Screening for a Human Single Chain Fv Antibody Against Epitope on Amyloid-Beta 1-40 from a Human Phage Display Library
Zhao Zhen-fu,Gao Guo-quan,Liu Shu,Zou Jun-tao,Xee Yao,Yuan Qun-fang,Wang Hua-qiao,Yao Zhi-bin
DOI: https://doi.org/10.1097/00029330-200711010-00016
2007-01-01
Abstract:A myloid-beta peptides (Aβ) are believed to be responsible for the mental decline in patients with Alzheimer's disease (AD). In 1999, Schenk et al1 reported that immunization with Aβ attenuated AD-like pathology in the PDAPP mouse, and developed a new vaccination approach to AD. Such vaccines were successfully tested in mouse models of AD for the reduction of Aβ plaque burden and the improvement of cognitive performance. However, 6% of AD patients developed symptoms of brain inflammation after vaccination that resembled encephalitis or meningitis, resulting in the termination of the phase IIa clinical trial. These adverse effects were considered to be related to Aβ1-42 toxicity and/or cell-mediated auto-immunity.2 One of the disadvantages of active immunization procedure is the generation of very robust cellular immune response that may not be desirable in elderly patients with AD.3 Currently, most scholars believe that passive immunotherapy could be more appropriate for AD patients. Peripheral administration of polyclonal and monoclonal antibodies against Aβ is reported to reduce plaque burden along with reduction of pathology in a mouse model for AD.4,5 These results indicate that in the absence of T-cell immunity, antibodies are sufficient to decrease amyloid deposition and AD-like pathology. However, passive immunization also demonstrates such adverse side effects as microhemorrhages.6 Efforts have been made to further screen antibodies that may be biologically functional without being associated with the development of adverse events.7 This could be achieved by using phage-displayed recombinant antibody fragment libraries, such as single chain variable fragments (scFv) libraries that result in the generation of a panel of specific antibodies quickly, easily and inexpensively in vitro. In the present study, we constructed an human phage-displayed scFv antibody library, isolated a scFv antibody with relatively high specificity and affinity by using Aβ1-40 protein as a target, and mapped the binding epitope on Aβ1-40. Our results provided an alternative way for diagnosis and passive immunotherapy of AD patients. METHODS Construction of the scFv antibody phage display library The B cells were isolated from peripheral blood lymphocytes of twenty AD patients, and total RNA of B cells was prepared by cDNA synthesis. The VH (heavy chain variable region) and VL (light chain variable region) genes were amplified from cDNA with polymerase chain reaction (PCR) using a specific primer8 which incorporated the restriction sites BssHII and SalI for VL and XhoI and NheI for VH. The PCR products obtained were gel purified using a rapid gel extraction system (Omega Bio-tech, USA), digested with the restriction enzymes above, and cloned into the similarly digested vector pDAN5 (provided by Professor YAN Xi-yun, the National Laboratory of Biomacromolecules, Institute of Biophysics, Beijing, China) respectively. The recombinant pDAN5 was then transformed into Escherichia coli XL1-blue by electroporation, and the transformants were plated on dishes with LB, containing 1% glucose, 100 μg/ml ampicillin, and 10 μg/ml tetracycline. The positive colonies were collected by scraping, rescued by helper phage M13K07 at a multiplicity of infection (MOI) of 20:1, and finally stored as a primary scFv library. Diversity assessment of the scFv library A number of individual clones from the primary library were randomly picked and their phagemids with scFv were prepared with a Plasmid Extraction Kit (Omega Bio-tech, USA). After amplifying with PCR, the individual scFv gene was digested by BstNI, and the fingerprinting of each colony was analyzed by 4% agarose gels. Screening of Aβ1-40 binders An amine-binding 96-well microtiter plate (Nunc, Rochester, NY, USA) was coated with Aβ1-40 (Sigma Bio-tech, USA) at 10 μg/ml in coating buffer (0.05 mol/L Na2CO3-NaHCO3, pH 9.6) overnight at 4°C, and another plate was coated with bovine serum albumin (BSA) as control. After washing with 0.1% PBST (0.1% Tween 20 in phosphate-buffered saline (PBS)), the plates were blocked for 2 hours at 37°C with 2% BSA in PBS, and the phage particles mixed with 1% BSA were added at a concentration of 1×1012/ml. After being incubated for 3 hours at room temperature, the plates were intensively washed with 0.1% PBST and PBS, and bound phage antibodies were eluted using 100 μl/well of 0.1 mol/L glycine/HCl, pH 2.2. The eluted phages were immediately neutralized with 1 mol/L Tris-HCl solution (pH 8.0), and the output titer was measured. In each round, the enriched phagemid library was rescued using helper phage M13K07. Four rounds of biopanning were performed. After the fourth round of panning, the elute was plated on the agar plates containing 100 μg/ml ampicillin and 1% glucose, and 96 individual clones were randomly picked up, rescued with M13K07 and evaluated by enzyme-linked immunosorbant assay (ELISA). Selection of the specific scFv antibody binding to Aβ1-40 The wells of a 96-well microtiter plate were coated overnight at 4°C with 10 μg/ml Aβ1-40 and blocked with 2% BSA for 2 hours in 37°C. Each individual phage scFv mixed with 1% BSA was added at a concentration of 1×1010 and incubated for 3 hours at room temperature. After intensive washes, the bound antibodies