ARTICLE
Long-term exposure to silica (SiO2) may induce silicosis as
well as extrapulmonary diseases such as scleroderma [1].
Scleroderma (SSc) is a multisystem disorder resulting in connective
tissue fibrosis at various anatomical sites including skin, the gastrointestinal
tract, lung, heart and kidney. In the skin, SSc is characterized by perivascular
infiltrates of CD3+/CD4+ T-lymphocytes, mononuclear
infiltrates expressing LFA-1 and ICAM-1, and EC expressing high amounts
of ICAM-1 [2]. Furthermore, changes in the dermal microvasculature were
observed that included endothelial cell damage, a reduction in the number
of vessels and finally, excessive production of collagen, especially near
the perivascular infiltrates, leading to dermal and subcutaneous fibrosis.
Earlier investigations suggested an involvement of CAMs mediating cell-cell
interactions in early pathogenetic events in scleroderma [2, 3].
The exact pathogenesis of SSc is still unknown, but various mechanisms
including primary angiopathy, autoimmunity and dysregulation of fibroblast
metabolism have been proposed [1, 4, 5]. Infiltration of mononuclear cells
and release of proinflammatory cytokines by these cells has been suggested
for the development of inflammatory and immunological events, typical
for scleroderma as well as for silica-induced scleroderma [1, 6, 7]. We
showed that silica-exposed blood monocytes liberate interleukin (IL)-6
and IL-1ß in a dose-dependent manner [8]. IL-6 liberated into the
connective tissue possibly induces the inflammatory stages of the disease
and is able to initiate a self-perpetuating cycle of activation in fibroblasts
[1]. IL-1ß is known to be a primary activator of endothelium as
well as fibroblasts [9].
We were able to demonstrate an upregulation of interstitial collagenase
I-mRNA in human dermal fibroblasts after direct exposure to silica [10].
In three-dimensional culture systems missing collagen bundles around silica-treated
fibroblasts support the hypothesis of increased levels of active collagenase
around the fibroblasts.
Another cell system that takes part in the pathogenesis of systemic
as well as silica-induced scleroderma is the endothelium. Several authors
demonstrated EC-activation and elevated expression of adhesion molecules
in SSc skin [11-13]. Rudnicka et al. [14] demonstrated an increased
adhesion of PBMC to vascular endothelium in SSc-patients. These cell-cell
interactions can be at least partly mediated by ICAM-1-LFA-1-interactions.
Recent studies have demonstrated that, early in the course of SSc, activated
fibroblasts expressing high levels of types I and III collagen mRNA are
located adjacent to blood vessels that are surrounded by mononuclear cells,
suggesting a possibly causal interaction of these cell types in the skin
[15, 16]. Soluble mediators resulting from interactions of endothelial
and mononuclear cells may directly activate fibroblasts to migrate, proliferate,
and produce excess amounts of matrix proteins.
Ziegler et al. [17] demonstrated deposits of silica-crystals
in the skin of SSc-patients who had occupational contact with silica dust.
It is important to mention the persisting action of silica that remains
in the tissue for many cell cycles and can locally influence the tissue
over a long period of time. Therefore, we investigated the direct influence
of silica on endothelial cells using cultures of human umbilical vein
endothelial cells (HUVEC) as an experimental model and microvascular human
dermal endothelial cells (HDMEC). Silica-mediated activation of EC was
studied at the levels of mRNAs coding for cell surface molecules, receptors
and proteins related to the extracellular matrix and at the respective
protein levels.
Materials and methods
Silica with a crystal size of < or = 5 µm (charge: DQ12 obtained
from the Institut F. Arbeitsmedizin, University of Düsseldorf, Germany)
was used for all investigations. RNA-polymerases for labeling by in
vitro transcription were purchased from Promega (purchased from Serva,
Heidelberg, Germany). 11-DIG-UTP was from Boehringer-Mannheim (Germany),
CSPD from Tropix (purchased from Serva, Heidelberg, Germany). The antibodies
used for flow cytometrical analyses were purchased from Coulter-Immunotech
(Hamburg, Germany). All chemicals used were of analytical or molecular
biology grade.
