嚴重急性呼吸道症候群的免疫致病機轉（第二年）─（子計畫四）SARS病人抗上皮細胞及內皮細胞自體抗體致病角色之探討

行政院國家科學委員會專題研究計畫 成果報告

（子計畫四）SARS 病人抗上皮細胞及內皮細胞自體抗體致病

角色之探討

中 華 民 國 94 年 7 月 20 日

Autoantibodies against Human Epithelial Cells and Endothelial Cells

after Severe Acute Respiratory Syndrome (SARS)-Associated

Coronavirus Infection

Yao-Hsu Yang, Yu-Hui Huang, Ya-Hui Chuang, Chung-Min Peng, Li-Chieh

Wang, Yu-Tsan Lin, Bor-Luen Chiang*

Department of Pediatrics, National Taiwan University Hospital, College of

Medicine, National Taiwan University, Taipei, Taiwan.

Running title: Autoantibodies after SARS-CoV infection

Reprint requests and correspondence: Dr Bor-Luen Chiang, Department of

Pediatrics, National Taiwan University Hospital, No. 7 Chung-Shan South Road,

Taipei, Taiwan.

TEL: 886-2-2312-3456 Ext. 5127

FAX: 886-2-2397-2031

Abstract

The severe acute respiratory syndrome (SARS) is caused by infection with the

SARS-associated coronavirus (SARS-CoV) and characterized by severe pulmonary

inflammation and fibrosis. In this study, the development of autoantibodies against

human epithelial cells and endothelial cells in patients with SARS at different time

periods (the first week: phase I, one month after the disease onset: phase II/phase III)

were investigated. Antibodies in sera of patients and healthy controls against 1) A549

human pulmonary epithelial cell-line, 2) human umbilical venous endothelial cells

(HUVEC), 3) primary human pulmonary endothelial cells (HPEC) were detected by

cell-based ELISA and indirect immunofluorescence staining. The results revealed that

serum levels of IgG anti-A549 cells antibodies, IgG anti-HUVEC antibodies, and IgM

anti-HPEC antibodies were significantly higher in SARS patients at phase II/phase III

than those in healthy controls. Sera from SARS patients at phase II/phase III could

mediate complement dependent cytotoxicity against some A549 cells and HPEC. It is

concluded that some autoantibodies against human epithelial cells and endothelial

cells would be developed after SARS-CoV infection and this phenomenon may

indicate post-infectious cellular injury and also induce SARS-induced

Introduction

In early 2003, a new infectious disease, the severe acute respiratory syndrome (SARS)

swept the world including Taiwan [Tsang et al., 2003; Twu et al., 2003]. The pathogen

was later identified as SARS-associated coronavirus (SARS-CoV) and spread through

close contact of droplets [Drosten et al., 2003; Yeh et al., 2004]. Those who were

infected by this virus presented with persistent high fever, cough, dyspnea, and the

disease may eventually progress to respiratory and/or multiple organs failure [Fowler

et al., 2003; Tsang et al., 2003]. Autopsies of patients died from SARS have revealed

extensive pulmonary consolidation, localized hemorrhage and necrosis, proliferation

and desquamation of alveolar epithelial cells, monocytes, lymphocytes and plasma

cells infiltration in alveoli, and hyaline membrane formation [Ding et al., 2003;

Franks et al., 2003]. Systemic vasculitis was also found and characterized by edema,

thrombosis, localized fibrinoid necrosis, and infiltration of monocytes, lymphocytes,

and plasma cells into vessel walls in many organs including heart, lung, liver, kidney,

and the stroma of striated muscles [Ding et al., 2003]. All these pathological changes

are now thought to be mediated by direct viral destruction and followed by

immune-mediated processes [Peiris et al., 2003a].

