Antibodies to Zika virus, Campylobacter jejuni and gangliosides in Guillain–Barre syndrome: A prospective single-center study from southern India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.241402
Source of Support: None, Conflict of Interest: None
Background: The objective of this article was to study the presence of antibodies against Zika virus (ZIKV), Campylobacter jejuni, and gangliosides in patients with Guillain–Barre syndrome (GBS).
Keywords: Anti-ganglioside antibodies, Campylobacter jejuni, Guillain–Barre syndrome, India, Zika virus
Guillain–Barre syndrome (GBS) is an immune-mediated polyradiculoneuropathy. Antibodies formed during a prior infection cross reacts against the gangliosides present on the peripheral nerve, causing demyelination and/or axonal damage. Prior infection by Campylobacter jejuni (C. jejuni) is the most common, especially in axonal variants of GBS. Recent reports from Pacific islands, South America, and Puerto Rico  have shown an association of GBS with Zika virus (ZIKV) infection. Indian studies on the role of C. jejuni are scarce and there have been no reports of ZIKV and GBS from India. Hence, this study was conducted to assess the presence of antibodies against ZIKV, C. jejuni and gangliosides and to determine whether or not these antibodies have any association with clinical outcome.
This was a prospective observational study, conducted from July 2015 to August 2016, at the Department of General Medicine, Jawaharlal Institute of Postgraduate Medical Education and Research (JIPMER), a tertiary care center in southern India. All consecutive patients (of age more than 12 years) diagnosed with GBS as per the Asbury Cornblath criteria  were included. These patients were followed up until their discharge and again at a 3-month follow-up visit.
Baseline demographic and clinical details [including nerve conduction study (NCS) and cerebrospinal fluid (CSF) results] were collected from patients after obtaining informed consent for participation in the study. The disability score using Hughes GBS disability scale  was calculated at admission. The disability score was again calculated at discharge and at 3 months of follow-up. A poor outcome was described as a disability score ≥3; no improvement at 3 months of follow-up; or, death during the study period. Nerve conduction studies (NCS) were used to classify GBS into various types – acute inflammatory demyelinating polyradiculoneuropathy (AIDP), acute motor axonal neuropathy (AMAN), and acute motor sensory axonal neuropathy (AMSAN), based on the electrophysiological criteria proposed by Hadden et al.
Serum samples were collected before the initiation of treatment, centrifuged for 20 min at 3000 rpm within 30 min of collection, and stored at −20°C to −80°C. The following tests were done – anti-C. jejuni IgG antibody, anti-ganglioside IgG antibody panel, and anti-Zika IgM antibody. Those samples which were positive for anti-Zika IgM were further tested for ZIKV polymerase chain reaction (PCR), dengue IgM antibody, and Japanese encephalitis IgM antibody.
Laboratory techniques in brief (details given in appendix)
Anti-C. jejuni (IgG) antibody was tested using Euroimmun (Euroimmun, Lubeck, Germany) anti-C. jejuni enzyme-linked immunosorbent assay [ELISA] (IgG) [antigen: strain – ATCC 33291]. Samples with IgG values ≥22 RU/mL were taken as positive, between 16 and 22 RU/mL as equivocal, and below 16 RU/mL as negative [Appendix e-2].
Anti-ZIKV antibodies were analyzed using MYBIO Kit – MBS109003, Qualitative Human ZIKV IgM (ZV-IgM) ELISA kit (MyBioSource, San Diego, CA, USA). Interpretation of results was based on the calculation of critical cut-off value, which is the average negative control value plus 0.15. Samples with optical density more than the critical value were taken as positive and those with less than the critical value were taken as negative [Appendix e-3].
IgM Capture ELISA (NIV, Pune, India) was used to detect antibodies to dengue serotypes 1–4 [Appendix e-4]and IgM Capture ELISA (MAC-ELISA; InBios) to detect antibodies to Japanese encephalitis [Appendix e-5].
Anti-ganglioside antibodies were tested using EUROLINE (Euroimmun) Anti-Ganglioside Profile 2 (IgG), a qualitative in vitro Western blot assay. The blot strips were incubated with diluted patient samples and visual evaluation was done by comparing the function control at the bottom of the strip [Appendix e-1].
