Altered nociceptive behavior and emotional contagion of pain in mouse models of autism

Abstract Individuals with autism spectrum disorder (ASD) have altered sensory processing but may ineffectively communicate their experiences. Here, we used a battery of nociceptive behavioral tests to assess sensory alterations in two commonly used mouse models of ASD, BTBR T+ Itpr3 tf /J (BTBR), and fragile‐X mental retardation‐1 knockout (Fmr1‐KO) mice. We also asked whether emotional contagion, a primitive form of empathy, was altered in BTBR and Fmr1 KO mice when experiencing pain with a social partner. BTBR mice demonstrated mixed nociceptive responses with hyporesponsivity to mechanical/thermal stimuli and intraplantar injections of formalin and capsaicin while displaying hypersensitivity on the acetic acid test. Fmr1‐KO mice were hyposensitive to mechanical stimuli and intraplantar injections of capsaicin and formalin. BTBR and Fmr1‐KO mice developed significantly less mechanical allodynia following intraplantar injections of complete Freund's adjuvant, while BTBR mice developed slightly more thermal hyperalgesia. Finally, as measured by the formalin and acetic acid writhing tests, BTBR and Fmr1‐KO mice did not show emotional contagion of pain. In sum, our findings indicate that depending on the sensation, pain responses may be mixed, which reflects findings in ASD individuals.


| INTRODUCTION
Autism spectrum disorders (ASD) are a highly prevalent class of neurodevelopmental disorders characterized by social and communicative impairments and restricted behavior. 1 In addition to the core social deficits, a growing body of work also implicates abnormal sensory responses as a critical symptom of ASD in multiple forms and across modalities. [2][3][4] Tactile hypersensitivity appears to be especially common, appearing in up to 95% of cases, 5 and often presents itself as defensiveness or avoidance and interferes with social behavior that involves interpersonal touch. 6 While as many as 44% of ASD patients engage in self-injurious behaviors such as head banging, hair pulling, skin picking, and scratching, the experience of pain remains poorly understood in ASD individuals. 7 Evaluation of sensory phenotypes has shown that ASD individuals display altered nociceptive behavior but depending on the type of noxious stimulation, pain responses may increase or decrease. 4 For instance, adolescents with ASD exhibit decreased thermal sensitivity, 8 while high functioning ASD children reportedly have increased pressure and mechanical sensitivity compared to typically developed children. 9 In addition, the lack of communication skills in ASD may not allow these individuals to communicate the degree and severity of pain experienced effectively. At the same time, ASD individuals may also not fully understand pain signals from others within their social environment. 10 Animal models have proven invaluable for investigating the core features and genetic bases of ASD, 11 with some recent studies also investigating altered sensory function. 12 For instance, mutations in the major SHANK isoforms (SH and multiple ankyrin repeat domains), a family of scaffolding proteins, have been associated with ASD in humans. [13][14][15] Mice harboring mutations of the Shank2 gene exhibit social impairments, 16 while Shank3 null mice display decreased social communication as measured by ultrasonic vocalizations but normal levels of sociability. 17 Null deletion of Shank2 leads to reduced mechanical and thermal pain sensitivity, 18 while Shank1 and Shank3 null mice display normal thermal pain responses 19,20 ; however, global or sensory neuron-specific deletion of Shank3 impairs heat hyperalgesia. 20 Heterozygous loss of either Tsc1 or Tsc2 (tuberous sclerosis complex), two genes associated with ASD in humans, 21 leads to social deficits, repetitive rearing, and learning and memory impairment in mice, but intact sensory function. 22 There is also evidence that deletion of Mecp2 (methyl CpG binding protein 2), a model of human Rett syndrome commonly classified as an ASD, is associated with decreased heat responsiveness, 23 while also linked to increased mechanical sensitivity. 24 Fragile X syndrome, another common syndromic form of ASD, is associated with pervasive intellectual disability, repetitive behaviors, social deficits, and increased anxiety. 25 Null deletion of the Fmr1 gene, which encodes for the fragile X mental retardation 1 protein (FMRP), yields social deficits 26 and decreased responses to inflammatory and neuropathic pain, 27 but hypersensitivity to tactile stimuli. 28 In another study, Fmr1 knockout mice and BTBR T + Itpr3 tf /J (BTBR) mice, another model of ASD, showed hyporesponsiveness to thermal stimuli and hyperresponsiveness to intraperitoneal injections of acetic acid. 29 The BTBR mouse model is interesting because several ASD-relevant mRNAs are altered, including Neurexin-1 and Homer31, which bind 30 to Shank1 and Shank3 and has been linked to over activation of the mechanistic target of rapamycin (mTOR) pathway as observed in syndromic forms of ASD such as tuberous sclerosis complex and fragile X syndrome. 31 In the current paper, we assessed nociceptive behaviors of BTBR, an idiopathic model of ASD, and Fmr1 null-mutant mice (Fmr1-KO), a monogenic model of ASD using a battery of innocuous and noxious pain tests. We also studied the development of chronic inflammatory pain and whether hypersensitivity was altered in these models of ASD. Finally, given that BTBR, and Fmr1-KO mice have impaired social behavior, we assessed whether the emotional contagion of pain was impaired in these mice. Emotional contagion-a primitive form of empathy-enhances pain behavior when familiar mice are tested together 32,33 ; however, it remains unclear whether the emotional contagion of pain is altered in ASD mouse model.

