INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
University Microfilms International A Bell & Howell Information Company 300 North Zeeb Road. Ann Arbor, Ml 48106-1346 USA 313/761-4700 800/521-0600
Order Number 9316190
Heterodimeric deoxyguanosine/deoxyadenosine kinase from Lactobacillus acidophilus R-26: Affinity protein purification, molecular cloning, sequence of the genes, and expression in Escherichia coli
Ma, Grace Tak-Yi, Ph.D.
The Ohio State University, 1993
Copyright ©1993 by Ma, Grace Tak-Yi. All rights reserved.
UMI 300 N. ZeebRd. Ann Arbor, MI 48106
HETERODIMERIC DEOXYGUANOSINE/DEOXYADENOSINE KINASE
FROM LACTOBACILLUS ACIDOPHILUS R-26:
AFFINITY PROTEIN PURIFICATION,
MOLECULAR CLONING, SEQUENCE OF THE GENES,
AND EXPRESSION IN E. COLI
DISSERTATION
Presented in Partial Fulfillment of the Requirement for
the Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
By
Grace Ma, B.S.
* * * « *
The Ohio State University
1992
Dissertation Committee: Approved by
D.H. Ives
G.A. Marzluf
C.A. Breitenberger Adviser,
M.D. Tsai Biochemistry Program,
The Ohio State University Copyright by
Grace Tak-Yi Ma
1993 To My Parents
and
My Husband Sik Kwan ACKNOWLEDGEMENTS
Words alone cannot express my thanks to my adviser Professor David H.
Ives. His guidance, assistance, understanding, support and patience will always be gratefully appreciated.
My special thanks go to Dr. Seiichiro Ikeda for his assistance and friendship during this work. Many thanks also go to all my friends in my laboratory-Dr. Min Young Kim, Dr. Ning Ma, Young Soo Hong, Pamphinol
Bunnag, and others, whose help and company made my years in the laboratory enjoyable.
I would like to gratefully acknowledge the advice and assistance provided by the faculty members of the Department of Biochemistry, especially Professors
George Marzluf and Caroline Breitenberger.
To my parents, my everlasting appreciation and love go to them for their unshakable, continuous support and love.
Finally, I express my deepest thanks, indebtedness and love to my husband,
Sik Kwan, whose love, support, encouragement, comfort and lots of patience made this possible. VITA
February 26, 1962 Bom - Hong Kong
1985 Bachelor of Science The Pennsylvania State University University Park, Pennsylvania
1986 - present Graduate Teaching/Research Associate, Department of Biochemistry The Ohio State University
PUBLICATION
1. Grace T. Ma and David H. Ives. "Cloning, Sequence and Expression of the Heterodimeric Deoxyadenosine Kinase/Deoxyguanosine Kinase from Lactobacillus acidophilus R26." Manuscript in preparation.
2. Seiichiro Ikeda, Grace T. Ma and David H. Ives. "Control of Competing Ionic and Hydrophobic Forces in Affinity Chromatography of the Deoxyguanosine Kinase/Deoxyadenosine Kinase Complex from Lactobacillus acidophilus R-26." Submitted.
3. Seiichiro Ikeda, Grace T. Ma and David H. Ives. "Heterodimeric Deoxynucleoside Kinases of Lactobacillus acidophilus R26: Assignment of Subunit Function by Direct Photoaffinity Labeling with dNTP Analogs and Comparison of N-Terminal Sequences." Manuscript in preparation.
v FIELDS OF STUDY
Major Field: Biochemistry
Studies in Protein Biochemistry Professor David H. Ives
Studies in Molecular Biology Professors David H. Ives, George A. Marzluf, Caroline A- Breitenberger, and Donald H. Dean TABLE OF CONTENTS
DEDICATION ...... iii
ACKNOWLEDGEMENTS ...... iv
VITA...... v
LIST OF TABLES...... xi
UST OF FIGURES...... xii
ABBREVIATIONS ...... xiv
ABSTRACT ...... xvii
CHAPTER PAGE
I. INTRODUCTION...... 1
A. Deoxynucleoside Kinases From L. acidophilus R-26 ...... 1
1. Overview ...... 1 2. Previous studies ...... 2
B. Deoxynucleoside Kinases From Other Sources ...... 17
1. Other Bacterial Deoxynucleoside K inases ...... 17 2. Mammalian Deoxynucleoside Kinase ...... 18 3. Viral Thymidine Kinases ...... 20
C. Affinity M edia ...... 21 D. Cloning Vector And Host Cell L in e ...... 22
1. Cloning Vector ...... 22 2. Host C e ll...... 25
II. MATERIALS AND METHODS...... 27
A. Materials...... 27
B. Methods ...... 28
1. Anion Exchange HPLC ...... 28 2. DEAE Trisacryl M ...... 28 3. Inorganic Phosphate Determination ...... 29 4. Culturing L. acidophilus R-26 ...... 30 5. Extraction of Enzymes ...... 31 6. Protein A ssay ...... 31 7. Enzyme Activity Assay ...... 32 8. SDS-Polyacrylamide Gel Electrophoresis ...... 33 a. The Laemmli Buffer System ...... 33 b. The pH 7.28 MZE 3328.IV buffer system ...... 34 9. Electroblotting ...... 34 10. N-Terminal Protein Sequence Analysis ...... 36 11. Chromosomal DNA Preparation ...... 36 12. Oligonucleotides ...... 37 13. Polymerase Chain Reaction (P C R ) ...... 38 14. Preparation of ^P-labeled DNA P ro b e ...... 38 15. Synthesis of Biotinylated DNA Probe by P C R ...... 39 16. Gel Electrophoresis for D N A ...... 39 17. Elution of DNA from Agarose G e l ...... 40 18. Southern Transfer and Hybridization ...... 41 19. Nonisotopic Nucleic Acid Detection ...... 41 20. Plasmid DNA Preparation ...... 42 21. Competent Cells and Transformation ...... 42 22. Colony Hybridization ...... 43 23. Genomic Library Screening by Polymerase Chain Reaction ...... 44
viii 24. DNA Sequencing of Cloned D N A ...... 44 25. Asymmetrical PCR and DNA Sequencing ...... 45 26. Extraction and Purification of Enzymes from E.coli clones ...... 46
III. AFFINITY PURIFICATION OF DEOX Y GU ANOSINE/ DEOXYADENOSINE KINASES FROM L. ACIDOPHILUS 47
A. Construction Of dAp 4 -Sepharose Affinity M edium ...... 47
1. Development ...... 47 2. Chemical Synthesis and Characterization ...... 49
B. Purification Of Deoxyguanosine Kinase/Deoxyadenosine Kinase By dAp 4 -Sepharose Affinity Chromatography ...... 58
C. Characterization Of Affinity Purified Deoxyguanosine Kinase/Deoxyadenosine K inase ...... 62
1. Enzyme Purity and Subunit Molecular W eight ...... 62 2. N-Terminal Protein Sequence Analysis ...... 65
IV. DEVELOPMENT OF CLONING PROBES BY POLYMERASE CHAIN REACTION...... 70
A. Construction Of Cloning Probe By P C R ...... 70
B. Analysis Of PCR Products ...... 75
1. Polyacrylamide Gel Electrophoresis ...... 75 2. Southern Transfer and Hybridization ...... 77 3. DNA Sequencing ...... 79 a. asymmetrical P C R ...... 79 b. cloning into pBluescript ...... 81
ix C. Preparation Of Biotin-Labeled Probe By PC R ...... 82
1. Reamplification of 93mer Fragment ...... 82 2. Biotinylation ...... 82
V. GENE CLONING OF DEOXYNUCLEOSIDE KINASES FROM L. ACIDOPHILUS IN E. C O L I...... 86
A. Identification Of Deoxynucleoside KinasesGenes ...... 86
1. Southern Analysis of Restriction Digested Genomic DNA ...... 86 2. Xba I Restricted Partial Genomic Library ...... 88 3. Kpn I Restricted Partial Genomic Library ...... 91
B. Nucleotide Sequence Analysis ...... 94
1. Open Reading Frames ...... 94 2. Comparison of DNA Sequences of Cloned Genes and Known N-terminal Peptide Sequences ...... 97 3. Promoter, Ribosome Binding Site and Terminator 99 4. Calculated pi Values ...... 101 5. Sequence Comparison of dAdo Kinase and dGuo Kinase ...... 101 6. Sequence Comparison with Related Kinases ...... 104 7. Codon Usage of Lactobacillus...... 110
VI. EXPRESSION AND CHARACTERIZATION OF CLONED DEOXYADENOSINE KINASE/DEOXYGUANOSINE KINASE...... 117
A. Enzyme Activities ...... 117
B. Kinetic Characteristics ...... 118
C. Purification by dAp 4 -Sepharose...... 120
x D. Subunit Molecular Weight by SDS-PAGE ...... 122
E. N-Terminal Protein Sequences ...... 127
VII. CONCLUSIONS AND RECOMMENDATIONS...... 129
BIBLIOGRAPHY...... 141
xi LIST OF TABLES
Table 1. Phosphate donor specificity of dCK/dAK...... 7
Table 2. Specificity of phosphate acceptor of deoxynucleoside kinase ...... 8
Table 3. Kinetic constants for inhibition of deoxynucleoside kinase by dNp4A ...... 12
Table 4. Kinetic constants for inhibition of deoxynucleoside kinases by dN T P ...... 13
Table 5. List of primers used in PCR and sequencing ...... 92
Table 6. List of Lactobacillus genes used in the codon usage tables (Table 7 to 9) ...... I ll
Table 7. Comparison of codon usages in genes among different species of Lactobacilli : acidophilus, brevis, bulgaricus, casei, confusus and delbrueckii...... 113
Table 8. Comparison of codon usages in genes among different species of Lactobacilli : fermenti, helveticus, hilgardii, lactis and paracasei ...... 114
Table 9. Comparison of codon usages in genes among different species of Lactobacilli: pentosus, plantarum, reuteri, sake, and 30a ...... 115
Table 10. Codon usage of Lactobacillus genes...... 116
Table 11. Specific activities of crude extracts from E. coli clone and from L. acidophilus...... 119
Table 12. Kinetic characterization of cloned dAK/dGK ...... 121 LIST OF FIGURES
Figure 1. Predominant pathways of deoxynucleoside metabolism in L. acidophilus ...... 3
Figure 2. Blue Sepharose affinity chromatography: Elution by ATP- Mg gradients...... 5
Figure 3. Blue Sepharose affinity chromatography: Elution by bisubstrate mixture...... 10
Figure 4. Putative modes of binding of substrates and multisubstrate analogs at the active sites of deoxynucleoside kinases 14
Figure 5. Partial amino acid sequence of dCyd kinase/dAdo kinase from L. acidophilus...... 16
Figure 6. Diagrammatic representation of pBluescript cloning vector ...... 24
Figure 7. Chemical structure of dAp 4 -Sepharose...... 50
Figure 8. Synthesis scheme of P^-(6-aminohex-l-yl)-deoxyadenosine tetraphosphate...... 51
Figure 9. FPLC Mono-Q anion-exchange elution profile of compound (III) and by-products ...... 55
Figure 10. DEAE Trisacryl M elution profile of compound (III) and by-products ...... 56
Figure 11. UV absorption spectrum of dAp 4 -Sepharose and deoxyadenosine ...... 59
Figure 12. Purification of dGuo kinase/dAdo kinase by dAp 4 -Sepharose affinity chromatography (20 ml dAp 4 -Sepharose)...... 61
xiii Figure 13. SDS-polyacrylamide gel (Laemmli buffer system) analysis of dAp 4 -Sepharose purified dGuo kinase/dAdo kinase ...... 63
Figure 14. Determination of dGuo kinase/dAdo kinase subunit molecular weight by SDS-PAGE...... 64
Figure 15. SDS-polyacrylamide gel (pH7 MZE 3328.IV buffer system) analysis of dAp 4 -Sepharose purified dGuo kinase/dAdo kinase ...... 66
Figure 16. N-terminal protein sequence of dGuo kinase and dAdo kinase ...... 67
Figure 17. Design of PCR primers for amplification of dCyd kinase/dAdo kinase specific probe ...... 72
Figure 18. 6% polyacrylamide gel analysis of polymerase chain reaction products ...... 76
Figure 19. Analysis of PCR products by Southern blot and hybridization ...... 78
Figure 20. DNA sequence and derived amino acid sequence of 93 base pairs PCR product ...... 80
Figure 21. Reamplification of the 93 base pairs PCR product ...... 83
Figure 22. Analysis of the biotinylated 93 base pairs PCR product. A) 4% agarose gel. B) Southern blot and biotinylated DNA detection ...... 84
Figure 23. Southern and hybridization analysis of restriction digested chromosomal DNA from L. acidophilus. (Probe: biotinylated 93mer PCR fragment)...... 87
Figure 24. Partial restriction map of Xba I clone ...... 89
xiv Figure 25. Southern and hybridization analysis of restriction digested genomic DNA from L. acidophilus. (Probe: biotinylated 117mer PCR fragment) ...... 93
Figure 26. Partial restriction map and sequencing strategy of Kpn I clone ...... 95
Figure 27. Sequence of dGuo kinase gene and dAdo kinase gene from L. acidophilus...... 96
Figure 28. Comparison of N-terminal amino acid sequences obtained by peptide and DNA sequencing ...... 98
Figure 29. Comparison in DNA sequence of dAdo kinase gene and dGuo kinase gene ...... 102
Figure 30. Comparison of derived amino acid sequence of dAdo kinase and dGuo kinase ...... 103
Figure 31. Sequence comparison of dAdo Kinase/dGuo kinase from L. acidophilus and Herpeviral T K ...... 105
Figure 32. SDS-polyacrylamide gel (Laemmli buffer system) analysis of expressed proteins in E. coli crude extract ...... 123
Figure 33. SDS-polyacrylamide gel (Laemmli buffer system) analysis of dAp 4 -Sepharosepurified dGuo kinase/dAdo kinase from the E. coli clone ...... 125
Figure 34. SDS-polyacrylamide gel (pH7 buffer system) analysis of dAp 4 -Sepharose purified dGuo kinase/dAdo kinase from E. coli clone ...... 126
Figure 35. Arrangement of the dAdo kinase/dGuo kinase genes from L. acidophilus R-26...... 137
xv ABBREVIATIONS
ado adenosine
ADP adenosine 5’-diphosphate
AMP adenosine S’-monophosphate
ATP adenosine 5’-triphosphate
BHV-1 bovine herpesvirus type 1
BSA bovine serum albumin cAK chicken muscle adenylate kinase cyd cytidine dAdo 2’-deoxyadenosine dAdo kinase (or dAK) deoxyadenosine kinase dAp4 2’-deoxyadenosine 5’-tetraphosphate dATP 2’-deoxyadenosine 5’-triphosphate dCDP 2’-deoxycytidine 5’-diphosphate dCp4 2’-deoxycytidine 5’-tetraphosphate dCTP 2’-deoxycytidine 5’-triphosphate dCyd 2’-deoxy cytidine dCyd kinase (or dCK) deoxycytidine kinase dGTP 2’-deoxyguanosine 5’-triphosphate dGuo 2’-deoxyguanosine dGuo kinase (or dGK) deoxyguanosine kinase dThd 2’-deoxythymidine
DNA deoxyribonucleic acid
DTE dithioerythritol
DTT dithiothreitol
EBV Epstein-Barr virus
EcAK E. coli adenylate kinase
EDTA Ethylenediamintetraacetate
EHV-1 equine herpesvirus type 1
FHV feline herpesvirus
FPLC fast protein liquid chromatography
HPLC high pressure liquid chromatography
HSV-1 herpes simplex virus type 1
HSV-2 herpes simplex virus type 2
HVS herpesvirus saimiri
LBA Lactobacillus acidophilus
MDV Marek’s disease virus MHV marmoset herpesvirus
PAK porcine adenylate kinase
PCR polymerase chain reaction
PMSF phenylmethylsulfonyl fluoride
RNA ribonucleic acid
RPV pseudorabies virus
SDS sodium dodecyl sulfate
THV turkey herpesvirus
Tris Tris-(hydroxymethyl)-aminomethane
TTP thymidine S’-triphosphate
VZV varicella-zoster virus
xviii ABSTRACT
HETERODIMERIC DEOXYGUANOSINE/DEOXYADENOSINE KINASE
FROM LACTOBACILLUS ACIDOPHILUS R-26:
AFFINITY PROTEIN PURIFICATION,
MOLECULAR CLONING, SEQUENCE OF THE GENES,
AND EXPRESSION IN E. COLI
By
Grace Tak-Yi Ma
The Ohio State University, 1992
Professor David H. Ives, Adviser
A new affinity medium, dAp4-Sepharose, for deoxyadenosine (dAdo) kinase was constructed. This dAp4-Sepharose column has increased the specific activities of both deoxyguanosine (dGuo) kinase (2150 units/mg) and dAdo kinase
(280 units/mg) by 2,700-fold, and efficiently purified the enzyme to homogeneity.
The enzyme appeared to be a heterodimer with similar subunit molecular weights of 26,000 daltons. The N-terminal protein sequences of the two subunits reflected high homology to each other, and to the deoxycytidine (dCyd) kinase/deoxyadenosine (dAdo) kinase, except for the initial amino acids.
Based on the known N-terminal protein sequence of dCyd kinase/dAdo
kinase, highly selective cloning probes specific for the dCyd kinase gene and the
dGuo kinase gene was constructed by employing the DNA amplification method
of the polymerase chain reaction (PCR). A clone containing an intact dAdo
kinase gene and a dGuo kinase gene was identified by the PCR probe, from a
Kpn I restricted partial genomic library of L. acidophilus R-26, constructed in the
pBluescript vector. The DNA sequence revealed two tandem genes, which were
separated by a 21 bp spacer, consisting of sequence homologies with the known
N-terminal amino acid sequence of dAdo kinase/dCyd (or dGuo) kinase. The upstream gene (later identified as the dAdo kinase gene) and the downstream gene (later identified as the dGuo kinase gene), respectively, encoded a 25 KDa polypeptide and a 26 KDa polypeptide, which agree with the subunit molecular weights of the purified Lactobacillus protein. Comparison of the two genes revealed 65% overall homology in DNA sequence, and 60.9% identity in the derived amino acid sequence. Consensus sequences of promoter, ribosome binding site and transcription terminator have been identified. Codon biases of the dAdo kinase/dGuo kinase genes of L. acidophilus were compared with those of other Lactobacillus genes.
Expression of the genes, using their single endogenous promoter, in E. coli results in dAdo and dGuo phosphorylation specific activities (in crude extract)
1000 to 2000-fold higher than background, and 10-fold higher than L. acidophilus.
xx Patterns of substrate phosphorylation, allosteric interaction, and end-product inhibition confirm the identities of the gene products as the dAdo kinase/dGuo kinase of L. acidophilus R-26. CHAPTER I
INTRODUCTION
A. DEOXYNUCLEOSIDE KINASES FROM L. acidophilus R-26
1. Overview
Deoxynucleoside kinases catalyze the transfer of a phosphoryl group from
MgATP to deoxynucleosides to form deoxynucleotide 5’-monophosphates and
ADP. Lactobacillus acidophilus R-26 is unique in that all four deoxynucleoside kinases are present and, because there is no functional ribonucleotide reductase, are essential to generate the four deoxynucleotide precursors of nucleic acid synthesis. The four deoxynucleoside kinases are namely:
Deoxyadenosine kinase: 2’-deoxyadenosine + ATP -» 5’-dAMP + ADP
Deoxycytidine kinase: 2’-deoxycytidine + ATP 5’-dCMP + ADP
Deoxyguanosine kinase: 2’-deoxyguanosine + ATP -*■ 5’-dGMP + ADP
Thymidine kinase: 2’-deoxythymidine + ATP -*■ 5’-dTMP + ADP
Most organisms possess both the de novo pathway to synthesize nucleotides from small precursors, and salvage pathways to synthesize nucleotides from preformed nucleosides. The de novo pathway seems to be the same in all organisms, while the salvage pathways, on the other hand, seem to be more
1 2 diverse. Unlike other bacteria, L. acidophilus has an absolute growth requirement for a purine, a pyrimidine and a single deoxyribonucleoside of any sort, since it lacks the ribonucleotide reductase (1) to couple de novo and salvage pathways. A single deoxynucleoside is sufficient for growth, because the bacterium is capable of deoxyribosyl transfer between bases. The predominant pathways of deoxynucleoside metabolism in L. acidophilus (2) are shown in
Figure 1.
2. Previous studies
L. acidophilus has previously been shown by Hoff-Jorgensen to be able to grow on deoxycytidine as its sole source of deoxyribose (3). This suggested the presence of deoxynucleoside kinases in the organism.
Durham and Ives (1) partially purified and characterized all four deoxynucleoside kinases from L. acidophilus, and found the following primary characteristics:
1. Thymidine kinase could be physically separated from the other three
kinase activities.
2. Phosphorylation of each deoxynucleoside is inhibited most effectively by
its homologous deoxynucleoside triphosphate.
3. dCTP, beside inhibiting deoxycytidine kinase, also inhibits deoxyguanosine
kinase activity, but stimulates deoxyadenosine and thymidine
phosphorylation. / dAMP dGMP dCMP ►dUMP dTMP
{ NH3 i i i
dAK dGK dCK TK TK
3 4 dAdo, dGuo dCyd dUrd dThd
U C Figure 1. Predominant pathways of deoxynucleoside metabolism in L acidophilus R-26. (Davis and Ives, 2). Enzymes involved are: (1) dCMP aminohydrolase, (2) thymidylate synthetase, (3) and (4) nucleoside deoxyribosyl transferase. 4 4. Deoxyadenosine phosphorylation is stimulated by deoxycytidine and
deoxyguanosine.
5. Deoxyadenosine kinase is less stable to heat and dilution than
deoxycytidine kinase and deoxyguanosine kinase.
6. Deoxycytidine kinase requires a divalent cation and a nucleoside
triphosphate for activity.
7. The pH profile of deoxycytidine kinase is biphasic, with optima at pH 7.6
and 10.5.
Deibel and Ives (4,5,6) resolved the three deoxynucleoside kinases into two paired associated activities, i.e. deoxycytidine kinase/deoxyadenosine kinase and deoxyguanosine kinase/deoxyadenosine kinase, by affinity chromatography on Blue Sepharose CL-6B. Characterization of the enzymes, concentrating mostly on dCyd kinase/dAdo kinase, has shown the following:
1. Two major enzyme fractions were recovered from Blue Sepharose affinity
column by differential elution with Mg-ATP. One fraction shows
enhanced specificity towards dCyd and dAdo, and the other towards dGuo
and dAdo. See Figure 2.
2. The molecular weights of both enzymes, dCyd kinase/dAdo kinase and
dGuo kinase/dAdo kinase, determined by gel filtration, are 50,000 ±
4,000. glycerol.Ives,&(Deibel 4). Figure Blue 2.sepharose affinity chromatography: Elution by ATP-Mg gradients, lf) - m; rgt 31 M i 1 m oasu hsht, pH phosphate, 15 mMpotassium in mM, 3-10 (right) 0-3 mM; (left)
nmoles dNMP 0.80 0.32 Q64 0.48 0.16 r r ^ 2 £", ■ ' ■ s£="^,a dCyd // I J// X A F Jt ! \ ±. , _ d do .dA ^dGuo
4 10 —i i— 8 6 . 6 15% , 5 3. dCyd kinase and dAdo kinase activities, (as the dGuo kinase and dAdo
kinase activities), are physically inseparable by gel electrophoresis and ion-
exchange chromatography.
4. Phosphate donor specificity and phosphate acceptor specificity of the dCyd
kinase/dAdo kinase are shown in Table 1 and Table 2.
5. pH optima for dCyd and dAdo phosphorylation were 7.8 and 9.5,
respectively. However, in the presence of the effector dCyd, the dAdo
kinase pH optimum changed to 7.8 also.
6 . The kinetic mechanism of the dCyd kinase was found to be rapid
equilibrium random bi bi; dAdo kinase activity also showed sequential
mechanism.
7. dCyd kinase and dAdo kinase sites exhibit allosteric interaction: dAdo
phosphorylation is noncompetitively stimulated (about 5 fold) by dCyd, i.e.
occupancy of the dCyd site stimulates dAdo kinase.
8 . dAdo is a weak competitive inhibitor of dCyd phosphorylation.
9. Allosteric interaction and differential responses of dCyd kinase and dAdo
kinase activities toward thermal stability, differential sensitivity to anions
and rose bengal show that the dCyd kinase and the dAdo kinase are on
separate catalytic sites on a single protein.
Chakravarty et aL (7) further purified and characterized the dGuo kinase/dAdo kinase as follows: 7
Table 1. Specificity of Phosphate Donor to Deoxynucleoside Kinase Activities. Deibel and Ives ( 6 ). Enzyme was 0.51 Mg of Fraction V-A, and the concentration of the donor nucleoside triphosphates was 10 mM. The concentration of Mg+ was 12 mM, and the acceptor deoxynucleoside concentration was 0.5 mM.
Relative Activity, % Donor dAdo dCyd dGuo
ATP 1 0 0 %’ 1 0 0 %* 1 0 0 %*
dATP 0 % 104% 32% CTP 262% 59% 72% dCTP 192% 17% 13%
GTP 83% 45% 1 2 %
dGTP 452% 23% 0 %
dTTP 6 8 % 38% 27%
*100% Activity corresponds to: dAdo Phosphorylation 10.3 nmoles/min/mg dCyd Phosphorylation 48.2 nmoles/min/mg dGuo Phosphorylation 39.0 nmoles/min/mg 8
Table 2. Specificity of Phosphate Acceptor of Deoxynucleoside Kinase. Deibel and Ives ( 6 ).
Relative Activity, %
Acceptor 0.020 mM Substrate 0.500 mM Substrate V-A V-C V-A V-C dAdo 15.1 25.8 25.6 13.9
dCyd 1 0 0 .0 * 7.0 1 0 0 .0 * 14.7
dGuo 9.4 1 0 0 .0 * 38.7 1 0 0 .0 *
dThd ND ND 0 . 1 0 . 1
Ado 0 . 2 ND 1 . 8 2.5
Cyd ND ND 3.8 0 . 1 Guo ND ND ND 2.5
Urd ND ND 0 . 1 0 . 1
The concentration of ATP-Mg (1:1) was 10 mM, and the nucleoside concentrations were varied as indicated. 1.2 Mg of Fraction V-A protein were used to assay 0.5 mM substrates, and 0.2 Mg were used to assay 0.02 mM substrates. 0.7 Mg of Fraction V-C protein and 0.07 Mg were used to assay 0.5 mM and 0.2 mM substrates, respectively.
