Abstract

In the prefrontal cortex of subjects with schizophrenia, markers of the synthesis and re-uptake of GABA appear to be selectively altered in a subset of interneurons that includes chandelier cells. Determining the effect of these disturbances in presynaptic GABA markers on inhibitory signaling requires knowledge of the status of GABAA receptors at the postsynaptic targets of chandelier cells, the axon initial segments (AIS) of pyramidal neurons. Because the α2 subunit of the GABAA receptor is preferentially localized at pyramidal neuron AIS, we quantified α2 subunit immunoreactive AIS in tissue sections containing prefrontal cortex area 46 from 14 matched triads of subjects with schizophrenia, subjects with major depression and control subjects. Systematic, random sampling revealed that the mean number of α2-labeled AIS per mm2 in subjects with schizophrenia was significantly (P = 0.007) increased by 113% compared to control subjects and non-significantly increased compared to subjects with major depression. Furthermore, within subjects with schizophrenia, the density of α2-labeled AIS was negatively correlated (r = –0.49, P = 0.038) with the density of chandelier axon terminals immunoreactive for the GABA membrane transporter. These data suggest that GABAA receptors are up-regulated at pyramidal neuron AIS in response to deficient GABA neuro-transmission at chandelier axon terminals in schizophrenia. Thus, disturbances in inhibition at the chandelier neuron–pyramidal neuron synapse may be a critical component of prefrontal cortical dysfunction in schizophrenia.

Introduction

The most persistent and debilitating features of schizophrenia involve disturbances in cognitive abilities, such as working memory, which appear to reflect dysfunction of the dorsal prefrontal cortex (PFC) (Weinberger et al., 1986; Goldman-Rakic, 1994). This dysfunction may be associated with alterations in the inhibitory circuitry of the PFC. For example, in post-mortem studies of schizophrenia, genes responsible for the synthesis of GABA, glutamic acid decarboxylase (GAD67), and for the re-uptake of GABA into the nerve terminal, GABA membrane transporter (GAT-1), show decreased mRNA expression in the PFC and these alterations are restricted to a subset of inhibitory neurons (Akbarian et al., 1995a; Guidotti et al., 2000; Volk et al., 2000, 2001). The affected subset of GABA neurons appears to include chandelier cells, since decreased expression of GAT-1 mRNA is associated with decreased GAT-1 protein immunoreactivity selectively in the axon terminals of chandelier neurons (Woo et al., 1998; Pierri et al., 1999) (Fig. 1E).

The axons of chandelier cells form vertical arrays of terminals, termed ‘cartridges’ (Lewis and Lund, 1990), which synapse at the axon initial segments (AIS) of pyramidal neurons (Somogyi, 1977; Freund et al., 1983; Peters, 1984). The localization of these synapses near the site of action potential generation suggests that chandelier cells are positioned to regulate powerfully the output of pyramidal neurons. However, it remains to be determined whether a reduction in GAT-1 in chandelier axon cartridges is actually associated with altered inhibitory neurotransmission at the AIS of pyramidal neurons in schizophrenia.

The status of GABA activity at the AIS of pyramidal neurons may be further informed by an analysis of postsynaptic GABAA receptors at the AIS. Indeed, compensatory changes in GABAA receptors in response to alterations in extracellular GABA levels have been previously reported (Montpied et al., 1991a; Mhatre and Ticku, 1994). However, previous radioligand binding and mRNA expression studies of PFC GABAA receptors in schizophrenia (Hanada et al., 1987; Akbarian et al., 1995b; Benes et al., 1996; Huntsman et al., 1998; Impagnatiello et al., 1998; Dean et al., 1999; Ohnuma et al., 1999) have not selectively analyzed the pyramidal neuron AIS.

GABAA receptors are composed of pentamers of subunits, most commonly including at least one of each of the α, β and γ subunit classes (Sieghart et al., 1999). In addition, the different GABAA receptor α subunits have distinctive subcellular distributions (Fritschy and Mohler, 1995). For example, in the superficial layers of human cerebral cortex, the α2 subunit is prominently localized at pyramidal neuron AIS (Loup et al., 1998). Indeed, although associated with only ∼15% of all GABAA receptors in the cortex (Fritschy and Mohler, 1995), the α2 subunit is found at >80% of inhibitory synapses onto pyramidal neuron AIS, at least in rat hippocampus (Nusser et al., 1996; Nyíri et al., 2001). Furthermore, only 0.1% of all GABA synapses are found at pyramidal AIS, whereas nearly 25% of all α2- immunoreactive synapses are found at pyramidal AIS. Finally, GABAA receptors containing the α2 subunit have a higher affinity for GABA and faster activation and slower de-activation times, compared to GABAA receptors containing the more commonly expressed α1 subunit (Levitan et al., 1988; Lavoie et al., 1997). Thus, GABAA receptors containing the α2 subunit appear to be anatomically positioned and functionally adapted to mediate a potent inhibitory influence on the output of pyramidal neurons and, thus, may provide critical insight into the status of inhibitory activity at pyramidal neuron AIS.

Therefore, we examined human post-mortem brain tissue containing PFC area 46 in order to determine: (i) whether the density of pyramidal neuron AIS immunoreactive for the GABAA receptor α2 subunit (a2-AIS) is altered in subjects with schizophrenia; (ii) whether changes in a2-AIS density are specific to the diagnosis of schizophrenia; and (iii) whether a2-AIS density is related to presynaptic GABA markers in chandelier axon cartridges in subjects with schizophrenia.

Materials and Methods

Characteristics of Study Subjects

Brain specimens were obtained during autopsies conducted at the Allegheny County Coroner’s Office after obtaining consent from the surviving next of kin. The 14 triads of subjects used in this study each consisted of one subject with schizophrenia matched to one control subject and one subject with major depressive disorder (MDD) for sex, and as closely as possible for age and post-mortem interval (PMI; Table 1). All subjects with schizophrenia, all control subjects and all but five subjects with MDD (613, 689, 693, 698 and 803; Table 1) were included in previous studies of chandelier axon cartridges immunoreactive for GAT-1 — ‘GAT-1-cartridges’ (Woo et al., 1998; Pierri et al., 1999). All subjects were under the age of 70 years and all subjects with schizophrenia were previously determined to have at least a 10% decrease in the density of GAT-1-cartridges compared to matched control subjects.

An independent panel of experienced clinicians arrived at consensus DSM-III-R diagnoses after reviewing medical records and/or the results of structured interviews conducted following written, informed consent with family members of the deceased (Glantz and Lewis, 1997). No psychiatric disorders were present in the control subjects. Six schizophrenic subjects and nine subjects with MDD had a history of an alcohol and/or substance abuse disorder (Table 1). Four schizophrenic subjects had been off antipsychotic medications prior to death (234, neurolepticnaïve; 450, discontinued for unknown length of time; 537, 10 months; 622, 7 months; Table 1) and two subjects with MDD were on antipsychotic medications at time of death. Three schizophrenic subjects and four subjects with MDD were receiving benzodiazepines at time of death. The mean (±SD) ages of the subjects with schizophrenia and MDD at the onset of illness were 29.7 ± 9.4 and 41.6 ± 10.7 years, respectively, and the average durations of illness were 23.3 ± 8.4 and 11.7 ± 9.3 years, respectively. All procedures were approved by the University of Pittsburgh’s Institutional Review Board for Biomedical Research.

