Portal de Periódicos da CAPES

Nitric Oxide Synthase in Toxic Epidermal Necrolysis and Stevens–Johnson Syndrome

2000; Elsevier BV; Volume: 114; Issue: 1 Linguagem: Inglês

10.1046/j.1523-1747.2000.00816.x

1523-1747

Lisa H. Lerner, Abrar A. Qureshi, Bhaskar V. Reddy, Ethan A. Lerner,

Urticaria and Related Conditions

Toxic epidermal necrolysis and Stevens–Johnson syndrome are severe cutaneous drug reactions of unknown mechanism. Nitric oxide can cause apoptosis and necrosis. The inducible form of nitric oxide synthase generates large amounts of nitric oxide and has been described in human skin. We propose that a large burst of nitric oxide in toxic epidermal necrolysis and Stevens–Johnson syndrome may cause the epidermal apoptosis and necrosis. Skin biopsies were taken from seven patients with actively progressing Stevens–Johnson syndrome or toxic epidermal necrolysis. Expression of inducible nitric oxide synthase was examined by reverse transcription–polymerase chain reaction and by immunoperoxidase staining for inducible nitric oxide synthase protein. Messenger RNA for inducible nitric oxide synthase was detected by reverse transcription–polymerase chain reaction and confirmed by the sequencing of polymerase chain reaction products. Strong staining for inducible nitric oxide synthase was observed in inflammatory cells in the lower epidermis and upper dermis. Diffuse, weaker staining was observed in keratinocytes. Expression of inducible nitric oxide synthase is consistent with the hypothesis that nitric oxide mediates the epidermal necrosis in toxic epidermal necrolysis and provides a potential target for therapeutic intervention. Toxic epidermal necrolysis and Stevens–Johnson syndrome are severe cutaneous drug reactions of unknown mechanism. Nitric oxide can cause apoptosis and necrosis. The inducible form of nitric oxide synthase generates large amounts of nitric oxide and has been described in human skin. We propose that a large burst of nitric oxide in toxic epidermal necrolysis and Stevens–Johnson syndrome may cause the epidermal apoptosis and necrosis. Skin biopsies were taken from seven patients with actively progressing Stevens–Johnson syndrome or toxic epidermal necrolysis. Expression of inducible nitric oxide synthase was examined by reverse transcription–polymerase chain reaction and by immunoperoxidase staining for inducible nitric oxide synthase protein. Messenger RNA for inducible nitric oxide synthase was detected by reverse transcription–polymerase chain reaction and confirmed by the sequencing of polymerase chain reaction products. Strong staining for inducible nitric oxide synthase was observed in inflammatory cells in the lower epidermis and upper dermis. Diffuse, weaker staining was observed in keratinocytes. Expression of inducible nitric oxide synthase is consistent with the hypothesis that nitric oxide mediates the epidermal necrosis in toxic epidermal necrolysis and provides a potential target for therapeutic intervention. inducible nitric oxide synthase nitric oxide synthase Stevens–Johnson syndrome Stevens Johnson syndrome/toxic epidermal necrolysis overlap toxic epidermal necrolysis Toxic epidermal necrolysis (TEN), Stevens–Johnson syndrome (SJS), and overlap syndromes (SJS,TEN,Bastuji-Garin et al., 1993Bastuji-Garin S. Rzany B. Stem R.S. Shear N.H. Naldi L. Roujeau J.C. Clinical classification of cases of toxic epidermal necrolysis, Stevens–Johnson syndrome, and erythema multiforme.Arch Dermatol. 1993; 129: 92-96Crossref PubMed Scopus (1312) Google Scholar) are severe blistering drug reactions of unknown mechanism with no effective treatment (Roujeau et al., 1990aRoujeau J.C. Chosidow O. Saiag P. Guillaume J.C. Toxic epidermal necrolysis (Lyell syndrome).J Am Acad Dermatol. 