*100.0% Activity corresponds to 41.6, 69.7, 49.8, and 135.2 nmoles/min/mg for Fractions V-A and V-C (0.020 mM), and Fractions V-A and V-C (0.500 mM), respectively. 1. Base-line separation of the dCyd kinase/dAdo kinase and dGuo
kinase/dAdo kinase activities is achieved by eluting from Blue Sepharose
affinity column with a bisubstrate mixture: dCyd plus ATP to release
dCyd kinase/dAdo kinase, and dGuo plus ATP to elute dGuo
kinase/dAdo kinase. (See Figure 3).
2. Further purification of dGuo kinase/dAdo kinase by UDP-Sepharose
affinity chromatography and anion-exchange HPLC cannot separate the
dGuo kinase and dAdo kinase activities.
3. Mutual noncompetitive activation of dGuo kinase and dAdo kinase
activities: dAdo phosphorylation is stimulated more than 3 fold by dGuo,
and dGuo kinase is also activated about 20% by dAdo.
4. Rates of turnover of dGuo kinase and dAdo kinase are additive when both
nucleosides, dGuo and dAdo, are present.
5. Kinetic mechanism of dGuo kinase activity is rapid equilibrium random bi
bi, and dAdo kinase shows ordered bi bi kinetic patterns, with ATP being
the leading substrate.
6 . Chemical inactivation of the dGuo phosphorylation site by 5’-[p-
(fluorosulfonyl)benzoyl]adenosine, an affinity probe, abolishes the
stimulation effect of dAdo kinase by dGuo.
Their results strongly suggest that dGuo kinase and dAdo kinase activities are situated on separate sites on a single protein. iue . le ehrs afnt crmtgah: lto b bisubstrate by Elution 7).al., et (ChakravartydCyd and kinase chromatography: dGuo affinityof Separation Sepharose mixture. Blue 3. Figure activity dGK o A dCK activity dCK A • Absorbance at 280nm at Absorbance • Units/Fraction 10 2 0 4 6 8 ^oQoxwftOfloaoQoaoQc” ^ A • •A S -Hi- \ 0 0 10 0 250 200 150 100 50 H H m 1 M . m dy 1 M dGuo mM 1 MgATP mM 5 MgATP mM 1 dCyd mM 0.5 AMP NADH mM 1 2mM 8.0 pH 6.6 pH A -X- -X rcin No. Fraction X X- -X 300 I c 0.1 0.2 0.3 0.4 0.5 0 280 10 11
Ikeda et a l (8 ) synthesized bisubstrate-type analogs, deoxynucleoside 5’- adenosine 5 ”’-P*,P^-tetraphosphate(dNp 4 A), and used them in combination with natural deoxynucleotide triphosphate (dNTP) end products, to study the deoxynucleoside kinases from L. acidophilus. Their results of inhibition specificity, inhibition patterns, and Kj(app) of dNp 4 A and dNTP are shown in
Table 3 and Table 4, respectively. The following conclusions were made:
1. Both dNTP and dNp 4 Abind to the active site of the corresponding kinase
through multiple binding determinants.
2. dNTPs are potent end product inhibitors of the corresponding kinases.
The high affinity of dNTP would be a combined effect of: (i) the optimal
fit of the deoxynucleoside moiety of dNTP at the deoxynucleoside binding
site, which provides the basis for its inhibition specificity, and, (ii) the
triphosphate group, which interacts with the ATP binding site.
3. dNp 4 A is a much weaker inhibitor than expected for an ideal
multisubstrate analog. The adenosine moiety does not appear to fit
optimally within the active site, whereas the tetraphosphate group is the
second binding determinant.
4. Putative modes of binding of substrates and multisubstrate analogs at the
active site are shown in Figure 4.
5. The kinetic mechanisms of dCyd kinase and dGuo kinase are each rapid
equilibrium random bi bi, whereas the dAdo kinase associated with dCyd
kinase or dGuo kinase appears to be steady state ordered bi bi. Table 3. Kinetic Constants for Inhibition of Deoxynucleoside Kinases by dNp4A
Kinetic constants for inhibition of deoxynucleoside kinases by dNp*A Kinase Substrate Inhibition Substrate Kinetic Inhibitor k : K.' assayed varied type* fixed mechanism* nM dCp+A dCyd dCyd 85(3) C 9.2 MgATP 1020 Random Bi Bi MgATP 1000 (3) C 8.0 dCyd 62
dGp«A dGuo dGuo 67 (4) c 1.4 MgATP 3000 Random Bi Bi MgATP 1700 (4) c 2.0 dGuo 54
dApiA dAdo I* dAdo NA' NC 2.3 MgATP 120 Ordered Bi Bi MgATP 86(5) c 2.9 dAdo NA dAp Kinase Substrate Inhibition Substrate Kinetic Inhibitor k : K, assayed varied type* fixed K; mechanism 11 fiM tlM mm dCTP dCyd dCyd 85 C 1.5 MgATP 950 Random Bi Bi MgATP 1000 C, NL 1.0 dCyd 72 dGTP dGuo dGuo 67 C 0.5 MgATP 1900 Random Bi Bi MgATP 1700 C, NL 0.5 dGuo 62 dATP dAdo I* dAdo NA/ NC 0.4 MgATP 85 Ordered Bi Bi MgATP C, NL 0.9 dAdo NA 86 dATP dAdo IP dAdo NA NC 0.5 MgATP 110 Ordered Bi Bi MgATP 110 C, NL 0.9 dAdo NA * Dissociation constant derived from initial velocity studies in previous papers (3-5). 6 The abbreviations are: C, competitive; NC, noncompetitive; NL, nonlinear. c Dissociation constant for fixed substrate, obtained from Figs. 6-9. d Kinetic mechanism, deduced from this study. * The dAdo kinase associated with dCyd kinase (presumed multifunctional enzyme). / NA, not applicable to second substrate in ordered mechanisms. * The dAdo kinase associated with dGuo kinase (multifunctional enzyme). 14 Figure 4. Putative modes of binding of substrates and multisubstrate analogs at the active sites of deoxynucleoside kinases. (Ikeda et al., 8 ). (i) substrates (deoxynucleoside and MgATP); (ii) dNp 4A; (iii) dNp 3A; and (iv) dNTP. Symbols: A = adenine, B = any base. R = ribose, dR = deoxyribose, and P = phosphate group. Ikeda et al (9) later utilized the multisubstrate analog, dCTP, to synthesize a new affinity medium, dCp 4 -Sepharose, and purified dCyd kinase/dAdo kinase to homogeneity with 60% recovery. Their studies demonstrated that: 1. dCyd kinase and dAdo kinase are confirmed to be associated on a single protein by native polyacrylamide gel electrophoresis. 2. SDS-polyacrylamide gel electrophoresis shows that the enzyme is composed of two subunits of similar size. 3. The N-terminal amino acid sequence of the intact protein was determined up to 28 amino acid residues from the N-terminus. See Figure 5. A single amino acid species per sequenation cycle appeared up to the 16th residue. However, at the 17th, 18th, 20th, 21st, 26th, and 27th residue positions of the sequence, two different amino acids in almost equal quantities were detected. 4. Residues 6 through 13 contain the highly conserved Gly-X-X-Gly-X-Gly- Lys sequence found at the active sites of kinases and other nucleotide- binding proteins. 16 123456789 10 Met-Ile-Val-Leu-Ser-Gly-Pro-Ile-Gly-Ala- 11 12 13 14 15 16 17 18 19 20 Gly-Lys-Ser-Ser-Leu-Thr-Ser-Leu-Leu-Ala- Gly lie Ser 21 22 23 24 25 26 27 28 Glu-Tyr-Leu-Gly-Thr-Gln-Ala-Phe Lys Asn Pro Figure 5. Partial N-terminal amino acid sequence of dCyd kinase/dAdo kinase from L. acidophilus. 17 B. DEOXYNUCLEOSIDE KINASES FROM OTHER SOURCES 1. Other Bacterial Deoxynucleoside Kinases Most bacteria possess thymidine kinase (TK), but the other three deoxynucleoside kinases, i.e. deoxycytidine kinase (dCyd kinase), deoxyadenosine kinase (dAdo kinase) and deoxyguanosine kinase (dGuo kinase), do not appear to be widely distributed. Besides L. acidophilus, all four deoxynucleoside kinases activities, so far, have only been found in extracts of Lactobacillus leichmamii (10) and Bacillus subtilis (11). However, B. subtilis does not seem to require all four kinase activities to grow, since a spontaneous mutant lacking dCyd kinase and dAdo kinase activities continues to grow. Pneumococci extracts (12) were reported to readily phosphorylate purine deoxynucleosides; however, no dCyd kinase was detected. dCyd kinase was also reported missing from Salmonella typhimurium (13). Deoxycytidine kinase activity was found in crude extracts of vegetative Bacillus subtilis (14,15), Bacillus megaterium KM (16), and a number of species of the class Mollicutes (17). Among the Mollicutes, only Spiroplasma ciri contained dGuo kinase; none has ATP-dependent dAdo kinase. All three kinases, dAdo kinase, dCyd kinase and dGuo kinase, are reported missing from cell-free extracts of Escherichia coli B (18). A highly specific thymidine kinase isolated from Escherichia coli (19,20,21,22,23) was shown to be allosterically inhibited by TTP and activated by dCDP. This enzyme has recently been cloned and sequenced (24). Thymidine kinase in 18 Bacillus stearothermophilus (25) was purified to homogeneity, and also found to be subject to end product inhibition. Sequence information regarding deoxynucleoside kinase from bacterial sources is limited; only thymidine kinase from E. coli has been cloned and sequenced as mentioned above. The E. coli TK amino acid sequence shows high homology to TK of vertebrates and large DNA viruses, but no similarity to known herpesviral TKs (24). 2. Mammalian Deoxynucleoside Kinases A number of deoxynucleoside kinases have been isolated from various mammalian system. Considerable variability in the properties of these enzymes has been observed between species and tissues. dCyd kinase has been isolated from calf thymus (26,27,28), murine neoplasm P815 (29), human lymphoblasts and myeloblasts (30,31,32), human leukemic spleen (33), and leukemic human T-lymphoblast (34,35). Some of them have broad deoxynucleoside substrate specificity. dCyd kinase purified from calf thymus cytoplasm (26,27,28) phosphorylates not only dCyd, but also Cyd, dGuo and dAdo. These nucleosides mutually inhibit the phosphorylation of one another. Human lymphoblasts and myeloblasts (30) contain dCyd kinase that also phosphorylates purine deoxynucleosides, even though dCyd is the preferred substrate. Similarly, dCyd kinase from human T-lymphoblasts (32) phosphorylates 19 dCyd, as well as dAdo and dGuo, but with much higher Km and kcat for the purine substrates, apparently at a common site. The purified enzyme has a molecular weight of 60,000, and a dimeric subunit molecular weight of 30,500. Recently, dCyd kinase from human T-lymphoblast MOLT-4 cells (36) has been cloned and sequenced. dCyd kinase from human leukemic spleen (33), which is strongly inhibited by dCTP, also phosphorylates dAdo and dGuo. The molecular weights of the native protein and subunits are similar to those of T-lymphoblast. dCyd kinase has been purified from both cytoplasm and mitochondria of acute myelocytic human leukemia blast cells (31). The two isozymes exhibit different substrate specificity; only the cytoplasmic dCyd kinase can phosphorylate cytidine, even though both purified isozymes have similar molecular weights, activation energies, and both catalyze the reaction by sequential mechanism. Another kinase of narrow specificity, the dCyd kinase purified from a murine neoplasm P815 (29), does not phosphorylate purine deoxynucleosides. dGuo kinase has been isolated from calf thymus mitochondria (28), skin extracts from pig (37), neonatal mouse (38) and human placenta (39). Some of these enzymes appear to have the capacity to phosphorylate dAdo as well, but are not yet pure, so phosphorylation of the deaminated product of dAdo is also possible. Rat liver mitochondria contain dThd, dCyd, dAdo, and dGuo kinase activities (40), but few details have been provided. Both dCyd and dGuo were 2 0 phosphorylated by a cell-free extract of mouse L1210 cells (41), and they were probably phosphorylated by the same enzyme. T-lymphoblast-specific nucleoside kinase (TSK) (42) partially purified from human MOLT 4F T cell extract, was also reported to phosphorylate dCyd, dAdo and dGuo. Thymidine kinase genes from human (43,44), chicken (45) and Chinese hamster (46) cells have been cloned and sequenced. There is about 70% homology between the coding regions of the human and chicken TK genes. 3. Viral Thymidine Kinases Many DNA viruses encode thymidine kinase. Since antiherpesvirus chemotherapy using nucleoside analogues depends on preferential activation by virus-encoded TK or preferential trapping of the drug by viral TK, a great deal of study has been targeted on herpesviral TK. The nucleotide sequences of many TK genes have been reported, including the TK gene of herpes simplex virus type 1 (47) and type 2 (48), marmoset herpesvirus (49), Marek’s disease virus, turkey herpesvirus (50), monkeypox and variola viruses (51). Thymidine kinase gene from Marek’s disease virus and turkey herpesvirus gene (50) derived amino acid sequences exhibit 58.2% identity. Compared to other TK genes, these sequences reveal greater homology to those of alpha- herpesviruses than those of gamma-herpesviruses, but exhibit no overall homology with chicken TK. 2 1 Balasubramaniam et al have compared regions of sequence homology among the herpesviral thymidine kinases (52). They observed 6 highly conserved sites and predicted secondary structure of the enzyme (will be discussed later in Chapter V). Site-directed mutagenesis was performed on the TK of vaccinia virus to identify the ATP-binding domain near the N-terminus (53), and the potential role of the aspartate residue in a conserved domain for magnesium binding. (54) The ATP-binding site of HSV-1 thymidine kinase was also identified by site-directed mutagenesis (55). Artificial mutants, which produced HSV-1 TK with major changes in protein stability and more temperature-sensitive were isolated (56). C. AFFINITY MEDIA Ikeda et al (57) have developed several affinity media for the purification of deoxynucleoside kinases. These media have been synthesized by linking deoxynucleosides to Sepharose through the 3’-hydroxyls or through various positions on the purine or pyrimidine bases. Although these media have successfully purified deoxynucleoside kinases from other sources, none of them retained the L. acidophilus deoxynucleoside kinases effectively. Multisubstrate- type affinity media (dNp 4 A-Sepharose), directed specifically toward the dCyd, 2 2 dAdo, or dGuo substrate sites (58), gave somewhat improved retention, but also have relatively limited capacity for bacterial kinases. It seems that the bulky adenosine portion of the dNp 4 Adoes not fit optimally at the ATP binding site of the bacterial kinases. The natural triphosphate end products (dNTP) were found to bind the active sites of the corresponding bacterial kinases relatively tightly (Kj = 0.4-3 juM). They inhibit the bacterial kinases even more strongly than do the synthetic dNp 4 A bisubstrate analogues (Kj = 1.4-9.2 pM) (8 ). Ikeda et aL proposed that the deoxynucleoside moiety of dNTP fits optimally at the deoxynucleoside binding site, while the triphosphate group of dNTP fits at the ATP binding site. These multiple binding determinants reinforce the affinity of dNTP and make it a potent end-product inhibitor. An affinity medium, dCp 4 -Sepharose,has been constructed by linking the dCTP to Sepharose through its terminal phosphate (9). This new affinity medium has been used to successfully purify dCyd kinase/dAdo kinase to homogeneity from L. acidophilus, with up to 60% recovery. D. CLONING VECTOR AND HOST CELL LINE 1. Cloning Vector The cloning vector used in preparing genomic libraries and subcloning for sequencing is the Bluescript phagemid vector, pBS(+)KS, from Stratagene. Bluescript is a high copy number plasmid. A diagrammatic representation of the 23 vector is shown in Figure 6 . Bluescript vector consists of the following major constructs: multiclonal site (MCS) polylinkers, T 7 and T 3 RNA polymerase promoters, an ampicillin resistant gene, a LacZ gene fragment, and the phage intergenic region (M13 related). As shown in Figure 6 , the MCS polylinkers contain 26 unique restriction enzyme sites. Flanking the polylinkers are the T 7 and T 3 RNA polymerase promoters which can be used to synthesize RNA in vitro. The choice of which strand of the insert cloned into the MCS is to be expressed, can be controlled by initiating the choice of promoter used for transcription. The polylinkers, as well as the Tyand T 3 promoters, are located in the N-terminal portion of a LacZ gene fragment. Expression of /3-galactosidase from the LacZ gene fragment of pBluescript results in blue colonies of the appropriate bacterial strains, when grown on X-gal and IPTG. But insertion of genes into the polylinker interrupts the coding region of the LacZ gene fragment on the vector, thus resulting in white colonies. Recombinants can easily be differentiated from non-recombinants by color selection. On the other hand, the bacterial cells transformed with the pBluescript phagemid, which will be ampicillin resistant, can be selected by ampicillin. Bluescript phagemids replicate autonomously as plasmids in bacteria; therefore, colonies will be obtained after transformation. When the bacteria have been infected by a helper phage, Bluescript will be rescued as single-stranded DNA from F-episome bearing bacteria. Bluescript vectors contain a 454 pBluescript SK +/-' Co/£i oh VAACAOCTATOACCATO* S' ATTAACCCTCACTAAAOS* S TCTAOAACTAQTOOATC* MET tmfN M M CM I M l XAftl O nN MfN I aattcqatatcaaocttatcqataccqtcoacctcoaocmmooocccmtacccaattcoccctataotoaotcotattacaattcactooccotcottttacaa X OCTATAOTTCaAATAOCTATOOCAOCTOOAOCTCCCCCCCOOOCCATOOQTTAAOCOOOATATCACTCAOCATAATQTTAAQTOACCOOCAOCAAAATarT S XOATATCACTCAOCATAA S’ Figure 6 . Diagrammatic representation of pBluescript cloning vector. 25 nucleotide intergenic region of Fj phage, which encodes all the cis-acting functions phages require for packaging and replication. pBluescript vectors are offered with the intergenic region in either of two orientations, i.e. M13+ and M13-. Bluescript M13+ is replicated such that the coding strand of fi- galactosidase gene is secreted in the phage particles; Bluescript M13- is replicated such that the non-coding strand is secreted. Both Bluescript M13+ and Bluescript M13- are offered in two polylinker orientations, KS and SK. The outermost restriction sites on the two ends of the polylinker are Kpn I and Sac I. In the KS orientation, the Kpn I restriction site is nearest the LacZ promoter, while the Sac I site is nearest the LacZ promoter in the SK orientation. Single-stranded DNA of Bluescript can be used for site-directed mutagenesis and constructing nested deletions. There are advantages to the use of Bluescript for site-directed mutagenesis. Since Bluescript plasmid does not replicate via the M13 cycle, the tendency toward deletion of the DNA insert would be avoided. Packaging of Bluescript containing inserts would be efficient because of its small size compare to the wild type M13. The resultant mutant transcript can be synthesized without subcloning. 2. Host Cell All three deoxynucleoside kinases, i.e. dCyd kinase, dGuo kinase and dAdo kinase, were reported missing from cell-free extracts of E. coli B (18). Therefore, 26 E. coli would be an appropriate recipient host for the genomic library construction. Escherichia coli XLl-Blue cells (59), developed at Stratagene, were used as the recipient strain in plasmid transformation. The genotype of XLl-Blue is: recA' (recAl lac' endAl gyrA96 thi hsdR17 supE44 relAl (F proAB lacfilacZAM15 TnlO}) Color selection of recombinant colonies containing Bluescript plasmids plus inserts can be performed by plating the cells on LB plates with ampicillin, tetracycline, X-gal and IPTG. XLl-Blue cells which did not receive the pBluescript could not survive in the presence of ampicillin. Colonies containing plasmids with no insert will turn blue with the supplements of X-gal and IPTG; while colonies containing plasmids with inserts will remain white. Since the AM15 lac gene carried on the F episome is needed for the blue/white color selection, host bacteria that have lost the F episome will remain as white colonies on an X-gal/IPTG plate even if the Bluescript plasmid contains no insert. Addition of tetracycline will reduce the false positives. CHAPTER II MATERIALS AND METHODS A. MATERIALS l,r-Carbonyldiimidazole, 6 -aminohexanol 1-phosphate and cyanogen bromide were purchased from Aldrich Chemical Co.. Ethyl trifluorothiol acetate and 10% aqueous purified Triton X-100 were obtained from Pierce Chemical Co.. The disodium salt of dATP was from American Research Products Company. N.N-dimethylformamide (Fisher) and methanol (Mallinckrodt) were distilled over CaH2 before use. Tributylamine (Aldrich Chemical Co.) was distilled over ninhydrin. Sepharose CL- 6 B and SephadexG-10 were from Pharmacia. Trisacryl M DEAE was obtained from IBF Biotechnics. Tritiated nucleosides were from ICN. The Photogene Nucleic Acid Detection Kit and biotin-7-dATP was purchased from Bethesda Research Laboratories. The modified Sequenase Kit T7 DNA Polymerase was purchased from United States Biochemical Corp. [a- '1C S]dATP was purchased from Amersham Corp. Biotin-7-dATP was from Bethesda Research Laboratories. Restriction endonucleases and DNA modifying enzymes were purchased from Boehringer Mannheim or Bethesda Research Laboratories. "Taq" polymerase was purchased from Amersham, Promega or 27 28 Perkin-Elmer/Cetus. Deoxynucleotides were from Perkin-Elmer/Cetus. Plasmid DNA purification cartridges were purchased from Qiagen. Plasmid pBluescript(+)KS was from Stratagene. NENSORB-PREP oligonucleotide purification cartridges were purchased from Dupont. Oligonucleotide primers were synthesized in the Biochemical Instrument Center of Ohio State University. Bradford reagent was purchased from Bio-Rad Laboratories. Reagents for gel electrophoresis were supplied by Bio-Rad. B. METHODS 1. Anion Exchange HPLC During the course of chemical synthesis of the affinity ligand, deoxyadenosine tetraphosphate, the products and by-products were analyzed by anion exchange HPLC, using a Pharmacia Fast Protein Liquid Chromatography (FPLC) system equipped with a Mono-Q column. The buffer used was a 0 - 0.5 M gradient of NaCl in 0.01 N HC1. The flow rate was 1.0 ml/min. 2. DEAE Trisaciyl M After coupling the activated imidazolide of dATP (I) to N-trifluoroacetyl- 6 - aminohexanol 1-phosphate (II), the product, P^-( 6 -(N-trifluoroacetyl)aminohex-l- yl)-deoxyadenosine tetraphosphate (III), was separated from by-products on a DEAE Trisacryl M column. The DEAE Trisaciyl M (370 ml) was packed in a 29 column. The column was washed with 2 liter of 2 M ammonium formate, then equilibrated with H2 O. Samples were dissolved in 10 ml of 50% methanol. One- half of the product m was applied to the column for each run. The column was then washed with 100 ml of 50% methanol, and then with 500 ml of H 2 O. The samples were eluted with a linear gradient generated from 0.2 M and 0.8 M ammonium formate, pH 4.5 ( 6 liter total). The flow rate was set at 50 ml/hr. and 2 2 ml fractions were collected. 3. Inorganic Phosphate Determination Total hydrolysis and inorganic phosphate determination was carried out as described by Clark and Switzer (60). A drop of 10% Mg(N 0 3 )2 in ethanol was added to the sample in a test tube. The solution was dried over a flame until white ash remained. After cooling, 0.5 ml of 1 N HC1 was added. The tube was capped and heated in boiling water bath for 15 minutes, then neutralized with 0.5 ml of 1 N NaOH. Aliquots of the hydrolysed sample and phosphate standards (0.1 - 0.5 umoles ofKH 2 P04)were adjusted to 0.3 ml and analyzed for inorganic phosphate content. To the 0.3 ml solution, 0.1 ml of acid molybdate, 0.1 ml of reducing reagent, and 0.5 ml of H 2 O were added and mixed well. After 20 minutes of incubation, absorbance at 660 nm was measured. Phosphate concentration of the sample was determined by interpolation from a standard curve. 30 Acid molybdate reagent: 136 ml concentrated H 2 S04, 25 g (NH^MoyC^I^O in 1 liter of water. Reducing reagent: 15 g NaHS0 3 , 5 g p-methylaminophenol (Elon) in 500 ml of water. 4. Culturing L. acidophilus R-26 Lactobacillus acidophilus R-26 stock was obtained from the American Type Culture collection (ATCC 11506). The bacterial cells were stored lyophilized at -20°C or maintained on 2% agar slants containing growth medium at 4°C. The bacteria were grown at 37°C as described previously (5) in the following growth medium: Ingredients gram/liter Bactopeptone 20 Bacto yeast extract 10 Tween 80 1 sodium citrate 0.25 manganese sulfate 0.002 k h 2 po 4 4.76 } results in pH 6 . 8 k 2 h p o 4 6.09 dextrose 20 Cells were harvested when growth reached late log phase (A 5 5 Q = 2.2). Cells were stored frozen at -70°C. 31 5. Extraction of Enzymes Enzymes were extracted from L. acidophilus R-26 as described previously (5,7). 