CHAPTER 2

CONSTRUCTION OF ORFeome LIBRARY

OF M. tuberculosis H37Rv, ITS

CHARACTERIZATION AND TRANSFER

TO AN EXPRESSION VECTOR

43

2.1 INTRODUCTION Phage display (Smith et al., 1985) is a well-established and powerful approach for the presentation of recombinant polypeptides or proteins on the surface of a bacteriophage. The strength of the system emanates from the direct physical linkage between the phenotype i.e., the characteristics of displayed protein and the corresponding encapsulated DNA sequence. Phage display technology, which started with the identification of peptide epitopes recognized by monoclonal antibodies (Scott and Smith, 1990) has grown into an approach for studying ligand-receptor interactions, elucidating signal transduction pathways, delineating contact residues in interacting proteins, and isolating peptide inhibitors (Zozulya et al., 1999, Cortese et al., 1996, Rader et al., 1997). Other applications of this technology include the construction of gene/genome-fragment and cDNA libraries (Kuwabara et al., 1999, Santi et al., 2000, Santini et al., 1998) displaying virtually every possible encoded peptide/protein that can be used for identifying specific interacting sequences. The use of peptide libraries, cDNA libraries and genomic fragment libraries has shown promising results in the screening of genome-wide protein interactions. Peptide libraries represent a source of peptide ligands, which can help in identifying candidate interacting proteins in a sequenced genome based on motif similarity (Kay et al., 2000), mapping protein-protein interactions and identifying peptide recognition modules (Tong et al., 2002). Whereas cDNA libraries (Crameri and Kodzius et al., 2001) represent only genome sequences corresponding to expressed genes at a specific cellular state, fragmented genome display libraries have the potential to cover all genes, irrespective of high or low expression levels. Moreover, fragment libraries may present a better-exposed proteome as protein domains, thus avoiding expression and exposition of bulky full-sized proteins. There are reports on whole genome phage display libraries that include applications in the area of epitope screening (Fehrsen and du Plessis, 1999; Palzkill et al., 1998; Robben et al., 2002; Wilson et al., 1998), interaction with fibronectin and IgG (Jacobsson and Frykberg, 1996, 1998; Lindmark and Guss, 1999), interaction with fibrinogen (Jacobsson and Frykberg, 1996; Nilsson et al., 1998), ligand identification of the

44

yeast Abp1-SH3-domain (Fazi et al., 2002) and identification of Gal80p interacting proteins using Saccharomyces cerevisiae whole genome phage displayed library

(Hertveldt et al., 2003). Studies have also been carried out to identify immunogenic polypeptides using genomic phage display libraries (Hust et al., 2008). But a disadvantage of using randomly generated DNA fragments is the presence of non-coding stretches and non-functional inserts due to frame shifts, stop codons or incorrect orientation, increasing the required size of representative libraries. Although, this high rate of non-functional inserts may be tolerable when starting with DNA from single gene or even small gene rich genome, in which complete functional representation can be obtained with relatively small libraries but it may become impractical if using more complex DNA sources. Therefore, there is a need to filter DNA fragments encoding ORFs away from those that do not enrich the library. In this chapter, we have described a method for the construction of small

fragment (100 - 300 bp) and large fragment (300 - 800 bp) genomic libraries of

Mycobacterium tuberculosis H37Rv, a clinically significant pathogen that would serve as a useful resource to select and identify the interacting mycobacterial proteins and to delineate the interacting major epitopes recognized by monoclonal antibodies raised against mycobacterial antigens. Here, we also present a helper phage based novel method for the enrichment of ORFs within the phage display system without additional efforts like subcloning of the selected ORF to obtain a whole collection of phage displayed ORFeome (genic and non genic ORFs). The ORFeome library was characterized by selecting epitopes recognized by monoclonal antibodies specific for mycobacterial antigens. Further, we have described methods for high fidelity transfer of the cloned ORFs from the master phagemid library to any desired expression vector that would allow the expression of ORF-encoded domains or peptides with tags at either the N- or the C-terminus or into any other vector depending on the application.

45

2.2 MATERIALS AND METHODS 2.2.1 Materials

TOP10F’ strain {F' [lacIq Tn10 (tetR)] mcrA Δ(mrr-hsdRMS-mcrBC)

φ80lacZΔM15 ΔlacX74 deoR nupG recA1 araD139 Δ(ara-leu)7697 galU galK

rpsL(StrR) endA1 λ-} was obtained from Invitrogen (Life Technologies Corporation, California USA). TOP10F’ cells are MC1061 derivatives and have tetracycline, streptomycin resistance marker genes.

TG1 strain [F', traD36 proAB+ lacIq lacZΔM15] supE thi-1 Δ(lac-proAB)

Δ(mcrB-hsdSM)5, (rK-mK

-)] was obtained from New England Biolabs (Beverly, MA).

BL21 (λDE3) RIL strain {B F– ompT hsdS (rB– mB

–) dcm+ Tetr gal l (DE3) endA Hte (argU ileY leuW Camr)} was obtained from Stratagene (La Jolla, CA, USA).

pVCEPI13959 (6305 bp) is a high copy number phagemid vector having backbone derived from pUC119 plasmid and carries under the control of lac promoter operator (lacPO), within HindIII and EcoRI sites a DNA cassette comprising XbaI site, ribosome binding site (RBS), codons of Pectate lyase B (PelB) signal sequence, 2.2 Kb stuffer flanked by unique restriction sites like NheI

and Bsu36I followed by a trypsin cleavage site, full length gene III (2 – 405 residues), so that the gene cloned in place of stuffer portion can be displayed as N-terminal fusion with gene III. The vector backbone further comprises of the

ColE1 origin of replication (ori), intergenic region of phage M13 (f-ori) and β-lactamase gene as selection marker. The lacPO is preceded by a tHP transcriptional terminator from glutamine permease operon (glnHPQ) of E. coli. This vector consisted of BsaI restriction site at 5255 bp position within the β-

lactamase gene and an additional Bsu36I site at position 6301 bp between the tHP terminator and lacPO. In this chapter, this vector was used to obtain full-length gene III protein, preceded by a trypsin cleavage site.

46

pVCLDc102 (3265 bp) is a high copy number vector that contains sequence encoding for a N-terminus Strep-tag, full length gpD of lambda followed by GGSG spacer(s), a small stuffer segment with multiple cloning sites, a deca-histidine tag at the C-terminus and T7 terminator sequence , under the control of the lacPO. The backbone comprises of the ColE1 origin of replication, phage origin of replication (f-ori), and β-lactamase gene as a selection marker without any BsaI restriction site. In this chapter this vector was used as the source of the

high copy backbone without BsaI and Bsu36I restriction sites.

pVLExp4032 (6639 bp) is a T7 promoter - lac operator (T7-Lac) based

expression vector derived from pVNLEBAP1306 (Chowdhury et al., 1994) with backbone from pET11a. It consists of a DNA cassette encoding for the N-terminus Strep-tag, a unique NheI site, 1.8 Kb stuffer flanked by two

appropriately oriented BsaI(a) and (b) sites, Bsu36I site followed by C-terminus deca-histidine tag. In this chapter, this vector was used to obtain 1.8 Kb stuffer flanked by BsaI sites.

pVLExp4337 (6672 bp) is a low copy T7 promoter - lac operator (T7-lac) based expression vector containing sequence encoding N-terminus deca-histidine tag, followed by a TEV protease cleavage site, 1.8 Kb stuffer flanked by two appropriately oriented BsaI(a) and (c) sites and a translational stop. In between the functional tag / protease site there are spacer sequences of glycine and serine

residues. In this chapter, pVLExp4337 was used for transferring ORF selected fragments of M. tuberculosis for expression analysis.

Genomic DNA of M. tuberculosis H37Rv strain was procured from Prof. Anil. K. Tyagi, Department of Biochemistry, University of Delhi South Campus, New Delhi. Concentration of genomic DNA was ~ 1 mg/ml.

2.2.2.a. Construction of phagemid vector pVCEPI23961 for the making of gene fragment library

47

DNA fragment encoding a stuffer fragment flanked by BsaI(a) and (b)

restriction sites and followed by a unique Bsu36I restriction site, a deca-histidine tag (H10) and T7 terminator sequence was amplified from purified DNA of

pVLExp4032 as template using a set of three 5’ overlapping primers and one 3’ primer (Table II.1A). The 5’ primers have been designed to incorporate unique XbaI restriction site followed by ribosome binding site (RBS) and PelB signal sequence (PelBss) of Erwinia carotovora at the N-terminus to allow translocation of the gIIIp-fusion protein into the periplasm. The PCR was set up in 100 µl reaction volume containing 10 ng of template, three 5’ primers PelBss 01-51 (2 pmoles),

PelBss 02-51 (2 pmoles) and PelBss 03-51 (50 pmoles), 3’ primer T7Tn (50 pmoles)

and 3 U of Expand long template enzyme (Roche Molecular Biochemicals, Mannheim, Germany). The amplification steps involved initial heating at 95oC

for 4 min followed by 30 cycles of denaturation at 95oC for 30 sec, annealing at

55oC for 30 sec and polymerization at 68oC for 2 min 30 sec with 2 sec extension in each cycle. Final polymerization was carried out at 68oC for 6 min. PCR product was purified using QIAquick PCR purification kit (Qiagen, Hilden, Germany)

and then digested with XbaI - Bsu36I restriction enzymes. DNA encoding the full-length gene III protein (gIIIp) of filamentous phage from 2 - 405 residues, preceded by a trypsin cleavage site and a short flexible spacer segment was

obtained by digesting the phagemid vector pVCEPI13959 with Bsu36I and EcoRI

restriction enzymes in a reaction volume of 300 µl following incubation at 37oC

for 3 hrs further followed by dephosphorylation using 2 U calf intestinal

phosphatase (CIP; Roche Molecular Biochemicals, Mannheim, Germany) at 37oC

for 30 min. The vector backbone was obtained by digesting pVCLDc102 plasmid

with XbaI and EcoRI restriction enzymes in a 300 µl reaction volume as described above. The three digested DNA were extracted with equal volume of phenol-chloroform with further two extractions using equal volume of chloroform and

precipitated in presence of 3 M Sodium acetate pH 5.2 and 100 % chilled ethanol. The precipitated DNA were resuspended and subjected to 1.2 % low melting

agarose to purify desired bands of XbaI - Bsu36I digested stuffer fragment (1834

bp), Bsu36I - EcoRI digested fragment containing gIIIp with trypsin cleavage site

TABLE II.1 Sequences of primers used for the cloning of phagemid vector pVCEPI23961 and pVCEPI23964. A. Set of three 5’ overlapping primers used to amplify 1.8 kb stuffer with incorporation of PelB signal sequence (PelBss). Nucleotide written in bold represents the sequence complementary to the template in PelBss 01-51 and overlapping primers PelBss 02-51 and PelBss 03-51. Underlined sequences represents the restriction site incorporated using oligonucleotides.

B. Primers used for the amplification of stuffer to introduce modification at the 3’ BsaI site. Nucleotide written in bold and blue font color represents the changed nucleotide used to modify the respective restriction site. Underlined sequences represents the restriction site contained within the oligonucleotides. plus sign followed by a number shows the residue position of protein. a, The sequence between the first and the last codon with amino acids in single letter code in bold letters. *, Translational stop.

PRIMERS USED FOR CLONING OF PHAGEMID VECTOR

S.No. Plasmid (template)

Cloning Vector (cloning

sites)

Primers used for amplification (Sequence between first and the last codon)a

A

pVLE

xp4032

pVC

LDc1

02 (X

baI-E

coRI

) pV

CEP

I139

59 (B

su36

I-Eco

RI)

PelBss 01-51 - 5' GGCAGCCGCTGGCTTGCTGCTGCTGGCAGCTCAGCCAGCGATGGCTGAGGCTAGCGGCAGAGACCGATGACGCCGAAAAAGGCCCCA 3' A A A G L L L L A A Q P A M A E PelBss PelBss 02-51 - 5’ AGCTTGGGTCTAGATAAGGAGGAATAACATATGAAATACCTCTTGCCTACGGCAGCCGCTGGCTTGCTGCTGCTGGCAGCTCAGCC 3' M K Y L L P T A A A G L L L L A A Q Pectate Lyase signal sequence (PelBss) PelBss 03-51 - 5' AGCTTGGGTCTAGATAAGGAGG 3' T7Tn - 5’GTCGGTTGAGTCGAAGGAAA 3' A S G R D R * G G A S G T S G A Q (H)10 * (5’ GCTAGCGGCAGAGACCGATGACGCTGACGCCGAAAAAGGCCCCA--STUFFER—GGCGGCGCCTCAGGCACTAGTGGCGCGCAG(CAC)10TAA 3’)Template (+92 bp of pVLExp4032)

B pV

LExp

4032

pV

CEP

I239

61

(Nhe

I-Bsu36

I)

T7 promoter- 5' TAATACGACTCACTATAGGGAGA 3' Bsa2-31 – 5’ CACTAGTGCCTGAGGCGCCTCCAGAGACCTTAAACTC 3’ G G A S G T S G A (5’ GAGTTTAAGGTCTCTGGCGGCGCCTCAGGCACTAGTGGCGCG 3’) Template

XbaI

XbaI

48

49

and spacer (1270 bp) and XbaI - EcoRI vector backbone from pVCLDc102 (2730 bp). DNA was extracted using the QIAquick gel extraction kit (Qiagen, Hilden, Germany). A three-fragment ligation was set up that contained one mole part of vector backbone and three mole parts each of stuffer fragment and gIIIp encoding fragment in the presence of 5 U T4 DNA Ligase (5 U/µl; Roche Molecular Biochemicals, Mannheim, Germany). Various control reactions were also set up to see the ratio of vector background with one or no insert. The resultant

recombinant was named as pVCEPI23961 and characterized by sequencing of

plasmid DNA with ori2 (5’ ACCTCTGACTTGAGCGTCGA 3’), MTB81-56,

g3CTR1 (5’ CTGTAGCGCGTTTTCATCGGCAT 3’) and T7Tn (5’

GTCGGTTGAGTCGAAGGAAA 3’) primers using Big Dye terminator chemistry

and automated DNA sequencer (ABI Prism 3730). 2.2.2.b. Construction of pVCEPI23964 containing modified BsaI restriction site at the 3’ end

To construct pVCEPI23964, DNA fragment containing 1.8 Kb stuffer flanked by BsaI(a) and (b) restriction sites, a unique NheI restriction site at the 5’

end and Bsu36I at 3’ end was amplified from plasmid pVLExp4032 using 5’

primer T7 promoter and 3’ primer Bsa2-31 (Table II.1B). The 3’ primer was designed to create modification within the cleavage sequence of the BsaI(b) site located towards the deca-histidine tag at position 1826 of the template by

changing the nucleotide at 1832 position from ‘C’ to ‘A’ such that digestion of the vector with BsaI restriction enzyme would create four-base overhang of GGAG in place of GGCG as created at the 5’ end of BsaI(b) site. 100 µl PCR was set up

containing 10 ng of pVLExp4032 DNA and 50 pmoles each of 5’ primer - T7

promoter and 3’ primer Bsa2-31 using Expand HF enzyme (Roche Molecular Biochemicals, Mannheim, Germany). The amplification steps involved initial

heating at 95oC for 4 min followed by 25 cycles of denaturation at 95oC for 30 sec,

annealing at 55oC for 30 sec and polymerization at 72oC for 2 min with 2 sec extension in each cycle. Final polymerization was carried out at 72oC for 4 min.

50

The PCR product of 1917 bp was purified using QIAquick PCR purification kit,

digested with NheI and Bsu36I restriction enzymes and then purified on 1.2 % low melting agarose gel to isolate the desired band. The purified insert was then

ligated to similarly digested and dephosphorylated pVCEPI23961 vector to obtain

a recombinant named as pVCEPI23964 containing modified BsaI(c) site. The recombinants were confirmed by sequencing of plasmid DNA with 5’ primer

M13R (5' AGCGGATAACAATTTCACACAGGA 3') and 3’ primer U251 (5'

AGTTTTGTCGTCTTTCCAGACGT 3'). 2.2.3. Construction of phage displayed whole genome fragment library of M. tuberculosis H37Rv 2.2.3.1. Fragmentation of M. tuberculosis H37Rv genomic DNA Sonication was used to generate random DNA fragments in the size range 100 bp - 1200 bp. Sonication was optimized with 25 µg genomic DNA in a volume of 250 µl 0.1 X T10E1 (10 mM Tris, 1 mM EDTA, pH 8.0) contained in 1.5 ml microfuge tube for different pulses of 10 seconds each i.e., 1 x 10 sec pulse to 11 x 10 sec pulses with 2 minutes rest on ice between each successive 10 sec pulse

using microtip sonicator (W-385 Ultrasonics with microtip, Misonix Inc.). 10 µl sonicated sample was removed after each 10 sec pulse for analysis on 1.0 % TAE-agarose gel.

For preparative work, ten individual sonications (= 250 µg DNA) were carried out under the conditions optimized from analytical sonication using 25 µg of genomic DNA in a volume of 250 µl per sonication. Following sonication, the DNA was pooled, precipitated using ethanol and finally suspended at a concentration of 1 µg/µl in 0.1 X T10E1. The resuspended DNA was electrophoresed on 1.0 % TAE-Sea-Kem GTG agarose gel and gel containing DNA fragments of different sizes were excised with a consecutive difference of 100 - 150 bp. Following DNA segments were excised from the agarose gel:

51

1. 100-200 bp

2. 200-300 bp

3. 300-450 bp 4. 400-600 bp 5. 550-700 bp 6. 700-800 bp 7. 800-900 bp 8. 900-1100 bp 9. Above 1100 bp DNA was extracted from each segment using QIAquick gel extraction kit

(Qiagen, Hilden, Germany) and an aliquot of each eluate was estimated on 1.2 % TAE-agarose gel electrophoresis and nanodrop spectrophotometer (Thermo Scientific, NanoDrop products, Wilmington, USA).

