Virulent spores of Bacillus anthracis and other Bacillus species deposited on solid surfaces have similar sensitivity to chemical decontaminants


Jose-Luis Sagripanti, US Army RDECOM Attn: AMSRD-ECB-RT Bldg. 3150 Aberdeen Proving Ground, MD 21010-5424. E-mail:


Aims:  To compare the relative sensitivity of Bacillus anthracis and spores of other Bacillus spp. deposited on different solid surfaces to inactivation by liquid chemical disinfecting agents.

Methods and Results:  We prepared under similar conditions spores from five different virulent and three attenuated strains of B. anthracis, as well as spores of Bacillus subtilis, Bacillus atrophaeus (previously known as Bacillus globigii), Bacillus cereus, Bacillus thuringiensis and Bacillus megaterium. As spore-surface interactions may bias inactivation experiments, we evaluated the relative binding of different spores to carrier materials. The survival of spores deposited on glass, metallic or polymeric surfaces were quantitatively measured by ASTM standard method E-2414-05 which recovers spores from surfaces by increasing stringency. The number of spores inactivated by each decontaminant was similar and generally within 1 log among the 12 different Bacillus strains tested. This similarity among Bacillus strains and species was observed through a range of sporicidal efficacy on spores deposited on painted metal, polymeric rubber or glass.

Conclusions:  The data obtained indicate that the sensitivity of common simulants (B. atrophaeus and B. subtilis), as well as spores of B. cereus, B. thuringiensis, and B. megaterium, to inactivation by products that contain either: peroxide, chlorine or oxidants is similar to that shown by spores from all eight B. anthracis strains studied.

Significance and Impact of the Study:  The comparative results of the present study suggest that decontamination and sterilization data obtained with simulants can be safely extrapolated to virulent spores of B. anthracis. Thus, valid conclusions on sporicidal efficacy could be drawn from safer and less costly experiments employing non-pathogenic spore simulants.


Bacillus spores are among the life forms most resistant to inactivation, with examples of spores revived from amber 25–40 million old (Cano and Borucki 1995) or from brine inclusions dated 250-million years old (Vreeland et al. 2000). Spores of Bacillus anthracis have been considered to be potentially effective biological weapons, and at different times this pathogen has been included in the biological arsenals of several nations (Sherman 1995). The resilience of spores of B. anthracis can make the decontamination of surfaces very difficult, making imperative the availability of chemical disinfectants whose efficacy is well known. Remediation of contaminated buildings after the delivery of anthrax spores via the US mail (Dewan et al. 2002) involved multimillion-dollar budgets (with the Trenton and Brentwood postal facilities decontaminated at a cost estimated in $200 million; reviewed in Canter 2005). Scientific issues and commercial considerations promoted a recent increase on the number of products that reportedly inactivated B. anthracis spores. However, the vast majority of these products have been tested against Bacillus spores others than B. anthracis (Spotts-Whitney et al. 2003).

Many genes encoding structural and regulatory proteins are similar in all Bacilli (Driks 2002). In particular, Bacillus subtilis and Bacillus atrophaeus (formerly named Bacillus globigii) spores are extremely similar because of their close phylogenic relationship (Priest 1993). However, there are structural and molecular differences between spores of B. anthracis and B. atrophaeus or B. subtilis spores. These differences could be important as B. atrophaeus or B. subtilis are generally used as simulants of B. anthracis in decontamination studies. Spores of B. anthracis differ from spores of B. subtilis and B. atrophaeus in the composition of proteins in the outer coat (Driks 2002; Kim et al. 2004). In addition, spores of B. anthracis are surrounded by an exosporium which is absent in spores of B. subtilis or B. atrophaeus. These differences in outer coat composition and in the presence or absence of exosporium could potentially result in differences in sensitivity to chemical inactivation between B. anthracis and B. subtilis or B. atrophaeus.

Given less stringent biosafety requirements, abundant data are available on decontamination of spores derived from non-pathogenic Bacillus species (reviewed in Block 2001). Bacillus spores exposed to biocides in commonly used sporicidal formulations, including glutaraldehyde, formaldehyde, peracetic acid, hydrogen peroxide, chlorine, phenol and heavy metals showed various degrees of inactivation, from a relatively high level (reducing spore contamination by one-million fold which is considered a 6 log reduction) or more, to practically negligible (with survival similar to spores exposed to water as a control) (Sagripanti 1992; Sagripanti and Bonifacino 1996a,b; 1997). Data on the relative efficacy of various sporicidal commercial products on Bacillus spores suggested that commercial liquid sterilants and disinfectants were less effective on contaminated surfaces tan generally acknowledged (Sagripanti and Bonifacino 1999).

