DCC Confers Susceptibility to Depression-like Behaviors in Humans and Mice and Is Regulated by miR-218

Abstract

Backgroud: Variations in the expression of the Netrin-1 guidance cue receptor DCC (deleted in colorectal cancer) appear to confer resilience or susceptibility to psychopathologies involving prefrontal cortex (PFC) dysfunction.

Methods: With the use of postmortem brain tissue, mouse models of defeat stress, and in vitro analysis, we assessed microRNA (miRNA) regulation of DCC and whether changes in DCC levels in the PFC lead to vulnerability to depression-like behaviors.

Results: We identified miR-218 as a posttranscriptional repressor of DCC and detected coexpression of DCC and miR-218 in pyramidal neurons of human and mouse PFC. We found that exaggerated expression of DCC and reduced levels of miR-218 in the PFC are consistent traits of mice susceptible to chronic stress and of major depressive disorder in humans. Remarkably, upregulation of Dcc in mouse PFC pyramidal neurons causes vulnerability to stress-induced social avoidance and anhedonia.

Conclusions: These data are the first demonstration of microRNA regulation of DCC and suggest that, by regulating DCC, miR-218 may be a switch of susceptibility versus resilience to stress-related disorders.

Keywords: Chronic social defeat stress; Guidance cue; Major depressive disorder; Neurodevelopment; Resilience; microRNA.

Figure 1

Opposite DCC and miR-218 regulation in PFC in MDD. (A) Summary of the information about the age and gender of the human subjects in the validation cohort. Brain samples were compared for age, RNA integrity number (RIN), post-mortem intervals (PMI), and pH value. There are no differences between groups (Age: t₍₂₁₎=0.39; RIN: t₍₂₁₎=0.55; PMI: t₍₂₁₎= 1.44; pH: t₍₂₁₎=0.92, p values >0.05). (B) DCC mRNA in the PFC is increased in MDD versus controls (t₍₂₁₎=2.50;*p=0.02) in the validation cohort. (C) miRNAs predicted to bind DCC. (D) Summary of the information about subjects in the original cohort. There are no differences between groups (Age: t₍₅₇₎=0.71; RIN: t₍₅₇₎=t=0.84; PMI: t₍₅₇₎= 0.33; pH: t₍₅₇₎=1.25). (E) Reduction in miR-218 in depressed versus control groups (t₍₅₇₎=2.71;*p=0.008) in original cohort (4). (F) Negative DCC-miR-218 correlation in MDD. (G) Decreased miR-218 in PFC of depressed versus control groups in validation cohort (t₍₂₁₎=2.30;*p=0.031). (H) DCC+ neurons and (I) DCC and miR-218 co-expression in PFC neurons. Scale bar= 50μm. The error bars represent the SEM.

Figure 2

Opposite expression of Dcc and miR-218 in PFC is associated with susceptibility to stress-induced depression-like behaviors. *† indicate different from control (CON) and resilient (RES), respectively. (A) CSDS protocol. (B) Social interaction ratio in CON (n=26), susceptible (SUS; n=27) and RES (n=27). F_(2,77)=94.37;p< 0.0001;*†p<0.01. (C) Increased PFC Dcc expression in SUS (n=12–13; F_(2,34)=6.725;p=0.003,*†p<0.01). Decreased PFC miR-218 expression in SUS (n=10–11; F_(2,28)=5.85; p=0.0075,*p<0.01; †p<0.05). Elevated PFC DCC protein in SUS (n= 4; F_(2,9)= 8.085; p=0.098, *†p<0.05). The mRNA, microRNA, and protein data were derived from tissue samples of susceptible, resilient and control mice collected from 3 different and independent CSDS experiments.(D) Co-expression of DCC/miR-218 in PFC neurons. Scale bars =50 μm.

Figure 3

Knocking out Dcc in PFC pyramidal neurons protects against stress-induced depression-like behaviors. (A) DCC expression in SMI-32+ medial PFC pyramidal neurons in adult stress-naïve mice (white arrows). (B) Timeline of infection experiments and injection site. PFC pyramidal neuron Dcc deletion prevents social avoidance induced by CSDS. (C) Ratio: Two-way ANOVA: stress: F_(1,30)=10.91; p=0.0025; stress by virus interaction: F_(1,30)=8.87; p=0.0057; Tuckey test: *^¤†p<0.05. (D) Corners: Two-way ANOVA: stress: F_(1,30)=8.77; p=0.0059; stress by virus interaction: F_(1,30)=7.83; p=0.0089;*†p<0.05, ^¤p<0.01. Symbols indicate different from *control-AAV-CaMKIIα-GFP (n=7), ^¤control-AAV-CaMKIIα-Cre-GFP (n=10) and †defeat-AAV-CaMKIIα-Cre-GFP (n=9); p<0.05.(E) Locomotor activity during the social interaction test is similar across all groups: Two-way ANOVA: stress by virus interaction: F_(1,30)= 0.004; p= 0.94, group: F_(1,30)= 0.36; p= 0.55, and virus: F_(1,30)=0.48; p=0.49).(F) Sucrose preference: Two-way ANOVA: stress by virus interaction: F_(1,26)=5.26; p=0.030. (G) Deletion of Dcc does not prevent anxiety-related behaviors induced by CSDS: Two-way ANOVA: significant main effect of stress, F_(1,30)=15.81; p=0.0004; *different from control; p<0.05). (H) Deletion of Dcc in individual GFP+/SMI-32+ neurons (I) Stereological analysis show that Dcc deletion does not compromise the survival of medial PFC SMI-32+ neurons (t₍₈₎= 0.57, p=0.58). Scale bars =50 μm.

Figure 4

miR-218 regulates DCC expression.(A) Immunofluorescence and Western blot experiments showing endogenous DCC expression in the IMR-32 cell line. Numbers on top of the DCC bands represent micrograms of protein loaded. Scale bar= 50μm. (B) Significant decreased of DCC mRNA expression in IMR-32 cells 24h after transfection with a miR-218 mimic (n=3) in comparison to transfection with reagent alone (mock; n=3) or mimic scramble (miR-Scr; n=3) (F_(2,6)=11.75; p=0.008; *different from mock; †different from miR-Scr, *†p<0.05). (C) Transfection with mimic in comparison to transfection reagent (mock) or mimic scrambled (miR-Scr) increases miR-218 in IMR-32 cells 24h after. One-way ANOVA: F_(2,6)=17.25; p<0.01; *different from mock; †different from miR-Scr; *†p<0.001). (D) miR-218 and DCC base pairing at the two binding sites. (E) Target protector experiment. (F) Decreased DCC expression in IMR-32 cells by miR-218 mimic 24 after transfection is reversed by target protectors (F_(6,16)=5.493; p= 0.0030; *different from mock; †different from negative control *p<0.05). n=3–5. (G) miR-218-induced downregulation of ROBO1 in IMR-32 cells is not prevented by DCC-specific target protectors (F_(6,16)= 4.546; p= 0.0071; *different from mock. *p<0.05). n=3–5.