Dissociable effects of advanced age on prefrontal cortical and medial temporal lobe ensemble activity

Abstract

The link between age-related cellular changes within brain regions and larger scale neuronal ensemble dynamics critical for cognition has not been fully elucidated. The present study measured neuron activity within medial prefrontal cortex (PFC), perirhinal cortex (PER), and hippocampal subregion CA1 of young and aged rats by labeling expression of the immediate-early gene Arc. The proportion of cells expressing Arc was quantified at baseline and after a behavior that requires these regions. In addition, PER and CA1 projection neurons to PFC were identified with retrograde labeling. Within CA1, no age-related differences in neuronal activity were observed in the entire neuron population or within CA1 pyramidal cells that project to PFC. Although behavior was comparable across age groups, behaviorally driven Arc expression was higher in the deep layers of both PER and PFC and lower in the superficial layers of these regions. Moreover, age-related changes in activity levels were most evident within PER cells that project to PFC. These data suggest that the PER-PFC circuit is particularly vulnerable in advanced age.

Keywords: Arc; CA1; Cognition; Infralimbic cortex; Prelimbic cortex.

Fig. 1.

Behavioral paradigm on final testing day. (A) Experimental timeline. (B) Objects used during the OPPA task are shown. During the OPPA task, rats traversed back and forth between the two arms of the testing apparatus, choosing an object in each arm. In the left arm, the chicks object was correct, but in the right arm the frog object was correct. Following a 20-minute rest in their home cages, animals then performed an alteration task in which they traversed back and forth between the two start platforms for reward, but did not perform the object discrimination component. Order of testing was counter balanced across rats. (C) Representative images of subcellular distribution of Arc within neurons. Subcellular location was used to infer which behavioral epoch a neuron was active in. Scale bar is 20 μm.

Fig. 2.

(A) Location of regions imaged (PL: lightblue, IL: dark blue, CA1: pink, PER: purple). Black squares indicate deep andred squares indicate superficial areas of interest. (B) Representative images of Arc (red) and CTB (green) labeling within the different brain regions analyzed. Scale bars represent 20mm. (C) Representative image of PFC with CTB (green) injection and schematic of injection site spread.

Fig. 3.

OPPAand alternation task performance. (A) During OPPA task acquisition, aged rats made significantly more incorrect trials before reaching stable criterion performance (p < 0.01). (B) Behavioral performance and (C) number of trials performed across age groups did not differ on the final day of the task. All values represent the mean ± the standard error of the mean (SEM) and * indicates p value ≤0.05 across age groups.

Fig. 4.

Baseline percentage of cells expressing Arc in young and aged caged control rats for (A) perirhinal cortex (PER), (B) CA1, (C) prelimbic cortex (PL), and (D) infralimbic cortex (IL). There was a significant age-related increase in activity within the superficial layers of PL (p = 0.01) and within both layers of IL (p = 0.03). All values represent the mean ± the standard error of the mean (SEM), ^# indicates main effect of age across both layers and * indicates significant effect of layer.

Fig. 5.

Percentof behaviorally induced Arc positive cells normalized to caged control levels within (A, B) deep and superficial layers of the perirhinal cortex (PER), (C, D) dorsal and ventral CA1, (E, F) deep and superficial layers of the prelimbic cortex (PL), and (G, H) deep and superficial layers of the infralimbic cortex (IL). There is a significant age-related increase in Arc activity within deep layers of both the PER and PL (p ≤ 0.04 for both) and a significant age-related increase in Arc activity within the superficial layers of both the PER and PL (p ≤ 0.1 for both). All values represent the mean ± the standard error ofthe mean (SEM) and # symbol indicates main effect of age across both tasks.

Fig. 6.

Proportion of cells with both cytoplasmic and nuclear foci labeling for Arc expression with (A) perirhinal cortex (PER), (B) CA1, (C) prelimbic cortex (PL), and (D) infralimbic cortex (IL). There is a significant decrease in expression within the IL and PL (p < 0.001 for both) with age. There is also a significant difference across layers/subregions within PER, CA1 and PL (p ≤ 0.03 for all) and a significant interaction between age and layer within PL (p < 0.001). All values represent the mean ± the standard error of the mean (SEM), * indicate main effect of layer, ^# indicates main effect of age across both layers and ^† indicates interaction between age and layer.

Fig. 7.

Similarity scores for Arc expression with (A) perirhinal cortex (PER), (B) CA1, (C) prelimbic cortex (PL), and (D) infralimbic cortex (IL). There is a significant decrease in similarity score across PFC with age as a whole (p < 0.03), reaching significance individually within the IL (p = 0.007) and PL (p = 0.03) regions. All values represent the mean ± the standard error of the mean (SEM), and ^# indicates main effect of age across both layers.

Fig. 8.

Comparison of Arc activity restricted to within CTB− cells vs. activity restricted to within CTB+ projection cells demonstrated elevated signaling within CTB+ projection neurons within (A, B) deep and superficial perirhinal cortex (PER) and (C, D) dorsal and ventral CA1. All values represent the mean ± the standard error ofthe mean (SEM) and * indicates a significant effect of age across both tasks.