Impaction samplers are the most common outdoor allergen samplers. Rotating arm impaction samplers have two vertical, adhesive-coated collecting arms mounted on a crossbar, which is rotated by a vertical motor shaft. Small particles, particularly pollen grains, are prone to blowing in the wind in a way that interferes with gravitational settling. They become more likely to impact on an adhesive surface. The sampler rotates up to several thousand revolutions per minute to overcome the effects of wind. However, at this speed turbulence may “push” the pollen away and decrease sampling. For this reason the sampling surface is small (1 to a few millimeters) to get the highest rate of impaction. Small surface areas, however, are rapidly overloaded, causing a decrease in the efficiency of capture. These samplers usually are run intermittently (20 to 60 seconds every 10 minutes) to reduce overloading. In some models, the impacting arms are retracted or otherwise protected while not in use. The Rotorod sampler (Fig. 6.1) is a commercially available impaction sampler and has been shown to be over 90% efficient at capturing pollen particles of approximately 20 μm diameter. It is much less efficient at capturing smaller particles, especially those <5 μm diameter.

Suction samplers employ a vacuum pump to draw the air sample into the device. Although suitable for pollens, they are more commonly used to measure smaller particles such as mold spores. Disorientation with wind direction and velocity skews the sampling efficiencies of particles of different sizes. For example, if the wind
velocity is less than that generated by the sampler, smaller particles are collected in greater concentrations than exist in the ambient air. The reverse is true for greater wind velocities. The Hirst spore trap is an inertial suction sampler with a clock mechanism that moves a coated slide at a set rate along an intake orifice. This enables discrimination of diurnal variations. A wind vane orients the device to the direction of the wind. The Burkard spore trap collects particles on an adhesive-coated drum that takes 1 week to make a full revolution around an intake orifice. Both of these spore traps are designed to measure nonviable material. Spore traps are the most flexible devices for sampling particles over a wide range of sizes.

The Anderson sampler is another suction device, but it is unique in its adaptability for enumerating viable fungal spores. Air passes through a series of sieve-like plates (either two or six), each containing 400 holes. Although the air moves from plate to plate, the diameter of the holes decreases. The larger particles are retained by the upper plates and the smaller ones by successive lower plates. A Petri dish containing growth medium is placed beneath each sieve plate, and the spores that pass through the holes fall onto the agar and form colonies. This method has value for identifying fungi whose spore morphologic features do not permit microscopic identification. In general, however, nonviable volumetric collection techniques more accurately reflect the actual spore prevalence than do volumetric culture methods. The volume of air sampled is easy to calculate for suction devices because the vacuum pumps may be calibrated. In the case of rotation impaction samplers, there are formulas that depend on the surface area of the exposed bar of slide, the rate of revolution, and the exposure time. After the adherent particles are stained and counted, their numbers can be expressed as particles per cubic meter of air. Gravitational samplers cannot be quantified volumetrically.

The location of samplers is important. Ground level is usually unsatisfactory because of liability, tampering, and similar considerations. Rooftops are used most frequently. The apparatus should be placed at least 6 m (20 ft) away from obstructions and 90 cm (3 ft) higher than the parapet on the roof.

Fungi also may be studied by culture techniques. This is often necessary because many spores are not morphologically distinct enough for microscopic identification. In such cases, characteristics of the fungal colonies are required. Most commonly, Petri dishes with appropriate nutrient agar are exposed to the air at a sampling station for 5 to 30 minutes. The plates are incubated at room temperature for about 5 days, and then inspected grossly and microscopically for the numbers and types of colonies present. Cotton-blue is a satisfactory stain for fungal morphologic identification. Potato-dextrose agar supports growth of most allergenic fungi, and rose bengal may be added to retard bacterial growth and limit the spread of fungal colonies. Specialized media such as Czapek agar may be used to look for particular organisms (e.g., Aspergillus or Penicillium).

