Figure 2 Freezing Behaviour During TMT Presentation

Kappa-opioid receptor antagonism modulates basal and stress-induced freezing. (A) CPP was conducted such that each mouse was exposed to TMT for 5-minutes on their preferred side of the behavioural testing apparatus. (B) NorBNI administration 24hours before testing increases basal freezing and decreases freezing during TMT presentation. (C) The number of freezing episodes was increased at Neutral by NorBNI administration. (D) NorBNI did not produce any changes in the average duration of freezing episode under basal conditions or during TMT exposure. All mice exhibit an increase in the % Freezing between the Neutral and TMT sessions (E), but only the saline group increased the number of freezing episodes between the two sessions (F). NorBNI administration did not affect the increase in average duration of freezing episode observed between the Neutral and TMT sessions (G). Data pressented as mean +/- SEM. * Indicates p < 0.05, *** indicates p < 0.001.

Figure 2.1: Kappa-opioid receptor antagonism modulates basal and stress-induced freezing. (A) CPP was conducted such that each mouse was exposed to TMT for 5-minutes on their preferred side of the behavioural testing apparatus. (B) NorBNI administration 24hours before testing increases basal freezing and decreases freezing during TMT presentation. (C) The number of freezing episodes was increased at Neutral by NorBNI administration. (D) NorBNI did not produce any changes in the average duration of freezing episode under basal conditions or during TMT exposure. All mice exhibit an increase in the % Freezing between the Neutral and TMT sessions (E), but only the saline group increased the number of freezing episodes between the two sessions (F). NorBNI administration did not affect the increase in average duration of freezing episode observed between the Neutral and TMT sessions (G). Data pressented as mean +/- SEM. * Indicates p < 0.05, *** indicates p < 0.001.

2.1 Statistical Analyses

2.1.1 Two-way ANOVA on Freezing behaivour

a <- aov(Frz_Tm ~ Drug * Task, data=train_data)
summary(a)

##             Df Sum Sq Mean Sq F value  Pr(>F)    
## Drug         1    340     340   0.555 0.45835    
## Task         1  95925   95925 156.610 < 2e-16 ***
## Drug:Task    1   6511    6511  10.631 0.00165 ** 
## Residuals   78  47776     613                    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

We quantified freezing behaviour both during the neutral pairing and the TMT exposure to investigate whether Nor BNI administration affected basal or stress-induced levels of freezing.

We computed a 2x2 ANOVA with Drug (saline or Nor BNI) as the between-groups factor and timepoint (Neutral vs TMT) as the within-subjects variable. The omnibus test for this model indicated a significant main effect of timepoint (F(1,39) = 79.09, p < 0.001) and a timepoint * Drug interaction (F(1,39) = 11.41, p < 0.001). Follow- up comparisons of the significant interaction indicated that NorBNI-injected mice froze more than did controls during the Neutral exposure (p < 0.001) and that Nor-BNI-injected mice spent less time freezing than controls during the TMT exposure (p = 0.042). Nevertheless, both groups spent more time freeing during TMT presentation than they had at Neutral (both p < 0.001), suggesting that both groups exhibited an innate defensive response during the single 5-minute TMT session.

a <- anova_test(data=train_data, dv=Frz_Tm, wid=ID, within = Task, between = Drug)
knitr::kable(get_anova_table(a))

Effect	DFn	DFd	F	p	p<.05	ges
Drug	1	39	0.522	0.474		0.007
Task	1	39	169.279	0.000	*	0.670
Drug:Task	1	39	11.353	0.002	*	0.120

b <- train_data %>%
  group_by(Drug)%>%
  pairwise_t_test(
  Frz_Tm ~ Task, paired = TRUE,
  p.adjust.method = "bonferroni"
  )

knitr::kable(b)

Drug	.y.	group1	group2	n1	n2	statistic	df	p	p.adj	p.adj.signif
Saline	Frz_Tm	Neutral	TMT	20	20	-10.722043	19	0e+00	0e+00	****
Nor-BNI	Frz_Tm	Neutral	TMT	21	21	-7.409239	20	4e-07	4e-07	****

c <- train_data %>%
  group_by(Task)%>%
  pairwise_t_test(
  Frz_Tm ~ Drug, paired = FALSE,
  p.adjust.method = "bonferroni"
  )

knitr::kable(c)

Task	.y.	group1	group2	n1	n2	p	p.signif	p.adj	p.adj.signif
Neutral	Frz_Tm	Saline	Nor-BNI	20	21	0.000182	***	0.000182	***
TMT	Frz_Tm	Saline	Nor-BNI	20	21	0.042000	*	0.042000	*

