help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2005-0853
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
147/5/2287    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Luo, N.
Right arrow Articles by Chua, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Luo, N.
Right arrow Articles by Chua, S., Jr.
Endocrinology Vol. 147, No. 5 2287-2295
Copyright © 2006 by The Endocrine Society

Allelic Variation on Chromosome 5 Controls ß-Cell Mass Expansion during Hyperglycemia in Leptin Receptor-Deficient Diabetes Mice

Na Luo, Shun Mei Liu, Hong Liu, Qiong Li, Qun Xu, Xi Sun, Brandi Davis, Jing Li and Streamson Chua, Jr.

Institute of Human Nutrition (N.L.) and Division of Molecular Genetics (H.L., Q.L., Q.X., X.S., B.D., J.L.), Department of Pediatrics, Columbia University, New York, New York 10032; and Division of Endocrinology (S.M.L., S.C.), Department of Medicine, Albert Einstein College of Medicine, New York, New York 10461

Address all correspondence and requests for reprints to: Streamson Chua, Jr., M.D., Ph.D., Belfer 701, Division of Endocrinology, Department of Medicine, Albert Einstein College of Medicine, 1300 Morris Park Avenue, New York, New York 10461. E-mail: schua{at}aecom.yu.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Leptin signaling is a critical component of normal insulin sensitivity. Overt hyperglycemia and type 2 diabetes mellitus can be manifested in states of leptin signaling deficiencies by the additional effects of other genetic factors. We have previously described the contrasting insulin sensitivities and glycemic states of two congenic diabetes (db/db) mouse strains. C57BL/6J db/db mice have mild insulin resistance and achieve euglycemia with mild hyperinsulinemia. FVB db/db mice have severe insulin resistance and are hyperglycemic despite escalating hyperinsulinemia with expanded pancreatic ß-cell mass. Analysis of obese progeny from the two reciprocal backcrosses suggests that genetic modifiers for insulin sensitivity are separable from loci that modulate ß-cell mass. A genome scan of the backcross to FVB suggests that one or more modifier genes are present on chromosome 5. This evidence is supported by the phenotypes of multiple incipient congenic strains wherein the hyperglycemia observed in obese FVB mice is reproduced. With similar degrees of hyperglycemia in obese mice of these strains, the haplotype at chromosome 5 is associated with ß-cell mass and circulating insulin concentrations. Finally, we offer arguments that production of multiple incipient congenic lines is an economical alternative to the production of speed congenic strains.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
AS DIABETES MELLITUS is a disease of the endocrine pancreas, numerous approaches have been used to advance our understanding of the regulation of pancreatic islet biology. The developmental approach tackled by examining the transcriptional regulation of islet-specific genes has identified numerous genes that are required for pancreatic development and islet differentiation (1, 2). A cell biological approach showed that islets have a slow rate of turnover of their constituent cells and that expansion of islet mass can occur by hyperplasia of existing islets and by recruitment of new islets from precursors that probably exist within pancreatic ducts (3, 4). This rich trove of information has led to genetic manipulations of mice that cause expansion of ß-cell mass or reduction of ß-cell mass—cyclin-dependent kinase 4 (Cdk4) (5), inducible cAMP early repressor {gamma} (6), hepatocyte growth factor (7, 8), PTHrP (9), cyclin D1 (10), among others. However, it is likely that subtle genetic variants, rather than complete deficiencies or massive overexpression, is responsible for a majority of diabetes in humans. Searches for human diabetogenic variants have been successful with the identification of most causes of maturity onset diabetes of the young (MODY) (11), as well as the identification of variation within calpain 10 (12), affecting diabetes susceptibility. It is notable that variation in the intronic sequences of the calpain 10 gene is linked to diabetes susceptibility.

Pathological mutations of the leptin/leptin receptor system cause a syndrome of early onset hyperphagia, nutrient partitioning favoring fat deposition, insulin resistance, and hypothalamic hypogonadism (13). Each aspect of this syndrome is affected by modifier genes, and this phenotypic variability was recognized by Coleman et al. (14) even in the earliest reports of the obese (ob, Lepob) and diabetes (db, Leprdb) mutations.The typical phenotype of these mutations is exemplified by the prototypical C57BL/6J strain wherein ob/ob and db/db mice show extreme obesity, near euglycemia due to hyperinsulinemia, and infertility of both sexes. In the C57BLKS/J strain, db/db mice (15) show obesity but the degree of obesity is limited by hypoinsulinemia and ß-cell loss with concomitant hyperglycemia. Obese mice of the FVB strain (both FVB db/db and FVB ob/ob mice) develop hyperglycemia and extreme hyperinsulinemia. FVB db/db mice develop obesity of a magnitude similar to that attained by C57BL/6J db/db mice due to their compensatory hyperinsulinemia despite the overt hyperglycemia. Thus, the extreme variability of phenotypic expression of diabetes-related phenotypes caused by defects in leptin signaling provide a rich resource to examine the genetic basis of phenotypic variability in diabetes and diabetes complications.

We have previously generated mice with leptin signaling deficiencies (FVB ob/ob and FVB db/db) on the FVB strain (16, 17). The enlarged islets and expanded ß-cell mass of hyperglycemic FVB db/db mice are unique traits in diabetes models because it is generally accepted that hyperglycemia causes death/loss of ß-cells and the maintenance of an expanded ß-cell mass over many months offers an opportunity to examine the genetic basis of this phenotype. Our data suggest that the apparent resistance of the FVB db/db mouse’s ß-cells to glucotoxicity is a genetically determined trait controlled by a region on mouse chromosome 5. This evidence is based on a genome wide scan of backcross progeny and is supported by data from multiple incipient congenic strains (MICS) carrying various haplotypes at chromosome 5.