Epithelial cells and endothelial cells, according to the pathological findings, may

Autoantibodies against epithelial cells have been detected in recurrent oral ulcer,

ulcerative colitis and prostate cancer [Ablin., 1972; Snook et al., 1991; Sun et al.,

2000], and anti-endothelial cell antibodies (AECA) have also been found in many

disorders such as systemic lupus eyrthematosus (SLE), Kawasaki disease,

Henoch-Schönlein purpura (HSP), Behcet's disease, and some post-infectious

immune-mediated diseases [Carvalho et al., 1999; Grunebaum et al., 2002; Lee et al.,

2003; Lin et al., 2003; Toyoda et al., 1999; Yang et al., 2002]. Although some of these

conditions appear as a result of inflammatory tissue injury, others have a pathogenic

potential to induce further damage. The aims of this study was to investigate the

development of autoantiboies against human epithelial cells and endothelial cells after

the SARS-coronavirus infection by using cell-based ELISA and indirect

Materials and Methods

Patients and controls

Twenty-two previously healthy Chinese adults suffering from SARS in early 2003

were included in this study. The diagnosis was confirmed by the typical clinical

presentations with fever, cough, and dyspnea, and positive viral PCR. Informed

consent and institutional approval were obtained for this study. Blood was sampled

during the first one-week (phase I) and one month after the disease onset (phase II or

phase III) [Peiris et al., 2003a]. Twenty healthy adults were enrolled as controls. In the

study by indirect immunofluorescence staining, patients with streptococcal

necrotizing pneumonia were also recruited as controls. For safety, serum samples

derived from patients were inactivated at 56℃ for 30 min before testing.

Antibodies against SARS-CoV nucleocapsid protein

To detect the presence of anti-SARS-CoV nucleocapsid (N) antibodies in SARS

patients, a 96-well microplate was coated with purified His-N protein at a

concentration of 5 µg/ml. Each well was then blocked by phosphate buffered saline

(PBS) containing 0.05﹪Tween-20 (PBS/Tween 20) (Sigma) and 5﹪bovine serum

albumin (BSA) at 37℃ for 2 hours. Diluted serum samples from SARS patients at

the wells at room temperature for 2 hours and then removed. Following the washing

procedure, peroxidase-conjugated mouse anti-human IgG, IgA and IgM (1: 5000 in 1 ﹪BSA) was added to each well and incubated at room temperature for 1 hour. The plates were washed by PBS/Tween 20 before adding teramethylbezidine (TMB) (KPL,

USA) substrate and the reactions were stopped by the addition of 2 N H2SO4. The

optical density (OD) of each well was read at a wavelength of 450nm minus 540 nm

by an ELISA reader. The serum levels of antibodies between patients and controls

were expressed as OD values.

Cells culture

A549 cells, a human pulmonary epithelial cell-line, were cultured with DMEM

supplemented with 10% heat-inactivated fetal calf serum (FCS), 2 mM L-glutamin,

150 mM HEPES, and 100 µg/ml penicillin/streptomycin. Primary human pulmonary

endothelial cells (HPEC) were cultured with EGM-2 MV (SingleQuots, USA)

supplemented with EBM (Cambrex Bio Science Walkersville, Inc. USA). Human

umbilical venous endothelial cells (HUVEC) were obtained from human umbilical

vein by collagenase (GIBCO BRL Life Technologies) digestion as described

previously [Jaffe et al., 1973]. The separated cells were seeded in 75 ml flasks

Technologies) supplemented with 15% heat inactivated FCS, heparin sulfate,

L-glutamine, endothelial cell growth factor (BM) (final concentration, 20 µg/mL), and

100 µg/ml penicillin/streptomycin. All cultures were incubated at 37℃ in 5% CO2,

and the cells were used between the 2nd and the 6th passage.

Cell-based ELISA

A549, HUVEC, and HPEC were prepared to detect autoantibodies in sera of SARS

patients. Cells were seeded on gelatin-coated 96-well microtitre plates (NuncTM,

Denmark) at a concentration of 1×105 cells/well. When the cellular growth became

confluent 3-4 days later, cells were fixed with 0.2% glutaraldehyde in PBS for 10 min

at room temperature and incubated with blocking buffer (1% BSA/0.05% azide/0.1 M

Tris in ddH2O) for 60 min at 37℃ to prevent non-specific binding. After washing

with PBS/Tween 20, the serum samples, diluted in blocking buffer at 1:200 for

IgG/IgM detection; 1:25 for IgA detection, were incubated for 2 h at 37℃. The sera

