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 comer 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 0201690
Membrane assembly and function Escherichiaof coli leader peptidase
Lee, Jong-In, Ph.D. The Ohio State University, 1991
UMI 300 N. Zeeb Rd. Ann Arbor, MI 48106
MEMBRANE ASSEMBLY AND FUNCTION OF ESCHERICHIA COLI LEADER PEPTIDASE
DISSERTATION
Presented in Partial Fulfillment of the Requirement for the Degree of Doctor of Philosophy in the Graduate School of The Ohio State University
By
JONG-IN LEE, M.S.
*****
The Ohio State University
1991
Dissertation Committee Approved by
Ross E. Dalbey
David H. Ives
James A. Cowan Advisor Department of Chemistry To my parents, wife, and all who supported me, in Lord Jesus Christ ACKNOWLEDGEMENTS
I express sincere appreciation to Dr. Ross Dalbey for his kind supports and guidence throughout the research. His advice and technical training were of great help.
I also give thanks to my collegues, Bill, Heng-Yi, Liming, Meesook and Shiyuan, and people in laboratories of Drs. Cowan, King, Klapper, Marshall and Tsai for their friendship, helpful discussions and sharing instruments, and appreciate Drs. Kuhn and von Heijn for their coorperation.
To my family, I always owe debt of love full of sacrifices. Above all, glory goes to Lord Jesus Christ. January 7,1960 Born in Seoul Korea
February, 1982 ...... B.S. in Chemistry, Seoul National University Seoul, Korea
February, 1984 ...... M.S. in Biochemistry Seoul National Univeraity Seoul, Korea
September, 1986- Present ...... Graduate Research Associate or Teaching Associate, Department of Chemistry, The Ohio State University.
PUBLICATIONS
1. Lee.J.I. and Yang,C.H. (1984), "Purification and Characterization of Pvull, a Restriction Endonuclease", Korean Biochem. J., 17, 357-364.
2. Bilgin.N., Lee.J.I., Zhu,H.Y., Dalbey,R.E. and von Heijne.G. (1990), "Mapping of Catalytically Important Domains in Escherichia coli Leader Peptidase", EMBO J., 9, 2717-2722.
Field of Study
Major Field: Chemistry Emphasis in Molecular and Cell Biology, Professor Ross E. Dalbey, Advisor. TABLE OF CONTENTS
PAGE DEDICATION i i
ACKNOWLEDGEMENTS...... i i i VITA...... iv
LIST of TABLES...... v ii LIST of FIGURES...... v iii LIST of ABBREVIATIONS x i
CHAPTER
I. INTRODUCTION...... 1
II. MATERIALS AND METHODS...... 16
2.1. Materials ...... 16 2.2. Bacterial Strains and Plasmids ...... 1 7 2.3. Preparation of Right-side-out Inner Membrane Vesicle ...... 1 8 2.4. Preparation of Spheroplasts ...... 20 2.5. Protease Mapping ...... 20 2.6. Preparation of Inverted inner membrane vesicles ...... 21 2.7. Preparation of S-100 ...... 21 2.8. Purification of Ribosome ...... 22 2.9. In Vitro Protein Transocation ...... 22 2.10. Preparation of Cell Lysate ...... 24 2.11. In Vitro Processing of Procoat ...... 24 2.12. Immunoprecipitation ...... 26 2.13. Immunoblotting ...... 27 2.14. SDS-PAGE and Fluorography ...... 27 2.15. Agarose gel electrophoresis ...... 28 2.16. Site-Directed Mutagenesis ...... 28 2.17. Transformation and transfection ...... 33
v 2.18. Preparation of Plasmid and RF D N A ...... 33 2.19. DNA Sequencing ...... 35 2.20. Subcloning ...... 37 2.20. Special Preparation of Solutions ...... 38 2.21. Construction of Pf3-Leader peptidase and Pf3-H1 ...... 40
III. In Vitro Translocation Studies of E. Coli Leader Peptidase ...... 43
3.1. Introduction ...... 43 3.2 Results ...... 45 3.3. Discussion ...... 52
IV. Distinct Domains of an Oligotopic Protein (Leader Peptidase) is Sec-dependent and Sec-independent 56
4.1. Introduction ...... 56 4.2. Results ...... 59 4.3. Discussion ...... 77
V. Mapping of Catalytically Important Domains in E. coli Leader Peptidase ...... 83
5.1. Introduction ...... 83 5.2. Results ...... 85 5.3. Discussion ...... 92
VI. Site-Directed Mutagenesis to Define the Limits of Sequence Variation Tolerated for Processing of the M13 Procoat Protein by the Escherichia coli Leader Peptidase ...... 96
6.1. Introduction ...... 96 6.2. Results ...... 99 6.3. Discussion ...... 110
References 124 LIST OF TABLES
TABLES
5-1 Catalytic Activity of Leader peptidase mutants ...... 87
6-1 List of Mutagenic Primers ...... 116
6-2 In vivo and in vitro processing of procoat with alterations at -1 ...... 117
6-3 In vivo and in vitro processing of procoat with an alteration at -3 ...... 118
6-4 In vivo and in vitro processing of procoat with an alteration at -6 ...... 119
6-5 In vivo and in vitro processing of procoat with an alteration a t+1 ...... 120
6-6 Processing of Procoat with an alteration at -2 ...... 121
6 -7 Processing of Procoat with an alteration at -4 ...... 122
6-8 Processing of Procoat with an alteration at -5 ...... 123
v i i LIST OF FIGURES
FIGURES
1-1 Representation of protein export to various compartments in an eukaryotic cell ...... 2
1 -2 Compartments of E. coli for protein export ...... 5
1-3 Possible transmembrane orientation of membrane proteins ...... 6
1-4 Integration segments within membrane proteins ...... 7
1-5 Sec-dependent and sec-independent pathways of membrane insertion of bacterial proteins ...... 1 0
1-6 Model of the membrane orientation of three proteins of the plasma membrane of E. coli ...... 11
1-7 Removal of the leader peptide of preprotein by leader peptidase during protein export in E. coli ...... 1 3
2-1 Subcloning of the Lep gene from M13mp8 into pING plasmid ...... 1 9
2 - 2 Genomic structure of pT712 vector ...... 25
2 -3 Construction of Pf3-leader peptidase ...... 41
3-1 In vitro post-translational translocation of pro-OmpA ...... 47
3-2 Urea dilution study of post-translational translocation of leader peptidase ...... 48
3-3 Urea dilution study of post-translational translocation of leader peptidase ...... 50 vi i i 3-4 Urea dilution study of post-translational translocation of leader peptidase ...... 51
3-5 Effects of antibodies to chapernones on the co- translational translocation of leader peptidase ...... 53
4-1 Topology of Pf3 coat protein, leader peptidase and Pf3-leader peptidase ...... 58
4-2 Pf3-leader peptidase is recognized by antibody to the Pf3 coat protein and the leader peptidase antibody ...... 62
4-3 Proteolysis of inner membrane vesicles - Total [35S]-labeled membrane proteins ...... 63
4-4 Immunoprecipitation of proteolysed inner membrane vesicles with antiserum to P f3 ...... 66
4 -5 Hybrid proteins between Pf3 peptide and leader peptidase ...... 67
4 -6 Protease mapping of Pf3-H1 ...... 68
4-7 Effects of SecA inactivation on the Insertion of the amino terminus of Pf3-leader peptidase ...... 71
4-8 Effects of membrane potential disruption on the insertion of the amino terminus of Pf3-leader peptidase ...... 74
4-9 Effects of positive charges on the translocation of Pf3- leader peptidase - Amino-terminal insertion in a SecA temperature-sensitive strain ...... 76
4-10 Effects of positive charges on the translocation of Pf3- leader peptidase - Translocation of the carboxyl-terminal domain in MC1061 ...... 78
i x 4-11 Working model of the membrane assembly of leader peptidase ...... 81
5-1 Domain structure of leader peptidase ...... 84
5-2 In vitro processing of pro-OmpA in strain IT41 at the non-permissive temperature with and without the pRD8 plasmid (expressing wild-type leader peptidase) ...... 86
5 -3 In vitro processing of phage M13 procoat protein in crude extracts of E. coli MC1061 earring plasmids expressing different Lep mutants ...... 91
5-4 Immunoblotting of MC1061 extracts containing different Lep mutants ...... 93
6-1 The sequence of wild-type M13 procoat and the 59 mutations created at positions +1, -1, -2, -3, -4, -5 and -6 near the cleavage site ...... 100
6 -2 In vivo processing of -1 and -3 procoat mutants by leader peptidase ...... 102
6-3 Representative results of in vitro processing of -1 and -3 procoat mutants by leader peptidase ...... 104
6 -4 Protease mapping o f -6 procoat mutants ...... 105
6-5 In vitro leader peptidase titration study of -6 procoat mutants ...... 107
6 -6 Representative results of In vitro processing of -2, -4 and -5 procoat mutants by leader peptidase ...... 109
x LIST OF ABBREVIATIONS
Apr ampicillin resistance ATP adenosine 5'-triphosphate bp base pair(s) CCCP Carbonylcyanide-m-chlorophenyl hydrazone d/ dATPaS 2'-deoxyadenosine 5'-[&- thio]triphosphate dCTP 2'-deoxycytosine 5'-triphosphate dGTP 2'-deoxyguanosine 5'-triphosphate DTT dithiothreitol dTTP 2'-deoxythymine 5'-triphosphate EDTA ethylenediamine tetraacetic acid EtBr ethidium bromide GuHCI guanidinium hydrochloride IPTG isopropyl-b-D-thiogalactopyranoside kb 1000-base pairs kDal 1000 Daltons LB Luria broth LPase Leader peptidase NADH nicotinamide adenine dinucleotide hydrogen OmpA outer membrane protein A PAGE polyacrylamide gel electrophoresis PMSF phenylmethylsulfonyl fluoride RF double-stranded replicative form SB sample buffer SDS sodium dodecyl sulfate Sec secretion Tris 2-amino-2-(hydroxymethyl)-1,3 propanediol
x i CHAPTER I
INTRODUCTION
1.1. Protein Export Cells have multiple compartments separated by lipid-rich membranes. Each compartment has a unique set of proteins and a specific biological function. These proteins must be exported to their compartments after they are synthesized in cytoplasm (Fig. 1- 1). Some proteins are assembled into a specific membrane, which will be further discussed in the following section. Others translocate across one or even several membranes to reach their final destination.
Eukaryotic cells have more compartments than prokaryotic cells due to many subcellular orgenelles, which include the nucleus, mitochondria, chloroplast, endoplasmic reticulum, golgi complex, lysosome, etc. Each of these organelles is divided into subcompartments by various membranes. On the other hand, prokaryotic cells have no subcellular organelle, and the aqueous compartments are separated only by inner and outer membranes. The endoplasmic reticulum (ER), the peroxisomal membrane, the inner
1 Lysosomes n vacuoles Mitochondrion
Peroxisome Golgi complex, Chloroplast
Precursor •o- polypeptides
Endoplasmic mRNA reticulum
Nucleus
Figure 1-1. Representation of protein export to various compartments in an eukaryotic cell. membrane of mitochondria, the inner and thylakoid membranes of chloroplasts and the bacterial plasma membrane can export many different proteins (Wickner et al., 1985). Among these membranes only mitochondrial inner membrane can transport proteins in both directions. The smooth ER, the Golgi complex, secretory vesicles, endosomes, lysosomes, the eukaryotic plasma membrane, and perhaps the nuclear inner membrane can translocate only a few special proteins. It is not known whether the outer membranes of mitochondria, chloroplasts, and Gram-negative bacteria can translocate proteins by themselves.
Most exported proteins in both prokaryotes and eukaryotes are synthesized as precursors with N-terminal leader peptide of 15-20 amino acids. Leader peptides interact with export machinery to initiate the protein translocation (Blobel, 1980; Walter et al., 1984,1986). These peptides usually have a positively charged N- terminus, and invariably have a hydrophobic core followed by a cleavage site for the leader peptidase (von Heijne, 1985, 1988; Watson, 1984). However, the primary sequences of leader peptides are very different from one another. After the preprotein translocates across the membrane, the leader peptide is removed by a leader peptidase .
Prokaryotes have advantages over eukaryotes since they are amenable to biochemical and genetic analysis. In Escherichia coli. a gram-negative bacteria, proteins are exported to four non-cytosolic compartments (Fig. 1-2): the inner membrane, periplasm, outer membrane and extracellular medium. The E. coli protein export machinery is composed of many proteins, which at least include (1) chaperonins such as SecB, GroEL and Trigger factor, (2) SecA, a peripheral membrane protein, (3) SecY, SecD and SecE, integral membrane proteins and (4) leader peptidase. Energy is also involved in protein export in the form of ATP and protein motive force.
1.2. Topogenesis of Membrane Proteins Membranes have a variety components which determine their biological functions within the cell. These proteins can be classified according to their topologies within the membrane as follows (Fig. 1-3): (1) Bitopic proteins are proteins that span the membrane once with their carboxy termini exposed on either the cytoplasmic (type 1) or extracytoplasmic surface (type II). (2) Oligotopic proteins span the membrane twice. Both their amino and carboxy terminus of these proteins are on the same side of the membrane. (3) Polytopic membrane proteins span the membrane multiple times and have more complex structures.