The chief disadvantage of the culture plate method is a gross underestimation of the spore count. This may be offset by using a suction device such as the Anderson or Burkard sampler. A microconidium containing many spores still grows only one colony. There may be mutual inhibition or massive overgrowth of a single colony such as Rhizopus nigricans. Other disadvantages are short sampling times, as well as the fact that some fungi (rusts and smuts) do not grow on ordinary nutrient media. Furthermore, avoiding massive spore contamination of the laboratory is difficult without precautions such as an isolation chamber and ventilation hood.

Numerous immunologic methods of identifying and quantifying airborne allergens have been developed. In general, these methods require more sophisticated instruments and thus are unlikely to replace the physical pollen count. The immunologic assays do not depend on the morphologic features of the material sampled, but on the ability of eluates of this material collected on filters to interact in immunoassays with human IgE or IgG or with mouse monoclonal antibodies (4,5). Studies at the Mayo Clinic have used a high-volume air sampler that retains 95% of particles larger than 0.3 μm on a fiberglass filter. The antigens, of unknown composition, are eluted from the filter sheet by descending chromatography. The eluate is dialyzed, lyophilized, and reconstituted as needed. This material is analyzed by RAST inhibition for specific allergenic activity or, in the case of antigens that may be involved in hypersensitivity pneumonitis, by interaction with IgG antibodies. The method is extremely sensitive. An eluate equivalent to 0.1 mg of pollen produced 40% to 50% inhibition in the short ragweed RAST. An equivalent amount of 24 μg of short ragweed pollen produced over 40% inhibition in the Amb a 1 RAST (6). The allergens identified using this method have correlated with morphologic studies of pollen and fungal spores using traditional methods and with patient symptom scores. The eluates also have produced positive results on prick skin tests in sensitive human subjects (7). These techniques demonstrate that with short ragweed, different-sized particles from ragweed plant debris can act as a source of allergen in the air before and after the ragweed pollen season. Unexpectedly, appreciable ragweed allergenic activity has been associated with particles less than 1 μm in diameter (8).

Use of low-volume air samples that do not disturb the air and development of a sensitive two-site
monoclonal antibody immunoassay for the major cat allergen (Fel d 1) have made accurate measurements of airborne cat allergen possible (5). These studies confirm that a high proportion of Fel d 1 is carried on particles smaller than 2.5 μm. During house cleaning, the amount of the small allergen-containing particles in the air approached that produced by a nebulizer for bronchial provocation (40 ng/m³). The results indicate that significant airborne Fel d 1 is associated with small particles that remain airborne for long periods. This is in contrast to prior studies with house dust mites (9) in which the major house dust mite allergen Der p 1 was collected on large particles with diameters greater than 10 μm. Little of this allergen remained airborne when the room was undisturbed.

Many pollen grains may be difficult to distinguish morphologically by normal light microscopic study. Immunochemical methods may permit such distinctions. Grass pollen grains collected from a Burkard trap were blotted onto nitrocellulose; then, by using specific antisera to Bermuda grass, a second antibody with a fluorescent label, and a fluorescent microscope study, Bermuda grass pollen grains could be distinguished from grass pollens of other species (10). These newer methods show promise because they measure allergenic materials that react with human IgE. Currently, immunochemical assays to quantify the major house dust mite allergens Der p 1 and Der f 1 and the major cat allergen Fel d 1 in settled dust samples are commercially available. Further studies with these techniques may lead to a better understanding of exposure–symptom relationships.

Some protein structures do seem to be more likely to be associated with allergenicity. One factor that may be important is the simultaneous exposure of multiple allergenic epitopes on a single structure to promote cross-linking of IgE. The major birch allergen, Bet v 1, has been compared to a naturally occurring, nonallergenic protein with significant sequence homology using three-dimensional computer modeling. The nonallergenic protein appears to have fewer epitopes on the exposed surface of the molecule and is more likely to be a monomer than Bet v 1 (13). Computer models that predict allergenicity suggest that the presence of multiple allergenic motifs on a protein makes it more likely to be allergenic (14). There is also evidence from computer modeling of known allergen sequences that a number of diverse allergens have a common structural motif, a groove inside an alpha-beta motif, which is also found in some toxins and defensins (15).