were then removed and the plates were washed, 100 µl of peroxidase-conjugated

rabbit antihuman IgG, IgM and IgA immunoglobulins were added to each well for a

further 2 h at 37℃. After washing, TMB solution was added for 15 min, and stop

solution (1M hydrochloric acid) for 5 min. The optical density of each well was read

immunofluorescence staining and ELISA had identified the patients with high

antibody binding activity to three cell types, and who were adopted as the positive

control (SLE serum for IgG and IgM detection; HSP serum for IgA detection). A

normal control serum with relative low binding activity was used as the negative

control. The results were expressed as ELISA ratio (ER) = 100×(S-A)/(B-A), where S

is absorbance of sample, A is absorbance of negative control and B is positive control.

Indirect immunofluorescence staining

A549, HPEC and HUVEC were prepared on 12-well Teflon-printed slides, fixed in

4% paraformaldehyde overnight at 4℃, and incubated with blocking buffer (5% fetal

calf serum in(PBS) for 30 min at 37℃. Cells on slides were then incubated with sera

of SARS patients, patients with necrotizing pneumonia, and healthy controls for 1 h at

37℃. The slides were washed three times by PBS and FITC-conjugated antihuman

immunoglobulins (CHEMICON, Australia), diluted in blocking buffer at 1:100, were

added to each well for a further 1 h at 37℃. The interactions of cells and PBS only

(without adding any serum) were as negative controls to establish backgrounds of

various immunofluorescence staining. The specimens were then washed three times,

Complement dependent cytotoxicity assay

Cells were seeded in 48-well culture plates at 1 × 104 cells/well overnight for cell

lysis assay. The culture medium was replaced by test medium (RPMI-1640

supplemented with 2 mM L-glutamine but without phenol red) before the addition of

patient sera. Patient and normal control sera were preheated at 56℃ for 30 min to

inactivate complement, diluted (1:25 dilution) and incubated with Low-Tox-M rabbit

complement (1:20 dilution; Celardane Laboratories Ltd., Hornby, Ontario, Canada) at

37℃ for 60 min before being added to the cells. After incubation for 48 hours, the

levels of lactate dehydrogenase activity in the culture supernatant were determined

using a Cytotoxicity Detection Kit (Boehringer Mannheim GmbH, Germany). The

absorbance of the sample was measured at 490 nm and the reference wavelength was

620 nm. To determine the cytotoxicity index, the absorbance values are substituted in

the following equation: cytotoxicity index (%) = (sample value-low control)/(high

control-low control) × 100%. Low control is the absorbance from the supernatant of

the cells cultured with test medium, and high control is the absorbance from the

supernatant of the cells cultured with 1% Triton X-100 in test medium.

Statistical analysis

SEM. Each two-group comparison was conducted using the Mann-Whitney U test. A

Results

Anti-N protein antibodies in SARS patients

Serum levels of antibodies against SARS-CoV nucleocapsid protein in SARS patients

at phase II/phase III and those in healthy controls were examined and compared. The

results showed that IgG and IgM anti-N protein antibodies elevated significantly in

SARS patients (IgG: 1.16 ± 0.10 vs 0.23 ± 0.04, p < 0.001, IgM: 0.84 ± 0.13 vs 0.46 ±

0.07, p = 0.03), and there was no statistical difference of IgA isotype between SARS

patients and healthy controls (0.69 ± 0.12 vs 0.42 ± 0.07, p = 0.08) (Fig 1).

Anti-epithelial cell antibodies (AEpCA) and anti-endothelial cell antibodies

(AECA) detection by cell-based ELISA

Figure 2 summarized the ELISA ratios of serum antibodies (IgG, IgA, IgM) against

A549 cells, HPEC, and HUVEC in healthy controls and SARS patients at different

time periods, phase I and phase II/phase III. During the first week (phase I), patients

presented with high fever, general malaise, myalgia and cough. The levels of these all

antibodies in this period were not different between patients and healthy controls.

