| Severe acute respiratory syndrome (SARS) is an emerging infectious disease of significant public health concern in the 21th century. The causative agent of SARS was identified as a novel coronavirus (SARS-CoV). Since the symptoms of SARS are not specific and SARS cannot be clinically differentiated from other acute respiratory illnesses, specific diagnosis of SARS relies on laboratory tests. Accurate and timely diagnosis of SARS is one of the critical steps for control of this disease at the point of management. Reverse transcription polymerase chain reaction (RT-PCR) has been used as rapid diagnostic test in most of the research centers during the last epidemic. However, the RT-PCR relies heavily on experienced technicians and specialized laboratory equipment, and sometimes provides false-positive results due to contamination. Although the detection of antibodies using indirect immunofluorescence assay (IFA) or using the culture extracts of SARS-CoV based enzyme-linked immunosorbent assay (ELISA) was the gold standard for confirmation of SARS-CoV infection, the median time to seroconversion in SARS patients is 17-20 days after the onset of symptoms. Serological tests cannot be used for early diagnosis. Therefore, early SARS diagnosis remains a problem in medical centres lacking sophisticated equipment for non-research laboratories. One of the worst fears during the outbreak of SARS is the incorrect cohorting of non-SARS febrile pneumonic patients with those who are really suffering from SARS. This is likely to happen because of insufficient single rooms for isolation and nursing staff during a major outbreak. Thus it is essential that a rapid and accurate diagnostic test isavailable at the point of patient care. Such a test would be important for any decisions on the use of antiviral treatment for cases or antiviral prophylaxis for the contacts in the management of SARS or in the setting of a clinical trial. In the present study, our aim to develop a rapid, accurate early diagnostic test that easy to perform, and can be carried out in less-well-equipped laboratory settings.The availability of the complete genomic sequence of SARS-CoV has allowed to the development of diagnostic tests for SARS. SARS-CoV contains four major structural proteins, including the nucleocapsid (N), spike (S), envelope (E), and membrane (M) proteins. Like other coronaviruses, the nucleoprotein is well conserved, immunogenic, and abundantly expressed during infection. Therefore it is an ideal target for detection. In this study, we attempt to demonstrate that the nucleocapsid protein of SARS-CoV can be used as an early diagnostic marker for SARS. We established a nucleocapsid antigen capture assay based on monoclonal antibodies against the N protein of SARS-CoV. The N antigen capture assay was evaluated in serum specimens collected from a large cohort of patients with SARS during the acute and convalescent phase of disease, to further demonstrate that it can be used to directly detect N antigen in acute phase serum of SARS patients. Antigenic cross-reactivity between SARS-CoV and the known human coronaviruses 229E and OC43 were also analyzed in this study.I. Expression and antigenic characterization of recombinant SARS-CoV N proteinThe full-length of nucleocapsid gene from SARS-CoV (HKU-39849) was cloned into the prokaryotic expression vector pQE30 in frame and upstream of the six histidine residue series and pGEX-5X-3 in frame and upstream of the GST coding sequence, respectively. After optimization of the induction of protein expression, the two N fusion proteins (His-N and GST-N) were expressed with soluble form. Using SDS-PAGE analysis after purification, the His-N and GST-N fusion protein band corresponding to their expected molecular mass were 47kDa and 74 kDa, respectively. To determine the antigenicity of the N protein, convalescent phase sera from SARS patients were reacted with the two recombinant N fusion proteins by Western Blot.The immunoreactive protein bands of the His-N and GST-N fusion proteins were shown with the same expected sizes. These results indicated that the recombinant N fusion protein could react to specific antibodies in SARS patients' sera. The N protein can be as an antigen for SARS diagnosis.II. The antibody responses to nucleocapsid protein of SARS-CoV in patients with SARSAn indirect ELISA assay was established using the purified recombinant N protein to further investigate the immunoreactivity of the nucleocapsid protein. The temporal profile of the specific antibody against N protein of SARS-CoV in SARS patients were also analyzed in this ELISA assay. To establish the baseline for the tests, serum samples from 745 healthy blood donors were tested in the ELISA. The mean of ELISA values (OD450) for detection of IgG antibody to nucleocapsid protein was 0.09 with standard deviations of 0.08. Absorbance value of 0.35 was selected as the cutoff values (equal to the sum of the mean values from the healthy control and three times the standard deviations). A total of 500 serum samples, including 250 clinical documented, 188 clinical suspected, and 62 clinical excluded patients with SARS were analyzed by the N protein ELISA. The results of the N protein ELISA were compared with results from a