Email: firstname.lastname@example.org. Present address: School of Veterinary Medicine and Science, The University of Nottingham, Sutton Bonington Campus, Leicestershire, UK. C.M. Mcgowan's present address is: Department of Clinical Veterinary Science, University of Liverpool, Cheshire, UK.
Reasons for performing study: Several conditions associated with laminitis in horses are also associated with insulin resistance, which represents the failure of glucose uptake via the insulin-responsive glucose transport proteins in certain tissues. Glucose starvation is a possible mechanism of laminitis, but glucose uptake mechanisms in the hoof are not well understood.
Objectives: To determine whether glucose uptake in equine lamellae is dependent on insulin, to characterise the glucose transport mechanism in lamellae from healthy horses and ponies, and to compare this with ponies with laminitis.
Methods:Study 1 investigated the effects of insulin (300 µU/ml; acute and 24 h) and various concentrations of glucose up to 24 mmol/l, on 2-deoxy-D-[2,6-3H]glucose uptake in hoof lamellar explants in vitro. Study 2 measured the mRNA expression of GLUT1 and GLUT4 transport proteins by PCR analysis in coronary band and lamellar tissue from healthy horses and ponies, ponies with insulin-induced laminitis, and ponies suffering from chronic laminitis as a result of equine Cushing's syndrome.
Results: Glucose uptake was not affected by insulin. Furthermore, the relationship between glucose concentration and glucose uptake was consistent with an insulin-independent glucose transport system. GLUT1 mRNA expression was strong in brain, coronary band and lamellar tissue, but was weak in skeletal muscle. Expression of GLUT4 mRNA was strong in skeletal muscle, but was either absent or barely detectable in coronary band and lamellar tissue.
Conclusions: The results do not support a glucose deprivation model for laminitis, in which glucose uptake in the hoof is impaired by reduced insulin sensitivity. Hoof lamellae rely on a GLUT1-mediated glucose transport system, and it is unlikely that GLUT4 proteins play a substantial role in this tissue.
Potential relevance: Laminitis associated with insulin resistance is unlikely to be due to impaired glucose uptake and subsequent glucose deprivation in lamellae.
Several conditions associated with laminitis in horses, such as equine Cushing's syndrome (pituitary pars intermedia dysfunction, PPID) (Johnson et al. 2004), equine metabolic syndrome (Johnson 2002; Treiber et al. 2005) and iatrogenic corticosteroid administration (Johnson et al. 2002) are also associated with altered glucose metabolism and the development of insulin resistance. The pathogenesis of laminitis associated with insulin resistance is not fully understood, but one hypothesis is that decreased uptake of glucose by the lamellae could result in laminitis (Pass et al. 1998; French et al. 2000). Glucose deprivation of hoof lamellar explants results in lamellar separation (Pass et al. 1998), and glucose consumption at rest exceeds that of the head (Wattle and Pollitt 2004). Thus, even a small decrease in the rate of glucose uptake within the lamellar tissue could be extremely damaging to the lamellar integrity, resulting in laminitis.
Glucose transport across the cell membrane is mediated by glucose transport proteins. The GLUT1 protein is responsible for basal glucose uptake in tissues such as the brain, erythrocytes, endothelial cells and skin keratinocytes, and can be upregulated by growth factors (e.g. IGF-1) and hypoxia in rats and man (Behrooz and Ismail-Beigi 1997). This protein is localised mainly in the cell membrane in the basal state, and does not require stimulation by insulin for upregulation and activation (Behrooz and Ismail-Beigi 1997). In contrast, the GLUT4 protein is largely localised within the intracellular organelles in the basal state, and is translocated to the cell membrane following stimulation with insulin in tissues such as the heart, skeletal muscle and adipose tissue in rats and man (Gould and Holman 1993). The equine GLUT4 gene has been characterised in equine skeletal muscle, and shares a high degree of homology with that of other mammals (Nout et al. 2003; van Dam et al. 2004; Jose-Cunilleras et al. 2005).
