Biomolecular phenotyping and heterogeneity assessment of mesenchymal stromal cells using label-free Raman spectroscopy.

Easy, quantitative measures of biomolecular heterogeneity and high-stratified phenotyping are needed to identify and characterise complex disease processes at the single-cell level, as well as to predict cell fate. Here, we demonstrate how Raman spectroscopy can be used in the difficult-to-assess case of clonal, bone-derived mesenchymal stromal cells (MSCs) to identify MSC lines and group these according to biological function (e.g., differentiation capacity). Biomolecular stratification is achieved using high-precision measures obtained from representative statistical sampling that also enable quantified heterogeneity assessment. Application to primary MSCs and human dermal fibroblasts shows use of these measures as a label-free assay to classify cell sub-types within complex heterogeneous cell populations, thus demonstrating the potential for therapeutic translation, and broad application to the phenotypic characterisation of other cells.

Effects of endothelial cells on human mesenchymal stem cell activity in a three-dimensional in vitro model.

An increasing body of data suggest that mesenchymal stem cells (MSCs) reside in a perivascular niche. To more closely mimic this in vivo microenvironment and for better understanding of its complexity, and the factors that regulate the MSC activity, human umbilical vein endothelial cells (HUVECs) were co-cultured with human bone marrow MSCs--using a novel three-dimensional (3D) spheroid co-culture system. Using confocal microscopy of fluorescently labelled cells, we observed HUVECs and MSCs to self-assemble and form organised structures with segregated cell-type partitioning. Under osteogenic conditions, the rate and extent of differentiation in MSC/HUVEC spheroids was significantly elevated compared to 3D co-cultures of MSCs and human dermal fibroblast controls as shown by alkaline phosphatase staining. Conversely, HUVECs inhibited adipogenic differentiation and the proliferation of MSCs in 3D co-cultures indicating that HUVECs suppressed MSC cycling and selectively promoted osteogenic differentiation in 3D. We have also shown that HUVECs enhanced activation of endogenous Wnt signalling and bone morphogenetic protein (BMP) signalling as shown by increased levels of active nuclear ß-catenin and pSmad 1/5/8 immunopositivity respectively. These data suggest strongly that endothelial cells regulate the MSC activity in simulated in vivo conditions, by maintaining quiescence and facilitating niche exit via osteogenic differentiation following appropriate cues. Our findings also underline the importance of 3D heterotypic cell-cell interactions in the regulation of MSC behaviour, suggesting that multicellular cocktails and/or 3D-based delivery strategies may be beneficial for bone repair.

Glutamate receptors on human melanocytes regulate the expression of MiTF.

Glutamate is the major excitatory neurotransmitter in the central nervous system but has also important functions in the epidermis. It is involved in keratinocyte barrier function and in re-epithelialization processes after wounding. Recently, glutamate signalling has been suggested to be implicated in the development of melanoma. The present study examined the expression and functionality of metabotropic and ionotropic glutamate receptors on normal human melanocytes. We found that cultured melanocytes expressed the ionotropic glutamate receptors GluR2 and 4 [alpha-amino-3-hydroxy-5-methyl-4-isoxsazolepropionic acid (AMPA) receptors] and N-methyl-d-aspartate (NMDA) receptors 2A and 2C and possibly the metabotropic glutamate receptor 1. Melanocytes were also found to express specific glutamate transporters and decarboxylases, but appeared neither to produce nor to release l-glutamate. Stimulation with 10 or 100 microM AMPA or NMDA elevated intracellular calcium concentrations in melanocytes, and thus demonstrated the functionality of the glutamate receptors. Millimolar concentrations of l-glutamate did not induce melanocyte toxicity and had no stimulating effect on melanin production. However, blockage of AMPA and NMDA receptors with CFM-2, memantine or MK801 caused a rapid and reversible change in melanocyte morphology, which was associated with disorganisation of actin and tubulin microfilaments. After 24 h of treatment with the AMPA receptor inhibitor CFM-2, there was a sharp reduction in the expression of the crucial melanocyte differentiation and proliferation factor MiTF. The results of this study demonstrate a role for glutamate in melanocyte regulation that may have implications in melanocyte associated disorders.

