Med Microbiol Immunol DOI 10.1007/s00430-015-0438-6

ORIGINAL INVESTIGATION

Development of an antibody capture ELISA using inactivated Ebola Zaire Makona virus Verena Krähling1,2 · Dirk Becker1,2 · Cornelius Rohde1 · Markus Eickmann1,2 · Yonca Erog˘lu1,2 · Astrid Herwig1 · Romy Kerber3 · Katharina Kowalski1 · Júlia Vergara‑Alert1,2 · Stephan Becker1,2 · the European Mobile Laboratory consortium

Received: 5 August 2015 / Accepted: 29 September 2015 © Springer-Verlag Berlin Heidelberg 2015

Abstract  The 2014 Zaire Ebola virus (ZEBOV) outbreak in West Africa represents an international public health concern. Highly sensitive and precise diagnostic tools are needed. In the present study, we developed a ZEBOV-specific enzyme-linked immunosorbent assay (ELISA) using inactivated ZEBOV isolate Makona from March 2014. Mock antigen was used to address nonspecific binding. Specificity, reproducibility and precision were determined to measure assay performance. The ZEBOV ELISA proved to be specific (96 %), reproducible and precise (Intra-assay CV 8 %, Inter-assay CV 18 %). Using the human monoclonal antibody KZ52, we showed that the ELISA was able to detect conformation-specific antibodies. Monitoring antibody development in 29 PCR-positive EBOV disease (EVD) patients revealed seroconversion in all cases. In addition, the ELISA was used to detect ZEBOV glycoprotein (GP)-specific antibodies in a vaccinated volunteer from day 14 until 5 years post-vaccination with a VSV-ZEBOV candidate vaccine. The results demonstrate the high reproducibility, specificity and sensitivity of this newly developed ELISA, which is suitable for the detection of specific antibody responses directed against different ZEBOV proteins in EVD patients and against the ZEBOV surface glycoprotein GP in vaccinated individuals.

* Stephan Becker [email protected]‑marburg.de 1

Institute of Virology, Philipps University Marburg, Hans‑Meerwein‑Str 2, 35043 Marburg, Germany

2

German Center for Infection Research (DZIF), Partner Site Giessen-Marburg-Langen, Hans‑Meerwein‑Str 2, 35043 Marburg, Germany

3

Bernhard Nocht Institute for Tropical Medicine, Bernhard‑Nocht‑Str. 74, 20359 Hamburg, Germany





Keywords  Ebola virus · ELISA · Clinical diagnostics · Ebola virus vaccination · Seroconversion

Introduction Ebola virus (EBOV), a member of the family of Filoviridae, is among the most virulent pathogens. This virus infects humans and nonhuman primates, causing severe hemorrhagic fever, with case fatality rates of up to 90 % in humans [1–3]. Five distinct EBOV species have been reported to date, including Zaire ebolavirus (ZEBOV), Sudan ebolavirus, Tai Forest ebolavirus, Reston ebolavirus, and Bundibugyo ebolavirus [3, 4]. In March 2014, the World Health Organization (WHO) was notified of an outbreak of ZEBOV in a remote area of Guinea. The outbreak then spread to the capital Conakry and to the neighboring countries of Liberia and Sierra Leone (WHO, 2015). To date, this outbreak is the largest Ebola virus disease (EVD) epidemic, with more than 27,000 cases, including 11,294 deaths, in West Africa (WHO, Ebola situation report—July 29, 2015). Currently, no drugs or vaccines are licensed for the treatment or prevention of EVD; however, potential candidates have been developed and successfully tested in preclinical models [5–7]. Furthermore, phase I—III clinical trials have been launched to analyze the safety and efficacy of a recombinant adenovirus and vesicular stomatitis virus (VSV) expressing the ZEBOV glycoprotein (GP) [8–12] as vaccine candidates. In addition to the use of rapid diagnostics to detect ZEBOV in suspected cases, serological tools are necessary to monitor the development of anti-ZEBOV antibodies in patients and to perform seroepidemiological studies that will help to describe the outbreak. Moreover,

13



such serological assays are necessary to monitor antibody responses in vaccinated volunteers to evaluate the successful development of an immune response against the vaccine. Many ELISA protocols have been published to date that are either based on viral antigens or recombinantly expressed filoviral proteins for the detection of IgG-specific or IgM-specific antibodies [13–19]. The present study describes the key features of a newly developed ELISA for ZEBOV Makona, a virus currently circulating in West Africa. This ELISA detects IgG antibodies against ZEBOV proteins. Human serum samples can show nonspecific reactivity with cellular proteins, and this obstacle was addressed by the use of mock antigen as a control. Using whole inactivated virions, we were able to develop a highly specific and robust test to detect antibodies elicited against all ZEBOV proteins that is suitable for monitoring the development of ZEBOV-specific antibodies in EVD patients in in-house emergency diagnostics [20]. Additionally, evaluation of specific immune responses against the ZEBOV GP in vaccinated volunteers was possible.

