What are the important allergens in grass pollen that are linked to human allergic disease?

There is no doubt that grass pollen is a significant cause of immediate hypersensitivity among susceptible individuals in temperate climates, affecting more than 25% of the population [1]. Over the last three and a half decades, a considerable amount of important research has been conducted worldwide to gain a better insight into the complexity of the allergenic content of grass pollens and their clinical and immunological significance. Moreover, as a result of advances in molecular biology techniques, during the last two decades, significant progress has been made in relation to the cloning and production of recombinant allergens. Such studies, both at the protein and DNA level, are mandatory not only for establishing clinical and immunological significance of a certain grass pollen, and of a particular allergen within that pollen, but more importantly, for producing standardized extracts for use in diagnosis and treatment of disease. Cloning and elucidation of the primary structure (i.e. nucleotide and deduced amino acid sequences) of clinically significant allergens will allow us to gain a better understanding of the molecular basis of allergenicity and perhaps answer the questions ‘What makes an allergen, an allergen?’ and ‘What are the important allergens in grass pollen that are linked to human allergic disease?’ An exciting spin-off of such molecular studies is the prospect of the design and production of a new generation of therapeutics that will revolutionize the treatment of grass pollen allergies with enhanced specificity, efficacy and safety. Allergens are essentially IgE-binding proteins and this interaction with human IgE is the primary cause of allergic response. Clinical tests, such as radioallergosorbent test (RAST) and skin tests, use total grass pollen protein extracts to establish allergy. Therefore, when assessing what is a clinically important allergen from any given source, it is critical that an allergen's IgE-binding capacity, strength and frequency is evaluated by several standard immunoassay techniques. Unfortunately, biological functions of most grass pollen allergens still remains largely unknown and therefore no particular biological function (e.g. enzymic activity) can be attributed to their IgE-reactivity. As a result, this editorial will focus on the clinically significant grass pollens and the relative IgE-binding capacities of their allergens that have been identified and characterized up to date to address the question ‘What are the important allergens in grass pollen that are linked to human allergic disease?’ The answer to this question will no doubt be the underlying principle for use of recombinant allergens for diagnosis and treatment of human allergic disease. Grass pollen allergens were first reported in 1965 by Johnson and Marsh for rye grass pollen [2]. Pollen allergens from a number of other grasses have been reported since. These include barley [3], Bermuda grass [4], canary grass [5, 6], Johnson grass [7], Kentucky bluegrass [8], maize [9], timothy grass [10], orchard grass [11], rice [12] and Yorkshire fog or velvet grass [13]. Although all of these grasses are clinically significant, certain grasses may be more prevalent in certain geographical regions and consequently release more pollen into the atmosphere and are therefore a dominant source of allergens. For example, in Australia rye grass occurs widely, and can produce up to half a tonne of pollen per hectare contributing significantly to the airflora and therefore to grass pollen allergy [14]. On the other hand, the pattern of IgE-reactivity is different in Europe. A Dutch study has recently reported that the highest IgE responses, among 209 patients demonstrating grass pollen allergy, were observed against Poa pratensis (Kentucky bluegrass) followed by Festuca rubra (red fescue), Phleum pratense (timothy grass) and Dactlis glomerata (orchard grass) [15]. The lowest IgE responses (by a factor of 5) were against Phragmites communis (common reed), Cynodon dactylon (Bermuda grass) and Zea mays (maize). However, despite the difference in geographical dominance of certain grasses, there are extensive studies highlighting a high degree of allergenic similarity or cross-reactivity between the different grasses [16]. This means that a number of rye grass pollen allergens, for example, are homologous to those in other grasses. Occurrence of homologous allergens in different grasses has been observed by both structural studies, using proteolysis of purified allergens, as well as at the IgE-binding epitope level using immunoblotting analysis [16]. However, although it is now generally accepted that cross-reactive allergens occur in different grass pollen extracts, it should be emphasized here that a number of unique allergens are also present which may play a critical role in allergy and must be considered with as much caution and consideration as for cross-reactive allergens. It is hoped that the presence of cross-reactive allergens will simplify the diagnosis and treatment