Infraorbital nerve injury triggers sex-specific neuroimmune responses in the peripheral trigeminal pathway and common pain behaviours

Trigeminal neuropathic pain is emotionally distressing and disabling. It presents with allodynia, hyperalgesia and dysaesthesia. In preclinical models it has been assumed that cephalic nerve constriction injury shows identical molecular, cellular, and sex dependent neuroimmune changes as observed in extra-cephalic injury models. This study sought empirical evidence for such assumptions using the infraorbital nerve chronic constriction model (ION-CCI). We compared the behavioural consequences of nerve constriction with: (i) the temporal patterns of recruitment of macrophages and T-lymphocytes at the site of nerve injury and in the tri-geminal ganglion; and (ii) the degree of demyelination and axonal reorganisation in the injured nerve. Our data demonstrated that simply testing for allodynia and hyperalgesia as is done in extra-cephalic neuropathic pain models does not provide access to the range of injury-specific nociceptive responses and behaviours reflective of the experience of trigeminal neuropathic pain. Similarly, trigeminal neuroimmune changes evoked by nerve injury are not the same as those identified in models of extra-cephalic neuropathy. Specifically, the timing, magnitude, and pattern of ION-CCI evoked macrophage and T-lymphocyte activity differs between the sexes.


Introduction
Peripheral neuropathic pain arises from damage to somatosensory nerves.It is often difficult to treat in part due of the complexity of the neural changes that underly the pain state (Ghazisaeidi et al., 2023).Trigeminal neuropathic pain is described as particularly distressing and usually presents with spontaneous pain, 'evoked' pain including, heightened sensitivity to mechanical and thermal stimuli (allodynia and hyperalgesia) and dysaesthesia (Smith et al., 2013).The pivotal role of the head and face in almost every organismal function makes trigeminal neuropathy intensely disabling.Preclinical models of neuropathic pain largely focus on damage to spinal nerves or extra-cephalic peripheral nerves, and it is often assumed that cephalic counterparts are likely to show similar responses to injury or damage.An assumption challenged by findings that show distinct molecular, cellular, and immune changes between cephalic and extra-cephalic pain pathways (Krzyzanowska and Avendaño, 2012;Latrémolière et al., 2008).
In general, preclinical assessment of 'pain' relies on sensory-evoked motor responses elicited by the application of either non-nociceptive stimuli to measure allodynia or nociceptive stimuli to measure hyperalgesia (Sadler et al., 2022).When studying sensory-evoked behaviours in the orofacial region, one of the major difficulties is accurately identifying nociceptive or nocifensive responses.In extra-cephalic models, the application of mechanical or thermal stimuli evokes a hindlimb response that can be characterised by paw withdrawal latency, frequency, or duration.A similar approach has been employed in many cephalic models of neuropathic pain, where the application of mechanical or thermal stimuli have been used to evoke the equivalent of a hindlimb response, that is head withdrawal latency, frequency, or duration (Krzyzanowska and Avendaño, 2012;Liu et al., 2023;Sadighparvar et al., 2023).However, in the orofacial region, nociceptive responses may vary from swiping the stimulus, head withdrawals, to Abbreviations: CCI, chronic constriction injury; CD68-IR, CD68-immunoreactive; CGRP, calcitonin gene-related peptide ; ION, infraorbital nerve; ION-CCI, chronic constriction injury of the infraorbital nerve; LFB, luxol fast blue; MBP, myelin basic protein; NHS, normal horse serum; PBS, phosphate buffered saline; SEM, standard error of the mean; T-cell, T lymphocyte; TCRαβ, T-lymphocyte receptor αβ; V1, Ophthalmic Division,Trigeminal Ganglion;V2,Maxillary Division,Trigeminal Ganglion;V3,Mandibular Division,Trigeminal Ganglion.facial grooming (Vos et al., 1994).Each of these possible responses should be captured and analysed to fully appreciate the functional implications of injury to the orofacial region.A significant problem for these studies lies in the fact that similar behavioural responses also occur following stimulus detection in other sensory fields (visual, auditory, and olfactory), each of which may evoke movement of the head and neck, including active avoidance of the stimulus.A simple solution in many studies which has been reviewed in Krzyzanowska and Avendaño (2012) has been the use of restraint, which raises significant problems in interpretating behavioural outcomes, these include the production of restraint induced analgesia, overall movement impairment, and the restriction of behavioural responses.
With regards to trigeminal neuropathic pain, transection or compression/constriction of either the inferior alveolar nerve (mandibular division, trigeminal ganglion (V3)), the infraorbital nerve (maxillary division, trigeminal ganglion (V2)), the mental nerve (V3), the trigeminal ganglion, or the nerve trunk as it enters the pons, have produced clinically relevant models (Iwata et al., 2011;Krzyzanowska and Avendaño, 2012;Sadighparvar et al., 2023).Due to its easier accessibility, one of the most commonly used models of trigeminal neuropathic pain is the chronic constriction injury of the infraorbital nerve (ION-CCI), allowing for surgical manipulations to be executed with minimal blunt dissection.It was first described by Voss, Strassman and Maciewicz (1994) using a surgical approach based on earlier work by Gregg (1973), and Jacquin and Zeigler (1983).Vos and colleagues showed in a longitudinal study, the behaviour of male rats included changes in spontaneous activity (face grooming; exploratory behaviour) and stimulus-evoked behaviours.Vos et al proposed a 'response score' to describe the impact of the injury, which was based on increasing complexity of behavioural responses to stimulation of the face.We have expanded on these results, and in this study, we investigated the temporal development of evoked 'pain' behaviours in both male and female Sprague-Dawley rats in the ION-CCI model of neuropathic pain.Our initial analyses using the 'response scores' initially developed by Vos and colleagues (1994) did not adequately capture the full range of stimulusevoked behavioural changes that we observed.Thus, we report in more detail the individual behavioural responses to mechanical stimulation of the face, observed over the 28 days, post-ION-CCI period.
Additionally, the precise pathophysiological mechanisms for the development of chronic neuropathic pain states are still unclear, although the role of neuroimmune responses in modifying nociceptive signalling through macrophages and T-lymphocytes activation have been clearly highlighted (Ghazisaeidi et al., 2023;Midavaine et al., 2021).Sex differences in neuroimmune signalling also underly differences in the development of chronic pain states (Gregus et al., 2021;Midavaine et al., 2021;Rosen et al., 2017).Once again, the majority, of these findings are derived from studies in extra-cephalic models and any role in the orofacial region has been largely inferred.Therefore, following ION-CCI, we determined the temporal development of macrophage and T lymphocyte (T-cell) accumulation at the site of injury and the corresponding sensory ganglia (maxillary division of the trigeminal ganglia).We compared this to histological alterations (i.e., demyelination, axonal organisation) in the infraorbital nerve.

Materials and methods
All experimental procedures were carried out with the approval of the Animal Care and Ethics Committee at the University of Sydney (AEC no.2021 / 1878) and in accordance with the guidelines of the Code for the Care and Use of Animals in Research Australia, and Ethical Guidelines for Investigation Association for the Study of Pain (Zimmermann, 1983).

