Vaccines alter risk of diabetes in animals
Published data indicates that immunization starting at birth with common pediatric vaccines such as the DTP vaccine can prevent the development of diabetes in NOD mice (Mice Graph 1) and BB rats (Rat Graph 1) as well as lupus in MRL mice (Classen,JB, 1996c). Additional studies showed immunization of NOD mice with the pertussis vaccine starting at 2 months of life is associated with a tripling of the risk of diabetes (Mice Graph 2) (Classen,JB, 1996c). Interestingly if the pertussis vaccine is given starting at birth then it can be given at 2 months without an increased risk of diabetes. In both sets of experiments an effect could be seen with very low doses of vaccine, comparable to a human dose but decreased proportionally for a rodent.
Information reviewed below indicates the rodent models of diabetes may be a good model of human type I diabetes.
Rodent Models of Type I Diabetes
Nonobese diabetic, NOD, mice spontaneously develop insulin dependent diabetes and are a model of type I diabetes mellitus. Disease in these animals, as in humans, appears to be of a autoimmune etiology that is heavily influenced by both genetics and environment. Substantial evidence exists that observations made in these animals have potential clinical relevance to human type I diabetes. Much of this data has been reviewed in several recent publications (Kolb, 1987; Rossini, Handler, Greineret al.1991; Kikutani & Makino, 1992) so only a few key points are discussed below.
There are many similarities between diabetes in NOD mice and humans. Disease onset occurs commonly in young mice which develop hyperglycemia, glycosuria, polyuria, ketonuria, hypoinsulinemia and death results in 1-2 months if the mice do not receive insulin. Insulitis develops in the pancreas before the development of diabetes. Animals develop glutamic acid decarboxylase autoantibodies, insulin autoantibodies and islet cell autoantibodies. Diabetes can be prevented by the administration of immunosuppressants like cyclosporine during the prediabetic period. Genetic experiments show that diabetes is closely linked to the MHC class II genes in NOD mice as it is in humans. Some NOD mice develop autoimmune thyroiditis that is casually (Pontesilli, Carotenuto, Gazdaet al.1987; Bernard, Ertug & Margolese, 1992).
Experiments have been performed in NOD mice to provide further evidence supporting an autoimmune etiology of insulin dependent diabetes. Diabetes can be passively transferred by purified lymphocytes while neonatal thymectomy or anti-lymphocyte antibodies can prevent diabetes. Purified lymphokines like TNF, and IL-1 can also prevent diabetes (Jacob, Aiso, Michieet al.1990; Formby, Jacobs, Dubucet al.1992). Researchers have found evidence of retroviral particles in the beta cells of NOD mice that may explain the initiation of the insulitis (Suenaga & Yoon, 1988; Gaskins, Prochazka, Hamaguchiet al.1992).
Attempts have been made at locating immunological disorders in the NOD mice that may predispose them to develop diabetes however like humans, the search has been mostly inconclusive to date. The most profound defect is the failure of the mice to express the MHC class II I-E alleles. There is data suggesting total number of lymphocytes may be transiently reduced in NOD mice before 160 days (Pontesilli, Carotenuto, Gazdaet al.1987). Data also exists that there may be a defect in Lyt-2 "suppressor" cells (Hatamori, Yokono, Nagataet al.1989)
The similarity between diabetes in the NOD mice and humans is extensive however the influence of sex on the incidence of disease is a striking difference. The incidence of diabetes in female mice at 6 months is about 80% compared to 20% in males. This difference is not seen in all substrains of NOD mice however it occurs in most strains throughout the life of the animals. The sex difference can be explained by a sensitivity of the immune system to sex hormones. Castration increases the incidence of diabetes in male mice, ovariectomy decreases the incidence in female mice, while administration of testosterone decreases the incidence of diabetes (Fitzpatrick, Lepault, Homo-Delarcheet al.1991; Fox, 1992; Hawkins, Gala & Dunbar, 1993). Male sex hormones may be exerting an inhibitory effect on lymphocytes because thymectomy on day 3 of life increases the incidence of diabetes in males and decreases difference in incidence between the sexes (Classen,JB & Shevach, 1993).
BioBreeding, BB, rats also develop spontaneous insulin dependent diabetes and are a second rodent model of type I diabetes mellitus. The rats were discovered in 1974 at the BioBreeding Laboratories in Ottawa, Canada. Several review papers have been published on this model (Mordes, Desemone & Rossini, 1987; Parfrey, Prud'homme, Colleet al.1989; Crisa, Mordes & Rossini, 1992) and only a few findings are discussed below.
The development of diabetes in BB rats is very similar to that in humans and NOD mice. Many of the findings present in human type I diabetics and in NOD mice are found in BB rats leading many to believe diabetes in BB rats is also a autoimmune disorder. Insulitis develops in the pancreas of BB rats before the development of diabetes while antibodies develop to islet cells and possibly to insulin. Diabetes can be prevented by neonatal thymectomy as well as administration during the prediabetic period with cyclosporine, anti-lymphocyte antibodies, or purified lymphokines like TNF. Genetic experiments show that diabetes is closely linked to the MHC class II genes in BB rat as it is in humans. Many older rats develop autoimmune thyroiditis that is casually related to the development of diabetes.
The BB rats differ from the NOD mice in several ways. Male rats develop diabetes at a rate equal to females, as happen in humans but not NOD mice. The incidence of diabetes in the BB rat is not affected by gonadectomy, or the administration of androgens (Verheul, Schot & Schuurs, 1986). In contrast to humans and NOD mice, the BB/Wor rats, the most commonly used substrain, is severely lymphopenic. There is a marked decrease in mature T lymphocytes in peripheral blood, spleen and lymph nodes. The CD4+ subset is substantially reduced but the CD8+ subset is almost completely absent. Natural killer cells are relatively over expressed. The lymphopenia in BB/Wor rats may be in part due to the lack of cells expressing the RT6 antigen. Half of the CD4+ and two thirds of the CD8+ lymphocytes are RT6 positive in normal rats strains and special strains of BB rats that are diabetes resistant (Greiner, Mordes, Handleret al.1987). The lymphopenia may explain their susceptibility for certain infections, their propensity for developing lymphomas, and the altered responses seen in cell cultures. Interestingly the BB/Edinburgh, another substrain of BB rats, is not lymphopenic but still develops diabetes (Varey, Dean, Walker et al.1987).