Binding Characteristics of Two Oxytocin Variants and Vasopressin at Oxytocin Receptors from Four Primate Species with Different Social Behavior Patterns

Abstract

A clade of New World monkeys (NWMs) exhibits considerable diversity in both oxytocin (OT) ligand and oxytocin receptor (OTR) structure. Most notable is the variant Pro⁸-OT, with proline instead of leucine at the eighth position, resulting in a rigid bend in the peptide backbone. A higher proportion of species that express Pro⁸-OT also engage in biparental care and social monogamy. When marmosets (genus Callithrix), a biparental and monogamous Pro⁸-OT NWM species, are administered the ancestral Leu⁸-OT, there is no change in social behavior compared with saline treatment. However, when Pro⁸-OT is administered, marmosets’ sociosexual and prosocial behaviors are altered. The studies here tested the hypothesis that OTR binding affinities and OT-induced intracellular Ca²⁺ potencies would favor the native OT ligand in OTRs from four primate species, each representing a unique combination of ancestral lineage, breeding system, and native OT ligand: humans (Leu⁸-OT, monogamous, apes), macaques (Leu⁸-OT, nonmonogamous, Old World monkey), marmosets (Pro⁸-OT, monogamous, NWM), and titi monkeys (Leu⁸-OT, monogamous, NWM). OTRs were expressed in immortalized Chinese hamster ovary cells and tested for intact-cell binding affinities for Pro⁸-OT, Leu⁸-OT, and arginine vasopressin (AVP), as well as intracellular Ca²⁺ signaling after stimulation with Pro⁸-OT, Leu⁸-OT, and AVP. Contrary to our hypothesis, Pro⁸-OT bound at modestly higher affinities and stimulated calcium signaling at modestly higher potencies compared with Leu⁸-OT in all four primate OTRs. Thus, differences downstream from a ligand-receptor binding event are more likely to explain the different behavioral responses to these two ligands.

Introduction

Oxytocin (OT) is a nonapeptide neurohormone that is critical for mammalian parturition, lactation, and parental behavior (Ellendorff et al., 1982; Fuchs et al., 1982; McNeilly et al., 1983; Chan et al., 1996; Lee et al., 2009). OT binds and activates its canonical G protein–coupled receptor, the oxytocin receptor (OTR). Synthetic OT is used widely in clinical settings for inducing and accelerating labor. Because of its ability to modulate a wide variety of social behaviors (Lee et al., 2009), OT is currently being evaluated for its clinical use in disorders with a social component, such as autism spectrum disorder and schizophrenia (Bakermans-Kranenburg and van Ijzendoorn, 2013; Feifel et al., 2016; DeMayo et al., 2017; Parker et al., 2017). Considerable effort has been invested in engineering OT analogs and formulations for potential therapeutic use and to extend understanding of OT actions, particularly central nervous system and behavioral effects (Manning et al., 2012; Busnelli et al., 2013; Muttenthaler et al., 2017). Although these efforts with novel synthetic analogs have been reasonably successful, naturally occurring variants of the OT peptide could provide an alternate route to novel agents and therapies (Gruber et al., 2012).

The nonapeptide family of hormone ligands is ancient and is present in nearly all animal lineages (Beets et al., 2013; Lockard et al., 2017). OT-like ligands generally vary at the third, fourth, or eighth amino acid position (Gruber et al., 2012), and the amino acid at the eighth position strongly affects the activity of the peptide on its target organs (Sawyer and Manning, 1973; Manning et al., 2012; Muttenthaler et al., 2017). OT and the closely related nonapeptide arginine vasopressin (AVP) differ at amino acid positions 3 and 8 and have vastly different roles in mammalian physiology, even though the affinity of OT for OTRs is only 2-fold greater than the affinity of AVP for OTRs (Manning et al., 2012). Despite having only a 2-fold lower binding affinity for OTRs, AVP is over 30-fold less potent than OT for generating OTR responses (Manning et al., 2012). Variations among species in their nonapeptide receptors correlate with variations in their respective OT-like ligands, indicating ligand-receptor coevolution (Koehbach et al., 2013). The presence of OT-like peptides across diverse animal taxa suggests the universal importance of their functions, and their coevolution with their ligands suggests a tightly aligned signaling system for these functions.

