**4. Discussion**

Peripheral leukocyte inflammatory cytokine gene expression profiles were investigated in calves in response to cautery disbudding in this study. Both control and meloxicam groups showed significant changes in a variety of pro- and anti-inflammatory cytokine mRNA after disbudding. There was a significantly higher transcription of *TNF*α gene at 24 h post-disbudding in both groups, but no significant difference was detected at 4 h after disbudding, compared to respective baseline values. *TNF*α is mainly secreted by activated macrophages after tissue injury and plays a pivotal role in the initiation of acute phase response along with other early pro-inflammatory cytokines such as *IL1B*, *IL6*, and *IFN*γ [23]. Disbudding with hot iron involves the burning of a ring of tissue containing horn bud cells. The systemic inflammatory response following a burn injury in mice has been demonstrated to induce the release of a myriad of pro- and anti-inflammatory cytokines, with a peak of serum *TNF*α detected at 24 h and 48 h post-burn [24]. Only a few studies [18,19] have assessed the expression of cytokines and other inflammatory mediators in response to calf disbudding. In the present study, it is likely that the mRNA levels of *TNF*α have started rising in the window between 4 and 24 h and by 24 h after disbudding, a significant increase from the basal level was found. Nonsteroidal anti-inflammatory drugs have been shown to upregulate the production of *TNF*α from human peripheral blood leucocytes [25,26]. Results from the current study appear to support this as the *TNF*α mRNA of meloxicam group, similar to the control group, significantly increased from baseline by 24 h post injury.

*IL1B*, *TNF*<sup>α</sup>, and *IFN*γ have been postulated to be specific and early markers of inflammation and nociception. No significant differences in *IL1B*, *TNF*<sup>α</sup>, and *IL6* mRNA abundance between sham-handled and disbudded calves treated with a lidocaine block and IV meloxicam have been reported in one study [18]. In the present study, the mRNA transcripts of both *IL1B* and *INF*γ were found to be significantly elevated in the meloxicam group post-disbudding, compared to the respective baseline values and values of the control group at 4 h and 24 h. A previous study in calves reported a similar elevation of these early pro-inflammatory cytokine levels as early as 15 min after disbudding [19]. In the current study, it is likely that the mRNA of these pro-inflammatory cytokines had begun to rise at an earlier time point than 4 h, which was the first sampling time point after disbudding in our study. The finding that meloxicam did not attenuate the rise in early pro-inflammatory cytokine mRNA in this study appears to support the previous literature that the administration of only a systemic NSAID without local anaesthetic prior to disbudding/dehorning did not completely mitigate the acute phase response [9,19,27].

Calves in both groups showed a significantly elevated transcription of the *IL8* gene at 4 and 24 h post-disbudding (Figure 1a). *IL8* is a pro-inflammatory cytokine produced by a variety of tissues and

blood cells in response to inflammation, and it exhibits chemotaxic activities against neutrophils and T lymphocytes, drawing these cells to the site of inflammation [28]. *IL8* has been shown to evoke hyperalgesia in rats by a prostaglandin-independent mechanism [29]. More recently, plasma *IL8* levels were found to be positively correlated with the intensity of burning mouth syndrome pain in humans [30]. In the current study, the significantly lowered transcription levels of this gene in the meloxicam group compared to those in the control group at 4 h post-disbudding potentially indicate the ability of meloxicam to reduce the pain caused by noxious stimuli via its anti-inflammatory e ffect. This is supported by the observed significantly higher transcription of the *IL10* gene in the meloxicam group at this time point (Figure 1a). In the current study, the significantly lowered transcription levels of this gene in the meloxicam group compared to those in the control group at 4 h post-disbudding potentially indicate the ability of meloxicam to reduce the pain caused by noxious stimuli via its anti-inflammatory e ffect. This is supported by the observed significantly higher transcription of *IL10* gene in the meloxicam group at this time point (Figure 1a). A similar negative regulatory e ffect of *IL10* on *IL8* expression in human monocytes has been documented [31]. It is interesting to see markedly elevated transcription levels of *IL8* in the leukocytes of calves in both groups prior to disbudding. This could be due to the age of the calves, as it has been found that *IL8* levels in healthy infant humans are significantly higher than those in adults [32]. The high *IL8* levels in infants have been attributed to be a major T cell e ffector function that has the potential to activate antimicrobial neutrophils and γδ T cells.

