Subtle modifications to oxytocin produce ligands that retain potency and improved selectivity across species

Rationale for single atom modifications

Based on structural data of OT (beta-turn stabilized by disulfide) (4, 47, 53–55) and earlier studies (37, 49, 56, 57), it is known that OT/AVP binding and receptor activation are sensitive even to minor modifications (e.g. ring reduction by one sulfur atom or disulfide bond replacement by a dicarba bridge causes complete loss of activity (56); Gly9 to Val9 exchange causes agonist to antagonist switch (49); see further details in Supplementary Note 1 in SI). Additionally, traditional medicinal chemistry strategies that modified residues within the loop pharmacophore resulted in selectivity improvements that were not retained across species (Table 1) (37, 38). Based on these observations, we selected a range of single atom modifications of the pharmacophore framework (i.e. disulfide bridge and N/C-terminus modifications summarized in Figure 1A and SI Table S1) to test the hypothesis that subtle framework modifications can yield selectivity gains that are retained across species.

Peptide synthesis

Two building blocks [Boc-L-Sec(Meb)-OH and dSec(Meb)-OH] were synthesized to access the desired sulfur/selenium modifications (Figure 1A & B, SI Table S1). To acquire Boc-L-Sec(Meb)-OH in desirable quantities, the protocols of Chocat et al. (58), Tanaka et al. (59) and Oikawa et al. (60) were optimized with the new synthetic chemistry able to provide 28.7 g of pure Boc-L-Sec(Meb)-OH in only three steps (Figure 1B, 62% overall yield, four days). A new synthetic strategy was also devised for dSec(Meb)-OH yielding the compound in a two-step synthesis with an overall yield of 12% (Figure 1B). With these building blocks in hand, all of the OT/AVP analogues 1-16 (Figure 1A, SI Table S1) were synthesized via Boc-SPPS using the HBTU-mediated in situ neutralization chemistry (51, 61), folded in 0.1M NH₄HCO₃ and purified by RP-HPLC.

Functional response of novel OT/AVP ligands

Agonist activity at the human OT, V1a and V1b receptors was determined using a Fluorescent Imaging Plate Reader (FLIPR) Ca²⁺ mobilization assay, and activity at hV2R was determined by measuring the production of second messenger cAMP upon receptor stimulation (Figure 1C & D, SI Table S2).

OT was able to activate all four receptors, yet with EC₅₀ 10-300 fold higher at the hAVPRs (hOTR<hV2R<hV1bR<hV1aR) than at hOTR (0.52 nM). Deletion of the N-terminal amino group (dOT, 2) improved activation at all of the AVPRs. Replacement of the disulfide bond with the isosteric, yet slightly more hydrophobic diselenide bond, ([Se-Se]-OT (3), d[Se-Se]-OT (4)) was well tolerated and did not lead to significant changes in EC₅₀. Exchange of the C-terminal amide to acid in OT (OT-OH, 5) resulted in weaker potency at all receptors and truncations of the C-terminus of OT (8, 9, 10) yielded higher EC₅₀ at all receptors with complete loss of activity at hV2R (up to 10 μM).

AVP also activated all four receptors including hOTR (hV2R>hV1bR>hV1aR>hOTR). Deamination of the N-terminus (dAVP, 12) had little impact on receptor activation with EC₅₀ values similar to AVP. The disulfide to diselenide exchange ([Se-Se]-AVP (13) and d[Se-Se]-AVP (14)) was again well tolerated and did not change the potency/selectivity profile significantly. Truncations of the C-terminus of AVP (15, 16) led to significant loss in potency over all four receptors (Figure 1D, SI Table S2).