Despite variation across Animalia taxa, the OT ligand is highly conserved within eutherian mammals (Wallis, 2012). Recently, a nonsynonymous nucleotide substitution in the OXT gene coding for OT was discovered in four species of New World monkeys (NWMs), resulting in a proline at amino acid position 8 (Pro⁸-OT) in place of the typical leucine (Leu⁸-OT) (Lee et al., 2011). Subsequent screening showed that the Pro⁸-OT variant is present in at least 20 NWM species (Ren et al., 2015; Vargas-Pinilla et al., 2015). Additional OT variants were also identified, for a total of six different forms of OT in NWMs, with at least one species from each NWM clade exhibiting an OT variant (Ren et al., 2015; Vargas-Pinilla et al., 2015). OTRs also vary in NWMs, particularly in the N terminus (Ren et al., 2015; Vargas-Pinilla et al., 2015), which is important for binding to the tail of the OT ligand (Postina et al., 1996; Gimpl and Fahrenholz, 2001), and there is strong evidence for OT-OTR coevolution (Koehbach et al., 2013; Ren et al., 2015; Vargas-Pinilla et al., 2015). Moreover, OXTR variation is associated with social monogamy among primates (Ren et al., 2015), and OT ligand variation at position 8 is associated with litter size within the family Cebidae (Vargas-Pinilla et al., 2015). Both native and non-native OT ligands modulate social behavior in NWMs expressing Pro⁸-OT (French et al., 2016), but the native Pro⁸-OT is more effective at modulating behavior than the ancestral Leu⁸-OT in the marmoset, a monogamous and biparental NWM (Cavanaugh et al., 2014; Mustoe et al., 2015, 2018). Together these findings indicate that both OT ligand variation and the corresponding variations in OTRs among NWMs contribute to functional outcomes.

Based on the findings summarized above, we hypothesized that the binding affinities and signaling potencies of primate OT variants are different for different OTR variants, with each receptor variant preferring the ligand variant from the same species. The studies presented here test this hypothesis by measuring binding affinities and signaling potencies for Leu8, Pro8, and AVP at the OTRs from four primate species, each representing a unique combination of ancestral lineage, breeding system, and native OT ligand.

Materials and Methods

OTR Transfection and Cell Culture.

Chinese hamster ovary (CHO; female origin) cells were purchased from American Type Culture Collection (Manassas, VA) and cultured at 37°C with 5% CO₂ using Ham’s F-12 medium supplemented with 10% fetal bovine serum and 100 U/ml penicillin and 100 µg/ml streptomycin. Human, marmoset, and macaque OTR plasmids (Table 1) were purchased from GenScript (Piscataway, NJ) in a pcDNA3.1+ vector. The titi monkey plasmid was generated by amplifying and ligating the coding region of the titi OXTR from genomic titi monkey DNA (flanked with BamHI and XhoI restriction sites) and ligating it into a T vector (pMD19). Competent Escherichia coli were transformed using this vector, plated onto Luria-Bertani/ampicillin/isopropyl β-d-1-thiogalactopyranoside/X-gal plates, and incubated overnight at 37°C. White colonies were selected, then plasmid DNA was purified and sequenced. Sequence-confirmed plasmids were then digested with BamHI and XhoI and ligated into a pcDNA3.1+ vector. CHO cells were then transfected using TurboFect (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer’s instructions and kept under selective pressure using 400 µg/ml G418 antibiotic. Individual clonal lines were generated by plating batch-transfected cells at approximately 10 cells/ml (1 cell/100 µl) into 96-well plates and then selecting wells for screening that originated from a single colony. Clonal lines were screened using an intact cell ¹²⁵I-ornithine vasotocin analog (OVTA) binding assay and selected for similar receptor expression across species, defined as total radioligand binding. All experiments were done in a single clone per species, except for the marmoset, in which two clones were used.

