1. Neuroscience
Download icon

Oxytocin-mediated social enrichment promotes longer telomeres and novelty seeking

  1. Jamshid Faraji
  2. Mitra Karimi
  3. Nabiollah Soltanpour
  4. Alireza Moharrerie
  5. Zahra Rouhzadeh
  6. Hamid lotfi
  7. S Abedin Hosseini
  8. S Yaghoob Jafari
  9. Shabnam Roudaki
  10. Reza Moeeini  Is a corresponding author
  11. Gerlinde AS Metz  Is a corresponding author
  1. University of Lethbridge, Canada
  2. Golestan University of Medical Sciences, Iran
  3. Exceptional Education Organization, Iran
  4. Babol University of Medical Sciences, Iran
  5. Islamic Azad University, Iran
  6. Avicenna Institute of Neuroscience, Iran
Research Article
  • Cited 1
  • Views 818
  • Annotations
Cite this article as: eLife 2018;7:e40262 doi: 10.7554/eLife.40262

Abstract

The quality of social relationships is a powerful determinant of lifetime health. Here, we explored the impact of social experiences on circulating oxytocin (OT) concentration, telomere length (TL), and novelty-seeking behaviour in male and female rats. Prolonged social housing raised circulating OT levels in both sexes while elongating TL only in females. Novelty-seeking behaviour in females was more responsive to social housing and increased OT levels than males. The OT antagonist (OT ANT) L-366,509 blocked the benefits of social housing in all conditions along with female-specific TL erosion and novelty-seeking deficit. Thus, females seem more susceptible than males to genetic and behavioural changes when the secretion of endogenous OT in response to social life is interrupted. Social enrichment may, therefore, provide a therapeutic avenue to promote stress resiliency and chances of healthy aging across generations.

https://doi.org/10.7554/eLife.40262.001

Introduction

The quality and quantity of social relationships in the modern era are rapidly changing, with increasing social isolation, and many individuals no longer living in extended families or missing close confidants. Restricted social relationships can endanger public health and increase mortality rate (Mikosz et al., 2016). Males and females, both human and animal, respond differently to social interactions. Differences may range from changes in neurobiological processes (Mikosz et al., 2016; Stack et al., 2010; Ngun et al., 2011; Cherif et al., 2003) to displaying distinctive gender-specific behaviours (Szell and Thurner, 2013; Jukka-Pekka Onnela et al., 2014; Palchykov et al., 2012; Trainor et al., 2011). Genetic and hormonal determinants regulate social behaviours (Choleris et al., 2018) and dictate morphological and behavioural variations in males and females. Robust sexual dimorphism was found in the neuropeptide oxytocin (OT) receptor density throughout the rat brain (Smith et al., 2017) and in OT receptor signaling in the medial prefrontal cortex (mPFC), which may modulate social interactions in females in response to OT (Nakajima et al., 2014). Primarily synthesized in the paraventricular nucleus (PVN) and supraoptic nucleus of the hypothalamus (Jurek and Neumann, 2018), OT is a key hormonal correlate of social behaviours in mammals (Anacker and Beery, 2013; Knobloch and Grinevich, 2014). OT exposure promotes social behaviours (Gamer et al., 2010; Guastella et al., 2008; Andari et al., 2010; Lim and Young, 2006; Bernaerts et al., 2017; McGraw and Young, 2010) and facilitates adaptive responses to stressors and stress resiliency (Olff et al., 2013). OT has also been shown to prevent premature aging of muscle tissue and facilitates muscle regeneration (Elabd et al., 2014). Hence, it can be hypothesized that OT exposure influences the chances of healthy aging.

A specific cellular index which differs according to age and sex indicates changes in telomere length (TL). Telomeres are DNA-repetitive nucleotide segments at the ends of each chromosome in mammals that protect genetic material from degradation during somatic cell division. Telomeric DNA in humans naturally shortens with age, and therefore TL represents a sensitive indicator of tissue-specific cellular aging (Puterman et al., 2016). As TL decreases along with proliferation in each cell division, telomere shortening can be considered a measure of biological aging (Sahin and DePinho, 2010; Blackburn, 2005). A variety of influences such as distress (Epel et al., 2004; Epel et al., 2010), childhood adversity (Puterman et al., 2016), interpersonal instability (Drury et al., 2014), social disadvantages and deprivation (Theall et al., 2013), and social isolation (Aydinonat et al., 2014) have been shown to be correlated with TL erosion. TL has also been shown to be positively associated with psychosocial factors, such as conscientiousness (Edmonds et al., 2015), optimism (Schutte et al., 2016), social support (Uchino et al., 2012), and social-pair mate choice (Johnsen et al., 2017).

It appears that females generally have longer telomeres than males (Gardner et al., 2014; Barrett and Richardson, 2011), and shorter telomeres in males are correlated with reduced social support (Zalli et al., 2014). This has led to the speculation that social experiences possibly make females more susceptible to telomere elongation than males, a lifestyle-dependent process that, in concert with sex-specific hormonal correlates results in longevity gender gaps (Merrill et al., 2017). Also, females show facilitated social learning compared to males (Ervin et al., 2015), along with estrogenic control of OT and OT receptor activity (Anacker and Beery, 2013). Accordingly, we have recently shown that prolonged social experiences change brain structure and behaviours in a sexually dimorphic manner in rats (Faraji et al., 2018).

Here, we investigated whether social experiences modulate genetic and behavioural repertoires through OT in male and female rats. Whereas OT facilitates novelty-seeking behaviour and TL, the OT antagonist (OT ANT) L-366,509 in socially raised male and female rats intermittently reduced OT secretion. The results suggest that extended social experiences modulate novelty seeking and TL through OT in a sex-dependent manner, with females being more susceptible than males.

Results

Experiment 1

Social experience increased plasma OT concentration in females

Figure 1A shows changes in plasma OT concentrations across different groups as a function of housing condition (standard, males: n = 23, females: n = 21; social, males: n = 22, females: n = 22). A significant effect of housing condition was observed in terms of circulating OT concentration (social vs. standard; F1,86=24.85, p ≤ 0.001) and Group (social males and females vs. standard males and females; F3,84=13.69, p ≤ 0.001), suggesting that long-term social experiences had a significant impact on plasma OT in socially raised animals (98.82 ± 3.94 vs. 71.00 ± 3.94 pmol/L). Post-hoc Tukey comparisons also indicated that social females showed higher OT concentration than standard females (108.55 ± 5.25 vs. 80.65 ± 5.37 pmol/L; p ≤ 0.002). Furthermore, a marginal difference was observed between social males and females (89.09 ± 5.25 vs. 108.55 ± 5.25 pmol/L; p = 0.05; Post-hoc Tukey). No effect was found in terms of litter.

Housing conditions influence plasma OT and telomere length.

(A) Social enrichment altered plasma OT levels in socially raised animals. Social females showed higher OT concentration than social males (n = 21–23/group). (B) Social experience increased telomere length only in females (n = 23) when compared with standard rats (n = 27–28) and their social male counterparts (n = 31). Asterisks indicate significant differences: *p ≤ 0.05; **p ≤ 0.01; one-way ANOVA. Filled circles: mean OT concentrations in individual rats. Horizontal bars: mean OT concentrations in each group. pM: picoMolar, OT: oxytocin, Tel: telomere, bp: basepair.

https://doi.org/10.7554/eLife.40262.002

Social experience increased telomere length (TL) in females but not males

Examination of TL in ear notch skin cells showed that socially raised rats (males: n = 31, females: n = 23) had greater telomere length than standard animals (males: n = 28, females: n = 27; F1,107 = 24.91, p ≤ 0.001). Also, a significant effect of group was observed (F3,105 = 65.84, p ≤ 0.001) while social female rats displayed greater TL than standard rats and their social male counterparts (all p ≤ 0.001; Post-hoc Tukey). No difference was found between social males and standard males and females (all p ≥ 0.05; Post-hoc Tukey; Figure 1B). No effect was found in terms of litter.

Novelty-seeking behaviour in females was significantly influenced by social experiences

An illustration of the corridor field task (CFT) with and without central object along with paths taken by rats in standard and social groups is shown in Figure 2A and B. No-central object CFT: panel (C) compares the exploratory behaviour in both groups (standard: n = 33; social: n = 39) within different zones of the no-central object CFT. There was a significant main effect of housing condition (two levels) in terms of the time spent in each zone (corridor: F1,70 = 39.60, p ≤ 0.001; open: F1,70 = 20.88, p ≤ 0.001; central: F1,70 = 16.43, p ≤ 0.001). When compared with the standard group, social rats spent less time in the corridor zone (243.02 ± 8.8 s vs. 325.60 ± 9.6 s) and more time in the open (177.76 ± 7.9 s vs. 123.84 ± 8.6 s) and central zones (59.20 ± 4.7 s vs. 30.54 ± 5.2 s). No effect was found in terms of litter, and no significant interactions between factors were found, except for sex × litter (p ≤ 0.03). Furthermore, between-subjects analysis indicated a significant effect of Group (four levels) for all zones (three levels; all p = 0.000) showing that social females spent significantly less times in the corridor zone than any other group (all p ≤ 0.02; Post-hoc Tukey) and made more visits to the open zone than standard rats (all p ≤ 0.05; Post-hoc Tukey). No significant difference was found between social males and females in open zone exploration (p ≤ 0.6; Post-hoc Tukey). Social females also spent more time in the central zone when compared with standard housed animals and their male counterparts (all p ≤ 0.01; Post-hoc Tukey).

Social experience alters novelty-seeking behaviour in the corridor field task (CFT).

