Dissection of the Human T-Cell Receptor γ Gene Repertoire in the Brain and Peripheral Blood Identifies Age- and Alzheimer's Disease-Associated Clonotype Profiles

Abstract

The immune system contributes to neurodegenerative pathologies. However, the roles of γδ T cells in Alzheimer's disease (AD) are poorly understood. Here, we evaluated somatic variability of T-cell receptor γ genes (TRGs) in patients with AD. We performed deep sequencing of the CDR3 region of TRGs in patients with AD and control patients without dementia. TRG clones were clearly detectable in peripheral blood (PB) and non-neuronal cell populations in human brains. TRG repertoire diversity was reduced during aging. Compared with the PB, the brain showed reduced TRGV9 clonotypes but was enriched in TRGV2/4/8 clonotypes. AD-associated TRG profiles were found in both the PB and brain. Moreover, some groups of clonotypes were more specific for the brain or blood in patients with AD compared to those in controls. Our pilot deep analysis of T-cell receptor diversities in AD revealed putative brain and AD-associated immunogenic markers.

Keywords: Alzheimer's disease; T-cell receptor γ genes; clonotype; immune repertoire; immunogenic marker.

Figure 1

Comparative analysis of the TRG repertoire in the cerebral cortex and peripheral blood. The top 20 TRG repertoire the cerebral cortex differed in V-segment composition from that of the peripheral blood: the TRGV9 segment was predominant in the blood (A), whereas the proportions of the TRGV2/4/8 segments significantly increased in the brain (B). The TRGV9-TRGJP clone type “CALW…LGKKIKVF” (where dots indicate any set of amino acids) was significantly more frequent in the blood (C), whereas other TRGV9 clones did not differ (D). A schematic of the V-segment distributions of clones in different data groups is presented in (E). The weighted average repertoire was more hydrophilic in the brain (F). The colors denote various data groups: red, AD samples; blue, control samples; gray, patients with Parkinson's disease. Data were analyzed by Kruskal-Wallis ANOVA with Dunn's post-test comparisons, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Figure 2

Age-related changes in the composition of the TRG repertoire in the blood and brain. Average numbers of unique clones after downsampling the data to 50,000 reads (A,B). Changes in Shannon entropy with age in the peripheral blood (C) and brain (D) (red, AD samples; blue, control samples; gray, patients with Parkinson's disease). Regression analysis of V-segment frequencies in the top 20 clones for the TRGV9 segment (E) and TRGV2/4/8 segments (F) in the cerebral cortex and peripheral blood (G,H). *P < 0.05, *** = < 0.001 and ****P < 0.0001.

Figure 3

TRG repertoires in Alzheimer's patients included clones with specific properties in the blood or brain. We analyzed clones found in at least two patients, but not in healthy individuals (A,B). Values in the matrices indicate numbers of clones found in the samples indicated along the axes. We compared the weighted CDR3 lengths (C,G), hydropathy indexes (D,H), volumes (E,I), and strengths (F,J) between shared AD repertoires, full AD repertoires, total norm repertoires, and shared norm repertoires. For brain samples, values of volume and hydropathy were chosen (K), and for blood, the CDR3 length and strength were chosen (L), as characteristics defining AD-specific clone features. Clones with these characteristics were significantly more represented in patients with Alzheimer's disease (M,O). Logoplots of CDR3 amino acid sequences for brain (N) and blood (P) were compiled (orange, small and weakly interacting amino acids; red, small; violet, small, weakly interacting, and hydrophilic; green, hydrophilic and weakly interacting; blue, hydrophilic). *P < 0.05, **P < 0.01, ***P < 0.001. Red dashed lines (K,L) denote median values for controls.