Deletion of Nicotinamide Nucleotide Transhydrogenase : A New Quantitive Trait Locus Accounting for Glucose Intolerance in C57BL/6J Mice

Mice were kept in accordance with U.K. Home Office welfare guidelines and project license restrictions under controlled light (12-h light and 12-h dark cycle), temperature (21 ± 2°C), and humidity (55 ± 10%) conditions. They had free access to water (25 ppm chlorine) and were fed ad libitum on a commercial diet (SDS maintenance diet) containing 2.6% saturated fat.

Transgenic mice were generated by pronuclear injection of closed circular BAC DNA (available at http://bacpac.chori.org/home.htm) into mouse C57BL/6J one-cell embryos. Briefly, 3-week-old C57BL/6J mice were superovulated with 5 units pregnant mare’s serum gonadotropin, followed 46 h later by 5 units human chorionic gonadotropin and mated with C57BL/6J males overnight. The next day, females with copulation plugs were culled and embryos harvested and treated with hyaluronidase to remove cumulus cells (8). DNA prepared using a Qiagen Large Construct kit was resuspended in polyamine buffer (10 mmol/l Tris-HCl pH 7.5, 0.1 mmol/l EDTA, pH 8.0, 100 mmol/l NaCl, 30 μmol/l spermine, and 70 μmol/l spermidine) immediately before microinjection. Between 100 and 200 embryos were injected with BAC DNA at a concentration of 1.0 ng/μl. Injected embryos were transferred on the same day to the oviducts of pseudopregnant CD1-recipient females. Litters arising from these transfers were genotyped at 3 weeks of age using ear biopsies to identify transgenic animals.

Genomic DNA was extracted from mouse tail tissue using a Qiagen DNeasy tissue kit. Mice were genotyped by PCR of missing exons 7 and 11 of the Nnt gene. Primers listed in 5′ to 3′ orientation are for Nnt exon 7: NntXn7 forward gtgcattgaaccctcaaagg and NntXn7 reverse caggtaagaaagctcctgtttt; for Nnt exon 11: NntXn11 forward tcctgctattcctcctcctg and NntXn11 reverse gctgccttgactttggatatt. PCR products were visualized on a 2% agarose gel run at 200V for 25 min.

Total RNA from mouse liver was extracted using an RNeasy mini-kit (Qiagen). cDNA generated by Superscript II enzyme (Invitrogen) underwent PCR amplification across the deleted fragment of Nnt and several introns. Primers listed in 5′ to 3′ orientation are for Nnt fragment 1 forward cagcacagctctgattccag and Nnt fragment 1 reverse gtcaaaccgaagaccgtggctgag; for Nnt fragment 2 forward gtgatgaaggatggcaaagtg and Nnt fragment 2 reverse gcaggtggttttctggtgactc.

Mice were fasted overnight, weighed, and a blood sample collected by tail venipuncture under local anesthetic (Lignocaine) using lithium-heparin microvette tubes (Sarstedt) to establish a baseline glucose level “T0.” They were then injected intraperitoneally with 2 g glucose (20% glucose in 0.9% NaCl) per kilogram body weight and blood samples taken at the designated times. For a fuller protocol, see EMPReSS (the European Mouse Phenotyping Resource for Standardized Screens designed by Eumorphia)-simplified IPGTT (available at http://empress.har.mrc.ac.uk). Plasma glucose was measured using an Analox Glucose analyzer GM9 (Analox, London, U.K.). Plasma insulin was measured using a Mercodia Ultra-Sensitive Mouse ELISA kit (Mercodia, Uppsala, Sweden) according to the manufacturer’s instructions.

Mice were killed by cervical dislocation, the pancreas removed, and islets isolated by liberase digestion and handpicking (5). Isolated islets were dispersed into single cells by incubation in calcium-free Hank’s solution (1 mmol/l EGTA) and trituration in RPMI tissue culture medium containing 5.5 mmol/l glucose (Life Technologies, Paisley, U.K.) supplemented with 10% FCS, 100 units/ml penicillin, and 100 μg/ml streptomycin. Cells were maintained in this medium at 37°C in a humidified atmosphere at 5% CO₂ in air and used 1–2 days after the isolation.

