Inflammation Research

, Volume 59, Issue 8, pp 627–634

A FQHPSFI peptide selectively binds to LPS-activated alveolar macrophages and inhibits LPS-induced MIP-2 production

Authors

    • Department of AnesthesiologyGuangzhou First Municipal People’s Hospital, Guangzhou Medical College
  • Hui Xiao
    • Department of Out-PatientGuangzhou First Municipal People’s Hospital, Guangzhou Medical College
  • Fang Wang
    • Department of MedicineShandong Binzhou Vocational College
  • Lixin Xu
    • Department of AnesthesiologyGuangzhou First Municipal People’s Hospital, Guangzhou Medical College
  • Shouzhang She
    • Department of AnesthesiologyGuangzhou First Municipal People’s Hospital, Guangzhou Medical College
Original Research Paper

DOI: 10.1007/s00011-010-0175-7

Cite this article as:
Ding, N., Xiao, H., Wang, F. et al. Inflamm. Res. (2010) 59: 627. doi:10.1007/s00011-010-0175-7

Abstract

Objective

The goal of this study was to identify peptides selectively binding to lipopolysaccharide (LPS)-activated alveolar macrophages (AMs) and to characterize their effects on the production of LPS-induced cytokines.

Methods

A phage display library was sequentially screened by binding phages to unmanipulated AMs and then to LPS-activated AMs. Individual phage clones were identified by cell-based ELISA. Positive phage clones were characterized by DNA sequencing and bioinformatics analysis. Binding specificity of the selected phage to LPS-activated AMs was tested using immunofluorescent staining. The selected candidate peptide was chemically synthesized to determine whether it could modulate LPS-induced cytokine production in AMs.

Results

Twenty-two out of 40 phage clones selected randomly after four rounds of biopanning bound selectively to LPS-activated AMs, and 12 of them displayed novel peptides. A phage clone displaying FQHPSFI peptide bound effectively to LPS-activated AMs, but not to other cells tested. Furthermore, the synthetic FQHPSFI peptide, but not seven point mutants tested, competitively inhibited the binding of the phage clone to LPS-activated AMs. Importantly, the FQHPSFI peptide significantly inhibited LPS-stimulated microphage inflammatory protein 2 (MIP-2) production in vitro.

Conclusions

Our data demonstrate that phage display technology is a powerful tool for the identification of bioactive peptides. The identified FQHPSFI peptide may be used for the modulation of LPS-stimulated MIP-2 production in AMs.

Keywords

Alveolar macrophagesPeptidePhage displayLipopolysaccharideInflammation

Introduction

Infection with Gram-negative bacteria in the respiratory system can lead to the development of sepsis, septic shock and associated organ failure [1]. Infection in the lung can result in severe inflammation, which may initiate the process of acute lung injury (ALI) and respiratory distress syndrome (ARDS) [2]. Currently, treatment of severe inflammation and associated ALI/ARDS is still challenging. Therefore, the discovery and development of new reagents for the modulation of severe inflammation and associated ALI/ARDS will be of great significance.

Lipopolysaccharide (LPS) is a potent inflammatory mediator and can induce the production of pro-inflammatory cytokines and chemokines by activating transcription factors [3]. Alveolar macrophages (AMs) are well known for their role in the inflammatory process and LPS-activated AMs have been thought to be crucial for the pathogenesis of ALI/ARDS. LPS-activated AMs can secrete many cytokines and chemokines, one of which is microphage inflammatory protein 2 (MIP-2) that can recruit neutrophils into the inflammatory area [4, 5]. Down-regulation of MIP-2 production should mitigate activated AM-related inflammation and associated ALI/ARDS.

Alterations in the expression of proteins in some cells are associated with human diseases [6]. The altered expression of proteins, particularly for some protein receptors, can be used for disease diagnosis and may serve as the targets for the development of therapies for the disease [7]. Natural and artificial ligands that bind preferentially to receptors on accessible cells may have therapeutic potential by blocking the biological effects of these receptors [8]. Similarly, neutralization of natural ligands for these receptors may also modulate disease progression. For example, treatment with VEGF-specific antibody has been demonstrated to inhibit the growth and metastasis of tumors in animal models and at the clinic [9]. A recent study showed that many new surface molecules/receptors were expressed by LPS-activated macrophages and might be used as the targets for the identification of new peptide ligands [10]. However, little is known about what kinds of ligands can selectively bind to LPS-activated AMs.

