The Ki67+ proliferation index correlates with increased cellular retinol-binding protein-1 and the coordinated loss of plakophilin-1 and desmoplakin during progression of cervical squamous lesions.

AIMS: To investigate the modulation of cellular retinol-binding protein (CRBP)-1 and the desmosomal plaque proteins plakophilin (PKP)-1 and desmoplakin (DP) in correlation with the Ki67+ proliferation index (PI) during the progression of cervical squamous intraepithelial lesions (SIL) to squamous cell carcinoma (SCC). METHODS: Using in situ imaging by brightfield and confocal laser scanning microscopy, the expression of CRBP-1 protein and transcripts, PKP-1, DP and the Ki67 PI were analysed in 38 low-grade (L) SIL, 56 high-grade (H) SIL, 49 SCC, 30 control cervices and 10 human papillomavirus-positive condylomatous lesions. RESULTS: CRBP-1+ cells increased from 11.4% in the normal cervix to 80.3% in LSILs, 92.3% in HSILs and slightly decreased to 78.3% in invasive SCCs (P = 0.0001) in close association with the Ki67 PI (r =0.41; P < 0.0001). PKP-1+ and DP+ cells were correlated (0.32; P < 0.0001) and decreased from normal (81% versus 92.3%) to LSIL (53.1% versus 85.3%), to HSIL (46.4% versus 67.5%) and SCC (35.1% versus 35.9%). The Ki67+ PI was inversely correlated with DP (-0.20, P = 0.0014) and PKP-1 (-0.19, P = 0.015). Condylomata retained low CRBP-1 and high expression of PKP-1 and DP. CONCLUSIONS: The gain of CRBP-1 and the loss of desmosomal proteins occur early in cervical carcinogenesis.

Cardiac myxoma cells exhibit embryonic endocardial stem cell features.

The origin of myxoma, the most frequent tumour of the heart, remains uncertain. Previous phenotypic characterizations have shown heterogeneous results and the most recent hypothesis suggests that cardiac myxoma originates from a primitive pluripotential cardiogenic cell. We investigated the expression of actin isoforms in 30 left atrial myxomas by immunohistochemistry and in eight consecutive tumours by RT-PCR. alpha-Smooth muscle actin (alpha-SMA) protein and/or transcripts were detected in all cases, whereas alpha-cardiac actin was observed in few cases and alpha-skeletal actin was always absent. Besides classical features, vessel-like structures were characterized by cells expressing CD34 and, less frequently, alpha-SMA. Confocal microscopy showed focal co-expression of CD34 and alpha-SMA in myxoma cells, suggesting a gradual loss of stem endothelial markers and the acquisition of myocytic antigens. In order to confirm this hypothesis, early cardiac differentiation markers were also investigated. RT-PCR documented the presence of transcripts for Sox9 (100%), Notch1 (87.5%), NFATc1 (37.5%), Smad6, metalloproteinases 1 and 2 alone or in variable combinations and the absence of ErbB3 and WT1. Myxoma cells maintained phenotypic heterogeneity in vitro, including the expression of alpha-SMA and the presence of stress fibres. These findings document in cardiac myxoma cells phenotypic markers of the embryonic endothelial-to-mesenchymal transformation that precedes terminal differentiation of endocardial cushions, supporting the hypothesis that cardiac myxoma cells may derive from adult developmental remnants.

Modulation of smooth muscle cell proliferation and migration: role of smooth muscle cell heterogeneity.

Proliferation and migration of smooth muscle cells (SMCs) from the media towards the intima are key events in atherosclerosis and restenosis. During these processes, SMC undergo phenotypic modulations leading to SMC dedifferentiation. The identification and characterization of factors controlling these phenotypic changes are crucial in order to prevent the formation of intimal thickening. One of the questions which presently remains open, is to know whether any SMCs of the media are capable of accumulating into the intima or whether only a predisposed medial SMC subpopulation is involved in this process. The latter hypothesis implies that arterial SMCs are phenotypically heterogenous. In this chapter, we will describe the distinct SMC phenotypes identified in arteries of various species, including humans. Their role in the formation of intimal thickening will be discussed.

