Ciliary Generation of a Peptidergic Sexual Signal

Peptidergic intercellular communication occurs throughout the eukaryotes, and regulates a wide range of physiological and behavioral responses. Cilia are sensory and secretory organelles that both receive information from the environment and transmit signals. Cilia derived vesicles (ectosomes), formed by outward budding of the ciliary membrane, carry enzymes and other bioactive products; this process represents an ancient mode of regulated secretion. Our previous study revealed the presence of the peptide amidating enzyme, peptidylglycine α-amidating monooxygenase (PAM), in cilia and its key role in ciliogenesis. Furthermore, PAM and its amidated products are released in ciliary ectosomes from the green alga Chlamydomonas reinhardtii. One amidated product (GATI-amide) serves as a chemotactic modulator for C. reinhardtii gametes, attracting minus gametes while repelling plus gametes. Here we dissect the complex processing pathway that leads to formation of this amidated peptidergic sexual signal specifically on the ectosomes of plus gametes. We also identify a potential prohormone convertase that undergoes domain rearrangement during ectosomal secretion as a substrate for PAM. Analysis of this pathway affords insight into how single-celled organisms lacking dense core vesicles engage in regulated secretion, and provides a paradigm for understanding how amidated peptides that transmit sexual and other signals through cilia are generated.


26
Cilia are membrane-delimited, microtubule-based cell extensions that protrude into the extracellular 27 space and function as key motile, sensory and secretory organelles in many eukaryotes (Marshall and 28 Basto, 2017). These complex organelles that were present in the last eukaryotic common ancestor both 29 receive and transmit signals (Carvalho-Santos et al., 2011; Malicki and Johnson, 2017). Proteins encoded 30 by approximately 5% of the human genome contribute to their assembly, structure and function (van Dam 31 et al., 2019). Mutations in many of these genes cause ciliopathies, with phenotypes ranging from 32 neurological malformations, skeletal abnormalities and kidney disease to obesity and insulin resistance 33 (Reiter and Leroux, 2017). The ciliary localization of receptors for peptides such as Wnt, Hedgehog, 34 insulin, somatostatin and α-melanocyte stimulating hormone (αMSH) plays an essential role in their 35 signaling ability ( mating type (termed minus and plus). Ectosome release increases rapidly when minus and plus gametes 64 are mixed (Cao et al., 2015). The interaction of their cilia triggers a complex intraciliary signaling pathway 65 that leads to loss of gametic cell walls, formation of mating structures, and cell fusion, yielding a 66 quadriciliate cell that ultimately develops into a diploid zygote. When nutrient conditions improve, the 67 zygote hatches, releasing haploid meiotic progeny (Harris, 2009). 68 Mass spectrometric analysis of mating ciliary ectosomes led to the identification of an amidated peptide, 69 derived from Cre03.g204500, that acts as a chemotactic modulator, attracting minus gametes while 70 repelling plus gametes (Luxmi et al., 2019). Amidated peptides from echinoderms, Hydra, vespids and 71 humans have also been reported to induce chemotaxis (Palma, 2006 expected of a prepropeptide (hereafter referred to as preproGATI) (Luxmi et al., 2019). Acting on the 74 proprotein (proGATI) created by removal of the signal sequence, a carboxypeptidase B-like enzyme could 75 remove three Arg residues, thus generating a substrate for PAM and production of an amidated C-76 terminus ending -Gly-Ala-Thr-Ile-NH 2 (GATI-NH2). 77 The C. reinhardtii genome encodes many proteins with the characteristics of prepropeptides. Mating 78 ciliary ectosomes contain proteins derived from several of these prepropeptides, along with the subtilisin-79 like enzymes needed for their cleavage, PAM and multiple amidated products (Luxmi et al., 2019). C-80 terminal amidation is often required for peptide bioactivity, as it can greatly enhance affinity for the 81 cognate receptor and confers resistance to proteolytic degradation (Luxmi et al., 2021). Our data suggest 82 that the mating type-specific production and release of bioactive products in ciliary ectosomes represent 83 an evolutionarily ancient path to achieving their regulated secretion. 84 Here we define the complex processing and amidation pathway leading to formation of the C. reinhardtii 85 chemotactic sexual signal and determine how it is trafficked through cilia and ultimately released into 86 ciliary ectosomes and the soluble secretome. We also find that one potential ciliary-localized prohormone 87 convertase is itself a PAM substrate and undergoes an alteration in domain organization during ciliary 88 trafficking, coincident with release of the peptidergic sexual signal. Analysis of proGATI, which yields 89 bioactive products, provides a route to understanding how regulated secretion can occur in a single celled 90 organism lacking peptide storage vesicles. As PAM, peptide processing enzymes, and cilia are broadly 91 conserved in eukaryotes, this study provides a paradigm for understanding how amidated products that 92 transmit chemotactic sexual and other signals through cilia can be generated. 93