Insulin prevents blood glucose concentrations from rising excessively after a meal by stimulating glucose uptake and glycogen accumulation in peripheral tissues, particularly skeletal muscle and adipose tissue (Pessin and Saltiel 2000). Insulin resistance occurs when either the translocation or functional capabilities of the GLUT4 protein are impaired (Garvey et al. 1998; Kim et al. 2001). Under these conditions, more insulin is required than normal to activate the glucose transporters and maintain euglycaemia.
Previous research has attempted to characterise the expression of GLUTs within equine lamellar tissue, with particular attention to the GLUT1 and GLUT4 transporters. Positive immunostaining for GLUT1 has been observed consistently within lamellar keratinocytes (Mobasheri et al. 2004; Wattle and Pollitt 2004) and has been reported to be reduced considerably in laminitic tissue (Mobasheri et al. 2004), indicating that the lamellae are highly metabolic tissues, and possibly suffer from impaired glucose metabolism during the laminitis disease process. In contrast, immunostaining for GLUT4 has produced inconclusive results, with one study reporting positive staining in normal lamellae, which was reduced in laminitic tissue (Mobasheri et al. 2004). Another study reported considerable cross reactivity and poor specificity for the GLUT4 antibody (Wattle and Pollitt 2004).
The hypothesis of the present study was that the hoof lamellae utilise a GLUT1-mediated glucose transport system, independent of insulin, and therefore impaired glucose uptake by lamellae due to insulin resistance is not a cause of laminitis. The aims of this study were to determine whether glucose uptake in equine hoof lamellar explants is dependent on insulin. Evidence for functional GLUT1 and GLUT4 was also sought by identifying the saturation point of glucose uptake in terms of glucose concentration. Finally, the presence of GLUT1 and GLUT4 transport proteins was sought via PCR analysis of mRNA expression in hoof tissue from normal, healthy horses and ponies, ponies with insulin-induced laminitis (Asplin et al. 2007), and ponies suffering from chronic laminitis due to PPID.
Materials and methods
Study 1: Glucose uptake response to insulin and varying concentrations of glucose
Tissue collection: Hooves were obtained from 9 normal, healthy horses killed at a commercial abattoir. The distal aspect of the left forelimb was disarticulated at the metacarpal-phalangeal joint within 10 min of death. Hooves were transported to the laboratory within 30 min, in 0.9% saline, warmed to 37°C. The hooves were sectioned with a bandsaw as described by Pollitt (1996), with minor adjustments. Briefly, a section at the midpoint between the coronary band and ground-bearing surface on the dorsal aspect of the hoof was obtained using a bandsaw and sharp dissection. A customised tool was used to obtain blocks of lamellar tissues (approximately 5 × 5 × 10 mm). Hoof lamellar explants (approximately 5 × 5 × 1 mm; 20 mg) extending from the inner hoof wall through the lamellar junction to the dermal connective tissue were obtained by sharp dissection.
Glucose uptake response to stimulation with insulin: Glucose uptake was estimated by measuring the uptake of nonmetabolisable 2-deoxy-D-[2,6-3H] glucose1. Hoof lamellar explants (15–20 mg) were preincubated in modified Kreb's-Henseleit Buffer (mKHB; 118.5 mmol/l NaCl, 4.7 mmol/l KCl, 2.5 mmol/l CaCl2, 1.2 mmol/l KH2PO4, 2.4 mmol/l MgSO4, 25 mmol/l NaHCO3, 35 mmol/l mannitol, 1 mmol/l 2-deoxy-D-glucose, 5.55 mmol/l d-glucose and 0.08% bovine serum albumin) at 37°C for 30 min under 5% CO2. Medium was removed, and explants were washed 3 times in 0.9% saline at 37°C. Explants were incubated in 5 ml mKHB containing 2-deoxy-D-[2,6-3H] glucose (0.25 µCi/ml), with 0 or 300 µU/ml porcine insulin2 for 10, 20, 40, 60, 90 or 120 min, or for 24 h, at 37°C under 5% CO2. The porcine insulin sequence differs from the equine sequence by only one amino acid residue (Harris et al. 1956), and was used due to the unavailability of equine insulin. Preliminary experiments were conducted to determine the effects of varying concentrations of insulin to determine the optimal insulin concentration for incubation (data not shown). Explants were then washed in ice-cold 0.9% saline to stop transport and remove unbound label, blotted, weighed and digested in 3 ml 0.3 mol/l NaOH for 48 h at 50°C in a shaking waterbath. Radioactivity was determined using a Wallac 1400 liquid scintillation counter3.