(A and B) Illustration of the no-central and central-object CFT protocol along with samples of paths taken by rats in standard and social groups. (C and D) Social life affected novelty-seeking behaviour in a sexually dimorphic manner. Socially reared females (n = 22) explored the open and central zones more than their social male counterparts (n = 17) or standard group (n = 16 and 17). (E and F) The rate of changes (ROC) indicated the most profound impact on social females than any other group from no-central object to central-object versions of the CFT in the open and central zones. Also, a significant regression equation was found only in social females (n = 22) where OT concentration significantly predicts the exploration in the corridor and central zones (G) when central object was not presented. (H) Analysis of linear regression indicated significant regression equations for the time spent in open and central zones only in social females (n = 22) by which the increased plasma OT levels significantly predicted exploration time in CFT when the central object was presented. (I) The TL elongation, also, was significantly associated with an increase in plasma OT level only in social females (n = 22).

https://doi.org/10.7554/eLife.40262.005

Central-object CFT: A significant effect of housing condition was observed in terms of the time spent in corridor (F1,70 = 154.88, p ≤ 0.001), open (F1,70 = 112.75, p ≤ 0.001) and central zones (F1,70 = 65.65, p ≤ 0.001) of the CFT with central object (Figure 2D). Again, social animals explored the corridor zone less than standard rats (175.23 ± 8 s vs. 322.93 ± 8.7 s) and spent more exploration time in the open (196.23 ± 5.27 s vs. 113.51 ± 5.7 s) and central zones (108.53 ± 5.4 s vs. 43.54 ± 5.9 s) when compared with standard rats. Also, a significant effect of group was found for the corridor (F3,68 = 115.42, p ≤ 0.001), open (F3,68 = 42.31, p ≤ 0.001), and central zones (F3,68 = 17.01, p ≤ 0.001), indicating that social females explored the corridor zone less and other zones more than any other group, even their male counterparts (all p ≤ 0.001; Post-hoc Tukey). No effect was found in terms of litter or interaction between factors (all p ≥ 0.05). Thus, social experience appeared to affect novelty-seeking explorative behaviours in the CFT in a sexually dimorphic manner, by which socially reared females explored the open and central zones more than social males and standard females. This behavioural flexibility was also supported by additional ROC analysis showing that social females experienced greater changes in novelty-seeking behaviours than other groups from no-central object to central-object versions of the CFT in the open and central zones (Figure 2E and F).

Simple linear regression analyses were conducted to predict novelty-seeking behaviour and TL based on plasma OT concentration. Plasma OT concentration and exploratory behaviour in no-central object CFT: a significant regression equation was found (F1,20 = 11.92, p ≤ 0.003) only in social females (n = 22) with an R2 = 0.373, through which the OT concentration significantly predicts the exploration in the corridor zone (420.34 + −1.84). The observed regression equation for exploration in the central zone was also significant (F1,20 = 9.08, p ≤ 0.007; R2 = 0.312; Figure 2G). Therefore, only in social females does plasma OT concentration predict a particular profile of CFT exploration through which in the presence of enhanced OT concentration, the corridor zone was less explored whereas the central zone was more explored in social females when compared to other groups.

Plasma OT concentration and exploratory behaviour in central object CFT: analysis of linear regression indicated significant regression equations for the time spent in open (F1,20 = 17.33, p ≤ 0.000; R2 = 0.464) and central (F1,20 = 30.01, p ≤ 0.000; R2 = 0.60) zones by which the increased plasma OT levels only in social females significantly predicted exploration time (Figure 2H). No significant changes were observed for the time spent in the corridor zone (p = 0.07; R2 = 0.148).

Plasma OT concentration and TL: a significant regression equation was found (F1,20 = 18.88, p ≤ 0.000; R2 = 0.486) only in social females through which the predicted TL was equal to 3816.88 + 8.53 (OT concentration) bp when OT concentration was measured in pM (Figure 2I). Thus, TL elongation was associated with higher plasma OT level only in social females.

Experiment 2

The OT antagonist L-366,509 reduced circulating plasma OT levels

Circulating OT concentration showed a significant effect of group (n = 10–13/group, F7,83 = 14.64, p ≤ 0.001), whereas injection of the OT antagonist L-366,509 resulted in a significant decrease in all ANT groups (all p ≤ 0.05, Post-hoc Tukey) except for standard OT ANT males (p = 0.475, Post-hoc Tukey; Figure 3A and B). Also, a significant effect of housing condition (n = 45 and 46, F1,89 = 9.72, p ≤ 0.002) was observed suggesting that social rats still had greater concentration of the plasma OT when compared with standard animals (80.55 ± 2.72 vs. 68.43±2.75 pmol/L). A significant effect of sex (n = 44 and 47, F1,89 = 11.08, p ≤ 0.001) indicated greater levels of OT in females than males (81.17 ± 2.76 vs. 68.36±2.67 pmol/L).

OT antagonist L-366,509 blocks the social experience phenotype in plasma OT concentration and telomere length.

(A and B) OT antagonist administration reduced plasma OT concentration in all OT ANT groups, except for standard OT ANT males (n = 10–13/group). Social control females displayed higher concentration of plasma OT than any other group. No difference was found between social control females and males. (C and D) OT antagonist L-366,509 reduced telomere length in social females. Asterisks indicate significant differences: *p ≤ 0.05; **p ≤ 0.01; one-way ANOVA. Filled circles: mean OT concentrations in individual rats. Horizontal bars: mean OT concentrations in each group. ANT: antagonist, bp: basepair, OT: oxytocin, pM: picoMolar, Tel: telomere.

https://doi.org/10.7554/eLife.40262.007

Social control females displayed higher plasma OT concentration than any other group (all p ≤ 0.05, Post-hoc Tukey) except for social control males (p = 0.07, Post-hoc Tukey). There were also significant differences in plasma OT with social control males having significantly higher OT concentration than other groups (p ≤ 0.05, Post-hoc Tukey) except for standard control females and social OT ANT females (all p ≥ 0.05, Post-hoc Tukey). A significant effect of litter was also found in terms of OT concentration (F10,81 = 11.46, p ≤ 0.03), in which litters 4 and 7 displayed higher OT levels than other litters (all p ≤ 0.05, Post-hoc Tukey). Thus, the OT antagonist administration had a significant impact on the plasma OT concentration in OT ANT groups, particularly in females regardless of their housing condition.

OT antagonist in social females significantly reduced telomere length

Despite significant effects of housing condition (n = 48 and 45/group, F1,91 = 4.80, p ≤ 0.03), sex (n = 47 and 46/group, F1,91 = 14.63, p ≤ 0.001), and group (n = 10–13/group, F7,85 = 7.06, p ≤ 0.001), the administration of the OT antagonist L-366,509 had no significant effect on the TL in OT ANT groups, except for social OT ANT females (Figure 3C and D). Only social females had greater TL than other groups (all p ≤ 0.05, Post-hoc Tukey), replicating the results of Experiment 1. In turn, the OT antagonist abolished the benefits of social housing by reducing circulating OT and reducing TL in females only. No effect of litter was found.

OT antagonist significantly influenced novelty-seeking behaviour in all groups

Figure 4A and C compare novelty-seeking behaviour in the CFT in animals with and without OT antagonist L-366,509 treatment. No-central object CFT: administration of L-366,509 affected novelty-seeking behaviour within all zones of the CFT across groups. A significant effect of housing condition (F1,53 = 11.63, p ≤ 0.001) indicated that socially raised rats (n = 27) still explored the corridor zone less than standard animals (n = 28; 276.77 ± 8.99 s vs. 319.75 ± 8.82 s). However, when compared with standard rats, social rats explored open and central zones of the CFT more frequently (open: 153.66 ± 6.48 s vs. 119.10 ± 6.3 s; central: 49.55 ± 5.19 s vs. 41.14 ± 5.10 s). The group effect was also significant for all three zones of the CFT (all p ≤ 0.01). Only social rats, however, were significantly impacted by the OT antagonist in terms of exploratory behaviour. Essentially, social control females significantly explored the central zone more than standard control females, whereas their corridor time diminished (all p ≤ 0.01, Post-hoc Tukey). Social control females also spent less time in the corridor and more time in the open and central zones when compared with standard OT ANT females (all p ≤ 0.01, Post-hoc Tukey). Moreover, when compared with social OT ANT females, the social control females explored the corridor zone less and spent more time in the central zone (all p ≤ 0.05, Post-hoc Tukey). Thus, the OT antagonist L-366,509 changed the exploration pattern particularly in social females. There was no difference between social control males and females in terms of CFT exploration (all p ≥ 0.05, Post-hoc Tukey). When compared to standard control males, however, the social control males spent more time in the open zone (p ≤ 0.05, Post-hoc Tukey). The social control males also spent less time in the corridor zone but explored the open zone more than standard OT ANT males (all p ≤ 0.05, Post-hoc Tukey). No difference was observed between social control and social OT ANT males (all p ≥ 0.05, Post-hoc Tukey). All interactions between factors were insignificant, except for sex × litter and sex × litter × housing condition (all p ≤ 0.05). Thus, although exploration in the no-central object CFT version was somehow affected by L-366,509, CFT exploration in social females was more impacted by the OT antagonist than any other group.

Behavioural consequences of the administration of OT antagonist L-366,509 in the corridor field task (CFT).

(A and B) The OT antagonist affected novelty-seeking behaviour in all zones of the no-central object CFT (standard: n = 28; social: n = 27). (C and D) The OT antagonist had a significant impact on the social females’ exploration in the central-object CFT. Novelty-seeking behaviour in socially raised females was more influenced by reduced OT levels than any other group (standard: n = 27; social: n = 27). (E–H) Rate of changes (ROC) within no-central and central-object CFT in response to OT antagonist. Note that social OT ANT females experienced the fewest changes in novelty-seeking behaviour from no-central object to central-object versions of the CFT. Asterisk indicates significant differences: p ≤ 0.05; MANOVA. Symbols denote comparisons: social control females: * relative to standard control females, # relative to standard OT ANT females, $ relative to social OT ANT females; social control males: * relative to standard control males, # relative to standard OT ANT males. Error bars show ± SEM. OT: oxytocin, ANT: antagonist.

https://doi.org/10.7554/eLife.40262.010

Central-object CFT: Figure 4C and D illustrates novelty-seeking behaviour within the CFT with and without OT antagonist L-366,509. Again, a significant effect of housing condition indicated a different profile of novelty-seeking behaviour in social rats (n = 27). Socially raised rats explored the corridor zone less than standard animals (n = 27, F1,52 = 5.45, p ≤ 0.02). Social animals, however, explored the open zone more than standard rats (F1,52 = 10.97, p ≤ 0.002), whereas they showed no difference in the central zone (p ≥ 0.05). Also, significant effects of group for all zones (all p ≤ 0.05) showed that the OT antagonist L-366,509 had a greater impact on social female exploration (all p ≤ 0.05, Post-hoc Tukey). The inhibitory effect of L-366,509 on OT concentration, and consequently the novelty-seeking behaviour in socially raised males did not reach the levels of social females (all p ≤ 0.05, Post-hoc Tukey). No effect of litter, and no interactions between factors were found, except for litter × sex (p ≤ 0.04). Therefore, novelty-seeking behaviour in socially raised females was significantly more influenced by reduced OT levels than any other group. Further analysis of exploratory behaviour through the ROC also indicated that L-366,509 had greater impact on the novelty-seeking behaviour in social females in both no-central and central-object CFT (Figure 4E–H).