Apolar integration sequences specify the information for the orientation of these proteins within the membrane. Four of these sequences have been characterized thus far (Fig. 1-4): (1) Cleavable amino-terminal leader sequences mediate the protein translocation 5
Outer membrane ■> Periplasm
Inner membrane
Figure 1-2. Compartments of E. coli for protein export. Type II
N C r s A C
Membrane
N C V N V W
• V------y------/ V. y J Oligotopie Polytopic Cytoplasm Bitopic
Figure 1-3. Possible transmembrane orientation of membrane proteins rnbranen P Poro^e'nS' me v/tth'n rat'on■ ^ sey cLeQrnen*s \nteg Figure 1 - 4 . 8 and is cleaved by leader peptidase (Blobel, 1980; Wickner and Lodish, 1985). (2) Uncleaved signal sequences also mediate protein translocation, but remains in the mature protein serving as a transmembrane anchor (Blobel, 1980). (3) Stop-transfer domains function to terminate the translocation of the polypeptide chain across the membrane by anchoring the protein to the membrane (Davis and Model, 1985). (4) Insertion domains which are composed of two hydrophobic regions separated by a polar region spontaneously insert across the membrane (Kuhn £ la l, 1986). Both uncleaved signals and leader peptides contain a polar and a basic amino terminus, followed by a hydrophobic domain (von Heijne, 1983; Perlman and Halvorson, 1983; Zerial fita l-, 1986; Spiess and Lodish, 1986). While uncleaved signals each have their amino terminus cytoplasmic and their carboxy terminus periplasmic, stop- transfer peptides have the reverse orientation. Hydrophobic character is important for the function of these sequences.
Proteins can be inserted into the membrane by several different pathways. E. coli leader peptidase and M13 coat protein are good examples. These two proteins are different in that the former does not have a cleavable leader peptide but the latter does. They also have different requirements for their insertion into the membrane. Leader peptidase requires SecA, SecY (Wolfe e la l, 1985) and the membrane electrochemical potential (Wolfe and Wickner, 1984) for translocation of its periplasmic large carboxyl- terminal domain. M13 coat protein precursor, on the other hand, does not require Sec proteins for its transient potential-dependent insertion across the inner membrane (Wolfe a ia L , 1985). Coat protein, with its N-terminus facing the periplasm, is produced upon removal of leader peptide by leader peptidase. Fig. 1-5 shows sec- dependent and sec-independent pathways of membrane insertion of bacterial proteins.
Positive charges are considered to play an important role in determining the topology of membrane proteins (Dalbey, 1990). Strong correlation has been found in bacterial membrane proteins between the orientation of the transmembrane segment and the positive charge distribution around the segment. The cytoplasmic side of the segment usually has more positive charges than the periplasmic side (Fig. 1-6). The positive charges at the amino- terminal end of leader peptides are necessary for the efficient protein translocation (Inouye et al.. 1982; lino et al.. 1987; Yamane and Mizushima, 1988). Manipulation of positive charge distribution around the transmembrane segment can affect its orientation (Szczesna-Skopupa a ia l, 1988). When positively charged residues are introduced after leader peptides, translocation of the protein was prevented (Li a ia l, 1988; Yamane and Mizushima, 1988; H. Y. Zhu, A. Kuhn and R. Dalbey, manuscript in preparation). Similar effects also have been found with leader peptidase, which lacks a leader peptide. "Translocation poison" sequence in E. coli leader Pathway 1: Sec gene-dependent
Signal Periplasm
Sec Y ATP
Plasma SecA membrane
(Chao Cytoplasm
Pathway 2: Sec gene-independent
Plasma ATP membrane
Ihapl - insertion domain
Figure 1-5. Sec-dependent and sec-independent pathways of membrane insertion of bacterial proteins. Hydrophoibic segments are indicated by dark rectangles. Chap refers to melecular chaperonins which bind to newly synthesized proteins to prevent incorrect folding of hydrophobic domains in the hydrophilic environment of the cytoplasm. MalF Cytoplasm Leader peptidase
Figure 1-6. Model of the membrane orientation of three proteins of the plasma membrane of E. coli. 1 2 peptidase prevents the preceding apolar domain from functioning as a signal peptide (von Heijne fila l. 1988) due to the positively charged residues within the polar region (Laws e ia L , 1989).
E. coli leader peptidase provides a good model system to study the assembly of membrane proteins. This protein has an uncleaved signal sequence (Dalbey and Wickner, 1986; Dalbey and Wickner, 1987; Dalbey et al.. 1987) and a translocation poison domain (von Heijne fila l, 1988; Laws and Dalbey, 1989; San Millan fital-. 1989). Using protease mapping studies, it has been shown that leader peptidase spans the inner membrane twice with a large carboxyl-domain exposed to the periplasm and its amino terminus presumably also facing periplasm, (Wolfe sla l-, 1983; Moore s la i-, 1987).
1.3. E. coli Leader Peptidase Leader Peptidase is one of central enzymes in protein export machinery. This enzyme functions to remove the N-terminal leader peptides after the preprotein has crossed the membrane (Fig. 1-7) to release exported proteins into the periplasm or to propel them on to the outer membrane of the bacteria.
Two processing enzymes, lipoprotein signal peptidase and leader peptidase, have been identified in the E. coli inner membrane. While leader peptidase cleaves in vivo the precursor to the M13 coat 1 3
Cytoplasm
Inner Membrane
Leader Peptidase Preprotein Periplasm
-6 -5 -4 -3 -2 -1 +1 —Pro — X — X — Ala —• X — Ala — X
Cleavage site
Figure 1-7. Removal of leader peptide of preprotein by leader peptidase during the protein export in E. coli. 1 4 protein (inner membrane), maltose binding protein (periplasm), and the outer membrane protein A (outer membrane) (Dalbey and Wickner, 1985), lipoprotein signal peptidase cleaves only lipoproteins (Tokunaga e ta l, 1982). Both of them consists of 1 polypeptide chain with a molecular mass of 18Kd for lipoprotein signal peptidase and 36 KDa for leader peptidase.
Wickner and colleagues have made a major contribution to the study of leader peptidase. They have purified this enzyme using in vitro assay system (Wolfe sla l-, 1982). The gene was then cloned and sequenced (Date s la t, 1981), the gene-encoded protein overexpressed under arabinose regulation (Dalbey and Wickner, 1985), and purified in large quantities. Gene cloning and mutagenesis techniques were also employed to ascertain the physiological role (Dalbey and Wickner, 1985) and membrane biogenesis (Wolfe and Wickner, 1984; Dalbey and Wickner, 1986; Dalbey and Wickner, 1987) of this enzyme. The active site of this enzyme is thought to be located in the periplasmic C-terminal domain since deletions in this region abolish the enzymatic activity.
Even though leader peptidase can cleave a wide variety of pre-proteins (Watts elal-, 1983) including bacterial exported proteins, yeast pre-acid phosphatase, honeybee pre-promellitin, and human pre-hormones, there is no common sequence in these substrate preproteins around their cleavage sites (von Heijne, 1983). Their cleavage sites, however, do have a common feature: -1 and -3 position with respect to the cleavage sites have small amino acids. This feature alone is not sufficient to explain the accuracy of leader peptidase cleavage since leader peptidase cleaves the target sequence of only preproteins. The target sequence, along with the two membrane spanning regions of leader peptidase, may define the accuracy of leader peptidase cleavage. Thus leader peptidase seems to recognize certain features of the preprotein rather than a specific sequence. CHAPTER II
MATERIALS AND METHODS
2.1. Materials All salts, amino acids, phenylmethylsulfonyl fluoride, dithiothreitol, phospho(enol)pyruvate, putrescine, folinic acid, 0- nicotinamide adenine dinuclotide reduced form (NADH), hen egg white lysozyme (54,000 units/mG) and spermidine were from Sigma.
Nucleotides (dATP.dTTP, dGTP and dCTP), pT712 vector, T7 RNA polymerase, SDS, bromophenol blue, acrylamide, ammonium persulfate and ultra pure TEMED were obtained from BRL.
Trypsin (L-1-tosylamido 2-phenylethyl chloromethylketone treated) and soybean trypsin inhibitor were purchased from Worthington. DNA polymerase I (Klenow fragment) and proteinase K were from Boehringer Mannheim. All restriction enzymes were from Bethesda Research Laboratories and New England Biolabs. DNA restriction enzymes were from Promega, Bethesda Research Laboratories, and Pharmacia.
Trans 35S-label, a mixture of 85% [35S ] methionine and 15% [35S ] cysteine, 1000 Ci/mmol were from ICN and K&K Laboratories 16 1 7
Inc. (Cleveland, OH). [35S]-dATP was obtained from New England Nuclear.
Yeast extract, bacto-agar, bacto-tryptone, fructose and glucose were purchased from Difco Laboratories. Sucrose was from ICN Biomedicals. Dioxane-free IPTG was from Research Organics.
Urea and sodium sulfite (reagent grade) were from Aldrich Chemical Company. N,N'-methylene-bis-acrylamide and Coomassie brilliant blue G-250 were purchased from Bio-Rad. Pansorbin was from Calbiochem. All oligonucleotides for site-directed mutagenesis were from the Biochemical Instrument Center at the Ohio State University.
2.2. Bacterial Strains and Plasmids E. coli strains HJM114 (Alacpro) F'(lac pro), MC1061 (AlacX74, araD139, A (ara, leu)7697, gal U, galK, hsr, hsm, strA), JM103 (Alacpro), thi, strA, supE, endA, sbcB, hsdR, traD36, proAB, lac I9Z M15) and RZ1032 were from our collection. CJ105 (HJM114, SecAts, Ieu82 :: Tn10) and CJ 107 (HJM114, SecYts^ were obtained from Dr. William Wickner. E. coli strain IT41, containing a temperature-sensitive amber mutation (lep-9) in the lep gene is described in Inada et al, (1989).
pING plasmid containing the araB promoter and the arabinose regulatory elements were described by Johnston et al., 1985. Cloning of the lep gene into the pING to produce wild-type plasmid pRD8 (Fig.2-1) were described by Dalbey and Wickner (1985). The pT712 plasmid containing the procoat gene under the control of the T7 promoter was described by Bilgin e la i- (1990).
2.3. Preparation of Right-side-out Inner Membrane V e s ic le MC1061 cells bearing plasmids encoding wildtype leader peptidase or pf3-leader peptidase were grown until mid-log phase in 1 liter of M9 minimal media containing 0.5% fructose and 19 amino acids (except methionine). Plasmid-encoded proteins were induced by arabinose for 1 hour and labeled with 35S-Methionine for 2min. Cells were harvested by centrifuging at 10,000 rpm for 5 min at 0 - 4°C and resuspended in 0.75 M sucrose-10 mM Tris, pH 7.8. Lysozyme (2mg/ml) was added to final concentration of 100 pg/ml and the sample was incubated on ice for 2 min followed by a dilution with 2 volume of cold 1.5 mM EDTA, pH 7.5, which was added at a constant rate over a period of 8 - 10 min in ice bath. Then cells were sonicated for 4 - 10 min and centrifuged for 20 min at 1200g at 0 - 4°C. The supernatant was again centrifuged for 2 hour at 360,000g at 2°C, and the pellet was suspended in 1 - 6 ml of 25% sucrose(w/w), 5mM EDTA, pH 7.5. The sample was then loaded on a stepwise gradient of 2 ml each of 30%, 35%, 40%, 45% and 50%, and 0.5 ml of 55% sucrose and 5mM EDTA, pH 7.5. Centrifugation was 19
Bam HI BomHI I AmP^i-'\Lep
I pT0l25 J V. ^/^Bglll
BomHI BamHI Isolate linear MI3mp8 Bqlll Isolate lep T4Ligase Lep Sma I
M13mp8Lep
Sma I Sma I Sail S ail . Isolate Lep fragment i r Isolate pING-1 T4Ligase
ara B1
ara C EcoRI Isolate large pRD8 fragment T4 Ligase Lep
Figure 2-1. Subcloning of Lep gene from M13mp8 into pING plasmid. 20 carried out at 38,000 rpm using a Beckman SW 40 rotor and the inner membrane fraction was pooled out using a syringe. The fraction was diluted with 9 volumes of 10 mM Tris-HCI, pH 7.6, 50mM KCI, 10 mM Mg(CH3COO)2 and 1mM DTT and centrifuged at 42,000 rpm for 1.5 hours using a Ti 50 rotor. The pellet was resuspended in a appropriate volume of the above buffer.
2.4. Preparation of Spheroplasts Spheroplasts were prepared as described (Randall and Hardy, 1986). E. coli strains bearing the plasmids encoding Pf3-leader peptidase, Pf3-H1, RR, RRA, A, or various procoat mutants were grown to the early log phase in M9 minimal media containing 0.5% fructose. Plasmid-encoded proteins were induced with arabinose (IPTG for procoat mutants) and labeled with [35s]-methionine for 2 min. After pulse-labeling, 1ml of the cells were chilled on ice, centrifuged, and resuspended in 250pl of 100mM Tris acetate, pH 8.2, 50 mM sucrose and 5mM EDTA. 20pl of 2mg/ml lysozyme was added and immediately mixed with 250pl of cold H20. After 5 min of incubation on ice, 50 jil of 0.2 M MgS04 was added to stabilize the sphroplasts, which was then centrifuged. The pellet was resuspended in 500 pi of 50 mM Tris-acetate, pH 8.2, 0.25 M sucrose and 10 mM MgS04
2.5. Protease Mapping Aliquots of spheroplasts or right-side-out inner membrane 21 vesicles were incubated for 1 hour either with or without proteinase K or trypsin (1 mg/ml at OoC for spheroplasts, 0.2 mg/ml at 370C for membrane vesicles). Where indicated, a portion of the aliquots were treated with Triton X-100 prior to the addition of proteases to lyse the cells. Subsequently, PMSF (5 mM final concentration) or soybean trypsin inhibitor was added to inhibit the proteases.
2.6. Preparation of Inverted Inner Membrane Vesicle E. coli strain D10 was grown in TYE media to the early mid-log phase, harvested, and resuspended in Buffer A (50 mM triethanolamine acetate. pH 7.5, 250 mM sucrose, 1 mM EDTA, 1 mM DTT and 0.5 mM PMSF). The suspension was passed through a French press three times at 8,000 psi with chilling on ice between presses, and centrifuged for 2 hours at 15,000 X g. The pelleted cells were then resuspended in buffer B (buffer A without EDTA and PMSF) using a Dounce homogenizer, and again centrifuged at 12,000 X g. The supernatant was subjected to a sucrose step gradient centrifugation (Schnaitman, 1970). The pelleted inverted inner membranes were resuspended and stored at -80°C.