Mating ectosomes contain proGATI along with N-terminal and C-terminal fragments of proGATI 95
Tryptic peptides derived from preproGATI, the protein encoded by Cre03.g204500 and consisting of 908 96 residues, were identified in both mating ectosomes and the soluble secretome (Luxmi et al., 2018;Luxmi 97 et al., 2019). Interestingly, a proGATI peptide that had been α-amidated and terminated with the 98 sequence -Gly-Ala-Thr-Ile-amide (GATI-amide) was identified in mating ectosomes (Luxmi et al., 2019) and 99 in one of six secretome samples analyzed previously (Luxmi et al., 2018). Removal of the N-terminal signal 100 sequence from preproGATI would yield proGATI, with a calculated molecular mass of 90.6 kDa. The 101 amidation of proGATI requires the removal of three Arg residues by a carboxypeptidase B-like 102 exoprotease (Fig. 1A), generating a Gly-extended protein that can serve as a PAM substrate; following α-103 hydroxylation of this Gly residue by PHM, PAL-mediated cleavage produces a protein terminating with a 104 C-terminal Ile-amide (Fig. 1A). 105 To explore the biosynthesis, post-translational processing, trafficking and secretion of products generated 106 by the endoproteolytic cleavage of proGATI, we prepared antibodies to a synthetic peptide located near 107 its N-terminus (N-ter peptide) and to a peptide that included the amidated C-terminus (C-ter peptide) (Fig.  108 1A). Three rabbits were injected with a mixture of carrier-conjugated synthetic peptides and mating 109 ectosomes were used to evaluate the sera. Bands of similar apparent molecular mass were visualized in 110 varying amounts by all three sera; the most prominent appeared at ~250 kDa, ~120 kDa, ~75 kDa and ~63 111 kDa (Fig. 1B); post-translational modifications such as N-glycosylation and O-glycosylation can have a 112 dramatic effect on the apparent molecular mass of proteins (Bollig et al., 2007;Voigt et al., 2007). 113 To test whether these bands were specific, sera were pre-incubated with N-ter peptide, amidated-C-ter 114 peptide or a mixture of both. Pre-incubation with N-ter peptide eliminated the 250-kDa, 120-kDa and 63-115 kDa signals. Pre-incubation with the amidated-C-ter peptide eliminated the 250-kDa signal and the 75-116 kDa signal (Fig. 1B). The ability of both peptides to block the appearance of the 250-kDa band suggests 117 that it is an extensively modified form of proGATI. The presence of multiple smaller products indicates 118 that proGATI is subjected to endoproteolytic cleavage. 119 To determine if the signal produced by the C-ter antibody required amidation, serum was pre-incubated 120 with the amidated C-ter peptide (GATI-NH2), the Gly-extended peptide (GATI-Gly) or GATI-OH, which has 121 a free carboxyl group at its C-terminus (Fig. 1C). For both the 250-kDa proGATI band and the 75-kDa band, 122 the signal was greatly reduced by pre-incubation with the GATI-NH2 peptide, partially attenuated by pre-123 incubation with the GATI-Gly peptide and unaffected by the GATI-OH peptide. These data indicate that at 124 least a fraction of the 250-kDa proGATI and 75-kDa product in mating ectosomes is amidated (Fig. 1C). 125 Affinity-purification was used to prepare antibodies that recognized either the N-terminal region or the 126 amidated C-terminus of proGATI. In agreement with the peptide blocking experiments, affinity-purified 127 N-ter antibody recognized the 250-kDa, 120-kDa and 63-kDa bands in mating ectosomes while C-ter 128 antibody affinity-purified using the GATI-NH2 peptide recognized the 250-kDa and 75-kDa bands ( Fig.  129 S1A). The specificity of the affinity-purified antibodies was quantified using solid phase assays (Figs. S1B 130 and C). 131 These data suggest that the 250-kDa protein visualized by both antibodies is a heavily modified version of 132 proGATI, a significant fraction of which is α-amidated. Endoproteolytic cleavage could generate an 133 amidated 75-kDa C-ter fragment along with a 120-kDa N-ter fragment. An additional cleavage could yield 134 a 63-kDa N-ter fragment along with a fragment that would not be recognized by either antibody. 135 Endoproteolytic cleavage of proGATI generates a heavily glycosylated 75 kDa product that contains the 136 amidated chemomodulatory peptide. 137 We next used mass spectrometry to identify the proGATI region included in the amidated 75-kDa C-ter 138 fragment immunoprecipitated from mating ectosomes (Fig. 1D) Treatment of mating ectosomes with PNGase F, which removes many mammalian N-linked sugars,  155 reduced the apparent molecular mass of a small fraction of the 75-kDa product detected by the C-ter 156 antibody (Fig. S1E); treatment with an O-glycosidase/neuraminidase cocktail was without effect. The 157 mobility of the 120-kDa proGATI fragment recognized by the N-ter antibody was unaltered by either 158 treatment (Fig. S1E). The non-canonical lipid-linked oligosaccharide identified in C. reinhardtii, along with 159 its lack of N-acetylglucosaminyltransferase I, which is required for maturation of N-linked extracts, but not in spent medium ( Fig. 2A). The fact that spent medium contains a single 170-kDa protein 174 recognized by both N-ter and C-ter antibodies led to its identification as HEK-proGATI; differences in the 175 N-and O-linked oligosaccharides attached to proGATI produced by HEK cells and by C. reinhardtii would 176 account for the difference in apparent molecular mass. 177 To test whether HEK-293 cells amidate proGATI, bathocuproine disulfonate (BCS) was used to deplete 178 cellular copper, inhibiting the activity of the amidating enzyme, PAM (Bonnemaison et al., 2015). While 179 the 170-kDa N-ter signal was unaltered following BCS treatment, the 170-kDa C-ter signal fell dramatically 180 (Fig. 2B). To account for any differences in secretion rate, the ratio of 170-kDa C-ter signal to 170-kDa N-181 ter signal was calculated. BCS treatment caused a four-fold reduction in this ratio, consistent with the 182 conclusion that HEK-293 cells amidate the C-terminus of the proGATI that they secrete (Fig. 2C). 183 ProGATI includes six potential N-glycosylation sites (Asn-X-Ser/Thr) and several potential O-glycosylation 184 sites (-Ser/Thr and HyP) (Fig. S1D). Digestion with either PNGase F or a mixture of O-glycosidase and 185 neuraminidase reduced the apparent molecular mass of secreted HEK-proGATI by ~15-20 kDa, consistent 186 with the occurrence of extensive N-and O-glycosylation (Fig. 2D). 187 Successful ectosome-mediated delivery of a chemomodulatory peptide such as GATI-NH 2 would require 188 it to be resistant to proteolysis. Spent medium containing HEK-proGATI was used to test this hypothesis. 189 Exposure to increasing amounts of trypsin eliminated the N-ter signal and generated a series of smaller 190 products recognized by the C-ter antibody (Fig. 2E). Cleavage at the single Lys residue in the N-ter peptide 191 is consistent with this result (Fig. 1A). Trypsin produced a sequence of smaller products detected by the 192 C-ter antibody. C-ter signal intensity was not diminished, with essentially complete conversion of 170 kDa 193 HEK-proGATI into a 50 kDa and then a 37-kDa product, which may resemble the amidated 75 kDa C-ter 194 fragment found in mating ectosomes (Fig. 2E). 195