Experimental design
This study was designed to assess the development of pain behaviours and neuroinflammatory events in the infraorbital nerve (ION) and trigeminal ganglion of male and female Sprague-Dawley rats following an ION-CCI.Adult male (n = 94) and female (n = 108) Sprague-Dawley rats (160-180g on arrival, 6-7 weeks of age) were sourced from the Animal Resource Centre (Perth, WA).On arrival rats were group housed in same sex cages (n = 3 / cage) in large individually ventilated cages with ad libitum access to standard laboratory chow and water on a reverse light-dark cycle (12:12hr, lights OFF at 07:00).Housing lights brightened and dimmed gradually over 30 min to simulate dawn and dusk conditions, and the room was maintained at 22 ± 1 • C and 4-75 % humidity.All husbandry, monitoring and experimental procedures were performed during the dark phase and conducted under dim red-light conditions (Lee® Filters, Medium Red, allowing wavelengths > 600 nm to which Sprague-Dawley rats are least sensitive).Housing and experimental procedures for male and female rats were conducted independently of each other and rats were handled daily from arrival.
Rats underwent three baseline mechanical sensory tests using von-Frey filaments on − 5 days, − 3 days and − 1 day prior to ION-CCI (see Fig. 1A).On surgery day ION-CCI rats had two ligatures tied around their right infraorbital nerve, sham rats received the surgical procedure with no ligatures applied to the right infraorbital nerve and naïve rats were left in their home cages undisturbed.Rats received bilateral mechanical sensory testing to the area of the skin innervated by the ION on post ION-CCI days 1, 3, 5, 9, 13, 17, 21 and 25 and were euthanised according to their allocated endpoint (see Fig. 1C).All test procedures were performed in the dark phase of the rats under dim red-light conditions.

Surgical Procedures: Infraorbital nerve constriction injury (ION-CCI)
Rats were anaesthetised by induction with isoflurane (5 % in 100 % oxygen), delivered via an airtight induction chamber.Surgical level anaesthesia was maintained with 2-3 % isoflurane in 100 % oxygen administered via a custom-made facemask.Body temperature was maintained by a thermal blanket.ION-CCI was performed as described by Vos et al. (1994).The following procedures were performed under x3.5 magnification using Galilean Loupes (28050-35 Fine Scientific Tools, Vancouver Canada).The midline above the nasal bones and between the eyes was shaved and sterilised with povidone-iodine.Once the blink reflex and pedal reflex were abolished, an incision was made approximately 12 mm long, and 2 mm medial to the supraorbital ridge of the right orbit, following the curvature of the frontal bone.The fascia and muscle were gently teased from the bone using blunt dissection until the contents of the orbit could be gently retracted laterally.Following retraction of the orbital contents, the ION was visualised approximately 8 mm deep within the orbit, lying within the infraorbital canal of the maxillary bone.Once visualised, 5 mm of the ION was gently freed from the surrounding connective tissue via blunt dissection.Two chromic gut ligatures (Chromic Gut 5-0 sutures, #687G, Ethicon, Inc.) were loosely tied around the ION using a square knot, tight enough to slightly depress the nerve, but not enough to occlude epineural circulation.The ligatures were spaced approximately 2 mm apart, and the ION was repositioned in the infraorbital canal of the maxillary bone.The incision was sutured closed with nylon sutures (Nylon Sutures 7-0, #1696G, Ethicon Inc.).The wound was cleaned and an antibiotic powder (Tricin®, Jurox) mixed with Vaseline applied to the suture site.The rat was moved to a heated recovery cage and monitored closely until ambulatory and eating and drinking.

Behavioural Testing: Mechanical von Frey testing
Rats were habituated to the test room and sensory testing cages (100 x 142 x 200 mm) for 20 mins for the two consecutive days prior to sensory testing.All habituation and sensory testing were performed in a dark room with red light illumination of the testing cages only (Fig. 1B).On test days, rats were allowed to habituate for 20 mins in the testing cages.This was followed by habituation to the presence of a 1 g von Frey filament (#12-1647, Baseline®, USA), which was introduced into the test cage and applied to the walls of the cage.The ION-innervated facial region was subsequently stimulated using the von-Frey filament when the rat was still, with either four paws placed on the ground, or two hind paws on the ground whilst rearing to sniff out of one of the two openings in the lid of the test apparatus (Fig. 1B).Each rat received three filament applications to both sides of the face in a counter-balanced fashion, which lasted approximately 30-40 mins.All behavioural tests were video recorded starting from the commencement of habituation for later analyses.Behavioural responses were scored based on the frequency and temporal order of behaviours performed.The behavioural responses evoked by mechanical facial stimulation are described in Table 1.

Euthanasia and perfusion
Rats were deeply anaesthetised with sodium pentobarbitone (Lethabarb, 130 mg/kg in 0.9 % (w/v) saline, i.p.) and perfused trans-cardially with 400 mL of ice cold heparinised 0.9 % (w/v) saline.Following the perfusion, the brains were removed and rapidly frozen in liquid nitrogen and stored at − 80 • C. The left and right infraorbital nerves and trigeminal ganglia were dissected and post-fixed in 4 % paraformaldehyde acetate-borate buffer, pH 9.6 for two hours at room temperature, then cryoprotected in 10 % (w/v) sucrose in 0.1 M phosphate buffered saline, pH 7.4 (PBS) and stored at 4 • C until processing.Fig. 1.Experimental design for infraorbital nerve chronic constriction injury of male and female rats and mechanical stimulation-evoked behavioural measures.(A) On arrival rats were group housed in same sex groups (n = 3 / cage) in large individually ventilated cages with ad libitum access to standard laboratory chow and water.The housing wason a reverse light-dark cycle (12:12hr, lights OFF at 07:00).Housing and experimental procedures for male (♂) and female (♀) rats were conducted independently of each other and rats were handled daily from arrival.Rats underwent three baseline mechanical sensory tests using von-Frey filaments on − 5 days, − 3 days and − 1 day prior to chronic constriction injury of the infraorbital nerve (ION-CCI).On surgery day ION-CCI rats had two ligatures tied around their right infraorbital nerve, Sham rats received the surgical procedure with no ligatures applied to the right infraorbital nerve and Naïve rats were left in their home cages undisturbed.On test days (B) rats were habituated to the test room and sensory testing cages (100 x 142 x 200 mm) for 20 mins before the commencement of mechanical stimulation testing with a 1 g von-Frey filament or vibrissae deflection on post ION-CCI days 1, 3, 5, 9, 13, 17, 21 and 25 and were arbitrarily allocated to surgical conditions and endpoints as seen in (C).All habituation and sensory testing were performed in a dark room with red light illumination of the testing cages only.

Tissue processing
Infraorbital nerves and trigeminal ganglia were cut using a cryostat (Leica, CM1950, Germany) into 14 µm or 10 µm thick transverse sections, respectively, as a 1 in 10 series onto 2 % (w/v) gelatine coated glass slides and stored at − 20 • C until processing using histological or immunofluorescence procedures.

Luxol fast blue (LFB)
Luxol Fast Blue (0.1 % [w/v] in 95 % ethyl alcohol) staining was performed on the ION which stains specifically for myelin and allows the characterisation of nerve damage.Slides were hydrated in the following order: 100 % ethanol for 2-mins, followed by an additional 1-min, 95 % ethanol for 1-min.Slides were placed in LFB for 16-hours at 56 • C. Excess LFB was rinsed off with 95 % ethanol, followed by distilled water (dH 2 O).Sections were then differentiated in 0.05 % (w/v) lithium carbonate for 30-seconds, followed by 2x1-min 70 % ethanol wash.Slides were rinsed with dH 2 O and counterstained in cresyl violet for 30-secs rinsed with dH 2 O, and differentiated in 95 % ethanol for 5-mins, finally slides were agitated in 100 % ethanol for 30-secs and cleared in histolene (Chem Supply, Australia), before being cover slipped with DPX mounting medium (06522-100ML, Sigma-Aldrich, Australia).