Basal glucose uptake was taken as 2-deoxyglucose uptake in the absence of insulin. In each experiment, 4 replicates were used to determine each point. Data are expressed as nmol 2-deoxyglucose per mg wet weight of explant per min incubation time (nmol/mg/min). The small amount of extracellular 2-deoxyglucose present in the intracellular space was measured and corrected for using [14C]-inulin1 (0.3 µCi/ml). This determines whether any cell death or loss of membrane integrity has occurred, which would result in a false estimate of glucose uptake.
Glucose uptake with increasing glucose concentration: Hoof explants were pre-incubated for 30 min in Dulbecco's modified Eagles medium (DMEM) (9 mmol/l D-glucose) supplemented with gentamycin (0.1 mg/ml) at 37°C under 5% CO2 for 30 min. The medium was removed and the explants washed 3 times in 0.9% saline at 37°C. Explants were incubated in DMEM supplemented with gentamycin (0.1 mg/ml), containing varying concentrations of D-glucose (0–24 mmol/l) spiked with 2-deoxy-D-[2,6-3H] glucose (0.25 µCi/ml), for 24 h at 37°C under 5% CO2. The explants were digested and measured for radioactivity as described above.
Study 2: GLUT1 and GLUT4 mRNA expression in lamellae, following induction of laminitis with prolonged insulin infusion
Tissue collection: Normal hoof tissue was obtained fresh from 4 horses killed at a commercial abattoir, and from frozen samples obtained from 4 control ponies who participated in the study described by Asplin et al. (2007). Frozen lamellar samples obtained from 5 ponies in which laminitis was induced with prolonged insulin infusion (Asplin et al. 2007) were also included in this study. Finally, tissue was also obtained from 2 aged ponies with a history of chronic laminitis and clinical signs of PPID. These ponies were subjected to euthanasia with pentobarbitone sodium (162.5 mg/kg bwt i.v.) and hoof lamellar tissue harvested. PPID was confirmed via histopathological assessment of the pituitary gland. Samples of brain (cerebral cortex) and skeletal muscle (gluteus medius) were collected from a horse subjected to euthanasia for clinical reasons by a registered veterinarian to provide positive control tissues expected to contain GLUT1 and 4 transporters, respectively.
Lamellar necropsy samples were obtained from the abattoir within 10 min of death, by drilling4 through the outer hoof wall and dissecting out the lamellar tissue. All other hooves were sectioned as described above. Hoof lamellar tissue (approximately 5 mm3) was sectioned as described above and transported on dry ice to the laboratory for storage at -80°C before processing. Coronary band tissue (approximately 5 mm3) was obtained at the point where the hoof tissue joined with hair on the dorsal aspect of the foot.
