Plasma and CSF oxytocin levels after intranasal and intravenous oxytocin in awake macaques

Psychoneuroendocrinology 66 (2016) 185–194

Contents lists available at ScienceDirect

Psychoneuroendocrinology journal homepage: www.elsevier.com/locate/psyneuen

Plasma and CSF oxytocin levels after intranasal and intravenous oxytocin in awake macaques Sara M. Freeman a , Sridhar Samineni b , Philip C. Allen b , Diane Stockinger b , Karen L. Bales a,c , Granger G.C. Hwa b,∗ , Jeffrey A. Roberts b a

California National Primate Research Center, University of California-Davis, Davis, CA 95616, USA Valley Biosystems, Inc., West Sacramento, CA 95605, USA c Department of Psychology, University of California-Davis, Davis, CA 95616, USA b

a r t i c l e

i n f o

Article history: Received 8 October 2015 Received in revised form 21 December 2015 Accepted 14 January 2016 Keywords: Neuropeptide Rhesus macaque Cerebrospinal fluid Nasal spray ELISA

a b s t r a c t Oxytocin (OT) is a neuropeptide that mediates a variety of complex social behaviors in animals and humans. Intranasal OT has been used as an experimental therapeutic for human conditions characterized by deficits in social functioning, especially autism spectrum disorder and schizophrenia. However, it is currently under intense debate whether intranasal delivery of OT reaches the central nervous system. In this study, four female rhesus macaques were implanted with chronic intrathecal catheters and used to investigate the pharmacokinetic profile of OT in the central nervous system and the peripheral vasculature following intravenous (IV) and intranasal (IN) administration of OT. In a randomized, crossover design, OT was given to four awake monkeys at three different doses based on body weight (0.1 IU/kg; 1 IU/kg; 5 IU/kg). A time course of concurrent cerebrospinal fluid (CSF) and plasma samples were taken following administration. We found a dose-dependent effect of IV OT treatment on plasma OT levels, which peaked at 5 min post-dose and gradually returned to baseline by 120 min. In contrast, a change in CSF OT was only observed at the highest IV dose (5 IU/kg) at 15 min post-dose and gradually returned to baseline by 120 min. After IN administration, there was no significant change in plasma OT at any of the three doses. However, at the highest dose level, we found a significant increase in CSF OT at 15–30 min post- dose. The results of this study in light of recent, similar publications highlight the importance of methodological consistency across studies. This study also establishes a non-human primate model that can provide a stable platform for carrying out serial sampling from the central nervous system and peripheral vasculature concurrently. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction Oxytocin (OT) is a nine amino acid neuropeptide that has been shown to modulate social behavior in both animals and humans. Due to the ability of this peptide to promote prosocial behaviors in animals, such as pair-bonding and affiliation, there has been an increasing number of studies investigating the effects of OT on human behavior (Young and Flanagan-Cato, 2012). However, OT has been shown to be too large to effectively cross the blood–brain barrier (BBB) and access neural structures when delivered peripherally, such as after intravenous (IV) administration (Ermisch et al., 1985). Thus, the overwhelming majority of OT studies in humans have employed an intranasal (IN) dosing paradigm in order to (hypothetically) bypass the BBB by delivering OT directly to the

∗ Corresponding author at: 1265 Triangle Court, West Sacramento, CA 95605, USA. E-mail address: [email protected] (G.G.C. Hwa). http://dx.doi.org/10.1016/j.psyneuen.2016.01.014 0306-4530/© 2016 Elsevier Ltd. All rights reserved.

olfactory and respiratory epithelia in the nasal cavity, where OT is thought to enter the brain via the olfactory neurons or trigeminal nerve endings (Quintana et al., 2014). Administering intranasal OT (IN-OT) to healthy humans has been shown to affect a suite of social behaviors, such as trust (Kosfeld et al., 2005), eye contact (Guastella et al., 2008), emotion recognition (Domes et al., 2007), socially-reinforced learning (Hurlemann et al., 2010), and pair-bonding-related behaviors (Scheele et al., 2012, 2013). More recent efforts have expanded these findings by administering IN-OT to clinical populations with deficits in social function, such as individuals with autism spectrum disorder (ASD). Indeed, there are now several studies showing that IN-OT can ameliorate several aspects of social function in adults and children with ASD, including enhancing the saliency of human faces (Domes et al., 2013), improving emotion recognition (Guastella et al., 2010), and increasing feelings of social reciprocity and increasing gaze to the eyes (Andari et al., 2010).

186

S.M. Freeman et al. / Psychoneuroendocrinology 66 (2016) 185–194

Despite the progress that has been made in the clinical application of OT, the pharmacokinetics of IN-OT within the central nervous system (CNS) remain unclear. Several studies have begun to characterize the plasma response to IN-OT in humans (Burri et al., 2008; Gossen et al., 2012; Kirkpatrick et al., 2014; Striepens et al., 2013), and one study has reported changes in CSF OT levels in humans after IN-OT treatment (Striepens et al., 2013). These studies have all reported significant increases in plasma OT after treatment with IN-OT at doses between 24–40 IU, with peaks around 15–30 min post-administration and a return to baseline levels by 75–90 min. One recent study delivered IN-OT with a novel, breathpowered device that closes the soft palate during administration and found significant increases in plasma OT levels at a much lower dose of 8 IU (Quintana et al., 2015). In the only study that has assessed the effect of IN-OT on CSF OT levels in humans, there was only a significant increase at 75 min post-administration (Striepens et al., 2013). In order to extend the results from the human literature, there are three studies that have sought to characterize the effects of IN-OT treatment on cerebrospinal fluid (CSF) levels of OT in the non-human primate (NHP) (Chang et al., 2012; Dal Monte et al., 2014; Modi et al., 2014). However, there were limitations in their respective study designs as well as variation in methodologies across studies, which may underlie the differing outcomes reported by these studies. First, these studies were all based on acute cisterna or lumbar puncture to collect CSF samples. The method of acute dural puncture poses an inherent limitation on sampling frequency, because it can cause post-dural puncture headache, CSF leak, intracranial hypotension (Burns and Scheinfeld, 2013; Gaiser, 2006) or even venous sinus thrombosis (Wilder-Smith et al., 1997). Second, the samples were taken at only one or two time points post-administration. Furthermore, the way in which OT was delivered to the nasal passage differed among these previous studies, with some using a nasal spray (IN-OT), which is consistent with the method used in human studies, and others using aerosolized OT that is inhaled via a nebulizer and nose cone. Finally, two of the studies used monkeys that were anesthetized during dosing, and one used awake monkeys. In the current study, we established an awake NHP model that allows for frequent and concurrent sampling of CSF and whole blood to measure the OT levels within the CNS and the peripheral vasculature following IN or IV administration of OT. The experiments were carried out in rhesus macaques that were chronically implanted with an intrathecal catheter so that the animals could be sampled repeatedly in the awake state over an extended period of time. We included IV dosing (IV-OT) because it is the principal route of OT administration for labor induction and postpartum hemorrhage (Prata et al., 2013; Stubbs, 2000) and the treatment of other conditions in humans (Hollander et al., 2003). Further, it has been reported that IV-OT can penetrate into the CNS in rodents (Mens et al., 1983).

