Proliferating Keratinocytes Are Putative Sources of the Psoriasis Susceptibility-Related EDA+(Extra Domain A of Fibronectin) Oncofetal Fibronectin
2004; Elsevier BV; Volume: 123; Issue: 3 Linguagem: Inglês
10.1111/j.0022-202x.2004.23224.x
ISSN1523-1747
AutoresMárta Széll, Zsuzsanna Bata‐Csörgõ, Andrea Koreck, Andor Pivarcsi, Hilda Polyánka, Csilla Szeg, Magdolna Gaál, A. Dobozy, Lajos Kemény,
Tópico(s)Protease and Inhibitor Mechanisms
ResumoThe extra domain A of fibronectin (EDA)+ oncofetal isoform of fibronectin was recently reported to be overexpressed in psoriatic uninvolved epidermis. It has been proposed that the abnormal presence of EDA+ oncofetal protein at the dermal–epidermal junction in the uninvolved skin may provide the "psoriatic" environment in which keratinocytes are in a preactivated state with regard to mitogenic signals (e.g., T cell lymphokines). To determine the possible sources of cellular fibronectin in the non-lesional psoriatic skin, we aimed to investigate whether keratinocytes could produce the EDA+ oncofetal form of fibronectin. RT-PCR studies revealed that both cultured normal keratinocytes and HaCaT cells express the EDA+ splice variant of fibronectin mRNA, and in HaCaT cells the EDA+/EDA− transcript ratio was elevated compared with normal keratinocytes. Cultured keratinocytes and HaCaT cells showed intracytoplasmic staining with an EDA+ fibronectin-specific antibody and among the positively stained cells many showed mitosis. Using RT-PCR, western blot analysis, and flow cytometry, we showed that in synchronized HaCaT cells the amount of both total fibronectin and its EDA+ isoform change with the proliferation/differentiation state of HaCaT cells and peak in highly proliferating cells. We show that in short-term ex vivo cultures, a small population of EDA+ fibronectin containing cell population appear among psoriatic uninvolved, but not normal epidermal cells. We also demonstrate that cell attachment has a strong influence on the expression of both total and EDA+ fibronectin. Our results suggest that proliferating keratinocytes could be the sources of the psoriasis susceptibility-related EDA+ oncofetal fibronectin in the epidermis. The extra domain A of fibronectin (EDA)+ oncofetal isoform of fibronectin was recently reported to be overexpressed in psoriatic uninvolved epidermis. It has been proposed that the abnormal presence of EDA+ oncofetal protein at the dermal–epidermal junction in the uninvolved skin may provide the "psoriatic" environment in which keratinocytes are in a preactivated state with regard to mitogenic signals (e.g., T cell lymphokines). To determine the possible sources of cellular fibronectin in the non-lesional psoriatic skin, we aimed to investigate whether keratinocytes could produce the EDA+ oncofetal form of fibronectin. RT-PCR studies revealed that both cultured normal keratinocytes and HaCaT cells express the EDA+ splice variant of fibronectin mRNA, and in HaCaT cells the EDA+/EDA− transcript ratio was elevated compared with normal keratinocytes. Cultured keratinocytes and HaCaT cells showed intracytoplasmic staining with an EDA+ fibronectin-specific antibody and among the positively stained cells many showed mitosis. Using RT-PCR, western blot analysis, and flow cytometry, we showed that in synchronized HaCaT cells the amount of both total fibronectin and its EDA+ isoform change with the proliferation/differentiation state of HaCaT cells and peak in highly proliferating cells. We show that in short-term ex vivo cultures, a small population of EDA+ fibronectin containing cell population appear among psoriatic uninvolved, but not normal epidermal cells. We also demonstrate that cell attachment has a strong influence on the expression of both total and EDA+ fibronectin. Our results suggest that proliferating keratinocytes could be the sources of the psoriasis susceptibility-related EDA+ oncofetal fibronectin in the epidermis. extra domain A of fibronectin proliferating cell nuclear antigen Psoriasis is a multifactorial hyperproliferative inflammatory skin disease that affects approximately 2%–3% of the European population. Both genetic and environmental factors contribute to the precipitation of psoriatic lesions. It is generally accepted that the genetic background for psoriasis susceptibility is pivotal for the appearance of the symptoms. Intensive family studies since the early 1950s and linkage analysis studies pointed out several genetic loci that play a role in psoriasis (Bhalerao and Bowcock, 1998Bhalerao J. Bowcock A.M. The genetics of psoriasis: A complex disorder of the skin and immune system.Hum Mol Genet. 