The phage display technique has been successfully used in the identification of protein–protein interactions [11] and specific inhibitors/modulators [12]. The whole-cell phage display has been used for the identification of phage-displayed peptides that bind to previously unknown targets, inhibiting inflammation [13, 14]. In the present study, we used a sequential screening strategy of whole-cell phage display to isolate peptides that bound selectively to LPS-activated AMs. We found that one of the newly identified peptides (sequence FQHPSFI) inhibited the production of MIP-2 in LPS-activated AMs.

Materials and methods

Animals and reagents

Male Sprague-Dawley (SD) rats at 8–10 weeks of age and weighing 200–250 g were obtained from the Experimental Animal Centre of Southern Medical University (Guangzhou, China). The Ph.D.-C7C library and E. coli host strain ER2738 were from New England Biolabs (Beverly, MA). N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES) was from Gibco BRL (Grand Island, NY). RPMI 1640 and fetal bovine serum (FBS) were from HyClone (Logan, UT). Horseradish peroxidase (HRP)-conjugated anti-M13 antibody (anti-M13/HRP) and anti-M13 antibody were from Amersham Pharmacia Biotech (Uppsala, Sweden). Alexa Fluor 488 conjugated antibody was from Molecular Probes (Eugene, OR).

Isolation and culture of AMs

AMs were isolated from rats, as described previously [14]. The cells were re-suspended in RPMI 1640. Isolated AMs with a purity of >94% and viability of 96% were used for our experiments. All experimental protocols were approved by the Animal Care and Use Committee of Guangzhou Medical College, in accordance with the National Guidelines for the Animal Care and Use.

Stimulation of AMs

AMs were cultured in RPMI 1640 medium at 37°C for 2 h and the unattached cells were removed. The cells were stimulated with LPS (100 ng/ml) in RPMI 1640 media for 24 h at 37°C in a 5% CO2 atmosphere and used as LPS-activated AMs. Cells cultured under the same conditions in the absence of LPS stimulation were used as the unmanipulated AMs (uAMs) for prescreening or controls.

Screening of peptides selectively binding to AMs

A strategy of sequential screening was applied by firstly binding the phage library to uAMs and then binding the flow-through phages to LPS-activated AMs, as described previously [15]. The procedure was repeated for three additional times using the phage obtained from the previous round. After biopanning for 4 rounds, 40 phage clones were randomly selected for further analysis.

Cell-based ELISA

The binding properties of selected individual phage clones to different types of cells were determined by cell-based ELISA [16]. Briefly, the LPS-activated AMs and uAMs were cultured in RPMI 1640 media in 96-wells plates for 2 h. After gentle washing with DPBS, the cells were treated with 0.3% hydrogen peroxide in DPBS for 10 min and blocked with 1% BSA in DPBS (PBSB). Subsequently, the cells were incubated in triplicate with individual phage clones in PBSB for 2 h. After washing, the bound phages were probed with anti-M13/HRP (1:5,000) and detected with TMB. The reaction absorbency (A) was read at 450 nm on a microplate reader (Bio-Tec Instruments, Winooski, VT).

DNA sequencing analysis

Individual phage clones were amplified in E. coli ER2738 and the amplified phages were collected by treatment with PEG 8000/NaCl, followed by centrifuging at 10,000g for 10 min. The phage pellet was suspended in iodide buffer. The single-stranded phage DNA was extracted by ethanol and further purified, followed by DNA sequencing using the −96 pIII sequencing primer (5′-CCCTCATAGTTAGCGTAACG-3′) on an ABI 3100 automated DNA sequencer (Applied Biosystems, Foster City, CA). The DNA sequences were aligned with known proteins using the BLAST program.

Phage binding assay

The binding of selected candidate phages to LPS-activated AMs was evaluated by the phage-binding assay. The mixture of selected phages and wild-type M13 phages (at a ratio of 1:1) was incubated with LPS-activated AMs for 2 h on ice, followed by centrifuging at 500g for 10 min. The phages bound to LPS-activated AMs were rescued by infecting E. coli hosts using X-gal for selection, because the selected phages contained the lacZ gene, and titrated by counting the blue plaques. The binding ability of the selected phages was compared to that of wild-type M13 phages alone that formed white plaques.

