The Microglia-activating Potential of Thrombin
2004; Elsevier BV; Volume: 279; Issue: 50 Linguagem: Inglês
10.1074/jbc.m408318200
ISSN1083-351X
AutoresUwe‐Karsten Hanisch, Denise van Rossum, Yiheng Xie, Klaus Gast, Rolf Misselwitz, Seppo Auriola, Gundars Goldsteins, Jari Koıstınaho, Helmut Kettenmann, Thomas Möller,
Tópico(s)Renin-Angiotensin System Studies
ResumoThe serine protease thrombin is known as a blood coagulation factor. Through limited cleavage of proteinase-activated receptors it can also control growth and functions in various cell types, including neurons, astrocytes, and microglia (brain macrophages). A number of previous studies indicated that thrombin induces the release of proinflammatory cytokines and chemokines from microglial cells, suggesting another important role for the protease beyond hemostasis. In the present report, we provide evidence that this effect is not mediated by any proteolytic or non-proteolytic mechanism involving thrombin proper. Inhibition of the enzymatic thrombin activity did not affect the microglial release response. Instead the cyto-/chemokine-inducing activity solely resided in a high molecular weight protein fraction that could be isolated in trace amounts even from apparently homogenous α- and γ-thrombin preparations. High molecular weight material contained thrombin-derived peptides as revealed by mass spectrometry but was devoid of thrombin-like enzymatic activity. Separated from the high molecular weight fraction by fast protein liquid chromatography, enzymatically intact α- and γ-thrombin failed to trigger any release. Our findings may force a revision of the notion that thrombin itself is a direct proinflammatory release signal for microglia. In addition, they could be relevant for the study of other cellular activities and their assignment to this protease. The serine protease thrombin is known as a blood coagulation factor. Through limited cleavage of proteinase-activated receptors it can also control growth and functions in various cell types, including neurons, astrocytes, and microglia (brain macrophages). A number of previous studies indicated that thrombin induces the release of proinflammatory cytokines and chemokines from microglial cells, suggesting another important role for the protease beyond hemostasis. In the present report, we provide evidence that this effect is not mediated by any proteolytic or non-proteolytic mechanism involving thrombin proper. Inhibition of the enzymatic thrombin activity did not affect the microglial release response. Instead the cyto-/chemokine-inducing activity solely resided in a high molecular weight protein fraction that could be isolated in trace amounts even from apparently homogenous α- and γ-thrombin preparations. High molecular weight material contained thrombin-derived peptides as revealed by mass spectrometry but was devoid of thrombin-like enzymatic activity. Separated from the high molecular weight fraction by fast protein liquid chromatography, enzymatically intact α- and γ-thrombin failed to trigger any release. Our findings may force a revision of the notion that thrombin itself is a direct proinflammatory release signal for microglia. In addition, they could be relevant for the study of other cellular activities and their assignment to this protease. Thrombin (EC 3.4.21.5, factor IIa) is a serine protease catalyzing the cleavage of fibrinogen and the activation of several other components of the blood coagulation cascade (1Gingrich M.B. Traynelis S.F. Trends Neurosci. 