Rapid identification of Human Serum Albumin binding
peptides from a phage display library：a back-of-the-envelope
assessment of positive binders using titer-based criteria
Yi-Feng Shi, Min Li, Jia-Di Zhang, Lei Bian
Human serum albumin (HSA) is the most abundant protein in blood and has a 19-day in
PrePrints
vivo half-life, the longest human blood protein. HSA has also been extensively studied as a
drug carrier in a wide variety of clinical applications. HSA-binding, compared with HSAfusion, is promising strategy for extending the plasma half-life of protein therapeutics. The
construction of albumin-binding drugs requires assessment of a large enough quantity of
HSA-binding peptide candidates for conjugation with therapeutic proteins. Here, we report
a back-of-the-envelope assessment method to facilitate phage display selection of HSAbinding peptides. With an experimentally determined number of phage titers, we can
calculate the specificity ratios and the recovery yields. The recovery yield is calculated
using the titers of eluted phage divided by the titers of input phage. The specificity ratio is
calculated using the titer of eluted phage from a target-coated plate divided by the titer of
eluted phage from a blank-control plate. These parameters are defined as quantitative
criteria for panning and characterization of binding phage clones. Consequently, this
approach may enable more rapid and low-cost phage display screening of HSA-binding
peptides, which could be used as candidates of HSA binders for conjugation with
therapeutic proteins.
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
2
3
4
5
PrePrints
6
7
Rapid identification of Human Serum Albumin binding peptides from a
8
phage display library：a back-of-the-envelope assessment of positive
9
binders using titer-based criteria
10
11
Yi-Feng Shi*, Min Li, Jia-Di Zhang, Lei Bian
12
Department of Biotechnology, School of Biological Engineering, Dalian Polytechnic
13
University, China
14
15
16
Correspondence: Yi-Feng Shi, Department of Biotechnology, School of Biological Engineering,
17
Dalian Polytechnic University, Street No.1 Qinggongyuan, District Ganjingzi, Dalian, Liaoning
18
116034, China.
19
Phone: (86) 13591838227
20
E-mail address: [email protected]
21
22
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
23
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
25
Abstract
26
Human serum albumin (HSA) is the most abundant protein in blood and has a 19-day in vivo
27
half-life, the longest human blood protein. HSA has also been extensively studied as a drug
28
carrier in a wide variety of clinical applications. Albumin-binding, compared with albumin-
29
fusion, is promising strategy for extending the plasma half-life of protein therapeutics. The
30
construction of albumin-binding drugs requires assessment of a large enough quantity of HSA-
31
binding peptide candidates for conjugation with therapeutic proteins. Here, we report a back-of-
32
the-envelope assessment method to facilitate phage display selection of HSA-binding peptides.
33
With an experimentally determined number of phage titers, we can calculate the specificity ratios
34
and the recovery yields. The recovery yield is calculated using the titers of eluted phage divided
35
by the titers of input phage. The specificity ratio is calculated using the titer of eluted phage from
36
a target-coated plate divided by the titer of eluted phage from a blank-control plate. These
37
parameters are defined as quantitative criteria for panning and characterization of binding phage
38
clones. Consequently, this approach may enable more rapid and low-cost phage display
39
screening of HSA-binding peptides, which could be used as candidates of HSA binders for
40
conjugation with therapeutic proteins.
41
Keywords: Binding peptide; Phage display technology; Quantitative criteria; Human serum
42
albumin; Phage titers
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
44
1. Introduction
45
Human serum albumin (HSA) is a 67 KDa globular protein with a concentration is of 3.4 to
46
5.4 g/dL in blood, accounting for about 60% of the total protein in blood serum. Its long half-life,
47
19~20 days, makes HSA an ideal carrier to extend the half-lives of therapeutic proteins
48
(Varshney et al., 2010; Dennis et al., 2002). Both HSA fusion and HSA binding technologies
49
have been developed to reduce the clearance rate of peptide and protein drugs. HSA fusion
50
technologies, e.g. granulocyte colony-stimulating factor-HSA (Albugranin) (Halpern et al., 2002),
51
growth hormone-HSA (Albutropin) (Osborn et al., 2002a), interleukin 2-HSA (Albuleukin)
52
(Melder et al., 2005) and interferon-HSA (Albuferon-α) (Osborn et al., 2002b), have been
53
achieved through expression of HAS-fused proteins through gene recombination. In contrast,
54
HSA- binding technologies attach an HSA binding peptide or protein domain rather than HSA
55
itself to therapeutic proteins, enabling them to bind to HSA through non-covalent interactions.