TABLE 1.

Representative primate species OTRs

Group	Human	Macaque	Marmoset	Titi Monkey
Lineage (family)	Old World (Hominidae)	Old World (Cercopithidae)	New World (Callitrichidae)	New World (Pithecidae)
Breeding system	Monogamous	Polygamous	Monogamous	Monogamous
Native OT ligand	Leu8-OT	Leu8-OT	Pro8-OT	Leu8-OT

Intact Cell Saturation Binding Assays.

CHO cells expressing primate OTRs were plated at 150,000 cells/ml (15,000 cells/well per 100 µl) into 96-well plates and grown to 80%–90% confluence. On the day of assay, growth medium was aspirated and cells were quickly washed once with 100 µl ice-cold high glucose HEPES-buffered Dulbecco’s modified Eagle’s medium containing 0.1% bovine serum albumin (HGH-BSA) and then placed on ice. Then 50 µl ice-cold ¹²⁵I-OVTA (PerkinElmer, Waltham, MA) in doubling concentrations from about 15 to 2000 pM was added in triplicate (technical replicates) to all wells and incubated for 3 hours on ice. At the end of the assay, an aliquot of the binding medium was collected to quantify free radioligand directly, eliminating any concerns about differential depletion of ligand due to differential receptor expression levels. Cells were then washed four times with 100 µl ice-cold HGH-BSA, solubilized with 100 µl 0.2 N NaOH, and counted on a gamma counter. Nonspecific binding was defined as ¹²⁵I-OVTA binding occurring in the presence of excess competitor (10⁻⁴ M Leu⁸-OT). Binding affinity (K_d) for ¹²⁵I-OVTA was determined using GraphPad Prism software (GraphPad Software Inc., La Jolla, CA) to fit the specific bound versus free ligand data to a single-site binding equation. These assays were done at least three times on 3 different days using fresh aliquots of ¹²⁵I-OVTA and competitor, and K_d values were averaged across three biologic replicates (five biologic replicates for marmoset).

Intact Cell Competition Binding Assays.

CHO cells expressing primate OTRs were plated at 150,000 cells/ml (15,000 cells/well per 100 μl) into 96-well plates and grown to 80%–90% confluence. On the day of assay, growth medium was aspirated and cells were quickly washed once with 100 µl ice-cold HGH-BSA and then placed on ice. Then 50 µl of roughly 50,000 cpm ice-cold ¹²⁵I-OVTA was added in triplicate (technical replicates) to all wells in the presence or absence of 10⁻¹¹ to 10⁻⁵ M Pro⁸-OT (CYIQNCPPG-NH2; Anaspec, Fremont, CA), Leu⁸-OT (CYIQNCPLG-NH2; Anaspec), or AVP (CYFQNCPRG-NH2; Anaspec) and incubated for 3 hours on ice. At the end of the assay, an aliquot of the binding medium was collected to quantify free radioligand directly. Cells were then washed four times with 100 µl ice-cold HGH-BSA, solubilized with 100 µl 0.2 N NaOH, and counted on a gamma counter. Binding affinities (IC₅₀) were determined by plotting bound ¹²⁵I-OVTA versus competitor concentration. IC₅₀ values were then calculated using the Cheng–Prusoff equation and each receptor’s affinity for ¹²⁵I-OVTA to produce K_i values. These assays were done at least three times on 3 different days using fresh aliquots of ¹²⁵I-OVTA and Leu⁸-OT, Pro⁸-OT, and AVP for three biologic replicates per clone.

Ca²⁺ Mobilization Assays.