Do specific carbohydrate determinants promote allergenicity? Serum from allergic patients often has IgE that interacts with cross-reacting carbohydrate determinants. Many of these cross-reacting carbohydrate determinants are present in a wide variety of proteins and across very different species. There may be some in vivo effects of these epitopes on an immunologic level. However, cross-reacting carbohydrate epitopes often appear to be a source of positive serologic results without clinical significance (16). Carbohydrates have recently been identified in IgE-mediated reactions, and further investigations are underway.

It has been hypothesized that proteins might be more allergenic because of their structural similarity to invasive organisms. Helminths are classically associated with high IgE levels and intuitively might be associated with allergy, but studies in animal models and human populations suggest that helminthic infection is generally protective against the development of allergies (17). An exception, the fish parasite Anisakis, is associated with allergenic symptoms during infection and has been reported as an occupational respiratory allergen in fish-processing factories (18).

It has been known for some time that the major house dust mite allergen, Der p 1 , has structural similarity to cysteine protease enzymes (19). The enzymatic activity of this allergen may have a role in the development of atopic sensitization and asthma. A series of experiments have been done in which mice are sensitized using enzymatically active or inactive Der p 1. In the presence of the active enzyme, the mice create more total IgE and Der p 1-specific IgE (20). Active Der p 1 also has been used as an adjuvant to increase production of ovalbumin-specific IgE (21). A third experiment involved nasal sensitization with either active or inactive Der p 1 . The active enzyme group had increased cellular inflammatory cells in the lungs and increased total IgE levels (22). Thus, the enzymatic activity appears to contribute to the allergenicity of the molecule.

Several mechanisms have been proposed to explain the relation between enzymatic function and the development of sensitization. Enzymatically active Der p 1 can disrupt tight junction between respiratory epithelia in cell cultures (23). This has been proposed as a mechanism by which the allergen can be delivered through the epithelium to the immune cells. Other dust mite allergens, Der p 3, 6, and 9, have been categorized as serine proteases. These allergens can also disrupt the tight junctions in respiratory epithelium (24).

Several immunologic mechanisms have also been proposed to explain the apparent allergenic effects of dust mite enzymes. The enzymatic activity of Der p 1 has been shown to cleave the low-affinity IgE FcR CD23 from human B cells, which may augment IgE synthesis. (25,26). It has also been shown that dendritic cells incubated with enzymatically active Der p 1 generate less IL-12, which would favor T_H2 cell responses (27).

The role of allergenic enzymes in activating protease activated receptors (PARs) has also been investigated. It appears that serine proteases, but not cysteine proteases are involved in the activation of protease activated receptors. The major mold allergen from penicillium, Pen c 13, is also a serine protease. Incubation of a respiratory epithelial cell line with Pen c 13 induces IL-8 release through a pathway that is dependent on activation of both PAR-1 and PAR-2 (28,29). Several other major mold allergens have been shown to increase IL-8 and IL-6 secretion in epithelial cell lines. Crude cockroach extracts also activate PAR-2 and stimulate IL-8 production in respiratory epithelial cell lines (30,31). Cysteine proteases, such as Der p 1, can activate respiratory epithelial cells to secrete IL-8, but the mechanism does not involve PAR-2 (32).

It is important to remember that pollen grains are complex structures designed to deliver plant reproductive material. They have many chemicals and proteins that are presented to the respiratory mucosa at the same time as the most allergenic proteins. The biochemical properties of these pollen grains also contribute to their allergenicity. It has been shown that birch, grass, and ragweed pollens contain both serine proteases and cysteine proteases and that these proteases can also disrupt epithelial tight junctions (33). Type 1 grass pollen allergens appear to migrate into the stratum corneum skin via the hair follicles as soon as 15 minutes. This has been proposed as a mechanism of sensitization in atopic dermatitis (34).

Pollen extracts also appear to have direct effects on the immune system. In cell culture, birch pollen extracts have been found to direct dendritic cells toward a more T_H2 type of antigen presentation and to recruit more T_H2 cells for antigen presentation (35).

