Biochemical features of a catalytic antibody light chain, 22F6, prepared from human lymphocytes

Abstract

Human antibody light chains belonging to subgroup II of germ line genes were amplified by a seminested PCR technique using B-lymphocytes taken from a human adult infected with influenza virus. Each gene of the human light chains was transferred into the Escherichia coli system. The recovered light chain was highly purified using a two-step purification system. Light chain 22F6 showed interesting catalytic features. The light chain cleaved a peptide bond of synthetic peptidyl-4-methyl-coumaryl-7-amide (MCA) substrates, such as QAR-MCA and EAR-MCA, indicating amidase activity. It also hydrolyzed a phosphodiester bond of both DNA and RNA. From the analysis of amino acid sequences and molecular modeling, the 22F6 light chain possesses two kinds of active sites as amidase and nuclease in close distances. The 22F6 catalytic light chain could suppress the infection of influenza virus type A (H1N1) of Madin-Darby canine kidney cells in an in vitro assay. In addition, the catalytic light chain clearly inhibited the infection of the influenza virus of BALB/c mice via nasal administration in an in vivo assay. In the experiment, the titer in the serum of the mice coinfected with the 22F6 light chain and H1N1 virus became considerably lowered compared with that of 22F6-non-coinfected mice. Note that the catalytic light chain was prepared from human peripheral lymphocyte and plays an important role in preventing infection by influenza virus. Considering the fact that the human light chain did not show any acute toxicity for mice, our procedure developed in this study must be unique and noteworthy for developing new drugs.

Keywords: Catalytic Antibody; DNase; Enzyme Catalysis; Human Light Chain; Influenza Virus; Influenza Virus Type A; Molecular Modeling; RNA Catalysis.

FIGURE 1.

SDS-PAGE analysis of purified light chain. M, marker; 1, reduced condition; 2, non-reduced condition. The purified light chain sample (250 ng) was separated by 12% acrylamide gel electrophoresis and visualized by silver staining. Under the non-reduced condition, the monomeric (26-kDa) and dimeric (51- and 43-kDa) forms of 22F6 were observed. A weak band at 43 kDa may be a different dimeric form from that of the band at 51 kDa. Under the reduced condition, no other bands except for the 31 kDa band of the monomeric form were detected, indicating high purity.

FIGURE 2.

Structure model for 22F6 light chain. A, side chains having the potential to form serine protease-like triads (Asp¹ (or Asp³⁴), Ser^27a, and His^27d) are shown. Asp, His, and Ser are colored red, blue, and green, respectively. The distance between Cα of His^27d and Ser^27a is 8.26 Å, which is close to the Cα distance (8.42 Å) between them in a catalytic triad of trypsin. B, Tyr³² is visualized. The residue is situated between Asp³⁴ and His^27d and may be concerned with the nuclease activity. Models for the light chain were visualized using WebLab ViewerLite (Accelrys Inc., San Diego, CA).

FIGURE 3.

Hydrolysis of peptide-MCA substrates and nucleic acids by 22F6 light chain. A, hydrolysis of peptide-MCA substrates. The 22F6 light chain (5 μm) was mixed with an equal volume of 50 μm synthetic substrate QAR-MCA (solid circle), EAR-MCA (open circle), EKK-MCA (solid triangle), VPR-MCA (open triangle), K-MCA (solid square), or R-MCA (open square) and incubated at 37 °C. QAR-MCA and EAR-MCA were cleaved with a turnover value of 0.018 and 0.007/h, respectively. B, hydrolysis of a phosphodiester bond of the plasmid DNA pBR322. The 22F6 light chain inserted a nicked and a linear form of the plasmid. The nicked form as well as the linear form became clearer, and the supercoiled form became fainter at 30 min of incubation time. The hydrolysis did not occur without magnesium ion. C, kinetic analysis. The decrease of the supercoiled form obeyed the Michaelis-Menten equation (k_cat = 1.01 × 10⁻²/min and K_m = 4.36 × 10⁻⁷ m). These values are comparable with those of Rodkey's catalytic antibody BV 04-01. D, cleavage of genome RNA from Noda virus. Noda virus possesses RNA-1 and RNA-2. In lane 3, RNA-1 at 3.1 kb (open arrow) disappeared at a concentration of 46.5 μg/ml 22F6. In lane 4 (465 μg/ml 22F6), several fragments (closed arrows) generated from the parent RNA-1 could be observed at around 1,000, 760, and 630 kb. The cleavage was dependent on the concentration of 22F6. M, marker. Lanes 1–4, 0, 4.65, 46.5, and 465 μg/ml 22F6. Reaction conditions were as follows: Noda virus, 200 ng; temperature, 25 °C; period, 24 h; volume, 20 μl.

