Cell-specific Expression and Regulation of a Glucokinase Gene Locus Transgene*

Transgenic mice containing one or more extra copies of the entire glucokinase (GK) gene locus were generated and characterized. The GK transgene, an 83-kilobase pair mouse genomic DNA fragment containing both promoter regions, was expressed and regulated in a cell-specific manner, and rescued GK null lethality when crossed into mice bearing a targeted mutation of the endogenous GK gene. Livers from the transgenic mice had elevated GK mRNA, protein, and activity levels, compared with controls, and the transgene was regulated in liver by dietary manipulations. The amount of GK immunoreactivity in hepatocyte nuclei, where GK binds to the GK regulatory protein, was also increased. Pancreatic islets displayed increased GK immunoreactivity and NAD(P)H responses to glucose, but only when isolated and cultured in 20 mm glucose, as a result of the hypoglycemic phenotype of these mice (Niswender, K. D., Shiota, M., Postic, C., Cherrington, A. D., and Magnuson, M. A. (1997) J. Biol. Chem. 272, 22604–22609). Together, these results indicate that the region of the gene from −55 to +28 kilobase pairs (relative to the liver GK transcription start site) contains all the regulatory sequences necessary for expression of both GK isoforms, thereby placing an upper limit on the size of the GK gene locus.

Glucokinase (GK) 1 plays an essential role in maintaining euglycemia, both in humans and rodents (2)(3)(4)(5)(6)(7). GK gene expression is cell type-specific and involves two alternate promoters that are differentially regulated (8 -10). Expression of the islet GK isoform, which is determined by the upstream promoter, is a key determinant of the insulin secretory response to glucose (11,12). The upstream promoter is also expressed in several other rare neural/neuroendocrine cell types in the brain and gut, including pituitary corticotropes, although the functional significance of GK in these locations remains to be determined (13). The downstream GK promoter determines expres-sion of the hepatic GK isoform, which is thought to be a ratedetermining step for hepatic glucose utilization (11,14,15).
Studies of the cis-regulatory elements involved in cell-specific expression of islet GK have shown that a 294-bp DNA fragment containing the upstream promoter is sufficient for reporter gene expression in a variety of neural/neuroendocrine cell types, including pancreatic ␤ cells (13). Elements in the proximal promoter region have been identified that contribute to transcription in both insulinoma cells (16,17) and AtT-20 cells, a pituitary corticotrope cell line (18). Distal elements may also be involved, although none have been reported (17). GK gene transcription in the islet appears to be largely constitutive, although glucose acts at a post-transcriptional level to regulate islet GK activity. Increased glucose leads to a 3-4-fold elevation in GK activity, islet glucose usage, and insulin secretion, without a parallel increase in islet GK mRNA (19 -21).
The regulation of the downstream GK promoter, which is expressed only in hepatocytes, is both more complex and less well understood than that of the upstream promoter. Insulin increases hepatic GK gene transcription, whereas glucagon, acting via cAMP, decreases transcription (8,(22)(23)(24). Thyroid hormone and biotin also increase GK gene transcription in liver (25,26). The cis-regulatory elements that determine hepatocyte-specific expression and regulation of the hepatic GK isoform are largely undefined as a DNA fragment able to confer both hepatocyte-specific and hormone-regulated expression to a reporter gene has not yet been identified. While Iynedjian et al. (27) have recently reported that a 1-kb fragment of the rat downstream promoter is transciptionally active in primary hepatocytes, it is not regulated by insulin. Furthermore, in transgenic mice a 7.5-kb fragment of the rat downstream promoter DNA does not confer position-independent expression on a reporter gene (28).
We have previously reported the cloning and partial sequencing of an 83-kb DNA fragment of the mouse GK gene locus (28). These studies revealed that the two promoters in the mouse GK gene are separated by 35 kb, and that the gene contains at least eight liver-specific DNase I-hypersensitive sites that span over 20 kb of DNA (28). Given the complexity of the GK gene locus, both in terms of the widely separated promoter regions, and the multiple, widely dispersed hypersensitive sites, we sought to place an upper limit on its size by inserting the 83-kb fragment of cloned genomic DNA into the genome of mice. The studies presented here indicate that this DNA fragment, which spans from Ϫ55 to ϩ28 kilobase pairs relative to the liver transcription start site, is expressed and regulated both in the liver and in the islet, strongly suggesting that it contains the entire mouse GK gene locus. In addition to placing an upper limit on the size of the GK gene locus, these studies have generated a novel animal model for determining the metabolic ramifications of increased GK gene copy number, as are described separately (1).