CHO cells expressing primate OTRs were plated into 96-well plates and grown to 80%–90% confluence. On the day of assay, growth medium was aspirated and cells were incubated at 37°C with 100 µl Fluo-4 Direct dye mixed in Fluo-4 Direct Ca²⁺ Assay Buffer (Thermo Fisher Scientific) with 5 mM probenecid for 1 hour. At the end of 1 hour, baseline fluorescence was measured at 37°C followed by stimulated fluorescence in the presence or absence of 10⁻¹¹ to 10⁻⁶ M Pro⁸-OT, Leu⁸-OT, or AVP (3 × technical replicates). Peak fluorescence minus baseline fluorescence was then plotted as a function of ligand concentration to determine EC₅₀ values. These assays were done at least three times on 3 different days using fresh aliquots of Leu⁸-OT, Pro⁸-OT, and AVP for three biologic replicates per clone.

Data Analysis.

Binding affinities (K_d) for ¹²⁵I-OVTA at each primate OTR were calculated by subtracting nonspecific binding and then plotting bound ¹²⁵I-OVTA versus free ¹²⁵I-OVTA.

Because concentrations of ¹²⁵I-OVTA were not identical from experiment to experiment, technical replicates within each experiment (n = 3) were normalized and binding affinities (K_i) were calculated using the Cheng–Prusoff equation and the measured binding affinity for ¹²⁵I-OVTA. Technical replicates were then averaged and used as biologic replicates (n = 3 per clone) to determine and compare K_i values for each ligand within species. A Bonferroni-corrected cutoff (P = 0.05 ÷ 3 = 0.0167) was used to determine statistically significant differences in K_i values.

Within-species differences in Ca²⁺ mobilization potency (EC₅₀) were determined by normalizing and averaging each technical replicate (n = 3) and then using the biologic replicates (n = 3) to assess ligand comparisons (Pro⁸-OT vs. Leu⁸-OT, Pro⁸-OT vs. AVP, and Leu⁸-OT vs. AVP). A Bonferroni-corrected cutoff (P = 0.05 ÷ 3 = 0.0167) was used to determine statistically significant differences in K_i values.

All data were analyzed using the nonlinear least-squares curve-fitting capabilities of GraphPad Prism software.

Results

Saturation Binding Assays.

Saturation assays were performed on 96-well plates with 50 µl binding medium per well. Representative saturation curves for all receptors are shown in Fig. 1. All of the binding and signaling assays for the human receptor were conducted with a single clone with a B_max value of 17 ± 6 fmol/well (n = 3). All assays for the macaque receptor were with a clone with a B_max value of 12 ± 5 fmol/well (n = 3). All assays for the titi monkey receptor were with a clone with a B_max value of 44 ± 12 fmol/well (n = 3). For the marmoset receptor, some assays were performed with a significantly higher expressing clone, R9, with a B_max value of 91 ± 20 fmol/well (n = 2); additional experiments were performed with a clone with lower expression, R10, with a B_max value of 33 ± 5 fmol/well (n = 4).

Fig. 1.

Representative saturation assays for ¹²⁵I-OVTA binding to OTRs from each of the four species. Cells on 96-well plates were incubated on ice in 50 µl binding medium with the indicated concentrations of ¹²⁵I-OVTA for 3 hours, and specific binding was then quantified. Data are from a single experiment with all four receptors tested side by side in triplicate. Values for this experiment are in good agreement with the average values in the Results and Table 1: humans, B_max = 9.7 fmol/well and K_d = 0.12 nM; macaques, B_max = 6.9 fmol/well and K_d = 0.144 nM; marmosets (R10), B_max = 34 fmol/well and K_d = 0.30 nM; and titi monkeys, B_max = 30 fmol/well and K_d = 0.24 nM.

Saturation binding analyses revealed only relatively small differences in binding affinities for the radioligand ¹²⁵I-OVTA among the four species, ranging from 161 to 481 pM (Table 2). The human and macaque OTRs exhibited very similar affinities that were somewhat higher than those for the titi monkey and marmoset, with the marmoset exhibiting the lowest affinity.