of grass pollen allergy using standardized extracts containing recombinant allergens. Gel electrophoresis has contributed significantly to the analysis of grass pollen extracts and identification of the number and the strength and frequency of IgE-binding activity of allergens. When grass pollen extracts are resolved on a polyacrylamide gel (PAGE), under reducing and denaturing conditions (i.e. SDS-PAGE), the complexity of the protein profile is visualized by staining of all of the polypeptides with protein specific dyes, like coomassie brilliant blue. There can be as few as three protein bands to more than 30 depending on the grass species. This complexity is further realized by two-dimensional polyacrylamide gel electrophoresis (2D-PAGE). In SDS-PAGE (1D-PAGE), proteins are separated according to their molecular mass only while in 2D-PAGE, proteins are first separated according to their charge followed by molecular mass separation. 2D-PAGE gives a more informative protein profile displaying the presence of isoforms. These are a group of related, but immunologically different, proteins with slight amino acid sequence differences contributing not only to a different mass but also to a different charge (i.e. isoelectric point). They therefore resolve as discrete protein spots. Provided that the relevant equipment is readily available, 2D gel analysis should be considered as the best tool to characterize complex protein profiles and identify which proteins are the allergens. After 1D or 2D-PAGE, grass pollen allergens are identified by transferring resolved proteins onto a solid nitrocellulose matrix and probing with grass pollen allergic patient serum (i.e. immunoblotting analysis). Detection of IgE-binding proteins is achieved with labelled antihuman IgE antibodies. Once developed, the intensity and frequency of IgE binding among individual patient sera is used to assign such proteins as strong and major allergens, respectively. Indeed, there have been many studies identifying major and minor allergens and these have been summarized in Table 1. As a general rule, major allergens are those that possess greater than 50% of IgE-binding frequency when tested with a panel of patient sera. However, this does not mean that any IgE-binding frequency lower than this should be dismissed as any IgE-binding allergen may provoke an allergic reaction in certain individuals. Therefore, strength of IgE-binding is also important to identify which allergens are clinically significant among such individuals. However, one common problem with the use of immunoassay techniques such as immunoblotting analysis of SDS-PAGE-resolved grass pollen extracts is the omission of the necessary negative controls. That is, duplicate blots need to be incubated in the antihuman IgE detection antibodies separately to assess the specificity of the observed IgE-reactivity. Such detection antibodies are raised in animals which may be exposed to the same proteins and hence also possess specific antibodies. Another important control is the assessment of IgE-reactivity of nonatopic human serum. A limitation of the SDS-PAGE itself is the fact that the proteins are resolved under reducing and denaturing conditions which may affect conformational epitopes that are formed as a result of di-sulphide bonding. Thus, IgE-binding epitopes observed with the SDS-PAGE resolved proteins may in fact not occur within the context of the natural protein. In these instances, we may be identifying proteins as allergens that do not exist naturally in the pollen grain. SDS-PAGE results can be clarified by using natural extracts as inhibitors in inhibition immunoassays and/or native gels. However, there are reports that claim renaturation of proteins with the removal of SDS during transfer of proteins to the nitrocellulose [17]. These limitations highlight the importance of molecular cloning of major grass pollen allergens since recombinant allergens can be used as potential inhibitors of IgE binding to the natural extracts in different immunoassays. Once a protein is identified as an allergen, it is then important to assess the degree or frequency of IgE binding among a population of atopic and nonatopic individual patient sera. Using such methods of analysis a number of allergens from different grass pollens have been identified, characterized, cloned and named according to the International Union of Immunological Societies (IUIS) system of nomenclature [18] (Table 1). Allergens with known partial or complete amino acid sequences are named according to the accepted taxanomic name of their source as follows: the first three letters of the genus; space; the first letter of the species name; space; and an Arabic numeral indicating the order of their identification [18]. So far, as a result of molecular cloning (i.e. sequence comparisons) and presence of cross-reactivity among different grass pollens, a number of allergen groups have been identified (Table 1). For example, group 1 allergens include Lol p 1 from rye grass pollen, Cyn d 1 from Bermuda grass pollen, Phl p 1 from timothy grass and so on. When assessed by immunoblotting analysis, different grass pollen allergens belonging to one group generally have a similar molecular mass. Once groups are assigned for one grass pollen, it does not necessarily mean that an allergen belonging to the same group will also be present in another grass pollen. For some grass pollens, certain allergen groups have not yet been filled, yet for others, new group numbers are created each time a novel allergen is cloned. Different grass pollens have different numbers of allergen groups. Since not all allergen groups have been characterized for all clinically significant grass pollens, it may be premature to answer the question ‘What are the important allergens in grass pollen that are linked to human allergic disease?’ with certainty. If one were asked to choose the most important allergens in grass pollen that are linked to human allergic disease, from our current knowledge and understanding, the obvious choice would be groups 1 and 5. There are several reasons for this choice. Firstly, these groups of allergens have been shown to possess the highest frequency and strength of IgE-reactivity (Table 1). Secondly, these allergen groups are highly conserved in most of the clinically significant grasses [12345678910111213141516]. Finally, these allergen groups exist as micronic particles in the atmosphere and therefore have the potential to trigger allergic asthma [1920]. If the majority of subjects under investigation are primarily allergic to allergen groups 1 and 5, then treatment with standardized recombinant Lol p 1 and Lol p 5, for example, should be adequate. It has been recently reported that while Lol p 1 and Lol p 5 accounted for an average of 81% of total antigrass pollen IgE, only 57% was observed with Phl p 1 and Phl p 5 [15]. Moreover, greater than 90% of the grass pollen positive sera reacted with purified natural Lol p 1 and/or Lol p 5 with 80% of the IgE response directed to these major allergens [15]. As a result, it may be possible to use only one grass pollen, and a selected few major allergens thereof (i.e. Lol p 1 and Lol p 5) for successful diagnosis and therapy of grass pollen allergy for the majority of the grass pollen sensitive population. It should be stressed here that certain individuals may be sensitive to specific allergens which may be classified as minor in a large population setting and in such cases, specific diagnosis and therapy may be needed. It should also be stressed here that, although cross-reactive allergens are present in different grass pollens, the use of recombinant allergens of one grass to diagnose/treat allergy to different grass pollen may not be effective since different cross-reactive allergens may also have unique/additional IgE-binding epitopes with exclusive IgE-reactivity among certain individuals. Indeed, several epitope mapping studies have located multiple human IgE-binding epitopes on some grass pollen allergens [21]. In order to determine which allergens are important in human allergic disease, due to the polyclonal nature of IgE-binding epitopes, it would be necessary to use purified and standardized natural or recombinant allergens in inhibition immunoassays to determine their capacity to inhibit IgE binding to the total pollen extracts of different grasses among a large population of grass pollen allergic individuals with defined allergen reactivity. Apart from looking at what allergens are clinically important, both among individual patients and in a large population, it is also important to assess the IgE-reactivity of the recombinant allergens in comparison to their natural equivalents; especially, since recombinant allergens are constantly suggested for use as standardized preparations in diagnosis and treatment of grass pollen allergy. Up to now, all cloned allergens reported in literature have some degree of IgE-reactivity (Table 1). However, not all studies have conducted such recombinant vs. natural allergen comparisons among a large population setting. This should be a prerequisite in our future studies as it generates valuable information of clinical relevance of recombinant allergens. Although it is important to have a recombinant allergen which possesses the same IgE-binding strength and frequency as its natural counterpart, its usefulness may primarily extend to diagnosis since presence of IgE reactivity can give rise to systemic reactions during immunotherapy. Availability of the allergen encoding cDNA allows one to mutate IgE-binding epitopes via site-directed mutagenesis while maintaining T-cell epitopes in an attempt to produce hypoallergenic variants for use in immunotherapy. Indeed, such studies have been conducted for a number of allergens where IgE-binding has been successfully abrogated and such mutants were able to retain T-cell stimulatory capacity [22232425]. By maintaining critical T-cell epitopes and allowing T-cell proliferation, following an immunotherapy regimen it is believed that the cytokine profile is changed from TH2 (allergic response) to TH1 (nonallergic), but the precise mechanism for this is not fully understood yet [26]. However, to allow the production