The smooth muscle cell (SMC) is the only cell type present in the healthy mammalian arterial media. During development, these cells synthesise extracellular matrix components, proliferate and have the capacity to migrate. Maturation of the vessel is associated with a relative loss of these functions and 'differentiation' into a cell whose primary function is contraction. This maturation is morphologically associated with a loss of synthetic organelles and a concomitant increase in contractile myofilaments. In contrast to most mature cell types, which are terminally differentiated, vascular smooth muscle has the capacity to reversibly modulate from the 'contractile' state to the morphologically less differentiated 'synthetic' phenotype in order to effect repair following injury. This phenotypic modulation is also an early and determining event in the process of atherogenesis. Similarly, when SMCs are freshly dispersed from a mature aorta and placed into culture they reversibly modulate from the contractile' to 'synthetic' phenotype in which they have an increased capacity to divide, migrate and synthesise extracellular matrix components akin to their immature or de-differentiated counterparts in vivo.
The phenotypic modulation of both vascular and visceral SMCs is associated with altered expression of various contractile, cytoskeletal and adhesion-related proteins. However, despite extensive knowledge examining the quantitative alterations of these proteins during this process, very little is known about their reorganisation.
Utilizing Western analysis, double-label immunofluorescence and confocal microscopy, in Chapter 3 it was investigated whether changes in phenotypic expression of rabbit aortic SMCs in culture could be correlated with alterations in expression and distribution of structural proteins. 'Contractile' state SMC (days 1 and 3 of primary culture) showed distinct sorting of proteins into subcellular domains, consistent with the theory postulated by Small (1995) that the SMC structural machinery is compartmentalised within the cell. Proteins specialised for contraction (a-smooth muscle (SM) actin, smooth muscle-myosin heavy chain (SM-MHC), and calponin) were highly expressed and concentrated in the upper central region of the cell. Vimentin was confined to the body of the cell, providing support for the contractile apparatus but not co-localising with it. In line with its role in cell attachment and motility, β-non muscle (NM) actin was localised to the cell periphery and basal cortex. The dense body protein α-actinin was concentrated at the cell periphery, possibly stabilising both contractile and motile apparatus. Vinculin-containing focal adhesions were well developed, indicating the cells' strong adhesion to substrate. In synthetic' state SMC (passages 2-3 of culture), there was decreased expression of contractile and adhesion (vinculin) proteins with a concomitant increase in cytoskeletal proteins (β-NM actin and vimentin). These quantitative changes in structural proteins were associated with dramatic changes in their distribution. The distinct compartmentalisation of structural proteins observed in 'contractile' state SMC was no longer obvious, with proteins more evenly distributed throughout the cytoplasm to accommodate altered cell function. The results of this experiment support the hypothesis that the contractile and cytoskeletal proteins of SMC occupy separate domains of the cell and the relative importance of these domains is phenotype-dependent. In addition to quantitative changes in protein expression, phenotypic modulation appears to be associated with a reorganization of structural proteins.
This reorganisation of the actin cytoskeleton during SMC phenotypic change, together with the knowledge that the cytoskeleton regulates cell function by acting as a spatial regulator of intracellular signalling, suggests a role for the actin cytoskeleton in phenotype regulation. The Rho family of small GTPases are regulatory molecules that link surface receptors to the organization of the actin cytoskeleton in a number of cell types. One member of this family in particular, RhoA, has been implicated in the development of stress fibres and maturation of focal adhesion sites.