Tissue Processing

Upon retrieval, brain specimens were blocked coronally, placed into 4% paraformaldehyde for 48 h, cryoprotected and then stored at –30°C. Subject groups did not significantly differ in mean storage time [F(2,42) = 2.36, P = 0.11; Table 1]. Blocks containing the left middle frontal gyrus were sectioned coronally at 40 μm on a calibrated cryostat. Nissl-stained sections were used to identify PFC area 46 in each subject using cytoarchitectonic criteria (Daviss and Lewis, 1995; Rajkowska and Goldman-Rakic, 1995). Neuropathological examination of each brain revealed no abnormalities in the region of interest and Alzheimer’s disease was ruled out in each subject using clinical and neuropathological criteria.

Four tissue sections containing PFC area 46, in serial order and ∼400 mm apart, from each subject were processed in a randomized block design (i.e. with one section from each subject in a triad always processed together, and with different combinations of triads in each run). Using an antibody raised in guinea pig against the cDNA-derived N-terminal sequence (amino acid residues 1–9) of the human GABAA receptor α2 subunit (Marksitzer et al., 1993), sections were processed using the avidin–biotin procedure, followed by diaminobenzidine and hydrogen peroxide as previously described (Woo et al., 1998). Reaction product was intensified through serial immersions in aqueous osmium tetroxide and thiocarbohydrizide, followed by silver nitrate and gold chloride (Pucak et al., 1996).

The specificity of the antibody for the GABAA receptor α2 subunit was previously demonstrated by Western blot experiments whereby immunoprecipitation of GABAA receptor α2 subunit by the antibody revealed a single band whose signal intensity was reduced in a dose-dependent manner following preadsorption with increasing concentrations of GABAA receptor α2 subunit peptide (Marksitzer et al., 1993). Specificity was also confirmed by anatomical observations described in the Results section.

Quantification

All quantification was conducted by one rater (D.W.V.), who was blinded to diagnosis and subject number. Using the Stereo Investigator fractionator program (MicroBrightField Inc., Colchester, VT), a region of the middle frontal gyrus containing area 46 and of uniform cortical depth was sampled in each subject. Using a 5× objective, a contour containing layers 2–3a, defined as 25% of the total cortical width immediately below the layer 1–2 border, was drawn on each section (mean ± SD total contour area per subject: 10.7 ± 2.6 mm2). This zone was chosen because it contains ∼80% of the total number of α2-AIS in the PFC and in order to maintain consistency with a previous study on the density of GAT-1-cartridges in this laminar location in the same cohort of subjects (Pierri et al., 1999). Between 40 and 50 counting frames, 70 × 70 μm, were systematically and randomly placed within the contour for each tissue section and two sides of each counting frame were identified as exclusion boundaries. Using a 100× oil-immersion objective (NA 1.4), all a2-AIS (see Results for criteria) within each counting frame were identified on a video monitor at a final magnification of 2450×. The total number of a2-AIS counted in each subject ranged from 4 to 616 and the mean (±SD) coefficient of error for a2-AIS counts per sampling frame in each subject was 0.13 ± 0.08. Assessments of intra-rater reliability in identifying a2-AIS throughout the course of the quantification procedure revealed an intraclass correlation coefficient of 0.99 (95% CI = 0.59–1.0).

In a separate analysis using the Neurolucida program (MicroBright-Field Inc., Colchester, VT), mean length of a­2-AIS was also determined in each schizophrenic and control subject. In one random tissue section from each subject, the length of every detectable a2-AIS was measured in a series of 100 mm wide cortical traverses. The average (±SD) number of a2-AIS measured in each subject was 54 ± 21.

Statistical Analysis

Measures of the density of a2-AIS per mm2 in each of the four tissue sections for each subject were treated as four correlated observations. A multivariate analysis of covariance (MANCOVA) model assuming a compound symmetric covariance structure (Neter et al., 1996) was employed to test for a main effect of diagnosis, with triad and immuno-histochemistry run included as blocking factors and tissue storage time as a covariate. In this model, the inclusion of triad accounted for matching of subjects for sex, age and PMI. Because regression analyses revealed a potential effect of PMI on α2-AIS density, two additional MANCOVA models were run. One model included both diagnosis and triad factors, with PMI and tissue storage time as covariates and run as a blocking factor; the other model included a diagnosis factor with sex, age, PMI and tissue storage time as covariates, and run as a blocking factor. As all three models yielded similar results, only the results of the primary model are reported. Comparisons between any two of the diagnostic groups were also conducted using the primary model. In addition, the effects of sex, psychotropic medication at time of death and/or alcoholism on a2-AIS density in subjects with schizophrenia were assessed using ANCOVAs with the difference in a2-AIS density within matched pairs of schizophrenic and control subjects as the dependent variable and PMI as a covariate. Finally, a one-tailed Pearson correlation was used to test the hypothesis that GAT-1-cartridge density was inversely related to a2-AIS density in subjects with schizophrenia. Analyses were implemented in SAS PROC Mixed (Littell et al., 1996) and all statistical tests were conducted with α = 0.05.

Results

Immunoreactivity for the GABAA Receptor α2 Subunit in Human PFC

As shown in Figure 1A, immunoreactivity for the GABAA receptor α2 subunit in human PFC was greatest in the supragranular layers, particularly in layer 2. This pattern directly parallels previous reports of a2-immunoreactivity in the medial PFC of rat (Dunn et al., 1996) and of α2 mRNA expression in human PFC (Akbarian et al., 1995b). At higher magnification, α2 subunit-immunoreactivity was prominently found in pyramidal neuron AIS — a2-AIS (Nusser et al., 1996; Loup et al., 1998; Nyíri et al., 2001) — identified as intensely immunoreactive, discrete, non-branching, vertically oriented processes located below unlabeled cell bodies, ranging in length from 8 to 40 μm and usually slightly tapered from superficial to deep (Figure 1BD).

α2-AIS Density in Schizophrenia and Major Depression

Statistical analysis using the primary MANCOVA model revealed a significant main effect of diagnosis [F(2,25) = 4.88, P = 0.016] on α2-AIS density. The mean (±SD) number of a2-AIS per mm2 was significantly [F(1,12) = 10.60, P = 0.007] increased by 113% in subjects with schizophrenia (302 ± 184) compared to control subjects (141 ± 119). Furthermore, in 12 of the 14 matched pairs of schizophrenic and control subjects, α2-AIS density was higher in the schizophrenic subject (Fig. 2). As illustrated in Figure 3, the mean increase in a2-AIS density in individual subjects with schizophrenia compared to their matched controls did not significantly differ when the subjects with schizophrenia were subdivided into groups based on gender [F(1,11) = 0.007, P = 0.934], treatment with psychotropic medications at time of death [F(1,11) = 0.201, P = 0.663], or history of an alcohol-related disorder [F(1,11) = 0.002, P = 0.968]. In addition, mean α2-AIS density was virtually identical in the three subjects with schizoaffective disorder (295 ± 203) and the 11 subjects with ‘pure’ schizophrenia (304 ± 189; Fig. 4).