2.7. Preparation of S-100 S-100 was prepared from 80 g of E. coli D10 paste. Cells were broken by blending with glass beads in 80 ml of 22 mM ammonium acetate, 10 mM TRIS-CI pH 7.5, 10 mM magnesium acetate and 1 mM DTT using waring blendor. The sample was centrifuged in a JA 20 rotor for 45 min at 15,000 rpm at 0°C, and the supernatant was again subjected to ultracentrifugation in a Ti 60 rotor at 56,000 rpm for 1 hour. The supernatant was loaded on a DE52 column pre equilibrated with the above buffer, and the fractions displaying yellow color were collected.
2.8. Purification of Ribosomes 80 g of D10 nuggets were thawed without warming and ground up using acid washed glass beads and waring blendor. Buffer A (22 mM ammonium acetate, 10 mM Tris-CI pH 7.5, 10 mM magnesium acetate and 1 mM DTT) was added up to 320 ml, and the sample was centrifuged for 30 min at 20,000 rpm using JA20 rotor. The supernatant was subjected to ultracentrifugation at 56,000 rpm for 1 hour at 0°C, and the pellet was homogenized in 5 ml of the buffer A. The suspension was then subjected to isopycnic ultracentrifugation through a 5-20% sucrose gradient in Beckman SW 27 rotor. The diffused blue ribosome band was colleced using syringe and stored at -70°C.
2.9. In Vitro Protein Translocation For post-transloational translocation, leader peptidase and pro-OmpA were synthesized in vitro in a mixture (25 pi) of 10 pCi of [35S] methionine, 2.2 mM ATP, 0.56 mM each of UTP, CTP and GTP, 30 mM creatine phosphate, 2.5 pg each of NADP, FAD, and folinic acid, 1.4 mM DTT, 2.5 pg of E. coli tRNA mixture, 0.35 mM each of 19 amino acids, 7 mM calcium acetate, 72 mM potassium acetate, 40 mM Tris-acetate, pH 8.0, 33 mM ammonium acetate, 13 mM magnesium acetate, 0.8 mM spermidine, S-100, purified ribosome, and 1 pg of plasmid DNA. The mixture was incubated for 60 min for the protein synthesis, and thesalts were removed by passing through a Sephadex G-25 column. The fractions containing the radioactively-labeled proteins were combined and an aliquot (50 pi) was added to the post-translational translocation system, which is composed of 5 pi of 0.5 M KCI, 1 pi of 66 mM ATP, 1 pi of freshly prepared 330 mM NADH in 20 mM Tris-CI, pH 8.0, 4 pi of polyamine mix (15 pi of 1 M spermidine- HCI and 9.3 mg of putricine-HCI mixed with 85 pl of water), and 1 pi of inverted inner membrane vesicle. The translocation was continued for 20 min at 37°C, and a portion (50 pl) of the mixture was subjected to protease mapping by incubating with 5.5 pl of 5 mg/ml proteinase K for 13 min on ice. The digestion was terminated by incubating with 11 pl of saturated PMSF for 20 min on ice and then adding 16 pl of sample buffer. Samples were then analyzed by SDS-PAGE and fluorography.
For urea dilution studies, urea was added to the translation mixture (after the desalting on Sephadex G-25) to a final concentration of 8M. Leader peptidase in 8 M urea was added to the post-translational translocation system so that urea concentration in the system was 0.8 M. 24
For co-translational translocation, 1 jil of inverted inner membrane vesicles were added to the translation mixture (30 pl) containing trans [35s]-methionine (30 |iCi), T7 RNA polymerase, Triton X-100 (0.4%) and S150-2, and 1 pg plasmid. After 30 min at 37°C, proteinase mapping was earned out as described above with proteinase K, and samples were analyzed by SDS-PAGE and fluorography.
2.10. Preparation of Cell Lysate E. coli MC1061 cells bearing a plasmid encoding leader peptidase or its mutants were grown on M9 minimal media containing fructose and amino acids to the mid-log phase, and the plasmid-encoded proteins were induced by arabinose. The cells were mixed with equal amount of lysis buffer (10mM EDTA, 10 mM Tris, pH 8.0, 20% sucrose, 1% triton X-100, 1 mg/ml fresh lysozyme, 5pg/ml DNase and 1pg/ml RNase) and incubated for 30 min at room temperature to lyse the cells. The cell lysates were properly diluted with 50 mM Tris, pH 8.0 and 0.1% Triton X-100.
2.11. In Vitro Processing of Procoat Cell-free synthesis of procoat labeled with [35s]-methionine was performed in an E. coli extract containing the pT712 plasmid encoding the procoat gene. Figure 2-2 shows the genomic structure of pT712 vector system. In this system each reaction (30 pl) contained DNA, trans [35s]-methionine (30 pCi), T7 RNA polymerase, 25
2812/1
1 (EcoR I-Pvu II) 137 (Acc I, 41 bp delation ) Hind III 331 273 Sph I Amp Pat I 497 +1 T7 transcript 331 Hind III MCS Acc I 576 EcoR I MCS , Hlnc II 2812 and I
Sat I EcoR I 576 orl
PT712 Promoter/Multiple Clonina Slta Sequence: + 1_ 480. | AATTAATACGACTCACTATAGGGAGACCACAACGGTTTCCCTCTAGCGGATCGC I______I 17 Promoter
5*0. .— 580. aagcttgggctgcaggtcgactctaga\x ;atccccgggcgagctcgaattcgtaatcatg Hind III Pat I Xba I X u I Sat I _____ I Sal I Sue I EcoR I I Acc I BaoH I I I Bine II j
Multiple Cloning Sit#
Figure 2-2. Genomic structure of PT712 vector 26
Triton X-100 (0.4%) and S150-2 (Yamane e ia L 1987). After 30 min at 37°C, synthesis was stopped by diluting in 50 mM Tris pH 8.0, 0.1% Triton X-100 (0.6 ml). To see if procoat was converted to the coat protein, an aliquot (10 pl) was treated with 2 pl of leader peptidase (1 mg/ml) and further incubated for 30 min at 37°C.
2.12. Immunoprecipitation Sample was mixed with equal amount of 20% TCA and incubated on ice for 15 min, and the precipitates were collected by centrifugation for 5 min in cold. The supernatant was aspirated away with a Pasteur pipet. After addition of 1 ml ice-cold acetone, the sample was again centrifuged for 5 min in the cold, and the acetone was aspirated away. The pellets were dried in heating block for 2 min and 100 pl of 10 ml Tris-HCI and 2% SDS pH 8.0 was added. Heating was continued for at least 8 min with intermittent vortexing. Centrifugation was performed to pellet debris for 5 min at room temperature. After addition of 1 ml of immunobuffer to the supernatant, the sample was mixed with 25 pl of Staph A followed by incubation for 15 min on ice for presorbing. The presorbed Staph A was removed by brief (15 seconds) centrifugation and antibody was added to the supernatant. After 30 min of incubtion on ice, 25 pl of Staph A was added to the sample and incubation was continued for another 30 min. The immuno-complex was then centrifuged down and washed with immunobuffer twice by successive resuspension and centrifugation. The washed pellet was resuspended in 2X 27 sample buffer and heated at 95°C for 5 min just prior to loading on a SDS-polyacrylamide gel.
2.13. Immunoblotting Immunoblotting was carried out according to the method of Towbin et al. (1979). Electrophoresed proteins were transferred onto nitrocellulose paper from the gel. The paper was then agitated in 10% nonfat dry powdered milk in H20 for 1 hour to block protein binding sites, placed in seal-a-meal-bag with serum diluted 1:500, and agitation continued for additional 2-16 hours. The paper was briefly rinsed with ddH20, and washed in TBS (5X TBS is 2 L water solution of 24.2 g Tris and 292.4 g NaCI, pH 7.5) for 1 hour with 2 washes. Then it was transferred to a bag with the goat anti-rabbit IgG horseradish peroxidase (diluted 1:3000 ) and incubated for 1 hour at room temperature. It was again briefly rinsed with ddH20 and washed 2 times in TBS containing 0.1% Triton X-100 for 1 hour with 2 washes. HRP color development solution (60 mg in 20 ml ice-cold methanol) was prepared in the dark and 60 pl of ice cold H20 2 was added just prior to use. The paper was immersed into the development solution and bands appeared on the transfer side of the paper. Development was stopped by incubating in distilled water for 10 min. The paper was then air-dried.
2.14. SDS-PAGE and Fluorography Samples in SB buffer ( 10 mM Tris-HCI, pH 8.0, 1 mM EDTA, 2% SDS, 5% B-mercaptoethanol and 0.001% bromophenol blue) were loaded on a gel immediatly after heating at 95°C for 5 min. Gel (19% for leader peptidase or 23% for procoat) was prepared according to the Table 2-1. Electrophoresis was performed at 30 - 40 mA for 2 - 3 hours in running buffer (0.19 M glycine, 0.024 M Tris, pH 7.5, 0.0034 M SDS). Then the gel was fixed in fixing solution (50% methanol and 7% acetic acid) for 45 min and treated with salicylate solution (1 Kg Sigma salicylic acid and 297 g NaOH pellets in 8 L water) for 15 min. Gel was dried and an X-ray film was put on the gel for fluorography.
2.15. Agarose gel electrophoresis 1% agarose gel was prepared for DNA analysis by dissolving ultra pure agarose in TBE buffer using a microwave oven. After addition of ethidium bromide to a final concentration of 0.5 pg/ml when the solution is around 65°C, the heated solution was poured into a casting tray. Electrophoresis was performed at 100 mA, and the DNA bands were visualized by UV light and subsequently photographed.
2.16. Site-Directed Mutagenesis Oligonucleotide-directed mutagenesis (Zoller and Smith, 1983) was used to generate mutations in either the leader peptidase and procoat. Each oligonucleotide was synthesized on an Applied Biosystem Model 380B Instrument by The Ohio State University 29
Table 2-1 Preparation of gels for SDS-PAGE
Separating gel (19%)
60% aery lam ide (w/v) 4.8 ml
1% bis-acrylamide (w/v) 1.1 ml
1M Tris-HCI (pH 8.725) 5.0 ml
Urea 5.4 g
10% SDS 0.4 ml
TEMED 5 pl
10% APS 50pl
Separating gel (23%)
60% aery lam ide (w/v) 15.3 ml
2% bis-acrylamide (w/v) 1.76 ml
2 M Tris-HCI (pH 8.725) 8.0 ml
Urea 14.4 g
10% SDS 0.4 ml 1 M NaCI 3.6 ml
TEMED 11 pl 10% APS 110 30
Table 2-1 Preparation of gels for SDS-PAGE
Stacking gel (5%)
30% acrylamide (w/v) 1.67 ml
1% bis-acrylamide (w/v) 1.3 ml
1 M Tris-HCI (pH 6.8) 0.625 ml
H2O 3.68 ml
Urea 3.6 g
10% SDS 0.1 ml
TEMED 5 pl
10% APS 50 pl Biochemical Instrument Center (Table 1). After annealing the mutagenic oligonucleotide and the primer to the single-stranded M13mp8 (for leader peptidase) or M13mp19 (for procoat) DNA containing the gene of interest, DNA polymerase I (large fragment) and T4 ligase were used to prepare the double-stranded DNA. To increase the efficiency of mutagenesis, the template DNA was prepared from phage grown on E. coli strain RZ1032 (ung- dut). After transformation into E. coli JM103 (ung +), the phage was isolated from single plaques and the mutation identified by dideoxynucleotide sequencing (Sanger e ia l, 1977). The replicative form of the M13mp8 or M13mp19 DNA containing the mutation was digested with restriction enzymes, and the fragment was inserted into either the plNG-1 or the pT712 plasmid.
Annealing reaction was carried out by mixing 5 pl of kinased oligonucleotide (40 pg/pl), 5 pl of DNA template (1 mg/ml), 1 pl of kinased primer (40pg/ml) and 1 pl of annealing buffer (0.2 M Tris- HCI, pH 7.5, 0.1 M MgCI2, 0.5 M NaCI and 0.01 M DTT), incubating for 7 min at 65°C and slowly cooling down to room temperature. Elongation and ligation was performed by mixing 12 pl of the above annealed DNA, 1 pl each of 10 mM dATP, 10 mM dTTP, 10 mM dGTP and10 mM dCTP, 2 pl of T4 DNA ligase and Klenow and 2 pl of ligation buffer (700 mM Tris-HCI, pH 7.5, 70mM MgCI2, 7 mM ATP and
100 mM DTT) followed by incubation overnight at 14 °C Oligonucleotides suspended in ammonium bicarbonate was spun in a speed vacuum condenser to evaporate the buffer. After resuspending the dried oligonucleotide in 0.5 ml of TEAB buffer (50mM triethanolamine bicarbonate, pH 8.5), the sample was desalted by a Sephadex G-25 column chromatography which had been preeqilibrated with TEAB buffer. DNA concentration of each 1 ml fraction was measured by a UV spectrophotometer at 260 nm wavelength.