56
For example, AB-Fab is trastuzumab-Fab fused to HSA binding peptide (AB) (Nguyen et al.,
57
2006), scDb-ABD is the albumin-binding domain of streptococcal protein G (ABDs) fused to a
58
bispecific single chain diabody (Stork et al., 2007) and AlbudAb/IL-1ra is interleukin-1 receptor
59
antagonist fused to serum albumin-binding domain antibodies (AlbudAbs) (Holt et al., 2008).
60
Which HSA technology is right for extending in vivo life of a protein drug? Walker et al.
61
compared such two forms of recombinant IFN-α2b: serum albumin-binding domain antibodies
62
fused to IFN-α2b (AlbudAbs- IFN-α2b) and HSA directly fused to IFN-α2b (HAS-IFN-α2b).
63
They found that AlbudAbs-IFN-α2b showed 18-fold higher potency in cell reporter assays, 1.5-
64
fold increased half-life in rat PK studies and 5.8-fold greater antiviral efficacy in the EMCV
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
65
assays (Walker et al., 2010). Shi et al. compared a series of protein drugs created by fusion
66
(covalent) and binding (non-covalent) to HSA and found that the HSA-fusion proteins showed a
67
more significant loss in binding capacity to their receptor, which related directly to
68
their pharmacological action. This effect was different than what was observed for HSA binding
69
proteins because the HSA molecule is much larger than HSA binders resulting in steric
70
hindrance and structural heterogeneity. In addition, the HAS-binding peptides are smaller in size
71
and do not greatly increase the size of recombinant proteins, thus often present an advantage in
72
their ability to diffuse rapidly into tissues (Shi et al., 2008).
73
Since HSA-binding technologies have been shown to effectively improve the half-life of
74
protein drugs, it is essential to obtain high affinity and specificity of HSA binding peptides or
75
HSA binders. Phage display technology has been successfully applied to antibody library and
76
random peptide library technologies since its emergence in 1985 (Smith, 1985; Kehoe et al.,
77
2005; Zhang et al., 2010), and provides a convenient platform for screening HAS- binding
78
peptides with the required properties. The use of the linear epitope mapping method has allowed
79
researchers to discover many HSA-binding peptides (Dennis et al., 2002). However, it may be
80
not efficient to select these peptides based on sequence identity via randomly sequencing
81
individual phage clones from enrichment pools. In this study, we described the efforts to isolate
82
HSA-binding peptides based on phage titer and focus on developing criteria for the performance
83
of rapid and cost-effective phage display screening.
84
85
2. Materials and Methods
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
86
2.1 Phage library and bacteria strains
87
The Ph.D.-7 Phage Display Library Kit including the Heptapeptide Phage Display Library
88
and E. coli ER2738 host strain was purchased from New England Biolabs. ER2738 with F
89
factor, which confers tetracycline resistance, can grow in LB-Tet medium (1% bacto-tryptone,
90
0.5% yeast extract, 0.5% NaCl, 20 g/mL tetracycline) and is used for M13 phage propagation.
91
Since the library phage carries the lacZ gene, phage plaques appear blue when the phage-
92
infected ER2738 cells plate on LB-Tet-IPTG-X-gal media (1% bacto-tryptone, 0.5% yeast
93
extract, 0.5% NaCl, 20 μg/mL Tetracycline, 50 μg/mL IPTG, 40 μg/ml X-gal, 1% agarose).
