PHAGE DISPLAY SCREENING AND EXPRESSION IN PLANTS
OF PEPTIDE APTAMERS THAT BIND TO PTHA
A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE
UNIVERSITY OF FLORIDA
TO MY PARENTS
I would like to thank my advisor Dr. Dean W. Gabriel, who instructed me and
always supported me. Also, I am greatly indebted to my graduate advisory committee
members, Dr. Robert R. Schmidt, Dr. Alice C. Harmon and later on Dr. L. Curtis
Hannah, who gave me a lot of valuable advice.
Thanks go to the people in the lab: Gary C. Marlow, who gave me a lot of
technical support; Dr. Yongping Duan, who gave me many good comments, and to the
graduate students in the lab, Adriana Castaiieda, Asha Huisden-Brunings and
Thanks go to the people outside of the lab: Mark Elliott helped me to do ELISA
microplate reading; the ICBR DNA Sequencing Core staff did all of the phage
sequencing and the Division of Plant Industry, Florida Department of Agriculture and
Consumer Services, provided the quarantine greenhouses for pathogenic inoculation of
I would like to thank Prof Yingchuan Tian from the Institute of Microbiology,
Chinese Academy of Sciences, in Beijing, China. He showed me a lot of experimental
techniques during his visit to our lab and initiated some parts of this project.
Finally, great thanks go to my parents who always support me.
TABLE OF CONTENTS
A CKN OW LED GM EN TS .................................................... ........ .. iv
LIST OF TABLES ................... .................. ................. .. .......... viii
LIST OF FIGURES ................... ................... .................. .. ......... ix
ABSTRACT ..................................... ....................................... xi
1 INTRODUCTION .................. ................... ................. 1
C itrus C anker H history ..................................................... ........... 1
Xanthomonas avr/pth Family ....................................................... 3
PthA Protein Function ................................................................. 6
Fusion Protein Construction, the pET-19b Expression
System and the GST Gene Fusion System ............... ................ 10
Phage Display Peptide Library Biopanning ....................................... 12
PCR-UDG Cloning Technique .......... ..................... .................. 16
Agrobacterium tumefaciens Mediated Transient
Expression in Plants .......... ................. ................. 17
Experimental Design .................................. .. ....................... 19
2 MATERIALS AND METHODS .......... ....................... ............ 21
Expression and Purification of PthA
Full-length Protein by His.Tag ............ .............................. 21
Expression and Purification of GST Fusion PthA
C-terminal 200 aa Truncated Protein ...................................... 23
Phage Display Peptide Library Biopanning ........................ ........... 25
In Vitro Binding Affinity of Aptamers ELISA ........... .................. 27
PCR-UDG Cloning of Aptamers .................. .............. ... ... ............. 28
Agrobacterium tumefaciens Mediated Transient
Expression of Aptamer Constructs in Bean,
Sweet Orange and Tomato ........... .......... .................. 31
3 RESULTS .......................... .................... ........ 34
Expression and Purification of PthA Protein by His.Tag ......................... 34
Expression and Purification of the PthA
COOH-terminal 200aa Truncated Protein
by GST Gene Fusion System ................ ......................... 37
Phage Display Peptide Library Biopanning .......................................... 39
In Vitro Binding Assay ELISA ................... ........................ 42
PCR-UDG Cloning of Aptamers ............ ....... ........... ............ 45
Agrobacterium tumefaciens Mediated Transient
Expression of Aptamers in Bean, Sweet
Orange and Tomato Leaves ...................................... 49
4 D ISC U SSIO N ..................... ....................................... .......... 54
5 SUMMARY AND CONCLUSIONS .................................. .......... 62
A MEDIA AND STRAINS USED .................. ............... .............. 64
M edia ................... ........................................ .... ......... 64
Strains ................... ................... .................. ... ......... 66
B PURIFICATION OF HIS.TAG PROTEIN
(FULL-LENGTH PTHA) .............. ............... .................. 68
C PURIFICATION OF GST FUSION PROTEIN
(PTHA C-TERMINAL 200AA
TRUNCATED PROTEIN) ............ .................... 76
D PHAGE DISPLAY PEPTIDE LIBRARY
BIOPANNING SCREENING OF APTAMERS ......................... 80
Phage Display Biopanning ................................ ........... ............ 80
Phage Amplification .............. ........ ....... ................. ............ 83
Phage Titer ............ ... ....................... ................... .... ......... 84
Characterization of Binding Clones ........................................ 86
E IN VITRO BINDING AFFINITY OF
APTAMERS ELISA ........... ................................... 88
F PCR-UDG CLONING OF APTAMERS ...................................... 93
G AGROBACTERIUM TUMEFACIENS MEDIATED
TRANSIENT EXPRESSION OF APTAMERS
IN BEAN, SWEET ORANGE AND
TOMATO LEAVES .............................. ................ 102
R EFEREN CE S ................... .................. .. .................. ..... ........ 109
BIOGRAPHICAL SKETCH ................. ....................................... 116
LIST OF TABLES
1. The genetic code for phage display peptide library biopanning ................ 13
2. The codon preference table of Citrus sinensis ........... ............ .......... 20
LIST OF FIGURES
1. Citrus canker symptoms on sweet orange ............................... ................ 2
2. The gene-for-gene "quadratic check"................ ....................... .............. 3
3. PthA full length peptide sequence .......................... ................ ........... 6
4. The predicted functional domains of PthA .................. ............. 7
5. The predicted peptide structure of PthA ......................... ....... .. ........ 8
6. N-terminal sequence of random, seven peptide
gill coat protein fusion ............ ............... ............... ............ 13
7. PCR-UDG cloning technique diagram ............................ ............. 15
8. M ap of pYY50.13 plasmid .................. ............... ...................... 22
9. Map of pGNLS3-2 plasmid .................. ......................... ............ 24
10. The design of PCR-UDG oligonucleotides to amplify
aptamers YPASYMQ and HPYTFLN .................................. ........ 29
11. Time course of PthA full length protein expression
induced by 0.1 m M IPTG ................................... ............ .......... 35
12. The purification of PthA full length protein by His.BindTM
affinity column chromatography ..... .............. ........................... 36
13. The expression of PthA COOH-terminal 200 aa truncated
protein induced by 0.1 mM IPTG at different times ......... .......... 37
14. Electro-elution of the COOH-terminal end of PthA,
fused to G ST ................... ......................................... ...... 38
15. Characterization of M13 PthA-binding clones ............ ............. 40
16. The DNA sequences of the M13 PthA-binding clones .......... ............. 41
17. In vitro Binding Assay of Aptamers .............. ...... ..... ........ ... ........... 43
18. The in vitro binding affinity of aptamers P7 (HPYTFLN),
N4 (YPASYM Q) and P10 (HPHTFLN) ............................................. 44
19. PCR amplification of aptamers P7 and N4 ................................. ......... 46
20. Uracil DNA glycosylase treatment of PCR products ................................. 47
21. 15.0% polyacrylamide gel electrophoresis of apatmers ...................... 47
22. Plasmid maps of pGZ7.5 and pGZ7.6 ......... ....................... .............. 48
23. Aptamers P7 and N4 cloned in Agrobacterium binary
vector pYD40.1 and pYD40.2 ............................ ................ 50
24. The plasmid maps of pGZ8.1, pGZ8.2, pGZ8.3
and pGZ8.4 ............. .. ...................... .......... ........ 51
25. Inoculation of Agrobacterium tumefaciens strain GV2260
constructs to the leaves of bean (panel A), citrus
(panel B) and tomato (panel C) .............. ................................... 52
26. The predicted Antigenicity Index of PthA ................ ...... ........ 60
27. The predicted KD Hydrophilicity and KD
Hydrophobicity of PthA ............. .......................... ............ 60
28. The predicted Surface Probability of PthA ................ ............. 61
Abstract of Thesis Presented to the Graduate School of the University of Florida
in Partial Fulfilment of the Requirement for the Degree of Master of Science
PHAGE DISPLAY SCREENING AND EXPRESSION IN PLANTS
OF PEPTIDE APTAMERS THAT BIND TO PTHA
Chairperson: Dean W. Gabriel
Major Department: Plant Molecular and Cellular Biology
The bacterial genus Xanthomonas is comprised of plant-associated bacteria, many
of which cause severe plant diseases. One of those is citrus canker disease, caused by
Xanthomonas citri. X citri causes citrus canker by injecting a protein signal molecule,
PthA, into citrus cells, and it is thought that several other severe Xanthomonas diseases
are similarly caused by other members of the avrBs3/pthA gene family. This work was to
determine if peptide aptamers could be selected that bind to PthA. A Phage Display
Peptide Library was screened to select 7-peptide aptamers that specifically bind to full-
length PthA and also to a truncated PthA consisting of the C-terminal 200 aa. Three
different aptamer sequences were identified that bound to both full-length PthA and
truncated PthA protein. ELISA tests indicated that there were no significant differences
in the binding affinities of the three aptamers. Based on the citrus codon usage
preference table, four PCR primers were designed and used to clone the aptamer
sequences into the plant expression vector pBI221. The aptamers were genetically
engineered as leader sequences fused to GUS and driven by a CaMV 35S promoter. The
engineered constructs were transferred into Agrobacterium tumefaciens strain GV2260
and inoculated onto California light-red kidney bean and sweet orange leaves. The
results showed that one aptamer, of sequence YPASYMQ, had a strong effect to block
the Hypersensitive Response (HR) normally elicited by pthA expressed in beans, and the
pathogenic response normally elicited by pthA on citrus. Another aptamer, of sequence
HPYTFLN, had reduced, but significant similar effects. Neither aptamer affected the
HR symptom elicited by expression of pthA in tomato cells. This result is in consistent
with other published reports that the COOH-terminal region of other members of the
pthA gene family are not important for elicitation of the HR in tomato. These results
confirmed that the aptamers exerted their effect by binding to the COOH-terminal end of
pthA, and indicated that aptamers might be used to control citrus canker disease and
several other severe diseases caused by Xanthomonas.
Citrus Canker History
According to the Florida Citrus Outlook 1998-99 Season Report (The Economic
and Market Research Department, Florida Department of Citrus, 1998), total Florida
citrus production, including round orange, specialty-citrus and grape fruit is expected to
be 248.8 million 90-lb boxes. The total 1998-1999 earnings are projected to be $971.5
million to $1,221.2 million. The Florida citrus industry is the largest among all US states
- about 3 times more than the total citrus production in California, the second largest
citrus provider in the US.
This important industry to the state of Florida is now threatened by citrus canker -
a world-wide citrus disease, widely distributed in Southeast Asia, Japan, the Middle East,
Africa and South America (Swarup et al., 1991). Last year, Asiatic citrus canker
infected large areas of southern Florida, including Hendry, Dade and Broward counties.
About $200 million this year was set aside for canker eradication in Florida.
A tree infected by citrus canker will drop fruit prematurely and exhibit lesions on
leaves, stems and fruit (De Feyter et al., 1993) (Figure 1). The bacteria responsible for
the canker, Xanthomonas citri (Gabriel et al., 1989), is spread primarily by wind-blown
rain, but also can be spread by contaminated tools and equipment, by people and by
animals. There is no effective resistance in grapefruit and little resistance in sweet
orange. Control by chemical sprays is only partially effective. Control by eradication
has been effective and X citri is a quarantined pathogen, subject to eradication. Both
infected and exposed trees are destroyed as part of a typical eradication program. If
citrus canker cannot be contained in Florida, losses of $8.55 billion in revenue is
expected according to the Florida Citrus Outlook 1998-99 Season Report. As many as
the 121,000 people could be unemployed, and 845,260 acres of citrus trees could be
destroyed. In 1985, an earlier outbreak of citrus canker in Florida forced the destruction
of 20 million citrus trees.
Figure 1. Citrus canker symptoms on sweet orange.
Due to this situation, researchers at USDA, the University of Florida, are working
to find a way to stop citrus canker. The primary purpose of this research was to try to
find a molecular biological approach to cure citrus canker disease and a possible way to
genetically engineer "immune" or "resistant" citrus trees.
Host Cultivar With
Pathogen R&- ri
Figure 2. The gene-for-gene "quadratic check".
Xanthomonas avr/pth Family
The genus Xanthomonas is unique among bacterial plant pathogens in exhibiting
a high degree of host-range specificity. The basis for the specifications is not clear.
Based on the classic "The gene-for-gene quadratic check" (Figure 2) (Ellingboe, 1976),
avirulence genes (avr) are considered to contribute to host range specificity in
Xanthomonas (Gabriel, 1997). Resistance in plants or avirulence in pathogens is usually
assayed by inoculation of pathogen into plants and is usually seen as a hypersensitive
response (HR) of the plant (Alfano and Collmer, 1996, Leach and White, 1996).
Surprisingly, most of the X campestris pv. malvacearum avr genes are also pathogenicity
(pth) genes (Yang et al., 1996). Even more surprisingly, these genes are members of the
same avr/pth gene family as the genes of X citri that are known to cause citrus canker
(Swarup et al., 1992, Duan et al., 1999). Most of the members of this avr/pth gene
family were originally isolated as avr genes and there was no evidence ofpth functions.
