 |
|
|
|
|
|
 |
|
|
|
SEROLOGICAL STUDIES ON A HUMAN COHORT FROM AN AREA OF SOUTH INDIA ENDEMIC FOR JAPANESE ENCEPHALITIS VIRUS Courtesy : Festschrift - Dr. S. Ramachandran |
|
|
|
Introduction The presence of Japanese encephalitis virus (JEV) in India was first inferred through a serological study in 1951, which demonstrated the presence of neutralizing antibodies to the virus in humans (4) but the disease was only recognized in 1956 with JEV being isolated from mosquitoes and human brain specimens in 1958 in Vellore (5). Although a history of clinical exposure and certain clinical features may suggest JE, laboratory confirmation, usually by serological tests is absolutely essential. JE virus occasionally can be recovered from blood but only in the pre-neuroinvasive phase (up to 3 to 7 days after onset) but no peripheral viremia is detected in the encephalitic phase. Virus can be recovered from the cerebrospinal fluid (CSF) of encephalitic patients (6). Viral detection methods involve inoculation of the test samples into cells in culture and scoring for cytopathic effects or for viral antigen by immunofluorescence (IF) using antibodies to the E protein. The most easy and widely used diagnostic method is the IgM-capture enzyme-linked immunosorbent assay (MAC-ELISA) first introduced in 1982 (7). A modification of the original MAC-ELISA incorporating a color reaction is the most popular and currently used format (8, 9). Specific IgM can be detected in CSF, serum or both in approximately 75% of patients within the first 4 days after onset of illness, and nearly all patients are positive 7 days after onset (10). Both fluids are tested to maximize sensitivity. In approximately 30% of the cases, antigen-bearing infected cells can be identified in CSF by IF antibody before intrathecal IgM is detected, yielding a specific diagnosis within hours of a lumbar puncture. However, the procedure’s sensitivity (58%) was lower than that of MAC-ELISA testing of acute CSF and serum samples (84%). Human subjects:
The human volunteers chosen for the study consisted of 11 healthy contacts (HC) and 16 convalescent JE-patients (CP) from a JE-endemic area of South India. The E gene of JEV (nucleotides 978 to 2478 of the JEV genome) was obtained by PCR of the recombinant plasmid PM7 (a kind gift of Dr. Peter Mason; 12) using the following primer pairs:
Radioimmunoprecipitation and western blot analysis for detecting serum antibodies: Antibodies that recognize the native E protein of JEV were detected by RIP as described (13). Briefly, PS cells infected with the P20778 strain of JEV were metabolically labeled with [35S]-methionine (EXPRE35S35S, NEN Life Science Products Inc., Boston, MA, 1175.0 Ci/mmol) 24 h post infection, harvested by scraping 8 h later and washed twice with PBS. The cells were lysed by sonication in the presence of protease inhibitors in RIPA buffer (10 mM Tris-Cl, pH 7.4; containing 150 mM NaCl, 0.1% SDS, 1% DOC and 1% triton X-100), clarified by centrifugation and preadsorbed to protein A-sepharose beads (Amersham Pharmacia Biotech Asia Pacific Ltd., Buckinghamshire, England) coated with a pool of serum samples from 18 donors ascertained to be negative for antibodies to JEV. The lysate was then incubated with the test serum sample bound to protein A-sepharose beads overnight at 4oC in RIPA buffer (13) with constant mixing. The beads were washed, boiled in SDS-sample buffer and electrophoresed on a SDS-PAGE and the gel bearing the immunoprecipitated proteins was dried and exposed to an X-ray film. To study the potential of the test serum samples in reacting with the denatured E protein, purified inclusion body preparations from E.coli expressing recombinant E were electrophoresed on 10% SDS-polyacrylamide gels and blotted on to a nitrocellulose (NC) membrane. The NC-membrane was subsequently cut into strips and each strip, incubated with the test serum at a 1:100 dilution in blocking buffer (0.5% gelatin in 10 mM Tris, pH 8.0, 150 mM NaCl and 0.05% Tween-20). Colour was developed using protein-A-HRP (Amersham Pharmacia Biotech Asia Pacific Ltd., Buckinghamshire, England) and diaminobenzidine as substrate. Antibodies to the NS1 protein of JEV in the serum samples