Invited Review: The spectrum of neuropathology in COVID-19.

There is increasing evidence that patients with Coronavirus disease 19 (COVID-19) present with neurological and psychiatric symptoms. Anosmia, hypogeusia, headache, nausea and altered consciousness are commonly described, although there are emerging clinical reports of more serious and specific conditions such as acute cerebrovascular accident, encephalitis and demyelinating disease. Whether these presentations are directly due to viral invasion of the central nervous system (CNS) or caused by indirect mechanisms has yet to be established. Neuropathological examination of brain tissue at autopsy will be essential to establish the neuro-invasive potential of the SARS-CoV-2 virus but, to date, there have been few detailed studies. The pathological changes in the brain probably represent a combination of direct cytopathic effects mediated by SARS-CoV-2 replication or indirect effects due to respiratory failure, injurious cytokine reaction, reduced immune response and cerebrovascular accidents induced by viral infection. Further large-scale molecular and cellular investigations are warranted to clarify the neuropathological correlates of the neurological and psychiatric features seen clinically in COVID-19. In this review, we summarize the current reports of neuropathological examination in COVID-19 patients, in addition to our own experience, and discuss their contribution to the understanding of CNS involvement in this disease.

Neuregulin-1 (NRG1) polymorphisms linked with psychosis transition are associated with enlarged lateral ventricles and white matter disruption in schizophrenia.

BACKGROUND: Two single-nucleotide polymorphisms (SNPs) (rs4281084 and rs12155594) within the neuregulin-1 (NRG1) gene have been associated with psychosis transition. However, the neurobiological changes associated with these SNPs remain unclear. We aimed to determine what relationship these two SNPs have on lateral ventricular volume and white matter integrity, as abnormalities in these brain structures are some of the most consistent in schizophrenia. METHODS: Structural (n = 370) and diffusion (n = 465) magnetic resonance imaging data were obtained from affected and unaffected individuals predominantly of European descent. The SNPs rs4281084, rs12155594, and their combined allelic load were examined for their effects on lateral ventricular volume, fractional anisotropy (FA) as well as axial (AD) and radial (RD) diffusivity. Additional exploratory analyses assessed NRG1 effects on gray matter volume, cortical thickness, and surface area throughout the brain. RESULTS: Individuals with a schizophrenia age of onset ⩽25 and a combined allelic load ⩾3 NRG1 risk alleles had significantly larger right (up to 50%, p adj = 0.01) and left (up to 45%, p adj = 0.05) lateral ventricle volumes compared with those with allelic loads of less than three. Furthermore, carriers of three or more risk alleles, regardless of age of onset and case status, had significantly reduced FA and elevated RD but stable AD in the frontal cortex compared with those carrying fewer than three risk alleles. CONCLUSIONS: Our findings build on a growing body of research supporting the functional importance of genetic variation within the NRG1 gene and complement previous findings implicating the rs4281084 and rs12155594 SNPs as markers for psychosis transition.

Predicting the diagnosis of autism spectrum disorder using gene pathway analysis.

Autism spectrum disorder (ASD) depends on a clinical interview with no biomarkers to aid diagnosis. The current investigation interrogated single-nucleotide polymorphisms (SNPs) of individuals with ASD from the Autism Genetic Resource Exchange (AGRE) database. SNPs were mapped to Kyoto Encyclopedia of Genes and Genomes (KEGG)-derived pathways to identify affected cellular processes and develop a diagnostic test. This test was then applied to two independent samples from the Simons Foundation Autism Research Initiative (SFARI) and Wellcome Trust 1958 normal birth cohort (WTBC) for validation. Using AGRE SNP data from a Central European (CEU) cohort, we created a genetic diagnostic classifier consisting of 237 SNPs in 146 genes that correctly predicted ASD diagnosis in 85.6% of CEU cases. This classifier also predicted 84.3% of cases in an ethnically related Tuscan cohort; however, prediction was less accurate (56.4%) in a genetically dissimilar Han Chinese cohort (HAN). Eight SNPs in three genes (KCNMB4, GNAO1, GRM5) had the largest effect in the classifier with some acting as vulnerability SNPs, whereas others were protective. Prediction accuracy diminished as the number of SNPs analyzed in the model was decreased. Our diagnostic classifier correctly predicted ASD diagnosis with an accuracy of 71.7% in CEU individuals from the SFARI (ASD) and WTBC (controls) validation data sets. In conclusion, we have developed an accurate diagnostic test for a genetically homogeneous group to aid in early detection of ASD. While SNPs differ across ethnic groups, our pathway approach identified cellular processes common to ASD across ethnicities. Our results have wide implications for detection, intervention and prevention of ASD.

