells, which were analyzed by immunofluorescence with antibodies against EGFP (green), acetylated tubulin (AcTub, left) or ARL13B (right) (red) and gamma-tubulin (γTub, magenta). Cells were also stained with DAPI (DNA). All HTR6 constructs clearly label the plasma membrane. Scale bar, 5 μm.

Figure 4.. An LPG motif is critical for the ciliary targeting function of HTR6’s CTS2.

(A) Top: schematic depiction of HTR6 (440 aa, purple) and HTR7 (448 aa, green) with transmembrane helices displayed as boxes. Notice how HTR6 C-terminal tail (CT) is twice as long as HTR7-CT (115 aa versus 58 aa). Bottom: alignment of HTR6-CT and HTR7-CT. The former is 10-fold richer in prolines (17% versus 1.7%) and its latter half (aa 391–440) has no homologous counterpart in HTR7. LPG motif inside HTR6-specific tail is underlined. (B) Schematic representation of Chimera J and Chimera Q. They are identical except that Chimera Q lacks the HTR6 residues indicated in Chimera J, and contains instead the HTR7 residues indicated in Chimera Q. (C) Schematic showing HTR6-CT on top (present in Chimera J), and the deletions that were introduced into Chimera J, covering all residues that had not been substituted in Chimera Q. Indicated at the bottom is the critical region required for ciliary targeting of Chimera J. (B, C, D) IMCD3 cells expressing the constructs from (B, C), all fused to EGFP in their C termini, were analyzed by immunofluorescence with antibodies against EGFP (green), acetylated tubulin (AcTub, red) and gamma-tubulin (γTub, blue). Scale bar, 5 μm. (D, E) Percentage of G protein–coupled receptor (GPCR)–positive cilia relative to total transfected-cell cilia was quantitated from (D). (C, F) Nine separate mutations (CT-mut1 to CT-mut9) were introduced into critical region from (C), with all indicated residues replaced by alanines. (D, G) IMCD3 cells expressing C-terminally EGFP-tagged mut1-mut9 Chimera J mutants were analyzed as in (D). Scale bar, 5 μm. (E, G, H) Percentage of GPCR-positive cilia from (G) was quantitated as in (E). (D, I) The residues mutated in CT-mut3 and CT-mut4 were individually substituted to alanine and analyzed as in (D). (E, I, J) Percentage of GPCR-positive cilia from (I) was quantitated as in (E). In all quantitations, data are mean ± SEM of n = 3–5 independent experiments per construct, with at least 50 cilia counted per construct and experiment. Statistical analysis was performed by one-way ANOVA followed by Tukey’s multiple comparisons tests. Significance is indicated as P < 0.05(*), P < 0.001(***), or P < 0.0001(****).

Figure S3.. CDK5 phosphorylation of Ser-352 in HTR6-CT does not affect HTR6 ciliary targeting.

(A) IMCD3 cells expressing C-terminally EGFP-tagged Chimera J with the S352A or S352D mutations were analyzed by immunofluorescence with antibodies against EGFP (green), ARL13B (red) and γ-tubulin (γTub, magenta). DNA was stained with DAPI. Arrows indicate cilia. Scale bar, 5 μm. Serine-352 is the mouse HTR6 equivalent of human HTR6 Serine-350, shown to be a target of CDK5 phosphorylation (17). S352A and S352D are non-phosphorylatable and phosphomimetic S352 mutants, respectively. (A, B) Quantification of ciliary localization from (A). Data are mean ± SEM of n = 3 independent experiments per construct, in each of which at least 50 transfected-cell cilia were counted for each G protein-coupled receptor. No significant differences were found by one-way ANOVA.

Figure 5.. An RKQ motif is critical for IC3-dependent HTR6 ciliary targeting.

(A) Sequence of HTR6’s IC3 loop and its mutants used here. (B) Schematic of HTR6 wild type and its mutants used here. CT-mut3 is the mut3 CTS2 mutation from Fig 3. The IC3 mutations from (A) were combined with CT-mut3. Mutations shown as red spots. CTS1 and CTS2 encircled with dashed lines when intact. (C) IMCD3 cells expressing the indicated versions of HTR6, all fused to C-terminal EGFP, were analyzed by immunofluorescence with antibodies against EGFP (green), ARL13B (red) and gamma-tubulin (γTub, blue). Scale bar, 5 μm. (C, D) Percentage of G protein-coupled receptor-positive cilia relative to total transfected-cell cilia was quantitated for the constructs in (C). (C, E) The indicated HTR6 mutants were analyzed as in (C). (D, E, F) Ciliary targeting of HTR6 mutants from (E) was quantitated as in (D). (C, G) The indicated HTR6 mutants were analyzed as in (C). (D, G, H) Ciliary targeting of HTR6 mutants from (G) was quantitated as in (D). Data in (D, F, H) are mean ± SEM of n = 3–4 independent experiments per construct, with at least 50 cilia counted per construct and experiment. Data were analyzed by one-way ANOVA followed by Tukey’s multiple comparisons tests. Unless otherwise indicated, significance is shown relative to control sample (black column) with P < 0.001(***) or P < 0.0001(****).

