Neurodegenerative fluid biomarkers are enriched in human cervical lymph nodes

Abstract

In animal models, brain neurodegeneration biomarkers drain into cervical lymph nodes (CLNs), and this drainage function is reduced with ageing. If this occurred in humans, CLNs may provide a readily accessible measure of this aspect of protein clearance. We tested this hypothesis in people using ultrasound-guided fine needle aspiration. We measured amyloid-beta 40 and 42, phosphorylated tau 181 (pTau181), glial fibrillary acidic protein and neurofilament light using single molecule array in CLN aspirates and plasma from: (i) a discovery cohort of 25 autoimmune patients; and (ii) plasma, CLNs and capillary blood in four healthy volunteers, an optimization cohort. Ultrasound-guided fine needle aspiration was well-tolerated by all participants. In both cohorts, all biomarkers were detected in all plasma and CLN samples, other than neurofilament light (8/17 of discovery cohort). CLN biomarker concentrations were significantly greater than plasma concentrations for all except neurofilament light, most markedly for pTau181 (266-fold; P < 0.02), whose CLN concentrations decreased with age (Spearman r = -0.66, P = 0.001). This study presents the first evidence that neurodegenerative biomarkers are detectable in human CLNs. Raised CLN:plasma biomarker ratios suggest their concentration in CLNs may offer a distinct compartment for minimally-invasive measurement of brain clearance and lymphatic drainage, with potential applicability to study of ageing and future clinical trials.

Keywords: biomarker; cervical lymph nodes; dementia; meningeal lymphatics.

Figure 1

Corrected concentrations of dementia fluid biomarkers in plasma versus capillary and cervical lymph node supernatants. (A) Schematic diagram illustrates sample sources (venous blood as plasma and cervical lymph node aspirate supernatant) from 25 clinical donors with autoimmune neurological diseases, which acted as a discovery cohort. (B) Dot and box plots indicate the concentrations of five dementia fluid biomarkers from two colour-coded bio-samples from 25 individuals on a logarithmic axis (plasma = blue; cervical lymph node supernatant = green). The box spans the minimum to maximum value with a horizontal line at the median. Lines between dots indicate the same individual across the two sample types. A dotted line runs across the graphs to indicate the limit of assay detection. The result of a Wilcoxon signed-rank test for each protein is indicated above the plots. (C) Schematic diagram illustrates sample sources (venous blood as plasma, capillary bed blood as supernatant, and cervical lymph node aspirate supernatant) from four healthy donors, which acted as an optimization cohort. (D) Dot and box plots indicate the concentrations of five dementia fluid biomarkers from three colour-coded bio-samples from four individuals on a logarithmic axis (capillary blood supernatant = pink; plasma = blue, cervical lymph node supernatant = green). The box spans the minimum to maximum values with a horizontal line at the median. Lines between dots indicate the same individual across each sample type. A dotted line runs across the graphs to indicate the limit of assay detection. The result of a two-tailed paired t-test for each protein is indicated above the plots. The results for the above protein are summarized below for each donor and sample type. Cells are filled with the same colour scheme to indicate a positive result and ‘X' indicates that this was below the limit of detection. Aβ = amyloid-beta peptide; Cap. = capillary blood supernatant; CLN = cervical lymph node supernatant; GFAP = glial fibrillary acidic protein; NfL = neurofilament light; ns = not significant; pTau181 = phosphorylated tau protein 181. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. Illustrations in A and C are modified from Provine et al.

Figure 2

Association of phosphorylated tau 181 (pTau181) concentration in cervical lymph node and plasma with age. (A) Correlation of corrected concentration of pTau181 in cervical lymph node supernatant and plasma is plotted against age (cervical lymph node = green; plasma = blue). (B) Correlation of the ratio between corrected concentration of pTau181 in cervical lymph node supernatant to corresponding plasma ratio is plotted against age. Each dot is one sample or ratio, respectively; a line of best fit is drawn to summarize the relationship with dotted lines indicating the 95% confidence interval. The Spearman correlation coefficient (r) and statistical significance (P) for each line are given below the graphs. CLN = cervical lymph node.