1990; 23: 1039-1058Abstract Full Text PDF PubMed Scopus (316) Google Scholar). Histologically, TEN and SJS are characterized by apoptosis and necrosis of keratinocytes in the epidermis (Paul et al., 1996Paul C. Wolkenstein P. Adie H. Wechsler J. Garchon H.J. Revuz J. Roujeau J.C. Apoptosis as a mechanism of keratinocyte death in toxic epidermal necrolysis.Br J Dermatol. 1996; 134: 710-714Crossref PubMed Scopus (266) Google Scholar) with separation of the epidermis from the underlying dermis, resulting in bullae. No animal models of TEN or SJS have been developed and the pathophysiology of these reactions is not known, although immune mechanisms (Villada et al., 1992Villada G. Roujeau J.C. Clerici T. Bourgault I. Revuz J. Immunopathology of toxic epidermal necrolysis. Keratinocytes, HLA-DR expression, Langerhans cells, and mononuclear cells: an immunopathologic study of five cases.Arch Dermatol. 1992; 128: 50-53Crossref PubMed Scopus (107) Google Scholar;Correia et al., 1993Correia O. Delgado L. Ramos J.P. Resende C. Torrinha J.A. Cutaneous T-cell recruitment in toxic epidermal necrolysis. Further evidence of CD8 + lymphocyte involvement.Arch Dermatol. 1993; 129: 466-468Crossref PubMed Scopus (191) Google Scholar) and altered metabolism of drugs (Friedmann et al., 1994Friedmann P.S. Strickland I. Pirmohamed M. Park B.K. Investigation of mechanisms in toxic epidermal necrolysis induced by carbamazepine.Arch Dermatol. 1994; 130: 598-604Crossref PubMed Scopus (140) Google Scholar;Wolkenstein et al., 1995Wolkenstein P. Charue D. Laurent P. Revuz J. Roujeau J.C. Bagot M. Metabolic predisposition to cutaneous adverse drug reactions. Role in toxic epidermal necrolysis caused by sulfonamides and anticonvulsants.Arch Dermatol. 1995; 131: 544-551Crossref PubMed Scopus (145) Google Scholar) have been postulated. We suggest that nitric oxide (NO) contributes to the pathogenesis of these conditions. NO is a small, diffusible molecule produced in most human tissues. This gas functions as a regulator of vascular tone, neurotransmitter, immune regulator, and cellular toxin for infectious organisms and tumors (Snyder and Bredt, 1992Snyder S.H. Bredt D.S. Biological roles of nitric oxide.Sci Am. 1992; 266 (68-71, 74-77)Crossref PubMed Scopus (551) Google Scholar;Vanin, 1998Vanin A.F. Biological role of nitric oxide: history, modern state, and perspectives for research.Biochemistry (Mosc). 1998; 63: 731-733PubMed Google Scholar). It can cause cell death either by necrosis or the induction of apoptosis (Bonfoco et al., 1995Bonfoco E. Krainc D. Ankarcrona M. Nicotera P. Lipton S.A. Apoptosis and necrosis: two distinct events induced, respectively, by mild and intense insults with N-methyl-D-aspartate or nitric oxide/superoxide in cortical cell cultures.Proc Natl Acad Sci USA. 1995; 92: 7162-7166Crossref PubMed Scopus (1860) Google Scholar;Brune et al., 1998Brune B. von Knethen A. Sandau K.B. Nitric oxide and its role in apoptosis.Eur J Pharmacol. 1998; 351: 261-272https://doi.org/10.1016/s0014-2999(98)00274-xCrossref PubMed Scopus (0) Google Scholar). Recently, it was shown that application of a topical NO releaser to human skin resulted in apoptotic and cytotoxic changes, apparently limited to the epidermis (Ormerod et al., 1999Ormerod A.D. Copeland P. Hay I. Husain A. Ewen S.W. The inflammatory and cytotoxic effects of a nitric oxide releasing cream on normal skin.J Invest Dermatol. 1999; 113: 392-397https://doi.org/10.1046/j.1523-1747.1999.00692.xAbstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar). NO is unstable, with a tissue half-life of seconds, during which time it can diffuse over 100 μm before reacting with other molecules such as hemoglobin, oxygen, lipids, or proteins (Lancaster, 1994Lancaster Jr, J.R. Simulation of the diffusion and reaction of endogenously produced nitric oxide.Proc Natl Acad Sci USA. 1994; 91: 8137-8141Crossref PubMed Scopus (623) Google Scholar;Beckman and Koppenol, 1996Beckman J.S. Koppenol W.H. Nitric oxide, superoxide, and peroxynitrite: the good, the bad, and ugly.Am J Physiol. 