When the disease progressed, patients received combined therapy of ribavirin,

intravenous immunoglobulin, and steroid. Although serum samples in this period were

IgM anti-HPEC antibodies, and IgG anti-HUVEC antibodies were significantly

increased when compared with healthy controls (IgG anti-A549 cells: 45.44 ± 6.26 vs

22.63 ± 4.57, p = 0.009; IgM anti-HPEC antibodies: 65.45 ± 7.38 vs 42.94 ± 6.67, p =

0.036; IgG anti-HUVEC antibodies: 27.12 ± 4.28 vs 13.47 ± 2.92, p = 0.025). The

levels of IgA anti-HPEC antibodies and IgA anti-HUVEC antibodies in patients, no

matter at phase I or phase II/phase III, were not different statistically from healthy

controls; however, these two antibodies were decreased significantly after the acute

phase (comparisons between different phases, IgA anti-HPEC antibodies: 25.04 ±

9.17 vs 7.94 ± 3.26, p = 0.018; IgA anti-HUVEC antibodies 25.13 ± 7.52 vs 11.75 ±

5.51, p = 0.018) (Fig 2(B), (C)).

Anti-epithelial cell antibodies (AEpCA) and anti-endothelial cell antibodies

(AECA) detection by indirect immunofluorescence staining

In the study of autoantibodies against A549 cells, Fig 3(A), (B), and (C) showed that

IgG anti-A549 cells, IgM anti-HPEC, and IgG anti-HUVEC antibodies existed in

SARS patients during phase II/phase III, but not in healthy controls and patients with

necrotizing pneumonia that also had severe pulmonary inflammation and damage.

In the present experiments, purified IgG and IgM immunoglobulins were not available

due to the limitations regarding blood sampling from SARS patients. Therefore, in

order to investigate if these autoantibodies have pathogenic effects, sera from SARS

patients with high-level autoantibodies were used for the cytotoxic assay. The results

showed that in the presence of complement, sera from patients at phase II/phase III

induced more A549 cells and HPEC lysis than sera from healthy controls (cytotoxicity

index %: A549 cells, 30.06 ± 3.54 vs 11.77 ± 2.65, p = 0.002; HPEC, 30.52 ± 4.67 vs

11.35 ± 3.19, p = 0.005) (Fig 4 (A), (B)). For HUVEC, cytotoxicity indexes between

patients and healthy control were not different significantly (17.8 ± 6.31 vs 10.55 ±

Discussion

SARS is a new emerging infectious disease with global impact. The diagnosis is

confirmed by the positive viral PCR, and patients were also found to have elevated

IgG and IgM antibodies against SARS-CoV nucleocapsid protein at later phases.

Although the pathogen has been identified, the underlying pathogenesis is yet to be

determined. A prospective study by Peiris et al (2003a) concluded that the clinical

progression of SARS had a tri-phasic pattern according to the clinical presentations

and pathological changes. Phase I (the first week), characterized by fever, myalgia

and other systemic symptoms was supposed to be the effect of viral rapid replication

and cytolysis. As the disease progressed into phase II and phase III, the rates of viral

shedding from nasopharynx, stood, and urine decreased gradually, however, severe

clinical worsening often occurred at this time [Peiris et al., 2003a; Poon et al., 2004].

In addition to pulmonary damage, some autopsies also revealed systemic vasculitis

[Ding et al., 2003; Lang et al., 2003]. Taken together with some therapeutic effects of

immunoglobulin and steroids to block disease progression [Chiang et al., 2003; Ho et

al., 2003], these findings suggest that the later phases of SARS are related to

immunopathological damage.

Focusing on the immune-mediated pathogenesis after SARS-CoV infection, It was

onset of the disease; including IgG anti-A549 cell antibodies, IgM anti-HPEC

antibodies, and IgG anti-HUVEC antibodies. There are many well-established

methods to detect antibodies against whole cells. The use of fluorescein-conjugated

antisera to human immunoglobulins (indirect immunofluorescence staining) is the

standard method with high specificity; however, this method is limited by low

sensitivity [Lindquist, Osterland., 1971; Praprotnik et al., 2001; Tan, Pearson., 1972].