| Mice
All experiments were performed on young (6-12 weeks) adult male, C57BL/6J, BTBR T + Itpr3 tf /J (Jax stock, 002282) or Fmr1-KO mice (C57BL6/J background, Jax stock, 003025) originally purchased from Jackson Laboratories. Mice were bred in-house for several generations at our animal facility at the University of Toronto Mississauga.
For some experiments, Fmr1-KO mice (C57BL6/J background) were generously provided by Dr. David Hampson (University of Toronto).
All mice were housed with the same sex in groups of four mice per cage, maintained in a temperature-controlled (20 ± 1 C) environment with 12:12-h light:dark cycle with access to food (Harlan Teklad 8604) and water ad libitum. Experiments were conducted only during the light period, and mice were habituated to the testing environment for at least 15 min in every assay before testing commenced. All procedures were approved by the University of Toronto animal care committee and in accordance with the Canadian Council on Animal Care.

| von Frey test
An automated von Frey test (Ugo Basile Dynamic Plantar Aesthesiometer) was used to assess mechanical nociceptive thresholds. Mice were placed in custom-constructed Plexiglas cubicles (6.3 Â 5.5 Â 10 cm) on a perforated metal floor and allowed to habituate for 1 h before testing. A blunt probe was raised toward the plantar surface of the hind paw, upon which pressure was gradually increased until the mouse withdrew its hind paw; the maximal pressure displayed at that point was then recorded. The average of three trials per mouse per paw was used as the measure of mechanical sensitivity.

| Tail clip
A small alligator clip (force, 700Âg) was applied at 1 cm from the base of the tail as we have previously performed. 34 The latency to attack/ bite the clip was measured to the nearest 0.1 s. Mice were only tested once on the tail clip test.

| Cold plantar
We followed the procedure as described in Brenner, Golden, Gereau 35 to measure cold sensation. Powdered dry ice was packed into a modified syringe, and the open end of the syringe was held against a glass surface while depressing the plunger to form a dry ice pellet that was extended past the end of the syringe and pressed to the glass underneath the hind paw using light, but consistent pressure applied to the syringe plunger. Care was taken to ensure that the hind paw was touching the glass surface. Latency to withdraw the hind paw from the stimulus was measured to the nearest 0.1 s. The average of three trials per mouse per paw was used as the measure of cold sensitivity.

| Radiant heat paw-withdrawal test
Mice were placed on a glass floor within small Plexiglas cubicles (9 Â 5 Â 5 cm high). Following habituation, a focused high-intensity projector lamp beam was shone from below onto the mid-plantar surface of the hind paw. 36 The radiant heat device (IITC Model 336) was set to 20% active intensity. Latency to withdraw the hind paw from the stimulus was measured to the nearest 0.1 s. The average of three trials per mouse per paw was used as the measure of heat sensitivity.

| Hotplate
Mice were placed into a clear Plexiglas cylinder atop a hotplate (Columbus Instruments) maintained at 50 C. The latency to lick or shake either hind paw was measured to the nearest 0.1 s. Mice were only tested once on the hotplate.

| Capsaicin
Mice were placed on a glass floor within Plexiglas cylinders (30 cm high; 30 cm diameter) and allowed to habituate for 15 min. Mice then received a subcutaneous injection of capsaicin (2.5 μg; Sigma) into the plantar left hind paw (20 μl) and were digitally videotaped for 10 min.
Video files were later scored for the total duration (s) of licking/biting of the injected paw.