We also computed ANOVA with the number of freezing episodes during each session entered as the dependent variable. This model also indicated a Drug * Timepoint interaction (F(1,39) = 13.17, p < 0.001). Saline-injected mice froze more often during the TMT session than the Neutral session (p < 0.001) whereas there was no difference in freezing frequency between the two sessions for NorBNI-injected mice (p = 0.28). Moreover, NorBNI-injected mice froze more often than did saline controls during the neutral exposure (p < 0.001), but there was no difference in freezing frequency between the groups during the TMT session (p = 0.09).

a <- anova_test(data=train_data, dv=Frz_Freq, wid=ID, within = Task, between = Drug)
get_anova_table(a)

## ANOVA Table (type III tests)
## 
##      Effect DFn DFd      F        p p<.05   ges
## 1      Drug   1  39  1.602 2.13e-01       0.019
## 2      Task   1  39 29.078 3.61e-06     * 0.286
## 3 Drug:Task   1  39 13.170 8.15e-04     * 0.153

b <- train_data %>%
  group_by(Drug)%>%
  pairwise_t_test(
  Frz_Freq ~ Task, paired = TRUE,
  p.adjust.method = "bonferroni"
  )

knitr::kable(b)

Drug	.y.	group1	group2	n1	n2	statistic	df	p	p.adj	p.adj.signif
Saline	Frz_Freq	Neutral	TMT	20	20	-7.644742	19	3.00e-07	3.00e-07	****
Nor-BNI	Frz_Freq	Neutral	TMT	21	21	-1.105378	20	2.82e-01	2.82e-01	ns

c <- train_data %>%
  group_by(Task)%>%
  pairwise_t_test(
  Frz_Freq ~ Drug, paired = FALSE,
  p.adjust.method = "bonferroni"
  )

knitr::kable(c)

Task	.y.	group1	group2	n1	n2	p	p.signif	p.adj	p.adj.signif
Neutral	Frz_Freq	Saline	Nor-BNI	20	21	0.00065	***	0.00065	***
TMT	Frz_Freq	Saline	Nor-BNI	20	21	0.09460	ns	0.09460	ns

ANOVA on the average duration of freezing episode indicated a significant main effect of timepoint (F(1,39) = 103.45, p < 0.001) that did not interact with drug treatment (p = 0.24). The average duration of freezing episode was increased by TMT.

a <- anova_test(data=train_data, dv=Av_Dur, wid=ID, within = Task, between = Drug)
get_anova_table(a)

## ANOVA Table (type III tests)
## 
##      Effect DFn DFd       F        p p<.05      ges
## 1      Drug   1  39   0.029 8.66e-01       0.000396
## 2      Task   1  39 103.456 1.58e-12     * 0.553000
## 3 Drug:Task   1  39   1.452 2.35e-01       0.017000

b <- train_data %>%
  group_by(Drug)%>%
  pairwise_t_test(
  Av_Dur ~ Task, paired = TRUE,
  p.adjust.method = "bonferroni"
  )

knitr::kable(b)

Drug	.y.	group1	group2	n1	n2	statistic	df	p	p.adj	p.adj.signif
Saline	Av_Dur	Neutral	TMT	20	20	-6.971897	19	1.2e-06	1.2e-06	****
Nor-BNI	Av_Dur	Neutral	TMT	21	21	-7.589131	20	3.0e-07	3.0e-07	****

c <- train_data %>%
  group_by(Task)%>%
  pairwise_t_test(
  Av_Dur ~ Drug, paired = FALSE,
  p.adjust.method = "bonferroni"
  )

knitr::kable(c)

Task	.y.	group1	group2	n1	n2	p	p.signif	p.adj	p.adj.signif
Neutral	Av_Dur	Saline	Nor-BNI	20	21	0.00381	**	0.00381	**
TMT	Av_Dur	Saline	Nor-BNI	20	21	0.50100	ns	0.50100	ns

Taken together, these findings indicate that NorBNI administration increases time spent freezing at Neutral by increasing the frequency of bouts of freezing. TMT exposure increases both the frequency and average duration of bouts of freezing among naive mice, and the magnitude of the change in freezing behaviour is reduced among NorBNI treated mice.

