Effects of fungal species, cultivation time, growth substrate, and air exposure velocity on the fluorescence properties of airborne fungal spores

Transcription

1 Indoor Air 2015; 25: wileyonlinelibrary.com/journal/ina Printed in Singapore. All rights reserved 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd INDOOR AIR doi: /ina Effects of fungal species, cultivation time, growth substrate, and air exposure velocity on the fluorescence properties of airborne fungal spores Abstract Real-time bioaerosol monitoring is possible with fluorescence based instruments. This study provides information on major factors that can affect the fluorescence properties of airborne fungal spores. Two fluorescence-based bioaerosol detectors, BioScout, and ultraviolet aerodynamic particle sizer (UVAPS), were used to study fluorescent particle fractions (FPFs) of released spores of three fungal species (Aspergillus versicolor, Cladosporium cladosporioides, and Penicillium brevicompactum). Two culture media (agar and gypsum board), three ages of the culture (one week, one month, and four months), and three aerosolization air velocities (5, 15, and 27 m/s) were tested. The results showed that the FPF values for spores released from gypsum were typically lower than for those released from agar indicating that poor nutrient substrate produces spores with lower amounts of fluorescent compounds. The results also showed higher FPF values with lower air velocities in aerosolization. This indicates that easily released fully developed spores have more fluorescent compounds compared to forcibly extracted non-matured spores. The FPFs typically were lower with older samples. The FPF results between the two instruments were similar, except with four-month-old samples. The results can be utilized in field measurements of fungal spores to estimate actual concentrations and compare different instruments with fluorescence-based devices as well as in instrument calibration and testing in laboratory conditions. S. Saari 1, J. Mensah-Attipoe 2, T. Reponen 2,3, A. M. Veijalainen 2, A. Salmela 2, P. Pasanen 2, J. Keskinen 1 1 Department of Physics, Tampere University of Technology, Tampere, Finland, 2 Department of Environmental Science, University of Eastern Finland, Kuopio, Finland, 3 Department of Environmental Health, University of Cincinnati, Cincinnati, OH, USA Key words: Fungal spores; Bioaerosol; Fluorescence; Ultraviolet aerodynamic particle sizer; BioScout; Real-time detection. S. Saari Tampere University of Technology Department of Physics Korkeakoulunkatu Tampere Finland Tel.: Fax: sampo.saari@tut.fi Received for review 12 May Accepted for publication 30 September Practical Implications Fluorescence-based instruments are the only choice for real-time detection of fungal spores at the moment. In general, all fluorescence-based bioaerosol instruments are tested against known bacterial and fungal spores in laboratory conditions. This study showed that fungal species, growth substrate, age of culture, and air current exposure rate have an effect on detection efficiency of fungal spores in the fluorescence-based instruments. Therefore, these factors should be considered in the instrument calibration process. The results are also important when interpreting results of fluorescence-based field measurements of fungal spores. Introduction Water-damaged building structures enable mold to grow and subsequently cause problems in indoor air quality. Fungal spores can be released from contaminated materials into the air, and if inhaled, they may cause adverse health effects for people and animals (Burge and Rogers, 2000; Mendell et al., 2011; Peccia et al., 2008). Traditional bioaerosol detection methods such as Andersen impactor and filter sampling require a separate step for analysis before concentration can be determined and result in relatively low time resolution (Reponen et al., 2011). Laser-induced fluorescence (LIF) is a modern technique for real-time bioaerosol detection (Hill et al., 1995; Kaye et al., 2005; P ohlker et al., 2012). The LIF technique is an effective method for detecting biological molecules such as tryptophan, NAD(P)H, and flavins that are typically present in microbial cells (Lakowicz, 2006). The LIF enables the differentiation of bioaerosols from other particles through their fluorescence ability (Hill et al., 2013; P ohlker et al., 2012). Saari et al. (2013) demonstrated 653