Information on the inactivation of B. anthracis spores is largely derived from the effect of chlorination treatment on spores in suspension. An earlier report suggested that B. atrophaeus spores in suspension could be more resistant to chlorine than B. anthracis (Brazis et al. 1958). Additional studies have suggested slight differences in sensitivity to chlorine between spores suspensions of B. anthracis Ames strain (virulent) and the attenuated Sterne strain (Rose et al. 2005). Differential sensitivity has also been reported between B. anthracis Sterne spores and spores of Bacillus thuringiensis or B. anthracis Ames strain (Rice et al. 2005). It is difficult to correlate previous data obtained with spores in liquid suspensions to the sensitivity of dry spores on contaminated surfaces as it has been shown that some bacteria are on average 300-fold more resistant to germicides when deposited on contaminated surfaces than in suspension (Sagripanti and Bonifacino 2000).

A review by the Centers for Disease Control and Prevention on available data from 1930 to 2002 made evident the lack of quantitative data comparing the sensitivity of B. anthracis spores to that of other Bacillus spores grown and analysed under similar conditions (Spotts-Whitney et al. 2003). In addition, (i) the use of spore preparations containing vegetative bacteria or germinated spores, (ii) the potentially different binding to and recovery from carrier materials, and (iii) the use of methods that do not account for all challenged spores or that have unknown recovery may further compromise the limited information available.

It remains unclear whether decontamination protocols used in building and environmental remediation or in medical sterilization/disinfection procedures to be used after a biological attack will be effective in inactivating spores of B. anthracis. Great savings in effort and speed in the development of knowledge and countermeasures could be accomplished if all members of the Bacillus family were shown to have similar sensitivity to sporicidal agents. In contrast, grave risk would be taken if assumptions drawn from experiments with simulants proved not to be valid for pathogenic anthrax. The goal of this study was to compare the sensitivity of virulent and attenuated spores of B. anthracis, as well as to establish the relative sensitivity of other Bacillus spores grown under similar conditions to inactivation by chemical agents that may be used to decontaminate civilian and military assets after a biological attack.

Materials and methods


Decon-Green consisting in a mixture of 0·090 g of K2CO3, 0·024 g of K2MoO4, 1 ml of 50% H2O2, 2·8 ml of propylene carbonate and 1 ml Triton X-100 was prepared and used undiluted as previously described (Wagner and Yang 2002). Sodium hypochlorite 6% (commercial Clorox, The Clorox Company, Oakland, CA, USA) was diluted with distilled water and used at a concentration of 5% (v/v chlorine, without adjusting pH) as recommended in the Handbook Medical Management of Biological Casualties (Eitzen et al. 1998). DF100 and DF200 are formulations developed by Sandia National Laboratory, US patent number 6566·574 B1 and commercialized by EnviroFoam Technologies, Inc. (Huntsville, AL, USA). These products were used as recommended by the manufacturer on the product label (



Black rubber material was obtained from the exterior and interior of the face piece of M-40 series military gas protective masks (meeting ECBC/US Army Specification EA-F-1379). The rubber material is made of a proprietary silicone and butyl rubber blend, formulation ‘2J02’ produced by ILC Dover Corporation (Frederica, DE 19946-2080) or formulation ‘2G06’ manufactured by Mine Safety Appliances (Pittsburgh, PA, USA). A number of protective masks were randomly selected, marked with a ruler and cut into 5 × 5 mm using a pair of scissors. The coupons were washed with ethanol (70%) and rinsed with distilled water before storing them. The carriers (together with biosterility markers) were sterilized in an autoclave at 121°C for a minimum of 15 min.


Light armour used to protect high mobility multipurpose-wheeled vehicles (HMMWV) was obtained by the Engineering Directorate (Edgewood Chemical Biological Center, ECBC, Aberdeen Proving Ground, MD, USA) from the manufacturer AM General Corporation (South Bend, IN, USA, The exterior of this material consisted in an aluminium alloy 5052-H34 camouflage coated with polyurea/polyurethane paint (Chemical Agent Resisting Coating, CARC military specification DIL 64159). A piece of light armour plate was randomly chosen from a large supply and custom-cut at the machine shop of the Aberdeen Proving Ground into 5 × 5 × 1 mm pieces. The metal coupons were cleaned with ethanol, rinsed with distilled water and sterilized in the same way as described for the rubber carriers.


Clear microscopy glass slides were custom-cut into 5 × 5 × 1 mm pieces by Erie Scientific Company (Portsmouth, New Hampshire, USA). Before use, the carriers were washed with ethanol, rinsed with distilled water, and then autoclaved in the same way described for the other carriers.