We discuss certain features of multiple incipient congenic strains that make MICS an economical yet powerful alternative to the development of speed congenic strains or full congenic strains. Typically, incipient congenic strains are developed to test the phenotype conferred by a specific phenotype. However, a significant drawback is the time required to produce a full congenic strain (10 backcross generations) or the expense required to generate a speed congenic strain. The development of two or more independent incipient congenic strains permits the analysis of robust effects of haplotypes with minimal increases in expense and effort. We also discuss the utility of a second type of incipient congenic strain that is homozygous for the haplotype of the host strain. This strain is readily developed at any generational level during the development of the incipient congenic strains and serve as the appropriate counterstrain for analysis of the incipient congenics that are homozygous for the haplotype of the donor strain. The use of MICS maintains the simplicity of breeding conventional congenic strains while avoiding the labor-intensive effort and expense of large-scale genotyping that is entailed in the production of speed congenic lines. Finally, further backcrossing of the MICS with subsequent narrowing of the critical interval provides the means for fine mapping and eventual identification of the genetic variants involved in differential strain responses to hyperglycemia.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Animals
The FVB-db and B6-db3J congenic strains have been previously described (17). Briefly, the db allele from the C57BLKS/J strain was backcrossed to the FVB/N strain for 10 generations. Whereas the original C57BLKS/J db/db mice are hyperglycemic due to the development of hypoinsulinemia and loss of ß-cell mass, the hyperglycemia of FVB db/db appears to be due to severe insulin resistance with continual increases in insulin secretory capacity from ß-cell mass expansion. The db3J allele, originating on the 129/J strain, was backcrossed to the C57BL/6J-Aw strain for 10 generations. The original 129 db3J/db3J mice characteristically have early hyperglycemia that switches to hypoglycemia due to massive islet expansion and hyperinsulinemia (18). Backcrosses between the two strains were performed with female F1 animals. All mice were maintained on Picolab diet 5058 (PMI Nutrition International, Richmond, IN) with a minimum of 9% fat content by weight. Glucose was measured on whole blood obtained from tail vein nicks using the glucose oxidase method (Glucometer; Bayer HealthCare AG, Elkhart, IN). The models used for glucose determinations had an upper limit of 600 mg/dl and any glucose values above this range are recorded as 600 mg/dl in this report. Insulin was measured by ELISA (ALPCO Diagnostics, Windham, NH) from serum samples of ad libitum-fed sampled blood. Because LEPR-null mice are hyperphagic, a randomly obtained blood sample would most likely be post-prandial. Mice were killed by carbon dioxide asphyxiation and organs were harvested for histological examination and for DNA analysis. Backcrosses to generate the incipient congenic strains were done by mating males with the desired genotypes to FVB mates, the desired genotypes being heterozygous for db3J and heterozygous for three markers that spanned mouse chromosome 5 (D5Mit233 at 19.7 cM; D5Mit24 at 45.9 cM; D5Mit43 at 77.6 cM). Subsequent testing of the N5 and N4 generations with markers within the interval showed no double crossovers, suggesting that the presence of FVB alleles within the introgressed interval is highly unlikely. The markers used were as follows: D5Mit309; D5Mit208; D5Mit239; D5Mit277; D5Mit240; D5Mit24; D5Mit158; D5Mit188; D5Mit431; D5Mit213; D5Mit168; and D5Mit169. All procedures were approved by the Columbia University Institutional Animal Use and Care Committee and conformed to guidelines established by the U.S. National Institutes of Health for humane vertebrate animal use.

Insulin assays were performed with a mouse insulin ELISA kit (Alpco Diagnostics, Salem, NH). Due to the potential for altered insulin processing due to increased insulin secretory demand, the sera of lean and FVB db/db mice were analyzed for the presence of proinsulin. Serum samples were processed in a manner similar to isolated islets and pancreata for insulin content (19)—material that was soluble in acidified alcohol was lyophilized followed by solubilization and fractionation by HPLC (20). Fractions were analyzed for insulin immunoreactivity by RIA with a hamster antiinsulin serum that recognizes proinsulin (40% cross-reaction). The column was also run with radiolabeled insulin and proinsulin from a mouse ß-cell line, ßTC3 to determine their respective elution profiles. All material from lean and db/db mice coeluted with mature insulin, whereas there was no immunoreactive material that coeluted with proinsulin.

Genotype analysis
DNA was isolated from ear clips or organs obtained after euthanasia. Allelic constitution at various microsatellite markers was determined by Taq polymerase-mediated amplification and agarose gel electrophoresis of the amplicons. Alleles were scored by comparison to amplicons derived from C57BL/6J and FVB/NJ DNA.

The markers used for the genome scan are listed: D1Mit76, D1Mit80, D2Mit450, D3Mit164, D3Mit49, D3Mit169, D3Mit17, D3Mit116, D4Mit206, D5Mit233, D5Mit24, D5Mit43, D6Mit9, D6Mit74, D8Mit4, D8Mit190, D9Mit75, D10Mit3, D10Mit13, D11Mit250, D12Mit2, D12Mit11, D13Mit3, D14Mit121, D15Mit15, D16Mit4, D16Mit19, D17Mit45, D18Mit188, D18Mit202, and D19Mit41.