1996; 271: Cl424-37Google Scholar). In tissue, hemoglobin is thought to be the major scavenger for NO. The toxic effects of NO may be greatest in avascular areas such as the epidermis where NO reacts with molecules other than hemoglobin (Qureshi et al., 1996bQureshi A.A. Lerner L.H. Lerner E.A. From bedside to the bench and back. Nitric oxide and the cutis.Arch Dermatol. 1996; 132: 889-893Crossref PubMed Google Scholar). NO is produced by one of three forms of the enzyme NO synthase (NOS,Michel and Feron, 1997Michel T. Feron O. Nitric oxide synthases: which, where, how, and why?.J Clin Invest. 1997; 100: 2146-2152Crossref PubMed Scopus (846) Google Scholar). Neuronal and endothelial NOS are constitutively expressed and their activity is dependent on the presence of calcium. In contrast, activity of the inducible form of NOS (iNOS) is independent of calcium concentration and has the capacity to generate large amounts of NO (Nathan, 1997Nathan C. Inducible nitric oxide synthase: what difference does it make?.J Clin Invest. 1997; 100: 2417-2423Crossref PubMed Scopus (837) Google Scholar). All NOS isoforms have been described in human skin (Qureshi et al., 1996aQureshi A.A. Hosoi J. Xu S. Takashima A. Granstein R.D. Lerner E.A. Langerhans cells express inducible nitric oxide synthase and produce nitric oxide.J Invest Dermatol. 1996; 107: 815-821Crossref PubMed Scopus (76) Google Scholar;Bruch-Gerharz et al., 1998Bruch-Gerharz D. Ruzicka T. Kolb-Bachofen V. Nitric oxide in human skin: current status and future prospects.J Invest Dermatol. 1998; 110: 1-7Crossref PubMed Scopus (194) Google Scholar). We propose that NO produced by iNOS contributes to the apoptosis and necrosis observed in TEN and SJS. The action of hemoglobin as a scavenger of NO in the dermis could explain why the damage is limited to the epidermis. To test our hypothesis, we examined the skin of seven patients with actively progressing SJS or TEN for expression of iNOS mRNA and protein. Under a protocol approved by the institutional review board, skin biopsies were obtained from seven consecutive patients (see Table 1) with actively progressing reactions meeting the criteria for SJS, TEN, or the overlap syndrome SJS/TEN as previously defined (Bastuji-Garin et al., 1993Bastuji-Garin S. Rzany B. Stem R.S. Shear N.H. Naldi L. Roujeau J.C. Clinical classification of cases of toxic epidermal necrolysis, Stevens–Johnson syndrome, and erythema multiforme.Arch Dermatol. 1993; 129: 92-96Crossref PubMed Scopus (1312) Google Scholar). In all cases, medications were suspected of causing the eruption. Two patients later died of complications from TEN. Discarded material from skin surgery served as normal controls.Table 1Clinical information on study patientsPatient numberAge and sexCausative drugMajor comorbid diseases prior to SJS or TENOutcome182 WF? perioperative antibioticESRD s/p AVFbEnd stage renal disease for which an arteriovenous fistula was placed 2 wk prior to the onset of the skin eruption. Details of the procedure and medications administered were not available.Died of ischemic bowel complicating TEN263 WFPhenytoinMetastatic breast cancerSurvived acute TEN326 WFLamotrigineSeizure disorderSurvived455 WFBuprion (Zyban)noneSurvived, chronic ocular sequelae552 HFPhenytoinHydrocephalus VP shuntcVentriculoperitoneal shunt revision complicated by post-operative seizure treated with phenytoin. Patient had prior history of skin eruption to phenytoin.Survived652 WFAllopurinolCAD, CRFdCoronary artery disease and chronic renal failure.Died of myocardial infarction during TEN735 BFIbuprofennoneSurvived with severe liver failure, on