Large numbers of airborne microorganisms, mostly spores of fungi and actinomycetes, can occur in agricultural and other working environments and cause respiratory disease. Development of occupational asthma is characteristically determined by the atopic state of the individual rather than the abundance of spores in his environment but extrinsic allergic alveolitis (hypersensitivity pneumonitis) and organic dust toxic syndrome (RYLANDER, 1986; DOPICO, 1986) require heavy exposure to airborne spores.
Concentrations usually exceed 106 m-3 air and often reach 101 0 m -3. (LACEY et al., 1972; RYLANDER, 1986). Sampling such highly contaminated environments presents many problems to the microbiologist.
Traditionally, air sampling methods that deposit spores directly onto the surface of agar plates have been used, e.g., the Andersen sampler (ANDERSEN, 1958) and slit sampler (BOURDILLON et al., 1941). However, their usefulness is restricted by the ease with which plates become overloaded, necessitating short sampling times with a consequent large variability between samples.
Sampling into liquid, using impingers, allows dilution of the catch but cells may be damaged where the impinger jet also functions as a critical orifice.
Various methods of overcoming these limitations have been suggested, including homogenisation of agar from exposed plates, dilution and replating (BLOMQUIST et al., 1984a, b) while PALMGREN et al. (1986) used filtration through polycarbonate membrane filters (Nuclepore) housed in polystyrene aerosol monitors and operated by personal sampler pumps.
Methods for enumerating and classifying mould spores by growing them in culture usually have poor precision and accuracy (LACEY and DUTKIEWICZ, 1976 ; MORRING et al., 1983). Results may be affected by the viability of the organisms, choice of nutrient medium, conditions for growth especially incubation temperature, and interactions between different organisms, especially when plates are overcrowded. However, microorganisms that grow can
-39-
usually be identified to species level. Because all spores, both living and dead, may cause disease, all spores should be counted. Methods used have included light microscopy (GREGORY and LACEY, 1963), epifluorescence microscopy (PALMGREN et al., 1986) and scanning electron microscopy (EDUARD et al., 1988).
Filtration using membrane filters offers several advantages over other methods of sampling, chiefly its extreme flexibility and portability. It can be used either for area or for personal sampling, if necessary remote from the laboratory when the exposed monitors can easily be returned by mail. The catch may also be assessed in several different and complementary ways:
directly, by light or scanning electron microscopy; after resuspension, filtration and staining by fluorescence microscopy; after resuspension and dilution, by culturing on different agar media; by analysis for mycotoxins, assay for endotoxins or by testing for specific allergens. For most of these purposes, polycarbonate membrane filters offer advantages over cellulose ester membranes (POPENDORF, 1986). However, methods of clearing polycarbonate filters for direct counting by light microscopy have not yet been tested.
This paper describes collaborative studies to evaluate different microscopic methods for assessing the load of spores of fungi, actinomycetes and bacteria in parallel filter samples collected from different occupational environments and their attempted harmonization by discussion between the laboratories concerned. Five laboratories participated: the National Institute of Occupational Health, Oslo, Norway (AMY); the National Institute of Occupational Health, Solna, Sweden (IAS) and Umea (IAU); the Swedish University of Agricultural Science, Uppsala, Sweden (SLU); and the A.F.R.C.
Institute of Arable Crops Research, Rothamsted Experimental Station, Harpenden, U.K. (RES).
2.3 MATERIALS AND METHODS
2.3.1. Sampling sites and distribution of samples.
Sampling sites (Table 1) were representative of different types of microbial exposure by workers on farms. All samples were taken by courier to Stockholm with minimal disturbance. They were then randomized and two samples from each site were sent, without identification of their source or indication as to which were parallel samples, to the five collaborating laboratories, again by courier. Samples for culture were mailed to RES.
Table 1. Sampling sites, periods of sampling and laboratory collecting samples.
Work environment Sampling period Laboratory1
(min)
Saw mill sorting and trimming plant 60 AMY
Woodchip handling2 5 IAU
Grain elevator2 10 IAU
Pig confinement building 60 SLU Straw handling 75 SLU
1, For abbreviations, see text.
2, Four samples were collected with filter-holders fitted with 1.3 cm extension rings instead of 5cm. These were all examined by fluorescence microscopy at IAU and SLU.
2.3.2. Collection of samples.
Series of ten parallel samples were collected at each of the five sites, normally using 37 mm diameter, closed-face filter holders (Nuclepore), each with a 5 cm extension ring. The intake velocity of the closed-face filter holders was 1 m s -1 and the long extension tube corresponds to British Health and Safety Executive sampling recommendations (Health and Safety Executive, 1984) and was assumed to assist the even distribution of particles on the filter. However, one filter holder broke open during shaking of a long filter
-41-
holder to resuspend spores for fluorescence microscopy, causing the sample to be lost. Consequently, filter holders with 1.3 cm extension rings were used for later samples (Table 1). The filter holders were mounted in the apparatus shown in Fig. 1 which was designed to provide good parallel samples. Air flow through the filters was provided by applying a vacuum to the plenum chamber behind the holders with a vacuum pump. Flow through the individual filters was controlled to about 1.5 1 min-1 using critical orifices and was checked at the start and finish of each sampling period. The calculated intake velocity of the sampling apparatus was ca 5 cm s_ 1 so that particles larger than 40 urn should not have been sampled. Within the sampling apparatus, air velocity decreased to ca 4 mm s-i, assuming laminar flow, and particles >10 mm were probably lost through elutriation.
vacuum pump
volume 10dm
Fig. 1. Apparatus used to collect parallel samples.
