sign in sign up
<<
Home , Adrenergic receptor, Biogenic amine, Biogenic amine receptor, Decarboxylation, Dimethylethanolamine, Dimethylglycine

1 Non-classical recognition evolved in a large clade of olfactory receptors

2

3 Qian Li1, Yaw Tachie-Baffour1, Zhikai Liu1, Maude W. Baldwin2, Andrew C. Kruse3, and Stephen

4 D. Liberles1*

5

6 1Department of Biology, Harvard Medical School, Boston, MA 02115, USA;

7 2Department of Organismic and Evolutionary Biology, Harvard University and the Museum of

8 Comparative Zoology, Cambridge, MA 02138, USA;

9 3Department of Biological Chemistry and Molecular , Harvard Medical School,

10 Boston, MA 02115, USA

11

12 *Correspondence to [email protected],

13 phone: (617) 432-7283, fax: (617) 432-7285

14

15 Competing interests

16 The authors declare that no competing interests exist.

17

18

19 20 Biogenic are important signaling molecules, and the structural basis for their

21 recognition by G -Coupled Receptors (GPCRs) is well understood. Amines are

22 also potent , with some activating olfactory -associated receptors

23 (TAARs). Here, we report that teleost TAARs evolved a new way to recognize amines in a

24 non-classical orientation. Chemical screens de-orphaned eleven zebrafish TAARs, with

25 including , , , 2-phenylethylamine, , and

26 . Receptors from different clades contact ligands through aspartates on

27 transmembrane α-helices III (canonical Asp3.32) or V (non-canonical Asp5.42), and diamine

28 receptors contain both aspartates. Non-classical monoamine recognition evolved in two

29 steps: an ancestral TAAR acquired Asp5.42, gaining diamine sensitivity, and subsequently

30 lost Asp3.32. Through this transformation, the fish dramatically

31 expanded its capacity to detect amines, ecologically significant aquatic odors. The

32 of a second, alternative for amine detection by olfactory receptors

33 highlights the tremendous structural versatility intrinsic to GPCRs.

34

35

2 36 INTRODUCTION

37

38 Biogenic amines are key intercellular signaling molecules that regulate and behavior.

39 In vertebrates, biogenic amines include , serotonin, , histamine, and

40 , and these biogenic amines can be detected by -Coupled Receptors

41 (GPCRs) expressed on the plasma membrane of target cells. The biochemical and structural

42 basis for recognition by GPCRs is well understood and involves an essential

43 bridge between the amino group and a highly conserved on the third

44 transmembrane α-helix of the (Asp3.32; Ballesteros-Weinstein indexing)1,2. of

45 Asp3.32 in adrenaline, serotonin, acetylcholine, histamine, and dopamine receptors eliminates

46 amine recognition2,3, and GPCRs that recognize amines through another motif have not been

47 identified.

48

49 Trace amine-associated receptors (TAARs) are a family of vertebrate GPCRs distantly related

50 to biogenic amine receptors4. There are 15 TAARs in , 6 in , and 112 in zebrafish,

51 and in these or related , all TAARs except TAAR1 function as olfactory receptors5-7. All

52 human TAARs (6/6), most mouse TAARs (13/15), and some zebrafish TAARs (27/112) retain

53 Asp3.32, suggesting that many TAARs would recognize amines. High throughput chemical

54 screens yielded ligands for TAAR18-10 and many olfactory TAARs that retain Asp3.32, including

55 TAAR3, TAAR4, TAAR5, TAAR7s, TAAR8s, TAAR9, and TAAR13c7,11,12. Each of these

56 receptors detects different amines, some of which evoke innate olfactory behaviors12-16. In

57 mouse, TAAR4 detects the aversive carnivore 2-phenylethylamine14, while TAAR5 detects

58 the attractive and sexually dimorphic mouse odor trimethylamine7,15. Knockout mice lacking

59 TAAR4 and TAAR5 lose behavioral responses to TAAR ligands identified in cell culture

60 studies13,15.

61

3 62 The selectivity of olfactory TAARs for particular amine odors suggests structural variability within

63 the ligand-binding pocket across the TAAR family. One clade of TAARs (including TAAR1,

64 TAAR3, and TAAR4) detects primary amines derived by , while a

65 second clade (including TAAR5, TAAR7s, TAAR8s, and TAAR9) prefers tertiary amines11.

66 Structural models and mutagenesis studies of TAAR1, TAAR7e, and TAAR7f predicted that

67 ligand binding occurs in the membrane plane, Asp3.32 forms a salt bridge to the ligand amino

68 group, and other transmembrane residues provide a scaffold that supports ligand binding and

69 enables selectivity11,17.

70

71 A third TAAR clade (clade III TAARs) arose in fish, and includes the majority (87/112) of

72 zebrafish TAARs6. Most clade III TAARs lack Asp3.32 (85/87), and it has been proposed that

73 these receptors may not detect amines6,18. However, ligands have not been identified for any of

74 these 85 receptors. More generally, ligands are known for only one zebrafish TAAR, TAAR13c,

75 which retains Asp3.32 and detects the carrion odor cadaverine12. Structure-activity analysis

76 revealed that TAAR13c preferentially recognizes odd-chained, medium-sized diamines, and

77 likely contains two distinct cation recognition sites within the ligand-binding pocket12. However,

78 the structural basis for diamine recognition by TAAR13c was unknown.

79

80 Here, we find that TAAR13c recognizes through two remote transmembrane

81 aspartic acids, Asp3.32 and Asp5.42. A TAAR13c variant with a charge-neutralizing mutation at

82 Asp5.42 has an altered ligand preference for monoamines rather than diamines. We propose that

83 Asp3.32 and Asp5.42 together form a general di-cation recognition motif, and note this motif to be

84 present in several zebrafish, mouse, and human TAARs. Surprisingly, Asp5.42 is retained in 84

85 out of 85 zebrafish TAARs that lack Asp3.32, raising the possibility that these receptors recognize

86 amines positioned in a non-canonical inverted orientation with the amino group pointing towards

87 transmembrane α-helix V. Using a high throughput chemical screening approach, we identified

4 88 ligands for eleven additional zebrafish TAARs, and found that different receptors detect amines

89 through salt bridge contacts at Asp3.32 and/or Asp5.42. TAAR10a, TAAR10b, TAAR12h, and

90 TAAR12i are Asp3.32-containing clade I TAARs that detect serotonin, tryptamine, 2-

91 phenylethylamine, and 3-methoxytyramine, while TAAR16c, TAAR16e, and TAAR16f are

92 Asp5.42-containing clade III TAARs that detect N-methylpiperidine, N,N-

93 dimethylcyclohexylamine, and isoamylamine. TAAR13a, TAAR13d, TAAR13e, and TAAR14d

94 have both Asp3.32 and Asp5.42, and recognize the di-cationic molecules putrescine, histamine,

95 and agmatine.

96

97 Based on these findings, we propose that a large clade of olfactory GPCRs evolved a novel way

98 to detect amines that involves a non-classical salt bridge to transmembrane α-helix V.

99 Furthermore, new TAAR ligands include several biogenic amines for which other receptors were

100 identified, including serotonin, histamine, and 2-phenylethylamine. These findings illustrate that

101 different members of the GPCR superfamily can evolve divergent ligand binding pockets for

102 recognition of the same .

