Articles & Information
FROM THE HOOFCARE & LAMENESS ARCHIVES
Microanatomy of the Hoof Wall
Note: this article was originally published in Hoofcare & Lameness in 1994.
In order to more fully understand the pathophysiology of equine laminitis we have been studying the normal structure and function of the inner hoof wall. Using microvascular corrosion casts and the scanning electron microscope the lamellar microcirculation was shown to consist of the expected circulatory elements of arteries, capillaries and veins with the addition of a surprisingly large number of direct shunts between arteries and veins (arteriovenous anastomoses or AVAs). The AVAS were assumed to exist in the equine foot for the same reason as they exist in the extremities of other cold adapted animals; to periodically warm the foot when a critically low temperature is reached. It was concluded that a double circulation, under central hypothalamic control, existed. One slow, nutritive, capillary circulation to maintain the metabolism of the lamellar epidermis and a second, fast, AVA circulation to rapidly bring warm arterial blood close to the surface of the hoof (Pollitt and Molyneux 1991) Thus, the temperature of horses hooves, varying as much as 20 degrees C, must still be considered normal. We have confirmed this notion by continuously monitoring the hoof temperature of horses kept for several days in a cool climate laboratory.
Serial section studies with the light microscope, show that the AVAs are associated with neurovascular bundles and are clustered on either side of each primary epidermal lamella. The walls of AVAs contain both longitudinal and circular smooth muscle cells, are densely innervated with peptidergic as well as adrenergic nerves (more so than the adjoining arteries) and have characteristically tall endothelial cells projecting into the vessel lumen. Studies of AVAs with the transmission electron microscope (TEM) shows that the endothelial cell cytoplasm contains a high organelle concentration thus suggesting that these cells are metabolically quite active (Molyneux, Mogg, Haller and Pollitt, 1994). Just what these activities are has yet to be fully investigated.
More recently we have focussed our attention on the structure of the connective tissue which suspends the distal phalanx to epidermal cells of the inner hoof wall. Our studies of the pathology of early laminitis indicates that one of the first structures to show any significant change is the basement membrane. The basement membrane (BM) is a thin, unbroken, sheet of extracellular material, partitioning the dermis from the epidermis, and lines the entire inner hoof wall sole and frog (Pollitt, 1994). Thus, all the dermal papillae of the periople, coronary groove, sole and frog as well as the dermal lamellae of the inner hoof wall and bars have a BM lining.
The BM is thus a structural boundary between the vascular dermal compartment and the avascular epidermal compartment but nevertheless co-ordinates a number of critically important biological processes between the two tissues. For instance the BM is a scaffold necessary for the patterned development, maturation and continued maintenance of correctly oriented hoof wall epidermal cells. The BM organises the cytoskeletal framework of the epidermal cells and influences the exchange of nutrients, macromolecules and growth regulating factors. The equine hoof is unique in that a single digit on each limb supports the entire weight of the animal and withstands a biomechanical load unequalled by less athletic ungulates. Since the BM is a layer of material connecting the densely keratinised hoof and the fibrous connective tissue emanating from the distal phalanx, two tissue compartments renowned for their durability, the BM too must have remarkable properties of strength and resilience.
We used a variety of laboratory techniques to investigate the properties of the equine hoof BM. For instance the PAS stain, which is known to react strongly with the carbohydrate moiety of proteins clearly outlined the glycoprotein components of the BM of the hoof wall lamellae, as a fine magenta line. The periodic acid silver methenamine (PASM) stain clearly outlined the collagen component of the BM as a fine black line as well as strands of collagen in the connective tissue of the primary dermal lamellae.
Staining sections with PAS and PASM clearly showed the BM at the dermal-epidermal junction of the equine foot and for this purpose were superior to H&E. The BM closely followed the contours of the epidermal lamellae and papillae and there were no spaces between the BM and the basal plasmalemma of the epidermal germinal cells. The use of the special connective stains highlighted the BM and showed that the tips of the SELs were always rounded and that tips of SDLs were always tapered. The BM at the tip of the SDLs, even at the light level, could be seen to penetrate between adjacent epidermal cells deep into the crypts between pairs of SELs. The tips of the SDLs were uniformly arranged on either side of PEL bases and approached the cornified axial spine of the PEL quite closely (a distance equivalent to the length of 12 basal cells).