cages in 12:12 h light dark cycle room. Animals were monitored at all times for distress or other signs of adverse reactions. In a randomized, crossover study design, animals received OT via IV or IN routes with minimum of a one-week washout period between each administration. The same four animals were used for all IV and IN dosing procedures. Animals were dosed awake, and blood and CSF samples were drawn at predetermined time points (0 min [predose], 5, 15, 30, 60, and 120 min). All monkeys were implanted with chronic CSF catheters as described below. Because there is evidence that plasma OT changes during the course of the menstrual cycle in macaques (Falconer et al., 1980), this entire study took place during the nonbreeding season (March–September), when the macaque menstrual cycle is weak, irregular, and least likely to impact OT levels (Du et al., 2010). Based on the timing of the study and the randomized, crossover design, the effects of menstrual cycle on OT levels were not assessed.

2. Materials and methods

2.4. Intravenous OT administration

2.1. Animals

Animals were chair-restrained and given an intravenous bolus of OT at either 0.1, 1, or 5 IU/kg body weight using right cephalic vein. Based on average body weights over the course of the study, these doses are the equivalent of 0.58–0.73 IU, 5.5–7.3 IU, or 29–36.5 IU, depending on the individual animal’s body weight. There was a minimum of a one-week washout period between each dose administration. Following IV-OT administration, timed venous blood samples were collected in EDTA tubes from the left arm cephalic vein. Concurrent blood (1.0–1.5 mL) and CSF (300 ␮L) samples were collected pre-dose (0 min) and at 5, 15, 30, 60, and 120 min post OT administration. Blood was kept on ice until cen-

Four adult female rhesus monkeys (Macaca mulatta) between the ages of 6 and 8 years (mean body weight, 5.8–7.3 kg) were used for this study (Valley Biosystems, CA). The animals were fed a standard laboratory primate chow (TekLad, Madison, WI) twice daily and housed in accordance with Guide for the Care and Use of Laboratory Animals. All animal studies were performed as approved by the Valley Biosystems Institutional Animal Care and Use Committee. Animals were pair housed during the day except during day of dosing and sample collection, when they were housed in squeeze

2.2. Behavioral habituation and training for awake dosing Animals were habituated to chair restraint using positive reinforcement for a minimum of 2 weeks prior to study. The animals did not show any obvious signs of stress during the dosing procedures. 2.3. Implantation of chronic CSF catheter The animals were habituated and chair-trained for a minimum of two weeks prior to implantation. They were fasted overnight before surgery. On the day of surgery, the animals were immobilized with ketamine (10 mg/kg; Phoenix), dexmedetomidine (15 ␮g/kg; Pfizer) and Buprenorphine (0.03 mg/kg; Reckilt Benckider Pharmaceutics) intramuscularly. Animals were intubated and maintained on isoflurane (1–1.5%) anesthesia with monitoring of blood pressure, heart rate, respiration, and SPO2 . The skin was surgically prepped from mid cranium to the lumbar spine. An incision was made of the occipital ridge of the skull to approximately C4. Blunt dissection was used to expose atlantooccipital membrane at the level of cisterna magna. The atlantooccipital membrane was punctured with an 18 g needle and a custom-made polyurethane catheter was inserted. A second incision was made lateral to the lumbar spine, and the catheter was fed underneath the skin to this incision. The catheter was then connected to an access port. Once CSF flow was established, the catheter was secured to the surrounding muscle using 2–0 silk (Ethicon). Both skin incisions were closed with 3–0 Vicryl (Ethicon) in a subcuticular pattern. The animals were reversed with Antiseden (Pfizer) and allowed to recover and returned to their home cages. Post-operative analgesia included Buprenorphine (0.03 mg/kg), and either Rimadyl (2.2 mg/kg; Pfizer Animal Health) or ketofen (2 mg/kg; Fort Dodge). Cefazolin (25 mg/kg; Walgreens Critical Care) was given as a perioperative antibiotic. Animals were given a minimum of two weeks to recover from surgery prior to initiation of study. All CSF was collected from the port sterilely.

S.M. Freeman et al. / Psychoneuroendocrinology 66 (2016) 185–194

trifuging. Plasma was separated by centrifuging the blood sample at 15,000 rpm for 15 min at 4 ◦ C. Plasma and CSF samples were stored at −80 ◦ C. 2.5. Intranasal OT administration Nasal spray bottles that were designed to deliver a specific volume of drug per spray were used for IN dose administration. Briefly, animals were chair restrained, and the head and jaw were manually restrained and stabilized. OT was administered into the nostrils at 0.1, 1, or 5 IU/kg body weight with a nasal spray bottle, alternating nostrils between sprays. Nasal spray bottles were weighed pre- and post-dose to verify administration of specific volume of OT. All animals were breathing normally, and no sneezing was observed during or after dosing. There was a minimum of a oneweek washout period between each dose administration. Following OT administration, timed venous blood samples were collected in EDTA tubes from the left arm cephalic vein. Concurrent blood (1.0–1.5 mL) and CSF (300 ␮L) samples were collected pre-dose (0 min) and at 5, 15, 30, 60, and 120 min post OT administration. Blood was kept on ice until centrifuging. Plasma was separated by centrifuging the blood sample at 15,000 rpm for 15 min at 4 ◦ C. Plasma and CSF samples were stored at −80 ◦ C. 2.6. Measurement of OT concentration in samples Levels of OT in plasma and CSF were measured using commercially available OT ELISA kits (Enzo Life Sciences, Farmingdale, NY). This assay has been previously validated for use in rhesus macaques (Bales et al., 2005). Assays were performed following the manufacturer’s protocols. Prior to assay, plasma was diluted with assay buffer using a 1:8 ratio. CSF samples were not diluted prior to assay. No samples went through extraction prior to assay. Limits of detection were 15.6 pg/mL to 1000 pg/mL. Values falling below or above the limits of detection were set to 15.6 pg/mL or 1000 pg/mL, respectively. Intra- and interassay CVs for the CSF samples were 5.4 and 10.1%, respectively. Intra- and interassay CVs for the plasma samples were 4.4 and 7.6%, respectively. 2.7. Statistical analysis Data from plasma and CSF samples were analyzed separately. Data was analyzed using a general linear model (GLM) in SAS (version 9.2, SAS Institute, Cary, NC) with fixed effects for administration method (IV or IN), dose (0.1 IU/kg, 1 IU/kg, or 5 IU/kg), time (0, 5, 15, 30, 60, 120 min), and the interaction of dose × time. We also included a random factor for animal ID to account for individual differences. Alpha was set to 0.05. Planned, post-hoc t-tests were carried out in the event that a significant main effect of the interaction of dose × time was found. In these cases, each time point (5, 15, 30, 60, and 120) was compared to the baseline value within each dose in order to determine the effect of time on OT levels. To correct for these multiple comparisons, the alpha level (0.05) was divided by the number of comparisons (15), which resulted in a new critical value of 0.003 that was used to determine significance. In order to better represent the results in one subset of data (CSF OT levels after IN-OT treatment), the data was normalized across individuals by calculating the percent change from baseline and analyzed using a two- way ANOVA. In addition, the subset of data that includes only the baseline plasma and CSF samples (n = 24; six baseline samples were taken for each of the four animals) were analyzed in a separate GLM analysis to determine whether baseline OT levels were related between the two biological compartments—a standing question in the field of OT research.

