Can asthma be studied in the urine?

There is a great need to expand the research armament for evaluation of airway inflammation in asthmatics. In particular, it would be desirable to have reliable and sensitive markers of inflammation which could be collected frequently but with simple procedures. Such measurements would be useful in studies of children and other groups of subjects who for different reasons are unsuitable for invasive means of sample collection such as bronchoscopy. Likewise, simple non-invasive strategies are desirable in epidemiology and other situations when large populations are to be investigated. Nasal lavage [1] or sputum induction [2] represent important efforts to develop more accessible means for obtaining samples for measurements of markers of inflammation. Analysis of urine has recently received attention as yet another new approach to monitor the activation of inflammatory cells and their mediators. The idea of measuring mediators in urine has emerged as a natural part of efforts over the last decade aimed at defining the role of leukotrienes (LTs) in asthma. For LTs and other arachidonic acid derived mediators (eicosanoids), the established strategy for assessment of many endogenous compounds [3, 4] is to measure a suitable indicator metabolite in the urine. This is because there are many and sometimes insurmountable problems which occur with eicosanoids during collection of blood, plasma or other body fluids [5]. Thus, the collection procedure itself is often a stimulus for de novo enzymatic biosynthesis of eicosanoids due to ex vivo activation of recovered cells and enzymes. In addition to artifactual biosynthesis of eicosanoids during sampling, non-enzymatic transformations of eicosanoids may spontaneously occur in fluid samples. There are, accordingly, several mechanisms which may lead to false high or low levels of the compound targeted for measurement, as well as the detection of compounds which have been formed ex vivo rather than endogenously. For example, large amounts of thromboxane B2(TXB2) may be formed in blood during sampling as a result of activated biosynthesis of thomboxane A2 (TXA2) by the collection procedure [6]. Urinary metabolites are therefore preferred for estimations of whole body production of the platelet-derived trombogenic product TXA2 [5]. The urinary excretion of 11-dehydro-TXB2 is a reliable measure because its biosynthesis cannot occur ex vivo [4, 7], whereas urinary TXB2 mainly reflects renal generation of TXA2 [5]. For such reasons, measurements of LTs and other eicosanoids in spot samples of, for example, bronchial lavage fluid may often be inappropriate for assessment of function. Measurement of an urinary metabolite which reliably reflects the endogenous compound to be studied is therefore most often the preferred strategy. Biosynthesis of LTs occur in mast cells, eosinophils and other inflammatory cells of myeloid origin by initial 5-lipoxygenation of arachidonic acid [8]. Leukotriene A4 is the first product and it may be transformed into LTB4 or LTC4. Studies of asthmatics using drugs which interfere with the formation or actions of LTs suggest that the pathway leading to formation of LTC4 is primarily involved in asthmatic airway inflammation [9]. The biosynthesis of LTC4 involves conjugation of LTA4 with glutathione at C6 in a reaction catalysed by LTC4 synthase [10]. Subsequent sequential cleavage of amino acids from the peptide side chain of LTC4 leads to the formation of LTD4 and LTE4 by the action of enzymes which are present in most tissues including the lung [11]. Together, LTC4, LTD4 and LTE4 are named the cysteinyl-LTs and they make up the biological activity previously implicated in asthma as slow reacting substance of anaphylaxis (SRS-A) [8]. Indeed, a strong body of evidence now supports the view that cysteinyl-LTs are significant mediators in asthma [9]. Drugs which inhibit the LTs are currently being introduced as a new treatment for asthma [9]. Interestingly, one of the first indications that LTs might be important in asthmatics in vivo in fact came from measurements of urinary excretion of LTs by Taylor et al. [12]. They found increased urinary LTE4 excretion following allergen bronchoprovocation and acute attacks of asthma. Subsequent work has confirmed the usefulness of measuring urinary excretion of LTE4 in studies of asthmatic responses [13-17]. For example, the first indications that LTs had a pivotal role in aspirin-intolerant asthma were obtained in studies where basal urinary LTE4 excretion was found to be elevated in aspirin-intolerant asthmatics, followed by a further increase in response to