103

104 RESULTS

105

106 Two remote amine recognition sites in the cadaverine receptor

107

108 To gain structural insights into diamine recognition by the cadaverine receptor TAAR13c, we

109 generated a TAAR13c model (Figure 1A) based on the X-ray crystal structure of

19 110 agonist-bound β2 receptor (Protein Data Bank Entry 4LDE) . The β2 adrenergic

111 receptor is the closest homolog (33% amino acid identity) of TAAR13c for which an active-state

112 structure is currently available. The homology model of TAAR13c indicated a canonical fold of

5 113 seven transmembrane α-helices with ligand bound at the core. Using the homology model, we

114 performed computational docking of cadaverine to the orthosteric binding site. One cadaverine

115 amino group of the docked ligand was located near Asp1123.32, which corresponds to the highly

116 conserved amine-contact site in biogenic amine receptors. The other cadaverine amino group

117 was positioned at the opposing end of the ligand-binding pocket, 3 Å away from another

118 aspartate, Asp2025.42. The second amino group was also within 5 Å of side chains from

119 Thr2035.43 and Ser2766.55, Leu192 and Phe194 on extracellular loop 2, and the backbone of

120 Ser1995.39 (additional nearby residues are depicted in Figure 1-figure supplement 1). Based on

121 its proximity and charge, Asp2025.42 was deemed a prime candidate to function as a counterion

122 that forms a salt bridge to the second cadaverine amino group. Position 5.42 in several other

123 GPCRs directly contacts ligands through bonds, salt bridges, or van der Waals

124 interactions3, and our homology model suggests a role for this amino acid in ligand recognition

125 by TAAR13c.

126

127 We used an established reporter assay7,11 to query the response properties of TAAR13c

128 variants with charge-neutralizing at Asp3.32 (TAAR13cD112A) and Asp5.42

129 (TAAR13cD202A). HEK-293 cells were co-transfected with plasmids encoding TAARs and a

130 cAMP-dependent reporter gene encoding secreted alkaline phosphatase (Cre-Seap).

131 Transfected cells were exposed to test ligands, and assayed for reporter activity using a

132 fluorescent phosphatase substrate. As previously published with this paradigm12, cadaverine

133 robustly activated TAAR13c, with a half maximal response occurring at ~10 μM. In contrast,

134 cadaverine failed to activate either the TAAR13cD112A or TAAR13cD202A mutant at any

135 concentration tested, up to 1 mM (Figure 1, Figure 1-figure supplement 1).

136

6 137 To ensure that mutation of Asp2025.42 did not impair protein folding or G protein coupling, we

138 asked whether the mutant receptor could be activated by other ligands. We tested 50 chemicals

139 for the ability to activate TAAR13cD202A, and identified the monoamine 3-methoxytyramine as an

140 agonist (Figure 1D). 3-methoxytyramine failed to activate wild type TAAR13c at any

141 concentration tested. Thus, charge-neutralizing mutation of Asp5.42 transformed TAAR13c from

142 a diamine receptor into a . The altered ligand-binding preference of the

143 TAAR13cD202A mutant suggests that Asp5.42 contributes directly to the ligand-binding pocket, and

144 homology modeling of TAAR13cD202A bound to 3-methoxytyramine (Figure 1-figure supplement

145 1) supports a direct interaction between this site and ligand. Based on modeling and

146 mutagenesis studies, we conclude that Asp5.42 in TAAR13c forms a salt bridge to a cadaverine

147 amino group.

148

149 Asp5.42 is widely conserved in clade III TAARs

150

151 We next asked whether Asp5.42 was present in other GPCRs, and might function more generally

152 in amine detection. Biogenic amine receptors use Asp3.32 to contact primary amines2,3, and

153 Asp5.42 is not present in any of 12 serotonin, 5 dopamine, 5 acetylcholine, or 9 adrenergic

154 receptors in mouse (Figure 2). Asp3.32 is also observed in all 4 mouse histamine receptors, while

155 Asp5.42 is observed in one ( H2). Histamine is di-cationic at low pH, and the

156 polar group of the histamine ligand is predicted to contact a transmembrane α-helix 5

157 asparagine in histamine receptor H1 (Asn5.46) and transmembrane α-helix 5 anions in histamine

158 receptors H2 (Asp5.42), H3 (Glu5.46), and H4 (Glu5.46)20.

159

160 Most mammalian TAARs retain Asp3.32, including 6/6 human TAARs, 13/15 mouse TAARs, and

161 15/17 TAARs. In mouse and rat, two TAAR7s have conservative changes of Asp3.32 to

7 162 Glu3.32; the effect of this substitution is unknown as ligands have not been found for Glu3.32-

163 containing TAARs. A small number of mammalian TAARs have an aspartate at position 5.43

164 including two in (TAAR6, TAAR8), mice (TAAR6, TAAR8b), and (TAAR6,

165 TAAR8a). Sequence alignments indicate that Asp5.43 of these receptors corresponds to Asp5.42 of

166 TAAR13c, and Asp5.43 may be similarly positioned in the agonist-binding pocket. These

167 mammalian TAARs are candidates to function as diamine receptors; however, ligands have not

168 been identified for these receptors, perhaps due to difficulties in achieving their functional

169 expression in heterologous cells. Mice lack a TAAR13c ortholog, yet a role for other TAARs in

170 diamine recognition is supported by the finding that mice lacking all olfactory TAARs fail to avoid

171 cadaverine odor13.

172

173 Analysis of the zebrafish TAAR repertoire surprisingly revealed that in contrast to mammalian

174 TAARs and biogenic amine receptors, the vast majority of zebrafish TAARs (91/112) have an

175 aspartate at position 5.42. Some of these receptors (7/91) also retain Asp3.32, including

176 TAAR13c, while most (84/91) have lost Asp3.32. In several other fish species such as medaka,

177 stickeback, fugu, and salmon, most TAARs also contain Asp5.42 but lack Asp3.32 (Figure 2-figure

178 supplement 1). Phylogenetic analysis indicates that TAARs containing Asp5.42 but not Asp3.32 are

179 clade III TAARs found in teleosts, and constitute a large branch of the TAAR family tree (Figure

180 2, Figure 2-figure supplement 2). Zebrafish lack clade II TAARs, and the teleost TAAR family

181 evolved differently from mammalian TAARs by forming the large cohort of clade III TAARs.

182 Several TAARs with both Asp5.42 and Asp3.32 reside between clade I TAARs and clade III TAARs

183 phylogenetically (including TAAR13s in zebrafish), suggesting that clade III TAARs derived from

184 clade I TAARs by first gaining Asp5.42 and then losing Asp3.32.

185

186 Ligands for clade III TAARs were unknown, and since they lack the canonical amine recognition

187 site, were proposed to detect chemicals other than amines6,18. However, the widespread

8 188 retention of Asp5.42 in clade III TAARs lacking Asp3.32 raised the possibility that the

189 corresponding 84 olfactory receptors in zebrafish indeed detect amines, but do so with ligands

190 positioned in a non-classical orientation.