Are unconventional NMDA receptors involved in slowly adapting type I mechanoreceptor responses?

Specific immunohistochemical staining for NMDA receptor NR2A/B subunits was found in the outer root sheath layer of rat sinus hair (whisker) follicle. Co-localization with CK 20 confirmed that Merkel cells were stained. The NR2A/B staining seen on Merkel cells was pericellular. In addition it appeared that NF70-positive staining was in close proximity to, but did not colocalise with NR2A/B immunoreactivity, indicating that NR2A/B was only expressed by Merkel cells and not their adjacent nerve terminals. Merkel cells and the nerve terminals have previously been associated with electrophysiological recordings from slowly adapting type I (St I) mechanoreceptor unit activity. Pharmacological experiments with isolated sinus hairs using a wide range of ionotropic glutamate receptor antagonists found that only certain NMDA receptor blockers depressed St I unit responses to mechanical stimuli. AMPA/kainate receptor antagonists (CNQX and NBQX, 100 microM) had no effect, nor did classical competitive NMDA receptor antagonists, D-AP5 (600 microM) and R-CPP (100 microM), nor the NMDA glycine site antagonist 5,7-dichlorokynurenic acid (100 microM). The only effective NMDA receptor blockers were those selective for the polyamine site: ifenprodil (IC50 20 microM) and Ro 25-6981 (IC50 approximately 50 microM), and the associated ion channel: MK 801, ketamine and (+/-)-1-(1,2-diphenylethyl)piperidine (IC50 < 100 microM). The two enantiomers of MK 801 were equipotent. All effects were long lasting, consistent with their non-/uncompetitive actions. The most potent drug tested, ifenprodil, at an effective dose of 30 microM, had a mean recovery time of 74 min. A three-fold increase in drug concentration was required to depress St II units (associated with non-synaptic lanceolate endings). Changes in Zn2+ did not affect St I unit responses. These data suggest that unconventional NMDA receptors are involved in St I unit responses, but question the notion of a glutamatergic synapse between the Merkel cell and nerve terminal.

The NMDA type glutamate receptors expressed by primary rat osteoblasts have the same electrophysiological characteristics as neuronal receptors.

Cells of mammalian bone express glutamate receptors. Functional N-methyl-D-aspartate (NMDA) receptors have been demonstrated in human, osteoblastic MG-63 cells, but currents in these cells, unlike those of mammalian neurons, are blocked by Mg(2+) in a voltage-insensitive manner. Differences between the characteristics of NMDA currents in bone cells and in neurons may reflect molecular variation of the receptors or associated molecules, with implications for the role(s) of glutamate in these different tissues and for targeting of ligands/antagonists. To determine whether NMDA receptors in primary bone cells are functional, and whether the currents carried by these receptors resemble those of MG-63 cells or those of mammalian neurons, we have applied the whole cell patch clamp technique to primary cultures of rat osteoblasts. In 0-Mg(2+) saline, 25% of cells showed a slowly developing inward current in response to bath perfusion with 1 mM or 100 microM NMDA. Antibodies against NMDA receptors stained approximately 26% of cells. When NMDA was applied by rapid superfusion, kinetics of the currents were similar to those of neuronal NMDA currents, reaching a peak within 20-30 ms. 1 mM Mg(2+) reduced current amplitude at negative holding potentials and caused the I-V relationship of the currents to adopt a 'J' shape rather than the linear relationship seen in the absence of added Mg(2+). Co-application of glycine (20 microM) with NMDA increased current amplitude by only 18%, suggesting that glycine is released from cells within the cultures. Currents were blocked by (+)-MK-801 and DL-2-amino-5-phosphonovaleric acid. Fluorimetric monitoring of [Ca(2+)](i) using fura-2 showed that, in Mg(2+)-free medium, NMDA caused a sustained rise in [Ca(2+)](i) that could be reversed by subsequent application of MK-801. We conclude that rat femoral osteoblasts express functional NMDA receptors and that these receptors differ from those previously identified in MG-63 cells. NMDA receptors of primary osteoblasts show a 'classical' voltage-sensitive Mg(2+) block, similar to that seen in neuronal NMDA receptors, and will therefore function as detectors of coincident receptor activation and membrane depolarization.