2000; 23: 399-407Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar, 2Bode W. Turk D. Karshikov A. Protein Sci. 1992; 1: 426-471Crossref PubMed Scopus (645) Google Scholar, 3Grand R.J. Turnell A.S. Grabham P.W. Biochem. J. 1996; 313: 353-368Crossref PubMed Scopus (325) Google Scholar). The proteolytically active Arg-specific enzyme of about 39 kDa derives from its 72-kDa zymogen (prothrombin, factor II) via cleavage by the factor Xa-containing prothrombinase complex, while certain inhibitors can serve in its inactivation. The molecular mechanisms and the physiological consequences of thrombin activity have been intensively studied (3Grand R.J. Turnell A.S. Grabham P.W. Biochem. J. 1996; 313: 353-368Crossref PubMed Scopus (325) Google Scholar, 4Coughlin S.R. Nature. 2000; 407: 258-264Crossref PubMed Scopus (2103) Google Scholar). Nevertheless physiological substrates are not restricted to soluble proteins.Proteinase-activated receptors (PARs) 1The abbreviations used are: PAR, proteinase-activated receptor; BBB, blood-brain barrier; IL, interleukin; KC, mouse equivalent of GROα (CXCL1); MCP-1, monocyte chemoattractant protein 1 (CCL2); MIP-1α/-1β, macrophage inflammatory protein 1α/1β (CCL3/CCL4); PPACK, d-Phe-Pro-Arg-chloromethyl ketone; RANTES, regulated on activation, normal T-cell expressed and secreted (CCL5); TNFα, tumor necrosis factor α (TNFSF1A); CNS, central nervous system; DMEM, Dulbecco's modified Eagle's medium; FCS, fetal calf serum; LPS, lipopolysaccharide; Z, benzyloxycarbonyl; FPLC, fast protein liquid chromatography; MS, mass spectrometry.1The abbreviations used are: PAR, proteinase-activated receptor; BBB, blood-brain barrier; IL, interleukin; KC, mouse equivalent of GROα (CXCL1); MCP-1, monocyte chemoattractant protein 1 (CCL2); MIP-1α/-1β, macrophage inflammatory protein 1α/1β (CCL3/CCL4); PPACK, d-Phe-Pro-Arg-chloromethyl ketone; RANTES, regulated on activation, normal T-cell expressed and secreted (CCL5); TNFα, tumor necrosis factor α (TNFSF1A); CNS, central nervous system; DMEM, Dulbecco's modified Eagle's medium; FCS, fetal calf serum; LPS, lipopolysaccharide; Z, benzyloxycarbonyl; FPLC, fast protein liquid chromatography; MS, mass spectrometry. serve in the direct control of cellular functions by proteases, including thrombin (4Coughlin S.R. Nature. 2000; 407: 258-264Crossref PubMed Scopus (2103) Google Scholar, 5Wang H. Reiser G. Biol. Chem. 2003; 384: 193-202Crossref PubMed Scopus (112) Google Scholar, 6Cocks T.M. Moffatt J.D. Trends Pharmacol. Sci. 2000; 21: 103-108Abstract Full Text Full Text PDF PubMed Scopus (200) Google Scholar). By a very unusual mechanism of limited proteolysis, the N-terminal portion of a PAR can be cleaved off, resulting in the unmasking of a new N terminus, which serves as an intramolecular ligand. The autostimulation of PARs by their tethered ligand is coupled to cytosolic signaling events including G protein families, phospholipase C, or several protein kinases such as Src, p38, and p44/42 mitogen-activated protein kinase (1Gingrich M.B. Traynelis S.F. Trends Neurosci. 2000; 23: 399-407Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar, 3Grand R.J. Turnell A.S. Grabham P.W. Biochem. J. 1996; 313: 353-368Crossref PubMed Scopus (325) Google Scholar, 4Coughlin S.R. Nature. 2000; 407: 258-264Crossref PubMed Scopus (2103) Google Scholar, 5Wang H. Reiser G. Biol. Chem. 2003; 384: 193-202Crossref PubMed Scopus (112) Google Scholar, 7Suo Z. Wu M. Citron B.A. Gao C. Festoff B.W. J. Biol. Chem. 2003; 278: 31177-31183Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar, 8Kim K.Y. Kim M.Y. Choi H.S. Jin B.K. Kim S.U. Lee Y.B. Neuroreport. 2002; 13: 849-852Crossref PubMed Scopus (35) Google Scholar). Thrombin thereby affects the growth, adhesion, chemotaxis, and release functions of several cell types such as endothelial and epithelial cells, vascular smooth muscle cells, fibroblasts, platelets, granulocytes, lymphocytes, and macrophages/monocytes (1Gingrich M.B. Traynelis S.F. Trends Neurosci. 