The current study provides the evidence that prenatal exposure to a bacterial endotoxin (LPS) during midgestation in rats leads to significant age-dependent changes in the postnatal development of mesolimbocortical DA system in the offspring. Specifically, LPS treated offspring showed lower levels of D2 receptor in the PFC at pre and post pubertal ages (P35 and P60 respectively) as assessed by ligand autoradiography. Analysis of PFC subregions reveals that the effect is limited to the IL/PL subregions of the mPFC as the animals grow to post-pubertal age. The decreased expression of D2R was also confirmed by immunohistochemical studies. D2 immunopositive cells were found to be decreased in IL/PL mPFC at P60 while the total number of mature neurons, indicated by NeuN immunolabelling, was not altered. We also found that the number of PV-positive cells in mPFC was not altered due to prenatal LPS treatment. However, there was a significant decrease in the number of PV neurons which co-express D2R. In the subcortical regions, the only significant finding was a decrease in the level of DAT binding in the Nacc of LPS exposed animals at pre pubertal age.
Our results on mPFC D2R in GD15/16 LPS-treated rats are consistent with data reported by Meyer et al who also observed decreased D2R immunoreactivity in the mPFC of adult mice challenged prenatally with poly (I:C) at GD9 
. However, these authors also reported a reduction in mPFC D1R immunoreactivity in poly (I:C) treated mice. Previous findings on the expression of D1 and D2 receptors in striatal and accumbal regions in maternal poly (I:C) model appear more complex and contradictory. For example, Ozawa et al using GD12/17 poly (I:C) treatment in mice reported decreased D2 in the striatum with no changes in D1 receptors 
. On the other hand, increased D1R in dorsal striatum and Nacc shell and increased D2R in Nacc shell has been reported in adult offspring of GD9 poly (I:C) treated mice 
. With respect to DAT expression, our data showing decreased DAT binding in Nacc core and shell in pre-pubertal rats is consistent with those reported in poly (I:C) mice model by Vuillermot et al who also found decreased DAT immunoreactivity in P35 dorsal striatum as well as Nacc core and shell 
. Thus, our data on maternal LPS-treated offspring suggest that while some DA-related long-term effects of maternal immune challenge may be specific to immune activating agents used, gestational time of insult or species, decreased levels of D2R in the PFC and DAT in the Nacc may be interesting common effects of maternal immune activations.
These alterations in D2 receptor and transporters could well be the substrate of DA-related behavioural changes reported in maternal infection models. For example, prenatal administration of LPS and poly (I:C) in rodents causes significant deficits in prepulse inhibition of acoustic startle and increases in amphetamine induced locomotion in the adult offspring 
. The significance of our D2R finding may be discussed in the context of the role of DA in mPFC-related cognitive functions such as working memory 
which is also found to be impaired in maternal infection models 
To assess the cellular sites of D2R reduction in the PFC of prenatal LPS-treated animals, our focus was on PV-expressing putative interneruons. It is known that D2R is expressed in both pyramidal as well as GABAergic interneurons 
and thus exerts direct and indirect GABA-mediated DA actions on pyramidal cells. Further, a widely-replicated postmortem finding in schizophrenia, i.e., reduction in GABA markers such as GAD67 occur in a subset of interneurons that express PV 
. In the current study, we did not find any change in the number of PV-immunopositive cells in the mPFC of LPS treated adult offspring, unlike for example in prenatal poly(I:C) model in mice where Meyer et al showed a reduction in the number of PV cells in the mPFC 
. Our finding was a bit unexpected as decreased number of parvalbumin neurons have previously been reported in other prenatal infection models, e.g., in the hippocampus of mice offspring exposed to poly I:C at GD9 
as well as rat offspring exposed poly I:C at GD15 
. To our knowledge, our study is the first one that specifically examined total number of PV interneurons in the prefrontal cortex of rat offspring exposed to LPS during mid gestation. However, it should be pointed out that there are several key differences in the methodology and experimental design of the studies reporting changes in PV neurons and our current experiments, e.g., the timing of prenatal immune activation (GD 15/16 in our experiments vs GD 9–17), type of immune activation (LPS vs polyI:C) and animal species (rat vs mouse). In addition, our analysis of mPFC included both IL and PL subregions whereas previous studies focused on changes primarily in the IL mPFC.
Despite no significant change in the total number of PV neurons, we however, found that the percentage of the PV cells which co-express D2R was significantly decreased in the mPFC of LPS offspring. This indicates that at least a part of D2R reduction that we observed using autoradiography and immunohistochemistry occurs in the PV containing interneurons. Despite making a relatively small percentage of neuronal population in the cerebral cortex, PV containing GABAergic interneurons represent a critical regulatory element in PFC physiology and are involved in several psychiatric disorders 
. PV expressing neurons are a major class of fast spiking interneurons that play a role in PFC gamma oscillatory activity and synchronize the activity of excitatory pyramidal neurons, essential for working memory 
. D2R enhances the excitability of these fast spiking, interneurons, that mature at late adolescence 
. Interestingly, a loss of D2 modulation of fast-spiking interneurons, presumably PV-interneurons, has also been reported in the PFC of neonatal ventral hippocampus-lesioned rat model of schizophrenia 
. Thus, our observation of a reduced D2R expression in PV cells may be another mechanism by which maternal infections affect adult PFC mediated cognitive functions.
In summary, our findings suggest that prenatal immune activation alters the development of mesolimbic and mesocortical dopaminergic system in an age specific manner. For example, decreased level of DAT expression in the nucleus accumbens is only observed at pre pubertal age in the LPS treated offspring; while at post pubertal time, we found significant decrease in D2 receptor expression in the PV interneurons of mPFC. Although extrapolation of findings from animal studies to clinical conditions must be made with caution, in the context of our data, we believe that clinical longitudinal studies on the trajectory of DA related brain changes in individuals at risk for neuropsychiatric disorders would be very informative for early detection and/or intervention purposes.