Amyloid and tau cerebrospinal fluid biomarkers in HIV infection

Abstract

Background: Because of the emerging intersections of HIV infection and Alzheimer's disease, we examined cerebrospinal fluid (CSF) biomarkers related of amyloid and tau metabolism in HIV-infected patients.

Methods: In this cross-sectional study we measured soluble amyloid precursor proteins alpha and beta (sAPPalpha and sAPPbeta), amyloid beta fragment 1-42 (Abeta1-42), and total and hyperphosphorylated tau (t-tau and p-tau) in CSF of 86 HIV-infected (HIV+) subjects, including 21 with AIDS dementia complex (ADC), 25 with central nervous system (CNS) opportunistic infections and 40 without neurological symptoms and signs. We also measured these CSF biomarkers in 64 uninfected (HIV-) subjects, including 21 with Alzheimer's disease, and both younger and older controls without neurological disease.

Results: CSF sAPPalpha and sAPPbeta concentrations were highly correlated and reduced in patients with ADC and opportunistic infections compared to the other groups. The opportunistic infection group but not the ADC patients had lower CSF Abeta1-42 in comparison to the other HIV+ subjects. CSF t-tau levels were high in some ADC patients, but did not differ significantly from the HIV+ neuroasymptomatic group, while CSF p-tau was not increased in any of the HIV+ groups. Together, CSF amyloid and tau markers segregated the ADC patients from both HIV+ and HIV- neuroasymptomatics and from Alzheimer's disease patients, but not from those with opportunistic infections.

Conclusions: Parallel reductions of CSF sAPPalpha and sAPPbeta in ADC and CNS opportunistic infections suggest an effect of CNS immune activation or inflammation on neuronal amyloid synthesis or processing. Elevation of CSF t-tau in some ADC and CNS infection patients without concomitant increase in p-tau indicates neural injury without preferential accumulation of hyperphosphorylated tau as found in Alzheimer's disease. These biomarker changes define pathogenetic pathways to brain injury in ADC that differ from those of Alzheimer's disease.

Figure 1

Amyloid and tau CSF biomarker concentrations in the different patient groups and their interactions. Panels A-E plot concentrations of different markers for the 6 subject groups. Boxes encompass interquartile ranges with median (line) and mean ("+"), while whiskers designate 10^th to 90^th percentiles. Results of statistical analysis of these data are given in Table 2A. Panel F shows the principal component analysis (PCA) score scatter plot of subjects identified by disease grouping with the superimposed variable loadings of the biomarkers. The data in Panels A - E were used to construct a two-dimensional model along the two vectors. The individual subjects are represented by the small symbols as indicated in the key, and their placement is determined by the combination of their marker values in the model. The placement of the five variables (large symbols as indicated) gives information about how each affects the placement of the individuals in the plot, both with respect to direction along each vector axis and the relative magnitude of this effect as indicated by their distance from the origin. For example, high p-tau and low sAPPβ values contribute importantly to the position in the upper right and left quadrants, respectively. Note, however, that all variables influence where each subject is located in the plot so that the plot provides a visual representation of how the subjects are separated by the results of all five biomarkers. For more detailed information on methods, see reference [35]. Abbreviations: NA, HIV+ neuroasymptomatics; ADC, AIDS dementia complex; CNS OIs, CNS opportunistic infections; Contr y, younger HIV- controls; AD, Alzheimer's disease; Contr o, older HIV- controls.

Figure 2

CSF biomarkers in opportunistic infections. The results for each of the biomarkers in the four infection subgroups are compared to the results of the HIV+ neuroasymptomatic subjects. The symbols show individual results while bars show medians and, for the neuroasymptomatics, the interquartile ranges. Abbreviations: NA, HIV+ neuroasymptomatics; CMV-E, CMV enecephalitis; Crypto, cryptococcal meningitis; Toxo, cerebral toxoplasmosis; PML, progressive multifocal leukoncephalopathy.

Figure 3

Selected marker correlations across patient groups. The four panels show relationships between markers: Panel A between sAPPα and sAPPβ, Panel B between Aβ_1-42 and sAPPβ, Panel C between Aβ_1-42 and sAPPα, and Panel D between p-tau and t-tau. Symbols for all four panels are defined in Panel A. The diagonal lines show linear regressions with 95% confidence intervals; each panel lists the P value and r² for the regression. Abbreviations are those given in Figure 1.

Figure 4

Hypothesis of alterations in amyloid metabolism in ADC and opportunistic infections compared to AD. The diagram shows proteolytic cleavages of the largest isoform of APP (APP770) and interprets the observed differences in effects of HIV and opportunistic infections from those of Alzheimer's disease on these pathways. In the non-amyloidogenic pathway (a.), cleavage by α-secretase occurs after residue 687, which enables the secretion of the large, soluble ectodomain of APP (sAPPα) into the medium and retention of the 83-residue C-terminal fragment (α-CTF) in the membrane. The α-CTF fragment can undergo cleavage by γ-secretase at residue 711 or 713 to release the p3 peptides (c.). In the amyloidogenic pathway (b.), β-secretase cleaves after residue 671, which causes the secretion of the slightly truncated sAPPβ molecule and the retention of a 99 residue C-terminal fragment (β-CTF). This fragment can undergo further cleavage by γ-secretase to release 40 or 42 aminoacid-long Aβ fragments (c.). Cleavage of both α- and β-CTF by γ-secretase releases the APP intracellular domain (AICD) into the cytoplasm. Hypothesized differences in the steps leading to CSF biomarker changes include: 1. an effect of CNS infection or immune activation on an early step in APP metabolism, in contrast to 2. deposition of Aβ_1-42 in Alzheimer's disease[41].