Single-stranded DNA template was prepared as follow: E. coli RZ1032 cells were infected with M13 phage in 1 I of TYE media containing 0.25 mg/ml uridine and 12.5 pg/ml TET and grown overnight at 37°C. The next day, the culture was centrifuged at 10,000 rpm for 10 min at 4°C. The supernatant was mixed with 250 ml of 20% PEG-6000-2.5 M NaCI and incubated at 0°C for 1 hour. After centrifugation at 12,000 rpm for 10 min at 0°C, the phage pellet was resuspended in 15 ml of TYE media. 5 ml of 20% PEG-2.5 M NaCI was added and incubated for 30 min at 0°C to precipitate the phage again. The sample was centrifuged, and the phage pellet was resuspended in 4.5 ml of TE buffer and subjected to CsCI density gradient centrifugation. The phage band was pooled out by a syringe, and the M13 coat protein removed by phenol extraction. After extracting the phenol-treated DNA twice with diethyl ether, the phage DNA was precipitated by ethanol, dried by speed vacuum condenser, and resuspended in TE buffer. 33
2.17. Transformation and transfection E. coli cells were grown to early mid-log phase in 80 ml of TYE media, and the culture was centrifuged for 5 min at 6,000 rpm. The pelleted cells were suspended in 10 ml of cold 50 mM CaCI2 and incubated for 15 min on ice. Cells were pelleted again by centrifugation, resuspended in 2 - 3 ml of 50 mM CaCI2 and incubated for another 15 min on ice. 200 pl of the competent cells were mixed with plasmid or the ligated DNA, incubated on ice for 30 min, and then heat-treated for 2 1/2 min at 37°C. 1 ml of TYE media was added and the transformed cells were grown by shaking in a water bath for 40 min at 37°C. 200 pl of the culture was added to a TYE-agar plate containing 50 pg/ml ampicillin.
Transfection was carried out by mixing 200 pl of the competent cells with 5 pl of RF (Replicative form) DNA, incubating for 30 min on ice followed by heat-treating for 2 min at 37°C. The heat-treated cells were mixed with 0.5 ml of TYE media, 100 pl of which was then mixed with 0.7 ml of JM103 overnight culture. The mixture was then mixed with 3 ml of TYE-top agar solution at 47°C and poured onto TYE-agar plate.
2.18. Preparation of RF DNA and Plasmid For RF DNA preparation, 50 pl of phage and 100 pl overnight- grown JM103 were added to 35 ml of TYE, shaken overnight at 37°C, and the culture was centrifuged. 25ml of the supernatant containing the phage was added to a JM103 culture (20 ml of overnight-grown JM103 was inocculated into 1 L of TYE media and grown for 1.5-2 hours prior to the addition of the supernatent) and growing was continued for 2 hours. Cells (for plasmid preparation, overnight- grown E. coli cells containing plasmids) were harvested by centrifugation at 8,000 rpm for 10 min at 4°C using JA10 rotor. The pelleted cells were resuspended in 13.5 ml of 15% sucrose, 0.05 M EDTA, and 1 mg/ml freshly prepared lysozyme. After incubation at room temperature for 10 - 60 min, 14.5 ml of Triton solution (0.1% Triton X-100, 0.05 M Tris, pH 8.0 and 0.05 M EDTA) was added and incubation was continued for another 10-20 min at room temperature (30 min at 37°C if lysis does not occur). The sample was centrifuged at 19,000 rpm for 1 hour in a JA 20 rotor at 5°C . 27.5 ml of the supernatant was mixed with 26.13 g of CsCI and 0.5 ml of 10 mg/ml ethidium bromide was added (Refractive index should be between 1.39 and 1.396 or at a density of 1.59 g/ml). Ultracentrifugation was performed at 40,000 rpm for 24 hours in a vertical rotor at 20°C. Bands were visualized by long-wave UV illumination, and the lower band (covalently closed supercoiled DNA) was pooled out by syringe. Ethidium bromide was removed by extracting with 3 ml isobutanol three times. The bottom (water) phase was mixed with 15 ml TE buffer, 2 ml of 3 M NaOAc and 20 ml of iso-propanol and incubated at -70°C for 1 hour (or at -20°C for 2 hour to overnight) to precipitate the DNA. The DNA was pelleted by centrifugation at 17,000 rpm for 50 min at 4°C in JA20 rotor, 35 dissolved in 0.5 ml of TE buffer and transfered to 2 ml Eppendorf tube. The DNA solution was mixed with 50 pl of 3 M NaOAc and 0.5 ml of iso-propanol, and incubated at -70°C for 0.5 - 1 hour to reprecipitate the DNA. The DNA was again pelleted by centrifugation for 10 min in cold using microfuge, washed with 70% ethanol, dried using speed vacuum condenser, and finally dissolved in 0.5 ml of TE buffer.
To prepare plasmids or RF DNA rapidly on a small scale, 1 ml of the cells were pelleted by centrifugation for 20 sec in an Eppendorf microfuge. The supernatant was removed by aspiration, and the cells were resuspended in 100 pl of GTE buffer (50 mM glucose, 25 mM Tris pH 8.0 and 10 mM EDTA). After incubation for 5 min at room temperature, 200 pl of alkali-SDS mix (0.2 M NaOH and 1% SDS, freshly mixed) was added and incubation was continued for 5 min on ice. Then the sample was mixed with 150 pl of acetate solution (5 M acetic acid and 5 M potassium acetate, pH 4.8), incubated for 5 min on ice, and centrifuged for 1 min. The supernatant was mixed with 0.9 ml of 100% ethanol, incubated for 20 min at -70°C and centrifuged for 10 min at 4°C. The DNA pellet was washed with 0.9 ml of 70% ethanol, dried for 5 min in a Speedvac, and resuspended in 20 pl of sterile TE- buffer.
2.19. DNA Sequencing DNA was sequenced by the dideoxy method described by Sanger et al. (1977). Annealing reaction was carried out by mixing 7 pi of single-stranded DNA template, 1 pi of primer (OD = 0.05 at 260 nm) and 2 pi of the sequencing buffer (100 mM Tris-HCI, pH 7.5), incubating at 65°C for 5 min, and then slowly cooling to room temperature. The annealed DNA sample was mixed with 2 pi of labeling mixture, 1 ml of 0.1 M DTT, 2 pi of [&-35S]dATP and 1 pi of properly diluted sequenase, incubated for 2 - 5 min at room temperature. 3.5 pi portions of the reaction mixture were mixed with 2.5 pi each of ddATP, ddTTP, ddGTP termination mixes and ddCTP termination mixes, and incubated for 5 min at 37°C. The reaction was terminated by adding 4 pi of sequencing dye. Samples were heated for 3 min at 90 °C just prior to loading on a sequencing gel.
The single-stranded templates were prepared as following: Plaques were picked by Pasteur pipet from the transfection plate, inoculated into 1 ml of TYE media, and grown overnight. Centrifugation was carried out in Eppendorf for 1 min to pellet the cells, and the supernatant containing the phage was mixed with 250 pi of 20% PEG-6,000 and 2.5 M NaCI and incubated for 30 min on ice. The precipitated phage were pelleted by centrifugation for 5 min and resuspended in 100 pi of TE buffer. The phage solution was vortexed with 50 pi of Tris-saturated phenol and centrifuged for 15 sec. The aqueous phase containing the single-stranded M13 DNA was extracted twice with 500 pi of diethyl ether, mixed with 10 pi of 3 M NaOAc, pH 5.5 and 300 pi of absolute ethanol, and incubated at 70°C for at least 1 hour to precipitate the phage DNA. The DNA pellet was obtained by centrifugation for 10 min in cold, washed with cold 80% ethanol, dried in Speedvac, and finally dissolved in 50 pi of TE buffer.
6% polyacrylamide gel electrophoresis was performed for separation of the DNA in the sequencing mixture. To prepare a gel, 36.04 g of urea was dissolved in 15 ml of 5X TBE buffer and 11.25 ml of acrylamide-bisacrylamide (40:1.33) solution, and water was added up to 75 ml. After filtration, the solution was mixed with 200 pi of 10% APS and 30 pi of TEMED and poured between 2 glass plates. Polymerization was carried out at room temperature for at least 1 hour, and the gel was heated by pre-running at 1500 V for 30 min just prior to loading samples. Samples were electrophoresed at 1500 - 2000 V for the appropriate time.
2.20. Subcloning Restriction digestion was performed by mixing 60 pi of vector DNA or RF DNA (containing the inserted DNA), 10 pi of 10X restriction buffer, 20 pi of water and 5 pi each of two restriction enzymes followed by incubation at 37°C overnight. The reaction was terminated by adding 9 pi of TBE dye and the sample was run on agarose gel at 150 V. DEAE-cellulose papers were placed on both sides of the interested DNA band, and the DNA was electrophoresed 38 backwards onto the upper filter. The filter was washed w ith 250 pi of low salt NET buffer (0.15 M NaCI, 0.1 mM EDTA and 20 mM Tris, pH 8.0), and the DNA was extracted with high salt NET buffer (1.0 M LiCI, 0.1 mM EDTA and 20 mM Tris, pH 8.0) for 1 hour in 65 °C (vortex every 15 min). 0.5 ml of phenol was added to the sample, vortexed, centrifuged for 15 sec, and the aqueous phase was transferred to a new microfuge tube. The sample was then extracted twice with 0.5 ml ether, mixed with 2 volumes of ethanol, and incubated for 45 min at -70°C (or overnight at -20°C). The precipitated DNA was collected by centrifugation for 10 min at 4oC, dried in the Speedvac, and resuspended in 100 pi of TE buffer. During this time, a spin column was prepared by pinching a tiny hole in a small Eppendorf tube with 25 X 5/8 needle, filling the Eppendorf tube with Sephadex G-25 in TE buffer, spinning for 1 min, filling again with Sephadex G- 25, spinning for 3 min, filling with TE buffer, and spinning until dry. The DNA solution was applied to the spin column and centrifuged for 3 min, and the flow-through containing pure DNA was obtained. 2.5 pi of the sample was mixed with 3 pi of TBE dye and run on 0.7% agarose gel to check yield.
For ligation, 2 pi of the isolated linear vector, 5 pi of DNA to be inserted, 1 pi of 10X ligation buffer and 1 pi of T4 DNA ligase was mixed and incubated at 14°C overnight.
2 .20. Special Preparation of Solutions 39
(a) Phenol-Saturated Phenol was removed from the freezer, allowed to warm to room temperature, and melted at 68°C. 8-hydroxyl-quinoline was added to a final concentration of 0.1%, and the melted phenol was extracted several times with 1.0 M Tris, pH 8.0 and 0.2% (3- mercaptoethanol. The phenol solution was stored at 4°C under equilibration buffer for a period of up to 2 month.
(b) 3M Na Acetate 408.1 g of sodium acetate(3H20) was dissolved in 800 ml of water. pH was adjusted to 5.2 with glacial acetic acid and the volume was adjusted to 1 L.
> (c) 1 M TEAB, pH 8.5 1 mole of triethanolamine and water was mixed to make 800 ml solution. Dry ice was added to the solution until pH fell to pH 8.5. Volume was adjusted to 1 L.
(d) 1 M Tris-CI for protein gel 121.1 g of Tris base was dissolved in 800 ml of water and stirred till it reaches room temperature. After adjusting the pH meter dial to the temperature, HCI was added until the pH decreased to 8.4. Concentrated NaOH was added until the pH reached 8.72 and water was added such that the final volume of the solution was 1 L. 40
2.21. Construction of Pf3-Leader peptidase and Pf3-H1 Hybrid proteins between Pf3 peptide and leader peptidase. A DNA fragment encoding the first 18 amino acids of Pf3 coat protein (Rohrer and Kuhn, 1990) was joined in frame with the beginning of the leader peptidase gene (Fig. 2-3). This was achieved using oligonucleotide-directed mutagenesis to introduce a unique Sea 1 restriction site near codons 2 and 3 of the leader peptidase gene on an M13mp8 vector. The synthetic oligonucleotide 5' CCCTTAGGAGTTGGCATGAGTACTATGTTTGCCCTGATT was prepared on an Applied Biosystem Model 380B Instrument by the Ohio State University Biochemical Instrument Center. The oligonucleotide was annealed to a single-stranded DNA from M13mp8 containing the leader peptidase gene. After elongation of the template and ligation (Dalbey and Wickner, 1987), the double-stranded DNA was used to transform E. coli JM103. The mutant phage within single plaques was identified by DNA sequencing. After subcloning the mutant gene into the pING plasmid (Johnston elai-, 1985), pRD13 was produced. After digestion of this plasmid with Sal 1 and Sea 1, and purification of the large linear vector fragment (coding for all but three amino acids of the leader peptidase gene), the DNA was religated to the 250 bp Sal 1-Eco RV fragment of the Pf3 coat gene encoding the first 18 residues (Rohrer and Kuhn, 1990) to produce a plasmid encoding an in-frame fusion of Pf3 coat and leader peptidase. Finally, four mutagenic primers and the changes they encode are: CTGACAGCGGTGCAACGTCGTACTATGTTTGCCCTG, PLEASE NOTE:
Page{s) missing in number only; text follows. Filmed as received.
UMI A la ^ G lu 18 to A rg^A rg^S ;
TTCTTTTTCGCACCTCAGGCAGCGGCGCAG, 30LySArgArgGluArg34 are deleted; GCGGTGCAAGCTCGTACTATGTTTGCC, Glu18 to Arg18; CGGCGGGAACGTTAGTAGGCGGCGCAG, Gln35Ala36 to amber codons. CHAPTER III
IN VITRO TRANSLOCATION STUDIES OF E. COLI LEADER PEPTIDASE
3.1. INTRODUCTION Proteins fold spontaneously to form their native compact three dimensional structure during or immediately after their synthesis. Exported proteins destined to traverse membrane(s) to reach their final destination, however, are believed to fold into a distinct conformation which is stable and loose enough for membrane translocation (Randall and Hardy, 1986; Cover et al., 1987; Eilers and Schatz, 1986). Thus, some soluble factor(s) had been expected to form a stable complex with a precursor of an exported protein in cytoplasm to allow a ’translocation-competent’ conformation.
Recently, a group of soluble proteins known as 'chaperones' have been identified and found to promote proper assembly of oligomeric complexes (Ellis, 1987; Hemmingsen et al., 1988). Chaperones were also found to be able to associate with newly made exported proteins. This association allows exported proteins to be in a translocation-competent conformation prior to the 43 44 translocation across membrane, both in eukaryotes and prokaryotes.
In eukaryotic cells, hsp70 support protein translocation into dog pancreas microsome (Zimmermann et al., 1988). In yeast, protein translocation across the endoplasmic reticulum and into the mitochondria requires four heat shock proteins encoded by SSA1-4 are required (Deshaies et al., 1988; Chirico et al., 1988). Also a signal recognition particle (SRP), consisting of six polypeptides and one RNA molecule, is known to associate with the emerging protein, arrest the translation, and bring the protein to the ER for translation coupled translocation across the membrane (Walter et al., 1984).
