The cedar-leaf oil was obtained from Tree of Life Essential Oil, Port Hardy, BC, Canada. The product was obtained by steam distillation of Thuja plicata (Western red cedar) leaves, and contained >80% thujone.

All the test organisms were ATCC strains, obtained from PML Microbiologicals, Oregon. The bacteria, Bacillus subtilis, Streptococcus pyogenes, Hemophilus influenzae, Acinetobacter baumannii, Enterococcus fecalis, Salmonella enteritidis, and Escherichia coli, were propagated and assayed, as colony-forming units (CFU), in Mueller-Hinton broth and agar plates. Candida albicans and Aspergillus niger, also ATCC strains, were maintained and assayed on Sabouraud-dextrose agar plates. Human lung epithelial cells (A549), were obtained from ATCC, and propagated in flasks in Dulbecco MEM medium supplemented with 5% fetal bovine serum, without anti-microbial agents, in a 5% CO₂ incubator at 37 °C.

Bacterial working stocks were obtained for each experiment by selecting 4–6 colonies from an agar plate and growing them in Mueller-Hinton broth. Cells were pelleted by centrifugation, washed and resuspended in PBS (phosphate buffered saline) to approximately 10⁶ cfu/mL. For the reactions with liquid CLO 20 or 100 µL of diluted bacteria were added to replicate Eppendorf tubes and mixed with the appropriate volume of CLO, at ambient temperature (22 °C). Individual tubes were taken at various times and the bacteria were pelleted in a microfuge, washed with PBS and assayed for cfu on agar plates. The centrifugation time in the microfuge was less than a minute; consequently this was not a significant addition to the exposure time during the reactions. Controls consisted of identical numbers of bacteria mixed with phosphate buffered saline (PBS) and processed in the same way without exposure to CLO. In some experiments CLO was replaced by tea tree oil (TTO) as a positive control. In experiments to evaluate the possible effect of light on the anti-bacterial activity, half of the Eppendorf reaction tubes were completely enclosed in aluminum foil to preclude ambient light (dark reactions [18]).

Dried films of bacteria were obtained by placing 5 µL of the stock bacterial suspension on the inner side of an Eppendorf tube cap and allowing to dry. A separate Eppendorf tube was then used for the oil “reservoir”, onto which the cap with dried bacteria was inverted. After an appropriate exposure time the bacteria were reconstituted by washing them off the cap with PBS, pelleted and washed, followed by assay for cfu. In some experiments 50 µL aliquots of bacteria were dried onto glass slides, which were then inverted over a jar containing the oil, followed by reconstitution in PBS, washing and assaying for cfu. In all experiments, only a fraction of the CLO actually evaporated during the reactions; thus continuous exposure to the dried films was obtained for many hours without replenishing the oil, an observation that is relevant to proposed use in the field.

Endospores of B. subtilis were prepared by growth of the bacteria on agar plates, followed by induction of spore formation by continued incubation with the plates sealed in parafilm. This was followed by rinsing off the upper spore layer, and treatment with distilled water at 80 °C [16]. Further purification was achieved by the histodenz centrifugation method described by Sorg and Dineen [19]. The pellet obtained with 45% histodenz was essentially free of vegetative/dead cells and debris, as determined by malachite green-safranin staining.

Candida albicans yeast cells were grown in Sabouraud-dextrose broth and treated in an analogous fashion to the bacteria, in liquid and dry film reactions, followed by assay for cfu on Sabouraud-dextrose (SB) agar plates. Aspergillus niger was grown on SB plates, and samples were removed for resuspension in PBS. Dried films were prepared as described above, and these were treated and assayed by inspection for growth or no growth on SB plates.

Cell viability was measured on monolayers of human lung A549 cells grown to confluence in 6-well trays. For treatment the media were removed by aspiration, and the moist cells were exposed for various times to CLO vapor. Following a further 24 h incubation in normal medium, cell viability was measured by the lactate dehydrogenase (LDH) method, using the kit obtained from Sciencell (Carlsbad, CA, USA), and following the instructions provided. The final assay reactions were measured in a micro-plate reader at a wavelength of 450 nm.

A total of seven bacteria, known to be potentially pathogenic to humans and recognized as environmental contaminants, were selected for evaluation. It was found that most of the materials used routinely in the laboratory were appropriate for tests (i.e., they were resistant to CLO liquid and vapor), with the exception of polystyrene culture vessels (trays, tubes, flasks), which were degraded by direct contact with CLO. The resulting degradation product was toxic to bacteria and to cultured cells. Consequently polypropylene and glass vessels were used in the experiments. Other materials tested, such as metals, wood, hard plastic, glass and fabrics, appeared to be resistant to CLO.

Initial tests were carried out with B. subtilis, using TTO, a well-known antibacterial oil [13] as a positive control, and an equivalent number of unexposed bacteria as a negative control. The antibacterial efficacy was affected by time of exposure and concentration of CLO, as expected, but was not significantly influenced by the number of bacteria, as shown in Figure 1. This latter point was important to establish since the number of viable organisms in the field are likely to vary considerably from a few bacteria to many thousands. B. subtilis was readily killed by CLO after several minutes contact with the oil (Figure 1). Concentrations of CLO down to 0.1% (in saline) completely killed the bacteria within one hour (Figure 1 and Figure 2). Further dilutions showed successively less efficacy.