In contrast to the group differences in a2-AIS density, mean a2-AIS length (in mm) was similar in both schizophrenic and control subjects (14.8 ± 5.6 and 14.7 ± 5.2, respectively). However, a2-AIS length was positively correlated with a2-AIS density in subjects with schizophrenia (r = 0.548, P = 0.042), but not in control subjects (r = 0.124, P = 0.674).

The mean density of a2-AIS was also increased in the subjects with schizophrenia by 37% compared to subjects with MDD (Fig. 2). Although this difference was not statistically significant [F(1,12) = 1.68, P = 0.22], 10 of the 14 subjects with schizophrenia had an increased density of α2-AIS compared to the matched subject with MDD. Furthermore, one subject with MDD (689; Table 1) appeared to be an outlier in that the density of a2-AIS in this subject (a2-AIS/mm2 = 731) was 2.6 SD above the mean α2-AIS density for all subjects with MDD (Fig. 2). Consequently, this outlier subject was included in the main statistical analysis, but excluded from subsequent analyses in which diagnosis groups were subdivided into smaller groups on the basis of demographic variables. Importantly, no demographic variable, such as psychotropic medication at time of death, history of alcoholism, presence of psychotic features, tissue storage time, or PMI, appeared to contribute to the increased a2-AIS density in this subject, or any other subject. When triad 9, which included subject 689, was excluded from analysis, the mean density of a2-AIS in subjects with schizophrenia was increased by 59% compared to subjects with MDD [182 ± 138, F(1,11) = 2.86, P = 0.12]. Furthermore, mean α2-AIS density did not significantly differ between control subjects and subjects with MDD [F(1,11) = 0.53, P = 0.48]. Finally, mean α2-AIS density in the four subjects with MDD with psychotic features did not differ from either the MDD subjects without psychotic features or the control subjects (Fig. 4).

Reciprocal Changes in a2-AIS and GAT-1-Cartridge Density in Schizophrenia

The density of GAT-1-cartridges in area 46 was previously determined for all subjects with schizophrenia, all control subjects and nine subjects with MDD examined in the present study (Woo et al., 1998; Pierri et al., 1999). Whereas the mean α2-AIS density in the subjects with schizophrenia was approximately twice that in the matched control subjects, the mean GAT-1-cartridge density in the same schizophrenic subjects was less than half that in the matched control subjects (Fig. 5). Furthermore, a significant inverse relationship (r = –0.49, P = 0.038) between the densities of α2-AIS and GAT-1-cartridges was present in the subjects with schizophrenia. In contrast, in the nine subjects with MDD included in the previous GAT-1-cartridge study, neither the mean density of a2-AIS nor of GAT-1-cartridges differed significantly from the nine matched control subjects (Fig. 5).

Discussion

In this study, a2-AIS density in the superficial PFC layers of subjects with schizophrenia was significantly increased compared to control subjects and was also increased, though non-significantly, compared to subjects with MDD. Furthermore, a2-AIS density was inversely related to GAT-1-cartridge density in subjects with schizophrenia. These data suggest that presynaptic alterations in chandelier neurons are associated with post-synaptic changes in GABAA receptors at the AIS of pyramidal neurons in schizophrenia.

Specificity of Findings for Schizophrenia

The increase in a2-AIS density appears to be relatively specific to the diagnosis of schizophrenia, or at least not a common feature of MDD, even when that disorder is accompanied by psychosis. Consistent with these findings, previous studies of the frontal cortex in subjects with MDD have failed to find differences in the concentration, synthesis, or re-uptake of GABA (Cheetham et al., 1988; Korpi et al., 1988; Sundman et al., 1997). Thus, alterations in PFC GABA neurotransmission at the chandelier cell–pyramidal neuron synapse appear to be a characteristic of schizophrenia that is not associated with depression or psychosis, per se.

Several lines of evidence suggest that treatment with antipsychotic medications and benzodiazepines or a history of alcoholism do not account for the changes in PFC GABA markers in subjects with schizophrenia. First, a2-AIS density was increased in subjects with schizophrenia, regardless of the use of psychotropic medications at time of death or history of an alcohol-related disorder. Consistent with these observations, studies of haloperidol-treated monkeys revealed no changes in the expression of GAD67 or GAT-1 mRNAs or the density of GAT-1-cartridges in the PFC (Pierri et al., 1999; Volk et al., 2000, 2001). In addition, pharmacological manipulations in rats have found that: (i) antipsychotic medications result in decreased muscimol binding to GABAA receptors (Johnson et al., 1994; Farnbach-Pralong et al., 1998); (ii) benzodiazepine treatment causes no changes in cortical GABAA receptor α2 subunit mRNA or protein expression levels (O’Donovan et al., 1992; Holt et al., 1996, 1997; Impagnatiello et al., 1996; Chen et al., 1999; Tietz et al., 1999); and (iii) ethanol results in either decreases or no change in cortical GABAA receptor α2 mRNA expression (Montpied et al., 1991b; Mhatre et al., 1993; Chen et al., 1998)

Methodological Considerations

The stereological principle of systematic, random sampling was employed in this study to reduce sampling bias; however, the lack of clear boundaries of PFC area 46 precluded the determination of the total number of a2-AIS in this region. Importantly, measures of a2-AIS density were not confounded by differences in a2-AIS length between the schizophrenic and control groups. Furthermore, given that a2-AIS and GAT-1-cartridge densities were changed in opposite directions in nearby tissue sections from the same subjects with schizophrenia, neither the method of quantification nor potential differences in cortical volume across subject groups appear to have introduced a systematic confound.

We interpret the greater density of a2-AIS in schizophrenia as reflecting an increase in the amount of a2 subunit and the total density of GABAA receptors, per AIS, resulting in a greater number of AIS with detectable levels of immunoreactivity. Unfortunately, direct quantification of GABAA receptor subunits in pyramidal neuron AIS would require the use of procedures — for example immunogold-labeling and electron microscopy or immunofluorescence at the light microscopic level (Sutoo et al., 1998) — that are technically challenging in post-mortem human brain. However, it seems unlikely that the increased α2-AIS density reflects an increase in pyramidal neuron density. First, in previous studies, PFC neuron density was reported to be either unchanged (Akbarian et al., 1995a), or, at ∼20%, not increased enough (Selemon et al., 1995, 1998) in schizophrenia to account for the present findings. Second, in the same subjects examined in the present study, the density of GAT-1-cartridges was actually reduced by >50%, which further argues against an increase in pyramidal neuron density. Finally, re-analysis of data from a previous study (Pierri et al., 2001) revealed no difference in deep layer 3 pyramidal neuron density between the same cohort of subjects with schizophrenia (48.4 ± 7.7 cells/0.001 mm3) and control subjects (49.1 ± 8.7 cells/0.001 mm3) examined in the present study.