Figure 6.. HTR6 CTS1 and CTS2 are both sufficient for ciliary targeting.

(A) Schematic of HTR7 with or without the CTS2-containing C-terminal tail of HTR6 (HTR6-CT) fused to its C terminus. (A, B) IMCD3 cells expressing C-terminally EGFP-tagged constructs from (A) were analyzed by immunofluorescence with antibodies against EGFP (green), acetylated tubulin (AcTub, red) and gamma-tubulin (γTub, blue). Scale bar, 5 μm. (B, C) Percentage of G protein-coupled receptor-positive cilia relative to total transfected-cell cilia was quantitated from (B). (D) Schematic of CD8α(1-206) chimeras, containing extracellular and transmembrane regions of CD8α fused to HTR7-CT, HTR6-CT (containing CTS2) or HTR6-IC3 (containing CTS1). (B, D, E) IMCD3 cells expressing C-terminally EYFP-tagged constructs from (D) were analyzed as in (B). Scale bar, 5 μm. (E, F) Percentage of G protein–coupled receptor-positive cilia relative to total transfected-cell cilia was quantitated from (E). Data in (C, F) are mean ± SEM of n = 4–5 independent experiments per construct. In each experiment, at least 50 cilia were counted per condition. (C, F) Data were analyzed by unpaired two-tailed t test (C) or by one-way ANOVA followed by Tukey’s multiple comparison tests (F). Significance in both cases is shown as P < 0.0001 (****).

Figure S4.. Membrane association is needed for HTR6-CT to function as a ciliary targeting sequence.

(A) IMCD3 cells expressing C-terminally EGFP-tagged HTR6-CT were analyzed by immunofluorescence with antibodies against EGFP (green), acetylated tubulin (AcTub, red) and γ-tubulin (γTub, magenta). DNA was stained with DAPI. Arrows indicate cilia. Scale bar, 5 μm. (A, B) Quantification of ciliary localization from (A). Data are mean ± SEM of n = 5 and n = 3 independent experiments for CD8α(1-206)-CT(HTR6)-EYFP and CT(HTR6)-EGFP, respectively. In each experiment, at least 50 transfected-cell cilia were counted for each construct. Significance in unpaired two-tailed t test is shown as P < 0.0001(****).

Figure 7.. SSTR3 ciliary targeting also depends on redundant ciliary targeting sequences in IC3 and CT.

(A) Schematic of CD8α(1-206) chimeras fused to SSTR3-IC3 or SSTR3-CT. (A, B) IMCD3 cells expressing C-terminally EYFP-tagged constructs from (A) were analyzed by immunofluorescence with antibodies against EGFP/EYFP (green), ARL13B (red), and gamma-tubulin (γTub, blue). Scale bar, 5 μm. (B, C) Percentage of G protein-coupled receptor-positive cilia relative to total transfected-cell cilia was quantitated for the constructs in (B). Data are mean ± SEM of n = 4, 8 (IC3, CT) independent experiments per construct. Significance in unpaired two-tailed t test shown as P < 0.01(**). (D) SSTR3-IC3 wild-type sequence (top) and its mutated versions used below. The two reported Ax(A/S)xQ motifs are underlined in wild-type sequence. (E) SSTR3-CT wild-type sequence (top) and its mutated versions used below. (F) Schematic of SSTR3 and its mutants used below. CTS1 and CTS2 are encircled where intact. Mutations shown as red spots. (B, F, G) Ciliary targeting of SSTR3 mutants from (F) was analyzed as in (B). (C, G, H, I) Percentage of positive cilia for each of the indicated SSTR3 constructs from (G) was quantitated as in (C). (J) Intensity of ciliary staining was quantitated for the indicated SSTR3 constructs. (K) Percentage of cells with no detectable plasma membrane or ciliary staining was quantitated for indicated constructs. Data in (H, I, J, K) are mean ± SEM and were analyzed by one-way ANOVA followed by Tukey’s multiple comparison tests. Significance shown as P < 0.05(*), P < 0.01(**), P < 0.001(***), P < 0.0001(****) or not significant (n.s.). For (H), numbers of independent experiments per construct from left to right were n = 10, 3, 4, 3, 3, 3, 4, 4, 3, 4, 10. Equivalent numbers for (I) were n = 3, 4, 4, 5, 5, 5, 4, 4. For both (H, I), at least 50 cilia were counted per construct and experiment. For (J), intensity was measured in n = 26–59 cilia per condition in one representative experiment. For (K), n = 4 independent experiments per construct with at least 200 transfected cells assessed per construct and experiment.