*IL6* is a Janus-faced complex cytokine. It is one of the pro-inflammatory cytokines released early in the cascade [33] and induces the production of acute phase proteins. It also acts as an anti-inflammatory cytokine. *IL6* production has been shown to be significantly upregulated in skin cells close to a heat-induced injury site in rats at 24 h post-injury [34]. The anti-inflammatory e ffect of IL6 is thought to be mediated, in part, through the induction of prostaglandin E2 (PGE2) synthesis, which in turn leads to the inhibition of *TNF*α and *IL1* receptors and subsequently their production [35,36]. In the present study, *IL6* mRNA counts did not significantly change in the meloxicam group, but the control group exhibited a significant increase at 24 h after disbudding. The precise time point at which the anti-inflammatory e ffect of *IL6* initiates in the course of inflammation relative to tissue injury is unknown. Treatment with NSAIDs for 3–6 days after injury in humans has been shown to result in lesser concentrations of serum *IL6* compared to a control group [37]. Thus, it appears that meloxicam had stabilized the *IL6* mRNA levels compared to the control group post-disbudding in the current study.

In the present study, the mRNA counts of *PGHS2*, *NOS1*, substance P (*TAC1*), *CRH*, and *NGF* showed no significant di fferences in transcription between pre-and post-disbudding time points in the meloxicam group. The control group showed a significant increase in the mRNA counts of these neuroactive ligand–receptor interaction molecules at 24 h post-disbudding. The *PGHS2* (aka cyclooxygenase 2, cox-2) gene encodes the inducible isozyme cox-2, which is predominantly involved in inflammatory prostanoid biosynthesis in response to injury [38]. Meloxicam is a preferential cox-2 inhibitor [39] and thus inhibits inflammatory prostanoids to produce its therapeutic e ffects in injured animals. Our finding of lower *PGHS2* mRNA counts in the meloxicam group could possibly reflect decreased ex vivo prostaglandin E2 synthesis in the plasma after cautery dehorning [40].

*TAC1* (a precursor of Substance P, SP) is a neuropeptide member of the tachykinin family and is widely distributed in the central, peripheral, and enteric nervous systems. Neuronal SP is released from sensory neurons on noxious stimulation and activates several immune cells such as macrophages, mast cells, and T lymphocytes [41]. Sensory neuropeptide activated immune cells release inflammatory mediators such as histamine, arachidonic acid derivatives, and cytokines/chemokines. The synergistic interplay between SP and prostaglandins has been demonstrated in various inflammation and pain models [42]. Lower plasma SP concentrations compared with a control group have been reported in meloxicam-treated calves following scoop dehorning without local anaesthesia [9], which is in line with the lower mRNA counts of *TAC1* in the meloxicam group at 24 h after disbudding in our study.

Nitric oxide synthases (*NOS1*) catalyse the oxidation of the amino acid L-arginine to produce the free radical, nitric oxide (NO). Nitric oxide has been shown to activate the cox enzymes for prostanoid biosynthesis [43] and facilitate prostaglandin-induced hyperalgesia in rats [44]. The inducible isoform of NOS (iNOS) is released from macrophages in response to inflammatory stimuli [45]. Nonsteroidal anti-inflammatory drugs have been shown to inhibit the expression of iNOS mRNA in rat in vitro studies [45]. Likewise, the co-administration of meloxicam and iNOS inhibitor produced a synergistic anti-inflammatory effect in carrageenan-induced acute inflammation in rats [46]. The findings from the current study support those of previous studies demonstrating the efficacy of NSAIDs in inhibiting the NO synthesis, with significantly lower mRNA of *NOS1* in the meloxicam-treated calves compared to the control calves.

The *CRH* gene encodes a member of the corticotropin-releasing hormone family and is mainly expressed by the hypothalamic paraventricular nucleus and secreted into the hypophyseal portal system [47]. It plays a crucial role in eliciting the stress response through stimulation of the hypothalamic, pituitary, adrenal axis and the secretion of cortisol in response to noxious stimulation [48]. In the present study, post-disbudding *CRH* mRNA were significantly higher in the control group than the meloxicam group at 4 h, and no difference was found in both groups at 24 h after disbudding. Plasma cortisol concentration peaks within 30 min and returns to pre-treatment levels 6–8 h after cautery disbudding [6]. It is likely that more obvious changes in *CRH* mRNA have occurred earlier than 4 h post-disbudding, and only a slight increase could be detected in the control group at 4 h in the current study.

*NGF*, a member of the neurotrophin family, is essential for the development and maintenance of both central and peripheral nervous systems [49]. Rats injected with *NGF* into paws showed rapid and prolonged hypersensitivity to noxious thermal stimulation, confirming its role in inflammatory pain [50]. This inflammatory hyperalgesic effect of *NGF* is thought to be primarily mediated via tropomyosin receptor kinase A (TrkA) receptors [51]. In the current study, it is interesting to note that while there was a significantly higher transcription of NGF in the control group calves at 4 h post-disbudding, its transcription in meloxicam-administered calves at 4 and 24 h after disbudding remained relatively unchanged compared to baseline values.