were detected by a mouse monoclonal anti-M13 antibody (Amersham Pharmacia Biotech, Sweden) diluted to 1/5000, followed by a horse-radish peroxidase (HRP) conjugated anti-mouse IgG as the second antibody using the 3,3',5,5'-tetramethylbenzidine (TMB) as a substrate. The color reaction was stopped with H2SO4 and the OD450 value was measured in a Bio-Rad ELISA reader (Hercules, CA, USA). DNA sequencing The selected positive scFv-expressing phagemid clone was cultured and rendered to Invitrogen Biotechnology Co., Ltd., and the DNA sequences were determined. The amino acid sequence and the complementarity determining regions (CDRs) of the selected scFv antibody were determined by providing the DNA sequences to the Immunogenetics Database (IMGT) of the European Bioinformatics Institute (EBI). Mapping the epitope of Aβ1-40 binding to the specific antibody by pre-absorption test and competitive ELISA The wells of a 96-well plate were coated overnight with 10 μg/ml Aβ1-40 in coating buffer. After blocking, 100 μl selected phage-scFv antibodies at the concentration of 1×1010 were loaded directly or after incubation with their inhibitors including Aβ1-40, Aβ1-16 and Aβ17-28 (Sigma Bio-tech) for 2 hours at 37°C. The plate was washed and loaded with secondary antibodies as described above, and the OD450 value was determined. Statistical analysis The data were expressed as the mean±standard deviation (SD). One way analysis of variance (ANOVA) was used to compare the results among the groups. A P value less than 0.05 was considered statistically significant, and the SPSS 11.0 software was used. RESULTS Construction of the human anti-Aβ scFv antibody phage display library To make a large diversity of a scFv library with a high affinity for Aβ1-40, we used pDAN5 as a phage-display vector and twenty AD patients' B cells (from about 300 ml peripheral blood) as a repertoire of antibodies. Using the PCR technique, seven groups of VH and thirteen groups of VL (four for Vκ and nine for Vλ) were amplified, and the sizes of the VH and VL were about 350 bp (Fig. 1A). The amplified VH and VL were inserted into the vector respectively and then electroporatedly transformed into E. coli XL1-blue to make a primary scFv library.Fig. 1.: Construction of the scFv library. A: Seven groups of VH and thirteen groups of VL (four for Vκ and nine for Vλ) were amplified by RT-PCR; B: 10 scFv clones were randomly selected from the primary library and their genes were amplified by PCR; C: Fingerprinting of the 10 scFv genes digested 2 hours at 60°C by BstNI and then analyzed in 4% agarose gel, showing a diversity of the primary antibody library.To assess the diversity of the scFv library, 10 clones from the primary library were randomly selected and their scFv genes were amplified with PCR, then the scFv genes were digested with BstNI since a number of BstNI sites randomly exist in the various regions of the antibody. All of the 10 clones contained a similar size of full length scFv around 750 bp as shown in Fig 1B. However, each scFv clone showed a unique BstNI-digested fingerprinting pattern (Fig. 1C). These data indicate that individual clones in the primary library are different. This library was calculated to have a diversity of 1.5 ×106 members. Selection of the specific scFv antibody binding to Aβ1-40 Four rounds of bio-pannings were performed with stringent conditions to enrich phage scFv for Aβ1-40 aforementioned, and the bound phages (output) were calculated after each panning by infecting E. coli cells and using serial dilution on agar plates containing ampicillin and glucose. The output/input ratio was gradually increasing, especially after the third round, and it was dramatically increasing in the fourth round. However, the phage scFv to immobilized control BSA did not show any significant changes (Fig. 2).Fig. 2.: Four panning rounds showing an obvious enrichment of scFv/phage to Aβ1-40. There is no significant change in immobilized control BSA during the third and fourth round.From the fourth round of panning, 96 phage clones were randomly selected for ELISA to evaluate their binding activity to Aβ1-40, and 77 clones were identified with specifically binding to Aβ1-40. Ten clones with higher ELISA readings were selected, and another ELISA was done, and one referred as H1 was selected as a specific binder because of the highest ELISA readings against Aβ1-40. DNA sequencing and amino acid sequence of H1 The DNA sequence of H1 was determined by providing the clone to Invitrogen Biotechnology Co., Ltd., using the upstream primer 5′-GCAGCCGCTGGATTGTTATTA-3′ and the downstream primer 5′-TTGTCGTCTTTCCA-GACGTTA-3′. The amino acid sequence and the complementarity-determinging regions (CDRs) of the selected scFv antibody were determined by providing the DNA sequences to the Immunogenetics Database (IMGT) of the EBI (Table). According to its sequence, the scFv H1was identified as the VH2 family and Vκ1 family.Table: The complementarity-determining regions (CDRs) of deduced amino acids sequenceBinding epitope mapping In order to map the epitope of Aβ1-40 binding to the specific antibody H1, pre-absorption test and competitive ELISA were performed as described above. The results showed an inhibition rate > 50% with Aβ1-40 and an inhibition rate > 50% with Aβ1-16 in contrast to no inhibition with Aβ17-28 (P<0.01) (Fig. 3). These results indicated that the binding epitope of the scFv antibody, H1, was located between amino acids 1 and 16 of Aβ1-40.Fig. 3.: The binding activity of H1 to coated Aβ1-40 and pre-absorption with