Purification and domain organization of proGATI 196
Since HEK-proGATI is amidated and secreted rapidly (Fig. S2B), we undertook its purification from spent 197 medium ( Fig. 3A and S2C) and analysis using mass spectrometry. Although the N-and O-glycans attached 198 to HEK-proGATI will differ from those attached to proGATI produced by C. reinhardtii, the sites available 199 to enzymes involved in N-and O-glycosylation are expected to be the same. Purified native and 200 deglycosylated HEK-proGATI were analyzed, using a cocktail of enzymes designed to remove both N-and 201 O-linked glycans. Deglycosylation reduced its apparent molecular mass by ~30 kDa (Fig. 3B). 202 Mass spectrometry of native HEK-proGATI identified four O-glycosylated Ser residues and two O-203 glycosylated HyP residues. Analysis of enzymatically deglycosylated HEK-proGATI identified five N-204 glycosylation sites. The amidated C-terminus (-GATI-NH2) was found in both samples; C-terminal peptides 205 ending in -Gly and -Gly-Arg were also identified indicating that carboxypeptidase processing and 206 amidation had not gone to completion ( Fig. 3C and D). Peptides spanning the entire sequence of HEK-207 proGATI were identified (87.3% coverage) (Fig. 3D). 208 The ability of trypsin to convert amidated HEK-proGATI into stable, amidated products as small as 37 kDa 209 (Fig. 2E), is consistent with the presence of stable domains. To explore this possibility, a structural model 210 of proGATI was generated using RoseTTAFold (Baek et al., 2021). The proGATI prediction includes three 211 well-folded domains connected by highly extended, Pro-rich flexible linkers (Fig. 3E). The signal sequence 212 was not included in the structural model. N-terminal domain 1 contains 323 residues (green), terminating 213 just before a Pro-rich region. Domain 2 includes 153 residues and domain 3 has 213 residues, ending at 214 the C-terminus. The 70-residue linker between domains 1 and 2 contains 37 Pro residues and a furin-like 215 site (K 407 PRK), while 43 of the 99 residues in the second linker are Pro residues; these very Pro-rich regions 216 likely contribute to the abnormal migration of proGATI during SDS-PAGE. Domain 3 forms an antiparallel 217 β-sandwich and has a nominal molecular mass of 23 kDa with a pI of 10 ( Fig. 3E). This domain corresponds 218 precisely to the C-terminal region identified by mass spectrometry of the 75-kDa amidated product in 219 mating ectosomes, and is immediately preceded by a furin-like cleavage site. Cleavage at this site alone 220 would release domains 1 and 2 (predicted to represent the 120-kDa N-terminal fragment), while further 221 proteolysis at K 407 might generate the 63-kDa N-terminal product. Domain 3, which includes four Cys 222 residues, has a single predicted disulfide bond (C 739 and C 745 ); although C 742 and C 817 are located close to 223 each other, a significant rearrangement would be needed for disulfide bond formation. Importantly, the 224 experimentally confirmed C-terminal amidation site, the Pro residue that is O-glycosylated (P 896 ) and both 225 Asn residues that are N-glycosylated (Asn 814 ThrThr and Asn 833 GlnThr) are completely exposed and 226 accessible for modification in the model structure (Fig. S3). 227

Ciliary localization and mating type-specific processing of proGATI 228
Under nutrient deprivation conditions, C. reinhardtii cells differentiate into minus and plus gametes that 229 expresses mating type-specific genes, enabling them to recognize each other. Our previous study showed 230 that CrPAM expression increased during gametogenesis, and that the C-ter antigenic peptide and a longer 231 synthetic amidated peptide (VLYPNDPAAYAAYAPGTGGGATI-NH2) produced a mating type-specific 232 chemotactic response, attracting minus gametes and repelling plus gametes (Luxmi et al., 2019). These 233 observations prompted investigation of proGATI in gametes. Cells, deciliated cell bodies and cilia were 234 subjected to immunoblot analysis. Use of the N-ter and C-ter antibodies revealed enrichment of 250-kDa 235 proGATI in the cilia of both minus and plus gametes (Fig. 4A). In contrast, the C-ter antibody detected a 236 75-kDa band only in plus gametes; while detectable in plus gamete cells, the 75-kDa band was enriched in 237 plus gamete cilia and essentially undetectable in deciliated cell bodies. Strikingly, production of amidated 238 75-kDa GATI is specific to plus gametes ( Fig. 4A and 4B). Mass spectrometric analysis of 75-kDa GATI 239 immunoprecipitated from plus gamete cell lysates confirmed complete amidation of its C-terminus and 240 the presence of peptides like those identified in 75-kDa GATI immunoprecipitated from mating ectosomes 241 (Figs. 1E, F and S4). 242 We next used immunofluorescence microscopy to determine the subcellular localization of GATI-derived 243 proteins in resting gametes. Maximal Z-projection confocal images of minus and plus gametes showed 244 that the C-ter GATI signal was localized in discrete puncta throughout the cytoplasm (Fig. 4C); this signal 245 could represent intact 250-kDa proGATI and/or the 75-kDa C-ter product derived from it. In contrast, 246 simultaneous visualization of FMG1, a ciliary membrane glycoprotein, revealed more signal in the cilia and 247 around the margins of the cell body. Single Z-stack images showed the accumulation of C-ter GATI signal 248 at the cell surface, co-localized with FMG1 ( Fig. 4C, inset). Diffuse C-ter GATI signal along the length of the 249 cilia was also observed in both mating types (Fig. 4C). To confirm staining specificity, the C-ter antibody 250 was pre-incubated with antigenic peptide; signal intensity (green) was greatly reduced in the cell body, 251 on the cell surface and in cilia of plus gametes ( both mating ectosomes and the secretome. We utilized affinity-purified proGATI antibodies and 260 immunogold-electron microscopy to determine its ectosomal localization. 261 Ectosomes from mating gametes were embedded in agarose and imaged by thin section transmission EM 262 (Fig. 5A); the vesicles range from ~80 nm to ~260 nm in diameter. Following incubation with intact 263 ectosomes, affinity-purified proGATI antibodies were visualized using a gold-tagged anti-rabbit secondary 264 antibody and negative stain EM; signals obtained with both antibodies were localized on the ectosome 265 surface (Fig. 5A). Ectosomes incubated only with gold-conjugated secondary antibody served as a negative 266 control. 267 Ectosomes, deciliated mixed gamete cell bodies and cilia prepared 1 h after the initiation of mating were 268 subject to immunoblot analysis (Fig. 5B). Based on use of both the N-ter and C-ter antibodies, cilia 269 contained only 250-kDa proGATI. In contrast, mating ectosomes contained 250-kDa proGATI along with 270 the 120 kDa N-ter fragment and the 75 kDa C-ter fragment, whereas proGATI products were not detected 271 in the cell bodies ( cilia; although mating ectosomes contained CrPAM, its ectosomal levels did not exceed those in the cell 280 body (Fig. 5D). FMG1 levels in cilia and mating ectosomes greatly exceeded those in cell bodies, but FMG1 281 levels in mating ectosomes did not exceed those in cilia (Fig. 5E). Thus, the cell bodies of mating gametes 282 were essentially devoid of proGATI while both CrPAM and FMG1 were readily detected. 283