Cellular retinol-binding protein-1 expression in endometrial stromal cells: physiopathological and diagnostic implications.

AIMS: Cellular retinol-binding protein-1 (CRBP-1) contributes to the maintenance of the differentiated state of the endometrium through retinol bioavailability regulation. The aim was to analyse CRBP-1 expression in endometrial stromal cells at eutopic and ectopic sites in different physiopathological conditions. METHODS AND RESULTS: Antibodies to CRBP-1, CD10 and alpha-smooth muscle actin were applied to proliferative (n = 10), secretory (n = 9) and atrophic (n = 7) endometrium, decidua (n = 4), adenomyosis (n = 5), endometriosis (n = 10), endometrial polyps (n = 9), simple endometrial hyperplasia (n = 6), well-differentiated endometrioid carcinoma (n = 6) and submucosal leiomyomas (n = 5). In some cases, Western blotting and reverse transcription-polymerase chain reaction were also applied. CRBP-1 was expressed by eutopic and ectopic endometrial stromal cells more markedly during the late secretory phase and in decidua of pregnancy. CRBP-1 expression was low in the stroma of atrophic endometrium and absent in myometrium, leiomyomas and cervical stroma. CD10 immunoreactivity was weak in atrophic endometrium and in decidua. CONCLUSIONS: CRBP-1 expression characterizes endometrial stromal cells at eutopic and ectopic sites and appears to be more specific than CD10. The level of CRBP-1 varies in intensity according to hormonal variations, reaching its maximum in predecidua and decidua. Thus, immunodetection of CRBP-1 may help to elucidate the physiopathological changes which occur in endometrial stroma and can also be applied as an adjuvant stromal marker.

The myofibroblast in wound healing and fibrocontractive diseases.

The demonstration that fibroblastic cells acquire contractile features during the healing of an open wound, thus modulating into myofibroblasts, has open a new perspective in the understanding of mechanisms leading to wound closure and fibrocontractive diseases. Myofibroblasts synthesize extracellular matrix components such as collagen types I and III and during normal wound healing disappear by apoptosis when epithelialization occurs. The transition from fibroblasts to myofibroblasts is influenced by mechanical stress, TGF-beta and cellular fibronectin (ED-A splice variant). These factors also play important roles in the development of fibrocontractive changes, such as those observed in liver cirrhosis, renal fibrosis, and stroma reaction to epithelial tumours.

[Effect of actomyosin contractility on focal contacts of myofibroblasts and structure of stress fibers]. / Vliianie aktomiozinovoi sokratimosti na fokal'nye kontakty miofibroblastov i strukturu stress-fibrill.

Myofibroblasts from rat lung were cultivated. These cells in addition to beta- and gamma-cytoplasmic actins, expressed alpha-smooth muscle actin (alpha-SMA) and formed a system of "supermature" focal contacts, which were connected with thick stress-fibers expressing alpha-SMA and myosin II. Reduction of actin-myison contractility by inhibitors BDM and ML-7 lead to stress fiber reorganization, e.g., decrease in their thickness, a selective disappearance of alpha-SMA expression and myosin translocation from bundles to the cytoplasm. Using immunofluorescence, interference-reflection microscopy and morphometry, we have demonstrated that an inhibition of actin-myosin contractility also leads to dispersion of myofibroblastic focal contacts. Phase-contrast and DIC video-enhanced microscopy of live cells showed morphological reorganization at the leading edge after inhibitory treatment. Thus, actin-myosin contractility controls the structure of "supermature" focal contacts of myofibroblasts and alpha-SMA expression in stress fibers.

Focal adhesion features during myofibroblastic differentiation are controlled by intracellular and extracellular factors.

