- Top of page
- Materials and methods
The association between chronic pain and depression is widely recognized, the comorbidity of which leads to a heavier disease burden, increased disability and poor treatment response. This study examined nociceptive responding to mechanical and thermal stimuli prior to and following L5-L6 spinal nerve ligation (SNL), a model of neuropathic pain, in the olfactory bulbectomized (OB) rat model of depression. Associated changes in the expression of genes encoding for markers of glial activation and cytokines were subsequently examined in the amygdala, a key brain region for the modulation of emotion and pain. The OB rats exhibited mechanical and cold allodynia, but not heat hyperalgesia, when compared with sham-operated counterparts. Spinal nerve ligation induced characteristic mechanical and cold allodynia in the ipsilateral hindpaw of both sham and OB rats. The OB rats exhibited a reduced latency and number of responses to an innocuous cold stimulus following SNL, an effect positively correlated with interleukin (IL)-6 and IL-10 mRNA expression in the amygdala, respectively. Spinal nerve ligation reduced IL-6 and increased IL-10 expression in the amygdala of sham rats. The expression of CD11b (cluster of differentiation molecule 11b) and GFAP (glial fibrillary acidic protein), indicative of microglial and astrocyte activation, and IL-1β in the amygdala was enhanced in OB animals when compared with sham counterparts, an effect not observed following SNL. This study shows that neuropathic pain-related responding to an innocuous cold stimulus is altered in an animal model of depression, effects accompanied by changes in the expression of neuroinflammatory genes in the amygdala.
Clinical comorbidity of depression and pain is widely recognized, with over 50% of chronic pain patients experiencing depression, while patients with depression are over twice as likely to develop chronic pain (Bair et al. 2003; Gameroff & Olfson 2006). Animal models provide an important means of understanding the neurobiological basis of depression–pain comorbidity. Depressive-like behaviour has been observed in several animal models of neuropathic pain (Fukuhara et al. 2011; Hu et al. 2009; Suzuki et al. 2007; Wang et al. 2011). Conversely, reserpine-induced monoamine depletion in rats elicits both depressive-like behaviour and mechanical allodynia (Arora et al. 2011; Nagakura et al. 2009). Chronic mild stress, chronic restraint stress and Wistar-Kyoto rat models of depression exhibit hyperalgesia to formalin and complete Freund's adjuvant, models of persistent inflammatory pain (Bardin et al. 2009; Shi et al. 2010b; Wang et al. 2012). Furthermore, mechanical allodynia following peripheral nerve injury is enhanced in Wistar-Kyoto rats (Zeng et al. 2008) and following chronic restraint stress (Norman et al. 2010b) or social isolation (Norman et al. 2010a). The olfactory bulbectomized (OB) rodent is a well-validated animal model of depression (Willner & Mitchell 2002), which exhibits behavioural, neurotransmitter, neuroendocrine and immune changes resembling those reported clinically (Kelly et al. 1997; Song & Leonard 2005). The OB-induced behavioural changes include anhedonia, decreased social behaviour, learning and memory deficits, novelty-induced hyperactivity and reduced sexual behaviour, which are selectively reversed by chronic, but not acute, antidepressant treatment (Kelly et al. 1997; Song & Leonard 2005). Recent studies have shown that OB rats exhibit mechanical allodynia, enhanced formalin-evoked inflammatory pain (Burke et al. 2010; Su et al. 2010) and increased pain responding to electrical stimulation of the dura mater (Liang et al. 2011). However, neuropathic pain responding has not been evaluated in the model, and therefore, the effect of bulbectomy on nociceptive responding to mechanical and thermal stimuli prior to and following spinal nerve ligation (SNL) was evaluated in this study.
