Journal of Biotechnology 187 (2014) 43–50

Contents lists available at ScienceDirect

Journal of Biotechnology journal homepage: www.elsevier.com/locate/jbiotec

Identification of peptides that selectively bind to myoglobin by biopanning of phage displayed-peptide library Guruprasath Padmanaban a,b,c , Hyekyung Park a , Ji Suk Choi d , Yong-Woo Cho d , Woong Chol Kang e , Chan-Il Moon f , In-San Kim g , Byung-Heon Lee a,b,c,∗ a

Department of Biochemistry and Cell Biology, School of Medicine, Kyungpook National University, Daegu, Republic of Korea Cell & Matrix Research Institute, Kyungpook National University, Daegu, Republic of Korea c BK21 Plus KNU Biomedical Convergence Program, Department of Biomedical Science, Kyungpook National University, Daegu, Republic of Korea d Department of Chemical Engineering, Hanyang University, Gyeonggi-do, Republic of Korea e Department of Cardiology, Gil Medical Center, Gachon University of Medicine and Science, Incheon, Republic of Korea f Department of Cardiology, School of Medicine, Hanyang University, Seoul, Republic of Korea g Biomedical Research Institute, KIST, Seoul, Republic of Korea b

a r t i c l e

i n f o

Article history: Received 12 April 2014 Received in revised form 28 June 2014 Accepted 18 July 2014 Available online 29 July 2014 Keywords: Cardiac biomarkers Sandwich assays Myoglobin Peptide Phage display

a b s t r a c t Biopanning of phage displayed-peptide library was performed against myoglobin, a marker for the early assessment of acute myocardial infarction (AMI), to identify peptides that selectively bind to myoglobin. Using myoglobin-conjugated magnetic beads, phages that bound to myoglobin were collected and amplified for the next round of screening. A 148-fold enrichment of phage titer was observed after five rounds of screening relative to the first round. After phage binding ELISA, three phage clones were selected (3R1, 3R7 and 3R10) and the inserted peptides were chemically synthesized. The analysis of binding affinity showed that the 3R7 (CPSTLGASC) peptide had higher binding affinity (Kd = 57 nM) than did the 3R1 (CNLSSSWIC) and 3R10 (CVPRLSAPC) peptide (Kd = 125 nM and 293 nM, respectively). Cross binding activity to other proteins, such as bovine serum albumin, troponin I, and creatine kinase-MB, was minimal. In a peptide-antibody sandwich ELISA, the selected peptides efficiently captured myoglobin. Moreover, the concentrations of myoglobin in serum samples measured by a peptide–peptide sandwich assay were comparable to those measured by a commercial antibody-based kit. These results indicate that the identified peptides can be used for the detection of myoglobin and may be a cost effective alternative to antibodies. © 2014 Elsevier B.V. All rights reserved.

1. Introduction During acute myocardial infarction (AMI), detection and measurement of creatine kinase-MB (CK-MB), cardiac troponin I or T, and myoglobin are commonly used for the diagnosis of AMI. CK-MB and cardiac troponin I are elevated after 3–6 h of onset of symptoms. Myoglobin, a 17.8 kDa heme protein normally present in muscles, is elevated within 1 h to 3 h and peaks within 6–9 h after the onset of AMI. The early increase of serum myoglobin may help the measurement of myoglobin as a major diagnostic utility for the early detection of AMI (de Winter et al., 2000; Moe and Wong, 2010;

∗ Corresponding author at: Department of Biochemistry and Cell Biology, School of Medicine, Kyungpook National University, 680 Gukchaebosang-ro, Jung-gu 700-842 Daegu, Republic of Korea. Tel.: +82 53 420 4824. E-mail address: [email protected] (B.-H. Lee). http://dx.doi.org/10.1016/j.jbiotec.2014.07.435 0168-1656/© 2014 Elsevier B.V. All rights reserved.

Sallach et al., 2004). As per National academy of clinical biochemistry (NACB) laboratory medicine practice guidelines, clinicians test for myoglobin in conjunction with cardiac troponin during AMI, and the combined use of myoglobin with cardiac troponin or CK-MB is helpful in early diagnosis of myocardial infarction (Morrow et al., 2007). Currently antibody-based immunoassays are a widely-used tool for diagnosing various biomolecules (Borrebaeck, 2000). This is due to its high affinity and selectivity against the antigen. However, antibody-based immunoassays seem to have disadvantages. For example, polyclonal antibodies have cross reactivity to antigens if the animal has exposed to various antigens in the past, while monoclonal antibodies are highly specific but sensitive to different environmental conditions (Petrenko and Vodyanoy, 2003; Shone et al., 1985). Other drawbacks of antibodies are high production cost and less chemical stability (Ruigrok et al., 2011). Nevertheless, antibody-based immunoassays are still commonly used for diagnosing cardiac biomarkers during AMI.