1998; 7: 1537-1545Crossref PubMed Scopus (230) Google Scholar). In the last decade, a molecular biology approach emerged to identify abnormally expressed genes and proteins contributing to psoriasis (Jackson et al., 1999Jackson M. Howie S.E. Weller R. Sabin E. Hunter J.A. McKenzie R.C. Psoriatic keratinocytes show reduced IRF-1 and STAT-1alpha activation in response to gamma-IFN.FASEB J. 1999; 13: 495-502PubMed Google Scholar;Chen et al., 2000Chen S.H. Arany I. Apisarnthanarax N. et al.Response of keratinocytes from normal and psoriatic epidermis to interferon-gamma differs in the expression of zinc-alpha(2)-glycoprotein and cathepsin D.FASEB J. 2000; 14: 565-571Crossref PubMed Scopus (53) Google Scholar). Previously, we have shown (Bata-Csorgo et al., 1995Bata-Csorgo Z. Hammerberg C. Voorhees J.J. Cooper K.D. Kinetics and regulation of human keratinocyte stem cell growth in short-term primary ex vivo culture. Cooperative growth factors from psoriatic lesional T lymphocytes stimulate proliferation among psoriatic uninvolved, but not normal, stem keratinocytes.J Clin Invest. 1995; 95: 317-327Crossref PubMed Scopus (181) Google Scholar) that the clonogenic basal stem cell population in psoriatic uninvolved skin is more sensitive to lymphokines than basal cells derived from normal epidermis. The T cell lymphokine, interferon-γ, has a pronounced growth-stimulatory effect on uninvolved psoriatic keratinocytes and its effect is synergistically elevated by fibronectin. The keratinocyte receptor for fibronectin, α5β1 integrin, is upregulated both in uninvolved and involved skin of psoriatic patients (Pellegrini et al., 1992Pellegrini G. De Luca M. Orecchia G. et al.Expression, topography, and function of integrin receptors are severely altered in keratinocytes from involved and uninvolved psoriatic skin.J Clin Invest. 1992; 89: 1783-1795Crossref PubMed Scopus (105) Google Scholar). Moreover, we demonstrated that keratinocytes isolated from uninvolved skin of psoriatic patients overexpress the α5, but not α2 and α3 integrins, on their membranes (Bata-Csorgo et al., 1998Bata-Csorgo Z. Cooper K.D. Ting K.M. Voorhees J.J. Hammerberg C. Fibronectin and alpha5 integrin regulate keratinocyte cell cycling. A mechanism for increased fibronectin potentiation of T cell lymphokine- driven keratinocyte hyperproliferation in psoriasis.J Clin Invest. 1998; 101: 1509-1518Crossref PubMed Scopus (59) Google Scholar). In addition to the elevated α5 integrin protein, its ligand, fibronectin, is overexpressed in the uninvolved skin of psoriatic patients in an alternative spliced variant (extra domain A of fibronectin (EDA)+ or oncofetal fibronectin) form. RT-PCR and immunohistochemistry showed that the EDA+ form of fibronectin is overexpressed at the dermal–epidermal junction (DEJ) of uninvolved skin, but not in control skin (Ting et al., 2000Ting K.M. Rothaupt D. McCormick T.S. et al.Overexpression of the oncofetal Fn variant containing the EDA splice-in segment in the dermal–epidermal junction of psoriatic uninvolved skin.J Invest Dermatol. 2000; 114: 706-711Crossref PubMed Scopus (45) Google Scholar). Rothaupt et al have demonstrated that the EDA+ isoform of fibronectin is co-localized with CD11c+ cells at the DEJ of lesional psoriatic skin and hypothesized that the macrophage-derived EDA+ fibronectin might contribute to the initiation of keratinocyte hyperproliferation. Although in the non-lesional skin bone marrow-derived (CD45RO+) cells also showed EDA+ staining in the dermis, given their very low number it is likely that there could be other sources of cellular fibronectin in the non-lesional tissue (Rothaupt et al., 2000Rothaupt D.B. McCormick S.A. Stevens S.R. Cooper K.D. EDA− fibronectin is co-localized with CD11c+ cells at the dermal–epidermal junction in psoriatic tissue.J Invest Dermatol. 