Asymmetric c-Fos Expression in the Ventral Orbital Cortex is Associated with Impaired Reversal Learning in a Right-Sided Neuropathy

Mechanical allodynia

Using the up-and-down assay to assess the 50% withdraw threshold we firstly showed that 3 weeks after the installation of the SNI model, there were marked differences to Von Frey monofilament probing between the three experimental groups (F_2,21 = 1518.301; P < 0.001). Post hoc analyses revealed a statistically significant threshold drop in the SNI groups (P < 0.001 in both sham vs SNI-L and sham vs SNI-R comparisons). No differences were however found between the two neuropathic groups SNI-L and SNI-R (Figure 1).

OPEN IN VIEWER

Figure 1

Mechanical allodynia. 3-weeks after SNI installation, a significant decrease in the monofilament-induced limb withdrawa threshold was observed independently of the injured side. **P < 0.01. Data presented as mean ±S.E.M.

Attentional set-shifting task

No differences were observed in the simple discrimination tasks for odor (SD_odo; F_2,21 = 1.589, n.s.) and texture (SD_tex; F_2,21 = 0.674, n.s.) indicating that all groups are able to equally discriminate both perceptual dimensions (Figure 2B). In the ASST proper, however, a group effect was observed in the reversal learning tasks Rev 1–3 (F_2,21 = 7.561, P < 0.01; F_2,21 = 4.611, P < 0.05; F_2,21 = 13.470; P < 0.001, respectively). In these cases, SNI-R required more trials to accomplish the criteria than sham or SNI-L groups as revealed by post hoc analyses (Figure 2B). In Rev 4, the same tendency was observed, although it did not reach statistical significance (F_2,21 = 2.045; n.s.). In all other steps of the ASST, compound discrimination, intradimensional shift (IDS1–2) and extradimensional shift (EDS) no significant differences were detected (CD, F_2,21 = 2.475; IDS1, F_2,21 = 0.337; EDS, F_2,21 = 1.774; IDS2, F_2,21 = 0.821) (Figure 2B).

OPEN IN VIEWER

Figure 2

Attentional set-shifting task. (A) The ASST was performed in a custom-built rectangular 70 × 40 × 20 cm (L, W, H) arena. (B) SNI-R animals required a significantly higher number of trials to criteria than sham and SNI-L animals to accomplish reversals (Rev) 1, 2 and 3 of the ASST task. No differences were found between groups in the other ASST steps. **P < 0.001. Data presented as mean ± S.E.M.

c-fos⁺ cell density in medial and orbital prefrontal cortices

c-fos⁺ cell densities were analyzed in medial and orbital prefrontal (mPFC and OFC, respectively) regions from both hemispheres and expressed as c-fos⁺/mm² (Figure 3B). Statistically significant differences were observed between the three groups in the mPFC areas cingulate (Cg) (F_2,43 = 5.446; P < 0.01), prelimbic (PrL) (F_2,43 = 5.314; P < 0.01) and infralimbic (IL) (F_2,43 = 5.657; P < 0.01) and in the OFC areas ventral orbital (VO) (F_2,43 = 11.640; P < 0.001) and lateral orbital (LO) (F_2,43 = 5.059; P < 0.05). In all cases, the differences resulted from a generalized increase in c-fos expression in the SNI groups (Figure 3B). Post hoc tests revealed statistically significant differences in c-fos⁺ cells density between the sham and SNI-L animals in the Cg cortex (P = 0.006) but not between sham and SNI-R (P = 0.160). In the ventral mPFC regions the inverse was observed, i.e., statistical significant differences were present between sham and SNI-R groups (PrL and IL, P < 0.001) but not in sham/SNI-L comparisons (PrL, P = 0.122; IL, P = 0.357). In the OFC, however, post hoc analyses revealed significant differences between the sham operated group and both SNI groups (VO, P_sham/SNI-L< 0.001 and P_sham/SNI-R = 0–01; LO, P_sham/SNI-L and P_sham/SNI-R < 0.05). Similar analyses in the primary motor cortex (M1) failed to reveal any difference between the experimental groups (F_2,43 = 0.178; N.S.). In any of the areas analyzed, neither the hemisphere side nor the interaction between the experimental group and side had a statistically significant effect. Within group comparisons of c-fos⁺ cell densities between left and right hemispheres also failed to reveal differences in any of the mPFC, OFC areas and the control M1 cortex.