Med Microbiol Immunol

ultracentrifugation (2 h, 76,000×g at 4 °C). Pellets were resuspended in PBS and inactivated by addition of a final concentration of 1 % SDS, and boiling for 10 min at 99 °C before they were removed from BSL4 laboratory. Supernatants of ZEBOV-infected Vero E6 cells (five 175-cm2 cell culture flasks with 45 ml medium each) were concentrated and suspended in 1.4 ml of PBS. This amount was sufficient to coat approximately 450 ELISA plates. Protein concentration of viral antigen was 1.9 mg/ml determined by bicinchoninic acid (BCA) protein assay (Thermo Scientific, Prod# 23225) according to the manufacturer’s instructions. Purity of ZEBOV antigen preparation was analyzed by Coomassie Brilliant Blue staining (0.01 %). Mock antigen was prepared following the same procedure as for ZEBOV antigen from mock-infected Vero E6 cells after 6 days of incubation. Small volumes of antigen preparations were stored at −20 °C until use. In accordance with federal and state safety requirements, work with infectious ZEBOV was performed in the biosafety level 4 (BSL4) laboratory at the Philipps-Universität of Marburg. Inactivated virus (boiling for 10 min in 1 % SDS) was handled under BSL2 conditions. Human sera

Materials and methods Cell culture and antibodies Vero E6 cells and HuH7 cells were maintained in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10 % fetal calf serum (FCS), penicillin (50 units/ml), and streptomycin (50 µg/ml). A mouse monoclonal antiZEBOV GP antibody (3B11, [21]), a chicken anti-ZEBOV nucleoprotein (NP) antibody, produced as described in Pauly et al. [22], as well as a goat serum directed against ZEBOV were used to detect viral antigens. A human monoclonal anti-ZEBOV GP antibody (KZ52, dilution 1:200, [23]) was purchased from IBT Bioservices (Integrated Biotherapeutics Inc, North Potomac). Polyclonal secondary antibodies were purchased from Dako (Dako, Denmark) and used in a dilution of 1:1000 (ELISA) or 1:40,000 (Western blot). Antigen preparation To prepare antigen for the ELISA, EBOV subtype Zaire [ZEBOV], isolate Makona (passage 4 after isolation from the patient C07, GenBank accession number KJ660347 [24]) was propagated in Vero E6 cells. Supernatants of ZEBOV-infected Vero E6 cells (MOI 0.01) were collected 6 days post-infection (p.i.) and clarified from cell debris by low-speed centrifugation (10 min, 2500 rpm, 4 °C). Viral particles were subsequently concentrated and purified by

13

A highly positive human anti-ZEBOV IgG convalescent serum and multiple sera during acute phase of infection (patient 1, voluntary donor, University Hospital Frankfurt [20]) were analyzed and used as a positive control and as a set of calibrator solutions. A total of 113 European human serum samples from healthy anonymous donors were used to confirm specificity of the test. Serum from a healthy voluntary donor (Eberhard Karls University, Tübingen) was used as negative control. Serum from an EVD case (patient 2, voluntary donor, Bernhard Nocht Institute for Tropical Medicine (BNITM), Hamburg [25]) was analyzed. Serum samples of a VSVΔG ZEBOV GP vaccinated person [26] 14 and 31 days or more than 5 years post-vaccination were analyzed (BNITM, Hamburg). Plasma sample pairs of 27 Ebola virus-infected patients were obtained from the European Mobile Lab Consortium. One sample was collected during the acute phase of the disease (RT-PCR positive) and a second sample during convalescence. Each serum/ plasma was analyzed at a final dilution of 1:200 in the ZEBOV antigen ELISA. Informed consents to use serum and plasma for scientific research were obtained from EVD patients and the vaccinated person. ZEBOV antigen‑based ELISA High binding single-break strip microtiter plates (Greiner bio-one, Cat. No. 705073 and 705075) were coated with

Med Microbiol Immunol

50 µl ZEBOV or mock antigen (both diluted 1:1000 in PBS) and incubated for 16–20 h at 4 °C. Further incubations were performed at room temperature (RT). ELISA plates were washed three times with PBS/0.1 % Tween®20 (PBST) and then blocked for 45 min with PBS containing 5 % milk powder. Washing procedure was repeated three times with PBST. Human sera/plasma and controls were diluted 1:200 in PBST containing 1 % milk powder, and allowed to react with ZEBOV and mock antigen for 1 h. After washing plates three times with PBST, polyclonal HRP-coupled antibodies (DAKO) were used for detection (dilution 1:1000, 30 min of incubation). Following another round of washing (two times with PBST and two times with PBS), 100 µl of 3,3′,5,5′-tetramethylbenzidine (TMB) substrate solution (SureBlue™ TMB Microwell Peroxidase Substrate, KPL Inc., Maryland) was added to each well and allowed to react for 10 min protected from light. The reaction was stopped with 100 µl/well of TMB-Stop Solution (KPL Inc., Maryland), and the optical density (OD) was determined at 450–630 nm using an automated spectrophotometer (PHOmo, Autobio Labtec Instruments Co., Ltd.) within 5 min. Each control and serum were analyzed in duplicate, and mean OD value of each sample on mock antigen is subtracted from OD value on ZEBOV antigen to obtain corrected OD values. To calculate arbitrary ELISA units (AEU), the straight line equation of the standard curve on each plate is determined by linear regression analysis. Six standards were prepared which represented serial dilutions of the patient antiserum. The highest dilution (standard 6) was set to 1000 AEU, standard 5 to 2000 and further on, until standard 2 had 16,000 AEU. Then, units were calculated from corrected OD values of the respective sample by using the straight line equation. Positive samples had an AEU of 1000 and higher, and negative samples were set to 500 AEU. One example of this calculation is shown in Fig. 5. ELISA analyses of EVD patient sera were analyzed under BSL4 conditions. All other sera were tested under BSL2 conditions. Determination of key performance indicators of ZEBOV antigen‑based ELISA The coefficient of variation (CV) is the ratio of the standard deviation (σ) to the mean (µ): [CV = σ/µ] and indicates variability of the assay. Intra-assay variability was determined by testing one negative (NHS), one positive (PHS), and one marginal human serum (MHS) sample 12 times within the same assay. Inter-assay variability was determined by evaluating the same serum samples in consecutive assay runs, and by different operators. In both cases, the results were obtained by using the following formula:

[% CV = (σ/mean) × 100]. In each assay, an internal set of calibrators was used to validate the assay. Specificity of the assay was determined by using 113 human serum samples without EVD history. The specificity of the ELISA was calculated as 100 % minus the percentage of sera reacting positively with the ELISA. Molecular cloning ZEBOV Makona GP (nucleotides 6021-8088 of reference sequence KJ660347) was amplified from viral RNA by reverse transcription-PCR using specific oligonucleotides with integrated restriction sites. Resulting cDNA was digested, purified and ligated into pCAGGS vector prepared accordingly. Sequencing revealed one mismatch compared with published sequence at position 7691. Additionally, sequencing of PCR products from viral RNA showed that both the obtained sequence and the published sequence were present in the virus stock. Therefore, at position 7691 of the viral genome, an adenosine instead of a guanine was cloned into the ZEBOV Makona GP expression construct, resulting in an amino acid change in GP from aspartic acid to asparagine. Subsequent to the cloning of ZEBOV Makona GP, the mutagenesis of the plasmid was performed to insert an additional A into the transcriptional editing site of GP to express the full-length surface GP (GP8A) and not the soluble sGP. Native surface staining HuH7 cells (2 × 105 cells/6-well) were seeded on fibronectin-coated cover slips (1.5 µg/cm2, ⌀ 12 mm). Cells were transfected with 1 µg empty vector (mock) or 1 µg pCAGGS-ZEBOV Makona GP8A in DMEM supplemented with 3 % FCS, l-glutamine (2 mM) and penicillin and streptomycin (50 IU/ml) by using Trans-IT®-LT1 (Mirus Bio LLC) according to the manufacturer’s instructions. Twenty-four hours post-transfection cells were incubated with a goat anti-ZEBOV serum or a test serum diluted 1:50 in blocking buffer (as described by Kolesnikova et al. [27]). Then, cells were fixed with 4 % paraformaldehyde in DMEM for 20 min. Cells were treated 10 min with 100 mM glycine in PBS followed by 10 min incubation with blocking buffer. Secondary Alexa Fluor 488® antibody (1:400, Invitrogen) and DAPI (1 mg/ml, 1:2000) were used for detection. Between each step, cells were washed three times with PBS, and all steps were performed at 4 °C. Cover slips were fixed with Fluoprep (bioMerieux) containing 3 % DABCO (Sigma-Aldrich). Microscopic analysis was performed with a fluorescence microscope (Zeiss Axiophot, 63× objective).

13



Western blot analysis

Med Microbiol Immunol

Results

Makona-infected Vero E6 cells and characterized by Coomassie blue staining (Fig. 1a). The protein profile showed the presence of all viral proteins except for polymerase L, maybe because of its low abundance in viral particles. The final protein concentration of the ZEBOV antigen, as determined by BCA protein assay, was 1.9 mg/ml (data not shown). Simultaneously prepared mock antigen from the supernatant of noninfected cells revealed the presence of residual amounts of bovine serum albumin (Fig. 1a, BSA). Serial dilutions of the ZEBOV antigen were used for coating the ELISA plates to determine the necessary amount to receive optimal signal strength. All antigen dilutions showed OD values higher than 2.5, and we selected a 1:1000 dilution for the development of the ELISA (Fig. 1b). Thus, the protein concentration of ZEBOV antigen used for coating was approximately 2 µg/ml. The stability of the antigen was demonstrated by a comparison of freshly prepared antigen versus antigen that underwent 10 cycles of freeze/thawing. The OD values obtained using the two preparations were not significantly different (Fig. 1c).

Optimization and standardization of viral antigen

Key performance indicators of ZEBOV antigen ELISA

To establish a ZEBOV ELISA based on whole virions, antigen was prepared from the supernatant of ZEBOV strain

To monitor plate-to-plate consistency, an internal set of calibrators was established for use in each assay. We

Five microliters of ZEBOV Makona antigen, the same as used to coat ELISA plates, or 20 µg recombinantly expressed EBOV rGPΔTM (IBT Bioservices, Catalog# 0501-016) was separated on 10 or 8 % denaturing, preparative SDS-PAGE and transferred onto nitrocellulose membranes. Membranes were cut into small strips to allow detection of viral proteins by a small volume of human sera. Immunostaining was performed with 1:100 dilutions of primary antibody or human sera in PBST containing 1 % (w/v) milk powder. Western blot detection was performed with HRP-conjugated secondary antibodies using the Image Lab™ software and the ChemiDoc™ XRS+ Systems (BIO-RAD) for quantification. Western blot analysis of EVD patient sera was performed under BSL4 conditions.

Fig. 1  Characterization of ZEBOV antigen. a Antigen preparations (mock and ZEBOV) were analyzed by 10 % SDS-PAGE with Coomassie staining and Western blotting (WB) using a goat anti-ZEBOV serum and an anti-goat HRPcoupled secondary antibody for detection. ZEBOV proteins and bovine serum albumin (BSA) are indicated on the right. b Different antigen dilutions were used to coat ELISA plates. To detect viral antigen, antiZEBOV goat serum (1:1000) and a corresponding anti-goatHRP antibody were used. Corrected OD values are shown. c The coating of the ELISA plates was performed using ZEBOV antigen that was thawed one or ten times. Anti-ZEBOV goat serum (1:10,000), a monoclonal anti-GP antibody (3B11, 1:1000), one negative and two different positive human serum samples (NHS, PHS 1 and 2) were used in combination with respective HRP-coupled secondary antibodies (1:1000) for detection. Corrected OD values are shown