2000; 114 (864, (Abstract)): 864Google Scholar). In a recent report, it was demonstrated that basal keratinocytes from uninvolved psoriatic skin exhibit a significantly higher level staining for focal adhesion kinase (FAK) in response to fibronectin. Moreover, FAK tyrosine phosphorylation also had a greater degree in uninvolved psoriatic keratinocytes than in normal keratinocytes. Based on these observations it has been proposed that keratinocytes in the non-lesional psoriatic skin are in a "pre-activated" state to signalling through integrin interactions such as α5β1 integrin signalling (Chen et al., 2001Chen G. McCormick T.S. Hammerberg C. Ryder-Diggs S. Stevens S.R. Cooper K.D. Basal keratinocytes from uninvolved psoriatic skin exhibit accelerated spreading and focal adhesion kinase responsiveness to fibronectin.J Invest Dermatol. 2001; 117: 1538-1545Crossref PubMed Scopus (14) Google Scholar). Ting et al., 2000Ting K.M. Rothaupt D. McCormick T.S. et al.Overexpression of the oncofetal Fn variant containing the EDA splice-in segment in the dermal–epidermal junction of psoriatic uninvolved skin.J Invest Dermatol. 2000; 114: 706-711Crossref PubMed Scopus (45) Google Scholar have demonstrated that the source of the EDA+ fibronectin at the DEJ of uninvolved skin was not the dermal fibroblast.Rothaupt et al., 2000Rothaupt D.B. McCormick S.A. Stevens S.R. Cooper K.D. EDA− fibronectin is co-localized with CD11c+ cells at the dermal–epidermal junction in psoriatic tissue.J Invest Dermatol. 2000; 114 (864, (Abstract)): 864Google Scholar attributed the origin of EDA+ fibronectin to macrophages at the basement membrane zone. Since the presence of EDA+ fibronectin could be responsible for the hyper-responsiveness of uninvolved psoriatic keratinocytes, it is very important to know whether keratinocytes themselves could be the sources of EDA+ fibronectin. We aimed to demonstrate the presence of EDA+ fibronectin in cultured keratinocytes and compare the ratio of EDA+ and EDA− fibronectin mRNA in cultured dermal fibroblasts, in cultured keratinocytes, and in HaCaT keratinocytes. We have also investigated the correlation between the proliferation/differentiation state of keratinocytes and the ratio of EDA+/EDA− fibronectin mRNA and protein expression using synchronized HaCaT keratinocytes that resemble the characteristics of epidermal keratinocytes in different stages of proliferation and differentiation (Pivarcsi et al., 2001Pivarcsi A. Szell M. Kemeny L. Dobozy A. Bata-Csorgo Z. Serum factors regulate the expression of the proliferation-related genes alpha5 integrin and keratin 1, but not keratin 10, in HaCaT keratinocytes.Arch Dermatol Res. 2001; 293: 20-213Crossref Scopus (31) Google Scholar). In order to study the fibronectin mRNA profile of epidermal cells, cultured keratinocytes, HaCaT cells, and fibroblasts, a primer pair was designed that border the EDA motif of fibronectin (Figure 1). After RT-PCR analysis, the resulting bands were easily distinguishable: the 847 bp band corresponds to the EDA− form of fibronectin, whereas the 1221 bp indicates the EDA+ oncofetal form of fibronectin. After densitometry, both total fibronectin mRNA accumulation and EDA+/EDA− splice variant ratio (Figure 2) were compared in the three different cell types. All three cell types were studied at the same proliferation stage: the culture flasks were approximately 70%–80% covered with the cells at the time of sample collection. G3PDH-specific RT-PCR analysis proved the equal quality of total RNA preparations from three different cell types (data not shown). Normal human keratinocytes express a dramatically lower level (0.35±0.02%, n=3) of fibronectin mRNA compared with cultured fibroblasts taken as 100% (n=2). The immortalized HaCaT keratinocytes exhibit a significantly higher level of total fibronectin mRNA when compared with normal keratinocytes, that is 1.69%±0.41% (n=4) of the fibronectin mRNA accumulation of fibroblasts. In addition to differences in the total fibronectin mRNA levels in the studied cell types, the ratio of the EDA+/EDA− splice variants is also different in each cell type. Both fibroblasts and cultured normal human keratinocytes express the two splice variants at about the same level at the studied state of proliferation: the ratio of EDA+/EDA− splice variants is 0.91±0.24:1 in fibroblasts and 1.21±0.12:1 in keratinocytes. In contrast, in HaCaT keratinocytes, the EDA+ oncofetal form of fibronectin clearly dominates; the EDA+/EDA− ratio is 2.55±0.20:1. Immunocytochemical