Student Theses and Dissertations


Wendy Chung

Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)

RU Laboratory

Leibel Laboratory


Non-insulin dependent diabetes mellitus (NIDDM) is perhaps the most common metabolic disorder in humans. In its essence, NIDDM is a disorder of glucose homeostasis which results from a relative deficiency of circulating insulin. Approximately 15% of all Americans over the age of 60 and 100 million individuals worldwide are affected by this disease whose micro- and macrovascular complications include blindness, nephropathy, neuropathy, and atherosclerosis. There is a close association between obesity and NIDDM, due in part to the increased insulin resistance imposed by increased body fat. The large genetic contribution to susceptibility to NIDDM is evidenced by the nearly 100% concordance rate for NIDDM in identical twins. The disease endpoint resolves (in a phenotypic sense) complex interactions of genes and environmental factors affecting glucose transport, production and pancreatic beta cell function. Due to these complexities, only a small number of the genes, usually in rare sub-types of NIDDM, have been identified in humans. Because inbred rodent strains are inherently more tractable genetically and can be employed in cases controlled for relevant environment, they have been used as a model system for studying the underlying genetic basis of NIDDM. Intercrosses of mouse (and rat strains) with divergent inherent diabetes susceptibilities in matings simultaneously segregating for one of four mouse obesity mutations (ob, db, fat, or tub) or the rat mutation fa were used. Obese F2 progeny, as expected, demonstrated widely varying degrees of obesity and diabetes. The obese progeny were characterized at approximately 100-120 days of age for body weight, body mass index, fasting plasma [glucose] and [insulin], HbAlc, pancreatic hormone content and pancreatic islet morphology using sections immunohistochemically stained for glucagon or insulin. These phenotypes intentionally emphasized the identification of genes which ultimately affect pancreatic beta cell function because primary beta cells defects have been associated with single gene mutations causing diabetes in humans. By selectively genotyping the most extremely affected 5% of the progeny with markers spaced approximately 20 cM apart throughout the genome, regions of the genome demonstrating deviation from the expected Mendelian ratios were identified and subsequently tested in all obese F2 progeny for association with obesity and diabetes phenotypes. When possible, additional genetic markers were used to more precisely position the quantitative trait loci (QLTs). QTLs identified in one set of F2 progeny were tested for replicability either in F3 or N2F1 progeny of the same cross or in other F2 progeny segregating for the same or different obesity mutations between the same of other parental strains. Using this technique in seven sets of obese F2 progeny and two set of obese F3 progeny, 23 separate QTLs on 18 mouse autosomes and two QTLs on two rat autosomes relating either to obesity and/or diabetes have been identified and (for the most part) replicated in multiple crosses. Candidate genes for the QTLs have been identified and in some cases directly evaluated by screening for mutations or allelic variants. Developmental studies have indicated the likely times of expression of four of the QTLs as well as likely points in NIDDM pathogenesis at which the QTLs act. Introgession techniques and strategies for cloning the genes underlying the QTLs and testing their relevance in human populations are described. Finally, evidence for the potential role of mitochondrial genes and the unanticipated differences of obesity penetrance in ob/ob and db/db mice are presented.


A thesis presented to the faculty of The Rockefeller University in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Included in

Life Sciences Commons