Evidence for targeted vesicular glutamate exocytosis in osteoblasts.

Regulated intercellular signaling is essential for the maintenance of bone mass. In recent work we described how osteoblasts and osteoclasts express functional receptors for the excitatory amino acid, glutamate, indicating that a signaling pathway analogous to synaptic neurotransmission exists in bone. Here, we show that osteoblasts also express the essential molecular framework for regulated glutamate exocytosis to occur as is present in presynaptic neurons. A combination of reverse transcription-polymerase chain reaction (RT-PCR) and northern and western blotting is used to show expression of the target membrane-SNARE (soluble NSF attachment protein receptor), proteins SNAP-25 and syntaxin 4 and the vesicular-SNARE protein VAMP (synaptobrevin), the minimum molecular requirements for core exocytotic complex formation. Immunofluorescent localizations reveal peripheral SNAP-25 expression on osteoblastic cells, particularly at intercellular contact sites, colocalizing with immunoreactive glutamate and the synaptic vesicle-specific protein, synapsin I. We also identify multiple accessory proteins associated with vesicle trafficking, including munc18, rSec8, DOC2, syntaxin 6, and synaptophysin, which have varied roles in regulated glutamate exocytosis. mRNA for the putative Ca(2+)-dependent regulators of vesicle recycling activity, synaptotagmin I (specialized for fast Ca(2+)-dependent exocytosis as seen in synaptic neurotransmission), and the GTP-binding protein Rab3A are also identified by northern blot analysis. Finally, we demonstrate that osteoblastic cells actively release glutamate in a differentiation-dependent manner. These data provide compelling evidence that osteoblasts are able to direct glutamate release by regulated vesicular exocytosis, mimicking presynaptic glutamatergic neurons, showing that a process with striking similarity to synaptic neurotransmission occurs in bone.

Glutamate signalling in non-neuronal tissues.

Since the discovery of its role in the CNS, glutamate, together with its involvement in signalling at synapses, has been the subject of a vast amount of research. More recently, it has become clear that glutamate signalling is also functional in non-neuronal tissues and occurs in sites as diverse as bone, pancreas and skin. These findings raise the possibility that glutamate acts as a more widespread 'cytokine' and is able to influence cellular activity in a range of tissue types. The impact of these discoveries is significant because they offer a rapid way to advance the development of therapeutics. Agents developed for use in neuroscience applications might be beneficial in the modulation of pathology peripherally, impacting on conditions such as osteoporosis, diabetes and wound healing.

Best5: a novel interferon-inducible gene expressed during bone formation.

Regulation of bone formation is important in the pathogenesis of many conditions such as osteoporosis, fracture healing, and loosening of orthopedic implants. We have recently identified a novel rat cDNA (best5) by differential display PCR that is regulated during osteoblast differentiation and bone formation in vitro and in vivo. Expression of best5 mRNA is induced in cultures of osteoblasts by both interferon-alpha (IFN-alpha) or IFN-gamma. Whereas IFN-alpha induced a rapid, transient induction of best5 expression peaking at 4-6 h poststimulation, IFN-gamma elicited a more prolonged induction of best5 expression, which remained elevated 48 h poststimulation. A polyclonal antibody generated to a peptide derived from the best5 coding region recognized a 27 kDa protein on Western blot analysis of osteoblast lysates. We localized BEST5 protein in osteoblast progenitor cells and mature osteoblasts in sections of rat tibiae and in sections of bones loaded in vivo to induce adaptive bone formation. Best5 may therefore be a fundamental intermediate in the response of osteoblasts to stimuli that modulate proliferation/differentiation, such as interferons or mechanical loading. These findings highlight the close interactions between the immune system and bone cells and may open new therapeutic avenues in modulating bone mass.