OPEN IN VIEWER

Figure 3

c-fos expression in the prefrontal cortex. (A) Exemplificative c-fos stained sections, the respective Paxinos and Watson atlas diagrams 7] and respective distance to Bregma (mm) are given. Scale bar, 2 mm (100 μm inset). (B) SNI groups present a higher c-fos density when compared to sham controls in the mPFC and OFC but not in the control area M1. The hemisphere side was not a determining factor for the differences between the groups. (C) SNI-R contrary to sham/SNI-L presented a leftward shift in the laterality index in the VO following ASST completion. The functional integrity of this area is critical to Rev execution. *P<0.05, **P < 0.01 and ***P < 0.001. Data presented as mean ±S.E.M.

To further detail the interhemispheric expression of c-fos within each experimental subject after ASST, we calculated and compared the laterality index (Figure 3C). Comparisons between the three experimental groups revealed a statistical significant effect in the VO (F_2,21 = 4.892; P = 0.019) but not in any other area analyzed. Opposing lateralization biases between sham/SNI-L and SNI-R to the right and left hemispheres, respectively, accounted for much of this effect. Additionally, the analyses of M1 laterality index also provided no indication of lateralized effects (F_2,21 = 0.754; N.S.).

Experimental subjects

3 months old, male Wistar Han rats (Charles River Laboratories, Barcelona, Spain) were used in all experiments. Food availability was restricted to 1 h/day (7:00–8:00 p.m.) during the time frame of the attentional set-shifting task (ASST) training and execution; animals' weight was controlled to ensure that it was maintained above 10–15% of the initial weight. All procedures with animals were performed according to the guidelines of European Communities Council Directive 2010/63/EU.

Neuropathy model and pain assessment

The SNI model of peripheral neuropathy was used in this study [6]. The model was installed in the left (SNI-L; N = 7) or right (SNI-R; N = 8) hindpaw following previously described procedures [4]. A group of left/right sham-operated animals was also included (N = 7). At the third week post surgery, mechanical allodynia was assessed using the up-and-down method [22].

Attentional-set shifting task

At the 4^th week post surgery the ASST was initiated. The paradigm was performed in a rectangular arena. In the first third a wall separated two compartments each containing a bowl with sawdust. In the beginning of each trial, the wall separating the last third of the arena was lifted, allowing the animals to reach the bowls (Figure 2A). A successful trial was considered if the animal dug the correct bowl and retrieved the reward (cheerio cereal). In the compound discrimination (CD) tasks, 2 odors (aromatic oils rubbed in the bowls' rim) and 2 textures (placed close to the bowls) were simultaneously present. In the first CD task, odor was the relevant cue; this was followed by its reversal (Rev 1) i.e. the previously irrelevant odor became the signal for reward. A new pair of odors and textures was then introduced maintaining odor as the relevant dimension (intradimensional shift; IDS1); Rev 2 followed IDS1. The relevant dimension was then shifted from odor to texture (extradimensional shift; EDS) followed by Rev 3, IDS 2 and Rev 4. Tasks were terminated when animals preformed 6 successful trials consecutively. The organization of the trials is given in Table 1.