10 grams of cells (wet weight) were resuspended in 100 ml of ice-cold extraction buffer (0.1 M Tris-HCl, pH 8.0, 3 mM EDTA) containing 0.1 mM phenylmethylsulfonylfluoride (PMSF). Cells were broken by flailing in a Biospec Bead Beater containing about 200 ml of glass beads, for a total of 20 minutes. The container was jacketed with ice. The contents were cooled for 5 minutes between each 5-minute run. The homogenized cell suspension was centrifuged to remove cell debris. The supernatant was treated with neutralized 10% streptomycin sulfate, at a ratio of 0.1 ml/ml of extract, to precipitate DNA. Then, the supernatant fraction was subjected to ammonium sulfate fractionation by bringing to 65% saturation, redissolving the sediment in 50 ml of extraction buffer, and reprecipitating in 65%-saturated ammonium sulfate. Finally, the pellet was dissolved in 20 ml of extraction buffer containing 50% glycerol. This ammonium sulfate fraction, IV, was stored at -20°C. 6. Protein Assay Protein concentration was measured by the Bradford method (61). Concentrated Bio-Rad dye reagent (0.2 ml) was added to 0.8 ml of protein solution. Absorbance at 595 nm was measured, and the protein concentration was determined by interpolation from a standard curve, using bovine plasma gamma globulin as a standard protein. 32 7. Enzyme Activity Assay Deoxynucleoside kinases activities were assayed using the disk anion- exchange method as described by Ives (62). The final concentration of reaction components for the dAdo kinase assay were: 10 mM ATP, 12 mM MgCl 2 ,20 mM dAdo, 0.5 f id ^H-dAdo and 0.1 M Tris-HCl, pH 8.0. For dGuo kinase and dCyd kinase assay, the substrates dAdo/^H-dAdo were substituted by dGuo/^H-dGuo and dCyd/ H-dCyd, respectively. The total assay volume was 40 /a 1. The reaction mixture was incubated at 20°C for 30 minutes. The reaction was stopped by adding 0.1 ml of 0.1 M formic acid. Aliquots of reaction mixture (20 /al) were then spotted onto 1 cm x 1 cm squares of Whatman DE-81 anion-exchange paper. These paper disks were washed free of unreacted substrate with circulating water for 30 minutes. After drying, the paper disks were placed into plastic scintillation vials. To elute radioactive nucleotide product, 0.2 ml of 0.1 M HC1 / 0.2 M KC1 were added and agitated for 15 minutes. Finally, 2 ml of liquid scintillation cocktail were added and radioactivity was counted in a scintillation counter. Scintillation cocktail was prepared by mixing 16.5 g of RPI preblend 2a60 (91% PPO and 9% bis-MSB) with 2 L of toluene and 1 L of Triton X-100. The dpm retained on the ion-exchange disk was directly proportional to the product the kinase produced, and the kinase activity was calculated as follows: % conversion = (cpm of sample - cpm of washed control) / cpm of unwashed control x 1 0 0 33 Enzyme Activity (nmole/min./^l) = fraction converted x nmole deoxynucleoside / 30 min. / n 1 enzyme One enzyme unit is defined as the amount of enzyme that catalyses the formation of 1 nmole of deoxynucleoside 5’-monophosphate per minute. 8 . SDS-Polyacrylamide Gel Electrophoresis Two different discontinuous buffer systems have been used for sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) using the Bio- Rad model 360 mini vertical slab-cell. a. The Laemmli Buffer System: Analytical SDS-PAGE was carried out according to Laemmli (63) . The stacking gel was 4% acrylamide / 0.125 M Tris-Cl, pH 6 . 8 and the separating gel was 12% acrylamide / 0.375 M Tris-Cl, pH 8 .8 . Protein samples were diluted 4X with sample buffer, and heated at 95°C for 4 minutes to denature. Sample buffer contained 62.5 mM Tris-Cl, (pH 6 .8 ), 2% (w/v) SDS, 8 mM DTT, 10% (v/v) glycerol, and 0.0013% (w/v) bromophenol blue. Running buffer was prepared as 5X stock by combining 4.5g Tris base, 21.6g glycine, 1.5g SDS, yielding a pH of 8.3. After electrophoresis, the gel was agitated in fixative for 30 minutes, then stained for 30 minutes, and finally destained overnight. The fixative contained 34 40% methanol/10% acetic acid (v/v); stain solution contained 0.1% Coomassie Blue R-250 in 40% methanol/10% acetic acid; destaining solution is 10% methanol/7.5% acetic acid. b. The pH 7.28 MZE 3328.IV buffer system: A modified SDS-PAGE with neutral discontinuous buffer system was used to separate the subunits of dGuo kinase/dAdo kinase for N-terminal protein sequence analysis as described by Moos (64,65,66,67). The 12% actylamide separating gel and 4% acrylamide (or 1.5% agarose) stacking gel was prepared the same way as above, except that the following buffers were substituted. Separating gel buffer: 0.12 M BisTris-HCl, pH 6.61 Stacking gel buffer: 0.125 M Tris-HCl, pH 6.80 Upper tank buffer: 0.113 M BisTris, 44 mM TES, pH 7.25 Lower tank buffer: 63 mM BisTris-HCl, pH 5.9 Pre-electrophoresis of separating gel was performed in separating gel buffer containing 0.1 mM sodium thioglycolic acid and 0.1% (w/v) SDS at 8 mA for 30 minutes. Sample were prepared as described above by incubating in sample buffer of the Laemmli buffer system at 95°C for 4 minutes. Electrophoresis was run at 8 to 12 mA. Protein bands in the gel were visualized by Coomassie Blue staining as described above. 35 9. Electroblotting Proteins or polypeptides resolved by SDS-PAGE were electrophoretically transferred onto a membrane, and polypeptide bands were cut out and submitted separately for N-terminal protein sequence analysis. Electroblotting of polypeptides from acrylamide gel to Millipore Immobilon-P membrane was performed as described by Kyhse-Andersen ( 6 8 ). The buffer reservoir was the stacked, soaked Whatman No. 1 filter paper. The transfer unit was assembled as following from bottom to top: 1 - (+ ) graphite electrode 2. stack of filter paper soaked in anode no. 1 buffer (0.3 M Tris, 20% (v/v) methanol, pH 10.4) 3. stack of filter paper soaked in anode no. 2 buffer (0.025 M Tris, 20% (v/v) methanol, pH 10.4) 4. Immobilon membrane 5. acrylamide gel 6 . stack of filter paper soaked in cathode buffer (40 mM 6 -amino-n- caproic acid, 25 mM Tris, 20% (v/v) methanol, 0.005% SDS, pH 9.4) 7. (-) graphite electrode Electroblotting was run at 0.8 mA / cm^ for 2 hours. The membrane was stained in 0.1% (w/v) Coomassie Blue R-250 / 40% (v/v) methanol / 10% (v/v) acetic acid for 15 minutes. Destaining was performed in 40% methanol / 10% 36 acetic acid for 15 minutes, then in 90% methanol / 7% acetic acid for seconds. 10. N-Terminal Protein Sequence Analysis Subunits of dGuo kinase/dAdo kinase separated by pH 7 SDS-PAGE were blotted onto Immobilon membrane (polyvinylidene difluoride). N-terminal protein sequence analysis of blotted protein was carried out as described by Matsudaira (69,70). Polypeptides blotted on membrane were excised and submitted to the Ohio State University Biochemical Instrument Center. N- terminal Protein sequencing, which was based on the Edman degradation chemistry, was performed on an Applied Biosystems Incorporated Model 470A Protein/Peptide Sequencer with on-line Model 120A PTH Analyzer and Model 900A Data Analysis Module. 11. Chromosomal DNA Preparation Chromosomal DNA was isolated from L. acidophilus R-26 according to Marmur (71) with modifications. A 2g to 3g (wet weight) portion of bacterial cells was suspended in 15 ml of SET buffer (20% w/v sucrose, 50 mM Tris-HCl pH 7.6, and 50 mM EDTA). Cells were treated with lysozyme and lysed by SDS at 60°C. The cell lysate was extracted with chloroform/isoamyl alcohol (24:1, v/v) to denature protein and extract lipids, and the nucleic acids were precipitated from the aqueous phase with ethanol. Chromosomal DNA was spooled with a glass rod and redissolved in TEN buffer (10 mM Tris-HCl, pH 7.6,1 mM EDTA, 37 and 10 mM sodium chloride). The yield was about 20 mg. Chromosomal DNA was further purified by cesium chloride-ethidium bromide equilibrium centrifugation according to the Maniatis manual (72). 12. Oligonucleotides Oligonucleotides were synthesized on an Applied Biosystems Incorporated Model 380B oligonucleotide synthesizer, utilizing /3-cyanoethyl phosphoramidite chemistry, at the Ohio State University Biochemical Instrument Center. The 5’- trityl group was left on the oligonucleotides when hydrolyzed from the controlled- pore resin on the instrument, facilitating later purification using the NENSORB PREP cartridge. Deblocking was carried out by incubation in concentrated ammonium hydroxide at 55°C for 6 to 12 hours to remove the protecting groups on the nucleotide bases. Sodium hydroxide was added to a final concentration of 5 mM. The volatile solvent was evaporated in a rotary evaporator under vacuum created by water aspiration. Oligonucleotides were purified on a Dupont NENSORB PREP oligonucleotide purification cartridge according to the protocol described by the manufacturer. Trityl-on oligonucleotides were applied to the cartridge in 0.1 M triethylammonium acetate, pH 7.0 (TEAA). Trityl-on oligonucleotides would bind to the cartridge while salts, failure sequences, and synthetic by-products were washed away by acetonitrile/TEAA, pH 7.0 (1:9, v/v). The trityl group was then hydrolyzed from the 5’ end of the oligonucleotide with 0.5% trifluoroacetic acid. Purified oligonucleotides were eluted with 35% 38 methanol. 13. Polymerase Chain Reaction (PCR) The method of polymerase chain reaction DNA amplification using thermostable Taq DNA polymerase was first described by Saiki et aL (73). To amplify the portion of dCyd kinase gene encoding the N-terminal 26 amino acids, 1 fig of total genomic DNA was subjected to PCR for 30 cycles in a total volume of 50 fil with 3 nM to 5 mM degenerate primers. The reaction mixture also contained a final concentration of: 200 fiM each of dNTP (dATP, dCTP, dGTP and dTIT); 2.5 mM MgC^; 50 mM KC1; 10 mM Tris-HCl, pH 8.4 at room temperature); and 1.25 units of Taq polymerase. PCR was carried out in a Perkin Elmer Cetus DNA Thermal Cycler or EriComp TwinBlock DNA Thermal Cycler. After initial denaturation at 94°C for 2 minutes, the samples were run for 30 cycles as follows: denatured at 94°C for 1 minute, annealed at 37°C for the first five cycles and 50°C for the subsequent 25 cycles for 1 minute, and extended at 72°C for 30 seconds. At the end of the last cycle, the reaction was extended at 72°C for an additional 7 minutes. The PCR products were separated by agarose gel electrophoresis. The desired size fragment was extracted from the gel and reamplified by PCR under the same conditions as described above, except for annealing at 50°C throughout all 30 cycles. 39 14. Preparation of 32P-labeled DNA Probe Synthetic oligonucleotide [+1], see Table 5, was labeled with [y- 3 2 P]-ATP for use as an internal probe for hybridization to the PCR product. In a 30 pi labeling reaction, 0 . 1 /xg of degenerate oligonucleotides was incubated with 2 0 0 /xCi of [y-3 2 P]-ATP, IX kinase buffer, and 5 units of T4 polynucleotide kinase at 37°C for 30 minutes. 10X kinase buffer was 0.7 M Tris-HCl, pH 7.6, 0.1 M magnesium chloride, 1.0 M potassium chloride, 50 mM DTE, and 5 mg/ml bovine serum albumin. J P-labeled oligonucleotides were then purified on a Dupont NENSORB-20 cartridge to remove unreacted 3 2 P-ATP, salts and proteins according to the manufacturer. The specific activity of the labeled oligonucleotides was typically 7x10^ cpm/pg. 15. Synthesis of Biotinylated DNA Probe by PCR Biotin-7-dATP was incorporated into DNA cloning probes, used for noniostopic nucleic acid detection, by PCR. PCR conditions were the same as described above except for using 150 fiM dATP and 50 /iM biotin-7-dATP. One ng of template DNA was subjected to 20 to 30 cycles of amplification. The biotinylated PCR product was purified on a QIAGEN tip-5 anion-exchange column to remove primers and unincorporated deoxynucleotides, as described by the manufacturer. 40 16. Gel Electrophoresis for DNA Agarose gel or acrylamide gel electrophoresis were carried out as described in the Maniatis manual (72). Agarose gels of 0.8% to 4% were run using the Hoefer Scientific Instruments MINNIE submarine agarose gel unit Model HE 33. Electrophoresis in acrylamide gels of 6 % to 8 % were performed in the Bio-Rad Model 360 mini vertical slab-cell. TBE buffer (89 mM Tris- borate, 89 mM boric acid, and 2 mM EDTA) was used in both the gel buffer and tank buffer. Just before loading, 6 X loading buffer (0.25% bromophenol blue and 40% sucrose, w/v) was added to the DNA sample. Electrophoresis was run between 20 volts to 100 volts at constant voltage. DNA bands were visualized by staining in a 0.5 /zg/ml ethidium bromide solution. 17. Elution of DNA from Agarose Gel DNA resolved by agarose gel electrophoresis was electroeluted utilizing the IBI Model-UEA Unidirectional Electroelutor according to the manufacturer. Running buffer A was 20 mM Tris-HCl, pH8.0, 0.2 mM EDTA, and 5 mM sodium chloride. Electroelution was performed at 100 volts (constant voltage) for 50 minutes. Once eluted from the gel, DNA would be retarded by the high salt buffer (3 M sodium acetate, pH 5.1, 0.01% bromophenol blue) in the V-shape channel. Another method of gel elution utilized low melting point agarose gel. The gel piece containing the desired size of DNA was melted at 65°C and the extract 41 was purified on a QIAGEN tip-5 anion-exchange column as described by the manufacturer. 18. Southern Transfer and Hybridization Chromosomal DNA digested with restriction enzymes was resolved in a 0.8% agarose gel and transferred onto a Photogene nylon membrane (Bethesda Research Laboratories) by capillary action, according to Southern (74,72). After transfer, the membrane was baked at 80°C in a vacuum oven for 1 hour to fix DNA onto the membrane. Prehybridization was carried out in 6 xSSPE / 5x Denhardt’s solution / 1% SDS / 200 Mg/ml herring sperm DNA at 51°C for 2 to 4 hours. The DNA probe, labeled with biotin by PCR as described above, was added to the prehybridization solution at a final concentration of 50 ng/ml. Hybridization was carried out in an agitating water bath at 51°C for more then 12 hours. Blots were washed twice at 60°C in 2x SSC / 0.1% SDS for 10 minutes, followed by one 60°C wash in O.lx SSC / 0.5% SDS for 30 minutes. Signals were detected photochemically (see below). 19. Nonisotopic Nucleic Acid Detection DNA immobilized on nylon membrane was detected non-radioactively by the Gibco BRL Photogene Nucleic Acid Detection System according to the manufacturer’s instruction. A biotin-labeled DNA probe was first hybridized to 42 the immobilized DNA. Streptavidin-alkaline phosphatase conjugate was then bound to the biotin group. After that, a substrate for the alkaline phosphatase, 4-methoxy-4-(3-phosphatephenyl)spiro[lm2-dioxetane-3,2’-adamantane], which luminesced when dephosphorylated, was incubated with the membrane. The emitted light could be detected on X-ray film (Kodak X-OMAT AR). 20. Plasmid DNA Preparation The cloning vector pBluescript( + )KS from Stratagene was used for library construction and subcloning. Plasmid DNA was isolated and purified either by cesium chloride-ethidium bromide equilibrium centrifugation as described in the Maniatis manual (72), or by QIAGEN tip-500 anion-exchange chromatography, according to the manufacturer. 21. Competent Cells and Transformation E. coli XLl-Blue cells (Stratagene) were used as the recipient cells in the construction of genomic libraries and subcloning. To prepare competent cells, XLl-Blue cells were grown in 250 ml LB broth (1% bactotryptone, 0.5% bacto yeast extract, and 1% sodium chloride; w/v)(75) at 37°C until cell density reached A ^ q = 0.2. Cells were harvested and resuspended in 100 mM magnesium chloride. Then, cells were treated with 100 mM calcium chloride and finally resuspended in 2.5 ml of 100 mM calcium chloride / 15% (v/v) glycerol. Competent cells were stored at -70°C in 0.2 ml aliquots for months. 43 Competent XLl-Blue cells were transformed with pBluescript, with or without inserts, by incubating 1 to 10 ng of DNA with 20 n\ of cells at 4°C for 30 minutes. After a 2-minute heat shock at 42°C, 80 fil LB was added and incubated at 37°C for 1 hour. Cells were spread on LB plates containing 75 fig/ml ampicillin and 15 Mg/ml tetracycline. Typical transformation efficiency was about 400,000 tranformants per ng of DNA 22. Colony Hybridization The Xba I and Kpn I genomic libraries were screened by colony hybridization. Bacterial colonies were replicated on Gibco BRL Photogene nylon membrane by contact and grown to about 1 mm in diameter as described in the Miniatis manual (72). In order to reduce the background signal in colony hybridization, the blots were scrubbed in 5x SSC / 0.5% SDS with kimwipes to remove cell debris after baking in a vacuum oven at 80°C. The blots were then treated with proteinase K (200 mg/ml in 100 mM Tris-HCl, pH 7.5 / 150 mM NaCl) at 37°C for 1 hour (76) prior to pre-hybridization. Sonicated and denatured pBluescript plasmid DNA (5 ng/mi) was added to the prehybridization solution to minimize nonspecific binding of the probe. The DNA probes (93mer or 117mer PCR products) were labeled non-isotopically with biotin as described above. Prehybridizaion and hybridization was carried out the same way as for the Southern blot (see above, Southern Transfer and Hybridization). Signals were detected nonisotopically by the Gibco BRL Photogene Nucleic Acid Detection 44 System (see above). 23. Genomic Library Screening by Polymerase Chain Reaction Colonies, picked from replica plates, were grown to 2 to 3mm in diameter on LB plates (1% bactotryptone, 0.5% bacto yeast extract, and 1% sodium chloride; w/v) containing ampicillin. These replica colonies were scraped with sterile toothpicks and resuspended in 50 nl of TTE buffer (1% Triton X-100, 20 mM Tris-HCl, pH8.5 and 2 mM EDTA). 10 to 20 colonies were screened per PCR reaction. After boiling for 5 minutes to lyse the cells, cell debris was pelleted by centrifugation. Aliquots (5 pi) of the supernatant fraction were used directly to provide templates for PCR, amplifying for 35 cycles in a total volume of 10 pi. PCR conditions have been described above. DNA was denatured at 94°C for 1 minute, annealed at 37°C for 1 minute for the first 5 cycles, and 50°C for the subsequent 30 cycles, and then extended at 72°C for 30 seconds. Primers [+2] and [-3] were used to screen the Xbal library; primers [+4] and [-5] were used for the Kpnl library. 24. DNA Sequencing of Cloned DNA Plasmid DNA templates used for double-stranded sequencing were prepared from E. coli XLl-Blue cells on a QIAGEN tip-20 anion-exchange column. DNA sequences were determined by employing the dideoxynucleotide chain termination method of Sanger et al (77). The Sequenase 2.0 kit from 45 United States Biochemical Corporation was used, and DNA was labeled with [a- -3C S]-dATP. Each strand of the gene was repeatedly sequenced by subcloning or gene-walking. Restriction fragments of the positive clones were subcloned into pBluescript and sequencing reactions were primed with either the universal M13(- 40) or the reverse M13(-48) sequencing primers. Sequence-specific primers were also used for gene-walking. Regions of compression were resolved by using dlTP in place of dGTP in the reactions. DNA was resolved by denaturing gel electrophoresis on either 8 % acrylamide / 8 M urea/IX TBE gel or6 % Long Ranger/7 M urea/1.2X TBE gel. HydroLink Long Ranger gel solution was from AT Biochem, Inc. and was prepared according to the manufacturer. 25. Asymmetrical PCR and DNA Sequencing To sequence the PCR product directly, asymmetrical PCR was employed to generate single-strand enriched DNA (78,79,80). Sense primer and antisense primer used in the reaction were at a 1 to 2 0 molar ratio, or vice versa. The lower concentration primer was set at 10 nM so it would be depleted after 10 to 15 cycles. 1 to 10 ng of template DNA was used. Other reaction components were the same as for the regular PCR described above, in a total volume of 100 /xl. DNA was initially denatured at 94°C for 1 minute, followed by 25 cycles of: 94°C denaturation for 1 minute, 50°C annealing for 1 minute, and 72°C extension for 30 seconds. An extra step of 72°C for 7 minutes was added 46 after the last cycle to allow full extension of the PCR products. Single-strand enriched PCR products were purified on a QIAGEN tip-5 anion-exchange column to remove primers and nucleotides. DNA sequencing was performed, using the lower concentration PCR primer as the sequencing primer. Again, the dideoxynucleotide chain termination sequencing method of Sanger et aL was used (77). 26. Extraction and Purification of Enzymes from K coli clones. E. coli XLl-blue cells transformed with pBluescript, with or without inserts, were grown at 37°C in LB broth containing ampicillin until the cell density reached an absorbance at 600 nm of about 1.0. Where applicable, IPTG was added when cell density reached an absorbance of 0.2. Cells were harvested by centrifugation in a Beckman microfuge for 10 minutes at 4°C and resuspended in extraction buffer (0.1 M Tris-HCl,pH 8.0,3 mM EDTA, 20% glycerol). PMSF was added to a final concentration of 0.2 mM. Immediately, cells were opened by sonication at a 20% pulse setting on ice for a total of 6 minutes, allowing a 5- minute cooling between each 2-minute run. Cell debris were removed by centrifugation in a Beckman Microfuge Model E for 10 minutes at 4°C. Crude extracts were further purified by streptomycin fractionation, ammonium sulfate fractionation and affinity chromatography, following the same procedure as for the wild type L. acidophilus dGuo kinase/dAdo kinase (5,81). CHAPTER III AFFINITY PURIFICATION OF DEOXYGUANOSINE/DEOXYADENOSINE KINASES FROM L. ACIDOPHILUS The dAp 4 -Sepharose affinity medium synthesis and protein purification works were done in collaboration with Dr. Seiichiro Ikeda. A. CONSTRUCTION OF dAp4-SEPHAROSE AFFINITY MEDIUM 1. Development The paired enzymes, deoxyguanosine kinase (dGuo kinase)/deoxyadenosine kinase (dAdo kinase) from L. acidophilus have been purified by Chakravarty (7) using conventional purification procedures: Blue Sepharose affinity chromatography, UDP-Sepharose affinity chromatography and anion-exchange HPLC. However, only an extremely small amount (0.1 Mg) of the pure protein could be obtained after the tedious procedures, with only 2 % recovery. Since the bacterial dGuo kinase/dAdo kinase comprises only a small portion of the total cellular protein, the synthesis of a suitable affinity media, that retains the enzymes specifically, would be critical for further study of the enzymes. Several affinity medium have previously been constructed in our laboratory: affinity media linking deoxynucleosides to Sepharose (57), and 47 48 dNp 4 A-Sepharose (58) (discussed in Chapter I). But none of them quantitively retain L. acidophilus deoxynucleoside kinases. The natural triphosphate end products (dNTP) were found to bind the active sites of the corresponding bacterial kinase tightly (Kj = 0.4-3 /xM), inhibiting even more strongly than the synthetic dNp 4 A bisubstrate analogues (Kj = 1.4-9.2 pM) (8 ). Ikeda et al proposed that the deoxynucleoside moiety of dNTP fits optimally at the deoxynucleoside binding site, while the triphosphate group of dNTP fits at the ATP binding site. These multiple binding determinants reinforce the affinity of dNTP and make it a potent end-product inhibitor. They have then constructed an affinity medium, dCp 4 -Sepharose, by linking the dCTP to Sepharose through its terminal phosphate (9). This affinity medium has been used to successfully purify dCyd kinase/dAdo kinase to homogeneity from L. acidophilus, with 60% recovery. To purify the paired enzyme dGuo kinase/dAdo kinase from L. acidophilus, another affinity medium (dAP 4 -Sepharose), analogous to dCp 4 * Sepharose, was newly synthesized. The potent natural inhibitor of dAdo kinase, dATP, was linked to the Sepharose through its terminal phosphate. Therefore, this medium should interact specifically with dAdo kinase. The ammonium sulfate fraction of L. acidophilus was first pass through the dCp4 ~Sepharose to remove dCyd kinase and the associating dAdo kinase. dGuo kinase/dAdo kinase in the run through fraction from dCp 4 -Sepharose was then purified by dAp 4 - Sepharose. 