Induction of laminitis with prolonged insulin infusion: Laminitis was induced in 5 clinically normal, healthy ponies via a prolonged euglycaemic-hyperinsulinaemic clamp technique as described by Asplin et al. (2007). Briefly, 9 ponies were allocated to either a treated (n = 5) or control (n = 4) group, to receive infusions of recombinant human insulin (Humulin R)5, plus glucose, or an equivalent volume of isotonic saline, respectively. Treated ponies received a priming dose of insulin (45 mU/kg bwt in 50 ml of 0.9% saline) i.v. as a bolus injection (DeFronzo et al. 1979), followed by infusion at a steady rate of 6 mU/min/kg bwt. An infusion of glucose solution (50% w/v; Baxter) was adjusted constantly to maintain euglycaemia. The ponies were subjected to euthanasia with pentobarbitone sodium (162.5 mg/kg bwt i.v.) at the onset of Obel grade 2 laminitis (Obel 1948), or after 72 h of saline infusion. Laminitis was confirmed in all treated ponies by histopathological examination, whereas tissue obtained from all control ponies was judged to be normal.
GLUT1 and GLUT4 mRNA expression in hoof lamellae: Frozen lamellar, coronary band, skeletal muscle and brain tissue (approximately 50 mg) were homogenised (OmniTip Homogenizing kit)6 in 1 ml TRIzol reagent. Isolation of RNA was performed using the TRIzol7 method and reconstituted in 20 µl RNase-free water. RNA was separated from the protein by adding 0.2 ml chloroform to each tube, shaking for 15 s, incubating for 3 min on ice and centrifuging at 12,000 g for 15 min at 4°C. The upper aqueous phase containing RNA was collected, 0.5 ml isopropanol added and the tubes placed on ice for 10 min, then centrifuged at 12,000 g for 10 min at 4°C. The RNA precipitate was washed with 75% ethanol (-20°C), vortexed and centrifuged at 7500 g for 5 min at 4°C. The supernatant was carefully decanted, and the RNA pellet dried at 65°C. Total RNA was reconstituted in 20 µl RNase-free water and quantified using a UV-visible spectrophotometer; its integrity was confirmed visually using gel electrophoresis. RNA was stored at -80°C until used.
Two-step RT-PCR reactions were carried out using a commercial kit (SuperScript First-Strand Synthesis System for RT-PCR)7. An initial denaturation step was performed by incubating 2 µg total RNA, 1 µl random primer (50 ng), 1 µl 10 mmol/l deoxynulceotide mix plus sterile water to 20 µl, at 65°C for 5 min, then placing the tubes on ice for 1 min, prior to the addition of 2 µl 10x reaction buffer, 4 µl 25 mmol/l MgCl2, 2 µl 0.1 mol/l DTT and 1 µl RNA inhibitor. This was followed by annealing at 25°C for 2 min, and the addition of 1 µl SuperScript II reverse transcriptase (50 units). First strand synthesis was carried out at 25°C for 10 min, 42°C for 50 min, followed by denaturation at 70°C for 15 min and then cooling to 4°C for 5 min. RNA was removed with 1 µl of RNase H, followed by incubation at 37°C for 20 min. The reverse transcriptase was omitted in control reactions. mRNA was stored at -20°C until processed.
Equine specific oligonucleotide primers7 based on the published sequences for GLUT1 and 4 glucose transport proteins, and β-actin (GenBank Accession Nos. DQ139875, AF531753 and NM_001081838, respectively) were designed to amplify suitably sized fragments (Table 1). Polymerase chain reaction (PCR) was carried out using a BioTaq DNA polymerase (BioLine). The addition of 10% DMSO was required to disrupt GC disulphide bonds prior to amplification of GLUT4 mRNA. PCR cycles were performed in a GeneAmp PCR System 2700 thermal cycler8. This consisted of an initial denaturation step at 94°C for 2 min, followed by amplification of target cDNA carried out for 35 cycles using a 3-step cycling protocol of denaturing at 92°C (GLUT1) or 94°C (GLUT4) for 30 s, annealing at 62°C (GLUT1) or 70°C (GLUT4) for 30 s and an extension phase at 72°C for 2 min. A final extension phase was performed at 72°C for 5 min after the last cycle.
