Neonatal Handling Produces Sex Hormone-Dependent Resilience to Stress-Induced Muscle Hyperalgesia in Rats

Animals

Primiparous timed-pregnant female Sprague Dawley rats (Crl:CD) were obtained from Charles River (Hollister, CA). Dams were housed with their litter in standard cages on postnatal days 0–1. On day 2, litters were assigned to handled (see below) or standard care (control) conditions. Behavioral experiments in adults were performed on 250–400 g (age: 50–75 days) male and female rats from these litters. Animals were housed in the Laboratory Animal Resource Center of the University of California, San Francisco, under a 12-hours light/dark cycle (lights on 7 am) and environmentally controlled conditions; ambient room temperature was 21–23 °C, and food and water were available ad libitum. Upon completion of experiments, rats were euthanized by CO₂ induced asphyxia followed by bilateral thoracotomy. Animal care and use conformed to NIH guidelines (NIH Guide for the Care and Use of Laboratory Animals) and to the guidelines of the Committee for Research and Ethical Issues of the International Association for the Study of Pain. All experimental protocols were approved by the Institutional Animal Care and Use Committee of the University of California, San Francisco.

Gonadectomy and drug treatment

Gonadectomy was performed on postnatal day 25–26, under general anesthesia (isoflurane 2.5%) and infiltration of the incision area with local anesthetics (0.25% bupivacaine, Marcaine^®, Hospira, Lake Forest, IL); all animals received preoperative carprofen (Rimadyl^®, Zoetis, Kalamazoo, MI). Briefly, ovaries were accessed by means of bilateral cutaneous and peritoneum incisions. Once located, ovaries and their vascular bundles were ligatured with 4-0 silk suture (Perma-Hand Silk^® Ethicon, Johnson & Johnson, Somerville, NJ). Ovaries were then excised, and the peritoneal and cutaneous incisions closed with 5-0 silk suture (Perma-Hand Silk^® Ethicon, Johnson & Johnson, Somerville, NJ). The testes were accessed by means of a single cutaneous incision made through the scrotum. Their vascular bundles were tied off with 4-0 silk suture, and the testes removed. The cutaneous incision was closed with 5-0 silk suture.

The selective estrogen receptor degrader (SERD) fulvestrant (ICI 182,780, Sigma-Aldrich, St. Louis, MO) was dissolved in DMSO (5%) and corn oil (95%), and administered by means of osmotic pumps (ALZET^® mini-osmotic pump model 2004, delivering 0.25 μl/hr, 200 μl reservoir; DURECT Corporation, Cupertino, CA) at a rate of 0.1 mg/kg/day, as previously reported.¹⁵ To ensure adequate fulvestrant levels throughout the experiments, rats were implanted on postnatal day 25–26 (at the time of ovariectomy) and again on postnatal day 52–53. Control rats were implanted with an inert implant. Nociceptive thresholds were measured beginning 2 weeks after the last implantation surgery.

Neonatal handling

NH experiments were performed by using a well established neonatal protocol that induces resilience to stress in the adult rat.^{3, 30, 36, 38} This protocol involves removing the pups from the home cage, placing them in a separate container and carefully handling them for 15 min. Litters were handled daily on postnatal days 2–9. On postnatal day 21, pups were weaned and same-sex housed, 3 per cage.

Water avoidance-induced stress

The WA protocol in the rat,⁷ which produces mechanical hyperalgesia in the skeletal muscle^{3, 10, 19} was used as a psychological chronic stressor. Briefly, adult rats (aged 7–12 weeks) were placed on an acrylic platform (8 × 8 cm, 10 cm high) in the center of a clear plastic tank (45 cm length × 25 cm width × 25 cm height) filled with room temperature tap water to a depth of 9 cm for 1 hr/day for 10 consecutive days. One day after the last stress exposure, rats were tested for mechanical hyperalgesia in the gastrocnemius muscle.