44

G. Padmanaban et al. / Journal of Biotechnology 187 (2014) 43–50

As an alternative approach, synthetic peptides may have advantages over antibody-based diagnostics, since it is easy to produce, cost less, and is readily modified chemically (Ladner et al., 2004; Meloen et al., 2003). Using the diversity of phage library, varieties of peptides have been produced in the past two decades. The selection process using a phage peptide library or biopanning is simple and cost effective. Based on an affinity selection, phages bound to the immobilized target were enriched and amplified. After several rounds of biopanning, the selected phage clones were picked and analyzed individually (Paschke, 2006; Willats, 2002). Using phage displayed-peptide library, we have identified a peptide that binds to the interleukin-4 receptor that is exposed on atherosclerotic plaques and cancer cells and employed it for imaging and drug delivery (Hong et al., 2008; Namgung et al., 2014). Also, we have identified a peptide that binds to histone H1 that is exposed on the surface of apoptotic cells and employed it for in vivo imaging of apoptosis and drug delivery (Wang et al., 2011, 2010). The aim of this study was to identify peptides that selectively bind to myoglobin and develop a peptide-based diagnostic assays for the detection of serum myoglobin. For this, we screened phage-displayed-peptide library to identify peptides that bind to purified myoglobin. Rather coating the target protein onto micro titer plates, we conjugated myoglobin to magnetic beads to perform the screening process three-dimensionally and also to employ the magnetic separation to remove the unbound phages. The binding of selective clones was confirmed by phage binding ELISA and their selectivity or binding affinity of the positive clones were confirmed by ELISA using synthetic peptides. Further, peptide-based sandwich assay was developed using clinical serum samples to validate the measurement of myoglobin by the selected peptides. 2. Materials and methods 2.1. Materials Ph.D-C7C phage peptide library was purchased from New England Biolabs (#E8110SC; New England Biolabs, Beverly, MA, USA). This contains a structurally constrained 7-mer random peptide library with complexity of 1.2 × 109 and E. coli ER 2738 as a host cell. Surface activated Dynabeads M-270 Epoxy was purchased from Life Technologies (Carlsbad, California, USA). Human myoglobin was purchased from Abcam (#ab96036; Cambridge, Massachusetts, USA). Anti-myoglobin monoclonal antibody (#sc65982) and goat anti-mouse IgG-HRP (#sc-2005) were purchased from Santa Cruz Biotechnology (Dallas, Texas, USA). Myoglobin ELISA kit was purchased from DRG International Inc. (#EIA 3955, Springfield, New Jersey, USA). ELISA plates were purchased from Corning (#3590; New York, USA), black-colored ELISA plates for fluorescent assay were from SPL life sciences Korea (Gyeonggi-do, Korea), bovine serum albumin (BSA) was from Bovogen (Bovostar #BSA100; East Keilor, Australia), gelatin was from Sigma (St. Louis, Missouri, USA), horse radish peroxidase (HRP)-conjugated antiM13 antibody was from GE Healthcare (#45-001-419, New Jersey, USA), and 3,3 ,5,5 -tetramethylbenzidine (TMB) substrate was from Komabiotech (#K0331070; Seoul, Korea). Absorbance was measured using the Sunrise Basic microplate reader (Tecan group Ltd, Männedorf, Switzerland). 2.2. Overall experimental setup The overall experimental setup is shown in Fig. 1. This consists of five parts: (1) biopanning of phage library, (2) Sequencing and amino acid sequence analysis, (3) evaluation of phage binding, (4) evaluation of peptide binding, and (5) development of peptidebased sandwich assays.

Fig. 1. A flow chart of experiments.

