A field usable qualitative anti-protective antigen enzyme-linked immunosorbent assay for serodiagnosis of human anthrax



Ajay Kumar Goel, Biotechnology Division, Defence Research and Development Establishment, Gwalior-474 002, India. Tel: 91 751 2233742; fax: 91 751 2341148; email: akgoel73@yahoo.co.uk


Although all mammals, including humans, are vulnerable when they come into direct contact with infected animals, anthrax is primarily a disease of herbivorous animals. In countries like India, cutaneous anthrax is a public health problem in several regions. Hence, a simple and efficacious serodiagnostic assay for large scale surveillance of endemic populations is required. In the present study, a field-usable, qualitative ELISA was developed for serodiagnosis of human anthrax. Results are assessed on a visual basis and no sophisticated instruments are required. Anti-protective antigen (PA) IgG was determined by visual examination of ELISA results of 225 human serum samples (160 from healthy humans, 5 from PA vaccinated individuals and 60 from confirmed anthrax cases). Comparison of the ELISA results with the results obtained from optical density values showed compatible sensitivity and specificity. Assay sensitivity, specificity, and positive and negative predictive values were found to be 100%. The developed assay could be a very useful tool for serological diagnosis of anthrax infection in humans.

List of Abbreviations
B. anthracis

Bacillus anthracis


diagnostic sensitivity


diagnostic specificity


edema factor


edema toxin


false negative


false positive


horseradish peroxidase




lethal factor


lethal toxin


negative predictive value


protective antigen


positive predictive value


recombinant protective antigen


room temperature


skim milk powder




true negative


true positive

Bacillus anthracis, the causative agent of anthrax, is a gram-positive, spore-forming, rod-shaped bacterium that primarily affects herbivorous livestock and wildlife species. Humans can acquire this disease through contact with infected animals and their products under natural conditions [1]. Although the number of anthrax cases is diminishing in developed countries, it remains a public health problem in several developing countries in which the main source of income is farming or where communities live in areas of interface with wildlife [2, 3]. India has the largest livestock population in the world and cutaneous anthrax is endemic in several regions [4, 5]. However, because no suitable serodiagnostic assay is available, most cases remain undiagnosed. Hence, a simple and reliable serodiagnostic assay for human anthrax is of utmost importance.

The toxigenicity and pathogenicity of B. anthracis are due to a tripartite toxin and the poly-γ-D-glutamic acid capsule, respectively [6]. B. anthracis bacteria form capsules of poly-D-glutamic acid that impede the host immune system and inhibit macrophages from engulfing and destroying the bacteria [7]. Anthrax toxins are secreted as three distinct proteins, namely PA, LF and EF, the activities of which have been well described [8, 9]. The exotoxins are binary and composed of a B (binding protein) and an A (enzymatically active) protein. PA acts as a B component and combines with EF and LF to form the binary toxins ETx and LTx, respectively [10]. Studies in animal models have confirmed that the immune response to PA is central to protection against B. anthracis [11, 12]. Therefore, presence of anti-PA IgG in human serum is an accurate indicator of anthrax exposure. Simultaneously, it can confirm the efficacy of vaccines in both humans and animals. Earlier reports have also shown that anti-PA IgG ELISA is a valuable tool for confirmation of cases of cutaneous and inhalational anthrax [13, 14].

In this study, we report the development of a visual ELISA for detection of anti-PA IgG in human serum samples. The assay does not require sophisticated laboratory facilities or the associated plate washers and plate readers, which are costly pieces of equipment. Therefore, this assay is a very useful tool for serodiagnosis of anthrax.


Protective antigen

Recombinant protective antigen was obtained from Alpha Diagnostics International, San Antonio, TX, USA. Lyophilized antigen was reconstituted in ultra pure water and stored frozen at −80°C in small aliquots (50 µL, 1 mg/mL) in 5 mM HEPES, pH 7.3.

Human test and control sera

In all, 225 serum samples were obtained from volunteers with their consent and following the rules prescribed by the Institutional Ethical Committee. The sera were allocated to the following five distinct groups:

  1. Group I (n = 100): samples from healthy persons from non-endemic anthrax regions

    These samples were obtained from healthy blood donors in northern and central India, who are representative of the anthrax non-endemic Indian population. The selected donors had no prior exposure to anthrax or related infections or vaccinations and belonged to various age groups. These samples were negative for anti-PA IgG [14].

  2. Group II (n=50): samples from healthy persons from anthrax endemic regions

    These samples were obtained from healthy blood donors from anthrax endemic areas (southern India). The selected donors had no prior exposure to anthrax or related infections or vaccinations. Theses samples were negative for anti-PA IgG [14].

  3. Group III (n=10): samples from patients with clinically confirmed non-anthrax infections

    These samples were obtained from clinically proven non-anthrax patients from anthrax endemic areas. The selection of donors was made on the basis that they had ailments other than anthrax. Theses samples were negative for anti-PA IgG [14].