Interestingly, the combination of diselenide replacement and C-terminal amide to acid change yielded a partial (E_max=52%) agonist ([Se-Se]-OT-OH, 6) with low nanomolar potency at hOTR (7.3 nM), no activation of hV1aR (>10 μM), a 600-fold higher EC₅₀ at hV1bR (4319 nM) and a 15-fold higher EC₅₀ at hV2R (109 nM). [Se-Se]-OT-OH (6) was found to be a weak antagonist at hV1aR (17% inhibition), with an IC₅₀ of 132 nM (SI Figure S1), which was confirmed in a radioligand displacement assay (Figure 2D). Based on earlier SAR studies (37, 57, 62) it is known that N-terminal deamination of OT analogues often leads to improved OTR activation; hence we synthesized and assessed whether d[Se-Se]-OT-OH (7) potency and selectivity for hOTR was improved. d[Se-Se]-OT-OH (7) (again a partial agonist at hOTR with E_max=59%), however, did not improve potency at OTR, yet improved potency at all three AVPRs thus yielding a less interesting selectivity profile than [Se-Se]-OT-OH (6) (Figure 1C, SI Table S2).

OT analogues 1-7, AVP and dAVP (11,12) were further tested using the homogeneous time-resolved fluorescence (HTRF) inositol-1-phosphate (IP₁) functional assay to allow comparison with the downstream intracellular calcium changes measured using the FLIPR (Figure 1E, SI Table S3). The HTRF-IP₁ assay correlated well with the FLIPR assay (SI Table S5 and Supplementary Note 2 in SI) and the potency and selectivity trends, particularly with lead compound [Se-Se]-OT-OH (6) (Figure 2C), were confirmed. [Se-Se]-OT-OH (6) displayed again partial agonism (E_max=60%) at hOTR with an EC₅₀ of 38 nM (2.6-fold less potent than OT) and no activity at hV1aR and hV1bR at concentrations up to 10 μM.

Radioligand displacement assay at human OT, V1a, V1b, and V2 receptors

Binding data for compounds 1-7 and 11, 12 were obtained in radioligand displacement assays measuring the competition with ³H-OT at hOTR,¹²⁵I-V1a antagonist at hV1aR, and ³H-AVP at the human V1b and V2 receptors in cell membrane preparations of heterologously expressed receptor (SI Table S4). The experimental inhibition constants (K_i) correlated well with the functional data (SI Table S2 & S3). [Se-Se]-OT-OH (6) displaced ¹²⁵I-V1a antagonist at hV1aR with a K_i of 57 nM, confirming the weak antagonistic effects observed in the FLIPR assay (Figure 2D and SI Table S4).

Schild regression analysis for binding mode

To confirm competitive binding of [Se-Se]-OT-OH (6) at hOTR, we performed a series of Schild regression experiments with OT (1) and [Se-Se]-OT-OH (6) against the well-characterized hOTR antagonist atosiban (AT). Results of these experiments revealed that all three compounds were competitive agonists at hOTR (SI Figure S2).

[Se-Se]-OT-OH - a stable and selective OTR agonist for translational studies

Considering that [Se-Se]-OT-OH (6) is one of the most selective OTR agonists to date and represents an important molecular probe to study OTR, it was important to (i) prove the hypothesis that subtle pharmacophore framework modifications can yield selectivity gains that are conserved across murine and human receptors, (ii) characterize its activity in a physiological and clinical relevant setting, (iii) determine its metabolic stability compared to OT (which is used in the clinic) and (iv) assess its cardiovascular side effect profile. We therefore carried out the pharmacological characterization of 6 across the four murine receptors (i), analyzed its role in central modulation of social anxiety/fear and tested it in one of the most important clinical applications of OT, namely labor induction/augmentation (ii), determined its stability in human serum (iii), and characterized its ability to activate cardiomyocytes that are linked to known cardiovascular side effects (iv).

Binding and functional analysis at murine OT, V1a, V1b and V2 receptors (i)

To determine if the pharmacological profile observed in human receptors is retained in mice, we tested [Se-Se]-OT-OH (6) and d[Se-Se]-OT-OH (7) in functional (SI Table S3, SI Figure S3) and binding assays (SI Table S4, SI Figure S4) across the four murine receptor subtypes.