Different changes in cortical tumor necrosis factor-α-related pathways in schizophrenia and mood disorders.

The growing body of evidence implicating tumor necrosis factor-α (TNFα) in the pathophysiology of psychiatric disorders led us to measure levels of that protein in the cortex of subjects with major depressive disorders (MDD). Having reported an increase (458%) in the levels of the transmembrane (tmTNFα), but not the soluble (sTNFα), form of the protein in Brodmann's area (BA) 46, but not 24, in people with the disorder, we decided to examine additional components of TNFα-related pathways in the same regions in people with MDD and extend our studies to the same cortical regions of people with schizophrenia (Sz) and bipolar disorders (BD). Using postmortem tissue, western blots and quantitative PCR, we have now shown there is a significant increase (305%) in tmTNFα in Brodmann's area 24, but not 46, from subjects with BD, and that levels of the protein were not altered in Sz. Levels of sTNFα were not altered in BD or Sz. In addition, we have shown that levels of TNF receptor 1 (TNFR1) mRNA are increased in BA 24 (53%) and BA 46 (82%) in people with Sz, whereas levels of TNFR2 mRNA was decreased in BA 46 in people with mood disorders (MDD=-51%; BD=-67%). Levels of proteins frequently used as surrogate markers of neuronal, astrocytic and microglia numbers, as well as levels of the pro-inflammatory marker (interleukin 1ß), were not changed in the cortex of people with mood disorders. Our data suggest there are differential changes in TNFα-related markers in the cortex of people with MDD, BD and Sz that may not be related to classical inflammation and may cause changes in different TNFα-related signaling pathways.

Effects of neuregulin-1 genetic variation and depression symptom severity on longitudinal patterns of psychotic symptoms in primary care attendees.

A better understanding of the factors associated with psychotic symptoms could aid early identification and treatment of psychotic disorders. Previous studies have typically utilized cross-sectional study designs and have focused on individuals with psychotic disorders. Thus, examination of promising correlates of psychotic symptoms using longitudinal designs among more broadly defined populations is warranted. Two such correlates are neuregulin-1 (NRG1) genotypic variation and depression symptom severity. Both NRG1 and depression symptom severity have cross-sectional evidence for an association with psychosis but their affect on longitudinal patterns of psychotic symptoms and their potential interaction effects are less clear. Using repeated measures analysis of variance and covariance we modeled the main and interaction effects of NRG1 genotypic variation and depressive symptom severity on longitudinal psychotic symptom patterns in 301 primary care attendees assessed annually over 4 years. One-fifth (19.9%) of the participants reported one or more psychotic symptoms over the 4-year assessment period. We observed a curvilinear (i.e., cubic) association between depression symptom severity at baseline and longitudinal patterns of psychotic symptoms but did not observe a main effect for NRG1 genotypic variation on psychotic symptom patterns. However, NRG1 rs6994992 genotype moderated the curvilinear association between depression symptom severity and psychotic symptom patterns. Specifically, depression symptom severity had less of an effect on longitudinal psychotic symptoms among carriers of the rs6994992 TT genotype compared to CC and CT carriers. Our findings suggest a curvilinear association between depression symptom severity and longitudinal patterns of psychotic symptoms that is moderated by NRG1 genotype.

Genome-wide association analyses of symptom severity among clozapine-treated patients with schizophrenia spectrum disorders.