2000; 23: 399-407Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar, 4Coughlin S.R. Nature. 2000; 407: 258-264Crossref PubMed Scopus (2103) Google Scholar, 5Wang H. Reiser G. Biol. Chem. 2003; 384: 193-202Crossref PubMed Scopus (112) Google Scholar, 9Cromack D.T. Porras-Reyes B.H. Wee S.S. Glenn K.C. Purdy J.A. Carney D.H. Mustoe T.A. J. Surg. Res. 1992; 53: 117-122Abstract Full Text PDF PubMed Scopus (18) Google Scholar). Consequently thrombin is considered not only a clotting but a growth and wound-healing factor exhibiting a complex spectrum of restorative activities.On the other hand, thrombin and other serum proteases are suspected to cause severe damage to CNS tissue upon disruption of the blood-brain barrier (BBB) (1Gingrich M.B. Traynelis S.F. Trends Neurosci. 2000; 23: 399-407Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar, 7Suo Z. Wu M. Citron B.A. Gao C. Festoff B.W. J. Biol. Chem. 2003; 278: 31177-31183Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar, 10Carreno-Muller E. Herrera A.J. de Pablos R.M. Tomas-Camardiel M. Venero J.L. Cano J. Machado A. J. Neurochem. 2003; 84: 1201-1214Crossref PubMed Scopus (64) Google Scholar, 11Wagner K.R. Packard B.A. Hall C.L. Smulian A.G. Linke M.J. De Court Packard L.M. Hall N.C. Dev. Neurosci. 2002; 24: 154-160Crossref PubMed Scopus (71) Google Scholar, 12Friedmann I. Hauben E. Yoles E. Kardash L. Schwartz M. J. Neuroimmunol. 2001; 121: 12-21Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar, 13Xue M. Del Bigio M.R. Stroke. 2001; 32: 2164-2169Crossref PubMed Scopus (120) Google Scholar). The BBB and blood-cerebrospinal fluid barrier ensure an organized exchange of molecules and limit the passage of cells between neural and extraneural compartments. Most serum proteins are normally denied CNS entry (5Wang H. Reiser G. Biol. Chem. 2003; 384: 193-202Crossref PubMed Scopus (112) Google Scholar, 14van Rossum D. Hanisch U.K. Delgado-Garcia J.M. Herdegen T. Brain Damage and Repair: From Molecular Research to ClinicalTherapy. Kluwer Academic Publishers, Dordrecht, Netherlands2004: 181-202Crossref Google Scholar). However, when the BBB integrity is compromised due to trauma, stroke, viral and bacterial infection, autoimmune diseases, or neurodegenerative processes, blood content can more or less inundate the tissue (1Gingrich M.B. Traynelis S.F. Trends Neurosci. 2000; 23: 399-407Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar, 5Wang H. Reiser G. Biol. Chem. 2003; 384: 193-202Crossref PubMed Scopus (112) Google Scholar, 12Friedmann I. Hauben E. Yoles E. Kardash L. Schwartz M. J. Neuroimmunol. 2001; 121: 12-21Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar, 13Xue M. Del Bigio M.R. Stroke. 2001; 32: 2164-2169Crossref PubMed Scopus (120) Google Scholar, 15Lee K.R. Kawai N. Kim S. Sagher O. Hoff J.T. J. Neurosurg. 1997; 86: 272-278Crossref PubMed Scopus (340) Google Scholar). Proteases, protease inhibitors or their complexes, lipid-loaded albumin, complement factors, immunoglobulins, or cytokines may then gain access to neurons and glial cells (1Gingrich M.B. Traynelis S.F. Trends Neurosci. 2000; 23: 399-407Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar, 3Grand R.J. Turnell A.S. Grabham P.W. Biochem. J. 1996; 313: 353-368Crossref PubMed Scopus (325) Google Scholar, 5Wang H. Reiser G. Biol. Chem. 2003; 384: 193-202Crossref PubMed Scopus (112) Google Scholar, 14van Rossum D. Hanisch U.K. Delgado-Garcia J.M. Herdegen T. Brain Damage and Repair: From Molecular Research to ClinicalTherapy. Kluwer Academic Publishers, Dordrecht, Netherlands2004: 181-202Crossref Google Scholar, 16Wang H. Ubl J.J. Reiser G. Glia. 2002; 37: 53-63Crossref PubMed Scopus (163) Google Scholar, 17Ubl J.J. Sergeeva M. Reiser G. J. Physiol. 