Cultures of HUVEC
Cells were prepared from umbilical cord by the standard procedure described
by Jaffe et al. [18]. Culture medium M199 (GibcoBRL, Eggenstein,
Germany) was used with the following additives: penicillin (25 µ/ml),
streptomycin (25 µg/ml), ascorbic acid (12.62 µM), glutathione
(12,62 µM), sodium-pyruvate (1 mM), non-essential amino acids (1%),
L-Glu (2 mM), HEPES (10 mM), endothelial cell growth substance (20 µg/ml),
heparin (50 µg/ml), human serum (10%) and fetal calf serum (10%).
The cells were grown in 10 ml culture flasks and the medium was changed
every third day. For the incubation with silica, subconfluent cultures
were used. Cells were incubated with fresh medium containing various concentrations
of silica for different of incubation times. The purity of the cultures
was tested by FACS-analysis using the anti-CD31 antibody as positive and
the fibroblast specific antibody AS02 [19] as negative control. The cells
were used between 3-7 passages in at least 3 parallel experiments. Data
are presented with standard deviations.
Cultures of HDMEC
Human dermal microvascular endothelial cells from adult skin were purchased
from Clonetics (EndoPackTM-MV, purchased from CellSystems,
Remagen, Germany). The EGM-MV Bullekit was used for culture media in accordance
with the original protocol. Exposure to silica was performed in hydrocortisone-free
medium according to the same regime used for the HUVEC. Additionally to
the certificate from Clonetics, the purity of the cultures was tested
by FACS-analyis using the anti-CD31 antibody as positive and the fibroblast
specific antibody AS02 [19] as negative control. The cells were used between
4-6 passages in at least 3 parallel experiments.
RNA analysis
RNA for Northern blot analysis was prepared with the Micro spin mRNA
purification kit (Pharmacia Biotech, Freiburg, Germany) in accordance
with the original protocol. RNA yield was photometrically quantified using
a GeneQuant Photometer (Pharmacia Biotech). 300-400 ng mRNA per lane were
loaded onto a formaldehyde denaturing 1.2% agarose gel according to Sambrook
et al. [20]. The RNA was blotted on an uncharged nylon membrane
and uv-crosslinked. Northern blots were hybridized with Digoxigenin-labeled
in vitro-transcription products from cDNA-clones for ICAM-1 and
interstitial collagenase I. A GAP-DH-probe was used as internal standard
for mRNA-quantity. The cDNA-probes for alpha(1) collagen type I (clone
Hf 677 from Pr. T. Krieg, Cologne) and alpha(1) collagen type III (purchased
from ATCC, Rockville, MD, USA) were used as negative controls to confirm
the purity of the HDMEC cultures at the mRNA level.
The detection of hybridization bands was performed by incubation of
the membranes with anti-DIG-Fab fragments, conjugated to alkaline phosphatase
(Boehringer Mannheim, Germany). CSPD™ (Tropix purchased
from Serva, Heidelberg, Germany) was used as substrate for alkaline phosphatase
to generate a chemiluminescense signal. The densitometric analysis was
performed using a CCD-camera and the BioProfil™ Software
from Vilber-Lourmat (Marne-la-Vallée, France).
FACS-analysis
Cell monolayers were detached by 0.05% trypsin, 0.02% EDTA (Gibco) and
washed twice with PBS. In earlier investigations we demonstrated that
trypsin-treatment did not alter the detectable expression of proteins
on the cell-surface compared to harvesting the cells by mechanical scraping
[19].
Cells (2 x 105) were incubated with 20 µl of the antibody
(anti-CD31, mAb AS02, anti-CD54 and anti-CD62e, stock: 200 µg/ml)
for 45 min at 4° C. After washing with PBS/10% Gelafusal, the cells
were incubated 45 min at 4° C with a goat-anti-mouse antibody-FITC
(fluorescein isothiocyanate) conjugated, washed three times, and fixed
in PBS/10% Gelafusal with 1% formaldehyde. The final evaluation was performed
using flow cytometry with an EPICS™-Flowcytometer (Coulter,
Krefeld, Germany).
Analysis of soluble ICAM-1 and interleukins
After the incubation with silica the medium was collected and stored
at 70° C until the analysis of sICAM-1 and cytokines in ELISA
was performed according to the manufacturer's protocol. The ELISA for
sICAM-1, IL-1ß and IL-6 were purchased from DPC (Bad Nauheim, Germany).