[Se-Se]-OT-OH (6) retained its functional selectivity for OTR in mice and was inactive at mV1aR and mV1bR up to concentrations of 10 μM (SI Table S3, SI Figure S3). It was able to activate mOTR with an EC₅₀=11.9 nM (E_max=43%), which is only slightly less potent in respect to OT (63). [Se-Se]-OT-OH (6) fully activated mV2R similarly to the human receptors – with an EC₅₀ (183 nM) that is ~1000-fold larger compared to AVP (64) and 60-fold compared to OT (64). [Se-Se]-OT-OH (6) bound to the mOTR with a K_i=32.6 nM and we observed bi-phasic binding curves for mV2R (K_i1=0.71 nM and K_i2=1855 nM) and mV1aR (antagonist; K_i1=16.3 nM, K_i2 >10 μM). [Se-Se]-OT-OH (6) was characterized by very low affinity binding for mV1bR (K_i >10 μM), which correlates well with the absence of activation (SI Table S4, SI Figure S3).

d[Se-Se]-OT-OH (7) also retained its selectivity profile in mice, again being less selective than [Se-Se]-OT-OH (6). Binding and functional activation was observed at all four murine receptors subtypes, including V1b, which displayed no binding/activity with [Se-Se]-OT-OH (6) (SI Table S3 & S4). d[Se-Se]-OT-OH (7) is a partial agonist at mOTR (EC₅₀=5.9 nM; E_max=38%) and at the mV1bR (EC₅₀=438 nM; E_max=68%), and a full agonist at mV1aR (EC₅₀=438 nM) and at mV2R (EC₅₀=41.6 nM). The binding affinity for mOTR was in the nanomolar range (K_i=5.9 nM), similar to OT (K_i=0.83 nM, (63)), and binding affinities for mV1aR and mV1bR were less potent, with K_i=293 nM and K_i=438 nM respectively. The binding of d[Se-Se]-OT-OH (7) to mV2R was characterized by a bi-phasic binding curve with K_i1 of 0.71 nM and K_i2 of 1855 nM.

Human serum stability assay (ii)

A human serum stability study was carried out on [Se-Se]-OT-OH (6), revealing that it was more stable than OT with a half-life of 25 h in human serum compared to 12 h of OT (Figure 3B).

Behavioral efficacy in a mouse model of social fear (iii)

[Se-Se]-OT-OH (6) was tested in the well-established social fear conditioning (SFC) paradigm (65), characterized by reduced social exploration of conspecifics 24 h after SFC. [Se-Se]-OT-OH (6) reduced social fear, as reflected by a facilitated extinction curve, 10 min after its intracerebroventricular (icv; 250 μM/2 μl/mouse) infusion compared with vehicle-treated social fear conditioned (SFC⁺) mice. This effect was similar to that of synthetic OT, which also potently reversed social fear, when infused icv (250 μM/2 μl/mouse). All control mice, i.e. unconditioned (SFC^-) mice treated with either vehicle, [Se-Se]-OT-OH (6) or synthetic OT showed a similar level of social preference behavior as reflected by a high level of exploration of conspecifics. All SFC⁺ and SFC^-groups, independent of subsequent treatment, also showed similar investigation of a non-social stimulus (small empty cage) indicating similar non-social anxiety (Figure 4).

Human myometrial contractility assays reflecting induction and augmentation of labor (iii)

To determine if the partial agonism observed with [Se-Se]-OT-OH (6) and d[Se-Se]-OT-OH OT (7) in the pharmacological studies would affect its physiological activity compared to the full agonist OT (1), we tested 1, 6 and 7's ability to induce contraction via activation of hOTR in myometrial human telomerase reverse transcriptase (hTERT-HM) cells (66) in a collagen gel contractility assay (67, 68). Both [Se-Se]-OT-OH (6) and d[Se-Se]-OT-OH (7) induced a contractile response comparable to OT (SI Figure S5). 6 and 7 increased contractility by 7.4% (P < 0.01) and 9% (P < 0.001) respectively, compared to 9% (P < 0.001) for OT. This was then confirmed in tissue bath studies using strips of pregnant human myometrium, where [Se-Se]-OT-OH (6) increased contraction amplitude in a dose-dependent manner that was comparable to OT (1). EC₅₀ for [Se-Se]-OT-OH and OT were 5.89 ± 3.08 nM (N=5) and 1.14 ± 0.56 nM (N=8) respectively (P > 0.05, P=0.3718, F-test; Figure 3C & D). The contraction profile with [Se-Se]-OT-OH, however, was markedly different to OT, as it maintained a more phasic and frequent contractile activity pattern compared to a more tonic-like activity under OT at the respective concentrations (Figure 3C & D).