TABLE 2.

Binding affinities for ligands at various primate OTRs

Data are presented as means ± S.E.M. and means ± log S.E.M. for Kd and Ki values, respectively.

OTR	n	125I-OVTA Kd	Ki
			Pro8-OT	Leu8-OT	AVP
		nM
Human	3	0.161 ± 0.019	22.78 ± 0.10a	71.84 ± 0.10	541.1 ± 0.07a,b
Macaque	3	0.199 ± 0.036	43.36 ± 0.07	74.99 ± 0.08	474.8 ± 0.10a,b
Marmoset	5, 6	0.481 ± 0.041	81.31 ± 0.11	170.2 ± 0.10	1093 ± 0.13a,b
Titi monkey	3	0.289 ± 0.031	146.9 ± 0.11a	894.5 ± 0.12	1924 ± 0.16b

^aIndicates a significant within-species difference compared with Leu⁸-OT using a Bonferroni-corrected cutoff of P < 0.0167.

^bIndicates a significant within-species difference compared with Pro⁸-OT using a Bonferroni-corrected cutoff of P < 0.0167.

Competition Binding Assays with OT Variants and AVP.

In competition binding assays, Pro⁸-OT exhibited a higher binding affinity than Leu⁸-OT for all four species, with a 1.5-fold difference for macaques, a 2-fold difference for marmosets, a 3-fold difference for humans, and over 6-fold difference for titi monkeys. Only for titi monkeys and humans was the difference in binding affinity statistically significant [F(1, 42) > 12.1, P < 0.016]. For the human OTR, the difference was due to greater affinity for Pro⁸-OT; for the titi OTR, the difference was due to lower affinity for Leu⁸-OT, rather than higher affinity for Pro⁸-OT compared with the other species (Fig. 2; Table 2). For both OT variants, the absolute binding affinities were 3- to 5-fold higher for humans, macaques, and marmosets than for titi monkeys.

Fig. 2.

Competition curves for Pro⁸-OT and Leu⁸-OT for each primate species OTR. Increasing concentrations of competitor ligand (Pro⁸-OT, Leu⁸-OT, or AVP) were added to a constant concentration of ¹²⁵I-OVTA in intact CHO cells expressing one of four primate OTRs. All values are expressed as the percentage of the maximal binding in the absence of OT or AVP.

Binding affinities for AVP were assessed alongside the two OT variants for all of the receptors. Compared with Pro⁸-OT and Leu⁸-OT, respectively, binding affinity for AVP was 20- and 8-fold lower for humans, 11- and 6-fold lower for macaques, and 13- and 6-fold lower for marmosets, but 13- and only 2-fold lower for titi monkeys. In fact, the affinity of the titi OTR for Leu⁸-OT was not significantly higher than that for AVP [F(1, 42) = 2.07, P = 0.157]. However, the rank order of potencies was the same for all species, with affinities for Pro8 > Leu8 > AVP.

Ca²⁺ Signaling Assays.

In Ca²⁺ mobilization assays, the rank order of potencies was the same for all species and with the same pattern as for binding, Pro8 > Leu8 > AVP; however, the magnitude of the differences was smaller for signaling than for binding. Pro⁸ and Leu⁸-OT were roughly equipotent for all species, with only 1.5-fold greater potency of Pro8-OT versus Leu⁸-OT for humans, macaques, marmosets, and titi monkeys (Fig. 3; Table 3). Pro⁸-OT consistently exhibited a slightly lower maximal response than Leu8 for all species except the marmoset.

Fig. 3.

Intracellular Ca²⁺ increases for each primate species OTR. Increasing concentrations of Pro⁸,-OT, Leu⁸-OT, or AVP were used to stimulate intracellular Ca²⁺ mobilization in CHO cells expressing one of four primate OTRs. All values are expressed as the percentage of the maximal response to 10⁻⁶ M Leu⁸-OT for each species.