Table 1. GLUT1 and GLUT4 primer sequences used for PCR in equine skeletal muscle, lamellae and coronary band tissue
5′-CAC TGG AGT CAT CAA CGC CC- 3′
5′-CCA CGA TCA GCA TCT CAA AG-3′
5′-TGG GCT CTC TCC GTG GCC ATC TT-3′
5′-GCT GCT GGC TGA GCT GCA GCA-3′
PCR products were resolved in 2% agarose gels using a horizontal submarine gel electrophoresis system9 and visualised with ethidium bromide stain. Parallel amplifications of cDNA for the housekeeping gene, β-actin, were used as an internal control.
Presentation of the results and statistical analysis
Results are expressed as means ± s.e. In Study 1, the distribution of observations for total glucose uptake was not normal, so natural logarithmic transformation was performed to achieve normality. Multivariable linear regression analyses were conducted to determine whether total glucose uptake was influenced by stimulation with insulin or increasing concentrations of glucose, and whether a time by treatment and/or horse by treatment interaction was significantly related to total glucose uptake. Significance was accepted as P<0.05. Statistical analyses were performed using the Stata v.10 for Windows software10.
Study 1: Glucose uptake response to incubation with insulin and varying concentrations of glucose
Glucose uptake response to stimulation with insulin: Total glucose uptake in hoof lamellar explants was investigated over time in the presence or absence of insulin (10–120 min Fig 1a, and at 24 h Fig 1b). Total glucose uptake increased significantly from 10–90 min of incubation time (P<0.05). However, insulin did not cause a significant change in glucose uptake when compared to basal values at any time point.
Glucose uptake with increasing glucose concentration: There was a significant effect of glucose concentration on total glucose uptake (P<0.001) (Fig 2). In particular, total glucose uptake increased from approximately 7 nmol/mg tissue at 3 mmol/l glucose, to a peak of 31 nmol/mg tissue at 15 mmol/l glucose, before plateauing and beginning to decrease at approximately 21 mmol/l.
Study 2: GLUT1 and GLUT4 mRNA expression in hoof lamellae
The expression of GLUT1 and GLUT4 mRNA in equine skeletal muscle, hoof lamellae and coronary band tissues in healthy ponies (n = 4; 7.0 ± 2.8 years) and ponies with insulin-induced laminitis (n = 5; 5.9 ± 1.7 years) is presented in Figure 3.
The expression of GLUT1 and GLUT4 mRNA in equine skeletal muscle, hoof lamellae and coronary band tissues in healthy horses (n = 4) and ponies with PPID and a history of chronic laminitis (n = 2) is presented in Figure 4.
Expression of GLUT1 mRNA was weak in skeletal muscle both from ponies with insulin-induced laminitis and PPID, and was not different from controls. Further, GLUT1 mRNA expression was strong both in lamellae and coronary band tissue in both groups of laminitic ponies, and was not different from controls.
Expression of GLUT4 mRNA was very strong in skeletal muscle from both groups of laminitic ponies, and was not different to controls. Very weak GLUT4 mRNA expression was evident in lamellae from only one out of 6 control animals compared with 2 out of 5 ponies with insulin-induced laminitis and 2 out of 2 ponies with PPID and chronic laminitis. Similarly, very weak GLUT4 mRNA expression was evident in coronary band tissue from 6 out of 7 control animals compared with 2 out of 5 ponies with insulin-induced laminitis and 2 out of 2 ponies with PPID and chronic laminitis.
The results of this study are consistent with glucose uptake in healthy lamellar tissue using a GLUT1-mediated glucose transport system, which is not responsive to insulin. In Study 1, neither short- nor long-term stimulation with insulin had any effect on glucose uptake. Other tissues in the body that are insulin-independent are typically those with a high requirement for glucose such as the brain (Vannucci et al. 1998; Lee et al. 2000), and in this case it is worth noting the findings of Wattle and Pollitt (2004), which highlighted the very high basal glucose requirement of the hoof. These results are consistent with studies in a variety of epithelial cell types, such as skin, human eye retina, corneal cells and absorptive cells in the small intestine (Kumagai et al. 1994; Voldstedlund and Dabelsteen 1997; Shen et al. 2000).