187

3. Results 3.1. Plasma We found a significant main effect of method (IN vs IV) (F1,139 = 47.74; p < 0.001; 2 = 0.093), with plasma OT levels significantly higher in the IV samples compared to IN (Fig. 1A). We also found a significant main effect of animal ID (F3,139 = 6.38; p < 0.001; 2 = 0.232), indicating individual differences across animals. Within the plasma samples taken after IV-OT administration, we found a significant main effect of dose (F2,51 = 84.37; p < 0.001; 2 = 0.224; Fig. 1B), time (F5,51 = 82.38; p < 0.001; 2 = 0.546; Fig. 1C), and a significant interaction effect of dose × time (F10,51 = 9.41; p < 0.001; 2 = 0.125). Post-hoc tests determined that after IV-OT treatment with the lowest dose (0.1 IU/kg OT), plasma OT was only significantly elevated compared to baseline at 5 min posttreatment (Fig. 2A, caret). After IV-OT treatment with the medium dose (1 IU/kg OT), levels of plasma OT were significantly increased compared to baseline at 5, 15, and 30 min (Fig. 2A, crosshairs) but were not significantly different from baseline at 60 and 120 min. After IV-OT treatment with the highest dose (5 IU/kg OT), levels of plasma OT were significantly increased compared to baseline at all time points (Fig. 2A, asterisks), indicating that plasma OT levels had not yet returned to baseline by 2 h post-treatment. Within the plasma samples taken after IN-OT administration, we found a significant main effect of dose (F2,51 = 9.66; p < 0.001; 2 = 0.117; Fig. 1B) but no main effect of time (F5,51 = 0.89; p = 0.4939; Fig. 1C) and no interaction effect of dose × time (F10,51 = 1.09; p = 0.3906; Fig. 2B). It appears that the main effect of dose is being driven by a decreased mean in the 1 IU/kg group compared to the other two doses.

3.2. CSF We found a significant main effect of method (IN vs IV) (F1,139 = 22.25; p < 0.001; 2 = 0.381), with CSF OT levels significantly higher in the IV samples compared to IN (Fig. 1D). We also found a significant main effect of animal ID (F3,139 = 33.05; p < 0.001; 2 = 0.085), indicating stable individual differences across animals, which has also been found in humans (Levine et al., 2007), although the functional significance of individual differences is not well understood. Within the CSF samples taken after IV-OT administration, we found a significant main effect of dose (F2,51 = 31.35; p < 0.001; 2 = 0.212; Fig. 1E) and time (F5,51 = 7.73; p < 0.001; 2 = 0.131; Fig. 1F), and a significant interaction effect of dose × time (F10,51 = 2.51; p < 0.02; 2 = 0.085). Post-hoc tests determined that after IV-OT treatment with the lowest dose, there were no significant increases in CSF OT at any time point compared to baseline (Fig. 2C). After IV-OT treatment with the medium dose, levels of CSF OT were also not significantly increased compared to baseline at any time point (Fig. 2C). After IV-OT treatment with the highest dose, levels of CSF OT were significantly increased compared to baseline at all time points except 120 min (Fig. 2C, asterisks), indicating that CSF OT levels returned to baseline by 2 h post-treatment. Within the CSF samples taken after IN-OT administration, we found a significant main effect of time (F5,51 = 3.64; p < 0.01; 2 = 0.073; Fig. 1F) but no main effect of dose (F2,51 = 0.57; p = 0.5697; Fig. 1E) and no interaction effect of dose × time (F10,51 = 0.90; p = 0.5379; Fig. 2D). These results indicate a moderate but significant increase in CSF OT over time after IN-OT treatment (Fig. 1F). When these data are normalized across individuals by calculating the percent change from baseline, the highest dose of IN-OT causes a significant increase in CSF OT at 15 and 30 min post-treatment (Fig. 3).

188

S.M. Freeman et al. / Psychoneuroendocrinology 66 (2016) 185–194

Fig. 1. Summary of main effects. Analysis of the changes in plasma OT (A–C) and CSF OT (D–F) after treatment with either intravenous (IV) or intranasal (IN) OT treatment showed a variety of significant main effects. Error bars indicate SEM.

3.3. Relationship between baseline plasma and CSF levels We also sought to determine whether there was a relationship between baseline OT levels between the two biological compartments. However, we found that baseline plasma OT levels did not significantly predict baseline CSF OT levels (F = 0.43; p = 0.5211). 4. Discussion This study is the first to use rhesus macaques with chronically implanted CSF catheters in order to study the pharmacodynamics of OT after exogenous OT administration. This study design permitted a more thorough characterization of the effect of IV-OT and IN-OT treatment in rhesus macaques than has been previously published. Specifically, the current study is the first of this kind to (1) use awake monkeys for dosing and CSF sample collection, (2) directly compare more than two different doses of exogenous OT treatment

(and use weight adjusted doses), and (3) collect five concurrent post-treatment plasma and CSF samples within a 2-h time course. The current study is also the first to use female rhesus macaques to evaluate the effect of IN-OT and IV-OT treatment on endogenous levels of OT in plasma and CSF. While there have been studies that measured baseline CSF OT in male macaques or in female macaques, there has not been a study to our knowledge that directly evaluates sex differences in baseline CSF OT levels in macaques. Thus, it is not possible at this time to know how the results from our study in females compare to the previous studies, which all used males (Table 1). Although plasma OT changes across the macaque menstrual cycle (Falconer et al., 1980), our study took place during the nonbreeding season, when the menstrual cycle in macaques is weak and irregular (Du et al., 2010). The timing of our study, combined with the randomized, crossover design, should prevent any effects of menstrual cycle on our plasma results. Furthermore, several studies suggest that CSF OT levels are unaffected by hormonal

Table 1 Summary of studies investigating the effect of OT treatment on plasma and/or CSF OT levels in rhesus macaques. Sedation state

Sample

Study design

Delivery method

OT dose (IU)

Sampling timepoints (min)

Effect on plasma OT

Effect on CSF OT

Sample collection site

Assay type

Sample extraction?