Specific binding assay

The binding specificity of the selected phage clone 15 to LPS-activated AMs was tested using immunofluorescent staining. The LPS-activated AMs, uAMs, LPS-treated and untreated SMMC 7721 cells, HEK 293 cells and NIH3T3 cells were cultured on Petri dishes (1 × 104 cells/dish, in duplicate) overnight and washed, followed by blocking with 3% BSA. The cells were then incubated with the phage clone 15 for 2 h on ice. After washing, the cells were fixed with 4% paraformaldehyde in DPBS and permeabilized with 0.1% Triton X-100 in DPBS. The bound phages were probed with anti-M13 antibody at 4°C overnight and visualized using Alexa Fluor 488-conjugated antibody under a Leica DMRA2 microscope (Leica Microsystems AB, Sollentuna, Sweden).

Peptide synthesis

The peptide FQHPSFI displayed by the phage clone 15 and its point mutants (VQHPSFI, FTHPSFI, FQRPSFI, FQHMSFI, FQHPGFI, FQHPSWI and FQHPSFA) were chemically synthesized by Shanghai Bioengineering on a Symphony/Multiplex automated peptide synthesizer (Rainin Instrument Co., Woburn, MA) and characterized by high performance liquid chromatography and mass spectrometry. Peptides with a purity of >95% were used for our experiments.

Competitive binding assay

The selected phage clone 15 (1 × 1010 TU/ml) was mixed with various concentrations (0.01–100 μg/ml) of individual peptides of FQHPSFI, VQHPSFI, FTHPSFI, FQRPSFI, FQHMSFI, FQHPGFI, FQHPSWI or FQHPSFA, respectively. The mixtures were incubated simultaneously with LPS-activated AMs for 2 h on ice. The phages bound to LPS-activated AMs were quantified by the cell-based ELISA.

Bioactivity of FQHPSFI peptide

The LPS-activated AMs and uAMs were in triplicate treated with different concentrations (0, 1, 10 and 100 μg/ml) of individual peptides (FQHPSFI, VQHPSFI, FTHPSFI, FQRPSFI, FQHMSFI, FQHPGFI, FQHPSWI and FQHPSFA), respectively, in RPMI 1640 medium at 37°C for 24 h. Their supernatants were harvested and the concentrations of tumor necrosis factor α (TNF-α), interleukin-1β (IL-1β), IL-6 and MIP-2 in the supernatants were determined by the LiquiChip system (Qiagen, Hilden, Germany), as described previously [17].

Statistical analysis

Data are expressed as mean ± SD and the difference between experimental and control groups was statistically analyzed by ANOVA. A value of P < 0.05 was considered to be statistically significant.

Results

Screening of phages binding to LPS-activated AMs

Following four rounds of screening enrichment, the phages selectively binding to LPS-activated AMs were enriched, evidenced by 1,328-fold increases in the recovery rate of phages (the rate of LPS-activated AMs bound phages in the loaded phages). The binding selectivity of phages for LPS-activated AMs (the ratio of LPS-activated AMs bound phages to uAMs bound phages) also increased dramatically. The number of phages bound selectively to LPS-activated AMs was 220-fold greater than that of uAMs.

Assessment of the binding of selected phage clones to LPS-activated AMs

Forty phage clones were randomly selected and their abilities to bind to LPS-activated AMs and uAMs were measured, as shown in Fig. 1. According to a determination of threefold higher binding to LPS-activated AMs than to uAMs, 22 out of 40 phage clones selectively bound to LPS-activated AMs. Analysis of phage clones 4, 15, 17 and 31 revealed that those clones of phages had 10.7, 20.8, 6.8 and 11.8-fold greater yields than wild-type M13 phage, respectively. Therefore, those clones of phages selectively bound to LPS-activated AMs.
https://static-content.springer.com/image/art%3A10.1007%2Fs00011-010-0175-7/MediaObjects/11_2010_175_Fig1_HTML.gif
Fig. 1