Antisense oligodeoxynucleotides

The androgen receptor (AR) is expressed in dorsal root ganglion (DRG) neurons.²⁵ To evaluate the role of androgens, acting at their receptor on nociceptors innervating the gastrocnemius muscle, the expression of the AR was attenuated by administration of antisense oligodeoxynucleotide (AS ODN) directed against a unique sequence of the respective encoding mRNA, in the rat: 5′-GTG CAA TCA TTT CTG-3′. The corresponding GenBank Accession Number and ODN position within the cDNA sequence are NM_012502.1 and 2733–2747, respectively in Rattus norvegicus. The mismatch (MM) ODN sequence, 5′-GTA CAA TCC TTT GTG-3′, corresponds to the AR antisense sequence with 3 bases mismatched (indicated in bold typeface). The AS and MM ODNs were synthesized by Invitrogen (Carlsbad, CA). The AS ODN sequence has been shown to produce significant knockdown of AR.²¹ We have previously shown that the intrathecal administration of AS ODN is able to block protein expression in sensory neurons and modulate muscle hyperalgesia.^{2, 20} The ODNs were reconstituted in sterile 0.9% NaCl (10 μg/μl), aliquoted and stored at −20°C until use. Before injections, ODN aliquots were diluted in sterile 0.9% NaCl to a final concentration of 2 μg/μl. Rats were briefly anesthetized with 2.5% isoflurane and a 29-gauge hypodermic needle (Becton Dickinson, Franklin Lakes, NJ) inserted into the subarachnoid space between the L4 and L5 vertebrae and a dose of 40 μg (volume 20 μl) of ODN injected. Intrathecal location of injections was confirmed by observing a tail-flick, as previously reported.⁴¹ This procedure was repeated daily for 3 consecutive days and then every other day, for a total of 8 ODN injections.

Muscle mechanical nociceptive threshold

Mechanical nociceptive threshold was quantified in the gastrocnemius muscle using a Chatillon^® digital force transducer (model DFI2, Amtek Inc., Largo, FL). Rats were placed in cylindrical acrylic restrainers designed to provide adequate comfort and ventilation, allow extension of the hind leg from the cylinder, and minimize restraint stress. All rats were acclimatized to the testing procedure, and testing performed in parallel across groups. Rats were adapted to the restrainer for 1 hr prior to starting each study and for 30 min prior to experimental manipulations. To measure nociceptive threshold, a 6 mm-diameter probe attached to the force transducer applied to the gastrocnemius muscle, to deliver a compression force that increased with time. The nociceptive threshold was defined as the force, in milliNewtons (mN), at which the rat withdrew its hind leg from the stimulus. Results from each rat correspond to the mean of 3 readings per gastrocnemius muscle, taken at 5 min intervals. Rats were assigned randomly to experimental groups and behavioral readings were taken blind to treatment.

Statistics

Group data are expressed as mean ± SEM of n distinct rats. Statistical comparisons were made by means of two- or three- way repeated measures analysis of variance (ANOVA), followed by Bonferroni’s or independent samples t-tests, as appropriate. Statistical significance was set at P < 0.05. Statistical Package for the Social Sciences (SPSS) version 24 (International Business Machines Corporation, Armonk, NY) and Prism 6.0h (GraphPad Software, Inc., La Jolla, CA) were used to perform data analysis and graph plotting.

Effect of neonatal handling in gonad-intact rats

Three-way repeated measures ANOVA allowed evaluating within-subjects effect of time (pre-WA or post-WA, F_1,20= 644.516, P < 0.001), between-subjects sex (male or female, F_1,20= 529.630, P < 0.001), and treatment (control or NH, F_1,20=160.157, P < 0.001) effects. This analysis showed a significant time × sex × treatment interaction (F_1,20= 41.715, P < 0.001), indicating that the effect of the NH on muscle nociceptive threshold observed over time depends on sex.

Post-hoc analysis showed no significant differences in nociceptive threshold between control and NH males at baseline (P = 0.376). However, consistent with our previous observations (Alvarez et al., 2015), control males did differ from NH males in nociceptive threshold after WA (n = 6/group, P < 0.001). Similarly, while control females did not differ from NH females at baseline (P = 0.382), control and NH females exhibited differences in nociceptive threshold after WA (n = 6/group, P = 0.006, Fig. 1A). However, the magnitude of mean difference (Control post-WA vs NH post-WA) was much greater for males (729.1) than it was for females (154.0), indicating a sex difference in the protective effect of NH in gonad-intact rats (Fig. 1A).