Bacillus species and strains

Several virulent strains were generously provided by Melissa Longnecker (US Army Research Institute of Infectious Diseases [USAMRIID], Ft. Detrick, MD, USA) including: (i) B. anthracis USAMRIID ba 1087; (ii) B. anthracis USAMRIID ba 1029; and (iii) B. anthracis LA1 (know also as USAMRIID ba 1088). Some of these strains have been used previously in research at USAMRIID (Little and Knudson 1986). Bacillus anthracis Ames was generously provided by Robert Buell [Biological Defense Research Division, US Navy, Washington, DC,]. Bacillus anthracis Vollum 1B (V1B) was provided under contract by Amanda Schilling (Naval Surface Warfare Centre, Dahlgren, VA, USA). Attenuated B. anthracis strains included Sterne and delta-Sterne provided by Dr Lisa Collins (Edgewood Chemical Biological Center) and Pasteur USAMRIID ba 3132 provided by USAMRIID (Fort Detrick). Other strains used in this study included B. subtilis 1031, B. atrophaeus ATCC B-385 (formerly known as B. globigii), Bacillus cereus ATCC 10702, B. thuringiensis 4055 (Microbial Genomic and Bioprocessing Research Unit, NCAUR, Peoria, IL, USA), and Bacillus megaterium CDC 684 (Carolina Biological Supply Company, Burlington, NC, USA). The identity of stocks of microbial strains was confirmed by analysis with The Crystal Identification System (Becton-Dickenson, Sparks, MD, USA) and by gas chromatographic analysis of fatty acids using instrumentation and software purchased from MIDI Inc (Newark, DE, USA). The plasmid composition of B. anthracis strains was confirmed by PCR analysis and it is indicated in Table 1.

Table 1.   Characteristics of Bacillus anthracis strains used in this study
Strain denominationPathogenesisPlasmids*Origin†
  1. *The presence (+) or absence (−) of capacity to synthesize capsule and toxin are indicated, respectively.

  2. †Origins as reported by Little, S.F., Knudson, G.B. (1986), and by Keim et al. (1997) and Price et al. (1999).

Ames Virulent+/+Originally isolated in Texas, USA
Vollum 1BV1BVirulent+/+Derived from Vollum which was isolated in the UK from a cow with anthrax in 1944
AlbiaUSAMRIID ba 1029Virulent+/+Albia, Iowa, 1963. Originally distributed by Iowa State University. With relatively lower virulence and forming rough colonies
ba 1087USAMRIID ba 1087Virulent+/+Dundee, Scotland. Isolated from a child treated for cutaneus anthrax
LA1USAMRIID ba 1088Virulent+/+Isolated in 1983 from an elephant (Loxodonta africana = LA) with anthrax in Etosha, Namibia
PasteurUSAMRIID ba 3132Attenuated−/+Derived from the original strain attenuated by Pasteur and used as vaccine in 1881
Sterne Attenuated+/−South Africa, Isolated by Sterne in 1937 and used as vaccine in livestock
Delta-Sterne Attenuated−/−As Sterne after the removal of the remaining plasmid

Preparation of spores

Pathogenic B. anthracis spores were prepared in the BSL3 facility of the Edgewood Chemical Biological Center. All strains of B. anthracis and all Bacillus species studied here were grown under comparable conditions as previously described (Carrera et al. 2006). Fresh overnight cultures of each Bacillus species were incubated by rotation at 37°C in 5–10 ml tryptic soy agar (TSA, Difco, Kansas City, MO, USA). Aliquots (400 μl) were spread over the surface of each 150 mm plates (six per strain) containing a modified medium derived from the Schaeffer Sporulation medium (described as sporulation medium S in Schaeffer et al. 1965). The agar plates were incubated at 25–37°C until 90–99% phase-bright spores were observed by phase-contrast light microscopy (see below). Spores were harvested and washed with cold sterile distilled ionized (DI) water as previously described (Carrera et al. 2006) and stored in DI water at 4°C until use for up to 2 weeks, changing the water at least once a week, or in the freezer at −20°C for up to a month.

Quality control of spores

The quality of spores was determined by two complementary criteria previously established to validate the presence of dormant spores (Sagripanti and Bonifacino 1996a; ASTM 2414-05, 2005). The criteria consisted in the evaluation of (i) the absence of vegetative cells (rods) determined by microscopic examination as described below, and (ii) the survival of spores in hydrochloric acid (2·5 N).

Microscopic analysis

Phase-contrast microscopy was performed using a Leica DMR optical microscope (Leica Microsystems Inc. Bannockburn, IL, USA) to distinguish spores at early stages of germination, which appeared phase dark, from dormant spores, which appeared phase bright. Imaging analysis was achieved with a Leica DC-480 camera (Leica Microsystems Inc. Bannockburn, IL, USA) and Image Pro Express software (Media Cybernetics L.P Silver Spring, MD, USA) as previously described (Carrera et al. 2006). Digital pictures were taken of every spore preparation and 200 microbial particles in each preparation were classified as vegetative cells or spores either in phase bright or in phase dark. All preparations used in this study contained less than 11% germinated spores, vegetative- or sporulating-cells, and consisted in 89% or more spores in phase bright as examined by phase-contrast light microscopy.

