A mobile biosafety microanalysis system for infectious agents

Biological threats posed by pathogens such as Ebola virus must be quickly diagnosed, while protecting the safety of personnel. Scanning electron microscopy and microanalysis requires minimal specimen preparation and can help to identify hazardous agents or substances. Here we report a compact biosafety system for rapid imaging and elemental analysis of specimens, including powders, viruses and bacteria, which is easily transportable to the site of an incident.

which could otherwise obscure the virus particles. This makes this approach applicable to both samples grown in tissue culture (as presented in this report), as well as any biological fluid from a patient sample. For example we have previously used this method to identify pathogens in urine and blood samples. Pore sizes ranging from 10 nm can catch the smallest viruses, while a pore size of up to 20 mm in diameter is suitable for larger bacteria. Very small virus particles (smaller than 30 nm in diameter) cannot be easily seen by SEM. However, all of the category A agents are above this size range and are thus able to be resolved by SEM.
Our results show that the system was able to distinguish between two families of poxviruses, (Fig. 2). Vaccinia, an orthopox virus, is brick-shaped and larger (approximately 360 3 270 3 250 nm) than pseudocowpox virus (a parapoxvirus) which also has a more cylindrical pill shape (160-190 diameter, 250-300 nm long). Ebola virus was also easily identified in SEM by its distinctive filamentous morphology 10,11 and the presence of comma-shaped virions (Fig. 2).
Spores of Bacillus cereus (as a proxy for anthrax) were clearly identifiable using the SEM and high resolution details such as the distinctive exosporium were observed equally as well as with TEM (Fig. 2). Bacterial samples of Salmonella and Listeria monocytogenes were easily recognizable by their morphology and surface features such as the 20 nm flagellae were also apparent ( Supplementary Fig.  4).
All specimens were also observed by negative-stain transmission electron microscopy (TEM) 1,12 for comparison (Fig. 2, Supplementary   Fig. 4). SEM has the advantage that the surface textures of the specimen are easily seen, and it can be used to directly observe bulky materials at low magnification. The traditional TEM technique has the disadvantage that it can only be used with dehydrated thin specimens, less than 200 nm thick. Many of the particles in powders are much thicker than this. For example, table salt particles are approximately 5 mm in diameter, more than 2000 times bigger than the thickest sample that can be observed by TEM. Crumbly or volatile specimens also present a problem since they are unstable and can contaminate the ultra-high vacuum of a TEM, but these types of specimen can be well tolerated using the higher specimen chamber pressures at which modern SEMs can operate. The additional specimen preparation steps required for TEM take time, and may also alter the specimen or cause artifacts. Moreover, a TEM is a large piece of equipment that cannot be moved without extensive disassembly, requires stringent stable environmental parameters for operation, and cannot be easily contained in a biosafety enclosure.
Several ''white powders'' were analysed as examples of the type of specimens that might be encountered during investigation of suspect bioterrorism events. These included table salt, domestic sugar, artificial sweeteners (sodium cyclamate, sucralose), gypsum board, and dried milk powder. (Fig. 2, Supplementary Figs. 5, 6, 7). The SEM images show the distinctive crystalline forms, the presence of fibres, and varied particle sizes of the different powders. The X-ray spectra demonstrate the relative elemental abundance profiles in the specimen (Fig. 2, Supplementary Fig. 5), and elemental images show the location of the specific elements within a sample ( Supplementary  Fig. 7), both of which are useful as a signature for forensic identification of the substance. Toxic heavy metals and other elements of high atomic weight are readily apparent in X-ray spectra. For example, the domestic sugar sample contains traces of calcium, silicon, sodium, magnesium, sulphur and aluminum from impurities or additives that are readily detectable (Fig. 2).