The PAS stain showed the BM as an isolated dark magenta line and presumably was locating the non collagenous glycoprotein components of the BM. During the past 30 years a number of non collagenous components have been identified in basement membranes, namely the glycoproteins laminin, fibronectin, nidogen and amyloid P component, the sulfated glycoprotein entactin and heparin sulfate proteoglycan (Inoue, 1989).
On the other hand the PASM stain is considered to principally stain the collagen component of the BM namely collagen type IV (Rambourg and Leblond, 1967). Since both the PAS and PASM stains resolve the BM as the same single membrane it has been concluded from this and much immunolabelling evidence that the BM is a complex lattice work of both collagen and glycoprotein molecules (Inoue, 1989). This combination of connective tissue stains in conjunction with routine H&E staining should provide additional information about the lamellar pathology which occurs during laminitis.
The glycoprotein laminin, one of the major non collagenous components of BMs, was identified in equine lamellar BM by immunolabelling with the monoclonal mouse anti-human laminin. Since laminin plays an important role in the differentiation and attachment of the closely associated epidermal cells, and heparin sulfate proteoglycan is believed to influence the passage of ions across the BM (Leblond and Inoue, 1989) the loss of these molecules from the lamellar BM, such as may occur during developmental laminitis, would have important implications in laminitis pathophysiology. Since plasminogen activator and plasminogen is synthesised in mammalian epidermal basal cells (Lazarus, 1985) and plasmin, which can digest laminin (Liotta et al, 1981), may attain high concentrations in plasma during disease states such as endotoxaemia, an investigation into the role of plasmin and other circulating proteases may be fruitful tor students of laminitis.
Transmission electron microscopic examination of lamellar tissues showed that a typical three layered BM followed the plicated contour of the epidermal basal cells. Adjoining the plasmalemma of the epidermal cells was a pale staining layer, the lamina lucida. Fine strands crossed the lamina lucida, seeming to connect the plasmalemma of the epidermal basal cell to the next layer of the BM, the opaque lamina densa. At the tip of many SELs the lamina densa gave off dense staining extensions into the adjoining connective tissue. Surrounding the extensions and the lamina densa was a network of fine strands which formed the final layer of the BM, the pars fibro reticularis. The lace work of fine strands forming the pars fibro reticularis enveloped the numerous collagen fibrils of the surrounding connective tissue and merged with them. Among the network of fine strands were many larger diameter, banded, anchoring fibrils which formed a mesh of recurving loops. When hemidesmosomes were present in the plasmalemma of the basal epidermal cells the number of the fine cords crisscrossing the lamina lucida increased. At the tips of many SDLs the BMs merged and became double, penetrating a short distance between adjacent epidermal basal cells. The double BMs had a central lamina densa and a lamina lucida on either side.
The three layered ultrastructural nature of the lamellar BM was similar to that described for the rat foot pad BM by Inoue (1989) and Leblond and Inoue (1989). Extensions of the lamina densa into the underlying collagen connective tissue of the SDL were common around the tips of the SELs more so than in the rat foot pad. Since the extensions increase the surface area of attachment between the BM and the surrounding connective tissue, the provision of many extensions would seem logical in the horses foot, considering the load experienced by a relatively large ungulate, weight bearing, on single digits. A feature of the pars fibroreticularis in the equine hoof BM were the large numbers of banded anchoring fibrils. They formed a mesh of recurving loops over the dermal surface of the lamina densa which intertwined with the surrounding dermal connective tissue.
The presence of double basement membranes at the tips of SDLs is unusual and appears not to have been described before. Double BMs are usually associated with capillaries, where the BM peripheral to the endothelium of a capillary is closely applied to the BM of an epithelial cell layer and the BMs merge (Leblond and Inoue, 1989). The double BMs found in the lamellae of the equine digit may confer extra strength.