The quantification was performed photometrically using a Dynatech MR5000-reader.
Results
In our experiments, silica was directly added to adherently growing
human endothelial cells. The crystals attached closely to the cells. Silica
was also found at the cell-surface, after intensive washing during trypsinization
without adding new silica to the trypsinated cells. We did not find granules
within the EC using phase-contrast microscopy; the cells grew normally
in the presence of silica and cytotoxicity was excluded using a Cytotox-Assay
(Promega-Serva, Heidelberg, Germany, data not shown). Therefore, we suppose
an attachment-mediated effect of silica on endothelial cells. In general,
HUVEC showed more intensive induction of ICAM-1 than HDMEC in the experiments
demonstrated here. Because of the minor relevance of HUVEC in the pathogenesis
of scleroderma, the data for HUVEC will only be mentioned in the text.
Induction of ICAM-1 mRNA expression by silica in
EC
In Figure 1, a significant
induction of ICAM-1 is shown in silica-treated HDMEC after 24 h incubation.
Shorter incubation times (30 min or 4 h) did not show any effects on the
mRNA level (data not shown).
In contrast to HUVEC, the mRNA steady-state levels for interstitial
collagenase I were induced in HDMEC in a dose-dependent manner. Collagen
type I and type III mRNA were not detectable in Northern hybridization
of HDMEC matching with known characteristics of these cells (data not
shown). The mRNA-levels for interstitial collagenase I remained unchanged
in HUVEC.
Enhanced expression of ICAM-1 on the cell surface
of EC
The mRNA data for an elevated ICAM-1 expression were proved at the protein
level using FACS-analysis. In Figure
2A the increased expression of ICAM-1 on HDMEC is demonstrated.
Trypsinization did not influence the detected levels of membrane-associated
ICAM-1 indicated by investigations with scraped cells (data not shown).
The effects observed were specific to silica because TiO2 did
not induce ICAM-1 on these EC.
These effects were observed in HDMEC as well as in HUVEC, but the increase
of both cell-bound and soluble ICAM-1 was higher in HUVEC (data not shown).
Determination of soluble ICAM-1 in supernatants
of silica-treated EC
To confirm the results from RNA- and FACS-analysis by an additional
method we checked the levels of soluble ICAM-1 in supernatants of silica-exposed
HDMEC and HUVEC. In general, only very small amounts of sICAM-1 could
be detected in supernatants of EC (Fig.
2B). In HUVEC, a significant induction was seen after 16 h of
incubation with silica (data not shown). In HDMEC, only after 48 h of
silica-treatment could sICAM-1 be detected in ELISA. The absolute amounts
of sICAM-1 found in supernatants of HDMEC were < 1 ng/ml, but the increase
due to 100 µg/ml silica reached 150% of the control without addition
of crystals (0.91 ± 0.08 ng/ml versus 0.62 ± 0.07 ng/ml).
This increase is comparable with the increase found in HUVEC (approximately
200% at 100 µg/ml silica).
According to the time course, these results correspond to previously
published data showing maximum expression of ICAM-1 by endothelial cells
in vitro, 24 h after stimulation with TNFalpha and IL-1 [21].
Induction of cytokines in silica-treated, endothelial
cells
Beside adhesion molecules, cytokines play a major role in the activation
of cells and in the regulation of important cell-cell interactions.
Interleukin-6, a pro-inflammatory cytokine produced by many cells taking
part in the development of SSc, was increased significantly in a dose-dependent
manner in supernatants of silica-treated HDMEC (Fig.
3) as well as in supernatants of HUVEC. The differences found
in the absolute amount of IL-6 (i.e. for 100 µg SiO2/ml:
55 ± 3 pg/ml in HDMEC versus 3,900 ± 452 pg/ml in HUVEC
after 16 h of incubation) should be based on the differing characteristics
of these cells and the cell densities reached in the cultures.
Interleukin-1ß could be determined in supernatants of HUVEC whereas
in supernatants of HDMEC the cytokine was not found. Induction of IL-1ß
takes place only after longer incubation, at least 12 h. Shorter incubations
from 30 min up to 4 h did not show any induction. However, the dose-dependent
increase found in supernatants of HUVEC after 24 h of incubation indicates
a specific effect of silica on these EC.