2000; 525: 319-330Crossref PubMed Scopus (35) Google Scholar, 18Striggow F. Riek-Burchardt M. Kiesel A. Schmidt W. Henrich-Noack P. Breder J. Krug M. Reymann K.G. Reiser G. Eur. J. Neurosci. 2001; 14: 595-608Crossref PubMed Google Scholar, 19Donovan F.M. Pike C.J. Cotman C.W. Cunningham D.D. J. Neurosci. 1997; 17: 5316-5326Crossref PubMed Google Scholar, 20Nishino A. Suzuki M. Ohtani H. Motohashi O. Umezawa K. Nagura H. Yoshimoto T. J. Neurotrauma. 1993; 10: 167-179Crossref PubMed Scopus (206) Google Scholar, 21Nakamura Y. Si Q.S. Takaku T. Kataoka K. J. Neurochem. 2000; 75: 2309-2315Crossref PubMed Scopus (10) Google Scholar, 22Si Q. Nakamura Y. Kataoka K. Exp. Neurol. 2000; 162: 89-97Crossref PubMed Scopus (23) Google Scholar). Blood-borne proteases seem to have a tremendous impact on the tissue. Brain cell injury associated with hemorrhage differs from pure ischemia, and the extent of CNS damage, for example upon stroke, is drastically reduced when certain protease activities are eliminated (13Xue M. Del Bigio M.R. Stroke. 2001; 32: 2164-2169Crossref PubMed Scopus (120) Google Scholar, 23Ohyama H. Hosomi N. Takahashi T. Mizushige K. Kohno M. Brain Res. 2001; 902: 264-271Crossref PubMed Scopus (43) Google Scholar, 24Vivien D. Buisson A. J. Cereb. Blood Flow Metab. 2000; 20: 755-764Crossref PubMed Scopus (82) Google Scholar).Microglial cells are major sensors and response elements in neuropathological scenarios of most heterogeneous etiology (25Hanisch U.K. Glia. 2002; 40: 140-155Crossref PubMed Scopus (1276) Google Scholar, 26Streit W.J. Glia. 2002; 40: 133-139Crossref PubMed Scopus (772) Google Scholar). Microglia represent a CNS-intrinsic population of macrophage-like cells safeguarding innate defense. Inducible synthesis of chemoattractive and immunoregulatory factors and the ability to present antigen help the recruitment of leukocytes and the engagement of adaptive immune responses. While support for neuronal functions and protection of tissue integrity are beneficial contributions, deregulated or dysfunctional microglia can be critical or at least instrumental in a harmful way. Excessive or chronic reactions fuel destructive cascades upon trauma or during inflammatory and neurodegenerative processes (25Hanisch U.K. Glia. 2002; 40: 140-155Crossref PubMed Scopus (1276) Google Scholar, 26Streit W.J. Glia. 2002; 40: 133-139Crossref PubMed Scopus (772) Google Scholar).Both the measures of defense as well as the detrimental consequences require the normally "resting" microglia to transform into alerted and finally reactive states. Challenges by foreign material or disturbances in the CNS homeostasis deliver the required signals (14van Rossum D. Hanisch U.K. Delgado-Garcia J.M. Herdegen T. Brain Damage and Repair: From Molecular Research to ClinicalTherapy. Kluwer Academic Publishers, Dordrecht, Netherlands2004: 181-202Crossref Google Scholar, 25Hanisch U.K. Glia. 2002; 40: 140-155Crossref PubMed Scopus (1276) Google Scholar). Molecularly these signals derive from bacterial cell walls and DNA, viral envelopes, or CNS endogenous factors normally not seen by microglia or usually not found at such concentration (14van Rossum D. Hanisch U.K. Delgado-Garcia J.M. Herdegen T. Brain Damage and Repair: From Molecular Research to ClinicalTherapy. Kluwer Academic Publishers, Dordrecht, Netherlands2004: 181-202Crossref Google Scholar, 25Hanisch U.K. Glia. 2002; 40: 140-155Crossref PubMed Scopus (1276) Google Scholar). However, only a few factors are identified thus far as to their (bio)chemical structure and mode of action.In principle, serum components could immediately inform microglia about disturbed BBB function or tissue injury (14van Rossum D. Hanisch U.K. Delgado-Garcia J.M. Herdegen T. Brain Damage and Repair: From Molecular Research to ClinicalTherapy. Kluwer Academic Publishers, Dordrecht, Netherlands2004: 181-202Crossref Google Scholar). Indeed several serum proteins fulfil accessory functions in the cell surface interaction of bacterial toxins or themselves carry some microglia-activating potential. We previously showed in vitro that microglia are, indeed, a target of thrombin and that these cells express PARs (27Balcaitis S. Xie Y. Weinstein J.R. Andersen H. Hanisch U.K. Ransom B.R. Möller T. Neuroreport. 