Differential Effects of the Inactivation of Anterior and Posterior Orbitofrontal Cortex on Affective Responses to Proximal and Distal Threat, and Reward Anticipation in the Common Marmoset

Abstract Structural and functional abnormalities of the orbitofrontal cortex (OFC) have been implicated in affective disorders that manifest anxiety-related symptoms. However, research into the functions of primate OFC has predominantly focused on reward-oriented rather than threat-oriented responses. To redress this imbalance, the present study performed a comprehensive analysis of the independent role of 2 distinct subregions of the central OFC (anterior area 11; aOFC and posterior area 13; pOFC) in the processing of distal and proximal threat. Temporary inactivation of both aOFC and pOFC heightened responses to distal threat in the form of an unknown human, but not to proximal threat assessed in a discriminative Pavlovian conditioning task. Inactivation of the aOFC, however, did unexpectedly blunt conditioned threat responses, although the effect was not valence-specific, as conditioned appetitive responses were similarly blunted and appeared restricted to a discriminative version of the task (when both CS− and CS+ are present within a session). Inactivation of the pOFC did not affect conditioned responses to either proximal threat or reward and basal cardiovascular activity was unaffected by manipulations of activity in either subregion. The results highlight the contribution of aOFC and pOFC to regulation of responses to more distal uncertain but not proximal, certain threat and reveal their opposing contribution to that of the immediately adjacent medial OFC, area 14.

pairwise comparisons revealing a significant difference between the first and last session (p<0.001) and no significant difference between the last and penultimate session (p=0.515). Bars represent the mean, with error bars representing the standard error of the mean. The individual data points for each animal are also presented as designated in Table 1. Significance symbols: n.s., p>0.05; **, p<0.01; ***, p<0.001, pairwise comparisons (session), post hoc for ANOVA (as above).

Supplementary Figure 2 -Marmosets show evidence of successful discriminative conditioning to threat. Data
presented is the mean of responses of each subject over the 3 discriminative conditioning sessions immediately before drug treatment began. Since the number of CStrials varied across session types, the responses to the CSfor each session were combined as a mean. This is presented for CS-directed behaviour (A) and cardiovascular responses (HR; B) for both CSand CS + . The US + -directed cardiovascular responses (US-CS HR) are also presented (C), confirming responsiveness to the threat itself. Paired t-tests confirmed a significant difference between the mean CS + and CS --directed responses for both behavioural and cardiovascular measures (CS + vs CS -; behaviour: t=14.39, p<0.001; HR: t=4.08, p=0.027). A one-sample two-tailed t-test was also employed to test the mean US + response, revealing a significant rise in HR exceeding 0 (t=11.81, p=0.001). The bars represent the mean, with the error bars representing the standard error of the mean. Individual data points are presented for each animal as designated in Table 1 (n=4 for all). Significance symbols: * is p<0.05, ** is p<0.01, *** is p<0.001, paired ttests (as above).
Supplementary Figure 3 -Heart rate showed a consistent and significant recovery following CS + presentation with US + omission across repeated control sessions (discriminative conditioning to threat). Graphs present the results from the experiment using a standard CS -/CS + /CSsession, where the CS + is presented for 20 seconds without US + presentation. Data from two saline sessions (aOFC and pOFC) is compared, analysed based on the order in which they took place rather than the target area (there was an inactivation session in-between the two saline sessions).
A. presents the CS + -directed HR responses and B demonstrates the recovery of HR responses across the recovery period. The 60-second recovery period is split into 3 bins (Rec20, Rec40, Rec60), with HR presented relative to the CS-directed response. The CS-directed response is presented for reference and was not included in the analysis of recovery. Repeated probe sessions with US + omission did not significantly affect CS-directed responses: the Wilcoxon test on HR responses showed no significant difference between the two control infusions (W=10.5, p=0.561). CS-directed responses showed a significant rise above baseline (one-sample tests against 0;

Supplementary Figure 4 -Marmosets show evidence of successful discriminative conditioning to reward. Data
presented is the mean of responses of each subject over the 3 discriminative conditioning sessions immediately before drug treatment began. This is presented for CS-directed cardiovascular (sysBP; A), and behavioural responses (B) for both CSand CS + . The US + -directed cardiovascular response is also presented (sysBP; C), demonstrating a sustained increase in sysBP during marshmallow consumption. Successful discrimination between CS + and CSwas determined using paired t-tests (sysBP: t=10.2, p<0.001; behaviour: t=11.91, p<0.001).
The same test conducted on cardiovascular responses to the US relative to the preceding CS revealed a significant difference between the US + and the US -(t=4.33, p=0.008). The bars represent the mean, with the error bars representing the standard error of the mean. Individual data points are presented for each animal as designated in Table 1 (n=6 for all). Significance symbols: ** p<0.01, *** p<0.001, paired t-tests (as above).