Alternatively, an increased amount of a2 subunit at the AIS could be coupled with reductions in other subunits reported to be located at pyramidal neuron AIS, such as the a1 and a3 subunits (Nusser et al., 1996; Loup et al., 1998), without changes in the total number of GABAA receptors. Such ‘subunit switching’ between the α2 and a1 subunits occurs in rat hypothalamic oxytocin neurons during pregnancy (Brussaard and Herbison, 2000). Unfortunately, quantification of a1-labeled AIS in schizophrenia is not possible, since the much greater density of a1 subunit at other locations (Nusser et al., 1996) prohibits the resolution of α1-labeled AIS by light microscopy. In addition, we were not able to detect a3-immunoreactivity in pyramidal neuron AIS in our human tissue samples. Yet, even if ‘subunit switching’, without a change in total receptor number, occurs in schizophrenia, an increased proportion of GABAA receptors containing the α2 subunit would still likely confer a greater inhibitory response to GABA at pyramidal neuron AIS (Levitan et al., 1988; Lavoie et al., 1997). Thus, the increase in α2-AIS density, whether reflecting an increased total density of GABAA receptors or an increased proportion of GABAA receptors containing the α2 subunit, suggests that postsynaptic GABAA receptors at pyramidal neuron AIS are functionally up-regulated in schizophrenia.

Pathophysiological Significance

The inverse relationship between the densities of a2-AIS and GAT-1-cartridges in the same subjects with schizophrenia suggests that GABAA receptors are up-regulated at pyramidal neuron AIS in response to deficient GABA activity in chandelier axon terminals in schizophrenia (Fig. 6). Although experimental models of a selective decrease in GABA release at chandelier axon terminals are not available, previous studies (Mhatre and Ticku, 1994) have demonstrated that GABAA receptor antagonists produce increased α2 subunit mRNA expression in embryonic chick neurons. Furthermore, in subjects with temporal lobe epilepsy, a loss of GABA neurons is associated with a substantial increase in a2 subunit immunoreactivity in pyramidal cells (Loup et al., 2000).

The mRNA expression for presynaptic inhibitory markers, including GAD67 and GAT-1, is decreased in the PFC of subjects with schizophrenia (Akbarian et al., 1995a; Ohnuma et al., 1999; Guidotti et al., 2000; Volk et al., 2000, 2001). In contrast, mRNA levels of the major postsynaptic GABAA receptor subunits appear to be unchanged in schizophrenia (Akbarian et al., 1995b), although some studies have reported increases in the α1 subunit (Impagnatiello et al., 1998; Ohnuma et al., 1999) and a reduction in the short isoform of the γ2 subunit (Huntsman et al., 1998). Of particular interest, α2 subunit mRNA expression was unchanged in PFC area 46 of subjects with schizophrenia (Akbarian et al., 1995b). Although other explanations are possible, the increase in α2-AIS density may reflect a locally regulated increase in a2 subunit at pyramidal neuron AIS. However, alterations in a2-immunoreactivity at locations other than the AIS, such as pyramidal neuron soma (Nyíri et al., 2001), cannot be excluded. Furthermore, radiolabeled ligand binding studies of PFC GABAA receptors have revealed increases in muscimol binding (Hanada et al., 1987; Benes et al., 1996; Dean et al., 1999) that were most prominent at pyramidal neuron soma (Benes et al., 1996). Together, these data suggest that GABAA receptors located at the soma and AIS of pyramidal neurons are locally up-regulated in schizophrenia in response to a reduction in proximal inhibitory input.

The pathophysiological mechanism that selectively initiates disturbances in a subset of GABA neurons that appears to include chandelier neurons, but not in the majority of PFC GABA neurons (Woo et al., 1998; Volk et al., 2000, 2001), remains unclear. One possibility is that reduced inhibitory neuro-transmission in chandelier neurons reflects an abnormality intrinsic to this cell class. Alternatively, inhibitory neurotransmission may be reduced in chandelier neurons in response to an alteration in their excitatory inputs. For example, morphological alterations in layer 3 pyramidal neurons in PFC area 46 (Garey et al., 1998; Rajkowska et al., 1998; Glantz and Lewis, 2000; Pierri et al., 2001) and reduced neuronal number in the mediodorsal thalamic nucleus (Pakkenberg, 1990; Popken et al., 2000; Young et al., 2000; Byne et al., 2002) have been reported in subjects with schizophrenia. Interestingly, in monkey PFC area 46, the dendrites of parvalbumin-containing cells, which include chandelier neurons, receive synaptic inputs from local pyramidal neurons (Melchitzky et al., 2001) and from the mediodorsal thalamic nucleus (Melchitzky et al., 1999), whereas other classes of GABA cells do not (Melchitzky and Lewis, 2000). Further studies of the mechanism(s) responsible for altered chandelier-cell–pyramidal-neuron inhibitory neurotransmission may lead to novel therapeutic strategies for remediating prefrontal cortical dysfunction in schizophrenia.

Table 1

Control subjects, subjects with schizophrenia and subjects with MDD used in this study