Zika virus polymerase chain reaction [Appendix e-6]
RNA was extracted from serum samples using Qiagen viral RNA extraction kit (Cat no. 154041472) and performed according to the manufacturer's instructions. All the extracted RNA was stored at −20°C.
Reverse transcription (RT)-Zika polymerase chain reaction (PCR) was done for viral antigen NS-5 using forward primer – CCTTGGATTCTTGAACGAGGA and reverse primer – AGAGCTTCATTCTCCAGATCAA (NIV; as described by Balm et al.). The final reaction volume was 25 μL.
Complementary DNA (cDNA) conversion was done at 50°C for 30 min, with an initial denaturation at 95°C for 15 min, followed by 45 cycles of denaturation at 94°C, annealing at 50°C for 25 s, and extension at 72°C for 20°C. The amplified product was detected by electrophoresis on agarose gel (2%) with ethidium bromide, and visualized with gel documentation system (Bio-Rad's Gel Doc XR system, USA).
Amplification product of expected range of 192 bp was considered positive.
Standard protocol approvals, registrations, and patient consents were obtained.
This study was approved by Institute Ethics Committee. Written consent was obtained from all patients.
The independent variables analyzed were anti-ganglioside antibodies, anti-C. jejuni antibody, and anti-ZIKV positivity and type of GBS. The outcome variable was the presence or absence of a poor outcome. All the variables were analyzed by Fisher's exact test of statistical significance using the statistical package for the social sciences (SPSS) software, version 19.0 (IBM).
Of the 90 patients included in the study, 23 recovered without any therapy. Of the remaining 67 patients, 11 had a poor outcome (8 died and 3 had persistent weakness). Nearly a third of the patients were in the age group of 30–40 years, with 64.4% males and 35.6% females.
A history suggestive of prior infections was present in 54.4% of the patients, with a median interval of 1 week before the symptom onset. The most common was acute gastroenteritis (23 patients) followed by upper respiratory tract infection and fever.
Most of the study patients presented with motor symptoms (96.6%), which was predominantly quadriparesis followed by paraparesis. Mechanical ventilation was required in 13 patients and 8 of these patients died due to ventilator-associated pneumonia. The demographic details and clinical features are summarized in [Table 1].
Cerebrospinal fluid (CSF) analysis was done in 60 of 90 patients, of whom 31.7% showed albuminocytological dissociation. AIDP was the predominant type of GBS (56.7%), whereas the axonal forms (AMAN and AMSAN) constituted 38.9%. The rest had features of both axonal and demyelination patterns. The laboratory findings are summarized in [Table 2].
Anti-C. jejuni antibodies were present in 42 (46.6%) patients and there was no correlation with the NCS types [Table e-1] or the outcome [Table e-2].
In both AIDP and AMSAN, the most common anti-ganglioside antibody was GT1b, followed by GM-1, GD1a in both AIDP and in AMSAN, and GM-1 in AMAN [Table e-3].
The clinical and laboratory profile of patients with anti-ZIKV antibody can be seen in [Table 3]. Fourteen patients (15.5%) tested positive for anti-ZIKV IgM.
The history of upper respiratory infection was present in five patients and acute gastroenteritis in three patients. The presenting symptoms were quadriparesis in 11 patients and paraparesis in 3 patients. Only one patient had respiratory failure with requirement of mechanical ventilation.
The NCS types were AMAN (n = 6), AIDP (n = 6), and AMSAN (n = 2). Anti-ganglioside antibodies were positive in seven patients, with GM-1 and GM-2 being the most common.
Of these 14 patients, 8 were positive only for Zika antibody, whereas 4 were positive for anti-dengue IgM and 5 were positive for anti-C. jejuni IgG antibody. All three antibodies were present in three patients.
Serum ZIKV conventional PCR and anti-Japanese encephalitis (IgM) antibody were negative in all patients.
In our study, anti-C. jejuni antibodies were positive in 42 patients (46.6%). This is higher than the incidence found in previous studies from northern India (26%–35%).,, This may be due to the higher incidence of the axonal variant of GBS (38%) in our study, which has been shown to have a greater association with prior C. jejuni infection.
