Hyaluronan levels are increased systemically in human type 2 but not type 1 diabetes independently of glycemic control

Abstract

Hyaluronan (HA), an extracellular matrix glycosaminoglycan, is implicated in the pathogenesis of both type 1 diabetes (T1D) as well as type 2 diabetes (T2D) and has been postulated to be increased in these diseases due to hyperglycemia. We have examined the serum and tissue distribution of HA in human subjects with T1D and T2D and in mouse models of these diseases and evaluated the relationship between HA levels and glycemic control. We found that serum HA levels are increased in T2D but not T1D independently of hemoglobin-A1c, C-peptide, body mass index, or time since diabetes diagnosis. HA is likewise increased in skeletal muscle in T2D subjects relative to non-diabetic controls. Analogous increases in serum and muscle HA are seen in diabetic db/db mice (T2D), but not in diabetic DORmO mice (T1D). Diabetes induced by the β-cell toxin streptozotozin (STZ) lead to an increase in blood glucose but not to an increase in serum HA. These data indicate that HA levels are increased in multiple tissue compartments in T2D but not T1D independently of glycemic control. Given that T2D but not T1D is associated with systemic inflammation, these patterns are consistent with inflammatory factors and not hyperglycemia driving increased HA. Serum HA may have value as a biomarker of systemic inflammation in T2D.

Keywords: Biomarker; Diabetes; Hyaluronan; T1D; T2D.

Figure 1:. Serum and muscle HA concentrations are not increased in a T1D mouse model.

A: Blood glucose of DORmO and BalbC mice from 4 to 20 weeks of age. B: Representative images of pancreatic islets from DORmO and BalbC mice stained for HA, in brown. C: Quantification of HA accumulation in pancreatic islets from IHC images. D: Representative images of skeletal muscle images from DORmO and BalbC mice stained for HA, in brown. E: Quantification of HA accumulation in skeletal muscle from IHC images. F: Quantification of HA content of skeletal muscle from DORmO and BalbC mice. G: Serum HA level from DORmO and BalbC mice. N = 5–10 mice per group. Data represent mean +/− SEM, *P < 0.05 vs the respective control by unpaired t test.

Figure 2:. Serum and muscle HA concentrations are significantly elevated in a T2D mouse model.

A: Blood glucose of db/db and C57Bl6 mice from 4 to 20 weeks of age. B: Serum HA level from db/db and C57Bl6 mice. C: Representative images of skeletal muscle images from db/db and C57Bl6 mice stained for HA, in brown. D: Quantification of HA accumulation in skeletal muscle from IHC images. E: Quantification of HA content of skeletal muscle from db/db and C57Bl6 mice. N = 5–10 mice per group. Data represent mean +/− SEM, *P < 0.05 vs the respective control by unpaired t test.

Figure 3:. STZ treatment induces hyperglycemia in mice but HA serum levels are not significantly elevated.

A: Blood glucose of C57Bl6 mice with and without STZ treatment. B: Representative images of pancreatic islets from C57Bl6 mice with and without STZ treatment. C: Quantification of HA accumulation in pancreatic islets from IHC images. D: Serum HA measurements from STZ treated C57Bl6 mice and their untreated controls. N = 5–10 mice per group. Data represent mean +/− SEM, *P < 0.05 vs the respective control by unpaired t test.

Figure 4:. Serum and muscle HA concentrations are not elevated in T1D patients.

A: Serum HA measurements in T1D and healthy control subjects. B: Relationship between serum HA and HbA1c. Mean serum HA values for each individual in the study are shown in conjunction with the HbA1c level measured for that given individual at the time of blood draw. C: BMI and (D) days from diagnosis are shown in correlation with HA. N = 1720 per group. Data in A represent mean +/− SEM, data in B-D are shown as regression plots. *P < 0.05 vs control by unpaired t test.

Figure 5:. Serum and muscle HA concentrations are elevated in T2D patients.

A: Serum HA measurements in T2D and healthy control subjects. B: Relationship between serum HA and HbA1c and (C) c-peptide. Mean serum HA values for each individual in the study are shown in conjunction with the HbA1c and c-peptide level measured for that given individual at the time of blood draw. D: BMI and (E) days from diagnosis are shown in correlation with HA. N = 17–20 per group. F: Representative images of muscle sections from healthy donors and T2D patients stained for HA, shown in brown. G: Quantification of HA muscle staining shown as % area fraction. N = 5 individuals per group. Data in A and G represent mean +/− SEM, data in B-E are shown as regression plots. *P < 0.05 vs control by unpaired t test.