Directed evolution of an LBP/CD14 inhibitory peptide and its anti-endotoxin activity

Abstract

Background: LPS-binding protein (LBP) and its ligand CD14 are located upstream of the signaling pathway for LPS-induced inflammation. Blocking LBP and CD14 binding might prevent LPS-induced inflammation. In previous studies, we obtained a peptide analog (MP12) for the LBP/CD14 binding site and showed that this peptide analog had anti-endotoxin activity. In this study, we used in vitro directed evolution for this peptide analog to improve its in vivo and in vitro anti-endotoxin activity.

Methods: We used error-prone PCR (ep-PCR) and induced mutations in the C-terminus of LBP and attached the PCR products to T7 phages to establish a mutant phage display library. The positive clones that competed with LBP for CD14 binding was obtained by screening. We used both in vivo and in vitro experiments to compare the anti-endotoxin activities of a polypeptide designated P1 contained in a positive clone and MP12.

Results: 11 positive clones were obtained from among target phages. Sequencing showed that 9 positive clones had a threonine (T) to methionine (M) mutation in amino acid 287 of LBP. Compared to polypeptide MP12, polypeptide P1 significantly inhibited LPS-induced TNF-α expression and NF-κB activity in U937 cells (P<0.05). Compared to MP12, P1 significantly improved arterial oxygen pressure, an oxygenation index, and lung pathology scores in LPS-induced ARDS rats (P<0.05).

Conclusion: By in vitro directed evolution of peptide analogs for the LBP/CD14 binding site, we established a new polypeptide (P1) with a threonine (T)-to-methionine (M) mutation in amino acid 287 of LBP. This polypeptide had high anti-endotoxin activity in vitro and in vivo, which suggested that amino acid 287 in the C-terminus of LBP may play an important role in LBP binding with CD14.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1. Establishing the LBP mutant phase library.

A. The first and second lanes are the PCR products obtained after annealing at 59°C. The third and fourth lanes are the PCR products obtained after annealing at 57°C and these were collected. B. Restriction enzyme digestion after ligation of the PCR products with PTA2. The target band appeared in the 777-bp fragment (lane 1).

Figure 2. Chemiluminescence assay for phage binding activity with CD14.

A. The luminescent particles were single clones with binding activity (3 independent experiments done in duplicate). A1–3, and B1–3 to H16–18 (3 independent experiments done in duplicate) were for samples 1 to 48, respectively. B. Chemiluminescence plate in (A). Corresponding single clone numbers and binding activity status. “+” indicates binding activity, “−” indicates no binding activity.

Figure 3. Chemiluminescent detection of competitive inhibition of LBP binding to CD14 by phage clones.

A. Non-luminescent particles indicated clones with competitive binding capacity. B. Number of single clones corresponding to the chemiluminescence plate labels and their competitive inhibition status. “+” indicates competition with LBP, “−” indicates no competition.

Figure 4. Effects of polypeptides P1 and MP12 on LPS-induced TNF-α mRNA expression and secretion by U937 cells.

A. Agarose gel electrophoresis results for TNF-α mRNA expression by RT-PCR. B. The optical density ratio for target gene/β-actin was determined. Results are means ± standard deviations. C. Cell culture supernatant was used to detect the concentration of TNF-α(with ELISA) ▴ compared with the control: P compared with P1 treatment: P<0.05; ▵ compared with P1 treatment:P<0.01.

Figure 5. Effects of P1 and MP12 on NF-κB activity in U937 cells.

A. NF-κB binding was significantly increased by LPS treatment. After treatment with P1 or MP12, NF-κB binding decreased; a more significant decrease was found with P1 treatment compared to MP12 treatment. B. IOD values were obtained using TotalLab Quant software. ▴ compared with the control: P compared with P1 group: P<0.05.

Figure 6. Effects of P1 and MP12 on LPS rat arterial blood gas analysis.

PaO2 and PaO2/FiO2 decreased significantly in the LPS group compared to the normal group; with LPS, these were 60 mmHg and >300 mmHg, respectively. The outcome was better in the P1 group compared to the MP12 group. ▴ compared with the control group: P

Figure 7. Pathological evaluations and scoring of rat lung tissues.

I. Rat lung tissue under a light microscope. A. Rat lung tissue had a relatively complete structure in the normal control group. B. Rat lung tissue in rats with LPS-induced injury. C. Rat lung tissue in P1 group. D. Rat lung tissue in MP12 group. Edema: black arrow; congestion: white arrow; inflammatory cell infiltration: gray arrow. II. Pathological scoring for lung tissue:(the scores of control group is zero) ▴compared with the LPS group: P