Human cardiomyocyte assay to assess cardiovascular safety profile (iv)

We tested the functional effects of OT (1), AVP (11) and [Se-Se]-OT-OH (6) on maximal intracellular Ca²⁺ levels in human cardiomyocytes (Figure 5) to determine if the improved selectivity of [Se-Se]-OT-OH (6) for hOTR over hV1aR would have reduced cardiovascular liability (which is regulated via hV1aR (69–71)) compared to OT. Both AVP and OT concentration-dependently caused an increase in maximal intracellular Ca²⁺ in human cardiomyocytes with EC₅₀ values of 121 nM and 174 nM, respectively. OT only had a partial effect (53 ± 11 %) compared to AVP, which correlates well with our pharmacological results of hV1aR activation by OT (SI Table S2 & S3). Consistent with its improved selectivity, [Se-Se]-OT-OH had no effect on the maximal intracellular Ca²⁺ level in cardiomyocytes up to 10 μM.

Subtle pharmacophore framework modifications - a new strategy to retain selectivity across murine and human receptors

This SAR study (see Supplementary Note 1 in SI for a more in depth discussion of the SAR study) demonstrated that single atom modifications to the pharmacophore framework (C/N-terminus, disulfide bond) of the endogenous ligands OT/AVP can tune selectivity and yield ligands ([Se-Se]-OT-OH (6) and d[Se-Se]-OT-OH (7)) with an improved selectivity profile that is conserved across species (mouse and human, SI Table S3 and Figure 3A). In particular, the replacement of the disulfide bond with the slightly more hydrophobic and longer diselenide bond (51, 52) in combination with a C-terminal amide to acid change yielded the most pronounced selectivity gains for OTR ([Se-Se]-OT-OH (6), Figure 3A). This selectivity gain renders [Se-Se]-OT-OH (6) an excellent probe to delineate OTR function in the CNS in behavioral animal models (where V2R is not expressed) and a promising lead molecule for the clinic, which we further explored herein. Considering all pharmacological and physiological data from this study we recommend an effective dose of 50 nM or, for comparison studies with OT, a 10-fold higher concentration than OT to be used for future functional studies to ensure OTR-mediated action while retaining OTR preference.

General conclusions and guidelines for future OT/AVP design

(i) N-terminal deamination of OT ligands improves binding and potency at all four receptors, renders OT slightly more hydrophobic and improves its stability against proteases (62). N-terminal deamination of AVP shows little impact on potency and selectivity, yet we recommend deamination nevertheless since it offers enhanced stability against proteases. (ii) Recent development in the area of disulfide mimetics recommends the use of the superior seleno-ether bond instead of the disulfide or diselenide bond, mainly due to its enhanced metabolic properties against reduction (4). (iii) The underexplored C-terminus of OT and AVP has already yielded two interesting studies (conopressin agonist to antagonist switch (49) and this study) which form a good base for future antagonist designs: we envision more SAR studies around the C-terminus towards the development of a new class of OT/AVP antagonists that might be able to overcome species differences by harvesting antagonist design using the C-terminus in contrast to current antagonists that obtain antagonistic effects through sterically hindered modifications at the N-terminus. (iv) In terms of pharmacological characterization, the FLIPR is a valuable instrument for the primary screen at the OT/AVP receptors since it rapidly provides agonist/antagonist information with high-throughput capabilities. Follow-up characterization of selected leads by radioligand displacement, HTRF and/or Bioluminescence Resonance Energy Transfer (BRET) assays can then be used to better understand ligand kinetics and secondary messenger signaling pathways. The SAR study presented here is to our knowledge the first assessing affinity and functional data (FLIPR and IP₁) of the control compounds OT/dOT and AVP/dAVP over all four receptor subtypes in a single study, hence forming an important reference for future OT/AVP ligand development.