2.2.3.2. Mixing of fragments in a suitable molar ratio to obtain equal representation of fragments in whole genome fragment library

Purified DNA fragments were mixed in varying ratio to obtain two

mixtures containing fragments of 100 - 300 bp and 300 - 800 bp. For mixture of

100 - 300 bp, 100 - 200 bp and 200 - 300 bp fragments were mixed in a molar ratio

of 1:1.5 (Table II.2A, MTBLIB25). Another mixture of fragments ranging from 300 to 800 bp was prepared by mixing fragments in the molar ratio as shown in (Table II.2B, MTBLIB27). 10 µg of total fragment mix was obtained for each kind of library. 10 µl of the mix was loaded on 1.2 % TAE-agarose gel to check the size distribution and concentration of the mix of fragments. The resultant mix of fragments for both the libraries (small size and large size) was handled separately thereafter.

2.2.3.3. End polishing of fragment mix to generate blunt ends

10 µg of DNA fragment mix of each library was used for end polishing with Quick Blunting Kit (New England Biolabs, Beverly, MA). This kit comprises

A. For MTBLIB25 (100 - 300 bp) Library B. For MTBLIB27 (300 - 800 bp) Library

Fragment Molar Ratio Amount in pmoles Fragment Molar Ratio Amount in

pmoles 300-450 bp 1.0 4.3

100-200 bp 1 30.0 400-600 bp 1.5 6.5 200-300 bp 1.5 45.0 600-700 bp 2.0 8.7

700-800 bp 2.5 11.0 Total (pmoles) 75.0 (10 µg) Total (pmoles) 30.0 (10 µg)

Table II.2. Molar ratio and amount (in pmoles) of different fragment segments used to prepare fragment mix of desired size range A. Analysis of MTBLIB25 library (100 - 300 bp); B. Analysis of MTBLIB27 library (300 - 800 bp) Fragments were mixed in appropriate molar ratio for two different libraries to allow unbiased representation of clones of variable sizes and to obtain a total amount of 10 µg in each library.

MIXING OF SONICATED GENOMIC DNA FRAGMENTS TO CONSTRUCT GENE FRAGMENT LIBRARY

52

53

of two enzymes T4 DNA polymerase and T4 DNA kinase, which allow the blunting and phosphorylation of the fragments. 100 µl reaction was set up

containing 10 µg of fragment mix, 200 µM dNTPs and 40 U (8 U/µl) blunting

enzyme in the presence of 1 x blunting buffer (New England Biolabs, MA) with 30 min incubation at room temperature (25oC) followed by inactivation of enzyme at 70oC for 10 min. The polished phosphorylated products were purified using QIAquick PCR purification kit and an aliquot was estimated on 1.2 % TAE-agarose gel.

2.2.3.4. Ligation of adaptors to blunted DNA fragment 2.2.3.4.a. Annealing of complementary oligonucleotides to obtain adaptors

Following fragmentation, blunt ending and kinasing of the sonicated

genomic DNA fragments, adaptor sequences were ligated to the DNA fragments. To construct the adaptors, two sense oligonucleotides BioD1 and E7 containing attB1 and attB2 attachment sites, respectively, with additional sequences, and respective complementary antisense strands D2 and E8 were synthesized and obtained from IBA GmbH, Germany in lyophilized form. Each oligonucleotide was HPLC-purified to ensure there are no contaminating or spurious sequences in the synthetic oligonucleotide preparation.

The adaptors were produced by annealing the two complementary

oligonucleotides in equal molar ratio in two separate reactions. The oligonucleotides were first dissolved in 0.1 X T10E1 (10 mM Tris, 1 mM EDTA, pH-8.0) to obtain a concentration of 1 nmole/µl. 10 nmoles of each complementary oligonucleotide BioD1 and D2 (for adaptor BioD1D2) and E7 and E8 (for adaptor E7E8) were mixed in two separate microfuge tubes containing annealing buffer (10 mM Tris-HCl, 50 mM NaCl, 10 mM MgCl2, 1 mM Dithiothreitol) in a 50 µl

reaction volume. The mix were stored as two aliquots of 3 nmoles each (15 µl x 2 aliquots) and four aliquots of 1 nmoles each (5 µl x 4 aliquots) at - 20oC. For annealing, the mix was thawed, heated at 95oC for 5 min in a thermocycler

54

(Perkin Elmer, Applied Biosystems) followed by gradual cooling (- 1oC/cycle) for 70 cycles until the oligonucleotides reached room temperature (25oC). 2.2.3.4.b. Ligation of adaptors to blunted DNA fragments

The two adaptors BioD1D2 and E7E8 were ligated in 30 fold molar excess over the inserts to prevent chimera formation of the inserts. For 10 µg sonicated

fragment mix of small library of 100 - 300 bp (equivalent to 33 pmoles), 1 nmole of each adaptor was used for ligation whereas for 10 µg sonicated fragment mix of

large library of 300 - 800 bp (equivalent to 100 pmoles), 3 nmoles of each adaptor was used. 100 µl ligation reaction was set up containing 10 µg sonicated DNA

fragments, 1 nmoles of each adaptor for small library and 3 nmoles of each adaptor for large library in presence of 40 U of T4 DNA ligase (5 U/µl, Roche Molecular Biochemicals, Mannheim, Germany) and 1 x Ligation buffer. Ligation was carried at 25oC for 16 hrs followed by purification using QIAquick PCR Purification kit to remove major excess of unligated adaptors and DNA fragments were eluted in 60 µl elution buffer (EB). 2.2.3.5. Isolation of Ligation Products and removal of excess oligonucleotides

Agarose gel electrophoresis was used to separate and isolate the adaptor ligated DNA fragments from the remaining un-ligated single adaptors or adaptor dimers. For this 60 µl adaptor-ligated DNA fragments were loaded in two wells on 1.2 % - TAE Sea-Kem low melting agarose gel. 100 bp marker (Invitrogen, Life Technologies Corporation, California, USA) was loaded in the adjacent lanes for

estimating the fragment size. Region of the gel corresponding to 168 - 368 bp for

small size library and 368 - 868 bp for the large size library was excised and double-stranded DNA of appropriate size range were extracted using QIAquick gel extraction kit in 100 µl EB. An aliquot of 2 µl was analyzed on 1.2 % TAE - agarose gel to estimate concentration.

55

2.2.3.6. Nick Repair

The remaining 98 µl of eluate containing adaptor ligated DNA fragments was treated with 24 U of BstI polymerase (8 U/µl, New England Biolabs, MA) in the presence of 250 µM dNTPs, BSA and 1 x thermopol buffer in 100 µl reaction

volume. The reaction was carried out at 65oC for 30 min followed by purification using QIAquick PCR purification kit and elution in 100 µl EB. 2.2.3.7. Isolation of single-stranded DNA (ss-DNA) using Streptavidin magnetic beads

To remove the unwanted population having same adaptors at either ends of the insert, the nick repaired adaptor ligated fragments (containing ~ 8 - 9 µg DNA) were mixed with Streptavidin magnetic beads (Dynal M-280 Streptavidin beads, 2.8 µm diameter, Invitrogen, USA). The binding capacity of these beads is reported to be 20 µg of double-stranded DNA per mg of beads, which is equivalent to 100 µl of the beads.

100 µl of beads were taken in a microfuge tube and washed thrice with 200 µl 1 x BW (Bead Wash) buffer (5 mM Tris, 0.5 mM EDTA, 1 M NaCl) by vortexing the beads in the wash buffer, capturing the beads with Magnetic Particle Concentrator (MPC), drawing the solution off from the captured beads and then repeating the process. After third wash, the beads were activated in the presence of 200 µl of 2 x BW buffer (10 mM Tris, 1 mM EDTA, 2 M NaCl) followed by addition of 100 µl nick repaired DNA fragment library containing approximately 8 - 9 µg DNA and 100 µl autoclaved double distilled water. The contents were mixed by mild vortexing and inversion, with incubation at room temperature (25oC) for 20 min to allow binding of the biotinylated DNA fragments to the streptavidin beads. The unbound DNA fragments were washed off twice with 400 µl 1 x BW buffer followed by washing twice with 200 µl autoclaved distilled water. The final water wash was removed from the bead pack using MPC and the single stranded DNA library was isolated by incubating washed beads with

56

fragments in 250 µl Melt buffer (100 mM NaCl, 125 mM NaOH) for 10 min at room temperature. This allows the denaturation of the bound double stranded DNA and release of complementary unbound non-biotinylated strand. Meanwhile, in a fresh 2.0 ml self-standing tube (Axygen Inc.), 1250 µl of buffer PB (from QIAquick PCR Purification Kit) was neutralized through the addition of 9 µl 20 % aqueous acetic acid. Using MPC, the beads in the Melt buffer were pelleted and 250 µl of supernatant (containing the single stranded DNA library) was carefully collected, transferred to the self-standing tube containing freshly prepared neutralized buffer PB and mixed. This 1500 µl neutralized single stranded library was concentrated over a single column using QIAquick PCR purification kit and eluted in 60 µl of EB pre-warmed at 55oC. The quantity of the resultant single stranded DNA library was estimated on nanodrop spectrophotometer. 2.2.3.8. Amplification of single stranded DNA library (ss-DNA library) to obtain double stranded DNA library

The single stranded DNA library was amplified to obtain double-stranded DNA using primers D1-s forward primer and E7-s reverse primer. Both these primers were obtained from IBA GmBH, Germany in lyophilized form and

carried phosphorothioate linkage up to two penultimate bases at the 3’ end.

First, a small scale analytical PCR was carried out using variable template concentrations to determine the lowest possible number of amplification rounds suitable to yield sufficient DNA for subsequent steps. Four different dilutions of ss-DNA library 1:10, 1:20, 1:40 and 1:80 were prepared serially in autoclaved double distilled water. These diluted samples were used as template and

amplified for 20, 25 and 30 cycles each. 20 µl analytical PCR was set up with 2 µl of each template dilution using 2 U of Pfu Ultra II fusion HS polymerase, 250 µM dNTPs and 10 pmoles each of forward primer D1-s and reverse primer E7-s. The PCR was performed with initial heating at 95oC for 2 min followed by different

cycles 20/25/30 of denaturation at 95oC for 30 sec, annealing at 55oC for 30 sec

57

and polymerization at 72oC for 20 sec for 100 - 300 bp library and 45 sec for 300 - 800 bp library with 2 sec extension in each cycle. The final polymerization was carried out for 10 min. To estimate the concentration of the amplified product, entire 20 µl PCR product was loaded on 1.2 % TAE-agarose gel. 2.2.3.9. Large-scale amplification of single-stranded DNA library to obtain double-stranded library

Based on the results of analytical run, two sets of 4 x 50 µl PCR were set up

with 5 µl of 1:40 diluted 100 - 300 bp and 300 - 800 bp ss-DNA libraries as template, respectively, in presence of 5 U of Pfu Ultra II fusion HS polymerase, 250 µM dNTPs and 25 pmoles of each primer D1-s and E7-s. The PCR conditions were same as described for analytical PCR with 25 cycles of denaturation,

annealing and polymerization at 72oC for 20 sec for 100 - 300 bp and 45 sec for 300 - 800 bp library with 2 sec extension in each cycle. For each library, the PCR product was pooled from four tubes and an aliquot of pooled PCR product from was analyzed on 1.2 % TAE-agarose gel. The amplified products were purified using QIAquick PCR purification kit and the eluate was loaded on 1.2 % Sea-Kem GTG agarose gel to remove the unused primers or non-specific PCR products.

The double stranded amplified library of size 168 - 368 bp and 368 - 868 bp, for MTBLIB25 and MTBLIB27 library respectively, were excised, extracted using QIAquick gel extraction kit and eluted in 100 µl EB. 2.2.3.10. T4 DNA Polymerase treatment of library to create vector compatible ends 5 µg of size selected double-stranded DNA was treated with 4.5 U of T4

DNA Polymerase (3 U/µl, New England Biolabs) in presence of 500 µM dTTP in 100 µl reaction volume at 15oC for 60 min. The T4 DNA polymerase treated fragments were purified by extracting once with equal volume of phenol, followed by one extraction with equal volume of phenol-chloroform with further two extractions using equal volume of chloroform. DNA was precipitated in

58

presence of 3 M Sodium acetate, pH 5.2 and 100 % chilled ethanol and resuspended in 60 µl 0.1 X TE followed by estimation of on 1.2 % TAE - agarose gel. 2.2.3.11. Ligation of inserts to BsaI-HF digested pVCEPI23764 2.2.3.11.a Digestion of vector pVCEPI23764 with BsaI-HF

20 µg of vector pVCEPI23764 was digested with 30 U BsaI-HF enzyme (10

U/µl, New England Biolabs) in reaction volume of 300 µl at 37oC for 2 hrs. Digested DNA was given phenol-chloroform treatment and precipitated as

described in section 2.2.2.a. Precipitated DNA was suspended in 30 µl 0.1 x TE and loaded on 1 % Sea-Kem GTG agarose to separate the desired band. Band was excised and DNA was extracted using QIAquick gel extraction kit followed by estimation on 1.2 % TAE - agarose gel electrophoresis. The linearized BsaI digested vector was stored as aliquots at – 20oC. 2.2.3.11.b Ligation and electroporation

Ligation was carried out with BsaI-HF digested pVCEPI23764 phagemid

vector using 3-fold molar excess of small size library (100 - 300 bp) or large size

library (300 - 800 bp) inserts. 10 µl analytical ligation reaction was set up

containing 100 ng of BsaI-HF digested vector, 15 ng of 150 - 350 bp T4 DNA polymerase treated inserts and 5 U T4 DNA ligase (5 U/µl, Roche Molecular Biochemicals, Mannheim, Germany). For large size library, 100 ng of digested

vector was ligated with 35 ng of 300 - 800 bp T4 DNA polymerase treated inserts.

Ligation was carried out for 16 hrs at 16oC followed by incubation for 1 hr at 37oC further followed by heat inactivation of ligase for 10 min at 65oC.

To optimize electroporation of ligated DNA, 1 µl ligated DNA was diluted

ten folds in autoclaved double distilled water and then 1 µl of diluted ligation sample was electroporated in 25 µl TOP10F’ competent cells (Efficiency 1 x

59

1010/µg) using 0.1 cm electroporation cuvettes (BIORAD, Gene Pulser cuvette, Hercule, CA, USA). Another 4 µl of undiluted ligation sample containing 40 ng of ligated DNA was electroporated in 100 µl TOP10F’ electrocompetent cells using 0.2 cm electroporation cuvettes (BIORAD, USA). After regeneration in SOC media

for 1 hr at 220 rpm, 37oC, electroporated cells were plated on LBAmp100 glu1% plates (LB agar medium containing 100 µg/ml ampicillin and 1 % w/v glucose).

The plates were incubated at 37oC for ~ 12 - 14 hrs. 2.2.3.12. PCR based analysis of library clones PCR analysis and DNA sequencing was used to screen for recombinant clones. For this, 96 randomly selected colonies were inoculated in 150 µl LB medium containing 100 µg/ml ampicillin and 1 % w/v glucose in 96-well flat bottom microtitre plate (MT-96FB, Laxbro Bio-Medical Aids Pvt. Ltd) and grown

at 37oC, 200 rpm for 3 - 4 hrs. 7.5 µl culture was diluted 20-fold in water and was

used as template in 15 µl PCR using 5’ primer M13R (5'

AGCGGATAACAATTTCACACAGGA 3') and 3’ primer U251 primer (5'

AGTTTTGTCGTCTTTCCAGACGT 3') with Expand HF enzyme. The PCR

products obtained were sequenced using M13R and U251 primers by ABI PRISM

dye terminator cycle sequencing ready reaction kit on Applied Biosystems 3730 automated sequencer. The nucleotide sequence of recombinants was analyzed using BLASTn program (NCBI, Nucleotide blast). 2.2.3.13. Large-scale library construction

To represent the whole of the 4.4 Mb genome of M. tuberculosis, genomic libraries containing fragments of different sizes were constructed by setting up

ligation reaction of 2 µg of pVCEPI23764 vector with 3-fold molar excess of the inserts.

Two sets of twenty 10 µl individual ligation reactions were set up, each

containing 100 ng BsaI-HF digested vector, 15 ng DNA fragments (100 - 300 bp)

60

or 35 ng DNA fragments (300 - 800 bp) respectively, and 5 U T4 DNA ligase (5 U/µl, Roche Diagnostics). The ligation mix was incubated similarly as described for analytical ligation. Finally, ligation mix from each of the twenty tubes was pooled (200 µl) followed by electroporation into E. coli TOP10F’ cells. A total of fifty electroporation were carried out in ten batches of five electroporations each so that the time during subsequent plating does not exceed 10 min. For each batch, 500 µl of electrocompetent TOP10F’ cells were mixed gently with 20 µl (200 ng) of the pooled ligation mixture followed by five electroporations of 100 µl cells-DNA mix using 0.2 cm electroporation cuvettes. The electroporated cells from each batch of electroporation were regenerated in 20 ml (4 ml per 100 µl

cells) SOC medium for 1 hr at 37oC, 250 rpm. Regenerated cells obtained from ten batches i.e., fifty electroporations were pooled (total 200 ml) and 2.0 ml regenerated mix was plated on each of the one hundred 150 mm LBAmp100 glu1% plates. To determine the size of primary library, an aliquot of the pooled regenerated mix was diluted appropriately and plated on 100 mm LB Amp100

glu1% plates. Following overnight incubation at 37oC the transformants from 150 mm plates (a total of 100 plates) were scraped in 200 ml 2 x YT, mixed with an equal volume of glycerol storage solution (65 % glycerol, 0.1 M Tris pH 8.0, 25 mM MgSO4) and stored as 166 aliquots of 1.2 ml each at - 80oC. An aliquot of the total pooled transformants was appropriately diluted and then plated on LB Amp100glu1% plates to determine cell count. The primary cells of the two libraries

of 100 - 300 bp and 300 - 800 bp were named as MTBLIB25C01 and MTBLIB27C01, respectively.