Discussion

The present rodent study indicates that persistent social experience is causally associated with increased OT level, elevated exploratory behaviour, and telomere elongation. These changes occurred in a sex-dependent manner to support three main conclusions: (i) social experience increases plasma OT levels, (ii) enhanced novelty-seeking behaviour and TL in socially housed females is mediated by higher OT levels, and (iii) social females respond to interrupted endogenous OT secretion with TL erosion and novelty-seeking deficit. Thus, OT is arguably involved in mediating beneficial effects of social experience on behaviour and TL. The findings reveal a dynamic interplay between social interactions and OT hormone to promote genetic and behavioural resiliency in female rats (Figure 5).

Representation of the experimental design and hypothetical mechanisms of social experience in rats.

Male and female rats were raised in either (a) standard- or (b) social-housing units for 84–90 d. (b1) Prolonged social housing (b2) increased telomere length in females (TL) while enhancing plasma oxytocin (OT) in both sexes. Novelty-seeking behaviour in females more than males was responsive to social housing. (b3) Higher OT levels amplify social bonding and interaction through enhanced sociality. (c) Social interaction modulates novelty-seeking behaviours, OT, and TL along with HPA axis activity as a function of sex hormone status. Other hypothetical mechanisms to modulate social experience-dependent behaviour and neuroplasticity may include neurotrophic factors, such as brain-derived neurotrophic factor (BDNF).

https://doi.org/10.7554/eLife.40262.012

Prolonged social experiences increase OT in a sex-specific manner

The pivotal role of OT in social behaviour has become a target of many experimental efforts (Churchland and Winkielman, 2012). However, the majority of data linking OT to social behaviours represent correlational conclusions (Barraza and Zak, 2009), or depict a hormone-to-behaviour approach in which subjects were exposed to OT, and then were monitored for OT-induced behavioural alterations (Bernaerts et al., 2017). Using a causal, behaviour-to-hormone approach in the present experiments, we examined the direct impact of prolonged social interactions across development upon OT-mediated effects. In previous discussions (Cerulo, 2009; Schilbach et al., 2013; Hari et al., 2015) social interaction was regarded as a critical contributor to health and longevity (Umberson and Montez, 2010; Yang et al., 2016) and it was assumed that OT is involved in social behaviours and social memory (Gamer et al., 2010; Bernaerts et al., 2017; Kirsch et al., 2005; Baumgartner et al., 2008; Kosfeld et al., 2005; Auyeung et al., 2015). The present findings expand these earlier studies by demonstrating that social rearing elevates plasma OT concentration, and this hormonal response to social experience appears more prominent in females than males. Experiment 2 replicated these findings from Experiment 1.

There are two possible mechanisms that may underlie the dynamic interaction between social experience and OT. First, social interaction is associated with persistent sensory stimulation during development which may promote hypothalamic-pituitary-adrenal (HPA) axis function and brain plasticity (Zucchi et al., 2014). This assumption is supported by a recently published report (Zheng et al., 2014) in which sensory experiences in mice promoted cortical development by up-regulation of synthesis and secretion of OT. Second, ongoing emotional sharing and social support foster prolonged social relationships (Beery and Kaufer, 2015) along with activation of the brain’s dopamine reward system (Rilling et al., 2002). Both, regulating stress responses and rewarding social interaction, are vital corollaries in OT-mediated social engagement.

While OT is produced both centrally and peripherally, social cues generally cause OT releases centrally in the brain, which in turn modulates OT-receptor activities in the socially relevant circuitry. Sex-specific differences (Joel et al., 2015) in cortical regions including mPFC revealed higher OT receptor density and signaling in females (Smith et al., 2017; Nakajima et al., 2014). Accordingly, compared with males, vertebrate females appear sensitized to social support and interactions (Beery and Kaufer, 2015; Ozbay et al., 2007) in terms of emotional expressions (Kring and Gordon, 1998) and enhanced social cognition (Gur et al., 2012). The neurophysiological sequelae (including changes in cardiovascular, neuroendocrine, and immune function) (Cohen and Wills, 1985; House et al., 1988; Lincoln et al., 2000) via providing social supports, reduce the impact of stress responses to promote stress resiliency, especially in females (Uchino, 2006; Charney, 2004). Social behaviours and OT are linked to stress regulation through contextual (e.g., the duration of social life or presence of a familiar individual) and inter-individual factors (e.g., sex) (Bartz et al., 2011). The present changes after social experience for 3 mo address both types of intervening factors resulting in sex-specific regulation of social behaviour and exacerbated hormonal responses in female.

Social experiences promote novelty-seeking behaviour

The present experiments provide evidence for the involvement of OT in exploratory and novelty-seeking behaviour (Kazlauckas et al., 2005) encouraged by the novel central object in the CFT. In agreement with our previous findings, female rats that experienced social enrichment across development showed extended exploration of the central areas in the CFT, thus increasing the radius of exploratory activity (Faraji et al., 2014). The latter indicates lowered stress response (Faraji et al., 2014), which agrees with the finding that socially raised female rats display lower HPA axis response and reduced anxiety-related behaviours (Faraji et al., 2018). Accordingly, intensified OT action through social enrichment enhances social bonding and reduces stress responses (Swain et al., 2014). Because OT may mediate sexual dimorphisms in regional neuroplasticity (Hillerer et al., 2014) and exert anxiolytic effects (Ayers et al., 2011), social experiences may affect brain anatomy and novelty-seeking behaviours especially in females.

OT-mediated enhancement in novelty-seeking behaviour may stem from the close relationship between OT receptors and central dopamine reward systems in mediating incentive response to novelty (Bardo et al., 1996). Although it has been suggested that females display generally lower levels of novelty seeking than males (Hughes, 1968), the present findings suggest that early life experiences and the endocrine consequences of social bonding may determine sex-dependent differences in novelty-seeking behaviours. Moreover, social relationships play key roles in the formation of emotional and social responsiveness that reduce aversive outcomes (e.g. fear and anger) in social interactions (Ozbay et al., 2007; Averbeck, 2010; Sippel et al., 2015). It appears that OT reciprocally influences the perceived social cues in a sex-dependent manner through these behavioural phenotypes (Olff et al., 2013).

Social experiences increase TL in females via OT enhancement

Telomeres act to protect the ends of chromosomes from degradation, and consequently, control chromosomal stability and cellular senescence (Mitchell et al., 2017; Drury et al., 2012). Here, socially reared female rats, in addition to increased plasma OT concentration, had significantly longer TL than social males (~638.01 bp). By contrast, interruption of endogenous OT release induced TL attrition (~442.66 bp) in social females only. Despite females appearing to have longer TL than males (Merrill et al., 2017), TL elongation in social females offers a potential predictor for extended lifespan in humans (Holt-Lunstad et al., 2010) and rodents such as rats (Yee et al., 2008), who typically engage in social relationships. Interestingly, TL was recently shown to positively correlate with the number of surviving children born to a woman (Fagan et al., 2017), suggesting a possible parallel link between reproductive success and longevity in females.

The pathways through which telomere elongation is influenced by peripheral OT enhancement or inhibition are not yet fully understood. For the present results, axo-vasal contacts for endocrine neurosecretion (Knobloch and Grinevich, 2014), a prominent central-peripheral axis which releases OT, vasopressin (VP), and their homologues into systemic blood circulation, may explain how alterations in peripheral OT can influence TL measured in the samples collected from ear. Axonal release of OT in the brain and in the circulatory system seem to be coordinated as central OT and peripheral OT in animals can be released simultaneously (Wotjak et al., 1998). In humans, a correlation was shown between plasma and brain OT indicating that peripheral OT represents a suitable proxy of central OT (Lefevre et al., 2017). It is also important to consider that, as they share a common ontogenetic origin from ectoderm, both brain and skin tissues contain similar collagenous proteins, reticulin, vasculature, and connective tissues such as fibroblasts and adipocytes. Therefore, changes in peripheral TL in general, and skin cells in particular, predict TL alterations found in brain tissue (Hehar and Mychasiuk, 2016).

The present results revealed that exposure to the OT antagonist across development and during social experiences accelerates TL attrition only in females. A causal explanation of how social and neurohormonal processes interact to induce a sex-dependent effect on TL was beyond the scope of the experiments. Three lines of evidence, however, support the observation of TL alteration in socially enriched females: (1) correlation between OT and social behaviours in females is stronger than males (Barraza and Zak, 2009); (2) both social experiences and OT hormone blunt HPA axis activity (Sippel et al., 2015; Ditzen et al., 2009); and more importantly, (3) TL shortening is associated with dysregulation of HPA axis activity and abnormal levels of cortisol (Schutte et al., 2016). Hence, the pathways involved in social experience-induced OT responses may contribute to sex differences in TL in close interaction with the HPA system. Notably, social experiences in our previous findings (Faraji et al., 2018) reduced HPA axis activity in F0 female rats and the F1 non-social housing offspring born to social mothers, which suggests an intergenerational impact of social rearing history through regulation of the HPA response in female rats. Various other genetic mechanisms not investigated here may also determine social behaviours. Immediate-early genes, such as zif268 mRNA expression in the dorsal and ventral medial mPFC in rats, appear to be sexually dimorphic (Stack et al., 2010). Because zif268 expression can activate specific signaling pathways, it may lead to long-lasting synaptic changes in the brain. For example, rats show sexual dimorphism in neuronal firing rates in the barrel cortex in response to social interaction components (Jurek and Neumann, 2018).

Taken together, the present results suggest that a socially stimulating environment is especially critical to female health trajectories. Even short periods of OT inhibition make females vulnerable to TL shortening via OT-mediated dysregulation of HPA system activity. Although influenced by sex and gender (Mather et al., 2011), TL measurement may be a useful tool in future studies to understand and predict health and longevity particularly in females.

Conclusion

Social experiences provide a source of psychoneurophysiological empowerment in both humans and animals and promote healthy development and successful aging (Yang et al., 2016). The present data causally associate social stimulation with genetic and behavioural improvements possibly as a function of endogenous OT. The findings indicate that early social experiences may last far into adulthood and manifest in TL elongation and novelty-seeking enhancement as a function of sex. The impact on cortical neuroplasticity, neurohormonal and behavioural outcome may be transmitted to the F1 non-social offspring in a sexually dimorphic manner. Thus, a socially stimulating environment may promote stress resiliency and mental health not only in exposed individuals, but also in their intergenerationally programmed descendants. Social enrichment may therefore provide a therapeutic avenue to promote stress resiliency and chances of healthy aging across generations.

