Key points are not available for this paper at this time.
High mobility group protein 1 (HMGB1) has been implicated in diverse cellular functions, including determination of nucleosomal structure and stability and binding of transcription factors to their cognate DNA sequences (1)(2)(3)(4). HMGB1 is also present in a membrane-associated form, termed amphoterin, that mediates neurite outgrowth (5). Amphoterin can interact with macrophage cell surface receptors for advanced glycation end products to enhance expression of tissue-type plasminogen activator (6). Recently, HMGB1 was identified as a late mediator of endotoxin lethality (7). Mice had increased serum HMGB1 concentrations after exposure to endotoxin, and sepsis patients who succumbed to infection also had increased serum HMGB1. It would therefore be useful to develop an easy and highly sensitive method to measure serum HMGB1. However, this study revealed that HMGB1 and HMGB2, with extremely high homology (81%) to HMGB1, coexist in the serum. We report an ELISA method we have developed that measures only HMGB1 without simultaneous determination of HMGB2. To prepare an anti-peptide monoclonal antibody reacting only with HMGB1, we selected a peptide sequence (peptide 1; GKGDPKKPRGK) with high antigenicity and different from that of HMGB2. The monoclonal anti-calf HMGB1 antibody was prepared against calf thymus-derived HMGB1, which was 98% homologous to human HMGB1. Each protein sample used for the analysis was prepared as follows. The peptides were purified and separated by HPLC. The peptide synthesized was added to maleimidobenzoyl-N-hydroxysuccinimide ester (Pierce Chemical Co.)-labeled keyhole limpet hemocyanin (Calbiochem) or maleimidobenzoyl-N-hydroxysuccinimide ester-bovine serum albumin (BSA; Sigma). Calf HMGB1 and HMGB2 were obtained from calf thymus by the method of Sanders (8). Human HMGB1 and HMGB2 were purified from HL60 cells (9). As a result, the anti-peptide 1 monoclonal antibody (no. 77) reacted with human HMGB1 but not with human HMGB2. The anti-calf thymus HMGB1 monoclonal antibody (no. 03E5) reacted with both human HMGB1 and HMGB2 (Fig. 1A ). Immunoreactivity of anti-HMGB1 antibodies (A), Western blotting results for HMGB1 (B), and linear regression analysis of ELISA vs Western blot (C). (A), immunoreactivity of anti-calf thymus HMGB1 (no. 03E5) and anti-peptide 1 (no. 77) monoclonal antibodies determined by Western immunoblot. Lane 1, anti-peptide 1 monoclonal antibody (no. 77); lane 2, peroxidase-conjugated anti-mouse immunoglobulin polyclonal antibody; lane 3, anti-calf thymus HMGB1 monoclonal antibody (no. 03E5). Before immunoblotting, purified human HMGB1 (1 ng/lane) and HMGB2 (1 ng/lane) in all lanes were subjected to 12.5% sodium dodecyl sulfate–polyacrylamide gel electrophoresis. (B), Western blotting of serum samples with monoclonal antibodies (nos. 03E5 and 77). Lane 1, normal human serum; lanes 2–4, sera from ulcerative colitis patients; lane 5, purified human HMBG1 + purified human HMGB2; lanes 6–13, sera from sepsis patients. Antibodies used were peroxidase-conjugated anti-mouse immunoglobulin polyclonal antibody (lane 1), monoclonal antibody no. 03E5 (lanes 2–10), and monoclonal antibody no. 77 (lanes 11–13). There was no difference with the two different monoclonal antibodies (nos. 03E5 and 77) or the peroxidase-conjugated anti-mouse immunoglobulin polyclonal antibody (data not shown). (C), linear regression analysis of ELISA vs Western blot. Linear regression analysis of human serum HMGB1 concentrations in 25 sepsis patients infected with unidentified pathogens. We next detected HMGB1 in the serum of sepsis patients and ulcerative colitis patients by Western blotting. Serum was fractionated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis, and proteins were transferred electrophoretically to nitrocellulose membranes (Millipore). After the membranes were blocked, the monoclonal antibody to HMGB1 (no. 77 or 03E5) was dissolved in phosphate-buffered