2.2.3.14. Phage production and purification from the primary library cells

An aliquot of pooled transformants containing 2.5 x 109 cells was diluted in

200 ml of 2 x YT glu1 % (2 x YT containing 1 % w/v glucose) to obtain OD600nm of

0.1 – 0.2 and grown at 37oC, 200 rpm, for 30 min. After 30 min of growth, 400 µl of ampicillin (50 mg/ml) was added to a concentration of 100 µg/ml followed by

growth at 37oC, 220 rpm, till OD600nm reaches 0.4-0.5. At OD600nm of 0.4, the culture

was kept for slow shaking at 100 rpm for 30 min and then 50 ml culture (~ 2.5 x

61

1010 cells) was infected with helper phage AGM13 at multiplicity of infection

(MOI) of 20 (5.0 x 1011 phages). The culture was kept at 37oC, without shaking for

30 min, followed by growth at 37oC, 100 rpm for 1 hr. The infected culture was diluted 10 folds in 450 ml of 2 x YT Amp100Kan50 (2 x YT medium containing 100

µg/ml ampicillin and 50 µg/ml kanamycin) and grown at 32oC, 220 rpm for 16 hrs. The phages were obtained from the culture supernatant, by two successive centrifugation carried out at 12,000 rpm for 10 min at 4oC, (GSA rotor Sorvall RC5B). The crude phage supernatant was stored at - 20oC as 18 aliquots of 1.0 ml each. The remaining phage supernatant was precipitated with 0.15 volumes of

16.67 % polyethylene glycol (PEG 6000-8000) and 3.3 M NaCl for 4 hrs at 4oC

followed by centrifugation at 14,000 rpm for 30 min at 4oC (SLA 1500 rotor Sorvall RC5B). The phage pellet was suspended in 6.0 ml NET buffer (100 mM NaCl, 1 mM EDTA, 10 mM Tris, pH – 8.0) and re-centrifuged at 15,000 rpm for 10 min at

4oC, (SS 34-rotor, Sorvall RC5B) to remove insoluble material. The phage particles in clear supernatant were finally purified on Sepharose 6B-CL (GE-Amersham Health Sciences, Uppsala, Sweden) packed in disposable 10 ml polypropylene columns (Pierce Biotechnology, Rockford, USA, Cat no. 29924). Before phage loading, the column was washed with 2 CV of autoclaved double distilled H2O followed by column equilibration with 2 CV NET buffer. Supernatant containing

phages was loaded on single column in two batches (each of 3 ml). In summary,

3 ml phage supernatant was loaded and flow-through was discarded. Column was washed with 0.2 ml NET buffer and phages were eluted from the column

using 3.8 ml NET buffer. Column was regenerated after washing with 4 CV NET

buffer, and then the remaining 3 ml of phage supernatant was then purified as described above. The total of 7.6 ml eluted phages were stored at - 20oC into 75 aliquots of 0.1 ml each. The phage titre was determined as transducing units (cfu) by infecting E. coli TOP10F’ cells with the appropriate phage dilutions. The phage

libraries of 100 - 300 bp and 300 - 800 bp were named as MTBLIB25P01 and MTBLIB27P01, respectively. 48 randomly selected transductants obtained after infecting E. coli TOP10F’ cells with primary phage library were used for screening by PCR analysis and DNA sequencing. Colony PCR was set up as described in

62

section 2.2.3.12 using 5’ primer M13R and 3’ primer U251 followed by sequencing and analysis of nucleotide sequence of recombinants using BLASTn program. 2.2.4 Selection of clones in Open Reading Frame (ORF) using a helper phage AGM13 2.2.4.1. Optimization of trypsin concentration for selection of clones in ORF with respect to PelB signal sequence and gIIIp For optimization of trypsin concentration, 1 x 1012 column purified phage particles of primary unselected phage library were diluted in 1 x PBS buffer followed by incubation with 0, 10, 100 and 200 µg/ml of trypsin as final

concentration (Cat no. T-8642, Sigma, St Louis, MO) in 1.0 ml total volume for 30

min at 37oC. Appropriate dilutions of trypsin untreated and treated phages were used to infect TOP10F’ cells to determine the phage titer. 48 randomly selected transductants were analyzed from each condition (different trypsin

concentrations 0, 10, 100 and 200 µg/ml treated phages) by PCR using M13R and U251 primers followed by DNA sequencing and analysis of nucleotide sequence of recombinants using BLASTn program as described above. 2.2.4.2. Selection of phages displaying ORF with respect to PelB signal sequence and gIIIp

For ORF selection, 1 x 1012 column purified phages of primary library were diluted 5-fold in 1 x PBS and then treated with 10 µg/ml of trypsin in a total

volume of 1.0 ml for 30 min at 37oC. Trypsin treated phages were then used to infect 10-fold excess of TOP10F’ cells (1 x 109 cells) of OD600nm0.4 - 0.5. The

infection was carried out at 37oC for 30 min without shaking followed by slow

shaking at 100 rpm for 30 min at 37oC. The infected cells were pelleted by centrifugation at 4,000 rpm (SIGMA centrifuge, 1-15, Germany), room

temperature (23 - 25oC) for 5 min, followed by twice washing of cell pellet with 5 ml of 2 x YT Amp100glu1% medium. The cell pellet thus obtained was resuspended

63

in 2 ml 2 x YT Amp100glu1% and then volume made up to 20 ml with 2 x YT Amp100glu1% followed by plating of suspension (1.0 ml per plate) on twenty 150

mm LBAmp100 glu1% plates and grown overnight at 37oC. The colonies

(transductants) from twenty plates were scraped in 30 ml 2 x YT and stored at – 70oC in presence of glycerol storage solution (65 % glycerol, 0.1 M Tris pH 8.0, 25

mM MgSO4) as 30 aliquots of 1.0 ml each. The transductants obtained for 100 -

300 bp and 300 - 800 bp libraries were named as MTBLIB25C02 and MTBLIB27C02, respectively. An aliquot of ORF selected transductants containing

1.0 x 1010 cells was diluted in 200 ml 2 x YT glu1 % and grown at 37oC, 200 rpm for

30 min. Following 30 min incubation, ampicillin was added to a final

concentration of 100 µg/ml and incubation was continued at 37oC, 220 rpm till OD600nm reaches 0.4-0.5. At OD600nm of 0.4 - 0.5, the culture was kept for slow

shaking at 100 rpm for 30 min and then infected with helper phage AGM13 at

MOI of 20. The culture was kept at 37oC, without shaking for 30 min, followed by

growth at 37oC, 100 rpm for 30 min. Following this, 50 ml of the culture was diluted 10 folds in 450 ml of 2 x YT Amp100Kan50 in two liter flask and grown at

32oC, 200 rpm for 16 hrs. The cell free phage supernatant was collected and

purified as described in section 2.2.3.14. The column-purified phages eluted in a final volume of 7.6 ml were stored at - 20oC as 75 aliquots of 0.1 ml each. The phage titre was determined as transducing units (cfu) by infecting E. coli TOP10F’ cells with the appropriate dilutions. These ORF enriched phage libraries of 100 -

300 bp and 300 - 800 bp were named as MTBLIB25P02 and MTBLIB27P02, respectively.

2.2.5. Characterization of ORF selected genome fragment phage display libraries 2.2.5.1 Affinity selection of phages displaying epitopes recognized by monoclonal antibodies (MAbs) against mycobacterial antigens

64

For affinity selection by panning, eight wells of maxisorp microplate strips (Nunc Inc.) were coated with 2 µg of purified MAbs 85-12, 1912 or 1905 and

1:5000 dilution of ascitic fluid of MAb U033 and incubated at 4oC for 16 hrs. The wells were blocked with 2 % BSA in PBST (PBS containing 0.1 % Tween 20) for 1

hr at 37oC. After washing three times with PBST, 1 x 1010 column purified phages of MTBLIB25 or MTBLIB27 library in PBSTB (PBST containing 1 % BSA) were

added to all wells and incubated at 37oC for 2 hrs. The unbound phages were removed by ten washes with PBST followed by three washes with PBS. All the washings were carried out in a microplate washer (Model Columbus, TECAN Austria GmbH, Grodig, Austria). The bound phages were eluted by addition of 100 µl of 0.1 N HCl (adjusted to pH 2.2 with glycine and BSA added to 1 mg/ml) to each well and incubation for 10 min at RT (25 - 27oC). The eluates from eight wells were collected and neutralized by adding 64 µl 2 M Tris solution (pH unadjusted). The phage titre for each of the eluate was determined as cfu by infecting E. coli TOP10F’ cells with appropriate phage dilutions. The remaining eluted phages (~ 1 x 107 phages) were amplified by infecting 10-fold excess of E.

coli TOP10F’ cells followed by super infection with helper phage AGM13 at MOI of 20 (Appendix III.14). The culture was diluted 10-fold in 2 x YT Amp100Kan50

media and grown at 32oC, 220 rpm for 16 hrs. The cell-free supernatant containing phages was used for the second and third round of affinity selection on immobilized monoclonal antibody. Bound phages were eluted and amplified as described above. Randomly selected transductants were analyzed by PCR and

DNA sequencing, as described in section 2.2.3.12. 2.2.6. Transfer of the ORF selected clones form MTBLIB27 library into an E. coli expression vector pVLExp4337 2.2.6.1 Insert preparation

100 µl secondary cells containing 1 x 109 ORF-selected transductants of library MTBLIB27 (MTBLIB27C02) were inoculated in 100 ml of ZYM505 medium

and grown at 37oC for 30 min. After 30 min of growth, 200 µl of ampicillin (50

65

mg/ml) was added to a final concentration of 100 µg/ml followed by overnight

growth at 37oC at 220 rpm. 4 x 1.5 ml of the overnight grown culture was used for the isolation of plasmid DNA using QIA miniprep kit (Qiagen, Germany) and estimated through agarose gel electrophoresis. The isolated plasmid was used as a template for the PCR amplification of the cloned ORFs using the two primers D1-s and E7-s. Before setting up a large-scale amplification, analytical PCR was carried out using variable concentration of template DNA to optimize the lowest possible number of amplification rounds. Two different concentrations, 5 ng and 2.5 ng, of the plasmid DNA were used as template for amplifying the ORFs using D1-s forward and E7-s reverse primer in the presence of 200 µM dNTP’s, and 1 U of Pfu Ultra II fusion HS polymerase in a 50 µl reaction volume. PCR was

performed with initial heating at 95oC for 3 min followed by variable number

(15/16/18/20) of amplification cycles, involving denaturation at 95oC for 30 sec,

annealing at 55oC for 30 sec and polymerization at 72oC for 45 sec with extension of 2 sec in each cycle. Final polymerization was done at 72oC for 5 min. An aliquot of the amplified products was electrophoresed on 1.2 % TAE-agarose gel to check the size distribution and yield obtained. Based on the analytical results, 4 x 50 µl PCR were set up using 5 ng of plasmid DNA as template per 50 µl reaction volume amplified for 15 cycles. The PCR conditions were used as described above. The amplified products from four reactions (=200 µl) were

pooled, digested with 2 U DpnI enzyme (1 U/µl, New England Biolabs) for 30

min at 37oC to remove the template DNA, and then purified using QIAquick PCR purification kit with final elution in 100 µl of EB. The amount of purified product was estimated by running on 1.2 % TAE-agarose gel. To prepare the inserts for further cloning, 5 µg of the purified PCR product was treated with 4.5 U of T4

DNA Polymerase in presence of 500 µM dTTP in a reaction volume of 100 µl at 15oC for 60 min. The treated inserts were purified using phenol - chloroform

extraction and estimated by gel electrophoresis as described in section 2.2.3.10. 2.2.6.2 Vector preparation

66

The T7 promoter based expression vector pVLExp4337 was prepared by digesting 10 µg of plasmid DNA with 20 U BsaI-HF enzyme (10 U/µl, NEB), at

37oC for 2 hrs in a reaction volume of 300 µl. The digested DNA was extracted using phenol - chloroform procedure and purified on 1.0 % Sea-Kem low melting agarose gel. The desired DNA was extracted using QIAquick gel extraction kit and eluted in 100 µl EB, the concentration was estimated through 1.2 % TAE - agarose gel electrophoresis. 2.2.6.3 Ligation and electroporation

10 µl reaction was set up containing 100 ng of BsaI-HF digested vector

pVLExp4337, 30 ng of 300 - 800 bp T4 DNA polymerase treated inserts and 5 U T4 DNA ligase. Ligation was carried out for 16 hrs at 16oC followed by 1 hr

incubation at 37oC followed by heat inactivation at 65oC for 10 min. 1 µl of the ligated DNA was diluted 10-fold in autoclaved double distilled water and then 1 µl of the diluted sample was electroporated in 25 µl electrocompetent BL21

(λDE3) RIL cells (Efficiency 7 x 109 / µg of supercoiled DNA) using 0.1 cm electroporation cuvette (BIORAD, USA). After regeneration in 1 ml SOC medium

at 37oC, 220 rpm for 1 hr, electroporated cells were plated on MDAG Amp100Cm30

plates (MDAG agar medium containing 100 µg/ml ampicillin and 30 µg/ml

chloramphenicol) and incubated overnight at 37oC.

2.2.6.4 Screening and expression of clones

96 randomly selected transformants were picked and inoculated in 150 µl

MDAG medium containing 100 µg/ml ampicillin and 30 µg/ml chloramphenicol

and grown for 3 hrs. Colony PCR was done as described in section 2.2.3.12 using

5’- T7 promoter primer and 3’- T7Tn primer with Expand HF enzyme. The amplified products were analyzed on 1.2 % - TAE agarose gel, followed by DNA sequencing with T7P and T7Tn primers and then analysis of nucleotide or

peptide sequence using Blastn/Blastp programs. For expression studies, 3 hrs

67

grown culture from the 96-well flat bottom plate was diluted 1:100 in 500 µl

ZYM5052Amp100Cm30 in 96 well 2 ml deep well plate (Whatman) followed by

growth at 30°C, 250 rpm for 20 - 22 hrs. For analysis of sample for total cell, 50 µl induced culture was mixed with 50 µl of 2 x Laemmeli sample buffer and then 10 µl sample was analyzed on 15.0 % SDS-PAG, protein bands were visualized by Coomassie blue staining.

2.3. RESULTS

The purpose of work described in this chapter relates to construction of gene fragment libraries encompassing bacterial genome, in specifically designed

phage display vectors, which in conjunction with a novel helper phage AGM13 is subjected to selection of clones that carry gene fragments in frame with the signal

sequence and gIIIp coat protein of phage M13 with purpose to select clones with open reading frame (ORF). Further, sequences flanking each insert are such that the entire ORF selected library can be transferred to another vector using different strategies where one protocol is based on classical lambda site-specific recombination based Gateway technology, and the second is an indigenously developed PCR based high efficiency restriction enzyme free cloning technology. M. tuberculosis genome has been selected as the model system. Accordingly, the results are described in four subsections:

1. Construction of genome-wide gene fragment libraries. 2. Genome-wide selection of the Open Reading Frame (ORF) using a novel

helper phage AGM13. 3. Characterization of ORF selected libraries. 4. Genome-wide transfer of ORF selected inserts from the phage display vector

to expression vector.

2.3.1. Construction of genome-wide gene fragment libraries.

This section includes designing of a phagemid vector for constructing large libraries using a restriction enzyme free ligation method along with designing of

68

adaptors which are attached on two ends of each insert so that following ORF selection, inserts can be transferred into other vector using restriction enzyme free or ligation free recombination based methods.