Similarly, pth genes were cloned by screening of pathogenecity, without evidence of avr
functions. Currently, 17 members of this gene family has been published, including X
campestris pv. malvacearum (cotton blight) (Yang et al., 1996), X citri (Asiatic citrus
canker) (Swarup et al., 1991), X campestris pv. aurantifolii (false citrus canker and
Mexican lime cancarosis) and X oryzae (rice blight) (Leach et al., 1996). The genes
cloned include avrBn (Gabriel et al., 1986), avrb6 (water soaking on cotton), avrB4,
avrb7 (water soaking on cotton), avrBIn (water soaking on cotton), avrBlO1 (water
soaking on cotton), avrB102 (water soaking on cotton) (De Feyter et al., 1993), avrB103,
avrB104 (water soaking on cotton), avrB5 (water soaking on cotton) (Yang et al., 1996),
pthN (water soaking on cotton), pthN2 (water soaking on cotton) (Chakrabarty et al.,
1997), avrBs3 (Bonas et al., 1989), pthA (canker on citrus) (Swarup et al., 1992), pthB
(canker on citrus), pthC (canker on citrus) (Yuan and Gabriel, unpublished data), avrXa5,
avrXa7 (elongated lesion on rice) and avrXalO (Hopkins et al., 1992).
The avr/pth gene encode signals affecting plant cell programs (Gabriel, 1999),
including "programmed death" (Dangl et al., 1996). All genes of the avr/pth family
found until today are from biotrophs which produce few degrading enzymes that could
assist releasing nutrient from living plant cells but not kill them (Vivian and Gibbon,
1997). A type III protein secretion system encoded by hrp genes (the name came from
affecting both hypersensitive response and pathogenicity of the bacterial strains) is
essential to deliver the avr/pth signals into plant cells (Van Gijsegem et al., 1995).
Mutation of this system abolish pathogenecity in biotrophic plant pathogenic Erwinia,
Pseudomonas and Xanthomonas (Alfano and Collmer, 1996). All of these bacterial
pathogens have very similar hrp genes (Galan, 1996).
Only a few molecules have been determined to be secreted by the hrp system.
One of these is a class of toxin-like molecules called harpins (Hoyos et al., 1996).
Harpins directly elicit plant symptoms on both host and nonhost plants. Harpins are
glycine rich proteins that seems to cause direct nutrient leakage from plant cells by
alkalization of the apoplast.
Another group of secreted molecules are the avr/pth gene family (Mecsas and
Strauss, 1996). Functional nuclear localization signals (NLSs) were found in all
members of Xanthomonas avr/pth gene family (Yang and Gabriel, 1995). This is
evidence that the signaling protein is directly targeted to plant cell nuclei. Site-directed
mutagenesis of NLS sequences abolished the pathogenicity phenotypes expressed by
these genes (Gabriel et al., 1996).
PthA Protein Function
Citrus canker disease has been historically described as having different "forms",
including Asiatic citrus canker (X. citri or "A form"), false citrus canker (X. campestris
pv. aurantifolii form B or "B form") and Mexican lime cancrosis (X. campestris pv.
aurantifolii form C or "C form"). A pathogenicity gene, pthA, was first cloned by on
pZit45 and confirmed necessary to for Xanthomonas citri to cause citrus canker disease,
the hyperplasia of citrus tissue (Swamp et al., 1991). The full length peptide sequence of
961 qchshpaqaf ddamtqfgms
rakpsptstq tpdqaslhaf adslerdlda
qsfevrapeq rdalhlplsw rvkrprtsig
gaaddfpafn eeelawlmel Ipq
Figure 3. PthA full length peptide sequence (1163 amino acids in total, the carboxyl
terminal 200 aa truncated protein is indicated by bold-italic).
Figure 4. The predicted functional domains of PthA.
PthA is given in Figure 3 (Swamp et al., 1992). There are 1163 amino acids. pthA,
avrb6 and avrBs3 are members ofXathomonas avr/pth family (Yang et al., 1994). The
complete DNA sequence of pthA is 97% identical to avrb6 and avrBs3. The central
regions of the predicted PthA protein is comprised of 17.5 tandem repeats each 34 amino
acids in length (Figure 4). These repeats are leucine-rich, indicating a potential protein-
protein interaction. Also, leucine zipper-like heptad repeats are present, that may also
indicate protein-protein or protein-DNA interactions. Site-directed mutagenesis of such
repeats proved that they are critical for the function of cultivar specific avirulence and
species-specific pathogenicity (Yang et al., 1994). The repeat regions of pthA and avrb6
determined hyperplasia of citrus and water soaking of cotton, respectively. Fusion
experiments using pthA and avrb6 repeat regions showed that the direct tandem repeats
determine both phenotypes, and are host-specific. The coding sequences of all members
of the gene family are flanked by nearly identical 62bp terminal inverted repeats, and the
Figure 5. The predicted peptide structure of PthA (including KD Hydrophilicity, Surface
Probability, Antigenic Index, CF Turns, Alpha Helices, Beta Sheets and Glycosyl Sites).
terminal 38bp of both inverted repeats are highly similar to the 38bp consensus terminal
sequence of the Tn3 family of transposons (De Feyter et al., 1993). Therefore, these
genes may have the capability to transpose.
The carboxyl terminal portion of PthA encodes three nuclear localization signals
(K-R/K-X-R/K) that are critical for PthA function and for localization to the host cell
nucleus (Yang and Gabriel, 1995). These three putative nuclear localization sequences
are at positions 1020-1024 (K-R-A-K-P) (nlsA), 1065-1069 (R-K-R-S-R) (nlsB), and
1101-1106 (R-V-K-R-P-R) (nlsC) in PthA. Site-directed mutagenesis of the nuclear
localization signal reduced the pthA localized in onion cell nuclei and also reduced the
pathogenicity of citrus canker symptoms (Gabriel et al., 1996). Mutation of all three
NLSs reducedpthA localization to onion cell nuclei. Mutations of any two NLSs
abolished the ability the pthA to elicit cankers on citrus, but did not abolish its ability to
elicit a non-host HR on tomato. Another member of the gene family from X campestris
pv. vesicatoria, avrBs3-2, does not require the 3' end (encoding all three NLSs) to elicit a
HR on tomato. Because avrBs3-2 and pthA are in the same gene family, and the DNA
sequence are very similar, it is predicted that PthA C-terminal region is also not
important to elicit a HR on tomato (Gabriel et al., 1996). The DNA coding sequence for
the C-terminal regions of pthA was fused to a P-glucuronidase (GUS) reporter gene.
Figure 5 shows the predicted structure of PthA full-length protein including KD
hydrophilicity, surface probability, antigen index, CF turns, alpha helics, beta sheets, and
glycosyl sites, etc. Both the N-terminus and Carboxyl terminus of PthA are likely to be
exposed outside of the naturally folded protein, thus, the COOH-terminus of PthA may
have the chance to interact with other protein or DNA molecules.
Fusion Protein Construction, the pET-19b Expression System and the GST Gene Fusion
The pET system used in the expression of recombinant proteins in E. coli was
originally constructed by Studier and colleagues (Studier and Moffatt, 1986, Studier et
al., 1990). The pET19b vector provides a translational N-terminal fusion of a cloned
gene to a cleavable His.Tag sequence for rapid affinity purification of the protein
product. His.Tag is an oligohistidine domain that allows convenient, economical
purification by His.Bind affinity resin. The target gene is cloned in the pET plasmids
under the control of a T7 promoter which is induced by T7 RNA polymerase in the host
cells (Dubendorff and Studier, 1991). T7 RNA polymerase is highly selective, which
converts almost all of the cell's protein production to target gene expression (Derman et
al., 1993). The expected yield of target protein can be more than 50% of the total protein
production after induction. Target genes are initially cloned in E. coli hosts that do not
contain the T7 RNA polymerase gene, i.e., the target genes are silent in the non-
expression host. When these plasmids are transferred into the expression host, E. coli
BL21(DE3)pLysS, which contains a chromosomal copy of the T7 RNA polymerase gene,
under lacUV5 control, expression of the target gene is strongly induced by the addition of
IPTG (Doherty et al., 1995). Since some proteins naturally occurred in E. coli host cell
having Histidine rich sequences, for some His.Tag constructs, they are not quiet efficient
to bind to His.Bind affinity resin or have very high background which create the
difficulties for the further purification of His.Tag proteins. Also, the very short peptides
are not very efficiently purified by His.Tag approaches (Studier et al., 1990).
The Schistosomajaponicum Glutathione S-transferase (GST) gene encodes a
26kDa protein that is expressed in E. coli with full enzymatic activity (Parker et al.,
1990). Recombinant genes are constructed in pGEX vectors, designed for inducible,
high-level intracellular expression of genes in fusion with GST (Maru et al., 1996).
GST fusion proteins have GST enzymatic activity (Ji et al., 1992). Gene expression is
under the control of the tac promoter, which is induced using the lactose analog IPTG
(Kaelin et al., 1992). After lysis of bacterial cells, the lysates are incubated with
Glutathione Sepharose 4B affinity resin. The glutathione group of GST is attached to
Sepharose 4B by coupling to the oxirane group using epoxy-activation. The GST fusion
protein can be 90% recovered using Glutathione Sepharose 4B in a single
chromatographic step. Fusion proteins are recovered from the matrix under mild elution
conditions (10mM glutathione) that preserve antigenicity and functionality of the
proteins. Specific protease cleavage sites allow separation of the GST and the fusion
protein to yield separate products. The GST fusion protein system proved to be very
efficient for short peptide expression (McTigue et al., 1995).
Phage Display Peptide Library Biopanning
There are several rapid and efficient methods using combinatorial biology for
screening random peptide epitopes binding to specific protein targets (Scott and Smith,
1990). The Ph.D.-7 Phage Display Peptide Library (New England BioLabs) is based on
a library of random 7-mer peptides fused to a minor coat protein (pIII) of the filamentous
coliphage M13. The Ph.D-7 peptide library is comprised of 2 x 109 sequences. By
comparison, theoretical numbers of possible 7-mers in a combination of 20 different
amino acids is 207 = 1.28 x 109 possible 7-residue sequences. Each amplification cycle
yields about 100 copies of each amplified sequence in 10pl of the supplied phage. The
displayed hetapeptides are expressed directly at the N-terminus of the pIII coat protein
when the fusion is expressed with a leader sequence removed during secretion, therefore,
the first residue of the mature protein is the first randomized position (Figure 6). The
peptide is followed by a short spacer (Gly-Gly-Gly-Ser) and then the wild type pIII
sequence. The genetic code of the Ph.D. phage display library is reduced, using only 32
codons encoding 20 different amino acids (Table 1).
Biopanning is a in vitro selection process that allows rapid identification of
peptide ligands for a variety of target molecules (such as antibodies, enzymes, cell-
surface receptors, etc) (Schumacher et al., 1996, Goodson et al., 1994, Barry et al., 1996,
Parmley and Smith, 1988). Biopanning is carried out by incubating a large random
library of unselected phage-displayed peptides with a polystyrene plate coated with the
target. After washing away the unbound phage, some peptides stick to the target and
Table 1. The genetic code for phage display peptide library biopanning.
pIII leader sequence KpnI
5'- ... TTA TTC GCA ATT CCT TTA GTG GTA CCT TTC
3'- ... AAT AAG CGT TAA GGA AAT CAC CAT GGA AAG
... Leu Phe Ala Ile Pro Leu Val Val Pro Phe
Start of mature 7 peptide gIII fusion EagI
NNK NNK NNK NNK NNK NNK NNK GGT GGA GGT TCG GCC
NNM NNM NNM NNM NNM NNM NNM CCA CCT CCA AGC CGG
Xxx Xxx Xxx Xxx Xxx Xxx Xxx Gly Gly Gly Ser Ala
AGT TGT TTA GCA AAA TCC CAT ACA GAA AAT TCA TTT
TCA ACA AAT CGT TTT AGG GTA TGT CTT TTA AGT AAA
Ser Cys Leu Ala Lys Ser His Thr Glu Asn Ser Phe
- -28gIII sequencing primer
K = G or T; M = A or C
ACT AAC ... -3'
TGA TTG ... -5'
Thr Asn ...
Figure 6. N-terminal sequence of random, seven peptide gill coat protein fusion
(The mature protein starting site is indicated by bold).
2nd T 2nd C 2ndA 2nd G
1st T Phe (F) Ser (S) Tyr (Y) Cys (C) 3rd T
Leu (L) Ser (S) Gin (Q) Trp (W) 3rd G
1st C Leu (L) Pro (P) His (H) Arg (R) 3rd T
Leu (L) Pro (P) Gin (Q) Arg (R) 3rd G
1st A Ile (I) Thr (T) Asn (N) Ser (S) 3rd T
Met (M) Thr (T) Lys (K) Arg (R) 3rd G
1st G Val (V) Ala (A) Asp (D) Gly (G) 3rd T
Val (V) Ala (A) Glu (E) Gly (G) 3rd G
remain bound, some with high affinity. Specifically-bound phage remains and is then
released under elution conditions. The eluted phage is then amplified in E. coli, the
enriched phage is collected, and then used to run additional binding/amplification cycles
to enrich the pool in a favor of binding sequences. After 3-4 rounds of enrichment, only
a few clones remain, the those are characterized by DNA sequencing.
The phage clones are also used for in vitro binding affinity assay with the original
target. Enzyme-linked immunosorbent assay (ELISA) is widely used for detection of
protein binding affinity in the phage display library screening, and the detection level can
be as low as 10-9 g of protein (Hoess et al., 1994). The method is to fix the specific target
protein on the supporting surface and add the selected binding peptide ligands, i.e., the
phage. After short incubating and washing, the unbound phage are washed off. At this
point, HRP conjugated anti-M13 antibody is added. The HRP converted colorless OPD
substrate into fluorescent yellow-orange color. Only the bound phage can react to anti-
M13 antibody and change the substrate color. The color level of the ELISA plate can be
read at 492nm.