TABLE 3.

Ca²⁺ mobilization potencies for ligands at various primate OTRs

Data are presented as means ± log S.E.M.

OTR (n = 3)	Ca2+ EC50
	Pro8-OT	Leu8-OT	AVP
	nM
Human	0.072 ± 0.12	0.127 ± 0.10	0.864 ± 0.07a,b
Macaque	1.341 ± 0.11	2.025 ± 0.11	7.981 ± 0.12a,b
Marmoset	0.115 ± 0.15	0.176 ± 0.14	0.459 ± 0.14b
Titi monkey	0.595 ± 0.09	1.010 ± 0.09	4.821 ± 0.09a,b

^aIndicates a significant within-species difference compared with Leu⁸-OT using a Bonferroni-corrected cutoff of P < 0.0167.

^bIndicates a significant within-species difference compared with Pro⁸-OT using a Bonferroni-corrected cutoff of P < 0.0167.

Ca²⁺ mobilization potencies for AVP were assessed alongside the two OT variants for all of the receptors. Compared with Pro⁸-OT and Leu⁸-OT, respectively, potency for AVP was 12- and 7-fold lower for humans, 6- and 4-fold lower for macaques, 5- and 2-fold lower for marmosets, and 8- and 5-fold lower in titi monkeys. The absolute potencies for Pro⁸-OT and Leu⁸-OT for humans and marmosets were similar, but the potency of AVP for the marmoset receptor was nearly 2-fold higher than it was for the human receptor. Potencies for each ligand across species were higher for human and marmoset OTRs compared with the macaque and titi monkey.

Ca²⁺ mobilization potencies relative to binding affinities were also computed as a metric of coupling efficiency (Table 4). In general, efficiencies within species were similar, with signaling EC₅₀ values exhibiting potencies over 2 log units higher than the binding affinities for all three ligands. The macaque OTR was the least efficient, signaling at potencies less than 2 log units higher than the binding affinity for all three ligands. Notably, in all species except the titi monkey, AVP was equally or more efficient at mobilizing Ca²⁺ than Pro⁸-OT or Leu⁸-OT, per unit of binding affinity.

TABLE 4.

Coupling efficiencies for ligands at various primate OTRs

OTR	Potency/Affinity Ratio −Log(Ca2+ IC50/Ki)
	Pro8-OT	Leu8-OT	AVP
Human	2.51	2.75	2.80
Macaque	1.51	1.57	1.77
Marmoset	2.85	3.00	3.38
Titi monkey	2.39	2.95	2.60

Discussion

The studies here tested the hypothesis that the coevolution between Pro⁸-OT and OTRs in NWMs (Ren et al., 2015; Vargas-Pinilla et al., 2015) would confer greater selectivity in binding and signaling for Pro⁸-OT over the ancestral Leu⁸-OT at receptors from Pro⁸-OT–expressing species, and conversely higher selectivity for Leu⁸-OT at receptors from species expressing Leu⁸-OT. The binding and signaling data in this study show that this hypothesis is at best only partially supported. For the marmoset OTR, the species-native Pro⁸-OT bound with only modestly higher affinity and induced Ca²⁺ mobilization with higher potency than Leu⁸-OT. In humans and titi monkeys, the species–non-native ligand Pro⁸-OT also bound with higher affinity than the species-native ligand Leu⁸-OT. For receptors from all three Leu⁸-OT–expressing species, the two ligands were equipotent at mobilizing Ca²⁺. The higher binding affinity for Pro⁸-OT for all of the species, including those whose native hormone is Leu⁸-OT, was unexpected and not consistent with our hypothesis of binding affinities for each species correlating with their native ligand. One explanation for the observed preference for Pro⁸-OT over Leu⁸-OT in all species may be that the flexible (Kotelchuck et al., 1972; Brewster et al., 1973) tail of Leu⁸-OT can orient into a conformation that is similar to the more rigid structure of Pro⁸-OT (Lee et al., 2011) for only a smaller percentage of ligand-receptor interactions than Pro⁸-OT, and that the optimal conformation for Leu⁸-OT is one that is similar to the structure of Pro⁸-OT. The lack of significant preferences for the endogenous ligand in terms of signaling potencies was similarly unexpected. Thus, differences in other factors, perhaps downstream of the initial receptor binding and activation steps, are now the more likely explanations for the differential behavioral responses to OT in Leu⁸-OT– versus Pro⁸-OT–expressing species.