Cell-based ELISA is now the method used most widely. In this assay, whole cells

from different sources are used as the substrate and fixed by glutaraldehyde treatment.

The procedure of cell fixation may lead to false-positive results, probably because that

autoantibodies reacting to intracellular antigens are also detected as well, therefore,

the specificity of this method is limited [Meroni et al., 1995; Praprotnik et al., 2001].

Each test described above has its own advantages and limitations. In order to confirm

the results of our study; we used these two methods to obtain and confirm the

laboratory data.

Previous studies of AEpCA were limited to some certain epithelial cells from

different tissue such as mucosal epithelial cells in pemphigus and recurrent oral ulcer,

and intestinal epithelial cells in ulcerative colitis [Colon et al., 2001; Snook et al.,

1991; Sun et al., 2000]. There is no literature concerning AEpCA detection in those

cells. A549 cells, an easily available and commonly used human respiratory epithelial

cell-line, were used as the substrate in this study. The results showed the development

of IgG anti-A549 cell antibodies in SARS patients but not in patients with

streptococcal necrotizing pneumonia, and this is the first report of the association

between autoantibodies development and infectious pulmonary disorders. In contrast

to AEpCA, anti-endothelial cell antibodies (AECA) are extensively studied. AECA

have been found in a wide range of diseases, especially in systemic autoimmune

diseases and primary autoimmune vasculitis [Carvalho et al., 1999; Grunebaum et al.,

2002; Lee et al., 2003; Lin et al., 2003; Praprotnik et al., 2001; Toyoda et al., 1999;

Yang et al., 2002]. Vasculopathy or vasculitis may develop after some viral infections

including hepatitis C virus (HCV) [Cacoub et al., 1999], cytomegalovirus (CMV)

[Toyoda et al., 1999], and dengue virus [Lin et al., 2003]. In these conditions, AECA

could also be detectable. SARS primarily affects lung, but vasculopathy/vasculitis of

other organs can also be found as the disease progresses [Ding et al., 2003; Lang et al.,

2003]. This phenomenon indicates that SARS-CoV like HCV, CMV, or dengue virus

may have the ability to damage vessels directly or indirectly, and this may be the

reason why those IgM anti-HPEC antibodies and IgG anti-HUVEC antibodies could

be detected in SARS patients. Another relevant finding in this study revealed that IgA

at phase I, decreased significantly when the disease progressed to phase II/phase III.

This phenomenon may be explained by the invasion of SARS-CoV that activates

mucosa-associated immune system, a specialized system for IgA globulins production,

and induces the formation of IgA AECA that decline gradually when the viral load is

decreased.

The mechanisms of these autoantibodies development in SARS patients were

speculated: after the contact of SARS-CoV, possibly through the epithelial cell surface

receptor recently identified as angiotensin-converting enzyme 2 [Li et al., 2003], the

virus invades into the epithelial cells, and that can be directly observed by the electro

microscope [Peiris et al., 2003b]. During the first week (phase I), SARS-CoV

replicates rapidly and induces cytolysis. At the same time, macrophages accumulate

around local inflammatory site; and these activated macrophages and other cells may

release tumor necrosis factor α, interleukin-1, and other proinflammatory cytokines

are increased after SARS-CoV infection [Beijing Group of National Research Project

for SARS., 2003; Ng et al., 2004]. The damaged cells and the stimulation by

proinflammatory cytokines may reveal some cryptic autoantigens. Macrophages

infiltrated around the lesion may play another role as the antigen presenting cells,

initiate the process of adaptive immunity, and lead to the formation of autoantibodies.

cells damage and increased proinflammatory cytokines did not have the same

phenomenon. Another possibility may be that SARS-CoV shares some specific

antigenic determinants with epithelial cells and endothelial cells individually. The

antibodies primarily against the virus then cross-react with these cells due to

molecular mimicry. Autoantibodies against cells like anti-endothelial cell antibodies

(AECA) are functionally heterogeneous, most probably depending on their specificity

[Bordron et al., 2001]. They may only be epiphenomenon of pulmonary epithelial and

vascular injury, but they could also have pathogenic roles to cause further cellular

damage by apoptosis, complement or antibody-dependent cytotoxic pathway

[Bordron et al., 2001; Lin et al., 2003; Worda et al., 2003]. In SARS patients, it was

found that those autoantibodies binding to epithelial cells and endothelial cells could

activate the complement system and induce some of these cells lysis.