EXPERIMENTAL PROCEDURES
Transgene Preparation-The insert from P1-305, which contains all GK coding sequences and both promoter regions (28), was isolated using a Qiagen Maxi plasmid purification kit (Qiagen, Chatsworth, CA). Restriction of P1-305 DNA with NruI left 392 bp of vector sequence on the 5Ј end and 202 bp of vector sequence on the 3Ј end of the insert, thereby enabling the transgene to be distinguished from the endogenous mouse GK gene locus. The 83-kb genomic DNA fragment was isolated as a single fragment by pulsed-field gel electrophoresis and agarase digestion (Gelase, Epicentre, Madison, WI) of a slice of gel. The DNA solution was dialyzed against TE, concentrated with a microconcentrator tube (Amicon, Beverly, MA), and quantitated by Hoechst dye fluorescence.
DNA Pronuclear Injection, Identification of Transgenic Animals, Transgene Integrity, and Copy Number-Mouse zygotes produced by mating two (C57BL/6J ϫ DBA/2J) F 1 hybrid mice were microinjected with the P1-305 DNA fragment (ϳ3 ng/l) then transferred into pseudopregnant ICR females (29). Transgenic mice were identified by PCR and Southern blot analysis of genomic DNA prepared from tail biopsies (29). PCR analysis was performed using primers that detected either a 355-bp fragment from the 5Ј end vector sequence (5Ј-AGACGTAGC-CCAGCGCGTCGGC and 5Ј-GCGGCCGCAAATTTATTAGAGC) or a 170-bp fragment from the 3Ј end vector sequence (5Ј-CGACGGCCAAT-TAGGCCTAC and 5Ј-GGCTTGAGTCGCTCCTCCTG). Southern blot analysis was performed using standard procedures (30) with the 5Ј end-specific PCR fragment used to routinely identify transgenic animals. For copy number analysis, probe 1, which lies near the 5Ј end of the transgene, was a 2.3-kb SacI fragment from plasmid mGK28.H2.S1. Probe 2, which is located near the middle of the transgene, was a 1.3-kb SacI-BamHI fragment from plasmid mGK40.SB. Probe 3 was a 500-bp BamHI-EcoRI fragment from pmGK18.BS, which contains sequences near the 3Ј end of the transgene. Mice bearing the targeted mutation of the endogenous GK gene (7) were identified using a 268-bp PCR fragment amplified from pmGK28.H2 using the primer pair 5Ј-AGCTTTG-GCAAATAGACAT and 5Ј-TGATGTAACTCATAAGGT. The DNA was cut with BglII and analyzed by Southern blot analysis. A 2.5-kb band was specific for the inactivated GK allele, and a 9-kb band indicated a wild type GK allele, as described previously by Bali et al. (7). An ϳ1-kb DNA fragment (generated by PCR using the primers 5Ј-CGCTCTA-GAACTAGTGGATCC and 5Ј-GAACAAGCTAGACACAGGTAG) containing sequences from a neomycin resistance gene was used as an internal probe for the targeted GK locus. Quantitation of band intensity was performed by phosphorimager analysis (GS-250 Molecular Imager, o-Rad).
All mice were housed in specific pathogen-free barrier facilities, maintained on a 12-h light/dark cycle, and fed a standard rodent chow (Purina Mills, Inc., St. Louis, MO). Females from both transgenic lines were used for most analyses. The gene rescue experiment was conducted using animals from line 37 only. Non-transgenic offspring of similar genetic backgrounds from within the same mouse colony served as controls. Animals were killed at 0600, which corresponds to a fed state (31), except for those in Fig. 5, which were killed in the afternoon.
Northern Blot Analysis-Total RNA was prepared by acid guanidinium thiocyanate extraction (32). RNA gels, blots, and hybridizations were performed as described previously (28). A 789-bp rat GK cDNA fragment from pGK.Z1 (24) was used as the probe. The loading control probe was an ϳ700-bp HindIII-EcoRI cyclophilin DNA fragment (33). Densitometric analysis of autoradiograms was performed using NIH Image software (National Institutes of Health, Bethesda, MD) after digital scanning. GK values were corrected for loading differences with the cyclophilin value.
