Skip directly to site content Skip directly to page options Skip directly to A-Z link Skip directly to A-Z link Skip directly to A-Z link
Volume 15, Number 10—October 2009

Immunologic Response of Unvaccinated Workers Exposed to Anthrax, Belgium

Pierre WattiauComments to Author , Marc Govaerts, Dimitrios Frangoulidis, David Fretin, Esther Kissling, Mieke Van Hessche, Bernard China, Martine Poncin, Yvo Pirenne, and Germaine Hanquet
Author affiliations: Veterinary and Agro-chemical Research Centre, Brussels, Belgium (P. Wattiau, M. Govaerts, D. Fretin, M. Van Hessche); Bundeswehr Institute of Microbiology, Munich, Germany (D. Frangoulidis); Institute of Public Health, Brussels, Belgium (E. Kissling, B. China, G. Hanquet); Occupational Medicine PROVIKMO, Verviers, Belgium (M. Poncin); Medical Inspection, Angleur, Belgium (Y. Pirenne); European Centre for Disease Control and Prevention, Stockholm, Sweden (E. Kissling)

Cite This Article


To determine immunologic reactivity to Bacillus anthrax antigens, we conducted serologic testing of workers in a factory that performed scouring of wool and goat hair. Of 66 workers, ≈10% had circulating antibodies or T lymphocytes that reacted with anthrax protective antigen. Individual immunity varied from undetectable to high.

Industrial anthrax, also known as woolsorter’s disease, was a serious threat in the 19th and early 20th centuries when the wool industry was flourishing. The causal agent, Bacillus anthracis, was brought into factories in sporulated form with the organic matter that was contaminating the animal fibers. The pathogen provoked the characteristic necrotic lesions on the skin of the wool workers (cutaneous anthrax), but it could also cause a respiratory disease through airborne transmission (inhalational anthrax). In 1950, 90% of those with the latter form died, although the proportion of deaths could be lowered to 50% with the aggressive therapy that was later used to treat the victims of the deliberate release of anthrax in the United States in 2001 (1,2).

Today, industrial processing of wool and goat hair has almost disappeared from Western industrialized countries. Cases of human anthrax have become rare in Europe (3,4), but they can sometimes result from contact with imported contaminated material (5,6). Apart from the 2001 attacks (7), the most recent human anthrax epidemic in the United States was reported in 1957 in a large goat hair–processing mill in Manchester, New Hampshire (1). In a recent study, we investigated the microbiologic flora of a Belgian factory (in operation since 1880) that processes and scours wool and goat hair from all over the world. Living anthrax spores were demonstrated in goat hair fibers, air dust, and unprocessed wastewater produced from goat hair scouring (8). No clinical case of anthrax was recorded among the employees of this company except for a possible cutaneous lesion, reported by a worker in 2002, the cause of which remained unconfirmed. In the current study, we investigated the immunity of the factory workers. Since none of these workers had been vaccinated against anthrax, we assumed that immunologic reactivity to anthrax antigens, if any, would very likely demonstrate exposure to B. anthracis.

The Study

Figure 1

Thumbnail of Representative examples of lymphocyte proliferation results. Proliferation was assayed by measuring 3H-thymidine incorporation (counts per minute [cpm]) of culture lymphocytes stimulated with different antigens and by determining the respective proliferative Indexes. The latter were calculated by dividing the cpm induced by a given antigen by the cpm induced by a negative control antigen (phosphate-buffered saline (PBS), white boxes). The proliferative index is a parameter that refl

Figure 1. Representative examples of lymphocyte proliferation results. Proliferation was assayed by measuring 3H-thymidine incorporation (counts per minute [cpm]) of culture lymphocytes stimulated with different antigens and by determining the respective proliferative Indexes....

Figure 2

Thumbnail of Graph showing anti–protective antigen (PA) immunoglobulin G (IgG) titers plotted against 3H-thymidine counts per minute (cpm) derived from PA-stimulated blood cell cultures conducted in year 2. The vertical lines define the ELISA borderline and upper thresholds (10 IU/mL and 15 IU/mL, respectively), which were defined as PA titers by the ELISA kit manufacturer, i.e., titers supposed to confer protection after vaccination. Samples testing below the borderline threshold are considered

Figure 2. Graph showing anti–protective antigen (PA) immunoglobulin G (IgG) titers plotted against 3H-thymidine counts per minute (cpm) derived from PA-stimulated blood cell cultures conducted in year 2. The vertical lines define the...

Blood samples were obtained from 66 of the 67 factory workers, including the administrative employees. Serologic testing was carried out at 2 time points (December 2006 and December 2007) by using a commercial ELISA (Serion, Würzburg, Germany) based on plates coated with purified anthrax protective antigen (PA) (Technical Appendix). The first year, 3 workers had immunoglobulin (Ig) G titers above the threshold recommended by the manufacturer for vaccine protection (>15 IU/mL), and titers for another worker were considered borderline (10–15 IU/mL). All 4 workers had positive results by Western blot or dot blot analysis with pure recombinant anthrax PA and lethal factor (LF). One year later, 54 workers were sampled (2 were new employees). The second round of testing gave similar results, except for 3 additional borderline cases which could also be confirmed by Western blot/dot blot analysis (Table). Lymphocyte proliferation assays were performed concurrently by using fresh, heparinized, whole blood samples to evaluate the cell-mediated immunity of the workers (9). This technique measures the ability of lymphocytes placed in short-term tissue culture to undergo clonal proliferation when stimulated in vitro by a foreign antigen. Cell proliferation was determined by measuring the incorporation of 3H-thymidine into chromosomal DNA. The release of interferon-gamma (IFN-γ) in the course of lymphocyte stimulation was also measured to assess antigen-specific, cell-mediated reactivity. The antigens used here were pure recombinant PA and LF, along with positive control (concanavalin A) and negative control (phosphate buffer) stimulants. As shown in the Table, 2 cultures were positive in proliferation assays. Of these 2 cultures, 1 reacted with PA and LF, and 1 reacted with PA only. When added together, PA and LF suppressed the proliferative effect of the individual antigens, consecutive to the probable cytotoxicity induced by the 2 assembled antigens (porin + toxin). Typical examples of lymphocyte proliferation results are shown in Figure 1. The lymphocyte cultures found to be responsive to pure anthrax antigens originated from workers who had little circulating anti-PA IgG (<15 IU/mL), as tested by ELISA (Figure 2). However, their serum tested positive by Western blot analysis (Table). Moreover, supernatants of PA-stimulated lymphocyte cultures derived from the blood of these workers with positive results by determining counts per minute, were confirmed by IFN-γ release assay.


Although some progress made in improving the biologic safety of the industrial processing of wool and goat hair (for example, systematic disinfection, air filtering, and protective gear for employees working in closed areas), this study shows that B. anthracis still poses a health risk to modern wool workers. Handling nondisinfected, raw animal fibers from areas where anthrax is endemic, such as the southern Caucasus region and the Middle East, has been and remains an at-risk activity. The presence of circulating antibodies and T lymphocytes that react with antigens expressed only by vegetative cells of B. anthracis in unvaccinated wool workers confirms several previous findings. First, these findings support the conclusions that anthrax spores are able to germinate into vegetative cells at the sites of exposure (skin, mucosa, respiratory tract) and cause asymptomatic infection (10,11). Second, the extent to which the human immune system responds to exposure to anthrax spores from the surrounding environment as well as the nature of this response varies tremendously from person to person. This conclusion was well exemplified by the situations of 2 workers. Results from 1 worker (T29) displayed a high IgG titer (>100 U/mL) but little or no cell-mediated reactivity. Results from the other worker (T2) showed significant lymphocyte reactivity (3H-thymidine counts >700 counts per minute, which corresponds to a proliferative index of 6, p>0.01), but a low IgG titer (Figure 2), which reflects reciprocal T- and B-cell responses. None of the persons whose samples tested positive by ELISA reported a past episode of anthrax (according to face-to-face interviews conducted when blood was sampled). Hence, their seroconversion most likely resulted from asymptomatic B. anthracis infection. One worker reported having had a skin lesion possibly compatible with cutaneous anthrax 4 years before the study. That worker’s samples tested positive by lymphocyte proliferation assay, Western blot, and dot blot, but not by anti-PA ELISA.

Notably, samples from many workers from the same factory, who had been exposed to goat hair for years in similar conditions, did not display positive serologic results when tested by ELISA. During our study, however, we noticed that serum samples from 3 workers had seroconverted from negative to partially protective (borderline) IgG titers at some point between the 2 blood samplings as determined by anti-PA ELISA. Given the long history of these workers at the company, the apparent lack of anti-PA antibodies at the first blood sampling may have been misestimated due to the high threshold defined for seropositivity by the commercial ELISA used in the study. This commercial kit is indeed primarily aimed at evaluating the efficacy of anthrax vaccination rather than at detecting antibody responses after exposure to subinfectious doses of anthrax spores (12). Accordingly, we noticed that of the 3 workers who seroconverted, 2 tested positive by Western blot, and 1 tested positive by dot blot when tested retrospectively at year 1. Blotting techniques seem thus more sensitive than the presently used ELISA seropositivity threshold for detecting low anti-PA antibody titers. The low sensitivity of the method used in this work to assess cell-mediated immunity (whole blood proliferation assay) may have also underestimated the actual number of workers who exhibited cell-mediated immunity against B. anthracis, and the results should be regarded as indicative rather than representative.

PA-based anthrax vaccines are available to protect professionally exposed people, such as the US anthrax vaccine adsorbed or the UK anthrax vaccine. These vaccines are efficient and elicit humoral responses that protect the vaccinees against toxin-associated death (13). They do require long clinical protocols and yearly boosters (14) and are not officially licensed in European Union member states (except the United Kingdom). According to some authors, these vaccines might not protect wool-workers efficiently against subclinical infection, spore germination, or bacteremia (13,15). Anthrax vaccines that confer long-term immunity of both the humoral and cellular type are not yet available for the general public. Vaccines with such characteristics would be highly desirable to better protect persons who work with animal products that are possibly contaminated with anthrax spores.

Dr Wattiau is a molecular microbiologist specialized in the diagnosis of zoonotic pathogens at the Department of Bacterial Diseases of the Veterinary and Agro-chemical Research Centre. His research interests are focused on the distribution of highly pathogenic bacteria in the environment and on the molecular diagnosis of infectious agents.



We acknowledge the director and the employees of the scouring company for their constructive collaboration, W.D. Splettstoesser and M. Ehrle (Bundeswehr) for expert technical assistance, and J.-M. Henkinbrant and Y. Nizet for helpful discussions.

Analysis costs were supported by the Occupational Medicine group PROVIKMO and by the Veterinary and Agro-chemical Research Centre. European Cooperation in the field of Scientific and Technical Research action B28, initiated by P. Butaye, is also acknowledged.



  1. Brachman  PS, Plotkin  SA, Bumford  FH, Atchisson  MM. An epidemic of inhalation anthrax: the first in the twentieth century. II. Epidemiology. Am J Hyg. 1960;72:623.PubMedGoogle Scholar
  2. Bossi  P, Tegnell  A, Baka  A, Van Loock  F, Hendriks  J, Werner  A, Bichat guidelines for the clinical management of anthrax and bioterrorism-related anthrax. Euro Surveill. 2004;9:E34.PubMedGoogle Scholar
  3. European Commission. Health and consummers directorate C—Public health and risk assessment. Cases of anthrax reported to EU, 1997–2006 [cited 2009 Aug 31]. Available from
  4. Kreidl  P, Stifter  E, Richter  A, Aschbachert  R, Nienstedt  F, Unterhuber  H, Anthrax in animals and a farmer in Alto Adige, Italy. Euro Surveill. 2006;11:E0602163.PubMedGoogle Scholar
  5. Eurosurveillance Editorial Team. Probable human anthrax death in Scotland. Euro Surveill. 2006;11:E0608172.
  6. Van den Enden  E, Van Gompel  A, Van Esbroeck  M. Cutaneous anthrax, Belgian traveler. Emerg Infect Dis. 2006;12:5235.PubMedGoogle Scholar
  7. McCarthy  M. Anthrax attack in the USA. Lancet Infect Dis. 2001;1:2889. DOIPubMedGoogle Scholar
  8. Wattiau  P, Klee  SR, Fretin  D, Van Hessche  M, Menart  M, Franz  T, Occurrence and genetic diversity of Bacillus anthracis strains isolated in an active wool-cleaning factory. Appl Environ Microbiol. 2008;74:400511. DOIPubMedGoogle Scholar
  9. Rosseels  V, Marche  S, Roupie  V, Govaerts  M, Godfroid  J, Walravens  K, Members of the 30- to 32-kilodalton mycolyl transferase family (Ag85) from culture filtrate of Mycobacterium avium subsp. paratuberculosis are immunodominant Th1-type antigens recognized early upon infection in mice and cattle. Infect Immun. 2006;74:20212. DOIPubMedGoogle Scholar
  10. Doolan  DL, Freilich  DA, Brice  GT, Burgess  TH, Berzins  MP, Bull  RL, The US capitol bioterrorism anthrax exposures: clinical epidemiological and immunological characteristics. J Infect Dis. 2007;195:17484. DOIPubMedGoogle Scholar
  11. Norman  PS, Ray  JG Jr, Brachman  PS, Plotkin  SA, Pagano  JS. Serologic testing for anthrax antibodies in workers in a goat hair processing mill. Am J Hyg. 1960;72:327.PubMedGoogle Scholar
  12. Grunow  R, Porsch-Ozcurumez  M, Splettstoesser  W, Buckendahl  A, Hahn  U, Beyer  W, Monitoring of ELISA-reactive antibodies against anthrax protective antigen (PA), lethal factor (LF), and toxin-neutralising antibodies in serum of individuals vaccinated against anthrax with the PA-based UK anthrax vaccine. Vaccine. 2007;25:367983. DOIPubMedGoogle Scholar
  13. Zhang  Y, Qiu  J, Zhou  Y, Farhangfar  F, Hester  J, Lin  AY, Plasmid-based vaccination with candidate anthrax vaccine antigens induces durable type 1 and type 2 T-helper immune responses. Vaccine. 2008;26:61422. DOIPubMedGoogle Scholar
  14. Centers for Disease Control and Prevention. Vaccines and immunizations. [cited 2009 Aug 31]. Available from
  15. Demicheli  V, Rivetti  D, Deeks  JJ, Jefferson  T, Pratt  M. The effectiveness and safety of vaccines against human anthrax: a systematic review. Vaccine. 1998;16:8804. DOIPubMedGoogle Scholar




Cite This Article

DOI: 10.3201/eid1510.081717

Table of Contents – Volume 15, Number 10—October 2009

EID Search Options
presentation_01 Advanced Article Search – Search articles by author and/or keyword.
presentation_01 Articles by Country Search – Search articles by the topic country.
presentation_01 Article Type Search – Search articles by article type and issue.



Please use the form below to submit correspondence to the authors or contact them at the following address:

Pierre Wattiau, Veterinary and Agrochemical Research Centre, Department of Bacterial Diseases, Groeselenberg 99, B-1180 Brussels, Belgium

Send To

10000 character(s) remaining.


Page created: December 09, 2010
Page updated: December 09, 2010
Page reviewed: December 09, 2010
The conclusions, findings, and opinions expressed by authors contributing to this journal do not necessarily reflect the official position of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors' affiliated institutions. Use of trade names is for identification only and does not imply endorsement by any of the groups named above.