2 Saari et al. that bacterial and fungal spores may be distinguished from each other through their dissimilar fluorescence spectra. The most well-known real-time LIF instrument is the ultraviolet aerodynamic particle sizer (UV- APS TM, TSI Inc., St. Paul, Minnesota), introduced by Hairston et al. (1997). The UVAPS is able to measure both aerodynamic particle size and autofluorescence of a single particle. UVAPS has been used for monitoring of airborne biological particles in swine confinements (Agranovski et al., 2004) and in outdoor environments (Huffman et al., 2010; Huffman et al., 2012). Recently, the UVAPS was used to characterize total and fluorescent biological aerosol particle levels in a classroom by continuously monitoring with 5-min resolution during eighteen occupied and eight unoccupied days distributed throughout a one-year period (Bhangar et al., 2014). A two-wavelength excitation bioaerosol instrument, Wide Issue Bioaerosol Sensor (WIBS, Kaye et al., 2005), has recently been used in several studies of atmospheric primary bioaerosol particles (e.g., Gabey et al., 2010; Healy et al., 2014). The WIBS enables multiband fluorescence analysis of single particles using two excitation wavelengths (280 and 370 nm) and two fluorescence bands, and may be a helpful instrument to find differences between the different types of bioaerosols. However, to our knowledge, no indoor applications of the WIBS have been presented yet. For more accurate interpretation of results on real-time LIF detection of fungal spores, fluorescence properties of spores should be well known in various conditions. In general, only a fraction of biological particles emit enough fluorescence light to be detected by the instrument. This fraction is here referred to as the fluorescent particle fraction (FPF) and represents the detection efficiency of the instrument. It is important to note that different instruments (UVAPS, WIBS, etc.) use different light sources and fluorescence thresholds resulting in instrument-specific detection efficiencies. Unknown detection efficiency causes miscounting spores in real environment. The performance of the UVAPS against two types of fungal spores has been reported by Kanaani et al. (2007, 2008). Their results showed that the FPF of spores of Aspergillus niger and Penicillium species varies between 48 and 99%, depending on the species and the physiological stage of the spores as well as the frequency of air current that releases the spores. They found decreasing FPF values with longer cultivation times and after multiple air current exposures. Lee et al. (2010) reported increasing FPF values for fungal spores (Aspergillus versicolor) when they were heated up to 400 C for 0.15 s. UV light exposure was shown to shift the fluorescence spectrum of Aspergillus niger spores from blue to yellow (Raimondi et al., 2009). The spores showed initially faint blue fluorescence emission that was assumed to come from NAD(P)H. The reason for yellow fluorescence emission at 540 nm after UV exposure was explained by the photooxidation of melanin on spore cell wall. Our recent study showed that a simple, 405-nm diode laser-based bioaerosol detector, BioScout (ENVI BioScout TM, Environics Ltd., Mikkeli, Finland), may be even more sensitive against common bioaerosols than the UVAPS (Saari et al., 2014). NAD(P)H, flavins, melanin, carotenoids, phenols, terpenoids, and DNA are the major fluorescent compounds in fungal spores and can absorb light at 405 nm (Raimondi et al., 2009; O Connor et al., 2011; P ohlker et al., 2012; Saari et al., 2013). In the present study, the two instruments are used side-by-side to study the effects of fungal species, growth substrate, cultivation time, and the aerosolization air velocity on fluorescence properties of fungal spores. Different fungal species have characteristic structures, and thus their fluorescent biochemical composition may be different. The growth substrate provides nutrients for the fungal growth and may affect the production of fluorescent compounds in spores. Using different air exposure velocities in the release process gives information on how the needed kinetic energy is associated with the fluorescence properties of released spores. This study provides information on major factors that can affect the fluorescence properties of airborne fungal spores. The results are useful for instrument calibration and interpretation of LIF-based field results on fungal spores. Materials and methods Fungal samples Three fungal species, Aspergillus versicolor (Culture collection of the Institute for Health and Welfare, Finland: HT31), Cladosporium cladosporioides (German collection of Microorganisms and Cell Cultures: DSMZ 62121), and Penicillium brevicompactum (American Type Cell Collection: ATCC 58606), were used in the experiments. These strains are common in indoor air worldwide (Hyv arinen et al., 2002; Reponen et al., 2012). The purity of the fungal aerosol was checked from TEM images (Figure S1). Furthermore, these strains were previously shown to produce pure fungal growth without any contamination when cultivated on different building materials in the incubation chamber (Mensah-Attipoe et al., 2015). The spore concentration was adjusted to about spores/ml before inoculating 0.5 ml of fungal suspension onto either agar plates or gypsum boards ( cm). Malt extract agar (ME) (LabM, Lancashire, UK) was used for A. versicolor and C. Cladosporioides, whereas dichloran glycerol 18% agar (DG18) (Merck, Darmstadt, Germany) was used for P. brevicompactum. Altogether 54 material samples were prepared: 27 agar plates and 27 gypsum board samples (three fungal species 9 two types of materials 9 three time 654