Selective binding of the phage clones to AMs. The LPS-activated AMs and uAMs were incubated with individual phage clones for biopanning, respectively. After washing, the bound phages were probed with HRP-conjugated anti-M13 monoclonal antibody and TMB substrate, followed by reading at A450. Binding of 40 randomly selected phage clones to LPS-activated AMs (filled bars) and uAMs (open bars) were analyzed in an automated ELISA plate reader. Data are expressed as means of three independent experiments

Analysis of the peptide displayed

Analysis of the peptides displayed by the 22-phage clones indicated that 12 peptides with variable sequences were identified, as shown in Table 1. Interestingly, 7 out of 22 phage clones displayed the peptide of FQHPSFI, representing the most frequent peptide. Three, 2 and 2 out of 22 clones presented the peptide SGYPRHY, RQSHLKV and KYNTHHA, respectively. The remaining clones displayed their individual specific peptides (Table 1). This suggested that specific phage clones were selected during the biopanning process. Notably, these peptide sequences had little homology among them and no homology with any of the known proteins that bind to LPS-activated AMs, determined by multiple alignment analyses using the CLUSTAL W program.
Table 1

The peptide sequences displayed by selected phage clones

Peptide

Sequence

Frequency

Phage clone

Peptide 1

FQHPSFI

7

4, 8, 15, 26, 28, 35, 39

Peptide 2

SGYPRHY

3

17, 23, 37

Peptide 3

RQSHLKV

2

10, 31

Peptide 4

KYNTHHA

2

20, 22

Peptide 5

TGSPELH

1

38

Peptide 6

ILGRIFS

1

12

Peptide 7

NQHTALS

1

30

Peptide 8

QMPEMKA

1

1

Peptide 9

WPNLTKE

1

18

Peptide 10

TVSTSIR

1

33

Peptide 11

PASKTLT

1

34

Peptide 12

LPSSGAA

1

21

Specific binding assay

To investigate the binding specificity of the selected phages to LPS-activated AMs, the phage clone 15 was further characterized for its binding to different types of cells by immunofluorescent assay, as shown in Fig. 2. Clearly, the phage clone 15 bound only to LPS-activated AMs (Fig. 2l), but not to uAMs (Fig. 2p). Importantly, this clone of phages bound neither to unmanipulated SMMC 7721 cells, HEK 293 cells and NIH3T3 cells, nor to LPS-treated SMMC 7721, HEK 293 and NIH3T3 cells in vitro. These data indicated that the phage clone 15 selectively bound to LPS-activated AMs.
https://static-content.springer.com/image/art%3A10.1007%2Fs00011-010-0175-7/MediaObjects/11_2010_175_Fig2_HTML.jpg
Fig. 2

The specific binding of the phage clone 15 to LPS-activated AMs. The clone 15 of phages was incubated with uAMs, LPS-activated AMs, LPS-treated and untreated SMMC 7721, HEK 293 and NIH3T3 cells. After washing, the bound phages were characterized by anti-M13 and Alexa Fluor 488-goat anti-mouse IgG antibody. The cells were examined under a light microscope (a, c, e, g, i, k, m, or o) or a Leica DMRA2 fluorescent microscope (b, d, f, h, j, l, n, or p). Data shown are representative images (×400 magnificent). a, b LPS-treated SMMC 7721 cells; c, d LPS-treated HEK 293 cells; e, f LPS-untreated SMMC 7721 cells; g, h LPS-untreated HEK 293 cells; i, j LPS-treated NIH3T3 cells; k, l LPS activated AMs; m, n untreated NIH3T3 cells; and o, p uAMs

Competitive binding assay

Next, we tested whether the peptide FQHPSFI and its mutants could compete for the binding of the phage clone 15 that displayed the peptide by competitive binding assay. The phages were incubated with LPS-activated AMs in the presence of indicated concentrations of the FQHPSFI peptide or individual mutants of VQHPSFI, FTHPSFI, FQRPSFI, FQHMSFI, FQHPGFI, FQHPSWI and FQHPSFA, respectively. The bound phages were analyzed by cell-based ELISA. As shown in Fig. 3, the binding of phage clone 15 to LPS-activated AMs was dramatically inhibited by the FQHPSFI peptide, but not by its point mutants. The inhibitory effects of different concentrations of the FQHPSFI peptide on the binding of phage clone 15 to LPS-activated AMs appeared to be concentration-dependent.
https://static-content.springer.com/image/art%3A10.1007%2Fs00011-010-0175-7/MediaObjects/11_2010_175_Fig3_HTML.gif
Fig. 3