Figure S5.. Plasma membrane targeting of SSTR3 mutants.

(A) The indicated SSTR3 constructs, all containing C-terminal EGFP, were expressed in IMCD3 cells and analyzed by immunofluorescence with antibodies against EGFP (green), ARL13B (red) and γ-tubulin (γTub, magenta). Cells were also stained with DAPI (DNA). All six constructs clearly reach the plasma membrane. (A, B) Indicated constructs were analyzed as in (A). IC3-mut1+CT-D335-428 mutant, lacking all but the first 10 residues in SSTR3-CT, consistently accumulates intracellularly and fails to reach plasma and ciliary membranes (left panels). In contrast, CT-mut1+IC3-D5, despite its intracellular retention frequency being much higher than normal (middle panels and Fig 7K), can still reach ciliary and/or plasma membrane in about 60% of transfected cells (right panels). Scale bars, 5 μm.

Figure 8.. Ciliary targeting of HTR6 and SSTR3 in hippocampal neurons also depends on redundancy between CTS1 and CTS2.

(A) Schematic of the HTR6 constructs used here. (A, B) Cultured hippocampal neurons expressing C-terminally EGFP-tagged constructs from (A) were analyzed by immunofluorescence with antibodies against adenylate cyclase 3 (ADCY3, red). DNA was stained with DRAQ5 (blue) and EGFP fluorescence was directly visualized. (C) Schematic of the SSTR3 constructs used here. (B, C, D) Cultured hippocampal neurons from (C) were analyzed as in (B). Scale bars, 10 μm.

Figure 9.. HTR6 and SSTR3 CTs associate with ciliary trafficking adapter TULP3.

(A, B) Schematic and protocol of BioID2 proximity labeling assay. HEK293T cells were cotransfected with plasmids encoding EGFP-TEV-Stag-TULP3 and a fusion protein containing the extracellular and transmembrane regions of CD8α (aa 1–206), the C-terminal tail (CT) or third intracellular loop (IC3) of a G protein-coupled receptor, the BioID2 biotin ligase, and an HA epitope. In presence of biotin (50 μM, 16 h), BioID2 biotinylates surrounding proteins in a proximity-dependent manner. After cell lysis, TULP3 was affinity purified by two sequential immunoprecipitations (IP) and its biotinylation assessed by Western blot (WB). (C) SDS–PAGE and WB analysis of immunoprecipitated S-tagged TULP3 (top two panels) and of the cleared cell lysates (bottom two). In the IPs, NeutrAvidin-HRP was used to detect TULP3 biotinylation (top) and anti-Stag antibody to detect its total levels. In the lysates, anti-EGFP and anti-HA tag antibodies were used to detect EGFP-TEV-Stag-TULP3 and the CD8 fusions, respectively. Molecular weight markers are indicated on the right (kD). (C, D, E) Biotinylated TULP3 signal, relative to total TULP3 signal in IPs, was quantitated from n = 5 independent experiments like the one in (C). Biotinylation by IC3 constructs (D) and by CT constructs (E) was normalized relative to β2AR-IC3 and β2AR-CT, respectively. Data are mean ± SEM and were analyzed by one-way ANOVA followed by Dunnett’s multiple comparison tests. Significance shown as P < 0.0001 (****). Source data are available for this figure.

Figure S6.. The Tubby domain of TULP3 is responsible for its association to HTR6-CT.

(A) Schematic of TULP3 protein depicting its N-terminal (NTD) and C-terminal (CTD) domains. The NTD contains an IFT-A-binding site, whereas the CTD consists of the phosphoinositide-binding Tubby domain. (B) Proximity biotinylation assay as done in Figs 9 and 10. In this case, biotinylation was analyzed for tandem immunoprecipitated S-tagged TULP3-NTD (aa 1–183) or TULP3-CTD (aa 184–442) (top two panels). Analysis of cleared cell lysates is shown in bottom panels. Molecular weight markers are indicated on the right (kD).