staining with a mouse monoclonal antibody specific for the EDA motif demonstrates that actively proliferating, subconfluent cultures of fibroblasts, normal keratinocytes, and HaCaT cells express the oncofetal isoform of fibronectin (Figure 3). The level of staining is apparently different in the three examined cell types: all fibroblasts showed uniform intracytoplasmatic staining, but among keratinocytes and HaCaT cells, mostly mitotic cells revealed positive intracytoplasmic staining. After detecting a shifted EDA+/EDA− splice variant ratio in subconfluent HaCaT cells, compared with cultured subconfluent keratinocytes, we aimed to study whether the proliferation/differentiation stages of HaCaT cells affect the fibronectin mRNA profile. Using propidium iodide DNA staining, and α5 integrin, keratin 1, and keratin 10 expression as proliferation and differentiation markers,Pivarcsi et al., 2001Pivarcsi A. Szell M. Kemeny L. Dobozy A. Bata-Csorgo Z. Serum factors regulate the expression of the proliferation-related genes alpha5 integrin and keratin 1, but not keratin 10, in HaCaT keratinocytes.Arch Dermatol Res. 2001; 293: 20-213Crossref Scopus (31) Google Scholar have demonstrated that discrete stages of a synchronized HaCaT culture after release from cell quiescence resemble different populations of epidermal keratinocytes. The serum-starved, contact-inhibited HaCaT cells resemble suprabasal non-proliferating differentiated (K1/K10+) keratinocytes of normal epidermis, whereas the highly proliferative HaCaT cells (α5 integrin+, K1/K10+) after release from cell quiescence resemble the activated, differentiated, transiently amplifying keratinocytes. Here, we demonstrate that the serum-starved, contact-inhibited HaCaT cells express a very low-level fibronectin mRNA that increases dramatically with time after passaging and serum re-addition (Figure 4a and B). The highest levels of total fibronectin mRNA (4.64±1.14-fold and 4.12±0.71-fold, compared with 0 h, n=2) coincide with the highest proliferation rates of HaCaT culture (48 and 72 h), demonstrated by PI staining and the highest levels of α5 integrin mRNA as well as protein expression (Pivarcsi et al., 2001Pivarcsi A. Szell M. Kemeny L. Dobozy A. Bata-Csorgo Z. Serum factors regulate the expression of the proliferation-related genes alpha5 integrin and keratin 1, but not keratin 10, in HaCaT keratinocytes.Arch Dermatol Res. 2001; 293: 20-213Crossref Scopus (31) Google Scholar). As the proliferation of the cells slows, the amount of fibronectin mRNA decreases and in the one-week-old culture it almost reaches the level of the serum-starved HaCaT cells (1.22±0.39-fold, compared with 0 h level, n=2). In addition to changes in total fibronectin mRNA, the ratio of EDA+/EDA− splice variants also varies in the synchronized HaCaT cell culture. The serum-starved HaCaT cells exhibit approximately the same amount of EDA+ and EDA− forms of fibronectin mRNA (Figure 4a and C). Twenty-four hours after release from cell quiescence, when the HaCaT cells have already attached to the surface of the culture flask and start to proliferate, there is three times more EDA+ fibronectin mRNA splice variant than EDA−. The most pronounced difference between the two splice variants (5.70±1.42:1) appears at 24 h that precedes the highest overall fibronectin mRNA level and the highest proliferation rate of HaCaT culture at 48 and 72 h. In parallel with the dramatic decrease of total fibronectin mRNA, the ratio between the EDA+/EDA− splice variants reduces from the 72nd h in the culture, and in the 1-wk culture, it reaches a level similar to the serum-starved cells at 0 h. In order to follow changes in total fibronectin at the protein levels in synchronized HaCaT cells after release from cell quiescence, western blot analysis was carried out. Goat polyclonal antibody was used to detect the amount of total fibronectin in cell lysates. Total fibronectin (Figure 5) was undetectable in the 0 and 12 h samples; its expression appeared in the 24 h sample (1.28±0.58 normalized band intensity in relative unit, (n=2)) and was highest in the 36–48 h samples (1.77±0.45 and 1.48±0.33 normalized band intensities in relative units, respectively (n=2)), and then gradually decreased to a still detectable level (0.46±0.20 normalized band intensity in relative