Clozapine is the most effective antipsychotic for patients with treatment-resistant schizophrenia. However, response is highly variable and possible genetic underpinnings of this variability remain unknown. Here, we performed polygenic risk score (PRS) analyses to estimate the amount of variance in symptom severity among clozapine-treated patients explained by PRSs (R2) and examined the association between symptom severity and genotype-predicted CYP1A2, CYP2D6, and CYP2C19 enzyme activity. Genome-wide association (GWA) analyses were performed to explore loci associated with symptom severity. A multicenter cohort of 804 patients (after quality control N = 684) with schizophrenia spectrum disorder treated with clozapine were cross-sectionally assessed using the Positive and Negative Syndrome Scale and/or the Clinical Global Impression-Severity (CGI-S) scale. GWA and PRS regression analyses were conducted. Genotype-predicted CYP1A2, CYP2D6, and CYP2C19 enzyme activities were calculated. Schizophrenia-PRS was most significantly and positively associated with low symptom severity (p = 1.03 × 10-3; R2 = 1.85). Cross-disorder-PRS was also positively associated with lower CGI-S score (p = 0.01; R2 = 0.81). Compared to the lowest tertile, patients in the highest schizophrenia-PRS tertile had 1.94 times (p = 6.84×10-4) increased probability of low symptom severity. Higher genotype-predicted CYP2C19 enzyme activity was independently associated with lower symptom severity (p = 8.44×10-3). While no locus surpassed the genome-wide significance threshold, rs1923778 within NFIB showed a suggestive association (p = 3.78×10-7) with symptom severity. We show that high schizophrenia-PRS and genotype-predicted CYP2C19 enzyme activity are independently associated with lower symptom severity among individuals treated with clozapine. Our findings open avenues for future pharmacogenomic projects investigating the potential of PRS and genotype-predicted CYP-activity in schizophrenia.

Decreased expression of AMPA receptor messenger RNA and protein in AIDS: a model for HIV-associated neurotoxicity.

HIV infection can cause extensive neuronal loss and clinically a severe dementia. The cause of the neurotoxicity remains unclear as neurons are not infected, but disturbance of glutamate-linked calcium entry has been implicated. In this study, we have shown a decrease in HIV-infected brain of the expression of mRNA and protein of the GluR-A flop subtype of alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) glutamate receptor in cerebellar Purkinje cells. Although Purkinje cells are relatively resistant to loss, the observed disturbance of AMPA receptors may contribute to the neurotoxic process in other vulnerable brain regions and clinically to the development of dementia.

Pathway-wide association study identifies five shared pathways associated with schizophrenia in three ancestral distinct populations.

Genome-wide association studies have confirmed the polygenic nature of schizophrenia and suggest that there are hundreds or thousands of alleles associated with increased liability for the disorder. However, the generalizability of any one allelic marker of liability is remarkably low and has bred the notion that schizophrenia may be better conceptualized as a pathway(s) disorder. Here, we empirically tested this notion by conducting a pathway-wide association study (PWAS) encompassing 255 experimentally validated Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways among 5033 individuals diagnosed with schizophrenia and 5332 unrelated healthy controls across three distinct ethnic populations; European-American (EA), African-American (AA) and Han Chinese (CH). We identified 103, 74 and 87 pathways associated with schizophrenia liability in the EA, CH and AA populations, respectively. About half of these pathways were uniquely associated with schizophrenia liability in each of the three populations. Five pathways (serotonergic synapse, ubiquitin mediated proteolysis, hedgehog signaling, adipocytokine signaling and renin secretion) were shared across all three populations and the single-nucleotide polymorphism sets representing these five pathways were enriched for single-nucleotide polymorphisms with regulatory function. Our findings provide empirical support for schizophrenia as a pathway disorder and suggest schizophrenia is not only a polygenic but likely also a poly-pathway disorder characterized by both genetic and pathway heterogeneity.

Meta-analysis reveals associations between genetic variation in the 5' and 3' regions of Neuregulin-1 and schizophrenia.

Genetic, post-mortem and neuroimaging studies repeatedly implicate neuregulin-1 (NRG1) as a critical component in the pathophysiology of schizophrenia. Although a number of risk haplotypes along with several genetic polymorphisms in the 5' and 3' regions of NRG1 have been linked with schizophrenia, results have been mixed. To reconcile these conflicting findings, we conducted a meta-analysis examining 22 polymorphisms and two haplotypes in NRG1 among 16 720 cases, 20 449 controls and 2157 family trios. We found significant associations for three polymorphisms (rs62510682, rs35753505 and 478B14-848) at the 5'-end and two (rs2954041 and rs10503929) near the 3'-end of NRG1. Population stratification effects were found for the rs35753505 and 478B14-848(4) polymorphisms. There was evidence of heterogeneity for all significant markers and the findings were robust to publication bias. No significant haplotype associations were found. Our results suggest genetic variation at the 5' and 3' ends of NRG1 are associated with schizophrenia and provide renewed justification for further investigation of NRG1's role in the pathophysiology of schizophrenia.