3 Fluorescence properties of fungal spores points 9 three replicates). Inoculated material samples were incubated at room temperature (21 2 C) in conditioned chambers (24 l) at a relative humidity of (95 97%) for, 1, and s. The humidity of 95 97% was achieved by placing a saturated K 2 SO 4 solution (150 g/l) on the bottom of the chamber (Korpi et al., 1997). A fan was placed in the chamber to circulate air uniformly. Temperature and humidity were measured once a week with a thermohygrometer (Vaisala Oyj, Helsinki, Finland). Experiments were carried out during winter (2013). The transparent incubation chambers were kept on the countertop in the laboratory. The chambers were exposed to daylight (6 h) and darkness (18 h) through the growth period. Samples were not stored but taken for aerosolization directly from the incubation chamber. Thus, the results as seen in the present study incorporate the combined effects of light and darkness on the fluorescence spectra of the spores released from the fungal growth. Before aerosolization experiments, the growth media were kept in dry (relative humidity of 18 2%) chambers overnight to dry out the fungal growth for easier aerosolization. Aerosolization The experimental setup is shown in Figure 1. The spores were released using the Fungal Spore Source Strength Tester (FSSST, Sivasubramani et al., 2004) either directly from growth on agar plate or from gypsum board samples. Three flow rates (5, 15, and 25 lpm) in FSSST were used to study the effect of extraction energy of spores. Effective air velocities on the growth surface under the FSSST nozzles were 5, 16, and 27 m/ s, respectively. The air velocity was increased step by step from 5 m/s to 27 m/s, and the total concentration, size distribution, and fluorescence properties of the aerosol were monitored for two minutes at each air velocity. After this flow ramp, the sample was changed in the FSSST and aerosolization was repeated two more times. The total flow rate in the setup was always 30 lpm. Dilution ratio between the flow rates was corrected. Fig. 1 The schematic experimental setup. Sample flow to the aerosol generator (FSSST) was adjusted using mass flow controllers (MFC). Released spores were measured by the ultraviolet aerodynamic particle sizer and the BioScout Instrumentation The fluorescence properties of released fungal spores were measured with two LIF-based real-time instruments, the BioScout and the UVAPS. BioScout. The operation principle of the BioScout is described in detail earlier (Saari et al., 2014). In brief, the BioScout has a 405-nm continuous wave laser diode with 200-mW optical power to excite autofluorescence from individual bioparticles. Particle size measurement is based on scattered light. The operative particle size range is optimized between 0.3 and 5 lm. The optical system of the BioScout was developed at the Tampere University of Technology and the instrument was commercialized by Environics Ltd. UVAPS. The current version of UVAPS (TSI Model 3014) measures aerodynamic diameter of particles between 0.5 and 15 lm with 52 channels. The UVAPS also measures the autofluorescence emission of individual particles using a pulsed 355-nm UV laser with 80 mw maximum power as an excitation source. In this study, the UV pulse energy and fluorescence PMT gain were set to their default values. Data analysis The BioScout data were analyzed using MATLAB software (MathWorks Inc., Natick, MA, USA). All particles in fluorescence channels 2 16 were classified as fluorescent and particles in channel 1 as non-fluorescent. Background levels from blank agar and gypsum samples were measured and deducted from the results. The concentrations of fluorescent particles and total particle concentration were calculated in the spore size range (1 5 lm). Fluorescent particle fraction (FPF) was calculated by dividing the concentration of fluorescent particles with the total particle concentration. The FPF represents the detection efficiency of the instrument, assuming that only fluorescent particles are bioaerosol particles. The UVAPS data were analyzed similarly, and all particles in fluorescence channels 2 64 were classified as fluorescent. The UVAPS fluorescence channel 2 was chosen as the limit for fluorescent particles, following the procedure used in laboratory studies by Agranovski et al. (2003), Kanaani et al. (2007), and Saari et al. (2014). Statistical significance of different factors was first analyzed using 4-way ANO- VA method. The first two models included the effects of fungal species, age of the culture, aerosolization air velocity, and growth substrate separately for BioScout and UVAPS data. Next, three stratified models included the effects of instrument, age, aerosolization air velocity, and growth substrate separately for each of the three species. After that, a more detailed analysis of different factors was conducted using 1-way ANOVA. 655