  4. Group IV (n=5): samples from persons vaccinated with anthrax vaccine adsorbed

    Five sera were obtained from persons who had been vaccinated with AVA. These sera served as positive controls for the presence of anti-PA IgG. These samples were positive for anti-PA IgG [14].

  5. Group V (n=60): samples from patients with clinically and immunologically diagnosed cutaneous anthrax

These samples were obtained from clinically proven anthrax patients from anthrax endemic areas. These patients lived in southern India and their diagnosis were confirmed clinically and epidemiologically. The presence of anti-PA IgG was confirmed in their sera by western blot analysis. These samples were positive for anti-PA IgG [14].

Qualitative anti-protective antigen immunoglobulin G ELISA

Maxisorp flat bottom plates (Nalge Nunc International, Roskilde, Denmark) were coated with 100 µL/well of PBS, pH 7.4 containing 1 µg/mL of rPA or 1 µg/mL of SMP in separate wells. After 1 hr incubation at RT, the wells were washed manually three times with washing buffer (PBS containing 0.1% Tween 20, pH 7.4). The wells were blocked with 300 µL of blocking buffer (3% SMP in PBS, pH 7.4) for 1 hr at RT. After decanting the blocking buffer, the plates were blotted dry on paper towels but were not washed during this step. A series of positive, negative and test sera were diluted to 1:250 in PBS containing 1% SMP, pH 7.4. The final volume of serum samples added to each well was 100 µL. Each sample was added to paired wells, that is, one coated with rPA and another coated with SMP. After incubation for 1 hr at RT, the plates were washed three times with wash buffer and blotted dry on paper towels. Goat anti-human IgG (Fc-specific) conjugated with horseradish peroxidase (Sigma–Aldrich, St Louis, MO, USA) was diluted to 1:20,000 in PBS containing 1% SMP and 100 µL of this was added to each well. After 1 hr incubation at RT, the plates were washed three times with wash buffer. 100 µL of TMB (Sigma–Aldrich) containing hydrogen peroxide was added to each well as substrate. Color development was observed visually after 15 min of incubation at RT. All tests were performed in duplicate and each serum sample from each individual was tested in PA-coated as well as in SMP-coated wells. The samples were also analyzed by a previously described laboratory-based ELISA for comparison purposes [14].

Assay acceptance criteria

Assays were accepted on the basis of evaluation of a series of positive and negative sera. The serum samples were blinded and the analyst did not know their source or status. The criteria for acceptance were as follows. Samples that developed intense blue color in PA-coated wells and no color in the corresponding SMP-coated wells were classified as positive for anti-PA IgG, whereas samples developing no color in either well were classified as negative for anti-PA IgG. The tests were classed as invalid when intense blue color developed in only the SMP-coated well or in both PA-coated and SMP-coated wells. The latter type of color development can be caused by non-specific binding of the serum proteins to the plate wells.

Robustness of the assay

Robustness refers to an assay's capacity to be unaffected by the minor variations in test situations that may occur over the course of testing. In this assay, the effects of incubation temperature (25–30°C) and incubation time (60 ± 5 min) on the performance of the assay were studied.

Statistical analysis

The DSN, DSP, PPV and NPV of the qualitative ELISA were determined by a categorical scale (dark blue to pale blue and transparent blue) that was converted to a binary scale (positive or negative). Contingency tables based on the status of each patient at the time of blood sample collection and the laboratory ELISA results were constructed for each ELISA to calculate the qualitative ELISA sensitivity, specificity, PPV, NPV and accuracy, [14]. The sensitivity was calculated as [TP/(TP + FN)] × 100, where TP = true positives and FN = false negatives, and specificity was calculated as [TN/(TN + FP)] × 100, where TN = true negatives and FP = false positives. The PPV and the NPV were calculated as TP/(TP + FP) and TN/(FN + TN), respectively. High levels of DSN and DSP together with the laboratory comparison indicate that the test assay data are accurate and reliable [15]. Agreement between field and laboratory analysis was evaluated by using kappa values using GraphPad software (GraphPad, San Diego, CA, USA).


The qualitative ELISA described in the present study yielded visual results that allowed determination of anti-PA IgG in 225 serum samples from five different groups of subjects (Table 1). On the basis of these visual readings, all 160 control sera samples (Groups 1, II, and III) and the 65 sera samples from vaccinated (Group IV) and anthrax infected individuals (Group V) were identified correctly.

Table 1. Evaluation of human anti-protective antigen immunoglobulin G visual enzyme-linked immunosorbent assay of serum samples from different groups of subjects
GroupNo. of samplesNo. of positive tests for anti-PA IgG
ELISA [14]Visual ELISA

Overall, sera from all control groups (Group I + Group II + Group III, n = 160) were visually transparent blue or colorless in both PA-coated and SMP-coated wells. However, strongly to moderately positive sera (Group IV + Group V) developed dark blue to pale blue color in PA-coated wells and transparent blue color in SMP-coated wells after 15 min of incubation at RT. The colors were stable for 1–2 hrs after exposure to light and overnight when stored in the dark.