Meta-analysis supports GWAS-implicated link between GRM3 and schizophrenia risk.

Genome-wide association study (GWAS) evidence has identified the metabotropic glutamate receptor 3 (GRM3) gene as a potential harbor for schizophrenia risk variants. However, previous meta-analyses have refuted the association between GRM3 single-nucleotide polymorphisms (SNPs) and schizophrenia risk. To reconcile these conflicting findings, we conducted the largest and most comprehensive meta-analysis of 14 SNPs in GRM3 from a total of 11 318 schizophrenia cases, 13 820 controls and 486 parent-proband trios. We found significant associations for three SNPs (rs2237562: odds ratio (OR)=1.06, 95% confidence interval (CI)=1.02-1.11, P=0.017; rs13242038: OR=0.90, 95% CI=0.85-0.96, P=0.016 and rs917071: OR=0.94, 95% CI=0.91-0.97, P=0.003). Two of these SNPs (rs2237562, rs917071) were in strong-to-moderate linkage disequilibrium with the top GRM3 GWAS significant SNP (rs12704290) reported by the Schizophrenia Working Group of the Psychiatric Genomics Consortium. We also found evidence for population stratification related to rs2237562 in that the 'risk' allele was dependent on the population under study. Our findings support the GWAS-implicated link between GRM3 genetic variation and schizophrenia risk as well as the notion that alleles conferring this risk may be population specific.

PET imaging of putative microglial activation in individuals at ultra-high risk for psychosis, recently diagnosed and chronically ill with schizophrenia.

We examined putative microglial activation as a function of illness course in schizophrenia. Microglial activity was quantified using [11C](R)-(1-[2-chrorophynyl]-N-methyl-N-[1-methylpropyl]-3 isoquinoline carboxamide (11C-(R)-PK11195) positron emission tomography (PET) in: (i) 10 individuals at ultra-high risk (UHR) of psychosis; (ii) 18 patients recently diagnosed with schizophrenia; (iii) 15 patients chronically ill with schizophrenia; and, (iv) 27 age-matched healthy controls. Regional-binding potential (BPND) was calculated using the simplified reference-tissue model with four alternative reference inputs. The UHR, recent-onset and chronic patient groups were compared to age-matched healthy control groups to examine between-group BPND differences in 6 regions: dorsal frontal, orbital frontal, anterior cingulate, medial temporal, thalamus and insula. Correlation analysis tested for BPND associations with gray matter volume, peripheral cytokines and clinical variables. The null hypothesis of equality in BPND between patients (UHR, recent-onset and chronic) and respective healthy control groups (younger and older) was not rejected for any group comparison or region. Across all subjects, BPND was positively correlated to age in the thalamus (r=0.43, P=0.008, false discovery rate). No correlations with regional gray matter, peripheral cytokine levels or clinical symptoms were detected. We therefore found no evidence of microglial activation in groups of individuals at high risk, recently diagnosed or chronically ill with schizophrenia. While the possibility of 11C-(R)-PK11195-binding differences in certain patient subgroups remains, the patient cohorts in our study, who also displayed normal peripheral cytokine profiles, do not substantiate the assumption of microglial activation in schizophrenia as a regular and defining feature, as measured by 11C-(R)-PK11195 BPND.

Aberrant expression of microRNAs as biomarker for schizophrenia: from acute state to partial remission, and from peripheral blood to cortical tissue.

Based on our previous finding of a seven-miRNA (hsa-miR-34a, miR-449a, miR-564, miR-432, miR-548d, miR-572 and miR-652) signature as a potential biomarker for schizophrenia, this study aimed to examine if hospitalization could affect expressions of these miRNAs. We compared their expression levels between acute state and partial remission state in people with schizophrenia (n=48) using quantitative PCR method. Further, to examine whether the blood and brain show similar expression patterns, the expressions of two miRNAs (hsa-miR-34a and hsa-miR-548d) were examined in the postmortem brain tissue of people with schizophrenia (n=25) and controls (n=27). The expression level of the seven miRNAs did not alter after ~2 months of hospitalization with significant improvement in clinical symptoms, suggesting the miRNAs could be traits rather than state-dependent markers. The aberrant expression seen in the blood of hsa-miR-34a and hsa-miR-548d were not present in the brain samples, but this does not discount the possibility that the peripheral miRNAs could be clinically useful biomarkers for schizophrenia. Unexpectedly, we found an age-dependent increase in hsa-miR-34a expressions in human cortical (Brodmann area 46 (BA46)) but not subcortical region (caudate putamen). The correlation between hsa-miR-34a expression level in BA46 and age was much stronger in the controls than in the cases, and the corresponding correlation in the blood was only seen in the cases. The association between the miRNA dysregulations, the disease predisposition and aging warrants further investigation. Taken together, this study provides further insight on the candidate peripheral miRNAs as stable biomarkers for the diagnostics of schizophrenia.