94
95
2.2 Phage titer analysis
A single E. coli ER2738 colony was inoculated into 10mL LB-Tet medium
and incubated
96
at 37 ℃ by shaking until mid-log phase (OD600 ~ 0.5) for phage titering. Phage solutions were
97
diluted 10-fold serially in TBS (50 mM Tris-HCl, pH 7.5, 150 mM NaCl) and aliquots of these
98
solutions (50 μL) were added to a suspension of the prepared ER2738 cells (450 μL), 1 for each
99
phage dilution. The phage infected ER2738 cells (100 μL) were then poured from each dilution
100
tube to a LB-Tet-IPTG-X-gal plate and the plates were incubated overnight at 37 °C. We
101
determined the number of phage particles contained in the original stock phage culture by
102
counting the number of plaques formed on the seeded agar plate and multiplying this by the
103
dilution factor.
104
2.3 Panning procedure
105
96-well microplates (Polypropylene, Xinhua Glass Instruments Factory, Sanhe, Haimen,
106
China) were coated overnight at 4 ℃ with 100 μL of the HSA target (Sigma, 100 μg/mL in 0.1
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
107
M NaHCO3, pH 8.6) and blank control (0.1 M NaHCO3, pH 8.6) respectively. After rapid
108
washing 3 times with TBST (TBS + 0.1% [v/v] Tween-20), 100 μL of the diluted phage library
109
(~2x1011 phage ) was added to the well of the coated plate and rocked gently to let them absorb
110
for 60 min. Following 10 washes with TBST, the bound phages were eluted with 0.2 M Gly-HCl,
111
pH 2.2. The 100 μL that was eluted was neutralized with 15 μL Tris-HCl (1 M, pH 9.1)
112
immediately. A small amount of eluted phage was titered as described above and the rest of
113
elutes were added to 20 ml ER2738 culture (OD600 ~ 0.5) to be amplified 4.5 h at 37℃ for the
114
next round.
115
2.4 Recovery yield and Specificity ratio
116
To evaluate the efficiency of the phage display panning procedure, we created two selection
117
criteria: the recovery yield and the specificity ratio. The recovery yield is the ratio of the titer of
118
the eluted phage to the input phage. The specificity ratio is the titer of the eluted phage from the
119
target coated well to the blank control well.
120
Recovery Yield (%)＝
121
Specificity Ratio＝
122
Elute phage titer from target
×100
Input phage titer
Elute phage titer from target
Elute phage titer from blank
2.5 Isolation and purification of phage
123
The amplified culture of phage clones was transferred to a 50 mL centrifuge tube and
124
centrifuged for 10 min at 10,000 rpm at 4 °C with a high-speed refrigerated centrifuge (Himac
125
CR21G). The supernatant was decanted to fresh tube and precipitated by 1/5 volume of PEG/ -
126
NaCl (20% [w/v] Polyethylene Glycol–8000, 2.5 M NaCl) at 4 °C overnight and centrifuged for
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
127
20 min at 10,000 rpm. The supernatant was discarded and the sediment containing phage was
128
suspended with TBS. To precipitate the phage again, a 1/5 volume of PEG/ NaCl was added to
129
the culture and it was stored at 4 °C for least 60 min. The PEG suspension was centrifuged for 10
130
min at 10,000 rpm, the phage precipitate was suspended in 500 µL of TBS.
131
2.6 DNA sequencing for characterization of specific binding peptides
132
After 3 rounds of bio-panning, individual phage clone (blue plaque) was isolated, amplified
133
and purified to check the specificity ratio of binding to HSA. The sequences of HSA specific
134
binding peptides were determined by DNA sequencing of the selected phage using -96 gШ
135
sequencing primer 5'-HOCCC TCA TAG TTA GCG TAA CG-3 ' in the Ph.D-7 phage library.