13

Med Microbiol Immunol

Fig. 2  Standard curve of ZEBOV antigen ELISA. a Calibrator solutions (S1–S6) were prepared by twofold serial dilutions of a human convalescent serum starting at a dilution of 1:8. As the other samples, the calibrator solutions were further diluted (1:200) for ELISA. Detection was performed using an anti-human HRP-coupled secondary antibody (1:1000). ELISA was repeated six times by different operators on consecutive days, and the standard deviation of the corrected OD values is shown. b A standard curve (S6–S2, left to right) was constructed by linear regression analysis. The lowest standard (S6) was set to 1000 AEUs, and the highest (S2) was set to 16,000 AEUs. The R2 value and straight line equation of the standard curve are depicted in the graph. OD values of the calibrator solutions were determined for each ELISA plate to ensure plate-to-plate consistency and to calculate AEUs, as shown in Fig. 5 Table 1  Intra- and inter-assay variability Intra-assay CV

Inter-assay CV

Mean (%) Mean OD SD

SD/mean (%)

Negative human serum

14

0.114

0.034 30

Marginal human serum

5

0.532

0.078 15

5

2.688

0.214

Positive human serum

8

8 18

The intra-assay coefficient of variability (CV) was determined by calculating the standard deviation of the means (SD)/the means of 12 replicates for each serum sample. The mean of five independent assays performed by different operators is shown. Inter-assay variability was calculated using the mean OD values of 12 replicates for each serum sample determined by five independent assays performed by four different operators on different days

Fig. 3  Specificity of ZEBOV antigen ELISA. a A total of 113 randomly selected anonymous human serum samples without EVD history (1:200 dilution) were analyzed by ZEBOV antigen ELISA. Binding to the mock (x axis) and ZEBOV (y axis) antigens is shown, and the binding of the standard 2 (S2) and 5 cross-reactive serum samples to the ZEBOV antigen (inside the triangle) are highlighted. b Five cross-reactive serum samples (#6, #43, #59, #63, and #113; 1:100 dilution) were analyzed for their recognition of ZEBOV proteins in Western blot analysis and compared with a goat anti-ZEBOV serum (1:30,000, positive control). HRP substrate was used to detect the enzymatic activities of HRP-coupled secondary antibodies. Binding of antibodies to viral proteins of ZEBOV Makona and to recombinantly expressed GP of ZEBOV was analyzed. A low-sensitivity HRP substrate was used, with 15 s of exposure. The asterisks indicate the detection of filoviral proteins

used a twofold serial dilution of serum from a convalescent EVD patient to cover the linear range of our ELISA (Fig. 2). Analyses of standard curves by different operators on consecutive days showed reproducible results (Fig. 2a). The straight line equation of the standard curve was used to calculate arbitrary ELISA units (AEUs) for each serum sample on each plate. The lowest standard (S6) was set to 1000 AEUs, and the highest (S2) was set to 16,000 AEUs (Fig. 2b). A 1:8 dilution of the patient serum (S1) was used as a positive control. To determine the intra-assay variability of the newly developed ZEBOV antigen ELISA, one negative, one positive, and one marginal sample were tested in 12 replicates. The calculated intra-assay variability was 8 %. The

13



Med Microbiol Immunol

Fig. 4  Analysis of Ebola virus disease patient sera. a Serum samples from two EVD patients (1:200) were analyzed on different days following the onset of symptoms by ZEBOV antigen ELISA. Seroconversion was documented by the detection of IgG-specific antibodies with an HRP-coupled secondary antibody (1:1000). Corrected OD values are shown. b The composition of antibodies in the serum from both patients at different time points following onset of symptoms (dilution 1:100) was determined on Western blot strips. Binding of antibodies to viral proteins of ZEBOV Makona and to recombinantly expressed GP of ZEBOV was analyzed. A low-sensitivity HRP substrate was used, with 15 s of exposure. c Native immunofluorescence

analysis of mock and GP-expressing HuH7 cells. At 24 h post-transfection, the convalescent serum from patient 1 (day 33 following the onset of symptoms), goat anti-ZEBOV serum (positive control) and a respective secondary anti-goat or anti-human Alexa Fluor® 488-coupled secondary antibody were applied. DAPI staining was used to visualize nuclei. d The antibody levels in the serum (black) and plasma (gray) samples from patient 1 (day 33, convalescent phase) were compared by analyzing twofold serial dilutions of both fluids, starting at 1:1600, by ZEBOV antigen ELISA. Corrected OD values are shown

inter-assay variability determined by testing the same set of samples on three different days by four independent operators (a total of 5 assays) was 18 % (Table 1). Among 113 negative European serum samples, five (4.4 %) reacted with the ZEBOV antigen but not with the mock antigen (Fig. 3a, dots in triangle). Therefore, the specificity of the ELISA was calculated to be 96 %. AEUs for the serum samples that reacted with the ZEBOV antigen

were calculated as 3386 (#6), 6336 (#43), 1861 (#59), 3738 (#63) and 1188 (#113). Those serum samples showed the presence of anti-ZEBOV VP40 (serum samples #6, #59, and #113), anti-NP (serum samples #6, #59, and #113), or anti-GP (serum sample #43) antibodies (Fig. 3b). Western blot analysis of serum #63 was not possible because of high unspecific background staining. Thirteen of the tested serum samples were lipemic, and four were hemolytic. Two