Filters to be examined by fluorescence microscopy (FM) and scanning electron microscopy (SEM) were collected on polycarbonate membrane filters (Nuclepore), with 0.4 mm pores, and those for light microscopy (LM) on cellulose ester membrane filters (Millipore), with 0.8 mm pores. Separate
samples were usually collected simultaneously with the parallel samples on cellulose ester membranes (Millipore), 0.8 pro pore size, for isolation of microorganisms in culture. However, the saw mill sample for culturing was collected simultaneously with those for the transportation loss study.
2.3.3. Distribution of particles in conducting filter-holders and transportation losses.
To investigate the distribution of particles on filters held in conducting filter-holders, one series of ten replicate parallel samples was collected, using the sampling apparatus shown in Fig. 1, in the sorting and trimming department of a saw mill (site A ) . Open-face, 25 mm diameter, graphite-filled polypropylene filter-holders (Nuclepore) were used. However, the 5 cm extension ring supplied was replaced with a specially prepared 1.3 cm extension ring with a shape identical to that of the standard three-piece non-conducting aerosol monitor. Three filters, in their holders, were transported to the laboratory by each of three methods:
(a) by car as carefully as possible;
(b) "carelessly" by car;
(c) by mail.
Counts were then made by SEM at AMY while the tenth filter was cultured at RES.
2.3.4. Losses on filter-holder walls.
After removing the filters from four sets of the duplicate air samples analysed by AMY, further clean polycarbonate filters were inserted into the filter-holders. Ten ml of 0.1% Tween 80 was then added to long 37 mm diameter filter-holders or 5 ml to 6hort 25 mm diameter filter-holders. Spores deposited on the walls were then resuspended by agitating in an ultrasonic bath for 2 min. The suspension was filtered through 25 mm diameter
-43-
polycarbonate filters which were examined by SEM at AMY. Three conducting filter holders used in the transportation study, one from each transportation method, were analysed in the same way.
2.3.5. Preparation and examination of samples.
2.3.5.1. Light microscopy (RES).
Filters were cut into two well clear of the mid-line and the larger segment mounted in glycerol triacetate (DARKE et al. , 1976) on a standard glass microscope slide and covered with a 26 x 50 mm cover slip. Counts were made of about 60 viewing fields 1 mm apart along three radii using a magnification of x625. The counting area was defined using an eyepiece graticule which divided the field of view into 18.1 mm squares. One square at the centre of the field of view was counted in samples with dense deposits of actinomycete spores but in samples with less dense deposits and for fungal spores 12 squares in a line at right angles to the radius were counted.
2.3.5.2. Fluorescence microscopy (IAU, SLU).
Filter deposits were dispersed in 0.01% sterile aqueous Tween 80 solution using the following procedure. First, 1ml of the solution was pipetted into the outlet port of the monitor to wet the cellulose support pad. Then 5 ml was pipetted through the inlet port and, after replugging both ports, the monitor was sliaken for 15 min. on a shaking table. The spores were then fixed in formaldehyde (1% w/v) and a suitable volume was taken, stained with acridine orange and filtered through a 25 mm diameter black Nuclepore filter and mounted in Cargill oil type A (ZIMMERMAN & MEIER-REIL, 1974). Filters were examined with an epifluorescence microscope at a magnification of xlOOO.
Usually, 40 random fields were counted, to give a total of about 400 microorganisms, but if the density of the deposit greatly exceeded ten spores per field, fewer fields were counted (PALMGREN et al., 1986; POPENDORF, 1986).
2.3.5.3. Scanning electron microscopy (AMY, IAS).
A 90° segment, cut from the filter with a generous margin, was mounted on a 32 mm diameter aluminium stub with carbon cement or silver paint and coated with gold in a vacuum (< 30Pa) with a Jeol JFC-1100 sputter coater. Specimens were examined in a JSM-35 scanning electron microscope. Counts were made of about 270 fields spaced about 1 mm apart in a square grid pattern, covering the whole surface of the filter specimen. The magnification was chosen so that a minimum of 250 spores were counted, providing there were sufficient spores on the segment.
2.3.6. Counting procedure.
With all methods, counts were made of total spores, spore containing units (SCU; includes single spores, aggregates of spores and other particles carrying microorganisms), single spores and of spores in aggregates within defined counting areas. All spores falling totally within the viewing fields were counted but where spores or aggregates touched the margins of the viewing field, only those touching or crossing the margins on two 6ides of the field were counted (+, Fig. 2) and those touching or crossing the other two margins (-, Fig. 2) were ignored. Fig. 2 also shows the methods used to record aggregates crossing the + and - boundaries of the field of view.