Differential release of proGATI products from mating ectosomes 284
We previously found that both CrPAM protein and enzyme activity associate with the ciliary axoneme; 285 this interaction is disrupted by treatment with 0.6 M NaCl following detergent extraction (Kumar et al., 286 2016a). To explore the ciliary distribution of proGATI and its products, we isolated cilia from resting 287 gametes of both mating types and from 1 h mixed gametes. Isolated cilia were first treated with Triton X-288 100 to release membrane proteins and soluble matrix components. This was followed by treatment with 289 0.6 M NaCl to extract proteins that were tightly bound to the axoneme; the resulting extracted axoneme 290 pellet was solubilized in 1× SDS buffer. 291 The amidated 75-kDa GATI product was detected in the cilia of plus but not minus gametes ( terminal product and N-terminal 63-kDa segment were both released by washing with low ionic strength 305 HEPES buffer; release did not occur following chelation of divalent cations with EDTA. Although not 306 released by buffer alone or by EDTA treatment, the N-terminal 120-kDa GATI fragment was partially 307 displaced from ectosomes by 10 mM DTT. This effect was DTT-specific; treatment with 10 mM β-308 mercaptoethanol had no effect (Fig. S5). These results suggest that all three domains individually mediate 309 associations with the ectosomal surface. This tripartite attachment mechanism likely explains why release 310 of 250-kDa amidated proGATI was not observed under any conditions. 311

Distribution, processing and amidation of putative prohormone convertases in cilia 312
The appearance and accumulation of the 75-kDa amidated proGATI product on plus (but not minus) 313 gamete cilia, and of both 120-and 75-kDa proGATI products on mating ectosomes, suggested that 314 proteolytic processing occurs on the ciliary and/or ectosomal surface or during the sorting and transit of 315 the precursor from cilia into nascent ectosomes. Mating ectosomes contain two subtilisin-like proteases, 316 SUB14 and VLE1; they are the closest Chlamydomonas homologs of mammalian prohormone convertases 317 PC2 and PCSK7, respectively (Luxmi et al., 2019). To address the ciliary distribution of these putative 318 prohormone convertases, we performed comparative proteomics of cilia from vegetative and gametic 319 cells of both mating types. This confirmed the presence of VLE1 in vegetative cilia of both mating types 320 (Kubo et al., 2009); SUB14 was not detectable in vegetative cilia ( Fig.7A and Supplemental Data File 1). 321 Strikingly, VLE1 was identified in the cilia of plus gametes, but was not detected in minus gamete cilia. 322 SUB14 expression was also mating type specific, but it was present in the cilia of minus, but not plus, 323 gametes (Fig.7A). In consequence, VLE1 is the only putative prohormone convertase present in ciliary 324 samples that contain proteolytically processed proGATI products. 325 Peptides from the cytosolic, pro-, S8 and C-terminal domains of VLE1 (Fig. 7B) were identified in cilia from 326 vegetative and plus gamete cells. In contrast, mating ectosomes and the secretome contained only 327 peptides from the S8 and C-terminal domains (Fig. 7B). Activation of subtilisin-like prohormone 328 convertases generally requires autoproteolytic cleavage and subsequent dissociation of the pro-domain 329 (Shakya and Lindberg, 2020). Clustal analysis identified the -Gly-Arg-Arg site that immediately precedes 330 the catalytic domain as the likely site for autoactivation. Autoproteolytic cleavage at this site, followed 331 by exoproteolytic removal of the two Arg residues would produce an amidation site. Mass spectrometry 332 revealed that all of the ciliary VLE1 had been proteolytically processed at this site and was amidated 333 (Fig.7B); partially processed peptides derived from this region of VLE1 and ending in -Gly, -Gly-Arg or -334 Gly-Arg-Arg were not observed. Detailed analysis of the predicted VLE1 structure (Fig. 7C) and peptides 335 identified in cilia suggests that the pro-domain remains associated with the S8 domain, tethering it to the 336 ciliary membrane even after autoproteolytic cleavage and amidation. 337