Transforming growth factor beta (TGFbeta), the most established promoter of myofibroblast differentiation, induces ED-A cellular fibronectin and alpha-smooth muscle actin expression in fibroblastic cells in vivo and in vitro. ED-A fibronectin exerts a permissive action for alpha-smooth muscle actin expression. A morphological continuity (called fibronexus), a specialized form of focal adhesion, has been described between actin stress fibers that contain alpha-smooth muscle actin, and extracellular fibronectin, which contains the ED-A portion, in both cultured fibroblasts and granulation tissue myofibroblasts. We have studied the development of these focal adhesions in TGFbeta-treated fibroblasts using confocal laser scanning microscopy, three-dimensional image reconstruction and western blots using antibodies against focal adhesion proteins. The increase in ED-A fibronectin expression induced by TGFbeta was accompanied by bundling of ED-A fibronectin fibers and their association with the terminal portion of alpha-smooth muscle actin-positive stress fibers. In parallel, the focal adhesion size was importantly increased, and tensin and FAK were neoexpressed in focal adhesions; moreover, vinculin and paxillin were recruited from the cytoplasmic pool into focal adhesions. We have evaluated morphometrically the length and area of focal adhesions. In addition, we have evaluated biochemically their content of associated proteins and of alpha-smooth muscle actin after TGFbeta stimulation and on this basis suggest a new focal adhesion classification, that is, immature, mature and supermature. When TGFbeta-induced alpha-smooth muscle actin expression was blocked by soluble recombinant ED-A fibronectin, we observed that the fragment was localised into the fibronectin network at the level of focal adhesions and that focal adhesion supermaturation was inhibited. The same effect was also exerted by the ED-A fibronectin antibody IST-9. In addition, the antagonists of actin-myosin contractility BDM and ML-7 provoked the dispersion of focal adhesions and the decrease of alpha-smooth muscle actin content in stress fibers of pulmonary fibroblasts, which constitutively show large focal adhesions and numerous stress fibers that contain alpha-smooth muscle actin. These inhibitors also decreased the incorporation of recombinant ED-A into fibronectin network. Our data indicate that a three-dimensional transcellular structure containing both ED-A fibronectin and alpha-smooth muscle actin plays an important role in the establishment and modulation of the myofibroblastic phenotype. The organisation of this structure is regulated by intracellularly and extracellularly originated forces.

Mechanical tension controls granulation tissue contractile activity and myofibroblast differentiation.

We have examined the role of mechanical tension in myofibroblast differentiation using two in vivo rat models. In the first model, granulation tissue was subjected to an increase in mechanical tension by splinting a full-thickness wound with a plastic frame. Myofibroblast features, such as stress fiber formation, expression of ED-A fibronectin and alpha-smooth muscle actin (alpha-SMA) appeared earlier in splinted than in unsplinted wounds. Myofibroblast marker expression decreased in control wounds starting at 10 days after wounding as expected, but persisted in splinted wounds. In the second model, granuloma pouches were induced by subcutaneous croton oil injection; pouches were either left intact or released from tension by evacuation of the exudate at 14 days. The expression of myofibroblast markers was reduced after tension release in the following sequence: F-actin (2 days), alpha-SMA (3 days), and ED-A fibronectin (5 days); cell density was not affected. In both models, isometric contraction of tissue strips was measured after stimulation with smooth muscle agonists. Contractility correlated always with the level of alpha-SMA expression, being high when granulation tissue had been subjected to tension and low when it had been relaxed. Our results support the assumption that mechanical tension is crucial for myofibroblast modulation and for the maintenance of their contractile activity.

Alpha-smooth muscle actin expression upregulates fibroblast contractile activity.

