Sharon L. Eastwood

The relative importance of premortem acidosis and postmortem interval for human brain gene expression studies : selective mRNA vulnerability and comparison with their encoded proteins

To help account for the variable quality and quantity of RNA in human brain, we have studied the effect of premortem (agonal state) and postmortem factors on the detection of poly(A)+mRNA and eight mRNAs. For comparison, the influence of the same factors upon gene products encoded by the mRNAs was studied immunocytochemically or by receptor autoradiography. Brain pH declined with increasing age at death and was related to agonal state severity, but was independent of postmortem interval and the histological presence of hypoxic changes. By linear regression, pH was significantly associated with the abundance of several of the RNAs, but not with poly(A)+mRNA, immunoreactivities, or binding site densities. Postmortem interval had a limited influence upon mRNA and protein products. Freezer storage time showed no effect. Parallel rat brain studies showed no relationship between postmortem interval (0-48 h) and amounts of total RNA, poly(A)+RNA, or two individual mRNAs; however, RNA content was reduced by 40% at 96 h after death. pH is superior to clinical assessments of agonal state or mode of death in predicting mRNA preservation. It provides a simple means to improve human brain gene expression studies. pH is stable after death and during freezer storage and can be measured either in cerebrospinal fluid or in homogenised tissue.

5-HT1A and 5-HT2A receptor mRNAs and binding site densities are differentially altered in schizophrenia

We have investigated 5-HT1A (serotonin1A) and 5-HT2A (serotonin2A) receptor mRNA abundance and binding site densities in various neocortical and hippocampal regions of schizophrenics and control subjects. Age, agonal state (brain pH), and post mortem interval were included where necessary as covariates in our analyses. In schizophrenics, 5-HT1A binding site densities, determined autoradiographically by [3H]8-hydroxy-2,3-(dipropylamino)-tetralin ([3H]8-OH-DPAT), were significantly increased (+23%) in the dorsolateral prefrontal cortex, with a similar trend in anterior cingulate gyrus. These increases were not accompanied by any change in 5-HT1A receptor mRNA. No differences between the groups in [3H]8-OH-DPAT binding or 5-HT1A receptor mRNA were seen in superior temporal gyrus, striate cortex, or hippocampus. 5-HT2A binding sites, determined by [3H]ketanserin, were decreased in the dorsolateral prefrontal cortex (−27%) and parahippocampal gyrus (−38%) of schizophrenics, with a similar trend in cingulate gyrus, but not in superior temporal gyrus or striate cortex. 5-HT2A receptor mRNA abundance was reduced in schizophrenics in the dorsolateral prefrontal (−49%), superior temporal (−48%), anterior cingulate (−63%) and striate (−63%) cortices, but not in parahippocampal gyrus. Parallel analyses of rat brain tissue showed no changes in 5-HT1A or 5-HT2A receptor mRNAs or binding site densities after chronic administration of haloperidol. These data show that schizophrenia is associated with alterations in the expression of central 5-HT1A and 5-HT2A receptors. They confirm reports of increased 5-HT1A and decreased 5-HT2A binding site densities in prefrontal cortex, and reveal more extensive decreases in 5-HT2A receptor gene expression at the mRNA level. The resulting imbalance in the 5-HT1A to 5-HT2A receptor ratio, when considered in terms of the chemoarchitectural distribution of these receptors, may contribute to an impairment of corticocortical association pathways. The apparent dissociation of the normal relationships between the abundance of each 5-HT receptor and its mRNA in schizophrenia introduces a separate complexity to the data, which may give clues to the underlying molecular mechanisms.

