4 Discussion
Several studies on depression have compared PrL and IL based on differences in cytoarchitecture, function and connectivity(Fullana, Covelo et al., 2019; Garro-Martinez, Fullana et al., 2021; Liu, Ota et al., 2015). In this study, we found that ICV-STZ activates IDO in both PrL and IL, but its regulatory pathways of depressive behaviors in the PrL and IL are different, with a remarkable attenuation of astrocytes in the PrL and overactivation of microglia in the IL as shown in Figure 10. The alteration of Trp levels in the PrL and IL was different, which hinted at the regional specificity of Kyn pathways between two subregions. Although the reason is unknown, a possible explanation is that the up-regulation of Trp levels in the IL was a compensatory increase induced by neuroinflammation. Furthermore, the increase of pro-inflammatory cytokines was observed only in the IL but not PrL, reflecting the neuroinflammatory region specificity in the ICV-STZ induced depression.
There is substantial evidence of depression emerging from abnormalities of astrocytes and microglia(Tao, Yan et al., 2020; Yang, Zhou et al., 2020), and in the central nervous system, glia can respond positively to neuroinflammation via regulating the kynurenine pathway balance(Dezsi, Tuka et al., 2015; Garrison, Parrott et al., 2018). Although no differences were observed in the structural distribution of glial cells in the PrL and IL(Banqueri, Mendez et al., 2019; Gosselin, Gibney et al., 2009), they appear to play different roles in the mediation of depressive-like behaviors(Fullana, Covelo et al., 2019). The present study illustrated that ICV-STZ induced depressive behavior was paralleled by astrocytic defects in the PrL and microglia activation in the IL. In the PrL, astrocytes were shown deficit and dysfunction characterized by the decrease in astrocyte number, complexity of branches and levels of GLT-1, GLAST and GFAP, with a concomitant reduction in the neuroprotective metabolite KA. In the IL, microglia activation was characterized by increased microglia numbers, Iba1 levels, pro-inflammatory cytokines and neurotoxicity metabolite 3-HK with shortening of cellular processes and enlargement of the soma. The level of KA in the IL was also reduced despite there were no astrocyte deficits, which might be due to Kyn being more metabolized into 3-HK induced by ICV-STZ. However, we did not observe activation of microglia in the PrL and defect of astrocytes in the IL in ICV-STZ rats. All these dysregulations were restored by regional selective inhibition of IDO, suggesting that the antidepressant effects of 1-MT in the PrL might be through normalizing astrocyte deficit and dysfunction, while the antidepressant effects of 1-MT in the IL might be mediated, at least for a part, by local anti-inflammatory processes.
Studies in human and animals indicated that the central inflammatory response in depression may be associated with the reduction of total volume, the shrinkage of dendrites and loss of spines within the mPFC(Fan, Song et al., 2018; Meier, Drevets et al., 2016). Some studies have shown that under chronic or acute stress, the dendritic branches of neurons in the PrL(Moench & Wellman, 2017) and IL(Moench, Maroun et al., 2016) can be seen retraction. In the present study, the dendritic spine density and proportion of mushroom type in the PrL and IL were significantly decreased, and this effect was reversed by local treatment with 1-MT. The mushroom type was identified as the main type involved in the regulation of synaptic plasticity because it is more stable and has larger postsynaptic densities and contact areas than other types(Helm, Dankovich et al., 2021). Both neuroplasticity and BDNF deficiency in the PrL and IL suggested an additional mechanism implicated with IDO activation in ICV-STZ rats and the effects of 1-MT on basal dendrites. Noteworthy, enhanced thin type and filopodia type proportions distributed on the dendrites of PrL were observed in ICV-STZ rats, while in the IL, a larger proportion of stubby type was mainly detected. Although ICV-STZ differentially altered structural plasticity in the PrL and IL, these structural alterations were reversed by local treatment with 1-MT, suggesting that 1-MT ameliorates depressive behaviors involving the structural remodeling of dendritic spines.
The model of ICV-STZ was highly prevalent and has been used to resemble sAD in decades(Kalafatakis & Zarros, 2014). Over time, ICV-STZ rats showed typical pathological changes similar to sAD, such as the deposition of Aβ and the hyperphosphorylation of Tau in the brain, and started to develop learning and cognitive impairment on day 14(Song, Cui et al., 2018). According to our study, although there were no such pathological changes and learning and memory impairments on day 7, depressive-like behaviors appeared. Evidence has reported that some patients with sAD have depression in the early stage, and depression might be a prodromal symptom of sAD(Burke, Cadet et al., 2018; Gatchel, 2021). Therefore, it is necessary to further study whether this animal model can simulate the clinical processes of sAD. Although a clear explanation of the mechanism of sAD formation is still lacking, the ICV-STZ model might provide a valuable contribution to the field of sAD model research. Additionally, our data show for the first time that the regionally selective blockade of IDO by 1-MT within the mPFC, either PrL or IL, protected rats from ICV-STZ induced depressive behaviors but failed to produce an anxiolytic effect, suggesting that the anxiety-like behaviors induced by ICV-STZ might not be due to the IDO activation in the PrL and IL, which need further confirmation.