The equine lamellar BM appears to have an increased proportion of those features which confer strength and resiliency to the BM and its adjoining tissues. Thus, where hemidesmosomes occurred there was an increased density of fine cords crossing the lamina lucida and connecting the cytoskeleton of the epidermal cell (via the tonofilaments converging on the attachment plaque of the hemidesmosome) directly to the lamina densa. Extensions of the lamina densa around the tips of SELs, many anchoring fibrils in the pars fibroreticularis and the presence of double Bms at the tips of SDLs appear to increase the surface area of attachment between the epidermal cells of the hoof, the subjacent connective tissue and ultimately the surface of the distal phalanx.
The current concept of the BM is that, instead of it being an amorphous cement layer, as was originally thought (Vracko, 1974), it is instead, a three dimensional anastomosing lattice work of fine interconnecting cords (Leblond and Inoue, 1989). The axial skeleton of the cord network consists of filaments of collagen IV. The collagen IV filaments are ensheathed with glycoproteins, in particular laminin, to become the more electron dense network of cords known as the lamina densa. The digestion of BM with plasmin exposes the collagen IV axial core and reduces the network of cords to a network of finer filaments (Leblond, 1989). If exposure of the lamellar BM to plasmin and other proteases were to occur during developmental laminitis the possibility exists that the BM could be weakened and thus contribute to the acute lamellar degeneration which occurs. Transmission electron microscopy and immunolabelling with antibodies directecl agdinst BM structural proteins woulcl appear to be useful techniques to explore this possibility.
To study the equine hoof basement memhrane and the connective tissue associated with it in more detail we removed all of the cytoplasmic and nuclear components of the region with a detergent/enzyme treatment. Examination with the scanning electron microscope of mid dorsal hoof wall lamellar tissue blocks showed that the detergent/DNAse treatment had rendered the tissue acellular. At higher magnification, holes in the cut surface of some epidermal cells showed where DNAse had removed cell nuclei. The detergent treatment had solubilized the cell membranes and cytoplasm between adjoining basal cells leaving spaces between the more densely keratinised portions of the cells. The BM remained firmly attached to the connective tissue of each waferlike SDL so only the epidermal surface of the 13M was visible. The surface of the exposed BM was generally smooth with a few folds and wrinkles.
Examination of ultra thin sections of the enzyme/detergent treated tissue with the transmission electron microscope (TEM) confirmed that the epidermal basal cells were anuclear and had lost their cytoplasm and cell membranes. The dark staining lamina densa of the BM had survived the treatment and had separated intact with the connective tissue of the SDLs.
Examination of the surfaces of the acellular tissue blocks with the SEM, showed where large sections of the BM and the attached dermal connective tissue had separated from the epidermis. The exposed SEL basal cells, lacking cytoplasm, cell membranes or nuclei and consisting only of keratinised extracellular matrix retained their shape and architecture. On the other hand the BM and the dermis being softer and more compliant was often located beside the epidermis, in collapsed folds, resembling fallen curtains.
Oblique sectioning of the lamellar tissue blocks, near the tips of the SELs, sometimes caused the detergent DNAse treatment to completely remove the SEL tips. The remaining SDLs resembled empty shells and showed how the BM was able to preserve the shape of the lamellar tips despite being unsupported by the cornified epidermal cells. These sections provided the opportunity to examine the cut edges of SDL at higher magnification and showed the interconnecting cords of fine connective tissue which lined the dermal surface of the BM.
To simply separate the hoof wall epidermis from the underlying dermis tissue blocks cut from the lamellae and coronary papillae were soaked a solution of sodium bromide. After incubation of hoof wall tissue blocks with sodium bromide for 24 hours the dermis was easy to separate from the epidermis. When the papilliform hoof wall dermis was fixed in situ the papillae retained their natural shape, even after critical point drying, and resembled a brush border of tapering hairs.
The architecture of the hoof wall dermis, covered with an intact BM, could be viewed "en face" with the SEM. In the lamellar region the shape of the PDLs and SDLs was well preserved despite the absence of the more rigid epidermal framework. The bases of the PDLs were adjacent to the large diameter sublamellar veins. The SDLs, which in life interdigitate with the SELs of the PEL tips, were relatively short and were spaced well apart to accommodate the rounded club shaped SELs in this region. The remainder of the SDLs, in the mid lamellar region and at the PDL tips were longer and resembled wafers.
