Assessment of paclitaxel induced sensory polyneuropathy with "Catwalk" automated gait analysis in mice

Abstract

Neuropathic pain as a symptom of sensory nerve damage is a frequent side effect of chemotherapy. The most common behavioral observation in animal models of chemotherapy induced polyneuropathy is the development of mechanical allodynia, which is quantified with von Frey filaments. The data from one study, however, cannot be easily compared with other studies owing to influences of environmental factors, inter-rater variability and differences in test paradigms. To overcome these limitations, automated quantitative gait analysis was proposed as an alternative, but its usefulness for assessing animals suffering from polyneuropathy has remained unclear. In the present study, we used a novel mouse model of paclitaxel induced polyneuropathy to compare results from electrophysiology and the von Frey method to gait alterations measured with the Catwalk test. To mimic recently improved clinical treatment strategies of gynecological malignancies, we established a mouse model of dose-dense paclitaxel therapy on the common C57Bl/6 background. In this model paclitaxel treated animals developed mechanical allodynia as well as reduced caudal sensory nerve action potential amplitudes indicative of a sensory polyneuropathy. Gait analysis with the Catwalk method detected distinct alterations of gait parameters in animals suffering from sensory neuropathy, revealing a minimized contact of the hind paws with the floor. Treatment of mechanical allodynia with gabapentin improved altered dynamic gait parameters. This study establishes a novel mouse model for investigating the side effects of dose-dense paclitaxel therapy and underlines the usefulness of automated gait analysis as an additional easy-to-use objective test for evaluating painful sensory polyneuropathy.

Figure 1. Dose-dense paclitaxel treatment induces a sensory polyneuropathy in mice.

(A) Schedule of paclitaxel treatment and behavioral as well as electrophysiological testing. Animals received paclitaxel at a dose of 20 mg paclitaxel/kg BW three times per week over four weeks with a cumulative dose of 240 mg/kg BW. (B) Paclitaxel treated animals gained less weight than controls, but their weight quickly normalized after the last injection. (C) Motor coordination in the rotarod test was comparable in paclitaxel and vehicle injected animals. (D) Paclitaxel treated animals developed mechanical allodynia with a significantly reduced mechanical withdrawal threshold of the hind paws as well as a (E) predominantly axonal neuropathy with a diminished sensory nerve action potential amplitude (SNAP) of the caudal nerve. (F) Alterations of the caudal nerve SNAP showed a strong positive correlation with the mechanical withdrawal threshold (dashed line depicts the 95% prediction interval). *** p<0.001 compared to the control group at the same time point (One-way ANOVA).

Figure 2. Paclitaxel induced polyneuropathy can be detected with the Catwalk method.

(A) A typical Catwalk run; the animal crosses from right to left. Paws are automatically detected and labeled by the program, in this example right hind paw (RH) and the left forepaw. Examples of gait parameters for the right hind paw are displayed at the bottom of the graph in respect to the step cycle. (B) Animals treated with paclitaxel develop distinct gait alterations compared to vehicle controls. The swing phase (squares, long dashed line) increases, while the stance phase (open triangles, dotted line) and the duty cycle (circles, solid line) decreases. Duty cycle expresses the stance phase as a percentage of the entire step cycle (stand + swing). (C) The print area of the hind paws is also significantly reduced in neuropathic animals. (D) The duty cycle of the hind paws shows a small correlation with the mechanical withdrawal threshold of the animals. Similar results are obtained for the correlation of the (E) hind paw stance phase duration with the caudal nerve SNAP. Solid lines in (D-E) signify linear regression lines, while medium dashed lines depict the 95% prediction interval. * p<0.05; *** p<0.001; both compared to the control group at the same time point (One-way ANOVA).

Figure 3. Effects of gabapentin on mechanical allodynia and gait alterations.

(A) Schedule of gabapentin (GBP) treatment after the induction of neuropathy with paclitaxel (PTX) and behavioral testing (Test). Animals were allocated to four groups: two groups received paclitaxel and two vehicle (VEH) injections. After the induction of neuropathy, one vehicle and one paclitaxel group were treated with daily gabapentin injections, whereas the other two groups received vehicle. (B) Paclitaxel induced mechanical allodynia was significantly improved after 8 days of gabapentin treatment as well as (C) the dynamic gait parameter duty cycle, whereas (D) the gait parameter hind paw print area was not affected by gabapentin treatment. * p<0.05; *** p<0.001 (Two-way ANOVA).