338
Identification of an amidated peptide that has a mating-type specific effect on C. reinhardtii mobility led 339 us to explore the properties of its putative precursor, the manner in which this precursor might be 340 converted into smaller products, and the regulated secretion of its product peptides. With the endoproteolytic cleavage of proGATI limited to the surface of mating ectosomes (Fig. 5) 3 corresponds precisely to the 75-kDa C-ter fragment (Fig. 3E). N-glycosylation of the two potential sites 365 in domain 3, along with O-glycosylation of a hydroxy-Pro located nine residues from the amidated C-366 terminus likely accounts for the ~50 kDa discrepancy between its apparent molecular mass and the mass 367 of its polypeptide chain (23 kDa) (Figs 3 and 8A). The endoproteolytic cleavage that produces 75-kDa C-368 ter fragment in ectosomes would also produce the 120-kDa N-ter fragment. Although the C-terminus of 369 proGATI is accessible to PAM, converting its C-terminus from -GATI-Gly to -GATI-NH 2, the amidated C-370 terminus is trypsin resistant (Fig. 2) and stable when exposed on the ciliary membrane and the surface of 371 mating ectosomes. 372 Examination of the first C. reinhardtii protein known to serve as a peptide precursor indicates that it shares 373 many similarities with vertebrate peptide precursors. However, its larger size, more complex domain 374 organization and extensive modifications suggest that this precursor carries additional information 375 needed to ensure that its signaling task can be accomplished. 376 Controlling the endoproteolytic cleavage of proGATI. ProGATI cleavage is linked to both mating type and 377 subcellular location (Fig. 8B). The cell bodies of plus and minus gametes contain intact proGATI, but the 378 75-kDa C-ter fragment is found only in the cilia of plus gametes. Both N-and C-ter proGATI fragments 379 accumulate in mating ectosomes. In metazoans, the cell type-specific cleavage of propeptides such as vegetative ectosomes provides access to the mother cell wall, which it degrades, allowing release of 386 mitotic progeny. A matrix metalloproteinase (gametolysin), not VLE1, cleaves the gametic cell wall prior 387 to fusion (Kinoshita et al., 1992), suggesting that VLE1 has additional targets on plus gamete cilia and/or 388 in the extracellular milieu. VLE1 cleaves to the C-terminal side of basic residues, although the required 389 sequence context is poorly understood (Matsuda et al., 1995). Endoproteolytic cleavage of proGATI after 390 a basic residue within a furin-like cleavage site (R 693 FSR↓) produces the amidated 75 kDa C-ter fragment 391 (Figs. 1F and 8A). 392 In metazoans, luminal pH plays a central role in controlling prohormone convertase activation and the 393 storage of product peptides in secretory granules (Halban, 1991 (Fig. 7C) (Shakya and Lindberg, 2020). The VLE1 pro-domain consists of an α/β fold that makes 407 extensive contact with one face of the S8 domain. Emanating from this α/β region is an extended strand 408 that arches over the active site, occluding it; the -Gly-Arg-Arg cleavage/amidation site is exposed on the 409 surface. Given the large surface area buried by the pro-domain, cleavage at the -Gly-Arg-Arg site seems 410 unlikely to result in pro-domain release from the catalytic core. 411 For amidation to occur, the extended strand must swing away from the catalytic site, enabling 412 carboxypeptidases to remove remaining Arg residue(s) and allowing PAM to access the exposed Gly The entry of ciliary proteins into ectosomes is also regulated. Differences in the ectosomal trafficking of 429 PAM, VLE1 and proGATI illustrate key features of this regulatory step (Fig. 8B). CrPAM is found in mating 430 ectosomes but not in vegetative ectosomes (Luxmi et  is linked to its release in mating ectosomes, where both N-ter and C-ter fragments accumulate. Although 437 metazoan secretory granules generally store mature product peptides, the cleavage of proatrial 438 natriuretic factor by corin, a type II plasma membrane enzyme like VLE1, is tied to the exocytosis of atrial 439 granules (Glembotski et al., 1988).  Table   465 466

Preparation of ectosomes, cilia and cell lysates from mating gametes 474
Gametes of both mating types were resuspended in 10 ml of fresh nitrogen-free M-N medium at a density 475 of 5×10 6 cells/ml. An equal number of mating type minus and plus gametes were mixed for 1 h; after 476 incubation, ectosomes were isolated by differential centrifugation as described previously (Luxmi et  allowed to adhere to 0.1% polyethyleneimine-coated coverslips for 10 min and then treated with methanol 510 for 10 min at -20˚C. Subsequent blocking and antibody incubation were done as described (Luxmi et al.,511 2019). Primary antibodies used were affinity-purified rabbit N-ter and C-ter proGATI antibodies (from 512 CT327; 1:500) and mouse FMG1 (1:1000). Alexa 488 anti-rabbit (Life Technologies, Thermo Fisher 513 Scientific) (1:500) and Cy3 anti-mouse (Jackson ImmunoResearch Laboratories, West Grove, PA) (1:2000) 514 conjugates were used as secondary antibodies. Images were obtained using a Zeiss 880 confocal 515 microscope with a 63× oil objective. 516

Electron microscopy analysis 517
Immuno-gold labeling of ectosomes was performed as described previously (Luxmi et  in serum-free medium at 37°C with 5% CO2. Cell lysates were prepared in 1x SDS lysis buffer with 1x 544 protease inhibitor cocktail (Sigma, # P8340) and 0.3 mg/ml PMSF. Soluble fractions (equal protein) were 545 analyzed using standard electrophoretic and immunoblotting techniques. 546

