Journal of Applied Microbiology ISSN 1364-5072


Standard method for deposition of dry, aerosolized, silicacoated Bacillus spores onto inanimate surfaces D. Harnish1, B.K. Heimbuch1, M. McDonald1, K. Kinney1, M. Dion2, R. Stote2, V. Rastogi3, L. Smith3, L. Wallace3, A. Lumley1, H. Schreuder-Gibson2 and J.D. Wander4 1 2 3 4

Engineering Science Division, Applied Research Associates, Panama City, FL, USA U.S. Army Natick Soldier Research, Development, and Engineering Center, Natick, MA, USA Biodefense Branch, Edgewood Chemical Biological Center, Aberdeen Proving Grounds, MD, USA Air Force Research Laboratory, Tyndall Air Force Base, FL, USA

Keywords aerosol, anthrax, antimicrobial, Bacillus, bioaerosol, deposition, infectious agents, spores. Correspondence Delbert Harnish, 430 W 5th St, Ste 700, Panama City, FL 32401, USA. E-mail: [email protected] 2013/2425: received 16 December 2013, revised 18 March 2014 and accepted 19 March 2014 doi:10.1111/jam.12509

Abstract Aims: To evaluate a standard aerosolization method for uniformly depositing threat-representative spores onto surfaces. Methods and Results: Lyophilized Bacillus anthracis DSterne spores, coated in silica, were aerosolized into a containment chamber and deposited onto nine surface types by two independent laboratories. Laboratory A produced a mean loading concentration of 178 9 105 CFU cm2; coefficient of variation (CV) was 104 colony-forming units (CFU) per cm2 is generally the threshold level (U.S. Army

Aerosol deposition of Bacillus spores

Research Office 2004; JSTO-CBD 2007) when evaluating decontamination technologies. Brown et al. (2007) loaded coupons with B. atrophaeus spores to evaluate sampling efficiency. They presented no data on loading variability and described loading densities in ranges: 102–103 or 104–105 CFU cm2. Baron et al. (2008) loaded only agar plates, but their chamber could load other substrates. They intended to load surfaces with low concentrations (approx. 003, 03, 3 CFU cm2) to be used in subsequent sampling studies. Loading variability was lowest at the highest density, coefficients of variation (CVs) averaging 177% over three runs. Based on experience, a DQO of 105 CFU cm2, deposited as a narrow PSD of aerosol particles in the respirable range, and to provide sample-to-sample variation with CV 80% spores viable by phase-contrast microscopy, >1010 CFU g1 spores after lyophilization and aerosolization of fluidized spores with a CMD of 20  10 lm. Materials and methods Spore preparation Bacillus anthracis DSterne (BaDS) spores were prepared by the method of Buhr et al. (2008). Spores were grown overnight on a tryptic soy agar (TSA) plate at 37°C. A sterile loop collected a colony and inoculated a 50-ml aliquot of tryptic soy broth (TSB) in a 250-ml Erlenmeyer flask. Incubating the TSB mixture for 25 h at 37° C and 220 rev min1 on a rotary shaker achieved a midlog growth phase culture. Sporulation broth was prepared by autoclaving 500 ml of 5% Difco nutrient broth (Becton–Dickinson; Franklin Lakes, NJ) in deionized (DI) water. After autoclaving, sterile-filtered components were added to achieve final concentrations of 200 mM pH-adjusted L-glutamate, 13 mM KH2PO4, 28 mM K2HPO4, 1 mM CaCl22H2O, 01 mM MnCl24H2O, 1 mM MgCl2, 005 mM ZnCl2 and 0001 mM FeCl36H2O in sporulation broth diluted to a final volume of 100 l with sterile DI water. A 1-l baffled flask containing 100 ml of sporulation broth was inoculated with a loopful of mid-log growth phase culture and incubated at 37°C and 300 rev min1 on a rotary shaker. After day 3, spores were observed daily under phase-contrast microscopy until the culture demonstrated >80% phase-bright spores. Spores were then harvested and purified by dividing the spore solution into two 50-ml conical polypropylene tubes and centrifuging for 10 min at 3000 9 g at room temperature. The supernatant was discarded, and each spore pellet was resuspended in 20 ml of sterile DI water at 4°C and centrifuged again for 10 min at 3000 9 g at 42