The robustness of the assay was assessed by slightly changing the incubation temperatures and times from 25 to 30°C and 60 ± 5 min, respectively. No significant differences were apparent visually as a result of these variations.

The DSN of the visual ELISA, a measure of the ability of the assay to correctly detect infection was in agreement with the results of laboratory-based ELISA. The DSP, a measure of the ability of the test to detect a specific infection, was also found to be 100% (95% CI, 97–100) with no false-positive or false-negative results when compared to ELISA results [14]. Both PPV and NPV of the visual qualitative ELISA were 100%. In terms of all the tests that gave correct results, the efficiency or test accuracy was 100%. As determined by kappa values, the results of visual ELISA were in 100% agreement with the results of laboratory-based ELISA [14]. Thus, extremely good agreement on determination of anti-PA IgG was observed when the results of this visual ELISA were compared with those of the laboratory-based ELISA [14].


Anthrax is a zoonotic disease caused by B. anthracis. Cutaneous anthrax is a skin infection caused by direct contact with the bacterium. Depending on the route of exposure to B. anthracis, anthrax takes three forms, namely cutaneous, gastrointestinal and inhalational. If untreated, the case-fatality rates of these forms of anthrax in humans are 1–20%, 25–60% and 86–89%, respectively, [16]. The cutaneous route accounts for 95% and the inhalational route for 5% of all reported anthrax cases whereas gastrointestinal anthrax is quite rare [17]. In addition, anthrax is an agent of biological warfare and can be used as a military or terrorist weapon. The virulence of anthrax depends on its tripartite exotoxins comprising PA and two enzymatically active subunits, LF and EF.

The long time required for seroconversion to occur after onset of anthrax limits the diagnostic usefulness of serological tests. A diagnosis of cutaneous anthrax is traditionally established by microbiological methods such as demonstrating gram-positive, encapsulated bacilli in smears from lesions or isolating B. anthracis in cultures. However, Gram stain and culture for B. anthracis can be negative in patients who have received antibiotic therapy before collection of samples [18]. Hence, serological testing is an important tool for confirming the diagnosis in cases in which direct isolation of the anthrax bacillus has not been achieved, for assessing vaccine efficiency and for epidemiological investigations. Protection against anthrax can be achieved by active induction of antitoxin antibody responses or passive administration of antitoxin antibodies. Studies in animal models have confirmed that the immune response to PA is central to protection against B. anthracis [12]. Antibodies against PA reportedly appear as early as 11 days after onset of the disease or 15 days after likely exposure to B. anthracis [19]. Because the immune response against PA of B. anthracis persists for as long as 1–2 years after infection, detection of anti PA IgG is a good candidate for the surveillance of anthrax infection [20]. Serology was particularly important in diagnosing cases of cutaneous anthrax after the 2001 bioterrorist attacks, because serology was the only laboratory test that produced positive results in 3 of the 10 cutaneous anthrax cases [21]. Serology is generally performed by plate ELISA under laboratory conditions. However, a simple ELISA that can be used under field conditions is required. Therefore, we have developed a qualitative ELISA for measuring anti-PA antibodies in human subjects for continuous monitoring as well as quick detection of cutaneous infections. This qualitative ELISA does not require any sophisticated instruments such as plate washers or plate readers because color development in the test (coated with PA) and control wells (coated with SMP) is observed with the naked eye. The antigen-coated plates and other reagents can be stored in refrigerators, which are readily available at point of care centers. Alternatively, freeze-dried PA can be reconstituted for fresh coating of the ELISA plates at point of care centers.

The DSN, specificity, PPV and NPV of this assay were all found to be 100%. The results were in extremely good agreement with the results of laboratory-based ELISA [14]. The sensitivity of the assay depends on the proportion of actual positives that it correctly identifies whereas the specificity depends on the proportion of negatives that it correctly identifies. From a practical point of view, qualitative results are adequate because it is usual for the first stage of routine surveillance studies to determine whether antibodies are circulating in a particular population or not and if they are, for the second step to measure their concentrations. Therefore, instead of making quantification of antibody concentrations or isolation of causative agent as the first goal, samples can be first analyzed by this ELISA. Thus, this assay could be a very useful tool for surveillance and serodiagnosis of anthrax infection in humans.


We thank the Director of the Defense Research and Development Establishment, Gwalior for providing necessary facilities and funds for this research work. N.G. thanks the Indian Council of Medical Research for providing a Senior Research Fellowship. We acknowledge help received from Dr. Harivadan Lukka, Dr. Swarajya Laxmi and Dr. T.N. Rao, Andhra Medical College, Visakhapatnam, Andhra Pradesh, India during sample collection.


None of the authors has any commercial associations or financial disclosures.