unit (n=2)) at 168 h. This pattern of total fibronectin expression was identical in four independent experiments, with the exception of the 0 h samples, in which a small amount of fibronectin was detectable if the protein was harvested directly from the plates (n=2), but no detectable fibronectin was present when cells were lysed after removal from the tissue culture plates (n=2). This observed difference indicates that HaCaT cells discharge some amount of fibronectin during the 2-wk period of synchronization. To follow the changes of EDA+ oncofetal fibronectin protein in HaCaT cells after release from cell quiescence, western blot and flow cytometry experiments were carried out. Although the monoclonal antibody specific for EDA+ fibronectin showed very faint bands at the appropriate molecular weight in the 48, 72, and 96 h samples on western blot in four independent experiments, its sensitivity was not sufficient in HaCaT cells (data not shown). Therefore, we used flow cytometric analysis to follow the EDA+ fibronectin protein expression. In repeated experiments (n=3), the level of EDA+ fibronectin was indistinguishable from isotype staining in serum-starved, contact-inhibited HaCaT cells (0 h) (Figure 6a). As cells started to proliferate in the culture, FITC fluorescence above the isotype level appeared. The maximal EDA+ fibronectin expression was seen at 72 h (12.56%, 6.54%, and 6.66% of cells were found above the isotype level in the three different experiments) (Figure 6b). In one experiment, double staining with the proliferating cell nuclear antigen (PCNA) mAb showed that all the EDA+ fibronectin-containing cells were among the PCNA+ population (Figure 6b). We also stained freshly separated epidermal cells from normal epidermis, psoriatic non-lesional, and lesional epidermis with antibodies against EDA+ fibronectin (n=4) and PCNA (n=2). We could not clearly detect EDA+ fibronectin cells in these freshly separated epidermal samples. Although EDA+ fibronectin fluorescence always exceeded that of the normal in both non-lesional and lesional psoriatic samples, almost similar shifts were detectable in the isotype control samples (data not shown). Next, we compared differences in EDA+ fibronectin production between normal and non-lesional psoriatic epidermal cells in short-term (72 h) ex vivo cultures. In the psoriatic non-lesional cultures, a small cell population (3.06% of all plated cells) (Figure 7c), appearing clearly above the isotype level (Figure 7d), was detected, indicating the presence of EDA+ fibronectin in these cells. This cell population was not seen in normal epidermal cell cultures (Figure 7a and b). These data indicate that in the non-lesional psoriatic epidermis, there are cells that are more capable of producing EDA+ fibronectin than cells in normal epidermis under certain conditions such as cell–cell contact disruption followed by in vitro culturing. To test how cell attachment affects the expression of fibronectin mRNA and the EDA+/EDA− fibronectin mRNA splice variant ratio, different cell culture surfaces were used. Pre-confluent, proliferating HaCaT cells were seeded on tissue culture plates, fibronectin-coated tissue culture plates, and Teflon and bacterial plates and then cultured for 72 h. (In a preliminary experiment, cells were harvested either at 24 or at 72 h after plating and the 24 and 72 h harvests gave similar results.) When semiconfluent HaCaT cells were seeded on tissue culture dish (TC) and on fibronectin-coated tissue culture plates (F), there was no drastic change in the total amount of fibronectin mRNA (84.53%±0.70% and 95.78%±13.90%, compared with the 0 h sample, n=2, 24 h harvests), and we detected a slight increase in the EDA+ fibronectin mRNA splice variant ratio (1.26±0.05:1 and 1.25±0.13:1, respectively, n=2, 24 h harvests). In contrast to this, on Teflon (TF) and bacterial plates (B), both total fibronectin mRNA (37.20%±7.86% and 25.54%±3.77%, compared with the 0 h sample, n=2, 24 h harvests) and the EDA+ isoform production were downregulated. The ratio of EDA+/EDA− fibronectin mRNA splice variants was higher on Teflon (1.40±0.09:1) than on the bacterial plate (0.64±0.01:1) (Figure 8a and b). These data demonstrate that without sufficient attachment, the cells are not able to express fibronectin mRNA and its EDA+ splice