136
137
3. Results and Discussion
138
3.1 The recovery yield and the specificity ratio
139
To analyze the effectiveness of our phage display experimental technique, two operating
140
parameters, the recovery yield and the specificity ratio, were developed. These parameters
141
assessed the initial efficiency of phage display screening. The recovery yield was calculated by
142
the titer of eluted phages divided by the titer of input phages in the same target-coated well. The
143
value of recovery yield quantifies the enrichment of target-binding phages. Ideally, the recovery
144
yield will rise as the phage display proceeds and the enrichment of phages after each round of
145
panning progresses so that a greater and greater quantity of target-binding phages is achieved.
146
However, the specificity ratio is more necessary than the recovery yield in monitoring the phage
147
display screening experiment. The specific ratio is employed to identify positive clones of
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
148
target-binding phages, as compared with the blank control. It is calculated by the titer of eluted
149
phages from the target-coated well divided by the titer of eluted phages from the blank control
150
well. The specificity ratio should be larger than 1, for a positive result indicating that target-
151
binding phages are present. High specificity ratios indicate large amount of target-binding
152
phages and imply that there may be high affinity of phage clones present.
153
3.2 HSA-immobilizing microplates and blocking agents
154
Unlike the instructions on many of phage display peptide library kits, we do not need to coat
155
the well a blocking agent protein such as BSA, skim milk, et al. which process is common with
156
ELISA assays. Blocking proteins are a mixture that may cause more complex binding results in
157
phage display screening. If the number of phages binding to blocking proteins is larger than to
158
target proteins, the recovery yield still rises in each round of panning and may cause pseudo-
159
positive results in phage display screening. Thus we use a blank well as the negative control. The
160
specificity ratio is used to measure how much the number of target binding phages is larger than
161
the number of blank binding phage in each round of the panning process. Therefore no matter
162
what kind of microplate, with low or high protein-binding capacity, is used to immobilize the
163
target protein molecule, the specificity ratio is generally suited for assessment of the presence of
164
target-binding phages.
165
3.3 Panning against HSA
166
We examined the effects of target concentration screening HSA-binding peptides from the
167
Ph.D.-7 phage display peptide library, using 100 μg/mL and 200 μg/mL HSA concentration,
168
respectively,. Because blank micro-plates only function to distinguish target-binding phages
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
169
from microplate--binding phages in panning results, the first round of panning using the original
170
phage display peptide library against the blank microplate is not necessary and should be
171
excluded from the panning process. As can be seen from the phage titer results in the panning for
172
100μg/ml HSA coated microplates shown in Table 1 and the corresponding recovery yields and
173
specificity ratios shown in Table 2, the recovery yield rose significantly in the constant
174
concentration target. Specifically, the third round is 31.9-fold greater than the first round, while,
175
the specificity ratio improved, from 1.65 in the second round to 2.12 in the third round. The
176
increase in the recovery yields after each round indicates that the pools of high affinity phages
177
binding to HSA has been enriched.
178
It is necessary to use specificity ratio to avoid background interference generated by non-
179
specific binding (Vodnik et al., 2011). This is because eluted phages often include both target-
180
binding and microplate surface-binding phages, resulting in false-positive panning. Importantly,
181
the specificity ratio in panning rounds 2 and 3 that is larger than 1 indicated that the eluted
182
phages preferred binding to HSA in the target coated well over the blank in the target-uncoated
183
well. The increase of specificity ratio after each round of panning also implies that phages with
184
high affinity for HSA may have been amplified. However, low specificity ratios and recovery
185
yields may result from the presence of a large number of plate-surface-binding phages.
186
Our data of 100 μg/ml and 200 μg/ml HSA target together demonstrated both a recovery
187
yield and specificity ratio increase by each round panning. In 200 μg/mL HSA coated microplate
188
(Table 1), the recovery yield also increased 3.6-fold in the third round, as in the first round, and
189
the specificity ratios improved significantly from 2.35 in the second round up to 7.78 in the third
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
190
round (Table 2). This was larger than in the 100 μg/mL HSA coated plate (Table 2). In addition,
191
increasing the initial target concentration from 100 μg/mL to 200 μg/mL should be viewed as
192
reducing relative background (plate surface) interference, which contributed to panning
193
efficiency in the 200 μg/mL HSA plate better than in 100 μg/mL HSA plate.