Figure 10.. HTR6 CTS2 antagonizes TULP3 association to HTR6-CT.

(A) Schematic of the CD8α (aa 1–206)-(HTR6-CT)-BioID2-HA constructs used here, showing only the HTR6-CT moiety. The CTS2 is shown as a blue oval and red crosses indicate the CT-mut3 mutation. Dashed lines indicate deleted regions. The intensity of TULP3 association is displayed on the right. At bottom, the regions in HTR6-CT promoting TULP3 association are shown in green, and the CTS2-containing region antagonizing TULP3 association is shown in red. (B) SDS–PAGE and WB analysis of tandem immunoprecipitated S-tagged TULP3 (top two panels) and of cleared cell lysates (bottom two). In the IPs, NeutrAvidin-HRP was used to detect TULP3 biotinylation (top) and anti-Stag antibody to detect its total levels. In the lysates, anti-EGFP and anti-HA tag antibodies were used to detect EGFP-TEV-Stag-TULP3 and CD8 fusions, respectively. Molecular weight markers on the right (kD). (B, C) Quantitation from (B) of biotinylated TULP3 signal, relative to total TULP3 in IP, and normalized to β2AR-CT sample. Data are mean ± SEM (n = 7, 6, 7, 5, 5, 3, 5, 4, 4 independent experiments, from left to right). Data were analyzed by one-way ANOVA followed by Tukey’s multiple comparisons tests. Significance shown as P < 0.05(*), P < 0.001(***), or P < 0.0001(****). Where not explicitly indicated, asterisks represent significance relative to β2AR-CT. Source data are available for this figure.

Figure 11.. TULP3 regulates both CTS1 and CTS2 function.

(A) Genomic characterization of the two CRISPR-generated TULP3-KO IMCD3 clones used in this figure. Allele nomenclature corresponds to Ensembl mouse transcript Tulp3-201 and follows Human Genome Variation Society (HGVS) guidelines (52). (B) TULP3-KO clones still form cilia, as seen by immunostaining with IFT88 (green) and g-tubulin (red) antibodies. DAPI in blue. Scale bar, 10 μm. (B, C) Percentage of ciliated cells was quantitated from (B). Data are mean ± SEM of n = 33 fields of cells, each containing at least 30 cells, from two coverslips. No significant differences were found by one-way ANOVA. (D) CD8α-IC3(HTR6)-Stag-TEV-EYFP was transfected into TULP3-KO clones, or WT IMCD3 cells as control, and its ciliary localization assessed by immunostaining with Stag (green), IFT88 (red), and g-Tubulin (magenta) antibodies. DAPI-stained nuclei in blue. CD8α-IC3(HTR6) levels are very low or undetectable in TULP3-KO cilia (arrows). Scale bar, 10 μm. (D, E) Quantitation of CD8α-IC3(HTR6) ciliary intensity from (D). The percentage of cilia in each of the indicated categories is shown. The number of transfected cell cilia counted in each condition is displayed at the bottom, together with statistical significance from chi-square tests comparing each mutant distribution to that of WT (P < 0.0001 (****)). (D, F) Same analysis as in (D) was performed for CD8α-CT(HTR6)-Stag-TEV-EYFP, which again localizes to WT but very weakly or not at all in TULP3-KO cilia (arrows). (E, F, G) Quantitation of CD8α-CT(HTR6) ciliary intensity from (F) was performed and analyzed as in (E).

Figure 12.. RABL2 regulates HTR6 ciliary targeting mostly via CTS1.