191

192 Identifying ligands for eleven 'orphan' zebrafish TAARs

193

194 We used the high throughput reporter gene assay to identify agonists for additional zebrafish

195 TAARs (Figure 3, Figure 3-figure supplement 1). Responses of 63 zebrafish TAARs were

196 examined, including at least one representative from each TAAR subfamily in the zebrafish

197 olfactory system (TAAR10 to TAAR20). Zebrafish TAARs were expressed as fusion

198 containing the N-terminal 20 amino acids of bovine , which enhances cell surface

199 expression of some chemosensory receptors in heterologous cells21. As above, HEK-293 cells

200 or Hana3A cells22 (Hana3A cells express olfactory chaperones for promotion of receptor

201 expression and were used for studies involving TAAR10b, TAAR12i, and TAAR16e) were

202 transfected with plasmids encoding zebrafish TAARs and Cre-Seap, incubated with test

203 chemicals, and assayed for reporter phosphatase activity. TAAR ligand identification in

204 heterologous cells has been extensively validated; TAAR ligands identified by this technique

205 also activate TAAR-containing in vivo12,13,15,23, and behavioral responses to identified

206 TAAR ligands are lost in TAAR knockout mice13,15. Here, we used the heterologous expression

207 approach to identify the first ligands for zebrafish TAAR10a, TAAR10b, TAAR12h, TAAR12i,

208 TAAR13a, TAAR13d, TAAR13e, TAAR14d, TAAR16c, TAAR16e, and TAAR16f.

209

210 Ligands for clade I TAARs. TAAR10a, TAAR10b, TAAR12h, and TAAR12i are clade I TAARs

3.32 5.42 211 that retain Asp but lack Asp . TAAR10a sensitively detects serotonin (EC50 = ~0.5 μM), has

212 significantly reduced affinity for the related amines 5-methoxytryptamine (EC50 = ~20 μM)

9 213 and tryptamine (EC50 = ~70 μM), and does not detect other chemicals examined (Figure 3-figure

214 supplement 2). The highest affinity ligand identified for TAAR12h was 2-phenylethylamine (EC50

215 = ~0.3 μM), and TAAR12h also detects other alkyl amines with reduced sensitivity (Figure 3-

216 figure supplement 2). TAAR10b and TAAR12i detect the primary amines tryptamine and 3-

217 methoxytyramine respectively (Figure 3-figure supplement 1).

218

219 Serotonin and 2-phenylethylamine are both biogenic amines that can be recognized by other

220 GPCRs. Zebrafish lack an ortholog of mouse TAAR4, which detects 2-phenylethylamine14, and

221 mouse TAAR4 shares 41% identity with zebrafish TAAR12h. Likewise, zebrafish TAAR10a is

222 distantly related to mouse serotonin receptors, sharing 27-35% identity. It is thought that the

223 TAAR family derived from an ancestral duplicate of serotonin receptor subtype HTR424, and it is

224 possible that either TAAR10a retains the ligand recognition properties of the ancestral TAAR or

225 that serotonin responsiveness was lost and then re-gained in this lineage.

226

227 Ligands for putative di-cation receptors. Seven zebrafish TAARs retain both Asp3.32 and

228 Asp5.42, including all five TAAR13s and two TAAR14s (TAAR14c and TAAR14d). We previously

229 found that zebrafish TAAR13c detects the diamine cadaverine12, and now report that zebrafish

230 TAAR13a, TAAR13d, TAAR13e, and TAAR14d also detect chemicals with two cations.

231

232 TAAR13d prefers medium-chained alkyl diamines with highest affinity for putrescine (EC50 = ~1

233 μM), and progressively reduced affinity for diamines with additional groups (Figure

234 4A). TAAR13d has ~3,000-fold reduced sensitivity for diaminopropane, which is only one

235 methylene group shorter than putrescine, indicating a strict minimal agonist length. The differing

236 preference of TAAR13c and TAAR13d for cadaverine and putrescine respectively is consistent

10 237 with cross-adaptation studies indicating that separate olfactory receptors detect these two

238 diamines25.

239

240 Histamine is an agonist for both TAAR13a (EC50 = ~20 μM) and TAAR13d (EC50 = ~7 μM).

241 TAAR13d recognizes other alkyl diamines of medium length (see above), while TAAR13a is

242 selective for histamine among tested ligands. TAAR13a and TAAR13d share only 18-28%

243 identity with mouse histamine receptors, and the recognition of histamine by olfactory receptors

244 is likely due to convergent evolution within the GPCR family. Agmatine also activates multiple

245 TAARs, including TAAR13e with higher affinity (EC50 = ~1 μM) and TAAR14d with reduced

246 affinity (EC50 = ~100 μM). Agmatine is derived from by decarboxylation, reminiscent of

247 several other TAAR ligands that are biogenic amines similarly derived from natural amino

248 acids4. TAARs that detect agmatine, histamine, serotonin, 2-phenylethylamine, or isoamylamine

249 (see below) are not activated by the natural amino acids from which their ligands were derived

250 (Figure 4-figure supplement 1).

251

252 Ligands for clade III TAARs. Next, we found ligands for three clade III TAARs that lack Asp3.32.

253 Despite the absence of the classical amine recognition motif present in all known amine-

254 detecting GPCRs, zebrafish TAAR16c, TAAR16e, and TAAR16f detect amines. Zebrafish

255 TAAR16c detects N-methylpiperidine (EC50 = ~10 μM), zebrafish TAAR16e detects N,N-

256 dimethylcyclohexylamine (EC50 = ~30 μM), and zebrafish TAAR16f detects isoamylamine (EC50

257 = ~200 μM).

258

259 Removing the amino group of TAAR16c and TAAR16f ligands abolished responses (Figure 4,

260 Figure 4-figure supplement 2). TAAR16c also detected N-methylpyrrolidine (EC50 = ~70 μM),

261 and the corresponding secondary amines piperidine and pyrrolidine were partial agonists. In

11 262 contrast, -containing analogs, tetrahydropyran and tetrahydrofuran did not activate

263 TAAR16c at any concentration tested, indicating that the amine is essential for receptor

264 agonism. Likewise, TAAR16f failed to detect isoamyl , and thus TAAR16f agonism also

265 required a ligand amine. These studies indicate that clade III TAARs are in fact amine receptors,

266 and that the ligand amino group is an essential moiety recognized by the receptor.

267

268 Transmembrane aspartates are required for TAAR agonism

269

270 We showed that Asp3.32 and Asp5.42 are essential for cadaverine detection by TAAR13c (Figure

271 1), and here asked if the corresponding residues are required for activation of other fish TAARs

272 (Figure 5). Charge-neutralizing mutation of Asp3.32 in clade I TAARs, TAAR10a and TAAR12h,

273 completely eliminated responses to serotonin and 2-phenylethylamine respectively, while

274 charge-neutralizing mutation of Asp5.42 in clade III TAARs, TAAR16c and TAAR16f, completely

275 eliminated responses to N-methylpiperidine and isoamylamine respectively. Similarly, mutation

276 of either Asp3.32 or Asp5.42 in TAAR13a, TAAR13d, TAAR13e, and TAAR14d reduced or

277 abolished responses to all di-cationic ligands. The asymmetric diamines histamine and

278 agmatine could be positioned in either orientation, with the primary amine contacting either

279 Asp3.32 or Asp5.42. These mutagenesis studies reveal that clade I TAARs and clade III TAARs

280 require different transmembrane aspartates for amine recognition.

281

282 Homology models of serotonin-bound TAAR10a and N-methylpiperidine-bound TAAR16c were

283 generated (Figure 6), as described previously for cadaverine-bound TAAR13c (Figure 1A).