Short peptide libraries displayed on phage have been used in a number of
applications, such as selection of phage-display peptides that bind to cucumber mosaic
virus coat protein (Gough et al., 1999), differentiating insecticidal activity of soybean
cystatins (Koiwa et al., 1998), and characterizing acquired resistance in lesion-mimic
transgenic potato expressing bacterio-opsin (Abad et al., 1997).
Figure 7. PCR-UDG cloning technique diagram.
PCR-UDG Cloning Technique
A novel ligation-independent method of site-directed mutagenesis of DNA
sequences using PCR amplification followed by uracil DNA glycosylase treatment
(termed UDG cloning) has proved to be very efficient and flexible (Rashtchian, 1995).
The method is to use two nucleotide primers that contain the desired DNA sequences and
which overlap at their 5' ends (Figure 7). The Thymine (dT) residues in the overlap
region are substituted with deoxyuracil (dU) in the primer nucleotide sequences. The
primers are designed to PCR amplify the entire plasmid to form a linear DNA fragment.
The desired base changes are in the primer. The two primers should be complementary
to each other (Smith et al., 1993). The presence of dU in the PCR products results the 5'
end of the amplified DNA fragments very susceptible to Uracil DNA glycosylase (UDG).
The UDG treatment of PCR products leads to excision of dU residues to produce 3'
protruding sticky ends. Because the two set of primers are overlapped, the two 3'
protruding ends are complementary each other and result in annealing of the PCR
products, i.e., the entire plasmid and the mutation sequence of the primer. The circular
PCR products are transformed to competent E. coli cells and any nicks of circularized
plasmid are filled by the E. coli in vivo repair system. The new plasmids are identical to
the wild-type parental plasmid except for the desired site-directed mutations designed on
the primers (Rashtchian et al., 1992).
One research group applied UDG cloning techniques to insert a chloramphenicol-
resistance gene into a SacI vector site and three segments of the lacZ PCR gene into a
correct orientation (Watson and Bennett, 1997). Another group designed primers with an
11 base overhang as an adaptor to PCR amplify the vector, while the PCR products were
treated by UDG to produce overhangs to yield a large numbers of inserts from human
cDNA cosmids. This method approved to be a highly efficient way to characterize genes
for shotgun sequencing of multiple shorter fragments in a large scale (Andersson et al.,
1994). Aptamer sequences selected from phage display peptide library biopanning only
comprise 21 base pairs (encoding 7 amino acids), and such a short DNA fragment is very
difficult to insert into a cloning vector and fuse with other stabilizing reporter genes in
the right orientation and in frame. UDG cloning techniques provide good approach to
genetic engineering of aptamers into the desired cloning vectors.
Agrobacterium tumefaciens Mediated Transient Expression in Plants
Agrobacterium tumefaciens can infect plants and introduce foreign genes into
plant cells (Hansen et al., 1994). At the site of infection, a crown gall (tumor tissue)
forms and synthesizes opines, which are novel amino acids that can only be metabolized
by Agrobacterium (Cho et al., 1997). The tumor is formed, and opines produced,
because of genes on the Ti plasmids (tumor-inducing plasmids) carried by
Agrobacterium. A small portion, usually only a 20kb segment, of the Ti plasmid, called
T-DNA (transferred DNA), carries the tumor-inducing and opine-producing genes and
integrates into the genome of the infected plant cell (Baron and Zambryski, 1996). The
bacteria can infect a susceptible plant only at the site of a wound, where plant cells secret
phenolic compounds that include acetosyringone and help activate the Ti plasmid genes
(Chumakov and Kurbanova, 1998). Scientists use molecular cloning techniques to insert
foreign DNA fragments between the left and right borders of the T-DNA and use
Agrobacterium Ti plasmids as a carrier to deliver cloned DNA into plant (Vaquero et al.,
1999). The ability of Agrobacterium T-DNA to integrate stably into its host's genome
gives plant molecular biologist a powerful tools to understand gene function. However,
transformation of plants take a lot of laboratory effort, and is not successful in every plant
Even without stable transformation, Agrobacterium Ti plasmids can be used to
achieve transient expression of genes in plants. Non-integrated T-DNA copies transiently
stay in the nucleus, where they are transcribed (Kapila et al., 1997). The transient
expression rate is much higher (at least 1,000 times) than that of the stable expression rate
in petunia leaf discs (Janssen and Gardner, 1989). Transient gene expression systems are
widely used because gene expression can be measured very shortly after T-DNA
delivery, there is no position effect, and the gene expression can be tested without
regeneration of a transgenic plant (Tai et al., 1999). A number of other techniques are
used to transfer DNA into intact plant tissue to achieve transient expression, including
electroporation (Lindsey and Jones, 1987, Dekeyser et al., 1990), particle bombardment
(Russell and Fromm, 1995), microinjection (Bilang et al., 1993), syringe injection
(Robinette and Matthysse, 1990) and vacuum infiltration (Kapila et al., 1997). Several
Agrobacterium tumefaciens host strains such as GV2260 (Deblaere et al., 1985), AGL1
(Lazo et al., 1991) and LBA4404 (Ooms et al., 1982) are commonly used in plant
transformation, and in transient expression assays. I used Agrobacterium tumefaciens
GV2260 with binary vector constructs to inject DNA into citrus, tomato and bean leaves
to assay pthA gene function and the interactions between the PthA and selected apatmers.
The first part of this thesis project was to translationally fusedpthA with His.Tag,
so that the recombinant PthA::His.Tag protein could be purified by His.Bind resin. The
COOH terminal 600bp end of pthA, encoding all three nuclear localizing signals (NLSs)
was translationally fused with GST and cloned into pGEX-4T-3. The recombinant GST
fusion protein was purified by Glutathione Sepharose 4B. The two purified recombinant
proteins, PthA full length protein and PthA COOH terminal 200 amino acid peptide, were
used as the targets to screen the Ph.D.-7 Phage Display Peptide Library to select 7-mer
amino acid aptamers by biopanning. The interesting M13 phage clones were DNA
sequenced and confirmed to bind PthA by in vitro binding affinity assay ELISA. The
aptamer clone sequences were fused to a GUS reporter gene by PCR-UDG cloning. The
codon of each amino acid was modified based on the codon preference table of citrus
(Table 2). The constructs were then cloned into an Agrobacterium binary vector with
and without the pthA gene and used to inoculate the Agrobacterium tumefaciens strains
on citrus, bean and tomato leaves. Transient expression assays were used to identity the
function of the aptamers in plants.
Table 2. The codon preference table of Citrus sinensis (the aptamer coding sequence
from phage display biopanning was translated into peptide sequence and replaced by
citrus preference codon).
Citrus sinensis [gbpln]: 17 CDS's (6141 codons)
fields: [triplet] [frequency: per thousand] ([number])
MATERIALS AND METHODS
Expression and Purification of PthA Full-length Protein by His.Tag
Full length pthA was cloned into the BamHI site of pET-19b to form pYY50.13
(Yang et al., 1995) (Figure 8). pYY50.13/E. coli DHSa single colonies were inoculated
into LB medium (Bacto tryptone 10g/l, Bacto yeast extract 5g/l, NaCl 5g/l, pH 7.5) with
100 [ig/ml of ampicillin and grown overnight. DNA mini-preps were made (Sambrook et
al., 1989) and used to transform E. coli protein expression host strain BL21(DE3)pLysS
(Studier et al., 1990). Fresh pYY50.13/E. coli BL21(DE3)pLysS colonies were
inoculated into LB medium with 50[lg/ml of carbenicillin and 34[lg/ml chloramphenicol
at 370C and shaken on a rotary shaker at 220 rpm until the cell density reached
OD600=0.5. Protein expression was evaluated by using 0.1 mM, ImM and 100mM IPTG
at 280C with shaking at 220 rpm for 3 hours. The cell pellet was collected and suspended
in IX Binding Buffer (5 mM Imidazole, 0.5 M NaC1, 20 mM Tris-HC1, pH 7.9) plus
0.1% NP-40 and 1 mM PMSF. Cells were then sonicated 60 seconds on ice to disrupt the
membranes using a SONICATOR Model W-225 (%duty cycle = 50, out put control=6,
pulsed). His.BindTM resin columns were set up by washing with 3 bed volumes of sterile
Figure 8. Map of pYY50.13 plasmid (pthA gene in pET-19b vector fused with His.Tag).
deionized water, charging with 5 bed volumes of 1X Charge Buffer (50mM NiSO4) and
equilibrating with 3 bed volumes of IX Binding Buffer. His.Tag fusionpthA protein
crude extract was loaded onto His.Bind resin column. The column was washed by 10
bed volumes of IX Binding Buffer, 6 bed volumes of IX Washing Buffer (60mM
Imidazole, 0.5 M NaC1, 20 mM Tris-HC1, pH 7.9). His.TagpthA protein was eluted with
lX Elute Buffer (1 M Imidazole, 0.5 M NaC1, 20 mM Tris-HC1, pH 7.9). The protein
elute was dialyzed against 50 mM NH4HCO3 at 40C overnight and lyophilized by
vacuum freeze drying. The freeze driedpthA protein was redissolved in 200 p1 1X
Treatment Buffer (0.125 M Tris.HC1, pH 6.8, 4% SDS, 20% glycerol, 0.2M DTT, 0.02%
Bromophenol Blue, pH 6.8) and examined for purity by SDS polyacrylamide gel
electrophoresis. The SDS-PAGE gel was soaked in 0.2 M ice cold KC1 and the 130 kDa
protein band excised with a sharp razor. The PthA protein was electro-eluted out of the
polyacrymide gel slice with Protein Elution Buffer (25 mM Tris base, 192 mM glycine,
0.1% SDS), using a setting of 10mA per tube for 8 hr. The Protein Elution Buffer was
exchanged for the same buffer without SDS and electro-elution was continued for
another 30 min to remove the SDS from the protein sample. The electro-eluted sample
was dialyzed against 50 mM NH4HCO3 overnight at 40C and freeze dried. Gel eluted
PthA was re-dissolved in 100 p of 0.1 M NaHCO3 (pH 8.6). The optical density of the
sample at 260 nm and 280 nm was measured and used to calculate the protein
concentration. The sample was stored at -200C until ready for use. Details are in
Expression and Purification of GST Fusion PthA C-terminal 200 aa Truncated Protein
The DNA encoding the carboxyl terminal 200 amino acids of PthA was cloned
from pthA in pYY50.13 by PCR amplification of a 660 bp gene fragment into the pGEX-
4T-3 vector (Pharmacia) by Prof. Yingchuan Tian (Gabriel and Tian, unpublished). This
resulted in a fusion of the COOH-terminus of PthA to Glutathione S-transferase (GST).
The clone was named pGNLS3-2 (Figure 9). Single colonies of pGNLS3-2/E. coli
Figure 9. Map of pGNLS3-2 plasmid (pthA gene COOH-terminal 660 bp fused with
GST in pGEX-4T-3 vector).
DH5ca were used to inoculate LB medium with ampicillin to a final concentration of
100[lg/ml and mini-preps of pGNLS3-2 plasmid DNA were used to transform E. coli
protein expression strain BL21(DE3)pLysS competent cells. pGNLS3-2/E. coli
BL21(DE3)pLysS cells were grown in a rotary shaker at 370C at 220 rpm in 2X YT
medium (Tryptone 16g/l, yeast extract 10g/l, NaCl 5g/l, pH 7.0) with 50[lg/ml
carbenicillin and 34[lg/ml chloramphenicol until growth reached the mid log phage
(OD600=0.5). Protein expression was induced by addition of 0.1 mM IPTG with
continued incubation at 300C at 220 rpm for 3 hours. Cells were collected by
centrifugation at 7,000 rpm at 4C for 10 min and suspended in 1X PBS (140 mM NaC1,
2.7 mM KC1, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.3) plus 0.1% NP-40 and 1 mM
PMSF. Cells were disrupted by sonication on ice. The lysate was centrifuged at 12,000
rpm for 20 min at 4C, the supernatant was collected and passed through a 0.45 micron
membrane filter. One ml of this crude protein extract was combined with 100 pl1
Glutathione Sepharose 4B (a w/v 50% slurry in IX PBS), and slowly inverted to mix
using a rotary platform shaker with gentle agitation at 4C for 30 min. The GST
(Glutathione S-transferase) fusion protein was eluted using 1 ml of Glutathione Elution
Buffer (10 mM reduced glutathione, 50mM Tris-HC1, pH 8.0) at 4C for 30 min. The
eluted protein mixture was centrifuged at 500 x g for 10 min at 4C to pellet the protein.
The protein pellet was lyophilized as described previously for PthA. The truncated PthA
COOH-terminal 200 aa GST fusion protein was redissolved into 100 p1 of 0.1 M
NaHCO3 (pH 8.6). Details are in Appendix C.
Phage Display Peptide Library Biopanning
The full length PthA (His.Tag fusion) and PthA COOH-terminal 200 amino acids
truncated protein (GST fusion) were diluted to 100 plg/ml in 0.1 M NaHCO3 (pH 8.6).
100 p[g/ml of Glutathione S-transferase in 0.1 M NaHCO3 (pH 8.6) was prepared as pre-
absorption blocking agent. Also, 100 [lg/ml of Bovine Serum Albumin (BSA) was used
as the background control. A 96-well microtiter plate was coated with 150 p|l of each
target per well, with gentle agitation at 40C overnight. Each well was rinsed with
Blocking Buffer (NaHCO3, pH 8.6, 0.1M, BSA, 5mg/ml, NaN3, 0.02%) for at least 1 hr
at 4C and then washed 6 times by IX TBST (Tris.HC1, pH 7.5, 50 mM, NaC1, 150 mM,
Tween-20 0.1% v/v) at room temperature. The original unselected library (New England
BioLabs, Ph.D.-7TM Phage Display Heptapeptide Library, 2 x 1013 pfu/ml) was diluted to
2 x 1011 phage in 1 ml of 1X TBST and incubated on the coated and blocked plates at
room temperature for 60 min with gentle agitation.