Glucose is capable of downregulating its own transport and metabolism via GLUT1 transporters in various tissues, including mouse skin, human and rat skeletal muscle, and rat aorta and lung, such that abnormally high glucose levels result in reduced glucose uptake (Simmons et al. 1993; Ciaraldi et al. 1995; Howard 1996; Spravchikov et al. 2001). Thus, in a tissue that is not reliant on insulin to regulate glucose transport, such as hoof lamellae, the ability of glucose per se to regulate its own metabolism is an important factor when considering the effects of glucose metabolism on its structural integrity. The results of this study show clearly that as hoof lamellae are exposed to increasing concentrations of glucose, glucose transport into the tissue begins to increase from about 7 nmol/mg tissue at 3 mmol/l to a peak of 31 nmol/mg tissue at 15 mmol/l, before plateauing and beginning to decrease at approximately 21 mmol/l. These results are consistent with those obtained in proliferating or differentiating skin keratinocytes, which showed decreased glucose uptake at 20 mmol/l glucose in primary murine keratinocytes (Spravchikov et al. 2001). This decrease in glucose uptake probably reflects downregulation of GLUT1 proteins as the result of a glucotoxic state (Spravchikov et al. 2001), but could also be the result of a necessary mechanism to prevent glucotoxicity.
Pass et al. (1998) observed that lamellar explants subjected to inadequate glucose supply in vitro underwent separation similar to that observed in tissue obtained from laminitic horses post mortem, and hypothesised that reduced glucose uptake resulting from insulin resistance could result in laminitis. However, the results of the present study indicate that the hoof lamellae are not insulin responsive, making this an unlikely scenario. Instead, the results obtained by Pass et al. (1998) may simply be a reflection of the removal of the available energy source, resulting in starvation and death of the tissue.
The strong expression of GLUT1 mRNA in lamellar and coronary band tissue indicates that an insulin-independent glucose transport system probably predominates in both regions of the hoof. Expression of GLUT4 mRNA was either not detectable or barely detectable in lamellae and coronary band samples in the present study. The results obtained here suggest that it is unlikely that the GLUT4 gene is expressed in abundance and is unlikely therefore to play a substantial role in glucose uptake in the hoof lamellae.
The present results support the hypothesis that equine hoof lamellae and the proliferative coronary band utilise a predominantly insulin-independent, GLUT1-mediated glucose transport mechanism. The implications of the present study are that it is unlikely that lamellar damage seen during laminitis in insulin resistant animals is the result of impaired GLUT4 translocation or function. Further research should determine the relative proportions of GLUT1 and GLUT4 proteins, as well as the effects of other glucoregulatory and stimulatory hormones on glucose transport in equine hoof lamellae.
This work was funded in part by a grant and PhD scholarship to K. Asplin from the Rural Industries Research and Development Corporation, Australia. The assistance of John Morton for statistical advice, and Kellie Munn, Johanna Barclay and Christopher Owens for technical support, is gratefully acknowledged.
1 Amersham Pharmacia Biotech, Baulkham Hills, New South Wales, Australia.
2 Sigma-Aldrich, Castle Hill, New South Wales, Australia.
3 Wallac, Turku, Finland.
4 Dremel, Racine, Wisconsin, USA.
5 Ely Lily, Indianapolis, Indiana, USA.
6 Omni International Inc., Marietta, Georgia, USA.
7 Invitrogen, Carlsbad, California, USA.
8 Applied BioSciences, Foster City, California, USA.
9 Bio-Rad, Hemel Hempstead, Hertfordshire, UK.
10 StataCorp, College Station, Texas, USA.
Author contributions All authors contributed to the initiation, planning and writing of this study. Its execution and statistics were conducted by K.E.A.