2.3. Biopanning of phage library Myoglobin was conjugated to surface activated Dynabeads M270 Epoxy for biopanning. After Dynabeads were resuspended and washed as per the manufacturer’s instruction, 60 ␮g of myoglobin was immobilized to 105 ␮l of beads (2 × 108 ), and 60 ␮l of 3 M ammonium sulfate is added to make a final volume of 180 ␮l. The mixture was incubated at 4 ◦ C for 16–24 h in Eppendorf tube rotator. After incubation, the myoglobin coated beads were collected using magnetic separator and the supernatant was removed and the coated beads were washed for 4 times with 1 ml of 0.1% PBST. Then the coated beads were blocked for 30 min at room temperature (RT) using 0.5% BSA in phosphate-buffered saline (PBS). Each biopanning round consists of negative selection or subtraction of phages that nonspecifically binds to the Dynabeads and subsequent positive selection of phages that binds to myoglobin and amplification of the eluted phages. In the first round, 5 ␮l of 1 × 1011 M13 phage library in PBS was added to the 0.5% BSA-blocked Dynabeads and incubated at 4 ◦ C for 1 h with gentle shaking for subtracting the phages that binds to the beads. After incubation, unbound phages in the supernatant were collected using magnetic separator and were incubated with myoglobin coated on the beads at 4 ◦ C for 1 h with gentle shaking. After incubation, the unbound phages were extensively washed using PBS containing Tween (PBST) and the concentration of Tween was gradually increased from 0.1% up to 0.5% in further rounds, in order to minimize the non-specific binding of phages. The bound phages were eluted by incubating with 0.1 M citrate (pH 3.1) at RT for 2 min and immediately neutralized with 1 M TrisHCl (pH 9.1). The phage particles in the elute were precipitated using 20% polyethyleneglycol (PEG)/2.5 M NaCl and then suspended in Tris-buffered saline (TBS). The phage titration was performed by serially diluting the elute and plated on LB media containing isopropyl-␤-D-thiogalactoside and 5-bromo-4-chloro3-indolyl-␤-D-galactoside. The remained phages were amplified

G. Padmanaban et al. / Journal of Biotechnology 187 (2014) 43–50

45

Fig. 2. A scheme of the biopanning experiment. Phage library was incubated with myoglobin coupled on magnetic beads. Beads were separated with using a magnet and phages bound to beads were eluted and amplified for next round of biopanning.

using ER2738 E. coli and used for next round of screening. The same amount of input (1 × 1011 pfu of phages) was maintained in the subsequent rounds, and also the subtraction step was included (Fig. 2). 2.4. DNA and amino acid sequence analysis After five rounds of biopanning, clones with inserts were confirmed by polymerase chain reaction. The DNA inserts of forty seven individual clones were sequenced at Macrogen (Seoul, Korea). Sequence similarity was identified and aligned using the Clustal W program available on-line. 2.5. Phage binding ELISA assays ELISA plates were coated with 100 ␮l of myoglobin (2 ␮g/ml) and BSA as a negative control in 0.1 M NaHCO3 pH 9.6 at 4 ◦ C overnight. The plates were blocked with 5 mg/ml of BSA in 0.1 M NaHCO3 pH 9.6 at 4 ◦ C for 2 h. After washing the wells for 6 times with TBS containing 0.1% Tween (TBST), 100 ␮l of phage clones (1 × 1010 pfu/well) were added in 0.1% TBST and incubated at RT for 1 h. After washing, HRP-conjugated anti-M13-antibody (diluted in blocking buffer at 1:10,000 ratio) was added and incubated at RT for 1 h. After the plates were washed for 6 times, TMB substrate was added and incubated at RT for 5–15 min. Finally the reaction was stopped using 2 M H2 SO4 and plates were read at 450 nm using a microplate reader.

2.7. Cross binding and saturation binding assays To examine cross binding of peptides to control proteins, such as troponin I, CK-MB, and BSA, each microplate well was coated with 100 ␮l of each protein (2.5 ␮g/ml) in 0.1 M NaHCO3 pH 9.6 at 4 ◦ C overnight. The plates were blocked with TBS containing 0.5% gelatin at 37 ◦ C for 2 h. After washing the plates for 6 times with 0.1% TBST, 100 ␮l of biotinylated peptide (5 ␮g/ml) in 0.1% TBST containing 0.5% gelatin was added and incubated at 37 ◦ C for 2 h. After washing, 1:10,000 dilution of streptavidin-HRP conjugate (SNN 2004; Invitrogen, Carlsbad, California, USA) in 0.1% TBST containing 0.5% gelatin was added and incubated at 37 ◦ C for 1 h. The plates were washed for 6 times with 0.1% TBST, and TMB substrate was added and incubated at RT for 5–15 min. Finally the plates were read at 450 nm using a microplate reader. To measure the affinity of peptide binding to myoglobin, saturation binding assays were performed by incubating various concentrations of biotinylated peptides in 0.1% TBST containing 0.5% gelatin to myoglobin-coated plates at 37 ◦ C for 2 h. The platebound biotinylated peptide was detected by streptavidin-HRP. Based on the HRP activity, total binding of each peptide to myoglobin was quantified. The Specific binding of each peptide to myoglobin was calculated by subtracting the values of non-specific binding of each peptide to BSA-coated wells from the total binding of each peptide to myoglobin. Binding assays were performed at least for three times in triplicate, yielding similar results. The affinity of each peptide to myoglobin was calculated by non-linear regression and transformed to Scatchard plot using GraphPad Prism 6 program.