Discussion
Endothelial cells and microvasculature are some of the earliest targets
in the pathogenesis of SSc. For example, Raynaud's phenomenon may precede
the disease by many years [21, 22]. In addition, perivascular infiltrates
of mononuclear cells in the skin also belong to the early events of SSc.
Therefore, endothelial cells play an important role when mononuclear cells
leave the capillaries.
Silica-induced scleroderma is a disease that is known in miners and
other people who have had long-term occupational contact with high concentrations
of silica dust by inhalation and percutaneous penetration [23]. Raynaud's
phenomenon of the fingers (where possibly silica penetrates into the dermis)
often precedes skin lesions. In addition, silica was found even in the
dermis of those patients [17], and the resulting microtraumata probably
represent events related to direct contact between silica and EC. As its
characteristics are well-known, silica-induced scleroderma may serve as
a quite good model for investigating of the pathophysiology of SSc.
The incubation of EC with silica resulted in an activation of these
cells. This process was studied at the levels of adhesion molecules, cytokines
and proteolytic enzymes. Enhanced attachment of mononuclear cells may
be the result of an increased expression of ICAM-1 at the inner surface
of the microvessels and attached monocytes themselves could initiate a
further induction of ICAM-1 in the surrounding EC [24-26].
Several authors [11-13] described increased
expression of ICAM-1 in the dermis of SSc patients. The induction of ICAM-1
in endothelial cells in vivo may enable the inflammatory cells
to adhere to and to permeate the capillary wall forming inflammatory infiltrates
around the microvessels in SSc.
In contrast to the rapid and transient expression of ICAM-1 mRNA on
EC stimulated by cytokine or monocyte-attachment [26], the effect of silica
was more moderate and took 24 h for the detectable induction of ICAM-1.
In the case of silica-induced scleroderma, silica possibly exerts a permanent
influence on EC in contact with the crystals. Beside ICAM-1, other cell
adhesion molecules such as VCAM or ELAM-1 also play a role in the attachment
of mononuclear blood cells to the endothelium [3]. In contrast to cytokine-mediated
processes, the induction of ICAM-1 was not preceded by detectable expression
levels of ELAM-1 or VCAM in the EC. Neither in FACS-analysis and immunohistochemical
studies of cultured EC nor in ELISA of supernatants could ELAM-1 and VCAM
be detected in vitro on EC (data not shown).
Interleukin-6 and IL-1ß are two cytokines playing an essential
role in the activation of various cell types. Elevated IL-6 levels have
been reported to be associated with a variety of diseases, including autoimmune
diseases. IL-6, which is found to be typically increased in the plasma
of SSc-patients [27, 28] was expressed by EC in vitro after treatment
with silica. The silica-induced expression of high amounts of IL-6 by
EC might, in vivo, influence blood cells as well as fibroblasts.
In the vessels, IL-6 stimulates B-cell differentiation and affects T-cells
resulting in IL-2-receptor expression that is a hallmark of the T-cell
response in autoimmune diseases [21, 28]. In the dermis of SSc-patients,
IL-6 is expressed with high intensity and it is thought to act as a proinflammatory
cytokine maintaining the inflammation in the tissue. Additionally, IL-6
enhances fibroblast proliferation and may induce a perpetuating autocrine
mechanism of IL-6 expression in dermal fibroblasts [27-29]. Beside IL-6,
IL-1ß is another powerful proinflammatory cytokine found during
the initial steps of many diseases and being involved in essential defense
and repair mechanisms [30, 31]. In contrast to the significant induction
of IL-6 by silica in EC, IL-1ß was induced by silica in HUVEC only
and was not detectable in supernatants of HDMEC. However, an induction
of IL-1ß has been demonstrated earlier for silica-treated monocytes
[8], and the induction of adhesion molecules by EC after exposure to IL-1ß
has been reported [32].