284 TAAR10a is a clade I TAAR and homology modeling indicated proximity of the serotonin amino

285 group to Asp1173.32. In contrast, TAAR16c is a clade III TAAR, and homology modeling

286 indicated ligand inversion with the N-methylpiperidine amino group near Asp1925.42. Inverting

287 the ligand orientation in clade III TAARs would be expected to impact other ligand-receptor

12 288 contacts; to test this model, we attempted to engineer such a ligand inversion in a canonical

289 , serotonin receptor 6 (HTR6). We analyzed the ligand preference of an

290 HTR6 double mutant (D3.32A; A5.42D) in which the canonical amine recognition site at Asp3.32 was

291 neutralized and a non-canonical amine recognition site at Asp5.42 was introduced. Ligand-

292 identification studies revealed that the HTR6 double mutant no longer recognized serotonin, but

293 instead recognized a different amine, (Figure 6-figure supplement 1). Re-orienting

294 the ligand amine towards Asp5.42 presumably alters other receptor interactions dramatically,

295 allowing Asp5.42-containing receptors to sample structurally distinct ligands. Together, these

296 studies indicate that Asp5.42 forms a non-classical amine interaction site in GPCRs, and this site

297 is present throughout a large clade of olfactory receptors.

298

299 DISCUSSION

300

301 Sensory receptors define the capacity of an organism to perceive its external environment.

302 Olfactory and receptor families can rapidly evolve, with dramatic species-specific

303 variations in repertoire size and function26,27. In some lineages, receptor families can be reduced

304 or die out, such as vomeronasal receptors in humans28,29 or sweet receptors in carnivores30,

305 while in other lineages new receptor families can be born, such as formyl receptors in

306 rodents31,32. Functional evolution of families often involves a pattern of gene

307 duplication and subsequent mutation11, although isolated functional transformations can also

308 occur, such as re-purposing in that led to a novel receptor for

309 found in nectar33.

310

311 Fish are aquatic animals, and efficient detection of water-soluble odors such as amines is

312 important for their ecological niche. Zebrafish generally use the same families of olfactory

313 receptors as other vertebrates, including odorant receptors, vomeronasal receptors, and

13 314 TAARs34. The TAAR repertoire in mice and humans is relatively small, representing 1-2% of all

315 main olfactory receptors. However, zebrafish have comparable numbers of TAARs (112) and

316 odorant receptors (143), and expansion of the teleost TAAR family suggests a relatively

317 expanded role for TAARs in fish chemosensation. However, prior to this study, ligands were

318 known for only one fish TAAR12. Moreover, the vast majority of zebrafish TAARs lack the

319 canonical amine recognition motif, and the classes of ligands they might detect were unknown.

320 Here, we show that almost all zebrafish TAARs (111/112) contain either a canonical amine-

321 detection site on transmembrane α-helix III (Asp3.32), a non-canonical amine-detection site on

322 transmembrane α-helix V (Asp5.42), or both. Thus, teleost TAARs likely represent a very large

323 family of diverse amine receptors.

324

325 Several zebrafish TAAR ligands are biogenic amines produced during decomposition35.

326 Cadaverine, histamine, agmatine, putrescine, tryptamine, 2-phenylethylamine, and

327 isoamylamine are derived in one step from the natural amino acids , , arginine,

328 , , , and leucine by microbe-mediated decarboxylation. Levels

329 of each of these amines are measured in commercially sold fish as indicators of quality and

330 freshness35; for example, excessive histamine in consumed fish causes human scombroid

331 poisoning36. While some of these biogenic amines are aversive to mammals13,14,37, they are

332 reported to evoke variable behaviors in different fish species. For example, agmatine,

333 putrescine, and cadaverine increase feeding-associated pecking behavior in goldfish25,38 while

334 cadaverine and putrescine evoke avoidance behavior in zebrafish12. Future studies are needed

335 to examine the roles of each TAAR and its ligands in fish behavior.

336

337 The evolution of divergent to the problem of amine detection by GPCRs reveals the

338 extensive landscape of change possible within the GPCR superfamily. Here, we propose a two-

14 339 step model for the functional evolution of clade III TAARs that is based on phylogenetic

340 analysis, ligand identification studies, structural modeling, and mutagenesis (Figure 6). First, an

341 ancestral clade I TAAR gained Asp5.42, and thus the ability to recognize di-cationic chemicals

342 like diamines. Ligands were identified for five zebrafish TAARs that retain Asp3.32 and Asp5.42,

343 and both aspartates are important for ligand recognition in each receptor. Second, a diamine

344 receptor lost Asp3.32, leading to clade III TAARs that display non-classical monoamine

345 recognition through a transmembrane α-helix V salt bridge to Asp5.42.

346

347 Flipping the amine orientation dramatically changes GPCR-ligand contacts, and presumably

348 allows clade III TAARs to sample a distinct repertoire of structurally divergent amine odors.

349 Once the new amine recognition motif was established, clade III TAARs dramatically expanded

350 through a pattern of gene duplication and subsequent mutation. It seems likely that the massive

351 expansion of this clade increased the capacity of the fish olfactory system to

352 detect amines, which are water-soluble and ecologically important stimuli for aquatic animals.

353

354 MATERIALS AND METHODS

355

356 Cloning of receptor . Zebrafish Taar genes were cloned from zebrafish genomic DNA

357 (AB strain), and inserted into a modified pcDNA3.1- (Invitrogen) vector containing a 5’ DNA

358 extension encoding the first 20 amino acids of bovine rhodopsin followed by a cloning linker

359 (GCGGCCGCC). Sequencing Taar genes revealed several AB strain polymorphisms as

360 compared with Tübingen strain-derived genomic sequences deposited at NCBI. Amino acid-

361 altering polymorphisms identified were as follows: Taar10a (NM_001082898): A467G (Ile to Val)

362 and G545A (Val to Ile); Taar10b (NM_001082903): A338G (His to Arg), G496A (Ala to Thr),

363 G544A (Val to Ile), and G775A (Val to Met); Taar12h (NM_001082909): A23G (Ile to Val), T26C

364 (Tyr to His), C62T (Pro to Ser), C285A (Ser to Tyr), T743A (Phe to Ile), C874T and G875A (Gly

15 365 to Ile); Taar12i (NM_001083085): T23C (Val to Ala), G313T, T314A (Val to Tyr), and T583G

366 (Phe to Val); Taar13a (NM_001083102): C427G (Thr to Arg), T680A (Asn to Lys), T803G (Ile to

367 Met), T882C (Phe to Leu), and T948C (Phe to Leu); Taar13d (NM_001083041): A112G (Ile to

368 Val), A196C (Thr to Pro), G382A (Val to Ile), and G559A (Val to Ile); Taar13e (NM_001083043):

369 A643G (Ile to Val) and G796A (Val to Ile); Taar16c (XM_009307036): A797C (Asn to Thr);

370 Taar16e (XM_009307037): T103C (Val to Ala), G711A, C712T (Ala to Ile), and C844T (Ala to

371 Val); Taar16f (XM_009305637): A23T (Asn to Ile), A59G (Asn to Ser), C190A (His to Lys),

372 A202T (Thr to Ser), and A951C (Leu to Phe). Htr6 was cloned from mouse cDNA and

373 inserted into pcDNA3.1-. Taar mutants were generated by overlap extension PCR and the Htr6

374 mutant was generated using TagMaster site-directed mutagenesis kit (GM Biosciences).