The distribution of 5-HT1A and 5-HT2A receptor mRNA in human brain

We have examined the distribution of 5-HT1A and 5-HT2A receptor mRNAs in post-mortem human hippocampus, neocortex, raphe nuclei, cerebellum and basal ganglia using in situ hybridization histochemistry. Receptor transcripts in brains from two males and two females (mean age +/- S.D. = 70 +/- 4 years; post-mortem interval = 29 +/- 6 h) were visualised with 35S-radiolabelled synthetic oligodeoxyribonucleic acid probes. In the hippocampus, 5-HT1A receptor mRNA was present in all fields, especially CA1. In the parahippocampal gyrus and neocortical regions 5-HT1A receptor mRNA was enhanced in superficial and middle laminae. 5-HT1A receptor mRNA was particularly abundant in the raphe and other serotonergic cell groups of the brainstem. The analysis of emulsion dipped sections showed 5-HT1A receptor mRNA to be concentrated in pyramidal neurons, together with the granule cells of the dentate gyrus. In neocortical areas lamina III pyramidal neurons were more heavily labelled than those in lamina V. There was no evidence of glial expression of 5-HT1A receptor mRNA in grey matter or white matter compartments. 5-HT2A receptor mRNA was present in all neocortical areas examined, where it was located in pyramidal neurons, of lamina V more than in those of lamina III, as well as in putative interneurons, especially within lamina IVc of the striate cortex. 5-HT2A receptor mRNA was observed at minimal levels in the hippocampus and not in the raphe. Neither 5-HT1A nor 5-HT2A receptor mRNA were detected in the cerebellum, substantia nigra or striatum. The ability to detect these transcripts at the regional and cellular level will help reveal important details of the 5-HT receptor system in the human brain. This includes the investigation of their putative roles in the normal chemoarchitecture and in pathophysiological brain processes.

Synaptic pathology in the anterior cingulate cortex in schizophrenia and mood disorders. A review and a Western blot study of synaptophysin, GAP-43 and the complexins

There are several reports of ultrastructural and protein changes affecting synapses in the anterior cingulate cortex in schizophrenia. Altered cytoarchitecture has also been described in this region in schizophrenia as well as in mood disorders. In this paper we review the literature and present a new study investigating synaptic abnormalities in the anterior cingulate cortex (area 24) in the Stanley Foundation brain series. We used Western blotting to assess four synaptic proteins: synaptophysin, growth-associated protein-43 (GAP-43), complexin I and complexin II, which inform about somewhat different aspects of the synaptic circuitry. Synaptophysin, complexin II and GAP-43 were reduced in bipolar disorder. The decreases correlated with the duration of illness and tended to be greater in subjects without a family history. Complexin II was also reduced in major depression. Complexin I and the housekeeping protein beta-actin did not differ between groups. None of the proteins changed significantly in schizophrenia. The results indicate the presence of a synaptic pathology in the anterior cingulate cortex in mood disorders, especially bipolar disorder. The abnormalities may contribute to the dysfunction of cingulate neural circuits. The loss of synaptophysin is suggestive of decreased synaptic density whilst the decrease in GAP-43 may denote impaired synaptic plasticity and the reduction of complexin II but not complexin I implies that the alterations particularly affect excitatory connections. The reductions may be progressive.

Altered synaptophysin expression as a marker of synaptic pathology in schizophrenia.

It has been proposed that synaptic density or synaptic innervation may be altered in schizophrenia as a correlate of the neurodevelopmental pathology of the disease. Synaptophysin is a synaptic vesicle protein whose distribution and abundance provides a synaptic marker which can be reliably measured in post mortem brain. We have used in situ hybridization histochemistry and immunoreactivity to assess the expression of synaptophysin messenger RNA and protein respectively in medial temporal lobe from seven schizophrenics and 13 controls. In the schizophrenic cases, synaptophysin messenger RNA was reduced bilaterally in CA4, CA3, subiculum and parahippocampal gyrus, with a similar trend in dentate gyrus but no change in CA1. It was also decreased in terms of grains per pyramidal neuron in the affected subfields. In parahippocampal gyrus, the loss of synaptophysin messenger RNA per neuron in schizophrenia was greater in deep than superficial laminae. A parallel study in rats showed no effect of haloperidol treatment upon hippocampal synaptophysin messenger RNA, suggesting that neuroleptic treatment does not underlie the reductions found in schizophrenia. In the right medial temporal lobe of schizophrenics, we confirmed the correlation of synaptophysin messenger RNA abundance between ipsilateral subfields seen in both hemispheres of control brains. However, these correlations were not observed in the left medial temporal lobe of the schizophrenic cases. Synaptophysin immunoreactivity in schizophrenia showed no significant differences in any subfield compared to controls. Our data support the broad hypothesis that synaptic pathology occurs in schizophrenia. In so far as synaptophysin expression is a marker for synaptic density, the data suggest that pyramidal neurons within the medial temporal lobe may form fewer synapses. However, the lack of any significant differences in synaptophysin immunoreactivity despite the loss of encoding messenger RNA means that this conclusion must be drawn cautiously. There are several plausible explanations for the preservation of synaptophysin immunoreactivity despite reductions in transcript abundance; one possibility is that the inferrred loss of synapses occurs in extra-hippocampal sites to which the affected pyramidal neurons project. For example, the reduction in synaptophysin messenger RNA in subicular neurons may be accompanied by decreased density of synaptic terminals in the nucleus accumbens. Such differences in the efferent synaptic connectivity of the hippocampus have previously been hypothesized to be an important component of the circuitry underlying schizophrenia.