Acid resistance

Ten microlitres of each spore suspension was mixed with 90 μl of HCl 2·5 N and incubated for 5 min (vortexing every minute) and immediately neutralized with 900 μl of Luria Bertani's (LB) broth + 90 μl NaOH 2·5 N. The titre of spores treated with acid was compared with the titre of spores without acid treatment and incubated in distilled sterile water as a control. Spore preparations were acceptable if 90% of spores challenged survived acid treatment.

Sporicidal testing

The efficacy of decontaminant agents was evaluated by employing the ASTM standard E 2414-05 (ASTM 2005) which is a quantitative three-step method (TSM) to determine the sporicidal efficacy of liquids, liquid sprays and vapour and gases on contaminated carrier surfaces (Fig. 1). This method fully recovers treated spores by differential elution (in fractions A, B and C) with increasing stringency (nearly 100% spore recovery calculated as previously reported by the ratio of [the sum spores in fractions A + B + C after treatment with water as a control, divided by the number of spores loaded on each device] × 100, Sagripanti and Bonifacino 1996a,b, 1999). The forces to dislodge spores in each step are different and not interchangeable. Spores loosely attached to carriers are released by washing in A. Those spores bound with higher affinity are released by sonication in B, and those spores still remaining on the coupons are recovered after incipient germination in C (Fig. 1). Briefly, each clean and sterile carrier received 10 μl of a spore suspension containing between 1 × 109 and 5 × 109 organisms ml−1 (resulting in a microbial load between 1 and 5 × 107 spores per carrier) and was then dried during 2–4 h at 20–25°C. The carrier loaded with spores was placed inside of a 1·5-ml microcentrifuge tube (labelled A). The disinfectant was added to this tube assuring that the inoculum in the carrier was completely submerged in the fluid. Control carriers did not receive disinfectant but instead received an equal volume of sterile DI water. After 30-min incubation with the disinfectant at room temperature (21 ± 3°C), ice-cold LB medium was added. Each carrier was immediately transferred to a new 1·5-ml microcentrifuge tube (labelled B) containing sterile DI water at room temperature and sonicated for 5 min in a low power water-bath sonicator (rated at 400–500 watts, and generally used for cleaning jewellery and other small objects). Ice-cold LB medium was added after which, the carrier was transferred to a new 1·5-ml microcentrifuge tube (labelled C) with LB medium. The tubes were incubated in a rotator inside of an incubator at 37°C for 30 min. Ice-cold LB was added to the tube (C) and the carrier, free from remaining spores, was discarded. The surviving spores in each fraction (A, B and C) were titrated by serial dilution and spread on petri dishes containing nutrient agar medium. Culture plates were incubated overnight at 37 ± 1°C and colonies were counted. Total spores surviving treatment with disinfectant were calculated by adding the spores counted in fraction A, plus spores in fraction B, plus spores in fraction C. The log10 reduction (that is 90% spore inactivation corresponds to 1 log10 reduction, 99% spore inactivation to 2 log10, etc.) of the total spores exposed to the disinfectant was calculated by subtracting the total number of surviving spores from the total number of spores in the controls incubated with sterile water. The assay allowed measuring a 107-fold reduction (7 log10) in spore survival relative to those in the untreated controls (Sagripanti and Bonifacino 1996a).

Figure 1.

 Schematics of the three-step sporicidal method used in this work. Reprinted, with permission, from E2414-05 Standard Test Method for Quantitative Sporicidal Three-Step Method to Determine Sporicidal Efficacy of Liquids, Liquid Sprays, and Vapour or Gases on Contaminated Carrier Surfaces, copyright ASTM International, 100 Barr Harbor Drive, West Conshohocken, PA 19428.


Quality of spores

To properly compare spores from diverse Bacillus species and different strains of B. anthracis (described in Table 1), we prepared spores in various media until we identified one (Medium S which is a modification of Schaeffer et al. 1965 as described in ‘Materials and methods’) able to sustain efficient growth and sporulation of all Bacillus species studied. A series of techniques involving a variety of reagents, including lysozyme (Prentice et al. 1972) and renographin (Tamir and Gilvarg 1966), have been used in other studies to purify spores from their plate or liquid cultures, separating the cells and the germinated spores from the dormant ones. To prevent any reagent from altering the true sensitivity of spores to decontaminating agents, we eliminated cells and accompanying culture debris from our preparations by repeated centrifugation and washing of spore pellets with sterile DI water. A high concentration of cells in logarithmic phase at the time of inoculation in sporulating media was critically necessary in order to obtain spore preparations that passed our quality criteria (as described in ‘Materials and methods’) with the relatively high proportion of spores shown in Table 2. Acid resistance and microscopic analyses demonstrated that the spores to be challenged with decontaminating agents consisted largely of (phase bright) dormant spores (Fig. 2). Preparing spores of good quality and nearly free of vegetative cells was essential in obtaining reproducible data on the sensitivity of spores to disinfecting agents.

