Dendrimers are highly branched and regular macromolecules with well-defined structures that attract considerable interest due to their potential applications in many fields of science. The three-dimensional architecture of dendrimeric systems confers them various intrinsic features such as structural homogeneity, integrity, controlled composition and high-density multidentate homogeneous terminal groups, ready for conjugation. These characteristics, added to their stability and versatility, mean that dendrimers have been used for many applications, such as sensing, catalysis, molecular electronics and photonics [1
]. Moreover, dendrimer and dendron nanostructures represent ideal scaffolds for numerous bioapplications and hold great promise for the future of nanomedicine [2
]. One important application is related to the study of allergic drug reactions.
Allergic drug reactions are one of the most important health problems nowadays. Betalactam (BL) allergy is self-reported by approximately 10% of adverse drug reaction sufferers. A high proportion of these cases are mediated by immunoglobulin E (IgE), leading to a range of symptoms, from simple skin involvement to anaphylactic shock [4
]. However, a variety of factors make the study of these reactions difficult, such as a lack of knowledge of the actual drug derivatives involved, changes to the pattern of hapten recognition over time in selected populations, the possibility of cross-reactivity between related chemical structures and a general increase in adverse patient responses due to environmental and genetic factors [6
In addition, BL allergy has a complex diagnosis which is still not correctly addressed. Many BLs with different chemical structures exist and each patient has a unique IgE response. An individual can be allergic to one BL only, or cross-reactive, being allergic to BLs with the same or similar side chains, or to multiple, potentially structurally diverse BLs from different families [9
]. A complete diagnostic procedure includes a detailed clinical history, which can be unreliable, followed by in vivo tests, including skin tests, which can have low sensitivity, and drug provocation testing, which poses some patient-risk, especially for severe reactions [12
In vitro testing represents a more rational alternative, as it can potentially identify the drug responsible, allowing the physician to find a safe alternative and reducing the need to perform drug provocation testing. However, it is necessary to confirm the sensitivity, specificity and negative and positive predictive values for these in vitro tests in order to enable their implementation in clinical practice. The most common in vitro tests are based on the detection of specific IgE, either in serum (immunoassays) or bound to receptors on the surface of effector cells (basophil activation tests) [13
BLs are haptens that need to be bound covalently to a carrier protein to induce an immune response. This hapten-carrier conjugate is used in immunoassays to quantify drug specific IgE (sIgE) in serum. The immunoassay is the most widely used in vitro test for the diagnosis of drug allergy. However, it has certain disadvantages, such as low sensitivity and the limited availability of commercial kits for a small range of BLs. Among non-commercial ones, Radio Allergo Sorbent Test (RAST) is the most commonly employed in research laboratories.
The nature of the carrier molecule is important for the development of an in vitro test for the detection of drug sIgE antibodies, as the conjugate it forms can influence IgE recognition. Poly-l
-Lysine is the most widely used carrier molecule for the RAST, due to its accessibility, multivalency and ease of chemical functionalization for different haptens [14
]. However, its inherent polydispersity complicates its precise chemical characterization and affects the reproducibility of the formed conjugate. To avoid these handicaps and to produce dense and reproducible hapten-carrier conjugates we proposed the use of PAMAM dendrimers [16
] as carrier molecules. Their potential for emulating the carrier protein in hapten-carrier conjugates for IgE recognition in BL allergy has been confirmed using dendrimeric antigens (DeAns). DeAns were synthetized by incorporating benzylpenicilloyl (BPO, the antigenic determinant of benzylpenicillin) or amoxicilloyl (AXO; the equivalent of amoxicillin) groups in the dendrimer periphery. Bi-epitope DeAns have also been designed, including both BPO and AXO on the same macromolecule, which enabled the detection of sIgE from selective and cross-reactive patients [17
]. The coupling of DeAns to different solid supports (cellulose disks, zeolites and silica particles) have allowed the determination of sIgE to penicillins by RAST, and this has been shown to have potential for diagnosis [18
Although the nanotechnological advances achieved in immunoassay design are very promising [24
], RAST assays have certain limitations, such as dependencies on a radioactive isotope, specific facilities and trained personnel. Thus, other techniques avoiding radioactivity are preferred by many research groups. Consequently, the basophil activation test (BAT), another in vitro diagnostic technique, has received increasing attention for the diagnosis of drug allergy [12
Like other functional assays, BAT tries to mimic in vivo IgE-mediated cell activation and mediator release. This test is useful for evaluating IgE-mediated reactions for a variety of injectable drugs since there is no need to use drug-carrier conjugates. BAT is also valuable for the identification of the drug responsible for a reaction. However, its sensitivity depends on the drug involved, with values of around 55% reported for BLs [26
BAT is based on the determination of activation markers expressed on basophil surface after the interaction of the drug with sIgE [13
]. However, the lack of knowledge of activation mechanisms has hampered a wider clinical application. BLs are not capable of activating basophils by themselves. They require conjugation to a carrier molecule, generally present in the blood, that is big enough to allow cross-linking of two sIgE bound to the basophil surface. However, no information about the size and compositions of these conjugates is available. The use of well-defined hapten-carrier conjugates would be a valuable tool for the investigation of the mechanism through which the activation occurs.