49 Since the charged groups of dNTP seem to be important in binding to the kinase (8,9), an extra phosphate group was inserted between the dATP and a hexyl group to compensate for the fourth negative charge of triphosphate lost in the covalent attachment. Cyanogen bromide activation was chosen as the means of attaching spacer arms to Sepharose, because the linkage is stable. The structure of dAp 4 -Sepharose is shown in Figure 7. 2. Chemical Synthesis and Characterization The new affinity medium, dAp 4 -Sepharose, was constructed by employing the method used for the preparation of affinity ligands for dCyd kinase (9) and dGuo kinase (82) described previously. The tetraammonium salt of dATP was first derivatized by carbonyldimidazole to form the imidazolide of dATP (I). This compound was coupled to N-trifluoroacetyl- 6 -aminohexanol 1-phosphate (II) to form P^-( 6 -(N- trifluoroacetyl)aminohex-l-yl)-deoxyadenosine tetraphosphate (III). By-products of the reaction were separated by means of a DEAE Trisacryl M column. The product P^-( 6 -aminohex-l-yl)-deoxyadenosine tetraphosphate (IV) was obtained by removing the protective trifluoroacetyl group under alkaline conditions at room temperature overnight. This ligand was finally coupled by reaction of the amine derivative with CNBr-activated Sepharose, to form an affinity medium containing dAdo tetraphosphate covalently attached via a six-carbon chain to the matrix. The scheme of synthesis is outlined in Figure 8 . The details of synthesis and o® 0® 0® 0® NH2 -o—C-NH—CH2 Figure 7. Chemical structure of dAp4-Sepharose affinity medium. 51 dATP (sodium salt) 6 -aminohexanol- 1 -phosphate 1 1 - ethyl trifluorothiol acetate dATP (pyridinium salt) N-trifluoroacetyl- 6 -aminohexanol- 1 -phosphate (in methanol) i 1 dATP (tributylamine salt) N-trifluoroacetyl- 6 -aminohexanol- 1 *- 1 , l’-carbonyldiimidazole l-phosphate(tributyiamine salt) (II) imidazolide dATP (I) P^-(6 -(N-trifluoroacetyl)aminohex-l-yl)-deoxyadenosine tetraphosphate (III) + by-products I DEAE Trisacryl M (370 ml) purified (III) i pH 11.5 P^-(6 -aminohex-l-yl)-deoxyadenosine tetraphosphate (IV) I Sephadex G-10 desalted (IV), P^-(6 -aminohex-l-yl)-deoxyadenosine tetraphosphate Figure 8 . Synthesis scheme of P^-( 6 -aminohex-l-yl)-deoxyadenosine tetraphosphate. 52 characterization will be discussed below. Preparation of imidazolide of dATP (Ik The disodium salt of dATP, 0.762 mmoles dissolved in water, was converted to the pyridinium salt by passing through a cation-exchange Dowex-50 column in the H +-ion form. Three molar equivalents of tributylamine were added to the dATP (evaporated and redissolved in methanol) and stirred for 20 minutes at room temperature. The solution were concentrated on a rotary evaporator (warmed in a 40°C water bath) and the syrupy residues were dried by repeated addition of anhydrous dimethylformamide (DMF) and evaporation. To the dATP redissolved in 10 ml of anhydrous DMF, 3.819 mmoles of l,l’-carbonyldiimidazole (about five molar equivalent) were added and stirred for 4 hours at room temperature in a desiccator over phosphorous pentoxide. Then, 0.154 ml of dry methanol (3.810 mmoles) was then added and stirred for an additional 30 minutes. This activated imidazolide of dATP (I) was ready to couple to N-trifluoroacetyl- 6 -aminohexanol 1-phosphate (II). Preparation of N-trifluoroacetyl- 6 -aminohexanol 1-phosphate (II): Before activating the 6 -aminohexanol 1-phosphate, the primary amino group first has to be blocked with an acetylating reagent, according to Barker et al (83). Ethyl trifluorothiol acetate was added in molar excess to 1.52 mmoles of 6 -aminohexanol 1-phosphate. During the course of the amino-blocking 53 reaction, the pH of the solution was maintained at pH 9.5 by additions of 5N lithium hydroxide. The completion of the blocking reaction was monitored by analyzing small aliquots of reaction mixture with ninhydrin. Once the reaction was completed, the reaction mixture was adjusted to pH 5 with trifluoroacetic acid. The solution was concentrated on a rotary evaporator (40°C water bath), redissolved in water, and passed through a Dowex 50-X8-H+ column to remove unreacted amine. A yield of 1.50 mmoles of N-trifluoroacetyl- 6 -aminohexanol 1- phosphate were obtained, as determined by inorganic phosphate analysis (60), as described in Chapter II. The compound (in 25 ml methanol) was then converted to tributylammonium salt by stirring with equal molar of tributylamine (0.362 ml) at room temperature. The solution was concentrated as above, and dried by repeated addition and evaporation of anhydrous DMF. This tributylammonium salt of N-trifluoroacetyl- 6 -aminohexanol 1-phosphate (II) was ready for coupling with (I). Preparation of P^-aminohex-l-vn-deoxvadenosine tetraphosphate (IV): The tributylammonium salt of N-trifluoroacetyI- 6 -aminohexanol 1- phosphate (II) was redissolved in 5 ml of DMF. This solution (II) was added to the activated imidazolide of dATP (I) and stirred for 7 days in a vacuum desiccator over phosphorous pentoxide. The formation of the coupled product P^-(6 -(N-trifluoroacetyl)aminohex-l-yl)-deoxyadenosine tetraphosphate (III), was detected, along with several by-products, by high performance liquid anion 54 exchange chromatography (Pharmacia FPLC system, Mono-Q column), using a linear 0 - 0.5 M gradient of NaCl in 0.01 N HC1. The FPLC elution profile showed the presence of compound (III), and dATP, as well as significant amounts of other by-products (Figure 9). After the solvent was evaporated, the residue was dissolved in 10 ml of methanol/H 2 0 (1:1, v/v). Half of the reaction products at a time was applied to the DEAE Trisacryl M (370 ml, formate form) column at 4°C. The column was washed with 100 ml of 50% methanol, and then with 500 ml of H 2 O. The products were eluted with a gradient generated from 0.2 M - 0.8 M ammonium formate (pH 4.53), Figure 10. Fractions of the major UV absorbing peaks were analyzed by FPLC as above. The fractions containing compound (III) were combined and lyophilized. The residue was dissolved in H 2 O and incubated at pH 11.5 at room temperature overnight. The resulting free amine (IV) was desalted through a Sephadex G-10 column and the aqueous solution was lyophilized to dryness. The total yield of (IV) was 0.288 mmoles, a yield of 30.5% based on the amount of dATP used. As determined by Dr. Ikeda (84), compound (III) gave 3.8 and 0.92 molar ratios of total phosphate and amine, respectively, per mole of adenine, based on the UV absorption of the base. The compound was found to be stable to alkaline phosphatase, confirming the proposed structure, which has no terminal phosphate. It inhibits L. acidophilus dAdo kinase activity (of an ammonium 55 o ' dATP-PHA-COCF: dATP Ht o Fraction No, Figure 9. FPLC Mono-Q anion exchange elution profile of (III) and by-products. Buffer used was 0 - 0.5 M gradient of NaCl in 0.01 N HC1. The scale was 0.1 O.D. 56 dATP-PHA-COCF Fraction No, Figure 10. DEAE Trisaciyl M elution profile of (III) and by-products. Buffer used was 0 - 1 M ammonium formate, pH 4.5. 57 sulfate fraction) noncompetitively versus dAdo (Kj(app) = 37 /uM at 0.1 mM MgATP). Preparation of affinity adsorbent. dAp/|-bound Sepharose: Compound (IV) was coupled to Sepharose CL- 6 B according to the simplified CNBr activation method of March, et aL (85). Activation of Sepharose (2 - 20 ml) by CNBr (0.2 g per ml of Sepharose) was carried out at 4 °C for 2 minutes. Ligands were coupled to activated Sepharose at 4°C for 1 to 3 days under conditions described below. After the coupling reaction, 1 ml to 21 ml ligand- bound Sepharose was packed in columns (10 - 16 mm diameter) and washed extensively with cold water. The columns were stored at 4 °C. The concentration of ligand bound to Sepharose was calculated from the decrease in the absorbance of the supernatant of the coupling reaction mixture. Different batches of dAp 4 -Sepharose with varying amount of ligand bound to Sepharose were obtained by changing the concentration of soluble ligand in the coupling reaction and the coupling time. The substitution of dAp 4 -Sepharose preparation obtained under certain conditions are as follows: (i) 1.32 Mmole bound ligand/ml Sepharose from 3.58 Mmole soluble ligand/ml Sepharose by 20 hrs coupling reaction; (ii) 2.04 Mmole bound ligand/ml Sepharose from 2.78 Mmole soluble ligand/ml Sepharose by 43 hrs. coupling; (iii) 1.50 Mmole bound ligand/ml Sepharose from 2.73 Mmole soluble ligand/ml Sepharose by 72 hrs. coupling. 58 The UV absorption spectrum (from 230 nm to 350 nm) of dAp^Sepharose was taken directly, with underivatized Sepharose as a blank as described by Ikeda et aL (57). Moist gel, 56 mg, with or without ligand bound, was suspended in 2 ml of 0.05 M sodium phosphate buffer, pH 7.0, containing 75% glycerol. The UV absorption spectrum of deoxyadenosine was also measured in the same way. As shown in Figure 11, dAP^Sepharose exhibited the absorption maximum at 259 nm characteristic of dAdo derivatives. The concentration of ligand measured from the spectrum was found to be 50 - 80 % of the value calculated above. B. PURIFICATION OF DEOXYGUANOSINE KINASE/ DEOXYADENOSINE KINASE BY dAp4-SEPHAROSE AFFINITY CHROMATOGRAPHY The ammonium sulfate fraction of L. acidophilus, prepared as described in Chapter II, was first applied to a dCp^Sepharose affinity column, to retain dCyd kinase/dAdo kinase according to Ikeda et aL( 9). The run-through fraction from the dCp^Sepharose column had been stored at -20 °C and was used as a source for further purification of dGuo kinase/dAdo kinase since it had been proved to contain nearly 100% of the dGuo kinase activity, but only 1-3% of the dCyd kinase activity of the original activities found in the ammonium sulfate fraction (9). dAp 4 ~Sepharose columns ranging from 1 ml to 21 ml have been run. Each dAp 4 -Sepharose column was equilibrated with Buffer A (15 mM potassium 59 259 nm 56 mg dAP4-Sepharose dAdo 230 250 270 290 310 330 350 nm Figure 11. UV absorption spectra of dAp4-Sepharose and deoxyadenosine. 60 phosphate buffer, pH 8.0, and 20% glycerol). The run through fraction from dCp 4 -Sepharose was applied to the dAp 4 -Sepharose column. Then, the column was washed with (i) Buffer A, (ii) 0.1 M KC1 in Buffer A, and (iii) Buffer A, to remove non-specifically bound impure proteins. dGuo kinase/dAdo kinase activities were specifically eluted with 0.3 mM dATP in Buffer A. A typical elution profile is shown in Figure 12. After each run of affinity chromatography, the column of dAp 4 -Sepharose was regenerated by washing with 10 column volumes each of 1 M KC1 in 15 mM K-phosphate (pH 8.0), 0.5% Triton X-100 in 15 mM K-phosphate (pH 8.0), and 6 M guanidine-HCl (pH 7.0), successively, and reequilibrated with buffer A. dGuo kinase/dAdo kinase active fractions were combined and concentrated to a minimum volume by Amicon Centricon ultrafiltration (10 KDa cutoff). The purified enzyme preparations were stored at -20 °C in the presence of dATP to stabilize the enzyme. The enzyme survives years of long term storage at -70°C with most of the activities intact. Both dGuo kinase/dAdo kinase activities were specifically eluted with 0.3 mM dATP, confirming that the two activities remained associated with each other. The activity ratio of dGuo kinase to dAdo kinase remained almost unchanged throughout the affinity purification (dGuo kinase/dAdo kinase = 5 - 8 in the absence of any activator or inhibitor). The final preparation of dGuo kinase/dAdo kinase after dAp 4 -Sepharose was shown to be pure by SDS-PAGE (see below). The specific activity increased by about 2,700 fold for both dGuo kinase (2,150 units/mg) and dAdo kinase (280 61 0 3mM 0.1M KCI Buffer dATP in Run thru Buffer A in Buffer A A Buffer A 0.5 T 0.4 31% Zs J > S 0.3 • ^ "o « 0.1 15% 0.0 (~^'Q~g k > 0 "Q— o 100 200 300 Fraction Number Figure 12. Purification of dGuo kinase/dAdo kinase by dAp4-Sepharose affinity chromatography. A 20 ml dAp4-Sepharose was equilibrated with Buffer A (15 mM potassium phosphate buffer, pH 8.0, 5 mM EDTA, 20% glycerol.) 75 ml of run through fraction from dCp4-Sepharose affinity column (13 units dAdo kinase) was applied to the dAp4-Sepharose column. 1.2 ml fractions were collected. 62 units/mg). Unfortunately, dAp 4 -Sepharose tended to degenerate relatively rapidly, compared with dCp 4 -Sepharose. Addition of 5 mM EDTA to all column buffers slowed down the decay, suggesting enzymatic action, but could not completely prevent the degeneration. Moreover, the properties of the column also gradually changed with use, i.e. impurities were eluted increasingly with 0.3 mM dATP, along with dAdo kinase/dGuo kinase. Increasing the concentration of KC1 in the washing step did not remove these impurities, but, even worse, reduced the recovery of the enzyme in the elution step. C. CHARACTERIZATION OF AFFINITY PURIFIED DEOXYGUANOSINE KINASE/DEOXYADENOSINE KINASE 1. Enzyme Purity and Subunit Molecular Weight dGuo kinase/dAdo kinase preparations purified from dAp 4 -Sepharose columns have been run on SDS-polyacrylamide gels with the Laemmli buffer system (63), and with the pH 7 MZE 3328.IV buffer system (64). By the Laemmli method, a single protein band was detected when visualized by Coomassie blue staining or silver staining (Figure 13), indicating that the dAp 4 - Sepharose was very specific to dAdo kinase, and impure proteins were effectively washed out. The molecular weight of the denatured protein band determined from SDS-PAGE was 26,000 Da (Figure 14). 63 5 6 Daltons 97.400 - 66 ,20 0 - 45,000" yy.b>w\ 31,000 " 21,500 " 14.400 Figure 13. SDS-polyacrylamide gel analysis of purified dGuo kinase/dAdo kinase. (12% gel, Laemmli buffer system). Gel stained by Coomassie Blue. Lane 1 & 7, protein markers; lane 2, ammonium sulfate fraction IV; lane 3, run through fraction from dCp4-Sepharose column; lane 4-6, dAp4-Sepharose purified dGuo kinase/dAdo kinase (0.1 fig, 0.2 fig, and 1 fig in lane 4,5, and 6 , respectively). The arrow indicates the position of the dGuo kinase/dAdo kinase. 64 5.2 5.0 4.8 4.4 4.2 4.0 0.0 0.2 0.4 0.6 0.8 1.0 Rf Figure 14. Determination of dGuo kinase/dAdo kinase subunit molecular weight by SDS-polyacrylamide gel. Standard proteins are: a, phosphorylase b, 97,400 Da; b, bovine serum albumin, 66,200 Da; c, ovalbumin, 45,000 Da; d, carbonic anhydrase, 31,000 Da; e, soybean trypsin inhibitor, 21,500 Da; f, lysozyme, 14,400 Da. E indicates the Rf of the purified dAdo kinase/dGuo kinase. 65 Similar to dCyd kinase/dAdo kinase, the two subunits of dGuo kinase/dAdo kinase were resolved by SDS-PAGE under the pH7 MZE 3328.IV buffer system. A doublet of protein bands was observed when stained with Coomassie Blue (Figure 15), indicating the two subunits were nonidentical. As mentioned above, the properties of the column gradually changed with use. In the later preparations, a 55,000 Da protein band was consistently coeluted with the 26,000 Da protein band (see Figure 15). Since this high molecular weight band was about twice the size of the 26,000 Da band, there was a possibility that the 26,000 Da peptide was a proteolytic product of the 55,000 Da protein. This would agree with results of Chakravarty, that the enzyme consisted of a monomeric polypeptide of 56,000 Da (7). All three polypeptides on Figure 15, therefore, were submitted for N-terminal protein sequence analysis (see below). 2. N-Terminal Protein Sequence Analysis N-terminal amino acid sequences of the three polypeptides are shown in Figure 16. The 26,000 Da doublet of protein bands, as well as the 55,000 Da band, have been sequenced separately. Subunits of dGuo kinase/dAdo kinase were separated on a 12% SDS-polyacrylamide gel by the pH 7 MZE 3328.IV buffer system, and then electroblotted onto a Millipore Immobilon PVDF membrane. Portions of membrane containing each polypeptide were excised (about 3.5 Mg. or 125 pmole of each polypeptide) and submitted to The Ohio 66 Da 97,400 66,200 45.000 ( 31.000 Figure 15. SDS-polyacrylamide gel (pH7 MZE 3328.IV buffer system) analysis of purified dGuo kinase/dAdo kinase. (12% gel). Gel stained by Coomassie Blue. Lane 1, protein markers; lane 2, dAp4-Sepharose purified dGuo kinase/dAdo kinase. The arrow indicates the position of the dGuo kinase/dAdo kinase doublet bands. 67 Slow migrating 123456789 10 peptide (S) Met-Ile-Val-Leu-Ser-Gly-Pro-Ile-Gly-Ala- 11 12 13 14 15 16 17 18 19 20 Gly-Lys-Ser-Lys-Gln-Thr- ? - ? -Leu-Ala Fast migrating 1 2 3 4 5 6 7 8 9 10 peptide (F) Thr-Val-Ile-Val-Leu-Ser-Gly-Pro-Ile-Gly Figure 16. N-terminal protein sequence of dGuo kinase and dAdo kinase. 68 State University Biochemical Instrument Center. N-terminal protein sequence analysis was performed on an Applied Biosystems Model 470A Protein/Peptide Sequencer with an on-line 120A PTH Analyzer and a 900A Data Analysis Module. Up to 20 amino acid residues and 10 residues from the N-terminus, for the slow migrating 26,000 Da peptide and fast migrating 26,000 Da peptide, respectively, were sequenced as shown in Figure 16. The two polypeptides differ by the first two amino acid residues, but are identical from the 2 nd residue onward, and contain the conserved sequence of an ATP-binding site (9). The fast migrating peptide lacks a N-terminal methionine, but instead starts with a threonine and a valine. These two peptide sequences also reveal highly conserved homology with the N-terminal sequence of dCyd kinase/dAdo kinase. Since the amount of amino acid detected after the 13th cycle was small, the identities of residues thereafter were not exactly certain. N-terminal amino acid sequence of the 55,000 Da peptide (Met-Phe-Tyr- Pro-Gly-Val) showed that it was actually an unrelated contaminating protein, not a non-proteolyzed form of dGuo kinase/dAdo kinase. Since the native dGuo kinase/dAdo kinase was previously shown to have a molecular weight of 50,000 (4,7), it appears that the enzyme is a heterodimer, composed of two nonidentical subunits of about 26,000 Da, presumably with dGuo kinase on one polypeptide and dAdo kinase on the other. This is analogous to the case of dCyd kinase/dAdo kinase (9). Chakravarty has 69 previously purified dGuo kinase/dAdo kinase complex in very small amounts by rather conventional methods and claimed that the enzyme consisted of a monomeric polypeptide of 56,000 Da (7). Since the amount of pure enzyme obtained in this work is almost 1 0 0 times greater than in the previous case, it is probable that an impurity was mistakenly recognized as the enzyme in the previous work, especially since the silver staining used is notoriously inconsistent in staining intensity with different proteins. However, we cannot exclude the possibility that a form of the enzyme having fused subunits was isolated at that time. Chapter IV DEVELOPMENT OF CLONING PROBES BY POLYMERASE CHAIN REACTION Early attempts at cloning the genes of dCyd kinase, dAdo kinase and dGuo kinase in our laboratory have not been successful due to the lack of a specific probe of any kind for screening. The sequence of the N-terminal 28 amino acids of dCyd kinase/dAdo kinase was recently determined (9), Figure 5. However, this protein sequence contains many highly degenerate codons, so a mixed oligonucleotide longer than a 15mer would consist of 0.3%, or less, of the properly hybridizing oligomer. Moreover, a highly conserved ATP-binding site is present in the sequence. A mixed oligonucleotide cloning probe (16mer, primer + 1 in Figure 17), which was based on the limited sequence and codon- usage information available at that time, failed to specifically locate the genes on a Southern blot. A. CONSTRUCTION OF CLONING PROBE BY PCR The polymerase chain reaction (PCR) is a powerful molecular biology method which allows exponential amplification of DNA segments invitro through 70 71 a succession of incubation steps at different temperatures (86,87) with a thermostable DNA polymerase (73). PCR amplification steps involve (i) heat denaturation of template DNA, (ii) annealing of primers, which flank the target DNA fragment to be amplified, to their complementary target sequences, and (iii) extension of the annealed primer with DNA polymerase. By employing the polymerase chain reaction (PCR), a cloning probe with most of the sequence degeneracy eliminated and therefore with a higher degree of specificity has been constructed here. The degree of specificity of PCR amplification depends greatly on the sequence and amount of primers, the annealing temperature, the magnesium chloride concentration, the amount of template, and the pH of the buffer used in the reaction. Since the copy number of the target sequence is low, and the primers designed based on the amino acid are highly degenerate, all the parameters become critical in order to have the desired product successfully amplified. Various combinations of conditions have been studied and compared: primer concentration was varied from 1-10 /iM; annealing temperatures of 37°C, 42°C, 47°C, 50°C and 52°C were used; magnesium chloride concentration was varied from 0 - 9 mM; template amounts of 1 - 10 ng/jil; pH values of 8.3 and 9.0 were compared. PCR primers were designed based on the N-terminal 26 amino acids of dCyd kinase/dAdo kinase. Sequences of the degenerate primers were shown in Figure 17. The first pair of primers used was primer [+2] and primer [-3]. The 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Met-Ila-Val-Leu-Bar-Qlv-Pro-Ile-Qlv-Ala-Glv-I,va-8er-Bar- > primer [+1] 5'-ATXGGNGCNGGNAA(AG)T-3» (I6mer, degeneracy384) ------> Primer[+2] 5'-ATGATXGTN(TO)TN(TA)(GC)-3' (14mer, degeneracy 384) Xbal------> Primer [+4] 5'-tgctctagATGATXGTN(CT)TN(AT)(GC)NGG-3' (25mer, degeneracy 1536) 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Leu-Thr-Ser-Lau-Leu-Ala-Glu-Tyr-Leu-Gly-Thr-Gln-Ala-Phe Gly lie 8er Lye Aan Pro <------Primer[-3] 3•-AT(GA)(GA)ANCCNTGNGT-5' (14mer, degeneracy 256) <------EcoRl Primer [-5] 3'-TT(CT)AT(AG)(AG)ANCCNTGN(GT)Tcttaaggc-5' (25mer, degeneracy 1024) Figure 17. Design of PCR primers for amplification of dCyd kinase/dAdo kinase specific probes, arrows indicate the codons included in the primers. N = T+C+A+G, X = T+C+A. Restriction enzyme site linkers are written in lower case. The ATP-binding site is underlined. N> 73 sense primer [+2], which comprises the coding sequence of amino acids 1 to 5, is a 14mer with degeneracy of 384. The antisense primer (-3) contains the sequence complementary to the coding sequence of amino acids 25 to 2 1 , and also is a 14mer, with degeneracy of 256. The putative template sequence in the genomic DNA should be 74 base-pairs. This pair of primers, however, failed to produce the desired 74 bp product. Another pair of primers, primer [+4] and primer [-5], with longer lengths were then prepared. The sense primer [+4] comprised the coding sequence of amino acids 1 to 6 plus an Xba I restriction site added to the 5’ end (17+8mer, degeneracy of 1536). The antisense primer [-5] contained the complementary codon sequence of amino acids 26 to 21, with an EcoRI site added to its 5’ end (17+8mer, degeneracy of 1024). Since the N-terminal 28 amino acids represented the mixed sequences of two subunits, Lys at residue position 21 was arbitrarily selected in order to generate a probe specific to one of the subunits. Thus, the target template sequence in the genomic DNA should be 77 base pairs and the amplified PCR product should be 93 bp including the added restriction sites. As will be discussed in a later section, the desired target sequence has been successfully amplified. Even though this second pair of primers has a very high degeneracy, this factor seems to be less critical than the length of the primer. Furthermore, the exogenous restriction site sequences become incorporated into the PCR products. This not only facilitates subsequent molecular cloning of the product, but also serves as an extended template for the annealing of the primers 74 to the PCR products generated after the initial PCR cycles. This allows the use of biphasic annealing steps, that increase the stringency of annealing from 37°C to 50°C after the initial cycles, thus reducing nonspecific products (see below for PCR cycles). When tested with various primer concentrations (1-10 /iM), the amount of product generated was directly proportional to the primer concentration used in the reaction; reactions with less than 2 /iM primers yielded very small amounts of the desired 93 bp product. However, an increase of nonspecific products outgrew the 93 bp fragment when the primer concentration was higher than 5 /iM (probably due to the high degeneracy of the primers). The best magnesium chloride concentration ranged from 1.5 - 2.5 mM, but 2.5 mM magnesium chloride gave more consistent results. The least amount of total genomic DNA required was 0.5 /ig in a 50 /il reaction. pH 8.3 and 9.0 were compared: pH 9.0 produced more overall products, especially the nonspecific contaminants, (probably because the enzyme was more active at this pH); pH 8.3 eliminated many of the nonspecific products, while the amount of the 93 bp fragment decreased slightly. The optimal PCR conditions for amplifying the desired 93 bp product from L. acidophilus genomic DNA have thus been determined. A 50 pi reaction contains 0.5 pg of total genomic DNA, 5 /iM and 3 pM of degenerate primers [+4] and [-5], respectively, 200 /iM each of dNTP (dATP, dCTP, dGTP and dTTP), 2.5 mM MgCl2, 50 mM KC1, 10 mM Tris-HCl (pH 8.4 at room 75 temperature) and 1.25 Units of Taq polymerase. After initially denaturing at 94°C for 2 minutes, the samples were run for 30 cycles using a biphasic annealing routine as follows: denaturation at 94°C for 1 minute, annealing for 1 minute at 37°C for the first five cycles and at 50°C for the subsequent 25 cycles, and extension at 72°C for 30 seconds. At the end of the last cycle, the reaction was extended at 72°C for an additional 7 minutes. B. ANALYSIS OF PCR PRODUCTS The successful amplification of the desired PCR products was confirmed by three different methods: polyacrylamide gel electrophoresis to detect the presence of a product of the expected size, Southern blot and hybridization to an internal probe, and DNA sequencing of the putative fragment. 