The shifting patterns of HIV encephalitis neuropathology.

HIV infected macrophages infiltrate the nervous system early in the progression of HIV infection, leading to a complex set of neuropathological alterations including HIV encephalitis (HIVE), leukoencephalopathy and vacuolar myelopathy that in turn result in neurodegeneration of selective cellular populations and pathways involved in regulating cognitive and motor functioning. Rapid progress in the development of highly active antiretroviral therapy (HAART) has changed the patterns of HIV related neuropathology and neurological manifestations in the past 10 years. The prevalence of opportunistic infections and central nervous system (CNS) neoplasms has decreased, and some groups have proposed that the frequency of chronic forms of HIVE have been rising as the HAART-treated HIV population ages. Accordingly, clinical manifestations have shifted from severe dementia forms to more subtle minor cognitive impairment, leading to the suggestion of a classification of HIV associated neurological conditions into an inactive form, a chronic variety, and a 'transformed' variant. From a neuropathological point of view these variants might correspond to: a) aggressive forms with severe HIVE and white matter injury, b) extensive perivascular lymphocytic infiltration, c) 'burnt-out' forms of HIVE and d) aging-associated amyloid accumulation with Alzheimer's-like neuropathology. Factors contributing to the emergence of these variants of HIVE include the development of viral resistance, immune reconstitution, anti-retroviral drug toxicity and co-morbid factors (e.g., methamphetamine, HCV). More detailed characterization of these proposed variants of HIVE is important in order to better understand the pathogenesis of HIV-associated neurological damage and to design more effective treatments to protect the nervous system.

SELENBP1 expression in the prefrontal cortex of subjects with schizophrenia.

Selenium binding protein 1 (SELENBP1) messenger RNA (mRNA) has previously been shown to be upregulated in the brain and blood from subjects with schizophrenia. We aimed to validate these findings in a new cohort using real-time PCR in Brodmann's Area (BA) 9, and to determine the disease specificity of increased SELENBP1 expression by measuring SELENBP1 mRNA in subjects with major depressive disorder and bipolar disorder. We then extended the study to include other cortical regions such as BA8 and BA44. SELENBP1 mRNA was higher in BA9 (P = 0.001), BA8 (P = 0.003) and BA44 (P = 0.0007) from subjects with schizophrenia. Conversely, in affective disorders, there was no significant difference in SELENBP1 mRNA in BA9 (P = 0.67), suggesting that the upregulation may be diagnosis specific. Measurement of SELENBP1 protein levels showed that changes in mRNA did not translate to changes in protein. In addition, chronic treatment of rats with antipsychotics did not significantly affect the expression of Selenbp1 in the cortex (P = 0.24). Our data show that elevated SELENBP1 transcript expression is widespread throughout the prefrontal cortex in schizophrenia, and confirm that this change is a consistent feature of schizophrenia and not a simple drug effect.

HIV-associated brain pathology in the United Kingdom: an epidemiological study.

OBJECTIVES: To examine the epidemiology of HIV-associated neuropathology in the United Kingdom and to investigate whether the prevalence of different forms of HIV-associated brain pathology varies with exposure category. DESIGN: The study was a cross-sectional survey; data was analysed from the Medical Research Council National AIDS Neuropathology database. SETTING: Information was gathered from throughout England, Scotland and Wales. SUBJECTS: Individuals who died from AIDS in the United Kingdom and had a postmortem examination. The database comprised 7% of all AIDS deaths in the United Kingdom between 1982 and 1993. MAIN OUTCOME: Neuropathological diagnoses based on internationally accepted neuropathological terminology of AIDS-related brain lesions. RESULTS: HIV encephalitis was the most prevalent pathological diagnosis, occurring in 25.3% [95% confidence interval (CI), 21.0-29.6] of the study sample. Statistically significant independent associations for the occurrence of HIV encephalitis were found for injecting drug use (odds ratio, 6.86; 95% CI, 2.91-16.17), and age less than 30 years at death (odds ratio, 3.58; 95% CI, 1.99-6.44). Vascular lesions were significantly higher among blood product recipients, 95% of whom were haemophiliacs. CONCLUSIONS: This was the first epidemiological investigation of HIV-associated brain pathology in the United Kingdom. HIV encephalitis appeared to occur more frequently in injecting drug users and those who died younger. Whereas the findings must be interpreted cautiously, one hypothesis was that differences in the route of transmission may have affected the manifestation of HIV-associated brain damage.