Neuroinflammatory processes are well recognized to play important roles in the pathophysiology of both depression and chronic pain (Miller et al. 2009; Panigada & Gosselin 2011; Watkins & Maier 2005). For example, central interleukin-1 beta (IL-1β) plays a key role in chronic stress-induced depressive behaviour (Goshen et al. 2008; Koo & Duman 2008), IL-6 in the amygdala increases immobility in the forced swim test (Wu & Lin 2008), chronic stress exposure increases IL-1β production specifically in the amygdala (Porterfield et al. 2012) and repeated psychosocial stress increases microglial activation in the amygdala (Wohleb et al. 2011). Spinal inflammatory processes are essential for the development of neuropathic pain (Vallejo et al. 2010); however, the role of neuroimmune mediators in supraspinal sites such as the amygdala is less understood. Prostaglandin E2 production is increased in the amygdala in a postoperative pain model (Shavit et al. 2006), and recent data have indicated a role for tumour necrosis factor alpha (TNFα) in the amygdala in anxiety and persistent inflammatory pain (Chen et al. 2013). Reserpine-induced depression–pain syndrome is associated with enhanced central inflammatory cytokines (Arora et al. 2011) and central administration of IL-1ra ameliorates the effects of neuropathic pain on depressive behaviour (Norman et al. 2010b). Thus, glial activation and cytokines in key brain regions such as the amygdala are involved in both affective and nociceptive processing, and may be responsible for the altered nociceptive responding associated with depression. As such, a further aim of this study was to determine if interactions between OB and SNL at a behavioural level are associated with concomitant alterations in the expression of genes encoding for neuroimmune mediators in the amygdala.
- Top of page
- Materials and methods
This study shows that the OB rat model of depression displays mechanical and cold allodynia, and altered neuropathic pain behaviour to an innocuous cold stimulus following SNL surgery. The reduced latency and increased number of responses in the acetone drop test following SNL surgery were positively correlated with IL-6 and IL-10 expression in the amygdala, respectively. Furthermore, the OB-associated increase in the expression of CD11b, GFAP and IL-1β in the amygdala was not observed in SNL animals. This study provides a further preclinical model for studying the association between depression and neuropathic pain, and indicates that neuroimmune processes in the amygdala may, in part, underlie the behavioural changes observed.
Altered nociceptive responding in the OB model of depression
Prior to spinal nerve injury, OB rats exhibited mechanical allodynia, results consistent with those previously reported from our laboratory (Burke et al. 2010); however, to our knowledge, this is the first study to report that OB rats also exhibit cold allodynia. In comparison, nociceptive responding to heat stimuli was not altered in OB rats in this study, although increased paw withdrawal latency to a radiant heat source and transient thermal heat hyperalgesia in the hot plate, but not tail flick, test have been previously reported in OB rats (Burke et al. 2010; Su et al. 2010; Wang et al. 2010). Thus, OB animals appear to exhibit alterations in nociceptive responding to thermal stimuli depending on the test employed. The present findings correlate with those observed clinically where depressed patients exhibit decreased (Bar et al. 2007; Lautenbacher et al. 1994; Schwier et al. 2010), increased (Chiu et al. 2005; Gormsen et al. 2004; Strigo et al. 2008) and no change (Graff-Guerrero et al. 2008) in sensitivity to experimental pain, depending on the modality and intensity of the stimulus.
This study is the first to examine the effect of bulbectomy on neuropathic pain responding, showing that OB rats exhibit altered nociceptive responding to an innocuous cold, but not heat or mechanical, stimulus following SNL. Specifically, OB-SNL rats exhibited reduced latency to respond and reduced number of responses of the ipsilateral hindpaw to an innocuous cold stimulus. We propose that the reduction in response latency may reflect enhanced initial perception of cold stimuli, whereas the reduction in the number of responses might reflect a concomitant reduction in the duration or magnitude of the response in OB-SNL animals. The mechanisms mediating these alterations remain to be determined; however, it is possible that bulbectomy may result in disruption of nociceptive gating and descending pain pathways. Thus, following nerve injury, OB animals may exhibit enhanced gating of cold stimulus-related nociceptive information (reduced latency); however, following initial perception, OB animals may be capable of engaging the descending inhibitory pain pathway more effectively, resulting in reduced magnitude or severity (number of responses) to the stimulus. It is possible that functional alterations in key brain regions part of the descending pain pathway such as the amygdala, and changes in glutamatergic, serotonergic or noradrenergic neurotransmission (Kelly et al. 1997; Song & Leonard 2005) may underlie the enhanced initial perception and/or the decrease in the magnitude of the SNL-induced cold allodynia in OB rats. This idea is discussed further below in the context of the OB-related changes in IL-6 and IL-10 expression observed herein. In contrast to our findings here, SNL and partial sciatic ligation result in a paradoxical increase in mechanical and thermal (heat) thresholds in the unpredictable chronic mild stress (Shi et al. 2010a) and the Flinders sensitive line (Shir et al. 2001) models of depression, respectively. Thus, differential neurobiological mechanisms may underlie the effects of unpredictable chronic mild stress and OB on nociceptive responding following SNL.