375

376 TAAR functional assays. Reporter gene assays in HEK-293 cells or Hana3A cells22 (for

377 studies involving TAAR10b, TAAR12i, and TAAR16e) were performed as previously

378 described7,11 with minor modifications. Modifications included use of tissue culture-treated 96

379 well plates (BD Biosciences) pre-incubated with polylysine (250 ng per well); transfection using

380 Lipofectamine 2000 (Invitrogen) reagent according to manufacturer's protocols, and introduction

381 of test stimuli (see below) four to six hours after transfection.

382

383 Test stimuli for high throughput chemical screens. To identify TAAR agonists, chemicals

384 were simultaneously tested in mixtures (83.3 μM per chemical, DMEM). Mix 1 contained GABA,

385 β-, 2-aminopentane, benzylamine, butylamine, and cystamine; mix 2 contained

386 dibutylamine, , N,N-dimethylcyclohexylamine, , ethyleneamine, and

387 ethanolamine; mix 3 contained hexylamine, isoamylamine, isobutylamine, ,

388 indole, and cadaverine; mix 4 contained , N-methylindole, N-methylpyrrolidine, N-

389 methylpiperidine, pyrrolidine, and putrescine; mix 5 contained hexanal, ethyl butyrate,

16 390 tryptamine, histamine, 2-phenylethylamine, and ; mix 6 contained 2-

391 methylbutylamine, cysteamine, 1-amino propan-2-ol, 3-methylthiopropylamine, and agmatine;

392 mix 7 contained , N,N-dimethylethanolamine, , N,N-dimethylglycine, 5-

393 hydroxyindole-3-acetic acid, and 3-methoxytyramine; mix 8 contained 5-methoxytryptamine, 4-

394 methoxy , N,N-dimethylphenethylamine, 5-methoxy-N,N-dimethyltryptamine,

395 N,N-dimethylaniline, N,N-dimethylisopropylamine; mix 9 contained 1-dimethylamino propanol, 2-

396 dimethylaminoethanethiol, and 1-(2-aminoethyl)-pyrrolidine; mix 10 contained 1,7-

397 diaminoheptane, 1,8-diaminooctane, 1,6-diaminohexane, 1,10-diaminodecane, 1,3-

398 diaminopropane; and mix 11 contained , , dopamine, serotonin, and

399 adenine. TAARs were tested with all 11 chemical mixtures, except TAAR14, TAAR15, and

400 TAAR16 family members which were tested with mixes 1-10.

401

402 TAAR phylogenetic analysis. Amino acid sequences of 112 zebrafish TAARs, 15 mouse

403 TAARs, 17 rat TAARs, 6 human TAARs, 12 biogenic amine receptors (zebrafish HTR2B, HRH2,

404 and DRD2A; mouse HRH3, DRD3, HTR5, DRD1A, and ADRB1; and rat HTR2A, HRH2, DRD3,

405 and ADRB2) and 5 odorant receptors (zebrafish OR103 and OR131, mouse OR121 and

406 OR446, and rat ORI15) were aligned with MAFFT v7.01739. For Figure 2-figure supplements 1

407 and 2, published TAAR amino acid sequences from opossum, cow, chicken, frog, coelacanth,

408 medaka, stickleback, tetraodon, fugu, Atlantic salmon, and elephant shark, as well as sea

409 lamprey aminergic receptors, were analyzed6,40. A phylogeny was inferred using a maximum

410 likelihood approach implemented in RAxML (RAxML v7.7.1) using the JTT + Γ model of codon

411 substitution, empirical frequencies, and the rapid bootstrapping technique (100

412 replicates)41. Nodes with support values < 50 were collapsed in the program TreeGraph 2 and

413 visualized using FigTree v1.3.142.

414

17 415 TAAR structural modeling. The TAAR13c structural model was generated using SWISS-

416 MODEL workspace43 (http://swissmodel.expasy.org/interactive) and based on the X-ray crystal

19 417 structure of agonist-bound human β2 (Protein Data Bank entry 4LDE) .

418 Modeling involved an active-state template, in which agonist binding is favored. The sequence

419 alignment for homology modeling was verified by inspection to confirm proper alignment of

420 sequence landmarks including conserved sequence motifs (DRY/DRH and NPxxY) and

421 transmembrane residues. Homology models for TAAR10a (37% amino acid identity) and

422 TAAR16c (30% amino acid identity) were prepared similarly using the same template. Models

423 of receptor point mutants were prepared by side chain substitution in previously prepared

424 homology models and then subjected to energy minimization prior to ligand docking. Docking of

425 ligands into homology models was conducted with the Schrödinger suite44 using LigPrep version

426 3.4 and Glide version 6.7. First, the receptor model was prepared for docking by adding protons

427 and assigning charges to ionizable side chains. Asp3.32 and Asp5.42 were modeled in the

428 deprotonated state expected at physiological pH. The resulting model was then subjected to

429 energy minimization using the Schrödinger impref utility prior to docking. Ligands were docked

430 to the orthosteric binding pocket using Glide Extra Precision45, and figures were prepared in

431 PyMol46.

432

433 ACKNOWLEDGMENTS

434 We thank Xianchi Dong for assistance with GPCR modeling and Sean Megason for zebrafish

435 genomic DNA. This work was supported by a grant from the NIH (SDL, Award Number R01

436 DC013289).

437

438 REFERENCES

439 1 Katritch, V., Cherezov, V. & Stevens, R. C. Structure-function of the G protein-coupled 440 receptor superfamily. Annual review of pharmacology and toxicology 53, 531-556, 441 (2013).

18 442 2 Shi, L. & Javitch, J. A. The binding site of aminergic G protein-coupled receptors: the 443 transmembrane segments and second extracellular loop. Annual review of 444 pharmacology and toxicology 42, 437-467, (2002).

445 3 Surgand, J. S., Rodrigo, J., Kellenberger, E. & Rognan, D. A chemogenomic analysis of 446 the transmembrane binding cavity of human G-protein-coupled receptors. Proteins 62, 447 509-538, (2006).

448 4 Liberles, S. D. Trace amine-associated receptors: ligands, neural circuits, and behaviors. 449 Curr Opin Neurobiol 34C, 1-7, (2015).

450 5 Horowitz, L. F., Saraiva, L. R., Kuang, D., Yoon, K. H. & Buck, L. B. Olfactory receptor 451 patterning in a higher primate. J Neurosci 34, 12241-12252, (2014).

452 6 Hussain, A., Saraiva, L. R. & Korsching, S. I. Positive Darwinian selection and the birth 453 of an olfactory receptor clade in teleosts. Proceedings of the National Academy of 454 Sciences of the United States of America 106, 4313-4318, (2009).

455 7 Liberles, S. D. & Buck, L. B. A second class of chemosensory receptors in the olfactory 456 . Nature 442, 645-650, (2006).

457 8 Borowsky, B. et al. Trace amines: identification of a family of mammalian G protein- 458 coupled receptors. Proceedings of the National Academy of Sciences of the United 459 States of America 98, 8966-8971, (2001).

460 9 Bunzow, J. R. et al. , 3,4-methylenedioxymethamphetamine, lysergic acid 461 diethylamide, and metabolites of the are agonists of a 462 rat trace amine receptor. Mol Pharmacol 60, 1181-1188, (2001).