Chang et al., 2012

Awake

4 males

Vehicle controlled, crossover design (n = 2); vehicle only (n = 2)

Aerosolized

25

CSF:

35

N/A

CSF: cisterna magna

ELISA

CSF: no

Modi et al., 2014

Anesthetized

8 males

OT (n = 4); vehicle (n = 4)

Aerosolized

24

CSF:

0, 60, 120

Peak at 5 min; increased at all time points compared to baseline and to vehicle

Increased at 35 min compared to within-subjects and betweensubjects vehicle Increased at both time points compared to baseline but not compared to vehicle

CSF: Lumbar Plasma: femoral vein

RIA

CSF: yes Plasma: yes

6 males

Vehicle controlled, crossover design

Nasal spray

24

Plasma: CSF:

0, 5, 15, 60, 120 0, 60, 120

Intravenous

48

Plasma: CSF:

0, 60, 120 0, 60, 120

Aerosolized

48

Plasma: CSF:

0, 5, 15, 60, 120 0, 40

CSF: cisterna magna Plasma: femoral vein

ELISA

CSF: no Plasma: yes

Nasal spray

48

Plasma: CSF:

0, 10, 20, 30, 40 0, 40

Dal Monte et al., 2014

Anesthetized

15 males

Vehicle controlled, crossover design

Increased at both time points compared to baseline and compared to vehicle

No effect

Peak at 5 min; increased at all time points compared to baseline and compared to vehicle

No effect

No effect

Increased at 40 min compared to baseline

Increased compared to baseline at 40 min

Increased at 40 min compared to baseline

S.M. Freeman et al. / Psychoneuroendocrinology 66 (2016) 185–194

Reference

189

190

Table 1 (Continued) Sedation state

Sample

Study design

Delivery method

OT dose (IU)

Sampling timepoints (min)

Freeman et al. (current study)

Awake

4 females

Crossover design; baseline control; no vehicle

Nasal spray

0.61

Plasma: CSF:

0, 10, 20, 30, 40 0, 5, 15, 30, 60, 120

Plasma:

0, 5, 15, 30, 60, 120 0, 5, 15, 30, 60, 120 0, 5, 15, 30, 60, 120 0, 5, 15, 30, 60, 120

6.42

CSF: Plasma:

323

CSF:

Plasma: Intravenous

1

0.6

CSF:

Plasma: 2

6.4

CSF:

Plasma: 323

CSF:

Plasma: 1,2,3

0, 5, 15, 30, 60, 120 0, 5, 15, 30, 60, 120

0, 5, 15, 30, 60, 120 0, 5, 15, 30, 60, 120

0, 5, 15, 30, 60, 120 0, 5, 15, 30, 60, 120

Effect on plasma OT

Effect on CSF OT

Sample collection site

Assay type

Sample extraction?

No effect

No effect

CSF: cisterna magna Plasma: forearm vein

ELISA

CSF: no Plasma: no

No effect

No effect

No effect

Increased at 15 and 30 min compared to baseline

Peak at 5 min; increased compared to baseline at 5 min only

No effect

Peak at 5–15 min; increased compared to baseline at 5, 15, and 30 min

No effect

Peak at 5–30 min; increased compared to baseline at all time points; nearing baseline by 120 min

Peak at 5–30 min; increased compared to baseline at 5, 15, 30, and 60 min; returned baseline by 120 min

0, 5, 15, 30, 60, 120

These doses are converted from average IU/kg body weight to average total IU in order to be comparable across studies (1 0.1 IU/kg, 2 1 IU/kg, 3 5 IU/kg).

S.M. Freeman et al. / Psychoneuroendocrinology 66 (2016) 185–194

Reference

S.M. Freeman et al. / Psychoneuroendocrinology 66 (2016) 185–194

191

Fig. 2. Summary of interaction effects of dose × time. In samples collected at baseline and 5, 15, 30, 60, 120 min after treatment with either IV or IN OT, we found a significant interaction effect of dose × time in plasma and CSF OT levels only after IV treatment (A, C). There were no interaction effects in either biological compartment after IN treatment (B, D). Error bars indicate SEM. *p < 0.003 compared to baseline for 5 IU/kg. +p < 0.003 compared to baseline for 1 IU/kg. ˆp < 0.003 compared to baseline for 0.01 IU/kg.

reflect procedure-related effects on OT levels, such as an increase in endogenous OT release due to handling stress, which would become evident if a placebo control had been used. However, our main research focus was to build from previous studies, which have already used placebo-controlled designs to establish that OT treatment can selectively change OT levels in the body. Thus, we justified the use of a baseline-controlled design (along with more sampling time points and multiple doses) in order to provide a more complete characterization of the pharmacodynamics of OT treatment. Additionally, we feel that the extensive habituation of all animals to the sampling procedure, as well as the significant interaction effects of dose and time, argue strongly against any effects that may have been due to handling stress, or the dosing/sampling procedure. 4.1. Comparison to previous studies in nonhuman primates Fig. 3. Percent change from baseline in CSF OT after treatment with IN-OT. The data were transformed and plotted as the percent change from baseline in order to better represent the significant main effect of time in the CSF data after intranasal treatment. Error bars indicate SEM. *p < 0.05 compared to baseline for 5 IU/kg.

status in female macaques. Specifically, CSF OT levels are similar in lactating and nonlactating female macaques (Challinor et al., 1992) and in ovariectomized and intact female macaques (Amico et al., 1989). Based on these reports, it is unlikely that underlying hormone or sex differences would drastically impact the responsiveness of macaque CSF OT levels to IN-OT or IV-OT treatment, although further research in both males and females is needed. Furthermore, our decision to use a baseline-controlled design rather than a placebo-controlled design merits some discussion. We realize that the results from this design could potentially