In accordance with previous studies (Suzuki et al. 2007; Yalcin et al. 2011), SNL induced anxiety-related behaviour in sham rats in the open field test, exemplified by the reduction in time in the inner area. In comparison, OB-SNL rats did not exhibit anxiety-related behaviour but maintained the OB-induced increase in locomotor activity on exposure to the open field. Although it cannot be ruled out that SNL may induce anxiety-related behaviour in OB animals at times or in paradigms other than those examined in this study, the present data indicate that OB rats exhibit resilience to SNL-induced anxiogenesis in the open field.
Gene expression changes in the amygdala associated with altered nociceptive responding in the OB model
Limbic regions such as the amygdala play key roles in the processing of emotion and pain, and inflammatory mediators in discrete brain regions modulate both affective and nociceptive processing. Thus, neuroimmune processes in regions such as the amygdala may be the driving force for altered nociceptive responding associated with depression. Previous studies have shown that OB rats exhibit increases in TNFα and/or IL-1β levels in the prefrontal cortex, hippocampus and hypothalamus (Borre et al. 2012; Myint et al. 2007), increased phospholipase A2 and prostaglandin E2 in the hypothalamus (Song et al. 2009) and increased GFAP in the frontal cortex (Cizkova et al. 1997). To our knowledge, this is the first study to investigate the effect of bulbectomy on the expression of immune mediators in the amygdala, showing enhanced gene expression of CD11b and GFAP, markers of microglial and astrocyte activation, respectively, and of the proinflammatory cytokine IL-1β in the amygdala of OB rats. OB-induced hyperactivity has been shown to be associated with neurodegeneration within the amygdala (Jarosik et al. 2007; Wrynn et al. 2000), effects that may result from neuroinflammatory processes in this region following removal of the bulbs. In addition, increased glial activation and IL-1β within the amygdala may be responsible for OB-induced mechanical and cold allodynia, as it has been shown that intracerebroventricular administration of non-pyrogenic doses of IL-1β results in thermal (Oka et al. 1993) and mechanical (Yabuuchi et al. 1996) hyperalgesia. Interestingly, the increase in the expression of IL-1β in the amygdala of OB rats was not observed following SNL, suggesting that spinal nerve injury can attenuate the increases in amygdaloid IL-1β expression that result from central injury (OB). Indeed, SNL induced a decrease in IL-6 and an increase in IL-10 expression in the amygdala. The lack of effect of SNL on the expression of CD11b, GFAP or other proinflammatory cytokines (IL-1β and TNFα) may not be surprising given the time at which these were examined post-surgery (day 22). Previous studies have shown that GFAP, IL-1β and TNFα protein levels are increased in the brain between days 3 and 7 post-surgery (Liu et al. 2007; Marcello et al. 2013; Xie et al. 2006). In comparison, IL-10 protein levels in the brain have been reported to increase over time following SNL, with highest levels observed 21 days post-surgery (Xie et al. 2006), correlating with the present findings.
Latency to respond to a cold thermal stimulus was positively correlated with IL-6 expression in the amygdala; thus, enhanced initial cold perception observed in OB animals is associated with, and possibly mediated by, low IL-6 levels in the amygdala. The number of responses to the innocuous cold stimulus was positively correlated with IL-10 expression in the amygdala. As mentioned, IL-10 levels in the brain have been reported to be enhanced following SNL (Xie et al. 2006), although region-specific changes have not been investigated. Pharmacological and genetic deletion of IL-10 is associated with depressive-like behaviour (Mesquita et al. 2008) and increased thermal nociceptive thresholds (Tu et al. 2003). Interleukin-10 is known to protect against glial and neuronal cell death (Bachis et al. 2001; Strle et al. 2002), and thus changes in IL-10 expression may alter neuronal integrity and function within the amygdala. Such changes may increase descending inhibitory pain regulation and/or reduce descending facilitatory pain modulation, resulting in reduced magnitude or severity (number of responses) to the cold stimulus as observed in OB-SNL rats.
In conclusion, the present findings show that in the presence of a depressive-like phenotype, neuropathic pain behaviour is altered depending on stimulus modality, effects accompanied by alterations in inflammatory mediator gene expression within the amygdala. Increased understanding of the neurobiological substrates underlying depression, chronic pain and the interaction between these disorders may provide novel therapeutic targets for treating these debilitating comorbid disorders.