commercially available ELISA kit based on lysates of SARS-CoV as antigen. Among the 250 serum specimens form clinical documented patients with SARS, IgG responses to the recombinant N protein were detected in 4 of 116 (3.45%) at first week, 42 of 68 (61.76%) at second week , 18 of 22( 81.82% ) at third week, and 44 of 44 (100%) from day 22 to 93 after the onset of symptoms, respectively. When comparing this result with an indirect ELISA kit coated with lysates of SARS-CoV, the positive rates of the IgG against SARS-CoV were 4.31% (5/116), 55.88% (38/68), 77.27% (17/22) and 100.00% (44/44) in the same four phases, respectively. In the 188 serum specimens from clinical suspected SARS patients, 3.19% (6/188) were IgG positive responses for N and 2.66% (5/188) for SARS-CoV, respectively. The IgG positive rates of the N and SARS-CoV were the same results of 6.45% (4/62) in the specimens from clinical excluded SARS patients. Comparing the sensitivities of these two assays, the McNemar T test showed that thedifference was insignificant between the two assay (P=0.180). The result of the indirect ELISA based N-protein showed a 98.20% (491/500) agreement with that of the ELISA assay based lysates of SARS-CoV. Of unmatched 9 serum samples (8 samples were positive for SARS-CoV IgG by IFA and Western blot), 2 samples were positive for SARS-CoV IgG and negative for N-IgG, and 7 samples were positive for N-IgG and negative for SARS-CoV IgG The results indicate that the ELISA based on the SARS-CoV N protein has high sensitivity. The recombinant N protein of SARS-CoV may be a good diagnostic marker for diagnosis of SARS.m. Production and characterization of monoclonal antibodies and rabbit hyperimmune serum against N protein of SARS-CoV.In this study, we attempt to develop a rapid and efficient method for producing monoclonal antibodies (mAbs) against N protein of SARS-CoV. Recombinant N protein was used to immunize mice for the production of hybridomas. BALB/c mice were immunized by injection into hind foot pad with N protein emulsified with complete Freund's adjuvant for the first injection and Freund's incomplete Freund's adjuvant for the following three booster injections. Popliteal lymph nodes isolated 15 days after food-pad immunization were used as the source of antibody producing hybridomas. Nine hybridomas that produced mAbs against the N protein of SARS-CoV were successfully selected by IFA, ELISA, and Western blot assay. Six mAbs were of IgGl isotype, one was IgG2a, and two were IgG2b. The range of highly affinity constants (Ka) of nine mAbs was from 8.3*10~8 M to 8.8><10"9M, which provided suitable for immunoassays. Competitive inhibition assay showed that the nine mAbs recognized at least five distinct epitopes of N antigen, which were inhibition by SARS-CoV antibodies positive serum from SARS patients. The results indicated the antigenic epitopes recognized by the nine mAbs were all in the native N antigen.Rabbits were initially immunized with subcutaneous injections of purified N protein emulsified in an equal volume of complete Freund adjuvant at multi-sites, followed by a series of intravenous injections of purified N protein without adjuvant or subcutaneous injections.The rabbit immune serum against N protein of SARS-CoVwas obtained with higher antibodies titer of 1 x 10"7.The specificity of mAbs and rabbit hyperimmune serum against N protein of SARS-CoV was further identified. Western blot assay showed nine mAbs and rabbit immune serum specifically bound to N protein of SARS-CoV. The IFA showed that nine mAbs and rabbit immune serum strongly reacted with SARS-CoV infected Vero E6 cells, with no recognition of the other human coronavirus and animal coronaviruses, except for one mAb either bound to virus-infected cells or normal cells. The result indicated that mAbs and rabbit immune serum to SARS-CoV N protein was specific for the SARS-CoV with no cross-reactivity with other coronaviruses. By immunohistochemistry mAbs detected SARS-CoV N antigen in formalin-fixed tissues from patients who died of SARS. The result indicated that mAbs can be able to recognize SARS-CoV N antigen in patients with SARS.These results showed hybridomas producing high specificity and affinity monoclonal antibodies to N protein of SARS-CoV which bind to distinct epitopes and thus select mAb pairs that can be used to develop a highly sensitivity of antigen capture assay. Production of these mAbs also provides a good tool for screening antigenic domain of nucleocapsid protein.IV. Development and optimization of two antigen-capture assays based on monoclonal antibodies against the N protein of SARS-CoVDetection of viral antigen has been successfully used in the diagnosis in most virus infectious diseases. It is therefore not unreasonable to develop an antigen capture assay for early diagnosis of SARS. In this study, we developed two antigen capture assays for detection of N protein of SARS-CoV, N antigen-capture ELISA and N antigen-capture chemiluminescent ELISA (CELISA). To establish optimal antigen capture assay, capturing and detecting antibodies were selected by sandwich ELISA paired the mAbs and polyclonal antibodies one by one. A batch of three mAbs that bind to unique epitope was used as antigen capture antibody and