saline (PBS), and the membrane was soaked in the solution at room temperature overnight. After the membrane was washed, it was incubated at room temperature for 2 h in a solution of peroxidase-labeled anti-mouse immunoglobulin polyclonal antibody (Dako) diluted 30 000-fold with PBS containing 30 g/L BSA. After another washing step, the membrane was soaked in ECL reaction reagent (Pharmacia) according to the manufacturer’s instructions. Chemiluminescently labeled bands were visualized after 10-min exposure to Xomat XAR5 film (Eastman Kodak Co.). HMGB1 was determined by immunoblot analysis and quantified from the blots by measurement of the absorbance of the band (NIH 1.59 software; NIH) and comparison with a calibration curve constructed with purified human HMGB1 serially diluted in normal human serum. Western blotting results for serum samples from sepsis patients and ulcerative colitis patients are shown in Fig. 1B . A band that reacted with anti-calf thymus HMGB1 monoclonal antibody was detected at the same position as human HMGB2. In addition, this band also reacted with a commercial anti-HMGB2 polyclonal antibody (Santa Cruz Inc. and BD Biosciences) that reacts only with HMGB2 (data not shown). We thus found that both HMGB1 and -2 are present in the sera of sepsis and ulcerative colitis patients. HMGB2 was also present in the cerebrospinal fluid, synovial fluid, and culture medium in which HMGB1 was detected. In this study, we developed an ELISA method that specifically measured HMGB1 only, using two types of monoclonal antibodies that we prepared. Polystyrene microtiter plates (Nunc) were coated with 100 μL of anti-calf thymus HMGB1 monoclonal antibody (no. 03E5) in PBS. The remaining binding sites in the wells were blocked by incubating the plates for 2 h with 400 μL/well PBS containing 10 g/L BSA. After the plates were washed, 100 μL of each dilution of the calibrator and samples 1:1 dilutions in 0.2 mol/L bis-Tris (pH 6.5)–0.15 mol/L NaCl containing 10 g/L BSA was added to the wells. The microtiter plates were incubated for 2 h at room temperature. After the plates were washed, 100 μL of anti-peptide 1 peroxidase-conjugated monoclonal antibody (no. 77) was added to each well, and the plates were incubated at room temperature for 3 h. After another washing step, 100 μL of PS-1 (Luminogen) was added to each well. The luminescence was measured with a 9000D microplate luminescence reader (DIA-IATRON). The amount of purified human HMGB1 was determined by the method of Lowry et al. (10) as modified by Markwell et al.(11) with BSA as the calibrator. The working range of the assay is 2–100 μg/L. The lower detection limit of the ELISA method is defined as the mean (± 3 SD) minimum HMGB1 concentration at which there is no overlap. Accurate measurement of HMGB1 can be achieved at concentrations >2 μg/L. Using the ELISA method, we analyzed plasma samples with HMGB1 protein concentrations of 5.4 and 75 μg/L in 20 replicates for assessment of intraassay imprecision. Interassay imprecision was determined by assaying plasma samples of 5.4 and 75 μg/L over 10 separate assays (n = 3). The intraassay CV at the two concentrations was <5%, and the interassay CV was ∼10%. Recovery, calculated from data (n = 10) at several concentrations of purified human HMGB1 added to pooled serum and calculated as the ratio of the observed concentration to the expected concentration multiplied by 100%, was 92–111% in the HMGB1 ELISA assay. We measured HMGB1 concentrations in serum samples from 35 patients with sepsis at Sapporo Medical University by Western blot and ELISA. The correlation between the values obtained by both methods was good: y = 1.067x − 2.86 μg/L (r = 0.922; Fig. 1C ). This study revealed that HMGB1 and HMGB2 coexist in serum. Moreover, we developed a new monoclonal antibody that binds only HMGB1 without reacting with HMGB2 and used that antibody to develop an ELISA that can measure HMGB1 in serum.
Yamada et al. (Mon,) studied this question.