Comparison of two quantitative methods for the evaluation of neuronal number in the frontal cortex in Alzheimer disease.

How to assess the substantial neuronal loss in a neurodegenerative disease such as Alzheimer disease is still being debated. Recently, stereological procedures have been proposed that claim improved accuracy and statistical power, but the results of some of these investigations have been controversial. In this study we compared and correlated the cell density results calculated per unit of volume obtained by a stereological technique, the "selector," with the cell counts per unit area obtained by computer-aided image analysis morphometry, in the same sections of midfrontal cortex in Alzheimer disease and control cases. The "selector" revealed a significant decrease in neuronal density that correlated well with a similar fall in large neuronal counts per unit area, as estimated by image analysis morphometry. These results indicate that stereological techniques and image analysis morphometry are complementary methods in reliably assessing cellular populations in neurodegenerative disorders.

Amelioration of neurotoxic effects of HIV envelope protein gp120 by fibroblast growth factor: a strategy for neuroprotection.

Approximately two thirds of patients with human immunodeficiency virus encephalitis (HIVE) show cognitive impairment and neurodegeneration, while one third are cognitively unimpaired and their neuronal populations are preserved. Thus, it is possible that these individuals might have the capacity to produce neurotrophic factors capable of protecting neurons against the deleterious effects of HIV. In this context, the main objective of this study was to determine whether fibroblast growth factor 1 (FGF1) is protective against HIV. For this purpose levels of FGF1 immunoreactivity were determined in the frontal cortex of 35 AIDS cases subdivided into 4 groups according to the presence or absence of HIVE and neurodegeneration. In cases without both HIVE and neurodegeneration, mild to moderate levels of FGFI immunoreactivity were observed in pyramidal neurons, while in cases with HIVE but without neurodegeneration, levels were significiantly elevated. In contrast, individuals with both HIVE and neurodegeneration showed low levels of neuronal FGF1 immunoreactivity. Furthermore, studies in primary human neuronal cultures treated with the HIV envelope protein-gp120 in the presence or absence of FGF1 showed that FGF1 was protective against gpl20 neurotoxicity in a dose-dependent manner. Taken together, these results support the notion that upregulation of certain neurotrophic factors, such as FGF1, might protect the central nervous system from the neurotoxic effects of HIV.

Cortical synaptic density is reduced in mild to moderate human immunodeficiency virus neurocognitive disorder. HNRC Group. HIV Neurobehavioral Research Center.

Dendritic and synaptic damage (without frank neuronal loss) may be seen in milder human immunodeficiency virus (HIV)-related cognitive disorders. Synapse volume estimates, performed by stereological methods, could enhance the ability to detect subtle neuronal changes that may accompany cognitive impairment in HIV infection. For the present study, synaptic density and neuronal number were assessed by combined stereology/confocal microscopy and these measures were then correlated with ante-mortem levels of cognitive performance in AIDS patients. Three-dimensional stereological measures showed a significant correlation between reduced synaptic density and poor neuropsychological performance. To evaluate the specificity of any observed associations, additional variables including viral burden, astrogliosis and number of calbindin-immunoreactive neurons were measured. Factor analysis of a set of neuropathological variables revealed two factors; one defined by synaptic density and volume fraction, calbindin pyramidal neuronal densities and viral burden; the second by astrocytosis and calbindin interneuron density. Only the first factor correlated significantly with neuropsychological functioning during life. It is concluded that a combination of factors including synaptic damage, specific neuronal loss and increasing viral load underlies HIV-associated cognitive impairment. As synaptic damage is potentially reversible, early diagnosis and treatment of mild cogntive disorders may improve functioning and prevent the progression of brain disease.