of such hypoallergenic mutants, their B cell (IgE-binding) and T-cell epitopes need to be mapped first. It is hoped that such hypoallergenic mutants may become useful tools in the treatment of allergy in the near future. An unresolved issue, in relation to complex grass pollen protein profiles, is ‘obscure’ or ‘hidden’ allergens. As discussed previously, this problem relates to the low resolution ability of 1D-PAGE. It was recently reported that a novel high molecular mass allergen Phl p 13, from timothy grass pollen, was cloned and expressed [27]. In this issue of Clinical and Experimental Allergy, the same group report that the high molecular mass allergen fraction of timothy grass pollen between 50 and 60 kDa is comprised of both Phl p 4 and Phl p 13 [28]. This paper highlights the importance of molecular cloning, since cloning of Phl p 13 has given clues to pin-point its natural equivalent, and clearly illustrates the limitations of 1D-PAGE. In this regard, 2D-PAGE analysis is by far the best tool to obtain maximum protein resolution. If isoelectric points of the proteins are known, narrower pH gradients can be chosen to resolve the proteins of interest even further. 2D-PAGE would be an ideal method for resolving Phl p 4 and Phl p 13 as their isoelectric points are reported to differ by about two points, 7.5 and 9.3, respectively [28]. Moreover, it would be interesting to evaluate if these, and other allergens possess isoforms, which add to the complexity and clinical importance of the grass pollen allergen profile. The study by Suck et al. [28] also addresses the importance of cloning novel allergens and assessing their clinical significance. For example, Phl p 4 and Phl p 13 have been shown to be immunologically unrelated yet they each account for more than 50% of IgE-binding frequency among 306 of the grass pollen sensitive patients tested in Germany [28]. However, the study only deals with purified natural Phl p allergens and does not compare IgE-reactivities with the recombinant allergens. In regard to the IgE-binding immunoassays, Suck et al. [28] report on inconsistencies with the strength of IgE-binding to Phl p 13 between immunoblotting analysis and ELISA. As discussed in detail previously, the low IgE-reactivity of Phl p 13 when assessed by immunoblotting analysis may be attributed to the reducing and denaturation effect, while in ELISA the allergen may be more likely to assume its natural conformation and remain more IgE-reactive. The study also uses inhibition immunoassays to show that Phl p 4 and Phl p 13 are immunologically unrelated. Surprisingly, pre-incubation of sera with Phl p 13 inhibited IgE-binding to higher molecular mass bands between 90 and 110 kDa. Since Phl p 13 is a glycoprotein, the authors suggest that cross-reactivity between the carbohydrate determinants may be responsible for this observation. This observation may also be attributed to shared IgE-binding epitopes or possible dimerization. The authors also report that Phl p 13 inhibits IgE-binding to a ‘background smear’ produced by smaller bands and attribute this to breakdown products due to proteolysis. Generally, the stability of grass pollen extracts is a concern as presence of pollen proteases in extracts may contribute to proteolysis, but there have been no conclusive studies to suggest this. If proteolysis was a problem, one could extract pollen proteins in the presence of protease inhibitors. However, it should be remembered that protease inhibitors are proteins themselves and will therefore contaminate pollen extracts and add to the protein profile complexity. Indeed, Glu-C and Lys-C is reported to fragment Phl p 13 [28] and in order to validate this theory further, the authors could perhaps have conducted the same inhibition experiments with timothy grass pollen extracts including protease inhibitors. In conclusion, this study confirms the complexity of grass pollen extracts and the importance of molecular cloning in discovery of novel and clinically significant allergens. However, there is perhaps more to the complexity of the grass pollen protein profile than realized. For example, in our own experiments we routinely observe that purified Lol p 5, from rye grass pollen, dimerizes and runs on SDS-PAGE at around 60 kDa and strong reducing agents are required to avoid this dimerization [29]. Also, there may be protein–protein interactions that have not yet been identified. Therefore, when assessing what are the important allergens in grass pollen that are linked to human allergic disease, the answer is dependent on the complexity of the grass pollen extract and, more importantly, on the information gained from molecular cloning and characterization of the purified natural and recombinant allergens. Unfortunately, the answer cannot be simply group 1, 5 and so on, as all IgE-binding proteins are potential allergens. Such ongoing molecular and immunological studies will only contribute to our understanding of the molecular basis of the human IgE–allergen interaction and allow the selection of key allergens, or epitopes thereof, for simplified diagnosis and therapy.