HEK-proGATI purification 547
Stably transfected HEK-293 cells expressing preproGATI (HEK-GATI cells) were washed and cultured in 548 serum-free media lacking ITS and BSA for 16-18 h. Spent medium was collected and centrifuged at 100 xg 549 to remove cell debris. Protease inhibitor cocktail and 0.3mg/ml PMSF were added to the medium, which 550 was stored at -80°C. Spent medium pooled from multiple sequential collections was used for purification. 551 A weak anion exchange column, HiTrap ANX Sepharose FF (Cytiva # 17-5163-01; Sigma), was used to 552 concentrate the HEK-proGATI (pI = 6.04). Prior to sample loading, the pH of the spent medium was 553 adjusted to 7.5 and the sample centrifuged at 10,000 xg for 15 min to remove any insoluble material. The 554 HiTrap ANX Sepharose FF column was washed with water, and equilibrated with 20 mM Tris, pH 7.5 555 containing 100 mM NaCl and 5% glycerol. The sample was loaded with a peristaltic pump and the flow-556 through discarded. The column was washed with 20 mM Tris, pH 7.5 buffer containing 100 mM NaCl and 557 5% Glycerol until the phenol red from the spent medium was no longer visible. Proteins were then eluted 558 using an AKTA Purifier 10 FPLC System (GE Healthcare, Fairfield, CT), with a gradient of 100 mM to 1 M 559 NaCl in 20mM Tris buffer containing 5% glycerol, a flow rate of 1 ml/min and a total elution volume of 40 560 ml. The collected fractions were analyzed using 4-15% SDS-PAGE gels, immunoblotted and probed with 561 the C-ter antibody. Peak fractions were pooled and further purified by gel filtration using a Superdex 200 562 Increase 10/300 GL (GE Healthcare, 28-9909-44) column equilibrated with 20 mM HEPES, pH 7.4 563 containing 0.5 M NaCl (Fig. S2C). Fractions were pooled based on SDS-PAGE analysis; purified HEK-proGATI 564 (~5 µg) was then analyzed by mass spectrometry (see below). Approximately 5 mg of HEK-proGATI was 565 purified from 500 ml of spent medium. 566

BCS treatment of HEK-GATI cells 567
HEK-GATI cells plated into 24 well dishes were washed and incubated for 30 min in serum-free media, at 568 37°C with 5% CO2. Cells were then treated with serum-free media containing 50 µM bathocuproine 569 disulfonic acid (BCS, Sigma) as described by (Bonnemaison et al., 2015). Cells treated with medium only 570 were used as control. Spent medium was collected and centrifuged at 100 xg to remove cell debris. Cell 571 lysates (15 µg, ~20% of total) and spent media (15 µl, 5% of total) were fractionated in 4-15% SDS-PAGE 572 gels and analyzed by immunoblotting. 573

Antibody generation 574
Synthetic peptides (BioMatik, Kitchener, Ontario, Canada) from the N-terminal (YELGLDIDGKPAHPAAT-NH2, 575 1.5 mg) and C-terminal (YAPGTGGGATI-NH2, 1.5 mg) regions of proGATI were individually conjugated to 576 keyhole limpet hemocyanin (KLH; 3 mg; Sigma H-7017, Lot 110K4833). An additional Cys residue added to 577 the N-ter peptide allowed conjugation to KLH using m-maleimidobenzoyl-N-hydroxysuccinimide ester. KLH 578 conjugation of the C-ter peptide used glutaraldehyde, facilitating the generation of amide specific 579 antibodies. Three rabbits (CT327, CT330, and CT332) were immunized with a mixture of KLH-conjugated 580 N-ter and C-ter peptides by Covance Immunology Services (Denver, PA). Crude IgG was obtained by 581 ammonium sulfate precipitation from the sera of immunized rabbits and N-ter and C-ter antibodies further 582 purified by peptide affinity chromatography. The N-ter (pI 5.5) and C-ter (pI 9.9) peptides were conjugated 583 to Affi-Gel-10 (Bio-rad) agarose beads for affinity-purification. Recoveries during affinity purification and 584 cross-reactivity of purified antibodies were examined using solid phase assays. High-affinity binding 96-585 well plates coated with N-ter (5 ng) or C-ter (5 ng) peptide were prepared and serial 3-fold dilutions of each 586 sample were tested. 587

Immunoprecipitation 599
Immunoprecipitation was performed using with slight modifications of previous protocols (Miller et al.,600 2017). Cross-reactive proteins were immunoprecipitated from plus gametic cell lysates and from mating 601 ectosomes using affinity-purified C-ter antibody. Before immunoprecipitation, samples were denatured. 602 An equal volume of 1× SDS-P buffer (50 mM Tris pH 7.6, 1% SDS, 130 mM NaCl, 5 mM EDTA, 50 mM NaF, 603 10 mM NaPPi) containing 0.3 mg/ml PMSF, protease inhibitor cocktail and PhosStop (Roche) was added 604 to the TMT cell lysate (1 mg protein) or mating ectosomes (1 mg protein) and samples were heated at 605 55°C for 5 min. Samples were allowed to cool and incubated with 0.5 volume (for cell lysate) or 1.0 volume 606 (for ectosomes) of 15% NP-40 for 20 min on ice. Samples were then diluted with 5 volumes of TES-607 mannitol (TM) buffer containing protease inhibitor cocktail, 0.3 mg/ml PMSF and PhosStop. Each sample 608 was centrifuged at 15,000 xg for 15 min at 4°C to remove any insoluble material. For pre-clearing, washed 609 Protein A agarose beads (50 µl) (Thermo Fisher Scientific, #22810) were added, samples were tumbled for 610 30 min at 4°C and then centrifuged at 100 xg for 3 min. Affinity-purified C-terminal antibody (100 µl) was 611 then added to the pre-cleared supernatants, followed by Protein A beads (50 μl) that had been washed 612 with 1x TMT buffer containing 1x protease inhibitor cocktail, 0.3 mg/ml PMSF and 1x Phos Stop. After 613 overnight incubation at 4°C, beads were pelleted and the unbound fraction saved; beads were then 614 washed once with TMT buffer containing 0.5M NaCl and twice with TM buffer containing protease 615 inhibitor cocktail, 0.3 mg/ml PMSF and Phos Stop. Bound protein was eluted by boiling in 2x Laemmli 616 sample buffer (Bio-rad) and particulate material removed by centrifugation at 15,000 xg at room 617 temperature. The input (15 µg) and eluted proteins (~2% of IPT) were fractionated in 4−15 % SDS-PAGE 618 gels (Bio-rad) and analyzed by immunoblotting. For mass spectrometry, samples were fractionated by 619 SDS-PAGE and visualized using QC colloidal Coomassie stain (Bio-rad); the 75-kDa band was excised from 620 the plus gamete cell lysate and mating ectosome samples. 621