2.3.1.1. Construction of gIIIp based phagemid vectors

A phagemid based gIIIp display vector pVCEPI23961 was constructed to clone randomly generated gene fragments fused to gIIIp using restriction enzyme free cloning strategy developed in our laboratory and having several other features which were derived from three existing plasmid vectors used for different purposes. First, a 1.8 Kb stuffer flanked by two BsaI sites, BsaI(a) and

BsaI(b) along with other cloning sites was amplified from a template pVLExp4032 using a set of three overlapping primers for 5’ end (PelBss01-51, PelBss02-51 and

PelBss03-51) and a 3’ primer that anneals to the T7Tn sequence (Fig.II.1A). The PelBss01-51 primer anneals to the 5’ end of 1.8 Kb stuffer and adds sequence encoding part of PelBss, while the other two overlapping primers add remaining codons of PelBss, initiation codon, ribosome binding site (RBS) and XbaI

restriction site. The amplified product (1963 bp) was digested with XbaI - Bsu36I

to give 1834 bp fragment (A) comprising a 1.8 Kb stuffer segment flanked by two BsaI sites which have been designed in a manner that digestion with BsaI would result in non compatible 4-base overhangs. A 1270 bp segment (B) comprising full-length gIIIp preceded by trypsin cleavage site was obtained by digesting

another gIIIp based phage display vector, pVCEPI13959 with Bsu36I and EcoRI (Fig.II.1B). The third segment (C) comprising the vector backbone was derived

from another plasmid pVCLDc102 as 2370 bp XbaI - EcoRI fragment (Fig.II.1C) followed by its dephosphorylation. A three-fragment ligation resulted in the

plasmid pVCEPI23961. This vector carries under the lac promoter operator a gene fusion cassette containing DNA sequence encoding PelB signal sequence, a leader peptide for transport of fusion protein across the inner bacterial membrane, 1.8

Kb stuffer flanked by unique restriction sites like NheI, Bsu36I, followed by a

Fig.II.1.Schematic representation of cloning of phagemid vector pVCEPI23961. Only relevant genes and restriction sites are shown. The map is not to scale. lacPO, lac promoter and lac operator; 1.8 kbp stuffer, stuffer of 1.8 kbp; H10, sequence encoding for deca-histidine tag; T7Tn, T7 terminator; ST, strep-tag; D, recombinant lamda head protein; MCS, multiple cloning sites; pelB; pectate lyase signal sequence; 2.2 kbp stuffer, stuffer of 2.2 kbp; Tryp; trypsin cleavage site,; S, spacer; geneIII, gene III coding sequence of filamentous phage M13; Ampr, β-lactamase gene; Ori, ColE1 origin of replication; F+, phage M13 origin of replication; tHP, transcriptional terminator. (A) 1.8 Kb stuffer was amplified with set of three 5’ overlapping primers PelBss-01-51, 02-51, 03-51 and 3’ primer T7TN. The primer PelBss 01-51 and PelBss 02-51 were designed in a manner to generate an overlap of 36 bases to incorporate the Pectate Lyase signal sequence upstream of the stuffer. PelBss 03-51 primer, overlapping with 22 bases at 5’ of PelBss 02-51 was used to include the XbaI restriction site for cloning. The PCR product obtained was digested with XbaI - Bsu36I to obtain a fragment of 1834 bp. (B) pVCEPI13959 phagemid vector was digested with Bsu36I - EcoRI to obtain a band of 1270 bp. (C) Vector backbone was derived from pVCLDc102 using restriction enzyme sites XbaI - EcoRI as a fragment of 2730 bp. Three-fragment ligation was carried out to obtain a final recombinant pVCEPI23961.

CONSTRUCTION OF PHAGEMID VECTOR pVCEPI23961

NheI Bsu36 I EcoRIXbaIBsaI(b)BsaI(a)

F+Ampr

pVCEPI23961Ori

tHPlacPO geneIIIpelB TrypS1.8 kbpStuffer

KpnI

g3p

Trypsin site SpeI Bsu36I AGGTCTCTGGCGGCGCCTCAGGCACTAGTGGCGCGCCTGGATCCAAGGACATCCGTTCCGGAGGGGGCGGTACC TCCAGAGACCGCCGCGGAGTCCGTGATCACCGCGCGGACCTAGGTTCCTGTAGGCAAGGAATCCCCCGCCATGG G G A S G T S G A P G S K D I R S G G G G T stuffer

BssHII BamHI BsaI(b)

PCR with PelBss 01-51, PelBss 02-51, PelBss 03-51, T7TN

Xba I, Bsu 36I

PelBss 01-51

PelBss 03-51

PelBss 02-51

BsaI

(B) F+

Ampr

XbaI

pVCEPI13959Ori

tHPlacPO geneIIIpelB TrypS2.2 kbp stuffer

NheI Bsu36I EcoRIBsu36I

1270 bp

STryp gene III

Bsu36I EcoRI

Bsu36I, EcoRI

2730 bp

(C) pVCLDc102F+

AmprOri

lacPO T7tnST D H10MCS

XbaI EcoRI RI

XbaI, EcoRI pVCLDc102

F+AmprOri

lacPO

XbaI EcoRI RIT7Tn

tHP tHP

F+pVLExp4032

AmprOri

T7lac T7tn1.8kbp stuffer H10

BsaI(b)

1.8 Kbp stuffer T7TnH10

laci

NheIBsaI(a)

Bsu36I

1.8kbp stufferRBS pelB(A)

1834 bp

Bsu36IXbaI

stuffer

pectate lyase B signal sequence XbaI

RBS

TCTAGATAAGGAGGAATAACATATGAAATACCTCTTGCCTACGGCAGCCGCTGGCTTGCTGCTGCTGGCAGCTCAGCCGGCCATGGCTGAGGCTAGCGGCAGAGACCGAT AGATCTATTCCTCCTTATTGTATACTTTATGGAGAACGGATGCCGTCGGCGACCGAACGACGACGACCGTCGAGTCGGCCGGTACCGACTCCGATCGCCGTCTCTGGCTA M K Y L L P T A A A G L L L L A A Q P A M A E A S G R

NheI BsaI(a)

NheIBsaI(b)BsaI(a)

NheI

1963 bp

Bsu36IXbaI1.8kbp stufferRBS T7tnH10

pelB

Alkaline phosphatase

T7Tn

69

70

spacer codons (SGAPGS), trypsin cleavage site and a flexible spacer (SGGG) fused to gIIIp. The stuffer is also flanked by two appropriately oriented BsaI(a) and (b) sites, which were designed to contain different cleavage sequences, allowing directional cloning and preventing re-ligation of the cut vector. Digestion with BsaI restriction enzyme generates ends compatible to those created after T4 DNA polymerase treatment of inserts in the presence of single nucleotide dTTP. Stuffer continues into a full-length gene III protein (2 - 405 aa) such that the insert cloned in the stuffer portion can be displayed as N-terminal fusion with gene III. In between the C-terminus of stuffer and N-terminus of gene III is present a trypsin cleavage site ‘KDIR’ that plays a vital role in selection of ORFs and also allows rescue of bound phages after affinity selection. The vector backbone comprises of the ColE1 origin of replication, f-ori, a β-lactamase gene as marker for selection and tHP transcriptional terminator from glutamine permease operon (glnHPQ) of E. coli upstream of the cap-binding site of lac promoter operator.

This vector was designed to construct the whole genome fragment library

of M. tuberculosis H37Rv by cloning the inserts in place of the stuffer portion using restriction digestion free cloning strategy. Therefore, it was important to characterize the vector after digestion with BsaI to check for the ligation and transformation efficiency. For this, a test gene was cloned into the BsaI digested

pVCEPI23961 vector. The insert was prepared by amplifying the test gene using gene specific primers with 7-base extension of CGGCAGC at the 5’ end of sense primer and CGCCACC at the 5’ end of antisense primer. The amplified product was purified on column followed by treatment with T4 DNA Polymerase in the presence of dTTP to create 4-base 5’ overhang compatible with BsaI digested (a)

and (b) sites of pVCEPI23961. Ligation was carried out using three molar excess of insert over the vector in the presence of T4 DNA Ligase. Upon electroporation in TG1 cells, transformation efficiency of 5 x 106 / µg was obtained. PCR based

analysis of 48 randomly selected transformants using M13R and U251 primers revealed that ~ 94 % (45/48) were recombinants as PCR produced an 800 bp product. Amongst the rest, two clones were of wild type (produced 2 Kb

71

amplified product) whereas one gave a product of ~ 300 bp. Sequencing of the

300 bp product with M13R primer showed that it was generated due to self-ligation of the digested vector after removal of two bases each from either ends, thus making the remaining two ends compatible to each other as shown in Fig.II.2A. The false recombinants generated due to deletion and subsequent vector self-ligation consisted of a small coding sequence in frame with PelB signal sequence and gIII protein. Therefore, these spurious clones produced phages, displaying a small peptide fused to gIIIp, when rescued using a helper phage. However, later experiments showed that these clones were preferentially getting amplified during subsequent cycles of growth. Therefore, it was necessary that the overhangs created in BsaI digested vector should be modified in a manner that the removal of any number of bases of the overhang should not produce compatible ends for efficient ligation that simultaneously brings PelBss and gIIIp in frame.

To circumvent the problem of generation of clones after vector self ligation,

the BsaI(b) site at the 3’ end of the stuffer was modified to produce another

phagemid pVCEPI23964 (Fig.II.3). In this vector, after BsaI digestion 4-base 5’

overhang would be created having 5’-GCCG-3’ and 5’-GGAG-3’ ends. As shown in Fig.II.2B, removal of any number of bases from either end does not produce compatible overhangs and vector self-ligation would be possible only when all the 4 bases are removed from both the ends to produce blunt ends. But this ligation results in altered reading frame of gIIIp with respect to PelBss and hence, such clones will not produce functional wild type gIIIp to have any growth advantage. The restriction enzyme free cloning using BsaI digested vector with 4-base 5’overhangs and inserts with 4-base compatible ends generated after controlled T4 DNA polymerase treatment is a standard cloning strategy in our laboratory. In our laboratory, T7-promoter based expression vectors have been constructed where cloning is carried out using this strategy. In those vectors also, we faced the same problem due to removal of some bases of the vector

overhangs. Therefore, a vector pVLExp4037, was constructed by changing the BsaI(b) site, that produced GGCG overhang, to BsaI(c) site where enzyme

72

Fig.II.2. Schematic representation of problem associated with pVCEPI23961 and its solution. A. Deletion of two bases from the either 5’ overhangs after BsaI digestion generates compatible ends which get ligated to produce a self ligated vector that is in frame with respect to gIIIp. B. 3’ BsaI site was modified by changing the base (indicated in blue font) to construct pVCEPI23964 vector, where after digestion with BsaI, ligation is possible only after deletion of all four bases of the overhangs to create blunt ends. Even after blunt end ligation, the self ligated vector will be off frame with respect to gIIIp. Underlined and highlighted sequences represents the BsaI restriction sites contained within the vector, plus sign followed by a number shows the nucleotide position in vector. *, Translational stop.

A. PROBLEM OF SELF LIGATION ASSOCIATED WITH BsaI DIGESTED pVCEPI23961 VECTOR

ATGGCTGAGGCTAGCGGCAGAGACCGATGAACG----STUFFER----AGTTTAAGGTCTCTGGCGGCGCCTCAGGCACTAGTGGCGC TACCGACTCCGATCGCCGTCTCTGGCTACTTGC----STUFFER----TCAAATTCCAGAGACCGCCGCGGAGTCCGTGATCACCGCG M A E A S G R D R * G G A S G T S G

ATGGCTGAGGCTAG CGGCGCCTCAGGCACTAGTGGCGC TACCGACTCCGATCGC CGCGGAGTCCGTGATCACCGCG M A E A S G A S G T S G

ATGGCTGAGGCTAG GGCGGCGCCTCAGGCACTAGTGGCGC TACCGACTCCGATCGCCG CGCGGAGTCCGTGATCACCGCG M A E A S G G G A S G T S G

ATGGCTGAGGCTAG GGCGGCGCCTCAGGCACTAGTGGCGC TACCGACTCCGATCGCCG CGCGGAGTCCGTGATCACCGCG M A E A S G G G A S G T S G

Digestion with BsaI

5’ four base overhangs

Deletion of two bases from either ends

Ligation of compatible ends

ATGGCTGAGGCTAGCGGCGCCTCAGGCACTAGTGGCGC TACCGACTCCGATCGCCGCGGAGTCCGTGATCACCGCG M A E A S G A S G T S G

Reading frame is restored

+ 231 bp of pVCEPI23961 BsaI(a) BsaI(b)

B. MODIFYING 3’ BsaI SITE TO CIRCUMVENT THE PROBLEM OF SELF LIGATION

ATGGCTGAGGCTAG GGAGGCGCCTCAGGCACTAGTGGCGC TACCGACTCCGATCGCCG CGCGGAGTCCGTGATCACCGCG M A E A S G G G A S G T S G

ATGGCTGAGGCTAGCGGCAGAGACCGATGAACG----STUFFER----AGTTTAAGGTCTCTGGAGGCGCCTCAGGCACTAGTGGCGC TACCGACTCCGATCGCCGTCTCTGGCTACTTGC----STUFFER----TCAAATTCCAGAGACCTCCGCGGAGTCCGTGATCACCGCG M A E A S G R D R * G G A S G T S G

ATGGCTGAGGCTAG GGAGGCGCCTCAGGCACTAGTGGCGC TACCGACTCCGATCGCCG CGCGGAGTCCGTGATCACCGCG M A E A S G G G A S G T S G

ATGGCTGAGGCTAG GCGCCTCAGGCACTAGTGGCGC TACCGACTCCGATC CGCGGAGTCCGTGATCACCGCG M A E A S A S G T S G

Digestion with BsaI

5’ four base overhangs

Deletion of all four bases from either ends

Ligation of blunt ends

ATGGCTGAGGCTAGGCGCCTCAGGCACTAGTGGCGC TACCGACTCCGATCCGCGGAGTCCGTGATCACCGCG M A E A R R L R H * W R

Reading frame is altered

+ 231 bp of pVCEPI23964 BsaI(a) BsaI(c)

CONSTRUCTION OF NEW PHAGEMID VECTOR pVCEPI23964 WITH

MODIFIED 3’ BsaI RESTRICTION SITE

Fig.II.3 Schematic representation of cloning of phagemid vector pVCEPI23964. Only relevant genes and restriction sites are shown. The map is not to scale. T7lac, T7 promoter-lac operator; Ampr, β-lactamase gene; Ori, ColE1 origin of replication; F+, phage M13 origin of replication; H10, sequence encoding for deca-histidine tag; laci, lac repressor; lacPO, lac promoter and lac operator; T7Tn, T7 terminator; pelB; pectate lyase signal sequence; 1.8 kbp stuffer, stuffer of 1.8 kbp; Tryp; trypsin cleavage site; S, spacer; geneIII, gene III coding sequence of filamentous phage; tHP, transcriptional terminator. The 1.8 kb stuffer was amplified with specific set of primer T7 Promoter forward primer and Bsa2-31 reverse primer carrying modified BsaI restriction site and Bsu36I site for cloning. PCR product generated was digested with NheI and Bsu36I and cloned into NheI- Bsu36I digested pVCEPI23961 vector to obtain pVCEPI23964. The modified base in the cleavage sequence of BsaI (c) site has been shown in bold and blue font.

F+Ampr

pVCEPI23964Ori

tHP

BsaI(a)

lacPO geneIIIpelB TrypS1.8kbp Stuffer

BsaI(c)

T7Tn

Digested insert

PCR with 5ʼ-T7P 3ʼ- Bsa2-31

NheI, Bsu36I

Bsu36INheI

1.8kbp stufferF+

pVLExp4032AmprOri

T7lac T7tn1.8kbp stuffer H10

BsaI(a) BsaI(b)Bsu36I

laci

NheIT7P

Bsa2-31

Bsu36INheIBsaI(a)

1.8kbp stufferT7lac

BsaI(c)

Trypsin site

Stuffer

BsaI(c)

NheI, Bsu36I

F+Ampr

XbaI

pVCEPI23961Ori

tHPlacPO geneIIIpelB TrypS1.8 kbp stuffer

NheI Bsu36I EcoRI

T7Tn

Digested vector

F+AmprOri

tHP

NheI Bsu36I

pelBlacPO geneIIITrypS T7Tn

GGGAGTTTAAGGTCTCTGGAGGCGCCTCAGGCACTAGTGGCGCGCCTGGATCCAAGGACATCCGT CCCTCAAATTCCAGAGACCTCCGCGGAGTCCGTGATCACCGCGCGGACCTAGGTTCCTGTAGGCA G G A S G T S G A P G S K D I R

73

74

recognition did not change but overhang produced changed to GGAG (Fig.II.2B).

For this, the 1.8 Kb stuffer in pVLExp4032 was amplified using 5’ T7 promoter

primer and a newly designed 3’ primer Bsa2-31 that annealed to 3’ end of 1.8 Kb

stuffer, created a new cleavage sequence for BsaI {BsaI(c)} and a Bsu36I site. In fact, it changes only one base ‘C’ to ‘A’, which does not bring about any change in the reading frame or in the encoded amino acids. The amplified fragment was

digested with NheI and Bsu36I, as these are unique sites in all our vectors at proper places. The digested fragment when ligated to any of our vectors using

NheI - Bsu36I sites, converts those vector into BsaI cleavable vector that produce 4-base overhangs with new sequence that does not allow self-ligation after removal of some bases from the overhangs. Accordingly, this new insert was

cloned into NheI – Bsu36I digested pVCEPI23961 to give pVCEPI23964 (Fig.II.3). This vector was digested with BsaI and used for ligating inserts with compatible ends and no self-ligated vector was ever found. The phagemid vector

pVCEPI23964 was further modified to pVCEPI23764 to delete the NdeI site located within gIIIp (Singh A, Ph.D thesis, 2011) and this vector was used for the cloning of the gene fragments. 2.3.1.2 Construction of genome fragment library of M. tuberculosis H37Rv.

This includes generation of random fragments of M. tuberculosis genome, designing and attachment of two adaptor sequences for restriction enzyme free

cloning in gIIIp based phagemid vector, pVCEPI23764. The overall strategy of library construction has been schematically depicted in Fig.II.4.

The choice of fragment size was based on two major applications of phage display library. One application was to map the epitopes recognized by monoclonal antibodies (MAbs) raised against M. tuberculosis proteins and also to identify immunodominant regions of mycobacterial proteins against which antibodies might be present in sera of patients carrying active tuberculosis. For

this, a phage display library displaying peptides of 30 - 100 amino acids was considered sufficient to map both linear as well as small conformational epitopes.

75

Fig.II.4 Strategy for the construction of M. tuberculosis genome fragment library in phagemid based phage display vector I. Random fragmentation, adaptor ligation, isolation of single stranded DNA library using Streptavidin magnetic beads and PCR amplification of ss-DNA library to obtain double stranded DNA library. The white arrows within blue boxes indicate the orientation of the insert. The fragments shown within pink and green boxes with dashed lines show the desired kind of fragments with two different adaptors on each end.