2003; 14: 2373-2377Crossref PubMed Scopus (33) Google Scholar, 28Möller T. Hanisch U.K. Ransom B.R. J. Neurochem. 2000; 75: 1539-1547Crossref PubMed Scopus (158) Google Scholar). By a PAR-dependent mechanism, as indicated by hirudin sensitivity and kinetics of desensitization, thrombin stimulation resulted in intracellular calcium signals. Moreover the cells responded with proliferation and the release of cytokines, a major executive feature of activated microglia. Induction of cytokines by thrombin would agree with an assumed profile as a proinflammatory signal (5Wang H. Reiser G. Biol. Chem. 2003; 384: 193-202Crossref PubMed Scopus (112) Google Scholar, 6Cocks T.M. Moffatt J.D. Trends Pharmacol. Sci. 2000; 21: 103-108Abstract Full Text Full Text PDF PubMed Scopus (200) Google Scholar, 7Suo Z. Wu M. Citron B.A. Gao C. Festoff B.W. J. Biol. Chem. 2003; 278: 31177-31183Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar, 8Kim K.Y. Kim M.Y. Choi H.S. Jin B.K. Kim S.U. Lee Y.B. Neuroreport. 2002; 13: 849-852Crossref PubMed Scopus (35) Google Scholar, 10Carreno-Muller E. Herrera A.J. de Pablos R.M. Tomas-Camardiel M. Venero J.L. Cano J. Machado A. J. Neurochem. 2003; 84: 1201-1214Crossref PubMed Scopus (64) Google Scholar).However, studying the kinetics and pharmacological and biochemical features of microglial cyto- and chemokine induction by α- and γ-thrombin preparations, we obtained evidence against a "typical" recruitment of PARs. Further experiments rapidly led us to conclude that neither PAR activation nor proteolytic or ligand-like activities of thrombin proper were essential. In the present report, we provide evidence that the release-inducing capacity of thrombin preparations resided in a minor fraction of associated protein.EXPERIMENTAL PROCEDURES0Cell Culture Preparation and Treatment—Animals were kept and treated according to the Guidelines for Animal Care at the Max Delbrück Center for Molecular Medicine, Berlin, Germany. Primary microglial cultures were prepared from newborn mice (NMRI, purchased from Tierzucht Schönwalde GmbH, Schönwalde, Germany) and cultured in Dulbecco's modified Eagle's medium (DMEM) as described previously (29Häusler K.G. Prinz M. Nolte C. Weber J.R. Schumann R.R. Kettenmann H. Hanisch U.K. Eur. J. Neurosci. 2002; 16: 2113-2122Crossref PubMed Scopus (111) Google Scholar). After 10–14 days of primary cultivation, microglial cells were separated from other cell types by shaking and placed in 96- or 24-well plates or Petri dishes at densities of 104, 5 × 104, or 2 × 106 cells/cavity. After a 30-min attachment period, cells were extensively washed with DMEM containing 10% fetal calf serum (FCS) and kept in culture for 1–3 days before being used. Cultures routinely consisted of ∼98% microglial cells as determined by staining with Griffonia simplicifolia isolectin B4 (Vector Laboratories, Burlingame, CA). Experiments were either carried out in FCS-containing DMEM or under serum-free conditions using macrophage serum-free medium supplemented with astrocyte-conditioned medium (30Prinz M. Kann O. Draheim H.J. Schumann R.R. Kettenmann H. Weber J.R. Hanisch U.K. J. Neuropathol. Exp. Neurol. 