Two recent studies [52,53] revealed the *AGTR2* to be a promising target for therapeutics aimed at treating neuropathic pain. The studies showed that damage to the peripheral nerve might lead to pain hypersensitivity as a result of signalling through *AGTR2* found on peripheral macrophages infiltrating the site of injury, rather than those on sensory neurons. The activation of *AGTR2* on peripheral macrophages triggers the release of reactive oxygen/nitrogen, which in turn activates the transient receptor potential ankyrin 1 ion channel, thus leading to nociceptive signaling in sensory neurons [52]. The current study also revealed the significantly higher transcription of this gene at 4 and 24 h post-disbudding in the calves of both groups (Figure 1a), indicating that meloxicam administration could not attenuate the transcription of this gene in the peripheral macrophages.

*IL10* is a potent anti-inflammatory cytokine, reducing the expression of pro-inflammatory cytokines to balance the inflammatory response to injury [54]. A significant increase in *IL10* gene expression has been reported 12 h after the burdizzo castration of cattle [15]. The pre-emptive administration of an NSAID, carprofen, has been shown to have no effect on *IL10* mRNA expression [55]. Meloxicam has been shown to have no effect on the production of *IL10* from bovine lymphocytes [56] despite the involvement of cox-2 (target of meloxicam) in the production of *IL10* [57]. Similarly, in the current study, meloxicam did not affect the upregulation of IL10 mRNA. Taken together, it appears that NSAIDs such as meloxicam and carprofen do not inhibit cox-2 activity to the extent of suppression of different populations of immune cells that produce *IL10* [56].

In the current study, a representative *post-hoc* power analysis pertaining to the mRNA data of two genes, *IL8* (at 4 h) and *PGHS2* (at 24 h), revealed that there was adequate power (>0.9, except for between-group differences in the case of *IL8* at 4 h, which was 0.71) of detecting between-group and between time-point differences. In the case of *IL8*, a sample size of 10 per group would provide

0.8 power for between-group comparison. It is to be noted that these power analyses were based on assuming a simple t-test, but for the actual analysis of the study data, a more power linear mixed model, which would better account for correlation between repeated measures, was employed.

It has been shown in humans [58] that cytokine increases in cerebrospinal fluid (CSF) during peripheral surgery are more marked compared to those in blood, indicating the role of pro-inflammatory cytokines in increased central nervous system sensitivity to surgical pain. Hence, to check if a similar trend was evident in the current study, a 1–2 mL CSF was obtained from the atlanto-occipital joint of all calves prior to and 4 and 24 h post-disbudding. However, attempts to quantify the transcription of cytokine and neuroactive ligand-receptor genes (as well as housekeeping genes) in the RNA purified from those samples were unsuccessful due to the very low yield of RNA (<20 ng/sample) being inadequate for the employed nCounter gene expression assay. A volume of 8–10 mL of CSF might contain an adequate number of cells, yielding the required mRNA quantity for the nCounter gene expression assay.

nCounter gene expression assay (NanoString, Seattle, WA, USA) has been employed in this study to explore the amplification-free expression of peripheral leukocyte inflammatory cytokine genes. The transcription of genes, as actual mRNA counts in relation to the inbuilt positive controls as well as the three selected reference genes (*GAPDH, GUSB, and YWHAZ*) was successfully quantified. Although the expression of only 12 genes (plus 3 reference genes) was investigated in the current study, the expression of as many as 96 genes (of choice) in total could be simultaneously investigated using this assay. Using this technique, differential transcription of a few cytokine and neuroactive ligand-receptor genes was detected in the peripheral leucocytes of claves post-disbudding. Disbudding resulted in an increased transcription of pro-inflammatory cytokine genes (such as *IL1*β, *IFN*γ, *IL8*, and *TNF*α) in all calves (control as well as meloxicam groups), while meloxicam administration attenuated the upregulation of a few other genes (*IL6*, *PGHS2*, *TAC1*, *NOS1*, and *CRH*) involved in pain sensitisation pathways. These findings indicate that meloxicam alone would not be able to completely reduce the pro-inflammatory response following cautery disbudding. However, the nCounter assay seems to be an efficient tool to screen combinations of different analgesics for their ability to attenuate the pro-inflammatory response of a variety of genes. However, further studies need to be undertaken to validate the findings of this study. It would be interesting to see if this differential transcription of genes reflects in terms of actual protein levels, using MILLIPLEX® Cytokine/Chemokine panels (EMD Millipore Corporation, Billerica, MA, USA) that can simultaneously quantify up to 15 different cytokines. Similarly, corroboration of the differential expression of cytokines with either behaviour-based pain scores or other objective methods such as electroencephalogram variables would be useful.