M.tb genomic DNA

Nick Repair by BstI Polymerase

Streptavidin Magnetic Beads

Fragmentation by sonication

Blunting of fragments

Ligation of adaptors

E7 E8

D1 D2

2. Washing 3. Alkaline Denaturation

ss-lib amplified by D1-s & E7-s

ds-amplified product

Streptavidin Magnetic Beads

Biotin Tag

Nick

(A) Four possibilities of adaptor ligation

(B) Desired adaptor ligated fragments

1. Magnetic Particle Concentrator (MPC)

4. MPC

5. Collect supernatant

I. Tw

o -fo

ld e

nric

hmen

t of a

dapt

or li

gate

d fr

agm

ents

76

Fig.II.4 Strategy for the construction of M. tuberculosis genome fragment library in phagemid based phage display vector. II. Ligation of inserts to the phagemid vector and production of primary phage library rescued using helper phage AGM13. III. ORF selection of the primary phage library using AGM13 helper phage to obtain enriched secondary phage library.

II.

Phag

e Pr

oduc

tion

Kanr

Helper phage

Phagemid vector Ampr

g3

Phagemid vector

Ampr Phagemid Library

Phages displaying gIII-p fusion

gIII-p based Phagemid vector Ligation

Infection by Helper phage AGM13

T4 treatment + dTTP

Double stranded Library

Onl

y C

pop

ulat

ion

is

infe

ctiv

e.

A &

B p

opul

atio

n is

no

n-in

fect

ive.

II

I. O

RF S

elec

tion

A

B

C

gIIIp from helper phage genome

Recombinant gIIIp from phagemid genome

CT N2 N1

TRYPSIN Cleavage SITE

TRYPSIN

TREATMENT

CT N2 N1

TRYPSIN

77

The other application relates to functional genomics to study protein-protein interactions or protein-small molecule interactions. For this purpose, it would be difficult to display the entire proteome of M. tuberculosis as intact functional proteins due to very large range of sizes of functional proteins [e.g. 80 amino acid

for ESAT6 (Rv3875) to 3300 amino acid for PPE (Rv0355c)], it was conceptualized that for most interactions or binding functions a complete domain might be sufficient. Domain can have one function and a combination of several domains or even one domain produces a functional protein. The domains are generally constituted of 100 - 250 amino acids. Therefore, a phage display library displaying 100 - 250 amino acids was considered. Thus, for constructing two libraries, gene fragments of size 100 - 800 bp were required. Since ligation efficiency of an insert is dependent on the its size such that smaller fragment ligates more efficiently, fragment of different sizes were first prepared and then mixed in some arbitrary molar ratios with rationale that ligation is size dependent to obtain a library containing fragment with uniform distribution of all sizes. The fragmentation of M. tuberculosis genomic DNA was achieved by sonication. First, 25 µg genomic DNA in 250 µl volume was sonicated and an aliquot containing 1 µg DNA was removed every 10 sec with a total of 11 samples being collected and analyzed on agarose gel to identify sonication conditions that produced majority of the fragments in the range of 100 - 1000 bp (Fig.II.5A). It was observed that with every 10 sec sonication cycle there was a progressive decrease in both lower and the upper limit of size of fragments for initial periods but after 5 x 10 sec sonication the lower limit of fragment size was fixed at about 100 bp while the upper limit was about 1000 bp with maximum at 600 bp. Further, sonication did not change the lower limit but the upper limit was reduced to 600 bp with maximum being around 400 bp after 10-11 x 10 sec cycles of sonication. Therefore, for preparative sonication, 5 x 10 sec sonication cycle was selected and without changing the probe or volume, sonication of 10 x 25 µg (250 µl each) genomic DNA was performed and pooled. The preparative sonication resulted in fragmented DNA of size range 100 bp - 1200 bp with maxima being around 500 - 600 bp (Fig.II.5B). The preparative fragmented DNA

Fig.II.5. Sonication of genomic DNA of M. tuberculosis H37Rv A. DNA fragments obtained after analytical sonication of M. tuberculosis whole genome for different time were electrophoresed on 1.0 % TAE-ethidium bromide agarose gel. Lane M, 1 Kb DNA marker (Invitrogen); Lane 1, 1 µg unsonicated genomic DNA obtained at 0 sec; Lane 2-11, 1 µg sonicated genomic DNA electrophoresed after every 1x10 sec to 11x10 sec pulse, respectively. B. DNA fragments were electrophoresed after prep sonication of 250 µg genomic DNA for 5x10 sec pulse to obtain desired range of fragments from 100-1200 bp. Lane M1, 1 Kb marker (Invitrogen), Lane 1, 10 µl sample loaded after prep sonication for 5x10 sec pulse, Lane M2, 100 bp marker (Invitrogen).

ANALYTICAL & LARGE-SCALE SONICATION OF GENOMIC DNA OF M. tuberculosis

78

ESTIMATION OF DIFFERENT SEGMENTS OF SONICATED GENOMIC DNA OF

M. tuberculosis

Segment Fragment Size Range (bp)

Concentration (ng/µl)a

Total volume

Total Amount (µg)

Amount in pmoles

1 100-200 50 180 µl 9.0 90.9 2 200-300 92 180 µl 16.6 100.4 3 300-450 95 180 µl 17.1 69.1 4 400-600 120 180 µl 21.6 65.4 5 550-700 104 180 µl 18.7 45.4 6 700-800 110 180 µl 19.8 40.0 7 800-900 100 180 µl 18.0 32.1 8 900-1100 102 180 µl 18.4 27.8 9 Above 1100 97 180 µl 17.5 22.1 Total 156.6

a concentration as estimated on Nanodrop 1000 Spectrophotometer (Thermo Scientific) Table II.3. Range and yield of different segments of DNA obtained after large-scale sonication and gel extraction of M. tuberculosis H37Rv genomic DNA. 250 µg of genomic DNA of M. tuberculosis H37Rv was sonicated for 5x10 sec pulse and loaded on 1 % Sea-Kem to excise nine segments of different size range followed by analysis of extracted DNA by gel electrophoresis and estimation on Nanodrop 1000 Spectrophotometer.

Fig.II.6. DNA fragments obtained after prep sonication and isolation of DNA from different segments were electrophoresed on 1.2 % TAE-ethidium bromide agarose gel. Lane M, 100 bp marker (Invitrogen); Lane 1, 100-200 bp fragments; Lane 2, 200-300 bp fragments; Lane 3, 300-450 bp fragments; Lane 4, 400-600 bp; Lane 5, 550-700 bp; Lane 6, 700-800 bp; Lane 7, 800-900 bp; Lane 8, 900-1100 bp; Lane 9, above 1100 bp.

bp

500 400 300

100 200

600 800

1500 1000

M1

1

2 3 4

5 6 7

8 M 9

bp

500 400 300

100 200

600 700 800

M 1

2

Fig.II.7. Mix of different size fragments of sonicated DNA to obtain desired size range. 1 µg mix was electrophoresed on 1.2 % TAE-ethidium bromide agarose gel. Lane M, 100 bp marker (Invitrogen); Lane 1, 100 - 300 bp fragments; Lane 2, 300 - 800 bp fragments.

MIXTURES OF SONICATED DNA FRAGMENTS FOR CONSTRUCTION OF LIBRARIES

79

80

was resolved in 1.2 % agarose gel and nine gel segments were excised to contain fragments of different sizes within a range of 100 - 150 bp, the smallest being 100 - 200 bp and the two large ones comprised fragments of 900 - 1000 bp and above 1100 bp, respectively. The DNA from each gel segment was extracted using QIAquick gel extraction kit, estimated by nanodrop spectrophotometer and an aliquot of each eluted fragment was also run on agarose gel (Fig.II.6 and Table II.3). The quantification showed that maximum amount of DNA was obtained for

300 - 900 bp fragments but 100 - 200 bp and 200 - 300 bp were present in higher mole amounts due to small sizes. Following this, the fragments were mixed to

produce two mixtures, 100 - 300 bp and 300 - 800 bp. For 100 - 300 bp, fragments

of 200 - 300 bp were added in 1.5 times molar excess over 100 - 200 bp fragments to ensure that there were more fragments of larger size during ligation and the ligated DNA carried equal molar distribution of different size fragments. For

preparing fragment mix of, 300 - 800 bp, the fragments 700 - 800 bp were added in

2.5 mole excess over the concentration of 300 - 450 bp fragments (Table II.2A 7

2B). These mixing ratios were totally arbitrary and finally mixture of fragments

containing 10 µg DNA was obtained for each mix. In terms of mole, 100 - 300 bp

fragment mix contained 75 pmole and 300 - 800 bp fragment mix contained 30 pmoles. An aliquot of both mixtures was analyzed on agarose gel to show that each mixture contained fragments of desired sizes (Fig.II.7). 2.3.1.3. Design of adaptors and process of producing ligation compatible inserts The next step was to clone these fragments into the phage display vector,

pVCEPI23764. Since, the ligation of blunt ended fragments is inefficient, and there would be possibility of insert chimerization, we developed a new method to first ligate two adaptors on either ends of the inserts following through a process which allows selection of inserts having two different adaptors on each end. Further, for cloning of inserts in a directional manner the adaptors contained sequences that would produce 4-base overhangs upon T4 DNA polymerase treatment in the presence of dTTP so that insert can anneal with BsaI digested phage display vector with high efficiency. The adaptors were designed to carry

81

attB1 and attB2 recombination sites used in Gateway recombinational cloning. In addition, the two adaptors in total comprised of priming sequence so that two primers could be used to amplify cloned inserts from the entire library and then transfer into another vector following Gateway recombination technology or by our proprietary restriction free enzyme cloning strategy. For converting blunt ended gene fragments into fragments with cohesive ends, two adaptors were designed having features as mentioned above. One adaptor (BioDID2) was

synthesized by annealing oligonucleotides BioD1 of 34 bases and D2 of 30 bases

such that the adaptor has blunt end at the 3’ end with respect to BioD1 and a 4-base overhang CGGC at the 5’ end, which also carries a biotin residue (Table II.4A & 4B). The BioD1D2 carries attB1 site, a 21 bp recombination sequence employed in Gateway cloning (Table II.4C, in bold letters). Similarly, the second adaptor

E7E8 was produced by annealing 34 base long E7 and 30 base long E8

oligonucleotides such that it has blunt end at 3’ end and a 4-base overhang CTCC at the 5’ end with respect to oligonucleotide E7 (Table II.4D, 4E). The adaptor E7E8 encompasses 21 bases long attB2 site of Gateway recombination (Table II.4F, in bold letters). The sequences of two adaptors are quite similar but yet different and once a part of the insert in the library, it would be possible to amplify DNA fragments flanked by these two adaptors using two specifically designed primers. For adaptor ligation, the two mixtures of DNA fragments were first treated with Quick blunting enzyme (comprising of T4 DNA Polymerase and T4 DNA Kinase, NEB) to convert frayed ends into blunt ends and phosphorylate 5’ end of each fragment. Following purification, blunt ended and phosphorylated inserts were

ligated with 30-fold molar excess of a mixture of two adaptors, BioD1D2 and E7E8. Excess of adaptors, which do not carry 5’- phosphate ensures that blunt ended inserts do not chimerize and each insert ligates to two adaptors. But, there is no control as to which adaptor/s would ligate to an insert and thus, there would be following four possibilities: (i) BioD1D2 - insert - E7E8 (ii) E7E8 - insert - BioD1D2 (iii) BioD1D2 - insert - BioD1D2 (iv) E7E8 - insert - E7E8

S.No. Oligo ID Size (bp) Features Oligonucleotides used for adaptor synthesis

A BioD1 34

5’ Biotinylated with attB1 Recombination

site

B O B’ 5’BIO CGGCAGCACAAGTTTGTACAAAAAAGCAGGCTCT 3’ G S T S L Y K K A G S

B D2 30

Complementary to BioD1 with 4 base short at the 5’ end

5’AGAGCCTGCTTTTTTGTACAAACTTGTGCT 3’

C Annealed BioD1D2

B O B’ 5’BIO CGGCAGCACAAGTTTGTACAAAAAAGCAGGCTCT 3’ 3’ TCGTGTTCAAACATGTTTTTTCGTCGCAGA 5’

D E7 34

attB2 recombination site

5’CTCCACCGACCACTTTGTACAAGAAAGCTGATCC 3’

E E8 30

Complementary to

E7 with 4 base short at the 5’ end

B’ O B 5’GGATCAGCTTTCTTGTACAAAGTGGTCGGT 3’ G S A F L Y K V V G

F

Annealed E7E8

B’ O B 5’GGATCAGCTTTCTTGTACAAAGTGGTCGGT 3’ 3’CCTAGTCGAAAGAACATGTTTCACCAGCCACCTC 5’

Table II.4 Sequences of complementary oligonucleotides used for the making the two adaptor sets Bio- Biotin tag present at the 5’ end of the oligonucleotide BioD1. Bases in bold and underlined indicate the attB recombination site, attB1 in BioD1 and attB2 in E7 where O is the core sequence, B and B’ are the two flanking arms, highlighted in grey. Changes in the attB sequence with respect to wild type attB are shown in red. Bases in Blue indicate the additional 7-bases required to generate BsaI compatible ends after T4 DNA Polymerase treatment in presence of dTTP.

OLIGONUCLEOTIDE SEQUENCES TO SYNTHESIZE ADAPTORS

82

83

Thus, only two of the possibilities (i and ii) produced adaptor-insert, which has two different adaptors at each end of the insert and can be ligated in a directional manner into the vector. Therefore, the following procedure was followed to eliminate ligated products (iii) and (iv).

First, unincorporated excess of adaptors were removed by QIAquick PCR

purification kit using PB1 buffer (containing guanidine hydrochloride and isopropanol), which allows removal of DNA fragments of less than 100 bp. Finally, agarose gel electrophoresis was used to remove traces of adaptors and

other small size contaminants. After ligation of 34 bp adaptors at two ends, the

size of fragments in each library increased by 68 bp thus, fragments of 168 - 368

bp and 368 - 868 bp (approximate) were excised from the gel and DNA was extracted using QIAquick gel extraction kit.

Since the adaptors were not phosphorylated, there is a nick between the adaptor and the insert at each end. For nick repair, inserts were treated with BstI DNA polymerase in the presence of dNTPs. BstI DNA polymerase displaces the unligated strand of adaptor BioD1D2 (oligonucleotide D2) and adaptor E7E8 (oligonucleotide E8) and extends insert to synthesize new strand complementary to D1 and E7 to produce four kinds of blunt ended inserts, where two kinds of inserts carry BioD1D2 only at one of the end (i and ii), one kind contains BioD1D2 at both the ends (iii) and one kind contains no BioD1D2 (iv). The latter two, were eliminated using Streptavidin (Sv) coated magnetic beads. For this, nick-repaired mixture of adaptor ligated DNA fragments was mixed with Sv magnetic beads to capture any DNA fragment carrying a biotin at 5’ end. This process eliminated inserts ligated with E7E8 on both ends (no BioD1D2; iv). Following this, the magnetic beads were treated with high pH (alkaline conditions). This process denatured the duplex DNA bound to magnetic beads through the biotin residue at the 5’ end and consequently, the complementary unbiotinylated strand of inserts with only BioD1D2 at one end was released in the supernatant (i and ii).

While, both the strands of inserts carrying BioD1D2 at either ends remained bound to Sv beads. The supernatant was collected which contained

84

single stranded DNA, flanked by the sequences complementary to BioD1 and E8 on either ends. The single stranded library was neutralized with acetic acid and purified on Qia column. The concentration of single stranded DNA obtained

from 100 - 300 bp and 300 - 800 bp libraries was estimated to be 500 ng (5 ng/µl) and 520 (5.2 ng/µl), respectively using nanodrop spectrophotometer.

Each single stranded DNA molecule in both libraries was now flanked by

two different sequences, which were complementary to D1 and E8 adaptor sequences. As mentioned before, these two sequences are quite similar especially towards the junctions with the insert. Therefore, two complementary primers, D1-s and E7-s (Table II.5) were designed to contain phosphorothioate bonds at the

last two penultimate bases of each of 34 base primers, which were actually same

as BioD1 and E7 but contained phosphorothioate bonds at the 3’ end and did not contain biotin. The presence of phosphorothioate bond would prevent removal

of bases from the 3’ end of each primer by the exonuclease activity (3’ to 5’) of Pfu Ultra II fusion HS DNA polymerase, so that they did not prime non-specifically.

Using primers D1-s and E7-s, amplification conditions were optimized to

obtain enough yield of double stranded DNA from single stranded DNA in minimum number of cycles and without altering the range of fragments. Pfu Ultra II fusion HS DNA polymerase was used to ensure minimal PCR generated mutations. Different dilutions of single stranded DNA (1:10, 1:20, 1:40 and 1:80)

were subjected to PCR for 20, 25 and 30 amplification cycles. Agarose gel analysis showed that PCR conditions of 25 cycles with 1:40 dilution of single stranded

DNA as template produced required distribution of fragments (160 - 360 bp for

100 - 300 bp library and 360 - 860 bp for 300 - 800 bp library) with good yield of DNA (~ 50 ng/µl) (Fig.II.8A and B; lane 8). Other PCR conditions either produced very low amplification or fragments of smaller or larger size than expected, which might have arisen due to recombination within fragments or due to chimerization, respectively.