1999; 58: 1078-1089Crossref PubMed Scopus (85) Google Scholar).Primary cultures were stimulated for up to 24 h by addition of thrombin preparations (bovine, mouse, and human thrombin (catalog nos. T-3399, T-4648, T-7513, T-8397, and T-6884); human prothrombin (catalog no. F-5132); and bovine factor Xa (catalog no. F-2027) from Sigma and bovine and human α-thrombin, human γ-thrombin, bovine prothrombin, and bovine factor Xa from Enzyme Research Laboratories, Swansea, UK) or lipopolysaccharide (LPS, Escherichia coli K-235, Sigma). Thrombic activity is given in NIH units (except for γ-thrombin). In experiments on heat-inactivated thrombin, concentrated solutions of the protease (bovine α-thrombin) were incubated at 70 or 100 °C for various periods of time or at given temperature for 10 min, diluted to the final concentration, and then used to stimulate cells. Stimulations with thrombin preparations were also carried out in the presence of recombinant hirudin HV1 (Hirudo medicinalis, Calbiochem Merck Biosciences) and HV2 variants (Sigma), Z-d-Phe-Pro-methoxypropylboroglycinepinanediol ester, or d-Phe-Pro-Arg-chloromethyl ketone (PPACK) (Calbiochem).Enzymatic Assay for Thrombin Activity—H-d-Phenylalanyl-L-pipecolyl-L-arginine-p-nitroaniline (2× HCl, 500 μm in 50 mm Tris/HCl, pH 7.0, 140 mm NaCl) was used as a chromogenic substrate (Chromogenix, Milano, Italy and Hemochrom Diagnostica, Essen, Germany). Absorbance at 405 nm was measured in a microplate reader (1420 Victor, Wallac Oy). All tests were run in triplicate using substrate solution as a blank. Inhibitors were mixed with thrombin 15 min before the assay. Note that γ-thrombin, although it cannot properly bind fibrinogen, is able to cleave small synthetic substrates as used in this assay.Cyto- and Chemokine Measurements—Following microglial stimulations, culture supernatants were collected and stored at –70 °C for cyto- and chemokine measurements. Interleukin-6 (IL-6), total IL-12 (collecting the IL-12 forms p75, p40, and p402), KC (mouse equivalent of growth-related oncogene GROα (CXCL1)), monocyte chemoattractant protein 1 (MCP-1 (CCL2)), macrophage inflammatory protein 1 α (MIP-1α (CCL3)), MIP-1β (CCL4), MIP-2, RANTES (regulated on activation, normal T cell expressed and secreted (CCL5)), tumor necrosis factor α (TNFα (TNFSF1A)), and soluble TNF receptor II were measured by enzyme-linked immunosorbent assay based on mouse- and factor-specific antibody pairs and standards or complete enzyme-linked immunosorbent assay kits following the instructions of the manufacturer (R&D Systems, Wiesbaden, Germany) (29Häusler K.G. Prinz M. Nolte C. Weber J.R. Schumann R.R. Kettenmann H. Hanisch U.K. Eur. J. Neurosci. 2002; 16: 2113-2122Crossref PubMed Scopus (111) Google Scholar, 31Hanisch U.K. Prinz M. Angstwurm K. Häusler K.G. Kann O. Kettenmann H. Weber J.R. Eur. J. Immunol. 2001; 31: 2104-2115Crossref PubMed Scopus (69) Google Scholar). The color reaction was analyzed in a microplate reader (SLT, Spectra, LabInstruments Deutschland GmbH, Crailsheim, Germany). Total protein was determined using the MicroBCA protein assay (Pierce).Viability Assays—Metabolic activity was assayed using WST-1 reagent (4-[3-(4-iodophenyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1,3-benzene disulfonate) based on the enzymatic cleavage of WST tetrazolium salt to formazan by the succinate-tetrazolium reductase system of the respiratory chain of intact mitochondria. The assay was performed according to the instructions of the manufacturer (Roche Diagnostics), and the color reaction was measured in a microplate reader (1420 Victor, Wallac Oy) at 540 nm wavelength (29Häusler K.G. Prinz M. Nolte C. Weber J.R. Schumann R.R. Kettenmann H. Hanisch U.K. Eur. J. Neurosci. 