[Se-Se]-OT-OH modulating central actions (behavior) - reversal of social fear

OT has been linked to multiple high-profile disorders that are modulated centrally such as autism, bipolar disorders, schizophrenia, pain, anxiety and depression (5, 8, 16, 33). Recent studies showing that these central actions can be elicited via intranasal OT administration in humans spiked an enormous interest in this research area with numerous of clinical trials currently on-going that study the therapeutic effects of intranasal OT in these disorders. Considering that OT also activates V1aR and V1bR (SI Table S2-4), especially at high doses such as those used for intranasal application in humans, and that OTR can form homo and hetero dimers (50, 74–76), the exact target pharmacology and underlying mechanism of actions are still unclear and under debate. The receptor-specificity underlying these effects is of substantial interest given the partly opposite behavioral effects of OT and AVP especially in the context of anxiety and stress regulation (14). [Se-Se]-OT-OH (6) enables clear and simple distinction between the involvement of OTR versus V1aR and V1bR in the CNS without the use of time and cost intensive knock-out models or co-application of species selective V1a/V1bR antagonists. This is crucial for therapeutic target validation and important for future drug development efforts. We chose to behaviorally test our probe in the recently established mouse model of social fear conditioning, which has clinical implication to treat social anxiety disorders (15, 65, 77). SFC is a paradigm based on operant fear conditioning principles and generates robust social fear in mice without any confounding behavioral alterations, and icv infusion of OT has been shown to completely abolish social fear in mice (77). The use of [Se-Se]-OT-OH (6) in this study confirmed that this action is mediated via OTR with no influence of V1aR nor V1bR (Figure 4).

[Se-Se]-OT-OH modulating peripheral action - labor induction and augmentation

Selectivity against V1aR to reduce cardiovascular side effects is of particular importance in the periphery since it is the primary receptor subtype expressed in the vascular smooth muscle and heart (69–71, 78, 79). Clinically, OT is administered intravenously to induce labor, augment weak contractions in labor and treat postpartum hemorrhage, and intranasally to elicit lactation. Due to its activation of AVPRs, OT administration has only a limited therapeutic window of use and adverse effects reported in the mother include anaphylactic reaction, postpartum hemorrhage, cardiac arrhythmia, fatal afibrinogenemia, pelvic hematoma, subarachnoid hemorrhage, hypertensive episodes, rupture of the uterus, convulsions, coma and death due to water intoxication. In the neonate, adverse effects reported include bradycardia, cardiac arrhythmia, jaundice, retinal hemorrhage, permanent CNS or brain damage, seizures and death (source RxList and (45, 80)). Hence, there is a clear unmet need for better and more selective OTR agonists that are able to differentiate against the AVPRs to diminish side effects.

[Se-Se]-OT-OH (6) may be such a candidate as experiments showed no activity on human cardiomyocytes (in contrast to OT, which displayed nanomolar activity), where V1aR is highly expressed regulating many of the above-mentioned side effects (Figure 5). Furthermore, [Se-Se]-OT-OH (6) was able to induce augmentation of contractility in human myometrial preparations comparable to OT, though with a more regular contraction pattern (Figure 3C & D and SI Figure S4). By contrast, OT reduced contraction frequency in a dose dependent fashion but increased the contraction duration resulting in a much greater area under the tension curve (a measure of the total contractile activity/work done) (Figure 3C & D); this can lead to uterine hyperstimulation and rupture, which are a life-threatening events for mother and fetus (81, 82). [Se-Se]-OT-OH (6) inactivity at V1aR, which is alongside with OTR also highly expressed in the myometrium during pregnancy and involved in contraction (83, 84), in combination with its partial agonism at OTR results in a more controlled and safer alternative to induce and progress labor in humans. Thus, [Se-Se]-OT-OH (6) not only reduces AVPR-related side effects but also offers additional advantages to aid labor induction and progression via a more regular increase of contraction strength without the risk of uterine hyperstimulation or rupture. This improved side effect profile is expected to lower associated risks for mother and fetus significantly and expand the therapeutic window of use, thereby lowering the incidence of cesarean sections and its associated costs.

