Air sampling of fungal spores on filters. An investigation on passive sampling and viability† A° sa Na�sman,*ab Go�ran Blomquistb and Jan-Olof Levinab aUmea° University, Department of Public Health and Clinical Medicine, Occupational Medicine, S-901 85 Umea°, Sweden, E-mail: asa.nasman@niwl.se; Fax: +46 90 7865027; Tel:+46 90 7869071 bNational Institute for Working Life, Department of Chemistry, Box 7654, S-907 13 Umea°, Sweden Received 9th April 1999, Accepted 9th June 1999 In this study, glycerol was tested as a collection substrate for passive bioaerosol sampling.Filters (mixed cellulose acetate and nitrate) were soaked in glycerol and exposed for an aerosol from three diVerent fungal species: Penicillum commune, Aspergillus versicolor and Paecilomyces variotii.The passive sampling method was compared with a closed-face polycarbonate filter sampling method. Exposure was performed in an exposure chamber. The total number of spores was determined by microscopic techniques, and the cultivable number was determined by cultivation on Malt Extract Agar dishes. The glycerol soaked filter demonstrated a good correlation with the closedface sampler with regard to the total count.Spores stored in a pumped filter cassette were not aVected by storage for up to 7 days. On the other hand, the culturability of the spores was markedly decreased after 1 day when stored on glycerol soaked filters. liquid have been used.12 These instruments, however, are Introduction usually not suitable for personal sampling, and filters have The measurement of micro-organisms is an important tool for been used to monitor the exposure in the work environment.13 investigating diseases in the work environment.DiVerent tech- Currently, there is a need for small and easily handled niques, such as impaction, impingement or filtration, have samplers in the work environment. The sampler should not been used for sampling micro-organisms in diVerent work interfere with the working procedure, i.e.be small and light. environments.1 The determination of the number of micro- It should also be easy to use. The best situation would be if organisms in the sample is usually performed either by the worker was able to perform the sampling himself and send microscopy or by cultivation.2–4 Microscopic methods provide the sample by mail for analysis at the laboratory.In order to the total number of single micro-organisms, cultivable as well develop such a sampler, it is necessary to find a suitable as non-cultivable, while cultivable methods determine only collection substrate that works for passive sampling, and also colony forming units which may be composed of single spores to determine how storage aVects the culturability of the spores.as well as aggregates of spores. Several studies show that the Passive sampling has been used for sampling dust in diVerent cultivable number of spores is markedly lower than the total work environments. Brown et al.13 used electrostatically number.5–7 This may be caused by a loss of viability during charged substrates to collect airborne dust, while Vinzents14 sampling,8,9 an eVect of aggregation7 or by the use of inad- used a tape mounted on a sampler in three diVerent directions.equate growth substrate. Since there is always a fraction of So far these passive samplers have not been used to assess the non-cultivable spores that still may cause trouble for humans exposure of micro-organisms in the work environment.such as allergic reactions and fever, 10 it is important to make In this study, a collection substrate for passive sampling sure that all spores are taken into account. This implies that was investigated. A filter soaked in glycerol was used as a the sampling method used for assessing the exposure in the passive sampler. This sampling method was compared with work environment must not only collect precise aerosol frac- closed-face filter cassettes.Both the total count and the cultitions, as defined by the CEN standard EN 481,11 but also vable counts were compared after 0, 1, 2 and 7 days of storage. prevent the micro-organisms from losing their viability during sampling and analysis. How the viability of fungal spores is aVected by storage Materials and methods between sampling and analyses is poorly explored. The micro- Cultivation of fungal spores organisms may lose their viability due to stress during sampling (i.e.method of collection