FIGURE 4.

Western blot analysis for 22F6 light chain. Western blot of 22F6 light chain was performed against H1N1 virus proteins. A faint band was identified at around 26 kDa, which corresponds to the HA₂ domain of the hemagglutinin molecule. The experiment used 50.9 μg/ml 22F6, which suggests that 22F6 had a lower affinity than a normal whole antibody. We could detect only the band corresponding to HA₂, indicating that 22F6 recognized the hemagglutinin molecule.

FIGURE 5.

In vitro assays. The human catalytic light chain 22F6 (50 μg/ml) was first incubated with 500 pfu plus 0.2 ml of influenza virus H1N1 or H3N2 for 24 or 48 h at 25 °C, and then the mixture was inoculated into the MDCK cell monolayer. The experiments were repeated twice. Although the catalytic light chain 22F6 showed little effect on the suppression of infectivity for both influenza virus H1N1 and H3N2, the catalytic light chain clearly suppressed infection of influenza virus H1N1 at 48 h of incubation. Error bars, S.D.

FIGURE 6.

In vivo assays. A, negative control mice (n = 2 mice) that received nasal administration without influenza virus (only PBS was given; mock group) were examined as shown. The time courses of the body weight are presented in the top two graphs. The left graph shows the time course of the weight for each mouse. The right graph shows the time course for the average body weight of two mice. The time courses of body weight were normal, as shown in the graphs. Photographs of the mice were taken at 7 and 14 dpi. No abnormalities were observed. B, positive control mice infected with intact influenza virus (PR-8 strain) were investigated. The body weight of four mice decreased until 7 dpi. All (four) mice exhibited decreases in body weight until 7 dpi. Two of the four mice died at 8 dpi (50% survival rate), and the other two mice were able to survive and recover their body weight by the final day (21 dpi). The appearance of mouse 2 and 3 was abnormal. Their bodies became smaller, and their hair bristled up. C, effect of 22F6 catalytic light chain. Influenza virus H1N1 (PR-8 strain) was incubated with the catalytic light chain 22F6 at 25 °C for 48 h, and then 50 μl of the mixture was given to mice (BALB/cN Sea mice, 6 weeks old, female) via nasal administration (n = 4 mice). One mouse decreased in body weight by 7 dpi but recovered it after 12 dpi. The other three mice increased in body weight by 21 dpi. No mice died during the assay. The human catalytic light chain 22F6 exhibited a huge effect against influenza virus infection even in the in vivo assay. Error bars, S.D.

FIGURE 7.

Dose dependence. The dose dependence of 22F6 was investigated. In this experiment, the virus at 2,000 pfu/50 μl was employed. Administration of 50 μg/ml 22F6 did not show an effect (0% survival), but that of 465 μg/ml 22F6 substantially improved the survival rate (65%) of the mice.

FIGURE 8.

Serum titer. The antibody titer against influenza virus in the mice serum tested in the above in vivo assay was examined. The titer of control mice infected with intact virus gradually increased, depending upon the time course measured. At 21 dpi, the titer reached 5,000-fold, whereas mock mice inoculated with only PBS did not show any response. On the other hand, the titer of the mice infected with a solution including intact virus and 22F6 catalytic light chain was 2,500-fold at 21 dpi, which was significantly lower than that of the positive control mice. This means that some part of the antigenicity of the influenza virus was lost by contact with the catalytic light chain 22F6.

FIGURE 9.

Acute toxicity tests. The time courses of the body weight of mice administered 145.5 mg/kg at single dose are shown. No significant differences between the administered (●) and control (○) mice were observed. Error bars, S.D.