13

Med Microbiol Immunol Fig. 5  Analysis of plasma pairs of African Ebola virus disease survivors. a 27 plasma sample pairs of EVD cases were analyzed by ZEBOV antigen ELISA. One sample was collected early during infection (acute phase), the other sample of the same person was collected at a time, when RT-PCR results of the respective plasma were already tested negative (late phase). Shown are the corrected OD values. Cutoff of the assay is indicated

of the hemolytic serum samples and none of those that were lipemic reacted with the ZEBOV antigen, demonstrating the importance of careful analysis of hemolytic serum samples. Seventeen serum samples were positive for Epstein– Barr virus (EBV)-specific IgGs, and seven were positive for measles virus-specific IgGs. Among those, only one EBV positive sera (#63) reacted with the ZEBOV antigen. Our results are in accordance with previously published data, showing that 6.9 % of European human sera reacted positive with filovirus antigens [28]. For African human serum samples, seroprevalences seem to vary greatly from 1.4 % in Northeastern Gabon over 15.3 % among rural populations in Gabon to 22 % in Sierra Leone [29–31]. Detection of ZEBOV‑specific IgG antibodies To determine whether the ZEBOV ELISA was able to monitor the development of antibodies in EVD patients, serum samples from two EVD patients were analyzed between day 8 and day 27 following the onset of symptoms. The patients developed anti-ZEBOV IgG antibodies between day 11 and day 14. Antibody titers peaked on day 14 and day 16, respectively (Fig. 4a). The calculated AEUs for the samples obtained on day 14 (patient 1) and day 16 (patient 2) following onset of symptoms were 30,993 and 31,027, respectively. Western blot analyses were performed to determine which ZEBOV Makona proteins were recognized by day 9 to day 18 sera of both patients, and these proteins were revealed to be NP and VP40 (Fig. 4b). In addition, GP-specific antibodies were detected by Western blotting when large amounts of recombinant GP were blotted, suggesting that anti-GP antibodies recognizing linear epitopes were present (Fig. 4b). These results were confirmed by the reaction of a convalescent serum of patient 1 with cells expressing native recombinant GP on their surfaces (Fig. 4c). Serial dilutions of convalescent serum and plasma from patient 1 revealed that the endpoint titers

of both body fluids were the same, although the OD values detected in the serum sample were slightly higher at all tested dilutions (Fig. 4d). Furthermore, plasma sample pairs of 27 African EVD cases revealed an increase in IgG antibody titers against ZEBOV in each of the EVD cases from acute to late phase of infection (Fig. 5) underscoring the high sensitivity of the ELISA. Detection of ZEBOV GP‑specific IgG antibodies To confirm the utility of the ZEBOV ELISA for detecting GP-specific antibodies, we used a mouse monoclonal antibody (3B11, [21]) and a commercially available human monoclonal antibody (KZ52, [23]) directed against GP. Both antibodies recognized the ZEBOV antigen (Figs. 1c, 6a), indicating the presence of linear epitopes (3B11) and at least some conformational epitopes (KZ52). KZ52 antibody does not detect GP protein in Western blot analyses. Furthermore, we tested serum samples from an individual vaccinated with 5 × 107 PFU of a live attenuated, recombinant VSV expressing the ZEBOV GP (VSV∆G ZEBOV GP). VSV∆G ZEBOV GP was administered at 48 h after a suspected exposure to ZEBOV in March 2009 [26]. Postvaccination (p.v.) serum samples obtained on day 14 and day 31 as well as at 5 years after vaccination were assessed by ELISA. The serum samples showed clear reactivity on day 14 and day 31 p.v. that was still detectable at 5 years later (Fig. 6b). The calculated AEUs for the samples obtained on day 14 and day 31 p.v. were 6754 and 7105, respectively. At 5 years p.v., the amount of ZEBOV-specific antibodies was decreased to 1705 AEUs (Fig. 6b). High reactivity with mock antigen (concentrated supernatant of uninfected Vero cells) may be due to the fact that VSVΔG ZEBOV GP was propagated on Vero cells [26]. While serum obtained from the vaccinated individual on day 14 p.v. clearly recognized the native recombinant GP on cell surfaces, as shown by immunofluorescence analysis, serum

13



Med Microbiol Immunol

Fig. 6  GP specificity of ZEBOV antigen ELISA. a Polyclonal goat anti-ZEBOV serum (1:10,000), monoclonal anti-GP (3B11, 1:10,000, 0.12 µg/ml), polyclonal anti-NP (chicken, 1:1000), positive and negative human sera (PHS, 1:4000; NHS, 1:200) and a monoclonal human anti-GP antibody (KZ52, 1:200, 6 µg/ml) were analyzed by ZEBOV antigen ELISA. Detection of IgG-specific antibodies was performed with a respective HRP-coupled secondary antibody (1:1000). Corrected OD values are shown. b Serum from an individual vaccinated with VSVΔG ZEBOV GP was analyzed on day 14 or day 31 and at over 5 years post-vaccination (p.v.) by ZEBOV antigen ELISA (1:200 dilution). Binding to the mock (gray) and ZEBOV (black) antigens is shown. Calculation of arbitrary ELISA units (AEUs) is shown on

the right. Corrected OD values (=y) were used to calculate x values, and then the results were multiplied by 1000 to determine AEUs. c Native immunofluorescence analysis of mock and GP-expressing HuH7 cells. Antibodies in two serum samples from a vaccinated individual (both 1:50 dilutions) were detected with an anti-human Alexa Fluor® 488-coupled secondary antibody. DAPI staining was used to visualize nuclei. d The antibodies in the two sera at different time points post-vaccination (dilution 1:100) were determined on Western blot strips. Binding of antibodies to viral proteins of ZEBOV Makona and to recombinantly expressed GP of ZEBOV was analyzed. A lowsensitivity HRP substrate was used, with 9.6 s of exposure

from the same individual collected at 5 years p.v. was nonreactive in immunofluorescence analysis (Fig. 6c). However, residual amounts of GP-specific and NP-specific antibodies were present even 5 years p.v., as shown by Western blot analysis (Fig. 6d).