At higher magnifications the exposed surface of the lamellar BM, from which the epidermal basal cells had been removed, had an overall smooth appearance marked with numerous small indentations and fine wrinkles. No tears or rents in the BM were ever observed on the surfaces of the SDLs. Connective tissue fibres were only observed at the cut edges of PDL s and SDLs. The connective tissue at these sites closely resembled the fine fibres and cords seen at the cut surfaces of lamellae after detergent/DNAse treatment.
Occasionally, the separation caused by the sodium bromide treatment was incomplete and remnants of the plasmalemma of the basal cells created a reticular pattern on the BM surface. There were also smooth indentations on the BM surface where epidermal basal cells were attached. SDLs were absent on the BM surface of the proximal dermal lamellae close to the coronary/lamellar border. As the BM surface of dermal lamella was scanned in a proximo-distal direction SDLs appeared as raised folds, of gradually increasing height, on the surface of the BM. The longest, and the first to appear, were at the tips of the PDLs and the rest arose along an oblique dorsopalmar line until the entire surface of the PDL bore its full complement of SDLs.
The surface of the BM covering the papillae was strikingly different from the lamellar BM. On the surface of each long thin tapering papilla were numerous parallel ridges giving each papilla a fluted appearance. In the valleys between the ridges the BM surface bore smaller ridges which formed a branching reticular network.
At the distal end of each dermal lamella the SDLs lost their leaf shape and fused with the tapering terminal papillae. The basement membrane of the terminal papillae was folded into parallel ridges similar to those on the surface of the coronary papillae. Some of the SDLs were continuous with the parallel ridges of the terminal papillae. Some fine terminal papillae appear to have budded off from the peripheral edge of the SDLs. The discovery that the BM of the both the colonary and terminal papillae is folded into numerous ridges, parallel with the long axis of the papilla, adds to our perception of the microanatomy of the inner hoof wall. The longitudinal ridges of the papillae are analogous to the secondary folds of the lamellae (SELs) and probably play a similar role in increasing the surface area of attachment between the hoof and the distal phalanx. They may also act as guides or channels directing the keratinising daughter cells of the rapidly proliferating epidermal basal cells in a correctly oriented proximo-distal direction. Detailed light microscope studies of the proximal hoof wall by Leach (1980) showed that the most proximal lamellae were devoid of secondary lamellae. As sections were made in a more distal direction the BM became folded to form SELs, initially at the bases of the PELs and then progressively towards the tips. This study corroborates the findings of Leach (1980) by using an unrelated technique (sodium bromide treatment and scanning electron microscopy) to also show that the secondary lamellae originate as folds in the BM of the proximal primary lamellae.
Exposure of the hoof wall BM, by removing the overlying epidermal basal cells with either the detergent/enzyme or the sodium bromide technique, allows almost the entire surface of the BM to be examined in detail. The results of this study show that the BM will survive the rigours of processing for scanning electron microscopy remarkably intact and free of defects and this despite the absence of a supporting epidermal framework. If these m~w techniques are applie(l to hoof wall tissues affected hy laminitis and lesions in the BM and the underlying connective tissue are revealed, then these phenomenon can be safely attributed to the pathology of laminitis and not to artefacts of the technique.
This research was funded by a generous grant from O'Dwyer Horseshoes Pty, Ltd. Australia.
Abrahamson, D.R. (1986). Recent studies on the structure and pathology of basement membranes. J. Pathol.149: 257-278.
Carlson, E.C. and Kenney, M.C. (1980). Preparation and histoarchitecture of ultrastructurally pure glomerular basement membrane. Renal Physiol. 3: 280-287.
Inoue, S. (1989). Ultrastructure of basement membranes. Int. Rev. Cytol.117: 57-98.
Karnovsky, M.J. (1 965). A formaldehydeglutaraldehyde fixation of high osmolality for use in electron microscopy. J. Cell. Biol.2: 137A-138A.
Lazarus, G.S. (1985). Plasminogen activators in epithelial biology. In: Plasminogen activators from cloning to therapy. Eds; Abbate, R., Barni, T. and Tsafari, A. Raven Press, New York.