1 . Polyacrylamide Gel Electrophoresis PCR products were electrophoresed on 5% to 8% polyacrylamide gels and visualized by ethidium bromide staining to detect the presence of product of the expected size. The expected 74 bp product was not detected in any of the PCR amplifications in which the first pair of primers, [+2] and [-3], were used under various reaction conditions. However, when using the longer pair of primers, [+4] and [-5], the expected 93 bp fragment was produced, as shown in Figure 18, together with other nonspecific products. 76 Figure 18. 6 % polyacrylamide gel analysis of polymerase chain reaction (PCR) products. Lane 1, DNA molecular weight marker (pUC19/Msp I); lane 2, PCR products amplified from genomic DNA of L. acidophilus, reaction conditions refer to p. 70 of text. 93 bp fragment is indicated by an arrow. 77 2. Southern Transfer and Hybridization To examine whether the 93 bp product represents the coding sequence of amino acids 1 to 26 of dCyd kinase/dAdo kinase, the PCR products in gel were transferred to nylon membrane and hybridized to an internal probe. The oligonucleotide sequence [+1] in Figure 17, which consists of the coding sequence of amino acids 8 to 12, and does not overlap either of the PCR primer sequences, was used as an internal probe. The result of Southern blotting and hybridization in Figure 19 showed that the 93 bp fragment hybridized strongly to the internal probe, indicating the target DNA sequence which encodes the first 26 amino acids of dCyd kinase/dAdo kinase was indeed amplified. Except for products greater than 500 bp, the other PCR nonspecific products, although exhibiting similar or higher intensity on the ethidium bromide stained gel (Figure 18), did not hybridize to the internal probe. A speculation as to why there is binding of the probe to this >500 bp fragment would be that the genes coding for dCyd kinase and dAdo kinase are located next to each other in the chromosome. Given the high homology in N- terminal amino acid sequences of the two subunits of the enzyme, the primers could have annealed to the N-terminal coding sequence(s) of two different genes (one to dCyd kinase gene, the other to its associated dAdo kinase gene) under reduced stringency conditions (37°C) during the first five cycles of amplification. If this is the case, then one complete gene, the upstream one of the tandemly arranged genes, would have been amplified. This speculation was later shown to 78 1 2 Figure 19. Analysis of PCR products by Southern blot and hybridization analysis. Probe used is 32P-labeled internal probe [+1]. Lane 1, final concentration of primers [+4] and [-5] are 5 /iM and 3 /iM, respectively; lane 2, primers [+4] and [-5] are 8 /iM and 5 /iM, respectively. 79 be correct. The dAdo kinase gene and dGuo kinase gene, indeed, are arranged in tandem on the chromosome, such that the translation stop codon of the dAdo kinase gene is only 2 1 nucleotides upstream of the translation start codon of the dGuo kinase gene (see Chapter V). 3. DNA Sequencing DNA sequence analysis of the 93 bp fragment has been obtained by two different methods. First, the single-stranded PCR product was sequenced directly; second, the fragment was cloned into a plasmid vector and then sequenced. a. asymmetrical PCR Single-stranded DNA was generated by asymmetrical PCR (78,79) as described in Chapter II. Sense primer and antisense primer used in the reaction were at a 1 to 20 molar ratio, or vice versa. The conditions were chosen so that the less-concentrated primer would be essentially depleted after 1 0 to 15 cycles. The subsequent 10 to 15 cycles would thus result in PCR products rich in single stranded amplified DNA. This single-stranded 93mer was used directly as a template for DNA sequencing, employing the less-concentrated PCR primer as the sequencing primer. A total of 31 nucleotides was determined by DNA sequencing of both strands of the 93mer as shown in Figure 20. The derived amino acid sequence matches exactly with the N-terminal amino acids 9 to 18 of dCyd kinase/dAdo Known dCyd kinase/dAdo kinase N-terminal protein sequence: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Het-Ile-Vel*leu-Ser-Gly-Pro*Ile-Gly-Ala-Gly-Ly8-Ser-Ser-Leu-Thr-Ser-Leu-Leu-Ala-Gl.u-Tyr-Leu-Gly-Thf'Gln Gly He Ser lys Asn A. D eterm ined by a sy m n e tric a l PCR: From DNA 5 1- TT GGA GC? GGA ??A TCC AGT CTA ACA GGT AT - 3 ' Derived a.a. Gly-Ala-Gly- ? -Ser-Ser-Leu-Thr-Glv-He B. Determined by cloning into pBluescript: From DNA 5'-ATG ATC GTG CTT T7G GGG CCC ATT GGA GCG GGA AAA TCC AGT CTA ACA GGT ATC TTA TCT AAA TAC CTA GGC ACG CAG A A T A (A) G C A derived a.a. Met-Ile-Val-Leu- 7 -Gly-Pro-Ile-Gty-Ala-Gly-Lys-Ser-ser-Leu-Thr-Glv-lle-teu-Ser-Lvs-Tvr-Leu-Glv-Thr-Gln Figure 20. DNA sequence and derived amino acid sequence of 93 bp PCR product. A) sequence determined by asymmetrical PCR and single-stranded DNA sequencing; B) sequence determined by cloning into pBluescript vector and double-stranded DNA sequencing. Derived amino acid sequences are compared to the known N-terminal protein sequence. § 81 kinase, and is outside the PCR primers region. b. cloning into pBluescript Since the exogenous restriction sites (Xba I and EcoR I) were incorporated into the PCR products, the 93 bp fragment was digested with restriction enzymes, Xba I and EcoRI, and cloned into the plasmid vector pBluescript. Four clones were subjected to double stranded DNA sequencing priming with the universal M13(-40) forward primer and M13(-48) reverse primer. Sequence results obtained from these clones were consistent with those from the above asymmetrical PCR generated single-stranded 93mer, and correlated precisely with the amino acid sequence of dCyd kinase/dAdo kinase (Figure 20). The sequence of this 93 base pair fragment was non-degenerate, however, except within the primer region. Within the primer region, DNA sequence varied from clone to clone, indicating utilization of multiple oligonucleotides for priming the PCR reaction. When the intact genes of dAdo kinase/dGuo kinase were cloned (Chapter V), it became clear that the PCR has specifically amplified the corresponding N-terminal coding sequence of dCyd kinase (or dGuo kinase, whose N-terminal is expected to be the same as the dCyd kinase), which has codons for Gly, lie, Ser and Lys at the split amino acid residue positions 17, 18, 2 0 and 2 1 . 82 C. PREPARATION OF BIOTIN-LABELED PROBE BY PCR 1. Reamplification of 93mer Fragment Analysis of the PCR products showed that the desired 93 bp fragment, which encodes the N-terminal 26 amino acids coding sequence, was successfully amplified, along with other contaminating PCR products. The original 93 bp fragment was thus isolated by agarose gel electrophoresis, eluted, and reamplified by PCR under the same conditions as described above, except for annealing at 50°C throughout all 30 cycles. The reamplified PCR product revealed a single 93 bp band when analyzed by polyacrylamide gel electrophoresis, as shown in ?. 2. Biotinylation A non-radioactive nucleic acid detection method (Gibco BRL’s Photogene Nucleic Acid Detection System), which involves binding of alkaline phosphatase- linked streptavidin to biotinylated probe and subsequent signal detection by substrate-specific chemiluminescence, was used instead of *^P. The DNA probe, the 93 bp fragment, was labeled with biotin by incorporation of biotin-dATP into the PCR reaction, as described previously in Chapter II. As shown in Figure 22 (a), the biotinylated 93 bp PCR product migrated slightly more slowly on agarose gel than its parental 93 base pair fragment, as expected, due to the incorporation of biotin. To confirm the incorporation of biotin molecules, DNA in gel was Southern blotted onto nylon membrane and detected with the Photogene Nucleic Acid Detection System, Figure 22 (b). A strong signal corresponding to the 83 Figure 21. Reamplification of 93 bp polymerase chain reaction (PCR) product. Lane 1, DNA markers (pUC 19/Msp I); lane 2, reamplified 93 bp PCR product, indicated by arrow. Figure 22. Analysis of biotinylated 93 base pairs PCR product. (A) 4% agarose gel: lane 1, DNA marker; lane 2, biotinylated 93bp PCR fragment; lane 3, 93 bp PCR fragment. (B) Southern blot of (A), biotinylated DNA detection by Photogene Nucleic Acid Detection system. Positions of 93 bp fragment are indicated by arrows. 85 biotinylated 93 bp product was detected on the blot, indicating the incorporation of the biotin, while no signal appeared on the adjacent lane containing an equal amount of the parental 93 bp fragment. The biotinylated probe was purified by Dupont NENSORB-20 cartridge or Qiagen tip-5 column to remove primers and unincorporated biotinyl-dATP. Dot blot analysis showed that as little as 0.25 ng of the biotinylated probe could be detected by the chemiluminescence method. CHAPTER V GENE CLONING OF DEOXYNUCLEOSIDE KINASES FROM L. ACIDOPHILUS IN K CO U A. IDENTIFICATION OF DEOXYNUCLEOSIDE KINASES GENES 1. Southern Analysis of Restriction Digested Genomic DNA The biotinylated 93mer PCR product, constructed as described in the last chapter, was used as a probe in the identification of the dCyd kinase, dAdo kinase and/or dGuo kinase gene(s). Southern blots of chromosomal DNA from L. acidophilus digested with various restriction enzymes were hybridized to the biotinylated probe and detected by the BRL Photogene Nucleic Acid Detection System. The results of hybridization are shown in Figure 23. A single EcoR I fragment of 0.7 Kb was detected to hybridize strongly to the probe, indicating the specificity of the probe. However, this fragment was too small to contain the whole gene of both subunits, since the calculated gene size would be about 600 bases per subunit. A 4.5 Kb BamH I restricted fragment which hybridized to the probe might well include the intact genes. A high molecular size (> 10 Kb) BamH I band was probably an incompletely digested fragment. However, an attempt at construction of a BamH I restricted partial genomic library failed, due 86 87 Kb 23.13 9.42 6.56 4.36 2.32 2.03 0.56 Figure 23. Southern hybridization analysis of restriction digested chromosomal DNA from L. acidophilus. Probe used is biotinylated 93mer PCR fragment. Hybridization at 42°C; 0.1 X SSC, 1 % SDS wash at 60°C. Restriction enzymes used are: lane 1 , intact LBA R-26 chromosomal DNA no digestion; lane 2 , BamHI; lane 3, Xbal. 8 8 to difficulty in achieving exhaustive digestion of chromosomal DNA by the rather labile restriction enzyme. Digestion with Xba I, a more stable restriction enzyme, produced a 4 Kb restriction fragment which hybridized strongly to the probe (see Figure 23). Therefore, an Xba I digested partial genomic library was constructed. 2. Xba I Restricted Partial Genomic Library An Xba I-restricted partial genomic library from L. acidophilus was constructed in E. coli XLl-Blue cells. About 100 Mg of CsCl-ethidium bromide purified chromosomal DNA from L. acidophilus was digested with 200 units of Xba I and electrophoresed on 0.8 % agarose gel. Only the marker lanes were stained with ethidium bromide; the gel piece, containing restricted DNA fragments, corresponding in positions to 3.5 Kb to 4.7 Kb, was cut out. The DNA was electroeluted, purified on a Dupont Nensorb-20 cartridge, and ligated into the Xba I site of dephosphorylated pBluescript cloning vector. After transformation of XLl-Blue cells, 800 recombinants (white colonies), out of a total 1000 transformants, were obtained from 0.1 Mg of DNA. Colony hybridization, with the biotinylated 93mer probe, revealed two positive clones. Both clones contained a 3.4 Kb Xba I insert which hybridized to the probe. Restriction analysis of the two clones revealed an identical restriction map, as shown in Figure 24. Southern analysis demonstrated that a 0.7 Kb EcoR I fragment from near one end of the insert hybridized strongly with the probe. Furthermore, a 0.8 Kb Apal Apal Xbal Xbal EcoR Hindlll Kpnl EcoRI Eco I 1 1 I il v - dAK dCK 1 Kb scale (partial) Figure 24. Partial restriction map of Xba I clone. Xba I restricted genomic DNA of L. acidophilus containing dAdo kinase gene and partial dGuo (or dCyd) kinase gene was inserted into pBluescript vector. Vector portion is not shown. oo VO 90 EcoRI/Kpnl double restriction fragment mapping next to the 0.7 Kb EcoRI fragment also hybridized, but weakly, to the probe. Both fragments were subcloned into pBluescript and sequenced. A start codon and a derived amino acid sequence which agreed with the N-terminal 28 amino acid sequence of dCyd kinase/dAdo kinase, was found in the 0.7 Kb EcoRI fragment, but with only one each of the paired residues found earlier at positions 18,21,26 and 27 (Figure 5). At this point, based on the available N-terminal peptide sequences, we were unable to deduce whether the clone represents the dCyd kinase or the dGuo kinase genes. Further sequencing of this subclone and the parental clone revealed that the C-terminal portion of the second gene (about 1/3 of the gene) was missing from the parental Xba I clone. Just 21 nucleotides upstream of the partial dCyd kinase (or dGuo kinase) gene, an intact open reading frame of 215 codons was found. Its N-terminal sequence also matched exactly the known sequence of residues 2 - 28 of dCyd kinase/dAdo kinase, but with the other amino acids of the paired residues seen at position 18,21, 26 and 27 (Figure 5). This was presumed to correspond to the weaker signal of the 0.8 Kb EcoRI/Kpnl fragment seen on the Southern blot. This upstream gene is terminated by a single translational stop codon, TAG. The entire 3.4 Kb insert of the Xba I clone has been sequenced; no other open reading frame with an homologous region was found in the 2.0 Kb region upstream of the dAdo kinase gene. 91 3. Kpn I Restricted Partial Genomic Library Since the 3’ end of one of the dCyd kinase/dAdo kinase genes was missing, a new library was constructed. A more specific probe was prepared by PCR. A new sense primer [+6] (Table 5), a 20mer, which contained the intervening sequence between the genes of dCyd kinase and dAdo kinase, was used, along with the original antisense primer [-5]. This new probe was therefore a 117mer, with an exogenous 8-nucleotide EcoR 1 linker at its 3’ end. Southern analysis was performed on restriction digested genomic DNA (Figure 25). Knowing that the 5’ to y orientation of the genes was from Kpnl to Xbal, the 2.5 Kb Kpnl digested genomic fragment hybridizing to the probe should contain the intact dCyd kinase/dAdo kinase genes. Therefore, a partial genomic library was constructed in pBluescript from the 2 to 4 Kb Kpnl fragments, and transformed into the E. coli XLl-Blue cells. After initial colony hybridization screening of more than a thousand recombinants with the new probe, numerous positive signals showed up (most are false positives). A secondary PCR screening of 10 to 20 colonies at a time was carried out, as described previously in Chapter II. 10 - 20 individual colonies were scraped directly from a replica agar plate and used as templates for a reaction. Primer [+6] and primer [-5] were used, and thus a PCR product of 117 base pairs was expected. About 200 colonies were screened in 10 reactions. The expected 117 bp amplified fragment was detected in one of the reactions by 8% acrylamide gel electrophoresis and visualized by ethidium bromide staining. Table 5. List of primers used in PCR and sequencing Primer Nucleotide Sequence Length Nucleotide Degeneracy Name position [+ 1 ] 5’-ATXGGNGCNGGNAA(AG)T-3’ 16 mer 2 8 -4 3 384 [+ 2 ] 5’-ATGATXGTN(TC)TN(TA)(GC)-3’ 14 mer 1-3,10-20; 384 679 - 689 [-3] 5’-TGNGTNCCNA(GA)(GA)TA-3’ 14 mer 752 - 739 256 [+4] 5’-tgctctagATGATXGTN(CT)TN(AT)(GC)NGG-3’ 25 mer 10 - 23 ; 1536 679 - 692 [-5] 5’-cggaattcT(TG)NGTNCCNA(GA)(GA)TA(TC)TT-3’ 25 mer 752 - 736 1024 [+ 6 ] 5’-ACTAGTTAACGAATAGAAGG-3’ 2 0 mer 644-613 - [+7] 5’-ACTCTAGAAGATTACTACAA-3’ 2 0 mer 1135-1154 - [+ 8 ] 5’-TACTTGAGCGAATTAATCTT-3’ 2 0 mer -167—148 - [-9] 5’-AACCATATCCTGTTACTAAG-3’ 2 0 mer 1474-1455 - [-1 0 ] 5’-GATCGCGAATAGCTTGGTTA-3’ 2 0 mer 217 - 198 In the cases where degenerate oligonucleotides are used, N = T+C+A+G, X = T+C+A, and nucleotides in paranthesis as shown. Naming of primers: [+] indicates sequence corresponding to coding sequence, [-] indicates sequence complementaiy to coding sequence. Nucleotide positions refer to those in Figure 27. 8 93 Figure 25. Southern and hybridization analysis of restriction digested genomic DNA from L. acidophilus. Probe = biotinylated 117mer PCR fragment. Hybridization at 42°C; 0.1 X SSC, 1 % SDS wash at 60°C. Restriction enzymes used are: lane 1, Kpn I; lane 2, Sac I; lane 3, DNA markers (A/Hind HI); lane 4, Xba I; lane 5, Eco RI; lane 6 , DNA markers (A./Hind HI/EcoR I); lane 7, Hind HI; lane 8 , Pst I. 94 Then, each colony in the positive set of 20 was screened individually by PCR. A positive clone was identified, and its partial restriction map is shown in Figure 26. Judging from the map, the 2.5 Kb insert contained an intact dAdo kinase gene as well as a dCyd kinase (or dGuo kinase) gene. B. NUCLEOTIDE SEQUENCE ANALYSIS 1. Open Reading Frames The nucleotide sequence of the entire 2.5 Kb Kpnl insert was determined. Sequencing strategies involved subcloning and "gene-walking" as indicated by arrows in Figure 26. The nucleotide sequence and derived amino acid sequence are shown in Figure 27. Examination of this sequence revealed two open reading frames (ORFs). The first ORF is 648 bp in length, the 215 codons encode a peptide with a predicted size of 25 KDa; the next 675 bp ORF encodes a peptide of 224 amino acid residues with a predicted molecular weight of 26 KDa. The N-terminal sequences of both genes match exactly with the known amino acid sequence of residues 2 to 28 of dCyd kinase/dAdo kinase, and diverge at the mixed amino acid residues at positions 18, 21, 26 and 27. The first gene is terminated by a single translation stop codon (TAG), and is separated from the second gene by a 2 1 bp spacer. Ndel „ , Apal Hpal Apal EcoRI _ MiP Kpnl EcoRI I Sacl I Xbal f Kpnl * I H »i * 4 * II I dAK dGK 0.5 Kb scale Figure 26. Partial restriction map and sequencing strategy of Kpn I clone. Kpn I restricted genomic DNA of L. acidophilus containing the intact dAdo kinase gene and dGuo kinase gene is inserted into pBluescript. M13 forward and reverse primers, as well as sequence specific primers have been used to prime sequencing reactions. Sequencing strategies are indicated by dotted arrows (5’-*3’). Regions of insert containing dAdo kinase gene (dAK) and dGuo kinase gene (dGK) is marked in 5’ -► 3’ direction. 96 ggtaccagcat -271 ctatcttacaagatcacccagatgtaacagttatttgtgatgaagttgctgccgcaaaacttgatccaaaatatagaaactaatcctcat -181 taattagaccgtttacttgagcgaattaatcttttttagattagttcgtttttcttttttgcgatttttttgcgtgttcgcttgactttt -91 A duster *35 -10 S/D ttcttttatagcaaattaaaaaaccacttg ttttcataccagatattgaattacactattaaagcaaactatatatQtagaaBgaacgtg -1 atgacagttattgtattgagcgggcccattggagccggaaaatccagtttaaccagtcttcttgccgaacatttaggcactcaagccttt 90 MTVIVLSGPIGAGKSSLTSLLAEHLGTQAF tatgagggtgtagataacaatccaattttgccactttattataaagatatggctcattatacttttcttttaaatacttatcttctaaac 180 YEGVDNNPILPLYYKDHAHYTFLLNTYLLN caccgtttagctcaaattaaccaagctattcgcgatcataacagtgtgtctgatcgttcaatttatgaagatgccctatttttcaaaatg 270 HRLAQINQAIRDHNSVSDRSIYEDALFFKM -39 -10 -35 -10 aatottgatagtggtattgctgatcctacagaattcaagatttatQ ataQtttacttqagaatatgatggaacaaocacctootaatcca 360 NVDSGIADPTEFKIYDSLLENMNEOAPGNP -39 -10 agtaagaaaccagatcttttaatttatattcatgtttctcttgatactatgcttcaccgaattcaaaaacgtggtcgtaagtttgaacaa 450 SKKPDL L I Y I HVSLDTMLHR I QKRGRKFEO -35 -10 ttatcaaccgatccaagtttaaaaoattactatgctcgccttttatcatattacoagccttggtacgaaaagtataatgcatcccctaag 540 LSTDPSLKDYYARLLSYYEPUYEKYNASPK atgatgattgatggtgataaatacgactttgttgccaatgaagacgcaagaaggaaagttattaacgcaattgatcaaaaacttattgat 630 MMIDGDKYDFVANEDARRKVINAIDQKLID S/D atagggaatttaaactagttaacgaatagaaggaacgtgatgacegttattgtattaagcgggcccattggagccggaaaatccagtcta 720 IGNLN. HTVIVLSGPIGAGKSSL acaggtatcttatctaaatatttgggtactaatcccttttatgaaagtgtagatgacaatcctgttttgccattattctatgaaaaccct 810 TGILSKYLGTNPFYESVDDNPVLPLFYEHP aaaaagtatgcctttttactgcaagtttatttcttaaatactcgttttcggagtattaagtcagccttaactgatgataataatgtactt 900 KKYAFLLQVYFLNTRFRSIKSALTDDNNVL gaccgttctatctacgaagatgctcttttcttccaaatgaatgcagatattgggcgtgctactccagaagaagtcgatacttactatgag 990 DRSIYEDALFFOMNAOIGRATPEEVDTYYE ctcttgcacaatatgatgagtgaactagatcggatgcctaagaagaatcctgatctcctggttcatatcgatgtctcatatgatacaatg 1080 LLHNHMSELDRMPKKNPDLLVH IDVSYDTM ctcaagagaattcaaaaacgtggtcgtaactatgaacaattgagttatgatccgactctagaagattactacaagcgtctacttcgttat 1170 LKRIQKRGRNYEQLSYDPTLEDYYKRLLRY tacaaaccttggtatgcgaagtatgactattcaccaaaaatgactattgatggtgataaacttgatttcatggcaagtgaagaagatcgt 1260 YKPUYAKYDYSPKMTIOGDKLDFMASEEDR caagaagtcctaaatcaaattgtggctaagctcaaagaaatgggtaaacttgaagacgactggaaacctaatttagttaaataaagcaaa 1350 OEVLNQIVAKLKEMGKLEDDUKPNLVK. actgcggtccaatatggatcgcagtttttattttttaatcctaacaatcaatgtcttgatgcaccacttacagcatcagctgcttcacct 1440 gttaactgatttt 1453 Figure 27. Sequence of the dAdo kinase and the dGuo kinase genes from L. acidophilus. Lower case letters indicate the nucleotide sequence; upper case letters indicate amino acid sequence. Promoters (-35, -10, etc.) and ribosome binding sites (S/D) are underlined and indicated as marked. The transcription terminator is marked by dotted arrows. 