rabbit polyclonal antibodies were used for antigen detection antibody in this assay. Recombinant N protein was used as a standard to establish a detection sensitivity of approximated 50pg/ml for N antigen-capture ELISA and 20pg/ml for Nantigen-capture CELISA. The linear range of detection in N antigen-capture ELISA and N antigen-capture CELISA was from 50pg/ml to 1600 pg /ml and 02ng/ml to 500ng /ml, respectively. The result indicated that the assay can be developed a quantitative assay for analyzing level of N protein. The sensitivity of N antigen capture assay in analysis of produced N antigen in infected cell culture was equal to 6.3PFU/ml virus titer in cell culture. There was no cross-reaction with other human and animal coronaviruses in this assay. In conclusion, two sensitive and quantitative N antigen capture assays were established for the early diagnosis and disease monitoring of SARS-CoV infection.V. Applications of antigen capture assay for detection of SARS-CoV N protein in sera from SARS patients.Since SARS has clinical evidence of multi-system involvement and laboratory evidence of circulating viral RNA detectable by RT-PCR, it is therefore not unreasonable to develop a serum antigen capture assay for early diagnosis of SARS. Otherwise, antigen detection in serum by enzyme immunoassay would be an important advance over RT-PCR in most health care settings since collection of nasopharyngeal aspirates may be associated with some risk to health care workers.In this study, using serum specimens from 325 patients with serologically confirmed SARS-CoV infection, N protein in 100% of the samples (19 of 19) from neutralization test-positive patients and 92.9% of the samples (78/84) from serologically confirmed patients taken from days 1 to 5 was detected using the assay. At days 6 to 10, 100% of the samples (5 of 5) from neutralization test-positive patients and 69.8% of the samples (44 of 63) form serologically confirmed patients were still positive for the N protein, and the detection rate of the N protein decreased in later samples. In 400 sera from non-SARS patients collected 4 years prior to outbreak of SARS, only 1 was found to be weakly positive for N protein, which resulted in a test specificity of 99.8%. Evaluation of the two assays for detection of N protein was performed using sera from serologically confirmed SARS patients. The sensitivity and specificity of N protein-capture ELISA were shown to be equivalent to that of the N protein-capture CELISA.The profile of N protein detection with the appearance of antibodies was further investigated. The N protein in sera could be detected as early as day 1, and the higher positive rate (80%-100%) was observed between day 3 and day 8 after the onset of symptoms. The detection rate of the N protein decreased after day 10 and was never detected beyond day 19 after the onset of symptoms. Whereas, the N protein specific IgG and SARS-CoV specific IgG in sera tested negative at first week after the onset of symptoms and progressively increased from day 7 onwards. The high level of antibodies persisted after 3 weeks after the onset of symptoms.Comparison of serum N protein and respiratory tract viral RNA of SARS-CoV, 44 sets of serum and throat-wash specimens from 44 serologically confirmed SARS patients were tested for serum N protein and respiratory tract viral RNA to assess the diagnostic value of their detection to identify SARS-CoV infections at various stages of the disease. In the first 5 days after the onset of symptoms, 100% of serum specimens from 9 SARS patients were positive for N protein, this positive rate decreased to 63.6% (7/11), 18.6% (2/11), and 16.7 (1/6) at days 6-10, 11-15, and 16-20 after the onset of symptoms, respectively. Compared to viral RAN detection, in the first 5 days after the onset of symptoms, 88.9% (8/9) of throat-wash specimens in 9 SARS patients were positive for viral RNA, this positive rate decreased to 54.5% (6/11), 27.3% (3/11), and 16.7 (1/6) at days 6-10, 11-15, and 16-20 after the onset of symptoms, respectively. However, either N protein in the serum specimens or viral RNA in throat-wash specimens was not detected beyond day 21 after the onset of symptoms. Comparing the sensitivities of these two assays, of 44 SARS patients, 14 had both N protein and viral RNA detectable, 21 patients had both N protein and viral RNA not detectable, 5 patients had N protein detectable and viral RNA not detectable, and 4 patients had viral RNA detectable and N protein not detectable. The results demonstrated a relatively good correlation between the two assays, with 79.5% agreement. However, of the 20 SARS patients tested in the first 10 days of illness, 16 were positive for N protein in serum samples and 14 were positive for viral RNA in throat-wash specimens. Although there was no significant correlation between the rates of positivity for N protein in serum samples and viral RNA in the throat-washsamples, the rate of positivity (80%) for N protein was higher than that (70%) for viral RNA for the samples collected during the first 10 days after the onset of symptoms, indicating that N protein detection in serum specimens using the antigen capture ELISA is more useful than viral RNA using RT-PCR for early diagnosis of