OPEN IN VIEWER

Table 1

Attentional set-shifting task trial organization and sequence

trial	1		2		3		4		5		6		7		8		9		10
bowl	L	R	L	R	L	R	L	R	L	R	L	R	L	R	L	R	L	R	L	R
S Dodo	O2	O1	O1	O2	O1	O2	O2	O1	O1	O2	O2	O1	O2	O1	O2	O1	O1	O2	O1	O2
SDtex	T2	T1	T1	T2	T1	T2	T2	T1	T1	T2	T1	T2	T1	T2	T1	T2	T2	T1	T2	T1
CD	O3	O4	O4	O3	O4	O3	O4	O3	O3	O4	O3	O4	O4	O3	O4	O3	O3	O4	O3	O4
	T3	T4	T3	T4	T4	T3	T4	T3	T4	T3	T3	T4	T3	T4	T3	T4	T3	T4	T4	T3
Rev1	O3	O4	O3	O4	O3	O4	O4	O3	O4	O3	O3	O4	O4	O3	O4	O3	O3	O4	O4	O3
	T4	T3	T4	T3	T4	T3	T4	T3	T3	T4	T3	T4	T3	T4	T4	T3	T3	T4	T3	T4
IDS1	O5	O6	O5	O6	O5	O6	O6	O5	O6	O5	O6	O5	O6	O5	O6	O5	O5	O6	O5	O6
	T5	T6	T5	T6	T6	T5	T5	T6	T6	T5	T5	T6	T5	T6	T6	T5	T6	T5	T6	T5
Rev2	O6	O5	O5	O6	O5	O6	O5	O6	O6	O5	O6	O5	O5	O6	O6	O5	O6	O5	O5	O6
	T5	T6	T6	T5	T6	T5	T5	T6	T6	T5	T5	T6	T5	T6	T6	T5	T6	T5	T6	T5
EDS	O7	O8	O8	O7	O7	O8	O7	O8	O7	O8	O8	O7	O7	O8	O8	O7	O8	O7	O8T7	O7
	T7	T8	T8	T7	T7	T8	T7	T8	T8	T7	T8	T7	T8	T7	T8	T7	T7	T8	T7	T8
Rev3	O8	O7	O8	O7	O7	O8	O7	O8	O7	O8	O8	O7	O8	O7	O8	O7	O7	O8	O7	O8
	T7	T8	T7	T8	T7	T8	T8	T7	T7	T8	T8	T7	T8	T7	T8	T7	T7	T8	T8	T7
IDS2	O9	O10	O9	O10	O10	O9	O9	O10	O9	O10	O10	O9	O10	O9	O9	O10	O10	O9	O10	O9
	T9	T10	T10	T9	T9	T10	T9	T10	T10	T9	T10	T9	T10	T9	T10	T9	T9	T10	T9	T10
Rev4	O9	O10	O10	O9	O9	O10	O9	O10	O9	O10	O10	O9	O10	O9	O9	O10	O9	O10	O10	O9
	T9	T10	T9	T10	T9	T10	T10	T9	T9	T10	T9	T10	T10	T9	T10	T9	T9	T10	T10	T9

Odors and textures are presented separately (SD_odo, SD_tex) or combined (CD, Rev 1–4, IDS 1/2, EDS) in simple and compound discrimination steps, respectively. If at the 10^th trial the learning criteria were not successfully achieved, the sequence was restarted. Relevant dimension on each trial are bold labeled. Odors: O₁-strawberry; O₂-camellia; O₃-lemon; O₄-papaya; O₅-rose; O₆-vanilla; O₇-lotus; O₈-cinnamon; O₉-mango; O₁₀-eucalyptus; Textures: T₁-styrofoam; T₂-sponge; T₃-sandpaper; T₄-scourer; T₅-sponge cloth 1; T₆- sponge cloth 2; T₇-cardboard; T₈- velvet; T₉-carpet; T₁₀-foam. L-left compartment; R-right compartment.

Tissue collection and c-fos immunohistochemistry

90 minutes after ASST termination animals were deeply anaesthetized and perfused intracardially with PBS followed by 4% paraformaldehyde. Brains were then removed and 50 μm coronal sections were obtained for immunoreaction. Brain sections were then incubated (overnight; RT) with a 1:2000 anti-c-fos serum (PC38 Rabbit; Calbiochem®, Nottingham, UK). After incubation with a 1:200 biotinylated secondary antibody (polyclonal swine anti-rabbit; E0353, Dako, Glostrup, Denmark), immunoreactivity was visualized using a chromogen reaction (Vector Laboratories, Burlingame, USA).

Cell counting procedures

c-fos⁺ cells were counted using Visiopharm Integrator System software (version 2.123.0; Hoersholm, Denmark) and a motorized microscope (BX-51; Olympus) connected to a digital camera (U-TV1X-2; Olympus). Square probes (2500 μm²) were placed over the selected areas, evenly spaced, covering 20% of the total defined area. The random positioning of this grid by the software ensured an unbiased and effective sampling. Spheroid/ovoid brownish nuclei within the probes were considered to be c-fos⁺ and were counted. The experimenter performing the sampling was blind to the experimental condition of the subjects. The interhemispheric expression of c-fos was calculated by the laterality index defined as

L a t e r a l i t y i n d e x = d e n s i t y l e f t - d e n s i t y r i g h t d e n s i t y l e f t + d e n s i t y r i g h t