transplant wait listaW, white; H, hispanic; B, black; F, female.b End stage renal disease for which an arteriovenous fistula was placed 2 wk prior to the onset of the skin eruption. Details of the procedure and medications administered were not available.c Ventriculoperitoneal shunt revision complicated by post-operative seizure treated with phenytoin. Patient had prior history of skin eruption to phenytoin.d Coronary artery disease and chronic renal failure. Open table in a new tab aW, white; H, hispanic; B, black; F, female. Thin shave biopsies of skin were immediately placed in embedding medium (TissueTek, Miles Scientific, Napperville, IL), frozen on dry ice and stored at -80°C until use. For immunoperoxidase studies, 5 μm sections were cut and stained the same day or stored at -80°C until use. For reverse transcription–polymerase chain reaction (reverse transcription–PCR), 10 μm sections were placed into 1.5 ml Eppendorf tubes and processed immediately or stored at -80°C until use. Frozen sections were air dried and fixed in acetone or methanol. Primary monoclonal antibodies against iNOS (Transduction Laboratories, Lexington, KY) and leukocyte common antigen (Dako Corporation, Carpinteria, CA) were diluted to a final concentration of 1–2 ng per ml in 10 mM phosphate-buffered saline with 5% heat-inactivated fetal bovine serum and incubated with the tissue sections overnight at 4°C. Monoclonal antibody keratin 903 (Enzo Diagnostics, Farmingdale, NY) directed against high molecular weight cytokeratin was supplied ready to use without further dilution. Mouse IgG2a isotype control antibody (Pharmingen, San Diego CA) was diluted to the same concentration as the iNOS antibodies. Tissue sections were incubated with primary antibodies overnight at 4°C and endogenous peroxidase was blocked with 0.3% peroxide. Biotinylated secondary antibody and avidin–biotin complex (Vector Laboratories, Burlingame, CA) were followed by application of 3-amino-9-ethylcarbazole substrate. Total RNA was obtained from cases 1–5 using a modified guanidinium technique according to the instructions provided with the Trizol reagent (Gibco BRL, Grand Island, NY) and messenger RNA was isolated using a Qiagen Oligotex mini mRNA kit (Qiagen, Santa Clarita, CA). Reverse transcription was performed for 1 h at 42°C in a 100 μl reaction containing 20 μg per ml random hexamers (Pharmacia, Piscataway, NJ), 0.6 mM each of dATP, dCTP, dGTP, and dTTP, 0.01 M dithiothreitol, 80 units of RNAse inhibitor (Boehringer Mannheim, Indianapolis, IN), 50 units of AMV reverse transcriptase (Boehringer Mannheim) and buffer provided with the reverse transcriptase. PCR was performed in a 100 μl reaction containing 5 μl of cDNA according to the protocol provided with the Taq polymerase (Qiagen), with the exceptions that 1 μM of each primer and 5 units of Taq polymerase were used. A 281 base pair fragment of iNOS cDNA spanning exons 9–11 was amplified using the 5′ primer TGGCCGTGACCC-TGAGCTCTTC and the 3′ primer GCTTGTGCGTTTCCAGGCCC-ATTC in 35 cycles each consisting of 45 s at 94°C, 60 s at 68°C, and 90 s at 72°C. The corresponding genomic DNA contains two introns measuring approximately 1800 and 90 bp (Xu et al., 1996Xu W. Charles I.G. Liu L. Moncada S. Emson P. Molecular cloning and structural organization of the human inducible nitric oxide synthase gene (NOS2).Biochem Biophys Res Commun. 