To evaluate whether alpha-smooth muscle actin (alpha-SMA) plays a role in fibroblast contractility, we first compared the contractile activity of rat subcutaneous fibroblasts (SCFs), expressing low levels of alpha-SMA, with that of lung fibroblasts (LFs), expressing high levels of alpha-SMA, with the use of silicone substrates of different stiffness degrees. On medium stiffness substrates the percentage of cells producing wrinkles was similar to that of alpha-SMA-positive cells in each fibroblast population. On high stiffness substrates, wrinkle production was limited to a subpopulation of LFs very positive for alpha-SMA. In a second approach, we measured the isotonic contraction of SCF- and LF-populated attached collagen lattices. SCFs exhibited 41% diameter reduction compared with 63% by LFs. TGFbeta1 increased alpha-SMA expression and lattice contraction by SCFs to the levels of LFs; TGFbeta-antagonizing agents reduced alpha-SMA expression and lattice contraction by LFs to the level of SCFs. Finally, 3T3 fibroblasts transiently or permanently transfected with alpha-SMA cDNA exhibited a significantly higher lattice contraction compared with wild-type 3T3 fibroblasts or to fibroblasts transfected with alpha-cardiac and beta- or gamma-cytoplasmic actin. This took place in the absence of any change in smooth muscle or nonmuscle myosin heavy-chain expression. Our results indicate that an increased alpha-SMA expression is sufficient to enhance fibroblast contractile activity.

The effects of extremely low shear stress on cellular proliferation and neointimal thickening in the failing bypass graft.

OBJECTIVE: Previous studies demonstrating a correlation between low shear stress (tau = 5-15 dyne/cm(2)) and experimental vein graft neointimal thickening (NIT) support the role of low tau in vein graft failure. However, a simple linear relationship between low tau and NIT would underestimate the degree of NIT evident in high-grade occlusive lesions of failing human vein grafts. In this study we used a new experimental model that maintains patency at low tau (< 2 dyne/cm(2)), to delineate possible deviations from linearity in the low tau --> NIT hypothesis. METHODS: Thirty-two New Zealand White rabbits underwent creation of a common carotid vein patch with a segment of ipsilateral external jugular vein. Very low tau was created in 13 patches by ligation of the distal common carotid artery, leaving the only outflow through a small muscular branch. Normal tau was created in 11 patches by leaving the common carotid artery outflow intact. High tau was created in eight patches by ligation of the contralateral common carotid artery. Six patches were harvested after 2 weeks for measurement of cell cycle entry by proliferating cell nuclear antigen (PCNA) immunohistochemistry. The remaining 26 patches were harvested after 4 weeks, perfusion fixed, and excised for morphometric analysis. RESULTS: Mean blood flow and tau at implantation ranged from 0.5 to 41 mL/min and 0.07 to 15 dyne/cm(2), respectively. At the time of harvest, 30 of 32 patches remained patent, and the artificially created aberrations in blood flow were maintained (range, 0.7-41 mL/min). After 2 weeks PCNA immunohistochemistry showed a significantly higher level of cell cycling in patches exposed to low tau (40 +/- 5 vs 1.6 +/- 0.3 PCNA-positive cells per high-power field; P <.001), which is equivalent to approximately 20% of the total cells present. In patches harvested after 4 weeks, NIT ranged from 42 to 328 microm and significantly correlated with mean tau at implantation. Patches with very low tau exhibited histologic characteristics similar to those of failing human bypass grafts, including laminar thrombus and flow-limiting luminal stenosis. The relationship between tau and NIT was nonlinear in that extremely low tau (< 2 dyne/cm(2)) resulted in NIT beyond that predicted by a simple linear correlation (P =.003). CONCLUSION: Extremely low tau (< 2 dyne/cm(2)) stimulates high rates of smooth muscle cellular proliferation in arterialized vein patches. NIT is accelerated in these regions of low tau far beyond that predicted by a simple linear model. The nonlinear nature of the cellular proliferative response and NIT at tau less than 2 dyne/cm(2) may explain the rapid progression of neointimal lesions in failing bypass grafts.

Alpha-smooth muscle actin-positive myofibroblasts in endometrial stroma are not a reliable criterion for the diagnosis of well differentiated endometrioid adenocarcinoma in small tissue samples.