Materials and methods

Animals

Male and female Wistar rats (222–430 g), bred and raised at the local vivarium were used in the present experiments. All animals were housed in a constant-temperature (21–24°C) room on a 12 h light/dark cycle (lights on at 7:30 am) with ad libitum access to food and water. Rats were handled for approximately 3 min daily for 5 consecutive d prior to any experimental manipulations. Also, body weight in all animals was recorded every 4 d. The behavioural testing was performed during the light phase of the cycle, at the same time of day by three experimenters blind to the experimental groups. All procedures in this study were carried out in accordance with the National Institutes of Health Guide to the Care and Use of Laboratory Animals, and were approved by the institutional animal care committee (Protocol No. 004674BGH; Avicenna Institute of Neuroscience-AINS).

Experimental design

Experiment 1: pups and their mothers were left undisturbed from postnatal day (PND) 1–21. After weaning at PND 21, 48 pups (five to six rats per litter) gathered from nine different litters were randomly assigned to four experimental groups in two housing conditions: (1) standard housing (males, n = 12), (2) standard housing (females, n = 12), (3) social housing (males, n = 12), and (4) social housing (females, n = 12). Novelty-seeking behaviour analysis included seven standard animals (four males, three females) and six social animals (three males, three females) from a previous experiment (Barrett and Richardson, 2011) to increase sample size. After living in either standard- or social-housing conditions for 86–89 d, all animals were subjected to blood sampling (days 83–86) and behavioural assessment (days 85–90). Animals were euthanized when behavioural assessments were completed. Experiment 2: at weaning, pups from 14 litters were randomly selected for Experiment 2. To accommodate OT antagonist (OT ANT) administration, animals were split into eight groups (n = 10–11/group): (1,2) standard housing: male and female (CONTROL), (3,4) standard housing: male and female (OT ANT), (5,6) social housing: male and female (CONTROL), and (7,8) social housing: male and female (OT ANT). All animals were housed in either standard- or social-housing units for 84–90 d and were subjected to blood sampling on days 80–83 and behavioural assessments on days 84–89. Rats were euthanized when behavioural testing was completed on days 88–91.

Housing condition

A complete description of the housing condition was previously reported by this team (Faraji et al., 2018). Briefly, animals assigned to the social housing condition were housed and raised in groups of 10–11 within two separate social housing units (86 cm × 86 cm × 41 cm) with no additional environmental enrichment provided. Animals were constantly living in the same units until completion of the experiment. In contrast, rats assigned to standard housing conditions were housed and raised in non-sibling groups of two or three within Makrolon shoebox cages (86 cm × 86 cm × 41 cm). Animals were briefly removed from their environments when bedding material was changed. Animals in social and standard housing conditions were raised separated by sex.

Blood sampling and oxytocin (OT) assay

Peripheral levels of OT (OTp) are assumed to reflect central releases of OT (OTc) in rats (Wotjak et al., 1998) and humans (Lefevre et al., 2017) with some procedural exceptions. Blood samples (0.5–0.7 mL) were taken 1–2 d prior to behavioural assessments (Faraji et al., 2014) to measure plasma concentration of OT using solid phase radioimmunoassay (RIA) (Alburges et al., 2000). Blood sample tubes containing aprotinin (500 kallikrein inactivation units/mL blood) were centrifuged at 3000 rpm (1700 g) for 15 min at 4°C. Plasma was stored at –70°C until analysis. Sample extraction and concentration were performed according to the manufacturer’s manual provided with the kits (Phoenix Pharmaceuticals, Burlingame, CA), and a method previously described by Kobayashi et al. (Kobayashi et al., 1999). Intra- and inter-assay variability was 7% and 15%, respectively, as reported by the manufacturer. No behavioural testing was performed on blood sampling days.

Oxytocin (OT) inhibition

The non-peptidyl OT antagonist (OT ANT) L-366,509 (MedKoo Biosciences, Inc., Morrisville, USA) (Evans et al., 1992; Pettibone et al., 1993) was administered (50 mg/kg) subcutaneously into the scruff of the neck in the OT ANT groups in Experiment 2. Administration occurred every other day (between 11:00 am and 12:00 noon) for 42–43 d (in total, 42–43 doses/rat) to intermittently inhibit or reduce OT secretion. Administration of L-366,509 started within the first week of the experiment (day 5) and ended 14 h before perfusion. The L-366,509 dosage used in Experiment 2 was chosen based on the results in two pilot experiments (each n = 3–5) that were found to significantly reduce plasma OT levels, and a previous report (Kobayashi et al., 1999). Physiologic saline solution was injected subcutaneously in control groups (42–43 doses/rat) to avoid the confounding differential effect of stress resulting from repeated OT ANT injections.

Although there is no information available on the penetration of L-366,509 into the blood-brain barrier (BBB) in the literature, a new generation of spiroindenylpiperidine camphor-sulfinamide OT antagonist (i.e. L-368,899, (Williams et al., 1994)) which readily crosses the BBB (Smith et al., 2010) was developed from L-366,509 with the same structural and lipophilic molecular characteristics. Thus, it is likely that antagonist L-366,509 crossed the BBB.

Corridor field task (CFT)

Novelty-seeking exploratory behaviour in the CFT was assessed according to the method of Faraji et al. (Faraji et al., 2018). Briefly, the floor of the CFT was divided into three zones: (1) corridor zone comprising the zone between external and internal walls; (2) open zone comprising the area within the task excluding corridor and the central zones; and (3) central zone comprising the middle area of the arena. Both variations of the CFT (plain CFT without a central object, and CFT with a central object) were used in the present study to assess locomotor aspects of free exploration. All rats were individually allowed to freely explore the environment for 8 min. Animals’ performance was recorded under dim illumination by a ceiling-mounted camera (CCTV Auto tracking PTZ; SONY, Tokyo, Japan) and analyzed by a computer tracking system (SINA motiongraph, Ayers et al., 2011), Tabriz, Iran) through analysis of the time spent in each zone. The apparatus was cleaned with 70% alcohol between test sessions.

Telomere measurement

Assuming that OT receptors are expressed in rat ear (Kitano et al., 1997) and human skin (Deing et al., 2013), genomic DNA from ear notch samples was extracted by a commercial kit (Sigma-Aldrich, Tokyo, Japan) based on the manufacturer's protocol (Hehar and Mychasiuk, 2016). Telomere length (TL) was measured using a modified protocol for a quantitative Real-Time Polymerase Chain Reaction (qRT-PCR) assay based on previous reports (Hehar and Mychasiuk, 2016; Cawthon, 2002). The protocol enabled calculation of the ratio of telomere copy repeats to a single-copy reference gene (36B4) to determine relative TL. Accordingly, when the ratio was found equal to 1.00, the unknown DNA was assumed to be identical to the reference DNA, whereas the ratios greater to, or less than one, were - respectively - considered increased or decreased telomere repeat numbers. A pipetting robot was used for all assay runs to avoid pipetting errors and inconsistency in reactions, and PCR assays were performed by two independent technicians. The relative TL was determined based on the linear regression equation reported by Cawthon (Cawthon, 2002).

Statistical analysis

Effects of main factors (housing condition; two levels, group; four and eight levels, sex; two levels, Zone; three levels, litter; nine and 11 levels) were analyzed separately for the time spent in CFT, with and without central object by repeated-measure and one-way ANOVA, and multivariate analysis of variance (MANOVA). Also, post-hoc test (Tukey) was used to adjust for multiple comparisons. Familywise error was considered prior to the multiple post-hoc analyses if necessary. Data for OT levels and TL were also analyzed by separate ANOVA analyses with the main between-subject factors of group and sex. Linear regression analysis was performed in Experiment 1 to predict the outcome (dependent) variables (the novelty-seeking behaviour in the CFT and telomere length) through the predictor (independent) variable (OT levels). Dependent and independent sample t-tests were conducted when necessary. In all statistical analyses (SPSS 16.0, SPSS Inc., USA), a p-value of less than 0.05 was considered statistically significant. Values represent mean ±SEM.

References

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
  11. 11
  12. 12
  13. 13
  14. 14
  15. 15
  16. 16
  17. 17
  18. 18
  19. 19
  20. 20
  21. 21
  22. 22
  23. 23
  24. 24
  25. 25
  26. 26
  27. 27
  28. 28
  29. 29
  30. 30
  31. 31
  32. 32
  33. 33
  34. 34
  35. 35
  36. 36
  37. 37
  38. 38
  39. 39
  40. 40
  41. 41
  42. 42
  43. 43
  44. 44
  45. 45
  46. 46
  47. 47
  48. 48
    Using sociometers to quantify social interaction patterns
    1. BNW Jukka-Pekka Onnela
    2. A Pentland
    3. S Schnorf
    4. D Lazer
    (2014)
    Scientific Reports, 4, 10.1038/srep05604.
  49. 49
  50. 50
  51. 51
  52. 52
  53. 53
  54. 54
  55. 55
  56. 56
  57. 57
  58. 58
  59. 59
  60. 60
    Is telomere length a biomarker of aging? A review
    1. KA Mather
    2. AF Jorm
    3. RA Parslow
    4. H Christensen
    (2011)
    The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 66:202–213.
    https://doi.org/10.1093/gerona/glq180
  61. 61
  62. 62
  63. 63
  64. 64
  65. 65
  66. 66
  67. 67
  68. 68
    Social support and resilience to stress: from neurobiology to clinical practice
    1. F Ozbay
    2. DC Johnson
    3. E Dimoulas
    4. CA Morgan
    5. D Charney
    6. S Southwick
    (2007)
    Psychiatry (Edgmont 4:35–40.
  69. 69
  70. 70
    Identification of an orally active, nonpeptidyl oxytocin antagonist
    1. DJ Pettibone
    2. BV Clineschmidt
    3. MT Kishel
    4. EV Lis
    5. DR Reiss
    6. CJ Woyden
    7. BE Evans
    8. RM Freidinger
    9. DF Veber
    10. MJ Cook
    (1993)
    The Journal of Pharmacology and Experimental Therapeutics 264:308–314.
  71. 71
  72. 72
  73. 73
  74. 74
  75. 75
  76. 76
    How does social support enhance resilience in the trauma-exposed individual?
    1. LM Sippel
    2. RH Pietrzak
    3. Dennis S, Charney
    4. LC Mayes
    5. SM Southwick
    (2015)
    Ecology and Society, 20, 10.5751/ES-07832-200410.
  77. 77
  78. 78
  79. 79
  80. 80
  81. 81
  82. 82
  83. 83
  84. 84
  85. 85
  86. 86
  87. 87
  88. 88
  89. 89
  90. 90
  91. 91
  92. 92
  93. 93

Decision letter

  1. Peggy Mason
    Reviewing Editor; University of Chicago, United States
  2. Catherine Dulac
    Senior Editor; Harvard University, United States

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

[Editors’ note: this article was originally rejected after discussions between the reviewers, but the authors were invited to resubmit after an appeal against the decision.]