In summary, although more studies should be designed and performed to identify

the disease-specific autoantigens, the presence of AEpCA and AECA after

SARS-CoV infection may represent the severe pulmonary injury and vascular damage

in these SARS patients. These autoantibodies also seem to have the potential to

damage some epithelial cells and endothelial cells, and these reactions provide

another immunological clue for a better understanding of the pathogenesis of SARS.

exclude the possibility of cross-reactions to these primary cells in the development of

Acknowledgements

This study was supported by a grant from the National Science Council, Republic of

China (NSC-92-2751-B-002-001-Y). We also thank CDC of Republic of China and

Prof. Chang SC, department of internal medicine, National Taiwan University

References

Ablin RJ. 1972. Antiepithelial antibodies in prostatic cancer. Arch Dermatol 105:759.

Beijing Group of National Research Project for SARS. 2003. Dynamic changes in

blood cytokine levels as clinical indicators in severe acute respiratory syndrome. Chin

Med J (Engl) 116:1283-1287.

Bordron A, Revelen R, D'Arbonneau F, Dueymes M, Renaudineau Y, Jamin C,

Youinou P. 2001. Functional heterogeneity of anti-endothelial cell antibodies. Clin

Exp Immunol 124:492-501.

Cacoub P, Ghillani P, Revelen R, Thibault V, Calvez V, Charlotte F, Musset L,

Youinou P, Piette JC. 1999. Anti-endothelial cell auto-antibodies in hepatitis C virus

mixed cryoglobulinemia. J Hepatol 31:598-603.

Carvalho D, Savage CO, Isenberg D, Pearson JD. 1999. IgG anti-endothelial cell

autoantibodies from patients with systemic lupus erythematosus or systemic vasculitis

stimulate the release of two endothelial cell-derived mediators, which enhance

adhesion molecule expression and leukocyte adhesion in an autocrine manner.

Arthritis Rheum 42:631-640.

Chiang CH, Chen HM, Shih JF, Su WJ, Perng RP. 2003. Management of

hospital-acquired severe acute respiratory syndrome with different disease spectrum. J

Colon JE, Bhol KC, Razzaque MS, Ahmed AR. 2001. In vitro organ culture model for

mucous membrane pemphigoid. Clin Immunol 98:229-234.

Ding Y, Wang H, Shen H, Li Z, Geng J, Han H, Cai J, Li X, Kang W, Weng D, Lu Y,

Wu D, He L, Yao K. 2003. The clinical pathology of severe acute respiratory

syndrome (SARS): a report from China. J Pathol 200:282-289.

Drosten C, Gunther S, Preiser W, van der Werf S, Brodt HR, Becker S, Rabenau H,

Panning M, Kolesnikova L, Fouchier RA, Berger A, Burguiere AM, Cinatl J,

Eickmann M, Escriou N, Grywna K, Kramme S, Manuguerra JC, Muller S, Rickerts V,

Sturmer M, Vieth S, Klenk HD, Osterhaus AD, Schmitz H, Doerr HW. 2003.

Identification of a novel coronavirus in patients with severe acute respiratory

syndrome. N Engl J Med 348:1967-1976.

Fowler RA, Lapinsky SE, Hallett D, Detsky AS, Sibbald WJ, Slutsky AS, Stewart TE;

Toronto SARS Critical Care Group. 2003. Critically ill patients with severe acute

respiratory syndrome. JAMA 290:367-373.

Franks TJ, Chong PY, Chui P, Galvin JR, Lourens RM, Reid AH, Selbs E, McEvoy

CP, Hayden CD, Fukuoka J, Taubenberger JK, Travis WD. 2003. Lung pathology of

severe acute respiratory syndrome (SARS): a study of 8 autopsy cases from Singapore.