Luxol fast blue and myelin basic protein Immunoreactivity
Bright field images of LFB stained sections were captured on an Olympus BX51 microscope.The middle transverse section of each ION was chosen for imaging.For LFB ION-CCI nerves, a 400x magnification image was captured from regions proximal and distal to the ligations, and from the segment between the ligations (middle).For naïve and sham-injured nerves, one 400x magnification image was captured for each nerve.
A five-level Likert scale was used to assess different elements of infraorbital nerve disruption to myelination, organization, and cellular structure following CCI (Likert, 1932;Reips and Funke, 2008).Five equally 'distanced' item categories were selected to assess; (i) the degree of demyelination, (ii) change in axonal organization and (iii) changes in cellular composition.The symmetry of categories, that are equidistant, allows the Likert scale to behave like an interval-level measurement (Reips and Funke, 2008).The responses to the Likert scales were collected from 17 to 22 independent assessors using mentimeter (mentimeter.com).The responses were summed to allow parametric statistical analyses of variance (Norman, 2010).
To complement the LFB analysis, MBP-IR ION's were imaged on an Olympus BX51 microscope with a Gryphax Kapella camera (Jenoptik®, Jena, Germany).Similar, to the LFB imaged nerves a 200x magnification image was captured at the same anatomical regions for ION-CCI, sham and naïve nerves.Quantification of MBP-IR consisted of the % area overlap of MBP-IR and DAPI using FIJI (Image J) colour merge channel and colour threshold function.

Macrophages (CD68-IR)
Slides were imaged using an Olympus BX51 microscope with a Gryphax Kapella camera (Jenoptik®, Jena, Germany).For all infraorbital nerves, the sections chosen for imaging were the transverse sections of the nerve with the largest thickness, at the midpoint of the nerve.A total of 6 images were captured: a 100x and 200x image each

Table 1
Scoring criteria following mechanical stimulation to the face.

Behaviour Description
No Response Rat does not respond with any movements to the stimulus Head turn Orientation of the head in the direction of the stimulus, head turn and precise localisation of the stimulus with their snout, can involve sniffing the stimulus Headwithdrawal Rat turns head slowly away or pulls head briskly backwards when the stimulation is applied Swiping Single face swipe ipsilateral to the stimulated area Biting Biting directed towards the stimulus Reaching Extending one or two forepaws directed towards the stimulus Attack Offensive upright posture, rapid approach towards the stimulus resulting in biting or boxing with extended forepaws.After the attack rat returns to an offensive upright posture or offensive sideways posture Avoidance Rat avoids further contact with stimulus object, either passively, by moving its body away from the stimulating object to assume a crouching position against cage wall, sometimes with the head buried under the body or towards the opposite side of the cage Facial grooming Movements patterns in which paws contact facial areas.Licking the forepaws then moving it towards a facial area from the distal, middle, and proximal nerve segments.For nerves from naïve and sham rats, as the ligatures could not serve as segmental landmarks, images for the middle segment were captured first from the midpoint of the nerve segment, and then the distal and proximal images were captured 1.5 mm either side of this midpoint zone.For all ganglia, 2 sections per slide were imaged, selected from the middle transverse sections.A total of 6 images were captured per section: a 100x and 400x each from the ophthalmic division, trigeminal ganglion (V1), the maxillary division, trigeminal ganglion (V2), and the mandibular division, trigeminal ganglion (V3).The 10x image was captured as a landmark, and the 400x image was captured for analysis, chosen as having the most ganglion cells to fill the frame.600x images were captured for representative photomicrograph use in figurework.Quantification of CD68-IR was assessed by determining the % area of positive CD68-IR using FIJI (Image J) threshold function.This analysis was applied to 200x images of the infraorbital nerve and 400x images from the trigeminal ganglion as detailed above.

T-cells (TCRαβ)
Slides were imaged using a Nikon C2 confocal microscope with motorised stage.This allowed for the large image function to be used to "stitch" multiple 200x images into a 4x3 large image, encapsulating the ligatures at the injury site of the infraorbital nerve.For the infraorbital nerve, transverse sections with the largest thickness, at the midpoint of the nerve were chosen for confocal imaging.For each trigeminal ganglion, 3 sections from the middle of the ganglion were chosen for imaging.A total of 9 images were captured per section at 200x magnification: 3 images per division of the ganglion.
T-cell counts were conducted using FIJI (ImageJ).In infraorbital nerves from ION-CCI rats' analyses were conducted in three zones: (i) 1.5 mm proximal to the boundary formed by the proximal ligatures; (ii) between the boundaries of the proximal and distal ligatures; and (iii) 1.5 mm distal to the boundary formed by the distal ligatures.The analysis of nerves from naïve and sham rats was performed on all the tissue along a 7 mm span of the nerve at the same anatomical location as the CCI was made.T-cells were counted within each of these areas and displayed as a total t-cell/area (mm 2 ) count per nerve.Trigeminal ganglion analysis consisted of counting all T-cells within a representative 200x image from either the V1, V2 or V3.
The summed responses (dependent variables: myelination, myelination organisation, circular nuclei, elongated nuclei and frothy profiles) to the Likert scales were assessed using a two-way ANOVAs with Tukey's multiple comparisons post hoc test.Similar, to the LFB analysis a two-way ANOVA (surgery x sex) with Tukey's multiple comparisons post hoc.To assess the effect of surgical procedures or stimulation procedures to the facial region (vibrissae deflection or innervation of the skin) at different time points a Mixed-effect analyses with Tukey's multiple comparisons post hoc test was performed on the dependent variables described in Table 1; no response, head turn, head withdrawal, swipe, bite, reach, attack, avoidance, and facial grooming, % area (CD68-IR) or cell counts (T-cells).To further probe any effects of sex a three-way ANOVA [between (surgery x sex) and within (time)] was performed.Data are represented in the figures as bar graphs, with individual data points identified, error bars represent ± standard error of the mean (SEM).

Behavioural responses to mechanical (von Frey) stimulation
Analysis of the video recordings of the behavioural responses of uninjured rats to mechanical stimulation of the area of the facial skin innervated by the ION revealed a range of behavioural responses.In order of the most frequently observed, these responses were (i) head withdrawal; (ii) biting; (iii) reaching; (iv) swiping; (v) attack; (vi) head turns (vii) avoidance and (viii) facial grooming.Frequently, (~50 % of stimulations), the response to von Frey filament application resulted in a sequence of behaviours, which most often began with a head withdrawal followed by either swiping, biting, or reaching.Head withdrawal followed by attack behaviours or facial grooming was observed on only five occasions.To contrast, stimulation of the same facial region in ION-CCI rats during the early post-injury period (days 1-7) did not often evoke a behavioural response, whereas in later tests (days 9-25) the von Frey stimulation almost exclusively evoked single behavioural responses which were either head withdrawal, or attack.These behavioural sequences are illustrated in supplementary Figs. 1 & 2.