463 10 Revel, F. G. et al. TAAR1 activation modulates , 464 preventing hyperdopaminergic and hypoglutamatergic activity. Proceedings of the 465 National Academy of Sciences of the United States of America 108, 8485-8490, (2011).

466 11 Ferrero, D. M. et al. Agonists for 13 trace amine-associated receptors provide insight 467 into the molecular basis of odor selectivity. ACS Chem Biol 7, 1184-1189, (2012).

468 12 Hussain, A. et al. High-affinity olfactory receptor for the death-associated odor 469 cadaverine. Proceedings of the National Academy of Sciences of the United States of 470 America 110, 19579-19584, (2013).

471 13 Dewan, A., Pacifico, R., Zhan, R., Rinberg, D. & Bozza, T. Non-redundant coding of 472 aversive odours in the main olfactory pathway. Nature 497, 486-489, (2013).

473 14 Ferrero, D. M. et al. Detection and avoidance of a carnivore odor by prey. Proceedings 474 of the National Academy of Sciences of the United States of America 108, 11235-11240, 475 (2011).

476 15 Li, Q. et al. Synchronous evolution of an odor pathway and behavioral 477 response. Curr Biol 23, 11-20, (2013).

19 478 16 Li, Q. & Liberles, S. D. Aversion and Attraction through Olfaction. Curr Biol 25, R120- 479 R129, (2015).

480 17 Reese, E. A. et al. Exploring the determinants of trace amine-associated receptor 1's 481 functional selectivity for the stereoisomers of amphetamine and . J 482 Med Chem 57, 378-390, (2014).

483 18 Liberles, S. D. Mammalian . Annu Rev Physiol 76, 151-175, (2014).

484 19 Ring, A. M. et al. Adrenaline-activated structure of beta2-adrenoceptor stabilized by an 485 engineered nanobody. Nature 502, 575-579, (2013).

486 20 Seifert, R. et al. Molecular and cellular analysis of human histamine receptor subtypes. 487 Trends Pharmacol Sci 34, 33-58, (2013).

488 21 Krautwurst, D., Yau, K. W. & Reed, R. R. Identification of ligands for olfactory receptors 489 by functional expression of a receptor library. Cell 95, 917-926, (1998).

490 22 Saito, H., Kubota, M., Roberts, R. W., Chi, Q. & Matsunami, H. RTP family members 491 induce functional expression of mammalian odorant receptors. Cell 119, 679-691, 492 (2004).

493 23 Zhang, J., Pacifico, R., Cawley, D., Feinstein, P. & Bozza, T. Ultrasensitive detection of 494 amines by a trace amine-associated receptor. J Neurosci 33, 3228-3239, (2013).

495 24 Lindemann, L. & Hoener, M. C. A renaissance in trace amines inspired by a novel GPCR 496 family. Trends Pharmacol Sci 26, 274-281, (2005).

497 25 Rolen, S. H., Sorensen, P. W., Mattson, D. & Caprio, J. as olfactory stimuli 498 in the goldfish Carassius auratus. J Exp Biol 206, 1683-1696, (2003).

499 26 Nei, M., Niimura, Y. & Nozawa, M. The evolution of animal chemosensory receptor gene 500 repertoires: roles of chance and necessity. Nature reviews 9, 951-963, (2008).

501 27 Shi, P. & Zhang, J. Comparative genomic analysis identifies an evolutionary shift of 502 gene repertoires in the vertebrate transition from water to land. 503 research 17, 166-174, (2007).

504 28 Liman, E. R. Use it or lose it: molecular evolution of sensory signaling in primates. 505 Pflugers Arch 453, 125-131, (2006).

506 29 Zhang, J. & Webb, D. M. Evolutionary deterioration of the vomeronasal 507 transduction pathway in catarrhine primates. Proceedings of the National Academy of 508 Sciences of the United States of America 100, 8337-8341, (2003).

509 30 Jiang, P. et al. Major taste loss in carnivorous . Proceedings of the National 510 Academy of Sciences of the United States of America 109, 4956-4961, (2012).

511 31 Liberles, S. D. et al. Formyl peptide receptors are candidate chemosensory receptors in 512 the vomeronasal . Proceedings of the National Academy of Sciences of the United 513 States of America 106, 9842-9847, (2009).

20 514 32 Riviere, S., Challet, L., Fluegge, D., Spehr, M. & Rodriguez, I. - 515 like proteins are a novel family of vomeronasal chemosensors. Nature 459, 574-577, 516 (2009).

517 33 Baldwin, M. W. et al. Sensory biology. Evolution of sweet taste in 518 hummingbirds by transformation of the ancestral receptor. Science (New York, 519 N.Y 345, 929-933, (2014).

520 34 Kermen, F., Franco, L. M., Wyatt, C. & Yaksi, E. Neural circuits mediating olfactory- 521 driven behavior in fish. Frontiers in neural circuits 7, 62, (2013).

522 35 Shalaby, A. R. Significance of biogenic amines to food safety and human health. Food 523 Research International 29, 675-690, (1996).

524 36 Morrow, J. D., Margolies, G. R., Rowland, J. & Roberts, L. J., 2nd. Evidence that 525 histamine is the causative of scombroid-fish poisoning. N Engl J Med 324, 716- 526 720, (1991).

527 37 Wisman, A. & Shrira, I. The smell of death: evidence that putrescine elicits threat 528 management mechanisms. Frontiers in psychology 6, 1274, (2015).

529 38 Hara, T. J. Feeding behaviour in some teleosts is triggered by single amino acids 530 primarily through olfaction. Journal of Fish Biology 68, 810-825, (2006).

531 39 Katoh, K., Misawa, K., Kuma, K. & Miyata, T. MAFFT: a novel method for rapid multiple 532 sequence alignment based on fast Fourier transform. Nucleic Acids Res 30, 3059-3066, 533 (2002).

534 40 Tessarolo, J. A., Tabesh, M. J., Nesbitt, M. & Davidson, W. S. Genomic organization and 535 evolution of the trace amine-associated receptor (TAAR) repertoire in Atlantic salmon 536 (Salmo salar). G3 (Bethesda) 4, 1135-1141, (2014).

537 41 Stamatakis, A., Hoover, P. & Rougemont, J. A rapid bootstrap algorithm for the RAxML 538 Web servers. Syst Biol 57, 758-771, (2008).

539 42 Stover, B. C. & Muller, K. F. TreeGraph 2: combining and visualizing evidence from 540 different phylogenetic analyses. BMC Bioinformatics 11, 7, (2010).

541 43 Biasini, M. et al. SWISS-MODEL: modelling protein tertiary and quaternary structure 542 using evolutionary information. Nucleic Acids Res 42, W252-258, (2014).

543 44 Schrodinger, L. L. C. Schrödinger Release 2015-2, (2015).

544 45 Friesner, R. A. et al. Extra precision glide: docking and scoring incorporating a model of 545 hydrophobic enclosure for protein-ligand complexes. J Med Chem 49, 6177-6196, 546 (2006).

547 46 Schrodinger, L. L. C. The PyMOL Molecular Graphics System, Version 1.3r1, (2010). 548

549

21 550 FIGURE LEGENDS

551

552 Figure 1. Diamine recognition sites in the cadaverine receptor TAAR13c. (A) Structural

553 modeling of zebrafish TAAR13c bound to cadaverine (yellow) reveals two aspartates (D1123.32

554 and D2025.42) with carboxylates (red) predicted to form salt bridges to ligand amino groups

555 (blue). (B) Cartoon of zebrafish TAAR13c topology depicts the location of D1123.32 and D2025.42.