1996; 219: 784-788https://doi.org/10.1006/bbrc.1996.0311Crossref PubMed Scopus (45) Google Scholar). A control reaction for amplification of the housekeeping gene glyceraldehyde phosphate dehydrogenase (GAPDH) was performed for each cDNA sample. A negative control reaction in which the target cDNA was omitted was run with every reaction. PCR products were separated by agarose gel electrophoresis and stained with ethidium bromide. Reverse transcription and PCR amplification for iNOS yielded products of the expected 281 bp size in the five cases of SJS and TEN that were tested (Figure 1). To confirm that the amplified band corresponded to iNOS, reaction products were sequenced and found to be identical with the predicted sequence for iNOS. A band of the expected size for GAPDH was amplified in control reactions for each case. Parallel negative control reactions lacking target cDNA were always performed and yielded negative results. Immunoperoxidase studies demonstrated marked expression of iNOS protein in single and clustered cells in the epidermis and papillary dermis (Figure 2a). Diffuse weaker epidermal staining for iNOS and staining of dermal vessels were also observed. Staining for iNOS was most prominent in the region of the lower epidermis in cases 2–7. In these cases, the biopsies also showed the most prominent apoptosis and necrosis in the lower epidermis (Figure 2c). In contrast to the other cases, case 1 showed full thickness epidermal staining for iNOS without accentuation in the lower epidermis (Figure 2d) and the diagnostic biopsy revealed full thickness necrolysis of epidermis (Figure 2e). A biopsy of perilesional skin was obtained from patient 4. This area became clinically involved within the subsequent 24 h. Staining for iNOS in the perilesional skin (Figure 2b) showed microscopic foci in which cells in the lower epidermis and superficial dermis were markedly positive, resembling the staining pattern seen in the lesional skin of the same patient. Stains for iNOS, leukocyte common antigen, and cytokeratin (a keratinocyte marker) in case 7 (Figure 2f–h) demonstrate that the strongest iNOS staining parallels the staining for leukocyte common antigen in both the epidermis and dermis. The pattern of epidermal iNOS staining in SJS/TEN differs from the staining patterns of normal control skin and psoriasis (data not pictured). Whereas SJS and TEN show strong staining of the lower epidermis, normal skin shows minimal epidermal staining and psoriasis shows moderate staining of the upper epidermis. In all cases, isotype-matched control antibodies yielded negative results.Figure 2Immunoperoxidase staining and histology of skin biopsies from SJS/TEN patients. Lesional skin from patient 4 (a) shows intense iNOS staining of single and clustered cells in the lower epidermis and upper dermis. There is also diffuse staining of the entire epidermis. In the perilesional skin of patient 4 (b), foci of marked staining (indicated by the box and enlarged in the inset) show a pattern similar to that of the lesional skin in this patient. The staining in (a) and (b) is from iNOS as no counterstain was applied. The routine diagnostic biopsy from patient 4 (c, hematoxylin and eosin stain), shows detached epidermis with clumped dyskeratotic and necrotic keratinocytes that are most prominent in the lower epidermis. Diffuse staining for iNOS is present throughout the epidermis and in dermal inflammatory cells within the biopsy from patient 1 (d). The standard diagnostic biopsy from patient 1 (e, hematoxylin and eosin) shows detached epidermis with full thickness necrosis. In the biopsy from patient 7 (f–h), stains for iNOS (f, isotype control in insert) and leukocyte common antigen (g) show a similar distribution of positive cells in contrast to the stain for keratin 903 (h).View Large Image Figure ViewerDownload (PPT) These data demonstrate that iNOS is expressed in the skin of patients with TEN and SJS. It appears that iNOS may be induced in leukocyte common antigen-positive inflammatory cells (Figure 2f, g) and in keratinocytes (Figure 2a, d). The finding that iNOS is expressed in microscopic foci of clinically uninvolved skin that subsequently blistered suggests that induction of iNOS occurs in the early stages of lesion formation and is not a late secondary event. NO is one of many molecules that participate in the apoptotic pathway and is known to affect caspases and members of the death receptor subfamily of tumor necrosis factor-α, specifically Fas (CD95) and its ligand Fasl (Melino et al., 1997Melino G. Bernassola F. Knight R.A. Corasaniti M.T. Nistico G. Finazzi-Agro A. S-nitrosylation regulates apoptosis.Nature. 