4 Saari et al. Results The FPF values of spores released from agar ranged between 0.18 and 0.79 with the BioScout and between 0.08 and 0.73 with the UVAPS. Total particle concentration ranges are shown in Table S1 and size distributions in Figure S2. The high variation in the concentrations is mainly caused by the different species and cultivation times. The variation was much smaller within the repeats of each experiment, that is, when data were stratified for species, incubation time, and air velocity as shown in Figure S3. All the particle size distributions showed narrow peaks in the spore size range of 2 4 lm, which support the TEM results on the purity of fungal aerosols. For gypsum board, only the results with 27 m/s air velocity are presented, because particle release from gypsum samples with the two lower air velocities was not distinguishable from the background. Results on 4-way ANOVA analysis An overview on the effects of different factors on the FPF values are shown in Table 1. The level of significance of fungal species, age of the culture, aerosolization air velocity, and growth substrate was high for the UVAPS response. The same was true for the BioScout, with the exception that the growth substrate showed lower significance. When the data were stratified by species, all the investigated variables were highly significant for the FPF of A. versicolor spores. In contrast, only the instrument was a significant factor for Table 1 Summary of the results of 4-way ANOVA on the effect of different factors on the fluorescent particle fraction (FPF) of released fungal spores Instrument Species Variable P-value Significance BioScout All Species Culture age Air velocity Substrate * UVAPS All Species Culture age Air velocity Substrate Both A.v. Instrument Culture age Air velocity Substrate C.c. Instrument * Culture age Air velocity Substrate NS P.b. Instrument ** Culture age NS Air velocity NS Substrate NS A.v., Aspergillus versicolor; C.c., Cladosporium cladosporioides, P.b., Penicillium brevicompactum. The level of significance is marked with asterisks (, P < 0.001; **, P < 0.01; *, P < 0.05; NS, not significant). P. brevicompactum spores. For C. cladosporioides spores, the age, air velocity, and type of instrument were significant variables, whereas the type of substrate was not significant. The results showed that all the investigated factors can have considerable effect on the response of the fluorescence-based bioaerosol instruments. The results of more detailed analysis of different factors using 1-way ANOVA is discussed below. Effect of the species The FPF values for spores released from agar are shown in Figure 2. The results are averages of all the differently aged samples generated with 27 m/s air velocity. A. versicolor had the highest FPF value (0.61) with the BioScout whereas P. brevicompactum had the highest value (0.50) with the UVAPS. C. cladosporioides had the lowest FPF values (0.18 and 0.08) with both instruments (BioScout and UVAPS). The 1-way ANOVA showed statistically significant differences (P < 0.01) between A. versicolor and C. cladosporioides and between C. cladosporioides and P. brevicompactum with both instruments. Difference between A. versicolor and P. brevicompactum was significant (P < 0.05) only with the BioScout. The trend was similar with gypsum board samples at 27 m/s, as can be seen in Figure 3. Effect of the growth substrate The FPF values for spores from agar and gypsum board samples are shown in Figure 3. The results are averages of all the differently aged samples generated with 27 m/s air velocity. The growth substrate could be compared only using the results with the highest air velocity, because particle release from gypsum samples was low with the two lower air velocities. The UVAPS FPF values from agar varied between 0.08 and 0.50 and from gypsum between 0.11 and 0.39, whereas corresponding values for the BioScout from agar varied between 0.18 and 0.61 and from gypsum between 0.15 and The FPF of the spores was in most cases higher from agar plates, but this difference was significant only with A. versicolor cultures (P-values were and for UVAPS and BioScout, respectively). The trends were similar with both instruments. Effect of cultivation time Figure 4 shows fluorescent particle fractions of spores released from 1-week-, 1-month-, and 4-month-old agar samples. The values are averages of results obtained with all three air velocities used for spore aerosolization. The BioScout FPF values for all the different aged samples varied between 0.54 and 0.79, 0.23 and 0.33, and 0.33 and 0.50 for A. versicolor, C. cladosporioides, and P. brevicompactum, respectively. 656