room temperature. This rinse step was repeated twice. After the third cycle of centrifugation, the supernatant was discarded, and each spore pellet was resuspended in 10 ml of sterile DI water. The suspensions were heat-treated for 30 min in a water bath at 65°C and then centrifuged for 10 min at 3000 9 g at room temperature and the supernatant was discarded. Cycles of resuspend, centrifuge and discard steps were repeated as above until >80% of spores in the suspension were phase bright. The final spore pellets were suspended in 10-ml sterile 005% Polysorbate 20 (Fisher Scientific, Hanover Park, IL) and then combined into one suspension, which was sonicated for one min and then vortexed for 30 s. Phase-contrast microscopy was used to determine the final ratio of phase-bright spores to phase-dark spores, vegetative cells and debris. The spore stock suspension was then centrifuged at 3000 9 g for 10 min, the supernatant discarded, and the pellet resuspended in sterile water in volume equal to the discarded supernatant. Spores were freeze-dried according to the manufacturer’s instructions for the lyophilizer (Virtis 4KBTZL-105; SP Industries, Inc., Warminster, PA). Lyophilized spores were aliquotted into 15-ml conical polypropylene tubes and stored at 80°C. Viable plating was performed to determine spore count per gram. Aerosol system The biological dispersion system (BDS, Fig. 1) comprised a glass-walled chamber, an aerodynamic particle sizer (APS 3321; TSI Inc., Shoreview, MN) and a fluidized-bed aerosol generator (PITT-3; AlburtyLab Inc., Drexel, MO) (ASTM International 2013). The chamber was a 06-m2 base by 12-m-high enclosure with four primarily glass walls. Two doors were included for easy access—the left door opened the entire side; the right door accessed the lower portion of the chamber. The interior floor was a turntable on which test samples were loaded. A variable speed control on the turntable selected rotation speeds of 0–7 rev min1. The top and bottom, respectively, of the chamber had particle distribution manifolds including eight symmetrically placed inlet and discharge ports to promote even distribution of particles. During runs, air enters through the top and is drawn out of the bottom of the BDS at the same rate to prevent pressurization. The PITT-3 aerosolized particles using vibration and air flow. Particles were supported on a latex diaphragm and aerosolized using an audio speaker mounted below the diaphragm. Careful adjustments to the audio signal frequency, amplitude and waveform allowed tuning of aerosol characteristics and concentrations. The PITT-3 used an airstream to lift suspended particles vertically into the tubing leading to the chamber distribution manifold. The system contained HEPA filters

Journal of Applied Microbiology 117, 40--49 © 2013 The Society for Applied Microbiology This article has been contributed to by US Government employees and their work is in the public domain in the USA

D. Harnish et al.

Aerosol deposition of Bacillus spores

sample of each material type occupied a position at each of three distances from the centre and in each third of the chamber, to allow assessment of loading variability across the platform onto each material type (Fig. 2). In addition to samples of the eight materials, six liquid reservoirs—20-ml scintillation vial caps 30 cm in diameter (71 cm2 surface area)—were filled with 2 ml of extraction buffer (ASTM International 2013) and similarly distributed on the turntable. The premise was that spores will load equally onto a given surface area, so a comparison between the total spores settled on the liquid reservoirs and on test samples would enable a calculation of extraction efficiency. (c)

(b) (a) Figure 1 Biological dispersing system: (a) aerosol chamber, (b) aerodynamic particle sizer, (c) fluidized-bed aerosol generator.

on the inlet and exit to purify air entering or exiting the system. To provide for secondary containment, the system was designed to fit into a biological safety cabinet, measuring at least 20 m high 9 12 m wide 9 08 m deep.