85

Oligo ID

Size (bp) Primers used for amplification of single stranded library

D1-s 34

5’CGGCAGCACAAGTTTGTACAAAAAAGCAGGCTsCsT 3’ D2 E7 3’GCCGTCGTGTTCAAACATGTTTTTTCGTCCGAGA-NNN-CCTAGTCGAAAGAACATGTTTCACCAGCCACCTC 5’ ss - DNA library as template for D1-s primer complementary to D2 sequence to obtain double stranded library (ds - library*)

E7-s 34

5’CTCCACCGACCACTTTGTACAAGAAAGCTGATsCsC 3’ E8 D1 3’GAGGTGGCTGGTGAAACATGTTCTTTCGACTAGG-NNN-TCTCGGACGAAAAAACATGTTTGAACACGACGGC 5’ 5’CTCCACCGACCACTTTGTACAAGAAAGCTGATCC-NNN-AGAGCCTGCTTTTTTGTACAAACTTGTGCTGCCG 3’ E7 D2 ds - library* serves as template for E7-s primer complementary to E8 sequence in subsequent rounds of amplification

TABLE II.5. Primers to amplify the single stranded DNA library s denotes the two phosphorothioated bonds present at the 3’- terminal of both the primers up to two bases. The template sequence is shown below the primer sequence with complementary sequence shown in bold.

PRIMERS TO AMPLIFY SINGLE-STRANDED LIBRARY

Fig.II.8. Analytical PCR of single-stranded DNA fragment library using different template concentration and varied number of cycles DNA fragments obtained after analytical amplification of single stranded DNA library (A) 100-300 bp and (B) 300-800 bp by D1-s forward and E7-s reverse primers using different template concentration and different number of cycles. 20 µl sample was electrophoresed on 1.2 % TAE-ethidium bromide agarose gel. Lane M, 100 bp marker; Lane 1-3, 1:10 dilution of template amplified for 20, 25, 30 cycles, respectively; Lane 4-6, 1:20 dilution of template amplified for 20, 25, 30 cycles, respectively; Lane 7-9, 1:40 dilution of template amplified for 20, 25, 30 cycles, respectively; Lane 10-12, 1:80 dilution of template amplified for 20, 25, 30 cycles, respectively.

OPTIMIZATION OF AMPLIFICATION OF SINGLE STRANDED LIBRARY USING D1-s & E7-s PRIMERS

A. 12 M1

1

2 3 4

5 6 7

8 9 10 11

500 400 300

100 200

600

1500 bp

bp B.

M1

1

2 3 4

5 6 7

8 9 10 11 12

500 400 300

100 200

600

1500

86

Finally, using appropriate conditions, approximately 10 µg of double stranded DNA was obtained for each library. Each insert was flanked by two adaptors, which carry a sequence of 5’ CGGCAGC followed by attB1 site or 5’ CTCCACC followed by attB2 site, respectively. Upon treatment of the inserts with T4 DNA polymerase in the presence of dTTP nucleotide, four bases of the complementary DNA would be removed thus, leaving each fragment flanked by 4-base 5’- overhangs, 5’- CGGC and 5’- CTCC which are complementary to two ends of BsaI

digested phage display vector, pVCEPI23764. 5’- CGGC is complementary to BsaI(a) end and 5’- CTCC is complementary to BsaI(c) end. Since, the two ends of inserts or vector are not compatible with itself hence, neither vector can self ligate nor insert can concatamerise, and only the insert and vector would ligate in directional manner where 5’- CTCC end is towards gIIIp. However, due to random fragmentation the insert could be in any frame or any orientation, thus only 1 of 18 possibilities would match with the protein encoded by the source DNA (genic sequence).

Accordingly, using 2 µg of BsaI digested pVCEPI23764 with 3-fold molar

excess of inserts with 4-base overhangs (300 ng for 100 - 300 bp library and 700 ng

for 300 - 800 bp library), preparative ligation were set up and electroporated in high efficiency TOP10F’ cells.

The library of 100 - 300 bp contained 1 x 108 independent clones and was named MTBLIB25C01 (C01 for transformed cells containing unselected primary

clones). The library of 300 - 800 bp fragments contained 5 x 107 independent clones and was named MTBLIB27C01. The colony PCR and sequencing of the amplified products revealed that both libraries contained more than 97 % clones with sequences aligning to the M. tuberculosis genome (Table II.6A1 and B1). Detailed analysis revealed that MTBLIB25C01 library contained inserts within the

range of 90 - 300 bp with 97 % of the total recombinant clones (89/92) having

inserts in the range of 100 - 300 bp (Table II.7A1). In the case of MTBLIB27C01,

majority of the clones (91/92) carried inserts in the range of 300 - 800 bp with one insert larger than 800 bp. In this library, there was uniform distribution of inserts

A. Analysis of clones of library MTBLIB25 (100 - 300 bp library)

S.No. Stage Clones Analyzed Aligning to M.tb genome

ORF in frame with PelBss and gIIIp

Aligning to M.tb proteome

1. MTBLIB25 C01 95 92 (96.8)a 1 (1.05)b - 2. MTBLIB25 P01 47 47 (100)a 1 (2.1)b -

3. MTBLIB25 C02 48 48 (100)a 43 (89.6)b 20 (41.6)c 4. MTBLIB25 P02 48 48 (100)a 45 (93.7)b 25 (52.1)c

B. Analysis of clones of library MTBLIB27 (300 - 800 bp library)

S.No. Stage Clones Analyzed Aligning to M.tb genome

ORF in frame with PelBss and gIIIp

Aligning to M.tb proteome

1. MTBLIB27 C01 92 90 (97.8)a 1 (1.1)b 1 (1.1)c 2. MTBLIB27 P01 46 46 (100)a 1 (2.2)b -

3. MTBLIB27 C02 45 45 (100)a 43 (95.5)b 23 (51.1)c 4. MTBLIB27 P02 46 46 (100)a 45 (97.8)b 29 (63.0)c

a Percentage of recombinant clones that aligned to the M. tuberculosis genome b Percentage of clones that are in frame with PelB signal sequence and gIIIp of phagemid c Percentage of clones that aligned with the M. tuberculosis proteome using NCBI BLASTp program (Genic ORF) Table II.6. Analysis of randomly sequenced clones at various stages of library construction to represent ORF Selection. A. Analysis of MTBLIB25 library (100 - 300 bp); B. Analysis of MTBLIB27 library (300 - 800 bp) 1. Transformants obtained after large-scale electroporation of the ligation sample 2. Transductants obtained after infection of Top10F’ cells with the primary phage library 3. Transductants obtained after infection of Top10F’ cells with the trypsin treated ORF phages 4. Transductants obtained after infection of Top10F’ cells with the secondary phage library

ANALYSIS OF RANDOM CLONES AT DIFFERENT STAGES OF LIBRARY CONSTRUCTION

87

A. Analysis of clones of library MTBLIB25 (100 - 300 bp library) Size 1. MTBLIB25C01 (95) 2. MTBLIB25P01 (47) 3. MTBLIB25C02 (48) 4. MTBLIB25P02 (48)

Less than 100 bp 2 2 3 2 100 – 200 bp 42 25 24 29

200 – 300 bp 47 20 20 17

More than 300 bp 1 - 1 - B. Analysis of clones of library MTBLIB27 (300 - 800 bp library)

Size 1. MTBLIB27C01 (92) 2. MTBLIB27P01 (46) 3. MTBLIB27C02 (45) 4. MTBLIB27P02 (46)

Less than 300 bp - - - - 300 - 400 bp 21 11 15 20 400 - 500 bp 36 10 16 10 500 - 600 bp 20 9 8 9 600 - 700 bp 10 8 4 4 700 - 800 bp 4 4 2 3

More than 800 bp 1 4 - -

Table II.7. Size distribution of random sequenced clones at various stages of library construction. A. Analysis of MTBLIB25 library (100 - 300 bp); B. Analysis of MTBLIB27 library (300 - 800 bp) This table shows the size distribution of the randomly selected clones at different stages of the library construction to indicate no change in the size range of the clones before and after ORF selection. The total clones analyzed at each stage are indicated in the brackets.

1. Transformants obtained after large-scale electroporation of the ligation sample 2. Transductants obtained after infection of Top10F’ cells with the primary phage library 3. Transductants obtained after infection of Top10F’ cells with the trypsin treated ORF phages 4. Transductants obtained after infection of Top10F’ cells with the secondary phage library

SIZE DISTRIBUTION OF CLONES AT DIFFERENT STAGES OF LIBRARY CONSTRUCTION

88

89

in the range of 300 - 600 bp but clones carrying insert in the range of 600 - 800 bp were slightly less (Table II.7B1). This lower representation is seen despite using higher proportion of large size fragments during initial preparation of mixture of fragments to take into account reduced ligation efficiency of larger fragments. The uniformity can be maintained by adding higher proportion of large size

fragments. Alignment of sequenced fragments with M. tuberculosis H37Rv genome using BLASTn analysis (NCBI) showed that fragments of both libraries were uniformly distributed across the genome with no overlapping or duplicate clones (Fig.II.9). Phages were rescued from the two libraries using a novel helper

phage AGM13 (described in the next section) and the phage libraries were named MTBLIB25P01 and MTBLIB27P01 (Primary ORF-unselected phage library) for 100

- 300 bp and 300 - 800 bp fragment libraries, respectively. The phages were titrated on E. coli TOP10F’ and scored as ampicillin resistant (Ampr) transductants (Ampr colonies). The phage titre in both libraries was ~ 1011/ml. The colony PCR of Ampr transductants and sequencing of the PCR product revealed that in both the libraries 100 % clones were recombinants and contained fragments aligning with M. tuberculosis genome (Table II.6A2 and B2). The size distribution of fragments of both libraries was comparable to C01 and the fragments were uniformly distributed across the M. tuberculosis genome. In this regard, it may be noted that despite containing 100 % recombinants and fragments with DNA sequences aligning to the M. tuberculosis genome, only 1 - 2 % clones of the library contained fragments in frame with PelBss and gIIIp to encode the fusion protein that can get incorporated in the extruding phage particles. Theoretically, this number should be about 5.55 %, as only 1 out 18 possible frames of the insert would produce complete fusion protein comprising the PelBss, cloned protein and full length gIIIp. Hence, strategy was developed to eliminate phages that do not carry inserts in frame with PelBss and gIIIp using a novel helper phage

AGM13 that is described in the next section.

90

DISTRIBUTION OF PHAGEMID CLONES OF MTBLIB27 (C01) LIBRARY ACROSS THE GENOME CARRYING M. tuberculosis

H37RV GENE FRAGMENTS

Fig.II.9. Diagrammatic representation of distribution of M. tuberculosis H37Rv gene fragments at C01 stage of MTBLIB27 phagemid library. M. tuberculosis genome is ~ 4.4 Mb consisting of ~ 4000 genes. The map is up to scale and the inserts have been shown as blue arrows. The direction of the arrow indicates the gene orientation.

91

2.3.2. ORF selection using Helper phage AGM13 2.3.2.1. Characteristics of Helper phage AGM13

AGM13 helper phage was derived from wild type VCSM13 helper phage by inserting a four amino acid sequence KDIR (Lys-Asp-Ile-Arg) with trypsin

cleavage site after 237th amino acid residue between N2 and CT domain of wild type gIIIp. In the wild type phage, the gIIIp consists of N-terminal N1 and N2 domains involved in phage infectivity and C-terminal CT domain essential for

phage assembly. In E. coli TG1, AGM13 produced titre of 2 x 1011/ml, which is

similar to VCSM13. The gIIIp encoded by AGM13 helper phage was very sensitive to trypsin as even 1 µg/ml trypsin led to more than 104 times loss in phage infectivity due to removal of N1 and N2 domains, suggesting that the trypsin cleavage site was fully exposed. This was further confirmed by western blot analysis using anti-gIIIp polyclonal and monoclonal antibodies that in

trypsin-treated AGM13, the loss of infectivity was due to cleavage of gIIIp and

most of the gIIIp was cleaved even at 1 µg/ml trypsin. AGM13 was comparable

to VCSM13 in producing phage particles from a gIIIp based phagemid vector, as

the yield of phages producing Ampr transductants was same for both AGM13 and

VCSM13. Further, the phages produced by AGM13 showed comparable

reactivity in phage ELISA to phages produced by VCSM13. These experiments clearly demonstrated that insertion of a four amino acid sequence with trypsin

cleavage site did not alter the “helper” properties of AGM13.

The rationale of AGM13 (trypsin sensitive helper phage) for ORF selection is based on the premise that once a DNA fragment is cloned in between the signal sequence and the coding sequence of gIIIp, the gIIIp fusion protein will be produced only by those clones where the cloned DNA fragment is in frame with signal sequence on one end and with the gIIIp on the other end as a complete signal sequence and the gIIIp are necessary for assembly of gIIIp into the

extruding phages. Therefore, when AGM13 is used as helper phage, clones

92

carrying the DNA fragment in frame with signal sequence and the gIIIp would produce phages having fusion with trypsin-resistant gIIIp. This gIIIp fusion protein would get incorporated on the extruding phage along with trypsin-

sensitive gIIIp supplied by AGM13 (Fig.II.10C). In contrast, clones carrying a DNA fragment, that is not in frame with signal sequence and gIIIp, will not produce functional trypsin-resistant gIIIp fusion protein and hence phages would

carry only the trypsin-sensitive gIIIp supplied by AGM13 (Fig.II.10B). When such phage population is treated with trypsin, only those phages remain infectious which display trypsin-resistant gIIIp carrying the fusion (Fig.II.10). Thus, these phages would carry inserted DNA in a desired reading frame (ORF).

Before ORF selection of library, the trypsin treatment conditions were

optimized. In this assay, phages from primary unselected library (P01) were incubated with different concentrations of trypsin and then titrated using TOP10F’ cells. As already mentioned, only phages carrying inserts in frame with signal sequence and native gIIIp would produce transductants because of the presence of trypsin-resistant gIIIp fusion protein on phage surface. As low as 10 µg/ml trypsin treatment resulted in more than 94 % transductants that contained insert in frame with PelBss and the gIIIp coding sequence. At this concentration, there was about 104-fold fall in the titre as tested above (Table II.8A and B). This fall in titre can be due to two reasons, first, theoretically only 1 out of 18 clones should be in frame, and second that in the phagemid system only a very small percentage of phages display the gIIIp-fusion protein encoded by the phagemid vector. Parallely, we have found that in antibody display, about 4 - 5 % phages display the phagemid encoded gIIIp-antibody fusion protein (data not shown). But, the two situation cannot be compared as in the case of complex library, a variety of clones are present and would be having varying expression and display characteristics, while for scFv and Fab displaying phages, these variation would be relatively less. Since higher concentration of trypsin resulted in much more drop in the titre without any significant improvement in number of transductants with insert in frame with PelBss and gIIIp, 10 µg/ml trypsin concentration was used for all further experiments.

93

Fig.II.10. Schematic representation of ORF selection of phages rescued with AGM13 helper phage. The phage coat is depicted as a black rectangle, domains N1, N2 and CT of coat protein pIII as labeled circles in pink, shaded box indicates off frame insert, blue rectangle as in frame insert and displayed protein as a blue square. The zig-zag line indicates the trypsin cleavage site. (A, B and C) Phages rescued with AGM13 helper phage. A minor population displays in frame protein (C), while the majority contains only wild-type pIII encoded by AGM13 helper phage (A and B). All forms are infective due to the presence intact of N1 and N2. After trypsin treatment the majority phages (A and B) become non-infective due to removal of lack N1 and N2 infectivity domains but phages with displayed protein are infective due to the presence of N1 and N2.

ORF SELECTION OF PHAGES USING NOVEL HELPER PHAGE, AGM13

gIII from helper phage genome

Recombinant gIII from phagemid genome

A

B

C

TRYPSIN

TREATMENT

CT

N2 N1

TRYPSIN

CT

N2 N1

TRYPSIN Cleavage SITE

Only C population is infective.

A & B population is non-infective.

A. Analysis of clones of library MTBLIB25 (100 - 300 bp library) Trypsin

concentration Fold reduction

in titre Clones

Analyzed Aligning to M.tb

genome ORF in frame with

PelBss and gIIIp Aligning to M.tb

proteome

Without trypsin - 32 32 - - 10 µg/ml 1 x 104 32 32 30 (93.7)a 12 (37.5)b

100 µg/ml 4 x 104 32 32 31 (96.8)a 10 (31.25)b 200 µg/ml 106 32 32 30 (93.7)a 10 (31.25)b

B. Analysis of clones of library MTBLIB27 (300 - 800 bp library)

Trypsin concentration

Fold reduction in titre

Clones Analyzed

Aligning to M.tb genome

ORF in frame with PelBss and gIIIp

Aligning to M.tb proteome

Without trypsin - 47 47 - - 10 µg/ml 1.3 x 104 44 44 44 (100)a 29 (65.9)b

100 µg/ml 1.7 x 104 44 44 41 (93.2)a 22 (50)b 200 µg/ml 105 44 43 42 (95.4)a 24 (62.5)b

a Percentage of clones that are in frame with PelB signal sequence and gIIIp of phagemid. b Percentage of clones that aligned with the M. tuberculosis proteome using NCBI BLASTp program Table II.8. Analysis of random sequenced clones after treatment of primary phages with various trypsin concentrations. A. Analysis of MTBLIB25 library (100 - 300 bp); B. Analysis of MTBLIB27 library (300 - 800 bp) 1012 primary phages were incubated with four different concentrations of trypsin; 0,10,100 and 200 µg/ml and then used to infect E. coli TOP10F’ cells to determine the fall in titre. Randomly selected clones were analyzed by PCR and DNA sequencing to estimate the percentage of ORF (genic and non-genic ORF) and non-ORF clones.