Hum Pathol 34:743-748.

Shoenfeld Y. 2002. The role of anti-endothelial cell antibodies in Kawasaki disease -

in vitro and in vivo studies. Clin Exp Immunol 130:233-240.

Ho JC, Ooi GC, Mok TY, Chan JW, Hung I, Lam B, Wong PC, Li PC, Ho PL, Lam

WK, Ng CK, Ip MS, Lai KN, Chan-Yeung M, Tsang KW. 2003. High-dose pulse

versus nonpulse corticosteroid regimens in severe acute respiratory syndrome. Am J

Respir Crit Care Med 168:1449-1456.

Jaffe EA, Nachman RL, Becker CG, Minick RC. 1973. Culture of human endothelial

cells derived from umbilical veins. Identification by morphologic and immunologic

criteria. J Clin Invest 52:2745-2756.

Lang Z, Zhang L, Zhang S, Meng X, Li J, Song C, Sun L, Zhou Y. 2003. Pathological

study on severe acute respiratory syndrome. Chin Med J (Engl) 116:976-980.

Lee KH, Chung HS, Kim HS, Oh SH, Ha MK, Baik JH, Lee S, Bang D. 2003. Human

alpha-enolase from endothelial cells as a target antigen of anti-endothelial cell

antibody in Behcet's disease. Arthritis Rheum 48:2025-2035.

Li W, Moore MJ, Vasilieva N, Sui J, Wong SK, Berne MA, Somasundaran M,

Sullivan JL, Luzuriaga K, Greenough TC, Choe H, Farzan M. 2003.

Angiotensin-converting enzyme 2 is a functional receptor for the SARS coronavirus.

Nature 426:450-454.

Antibodies from dengue patient sera cross-react with endothelial cells and induce

damage. J Med Virol 69:82-90.

Lindquist KJ, Osterland CK. 1971. Human antibodies to vascular endothelium. Clin

Exp Immunol 9:753-760.

Meroni PL, Khamashta MA, Youinou P, Shoenfeld Y. Mosaic of anti-endothelial

antibodies. 1995. Review of the first international workshop on anti-endothelial

antibodies: clinical and pathological significance. Lupus 4:95-99.

Ng PC, Lam CW, Li AM, Wong CK, Cheng FW, Leung TF, Hon EK, Chan IH, Li CK,

Fung KS, Fok TF. 2004. Inflammatory cytokine profile in children with severe acute

respiratory syndrome. Pediatrics 113(1 Pt 1):e7-14.

Peiris JS, Chu CM, Cheng VC, Chan KS, Hung IF, Poon LL, Law KI, Tang BS, Hon

TY, Chan CS, Chan KH, Ng JS, Zheng BJ, Ng WL, Lai RW, Guan Y, Yuen KY;

HKU/UCH SARS Study Group. 2003a. Clinical progression and viral load in a

community outbreak of coronavirus-associated SARS pneumonia: a prospective study.

Lancet 361:1767-1772.

Peiris JS, Lai ST, Poon LL, Guan Y, Yam LY, Lim W, Nicholls J, Yee WK, Yan WW,

Cheung MT, Cheng VC, Chan KH, Tsang DN, Yung RW, Ng TK, Yuen KY; SARS

study group. 2003b. Coronavirus as a possible cause of severe acute respiratory

Poon LL, Chan KH, Wong OK, Cheung TK, Ng I, Zheng B, Seto WH, Yuen KY,

Guan Y, Peiris JS. 2004. Detection of SARS coronavirus in patients with severe acute

respiratory syndrome by conventional and real-time quantitative reverse

transcription-PCR assays. Clin Chem 50:67-72.

Praprotnik S, Blank M, Meroni PL, Rozman B, Eldor A, Shoenfeld Y. 2001.

Classification of anti-endothelial cell antibodies into antibodies against microvascular

and macrovascular endothelial cells: the pathogenic and diagnostic implications.

Arthritis Rheum 44:1484-1494.