In Chapters 4 and 5, the role of RhoA in the process of SMC phenotypic modulation was investigated in cultured rabbit aortic SMC. RhoA transcription was elevated in the first 3 days of primary culture, and protein expression increased peaking at 2 days post-confluence when SMCs morphologically return to a more 'contractile' state. However, RhoA showed augmented activation (indicated by translocation to the membrane) at three time-points in primary culture: the transition point when SMCs are entering a phase of logarithmic growth and increased motility, upon reaching senescence, and when they return to a more 'contractile' state. In addition, thrombin, heparin and TGF-β, three factors implicated in the regulation of smooth muscle phenotype, were shown to activate RhoA (demonstrated by 'pulling' down GTP-bound Rho using the Rho binding domain of rhotekin) in serum-starved passaged SMCs. In Chapter 5, transient transfection of 'synthetic' state rabbit SMC with constitutively active Rho (val14RhoA) caused a dramatic decrease in cell size and reorganization of cytoskeletal proteins, reminiscent of the 'contractile' phenotype: α-SM actin and SMMHC adopted a tightly packed, highly organised arrangement while vimentin localised to the immediate perinuclear region and focal adhesions were enlarged. Although this phenotype bore a resemblance to the 'contractile' phenotype, the sorting of actin, myosin and calponin along the stress fibres in RhoA-transfected cells was not apparent. Conversely, specific inhibition of endogenous Rho by expression of C3 transferase resulted in the complete loss of actin and myosin filaments, again without affecting the distribution of vimentin. Focal adhesions were dramatically reduced in both size and number. Thus the results of these chapters suggest that RhoA might in fact play a key role in regulating SMC phenotype, with possible preference for a more 'contractile' phenotype.
Though SMC phenotypic modulation has been well characterised, and many factors which inhibit or accelerate this process in primary culture have been identified, inducing passaged vascular SMCs to return to the contractile phenotype in subsequent passages has remained elusive. To address this question. Chapter 6 examined the effect of long-term (12-15 days) confluence and serum deprivation on the phenotype of passage 1-2 rabbit aortic SMCs, as well as the effect of heparin and thrombin - two factors known to maintain SMCs in their contractile state in primary culture and shown in the previous chapter to activate RhoA. We showed that confluence and serum deprivation, with or without exposure to heparin (100 μg/ml) or thrombin (0.5 and 5 U/ml), induced a more contractile morphology. These cells were smaller, significantly rounder and more compact than those treated with 10% FCS (24 hrs post passaging). Western blotting, FACs analysis and immunofluorescence showed that expression of SMC contractile markers calponin, SM-MHC and its SM-2 isoform were slightly (although not significantly) increased. Although α-SM actin expression did not alter significantly with these treatments, β-NM actin levels decreased. Heparin (100 µg/ml), thrombin (0.5 U/ml) and serum starvation inhibited cell migration following wounding of the confluent cell layer compared with cells treated with 10% FCS.
The identification of a role for the small GTP-binding protein Rho in SMC phenotypic modulation (together with knowledge that it regulates SMC contraction, the actin cytoskeleton and focal adhesion expression), led us to examine the impact of C3 transferase (a specific inhibitor of Rho) on SMC phenotype. This enzyme caused a marked decrease in expression of all cytoskeletal proteins and enhanced cell motility. The lipid signalling molecules, sphingosine-1-phosphate and lysophosphatidic acid (both known activators of Rho) were found to enhance the proportion of cells modulating to a more 'contractile' phenotype under conditions of long-term confluence and serum deprivation. However they did not greatly alter the expression of contractile markers in SMCs under these conditions. Neither sphingosine-1-phosphate nor lysophosphatidic acid appeared to enhance motility (wound healing) in SMCs under serum-deprived conditions. This study demonstrates that serum deprivation and longterm confluence alone induce synthetic state' SMCs to return to a more 'contractile' phenotype and that it is enhanced by factors such as heparin and thrombin. It also suggests that Rho may play a role in SMC phenotypic regulation.
In summary, this thesis shows that SMC phenotypic modulation to the 'synthetic' phenotype has both a quantitative and distributional change in contractile-associated, cytoskeletal and adhesion proteins. The reorganisation of the actin cytoskeleton and focal adhesions suggests a possible role for Rho in phenotypic regulation. However, examination of RhoA expression and activation during SMC modulation indicate the importance of Rho activation in SMC of both the 'contractile' and 'synthetic' phenotypes. The inability of sphingosine-1-phosphate and lysophosphatidic acid to elevate contractile protein expression and effectively redifferentiate cultured SMC indicate that Rho (under these conditions) does not have the ability to 'drive' SMC modulation to the fully 'contractile' phenotype. Thus although RhoA is clearly important in SMC biology, a greater knowledge of other signal transduction pathways and their interactions is needed to elucidate how and to what extent they interact with RhoA to determine its biological effect in 'contractile' and 'synthetic' SMCs.