Thank you for submitting your work entitled "Oxytocin-mediated Social Enrichment Promotes Longevity and Novelty Seeking" for consideration by eLife. Your article has been reviewed by four peer reviewers, including Peggy Mason as the Reviewing Editor and Reviewer #1, and the evaluation has been overseen by a Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Takefumi Kikusui (Reviewer #3).

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we regret to inform you that your work will not be considered further for publication in eLife.

The reviewers' chief concern is the lack of an OXT antag control. It is plausible that the stress of the non-trivial number of injections was the active agent here rather than the pharmaceutical compound. If the authors came back with such a control, we would be interested in such a (new) submission. The authors may also benefit from a number of other comments from the reviewers including walking back some of the conclusions to acknowledge the correlational nature of the data as well as the lack of a compelling direct path from housing to telomere length regulation.

Reviewer #1:

There is a lot to like and a fair amount to easily worry about in this study. The big positives are the large groups compared to normal social housing. Comparing normal (yet deprived in the natural scheme of things) to enriched (probably close to true normal for rats) is a strong design, far better than comparing 2-4 rodent groups (deprived) to social isolation (a ridiculous condition with no ecological correlate). The investigation of brave behavior (going in to the OF center with or without an object there) across the sexes and housing conditions is an interesting design aimed at important questions regarding early social development and sexual dimorphisms. The biggest issue with this paper is the lack of an injection control for the OXT antag injections. The results obtained – less exploration of the center after the antag – could easily result from a differential effect of stress (the stress of 28 injections!!) on females compared to males, Is this issue surmountable with the data in hand? I fear no but look forward to seeing what the other reviewers think.

At the outset of the results it is necessary to state how many rats are in social (n=10-11) and standard (n=2-3) groups. It would also be useful to state how the rats from different litters were assigned to the groups. The no difference reported across litter is not interpretable without this information.

Then the authors need to be far more careful about the use of the word "group." They state "…across different groups (n=21-23/group)" which appears to say that there are 21-23 in the group housed cages but in point of fact this appears to be the number of animals in the social or standard conditions. Which would imply that two Social groups (n=20-22 rats), one for each sex, were compared to 8-10 Standard groups, half for each sex? This should be explicitly stated. Then the number of rats in Figure 1A Social as 44 and in 1B as 54. Please make sense of these various numbers.

A significant effect of Housing Condition (F1,86=24.85, p{less than or equal to}0.001) and Group (F3,84=13.69, p{less than or equal to}0.001). By Group do the authors mean sex? I thought the conditions were the groups. This is very confusingly written. I think the authors are saying there are 4 groups whereas it should be a 2x2 factor design with housing – 2 conditions – and sex – 2 groups.

It appears that the groups are mixed males and females. Are some of the females then pregnant? If this is the case, then some of the females must be pregnant which would obviously change their behavior. Please explain.

There is no saline control for the OT antag injections. This is an experimental choice that must be justified and is possibly a deal-killer. The injections were "every other day (between 11:00-12:00 am) for 13 days (in total, 28 doses/rat)…" Clearly every other day for 13 days does not yield 28 doses per rat. So, this needs to be rewritten into careful English so that the reader is told what was done.

Reviewer #2:

This is a terrific paper that should be published as soon as possible. I believe it stands to make a substantive contribution to the field and will be heavily cited. It explains the interactions among social relationships, Oxytocin, telomeres and longevity convincingly. Links to HPA axis function are also very interesting. The implications for human relationships are very intriguing. The background and discussion are soundly presented. The analytic methods are sound, and the results are compelling.

I am less able to review the data collection methods, as I am not an animal scientist, so I trust other reviewers to do so.

There were no grammatical or typos to note.

Again, I recommend publication as soon as possible.

Reviewer #3:

This paper demonstrated that long-term social housing in the juvenile period enhanced peripheral OT secretion, together with longer TL. Novelty-seeking behavior was also increased by social housing. These housing effects were more robust in females. The data are interesting because the authors examined causality between OT secretion, TL and behavior. Even though there is no neuron/ brain region analysis conducted, the findings would reveal a new function of OT.

Here are my concerns,

1) Ear notch samples were used to assess TL, but is there any evidence showing that these cells express oxytocin receptor? If so, please add the reference in the Materials and methods section. If not, the authors need to include the discussion how blood OT can affect TL in these OTR-lacking cells.

2) OT antagonist L-366,509 was used to inhibit OT system in the CNS. Please add the information about penetration of this chemical through blood brain barrier in the Materials and methods section.

3) The authors mentioned the function of OT on the HPA axis, but this paper did not include the data of corticosterone. If the blood samples remain, please measure corticosterone, to make the conclusion more convincing.

4) Related to #2 and #3, higher corticosterone exposure will shorten the TL in peripheral cells. Please add the corticosterone data if possible.

5) Were the changes observed in social-housing rats caused by social interactions? If the rats were housed in a large area, the locomotor activity can increase. In this experimental setting, the authors cannot conclude that the changes were caused by social interactions. There might be an appropriate control group needed, such as similar locomotor activity without social interaction. If the authors cannot exclude the effect of activity (physical exercise can facilitate OT secretion, TL etc.), the discussion should be modified.

6) The social housing was conducted just after the weaning, so the effects of social housing would be robust (still in the developmental period). If the social housing were to be conducted in the adulthood or aged rats, could the same positive affects be observed? In humans, social loneliness in elderly is one of the important social issues related to the health problems.

Reviewer #4:

In manuscript by Faraji et al., the authors reported correlations between circulating oxytocin concentration and prolonged social housing, in both male and female rats. They further showed that prolonged social housing correlated with elongated telomere length in only in female rats, an effect blocked by the oxytocin antagonist L-366,509. While the correlation between oxytocin and telomere length is potentially interesting, I believe that there are some gaps of logic in the author's interpretation of the results.

1) The title states that "oxytocin-mediated social enrichment promotes longevity and novelty seeking". The authors did not measure life span of their experimental animals, only telomere length. To my understanding, telomere length is not equivalent to longevity and thus I find the title misleading.

2) This is my main concern: the authors showed that social housing increased oxytocin (OT) level in both male and female rats, but only increased telomere length in female rats. Administration of the OT antagonist blocked the effect of social housing in both males and females, but only blocked TL increase in social females. Based on these results, the authors concluded that "…extended social experiences modulate TL and novelty seeking through the OT system in a sex-dependent manner…" (Abstract). These is extensive literature reporting oxytocin as a "social hormone", thus the correlation between oxytocin level and social housing is consistent with other published results. To my knowledge, the link between oxytocin level and telomere length has not been previously reported and certainly there are no proposed mechanisms. Under these circumstances, I would be more willing to accept the authors' correlative results as potentially interesting and significant, if the correlations between oxytocin level and telomere length were similar to those of social housing, i.e. either both correlations held for males and females, or both held for only females.

3) I do not know the telomere field, but is telomere length in the ear representative of the entire body? Would telomere length in multiple cell types need to be measured to reach a conclusion?

4) I do not agree with the authors' interpretation of their results, in statements such as: "The present study demonstrates a key role of OT in socially mediated behavioral fitness and telomere length as an indicator of overall health and longevity" (Discussion section); "…Thus OT is causally involved in mediating beneficial effects of social experience on behavior and TL…" (Discussion section), and "The present data causally associates social stimulation with genetic and behavioral improvements as a function of endogenous OT" (subsection “Conclusion”). The authors did not measure behavior fitness, health or longevity. Based on arguments presented in point 2 above, I don't think the results provided sufficient evidence to demonstrate a causal link between OT and TL. In order to demonstrate a causal link, at a minimum, the authors would need to demonstrate that in vivo OT administration can promote telomere length, and that the effect of social housing on telomere length are abolished in oxytocin knockout mice. Telomere length needs to be measuring in multiple tissues and not just ear punches. Also, no statements can be made regarding behavior fitness, health or longevity unless data is provided.

5) Related to points 2 and 4 and given the discussion regarding how accurately oxytocin level in the plasma measured using RIA or ELISA reflects oxytocin level in the brain, the authors could measure oxytocin mRNA level or immunostain for oxytocin in brain sections, to further establish the correlation between oxytocin level and telomere length.

6) Results presented in Figure 2A-D are very similar to that presented in a previous publication from the same lab (Faraji et al., 2018). The authors should reference their previous results in the Results section, and not just the Materials and methods section.

7) Statements such as "…Thus, social females' exploration in the corridor zone decreased -1.84 for each pM of OT concentration" (subsection “Novelty-Seeking Behaviour in Females was Significantly Influenced by Social Experiences”), and "…TL, therefore, only in social females increased 8.53 bp for each pM of OT level" (subsection “Novelty-Seeking Behaviour in Females was Significantly Influenced by Social Experiences”), do not make sense to me, as the data points for TL range from 4000 to 5700 bp, and that for OT range from around 40 to 125 pM. In light of huge differences in the levels of OT and TL, these precise interpretations do not make much sense to me.

[Editors’ note: what now follows is the decision letter after the authors submitted for further consideration.]

It is great that you did saline controls. Of course, it is a problem that none of the reviewers realized that. Please address this by making it far more evident in the manuscript. Also make sure that the saline groups make their way into the figures. Finally, please respond to the other issues raised by the reviewers. We look forward to your resubmission.

https://doi.org/10.7554/eLife.40262.015

Author response

[Editors’ note: the author responses to the first round of peer review follow.]

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we regret to inform you that your work will not be considered further for publication in eLife.