Chaperones also participate in bacterial protein export. E. coli inner membrane is the most extensively studied prokaryoic membrane protein translocation system (Kumamoto and Beckwith, 1985; Bochkareva et al., 1988; Collier et al., 1988; Crook et al, 1988). Thus far, SecB (Collier et al., 1988; Kumamoto and Gannon, 1988; Weiss et al., 1988), trigger factor (Crooke and Wickner, 1987; Crooke et al., 1988), and GroEL have been identified as chaperones involved in this transport system.
We studied the translocation of the C-terminal domain of EL coli leader peptidase across the inner membrane, in the hope that SecB, trigger factor, GroEL, or some other soluble fators might be involved in this translocation. Biochemical studies using in vitro 45 translocation systems have been essential to identify those factors.
We first tried to reconstitute both co- and post-translational translocation of leader peptidase in vitro in order to identify any soluble factor(s) involved in this translocation. We checked pro- OmpA as a control. The pro-OmpA traslocation was successfully reconstituted in both co- and post-translational translocation systems. Leader peptidase translocation was successful only in the co-translational system. In the post-translational system, however, leader peptidase did not translocate even under the condition that pro-OmpA did. Using the co-translational translocation system, we attempted to identify soluble factor(s) involved in leader peptidase translocation across the E. coli inner membrane.
3.2. RESULTS Post-Translational Translocation of Leader Peptidase In Vitro It is commonly believed that bacterial membrane proteins insert across the membrane after their synthesis has been completed. To test whether E. coli leader peptidase also can translocate across the inner membrane post-translationally, leader peptidase was first synthesized in vitro and then added to the post- translational translocation system. After analysis by protease mapping, the bands protected by the membrane were checked to see if translocation into the inner membrane vesicle occured. We found that pro-OmpA was processed to OmpA by inner membrane vesicle in 46 an energy-dependent manner (Fig. 3-1); The OmpA band did not appear in the absence of ATP or NADH. Translocation of the C- terminus of leader peptidase across the inverted membrane vesicle was monitored by appearance of a down-shifted band of the protected C-terminus after protease mapping. No protected band of leader peptidase, however, was detected under the condition that allows the translocation of OmpA nor did manipulation of concentration of ATP and NADH help leader peptidase translocation.
To prevent conformational change to a translocation- incompetent form due to the time interval between the synthesis and translocation reaction, we used a urea dilution technique. Leader peptidase in 8.0 M urea was added to the post-translational translocation system so that the urea concentration in the system was 0.8 M, under which condition translocation of pro-OmpA was not affected. However leader peptidase still did not translocate across the membrane (Fig. 3-2). Addition of purified SecA also was not helpful. Similar experiments were carried out with affinity- purified [ 35 S]-labeled leader peptidase, which gave the same results.
All together, these results indicate the membrane translocation of the large C-terminal domain of leader peptidase is impossible or very inefficient after the protein has been completely synthesized. 47
Membrame +
Proteinase K . +
Figure 3-1. In vitro post-translational translocation of pro- OmpA. 48
Leader Peptidase
Proteinase K
Figure 3-2. Urea dilution study of post-translational translocation of leader peptidase. 49
Co-translational translocation of leader peptidase Translocation of leader peptidase and pro-OmpA was also tried while the translation was occuring. We developed a very efficient in vitro translation system using the soluble fraction S 150-2 (Yamane fita i-, 1987) and pT712 plasmid encoding leader peptidase or pro-OmpA. Co-translational translocation was performed by earring out the translation reaction in the presence of inverted membrane vesicle. In this system, the translocation of both leader peptidase and pro-Omp was very successful. The translocation efficiency of leader peptidase was about 30% (Fig. 3-
3).
Searching for Soluble Factors for Membrane Translocation of Leader Peptidase Using co-translational translocation system, we searched for soluble factor(s) involved in membrane translocation of leader peptidase. We used two strategies: (1) Heat-inactivation of possible soluble factors in S150-2 followed by addition of fractionated S150-2 to restore the translocation. (2) Use of antibodies to known E. coli chaperones (SecB, trigger factor and GroEL) to abolish the translocation.
S150-2 was heat-treated at various temperatures for various times in the hope that we could inactivate soluble factor(s) while not disturbing translation. Under all conditions that support translation, however, translocation activity was not lost (Fig. 3-4). 50
S'
Leader Peptidase
Membrane-protected C-terminal domain of leader peptidase
+ Proteinase K
Figure 3-3. In vitro co-translational translocation of leader peptidase. 51
. * I " | i ra il a i • r .. siL Leader Peptidase
Membrane- protected C -term in al domain of LPase
1 Hit -S i Heating time of S150-2
Proteinase K
Figure 3-4. Effect of heat-treatment of soluble fraction on the co-translational translocation of leader peptidase. 52
While this strategy turned out not to be useful for our purpose, the result suggests that, if there are soluble factors involved in the membrane translocation of leader peptidase, they are heat-stable.
Antibodies to SecB, trigger factor and GroEL were added to the co-translational translocation system to see their effects on leader peptidase translocation. None of the antibodies were able to abolish the translocation activity of S150-2 on leader peptidase (Fig. 3-5). These results suggests that leader peptidase does not require these chaperones, and may require no soluble protein at all.
3.3 Discussion E. coli leader peptidase requires SecA, SecY and electrochemical potential for the translocation of its large periplasmic C-terminal domain across the inner membrane. However it was not known whether soluble protein(s) are involved in the translocation to stabilize and allow leader peptidase to be in a translocation- competent conformation. We tried to identify them using co- and post-translational translocation systems.
Normally, the membrane translocation of leader peptidase occurs co-translationally in vivo. However, it had been shown that the translocation could be forced to occur post-translationally under certain conditions; when the electrochemical potential was abolished by an uncoupler, CCCP (Carbonyl cyanide-m-chlorophenyl hydrazone), leader peptidase remained on the cytosolic side of the 53
Leader Peptidase
1 Membrane- protected C-term inal domain of LPase
Proteinase K
SecBa n t|- antianti- ant(_ trigger factor gr 0 EL
Figure 3-5. Effects of antibodies to chapernones on the co- translational translocation of leader peptidase. 54 membrane without further translocation. Upon the restoration of the electrochemical potential by DTT (which destroys CCCP), the translocation of C-terminal domain was completed. This previous study led us to pursue the reconstitution of post-translational translocation of this protein in vitro, since this system is preferable to the co-translational system to study soluble factors and energy requirements of the protein translocation. While co- translational translocation system contains all the proteins and energy sources necessary for translation, the post-translational translocation system is free of any soluble protein and/or energy sources. We can add purified precursor proteins and any suspected soluble factor or energy sources to the system and examine the effect on protein translocation. However our results indicate leader peptidase translocation is unlikely to occur in vitro.
Leader peptidase may interact directly with the membrane by means of its N-terminal hydrophobic domain(s) and surrounding positive charges as it is being on the ribosome. The positive charges may interact with the negatively charged head-groups of membrane phospholipids and the hydrophobic domain(s) may partition into the membrane. And the translocation of its C-terminal domain may occur, with the help of SecA, SecY and electrochemical potential, after the N-terminus has inserted. The presence of the fully- synthesized large C-terminal domain may prevent these interactions by shielding the N-terminus from the membrane. Thus, in urea dilution study, the renaturation may be so fast that the shielding happen before the protein reaches the membrane. Our results that antibodies to chaperones (SecB, trigger factor and GroEL) did not abolish the translocation activity of S150-2 on leader peptidase also support the direct interaction between leader peptidase and the membrane. This also may be the way leader peptidase avoids aggregation by hydrophobic interaction in vivo. CHAPTER IV
DISTINCT DOMAINS OF AN OLIGOTOPIC PROTEIN ARE SEC- DEPENDENT AND SEC- INDEPENDENT FOR MEMBRANE INSERTION
4.1. INTRODUCTION The information that dictates the membrane orientation of proteins is specified by apolar integration segments in the protein sequence (Wickner and Lodish, 1985; Verner and Schatz, 1988; Dalbey, 1990). Amino-terminal cleavable leader peptides (Michaelis and Beckwith, 1982; Carlson and Botstein, 1982) and uncleaved signal sequences (Blobel, 1980) promote carboxyl-terminal insertion of proteins. These sequences span the membrane with their amino termini cytoplasmic and their carboxy termini on the lumenal side. While leader peptides are removed by a membrane-bound leader peptidase after the exported protein crosses the membrane, uncleaved signals remain in the mature protein, usually as transmembrane anchors. Spanning the membrane in the opposite direction are stop-transfer sequences (Davis and Model, 1985) that halt the protein translocation initiated by a signal peptide. In addition to these integration segments, some proteins have signal sequences (so called signal-halt peptides) that initiate amino- terminal insertion (von Heijne and Gavel, 1988). They are oriented 56 57 with their carboxy termini facing the cytosol, which is an unusual orientation for signal sequences. Molecular genetic studies have shown that the hydrophobic character is important for each of the integration peptides (Bassford M M -, 1979; Emr M M -. 1978; Spiess and Handschin, 1987; Dalbey and Wickner, 1988; Zhu and Dalbey, 1989; Davis M M -. 1985; Hull M a i, 1988).
We are studying leader peptidase, a transmembrane protein of the plasma membrane of E. coli. as a model to define the structural information for membrane assembly. Leader peptidase spans the lipid bilayer twice (Fig. 4-1) and has a small domain near its amino terminus facing the cytoplasm and a large hydrophilic region exposed to the periplasm (Wolfe M ai-. 1983; Moore and Miura, 1987). Like many periplasmic and outer membrane proteins, leader peptidase requires the SecA and SecY gene-encoded proteins (Wolfe M ai-, 1985) and the electrochemical membrane potential (Wolfe and Wickner, 1984) for the membrane insertion of its large polar carboxyl-terminal domain. In contrast, as seen for most proteins of the plasma membrane, leader peptidase is synthesized without an amino-terminal leader peptide. Genetic studies have shown that this protein does have an internal, uncleaved signal peptide, which is essential for the translocation of the large carboxyl-terminal domain and also functions as a membrane-spanning segment to connect the two polar domains (Dalbey and Wickner, 1987; Dalbey M aL. 1987; Dalbey and Wickner, 1988; Zhu and Dalbey, 1989). 58
Pf 3 peptide
Periplasm
Inner Mem brane
Cytoplasm
+ — Pf 3 Coat Leader Pf 3 peptidase Leader peptidase
Figure 4-1. Topology of Pf3 coat protein, leader peptidase and
Pf3-leader peptidase. 59
How does leader peptidase insert across the membrane to expose its amino terminus to the periplasm when it does not contain a cleavable leader peptide? It has been proposed that uncleaved proteins such as leader peptidase contain a signal-halt sequence that inserts the amino terminus across the membrane and then halts the transfer (Monier £iai-, 1988). Furthermore, it has been suggested that these transmembrane segments either spontaneously insert into the membrane (Wickner and Lodish, 1986) or they insert by a mechanism requiring receptor proteins (Anderson e ia l, 1983).
4.2 RESULTS To determine how this transfer process occurs, we first developed a method to monitor insertion, which involves introducing a small antigenic peptide at the beginning of leader peptidase and showing that this segment was accessible to proteases added from the outside of the cell. We chose the first 18 residues of the Pf3 coat protein to fuse to the beginning of leader peptidase because 1) this peptide is immunoprecipitated by antibody to the Pf3 coat protein, and 2) it is digested by proteinase K but not by trypsin (since it contains no lysine or arginine residues). Figure 4-1 depicts the predicted membrane topology of the fusion protein. To prove that our construction indeed contains the Pf3 peptide, we tested whether Pf3-leader peptidase is recognized by antiserum generated against the Pf3 bacteriophage coat protein. Exponentially growing cells 60 were induced with arabinose and labeled with [35s]-methionine for 1 min. Figure 4-2 shows that a band is detected with Pf3 antiserum (lane 2) and leader peptidase antiserum (lane 3) but not with prebleed serum (lane 1).
To establish that the Pf3 peptide is localized in the periplasm, we isolated right-side-out membrane vesicles and treated them with proteases. Previously, Moore and Miura (1987) showed that proteinase K and trypsin treatment of these vesicles containing wild-type leader peptidase yields amino-terminal protected fragments of approximately 82 and 105 amino acid residues, respectively. Our cells were labeled with [35s]-m ethionine, converted to spheroplasts, then sonicated and analyzed by isopycnic sucrose density centrifugation (Osborn e la i-. 1972). The isolated inner membrane vesicles were digested with trypsin or proteinase K. As expected with wild-type leader peptidase (Moore and Miura, 1987), protease-resistant fragments of 11,000 KDa (lane 3) were generated with trypsin, and fragments of 9,000 KDa (lane 5) with proteinase K (Fig. 4-3). When proteolysis was carried out in the presence of detergent, these fragments were not detected (data not shown). Trypsin treatment of vesicles containing Pf3-leader peptidase yields a protected fragment slightly larger than the 11 kd fragment because the 18 residues that have been added to the amino terminus of leader peptidase do not include arginine or lysine amino acids (Fig. 4-3, lane 4). Proteinase K yields a peptide with the same Figure 4-2. Pf3-leader peptidase is recognized by antibody to the Pf3 coat protein and the leader peptidase antibody. Cells bearing pJF1 encoding Pf3-leader peptidase were grown at 37°C to mid-log phase in M9 minimal medium with 0.5% fructose and all amino acids except methionine, incubated with arabinose ( 0.2%) for 1 h, and then labeled with [ 35 S]-methionine for 2 min. Samples were immunoprecipitated (Wolfe M a i., 1982) with antiserum to Pf3 (lane 2), leader peptidase (lane 3) or prebleed serum (lane 1), and then analyzed by SDS-PAGE and fluorography (Ito M a i-, 1980).