Pancreatic islet analyses
Pancreatic sections were immunostained for insulin (17). Images were obtained with an Eclipse 400 microscope (Nikon Instruments, Inc., Melville, NY) equipped with a Spot Insight digital imaging system (Diagnostic Instruments, Sterling Heights, MI). Islet cellularity was determined by manual counting of insulin-positive cells with a minimum of 15 islets per mouse. Islet areas were calculated with Image Pro Plus (Media Cybernetics, Silver Spring, MD) from an entire section with a minimum of 15 islets per mouse.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Hyperglycemic FVB db/db mice have hyperinsulinemia and expanded ß-cell mass
We characterized the diabetes phenotype of LEPR-deficient mice on the FVB inbred strain at 3 months of age (Fig. 1Go). LEPR-deficient mice of this strain exhibit long-term hyperglycemia along with high circulating concentrations of insulin that is supported by an expansion of pancreatic ß-cells. This is in contrast to the diabetes phenotypes of the two prototypic strains: 1) hyperglycemia and relative hypoinsulinemia of LEPR-deficient mice of the C57BLKS/J-db strain; and 2) near euglycemia and hyperinsulinemia of LEPR-deficient mice of the C57BL/6J-db3J strain. The diabetes phenotype of the C57BL/6J db3J/db3J mice are identical with the phenotype of C57BL/6J db/db mice, despite the fact that the db mutation results in the loss of only the LEPR-B isoform, whereas all membrane bound isoforms of LEPR are lost due to a 17-bp exonic deletion in the db3J mutation. Consideration of the data in Fig. 1Go would suggest that the endocrine pancreas of FVB-db and C57BL/6J-db3J mice is responding to insulin resistance and hyperglycemia by expansion of total ß-cell mass. However, there is a difference in the glucose concentrations achieved for the respective insulin concentrations. With lower insulin concentrations, the C57BL/6J db3J/db3J mice are able to achieve near euglycemia, whereas even an 8-fold higher insulin concentration (mean of 85 ng/ml in females) in FVB-db mice is unable to prevent overt hyperglycemia.


Figure 1
View larger version (13K):
[in this window]
[in a new window]
  		FIG. 1. Characteristics of three congenic db strains. Fed glucose and insulin values (mean and SDs) are provided for obese 3- to 4-month-old mice (db/db) of three congenic strains: C57BL/6J (black symbols, B6), C57BLKS/J (gray, BLKS), and FVB (unfilled, FVB). Males are symbolized by squares, and females by circles.

 
Within the pancreas, there is a difference in the distribution of islet sizes between C57BL/6J db3J/db3J and FVB db/db males (Fig. 2Go). A significant fraction (~45%) of the C57BL/6J db3J/db3J islets contain fewer than 100 cells. This stands in contrast to the islets of FVB db/db males where nearly 80% contain more than 100 cells.


Figure 2
View larger version (43K):
[in this window]
[in a new window]
  		FIG. 2. Islets from obese males of the two congenic strains and obese males of the F1 generation at 5 months of age. A, Relative frequencies of islet sizes in obese male mice of the two congenic strains and the F1 cross. Islet cellularity was determined for individual mice (four to five male mice per strain). Islets were binned into three size classes (1–100 cells; 101–500 cells; and > 500 cells). Male FVB db/db mice have a significantly different size distribution of islets compared with C57BL/6J db3J/db3J males and BxF F1 db/db3J males; signified by # (P < 0.05, {chi}2 test). B, Representative islets from the three types of obese male mice. Islets from B6 db3J/db3J mice are ovoid and smooth in their outlines. Islets from FVB db/db mice are typically larger, irregular, and usually associated with ductal structures. The islets from BxF F1 db/db3J mice are similar in size to islets of B6-db3J/db3J mice but are irregular in their outline and are associated with ductal structures. Scale bar, 200 µm.

 
Genetic transmission of insulin resistance and ß-cell response to hyperglycemia
To define the genetics that determine the phenotypic differences between the C57BL/6J and FVB strains, we performed a set of genetic crosses between the FVB db and C57BL/6J db3J strains. We generated two sets of F1 outcross progeny (FVB-db x C57BL/6J-db and FVB-db x C57BL/6J-db3J). The obese F1 male progeny (db/db and compound heterozygotes db/db3J) were obese and hyperglycemic (Fig. 3Go) and there was no difference between db/db and db/db3J animals. The glucose concentrations were in the hyperglycemic range and identical with those observed in FVB db/db mice, whereas their insulin concentrations were similar to those observed for C57BL/6J db3J/db3J mice. The pancreatic islets were small, and their size distribution is similar to that observed in obese B6 males (Fig. 2Go). Thus, it is likely that the severe insulin resistance of the FVB db/db mice is a dominant trait. However, the low insulin concentrations would suggest that the FVB strain’s ability to expand ß-cell mass is a trait that is recessive to the alleles found in the C57BL/6J genome. These hypotheses were further tested by generating N2 populations from two backcrosses (F1-db3J/+ x FVB-db/+ and F1-db3J/+ x C57BL/6J-db/+).


Figure 3
View larger version (11K):
[in this window]
[in a new window]
  		FIG. 3. Characteristics of obese male progeny of the F1 generation (FxB F1) and the two reciprocal backcrosses—(FxB) F1 x F and (FxB) F1 x B. The data represent glucose and insulin values from individual obese mice from the three crosses (F1, filled diamonds; F1xF, open triangles; F1xB, x symbol) between 3 and 4 months of age.

 
The backcross to C57BL/6J (F1-db3J/+ x C57BL/6J-db/+, F1xB) enriched for C57BL/6J alleles such that any given locus in any given mouse is either homozygous for the B6 allele or heterozygous carrying one FVB and one B6 allele. The exception is in the Lepr locus because only db3J/db compound heterozygotes are obese. It is notable that the obese mice for this cross are obligate heterozygotes (FVB-C57BL/6J) for the region around Lepr on mouse chromosome 4. Most of these mice exhibit near euglycemia with a large variation in insulin concentrations (Fig. 3Go). This is consistent with our hypothesis that the C57BL/6J genome carries alleles recessive to FVB alleles and that these recessive alleles confer improved insulin sensitivity.

The backcross to FVB (F1-db3J/+ x FVB-db/+; F1xF) enriched for FVB alleles. Any given mouse at any given locus is either homozygous for FVB alleles or heterozygous carrying a C57BL/6J and a FVB allele. It is notable that these mice are also obligate heterozygotes (FVB-C57BL/6J) for the region around Lepr, as in the backcross to C57BL/6J, making it unlikely that any phenotypic differences between the reciprocal backcrosses could be due to alleleic differences in genes in the vicinity of Lepr. Most of these mice are hyperglycemic with a great variation in insulin concentrations (Fig. 3Go). This is consistent with our postulate that the FVB genome carries alleles dominant to C57BL/6J alleles that confer increased insulin resistance. Moreover, the hyperglycemic status of these mice allows us to test two postulates: 1) the FVB genome has allelic variants that control ß-cell responses to hyperglycemia; and 2) these variants are recessive to the C57BL/6J variants.

