Cross-sectional, interventional, and causal investigation of insulin sensitivity using plasma proteomics in diverse populations

Abstract

Background: We previously reported significant correlations between a direct measure of insulin sensitivity (IS) and blood levels of proteins measured using the Proximity Extension Assay (PEA) in two European cohorts. However, protein correlations with IS within non-European populations, in response to short-term interventions that improve IS, and any causal associations with IS have not yet been established.

Methods: We measured 1,470 proteins using the PEA in the plasma of 1,015 research participants at Stanford University who underwent one or more direct measures of IS. Association analyses were carried out with multivariable linear regression within and across Stanford subgroups and within each of the two European cohorts. Association statistics were also meta-analyzed after transformation and harmonization of the two direct measures of IS. Lastly, we performed genome-wide association studies of IS and used genetic instruments of plasma proteins from the UK Biobank to identify candidate causal proteins for IS through Mendelian Randomization (MR) analysis.

Results: In age and sex adjusted model, 810 proteins were associated with baseline IS among 652 self-reported European participants in the Stanford cohort at a false discovery rate (FDR) < 0.05. Effect sizes for these proteins were highly correlated with those observed in 122 South Asian, 92 East Asian, 85 Hispanic, and 52 Black/African American persons (r= 0.68 to 0.83, all P≤4.3×10^-113). Meta-analysis of the full Stanford cohort with the two European cohorts (N=2,945) yielded 247 significant protein associations (FDR < 0.05), with 75 remaining significant after further adjustment for body mass index. In a subset of Stanford participants undergoing insulin sensitizing interventions (N=53 taking thiazolidinediones, N=66 with weight loss), 79.6% of protein level changes were directionally consistent with the respective baseline association (observed/expected p=6.7×10^-16). MR analyses identified eight candidate causal proteins for IS, among which were SELE and ASGR1, proteins with established drug targets currently under investigation.

Conclusion: Plasma proteins measured using the PEA provide a robust signature for IS across diverse populations and after short-term insulin sensitizing interventions highlighting their potential value as universal biomarkers of insulin resistance. A small subset of markers provided insights into potential causal molecular mechanisms and therapeutic targets.

Keywords: Insulin sensitivity; Mendelian Randomization; causal inference; plasma protein; thiazolidinediones; weight loss.

Figure 1.. Comparison of regression coefficients for different populations in (a) BMI-unadjusted and (b) BMI-adjusted models, using White/Europeans in the Stanford study as reference.

The comparisons are based on the proteins associated with SSPG in Whites. Pearson correlations between each pair of populations are also reported.

Figure 2.. Manhattan plot of meta-analysis combining cross-sectional studies of plasma proteins and M values from RISC, ULSAM and Stanford.

Top panel shows the BMI-unadjusted model and the bottom panel shows the BMI-adjusted model. Proteins are color-coded according to their Olink panel. Red dotted lines indicate Bonferroni-corrected P-values (P < 0.05/1,471 proteins tested = 3.4×10⁻⁵) while blue dotted lines indicate FDR-corrected P-values. Proteins that passed Bonferroni correction are annotated, and the direction of effect of each protein on M values is indicated by the direction of the triangle.

Figure 3.. Correlation of regression Z-score between meta-analysis of cross-sectional studies and Stanford intervention study of (a) all participants, (b) those who underwent TZD intervention and (c) those who underwent weight loss intervention, using 255 candidate proteins identified in the meta-analysis of cross-sectional studies at FDR < 0.05.

Insulin sensitivity was measured as M values in meta-analysis of cross-sectional studies, and as SSPG in Stanford Intervention study. Proteins with significant associations at P < 0.05 were annotated.

Figure 4.. Mendelian randomization analysis of cis-pQTLs only of candidate causal proteins on M values.

The proteins shown are those with evidence of causal associations in MR primary analyses at P < 0.05. Candidate causal proteins supported by at least one sensitivity analysis (where available) at P < 0.05 are marked with an asterisk.