To date, only three studies in NHP have sought to characterize whether IN-OT treatment results in elevated CSF levels of OT (Chang et al., 2012; Dal Monte et al., 2014; Modi et al., 2014). Similar to the current study, these studies also administered intranasal OT to rhesus macaques followed by a time course of samples of blood and CSF. Among these studies, there were several methodological differences, which we will outline briefly here and which are also summarized in Tables 1 and 2. The main differences in experimental design are (1) the sedation state of the animals during dosing, (2) the method of intranasal delivery, (3) the dose, and (4) the time points for sample collection. First, two out of the three studies used anesthetized macaques during dosing (Dal Monte et al., 2014; Modi et al., 2014). Second, these studies have used both the handactuated nasal spray pumps, which are the most common delivery method for human studies, as well as aerosolized delivery using a

192

S.M. Freeman et al. / Psychoneuroendocrinology 66 (2016) 185–194

Table 2 Examples of variations in experimental parameters across studies. Dose (IU) 24 48 weight-adjusted Sedation during dosing Awake Anesthetized IN delivery method Nasal spray Aerosolized

CSF time course (min) 0, 60, 120 0, 40 0, 5, 15, 30, 60, 120 CSF collection site Lumbar Cervical Type of OT assay ELISA RIA

Plasma time course (min) 0, 10, 20, 30, 40 0, 5, 15, 60, 120 0, 5, 15, 30, 60, 120 Plasma collection site Femoral vein Forearm vein Sample extraction? Yes No

nebulizer and nose cone/mask. Third, two doses were used across the three studies: 24–25 IU or 48 IU. Finally, two of the studies only sampled CSF at one time point post-treatment (Chang et al., 2012; Dal Monte et al., 2014), and the third sampled CSF at only two time points (Modi et al., 2014). While some of our results agree with previous reports, there were also some unexpected findings. First, we expected that the IVOT treatment would increase plasma OT in a dose dependent way, which we found. We also expected that treatment with an OT nasal spray would also increase plasma OT levels, as that result has been previously reported in the literature for anesthetized macaques (Dal Monte et al., 2014; Modi et al., 2014) as well as humans (Burri et al., 2008; Gossen et al., 2012; Kirkpatrick et al., 2014; Quintana et al., 2015; Striepens et al., 2013). However, we found that the OT nasal spray had no effect on plasma levels of OT, regardless of dose. This conflicting result could be due to the awake and upright posture of the animal, which could impact the delivery of the dose into the nasal cavity and prevent absorption into the bloodstream via the capillaries of the nasal epithelium. Although human studies have also used awake and upright intranasal dosing with a nasal spray and shown elevations in plasma OT after treatment, these studies in humans can use verbal directions to coordinate a deep inhale/sniff at the time of spray delivery. In awake macaques, it is not possible to direct the animals to inhale deeply through the nose, so it is possible that the delivery of OT to the nasal epithelium in our study may have been missed, causing a lack of significant absorption into the capillaries of the nasal passage. This idea has been proposed already; misaligned nasal spray delivery of OT in the anterior portion of the nasal cavity would bypass the epithelium and instead drain into the throat, where it would then be swallowed and digested (Quintana et al., 2014), which is one possible reason why we did not detect elevations in plasma OT after IN-OT administration. If this reasoning is true, then the use of a hand-actuated nasal spray for the delivery of IN-OT in awake macaques may not be the most effective delivery method for elevating plasma levels of OT, which agrees with the recommendations from one of the studies in anesthetized macaques (Modi et al., 2014). We detected a modest, yet significant increase in CSF OT at 15 and 30 min after treatment with our highest dose of IN-OT (29–36.5 IU, depending on the weight of the animal), a result that is corroborated by previous studies. Dal Monte et al. (2014) reported an increase in CSF OT at 40 min post-treatment with 48 IU of INOT, and Chang et al. (2012) detected an increase in CSF OT at 35 min post-treatment with 25 IU of aerosolized OT. The significant increase from baseline at 30 min detected in the current study would likely remain elevated had we sampled five and ten minutes later, at the time points used in Chang et al. (2012) and Dal Monte et al. (2014), respectively. Modi et al. (2014) found no change in CSF OT at 60 min or 120 min post-treatment with 24 IU IN-OT, which agrees with our finding that CSF OT levels were not significantly different from baseline at these same two time points. However, Modi et al. (2014) found increased CSF OT at 60 min and 120 min after treatment with aerosolized OT, which suggests that aerosolized OT

may be more effective than IN-OT at elevating central levels of OT for a longer period of time. Interestingly, we found that treatment with the highest dose of IV-OT significantly increased CSF levels of OT in our animals in a time- dependent manner, which returned to baseline levels by 120 min post-administration. This result was unexpected, as OT does not cross the blood–brain barrier with much efficacy. In fact, only 1% of an injected, peripheral dose of OT crosses the blood–brain barrier in rats (Ermisch et al., 1985). Furthermore, Modi et al. (2014) found that treatment with 48 IU IV-OT did not increase CSF OT in anesthetized macaques. However, it is possible that elevated levels of OT in the bloodstream, as a result of OT treatment, could result in central release of OT through peripheral feedback mechanisms (Quintana et al., 2014). 4.2. Comparison to previous studies in humans There have been several studies in humans that have investigated the relationship between IN-OT treatment and plasma levels of OT (Burri et al., 2008; Gossen et al., 2012; Kirkpatrick et al., 2014; Quintana et al., 2015), but there has only been one study that examined CSF levels of OT in humans after IN-OT (Striepens et al., 2013). To our knowledge, there are no studies in humans that have examined the effects of IV-OT on CSF levels of OT, so our finding that 5 IU/kg IV-OT significantly increased CSF OT is a novel finding that has interesting implications for the interpretation of human studies that employed IV-OT. There have been reports in humans of behavioral changes following treatment with IV-OT, including an amelioration of some symptoms of ASD (Hollander et al., 2003, 2007), although one recent paper detected no behavioral differences after a low dose of IV-OT (Quintana et al., 2015). Peripheral OT in circulation is too large to effectively cross the blood–brain barrier (BBB) and access neural structures (Ermisch et al., 1985), but we found that our highest dose of IV-OT caused significant elevation in CSF OT over time. It has also been suggested that cell bodies in circumventricular organs like the neurohypophysis, where OT is synthesized and secreted into the bloodstream, can be reached by circulating peptides like OT (Ermisch et al., 1985). Thus, there is a potential feed-back mechanism whereby peripheral elevations in OT could trigger the release of endogenous OT by acting on oxytocinergic cells in the neurohypophysis. This theoretical mechanism has not yet been directly tested to our knowledge, but recent publications have also suggested this potential mechanism of action for plasma OT to affect the brain (Churchland and Winkielman, 2012; Macdonald and Feifel, 2013; Quintana et al., 2014). Based on this idea, the fact that the highest dose of IV-OT (5 IU/kg) in the current study significantly elevated levels of CSF OT could have occurred in one of two ways. First, elevated peripheral OT from the IV dose is reaching the CSF and being detected directly in our samples, which is unlikely given previous studies. Alternatively the increased levels of OT in circulation after IV-OT treatment is triggering OT neurons in the neurohypophysis to release OT into the central nervous system, and this endogenously released OT is being detected in the CSF samples in our study. It is unclear whether exogenous or endogenous OT being detected, so further research using tagged, exogenous OT is needed to disentangle the mechanism. In either case, if peripheral treatment with OT can increase central levels of OT, then it could explain the findings of our study as well as the behavioral changes reported in humans following IV-OT treatment. Finally, one outstanding question in the field of OT research is whether the levels of OT in circulation in the bloodstream are predictive of the levels of OT in CSF. If it is the case that these values are significantly related, then blood measures of OT can be used to infer relative levels of OT in the central nervous system, which bypasses the need for a spinal tap. This question has been investi-