Snook JA, Lowes JR, Wu KC, Priddle JD, Jewell DP. 1991. Serum and tissue

autoantibodies to colonic epithelium in ulcerative colitis. Gut 32:163-166.

Sun A, Hsieh RP, Liu BY, Wang JT, Leu JS, Wu YC, Chiang CP. 2000. Strong

association of antiepithelial cell antibodies with HLA-DR3 or DR7 phenotype in

patients with recurrent oral ulcers. J Formos Med Assoc 99:290-294.

Tan EM, Pearson CM. 1972. Rheumatic disease sera reactive with capillaries in the

mouse kidney. Arthritis Rheum 15:23-28.

Toyoda M, Petrosian A, Jordan SC. 1999. Immunological characterization of

anti-endothelial cell antibodies induced by cytomegalovirus infection.

Transplantation 68:1311-1318.

Yam LY, Cheung TM, Wong PC, Lam B, Ip MS, Chan J, Yuen KY, Lai KN. 2003. A

cluster of cases of severe acute respiratory syndrome in Hong Kong. N Engl J Med

48:1977-1985.

Twu SJ, Chen TJ, Chen CJ, Olsen SJ, Lee LT, Fisk T, Hsu KH, Chang SC, Chen KT,

Chiang IH, Wu YC, Wu JS, Dowell SF. 2003. Control measures for severe acute

respiratory syndrome (SARS) in Taiwan. Emerg Infect Dis 9:718-720.

Worda M, Sgonc R, Dietrich H, Niederegger H, Sundick RS, Gershwin ME, Wick G.

2003. In vivo analysis of the apoptosis-inducing effect of anti-endothelial cell

antibodies in systemic sclerosis by the chorionallantoic membrane assay. Arthritis

Rheum 4:2605-2614.

Yang YH, Wang SJ, Chuang YH, Lin YT, Chiang BL. 2002. The level of IgA

antibodies to human umbilical vein endothelial cells can be enhanced by TNF-alpha

treatment in children with Henoch-Schonlein purpura. Clin Exp Immunol

130:352-357.

Yeh SH, Wang HY, Tsai CY, Kao CL, Yang JY, Liu HW, Su IJ, Tsai SF, Chen DS,

Chen PJ, Chen DS, Lee YT, Teng CM, Yang PC, Ho HN, Chen PJ, Chang MF, Wang

JT, Chang SC, Kao CL, Wang WK, Hsiao CH, Hsueh PR. 2004. Characterization of

severe acute respiratory syndrome coronavirus genomes in Taiwan: Molecular

Legends

Figure 1. The comparisons of serum IgG, IgA, and IgM antibodies against SRAS-CoV

nucleocapsid protein between SARS patients at phase II/phase III and healthy controls.

The relative serum levels of these immunoglobulins were expressed as optical density

(OD) values. (* P < 0.05, ** P < 0.01)

Figure 2. Quantitative analysis of serum levels of IgG, IgA, and IgM autoantibodies

against (A) A549 cells, (B) human pulmonary endothelial cells (HPEC), and (C)

human umbilical venous endothelial cells (HUVEC) in healthy controls (…) and

SARS patients at phase I (hatched bars) and phase II/phase III („). The levels of

antibodies (ELISA ratios) are expressed as mean ± SEM. (* P < 0.05, ** P < 0.01)

Figure 3. Immunofluorescence analysis for the binding activities of (A) IgG

antibodies against A549 cells, (B) IgM antibodies against HPEC, and (C) IgG

antibodies against HUVEC in patients with streptococcal necrotizing pneumonia (2nd

row), healthy controls (3rd row), and SARS patients at phase II/phase III (4th row).

PBS was used in this test as a control to eliminate non-specific bindings (1st row).

at phase II/phase III with autoantibodies and from healthy controls (n = 9) were used

against (A) A549 cells, (B) HPEC, and (C) HUVEC. Low-Tox-M rabbit complement

(1:20 dilution) was incubated with pre-inactived sera, and the cytotoxic effects were

evaluated by the supernatant levels of lactate dehydrogenase. Data were calculated