Discussion

13

In the present study, we developed an IgG ELISA using antigen derived from inactivated ZEBOV Makona particles. While several serological assays are available to

Med Microbiol Immunol

detect ZEBOV-specific antibodies [13–19], we established an ELISA based on the current circulating ZEBOV strain using sera from 29 EVD patients and one vaccinated individual [20, 25, 26]. The use of only one filoviral protein as an antigen is disadvantageous for detecting antibodies in EVD. We found that the serum of EVD patients contains antibodies against at least three different viral proteins (GP, NP, and VP40); thus, ELISA for detecting only one of them will show reduced reactivity and will detect seroconversion at a later time point. To address the question of whether residual cellular proteins in the antigen preparation reacted nonspecifically with serum antibodies, we included a mock antigen in the ELISA. This step was important because the detection of cellular components greatly varied among different human sera (Figs. 3a, 6b). The parallel testing of the sera with ZEBOV antigen and mock antigen allowed for the filtering of these variabilities. The precision of the ZEBOV antigen ELISA was determined by measuring the intra- and inter-assay variations (CV) for positive, marginal, and negative samples. The results showed very good reproducibility and comply with the generally accepted standards of intra-assay and interassay variabilities of no greater than 15 and 20 %, respectively [32]. The specificity of the ZEBOV ELISA was calculated to be 96 %. Sensitivity of the assay was tested by analysis of 29 paired samples of ZEBOV-infected patients that survived the infection. An increase in IgG titers from the PCR positive to the PCR negative plasma sample of the same individual was monitored in each of the 29 pairs indicating a sensitivity of 100 % (Fig. 4, 5). Furthermore, we determined that the seroconversion of the two EVD patients hospitalized in Germany occurred at approximately day 11–day 14 following the onset of symptoms. For most of the African EVD patients, the exact date of the onset of symptoms could not be determined. Therefore, no statistical reliable statement can be made of when after the onset of symptoms seroconversion takes place. Specific antibodies against ZEBOV GP were also detected in one vaccinated individual on day 14 postvaccination. Interestingly, anti-GP antibodies could be detected even at 5 years post-vaccination, which is the first indication that the VSVΔG ZEBOV GP-induced immune response is long lasting in humans. This finding supports data from nonhuman primate studies showing that the animals were protected against a challenge with MARV at 1 year after vaccination with VSVΔG MARV GP [33]. This result is important with respect to phase III clinical studies of VSVΔG ZEBOV GP in West Africa. The anti-GP KZ52 antibody binds to a conformational epitope composed of amino acids from both GP subunits, GP1 and GP2 [23]. KZ52 recognized the ZEBOV antigen in the ELISA performed in this study, whereas this

antibody is not suitable for detection on denaturing SDSPAGE, demonstrating that the ZEBOV ELISA displayed conformational GP epitopes. This finding was conceivable because although the ZEBOV antigen coated to the ELISA plates was boiled in 1 % SDS, it was not treated with reducing agents, which would have destroyed the disulfide bonds connecting GP1 and GP2 [34]. The recognition of KZ52 is in accordance with the notion that the ZEBOV antigen is partially refolded during incubation with isotonic buffer [35] and may be stabilized by the binding of the antibody. In summary, we have developed a highly specific and sensitive ELISA to detect IgG antibodies directed against different ZEBOV proteins in the sera of EVD patients and vaccinated individuals. The nonspecific binding of human antibodies to cellular proteins was filtered using mock antigen as a control. The results underscore the specificity, sensitivity, and precision of the ZEBOV antigen ELISA, which is suitable for emergency diagnostics and for assessing the immunogenicity of current vaccine candidates against ZEBOV. Acknowledgments  This work was supported by the German Center for Infection Research (DZIF), the Wellcome Trust Foundation through the World Health Organization: “Immunogenicity of Ebola Virus vaccine” (Grant Number: SPHQ14-LOA-296), the Federal Ministry of Education and Research (EBOKON), the European Union: EVIDENT “Ebola Virus Disease—correlates of protection, determinants of outcome, and clinical management” (Horizon 2020 Grant Agreement No.: 666100 and service contract IFS/2011/272372), as well as the Jürgen-Manchot-Foundation through a stipend to CR and the German Research Foundation (DFG, SPP1596). We especially thank Gotthard Ludwig and Michael Schmidt for their expert technical support with the BSL4 procedures at Philipps University of Marburg. We thank Carsten van Hammel, Ronny Kohlhoff, and Jörg Schmidt for the technical and administrative support. We also thank all of the voluntary donors for generously providing serum samples for scientific analyses as well as Petra Emmerich, Timo Wolf, and Marylyn M. Addo for providing some of the samples. We thank Stephanie Wurr und Elisa Pallasch for excellent assistance with biobanking in the BSL-4 laboratory at BNITM. The European Mobile Lab consortium: Miles W. Carroll, Roger Hewson, Joseph Akoi Bore, Raymond Koundouno, Saïd Abdellati, Babak Afrough, John Aiyepada, Patience Akhilomen, Danny Asogun, Barry Atkinson, Marlis Badusche, Amadou Bah, Simon Bate, Jan Baumann, Dirk Becker, Beate Becker-Ziaja, Anne Bocquin, Benny Borremans, Andrew Bosworth, Jan Peter Boettcher, Angela Cannas, Fabrizio Carletti, Concetta Castilletti, Simon Clark, Francesca Colavita, Sandra Diederich, Adomeh Donatus, Sophie Duraffour, Deborah Ehichioya, Heinz Ellerbrok, Maria Dolores Fenandez-Garcia, Alexandra Fizet, Erna Fleischmann, Sophie Gryseels, Antje Hermelink, Julia Hinzmann, Ute Hopf-Guevara, Yemisi Ighodalo, Lisa Jameson, Anne Kelterbaum, Zoltan Kis, Stefan Kloth, Claudia Kohl, Miša Korva, Annette Kraus, Eeva Kuisma, Andreas Kurth, Britta Liedigk, Christopher H. Logue, Anja Lüdtke, Piet Maes, James McCowen, Stéphane Mély, Marc Mertens, Silvia Meschi, Benjamin Meyer, Janine Michel, Peter Molkenthin, César Muñoz-Fontela, Doreen Muth, Edmund N. C. Newman, Didier Ngabo, Lisa Oestereich, Jennifer Okosun, Thomas Olokor, Racheal Omiunu, Emmanuel Omomoh, Elisa Pallasch, Bernadett Pályi, Jasmine Portmann, Thomas Pottage, Catherine Pratt, Simone Priesnitz, Serena Quartu, Julie Rappe, Johanna Repits, Martin Richter, Martin Rudolf, Andreas Sachse, Kristina Maria Schmidt, Gordian

