Acute Bidirectional Manipulation of Muscle Glucose Uptake by In Vivo Electrotransfer of Constructs Targeting Glucose Transporter Genes
http://www.100md.com
糖尿病学杂志 2005年第9期
1 Diabetes and Obesity Program, Garvan Institute of Medical Research, Sydney, Australia
2 Benitec, St. Lucia South, Queensland, Australia
3 School of Biotechnology and Biomolecular Sciences, Faculty of Science, University of New South Wales, Sydney, Australia
4 St. Vincent's Clinical School, Faculty of Medicine, University of New South Wales, Sydney, Australia
ABSTRACT
Analysis of conventional germ-line or tissue-specific gene manipulation in vivo is potentially confounded by developmental adaptation of animal physiology. We aimed to adapt the technique of in vivo electrotransfer (IVE) to alter local gene expression in skeletal muscle of rodents as a means of investigating the role of specific proteins in glucose metabolism in vivo. We utilized a square-wave electroporator to induce intracellular electrotransfer of DNA constructs injected into rat or mouse muscles and investigated the downstream effects. In initial studies, expression of green fluorescent protein reporter was induced in 53 ± 10% of muscle fibers peaking at 7 days, and importantly, the electrotransfer procedure itself did not impact upon the expression of stress proteins or our ability to detect a reduction in 2-deoxyglucose tracer uptake by electroporated muscle of high-fat-fed rats during hyperinsulinemic-euglycemic clamp. To demonstrate functional effects of electrotransfer of constructs targeting glucose transporters, we administered vectors encoding GLUT-1 cDNA and GLUT-4 short hairpin RNAs (shRNAs) to rodent muscles. IVE of the GLUT-1 gene resulted in a 57% increase in GLUT-1 protein, accompanied by a proportionate increase in basal 2-deoxyglucose tracer uptake into muscles of starved rats. IVE of vectors expressing two shRNAs for GLUT-4 demonstrated to reduce specific protein expression and 2-deoxyglucose tracer uptake in 3T3-L1 adipocytes into mouse muscle caused a 51% reduction in GLUT-4 protein, associated with attenuated clearance of tracer to muscle after a glucose load. These results confirm that glucose transporter expression is largely rate limiting for glucose uptake in vivo and highlight the utility of IVE for the acute manipulation of muscle gene expression in the study of the role of specific proteins in glucose metabolism.
In vivo genetic manipulation is a valuable modality to identify the roles of candidate genes in the development of insulin resistance and the metabolic syndrome. Genetic manipulation in vivo has been largely carried out to date using murine germ-line manipulation, resulting in the "knocking in" or "knocking out" of functional genes for the lifetime of the animal. However, interpretation of such data may be hampered by developmental adaptation to the genetic manipulation. Furthermore, the phenotype of global or whole-tissue knockouts or transgenics can be pronounced, leading to doubt regarding the true physiological relevance of the gene in question. Muscle-specific overexpression of GLUT-1 (1eC3), overexpression of GLUT-4 (2, 4, 5), or targeted disruption of GLUT-4 in muscle (6, 7) all resulted in profound changes in whole-body glucose homeostasis and/or insulin sensitivity that were apparent from an early age and often associated with specific secondary changes in nontarget tissues. These findings have formed a significant component of the evidence used in support of the hypothesis that glucose transport is rate determining in the process of glucose disposal into muscle. However, it could potentially be more informative to examine the effects of more acute and subtle changes in GLUT expression on glucose transport. In this and other areas, therefore, there is considerable scope for the development of a reliable technique that permits acute genetic modification of single muscles in postnatal or mature animals.
A promising approach to local genetic manipulation is through direct injection of a mammalian expression vector encoding a cDNA of interest combined with in vivo electrotransfer (IVE) (8eC10). However, although a few authors have utilized this technique to overexpress genes of interest in a research context (11, 12), its applicability to genetic manipulation of muscle in vivo as a tool in metabolic research has not been fully evaluated, especially with regard to its compatibility with standard physiological methods for evaluating tissue insulin resistance. Additionally, IVE could have additional potential to knock down expression of candidate insulin resistance genes in muscle through the use of RNA interference, mediated using specific short interfering RNAs (siRNAs). siRNAs are short RNA sequences that when introduced intracellularly are responsible for degradation of complementary messenger RNA sequences, resulting in a reduction in the corresponding protein (13). RNA interference may also be triggered using short hairpin RNAs (shRNAs) generated from RNA polymerase III (14). Since shRNAs are encoded by a DNA vector, they should achieve a more chronic time course of gene silencing than might be possible after a single administration of siRNAs.
Thus, in this study we aimed to determine whether the IVE procedure itself altered glucose metabolism, and to what extent acute overexpression or knockdown of specific glucose transporters by this means would alter glucose disposal to rodent skeletal muscle.
RESEARCH DESIGN AND METHODS
Vector construction.
pCMS-EGFP vector was purchased from Clontech (BD Biosciences, New South Wales, Australia). pCB6-hGLUT-1 was as previously reported (15). EH114 vector was a gift from Dr. Edna Hardeman. This vector was created by subcloning the 2.2-kb human skeletal muscle actin promoter (16) into the HindIII site of pcDNA3.1(+) (Invitrogen Australia, Victoria, Australia) to render expression from this vector skeletal muscle specific. EH114-EGFP vector was created by excising the EGFP sequence from pCMS-EGFP and ligating it into the ECoR I site of EH114. Molecular reagents were supplied by Promega (New South Wales, Australia) and New England Biolabs (Genesearch, Queensland, Australia).
Silencing constructs were generated by insertion of the shRNA cassette into either pU6.cass (17) or pBABE puro (18) and pBluescript SK+ (Stratagene, La Jolla, CA) using the long-range cloning technique (17). Each shRNA cassette contained a U6 promoter, the target hairpin, and an RNA polymerase III terminator potentially encoding a 3' UU sequence. The constructs were designed to express the following shRNAs targeting mouse GLUT-4: GH-1, 5'-GGUGAUUGAACAGAGCUACAAUGCAACGUGCUACACAAAUU-3'; GI-2, 5'-GUGAUUGAACAGAGCUACACCUGAAAAGGUU-3' (antisense sequences targeting GLUT-4 mRNA are underlined). Two control constructs were also prepared: one, EGFP-A, targeting EGFP (Invitrogen), 5'-GCUGACCCUGAAGUUCAUCUUCAAGAGACAGCUU-3'; the other, GD-1, an inactive construct targeting a mouse sequence, 5'-AACAUAUUCUCAGUGCCUGAGAUUGUGAGUUCAAGAGAUU-3'. Expression cassettes based on pU6.cass were used for IVE, while constructs based on pBABE puro were used to generate retrovirus and infect 3T3-L1 fibroblasts (19).
3T3-L1 transfection and immunofluorescence and immunoblotting.
Culture, differentiation, and infection of 3T3-L1 fibroblasts with retrovirus were as previously described (19). 3T3-L1 adipocytes infected with GLUT-4 and control shRNA constructs were grown on coverslips and used for microscopy as described (20). Immunofluorescence utilized monoclonal anti-GLUT4 antiserum (1:100 dilution) and Alexa 488-conjugated goat anti-mouse antibody (Molecular Probes, Eugene, OR). Control images were produced by incubation of the Alexa 488 antibody alone. The images were acquired using a Zeiss AxioVert 200M inverted microscope (Thornwood, NJ). Electrophoresis, immunoblotting, and enhanced chemiluminescence visualization of cell lysates were as previously described (19).
In vitro 2-deoxyglucose uptake assay.
Retroviral infected 3T3-L1 adipocytes plated in 12-well culture dishes were washed once and incubated for 2 h at 37°C with Krebs-Ringer phosphate buffer (25 mmol/l HEPES, pH 7.4, 120 mmol/l NaCl, 6 mmol/l KCl, 1.2 mmol/l MgSO4, 1 mmol/l CaCl2, 0.4 mmol/l NaH2PO4, and 0.6 mmol/l Na2HPO4) containing 0.2% (wt/vol) BSA. The incubation buffer was replaced with fresh Krebs-Ringer phosphate buffer, and some wells were stimulated with 100 nmol/l insulin. The assay was initiated by adding 9.25 MBq/sample 0.02 mmol/l 2-deoxy-D-[2,6-3H]glucose (Amersham Biosciences, Buckinghamshire, U.K.) 29 min after insulin stimulation. Adipocytes were then incubated at 37°C for 1 min, and the assay was terminated by washing the cells three times with ice-cold PBS. Adipocytes were solubilized with 1% Triton X-100 in PBS for 15 min, and incorporated radioactivity was determined by scintillation counting.
Animal maintenance and surgery.
Male Wistar rats or Swiss mice were obtained from the Animal Resources Centre (Perth, Australia) and acclimatized to their new surroundings for 1 week. Animals were maintained at 22 ± 0.5°C under a 12-h day/12-h night cycle and were fed a standard diet (Norco, Kempsey, Australia) (18% fat, 33% protein, and 48% carbohydrate as a percentage of total dietary energy) ad libitum. Where a high-fat diet was employed, half of the subject animals were fed a homemade diet contributing 45% fat, 20% protein, and 35% carbohydrate as energy for 4 weeks ad libitum, while control animals continued to eat chow. Approximately 1 week before study, the right and left jugular veins of rats designated to undergo clamp studies, or the right jugular alone for rats designated to undergo basal tracer uptake studies, were cannulated as previously described (21). Anesthesia was induced with 5% and maintained with 1eC2% halothane in oxygen, the surgical site was irrigated with bupivicaine (0.5 mg/100 g) before closure, and 5 mg/kg ketoprofen was administered to provide postoperative analgesia. Rats were single housed and handled daily for the following week to minimize stress. Body weight was recorded daily, and only those rats that had fully recovered their presurgery weight were subsequently studied. All experimental procedures were approved by the Garvan Institute/St. Vincent's Hospital Animal Experimentation Ethics Committee and were in accordance with the National Health and Medical Research Council of Australia Guidelines on Animal Experimentation.
