Study of protein haptenation by amoxicillin through the use of a biotinylated antibiotic

Abstract

Allergic reactions towards β-lactam antibiotics pose an important clinical problem. The ability of small molecules, such as a β-lactams, to bind covalently to proteins, in a process known as haptenation, is considered necessary for induction of a specific immunological response. Identification of the proteins modified by β-lactams and elucidation of the relevance of this process in allergic reactions requires sensitive tools. Here we describe the preparation and characterization of a biotinylated amoxicillin analog (AX-B) as a tool for the study of protein haptenation by amoxicillin (AX). AX-B, obtained by the inclusion of a biotin moiety at the lateral chain of AX, showed a chemical reactivity identical to AX. Covalent modification of proteins by AX-B was reduced by excess AX and vice versa, suggesting competition for binding to the same targets. From an immunological point of view, AX and AX-B behaved similarly in RAST inhibition studies with sera of patients with non-selective allergy towards β-lactams, whereas, as expected, competition by AX-B was poorer with sera of AX-selective patients, which recognize AX lateral chain. Use of AX-B followed by biotin detection allowed the observation of human serum albumin (HSA) modification by concentrations 100-fold lower that when using AX followed by immunological detection. Incubation of human serum with AX-B led to the haptenation of all of the previously identified major AX targets. In addition, some new targets could be detected. Interestingly, AX-B allowed the detection of intracellular protein adducts, which showed a cell type-specific pattern. This opens the possibility of following the formation and fate of AX-B adducts in cells. Thus, AX-B may constitute a valuable tool for the identification of AX targets with high sensitivity as well as for the elucidation of the mechanisms involved in allergy towards β-lactams.

Figure 1. Synthetic procedure for AX-B.

Figure 2. ¹H-NMR spectra of (A) AX-B and (B) AX recorded in D₂O.

Figure 3. Reactivity of AX and AX-B.

¹H NMR of each amoxicillin: (A) AX-B and (B) AX, in presence of butylamine in D₂O after 15 min. Arrows point at signals corresponding to H5 and H6, used to monitor the reaction.

Figure 4. Modification of HSA by AX and AX-B.

(A) HSA was incubated in the presence of the indicated concentrations of Ax or Ax-B and adduct formation was assessed by western blot and detection with an anti-AX antibody (AO3.2) or with HRP-streptavidin. Exposure times were 5 minutes for AX and one second for AX-B detection, respectively. (B) HSA was incubated with AX or AX-B at 0.5 mg/ml for 16 h at 37°C. Aliquots of the incubation mixture containing the indicated amounts of total protein were analyzed by SDS-PAGE and adducts formed were detected as above. Exposure times were 2 minutes for AX and one second for AX-B detection, respectively.

Figure 5. Competition between AX and AX-B for modification of HSA.

(A) HSA was incubated overnight with a 9-fold molar excess of AX or AX-B in bicarbonate buffer and detection of adducts was achieved by western blot with an anti-AX antibody (AO3.2) or with HRP-streptavidin, as indicated. (B) HSA was incubated in the absence or presence of AX-B, as above, followed by an overnight incubation with AX. Binding of AX was assessed by western blot with AO3.2 antibody. (C) HSA was incubated for 2 h with 80 μM AX-B, after a 16 h pre-incubation with the indicated concentrations of AX, expressed in molar excess with respect to AX-B. Aliquots of the incubation containing 2 μg of protein were subjected to SDS-PAGE and transfer, and incorporation of AX-B was assessed by detection with HRP-streptavidin. Blots shown in every case are representative of three independent assays with similar results.

Figure 6. Recognition of AX and AX-B by the sera of allergic patients.

Shown are RAST inhibition assays using in the solid phase AXO-PLL discs and in the fluid phase AX and AX-biotin, at the indicated concentrations. Results are from selective allergic patients to AX (A, patients 1–4) and from cross-reactive allergic patients (B, patients 5–8).

Figure 7. Modification of serum proteins by AX-B.

Total human serum was incubated in the presence of the indicated concentrations of AX or AX-B, for 16 h, after which, 70 μg of protein were analyzed by 2D electrophoresis, as described in the experimental section. (A) and (B) AX and AX-B-modified proteins were visualized by detection with AO3.2 antibody or HRP-streptavidin, as indicated. Exposure times were 3 min and 5 seconds, respectively. (C) Coomassie staining of one gel showing the positions of the identified proteins. 1, HSA; 2, transferrin; 3, 4 heavy chains of Igs; 5, 6, light chains of Igs; 7, 8, haptoglobin 2; 9, apolipoprotein A1. (D) Total human serum was incubated with 0.1 mg/ml AX-B (0.16 mM) for 2 h after a 16 h pre-incubation with the indicated concentrations of AX, expressed in molar excess with respect to AX-B. Aliquots of the incubations containing 0.5 μg of protein were resolved on a 10% polyacrylamide gel, before transfer and detection with HRP-streptavidin. The positions of transferrin, HSA and heavy and light chains of Igs, from top to bottom, are marked by asterisks.

Figure 8. Detection of intracellular AX-B-protein adducts.

(A) Raw264.7 murine macrophages were incubated in the absence or presence of AX for 24 h, after which, formation of AX-protein adducts was assessed by SDS-PAGE of cell lysates and detection with AO3.2 antibody. The position of two of the bands detected above background is marked by arrowheads. (B) Cells were incubated with the indicated concentrations of AX-B and protein adducts present in cell extracts were detected by SDS-PAGE, transfer and detection with HRP-streptavidin. The position of endogenous biotinylated proteins is marked by arrows and that of the major labeled proteins is indicated by asterisks. (C) Cells were incubated with biotin and the presence of biotinylated proteins was assessed as in (B). (D) Cells were incubated with the indicated concentrations of AX for 24 h, after which, AX-B was added at 0.3 mg/ml and incubation was continued for an additional 24 h. Incorporation of AX-B into proteins was assessed as above.

Figure 9. Localization and cell-type specificity of AX-B adducts.

(A) RAW264.7 macrophages were cultured in the absence or presence of 0.5 mg/ml AX-B in serum-free medium for 24 h, after which, cells were fixed with paraformaldehyde and biotin positive structures were visualized by staining with Alexa-488-streptavidin. Nuclei were stained with DAPI. Shown is a single z section at mid cell height. DIC, differential interference contrast. Bar, 20 μM. (B) RAW264.7 cells (left panel), monocytic THP-1 (middle panel) and B lymphoma cells (right panel) were incubated with AX-B and the presence of AX-B-protein adducts was analyzed by SDS-PAGE and biotin detection with HRP-streptavidin and ECL. Typically, aliquots from total protein extracts containing 15 μg of protein were analyzed. Endogenous biotinylated polypeptides are marked by arrows and the position of the polypeptides showing higher biotin incorporation after AX-B treatment is indicated by asterisks.