Localization of ADAM10 and Notch receptors in bone.

In Drosophila melanogaster, the role of the metallodisintegrin, Kuzbanian (kuz), is thought to involve activation of the Drosophila Notch receptor that plays a role in cell-fate determination during neurogenesis and myoblast differentiation. To understand the possible function(s) of a-disintegrin and metalloproteinase (ADAM10), the mammalian ortholog of kuz, in the skeleton, we studied its expression as well as the messenger RNA (mRNA) encoding one candidate substrate, the mammalian Notch2 receptor in bone, bone cells, and cartilage. In sections of neonatal rat tibiae, ADAM10 is expressed in specific regions of articular cartilage and metaphyseal bone. Expression of ADAM10 in articular cartilage occurs predominantly in superficial chondrocytes and becomes more sporadic with increasing distance from the articular surface. In bone, ADAM10 is expressed by periosteal cells, osteoblasts, and osteocytes at locations of active bone formation. Osteoclasts did not express ADAM10. Notch2 mRNA expression was not detectable in superficial chondrocytes. However it colocalized at all sites of ADAM10 expression in bone cells. In vitro, both primary human osteoblasts and osteoblast cell lines expressed a single 4.5 kb and 7.5 kb transcript of ADAM10 and the Notch2 receptor homolog, respectively. Subcellular localization of the ADAM10 protein in MG-63 cells was determined using immunofluorescent techniques. These observations showed clearly that the ADAM10 protein was expressed in the trans-Golgi network and on the plasma membrane. Western blot analysis of fractionated cells showed that, in the plasma membrane fraction, the previously characterized 58 kDa and 56 kDa isoforms were present, whereas, in the trans-Golgi network, the ADAM10 protein was present in several additional bands, possibly indicative of further interdomain processing of the ADAM10 protein. The metallodisintegrins (ADAMs) have several putative functions, including modulation of cell adhesion, membrane-associated proteolysis, and cell-cell signaling. These observations suggest that, in bone but not cartilage, ADAM10 has catalytic activity within the transGolgi network and may play a role in the activation of Notch receptor homologs. This implicates ADAM10 in cell-fate determination of osteoblast progenitor cells, possibly during skeletal development and normal bone remodeling. Plasma-membrane-associated ADAM10 may confer alternative functions.

Osteoblast-derived acetylcholinesterase: a novel mediator of cell-matrix interactions in bone?

The adhesive interactions that occur between bone cells and the developing matrix during bone formation help guide coupled remodeling and the maintenance of bone mass. Here, we provide evidence that acetylcholinesterase (AChE) is a novel osteoblast-derived mediator of cell-matrix interactions in bone. These findings complement an increasing body of evidence which suggests that AChE, in addition to its role in terminating cholinergic signaling, may be instrumental in regulating cellular differentiation and adhesion. We have shown, using RT-PCR, that osteosarcoma cell lines and primary cultures of osteoblasts express AChE mRNA. Expression appeared to be differentiation-dependent, and restricted to AChE splice variants containing the T subunit (exon 6). Immunofluorescent localization demonstrated that these osteoblastic cells expressed protein for AChE with an intracellular vesicular distribution. Immunohistochemistry on tissue sections confirmed AChE expression by osteoblasts in vivo, and revealed the presence of AChE along cement lines, also identified by enzyme histochemistry. In vitro functional studies indicated that osteoblast-like cells adhered specifically to and spread on AChE substrates, but did not interact with butyrylcholinesterase, a closely related protein. Our evidence strongly implicates AChE as a novel bone matrix protein, capable of mediating cell-matrix interactions, and as such may be a principal participant in organized bone formation and the regulation of remodeling.