SARS.Furthermore, of 17 sera collected from four patients during the SARS recurrence in Guangzhou, 2003/2004, and 118 serum specimens collected from 115 contacts of the index four patients were tested using N antigen-capture ELISA and N antigen-capture CELISA, respectively. Five sera collected from 7 to 9 days after the onset of symptoms were positive for the N protein by the two assays. In addition, N antigen-capture CELISA seems more sensitive than the ELISA in this study, being able to detect the N antigens in serum samples collected on the 9 and 10 day of onset from the 2nd patient that was negative by the N antigen-capture-ELISA. This result was not surprising since enzyme-amplified chemiluminescent chemistry is generally more sensitive than the conventional ELISA. Of the 118 serum specimens from the 115 contacts, one sample collected on January 6, 2004 from a female contact of the second index patient was found to be positive for N antigen using N antigen-capture CELISA and the asymptomatic SARS-CoV infection was further confirmed by the serological tests.These results demonstrate that the high sensitivity of the capture assay in detecting SARS-CoV N protein in serum samples makes it a useful early laboratory diagnosis of SARS. This method overcomes several problems associated with the RT-PCR approach. For example, the problems due to contamination of RT-PCR products are not an issue in antigen detection assays. In addition, ELISA is rapid, easy to perform, and can be carried out in less-well-equipped laboratory settings. Widely employing this assay will enhance our ability to provide a rapid response in the event of return outbreak of SARS.VI. Antigenic cross-reactivity between SARS-CoV and human coronaviruses 229E and OC43Cross-reactivity between antibodies to different human coronaviruses (HCoVs)has not been systematically studied. Using Western blot analysis, IFA, and ELISA, antigenic cross-reactivity between SARS-CoV and the two HCoVs, 229E and OC43, in immunized animal and human sera was demonstrated. No cross-reactivity towards viral nucleocapsid proteins was found between SARS-CoV and the two known HCoVs when the rabbit immune sera were used. In 5 of 11 and 10 of 11 SARS patients, using IFA, paired sera showed a fourfold or greater antibody rise to 229E and OC43. None of these paired sera had an increase in antibody titer towards the OC43 nucleocapsid protein based-ELISA, indicating that major antigenic cross reactivity to OC43 is not related to the nucleocapsid protein. The paired sera from SARS patients against 229E showed a partial cross-reactivity. However, there was no significant correlation between IFA titer using whole virus-infected cells and ELISA ratios using nucleocapsid protein, indicating that any antigenic relationship between 229E and SARS-CoV is variable. Furthermore, antibodies against SARS-CoV could not be detected by using the IFA or a nucleocapsid protein-based Western blot analysis, in sera from healthy donors with antibodies positive to 229E and OC43. Overall, serum from convalescent SARS patients had a one-way cross reactivity to the two known HCoVs. Antigens of SARS-CoV and OC43 were more cross-reactive than those between SARS-CoV and 229E. These findings have important implications for the use of serological tests in epidemiology studies of coronaviruses infecting humans.Summarization:We analyzed the profile of nucleocapsid protein of SARS-CoV and antibody responses to the nucleocapsid protein in serum of SARS patients, first time put forward and confirm that the nucleocapsid protein of SARS-CoV in serum can be used as an early diagnostic marker for SARS, and established a simple assay for targeting SARS-CoV nucleocapsid protein, which can be used to directly detect virus antigen in acute phase serum of SARS patients. We have developed both nucleocapsid antigen capture detection kits based on the monoclonal antibodies, an enzyme-linked immunosorbent assay and an enhanced chemiluminescentimmunoassay, which have acquired new-drag certificate and production license issued by State Food and Drag Administration. Clinical studies have approved that the assys can detect nucleocapsid protein of SARS-CoV as early as day 1 after disease onset, the detection rate is above 90% between day 1 and day 5 after the onset of symptoms, and the approach resulted in a test specificity of 99%, demonstrating the assays with high sensitivity, specificity and stability for detection of serum nucleocapsid antigen. The asssy is easy to manipulate and economic, has been used in medical departments lacking sophisticated equipment as a screening reagents for acute respiratory infectious diseases. As a result of compatibility of specimen from different species, the assay will be useful in mass screening to tract the origins of SARS. In addition, we demonstrated that SARS patients may represent an anamnestic reaction to previous infections with the two known HCoVs 229E and OC43. The major antigenic cross reactivity towards two known HCoVs in the serum of convalescent SARS patients is not due to nucleocapsid proteins, but other viral components, which have important implications for the use of serological tests in epidemiology studies of coronavirases infecting humans. |
PDF Full Text Request