Table 2.   Quality control of spore preparation*
Species and strainsSpores phase bright (%)Spores phase dark (%) Cells (%)
  1. *Spores in early stages of germination (which appear phase dark), dormant spores (which appear as phase bright) and vegetative bacteria (rod shaped) were distinguished by microscopic observation and photographic analysis as described in ‘Materials and methods’.

Bacillus anthracis ba 1029901·09·0
B. anthracis LA-1891·010·0
B. anthracis Vollum V1B900·59·5
B. anthracis ba 1087971·02·0
B. anthracis Ames901·09·0
B. anthracis Sterne971·02·0
B. anthracis Delta-Sterne961·52·5
B. anthracis Pasteur961·03·0
Bacillus cereus990·50·5
Bacillus thuringiensis951·04·0
Bacillus megaterium980·51·5
Bacillus subtilis953·02·0
Bacillus atrophaeus962·51·5
Figure 2.

 Quality of spores. Phase contrast microscopy at 1000× (total magnification) of Bacillus anthracis delta Sterne, B. anthracis ba 1087, Bacillus cereus, and Bacillus subtilis showing more than 95% of phase bright dormant spores.

All B. anthracis sporulated after 5–6 days of plating. In contrast, B. cereus and B. megaterium sporulated quite rapidly, achieving 90–95% sporulation between 48 and 72 h after plating. By growing bacteria in TSB media and sporulating in medium S, yields ranged from 6·0 × 109 spores plate−1 (B. anthracis LA1) to 2·2 × 1010 spores plate−1 (B. megaterium).

Two or more batches of each Bacillus spores were prepared and tested below to account for any difference in sporulation between batches.

Effect of surface material

To quantitatively evaluate the interaction of various Bacillus spores with surface materials, we exposed the contaminated carriers to water as a non-sporicidal control and released the spores from the carriers by three steps of increasing stringency (fractions A, B and C). The fractioned elution of B. subtilis spores dried onto glass carriers after exposure to water was A > B > C, as expected from a relatively smooth and low-binding material, with 1·6 log10 and 1·1 log10 difference between steps, respectively (empty bars in Fig. 3).

Figure 3.

 Effect of surface materials. Bacillus subtilis spores dried onto glass (empty bars), metal (grey bars), or rubber (black bars) carriers were treated with water (in absence of disinfectants) and eluted in three steps of increasing stringency. Each fraction A, B and C was titrated separately as described in ‘Materials and methods’. Bar height represents the mean log of spore survival and the bracket over the bars indicate the standard error obtained in triplicate determinations.

A survey (Engineering Directorate, ECBC, US Army Material Command) revealed that gas masks (for their expected protective role) and light armour (for its wide distribution on military vehicles) were materials whose decontamination was of critical importance. Therefore, we dried spores onto silicone rubber employed in military protective gas mask production and onto a painted metal aluminium alloy used as light armour in military vehicles. The elution profile of B. subtilis from glass, metal and rubber is shown in Fig. 3. In both military materials, the mean number of spores in fractions A to B remained relatively constant in contrast to the progressive decrease observed in glass.

Sequential elution of virulent B. anthracis spores after drying in military materials and exposure to water is shown in comparison to B. athropheus in Fig. 4. In addition, spores from attenuated strains of B. anthracis and the other Bacillus species studied were also eluted with increasing stringency from metal and rubber carriers (data not shown). To compare any effect of the carrier material, we first counted the number of spores recovered in each fraction (A, B or C) after water treatment of each Bacillus species or strain (listed in Tables 1 and 3). Then, we calculated the average log10 number for each fraction (A, B or C) eluted from either metal or rubber among all spore strains and species tested. The log10 averages (± standard deviation, SD, in a number of experiments n = 12) from metal and from rubber carriers were 7·25 ± 0·52 and 7·30 ± 0·44 for fraction A; 6·58 ± 0·48 and 6·59 ± 0·53 for fraction B; and 4·82 ± 0·59 and 4·98 ± 0·91, respectively. These similar results obtained for each fraction ruled out a systematic effect of the carrier material in the recovery of spores from metal or rubber. The SD of the mean number of spores (obtained for all 13 different Bacillus spores) eluted in each fraction (A, B and C) by water from metal or rubber was 0·3 log10 (n = 75). Spores of B. anthracis Vollum V1B released relatively easily from rubber (most spores in fractions A and B and fewer in C, Fig. 3). In contrast, the number of spores recovered in fraction C from rubber remained relatively high for B. subtilis (Fig. 2) and for B. cereus (data not shown) suggesting a relatively stronger interaction between these spores and this particular carrier material.