In vivo electrotransfer.
The electrotransfer procedures used were adapted from those of Mir et al. (10) and Satkauskas et al. (22). For electrotransfer, vectors were propagated in selective media and DNA extracted, purified using Endotoxin-free Maxi- or Mega-Prep kits (Qiagen, Victoria, Australia), and resuspended in sterile 0.9% saline. Rats or mice to be electroporated were anesthetized as above and their hindlimbs shaved and prepared with a chlorhexidine-ethanol solution. Tibialis cranialis (TC) muscles were injected in oblique fashion transcutaneously along their length with 0.5 or 1 mg/ml total DNA in saline using an insulin syringe. Rats received six spaced 50-e蘬 injections and mice one 30-e蘬 injection. This was followed by the application of a pair of tweezer electrodes across the distal limb connected to an ECM-830 electroporator device (BTX, Holliston. MA). Two alternative protocols were used: eight 20-ms pulses of 200 V/cm at a frequency of 1 Hz (protocol 1) and one 800-V/cm, 100-ms pulse followed by four 800-V/cm, 100-ms pulses at 1 Hz (protocol 2). Animals were killed by carbon dioxide inhalation, and muscles were rapidly removed.
Muscle sections.
Where appropriate, transverse and longitudinal portions of muscle were dissected, mounted in Tissue-tek (Sakura Finetechnical, Tokyo, Japan), and snap frozen in liquid nitrogen-cooled isopentane. Ten-micrometer sections were cut on a cryostat and placed on Superfrost Plus slides, fixed in acetone and examined using a Zeiss AxioVert 200M fluorescence microscope. To quantify fiber transfection rate, five random fields were counted for positive or negative fluorescence per transverse section for each muscle.
Hyperinsulinemic-euglycemic clamp and tracer infusion.
Conscious rats were studied after 5eC7 h of fasting. Basal tracer uptake studies were conducted using a single jugular cannula connected to a sampling line between 0900 and 1000, while a second jugular cannula was also connected to an infusion line when clamping was required. Rats were allowed to acclimatize to the study cage for 30eC40 min. Hyperinsulinemic-euglycemic clamping of conscious rats or basal tracer uptake studies were performed using a protocol previously established in this laboratory (21), incorporating administration of a bolus injection of 2-deoxy-D-[2,6-3H]-glucose (Amersham Biosciences). At the end of each study, rats were killed by intravenous injection of pentobarbitone sodium (Nembutal; Abbott Laboratories, Sydney, Australia) and their muscles rapidly dissected and freeze-clamped using liquid nitrogen-cooled tongs. Plasma tracer disappearance was used to calculate Rd, and hepatic glucose output was estimated from the difference between clamp glucose infusion rate and Rd. The area under the tracer disappearance curve of 2-deoxy-D-[2,6,-3H] glucose together with the counts of phosphorylated [3H] deoxyglucose from individual muscles were used to calculate Rg' (23).
During clamps, plasma glucose was determined immediately using a glucose analyzer (YSI 2300; YSI, Yellow Springs, OH), with the remaining plasma being frozen and subsequently used for plasma insulin determination by radioimmunoassay (Linco Research, St. Charles, MO).
Combined intraperitoneal glucose tolerance test (IPGTT) and 2-deoxyglucose tracer uptake.
A 7.4 MBq/ml solution of 2-deoxy-D-[1-14C] glucose (Amersham Biosciences) in 50% glucose/0.9% saline was prepared and injected at 2 g/kg body weight, 0.37 MBq into overnight-starved mice (24). Blood was collected from the tail tip at 0, 15, 30, 45, 60, and 90 min postinjection and blood glucose measured immediately using an Accu-Check Advantage meter (Roche Diagnostics, New South Wales, Australia). Blood radioactivity was subsequently determined at each time point by liquid scintillation counting, and the area under the curve was calculated. This, together with the disintegrations per minute of phosphorylated 2-deoxy-D-[1-14C] glucose from individual muscles was used to calculate the clearance of tracer into each muscle.
Muscle lysates, SDS-PAGE, and immunoblotting.
Protein expression of molecules present in muscle or adipocytes was assessed by SDS-PAGE and quantification of Western blots of cell lysates. Whole-tissue lysates were prepared by Polytron or manual homogenization in radioimmunoprecipitation assay buffer (65 mmol/l tris, 150 mmol/l NaCl, 5 mmol/l EDTA, pH 7.4, 1% (vol/vol) NP-40 detergent, 0.5% (wt/vol) sodium deoxycholate, 0.1% (wt/vol) sodium dodecyl sulfate, 10% (v/v) glycerol, containing 25 e/ml leupeptin, 10 e/ml aprotinin, 2 mmol/l sodium orthovanadate, 1 mmol/l sodium pyrophosphate, 1 mmol/l ammonium molybdate, 10 mmol/l NaF, and 1 mmol/l polymethylsulphonyl fluoride), followed by incubation for 90 min at 4°C and centrifugation for 10 min at 12,000g. Protein content of supernatants was quantified using the Bradford method (Protein Assay kit; Bio-Rad laboratories, New South Wales, Australia), and aliquots containing 10eC25 e protein were denatured in Laemmli buffer for 5 min at 95°C or 30 min at 37°C. Proteins were resolved by SDS-PAGE electrophoresis and electrotransferred as previously described (25). Immunoblotting and quantitation were also as previously described (25). Green fluorescent protein (GFP) antibody was purchased from Molecular Probes, -actin antibody from Sigma (St. Louis, MO), and Akt and HSP-70 antibodies from Cell Signaling Technology (Beverley, MA). Antibodies against GLUT-1 and GLUT-4 have been described previously (26, 27).
Statistics.
All data are quoted as means ± SE. Comparisons between two treatment groups were made using Student's t test or the Mann-Whitney rank-sum, using paired statistics where appropriate. Comparisons between data from multiple treatment groups were made using one- or two-way ANOVA followed by Holm-Sidak post hoc analysis with repeated measures where appropriate. Non-normally distributed data were analyzed using Kruskal-Wallis one-way ANOVA on ranks followed by Dunn's test. Analyses were conducted using Sigma Stat v3.00 (SPSS, Chicago, IL), with P < 0.05 regarded as significant.
RESULTS
Expression of reporter gene in muscle fibers in vivo by electrotransfer.
Groups of rats were electroporated after injection of pCMS-EGFP vector and killed at various time points up to 3 weeks later. Fluorescence microscopy of transverse (Fig. 1A) and longitudinal (Fig. 1B) muscle sections demonstrated a muscle fiber transfection rate of 53 ± 10% (n = 5). Based upon the results from several cohorts of animals, a 40eC50% transfection rate is readily achievable. GFP measurement in muscle lysates 2, 7, and 12 days post-IVE (protocol 2) indicated that expression levels of electrotransfered protein peak around the 7-day time point (P = 0.013 effect of time since electrotransfer, P < 0.05 7 days vs. 2 days; n = 4 per group) (Fig. 1C).
Compatibility of the electrotransfer technique with physiological measures of glucose uptake.
Measurement of glucose tracer disposal under basal or euglycemic-hyperinsulinemic clamp conditions is the preferred method for assessment of muscle insulin sensitivity and glucose uptake. We wished to determine therefore if the IVE procedure itself confounded subsequent tracer uptake measurements in groups of rats fed either high-fat or standard laboratory diets for 4 weeks. Single TC muscles in each rat were injected with saline and electroporated (protocol 2), and glucose tracer disposal to paired TC muscles was measured under clamp conditions 7 days after electrotransfer. The presence of whole-body insulin resistance was confirmed in the high-fat-fed rats, as glucose infusion rate required to maintain euglycemia (25.1 ± 1.4 vs. 19.1 ± 1.5 mg · kgeC1 · mineC1; P = 0.008) and whole-body Rd (28.4 ± 2.6 vs. 22.2 ± 1.9 mg · kgeC1 · mineC1; P = 0.076) were reduced by high-fat feeding, while hepatic glucose output was increased (2.0 ± 1.0 vs. 4.6 ± 0.6 mg · kgeC1 · mineC1; P = 0.047). Clamp plasma insulin was not different between groups, being 167 ± 9 mU/l in high-fat-fed and 150 ± 5 mU/l in standard diet rats (P = 0.12, n = 8eC9 per group). High-fat feeding reduced tracer Rg' into TC muscles of rats by a mean 31% (P = 0.027) (Fig. 2A); however, there was no effect of IVE per se upon this. The Rg' data for paired TC muscles indicates this more clearly (P = 0.47) (Fig. 2B). Hence, the IVE procedure itself had no effect on the physiological assessment of insulin sensitivity in muscle 1 week later. Furthermore, at the 1-week time point there were no differences detected in expression of two proteins, which can be regarded as indexes of muscle stress between electroporated and untreated muscle (HSP-70 [Fig. 2C] or total and serine 473-phosphorylated Akt protein [Fig. 2D]), suggesting minimal induction of nonspecific pathophysiology in the tissue.
Functional overexpression of GLUT-1 in rat muscle in vivo.
We electroporated the right TC muscle of rats after injection of pCB6-hGLUT-1 and pCMS-EGFP and the left muscle after injection of pCB6 vector as control (protocol 1). In overnight-fasted rats after 7 days, GLUT-1 expression was increased in all the electroporated muscles versus paired controls (by a mean 57%; P = 0.016, n = 7) (Fig. 3A). Rg' was increased by a mean 46% (P = 0.05) in hGLUT-1 muscle versus control muscle (Fig. 3B), implying that the overexpression of GLUT-1 protein was associated with increased basal glucose uptake in TC muscle. There was a close correlation between the increase in GLUT-1 protein and the increase in Rg' between test and control muscles (r2 = 0.58, P = 0.046; Fig. 3C), further suggesting a causal link between the two. Furthermore, the level of coelectrotransfered GFP protein expressed by the test muscle and the difference in Rg' between the test and control muscles were also correlated (r2 = 0.61, P = 0.04; Fig. 3D), suggesting that both plasmids were transfected into muscle fibers with similar efficacy and that transcription from each promoter occurred in a constant ratio.