(A) Lysates from HEK293T cells expressing the proteins indicated at the top were immunoprecipitated with anti-EGFP antibodies and analyzed by Western blot with anti-Myc and anti-EGFP antibodies, as indicated. At bottom, Myc signal immunoprecipitated by EGFP-RABL2 is quantitated relative to Myc signal in the corresponding lysate, and normalized relative to Myc-HTR6 (100%). Molecular weight markers are shown on the right. Input is 2.5% of lysate used for IP. (A, B) Co-IP experiment as in (A), but with the constructs indicated at the top. (A, C) Co-IP experiment as in (A), but with the constructs indicated at the top. In the quantitations below, each mutant is normalized to its respective control (and β2AR-IC3 is normalized to HTR6). (D) Genomic characterization of the two CRISPR-generated RABL2-KO IMCD3 clones used in this figure. Allele nomenclature corresponds to Ensembl mouse transcript Rabl2-201 and follows Human Genome Variation Society (HGVS) guidelines (52). (E) Ciliogenesis is strongly reduced in RABL2-KO clones, as seen by immunostaining with ARL13B (red) antibodies. DAPI in blue. Scale bar, 10 μm. (E, F) Percentage of ciliated cells was quantitated from (E). Data are mean ± SEM of n = 12 fields of cells, each containing at least 30 cells, from two coverslips. Data were analyzed by one-way ANOVA with Tukey’s multiple comparisons tests (P < 0.0001 (****)). (G) EYFP-tagged CD8α-IC3(HTR6) was transfected into RABL2-KO clones, or WT IMCD3 cells as control, and its ciliary localization assessed by immunostaining with EGFP (green) and ARL13B (red) antibodies. DAPI-stained nuclei in blue. CD8α-IC3(HTR6) levels are very low or undetectable in RABL2-KO cilia (arrows). Scale bar, 10 μm. (G, H) Quantitation of CD8α-IC3(HTR6) ciliary intensity from (G). Percentage of cilia in each of the indicated categories is shown. Number of transfected cell cilia counted in each condition is displayed at the bottom, together with statistical significance from chi-square tests comparing each mutant distribution to that of WT (P < 0.0001 (****)). (G, I) Same analysis as in (G) was performed for CD8α-CT(HTR6)-EYFP. Arrows point to cilia. Scale bar, 10 μm. (H, I, J) Quantitation of CD8α-CT(HTR6) ciliary intensity from (I) was performed and analyzed as in (H). (A, K) Co-IP experiment as in (A) but with the indicated constructs and antibodies. (L) Model of HTR6 ciliary targeting. Double-headed arrows represent physical interactions. Single-headed arrows represent positive effects. Also shown is the antagonism of CTS2 on TULP3 binding to HTR6-CT. We hypothesize CTS2 promotes intraciliary dissociation of TULP3, thereby freeing it for further rounds of transport. Whether HTR6-IC3 directly binds to IFT-A, as shown for SSTR3-IC3, remains unknown (48). Source data are available for this figure.

Figure S7.. CTS2 mutation reduces RABL2B binding to Chimera J but not to wild-type HTR6.

(A) Lysates from HEK293T cells expressing the proteins indicated on top were immunoprecipitated with anti-Flag antibodies and analyzed by Western blot with anti-HTR6 and anti-Flag antibodies, as indicated. (A, B) Experiment as in (A) except that effect of CT-mut3 was tested in myc-HTR6 rather than myc-Chimera J, as indicated. Molecular weight markers are shown on right of all panels.

Figure S8.. Evolutionary conservation of HTR6 and SSTR3 ciliary targeting sequences.

(A) HTR6 orthologs from indicated species were aligned using Clustal Omega and the resulting alignment visualized with MView. The displayed region includes the first half of HTR6-IC3, as indicated on top. The RKQxxxV motif we identified in mouse HTR6 is shown in the orange box below and the mouse-specific ATAGQ motif is marked with a red rectangle. The 80% consensus sequence is also shown. (A, B) HTR6 orthologs were aligned and visualized as in (A) and the C-terminal tail region containing the LPG motif (orange box) is shown here. (A, B, C) SSTR3 orthologs were aligned and visualized as in (A, B). The orange box shows the residues we found involved in ciliary targeting of mouse SSTR3, with A-Q motifs underlined. (C, D) SSTR3 orthologs were aligned and visualized as in (C). The juxtamembrane region of SSTR3-CT is shown. The regions we found to affect mouse SSTR3 ciliary targeting are shown in the orange box below, with the FK and LLxP motifs underlined. Sequence identifiers for HTR6 orthologs are: tr|E9QFU3|, tr|A0A0G2JYX7|, sp|Q9R1C8|, tr|E1BE22|, sp|P50406|, sp|Q5IS65|, tr|A0A1L8FPD2|, tr|A0A151MJI5|, and tr|D2XUT1| for zebrafish, rat, mouse, bovine, human, chimp, toad, alligator and chick, respectively. For SSTR3 orthologs they are: tr|A0A1L8GH74|, tr|A2BHH9|, sp|P30936|, sp|P30935|, tr|G3MWT4|, sp|P32745|, tr|A0A2J8QPP4|, tr|Q4ZJF2|, and tr|A0A151NHS7| for toad, zebrafish, rat, mouse, bovine, human, chimp, chick and alligator, respectively. The lowercase letters in the consensus rows stand for: hydrophobic (h), small (s), aliphatic (l), polar (p), turnlike (t), tiny (u), charged (c), alcohol (o), and positive (+).