Effect of neonatal handling in gonadectomized rats

Our three-way repeated measures ANOVA allowed evaluation of within-subjects effect of time (pre-WA or post-WA, F_1,20= 1339.822, P < 0.001), between-subjects sex (male or female, F_1,20= 7.201, P = 0.013) and treatment (control or NH, F_1,20=258.373, P < 0.001) effects. This analysis showed a significant time × sex × treatment interaction (F_1,20= 12.912, P = 0.001), indicating that the effect of the NH on muscle nociceptive threshold observed over time depends on sex, in gonadectomized rats.

Post hoc analysis showed that, in gonadectomized males, the control group (n = 6) differed from the NH group (n = 9) at baseline (P < 0.001). After WA exposure, gonadectomized control and NH males also showed significant differences in nociceptive threshold (P < 0.001, Fig. 1B). In gonadectomized females, the control group (n = 6) differed from the NH group (n = 9) at baseline (P < 0.001). After WA exposure, gonadectomized control and NH females also exhibited significant differences in nociceptive threshold (P < 0.001, Fig. 1B).

Effect of long-term fulvestrant

Given that the attenuation of WA-induced hyperalgesia by NH is enhanced by ovariectomy, and that extra-ovarian sources of estrogen⁶⁰ could contribute to reduce the protective effect of NH, we explored the effect of long-term administration of the anti-estrogen drug fulvestrant in gonadectomized NH females. Two-way repeated measures ANOVA showed a significant effect for time (pre-WA, post-WA, F_2,14= 87.40, P < 0.0001), but not for treatment (control or fulvestrant, F_1,7= 0.406, P < 0.0001), or time × treatment interaction (F_2,14= 0.217, P < 0.0001) on muscle nociceptive threshold, suggesting that compared to control implants (n=4), fulvestrant (n=5) did not enhance the effect of gonadectomy in NH female rats, measured 1 day after last exposure to WA stress (P > 0.05; Fig. 2).

Effect of androgen receptor antisense

Since orchiectomy abolished the effect of NH in males we sought to explore whether a protective effect of male sex hormones (presumably testosterone) is dependent on its effect at its cognate receptor on nociceptors. Therefore, we administered AS ODN directed against AR mRNA intrathecally, to attenuate the expression of the AR receptor in sensory neurons. Two-way repeated measures ANOVA revealed a significant effect for treatment (AS ODN or MM ODN, F_1,7= 13.12, P = 0.0085), time (baseline, post ODN×3, postWA (post ODN×8), F_2,14= 47.06, P < 0.0001) and treatment × time interaction (F_2,14= 15.18, P = 0.0003) in NH male rats 1 day after last exposure to WA stress. Bonferroni’s post-hoc test showed that, while the AS ODN intervention (n=5) did not significantly change baseline nociceptive threshold compared to MM ODN (control) treatment (n=4, P > 0.05, Fig. 3), it significantly decreased the nociceptive threshold after exposure to WA stress (i.e., increased muscle hyperalgesia) compared to MM ODN treatment (P < 0.05; Fig. 3).

1. Sex differences in NH protection against WA-induced muscle hyperalgesia

While NH fully prevented WA stress-induced muscle hyperalgesia in males, its protective effect was small, albeit statistically significant, in females. Sex differences of NH have been observed in other paradigms exploring its effect on behavioral and neuroendocrine responses to stress in adults. For example, in contrast to NH females, NH males exhibited shorter immobility time to chronic forced swimming stress suggesting enhanced capacity to cope with chronic stressors.⁵¹ Furthermore, NH reduced restraint stress-induced increases in adrenocorticotropic hormone and corticosterone levels in adolescent male, but not female rats.⁵²

Two mechanisms could be involved in the induction of sex differences observed in NH: first, evidence consistently shows that dams spend significantly more time caring for their male than female offspring^{43, 44, 56}, which could limit the effect of NH in females; second, sex hormones display important organizational effects on the developing HPA axis during early neonatal life.^{37, 45, 48} Indeed, neonatal orchiectomy increases corticosterone secretion in response to stress, whereas hormone replacement in adults did not significantly reverse this change, compatible with the suggestion that the HPA axis is less responsive to testosterone in neonatally-gonadectomized males.³⁸ In this case, the induction of the protective effect of NH would depend on an appropriate sex hormone context during the critical period, namely the influence of androgens. In addition to these mechanisms, the expression of the NH phenotype is also likely due to ‘activational’ mechanisms, which depend on ongoing secretion of androgens (see below).