In vitro characterization of shRNAs targeted against GLUT-4 sequences.
To examine the effect of an acute reduction in GLUT-4 expression on glucose metabolism in muscle, we designed vectors that would encode shRNAs targeting GLUT-4 protein sequences. To verify their efficacy, GH-1 or GI-2 constructs were transfected into 3T3-L1 fibroblasts, which were subsequently differentiated into adipocytes. Immunofluorescence demonstrated that the expression level of GLUT-4 protein was reduced in GH-1eCtransfected cells compared with cells transfected with the control construct (Fig. 4A and B). This result was confirmed by immunoblotting of whole-cell lysates for GH-1, as GLUT-4 expression was 31 ± 7% of control (Fig. 4C). A similar suppression of GLUT-4 protein was achieved using GI-2, with a residual GLUT-4 expression of 27 ± 4% of control (Fig. 4D). 2-deoxyglucose uptake into cells transfected with GH1-siRNA was correspondingly reduced by 40% versus untransfected cells in three independent experiments (Fig. 4E).
Electrotransfer of GLUT-4 shRNAs into muscles of mice suppresses GLUT-4 protein expression and attenuates 2-deoxyglucose uptake.
We then aimed to assess the utility of the electrotransfer technique for functional suppression of GLUT-4 protein levels in muscle in vivo. Right TC muscles of mice were injected with 15 e GD-1 and left muscles with 15 e GH-1, both also receiving pCMS-EGFP and electroporated using protocol 1. Figure 5 shows GLUT-4 expression in paired muscles 1 (Fig. 5A), 7 (Fig. 5B), and 12 (Fig. 5C) days postelectrotransfer. Across all time points there was a mean 26% reduction in GLUT-4 expression (P = 0.007, n = 4 per group), which was most marked at the 7-day time point. In a later experiment, a second cohort of mice received 15 e GD-1 and 15 e GI-2, respectively, into right and left TC muscles using the same protocol, and protein expression of both GLUT-4 and GLUT-1 were assessed after 7 days. GLUT-4 protein was reduced by a mean 28% (P = 0.04, n = 7), whereas GLUT-1 protein was not altered, verifying the specificity of the construct and indicating that no compensatory overexpression of GLUT-1 occurred. Thus, both constructs were independently successful in knocking down endogenous GLUT-4 in vivo.
Next we assessed the effect of acute suppression of GLUT-4 protein on uptake of glucose tracer as part of an IPGTT in muscle injected with both GH-1 and GI-2 (15 e each) constructs versus contralateral muscles injected with GD-1 and EH114-GFP constructs and electroporated. GLUT-4 expression was consistently reduced in test muscles, by a mean of 51% (P = 0.004, n = 8), in the absence of any nonspecific effect upon housekeeping gene -actin (P = 0.23) (Fig. 6A). This effect was approximately additive over that of the two constructs alone. The IPGTT resulted in the expected effects on plasma glucose and tracer disappearance tissues over 90 min (Fig. 6B). Clearance of tracer to muscles transfected with the two GLUT-4 targeting constructs was reduced versus contralateral muscles by a mean 30% (P < 0.001) (Fig. 6C). Furthermore, there was a clear correlation between the ratios of GLUT-4 protein and glucose tracer clearance between paired TC muscles (r2 = 0.68, P = 0.012), implying a causal relationship between GLUT-4 suppression and attenuated glucose uptake. Thus, expression of a key muscle glucose disposal gene can be successfully knocked down in vivo using constructs encoding shRNAs, resulting in an altered physiological end point.
DISCUSSION
This study demonstrates the utility of IVE for both functional overexpression and suppression of glucose transporters in rodent skeletal muscle. Specifically, local direct manipulation of glucose transporter availability, in the absence of a perturbation of whole-body metabolism, resulted in substantial changes in basal and insulin-stimulated in vivo muscle glucose disposal. Furthermore, the work encompasses the first demonstration of the combination of IVE with gene silencing to produce functional changes in a target metabolic pathway in muscle in vivo.
IVE is emerging as a potential gene therapy modality in multiple tissues, despite its spatially and temporally limited effects (8), but has as yet been underutilized in basic research. Skeletal muscle is especially well suited to this procedure, being readily accessible and capable of regenerating any damage inflicted by a brief insult (28). Herein, we have demonstrated that transfection of 50% of fibers in a target muscle is readily achievable and that the resultant protein persists in the muscle for a minimum of 3 weeks, with peak expression around 7 days postelectroporation, consistent with previous reports (10, 29eC32). One to 2 days after electroporation, our results indicate that expression of electroporated cDNAs is relatively low, and it is undesirable to study the downstream effects of genetic manipulation before 7 days, as a degree of acute inflammation and vascular reaction is generated following electroporation (30, 33). However, as the electroporation procedure per se did not affect the expected attenuation of tracer uptake into muscle of high-fat-fed rats or alter the expression of indexes of muscle stress (Akt and HSP-70), we can conclude that any muscle damage is insufficient to confound downstream physiological measurements at this time point. Furthermore, although the data as presented here and our experience with other target genes confirms that transfection efficiency can be quite variable between animals, the use of a reporter gene and/or the restriction of downstream analysis to animals achieving a minimum level of gene overexpression or silencing should circumvent this as an issue.
Few reports have described acute silencing of genes in vivo by RNA interference (34eC37). Only one of these reports documented successful silencing in skeletal muscle and this only the very acute downregulation of a reporter gene (34). Herein, we have characterized two constructs encoding shRNAs targeting distinct GLUT-4 mRNA sequences, both in vitro and in vivo. This approach was adopted as it was likely that ongoing expression of shRNAs from a vector would result in a longer duration of silencing, as individual siRNA molecules are likely to be degraded within a short period of time (13, 37). In support of this, gene silencing by siRNAs was only recorded previously 1eC2 days after treatment (34, 36), while it was present after 10 days when vectors that expressed double-stranded siRNAs were used (35). Consistent with this, we found that GLUT-4 protein suppression by one shRNA constuct peaked 7 days postelectroporation, analogous to the results obtained for gene overexpression from similar vectors. Moreover, independent electrotransfer of a second shRNA construct targeting a distinct mRNA sequence achieved a similar effect upon GLUT-4 protein. When the two vectors were administered together, the mean protein suppression achieved was approximately additive, suggesting that a multiple targeting strategy might thus be generally more efficacious. However, these differences could also be accounted for by variations in electrotransfer efficiency.
Glucose disposal to skeletal muscle is largely responsible for clearance of a glucose load postprandially, and transport into myofibers is principally accomplished using the insulin-stimulated facilitative glucose transporter GLUT-4, while constitutive transport of glucose into muscle is thought to be mediated via GLUT-1 (38). Studies of transgenic mice overexpressing GLUT-1 in skeletal muscle have helped to establish this (2, 3). Consistent with these reports, we observed a close correlation between the degree of GLUT-1 overexpression in muscle of starved rats and the increase in glucose transport generated as a result.
There has been much debate over which component of the insulin-stimulated uptake and storage of glucose as glycogen in skeletal muscle is rate limiting. Of the three components of glucose disposal that have been considered, glucose transport, phosphorylation, and glycogen synthesis, glucose transport has been most frequently implicated. The close correlation between GLUT-4 suppression and the reduction in glucose tracer uptake observed in this study is consistent with this conclusion, although the relatively greater magnitude of the former change could imply additional levels of regulation are important, as has been previously suggested (5, 39). Some of the strongest evidence for glucose transport being rate limiting has come from transgenic overexpression or targeted disruption of GLUT-4 in mouse skeletal muscle. These genetic manipulations resulted in markedly increased (2, 4, 5) or decreased (6, 7) glucose disposal to muscle in mice. However, substantial whole-body alterations in glucose metabolism and/or insulin sensitivity were generated as part of each transgenic phenotype, often resulting in secondary effects on metabolism in other tissues. The IVE technique has the considerable advantage that individual muscles can be studied in an in vivo setting, but in the absence of whole-body physiological changes. Furthermore, acute genetic alteration can be carried out in mature animals, thus avoiding the potential confounding of interpretation caused by developmental adaptation to germ-line manipulation. Finally, IVE permits direct comparison of test muscles with their contralateral equivalents that have undergone electrotransfer of control DNA, permitting the use of powerful paired statistical methods to analyze results in each animal.
In summary, IVE is a technique that can be readily applied to acutely increase or decrease gene expression in single rodent skeletal muscles to study gene dosage effects in the absence of the confounding influences of developmental adaptation and alterations in whole-body physiology. Its compatibility with appropriate physiological measurements in muscle and with standard models of insulin resistance make it particularly suitable for the interrogation of candidate insulin sensitivity genes in this tissue. Using this technique, we have confirmed that under basal and insulin-stimulated conditions, the availability of glucose transporters is largely rate-limiting for glucose disposal into skeletal muscle.
ACKNOWLEDGMENTS
The authors are grateful to the National Health and Medical Research Council of Australia, the Diabetes Australia Research Trust, and the Rebecca L. Cooper Medical Research Foundation for their funding.
We thank Drs. Edna Hardeman and Mark Corbett, Children's Medical Research Institute, Westmead, NSW, Australia, for initial valuable advice and the EH114 vector. Also, we thank Dr. Lluis Mir, Institut Gustave-Roussy, Villejuif Cedex, France, for additional discussion regarding electroporation methodology; Dr. Sally Dunwoodie, Victor Chang Cardiac Research Institute, Darlinghurst, NSW, Australia, for loan of an Electroporator for initial studies; and to Jane Radford, University of Sydney, Australia, for sectioning of tissues. We are grateful to the staff of the Garvan Biological Testing Facility for animal care and to Andy Muirhead, Elaine Preston, and Nicolas Dzamko for technical assistance.