2.3.7. Classification of spores.
Spores counted by light microscopy were classified by their size morphology and colour into different taxonomic groups: "actinomycetes + bacteria", spherical spores about 1 mm diameter; A s p e r g i l l u s / Penicillium, spherical spores >2.5 mm; Aspergillus glaucus, ovoid spores about 5 mm diameter with roughened surface; Humicola lanuginosa, dark brown spherical spores with warty surface 6-10 mm diameter; Wallemia sebi, small cuboid spores mostly 2-2.5 mm diameter; Mucoraceae, spherical or ovoid, thin-walled smooth hyaline spores; Rhizopus, spherical, ovoid or angular, lightly coloured,
-45-
2/(6)
3/6(8)
Fig. 2. Counting rules and methods used to record aggregates.
Key: +, spores and aggregates touching edge of field counted -, spores and aggregates touching edge of field not
counted
1, counted as 2/(6); two spores forming part of an aggregate of six crossing a - side
2, counted as 4/(6); four spores forming part of an aggregate of six crossing a + side
3, counted as 3/6(8); three spores forming part of an aggregate of eight, two of which cross a - side
sometimes striate spores, variable in size. These and other 6pore types are illustrated in GREGORY (1973). Spores counted by SEM could be classified into actinomycetes + bacteria and fungal spores on the basis of size.
Actinomycetes and bacteria were smaller than 1.5 pm. Fungal spores from saw mills were further classified by morphology as: Rhizopus microsporus var.
rhizopodiformis, spores ovoid or angular with fine protrusions, 2.5 - 6 pm diameter; Paecilomyces variotii, spores ellipsoid, smooth with distinct, scars, 1.5-3.5 x 3.5-6 pm; Aspergillus fumigatus, spores appearing cubical or spherical with distinct protrusions 1.5-3 mm diameter (EDUARD et al.., 1988).
Visual classification was supported and extended to species level by identification of colonies grown in culture using standard texts (see LACEY et al., 1980).
2.3.8. Calculation of results.
Concentrations of total airborne spores, single spores and of spores in SCUs in each environment were calculated from the mean counts obtained by each method. However, account was also taken of the position of spores on the
filters when counting by LM at RES and fields were combined in concentric rings of 1 ran width. Distributions of spores on the filters were then assessed and results calculated using different ring widths to assess errors from uneven distributions of particles.
2.3.9. Calculation of counting precision.
Precision can be calculated from counts of duplicate filters using equation ( 1 ) :
where di is the difference and xi the average concentration of the i th duplicate.
The precision of an analytical result obtained by counting is dominated by counting precision when small numbers of particles are counted. Total precision can therefore be approximated by counting precision. Counts of randomly distributed particles follow a Poisson distribution and the relative precision (rsc) can then be predicted by equation ( 3 ) :
(2)
(4) where d is the difference between duplicates and n is the number of duplicates (YOUDEN, 1951).
The results of four series were pooled for each laboratory to obtain an estimate of the precision. Since spore concentrations ranged from 1 07 to 1 09
spores m-3, the relative differences were pooled using equation ( 2 ) . (1)
(3) where N is the number of particles counted (single spores or S C U ) .
Spores collected on filters are not randomly distributed but many are deposited in aggregates which greatly affect counting precision. EDUARD and AALEN (1988) found that the relative counting precision of the total spores could be predicted by equation ( 4 ) .
-47-
where f(a) = the relative frequency of aggregates with size a and Ns is the number of spores counted.
2.3.10. Culture of microorganisms from filter samples.
The cellulose support pad was first wetted by adding 1ml of suspension fluid (0.1% bacteriological peptone, 0.05% Tween 80, 2% inositol) through the exit port before a further 5 to 8 ml of suspension fluid was added to the membrane filter through the inlet port. The whole monitor was then shaken on a Whirlimixer (Fisons) for 1 min. Serial dilutions were then prepared by taking 1 ml of suspension, first from the monitor and then from successive dilutions after shaking on a Whirlimixer, and adding them to 9 ml 1/4-strength Finger's solution. Dilutions down to 10-7 to 10-11 were prepared and at least two agar plates of each dilution/medium/temperature combination were then inoculated with 0.1 ml of suitable dilutions. This was spread over the agar surface with a bent glas6 rod using a turntable. For isolating fungi, 2% malt extract agar containing penicillin (20 I.U. ml- 1) and streptomycin (40 units ml-1) was used; for xerophilic fungi, dichloran-glycerol agar (DG18, Oxoid);
for bacteria and actinomycetes, 1/2-strength nutrient agar (Qxoid) plus actidione (50 mg ml-i); and for thermophilic bacteria and actinomycetes, 1/2-strength tryptone-soya agar (Qxoid) plus 0.4% casein hydrolysate (Qxoid) and actidione (50 mg m l- 1) . Plates were incubated at either 55 (thermophilic actinomycetes only), 40 or 25°C.