2.6. Peptide synthesis 2.8. Peptide-antibody sandwich ELISA Peptides were synthesized by standard Fmoc method and conjugated with biotin at C-Terminal and purified by mass spectrometry (Peptron, Daegeon, Korea). Each peptide was cyclized using cysteine at C- and N-terminal of CX7 C sequence (disulfide bonding). Using GGGS spacer and lysine as a linker, biotin was conjugated to the amidated C-terminal of CX7 C sequence, and alanine was added to the N-terminal of CX7 C sequence, resulting in a biotinylated peptide sequence of ACX7 CGGGSK-biotin.

To see the capturing efficiency of peptides, wells were coated with peptides and then myoglobin bound to peptides were detected by anti-myoglobin antibody. For this, wells were coated with 10 ␮g/ml of streptavidin in 0.1 M NaHCO3 pH 9.6 at 4 ◦ C overnight. The plates were then blocked with TBS containing 0.5% gelatin at 37 ◦ C for 2 h. After washing, 100 ␮l of biotinylated peptide (10 ␮g/ml) in 0.1% TBST containing 0.5% gelatin was added and

46

G. Padmanaban et al. / Journal of Biotechnology 187 (2014) 43–50

incubated for 2 h at 37 ◦ C. After washing, myoglobin was added to wells at various concentrations in dilution buffer and incubated at RT for 2 h. The plates were incubated with anti-myoglobin antibody (1 ␮g/ml) in 0.1% TBST containing 0.5% gelatin at RT for 2 h. The wells were incubated with HRP-conjugated goat anti-mouse IgG (1:2000) and the absorbance was measured using a microplate reader.

Table 1 The ratio of output phage titer over input phage titer throughout biopanning rounds. Round

Input (pfu)

Output (pfu)

Ratio (output/input)

1st Round 2nd Round 3rd Round 4th Round 5th Round

1 × 1011 1 × 1011 1 × 1011 1 × 1011 1 × 1011

1.93 × 105 5.63 × 104 4.01 × 105 6.98 × 106 2.85 × 107

1.93 × 10−6 5.63 × 10−7 4.01 × 10−6 6.98 × 10−5 2.85 × 10−4

2.9. Serum sample collection

2.10. Peptide–peptide sandwich assay In order to determine the concentrations of serum myoglobin using a peptide–peptide sandwich assay, serum samples were 10 fold diluted with 0.1% TBST buffer containing 0.5% gelatin. A standard curve was plotted using myoglobin diluted in normal human serum. The peptide–peptide sandwich assay was performed by coupling a biotinylated, primary peptide (100 ␮g/ml) diluted in 0.1% TBST buffer containing 0.5% gelatin to streptavidincoated ELISA plates and then by incubating the plates with the serum samples at RT for 2 h. The wells were washed and incubated with a fluoroisothiacyanate (FITC)-labeled, secondary peptide (100 ␮g/ml) diluted in 0.1% TBST buffer containing 0.5% gelatin for 2 h at RT. The fluorescence intensity was measured using a fluorescence microplate reader. The peptide–peptide based sandwich assays were performed in triplicate at least three times yielding similar results. In comparison, an antibody-based, commercial myoglobin ELISA was performed as per the manufacturer’s instruction (DRG International Inc.).

148.1

300 250

phage titer ( x 105 pfu)

Serum samples from 12 patients with a confirmed diagnosis of myocardial infarction were collected from Gachon University Hospital (IRB No. GAIRB2013-145). Serum samples from 12 healthy volunteers were collected in the lab. All samples were stored at −80 ◦ C until use. These collected serum samples were used to perform sandwich assays for the detection of myoglobin using a peptide-based sandwich assay or antibody-based commercial kit.

200 150 100 36.2 50

1.0

0.3

2.3

0 1R

2R

3R

4R

5R

Panning rounds Fig. 3. Biopanning and enrichment of phage library. Five rounds of biopanning against myoglobin were performed, and the phage titer (pfu) after each round was measured. Numbers on bars are the enrichment fold of the phage titer over the first round. Table 2 Amino acid sequences and frequency of the selected peptides. Phage clone

Amino acid sequence

Frequency

3R1 3R7 3R10 3R3 3R26 3R37

CNLSSSWIC CPSTLGASC CVPRLSAPC CVATLPAGC CHPLRSAFC CSHTSLESC

24/47 (51%) 7/47 (15%) 7/47 (15%) 5/47 (11%) 3/47 (6%) 1/47 (2%)

2.11. Statistical analysis To evaluate the statistical significance of differences among groups, a non-parametric Kruskal-Wallis ANOVA analysis was performed using the GraphPad Prism 6 program and the p-values ≤0.05 considered as statistically significant.

the third round of biopanning, with the frequency of 51%, 15%, 15%, and 11%, respectively.