The ability of AVP to bind and activate primate OTRs was also tested, because AVP binds and activates OTRs, and AVP is known to affect social behavior in primates, including titi monkeys and marmosets (Caldwell et al., 2008; Jarcho et al., 2011; Taylor and French, 2015; Taylor et al., 2017). In all species except the titi monkey, AVP bound with much lower affinity to the OTR than Pro⁸-OT or Leu⁸-OT, and in all species AVP had lower potency for mobilizing Ca²⁺ than Pro⁸-OT or Leu8-OT. These results show that the coevolution of Pro⁸-OT and the OTR in marmosets has not altered selectivity for AVP versus the two OT variants.

To our knowledge, this study is the first description of NWM OT ligand variant binding in nonhuman primates, and the Ca²⁺ mobilization data inform the recent research investigating OTR signaling and NWM ligand variants in rodent models. Parreiras-E-Silva et al. (2017) found no differences in Ca²⁺ signaling at the human OTR between Pro⁸-OT, Leu⁸-OT, or the additional NWM OT variant Val³Pro⁸-OT. Our data also partially replicate those obtained by our collaborators (Pierce et al., 2016, unpublished observations), who found no difference in Ca²⁺ signaling between Pro⁸-OT and Leu⁸-OT at the human OTR but that Pro⁸-OT was more efficacious than Leu⁸-OT at inducing Ca²⁺ mobilization at the marmoset OTR. Taken together, these data indicate that the substitution of proline in place of leucine does not inhibit the G protein–coupled activity in species that express the ancestral Leu⁸-OT. Perhaps more importantly, the Pro⁸ substitution confers equal or greater potency for Ca²⁺ mobilization in a species in which Pro⁸-OT is the native ligand.

These binding and signaling data provide a functional link between the genetic surveys of the OT system in NWMs and the growing body of work comparing the behavioral effects of intranasal treatment with Pro⁸-OT and Leu⁸-OT in marmosets. Pro⁸-OT, but not Leu⁸-OT, enhances a variety of pairmate-directed social approach behaviors (Cavanaugh et al., 2014, 2018). Moreover, Pro⁸-OT increases the amount of social behavior that an OT-treated marmoset receives from its mate and reduces sociosexual behavior directed toward individuals other than the pairmate (Cavanaugh et al., 2014; Mustoe et al., 2015). Leu⁸-OT does affect some social behavior in marmosets, but Leu⁸-OT never enhances a social behavior that Pro⁸-OT does not also enhance (Mustoe et al., 2018). The binding and signaling data support these behavioral findings. Pro⁸-OT not only bound to the marmoset OTR with greater affinity but was also modestly more potent at stimulating Ca²⁺ mobilization. Although it is unlikely that this difference in signaling between Pro⁸-OT and Leu⁸-OT is the only contributing factor to the behavioral differences between treatment with Pro⁸-OT and Leu⁸-OT in marmosets, it is likely at least one contributing factor.

