Type 2 diabetes and tooth loss are linked both epidemiologically and pathophysiologically. We applied label-free differential protein expression analysis using multidimensional liquid chromatography/tandem mass spectrometry (2D-LC-MS/MS) to explore the proteomic profile of saliva samples collected from selected type 2 diabetic edentulous patients and non-diabetic controls. Ninety-six peptides corresponding to 52 proteins were differentially expressed between the diabetic edentulous patients and controls (p < 0.05). Some diabetes-related inflammatory biomarkers including glyceraldehyde-3-phosphate dehydrogenase and serum amyloid A were detected with levels increased in diabetic samples. Other biomarkers including amylase, palate, lung and nasal epithelium associated protein (PLUNC), and serotransferrin levels were decreased in diabetic samples. In contrast with previous findings, salivary carbonic anhydrase 6 and alpha-2 macroglobulin levels, however, were decreased in this diabetic patient population. Cluster analysis and principle component analysis demonstrated a differential pattern of protein biomarker expression between diabetic and control subjects. Western blot analysis was completed to confirm the relatively lower expression level of two biomarkers, including PLUNC and amylase in the diabetic group compared to control subjects. The presence of salivary biomarkers specific for diabetes in edentulous subjects mimics those in serum, especially those related to inflammatory/lipid metabolism. While this exploratory study requires further validation with a larger population, it provides proof-of-principle for salivary proteomics for edentulous subjects with diabetes.