| Formalin test
Formalin injection produces a biphasic response: an acute, nociceptive "early" phase and a tonic, inflammatory "late" phase, separated by a quiescent period in which there is reduced pain behavior. 37 Mice were placed on a glass platform within Plexiglas cylinders (30 cm high;

| Acetic acid
Mice were habituated for at least 30 min to an observation chamber (15 cm diameter; 22.5 cm high), placed atop a glass surface suspended over high-resolution video cameras. Mice were injected intraperitoneally (10 ml/kg) with 0.9% acetic acid and videotaped digitally for 30 min after the injection. The videotapes were later scored offline by a different experimenter who was blind to experimental details. Video files were coded for the number of lengthwise constrictions of the abdominal musculature ("writhes") using a sampling procedure (1 sample every 20 s) as we have previously performed. 32 For the social modulation of pain studies, mice were tested as described for the formalin experiments, the only difference being that acetic acid was used as the pain stimulus.

| Complete Freund's adjuvant
Complete Freund's adjuvant (CFA; 50%; Sigma) was injected subcutaneously in a volume of 20 μl into the left plantar hind paw using a 100-μl microsyringe with a 30-gauge needle. Mice were tested for radiant heat paw withdrawal or sensitivity to von Frey filaments of both hind paws as described above, before, 3-, 7-or 10 days post-CFA injection. The percentage of allodynia/hyperalgesia for Day 3 was calculated as a function of baseline (i.e., decrease from baseline threshold) and reported as percentage change.

| Three chamber test
We followed the experimental protocol described by Yang et al. 38 and as we have previously performed. 39 Briefly, we habituated test mice to the center of the three chambered apparatus for 10 min. Following initial habituation, mice were allowed to freely explore all chambers for an additional 10 min (baseline). A single naive mouse was then placed in an inverted wire pencil cup in one side chamber (in a counterbalanced fashion). These "stimulus" mice were previously habituated to the pencil cup to reduce excessive movement while in the cup. Test mice were videotaped for 10 min in the presence of the stimulus mice. The total time spent in each side chamber (one containing the stimulus mouse and the other a novel object) was coded by a blinded experimenter.

| Statistical analysis
Data were analyzed by a 2-tailed Student's t-test (unless otherwise indicated) or two-way ANOVA for CFA and pain contagion experiments, followed by Tukey's honest significant difference (HSD) post hoc tests. For CFA time-course data, we conducted post hoc testing between mouse strains at each time point. For the pain contagion experiments, post hoc testing was conducted within strain and comparisons between the alone, cagemate, and stranger conditions were made. A p-value of less than 0.05 was used to determine statistical significance. All data were analyzed using SPSS v 27.  Figure 3F).  Figure 4C; maximal hyperalgesia, t 22 = 1.892, p = 0.07, Figure 4D).

| Reduced sociability and emotional contagion of pain in BTBR and Fmr1-KO mice
We previously showed that pain behavior is increased in humans and rodents when conspecifics observe and experience pain with a familiar individual, a phenomenon known as emotional contagion. 32,33 As ASD mouse models have reduced sociability, 11 we next wanted to determine whether pain behavior was modulated in the presence of a social partner. To confirm that BTBR and Fmr1-KO mice showed a lack of social preference, we first used the three-chambered test of sociability. We found that, unlike C57BL6/J mice, BTBR and Fmr1 KO mice did not spend significantly more time within the chamber containing a novel mouse (two-way ANOVA, main effect of strain: strain Â chamber interaction: F 2,31 = 13.54, p < 0.001; Figure 5A).
Next, we tested whether the emotional contagion of pain was altered in BTBR and Fmr1 KO mice. When tested in the presence of a familiar cagemate, C57BL6/J, but not BTBR or Fmr1-KO mice, showed enhanced nociceptive sensitivity during the formalin test ( Figure 5B,C, D). The time-course of nocifensive behavior during the formalin assay is shown in Figure 5B. Analysis of the early phase of formalin (0-10 min) revealed enhanced licking behavior in C57BL6/J mice when tested in the presence of a cagemate and compared with mice tested with a stranger or alone. This effect was not present in BTBR or   Figure 5E).  29 and some indicating no change. 41 These findings are also reflected in the human literature, Mechanical behavioral phenotypes have previously been identified for monogenic ASD mouse models, including Fmr1, 28 Ube3a, 43