194
3.4 Identification of positive HSA-binding phages
195
Finally, the specificity ratio was used to determine if positive phage clones with high HSA-
196
binding affinity were isolated. From the final round of panning plates, we isolated individual
197
clones to amplify, purify and measure the specificity ratio for binding to HSA. We tested the
198
selected clones with significant specificity ratios as potential HSA-binders, after screening many
199
phage clones that consisted of microplate-surface-binding phages, the specific and non-specific
200
target-binding phages, from the target-binding plate as shown in Table 3. The three selected
201
clones, HSA-3, HSA-4 and HSA-5, show specificity ratios of 7.4, 9.8 and 6.1, respectively,
202
confirming that our method in the panning process is effective. These phage-fused peptides were
203
subjected to DNA sequencing and their corresponding heptapeptide sequences are shown in
204
Table 3.
205
According to the instruction manual for the phage display peptide library kit by the New
206
England Biolabs (Instruction manual of Ph.D.™-7 Phage Display Peptide, New England
207
Biolabs), large-scale randomized DNA sequencing in the enrichment pool of eluted phages is
208
required to uncover the binding motif among target-binding peptides. But the surface of HSA for
209
binding drugs may not be limited to the motif-specific area. This method focusing on phage-
210
fused peptide clones with greater binding capacity rather than the HSA-binding motif may be a
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
211
more rapid and cost-efficient means of determining which HSA-binding peptides are useful and
212
effective for conjugation with a bioactive protein.
213
3.5 Intended HSA-binding affinity of HSA-binding peptide fusions rather than HSA-
214
binding peptide
215
Although we and others have developed methods to determine the relative dissociation
216
constants (kD) between phage displayed peptides and their respective targets via phage titer (Lu
217
et al., 2010; Dyson et al., 1995), the HSA-binding peptides rather than their fused phages should
218
be validated for HSA-binding affinity. When the HSA-binding peptides are fused with bioactive
219
proteins, their affinity for HSA would most likely decrease owing to shielding effects and
220
folding problems in their corresponding fusion proteins. The accurate affinity of HSA binding
221
peptide fusions for HSA, as measured by their Surface Plasmon Resonance(SPR)-measured KD,
222
may not be the same as that of free HSA-binding peptides binding to HSA. Thus, the specificity
223
ratio can be employed as an initial criterion, or a back-of-the-envelope assessment, to select
224
HSA-binding peptide candidates for fusion with the intended protein drugs. Our proposed
225
strategy is to select the candidate HSA-binding peptides for conjugation with the therapeutic
226
proteins using the specificity ratio, then subjected these to SPR to determine the KD between the
227
fusion protein and HSA.
228
Summary and outlook
229
In summary, both the recovery yield and the specificity ratio are designed to monitor the
230
phage display panning process. The specificity ratio is a critical recognition feature used to
231
discriminate target binding from background interference. It also serves as criteria to
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
232
demonstrate the target binding affinity and relative specificity of isolated clones from the phage
233
display peptide library. Use of the specificity ratio and recovery yield facilitates phage panning
234
and identification of positive phage clones. Therefore, these two quantitative parameters are
235
amenable to a wide variety of phage display screening applications.
236
The selected HSA-binding peptides will be attached to bioactive proteins to prolong their
237
half-lives through non-covalent association with HSA in vivo. The small size makes HSA-
238
binding peptides ideal fusion partners for recombinant proteins because they avoid decreasing
239
biological activity and maintain the native capacity to penetrate tissues. In addition, the
240
extremely high plasma concentration of HSA confers the fusion proteins more opportunities to
241
bind HSA, even with a relatively low affinity HAS-tag. Thus, small HSA-binding peptides are
242
attractive candidates for the engineering of pharmacokinetic properties of protein therapeutics.