Although a desmoplastic stromal reaction in well-differentiated endometrioid adenocarcinoma is considered a major criterion in the differential diagnosis with atypical hyperplasia, this histologic feature has not met with universal approval. Since alpha-smooth muscle (alpha-SM) actin positive myofibroblasts characterize the desmoplastic stromal response in a variety of neoplasms, the present study was undertaken in order to establish whether these cells are also prominent in the stroma of endometrioid carcinoma and if present could be used as a valid criterion in the differential diagnosis between benign and malignant lesions. The present study of 100 endometrial samples showed focal desmoplastic stromal reaction with alpha-SM actin positive myofibroblasts in 30% of small samples and in 50% of hysterectomy specimens with endometrioid carcinoma. In normal endometrium and in benign lesions lacking a desmoplastic reaction, focal stromal alpha-SM actin positivity was a very common finding. Stromal alpha-SM actin-positive cells were also frequently seen in nondesmoplastic stroma of endometrioid carcinoma. Thus the common presence of alpha-SM actin-positive myofibroblasts in normal endometrial stroma and in benign and malignant lesions precludes its usefulness in the diagnosis of well differentiated endometrioid adenocarcinoma, especially in small tissue samples.

Keloids and hypertrophic scars of Caucasians show distinctive morphologic and immunophenotypic profiles.

The aim of this study was to identify possible morpho-phenotypic differences between keloids (K) and hypertrophic scars (HS) in a Caucasian population. Young HS (< or =1 year of age) presented a high number of diffusely distributed spindle-shaped cells (alpha-smooth-muscle actin+ and fibronectin+). Fully developed HS (> 1 year of age and <3 years of age) were characterized by the frequent presence of distinct collagenous cellular nodules (cells: alpha-smooth-muscle actin+ and fibronectin+). Old HS (> or =3 years of age) showed widespread collagenization phenomena. The histological profile of K was not related to the age of the lesion and was characterized by the almost constant presence of abnormally thick, hyalinized collagen fibers, the presence of collagenous cellular nodules, and variable--albeit lower than in HS-- expression of alpha-smooth-muscle actin and fibronectin. Ultrastructurally, myofibroblasts were the predominant cell type in young and fully developed HS and in K. The immune-cell infiltrate was composed of CD3+, CD45RO+, CD4+, human lymphocyte antigen (HLA)-DR+, and lymphocyte function associated antigen (LFA)-1+ T lymphocytes, strictly associated with CD1a+/ CD36+, HLA-DR+, and intercellular adhesion molecule (ICAM)-1+ dendritic cells, both in HS and K. However, different amounts of immune cells were observed in relation to the type and age of the lesion, and these findings support the hypothesis that cell-mediated, major histocompatibility complex (MHC)-class II-restricted immune responses play an important role in the development of HS and K.

Cultured arterial smooth muscle cells maintain distinct phenotypes when implanted into carotid artery.

Cultured arterial smooth muscle cells (SMCs) with distinct phenotypic features have been described by several laboratories; however, it is not presently known whether this phenotypic heterogeneity can be maintained within an in vivo environment. To answer this question, we have seeded into the intima of denuded rat carotid artery 2 SMC populations with well-established distinct biological features, ie, spindle-shaped, not growing in the absence of serum, and well differentiated versus epithelioid, growing in the absence of serum, and relatively undifferentiated, derived from the aortic media of newborn rats (aged 4 days) and old rats (aged >18 months), respectively. We show that these 2 populations maintain their distinct biochemical features (ie, expression of alpha-smooth muscle actin, smooth muscle myosin heavy chains, and cellular retinol binding protein-1) in the in vivo environment. The old rat media-derived SMCs continue to produce cellular retinol binding protein-1 but little alpha-smooth muscle actin and smooth muscle myosin heavy chains, whereas the newborn rat media-derived SMCs continue to express alpha-smooth muscle actin and smooth muscle myosin heavy chains but no cellular retinol binding protein-1. Our results reinforce the notion of arterial SMC phenotypic heterogeneity and suggest that in our model, heterogeneity is controlled genetically and not by the local environment.