Triad Control subjects Subjects with schizophrenia Subjects with major depression (MDD) 
 Case Sex/age PMIa Storage timeb Cause of deathc Case Diagnoses and medications Sex/age PMIa Storage timeb Cause of deathc Case Sex/age PMIa Storage timeb Cause of deathc  
aPMI indicates post-mortem interval in hours. bStorage time is in months. cASCVD, atherosclerotic coronary vascular disease. dAntipsychotic medication current at time of death. eBenzodiazepine medication current at time of death. fAlcohol dependence, current at time of death. gAlcohol dependence, in remission at time of death. hAlcohol abuse, current at time of death. iAlcohol abuse, in remission at time of death. jOther substance dependence, current at time of death. kOther substance dependence, in remission at time of death. lOther substance abuse, current at time of death. +P, presence of psychotic features in subjects with MDD. 
    1 250 F/47 5.3 110 ASCVD 398 Schizoaffective disorderd,e F/41 10.3 88 Pulmonary embolism 210 MDD F/50 4.7 117 Suicide by drowning 
    2 270 M/62 3.3 107 ASCVD 131 Chronic undifferentiated schizophreniad,i M/62 3.9 134 Pneumonia 249 MDD+P M/57 4.3 111 ASCVD 
    3 451 M/48 12.0 70 ASCVD 317 Chronic undifferentiated schizophreniad M/48 8.3 101 Pneumonia 511 MDD M/43 17.9 56 ASCVD 
    4 178 M/48 7.8 121 ASCVD 377 Chronic undifferentiated schizophreniad,f M/52 10.0 92 GI bleeding 505 MDDf M/57 12.8 57 Suicide by gun shot 
    5 420 F/67 19.5 81 Accidental CO poisoning 333 Chronic undifferentiated schizophreniad F/66 17.9 99 ASCVD 803 MDD F/65 18.0 11 Trauma 
    6 344 M/50 6.8 98 ASCVD 422 Chronic paranoid schizophreniad,e M/54 11.0 81 ASCVD 698 MDD+Pd M/59 13.0 29 Suicide by hanging 
    7 449 F/47 4.3 70 Accidental CO poisoning 517 Chronic disorganized schizophreniad,f F/48 3.7 56 Intracerebral hemorrhage 248 MDDe,l F/48 6.3 111 Suicide by hanging 
    8 412 M/42 14.2 84 Aortic stenosis 466 Chronic undifferentiated schizophreniad M/48 19.0 68 ASCVD 421 MDDi M/44 16.0 81 ASCVD 
    9 592 M/41 22.1 44 ASCVD 450 Chronic undifferentiated schizophreniag,k M/48 22.0 70 Suicide by jumping 689 MDD+Pd,e,i M/45 24.4 30 Suicide by acid ingestion 
10 681 M/51 11.6 32 Cardiomyopathy 234 Chronic paranoid schizophrenia M/51 12.8 113 Cardiomyopathy 602 MDDf M/56 11.8 43 Suicide by gun shot 
11 567 F/46 15.0 48 Mitral valve prolapse 537 Schizoaffective disorder F/37 14.5 53 Suicide by hanging 693 MDD+Pe,j F/42 12.6 30 Suicide by propoxyphene overdose 
12 568 F/60 9.5 48 ASCVD 559 Schizoaffective disorderd,f F/61 16.8 50 ASCVD 565 MDDh,k F/62 12.4 49 Suicide by gun shot 
13 620 M/64 17.3 40 Accidental drowning 566 Chronic undifferentiated schizophreniad,e,i M/63 18.3 49 ASCVD 613 MDD+Pi M/59 15.6 41 Suicide by gun shot 
14 551 M/61 16.4 51 Cardiac tamponade 622 Chronic undifferentiated schizophrenia M/58 18.9 39 Right MCA infarction 619 MDDe,k M/55 18.8 40 Suicide by gun shot 
Mean  52 11.8 72    53 13.4 78    53 13.7 58  
SD  5.9 29    5.8 28    5.7 34  
Triad Control subjects Subjects with schizophrenia Subjects with major depression (MDD) 
 Case Sex/age PMIa Storage timeb Cause of deathc Case Diagnoses and medications Sex/age PMIa Storage timeb Cause of deathc Case Sex/age PMIa Storage timeb Cause of deathc  
aPMI indicates post-mortem interval in hours. bStorage time is in months. cASCVD, atherosclerotic coronary vascular disease. dAntipsychotic medication current at time of death. eBenzodiazepine medication current at time of death. fAlcohol dependence, current at time of death. gAlcohol dependence, in remission at time of death. hAlcohol abuse, current at time of death. iAlcohol abuse, in remission at time of death. jOther substance dependence, current at time of death. kOther substance dependence, in remission at time of death. lOther substance abuse, current at time of death. +P, presence of psychotic features in subjects with MDD. 
    1 250 F/47 5.3 110 ASCVD 398 Schizoaffective disorderd,e F/41 10.3 88 Pulmonary embolism 210 MDD F/50 4.7 117 Suicide by drowning 
    2 270 M/62 3.3 107 ASCVD 131 Chronic undifferentiated schizophreniad,i M/62 3.9 134 Pneumonia 249 MDD+P M/57 4.3 111 ASCVD 
    3 451 M/48 12.0 70 ASCVD 317 Chronic undifferentiated schizophreniad M/48 8.3 101 Pneumonia 511 MDD M/43 17.9 56 ASCVD 
    4 178 M/48 7.8 121 ASCVD 377 Chronic undifferentiated schizophreniad,f M/52 10.0 92 GI bleeding 505 MDDf M/57 12.8 57 Suicide by gun shot 
    5 420 F/67 19.5 81 Accidental CO poisoning 333 Chronic undifferentiated schizophreniad F/66 17.9 99 ASCVD 803 MDD F/65 18.0 11 Trauma 
    6 344 M/50 6.8 98 ASCVD 422 Chronic paranoid schizophreniad,e M/54 11.0 81 ASCVD 698 MDD+Pd M/59 13.0 29 Suicide by hanging 
    7 449 F/47 4.3 70 Accidental CO poisoning 517 Chronic disorganized schizophreniad,f F/48 3.7 56 Intracerebral hemorrhage 248 MDDe,l F/48 6.3 111 Suicide by hanging 
    8 412 M/42 14.2 84 Aortic stenosis 466 Chronic undifferentiated schizophreniad M/48 19.0 68 ASCVD 421 MDDi M/44 16.0 81 ASCVD 
    9 592 M/41 22.1 44 ASCVD 450 Chronic undifferentiated schizophreniag,k M/48 22.0 70 Suicide by jumping 689 MDD+Pd,e,i M/45 24.4 30 Suicide by acid ingestion 
10 681 M/51 11.6 32 Cardiomyopathy 234 Chronic paranoid schizophrenia M/51 12.8 113 Cardiomyopathy 602 MDDf M/56 11.8 43 Suicide by gun shot 
11 567 F/46 15.0 48 Mitral valve prolapse 537 Schizoaffective disorder F/37 14.5 53 Suicide by hanging 693 MDD+Pe,j F/42 12.6 30 Suicide by propoxyphene overdose 
12 568 F/60 9.5 48 ASCVD 559 Schizoaffective disorderd,f F/61 16.8 50 ASCVD 565 MDDh,k F/62 12.4 49 Suicide by gun shot 
13 620 M/64 17.3 40 Accidental drowning 566 Chronic undifferentiated schizophreniad,e,i M/63 18.3 49 ASCVD 613 MDD+Pi M/59 15.6 41 Suicide by gun shot 
14 551 M/61 16.4 51 Cardiac tamponade 622 Chronic undifferentiated schizophrenia M/58 18.9 39 Right MCA infarction 619 MDDe,k M/55 18.8 40 Suicide by gun shot 
Mean  52 11.8 72    53 13.4 78    53 13.7 58  
SD  5.9 29    5.8 28    5.7 34  
Figure 1.

Immunoreactivity for the GABAA receptor a2 subunit in human PFC area 46. (A) Immunoreactivity for the a2 subunit is greatest in the superficial layers of the PFC, with the highest intensity in layer 2. The solid white lines indicate the zone sampled in this study and the dashed white line indicates the layer-6–white-matter border. Scale bar = 600 mm. (BD) Immunoreactivity for pre-and postsynaptic GABA markers at the axon initial segment of pyramidal neurons. In both control (B,C) and schizophrenic (D) subjects, axon initial segments (solid arrows) intensely labeled with an antibody against the GABAA receptor a2 subunit appear as vertically oriented processes below the unlabeled cell bodies of pyramidal neurons (P) and are occasionally slightly tapered from superficial to deep. (E) Chandelier neuron axon cartridge immunoreactive for GAT-1 (open arrow) outlines the unlabeled axon initial segment of a pyramidal neuron. Scale bar in (B) = 50 mm and in (C–E) = 20 μm.

Figure 1.

Immunoreactivity for the GABAA receptor a2 subunit in human PFC area 46. (A) Immunoreactivity for the a2 subunit is greatest in the superficial layers of the PFC, with the highest intensity in layer 2. The solid white lines indicate the zone sampled in this study and the dashed white line indicates the layer-6–white-matter border. Scale bar = 600 mm. (BD) Immunoreactivity for pre-and postsynaptic GABA markers at the axon initial segment of pyramidal neurons. In both control (B,C) and schizophrenic (D) subjects, axon initial segments (solid arrows) intensely labeled with an antibody against the GABAA receptor a2 subunit appear as vertically oriented processes below the unlabeled cell bodies of pyramidal neurons (P) and are occasionally slightly tapered from superficial to deep. (E) Chandelier neuron axon cartridge immunoreactive for GAT-1 (open arrow) outlines the unlabeled axon initial segment of a pyramidal neuron. Scale bar in (B) = 50 mm and in (C–E) = 20 μm.