Test procedure Two independent laboratories performed sample preparation and aerosolization studies using separate chambers and personnel. Three independent aerosol loading experiments were performed by each laboratory. Samples for aerosolization were prepared by adding 008 g of lyophilized BADS, 002 g hydrophobic fumed silica (Aerosilâ R812S; Evonik Industries, Theodore, AL) and 15 sterile 3-mm borosilicate glass beads to a 15-ml round-bottom polypropylene tube and securing the cap with Parafilmâ to create an airtight seal. The tube was vortexed for 2 min on high, with pauses at 30-s intervals to tap the bottom of the tube against a hard surface. The Parafilm was then removed under aseptic conditions, and the glass

Test materials Eight DoD-relevant materials were tested in this study— butyl rubber, COLPRO tent fabric, polycarbonate, hydrophilic cotton fabric, hydrophobic cotton fabric, surgical facemask fabric, concrete and stainless steel. Butyl rubber and polycarbonate were obtained from McMaster–Carr (Atlanta, GA). Stainless steel was grade 304, unpolished and no evidence of oil. Concrete was made from Quikrete grey mortar repair mix (Atlanta, GA), trowelled to a rough surface and cured for >4 weeks. Both cotton fabrics were of the same 100% cotton fabric sheet—the hydrophobic samples were made so by applying heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane and then microwave-treated. The facemask fabric was nonwoven and untreated. Circular coupons, 26 cm in diameter (53 cm2 surface area), were cast in 26-cm-diameter moulds (concrete) or cut from each material using either an arch punch (fabrics) or a water-jet cutter (stainless steel and polycarbonate). Coupons were arranged within the chamber so a

Figure 2 Coupon layout on the turntable in the aerosol chamber. liquid reservoir; □rubber; ○COLPRO tent; ♢polycarbonate; hydrophilic cotton; ▲hydrophobic cotton; tent; ♦facemask; ●concrete; ■steel.

Journal of Applied Microbiology 117, 40--49 © 2014 Society for Applied Microbiology. This article has been contributed to by US Government employees and their work is in the public domain in the USA


Aerosol deposition of Bacillus spores

D. Harnish et al.

Data analysis To demonstrate sample variability, the CV was calculated for each sample type of each run (Eqn 1). To demonstrate sample type variability, a one-way ANOVA using a Bonferroni post-test was used to compare the mean loading concentrations of all sample types in pairs within a given run. CV ¼ r=l


r = standard deviation; l = mean loading concentration. 44

To demonstrate run variability, a one-way ANOVA was used to compare the mean loading concentrations of all three runs. To compare results between laboratories, the loading data are reported as the ratio of CFU deposited (per coupon) to total CFU aerosolized (based on spore powder weight and concentration), expressed as percentage of total spore delivery. Two approaches were used to normalize the data from both laboratories to allow for comparison of the two data sets: (i) comparison of the ratio of total spores per coupon to the total spores aerosolized for each sample type of each run; (ii) comparison of the ratio of the mean loading concentration of each sample type to the mean loading concentration of the liquid reservoir for that run (RL), divided by the mean RL value for all samples of the run. A one-way ANOVA test was used to compare the normalized data for both laboratories. Results Laboratory A The lyophilized spore content used by Laboratory A was 23 9 1010 CFU g1; >80% phase-bright spores were observed prior to lyophilization. The mean CMD for all three spore aerosol runs was 119  008 lm (Fig. 3). The areas under the APS curves from Runs 1–3 were 1255, 3145, and 3870 lmN, respectively, providing a reference point for the total particle count for each run. The mean loading concentration for all three runs was 178  105 9 105 CFU cm2 (Table 1). CVs for each sample type in each run were 099). Discussion The purpose of this method is uniform deposition of spores onto surfaces at high enough concentrations to evaluate the efficacy of sporicidal technologies. Accessible requirements suggest that loading of biological agents on surfaces following a biological attack may exceed 104 CFU cm2 (U.S. Army Research Office 2004; JSTOCBD 2007). The mean loading concentration achieved by both laboratories exceeded this value, making the method threat representative for loading concentration. However, for testers and researchers within the DoD, a six-log reduction is often specified as the criterion for effectiveness of sporicides and other decontamination