To reduce non-specific binding of GST fusion protein, the unselected phage
library was first incubated with GST coated and blocked plates at room temperature for
60 min, and then transferred to the PthA COOH-terminal 200 aa GST fusion protein
coated plates and incubated for another 60 min at room temperature. After incubation,
all plates were washed 10 times using IX TBST. Bound phage was eluted with 100 pfl
of 100 plg/ml full length PthA protein or PthA COOH-terminal 200 aa truncated GST
fusion protein in IX TBS (Tris.HC1, pH 7.5 50 mM, NaC1, 150 mM) at room temperature
for 60 min. The eluted phage was amplified until the titer was up to 2 x 1013 pfu/ml.
These biopanning steps were repeated three times. After the second round of selection,
the Tween-20 concentration was increased to 0.5% v/v in the washing steps. For the
GST fusion protein, each round of selection used plates that were pre-absorbed by GST.
For the PthA full length protein, four rounds of phage display biopanning were carried
out. A fourth round of selection was carried out using full length PthA protein on the
library previously selected using only the truncated GST fusion PthA COOH-terminal
200 aa protein. At least 10 clones from both screened libraries were picked at random for
DNA sequencing. The DNA sequencing primer was -28gIII (5'-
GTATGGGATTTTGCTAAACAAC-3'). The DNA sequencing data was used to find out
the consensus sequence of binding clones. Interesting clones were stored at -200C.
Details are in Appendix D.
In Vitro Binding Affinity of Aptamers ELISA
Representative interesting clones were used for in vitro binding affinity assays.
The putative phage clone was amplified to a titer of 1013 pfu/ml. As a control, 1 1l of
M13 phage from the original unselected library (New England BioLabs, Ph.D.-7TM Phage
Display Heptapeptide Library, 2 x 1013 pfu/ml) was amplified to a titer of 1013 pfu/ml.
Two targets were used to coat 0.5ml microcentrifuge tubes: 150 1l of 100 [lg/ml full
length PthA protein in 0.1 M NaHCO3 (pH 8.6) or 150 p1l of 100 plg/ml PthA COOH-
terminal 200 aa truncated GST fusion protein in 0.1 M NaHCO3 (pH 8.6). As a target
background control, 150 p1 of 100 [lg/ml ofBSA in 0.1 M NaHCO3 (pH 8.6) was used.
The coated microcentrifuge tubes were incubated with gentle agitation at 4C overnight.
Each tube was then blocked using 200p1 Blocking Buffer at 4C for 2 hr. Serial dilutions
of phage in IX TBST were prepared in a dilution range of 107 to 1012. Each tube was
washed 6 times with IX TBST at room temperature. To reduce non-specific binding, the
serial dilutions of phage were first incubated in microcentrifuge tubes which were only
rinsed by Blocking Buffer for 60 min at room temperature and then the pre-absorbed
phage was transferred into the blocked target protein tubes to continue incubating at room
temperature for another 60 min. Each tube was then washed 6 times with IX TBS with
0.5% (v/v) Tween-20. Horse Radish Peroxidase/Anti-M13 Monoclonal Conjugate
(Pharmacia) was diluted 1:5000 in IX Blocking Buffer and dispersed 200 pLl to each
phage-treated tube and incubated at room temperature for 60 min with gentle agitation.
Each tube was washed 6 times in 1X TBS with 0.5% (v/v) Tween-20. Sigma FAST
OPDTM (O-Phenylenediamine Dihydrochloride) substrate tablet was reconstituted by
adding 20 ml distilled water and 200 Ipl was added to each tube to incubate in the dark at
room temperature for 30 min. To stop the reaction, 50 Ipl of 3M HC1 was added to each
tube. 150 p1l of solution from each tube was transferred to a 96-well ELISA microtiter
plate for comparisons. A Microplate Autoreader (model EL 309) Bio-Tek Instruments
was used to read absorbance of each well at 492nm. Details are in Appendix E.
PCR-UDG Cloning of Aptamers
A PCR-UDG cloning strategy was designed to clone the aptamer, YPASYMQ
and HPYTFLN, respectively, as the leading sequence and in-frame fusion with GUS
reporter gene in a plant expression vector generated from pBI221 (Figure 10). Each
amino acid codon was modified based on the citrus codon preference (Table 2). Full
length pthA gene was cloned into the BamHI site of plant expression vector pBI221 to
create a tranlational fusion with GUS reporter gene on the vector by Dr. Yongping Duan
(Duan et al., 1999). The resulting clone was named pYD12.9F. E. coli DH5ca was
transformed with this vector and single colonies were inoculated into LB medium with
ampicillin at a concentration of 100[ig/ml and grown at 370C in a rotary shaker at 220
Figure 10. The design of PCR-UDG oligonucleotides to amplify aptamers YPASYMQ
and HPYTFLN (The aptamer sequences are indicated by italic on the oligo).
rpm overnight. Plasmid DNA was extracted and digested with BamHI. After
phenol/chloroform extraction and ethanol precipitation, pYD12.9F/BamHI fragments
were resuspended in TE (pH 8.0) with the final concentration of 5ng/|ll. Two sets of
oligonucleotides were designed for UDG cloning according to the amino acid sequences
of selected aptamers (YPASYMQ and HPYTFLN) and codon preference table of Citrus
sinensis (Figure 10 and Table 2). Each codon coding the amino acid sequence of
aptamers was specifically converted to codon usage appropriate for citrus and the
aptamer sequences were included in the synthetic oligonucleotides. Overlaps of 18
nucleotides were designed with 4 to 8 uracil molecules and after Uracil DNA glycosylase
treatment, the Uracil residuals dissolved and anneal the overlaps. The oligonucleotides
were synthesized by GIBCO BRL and reconstituted in TE (pH 8.0). PCR reaction
mixtures were set up as 2[pl of 5ng/pl pYD12.9F/BamHI, 15 pil of 10X PCR buffer, 4.5pl
of 50 mM MgC12, 15pl of 2.5mM dNTP, 82.75pl of distilled water. Oligonucleotides
were added in two sets: 15 ptl each of primers GZ1 (4.36 piM) and GZ2 (4.32 pIM) or 15
p1l each of primers GZ3 (4.73 [pM) and GZ4 (4.82 [pM). Details of PCR cycle conditions
are in Appendix F. After PCR amplification, Uracil DNA Glycosylase (UDG) treatment
was carried out by incubation of 5pAl of crude PCR product, 2pl of 10X PCR buffer, 1~1l
of 1U/pl Uracil DNA glycosylase (GIBCO BRL) and 12 p1l of distilled water at 370C for
30 min followed by denaturation at 65 C for 10 min. The UDG treated PCR products
were transformed to E. coli DH5a competent cells. Single colonies were inoculated into
LB medium with ampicillin at a concentration of 100p[g/ml in a rotary shaker at 220 rpm
and grown at 370C overnight. The plasmid DNA was extracted and then separately
digested by Xbal/BamHI and Xbal/Smal and examined by 15.0% polyacrylamide gel
electrophoresis. Interesting clones were saved as glycerol (50% v/v) stock cultures at -
700C. GUS activity assays were carried out by resuspending pellets from 1 ml overnight
cultures into 570 p1 distilled water and 330 p1 of 0.3 mg/m 5-bromo-4-chloro-3-indolyl-
P-D-glucuronide (X-gluc). The suspensions were incubated at room temperature
overnight. Putative clones showing blue color as compared to the control of E. coli
DH5a cells were examined further. Plasmid DNA of putative clones was purified using
QIAGEN Midi Kits and DNA sequencing was carried out with the DNA sequencing
primer DG39R (5'- CATAAGGGACTGACCA -3'). Details are in Appendix F.
Agrobacterium tumefaciens Mediated Transient Expression of Aptamer Constructs in
Bean, Sweet Orange and Tomato
Single colonies of above PCR-UDG putative clones were inoculated into LB
medium with ampicillin at a concentration of 100p[g/ml in a rotary shaker at 220 rpm and
grown at 370C overnight. Plasmid DNA was extracted and digested by EcoRI. The
digested insert DNA fragments were precipitated by ethanol and redissolved in TE (pH
8.0) to a final concentration of 1l[g/p[l. Full length pthA gene was cloned into pGZ6.4
binary vector by Dr. Yongping Duan, and the resulting clone was named pYD40.1 (Duan
et al., 1999). As a control, pthA was deleted by digesting with BamHI and religation,
generated a control clone named pYD40.2 (Duan et al., 1999). E. coli DH5a/pYD40.1
and DH5a/pYD40.2 were inoculated into LB medium with kanamycin at the
concentration of 50p[g/ml, respectively, and incubated at 370C in a rotary shaker at a
speed of 220 rpm overnight. Plasmid DNA was extracted and digested by EcoRI.
The digested DNA fragments of pYD40.1 and pYD40.2 were separately extracted
by phenol/chloroform, precipitated by ethanol, and redissolved in TE (pH 8.0). 20 ng of
each EcoRI digest were incubated with 5 p[l of 10X buffer and 1 l1 of Shrimp Alkaline
Phosphatase (1:1 fresh diluted by IX Dilution Buffer) in a total volume of 50pl at 37C
for 60 min followed by 65 C denaturation for 20 min. Ligation reactions were set up
with 4 pil of EcoRI digest of PCR-UDG putative clones, 2[pl of the pYD40.1 or pYD40.2
EcoRI digests, 2[il of 5X ligation buffer, 1 il of T4 DNA ligase in a total volume of 10 pl
at 160C overnight. 2 pAl of the crude ligation mixture was used to transform E. coli DH5a
competent cells and transformants were selected on LB agar plates containing 50p[g/ml of
carbenicillin and 50[lg/ml of kanamycin.
Single colonies were inoculated into LB medium containing 50[lg/ml of
carbenicillin and 50[lg/ml of kanamycin and incubated in a rotary shaker at 370C with a
speed of 220 rpm overnight. Plasmid DNA extracts were digested by EcoRI to confirm
the insertion. The orientation of interesting clones was tested by digestion of SstI.
Putative clones were mated into Agrobacterium tumefaciens GV2260 by tri-parental
matings and incubated on LB agar plate at 280C overnight. Selection for transconjugants
was on LB agar plates containing 50[lg/ml of carbenicillin, 50[lg/ml of kanamycin and
75p[g/ml rifomycin. As a control, pYD40.1 and pYD40.2 were also mated into
Agrobacterium tumefaciens GV2260 and selected on LB agar plates containing 50p[g/ml
kanamycin and 75p[g/ml rifomycin. Single colonies of Agrobacterium tumefaciens
GV2260 and transconjugants were inoculated into 20 ml YEB medium (Beef extract 5g/l,
yeast extract lg/1, peptone 5g/l, sucrose 5g/l, MgSO4 2mM, MES 10mM, acetosyringone
20piM, pH 5.6) with the same antibiotics as used in the mating selections and incubated
in a rotary shaker at 280C and a speed of 60 rpm until the OD600 reached 0.8. Bacterial
cells were then collected by micro-centrifugation at 4C and a speed of 5,000 rpm and
the pellets were resuspended in MMA medium (IX MS salt, MES 10mM, sucrose 20g/l,
acetosyringone 200piM, pH 5.6) until OD600=0.5. The suspension was kept at room
temperature for 1 hr before use.
Ten-day-old young leaves of California light red kidney bean were infiltrated by
Agrobacterium tumefaciens GV2260 suspension using vacuum infiltration at a pressure
of 0.5 mbar for 20 min. The infiltrated leaves were incubated in 100mm petri-dishes with
three-layers of 3MM paper moistured by distilled water and kept at 220C under light for
36 hr. Also, cell suspensions ofAgrobacterium tumefaciens GV2260 (OD600=0.3) were
injected using the blunt end of tubular syringe the abaxial side of tender sweet orange
leaves or young tomato leaves (Kapila et al., 1997). Inoculated sweet orange plants were
grown at 280C for four weeks in a quarantined green house. The inoculated tomato
plants grew under low light for 24 hr in growth chambers. Details are in Appendix G.
Expression and Purification of PthA Protein by His.Tag
Full-length PthA protein expression was induced by adding IPTG. Three
different IPTG concentrations (0.1mM ImM and 100mM), seven induction times (Ohr,
lhr, 2hr, 3hr, 4hr, 5hr and 6hr) and three induction temperatures (280C, 300C and 370C)
were tested. The optimum IPTG induction concentration was 0. mM and the induction
temperature was 280C. 10l of each sample was mixed with 10 1l of IX Treatment
Buffer (0.125 M Tris.HC1, pH6.8, 4% SDS, 20% glycerol, 0.2M DTT, 0.02%
bromophenol blue), boiled 10 min, and centrifuged at 10,000 rpm for 10 min.
Supernatant was loaded on 8% (w/v) polyacryamide separating gel with 4% (w/v)
stacking gels. Figure 11 shows the time course of full-length PthA protein expression
induced by 0. ImM IPTG. A new 130 kDa protein band was very clear after IPTG
induction for 1 hr (Figure 11, band D). Maximum induction was obtained after 3hr
(Figure 11, band F). Compared to pYY50.13/E.coli BL21(DE3)pLysS control (Figure
11, band A) and no IPTG induction control (Figure 11, band B), the results obtained that
0. mM IPTG was optimum to induce His.Tag fusion protein expression.
