Introduction: With an estimated prevalence of 425 million individuals, type 2 diabetes (T2D) is a major challenge to health care systems worldwide. It is a multifactorial complex disease characterized by insulin resistance and hyperinsulinemia, and involves widespread complications. We performed hypothesis-free association and Mendelian randomization (MR) based causal inference analyses of T2D, investigating 1,100 plasma proteins in cross-sectional and longitudinal settings. Methods: The population-based cohort studies KORA (Cooperative health research in the Region of Augsburg) and HUNT3 (Third wave of the Nord-Trøndelag Health Study) were used for discovery and replication, respectively. Proteins were quantified using the SOMAscan proteomics platform in 999 KORA and 1007 HUNT participants. After quality control, we log2-transformed and (0,1)-standardized the proteomics data. We validated T2D using clinical and oral glucose tolerance test data in KORA, and clinical data only in HUNT3. We computed logistic regression models adjusted for age, sex, body mass index, smoking status and hypertension status and accounted for multiple testing using the Benjamini-Hochberg false discovery rate (FDR) method (FDR<0.05). Subsequently, we used two-sample bi-directional MR to infer the causal relationship between T2D and the replicated proteins. Results: The cross-sectional analysis with prevalent T2D (KORA 110 cases and HUNT 149) yielded 24 replicated proteins, of which the three yielding the highest odds ratios were aminoacylase 1, complement C2 and plasma protease C1 inhibitor. Longitudinal analysis with incident T2D (KORA 74 cases and HUNT 103) yielded three replicated proteins namely aminoacylase-1, growth hormone receptor and insulin like growth factor binding protein 2. The MR analysis testing the causal effect of T2D on the replicated proteins, using 120 SNPs as instrumental variables (IV), yielded nominally significant p-values (p<0.05) for cathepsin Z and renin. The estimated causal effects went into the same direction as the results for both proteins suggested. In the MR analyses testing the causal effects of the proteins on T2D, using cis-acting SNPs as IVs, sex hormone-binding globulin (SHBG) had nominally significant causal effect that went into the same direction as its result suggested. None of the MR analysis results was statistically significant at the FDR threshold. Conclusion: Using an aptamer-based technique, we replicated previously reported prevalent and incident T2D-protein associations, including SHBG, complement C2 and renin as well as identified novel ones, like aminoacylase-1. Causality analyses using MR identified SHBG as a potential protein directly involved in the pathogenesis of T2D, and suggested a causal effect of T2D on both cathepsin Z and renin, both known to have roles in the pathophysiological pathways of cardiovascular disease.