Markers on chromosome 5 are associated with insulin concentrations
We classified 22 male db/db3J mice of the N2 FVB backcross (F1-db3J/+ x FVB-db/+) according to their insulin concentrations in the ad libitum-fed state. According to our postulate, mice in the upper half of the insulin concentration range should have been enriched to be homozygous FVB at those loci that control ß-cell responses, whereas mice in the lower half would be enriched to be heterozygous C57BL/6J-FVB. We performed a genome scan with a set of 31 microsatellite markers that could distinguish between C57BL/6J and FVB alleles. The markers were picked to provide coverage of each chromosome with an average distance of 30 cM between markers. There were several promising markers on three chromosomes (3, 5 and 13) that appeared to show cosegregation with the insulin concentrations in the genome scan (Table 1Go). In the case of the chromosome 5 markers, we observed a high frequency of FVB-FVB genotypes in the high (insulin) group, whereas there was a high frequency of C57BL/6J-FVB genotypes in the low (insulin) group, consistent with the idea that the C57BL/6J allele was associated with low insulin concentrations.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Cosegregation of FVB and C57BL/6J alleles on mouse chromosome 5 with insulin concentrations in (BxF) x F N2 progeny

 
These loci were retested on an independent set of 24 male N2 progeny from the same backcross (Table 1Go), although their mean (insulin) were lower because the blood samples were collected at a younger age. None of the markers were significant when analyzed separately. However, when the data sets A and B were combined and analyzed together, three markers on chromosome 5 spanning 60 cM showed a consistent association of allelic enrichment according to insulin concentrations. This provided supportive evidence, although certainly not conclusive proof, that a locus or a set of loci on mouse chromosome 5 was responsible for the response of ß-cells to hyperglycemia. We have provisionally called this locus Modb1, modifier of diabetes 1.

Two types of incipient congenic strains for locus confirmation
At this juncture of the project, we assessed options for obtaining further evidence for the existence of diabetes modifier genes on mouse chromosome 5, taking into account that these modifier genes produced phenotypes only in obese rodents. In general, the generation of full congenic mouse strains is too lengthy a process (a minimum of 3 yr) to provide evidence for the existence of a modifier gene. Alternatively, the generation of speed congenics, while reducing the time required to produce the mice, may entail the commitment of resources that are not available. Therefore, we developed a strategy that enhances the power of conventional congenic strains while minimizing the time and effort for their production. We conceived of two types of congenic strains. The first type of congenic strain is produced by marker assisted selection to produce mice homozygous for the desired haplotype. These two founders and their progeny are backcrossed to the host strain for several backcrosses, up to N4 or N5. At the N4 level, the residual haploid donor genome still contains approximately 12.5% of the host strain. This is the major criticism of conventional congenic strains because it is difficult to exclude the potential of a modifier gene in the unmarked regions of the genome producing the observed phenotype. However, analysis of two independent congenic strains at N4 or N5 would considerably reduce this criticism. The donor genome that the two independently derived incipient congenic strains share within the unmarked regions is the product of the two fractions contained within each strain that is derived from the donor genome. For two N4 incipient congenic strains, it is 0.125 x 0.125, or.015625 (Fig. 4BGo). This is equivalent to approximately 25 cM in the mouse genome or a region that is considerably smaller than the region of the introgressed interval. Unfortunately, the total amount of the genome that is contained in the two incipient congenic strains is almost doubled. There is a formal possibility that two independent modifier genes may be present, one in each of the congenic strains. It is possible that multiple modifier genes are in play between any two inbred mouse strains.


Figure 4
View larger version (16K):
[in this window]
[in a new window]
  		FIG. 4. Breeding scheme for the generation of incipient congenic strains. A, The breeding is initiated with the outcross between two parental strains to generate the F1 progeny. The F1 progeny are backcrossed repeatedly to the same parental strain up to N4 or N5. At the N2 and subsequent generations, selection of mice with the appropriate genotypes/haplotypes is necessary. At N4 or subsequent generations, an intercross is made to generate mice that are homozygous at the desired loci for the two parental strains. B, The fraction of the residual haploid donor genome at various backcross generations is plotted for one congenic strain and for the common region between two independently derived congenic strains.

 
A second type of congenic strain should also be produced as a control strain, one that reconstitutes the haplotype of the host strain at the investigated interval. Ideally, this strain should be generated from the same parents that were used to produce one of the two incipient congenic strains. Either at the N4 or N5 generation, heterozygous breeders are paired to produce mice that are homozygous for the donor haplotype (Fig. 4AGo). Within the progeny, mice that are homozygous for the host haplotype will also be identified. These are the progeny that should be used to generate the control strain. The effects of potential modifier genes within the unmarked genomic regions will be accounted for because these modifier genes will be nearly identical between the control strain and one of the two incipient congenic strains. The second type of congenic strain, the control strain, can be produced at any backcross generation and does not incur any additional costs or effort until a cohort of animals is to be produced for analysis.