Effects of von Frey stimulation in naïve (uninjured) rats
Sensory testing was conducted on ten test days during the 33-day experimental period.In naïve male and female rats, application of the von Frey filament to the facial skin innervated by the ION always produced a behavioural response.In contrast, contact with the vibrissae by the von Frey handle, failed to evoke behavioural responses in 11-28 % of applications in males (F 1, 31 = 19.09,P < 0.001) and 5.5-28 % of applications in females (F 1, 28 = 54.82,P < 0.001) (Fig. 2A & Fig. 3A).Handle-contact with the vibrissae, evoked head turning on 53-85 % of tests in males and, 40-69 % of tests in females.Head turns were seldom seen (ie, 1 of 6 stimulus applications on one test day only) following mechanical stimulation of the facial skin (male: F 1, 31 = 764.1,P < 0.0001; females: F 1,28 = 345.3,P < 0.001).The next most common response was head withdrawal, which occurred in up to 14 % of tests in males and in up to 39 % of tests in females.Mechanical stimulation of the skin innervated by the ipsilateral (right) ION, evoked significantly more head withdrawals than vibrissae stimulation.The response frequency of head withdrawals in females was 52-74 % (F 1,28 = 66.61,P < 0.001), and 33 %-68 % in males (F 1, 31 = 108.3,P < 0.001), the frequency of head withdrawals decreased over the experimental period and biting of the filament increased in both females and males (female: 5-29 % and male: 2-29 %).Biting behaviour was not seen following vibrissae stimulation only (male: F 1, 31 = 40.84,P < 0.001; female: F 1,28 = 6.251,P < 0.05).Attack of the filament was never seen in females but was notable in males (~15 % of stimuli) (F 1, 31 = 14.37,P < 0.001) (Fig. 2G & Fig. 3G).Comparisons of the frequency of occurrence of each behaviour, between application of the von Frey filament to the facial skin innervated by the ION, or deflection of the vibrissae are shown in Figs. 2 and 3 (Suppl Table 1).
Further, a comparison of males and females showed a sex effect, specifically on the frequency of head turns, head withdrawals, and attack behaviours evoked (see Suppl Tables 3 & 4).
Further, a comparison of males and females showed an effect of sex, specifically on the frequency of head withdrawals, attack and facial grooming behaviours evoked (see Suppl Table 6).

Histological and immunofluorescence evaluation of the ION following CCI over 28-days
Histological examination of luxol fast blue stained IONs from sham male rats revealed that in a majority of cases, sham surgical procedures showed no significant effects on the myelination of the nerve when compared with nerves from naïve rats.We did note however in a few cases at day 14 (2/6) and day 28 (4/9) there were small patches of reduced luxol fast blue staining, indicative of myelin loss, likely due to the exposure of the nerve and wound suturing.In contrast, nerves from sham injured female rats did show a significant reduction in myelination when compared to naïve rats at 2 days, 7 days, and 14 days after surgical procedures (Fig. 6C, 6D, 6E).Although analysis showed a reduction in myelin in sham injured female rats, there were no changes in the myelin organization when compared to uninjured nerves (Fig. 6G-J).
To further assess the impact of ION-CCI and sham surgical procedures on myelin we assessed the changes in distribution of MBP, a key structural protein found in myelin.Previously, Lee and Zhang (2012) showed that following mental nerve ligation, MBP-IR particles were found in macrophages, suggesting the clearance of myelin debris.Our analysis determined the percentage co-localization of MBP-IR and nuclear stained DAPI, which suggests the presence of MBP within cellular profiles (Fig. 7).
The nerves of ION-CCI rats showed significant changes in gross structure and cellular composition, in both male and females distal, proximal, and between the two ligatures of the ION at each time point (suppl Fig. 3).Fig. 6C shows that 2 days post-CCI, the distal and proximal segments of male rats showed a moderate demyelination with the female rats showing a similar trend.This was associated with increased number of circular nuclei, as well as what appeared to be luxol fast blue stained myelin profiles (macrophage-like) (suppl Fig. 3A & 3I).In addition, there was a significant main effect of sex (F 1, 29 (sex) = 5.424, P < 0.05) and surgery (F 4, 29 (surgery) = 3.061, P < 0.05) on the distribution of MBP-IR/DAPI.The female sham and ION-CCI rats appeared to show increased levels of MBP-IR/DAPI, although post-hoc analysis did not reveal any statistical significance (Fig. 7A).
At 7 days post-ION-CCI, the distal and proximal nerve segments showed largely similar histological features, to those seen at 2 days although, while the nerve fibres in the distal segment were either straight or wavy, fibres in the proximal segment began to lose their laminar organisation (Fig. 6H).Reductions in luxol fast blue staining, indicative of myelin loss, were apparent in the distal nerve segment in male rats, in contrast to females which showed an even loss of myelin across all three nerve segments (distal, middle, and proximal) (Fig. 6D).Increased numbers of macrophage-like cells, containing luxol fast blue stained myelin were also observed (suppl Fig. 3J).The MBP-IR/DAPI staining supports this observation with the largest effect of surgery (F 4, 31 (surgery) = 5.6, P < 0.01) present in both male and females at this time point (Fig. 7B).
At 14 days and 28 days post-ION-CCI, in male rats distal and proximal nerve segments shared similar histological features to each other, however they differed from those at 2 days and 7 days.There was evidence of significant myelin loss (Fig. 6E & 6F) accompanied by increases in the number of circular nuclei, although the numbers of macrophagelike cells containing luxol fast blue stained myelin had decreased (Suppl Fig. 3C, 3D, 3 K & 3L).The co-localization of MBP-IR and DAPI were also the lowest at 14 days (F 4, 30 (surgery) = 3.0, P < 0.05) and 28 days (F 4, 31 (surgery) = 4.872, P < 0.01) after ION-CCI.At day 28 post ION-CCI only the proximal segment of male rats exhibited difference in MBP-IR/DAPI when compared to naïve controls (Fig. 7C & D).The numbers of wavy fibres had significantly diminished, and only straight or disordered nerve fibres were seen (Fig. 6I & 6 J).In females, the demyelination appeared to have stabilised at 7 days with no further loss of myelin apparent at 14 days, or 28 days (Fig. 6).It was noted that the degree of constriction in male and female IONs over the 28 days period were not different (Suppl Fig. 4).

Immune markers in the ION following CCI
We next sought to investigate the relationship of macrophages, with the observed sex and time specific histological effects in the infraorbital nerve following CCI.
In all rats, macrophage densities were calculated for the segment of  11).
We next compared these changes in CD68-IR with uninjured and sham injured ION over the 28-day period which showed a significant effect of ION-CCI and time (see Fig. 8 and Suppl Table 12).
In males, 2 days post ION-CCI, the CD68-IR of the proximal segments of the nerve were significantly greater than the distal segment (distal-1.108% vs proximal 3.67 %, P > 0.05, middle-3.136%), 7 days post ION-CCI, there was an overall decrease in the amount of CD68-IR in the middle and proximal segments (distal-1.3% vs middle-1.073% and proximal 1.997 %).CD68-IR increased once more 14 days post ION-CCI, and now included increase in CD68-IR in the distal segment, (distal-1.995% vs middle-2.063% vs proximal 2.905 %).By 28 days post ION-CCI, there were clear increases in CD68-IR in the distal and proximal segments, with the levels seen in middle segments remaining similar to those seen 14 days post ION-CCI (distal-2.912% vs middle-2.09% and proximal 4.93 %) (Fig. 8).Over the 28 days observation period the numbers of macrophages increased in the distal segment (1.108 % up to 2.912 %); in the middle segment the initial large increase in macrophage numbers seen on day 2 (3.136 %) stabilizes on ~ 2 % for the remaining period of observation; in the proximal segment a large increase 2 days post ION-CCI (3.67 %), wanes during days 7 and 14, but returns again on In females, at 2 days post ION-CCI, the CD68-IR of the middle, distal and proximal segments of the nerve were almost identical (distal-0.98% vs, middle-0.76% vs, proximal 0.54 %).By 7 days post ION-CCI, there was an overall increase in the amount of CD68-IR in both the proximal and distal segments (proximal 5.25 % vs, distal-5.62%; middle-2.71%).CD68-IR 14 days post ION-CCI remained similar to those 7 days post ION-CCI, (proximal 4.1 % vs, distal-4.56%: middle-1.66%).By 28 days post ION-CCI, the proximal/distal versus middle differences was further amplified (proximal 8.83 % vs, distal-7.28%: middle-1.31%) (Fig. 8).Over the 28 days observation period the numbers of macrophages increase in the distal segment (0.98 % up to 7.28 %); in the middle segment the low levels of macrophage accumulation remained remarkably stable (0.76 % to 2.71 %); in the proximal segment the accumulation of macrophages increases significantly from 2 days to 28 days with a range from 0.54 % to 8.83 % (Fig. 8).
Further, a comparison of males and females showed an effect of sex on CD68-IR on days 7, 14 and 28 (see suppl table 13).