Key Words: GFP, green fluorescent protein HSP, heat shock protein IPGTT, intraperitoneal glucose tolerance test IVE, in vivo electrotransfer shRNA, short hairpin RNA siRNA, short interfering RNA TC, tibialis cranialis
REFERENCES
Ren JM, Barucci N, Marshall BA, Hansen P, Mueckler MM, Shulman GI: Transgenic mice overexpressing GLUT-1 protein in muscle exhibit increased muscle glycogenesis after exercise. Am J Physiol Endocrinol Metab278 :E588 eCE592,2000
Marshall BA, Mueckler MM: Differential effects of GLUT-1 or GLUT-4 overexpression on insulin responsiveness in transgenic mice. Am J Physiol267 :E738 eCE744,1994
Ren JM, Marshall BA, Gulve EA, Gao J, Johnson DW, Holloszy JO, Mueckler M: Evidence from transgenic mice that glucose transport is rate-limiting for glycogen deposition and glycolysis in skeletal muscle. J Biol Chem.268 :16113 eC16115,1993
Ren JM, Marshall BA, Mueckler MM, McCaleb M, Amatruda JM, Shulman GI: Overexpression of Glut4 protein in muscle increases basal and insulin-stimulated whole body glucose disposal in conscious mice. J Clin Invest95 :429 eC432,1995
Tsao TS, Burcelin R, Katz EB, Huang L, Charron MJ: Enhanced insulin action due to targeted GLUT4 overexpression exclusively in muscle. Diabetes45 :28 eC36,1996
Zisman A, Peroni OD, Abel ED, Michael MD, Mauvais-Jarvis F, Lowell BB, Wojtaszewski JF, Hirshman MF, Virkamaki A, Goodyear LJ, Kahn CR, Kahn BB: Targeted disruption of the glucose transporter 4 selectively in muscle causes insulin resistance and glucose intolerance. Nat Med6 :924 eC928,2000
Kotani K, Peroni OD, Minokoshi Y, Boss O, Kahn BB: GLUT-4 glucose transporter deficiency increases hepatic lipid production and peripheral lipid utilization. J Clin Invest114 :1666 eC1675,2004
Andre F, Mir LM: DNA electrotransfer: its principles and an updated review of its therapeutic applications. Gene Ther11 (Suppl. 1) :S33 eCS42,2004
Gehl J: Electroporation: theory and methods, perspectives for drug delivery, gene therapy and research. Acta Physiol Scand177 :437 eC447,2003
Mir LM, Bureau MF, Gehl J, Rangara R, Rouy D, Caillaud JM, Delaere P, Branellec D, Schwartz B, Scherman D: High-efficiency gene transfer into skeletal muscle mediated by electric pulses. Proc Natl Acad Sci U S A96 :4262 eC4267,1999
Fujii N, Boppart MD, Dufresne SD, Crowley PF, Jozsi AC, Sakamoto K, Yu H, Aschenbach WG, Kim S, Miyazaki H, Rui L, White MF, Hirshman MF, Goodyear LJ: Overexpression or ablation of JNK in skeletal muscle has no effect on glycogen synthase activity. Am J Physiol Cell Physiol287 :C200 eCC208,2004
Ho RC, Alcazar O, Fujii N, Hirshman MF, Goodyear LJ: p38 Gamma MAPK regulation of glucose transporter expression and glucose uptake in L6 myotubes and mouse skeletal muscle. Am J Physiol Regul Integr Comp Physiol286 :R342 eCR349,2003
Hannon GJ, Rossi JJ: Unlocking the potential of the human genome with RNA interference. Nature431 :371 eC378,2004
Brummelkamp TR, Bernards R, Agami R: A system for stable expression of short interfering RNAs in mammalian cells. Science296 :550 eC553,2002
Inukai K, Shewan AM, Pascoe WS, Katayama S, James DE, Oka Y: Carboxy terminus of glucose transporter 3 contains an apical membrane targeting domain. Mol Endocrinol18 :339 eC349,2004
Muscat GE, Perry S, Prentice H, Kedes L: The human skeletal alpha-actin gene is regulated by a muscle-specific enhancer that binds three nuclear factors. Gene Expr2 :111 eC126,1992
Rice RR, Muirhead AN, Harrison BT, Kassianos AJ, Sedlak PL, Maugeri NJ, Goss PJ, Davey JR, James DE, Graham MW: Simple, robust strategies for generating ddRNAi constructs. Methods Enzymol324 :405 eC419,2005
Morgenstern JP, Land H: Advanced mammalian gene transfer: high titre retroviral vectors with multiple drug selection markers and a complementary helper-free packaging cell line. Nucleic Acid Res18 :3587 eC3596,1990
Shewan AM, Marsh BJ, Melvin DR, Martin S, Gould GW, James DE: The cytosolic C-terminus of the glucose transporter GLUT4 contains an acidic cluster endosomal targeting motif distal to the dileucine signal. Biochem J350 :99 eC107,2000
Shewan AM, van Dam EM, Martin S, Luen TB, Hong W, Bryant NJ, James DE: GLUT4 recycles via a trans-golgi network (TGN) subdomain enriched in syntaxins 6 and 16 but not TGN38: involvement of an acidic targeting motif. Mol Biol Cell14 :973 eC986,2003
Clark PW, Jenkins AB, Kraegen EW: Pentobarbital reduces basal liver glucose output and its insulin suppression in rats. Am J Physiol258 :E701 eCE707,1990
Satkauskas S, Bureau MF, Puc M, Mahfoudi A, Scherman D, Miklavcic D, Mir LM: Mechanisms of in vivo DNA electrotransfer: respective contributions of cell electropermeabilization and DNA electrophoresis. Mol Ther5 :133 eC140,2002
Kraegen EW, James DE, Jenkins AB, Chisholm DJ: Dose-response curves for in vivo insulin sensitivity in individual tissues in rats. Am J Physiol248 :E353 eCE362,1985
Crosson SM, Khan A, Printen J, Pessin JE, Saltiel AR: PTG gene deletion causes impaired glycogen synthesis and developmental insulin resistance. J Clin Invest111 :1423 eC1432,2003
Cleasby ME, Dzamko N, Hegarty BD, Cooney GJ, Kraegen EW, Ye JM: Metformin prevents the development of acute lipid-induced insulin resistance in the rat through altered hepatic signaling mechanisms. Diabetes53 :3258 eC3266,2004
James DE, Strube M, Mueckler M: Molecular cloning and characterization of an insulin-regulatable glucose transporter. Nature338 :83 eC87,1989
Piper RC, Hess LJ, James DE: Differential sorting of two glucose transporters expressed in insulin-sensitive cells. Am J Physiol260 :C570 eCC580,1991
Charge SB, Rudnicki MA: Cellular and molecular regulation of muscle regeneration. Physiol Rev84 :209 eC238,2004
Dona M, Sandri M, Rossini K, Dell’Aica I, Podhorska-Okolow M, Carraro U: Functional in vivo gene transfer into the myofibers of adult skeletal muscle. Biochem Biophys Res Commun312 :1132 eC1138,2003
Molnar MJ, Gilbert R, Lu Y, Liu AB, Guo A, Larochelle N, Orlopp K, Lochmuller H, Petrof BJ, Nalbantoglu J, Karpati G: Factors influencing the efficacy, longevity, and safety of electroporation-assisted plasmid-based gene transfer into mouse muscles. Mol Ther10 :447 eC455,2004
Vilquin JT, Kennel PF, Paturneau-Jouas M, Chapdelaine P, Boissel N, Delaere P, Tremblay JP, Scherman D, Fiszman MY, Schwartz K: Electrotransfer of naked DNA in the skeletal muscles of animal models of muscular dystrophies. Gene Ther8 :1097 eC1107,2001
Maruyama H, Sugawa M, Moriguchi Y, Imazeki I, Ishikawa Y, Ataka K, Hasegawa S, Ito Y, Higuchi N, Kazama JJ, Gejyo F, Miyazaki JI: Continuous erythropoietin delivery by muscle-targeted gene transfer using in vivo electroporation. Human Gene Ther11 :429 eC437,2000
Gehl J, Skovsgaard T, Mir LM: Vascular reactions to in vivo electroporation: characterization and consequences for drug and gene delivery. Biochim Biophys Acta1569 :51 eC58,2002
Kishida T, Asada H, Gojo S, Ohashi S, Shin-Ya M, Yasutomi K, Terauchi R, Takahashi KA, Kubo T, Imanishi J, Mazda O: Sequence-specific gene silencing in murine muscle induced by electroporation-mediated transfer of short interfering RNA. J Gene Med6 :105 eC110,2004
Matsuda T, Cepko CL: Electroporation and RNA interference in the rodent retina in vivo and in vitro. Proc Natl Acad Sci U S A 2003
Soutschek J, Akinc A, Bramlage B, Charisse K, Constien R, Donoghue M, Elbashir S, Geick A, Hadwiger P, Harborth J, John M, Kesavan V, Lavine G, Pandey RK, Racie T, Rajeev KG, Rohl I, Toudjarska I, Wang G, Wuschko S, Bumcrot D, Koteliansky V, Limmer S, Manoharan M, Vornlocher HP: Therapeutic silencing of an endogenous gene by systemic administration of modified siRNAs. Nature432 :173 eC178,2004
Layzer JM, McCaffrey AP, Tanner AK, Huang Z, Kay MA, Sullenger BA: In vivo activity of nuclease-resistant siRNAs. RNA10 :766 eC771,2004
Olson AL, Pessin JE: Structure, function, and regulation of the mammalian facilitative glucose transporter gene family. Annu Rev Nutr16 :235 eC256,1996
Fisher JS, Nolte LA, Kawanaka K, Han DH, Jones TE, Holloszy JO: Glucose transport rate and glycogen synthase activity both limit skeletal muscle glycogen accumulation. Am J Physiol Endocrinol Metab282 :E1214 eCE1221,2002(Mark E. Cleasby, Jonathan)
2 Benitec, St. Lucia South, Queensland, Australia
3 School of Biotechnology and Biomolecular Sciences, Faculty of Science, University of New South Wales, Sydney, Australia
4 St. Vincent's Clinical School, Faculty of Medicine, University of New South Wales, Sydney, Australia
ABSTRACT
Analysis of conventional germ-line or tissue-specific gene manipulation in vivo is potentially confounded by developmental adaptation of animal physiology. We aimed to adapt the technique of in vivo electrotransfer (IVE) to alter local gene expression in skeletal muscle of rodents as a means of investigating the role of specific proteins in glucose metabolism in vivo. We utilized a square-wave electroporator to induce intracellular electrotransfer of DNA constructs injected into rat or mouse muscles and investigated the downstream effects. In initial studies, expression of green fluorescent protein reporter was induced in 53 ± 10% of muscle fibers peaking at 7 days, and importantly, the electrotransfer procedure itself did not impact upon the expression of stress proteins or our ability to detect a reduction in 2-deoxyglucose tracer uptake by electroporated muscle of high-fat-fed rats during hyperinsulinemic-euglycemic clamp. To demonstrate functional effects of electrotransfer of constructs targeting glucose transporters, we administered vectors encoding GLUT-1 cDNA and GLUT-4 short hairpin RNAs (shRNAs) to rodent muscles. IVE of the GLUT-1 gene resulted in a 57% increase in GLUT-1 protein, accompanied by a proportionate increase in basal 2-deoxyglucose tracer uptake into muscles of starved rats. IVE of vectors expressing two shRNAs for GLUT-4 demonstrated to reduce specific protein expression and 2-deoxyglucose tracer uptake in 3T3-L1 adipocytes into mouse muscle caused a 51% reduction in GLUT-4 protein, associated with attenuated clearance of tracer to muscle after a glucose load. These results confirm that glucose transporter expression is largely rate limiting for glucose uptake in vivo and highlight the utility of IVE for the acute manipulation of muscle gene expression in the study of the role of specific proteins in glucose metabolism.