Reciprocal chromosome 5 congenic strains provide evidence for the existence of Modb1
We tested the existence of Modb1 by generating multiple reciprocal congenic lines carrying db3J and defined alleles of Modb1, Modb1-F (FVB allele), and Modb1-B (C57BL/6J allele) on the FVB background. By the nature of backcrossing to produce congenic strains, there is the potential of carrying donor alleles to the introgressed locus, in this case, the C57BLKS/J db mutation. Although the original db mutation arose on C57BLKS/J and causes hyperglycemia with associated ß-cell loss, it is a formal possibility that specific C57BLKS/J derived alleles in the vicinity of the Lepr locus could contribute to the phenotype of the FVB db/db mice. Therefore, we used our newly developed C57BL/6J db3J/db3J congenic strain as a donor of C57BL/6J-derived alleles. As mentioned in Materials and Methods, the original db3J mutation arose on the 129/J strain and was transferred to the C57BL/6J strain by 10 backcrosses. In this manner, the generation of a new incipient congenic strain would formally eliminate the possibility of a chromosome 4 C57BLKS/J allele contribution to the ß-cell mass expansion phenotype because the new incipient congenic strains would carry 129-derived alleles linked to the db3J mutation. This is one of the criticisms of strain susceptibility studies with mutations—the possibility of a linked locus contributing to the phenotype cannot be eliminated because only one allele is available. The availability of independently derived Lepr-db alleles on several inbred strains is an asset for the resolution of a difficult and thorny problem. These reciprocal congenic lines were generated in the following manner. Lean N2 progeny from the FVB backcross (F1-db3J/+ x FVB-db/+) were initially screened for heterozygosity for the db3J mutation. Heterozygotes (db3J/+) were then screened for three markers that spanned 60 cM of chromosome 5. Mice that were heterozygous FVB-C57BL/6J at all three markers were assumed to contain no meiotic recombinations over the 60-cM span (Fig. 4Go). Subsequent testing showed no evidence of FVB alleles within the introgressed B6 interval. This process was repeated twice (to N4 for one line) and three times (to N5 for a second line). The likelihood that an unmarked locus is responsible for the observed phenotype is the product of the fraction of the residual haploid C57BL/6J genome shared between the two lines. The residual haploid C57BL/6J genome at the N4 generation is 12.5% and is 6.25% at the N5 generation. The fraction of the unmarked haploid C57BL/6J genome that is shared between the two lines is approximately 0.0078, whereas the nominally 60-cM region on chromosome 5 that is tracked with the genetic markers represents approximately 0.02 of the haploid genome in each line. At the N4 or N5 generations, db3J/+ mice that were heterozygous at all three chromosome 5 markers were intercrossed. The N4F1 progeny were selected to be homozygous for the C57BL/6J alleles at all three chromosome 5 markers and db3J/+. The N5F1 progeny were similarly selected. In addition, we selected N5F1 progeny that were homozygous for FVB alleles at all three chromosome 5 markers (N5-ff). This line was generated as a control because it shares all of the unmarked C57BL/6J alleles as its sibling N5F-bb line but differs only in its allelic constitution at chromosome 5. Thus, any difference between these N5 lines can be directly attributed to the allelic differences carried on chromosome 5. The mice were then intercrossed within each line to generate db3J/db3J mice that differed in their Modb1 status, Modb1-ff (N5-ff) or Modb1-bb (N5-bb). These mice would provide a direct test of the biological relevance of Modb1 alleles.

The diabetes phenotype of the mice that are Modb1-bb db3J/db3J is quite remarkable and dramatically different from mice that are Modb1-ff db3J/db3J (Table 2Go). The obese mice (db3J/db3J) from both bb lines (N4-bb and N5-bb) are hyperglycemic and relatively hypoinsulinemic. Most of the bb db3J/db3J males show a dramatic weight loss and have a high mortality rate between 6 wk and 3 months of age (N4-bb: 8 males alive of 14; N5-bb: 2 males alive of 8). This is in comparison to the N5-ff db3J/db3J mice that show hyperglycemia, hyperinsulinemia, and very low mortality (12 of 12 N5-ff males alive at end of 3 months), a phenotype that is nearly identical with that of the obese FVB-db mice described previously. Of note, we observed effects of Modb1 in both males and females, an observation that is important because the original genome scan was performed only in obese males. The mean insulin concentrations of Modb1-ff db3J/db3J mice is at least 2- to 3-fold higher than the insulin concentrations of Modb1-bb db3J/db3J mice (of both N4-bb and N5-bb lines) at very similar blood glucose concentrations (Table 2Go).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Diabetes phenotype of reciprocal two locus congenic lines

 
We also analyzed the insulin data regarding an increase in circulating insulin concentrations between 2 and 3 months of age (Table 2Go). The N5-ff db3J/db3J mice show a statistically significant increase in insulin concentrations over the 1-month interval, suggesting that there is a dynamic response to the hyperglycemia. In contrast, the two bb lines showed no such increment in circulating insulin concentrations.

In the pancreas, we observe differences in ß-cell mass that parallel the differences in insulin concentrations. The mean islet areas of Modb1-ff db3J/db3J (N5-ff) mice are larger than for Modb1-bb db3J/db3J (N5-bb) mice (Fig. 5Go). Indeed, it is difficult to identify islets in male Modb1-bb db3J/db3J (N5-bb) mice without immunohistochemical staining due to their atypical appearance. In female N5-bb db3J/db3J mice, the islets are large but only a small fraction of islet cells stain positively for insulin, probably due to extensive degranulation. Quantification of the islet areas (Fig. 5Go) of the two N5 sibling congenic strains shows significant differences. For both male and female N5-ff db3J/db3J mice, the mean islet areas are significantly greater (nearly 2-fold) than for same sex N5-bb and N4-bb db3J/db3J mice.


Figure 5
View larger version (50K):
[in this window]
[in a new window]
  		FIG. 5. Islets of obese mice from the reciprocal incipient congenic strains differing at chromosome 5. A, Representative insulin stained sections from 3-month-old males and females from the bb and ff incipient congenic strains are shown. Bar, 240 µm; all images are at the same magnification. B, Quantification of islet areas for the bb and ff reciprocal incipient congenic strains. Mean islet areas are presented for obese (db3J/db3J) males and females of the three incipient congenic strains. There were six to 11 mice per group except for the N5 bb db3J/db3J males (two in the group due to high mortality; see text). *, Significant effect (P < 0.05) of haplotype on islet area.