and handling of sample), the type of Penicillum commune (Pegasus LAB AB, Uppsala, Sweden) and growth medium used, the aerosolization method, the tempera- Paecilomyces variotii (UPSC 1766) were cultured in Petri ture and the humidity.9 These eVects diVer between diVerent dishes (9 cm in diameter) on Malt Extract Agar (MEA), micro-organisms and sampling methods employed.Bacteria composed of 1.5% Bacto-Agar (Difco, Detroit, MI, USA), are usually more sensitive than spores of fungi and acti- 2% Malt Extract (Oxoid L39, Basingstoke, Hampshire, UK) nomycetes when sampled on filters.8 Therefore, samplers or and deionized water.Aspergillus versicolor (UPSC 2027) was impactors that collect the bacteria on a semisolid gel or in a cultured on MEA plates plus 20% sucrose, M20 plates (J. T. Baker B. V., Deventer, Netherlands). P. commune was cultured for 7 days at 22 °C and afterwards the cultivation plates were †Presented at AIRMON ’99, Geilo, Norway, February 10–14, 1999.J. Environ. Monit., 1999, 1, 361–365 361stored at +5 °C for further usage after 8–32 days. A. versicolor the filter cassettes was 1.88±0.06 l min-1. The filter cassettes were analysed immediately after sampling. was cultured for 14–21 days at 22 °C and then either used immediately or stored for 5–36 days at +5 °C. P. variotii was cultured for 7 days at 22°C and then the cultivation plates were stored at+5 °C for another 6–13 days.After the sampling Determination of cultivable and non-cultivable spores procedure, the spores from P. commune and A. versicolor were After the sampling period, the inlet and the outlet connections cultivated on MEA plates. P. variotii was cultured on MEA of the filter cassettes were plugged.All filters were moved to with antibiotics [3 ml l-1 penicillin–streptomycin 10 000 a smaller sealed chamber for storage at either +22 °C or IU ml–10 000 ml ml-1 (Gibco BLR, Life Technologies, Ta� by, +5 °C (P. commune). Four HA filters and two polycarbonate Sweden)]. filters were analysed each time. Analyses were performed after 0, 1, 2 and 7 days. Laboratory generation and sampling of spores The passive filters were placed in a sterile beaker, covered HA (mixed cellulose acetate and nitrate) filters with a pore with a lid, and 3 ml of sterile filtered 0.05% Tween-80 was size of 0.45 mm and measuring 25 mm in diameter (Millipore added.Five ml of sterile filtered 0.05% Tween-80 was added Co., Bedford, PA, USA) were first soaked in glycerol (BDH to the filter cassettes.All filters were then vortexed for 30 s in Laboratory Supplies, Poole, Dorset, UK), and left hanging order to suspend the fungal spores. The total number of spores for 1 h to remove excess glycerol. The remaining droplet was was then determined by counting in a Bu� rker chamber. Nine removed with a Kleenex tissue. These filters were then used A-squares were counted three times per sample.When the for passive sampling. total number exceeded 200 spores, three diagonal A-squares Polycarbonate filters, with a pore size of 0.45 mm and a were counted. diameter of 37 mm, inside polypropylene aerosol monitors After the extraction procedure, 1 ml was used for serial (MSI, Westboro, MA, USA) were used for pumped sampling. dilution. Cultivable counts were analysed by spreading 100 ml The mean air flow through the filter cassette during sampling of the original solution, diluted 10, 100 and 1000 times for the was 1.06±0.07 l min-1.passive filters, and 100, 1000 and 10 000 times for the pumped All filters were placed in an even pattern in a chamber made filters, on MEA plates. P. variotii was cultured on MEA plates of polyacrylate with a voluproximately 1 m3 with antibiotics. Three cultivation plates were used for each (120×100×80 cm).The filters were divided into three diVerent dilution. These were cultivated for 7 days at 22 °C before the positions in the box corresponding to the left, the middle and number of colony forming units was determined. This number the right (Fig. 1). Air was blown (50 l min-1) at the culture represents the cultivable part of the fungi.The cultivable dish, placed in the middle of the chamber, through a tube number was then divided by the total number of spores in ending approximately 5 cm above the culture dish. This was order to obtain the percentage of cultivable spores. done to produce an aerosol in the chamber. The exposure time was 1 h in all experiments with P.commune and 2 h