A B C D E F
Figure 11. Time course of PthA full length protein expression induced by 0.1 mM
A: pYY50.13/E. coli BL21(DE3)pLysS, B: no IPTG,
C: IPTG 0 hr, D: IPTG 1 hr, E: IPTG 2 hr, F: IPTG 3 hr,
G: IPTG 4 hr, H: IPTG 5 hr, I: IPTG 6 hr, J: MW marker.
After His.Bind affinity column chromatography, each fraction was collected and
10pl of each sample was mixed 1:1 with IX Treatment Buffer for SDS-PAGE. Figure 12
shows PthA His.Tag protein after His.Bind affinity column purification. Compared to
pYY50.13/E.coli BL21(DE3)pLysS control (Figure 12, band A), a good yield of full-
length PthA protein was achieved by 0. Im IPTG induction for 3 hr (Figure 12, band B).
After sonication, most PthA protein was found in the soluble protein fractions (Figure 12,
G H I
A B C D
Figure 12. The purification of PthA full length protein by His.BindTM
affinity column chromatography.
A: pYY50.13/BL21(DE3)pLysS no induction, B: 0.1 mM IPTG induction for 3 hr,
C: Soluble protein fraction after sonication, D: Soluble protein after His.Bind resin
binding, E: His.Bind column PthA protein elute, F: Electro gel elute PthA protein, G:
band C). After His.Bind affinity column chromatography, the absence of the 130 kDa
band in eluate indicated that most PthA protein was bound to the column (Figure 12,
band D). Most of the protein in the elute from the His.Bind column was PthA protein
(Figure 12, band E). After electro gel elution, the PthA full-length protein was very pure
and appeared as a single clear band at 130 kDa (Figure 12, band F).
E F G
Expression and Purification of the PthA COOH-terminal 200aa Truncated Protein by
GST Gene Fusion System
Similar experiments were carried out with the PthA COOH-terminal 200aa GST
fusion protein. Three different IPTG concentrations (0.1mM, ImM and 100mM), seven
induction times (Ohr, lhr, 2hr, 3hr, 4hr, 5hr and 6hr) and three induction temperatures
(280C, 300C and 370C) were tested. The optimum IPTG induction concentration was
0.1 mM and the induction temperature was 30C.
A B C D E F G H I J
Figure 13. The expression of PthA COOH-terminal 200 aa truncated protein
induced by 0.1 mM IPTG at different times.
A: pGNLS3-2/E. coli BL21(DE3)pLysS, B: no IPTG, C: IPTG 0 hr,
D: IPTG 1 hr, E: IPTG 2 hr, F: IPTG 3 hr, G: IPTG 4 hr,
H: IPTG 5 hr, I: IPTG 6 hr, J: MW marker.
Figure 14. Electro-elution of the COOH-terminal end of PthA, fused to GST.
A: Electro gel elution of PthA C-terminal 200 aa truncated protein, B: MW marker.
Figure 13 shows the expression pattern of GST fusion proteins at 300C after
induction by adding 0. 1mM IPTG. After 1 hr of 0. 1mM IPTG induction, a new 55kDa
protein band appeared (Figure 13, band D) compared to the pGNLS3-2/E. coli
BL21(DE3)pLysS control (Figure 13, band A), expression of the 55kDa protein reached
its peak at 2 hr (Figure 13, band E). Longer induction did not increase the expression
level of GST fusion protein (Figure 13, band F-I).
Figure 14 shows the purity of electro-gel-eluted PthA COOH-terminal 200aa GST
fusion protein. The GST fusion PthA COOH-terminal 200 aa peptide was found at the
expected size of 55 kDa (Figure 14 band A).
Phage Display Peptide Library Biopanning
After the fourth round of selection, unamplified phage was titered. Single
plaques were picked up in plates with about 100 plaques. Single stranded M13 phage
DNA was prepared and dissolved in 30pil of TE (pH 8.0) (Appendix D). l1[l of each
phage prep was tested by 0.7% agarose gel electrophoresis (Figure 15). Figure 15 Panel
A shows 10 randomly selected clones following biopanning selection using PthA full
length protein as a target for all four rounds of selection. Figure 15 Panel B shows 10
clones using PthA COOH-terminal 200 aa truncated protein as a target for three rounds of
selection and then using PthA full length protein as a target in the 4th selection round.
10pl of each selected M13 phage DNA was sequenced and the result is
summarized in Figure 16. Of 11 random clones sequenced from the phage population
after four rounds of selection, a DNA sequence encoding HPYTFLN appeared 5 times
(45.5%), a sequence encoding HPHTFLN appeared twice (18.2%) and a sequence
encoding YPASYMQ appeared 4 times (36.4%) (Figure 16, panel A). Of 10 random
clones from the phage population after four rounds of selection, a DNA sequence
encoding HPYTFLN appeared twice (20.0%) and a sequence encoding YPASYMQ
appeared 7 times (70.0%) (Figure 16, panel B).
A 1 2 3 4 5 6 7 8 91011
B 1 2 3 4 5 6 7 8 9 1011
Figure 15. Characterization of M13 PthA-binding clones.
Al:1 Kb DNA Ladder; A2 Al 1:10 clones using PthA full length protein as target in
phage display biopanning of all four rounds of selection; B B10: 10 clones using PthA
COOH-terminal 200 aa truncated protein as target till 3rd round selection and cross
react to PthA full length protein in 4th round selection; B 11: 1 Kb DNA Ladder.
A. Phage display sequences of binding clones after 4th round selection (all four rounds of
selection against PthA full length protein):
B. Phage display sequences of binding clones after 4th round cross-reaction against PthA
full length protein (previous selection against PthA C-terminal 200 aa truncated protein
till 3rd round)
(CTG CAT ATA CGA AGC AGG ATA)
(CTG CAT ATA CGA AGC AGG ATA)
(probably bound to polystyrene well)
(CTG CAT ATA CGA AGC AGG ATA)
(CTG CAT ATA CGA AGC AGG ATA)
(ATT CAG AAA CGT ATA CGG ATG)
(CTG CAT ATA CGA AGC AGG ATA)
(CTG CAT ATA CGA AGC AGG ATA)
(ATT CAG AAA CGT ATA CGG ATG)
(CTG CAT ATA CGA AGC AGG ATA)
Figure 16. The DNA sequences of the M13 PthA-binding clones (The conversion of
nucleotide sequence to peptide sequence is listed in Table 1).
(5 of 11, 45.5%)
(2 of 11, 18.2%)
(4 of 11, 36.4%)
In Vitro Binding Assay ELISA
Three representative clones were chosen for ELISA tests, P7.28g (P7) for the
HPYTFLN aptamer, N4.28g (N4) for the YPASYMQ aptamer and P10.28g (P10) for the
HPHTFLN aptamer. Each clone was amplified until the titer reached 1013 pfu/ml. The
unselected phage from the original phage display library was also amplified until the titer
reached 1013 pfu/ml for use as a control. Serial dilutions of each phage were prepared
with titers ranging from 1012 pfu/ml to 107 pfu/ml. The target control was Bovine Serum
Albumin (BSA). Figure 17 shows a 96-well ELISA plate loaded with 150 p1 of reaction
mixture, which was pipetted out of 0.5 ml microcentrifuge tubes and read at 492nm using
a Microplate Autoreader EL309. There was no reaction between phage (unselected
original M13 phage(WT), HPYTFLN aptamer (P7), YPASYMQ aptamer (N4) or
HPHTFLN aptamer (P10)) and BSA, the "pseudo-target". There also was no reaction
between unselected original M13 phage (WT) and PthA full-length protein (pthA) or the
COOH-terminal 200 aa truncated PthA GST fusion protein (NLS). HPYTFLN aptamer
(P7), YPASYMQ aptamer (N4) and HPHTFLN aptamer (P10) all reacted to PthA full-
length protein (pthA) and to the COOH-terminal 200 aa truncated PthA GST fusion
protein (NLS). The reaction of the aptamers to the full length PthA protein looked
stronger than to the truncated protein. Figure 18 is the summary of ELISA data read at
492nm. The result indicated that HPYTFLN (P7), YPASYMQ (N4) and HPHTFLN
(P10) aptamer phage all had strong binding affinity to PthA full-length protein and the
Figure 17. In vitro Binding Assay of Aptamers.
ELISA Test for different concentrations of WT (no selective random phage), P7
(HPYTFLN phage), N4 (YPASYMQ phage) and P10 (HPHTFLN phage) against
different targets including BSA (Bovine Serum Albumin), pthA (PthA full length protein)
and NLS (PthA COOH-terminal 200 aa truncated protein).
Figure 18. The in vitro binding afnity of aptamers: P7 (HPYTFLN),
N4 (YPASYMQ) and P10 (HPHTFLN).
COOH-terminal 200 aa truncated PthA protein. The binding affinity diminished with
dilution. There was not much difference in binding affinity among the three different
1012 1011 100 1109 108 107
Serial Dilution of Phage
Figure 18. The in vitro binding afnity of aptamers: P7 (TPYTFLN),
N4 (YPASYMQ) and PI 0 (HPHTFLN).
COOH-terminal 200 aa truncated PthA protein. The binding affinity diminished with
dilution. There was not much difference in binding affinity among the three different
PCR-UDG Cloning of Aptamers
Two aptamer sequences, encoding YPASYMQ and HPYTFLN, were synthesized
and amplified by PCR for UDG-cloning into pBI221 using the pYD12.9F/BamHI
fragment as the template (Figure 19, band C). PCR products were tested by 0.7%
agarose gel electrophoresis. Figure 19 band E shows the GZ1 and GZ2 primer amplified
PCR products with the sequence of YPASYMQ and Figure 19 band G shows the GZ3
and GZ4 primer amplified PCR products with the sequence of HPYTFLN. Figure 19
band D and band F are negative controls without template. The PCR products were at the
expected size of 5.8 kb. After Uracil DNA glycosylase treatment, the linear PCR
products were reclosed (Figure 20). Further testing by double restriction enzyme
digestion yielded the expected aptamer size of 45 base pairs between the Xbal site and
the BamHI sites (Figure 10). Figure 21 indicates that both pGZ7.5 and pGZ7.6 have 45
bp insertions (Figure 21, band D and band F). The 90 bp band is likely an aptamer
dimer. The insertion sequence was confirmed by DNA sequencing. Two clones were
saved for further testing. The pGZ7.5 for the YPASYMQ apatmer (Figure 20 band F)
and pGZ7.6 for the HPYTFLN aptamer (Figure 20 band M). Maps of pGZ7.5 and
pGZ7.6 are shown in Figure 22.
A B C D E F G H
Figure 19. PCR amplification of aptamers P7 and N4.
A: kDNA/HindIII fragments; B: pYD12.9F plasmid DNA, C: pYD12.9F/BamHI
template, D: pGZ1 primer + pGZ2 primer (no template), E: pGZ1 primer + pGZ2 primer
+ pYD12.9F/BamHI template, F: pGZ3 primer + pGZ4 primer (no template),
G: pGZ3 primer + pGZ4 primer + pYD12.9F/BamHI template, H: 1 Kb DNA Ladder.
ABCDEFGHI J KLMN
Figure 20. Uracil DNA glycosylase treatment of PCR products.
A: kDNA/HindIII fragments, B to G: pGZ1 primer and pGZ2 primer amplified
"YPASYMQ" clones (F clone saved as pGZ7.5), H to M: pGZ3 primer and pGZ4 primer
amplified "HPYTFLN" clones (M clone saved as pGZ7.6), N: 1 kb DNA Ladder.
A B C D E FG H
Figure 21. 15.0% polyacrylamide gel electrophoresis of apatmers.
A to C: Other PCR-UDG clones digested by Xbal+BamHI,
D: pGZ7.5+Xbal+BamHI, E: pGZ7.5+XbaI+SmaI, F: pGZ7.6+Xbal+BamHI,
G: pGZ7.6+XbaI+SmaI, H: 100 bp DNA Ladder.
Hhbl M PMI
Figure 22. Plasmid maps of pGZ7.5 and pGZ7.6.
Agrobacterium tumefaciens Mediated Transient Expression of Aptamers in Bean, Sweet
Orange and Tomato Leaves
Constructs of pGZ7.5 and pGZ7.6 cloned into Agrobacterium binary vector
(pYD40.1 and pYD40.2) are shown in Figure 23. EcoRI digestion confirmed the size of
insertion and SstI digestion indicated the orientation of inserted fragments. Four clones
were saved: pGZ8.1 carries the coding sequence for the YPASYMQ aptamer andpthA
(Figure 23, band 2), pGZ8.2 carries the coding sequence for the YPASYMQ aptamer
(Figure 23, band 17), pGZ8.3 carries the coding sequence for the HPYTFLN aptamer and
pthA (Figure 23, band 28) and pGZ8.4 carries the coding sequence for the HPYTFLN
aptamer (Figure 23, band 31). Maps of these four clones shown in Figure 24.
Figure 25 Panel A shows Agrobacterium tumefaciens GV2260 constructs after
infiltration into young California light red kidney bean leaves. Compared to the dark
brown Hypersensitive Reaction (HR) lesion of the positive control from pYD40.1
(expressing pthA), pGZ8.1 (expressing the YPASYMQ aptamer andpthA) and pGZ8.3
(expressing HPYTFLN aptamer andpthA) showed reduced HR symptoms. pGZ8.1
showed significantly reduced HR, indicating that the YPASYMQ aptamer may work
better in plant than the HPYTFLN aptamer.