The X-ray spectra and elemental maps were collected using the Xmax 80 mm 2 silicon drift detector (SDD) with the INCA EnergySEM 350 microanalysis software package (Oxford Instruments, Abingdon, UK). This system has 2048 channels with an energy resolution of 129 eV at Mn Ka, with a light element sensitivity that can detect beryllium. The system is capable of collecting spectra, elemental images, and quantitating the relative elemental composition in a specimen. The operation of this system in the SEM enclosure does not affect the energy resolution, or any of the other operations of the system as per the manufacturer's specifications. Some elements generate X-rays with overlapping peak positions (by both energy and wavelength) that are difficult to separate. Thus sensitivity varies according to the elemental composition of the specimen. In general, a useful X-ray detection limit is 0.1%, but can be as low as 0.01% for elements of large atomic mass against a low atomic mass background, or where the X-ray spectral peaks are well separated 13 . In general, substances such as toxic heavy metals added to an otherwise harmless substance (such as foods-which contain mostly elements of low atomic mass) can often be easy to detect with the technique.
The combination of X-ray microanalysis and SEM all within the same enclosure provides a safe and powerful forensic tool. The SEM images permit the morphological identification of a microorganism, and the X-ray microanalysis can give information on the elemental composition of both the organic and inorganic components. For example, elemental composition data such as a high calcium concentration, can be used to differentiate bacterial spores from other suspicious particles with a similar morphology 14 . In investigating a suspect anthrax powder, its morpohological characteristsics, such as having a high spore concentration, a uniform particle size, and the presence of anti-clumping agents might indicate a material that was deliberately prepared 16 . Thus, a full forensic analysis can be carried out, using a combination of nucleic acid amplification methods to identify the specific strain of agent present, along with microanalysis to determine the particle size, morphology, and elemental composition to help investigate how the material was formulated.
In conclusion, we describe the first electron microscope with elemental microanalysis within a compact class III biosafety cabinet that can be operated externally, while the system is biologically sealed. Along with novel methodology for specimen preparation that we have recently developed 9 , this system is ideal for microscopy and elemental microanalysis of biohazardous specimens. This has been achieved in a highly compact platform that can be easily moved from one room to another. For example, it could be temporarily operated within a biosafety level 3 laboratory, to further increase safety. The capability for operation at the site of incidents can help to avoid delays in transporting specimens to specialized high containment laboratories. For this type of field operation, it is envisaged that the micoranalysis system would form part of a mobile laboratory setup including personal protective equipment and a portable negative air pressure isolation unit for collecting and processing samples 3,15 .

Methods
Design of the cabinet. The best solution was to enclose the entire microscope in the class III biosafety high-efficiency particulate air (HEPA) filtered cabinet maintained at an internal air pressure that is negative to that of the surrounding room (Fig. 1,  Supplementary Fig. 1). This avoided problems with designs that contained only part of the microscope system (which would cause difficulties with sealing the joints around complicated parts of the system). Furthermore, the vacuum exhaust is within the containment enclosure, to avoid any potential aerosol hazards that could be created when the microscope chamber is pumped down. A full scale mock-up of the SEM enclosure design was constructed around the JEOL JCM-5700 ( Supplementary  Fig. 2). The SEM was given a full imaging and spectrum collection operation trial with the mock-up in place. This allowed adjustments to the design to be made before final manufacturing. Operation using the electrical bulkhead was tested to ensure correct electronic function, and adjustments were made to allow ergonomic positioning of the glove ports and pass through box for specimen exchange, aperture alignment and filament replacement.
The cabinet is equipped with three HEPA filters for air intake, exhaust, and the pass-through chamber which acts as an airlock for bringing infectious specimens into the microscope (Fig. 1c, Supplementary Figs. 1f, 3c). The filters can easily be removed while preventing any external contamination (known as ''bag in/bag out'' replacement). Ports are available for decontaminating the entire equipment after use, using vaporous hydrogen peroxide (Supplementary Fig. 3d). Sensors provide a read out of the temperature and pressure inside the enclosure, which is maintained at a lower level than outside, so that when the pass-through chamber is opened to insert specimens, air is drawn inwards, protecting the outside environment from contamination ( Supplementary Fig. 3a). Enclosing the equipment in an air-tight box created a problem with the air cooling of the microscope, which is designed to work in a laboratory room with adequate ventilation and temperature control systems. Six Peltier refrigerator units were included to provide sufficient cooling (Fig. 1b,  Supplementary Fig. 1f, blue arrows).