Figure 4.

 Elution of Bacillus anthracis spores from metal or rubber. The different strains of virulent B. anthracis indicated in the graphs were tested on metal carriers (grey bars) or rubber carriers (black bars). The elution profile of Bacillus atrophaeus is included for comparison. Bar height represents the mean log of spore survival and the bracket over the bars indicate the SE (n geqslant R: gt-or-equal, slanted 3).

Table 3.   Comparative inactivation sensitivity of Bacillus spores
 Log reduction
Decon greenCloroxSandia DF100Sandia DF200
  1. The log of spore reduction relative to the amount of spores in the controls (identically processed after exposure to water). In each independent experiment, the three-step method protocol was performed with spores of one Bacillus strain deposited on triplicate carriers of each material and exposed to each decontaminant. The values are the mean log reduction ± SD (standard deviation, n geqslant R: gt-or-equal, slanted 3).

  2. *> in the Table indicates the detection limit when no surviving colonies were obtained.

Bacillus anthracis 1029  6·61 ± 0·48  5·84 ± 0·10  6·99 ± 0·176·30 ± 0·230·05 ± 0·160·75 ± 0·36  7·09 ± 0·01  6·50 ± 0·25
B. anthracis V1B 6·10 ± 0·208·06 ± 0·017·41 ± 0·788·06 ± 0·013·33 ± 0·033·90 ± 0·357·86 ± 0·028·06 ± 0·01
B. anthracis Ames 4·97 ± 0·105·33 ± 0·496·32 ± 0·605·99 ± 0·770·49 ± 0·222·54 ± 0·076·36 ± 0·446·77 ± 0·17
B. anthracis 1087>7·67*7·13 ± 0·20>7·677·53 ± 0·400·16 ± 0·100·22 ± 0·117·51 ± 0·275·55 ± 0·13
B. anthracis LA-16·13 ± 0·406·19 ± 0·366·10 ± 0·936·16 ± 0·880·79 ± 0·160·85 ± 0·166·85 ± 0·306·91 ± 0·24
B. anthracis Sterne5·96 ± 1·02>7·065·94 ± 1·046·30 ± 0·972·04 ± 0·351·75 ± 0·03>6·97>7·06
B. anthracis D Sterne6·74 ± 0·245·73 ± 0·276·93 ± 0·326·70 ± 0·491·34 ± 0·040·87 ± 0·036·75 ± 0·215·92 ± 0·03
B. anthracis Pasteur6·69 ± 0·375·93 ± 0·487·05 ± 0·947·12 ± 0·081·70 ± 0·120·78 ± 0·10>7·60>8·05
Bacillus cereus6·32 ± 0·285·62 ± 0·096·33 ± 0·385·52 ± 0·091·45 ± 0·061·08 ± 0·046·40 ± 0·285·80 ± 1·06
Bacillus thuringiensis6·77 ± 0·176·36 ± 0·04>6·876·91 ± 0·071·40 ± 0·561·16 ± 0·016·7 ± 0·287·17 ± 0·54
Bacillus megaterium7·18 ± 0·646·44 ± 0·247·09 ± 0·187·02 ± 0·720·09 ± 0·120·02 ± 0·087·51 ± 0·246·71 ± 0·28
Bacillus subtilis5·51 ± 0·194·68 ± 0·576·30 ± 0·336·29 ± 0·231·90 ± 0·231·57 ± 0·116·18 ± 0·065·73 ± 0·74
Bacillus atrophaeus5·95 ± 0·116·34 ± 0·136·28 ± 0·376·71 ± 0·231·97 ± 0·741·76 ± 0·226·47 ± 0·746·52 ± 0·22

Sensitivity of Bacillus strains

The sensitivity of various strains of B. anthracis deposited in military surfaces to a common decontaminating agent (Chlorox) was compared with the sensitivity of B. subtilis and B. atrophaeus spores. The inactivation by chlorine was similar among all these spores as shown by the results presented in Fig. 5. We investigated whether these similarities would extend to spores of other Bacillus species and to treatment with chemically different decontaminating agents. Therefore, we determined the inactivation produced by three additional decontaminating agents that have been proposed for use in biodefense and with chemical compositions that included peroxides and other oxidants. We compared the effect on spores of the same strains of B. anthracis tested with chlorine and five additional Bacillus species deposited on silicone rubber (protective mask material) or on aluminium alloy (light armour) with the results shown in Table 3. The different spores showed similar resistance to inactivation by the different decontaminating agents. The total number of spores inactivated by each agent was also similar on spores dried on both materials. Three decontaminating agents currently considered for use in military decontamination had a high and similar efficacy (generally above 6 log killing in Table 2). A decontaminating formulation previously considered for use (DF100) showed relatively low sporicidal activity.