Figure 2.

Scatterplots showing the number of a2-AIS per mm2 in each subject. Solid horizontal lines indicate the mean value for each group. Subjects in matched triads are indicated by the same color and connected by lines.

Figure 2.

Scatterplots showing the number of a2-AIS per mm2 in each subject. Solid horizontal lines indicate the mean value for each group. Subjects in matched triads are indicated by the same color and connected by lines.

Figure 3.

Bar graphs showing the mean (±SD) increase in number of a2-AIS per mm2 in individual schizophrenic subjects compared to matched control subjects when schizophrenic subjects are divided into groups based upon sex, psychotropic medication at time of death, or history of an alcohol-related disorder.

Figure 3.

Bar graphs showing the mean (±SD) increase in number of a2-AIS per mm2 in individual schizophrenic subjects compared to matched control subjects when schizophrenic subjects are divided into groups based upon sex, psychotropic medication at time of death, or history of an alcohol-related disorder.

Figure 4.

Bar graphs showing the mean (±SD) number of a2-AIS per mm2 in control subjects, subjects with major depressive disorder grouped by the absence (MDD) or presence (MDD + P) of psychotic features, subjects with schizoaffective disorder (SA) and subjects with ‘pure’ schizophrenia (SCH). *One subject with MDD, 689, was determined to be an outlier (see Results) and, consequently, was not included in this figure.

Figure 4.

Bar graphs showing the mean (±SD) number of a2-AIS per mm2 in control subjects, subjects with major depressive disorder grouped by the absence (MDD) or presence (MDD + P) of psychotic features, subjects with schizoaffective disorder (SA) and subjects with ‘pure’ schizophrenia (SCH). *One subject with MDD, 689, was determined to be an outlier (see Results) and, consequently, was not included in this figure.

Figure 5.

Bar graph showing the percentage change in the mean densities of a2-AIS (solid bars) and GAT-1-cartridges (cross-hatched bars) in schizophrenic subjects (left) and in subjects with MDD (right) compared to the matched control subjects. The percentage changes in a2-AIS and GAT-1-cartridge densities in MDD were calculated from the nine subjects with MDD and the nine matched control subjects for whom GAT-1-cartridge densities were available.

Figure 5.

Bar graph showing the percentage change in the mean densities of a2-AIS (solid bars) and GAT-1-cartridges (cross-hatched bars) in schizophrenic subjects (left) and in subjects with MDD (right) compared to the matched control subjects. The percentage changes in a2-AIS and GAT-1-cartridge densities in MDD were calculated from the nine subjects with MDD and the nine matched control subjects for whom GAT-1-cartridge densities were available.

Figure 6.

Alterations in GABA neurotransmission at the chandelier neuron–pyramidal neuron synapse in schizophrenia. In a subset of GABA neurons that appears to include chandelier neurons (pink), the mRNA expression for GAD67 and GAT-1 are both reduced below a detectable level (Volk et al., 2000, 2001) and the protein level of GAT-1 is selectively reduced in chandelier neuron axon terminals (Woo et al., 1998; Pierri et al., 1999), which synapse at the AIS of pyramidal neurons. In postsynaptic pyramidal neurons (blue), immunoreactivity for the GABAA receptor α2 subunit is increased at the axon initial segment (light blue), suggesting that GABAA receptors are up-regulated in response to a reduction in inhibitory neurotransmission in chandelier neuron axon terminals.

Figure 6.

Alterations in GABA neurotransmission at the chandelier neuron–pyramidal neuron synapse in schizophrenia. In a subset of GABA neurons that appears to include chandelier neurons (pink), the mRNA expression for GAD67 and GAT-1 are both reduced below a detectable level (Volk et al., 2000, 2001) and the protein level of GAT-1 is selectively reduced in chandelier neuron axon terminals (Woo et al., 1998; Pierri et al., 1999), which synapse at the AIS of pyramidal neurons. In postsynaptic pyramidal neurons (blue), immunoreactivity for the GABAA receptor α2 subunit is increased at the axon initial segment (light blue), suggesting that GABAA receptors are up-regulated in response to a reduction in inhibitory neurotransmission in chandelier neuron axon terminals.

These studies were supported by NIH grants MH 43784, MH 45156 and MH 64320 and by a Scottish Rite Schizophrenia Fellowship Award.

References

Akbarian S, Kim JJ, Potkin SG, Hagman JO, Tafazzoli A, Bunney WE Jr, Jones EG (
1995
) Gene expression for glutamic acid decarboxylase is reduced without loss of neurons in prefrontal cortex of schizophrenics.
Arch Gen Psychiatry
 
52
:
258
–266.
Akbarian S, Huntsman MS, Kim JJ, Tafazzoli A, Potkin SG, Bunney WE Jr, Jones EG (
1995
) GABAA receptor subunit gene expression in human prefrontal cortex: comparison of schizophrenics and controls.
Cereb Cortex
 
5
:
550
–560.
Benes FM, Snyder-Marie A, Vincent S, Khan Y (
1996
) Up-regulation of GABA-A receptor binding on neurons of the prefrontal cortex in schizophrenic subjects.
Neuroscience
 
75
:
1021
–1031.
Brussaard AB, Herbison AE (
2000
) Long-term plasticity of postsynaptic GABAA-receptor function in the adult brain: insights from the oxytocin neurone.
Trends Neurosci
 
23
:
190
–195.
Byne W, Buchsbaum MS, Mattiace LA, Hazlett EA, Kemether E, Elhakem SL, Purohit DP, Haroutunian V, Jones L (
2002
) Postmortem assessment of thalamic nuclear volumes in subjects with schizophrenia.
Am J Psychiatry
 
159
:
59
–65.
Cheetham SC, Crompton MR, Katona CL, Parker SJ, Horton RW (
1988
) Brain GABAA/benzodiazepine binding sites and glutamic acid decarboxylase activity in depressed suicide victims.
Brain Res
 
460
:
114
–123.
Chen F, Rezvani A, Jarrott B, Lawrence AJ (
1998
) Distribution of GABAA receptors in the limbic system of alcohol-preferring and non-preferring rats: in situ hybridisation histochemistry and receptor autoradiography.
Neurochem Int
 
32
:
143
–151.
Chen S, Huang X, Zeng XJ, Sieghart W, Tietz EI (
1999
) Benzodiazepine-mediated regulation of a1, a2, b1–3, and γ2 GABAA receptor subunit proteins in the rat brain hippocampus and cortex.
Neuroscience
 
93
:
33
–44.
Daviss SR, Lewis DA (
1995
) Local circuit neurons of the prefrontal cortex in schizophrenia: selective increase in the density of calbindin-immunoreactive neurons.
Psychiatry Res
 
59
:
81
–96.
Dean B, Hussain T, Hayes W, Scarr E, Kitsoulis S, Hill C, Opeskin K, Vopolov DL (
1999
) Changes in serotonin2A and GABAA receptors in schizophrenia: studies on the human dorsolateral prefrontal cortex.
J Neurochem
 