S.M. Freeman et al. / Psychoneuroendocrinology 66 (2016) 185–194

gated in two studies in humans. One study did not find a significant association between levels of OT in plasma and in CSF (Kagerbauer et al., 2013), but the other study did find that plasma OT levels were significantly predictive of CSF levels of OT (Carson et al., 2015). We found that plasma OT does not predict CSF OT, which supports the former study in humans and is also supported by the findings of a previous study in macaques that concluded that the regulation of OT levels in plasma and CSF are functionally independent (Amico et al., 1990). Thus, evidence in macaques, at least, suggests that plasma levels of OT cannot be used as a reference for concurrent CSF levels of OT in the central nervous system. 4.3. Methodological comparisons across NHP studies While there are several methodological similarities among the current study and previous studies (Chang et al., 2012; Dal Monte et al., 2014; Modi et al., 2014), there are many notable differences across these studies (Table 2) that make it difficult to directly compare their results. These differences include: (1) the sedation state during dosing, (2) the intranasal delivery method, (3) the dose delivered, (4) the number and spacing of sampling time points, (5) the location of the CSF collection, (6) the type of OT assay used, and (7) whether the samples were extracted or not prior to assay. Many of these experimental variables likely contribute to the variation in the results reported across studies, which will be discussed briefly below. The current study is the first of this kind to use awake rhesus macaques, which were all extensively chair-trained and habituated to tolerate being dosed with a nasal spray. In doing so, we were uniquely capable of studying the efficacy of an OT nasal spray in a primate whose breathing, posture, and metabolism are normal and unaffected by sedatives or anesthesia, which more closely parallels the dosing regime used in human studies. Sedation is also required in order to acquire a CSF sample in monkeys via a lumbar or cervical puncture. All three of the previous studies to measure CSF OT after IN-OT or IV-OT treatment took CSF samples in anesthetized animals. It is unknown whether anesthesia affects the pharmacokinetics of OT in CSF, but the requirement of sedation in order to collect CSF drastically limits the frequency of CSF sampling. In order to complete a sampling time course for a study, it is necessary to either perform repeated CSF taps to acquire each time point after a wash-out period or to maintain the animal on extended anesthesia throughout the entire time course of sampling. Awake CSF sampling in macaques, which is made possible through surgical implantation of an intrathecal catheter, reduces the number of individual spinal punctures, reduces the risk of blood contamination of the sample, reduces the frequency and duration of anesthesia usage, and improves the quality of the data by capturing the pharmacodynamic profile of the drug of interest without sedatives or anesthetics on board. Our use of chronic implanted CSF catheters in combination with extensive behavioral habituation made it possible for the first time to study the pharmacokinetics of exogenous OT treatment in an awake nonhuman primate. There are also differences in the CSF sampling site across studies. Of the four studies to collect CSF from macaques after OT treatment, three studies (present study included) collected CSF from the cervical region (Chang et al., 2012; Dal Monte et al., 2014), but one used a lumbar location (Modi et al., 2014). This variation could contribute to differences in CSF OT across studies due to the increasing cervical-to-lumbar gradient in baseline CSF OT in macaques (Amico et al., 1989). Furthermore, there is also a circadian rhythm in CSF OT in macaques, which becomes more pronounced in lumbar samples compared to thoracic or cervical samples (Amico et al., 1989). The CSF sampling in the current study, as well Dal Monte et al. (2014), was performed at the same time of day across animals and experimental conditions, but the other two studies did not mention

193

whether or not they controlled for time of day. Thus it is important that future studies examining CSF OT are consistent in the location along the spinal column where samples will be drawn and in the time of day when sampling takes place. When assaying for OT in plasma samples, there has been great controversy over whether or not to extract the samples. A recent analysis of 47 publications that reported human baseline plasma OT levels found that sample extraction dramatically reduces variation in OT measurements across samples and across studies, although it also reduces the mean pg/mL measured in the assay (Christensen et al., 2014). Proponents of sample extraction argue that the elevated OT levels reported using non-extracted samples represent nonspecific binding of the antibody to other molecules present in the plasma. Thus, by removing these competing molecules by extracting OT from the samples, they achieve more specific measurements of OT. In contrast, proponents of assaying unextracted plasma samples argue that extraction causes artificial loss of OT, which results in OT values from extracted samples to fall within an artificially normalized range. It is also possible that there may be other bioactive forms OT (metabolites or conjugates) that are present in plasma that are detected by the assay’s antibody, which are lost in the extraction process (Martin and Carter, 2013). In either case, the results of the current study, which did not use extraction, are still valid to compare with one another within this study’s experimental groups. Until a consensus is reached on whether or not to extract OT from plasma or CSF samples, or both, and whether ELISA or RIA is preferable, the field will not be able to compare results across studies using these different methods. 4.4. Conclusion It is apparent that there is a need for methodological consistency across studies and more frequent CSF sampling in order to reliably characterize the effect of exogenous OT treatment on endogenous OT levels in a living primate system. The results of our study do not directly support any one potential mechanism for the action of IN-OT, but we feel that IN-OT and IV-OT may be impacting central levels of OT via feedback to the hypothalamus, which is triggering endogenous release of OT. Future studies are needed to disentangle this highly relevant research question. Furthermore, while it is important for future research involving nonhuman primates to determine which method of intranasal OT administration (aerosolized vs nasal spray) is the most effective for increasing central levels of OT in nonhuman primate species, it is also critical to use nonhuman primate models to better evaluate the effectiveness of the delivery method most commonly used in human studies and clinical use—the nasal spray. Thus, the current study sought to replicate and expand upon the results of previous studies by establishing a non-human primate model that can provide a stable platform for carrying out serial sampling from the central nervous system and peripheral vasculature concurrently. Conflict of interest The authors from Valley Biosystems, Inc. (S.S., P.C.A., D.S., G.G.C.H., and J.A.R.) declare that there are no other conflicts of interest. K.L.B. and S.M.F. report no conflicts of interest. Contributors All authors assisted with the designed the study. S.S., P.C.A., D.S., G.G.H., and J.A.R. carried out the animal experiments. S.M.F. performed the assays. S.M.F. and K.L.B. analyzed the data. S.M.F. was the first author of the manuscript, with involvement and approval from all authors.