ZIKV has been implicated in microcephaly and GBS. Three large studies have suggested a possible link between Zika and GBS. The first study was from French Polynesia by Cao-Lormeau et al. They found anti-IgM Zika antibodies in 98% (41 of 42) patients. The second study was from Colombia by Parra et al. Urine/CSF reverse transcription (RT) PCR detected ZIKV in 17 of the 46 patients (40%). All these patients had serological evidence of recent flavivirus infection, as confirmed by serum and CSF IgG and IgM dengue antibodies. The third study was by Dirlikov et al. Zika PCR positivity in serum/CSF was present in 10 of 56 patients (18%), whereas antibodies to Zika alone were present in 16 (29%) patients. Antibodies to both Zika and dengue were present in 8 (14%) cases.
Serological evidence of ZIKV has been documented in India previously.,,, Our study is the first from India to study Zika and GBS. Anti-Zika IgM antibodies were positive in 14 (15.5%) patients. Eight patients had isolated Zika IgM antibody positivity (they were negative for both dengue and C. jejuni antibodies). Of these eight patients, preceding events were present in four patients with acute gastroenteritis in two patients and upper respiratory tract infection in two patients. There was no significant difference in the course and follow-up of these patients with respect to others except that the requirement of ventilatory support and death occurred in one patient due to ventilator-associated pneumonia. All other patients had improvement at discharge and at follow-up assessment.
Four of the 14 patients (28%) also had IgM dengue positivity. This may be due to cross-reaction among flaviviruses or evidence of recent dengue infection. The only method to differentiate between acute Zika and dengue infection is by plaque reduction neutralization technique. This was not done due to lack of laboratory resources. In a study by Cao-Lormeau et al., 8 of the 42 patients (19%) had positivity for both IgM Zika and IgM dengue antibodies, while in a study by Parra et al., 8 of the 17 PCR-positive patients had IgM dengue positivity.
Our study had significant differences with these studies. We looked at all patients with GBS irrespective of prior viral illness. At the time of the study, there were no reports of acute Zika infection from India. Only recently, there have been three reports of acute Zika infection (two from northern India and one from southern India). This is in contrast to the other studies which were done during an established outbreak of ZIKV.
We used anti-IgM antibodies for Zika. This is because in a study by Cao et al., 93% of patients had IgM positivity, whereas only two patients were IgG-positive but IgM-negative. We did not do urine RT PCR for confirming Zika infection because this information was available only late in the course of our study.
Testing for dengue (IgM) was done only in those patients who tested positive for Zika IgM, whereas previous studies have tested dengue antibodies (IgM and IgG) in all patients. Anti-C. jejuni antibodies were present in 50% of our series and in 5 of the 14 Zika-positive patients. This is contrast to 0% positivity for C. jejuni in a study by Cao-Lormeau et al.
Anti-ganglioside antibodies were present in 56 (62.2%) patients with the predominant type being GT-1b, followed by GM-1 and GD-1b. Of the 14 positive patients, 7 (50%) had positive antibodies with GM1 being the most common. Anti-ganglioside antibodies were present in 31% in a study by Cao et al., with anti-GA1 (19%) and GD1a (12%) being the most common, whereas GM1 had 0% positivity.
The tests were done on stored serum samples. PCR and antibodies were not evaluated in urine and CSF samples.
Our study has shown the high prevalence of anti-C. jejuni and anti-ganglioside antibodies. Evidence of recent Zika infection, as evidenced by anti-IgM antibodies, was present in 14 patients with 4 of them testing positive for anti-dengue IgM antibody. Whether this represents cross-reaction with dengue or prior/co-infection with dengue could not be addressed in this study. Although no significant correlation was seen between these antibodies and outcome in this study, future studies are required to analyze the effect of these associations on prediction of outcome in larger populations of GBS patients.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
Appendix e-1: Anti-ganglioside antibody analysis
Antigangioside antibodies were tested using EUROLINE (Euroimmun, Lubeck, Germany) Anti-Ganglioside-Profile 2(IgG), a qualitative in-vitro Western blot assay.