5 Fluorescence properties of fungal spores Fig. 2 Fluorescent particle fraction of spores released from agar with 27 m/s air velocity (averages of all cultivation times). Variability bars represent standard deviations of three repeats. Asterisks indicate the level of statistical significance by 1-way ANOVA method of two data points (, P < 0.001; **, P < 0.01; *, P < 0.05) Fig. 3 Fluorescent particle fraction of spores released from agar and gypsum board with 27 m/s air velocity (averages of all cultivation times). Variability bars represent standard deviations of three repeats. Asterisks indicate the level of statistical significance by 1-way ANOVA method (, P < 0.001; **, P < 0.01) The corresponding ranges for the UVAPS FPF values were , , and The FPF typically was highest for the youngest samples. The exception was P. brevicompactum that had the lowest FPF values with the youngest samples. Statistically significant differences in the FPF values were found between 1-week-old and 1-month-old cultures as well as between 1-week- and 4-month-old samples, whereas 1-month- and 4-month-old samples exhibited typically similar values. Similar trends were observed for the FPF values for spores released from gypsum board samples at an air velocity of 27 m/s (data not shown). Effect of air velocity The FPFs of spores released from agar samples with different air velocities are shown in Figure 5. The values are averages of all samples with different cultivation times. The FPF values decreased with increasing air velocities for all the species, except P. brevicompactum measured by the BioScout that showed an opposite trend. The decreasing trend was statistically significant for A. versicolor and C. cladosporioides. Similar trends were observed for spores released from gypsum board samples (data not shown). Total concentrations of spores typically increased with higher air velocities, and upper concentration limit for the fluorescence measurement of the UVAPS (60 #/cm 3 ; Agranovski et al., 2003) was reached with P. brevicompactum with higher air velocities (16 and 27 m/s) (see Figure S3). This arguably affected the decreasing FPF values for P. brevicompactum with higher air velocities, so these data points are not comparable. 657

6 Saari et al. Fig. 4 Fluorescent particle fractions from 1-week (1w)-, 1-month (1 m)-, and 4-month (4 m)-old agar (average of all aerosolization air velocities). Variability bars represent standard deviations of three repeats, except the 1 m-old P.b. that had only 2 successful repeats. Asterisks indicate the level of statistical significance (1-way ANOVA) of two data points (, P < 0.001; **, P < 0.01; *, P < 0.05) Fig. 5 Fluorescent particle fractions of spores released from agar samples with different air velocities (average of all cultivation times). Variability bars represent standard deviations. Asterisks indicate the level of statistical significance (1-way ANOVA) of two data points (, P < 0.001; **, P < 0.01; *, P < 0.05) Instrument comparison Figure 6 shows differences of fluorescent particle fractions between the instruments. The values are averages of results from agar samples obtained with all three air velocities used for spore aerosolization. A. versicolor had higher FPF values with the BioScout whereas P. brevicompactum had higher FPF values with the UVAPS. This difference was statistically significant only with 4-month-old cultures (P < for A. versicolor and P < 0.01 for P. brevicompactum). Discussion Our results showed that FPF values were typically lower for spores released from gypsum board than from agar. This indicates that poor nutrient substrate produces spores with lower amounts of fluorescent compounds or the metabolic activity of spores is lower. Viability has also been reported to correlate with FPF Fig. 6 Comparison of the instruments. Fluorescent particle fractions from 1-week (1w)-, 1-month (1 m)-, and 4-month (4 m)-old agar (average of all air velocities). Variability bars represent standard deviations of three repeats. Asterisks indicate the level of statistical significance by 1-way ANOVA method (, P < 0.001; **, P < 0.01) 658