OPTIMIZATION OF TRYPSIN CONCENTRATION FOR THE ORF SELECTION OF PHAGES

94

95

For ORF selection of the library, 1 x 1012/ml of purified and concentrated primary

phages (P01) of each library were treated with trypsin (10 µg/ml) for 30 min at

37oC and then mixed with log phase grown E. coli TOP10F’ cells. After incubating

for 30 min at 37oC the cells were pelleted to remove unbound phage and residual trypsin, the cells were washed twice and resuspended in 2 x YT Amp100glu1% and then plated on twenty 150 mm plates containing LBAmp100glu1%. Simultaneously, a small aliquot of these cells was plated to get well-isolated colonies for analysis. The colonies obtained for each library were scraped from 150 mm plates in 10 ml media, mixed with glycerol and stored at – 80oC for further use and labeled as C02 (cells of ORF selected clones). As already mentioned, trypsin treatment

would have destroyed gIIIp derived from AGM13 thereby leaving only those phages infectious which displayed a protein fused to trypsin-resistant wild type gIIIp derived from phagemid and hence carried ORF-selected insert between PelBss and gIIIp. Hence, each colony at C02 stage shall carry a phagemid with insert in frame with PelBss and gIIIp (ORF-selected sequence) and would be capable of displaying a protein. Further, if C02 cells were to be used for phage

production by AGM13 or by any other helper phage such as VCSM13 or

M13KO7, every clone should produce phages carrying insert in frame with PelBss and gIIIp and display encoded peptide/ protein.

Analysis of C02 cells of both libraries showed that 100 % clones were

recombinant with DNA sequence aligning to M. tuberculosis genome (Table

II.6A3a and B3a) and approximately 90 % (89.6% in 100 - 300 bp library and 95.5 %

in 300 - 800 bp library) were in frame with PelBss and gIIIp, thus carrying the

ORF-selected clone (Table II.6A3b and B3b). However, only half of the total clones contained sequence, which coded for protein that aligned to the M. tuberculosis

proteome referred as genic clones; Table II.6A3c and B3c) (Bradbury et al., 2011). This number is not surprising as there would be several clones with non-genic frame but without any stop codon and would make a gIIIp fusion protein and hence, display a non-genic protein, which cannot be avoided. The large fragment

(300 - 800 bp) library has more genic clones than the small fragment library which is understandable as chances of having a proper translatable frame without a stop

96

codon in non-ORF reading frame in large fragments would be less and such clone would not display gIIIp-fusion protein and would have got eliminated during enrichment. The analysis of phages rescued from C02 cells of both libraries also showed that more than 95 % of the phages carry clones with inserts in frame with PelBss and gIIIp and the proportion of genic clone is also more than 50 % (Table II.6A4 and B4b/c). Such an ORF-selected library is expected to be many fold more efficient than ORF-unselected library. Another important feature of this ORF selection protocol, was that the size distribution of different fragment was largely

maintained as 100 - 300 bp library contained more than 90 % fragments that were

within 100 - 300 bp range with occasional clones having insert of less than 100 bp

or more than 300 bp (Table II.7A3 and A4). Similarly, the 300 - 800 bp library

contained more than 90 % clones carrying inserts in the range of 300 - 800 bp

(Table II.7B3 and B4). Further, alignment showed that clones of both libraries at every stage of selection protocol (C01/P01/C02/P02) were uniformly distributed across M. tuberculosis genome with no two overlapping clones (after analyzing 48 or 96 clones), further emphasizing on the robustness of the protocol (Fig.II.11). 2.3.3. Characterization of ORF-selected phage display library

The previous section describes the construction and ORF selection of two

libraries carrying genes to code for 30 - 100 and 100 - 250 amino acids fused to the gIIIp. It was also found that 50 % of the clones carried sequences that coded for peptide/ proteins aligning to the annotated sequence of M. tuberculosis genome. Following ORF selection, the libraries were characterized for identifying epitopes using monoclonal antibodies raised against various mycobacterial proteins.

Affinity selection (panning) was performed using PEG - NaCl and

Sepharose 6B-CL column purified phages of MTBLIB25P02 and MTBLIB27P02 ORF-selected phage displayed libraries. As described in methods, for panning the MAbs were immobilized on microtitre plate surface and ORF-selected phages were added. After binding, unbound phages were removed by washing and bound phages were eluted by low pH. An aliquot was used to obtain

97

DISTRIBUTION OF CLONES CARRYING M. tuberculosis H37Rv GENE FRAGMENTS IN MTBLIB27 LIBRARY ACROSS THE

GENOME

C.

A. B.

D.

Fig.II.11. Diagrammatic representation of distribution of M. tuberculosis H37Rv gene fragments in MTBLIB27 library. M. tuberculosis genome is ~ 4.4 Mb consisting of ~ 4000 genes. A) indicates the distribution of inserts in MTBLIB27C01 primary cells before ORF selection. (B) indicates the distribution of clones in MTBLIB27P01 after trypsin treatment (10 µg/ml) of primary phages. (C) indicates the distribution of inserts in MTBLIB27C02 secondary cells obtained after trypsin treatment (10 µg/ml) of primary phages for ORF selection. (D) indicates the distribution of clones in MTBLIB27P02 obtained from C02. The non-ORF selected inserts that align with the M. tuberculosis genome have been shown as blue arrows. The clones in frame with PelBss gIIIp and are indicated as red arrow and clones in frame with the M. tuberculosis proteome are indicated as green arrows. The direction of the arrow indicates the gene orientation. The map is up to scale.

98

transductants for analysis. The major eluted fraction was used for infection and amplification to obtain more phages, which were purified before being used for second round of panning. The eluate of the second panning was also analyzed as transductants for DNA sequence. These transductants were used to produce phages and tested in phage ELISA to identify clones displaying antibody-binding peptides. Finally, the sequences were aligned to the protein sequence and correlated with the reactivity in phage ELISA. The minimal epitope was theoretically calculated by deriving the small overlapping region of phage ELISA

positive clones. The panning was carried out on four different MAbs U033, Ag85-12, which probably recognize small linear epitopes and MAbs 1912 and 1905 which probably bind to a large fragment of 19 kDa protein. (i) MAb U033

This is an antibody against Mtb 81 kDa (Rv1837c). Two rounds of panning

on MAb U033 resulted in selection of nearly 32 % clones (14 out of 44; Table II.9A1), which aligned to the N-terminal region of the protein in the range of 11 to

111 amino acids and were reactive in phage ELISA on MAb U033. The overlapping region of these clones predicted 25 - 72 amino acids to comprise of

epitope recognized by MAb U033 (Fig.II.12A). (ii) MAb Ag85-12

This antibody binds to both Ag85A (Rv3804c) and Ag85B (Rv1886c) of M.

tuberculosis. Since both Ag85A (339 amino acids) and Ag85B (326 amino acids) have very similar sequence with 90 % similarity at amino acids level with many identical stretches, therefore it was interesting to know if this MAb binds to same sequence on Ag85A and Ag85B or to a sequence which is similar but has some differences between Ag85A and Ag85B. Two rounds of panning selected clones, which upon DNA sequence analysis, aligned to a region of Ag85A and Ag85B.

However, most of the clones ~ 78 % (35 out of 45 analyzed) aligned to Ag85B

(Table II.9A2) carrying inserts aligning to amino acid 224 - 325 of Ag85B. There

99

were some clones that carried sequence belonging to Ag85A in the region 217 -

327 amino acid residues. All the clones aligning in this region were reactive in phage ELISA on MAb Ag85-12. The overlapping sequenced region of various phage clones led to the conclusion that region between 279 - 288 amino acid residues (KFQDAYNAAG) of Ag85B might constitute the epitope recognized by Ag85-12 (Fig.II.12B). The corresponding sequence in Ag85A aligns to 282 - 291 (KFQDAYNAGG) amino acid residues and has only one residue that if different (as shown in bold). Since, these residues have no side chains, their contribution in antibody binding might be minimal. It may be noted that the epitope

recognized by MAb U033 binds to protein coded by Rv1837c and MAb Ag85-12

binds to protein coded by Rv3804c and Rv1886c, which are located far away on the M. tuberculosis genome. Therefore, successful identification of epitope through several overlapping clones suggests that the ORF selected library is quite complete with lot of overlapping fragments covering the entire genome. (iii) MAb 1905 and 1912

MTBLIB25P02 library failed to identify epitope recognized by two others

MAbs 1912 and 1905 which bind to 19 kDa antigen (Rv3763) and are indicated to have an epitope comprising ~ 100 amino acids that might be conformational in nature. Further, 19 kDa antigen used for producing MAb 1912 and 1905 consists

of 136 amino acids with two cysteine residues at 44th and 135th position of its protein sequence. This failure could be attributed to the composition of this library as it might contain none or a very few clones carrying 100 amino acid long

(300 bp) fragments. Therefore, epitope mapping was tried using MTBLIB27P02

library phages that carries fragments of 300 - 800 bp coding for more than 100 amino acids with maxima around 150 - 175 amino acids. After two rounds of

panning on MAb 1912 and 1905, only 4 out of 32 and 2 out of 32 clones aligned to

the 19 kDa sequence, respectively (Table II.9B.I.3 and 4). Therefore, third round of panning was carried out using the amplified Pan II eluates. This led to further enrichment such that majority of the clones (more than 65 %) now aligned to the

19 kDa sequence (Table II.9B.II.3 and 4) and were reactive in phage ELISA on

AFFINITY SELECTION OF GENOME FRAGMENT LIBRARY ON VARIOUS MONOCLONAL ANTIBODIES

Pan II Monoclonal Abs (1) U033 (2) Ag85-12 (3) 1912 (4) 1905 Test Control Test Control Test Control Test Control Input phage 1 x 109 1 x 109 1 x 109 1 x 109 1 x 109 1 x 109 1 x 109 1 x 109 Output phage 7 x 105 2 x 106 8 x 104 3 x 106 7 x 105 1 x 106 8 x 105 2 x 106 Fold enrichmenta 2.85 37.5 1.43 2.5 % Clones aligning with the specific M. tuberculosis protein 32 (14/44) 77.7 (35/45) - (0/47) - (0/47)

a Fold enrichment is calculated as Test titer/control titer. Table II.9A. Panning of MTBLIB25P02 phages on various MAbs against M. tuberculosis proteins.

(A) MTBLIB25 LIBRARY

(I) Pan II

Monoclonal Abs (1) U033 (2) Ag85-12 (3) 1912 (4) 1905 Test Control Test Control Test Control Test Control Input phage 1 x 109 1 x 109 1 x 109 1 x 109 1 x 109 1 x 109 1 x 109 1 x 109 Output phage 3.5 x 105 8.5 x 104 3 x 105 3 x 104 9 x 105 2 x 105 1 x 106 2 x 105 Fold enrichmenta 4.0 10.0 4.5 5.0 % Clones aligning with the specific M. tuberculosis protein 3.1 (1/32) 71.9 (23/32) 12.5 (4/32) 6.25 (2/32)

(1) U033

(II) Pan III

(2) Ag85-12

(3) 1912

(4) 1905 Test Control Test Control Test Control Test Control Input phage 1 x 109 1 x 109 1 x 109 1 x 109 1 x 109 1 x 109 1 x 109 1 x 109 Output phage 4 x 105 1 x 105 1 x 106 3 x 104 1 x 106 6 x 104 5 x 105 5 x 104 Fold enrichmenta 4.0 30.0 17 10.0 % Clones aligning with the specific M. tuberculosis protein 12.5 (4/32) 90.6 (29/32) 68.7 (22/32) 65.6 (21/32) a Fold enrichment is calculated as Test titer/control titer.

Table II.9B. Panning of MTBLIB27P02 phages on various MAbs against M. tuberculosis proteins.

(B) MTBLIB27 LIBRARY

100

101

CHARACTERIZATION OF MTBLIB25/27 LIBRARY USING MONOCLONAL ANTIBODIES AGAINST M. tuberculosis ANTIGENS

(B) Ag85-12 50

50

100

150

150

200

200

250

250

300

300

100

Ag85B

247-289

263-304 250-287

254-299

260-301 251-288

236-302

262-319 245-296

252-308

263-308 276-325

1-326

234-319 228-308

238-325

236-307 224-300

238-307

249-315 241-307

256-318 234-303

(279) KFQDAYNAAG (288)

1-738

(A) U033 81-kDa

(25) LPG----RRV (72)

100

100

200

200

300

300

400

400

500

500

600

600

700

700

46-111

16-69 22-80

16-69 29-79 28-79 40-79 44-81 36-79

102

Fig.II.12. Phage Display based epitope mapping of MAbs raised against Mycobacterial antigens (A) anti-Mtb81 MAb, U033 (B) anti-Ag85B MAb, Ag85-12 (C) anti19-kDa MAb, 1905, and (D) anti19-kDa MAb, 1912. Schematics of alignment of translated sequence of 81-kDa, Ag85B and 19-kDa antigens of M. tuberculosis is represented as solid bars along with the corresponding locations of the deduced peptide sequences displayed on phages affinity selected using respective monoclonal antibodies. Amino acid numbers of the coded peptide is shown in bold on the right of the corresponding clone bar. The common minimal overlapping sequences are shown below alignment of each sequence.

20

20

40

40

60

60

80

80

100

100

120

120

19-kDa 1-136

14-136 22-136

1-136

42-136 24-136

(C) 1905

20

20

40

40

60

60

80

80

100

100

120

120

19-kDa 1-136 12-136

19-136 22-136 24-136 28-136 35-136

14-136

40-136

(42) VVCTTA---------------EVTCS (136)

(D) 1912

103

respective antibodies. The alignment of different sequences recognized a

common sequence encompassing amino acid residues 42 - 136 of 19 kDa including two cysteine residues (Fig.II.12C and D).

A remarkable observation was that the library contained a large number of

clones, which end at 136th residue of 19 kDa antigen but have variable residues at the N-termini. It is very unlikely that the primary library (MTBLIB27P01) did not

have clones with longer sequences beyond 136th residue but such clones must

have got eliminated due to stop codon after 136th residue during ORF selection. Therefore, the results described above clearly demonstrate that the technology illustrated here allows, construction of large gene fragment libraries of varying sizes with selection of ORF that can be efficiently employed for different applications. 2.3.4. Genome wide transfer of ORF-selected inserts from phage display vector into other vectors For many applications, gIIIp based phage displayed ORF-selected libraries can be directly used as has been illustrated for epitope mapping. As discussed

earlier, gIIIp based M13 phage display vector have low display density and many applications might require high-density display such as lambda phage based display as described in the next chapter. The ORF-selected libraries are a great resource and can be transferred for expression studies into different systems. While working on the strategy for high efficiency cloning of inserts, the adaptors were designed to carry special features including possibility of transferring entire ORF-selected library to another vector either using Gateway recombinational

system (Fig.II.13.B) or by our proprietary restriction enzyme free cloning strategy

(Fig.II.13.A). These adaptor sequences were designed for amplification of cloned inserts at library scale, where treatment of PCR product with T4 DNA polymerase in the presence of dTTP produces 4-base overhangs, 5’- CGGC at 5’ end and 5’-

CTCC at the 3’ end that can be cloned in a vector having compatible ends. In such vectors, compatible ends can be created using two appropriately oriented

Fig.11.13. Schematic representation of transfer of ORFs from the phage display vector to any other vector by (A) PCR based method (B) Gateway technology based method. (A) In PCR based method; the cloned ORF is amplified using two primers D1-s and E7-s from the pVCEPI23764 vector as template. The amplified PCR product is purified and then treated with T4 DNA polymerase in the presence of dTTP to create BsaI compatible ends. The insert can be then ligated to any BsaI digested vector with compatible ends. (B) In the Gateway method, the phagemid vector harboring the ORFs flanked by attB sites (B1 and B2) are transferred into attP sites (P1 and P2) containing pDONR221 donor vector using BP recombination to create entry library (ORFs flanked by attL sites), that can be easily shuttled into destination vectors with attR sites using LR recombination.

METHODS FOR THE TRANSFER OF ORFs FROM THE PHAGE DISPLAY VECTOR TO OTHER VECTORS

Destination vectors

ORFR1 R2

C-terminal fusion

B. GATEWAY METHOD

F+Ori

tHP

Ampr

T7TnlacPO pelB ORFB1 B2 Tryp Gene III

ccdBP1 P2

ORFL1 L2

ORFR1 R2 ORFR1 R2

N-terminal fusion E. coli expression

BP Recombination (Int and IHF)

pDONR221

Entry Library

LR Recombination (Xis)

pVCEPI23764 with ORF

A. PCR BASED METHOD

ORF

F+Ori

tHP

Ampr

T7TnlacPO pelB ORFB1 B2 Tryp Gene IIID1-s

E7-s

pVCEPI23764 with ORF

F+pVLEXp4337

AmprOri

T7lac

laci

H10

TEV

PCR amplification with D1-s and E7-s primers

ORFB1 B2

Amplified ORF

T4 DNA polymerase + dTTP

Ligation to BsaI vector

BsaI cut pVLExp4337 vector

OR

Any BsaI cut vector

104

105

BsaI restriction enzyme sites, as has been done in pVCEPI23964. In this section, transferability by PCR based method combined with our restriction enzyme free cloning strategy has been established by transferring the library wide ORFs from MTBLIBC02 (cells of ORF selected library) to a T7-promoter based expression vector.