TCRαβ immunoreactivity (T-cells)
Following the investigation of macrophage accumulation in the ION.We investigated the relationship of T-cells as part of the adaptive immune response on, time, and sex.As they have been shown to play a regulatory role in nociceptive events.Identical to macrophage quantification, T-cell numbers were counted for the segment of the nerve between ligations, and in the segments of the nerves both proximal and distal to the ligations (see Fig. 9E).
In both male and female naïve rats, TCRαβ -IR cells (T-cells) were never encountered in the ION.In contrast, in male naïve rats very small numbers of T-cells (<29) were detected in 1/6 rats at day 2, 2/6 rats at day 7, 4/6 rats at day 14 and 5/9 rats at day 28.In sham female rats, remarkably similar numbers of T-cells (<29) were also detected in 2/6 rats at day 2, 6/6 rats at day 7, 5/7 rats at day 14, and 1/6 rats at day 28 (see Suppl Table 14).
In male ION-CCI rats, over the 28 days observation period the numbers of T-cells increased around the site of injury.At 2 days post ION-CCI, T-cells were not detected in any of the nerve segments.However, 7 days post ION-CCI, T-cells begin to appear in both proximal and middle segments (proximal 5.78 ± 4.349; middle 5.11 ± 3.364) and by day 14, in some rats, the numbers of T-cells increased markedly (proximal 33.33 ± 19.3; middle 25.33 ± 15.47).28 days post ION-CCI, T-cells were detected in all three segments of the injury site and were significantly different to naïve and sham-injured rats (Total: F 2, 61 (surgery) = 6.921,P < 0.01) (proximal 34.00 ± 18.53; middle 45.78 ± 15.8; distal 13.78 ± 9.11) (Fig. 9 and Suppl Table 14).
In female ION-CCI rats, over the 28 days period the numbers of Tcells detected were much smaller, although increases around the site of injury were still identified.2 days and 7 days post ION-CCI, T-cells were not detected in any of the nerve segments.However, on day 14, small numbers of T-cells appear in the distal and proximal segments of the injury, (proximal 9.5 ± 5.45; middle 18.33 ± 9.95; distal 6.67 ± 3.5), with a significant increase in T-cells in the middle segment (P < 0.001, Fig. 7B).On day 28 post ION-CCI, T-cells are still detected in small numbers in all three segments of the injury site, (proximal 7.33 ± 4.26; middle 8.6 ± 3.6; distal 5.4 ± 2.58) (Fig. 9 and Suppl Table 14).
Comparison of T-cell numbers in male and female showed a sex specific effect at 28-days after ION-CCI (Suppl Table 15).

TCRαβ immunoreactivity (TCR-IR cells)
In both male and female, naïve rats, very small numbers of TCR-IR cells, (T-cells) were found in all divisions of the ganglion during the testing period (see Suppl Table 14).In sham male rats, a slight increase in the number of T-cells were observed in V2 of the ganglion 7-days postsham surgery which were not seen in females (see Fig. 11).
In ION-CCI male and female rats, there was a significant increase in the number of T-cells at 14 days post injury (males: F 2, 38 (surgery) = 5.613, P < 0.01; females: F 2, 42 (surgery) = 6.143,P < 0.01) in the V1, with the V2 division showing similar trends, but did not reach statistical significance (see Fig. 11C).These increases were specific to this timepoint, as T-cell numbers had returned to baseline in both V1 and V2 by 28 days post-injury although there was still a main effect of surgery in male rats (F 2, 53 (surgery) = 3.293, P < 0.05) (Fig. 11).

Discussion
We investigated the gross histological and neuroimmune events in the infraorbital nerve and trigeminal ganglion following chronic constriction injury, in relation to the development of a range of stimulusevoked pain behaviours in individual male and female rats.In both male and female rats, at the site of injury, the temporal pattern of innate (macrophages) and adaptive (T-cells) immune responses were similar, however their magnitudes showed complementary differences; macrophage accumulation was greatest in females, whereas T-cell numbers were larger in males.To contrast, within the trigeminal ganglia of injured rats, macrophage accumulation in the maxillary division of the ganglion was observed only in males, not in females.T-cell numbers in the maxillary division of the ganglion in injured rats were similar in both males and females.Our observations revealed that ION-CCI triggers sexspecific differences in neuroimmune signalling at the site of injury and in the trigeminal ganglion, however the injury results in identical repertoires of stimulus-evoked pain behaviours.