Human serum stability of [Se-Se]-OT-OH

Enzymatic stability is a key characteristic for many peptide drug candidates, hence it was important to characterize the stability of [Se-Se]-OT-OH (6) compared to OT (1). Diselenide bonds are known to exhibit a lower redox potential than disulfide bonds, rendering such mimetics more difficult to reduce, a feature that can be exploited in improving the half-life of peptide drug candidates (4, 51, 52, 56, 85). C-terminal amides on the other hand are often used in nature to produce slightly more stable compounds compared to peptides that have C-terminal acids. The human serum stability study showed that [Se-Se]-OT-OH (6) displayed an improved half-life of 25 h compared to the half-life of 12 h of OT (Figure 3B), suggesting that the impact of the more redox stable diselenide bond outweighs the disadvantage of the C-terminal acid modification. While this increase in stability might only have limited effects for peripheral applications due to rapid renal clearance (minutes), it could be significant for potential therapeutic CNS applications through intranasal delivery (e.g. treatment of autism in children or migraines in adults) (21).

Selenocysteine building block synthesis

An optimized protocol with yield-improving modifications was developed based on Chocat et al. (58), Tanaka et al. (59) and Oikawa et al. (60) enabling the synthesis of the Boc-L-Sec(Meb)-OH at a scale of 20 g with an overall yield of 62%. A new two-step strategy was devised for the synthesis of dSec(Meb)-OH with an overall yield of 12% (Figure 1).

Peptide synthesis

Peptides 1-16 were assembled manually via Boc-SPPS using the HBTU-mediated in situ neutralization protocol (51, 61). The analogues were cleaved by HF (90), purified by RP-HPLC, folded over 24 h in 0.1M NH₄HCO₃ buffer at pH 8.2 (1 mg/10 mL, ~100 μM) and purified by RP-HPLC to >95% purity.

FLIPR assay for the human OT, V1a and V1b receptors

Ca²⁺ responses were assessed in COS-1 cells transfected with OTR, V1a or V1b using a FLIPR^TetraPlus fluorescent plate reader (Molecular Devices, Sunnyvale, CA) as previously described (4). In brief, cells were loaded with Calcium-4 No Wash dye (Molecular Devices) for 30 min and responses to addition of test compounds were measured using a cooled CCD camera with excitation at 470–495 nM and emission at 515–575 nM. Data was analyzed using ScreenWorks 3.2 (Molecular Devices) and plotted using GraphPad Prism 6.

cAMP assay for hV2R

Cloning of hV₂R, transfection and propagation of HEK293 cells, and assay to determine formation of ³H-cAMP was performed according to published protocols (47, 91, 92).

HTRF-IP₁ assay for the human receptors

COS-1 cells were cultured and transiently transfected with plasmid DNA encoding the human OTR, V1aR or V1bR following the manufacturer’s protocol (Lipofectamine 2000, Invitrogen or FuGENE HD, and Roche Applied Science) (56). Assays measuring IP₁ accumulation were performed 48 h post-transfection following the manufacturer’s instructions (IP₁-HTRF assay kit, Cisbio International).

Radioligand displacement assay for the human receptors

Receptor affinity assays were performed using FlashBLUE GPCR scintillation beads (PerkinElmer) (56). Reactions containing increasing concentrations of competing OT/AVP analogues (10 pM to 10 μM), FlashBLUE GPCR beads (100 μg for hOTR and hV1aR/200 μg for hV1bR and hV2R), human OT/V1a/V1b/V2 receptor membrane preparation (5 μg protein) and radioligand (³H-OT (2 nM)/³H-AVP (0.5 nM)/³H-AVP (3 nM)) in assay buffer (50 mM Tris-HCl, 10 mM MgCl₂ (5 mM MgCl₂ for OT), 0.1% BSA, pH 7.4) were established in 96-well white polystyrene plates with clear flat bottoms in a total reaction volume of 80 μL.