Expression of a functional N-methyl-D-aspartate-type glutamate receptor by bone marrow megakaryocytes.

Better understanding of hemostasis will be possible by the identification of new lineage-specific stimuli that regulate platelet formation. We describe a novel functional megakaryocyte receptor that belongs to a family of ionotropic glutamate receptors of the N-methyl-D-aspartate (NMDA) subtype responsible for synaptic neurotransmission in the central nervous system (CNS). Northern blotting and reverse-transcriptase polymerase chain reaction (RT-PCR) studies identified expression of NMDAR1 and NMDAR2D type subunit mRNA in rat marrow, human megakaryocytes, and MEG-01 clonal megakaryoblastic cells. Immunohistochemistry and in vivo autoradiographic binding of the NMDA receptor-specific antagonist MK-801 confirmed that megakaryocytes expressed open channel-forming NMDA receptors in vivo. Western blots indicated that megakaryocyte NMDAR1 was either unglycosylated or only glycosylated to low levels, and of identical size to CNS-type NMDAR1 after deglycosylation with endoglycosidase F/peptide-N-glycosidase F. In functional studies, we demonstrated that NMDA receptor activity was necessary for phorbol myristate acetate (PMA)-induced differentiation of megakaryoblastic cells; NMDA receptor blockade by specific antagonists significantly inhibited PMA-mediated increases in cell size, CD41 expression, and adhesion of MEG-01 cells. These results provide evidence for a novel pathway by which megakaryocytopoiesis and platelet production may be regulated.

Evidence for a novel glutamate-mediated signaling pathway in keratinocytes.

Phenotypic alterations in keratinocyte behavior are essential for maintaining epidermal integrity during growth and wound repair and rely on co-ordinated cell signaling events. Numerous growth factors and cytokines have been shown to be instrumental in guiding such changes in keratinocyte activity; here we provide evidence which proposes a novel epidermal signaling pathway mediated by the excitatory amino acid glutamate. Glutamate is the major excitatory neurotransmitter at synaptic junctions within the central nervous system; however, we have identified expression in vivo of several regulatory molecules associated with glutamate signaling in keratinocytes. In resting rat skin epidermis, different classes of glutamate receptors, transporters, and a recently described clustering protein were shown to display distinct distribution patterns, supportive of a multifunctional cellular communication pathway. Immunoreactive N-methyl-D-aspartate-type, alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate-type, and metabotropic-type glutamate receptors were colocalized with the specific glutamate transporter EAAC1 in basal layer keratinocytes, and GLT-1, a related transporter, was expressed suprabasally. In full-thickness rat skin wounds, marked modifications in the distribution of N-methyl-D-aspartate receptors and EAAC1 were observed during re-epithelialization, and alterations in N-methyl-D-aspartate receptor expression accompanied embryonic epidermal development, implicating glutamate signaling in these important biologic events. Furthermore, we provide evidence that these receptors are functional in vitro. These data provide strong evidence to support a role for glutamate in the control of epidermal renewal, and therefore suggest potentially novel therapeutic targets for the treatment of skin disease and enhancement of wound healing.

Expression of an N-methyl-D-aspartate-type receptor by human and rat osteoblasts and osteoclasts suggests a novel glutamate signaling pathway in bone.

Signaling between the various types of cells found in bone is responsible for controlling the activity of osteoblasts and osteoclasts, and therefore the regulation of bone mass. Our identification of a neuronal glutamate transporter in osteoblasts and osteocytes suggests the possibility that bone cells may use the excitatory amino acid glutamate as a signaling molecule. In these studies we report the expression of different subtypes of glutamate receptors in osteoblasts and osteoclasts in vitro and in vivo. We have identified expression in human and rat bone cells of N-methyl-D-aspartate receptor-1 (NMDAR-1) and 2D subunits and PSD-95, the NMDA receptor clustering protein associated with signaling in the central nervous system. In situ hybridization and immunohistochemistry localized NMDAR-1 expression to osteoblasts and osteoclasts in human tissue sections. These findings strengthen the suggestion that glutamate is involved in signaling between bone cells.