Mass spectrometry 622
Excised gel bands were destained using 40% ethanol and 10% acetic acid in water, equilibrated to pH 8 in 623 100 mM ammonium bicarbonate, reduced by incubation with 10 mM dithiothreitol in 100 mM ammonium 624 bicarbonate (1 hr at 37°C) and alkylated by incubation with 55 mM iodoacetamide in 100 mM ammonium 625 bicarbonate (45 min at 37°C in the dark). Gel bands were dehydrated using acetonitrile, dried, and then 626 rehydrated in a 12.5 ng/µL trypsin solution (Promega porcine sequencing grade trypsin) in 100 mM 627 ammonium bicarbonate. Proteolysis proceeded for 16 hr at 37°C. Tryptic peptides were extracted using 628 alternating washes with 100 mM ammonium bicarbonate and 5% formic acid in 50% acetonitrile and a 629 final wash cycle with 100 mM ammonium bicarbonate and 100% acetonitrile. Peptide solutions were 630 pooled, dried and peptides resuspended in 0.1% formic acid in water prior to mass spectrometry analysis. 631 Purified HEK-proGATI was diluted with 100 mM ammonium bicarbonate in water and subjected to 632 reduction and alkylation using 5 mM dithiothreitol in 100 mM ammonium bicarbonate (1.5 hr at 37°C) 633 and 10 mM iodoacetamide in 100 mM ammonium bicarbonate (45 min at 37°C in the dark), respectively. 634 Promega sequencing grade trypsin was added (1:20 w/w, enzyme:protein) and proteolysis proceeded for 635 16 hr at 37˚C. Digestion was quenched by addition of concentrated formic acid. Peptides were desalted 636 using high capacity C18 desalting spin columns (Pierce #89851; ThermoFisher). Desalted peptides were 637 dried to completion and resuspended in 0.1% formic acid in water prior to mass spectrometry analysis. 638 Resuspended peptides were analyzed using nanoflow ultra-high performance liquid chromatography 639 (UPLC) coupled to tandem mass spectrometry (MS/MS) using a Dionex Ultimate 3000 RSLCnano UPLC 640 system and Q Exactive HF mass spectrometer (ThermoFisher Scientific terminal cleavage specificity was set to "semi-specific C-ragged" at "KR" sites to identify C-terminal non-653 tryptic proteolysis sites and subsequent C-terminal peptide amidation. Peptide output option was set to 654 "automatic score cut" to allow 0-5% peptide level FDR filtering and protein FDR was set to 1%. All other 655 parameters were kept at default settings. Scaffold v4 or v5 (Proteome Software, Inc., Portland, OR) were 656 used for visualization and further analysis. 657 For comparative proteomics of VLE1 and SUB14, vegetative and gametic cilia were obtained from both 658 mating types by the dibucaine method (see above). Isolated cilia were separated into membrane/matrix 659 and axonemal fractions by extraction with 1% IGEPAL CA-630 and differential centrifugation. Samples were 660 electrophoresed in triplicate using a short SDS-PAGE gel protocol, stained with Coomassie blue and then 661 subject to tryptic digestion. Tandem

