Over the past several decades, “algesiometry” (pain testing) in rodent subjects has undergone a series of changes, or waves of development. The recent status quo involves the introduction of some injury (either inflammatory or neuropathic) followed by testing for reflexive withdrawal from experimenter-delivered noxious stimuli (heat, cold, or mechanical). This is a poor model of the clinical reality, in which the most prevalent and important symptom of clinical pain in humans is suprathreshold, non-reflexive, spontaneous pain. MOGILab was among the first to point out this disconnection, and has devoted much of its activities to developing more clinically relevant assays (i.e., injury models) and measures of pain. One of these new measures, the Mouse (and Rat) Grimace Scale, has been widely adopted for both research use and veterinary care.
Assays
Pain assays are methods by which states resembling clinical pain are introduced in animal models. MOGILab has participated in the refinement of existing assays, for example by studying optimal ways to score pain behaviours within the formalin test, abdominal constriction test, colonic irritation test, and cyclophosphamide cystitis test. We have also developed two novel preclinical pain assays. One of these relies on the measurement of naturally occurring orofacial behaviours (e.g., eye closure, head grooming, oculotemporal strokes, eye blinks, eye winking) in the R192Q and S218L mouse models of familial hemiplegic migraine, type 1 (FHM-1). In the second, a credible mouse model of vulvodynia (also known as provoked vestibulodynia) was developed using an existing human etiological hypothesis based on the observation that women with vulvodynia have a higher-than-average history of recurrent candidiasis (yeast infections). By repeatedly infecting female mice with Candida albicans and treating them with fluconazole, we produced long-lasting vulvar allodynia in a large subset of mice. Those mice developing allodynia also displayed hyperinnervation of the vulva by peptidergic nociceptors and sympathetic fibers, as has been reported in women with vulvodynia.
New Measures
MOGILab has been vocal in the pain community regarding the limitations of commonly used reflexive measures (i.e., limb withdrawal) to experimenter-delivered mechanical and thermal stimuli, both in terms of the mismatch between the prevalence of these symptoms and humans and their use in animals and also because of the close and prolonged experimenter-subject interaction required. We have also provided data challenging the validity of various nocifensive behaviours (hypolocomotion, licking, lifting, flinching, shaking, dynamic weight bearing) as measures of chronic, neuropathic pain in mice.
What is sorely needed are new measures of spontaneous pain, preferably ones that work in “real time” and that do not rely on direct human-animal interaction. MOGILab has successfully developed two such measures. “Grimace scales” were developed (the Mouse Grimace Scale [MGS] and the Rat Grimace Scale [RGS]) by adapting the Neonatal Facial Coding Scale (Grunau & Craig, Pain, 28:395-410, 1987). Grimace scales feature high reliability and accuracy, are potentially a measure of pain affect, track pain levels for up to several weeks, and are currently being fully automated using machine learning algorithms. Grimace scales have now been developed in at least 12 different mammalian species, and have been used to re-evaluate proper dosing of postoperative analgesia. A fully trained machine-learning algorithm is now publicly available to score the MGS automatically. In a paper published in 2021 with Dr. Rob Bonin’s lab, we introduce cage-lid hanging as a new measure of spontaneous pain in mice. A detailed “ethogram” of home cage behaviours in the mouse before and after persistent inflammatory and neuropathic injury revealed cage-lid hanging as the only behaviour featuring robust and long-term changes. The ease of automation suggests that this measure might prove extremely useful to the preclinical pain research community.
Current Projects
Current algesiometric efforts in MOGILab include assisting Dr. Mark Zylka’s laboratory in the machine learning-enabled full automation of the MGS for both albino and dark-furred strains. Initial steps have been taken to investigate optogenetic control of facial grimacing in an attempt to clarify what brain regions and neural circuits are involved. Finally, we are endeavouring to compare pain behaviours in standard laboratory (i.e., high stress) conditions compared to those occurring in the Eco-HAB housing setup (see Social Behaviour & Pain).