Principle: The test kit contains test strips coated with parallel lines of purified antigens. In the first reaction step, blot strips were incubated with diluted patient samples. In the case of positive samples, specific antibodies of the class IgG would bind to the antigens. To detect the bound antibodies, a second incubation with enzyme labelled antihuman IgG (enzyme conjugate) was done catalysing a colour reaction.
Procedure: The serum of the patients was diluted in working strength sample buffer in the ratio of 1:51. About 1.5 ml of this diluted sample was pipetted into the incubation channels in which the blot strips were arranged. The blot strips in the channels were incubated for a period of 120 min in a rocking shaker at room temperature. The fluid was aspirated and blot strips were washed 3 times with wash buffer. 1.5 ml of enzyme conjugate was added to the incubation channel and incubated in rocking shaker for 60 min. Again the channels were washed three times with wash buffer. 1.5ml of substrate solution was added to the incubation channel and kept in the rocking shaker for 10 min followed by aspiration and washing three times with distilled water.
Pictorial representation of arrangement of antigens on the membrane strips:
Intepretation of results: Visual evaluation was done following incubation of test strips in the patient serum by comparing the result obtained with the function control at the bottom of the strip.
Appendix e-2: Anti-Campylobacter jejuni antibody analysis:
Campylobacter jejuni antibodies were tested using Euroimmun (Euroimmun, Lubeck, Germany) anti-Campylobacter jejuni ELISA (IgG), a semiquantitative in vitro assay to test IgG antibodies in the serum.
Antigen: The strain “ ATCC 33291"of C. jejuni was utilized. The antigen mixture used contained mainly the 45 kD outer membrane protein.
Principle: The test kit contained microtitre strips, each with 8 breakoff reagent wells coated with Campylobacter jejuni antigens. In the first reaction, step diluted patient samples were incubated in the wells. In the case of positive samples, specific IgG antibodies would bind to the antigen. To detect the bound antibodies, a second incubation was carried out using an enzyme labelled antihuman IgG catalysing a colour reaction.
Procedure: Patient samples were diluted with sample buffer in the ratio of 1:101. 100 microlitres each of the calibrators, positive controls (PC), negative controls (NC) and the patient samples were transferred into the individual microtitre plates according to the pipetting protocol
The samples were incubated for 30 minutes at room temperature. The wells were emptied and subsequently washed 3 times using working strength washing buffer. 100 microlitres of enzyme conjugate (peroxidase labelled anti human IgG) was added into the microplate wells and incubated for 30 minutes at room temperature. After washing the wells, 100 microlitre of chromogen substrate solution was added into the wells and incubated for 15 minutes at room temperature. Stop solution was then added to the wells. Photometric measurement of colour density was made at 450 nm and was expressed as optical density values. A standard curve was plotted based on the extinction optical density (OD) values of calibrators 1, 2, 3 with fixed IgG RU (relative units) values of 200, 20 and 2, respectively.
Calibration: This was given in RU (relative units) as no international reference serum exists for antibodies against C. jejuni. For every group of tests performed, the extinction values of calibrators and RU determined for positive and negative controls must lie within the limits stated for the relevant test kit.
Interpretation: Optical density values of the samples were plotted in the above graph and their IgG RU values were obtained and interpreted as below:
<16 RU/ml: Negative
>/=16 to > 22 RU/ml: Borderline
>/=22 RU/ml: Positive
Appendix e-3: Anti-Zika virus antibody analysis:
Anti-Zika virus antibodies were analysed using MYBIO Kit- MBS109003, Qualitative Human Zika virus IgM (ZV-IgM) ELISA kit (MyBiosource San Diego (USA), which uses sandwich ELISA technique to qualitatively analyse Zika virus IgM antibodies in serum samples.
Procedure: All reagents and samples were brought to room temperature (18 – 25 degree centigrade) for 30 minutes before the start of the assay. The positive control, negative control and sample wells were set with PC 50 microlitre, NC 50 microlitre and diluted sample 50 microlitre in their wells, respectively. 100 microlitre of horseradish peroxidase (HRP) conjugate reagent was added to the PC, NC and sample wells, covered and incubated for 60 minutes at 37 degree centigrade. The microtitre plate was washed 4 times followed by addition of 50 microlitre of chromogen solution A and B into each well successively and incubated for 15 minutes at 37 degree centigrade. Then, 50 microlitre of stop solution was added to each well. The colour change from blue to yellow was noted. The OD (optical density) at 450nm using ELISA reader was obtained within 15 minutes of adding the stop solution.