The reviewers' chief concern is the lack of an OXT antag control. It is plausible that the stress of the non-trivial number of injections was the active agent here rather than the pharmaceutical compound. If the authors came back with such a control, we would be interested in such a (new) submission. The authors may also benefit from a number of other comments from the reviewers including walking back some of the conclusions to acknowledge the correlational nature of the data as well as the lack of a compelling direct path from housing to telomere length regulation.

Reviewer #1:

There is a lot to like and a fair amount to easily worry about in this study. The big positives are the large groups compared to normal social housing. Comparing normal (yet deprived in the natural scheme of things) to enriched (probably close to true normal for rats) is a strong design, far better than comparing 2-4 rodent groups (deprived) to social isolation (a ridiculous condition with no ecological correlate). The investigation of brave behavior (going in to the OF center with or without an object there) across the sexes and housing conditions is an interesting design aimed at important questions regarding early social development and sexual dimorphisms. The biggest issue with this paper is the lack of an injection control for the OT antag injections. The results obtained – less exploration of the center after the antag – could easily result from a differential effect of stress (the stress of 28 injections!!) on females compared to males, Is this issue surmountable with the data in hand? I fear no but look forward to seeing what the other reviewers think.

As stated in the manuscript (Materials and methods section), physiologic saline solution was injected in control groups. The new paragraph now reads:

“…Physiologic saline solution was injected subcutaneously in control groups (42-43 doses/rat) to avoid the confounding differential effect of stress resulted by repeated OT ANT injections.”

We also included the word “saline” in control group terminology in Figures 3 and 4 in order to highlight the saline injections in all control animals in Experiment 2 (please see Figure 3A-D, and Figure 4A-D, F and H).

At the outset of the results it is necessary to state how many rats are in social (n=10-11) and standard (n=2-3) groups. It would also be useful to state how the rats from different litters were assigned to the groups. The no difference reported across litter is not interpretable without this information.

We appreciate this comment and haveaddressed this concern by re-writing the first sentence of the Results section in the manuscript as the reviewer suggested. The new sentence now reads:

“…Figure 1A shows the changes in plasma OT concentrations across different groups as a function of housing condition (Standard, males: n=23, females: n=21; Social, males: n=22, females: n=22)…”.

Also:

“…Examination of TL in ear notch skin cells showed that socially-raised rats (males: n=31, females: n=23) had greater telomere length than standard animals (males: n=28, females: n=27;…” (Results section).

Moreover, to further address the reviewer’s concerns we have re-written the sampling method for different litters in the manuscript that now reads:

“…Pups and their mothers were left undisturbed from postnatal day (PND) 1–21. After weaning at PND 21, 48 pups (5-6 rats per litter) gathered from 9 different litters were randomly assigned to four experimental groups in two housing conditions:…”(Materials and methods section).

Then the authors need to be far more careful about the use of the word "group." They state "…across different groups (n=21-23/group)" which appears to say that there are 21-23 in the group housed cages but in point of fact this appears to be the number of animals in the social or standard conditions. Which would imply that two Social groups (n=20-22 rats), one for each sex, were compared to 8-10 Standard groups, half for each sex? This should be explicitly stated. Then the number of rats in Figure 1A Social as 44 and in 1B as 54. Please make sense of these various numbers.

We have re-written this part of the Results section to address the reviewer’s concerns that now reads:

“Figure 1A shows the changes in plasma OT concentrations across different groups as a function of housing condition (Standard, males: n=23, females: n=21; Social, males: n=22, females: n=22). A significant effect of Housing Condition was observed in terms of circulating OT concentration (Social vs. Standard; F1,86=24.85, p≤0.001) and Group (social males and females vs. standard males and females; F3,84=13.69, p≤0.001), suggesting…”..

A significant effect of Housing Condition (F1,86=24.85, p{less than or equal to}0.001) and Group (F3,84=13.69, p{less than or equal to}0.001). By Group do the authors mean sex? I thought the conditions were the groups. This is very confusingly written. I think the authors are saying there are 4 groups whereas it should be a 2x2 factor design with housing – 2 conditions – and sex – 2 groups.

Please, refer to the previous comments. Also, we have included further information into the Statistical Analysis about how we defined and analyzed variables in order to address this major concern. The new inclusion now reads:

“…Effects of main factors (Housing Condition; 2 levels, Group; 4 & 8 levels, Sex; 2 levels, Zone; 3 levels, Litter; 9 & 11 levels) were analyzed separately for the time spent in CFT, with and without …” (Materials and methods section).

It appears that the groups are mixed males and females. Are some of the females then pregnant? If this is the case, then some of the females must be pregnant which would obviously change their behavior. Please explain.

The authors apologize for not making this clear enough in the manuscript. Males and females were raised separately either in social or standard housing conditions. This is mentioned in the revised manuscript in the Materials and methods section for further explanation.

There is no saline control for the OT antag injections. This is an experimental choice that must be justified and is possibly a deal-killer.

We have addressed these concerns by including a new paragraph. The new paragraph now reads:

“…Physiologic saline solution was injected subcutaneously in control groups (42-43 doses/rat) to avoid the confounding differential effect of stress resulted by repeated OT ANT injections.” (Materials and methods section).

We also included the word “saline” in control group terminology in Figures 3 and 4 in order to highlight the saline injections in all control animals in Experiment 2 (please see Figure 3A-D, and Figure 4A-D, F and H).

The injections were "every other day (between 11:00-12:00 am) for 13 days (in total, 28 doses/rat)…" Clearly every other day for 13 days does not yield 28 doses per rat. So, this needs to be rewritten into careful English so that the reader is told what was done.

Many thanks for the comment, and we apologize for this mistake. We rephrased the sentence that now reads:

“…was administered (50 mg/kg) subcutaneously into the scruff of the neck in the OT ANT groups in Experiment 2. Administration occurred every other day (between 11:00-12:00 am) for 42-43 days (in total, 42-43 doses/rat) to intermittently inhibit or reduce OT secretion.…” (Materials and methods section).

Reviewer #3:

This paper demonstrated that long-term social housing in the juvenile period enhanced peripheral OT secretion, together with longer TL. Novelty-seeking behavior was also increased by social housing. These housing effects were more robust in females. The data are interesting because the authors examined causality between OT secretion, TL and behavior. Even though there is no neuron/ brain region analysis conducted, the findings would reveal a new function of OT.

Here are my concerns,

1) Ear notch samples were used to assess TL, but is there any evidence showing that these cells express oxytocin receptor? If so, please add the reference in the Materials and methods section. If not, the authors need to include the discussion how blood OT can affect TL in these OTR-lacking cells.

This is a wonderful suggestion which we take very seriously. Indeed, we found out that at least in three previously published papers by Kitano et al., (1997), Bolan and Goren, (1987) and Deing et al., (2013), OT receptors were found to be expressed in the rat inner ear and epididymal adipocytes, and human skin. Two of them are cited now in the manuscript (please see Materials and methods section) to address the reviewer’s concerns. Moreover, the authors hypothesize that one possible mechanism in bridging increased plasma OT concentration to telomere elongation in the present experiments likely relies upon axo-vasal contacts for endocrine neurosecretion (please see Knobloch and Grinevich, 2014). By its traditional definition, endocrine neurosecretion refers to release and delivery of OT, VP and their homologues to systemic blood circulation. It is traditionally assumed that blood stream carries these molecules (e.g. OT) to their receptors which are abundantly expressed in peripheral organs such as uterus, mammary glands, intestinal tracts, heart, and more specifically in this case, even in the skin. Thus, we anticipate that a major part of the observed changes in TL induced by prolonged social experiences and the following OT enhancement is influenced by the axo-vasal contacts, which reveals an alternative explanation of the dynamic dialogue between the central capacities and peripheral repertoires in response to social experiences.

Also, to further address the reviewer’s concerns, we included a new paragraph in the discussion that now reads:

“…The pathways through which telomere elongation is influenced by peripheral OT enhancement or inhibition are not yet fully understood. For the present results, axo-vasal contacts for endocrine neurosecretion14, a prominent central-peripheral axis which releases OT, vasopressin (VP) and their homologues into systemic blood circulation, may explain how alterations in peripheral OT can influence TL measured in the samples collected from ear…” (please see subsection “Social Experiences Increase TL in Females via OT Enhancement).

2) OT antagonist L-366,509 was used to inhibit OT system in the CNS. Please add the information about penetration of this chemical through blood brain barrier in the Materials and methods section.

We appreciate this thoughtful comment. To our best knowledge, there is no detailed proof available in the literature for the penetration of antagonist L-366,509 through the blood-brain barrier. However, based on experimental evidence, we assumed that L-366,509 crosses the blood-brain barrier. First, all lipophilic molecules readily cross the blood-brain barrier, even though hydrophilic molecules including OT itself do not (please see McEwen, 2004; Churchland and Winkielman, 2012). Because camphor-based non-peptide OT antagonists such as L-366,509 belong to bulky, lipophilic groups (see Evans et al., 1992) that have a high tolerance and considerably improved solubility in aqueous media, we anticipate that antagonist L-366,509 crosses the blood-brain barrier. Second and more importantly, the non-peptide OT antagonist L-368,899 is a newer member of structural classes of spiroindenes that is structurally derived from L-366,509 (see Williams et al., 1994). The most prominent difference between these two antagonists, however, refers to the higher affinity of L-368,899 rather than L-366,509 for OT receptors. Hence, as L-368,899 is reported to cross the blood-brain barrier with selective accumulation in specific areas of the limbic system where it might inhibit OT-induced social behaviours (see Smith et al., 2010), it allowed the authors to predict that antagonist L-366,509 can also cross the blood-brain barrier.

We fully agree that this issue has to be reported in the manuscript, and we apologize for this shortcoming. Therefore, we addressed the reviewer’s concern in the manuscript in a new paragraph that reads:

“…Although there is no information available on the penetration of L-366,509 into the blood-brain barrier (BBB) in the literature, a new generation of spiroindenylpiperidine camphor-sulfinamide OT antagonist (Williams et al., 1994) which readily crosses the BBB (Smith et al., 2010) was developed from L-366,509 with the same structural and lipophilic molecular characteristics. Thus, it is likely that antagonist L-366,509 crossed the BBB.” (please see Materials and methods section).

3) The authors mentioned the function of OT on the HPA axis, but this paper did not include the data of corticosterone. If the blood samples remain, please measure corticosterone, to make the conclusion more convincing.