Aeroallergen particle size is an important element of allergic disease. Airborne pollens are in the range of 20 μm to 60 μm in diameter; mold spores usually vary between 3 μm and 30 μm in diameter or longest dimension; house dust mite particles are 1 μm to 10 μm. Protective mechanisms in the nasal mucosa and upper tracheobronchial passages remove most of the larger particles, so only those 3 μm or smaller are thought to reach the alveoli of the lungs. Hence, the conjunctivae and upper respiratory passages receive the largest dose of airborne allergens. Despite this conventional wisdom, examination of tracheobronchial aspirates and surgical lung specimens has revealed whole pollen grains in the lower respiratory tract (36). These are considerations in the pathogenesis of allergic rhinitis and bronchial asthma as well as the effects of chemical and particulate atmospheric pollutants.

The development of asthma after pollen exposure is enigmatic because pollen grains are thought to be deposited in the upper airways as a result of their large particle size. Experimental evidence suggests that rhinitis, but not asthma, is caused by inhalation of whole pollen in amounts encountered naturally (37). Asthma caused by bronchoprovocation with solutions of pollen extracts
is easily achieved in the laboratory, however. Pollen asthma may be caused by the inhalation of pollen debris that is small enough to access the bronchial tree.

Ragweed asthma supports this hypothesis. The major ragweed allergen, Amb a 1, has been found in ambient air, even in the absence of whole pollen (7). Extracts of materials collected on an 8-μm filter that excludes ragweed pollen grains still appear to contain ragweed allergen based on skin testing and ragweed-IgG inhibition (38).

In Melbourne and London, severe outbreaks of asthma have been reported during some thunderstorms. This phenomenon has been referred to as thunderstorm asthma. People who had asthma exacerbations during a thunderstorm were more likely to be sensitive to grass pollen (39). Grass pollen is generally considered to be too large to access the smaller airways of the lungs. However, exposure of grass pollen grains to water creates rupture into smaller, respirable-size starch granules with intact allergens (40). These starch granules have been found to increase 50-fold during a rainstorm and thunderstorm asthma patients are more likely to be sensitive to the starch granules than other asthma patients (39,41). There is evidence for a similar effect of Alternaria spores. Thunderstorm asthma patients were more likely to be sensitive to Alternaria , and counts of broken Alternaria spores correlate with hospital admissions during a thunderstorm (42).

Trees may be gymnosperms or angiosperms. The gymnosperms include two orders, the Coniferales (conifers) and the Ginkgoales. Neither is of particular importance in allergy, but because of the prevalence of conifers and the incidence of their pollens in surveys, some comments are in order.

Conifers grow mainly in temperate climates. They have needle-shaped leaves. The following three families are germane to this discussion.

Most allergenic trees are in this group. The more important orders and families are listed here with relevant notations. Other trees have been implicated in pollen allergy, but most of the pollinosis in the United States can be attributed to those mentioned here.

Grasses are monocotyledons of the family Poaceae (or Gramineae). The flowers usually are perfect (Figs. 6.7 and 6.8). Pollen grains of most allergenic grasses are
20 μm to 25 μm in diameter, with one germinal pore or furrow and a thick intine (Fig. 6.9). Some grasses are self-pollinated and therefore noncontributory to allergies. The others are wind-pollinated, but of the more than 1,000 species in North America, only a few are significant in producing allergic symptoms. Those few, however, are important in terms of the numbers of patients affected and the high degree of morbidity produced. Most of the allergenic grasses are cultivated and therefore are prevalent where people live.

The grass family contains several subfamilies and tribes of varying importance to allergists. The most important are listed here.

A weed is a plant that grows where people do not intend it to grow. Thus, a rose could be considered a weed if it is growing in a wheat field. What are commonly called weeds are small annual plants that grow without cultivation and have no agricultural or ornamental value. All are angiosperms and most are dicotyledons. Those of interest to allergists are wind pollinated, and thus tend to have relatively inconspicuous flowers.