72
:
1593
–1599.
Dunn E, Fritschy J-M, Carter DB, Merchant KM (
1996
) Differential distribution of γ-aminobutyric acidA receptor subunit (α1, a2, a3, a5, and b2+3) immunoreactivity in the medial prefrontal cortex of the rat.
Neurosci Lett
 
210
:
213
–217.
Farnbach-Pralong D, Bradbury PR, Spurlock G, McGuffin P (
1998
) Clozapine and olanzapine treatment decreases rat cortical and limbic GABAA receptors.
Eur J Pharmacol
 
349
:
R7
–R8.
Freund TF, Martin KAC, Smith AD, Somogyi P (
1983
) Glutamate decarboxylase-immunoreactive terminals of Golgi-impregnated axoaxonic cells and of presumed basket cells in synaptic contact with pyramidal neurons of the cat’s visual cortex.
J Comp Neurol
 
221
:
263
–278.
Fritschy J-M, Mohler H (
1995
) GABAA-receptor heterogeneity in the adult rat brain: differential regional and cellular distribution of seven major subunits.
J Comp Neurol
 
359
:
154
–194.
Garey LJ, Ong WY, Patel TS, Kanani M, Davis A, Mortimer AM, Barnes TRE, Hirsch SR (
1998
) Reduced dendritic spine density on cerebral cortical pyramidal neurons in schizophrenia.
J Neurol Neurosurg Psychiatry
 
65
:
446
–453.
Glantz LA, Lewis DA (
1997
) Reduction of synaptophysin immuno-reactivity in the prefrontal cortex of subjects with schizophrenia: regional and diagnostic specificity.
Arch Gen Psychiatry
 
54
:
943
–952.
Glantz LA, Lewis DA (
2000
) Decreased dendritic spine density on prefrontal cortical pyramidal neurons in schizophrenia.
Arch Gen Psychiatry
 
57
:
65
–73.
Goldman-Rakic PS (
1994
) Working memory dysfunction in schizophrenia.
J Neuropsychiatry
 
6
:
348
–357.
Guidotti A, Auta J, Davis JM, Gerevini VD, Dwivedi Y, Grayson DR, Impagnatiello F, Pandey G, Pesold C, Sharma R, Uzunov D, Costa E (
2000
) Decrease in reelin and glutamic acid decarboxylase67 (GAD67) expression in schizophrenia and bipolar disorder.
Arch Gen Psychiatry
 
57
:
1061
–1069.
Hanada S, Mita T, Nishino N, Tanaka C (
1987
) [ 3H]Muscimol binding sites increased in autopsied brains of chronic schizophrenics.
Life Sci
 
40
:
239
–266.
Holt RA, Bateson AN, Martin IL (
1996
) Chronic treatment with diazepam or abecarnil differentially affects the expression of GABA A receptor subunit mRNAs in the rat cortex.
Neuropharmacology
 
35
:
1457
–1463.
Holt RA, Bateson AN, Martin IL (
1997
) Chronic zolpidem treatment alters GABAA receptor mRNA levels in rat cortex.
Eur J Pharmacol
 
329
:
129
–132.
Huntsman MM, Tran BV, Potkin SG, Bunney WE, Jones EG (
1998
) Altered ratios of alternatively spliced long and short gamma 2 subunit mRNAs of the gamma-amino butyrate type A receptor in prefrontal cortex of schizophrenics.
Proc Natl Acad Sci USA
 
95
:
15066
–15071.
Impagnatiello F, Pesold C, Longone P, Caruncho H, Fritschy J-M, Costa E, Guidotti A (
1996
) Modifications of γ-aminobutyric acidA receptor subunit expression in rat neocortex during tolerance to diazepam.
Mol Pharmacol
 
49
:
822
–831.
Impagnatiello F, Guidotti AR, Pesold C, Dwivedi Y, Caruncho H, Pisu MG, Uzunov DP, Smalheiser NR, Davis JM, Pandey GN, Pappas GD, Teuting P, Sharma RP, Costa E (
1998
) A decrease of reelin expression as a putative vulnerability factor in schizophrenia.
Proc Natl Acad Sci USA
 
95
:
15718
–15723.
Johnson AE, Liminga U, Liden A, Lindefors N, Gunne LM, Wiesel F-A (
1994
) Chronic treatment with a classical neuroleptic alters excitatory amino acid and GABAergic neurotransmission in specific regions of the rat brain.
Neuroscience
 
63
:
1003
–1020.
Korpi ER, Kleinman JE, Wyatt RJ (
1988
) GABA concentrations in forebrain areas of suicide victims.
Biol Psychiatry
 
23
:
109
–114.
Lavoie AM, Tingey JJ, Harrison NL, Pritchett DB, Twyman RE (
1997
) Activation and deactivation rates of recombinant GABAA receptor channels are dependent on α-subunit isoform.
Biophys J
 
73
:
2518
–2526.
Levitan ES, Schofield PR, Burt DR, Rhee LM, Wisden W, Kohler M, Fujita N, Rodriguez HF, Stephenson A, Darlison MG, Barnard EA, Seeburg PH (
1988
) Structural and functional basis for GABAA receptor heterogeneity.
Nature
 
335
:
76
–79.
Lewis DA, Lund JS (
1990
) Heterogeneity of chandelier neurons in monkey neocortex: corticotropin-releasing factor and parvalbumin immuno-reactive populations.
J Comp Neurol
 
293
:
599
–615.
Littell RC, Milliken GA, Stroup WW, Wolfinger RD (1996) SAS system for mixed models. Cary, NC: SAS Institute Inc.
Loup F, Weinmann O, Yonekawa Y, Aguzzi A, Wieser H-G, Fritschy J-M (
1998
) A highly sensitive immunofluorescence procedure for analyzing the subcellular distribution of GABAA receptor subunits in the human brain.
J Histochem Cytochem
 
46
:
1129
–1139.
Loup F, Wieser H-G, Yonekawa Y, Aguzzi A, Fritschy J-M (
2000
) Selective alterations in GABAA receptor subtypes in human temporal lobe epilepsy.
J Neurosci
 
20
:
5401
–5419.
Marksitzer R, Benke D, Fritschy J-M, Trzeciak A, Bannwarth W, Mohier H (
1993
) GABAA-receptors: drug binding profile and distribution of receptors containing the α2-subunit in situ.
J Receptor Res
 
13
:
467
–477.
Melchitzky DS, Sesack SR, Lewis DA (
1999
) Parvalbumin-immuno-reactive axon terminals in monkey and human prefrontal cortex: laminar, regional and target specificity of Type I and Type II synapses.
J Comp Neurol
 
408
:
11
–22.
Melchitzky DS, Lewis DA (
2000
) Intrinsic axon terminals of layer 3 pyramidal neurons in monkey prefrontal cortex: target selectivity for a subpopulation of GABA neurons.
Soc Neurosci Abstr
 
26
:
1236
.
Melchitzky DS, Gonzalez-Burgos G, Barrionuevo G, Lewis DA (
2001
) Synaptic targets of the intrinsic axon collaterals of supragranular pyramidal neurons in monkey prefrontal cortex.
J Comp Neurol
 