In vivo genetic manipulation is a valuable modality to identify the roles of candidate genes in the development of insulin resistance and the metabolic syndrome. Genetic manipulation in vivo has been largely carried out to date using murine germ-line manipulation, resulting in the "knocking in" or "knocking out" of functional genes for the lifetime of the animal. However, interpretation of such data may be hampered by developmental adaptation to the genetic manipulation. Furthermore, the phenotype of global or whole-tissue knockouts or transgenics can be pronounced, leading to doubt regarding the true physiological relevance of the gene in question. Muscle-specific overexpression of GLUT-1 (1eC3), overexpression of GLUT-4 (2, 4, 5), or targeted disruption of GLUT-4 in muscle (6, 7) all resulted in profound changes in whole-body glucose homeostasis and/or insulin sensitivity that were apparent from an early age and often associated with specific secondary changes in nontarget tissues. These findings have formed a significant component of the evidence used in support of the hypothesis that glucose transport is rate determining in the process of glucose disposal into muscle. However, it could potentially be more informative to examine the effects of more acute and subtle changes in GLUT expression on glucose transport. In this and other areas, therefore, there is considerable scope for the development of a reliable technique that permits acute genetic modification of single muscles in postnatal or mature animals.
A promising approach to local genetic manipulation is through direct injection of a mammalian expression vector encoding a cDNA of interest combined with in vivo electrotransfer (IVE) (8eC10). However, although a few authors have utilized this technique to overexpress genes of interest in a research context (11, 12), its applicability to genetic manipulation of muscle in vivo as a tool in metabolic research has not been fully evaluated, especially with regard to its compatibility with standard physiological methods for evaluating tissue insulin resistance. Additionally, IVE could have additional potential to knock down expression of candidate insulin resistance genes in muscle through the use of RNA interference, mediated using specific short interfering RNAs (siRNAs). siRNAs are short RNA sequences that when introduced intracellularly are responsible for degradation of complementary messenger RNA sequences, resulting in a reduction in the corresponding protein (13). RNA interference may also be triggered using short hairpin RNAs (shRNAs) generated from RNA polymerase III (14). Since shRNAs are encoded by a DNA vector, they should achieve a more chronic time course of gene silencing than might be possible after a single administration of siRNAs.
Thus, in this study we aimed to determine whether the IVE procedure itself altered glucose metabolism, and to what extent acute overexpression or knockdown of specific glucose transporters by this means would alter glucose disposal to rodent skeletal muscle.
RESEARCH DESIGN AND METHODS
Vector construction.
pCMS-EGFP vector was purchased from Clontech (BD Biosciences, New South Wales, Australia). pCB6-hGLUT-1 was as previously reported (15). EH114 vector was a gift from Dr. Edna Hardeman. This vector was created by subcloning the 2.2-kb human skeletal muscle actin promoter (16) into the HindIII site of pcDNA3.1(+) (Invitrogen Australia, Victoria, Australia) to render expression from this vector skeletal muscle specific. EH114-EGFP vector was created by excising the EGFP sequence from pCMS-EGFP and ligating it into the ECoR I site of EH114. Molecular reagents were supplied by Promega (New South Wales, Australia) and New England Biolabs (Genesearch, Queensland, Australia).
Silencing constructs were generated by insertion of the shRNA cassette into either pU6.cass (17) or pBABE puro (18) and pBluescript SK+ (Stratagene, La Jolla, CA) using the long-range cloning technique (17). Each shRNA cassette contained a U6 promoter, the target hairpin, and an RNA polymerase III terminator potentially encoding a 3' UU sequence. The constructs were designed to express the following shRNAs targeting mouse GLUT-4: GH-1, 5'-GGUGAUUGAACAGAGCUACAAUGCAACGUGCUACACAAAUU-3'; GI-2, 5'-GUGAUUGAACAGAGCUACACCUGAAAAGGUU-3' (antisense sequences targeting GLUT-4 mRNA are underlined). Two control constructs were also prepared: one, EGFP-A, targeting EGFP (Invitrogen), 5'-GCUGACCCUGAAGUUCAUCUUCAAGAGACAGCUU-3'; the other, GD-1, an inactive construct targeting a mouse sequence, 5'-AACAUAUUCUCAGUGCCUGAGAUUGUGAGUUCAAGAGAUU-3'. Expression cassettes based on pU6.cass were used for IVE, while constructs based on pBABE puro were used to generate retrovirus and infect 3T3-L1 fibroblasts (19).
3T3-L1 transfection and immunofluorescence and immunoblotting.
Culture, differentiation, and infection of 3T3-L1 fibroblasts with retrovirus were as previously described (19). 3T3-L1 adipocytes infected with GLUT-4 and control shRNA constructs were grown on coverslips and used for microscopy as described (20). Immunofluorescence utilized monoclonal anti-GLUT4 antiserum (1:100 dilution) and Alexa 488-conjugated goat anti-mouse antibody (Molecular Probes, Eugene, OR). Control images were produced by incubation of the Alexa 488 antibody alone. The images were acquired using a Zeiss AxioVert 200M inverted microscope (Thornwood, NJ). Electrophoresis, immunoblotting, and enhanced chemiluminescence visualization of cell lysates were as previously described (19).
In vitro 2-deoxyglucose uptake assay.
Retroviral infected 3T3-L1 adipocytes plated in 12-well culture dishes were washed once and incubated for 2 h at 37°C with Krebs-Ringer phosphate buffer (25 mmol/l HEPES, pH 7.4, 120 mmol/l NaCl, 6 mmol/l KCl, 1.2 mmol/l MgSO4, 1 mmol/l CaCl2, 0.4 mmol/l NaH2PO4, and 0.6 mmol/l Na2HPO4) containing 0.2% (wt/vol) BSA. The incubation buffer was replaced with fresh Krebs-Ringer phosphate buffer, and some wells were stimulated with 100 nmol/l insulin. The assay was initiated by adding 9.25 MBq/sample 0.02 mmol/l 2-deoxy-D-[2,6-3H]glucose (Amersham Biosciences, Buckinghamshire, U.K.) 29 min after insulin stimulation. Adipocytes were then incubated at 37°C for 1 min, and the assay was terminated by washing the cells three times with ice-cold PBS. Adipocytes were solubilized with 1% Triton X-100 in PBS for 15 min, and incorporated radioactivity was determined by scintillation counting.
Animal maintenance and surgery.
Male Wistar rats or Swiss mice were obtained from the Animal Resources Centre (Perth, Australia) and acclimatized to their new surroundings for 1 week. Animals were maintained at 22 ± 0.5°C under a 12-h day/12-h night cycle and were fed a standard diet (Norco, Kempsey, Australia) (18% fat, 33% protein, and 48% carbohydrate as a percentage of total dietary energy) ad libitum. Where a high-fat diet was employed, half of the subject animals were fed a homemade diet contributing 45% fat, 20% protein, and 35% carbohydrate as energy for 4 weeks ad libitum, while control animals continued to eat chow. Approximately 1 week before study, the right and left jugular veins of rats designated to undergo clamp studies, or the right jugular alone for rats designated to undergo basal tracer uptake studies, were cannulated as previously described (21). Anesthesia was induced with 5% and maintained with 1eC2% halothane in oxygen, the surgical site was irrigated with bupivicaine (0.5 mg/100 g) before closure, and 5 mg/kg ketoprofen was administered to provide postoperative analgesia. Rats were single housed and handled daily for the following week to minimize stress. Body weight was recorded daily, and only those rats that had fully recovered their presurgery weight were subsequently studied. All experimental procedures were approved by the Garvan Institute/St. Vincent's Hospital Animal Experimentation Ethics Committee and were in accordance with the National Health and Medical Research Council of Australia Guidelines on Animal Experimentation.
In vivo electrotransfer.