Detecting orofacial pain: challenges of using behaviour
There are several major challenges in measuring orofacial pain using mechanically evoked behavioural responses, these include the presence of the special sensory organs that detect the approach of a stimulus (auditory, visual), the chemosensory, olfactory system, and the trigeminal somatosensory system each of which can drive a range of behavioural responses.Since the first description of the ION-CCI model which explored a large range of stimulus evoked and non-evoked behavioural responses, there has not been significant consideration of the confounding effects of other cranial sensory systems in identifying painspecific behaviour/s.This challenge in determining the behavioural 'response' or the selection of a single behavioural 'response' in preclinical facial pain models persists [see reviews (Krzyzanowska and Avendaño, 2012;Liu et al., 2023;Sadighparvar et al., 2023)].
The challenges of modelling facial pain do not exist to the same extent in other models of spinally mediated pain.For example, peripheral nerve injury related sensory sensitivity is readily characterised in both hindlimbs and forelimbs by altered withdrawal thresholds, latencies and/or frequencies of response.In these extra-cephalic models uninjured and 'pre-injury' animals generally produce stable behavioural responses to mechanical stimuli, that are seemingly unaffected by auditory, visual, or olfactory cueing of the approach of the stimulus.See Sadler et al. (2022) for a recent review of preclinical sensory assays.
In naïve rats we found that despite multiple periods of habituation to the test cage, testing room, and experimenter, the initial 3-4 presentations of the von Frey handle evoked head turns (orient towards), or head withdrawals (orient away) although these habituated rapidly.These responses were most likely driven by auditory, visual, or chemosensory cues.When the von Frey handle contacted the vibrissae, ~70 % of the applications produced orienting head movements towards the handle.This reaction is also likely driven by somatosensory inputs, in addition to auditory, visual, chemosensory cues.To contrast, mechanical stimulation of the face with the von Frey filament evoked a range of behavioural responses that included, head turns (orient towards), head withdrawals (orient away), swipes, biting, reaching, defensive behaviours (attack or withdrawal), and facial grooming.Thus, mechanical stimulation of the skin innervated by the infraorbital nerve produces different behaviours when compared to deflection of the overlying vibrissae.
The behaviours we describe above, were first reported by Vos et al. (1994) following mechanical stimulation using von-Frey filaments in freely moving rats.Vos and colleagues created semi-quantitative rank scores categorised as: (i) no response; (ii) stimulus detection; (iii) aversive and (iv) sustained aversive.This system was also adapted in mouse models, see Krzyanowska et al, 2011, for example.While this rank scoring methodology has been useful in many respects, these categories have been interpreted in many studies as continuous variables, assumed to be symmetrical and equidistant, rather than categorical descriptions, which has led to unsupported assumptions of systematic differences in pain intensity.Because of this concern and our observation that the 'response scores' initially described by Vos et al., 1994 did not capture important subtleties of stimulus-evoked behavioural changes observed in ION-CCI male and female rats, we reported individual behavioural responses observed following mechanical stimulation.6.2.Stimulus-evoked behaviours following ION-CCI in male and female rats ION-CCI produced remarkably similar changes in stimulus-evoked behavioural responses in male and female rats.In both sexes, the behaviours followed the same patterns of change across the 25-day testing period.Immediately following ION-CCI, both male and female rats showed a loss of stimulus-evoked responses in the facial region innervated by the ligated infraorbital nerve for up to one week.This hyporesponsivity was first noted by Vos et al. (1994) in male Sprague-Dawley rats and was described as the 'early period'.This phase is also characterised by increases in non-evoked facial grooming.Similarly, 4 and 8 days after ION-CCI, reduced sensitivity to mechanical stimulation (normalised to baseline scores) was also observed in both male and female Sprague-Dawley rats (Korczeniewska et al., 2018).Follow up studies by Deseure and colleagues, also confirmed an increase in nonevoked facial grooming from post-ION-CCI days 4-6, as well as a reduction in burrowing behaviour, a measure of general well-being (Deseure and Adriaensen, 2004;Deseure and Hans, 2018).This initial loss of response to mechanical stimulation, hypoalgesia/anaesthesia, in injured rats suggests major sensory modulation akin to that which characterises passive emotional coping responses, that are usually recruited following traumatic injury, and inescapable stress (Keay andBandler, 2001, 2002).Vos et al., 1994.also describes a 'later period', 15-130 days post-ION-CCI.The sensory loss subsides replaced by stimulus evoked hyper-responsiveness, with an accompanying reduction in facial grooming.The emergence of distinct head withdrawal responses around two weeks following injury is consistently reported in freely moving ION-CCI rats, and importantly female rats present with increased head withdrawal frequency, or decreased withdrawal thresholds earlier than males (Korczeniewska et al., 2018;Korczeniewska et al., 2023;Vos et al., 1994).We also observed that head withdrawals are a major behavioural response to stimulation at the same time points, as well as confirming the sex differences reported by Korczeniewska and colleagues.Our data also revealed a sustained loss of biting responses directed at the filament immediately following ION-CCI in both male and female rats which was sustained throughout the testing period.Attack behaviours began to occur from post ION-CCI day 17.These behaviours all occurred during the timeframe described by Vos et al., 1994, as the 'later period' and add to our understanding of the consequences of ION-CCI between 15 and 130 days post-injury.
Our data also suggest that there is an 'intermediate' or transitionary period (12-21 days post-ION-CCI) in both male and female rats where stimulus-evoked behaviours appear to reflect an increasing aversiveness of the mechanical stimulation, characterised by the emergence of defensive behaviours and reduction head withdrawal thresholds (Korczeniewska et al., 2018).We now consider the neuroimmune responses observed in the injured nerve and its ganglion using this framework of behavioural changes.

Neuroimmune events in the "Early (first 48 h)" phase following ION-CCI
Infraorbital Nerve-During the first 48 h following injury in the peripheral nervous system, nerve fibres undergo Wallerian degeneration.Macrophages form part of the first responders of the innate immune system following injury and are recruited by Schwann cells which secrete cytokine/chemokines following axonal degeneration (Gaudet et al., 2011;Rotshenker, 2011).Resident macrophages located within the endoneurium, which account for 2-9 % of nucleated cells within uninjured peripheral nerves, proliferate and perform phagocytic activities approximately 24-48 h following injury, before blood-derived monocytes infiltrate the injury site (Gaudet et al., 2011;Mueller et al., 2001).The majority of studies looking at the relationship between macrophages and the development of neuropathic pain have focused on nerve injury models of the sciatic nerve (Austin and Moalem-Taylor, 2010), with the general consensus that macrophages play an important role at the site of injury and at the sensory ganglia in coordinating tissue repair, pro-nociceptive signalling and anti-nociceptive signalling which shifts over time (Fiore et al., 2023).We found that the initial demyelination, morphological and axonal disruption at the site of injury in the infraorbital nerve were consistent with observations that demyelination peaks in the distal site at the end of the early phase, around 6-9 days after CCI of the infraorbital nerve in male Wistar rats (Costa et al., 2016).The demyelination, disruption of axonal organisation and repair mechanisms serve to facilitate the generation of ectopic and spontaneous neuronal firing which manifests in signs of mechanical allodynia often observed in trigeminal neuralgia (Devor, 2006;Love and Coakham, 2001;Marinković et al., 2009).These 'activated' macrophages are dynamically engaged in the process of degeneration and the subsequent regeneration of injured peripheral nerves.The temporal relationship of macrophage proliferation, infiltration and dominant phenotype is thought to determine the clearance of myelin debris, repair, and development of neuropathic pain (Austin and Moalem-Taylor, 2010;Costa et al., 2016;Malcangio, 2019).
Trigeminal Ganglion-In mice, ganglionic macrophages, which proliferate rapidly after infraorbital nerve injury are tissue-resident, enter between satellite glial cells and neurons forming ring-like structures in direct contact with neurons.These macrophages exhibit an M2-like phenotype suggesting that they are related to tissue repair (Iwai et al., 2021).This has also been seen in female Wistar rats in the DRG 7 days following sciatic nerve transection (Hu and McLachlan, 2002).Our data, show a similar proliferation of ring-like macrophages surrounding the maxillary ganglion cells early after injury (2-7 days) in male Sprague-Dawley rats.