 
There was no evidence of an effect of the chromosome 5 haplotype on pancreatic islet cellularity in lean mice. Male bb mice with db3J/+ or +/+ genotypes had 30.3 ± 11.3 cells per islet compared with male ff mice (of the same Lepr genotypes) with 36.5 ± 8.5 cells per islet (Table 3Go). We did not observe any differences in body weight, glucose concentrations and insulin concentrations between lean males of these two haplotypes (Table 3Go). Female bb db3J/+ or +/+ mice had 39.0 ± 9.8 cells/islet compared with female ff db3J/+ or +/+ mice had 38.6 ± 7.0 cells per islet. There was no significant difference in islet cellularity between lean females of the two reciprocal congenic lines.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Phenotypic data of lean males from reciprocal two locus congenic lines

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Based on our observations of the disparate phenotypes of the two reciprocal backcrosses, we propose a heuristic model, albeit imprecise, that should prove useful at the current stage of analysis. Firstly, we propose that leptin is a critical component for maintenance of normal insulin sensitivity and that loss of leptin signaling results in a certain degree of insulin resistance. Secondly, modifier genes interact with the insulin resistance caused by loss of leptin signaling to produce transient hyperglycemia, as in the obese B6 mouse, or chronic hyperglycemia, as in the obese FVB mouse. Thirdly, insulin resistance and hyperglycemia cause the endocrine pancreas to respond by increasing insulin synthesis and secretion, along with ß-cell mass expansion. The stimulus on pancreatic ß-cells is also constrained by leptin signaling, via central effects on autonomic nervous system activity. Our strategy has been to maintain as a constant, loss of leptin signaling via analysis of obese db/db mice, and chronic hyperglycemia via analysis of the backcross to the FVB strain. In this manner, we believe that genetic linkage to those genes modulating pancreatic ß-cell mass would be identified.

Our search has been successful in isolating a segment of the mouse genome that harbors a gene or several genes that permit ß-cell mass expansion during hyperglycemia. This is a unique aspect of our model. Most other models analyze differences between euglycemic, insulin-resistant mice with expanded ß-cell mass and hyperglycemic mice with greatly diminished ß-cell mass: insulin receptor knockout (21, 22), leptin deficiency (23), leptin receptor deficiency (19), the New Zealand Obese (NZO) mouse (24, 25), the KK mouse strain (26), and the Goto-Kakizaki (GK) rat (27, 28). For these models, it was surmised that ß-cells were irreparably damaged by hyperglycemia. However, our model shows that loss of ß-cell mass is not an inevitable result of chronic hyperglycemia. It is remarkable that one chromomosal region is able to replicate the phenotypic differences between the two strains, given the large number of genes that have been shown to have an impact on ß-cell mass. The identified chromosome 5 interval may actually contain more than one locus and further subdivision of the region with subcongenic strains will be necessary to demonstrate the existence of multiple loci. Notwithstanding these potential complications, the current data favor the existence of naturally occurring genetic variants that control ß-cell responses to hyperglycemia. In support of this possibility, partial pancreatectomy in rats and mice results in mild hyperglycemia with growth of new islets via the recruitment of stem cells from intrapancreatic ducts (3, 4). Thus, a search for factors that permit ß-cell mass expansion, such as Cdk4 activation (5), may be a fruitful addition to approaches for the treatment of type 2 diabetes mellitus and diseases of the pancreatic ß-cell.

The lack of phenotypic differences in lean (leptin receptor intact) animals might be construed as a lack of relevance of Modb-1 to human pathophysiology except in the rare cases of deficiencies of leptin signaling. We would argue that Modb-1 may be relevant in cases of leptin resistance associated with obesity or lipodystrophy. Indeed, ß-cell mass and function are usually of interest under conditions of insulin deficiency or insulin resistance when insulin secretory capacity may not be able to meet physiological demands. The vast majority of inbred rodent strains do not show overt diabetes or gross abnormalities in glucose metabolism. However, studies of variations of ß-cell mass and function in standard rodent strains is justified by the implication that the genetic variation controlling phenotypic variation under standard rearing conditions can be extrapolated to the condition of type 2 diabetes mellitus. One major caveat to any of these studies, including the present one, is that genetic responses to different environments are difficult to predict. Norms of reaction actually need to be experimentally determined due to the complexity of the regulation of individual genes and physiological responses (29).

Regarding our genome scan, we limited the number of progeny that were analyzed. Our purpose was to identify a chromosomal interval of major effect that would be amenable to positional cloning rather than to identify most or all of the modifier genes segregating within the cross. Given the long-term prospects of identifying a modifier gene with an effect that was not generally believed to exist, we deemed it prudent to obtain firm evidence of such a modifier as well as to establish a tractable genetic system. Our data, comprised of the genome scan of the N2 progeny and the confirmatory phenotypes of multiple incipient congenic lines, affirms the value of our heuristic model that proposes modifier genes that independently affect insulin sensitivity and ß-cell mass. Eventual identification of the responsible genetic variants will be dependent upon the use of subcongenic strains that will refine the critical interval and reduce the quantitative trait (ß-cell mass) to a Mendelian trait (30).

Congenic lines have been used previously to successfully map modifier genes for diabetes in NOD mice and GK rats (31, 32). However, the development of speed congenics was a major emphasis in these projects. We propose that independent congenic lines for a given chromosomal region would be a useful adjunct to the tools of the geneticist. Two or three independent congenic lines can be readily accommodated in most breeding operations, unlike the cage requirements for speed congenic strain production. Multiple independent congenic lines for an individual locus also obviate the onerous chore of extensive genotyping at each backcross generation. Furthermore, depending on the size of the region being selected, female progeny can be used to propagate the backcross. However, this may only be possible in strains with highly fecund females, such as the FVB strain, because recombinant inbred lines are notorious for poor reproductive performance.