with A. versicolor and P. variotii. In order to investigate the cultivable number on a cultivation Results and discussion plate, a cultivation plate of P. variotii, which was matched Validation of the exposure chamber with the plate used for aerosolization, was rinsed with sterile filtered 0.05% Tween-80 (KEBO AB, Stockholm, Sweden) The mean value of the spore concentration in the exposure before the experiment was started.This sample is referred to chamber varied between 7×105 and 1×107 for the filter as the reference sample. The spore solution was then treated cassettes and between 3×105 and 1×106 for the passive filters as below. The same thing was done afterwards with P. in the diVerent P. commune experiments.In the A. versicolor commune or A. versicolor. Two plates of P. commune and A. experiments, the concentration varied between 8×105 and versicolor were used in each experiment; the spores were either 3×106 and 4.0×104 and 4.4×104 for the filter cassettes and rinsed by sterile filtered 0.05% Tween-80 (this sample is referred the passive filters, respectively. The corresponding values for to as the reference sample) or aerosolized in the exposure P.variotii were 3×105–8×105 and 3×104–6×104, respectchamber as above. During this experiment, four pumped ively. The number of spores that were aerosolized when air polycarbonate filters were used. The medium air flow through was blown at a cultivation plate varies considerably between diVerent species. Of the three diVerent species investigated in this study, P.commune was easiest to aerosolize. P. variotii and A. versicolor spores were harder to render airborne from the cultivation plate, and therefore the sampling time was increased to 2 h. Normalized values for the diVerent positions in the exposure chamber (Fig. 1) are shown in Tables 1 and 2. Positions closer to the walls gave lower spore counts.This is caused by the change in air flow close to the walls. No sampling was therefore performed closer than 20 cm to the wall. An ANOVA test was conducted at 95% confidence interval comparing the diVerent sampling positions in five P. commune experiments. The results showed that there was no significant diVerence between the diVerent positions for pumped sampling, however there was a significant diVerence between the passive sampling positions.The results are shown in Tables 3 and 4. As this study is based on the percentage of cultivable spores Fig. 1 The diVerent sampling positions in the exposure chamber. The sampled at each position, or mean values for a complete cones represent the filter cassette samplers and the circles represent experiment in comparison with other experiments, it is assumed the passive samplers.All samplers to the right are named R1–R9, the that the spread in concentration in the exposure chamber did samplers in the middle M1–M6 and the samplers to the left L1–L9 (see Tables 1 and 2). not aVect the results. 362 J. Environ. Monit., 1999, 1, 361–365Table 1 Distribution of spore concentrations at the diVerent sampling positions.Expressed as normalized values for passive sampling. Positions according to Fig. 1 Position P. commune A. versicolor P. variotii R1 0.4 1.8 0.9 R3 1.4 1.2 1.1 R5 1.0 1.3 0.8 R7 1.0 1.3 0.9 R9 1.0 1.5 1.2 M1 1.1 0.8 1.0 M2 0.7 0.7 0.8 M3 1.1 0.6 0.9 M4 1.5 1.5 1.6 Fig. 2 Correlation between filter cassette sampling and passive M5 1.1 1.0 1.7 sampling when total number was compared.Each point represents M6 1.0 0.8 1.1 the mean value of the total number from a single experiment, passive L1 0.7 0.9 0.8 sampling versus filter cassette sampling. Filled diamonds indicate L3 0.8 0.6 0.6 P. variotii, open squares A. versicolor and filled triangles P. commune. L5 1.0 0.9 0.8 L7 1.2 0.8 1.1 L9 1.0 0.9 0.8 Influence of sampling method on culturability Pumped sampling.Samples collected with the ordinary filter Table 2 Distribution of spore concentrations at the diVerent sampling cassette showed a decrease of about 50% in culturability positions. Values expressed as normalized values for filter cassette (Fig. 3, 5 and 6). The loss in culturability is seen immediately sampling. Positions according to Fig. 1 after sampling and is not significantly aVected by storage over several days. Previous studies have shown that the culturability Position P. commune A. versicolor P. variotii of fungal spores is decreased when a pumped sampling method is used.5–7 When the