| DISCUSSION
Mecp2, 24,44 and Gabrb3. 44 Our Fmr1-KO data support these findings and align with increased tactile defensiveness in Fmr1-KO mice, which may interfere with social behavior and interpersonal touch. 28 Further, Mecp2 and Gabrb3 deletion in low-threshold mechanoreceptors during development, but not adulthood, causes social interaction deficits and anxiety-like behavior. 44 This paper also showed that restoring autism. 45 Children with ASD become tense when touched, find touch aversive, and prefer to be touched on their own terms. 4 Autistic children also report significantly lower pleasantness ratings in response to tactile stimuli than typically developing children. 46 Hypersensitivity to touch has also been documented in adults with ASD. For example, adults with Asperger's display a significantly lower detection threshold for vibrotactile stimuli and described mild sensations applied to their hand as more "tickly" and intense than control subjects. 47 In autobiographical accounts, patients with high-functioning autism have described touch as "an intense feeling" that can be "overwhelming and confusing" and serve as the impetus for social withdrawal. 48 Conversely, BTBR mice were hyposensitive to mechanical and thermal stimuli, which may be related to the reduced conduction velocity of afferent nerve fibers. 42 Recent evidence has indicated that chronic pain perception may be implicated in the pathogenesis of poor health outcomes in children with ASD. 49 The prevalence of pain is twofold higher in ASD children compared with typically developed controls. 50  Our study tested only male mice because idiopathic ASD is four times more prevalent in males than in females, [58][59][60] and fragile X is twice as common in males. 61 However, this is a limitation of the current work, and future studies should assess pain phenotypes in female ASD mouse models. Although, we have tested female BTBR and Fmr1-KO mice on some of the pain tests used in the current paper (i.e., von Frey, hotplate, and formalin) and have not noticed any obvious sex differences (unpublished data). In addition, the lack of social pain contagion in strangers is only evident in male mice, thus necessitating the sole use of males for the cagemate/stranger comparison. 62 Further, some papers have tested pain phenotypes in ASD models using both sexes 44,63 ; however, no obvious sex differences have emerged. The initial characterization of nociceptive sensitivity in Fmr1-KO mice tested both male and female mice, but no sex differences were reported. 27 Shank3 deletion in peripheral mechanosensory neurons leads to tactile hypersensitivity, and region-specific brain abnormalities with no obvious sex differences reported. 63 Sex-specific characterization of prototypical features of ASD has been more commonly investigated than sensory abnormalities in mouse models. For instance, disruption of Mecp2 in the amygdala of male but not female rats resulted in a significant decrease of juvenile social play behavior. 64 A careful battery of behavioral tests conducted on female Fmr1-KO mice showed increased repetitive behaviors on the nose-poke task and enhanced coordination on the accelerating rotarod compared to female WT mice. In contrast, male Fmr1-KOs lacked these behavioral differences. 65 Combined with the original characterization of pain processing in Fmr1-KO mice that included behavior, anatomy, and electrophysiological responses, 27 our results and other reports 28,29 indicate that these mice lack sensitization. Although, in our study we do not find altered visceral sensitivity in Fmr1-KO mice. An increase in visceral pain behavior has been previously reported in Fmr1-KO, 29 and children with ASD suffer from gastrointestinal problems such as gastroesophageal reflux disease (GERD) and frequent abdominal pain. 66 Interestingly, the frequency and severity of visceral pain in ASD children have been linked to social withdrawal, stereotype, and hyperactivity compared with children who have no history of frequent GI symptoms. 67 In a previous study, BTBR mice showed enhanced abdominal constriction behavior compared to C57BL/6J mice. 29 At the same time, the application of capsaicin and inflammatory mediators increased excitability in jejunum tissue prepared from BTBR mice. 52 Thus, enhanced visceral sensitivity as observed in the BTBR mouse may partly be due to the enhanced firing of visceral primary afferents following activation by chemical stimuli. Regardless, critical mechanisms of pain regulation in the BTBR model of ASD remain to be uncovered. There are seemingly contradictory findings where these mice were hyposensitive on some tests while hypersensitive on others. These findings recapitulate reports in ASD individuals who show mixed pain responses depending on the sensation.
Finally, the ability of pain to modulate social behavior has been observed in many gregarious species, ranging from rodents to humans, suggesting that a relationship exists between sociability, empathy, and pain perception. [68][69][70] As pain-related expressions may communicate one's pain so that care and help may be provided, 71  show basic empathetic behaviors such as emotional contagion for pain. 75 In children with ASD, contagious yawning and laughter are impaired but moderated by familiarity with higher levels of contagion observed between ASD children and their parents. 76 Further, individuals with ASD score lower on self-reported measures of empathy.
However, they show similar levels of brain activation during the perception of facial pain expressions compared with controls suggesting that reappraisal may lead to a failure of appropriate empathic responding. 75 In contrast, other studies indicate that the response of an ASD individual to others' pain is dependent on stimulus modality. 10 The pain behaviors-foot licking or abdominal stretching-used to assess the emotional contagion of pain in BTBR and Fmr1-KO mice are distinct and obvious. Thus, BTBR and Fmr1-KO mice may lack an emotional contagion response due to an inability to perceive the pain expressions from their social partner rather than a true disruption of emotional state sharing. 77