Renal nerve ultrastructural alterations in short term and long term experimental diabetes

Abstract

Background: Despite the evidence that renal hemodynamics is impaired in experimental diabetes, associated with glomeruli structural alterations, renal nerves were not yet investigated in experimental models of diabetes and the contribution of nerve alterations to the diabetic nephropathy remains to be investigated. We aimed to determine if ultrastructural morphometric parameters of the renal nerves are affected by short term and/or long term experimental diabetes and if insulin treatment reverses these alterations. Left renal nerves were evaluated 15 days or 12 weeks (N = 10 in each group) after induction of diabetes, with a single injection of streptozotocin (STZ). Control rats (N = 10 in each group) were injected with vehicle (citrate buffer). Treated animals (N = 10 in each group) received a single subcutaneous injection of insulin on a daily basis. Arterial pressure, together with the renal nerves activity, was recorded 15 days (short-term) or 12 weeks (long-term) after STZ injection. After the recordings, the renal nerves were dissected, prepared for light and transmission electron microscopy, and fascicle and fibers morphometry were carried out with computer software.

Results: The major diabetic alteration on the renal nerves was a small myelinated fibers loss since their number was smaller on chronic diabetic animals, the average morphometric parameters of the myelinated fibers were larger on chronic diabetic animals and distribution histograms of fiber diameter was significantly shifted to the right on chronic diabetic animals. These alterations began early, after 15 days of diabetes induction, associated with a severe mitochondrial damage, and were not prevented by conventional insulin treatment.

Conclusions: The experimental diabetes, induced by a single intravenous injection of STZ, in adult male Wistar rats, caused small fiber loss in the renal nerves, probably due to the early mitochondrial damage. Conventional treatment with insulin was able to correct the weight gain and metabolic changes in diabetic animals, without, however, correcting and / or preventing damage to the thin fibers caused by STZ-induced diabetes. The kidney innervation is impaired in this diabetic model suggesting that alterations of the renal nerves may play a role in the development of the diabetic nephropathy.

Figure 1

Semithin transverse sections of renal nerves from acute control (A), chronic control (B), acute diabetic (C), chronic diabetic (D), acute diabetic treated with insulin (E) and chronic diabetic treated with insulin (F) animals. The nerve fascicles are enveloped by a well defined perineurium (arrowheads) even when more than one fascicle is present (C). Most of the endoneural space is occupied by unmyelinated fibers but few myelinated fibers (arrows) are present. No morphological differences were evident between groups at this resolution. V = capillary vessels, S = Schwann cell nucleus. Toluidine blue stained, bars = 10 μm.

Figure 2

Transmission electron microscopy cross sections of renal nerves from acute control (A), chronic control (B), acute diabetic (C), chronic diabetic (D), acute diabetic treated with insulin (E) and chronic diabetic treated with insulin (F) animals. No ultrastructural alterations were evident on the controls that show typical Schwann cell units. Myelinated fibers are indicated by “m”. Acute diabetic nerves showed mild destruction of axonal mitochondria (arrows) and unmyelinated fiber enlargement with signs of swelling and degeneration (*). Chronic diabetic nerves presented with severe destruction of axonal mitochondria (arrows), micro-axons formation (arrowheads) and thickening of the endoneural space with an increased density of collagen fibers. Insulin treated animals showed alterations of the endoneural blood vessels such as thickening of the basement membrane (*) and lumen reduction at early stages (E). In chronic stages, very small unmyelinated axons were present (arrowheads), edema of the endoneural space could be evidenced and unmyelinated fibers with atrophy (a) were frequent. Bar = 1 μm for B, C, D and F. Bar = 2 μm for A and E.

Figure 3

Myelinated fibers diameter distributions on acute (A) and chronic (B) treated and untreated diabetic groups and respective controls. All fibers from all nerves are represented. For all experimental groups, myelinated fibers distributions are bimodal in shape, with a clear point of separation between small and large fibers at 6.0 μm. The acute control group shows an homogeneous distribution of the small fibers while both diabetic groups, treated and untreated showed a larger percentage of small myelinated fibers, shifting the histograms to the left (One Way ANOVA on Ranks, p < 0.001). The chronic diabetic group also shows a shift of the distribution towards the small fibers, with a disorganization of the large fibers distribution (One Way ANOVA on Ranks, p < 0.001). The chronic treated group showed a reduced percentage of both small and large myelinated fibers (One Way ANOVA on Ranks, p < 0.001).

Figure 4

Myelinated fibers G ratio on acute (A) and chronic (B) treated and untreated diabetic groups and respective controls. All fibers from all nerves are represented. For all experimental groups the distributions are unimodal, with peaks at 0.6. Note that on acute diabetic group there is a slight shif of the distribution towards 1.0, suggestive of demyelination. This shift is more evident on the chronic diabetic groups, treated and untreated (One Way ANOVA on Ranks, p < 0.001).

Figure 5

Unmyelinated fiber distribution on acute (A) and chronic (B) treated and untreated diabetic groups and respective controls. Average values are represented. The distributions are unimodal in all groups with the peak at 0.7 μm for acute groups and chronic control group. For the chronic diabetic groups, treated and untreated, there is a shift (One Way ANOVA on Ranks, p < 0.001) of the peak to 0.9 μm, indicating a loss of the small fibers.