2002; 16: 2113-2122Crossref PubMed Scopus (111) Google Scholar, 31Hanisch U.K. Prinz M. Angstwurm K. Häusler K.G. Kann O. Kettenmann H. Weber J.R. Eur. J. Immunol. 2001; 31: 2104-2115Crossref PubMed Scopus (69) Google Scholar).Reverse Transcription-PCR/PCR—For detection of cytokine mRNA induction, microglial cells (2 × 106/Petri dish, 6-cm diameter, DMEM/FCS) were treated with LPS (100 ng/ml) or thrombin (10 units/ml) for 1 h. RNA was isolated using TRIzol reagent (Invitrogen). Briefly 2 ml of TRIzol reagent were used for the homogenization of the cells with the lysate pipetted several times to facilitate disruption. RNA was extracted by adding chloroform (0.2 volume eq), precipitated using 1 volume each of aqueous phase and isopropanol, and washed twice in 75% ethanol. Up to 5 μg of the isolated RNA were then used for cDNA synthesis. RNA was incubated with 1 μl of random primers (200–400 ng/μl) at 70 °C for 10 min. The mixture was then put on ice for 1 min. The reaction mixture consisting of 1 μl of 10 mm dNTP mixture, 4 μl of 5× Superscript reverse transcription buffer, 0.5 μl of RNasin (10 units), and 2 μl of 100 mm dithiothreitol (all purchased from Invitrogen) was added to the RNA/random primer mixture and left for 5 min at room temperature. The reaction was started by addition of 1 μl of Super-Script reverse transcriptase (200 units, Invitrogen). The final mixture was incubated for 50 min at 42 °C followed by 15 min at 70 °C to stop the reaction. To degrade the RNA of DNA-RNA hybrids, the mixture was treated with 1 μl of 2 units/μl E. coli RNase H (Invitrogen) at 37 °C for 20 min. PCR for cytokines was carried out with a CytoXPress Mouse Cytokine Sepsis Set 2 kit following the instructions of the manufacturer (BioSource, Camarillo, CA). PCR products were analyzed by 2% agarose gel electrophoresis.Microfiltration—Thrombin preparations were dissolved in 50 mm Na2HPO4/NaH2PO4, 150 mm NaCl, pH 7.0 (≤1 mg/ml). 500 μl of the solution were loaded on a Microcon YM-100 filter unit with a 100-kDa molecular mass cut-off (Millipore) and centrifuged at 10,000 × g until the complete volume had been passed through. The filtrate was stored, the filter unit was filled with 500 μl of buffer, shaken, and transferred onto a new test tube for another centrifugation (10,000 × g, 15 min). The washing cycle was repeated several times. At the end, the filter was placed upside down on a new tube, and the retentate was collected by centrifugation (1000 × g, 3 min). Aliquots of the original solution, the various filtrates, and the retentate were subsequently analyzed for enzymatic and release-inducing activity.SDS-PAGE—Thrombin preparations were analyzed by SDS-PAGE under non-reducing conditions using 12.5% gels and in-gel protein staining as described previously (31Hanisch U.K. Prinz M. Angstwurm K. Häusler K.G. Kann O. Kettenmann H. Weber J.R. Eur. J. Immunol. 2001; 31: 2104-2115Crossref PubMed Scopus (69) Google Scholar).FPLC—Thrombin preparations were separated by gel filtration on a Superose 12 HR 10/30 column using a FPLC system (Amersham Biosciences). Samples were prepared in elution buffer (50 mm Na2HPO4/NaH2PO4, 150 mm NaCl, pH 7.0), centrifuged, and filtered (0.2 μm). Aliquots were separated using elution buffer at a flow rate of 0.5 ml/min. Protein was detected by absorbance at 280 nm, and fractions were collected and stored for further analyses. A mixture of dextran blue (2000 kDa), bovine serum albumin (67 kDa), ovalbumin (43 kDa), cytochrome c (12.5 kDa), and aprotinin (6.5 kDa) was used for calibration. For studies on proteolytic stability, aliquots of the same sample were repeatedly separated under the same conditions.Mass Spectrometry—Microfiltration retentate from thrombin preparations was digested overnight at 37 °C in a solution of 25 mm ammonium hydrogen carbonate, pH 8, containing 