Western Blot Analysis-Western blot analyses were performed as described (34). To visualize GK antibody binding, the washed membranes were incubated in a 1:10,000 dilution of donkey anti-sheep horseradish peroxidase-conjugated antibody (Jackson Immunoresearch) for 1 h at room temperature. A solution consisting of 1.25 mM Luminol, 0.2 mM para-coumaric acid, and 0.009% H 2 O 2 in 100 mM Tris (pH 8.5) was used to initiate the chemiluminescence reaction. The membranes were then drip-dried, wrapped in polyvinyl chloride film, and exposed to autoradiography film.
GK Activity Measurements-GK activities in crude liver extracts (as prepared for Western blotting) were estimated by determining the difference in glucose phosphorylation in the presence of 100 mM and 0.5 mM glucose. The final assay mixture contained 50 mM triethanolamine hydrochloride, 20 mM MgCl 2 , 100 mM KCl, 1 mM dithiothreitol, 0.1% bovine serum albumin, 10 mM ATP, 1 mM NADP, either 0.5 or 100 mM glucose, 4 g/ml glucose-6-phosphate dehydrogenase (Boehringer Mannheim), and 100 l of liver extract. Glucose phosphorylating activity was taken as the increase in NADPH fluorescence measured at 340 nM after 4 min at 32°C.
Immunocytochemical Analysis-Tissues were processed for immunocytochemical analysis as described previously (34). Briefly, GK immunoreactivity was detected using an anti-GST-GK fusion protein antibody (34) with a CY3-conjugated donkey anti-sheep IgG secondary antibody (Jackson Immunoresearch, West Grove, PA). Samples were imaged using a Zeiss LSM410 confocal scanning laser microscope with the 543 nm line of a helium/neon laser, and digital images were quan-

FIG. 1. Detection and characterization of the GK gene locus transgene.
A, diagram of the GK gene fragment used. The insert was excised from the P1 vector using NruI. This left short vector DNA sequences on the 5Ј and 3Ј ends that enabled detection of the transgene. B, PCR detection of the 5Ј and 3Ј ends of the transgene in two of three founders. C, Southern blot analysis using probes isolated from three different parts of the gene (locations are shown in A). The same three probes were used to determine the transgene integrity and calculate copy number. This analysis was performed using mice that were both homozygous (hom.) and heterozygous (het.) for the transgene, as well as non-transgenic controls (cont.). titated using NIH Image software. When expression levels were to be compared, tissues were processed in parallel, sections cut to the same thickness, and immunostaining reactions were conducted in parallel with the same reagents. The sensitivity and base line of the photomultiplier tube in the confocal microscope was adjusted such that the full range of pixel intensities were collected without saturation. Hepatocyte nuclear GK quantification was performed by counting the number of immunofluorescent positive (nuclear signal above cytoplasmic signal) nuclei per high powered field (mm 2 ).
Cultured Islet Studies-Islets were isolated from heterozygous transgenic animals from line 37 and control mice by pancreatic distension and collagenase digestion of the splenic portion of the pancreas (35,36). Islets were cultured in RPMI 1640 medium (Life Technologies, Inc.) with 10% fetal bovine serum, 100 IU/ml penicillin, 100 g/ml streptomycin, and 20 mM glucose for 40 h. Whole islet immunostaining was performed using affinity-purified anti-GK IgG (34) at a 1:10 dilution. Two-photon excitation microscopy analysis of NAD(P)H autofluorescence in intact islets was performed as described (37).
Statistical Analysis-Results are presented as the mean Ϯ standard error of the mean. Statistical significance was determined by Student's t test. Differences are reported as statistically significant at an ␣ value of 0.05.

Generation of GK Gene Locus Transgenic Mice, Transgene
Integrity, and Copy Number-Pronuclear DNA microinjection experiments were performed to generate transgenic mice containing additional copies of the entire GK gene, including both promoter regions and all exons (28). Three founders were generated, two of which passed the transgene through the germline (animals 37 and 41). The founder that did not pass the transgene was presumed to be mosaic. Transgene inheritance patterns indicated that the exogenous GK gene sequences had integrated into the X chromosome in founder 37 and into an autosome in founder 41. Because of the large size of the transgene, both PCR and Southern blot experiments were performed to exclude fragmentation or rearrangement of the DNA upon integration into the genome. These analyses suggest that transgene integration occurred without any major rearrange-ments or deletions in both lines of mice (Fig. 1, A-C). Quantitation of Southern blot band intensities (Fig. 1C) indicated that line 37 had ϳ1 transgene copy per haploid genome, whereas line 41 had ϳ4 copies.
