6 Anthrax

Anthrax is caused by the bacterium Bacillus anthracis, a Gram-positive aerobic spore-forming bacillus, primarily infecting herbivores. Although rare in the developed world, the organism remains a threat to livestock in African and Asian countries where control depends on appropriate animal husbandry approaches such as vaccination and disposal/decontamination of carcasses. Animals are thought to contract anthrax by ingesting spores from contaminated soil while humans become infected via contact with diseased animals, their products or as a consequence of acts of bio-terrorism such as occurred in 2001. This unprecedented act has stimulated a burst of research, shedding new light on the biology of the organism and its ability to cause disease. It is to be hoped that through this renewed interest anthrax will once again regain the status of an exotic disease of antiquity.

Anthrax has been a scourge of man and animals since the first written history of disease. It may have been one of the plagues of Egypt in the time of Moses (c. 1250 BC) and accounts of its symptoms can be found in the writings of ancient scholars such as Homer (c. 1000 BC) and Galen (c. AD 200) demonstrating the disease was well known to the Greeks and Romans. The earliest scientific (as opposed to historical) reports are the descriptions of malignant pustules by Maret in 1752 and of the disease in animals by Chabert in 1780 (Wilson and Miles 1946). The nineteenth century work on anthrax has more than usual significance, underpinning a major turning point in the history of medicine. It was the first disease of man (Woolsorter’s disease) and animals shown to be caused by a microorganism, enabling Koch to established his postulates in 1877 by proving Bacillus anthracis (named by Cohn in 1875) was the cause of anthrax. Subsequent researchers such as Greenfield and Pasteur in the early 1880s demonstrated the feasibility of using attenuated live vaccines to protect livestock, establishing histories second bacterial vaccine. At the turn of twentieth century Metchnikoff employed B. anthracis to characterize the ability of macrophages to kill microbes and helped establish the field of immunology.

During the twentieth century the scientific fruits of these endeavours, particularly the development and extensive use of animal vaccines, saw the status of the organism reduced to an exotic disease responsible for occasional outbreaks in animals and rare secondary infections in humans. It is ironic that while science developed the tools to manage the threat posed by anthrax, it also facilitated its development as a biological weapon.

While there are reports of the use of anthrax against cavalry horses during the First World War, serious efforts to develop anthrax weapons did not begin until the outbreak of World War II. In 1942 trials off the coast of Scotland demonstrated the feasibility of using B. anthracis spores as a biological weapon, and while the Allies sought to develop anthrax bombs they were never used in anger. The UK offensive program was terminated in the 1950s, with the USA program following suite in 1969. The Japanese and Soviets also had offensive biological weapons programs during the war and though the Japanese program terminated, the Soviet effort continued (Zilinskas 2006).

B. anthracis has been weaponized by various nations, most recently Iraq in 1991. The simplicity of the production technology and the availability of the organism in nature have made anthrax an attractive terror option for extremist groups such as the Aum Shinrikyo cult (Keim et al. 2001). The ability of a small scale attack to disrupt the infrastructure of a whole country was amply demonstrated by the 2001 postal attacks in the USA.

Bacillus anthracis is the only obligate pathogen within the genus Bacillus, comprising the Gram-positive aerobic or facultative anaerobic spore-forming, rod-shaped bacteria. It is a member of the genetically closely related ‘Bacillus cereus group’ comprised of B. cereus, B. anthracis, B. thuringiensis, B. mycoides, B. pseudomycoides, and B. weinhenstephanensis. However, it can be clearly identified by phenotypic traits such as virulence (which can be lost), lack of motility, absence of haemolysis on sheep and horse blood agar, susceptibility to penicillin, and sensitivity to the diagnostic gamma bacteriophage.

At the genetic level B. anthracis can be clearly distinguished by a range of DNA-based approaches, such as multilocus sequence typing (MLST), which targets chromosomal sequences (Priest et al. 2004). While snap shot techniques such as MLST enable the relationship of large numbers of isolates to be rapidly assessed, it is now feasible economically to determine the entire genetic sequence of the bacterium. Analysis of the genome of the Ames strain of B. anthracis revealed a 5.23 megabase chromosome essentially identical to B.cereus and B.thuringiensis, suggesting a common insect pathogen ancestor that acquired additional plasmid borne virulence factors (Read et al. 2003). The diversity of phenotypes within this group is often mediated by plasmid encoded factors. In the case of B. anthracis these comprise two plasmids, pXO1 (182 kb) and pXO2 (96 kb), both encoding major virulence factors (Fig. 6.1).The ability of these virulence plasmids to be transferred to other members of the B.cereus group has been reported. An example of this is B.cereus G9241, which was isolated from individuals presenting with an infection clinically indistinguishable from inhalational anthrax (Hoffmaster et al. 2004). Genetic analysis revealed the presence of a homolog to pXO1 and a second plasmid, which although genetically distinct from pXO2, encoded a phenotypically similar anti-phagocytic capsule.

Given the central role of virulence plasmids in pathogenicity and their ability to move between closely related strains we should take care in dismissing all clinical isolates of B.cereus as environmental contaminants.

Under conditions that favour growth, B. anthracis forms large, non-motile Gram-positive square ended rods which produce opaque, white, non-haemolytic colonies when grown on sheep blood agar. The colony has a rough surface with an irregular edge appearing filamentous when viewed using a hand lens. In contrast, if cultured in the presence of 5% CO₂ the organism forms a capsule (Uchida et al. 1993) and the resulting colonies are smooth and mucoid.

Unfavourable growth conditions, such as nutrient limitation, result in the formation of highly resistant, oval spores clearly visible in the centre of the bacilli in stained smears (Fig. 6.2). Spore formation is a survival strategy which enables the organism to persist for decades and survive exposure to physical and chemical insults. The spore is constructed as an internal double membrane-bound compartment, called the forespore, over the course of several hours. Dormancy and resistance depend in part on the partial dehydration of the inner compartment of the spore, known as the core, which houses the chromosome. This core is surrounded by a thick layer of peptidoglycan, known as the cortex, which is further enveloped by the spore coat. The outer most layer of the spore, which is separated by a gap, is called the exosporium.

The various proteinaceous layers which comprise the spore protect the cortex from damage by mechanisms which still remain unclear. It is possible that the exosporium and spore coat act as a permeability barrier to some chemicals or that toxic agents react with the various layers reducing the amount of agent available to attack essential molecules such as enzymes and DNA located in the spore core.

In addition to contributing to physical protection, the exosporium also contains biologically active enzymes thought to play a role in the intracellular survival of the vegetative bacterium within the phagolysosome of the macrophage (Kang et al. 2005). These include enzymes which subvert the production of antibacterial radicals, such as superoxide and nitric oxide, in addition to enzymes that regulate in vivo germination (Baillie et al. 2005; Raines et al. 2006; Weaver et al. 2007).

The ability to regulate germination is due to presence of enzymes such as alanine racemase and an inosine preferring nucleoside hydrolase. Alanine racemase, an inhibitor of germination, converts L-alanine to D-alanine, a form not recognized by the alanine specific germination receptor of B. anthracis. It has been proposed that spores employ this mechanism to prevent germination under sub-optimal conditions such as high spore numbers in soil or non-permissive environments encountered during infection (Hu et al. 2006; Titball and Manchee 1987).

There is debate concerning the ability of B. anthracis to replicate outside an infected host. As a consequence, it has been regarded as an obligate pathogen whose environmental presence reflects contamination from an animal source rather than self-maintenance in the environment. The ability of the organism to germinate and replicate greatly depend on the local conditions such as availability of germinants and nutrients, pH, temperature, moisture, and the presence of microbial competitors and predators such as bacteriophages. However, while replication is uncommon, germination, gene transfer, and spore formation can occur under certain environmental conditions such as those found in the immediate vicinity of plant roots and in organically rich soils (Saile and Koehler 2006; Baillie Labs, unpublished data). However, once spores have been formed, it is well established that they can survive for long periods, and the time interval between host infection can be decades.

While culture based methods have been developed to identify the presence of B. anthracis in environmental samples, they are time consuming and can lack sensitivity due to the need to eliminate other closely related members of the B.cereus group. For this reason rapid antibody and DNA based assays have been developed which target unique signatures associated with the major virulence factors (toxins and capsule), the chromosome and the spore surface. While these assays are considerably faster than culture, giving results in hours versus days, there are concerns over sensitivity and specificity—particularly against environmental samples.

An additional complication is that the organism is normally present as a spore in nature. This means that the majority of rapid DNA based assays include an additional step to crack open the spore and access the DNA target. The methodologies adopted to achieve this focus primarily on physical and chemical disruption and while effective, they can detect <10 spores in a sample, the multi-step processes required are labour intensive and time consuming.

Alternative approaches target spore surface located factors such as BclA, a major glycoprotein of the exosporium, and cellular debris derived from the mother cell which includes both protein and DNA encoding virulence factors. By employing the polymerase chain reaction (PCR) to detect surface located DNA one can identify as few as 10² spores within 2 hrs. Indeed it is now feasible to detect DNA on the surface of a spores in as little as 30 seconds (Kadir et al. 2007).

The organism has two major virulence factors, a tripartite toxin and an antiphagocytic capsule encoded by genes carried on two plasmids pXO1 and pXO2. The loss of either plasmid results in a marked reduction in virulence (Little and Ivins 1999). Plasmid pXO1 encodes a tripartite toxin comprising Lethal Factor (LF-776 amino acids) a metaloprotease, Oedema Factor (EF-767 amino acids) a cyclic AMP modulator, and Protective Antigen (PA-735 amino acids) the non-toxic, cell binding component responsible for transporting LF and EF into the cell. This toxin accounts for the majority of the pathology, while pXO2 encodes an antiphagocytic capsule composed of poly-D-glutamic acid, thought to inhibit uptake by immune effector cells such as macrophages (Makino et al. 2002).

The tripartite toxin follows the AB model where the A moiety is comprised of catalytic subunits LF and EF, and the B moiety, PA, translocates EF or LF into the cytosol. The B moiety is named due to its role as the key protective immunogen in the current human vaccine. It binds to ubiquitous cell surface receptors, two of which have been identified; anthrax toxin receptor/tumour endothelial marker 8 and capillary morphogenesis protein 2. Upon binding, PA is cleaved by the cell-surface protease furin to expose the A moiety binding site (Bradley et al. 2001; Rainey and Young 2004). Following proteolytic activation, PA forms a membrane-inserting heptamer that translates LF and EF into the cytosol (Petosa et al. 1997). The current working model (Leppla 1991) of in vivo toxin uptake is shown in Fig. 6.3.

Experimental evidence indicates that this is not the only model of toxin interaction and uptake (Panchal et al. 2005). PA and LF can form biologically active complexes in serum capable of killing susceptible macrophages.

Irrespective of how the toxin complex enters the cell, Lethal Toxin (LT), the combination of LF and PA, is the central effector of shock and death (Smith and Keppie 1954). The toxin contains a thermolysin-like active site and zinc-binding consensus motif HExxH, which acts as a Zn²⁺ metalloprotease on a range of substrates, including peptide hormones and Mitogen-Activated Protein Kinases (MAPK) (Duesbery et al. 1998; Pellizzari et al. 1999). The MAPK cascade is essential for full induction of the oxidative burst and pro-inflammatory cytokine expression, and its disruption neutralizes macrophage activation favouring bacterial escape from lymph nodes during the initial phase of infection (Baldari et al. 2006).

The combination of EF and PA results in oedema toxin (ET) causing oedema through the elevation of cellular cyclic AMP (cAMP) concentrations in affected tissues. Once in contact with the cytoplasm, EF binds calmodulin (a eukaryotic calcium-binding protein) becoming enzymatically active, converting ATP to cAMP. The effects are the same as those caused by cholera toxin, with intoxicated cells secreting large amounts of fluid (Leppla 1991).

The overall contribution of ET to the infective process is ill defined. It is generally considered that the pathological changes seen in infected animals are due to the LT and that these acute effects mask any cAMP-mediated responses. However, purified ET has been shown to inhibit chemotactic response of polymorphonuclear leukocytes and subsequent phagocytosis (O’Brien et al. 1985; Wright and Mandell 1986).

Both LT and ET are expressed soon after germination and suppress superoxide and nitric oxide production, which are key antibacterial killing mechanisms of the macrophage (O’Brien et al. 1985; Pellizzari et al. 1999). Following escape from the macrophage the two toxins target all the cells of the innate and adaptive immune system, subverting cell signalling so as to suppress their ability to mount a protective immune response against the bacteria, resulting in massive bacteremia and toxemia (Baldari et al. 2006).

As infection progresses the accumulation of toxin induces the development of cytokine independent shock. This is thought to be related to the direct injurious effects of LT on the endothelial cell function which ultimately contributes to death (Moayeri and Leppla 2004; Sherer et al. 2007).

While the toxins play an important role in virulence, to be fully pathogenic an infecting strain of B. anthracis must also produce a capsule. This structure is composed of a polypeptide, poly-D-glutamic acid which is thought to inhibit phagocytosis and opsonisation of the bacilli by virtue of its negative charge. The genes controlling capsule synthesis, capB, capC and capA, are present as an operon on the pXO2 plasmid. Their expression is in part regulated by serum, CO₂ and temperature via an unclear mechanism involving the product of the atxA and acpA genes, which also regulate toxin expression.

In addition to the major factors already described, the bacterium expresses other plasmid and chromosome encoded genes which contribute to the overall pathogenesis of the organism (Baillie and Read 2001). Differential expression of any or all of these genes may explain why wild type strains differ in virulence. Candidate virulence factors include chromosomally encoded extracellular proteases, phospholipases such as cereolysin and S layer proteins. Interestingly PlcR, a global transcription regulator of a virulence regulon thought to play a major role in insect virulence is defective in B. anthracis, possibly due to the acquisition of pXO1 (Mignot et al. 2001).

Infection commences subsequent to entry of the spore into the body by one of a number of routes and is followed by germination and multiplication locally or after transport to the regional lymph nodes. Germination in vivo plays a key role in pathogenicity; while spores germinate poorly in serum, the process is considerably more efficient within professional phagocytic cells such as alveolar macrophages. Recent data suggests that exposure to antibacterial free radicals generated within the phagolysosome triggers germination by inactivating negative regulators of germination such as alanine racemase located on the surface of the spore (Baillie et al. 2005).

In cutaneous infections, germination, multiplication and production of the toxin result in the characteristic eschar invariably accompanied by extensive oedema. Neutrophils, rather than macrophages, are the first white blood cells recruited to the site of a cutaneous infection (Mayer-Scholl et al. 2005). These cells are particularly adept at combating B. anthracis and their effectiveness is thought to account for why the majority of cutaneous infections spontaneously resolve.

Inhalational anthrax differs from cutaneous infection because inhaled spores are taken up by alveolar macrophages and pulmonary dendritic cells rather than neutrophils and are then carried to mediastinal lymph nodes. Following germination bacteria are released and lymphadenitis develops with minor, often unreported symptoms including malaise, mild fever and a mild cough. Bacteria frequently overcome the lymph-node filter, and enter lymphatic and blood circulation, causing massive bacteremia and toxemia. This blood borne phase is accompanied by major symptoms including fever, enlarged lymph nodes, pulmonary oedema with acute dyspnea (laboured respiration) and cynosis (bluish discoloration of the skin caused by poor blood oxygenation). As infection progresses accumulation of toxin and other bacterial derived factors inactivate the innate and adaptive immune systems, damage endothelial cell function and induce cytokine independent shock which ultimately results in death.

The primary role of toxin in causing death has been demonstrated experimentally in laboratory animals. There is an inverse relationship between susceptibility to infection and susceptibility to toxin. Guinea-pigs are highly susceptible to infection, yet quite resistant to toxin, while the opposite is true for rats. Toxin sensitivity could be linked to polymorphisms in a gene called Nalp1b which has been identified in toxin sensitive laboratory mice. The gene encodes a protein involved in the recognition of microbial components and danger-associated host molecules (Boyden and Dietrich 2006). In contrast to mice and rats, human and primate cells have been reported to be relative resistant to lethal toxin possibly due to variations in the coding sequence of the primate version of Nalp1.

Anthrax is primarily a disease of warm blooded animals such as herbivores, particularly human food animals, and has a worldwide distribution. Prior to the advent of effective animal vaccines it caused heavy losses in cattle, sheep, goats, horses and donkeys. In 1923 in South Africa it was estimated that 30, 000–60, 000 animals died of anthrax (Sterne 1967). Species vary in their susceptibility to different aspects of the infectious agent, for example, while dogs are relatively resistant to spore challenge they are extremely sensitive to anthrax toxin (Lincoln et al. 1967).

The pathogenic process in humans is, understandably, ill defined. The extrapolation of disease responses in experimental animals to man is complex, although the limited data available indicate that the infective process in humans is similar to that seen in animals (Phipps et al. 2004).

Healthy individuals can tolerate low level exposure to anthrax spores. A study of mill workers found that individuals could inhale 600–1,300 spores during the working day with no ill effect, although no indication was given as to the size of the particles inhaled (Dahlgren et al. 1960). This is an important factor—animals studies have demonstrated that to gain access to the body via the lungs, aerosol particles should be <5 µm. Attempts to use primate data to determine an infectious dose for humans has proved challenging, not least because of the outbred, diverse nature of the general population with regards to age, size, and health status. As a general guide, a dose of spores lethal for 50% of the individuals challenged via the aerosol route has been estimated at between 250–5,500 spores (Inglesby et al. 2002). It should be noted that this is an average dose, it is estimated that Ottilie Lundgren, aged 94, the last fatal victim of the anthrax postal attacks, was exposed to a single spore.

Sudden death in a herbivore without prior symptoms or following a brief period of fever and disorientation should lead to suspicion of anthrax, and bloody fluid exuding from the nose, mouth, or anus of the dead animal is particularly suggestive. In pigs and carnivores, local oedemas, particularly in the neck region, are pathognomonic signs. At death, in most susceptible species, the blood contains 10⁷–10⁹bacilli mL^-1, provided the animal has not been treated (numbers may also be lower in immunized animals which succumb to the disease). Pigs are noted for being an exception and the bacterium may be undetectable in their blood at death.

The blood of an anthrax victim clots poorly and usually the small volume of blood necessary for a diagnostic smear and culture can be drawn with a syringe from a vein in reasonably fresh carcass. Where that is not possible, a small piece of tissue, traditionally (but not necessarily) an ear clipping because of its high capillary content, preferably with signs of blood on it, can be excised and used to make a smear and for culture.

Smears should be stained with polychrome methylene blue (M’Fadyean’s stain); large numbers of blue-black-staining bacilli, often square ended and in short chains, surrounded by a clearly demarcated pink capsule is fully diagnostic. Specimens for culture should be submitted to the appropriate diagnostic laboratory (Defra 2007).

If anthrax is suspected, the carcass should not be opened thus avoiding environmental contamination. In pigs, where confirmation may depend on obtaining the relevant lymph nodes (submandibular, suprapharyngeal, or mesenteric) for culture, appropriate precautions should be taken before dissection to avoid environmental contamination. For differential diagnosis, blackleg, botulism, toxicosis (e.g. toxic plants, heavy metals, snake bite), lightning strike, and peracute babesiosis may cause symptoms similar to those of anthrax (Turnbull 1998).

Three forms of the disease are recognized in humans: cutaneous, inhalational and gastrointestinal. The latter two forms are regarded as being most frequently fatal due to being unrecognized until it is too late to instigate effective treatment (Dixon et al. 1999). An occasional complication is meningitis. Subsequent infections in the same individual are rare (Heyworth et al. 1975).

Human anthrax is frequently differentiated into non-industrial or industrial anthrax depending on whether the disease is acquired directly from animals or indirectly during the handling and processing of contaminated animal products. Non-industrial anthrax usually affects people who work with animals or animal carcasses, such as farmers, abattoir workers, knackers, butchers, and veterinary personnel, and is almost always cutaneous, although occasionally intestinal if, as occurs in developing countries, the owners’ skin, butcher, and eat the meat (Baillie 1999).

Industrial anthrax, occurring as a result of contact with contaminated animal products is usually cutaneous but has a higher chance of being inhalational through exposure to spore-laden dust. Analysis of the aerosolized spore content of a goat hair processing mill in the USA, where cases of inhalational anthrax had previously occurred, found that the spore load varied from day to day with workers being exposed to significant numbers of spores without ill effects suggesting that other factors contribute to susceptibility (Dahlgren et al. 1960). Indeed the unfortunate bongo drum maker in Scotland who contracted a fatal case of inhalational anthrax from untanned animals hides in 2006 suffered from an underlying health problem which is likely to have increased his susceptibility to infection (Anaraki et al. 2008).

Entry of infecting spores occurs via a lesion in the skin with a small pimple occurring 3–5 days later. Over the next 2–3 days, the centre of the pimple ulcerates to become a dry, black, firmly adherent scab, surrounded by a ring of vesicles, the typical anthrax eschar. Despite its angry appearance, there is little pain; pain and pus only develop if there is secondary infection. Lesions vary greatly in size from about 2 cm to several centimetres across and are accompanied by pronounced oedema, which can become life threatening if located on the face or neck. In uncomplicated cases, the eschar begins to resolve about 10 days after the appearance of the initial papule; resolution takes 2–6 weeks, regardless of treatment leaving little trace. Complications arise when the organism spreads to the bloodstream resulting in an overwhelming infection in ∼20% of untreated cutaneous cases. Diagnosis is made by M’Fadyean-stained smears and/or culture of pretreatment specimens of vesicular fluid obtained from under the edge of the eschar. For differential diagnosis, boil, orf, primary syphilitic chancre, erysipelas, plague, glanders, and tropical ulcer should be considered (Turnbull 1998).

In pulmonary (due to inhalation of spores) and intestinal (due to ingestion of contaminated meat) forms of anthrax, the illness has an incubation period of 1–6 days, during which nonspecific symptoms of fever, sweats, fatigue, dyspenea, non-productive cough, and nausea can occur (Abramova et al. 1993; Jernigan et al. 2002). These symptoms persist for 2 or 3 days, and in some cases there is a short period of clinical improvement. This is followed by the sudden onset of increasing respiratory distress with dyspenea, stridor, cyanosis, increased chest pain, and sweating. Respiratory distress is typically followed by rapid onset of shock and death within 24–36 hours. Mortality rates of 45 to 100% have been reported (Jernigan et al. 2002; Phipps et al. 2004). The recent ‘outbreak’ of human anthrax in the USA in 2001 saw 22 cases of laboratory confirmed anthrax, half of which were inhalational in nature with five cases proving fatal (45% case fatality ratio).

Early recognition of individuals with inhalational anthrax is difficult as its prodrome is similar to many common acute respiratory illnesses. While there are significant chest radiograph differences between inhalational anthrax and community—acquired pneumonia such as mediastinitis, none of the changes taken alone are highly predictive of infection with B. anthracis in the absence of evidence of previous contact with the organism (Kyriacou et al. 2007).

In recent years a number of laboratory based assays have been developed to supplement the diagnostic potential of M’Fadyean-stained smears and microbiological culture. These approaches include PCR based detection of genes encoding virulence factors, immunohistochemical and ELISA based detection of toxin proteins, and detection of enzymically active lethal factor protein in biological samples by mass spectroscopy (Walsh et al. 2007).

The characteristics of human infection, as seen by gross and microscopic observation, are largely similar to those seen in multiple animal models (for a review see Phipps et al. 2004). Studies of the pathology of human inhalation anthrax have focused primarily on mediastinal, hemi-lymphatic, and pulmonary changes. Events in the lung are unspectacular, being limited to haemorrhage, oedema, and atelectasis. This is due to it not being a primary site of infection but rather a portal of entry into the body, although secondary infection can occur once septicaemia has been established. Spores are taken up from the alveoli by macrophages and transported to a regional lymph node. Once there, they multiple and lyse their host cells, and escape in to the bloodstream to establish a systemic infection (Barnes 1947; Ross 1957). In contrast, mediastinal changes in humans are more pronounced, consisting primarily of oedema and haemorrhage with similar changes within the parenchyma of mediastinal lymph nodes which exhibit haemorrhage, necrosis, and the presence of Gram-positive bacteria as a consequence of lymphocytolysis. Vasculitis within the mediastinal lymph nodes is characterized by fibrinoid necrosis and infiltration by neutrophils and histiocytes.

The basic disease mechanism is vascular injury with oedema, haemorrhage, and thrombosis. Vascular injury is probably the result of toxin acting directly on the endothelial cell membrane, making them highly permeable to plasma, and causing adhesion of the leucocytes and platelets with widespread intravascular thrombosis (Dalldorf et al. 1971). Lesions in other organs beside lymphoid tissue, lungs, and brain have been described, but many appear to be secondary to shock and agonal changes.

Most strains are sensitive to penicillin, erythromycin, chloramphenicol, gentamicin, ciprofloxacin and tetracycline. It is important that chemotherapy is administered early regardless of the antibiotic chosen, due to the accumulation of toxin. Cutaneous anthrax responds well to treatment, however eschar formation may still occur due to the presence of toxin in the primary lesion.

In contrast, the treatment of inhalation anthrax is usually ineffective as the disease is rarely recognized prior to the onset of bacteremia. Once treatment has commenced it must continue for prolonged periods. Spores can persist in the lungs of infected primates for 60 days and reinitiate infection if treatment is prematurely terminated. Thus exposed individuals should be vaccinated on commencement of antibiotic treatment to enable antibiotic cover to be halted once protective immunity has be established (Friedlander et al. 1993).

Following the postal attacks in the USA ∼32, 000 individuals were given ciprofloxacin and then, when antimicrobial sensitivities became available, where encouraged to change to doxycycline (Jernigan et al. 2002). Amoxicillin was provided for pregnant women, breast feeding mothers, and children, due to concerns over the potential toxicity of the first two agents. While adverse events associated with antimicrobial prophylaxis were common (57%), serious events requiring hospitalization were rare (7%). Only 44% of over 10, 000 individuals recommended to complete a 60 day course of treatment did so, suggesting that relapse may be an issues in the event of future attacks (Shepard et al. 2002).

Numerous animal studies have demonstrated that inhibiting toxin activity prevents morbidity. The development of antitoxins to treat anthrax has been recently reviewed by Rainey and Young (2004). Approaches currently being pursued include antibodies (the approach closest to a product), receptor decoys, dominant-negative inhibitors of translocation, small molecule inhibitors and substrate analogues.

Until recently sub-clinical cases of anthrax in animals could not be diagnosed and as a consequence, infection was regarded as inevitably fatal. The development and application of antibody based assays have shown that sub-clinical infections do occur in animals that appear healthy. Humans are thought to be relatively resistant to anthrax when compared to herbivores, and data from cutaneously or orally exposed but untreated individuals suggest that sub-clinical infection is not uncommon (Heyworth et al. 1975).

Due to the scarcity of human infection there is little data concerning the long term health consequences of infection. Follow up of a significant proportion of the 22 infected individuals who survived the USA Postal attacks revealed that 53% had not returned to work a year after the event and that all were receiving psychiatric supports. Many of the survivors reported significant health problems, psychological distress, poor life adjustment, and loss of functional capacity (Reissman et al. 2004).

While few countries are truly enzootic for anthrax, most have some cases of the disease in their livestock in any one year. Countries experiencing relatively high incidence are those in sub-Saharan Africa, the Indian subcontinent and Indonesia, certain provinces of China, parts of Turkey, and various countries of the former USSR. The incidence of the disease declined dramatically in Britain due to the adoption of control measures such as vaccination and an increase in the use of man-made alternatives to animal products. Nevertheless, specialized leather and woollen industries continue to depend on hides and wool from particular species or breeds raised in countries where anthrax is still endemic (Turnbull 1998).

Epidemics are of the point source type with animals acquiring infection as a consequence of grazing on spore contaminated land. Animal to animal transmission appears to be a rare event and when it does occur it is believed to be mediated by biting flies. Outbreaks have been linked to environmental changes, particularly flooding, which may result in the redistribution and concentration of anthrax spores in particular areas. Little is known as to how animals become infected or indeed the factors that determine why some animals in the same herd survive while others succumb to infection. It is likely that underlying host specific factors are important. Experimental infection studies have shown that spore levels far higher than those encountered in nature are required to initiate infection events in susceptible, healthy animals.

Human-to-human transmission is exceedingly rare but exceptions have been recorded (Heyworth et al. 1975). As indicated earlier, man normally contracts anthrax directly or indirectly from animals. The most common, cutaneous form of the disease occurs as a result of spores gaining access to the body via a lesion. Workers who carry hides or carcasses on their shoulders are susceptible to infection on the back of the neck; handlers of other animal materials or products tend to be infected on the hands, arms, or wrists. Gastrointestinal infection can also occur as a result of the consumption of contaminated meat when the nutritional value of meat outweighs the perceived risks of serious illness from eating.

Control of anthrax for both livestock and humans lies in the well supervised disposal of infected animals carcasses, the application of biocides which reduce spore numbers to an undetectable level, and the immediate vaccination/and or prophylactic treatment of other members of the affected herd or at risk individuals. Although official recommendations in most countries are that anthrax carcasses be buried or burnt, the legacy of contaminated land from past burials (often decades ago) shows that incineration is the only truly satisfactory option. Mobile blowtorch incinerators are available but complete destruction of a bovine carcass can take more than 24 hours. Some countries prefer rendering, although the problem of preventing contamination of environment and equipment during transport and loading into the rendering plant has to be addressed.

Decontamination strategies for anthrax spores have centred on the use of toxic biocides (formaldehyde, chlorine releasing agents such chlorine dioxide and hydrogen peroxide) or gamma radiation. While effective these approaches suffer from the dual handicap of toxicity to man and the environment and/or are extremely expensive.

The application of control criteria designed for domestic animals to wildlife is impractical particularly with regards to the disposal of infected carcasses, many of which die unobserved. As to vaccination of wildlife in enzootic areas there are number of hurdles, not least of which is the argument that vaccination would interfere with the natural balance of the ecosystem. At present, immunization depends on the use of the livestock vaccines which requires either direct dart gun administration or immobilization of the animals followed by administration using a syringe. Either approach is expensive, traumatic for the animals, and can only give cover to a small core of susceptible animals. Also, it must be remembered that the duration of effectiveness of the vaccine is thought to be only about a year. Considerations are being given to the development of suitable oral vaccines for this purpose, although numerous obstacles must be overcome before oral vaccines satisfy concerns over safety, environmental contamination, and efficacy.

Vaccination is the most cost-effective form of prophylactic treatment. For this reason a considerable amount of time and effort has been expended on developing safe and effective animal and human vaccines.

The Sterne attenuated live spore vaccine which comprises a toxin expressing (pXO1+) but capsule deficient (pXO2−) isolate derived from a case of bovine anthrax has been employed extensively to control the disease in livestock (Turnbull 2000). Immunization of humans with live spores similar to the Sterne vaccine has been limited to the former USSR and China. The UK and the USA use non-living subunit vaccines based primarily on PA due to concerns over the possibility of residual virulence. The UK vaccine, which is similar in principle to its US counterpart, is produced from an alum precipitate of the cell-free culture filtrate of the Sterne strain of B. anthracis. In addition to containing large amounts of PA, the UK vaccine also comprises trace amounts of LF and other bacterially derived, immunogenic antigens, which have been shown to stimulate antibody responses in recipients and may contribute to protection and the transient side effects reported by some individuals (Baillie 2006).

Given the shortfalls of the current vaccine, research is in progress to develop a next generation replacement which will be fully defined and thus free from any adverse effects. In addition, vaccine formulations capable of self administration via the oral, nasal, or dermal routes, which induce immunity following a single dose and are stable at room temperature, would be extremely attractive to authorities seeking to build stockpiles to respond to a large scale future threat (Baillie 2006).