0.1 μg/μl trypsin (Promega, Madison, WI). Resulting tryptic peptides were concentrated on a vacuum centrifuge and submitted to mass spectroscopy. The tryptic peptides were separated using the Ultimate/Famos capillary liquid chromatography system (LC Packings, Amsterdam, Netherlands). The sample was loaded onto a 300-μm-inner diameter × 1-mm C18 PepMap precolumn with a flow rate of 10 μl/min of 0.1% acetic acid using a Rheos 4000 liquid chromatography pump (Flux Instruments, Danderyd, Sweden). After 7 min of preconcentration and clean-up, the precolumn was automatically switched in-line with the 75-μm-inner diameter × 50-mm analytical column, and the peptides were separated with a gradient of 2–40% acetonitrile in 40 min (0.1% HCOOH) at a flow rate of 200 nl/min. The liquid chromatography system was connected to a mass spectrometer with a Protana platform (Protana, Odense, Denmark) using a 30-μm PicoTip from New Objective (Woburn, MA). Mass spectra were recorded using an LCQ quadrupole ion trap mass spectrometer (Thermoquest, San Jose, CA) using Triple-Play function. A first full scan mass spectrum was measured for a m/z 615–2000 range. A second scan was used to measure more precisely the molecular weight of the most abundant peptide signal in the first scan. A third scan was used to measure the collision-induced mass spectrometry (MS)/MS spectrum of the selected peptide. The spray needle was set to 2.4–3 kV in the positive ion mode. The inlet capillary temperature was 200 °C. Other source parameters and the spray position were optimized for a myoglobin tryptic digestion. The peptides were identified with Excalibur and Sequest programs (Thermoquest, San Jose, CA).RESULTSThrombin Preparations Induce Cyto- and Chemokine Release in Microglia—Incubation of mouse microglia with preparations of bovine, mouse, and human thrombin resulted in a dose-dependent release of multiple cytokines and chemokines (Fig. 1, A–E, and Table I). Bacterial LPS, a commonly used agent to activate macrophages/microglia and to mimic Gram-negative infection (29Häusler K.G. Prinz M. Nolte C. Weber J.R. Schumann R.R. Kettenmann H. Hanisch U.K. Eur. J. Neurosci. 2002; 16: 2113-2122Crossref PubMed Scopus (111) Google Scholar, 31Hanisch U.K. Prinz M. Angstwurm K. Häusler K.G. Kann O. Kettenmann H. Weber J.R. Eur. J. Immunol. 2001; 31: 2104-2115Crossref PubMed Scopus (69) Google Scholar), was used as a reference. Release induced by thrombin preparations was normalized to the LPS-inducible release and expressed as percentage. Absolute values are given in Table II. Exposure to thrombin did not result in impaired viability of the cells as determined in a WST-1 assay system (data not shown).Table ICyto- and chemokine-inducing activity of thrombin preparationsPreparationSupplier/product codeSpecific thrombin activityRelease induction activityunits/mgBovine α-thrombinSupplier A/product 195YesBovine α-thrombinSupplier A/product 240-165YesBovine α-thrombinSupplier A/product 32000NoBovine α-thrombinSupplier B/product 12031-2082NoMouse α-thrombinSupplier A/product 41000YesHuman α-thrombinSupplier A/product 52000YesHuman α-thrombinSupplier B/product 23100Yes/noaSome release induction was seen for certain but not for other cyto-/chemokines.Human γ-thrombinSupplier B/product 3Homogeneous by SDS-PAGEYesHuman prothrombinSupplier A/product 6NoBovine prothrombinSupplier B/product 4NoBovine factor XaSupplier A/product 7NoBovine factor XaSupplier B/product 5Yesa Some release induction was seen for certain but not for other cyto-/chemokines. Open table in a new tab Table IIRelease of microglial cyto- and chemokines as induced by thrombin preparationsCyto-/chemokineReleasepg/μgTNFα76.0 ± 5.4sTNF-R II122.7 ± 15.7IL-642.3 ± 7.4IL-12aIncluding I
Referência(s)