reference samples (samples not submitted R2 1.2 1.8 1.1 R4 0.9 1.1 0.8 to the sampling step) for all three species tested in this study R6 0.7 0.8 0.8 were analysed, the result showed a non-cultivable fraction of R8 1.2 1.3 0.7 about 30% (Fig. 4 and 6). This means that moulds consist of L2 0.7 1.0 0.7 quite a large fraction of spores that are already non-cultivable L4 1.4 0.9 1.8 before sampling is performed. A. versicolor and P. commune L6 0.9 0.7 1.0 showed no significant diVerence in culturability when the L8 0.9 0.4 1.0 reference sample was compared with spores collected by pumped sampling (Fig. 4). P. variotii showed a decrease in Table 3 ANOVA table for the diVerent positions in the exposure culturability of 30% after pumped sampling compared to the chamber for passive sampling. SS=square sum, df=degrees of free- reference sample (Fig. 6). A. versicolor and P. commune dom, MS=mean square, F=calculated F-value showed no decrease in culturability (Fig. 3 and 5) when sampled on pumped filters, compared with P. variotii that Source of variation SS df MS F P-value tended to loose culturability with storage time (Fig. 6). After Position 1.63×1011 15 1.1×1010 1.90 3.6×10-2 storage at +22 °C for 7 days, only half of the spores remained Experiment 2.08×1011 4 5.2×1010 9.3 6.2×10-6 cultivable.Error 3.35×1011 60 5.6×109 Total 7.05×1011 79 Passive sampling. All three species were aVected in a negative manner by glycerol. P. commune appeared to stand a short time of exposure to glycerol and was unaVected after 1 h of Table 4 ANOVA table for the diVerent positions in the exposure sampling (Fig. 7). After 1 day of storage, the cultivable chamber for filter cassette sampling.SS=square sum, df=degrees of fraction was 8% and decreased to zero after further storage. freedom, MS=mean square, F=calculated F-value A. versicolor and P. variotii were aVected immediately. P. variotii showed a cultivable fraction of 30% after 2 h of Source of variation SS df MS F P-value sampling, with no cultivable spores after 1 day of storage Position 4.3×1011 7 6.1×1010 0.76 6.2×10-1 (Fig. 8). A. versicolor was most aVected, i.e., after 2 h of Experiment 1.1×1012 4 2.7×1011 3.4 2.1×10-2 sampling only 14% was cultivable (Fig. 9). Further storage Error 2.2×1012 28 8.0×1010 reduced the cultivable fraction to zero. Total 3.8×1012 39 Passive filters soaked in glycerol reduce the culturability of EYciency of the sampling methods for total spore sampling The passive filter method was validated against a traditional pumped method, in order to compare two diVerent sampling techniques.Both the total count and cultivable count were compared. The total number of spores collected by the filter cassettes was 2–22 times larger than the amount collected on the passive filters. A coeYcient of correlation of 0.8 (Fig. 2) between the two sampling methods was achieved when the total number was compared.This indicates that these filters may have the potential to be used as collection substrate in a passive sampling device, if they are built into a sampling device Fig. 3 Cultivable number of P. commune for filter cassette sampling. Values expressed as the mean value of two experiments. and calibrated against a pumped sampler.J. Environ. Monit., 1999, 1, 361–365 363Fig. 7 Cultivable number of P. commune for passive sampling. Values expressed as the mean value of two experiments. Fig. 4 Cultivable number of reference sample and mean value of the four filter cassettes used in the same experiment. A, P. commune; B, A. versicolor. Fig. 8 Cultivable number of P. variotii for passive sampling.Values expressed as the mean value of two experiments. Fig. 5 Cultivable number of A. versicolor for filter cassette sampling. Values expressed as the mean value of two experiments. Fig. 9 Cultivable number of A. versicolor for passive sampling. Values expressed as the mean value of two experiments. Stability of experiment culture The purpose of the experiment was to study the influence of storage time and temperature on the culturability of the spores in order to investigate how a prolonged storage time before sampling and diVerent storage temperatures after sampling aVect the culturability.The cultivable fraction of P. commune was increased by 40% if the experiment culture was stored at +5 °C for 32 days before sampling (compared with 8 days), when pumped sampling was used (Fig. 3 and 10B). When passive