variant. Psoriasis is thought to be a multigenic disorder, and the susceptibility genes have been intensely investigated. Given the complexity of the disease, the modified expression of well-characterized genes in the psoriatic uninvolved epidermis might lead to the understanding of the molecular pathogenesis of this skin disorder. It has been demonstrated that the oncofetal EDA+ isoform of fibronectin is present in the uninvolved psoriatic, but not in normal epidermis (Ting et al., 2000Ting K.M. Rothaupt D. McCormick T.S. et al.Overexpression of the oncofetal Fn variant containing the EDA splice-in segment in the dermal–epidermal junction of psoriatic uninvolved skin.J Invest Dermatol. 2000; 114: 706-711Crossref PubMed Scopus (45) Google Scholar). The presence of the EDA+ fibronectin in the non-lesional epidermis may be the key in providing a "psoriatic" environment to the keratinocytes that renders them hypersensitive to different signals. In in vivo non-lesional skin, fibroblasts do not seem to be the source of cellular fibronectin; only CD45RO+ cells double stain with EDA+ fibronectin (Rothaupt et al., 2000Rothaupt D.B. McCormick S.A. Stevens S.R. Cooper K.D. EDA− fibronectin is co-localized with CD11c+ cells at the dermal–epidermal junction in psoriatic tissue.J Invest Dermatol. 2000; 114 (864, (Abstract)): 864Google Scholar). Here, we demonstrate the presence of EDA+ mRNA and EDA+ isoform of fibronectin protein in cultured fibroblasts, keratinocytes (both in third passage), and in HaCaT keratinocytes with RT-PCR analysis as well as with immunohistochemical staining. In contrast to normal cultured keratinocytes that express the EDA+/EDA− splice variants approximately at the same ratio, HaCaT cells exhibit a dominating level of EDA+ fibronectin mRNA. HaCaT cells, although immortalized and genetically abnormal, are considered to be good models for studying the proliferation/differentiation of human keratinocytes (Boukamp et al., 1988Boukamp P. Petrussevska R.T. Breitkreutz D. Hornung J. Markham A. Fusenig N.E. Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line.J Cell Biol. 1988; 106: 761-771Crossref PubMed Scopus (3270) Google Scholar;Ryle et al., 1989Ryle C.M. Breitkreutz D. Stark H.J. Leigh I.M. Steinert P.M. Roop D. Fusenig N.E. Density-dependent modulation of synthesis of keratin 1 and 10 in human keratinocyte line HACAT and in ras-transfected tumorigenic clones.Differentiation. 1989; 40: 42-54Crossref PubMed Scopus (133) Google Scholar;Pivarcsi et al., 2001Pivarcsi A. Szell M. Kemeny L. Dobozy A. Bata-Csorgo Z. Serum factors regulate the expression of the proliferation-related genes alpha5 integrin and keratin 1, but not keratin 10, in HaCaT keratinocytes.Arch Dermatol Res. 2001; 293: 20-213Crossref Scopus (31) Google Scholar). HaCaT cells, however, have the ability to dedifferentiate and this unique feature may be related to their immortalized nature. Here, we provide yet another feature that may explain why HaCaT cells are immortalized. The EDA+ fibronectin produced and secreted by HaCaT cells could provide an elevated stability for the receptor of fibronectin, α5β1 integrin, and self-sustain a high-level proliferation rate that is characteristic of HaCaT cells. Our data indicate a correlation between proliferation and EDA+ fibronectin production in HaCaT cells. The fact that only a minor population of highly proliferating PCNA+ cells contain EDA+ fibronectin suggest that EDA+ fibronectin is synthetized during a well-defined, short period of the cell cycle and is released relatively quickly from the cells. Both the amount and the EDA+/EDA− fibronectin mRNA splice variant ratio change with the proliferation/differentiation state of HaCaT keratinocytes: after release from cell quiescence, the synchronized cells show the highest level of total fibronectin expression and the highest EDA+/EDA− ratio when the cells actively proliferate. The fact that certain genes such as keratin 1 and keratin 10 are downregulated during this period (Pivarcsi et al., 2001Pivarcsi A. Szell M. Kemeny L. Dobozy A. Bata-Csorgo Z. Serum factors regulate the expression of the proliferation-related genes alpha5 integrin and keratin 1, but not keratin 10, in HaCaT keratinocytes.Arch Dermatol Res. 2001; 293: 20-213Crossref Scopus (31) Google Scholar) excludes the possibility that