Successful Islet Transplantation Does Not Prevent the Development of Neuropathy in Alloxan-Induced Diabetic Rats C.T. Spadella, J.L.M. Machado, C.A. Caramori, and E.A. Grego´ rio P ERIPHERAL neuropathy is one of the most common secondary complications of diabetes mellitus, causing severe and prolonged morbidity. 1,2 However, clinical and experimental studies have reported that careful glucose control may prevent, stabilize, and/or reverse neuropathy and other chronic diabetic complications. 3–5 Unfortunately, insulin therapy does not prevent the development or progression of chronic lesions in the ves- sels, kidneys, eyes, or nerves of the diabetic patient. 6–8 As a result, there is great interest in investigating other forms of endocrine pancreas-replacement therapy, such as trans- plantation of the whole pancreas or of the islets of Lang- erhans (IT). Pancreas transplantation has been performed for type I diabetic patients with advanced disease. The procedure corrects the metabolic abnormalities of diabetes and has been reported to control the secondary complications of diabetes. 9,10 However, pancreas transplant implies a major operation with several potential complications that pose important risks for the patient. Previous studies in our laboratory have shown that good long-term metabolic control of diabetes is achieved in alloxan-induced diabetic rats by IT. 11 Based on the hypoth- esis that perfect glucose homeostasis is possible, we inves- tigated whether IT could prevent the development of lesions in somatic peripheral nerves, hoping to contribute to a better understanding of diabetes and its treatment. MATERIAL AND METHODS Animals and Groups One hundred and fifty inbred male Lewis rats, approximately 3 months old, were randomly assigned to three experimental groups of 50 specimens each, coded and handled as follows: NC, nondia- betic control rats; DC, untreated diabetic control rats; and IT, diabetic rat recipients of pancreatic islet transplants prepared by collagenase digestion of glands from normal synergeneic donor Lewis rats and injected into the portal vein. Each group was further divided into five subgroups of 10 rats each that were sacrificed at 1, 3, 6, 9, and 12 months of follow-up. Diabetes was induced by a single intravenous administration of alloxan (42 mg/kg body weight Sigma Chemical Co, St. Louis, Mo, USA). Only rats showing a severe diabetic state were included in the experiment. Islets of Langerhans were prepared by collagenase according to the proce- dure originally described by Moskalewski 12 and subsequently mod- ified. 13 About 1500 cells were injected into the portal vein of IT rats 14 days after alloxan administration. Clinical and Laboratory Analysis Seven days before as well as 4 days and 1, 3, 6, 9, and 12 months after IT, rats were housed in metabolic cages for 24 hours to measure body weight, food and water intake, urine output, blood and urine glucose levels, and plasma insulin. Blood and urine glucose were determined by the enzymatic method (Celm Lab, Sa ˜o Paulo, Brazil) and plasma insulin levels by radioimmunoassay in solid phase (DPC, Los Angeles, Calif, USA). Tissue Preparation After 1, 3, 6, 9, and 12 months of follow-up, animals from the three experimental groups were killed. The right sciatic nerves of five rats in each subgroup were removed and processed for examination under light and electron microscopy. For histology tissue fragments from the sciatic nerve were fixed in 2.5% glutaraldehyde for 24 hours and routinely stained with hematoxylin-eosin and metilen. For electron microscopy the fragments were postfixed in 1% osmium tetroxide in 0.1 mol/L phosphate buffer at pH 7.4 for 1 hour. The tissue was then dehydrated in acetone and embedded in epoxy resin (Araldite). Two blocks of nerves were obtained from each animal, and ultrathin sections were prepared and stained with uranyl acetate and bismuth subnitrate. Morphometric Study The following parameters were assessed for morphometric analy- sis: the number of myelinated and non-myelinated nerve fibers, the cross-sectional areas of myelinated axons, the degree of axonal demyelination, and the density of intraaxonal glycogen deposits. To count the number of myelinated nerve fibers, five randomly se- lected, nonoverlapping fields from each of the two blocks were analyzed in a blind study. To calculate the cross-sectional areas of the myelinated axons, thin sections were analyzed in a immersion photomicroscope at a final magnification of 1440. The number and areas of nerve fibers were measured through a Digital Image From the Departments of Surgery (C.T.S., J.L.M.M.), Medical Clinic (C.A.C.), and Morphology (E.A.G.), School of Medicine, State University of Sao Paulo (UNESP), Botucatu, SP, Sa˜o Paulo, Brazil. This work was supported by FAPESP. Address reprint requests to C.T. Spadella, MD, Faculdade de Medicina de Botucatu—UNESP, Departamento de Cirurgia, 18618-970, Botucatu, Sa˜ o Paulo, Brasil. 0041-1345/02/$–see front matter © 2002 by Elsevier Science Inc. PII S0041-1345(02)02769-0 655 Avenue of the Americas, New York, NY 10010 1296 Transplantation Proceedings, 34, 1296 –1300 (2002)