Cindy Boer

236 | Chapter 5.2 Table 2: Results of multivariate linear regression analysis of gut microbiome relative abundancies and knee WOMAC-pain scores, in the Rotterdam Study, LifeLines-DEEP Rotterdam Study (RS) Taxonomy % N CoE SE p-value FDR Class Bacilli 27.3% 1,419 6.1 x10 -03 1.0 x10 -03 9.1 x10 -09 1.0 x10 -06 Order Lactobacillales 100% 1,417 6.1 x10 -03 1.1 x10 -03 7.6 x10 -09 1.0 x10 -06 Family Streptococcaceae 79.6% 1,402 4.9 x10 -03 9.3 x10 -04 1.5 x10- 07 8.7 x10 -06 Genus Streptococcus 98.7% 1,396 5.0 x10 -03 9.3 x10 -04 7.3 x10- 08 5.6 x10 -06 Life Lines Deep (LLD) Meta- Ana;ysis (RS + LLD) % N CoE SE p-value N p-value Class Bacilli 27.3% 867 5.4 x10 -03 1.3 x10 -03 3.6 x10 -05 2,286 1.1x10 -12 Order Lactobacillales 100% 864 4.9 x10 -03 1.3 x10 -03 2.4 x10 -04 2,281 8.3x10 -12 Family Streptococcaceae 79.6% 863 2.9 x10 -03 1.3 x10 -03 2.3 x10 -02 2,265 2.1x10 -08 Genus Streptococcus 98.7% 860 3.3 x10 -03 1.6 x10 -03 3.7 x10 -02 2,256 1.3x10 -08 Adjusted for age, sex and, technical covariates: DNA isolation batch and TimeInMail. RS: Rotterdam Study (n=1427) P-value, CoE and SE from MaAsLin, LLD: LifeLines-DEEP (n=867) P-value, CoE and SE fromMaAsLin analysis, Meta: Rotterdam Study and LifeLines Deep meta-analyzed together, sample size weighted inverse-variance meta-analysis in METAL. Taxonomy%=percentage of taxonomy is from one taxonomy level higher, ex. 23.7% of all Firmicutes are Bacilli. N=number of individuals in cohort where microbial abundancy is not zero for that taxonomy. FDR: P-value adjusted for multiple testing, Benja- min-Hochberg false discovery rate. CoE coefficient, SE standard error Quantitative determination of Streptococcus spp As 16S rRNA-sequencing derived relative microbiome data cannot provide information about the extent or directionality of changes in microbiome taxa abundance[30], we de- termined the absolute amount of Streptococcus spp. in the individuals of our study pop- ulation. For each sample in our cohort (n=1,427), we quantified the number of Strepto- coccus spp. using genus specific qPCR and the total microbial load using 16S rRNA qPCR. We calculated the absolute quantity of Streptococcus spp. and normalized for the total bacterial load in each samples as measured by 16S rRNA qPCR. The 16S rRNA-sequenc- ing results and qPCR Streptococcus spp. quantity yielded similar results (Spearman correlation, r =0.80, p-value=2.2×10 −16 , Supplementary Figure 2 ). Using the absolute abundance of Streptococcus measured by qPCR instead of the relative abundance de- rived from the 16S rRNA-sequencing profiles, we again found a significant association between higher knee WOMAC pain and greater absolute Streptococcus spp. abundance (beta=0.10, p-value=7.4×10 −03 , Poisson regression), also after adjustment for smoking, alcohol consumption, and BMI (beta=0.074, p-value=4.5×10 −02 , Poisson regression).