Neuroimmune events in the "Transitional" phase following ION-CCI
Infraorbital Nerve-Both males and females had increasing numbers of CD68-IR macrophages at the site of injury during this phase, although overall females had a significantly greater magnitude of macrophages than males.Interestingly in female sham rats, the presence of macrophages detected during the early phase was also sustained during the transitional phase.This observation suggests a heightened innate immune response to injury in females.Although we did not distinguish between infiltrating bone-marrow derived macrophages and resident tissue macrophages, evidence suggests that following peripheral nerve injury the majority of macrophages present at the site of injury and associated sensory ganglion are proliferating tissue resident macrophages (Guimarães et al., 2023;Ydens et al., 2020).In sciatic neuropathic pain models, macrophages are observed in abundance at 7 days (Austin et al., 2015), and 14 days post-injury (Grace et al., 2016).Macrophages, by producing pro-inflammatory mediators can amplify nociceptive signalling (TNF⍺, IL1β, IL6 nitric oxide, prostaglandins etc) (Ma and Eisenach, 2003;Sommer and Kress, 2004) or by producing antiinflammatory mediators (IL4, TGF β, opioid peptides) can be pain resolving (Celik et al., 2020;Fiore et al., 2023;Kiguchi et al., 2015;Labuz et al., 2021;Labuz et al., 2010;Tonello et al., 2020).Thus, macrophages play a critical role in establishing a cytokine/chemokine network which plays a key role in orchestrating the development and maintenance of neuropathic pain.
In the mental nerve of male Sprague-Dawley rats it was shown that following CCI MAC1 + monocyte-like macrophages expressed proinflammatory cytokines IL6 and CCL3/MIP1α and anti-inflammatory cytokines IL10 and TGFβ1.These macrophages were present around ligations beginning at 14 days post-injury (Lee and Zhang, 2012).This study also showed that ED1+ (CD68-IR) macrophages exhibited phagocytic activities determined by the co-localisation of myelin basic protein, but they did not co-express any cytokines/chemokines.These ED1+ (CD68-IR) macrophages also expressed calcitonin gene-related peptide (CGRP) at the 14 days post-injury timepoint (Lee and Zhang, 2012).Similarly, in the infraorbital nerve, 15 days after CCI, increases in substance P and IL1β have been described (Costa et al., 2016).The appearance of CGRP and substance P in nerves of the trigeminal system 14-15 days after an injury, coincides with the development of specific classes of stimulation evoked-behaviours and may reflect critical peptidergic nociceptive signals that evoke pain-behaviours.In addition to the production of CGRP, activated macrophages can also increase the concentrations of neurotrophic factors either directly or indirectly via non-neuronal cells.These factors enhance nociceptive signalling and thus the expression of pain related behaviours (Heumann et al., 1987a;Heumann et al., 1987b;Lindholm et al., 1987).These pro-nociceptive actions may be challenged by increased substance P production which has been shown to promote anti-nociceptive, M2-like polarisation of macrophages which have been shown to be pain resolving (Zhou et al., 2023).
T-cells reached peak quantity during the "transitional" phase of ION-CCI in a similar manner to injuries of the sciatic nerve (Moalem et al., 2004).T-cells shape this phase of the immune response by producing pro-or anti-inflammatory cytokines (Mosman et al, 1986).Type 1, helper T cells secrete pro-inflammatory cytokines (ie, TNF-α, IFN-γ) to activate nearby macrophages, neutrophils, and natural killer cells.Type 2 helper T cells release anti-inflammatory cytokines (ie, IL4, IL10) which serve to suppress the pro-inflammatory cascade (London et al, 1998;Xin J et al, 2011).Our data, show for the first time, in the ION that following CCI, T-cells infiltrate with a similar time frame to that seen in the sciatic nerve, with cell numbers peaking at both 14 and 28 days in males and females.The males exhibited a significantly higher quantity of T-cells in the ION when compared to the females.T-cells can mediate pain relief via the release of β-endorphins.In sciatic nerve CCI, β-endorphin containing T-cells accumulated at the site of injury and can reduce mechanical pain sensitivity through a corticotropin-releasing factordependent mechanism (Labuz et al., 2010).The interplay between proinflammatory and pain-resolving T-cells may underpin the development of neuropathic pain.The clinical importance of this suggestion is strongly emphasised by observations in people with chronic pain resulting from peripheral neuropathy, post-herpetic neuralgia, or orofacial pain who have an increased quantity of anti-inflammatory regulatory T-cells in their blood (Luchting et al., 2015).
Trigeminal Ganglion-Macrophages decreased in density during the transitional phase in the maxillary division of male rats.However, in both male and female sham rat's macrophage densities were maintained in the maxillary division, in fact in this group and at this time point, the maxillary division was the primary location at which macrophages were seen.This initially puzzling observation led us to suggest that macrophage behaviour is subject to tighter temporal and functional regulation in ligated nerves compared with nerves that are surgically isolated but not ligated.Our inclusion of uninjured rats, naïve to surgical procedures also highlights the fact that sham injury is far from an uninjured and potentially non-painful condition.Our data suggest that the neuroimmune response in peripheral trigeminal pathways is sex specific.Indeed, mRNA expression of proinflammatory cytokines IL1α and TNFα has been shown to be specifically increased in the trigeminal ganglion in female Sprague-Dawley rats, but not in males 21 days after an ION-CCI (Korczeniewska et al., 2018).Further, although cytokine/chemokines Csf1 and CXCR1 were upregulated in a similar manner in both male and female trigeminal ganglia 21 days after ION-CCI, there was a larger fold increase found in females (Korczeniewska et al., 2018).
In mice, sex differences in peripheral innate immune activation have been described with male and female bone marrow derived macrophages exhibiting different cytokine/chemokine profiles, macrophage proliferation, and polarisation following stimulation by TNFα (Friedman et al., 2023).Moreover, when TNFα activated macrophages were injected into the hindpaw a more prolonged hypersensitivity to mechanical stimulation was identified in female mice, when compared to males (Friedman et al., 2023).
Following sciatic nerve transection or spinal nerve transection in female Wistar rats, T-cells are found in the DRGs peaking at 7 days and maintain their presence during the 'transitional phase' these studies also reported that a range of different macrophage populations also peak in number in the DRG at 7 days once again being sustained during the 'transitional phase' (Hu andMcLachlan, 2002, 2003).Following sciatic nerve CCI in male Sprague-Dawley rats T-cells and macrophages in the DRG also peak at 7 days and then decrease at 70 days post-injury (Hu et al., 2007).Our data, suggest distinct temporal differences in the neuroimmune responses in the sensory ganglia and to a lesser extent the peripheral nerve, between cephalic and extra-cephalic nerve injury models.Specifically, T-cells showed a peak in accumulation in the trigeminal ganglion 14 days following ION-CCI, whereas data from dorsal root ganglia following sciatic nerve CCI show an earlier peak at around 7 days.
During the 'transitional period' we remain perplexed by our observation that the largest T-cell presence was found in the ophthalmic rather than the maxillary division of the trigeminal ganglion in all ION-CCI rats.In attempting to understand this observation, we have reflected on data previously shown by Hu and McLachlan, whose figures showed that T-cells were densely located along the border of the meningeal layer of the ganglion, and specifically in the subarachnoid angle, in injured (and uninjured) rats.The suggestion from their observations is that Tcells occupy anatomically specific niches in the ganglion, in part dictated by available space (Hu and McLachlan, 2002).If this is true, then it is possible that the T-cells we observe in the area of the ophthalmic division of the ganglion are related more to the invasion of T-cells into the trigeminal ganglion from the meningeal border which is greatest adjacent to the ophthalmic sub-region, in response to signals from macrophages in the maxillary division cells which are located deeper in the ganglion with a much shorter meningeal border.6.5.Neuroimmune events in the "Late" (post 28 days) phase following ION-CCI Infraorbital nerve -In the 'late phase', the immune response to injury appears to take on a pro-repair macrophage phenotype.In both males and females, we observed the greatest macrophage densities at 28 days post injury.Our data also show an overall increase in the amount of myelin in the nerves of female rats at 28 days post injury, although this effect was not seen in male rats, suggesting a possible repair to nerve damage in females rather than males during this phase.A similar sex difference was noted in T-cell numbers at this time, with females showing declining numbers of cells whereas males show the highest numbers of cells.Whether the numbers of T-cells in males begins to decline after this timepoint awaits empirical confirmation.Following mental nerve constriction injury in male Sprague-Dawley rats it was shown at 28 days post injury that MAC1 + monocyte-like macrophages expressed the pro-inflammatory cytokines IL6 and CCL3/MIP1α and the anti-inflammatory cytokines IL10 and TGFβ1 (Lee and Zhang, 2012).It appears likely then, that a balance of pro-and anti-inflammatory mediators at the injury site can support both neural repair and the maintenance of pain, through a blend of individual cytokines/chemokines each supporting either re-myelination or the modulation of activity in related sensory ganglia.
Trigeminal Ganglion-Macrophage densities generally decrease in the trigeminal ganglion in the late phase, and similar reductions are reported 70 days following sciatic nerve CCI, in the DRG of male rats (Hu et al., 2007), and 77 days after sciatic nerve transection in the DRG of female rats (Hu and McLachlan, 2003).Prior to this reduction, at 30 days post sciatic CCI, macrophages with an anti-inflammatory M2-like appearance predominated in the DRG of the injured nerve, in male rats compared to pro-inflammatory M1-like cells (Perera et al., 2015).Once again, the balance between the numbers and levels of activity of discrete classes of pro-inflammatory and anti-inflammatory cells can lead to both persistent pain alongside attempts to restore both the structure and function of the ganglion.In the late phase, T-cells were no longer present