556 (C) Charge-neutralizing mutation of D1123.32 and D2025.42 abrogates cadaverine responsiveness

557 of zebrafish TAAR13c expressed in HEK-293 cells. (D) D202A5.42 mutation transforms TAAR13c

558 from a diamine receptor into a monoamine receptor (mean ± sem, n = 3, **p < .01 based on

559 two-tailed unpaired Student's t-test).

560

561 Figure 2. Asp5.42 is highly conserved in clade III TAARs. (A) Cartoon depiction indicates the

562 location of positions 3.32 and 5.42 in GPCRs. (B) Bioinformatic analysis reveals the number of

563 receptors with anions at positions 3.32 (X) or 5.42 (Y) across various biogenic receptor

564 subfamilies in mouse, as well as TAARs in mouse, rat, humans, and zebrafish. Two mouse, rat,

565 and human TAARs have Asp5.43 instead of Asp5.42, and these are included in columns marked

566 Y=Asp (*). (C) Phylogenetic analysis of the TAAR family in zebrafish (solid lines), mice (dashed

567 lines), rats (dashed lines), and humans (dashed lines); scale bar = 0.5 substitutions per site.

568 TAARs containing only Asp3.32 (blue), only Asp5.42 (red), or both Asp3.32 and Asp5.42 (green) are

569 depicted. Mammalian TAARs with Asp3.32 and Asp5.43 instead of Asp5.42 are green. Glu3.32-

570 containing rodent TAARs and the one zebrafish TAAR, TAAR19f, that lacks both Asp3.32 and

571 Asp5.42 are depicted in black.

572

573 Figure 3. Identifying the first ligands for several zebrafish TAARs. HEK-293 cells were

574 cotransfected with Cre-Seap and plasmids encoding zebrafish TAARs, incubated with test

575 chemicals (100 μM), and assayed for phosphatase activity using a fluorescent substrate (mean

22 576 ± sem, n = 3). Zebrafish TAAR10a and TAAR12h are clade I TAARs containing Asp3.32 but not

577 Asp5.42 (blue), zebrafish TAAR13a, TAAR13c, TAAR13d, and TAAR13e, and TAAR14d contain

578 both Asp5.42 and Asp3.32 (green), and zebrafish TAAR16c is a clade III TAAR containing Asp5.42

579 but not Asp3.32 (red).

580

581 Figure 4. Structure-activity studies of zebrafish TAARs. (A) Zebrafish TAAR13d displays

582 highest affinity for putrescine among tested ligands, and reduced affinity for longer or shorter

583 diamines. (B) Zebrafish TAAR16c recognizes several structurally related amines but not

584 oxygen-containing analogs (1 mM). F: N-methylpiperidine; G: piperidine; H: N-methylpyrrolidine;

585 I: pyrrolidine; J: tetrahydropyran; K: tetrahydrofuran. (C) Dose-dependent responses of

586 TAAR16c for N-methylpiperidine and tetrahydropyran. (D) Zebrafish TAAR16f recognizes

587 isoamylamine (L) but not (M) at 1 mM. (E) Dose-dependent responses of

588 TAAR16f for isoamylamine and isoamyl alcohol. (mean ± sem, n = 3).

589

590 Figure 5. Dose-dependent responses of zebrafish TAARs and charge-neutralizing TAAR

591 mutants. (A) The identities of amino acids 3.32 and 5.42 in nine 'de-orphaned' zebrafish

592 TAARs, as well as preferred ligands and corresponding EC50s, are depicted. (B) Dose-

593 dependent activation of TAARs and TAAR mutants by ligands indicated (mean ± sem, n = 3).

594 Responses are shown in cells transfected with Cre-Seap alone (black) or together with wild type

595 TAARs (red), D3.32A mutant TAARs (blue), and D5.42A mutant TAARs (green). D3.32 is lacking in

596 clade III TAARs (TAAR16c, TAAR16f) and D5.42 is lacking in clade I TAARs (TAAR10a,

597 TAAR12h), so the corresponding D3.32A and D5.42A mutants could not be generated.

598

599 Figure 6. Modeling the structure and evolution of non-classical amine recognition. (A)

600 Structural modeling of zebrafish TAAR10a and TAAR16c bound to serotonin and N-

23 601 methylpiperidine (yellow) respectively. (B) A model for the birth of a large clade of olfactory

602 receptors with non-classical amine recognition. We propose that clade III TAARs evolved a non-

603 classical mode of amine recognition in two steps. First, an ancestral TAAR gained the ability to

604 recognize diamines by acquiring Asp5.42. Subsequently, a diamine-detecting TAAR lost the

605 canonical amine recognition site, Asp3.32, leading to non-classical amine recognition through a

606 transmembrane α-helix V salt bridge. Extensive gene duplication and mutation expanded and

607 diversified clade III TAARs, leading to a large clade of olfactory receptors with non-canonical

608 amine-detection properties.

609

610 Supplemental Figures

611

612 Figure 1-figure supplement 1. Functional analysis and structural modeling of TAAR13c

613 mutants. (A) Structural modeling of TAAR13c bound to cadaverine (yellow) reveals residues

614 within 5 Å of the amino group pointing towards transmembrane α-helix 5. (B) Charge-

615 neutralizing mutation of Asp3.32 eliminates cadaverine responsiveness in TAAR13c. HEK-293

616 cells were co-transfected with Cre-Seap and the D112A3.32 mutant of TAAR13c, incubated with

617 cadaverine or 3-methoxytyramine at various concentrations, and assayed for phosphatase

618 activity (mean ± sem, n = 3). (C) Structural modeling of wild type TAAR13c bound to cadaverine

619 (yellow) and the TAAR13c D202A5.42 mutant bound to 3-methoxytyramine (yellow).

620

621 Figure 2-figure supplement 1. The phylogenetic tree of tetrapod and fish TAARs. (A) As in

622 Figure 2, bioinformatic analysis reveals the number of receptors with anions at positions 3.32

623 (X) or 5.42 (Y). (B) Phylogenetic analysis of the TAAR family in fish (solid lines) and tetrapods

624 (dashed lines); scale bar = 0.8 substitutions per site. TAARs containing only Asp3.32 (blue), only

625 Asp5.42 (red), both Asp3.32 and Asp5.42 (green), and neither (black) are depicted.

24 626

627 Figure 2-figure supplement 2. Phylogenetic analysis of TAARs from different fish

628 species. Using the phylogenetic tree in Figure 2- figure supplement 1B as a template, TAARs

629 are highlighted from individual fish species indicated (solid, color), as well as from zebrafish and

630 tetrapods (dashed, grey); scale bar = 0.8 substitutions per site. TAARs containing only Asp3.32

631 (blue), only Asp5.42 (red), both Asp3.32 and Asp5.42 (green), or neither (black) are depicted.

632

633 Figure 3-figure supplement 1. Functional expression of TAAR10b, TAAR12i, and

634 TAAR16e in Hana3A cells. Hana3A cells were co-transfected with Cre-Seap and TAAR-

635 encoding plasmids, incubated with ligands, and assayed for phosphatase activity (mean ± sem,

636 n = 6).