97 2. Comparison of DNA Sequences of Cloned Genes and Known N-terminal Peptide Sequences Dr. Ikeda in our lab has recently resolved the two subunits of dCyd kinase/dAdo kinase using a pH 7 MZE discontinuous buffer system SDS-PAGE (similar to the separation of dGuo kinase/dAdo kinase as described in Chapter III). By employing affinity labeling with substrate-binding site specific probes, he identified the fast migrating peptide as dCyd kinase and the slow migrating peptide as dAdo kinase ( 8 8 ). N-terminal protein sequences of the subunits have been determined individually. The derived amino acid sequences of the two cloned genes were compared with the known N-terminal amino acid sequences of dCyd kinase/dAdo kinase and dGuo kinase/dAdo kinase. The sequence of the upstream gene agrees exactly with the amino acid sequence of dAdo kinase from residues 2 to 28, but contains two extra codons (threonine and valine) between the initial methionine and the isoleucine (Figure 28). The sequence of the downstream gene is identical to that of dCyd kinase from residues 2 to 28, also with the exception of two additional codons, threonine and valine, after the initial methionine. It is also identical to that of the dGuo kinase except the initial methionine is missing from the peptide. These two genes were later expressed in E. coli (as described in Chapter VI) and were identified as the genes encoding dAdo kinase and dGuo kinase, based on enzyme activities. (Protein) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 dAK o f dCK/dAK Met-Ile-val*Leu-Ser-Gly-Pro-Ile-GlyAla-Gly-Ly8*Ser*Ser-l.eiJ-Thr-Ser-Leu-Leu-Ala-Glu- ? -Leu-Gly-Thr-Gln 1 23456789 10 11 12 13 dAK of dGK/dAK Het-Ile-Val-Leu-Ser-Gly-Pro-Ile-Gly-Ala-Gly-lys-Ser (DNA) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 2B 1st gene Met-ThrM/ai.-lle-Val-Leu-Ser-Gly-Pro-l te-Gly-Ala-Glylys-Ser-Ser-Leu-Thr-Ser-Leu-Leu-Ala-Glu-His-Leu-Gly-Thr-Gln (Protein) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 dOC of dCK/dAK Met-Ile*Val-Leu-Ser-Gly-Pro-ne-Gly-Ala*Gly'Ly8-Ser-Ser-Lau-Thr-Gly-Ile-Leu-Ser-Ly*-Tyr-Leu-Gly-Thr*A8n 123456789 10 dGK of dGK/dAK Thr-Val-Ue-Val-Leu-Ser-Gly-Pro-Ile-Gly-Ala-Gly (DNA) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 2nd gene Het-Thp^Val-Ile-Val-leu-Ser-Gly-Pro-Ile-Gly-Ala-Gly-Lys-Ser-Ser-leu-Thr-Gly-Ile-Leu-Ser-lys-Tyr-Leu-Gly-Thr-Aan Figure 28. Comparison of N-terminal amino acid sequence obtained by peptide and DNA sequencing. 99 L. acidophilus R-26 seems to possess some sort of processing mechanism to remove the additional amino acid residues. Could this be a posttranslational modification? Could this be an activation mechanism to produce the active forms of the enzymes? Could splicing occur at the mRNA level? Or could the codon GUU, which normally codes for valine, code for the initial methionine in these genes? The reason for the discrepancies between the coding sequence in the genes and the amino acid sequences of the active enzymes in L. acidophilus would be interesting to study. 3. Promoter, Ribosome Binding Site and Terminator Weak "-10" and "-35" promoter elements precede the dAdo kinase gene, with the sequences TACACT beginning at nucleotide -39 and TTGTTT beginning at nucleotide -63, respectively, as marked in Figure 27. Furthermore, additional promoter elements conserved among gram positive bacteria are also found (89): The first is the "-45" A cluster (AAAAA) beginning at nucleotide -72; the second is the sequence T at nucleotide -75; finally is the sequence TG between "-10" and "-35" region at nucleotides -44 and -43. The sequence GAAAGA, 6 bases upstream from the translation start codon of the dAdo kinase gene, is likely to be the ribosome-binding site. Preceding the second gene (the dGuo kinase or dCyd kinase gene), sequences homologous to the "-10" and "-35" promoter elements are revealed in four different regions. These occur at 141,259,331 and 355 bases upstream from 100 the initiation codon of the second gene and overlap the coding region of the first gene. The sequence of these possible promoter elements seems to be more homologous to those of theE. coli than the sequences preceding the dAdo kinase gene. However, none of the additional conserved promoter elements of gram positive bacteria are present before the second gene. Therefore, these putative promoters preceding the second gene are probably not functional in L. acidophilus. A putative ribosome-binding site is 6 bases upstream from the initiation codon, with the sequence GAAGGA. Just 4 bases downstream from the translational termination codon (TAA) of the second gene, a sequence with characteristics of transcription terminators (90) is revealed. A GC-rich region of dyad symmetry, AAAACTGCGGTCCA and TGGATCGCAGTTCT (as marked by arrows in Figure 27), which may form a stem and loop hairpin structure, is followed by a stretch of uridines in the transcript (TTTTTT at positions 1531 - 1536, marked by underline). This sequence may function as a transcription terminator. No such characteristic sequence is observed close to the stop codon of dAdo kinase gene. Absence of a transcription terminator in the dAdo kinase gene (and probable absence of a promoter before the second gene) may suggest that a polycistronic message is encoded in the transcription step. However, whether any of the potential promoter sequences and the terminator sequence would function, or not, could only be determined by further experimental studies. 101 4. Calculated pi Values Based on the inferred amino acid sequences, the calculated pi values for dAdo kinase and dGuo kinase (or dCyd kinase) are 5.83 and 5.13, respectively. The calculated pi for the dimer (5.43) agrees exactly with the experimental pis of the wild type dAdo kinase/dGuo kinase and dAdo kinase/dCyd kinase (both were 5.4), determined previously by Deibel (4). 5. Sequence Comparison of dAdo Kinase and dGuo Kinase The two genes cloned from L. acidophilus are highly homologous in their sequences, both at the gene level and at the amino acid level. Comparison of the two genes, Figure 29, reflects 65% overall homology in DNA sequence. Furthermore, alignment of derived amino acid sequences from cloned dAdo kinase and dGuo kinase (or dCyd kinase), Figure 30, reveals 60.9% identity between the two peptides. There are three regions in the two peptide sequences that seem to be highly conserved among the two genes, as marked on Figure 30: (1) ATP-binding site: The N-terminus of each of the peptides contains the known conserved sequence of an ATP-binding site (9). This conserved region has made this cloning project a lot more challenging since oligonucleotide cloning probes which span this region would bind to genes of many ATP-binding proteins. (2) The region with the sequence Asp-Arg-Ser is about midway between the region ( 1 ) and region (3). Its function is not clear, but dAK ATGACAGTTATTGTATTGAGCGGGCCCATTGGAGCCGGAAAATCCAGTTTAACCAGTCTTCTTGCCGAACATTTAGGCACTCAAGCCTTTTATGAGGGTG 1 I I It I I I I I II I I I I I I I I I II I I I I I I I I 1 I I I I II I I I I I I It 1 Mil II I I I II Mil II III I 1 I I I I I I I I I III 111 ii 111111111111 111111111111111111111111111111 nil ii i i i ii m i ii in i i ii ii 111 ii ill dGK ATGACAGTTATTGTATTAAGCGGGCCCATTGGAGCCGGAAAATCCAGTCTAACAGGTATCTTATCTAAATATTTGGGTACTAATCCCTTTTATGAAAGTG *10 *20 *30 *40 *50 *60 *70 *80 *90 *100 dAIC TAGATAACAATCCAATTTTGCCACTTTATTATAAAGATATGGCTCATTATACTTTTCTTTTAAATACTTATCTTCTAAACCACCGTTTAGCTCAAATTAA him 1111111 11111111 i i hi n i i mi i in i i i mi i nil inn inn Mill IIIMM Mllllll I I III II I I III I III I I I Mil I Mil Mill lllll dGK TAGATGACAATCCTGTTTTGCCATTATTCTATGAAAACCCTAAAAAGTATGCCTTTTTACTGCAAGTTTATTTCTTAAATACTCGTTTTCGGAGTATTAA *110 *120 *130 *140 *150 *160 *170 *180 *190 *200 dAK CCAAGCTATTCGCGATCATAACAGTGTGTCTGATCGTTCAATTTATGAAGATGCCCTATTTTTCAAAATGAATGTTGATAGTGGTATTGCTGATCCTACA in i in nil i in in i n n ii i i n i n n i it n in m m m nil in tin h i i in i in nil i in in m im i i i i n i n n i n i i in iiiiiini m i in nil in i dGK GTCAGCCTTAACTGATGATAATAATGTACTTGACCGTTCTATCTACGAAGATGCTCTTTTCTTCCAAATGAATGCAGATATTGGGCGTGCTACTCCAGAA *210 *220 *230 *240 *250 *260 *270 *280 *290 *300 dAK GAATTCAAGATTTATGATAGTTTACTTGAGAATATGATGGAACAAGCACCTGGTAATCCAAGTAAGAAACCAGATCTTTTAATTTATATTCATGTTTCTC in ii i i in n i l l M i n i m n i i i i n I inn n m u i n nil nil ii in ii i i in n i i i Miiiiiii n i i i i i i i i n n i i i n n i i i nil ini ii dGK GAAGTCGATACTTACTATGAGCTCTTGCACAATATGATGAGTGAACTAGATCGGATGCCTAAGAAGAATCCTGATCTCCTGGTTCATATCGATGTCTCAT *310 *320 *330 *340 *350 *360 *370 *380 *390 *400 dAK TTGATACTATGCTTCACCGAATTCAAAAACGTGGTCGTAAGTTTGAACAATTATCAACCGATCCAAGTTTAAAAGATTACTATGCTCGCCTTTTATCATA n n n m i l i 1111111111111111111111 i iiiiimi inn i i ii mimimii n n i n n n n m i l i inniii 11ii111iiiiin i miiiiiii inn i i ii m i n i m n n i n dGK ATGATACAATGCTCAAGAGAATTCAAAAACGTGGTCGTAACTATGAACAATTGAGTTATGATCCGACTCTAGAAGATTACTACAAGCGTCTACTTCGTTA *410 *420 *430 *440 *450 *460 *470 *480 *490 *500 dAK TTACGAGCCTTGGTACGAAAAGTATAATGCATCCCCTAAGATGATGATTGATGGTGATAAATACGACTTTGTTGCCAATGAAGACGCAAGAAGGAAAGTT nil i iiiiimi i n n n i n n n nil i iii 11111111111 n n i n i n n n i i nil nil i iiiiimi i nnn i n n n nil nnnnnnni n n i n i n n n i i nil dGK TTACAAACCTTGGTATGCGAAGTATGACTATTCACCAAAAATGACTATTGATGGTGATAAACTTGATTTCATGGCAAGTGAAGAAGATCGTCAAGAAGTC *510 *520 *530 *540 *550 *560 *570 *580 *590 *600 dAK ATT AACGCAAT TGAT CAAAAACT T AH GAT AT AGGGAAT TT AAACT AG i l l 11111 lllll 11 11 ii 11 I I I ill lllll lllll II 11 II 11 I I I dGK CTAAATCAAATTGTGGCTAAGCTCAAAGAAATGGGTAAACTTGAAGACGACTGGAAACCTAATTTAGTTAAATAA *610 *620 *630 *640 *650 *660 *670 Figure 29. Comparison of nucleotide sequence of dAdo kinase (dAK) gene and dGuo kinase (dGK) gene from L. acidophilus. " |" indicates matching nucleotides. 8 (Site i) 10V 20V 30v 40v 50v 60v dAK MTVIVLSGPIGAGKSSLTSLIiAEHLGTQAFYEGVDNNPILPLYYKDMAHYTFLLNTYLLN MTVIVLSGPIGAGKSSLT::L:. . LGT::FYE:VD:NP:LPL:Y.: .Y:FLL:.Y:LN dGK MTVIVLSGPIGAGK8SLTGILSKYLGTNPFYESVDDNPVLPLFYENPKKYAFLLQVYFLN 10A 20A 30A 40A 50A 60A (Site ii) 70v 80v 90v lOOv llOv 12Ov dAK HRLAQINQAIRDHNSVSDRSIYEDALFFKMNVDSGIADPTEFKIYDSLLENMMEQAPGNP R: I: A: D:N:V DRSIYEDALFF:MN.D G A.P.E ..Y .LL:NMM.: P dGK TRFRSIKSALTDDNNVLDR8IYEDALFFQMNADIGRATPEEVDTYYELLHNMMSELDRMP 70A 80A 90A 100A 110A 120A (Site iii) 130V 140v 150v 160v 17Ov 180v dAK SKKPDLLIYIHVSLDTMLHRIQKRGRKFEQLSTDPSLKDYYARLLSYYEPWYEKYNASPK .K:PDLL:.I:VS DTHL.RIQKRGR::EQLS DPiL.DYY RLL.YY.PWY.KY: SPK dGK KKNPDLLVHIDVSYDTMLKRIQKRGRNYEQLSYDPTLEDYYKRLLRYYKPWYAKYDYSPK 130A 140A 150A 160A 170A 180A 19Ov 200v 210v dAK MHIDGDKYDFVANEDARRKVINAIDQKLIDIGNLNX M IDGDK DF:A:E:.R:.V:N.I .KL ::G:L:. dGK MTIDGDKLDFHASEEDRQEVLNQIVAKLKEMGKLEDDWKPNLVK 190A 200A 210A 220A Figure 30. Comparison of derived amino acid sequence of dAdo kinase (dAK) and dGuo kinase (dGk). 60.9% identity in 215 aa overlap (DNAStar AANW). Consensus regions, site (i), (ii) and (iii) (described in text) are indicated in bold. 104 by analogy with viral thymidine kinase (52), might be a nucleoside binding site (3) Arginine rich region: An arginine region located at about 2/3 of the way along the length of the peptides consists of the conserved sequence associated with substrate phosphorylation sites. These sites will be discussed in more detail and compared with related kinases in the next section. 6. Sequence Comparison with Related Kinases Beside the dAdo kinase/dGuo kinase from L. acidophilus, so far, no other dAdo kinase or dGuo kinase sequence information is available from any source that can be used for comparison. However, a rich lode of sequence data is available from viral, bacterial and vertebrate thymidine kinase. Recently, dCyd kinase from human T-lymphoblast MOLT-4 cells has also been cloned and sequenced (36). These deoxynucleoside kinases are expected to be closely related and therefore will be used to compare with the derived amino acid sequences of the newly-cloned dAdo kinase and dGuo kinase. Balasubramaniam et aL (52) has compared the amino acid sequences of twelve herpesviral thymidine kinases and has identified six highly conserved segments. Three of these segments are also conserved in the L. acidophilus dAdo kinase and dGuo kinase, as shown in Figure 31. Bit* i Sit* ii Bit* iii la dAK GPIGAGKS DRS RIQKRGR la dGK GPIGAGKS DRS RIQKRGR h dCK GNIAAGKS DRY RIYLRGR hsl TK GPHGMGKT DRH RLAKRQR hs2 TK GPHVFGKT DRH RLARRQR mhv TK GPHGVGKS DRH RLRARAR vzv TK GAYGIGKT DRH RVSKRAR fhv TK GAYGIGKS DRH RLRGRSR prv TK GAYGTGKS DRH RLRARAR ehl TK GVYGIGKS DRH RLRTRAR bhl TK GAHGLGKT DRH RLAARAR thv TK GPFGIGKT DRH RICSRDR mdv TK GSMGIGKT DRH RLSSRNR hvs TK GSIGVGKT DRH RVKKRNR ebv TK GAPGVGKT DRH RVKKRNR consensus G**G*GKS/T DR* R***R*R Figure 31. Sequence comparison of dAdo kinase and dGuo kinase from L. acidophilus with Herpeviral thymidine kinases. 106 Site (i): This segment is located near the N-terminus, amino acid residues 8 * 15, of the dAdo kinase and the dGuo kinase and contains the consensus sequence (GXXGXGKT/S) which has been found in most ATP binding enzymes. This site was first described by Pai et aL (91) as part of a substrate binding pocket in porcine adenylate kinase (PAK) and was initially identified in viral thymidine kinases by Otsuka and Kit (49) and by Gentry (92). Crystal structure analysis of guanine nucleotide binding protein EF-Tu and p21, as well as the adenine nucleotide binding protein adenylate kinase showed that this segment constitutes a loop that wraps around a phosphate group (93,94,95,96,97,98). This phosphate group is identified to be the fJ- phosphate of GDP and GppNp in p21 complexes (97,98) and phosphate 2 of the adenylate kinase-APgA complex (93,99). Based on the NMR and x-ray structure of PAK by Fry et aL (100), this glycine-rich flexible loop undergoes a conformational change upon ATP binding, and, together with the hydrophobic residues following the consensus sequence, GXXGXGKS/T, they form a pocket that binds the adenine moiety of ATP. Fry et aL suggested three functions of the glycine-rich loop in PAK: (a) control of accessibility to the substrate binding site. Since the loop is located in the middle of a cleft, its change in conformation, as evidenced in two crystal forms, demonstrates its capability of alternately blocking and opening the cleft, (b) modification of binding site affinity. The conserved lysine is located in PAK to interact with the a-phosphoryl group of bound MgATP. Thus, a different 107 orientation of this lysine induced by a conformational change of the loop could alter binding affinity of MgATP and possibly stabilize the binding of the product MgADP at this site, since the negative charge distribution among the phosphoryl groups is different for the two substrates, (c) relocation of catalytic groups to the reaction center. By a conformational change, the serine hydroxyl residue of the loop approaches the triphosphate chain of bound MgATP in PAK and could facilitate the phosphoryl transfer reaction, possibly by functioning as a hydrogen bond donor. The three conserved glycines are important components of the flexible loop. Liu and Summer (55) showed that site-directed mutagenesis of HSV-1 thymidine kinase in any one of the three glycines to valine abolished enzyme activities. The small sized glycine is thought to have a secondary structure breaking activity that is essential for the TK to function. How about the lysine in the consensus glycine-rich loop? It was inferred from crystal structure that the lysine residue in the glycine-rich loop of E. coli adenylate kinase (EcAK), which connects a /9-sheet and an a-helix, points toward the phosphate groups (93,99,98). Affinity labeling of chicken AK by substrate analogues like adenosine diphosphopyridoxal totally inactivated the enzyme, and the lysine in the glycine-rich loop was found to be chemically modified ( 1 0 1 ). However, this inactivation was prevented by substrates ATP or ADP, and to a small extent, by AMP. Site-directed mutagenesis of this lysine residue of HSV-1 TK to valine resulted in loss of enzyme activity (55). Similar inactivation results 108 were observed in the corresponding lysine mutants of F^-ATPase (102), Rho protein (103), and EcAK (104). The latter is of some interest, because the lysine to glutamine mutant was shown to drastically affect the transition state rather than substrate binding, so the lysine seems to be important in stabilizing the transition state (104). When the conserved threonine in the loop was changed to alanine, enzyme activity was lost. However, with a serine in place of the threonine, the enzyme retained activity, as would be expected since many nucleotide enzymes contain a serine at the corresponding position, including the dAdo kinase and dGuo kinase from L. acidophilus. Site (ii): This short segment, from amino acid residues 78 - 80, contains the sequence Asp-Arg-Ser. A similar sequence Asp-Arg-His is a highly conserved site in the herpeviral thymidine kinases (52) and is thought to be involved in thymine recognition (105,106,107). This sequence, except that the histidine is replaced by a tyrosine, is also observed in thymidylate synthetases of vaccinia virus (108) and of yeast (109), as well as in the deoxycytidine kinase of human T-lymphoblast MOLT-4 cells (36). Furthermore, several bacterial phosphotransferases using ATP as phosphate donor also contain the sequence (110). The aspartate of this motif may interact with the phosphate groups of ATP through a Mg^+ salt bridge (52). However, this consensus site is not found in any of the vertebrate (human, mouse, chicken), E. coli or large DNA viral thymidine kinases. 109 The function of this site in the dAdo kinase/dGuo kinase from L. acidophilus are being studied in our laboratory by site-directed mutagenesis. Site (iii): This segment is arginine-rich with the consensus sequence RXXXRXR at amino acid residue positions 140 - 146 of LBA dAdo kinase and dGuo kinase. This arginine-rich site is also highly conserved in the herpeviral thymidine kinases and has been recognized in Herpeviral TKs by its similarity to an arginine-rich site in PAK (92,106,52). In chicken muscle adenylate kinase (cAK), the third arginine is further to the right (Arg-138) when aligned to the herpeviral TKs (52). This arginine in cAK is shown to be essential for activity by site-directed mutagenesis and NMR studies by Yan et aL (111). Replacement of Arg-138 by methionine, or even lysine, which did not cause structural perturbation compared with wild type enzyme, resulted in inactivation of the enzyme. Their kinetic and structural results demonstrated that the Arg-138 stabilizes the transition state, and to a small extent, the ground state ternary complex, thus suggesting this Arg is a strong candidate to interact with the transferring phosphoryl group. Dreusike et aL (112) also suggested the arginines in this region of PAK bind substrate phosphoryl groups based on refined crystal structure. The arginines in this site of LBA dAdo kinase/dGuo kinase may have a similar function. In addition to the three arginines in the consensus sequence, a glycine and a glutamate, which are two residues to the right from the third consensus arginine, are also found to be highly conserved in herpeviral TKs (52). The 110 glutamate, but not the glycine, is also reflected in the sequence of human dCyd kinase (36), as well as the dAdo kinase and dGuo kinase of LBA. This acidic residue may also be important for some catalytic function of this site (52). The three arginines in this site are completely conserved between the LBA dAdo kinase and dGuo kinase, and herpeviral thymidine kinases. Within this arginine-rich site, the sequence of LBA dAdo kinase and dGuo kinase seem to be more homologous to the Epstein-Barr virus TK than to other herpeviral TKs. 7. Codon Usage of LactobadUus Years ago when I started the gene cloning project, only four genes from Lactobacillus were cloned and sequenced. Therefore, the codon usage information provided was insufficient to reduce the degeneracy in cloning probe design. At the present time, sequences of more than fifty genes from the genus Lactobacillus have been reported; thus it is possible to deduce codon preferences in Lactobacillus. Several sequence databases were searched, including GenBank and EMBL. Records were retrieved from databases at the National Center of Biotechnology Information (NCBI) in the National Library of Medicine of NIH. The accession numbers (Genbank) and references of the genes retrieved are listed in Table 6 . Codon usage of each Lactobacillus gene was generated using the computer software DNAStar, based on the sequences covered. Average codon usages of each species and all Lactobacillus species were calculated and summarized in Ill Table 6 . List of Lactobacillus genes used in the codon usage tables (Table 7 to 9). (* gene from plasmid.) ACCESSION Lactobeci1lus REFI J02613 30a 113 X13099 30a 114 M84564 b r e v is 115 H94058 S39818 bulgaricus 116 D00496 c a s e i 117 D01251 D125911 c a s e i 118 J05221 c a r e i 119 M10922 casei 120 N19653 c a s e i 121 M20150 c a s e i 122 M20151 c a s e i 123 M26929 c a s e i 124 N37466 M76708 c a s e i 125 H60851 c a s e i 126 H83993 S39433 c a s e i 127 S77578 c a s e i 128 X02734 c a s e i 129 X64542 c a s e i 130 M28050 M31425 c o n fu su s 131 D10020 delbrueckii 132 N23530 d e lb ru e c k i i 133 M55068 H38754 delbrueckii 134 M85224 delbruecki i 135 X60220 delbrueckii 136 X61190 delbrueckii 137 010605 ferm entLin 135 L02916 fermenti 139 M30121 H59360 h e lv e tic u s 140 H94059 S39821 helveticus 141 S41325 X57248 helveticus 142 S45542 X66723 h e lv e tic u s 143 H37906 H55222 h i lg a r d i i 144 J 05645 M60673 l a c t i s 145 M83946 S43733 p a ra c a s e i 146 M57384 S69823 p e n to s u s 147 S43327 pentosus 148 X62347 pentosus 149 D90339 plantarum 150 D90340 plantarum # 151 J03319 plantarum * 152 H31223 plantarum ‘ 153 M96175 p ls n ta r u n 154 S39954 p la n ta ru m 155 S39956 p la n ta ru m 156 H64090 r e u te r i 157 N84015 sa k e 158 112 Table 7 to Table 10. Table 7, Table 8 and Table 9 compare the codon usages in genes among different species of Lactobacilli. Table 10 shows the average codon usage ofLactobacillus, with and without the dGuo kinase/dAdo kinase genes of L. acidophilus, and is compared with the codon usage by E. coli. Examination of the redundant codons of each amino acid reveals that the dAdo kinase/dGuo kinase genes of L. acidophilus appear to have a preference for uracil and adenine in the third base of the codons. L. helveticus and L. lactis also have similar preferences. These tables should be useful in the cloning of other Lactobacillus genes and in the optimizing of their expression in E. coli. Table 7. Comparison of codon usages among different species of the genus Lactobacillus', acidophilus, brevis, bulgaricus, casei, confusus and delbrueckii. (Codon usage by casei and delbrueckii represents the total of all the genes reported for the species listed in Table 6 .) Codon | Amino aeldoph lus bftvls td ta r ic u i C4f*/ \coafuiui | dtlbruttkii Add U ua* % Uua* % Uua* I % Uua* | % U u j* ^ % U ua* % 1 I GCA !ALA (A) 6 1 36% 7; 1.56% 0 000% 132! 200% " ~ 1 2 | 3 86% 501 1 13% GCC :ALA (A) Q 1 81% 10! 2.22% 8 7 69% 1971 2 99% 6 193% 1681 3.80% GCG ALA (A) [ 1 1 0 23% 2 1 0 44% 1 096% 134 2 03% 8 257% 33 0 75% GCU ALA (A) 1 181% 26 5 78% 2! 192% 1911 2 90% 11 3 54% ^ 3 0 2 94% IAGA jARG (R) 2 045% 0 0.00% 0| 000% 8 012% 0 000% 17 0 38% Ia GG ARG (R) 1 0 23% 0 000% 0 ! 000% 7 0.11% 0 000% 3 007% CGA ARG (R) 1 023% 0 000% 0 ! 000% 28 043% 0 000% 3 007% CGC ARG (R) 2 0 45% 1 022% 2 1 92% 64 097% 4 1 29% 53 1 20% CGG ARG (R) 2 0 45% 5 1.11% 2 1 92% 64 097% 0 72 1 63% CGU ARG (R) 12 2 72% 9 2.00% Oi 000% 71 1 08% 9 28 063% AAC ASN(N) S 1 81% I 191 4 22% 2 1 92% 127! 1 93% 11 138 312% 18 4 08% 0 0 00% 1i 0.96% 131 199% 2 IGAC ASP (D) 7 1 59% 17 3 78% 6 5 77% 147 2 23% 12 GAU ASP (D1 32 726% 23 511% o> 000% 281 4 27% 15 F4B2%I 70! 1 58%l UGC CYS (C) 0 0.00% 1 022% 0 000% 8 0 12% 0 16 0.36% UGU CYC (O 0 000% 0 000% 0 000% 16 0 24% 0 5 011% CAA |GLN(G) 13 2 95% 16 356% 2 192% 172 2.61% 14 4.50% 59' 1 33% CAG Ig LN (Q) 0 000% 0 4 3 85% 125 1 90% 0 000% 1 95% GAA GLU (E) 20 4 54% 33 7 33% 7 6 73% 27? 4 19% 9 2 89% ^ ^ 8 3 GAG Ig LU (E) 4 0 91% 0 0 00% 0 0.00% 84 1 28% 5 1 61% 45l 1 02% GGA IGLY (G) 4| 0 91% 2 0 44% 1 0 96% 40 061% 5 161% 22 050% GGC GLY (G) 11 0 23% 13 2 69% 2 192% 166252% 6 193% 173 391% GGG iGLY (G) 41 091% 5 1 11% 4 3 85% 59 090% 2 0 64% 71 1 61% GGU GLY (G) 10 i 2 27% 14 311% 3 288% 185 2.81% 16 514% 66 149% CAC HIS (H) sl 0 66% 10 2 22% 1 0.96% 66 100% 6 1 93% 72 163% CAU IHIS (H) 5' 1 13% 4 0 89% 0 0 00% 104 1 58% 1 19 0 43% AUA IILU (1) ll 0 23% o1r 0 00% 0 0 00% 10 0 15% 0 2 0 05% ADC ILU (1) 3 068% 8] 1 78% 10 9 62% 193 293% 7 2 25% 143 323% AUU ILU (1) 22 4 99% 111 2.44% 2 192% 254 3.86% 15 4 82% 62 1 40% CUA LEU (L) 71 159% 2 0.44% 0 000% 38 : 0.58% 0 0 00% 10 023% cu e LEU (L) 4 1 091% 3 0 67% 2 1 92% 46 : 0 70% 1 0 32% 42 095% CUG LEU (L) 2! 0 45% 1! 