sampling was Fig. 6 Cultivable number of P. variotii for filter cassette sampling. used, a longer storage time at +5 °C gave an increase of 10% Values expressed as the mean value of two experiments. Noll= (Fig. 7 and 10A). The largest eVect was observed when the reference sample (not submitted to sampling). filters were stored for more than 1 day.On the seventh day of storage, 7% of the spores were still cultivable. A. versicolor was not aVected by a longer storage phase. Regarding temperature stability after sampling, when all the fungi and are therefore not suitable for use when it is necessary to identify the species. This is probably due to filters were stored at +5 °C instead of +22 °C after sampling, an increased culturability of about 31% was seen for both dehydration since the glycerol used contained no water.Other substrates, such as polymer solutions with their rather sampling methods (Fig. 3, 7 and 11). Passive sampling showed an increased culturability after 24 h of storage; on day 2, the high water content (80–90%),15 might be better in this consideration. culturability was 9% (Fig. 7 and 11A). 364 J. Environ. Monit., 1999, 1, 361–365important as a large fraction of the spores are non-cultivable already before sampling. This means, regardless of the sampling method, a large fraction of spores will be missed if culture-based methods are used for determination. On the other hand, if a non-culture-based method is used, it is diYcult to determine what species are being dealt with.Pumped filter cassettes show good characteristics for sampling of fungal spores in both aspects. The cultivable fraction is stable during storage and is of the same magnitude as for normal growing conditions. Glycerol impregnated filters may be used as a collection substrate for passive sampling as long as only the total number of spores is determined.In order to produce a passive sampler that works for the cultural fraction as well, other substrates need to be investigated. An alternative could be water-based polymers, such as polyethylene glycols. Acknowledgements The excellent assistance of Sven-Olof Westermark (National Institute for Working Life, Department of Chemistry, Umea°, Sweden) and the helpful advice of Hans Stenlund (Department Fig. 10 Cultivable number of P. commune. Cultivation plate stored at of Epidemiology and Public Health, Umea° University, Umea°, +5 °C for 29 days before aerosolized. A, passive sampling; B, filter Sweden) are gratefully acknowledged. cassette sampling. References 1 G. Blomquist, Analyst, 1994, 119, 53. 2 U. Palmgren, G. Stro�m, G. Blomquist and P. Malmberg, J. Appl.Bact., 1986, 61, 401. 3 Air Quality Monographs, vol. 2 Health Implications of Fungi in Indoor Environments, ed. R. A. Samson, B. Flannigan, M. E. Flannigan, A. P. VerhoeV, O. C. G. Adan and E. S. Hoekstra, Elsevier, Amsterdam, 1994, ch. 2. 4 W. Eduard and D. Heederik, Am. Ind. Hyg. Assoc. J., 1998, 59, 113. 5 B. Crook, S. J. Futter, W. D. GriYths, L. C. Kenny, S. Stagg, J. D. StancliVe and I.W. Stewart, Ann. Occup. Hyg., 1997, 41, 677. 6 K. Karlsson and P. Malmberg, Scand J. Work Environ. Health, 1989, 15, 353. 7 W. Eduard, J. Lacey, K. Karlsson, U. Palmgren, G. Stro�m and G. Blomquist, Am. Ind. Hyg. Assoc. J., 1990, 51, 427. 8 B. Crook, L. C. Kenny, S. Stagg, J. D. StancliVe, S. J. Futter, W. D. GriYths and I.W. Stewart, Ann. Occup. Hyg., 1997, 41, 647. 9 W.D.GriYths and G. A. L. DeCosmo, J. Aerosol Sci., 1994, 25, 1425. 10 P. Malmberg, Arbete och Ha�lsa, 1991, 50, 39. 11 Workplace Atmospheres—Size Fraction Definitions for Measurement of Airborne Particles, EN 48151993, European Fig. 11 Cultivable number of P. commune. Cultivation plate stored at Standard, Brussels, 1993. +5 °C for 35 days before aerosolized. Samples stored at +5 °C. 12 E. Welam-Henningsson and M. S. Ahlberg, J. Aerosol Sci., 1994, A, passive sampling; B, filter cassette sampling. 25, 1459. 13 R. C. Brown, M. A. Hemingway, D. Wake and A. Thorpe, Analyst, 1996, 121, 1241. Conclusions 14 P. S. Vinzents, Ann. Occup. Hyg., 1996, 40, 261. 15 P-A° . Albertsson, Partition of Cell Particles and Macromolecules, Since it is important to determine both the total number and Wiley, New York, 3rd edn., 1986, pp. 73–111. the cultivable number of fungal spores, it is necessary to find a method that is reliable in both aspects. This is especially Paper 9/02832C J. Environ. Monit., 1999, 1, 361–3