Mechanically regulated expression of a neural glutamate transporter in bone: a role for excitatory amino acids as osteotropic agents?

Without habitual exercise, bone is lost from the skeleton. Interactions between the effects of loading of bone and other osteotropic influences are thought to regulate bone mass. In an attempt to identify potential targets for therapeutic manipulation of bone mass, we used differential RNA display to investigate early changes in osteocyte gene expression following mechanical loading of rat bone in vivo. One gene found to be down-regulated by loading had high homology to a glutamate/ aspartate transporter (GLAST) previously identified only the mammalian CNS. RT-PCR analysis using primers targeted to the coding region of the published GLAST sequence amplified identical products from bone and brain (but not a range of other tissues). The amplicons were sequenced and found to be identical to the published CNS GLAST sequence. Northern analysis confirmed expression of GLAST mRNA in bone and brain, but not other tissues. In situ hybridization localized GLAST mRNA expression in rat bone to osteoblasts and osteocytes. A GLAST antibody localized high levels of protein expression to osteoblasts, and newly incorporated osteocytes. Interestingly, older osteocytes also expressed detectable levels of GLAST. Another neural glutamate transporter, GLT-1 was immunolocalized to the pericellular region of mononuclear bone marrow cells, while a further antibody to the EAAC-1 transporter failed to bind to bone cells. Five days after loading, GLAST protein expression was undetectable in osteocytes of loaded bone but present in control nonloaded sections, confirming the downregulation detected by differential display. On quiescent periosteal surfaces, GLAST expression was almost absent, while on surfaces where loading had induced cellular proliferation and bone formation, GLAST protein expression was elevated. In the CNS, the expression of glutamate transporters on neuronal membranes is associated with reuptake of released neurotransmitters at synapses, where they have a role in the termination of transmitter action. In this study, we describe for the first time, the expression of GLAST (and GLT-1) in bone, raising the possibility that excitatory amino acids may have a role in paracrine intercellular communication in bone. Manipulation of bone cell function by moderators of glutamate action could therefore provide novel treatments for bone diseases such as osteoporosis.

Integrin receptor involvement in actin cable formation in an in vitro model of events associated with wound contraction.

Actin cables have been reported to act in vivo as contractile 'purse strings' capable of closing embryonic wounds through generation of circumferential tension. Furthermore, their involvement in wounds within in vitro model systems suggests that actin cable contraction may be an important mechanism involved in the process of wound closure. The aim of this study therefore, was to investigate the appearance of actin cables in a contracting fibroblast populated collagen lattice, an in vitro model of events associated with wound contraction. Utilising this in vitro model, the time-course of actin cable production was investigated and the involvement of integrin receptors analysed using immunofluorescent labelling techniques. Over a period of hours distinct cellular cable-like structures developed at the edges of collagen lattices coinciding with the onset of contraction. Cellular organisation within the cable was evident as was polymerisation of actin microfilaments into elongated stress fibres forming a continuous cell-cell 'actin cable' around the circumference of the lattice. Immunolocalisation demonstrated that integrin receptor subunits beta 1 and alpha 2 but not alpha 5 were involved in apparent intimate cell-cell contact between juxtaposed fibroblasts within this actin cable. This study demonstrates the involvement of integrin receptors in actin cable formation within collagen lattice systems undergoing reorganisation. Such integrin involvement may enable participating cells to respond to the tensional status of their surrounding environment and via cell-cell communication, to permit a co-ordinated contraction of the cable. It is concluded that integrin receptor involvement in active actin cable contraction may be involved in the process of wound contraction.

Influence of vitamin D status on hyaluronan localization in bone.