Interpretation of results was based on the calculation of critical cut-off value which was the average negative control value, +0.15.
A positive result was when the sample OD > critical cut-off value
Negative result- sample OD < critical cut-off value
In the patients who were tested positive for anti Zika antibody, the following tests were performed
Appendix e-4: Dengue IgM serology analysis:
Dengue IgM capture ELISA was used for qualitative detection of IgM in serum samples of patients.
Principle: IgM antibodies in patient serum was captured by antihuman IgM coated on solid surface of wells. Dengue antigen was added, which binds to the captured human IgM. Unbound antigen was removed during the washing. In the subsequent step, biotinylated flavivirus anti DEN monoclonal antibodies were added followed by avidin HRP. Chromogen substrate was added and the reaction was stopped. Optical density was measured at 450 nm.
Procedure: Serum samples were diluted 1:100 times with the diluent. 50 microlitre of diluted samples were added to respective sample wells. 50 microlitre of dengue virus (DEN) IgM PC and NC were added to the respective wells. The microtitre plate was incubated at 37 degree centigrade for 1 hour followed by washing 5 times using wash buffer. 50 microlitre of DEN antigen was added to each of the wells followed by incubation and washing. Anti-DEN monoclonal antibodies were added followed by incubation. Avidin HRP followed by liquid 3,3', 5, 5'-tetramethylbenzidine (TMB) substrate was added to each well and incubated at 10 minutes in the dark. Optical density values were measured after termination of reaction.
Appendix e-5: Japanese encephalitis IgM serology analysis
This was done using Japanese encephalitis (JE) IgM antibody capture ELISA.
Principle: This is a 2-step sandwich-type immune assay. In this assay, JE negative control, JE positive control and serum samples were diluted with sample buffer and incubated in microtitre wells, which have been coated with antihuman IgM antibodies. This was followed by incubation with both JE virus derived recombinant antigen (JERA) and normal cell antigen (NCA) separately. After incubation and washing, the wells were treated with a JERA specific antibody labelled with HRP. After a third incubation and washing, the wells were incubated with TMB substrate. After stopping the reaction, optical density values were determined at 450 nm.
Procedure: Patient samples were diluted in the ratio of 1: 100 using sample buffer. 50 microlitre of diluted patient serum, JE NC and JE PC were added to the respective wells and incubated for one hour at 37 degree centigrade. After washing the wells, 50 microlitre of JERA were added to the wells and incubated for one hour. This was followed by addition of enzyme HRP conjugate and liquid TMB substrate. An incubation for one hour each was carried out successively with these agents. A stop solution was added to the wells, followed by optical density (value at 450 nm) analysis using a microplate reader.
Interpretation: The ratio of optical density values for JERA and NCA were calculated for both negative and positive controls, which is called as ISR (immune status ratio). A positive result was determined when the sample ISR was greater than 6; and a negative result was determined when the sample ISR was less than 4.
Appendix e-6: Zika virus PCR
Extraction: RNA was extracted from serum samples using Qiagen viral RNA extraction kit (Catalogue number: 154041472) and performed according to the manufacturer's instructions. All the extracted RNA were stored at -20 degree centigrade.
Amplification: Real time (RT)-Zika PCR was done for viral antigen NS-5 using a forward primer- CCTTGGATTCTTGAACGAGGA; and, a reverse primer-AGAGCTTCATTCTCCAGATCAA (National Institute of Virology, Pune- obtained from Balm et al.). The final reaction volume was 25 microlitre.
Cycling conditions: cDNA conversion was done at 50 degree centigrade for 30 min, initial denaturation at 95 degree centigrade for 15 min, followed by 45 cycles of denaturation at 94 degree centigrade, annealing at 50 degree centigrade for 25 sec, and extension at 72 degree centigrade for 20 sec. The amplified product was detected by electrophoresis on agarose gel (2%) with ethidium bromide and visualised with gel documentation system (Biorad gel doc XR system, USA).
Interpretation: The presence of an amplification product of the expected range of 192bp was considered as signifying a positive result.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]