challenge with aspirin [15, 16]. Measurements of urinary LTE4 has, in addition, provided indications of other pulmonary [18-20] and extrapulmonary diseases where cysteinyl-LTs may be involved [21-24]. The suitability of urinary LTE4 as a marker of endogenous formation of cysteinyl-LTs has been established in metabolic studies [25-28]. The initial metabolism of LTC4 into LTD4 and LTE4 is rapid in most cells, including lung tissue and airways [11]. Likewise, systemically injected LTC4 is largely metabolised to LTD4 and LTE4 during one passage through the pulmonary circulation [29]. Thereafter, however, LTE4 is predominantly eliminated through biliary and renal excretion [26-30]. Although the major route for elimination of LTE4 is through the liver, for serial measurements it is better to collect the portion (5–15%) which is excreted into the urine. Provided that subjects have a normal glomerular filtration rate and no cholestasis, changes in urinary excretion of LTE4 appear to reflect changes in whole body production of cysteinyl-LTs [30]. The relatively rapid turn-over of LTs may deserve attention. Thus, following inhalation of LTD4, urinary excretion of LTE4 is increased in the first 2 h following the challenge [15, 31], with the majority being excreted within the first hour. Likewise, in asthmatics displaying isolated early reactions to allergen-challenge, the increase in urinary LTE4 was confined to the first 2 h following challenge [15, 32]. It follows that the time-course of urinary LTE4 excretion may provide additional information about whether or not an observed change has bearing upon the investigated clinical or experimental state. Another factor which may determine if a response is associated with changes in urinary LTE4 seems to be the number of cells or the size of the tissue which is involved in the reaction. Thus, nasal provocation with allergen does not change urinary LTE4 [12], although LTs may be measured in lavage fluid following allergen challenge [33], and inhibitors of LTs attenuate allergen-induced nasal blockage [34]. More recently, measurements of urinary LTE4 have been combined with the detection of another eicosanoid marker in the urine, namely the PGD2 metabolite 9α,11β-PGF2 [35]. There is good evidence that this marker may introduce a new and simple means to assess mast cell activation in vivo. First, it is well established that mast cell activation in vitro is associated with release of PGD2 [36-38], and PGD2 has also been shown to increase in vivo after local segmental allergen challenge in asthmatics [39]. Second, in humans, mast cells appears to be the only cell of importance to consider as a source of PGD2 [35]. It should however be recognized that there are considerable species differences. For example, in the rat, biosynthesis of PGD2 occurs in many additional cells [40]. The approach to measure a PGD2 metabolite in urine originates from extensive studies of PGD2 metabolism in man [41, 42]. In the human lung, the NADPH-dependant enzyme 11-ketoreductase metabolises PGD2 into 9α,11β-PGF2 [43], which is rapidly excreted intact into the urine as a prominent major metabolite of PGD2. This makes 9α,11β-PGF2 a suitable target for assessment of PGD2 formation. Interestingly enough, 9α,11β-PGF2 retains considerable biological activity and is a fairly potent bronchoconstrictor [43, 44]. As a further indication of the relation between mast cell activation and PGD2, Roberts et al. discovered that patients with systemic mastocytosis excreted increased amounts of several PGD2 metabolites in urine [45]. More recent studies in healthy volunteers [35], found low but detectable levels of 9α,11β-PGF2 in the urine, and there was no diurnal variation. Allergen-induced bronchoconstriction was associated with increased urinary excretion of 9α,11β-PGF2, whereas histamine-induced bronchoconstriction was not [35]. These observations thus confirmed the hypothesis that release of PGD2 was a result of mast cell activation and not a consequence of the bronchoconstriction per se. It was also found that the urinary excretion of 9α,11β-PGF2 increased following aspirin provocations of aspirin-intolerant asthmatics [35], supporting that mast cells are activated during the intolerance reaction. Previous attempts to establish mast cell activation during aspirin-induced bronchoconstriction by measurements of histamine or tryptase have produced ambiguous results [46, 47]. In this issue of Clinical and Experimental Allergy, Nagakura et al. have used the strategy of measuring urinary 9α,11β-PGF2 to assess whether or not mast cells are activated during exercise-induced bronchoconstriction in