The electrotransfer procedures used were adapted from those of Mir et al. (10) and Satkauskas et al. (22). For electrotransfer, vectors were propagated in selective media and DNA extracted, purified using Endotoxin-free Maxi- or Mega-Prep kits (Qiagen, Victoria, Australia), and resuspended in sterile 0.9% saline. Rats or mice to be electroporated were anesthetized as above and their hindlimbs shaved and prepared with a chlorhexidine-ethanol solution. Tibialis cranialis (TC) muscles were injected in oblique fashion transcutaneously along their length with 0.5 or 1 mg/ml total DNA in saline using an insulin syringe. Rats received six spaced 50-e蘬 injections and mice one 30-e蘬 injection. This was followed by the application of a pair of tweezer electrodes across the distal limb connected to an ECM-830 electroporator device (BTX, Holliston. MA). Two alternative protocols were used: eight 20-ms pulses of 200 V/cm at a frequency of 1 Hz (protocol 1) and one 800-V/cm, 100-ms pulse followed by four 800-V/cm, 100-ms pulses at 1 Hz (protocol 2). Animals were killed by carbon dioxide inhalation, and muscles were rapidly removed.
Muscle sections.
Where appropriate, transverse and longitudinal portions of muscle were dissected, mounted in Tissue-tek (Sakura Finetechnical, Tokyo, Japan), and snap frozen in liquid nitrogen-cooled isopentane. Ten-micrometer sections were cut on a cryostat and placed on Superfrost Plus slides, fixed in acetone and examined using a Zeiss AxioVert 200M fluorescence microscope. To quantify fiber transfection rate, five random fields were counted for positive or negative fluorescence per transverse section for each muscle.
Hyperinsulinemic-euglycemic clamp and tracer infusion.
Conscious rats were studied after 5eC7 h of fasting. Basal tracer uptake studies were conducted using a single jugular cannula connected to a sampling line between 0900 and 1000, while a second jugular cannula was also connected to an infusion line when clamping was required. Rats were allowed to acclimatize to the study cage for 30eC40 min. Hyperinsulinemic-euglycemic clamping of conscious rats or basal tracer uptake studies were performed using a protocol previously established in this laboratory (21), incorporating administration of a bolus injection of 2-deoxy-D-[2,6-3H]-glucose (Amersham Biosciences). At the end of each study, rats were killed by intravenous injection of pentobarbitone sodium (Nembutal; Abbott Laboratories, Sydney, Australia) and their muscles rapidly dissected and freeze-clamped using liquid nitrogen-cooled tongs. Plasma tracer disappearance was used to calculate Rd, and hepatic glucose output was estimated from the difference between clamp glucose infusion rate and Rd. The area under the tracer disappearance curve of 2-deoxy-D-[2,6,-3H] glucose together with the counts of phosphorylated [3H] deoxyglucose from individual muscles were used to calculate Rg' (23).
During clamps, plasma glucose was determined immediately using a glucose analyzer (YSI 2300; YSI, Yellow Springs, OH), with the remaining plasma being frozen and subsequently used for plasma insulin determination by radioimmunoassay (Linco Research, St. Charles, MO).
Combined intraperitoneal glucose tolerance test (IPGTT) and 2-deoxyglucose tracer uptake.
A 7.4 MBq/ml solution of 2-deoxy-D-[1-14C] glucose (Amersham Biosciences) in 50% glucose/0.9% saline was prepared and injected at 2 g/kg body weight, 0.37 MBq into overnight-starved mice (24). Blood was collected from the tail tip at 0, 15, 30, 45, 60, and 90 min postinjection and blood glucose measured immediately using an Accu-Check Advantage meter (Roche Diagnostics, New South Wales, Australia). Blood radioactivity was subsequently determined at each time point by liquid scintillation counting, and the area under the curve was calculated. This, together with the disintegrations per minute of phosphorylated 2-deoxy-D-[1-14C] glucose from individual muscles was used to calculate the clearance of tracer into each muscle.
Muscle lysates, SDS-PAGE, and immunoblotting.
Protein expression of molecules present in muscle or adipocytes was assessed by SDS-PAGE and quantification of Western blots of cell lysates. Whole-tissue lysates were prepared by Polytron or manual homogenization in radioimmunoprecipitation assay buffer (65 mmol/l tris, 150 mmol/l NaCl, 5 mmol/l EDTA, pH 7.4, 1% (vol/vol) NP-40 detergent, 0.5% (wt/vol) sodium deoxycholate, 0.1% (wt/vol) sodium dodecyl sulfate, 10% (v/v) glycerol, containing 25 e/ml leupeptin, 10 e/ml aprotinin, 2 mmol/l sodium orthovanadate, 1 mmol/l sodium pyrophosphate, 1 mmol/l ammonium molybdate, 10 mmol/l NaF, and 1 mmol/l polymethylsulphonyl fluoride), followed by incubation for 90 min at 4°C and centrifugation for 10 min at 12,000g. Protein content of supernatants was quantified using the Bradford method (Protein Assay kit; Bio-Rad laboratories, New South Wales, Australia), and aliquots containing 10eC25 e protein were denatured in Laemmli buffer for 5 min at 95°C or 30 min at 37°C. Proteins were resolved by SDS-PAGE electrophoresis and electrotransferred as previously described (25). Immunoblotting and quantitation were also as previously described (25). Green fluorescent protein (GFP) antibody was purchased from Molecular Probes, -actin antibody from Sigma (St. Louis, MO), and Akt and HSP-70 antibodies from Cell Signaling Technology (Beverley, MA). Antibodies against GLUT-1 and GLUT-4 have been described previously (26, 27).
Statistics.
All data are quoted as means ± SE. Comparisons between two treatment groups were made using Student's t test or the Mann-Whitney rank-sum, using paired statistics where appropriate. Comparisons between data from multiple treatment groups were made using one- or two-way ANOVA followed by Holm-Sidak post hoc analysis with repeated measures where appropriate. Non-normally distributed data were analyzed using Kruskal-Wallis one-way ANOVA on ranks followed by Dunn's test. Analyses were conducted using Sigma Stat v3.00 (SPSS, Chicago, IL), with P < 0.05 regarded as significant.
RESULTS
Expression of reporter gene in muscle fibers in vivo by electrotransfer.
Groups of rats were electroporated after injection of pCMS-EGFP vector and killed at various time points up to 3 weeks later. Fluorescence microscopy of transverse (Fig. 1A) and longitudinal (Fig. 1B) muscle sections demonstrated a muscle fiber transfection rate of 53 ± 10% (n = 5). Based upon the results from several cohorts of animals, a 40eC50% transfection rate is readily achievable. GFP measurement in muscle lysates 2, 7, and 12 days post-IVE (protocol 2) indicated that expression levels of electrotransfered protein peak around the 7-day time point (P = 0.013 effect of time since electrotransfer, P < 0.05 7 days vs. 2 days; n = 4 per group) (Fig. 1C).
Compatibility of the electrotransfer technique with physiological measures of glucose uptake.
Measurement of glucose tracer disposal under basal or euglycemic-hyperinsulinemic clamp conditions is the preferred method for assessment of muscle insulin sensitivity and glucose uptake. We wished to determine therefore if the IVE procedure itself confounded subsequent tracer uptake measurements in groups of rats fed either high-fat or standard laboratory diets for 4 weeks. Single TC muscles in each rat were injected with saline and electroporated (protocol 2), and glucose tracer disposal to paired TC muscles was measured under clamp conditions 7 days after electrotransfer. The presence of whole-body insulin resistance was confirmed in the high-fat-fed rats, as glucose infusion rate required to maintain euglycemia (25.1 ± 1.4 vs. 19.1 ± 1.5 mg · kgeC1 · mineC1; P = 0.008) and whole-body Rd (28.4 ± 2.6 vs. 22.2 ± 1.9 mg · kgeC1 · mineC1; P = 0.076) were reduced by high-fat feeding, while hepatic glucose output was increased (2.0 ± 1.0 vs. 4.6 ± 0.6 mg · kgeC1 · mineC1; P = 0.047). Clamp plasma insulin was not different between groups, being 167 ± 9 mU/l in high-fat-fed and 150 ± 5 mU/l in standard diet rats (P = 0.12, n = 8eC9 per group). High-fat feeding reduced tracer Rg' into TC muscles of rats by a mean 31% (P = 0.027) (Fig. 2A); however, there was no effect of IVE per se upon this. The Rg' data for paired TC muscles indicates this more clearly (P = 0.47) (Fig. 2B). Hence, the IVE procedure itself had no effect on the physiological assessment of insulin sensitivity in muscle 1 week later. Furthermore, at the 1-week time point there were no differences detected in expression of two proteins, which can be regarded as indexes of muscle stress between electroporated and untreated muscle (HSP-70 [Fig. 2C] or total and serine 473-phosphorylated Akt protein [Fig. 2D]), suggesting minimal induction of nonspecific pathophysiology in the tissue.
Functional overexpression of GLUT-1 in rat muscle in vivo.
We electroporated the right TC muscle of rats after injection of pCB6-hGLUT-1 and pCMS-EGFP and the left muscle after injection of pCB6 vector as control (protocol 1). In overnight-fasted rats after 7 days, GLUT-1 expression was increased in all the electroporated muscles versus paired controls (by a mean 57%; P = 0.016, n = 7) (Fig. 3A). Rg' was increased by a mean 46% (P = 0.05) in hGLUT-1 muscle versus control muscle (Fig. 3B), implying that the overexpression of GLUT-1 protein was associated with increased basal glucose uptake in TC muscle. There was a close correlation between the increase in GLUT-1 protein and the increase in Rg' between test and control muscles (r2 = 0.58, P = 0.046; Fig. 3C), further suggesting a causal link between the two. Furthermore, the level of coelectrotransfered GFP protein expressed by the test muscle and the difference in Rg' between the test and control muscles were also correlated (r2 = 0.61, P = 0.04; Fig. 3D), suggesting that both plasmids were transfected into muscle fibers with similar efficacy and that transcription from each promoter occurred in a constant ratio.