3. Results

The binding efficiency of these frequently occurring clones (3R1, 3R7, 3R10, and 3R3) to myoglobin was validated by phage-binding ELISA. The 3R1, 3R7, 3R10 clones showed higher binding to myoglobin than the 3R3 clones, while all of the clones showed little

Five rounds of biopanning were performed to screen M13 phage library against myoglobin. The enrichment of phage was monitored by measuring titers of the output after each biopanning round and the fold enrichment relative to the titer of the first round. The phage titer has been increased from the first round (1.93 × 105 pfu) to the fifth round (2.85 × 107 pfu) and, at the end of five rounds of biopanning, the phage titer has been enriched to 148 fold over the first round (Table 1 and Fig. 3). The enrichment of phage titer suggests that the biopanning of phages that selectively bind to myoglobin is successfully achieved. Forty seven phage clones were randomly picked from phage clones of the third, fourth, and fifth round. These individual clones were subjected to polymerase chain reaction and DNA sequencing to know the sequence of the inserted peptides. After sequencing, the amino acid sequences were aligned and the frequency of these sequences were analyzed (Table 2). Phage clones that occurred at relatively high frequency were 3R1, 3R7, 3R10, and 3R3 clones from

0.8 0.7

OD at 450nm

3.1. Biopanning of myoglobin-binding phages

3.2. Validation of phage clones by phage-binding ELISA

0.6 0.5 0.4 BSA Myoglobin

0.3 0.2 0.1 0.0 Library

3R-1

3R-3

3R-7

3R-10

Phage Clones Fig. 4. Binding activity of phage clones to myoglobin. The binding activity of the four selected phage clones to myoglobin coated on plates was determined by phagebinding ELISA. Optical density (OD) was measured at a wavelength of 450 nm. BSA was used as a control protein. Phage library was used as a control phage.

G. Padmanaban et al. / Journal of Biotechnology 187 (2014) 43–50

47

Table 3 The affinity (Kd ) values of selected peptides.

Fig. 5. Diagram of biotin-labeled synthetic peptides. For mimicking a peptide displayed on phage (bottom), a synthetic peptide (top) was cyclized by disulfide bonding and was added with the GGGS as a linking sequence. The alanine residue was added to N-terminal, and lysine residue was inserted at C-terminal for labeling with biotin.

binding to BSA (Fig. 4). These results indicate the selective binding activity of these phage clones to myoglobin.

3.3. Validation of peptide binding to myoglobin The CX7 C peptide that was displayed on phage was constrained between a pair of cysteine residues and formed disulfide linkage during the assembly. Based on the result of phage-binding ELISA, the 3R1 (CNLSSSWIC), 3R7 (CPSTLGASC) and 3R10 (CVPRLSAPC) peptides were selected and synthesized. During synthesis, each peptide was added with alanine at N-terminal, was cyclized using a disulfide-bond between cysteine residues, added with a short GGGS spacer between the displayed peptide and PIII phage coat protein, and added with biotin at C-terminal using lysine as linker in order to mimic peptides as expressed on phage surface (Fig. 5). Various parameters were tested for optimizing peptide-binding ELISA conditions, such as myoglobin coating concentration, pH, temperature, peptide concentration, wash buffers, dilution buffers, and salt concentrations (data not shown). After optimization, binding of peptide to myoglobin and control proteins were examined. The binding of selected peptides to myoglobin was higher than their cross binding activity to other cardiac biomarker, such as troponin I, CK-MB, and BSA used as control proteins (Fig. 6). Among these peptides, the binding of 3R7 peptide to myoglobin was higher than that of 3R1 and 3R10. The binding of 3R10 to myoglobin was lower but was relatively more specific to myoglobin than that of 3R7 peptide (Fig. 6).

3.5 3.0

OD at 450nm

2.5 2.0

3R1 3R7

1.5

3R10

1.0 0.5 0.0

BSA

Troponin I

CK-MB

Myoglobin

Fig. 6. Validation of binding activity of peptides to myoglobin. The binding of 3R1, 3R7, and 3R10 peptides to myoglobin and their cross binding to troponin and CK-MB as control cardiac biomarkers were determined by peptide-binding ELISA. Myoglobin and control proteins were coated onto ELISA plates. Peptides that are bound to the coated proteins were detected by HRP-conjugated streptavidin and the enzyme substrate. Optical density (OD) was measured at a wavelength of 450 nm.