The distribution of hyaluronan was investigated in the proximal tibiotarsal bones of normal (vitamin D-treated) chicks, and chicks with rachitic lesions induced by vitamin D deficiency. Localization studies using a biotinylated hyaluronan-binding probe revealed that in vitamin D-treated chicks, a high level of hyaluronan staining was present in upper proliferative zone cartilage and upper hypertrophic zone cartilage. Hyaluronan staining was greatly reduced in the zone of provisional calcification. In the metaphyses and diaphyses of normal chicks hyaluronan was predominantly localized to the non-bone-forming surfaces of osteoblasts but was also present on the basolateral surfaces of lining cells and osteoclasts. Marked changes in hyaluronan distribution were observed in vitamin D-deficient chicks. The amount of hyaluronan present in proliferative zone growth cartilage was similar to control chicks, although with a more widespread distribution, extending into lower proliferative zone cartilage. In the zone of hypertrophy/calcification, biochemical analyses revealed that hyaluronan levels in rachitic chicks were about 3.6 times greater than in vitamin D-treated chicks; localization studies demonstrated that this increase was associated with the presence of hyaluronan-positive spindle-shaped cells in the metaphyseal vascular spaces. Intense hyaluronan staining was also associated with abundant spindle-shaped cells occupying the marrow spaces of rachitic diaphyseal bone. The distribution of hyaluronan in vitamin D-treated chick bone, and the alterations observed in rachitic tissue suggests a role for hyaluronan in endochondral bone formation.

Influence of the extracellular matrix on fibroblast responsiveness to phenytoin using in vitro wound healing models.

Recent reports indicate that the topical administration of phenytoin to cutaneous wounds can promote repair. However, isolated skin cells (keratinocytes and fibroblasts) in vitro have varied in their response to phenytoin, giving rise to apparently contradictory results. We have examined how the structure of the extracellular matrix in which human dermal fibroblasts are grown in vitro can influence the response of these cells to phenytoin. The results indicate that, when fibroblasts are embedded within freely-contracting, relaxed, types I collagen matrices, they are insensitive to phenytoin treatment. However, if fibroblasts are grown in collagen matrices which are nonretracting and under tension, phenytoin (5-50 micrograms/ml) significantly (P < 0.01) stimulates cell proliferation, and inhibits collagenase activity in a dose- and time-dependent manner. The fact that the effects of phenytoin on dermal fibroblasts are biphasic and influenced by the surrounding matrix may help to explain why in vitro investigations with phenytoin give rise to inconsistent data. It also suggests that the matrix alterations which accompany wound healing may modulate the effects of phenytoin on dermal fibroblasts.

The wounded dermal equivalent offers a simplified model for studying wound repair in vitro.

Dermal equivalents (DEs), fabricated by seeding fibroblasts into a collagen lattice, may be used as in vitro models for studying wounding and the remodelling processes observed in vivo. We investigated fibroblast responses to a wound stimulus in vitro by making small wounds in DEs. We found that, following wounding, fibroblasts appeared to migrate towards and into the wound defect and appeared to initiate the closure of the wound by bringing together the cut collagen surfaces over a period of 5-11 days. Fibroblast movement into the wound defects was significantly stimulated in the presence of EGF and PDGF (2.8- and 3.5-fold respectively) but not TGF-beta. Cell proliferation in wounded DEs was up to 21% greater than in non-wounded DEs and cell numbers were stimulated further by the addition of TGF-beta, EGF and PDGF (1.5-, 1.7- and 1.8-fold respectively). Wounded DEs also displayed a 2.1-fold increase in latent collagenase production followed by a 1.3-fold increase in active collagenase levels compared to non-wounded DEs. Staining actin fibers within fibroblasts using rhodamine-phalloidin showed that fibroblasts in DEs were under tension, but that this tension was lost upon wounding. Subsequently the stress fibers reappeared concomitantly with the observed "healing" process. Additionally a continuous cell-cell actin cable purse-string developed around the entire wound edge which may be involved in wound closure. The findings suggest that the wounded dermal equivalent offers a valuable model for studying wound healing in vitro.