2. Reduction of the protective effect of NH by female sex hormones

In female rats, ovariectomy modestly enhanced NH-induced antihyperalgesia, suggesting that estrogens can attenuate the expression of the protective effect of NH. Estrogen is known to enhance HPA axis responsiveness, producing greater increased secretion of corticosterone and adrenocorticotrophin at baseline and in response to stress in females compared to males.^{12, 26, 48} Furthermore, ovariectomized females have significantly lower basal and stress-induced corticosterone levels than gonad intact females.⁵⁹ These observations indicate that estrogen likely increases the responses of the HPA axis evoked by WA-induced stress, contributing to the reduced inhibitory effect of NH on WA-induced muscle pain. However, the administration of the antiestrogen fulvestrant, to gonadectomized females, did not further increase the protective effect of NH on WA-induced hyperalgesia. Thus, although estrogens partially counteract NH-induced antihyperalgesia, its absence does not uncover the NH protective effect observed in males, suggesting that it is not the presence of estrogens but the lack of androgens, which underlies the small protective effect of NH observed in females.

3. Androgens play a major role in expression of NH-induced resilience

Upon exposure to WA stress, gonadectomized NH males displayed muscle hyperalgesia of magnitude comparable to control (non-handled) WA-exposed males, supporting a major role for androgens in NH-induced resilience. Several mechanisms could be involved in the role of androgens in such resilience. While the effects of androgens may have been due to an organizational effect on the developing brain during the neonatal period⁴⁵, post pubertal orchiectomy eliminated the protective effect of NH, suggesting that ongoing secretion of androgens is needed for the expression of NH-induced resilience to WA-induced muscle pain in the adult. Androgens may also act on the HPA axis allowing reduced stress responses to attenuate WA-induced hyperalgesia. Indeed, experiments in knockout mice show a role for AR in regulating anxiety-like behaviors and HPA axis responses.⁹ Also, since ARs are expressed in sensory neurons,²⁵ androgens acting at this level may attenuate nociceptive inputs from the skeletal muscle nociceptors, contributing to the NH-induced phenotype. Indeed, masseter muscle inflammation-induced up-regulation of mu-opioid receptor (MOR) mRNA expression in the trigeminal ganglia is prevented by orchiectomy and restored by testosterone replacement.⁶⁸ Furthermore, TRPV1-positive trigeminal sensory neurons co-express AR, which transcriptionally regulates MOR expression in these neurons induced by inflammation.²⁹ These observations are consistent with our data from intrathecal antisense experiments, showing that the protective effect of NH is significantly attenuated after knockdown of the AR.

Finally, the contribution of androgens to NH-induced protection against stress-induced muscle hyperalgesia could also be due to an effect on neuroendocrine-immune responses triggered by stress. Indeed, testosterone inhibits macrophage activation and their production of pro-inflammatory cytokines,^{14, 55} and induces the synthesis of the anti-inflammatory/antinociceptive cytokine IL-10.¹⁴ Testosterone also modulates the immune response of lymphocytes, inhibiting the release of the pro-inflammatory cytokine interferon gamma and enhancing the release of anti-inflammatory IL-10.^{5, 34} Importantly in this regard, there is a major contribution of pro-inflammatory cytokines to WA stress-induced hyperalgesia.^{28, 47}

Conclusion

The resilience phenotype induced by NH in males is likely due to an interaction between decreased response of the HPA axis induced by enhanced maternal care in early-life and secretion of androgens. By acting on muscle nociceptors androgens contribute to the NH resilient phenotype to stress-induced muscle pain. In females, the resilience phenotype confers only a small protection to stress-induced muscle pain. Since sex hormones also play a role in neuropathic^{8, 32, 63} and inflammatory^{1, 6, 16, 35} pain, exploring the interaction between androgens and NH-induced resilience in preclinical models of these pain syndromes is an important future direction.