Figure 25 Panel B shows Agrobacterium tumefaciens GV2260 constructs
inoculated into young sweet orange leaves. Compared to the strong canker symptom
expressed by pYD40.1 (pthA), pGZ8.1 (expressing the YPASYMQ aptamer andpthA)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
21 22 23 24 25 26 27 28 29 30 313233 343536 37 38 39 40
Figure 23. Aptamers P7 and N4 cloned in Agrobacterium binary vector pYD40.1 and
1, 21: XDNA/HindIII fragments; 20, 40: 1 kb DNA Ladder; 2 to 10: pYD40.1+pGZ7.5;
11 to 19: pYD40.2+pGZ7.5; 22 to 30: pYD40.1+pGZ7.6; 31 to 39: pYD40.2+pGZ7.6.
Each group has three different clones. Each clone has three different treatments,
such as undigested plasmid DNA, digested by EcoRI and digested by SstI.
(Save No.2 clone as pGZ8.1, No.17 clone as pGZ8.2, No.28 clone as pGZ8.3,
and No.31 clone as pGZ8.4, respectively).
Figure 24. The plasmid maps of pGZ8.1, pGZ8.2, pGZ8.3 and pGZ8.4.
Figure 25. Inoculation of Agrobacterium tumefaciens strain GV2260 constructs to
the leaves of bean (panel A), citrus (panel B) and tomato (panel C).
pYD40.1 ispthA gene only in binary vector, pYD40.2 is the binary vector, pGZ8.1 is
YPASYMQ aptamer sequence withpthA gene in binary vector, pGZ8.2 is YPASYMQ
aptamer sequence only in binary vector, pGZ8.3 is HPYTFLN aptamer sequence with
pthA gene in binary vector and pGZ8.4 is HPYTFLN aptamer sequence only in vector.
and pGZ8.3 (expressing the HPYTFLN aptamer andpthA) showed reduced canker
symptoms. Also, pGZ8.1 reduced canker symptoms more than pGZ8.3, which confirmed
the observations in bean. pYD40.2 (binary vector only), pGZ8.2 (expressing the
YPASYMQ aptamer only) and pGZ8.4 (expressing the HPYTFLN aptamer only) showed
no effects on any plants tested.
Figure 25 Panel C shows inoculation of tomato young leaves using the same
Agrobacterium constructs. There was no difference in HR symptoms between pYD40.1
(expressing pthA), pGZ8.1 (expressing the YPASYMQ aptamer andpthA) and pGZ8.3
(expressing the HPYTFLN aptamer and pthA). This result indicated that the two
aptamers had no effect on HR expression in tomato, as expected. Published reports and
our own lab results (Gabriel et al., 1996) showed that the COOH-terminal region of
members of the pthA gene family are not important for elicitation of the HR in tomato.
The most plausible explanation for this results is that both the YPASYMQ aptamer and
the HPYTFLN aptamer bind to the COOH-terminal region of PthA protein, thus, it didn't
show the effect of block PthA function.
A Phage Display Peptide Library was used to screen for 7-peptide aptamers that
specifically bind to PthA full-length protein and a PthA C-terminal 200aa truncated
protein. Two recombinant protein were constructed, pthA full length gene in pET-19b
vector for His.Tag fusion and pthA COOH terminal 600bp gene fragment in pGEX-4T-3
vector for GST fusion.
The His.Tag fusion protein was best expressed by vigorously shaking at 370C
followed by 0. mM IPTG induction at OD600=0.5, changing the temperature to 280C for
3 hr. The best condition for GST fusion protein expression was vigorously shaking at
37C in 2X YT medium till OD600=0.5, followed by adding of 0. mM IPTG and shaking
at 300C for 3 hr. The low concentration of IPTG and lower temperature induction
appeared to make the fusion proteins more soluble and more easily purified by resin
affinity purification. Use ofE. coli BL21(DE3)pLysS as a host strain significantly
enhanced the expression rate of the fusion protein compared to non-host E. coli strain
DH5a. Also, since Ampicillin and Chloramphenicol had a synergistic effect,
Carbenicillin was used to replace Ampicillin in combination with chloramphenicol for
antibiotic selection. Proteinase inhibitors such as PMSF appeared to inhibit proteinase
activity during sonication and protected the fusion protein from degradation. Nonionic
detergents such like NP-40 increased the solubility of fusion proteins. All reactions were
kept on ice to obtain the maximum yield of fusion proteins. Quick sonication on ice
helped keep a high protein yield and high-speed centrifugation was used to collect
soluble proteins. For His.Tag protein purification, the His.Bind resin needed to be
regenerated before use and well washed, charged and equilibrated. Such treatment
significantly enhanced the efficiency of the affinity column. A flow rate of about 10
column volumes per hour was required and if necessary, the pass-through was collected
and reloaded on top of the column and run again to maximize the binding of His.Tag
protein to the column. The His.Tag protein always come out in the first 5 ml of eluate.
Later eluate contained other proteins with some affinity to the His.bind resin. Protein
was dialyzed and redissovled in as small a volume as possible to increase the
For GST fusion protein, low speed (500xg) centrifugation was necessary to
sediment the Glutathione Sepharose 4B without compacting the pellet. Gentle agitation
during elution was required. The Glutathione Elution Buffer was freshly made to keep it
in a reduced state. Dialysis against 50mM NH4HCO3 is one of the modifications used for
easy purification of lyophilized proteins. After freeze drying, the NH4HCO3 evaporated
and the protein sample was dissolved in any desired salt solutions without changing of
ion strength. After eight hours of electro elution at 10mA per tube, the Protein Elution
Buffer without SDS was continued another 30 min.
There were two target proteins in Phage display biopanning, PthA full-length
His.Tag protein and PthA COOH-terminal 200aa truncated GST fusion protein. Because
the pET-19b vector only had 10 Histidine repeats, compared to the size of PthA full-
length protein (130 kDa), the His.Tag region was very small. Therefore, PthA full length
His.Tag protein could be directly used as the target for phage display biopanning. But for
the GST fusion protein, the Glutathione S-transferase itself has a molecular weight of 26
kDa, and the size of PthA COOH-terminal 200 aa peptide was 29 kDa (Figure 3, bold
letters). In this case, Glutathione S-transferase could readily interact with the truncated
PthA peptide and have some side effect. Therefore, the Glutathione S-transferase could
be separated using a cleavage enzyme and the PthA COOH-terminal peptide could be
recovered by gel purification. In this experiment, during phage display biopanning, the
amplified phage was first incubated with pure Glutathione S-transferase. Any phage
bound to GST was eliminated during this pre-absorbing treatment, then, unbounded
phage was applied to against GST fusion protein in biopanning. After three rounds of
separate selection, a fourth round of cross-selection using full length PthA was carried
out. For PthA full length protein biopanning, PthA full-length protein was used as the
target in all selections. Three different aptamer encoding sequences were found:
YPASYMQ, HPYTFLN and HPHTFLN. YPASYMQ was found 4 times out of 11
(36.4%), HPYTFLN was found 5 times out of 11 (45.5%) and HPHTFLN was found 2
times out of 11 (18.2%). In the phage population from selection against the PthA
COOH-terminal 200 aa truncated GST fusion protein, YPASYMQ was found 7 times out
of 10 (70.0%) and HPYTFLN was found 2 times out of 10 (20.0%). No HPHTFLN
sequence was found in this case. Based on previous reports (Hoess et al., 1994), some
phage containing sequences with repeats of Phenylalanine or Histidine especially could
bind to polystyrene well. One such sequence, encoding the sequence ofHYGFPPP, wass
considered to be a pseudo-binding sequence. Based on the in vitro binding affinity assay,
ELISA readings indicated HPYTFLN and HPHTFLN had very similar reactivity both to
PthA full length protein and PthA COOH-terminal 200 aa truncated GST fusion protein,
so HPHTFLN could be considered as the subgroup of HPYTFLN with only one base pair
change in the DNA sequence. It is interesting that there were only two aptamer
sequences found in two independent selective pathways, and that they were equally
distributed within two selection phage populations.
In vitro binding affinity assay was carried out by ELISA reading at 492 nm. The
optimal reading value should be in the range of 0.5 to 2.0. M13 phage itself may bind to
ELISA plates. M13 phage always showed high background readings and it was very hard
to tell the difference between the phage binding to plate or the phage binding to target.
After a couple of trials, I found that 0.5 ml microcentrifuge tubes (usually used for PCR
reaction) were very good for reducing the background of M13 in ELISA reactions. To
reduce background, one set of tubes were blocked by Blocking Buffer. The phage was
first put into only blocked empty tubes to incubate for 60 min at room temperature, then,
pre-absorbed phage was transferred to the tubes with blocked target. This step helped to
eliminate non-specific phage only binding to the tube surface. The unbound phage was
used in binding to the target. As a target control, BSA was used. YPASYMQ and
HPYTFLN didn't bind to BSA. Also, as a phage control, original unselected phage was
used. The ELISA result showed that unselected phage didn't bind to PthA full length
protein or PthA COOH-terminal 200 aa truncated GST fusion protein. Since the original
library complexity is 2 x 1013 pfu/ml, the frequency of two aptamer sequences in the
unselected original phage population is almost zero, the reaction of unselected original
phage could be considered as the background. In the ELISA results, aptamers bound to
PthA full length protein with higher affinity than bound to PthA COOH-terminal 200 aa
truncated GST fusion protein. This indicated that the aptamers interacted more with the
full-length PthA protein.
PCR amplification was carried out by a program called "touch-down". Based on
the Tm of primers (GZ1 at 580C, GZ2 at 54C, GZ3 at 580C and GZ4 at 54C), a
program was designed with annealing temperatures ranging from 56C to 500C,
dropping one degree of centigrade once in a total of seven respective cycles, followed by
a regular 30 cycles of PCR carried out at 94C for 1 min, 65 C for 1 min and 720C for 5
min. Elongation times were determined by the length of the expected PCR amplified
molecules. For an expected PCR product of 5.8 Kb, elongation time was set up at 5 min.
Uracil DNA glycosylase was used to remove Uracil. The intervening bases was then
melted off at 65 C, and the 3' overhangs reannealed (See Figure 10). The aptamers in
pBI221 were engineered to form the peptides of sequence M-YPASYMQ-GGGS-
PGGQSLM-GUS and M-HPYTFLN-GGGS-PGGQSLM-GUS. The starting methionine
seemed to have no effect on aptamer function. The Glu-Glu-Glu-Ser forms a flexible
arm to ensure that the aptamer peptide protrudes out of the protein fusion complex. The
GUS protein provides both stability to the aptamer and is useful for detection of plant
transformation. The aptamer fusion sequence was given a CaMV35S promoter and
terminated by a NOS terminator, so it could be used in plant expression (Figure 22).
The two aptamer constructs were cloned into the Agrobacterium binary vector
pYD40.1 (with pthA) and pYD 40.2 (vector only), and mated into Agrobacterium
tumefaciens GV2260 strains, then vacuum infiltrated into California light-red kidney
bean young leaves and syringe inoculated into sweet orange and tomato leaves.
pYD40.1 (pthA alone) showed very strong Hypersensitive Reaction (HR) brown lesions
on California light red kidney bean leaves, while pGZ8.1 (YPASYMQ aptamer with
pthA) and pGZ8.3 (HPYTFLN aptamer with pthA) showed significantly reduced HR
symptoms. The YPASYMQ aptamer appeared to work better than the HPYTFLN
aptamer. A similar result also showed in sweet orange inoculations. Compared to the
strong canker symptom of pYD40.1 (pthA alone), the YPASYMQ aptamer with pthA and
the HPYTFLN aptamer with pthA showed reduced canker symptoms. Also, the
YPASYMQ aptamer reduced canker symptoms more than the HPYTFLN aptamer, which
confirmed the observations in bean. The aptamers alone, when expressed in plant cells,
showed no effect. Inoculating tomato showed there is no difference between pthA alone
orpthA with aptamer which indicated the aptamer may interact the COOH terminal of
pthA gene family.
Figure 26 shows the predicted antigen index of PthA full-length protein. Figure
27 shows the predicted KD hydrophilicity and KD hydrophobicity of PthA full-length
protein. Figure 28 shows the predicted surface probability of PthA full-length protein.
Figure 26. The predicted Antigenicity Index of PthA.
Figure 27. The predicted KD Hydrophilicity and KD Hydrophobicity of PthA.
Figure 28. The predicted Surface Probability of PthA.
From all three figures, the COOH-terminal has highest chance to be exposed to the
surface of the PthA protein in the natural folded state, and the COOH-terminal also has
the strongest antigenic and hydrophilic characteristics. It is predicted that the COOH-
terminal end of PthA full length protein plays a very important functional role in protein-
protein interactions and antigenic recognition. Since the aptamers appear to interact with
the COOH-terminal of PthA and block pathogenicity, the COOH terminal end is likely to
be critical in the canker and HR signal transduction pathways.
In the future, the aptamer sequence could be used for permanent transformation of
citrus, in attempt to engineer "resistant" or "immune" trees.