Leach, D.H. (1980). The structure and function of the equine hoof wall. Ph.D.dissertation. Western College Veterinary Medicine, University of Saskatchewan.
Leblond, C.P. and Inoue, S. (1989). Structure, composition and assembly of basement membrane. Amer. J. Anat.185: 367-390.
Linford, R.L. (1987). A radiographic, morphometric, histological and ultrastructural investigation of lamellar function, abnormality and the associated radiographic findings for sound and footsore thoroughbreds and horses with experimentally induced traumatic and alimentary laminitis. Thesis. University of California, Davis, USA.
Liotta, L.A., Goldforb, R.H., Brundage, R., Siegal, G.P., Terranova, V. and Garbisa, S. (1981). Effect of plasminogen activator (Urokinase), plasmin and thrombin on glycoprotein and collagenous components of basement membrane. Cancer Res.
Lowden, S. and Heath, T. (1992). Lymph pathways associated with Peyer's patches in sheep. J. Anat. 181: 209-217.
May, S. A., Spiers, S., Carter, S. D., Bennett, D and Edwards, G.B. (1992). Identification of tumour necrosis factor in the blood and peritoneal cavity of horses with colic. Equine vet. J. Suppl.13 :19-21.
Molyneux, G.S., Haller,C.J., Mogg, K. and Pollitt C.C. (1994) The structure, innervation a n d I ocat i on of arteriovenous anastomoses in the equine foot. Equine vet. J.26 (in press).
Pollitt, C.C. (1994) The basement membrane at the equine hoof dermal epidermal junction. Equine vet. J.26 (in press).
Pollitt, C.C. and Molyneux G.S. (1990) A scanning electron microscopic study of the dermal microcirculation of the equine foot. Equine vet. J. 22, 208-212.
Rambourg, A. and Leblond, C.P. (1967). Staining of basement membranes and associated structures by the periodic acid-Schiff and periodic acid-silver methenamine techniques.
Stump, J.E. (1967). Anatomy of the normal equine foot including microscopic features of the laminar region. l. Am. Vet. Med. Assoc.151: 1588-1598.
Tracey, K.J., Vlassara, H and Cerami, A. (1989). Cachectin/tumour necrosis factor. Lancet.i: 11221126.
Vracko, R. (1974). Basal lamina scaffold-anatomy and significance for maintenance of orderly tissue structure. Am. l. Pathol. 77: 314-346.
Warfel, K.A. and Hull, M.T. (1984). Migration of Iymphocytes through the cutaneous basal lamina in normal skin: an ultrastructural study. Anat. rec. 208
Dr Chris Pollitt is the senior lecturer in equine medicine at the School of Veterinary Science, University of Queensland, Australia. He was a general practioner for ten years until being awarded a PhD degree by the University of Queensland, in 1983,for his research into the blood types of horses. Dr Pollitt has presented his research on the structure, function and diseases of the horse's foot to the American Farrier's Association in 1990, the American Association of Equine Practitioners in 1991, the Bntish Equine Veterinary Association in 1992 and the World Equine Veterinary Association in Geneva, Switzerland in 1993. His scientific research papers have been published exclusively in the Equine Veterinary Journal. A colour atlas entitled "Diseases and Disorders of the Horses Foot" containing text and over 400 colour photographs of Dr Pollitt's work is scheduled to be published by Mosby-Wolfe, U.K., in December 1994. Chris is an enthusiastic amateur farrier and actively competes in endurance riding on horses shod by himself. In 1991 he was the Australian heavyweight points rider and heavyweight distance rider of the year. His Australian Stockhorse Monivae Warlock was Australian heavyweight points horse of the year. He is a member of the FEI Veterinary Commission for Endurance at the World Equestrian Games, 1994, at The Hague, Holland.
This article originally appeared in Hoofcare & Lameness: The Journal of Equine Foot Science and is available for your personal use only. Re-publication is prohibited without the express written permission of Hoofcare & Lameness.
Detailed information on this and many other hoofcare topics can be found in Hoofcare & Lameness publisher Fran Jurga's award-winning guide to hoofcare, "Understanding the Equine Foot".
Hoofcare & Lameness