0 22% 4 3 85% 107 1 62% 1 0 32% 159 360% CUU LEU Codon Amino l/M/wtteu* |thJHUfdU | (Ktff iptn tttti | Acid U ua* ' % U ua* % Uuq* % U ua* ! % Uujj*_ % j GCA AuA (A) 181 1 68% 51 2 46% 12 1 77% 4) 1 53% 63 286% GCC ALA (A) 36! 3 37% 15 0 72% 7 103% 41 1 53% 69 313% GCG ALA (A) 13: 122% 9 0 43% 7 103% 11 0 38% 51 232% GCU ALA (A) 34: 3 18% 76 3 67% ^ 1 7 2.51% 9l 3 44% 70 318% AGA ARG (R) 0 000% 23 111% 6 0 89% 3 1 15% 1 0 05% AGG ARG (R) 2 019% 1 005% 1 015% 0 000% 3 014% CGA ARG (R) 1 0.09% 8 039% 5 0 74% 3 1.15% 3 0.14% CGC ARG (R) 5 047% 6 0 29% 10 1 48% 0 000% 8 0 36% CGG ARG (R) 12 1 12% 2 010% 8 1 18% 0 000% 6 027% CGU ARG (R) 17 1 59% 40 1 93% 2 36% — £1 1 91% _^_8 036% AAC ! a SN(N) 27 2 53% 47 2 27% 16 2 36% 1 53% 55 2 50% AAU lASN(N) 211 1 96% 93 4 49% _ ^ ^ 9 4 28% 305% _ J ! 2 i 318% GAC !a SP(D) 27) 2 53% 46 222% 10! 1 48% 1 91% 70 318% GAU 'ASP (D) 42 393% 109 5 26% 35 517% 161 611% ^ ^ jo a 490% IIGC CYS(C) 3 0 28% 6 0 29% 2 ! 030% 1 038% 0 ugu !c y c Codon lAmino pcnroiui plMnurvm nanrt U l l I * ’* Acid U ua* : % U ua* . % U ua* I % U ua* l % 1 U u w l % i GCA 'ALA {A) 45 2 28% 31! 1 59% 21 0 75% 8 1 67% 14 2 87% GCC ALA (A) 21 1 06% 4SI 2 31% 2 0 75% 16 3 34% 0 000% GCG IALA (A) 42 213% 13 067% 0 000% 3 0 63% 3 062% GCU lALA (A) 50 2 54% 79 4 05% J 3 1 13% 11 2 30% 21 4 31% AGA jARG(R) 12 061% 12 0 62% 7 264% 1 021% 1 021% AGG IARG (R> 3 015% 0 0 00% 3 1 13% 0 000% 1 021% CGA ARG (R) 19 096% 5i 026% 1 038% 2 0 42% 4 082% CGC ARG (R) 10 051% 9 046% 0 000% 9 1 88% 1 021% CGG ARG (R) 28 1 42% 14 0 72% 0 0 00% 6 1 25% 1 021% CGU ARG (R) 23 1 17% 34 1 74% _ _ 1 038% 6 1 25% 11 226% AAC ASN(N) 24 1 22% 66 338% 9 340% 14 292% 13 2.67% AAU ASN1N1 62 314% 47 241% _ J 8 679% 13 2.71% 9 1 85% GAC ,ASP (D) 35 1 77% 77 395% 1 038% 5 1 04% 15 308% GAU lASP (D) 68 4 46% 75 3 84% _ 8l 302% 27 564% 28 575% UGC iCYS (C) 7 035% 2,010% 0 000% 0 000% 0 UGU 'CYC (C) 12 061% 9 046% 0 75% 0 000% 3 CAA GLN(Q) 75 380% 62 3 16% 264% " 30 626% 14 2.87% CAG IGLN (Q) 26 1 32% 5' 026% ^ _ i 0.38% 0 000% 0 IGAA GLU (E) 93 4 72% 104 533% 121 4 53% 26 543% 34 Igag GLU (E) 41 208% 12 062% 075% 2 042% 1 021% GGA IGLY (G) 31 1 57% 16 082% 2 075% 1 021% 3 062% GGC GLY (G) 30 1 52% 35 1 79% 1 038% 15 313% 7 1 44% GGG 'GLY (G) 19 096% 18 092% 1 038% 3 0.63% 0 0 00% GGU GLY (G1 50 2,54% 61 1343% 2 1 0 75% 13 2 71% 28 575% CAC HIS (H) 9 046% 13j 0 67% 1 0.38% 6 1.25% 3 062% CAU .HIS (H) 41 208% 151 0 77% 6 226% 6 1.25% 4 0 82% AUA [iLU (1) 5 025% ' 5026% 13 4.91% 1 0 21% 2 0.41% AUC ILU (1) 22 1 12% 56 287% 3 1 13% 8 167% 8 164% AUU IlLU (1) 90 4 56% SO 256% 15 566% 12 2 51% 23 4 72% CUA LEU (L) 16 081% 10| 051% 4 1 51% 3 063% 0 0 00% CUC 'LEU (L) 1 005% 41 0 21% 1 0 38% 10 209% 3 062% CUG LEU (L) 23 1 17% 4l 021% 0I 000% 2 042% 0 0 00% CUU LEU (L) 6 030% 191 0 97% 2 0 75% 8 167% 4 0 82% UUA ,LEU(L) 75 380% 92 472% 12 4 53% 10 209% 14 287% UUG ILEU(L) 63 319% 47 2 41% 3 1 13% 0 000% 9 185% AAA i LYS (K) 69 350% 78' 4 00% 26 9 81% 16 334% 12 246% AAG LYS IK) 68 345% 701 3 59% 4 1 51% 10 209% 19 390% AUG MET (M) 33 1 67% 451 2 31% 4 1 51% 11 230% 16 329% UUC PHE (F) 25 1.27% 471 2.41% 4 1 51% 17 355% 5 103% UUU !PHE(F) 52 264% 29! 1 49% 16 6 04% 7 1 46% 11 2 26% CCA ,PRO(P) 20 1 01% 36 1 85% 2 0 75% B 1 67% 20 4.11% CCC : PRO (P) 6 0.30% 2 0 10% 1 038% 10 209% 0 000% CCG PRO(P) 9046% 20 10% 0 000% 5 104% 1 021% CCU IPRO(P) 15 076% 20 1 03% 5 1 89% 4 084% 5 1 03% |AGC SER (S) 14 ' 0 71% 11 056% 4 1 51% 6 1 25% 6 1 23% UGU SER (S> 23 1 17% 18 0 92% 5 189% 5 1.04% 6 123% lUCA SER (S) 34 172% 36 1 85% 6 226% 7 1 46% 10 205% UCC SER (S) 4 020% 13 067% 2 0 75% 2 042% 2 0 41% UCG SER (S) 24 1 22% 4| 021% 2 0 75% 1 021% 0| 000% UCU SER (S) 20 1 01% 26! 1 33% 1 0 38% 6; 1 25% 7! 1 44% UAA STOP() I 5! 02554.1 6| 0 31% 2 075% o[ooo% 2 0 41% UAG STOP () 0 000%1 Ol 0 00% 0 0 00% ol 000% 0 0 00% UGA STOP() 0 000% I o: 0 00% oi 000% 1! 0 21% 0 ! 000% ACA THR (T) 30 j 152% 14, 072% 1 7; 2 64%| 5 J_04% 5 103% ACC THR (T) 21 106% 131 0 67% 1 0 38% 6 1 25% 4 0 82% ACG .THR (T) 311 157% 191 0 97% O' 000% 7 146% 1 0 21% ACU THR (T> 24! 1 22% 53! 2 72% 2 0 75% 9 1 88% 13 '267% UGG TRP(W> 301 1 52% 151 0 77% I 11 036%l 6 125% 7 144% UAC TYR (Y) 11 056% 491 2.51% 3 1 13% 11 230% 13 2 67% UAU TYR fY) 80I 406% 411 2 10% 3 77% 10 209% 11 226% GUA VAL (V) 191 0.96% 24. 1 23% 4 1 51% 0 000% 4 0 82% GUC 'VAL (V) 26 { 1 32% 171 0 87% o' 000% 16 3 34% 4 082% GUG !VAL (V) 32| 1 62% 221 1 13% 1 0 38% 3' 063% 2 0 41% GUU VAL (V) 50 2 54% 95 4 87% 7 r 2 64% 1 131 2 71%| 19 390% 116 Table 10. Codon usage of Lactobacillus genes. (* not include dGuo kinase/dAdo kinase of L. acidophilus.) Codon Amino L. acidophilus Lactobacillus * (Avg.) Total Lactobaclllus(Avg.) E. coll (Avg.) Acid Usaga i % Usaga % Usaga % Usaga 1 % i GCA iALA (A) 6 1 36% 10 1 86% 10 1 87% 28 : si% GCC ALU (A) e 1 81% 14 2 61% 14 2.62% 14 1 40% GCG ALA (A) 1 0 23% 7 1 30% 7 1 31% 28 I 281% GCU 'ALA (A) 8 1 81% 17 317% 17 318% 46! 4 61% AGA ARG (R) 2r 0 45% 2 0.37% 2 0 37% 1 010% AGG ARG (R) 1 0 23% 1 0 19% 1 019% 0 0 00% CGA ARG (R) 1 0 23% 2 0 37% 2 0 37% 1 0 10% CGC ARG (R) 2 0 45% 4 074% 4 0.76% 19 1 90% CGG ARG (R) 2 0 45% 5 0.93% 5 0 93% 1 0 10% CGU ARG (R) 12 2 72% 6 1 12% 7 1 31% 38 3.8151 AAC ASN(N) 8 1 81% 13 242% 13 243% 31 3115 AAU ASN(N) 18 4 08% 13 242% 13 2 43% 6 0 60% GAC ASP (O) 7 1 59% 16 2 98% 16 299% 31 3.11% GAU ASP (D) 32 7 26% 21 391% 22 4 11% 21 2 10% UGC CYS (C) 0 0 00% 1 0 19% 1 019% 4 0 40% UGU CYC (C) 0 0 00% 2 0 37% 2 0 37% 3 030% CAA GLN (O) 13 2 95% 14 261% 14 2.62% 9 0 90% CAG GLN (Q) 0 0 00% 7 130% 7 1.31% 31 311% GAA GLU (E) 20 4 54% 27 5 03% 26 4.86% so - 501% GAG GLU (E) 4 091% 5 0 93% 5 0 93% 18 1 80% GGA GLY (G) 4 0 91% 5 0 93% 5 0 93% 3 0.30% GGC GLY (G) 1 0 23% 13 2 42% 13 2 43% 31 311% GGG GLY (G) 4 0 91% 5 093% 5 093% 4 0 40% GGU GLY (G) 10 2 27% 16 2 98% 16 2 99% 41 4.11% CAC HIS (H) 3 0 68% 6 1 12% 6 1.12% 13 1 30% CAU HIS (H) 5 1 13% 6 1 12% 6 1 12% 7 0 70% AUA ILU (1) 1 0 23% 1 019% 1 019% 0 000% AUC ILU (1) 3 0 68% 13 2.42% 13 2 43% 41 4 11% AUU ILU (1) 22 4 99% 18 3 35% 19 3.55% 17 1 70% CUA LEU (L) 7 1 59% 3 056% 3 0 56% 1 010% c u e LEU (L) 4 0 91% 3 0 56% 3 0 56% 5 0 50% CUG LEU (L) 2 0 45% 8 1 49% 8 1 50% 62 6.21% CUU LEU (L) 16 3 63% 7 1 30% 7 1 31% 5 050% UUA LEU (L) 17 3 85% 12 2 23% 12 2 24% 5 0.50% UUG LEU (L) 6 1 36% 13 242% 13 2 43% 6 060% AAA LYS (K) 20 4 54% 16 298% 16 2 99% 52 5 21% AAG LYS (K) 13 2 95% 21 3 91% 21 3 93% 20 2.00% AUG MET (M) 17 385% 12 2 23% 12 2 24% 23 230% UUC PHE (F) 7 1 59% 10 1 86% 10 1 87% 20 2 00% UUU IPHE (F) 8 1 81% 11 2 05% 11 2 06% 10 100% CCA PRO (P) 8 1 81% 10 186% 10 1 87% 7 0.70% CCC PRO (P) 3 0 68% 2 037% 2 037% 2 0.20% CCG PRO (P) 1 0 23% 5 0.93% 5 0 93% 25 251% CCU PRO (P) 10 2 27% 4 074% 4 0.75% 3 030% AGC SER (S) 2 0 45% 6 1 12% 5 093% 10 1.00% AGU SER (S) 13 2 95% 5 0 93% 6 1 12% 4 0.40% UCA I SER (S) 6 1 36% 7 1 30% 7 1.31% 3 0.30% UCC SER (S) 3 0 68% 4 0 74% 4 0 75% 14 1 40% UCG SERJS) o 0 00% 2 0 37% 2 0 37% 3 0 30% UCU SER !S) 4 091% 5 0 93% 5 0 93% 17 1 70% UAA STOP() 1! 0 23% 1 0 19% 1 0 19% 0 0 00% UAG STOPO 1' 0 23% 0 0 00% 0 0 00% 0 0 00% UGA STOP() 0 0 00% 0 0 00% “ o 1 000% 0 0 00% ACA THR (T) 5 1 13% 6 1 12% 6 1 12% 4' 0 40% ACC |THR (T) 2 0 45% 9l 1 68% 9 1 68% 251 2 51% ACG THR (T) o 0 00% 7 1 30% 6 1 12% 5 0 50% ACU |THR(T) 11 2 49% 111 2 05% 11 206% 20 2 00% UGG ;TRP(W) 3 068% 6 1 12% 6 1 12% 7 0 7i% UAC !TYR (YJ 9 2 04% 10 1 86% 10 1 87% 15 1 50% UAU !TYR (Y) 24 5 44% 11 2 05% 11 2 06% 7 0 70% GUA iVAL (V) 5 1 13% 5 0 93% 5 093% 22 2.20% GUC VAL (V) 3 0 68% 9 1 68% 9 1 68% 9 0 90% GUG IVAL (V) 2 0 45% 6 1 12% 6 1 12% 18 1 80% GUU VAL (V) 10 2 27% 18 3 35% 18 3 36% 32 321% CHAPTER VI EXPRESSION AND CHARACTERIZATION OF CLONED DEOXYADENOSINE KINASE/DEOXYGUANOSINE KINASE Since the dAdo kinase gene and dGuo kinase (or dCyd kinase) gene from L. acidophilus contain consensus promoter elements and ribosomal binding sites homologous to that in E. coli in the 5’ upstream regions, also, the dGuo kinase (or dCyd kinase) gene contains a putative transcription terminator in the 3’ downstream region, expression of the cloned genes in E. coli could possibly occur using their endogenous promoters. Indeed, enzyme activities for dAdo phosphorylation and dGuo phosphorylation, which are absent from E. coli, were detected in transformed E. coli containing the Kpn I L. acidophilus clone. The cloned dAdo kinase/dGuo kinase protein has been purified and compared with the wild type kinases. A. Enzyme Activities Lysates of E. coli XLl-Blue cells containing the pBluescript vector with or without insert (control) were assayed for all three deoxynucleoside kinase activities: dCyd kinase, dGuo kinase and dAdo kinase. The three kinase 117 118 activities were detected neither in the Xba I clone, which accommodated an intact dAdo kinase gene and a partial dCyd kinase (or dGuo kinase) gene, nor in the cells bearing the control vector with no insert. However, enzyme activities of dGuo kinase and dAdo kinase, but not dCyd kinase, were observed in the Kpn I clone. The dAdo kinase activity and dGuo kinase activity were 1000 fold and 2000 fold higher, respectively, than the control, as shown in Table 11. Addition of IPTG to activate the lac promoter did not further increase the kinase activities, indicating the dAdo kinases/dGuo kinase was expressed using its endogenous promoter. Enzyme specific activities for both dAdo phosphorylation and dGuo phosphorylation in the crude extract of the Kpn I clone were more than 10-fold higher than those in the wild type L. acidophilus. The fact that expression of dGuo kinase and dAdo kinase is a property of the Kpn I clone (two intact genes), but not of the Xba I clone (1 gene plus 2/3 gene), suggests that both intact subunits might be needed to form an active enzyme, or that some sequence information in or downstream of the dGuo kinase gene--the terminator for example-could be important for expression. B. Kinetic Characteristics Wild type dGuo kinase/dAdo kinase and dCyd kinase/dAdo kinase from L. acidophilus were known to exhibit positive allosteric interactions (4,7); dAdo phosphorylation in dGuo kinase/dAdo kinase is stimulated more than 3 fold by dGuo; dAdo phosphorylation in dCyd kinase/dAdo kinase is activated about 5 119 Table 11. Specific activities of crude extracts from E. coli and from L. acidophilus. (units/mg) Kpn I clone Control LBA = pBS in XLl-Blue R-26 + IPTG - IPTG + IPTG - IPTG dAdo 0.7 0.8 0.0007 0.0007 0.06 kinase dGuo 4.8 6.7 0.0021 0.0025 0.33 kinase dCyd 0.09 0.00 0.05 0.00 0.32 kinase 120 fold by dCyd. On the other hand, dGTP and dATP are potent end product inhibitors, respectively, to dGuo kinase and dAdo kinase (8). Cloned dGuo kinase/dAdo kinase were tested for the kinetic characteristics of positive allosteric interaction and inhibition. Similar to the wild type kinases from Lactobacillus, the cloned dAdo kinase in crude extract was activated 4.6 fold by 0.1 mM dGuo, and inhibited strongly by 0.5 mM dATP (only 10% activity remained); dGuo kinase was also potently inhibited by 0.5 mM dGTP, with only 5% activity remaining. See Table 12. However, dCyd has no stimulating effect on dAdo phosphorylation, indicating the dAdo kinase cloned is likely to be the one associated with the dGuo kinase, and not with the dCyd kinase. C. Purification by dAp^-Sepharose Purification procedures similar to those used to purify the wild type dGuo kinase/dAdo kinase from L. acidophilus were applied to the cloned kinases. A crude extract of the Kpn I clone was partially purified by streptomycin and ammonium sulfate fractionation. Then, the ammonium sulfate fraction (IV) was applied directly to the dAp 4 -Sepharose affinity column. Since no dCyd kinase activity was detected in the crude extract, the dCp 4 -Sepharose affinity chromatography purification step prior applying to the dAp 4 ~Sepharose column was omitted. Details of culturing cells, protein extraction, and purification were presented previously in Chapter II. The cloned enzyme was purified to 121 Table 12. Kinetic characterization of cloned dAdo kinase/dGuo kinase dAdo kinase dGuo kinase Unit/mg Unit/mg No additions 0.73 7.05 dGuo, 0.1 mM 3.40 - dATP, 0.5 mM 0.08 - dGTP, 0.5 mM - 0.35 122 homogeneity by dAp^-Sepharose as analyzed by SDS-PAGE (see below). The specific activities of the purified enzyme were 65 units/mg and 1000 units/mg, respectively, for dAdo kinase and dGuo kinase from the clone, compared with 250 units/mg and 2000 units/mg from the wild type Lactobacillus cells. Compared with the wild type enzymes, the specific activities of the cloned enzymes were 4-fold lower in dAdo phosphorylation, and 2-fold lower in dGuo phosphorylation. Moreover, the activities of the cloned dAdo kinase/dGuo kinase, both the ammonium sulfate fraction and the dAp 4 -Sepharose purified fraction, are more sensitive to dilution than the wild type. These facts indicate that the cloned dAdo kinase/dGuo kinase somehow differ from their wild type counter parts, possibly in the N-terminal sequences, since E. coli might not be equipped with a processing mechanism to remove the two extra residues, threonine and valine, which follow the initial methionine as discussed above. This was tested by N-terminal amino acid sequence analysis (see below). D. Subunit Molecular Weight by SDS-PAGE Crude extracts of the Kpn I clone were run on an SDS-polyacrylamide gel (Laemmli buffer system) side by side with the crude extracts of XLl-Blue cells transformed with only pBluescript vector (control). A 26,000 dalton subunit band was seen in the Kpn I clone but not in the control, when visualized by Coomassie- blue staining (Figure 32). The dAdo kinase/dGuo kinase protein production in the crude was estimated to be 1 % of total cell proteins in the Kpn I E. coli 123 Daltons 97,400 66,200 45,000 31,000 21,500 14,400 Figure 32. SDS-polyacrylamide gel (laemmli buffer system) analysis of expressed proteins from E. coli crude extract. Lane 1 , protein markers; lane 2, crude extract of Kpn I clone; lane 3, crude extract from control. Position of the dAdo kinase/dGuo kinase protein band is indicated by an arrow. 124 clone, and was absent from E. coli cells transformed with vector lacking the insert. Wild type dAdo kinase/dGuo kinase has previously been shown in Chapter III to be a heterodimer with similar subunit molecular weight of 26,000 dalton. Cloned dAdo kinase/dGuo kinase was purified similarly as for the wild type enzyme by dAp^-Sepharose affinity medium. Electrophoresis of this preparation of the cloned enzyme on a SDS-polyacrylamide gel (Laemmli buffer system) reflected a single polypeptide band of 26,000 Da in size, which was identical to the wild type protein, when visualized by Coomassie Blue staining (Figure 33). This result indicates the cloned enzyme has similar subunit molecular weights to the wild type protein, and agrees with the size prediction of the gene products inferred by the nucleotide sequences of the genes. The two subunits of the cloned dAdo kinase/dGuo kinase were resolved by SDS-polyacrylamide gel electrophoresis (pH7 buffer system), as shown in Figure 34. The subunits of the cloned enzyme apparently migrated at the same rates as the wild type counterparts. Dr. Ikeda has identified the functional assignments of the two subunits of dAdo kinase/dGuo kinase, for both the wild type and the cloned enzymes, by affinity labeling with substrate-binding site specific probes and by differential tryptic digestion ( 8 8 ). The slow migrating subunit was identified as the dAdo kinase; the fast migrating subunit was identified as the dGuo kinase. It is interesting that the dGuo kinase, the fast migrating subunit, is a larger polypeptide than the dAdo kinase, as inferred by the nucleotide sequences of the 125 Daltons 1 2 3 97.400 66,200 45.000 31.000 21,500 14.400 Figure 33. SDS-polyacrylamide gel (laemmli buffer system) analysis of dAp^- Sepharose purified dGuo kinase/dAdo kinase from the E. coli clone. Lane 1, protein markers; lane 2, dAp^Sepharose purified dAdo kinase/dGuo kinase from E. coli clone; lane 3, dAp^Sepharose purified wild type dAdo kinase/dGuo kinase from L. acidophilus. Position of the dAdo kinase/dGuo kinase protein band is indicated by an arrow. 126 Figure 34. SDS-polyacrylamide gel (pH 7 buffer system) analysis of dAp4- Sepharose purified dGuo kinase/dAdo kinase from the E. coli clone. Lane 1, dAp4-Sepharose purified wild type dAdo kinase/dGuo kinase from L. acidophilus', lane 2, dAp4-Sepharose purified dAdo kinase/dGuo kinase from E. coli clone; lane 3, protein markers. Positions of the dAdo kinase/dGuo kinase subunit peptide bands are indicated by arrows. 127 genes. (The dGuo kinase gene encodes a 26 KDa polypeptide of 224 amino acids, and the dAdo kinase gene encodes a 25 KDa polypeptide of 215 amino acids.) Whether this inverted migration pattern is an indication of differential binding of SDS, or of posttranslational modification such as glycosylation or phosphorylation, has to be determined by further studies. In any case, since the cloned dAdo kinase/dGuo kinase subunits appear to behave like the wild type subunits in terms of the electrophoretic mobility, it seems unlikely that their processing by E. coli differs greatly from what occurs in Lactobacillus acidophilus. E. N-Terminal Protein Sequences As mentioned above, the DNA sequence of the cloned dAdo kinase gene revealed two extra codons, for threonine and valine, between the initial methionine and isoleucine codons (Figure 28). These two extra amino acids are absent from the known N-terminal amino acid sequence of the wild type dAdo kinase. Furthermore, the N-terminal sequence of the wild type dGuo kinase has no initial methionine, but a methionine codon is present in the cloned dGuo kinase gene. If E. coli is not equipped with a processing mechanism to remove the additional amino acid residues, the N-terminal amino acids of the cloned protein would differ slightly from the wild type enzyme. With the conserved ATP-binding site sequence so close (only five amino acid residues away), this small change might indeed affect the catalytic properties or stability of the enzyme. 128 Protein purified from the dAdo kinase/dGuo kinase clone was submitted for N-terminal protein sequence analysis. The native protein, which contained both subunits, was concentrated onto PVDF membrane and submitted to six sequencing cycles, and the remaining residues were submitted for amino acid analysis to determine the actual amount of protein on the membrane. Only one amino acid was detected in each sequencing cycle, in the order of: Thr, Val, lie, Val Leu and Ser. The sequence agreed exactly with that of wild type dGuo kinase, and also agreed with the inferred amino acid sequence of both cloned genes except for the absence of the initial methionine. The total amount of protein, based on the size of the native protein (56,000 dalton) but not the subunit size (26,000 dalton), was determined to be 80 pmoles. However, 40 pmoles, which was only 1/4 of the total amount on the membrane, was yielded in the first cycle. Whether the amino acid obtained in each cycle belonged only to one subunit or to both was not conclusive, but at least one of the cloned peptides, possibly the dGuo kinase, whose initial methionine was removed in E. coli, thus has an N-terminal sequence identical to the wild type dGuo kinase. The other subunit, regardless of whether the N-terminus was blocked or started with a threonine, was different from the N-terminus of the wild type dAdo kinase subunit. CHAPTER VII CONCLUSIONS AND RECOMMENDATIONS dAp 4 -Sepharose, which is chemically synthesized by linking the potent natural inhibitor of dAdo kinase (dATP) to Sepharose, has been used successfully to retain and purify dAdo kinase from L. acidophilus. dGuo kinase activity is co purified with the dAdo kinase activity, as expected, when specifically eluted with dATP, thus confirming the findings of Deibel (4,6) and Chakravarty (7) that native dGuo kinase is physically inseparable from dAdo kinase although they have separate substrate phosphorylation sites. The activity ratio of dGuo kinase to dAdo kinase, about 5-8 : 1 in the absence of activator or inhibitor, remained almost unchanged throughout the purification procedures. After the final affinity purification step with dAp 4 -Sepharose, the dGuo kinase/dAdo kinase appeared to be homogeneous by SDS-polyacrylamide gel electrophoresis when visualized by Coomassie Blue staining. The specific activities of both dGuo kinase (2150 units/mg) and dAdo kinase (280 units/mg) have increased by about 2,700 fold. When the purified dGuo kinase/dAdo kinase was applied on the SDS- polyacrylamide gel (Laemmli buffer system), a single protein band of 26,000 Da was observed. Since the native dGuo kinase/dAdo kinase has previously been 129 130 shown to have a molecular weight of 50,000 Da (4), it appears the enzyme is composed of two subunits of similar size. This is analogous to the case of dCyd kinase/dAdo kinase (9). However, this interpretation disagrees with the previous work of Chakravarty (7), in which dGuo kinase/dAdo kinase was claimed to be a monomeric polypeptide of 56,000 Da. In the previous case, the enzyme was purified by a rather lengthy conventional procedure, and the amount of pure enzyme obtained was only about 1 % of that prepared in this work, so it is probable an impurity was mistakenly identified as the enzyme, or possibly, the protein with fused subunits was isolated at that time. The two subunits of dGuo kinase/dAdo kinase were successfully resolved by SDS-PAGE under the pH7 MZE 3328.IV buffer system, suggesting that the two subunits are nonidentical. The N-terminal protein sequences of the two subunits have been determined (Figure 16). They are identical to each other except that the fast migrating peptide (identified as dGuo kinase by affinity labeling by Dr. S. Ikeda) lacks an initial methionine, but starts with a threonine followed by a valine, while these two amino acids are not present in the N- terminal sequence of the slow migrating peptide (identified as dAdo kinase, also by Dr. Ikeda). Except for uncertainty as to the identity of amino acid residues at positions 14,15,17 and 18 of dAdo kinase, and the initial threonine and valine of dGuo kinase, the subunits of dAdo kinase and dGuo kinase are identical to the known sequences of dAdo kinase and dCyd kinase, respectively (Figure 28). 131 Based on the N-terminal protein sequence of dCyd kinase/dAdo kinase, cloning probes have been constructed by employing the DNA amplification method of the polymerase chain reaction (PCR). These probes, PCR 93mer and PCR 117mer, have been shown to consist of the non-degenerate N-terminal coding sequence of the dCyd kinase gene (and the dGuo kinase gene). The probes generated by PCR, with their sequence degeneracy mostly eliminated and therefore having a higher degree of specificity, have overcome the limitation of using short degenerate oligonucleotide probes in cloning experiments, especially colony hybridization. Success in amplifying a target genomic DNA fragment, using the highly degenerate mixed oligonucleotide primers, has been shown to depend greatly on the conditions of PCR. The optimal conditions for PCR have been worked out, and the biphasic annealing temperature cycles have been shown to be most effective. The conditions giving the most satisfactory results are as follows (final concentrations): in a total volume of 50 /xl, 0.5 ng total genomic DNA, 5 juM primer [+4J, 3 fiM primer [-5], 200 juM (each) dNTP, 2.5 mM MgCl2 , 50 mM KC1, 10 mM Tris-Cl (pH 8.4 at room temperature), and 1.25 units of Taq polymerase. The most effective PCR cycles are: (i) an initial denaturation at 94°C for 2 minutes, then (ii) 30 cycles of denaturation at 94°C for 1 minute, annealing for 1 minute at 37°C for the first five cycles, and 50°C for the subsequent 25 cycles, and extension at 72°C for 30 seconds, and followed by (iii) a final extension at 72°C for 7 minutes. 