13

Schudt, Thomas Strecker, Ruth Thom, Stephen Thomas, Ekaete Tobin, Howard Tolley, Jochen Trautner, Tine Vermoesen, Inês Vitoriano, Matthias Wagner, Svenja Wolff, Constanze Yue, Maria Rosaria Capobianchi, Romy Kerber, Tatjana Avšicˇ-Županc, Andreas Nitsche, Marc Strasser, Giuseppe Ippolito, Stephan Becker, Kilian Stoecker, Martin Gabriel, Hervé Raoul, Antonino Di Caro (executive board), Roman Wölfel (executive board) and Stephan Günther (executive board).

References 1. Kiley MP, Bowen ET, Eddy GA et al (1982) Filoviridae: a taxonomic home for Marburg and Ebola viruses? Intervirology 18:24–32 2. Mahanty S, Bray M (2004) Pathogenesis of filoviral haemorrhagic fevers. Lancet Infect Dis 4:487–498 3. Sanchez A, Geisbert TW (2006) Feldmann H. Marburg and Ebola viruses. Fields virology, Filoviridae, pp 1409–1448 4. Towner JS, Sealy TK, Khristova ML et al (2008) Newly discovered Ebola virus associated with hemorrhagic fever outbreak in Uganda. PLoS Pathog 4:e1000212 5. Falzarano D, Feldmann F, Grolla A et al (2011) Single immunization with a monovalent vesicular stomatitis virus-based vaccine protects nonhuman primates against heterologous challenge with Bundibugyo ebolavirus. J Infect Dis 204(Suppl 3):S1082–S1089 6. Wong G, Audet J, Fernando L et al (2014) Immunization with vesicular stomatitis virus vaccine expressing the Ebola glycoprotein provides sustained long-term protection in rodents. Vaccine 32:5722–5729 7. Wong G, Richardson JS, Cutts T, Qiu X, Kobinger GP (2015) Intranasal immunization with an adenovirus vaccine protects guinea pigs from Ebola virus transmission by infected animals. Antiviral Res 116:17–19. doi:10.1016/j.antiviral.2015.01.001 8. Kanapathipillai R, Henao Restrepo AM, Fast P et al (2014) Ebola vaccine—an urgent international priority. N Engl J Med 371:2249–2251 9. Ledgerwood JE, DeZure AD, Stanley DA, Novik L, Enama ME, Berkowitz NM, Hu Z, Joshi G, Ploquin A, Sitar S, Gordon IJ, Plummer SA, Holman LA, Hendel CS, Yamshchikov G, Roman F, Nicosia A, Colloca S, Cortese R, Bailer RT, Schwartz RM, Roederer M, Mascola JR, Koup RA, Sullivan NJ, Graham BS; the VRC 207 Study Team (2014) Chimpanzee adenovirus vector ebola vaccine—preliminary report. N Engl J Med. doi:10.1056/ NEJMoa1410863 10. Rampling T, Ewer K, Bowyer G, Wright D, Imoukhuede EB, Payne R, Hartnell F, Gibani M, Bliss C, Minhinnick A, Wilkie M, Venkatraman N, Poulton I, Lella N, Roberts R, SierraDavidson K, Krähling V, Berrie E, Roman F, De Ryck I, Nicosia A, Sullivan NJ, Stanley DA, Ledgerwood JE, Schwartz RM, Siani L, Colloca S, Folgori A, Di Marco S, Cortese R, Becker S, Graham BS, Koup RA, Levine MM, Moorthy V, Pollard AJ, Draper SJ, Ballou WR, Lawrie A, Gilbert SC, Hill AV (2015) A monovalent chimpanzee adenovirus ebola vaccine—preliminary report. N Engl J Med. doi:10.1056/NEJMoa1411627 11. Agnandji ST, Huttner A, Zinser ME, Njuguna P, Dahlke C, Fernandes JF, Yerly S, Dayer JA, Kraehling V, Kasonta R, Adegnika AA, Altfeld M, Auderset F, Bache EB, Biedenkopf N, Borregaard S, Brosnahan JS, Burrow R, Combescure C, Desmeules J, Eickmann M, Fehling SK, Finckh A, Goncalves AR, Grobusch MP, Hooper J, Jambrecina A, Kabwende AL, Kaya G, Kimani D, Lell B, Lemaître B, Lohse AW, Massinga-Loembe M, Matthey A, Mordmüller B, Nolting A, Ogwang C, Ramharter M, Schmidt-Chanasit J, Schmiedel S, Silvera P, Stahl FR, Staines

