Endocrinological aspects of the pathophysiology of equine laminitis


Certain individual horses appear to be predisposed to recurrent episodes of laminitis, but the exact mechanism underlying their predisposition remains a fundamental question in laminitis research. It seems likely that there are certain phenotypic traits common to these individuals. Multiple variables have been evaluated as risk factors for laminitis with the findings being inconsistent between studies. Associations between the occurrence of laminitis and being a pony, the spring and summer months [1], being female, increasing age [2] and obesity [3, 4] have been demonstrated. The most likely endocrinological disorders that may play a role in this predisposition are those associated with altered glucose metabolism and the development of insulin resistance (IR). The risk factors of being a pony [5] and obesity [6] can be associated with IR, which are a feature of equine metabolic syndrome, a disease characterised by recurrent laminitis as well as regional or generalised adiposity [7]. In addition, laminitis is seen in association with pituitary pars intermedia dysfunction (PPID) [7] and iatrogenic corticosteroid administration [8] in which the laminitis may be triggered by antagonism of insulin by cortisol and hence the development of IR. The pathogenesis of laminitis associated with IR is not fully understood, but the initial hypothesis was that decreased uptake of glucose by the lamellae could result in laminitis.

The normal equine hoof consumes glucose, and glucose consumption at rest exceeds that of the head [9]. Initial research using hoof explants found that decreased glucose availability resulted in separation of the hoof lamellae between the basal epidermal cells and their basement membrane in vitro characteristic of the hoof separation that occurs in naturally occurring laminitis [10]. The authors of that study suggested that the integrity of the hoof explants was dependent on consumption of glucose and that this provided a possible explanation for the development of laminitis caused by conditions such as carbohydrate overload, acute inflammatory conditions, corticosteroid therapy and hyperlipaemia as these conditions would induce a major insulin-mediated metabolic shift away from glucose consumption by many peripheral tissues and if the metabolic change occurred faster than the hoof tissue could adapt to an alternative energy substrate, then hoof separation and laminitis would occur [10]. A second study revealed that the separation occurred due to the lack of glucose causing a reduction in hemidesmosome numbers until they disappeared and the basal cytoskeleton collapsed [11]. 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 (GLUTs). Equine research has focused on lamellar expression of the insulin-independent GLUT-1 and the insulin-dependent GLUT-4 transporters. Glucose transporter-1 has been consistently demonstrated within lamellar keratinocytes [9, 12] and was found to be reduced in laminitic tissue [12], indicating that the lamellae are highly metabolic tissues, and possibly suffer from impaired glucose metabolism during the laminitis disease process. In one study GLUT-4 was shown to be present in normal lamellae and reduced in laminitic tissue [12], while another study reported poor specificity for the GLUT-4 antibody [9]. However, a recent EVJ article (included in the EVJ laminitis virtual issue) has shown that glucose uptake by hoof explants was not affected by insulin and that the relationship between glucose concentration and glucose uptake was consistent with an insulin-independent glucose transport system [13]. In addition, while there was strong expression of the insulin-independent glucose transporter GLUT-1 mRNA in lamellar tissue, expression of the insulin-dependent glucose transporter GLUT-4 mRNA was absent or barely detectable [13]. Thus, these results do not support a glucose deprivation model for laminitis, in which glucose uptake in the hoof is impaired by reduced insulin sensitivity (or increased IR).

Equine IR is characterised by hyperinsulinaemia and normoglycaemia in the majority of animals. Further investigation into the role of insulin in the pathogenesis of laminitis resulted in the effects of hyperinsulinaemia on the lamellae being evaluated in a number of studies (including 2 of the publications in this virtual issue). Prolonged administration of insulin to induce hyperinsulinaemia while maintaining normal glucose concentrations resulted in the development of clinical and histological laminitis in all 4 feet of healthy ponies within 72 h [14] and of healthy Standardbred horses within 48 h [15] suggesting that IR and the associated hyperinsulinaemia place horses and ponies at risk of developing laminitis through direct insulin toxicity. However, it must be acknowledged that the serum insulin concentrations achieved in these studies were much higher than those commonly associated with PPID [16] and those reported in some cases of pasture-associated, endocrinopathic laminitis [17] and similarly high serum insulin concentrations have been documented in recurrently laminitic ponies in the absence of clinical signs of laminitis [18] such that very high serum insulin concentrations alone are not necessarily laminitis-inducing. In addition, subsequent histological examination of the laminitis lesions from the ponies with experimental insulin-induced laminitis revealed a lack of widespread basement membrane disintegration and instead the predominant lesion was one of apoptosis and mitosis in the axial regions [19]. Thus, the histological changes induced by insulin are not the same as those seen in the various carbohydrate-overload models of laminitis or in the naturally occurring disease.

Despite this histological discrepancy, if it is not through glucose deprivation, how then might insulin cause laminitis? Possibilities include effects on blood flow, inflammatory effects, glucose excess and MMP activation. Infusion of insulin into healthy Standardbreds resulted in the hoof wall surface temperature being higher and less variable once hyperinsulinaemia was established compared with control horses suggestive of increased digital perfusion [15]. A second study showed that vasodilation without hyperinsulinaemia was not sufficient to induce laminitis [20]. In contrast in other species, insulin activates endothelial vasodilatory and vasoconstrictive pathways; insulin stimulates nitric oxide (NO) synthase via the phosphatidylinositol 3-kinase (PI-3 K) pathway and endothelin-1 (ET-1) release through the mitogen-activated protein kinase (MAPK) pathway [21]; however, in IR, PI-3 K activation is blocked and the resultant NO:ET-1 imbalance favours vasoconstriction [21]. This IR-associated vasoconstriction is supported by the finding that following induction of IR, the normal relaxation responses of isolated rings of equine palmar digital arteries and veins were converted to contractions [22]. Thus, whether equine IR is associated with increased or decreased digital perfusion remains unclear.