Journal of Applied Microbiology 117, 40--49 © 2014 Society for Applied Microbiology. This article has been contributed to by US Government employees and their work is in the public domain in the USA


Aerosol deposition of Bacillus spores

D. Harnish et al.

Table 2 Laboratory B spore deposition data

Loading concentration (9106 CFU cm2)

Coefficient of variation (%)



Run 1

Run 1

Run 2

Run 1

Run 1

Run 2

Run 3

Run 1

Run 2

Run 3

Reservoirs* Butyl rubber COLPRO tent Polycarbonate Hydrophilic cotton Hydrophobic cotton Surgical mask Concrete Stainless steel Mean

595 414 609 137 474 414 647 496 ND 627

2558 1419 4876 10652 698 991 4974 3213 N/A 3673

3220 4771 899 1587 3141 1878 4448 359 N/A 2538

3917 2817 12224 1916 1360 10942 1823 1400 7000 4822

– 070 102 230 080 070 109 083 N/A 106

– 069 070 078 095 065 082 066 N/A 075

– 056 299 070 053 169 084 060 048 105

– 066 097 217 075 066 103 079 N/A 100

– 091 094 104 127 087 109 089 N/A 100

– 053 285 067 051 161 080 057 046 100


Run 2 152 059 297 146 033 041 322 159


127 870 895 995 121 829 104 844 ND 994

Run 3


42 415 081 158 38 156 46 030


785 436 235 552 418 133 656 471 377 819


308 123 288 106 057 146 120 066 264 645


ND, no data; N/A, not applicable; RL see Table 1. *Larger surface area of reservoirs accounted for.

Steel Concrete Facemask Hydrophobic Cotton Hydrophilic Cotton Polycarbonate COLPRO Tent Rubber Liquid Reservoir 0·0001


0·01 Deposition %

technologies. A six-log reduction in viability also appears in the medical literature as a criterion for sterilization (FDA 2002). The aerosolization method described herein consistently deposited challenges of viable spores that exceeded 105 CFU cm2. Had a larger coupon been used (e.g. 125-cm2), loading ≥106 CFU would have been achieved. The DQO of >80% spore purity was easily achieved by each laboratory. However, variations in the spore preparations were clearly evident in APS data taken during the test runs at the two laboratories (Figs 3 and 4). The APS curves collected by Laboratory B were very similar in shape to one another, but showed obviously different particle concentrations. This could have been enhanced by small differences in total spore powder added to the PITT-3 or slight fluctuations in the settings to the PITT3. The particle distributions among the three runs for Laboratory A all had different-shaped curves. These 46



Figure 5 Percentage spores deposited based on spores delivered for Laboratories A and B. Laboratory A; Laboratory B.

differences are likely due to inconsistency in the spore production methods. CMDs of all three runs fell within the DQO range, and APS variation in curve shape and particle concentration was observed to not aggravate loading variability. The mean CMDs of 119  008 and 111  000 lm for Laboratories A and B, respectively, show that the small-scale grinding/fluidization preparation yields consistent results. If the overall PSD is to be considered, the total area under the curve or the area larger than the CMD may provide a better indicator of the level of spore delivery than the CMD alone. However, improving uniformity of the spore preparation is expected to lead to methods that are more reproducible. Work will continue to better standardize small-scale spore production. Although we acknowledge that a single spore preparation used for all validation tests would have provided better data standardization, demonstration of the method’s ability to deposit spores uniformly using

Journal of Applied Microbiology 117, 40--49 © 2013 The Society for Applied Microbiology This article has been contributed to by US Government employees and their work is in the public domain in the USA

D. Harnish et al.

different spore batches highlights the robustness of this application. Distributing spores uniformly onto surfaces (

Standard method for deposition of dry, aerosolized, silica-coated Bacillus spores onto inanimate surfaces.

To evaluate a standard aerosolization method for uniformly depositing threat-representative spores onto surfaces...
601KB Sizes 0 Downloads 3 Views