Interstitial white matter neurons express less reelin and are abnormally distributed in schizophrenia: towards an integration of molecular and morphologic aspects of the neurodevelopmental hypothesis

Two main pieces of neurobiological evidence are adduced to support an early neurodevelopmental component to schizophrenia. Firstly, an abnormal distribution of neurons, especially interstitial white matter neurons (IWMNs). Secondly, decreased expression of reelin, a key developmental signalling molecule. Although influential, neither result is wholly established, and a possible link between them has not been examined. We addressed both issues, in superior temporal cortex, in 12 subjects with schizophrenia and 14 controls. The distribution and density of IWMNs, immunostained with the neuronal marker NeuN, was increased in the superficial white matter in schizophrenia (+16%; P=0.03). IWMN density in deep white matter was unaffected. Using in situ hybridization, reelin mRNA was found to be expressed by many IWMNs, layer I neurons, and scattered interneurons. Superficial IWMNs (P=0.008) and layer I neurons (P=0.036) both expressed less reelin mRNA per cell in schizophrenia, with a trend for deep IWMNs (P=0.055). In conclusion, we replicated findings of increased IWMN density, and of decreased reelin expression, in schizophrenia. The loss of reelin reflects, at least partly, its decreased expression by IWMNs. These findings together support neurodevelopmental theories of the disorder, and indicate a link between reelin and IWMNs in this process, consistent with evidence from the heterozygous reeler mutant mouse. The alterations may contribute to the aberrant synaptic connectivity seen in schizophrenia. However, the functional implications of the abnormalities, as well as the mechanisms involved, remain to be fully elucidated.

Decreased expression of mRNAs encoding non-NMDA glutamate receptors GluR1 and GluR2 in medial temporal lobe neurons in schizophrenia

Schizophrenia is associated with a complex pattern of alterations in the glutamatergic system of the brain. Previous studies have shown a reduced density of some hippocampal non-N-methyl-D-aspartate (non-NMDA) receptors which is accompanied by a loss of encoding receptor mRNA. We have extended this work using in situ hybridization histochemistry with oligonucleotide probes specific for two non-NMDA receptor transcripts, GluR1 and GluR2, in right and left medial temporal lobe sections from 9 schizophrenics and 14 matched normal controls. Both mRNAs were found to be decreased bilaterally and to a similar degree in the hippocampal formation in schizophrenia. Analysis of autoradiograms showed a regional loss of GluR1 and GluR2 mRNAs in dentate gyrus, CA4, CA3 and subiculum. GluR2 mRNA was also reduced in parahippocampal gyrus. These reductions ranged from 25% to 70% in terms of 35S nCi/g tissue equivalents. Additionally we measured grain density for the mRNAs over individual pyramidal neurons in each area. GluR1 and GluR2 mRNAs were less abundant per neuron in CA4 and CA3 in schizophrenia than in controls. GluR2 mRNA was also reduced significantly in parahippocampal gyrus neurons, with an increase in the proportion of GluR1 mRNA to GluR2 mRNA in this cell population. No asymmetries in expression of GluR1 and GluR2 were found in normal or schizophrenic brains. These data further the evidence for reduced non-NMDA receptor expression in the medial temporal lobe in schizophrenia. They confirm the decrease in GluR1 mRNA and show that there are similar losses of GluR2 mRNA in the hippocampal formation. The pattern of changes in the two mRNAs suggests a common mechanism which is unknown but which may be a correlate of the neurodevelopmental abnormalities postulated to underlie the disease. The reduction of GluR2 mRNA but not GluR1 mRNA in parahippocampal gyrus neurons in schizophrenia may have functional consequences given the calcium permeability of non-NMDA receptors lacking the GluR2 subunit.

Decreased expression of vesicular glutamate transporter 1 and complexin II mRNAs in schizophrenia: further evidence for a synaptic pathology affecting glutamate neurons.