Statistics and quantification 673
For each experiment, the number of biological replicates is indicated in the Figure  independent experiments. C. The proGATI antibody generated is amidation specific. Immunoblot of 854 mating ectosomes (10 µg/lane) probed with CT327 antiserum pre-incubated with peptides having -NH2 855 (GATI-amide), -Gly (GATI-Gly) or -OH (GATI-OH) at the C-terminus. Red arrowheads indicate that the 856 signals for the 250-kDa and 75-kDa bands are almost completely blocked by GATI-NH2 peptide, attenuated 857 by GATI-Gly and unaffected by GATI-OH. Similar results were obtained in two independent experiments. 858 D. Immunoprecipitation from mating ectosomes with affinity-purified C-ter antibody. The excised 75-kDa 859 fragment (red arrow) was analyzed by mass spectrometry. E. The location of peptides identified by mass 860 spectrometry is indicated (pink boxes). A furin-like cleavage site (black arrow) precedes the most N-861 terminally located peptide identified; potential paired basic cleavage sites (green) and predicted N-862 glycosylation and O-glycosylation sites are indicated. F. The C-terminal sequence of proGATI is shown. 863 Peptides identified by mass spectrometry are in blue. The furin-like cleavage site (yellow highlight), paired 864 basic residues (green), predicted N-glycosylation sites (pink), predicted O-glycosylation sites (Pro residues 865 subject to hydroxylation; orange) and amidated C-terminus (red) are indicated. 866 approximately 10 % of total) and spent medium (Mdm; 1% of total collected over an 18 h period) of 870 Control (non-transfected) and HEK-293 cells expressing preproGATI probed with affinity-purified C-ter 871 antibody. A 170 kDa (HEK-proGATI) band was detected in both cells and spent medium while 120 kDa and 872 37 kDa bands were detected only in cells (and see Fig. S2A). Non-specific bands identified in Controls are 873 marked (blue arrow). Secretion rate and cell content are quantified in Fig. S2B. B. Analysis of C-terminal 874 amidation of HEK-proGATI. Spent medium (5%) and cell lysates (15 µg, ~20% of total) of BCS-treated cells 875 and their respective Controls were analyzed. The C-ter signal for HEK-proGATI was reduced following BCS 876 treatment (red arrow), whereas the N-ter signal was unaffected. C. The C-ter/N-ter signal ratio for HEK-877 GATI  confocal images of minus and plus resting gametes stained with the C-ter proGATI antibody (green) and 906 antibody to FMG1 (red). Inset images show single Z-planes. Plus gametes probed with antibody pre-907 incubated with the GATI-NH2 peptide exhibit reduced staining (green) in cell bodies and cilia. Similar 908 localization of proGATI in gametes was obtained in three independent experiments. Scale bar = 5 µm. 909 agarose embedded ectosome pellet isolated from 1 h -/+ mixed gametes is shown. The inset shows a 912 higher magnification image of ectosomes that had been treated with Na2CO3 to remove peripheral 913 membrane proteins; scale bar = 100 nm. The right panels show immuno-gold-EM negative stain images 914 of intact ectosomes incubated with affinity-purified N-ter or C-ter antibodies and a gold-tagged secondary 915 antibody; both epitopes localized to the ectosomal surface. Ectosomes incubated with gold-tagged 916 secondary anti-rabbit antibody alone served as a negative control; scale bars = 500 nm (main image) and 917 100 nm (inset). Images are representative of three independent experiments. B. The deciliated cell bodies 918 (CB), cilia and ectosomes (Ecto) isolated from mixed gametes were fractionated by SDS-PAGE, blotted and 919 probed with the N-ter and C-ter antibodies against proGATI. C. Graph showing the enrichment of N-ter 920 and C-ter signals for proGATI and its fragments in ectosomes. Results are the average of two independent 921 experiments; mean is ± range. Asterisks indicate a statistically significant difference between two groups 922 (*P < 0.05, **P<0.001, ***P<0.0001). D & E. Immunoblot analysis showing CrPAM and FMG1 levels in 923 cells, cell bodies, cilia and ectosomes isolated from mixed gametes. Quantification of CrPAM and FMG1 924 protein levels is shown in the graphs. Results are average of duplicates and error bars indicate the ± range 925 (where *P < 0.05, **P<0.001). 926 sequentially treated with buffers containing 1% Triton X-100 (TMT) and 0.6 M NaCl (NaCl); the resulting 929 axonemal pellet (Axo) was solubilized in 1% SDS-buffer. The sub-ciliary fractions from resting minus (G124-930 ) and plus (G125+) gametes mixed gametes and mating ectosomes (Ecto) were fractionated by SDS-PAGE, 931 blotted and probed with affinity-purified N-ter and C-ter antibodies. Equal amounts of protein (20 µg) 932 were loaded for each sample. B. Immunoblot quantification of the 250-kDa and 75-kDa C-ter products. 933 Means are average of duplicates and error bars indicate ± range, where *P<0.05, **P<0.01. C. Freshly 934 isolated mating ectosomes (Input) were washed with buffer alone (10 mM HEPES, control) or with buffer 935 containing 10 mM dithiothreitol (DTT) or 10 mM EDTA; after centrifugation, the resulting supernatants (S) 936 and pellets (P) were analyzed for the presence of proGATI (using N-ter and C-ter antibodies), PAM and 937 FMG1. Red arrowheads mark the 120-kDa, 75-kDa and 63-kDa bands. Samples loaded represent the 938 pellets and corresponding supernatants derived from an initial 15 µg of ectosomes. D. Quantification of 939 120-kDa N-ter signal (n = 4) and 75-kDa C-ter signal (n = 3); means ± SEM are shown. Asterisks indicate 940 significant differences between the groups, *P<0.05, **P<0.01, ***P<0.001. 941 domain (blue), and a large C-terminal domain (yellow) that has a tri-partite organization with each lobe 954 consisting of two anti-parallel β sheets which exhibit considerable structural similarity to the CEA1 N-955 acetylglucosamine-binding adhesin from the methylotrophic yeast Komagataella pastoris (z score = 11.8, 956 RMSD = 4.0 Å; 5A3L). The cleavage/amidation site is indicated in magenta. The right-hand panel shows a 957 ribbon diagram of the active site. Side chains of the catalytic triad residues and the Asn that stabilizes the 958 transition state are shown. The pro-domain strand that arches across the active site is indicated in green. 959 occurs as it traffics through the ER and Golgi and subsequently enters cilia and ectosomes. As preproGATI 962 enters the ER, its signal peptide (orange box) is removed by signal peptidase. The addition of N-linked 963 sugars (red) begins in the ER, as does modification of Pro to HyP by prolyl hydroxylases. As proGATI moves 964 into the Golgi complex, more complex sugars and O-linked sugars on HyP (orange) and Ser/Thr (blue) 965 residues are added, leading to the higher apparent molecular mass (~250 kDa) of proGATI. A 966 carboxypeptidase trims the three C-terminal Arg residues and generates a substrate for PAM. PAM 967 converts the -Gly extended substrate into the amidated product (GATI-NH2) in a two-step reaction and 968 releases glyoxylate as a byproduct. This 250-kDa amidated proGATI form is then moved to the ciliary 969 membrane. Once on cilia, or as it moves from cilia into nascent ectosomes, 250-kDa proGATI is cleaved 970 by a subtilisin-like endoprotease (predicted to be VLE1) to yield the 120-kDa N-terminal region, and the 971 amidated 75-kDa C-terminal fragment. Cleavage of the 120-kDa product at either a second furin-like 972 cleavage site located in the linker between domains 1 and 2, or at a dibasic site at the C-terminal end of 973 domain 1, might then produce the 63-kDa N-terminal fragment and a second product containing domain 974 2 for which no probe currently exists. B. Diagram illustrating the presence and absence (red cross) of PAM, 975