13

Med Microbiol Immunol HM, Strecker T, Stubbe HC, Tsofa B, Zaki S, Fast P, Moorthy V, Kaiser L, Krishna S, Becker S, Kieny MP, Bejon P, Kremsner PG, Addo MM, Siegrist CA (2015) Phase 1 trials of rVSV Ebola vaccine in Africa and Europe—preliminary report. N Engl J Med. doi:10.1056/NEJMoa1502924 12. WHO (2015) Ebola vaccines, therapies, and diagnostics. http:// www.who.int/medicines/emp_ebola_q_as/en/. Accessed 06 July 2015 13. Huang Y, Zhu Y, Yang M, Zhang Z, Song D, Yuan Z (2014) Nucleoprotein-based indirect enzyme-linked immunosorbent assay (indirect ELISA) for detecting antibodies specific to Ebola virus and Marbug virus. Virol Sin 29:372–380 14. Ikegami T, Saijo M, Niikura M et al (2003) Immunoglobulin G enzyme-linked immunosorbent assay using truncated nucleoproteins of Reston Ebola virus. Epidemiol Infect 130:533–539 15. Ksiazek TG, West CP, Rollin PE, Jahrling PB, Peters CJ (1999) ELISA for the detection of antibodies to Ebola viruses. J Infect Dis 179(Suppl 1):S192–S198 16. Macneil A, Reed Z, Rollin PE (2011) Serologic cross-reactivity of human IgM and IgG antibodies to five species of Ebola virus. PLoS Negl Trop Dis 5:e1175 17. Nakayama E, Yokoyama A, Miyamoto H et al (2010) Enzymelinked immunosorbent assay for detection of filovirus speciesspecific antibodies. Clin Vaccine Immunol 17:1723–1728 18. Saijo M, Niikura M, Morikawa S et al (2001) Enzyme-linked immunosorbent assays for detection of antibodies to Ebola and Marburg viruses using recombinant nucleoproteins. J Clin Microbiol 39:1–7 19. Sobarzo A, Perelman E, Groseth A et al (2012) Profiling the native specific human humoral immune response to Sudan Ebola virus strain Gulu by chemiluminescence enzyme-linked immunosorbent assay. Clin Vaccine Immunol 19:1844–1852 20. Wolf T, Kann G, Becker S et al (2015) Severe Ebola virus disease with vascular leakage and multiorgan failure: treatment of a patient in intensive care. Lancet 385:1428–1435 21. Lucht A, Grunow R, Otterbein C, Möller P, Feldmann H, Becker S (2004) Production of monoclonal antibodies and development of an antigen capture ELISA directed against the envelope glycoprotein GP of Ebola virus. Med Microbiol Immunol 193:181–187 22. Pauly D, Chacana PA, Calzado EG, Brembs B, Schade R (2011) IgY technology: extraction of chicken antibodies from egg yolk by polyethylene glycol (PEG) precipitation. J Vis Exp 51:e3084. doi: 10.3791/3084 23. Lee JE, Fusco ML, Hessell AJ, Oswald WB, Burton DR, Saphire EO (2008) Structure of the Ebola virus glycoprotein bound to an antibody from a human survivor. Nature 454:177–182 24. Baize S, Pannetier D, Oestereich L et al (2014) Emer gence of Zaire Ebola virus disease in Guinea. N Engl J Med 371:1418–1425 25. Kreuels B, Wichmann D, Emmerich P et al (2014) A case of severe Ebola virus infection complicated by gram-negative septicemia. N Engl J Med 371:2394–2401 26. Günther S, Feldmann H, Geisbert TW et al (2011) Management of accidental exposure to Ebola virus in the biosafety level 4 laboratory, Hamburg, Germany. J Infect Dis 204(Suppl 3):S785–S790 27. Kolesnikova L, Mittler E, Schudt G, Shams-Eldin H, Becker S (2012) Phosphorylation of Marburg virus matrix protein VP40 triggers assembly of nucleocapsids with the viral envelope at the plasma membrane. Cell Microbiol 14:182–197 28. Becker S, Feldmann H, Will C, Slenczka W (1992) Evidence for occurrence of filovirus antibodies in humans and imported monkeys: Do subclinical filovirus infections occur worldwide? Med Microbiol Immunol 181:43–55

Med Microbiol Immunol 29. Becquart P, Wauquier N, Mahlakõiv T et al (2010) High prevalence of both humoral and cellular immunity to Zaire ebolavirus among rural populations in Gabon. PLoS ONE 5:e9126 30. Boisen ML, Schieffelin JS, Goba A et al (2015) Multiple circulating infections can mimic the early stages of viral hemorrhagic fevers and possible human exposure to filoviruses in Sierra Leone prior to the 2014 outbreak. Viral Immunol 28:19–31 31. Heffernan RT, Pambo B, Hatchett RJ, Leman PA, Swanepoel R, Ryder RW (2005) Low seroprevalence of IgG antibodies to Ebola virus in an epidemic zone: Ogooué–Ivindo region, Northeastern Gabon, 1997. J Infect Dis 191:964–968

32. Jacobson RH (1998) Validation of serological assays for diagnosis of infectious diseases. Rev Sci Tech 17:469–526 33. Mire CE, Geisbert JB, Agans KN et al (2014) Durability of a vesicular stomatitis virus-based marburg virus vaccine in nonhuman primates. PLoS ONE 9:e94355 34. Feldmann H, Klenk HD, Sanchez A (1993) Molecular biology and evolution of filoviruses. Arch Virol Suppl 7:81–100 35. Dornmair K, Kiefer H, Jähnig F (1990) Refolding of an integral membrane protein. OmpA of Escherichia coli. J Biol Chem 265:18907–18911

13