Regulation of alpha-smooth muscle actin and CRBP-1 expression by retinoic acid and TGF-beta in cultured fibroblasts.

We have reported that Cellular Retinol Binding Protein-1 (CRBP-1) is expressed de novo during skin wound healing by a proportion of fibroblastic cells which then differentiate into myofibroblasts and express alpha-smooth muscle actin. In fibroblasts cultured from different tissues we have shown that alpha-smooth muscle actin expression, mainly controlled by Transforming Growth Factor-beta (TGF-beta), is also regulated by retinoic acid and that CRBP-1, known to be a retinoic acid-responsive gene, is modulated by TGF-beta. The aim of the present study has been to investigate the relationships between retinoic acid and TGF-beta in regulating the expression of CRBP-1 and alpha-smooth muscle actin in cultured rat subcutaneous tissue fibroblasts. We have observed that the TGF-beta-induced, but not the retinoic acid-induced, alpha-smooth muscle actin expression is associated with a modulation of endogenous TGF-beta and TGF-beta receptors, suggesting that the action of retinoic acid on alpha-smooth muscle actin expression is not mediated by TGF-beta. The expression of CRBP-1 is regulated at the transcriptional level by TGF-beta and retinoic acid but not synergistically, suggesting a possible common pathway. However, retinoic acid, but not TGF-beta, increases the transcription of a transiently transfected chimeric construct containing the retinoic acid response element of the CRBP-1 promoter, indicating that TGF-beta does not influence CRBP-1 through the retinoic acid pathway. Our results indicate that distinct pathways regulate the genes involved in the appearance and evolution of the myofibroblastic cells. The characterization of these pathways will be helpful for the design of drugs influencing wound healing.

Chimeric smooth muscle-specific enhancer/promoters: valuable tools for adenovirus-mediated cardiovascular gene therapy.

Gene transfer with adenoviral vectors is an attractive approach for the treatment of atherosclerosis and restenosis. However, because expression of a therapeutic gene in nontarget tissues may have deleterious effects, artery-specific expression is desirable. Although expression vectors containing transcriptional regulatory elements of genes expressed solely in smooth muscle cells (SMCs) have proved efficient to restrict expression of the transgene, their use in the clinical setting can be limited by their reduced strength. In the present study, we show that low levels of transgene expression are obtained with the smooth muscle (SM)-specific SM22alpha promoter compared with the viral cytomegalovirus (CMV) enhancer/promoter. We have generated chimeric transcriptional cassettes containing either a SM (SM-myosin heavy chain) or a skeletal muscle (creatine kinase) enhancer combined with the SM22alpha promoter. With both constructs we observed significantly stronger expression that remains SM-specific. In vivo, reporter gene expression was restricted to arterial SMCs with no detectable signal at remote sites. Moreover, when interferon-gamma expression was driven by one of these two chimeras, SMC growth was inhibited as efficiently as with the CMV promoter. Finally, we demonstrate that neointima formation in the rat carotid balloon injury model was reduced to the same extent by adenoviral gene transfer of interferon-gamma driven either by the SM-myosin heavy chain enhancer/SM22alpha promoter or the CMV promoter. These results indicate that such vectors can be useful for the treatment of hyperproliferative vascular disorders.

Mechanisms of neointima formation and remodeling in the porcine coronary artery.