SUMMARY AND CONCLUSIONS
This work was to determine if peptide aptamers could be selected that bind to
PthA. Two different proteins were expressed, PthA full-length protein (constructed in
pET-19b His.Tag fusion expression vector) and PthA COOH-terminal 200 amino acids of
PthA fused to GST (constructed in pGEX-4T-3 expression vector). Phage display
library biopanning was performed to screen 7-peptide aptamers specifically binding to
full-length PthA protein and a truncated PthA consisting of the C-terminal 200 amino
acids. From 11 random clones chosen from the phage population after four rounds of
selection against full length PthA, the sequence encoding HPYTFLN appeared 5 times
(45.5%), the sequence encoding HPHTFLN appeared twice (18.2%) and the sequence
encoding YPASYMQ appeared 4 times (36.4%). From 10 random clones chosen from
the phage population after three rounds of selection against the COOH-terminal 200 aa of
PthA and a 4th round of selection against full-length PthA, the sequence encoding
HPYTFLN appeared twice (20.0%) and the sequence encoding YPASYMQ appeared 7
times (70.0%). ELISA tests indicated that all three aptamers, YPASYMQ, HPYTFLN
or HPHTFLN, can bind to full-length PthA as well as the COOH-terminal 200 amino
acids of PthA. There were no significant differences in binding affinities among these
three aptamers. Two of the aptamer sequences were cloned into pBI221. The two
aptamer constructs were YPASYMQ (pGZ7.5) and HPYTFLN (pGZ7.6). Both were
separately cloned into Agrobacterium binary vectors with and without pthA. They were
transferred into Agrobacterium tumefaciens strain GV2260, and inoculated into
California light-red kidney bean, sweet orange and tomato leaves. The results showed
that the YPASYMQ aptamer had a strong effect to block the Hypersensitive Response
(HR) normally elicited by pthA expressed in beans and canker elicited by pthA expressed
in citrus. The HPYTFLN aptamer had a reduced, but significant effect. In tomato,
neither aptamer reduced the HR symptom elicited by expression ofpthA in cells. This
result is in consistent with other published reports that the COOH-terminal region of
other members of the pthA gene family are not important for elicitation of the HR in
tomato. These results confirmed that the aptamers exerted their effect by binding to the
COOH-terminal end of pthA, and indicated that aptamers might be used to control citrus
MEDIA AND STRAINS USED
Bacto yeast extract
pH was adjusted to 7.5 by adding IN NaOH, distilled water was added up
to the final volume of 1 liter, the medium was autoclaved before use.
NaCl (5 M)
KC1 (1 M)
MgC12 (2 M)
MgSO4 (1 M)
Distilled water was added up to the volume of 990 ml, pH was adjusted to
7.0 by adding IN NaOH, the medium was autoclaved and equilibrated to
room temperature. At this point, 10 ml of 1M glucose (18%) was added.
The final volume was 1 liter.
pH was adjusted to 7.0 by adding IN NaOH, distilled water was added up
to final volume of 1 liter, the medium was autoclaved.
pH was adjusted to 5.6 by adding IN HC1, distilled water was added up to
the final volume of 1 liter, the medium was autoclaved and equilibrated to
room temperature. At this point, the following items was added:
MES 10 mM
Acetosyringone 20 [pM
MS salt stock (10X) 100 ml
MES 10 mM
Sucrose 20 g
Acetosyringone 200 [pM
pH was adjusted to 5.6 by adding appropriate IN HC1 or IN NaOH,
distilled water was added to the final volume of 1 liter. The medium was
E. coli DH5ca (competent cells):
supE44, AlacU169 (480 lacZAM15), hsdR17, recAl, endAl, gyrA96,
E. coli BL21(DE3)pLysS (fusion protein expression host):
F-, ompT, hsdSB (rB-, mB-), gal, dcm (DE3)pLysS, Cmr 34
E. coli ER2537 (M13 phage host):
F, lacIq, A(lacZ)M15, proA+B+/JhuA2, supE, thiA(lac-proAB), A(hsdMS-
E.coli HB101/pRK2073 (mating helper):
supE44, hsdS20(rB, mB-), recA13, ara-14, proA2, lacY1, galK2, rpsL20,
xyl-5, mtl-1, Spr 50
Agrobacterium tumefaciens GV2260:
C58C1, Rif 75
PURIFICATION OF HIS.TAG PROTEIN (FULL-LENGTH PTHA)
1. pYY50.13 Plasmid Preparation:
E. coli DH5cu/pYY50.13 (pthA full-length gene in pET-19b vector) was streaked
out on LB plates with ampicillin at a final concentration of 100[pg/ml. A single colony
was used to inoculate 2 ml of LB medium with ampicillin at a final concentration of 100
[pg/ml and the inoculated culture was incubated overnight at 37C in a rotary shaker at a
speed of 220 rpm. The bacterial culture was transferred into 1.5 ml microcentrifuge
tubes and centrifuged at 5,000 rpm for 5 min. The pellets were re-suspended in 200p1 of
Buffer P1, 200 p1 of Buffer P2 was added and mixed gently, and the mixture was
incubated at room temperature for 5 min. 200 p1 of chilled Buffer P3 was then added
and mixed immediately; the mixture was incubated on ice for 15 min. The mixture was
then centrifuged at 10,000 rpm for 15 min at 40C and the clean supernatant was
transferred into a fresh 1.5 ml microcentrifuge tube. DNA was precipitated at room
temperature with 0.7 volume of isopropanol for 5 min. The mixture was centrifuged at
10,000 rpm for 15 min at 40C and the supernatant was poured off. The DNA pellet was
washed with 200 pAl of 70% ethanol and the pellet was air-dried for 10 min. pYY50.13
DNA was re-dissolved into 30 p1 of TE (pH 8.0). The quality of plasmid DNA was
tested by agarose gel electrophoresis. The concentration of pYY50.13 DNA was adjusted
to 1 [pg/pl1.
Tris-HCl (pH 8.0) 50 mM
EDTA 10 mM
NaOH 200 mM
Potassium acetate (pH 5.5) 3.0M
2. Transformation toE. coliBL21(DE3)pLysS:
E. coli BL21(DE3)pLysS competent cells were thawed on ice and of 50 p11
transferred to a 1.5 ml microcentrifuge tube and mixed with 1 p11 of pYY50.13 DNA on
ice for 10 min without shaking. The mixture was heat shocked for 1 min at 420C. The
cells were put back on ice for 5 min. One ml of SOBG medium was added, and the
mixture was incubated at 370C in a rotary shaker at a speed of 220 rpm for 45 min. 150
1il of transformed culture was spread on LB agar plates with carbenicillin at a final
concentration of 50[pg/ml and chloramphenicol at a final concentration of 34 [pg/ml, and
the plates were inverted and incubated at 370C overnight.
3. Expression ofE. coli pYY50.13/BL21(DE3)pLysS (PthA full-length protein):
A single colony from the above transformants was inoculated in 2 ml of LB broth
with carbenicillin at a final concentration of 50 [pg/ml and chloramphenicol at a final
concentration of 34 [pg/ml as a starter culture and incubated in a rotary shaker at a speed
of 220 rpm at 370C overnight. 500 pil of the overnight culture was diluted 1:100 into 50
ml of LB with 50 [pg/ml carbenicillin and 34 [pg/ml chloramphenicol in a 250 ml flask,
and continuously incubated in a rotary shaker at a speed of 220 rpm for 3 to 5 hours at
370C till OD600=0.5. For incubation, 50pl of 100mM IPTG was added to 50 ml of early-
log-phase bacterial culture (the final concentration of IPTG was 0. 1mM) and the culture
was incubated in a rotary shaker at a speed of 220 rpm at 280C for 3 hr. The bacterial
culture was centrifuged at 7,000 rpm at 40C for 10 min. The pellet was washed once with
IX Binding Buffer. The cells were re-suspended in 5 ml of IX Binding Buffer. The
suspensions were kept on ice.
IX Binding Buffer:
Imidazole 5 mM
NaCl 0.5 M
Tris-HCl (pH 7.9) 20 mM
4. His.Bind Resin Preparation:
His.Bind Resin (Novagen) was gently mixed by inversion until the slurry was
completely suspended. The slurry was transferred into a 10 ml column (1cm in
diameter). The storage buffer (with 20% ethanol) was completely removed, and the
remaining resin occupied about 1 ml of settled bed volume. The column was washed,
charged and equilibrated with 3 volumes of sterile deionized water, 5 volumes of 1X
Charge Buffer and 3 volumes of IX Binding Buffer, respectively.
IX Charge Buffer:
NiSO4 50 mM
5. PthA Cell Extract Preparation:
0.1% NP-40 and ImM PMSF was added to the chilled cell suspension in 1X
Binding Buffer. The mixture was kept on ice for 10 min. The cell mixture was sonicated
with SONICATOR (Model W-225) (% duty cycle=50, out put control=6, pulsed cycle)
for 60 seconds in a ice bath until the sonicated solution was no longer viscous. The
sonicated mixture was kept on ice. The lysate was centrifuged at 12,000 rpm for 20 min
at 40C. The supernatant was transferred to a fresh tube and filtered through a 0.45 micron
membrane. About 5 ml of prepared crude extract was collected and kept on ice until the
6. His.Tag Column Chromatography to Purify PthA:
The IX Binding Buffer was allowed to drain to the top of the column bed. 5 ml
of prepared crude extract was loaded on the top of the column and the column was
adjusted to a flow rate of about 10 column volumes per hour. When all cell extract had
passed through, the column was washed with 10 volumes of IX Binding Buffer and 6
volumes of IX Wash Buffer. The bound PthA was finally eluted with 6 volumes of 1X
Elute Buffer. The PthA eluate (about 5ml) was collected in a fresh tube which was kept
on ice. The column was cleaned with 5 volumes of 1X Strip Buffer. Finally, the column
was washed with deionized water until equilibrium was achieved.
IX Wash Buffer:
Tris-HCl (pH 7.9)
IX Elute Buffer:
Tris-HCl (pH 7.9)
1X Strip Buffer:
Tris-HCl (pH 7.9)
7. Dialysis of PthA:
The PthA column eluate (about 5 ml) was transferred into SPECTRAPOR
membrane tubing which was dialyzed against 50 mM NH4HCO3 at 40C with gentle
agitation overnight. The dialysis solution was changed in every 2 hrs.
8. Lyophilization of PthA:
The dialyzed PthA was pipetted out of the dialysis bag and put into Falcon 14 ml
centrifuge tubes. The protein sample was frozen at -70C overnight. The dialyzed PthA
was freeze-dried in a SENTRY Freeze Dryer overnight.
9. SDS-Polyacrylamide Gel Electrophoresis of PthA:
The freeze dried PthA was re-suspended in 200 ptl of IX Treatment Buffer, The
mixture was boiled for 10 min and then centrifuged at 10,000 rpm for 10 min. Only the
upper 80% supernatant of treated samples was pipetted out for electrophoresis loading.
IX Treatment Buffer:
Tris-HCl (pH 6.8) 0.125 M
DTT 0.2 M
Bromophenol Blue (pH 6.8) 0.02%
Protein samples were separated by SDS-PAGE, using an 8% separating gel and
4% stacking gel.
8% Separating gel:
distilled water 16.37 ml
1.5 M Tris-HCl (pH 8.8) 8.75 ml
30% Acrylamide/Bis 9.3 ml
10% Ammonium persulfate
4% Stacking gel:
0.5 M Tris-HCl (pH 6.8)
10% Ammonium persulfate
75 p 1
30% Acyrlamide/Bis (30%T, 2.67%C) stock:
(Distilled water was added up to 100 ml. The solution was stored at 40C
SDS-PAGE was run at 100 V for 5 hr. After gel electrophoresis, the gel slice was
soaked into 200 ml of ice-cold 0.2 M KC1. The gel slice was kept on ice for 2 min.
Proteins appeared as white pale colored bands on the gel slice. A razor blade was used
to cut the 130 kD protein band out and the recovered gel slice was chopped into small
10. Gel Elution of PthA:
The recovered gel slices were loaded into Model 422 Electro-Eluter (Bio-Rad)
assembly. The gel slices were washed with distilled water for 5 min to eliminate excess
KC1. PthA was electroeluted at 10 mA per tube for 8 hours. The Protein Elution Buffer
without SDS was changed at this point and electrophoresis was continued for an
additional 30 min to remove SDS in the protein sample.
Protein Elution Buffer:
Tris base 25 mM
Glycine 192 mM
The PthA eluate was collected in a volume of 700 pl, which was transferred into
SPECTRAPOR membrane tubing and dialyzed against 50mM NH4HCO3 at 4C
overnight with gentle agitation. The dialyzed PthA electro gel eluate was freeze-dried
11. PthA Quantification by UV Spectrophotometry:
PthA was re-dissolved in 100 pAl of 0.1M NaHCO3 (pH 8.6), and 10 pAl was
diluted 1:30 in 300 ptl of 0.1M NaHCO3 (pH 8.6). The optical density was measured in a
Gilford spectrophotometer. The OD280 was 0.3363, the OD260 was 0.3286, the A280/A260
ratio was 1.023. The concentration of electro gel purified PthA was caculated by A280 X
Correction Factor = 0.3363 X 0.81 = 0.27 mg/ml. The original concentration of PthA in
0.1M NaHCO3 (pH 8.6) was 8.1 mg/ml. PthA were stored at -200C.
PURIFICATION OF GST FUSION PROTEIN (PTHA C-TERMINAL 200AA
1. pGNLS-3-2 Plasmid Preparation:
E. coli DH5cL/pGNLS-3-2 (pthA C-terminal 600bp truncated gene fused with GST
in pGEX-4T-3 vector) plasmid DNA was prepared as described in Appendix B1.