The observed induction of interstitial collagenase I after incubation
with SiO2 highlights the possible involvement of EC in perivascular
fibrotic reactions mentioned earlier [32]. Monocyte-derived TNFalpha is
also known to induce collagenase in HDMEC [33] supporting the hypothesis
of the involvement of macrophages in the development of SSc. Collagenolytic
activities play important roles during the growth of microvessels as well
as during fibrotic reactions. In SSc, a localization of collagen-mRNA
in the vicinity of microvessels is discussed as a causal agent in the
development of dermal fibrosis [16]. These facts underline the close relationship
between regulation of the connective tissue and the vascular system. However,
this induction of collagenase mRNA was restricted to HDMEC confirming
that their nature differs from that of HUVEC [9].
Taken together, our results show that beside cytokines, silica is also
able to directly activate HDMEC and HUVEC in vitro. In agreement
with the observed pathology, the activation by silica in vitro
is slower and less intense than the cytokine-mediated activation of EC.
Concluding our experimental data and data from the literature [1, 7, 8,
10], silica-associated scleroderma is possibly triggered by silica or
free silic acid found in the serum of these patients. Thereby, silica
activates, at least in vitro, various cell types (monocytes, fibroblasts
and EC) involved in the pathological development of scleroderma.
CONCLUSION
Acknowledgements
This work was supported by grants of the Deutsche Forschungsgemeinschaft
(Ha 2052/1-2) and the Sächsische Akademie der Wissenschaften zu Leipzig.
REFERENCES
1. Haustein U-F, Herrmann K. Environmental scleroderma. Clinics in
Dermatol 1994; 12: 467-73.
2. Gruschwitz M, von den Driesch P, Kellner I, Hornstein OP, Sterry
W. Expression of adhesion proteins involved in cell-cell and cell-matrix
interactions in the skin of patients with progressive systemic sclerosis.
J Am Acad Dermatol 1992; 27: 169-77.
3. Koch AE, Kronfeld-Harrington LB, Szekanecz Z, Cho MM, Harlow LA,
Strieter RM, Kunkel SL, Massa MC, Barr WG, Jimenez SA. In situ
expression of cytokines and cellular adhesion molecules in the skin of
patients with systemic sclerosis. Pathobiology 1993; 61: 239-46.
4. Varga J, Rudnicka L, Uitto J. Connective tissue alterations in systemic
sclerosis. Clin Dermatol 1994; 12: 387-96.
5. Gay S, Boulware DW, Gay RE. Oncogenes, growth factors, and matrix
in the pathogenesis of systemic sclerosis. Clin Dermatol 1994;
12: 419-24.
6. Krieg T, Meurer M. Systemic scleroderma. J Am Acad Dermatol
1988; 18: 457-81.
7. Koeger AC, Lang T, Alcaix D, Milleron B, Rozenberg S, Chaibi P, Arnaud
J, Mayaud C, Camus JP, Bourgeois P. Silica-associated connective tissue
disease. Medicine 1995; 74/5: 221-36.
8. Frank R, Giese T, Dummer R, Walther T, Rytter M, Ziegler V, Haustein
U-F. Silica-induced cytokine release in human monocyte cultures and its
possible involvement in the pathophysiology of silica-associated scleroderma.
Eur J Dermatol 1993; 3: 304-9.
9. Swerlick RA, Lawley TJ. Role of microvascular endothelial cells in
inflammation. J Invest Dermatol 1993; 100: 111-5.
10. Anderegg U, Vorberg S, Herrmann K, Haustein U-F. Increased expression
of interstitial collagenase in silica-treated fibroblasts. Eur J Dermatol
1996; 6/1: 51-5.
11. Sollberg S, Peltonen J, Uitto J, Jimenez SA. Elevated expression
of ß1 and ß2 integrins, intercellular adhesion molecule-1,
and endothelial leucocyte adhesion molecule 1 in the skin of patients
with systemic sclerosis of recent onset. Arthritis Rheum 1992;
35: 290-8.
12. Majewski S, Hunzelmann N, Johnson JP, Jung C, Mauch C, Ziegler-Heitbrock
HWL, Riethmüller G, Krieg T. Expression of intercellular adhesion
molecule-1 (ICAM-1) in the skin of patients with systemic scleroderma.
J Invest Dermatol 1991; 97: 667-71.
13. Claman HN, Giorno RC, Seibold JR. Endothelial and fibroblastic activation
in scleroderma: The myth of the "uninvolved skin". Arthritis Rheum
1991; 34: 1495-501.