BACKGROUND: To characterize the cells responsible for neointima formation after porcine coronary artery wall injury, we studied the expression of smooth muscle cell (SMC) differentiation markers in 2 models: (1) self-expanding stent implantation resulting in no or little interruption of internal elastic lamina and (2) percutaneous transluminal coronary angioplasty (PTCA) resulting in complete medial rupture and exposure of adventitia to blood components. METHODS AND RESULTS: The expression of alpha-smooth muscle (SM) actin, SM myosin heavy chain isoforms 1 and 2, desmin, and smoothelin was investigated by means of immunohistochemistry and Western blots in tissues of the arterial wall collected at different time points and in cell populations cultured from these tissues. The expression of smoothelin, a marker of late SMC differentiation, was used to discriminate between SMCs and myofibroblasts. Both stent- and PTCA-induced neointimal tissues and their cultured cell populations expressed all 4 markers. The adventitial tissue underlying PTCA-induced lesions temporarily expressed alpha-SM actin, desmin, and SM myosin heavy chain isoforms, but not smoothelin. When placed in culture, adventitial cells expressed only alpha-SM actin. CONCLUSIONS: Our results suggest that SMCs are the main components of coronary artery neointima after both self-expanding stent implantation and PTCA. The adventitial reaction observed after PTCA evolves with a chronology independent of that of neointima formation and probably corresponds to a myofibroblastic reaction.

Angiotensin II stimulates alpha-skeletal actin expression in cadiomyocytes in vitro and in vivo in the absence of hypertension.

Using a specific alpha-skeletal actin antibody, we have previously shown, that during hypertension-associated cardiac hypertrophy in the rat, the expression of alpha-skeletal actin in the myocardium is increased, but maintains focal distribution, compared to normotensive animals. In the present study, we have investigated whether alpha-skeletal actin expression can be induced in the absence of hypertension. For this purpose, we have examined transgenic mice overexpressing angiotensinogen exclusively in the heart. These animals are characterized by high cardiac angiotensin II levels and cardiac hypertrophy accompanied or not by high blood pressure depending on their genetic background, i.e. presence of one or two renin genes. Alpha-skeletal actin levels were highly increased in transgenic compared to wild-type myocardium independently of the number of renin genes, indicating that angiotensin II can stimulate alpha-skeletal actin expression in normotensive animals. Additional in vitro experiments using cultured mouse and rat cardiomyocytes showed that angiotension II not only increases alpha-skeletal actin expression but also induces an increase of its incorporation within II-bands compared to control cardiomyocytes. Angiotensin II increases also the expression of alpha-smooth muscle actin in sarcomeres of cardiomyocytes as well as in fibroblastic cells present within the culture.

Early activation of hepatic stellate cells and perisinusoidal extracellular matrix changes during ex vivo pig liver perfusion.

In previous works, we observed during liver transplantation procedure, the early activation of hepatic stellate cells (HSC) which acquire alpha-smooth muscle (SM) actin expression. In this study, we evaluated changes in HSC and in perisinusoidal extracellular matrix during ex vivo pig liver perfusion. Under general anesthesia, pig livers were flushed and removed, and then perfused ex vivo for 6 h with homologous blood. Liver biopsies were taken before and after washout, at 5 min perfusion, and then hourly. Tissues were processed for immunohistochemistry, immunofluorescence, confocal microscopy, in situ hybridization and electron microscopy. Before and after liver washout, alpha-SM actin was present in vessel walls but in very few lobular HSC. After 1 h perfusion, a strong reactivity for alpha-SM actin was present in HSC, particularly along dilated sinusoids. At the ultrastructural level, numerous microfilament bundles appeared in HSC cytoplasmic processes. During perfusion, type I and type IV collagens, type III procollagen, and fibronectin acquired a looser organisation in relation with the enlargement of perisinusoidal spaces; laminin appeared in perisinusoidal spaces around portal areas and fibrillin deposits increased. In situ hybridization studies showed an increase of the type I procollagen mRNA expression mainly in portal tracts and septa. Ex vivo liver perfusion induces: 1) an early activation of HSC which acquire the expression of alpha-SM actin, and 2) significant changes in the perisinusoidal extracellular matrix. These results are compatible with the view that HSC function as liver specific pericytes participating in the regulation of sinusoidal blood pressure.