194

S.M. Freeman et al. / Psychoneuroendocrinology 66 (2016) 185–194

Funding This work was supported by funding from Valley Biosystems, Inc. with no involvement from other organizations. Acknowledgement We would like to thank Trenton C. Simmons for assistance with the oxytocin bioassays. References Amico, J.A., Levin, S.C., Cameron, J.L., 1989. Circadian rhythm of oxytocin in the cerebrospinal fluid of rhesus and cynomolgus monkeys: effects of castration and adrenalectomy and presence of a caudal-rostral gradient. Neuroendocrinology 50, 624–632. Amico, J.A., Challinor, S.M., Cameron, J.L., 1990. Pattern of oxytocin concentrations in the plasma and cerebrospinal fluid of lactating rhesus monkeys (Macaca mulatta): evidence for functionally independent oxytocinergic pathways in primates. J. Clin. Endocrinol. Metab. 71, 1531–1535, http://dx.doi.org/10.1210/ jcem- 71-6-1531. Andari, E., Duhamel, J.-R., Zalla, T., Herbrecht, E., Leboyer, M., Sirigu, A., 2010. Promoting social behavior with oxytocin in high-functioning autism spectrum disorders. Proc. Natl. Acad. Sci. U. S. A. 107, 4389–4394, http://dx.doi.org/10. 1073/pnas.0910249107. Bales, K.L., Kramer, K.M., Hostetler, C.M., Capitanio, J.P., Mendoza, S.P., 2005. Validation of oxytocin and vasopressin blood assay for primates: what can blood tell us? Presented at the American Journal of Primatology, 73. Burns, J., Scheinfeld, M.H., 2013. Back to the scanner: expected and unexpected imaging findings following spinal puncture and access. Emerg. Radiol. 20, 291–297, http://dx.doi.org/10.1007/s10140-013-1111-2. Burri, A., Heinrichs, M., Schedlowski, M., Kruger, T.H.C., 2008. The acute effects of intranasal oxytocin administration on endocrine and sexual function in males. Psychoneuroendocrinology 33, 591–600, http://dx.doi.org/10.1016/j. psyneuen.2008.01.014. Carson, D.S., Berquist, S.W., Trujillo, T.H., Garner, J.P., Hannah, S.L., Hyde, S.A., Sumiyoshi, R.D., Jackson, L.P., Moss, J.K., Strehlow, M.C., Cheshier, S.H., Partap, S., Hardan, A.Y., Parker, K.J., 2015. Cerebrospinal fluid and plasma oxytocin concentrations are positively correlated and negatively predict anxiety in children. Mol. Psychiatry 20, 1085–1090. Challinor, S.M., Cameron, J.L., Amico, J.A., 1992. Pulses of oxytocin in the cerebrospinal fluid of rhesus monkeys. Horm. Res. 37, 230–235. Chang, S.W.C., Barter, J.W., Ebitz, R.B., Watson, K.K., Platt, M.L., 2012. Inhaled oxytocin amplifies both vicarious reinforcement and self reinforcement in rhesus macaques (Macaca mulatta). Proc. Natl. Acad. Sci. U. S. A. 109, 959–964, http://dx.doi.org/10.1073/pnas.1114621109. Christensen, J.C., Shiyanov, P.A., Estepp, J.R., Schlager, J.J., 2014. Lack of association between human plasma oxytocin and interpersonal trust in a Prisoner’s dilemma paradigm. PLoS One 9, e116172, http://dx.doi.org/10.1371/journal. pone.0116172. Churchland, P.S., Winkielman, P., 2012. Modulating social behavior with oxytocin: how does it work? What does it mean? Horm. Behav. 61, 392–399, http://dx. doi.org/10.1016/j.yhbeh.2011.12.003. Dal Monte, O., Noble, P.L., Turchi, J., Cummins, A., Averbeck, B.B., 2014. CSF and blood oxytocin concentration changes following intranasal delivery in macaque. PLoS One 9 e103677, http://dx.doi.org/10.1371/journal.pone. 0103677. Domes, G., Heinrichs, M., Michel, A., Berger, C., Herpertz, S.C., 2007. Oxytocin improves mind-reading in humans. Biol. Psychiatry 61, 731–733, http://dx.doi. org/10.1016/j.biopsych.2006.07.015. Domes, G., Heinrichs, M., Kumbier, E., Grossmann, A., Hauenstein, K., Herpertz, S.C., 2013. Effects of intranasal oxytocin on the neural basis of face processing in autism spectrum disorder. Biol. Psychiatry 74, 164–171, http://dx.doi.org/10. 1016/j.biopsych.2013.02.007. Du, Y., Fan, T.-Y., Tan, Y., Xiong, Z., Wang, Z., 2010. Seasonal changes in the reproductive physiology of female rhesus macaques (Macaca mulatta). J. Am. Assoc. Lab. Anim. Sci. 49, 289–293. Ermisch, A., Rühle, H.J., Landgraf, R., Hess, J., 1985. Blood–brain barrier and peptides. J. Cereb. Blood Flow Metab. 5, 350–357, http://dx.doi.org/10.1038/ jcbfm.1985.49. Falconer, J., Mitchell, M.D., Mountford, L.A., Robinson, J.S., 1980. Plasma oxytocin concentrations during the menstrual cycle in the rhesus monkey, Macaca mulatta. J. Reprod. Fertil. 59, 69–72. Gaiser, R., 2006. Postdural puncture headache. Curr. Opin. Anaesthesiol. 19, 249–253, http://dx.doi.org/10.1097/01.aco.0000192809.71408.ba. Gossen, A., Hahn, A., Westphal, L., Prinz, S., Schultz, R.T., Gründer, G., Spreckelmeyer, K.N., 2012. Oxytocin plasma concentrations after single intranasal oxytocin administration—a study in healthy men. Neuropeptides 46, 211–215, http://dx.doi.org/10.1016/j.npep.2012.07.001.

