Increased MMP-9 levels with strain-dependent stress resilience and tunnel handling in mice

Abstract

Increased perineuronal net (PNN) deposition has been observed in association with corticosteroid administration and stress in rodent models of depression. PNNs are a specialized form of extracellular matrix (ECM) that may enhance GABA-mediated inhibitory neurotransmission to potentially restrict the excitation and plasticity of pyramidal glutamatergic neurons. In contrast, antidepressant administration increases levels of the PNN-degrading enzyme matrix metalloproteinase-9 (MMP-9), which enhances glutamatergic plasticity and neurotransmission. In the present study, we compare pro-MMP-9 levels and measures of stress in females from two mouse strains, C57BL/6 J and BALB/cJ, in the presence or absence of tail grasping versus tunnel-associated cage transfers. Prior work suggests that C57BL/6 J mice show relatively enhanced neuroplasticity and stress resilience, while BALB/c mice demonstrate enhanced susceptibility to adverse effects of stress. Herein we observe that as compared to the C57BL/6 J strain, BALB/c mice demonstrate a higher level of baseline anxiety as determined by elevated plus maze (EPM) testing. Moreover, as determined by open field testing, anxiety is differentially reduced in BALB/c mice by a choice-driven tunnel-entry cage transfer technique. Additionally, as compared to tail-handled C57BL/6 J mice, tail-handled BALB/c mice have reduced brain levels of pro-MMP-9 and increased levels of its endogenous inhibitor, tissue inhibitor of metalloproteinase-1 (TIMP-1); however, tunnel-associated cage transfer increases pro-MMP-9 levels in BALB/c mice. BALB/c mice also show increases in Western blot immunoreactive bands for brevican, a constituent of PNNs. Together, these data support the possibility that MMP-9, an effector of PNN remodeling, contributes to the phenotype of strain and handling-associated differences in behavior.

Keywords: BALB/cJ; C57BL/6J; MMP-9; PNN; Stress; Tunnel handling.

figure 1.. Strain related differences in percent time spent in EPM closed arms

Time spent in closed arms is shown (figure 1A), as well as total distance traveled in all arms (figure 1B). The difference in percent time in the closed arms was significant at p<0.0001 (n=20 mice per group, t (38)=4.5, Student’s t-test). The difference in distance traveled was also significant at p< 0.0001 (n=20 mice per group, t (38)=11.09, Student’s t-test).

figure 2.. Four weeks of twice weekly tail or tunnel cage transfers on EPM behavior in C57BL/6J and BALB/c mice.

The percent time spent in the open and closed arms is shown for each strain and handling technique. Results for the percent closed arm time are shown in figure 2. A two-way ANOVA with Sidak’s post hoc testing showed that there was a significant difference between strains (n=9–10 per group, p < 0.0001), but no difference as a function of handling (p= 0.89 for C57BL/6J and p=0.32 for BALB/c). In addition, there was no significant interaction between the two variables (p=0.52).

figure 3.. Tunnel handling increases open field activity in BALB/c mice

Open field testing was also performed 4 weeks following twice weekly tail or tunnel cage transfers in C57BL/6J and BALB/c mice (n=8 animals per group). Total distance traveled in cm/60 minutes is show in figure 3A. Two-way ANOVA with Sidak’s post hoc testing of these data shows that while there is no interaction between the two variables of strain and handling, there is a significant effect of both strain (p=0.0001) and handling (p=0.0012). The difference between tail and tunnel handled C57BL/6J mice is not significant on post hoc testing but the difference between tail and tunnel handled BALB/c mice is significant (p=0.0065). In figures 3B and 3C we show raw data (traces) of activity for tunnel (3B) and tail (3C) handled BALB/c mice.

figure 4.. MMP-9 and TIMP-1 levels in brain lysates

MMP-9 and its endogenous inhibitor TIMP-1 were measured by ELISA in tail and tunnel-handled BALB/c and C57BL/6J mice. Results are shown in figure 4. Two-way ANOVA with Tukey’s post hoc testing of MMP-9 data shown in figure 4A demonstrate no interaction between handling and strain, but a significant effect of strain on MMP-9 levels [8–10 animals per group; F(1,35)=23; p<0.0001]. While the difference between tail-handled C57Bl/6J and BALB/c mice was significant (p<0.005), the difference between the two strains in the tunnel-handled groups did not reach significance. This is consistent with a greater effect of tunnel handling on the BALB/c mice, and though the effect of tunnel handling on BALB/c mice does not reach significance by two-way ANOVA, a Student’s t–test of BALB/c tunnel versus BALB/c tail handled mice demonstrates a significant difference between groups (p=0.024). TIMP-1 and MMP-9/TIMP-1 values are shown in figures 4B and C respectively. Two-way ANOVA with Tukey’s post hoc testing of these data again demonstrate no interaction between handling and strain, but a significant effect of strain [8–10 animals per group; p<0.0001].

figure 5.. High molecular weight forms of brevican are elevated in BALB/c mice

Western blot analyses of brain lysates from tail and tunnel handled C57BL/6J and BALB/c mice with quantification of the 150 and 130 kDa forms (arrows) as well as the main ponceau band (arrow). Blots and quantification of brevican/ponceau intensity, shown in figure 5A (tunnel) and 5B (tail), demonstrate reduced levels of 100 kDa brevican in C57BL/6J as compared to BALB/c mice. The differences are significant [p<0.029, t (7)=2.75 for tunnel-handled animals and p=0.006, t(7)=3.96 for tail-handled animals; Student’s t-test].