The reviewer is absolutely right that CORT data could potentially provide support for the current results. Unfortunately, due to economic instabilities in Iran and other practical constraints we were unable to do these analyses. We deeply regret this issue as we indeed had initially planned on these analyses. Instead, to compensate for the lack of information of CORT changes, we now discuss our results in the light of previous relevant work by others and by our team in the Discussion section. The considerable amount of existing evidence provides solid support for our current findings. Again, we very much apologize for this shortcoming.

4) Related to #2 and #3, higher corticosterone exposure will shorten the TL in peripheral cells. Please add the corticosterone data if possible.

Please, see comment #3.

5) Were the changes observed in social-housing rats caused by social interactions?

This is an excellent point for consideration, thank you. We agree with the reviewer that both experiments in the present study could be conducted with more controls to better control for potentially confounding variables, even though we have done our best to have most of the confounding variables controlled. Also, it is very difficult to mention that changes observed in the present experiments were only the results of social interactions, because other uncontrolled external factors simultaneously may confound and/or overshadow our results. However, we strongly believe that social interactions as the most prominent component of social-housing condition designed here played a key role in the changes reported in the present study. Please refer to our logics in the next part of our answer to this comment.

If the rats were housed in a large area, the locomotor activity can increase. In this experimental setting, the authors cannot conclude that the changes were caused by social interactions. There might be an appropriate control group needed, such as similar locomotor activity without social interaction. If the authors cannot exclude the effect of activity (physical exercise can facilitate OT secretion, TL etc.), the discussion should be modified.

As stated in the manuscript (please see Materials and methods section), the same size of cage (86 cm × 86 cm × 41 cm) was used for the social and standard housing conditions, thus varying only the number of cage mates (standard, n = 2–3 vs. social, n =10- 11) while keeping the size of the environments constant (please see Faraji et al., 2018 for further details on the housing conditions). For two reasons, therefore, we do not agree with the reviewer that the observed changes are influenced by the enhanced level of physical activities in social-housing condition. First, if physical activity could determine the incidence or the extent of changes, all of these changes should be normally observed in the standard animals (here, controls) who were raised in the large but less crowded units. Second, changes in OT levels in response to physical activity induced by voluntary wheel running in rats are not entirely evident, yet (for instance, please see Broderick et al., 2014 and Bakos et al., 2007). Moreover, physical activity, to our knowledge, has been shown to impact OT levels in humans only if associated with a more strenuous type of physical exercise (see also Frisen et al., 2017 and Landgraf et al., 1982, for example). No additional enrichment stimulus such as a running wheel was provided for rats in neither standard nor social housing conditions. Further follow-up studies that address these issues would make a timely contribution to the literature.

6) The social housing was conducted just after the weaning, so the effects of social housing would be robust (still in the developmental period). If the social housing were to be conducted in the adulthood or aged rats, could the same positive affects be observed? In humans, social loneliness in elderly is one of the important social issues related to the health problems.

This is a very interesting question and worthy a follow-up study. Unfortunately, the present data do not allow a conclusion on social enrichment effects in older age, although previous data suggest that this assumption is very reasonable and worthy of further consideration (For example, please see Okabayashi’s report about social interactions and their effects on subjective well-being among Japanese elders; Okabayashi and Houghman, 2014).

Reviewer #4:

In manuscript by Faraji et al., the authors reported correlations between circulating oxytocin concentration and prolonged social housing, in both male and female rats. They further showed that prolonged social housing correlated with elongated telomere length in only in female rats, an effect blocked by the oxytocin antagonist L-366,509. While the correlation between oxytocin and telomere length is potentially interesting, I believe that there are some gaps of logic in the author's interpretation of the results.

1) The title states that "oxytocin-mediated social enrichment promotes longevity and novelty seeking". The authors did not measure life span of their experimental animals, only telomere length. To my understanding, telomere length is not equivalent to longevity and thus I find the title misleading.

We appreciate this comment and changed the title. The revised title now reads:

Oxytocin-mediated social enrichment promotes longer telomeres and novelty seeking

2) This is my main concern: the authors showed that social housing increased oxytocin (OT) level in both male and female rats, but only increased telomere length in female rats. Administration of the OT antagonist blocked the effect of social housing in both males and females, but only blocked TL increase in social females. Based on these results, the authors concluded that "…extended social experiences modulate TL and novelty seeking through the OT system in a sex-dependent manner…" (Abstract). These is extensive literature reporting oxytocin as a "social hormone", thus the correlation between oxytocin level and social housing is consistent with other published results. To my knowledge, the link between oxytocin level and telomere length has not been previously reported and certainly there are no proposed mechanisms. Under these circumstances, I would be more willing to accept the authors' correlative results as potentially interesting and significant, if the correlations between oxytocin level and telomere length were similar to those of social housing, i.e. either both correlations held for males and females, or both held for only females.

We would like to address this thoughtful and challenging comment by discussing several main arguments. First and most importantly, we believe that the experimental design of the present study enables us to make causal conclusions for the most parts of the results. Therefore, although for some experimental constraints we did not identify and/or offer a causal mechanism to connect variations in dependent variables such as peripheral OT levels to the TL, our data are still able to profile a clear “causal effect” for the independent and dependent variables, beyond the mere correlational conclusion. Indeed, the presence of control group(s) in both experiments assures empirical correlation along with temporal priority of the independent variable, thus providing a solid causal effect. Like the reviewer, the authors are also concerned about spurious data interpretation. Moreover, heavily standardized conditions by which we raised the groups of animals in standard units and aged-matched groups in a social environment helped us to interpret most variations in dependent variables (behavioural, hormonal and genetic) in the light of a causal conclusion. We believe that if we could (1) provide evidence of empirical association between at least one independent variable and one dependent variable, (2) confirm temporal priority of the independent variable (here, housing condition) using control groups, and also (3) tried to have nonspurious changes which were not influenced by irrelevant variable(s), the minimum criteria for making an experimental casual effect are met. Therefore, we believe that as long as we are able to manipulate the independent variable in a systematic manner and address internal and external control variables using careful implementation and analysis, the causal conclusion (if A [where A is social experience], then B [where B is OT enhancement and/or telomere elongation) can be possible and suggested.

Second, we agree with the reviewer that “…these is extensive literature reporting oxytocin as a "social hormone", thus the correlation between oxytocin level and social housing is consistent with other published results…”. However, to our knowledge, most of these correlational reports are depicting a hormone-to-behaviour approach (as stated in the manuscript) through which subjects are exposed to OT first, and then they are monitored for possible behavioural impacts of OT. In the present experiment, by contrast, we have chosen an alternative, behaviour-to-hormone approach by which we examined the impact of prolonged social experience on OT levels. This approach provides a new argument of causality to our results and conclusions as opposed to most literature about the experimental administration of OT and its consequences.

Third, we regard our data as considerably valid to causally link social lifestyle in early development (the major independent variable) among females to behavioural, endocrine and genetic outcomes (dependent variables). This seems more important when such changes are questioned in response to interruptions in social relationships among females and their health consequences, or when factors that in general determine telomere length in early life are investigated. Also, simultaneous responses (enhanced novelty seeking, increased levels of OT and telomere elongation) of females to persistent social experiences have implications in future investigations to causally hypothesize these changes. Examples include sex hormones that were previously shown to play a key role in telomere length (see Calado et al., 2006, for example) or for neurotrophins (e.g. BDNF) that are confirmed to influence telomere function (see Niu and Yip, 2011) and OT receptors in women (see Chagnon et al., 2015).

3) I do not know the telomere field, but is telomere length in the ear representative of the entire body? Would telomere length in multiple cell types need to be measured to reach a conclusion?

These are very exciting questions. The answer to both questions is “not necessarily”, even though the TL in lifespan is gender-related and follows a tissue-specific regulation during development. Indeed, the TL within individual tissues is regulated independently. In humans, TL seems highly variable between tissues and individuals. However, samples taken from peripheral organs such as kidney, liver, lung, etc. for inter-tissue comparison in intact rats are showing the same profile of changes in the TL (please, see Cherif et al.,’s work, such as Cherif et al., 2003). Also, studies with various strains of mice did not detect any differences in TL between tissues such as liver, testes or kidney (for instance, please see Coviello-McLaughlin and Prowse, 1997), although the TL generally differs between species. Moreover, TL in hearts and lungs in rats are equally affected by physical stress (please, see Wang et al., 2018) meaning that peripheral organs are possibly showing similar telomeric responses to experimental manipulations. Therefore, because different tissue types may represent different deterioration processes along with variable mortality, inter-tissue analysis in humans can potentially create more valid assessment of TL alterations in response to developmental influences or environmental manipulations. However, this does not seem to be necessarily required in rats as mentioned above. The only exception is for the TL in the brain tissues in rats and mice that were shown to depict different rates of TL alterations when compared with, for instance, spleen or other peripheral organs. Importantly, the reasons by which we have chosen the ear notch samples in the present study, in addition to the ease of collection, was that samples from ear are emerged from ectoderm, the same lineage as brain tissue. Like brain tissue, skin samples also contain a mixture of collagenous proteins, reticulin, vasculature, and connective tissues such as fibroblasts and adipocytes. Therefore, changes in rat peripheral TL in general, and skin cells in particular not only may represent TL changes in other organs influenced by experimental interventions, but also arguably predict TL alterations in the brain tissue (please see Hehar and Mychasiuk, 2016). To address this concern, we have included a new statement in the Discussion section that now reads:

“…It is also important to consider that, as they share a common ontogenetic origin from ectoderm, both brain and skin tissues contain similar collagenous proteins, reticulin, vasculature, and connective tissues such as fibroblasts and adipocytes. Therefore, changes in peripheral TL in general, and skin cells in particular, predict TL alterations found in brain tissue (Hehar and Mychasiuk, 2016).

4) I do not agree with the authors' interpretation of their results, in statements such as: "The present study demonstrates a key role of OT in socially mediated behavioral fitness and telomere length as an indicator of overall health and longevity" (Discussion section); "…Thus OT is causally involved in mediating beneficial effects of social experience on behavior and TL…" (Discussion section), and "The present data causally associates social stimulation with genetic and behavioral improvements as a function of endogenous OT" (subsection “Conclusion”). The authors did not measure behavior fitness, health or longevity.

Thanks to the reviewer for these comments. We have made some changes in the sentences in the manuscript as the reviewer requested. We hope that these changes address the reviewer’s concerns:

- Old sentence: “…The present rodent study demonstrates a key role of OT in socially mediated behavioral fitness and telomere length as an indicator of overall health and longevity…”.