Four of the other bacteria were tested by the same protocols, and all of them were readily killed by CLO down to high dilutions, as shown in Figure 2, although the relative sensitivities varied. H. influenzae (Gram –) and S. pyogenes (Gram +) were as susceptible as B. subtilis (Gram +), whereas A. baumannii (Gram –) was slightly more resistant, and E. fecalis (Gram +) was substantially more resistant. Nevertheless even the latter was completely killed by exposure to 1% CLO. In addition S. enteritidis (Gram –) was very sensitive, but E. coli (Gram –) was slightly more resistant (data not shown). To determine if these activities of CLO were bactericidal or bacteriostatic, representative treated culture plates were returned to the incubator for several days, but no growth or additional growth occurred. In addition when the treated “bacteria” were pelleted by centrifugation, washed in saline and resuspended in broth, no growth was observed. Thus the activities were bactericidal.

Since CLO is immiscible with aqueous solutions, incubations of CLO with bacterial suspensions required frequent agitation or mixing, although efficient bacterial killing was achieved. Attempts were made to improve the stability of the emulsions by incorporating small amounts of the neutral detergent Tween 80 into the oil, as recommended by studies on Taxandria oil [20]. However the addition of either Tween 80 or Tween 20 did not improve the antibacterial efficiency of CLO any further and consequently they were omitted in subsequent tests.

Some plant-based antimicrobials are influenced by ambient light. In theory the antibacterial activity could be enhanced by light (due to the presence of photosensitizers, [18]), or reduced as a result of photo-degradation. Since the CLO applications in buildings could take place in either the presence or absence of ambient light, it was important to establish that the efficacy of the CLO would not be affected significantly. This was tested with B. subtilis, but the efficacy of killing by CLO was essentially the same in light or dark, as shown in Figure 3.

Conventional methods of producing spores for study often contain mixtures of spores and vegetative cells. In order to ensure that prospective antimicrobials are really effective against spores, it is desirable to experiment with preparations that are relatively free of dead and vegetative forms [10]. Recently a method of purifying spores was reported for Clostridium difficile (an anaerobic intestinal pathogen), which involved centrifugation of partly purified spores through gradients of Histodenz [19]. The spores of C. difficile sedimented through 50% Histodenz, leaving them free from the vegetative cells and dead organisms, which were found in the layers of less dense Histodenz.

In our experiments with different concentrations of Histodenz, it was shown that purification of B. subtilis spores worked best by sedimentation through 45% Histodenz (they did not sediment through 50%). This purified spore preparation was tested in a time course experiment, and the results are summarised in Table 1. Increasing exposure times and concentrations resulted in greater killing of the spores, although the rate and level of killing achieved was not as great as that observed with the vegetative B. subtilis.

In subsequent experiments the bacteria were used as dried films on glass slides, to represent normal conditions of surface contamination. The treated films were reconstituted in saline solution after exposure, followed by serial dilution and measurement of CFUs on agar plates. The films, containing different numbers of viable bacteria, were found to be just as vulnerable to liquid CLO as bacterial suspensions (data not shown).

The next series of experiments consisted of exposing similar dried films of bacteria to the vapor of CLO. This treatment was also very effective, although the time of exposure required for efficient killing was significantly longer than with the liquid oil. Results are shown in Figure 4 for the three bacteria, B. subtilis, H. influenzae, and S. pyogenes. In separate tests dried films of A. baumannii showed a similar level of susceptibility.

The pathogenic yeast C. albicans was exposed to CLO liquid or vapor and assayed on agar plates as CFU (analogous to bacterial cfu). The organism was readily killed by liquid CLO, with a sensitivity similar to B. subtilis and H. influenzae (data not shown). The results in Table 2 show that this organism is as susceptible to CLO vapor as the bacteria.

The fungus (mold) A. niger is a common environmental organism which can cause serious disease in humans with compromised immune systems. It can be grown and assayed conveniently on appropriate agar plates, although it tends to produce filamentous growth rather than discrete colonies. Quantitative measurements can be performed by serial dilutions and determination of growth end-point. The results indicated that this organism is also susceptible to CLO-vapor, although at this time we cannot be sure if it is irreversibly killed (fungicidal) or just growth-inhibited (fungistatic), since a few colonies did eventually appear on plates of treated A. niger following prolonged cultivation. Nevertheless even if eradication was not achieved, the prospect of its control by CLO vapor is realistic.

It is important to establish experimentally that short-term exposure to CLO vapor is harmless to building occupants and operators. This was evaluated by exposing cultures of human lung cells (A549 epithelial cell line), with culture medium removed to permit exposure of the monolayer of cells, to CLO vapor for different times. This was followed by microscopic examination of the cells for signs of cell toxicity or death, and assays for cell viability. No signs of cytopathology were evident, and the results of the LDH (lactate dehydrogenase) assay (Table 3) showed that exposures up to 60 minutes had no deleterious effects on the cells.