children [48]. The occurrence of post-exercise bronchoconstriction relates to the degree of water loss and heat change in the airways during exercise [49]. This suggests that changes in the local temperature and osmolarity of the airway mucosa triggers the reaction, and hyperosmolar solutions may indeed cause the release of mediators from isolated mast cell preparations [50] or during segmental bronchial challenge in vivo [51]. However, previous attempts to prove mast cell involvement in exercise-induced asthma by measurements of mediator release have been negative [49]. The study of Nagakura et al. [48] now provides support for release of mast cell mediators in exercise-induced bronchoconstriction. Fourteen children with exercise-induced bronchoconstriction showed an increase in urinary 9α,11β-PGF2 following exercise, whereas a group of children who underwent a negative exercise-test showed no significant change in urinary 9α,11β-PGF2. The findings of Nagakura et al. are obviously consistent with the pharmacology of exercise-induced bronchoconstriction. Thus, drugs which act on mast cells or their mediators, i.e. cromoglycate [52], antihistamines [52, 53] and LT-antagonists [54-56], attenuate exercise-induced bronchoconstriction. The investigation by Nagakura et al. [48] also illustrates the importance of selecting an appropriate target for measurement when attempting to assess the involvement of a particular cell. Although histamine and tryptase are excellent markers of mast cell activation in vitro, they have not been found particularly useful to measure in plasma. In fact, increased serum tryptase is primarily seen in anaphylaxis and other severe systemic reactions [57]. Therefore, previous negative data in exercise-induced bronchoconstriction with insensitive markers bear little weight compared with the present findings of Nagakura et al. [48]. Nagakura et al. in fact provide further evidence of the usefulness of 9α,11β-PGF2 as a mast cell marker [48]. They follow urinary 9α,11β-PGF2 in a group of 30 children admitted to hospital because of acute asthma. When treatment is instilled and the subjects improve clinically, the urinary levels of 9α,11β-PGF2 progressively decrease from those obtained at the admission. Therefore, it may be concluded that mechanisms in asthma can indeed be studied in urine samples. It goes without saying that the ease with which urine may be obtained makes this approach attractive, for example, in time-course studies or when assessing the effects of different environmental exposures. However, it should be recognized that the strategy is in its infancy and currently only established in a few research laboratories around the world. Although immunoassays are available [58], their application requires experience with measurements of this class of compounds and careful validation. This often involves purification by chromatography and other methods [8]. Nagakura et al. have used gas chromatography/mass spectrometry, which is very precise but the method is expensive and labour-intensive. It must indeed have been very time-consuming to assay the data they report on. An immunoassay for measurement of urinary 9α,11β-PGF2 has however recently been described [35]. Furthermore, although there is a growing interest in measuring urinary LTE4 in studies of asthmatics, there is currently very limited information about what are the normal values and how different manipulations or other diseases affect urinary LTE4. The few studies which have addressed such questions have generally included a very small number of subjects. There is clearly a great need to collect a lot of basic information about urinary excretion of LTE4, as well as the excretion of 9α,11β-PGF2. As for many other proposed markers of inflammation, it will be important to establish whether changes or abnormalities in urinary excretion of a mediator has predictive value in an individual as well as between groups of subjects. Given the easy and non-invasive nature of the sample collection, however, it appears worthwhile to carefully evaluate the use of urinary excretion of LTE4, 9α,11β-PGF2, and other metabolites excreted into the urine, as new potential markers of inflammation in asthma research. If such studies are successful, measurements of urinary markers may become useful in clinical diagnosis and for monitoring the effects of treatment. The authors are supported by the Karolinska Institutet and the following Swedish foundations: Medical Research Council (project 14X-9071), Heart Lung Foundation, the Foundation for Health Care Sciences and Allergy Research (Vårdal), and the Association Against Asthma and Allergy.