The inflammatory events associated with the onset of both carbohydrate-overload and black walnut experimental models are increasingly well documented [23, 24]. However, the role of cytokines in the development of endocrinopathic laminitis remains less clear. While laminar leucocyte immigration also occurs in the endocrinopathic model of laminitis, albeit less intensely in comparison with carbohydrate-overload models [19, 25], in one study there was no evidence of increased cytokine-mediated inflammation in obese IR horses [26] while in another the results suggested that an inter-relationship exists among obesity, inflammatory cytokines, and insulin sensitivity in the horse [27]. Thus once again, the association between IR and inflammation remains unclear.

In man, GLUT-1 is more sensitive to increased blood flow that leads to increased flow-mediated glucose uptake. Since the GLUT-1 receptor predominates in the lamellar vasculature then it was postulated that increased digital blood flow may result in glucose excess. This in turn results in the increased production of advanced glycation endproducts and reactive oxygen species that cause vascular dysfunction and possibly laminitis. However, in a recent study involving lamellar tissue from horses with experimental insulin-induced laminitis, there was no increase in reactive oxygen species production or advanced glycation end products receptor gene expression and accumulation of advanced glycation end products only appeared after 48 h (i.e. at the onset of clinical signs) rather than being an initiating event [28]. Thus, it would appear that glucose excess does not play a role in insulin-induced laminitis.

Activation of matrix metalloproteinases (MMPs)-2 and -9 was previously thought to be an initiating event in laminitis. However, more recent studies have shown this not to be the case in the oligofructose model of laminitis and other proteases such as ADAMTS-4 and regulators such as TIMP-2 may be involved in early lamellar damage instead [29]. In the hyperinsulinaemic model of laminitis, gene expression of the members of the MMP family MMP-2, MT1-MMP and ADAMTS-4 and the regulator TIMP-3 were not also increased during the developmental stage [30] suggesting that MMP upregulation does not play a role in insulin-induced laminitis.

Regardless of the exact role insulin may or may not play in the pathogenesis of laminitis, as far as the equine clinician is concerned, the ability to identify animals at increased risk of laminitis would be priceless, as it would enable the appropriate management countermeasures to be implemented prior to the onset of the disease. Previously it has been shown that meeting 3 or more of the following criteria: out of body condition score >6/9, plasma triglyceride concentration >570 mg/l, proxy measurements of insulin sensitivity reciprocal of the square root of insulin (RISQI) <0.32 [mu/l]-0.5 and insulin secretory response modified insulin-to-glucose ratio (MIRG) >5.6 muinsulin2/[10·l·mgglucose]) differentiated a group of previously laminitic ponies from a group of normal ponies with a total predictive power of 78% [31]. There have been several studies more recently attempting to devise a test or panel of tests capable of identifying individual animals at risk of pasture-associated laminitis rather than groups of animals and of identifying the imminent onset of an episode of disease (2 of which are included in this virtual issue).

Carter et al. [32] sought to identify variables and clinically useful cut-off values with reproducible diagnostic accuracy for the prediction of the future occurrence of laminitis following exposure to nutrient-dense pasture within a cohort of ponies from a single inbred herd. Variables with diagnostic accuracy for the prediction of an episode of clinical laminitis in individual ponies within the next 3 months included serum insulin concentration >32 mu/l, plasma leptin concentration >73 ng/ml, body condition score ≥7/9, cresty neck score ≥4 and neck circumference:height ratio >0.71 [32]. Combining tests did not result in higher diagnostic accuracy than individual tests of insulin or leptin.

Borer et al. [33] sought to determine whether proxy measurements of insulin sensitivity (RISQI and quantitative insulin sensitivity check index: QUICKI) and insulin secretory response (MIRG and modified insulin-to-glucose ratio for ponies: MIGRP) based on basal glucose and insulin concentrations could distinguish between outbred normal and previously laminitic ponies kept under identical management regimes. While RISQI, QUICKI and MIGRP, but not MIRG, differentiated between normal and previously laminitic as a group, none accurately identified individual laminitis-prone animals [33], thus limiting their usefulness as a clinical test. In a second study Borer et al. [34] evaluated the glycaemic and insulinaemic responses of normal and previously laminitic ponies to glucose, fructose and inulin in spring and autumn after 7 days adaptation to a pasture or hay diet. Measurement of peak serum insulin concentrations 2 h after feeding of a single dose of glucose (1 g/kg bwt) discriminated between individual normal and laminitis-prone animals, particularly when ponies are adapted to eating autumn pasture [34]. All normal ponies had peak insulin concentrations <350 miu/ml whereas all previously laminitic ponies had peak insulin concentrations >500 miu/ml in autumn when adapted to eating pasture. Thus this may be a clinically useful test of laminitis predisposition when an animal's history is unknown.

Thus, thanks to the phenomenal efforts of a large number of dedicated researchers, knowledge relating to the endocrinological aspects of the pathophysiology of laminitis has increased dramatically over the last few years. However, it has probably generated as many questions as answers and there is still a lot more that needs to be done before we have achieved our goal of fully understanding and therefore conquering this most frustrating of conditions. In addition it should be remembered that not all laminitis-prone animals are obese and/or insulin resistant, not all obese animals are insulin resistant and not all insulin resistant animals are laminitis-prone.