Nnt expression was verified using SDS-PAGE 12% gel and Western blot. Protein samples were heated for 3 min at 100°C in sample buffer (final concentration of 30 mmol/l Tris-HCl, pH6.8, 5% glycerol, 0.005% bromophenol blue, and 2.5% β-mercaptoethanol). Proteins were size separated by electrophoresis alongside a “SeeBlue” (Invitrogen, Paisley, U.K.) protein ladder to allow molecular weights to be estimated.

Proteins were transferred to hybond-P membrane (Amersham, Chalfont St. Giles, U.K.) using a Transblot SD system (Biorad) by passing 100 V across membrane/gel for 1 h. Membranes were probed with a custom-made polyclonal Nnt antibody raised against amino acids 10731-086 at the C-terminus of the peptide (Eurogentec, Seraing, Belgium) and actin antibody (control) from Santa Cruz (Santa Cruz, CA). Enhanced Chemiluminescence Plus (Amersham) was then used according to manufacturers’ instructions to allow visualization of protein on enhanced chemiluminescence film, developed using the Compact X4 (Xograph).

We obtained two independent transgenic lines, B6-TG1 and B6-TG2. BAC complementation was verified first by PCR of the BAC vector itself (Fig. 1A) and second at the transcriptional level by detection of cDNA missing exons 7–11 in the Nnt gene (Fig. 1B). This was also observed at the protein level by Western blotting. The amount of Nnt protein produced by both transgenic lines was similar to or higher than that produced by a C3H/HeH control strain that harbors wild-type Nnt (Fig. 1C). Nnt protein was completely absent in B6-TG–negative controls and stock C57BL/6J mice, as previously reported (6). Both the transgenes were transmitted through the germline, and lines were maintained by crossing to wild-type C57BL/6J mice.

In both lines, expression of wild-type Nnt on the C57BL/6J background in male mice induced improvements in glucose tolerance following a glucose challenge at 12 and 16 weeks of age compared with nontransgenic littermates (Fig. 2A and B). We therefore looked for differences in insulin secretion by carrying out an IPGTT taking samples at 0, 10, 20, and 30 min after glucose injection (Fig. 2C). Insulin secretion by B6-TG1 and B6-TG2 mice was significantly higher (for the same glucose challenge per gram bodyweight) than for B6-TG–negative mice. No difference in body weight was observed between the transgenic lines compared with the C57BL/6J littermates without the transgene. Similar results were seen in female C57BL/6J mice, although glucose intolerance in females is less pronounced (5) (online appendix d [available at http://diabetes.diabetesjournals.org]).

To confirm that a defect in β-cell function, due to a deletion in Nnt, underlies the glucose intolerance of C57BL/6J mice and that this is ameliorated in our transgenic rescue mice, we compared the ability of glucose to stimulate insulin secretion from islets isolated from B6-TG and B6-TG–negative mice. Figure 2D shows that there was no difference in basal insulin secretion (at 0 and 2 mmol/l glucose) between the two mouse strains but that insulin secretion in response to 20 mmol/l glucose was increased in B6-TG islets in comparison to B6-TG–negative islets (P < 0.05). The ATP-sensitive K⁺ channel blocker tolbutamide stimulated secretion comparably in both strains as expected because it bypasses the steps in metabolism secretion coupling that involve Nnt (7,9,10). Figure 2E clearly shows that B6-TG mice have normal glucose tolerance resembling that of C3H/HeH control mice, whereas both B6-TG–negative and C57BL/6J mice display significantly higher plasma glucose levels after 60 and 120 min following IPGTT. Of 23 mouse strains previously screened (5), including six other C57BL mouse strains, only C57BL/6J had the naturally occurring deletion in Nnt (online appendix a).

Our findings show that the deletion of Nnt observed in C57BL/6J mice accounts for their glucose insensitivity and reduced insulin secretion. This supports the concept that primary genetic defects in β-cell metabolism contribute to the pathogenesis of impaired glucose homeostasis (11,12). Our QTL mapping and that of Kayo et al. (13) suggested that this locus would account for ∼10% of the variance of the phenotype (5,13). However, this is in the F₂ genetic environment rather than a pure C57BL/6J genetic environment, where Nnt alone appears to be sufficient to account for the phenotype. Thus, the genetic compositions of the mice are different, and this illustrates the importance of genetic interactions in determining quantitative traits.

We thank the Medical Research Council (studentship to H.C.F.) and EU FP6 funding (LHSM-CT-2003-503041) for support for this work.

We thank the staff of the Mary Lyon Centre for their excellent technical support.