|
|
Blood, Vol. 109, Issue 8, 3521-3528, April 15, 2007
Generation of activation-specific human anti- Mß2 single-chain antibodies as potential diagnostic tools and therapeutic agents
Blood Eisenhardt et al.
109: 3521
Supplemental materials for: Eisenhardt et al, Vol 109, Issue 8, 3521-3528
Files in this Data Supplement:
- Table S1. List of primers used for the site-directed mutations of selected amino acids in the HCDR3 domain of MAN-1, MAS-1, and MAS-2 (PDF, 47.4 KB)
- Figure S1. Amplification of activation-specific anti–Mac-1 antibodies over 4 rounds of subtractive panning via phage display (JPG, 42.2 KB)
-
(A) Course of panning. After each round of panning, the rescued phages were used for infection of log-phase XL-1 blue E coli bacteria. The numbers of colonies, which represent the number of phage clones, were counted. The natural and synthetic libraries demonstrated a very similar course of panning. After low yields with around 2000 colonies in the first panning rounds, we had a significant increase of selected phages of up to 6000 clones in the final panning rounds. The increase of colonies in the final rounds represents the amplification of scFv clones, which bind to the activated receptor but not to the inactivated Mac-1 receptor. (B) Fingerprinting of natural clones by BstNI and RsaI digest. The diversity of the natural clones was evaluated by digestion with the restriction enzymes BstNI and RsaI. Phagemid DNA of 20 randomly picked natural clones of panning rounds 2, 3, and 4 was purified and digested with the restriction enzymes and separated in agarose gel–electrophoresis and stained with ethidium bromide. The DNA markers lambda DNA/HindIII (lane 1) and phiX174DNA/HaeIII (lane 2) are used as comparison. The restriction pattern of scFv clones differs widely after panning rounds 2 and 3. In contrast, after panning round 4, all investigated clones demonstrated an identical restriction pattern. These results demonstrate the power of positive clone amplification using the developed depletion/selection system over the course of consequent panning rounds. In the case of the synthetic library, where restriction pattern analysis does not work because of sequence identity outside the HCDR3 region, 10 randomly picked clones were sequenced, revealing 2 distinct clones, each represented 5 times.
- Figure S2. MAN-1 production and purification (JPG, 39.5 KB)
-
(A) Silver staining of SDS-PAGE. (B) Western blot probed with an anti–HIS-tag HRP-coupled antibody. Phagemid DNA was cloned into the expression vector pHOG-21 using the restriction enzymes NcoI and NotI and transformed into TG-1 E coli. These bacteria were grown at 37°C to an optical density of 0.8 in LB medium containing glucose (50 mM) and 100 µg/mL ampicillin. Bacteria were then transferred to LB medium containing 0.4 M sucrose, 100 µg/mL ampicillin, and 0.25 mM IPTG and incubated for 16 hours at 200 rpm at 23°C. For the isolation of the periplasma, bacteria were centrifuged at 3000g for 10 minutes and resuspended in 5 mL/mg of pellet 1X BugBuster (Novagen, Madison, WI) solution. After 30 minutes of incubation at room temperature and centrifugation for 30 minutes at 10 000g at 4°C, the supernatant containing the periplasmic proteins was run over a Ni-NTA agarose column (Qiagen). The column was washed twice with washing buffer containing 50 mM NaH2PO4, 300 mM NaCl, and 20 mM imidazole (pH 8.0), and then eluted with 500 µL elution buffer per liter bacterial culture, containing 50 mM NaH2PO4, 300 mM NaCl, and 250 mM imidazole (pH 8.0). Subsequently, the purified protein was dialyzed against PBS in Slide-A-Lyzer dialysis cassettes (Pierce, Rockford, IL) with a molecular mass cutoff at 10 000 Da. Production and purification were monitored by SDS-PAGE with silver staining (A) and Western blotting (B). Gel and blot show bacterial culture (lane I), supernatant of the first centrifugation step (lane II), lysate after BugBuster treatment of the bacterial pellet (lane III), flow through of the Ni-NTA agarose column (lane IV), first washout of the Ni-NTA agarose column (lane V), second washout of the Ni-NTA agarose column (lane VI), empty (lane VII), and eluate of the Ni-NTA agarose column (lane VIII). The bacterial suspension contains a low concentration of scFv (lane I). After centrifugation, the scFv is not detectable in the supernatant (lane II) but at a high concentration in the lysate of the bacterial pellet (lane III). The flow through of the Ni-NTA agarose column (shown in lane IV) shows a large amount of protein (silver staining), but no signal in the His-tag staining, indicating that the scFvs are bound to the Ni-NTA agarose, whereas unspecific proteins flow through. The following washing steps (lanes V-VI) demonstrate further washout of decreasing amounts of proteins that include only a small portion of scFv. The comparison between the bacterial suspension (lane I) and the final eluate (lane VIII) demonstrates the power of bacterial protein expression/isolation and the Ni-His-tag purification system.
- Figure S3. Sequences and alignment of activation-specific anti–Mac-1 single-chain antibodies (MAN-1, MAS-1, and 2) (JPG, 76.6 KB)
-
(A) The full sequence of MAN-1 is given. The CDRs of the light and heavy chains, the linker region, and the His-tag are marked. (B) The HCDR3 regions, as the central binding sites of the scFv’s were compared between the 3 generated activation-specific anti–Mac-1 clones. The analysis was performed with Clustal multiple sequence alignment program (HUSAR; German Cancer Research Institute, Heidelberg, Germany). Red indicates more than half of the amino acids of a column are identical or demonstrate strong similarities; gray, more than half of the amino acids of a column demonstrate weak similarities.