The FQHPSFI peptide competes with the binding of the phage clone 15 to LPS-activated AMs. The LPS-activated AMs were incubated with the phage clone 15 in 96-well plates in the presence of indicated concentrations of the FQHPSFI peptide or individual mutant peptides of VQHPSFI, FTHPSFI, FQRPSFI, FQHMSFI, FQHPGFI, FQHPSWI and FQHPSFA, respectively. After washing, the bound phages were determined by HRP-conjugated anti-M13 monoclonal antibody and TMB substrate, followed by reading at A450. Data are expressed as mean ± SD of the values from three independent experiments. *P < 0.05 compared with untreated control group

Effect of the FQHPSFI peptide on the production of cytokines

To determine whether the FQHPSFI peptide and its mutants could modulate LPS-stimulated cytokine production, LPS-activated AMs were treated with the indicated concentrations of the FQHPSFI peptide or individual mutants of VQHPSFI, FTHPSFI, FQRPSFI, FQHMSFI, FQHPGFI, FQHPSWI and FQHPSFA, respectively, for 24 h and the contents of TNFα, IL-6, IL-1β and MIP-2 in the supernatants were determined, as shown in Fig. 4. Treatment with any of the peptides did not significantly affect the production of TNFα, IL-6, and IL-1β in LPS-activated AMs (data not shown). Furthermore, treatment with the FQHPSFI peptide, but none of the other peptides tested, significantly reduced the levels of MIP-2 in a concentration-dependent manner. Treatment with the FQHPSFI peptide at 10 μg/ml reduced the LPS-stimulated MIP-2 production by 45% and treatment with 100 μg/ml of the peptide further increased the inhibitory effect of the peptide on LPS-stimulated MIP-2 production in AMs. In addition, we tested the impact of different concentrations of the FQHPSFI peptide on the survival of uAMs and LPS-activated AMs and found that treatment with 1–100 μg/ml of the peptide had no effect on the viability of both types of cells in vitro (data not shown). Therefore, the inhibitory effect of the peptide on LPS-stimulated MIP-2 was unlikely to have been caused by the peptide’s cytotoxicity, but rather by modulating the expression and secretion of MIP-2 in LPS-activated AMs.
https://static-content.springer.com/image/art%3A10.1007%2Fs00011-010-0175-7/MediaObjects/11_2010_175_Fig4_HTML.gif
Fig. 4

Effect of the FQHPSFI peptide on the production of MIP-2 by LPS-activated AMs. The LPS-activated AMs were treated with the indicated concentrations of the FQHPSFI peptide or individual mutant peptides of VQHPSFI, FTHPSFI, FQRPSFI, FQHMSFI, FQHPGFI, FQHPSWI and FQHPSFA, respectively, for 24 h. The levels of MIP-2 in the supernatants were determined by the LiquiChip system. Data are shown as mean ± SD of MIP-2 concentrations from five independent experiments. *P < 0.05 compared with untreated control group

Discussion

Despite recent advances in medical technology and critical care, the mortality of ALI/ARDS remains high. AMs have been thought to be important mediators for the development of ALI/ARDS. AMs in response to endotoxin produce pro-inflammatory cytokines and chemokines, which lead to inflammatory cascade [4, 5]. Therefore, the development of new therapeutic reagents targeting activated AMs may benefit patients with ALI/ARDS.