Peptide

Kd (nM)

Standard Error

R2

3R1 3R7 3R10

125 57 293

0.07 0.80 0.14

0.997 0.993 0.990

3.4. Saturation binding assays and measurement of binding affinity A saturation binding assay and Scatchard analysis of binding curve for the 3R1, 3R7 and 3R10 peptides revealed that the binding affinities (Kd ) of peptides to myoglobin coated on plates were in nanomolar ranges (Table 3). The Kd value was determined by non-linear regression analysis and Scatchard transformation using GraphPad Prism 6 program. The binding of these peptides to myoglobin was specific and hyperbolic. The 3R7 peptide showed higher binding affinity (Kd = 57 nM) than that of 3R1 (Kd = 125 nM) and 3R10 (Kd = 293 nM) (Fig. 7). The binding pattern of these peptides to myoglobin measured by saturation binding assays was consistent with the binding pattern measured by the peptide-binding ELISA. 3.5. Detection of myoglobin using peptide-antibody sandwich ELISA To see whether the selected peptides could efficiently capture and detect myoglobin, a peptide-antibody sandwich ELISA was performed. The sandwich assay was performed by coating biotin-labeled peptide to the streptavidin-coated ELISA plate to capture myoglobin and then by detecting the peptide-captured myoglobin with an anti-myoglobin monoclonal antibody. All of the 3R1, 3R7, and 3R10 peptides were able to capture myoglobin in a concentration-dependent manner, but the binding curve obtained by the 3R7 peptide was more linear and more steep in the concentrations of myoglobin (0–1000 ng/ml) compared to that of the 3R1 and 3R10 peptides (Fig. 8). 3.6. Measurements of serum myoglobin using peptide–peptide sandwich assays In order to examine whether the selected peptides could replace antibodies for the detection of myoglobin, peptide–peptide sandwich assays were performed. Different combinations of the selected peptides, as a biotin-labeled, primary peptide to capture myoglobin and a FITC-labeled, secondary peptide to detect the captured myoglobin, were examined to choose the best pair for the peptide-based sandwich assays. After several trials and errors, the pair of 3R10 and 3R7, as the primary and secondary peptide, respectively, showed better results than other combinations (data not shown). A standard calibration curve was plotted using the peptide–peptide sandwich assays, in which myoglobin at a concentration as low as 25 ng/ml could be detected (Fig. 9A). The concentrations of myoglobin from the serum of twelve AMI patients measured by the peptide–peptide sandwich assay were comparable to the corresponding values measured by a commercial antibody-based kit with the median values of 327 and 372 ng/ml, respectively (Table 4 and Fig. 9B). The serum concentrations of myoglobin of normal healthy volunteers were almost undetectable and significantly low compared to those of patients (Fig. 9B, P = 0.0001). 4. Discussion A diverse immunosensing method, including optical density, fluorescence, luminescence, surface plasmon resonance, Raman

48

G. Padmanaban et al. / Journal of Biotechnology 187 (2014) 43–50

Fig. 7. Analysis of binding affinity. The binding affinity of the selected peptides (3R1, 3R7, and 3R10) to myoglobin coated on plates was measured by the saturation binding assays. The Kd value was determined by non-linear regression analysis and Scatchard transformation using GraphPad Prism 6 program. 2.5

Table 4 The concentrations of myoglobin of serum samples measured by a commercial kit and peptide–peptide sandwich assay.

1 2 3 4 5 6 7 8 9 10 11 12

Age

M M M M M M M F M M M M

Sex

65 59 48 38 50 71 43 72 68 78 39 42

Myoglobin concentrations (ng/ml) by commercial kit 463 619 283 522 598 487 97 275 371 173 252 373

Myoglobin concentrations (ng/ml) by peptide–peptide sandwich assay 342 582 301 478 629 375 130 308 394 182 256 311

spectroscopy, and electrochemistry, have been used for the detection of cardiac biomarkers (Mohammed and Desmulliez, 2011). In most of these methods, antibodies have been used for the detection of myoglobin or other cardiac biomarkers. In this study, we have selected peptides that bind to early cardiac biomarker myoglobin and employed them for the detection of myoglobin. Unlike conventional biopanning using simple adsorption of a target protein to the solid surface, here we have conjugated the myoglobin to magnetic beads and performed the biopanning in

2.0

OD at 450nm

Sample No.

1.5 3R1

1.0

3R7 3R10

0.5

0.0

Myoglobin (ng/ml) Fig. 8. Capture of myoglobin by peptides (peptide-antibody sandwich ELISA). Increasing concentrations of myoglobin was incubated with the selected peptides pre-coated on plates and then incubated with anti-myoglobin antibody. Antibody was detected by HRP-conjugated anti-IgG and the enzyme substrate. Optical density (OD) was measured at a wavelength of 450 nm. Data represent mean OD ± standard deviation of assays performed in triplicates.

suspension. The beauty of this strategy is the screening of phage clones in suspension, which gives more room for reaction (Kim et al., 2005). The adsorption of a target to ELISA plates in the conventional method tends to undergo partial denaturation of protein due to non-specific hydrophobic interaction (Adey et al., 1995). Another feature of the biopanning strategy in this study was using a