61 62
1 2 3
Pf 3 Leader peptidase
Figure 4-2. Pf3-leader peptidase is recognized by antibody to the Pf3 coat protein and the leader peptidase antibody. Figure 4-3. Proteolysis of inner membrane vesicles - Total [ 35 S] labeled membrane proteins. Inner membrane vesicles containing wild-type leader peptidase or Pf3-leader peptidase were prepared from E. coli strain MC1061 bearing pRD 8 (Moore and Miura, 1987) or pJL1 encoding Pf3-leader peptidase as described and incubated for 1 h at 37 °C without further addition , or with proteinase K or trypsin (0.2 mg/ml). After addition of soybean trypsin inhibitor and PMSF, samples were analyzed by SDS-PAGE and fluorography.
63 64
1 2 3 4 5
-13k - 11k - 9k
. v \/ \/ ProtG3SG. None Trypsin Proteinase k
Figure 4-3. Proteolysis of inner membrane vesicles - Total [ 35 S] labeled membrane proteins. 65 mobility as the 9 KDa (lane 6 , compared with lane 5). This is strong evidence that the Pf3 region has been digested by proteinase K, suggesting that this region faces the periplasm. Figure 4-4 shows that only the tryptic peptide derived from Pf3-leader peptidase (lane 2), not the proteinase K peptide (lane 4), is immunoprecipitable by Pf3 antibody, confirming that the Pf3 region is out but not cleaved by trypsin in right-side-out vesicles.
To test if the first transmembrane segment of leader peptidase initiates translocation of the amino-terminus, we introduced two amber codons immediately after the DNA that codes for residue 35 of the protein (Fig. 4-5-bottom construction). Synthesis of this truncated protein, termed Pf3-H1, was assayed by labeling with [35s]-methionine, immunoprecipitation with antiserum to Pf3 coat protein, and SDS-PAGE and fluorography (Fig. 3-6, lane 1 or 4). After labeling, cells were converted to spheroplasts and treated with trypsin or proteinase K for 60 min at 0°C. Pf3-H1 was completely digested with proteinase K (Fig. 4-6, lanes 4 and 5), indicating the Pf3 peptide had inserted across the membrane. As a control, we confirmed that Pf3-H1 is not accessible to trypsin, as it does not contain any of the arginyl or lysyl residues necessary for trypsin cleavage (Fig. 4-6, compare lanes 1 and 2). In these experiments we determined that a cytoplasmic protein (ribulokinase) remains inaccessible to digestion, while mature outer membrane protein A (OmpA) is fully digested by protease. This data 66
3 4 5 Pf3 Leader- i Peptidase
r. I : i 13k- r, ' S ‘ i' . 9k- " ■ l *:u ■ ( \/ \ / Protease: None Trypsin Proteinase k Detergent: — - + - +
Figure 4-4. Immunoprcipitation of Proteolysed inner membrane vesicles with antiserum to Pf3. 67
Pf 3 Coat N-
Leader peptidase N 1 1— — — - — —f f 1 I— i------i— / / - C
Hybid Proteins
P( 3 Leader peptidase N ~ < " 'I **- — - f f ------1 ~i— i i— / £ - C
Pf 3 Leader peptidase RR N - ——i ------1 I ------// ------1------1—i------1—/ / _ Q
Pf 3 Leader peptidase A N——l _ l ------' ------7/------1---- "i—<----- \— {/—C
R 3 Leader peptidase RRA N———T- zm—cmz]—//—C
Pf 3 HI
Figure 4-5. Hybrid proteins between Pf3 peptide and leader peptidase. Figure 4-6. Protease mapping of Pf3-H1. E. coli MC1061 bearing plasmid encoding Pf3-H1 was grown to mid-log phase, induced with 0.2% arabinose for 1 h at 37°C, and labeled with 50 pCi of [ 35 S]- methionine for 1 min. Protease mapping was performed as described in Fig. 2. using trypsin and proteinase K. Soybean trypsin inhibitor and PMSF were added to inhibit the proteases. Samples were immunoprecipitated with antiserum to OmpA, leader peptidase or ribulokinase, and then analyzed by SDS-PAGE and fluorography.
68 lysis lysis Trypsin Proteinase k
Figure 4-6. Protease mapping of Pf3-H1. 70
establishes that the first apolar domain, along with the charged carboxyl-terminal region, is sufficient for translocation and therefore resembles a signal-halt sequence.
The carboxyl-terminal domain of leader peptidase requires the SecA and SecY proteins for translocation across the membrane (Wolfe etai., 1985). To investigate whether leader peptidase requires the Sec machinery for amino-terminal insertion, we transformed SecAts and SecYts strains with a plasmid encoding Pf3- leader peptidase. Protease-accessibility experiments (Dalbey and Wickner, 1986) were performed at the non-permissive temperature to determine whether the amino terminus of Pf3-leader peptidase translocates across the membrane. Cells expressing Pf3-leader peptidase were labeled with [35s]-methionine for 1 min, converted to spheroplasts, then incubated with proteinase K for 60 min. Samples were then immunoprecipitated with antibodies to outer membrane protein A or to leader peptidase. After addition of proteinase K, there was a slight shift in the molecular weight of Pf3-leader peptidase, showing that the Pf3 peptide crossed the membrane and was digested with protease (Fig. 4-7A and B). In addition, the shifted band was not immunoprecipitable with antibody to Pf3 coat protein (data not shown). These results show that the amino and the carboxyl termini of leader peptidase have different insertion requirements in terms of the Sec genes. Figure 3-7. Effects of SecA inactivation on the Insertion of the amino terminus of Pf3-leader peptidase. E. coli CJ105 bearing pJL1 coding for Pf3-leader peptidase was grown to mid-log phase at 30°C, shifted to 42°C, induced with 0.2% arabinose for 4 h, and then labeled with 500 jiCi of [3 5 S]-methionine for 1 min. Protease mapping was performed by converting cells to spheroplasts (Randall and Hardy, 1986), followed by incubating with or without proteinase K (final concentration, 1 mg/ml) for 1 h at 0°C. Where indicated, a portion of cells was treated with Triton X-100 prior to the addition of proteinase K to lyse the cells. Subsequently, PMSF (5 mM final concentration) was added to inhibit the protease. Samples were immunoprecipitated with antiserum to OmpA or leader peptidase, and then analyzed by SDS-PAGE and fluorography. (B) Effects of SecY inactivation. E. coli CJ107 bearing pJL1 was grown, labeled, and analyzed as above in A.
71 72
A. S e c A ls B. Sec Yt s 1 2 3 1 2 3
Pro-Omp A- Pro-Omp A- <
Pf3 Pf3 Leader- Leader- ^ Peptidase —r Peptidase Proteinase k - + + Proteinase k - + + lysis lysis
Figure 3-7. Effects of SecA inactivation on the Insertion of the amino terminus of Pf3-leader peptidase. We next tested whether the proton motive force is needed for assembly of the amino terminus since it is required for carboxyl- terminal translocation of almost all exported proteins, including leader peptidase (Wolfe and Wickner, 1984). Figure 4-8 shows the results of pretreatment of cells expressing Pf3-leader peptidase with CCCP, which collapses the membrane potential. In this study, samples were analyzed as described in Figures 4A and 4B. Strikingly, treatment with CCCP prior to [35S]-methionine labeling does not prevent the very amino terminus of leader peptidase from translocating across the membrane and being digested with externally added proteinase K. This is revealed by the slight shift in the molecular weight of Pf3-leader peptidase upon adding proteinase K to spheroplasts (Fig. 4-8). As a control, we confirmed that the proteinase K digested protein was not precipitated with Pf3 coat antiserum (data not shown). Thus, although the carboxyl-terminal domain of leader peptidase cannot translocate when the membrane electrochemical potential is depleted, the amino terminus of the protein molecule can still cross.
Previously, it has been demonstrated that the positively charged cluster, Lys-Arg-Arg-Glu-Arg, immediately downstream from apolar domain 1 prevents this transmembrane segment from orienting itself in the membrane with its carboxy terminus in the periplasm and keeps it from functioning as an export signal (von Heijne el at-. 1988; Laws and Dalbey, 1989; San Millan etaL, 1989). 74
1 2 3
Pro-Omp A -
Pf3 Leader - Peptidase
Proteinase k - + + lysis
Figure 4-8. Effects of membrane potential disruption on the Insertion of the amino terminus of Pf3-leader peptidase. E. coli MC1061 bearing pJL1 was grown to mid-log phase and induced with 0.2% arabinose for 1 h at 37°C, treated with 5 pi of 10 mM CCCP for 45 s, labeled with 500 pCi of [35s]-methionine for 1 min, and then analyzed as above. 75
The membrane orientation of leader peptidase can also be reversed by changing the charge distribution around the first transmembrane segment (von Heijne, 1989; Nilsson and von Heijne, 1990). In these earlier studies, there was no possibility of directly monitoring the location of the amino terminus of leader peptidase. Therefore, we examined the effect of positively charged residues bordering the first transmembrane segment on the insertion of this region. Oligonucleotide-directed mutagenesis was used either to delete residues 30-34 after H1 and/or to substitute the amino acids 18 and 19 before H1 with arginyl residues. Figure 4-5 shows the mutations that were made. To monitor insertion of the mutant Pf3-leader peptidase molecules, we studied their assembly properties in a SecA temperature-sensitive strain. Cells synthesizing Pf3-leader peptidase RR, RRA and A were pulse-labeled for 1 min at 42°C and then analyzed for translocation as described in Figure 4-7A. The amino terminus of Pf3-leader peptidase RR, with two positively charged residues at +17 and +18, does not translocate across the membrane since the molecular weight was not changed after proteinase K digestion (Fig. 4-9, left panel). Similarly, the full- length RRA is resistant to protease (Fig. 4-9, middle panel), indicating that the amino terminus does not cross the membrane. In contrast, the amino terminus of the A mutant does cross, establishing that the positively charged cluster is not absolutely essential for insertion. Two leader peptidase species were observed in the 1 min pulse-labeling for the RRA and A mutant (Fig. 4-9). We 76
RR RRA A 1 2 3 1 2 3 1 2 3
Pro-Omp A-Pro-Omp A- Pro-Omp A -
Pf3 Pf3 Pf3 Leader- Leader - Leader- Peptidase Peptidase i Peptidase
Proteinase k - + + Proteinase k - + + Proteinase k - lysis lysis lysis
Figure 4-9. Effects of positive charges on the translocation of Pf3-leader peptidase - Amino-terminal insertion in a SecA temperature-sensitive strain. E. coli CJ105 with a plasmid coding for RR, RRA, or A Pf3-leader peptidase was analyzed for insertion of its amino terminus, as described in Fig. 4A. 77 suspect that the lower molecular weight species is derived from the larger protein by degradation after it inserts across the membrane with the reverse orientation. These smaller species had also been observed previously when the membrane topology of leader peptidase was inverted (von Heijne, 1989; Nilsson and von Heijne, 1990). Figure 4-10 shows that the carboxyl-terminal domain of the leader peptidase mutants translocates across the membrane in cells with functional Sec genes. The smaller protein molecules for the RRA and A mutant are resistant to protease, consistent with their having the reverse orientation.
4.3. DISCUSSION Our results here with leader peptidase fused to the short Pf3 peptide show for the first time that the amino terminus of leader peptidase is periplasmic. Using this Pf3-leader peptidase fusion protein, we establish that there are striking differences between the insertion requirements of the amino and carboxyl-terminal portions within the same protein. The SecA and SecY proteins and the proton motive force are not required for insertion of the amino terminus of leader peptidase, whereas they are needed for insertion of the large hydrophilic carboxyl-terminal domain. The Sec- independent insertion is similar to that of the 44 residue Pf3 bacteriophage coat protein, which spans the membrane once (Rohrer and Kuhn, 1990) with a basic carboxyl-terminal region and a short 78
A RR RRA 2 3 2 3 1 2 3 PI3 PI3 PI3 Leader- — • Leader- Leader- Peptidase Peptidase j f e M Peptidase i* m ___
Proteinase k - + ♦ Proteinase k - + Proteinase k - lysis lysis lysis
Figure 4-10. Effects of positive charges on the translocation of Pf3-leader peptidase - Translocation of the carboxyl-terminal domain in MC1061. Cells containing a plasmid which encodes RR, RRA, and A Pf3-leader peptidase were pulse-labeled for 1 min and analyzed by proteinase K mapping, as described in Fig. 4. 79 acidic region (Fig. 4-1). These results support the idea that the Sec machinery is required only to translocate large polar regions and not short domains (Kuhn fila l, 1990; von Heijne, 1989).
The first 35 leader peptidase residues are sufficient to transfer the amino-terminal Pf3 peptide across the membrane, strongly suggesting that leader peptidase contains a signal peptide that initiates amino-terminal insertion. This unusual signal sequence has structural features different from those of a leader or uncleaved signal peptide that spans the membrane in the opposite direction: it has a short amino-terminal nonbasic extracytoplasmic domain, an apolar core of approximately 20 residues, and a basic carboxyl-terminal cytoplasmic region (Dalbey and Wickner, 1987). With the identification of this unusual signal peptide, we have found two apolar regions that are important for the membrane biogenesis of leader peptidase. The membrane-spanning region, which includes the second apolar domain, is an internal, uncleaved signal peptide that is essential for translocation of the large carboxyl-terminal domain (Dalbey e ia i-, 1986). These studies strongly suggest that leader peptidase contains two linear apolar integration segments that have distinct roles and can function out of their normal context.
