Aging is a complex process influenced by mechanisms operating at numerous levels of functioning. Multiple biomarkers of age have been identified, yet we know little about how the different alternative age indicators are intertwined. In the Berlin Aging Study II (n_min= 328; n_max= 1,517, women = 51%; 14.27 years of education), we examined how levels and seven-year changes in indicators derived from blood assays, MRI brain scans, other-ratings, and self-reports converge among older adults. We included eight epigenetic biomarkers (incl. five epigenetic “clocks”), a BioAge composite from clinical laboratory parameters, brain age, skin age, subjective age, subjective life expectancy, and future health horizon. We found moderate associations within aging domains, both cross-sectionally and longitudinally over seven years. However, associations across different domains were infrequent and modest. Notably, participants with older BioAge had correspondingly older epigenetic ages. Our results suggest that different aging clocks are only loosely interconnected and that more specific measures are needed to differentiate healthy from unhealthy aging.

The immune system plays a crucial role in many human diseases. In this context, genome-wide association studies (GWAS) offer valuable insights to elucidate the role of immunity in health and disease. The present multi-omics study aimed to identify genetic determinants of immune cell type distributions in the blood of healthy individuals and to assess whether the distributions of these cells may play a role for autoimmune and COVID-19 disease risk. To this end, the frequencies of different immune cells in 483 healthy individuals from the Berlin Aging Study II were quantified using flow cytometry, and GWAS was performed for 92 immune cell phenotypes. Additionally, we performed linear regression analyses of immune cell distributions using polygenic risk scores (PRS) based on prior GWAS for five autoimmune diseases as well as for COVID-19 infection and post-COVID syndrome (“long COVID”). We validated seven previously described immune loci and identified 13 novel loci showing genome-wide significant (α=5.00E-8) association with different immune cell phenotypes. The most significant novel signal was conferred by the SLC52A3 locus, encoding for a riboflavin transporter protein, which was associated with naïve CD57+ CD8+ T cells (p=4.13E-17) and colocalized with SLC52A3 expression. Several novel loci contained immunologically plausible candidate genes, e.g., variants near TBATA and B3GAT1 representing genes associated with T cell phenotypes. The PRS of type 1 diabetes were significantly associated with CD8+ T cells at different differentiation states (p≤7.02E-4), and PRS of long COVID were associated with early-differentiated CD4+ T cells (p≤1.54E-4). In conclusion, our extensive immune cell GWAS analyses highlight several novel genetic loci of likely relevance for immune system function. Furthermore, our PRS analyses point to a shared genetic basis between immune cell distributions in healthy adults and T1D (CD8+ T cells) as well as long COVID (CD4+ T cells).

The immune system likely plays a key role in Parkinson’s disease (PD) pathophysiology. Thus, we investigated whether immune cell compositions are already altered in healthy individuals at high genetic risk for PD. We quantified 92 immune cell subtypes in the blood of 442 individuals using multicolor flow cytometry. Polygenic risk scores (PGS) for PD were calculated based on genome-wide significant SNPs (n = 87) from a large genome-wide association study (n = 1,530,403). Linear regression analyses did not reveal significant associations between PGS and any immune cell subtype (FDR = 0.05). Nominally significant associations were observed for NKG2C+ B cells (p = 0.026) in the overall sample. Older participants at increased genetic PD risk also showed a higher proportion of myeloid dendritic cells (p = 0.019) and CD27+CD4+ memory T cells (p = 0.043). Several immune cells were nominally statistically associated in women only. These findings suggest that major alterations of immune cells only occur later in the progression of PD.

The immune system plays a crucial role in the pathogenesis of neurodegenerative diseases. Here, we explored whether blood immune cell profiles are already altered in healthy individuals with a genetic predisposition to amyotrophic lateral sclerosis (ALS) or Alzheimer’s disease (AD). Using multicolor flow cytometry, we analyzed 92 immune cell phenotypes in the blood of 448 healthy participants from the Berlin Aging Study II. We calculated polygenic risk scores (PGSs) using genome-wide significant SNPs from recent large genome-wide association studies on ALS and AD. Linear regression analyses were then performed of the immune cell types on the PGSs in both the overall sample and a subgroup of older participants (>60 years). While we did not find any significant associations between immune cell subtypes and ALS and AD PGSs when controlling for the false discovery rate (FDR = 0.05), we observed several nominally significant results (p < 0.05) with consistent effect directions across strata. The strongest association was observed with CD57+ CD8+ early-memory T cells and ALS risk (p = 0.006). Other immune cell subtypes associated with ALS risk included PD-1+ CD8+ and CD57+ CD4+ early-memory T cells, non-classical monocytes, and myeloid dendritic cells. For AD, naïve CD57+ CD8+ T cells and mature NKG2A+ natural killer cells showed nominally significant associations. We did not observe major immune cell changes in individuals at high genetic risk of ALS or AD, suggesting they may arise later in disease progression. Additional studies are required to validate our nominally significant findings.