7 Fluorescence properties of fungal spores values of spores (Kanaani et al., 2008), so our results indicate also lower viability for spores released from gypsum board. To our knowledge, this is the first study showing the effect of the growth material on the fluorescence properties of fungal spores. In general, all fluorescence-based bioaerosol instruments have been tested against known bacterial and fungal spores in laboratory conditions (e.g., Agranovski et al., 2003; Kanaani et al., 2008). In these cases, microorganisms have typically been grown on substrates rich in nutrients, such as agar and have resulted in good detection efficiency for the instruments. In real world, microbial growth often arises on surfaces with poor nutrient substances, such as wood and wall paper, in which case the detection efficiencies of the instruments may be lower. Therefore, spore concentrations may be underestimated with fluorescence-based instruments in real-world measurements. It is interesting that C. cladosporioides spores had clearly lower FPF values compared to the other species considering the fact that C. cladosporioides is very abundant strain both in indoor and outdoor environments (Despres et al., 2012; Healy et al., 2014; Hyv arinen et al., 2002). The results showed only 8% and 18% detection efficiencies for the UVAPS and the BioScout, respectively, indicating that C. cladosporioides concentrations will be underestimated in field measurements. This is consistent with atmospheric measurements with the WIBS and the UVAPS (Healy et al., 2014). The reasons for the decreasing FPF values with older A. versicolor and C. cladosporioides samples are not clear. The age of growth can affect fluorescent biochemical composition of spores, for example, metabolic activity may be lower with older spores or protective coating of spores may become thicker with older spores. Kanaani et al. (2007) reported decreasing FPF values for Aspergillus niger and Penicillium spores measured by the UVAPS when culturing time of agar samples increased from 2 to 21 days. As viability has been shown to correlate with FPF values of spores (Kanaani et al., 2008), our results indicate lower viability for older spores. The study by Raimondi et al. (2009) reported higher fluorescence from younger fungal spores at 440 nm, which was assumed to come from protein-bound NAD(P)H. This is consistent with the results for A. versicolor and C. cladosporioides presented here but differ from P. brevicompactum results. Our results indicate that the life cycle of A. versicolor and C. cladosporioides spores may be different compared to P. brevicompactum or that the major fluorescent compounds of the species differ. The BioScout and the UVAPS gave comparable FPF results, except for the 4-month-old A. versicolor and P. brevicompactum spores. For the older A. versicolor spores, the BioScout gave higher FPF values, whereas the opposite was true for the older P. brevicompactum spores. Exhibited fluorescence depends on the fluorescent compounds in fungal spore and the selected excitation wavelength as discussed in the previous studies (P ohlker et al., 2012; Saari et al., 2013). As the instruments have different spectral ranges, the results indicate different fluorescent compounds production for A. versicolor and P. brevicompactum in the late stage of the spores. A multiband fluorescence instrument, such as the WIBS, can be useful to find differences in fluorescence between the species especially for older spores. Saari et al. (2013) reported somewhat different fluorescence excitation and emission spectra for P. solitum and A. versicolor spores. A previous study reported that UV light exposure can shift fluorescence spectrum of Aspergillus niger spores to longer wavelengths (Raimondi et al., 2009). Taken together, these results indicate that fluorescent compounds in spores depend on fungal species and may change during the spore lifetime. This needs further studies, including fluorescence spectral measurements of fungal spores. The results showed higher FPF values with lower air velocities in the FSSST during aerosolization. This indicates that easily released, presumably fully developed spores have more fluorescent compounds compared to forcibly extracted non-matured spores. Similar observations were reported by Kanaani et al. (2007). They found that releasing spores from the same growth multiple times decreased the FPF value of spores. Thus, the amount and frequency of kinetic energy applied to spores during aerosolization influences the maturity of the released spores and thereby, their fluorescence properties. As discussed above, fluorescence-based bioaerosol instruments are typically tested in laboratory, in which case air velocity should be well chosen if dry generation method is applied for the aerosolization of fungal spores. When the FSSST and similar methods are used to estimate mold contamination in field experiments (e.g., Adhikari et al., 2010; Niemeier et al., 2006; Sivasubramani et al., 2004), the operating air velocity should be considered if fluorescence based method is used for spore analysis. Based on this study, the highest FPF values, but lowest particle concentrations, were shown with the lowest air velocity. On the other hand, higher air velocity released more spores, but they had lower FPF values indicating also lower viability. Differences in the concentration with different air velocities were clearly higher compared to differences in the FPF values. Based on these results, we suggest using a higher air velocity if maximum concentration of viable spores is needed. This study showed that several factors can have effect on fluorescence properties of fungal spores. Statistically significant differences were observed between fungal species, culture ages, aerosolization air velocities, growth substrates, and instruments. However, the variations of the FPF values were moderate, for example, 659