Synaptic protein gene expression is altered in schizophrenia. In the hippocampal formation there may be particular involvement of glutamatergic neurons and their synapses, but overall the profile remains unclear. In this in situ hybridization histochemistry (ISHH) study, we examined four informative synaptic protein transcripts: vesicular glutamate transporter (VGLUT) 1, VGLUT2, complexin I, and complexin II, in dorsolateral prefrontal cortex (DPFC), superior temporal cortex (STC), and hippocampal formation, in 13 subjects with schizophrenia and 18 controls. In these areas, VGLUT1 and complexin II are expressed primarily by excitatory neurons, whereas complexin I is mainly expressed by inhibitory neurons. In schizophrenia, VGLUT1 mRNA was decreased in hippocampal formation and DPFC, complexin II mRNA was reduced in DPFC and STC, and complexin I mRNA decreased in STC. Hippocampal VGLUT1 mRNA declined with age selectively in the schizophrenia group. VGLUT2 mRNA was not quantifiable due to its low level. The data provide additional evidence for a synaptic pathology in schizophrenia, in terms of a reduced expression of three synaptic protein genes. In the hippocampus, the loss of VGLUT1 mRNA supports data indicating that glutamatergic presynaptic deficits are prominent, whereas the pattern of results in temporal and frontal cortex suggests broadly similar changes may affect inhibitory and excitatory neurons. The impairment of synaptic transmission implied by the synaptic protein reductions may contribute to the dysfunction of cortical neural circuits that characterises the disorder.

[3H]WAY–100635 for 5–HT1A receptor autoradiography in human brain: a comparison with [3H]8–OH–DPAT and demonstration of increased binding in the frontal cortex in schizophrenia

WAY-100635 is the first selective, silent 5-HT1A (5-hydroxytryptamine1A, serotonin-1A) receptor antagonist. We have investigated the use of [3H]WAY-100635 as a quantitative autoradiographic ligand in post-mortem human hippocampus, raphe and four cortical regions, and compared it with the 5-HT1A receptor agonist, [3H]8-OH-DPAT. Saturation studies showed an average Kd for [3H]WAY-100635 binding in hippocampus of 1.1 nM. The regional and laminar distributions of [3H]WAY-100635 binding and [3H]8-OH-DPAT binding were similar. The density of [3H]WAY-100635 binding sites was 60-70% more than that of [3H]8-OH-DPAT in all areas examined except the cingulate gyrus where it was 165% higher. [3H]WAY-100635 binding was robust and was not affected by the post-mortem interval, freezer storage time or brain pH (agonal state). Using [3H]WAY-100635, we confirmed an increase of 5-HT1A receptor binding sites in the frontal cortex in schizophrenia, previously demonstrated with [3H]8-OH-DPAT. Compared to [3H]8-OH-DPAT, [3H]WAY-100635 has two advantages: it has a higher selectivity and affinity for the 5-HT1A receptor, and it recognizes 5-HT1A receptors whether or not they are coupled to a G-protein, whereas [3H]8-OH-DPAT primarily detects coupled receptors. Given these considerations, the [3H]WAY-100635 binding data in schizophrenia clarify two points. First, they indicate that the elevated [3H]8-OH-DPAT binding seen in the same cases is attributable to an increase of 5-HT1A receptors rather than any other binding site. Second, the enhanced [3H]8-OH-DPAT binding in schizophrenia reflects an increased density of 5-HT1A receptors, not an increased percentage of 5-HT1A receptors which are G-protein-coupled. We conclude that [3H]WAY-100635 is a valuable autoradiographic ligand for the qualitative and quantitative study of 5-HT1A receptors in the human brain.

Decreased synaptophysin in the medial temporal lobe in schizophrenia demonstrated using immunoautoradiography.

Synaptic alterations have been suggested, largely on theoretical grounds, to occur in the brain in schizophrenia. The messenger RNA encoding synaptophysin, a presynaptic terminal protein, is reduced in the medial temporal lobe in the disease, but immunocytochemical and immunoblotting data have not produced clear evidence for a loss of the encoded protein. Here we have used immunoautoradiography with an antisynaptophysin monoclonal antibody and a 35S-labelled secondary antibody in medial temporal lobe sections from 11 schizophrenics and 14 matched controls. In the schizophrenic cases there was an overall loss of synaptophysin (P < 0.02). Analysis by subfield showed significant reductions in the right dentate gyrus molecular layer, subiculum and parahippocampal gyrus, with similar trends in most other subfields. These data confirm that synaptophysin expression is decreased within the medial temporal lobe in schizophrenia. In the respect that synaptophysin is a marker of synaptic density, our findings suggest that reduced synaptic density may be a feature of the molecular neuropathology of the disease.