14. Rudnicka L, Majewski S, Blaszcyzk M, Skiendzielewska A, Makiela
B, Skopinska M, Jablonska S. Adhesion of peripheral blood mononuclear
cells to vascular endothelium in patients with systemic sclerosis (scleroderma).
Arthritis Rheum 1992; 35: 771-5.
15. Kahaleh MB. Endothelin, an endothelial-dependent vasoconstrictor
in scleroderma. Arthritis Rheum 1991; 34/8: 978-83.
16. Scharffetter K, Lankat-Buttgereit B, Krieg T. Localization of collagen
mRNA in normal and scleroderma skin by in situ hybridization. Eur
J Clin Invest 1988; 18: 9-17.
17. Ziegler V, Keyn J, Mehlhorn J, Kipping D, Haustein U-F. Quarznachweis
in der Sklerodermiehaut. Dermatol Mon Schr 1988; 174: 688-9.
18. Jaffe EA, Nachmann RL, Becker CG, Mimick CR. Culture of human endothelial
cells derived from umbilical veins. J Clin Invest 1973; 52: 2745.
19. Saalbach A, Anderegg U, Schnabel E, Herrmann K, Haustein U-F. A
novel fibroblast specific antibody. Properties and specificities. J
Invest Dermatol 1996; 106: 1314-9.
20. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory
manual. 1989. Cold Spring Harbor.
21. Pearson JD. Endothelium its role in scleroderma. Ann Rheum
Dis 1991; 50: 866-71.
22. Belch JJF. Raynaud's phenomenon: its relevance to scleroderma. Ann
Rheum Dis 1991; 50: 839-45.
23. Haustein U-F, Ziegler V, Herrmann K, Mehlhorn J. Silica-induced
scleroderma. J Am Acad Dermatol 1990; 22: 444-8.
24. Pober JS, Gimbrone MA, Lapierre LA, Mendrick DL, Fiers W, Rothlein
R, Springer TA. Overlapping patterns of activation of human endothelial
cells by IL-1, TNF and immune interferon. J Immunol 1986; 137/6:
1893.
25. Rothlein R, Czaijkowski M, O'Neill MM, Marlin SD, Mainolfi E, Merluzzi
VJ. Induction of intercellular adhesion molecule 1 on primary and continuous
cell lines by proinflammatory cytokines. J Immunol 1988; 141: 1665-9.
26. Combe C, Duplàa C, Couffinhal T, Moreau C, Bonnet J. Induction
of intercellular adhesion molecule-1 by monocyte adhesion to endothelial
cells in human culture system. J Cell Physiol 1995; 164: 295-303.
27. White Needleman B, Wigley FM, Stair RW. Interleukin-1, interleukin-2,
interleukin-4, interleukin-6, tumor necrosis factor alpha and interferon-gamma
levels in sera from patients with scleroderma. Arthritis Rheum
1992; 35/1: 67-72.
28. Ihn H, Sato S, Fujimoto M, Kiguchi K, Takehara K. Demonstration
of interleukin-2, interleukin-4, and interleukin-6 in sera from patients
with localized scleroderma. Arch Dermatol Res 1995; 287: 193-7.
29. Feghali CA, Bost KL, Boulware DW, Levy LS. Control of IL-6 expression
and response in fibroblasts from patients with systemic sclerosis. Autoimmunity
1994; 17: 309-18.
30. Dinarello CA. Blocking interleukin-1 receptors. Int J Clin Lab
Invest 1994; 24: 61-79.
31. Duncan R, Matthew, Bermann. Differential regulation of collagen
glycosaminoglycan, fibronectin, and collagenase activity production in
cultured human adult dermal fibroblasts by interleukin-1-alpha and beta
and tumor necrosis factor-alpha and beta. J Invest Dermatol 1989;
92: 699-706.
32. Mantovani A, Bussolino F, Dejana E. Cytokine regulation of endothelial
cell function. FASEB J 1992, 6: 2591-9.
33. Cornelius LA, Nehring LC, Roby JD, Parks WC, Welgus HG. Human dermal
microvascular endothelial cells produce matrix metalloproteinases in response
to angiogenic factors and migration. J Invest Dermatol 1995;
|
Printable version