The determining factor for the orientation of membrane proteins seems to be the positively charged residues that border the transmembrane segments (Dalbey, 1990). For example, the 80 orientation of the apolar domain 1 of leader peptidase can be inverted when charged residues are reversed on the two sides of the hydrophobic region (von Heijne e la i., 1988; Laws and Dalbey, 1989; San Millan fila i-, 1989). The evidence presented here with the Pf3- leader peptidase fusion protein (Fig. 4-9, left and middle panel) shows that the Pf3 segment (1 Met-Gln-Ser-Val-lle-Thr-Asp-Val- Thr-Gly-Gln-Leu-Thr-Ala-Val-Gln-Ala-Asp18) must have certain properties in order to be translocated. The substitution of alanine at position 17 and glutamic acid at position 18 with positively charged residues blocks translocation of the amino terminus. In addition, the positively charged cluster (Lys- Arg-Arg-Glu-Arg) at the carboxyl-terminal side of apolar domain 1 of leader peptidase is not essential for insertion of the amino terminus (Fig. 4-9, right panel). However, this basic cluster is important in preventing this hydrophobic domain from reversing its orientation (Fig. 5; Nilsson and von Heijne, 1990).
Figure 4-11 shows our current model of leader peptidase membrane topogenesis. Soon after synthesis, the first apolar domain targets leader peptidase to the membrane. This targeting, which does not require the Sec machinery or the membrane electrochemical potential, may involve both electrostatic and hydrophobic interactions between the unusual signal peptide and the lipid bilayer. After insertion of the amino terminus, this segment of the protein floats in the membrane until it encounters the Sec- 81
Periplasm
Inner Membrane PriA/ [SecEl ATP jSecYl Sec A
Cytosol
Figure 4-11. Working model of the membrane assembly of leader peptidase. machinery which is needed for carboxyl-terminal translocation. Translocation of the large carboxyl-terminal domain then occurs in a very complicated step, requiring ATP hydrolysis and the membrane potential. We suspect that the reason leader peptidase can avoid aggregation before membrane assembly is that it is targeted to the membrane immediately after synthesis of apolar domain 1. This speculative model is currently being tested by examining when the amino terminus and the large carboxyl-terminal domain of nascent leader peptidase polypeptide chains cross the membrane. CHAPTER V
MAPPING OF CATALYTICALLY IMPORTANT DOMAINS IN ESCHERICHIA COLI LEADER PEPTIDASE
5.1. INTRODUCTION Secreted proteins are initially made as a precusor with a short N-terminal leader peptide which signals the protein to enter the secretory pathway. The leader peptide is removed by a membrane protease which recognize a conserved feature of the leader peptide (von Heijne, 1986b). Two types of protease which cleave the leader peptide have been found so far: leader peptidase and lipoprotein signal peptidase. Of the two, leader peptidase processes most of the preproteins in E. coli. Leader peptidase is a inner membrane protein (Wolfe et al., 1983) which has 323 residues with a molecular weight of 37,000 Da. This enzyme is known to be insensitive to all known protease inhibitors (Zwizinski et al., 1981).
The topogenesis of this protein within the memtrane is well studied (Wickner,1988). Figure 5-1 shows the membrane topology of leader peptidase (Moore and Miura, 1987; von Heijn, 1989). This protein is composed of 4 major domains: (1) first hydrophobic transmembrane domains H1, (2) second hydrophobic domain H2, which is known as an internal signal sequence, (3) a highly charged 83 84
periplasm
cytoplasm
1 10 20 30 40 50 manmfalilviatlvtgilw cvdkfffapkrrfroaaaoaardsldkatl
KKVAPKPGWLETGASVFPVLAIVLIVRSFIYEPFOIPSGSMMPTLLIGDP
Figure 5-1. Domain structure of leader peptidase. 85 cytoplasmic domain P1 between H1 and H2, (4) a large periplasmic C-terminal domain P2. Both H1 and H2 are essential for the correct topology of this protein, and the charge balance around these two domains are known to play an important role in determining the topology (Dalbey and Wickner, 1987; Laws and Dalbey, 1989; von Heijne, 1989; von Heijne et al., 1988).
We studied the relationships between the structure and and the catalytic activity of the leader peptidase. First we developed jn vitro and in vivo assay systems which monitors the processing of pro-OmpA to mature OmpA and of M13 procoat to mature coat protein, respectively. Then we made a large panel of leader peptidase mutants and determined the kinetics of the processing in these systems. Our data suggest that H1 and P1 are not directly involved in catalysis, but that H2 and and the region immediatly following H2 may be in contact with the signal peptide and/or located spatially close to the active site of this enzyme.
5.2. RESULTS We used a E. coli strain IT41 which has a temperature- sensitive mutation in the chromosomal leader peptidase gene (Inada et al., 1989) to eliminate the background activity due to the chromosomal copy of leader peptidase. Figure 5-2 (top panel) shows slow processing of pro-OmpA to OmpA at non-permissive temperature (42oC) with a characteristic processing time t* (= 86
no plasmid
P m
chase (sec) 10 20 30 40 50 60 90 120
pRD8
chase (sec) 10 20 30 40 50 60 90 120
Figure 5-2. In vitro processing of pro-OmpA in strain IT41 at the non- permissive temperature with and without the pRD8 plasmid expressing wild-type leader peptidase. 87
Table 5-1. Catalytic Activity of Leader peptidase mutants
Mutant in vitro t* (s) (s) Growth at 42°C —arabinose + arabinose —/+ arabinose
Wild-type + 8 7 4- + / + + No plasmid — 90 94
A5-22 — 133 118 -/- A5-22,V6l — 16 7 4- + /+ + (A L )8 — 44 30 + / + A62-65 + 8 7 + + / + + A62-68 + 22 10 + + /+ + 76-VLIV-77 ND 25 8 + + / + + A4-50 — 13 7 -1- +I+ + A30-52 — 20 7 + + / + + A51-57 + 10 10 + + /+ + L77 — 74 42 -/- K77 ND 35 10 + + /+ + E l l ND 39 13 + /+ + N77 — 91 33 - / - R74.N77 — 45 16 + / + N77.R79 — 91 33 -/- 76-PG-77 ND 68 38 - / - 76-N-77 ND 44 12 + / + 4-D-5 + 10 7 + + / + + 4-K-5 — 67 18 -/-
Column 2 gives the results for in vitro processing of M 1 3 procoat protein. Columns 3 and 4 give the characteristic reaction-timcs r* = l/Aypp for the processing of pro-OmpA to OmpA in v/iw for uninduccu and induced IT41 cells (sec text). Column 5 givjjs the growth characteristics at 42 °C of IT4I cells transformed with the appropriate plasmid, both with and without arabinose present. ND: not determined. 88
1/kapp) of about 100 s. Then IT41 was transfromed with plasmids encoding wild type and various leader peptidase mutants under an ara promotor, and the kinetics of processing was tested. When IT41 expresses wildtype leader peptidase from the pRD8 plasmid (Dalbey and Wickner, 1985; Johnston et al., 1985), t* drops more than 10- fold to about 7 seconds (Fig. 5-2, bottom pannel) , even under the non-induced conditions where the expression level is very low. Table 5-1 summarizes the catalytic activity of various leader peptidase both in vivo and in vitro.
Leader peptidase A5-22, lacking the first hydrophobic domain H1, shows no activity in vivo. H1 plays important role for the membrane assembly of leader peptidase. When this region is deleted, H2 cannot act as a signal for the translocation of the P2- domain (von Heijne e ia l, 1988). No. activity of this mutant could be interpreted as a failure of the translocation of active domain. To check this possibility, translocation of P2 was restored by either changing Glu61 to Val or Arg77 to Leu. Translocation of P2 in these mutants were confirmed by protease mapping. These two mutants, surprisingly, showed full activity. These data indicate that H1 is not required for the leader peptidase activity, even though it is required for the translocation of leader peptidase.
Leader peptidase A62-68, lacking the first half of the second hydrophobic domain H2, and A62-65 have full activity in vivo. 89
However, when the entire H2 was replaced by an unrelated hydrophobic sequence composed of eight (Ala-Leu) repeats, it has very low activity (P2 domain of this mutant was translocated according to protease mapping). When H2 is lengthened by inserting Val-Leu-llu-Val between residues 76 and 77 (76-VLIV-77), this mutant also lost measurable activity in vivo. Theses data suggest that the second hydorphobic transmembrane domain H2 has a strong influence on the catalytic activity, but at least the first half of H2 is not required for the activity.
Leader peptidase A3-52, in which most of the cytoplasmic P1 domain was deleted, has a fairly high activity in vivo and in vitro. Mutant A51-57, where a few residues in front of H2 are deleted, has almost full activity. These mutants are known to assemble normally (Dalbey and Wickner, 1988; von Heijne fitai-, 1988). When both H1 and most of P1 is deleted (mutant A4-50), this leader peptidase also has almost full activity in vivo. This mutant assembles almost normally (Dalbey et al., 1987; Dalbey and Wickner, 1987), suggesting that the cytoplamic domain P1 is relatively unimportant for catalysis.
The mutations at and around the Arg77 has significant influence on the catalytic activity. A5-22.L77 has no measurable activity even though it has correct topology. K77, E77, and N77 have low activities. L77 has lost almost completely the activity. A 90 double mutant R74.N77 has a much higher activity than N77, but another double mutant N77.R79 shows almost no activity (like N77). Mutant 76-PG-77, which has a helix-breaking Pro-Gly dipeptide inserted between Val76 and Arg77, has a very low activity. Mutant 76-N-77 with an insertion of Asn, another helix-breaker, in the same position also shows a major decrease in the catalytic activity. These results might be due to the breaking of the S-helical structure of H2 (Jennings,1989), which may be extended to the Pro83, which is the first helix-breaker downstream of H2. Taken together, these data suggests that the region around Arg77 immediately downstream of H2 is very important for the catalytic activity.
Since leader peptidase is active in vitro (Dalbey and Wickner, 1987) we also wanted to test whether the activities of the different mutants as measured in vivo and in vitro correlate. Briefly, the in vitro assay is based on a lysate of MC1061 cells transformed with the same plasmids as used in the in vivo studies. Radiolabeled phage M13 procoat protein synthesized from a pT712-derived plasmid by an in vitro transcription/translation system is used as a substrate. Typical results obtained with untransformed MC1061 (carrying on the chromosomal leader peptidase gene) and with arabinose induced MC1061 transformed with plasmid pRD8 (earring a wild-type leader peptidase gene) are shown in figure 5-3. With no dilution of the lysate, about 80% of procoat is processed to the mature form under standard conditions when leader peptidase is overproduced from 91
no plasmid pRD8 A5-22 Procoat ■ Coat -
kl ‘T^
Dilution 1 5 10 20 40 1 5 10 20 40 1 5 10 20 40
A62-65 A30-52 N77 Procoat - 4.* I . £l . r - ' -e ^ ; >|P Coat * ymtf. . y-‘ •
Dilution: : 1 5 10 20 40 1 5 10 20 40 1 5 10 20 40
Figure 5-3. In vitro Processing of phage M13 procoat protein in crude extracts of E. coli MC1061 earring plasmids expressing different Lep mutants. 92 pRD8. The background activity from wild-type leader peptidase produced from the chromosomal gene is about 1/40 of this maximal activity. To control for possible differences in expression levels, the amount of each leader peptidase mutant in the cell extract was assayed by immunoblotting (Towbin et al., 1979) using an antiserum to leader peptidase, Figure. 5-4. The level of the mutants tested were found to be within a factor of four of each other.
Cell extracts containing mutants A5-22, A30-52, and N77 all have no more than background activities. An extract containing mutant A62-65, on the other hand, has almost full enzyme activity. The in vitro data for all the mutants ara summarized in Table 5-1. The only apparent differences between the in vivo and in vitro results are for two mutants where H1 has been deleted (A5-22.V61; A4-50) and one with a large deletion in P1 (A30-52), all three of which have high activities in vivo and grow well at 42oC in IT41, but have only background activity in vitro. Since all three are slightly but measurably less active than wild-type leader peptidase in vivo, it is possible that they score as negatives simply because of the limited sensitivity of the in vitro assay. Another possibility is that the processing rates of pro-OmpA and M13 procoat are differently affected by these particular mutations.
5.3. DISCUSSION This is the first mapping study of catalytically important 93
O c
Figure 5-4. Immunoblotting of MC1062 extracts containing different Lep mutants 94
domains in leader peptidase. Even though we used two totally different in vivo and in vitro assay systems involving different substrates, pro-OmpA and procoat, respectively, the results from the two assay agree well.
Our results strongly support a model which has the active site of leader peptidase in the periplasmic P2 domain, spacially close to the surface of the membrane and to the C-terminal end of H2. Even though H1 and P1 is clearly not part of the active site, these regions seems to be required for the full activity since there is a small but measurable decreae in the activity both, in vivo and in vitro, when these regions are removed. This might be due to the alterations in the distance or orientation of the active site with respect to the membrane and the substrate leader peptide.
Our results with the mutations at and around the Arg77 suggest this region immediate downstream of H2 is very important for the catalytic activity. Both the L77 and N77 mutants have very low catalytic activity. Even a conservative replacement of Arg with Lys seriously impair the activity. Leucine or asparagine at this position may pull residues 77-81 into the membrane by extending the uncharged H2 segment from residue 77-up to Glu82. This suggests the distance between the active site and the membrane surface and/or the cleavage site in the substrate leader peptide is important for the activity. It also suggests that an 3-helical structure may be required through out the H2 and and its C-terminal flanking region around Arg77, since the helix-breakers introduced between position 76 and 77 significant impaired the catalytic activity. Arg 77 might be primarily involved in the folding of the periplasmic domain or it might form part of the active site.
Our results showing that the second hydrophobic transmembrane domain H2 and the region immediately downstream of it are very important for catalytic activity is consistent with the loop model for the insertion of leader peptides into the bacterial membrane proposed by DiRienzo et al. (1989). A similar dissection of the remainder of the periplasmic P2 domain has not been attempted, since there is no obvious subdomains or stretches of residues that would serve as good targets for mutagenesis. Classical genetic techniques could be. used to search for catalytically defective mutations in this domain. However, it is necessary to have a structural model of this domain to interpret any such mutation. CHAPTER VI
SITE-DIRECTED MUTAGENESIS TO DEFINE THE LIMITS OF SEQUENCE VARIATION TOLERATED FOR PROCESSING OF THE M13 PROCOAT PROTEIN BY THE ESCHERICHIA COLI LEADER PEPTIDASE
6-1. INTRODUCTION Many secreted and membrane proteins are synthesized in a precursor form with an amino-terminal extension peptide of 15-30 amino acid residues. These leader (signal) sequences function to initiate protein translocation across the membrane and are removed by a membrane-bound leader (signal) peptidase during or after protein export.