It is likely that the successful replication of the phenotypes in our incipient congenics was due to the extremely large segment of chromosome 5 that was introgressed (from 20–78 cM). The N4 and N5 BB lines were essentially consomic lines or chromosome substitution lines (33), with their attendant advantages. Our incipient congenic lines would most likely capture any and all diabetes modifier loci on chromosome 5 and thus, we cannot make any conclusions regarding the number of diabetes modifier loci located on mouse chromosome 5. Nevertheless, we will use the provisional symbol of Modb1 until there is strong evidence of multiple loci. This is in contrast to congenic lines that have been constructed to contain small (~10–20 cM) introgressed intervals that, being highly dependent on accurate mapping of the critical interval, have shown variability in their ability to replicate phenotypes observed in F2 progeny (34). Finally, the analysis of sibling congenic strains of differing haplotypes that are derived from the same progenitors (and removed by no more than two or three generations) provides powerful experimental evidence in favor of the existence of modifier genes within the chromosomal interval. The importance of the appropriate control strain for the congenic strain bearing the introgressed interval has been previously emphasized in a study using congenic mouse strains to analyze the genetic control of bone structure (35).


   Acknowledgments
 
We acknowledge the assistance of Drs. Margarita Leiser, Dennis Shields, and Norman Fleischer for the insulin/proinsulin analysis.


   Footnotes
 
This work was supported by grants from the National Institutes of Health: DK057621, DK063306, DK063608, DK26687, and 5P60DK020541.

Current address for J.L.: Singapore Eye Research Institute, 11 Third Hospital Avenue, Singapore 168751.

The authors claim that they have not had, in the last 12 months, a relevant duality of interest with a company whose products or services are directly related to the subject matter of our manuscript.

First Published Online February 16, 2006

Abbreviations: db/db, Diabetes; MICs, multiple incipient congenic strains.

Received July 9, 2005.