Expression and Regulation of the Transgene in Liver-To determine whether the downstream GK promoter in the transgene is expressed in liver, GK mRNA, GK activity, and GK immunoreactivity were assayed in transgenic and control mice. In fed animals from line 37, GK mRNA levels were elevated 1.8-fold in heterozygotes and 2.6-fold in homozygotes, compared with the controls (Fig. 2A, left). Similar expression expression levels were evident in line 41; GK mRNA was 1.8-and 2.4-fold greater in the heterozygous and homozygous transgenic mice, respectively (Fig. 2D). Increased hepatic GK activity (Fig. 2, B and E) and increased GK protein (Fig. 2, C and F) were also observed in both lines of mice.
To assess the regulation of the transgene in liver, GK mRNA levels were determined in mice fasted for 34 h or fed diets that varied in carbohydrate content. In the 34-h fasted mice GK mRNA expression was decreased in both the transgenic and control mice to very low levels ( Fig. 2A). In addition, hepatic GK mRNA was decreased by ϳ70% in mice fed a low carbohydrate diet, compared with those fed a high carbohydrate diet (data not shown). These results indicate that the GK transgene is expressed in liver, that there is an association between gene copy number and gene expression within a given line, and that the expression of the transgene in liver is affected by dietary manipulations known to alter expression of the endogenous GK gene (8, 24, 38 -40). Identification of GK mRNA and GK immunoreactivity in the liver of a transgene-rescued GK null mouse (as shown below) also provides evidence that the downstream GK promoter in the transgene is transcriptionally active in hepatocytes.
The transmission pattern of the GK transgene in line 37 revealed that it had integrated into the X chromosome, as

FIG. 2. Expression and regulation of the GK gene locus transgene in liver.
A, hepatic GK mRNA analysis for line 37 (n ϭ 6 -7/group; #, p Ͻ 0.01 compared with the fed group; *, p Ͻ 0.05 compared with control). Mice were fed ad libitum overnight or were starved for 34 h. Relative band intensities were normalized to the values of overnight fed control animals. B, hepatic GK activities for overnight fed, line 37 animals (n ϭ 9 -14/group; *, p Ͻ 0.01 compared with control). C, immunoreactive hepatic GK for line 37 (n ϭ 4 -5/group; *, p Ͻ 0.05 compared with control). The same fed liver extracts used to estimate activity were also assayed by Western blot, and GK band intensity was quantitated by densitometry. Values were normalized to controls. D, hepatic GK mRNA analysis for line 41 (n ϭ 10 -12/group; *, p Ͻ 0.05 compared with control). E, hepatic GK activities for line 41 (n ϭ 8 -14/group; *, p Ͻ 0.05 compared with control). F, immunoreactive hepatic GK for line 41 (n ϭ 4 -5/group; *, p Ͻ 0.05 compared with control). For RNA blot analysis, variation in RNA loading was corrected with a rat cyclophilin control probe. RNA and immunoblot results were normalized to control mouse values. Except for the fasted animals in A, all were fed ad libitum overnight. noted above. In this line, hepatic GK mRNA was elevated 1.8-fold in heterozygotes (3 gene copies) and 2.6-fold in homozygotes (4 gene copies) compared with controls (2 gene copies). Since the increase in hepatic GK mRNA was nearly proportional to the number of GK genes, the GK transgene appears to escape X chromosome inactivation. Other X-linked transgenes have also been shown to escape inactivation (41)(42)(43). While both lines of mice showed increased GK gene expression in liver, thus suggesting position-independent expression of the downstream promoter in the transgene, the design of these studies, which did not use a reporter gene strategy, makes it difficult to conclude whether expression is also copy numberdependent. The combination of position-independent, and copy number-dependent transgene expression are accepted criteria by which the presence of a locus control region within a gene locus is defined (44 -47).