1997; 388: 432-433Crossref PubMed Scopus (375) Google Scholar;Chlichlia et al., 1998Chlichlia K. Peter M.E. Rocha M. et al.Caspase activation is required for nitric oxide-mediated, CD95 (APO-l/Fas) -dependent and independent apoptosis in human neoplastic lymphoid cells.Blood. 1998; 91: 4311-4320Crossref PubMed Google Scholar). Fas and Fasl were recently reported to contribute to the apoptosis of TEN (Viard et al., 1998Viard I. Wehrli P. Bullani R. et al.Inhibition of toxic epidermal necrolysis by blockade of CD95 with human intravenous immunoglobulin.Science. 1998; 282: 490-493Crossref PubMed Scopus (987) Google Scholar). Another recent study demonstrated that in vivo application of a cream that releases NO has cytotoxic and apoptotic effects on human epidermis (Ormerod et al., 1999Ormerod A.D. Copeland P. Hay I. Husain A. Ewen S.W. The inflammatory and cytotoxic effects of a nitric oxide releasing cream on normal skin.J Invest Dermatol. 1999; 113: 392-397https://doi.org/10.1046/j.1523-1747.1999.00692.xAbstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar). Taken together, these studies and our data support the hypothesis that NO may mediate the epidermal apoptosis and necrosis that occurs in SJS and TEN. The steps between exposure to a drug and activation of iNOS and Fasl in TEN, however, remain unclear. Acute cutaneous graft-versus-host disease resembles SJS and TEN in that the epidermis undergoes apoptosis (Gilliam et al., 1996Gilliam A.C. Whitaker-Menezes D. Korngold R. Murphy G.F. Apoptosis is the predominant form of epithelial target cell injury in acute experimental graft-versus-host disease.J Invest Dermatol. 1996; 107: 377-383Crossref PubMed Scopus (77) Google Scholar) that can be indistinguishable from drug-induced TEN (Roujeau et al., 1990aRoujeau J.C. Chosidow O. Saiag P. Guillaume J.C. Toxic epidermal necrolysis (Lyell syndrome).J Am Acad Dermatol. 1990; 23: 1039-1058Abstract Full Text PDF PubMed Scopus (316) Google Scholar). Increased circulating levels of NO metabolites have been reported in human graft-versus-host disease (Weiss et al., 1995Weiss G. Schwaighofer H. Herold M. Nachbaur D. Wachter H. Niederwieser D. Wemer E.R. Nitric oxide formation as predictive parameter for acute graft-versus- host disease after human allogeneic bone marrow transplantation.Transplantation. 1995; 60: 1239-1244Crossref PubMed Scopus (53) Google Scholar) and this disease can be ameliorated by NOS inhibitors in animal models (Hoffman et al., 1997Hoffman R.A. Nussler N.C. Gleixner S.L. et al.Attenuation of lethal graft-versus-host disease by inhibition of nitric oxide synthase.Transplantation. 1997; 63: 91-94Crossref Scopus (22) Google Scholar). It is possible that NOS expression is altered in the skin of patients with graft-versus-host disease. Epidermal apoptosis and necrosis in SJS and TEN may result directly or indirectly from local expression of iNOS. Trials of NOS inhibitors in patients with these conditions may help to elucidate further the role of NOS and have the potential to lead to specific and effective treatment. This work was supported by an agreement between Massachusetts General Hospital and the Shiseido Co. Ltd, National Institutes of Health Grants ROI-AR42005-01 and ROI-AR44510-01, and a Pfizer Pharmaceuticals Research Grant from the Dermatology Foundation. EAL is an Established Investigator of the American Heart Association.