The design also has a rear door through which the microscope can be wheeled out on ramps for servicing (Fig. 1b, Supplementary Fig. 3b). There is external control of all of the microscope's controls including motorized stage movement, magnification selection, beam intensity and focussing functions. The electronics for external control are connected via an electrical bulkhead (Fig. 1c, Supplementary Fig. 3e). A general purpose port (blue circle, in Supplementary Fig. 3e) was included for installation of additional detectors or accessories if required.
Wet sample preparation for SEM. All fluid samples were filtered using 13 mm diameter SPI-pore polycarbonate track etch filters (SPI supplies, West Chester Pennsylvania, USA), held in 13 mm SwinnexH filter holders (Millipore, Billerica, Massachusetts, USA) attached to syringes with Luer-LokH couplings to prevent sample leaks. In a Class II Biosafety Cabinet, bacterial suspensions were made form growth on agar plates. Approximately two loop-fulls of bacteria were suspended in 1 ml PBS. If the suspension was too turbid a 10X dilution was made. The filter was first wetted by passing 2 ml of PBS through the apparatus. Then 0.2 ml of the bacterial suspension was applied to the filter with a 1 ml syringe, followed by three www.nature.com/scientificreports SCIENTIFIC REPORTS | 5 : 9505 | DOI: 10.1038/srep09505 consecutive 2 ml washes of PBS using a 2 ml syringe. Finally 2 ml of 4% glutaraldehyde was applied with a 2 ml syringe. After one hour wait, the filter was washed with 2 ml 50% ethanol, 2 ml 70% ethanol, 2 ml 85% ethanol, 2 ml 95% ethanol, and then 2 ml of 100% ethanol and finally air dried. Syringes were either operated by hand, or with a Legato 200 syringe pump (KD Scientific, Holliston, Massachusetts, USA).
Gold coating. Filters were cut and mounted on an SEM stub using double-sided adhesive carbon disc and silver flash paint to create a contact between the stub and the filter paper. The samples were sputtered with gold using a Quorum Q150R S (Quorum Technologies, East Sussex, UK) containing a 0.1 mm gold target. The sample was pumped down, purged with argon and sputtered with gold for 120 sec while on a rotating stage.
Dry sample preparation for SEM. Working in a class II biosafety cabinet, powder samples were directly mounted onto double-sided adhesive carbon discs attached to metal specimen stubs, using a spatula to sprinkle small quantities. The powder was then gently pressed with the spatula to improve adherence. The stub was then inverted over a waste container and tapped to remove any excess loose particles.
Operation of the SEM in the biosafety enclosure. When the enclosure is turned on the negative pressure alarm sounds until the enclosure reaches the operational 0.50 H 2 O below ambient pressure which is usually achieved in less than one minute. This confirms both operational status of the system as well as the functioning of the alarm system. The negative pressure inside is indicated in red on the controller display ( Supplementary Fig. 3a). Temperature control is automatic and maintains an internal temperature of 20.5uC. The SEM (JEOL CarryScope, JEOL Ltd., Tokyo, Japan) is computer controlled, and equipped with motor drives for X, Y and Z motion which are connected to the electrical bulk head of the enclosure (Supplementary Fig. 3e). The only differences in the operation of the SEM is that any time the SEM has to be physically touched the glove ports or specimen pass through must be used (Fig. 1d,  Supplementary Fig. 1e). Therefore operations including turning on the SEM, opening and the closing stage for specimen insertion, specimen stage tilting, in-plane rotation of specimen in stage, alignment of condenser aperture, servicing of the electron source, and condenser aperture adjustment all require use of the glove ports. All other features of the SEM are electronically controlled externally (shift, focus, magnification, stigmatism, contrast, and brightness). The SEM was operated at 4 kV, with a 7 mm working distance, and with a 30 mm condenser aperture. Images (2560 3 1920 pixels) were collected using the secondary electron detector with an acquisition time of 160 seconds.