Altered Subcellular GK Localization in Hepatocytes-Recent studies have revealed that GK translocates from the cytoplasm to nucleus in certain metabolic states (48,49). Thus, to examine the subcellular location of GK in the livers of transgenic mice, we performed immunocytochemical studies. In livers from fed control mice, GK immunoreactivity displayed a peri-central zonation pattern typical of GK and other glycolytic enzymes (50) with only a few faintly immunopositive nuclei being observed (Fig. 3A). In contrast, the livers of fed transgenic mice showed both more pronounced nuclear GK immunostaining (Fig. 3B) and cytosolic GK immunoreactivity that was more intense than the controls (Fig. 3, compare A and B). Approximately 5-fold more nuclei stained positively for GK in both homozygous and heterozygous transgenic livers (Fig. 3C). In addition, the peri-central zone of both nuclear and cytoplasmic GK immunoreactivity was expanded in transgenic livers (data not shown). While the functional significance of GK localization in the nucleus of the hepatocyte remains to be determined, the enhanced nuclear localization of GK in transgenic hepatocytes appears to reflect nuclear sequestration of excess GK by binding to GKRP, since the latter has been found to be a nuclear protein (48,51). Given that alterations in GK gene expression profoundly affect the plasma glucose concentration (1), increased binding of GK to GKRP in these mice may be a mechanism that protects the animals from the physiologic effects of excess hepatic GK catalytic activity.
GK Transgene Expression in Islets-To assess expression of the upstream GK promoter in the transgene, we used quantitative methods to examine GK immunoreactivity in pancreatic islets. Given that small fragments of the upstream GK promoter have previously been shown to be expressed in both the islet and other rare neuroendocrine cell types (13,52), we expected that a transgene with ϳ15 kb of additional 5Ј-flanking sequence for the upstream promoter would cause increased islet GK gene expression. Surprisingly, in freshly isolated pancreata, the amount of immunoreactive GK in islets from both lines of transgenic mice was less than the controls: 66% and 74%, respectively, for line 37 and 41 (Fig. 4, A and B). Representative islets from a line 37 mouse, compared with a control mouse, are shown in Fig. 4 (C and D). However, since glucose is known to regulate islet GK content (20,53), and since these mice have a hypoglycemic phenotype, as is described in detail elsewhere (1), decreased islet GK content may reflect differences in the plasma glucose concentrations in transgenic mice. To test this possibility, GK immunoreactivity was assessed FIG. 3. Immunocytochemical analysis of GK subcellular localization in hepatocytes. A, control liver from overnight fed mouse. A representative 1-mm 2 field is shown with a central vein near the center of the image. Faint GK immunoreactivity can be seen in a few nuclei as well as diffuse cytoplasmic GK immunoreactivity. Scale bar ϭ 50 M. B, corresponding field from a transgenic liver. C, quantitation of nuclear GK immunoreactivity (n ϭ 2 mice/group, ϳ10 fields/animal; *, p Ͻ 10 Ϫ8 compared with control). GKRP immunostaining patterns and numbers of positive nuclei were not different between transgenic and control mice (data not shown).

FIG. 4. Quantitative immunofluorescence analysis of islet GK content.
Immunoreactive GK content in pancreata from transgenic and control mice was determined as described under "Experimental Procedures." A, quantitation of islet GK immunoreactivity from line 37 mice (*, p Ͻ 10 Ϫ5 compared with control). B, quantitation of islet GK immunoreactivity from line 41 mice (*, p Ͻ 10 Ϫ5 compared with control). For both A and B, 2 mice/group and 8 -15 islets/animal were analyzed. C, representative confocal micrograph of an islet from a line 37 mouse. D, representative confocal micrograph of a control islet.
after culturing islets under equivalent glucose conditions. After culturing control and transgenic islets from line 37 in 20 mM glucose, 1.5-fold more immunoreactive GK was present in the transgenic islets (Table I), a difference that parallels the addition of the single extra GK gene in these mice (3 versus 2 gene copies). To further validate this result, differences in glucose metabolism were quantitatively assessed using two photon excitation microscopy to measure changes in NAD(P)H autofluorescence (37). The 1.5-fold increase in GK immunoreactivity observed after culturing the transgenic islets in 20 mM glucose was matched by a 1.4-fold increase in glucose-stimulated NAD(P)H autofluorescence. Finally, detection of GK immunoreactivity in ␤ cells of the transgene-rescued mouse also provides clear evidence that the GK gene locus transgene is expressed in the islet (see below).
These results indicate that transgenic islets, when exposed to the same glucose concentrations as controls, express a proportionally greater amount of GK. Moreover, like the liver, the amount of GK immunoreactivity also appears to be GK gene copy number-dependent. While an extensive analysis of transgene expression in other neuroendocrine cell types known to express the islet GK isoform was not performed (13), GK mRNA levels were elevated to a similar extent (ϳ1.6-fold) in a sample of pooled transgenic pituitaries, suggesting that the GK transgene is probably expressed in all locations where the endogenous upstream GK promoter is transcriptionally active.