8 Saari et al. the BioScout FPF values from agar for A. versicolor varied between 0.54 and 0.79 and for C. cladosporioides and Penicillium brevicompactum between 0.23 and 0.33 and 0.33 and 0.50, respectively. Although detection efficiency of the fluorescence-based instruments is affected by various factors described above, they are still the only choice for real-time detection of fungal spores at the moment. Conclusions The effects of the fungal species, cultivation time, growth substrate, and air velocity on the fluorescent properties of fungal spores were studied using the UV- APS and the BioScout. A unique new finding was that the FPF values were typically lower for spores that had grown on poor nutrient substrate. The results also showed that C. cladosporioides spores had remarkably lower FPF values compared to A. versicolor and P. brevicompactum species, so C. cladosporioides spore concentrations may be easily underestimated with fluorescence-based instruments in field studies. The FPF typically decreased with older samples, except the P. brevicompactum samples that had an opposing trend. The results indicate different fluorescent compounds production between A. versicolor and P. brevicompactum in the late stage of spores. The higher FPF with lower aerosolization air velocity indicates that easily released fully developed spores have more fluorescent compounds compared to forcibly extracted raw spores. The current study gives new information about the fluorescence properties of fungal spores in various conditions. The results can be utilized in the real-world measurements of fungal spores as well as in instrument calibration and testing in laboratory conditions. Acknowledgements The work is financially supported by Doctoral School of TUT and TEKES FiDiPro project named BITEFA (decision number 40371/11). The private financial supports of Jenny and Antti Wihuri Foundation and Eemil Aaltonen Foundation are also gratefully acknowledged. Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. TEM images of aerosolized fungal spores collected onto copper grids (400 mesh) with holey carbon film (Agar S147-4). Figure S2. Typical total particle size distributions (UV- APS) of particles aerosolized with 16 m/s air velocity from one month old fungal cultures grown on gypsum board and agar. Figure S3. Total particle concentrations of particles (1 5 lm) aerosolized with different flow rates from fungal cultures grown on agar and measured by the UVAPS. Table S1. Ranges (min max) for concentration and total number of fungal spores measured by the Bio- Scout and the UVAPS per sample. References Adhikari, A., Lewis, J.S., Reponen, T., De- Grasse, E.C., Grimsley, L.F., Chew, G.L., Iossifova, Y. and Grinshpun, S.A. (2010) Exposure matrices of endotoxin, (1?3)-b-D-glucan, fungi, and dust mite allergens in flood-affected homes of New Orleans, Sci. Total Environ., 408, Agranovski, V., Ristovski, Z., Hargreaves, M., Blackall, P. and Morawska, L. (2003) Real-time measurement of bacterial aerosols with the UVAPS: performance evaluation, J. Aerosol Sci., 34, Agranovski, V., Ristovski, Z., Blackall, P.J. and Morawska, L. (2004) Size-selective assessment of airborne particles in swine confinement building with the UVAPS, Atmos. Environ., 38, Bhangar, S., Huffman, J.A. and Nazaroff, W.W. (2014) Size-resolved fluorescent biological aerosol particle concentrations and occupant emissions in a university classroom, Indoor Air, 24, Burge, H.A. and Rogers, C.A. (2000) Outdoor allergens, Environ. Health Perspect., 108, Despres, V.R., Huffman, J.A., Burrows, S.M., Hoose, C., Safatov, A.S., Buryak, G. and Jaenicke, R. (2012) Primary biological aerosol particles in the atmosphere: a review, Tellus, Series B: Chemical and Physical Meteorology, 64, art. no Gabey, A.M., Gallagher, M.W., Whitehead, J., Dorsey, J.R., Kaye, P.H. and Stanley, W.R. (2010) Measurements and comparison of primary biological aerosol above and below a tropical forest canopy using a dual channel fluorescence spectrometer, Atmos. Chem. Phys., 10, Hairston, P.P., Ho, J. and Quant, F.R. (1997) Design of an instrument for realtime detection of bioaerosols using simultaneous measurement of particle aerodynamic size and intrinsic fluorescence, J. Aerosol Sci., 28, Healy, D.A., Huffman, J.A., O Connor, D.J., P ohlker, C., P oschl, U. and Sodeau, J.R. (2014) Ambient measurements of biological aerosol particles near Killarney, Ireland: a comparison between realtime fluorescence and microscopy techniques, Atmos. Chem. Phys., 14, Hill, S.C., Pinnick, R.G., Nachman, P., Chen, G., Chang, R.K., Mayo, M.W. and Fernandez, G.L. (1995) Aerosol-fluorescence spectrum analyzer: real-time measurement of emission spectra of airborne biological particles, App. Optics, 34, Hill, S., Pan, Y.