Previous study has shown that the LPS-activated macrophages express many new molecules, such as scavenger receptor class A [18]. We employed a screening strategy by sequentially binding the phage-displayed random peptide library to uAM and LPS-activated AMs for the identification of new peptides that selectively bind to LPS-activated AMs. After four rounds of subtractive biopanning, 22 out of 40 phage clones selected randomly were found to bind selectively to LPS-activated AMs. Of the 22 clones, 7 displayed the peptide FQHPSFI, 3 displayed SGYPRHY, 2 displayed RQSHLKV and 2 displayed KYNTHHA. These clones bound to LPS-activated AMs with 6–20-fold higher activity than wild-type M13 phage. These data suggest that these phage clones that selectively bound to LPS-activated AMs can be enriched and selected through the biopanning process. Further analysis of the phage clone 15 revealed that it selectively bound to the surface membrane of LPS-activated AMs, but not to uAMs, nor to LPS-treated and untreated SMMC 7721 cells, HEK 293 cells and NIH3T3 cells. Importantly, the FQHPSFI peptide displayed by the phage clone 15, but not its point mutants or the control peptide, inhibited the binding of the phage clone 15 to LPS-activated AMs and its inhibitory effects appeared to be concentration-dependent. These suggested that the peptides displayed by the phage clones were responsible for the binding of phage clones to LPS-activated AMs. Notably, natural or artificial ligands binding to the surface receptors preferentially expressed on the cells may modulate the biological functions of the cells. We found that the FQHPSFI peptide presented by the phage clone 15, but not its mutant peptides, significantly inhibited MIP-2 production, although treatment with the peptide did not significantly affect LPS-stimulated TNFα, IL-1β and IL-6 production by LPS-activated AMs in vitro. Selective inhibition of the FQHPSFI peptide on MIP-2 production by LPS-activated AMs may be attributed to the peptide binding to a protein or receptor on LPS-activated AMs, which triggers downstream signaling that specifically inhibits MIP-2 expression. However, the precise mechanisms underlying the action of the FQHPSFI peptide in the LPS-induced MIP-2 expression in AMs remain to be further determined. Importantly, MIP-2 is a crucial chemokine for attracting neutrophils, which are important in the pathogenesis of bacterial infection-related inflammatory diseases [5]. Accordingly, the FQHPSFI peptide may be used for the modulation of MIP-2 production and activated AMs-related inflammation, such as ALI and ARDS. We are interested in further examining the bioactivity of this peptide and its potential therapeutic effect on inhibition of human inflammatory diseases.

The binding ability of a peptide displayed by the phage clone usually is dependent on its homology with the natural ligand. Indeed, a specific hepatopeptide from a phage display peptide library has been identified to bind selectively to primitive hematopoietic stem cells [19]. Sequence analysis revealed that the FQHPSFI peptide had no homology with the proteins of CD14, LPS binding protein (LBP), and heat shock protein 70 (HSP 70), which bind to the surface membrane of LPS-activated macrophages [20]. Therefore, it is possible that this peptide may represent a novel ligand for LPS-activated AMs. Our further studies will center on determining to which surface protein the peptide binds. Given that 12 novel peptides were found to selectively bind to LPS-activated AMs, we are interested in further investigating whether those phage clones can compete each other for binding to LPS-activated AMs and determining their potential targets on cellular membranes as well as the biological functions of these novel peptides.

Phage display technology has been widely used for the identification of receptor-specific ligands [21]. Several groups have used the whole cell-based phage display technique to identify specific peptides targeting unknown receptors [22, 23]. However, the whole cell-based phage display technique for screening phage libraries has some serious disadvantages. For example, the non-specific binding of the phages to the cells usually results in high background and low yield of phage recovery. The repeated washing for reducing the background is labor-intensive and inefficient due to the potential loss of the cells and peptide [15]. To address these technical issues, we used the biopanning and rapid analysis of selective interactive ligands (BRASIL) method and successfully identified the peptides binding selectively to LPS-activated AMs. As the BRASIL requires only a single step of centrifugation, this method is simpler and more convenient than other cell panning techniques [15].

In conclusion, we employed a screening strategy by sequentially binding the peptide display phage library to uAMs and then to LPS-activated AMs, and successfully identified the peptides binding selectively to LPS-activated AMs. Further analysis revealed that the novel FQHPSFI peptide, but not its point mutants, greatly inhibited the production of MIP-2 by LPS-activated AMs in vitro. Our data indicate that the phage display technology is a powerful tool for the identification of bioactive peptides. Our findings may provide a basis for the development of new peptide-based therapies for treatment of inflammatory diseases.

Acknowledgments

This study was supported by grants from the Medical Scientific Research Foundation of Guangdong Province (A2009497), and the Medical Science and Technology Program of Guangzhou (2008-YB-012, 2009-YB-021, 2009-ZDi-02, and 2008-ZDi-14).

Copyright information

© Springer Basel AG 2010