243
244
Acknowledgement
245
Y.-F. Shi is grateful to Prof. Jeffery Smith (Sanford-Burnham Medical Research Institute) for
246
his suggestion on HSA. We also thank the assistance of Yin-Min Hou and Yang Liu in preparing
247
the experiment. This work was supported by Liaoning Province Education Department
248
（Project L2013216）.
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
250
References
251
Dennis MS, Zhang M, Meng YG, Kadkhodayan M, Kirchhofer D, Combs D, Damico LA. 2002.
252
Albumin binding as a general strategy for improving the pharmacokinetics of proteins.
253
Journal of Biological Chemistry 277:35035-43.
254
Dyson MR, Germaschewski V, Murray K. 1995. Direct measurement via phage titre of the
255
dissociation constants in solution of fusion phage-substrate complexes. Nucleic Acids Research
256
23:1531-5.
257
Halpern W, Riccobene TA, Agostini H, Baker K, Stolow D, Gu ML, Hirsch J, Mahoney A,
258
Carrell J, Boyd E, Grzegorzewski KJ. 2002. Albugranin, a recombinant human granulocyte
259
colony stimulating factor (G-CSF) genetically fused to recombinant human albumin induces
260
prolonged myelopoietic effects in mice and monkeys. Pharmaceutical Research 19:1720-9.
261
Holt LJ, Basran A, Jones K, Chorlton J, Jespers LS, Brewis ND, Tomlinson IM. 2008. Anti-
262
serum albumin domain antibodies for extending the half-lives of short lived drugs.
263
Protein Engineering Design & Selection 21:283-8.
264
Lu Y, Shi Y-F. 2010. Screening Human Transferrin Binding Peptides by Phage Technology.
265
Chinese Journal of Biotechnology 30:89-94.
266
Kehoe JW, Kay BK. 2005. Filamentous phage display in the new millennium.
267
Chemical Reviews. 105:4056−72.
268
Melder RJ, Osborn BL, Riccobene T, Kanakaraj P, Wei P, Chen G, Stolow D, Halpern, WG,
269
Migone TS, Wang Q, Grzegorzewski KJ, Gallant G. 2005. Pharmacokinetics and in vitro and in
270
vivo anti-tumor response of an interleukin-2-human serum albumin fusion protein in mice.
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
271
Cancer Immunology, Immunotherapy. 54: 535-47.
272
Nguyen A, Reyes AE, Zhang M, McDonald P, Wong WL, Damico LA, Dennis MS. 2006. The
273
pharmacokinetics of an albumin-binding Fab (AB.Fab) can be modulated as a function of affinity
274
for albumin. Protein Engineering Design & Selection 19:291-7.
275
Osborn BL, Sekut L, Corcoran M, Poortman C, Sturm B, Chen G, Mather D, Lin HL, Parry TJ.
276
2002a. Albutropin: a growth hormone-albumin fusion with improved pharmacokinetics and
277
pharmacodynamics in rats and monkeys. European Journal of Pharmacology 456:149-58.
278
Osborn BL, Olsen HS, Nardelli B, Murray JH, Zhou JX, Garcia A, Moody G, Zaritskaya LS,
279
Sung C. 2002b. Pharmacokinetic and pharmacodynamic studies of a human serum albumin-
280
interferon-alpha
281
Journal of Pharmacology and Experimental Therapeutics 303:540-8.
282
Shi Y-F, Bian L. 2008. Progress on the long-lasting protein drugs with Human Serum Albumin
283
as carrier. Chinese Medicinal Biotechnology 3:454-6.
284
Smith GP. 1985. Fliamentous fusion phage: novel expression vectors that display cloned
285
antigens on the virion surface. Science 228:1315-7.
286
Stork R, Müller D, Kontermann RE. 2007. A novel tri-functional antibody fusion protein with
287
improved pharmacokinetic properties generated by fusing a bispecific single-chain diabody with
288
an
289
Protein Engineering Design & Selection 20:569-76.
290
Varshney A, Sen P, Ahmad E, Rehan M, Subbarao N, Khan RH. 2010. Ligand binding strategies
291
of human serum albumin: how can the cargo be utilized?
fusion
albumin-binding
protein
domain
in
from
cynomolgus
streptococcal
monkeys.
protein
G.