In this paper, we analyze BAT results in a group of patients with immediate allergic reactions to BLs using various DeAns as immunogens. We compare the results to those obtained using the free drug or hapten. We further analyze these results taking into account the structural features of the DeAns, using diffusion Nuclear Magnetic Resonance (NMR) and molecular dynamics simulation (MDS).
The diagnosis of penicillin allergy is complex and there is a large and unmet need from health-care professionals for better in vitro methods. The sensitivity of the tests can be influenced by the structure of the immunogen and is related to the underlying mechanisms involved in the allergic process. Activation of effector cells, mast cells and basophils, requires antigens of a certain size [32
] and may be negatively affected by the separation between the antigenic determinants [32
]. In the case of drugs, they are thought to act as haptens, as they are considered too small to induce allergy by themselves. To reach the adequate size to induce reactions, penicillins must bind proteins covalently, forming conjugates [34
]. The simultaneous recognition of a penicillin-protein conjugate by at least two sIgE molecules bound to adjacent FcεRI at the cellular surface is known as cross-linking. This induces degranulation of the effector cells, leading to the release of inflammatory mediators responsible for the reaction [35
In the BAT scenario, the free drugs (AX or BP) are assumed to bind proteins present in blood covalently through β-lactam reactivity, forming a big enough conjugate to achieve cross-linking. This approach attempts to emulate in vivo conditions, however it lacks information about the chemical composition of the conjugate inducing the activation. By using DeAns, one has more control over conjugate size, multivalence and the structure of peripheral antigenic determinants, allowing more reproducible assays. The use of the appropriate DeAn structure ensure the optimal interaction between the drug moieties and sIgE on the basophil surface, inducing more potent basophil degranulation and improving BAT sensitivity.
These relationships between cell activation and immunogen structure have been recently explored in in vitro studies performed in animal models using 2,4-dinitrophenyl (DNP) as a model hapten. It was found that the number of epitopes and the distance between them in synthetic nanostructures have different effects on mast cell degranulation [32
]. One of these studies, using dendrimers decorated with DNP epitopes, showed that larger DNP16
-dendrimers (64 Å) trigger mast cell degranulation by cross-linking IgE-receptor complexes, whereas smaller DNP-dendrimers are inhibitory [39
]. Other studies have analyzed these relationships in BLs using monovalent haptens, which could be recognized by IgE but unable to bind two adjacent antibodies simultaneously. These structures were shown to inhibit the development of an allergic reaction [31
], for both in vitro and in vivo tests in BP allergic patients, however this finding has not been further explored in a clinical setting. Most of our current knowledge regarding the activation of nanostructures in effector cell degranulation is based on studies performed with model ligands and animal mast cells. To the best of our knowledge, no studies employing this strategy have been used with real haptens or in human samples.
These studies have provided much needed information about the requirements necessary to activate effector cells, which cannot be deduced using the free drug alone. In fact, little is known about the nature of the adduct that activates basophils in the assay. In this context, the size of the actual carrier protein, the number of reactive sites, and the proximity between them are considered key factors that influence the cross-linking process, and their evaluation would be very complex. The use of well-defined nanostructures with consistent sizes and epitope density will provide a valuable tool to study the structural parameters required for these cell processes. We evaluated the ability of these nanostructures to stimulate basophils and we found that DeAns were able to induce activation in a selective and specific way. Interestingly, basophils from allergic patients follow similar patterns regarding generation: DeAn-G4 produces higher SI than DeAn-G2; this could be due to the size, valence or proximity between epitopes in the immunogen: compared to DeAn-G2, DeAn-G4 have a higher size (~20 Å vs. ~14 Å), an increased density of epitopes (64 vs. 16) and a higher proximity between them, favoring the IgE cross-linking on cell surface, and enhancing activation. This preliminary data suggests that using a size of approximately 20 Å as well as a higher density of epitopes may better resemble the IgE molecular recognition that occurs with penicillin-protein conjugates formed in vivo.