G. Padmanaban et al. / Journal of Biotechnology 187 (2014) 43–50

49

Fig. 9. Peptide–peptide sandwich assays for the measurement of serum myoglobin. (a) Standard calibration curve of myoglobin obtained by a peptide–peptide sandwich. Myoglobin was diluted to the concentrations of 25, 100, 250, and 500 ng/ml with normal human serum. Diluted myoglobin standards were incubated with the 3R10 peptide pre-coated onto plates and then incubated with FITC-labeled 3R7 peptide, as the primary and secondary peptide, respectively. Dotted lines represent standard deviations of data in triplicates (b) Plot analysis of myoglobin concentrations in serum samples obtained by a peptide–peptide (3R10-3R7) sandwich assay in comparison with a commercial kit. Serum samples of AMI patients were incubated with the 3R10 peptide pre-coated onto plates and then incubated with FITC-labeled 3R7 peptide. Serum myoglobin concentration was also measured by a commercial, antibody-based kit. Normal human serum was used as a control. Horizontal lines represent median values.

disulfide-bonded, cyclic form of peptide library. Such a structurally constrained phage library adds more advantage by increasing the target affinity and decreasing the entropy during ligand-receptor binding during molecular recognition of protein-peptide interaction (Lindner et al., 2011; Uchiyama et al., 2005). Using the disulfide-bonded, cyclic form of phage-displayed-peptide library, we have previously identified a peptide that binds to human IgG with a high binding affinity (Hien et al., 2012). Throughout biopanning rounds, we observed a gradual enrichment of phage titers and several peptide sequences with higher frequency and chose three peptide candidates, including 3R1 (CNLSSSWIC), 3R7 (CPSTLGASC), and 3R10 (CVPRLSAPC), with the affinities in nanomolar ranges and selective binding to myoglobin. Serum myoglobin levels in normal subjects range from 12 to 100 ng/ml and slightly increase by age (Tietz, 1995). It has been known that myoglobin concentration above 200 ng/ml is a cutoff value for diagnosis of AMI, and the combination of myoglobin with troponin provide more accurate diagnosis of AMI (Melanson et al., 2004). Throughout the development of peptide-based sandwich assays, we demonstrated that the combination of 3R10 and 3R7 myoglobin-binding peptides in sandwich assays could efficiently capture and detect serum myoglobin at a concentration of as low as 25 ng/ml. Moreover, the 3R10 and 3R7 peptide-based assays of serum myoglobin of AMI patients resulted in myoglobin concentrations that are similar to the corresponding values obtained by a commercial kit. These results suggest that the sensitivity and specificity of the peptide-based assay were quite comparable to the commercial kit and could be a useful diagnostic tool for AMI and that the selected peptides can be used as an alternative to antibody for the detection of myoglobin. A minor drawback of the peptide–peptide sandwich assay is that the whole procedure takes 2–4 h, which is longer than 1 h by antibody-based commercial kits. The clinical performance of the peptide–peptide sandwich assay remains to be further validated using a large numbers of patient serum samples.

5. Conclusions To our best knowledge, this is the first study to demonstrate the detection of early cardiac biomarker myoglobin using peptides that are identified by phage display. These peptides that selectively bind to biomarkers and detect them (“biomarker-sensing