New sentence: “…The present rodent study indicates that persistent social experience is causally associated with increased OT level, elevated exploratory behaviour, and telomere elongation….” (please see Discussion section).

- Old sentence: “…Thus OT is causally involved in mediating beneficial effects of social experience on behaviour and TL…".

New sentence: “…Thus OT is arguably involved in mediating beneficial effects of social experience on behaviour and TL…” (please see Discussion section).

- Old sentence: “…The present data causally associates social stimulation with genetic and behavioural improvements as a function of endogenous OT…".

New sentence: “…The present data causally associates social stimulation with genetic and behavioural improvements possibly as a function of endogenous OT…” (please see Discussion section).

Based on arguments presented in point 2 above, I don't think the results provided sufficient evidence to demonstrate a causal link between OT and TL. In order to demonstrate a causal link, at a minimum, the authors would need to demonstrate that in vivo OT administration can promote telomere length, and that the effect of social housing on telomere length are abolished in oxytocin knockout mice. Telomere length needs to be measuring in multiple tissues and not just ear punches. Also, no statements can be made regarding behavior fitness, health or longevity unless data is provided.

We agree with the reviewer that data in the present experiments cannot draw a direct line to connect changes in OT levels to TL, as the study was not originally designed for such questions. However, we still believe that the present data can make causal conclusion(s) for bridging the independent variable (i.e. social life) and dependent variables (i.e. enhanced novelty-seeking, increased levels of OT and/or telomere elongation). Please see our rationales in response to comment 2. Moreover, even though longevity was not measured in the present study, TL itself has been proposed as a potential biomarker for longevity in humans, mice, and other organisms in many studies (please, see Calado and Dumetrio, 2013, for example). More importantly, as shown by Monaghan’s team, TL in early life predicts lifespan (see Heidinger et al., 2012). From our viewpoint, results in our experiments significantly advance the understanding factors that may causally determine early life influences on TL.

5) Related to points 2 and 4 and given the discussion regarding how accurately oxytocin level in the plasma measured using RIA or ELISA reflects oxytocin level in the brain, the authors could measure oxytocin mRNA level or immunostain for oxytocin in brain sections, to further establish the correlation between oxytocin level and telomere length.

We appreciate this comment and apologize for our mistake. The authors confirm that 7 standard animals (4 males and 3 females) and 6 social animals (3 males and 3 females) from our previous experiment were included into the analysis of novelty-seeking behaviour in Experiment 1 in the present study. The inclusion of these animals has been conducted to increase sample size and to reduce within-group variation, thus achieving more precise estimates of average. To address the reviewer’s concern, we included a new sentence into the Material and Methods section which seemed more relevant that now reads:

“…Novelty-seeking behaviour analysis included 7 standard animals (4 males, 3 females) and 6 social animals (3 males, 3 females) from a previous experiment (Faraji et al., 2018) to increase sample size…”.

6) Results presented in Figure 2A-D are very similar to that presented in a previous publication from the same lab (Faraji et al., 2018). The authors should reference their previous results in the Results section, and not just the Materials and methods section.

Thank you for the great suggestion. Some of us (SR and RM) are analysing brain tissues for OT responses in the PVN and SON in the hypothalamus in response to social life, but unfortunately the results might not be reported in the present manuscript due to the extended amount of time required for such analyses. For the present study, based on previous findings in the literature we assumed that peripheral levels of OT can properly reflect central releases of OT in rats (for example, please see Wotjak et al., 1998). Furthermore, it seems that there is a coordinated axonal release of OT in the brain and in the circulatory system, and central OT and peripheral OT in animals can be released simultaneously. In humans, a correlation was shown between plasma and brain OT indicating that (at least under specific testing conditions) peripheral OT represents a suitable proxy of central OT (please see Lefvre and others 2017). Therefore, the authors believe that plasma OT concentration in the present study can be considered as a marker of its action in the brain. To address this concern, we have included a new statement in the Discussion section that now reads:

“…Axonal release of OT in the brain and in the circulatory system seem to be coordinated as central OT and peripheral OT in rats can be released simultaneously (Wotjak et al., 1998). In humans, a correlation was shown between plasma and brain OT indicating that peripheral OT represents a suitable proxy of central OT (Lefvre et al., 2017).

7) Statements such as "…Thus, social females' exploration in the corridor zone decreased -1.84 for each pM of OT concentration" (subsection “Novelty-Seeking Behaviour in Females was Significantly Influenced by Social Experiences”), and "…TL, therefore, only in social females increased 8.53 bp for each pM of OT level" (subsection “Novelty-Seeking Behaviour in Females was Significantly Influenced by Social Experiences”), do not make sense to me, as the data points for TL range from 4000 to 5700 bp, and that for OT range from around 40 to 125 pM. In light of huge differences in the levels of OT and TL, these precise interpretations do not make much sense to me.

We agree and are now convinced that the reviewer is absolutely right. Therefore, we omitted both sentences from the manuscript to address the reviewer’s concerns.

[Editors’ note: the author responses to the re-review follow.]

It is great that you did saline controls. Of course, it is a problem that none of the reviewers realized that. Please address this by making it far more evident in the manuscript. Also make sure that the saline groups make their way into the figures. Finally, please respond to the other issues raised by the reviewers. We look forward to your resubmission.

We appreciate this comment and have addressed these concerns by re-writing the corresponding sentence in the manuscript as the reviewer suggested. The new paragraph now reads:

“…Physiologic saline solution was injected subcutaneously in control groups (42-43 doses/rat) to avoid the confounding differential effect of stress resulted by repeated OT ANT injections.” (please, see Materials and methods section).

We also included the word “saline” in control group terminology in Figures 3 and 4 in order to highlight the saline injections in all control animals in Experiment 2 (please, see Figure 3A-D, and Figure 4A-D, F and H).

https://doi.org/10.7554/eLife.40262.016

Article and author information

Author details

  1. Jamshid Faraji

    1. Canadian Centre for Behavioural Neuroscience, University of Lethbridge, Lethbridge, Canada
    2. Faculty of Nursing & Midwifery, Golestan University of Medical Sciences, Gorgan, Iran
    Contribution
    Conceptualization, Methodology
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-1726-5836
  2. Mitra Karimi

    Inclusive-Integrated Education Program for Children with Special Needs, Exceptional Education Organization, Tehran, Iran
    Contribution
    Supervision, Investigation, Visualization, Project administration
    Competing interests
    No competing interests declared
  3. Nabiollah Soltanpour

    Department of Anatomical Sciences, Babol University of Medical Sciences, Babol, Iran
    Contribution
    Conceptualization, Data curation, Supervision, Investigation, Methodology, Project administration
    Competing interests
    No competing interests declared
  4. Alireza Moharrerie

    Department of Anatomy, Golestan University of Medical Sciences, Gorgan, Iran
    Contribution
    Formal analysis, Supervision, Investigation, Project administration
    Competing interests
    No competing interests declared
  5. Zahra Rouhzadeh

    Department of Psychology, Islamic Azad University, Sari, Iran
    Contribution
    Investigation, Project administration
    Competing interests
    No competing interests declared
  6. Hamid lotfi

    Department of Psychology, Islamic Azad University, Tonekabon, Iran
    Contribution
    Supervision, Investigation, Project administration
    Competing interests
    No competing interests declared
  7. S Abedin Hosseini

    Faculty of Nursing & Midwifery, Golestan University of Medical Sciences, Gorgan, Iran
    Contribution
    Investigation, Project administration
    Competing interests
    No competing interests declared
  8. S Yaghoob Jafari

    Faculty of Nursing & Midwifery, Golestan University of Medical Sciences, Gorgan, Iran
    Contribution
    Investigation, Project administration
    Competing interests
    No competing interests declared
  9. Shabnam Roudaki

    Department of Behavioural Studies, Avicenna Institute of Neuroscience, Yazd, Iran
    Contribution
    Formal analysis, Supervision, Investigation, Methodology, Project administration
    Competing interests
    No competing interests declared
  10. Reza Moeeini

    Department of Behavioural Studies, Avicenna Institute of Neuroscience, Yazd, Iran
    Contribution
    Conceptualization, Resources, Data curation, Formal analysis, Supervision, Validation, Methodology, Project administration, Writing—review and editing
    For correspondence
    reza.moeeni123@gmail.com
    Competing interests
    No competing interests declared
  11. Gerlinde AS Metz

    Canadian Centre for Behavioural Neuroscience, University of Lethbridge, Lethbridge, Canada
    Contribution
    Conceptualization, Resources, Supervision, Methodology, Writing—review and editing
    For correspondence
    gerlinde.metz@uleth.ca
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-5801-7716

Funding

Natural Sciences and Engineering Research Council of Canada (5519)

  • Gerlinde AS Metz

Avicenna Institute of Neuroscience (Basic Research Program 41108–010)

  • Reza Moeeini

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

The authors are grateful to Tanzi Hoover for productive comments on the manuscript. We thank B Rezazadeh, J Foroughi, and F Nejad-Ghorban for their assistance in behavioural testing, Dr. GR Namazi for technical assistance, and colleagues at the research ethics board at the Avicenna Institute of Neuroscience (AINS) for suggestions and comments. The authors would also like to thank the animal care staff at the AINS vivarium for assistance with animal husbandry. Special thanks to Dr. M Fakhamati for his suggestions and assistance with statistical analysis. Funding for this study was provided by a basic science research program-AINS (#41108–010) to RM, and by Natural Sciences and Engineering Research Council of Canada Discovery Grant #5519 to GM.

Ethics

Animal experimentation: All procedures in this study were carried out in accordance with the National Institute of Health Guide to the Care and Use of Laboratory Animals, and were approved by the institutional animal care committee (Protocol No. 004674BGH; Avicenna Institute of Neuroscience-AINS).

Senior Editor

  1. Catherine Dulac, Harvard University, United States

Reviewing Editor

  1. Peggy Mason, University of Chicago, United States

Publication history

  1. Received: July 19, 2018
  2. Accepted: November 12, 2018
  3. Accepted Manuscript published: November 13, 2018 (version 1)
  4. Version of Record published: December 3, 2018 (version 2)

Copyright

© 2018, Faraji et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 818
    Page views
  • 142
    Downloads
  • 1
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)

Further reading

    1. Neuroscience
    Matthew C W Oswald et al.
    Research Article
    1. Neuroscience
    2. Structural Biology and Molecular Biophysics
    Dipak N Patil et al.
    Research Article