430
:
209
–221.
Mhatre MC, Ticku MK (
1994
) Chronic GABA treatment downregulates the GABAA receptor α2 and a3 subunit mRNAs as well as polypeptide expression in primary cultured cerebral cortical neurons.
Mol Brain Res
 
24
:
159
–165.
Mhatre MC, Pena G, Sieghart W, Tietz EI (
1993
) Antibodies specific for GABAA receptor α subunits reveal that chronic alcohol treatment down-regulates α-subunit expression in rat brain regions.
J Neurochem
 
61
:
1620
–1625.
Montpied P, Ginns EI, Martin BM, Roca D, Farb DH, Paul SM (
1991
) γ-Aminobutyric acid (GABA) induces a receptor-mediated reduction in GABAA receptor α subunit messenger RNAs in embryonic chick neurons in culture.
J Biol Chem
 
266
:
6011
–6014.
Montpied P, Morrow AL, Karanian JW, Ginns EI, Martin BM, Paul SM (
1991
) Prolonged ethanol inhalation decreases γ-aminobutyric acidA receptor α subunit mRNAs in rat cerebral cortex.
Mol Pharmacol
 
39
:
157
–163.
Neter J, Kutner MH, Nachtsheim CJ, Wasserman W (1996) Applied linear statistical models. Chicago, IL: Irwin.
Nusser Z, Sieghart W, Benke D, Fritschy J-M, Somogyi P (
1996
) Differential synaptic localization of two major γ-aminobutyric acid type A receptor α subunits on hippocampal pyramidal cells.
Proc Natl Acad Sci USA
 
93
:
11939
–11944.
Nyíri G, Freund TF, Somogyi P (
2001
) Input-dependent synaptic targeting of a2-subunit-containing GABAA receptors in synapses of hippocampal pyramidal cells of the rat.
Eur J Neurosci
 
13
:
428
–442.
O’Donovan MC, Buckland PR, Spurlock G, McGuffin P (
1992
) Bi-directional changes in the levels of messenger RNAs encoding γ-aminobutyric acidA receptor α subunits after flurazepam treatment.
Eur J Pharmacol
 
226
:
335
–341.
Ohnuma T, Augood SJ, Arai H, McKenna PJ, Emson PC (
1999
) Measurement of GABAergic parameters in the prefrontal cortex in schizophrenia: focus on GABA content, GABAA receptor α-1 subunit messenger RNA and human GABA transporter-1 (HGAT-1) messenger RNA expression.
Neuroscience
 
93
:
441
–448.
Pakkenberg B (
1990
) Pronounced reduction of total neuron number in mediodorsal thalamic nucleus and nucleus accumbens in schizophrenics.
Arch Gen Psychiatry
 
47
:
1023
–1028.
Peters A (1984) Chandelier cells. In: Cerebral cortex (Jones EG, Peters A, eds), vol. 1, pp. 361–380. New York: Plenum Press.
Pierri JN, Chaudry AS, Woo T-U, Lewis DA (
1999
) Alterations in chandelier neuron axon terminals in the prefrontal cortex of schizophrenic subjects.
Am J Psychiatry
 
156
:
1709
–1719.
Pierri JN, Volk CLE, Auh S, Sampson A, Lewis DA (
2001
) Decreased somal size of deep layer 3 pyramidal neurons in the prefrontal cortex in subjects with schizophrenia.
Arch Gen Psychiatry
 
58
:
466
–473.
Popken GJ, Bunney WE Jr, Potkin SG, Jones EG (
2000
) Subnucleus-specific loss of neurons in medial thalamus of schizophrenics.
Proc Natl Acad Sci USA
 
97
:
9276
–9280.
Pucak ML, Levitt JB, Lund JS, Lewis DA (
1996
) Patterns of intrinsic and associational circuitry in monkey prefrontal cortex.
J Comp Neurol
 
376
:
614
–630.
Rajkowska G, Goldman-Rakic PS (
1995
) Cytoarchitectonic definition of prefrontal areas in the normal human cortex: I. Remapping of areas 9 and 46 using quantitative criteria.
Cereb Cortex
 
5
:
307
–322.
Rajkowska G, Selemon LD, Goldman-Rakic PS (
1998
) Neuronal and glial somal size in the prefrontal cortex: a postmortem morphometric study of schizophrenia and Huntington disease.
Arch Gen Psychiatry
 
55
:
215
–224.
Selemon LD, Rajkowska G, Goldman-Rakic PS (
1995
) Abnormally high neuronal density in the schizophrenic cortex: a morphometric analysis of prefrontal area 9 and occipital area 17.
Arch Gen Psychiatry
 
52
:
805
–818.
Selemon LD, Rajkowska G, Goldman-Rakic PS (
1998
) Elevated neuronal density in prefrontal area 46 in brains from schizophrenic patients: application of a three-dimensional, stereologic counting method.
J Comp Neurol
 
392
:
402
–412.
Sieghart W, Fuchs K, Tretter V, Ebert V, Jechlinger M, Hoger H, Adamiker D (
1999
) Structure and subunit composition of GABAA receptors.
Neurochem Int
 
34
:
379
–385.
Somogyi P (
1977
) A specific axo-axonal interneuron in the visual cortex of the rat.
Brain Res
 
136
:
345
–350.
Sundman I, Allard P, Eriksson A, Marcusson J (
1997
) GABA uptake sites in frontal cortex from suicide victims and in aging.
Neuropsychobiology
 
35
:
11
–15.
Sutoo D, Akiyama K, Yabe K (
1998
) Quantitative mapping analyzer for determining the distribution of neurochemicals in the human brain.
J Neurosci Methods
 
85
:
161
–173.
Tietz EI, Huang X, Chen S, Ferencak WF III (
1999
) Temporal and regional regulation of a1, b2, and b3, but not a2, a4, a5, a6, b1, or g2 GABAA receptor subunit messenger RNAs following one-week oral flurazepam administration.
Neuroscience
 
91
:
327
–341.
Volk DW, Austin MC, Pierri JN, Sampson AR, Lewis DA (
2000
) Decreased GAD67 mRNA expression in a subset of prefrontal cortical GABA neurons in subjects with schizophrenia.
Arch Gen Psychiatry
 
57
:
237
–245.
Volk DW, Austin MC, Pierri JN, Sampson AR, Lewis DA (
2001
) GABA transporter-1 mRNA in the prefrontal cortex in schizophrenia: decreased expression in a subset of neurons.
Am J Psychiatry
 
158
:
256
–265.
Weinberger DR, Berman KF, Zec RF (
1986
) Physiologic dysfunction of dorsolateral prefrontal cortex in schizophrenia. I. Regional cerebral blood flow evidence.
Arch Gen Psychiatry
 
43
:
114
–124.
Woo T-U, Whitehead RE, Melchitzky DS, Lewis DA (
1998
) A subclass of prefrontal gamma-aminobutyric acid axon terminals are selectively altered in schizophrenia.
Proc Natl Acad Sci USA
 
95
:
5341
–5346.
Young KA, Manaye KF, Liang C-L, Hicks PB, German DC (
2000
) Reduced number of mediodorsal and anterior thalamic neurons in schizophrenia.
Biol Psychiatry
 
47
:
944
–953.