peptides”) could be cost-effective alternatives of antibodies and a useful diagnostic tool for the detection of other serum biomarkers as well as cardiac myoglobin. Acknowledgements This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (No. 20080061891) and a grant NRF-2012M2A2A7035589. References Adey, N.B., Mataragnon, A.H., Rider, J.E., Carter, J.M., Kay, B.K., 1995. Characterization of phage that bind plastic from phage-displayed random peptide libraries. Gene 156, 27–31. Borrebaeck, C.A.K., 2000. Antibodies in diagnostics—from immunoassays to protein chips. Immunol. Today 21, 379–382. de Winter, R.J., Lijmer, J.G., Koster, R.W., Hoek, F.J., Sanders, G.T., 2000. Diagnostic accuracy of myoglobin concentration for the early diagnosis of acute myocardial infarction. Ann. Emerg. Med. 35, 113–120. Hien, T.B., Maeng, J.H., Lee, B.H., Seong, G.H., Choo, J., Lee, E.K., 2012. Potential application of antibody-mimicking peptides identified by phage display in immuno-magnetic separation of an antigen. J. Biotechnol. 161, 213–220. Hong, H.Y., Lee, H.Y., Kwak, W., Yoo, J., Na, M.H., So, I.S., Kwon, T.H., Park, H.S., Huh, S., Oh, G.T., Kwon, I.C., Kim, I.S., Lee, B.H., 2008. Phage display selection of peptides that home to atherosclerotic plaques: IL-4 receptor as a candidate target in atherosclerosis. J. Cell. Mol. Med. 12, 2003–2014. Kim, Y.G., Lee, C.S., Chung, W.J., Kim, E., Shin, D.S., Rhim, J.H., Lee, Y.S., Kim, B.G., Chung, J.H., 2005. Screening of LPS-specific peptides from a phage display library using epoxy beads. Biochem. Biophys. Res. Commun. 329, 312–317. Ladner, R.C., Sato, A.K., Gorzelany, J., de Souza, M., 2004. Phage display-derived peptides as a therapeutic alternatives to antibodies. Drug Discov. Today 9, 525–529. Lindner, T., Kolmar, H., Haberkorn, U., Mier, W., 2011. DNA libraries for the construction of phage libraries: statistical and structural requirements and synthetic methods molecules. Molecules 16, 1625–1641. Melanson, S.F., Lewandrowski, E.L., Januzzi, J.L., Lewandrowski, K.B., 2004. Reevaluation of myoglobin for acute chest pain evaluation: would false-positive results on first-draw specimens lead to increased hospital admissions? Am. J. Clin. Pathol. 121, 804–808. Meloen, R.H., Puijk, W.C., Langeveld, J.P., Langedijk, J.P., Timmerman, P., 2003. Design of synthetic peptides for diagnostics. Curr. Protein Pept. Sci. 4, 253–260. Moe, K.T., Wong, P., 2010. Current trends in diagnostic biomarkers of acute coronary syndrome. Ann. Acad. Med. Singapore 39, 210–215. Mohammed, M.I., Desmulliez, M.P.Y., 2011. Lab-on-a-chip based immunosensor principles and technologies for the detection of cardiac biomarkers: a review. Lab. Chip 11, 569–595. Morrow, D.A., Cannon, C.P., Jesse, R.L., Newby, L.K., Ravkilde, J., Storrow, A.B., Wu, A.H., Christenson, R.H., National Academy of Clinical, B., 2007. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines: clinical characteristics and utilization of biochemical markers in acute coronary syndromes. Circulation 115, e356–e375.

50

G. Padmanaban et al. / Journal of Biotechnology 187 (2014) 43–50

Namgung, R., Mi Lee, Y., Kim, J., Jang, Y., Lee, B.H., Kim, I.S., Sokkar, P., Rhee, Y.M., Hoffman, A.S., Kim, W.J., 2014. Poly-cyclodextrin and poly-paclitaxel nano-assembly for anticancer therapy. Nat. Commun. 5, 3702. Paschke, M., 2006. Phage display systems and their applications. Appl. Microbiol. Biotechnol. 70, 2–11. Petrenko, V.A., Vodyanoy, V.J., 2003. Phage display for detection of biological threat agents. J. Microbiol. Methods 53, 253–262. Ruigrok, V.J.B., Levisson, M., Eppink, M.H.M., Smidt, H., van der Oost, J., 2011. Alternative affinity tools: more attractive than antibodies? Biochem. J. 436, 1–13. Sallach, S.M., Nowak, R., Hudson, M.P., Tokarski, G., Khoury, N., Tomlanovich, M.C., Jacobsen, G., de Lemos, J.A., McCord, J., 2004. A change in serum myoglobin to detect acute myocardial infarction in patients with normal troponin I levels. Am. J. Cardiol. 94, 864–867. Shone, C., Wilton-Smith, P., Appleton, N., Hambleton, P., Modi, N., Gatley, S., Melling, J., 1985. Monoclonal antibody-based immunoassay for type A Clostridium

botulinum toxin is comparable to the mouse bioassay. Appl. Environ. Microbiol. 50, 63–67. Tietz, N.W., 1995. Clinical Guide to Laboratory Tests, 3rd ed. W.B. Saunders, Co., St. Louis MO, pp. 482. Uchiyama, F., Tanaka, Y., Minari, Y., Toku, N., 2005. Designing scaffolds of peptides for phage display libraries. J. Biosci. Bioeng. 99, 448–456. Wang, K., Na, M.H., Hoffman, A.S., Shim, G., Han, S.E., Oh, Y.K., Kwon, I.C., Kim, I.S., Lee, B.H., 2011. In situ dose amplification by apoptosis-targeted drug delivery. J. Control Release 154, 214–217. Wang, K., Purushotham, S., Lee, J.Y., Na, M.H., Park, H., Oh, S.J., Park, R.W., Park, J.Y., Lee, E., Cho, B.C., Song, M.N., Baek, M.C., Kwak, W., Yoo, J., Hoffman, A.S., Oh, Y.K., Kim, I.S., Lee, B.H., 2010. In vivo imaging of tumor apoptosis using histone H1-targeting peptide. J. Control Release 148, 283–291. Willats, W.G., 2002. Phage display: practicalities and prospects. Plant Mol. Biol. 50, 837–854.