132 A clone containing two intact genes, with a sequence matching the known N-terminal sequence dCyd kinase/dAdo kinase and also dGuo kinase/dAdo kinase, has been identified by hybridization with the PCR 117mer probe. This clone was selected from a Kpn I restricted genomic library of L. acidophilus, with the 2.5 Kb Kpn I average genomic fragments inserted into the pBluescript cloning vector and transformed into E. coli XLl-Blue cells. The DNA sequences of the two genes have been completely determined (Figure 27), from the combined sequence results of subclones of smaller fragments, as well as from sequencing reactions primed with sequence specific primers (gene walking). Compared with the known peptide sequences of dCyd kinase/dAdo kinase and dGuo kinase/dAdo kinase, the derived amino acid sequence of the upstream gene matches exactly with the peptide sequence from residues 2 to 28 of the two dAdo kinases, but contains two extra codons, threonine and valine (which are not present in the known peptide sequence), immediately after the initial methionine (Figure 30). Except for the absence of an initial methionine, the derived sequence of the downstream gene agrees exactly with the known peptide sequence of dGuo kinase; it also agrees with the known 28 amino acid sequence of dCyd kinase, except for the threonine and valine following the initial methionine. These observations, combined with later results with enzyme protein expressed by this clone, make it clear that the upstream gene is dAdo kinase and the downstream gene is dGuo kinase. 133 These two genes appear to be arranged in tandem on the chromosome such that the start of the dGuo kinase gene is only 21 nucleotides downstream from the stop codon of the dAdo kinase gene. The dAdo kinase gene is 648 base pairs in length, the 215 codons encode a peptide with a predicted size of 25 KDa. The next 675 bp open reading frame (dGuo kinase gene) encodes a peptide of 224 amino acid residues with predicted molecular weight of 26 KDa. The two genes are highly homologous to each other: with 65% overall homology in DNA sequence, and 60.9% identity in the derived amino acid sequence. The gene of dCyd kinase is not located in the upstream or downstream regions in both the Kpn I clone and Xba I clone; thus it is probably located elsewhere in the genome. Potential promoters, ribosome binding sites and transcription terminator have been identified (Figure 27). Preceding the dAdo kinase gene, but not the dGuo kinase gene, promoter consensus sequences common in gram positive bacteria are found. The sequences "GAAAGA" and " GAAGGA", which are 6 nucleotides upstream from the translation start codons of the dAdo kinase gene and dGuo kinase gene, respectively, are likely to be the ribosome binding sites. Four nucleotides downstream from the translational stop codon (TAA) of dGuo kinase gene is a GC-rich region of dyad symmetry, "AAAACTGCGGTCCA" and 'TGGATCGCAGTTTT', which is followed by a stretch of thymines (or uridines in the mRNA); this loop may function as a transcription terminator. Because the upstream gene (dAdo kinase) lacks a transcription terminator and the downstream gene (dGuo kinase) lacks a promoter of gram positive bacteria, it is 134 likely that a polycistronic message is transcribed in L. acidophilus. Whether a polycistronic or two separate messages are transcribed can be tested by a Northern blot, with DNA probes specific to each gene. Since the N-terminal sequences of the three deoxynucleoside kinases-upon which the cloning probe was based-are very similar, the identities of the two cloned genes cannot be assigned until enzyme activities of dAdo phosphorylation and dGuo phosphorylation are reflected in the cell-free extract of the clone. No dCyd kinase activity, however, is detected. The dAdo kinase and dGuo kinase activities detected in the crude extract of the clone, expressed using its endogenous promoter, are 1 0 0 0 -fold and 2 0 0 0 -fold, respectively, higher than background, and 10-fold higher than in the crude extract of L. acidophilus. The cloned dAdo kinase/dGuo kinase exhibit allosteric stimulation and inhibition patterns similar to those of wild type enzymes. dAdo kinase is activated about 5-fold by dGuo and is strongly inhibited by dATP (90% activity loss), while dGuo kinase is also strongly inhibited by dGTP (95%). The specific activities of the cloned dAdo kinase/dGuo kinase prepared by dAp 4 -Sepharose affinity purification are 65 units/mg and 1000 units/mg, respectively. Compared to the wild type enzymes, the specific activities of these purified cloned enzymes are 2-4 fold lower and they are more sensitive to dilution. Subunit molecular weights of the cloned enzymes (26,000) determined by SDS-PAGE appear to be identical to those of wild type. 135 Results of N-terminal peptide sequencing of the cloned enzyme showed a sequence which is identical to that of the wild type dGuo kinase, i.e., with the initial methionine processed out, but containing threonine and valine residues. However, the dAdo kinase sequence was not reflected in the sequence of the cloned protein. Possibly the N-terminus of the dAdo kinase is blocked-a possibility made greater by the fact that only 25% of the total protein was sequenceable. Thus, it is not conclusive whether the E. coli is also equipped with a processing mechanism to remove the extra threonine and valine codons present in the gene. It is quite likely that the two extra codons (threonine and valine) present in the gene are not processed in E. coli as they were in L. acidophilus. Because these two extra amino acids are located so close to the ATP-binding site at the N-terminus of dAdo kinase, they might affect the folding, substrate binding, catalytic activities and/or stability of the dAdo kinase, and might even affect the association with dGuo kinase, which shows allosteric interaction with dAdo kinase. Nevertheless, the possibility that the cloned enzymes might differ from the wild type enzymes in other respects should not be neglected. Three regions of consensus sequences, which are common in herpeviral thymidine kinases, thymidylate synthetases, human deoxycytidine kinase and adenylate kinases, are revealed in the derived amino acid sequence of the cloned dAdo kinase and dGuo kinase. Namely, (i) the glycine-rich loop (GXXGXGKT/S), located at the N-termini of the dAdo kinase and dGuo kinase, (ii) the DRH/Y/S fragment located at about 80 residues away from the N- 136 termini, and (iii) the arginine-rich site (RXXXRXR), situated at about 145 residues away from the N-termini. Regions (i) and (iii) have been extensively studied and shown to be important in nucleotide binding and substrate phosphorylation. Site (ii), implicated in nucleoside binding, is being studied in our laboratory by site-directed mutagenesis. Codon biases of the dAdo kinase/dGuo kinase of L. acidophilus R-26 has been calculated and compared with those of other Lactobacillus genes. For those amino acids with redundant codons, the dAdo kinase/dGuo kinase of L. acidophilus genes appear to have a preference in uracil and adenine in the third base of the codons. In this project, the enzymes dAdo kinase/dGuo kinase have been purified by the newly synthesized dAp 4 -Sepharose affinity medium to homogeneity, the genes of dAdo kinase and dGuo kinase have been successfully cloned and sequenced, and enzyme activities of the heterodimeric dAdo kinase and dGuo kinase were expressed in E. coli. The cloned enzymes, however, behave slightly differently from the wild type enzymes in terms of specific activity and stability. The arrangement of the dAdo kinase and dGuo kinase genes is summarized in Figure 35. It would be interesting to find out whether differences in these characteristics are caused by the extra threonine and valine codons at the N- terminus of the cloned dAdo kinase, or by other factors. Modification at the gene level could be made to remove the two extra codons, and the mutated enzymes P S/D S/D 5 * dAK dGK ORF ORF 648 bp 675 bp 215 amino acids 224 amino acids 25 KDa 26 KDa pH 7 SDS-PAGE: slow migrating peptide fast migrating peptide Figure 35. Arrangement of the dAdo kinase/dGuo kinase genes from L. acidophilus R-26. 138 could be compared structurally and kinetically with the wild type enzymes from L. acidophilus. Although the processing of the initial methionine in the case of dGuo kinase is not uncommon, no genes from any other source have been reported, so far, which give such deviations between the gene and protein sequences as in the dAdo kinase. Would the genes for dCyd kinase/dAdo kinase from L. acidophilus also contain the two extra codons? This question will soon be answered when their genes are cloned. Could the codon GUU of valine (nucleotide positions 7 to 9 in Figure 27) code for an initial methionine in the dAdo kinase? It is known that the valine codon GUG can act as an initial methionine codon in some cases, but GUU has never been reported to code for an initial methionine. By site- directed mutagenesis, the codon AUG (at nucleotide positions 1 to 3 in Figure 27) could first be mutated to determine if translation is initiated at this codon. If a protein was still translated, the codon GUU (nucleotide positions 7-9) could then be mutated to GUA or GUC, which normally codes for a valine, to examine the possibility of GUU serving as a start codon. If the discrepancy is not generated at the translation level, it will be interesting to explore the new processing mechanism. High level expression of cloned dAdo kinase/dGuo kinase by subcloning into an expression vector is being carried out in our laboratory. After the enzymes are overexpressed, structural studies such as x-ray crystallography, circular dichroism and fluorescence measurements, etc., which require large 139 amounts of protein, would be possible to perform. We know that in L. acidophilus, the dAdo kinase associates with dGuo kinase in the native state of the enzyme. However, we cannot exclude the possibility that dAdo kinase itself and/or dGuo kinase itself can form active monomers or dimers. With the dAdo kinase and dGuo kinase genes in hand, separate expression of each gene could be performed. dAdo kinase is found to be associated with dGuo kinase as well as with dCyd kinase. A fascinating question is whether dGuo kinase and dCyd kinase are sharing the same dAdo kinase gene product. If they are, would the dAdo kinase be produced from one gene be expressed at a level high enough to supply the dGuo kinase/dAdo kinase as well as the dCyd kinase/dAdo kinase? Or could there be another dAdo kinase gene situated with the dCyd kinase gene on the same cistron? The answers to these questions also have to await the cloning of the dCyd kinase gene. To clone the dCyd kinase gene, the ideal cloning probe would be one based on a protein sequence specific to dCyd kinase, such as the nucleoside binding site. However, no dCyd kinase sequences, other than those common to dGuo kinase, are yet available. The dCyd kinase gene does not seem to be located on the same operon as the dGuo kinase gene and dAdo kinase gene, and thus should stand elsewhere in the bacterial genome. Genomic DNA digestion by EcoR I revealed two bands (0.7 Kb and 2.5 Kb) when probed with the PCR 117mer (Figure 25), which is more specific to dGuo kinase but probably still binds to dCyd kinase to some extent. Since the N-terminal portion of the dGuo kinase gene was shown to be in the 0.7 Kb EcoR I fragment, the 2.5 Kb EcoR I fragment would be a good starting point for cloning the dCyd kinase gene. BIBLIOGRAPHY 1. Durham, J.P. and Ives, D.H. (1971) Biochim. Biophys. Acta 228, 9-25. 2. Davis, M.B. and Ives, D.H. (1976)/. Bacterial. 126, 1136-1140. 3. Hoff-Jorgensen, E. (1952) Biochem. J. 50, 400-403. 4. Deibel, M.R.,Jr. and Ives, D.H. (1977) /. Biol Chem. 252, 8235-8239. 5. Deibel, M.R,,Jr. and Ives, D.H. (1978) Methods Enzymol 51, 346-354. 6 . Deibel, M.R.,Jr., Reznik, R.R., and Ives, D.H. (1977) /. Biol Chem. 252, 8240-8244. 7. Chakravarty, R., Ikeda, S., and Ives, D.H. (1984) Biochem. 23, 6235-6240. 8 . Ikeda, S., Chakravarty, R., and Ives, D.H. (1986)/. BioL Chem. 2 6 1 , 15836- 15843. 9. Ikeda, S., Swenson, R.P., and Ives, D.H. (1988) Biochem. 27, 8648-8652. 10. Powell, J.W., and Wachsman, J.T. (1973) AppL Microbiol 25, 869-872. 11. Mollgaard, H. (1980) /. Biol Chem. 255, 8216-8220. 12. Firshein, W., and Hasselbacher, P. (1970) Biochim. Biophys. Acta 204, 60- 81. 13. Neuhard, J. (1968) /. Bacteriol 96, 1519-1527. 14. Tanooka, H., Terano, H., and Otsuka, H. (1971) Biochim. Biophys. Acta 228, 26-37. 15. Rimi, B.K., and Takahashi, L. (1977) /. Bacterial 129, 574-579. 16. Wachsman, J.T., and Morgan, D.D. (1973) Appl Microbiol 25, 506-508. 141 142 17. McElwain, M.C., Chandler, D.K.F., Barile, M.F., Young, T.F., Tryon, V.V., Davis, J.W.,Jr., Petzel, J.P., Chang, C.-J., Williams, M.V., and Pollack, J.D. (1988) Int. J. Stst. BacterioL 38, 417-423. 18. Karlstrom, H. (1970) Eur. J. Biochem. 17, 68-71. 19. Okazaki, R., and Komberg, A. (1964) 7. BioL Chem. 239, 269-274. 20. Okazaki, R., and Komberg, A. (1964) 7. BioL Chem. 239, 275-284. 21. Iwatsuki, N., and Ikazaki, R. (1967) J. MoL BioL 29, 155-165. 22. Voytek, P., Chang, P.K., and Prusoff, W.H. (1971)7. BioL Chem. 246, 1432- 1438. 23. Voytek, P., Chang, P.K., and Prusoff, W.H. (1972)7. BioL Chem. 2 4 7 ,367- 372. 24. Bockamp, E.-O., Blasco, R., and Vinuela, E. (1991) Gene 101, 9-14. 25. Kobayashi, S.-H., Hubbell, H.R., and Orengo, A. (1974) Biochem. 13,4537- 4543. 26. Durham, J.P. and Ives, D.H. (1970) 7. BioL Chem. 245, 2276-2284. 27. Ives, D.H., and Durham, J.P. (1970) 7. BioL Chem. 245, 2285-2294. 28. Gower, W.R., Jr., Carr, M.C., and Ives, D.H. (1979) 7. BioL Chem. 254, 2180-2183. 29. Meyers, M.B., and Kreis, W. (1976) Arch. Biochem. Biophys. 177, 10-15. 30. Samp, J.C., and Fridland, A. (1987) Biochem. 26, 590-597. 31. Cheng, Y.C., Domin, B., and Lee, L.S. (1977) Biochim. Biophys. Acta 481, 481-492. 32. Datta, N.S., Shewach, D.S., Hurley, M.C., Mitchell, B.S., and Fox, I.H. (1989) Biochem. 28, 114-123. 33. Bohman, C., and Eriksson, S. (1988) Biochem. 4258-4265. 34. Kim, M.Y., Ikeda, S., and Ives, D.H. (1988) Biochem. Biophys. Res. Commun. 156, 92-98. 143 35. Kim, M.Y., and Ives, D.H. (1989) Biochem. 28, 9043-9047. 36. Chottiner, E.G., Shewach, D.S., Datta, N.S., Ashcraft, E., Gribbin, D., Ginsburg, D., Fox, I.H., and Mitchell, B.S. (1991) Proc. NatL Acad. ScL USA 88, 1531-1535. 37. Green, FJ., and Lewis, R.A. (1979) Biochem. J. 183, 544-553. 38. Baker, J., and Lewis, R.A. (1981) Biochim. Biophys. Acta 658, 111-123. 39. Yamada, Y., Goto, H., and Ogasawara, N. (1982) Biochim. Biophys. Acta 709, 265-272. 40. Fabianowska-Majewska, K., and Greger, J. (1982) Enzyme 27, 124. 41. Schrecker, A.W. (1970) Cancer Res. 30, 632-641. 42. Yamada, Y., Goto, H., and Ogasawara, N. (1985) Int. J. Biochem. 17, 425- 428. 43. Flemington, E., Bradshaw, H.D.^Jr., Traina-Dorge, V., Slagel, V., and Deininger, L. (1987) Gene 52, 267-277. 44. Arcot, S.S., and Deininger, P.L. (1992) Gene 111, 249-254. 45. Kwoh, TJ., and Engler, J.A. (1984) NucL Acids. Res. 12, 3959-3971. 46. Lewis, J.A., Schimizu, K., and Zipser, D. (1983) Mol Cell Biol 3, 1815- 1823. 47. Wagner, MJ., Sharp, J.A., and Summers, W.A. (1992) Proc. Natl Acad. ScL USA 78, 1441-1445. 48. Swain, M.A., and Galloway, D.A. (1983)/. Virol 46, 1045-1055. 49. Otsuka, H., and Kit, S. (1984) Virology 135, 316-330. 50. Scott, S.D., Ross, N.L.J., and Binns, M.M. (1989) /. Gen. Virol 70, 3055- 3065. 51. Esposito, JJ., Kingsbury, R.C., Spence, A., and Smith, M. (1984) Virology 135, 561-567. 144 52. Balasubramaniam, N.K., Veerisetty, V., and Gentry, G.A. (1990) J. Gen. ViroL 71, 2979-2987. 53. Black, M.E., and Hruby, D.E. (1990)/ BioL Chem. 265, 17584-17592. 54. Black, M.E., and Hruby, D.E. (1992) J. BioL Chem. 267, 6801-6806. 55. Liu, Q., and Summers, W.C. (1988) Virology 163, 638-642. 56. Dube, D.K., Parker, J.D., French,D.C., Cahill, D.S., Dube, S., Horwitz, M.S.Z., Munir, K.M., and Loeb, L.A. (1991) Biochem. 30, 11760-11767. 57. Ikeda, S., Park, I., Gardner, P., and Ives, D.H. (1984) Biochem. 23, 1914- 1921. 58. Ikeda, S., and Ives, D.H. (1985) / BioL Chem. 260, 12659-12664. 59. Bullock, W.O. (1987) BioTechniques 5, 376-378. 60. Clark, J.M., and Switzo, R.L. (1977) Experimental Biochemistry, 2nd ed., Freeman and Co., p. 166-167. 61. Bradford, M. (1978) AnaL Biochem. 72 248-254. 62. Ives, D.H. (1984) AnaL Biochem. 136 416-420. 63. Laemmli, U.K. (1970) Nature 227 680-685. 64. Moos, M., Jr., Nguyen, M.Y. and Liu, T.Y. (1988) / BioL Chem. 263 6005- 6008. 65. Jovin, T.M. (1973) Biochemistry 12 871-879. 66. Jovin, T.M. (1973) Biochemistry 12 879-890. 67. Jovin, T.M. (1973) Ann. N.Y. Acad. ScL 209 474-496. 68. Kyhse-Andersen, J. (1984)/ Biophys. Biochem. Meth. 10 203-209. 69. Matsudaira, P. (1987) / BioL Chem. 262, 10035-10038. 70. LeGender, N., and Matsudaira, P. (1988) BioTechniques 6(2), 154-159. 71. Marmur, J. (1961) / MoL BioL 3, 208. 145 72. Sambrook, J.,Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, second edition, Cold Spring Harbor Laboratory Press. 73. Saiki, R.K, Gelfand, D.H., Stoffel, S., Scharf, S J., Higuchi, R. Horn, G.T., Mullis, KB., and Erlich, H.A. (1988) Science 239 487-491. 74. Southern, E. (1975) /. M ol BioL 98 503. 75. Miller, J.H. (1972) Experiments in Molecular Genetics, pp.431-435, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 76. Kincaid, R.L. and Nightingale, M.S. (1988) BioTechniques 43-48.6 77. Sanger, F., Nicklen, S., and Coulson, A.R. (1977) Proc. Natl Acad. ScL U.SA. 74 5463-5467. 78. Mihovilovic, M. and Lee, J.E. (1989) BioTechniques 7 14-16. 79. Gyllensten, U.B., and Erlich, H.A. (1988) Proc. Natl Acad. Scl USA 85, 7652-7656. 80. McCabe, P.C. (1990) PCR Protocols: A Guide to Methods and Application, edited by Innis, M.A, Gelfand, D.H., Sninsky, J.J. and White, TJ. Academic Press, Inc., p. 76-83. 81. Ikeda, S. and Ives, D.H. (1992) preparation in progress. 82. Hoffmann, P.J., and Blakley, R.L. (1975) Biochem. 14, 4804-4812. 83. Barker, R., Olsen, K.W., Shapa, J.H., and Hill, R.L. (1972) /. Biol Chem. 247, 7135-7147. 84. Ikeda, S., Ma, G.T., and Ives, D.H. (1992) submitted. 85. March, S.C., Parikh, I., and Cuatrecasas, P. (1974) Anal Biochem. 60, 149-152. 86. Saiki, R., Scharf, S., Faloona, F., Mullis, K, Horn, G., Erlich, H., and Arnheim, N. (1985) Science 230, 1350-1354. 87. Mullis, KB., and Faloona, F.A. (1987) Meth. Enzymol 155, 335-350. 88. Ikeda, S., Ma, G.T., and Ives, D.H. (1992) manuscript in preparation. 146 89. Graves, M.C. and Rabinowitz, J.C. (1986) /. BioL Chem. 261 11409-11415. 90. Platt, R. (1986) Ann. Rev. Biochem. 55, 339-372. 91. Pai, E.F., Sachsenheimer, W., Schirmer, R.H., and Schulz, G.E. (1977) J. M ol BioL 114, 37-45. 92. Gentry, G.A. (1985) Proc. Nat. Acad. ScL USA 82, 6815-6819. 93. Egner, U., Tomasselli, A.G., and Schulz, G.E. (1987) J. MoL BioL 195, 649-658. 94. Dreusicle, D., and Schulz, G.E. (1986) FEBS Lett. 208, 301-304. 95. Jurnak, F. (1985) Science 230, 32-36. 96. LaCour, R.F.M., Nyborg, J., Thirup, S., and Clark, B.F.C. (1985) EMBO J. 4, 2385-2388. 97. DeVos, A.M., Tong, L., Milbura, M.V., Matias, P.M., Jancarik, J., Noguchi, S., Nishimura, S., Miura, K., Ohtsuka, E., and Kim, S.-H. (1988) Science 239, 888-893. 98. Pai, E.F., Kabsch, W., Krengel, U., Holmes, K.C., John, J., and Wittinghofer, A. (1989) Nature (London) 341, 209-214. 99. Muller, C., and Schulz, G.E. (1988) J. MoL BioL 202, 909-912. 100. Fry, D.C., Kuby, S.A., and Mildvan, A.S. (1986) Proc. NatL Acad. Set USA 83, 907-911. 101. Tagaya, M., Yagami, T., and Fukui, T. (1987) J. BioL Chem. 262, 8257- 8281. 102. Rao, R., Pagan, J., and Senior, A.E. (1988) J. BioL Chem. 263, 15957- 15963. 103. Dombroski, AJ., Brennan, C.A., Spear, P., and Platt, T. (1988) J. BioL Chem. 263, 18802-18809. 104. Reinstein, J., Schlichting, I., and Wittinghofer, A. (1990) Biochem. 29, 7451-7459. 147 105. Darby, G.f Larder, B.A., and Inglis, M.M. (1986) J. Gen. Virol 67, 753- 758. 106. Mittal, S.K., and Field, HJ. (1989) /. Gen. Virol 70, 901-918. 107. Robertson, G.R., and Whalley, J.M. (1988) Nuc. Acids Res. 16, 11303- 11317. 108. Smith, G.L., De Carlos, A., and Chan, Y.S. (1989) Nuc. Acids Res. 17, 7581-7590. 109. Jong, A.Y.S., Kuo, C.-L., and Campbell, J.L. (1984) J. Biol Chem. 259, 11052-11069. 110. Brenner, S. (1987) Nature (London) 329, 21. 111. Yan, H., Shi, Z., and Tsai, M.-D. (1990) Biochem. 29, 6385-6392. 112. Dreusike, D., Karplus, P.A,, and Schulz, G.E. (1990) J. Mol Biol 9 9359- , 371. 113. Vanderslice, P., Copeland, W. and Robertus, J. (1986) J. Biol Chem. 261, 15186-15191. 114. Copeland, W.C., Domena, J.D. and Robertus, J.D. (1989) Gene 85, 259-265. 115. Bor, Y.-C., Oraes, C., Lee, S.-P., Crosby, W.L., Sinskey, A.J. and Batt, C.A. (1992) Gene 114, 127-131. 116. Duwat, P., Ehrlich, S.D. and Gruss, A. (1992) /. Bacteriol 174, 5171-5175. 117. Natori, Y., Kano, Y. and Imamoto, F. (1990) /. Biochem. 107, 248-255. 118. Taguchi, H. and Ohta, T. (1992) Unpublished. 119. Toy, J. and Bognar, A.L. (1990) J. BioL Chem. 265, 2492-2499. 120. Andrews, J., Clore, G.M., Davies, R.W., Gronenbom, A.M., Gronenbom, B., Kalderon, D., Papadopoulos, P.C., Schaefer, S., Sims, P.F. and Stancombe, R. (1985) Gene 35, 217-222. 121. Pinter, K., Davisson, VJ. and Santi, D.V. (1988) DNA 7, 235-241. 148 122. Alpert, C.-A. and Chassy, B.M. (1988) Gene 62, 277-288. 123. Porter, E.V. and Chassy, B.M. (1988) Gene 62, 263-276. 124. Lerch, H.-P., Bloecker, H., Kallwass, H., Hoppe, J., Tsai, H. and Collins, J. (1989) Gene 78, 47-57. 125. Kim, S.F., Baek, SJ. and Pack, M.Y. (1991) Appl Environ. MicrobioL 57, 2413-2417. 126. Alpert, C.-A. and Chassy, B.M. (1990) J. BioL Chem. 265, 22561-22568. 127. Heaton, M.P. and Neuhaus, F.C. (1992) J. BacterioL 174, 4707-4717. 128. Kim, S.F., Baek, S.J. and Pack, M.Y. (1991) Appl Environ. MicrobioL 57, 2413-2417. 129. Shimizu-Kadota, M., Kiwaki, M., Hirokawa, H. and Tsuchida, N. (1985) MoL Gen. Genet. 200, 193-198. 130. Klugbauer, N., Ludwig, W., Bauerlein, E. and Schleifer, K.H. (1992) Syst. AppL MicrobioL 15, 323-330. 131. Lerch, H.-P., Frank, R. and Collins, J. (1989) Gene 83, 263-270. 132. Ishino, Y., Morgenthaler, P., Hottinger, H. and Soil, D. (1992) AppL Environ. MicrobioL 58, 3165-3169. 133. Schmidt, B.F., Adams, R.M., Requadt, C., Power, S. and Mainzer, S.E. (1989) J. BacterioL 171, 625-635. 134. Leong-Morgenthaler, P.M., Zwahlen, M.C. and Hottinger, H. (1991) J. BacterioL 173, 1951-1957. 135. Kochhar, S., Hunziker, P.E., Leong-Morgenthaler, P. and Hottinger, H. (1992) J. BioL Chem. 267, 8499-8513. 136. Delcour, J. (1991) Unpublished. 137. Leong-Morgenthaler, P.M., Oliver, S. and Soli, D. (1991) Unpublished. 138. Suzuki, K., Takahashi, M., Imamura, S. and Ishikawa, T. (1992) Unpublished. 149 139. Gallo, K.A (1992) Unpublished. 140. Joerger, M. and Klaenhammer, T.R. (1990) J. BacterioL 172, 6339-6347. 141. Duwat, P., Ehrlich, S.D. and Gruss, A. (1992) /. BacterioL 174,5171-5175. 142. Mollet, B. and Pilloud, N. (1991) /. BacterioL 173, 4464-4473. 143. Kochhar, S., Hottinger, H., Chuard, N., Taylor, P.G., Atkinson, T., Scawen, M.D. and Nicholls, DJ. (1992) Ear. J. Biochem. 208, 799-805. 144. Josson, K., Soetaert, P., Michiels, F., Joos, H. and Mahillon, J. (1990) /. BacterioL 172, 3089-3099. 145. van Rooijen, RJ. and de Vos, W.M. (1990) J. BioL Chem. 265, 18499-18503. 146. Hoick, A. and Naes, H. (1992) J. Gen. MicrobioL 138, 1353-1364. 147. Lokraan, B.C., van Santen, P., Verdoes, J.C., Kruse, J., Leer, R J., Posno, M. and Pouwels, P.H. (1991) MoL Gen. Genet. 230, 161-169. 148. Leer, RJ., van Luijk, N., Posno, M., and Pouwels, P.H. (1992) MoL Gen. Genet. 234, 265-274. 149. Leer, RJ., Van Luijk, N., Posno, M., and Pouwels, P.H. (1991) Unpublished. 150. Ohta, T., and Taguchi, H. (1991) J. BioL Chem. 266, 12588-12594. 151. Ohta, T., and Taguchi, H. (1991) J. BioL Chem. 266, 12588-12594. 152. Skaugen, M. (1989) Plasmid 22, 175-179. 153. Bouia, A., Bringel, F., Frey, L., Kammerer, B., Belarbi, A, Guyonvarch, A, and Hubert, J.-C. (1989) Plasmid 22, 185-192. 154. Leer,R.J., Christiaens,H., Peters,L., Posno,M., Verstraete,W. and Pouwels,P.H. (1992) Unpublished. 155. Taguchi, H., and Ohta, T. (1991) I BioL Chem. 266, 12588-12594. 156. Taguchi, H., and Ohta, T. (1991) J. BioL Chem. 266, 12588-12594. 157. Connell,H., Tannock,G.W., Luchansky,J.B., Mercer,A.A., Miller,L.A and Klaenhammer,T.R. (1991) Unpublished. 158. Knauf, H.J., Vogel, R.F., and Hammes, W.P. (1992) Appl Environ. Microbiol 58, 832-839.