These binding and signaling data also shed new light on the clade-wise surveys of the OXT, OXTR, and AVP V1a receptor (AVPR1A) genes in NWMs. First and foremost, our data help to explain the finding that social monogamy and the OTR coevolved in NWMs (Ren et al., 2015; Vargas-Pinilla et al., 2015), with Pro⁸-OT binding and Ca²⁺signaling both enhanced in the socially monogamous species that expresses Pro⁸-OT natively. Moreover, Ca²⁺ signaling was not reduced by the substitution of proline for leucine at the eighth position in OTRs from Leu⁸-OT species. This suggests that the OTR is permissive for this substitution, providing a potential mechanism for the coevolution of Pro⁸-OT and OTRs in NWMs (Ren et al., 2015; Vargas-Pinilla et al., 2015). The single nucleotide substitution that produced Pro⁸-OT may have had modest consequences for neurotransmission, and thus the OTR may have evolved to accommodate this substitution. There is also a relationship between social monogamy and variation in the AVPR1A gene. Interestingly, in one of the only Leu⁸-OT NWMs that exhibits social monogamy (the titi monkey), Leu⁸-OT and AVP bound the OTR with similar affinity. These genetic and signaling data suggest that interrogation of the NWM AVP receptors (V1aR, V1bR, V2R) may provide new insights into nonapeptide signaling in primates, and these studies are currently in progress.

Alongside the work of Parreiras-E-Silva et al. (2017), this article constitutes a “first look” at the characteristics of the marmoset and titi monkey OTRs when bound to the Pro⁸-OT variant. As such, we only explored two facets of GPCR function: ligand binding and Ca²⁺ signaling. Other characteristics of these receptors, such as differential coupling to specific Gα subunits or bias for G proteins versus β-arrestin, are beyond the scope of this project but are nonetheless interesting future directions. The OTR is capable of coupling to a variety of Gα subunits resulting in a variety of cellular outcomes [see Gimpl and Fahrenholz (2001) and Mustoe et al. (2018) for detailed reviews], and there is a need for and a value to the production and characterization of ligands that are functionally selective at the OTR as tools to target specific signaling cascades. Indeed, even relatively small modifications to the OT ligand can alter the functional selectivity at the OTR, causing it to couple to different Gα subunits (Busnelli et al., 2012), and it is already known that Pro⁸-OT is less efficacious than Leu⁸-OT at promoting β-arrestin recruitment and internalization at the human receptor (Parreiras-E-Silva et al., 2017). It is possible that Pro⁸-OT and Leu⁸-OT may differentially promote coupling to specific Gα subunits or bias signaling via G proteins versus β-arrestin in marmosets and titi monkeys as well, and these experiments may provide more insight into the evolution of this system in NWMs. Another interesting possibility is that the OTRs may form dimers with various other GPCRs, and the Leu⁸-OT and Pro⁸-OT variants might exhibit selectivity for binding or activating one of these dimers versus another. Such dimer selectivity would not be detected in these assays with only the OTR expressed. Thus, multiple possible explanations for the ligand variation and its correlations with GPCR signaling and behavior remain to be explored.

A final potentially important outcome of these studies is that the higher binding affinity of Pro⁸-OT versus Leu⁸-OT at OTRs from all species, including humans, should presumably make Pro⁸-OT a better ligand for future binding studies, in either a radiolabeled or fluorescently tagged form. The 3-fold higher binding affinity would allow the use of 3-fold lower concentrations of the ligand to achieve the same fractional receptor occupancy, thus decreasing the amount of ligand required and the corresponding cost and usage of the ligand. The tighter binding of the Pro⁸-OT variant to the OTR may be useful in other contexts as well.

Abbreviations

AVP

arginine-8-vasopressin

CHO

Chinese hamster ovary

HGH-BSA

high glucose HEPES-buffered Dulbecco’s modified Eagle’s medium containing 0.1% bovine serum albumin

NWM

New World monkey

OT

oxytocin

OTR

oxytocin receptor

OVTA

ornithine vasotocin analog

Authorship Contributions

Participated in research design: Taylor, Schulte, French, Toews.

Conducted experiments: Taylor, Schulte.

Performed data analysis: Taylor, Schulte, Toews.

Wrote or contributed to the writing of the manuscript: Taylor, Schulte, French, Toews.