In vitro characterization of shRNAs targeted against GLUT-4 sequences.
To examine the effect of an acute reduction in GLUT-4 expression on glucose metabolism in muscle, we designed vectors that would encode shRNAs targeting GLUT-4 protein sequences. To verify their efficacy, GH-1 or GI-2 constructs were transfected into 3T3-L1 fibroblasts, which were subsequently differentiated into adipocytes. Immunofluorescence demonstrated that the expression level of GLUT-4 protein was reduced in GH-1eCtransfected cells compared with cells transfected with the control construct (Fig. 4A and B). This result was confirmed by immunoblotting of whole-cell lysates for GH-1, as GLUT-4 expression was 31 ± 7% of control (Fig. 4C). A similar suppression of GLUT-4 protein was achieved using GI-2, with a residual GLUT-4 expression of 27 ± 4% of control (Fig. 4D). 2-deoxyglucose uptake into cells transfected with GH1-siRNA was correspondingly reduced by 40% versus untransfected cells in three independent experiments (Fig. 4E).
Electrotransfer of GLUT-4 shRNAs into muscles of mice suppresses GLUT-4 protein expression and attenuates 2-deoxyglucose uptake.
We then aimed to assess the utility of the electrotransfer technique for functional suppression of GLUT-4 protein levels in muscle in vivo. Right TC muscles of mice were injected with 15 e GD-1 and left muscles with 15 e GH-1, both also receiving pCMS-EGFP and electroporated using protocol 1. Figure 5 shows GLUT-4 expression in paired muscles 1 (Fig. 5A), 7 (Fig. 5B), and 12 (Fig. 5C) days postelectrotransfer. Across all time points there was a mean 26% reduction in GLUT-4 expression (P = 0.007, n = 4 per group), which was most marked at the 7-day time point. In a later experiment, a second cohort of mice received 15 e GD-1 and 15 e GI-2, respectively, into right and left TC muscles using the same protocol, and protein expression of both GLUT-4 and GLUT-1 were assessed after 7 days. GLUT-4 protein was reduced by a mean 28% (P = 0.04, n = 7), whereas GLUT-1 protein was not altered, verifying the specificity of the construct and indicating that no compensatory overexpression of GLUT-1 occurred. Thus, both constructs were independently successful in knocking down endogenous GLUT-4 in vivo.
Next we assessed the effect of acute suppression of GLUT-4 protein on uptake of glucose tracer as part of an IPGTT in muscle injected with both GH-1 and GI-2 (15 e each) constructs versus contralateral muscles injected with GD-1 and EH114-GFP constructs and electroporated. GLUT-4 expression was consistently reduced in test muscles, by a mean of 51% (P = 0.004, n = 8), in the absence of any nonspecific effect upon housekeeping gene -actin (P = 0.23) (Fig. 6A). This effect was approximately additive over that of the two constructs alone. The IPGTT resulted in the expected effects on plasma glucose and tracer disappearance tissues over 90 min (Fig. 6B). Clearance of tracer to muscles transfected with the two GLUT-4 targeting constructs was reduced versus contralateral muscles by a mean 30% (P < 0.001) (Fig. 6C). Furthermore, there was a clear correlation between the ratios of GLUT-4 protein and glucose tracer clearance between paired TC muscles (r2 = 0.68, P = 0.012), implying a causal relationship between GLUT-4 suppression and attenuated glucose uptake. Thus, expression of a key muscle glucose disposal gene can be successfully knocked down in vivo using constructs encoding shRNAs, resulting in an altered physiological end point.
DISCUSSION
This study demonstrates the utility of IVE for both functional overexpression and suppression of glucose transporters in rodent skeletal muscle. Specifically, local direct manipulation of glucose transporter availability, in the absence of a perturbation of whole-body metabolism, resulted in substantial changes in basal and insulin-stimulated in vivo muscle glucose disposal. Furthermore, the work encompasses the first demonstration of the combination of IVE with gene silencing to produce functional changes in a target metabolic pathway in muscle in vivo.
IVE is emerging as a potential gene therapy modality in multiple tissues, despite its spatially and temporally limited effects (8), but has as yet been underutilized in basic research. Skeletal muscle is especially well suited to this procedure, being readily accessible and capable of regenerating any damage inflicted by a brief insult (28). Herein, we have demonstrated that transfection of 50% of fibers in a target muscle is readily achievable and that the resultant protein persists in the muscle for a minimum of 3 weeks, with peak expression around 7 days postelectroporation, consistent with previous reports (10, 29eC32). One to 2 days after electroporation, our results indicate that expression of electroporated cDNAs is relatively low, and it is undesirable to study the downstream effects of genetic manipulation before 7 days, as a degree of acute inflammation and vascular reaction is generated following electroporation (30, 33). However, as the electroporation procedure per se did not affect the expected attenuation of tracer uptake into muscle of high-fat-fed rats or alter the expression of indexes of muscle stress (Akt and HSP-70), we can conclude that any muscle damage is insufficient to confound downstream physiological measurements at this time point. Furthermore, although the data as presented here and our experience with other target genes confirms that transfection efficiency can be quite variable between animals, the use of a reporter gene and/or the restriction of downstream analysis to animals achieving a minimum level of gene overexpression or silencing should circumvent this as an issue.
Few reports have described acute silencing of genes in vivo by RNA interference (34eC37). Only one of these reports documented successful silencing in skeletal muscle and this only the very acute downregulation of a reporter gene (34). Herein, we have characterized two constructs encoding shRNAs targeting distinct GLUT-4 mRNA sequences, both in vitro and in vivo. This approach was adopted as it was likely that ongoing expression of shRNAs from a vector would result in a longer duration of silencing, as individual siRNA molecules are likely to be degraded within a short period of time (13, 37). In support of this, gene silencing by siRNAs was only recorded previously 1eC2 days after treatment (34, 36), while it was present after 10 days when vectors that expressed double-stranded siRNAs were used (35). Consistent with this, we found that GLUT-4 protein suppression by one shRNA constuct peaked 7 days postelectroporation, analogous to the results obtained for gene overexpression from similar vectors. Moreover, independent electrotransfer of a second shRNA construct targeting a distinct mRNA sequence achieved a similar effect upon GLUT-4 protein. When the two vectors were administered together, the mean protein suppression achieved was approximately additive, suggesting that a multiple targeting strategy might thus be generally more efficacious. However, these differences could also be accounted for by variations in electrotransfer efficiency.
Glucose disposal to skeletal muscle is largely responsible for clearance of a glucose load postprandially, and transport into myofibers is principally accomplished using the insulin-stimulated facilitative glucose transporter GLUT-4, while constitutive transport of glucose into muscle is thought to be mediated via GLUT-1 (38). Studies of transgenic mice overexpressing GLUT-1 in skeletal muscle have helped to establish this (2, 3). Consistent with these reports, we observed a close correlation between the degree of GLUT-1 overexpression in muscle of starved rats and the increase in glucose transport generated as a result.
There has been much debate over which component of the insulin-stimulated uptake and storage of glucose as glycogen in skeletal muscle is rate limiting. Of the three components of glucose disposal that have been considered, glucose transport, phosphorylation, and glycogen synthesis, glucose transport has been most frequently implicated. The close correlation between GLUT-4 suppression and the reduction in glucose tracer uptake observed in this study is consistent with this conclusion, although the relatively greater magnitude of the former change could imply additional levels of regulation are important, as has been previously suggested (5, 39). Some of the strongest evidence for glucose transport being rate limiting has come from transgenic overexpression or targeted disruption of GLUT-4 in mouse skeletal muscle. These genetic manipulations resulted in markedly increased (2, 4, 5) or decreased (6, 7) glucose disposal to muscle in mice. However, substantial whole-body alterations in glucose metabolism and/or insulin sensitivity were generated as part of each transgenic phenotype, often resulting in secondary effects on metabolism in other tissues. The IVE technique has the considerable advantage that individual muscles can be studied in an in vivo setting, but in the absence of whole-body physiological changes. Furthermore, acute genetic alteration can be carried out in mature animals, thus avoiding the potential confounding of interpretation caused by developmental adaptation to germ-line manipulation. Finally, IVE permits direct comparison of test muscles with their contralateral equivalents that have undergone electrotransfer of control DNA, permitting the use of powerful paired statistical methods to analyze results in each animal.
In summary, IVE is a technique that can be readily applied to acutely increase or decrease gene expression in single rodent skeletal muscles to study gene dosage effects in the absence of the confounding influences of developmental adaptation and alterations in whole-body physiology. Its compatibility with appropriate physiological measurements in muscle and with standard models of insulin resistance make it particularly suitable for the interrogation of candidate insulin sensitivity genes in this tissue. Using this technique, we have confirmed that under basal and insulin-stimulated conditions, the availability of glucose transporters is largely rate-limiting for glucose disposal into skeletal muscle.
ACKNOWLEDGMENTS
The authors are grateful to the National Health and Medical Research Council of Australia, the Diabetes Australia Research Trust, and the Rebecca L. Cooper Medical Research Foundation for their funding.
We thank Drs. Edna Hardeman and Mark Corbett, Children's Medical Research Institute, Westmead, NSW, Australia, for initial valuable advice and the EH114 vector. Also, we thank Dr. Lluis Mir, Institut Gustave-Roussy, Villejuif Cedex, France, for additional discussion regarding electroporation methodology; Dr. Sally Dunwoodie, Victor Chang Cardiac Research Institute, Darlinghurst, NSW, Australia, for loan of an Electroporator for initial studies; and to Jane Radford, University of Sydney, Australia, for sectioning of tissues. We are grateful to the staff of the Garvan Biological Testing Facility for animal care and to Andy Muirhead, Elaine Preston, and Nicolas Dzamko for technical assistance.