Chirality 22:77-87.
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
292
Walker A, Dunlevy G, Rycroft D, Topley P, Holt LJ, Herbert T, Davies M, Cook F, Holmes S,
293
Jespers L, Herring C. 2010. Anti-serum albumin domain antibodies in the development of highly
294
potent, efficacious and long-acting interferon. Protein Engineering Design & Selection 23: 271-8.
PrePrints
295
296
Vodnik M, Zager U, Strukelj B, Lunder M. 2011. Phage display: selecting straws instead of a
297
needle from a haystack. Molecules 16: 790-817.
298
Zhang JD, Shi YF. 2010. Phage display technology using in function genomics and protein-
299
protein interaction─compared with yeast two-hybrid system, Chinese Medicinal Biotechnology
300
5: 211-5.
301
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
Table Captions
304
Table 1 Phage titers during panning against HSA
305
Table 2 Recovery yields and specificity ratios during panning against HSA
306
Table 3 Sequences of phage clones selected for binding to HSA
PrePrints
303
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
308
Table 1 Phage titers during panning against HSA
PrePrints
Panning target
Input
HSA
(pfu/mL)
Blank
(pfu/mL)
Elute
HSA
(pfu/mL)
Blank
(pfu/mL)
HSA with 100 μg/mL concentration
Round 1
Round 2 Round 3
HSA with 200 μg/mL concentration
Round 1
Round 2
Round 3
2.00×101
3.20×101
1
1
2.00×101
3.20×101
1
1
/
2.40×104
5.60× 104
2.96×105
/
3.40×104
1.40×105
8.40×1011
/
4.40×1011
4.24×101
1
4.24×101
2.40×1011
1
2.40×1011
5.20×104
7.24× 104
1.12×105
/
3.08×104
1.44×105
309
310
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
PrePrints
312
313
Table 2 The recovery yields and the specificity ratios during panning against HSA
Panning target
HSA with 100 μg/ml concentration
HSA with 200 μg/ml concentration
Round 1
Round 2
Round 3
Round 1
Round 2
Round 3
Recovery yield
2.90× 10-6 2.80× 10-5 9.25× 10-5
1.18× 10-5 1.71× 10-5 4.25× 10-5
(%)
Specificity
/
1.65
2.12
/
2.35
7.78
ratio
314
The recovery yields were calculated using the titers of eluted phage divided by the titers of
315
input phage; the specificity ratios were calculated by the titers of eluted phage from target
316
molecule coated plate divided by the titers of eluted phage from blank control plate.
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
318
319
320
Table 3 Sequences of phage clones selected for binding to HSA.
Input
phage
titre
(pfu/mL)
Elute
phage
titre
(pfu/mL)
Blank
control
phage titre
(pfu/mL)
Specificity
Ratio
DNA sequence
Peptide
sequence
HSA-3
1.16×1012
9.16×105
1.24×105
7.4
ATTAAGCTTCTAACTCTCCGC
IKLLTLR
HSA-4
1.34×1013
5.12×106
5.22×105
9.8
ACGCAACGGAAGCGGCGTAAG
TQRKRRK
HSA-5
1.28×1011
1.84×105
3.02×104
6.1
CATATAAGTCGCAGGCAGCTG
HISRRQL
PrePrints
Phage
displyed
peptide
321
322
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015
324
Highlights
325

target-binding motif sequencing, we perform phage display screening;
326
327
328
Using the specificity ratio and recovery yield, parameters based on phage-titer rather than

Human serum albumin (HSA) binding peptides or HSA binders have been identified rapidly
and cost-effectively as potential tags to extend the half-lives of protein drugs in vivo.
PrePrints
329
PeerJ PrePrints | https://dx.doi.org/10.7287/peerj.preprints.1115v1 | CC-BY 4.0 Open Access | rec: 23 May 2015, publ: 23 May 2015