Rescue of GK Null Lethality-The effects of complete inactivation of the endogenous GK gene by gene targeting have been reported by two different groups (6,7). While the time of death varies greatly in the two different studies, the availability of these mice provided an opportunity to directly test the expression of the GK transgene in the absence of endogenous GK gene expression. GK transgenic mice were were first crossed with the heterozygous null GK mice generated and characterized by Bali et al. (7) to generate animals that were heterozygous for both the GK transgene and targeted endogenous GK alleles. The double heterozygous mice were then intercrossed, and 56 offspring were generated and characterized. Because the genomic Southern blot analysis was complicated by the presence of up to four GK genes in some mice (Fig. 5A), variable band intensities, and a random insertion of the targeting vector in the genome of the GK null mice (see extra band for the neomycin resistance gene probe in Fig. 5A), all candidate rescued mice were test-mated to a wild type mouse to verify transmission of one null GK allele to all offspring.
From these matings, a single female mouse was identified that passed one null GK allele to all offspring (n ϭ 7, p Ͻ 0.01), as shown in Fig. 5B. This animal was killed for tissue analysis at ϳ12 weeks of age and was euglycemic at that time (104 mg/dl). GK immunostaining in both liver and pancreatic tissue sections in the rescued mouse were indistinguishable from a control mouse (Fig. 5C). RNA blot analysis also indicated that there was a similar amount of hepatic GK mRNA expressed compared with the control (Fig. 5D). There was no evidence of transgene misexpression in a survey of several other tissues that do not express endogenous GK (Fig. 5D). More importantly, the transgene-rescued mouse lived well past the time when the GK null mice died (6, 7) and survived metabolic stresses such as weaning and pregnancy. While the results of this experiment need to be interpreted with caution, it suggests  lanes 1 and 2) as well as the neomycin resistance gene (BglII digests; lanes 5 and 6) showed twice the band intensity for mouse 1318 compared with mouse 1319, suggesting homozygosity for the mutant GK allele. The transgene probe was used on BamHI-digested DNA (lanes 3 and 4). B, genotypes of the test cross of 1318 with a wild type mouse showing transmission of one targeted GK allele to all offspring (n ϭ 7, p Ͻ 0.01). This result provides genetic proof that animal 1318 was homozygous null at the endogenous GK gene. C, GK immunostaining of tissues from animal #1318 (rescued) and a wild type mouse. The patterns of GK immunoreactivity in fasted liver (both low and high magnification) and islets were indistinguishable. The scale bars equal 50 M. D, Northern blot of total RNA from various tissues of a control mouse (lanes 1-5), a line 37 transgenic mouse (lanes 6 -10), and the transgene-rescued mouse (lanes [11][12][13][14][15]. Lanes 1, 6, and 11 contain liver RNA; lanes 2, 7, and 12 contain white adipose tissue RNA; lanes 3, 8, and 13 contain heart RNA; lanes 4, 9, and 14 contain skeletal muscle RNA; lanes 5, 10, and 15 contain kidney RNA.

TABLE I
Quantitation of GK immunoreactivity and NAD(P)H autofluorescence in isolated islets after culture in 20 mM glucose Transgenic and control islets were isolated as described and cultured for 40 h in equimolar glucose conditions. The NAD(P)H response to 30 mM glucose was first determined as described, and then islets were fixed and quantitatively immunostained for GK expression level determination. The number of animals in each group is indicated in parentheses (*, p Ͻ 10 Ϫ6 , **, p Ͻ 0.05).

Control
Transgenic Ratio GK immunofluorescence intensity 101 Ϯ 3 (6) 153 Ϯ 4 (6)* 1.5 Ϯ 0.04 NAD(P)H autofluorescence response to 30 mM glucose 1.76 Ϯ 0.08 (2) 2.47 Ϯ 0.005 (2)** 1.41 Ϯ 0.06 that the GK gene locus transgene is able to fully compensate for the lack of endogenous GK gene expression in the GK null mice, thus providing additional evidence that the 83-kb DNA fragment tested in these mice defines the mouse GK gene locus. Conclusion-The analyses performed here indicate that an 83-kb DNA fragment spanning from about Ϫ55 to ϩ28 kilobase pairs relative to the liver transcription start site defines a functional mouse GK gene locus. By all criteria applied, expression of the GK gene locus transgene fully recapitulates the patterns of expression and regulation that characterize the endogenous GK gene. Given this, these mice can be used to explore the physiological effects of increased GK gene copy on blood glucose homeostasis, as we describe in a related study (1).