of donors were detected by RIP as described (13) with an additional western blot included to confirm the identity of the NS1 protein. The RIP samples boiled in SDS-sample buffer were electrophoresed on a SDS-PAGE as described above. Proteins were blotted onto a nitrocellulose membrane, which was incubated with the rabbit antiserum raised to recombinant NS1 protein at a dilution of 1:1500. Blots were developed using HRP-conjugated goat anti-rabbit IgG (Bangalore Genei Pvt. Ltd., Bangalore, Karnataka, India) and diaminobenzidine (Sigma, St. Louis, MO). The same blot was then exposed to an X-ray film to visualize all the proteins of JEV immuno-precipitated by the serum samples. The relative intensities of the immunoprecipitated viral protein bands were quantitated in arbitrary units using a Bioimage Analyzer (Fuji, Tokyo, Japan). The glycosylation status of the NS1 protein was studied by inhibiting this process by treating JEV-infected cells with tunicamycin. PS cells were infected with JE virus at a moi of 5.0. 10 hours post infection the monolayer was washed and overlayed with fresh warm RPMI medium containing tunicamycin at a final concentration of 5ěg/ml. After incubation for 1 h at 37oC the cells were pulsed with 200 ěCI of 35S-methionine and incubation continued for the next 12 hours. The cells were harvested and RIP was performed as above. Of the 22 representative samples shown in Figure 1, 6 (lanes 2, 3, 11, 14,15 and 17) were obtained from the control group and the remaining were obtained from acute phase encephalitis patients. In order to determine whether the conformation of the E protein had any bearing on its recognition by E-specific antibodies, the reactivity of serum samples from JE-exposed individuals to denatured (Fig.1a) and the native (Fig.1a) forms of the E protein was compared. Figure 1 Serum samples from patients with JE-encephalitis predominantly recognize the native form of the viral envelope protein. A. Western blot analysis of individual serum samples for reactivity to recombinant envelope protein expressed in E.coli. Each lane represents a different serum sample. B. Radioimmunoprecipitation of metabolically labeled JE-infected PS cell lysates using the same serum samples shown in A. Lane M used mouse anti-JE serum as positive control. Three of the six control samples were found to have absolutely no reactivity to the E protein. Of the 16 patient samples, 10 had stronger reactivity to the native form of the E protein (Fig.1b). In fact, 7 of these 10 sera could not even weakly detect the denatured protein (Fig.1a, lanes 1, 5, 6, 10, 20, 21 and 22). Of the remaining 6 samples, 3 had good reactivities to both forms of E (Fig.1a, lanes 4, 7 and 16). Only three sera had stronger reactivities to the denatured protein (lanes 12, 18 and 19), but could weakly immunoprecipitate the native form, and the bands were visible only after a prolonged exposure of the gel to an X-ray film. Serum reactivities to denatured E protein were not dependent on the glycosylation status since western blots of JE-infected C6/36 cell lysates probed with the same serum samples did not improve the signal strength (data not shown). Figure 2
Reactivity of serum samples from JE-exposed individuals to the NS1 protein. A. Radioimmunoprecipitation of 35S-methionine labeled PS cell lysates with sera from JE-exposed individuals. Data with representative serum samples from acute (PA) and convalescent phase (PC) JE patients, as well as healthy contacts (HC), is shown. Lane labeled M used mouse anti-JE serum as positive control. Fig. 2A shows the reactivity of sera from four healthy contacts, four acute phase JE patients and eight convalescent JE patients to the NS1 protein of JEV detected by radioimmunoprecipitation. While all convalescent patients scored positive, sera from healthy contacts had at best very weak reactivity to this viral protein. Acute phase patient sera had weak reactivity to NS1 barely detectable except on long exposure of the autoradiogram although all four samples had reactivity to the envelope protein. The identity of the lower band detected by RIP was confirmed to be NS1 by western blotting the radioimmunoprecipitated material with NS1-specific rabbit