Accepted for publication February 8, 2006.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Lantz KA, Kaestner KH 2005 Winged-helix transcription factors and pancreatic development. Clin Sci (Lond) 108:195–204[Medline]
  2. Habener JF, Kemp DM, Thomas MK 2005 Minireview: transcriptional regulation in pancreatic development. Endocrinology 146:1025–1034[Abstract/Free Full Text]
  3. Bonner-Weir S, Toschi E, Inada A, Reitz P, Fonseca SY, Aye T, Sharma A 2004 The pancreatic ductal epithelium serves as a potential pool of progenitor cells. Pediatr Diabetes 5(Suppl 2):16–22
  4. Bonner-Weir S, Baxter LA, Schuppin GT, Smith FE 1993 A second pathway for regeneration of adult exocrine and endocrine pancreas. A possible recapitulation of embryonic development. Diabetes 42:1715–1720[Abstract]
  5. Rane SG, Dubus P, Mettus RV, Galbreath EJ, Boden G, Reddy EP, Barbacid M 1999 Loss of Cdk4 expression causes insulin-deficient diabetes and Cdk4 activation results in ß-islet cell hyperplasia. Nat Genet 22:44–52[CrossRef][Medline]
  6. Inada A, Hamamoto Y, Tsuura Y, Miyazaki J, Toyokuni S, Ihara Y, Nagai K, Yamada Y, Bonner-Weir S, Seino Y 2004 Overexpression of inducible cyclic AMP early repressor inhibits transactivation of genes and cell proliferation in pancreatic ß cells. Mol Cell Biol 24:2831–2841[Abstract/Free Full Text]
  7. Cozar-Castellano I, Weinstock M, Haught M, Velazquez-Garcia S, Sipula D, Stewart AF 2006 Evaluation of ß-cell replication in mice transgenic for hepatocyte growth factor and placental lactogen: comprehensive characterization of the G1/S regulatory proteins reveals unique involvement of p21cip. Diabetes 55:70–77[Abstract/Free Full Text]
  8. Garcia-Ocana A, Vasavada RC, Cebrian A, Reddy V, Takane KK, Lopez-Talavera JC, Stewart AF 2001 Transgenic overexpression of hepatocyte growth factor in the ß-cell markedly improves islet function and islet transplant outcomes in mice. Diabetes 50:2752–2762[Abstract/Free Full Text]
  9. Cebrian A, Garcia-Ocana A, Takane KK, Sipula D, Stewart AF, Vasavada RC 2002 Overexpression of parathyroid hormone-related protein inhibits pancreatic ß-cell death in vivo and in vitro. Diabetes 51:3003–3013[Abstract/Free Full Text]
  10. Zhang X, Gaspard JP, Mizukami Y, Li J, Graeme-Cook F, Chung DC 2005 Overexpression of cyclin D1 in pancreatic ß-cells in vivo results in islet hyperplasia without hypoglycemia. Diabetes 54:712–719[Abstract/Free Full Text]
  11. Bell GI and Polonsky KS 2001 Diabetes mellitus and genetically programmed defects in ß-cell function. Nature 414:788–791[CrossRef][Medline]
  12. Horikawa Y, Oda N, Cox NJ, Li X, Orho-Melander M, Hara M, Hinokio Y, Lindner TH, Mashima H, Schwarz PE, del Bosque-Plata L, Oda Y, Yoshiuchi I, Colilla S, Polonsky KS, Wei S, Concannon P, Iwasaki N, Schulze J, Baier LJ, Bogardus C, Groop L, Boerwinkle E, Hanis CL, Bell GI 2000 Genetic variation in the gene encoding calpain-10 is associated with type 2 diabetes mellitus. Nat Genet 26:163–175[CrossRef][Medline]
  13. Leibel RL, Chung WK, Chua Jr SC 1997 The molecular genetics of rodent single gene obesities. J Biol Chem 272:31937–31940[Free Full Text]
  14. Coleman DL, Leiter EH, Applezweig N 1984 Therapeutic effects of dehydroepiandrosterone metabolites in diabetes mutant mice (C57BL/KsJ-db/db). Endocrinology 115:239–243[Abstract]
  15. Hummel KP, Dickie MM, Coleman DL 1966 Diabetes, a new mutation in the mouse. Science 153:1127–1128[Abstract/Free Full Text]
  16. Haluzik M, Colombo C, Gavrilova O, Chua S, Wolf N, Chen M, Stannard B, Dietz KR, Le Roith D, Reitman ML 2004 Genetic background (C57BL/6J versus FVB/N) strongly influences the severity of diabetes and insulin resistance in ob/ob mice. Endocrinology 145:3258–3264[Abstract/Free Full Text]
  17. Chua Jr S, Liu SM, Li Q, Yang L, Thassanapaff VT, Fisher P 2002 Differential ß cell responses to hyperglycaemia and insulin resistance in two novel congenic strains of diabetes (FVB-Lepr (db)) and obese (DBA-Lep (ob)) mice. Diabetologia 45:976–990[CrossRef][Medline]
  18. Leiter EH, Coleman DL, Eisenstein AB, Strack I 1980 A new mutation (db3J) at the diabetes locus in strain 129/J mice. I. Physiological and histological characterization. Diabetologia 19:58–65[CrossRef][Medline]
  19. Chung WK, Zheng M, Chua M, Kershaw E, Power-Kehoe L, Tsuji M, Wu-Peng XS, Williams J, Chua Jr SC, Leibel RL 1997 Genetic modifiers of Leprfa associated with variability in insulin production and susceptibility to NIDDM. Genomics 41:332–344[CrossRef][Medline]
  20. Xu H, Shields D 1993 Prohormone processing in the trans-Golgi network: endoproteolytic cleavage of prosomatostatin and formation of nascent secretory vesicles in permeabilized cells. J Cell Biol 122:1169–1184[Abstract/Free Full Text]
  21. Kulkarni RN, Almind K, Goren HJ, Winnay JN, Ueki K, Okada T, Kahn CR 2003 Impact of genetic background on development of hyperinsulinemia and diabetes in insulin receptor/insulin receptor substrate-1 double heterozygous mice. Diabetes 52:1528–1534[Abstract/Free Full Text]
  22. Kido Y, Philippe N, Schaffer AA, Accili D 2000 Genetic modifiers of the insulin resistance phenotype in mice. Diabetes 49:589–596[Abstract]
  23. Stoehr JP, Nadler ST, Schueler KL, Rabaglia ME, Yandell BS, Metz SA, Attie AD 2000 Genetic obesity unmasks nonlinear interactions between murine type 2 diabetes susceptibility loci. Diabetes 49:1946–1954[Abstract]
  24. Plum L, Giesen K, Kluge R, Junger E, Linnartz K, Schurmann A, Becker W and Joost HG 2002 Characterisation of the mouse diabetes susceptibilty locus Nidd/SJL: islet cell destruction, interaction with the obesity QTL Nob1, and effect of dietary fat. Diabetologia 45:823–830[CrossRef][Medline]
  25. Reifsnyder PC and Leiter EH 2002 Deconstructing and reconstructing obesity-induced diabetes (diabesity) in mice. Diabetes 51:825–832[Abstract/Free Full Text]
  26. Suto J and Sekikawa K 2002 A quantitative trait locus that accounts for glucose intolerance maps to chromosome 8 in hereditary obese KK-A(y) mice. Int J Obes Relat Metab Disord 26:1517–1519[CrossRef][Medline]
  27. Wallace KJ, Wallis RH, Collins SC, Argoud K, Kaisaki PJ, Ktorza A, Woon PY, Bihoreau MT, Gauguier D 2004 Quantitative trait locus dissection in congenic strains of the Goto-Kakizaki rat identifies a region conserved with diabetes loci in human chromosome 1q. Physiol Genom 19:1–10[Abstract/Free Full Text]
  28. Wallis RH, Wallace KJ, Collins SC, McAteer M, Argoud K, Bihoreau MT, Kaisaki PJ, Gauguier D 2004 Enhanced insulin secretion and cholesterol metabolism in congenic strains of the spontaneously diabetic (type 2) Goto Kakizaki rat are controlled by independent genetic loci in rat chromosome 8. Diabetologia 47:1096–1106[Medline]
  29. Griffiths AJF, Gelbart WM, Miller JH, Lewontin RC 1999 Modern genetic analysis. New York: W. H. Freeman and Co.
  30. Lander ES, Botstein D 1989 Mapping mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121:185–199[Abstract/Free Full Text]
  31. Aitman TJ, Todd JA 1995 Molecular genetics of diabetes mellitus. Baillieres Clin Endocrinol Metab 9:631–656[CrossRef][Medline]
  32. Ktorza A, Bernard C, Parent V, Penicaud L, Froguel P, Lathrop M and Gauguier D 1997 Are animal models of diabetes relevant to the study of the genetics of non-insulin-dependent diabetes in humans? Diabetes Metab 23(Suppl 2):38–46
  33. Nadeau JH, Singer JB, Matin A, Lander ES 2000 Analysing complex genetic traits with chromosome substitution strains. Nat Genet 24:221–225[CrossRef][Medline]
  34. Grondin M, Eliopoulos V, Lambert R, Deng Y, Ariyarajah A, Moujahidine M, Dutil J, Charron S, Deng AY 2005 Complete and overlapping congenics proving the existence of a quantitative trait locus for blood pressure on Dahl rat chromosome 17. Physiol Genom 21:112–116[Abstract/Free Full Text]
  35. Shultz KL, Donahue LR, Bouxsein ML, Baylink DJ, Rosen CJ and Beamer WG 2003 Congenic strains of mice for verification and genetic decomposition of quantitative trait loci for femoral bone mineral density. J Bone Miner Res 18:175–185[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Endocr. Rev.Home page
S. M. Clee and A. D. Attie
The Genetic Landscape of Type 2 Diabetes in Mice
Endocr. Rev., February 1, 2007; 28(1): 48 - 83.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
147/5/2287    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Luo, N.
Right arrow Articles by Chua, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Luo, N.
Right arrow Articles by Chua, S., Jr.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals