Bacteriology

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Anthrax in Humans and Animals. 4th edition. Geneva: World Health Organization; 2008.

Anthrax in Humans and Animals. 4th edition.

Show details

Contents

< Prev Next >

6Bacteriology

6.1. Description

B. anthracis, the causative agent of anthrax, is a Gram-positive, aerobic or facultatively anaerobic, endospore-forming, rod-shaped bacterium approximately 4 µm by 1 µm, although under the microscope it frequently appears in chains of cells. In blood smears, smears of tissues or lesion fluid from diagnostic specimens, these chains are two to a few cells in length (see Fig. 8A); in smears made from in vitro cultures, they can appear as endless strings of cells – responsible for the characteristic tackiness of the colonies (see Fig. 8B) and for the flocculating nature of broth cultures. Also characteristic is the square-ended (“box-car” shaped) appearance traditionally associated with B. anthracis vegetative cells, although this may not always be very clear. In the presence of oxygen, and towards the end of the exponential phase of growth, one ellipsoidal spore (approximately 2 µm by 1 µm in size) is formed in each cell; this does not swell the sporangium and is generally situated centrally, sometimes subterminally (see Fig. 8C).

In the absence of oxygen and under a high partial pressure of CO₂ in the presence of bicarbonate (HCO₃-), the vegetative cell secretes its polypeptide capsule (Fig. 8D; see section 5.5.1) and it is one of the two established in vivo virulence factors of B. anthracis. The capsule is also a primary diagnostic aid (see section 4.4.2, 4.4.4.2, Annex 1, sections 8–10 and Fig.8A).

6.2. Detection and isolation

In appropriate blood or other body fluid samples, or tissue specimens collected within a few hours of death from animals (see section 3.5) or humans with anthrax, B. anthracis is readily visualized in capsule-stained smears and readily isolated in pure culture. The same applies to smears of fluid from cutaneous lesions of humans prior to treatment (see section 4.4.2). Procedures are given in Annex 1, sections 8–10.

In old or decomposed animal specimens, or processed products from animals that have died of anthrax, or in environmental samples, detection is likely to involve a search for relatively few B. anthracis within a background flora of other bacteria, many of which will probably be other Bacillus species, in particular the closely related B. cereus. In this case, selective techniques are necessary. Procedures for the isolation of B. anthracis from such specimens are given in Annex 1, sections 8 & 10.

Such is the nature of the properties of B. anthracis that few agents which differentially select between B. anthracis and other Bacillus species do so in favour of B. anthracis, and those that do only do so unconvincingly. Of the selective media that have been proposed, the most successful historically is polymyxin-lysozyme-EDTA-thallous acetate (PLET – Fig. 8E and Fig. 14) agar (Knisely, 1966), although care should be taken to prepare this correctly (Annex 2). Success has also been noted with the trimethoprim and sulfamethoxazole blood agar formulation (Annex 2).¹ As yet no selective enrichment broth system has been devised for B. anthracis, although significant attempts have been made (Bowen, 1999) and, pending development of such a system, the sensitivity of in vitro detection of B. anthracis by conventional means in environmental samples or specimens from old or decomposed animals, or from processed animal products, is in the order of 10 cfu/g or /ml depending on the sample processing protocol employed (Manchee et al., 1981; Dragon & Rennie, 2001). Justification today for the use of laboratory animals to isolate B. anthracis is exceedingly unusual and confined to “last resort” circumstances.

PCR detection systems (Annex 1, section 10.7.4) have been developed for B. anthracis (Beyer et al., 1996; Patra et al., 1996a; Sjöstedt et al., 1996, 1997; Jackson et al., 1998), but it will probably be some time before they become totally stand-alone and generally available for use in the non-specialist laboratory. A positive by this method is close to being accepted on a stand-alone basis now for clinical specimens and simple environmental samples such as tap water and air samples. In case the reason for a negative by PCR is the presence of inhibitors, negatives may need bacteriological support. Clinical specimens and more microbiologically “complex” samples, such as faeces, muddy water or soil, will generally need a DNA extraction stage and will also need bacteriological support to confirm positives or negatives.

Rapid immunoassays for detection or confirmation of identity of B. anthracis are available commercially.² The absolute specificities of these are, as yet, unproven and positives using these generally need to be confirmed bacteriologically or by PCR, or both. The updated immunochromatographic device described by Burans et al. (1996) and used by Tubbesing (1997) and Muller et al. (2004) utilizes a monoclonal capture antibody to the anthrax protective antigen (PA) bound to a nitrocellulose membrane, and a second monoclonal antibody specific for a different epitope of PA bound to colloidal gold particles which become visible when they accumulate at the capture sites. The assay can detect as little as 25 ng/ml of PA and is performed in a few minutes without the need for special reagents. It therefore lends itself to direct diagnosis of cases of anthrax by detection of PA in the blood or body fluids, or to retrospective analysis of extracts from the types of sample of animal origin for which the Ascoli test was designed (Annex 1, section 11.1). As such, it has excellent specificity and sensitivity, but it has not become commercially available.

6.3. Identification and confirmation

6.3.1. Practical differentiation

With rare exceptions, it is generally easy to identify B. anthracis and to distinguish it from other Bacillus species, including other members of the B. cereus group (B. cereus, B. thuringiensis, B. mycoides, B. pseudomycoides and B. weihenstephanensis). For all practical purposes, an isolate is B. anthracis if it:

has the characteristic colonial morphology (Parry et al., 1983) on nutrient or blood agar (matt appearance, fairly flat, similar to B. cereus but generally rather smaller, more tacky, white or grey-white on blood agar, and often (but by no means always) having curly tailing at the edges (Fig. 8B)
is non-haemolytic or (very rarely) only very weakly haemolytic on sheep or horse blood agar;
is non-motile;
is sensitive to the diagnostic ‘gamma’ phage and penicillin (Fig. 8F);
is able to produce the capsule in blood or when cultured on bicarbonate-serum agar under CO₂ (Fig. 8D; Annex 1, section 10.8.2).

Biochemical tests or commercial identification kits can contribute virtually nothing to this for identifying B. anthracis, although they may usefully supply the identity of an isolate initially, but mistakenly, thought possibly to be B. anthracis. Despite this, conventional biochemical and physiological identification characters for B. anthracis are generally included in microbiological textbooks, although the evidence is that this rarely represents first-hand experience on the part of the author(s). In addition, there is an increasing tendency in textbooks for B. anthracis to be listed simply as a member of “the B. cereus group” stating that, apart from being non-haemolytic, non-motile and pathogenic, it exhibits the same physiological and biochemical identification characteristics as B. cereus. This leads to these texts being of limited practical value, and possibly even misleading, for the microbiologist encountering a suspicious isolate. For the record, however, conventional physiological and biochemical characteristics for B. anthracis are given in Table 3 and sections 6.3.1.1 to 6.3.1.14. The profiles given in the majority of textbooks are largely derived from the related milestone monographs of Smith et al. (1952) and Gordon et al. (1973), although it should not be forgotten that these had been preceded by some excellent texts dating back to the previous century (Slater & Spitter, 1898; Whitby, 1928; St John-Brooks, 1930; Knight & Proom, 1950).These characteristics have been rechecked (Turnbull, unpublished data, 2005).

TABLE 3

Primary physiological and biochemical characteristics of B. anthracis (all incubations at 35 °C).

B. anthracis databases have been developed for the commercial API³ and Biolog⁴ systems (Baillie et al., 1995; Cogne et al., 1996). The API 50CHB system for the identification of Bacillus species is based on acidification of 49 carbohydrates (and is designed to be – and should be – used in conjunction with miscellaneous tests in the API 20E strip). The Biolog system is based on “carbon source metabolic fingerprints” in 96-well trays. These supply useful characterization data for comparing B. anthracis with other Bacillus species but a read-out giving the identity of an isolate as B. anthracis must be confirmed with either the simple tests listed in the first paragraph of this section and/or the PCR (section 6.3.2 and Annex 1, section 10.7.4)

6.3.1.1. Spore and cell morphology

In blood or tissue smears from infected humans or animals, the bacteria are seen in short chains of two to a few cells, frequently square-ended (box-car) or “vertebrate” in shape (not apparent in Fig. 8A). In contrast, when grown in laboratory media, long chains are often apparent under the microscope, although with quite wide-ranging strain-to-strain and media-to-media differences. Cell size, or chain width, and partial loss of Gram-positivity at 24–48 hours on nutrient or sporulation agar also vary from strain to strain. The long chains (Fig. 8C) are the reason for the characteristic tackiness of B. anthracis colonies and for the stringy growth frequently encountered with B. anthracis in broth cultures. Turnbull (unpublished data) found that chaining could be selected out by passage through semisoft agar; the resulting cells in the passaged culture were present singly or in short chains or small clumps, and the colonies no longer had the tackiness of typical B. anthracis colonies.

Strain-to-strain differences can be expected in the degree of sporulation on sporulation agar at any given time period, and in the size and dimensions (length and breadth) of the ellipsoidal spores, which do not swell the sporangia (Fig. 8C). The spores usually appear within 48 hours and are often apparent after 24 hours, but several (5 or 6) days should be allowed before concluding that spores are not formed if they are not seen at 48 hours. Lipid globules are apparently not so readily visible in the protoplast of the vegetative cells as with at least some B. cereus.

6.3.1.2. Haemolysis

Fellows (1996) showed that B. anthracis was haemolytic on blood agar made with sheep red cells that had been washed with buffered saline containing calcium and magnesium, and Popov et al. (2004) refer to the haemolytic activity and haemolytic proteins of B. anthracis. Similarly, the anthrolysin O of Shannon et al. (2003), Mosser et al. (2005) and Thomason et al. (2005) is presumably haemolytic. Reports are also occasionally encountered of haemolysis in blood or on agar containing blood of certain species, including human. It is unclear whether, when seen, this is equivalent in appearance to the strong haemolysis of B. cereus. (See also section 6.4.1.)

6.3.1.3. Lecithinase

B. anthracis either produces lecithinase to a lesser extent than its close relatives, B. cereus and B. thuringiensis, or produces a lecithinase with a lower activity (McGaughey & Chu, 1948; Zwartouw & Smith, 1956b). In the lecithinase test on egg yolk agar, the zone of opalescence or “halo” almost always seen around colonies or areas of growth of B. cereus and B. thuringiensis is only sometimes visible around B. anthracis, probably only becoming apparent at 48 hours (35–37 °C) and usually in a narrow band when present. Opalescence should be looked for under the colony/area of growth by scraping away some of the colony material. It can be seen here after 24 hours incubation at 35–37 °C. While B. anthracis grows well on conventional egg yolk agar, it grows less well than B. cereus on Kendall’s BC egg yolk-mannitol agar (Parry et al., 1983); the growth of B. anthracis on this medium (24–48 hours) is greyish as compared to the deep purple of B. cereus and, in contrast to B. cereus, a zone of opalescence does not form around the growth of B. anthracis. Once again, colony material must be scraped away to see the underlying LV (lecithovitellin) reaction.

6.3.1.4. Motility

As reviewed by Sterne & Proom (1957), a number of claims exist in the early literature of motile forms of B. anthracis. Sterne & Proom considered that most, if not all, of these claims could be explained in terms of contaminants or misidentification. Logan & Berkeley (1984) and Logan et al. (1985) found that, while 144 of their 149 strains of B. cereus were motile, none of 37 strains of B. anthracis they examined were motile. Although examination for motility is always listed as one of the primary identification tests for B. anthracis, it is doubtful that the test is often done on new isolates or checked more than rudimentarily and occasionally on culture collection cultures. Certainly, the first obvious appearance for diagnostic test purposes is lack of motility, but in view of the existence of genes associated with motility (see section 6.4.1), and even one recent report (Liang & Yu, 1999) that includes electron micrographs showing flagella, perhaps the “absoluteness” of non-motility in B. anthracis should be revisited.

6.3.1.5. Diagnostic phage

An anthrax phage was described as early as 1931 (Cowles, 1931). McCloy (1951) described a phage, designated phage W, isolated from strain W bacillus which she believed to be an atypical B. cereus or unusual B. anthracis strain. The “gamma” phage was first recorded by Brown & Cherry (1955) as a “new variant isolated from the original strain W bacteriophage”. This phage lysed all of 41 B. anthracis strains, but none of 89 B. cereus strains or 134 strains representing 8 other Bacillus species. The subsequent work of Buck et al. (1963) in the early 1960s with phages isolated from 25 lysogenic B. anthracis strains, and the work 30 years later by Redmond et al. (1996b) with almost identical results using a set of 25 phage isolates from different cultures of B. anthracis, have indicated that either there exists a family of closely related phages, of which the gamma phage is one, or that selective pressures have resulted in variants of one phage with slightly different properties.

In so far as the phages are spontaneously produced and plaques appear on culture lawns in an unpredictable manner, they are considered temperate. However, it has long been noted (McCloy, 1951, 1958; Buck et al., 1963) that, in being able to lyse their own original host and in that they can be grown in their homologous host, this gamma family “provides a striking exception to the rule of immunity” (McCloy, 1958). Buck et al. (1963) termed this phenomenon “dismunity” and seemed to attribute it to the formation of mutant forms of the phages.

Just to what extent the diagnostic phages used in various laboratories around the world and often referred to as “gamma phages” are unadulterated descendants of McCloy’s gamma phage is now uncertain. For this reason, the term “diagnostic phage” is preferred.

Some care needs to be taken in defining precisely what is meant by phage sensitivity. The titre of the phage suspension is of consequence, and interpretation of what is seen in the zone of effect is somewhat subjective. For the purposes of Table 3, the titre of the phage suspension is ≥ 10⁹ pfu/ml and any zone of effect, which may be graded from ± to 4+, represents sensitivity. (Note: this is different from the manner in which antibiotic sensitivity is read, where colonies growing in a zone of clearing result in an interpretation of “resistant”.) For phage sensitivity tests, a total absence of effect is regarded as true resistance. On this basis, in formal studies on susceptibility/resistance, Buck et al. (1963) found that 7 of 264 (2.7%) B. anthracis isolates were phage-resistant, and 3 of 64 (3.1%) non-anthrax Bacillus species were lysed by anthrax phage. Similarly, Redmond et al. (1996b) found that 1 of 87 (1.2%) strains of B. anthracis were phage-resistant, and 2 of 47 (4.2%) non-anthrax Bacillus species were lysed by anthrax phage. In further tests (unpublished), Turnbull and colleagues found no positives among 14 B. cereus, 10 B. megaterium, 6 B. pumilus, 5 B. subtilis, 5 B. circulans, 4 B. mycoides, 4 B. firmus, 4 B. sphaericus, 3 B. licheniformis, 3 B. thuringiensis, 2 B. amyloliquefaciens, 1 B. lentus, and 2 Brevibacillus brevis from collections at the Food Safety and Microbiology Laboratory, Health Protection Agency, Colindale, London, United Kingdom, and the Department of Biological and Biomedical Sciences, Glasgow Caledonian University, Glasgow, United Kingdom. This is also supported by the unpublished results of tests on many hundreds of isolates from routine environmental samples performed over a period of years by Turnbull and colleagues, in which only three phage-resistant B. anthracis isolates were encountered (from a tannery dump site in the United Kingdom, from a human case of anthrax in Zimbabwe, and from three animals in the Etosha National Park, Namibia). Likewise, the finding of a phage-sensitive non-anthrax Bacillus species was a very rare event.

Subsequent examination of the three “phageresistant” B. anthracis isolates retained in a culture collection indicated that this resistance was not so clear-cut, suggesting it had represented some type of lysogenic immunity at the time of isolation which has, since then, become “dismunity” (see above, this section) (Turnbull, unpublished observations).

It should be noted that these phages do not lyse capsulated B. anthracis cultures (McCloy, 1951; Meynell, 1963; Turnbull, unpublished data) and only attack vegetative cells during the multiplication phase. As discussed in section 2.1.2.3, the vegetative form of B. anthracis rarely exists or multiplies outside the animal host, and therefore is rarely exposed to exogenous phage. This raises the following question: what is the evolutionary role of the anthrax phage? One possibility is that lysis is a laboratory-induced side-effect and is not representative of the true and yet-to-be elucidated role of this virus, or group of viruses, in the life-cycle of B. anthracis.

6.3.1.6. Capsule

Spontaneous loss of ability by virulent B. anthracis to produce the capsule has been recognized since at least Sterne’s observations (1937a, 1937b) and it was out of these observations that Sterne’s livestock vaccine was developed (sections 5.5, 8.6.2). Chu (1952) attached numerical values to this, finding that laboratory stock cultures put out noncapsulated variants to extents ranging from 0.14% to 32.4%. This appears to result from spontaneous loss of pXO2, although the triggering event and point in the life-cycle at which this occurs has yet to be elucidated. The not infrequent presence of pXO1+/2- (and occasionally pXO1-/2-) forms of B. anthracis in environmental sites and samples with histories of potential anthrax contamination was noted by Turnbull et al. (1992b) and Turnbull (1996), and was attributed to natural curing of one or both plasmids by unidentified environmental stresses.

6.3.1.7. Voges-Proskauer reaction

Although definitely VP-positive, some difficulty may be experienced in this test due to occasional reluctance by B. anthracis to grow in glucose-phosphate broth. In the event of a negative test, the broth should be checked to determine whether growth took place, or whether the culture died out. Positivity manifests itself as a gentle rust colour as compared with the much deeper red of an Enterobacter cloacae control.

Logan (personal communication, 2005) found the VP test to be inconsistent within the genus Bacillus using the commercial API 20E strip, no matter how well controlled the variables were, but all B. anthracis strains were positive as compared with 93% of 149 B. cereus strains being positive (Logan & Berkeley, 1981; Logan & Berkeley, 1984; Logan et al., 1985).

6.3.1.8. Citrate utilization

This is given as “variable” (as compared with “positive” for B. cereus) by Gordon et al. (1973), Parry et al. (1983), Turnbull et al. (1990a) and Turnbull & Kramer (1991), as well as various editions of Bergey’s Manual of systematic bacteriology. Elsewhere (Turnbull & Kramer, 1995), B. anthracis and B. cereus are combined under “B. cereus group” and citrate utilization is again inscribed “variable”. In fact, using the procedure of Parry et al. (1983), in which slants are stabbed and streaked with a needle dipped into a saline suspension of culture with turbidity equivalent to MacFarland standard 0.5, B. anthracis appears to be unable to survive and is negative in this test.

Using the commercial API 20E system, Logan et al. (1985) and Logan & Berkeley (1984) found all 37 of their strains of B. anthracis to be negative as compared with 89% of their 149 B. cereus strains being positive for this character.

6.3.1.9. Growth in the presence of 5% and 7% salt

Textbooks generally link B. anthracis together with B. cereus in the “B. cereus group” as growing in the presence of 7% salt. In fact, B. anthracis generally seems to be reluctant to grow in the presence of 5% salt. One or two strains may show a little growth in the presence of 5% salt, but this is much reduced compared to the 0% control. Others will not show visible growth. Whether this growth is seen or not, the cultures in 5% (and 7%) salt tend to become non-viable within a week.

6.3.1.10. Nitrate

Kiel et al. (2000) discuss the fact that B. anthracis maintains a strong nitrate reductase activity and, unlike other facultative anaerobes, the presence of oxygen does not prevent the utilization of nitrate. They relate this to the initiation of sporulation through nitration of the SpoOA gene product when the vegetative cells shed by the infected animal encounter the oxygen of the air.

Logan et al. (1985) and Logan & Berkeley (1984) found all 37 of their strains of B. anthracis to be nitrate-positive when tested using the API 20E systems compared with only 81% of their 149 B. cereus strains.

6.3.1.11. Casein

Strain differences can be expected in the size and nature of the haloes which, after 48 hours at 35 °C, range from narrow (1–2 mm) opaque zones around the growth to totally translucent zones of about 4 mm surrounded by opaque zones of approximately 2 mm width.

6.3.1.12. Starch hydrolysis

As with casein hydrolysis, strain differences can be expected in the extent and visibility of the hydrolysis. Plates should be left 48 to 96 hours before reading. Rapid colony spreading occurs without alterations of these readings beyond this point. Flooding the plate with a 1:5 dilution of Lugol’s iodine solution is generally necessary to show up the hydrolysis.

6.3.1.13. Phenylalanine deamination

Gordon et al. (1973) and other texts, including various editions of Bergey’s Manual of systematic bacteriology, specify that B. anthracis (and B. cereus) fails to deaminate phenylalanine. In tests done by Turnbull (unpublished) using Method 2 in Cowan & Steel’s identification manual (Cowan, 1974), greening of the agar surface was found with 6 of 12 B. anthracis cultures and browning in a further 5. These were interpreted as positive tests.

6.3.1.14. Propionate utilization

The results of this test can vary with inoculum size. Following the procedure of Parry et al. (1983) described in section 6.3.1.8, B. anthracis is negative in this test. Positives may result in an inconsistent manner when larger inocula are used. In fact, in an international reproducibility trial of standardized characterization tests for Bacillus, Logan & Berkeley (1981) judged this test to be inherently unreliable.

6.3.2. PCR

PCR is becoming more widely available as a means of confirming the presence of the virulence factor (capsule and toxin) genes, and hence that an isolate is, or is not, virulent B. anthracis. For routine purposes, primers to one of the toxin genes (usually the Protective Antigen gene) and to one of the enzymes mediating capsule formation are adequate (Annex 1, section 10.7.4). In laboratories not equipped for PCR tests, if doubt remains at the end of the procedures given in Annex 1, sections 10.7.1 to 10.7.3 as to the definitive identity of a suspect B. anthracis isolate, inoculation into a mouse or guinea-pig may be the only way remaining to determine conclusively if it is virulent B. anthracis (Annex 1, section 12). However this should be a last resort procedure and confined to situations where a definitive identification is essential.

6.4. Molecular composition

6.4.1. The genome

The B. anthracis chromosome is a circular DNA molecule of 5 227 293 base pairs encoding 5508 predicted protein-coding sequences (Read et al., 2003).

Elucidation of the genome sequences of B. anthracis, B. cereus and B. subtilis has revealed that the majority of B. anthracis genes are homologues of genes found in B. cereus and other closely-related species. These include genes encoding such testable phenotypes as haemolysins, enterotoxins, phospholipases, beta-lactamases and other enzymes, and even genes associated with motility (Read et al., 2003). Lack of expression by B. anthracis of many phenotypes appears to lie primarily in a truncation of the PlcR positive regulator gene, which has been shown to increase production of many of these enzymes. In the case of motility, there are also deletions of genes in the flagellum operon. Low expression of lecithinase and haemolytic activity, or expression under specific inducing conditions, are explained in terms of alternative regulatory mechanisms (Klichko et al., 2003; Read et al., 2003).

The AtxA transcriptional regulator of the anthrax toxin genes located on the pXO1 plasmid (see section 5.5) appears to be a master regulator of genes on both the plasmids and chromosome of B. anthracis (Bourgogne et al., 2003). There is some evidence that PlcR and AtxA may be incompatible (Mignot et al., 2001, 2003).

While homologues of the plasmid-encoded toxin and capsule genes and their associated regulatory loci were not found on the B. cereus ATCC14579 (type strain) genome, the presence in other B. cereus strains of genes with pXO1 and pXO2 sequence identities supports evidence of mobility on the part of the anthrax plasmid genes (Read et al., 2003). Thus, comparative genome analysis was supportive of the hypothesis that B. anthracis evolved from a B. cereus ancestor. Other than the acquisition of the pXO1 and pXO2 plasmids, the major phenotypic differences between B. anthracis and its relatives appear, therefore, to be the result of altered gene expression rather than absence or presence of specific chromosomal genes (Read et al., 2003).

6.4.2. Strains

See section 2.3.

6.5. Spores

The position of the spore as the centre of the cycle of infection of B. anthracis and the environmental factors influencing sporulation, germination and survival of the emergent cells are discussed in section 2.1.2. From the public health and veterinary public health standpoints, the rate of sporulation of vegetative forms released by the animal that has died of anthrax is probably of more relevance than the rate of germination.

6.5.1. Sporulation

Reports on conventional studies of sporulation rates of B. anthracis were not found, but there is little reason to believe that the rate varies greatly from other Bacillus species. In a culture with synchronized growth, the seven stages of sporulation commence at the top of the exponential phase of growth and are complete 7–8 hours later (Doi, 1989; Nicholson & Setlow, 1990). Resistance to UV (ultraviolet) locks in at about 4–6 hours (stages 4 and 5) and resistance to heat and chemicals at around 5–7 hours (stages 5 and 6) after onset of the sporulation process. These are under optimal conditions; under suboptimal conditions it can take much longer (Davies, 1960). In a field study (Lindeque & Turnbull, 1994), the earliest time at which spores were detected in the soil at a site in the Etosha National Park (Namibia) where a springbok had just died was 8 hours; by 24 hours, spores accounted for 75% of the total count, and complete sporulation was achieved between 32 and 48 hours. The findings in soils and waterhole water inoculated in the laboratory with blood collected just after death are given in section 2.1.2.4, and other information on sporulation can be found in sections 2.1.2.2 and 2.1.2.4.

6.5.2. Germination

The process of germination is dependent upon the action of a germinant on a trigger site within the spore. Common germinants include L-alanine and ribosides, and sometimes germination can be induced by combinations of chemicals that are not germinants on their own (McCann et al., 1996). “Heat-shocking” the spores, i.e. exposing them to 60 °C–70 °C for several minutes (10–30), predisposes them to germination and results in more synchronous germination within a spore population. Heat-shock temperature-time combinations in the literature vary from 60 °C for 90 minutes to boiling for one minute. Turnbull et al. (2006) found evidence of damage to B. anthracis spores at 80 °C for 10 minutes, and almost complete killing at 90 °C for 10 minutes.

Germination is a rapid process. Fernelius (1960) found that, following germination of B. anthracis spores by alanine, tyrosine and adenosine and, as measured by loss of heat resistance, > 99% germination had occurred in 8 minutes at 30 °C and in 16 minutes at 15 °C. Based on loss of refractility and resulting optical density change, Titball & Manchee (1987) recorded germination as a rather slower process, with 61% to 63% of spores germinating in 90 minutes under optimal conditions, which consisted of low spore concentrations at 22 °C. This optimal temperature is some 10 °C lower than that recorded for other Bacillus species; the authors considered that it might represent selection for the best conditions for establishing cutaneous infection. Studying germination of B. anthracis spores in sera from 11 species, they found that serum alanine ranged from 0.01 to 0.35 mmol/l, and serum tyrosine from 0.02 to 0.08 mmol/l. Germination rates varied in the different species. They concluded, therefore, that there was no evidence of a relationship between germination ability and the innate resistance of an animal to anthrax, as had been proposed by Hachisuka (1969).

Lincoln et al. (1967) similarly failed to correlate spore germination within phagocytes with susceptibility of an animal to anthrax.

6.6. Other surface antigens: anthrax-specific epitopes and detection methodology

Reliable detection of B. anthracis in the environment has traditionally depended on selective culture and identification of the emergent bacilli. This is a slow process (two days or more) and there has long been a desire, particularly in relation to bioaggressive scenarios, for more instant methods of detecting anthrax spores. Numerous attempts were made in the 1960s, 1970s and 1980s to develop antigen-based rapid, species-specific spore-detection systems, but invariably cross-reactivity with other common environmental Bacillus species proved insurmountable. Hope that monoclonal antibodies to anthrax spore-specific epitopes might provide reliable bases for rapid detection tests arose in the late 1980s (Phillips et al., 1988). Progress on finding and validating monoclonal antibodies that had the necessary specificities, sensitivities, affinities, etc., for application in detection systems appears to have been slow, with the next published report being in 1998 (Park et al., 1998), and the most recent being that of Mangold et al. (2005) recording the development of anthrax-specific monoclonal antibodies to spore and vegetative cell antigens. Rapid detection being a major issue for defence concerns, it is certain that unpublicized systems are in use.

Claims now exist that anthrax spore-specific epitopes are present on at least one immunodominant exosporium protein (see section 5.5.6).

The target epitopes of some anthrax-specific monoclonal antibodies proposed or used for detection of anthrax spores are unidentified, but suspected of residing in the vegetative cell S-layer extractable antigen EA1 (section 5.5.5) retained on the spores. For general and veterinary public health detection purposes, antibody-based systems utilizing such monoclonals are likely to be satisfactory. In bioaggression scenarios, the possibility that the determined aggressor may have effectively cleaned his/her spores may have to be borne in mind.

6.7. Transport of clinical and environmental samples

Movement of infectious or contaminated materials from the site of origin to a diagnostic or reference laboratory, both within a country and internationally, presents a risk of spread of diseases if the materials inadvertently escape into the environment during transit.

The international regulations for the transport of infectious substances by any mode of transport are based upon the recommendations made by the Committee of Experts on the Transport of Dangerous Goods (UNCETDG), a committee of the United Nations Economic and Social Council. The recommendations are presented in the form of Model Regulations. The United Nations Model Regulations (UNMR) are reflected in international law through international model agreements.⁵

WHO has developed a document to help understand the current regulatory framework and support compliance with current international regulations for the transport of infectious substances and patient specimens by all modes of transport, Guidance on regulations for the transport of infectious substances 2007–2008, applicable as of 1 January 2007 (WHO, 2007). This document is regularly updated to reflect modifications made to relevant sections of the UNMR.

The reader is encouraged to refer to this WHO document for details on transport of clinical and environmental samples.

Footnotes

1: http://www.ourfood.com/Anthrax.html.
2: Examples can be found on www.osborn-scientific.com, www.responsebio.com and www.tetracore.com, and from New Horizons Diagnostics,Columbia, MD 21045, USA; e-mail: moc.loa@gaiDHN.
3: BioMérieux SA, 69280 Marcy-l’Etoile, France.
4: Hayward, California 94545, USA.
5: http://www.unece.org/trans/danger/danger.htm.

All rights reserved. Publications of the World Health Organization can be obtained from WHO Press, World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland (tel.: +41 22 791 3264; fax: +41 22 791 4857; e-mail: tni.ohw@sredrokoob). Requests for permission to reproduce or translate WHO publications – whether for sale or for noncommercial distribution – should be addressed to WHO Press, at the above address (fax: +41 22 791 4806; e-mail: tni.ohw@snoissimrep).

Bookshelf ID: NBK310474

Contents