the elevated level of fibronectin mRNA would reflect only a general increase in gene expression. We demonstrate that culture conditions that result in different cell attachment opportunities affect the total fibronectin mRNA expression as well as the EDA+/EDA− transcript ratio. Our data also suggest that attachment signals are necessary for triggering upregulation of fibronectin mRNA and its EDA+ isoform. That the state of proliferation may direct fibronectin splicing has been indicated in cultured fibroblasts and MDCK epithelial cells (Inoue et al., 1999Inoue T. Nabeshima K. Shimao Y. Koono M. Hepatocyte growth Factor/Scatter factor (HGF/SF) is a regulator of fibronectin splicing in MDCK cells: Comparison between the effects of HGF/SF and TGF-beta1 on fibronectin splicing at the EDA region.Biochem Biophys Res Commun. 1999; 260: 225-231Crossref PubMed Scopus (19) Google Scholar). Both in fibroblasts and MDCK epithelial cells, the EDA+/EDA− fibronectin splice variant ratio was higher at low cell density than at high cell density. The same authors have also demonstrated that during migration, MDCK cells produce a lesser amount of fibronectin mRNA with an elevated EDA+/EDA− ratio (Inoue et al., 2001Inoue T. Nabeshima K. Shimao Y. Meng J.Y. Koono M. Regulation of fibronectin expression and splicing in migrating epithelial cells: migrating MDCK cells produce a lesser amount of, but more active, fibronectin.Biochem Biophys Res Commun. 2001; 280: 1262-1268Crossref PubMed Scopus (17) Google Scholar). In contrast to the findings of elevated EDA+/EDA− ratios occurring due to different stimuli, a decreased EDA+/EDA− ratio has been demonstrated in interleukin-4-activated macrophages (Gratchev et al., 2001Gratchev A. Guillot P. Hakiy N. Politz O. Orfanos C.E. Schledzewski K. Goerdt S. Alternatively activated macrophages differentially express fibronectin and its splice variants and the extracellular matrix protein betaIG-H3.Scand J Immunol. 2001; 53: 386-392Crossref PubMed Scopus (217) Google Scholar). Both cis- and trans-acting factors play important roles in the regulation of fibronectin mRNA splicing. Purine-rich sequence tracks have been described within exon B, which are important for proper 5′ splice selection both in vivo and in vitro (Kuo and Norton, 1999Kuo B.A. Norton P.A. Accurate selection of a 5′ splice site requires sequences within fibronectin alternative exon B.Nucleic Acids Res. 1999; 27: 3945-3952Crossref PubMed Scopus (7) Google Scholar). Moreover, an exonic splicing enhancer (ESE) that binds splicing regulatory (SR) proteins was also identified in the promoter region of the fibronectin gene (Cramer et al., 1999Cramer P. Caceres J.F. Cazalla D. Kadener S. Muro A.F. Baralle F.E. Kornblihtt A.R. Coupling of transcription with alternative splicing: RNA pol II promoters modulate SF2/ASF and 9G8 effects on an exonic splicing enhancer.Mol Cell. 1999; 4: 251-258Abstract Full Text Full Text PDF PubMed Scopus (250) Google Scholar). The trans-acting factors that bind these cis-elements may act either as activators (T-Ag) or as inhibitors (VP16) of EDA inclusion (Kadener et al., 2001Kadener S. Cramer P. Nogues G. et al.Antagonistic effects of T-Ag and VP16 reveal a role for RNA pol II elongation on alternative splicing.EMBO J. 2001; 20: 5759-5768Crossref PubMed Scopus (104) Google Scholar). Comparing the above cis-elements and trans-acting factors in different cell lines (normal cultured keratinocytes versus HaCaT cells) and under various pathological conditions (e.g., psoriasis), where the EDA+/EDA− fibronectin mRNA splice variant ratios differ dramatically, may reveal variances in the fibronectin gene itself that affect splicing processes. So far, only two groups of growth factors are known that can affect the splicing of fibronectin: the hepatocyte growth factor/scatter factor (HGF/SF) (Inoue et al., 1999Inoue T. Nabeshima K. Shimao Y. Koono M. Hepatocyte growth Factor/Scatter factor (HGF/SF) is a regulator of fibronectin splicing in MDCK cells: Comparison between the effects of HGF/SF and TGF-beta1 on fibronectin splicing at the EDA region.Biochem Biophys Res Commun. 1999; 260: 225-231Crossref PubMed Scopus (19) Google Scholar) and the transforming growth factor β (TGF-β) (Li et al., 2000Li J. Tripathi B.J. Tripathi R.C. Modulation of pre-mRNA splicing
Referência(s)