Figure 5.

 Comparison of Bacillus anthracis and its simulants. Bar height represents the log reduction of the number of spores deposited either on metal carriers (grey bars) or rubber carriers (black bars). Log reduction was calculated by subtracting the total number of spores surviving treatment with sodium hypochlorite 5% (v/v) from the total number of spores recovered from carriers exposed to water as a control. Total spores surviving treatment with either hypochlorite or water were calculated by adding the spores counted in fraction A, plus spores in fraction B, plus spores in fraction C, respectively, as described in ‘Materials and methods’. The bracket over the bars represents the Standard Error obtained in each of the experiments from triplicate samples. ‘B. a’. represents B. anthracis spores of the strain specified in the x-axis. Gap separates B. anthracis strains from other Bacillus species (simulants).

The average inactivation from the eight treatments for B. anthracis spores (rows in Table 3) ranged between 4·8 log reduction for the Ames strain to 6·6 log reduction for the V1B strain. The average SD for the mean inactivation of all spores and all treatments in Table 3 was 0·31 log10 (n = 95), nearly identical to the SD obtained on the binding experiments discussed in the previous section. The average sensitivity (rows) of spores from the five (non-anthracis) Bacillus species ranged between 4·8 log10 for B. subtilis to 5·4 log10 for B. thuringiensis, within the range obtained for spores of B. anthracis. Therefore, the relative sensitivities to the tested disinfecting agents appeared similar for the various spores species and strains studied.


Some often overlooked parameters that can potentially bias the results from spore inactivation experiments include the use of preparations containing vegetative bacteria or germinated spores, and the use of tests that do not account for all challenged spores.

We subjected each spore preparation to quality-acceptance criteria before testing in order to avoid inactivation results from being confounded by the presence of germinated spores, by more sensitive vegetative cells, or by the chemical reactivity of disinfectants being scavenged by cell debris. As the goal of this study was to compare different spores and not to evaluate the effect of growth conditions, we employed the same growth and sporulation media to sustain the growth and sporulation of all Bacilli used in this study.

Full recovery of all the spores in the inoculums from contaminated negative controls required the fractionated elution of spores in three fractions: (i) consisting in spores loosely attached the surface, (ii) spores dislodged by sonication and (iii) spores released by a short incubation with agitation at 37°C. ASTM Standard E-2414-05 generally known as the TSM (Sagripanti and Bonifacino 1996a,b) was rapid, inexpensive, generated very little waste and quantitatively accounted for all spores challenged.

The lack of significant differences in the data pooled for all spores in rubber or in metal carriers precluded a difference in the relative binding of spores to surfaces that could bias subsequent decontamination studies in military gas mask or light armour. However, spores from different strains of B. anthracis and other Bacillus species seem to interact slightly different with each carrier, as shown by their elution profiles (Figs 3 and 4). The relative strength of spore binding to metal or rubber was independent of growth conditions (as B. anthracis and all other spores were prepared similarly) and was not correlated to virulence or presence of exosporium. Screening for spore binding by the sequential elution method described in this study could assist in identifying surfaces and materials better suited for microbial decontamination and in avoiding other materials where bacterial spores persist more readily.

We observed a similar sensitivity to chlorine among different strains of B. anthracis on contaminated surfaces (Fig. 5). This finding is consistent with the fact that strains of B. anthracis form a very monophyletic group as shown by genomic sequencing (Price et al. 1999). Our findings appear in disagreement with previous observations where the Ames strain (virulent) appeared slightly less susceptible to chlorination conditions used in water treatment than the attenuated Sterne strain (Rose et al. 2005), but lack of standard deviation and slight differences in initial inoculum and chlorine concentration make difficult to assess the statistical significance of the differences previously reported. In a subsequent study, spores of B. anthracis Sterne and B. cereus in suspension were more sensitive (between 1 and 2 log10) to chlorine than spores of B. thuringiensis ssp. Israelensis or B. anthracis Ames strain (Rice et al. 2005). There is apparent discrepancy between the similar sensitivity among strains of B. anthracis that we observed and the slight differences reported by others. Apparently contradicting results could be due either to (i) a differential sensitivity between spores in suspensions as reported previously and similar sensitivity on surfaces, as we observed; (ii) differences in inoculums or preparation conditions among strains or species in studies where these variables were not identical; or (iii) the differences previously reported could be below statistical significance.

