Quantification of pain in sickle mice using facial expressions and body measurements

Abstract

Pain is a hallmark feature of sickle cell disease (SCD). Subjects typically quantify pain by themselves, which can be biased by other factors leading to overtreatment or under-treatment. Reliable and accurate quantification of pain, in real time, might enable to provide appropriate levels of analgesic treatment. The mouse grimace scale (MGS), a standardized behavioral coding system with high accuracy and reliability has been used to quantify varied types of pain. We hypothesized that addition of the objective parameters of body length and back curvature will strengthen the reproducibility of MGS. We examined MGS scores and body length and back curvature of transgenic BERK sickle and control mice following cold treatment or following treatment with analgesic cannabinoid CP55,940. We observed that sickle mice demonstrated decreased length and increased back curvature in response to cold. These observations correlate with changes in facial expression for the MGS score. CP55,940 treatment of sickle mice showed an increase in body length and a decrease in back curvature concordant with MGS scores indicative of an analgesic effect. Thus, body parameters combined with facial expressions may provide a quantifiable unbiased method for objective measure of pain in SCD.

Keywords: Cannabinoid; Hyperalgesia; Mouse grimace scale; Pain; Sickle cell disease.

Figure 1

Changes in mouse facial expression in response to cold. Description and representative images for each AU at RT and on cold plate (4°C) are shown for the parameters measured for overall average MGS score. Abbreviations: AU, action units; RT, room temperature.

Figure 2

Facial features and body parameters proposed to quantitate pain. Diagram showing the facial action units (AUs) measured for the Mouse Grimace Scale (MGS) score and positioning of the ellipse used to determine curvature of the back.

Figure 3

Cold hyperalgesia leads to shortening of length and increased curvature of the back. (A) Cold exposure (red bars) reduced the length measurements in both sickle (SS) and control (AA) mice as compared to their measurements at RT (black bars). (B) Sickle mice showed significantly more reduced percent difference in length than control mice. (C) Cold exposure increased the curvature in both sickle and control mice compared to values at RT, curvature is inversely proportional to the eccentricity. (D) Sickle mice showed significantly greater percent difference in back curvature than control mice. Percent difference in lengths and curvature of sickle and control mice at 4°C are compared to their respective values at RT as 100%. Data given as mean ± SEM (n=10 per group). Mean age of each group of mice in months ± SEM for sickle mice: 5.4 ± 0.3 and for control mice: 5.5 ± 0.2. Two-way ANOVA followed by Bonferoni’s correction for multiple comparisons revealed a significant genotype x treatment interaction for the measured length but not eccentricity (A: F_{1, 18} = 23.61; P = 0.0001; C: F_{1, 36} = 2.265; p = 0.141). Unpaired t-test was used to determine significance between percent change values (B, D). * P < 0.01, ** P < 0.0001.

Figure 4

Cold exposure increases AUs and MGS score differently in male and female sickle mice. (A) Orbital tightening significantly increases in both male and female sickle mice at 4°C compared to RT. Both sickle groups also showed significant increases from their respective controls at 4°C. (B) Nose bulge increased significantly in response to cold only in female sickle mice and between male sickle and control mice. (C) Cheek bulge significantly increased in both male and female sickle mice and only in male control mice at 4°C compared to RT. There was a difference between the female sickle and female control mice at 4°C. (D) Ear position AUs only significantly changed in female sickle mice at 4°C compared to RT, and only between sickle and control females at 4°C. (E) Average MGS score significantly increased in both male and female sickle mice and only in male control mice at 4°C compared to RT. Both sickle groups also showed significant increases from their respective controls at 4°C. There was an overall significant increase of the MGS score for female sickle mice when compare to male sickle mice at 4°C. (F) The percent differences in MGS score significantly increased only in female sickle mice in comparison to control mice. These differences at 4°C are compared to their respective values at RT as 100%. Data are means ± SEM (n=10 per group). Mean age of each group of mice in months ± SEM for sickle male mice: 6.2 ± 0.2; sickle female mice: 4.6 ± 0.0; control male mice: 6.1 ± 0.1 and control female mice: 4.9 ± 0.2. Two-way ANOVA (A–E) or one-way ANOVA (F) followed by Bonferoni’s correction for multiple comparisons was used to determine significance. F test analyses of sickle males x sickle females, sickle males x control males, sickle females x control females, and control males x control females, showed significant treatment interaction between male and female sickle mice for ear position (D: F_{1, 23} = 6.750, P = 0.0161) and overall MGS score (E: F_{1, 25} = 10.07, P = 0.0040) and sickle and control females for cheek bulge (C: F_{1, 16} = 5.772, P = 0.0288), ear position (D: F_{1, 16} = 15.05, P = 0.0013), overall MGS score (E: F_{1, 15} = 20.05, P = 0.0004) and % difference in MGS score (F: F3, 37 = 4.277, P = 0.0109) * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001. Abbreviations: AU, action units; MGS, Mouse Grimace Scale

Figure 5

CP 55,940 treatment of sickle mice increases length and decreases curvature of the back. (A) Length measurements increased in treated sickle mice compared to vehicle group. (B) Percent difference in length of CP 55,940 treated sickle mice is significantly greater than that of vehicle. (C) Back curvature decreased significantly in sickle mice following treatment compared to vehicle group, after CP55,940 treatment. (D) Percent change in back curvature (eccentricity) following CP 55,940 treatment was higher in sickle mice than in vehicle treated mice. Percent difference in length and eccentricity of treated and vehicle groups are relative to their respective values before injection as 100%. Data are means ± SEM (n=8 per group). Mean age of each group of mice in months ± SEM for vehicle: 5.6 ± 0.2 and CP55,940: 5.4 ± 0.2. Unpaired t-test was used to determine significance. *P < 0.05, **P < 0.001, ***P < 0.0001

Figure 6

CP 55,940 treatment decreases AU and MGS score in sickle mice. Orbital tightening (A), cheek bulge (C), ear position (D), and overall MGS score (E) significantly decreased in CP55,940 treated sickle mice compared to vehicle treated sickle mice. Nose bulge (B) was not significantly different between the CP55,940 and vehicle treated groups. Data are means ± SEM (n=8 per group). Mean age of each group of mice in months ± SEM for vehicle: 5.6 ± 0.2 and CP55,940: 5.4 ± 0.2. Unpaired t-test was used to determine significance. *P < 0.05, **P < 0.01. Abbreviations: AU, action units; MGS, mouse grimace scale.

Figure 7

Correlation between body parameters and MGS in sickle mice. Pain determination using body parameters correlates to the MGS in sickle mice. (A) Length demonstrates an inverse correlation (R = −0.7178) with a value of R² = 0.5153, P < 0.0001. (B). Eccentricity demonstrates a moderate inverse correlation (R = −0.5780) and the resulting coefficient of value of R² = 0.3341, P < 0.0001. This inverse correlation demonstrates a tendency for high MGS scores to go with low body parameter values (and vice versa). Values are from all the mice used in this study including pre- and post-cold exposure and pre- and post-CP55,940 treatment. Abbreviations: MGS, mouse grimace scale.