X-ray microanalysis. X-ray spectra were collected using the X-max 80 mm 2 silicon drift detector (SDD) with the INCA microanalysis software package (Oxford Instruments, Abingdon, UK). The entire detector is electronically controlled externally and is connected to the computer through the electrical bulkhead of the biosafety enclosure ( Supplementary Fig. 3e). The X-max SDD is cooled by a Peltier cooler, and so does not require liquid nitrogen. For X-ray microanalysis the SEM was operated at 20 kV, with a 10 mm working distance, and 100 mm condenser aperture. Maps and line scans were collected at 512 3 352 pixels with 2048 X-ray channels and 50 frames per acquisition with a dwell time of 100 ms per pixel: each acquisition took 15 minutes.
Virus cultures. Zaire Ebola virus was propagated in Vero E6 cells and prepared as previously described 17 . Samples were analysed by SDS-Page and Western blotting, and rendered non-infectious by fixation with 4% paraformaldehyde. Excess fixative was removed by placing the fixed samples in a Slide-A-Lyzer G2 cassette with a 0.5 ml capacity, and a 10,000 MWCO (Thermo Scientific Pierce Protein Research Products, Rockford, Illinois, USA), followed by dialysis against PBS. All work with infectious Ebola virus (virus culture and purification) was performed in the biosafety level 4 laboratories at the National Microbiology Laboratory of the Public Health Agency of Canada, Winnipeg, Manitoba.
Baby hamster kidney fibroblast cells (BHK-21: ATCC) were grown in Dulbecco's Modified Eagle's Medium (Gibco) containing 10%fetal bovine serum (FBS) (Gibco) and 1X penicillin/streptomycin/L-glutamine (Gibco). BHK-21 cells were inoculated with Modified Vaccinia Ankara (MVA) (kindly provided by Dr. Jingxin Cao, National Microbiology Laboratory) for 1 hour at 37uC. After washing with PBS, complete growth medium was added and the cells incubated at 37uC for 48 hours. MVA was harvested by freeze-thawing cell cultures 3X, alternating between -80uC and room temperature. Following final thawing, the supernatant was clarified by centrifugation at 3000 3 g for 3 minutes to remove the cell debris.
Bacterial cell cultures. Bacillus cereus was cultured on CAB plates, and incubated at 37uC for 24 hours. 300 ml of 1/10 Columbia broth containing 0.1 mM MnSO 4 , were inoculated with a loopful of Bacillus cereus, and incubated at 37uC for 96 hours on an orbital shaker. The culture was centrifuged at 4900 g for 15 minutes at 4uC. The supernatant was decanted and the pellet was washed three times in sterile deionized water, followed by centrifugation at 4900 3 g for 15 minutes at 4uC. The entire spore preparation was resuspended in 40 ml of ethanol, and transferred to a 50 ml tube, then incubated at room temperature for 2 hours. Then it was centrifuged at 4900 g for 15 minutes at 4uC, and the ethanol was decanted off. This was followed by two washes with 40 ml of sterile deionized water, and centrifuging at 4900 g for 15 minutes at 4uC. The remaining wash was decanted off. The final spore pellet was resuspended in 10 ml of sterile deionized water. Non-Typhi Salmonella was grown overnight on a semi-solid agar plate at 35uC. Listeria monocytogenes was grown in tryptose phosphate agar (TPA) motility tubes at room temperature for two days, and then subcultured on to a TPA plate at room temperature for an additional two days.
Sample preparation for TEM. Fluid samples for transmission electron microscopy were fixed in 2% glutaraldehyde/1% paraformaldehyde. Samples were adsorbed to glow discharged carbon-coated formvar films on a 400-mesh copper grids for 1 min, and negatively stained with 2% methylamine tungstate (Nano-W; Nanoprobes, Yaphank, NY, USA). Specimens were observed at 200 kV in an FEI Tecnai 20 transmission electron microscope (FEI Company, Hillsboro, OR, USA) operated, and at instrument magnifications of X25,500 to X71,000. Digital images of the specimens were acquired using an AMT Advantage XR 12 CCD camera (AMT, Danvers, MA, USA). This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder in order to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ www.nature.com/scientificreports SCIENTIFIC REPORTS | 5 : 9505 | DOI: 10.1038/srep09505