Fig. 2 .
Fig. 2. Behavioural responses to mechanical stimulation of the facial skin or vibrissae deflection in naïve male rats.The percentage response frequency from experimental stimulation or vibrissae deflection are categorised as: (A) No Responses; (B) Head turn; (C) Head withdrawal; (D) Swipe; (E) Bite; (F) Reach; (G) Attack; (H) Avoidance or (I) Facial grooming.The changes in behavioural responses are depicted in rats that receive no stimulation, but vibrissae deflection (blue) and rats that receive a mechanical stimulation (1 g, green) over a baseline period of − 5, − 3 and − 1 days which corresponds to the equivalent baseline pre-injury days of rats that receives an ION-CCI, and a 25-day period corresponding to a post ION-CCI period.Significant differences are represented as *P < 0.05, **P < 0.01 and ***P < 0.001 when vibrissae deflection is compared to mechanical stimulation.Mixed-effects analysis, with Tukeys multiple comparisons Post hoc test.Error bars represent ± SEM. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3 .
Fig. 3. Behavioural responses to mechanical stimulation of the facial skin or vibrissae deflection in naïve female rats.The percentage response frequency from experimental stimulation or vibrissae deflection are categorised as: (A) No Responses; (B) Head turn; (C) Head withdrawal; (D) Swipe; (E) Bite; (F) Reach; (G) Attack; (H) Avoidance or (I) Facial grooming.The changes behavioural responses are depicted in rats that receive no stimulation, but vibrissae deflection (blue) and rats that receive a mechanical stimulation (1 g, green) over a baseline period of − 5, − 3 and − 1 days which corresponds to the equivalent baseline pre-injury days of rats that receives an ION-CCI, and a 25-day period corresponding to a post ION-CCI period.Significant differences are represented as *P < 0.05, **P < 0.01 and ***P < 0.001 when vibrissae deflection is compared to mechanical stimulation.Mixed-effects analysis, with Tukeys multiple comparisons Post hoc test.Error bars represent ± SEM. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4 .
Fig. 4. Behavioural responses to mechanical stimulation of the facial skin in naïve, sham and ION-CCI male rats.The percentage response frequency from experimental stimulation or whisker deflection are categorised as: (A) No Responses; (B) Head turn; (C) Head withdrawal; (D) Swipe; (E) Bite; (F) Reach; (G) Attack; (H) Avoidance or (I) Facial grooming.The changes in behavioural responses of naïve (green, transparent), sham-injured (orange) and ION-CCI (purple) are depicted over a baseline period of − 5, − 3 and − 1 days which corresponds to the equivalent baseline pre-injury days of rats that receives an ION-CCI, and a 25-day period corresponding to a post ION-CCI period.Significant differences are represented as *P < 0.05, **P < 0.01 and ***P < 0.001 when ION-CCI rats are compared to sham rats.Mixed-effects analysis, with Tukeys multiple comparisons Post hoc test.Error bars represent ± SEM. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5 .
Fig. 5. Behavioural responses to mechanical stimulation of the facial skin in naïve, sham and ION-CCI female rats.The percentage response frequency from experimental stimulation or whisker deflection are categorised as: (A) No Responses; (B) Head turn; (C) Head withdrawal; (D) Swipe; (E) Bite; (F) Reach; (G) Attack; (H) Avoidance or (I) Facial grooming.The changes in behavioural responses of naïve (green, transparent), sham-(orange) and ION-CCI (purple) are depicted over a baseline period of − 5, − 3 and − 1 days which corresponds to the equivalent baseline pre-injury days of rats that receives an ION-CCI, and a 25-day period corresponding to a post ION-CCI period.Significant differences are represented as *P < 0.05, **P < 0.01 and ***P < 0.001 when ION-CCI rats are compared to sham rats.Mixed-effects analysis, with Tukeys multiple comparisons Post hoc test.Error bars represent ± SEM. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 6 .
Fig. 6.Luxol Fast Blue in male and female infraorbital nerves.A 5-level Likert scale (1 to 5) was used to perform a semi-quantitative assessment of myelin with 1-no blue present in the image and 5-very intense blue to determine the degree of demyelination or myelin organisation with 1-disorganised myelin and 5-very wavy organised myelin.(A) shows a representation of the sum of levels which determined the range of blue intensity as a proxy for myelin concentration in the infraorbital nerve.The number above the micrographs represents the summed data over multiple assessors.(B) a diagrammatic representation of the infraorbital nerve chronic constriction injury (ION-CCI) within the peripheral trigeminal system, example luxol fast blue ION and trigeminal ganglion stained with haemotoxylin and eosin.The degree of blue and the degree of disruption of myelin organisation and integrity was determined in naïve (N), sham (S) and within the distal (D), middle (M) and proximal (P) infraorbital nerve of ION-CCI male (♂) and female (♀) rats on (C) 2 days, (D) 7 days, (E) 14 days and (F) 28 days after a CCI.*P < 0.05, **P < 0.01, and ***P < 0.001, when compared to naïve rats.Two-way ANOVA analysis, with Tukeys multiple comparisons post hoc test.Error bars represent ± SEM. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 7 .
Fig. 7. Myelin Basic Protein in male and female infraorbital nerves.The percentage area of MBP-IR and DAPI co-localization in naïve (N), sham (S) and within the distal (D), middle (M) and proximal (P) infraorbital nerve of ION-CCI male (♂) and female (♀) rats on (A) 2 days, (B) 7 days, (C) 14 days and (D) 28 days after a CCI.Panel (E) illustrates a high magnification photomicrograph of MBP-IR 7 days post ION-CCI and an example of MBP-IR/DAPI co-localization in cyan, where blue depicts DAPI and green depicts MBP-IR.Scale bar represents 100 µm *P < 0.05 and **P < 0.01, when compared to naïve rats.Two-way ANOVA analysis, with Tukeys multiple comparisons post hoc test.Error bars represent ± SEM. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 9 .
Fig. 9. TCR-immunoreactivity in male and female infraorbital nerves.The number of TCR-IR cells per mm 2 within the (A) distal, (B) middle, (C) proximal segments, and the (D) total cell count from the sum of all three segments of the infraorbital nerve of male and female rats were quantified on day 2, 7, 14 and 28 days after an ION-CCI.(E) depicts an illustration of the location images were taken within the different segments of the infraorbital nerve.(F) illustrates an example photomicrograph of TCR-IR cells within the infraorbital nerve, scale bar represents 100 µm.Right photomicrograph represents a 400x example of a TCR-IR cell, scale bar represents 50 µm.White solid arrows identify positively labelled TCR-IR cells.*P < 0.05, **P < 0.01 and ***P < 0.001, Mixed-effects analysis, with Tukeys multiple comparisons Post hoc test.Error bars represent ± SEM.

Fig. 11 .
Fig. 11.TCR-immunoreactivity in male and female trigeminal ganglia.The mean number of TCR-IR cells within the Ophthalmic trigeminal branch (V1), Maxillary trigeminal branch (V2), and Mandibular trigeminal branch (V3) at (A) 2-days post ION-CCI, (B) 7-days post ION-CCI, (C) 14-days post ION-CCI and (D) 28 days post ION-CCI of male (♂) and female (♀) rats (E) depicts an illustration of the location images were taken within the different trigeminal branches of the trigeminal ganglion.Panel (F) illustrates an example photomicrograph of CD68-IR in male and female rats 28 days post ION-CCI.White solid arrows identify positively labelled TCR-IR cells.*P < 0.05 and **P < 0.01, Mixed-effects analysis, with Tukeys multiple comparisons Post hoc test.Error bars represent ± SEM.