In previous studies, B. atrophaeus spores in suspension appeared to be more resistant (approximately 2 log10) to free active chlorine than B. anthracis spores up to pH 8·6, above which resistance of both species appeared to be equal (Brazis et al. 1958). However, when chlorine was expressed in terms of hypochlorous acid, the same concentration was required to produce similar inactivation. We also observed a similar sensitivity of B. anthracis and B. atrophaeus on contaminated surfaces to unadjusted chlorine (whose pH is near 10).

The mean log reduction of different spores from five different virulent and three attenuated strains of B. anthracis, as well as B. subtilis, B. atrophaeus and the near neighbours B. cereus, B. thuringiensis and B. megaterium inactivated by each decontaminant that we tested was similar and generally within 1 log10 of each other (Table 3). This similarity among Bacillus strains and species was observed after treatment with any of the three agents with high activity as well as after exposure to the product showing low sporicidal activity. Although sporadic and relatively small differences in mean spore reduction were obtained for a given species or strain under a single combination (e.g. the relatively lower value for B. subtilis on metal exposed to Decon Green), these differences were not apparent under other conditions, and hence, can be attributed to the statistical variation expected on a relatively large body of data.

Virulent B. anthracis Ames strain and B. subtilis spores on contaminated surfaces exhibited no significant differences to inactivation by gaseous hydrogen peroxide in five of seven surfaces used as interior building materials (Rogers et al. 2005). Thus, the similar sensitivity to liquid agents that we observed for spores on surfaces generally agrees with the similar sensitivity of B. anthracis and B. subtilis spores reported after gas inactivation. The difference in sensitivity to gaseous inactivation (approximately 1·5 log10) between B. anthracis Ames and B. subtilis previously reported for the other two substrates (industrial carpet and pine wood) paralleled 1 log reduction difference (10%) in the experimental recovery of both organisms obtained in the untreated controls (Rogers et al. 2005). Thus, the apparent difference in sensitivity to gaseous peroxide previously reported could relate to differential recovery, the impact effect of which on sporicidal testing has been discussed previously (Sagripanti and Bonifacino 1996a,b, 1999). Moreover, the previous report of differences to gaseous inactivation could be traced to different conditions reported to prepare spores (B. anthracis Ames using a BioFlo fermentor in the laboratory vs B. subtilis purchased from a commercial source, Rogers et al. 2005).

Overall, the data reported here indicate that the sensitivity of common simulants (B. atrophaeus and B. subtilis) to inactivation by products that contain peroxide, chlorine or oxidants is similar to that of all the B. anthracis strains studied. Our findings of similar spore sensitivity to chemical agents is consistent with the similar sensitivity to UV inactivation (same UV inactivation kinetics) exhibited by B. anthracis Sterne and B. subtilis spores, as long as both spores were prepared and assayed under identical conditions. (Nicholson and Galeano 2003).

The similar sensitivity that we observed with spores from different species and strains suggests that members of the Bacillus genera share an energetically comparable biochemical pathway that ultimately leads to spore inactivation. The comparative results of the present study suggest that decontamination and sterilization data obtained with simulants can be safely extrapolated to spores of B. anthracis and indicate that valid conclusions on sporicidal efficacy can be drawn from safer and less costly experiments employing non-pathogenic spore simulants. These findings should assist government agencies and commercial companies involved in biodefense to develop and evaluate more effective sporicidal products.


This work was supported by the US Department of Defense Chemical and Biological Defense program administered by the Defense Threat Reduction Agency. The testing of B. anthracis strain Vollum V1B performed under contract by Ms Amanda Schilling (Naval Surface Warfare Centre Dahlgren VA, USA) is acknowledged. The DECON Team of the Edgewood Chemical Biological Center is thanked for providing the military DECON products. The advice and guidance on selection and use of military materials received from Merlin Erickson, Engineering Directorate, ECBC (Aberdeen Proving Ground, MD, USA) is appreciated. Jim Church and Richard Dekker (Engineering Directorate, ECBC, Aberdeen Proving Ground, MD, USA), MAJ Dan Rusin and John Escarcega, Weapons and Materials Research Directorate, US Army Research Laboratory (Aberdeen Proving Ground, MD, USA), and Brent Starkey, Office of the Product Manager for Sets, Kits, Outfits, and Tools, Tank and Automotive Command (Rhode Island, IL, USA) are thanked for the information on the materials used in gas masks, light armour and CARC coatings. The information on strains of B. anthracis provided by Drs Arthur M. Friedlander (USAMRIID, Fort Detrick), Paul J. Jackson (Lawrence Livermore National Laboratory, CA, USA) and Paul Keim (Northern Arizona University, Flagstaff, AZ, USA) is appreciated.