antiserum (Figure 2B). We have encountered a few acute phase samples with reactivity to NS1; however, their small proportion rules out the feasibility of using NS1-specific antibodies as a diagnostic tool. Synthesis of viral proteins in individuals suffering from the disease would be expected to be at a much higher level than that obtaining in infected healthy contacts who successfully ward off disease, perhaps explaining the significant levels of anti-NS1 serum titers in patients compared to healthy exposed individuals. The basis for late onset of these antibodies could be of some interest. The ability of two patient sera tested (P1 and P2) to react with NS1 was independent of its glycosylation status. When we inhibited glycosylation by treating JE-infected cells with tunicamycin, we found that the patient sera immuno- precipitated both glycosylated and non-glycosylated forms of the NS1 protein, only from JE-infected but not control cell lysates (Fig. 3). Figure 3 JE patient sera recognize the NS1 protein independent of its glycosylation status. JE-infected PS cells were treated with tunicamycin to inhibit glycosylation and metabolically labeled lysates were immunoprecipitated using two serum samples that recognized the NS1 protein, as described in the Methods section.
Schematic representation of the MAC-ELISA (A) currently used in JE-diagnosis compared to the proposed format (B) using recombinant envelope protein. differentially diagnose encephalitis due to JE from other etiologies. We explored the possibility of using serum in the place of cerebrospinal fluid (CSF) in diagnosis of encephalitis due to JE. Our results clearly showed that serum of acute phase encephalitis patients could replace CSF in JE diagnosis, thus helping to avoid the invasive lumbar puncture procedure. Although a sizeable number of convalescent JE patients also did have NS1-reactive serum antibodies, their absence in acute phase sera precluded the use of NS1 as a diagnostic antigen. The MAC-ELISA (Fig. 4A, left panel) which is the diagnostic test currently used for the detection of JE-specific antibodies in the CSF of suspected JE-patients (8, 9) suffers from several demerits. Besides being extremely time-consuming and cumbersome due to multiple steps, the major impediment in setting up this assay is the skill that is required to generate the antigen used. Generation of JE-infected mouse brain homogenates necessitates the maintenance of large colonies of suckling mice from where the JE-infected brains are sourced. The preparation of this antigen not only involves animal sacrifice, but also a laborious procedure of inactivating the infectious virus followed by acetone extraction of the brain tissue. This process is also unduly risky since it exposes the individuals involved in antigen preparation to very high titers of extremely pathogenic virus. Moreover, the antigen being a crude viral preparation, quite often elicits non-specific reactions from human sera (13). The non-availability of skilled personnel and the high attendant costs of antigen preparation have resulted in very few hospitals having the expertise to set up this assay in India, a country endemic for JEV. As a result, samples from suspected JE patients are stored for extended periods of time in hospitals that admit JE cases and sent to the testing units only on collection of a sufficiently large number of samples. More often than not, the results are procured after the patient has been treated, based on the symptoms exhibited, and discharged from the hospital on recovery. Hence, the availability of easily produced, cost-effective, purified recombinant antigen devoid of contaminating host proteins would be an immense advancement in the sero-diagnosis of JEV. Furthermore, in cases where the detection of anti-JE antibodies in the test sample is the prime objective without the necessity for antibody-subtype determination, a 3 step direct ELISA (Fig. 4B) with the recombinant antigen should be easily able to replace the existing 5-step format. |
|
|
|
Authors Corresponding address: Dr. Priti Kumar |
|
|
|
The views expressed in this article are of the author(s), and any clarifications can be obtained from the author(s). |
|
|