-L., Williamson, C., Santarpia, J. and Hill, H. (2013) Fluorescence of bioaerosols: mathematical model including primary fluorescing and absorbing molecules in bacteria, Opt. Express, 21, Huffman, J., Treutlein, B. and P oschl, U. (2010) Fluorescent biological aerosol particle concentrations and size distributions measured with an Ultraviolet Aerodynamic Particle Sizer (UV-APS) in Central Europe, Atmos. Chem. Phys., 10, Huffman, J.A., Sinha, B., Garland, R.M., Snee-Pollmann, A., Gunthe, S.S., Artaxo, P., Martin, S.T., Andreae, M.O. and P oschl, U. (2012) Size distributions and temporal variations of biological aerosol

9 Fluorescence properties of fungal spores particles in the Amazon rainforest characterized by microscopy and real-time UV-APS fluorescence techniques during AMAZE-08, Atmos. Chem. Phys., 12, Hyv arinen, A., Meklin, T., Veps al ainen, A. and Nevalainen, A. (2002) Fungi and actinobacteria in moisture-damaged building materials concentrations and diversity, Int. Biodeterior. Biodegradation, 49, Kanaani, H., Hargreaves, M., Ristovski, Z. and Morawska, L. (2007) Performance assessment of UVAPS: influence of fungal spore age and air exposure, J. Aerosol Sci., 38, Kanaani, H., Hargreaves, M., Smith, J., Ristovski, Z., Agranovski, V. and Morawska, L. (2008) Performance of UVAPS with respect to detection of airborne fungi, J. Aerosol Sci., 39, Kaye, P., Stanley, W.R., Hirst, E., Foot, E.V., Baxter, K.L. and Barrington, S.J. (2005) Single particle multichannel bioaerosol fluorescence sensor, Opt. Express, 13, Korpi, A., Pasanen, A.L., Pasanen, P. and Kalliokoski, P. (1997) Microbial growth and metabolism in house dust, Int. Biodeter. Biodegr., 40, Lakowicz, J.R. (Ed.) (2006) Principles of Fluorescence Spectroscopy, 3rd edn, Berlin, Springer, Lee, B.U., Jung, J.H., Yun, S.H., Hwang, G.B. and Bae, G.N. (2010) Application of UVAPS to real-time detection of inactivation of fungal bioaerosols due to thermal energy, J. Aerosol Sci., 41, Mendell, M.J., Mirer, A.G., Cheung, K. and Douwes, J. (2011) Respiratory and allergic health effects of dampness, mold, and dampness-related agents: a review of the epidemiologic evidence, Environ. Health Perspect., 119, 748. Mensah-Attipoe, J., Reponen, T., Salmela, A., Veijalainen, A.M. and Pasanen, P. (2015) Susceptibility of green and conventional building materials to microbial growth, Indoor Air, 25, Niemeier, R.T., Sivasubramani, S.K., Reponen, T. and Grinshpun, S.A. (2006) Assessment of fungal contamination in moldy homes: comparison of different methods, J. Occup. Environ. Hyg., 3, O Connor, D.J., Iacopino, D., Healy, D.A., O Sullivan, D. and Sodeau, J.R. (2011) The intrinsic fluorescence spectra of selected pollen and fungal spores, Atmos. Environ., 45, Peccia, J., Milton, D.K., Reponen, T. and Hill, A.J. (2008) A role for environmental engineering and science in preventing bioaerosol-related disease, Environ. Sci. Technol., 42, P ohlker, C., Huffman, J. and P oschl, U. (2012) Autofluorescence of atmospheric bioaerosols fluorescent biomolecules and potential interferences, Atmos. Meas. Tech., 5, 37 71, Raimondi, V., Agati, G., Cecchi, G., Gomoiu, I., Lognoli, D. and Palombi, L. (2009) In vivo real-time recording of UVinduced changes in the autofluorescence of a melanin-containing fungus using a micro-spectrofluorimeter and a low-cost webcam, Optics Expr., 17, , Reponen, T., Willeke, K., Grinshpun, S. and Nevalainen, A. (2011) Biological particle sampling, Aerosol Measurement: Principles, Techniques, and Applications, Third Edition, Reponen, T., Lockey, J., Bernstein, D.I., Vesper, S.J., Levin, L., Khurana Hershey, G.K., Zheng, S., Ryan, P., Grinshpun, S.A., Villareal, M. and LeMasters, G. (2012) Infant origins of childhood asthma associated with specific molds, J. Allergy Clin. Immunol., 130, Saari, S.E., Putkiranta, M.J. and Keskinen, J. (2013) Fluorescence spectroscopy of atmospherically relevant bacterial and fungal spores and potential interferences, Atm. Environ., 71, Saari, S., Reponen, T. and Keskinen, J. (2014) Performance of two fluorescence based real-time bioaerosol detectors: Bio- Scout vs. UVAPS, Aerosol Sci. Technol., 48, Sivasubramani, S.K., Niemeier, R.T., Reponen, T. and Grinshpun, S.A. (2004) Assessment of the aerosolization potential for fungal spores in moldy homes, Indoor Air, 14,