Radioligand displacement assay for the murine receptors

Competitive binding assays were performed using ³H-OT and ³H-AVP (Perkin Elmer) and increasing concentration of the indicated peptides at 30°C on membranes prepared from HEK293 cells transfected with OTR, V1aR, V1bR and V2R murine receptors as in Busnelli et al. (63).

HTRF-IP₁ assay and cAMP determination for the murine receptors

IP₁ and cAMP accumulation were determined in HEK293 cells transiently transfected with DNA plasmids encoding the murine OT/V1a/V1b or V2 murine receptors using HTRF assays (IP₁ and cAMP assays, CisBio International, Bagnols-sur-Cèze, France) as previously described (63).

Stability assays

Peptides were added to human serum and aliquots were taken and analyzed by RP-HPLC and LCMS at 1, 2, 3, 4, 12, 24 and 48 h time points (56).

Social fear conditioning (SFC) mouse model

SFC was performed as previously described on adult mice (65). Briefly, on day 1 (social fear acquisition), the SFC^-mice were allowed to investigate a social stimulus (male conspecific in a small cage) in the conditioning chamber for 3 min without receiving any foot shocks, whereas SFC⁺ mice were given a 1-s electric foot shock (0.7 mA) each time they investigated (sniffed) the social stimulus. On day 2 (social fear extinction), SFC⁺ and SFC^-mice were first exposed to 3 non-social stimuli (empty small cages) to assess non-social investigation as a parameter of non-social fear and anxiety-related behavior. Mice were then exposed to 6 different conspecific males (social stimulus) in 3 min intervals, each in a different small cage to assess social investigation as a parameter of social fear. Mice received icv infusions of either vehicle (2 μl; sterile Ringer solution), OT (250 μM/2 μl; 500 pmol) or [Se-Se]-OT-OH (250 μM/2 μl; 500 pmol) 10 min prior to social fear extinction on day 2 as previously described (65, 77). All infusions were performed 10 min prior to social fear extinction on day 2 of the SFC paradigm. The dose and timing of OT was selected based on previous studies (77).

Human myometrial cell contractility assay

Human uterine myometrial smooth muscle cells (hTERT-HM) were cultured, collagen-gels were prepared, and the assays were performed as described previously (47, 68).

Tissue bath myometrial contractility assays

Strips of myometrium were dissected from biopsies obtained from the lower uterine incision site during Cesarean section (93). All women gave written informed consent to participate. The study was approved by the North West (Liverpool East) Research Ethics committee (Ref: 10/H1002/49) and by the Research and Development director at Liverpool Women’s Hospital NHS Foundation Trust, Liverpool, UK. [Se-Se]-OT-OH (6) or OT (1) was added to the organ baths and amplitude of contraction was measured and compared to control activity occurring immediately prior to application of the first concentration (100%). Data were analyzed using Origin Pro 9.0 and Graphpad Prism 5 (N ≥ 5) as described previously (47, 94, 95).

Human cardiomyocyte assay

Changes in intracellular Ca²⁺ levels were assessed in cardiomyocytes differentiated from WT WTC11 human induced pluripotent stem cells (hiPSCs) using a FLIPR^Tetra fluorescent plate reader (Molecular Devices, Sunnyvale, CA, USA). In brief, cells were loaded with Calcium-4 No Wash dye (Molecular Devices) according to the manufacturer's instructions and incubated at 37 °C for 30 min. Responses to addition of test compounds were measured using a cooled CCD camera with excitation at 470–495 nM and emission at 515–575 nM. Data was analyzed using ScreenWorks 3.2 (Molecular Devices) and plotted using GraphPad Prism 6.