- Figure S4. Evaluation of cross-reactivity with other related integrins (JPG, 148 KB)
-
(A) MAN-1 does not exhibit cross-reactivity with the  2-integrins LFA-1 ( L2; CD11a/CD18), p150/95 (X2; CD11c/CD18), and D2 (CD11d/CD18) MAN-1 cross-reactivity with 2-integrins other than Mac-1 was investigated in flow cytometry. Leukocytes in whole blood were activated, and MAN-1 binding was evaluated as described in “Materials and methods.” Blocking antibody clones TS1 (CD11a; ATCC, Manassas, VA), 2LPM19c (CD11b; Dako), BU15 (CD11c; Serotec, Raleigh, NC), and 240I (CD11d; kindly provided by ICOS, Bothell, WA) were added in various concentrations and incubated for 15 minutes. Subsequently, MAN-1 was added at a concentration of 5 µg/mL for 10 minutes, and binding was detected by an Alexa Fluor 488–conjugated anti–His-tag secondary antibody. In parallel, 2-integrin expression was assessed with the same antibodies and a secondary FITC-coupled goat antimouse antibody. The FITC-coupled goat antimouse antibody alone served as a control. Measurements were performed in triplicates. A typical result out of 3 experiments is shown. The blocking anti-CD11b antibody reduces MAN-1 binding to the background level at a concentration of 50 µg/mL. In contrast, at the same concentrations, the other blocking antibodies did not inhibit MAN-1 binding. These findings imply a selective binding of MAN-1 to Mac-1 without cross-reactivity to other 2-integrins. (B) scFv MAN-1 immunoprecipitates CD11b from monocyte lysates, but not CD11a, CD11c, or CD11d. Lysed monocytes were incubated with 10 µg/mL MAN-1 and anti-His(6) antibody (Novagen). Subsequently, protein G sepharose beads (Zymed, San Francisco, CA) were used to precipitate bound proteins. Samples were run on SDS-PAGE, western blotted on a nitrocellulose membrane (Millipore, Billerica, MA), and stained for either CD11a (clone 25.3.1; Immunotech), CD11b (Santa Cruz Biotechnology, Santa Cruz, CA), CD11c (clone BU15; Serotec) or CD11d (clone 169A; ICOS), and detected by secondary goat anti–mouse HRP (Pierce) antibody for CD11a, CD11c, and CD11d, or donkey anti–goat HRP (Santa Cruz Biotechnology) for CD11b. A chemiluminnescence substrate (Pierce) was used to make the bands visible. Results were documented on a Chemidoc Imager (Bio-Rad, Hercules, CA) and analyzed by Quantity One software (Bio-Rad). MAN-1 is able to precipitate CD11b. Faint bands for CD11a, CD11c, and CD11d are only visible in the monocyte lysate, but not in the MAN-1 immunoprecipitation. A typical result out of 3 immunoprecipitations is shown. (C) MAN-1 does not exhibit cross-reactivity to GPIIb/IIIa (IIb3; CD41/CD61). Cross-reactivity with another fibrinogen-binding integrin was assessed in flow cytometry using CHO cells expressing either a GFFKR-deleted and thereby activated GPIIb/IIIa, or the native and thereby nonactivated platelet integrin GPIIb/IIIa (details of cell generation have been described previously).1,2 The expression of the activated conformation was determined by the binding of the activation-specific antibody Pac-1. MAN-1 binding was measured by an Alexa Fluor 488–conjugated anti–His-tag secondary antibody. MAN-1 binds neither to nonactivated nor activated GPIIb/IIIa. Thus, no cross-reactivity to GPIIb/IIIa was seen. A typical result out of 3 experiments is shown.
- Figure S5. Binding of C3bi to activated Mac-1 can be inhibited by a blocking anti-CD11b antibody, but not by MAN-1 (JPG, 34.1 KB)
-
C3bi (10 µg/mL) was incubated with activated monocytes, and binding was detected by a biotinylated C3bi antibody and avidin-PE in flow cytometry. Blocking antibody clone 2LPM19c inhibits C3bi binding, whereas MAN-1 and a control antibody (against CD7) did not inhibit C3bi binding. The experiment was performed in triplicate. One representative result out of 3 experiments is shown.
- Figure S6. Inhibition of adhesion of Mac-1–expressing CHO cells on immobilized fibrinogen under flow conditions by MAN-1 (JPG, 73.8 KB)
-
Cells expressing the native, low-affinity Mac-1, and cells expressing the GFFKR-deleted, activated Mac-1 were examined under venous flow as well as under arterial flow conditions. A Mac-1–blocking mAb was used as an activation-nonspecific control. Quantification of adhering cells was performed by counting 5 fields. Mean and standard deviation of 5 experiments are given. MAN-1 inhibits the binding of cells expressing the activated Mac-1 receptor but not the native Mac-1, whereas the unspecific CD11b-blocking antibody clone 2LPM19c inhibits the binding of both cell lines to the fibrinogen matrix.
|
|