Key Words: GFP, green fluorescent protein HSP, heat shock protein IPGTT, intraperitoneal glucose tolerance test IVE, in vivo electrotransfer shRNA, short hairpin RNA siRNA, short interfering RNA TC, tibialis cranialis
REFERENCES
Ren JM, Barucci N, Marshall BA, Hansen P, Mueckler MM, Shulman GI: Transgenic mice overexpressing GLUT-1 protein in muscle exhibit increased muscle glycogenesis after exercise. Am J Physiol Endocrinol Metab278 :E588 eCE592,2000
Marshall BA, Mueckler MM: Differential effects of GLUT-1 or GLUT-4 overexpression on insulin responsiveness in transgenic mice. Am J Physiol267 :E738 eCE744,1994
Ren JM, Marshall BA, Gulve EA, Gao J, Johnson DW, Holloszy JO, Mueckler M: Evidence from transgenic mice that glucose transport is rate-limiting for glycogen deposition and glycolysis in skeletal muscle. J Biol Chem.268 :16113 eC16115,1993
Ren JM, Marshall BA, Mueckler MM, McCaleb M, Amatruda JM, Shulman GI: Overexpression of Glut4 protein in muscle increases basal and insulin-stimulated whole body glucose disposal in conscious mice. J Clin Invest95 :429 eC432,1995
Tsao TS, Burcelin R, Katz EB, Huang L, Charron MJ: Enhanced insulin action due to targeted GLUT4 overexpression exclusively in muscle. Diabetes45 :28 eC36,1996
Zisman A, Peroni OD, Abel ED, Michael MD, Mauvais-Jarvis F, Lowell BB, Wojtaszewski JF, Hirshman MF, Virkamaki A, Goodyear LJ, Kahn CR, Kahn BB: Targeted disruption of the glucose transporter 4 selectively in muscle causes insulin resistance and glucose intolerance. Nat Med6 :924 eC928,2000
Kotani K, Peroni OD, Minokoshi Y, Boss O, Kahn BB: GLUT-4 glucose transporter deficiency increases hepatic lipid production and peripheral lipid utilization. J Clin Invest114 :1666 eC1675,2004
Andre F, Mir LM: DNA electrotransfer: its principles and an updated review of its therapeutic applications. Gene Ther11 (Suppl. 1) :S33 eCS42,2004
Gehl J: Electroporation: theory and methods, perspectives for drug delivery, gene therapy and research. Acta Physiol Scand177 :437 eC447,2003
Mir LM, Bureau MF, Gehl J, Rangara R, Rouy D, Caillaud JM, Delaere P, Branellec D, Schwartz B, Scherman D: High-efficiency gene transfer into skeletal muscle mediated by electric pulses. Proc Natl Acad Sci U S A96 :4262 eC4267,1999
Fujii N, Boppart MD, Dufresne SD, Crowley PF, Jozsi AC, Sakamoto K, Yu H, Aschenbach WG, Kim S, Miyazaki H, Rui L, White MF, Hirshman MF, Goodyear LJ: Overexpression or ablation of JNK in skeletal muscle has no effect on glycogen synthase activity. Am J Physiol Cell Physiol287 :C200 eCC208,2004
Ho RC, Alcazar O, Fujii N, Hirshman MF, Goodyear LJ: p38 Gamma MAPK regulation of glucose transporter expression and glucose uptake in L6 myotubes and mouse skeletal muscle. Am J Physiol Regul Integr Comp Physiol286 :R342 eCR349,2003
Hannon GJ, Rossi JJ: Unlocking the potential of the human genome with RNA interference. Nature431 :371 eC378,2004
Brummelkamp TR, Bernards R, Agami R: A system for stable expression of short interfering RNAs in mammalian cells. Science296 :550 eC553,2002
Inukai K, Shewan AM, Pascoe WS, Katayama S, James DE, Oka Y: Carboxy terminus of glucose transporter 3 contains an apical membrane targeting domain. Mol Endocrinol18 :339 eC349,2004
Muscat GE, Perry S, Prentice H, Kedes L: The human skeletal alpha-actin gene is regulated by a muscle-specific enhancer that binds three nuclear factors. Gene Expr2 :111 eC126,1992
Rice RR, Muirhead AN, Harrison BT, Kassianos AJ, Sedlak PL, Maugeri NJ, Goss PJ, Davey JR, James DE, Graham MW: Simple, robust strategies for generating ddRNAi constructs. Methods Enzymol324 :405 eC419,2005
Morgenstern JP, Land H: Advanced mammalian gene transfer: high titre retroviral vectors with multiple drug selection markers and a complementary helper-free packaging cell line. Nucleic Acid Res18 :3587 eC3596,1990
Shewan AM, Marsh BJ, Melvin DR, Martin S, Gould GW, James DE: The cytosolic C-terminus of the glucose transporter GLUT4 contains an acidic cluster endosomal targeting motif distal to the dileucine signal. Biochem J350 :99 eC107,2000
Shewan AM, van Dam EM, Martin S, Luen TB, Hong W, Bryant NJ, James DE: GLUT4 recycles via a trans-golgi network (TGN) subdomain enriched in syntaxins 6 and 16 but not TGN38: involvement of an acidic targeting motif. Mol Biol Cell14 :973 eC986,2003
Clark PW, Jenkins AB, Kraegen EW: Pentobarbital reduces basal liver glucose output and its insulin suppression in rats. Am J Physiol258 :E701 eCE707,1990
Satkauskas S, Bureau MF, Puc M, Mahfoudi A, Scherman D, Miklavcic D, Mir LM: Mechanisms of in vivo DNA electrotransfer: respective contributions of cell electropermeabilization and DNA electrophoresis. Mol Ther5 :133 eC140,2002
Kraegen EW, James DE, Jenkins AB, Chisholm DJ: Dose-response curves for in vivo insulin sensitivity in individual tissues in rats. Am J Physiol248 :E353 eCE362,1985
Crosson SM, Khan A, Printen J, Pessin JE, Saltiel AR: PTG gene deletion causes impaired glycogen synthesis and developmental insulin resistance. J Clin Invest111 :1423 eC1432,2003
Cleasby ME, Dzamko N, Hegarty BD, Cooney GJ, Kraegen EW, Ye JM: Metformin prevents the development of acute lipid-induced insulin resistance in the rat through altered hepatic signaling mechanisms. Diabetes53 :3258 eC3266,2004
James DE, Strube M, Mueckler M: Molecular cloning and characterization of an insulin-regulatable glucose transporter. Nature338 :83 eC87,1989
Piper RC, Hess LJ, James DE: Differential sorting of two glucose transporters expressed in insulin-sensitive cells. Am J Physiol260 :C570 eCC580,1991
Charge SB, Rudnicki MA: Cellular and molecular regulation of muscle regeneration. Physiol Rev84 :209 eC238,2004
Dona M, Sandri M, Rossini K, Dell’Aica I, Podhorska-Okolow M, Carraro U: Functional in vivo gene transfer into the myofibers of adult skeletal muscle. Biochem Biophys Res Commun312 :1132 eC1138,2003
Molnar MJ, Gilbert R, Lu Y, Liu AB, Guo A, Larochelle N, Orlopp K, Lochmuller H, Petrof BJ, Nalbantoglu J, Karpati G: Factors influencing the efficacy, longevity, and safety of electroporation-assisted plasmid-based gene transfer into mouse muscles. Mol Ther10 :447 eC455,2004
Vilquin JT, Kennel PF, Paturneau-Jouas M, Chapdelaine P, Boissel N, Delaere P, Tremblay JP, Scherman D, Fiszman MY, Schwartz K: Electrotransfer of naked DNA in the skeletal muscles of animal models of muscular dystrophies. Gene Ther8 :1097 eC1107,2001
Maruyama H, Sugawa M, Moriguchi Y, Imazeki I, Ishikawa Y, Ataka K, Hasegawa S, Ito Y, Higuchi N, Kazama JJ, Gejyo F, Miyazaki JI: Continuous erythropoietin delivery by muscle-targeted gene transfer using in vivo electroporation. Human Gene Ther11 :429 eC437,2000
Gehl J, Skovsgaard T, Mir LM: Vascular reactions to in vivo electroporation: characterization and consequences for drug and gene delivery. Biochim Biophys Acta1569 :51 eC58,2002
Kishida T, Asada H, Gojo S, Ohashi S, Shin-Ya M, Yasutomi K, Terauchi R, Takahashi KA, Kubo T, Imanishi J, Mazda O: Sequence-specific gene silencing in murine muscle induced by electroporation-mediated transfer of short interfering RNA. J Gene Med6 :105 eC110,2004
Matsuda T, Cepko CL: Electroporation and RNA interference in the rodent retina in vivo and in vitro. Proc Natl Acad Sci U S A 2003
Soutschek J, Akinc A, Bramlage B, Charisse K, Constien R, Donoghue M, Elbashir S, Geick A, Hadwiger P, Harborth J, John M, Kesavan V, Lavine G, Pandey RK, Racie T, Rajeev KG, Rohl I, Toudjarska I, Wang G, Wuschko S, Bumcrot D, Koteliansky V, Limmer S, Manoharan M, Vornlocher HP: Therapeutic silencing of an endogenous gene by systemic administration of modified siRNAs. Nature432 :173 eC178,2004
Layzer JM, McCaffrey AP, Tanner AK, Huang Z, Kay MA, Sullenger BA: In vivo activity of nuclease-resistant siRNAs. RNA10 :766 eC771,2004
Olson AL, Pessin JE: Structure, function, and regulation of the mammalian facilitative glucose transporter gene family. Annu Rev Nutr16 :235 eC256,1996
Fisher JS, Nolte LA, Kawanaka K, Han DH, Jones TE, Holloszy JO: Glucose transport rate and glycogen synthase activity both limit skeletal muscle glycogen accumulation. Am J Physiol Endocrinol Metab282 :E1214 eCE1221,2002(Mark E. Cleasby, Jonathan)