Guastella, A.J., Mitchell, P.B., Dadds, M.R., 2008. Oxytocin increases gaze to the eye region of human faces. Biol. Psychiatry 63, 3–5, http://dx.doi.org/10.1016/j. biopsych.2007.06.026. Guastella, A.J., Einfeld, S.L., Gray, K.M., Rinehart, N.J., 2010. Intranasal oxytocin improves emotion recognition for youth with autism spectrum disorders. Biol. Psychiatry 67, 692–694. Hollander, E., Novotny, S., Hanratty, M., Yaffe, R., DeCaria, C.M., Aronowitz, B.R., Mosovich, S., 2003. Oxytocin infusion reduces repetitive behaviors in adults with autistic and Asperger’s disorders. Neuropsychopharmacology 28, 193–198, http://dx.doi.org/10.1038/sj.npp.1300021. Hollander, E., Bartz, J., Chaplin, W., Phillips, A., Sumner, J., Soorya, L., Anagnostou, E., Wasserman, S., 2007. Oxytocin increases retention of social cognition in autism. Biol. Psychiatry 61, 498–503, http://dx.doi.org/10.1016/j.biopsych. 2006.05.030. Hurlemann, R., Patin, A., Onur, O.A., Cohen, M.X., Baumgartner, T., Metzler, S., Dziobek, I., Gallinat, J., Wagner, M., Maier, W., Kendrick, K.M., 2010. Oxytocin enhances amygdala-dependent, socially reinforced learning and emotional empathy in humans. J. Neurosci. 30, 4999–5007, http://dx.doi.org/10.1523/ JNEUROSCI.5538-09.2010. Kagerbauer, S.M., Martin, J., Schuster, T., Blobner, M., Kochs, E.F., Landgraf, R., 2013. Plasma oxytocin and vasopressin do not predict neuropeptide concentrations in human cerebrospinal fluid. J. Neuroendocrinol. 25, 668–673, http://dx.doi. org/10.1111/jne.12038. Kirkpatrick, M.G., Francis, S.M., Lee, R., de Wit, H., Jacob, S., 2014. Plasma oxytocin concentrations following MDMA or intranasal oxytocin in humans. Psychoneuroendocrinology 46, 23–31, http://dx.doi.org/10.1016/j.psyneuen. 2014.04.006. Kosfeld, M., Heinrichs, M., Zak, P.J., Fischbacher, U., Fehr, E., 2005. Oxytocin increases trust in humans. Nature 435, 673–676, http://dx.doi.org/10.1038/ nature03701. Levine, A., Zagoory-Sharon, O., Feldman, R., Weller, A., 2007. Oxytocin during pregnancy and early postpartum: individual patterns and maternal-fetal attachment. Peptides 28, 1162–1169, http://dx.doi.org/10.1016/j.peptides. 2007.04.016. Macdonald, K., Feifel, D., 2013. Helping oxytocin deliver: considerations in the development of oxytocin-based therapeutics for brain disorders. Front. Neurosci. 7, 35, http://dx.doi.org/10.3389/fnins.2013.00035. Martin, W.L., Carter, C.S., 2013. Oxytocin and vasopressin are sequestered in plasma. In: Presented at the 10th World Congress of Neurohypopheseal Hormones, Bristol, UK. Mens, W.B., Witter, A., van Wimersma Greidanus, T.B., 1983. Penetration of neurohypophyseal hormones from plasma into cerebrospinal fluid (CSF): half-times of disappearance of these neuropeptides from CSF. Brain Res. 262, 143–149. Modi, M.E., Connor-Stroud, F., Landgraf, R., Young, L.J., Parr, L.A., 2014. Aerosolized oxytocin increases cerebrospinal fluid oxytocin in rhesus macaques. Psychoneuroendocrinology 45, 49–57. Prata, N., Bell, S., Weidert, K., 2013. Prevention of postpartum hemorrhage in low-resource settings: current perspectives. Int. J. Womens Health 5, 737–752, http://dx.doi.org/10.2147/IJWH.S51661. Quintana, D.S., Alvares, G.A., Hickie, I.B., Guastella, A.J., 2014. Do delivery routes of intranasally administered oxytocin account for observed effects on social cognition and behavior? A two-level model. Neurosci. Biobehav. Rev. 49C, 182–192, http://dx.doi.org/10.1016/j.neubiorev.2014.12.011. Quintana, D.S., Westlye, L.T., Rustan, Ø.G., Tesli, N., Poppy, C.L., Smevik, H., Tesli, M., Røine, M., Mahmoud, R.A., Smerud, K.T., Djupesland, P.G., Andreassen, O.A., 2015. Low-dose oxytocin delivered intranasally with breath powered device affects social-cognitive behavior: a randomized four-way crossover trial with nasal cavity dimension assessment. Transl. Psychiatry 5, e602, http://dx.doi. org/10.1038/tp.2015.93. Scheele, D., Striepens, N., Güntürkün, O., Deutschländer, S., Maier, W., Kendrick, K.M., Hurlemann, R., 2012. Oxytocin modulates social distance between males and females. J. Neurosci. 32, 16074–16079, http://dx.doi.org/10.1523/ JNEUROSCI.2755-12.2012. Scheele, D., Wille, A., Kendrick, K.M., Stoffel-Wagner, B., Becker, B., Güntürkün, O., Maier, W., Hurlemann, R., 2013. Oxytocin enhances brain reward system responses in men viewing the face of their female partner. Proc. Natl. Acad. Sci. U. S. A. 110, 20308–20313, http://dx.doi.org/10.1073/pnas.1314190110. Striepens, N., Kendrick, K.M., Hanking, V., Landgraf, R., Wüllner, U., Maier, W., Hurlemann, R., 2013. Elevated cerebrospinal fluid and blood concentrations of oxytocin following its intranasal administration in humans. Sci. Rep. 3, 3440, http://dx.doi.org/10.1038/srep03440. Stubbs, T.M., 2000. Oxytocin for labor induction. Clin. Obstet. Gynecol. 43, 489–494. Wilder-Smith, E., Kothbauer-Margreiter, I., Lämmle, B., Sturzenegger, M., Ozdoba, C., Hauser, S.P., 1997. Dural puncture and activated protein C resistance: risk factors for cerebral venous sinus thrombosis. J. Neurol. Neurosurg. Psychiatr. 63, 351–356. Young, L.J., Flanagan-Cato, L.M., 2012. Editorial comment: oxytocin, vasopressin and social behavior. Horm. Behav. 61, 227–229, http://dx.doi.org/10.1016/j. yhbeh.2012.02.019.