2. Transformation toE. coli BL21(DE3)pLysS:
E. coli BL21(DE3)pLysS competent cells were prepared as described in
3. Expression of E. coli pGNLS-3-2/BL21(DE3)pLysS (PthA C-terminal 200aa
E. coli pGNLS-3-2/BL21(DE3)pLysS cells were induced as described in
Appendix B3, except that the culture medium was 2X YT medium, the cells were washed
and resuspended in 1X PBS.
iX PBS (pH 7.3):
NaCl 140 mM
KC1 2.7 mM
Na2HPO4 10 mM
KH2PO4 1.8 mM
4. Preparation of Glutathione Sepharose 4B:
The bottle of Glutathione Sepharose 4B (Pharmacia Biotech) was gently shaken
to resuspend the matrix. The matrix was washed with IX PBS. A 50% Glutathione
Sepharose 4B slurry was prepared in a bed volume of 50 pl. A micropipette was used to
remove 100 p1 of slurry and the slurry was transferred to a 1.5 ml microcentrifuge tube.
The matrix was sedimented by centrifugation at 500 x g for 5 min. The supernatant was
carefully removed and decanted. The matrix was washed again with 700 p|l of cold 1X
PBS, inverted to mix, and centrifuged at 500 x g for 5 min. Again, the supernatant was
discarded. The matrix was kept at 40C ready for use.
5. GST Fusion Cell Extract Preparation:
The truncated PthA::GST fusion protein was prepared as described in Appendix
6. Glutathione Sepharose 4B Batch Purification of GST Fusion:
One ml of the prepared cell extract was added to the Glutathione Sepharose 4B
affinity matrix in a 1.5 ml microcentrifuge tube; the tube was slowly inverted to mix with
gentle agitation at 40C for 30 min. The slurry was centrifuged at 500 x g for 5 min and
the supernatant was carefully removed. One ml of cold IX PBS was added and inverted
to mix with gentle agitation at 40C for 10 min. The mixture was centrifuged at 500 x g
for 5 min and the supernatant was carefully removed. The washing steps was repeated
twice. One ml of Glutathione Elution Buffer was added, the tube was inverted to mix
with gentle agitation at 40C for 30 min. The slurry was centrifuged at 500 x g for 10 min.
The supernatant was kept on ice as "glutathione protein elute". The rest of Glutathione
Sepharose 4B matrix should be washed by 1X PBS at least three times to clean up.
Glutathione Elution Buffer:
Reduced glutathione 10 mM
Tris-HCl (pH 8.0) 50 mM
7. Dialysis of GST Fusion Protein:
The total Glutathione eluate (about 5 ml) was transferred into SPECTRAPOR
membrane tubing and dialyzed against 50mM NH4HCO3 at 4C overnight. Fresh
solution was changed every two hours. The solution was agitated at very low speed.
8. Lyophilization of GST Fusion:
The dialyzed Glutathione eluate was freeze-dried as described in Appendix B8.
9. SDS-PAGE of GST Fusion Protein:
The freeze-dried Glutathione protein sample was electrophoresed as described in
Appendix B9. The 55 kD protein band was recovered.
10. Gel Elution of GST Fusion Protein:
The GST fusion protein was electro-eluted as described in Appendix B10.
11. Protein Quantification by UV Spectrophotometry:
The concentration of PthA C-terminal 200aa truncated protein in 0.1M NaHCO3
(pH 8.6) was 10.0 mg/ml, determined as described in Appendix Bl1. The GST fusion
protein sample was stored at -20C.
PHAGE DISPLAY PEPTIDE LIBRARY BIOPANNING SCREENING OF
Phage Display Biopanning
1. A solution of 100 [lg/ml of the target in 0.1M NaHCO3 (pH 8.6) was prepared. There
were two targets: PthA full-length protein (His.Tag) and the PthA C-terminal 200aa
truncated protein (GST fusion protein). Also, a solution of 100 [lg/ml of Glutathione S-
transferase (GST) was prepared as a pre-absorb target in 0.1 M NaHCO3 (pH 8.6).
2. 150 |pl of each target was added to two separate microtiter wells. In a separate well,
150 |pl of GST was added. The microtiter plate was swirled repeatedly until the surface
was completely wet. The plate was incubated at 40C overnight with gentle agitation
within a humidified container.
3. E. coli ER2537 stock culture was inoculated into 10 ml of LB medium, and incubated
with vigorous shaking at a speed of 220 rpm at 370C overnight. The overnight culture
was 1:100 diluted into 100 ml LB, and continuously incubated at 370C in a rotary shaker
at a speed of 220 rpm for 3 hr until the OD600 reached 0.5.
4. The coating solution from each well was poured off and the plate was firmly slapped
face-down onto a clean paper towel to remove residual solution. Each plate or well was
completely filled with Blocking Buffer and incubated at least 1 hour at 40C with gentle
NaHCO3 (pH 8.6) 0.1 M
BSA 5 mg/ml
5. The Blocking Buffer was discarded by shaking. The well was washed 6 times with
TBST at room temperature.
Tris-HCl (pH 7.5) 50 mM
NaCl 150 mM
Tween-20 0.1% (v/v)
6. 2 x 1011 M13 phage from the original library (New England BioLabs, Ph.D.-7T
Phage Display Heptapeptide Library, 2 x 1013 pfu/ml) was diluted by 1 ml of TBST. The
diluted phage were pipetted onto coated plates and rocked gently for 60 min at room
temperature. For PthA full-length protein (His.Tag) screening, the diluted phage was
directly added. For the GST fusion protein screening, the diluted phage was first added
to the GST coating well and incubated for 60 min, then the pre-absorbed phage was
transferred to the GST fusion protein coating well and incubated for another 60 min at
room temperature. Such a pre-absorbing step was proved to be very efficient to eliminate
GST non-specific binding background.
7. The non-binding phage was discarded and the plate was washed 10 times with 1X
8. The bound phage were eluted with 100 Ipl of free target solutions, i.e., 100 [lg/ml of
PthA full-length protein (His.Tag) in TBS and 100 [lg/ml of GST fusion protein in TBS,
respectively. The plates were rocked gently for 60 min at room temperature. The eluate
was transferred to a 1.5 ml microcentrifuge tube.
Tris-HCl (pH 7.5) 50 mM
NaCl 150 mM
9. One Ipl of elute phage was titered based on the following steps. The remaining phage
elute was stored at 40C.
10. The elute phage was amplified till the titer up to 2 x 1013 pfu/ml.
11. The biopanning steps were repeated twice. The only difference was changing the
Tween-20 to 0.5% in washing steps after second round phage display biopanning. For
GST fusion protein, each round of selection was pre-absorbed by GST.
1. 10 ml of LB medium was inoculated with a single colony of E. coli ER2537 and was
shaken vigorously in a rotary shaker at a speed of 220 rpm at 370C overnight. The
overnight starting culture was diluted 1:100 into 100 ml of LB medium. The dilute
bacterial culture was incubated at 370C for 3 hr until the OD600 reached 0.5.
2. One microliter of phage eluate was added into 20 ml of early-log-phase E. coli
ER2537 culture, transferred to a 250 ml flask, and incubated at 370C in a rotary shaker at
a speed of 220 rpm for 4.5 hr.
3. The culture was centrifuged at 10,000 rpm for 15 min at 40C. The upper 80%
supernatant was transferred to a fresh tube and re-centrifuged briefly.
4. The upper 80% of the supernatant was pipetted out to a fresh tube and 1/6 volume of
PEG/NaCl was added. The M13 phage was precipitated at 40C overnight.
Polyethylene glycol-8000 20% (w/v)
NaCl 2.5 M
5. The precipitation mixture was centrifuged at 10,000 rpm for 20 min at 40C. The
supernatant was discarded. The residual supernatant was carefully removed by a pipette.
6. The pellet was re-suspended in 1 ml of TBS. The suspension was transferred to a 1.5
ml microcentrifuge tube and centrifuged at 10,000 rpm for 5 min at 40C to remove the
7. The supernatant was transferred to a fresh tube and re-precipitated by adding 1/6
volume of PEG/NaC1. The precipitation mixture was incubated on ice for 60 min then
centrifuged at 10,000 rpm for 10 min at 40C. The supernatant was discarded and the
residual supernatant was carefully removed by a micropipette.
8. The pellet was resuspended in 200 Ipl of TBS with 0.02% NaN3. The suspension was
centrifuged at 10,000 rpm for 5 min to pellet any remaining insoluble matter. The
supernatant was transferred to a fresh tube. This was the amplified phage.
9. The amplified phage was titered according to the next step. All the eluate were stored
1. Ten ml of LB medium was inoculated with a single colony of E. coli ER2537 and
incubated in a rotary shaker at a speed of 220 rpm at 370C overnight. The overnight
culture was 1:100 diluted into 100 ml LB medium and continuously incubated at 370C
for 3 hr to reach OD600=0.5.
2. Agarose Top Agar was melted in a microwave oven and 3 ml of melted top agar was
dispensed into a 15ml Falcon centrifuge tubes (one tube for one phage dilution). All the
tubes were equilibrated at 450C water bath ready for use.
Agarose Top Agar:
Yeast extract 5 g
NaC 5 g
MgC12.6H20 1 g
Agarose 7 g
pH was adjusted to 7.5, H20 was added up to volume of 1 liter, the
medium was autoclaved and stored in room temperature.
3. LB agar plates were pre-warmed at 370C (one plate per phage dilution).
4. Ten-fold serial dilutions of M13 phage were prepared in LB medium. The dilution
range was from 101 to 1012
5. 200 p1 of mid-log E. coli ER2537 culture was transferred into a 1.5 ml
microcentrifuge tube (one tube for one phage dilution).
6. 10 ptl of each phage dilution was pipetted to each microcentrifuge tube with bacteria,
mixed quickly and incubated at room temperature for 1 min (The infection time was very
7. The infected cells were transferred to the 450C pre-warmed Agarose Top agar tube,
mixed well, and were immediately poured onto a 370C pre-warmed LB agar plate. Top
agar was spread evenly by tilting plate for several times.
8. The plate was cooled at room temperature for 5 min, inverted and incubated at 370C
no longer than 18 hours. Longer incubation may cause deletion of the phage sequence.
9. The plates were inspected and the phage plaques was counted. A good range of
dilution was the plate having about 100 plaques. The dilution factor was multiplied to get
phage titer in plaque forming units (pfu) per ml.
Characterization of Binding Clones
1. 10 ml of LB medium was inoculated with a single colony of E. coli ER2537 and the
inoculated culture was incubated in a rotary shaker at a speed of 220 rpm at 370C
overnight. The overnight starting culture was diluted 1:100 into 100 ml of fresh LB
medium, and continuously incubated at 370C for 3 hr to reach OD600=0.5.
2. One milliliter of early-log-phase bacterial culture was transferred into 15 ml test tubes
(one for each clone to be characterized).
3. The phage titer plate with about 100 plaques was chosen. A sterile wooden stick was
used to stab each plaque and transfer to the bacterial culture tube.
4. The test tube with bacteria and phage mixture was incubated at 370C in a rotary shaker
at a speed of 220 rpm for 4.5 hr.
5. The culture was transferred to a 1.5 ml microcentrifuge tube and centrifuged at a
speed of 10,000 rpm for 1 min. The upper 80% of the supernatant was transferred to a
fresh tube and re-centrifuged for an additional 1 min. The upper 80% of the supernatant
was transferred to another fresh tube.
6. 300 ptl of the upper 80% of the supernatant was diluted 1:1 with 100% glycerol and
stored at -200C as the phage stock. It could be re-amplified as necessary.
7. The rest of 500 p1l of upper 80% supernatant was transferred to a fresh 1.5 ml tube.
8. 200 p.1 of PEG/NaCl was added, the tube was inverted several times to mix well, and
incubated in room temperature for 10 min.
9. The mixture was centrifuged at 10,000 rpm for 10 min and the supernatant was
10. The pellet was re-suspended in 100 p1l of Iodide Buffer and 250 p1l of 95% ethanol
was added later. The mixture was incubated at room temperature for 10 min.
Tris-HCl (pH 8.0) 10 mM
EDTA 1 mM
11. The mixture was centrifuged at 10,000 rpm for 10 min at 40C. The supernatant was
discanted. The pellet was carefully washed by 70% ethanol and dried by a vacuum.
12. The pellet was re-suspended in 30 p1l of TE (pH 8.0).
13. The single-stranded M13 phage DNA was tested by 0.7% agarose gel
14. 10 p1 of each sample was enough for DNA sequencing. The DNA sequencing
primer was -28gIII (5'-GTATGGGATTTTGCTAAACAAC-3') which was synthesized
by University of Florida ICBR DNA Synthesis Core Facility.
IN VITRO BINDING AFFINITY OF APTAMERS ELISA
1. Host strain E coli ER2537 was streaked out in a LB agar plate.
2. A single colony was inoculated to 2 ml of LB medium as starting culture. The culture
tube was incubated in a rotary shaker at a speed of 220 rpm at 370C overnight.
3. The overnight starting culture was diluted 1:100 into 20 ml of LB medium in a 250 ml
flask, incubated in a rotary shaker with a shaking speed of 220 rpm at 370C for about 3
hr, until the OD600 reached 0.5 (early-log-phase).
4. 1 1l of M13 amplified phage stock (phage upper 80% supernatant 1:1 mixed with
100% glycerol, stored at -200C) was transferred to the 20ml of early-log-stage E. coli
ER2537 culture, and the inoculated culture was incubated in a rotary shaker with a
shaking speed of 220 rpm at 370C for 4.5 hr.
5. The culture was transferred to a Falcon centrifuge tube and centrifuged at 10,000 rpm
at 40C for 10 min. The upper 80% of the supernatant was transferred to a fresh tube and
re-centrifuged at 10,000 rpm at 40C for 5 min.
6. The upper 80% supernatant was pipetted out to a fresh tube and 1/6 volume of
PEG/NaCl was added to the mixture. The M13 phage was precipitated overnight at 40C.