The plasmid DNA from MTBLIB27C02 cells was extracted and used as template for PCR using primers D1-s and E7-s with high fidelity enzyme Pfu Ultra II HS fusion polymerase. Since, MTBLIB27C02 was derived from MTBLIB27C01 that carried 5 x 107 independent clones, after ORF selection, theoretically, MTBLIB27C02 is supposed to contain 1/18th fraction of total clones. Thus, the library size of ~ 2.5 x 106 would represent the entire original library. Since, earlier analysis showed that MTBLIB27C02 contained about 50 % genic clones (clones with translated sequence that aligned to M. tuberculosis proteome) and remaining being non-genic clones (in frame but not coding proteins that match M. tuberculosis proteome), a library of 5 x 106 clones was conceived to represent the entire original library. To represent the whole repertoire of the library clone with 100 representatives of each, 5 x 108 (100 x 5 x 106) plasmid DNA molecules were considered to be sufficient. Initially, PCR conditions were optimized to obtain good amplification in minimum cycles using 5 ng and 2.5 ng as template in 50 µl total reaction volume. Based on this, the final preparative PCR of 200 µl containing 5 ng of total plasmid DNA (~ 1 x 109 plasmid molecules with insert of average size of 500 bp) as template, was set up for 15 cycles and 7 - 8 µg of amplified product was obtained. The amplified product was treated with T4 DNA polymerase in the presence of dTTP to generate 4-base 5’-overhangs.

These inserts were ligated in 3-fold molar excess with 100 ng of BsaI-HF digested

T7-promoter based expression vector pVLExp4337. The ligation efficiency was about 5 x 106 clones per µg vector DNA as tested by electroporating a diluted

aliquot of ligation sample in sexcompetent BL21 (λDE3) RIL strain. The clones were analyzed by colony PCR using T7 promoter and T7

terminator primer followed by sequencing of the amplified PCR products using

106

the same primers. It was observed that 100 % clones were recombinants and

contained fragments ranging from 300 - 700 bp with majority of the fragments of size range 400 - 500 bp. The sequence analysis further revealed that ~ 88 % (60/68) clones were in frame with respect to PelBss and gIIIp, of which ~ 56 %

contained the genic ORFs (34/60). About 19 % (13/68) clones had inserts in the

size range 300 - 350 bp, 29 % (20/68) clones were in size range of 350 - 400 bp and

31 % (21/68) clones had inserts in size range of 400 - 500 bp. The large size inserts of 500 - 700 bp were represented by ~ 18 % (12/68) of the total clones analyzed. This distribution of fragment sizes is similar to that obtained in MTBLIB27C02 or MTBLB27P02. Thus, the protocol described here is able to transfer library from one vector to another vector with high efficiency (library size ~ 5 x 106 / µg ligated DNA) and high fidelity (no change in distribution of fragments).

The clones were subjected for protein expression by auto induction in

ZYM5052 medium in 96-well deep well plate. The SDS - PAGE (Fig.II.14) showed that a large number of clones expressed very good amount of protein of varying

sizes from 15 kDa to 35 kDa corresponding to 140 to 310 amino acids. Since, this vector adds about 25 amino acid residues due to N-terminal H10 tag, spacers and TEV protease site and also 11 amino acids of adaptors on either ends of the encoded peptide, the sizes of insert encoded peptides would be approximately 90 - 250 amino acids. Careful sequence analysis of 68 clones revealed that most the clones carrying genic ORFs showed very good level of expression (Fig.II.14, labeled red), while non-genic ORFs showed weak expression (Fig.II.14, labeled green), while clones without ORF or vector alone showed no expression (Fig.II.14, labeled blue). For some of the clones, although analyzed for expression but sequencing data was not available (Fig.II.14, labeled black). The results obtained showed that the PCR amplification based method described here results in rapid, efficient and precise transfer of ORFs from one vector to any other compatible vector thus, making possible multiple options for creating various expression clones useful for post-genomic functional studies.

107

kDa

21

31

45 66 97

14

1 32 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0

16 17 18 19 20 21 290

23 22 249

25 26 270

28 30 31 M M kDa

66 45

31

21

97

14

64 44 45 46 47 0

48 49 50 51 52 53 61 55 54 569

57 58 590

60 62 63 M M 33 34 35 360

37 38 39 40 41 42 43 kDa

21

31

45 66 97

14 6

kDa

66 45

31

21

97

14 6

96 76 77 78 790

80 81 82 83 84 85 93 87 86 889

89 90 910

92 94 95 M M 65 66 67 68 69 70 71 72 73 74 75 kDa

66 45

31

21

97

14 6

kDa

21

31

45 66 97

14 6

Fig.II.14. E. coli BL21 (λDE3) RIL cells harboring the expression plasmid pVLExp4337 with random clones of library MTBLIB27 were grown in 0.5 ml of ZYM5052 Amp100 Cm30 and incubated at 30°C, 250 rpm for 22 hours. Cells after incubation were harvested and then an aliquot of total cell (TC) were boiled in Laemmli sample buffer for 3-5 min at 110oC and electrophoresed on 0.1% SDS – 15.0 % PAG. The protein bands were visualized with Coomassie blue R-250 staining. M, molecular weight markers, broad range (BIORAD, Hercules, CA) (shown in kDa). The clones numbered in red font are the genic ORFs; in green font are the non-genic ORFs; in blue font are the off frame clones without ORF; and black font are the clones for which sequencing data was unavailable.

PROTEIN EXPRESSION OF CLONES OBTAINED AFTER TRANSFER OF MTBLIB27 LIBRARY TO

EXPRESSION VECTOR

108

The plasmid isolated from MTBLIBC02, was also subjected to BP recombination reaction to produce entry clones using BP clonase mix (int and IHF) and pDONR221 (donor vector) as per the manufacturer recommendation. But the library size obtained was only ~ 5 x 104 / µg of donor DNA. This was highly inefficient in comparison to the PCR based method described above. The BP reaction is the first step for the transfer of library clones from one vector to another to create entry clones that in the next step are transferred to destination vectors (Expression vector) using LR reaction in the presence of LR clonase mix (Xis). The literature shows that LR reaction is much more inefficient as compared to BP reaction under standard conditions. Since, BP reaction was not sufficient to produce required number of clones (5 x 106) to ensure complete transfer of the library, further transfer of entry clones to destination vectors using LR reaction was not attempted.

2.4 DISCUSSION

Phage display is a very powerful technology that allows selection of desired interacting partners, rather than screening to identify the desired partner.

The use of gIIIp based M13 phage display system to construct cDNA or genome fragment library has been limited, as cloning of intact cDNA does not allow display due to presence of stop codons and cloning of gene fragments results in phages not displaying any protein due to incorrect frames. Therefore, it would be desirable if a system could be developed that allows high efficiency cloning of fragmented genomes, elimination of unwanted non-ORF clones for the enrichment of clones carrying ORF and genome-wide transfer of each ORF-selected clone to other application specific vectors. With the work described in this chapter, we have been able to achieve all the three objectives stated in the beginning of this chapter, which are, construction of high efficiency genome-wide gene fragment libraries in phage display vector, selection of clones carrying open reading frame (ORF) and high efficiency genome-wide transfer of ORF-selected clones to other vectors. Besides

109

this, the phage displayed ORF-selected library of M. tuberculosis genome could be successfully used for identification of linear as well as conformational epitopes recognized by MAbs raised against mycobacterial proteins. For constructing the gene-fragment library, a gIIIp based phagemid vector,

pVCEPI23964 was designed that allows cloning of gene fragments between PelB signal sequence (PelBss) and full-length gIIIp (aa 2 - 406) using a restriction enzyme free method developed in our laboratory as a general high efficiency cloning method.

The fragments of M. tuberculosis genome were prepared by sonication. Since cloning of blunt ended fragments is inefficient, a novel method was developed by attaching two adaptors, which are synthesized by annealing

two complementary oligonucleotides of 34 bases and 30 bases, respectively so that one end of the adaptor is blunt while the other has 4-base overhang.

Further, one of the 34 base oligonucleotides, D1 carries biotin at the 5’ end. The two adaptors contain 21 base long attB1 and attB2 site-specific recombination sites, based on Gateway recombinational system. The 5’ end of each adaptor is designed such that after T4 DNA polymerase treatment of inserts with adaptors in the presence of dTTP, 4-base overhangs are created which are compatible to the BsaI digested phagemid vector. Since ligation of adaptors results in four types of ligated products with only two kinds of desired product bearing two different adaptors at each end, a method was developed to eliminate products having same adaptors on both the ends of the insert and finally, a single stranded library of inserts flanked with two different adaptors is obtained. The double stranded DNA was obtained by PCR using primers that anneal to the end of the adapted inserts within the adaptor sequences and conditions of minimal cycles were optimized to obtain desired size and yield of double stranded fragments. It may be noted that the sequences at the two ends of these inserts are quite similar due to the presence of 21 base long attB1 and attB2 sites with only a few mismatches. Therefore, the primers for amplification were designed to carry phosphorothioate

linkage at the last two penultimate bases so as to prevent degradation of the 3’

110

end of the primers by the high exonuclease activity (3’ - 5’) of Pfu Ultra II fusion HS DNA polymerase, as degraded primers may prime non-specifically. Finally, the treatment of these inserts with T4 DNA polymerase in the presence of dTTP results in creation of 4-base overhangs for the efficient ligation into the phage display vector to obtain libraries of order 0.5 - 1.0 x 108 independent clones. Several other groups have prepared genome-wide libraries but they generally either employ blunt end ligation or amplify insert using random primers carrying sequences of restriction sites (tagged-PCR) (Bradbury et al., 2011; Beghetto et al., 2008). However, the method developed in this study does not only result in high efficiency libraries but also provides a handle to transfer cloned DNA to other vectors using two different strategies as described later. The cloning of random fragments in any expression vector would result in clones, which are not in frame with respect to the initiation codon (1 out of 6 will be in frame). In the case of gIIIp based phage display vectors, the problem is more critical as only 1 out of 18 cloned inserts would be in frame with respect to the signal sequence and gIIIp so that the complete gIIIp fusion protein is secreted into the periplasm and after processing of the signal sequence, the gIIIp with N-terminal peptide/protein encoded by the inserts gets assembled on the extruding phages. Therefore, it would be of interest to eliminate the rest 17 types of clones that would theoretically bring the signal sequence and the coding sequence of

gIIIp off frame. Here we have taken advantage of a novel helper phage AGM13, which carries a four amino acid sequence “KDIR” in the spacer between the N2 and CT domain of gIIIp and is very sensitive to trypsin such that 1 - 10 µg/ml of trypsin inactivates gIIIp derived from this phage. For ORF-selection, library phages are rescued from the transformed cells carrying plasmid with cloned

inserts using AGM13 helper phage. The phages thus produced, mostly encapsulate single stranded DNA derived from the phagemid and phenotypically carry two kinds of gIIIp. One is derived from the phagemid with the encoded peptide/protein and is trypsin resistant while other is derived from the helper phage genome that is trypsin sensitive. Therefore, in clones where the insert is

111

not in frame with respect to the signal sequence and coding sequence of gIIIp, there would be no trypsin resistant gIIIp fusion protein and all five copies would be trypsin sensitive gIIIp derived from the helper phage. Thus, treatment of this phage population with trypsin would eliminate all those phages, which carry off frame inserts; consequently, enriching the phages with insert in frame or the ORF-selected clones. These phages can be rescued by infecting E. coli cells to obtain transductants harboring plasmid of clones with insert in frame with signal sequence and gIIIp coding sequence. When these cells are used for producing phage particles, majority of phages display peptides. However, there may be clones which carry insert that are in frame with signal sequence and gIIIp but do not code for protein matching with the proteome of M. tuberculosis. These occur due to the absence of any stop codon or suppressible stop codon (TAG) in the non-genic frame. We have been able to eliminate some of these inserts containing suppressible stop codons using E. coli TOP10F’, a non-suppressor strain. In this study, after ORF selection process, nearly 100 % clones carried inserts of desired size in frame with signal sequence and the gIIIp. However, occurrence of other non-genic clones could not be avoided. In our study, with M. tuberculosis genome, about 50 % clones in the library are genic (carrying inserts to code for real mycobacterial proteins). Thus, in the present work we were able to construct large ORF-selected library of M. tuberculosis genome. The gIIIp based phagemid display system has another inherent problem that only a small percentage (1 - 10 %) of phages carry the displayed fusion protein and most of the phages carry only the helper phage encoded gIIIp protein. Such phages are non functional, as they do not display any peptide or protein. Various helper phages have been described that provide reduced amount of gIIIp or no gIIIp at all to facilitate

incorporation of only phagemid encoded gIIIp fusion protein (Kramer et al., 2003;

Rondot et al., 2001; Soltes et al., 2003). Thus, it would be interesting to produce a

derivative of AGM13 that not only allows ORF-selection but also avoids production of phage population without any display.

Since cloning, expression and purification of full-length proteins from the annotated sequence is an arduous task, there have been numerous efforts to

112

identify the domain and clone, express and purify domain for structural and functional analysis because domain may be easier to be produced in a soluble form. Therefore, ORF selection method is of great importance. Several efforts have been made to select ORF by cloning randomly fragmented DNA upstream of a reporter molecule GFP so that in frame clone would lead to expression of functional GFP (Waldo et al., 1999; 2005) but presence of cryptic translational start sites within the insert sequence can lead to enrichment of false positive functional GFP expressing clones carrying a non genic frame at the N-terminus. However, this problem has been addressed by cloning inserts between the signal sequence and coding sequence of β-lactamase to select out ampicillin resistant (ampr)

clones, which should contain insert in frame (Zacchi et al., 2003). Although, β-lactamase system not only allows filtering out the open reading frames but also can be used for selecting out better folding molecules. But in the context of phage display, using this system, the ORF-selected inserts have to be either sub cloned into a phage display vector or β-lactamase sequence is required to be deleted by cre-LoxP recombination, thus requiring an additional effort (Bradbury et al., 2011; Faix et al., 2004). The hyperphage based system of ORF-selection (Hust et al., 2006) allows direct selection of ORF in phage display format but when applied to selection of ORF at library scale, there was decrease in the size range of fragments than the original library. In the system proposed here, the ORF-selected library can be directly used for panning and there is no change in the size of the inserts at any stage of library preparation. This size range and distribution of fragments is

consistently maintained because, upon trypsin treatment not only the AGM13 derived trypsin-sensitive gIIIp is destroyed but the displayed protein fused to trypsin resistant gIIIp is also removed thus, leaving all ORF-selected phages with trypsin resistant gIIIp without any additional sequence and therefore, all phages are equally infectious despite carrying gene to code for different sizes of ORF-selected proteins.

Nevertheless, the genome wide ORF-selected inserts are of immense value.

These inserts can be cloned into expression vector to produce whole proteome in a test tube. These inserts can be transferred to eukaryotic expression vector and

113

can be used for DNA immunization (Johnston et al., 1997; 2005). The technology of genome-wide ORF selection would be of immense value to identify vaccine candidates using techniques such as Proteomic-based Expression library screening (PELS, Kudva et al., 2006), In vivo induced antigen technology (IVIAT, Rollins et al., 2005), which employ cloning of gene fragments in inducible expression vector and currently, randomly fragmented DNA is cloned. Further, currently to enhance the cloning of correct reading frames, libraries are constructed in three vectors, which place inserted DNA into three reading frames, thus increasing the effort. All this can be overcome by using ORF selected gene fragments. These ORF-selected inserts can be transferred to yeast two-hybrid vectors to undertake protein interaction studies or inserts can be transferred to another phage display system for high-density display as described in Chapter III.

While designing the adaptors for cloning inserts in the library we had kept

these applications in mind and adaptors were planned for genome wide transfer of the inserts from ORF-selected library to any other system. Since, the two adaptors flanking each insert carried attB1 and attB2 sites in proper frame, these inserts could be first transferred to pDONR vector to make the entry library using BP reaction of Gateway system, which in turn could be transferred to any destination vector following the LR reaction. We attempted to transfer ORF-selected library to pDONR221, but only 5 x 104/µg efficiency was obtained which is very low if we wish to transfer the entire library of 5 x 106 ORF-selected clones. But this low efficiency is not surprising as reported efficiency with single gene is also in the range of 1 - 2 x 105/µg DNA (Hartley et al., 2000, Gateway®

Technology Catalog nos. 12535-019 and 12535-027, version 2010). Since, LR reaction is further less efficient no more attempts were made to transfer clones using this method. Within the adaptors we had incorporated sequences so that two primers could be designed to amplify faithfully the entire genome-wide ORF-selected inserts and then clone into any other vector following creation of 4-base overhangs upon T4 DNA polymerase treatment. The amplification primers were designed to carry phosphorothioate linkages at the two penultimate bases so that

114

Pfu Ultra II fusion HS DNA polymerase does not remove these bases to allow non-specific annealing and amplification. Template amount and PCR conditions were optimized to determine minimal amplification cycles to get desired yield. PCR based methods have been used for transferring genes from one vector to another as in In-fusion technology (In-Fusion® Advantage PCR Cloning Kit - Clontech) where also PCR conditions needs to be optimized for minimal cycles. To illustrate, the genome-wide transfer, the ORF selected inserts were amplified from plasmids obtained from MTBLIB27C02 cells (ORF selected

transductants carrying 300 - 800 bp inserts) and cloned in a T7 promoter based expression system using restriction enzyme free strategy with very high ligation

efficiency which were directly transformed in the expression host BL21 (λDE3) RIL to obtain efficiency of 5 x 106 / µg vector DNA. The analysis of 96 clones showed that the transfer was nearly 100 % faithful with no change in size distribution of the inserts. All the clones with in-frame inserts expressed proteins. It was noted that genic clones expressed more protein than the non-genic clones. This transfer strategy is universal and has been used in transferring other ORF-selected library from a gIIIp phage display vector to a gVIIIp based phage display vector that was further characterized for epitope mapping. In that system also, following transfer the distribution of insert remained unchanged with successful identification of epitopes of various MAbs against mycobacterial proteins (data not shown). Thus, the work described here provides a complete technology for constructing large phage display libraries carrying ORF-selected inserts that can be directly used for selection or can be faithfully transferred to other vectors. In contrast, the existing methods might allow ORF-selection but do not allow easy transfer of ORF-selected clones between different vectors.