2.1.2 Regressions

## 
## Call:
## lm(formula = Perc ~ Frz_Freq * Drug, data = BL_data)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2.8267 -0.7702 -0.1686  0.3901  8.3884 
## 
## Coefficients:
##                      Estimate Std. Error t value Pr(>|t|)    
## (Intercept)           0.35976    1.03281   0.348 0.729567    
## Frz_Freq              0.23889    0.06017   3.970 0.000318 ***
## DrugNor-BNI          -1.78723    1.40227  -1.275 0.210422    
## Frz_Freq:DrugNor-BNI  0.15621    0.06761   2.311 0.026535 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.804 on 37 degrees of freedom
## Multiple R-squared:  0.8815, Adjusted R-squared:  0.8719 
## F-statistic: 91.78 on 3 and 37 DF,  p-value: < 2.2e-16

## 
## Call:
## lm(formula = Perc ~ Frz_Freq * Drug, data = b)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2.8267 -0.7702 -0.1686  0.3901  8.3884 
## 
## Coefficients:
##                     Estimate Std. Error t value Pr(>|t|)    
## (Intercept)         -1.42747    0.94851  -1.505   0.1408    
## Frz_Freq             0.39510    0.03082  12.820 3.51e-15 ***
## DrugSaline           1.78723    1.40227   1.275   0.2104    
## Frz_Freq:DrugSaline -0.15621    0.06761  -2.311   0.0265 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.804 on 37 degrees of freedom
## Multiple R-squared:  0.8815, Adjusted R-squared:  0.8719 
## F-statistic: 91.78 on 3 and 37 DF,  p-value: < 2.2e-16

A regression model with Percent freezing during the Neutral session entered as the dependent variable and freezing frequency entered as the independent variable accounted for 87% of the variability in freezing (F(3,37) = 91.78, Adjusted R^2 = 0.87, p < 0.001). Simple effects for this model indicated a significant main effect of Freezing Frequency (t = 3.97, p < 0.001) and a significant Frequency * Drug interaction (t = 2.33, p = 0.026). Evaluation of the simple slopes for each drug condition indicated that for saline-treated mice, a 10-unit increase in freezing frequency was associated with a 2.3% increase in total time freezing. Among NorBNI-treated mice, a 10-unit increase in freezing was associated with a 3.9% increase in total time freezing.

## 
## Call:
## lm(formula = Perc ~ Frz_Freq * Drug, data = TMTs)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -26.5697  -7.2100   0.5262   9.3678  23.8668 
## 
## Coefficients:
##                      Estimate Std. Error t value Pr(>|t|)   
## (Intercept)           26.5527     9.5304   2.786  0.00837 **
## Frz_Freq               0.3165     0.2349   1.347  0.18607   
## DrugNor-BNI          -12.8500    13.1699  -0.976  0.33554   
## Frz_Freq:DrugNor-BNI   0.1866     0.3551   0.525  0.60249   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 12.79 on 37 degrees of freedom
## Multiple R-squared:  0.2159, Adjusted R-squared:  0.1523 
## F-statistic: 3.395 on 3 and 37 DF,  p-value: 0.02779

# Model 2: X=Av_Dur, y=Perc
a <- lm(Perc~Av_Dur * Drug, data=BL_data)
summary(a)

## 
## Call:
## lm(formula = Perc ~ Av_Dur * Drug, data = BL_data)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -7.8633 -1.6178 -0.8294  1.2113 10.2455 
## 
## Coefficients:
##                    Estimate Std. Error t value Pr(>|t|)  
## (Intercept)         -0.1638     2.8444  -0.058   0.9544  
## Av_Dur               6.6422     4.2461   1.564   0.1263  
## DrugNor-BNI         -5.6693     4.2286  -1.341   0.1882  
## Av_Dur:DrugNor-BNI  12.0048     5.6131   2.139   0.0391 *
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.292 on 37 degrees of freedom
## Multiple R-squared:  0.6058, Adjusted R-squared:  0.5738 
## F-statistic: 18.95 on 3 and 37 DF,  p-value: 1.3e-07

# Re-order the "drug" variable to get the intercept for the NorBNI group.
b <- BL_data
b$Drug <- as.character(b$Drug)
b$Drug <- factor(b$Drug,levels = c("Nor-BNI","Saline"))

# A regression model on Neutral Freezing
a <- lm(Perc~Av_Dur * Drug, data=b)
summary(a)

## 
## Call:
## lm(formula = Perc ~ Av_Dur * Drug, data = b)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -7.8633 -1.6178 -0.8294  1.2113 10.2455 
## 
## Coefficients:
##                   Estimate Std. Error t value Pr(>|t|)    
## (Intercept)         -5.833      3.129  -1.864   0.0702 .  
## Av_Dur              18.647      3.671   5.079  1.1e-05 ***
## DrugSaline           5.669      4.229   1.341   0.1882    
## Av_Dur:DrugSaline  -12.005      5.613  -2.139   0.0391 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.292 on 37 degrees of freedom
## Multiple R-squared:  0.6058, Adjusted R-squared:  0.5738 
## F-statistic: 18.95 on 3 and 37 DF,  p-value: 1.3e-07