637

638 Figure 3-figure supplement 2. Structure-function studies of zebrafish clade I TAARs:

639 TAAR10a and TAAR12h. HEK-293 cells were co-transfected with Cre-Seap and TAAR-

640 encoding plasmids, incubated with ligands, and assayed for phosphatase activity.

641

642 Figure 4-figure supplement 1. Several TAARs detect amines but not amino acids. HEK-

643 293 cells were co-transfected with Cre-Seap and TAAR-encoding plasmids, incubated with

644 ligands, and assayed for phosphatase activity (mean ± sem, n = 3).

645

646 Figure 4-figure supplement 2. Structure-function studies of the zebrafish clade III TAAR,

647 TAAR16c. HEK-293 cells were co-transfected with Cre-Seap and plasmid encoding TAAR16c,

648 incubated with ligands, and assayed for phosphatase activity (mean ± sem, n = 3 - 6).

649

25 650 Figure 6-figure supplement 1. Re-engineering the amine contact site of HTR6. HEK-293

651 cells were co-transfected with Cre-Seap and plasmids encoding HTR6 or an HTR6 double

652 mutant (D3.32A; A5.42D), incubated with ligands (0, 10 or 100 μM), and assayed for phosphatase

653 activity (mean ± sem, n = 3, *p < .05, **p < .01 based on one-way ANOVA analysis and

654 Dunnett’s test to compare controls and ligand-activated responses).

26 A B TAAR13c D1123.32 D2025.42

TM7 TM6

D D

Cadaverine TM5 C 200 ** No ligand 5.42 1 mM cadaverine Asp1123.32 Asp202

TM3 TM4

TM2 Receptor activity

0 No receptor TAAR13c D112A3.32 D202A5.42 D Wild type D202A5.42 200

H2NNH2 Cadaverine

0 0 0.1 0.3 1 3 10 30 100 300 1,000 0 0.1 0.3 1 3 10 30 100 300 1,000 Concentration (+M) Concentration (+M)

120

O NH2

HO 3-methoxytyramine Receptor activity Receptor activity

0 0 0.1 0.3 1 3 10 30 100 300 1,000 0 0.1 0.3 1 3 10 30 100 300 1,000 Concentration (+M) Concentration (+M)

Figure 1 A

Y X=3.32 X TM3 TM5 Y=5.42

B Number X=Asp GPCR Species X=Asp X=Glu Y=Asp Other of receptors Y=Asp

Serotonin Mouse 12 12 0 0 0 0

Dopamine Mouse 505000

Histamine Mouse 4 401 1 0

Adrenergic Mouse 909000

Acetylcholine Mouse 550 000

TAARs Mouse 15 13 2 2* 2* 0

TAARs Rat 17 15 2 2* 2* 0

TAARs Human 6 6 0 2* 2* 0

TAARs Zebrafish 11227 0 7 91 1 C

19 20 12 Clade I 18 Clade III 17 2 3 4 11 1 10

16 1

15 outgroups

14 13

Asp3.32 Asp5.42 Blue + 9 5 Red + 6 8 Green Mammalian 7 + +

Zebrafish Clade II 0.5

Figure 2 1 2 OH 3

H2N No test chemical H2N

N H Serotonin Phenylethylamine

H2N 4 5 6 H N NH H2N NH2 2 NH2 N Histamine Cadaverine Putrescine

NH 7 8 9 H2N H2N N NH2 N H Agmatine N-methylpiperidine Isoamylamine

No receptor TAAR10a TAAR12h 150 150 800 Receptor activity Receptor activity Receptor activity 0 123456789 0 123456789 0 123456789

60 TAAR13a 800 TAAR13c 800 TAAR13d Receptor activity Receptor activity Receptor activity 0 0 0 123456789 123456789 123456789

800 TAAR13e 150 TAAR14d 80 TAAR16c Receptor activity Receptor activity Receptor activity 0 0 123456789 123456789 0 123456789

Figure 3 A 200 TAAR13d H2N NH2 Diaminopropane (A) A B C B H2N Putrescine (B) NH2 D E C H2N NH2 Cadaverine (C) D

Receptor activity E H2N Diaminohexane (D) NH2 A H N NH 2 2 Diaminoheptane (E) 0 00.1 0.3 1 3 10 30 100 300 1,000 Ligand concentration (+M) B TAAR16c C TAAR16c FGH I JK 200 F 200 J N N N N O O H H Receptor activity Receptor activity

0 01,0000.1 0.3 1 3 10 30 100 300 0 Ligand concentration (+M) -FGHIJK DE TAAR16f TAAR16f 100 L 100 L M NH2

M Receptor activity Receptor activity

OH

0 0 -LM 03,0000.3 1 3 10 30 100 300 1,000 Ligand concentration (+M)

Figure 4 A X=3.32 Y=5.42 Ligand EC50 (+M) TAAR10a Asp Phe Serotonin 0.5 TAAR12h Asp Cys Phenylethylamine 0.3 TAAR13a Asp Asp Histamine 20 TAAR13c Asp Asp Cadaverine 10 Y TAAR13d Asp Asp Putrescine 1 X TM3 TM5 TAAR13e Asp Asp Agmatine 1 TAAR14d Asp Asp Agmatine 100 TAAR16c Leu Asp N-methylpiperidine 10 TAAR16f LeuAsp Isoamylamine 200 B No receptor Wild type D3.32A D5.42A

400 TAAR10a 1,000 TAAR12h OH H2N

H2N Canonical monoamine N H Receptor activity Receptor activity

0 0 00.01 0.03 0.1 0.3 1 3 10 30 100 01000.01 0.03 0.1 0.3 1 3 10 30 Serotonin concentration (+M) Phenylethylamine concentration (+M) TAAR13a TAAR13d 300 1,000 H2N H2N NH2 NH N

Diamine Receptor activity Receptor activity

0 0 0 0.1 0.3 1 3 10 30 100 300 1,000 01,0000.1 0.3 1 3 10 30 100 300 Histamine concentration (+M) Putrescine concentration (+M) TAAR13e TAAR14d 600 300 NH NH

H2N H2N N NH2 N NH2 H H Diamine Receptor activity Receptor activity

0 0 00.01 0.03 0.1 0.3 1 3 10 30 100 0 0.1 0.3 1 3 10 30 100 300 1,000 Agmatine concentration (+M) Agmatine concentration (+M)

200 TAAR16c 100 TAAR16f H2N

N Inverted monoamine Receptor activity Receptor activity

0 0 0 0.1 0.3 1 3 10 30 100 300 1,000 0 0.3 1 3 10 30 100 300 1,000 3,000 N-methylpiperidine concentration (+M) Isoamylamine concentration (+M) Figure 5 A Canonical monoamine receptor Inverted monoamine receptor TAAR10a TAAR16c TM3 TM3 D1173.32 F2085.42 N-methylpiperidine L1073.32

D1925.42 Serotonin

TM5 TM5

Y296

F270 Y275 TM7 TM6 Y255 F269 W266 Y254 TM7 TM6 W251 B

H3N R +

3.32 5.42 Canonical monoamine D O-

X TM3 O TM5

+ Asp5.42

H3NNH( )x 3 ++

3.32 5.42

O

Diamine D O- - D

O TM3 O TM5

- Asp3.32

R NH3 +

3.32 5.42 Inverted monoamine - O D

X TM3 O TM5

Figure 6