different antigen including 2 μg Aβ1-40, Aβ1-16 and Aβ17-28.DISCUSSION A number of different immunization strategies for clearing Aβ in the brains of AD patients have been proposed or studied including active immunization with Aβ peptide or phage peptide and passive immunization with monoclonal or polyclonal antibodies.9 Both have been effective in preventing or removing amyloid-β deposits and improving cognitive performance in animal models of the disease and, potentially, in AD patients. However, the appearance of sub-acute meninggo-encephalitis in 6% of AD patients in the Elan Phase IIa active immunization trial3 underscores the safety concerns of immunotherapy approaches. One of the reasons of these adverse events is considered to be the eliciting of T-lymphocyte, which can lyse cells that overproduce Aβ and trigger an inflammatory response.9 Passive administration of antibodies is not expected to elicit T-cell responses and the appearance of meningoencephalitis is not related to plasma anti-Aβ titers, so is considered more appropriate for AD patients.5,9 However, passive immunization also demonstrates such adverse side effects as microhemorrhage.6 Recent studies have demonstrated that the fragments of monoclonal antibody, including F(ab)2 and scFv, also can effectively inhibit Aβ aggregation and prevent Aβ-induced neurotoxicity.4,5 These results suggest that direct disruption of fibrils, in addition to classical Fc-dependent phagocytosis, is involved in the antibody-mediated clearance of Aβ aggregates in vivo. Single chain antibody fragments do not include the Fc portion of the antibody, therefore they cannot initiate the complement cascade and limit the inflammatory response to binding to Aβ peptide. Also passive immunization with scFvs should not stimulate T-lymphocyte activity, nor should it activate microglia cells, which could also result in a widespread inflammatory response. Moreover, scFv has a much lower molecular weight than a full antibody, potentially facilitating its transfer into the brain, so antibodies of single chain fragments represent a promising immunotherapeutic approach for AD. Formerly, most antibody libraries were constructed by using B cells of the immunized mice, and the antibodies selected from such a library may evoke human anti-mouse antibodies (HAMA), which limited their application in human bodies as diagnostic and therapeutic reagents. Also some scholars screened antibodies from the human naive antibody library. But it is difficult to obtain an antibody with high specificity and affinity from such an antibody library unless the library is very large. In our experiment, we constructed first time an scFv library, which was derived from B cells of 20 AD patients. Because of the higher titers of Aβ peptides in peripheral blood of AD patients than those in normal men, the antibody must have been undergoing the maturing course, and the library could be considered to be an “immune” human scFv library. Compared with the naive or synthetic antibody library, the immune library is more likely to obtain nature-like antibody with high specificity and affinity.10 Thus, it is more feasible to select neutralizing antibody by panning against Aβ peptide from such an “immune” human scFv phage display library. The selected scFv H1 in our study showed a specific binding to Aβ1-40in vitro. Moreover, we demonstrated that the binding epitope of H1 is located between amino acids 1 and 16 of Aβ1-40 by competitive ELISA. The N-terminal region of Aβ peptide has been verified as an important epitope that elicits neutralization antibody in previous reports even through the antibodies against the center domain or the C-terminal region of Aβ peptide could also reverse the pathology and the Aβ-induced neurotoxicity. In general, the center domain and C-terminal region of Aβ peptide are considered to be the dominant T-cell epitope, and could elicit the cellular immune responses which are not expected. While the N-terminal region of Aβ contains a dominant B-cell epitope,11 and the majority of antibodies that reduce plaque burden and that have been isolated from patients in the immunization trials of Elan are directed towards the N-terminal residues of the peptide.14 Because of the obvious effects in preventing or interfering with Aβ-dependent neuropathology and subsequent progressive cognitive decline, active immunization continues to be an attractive therapeutic target for the prevention and/or treatment of AD in spite of the adverse reactions. Many researchers improved the method of active immunization after the cessation of the phase IIa clinical trial of Elan, and most of them selected the N-terminal residues of the peptide as the antigen to immunize the animal models, and obtained promising results.2,11 The present study mapped the epitope of the scFv antibody which was selected from a library using Aβ1-40 as a target, and demonstrated that the epitope was located in the N-terminal region of the peptide. This result additionally supported the strategy in designing a vaccine for current use. In summary, we constructed an “immune” human phage displayed scFv antibody library, and developed an effective method to obtain an antibody, which mimics natural antibody in immune sera. This study would be significantly helpful not only in developing a new human neutralizing antibody as an attractive alternative to more conventional antibody-based diagnosis and therapeutics to AD, but also in better understanding of the immunogenic characteristics of Aβ peptide and the designing of Aβ vaccine.