Despite extensive research over the last decade (Mollay fila i-, 1982; Lively and Walsh, 1983; Jackson and Blobel, 1980), only recently has signal peptidase from several different sources been isolated. Signal peptidase from dog pancreas consists of a complex of five polypeptide chains of molecular mass 25, 23/22, 21, 18 and 12 KD (Evans etaL. 1986). Whereas there are five proteins in canine pancreas, there are only two subunits in the signal peptidase from the egg oviduct (Baker and Lively, 1987). In contrast, each of the two bacterial leader peptidases that have been isolated have one 96 polypeptide chain. Lipoprotein signal peptidase (Innis elaL., 1984) processes several outer membrane proteins, termed lipoproteins, while leader peptidase (Dalbey e ia l, 1986) processes the other outer membrane proteins and all inner membrane and periplasmic proteins (Wolfe fila i-, 1983; Dalbey and Wickner, 1985). This peptidase, whose active site faces the periplasmic side of the membrane (Wolfe s la i-, 1982), has been purified to homogeneity (Zwizinski and Wickner, 1980; Wolfe e ia L , 1982). In addition, its gene has been cloned (Date and Wickner, 1981) and sequenced (Wolfe S lai., 1983), and the enzyme that it codes for overexpressed under inducible/repressible regulation. The physiological role of leader peptidase is to release periplasmic and outer membrane proteins from their membrane anchor (Dalbey and Wickner, 1985).
The E. coli leader peptidase, which has very wide substrate specificity, processes many pre-proteins (Watts g ia L , 1983; Dalbey and Wickner, 1985), including most bacterial exported proteins, yeast pre-acid phosphatase, honey bee pre-promellitin, and human pre-hormones (pre-proinsulin, pre-growth hormone, pre-interferon, etc.). These leader peptidase substrates do not have any sequence homology. Rather, they have a common pattern (von Heijne, 1983; Perlman and Halvorson, 1983) of small amino acids at residues - 1(ala, gly or ser) and -3 (ala, gly, ser, leu, val or ile), and a helix- breaking residue around position -6 (with respect to the cleavage site). Earlier studies have shown that the region immediately preceding the leader peptidase cleavage site is important for processing: (i) Mutations that alter the helix breaker proline to either a leucine or a serine residue block leader peptidase processing but not translocation of beta-lactamase (Koshland slat-, 1982); (ii) Mutations that change the amino acid at -1, -3 or -6 of the M13 procoat protein prevent cleavage (Kuhn and Wickner, 1985); and (iii) Leader peptidase can cleave small peptides, including the 16 amino acid peptide containing residues +5 to -9 of the procoat protein (Dierstein and Wickner, 1986) and the 5 amino acid peptide corresponding to positions -3 to +2 of the precursor to the maltose binding protein (Dev s la l- 1990)
In this report we have directly tested whether the strongly conserved residues at -1, -3 and -6 of the leader sequence represent the substrate specificity of this enzyme. Oligonucleotide-directed mutagenesis was used to generate over 57 mutant procoats with mutations in the region +1 through -6. Most of these proteins were characterized for their in vivo and in vitro processing, as well as for their protein translocation across the membrane. Our results show that the specificity determinants for processing of procoat are located at -1,-3 and -6. Small residues at -1 and -3 and only proline at -6 of procoat were rapidly cleaved by leader peptidase. All mutants with alterations at +1 (except for proline and threonine) were cleavable. In addition, leader peptidase has very broad substrate specificity at -2, -4 and -5 of procoat and can cleave all 99 the procoat mutants at this position. All mutations that we created at the -6 position (apart from glutamine and histidine which were slowly processed) completely abolished processing by leader peptidase in vivo, including glycine or other good helix breakers.
6.2. RESULTS Figure 6-1 shows the sequence of wild-type M13 procoat and the 59 mutations created at positions +1, -1, -2, -3, -4, -5 and -6 near the cleavage site. The mutations are designated by the one- letter code for each amino acid and are classified into three groups, depending on whether they are cleaved, slowly cleaved or not cleaved. In this study, both in vivo and in vitro assays for leader peptidase processing were used to characterize the mutants. With protease-mapping experiments, we verified that those unprocessed mutants were translocated across the. membrane.
Requirements of small residues at -1 and -3 positions. To determine whether leader peptidase requires small residues at positions -1 and -3 (relative to the cleavage site) in its substrates for processing, we isolated mutations at these positions using degenerate oligonucleotides. The specific mutation was identified by dideoxy sequencing of the single-stranded phage DNA containing the procoat gene. The plasmids coding for the procoat mutants, were designated OMXY or OLXY (for Oligonucleotide, Mature or Leader, position X of amino acid Y). To test if the procoat 100
Recognition sequence of leader peptidase
MKKSLULKI flSUflUftTLUPtlLSF ft IftEGOOPflKflftFNSLQftSflTE fv i GYflUfltlUUU I UGftT IG 1 iKLFKKfTSKA
Procoat nutants: not cleaved partially cleaved cleaved LVPMLSFA | AEG LVPMLSFA [ AEG LVPMLSFAjAEG i h u mm mm F PF F P OPEL P C FUStte F L PI DM T I A L S L C O T RT LOT V U K H VS P 0 5 R N TH E S I A S 0 0 S E 8 0
Figure 6-1. The sequence of wild-type M13 procoat and the 59 mutations created at positions +1, -1, -2, -3, -4, -5 and -6 near the cleavage site. 101 mutants were cleaved by leader peptidase, cells expressing these mutants were pulse-labeled with [35s]-methionine for 30 s, chased with cold methionine for 5 s or 1 min, then immurioprecipitated with antibody to procoat. Aliquots were analyzed by SDS-polyacrylamide gel electrophoresis and fluorography. Table 6-2 summarizes the results for mutations at -1. Procoat with either serine or glycine at -1 was almost completely processed, whereas mutant with either Asn, Thr, His, lie, Arg or Phe at -1 was not cleaved. The proline mutant at -1 was partially cleaved. The -3 specificity of leader peptidase was examined in procoat by making eight mutations at this position (Table 6-3). Only procoat with a small amino acid (glycine or valine) was efficiently cleaved in vivo: for the bigger amino acid side chains of threonine and leucine, cleavage was retarded; cleavage did not occur with Glu, Arg or Lys residue, or the helix-breaker proline. Typical results from this type of analysis for the -1 and -3 mutants are shown in Figure 6-2.
In parallel experiments, we tested whether the mutant procoats were substrates for leader peptidase in vitro. Each mutant was synthesized in a cell-free transcription/translation system and incubated with leader peptidase for 30 min at 37°C. As can be seen in Table 6-2, the in vitro data correlates well with the in vivo results. For instance, alterations at position -1 with alanine, glycine and serine were processed by leader peptidase in the detergent extracts, while even at very high levels of purified leader OL1S OL1 H OL3G OL3V OL3 L WT
procoat- coat- iftftawaxMa*
Figure 6-2. In vivo processing of -1 and -3 procoat mutants by leader peptidase. 103 peptidase, no processing was found with other substitutions. A similar correlation was found in the -3 position, except in the case of leucine, which was readily cleaved in the in vitro assay. Figure 6-3 shows some representative results obtained for the -1 and -3 mutants in this assay.
Requirement of a proline residue at -6. A characteristic feature of bacterial leader peptides is a helix-destabilizing residue such as proline or glycine between -4 and -6 within the leader peptide. To determine whether the helix-breaking proline at -6 of the procoat leader peptide is essential for cleavage, we created 11 alterations at this position. Almost all of the substitutions at this position blocked in vivo processing (Table 6-4), including glycine, a good helix breaker. One possible explanation for this is that the proline is required for translocation of the protein across the membrane but not for processing. To test this idea, protease- mapping studies were performed. Cells were pulse-labeled for 2 min with [35s]-methionine, converted to spheroplasts, and incubated at 0°C for 60 min with or without proteinase K. Aliquots were then immunoprecipitated with antibody to procoat and outer membrane protein A, and subjected to SDS-polyacrylamide electrophoresis and fluorography. In these experiments, we had less than 20% lysis, as judged by the accessibility of cytoplasmic proteins. Figure 6-4 shows that each of the nonprocessed mutants was accessible to proteinase K, indicating that they were translocated across the O L1S
procoat coat-
Figure 6-3. Representative results of in vitro processing of and -3 procoat mutants by leader peptidase. 1 2 4 5 7 8 10 11 12 O L 6 L O L 6 G O L6W O L 6 R 3ES
procoat —
Figure 6-4. Protease mapping of -6 procoat mutants. 106 plasma membrane.
In contrast to the in vivo processing results, several of the -6 mutant procoats were processed by leader peptidase in detergent extracts, indicating a difference between the two assay systems. This apparent discrepancy was resolved in a leader peptidase titration study (Fig. 6-5). While mutant procoats synthesized in vitro were processed to the mature form at very high levels of leader peptidase, they were not cleaved at low levels, where the wild-type procoat is processed. This data indicates that the mutations at -6 disturb the binding of procoat to leader peptidase, since increasing the concentration of leader peptidase drives processing.
Substitutions at +1 in the mature region. Sequence analysis of pre-proteins does not reveal any conserved residue at this position. To test the +1 specificity of leader peptidase, we created 12 mutations at this position within the procoat protein. Most of the +1 mutant procoats were processed in vivo, as well as in detergent extracts, except for those containing proline, threonine or cysteine (Table 6-5). Protease-mapping studies confirmed that procoat with a proline or a threonine residue, did not translocate across the membrane (data not shown).
Analysis of mutations at -2, -4 and -5. We also directly tested whether there was any specificity for leader peptidase 107
1 2 3 4 5 6 WT
OL6G
OL6 S
OL6 Q
LPase (units) 0 1 10 102 103 104
Figure 6-5. In vitro leader peptidase titration study of -6 procoat mutants. processing at the -2, -4 and -5 positions. All of the 27 mutants that we created at these positions were processed (Tables 6-6,7and 8). However, we found that there were two types of mutants. In the first class, procoat was completely processed to the coat protein. The second class of proteins, which showed delayed kinetics of processing (Fig. 6-6), were poorer substrates. This indicates that while there are no critical specificity determinants for leader peptidase at these positions, certain residues are markedly preferred for processing. 109
O L4DO L 4Q OL4R O L5R
procoat coat
Figure 6-6. Representative results of In vitro processing of -2, -4 and -5 procoat mutants by leader peptidase. 110
6.3. DISCUSSION
We report here a detailed study of the substrate specificity of leader peptidase, using the M13 procoat as a model substrate. This protein contains each of the conserved features found within the C- terminal domain of a leader sequence: a small, apolar amino acid at positions -1 (alanine) and -3 (serine), with respect to the cleavage site, and a helix destabilizing residue (proline) at -6. To determine the substrate specificity of leader peptidase, we created over 57 mutations within procoat from positions +1 to -6 near the leader peptidase cleavage site. In addition to assaying the mutants for in vivo processing and protein translocation across the membrane, we also analyzed most of them using an in vitro system.
In these experiments we found that the specificity determinants were located at -1 and -3 with respect to the cleavage site of the procoat protein. The small residues alanine, serine and glycine were processed at -1, whereas the larger residues were not processed, including asparagine, threonine, histidine, isoleucine, arginine, and phenylalanine. This is in excellent agreement with the extensive comparison studies of leader sequences by von Heijne (1983), and Perlman and Halvorson (1983). At -3, serine, glycine, theonine, valine and leucine, which are amino acids found at this position within leader peptides, were processed, although the kinetics of processing of the latter two were slow in vivo (Fig. 6-2). 111
Other amino acids not commonly present at this position within leader peptides were not processed (Table 6-3). The -2, -4 and -5 positions seem to be important for determining the efficiency of the processing. While all these mutants were processed, there were preferences for certain amino acids. There was also broad specificity at the +1 position (except for proline and threonine) of procoat (Table 6-5). Proline is never found in the +1 to -3 region of prokaryotic leader peptides (von Heijne, 1983). In addition, proline and threonine residues are known to affect the neighboring peptide architecture, which might explain their effect on the recognition sequence.
In addition to the conserved small apolar amino acids at -1 and -3, bacterial leader peptides have a conserved helix breaker at -6, usually proline or glycine. The proline residue at position -6 of procoat is absolutely essential for rapid in vivo processing by leader peptidase. The substitution of a glutamine for the proline residue slowed processing; histidine severely affected cleavage; and the rest of the changes at this position were not processed at all, including glycine, a good helix breaker (Chow and Fasman, 1974). This result reinforces the importance of the helix breaking residue (Pugsley, 1989) and re-emphasizes the novel properties of proline residues in biology. Proline is unique among the 20 naturally occurring amino acids because it does not have an amide hydrogen, and is found with the highest frequency at beta turns and with very 112 low frequency in alpha helices. In contrast, although glycine is a turn-inducing residue, it can be within an alpha helix, especially if it is surrounded by helix making residues in the procoat leader peptide.
We suspect that procoat is efficiently processed with a proline at -6 because proline is able to break the alpha helical hydrophobic segment of the leader peptide and bring the cleavage site of procoat close to the active site of leader peptidase. This may also occur with glutamine, and to a lesser extent, with histidine residues even though they are good helix breakers. This idea is currently being tested using N-15 NMR spectroscopy on these -6 mutant procoats However, clearly procoat does not have an absolute requirement for a helix breaker at this position to be processed since many mutant created at -6 were processed by leader peptidase in detergent extracts with 105 times the normal levels of leader peptidase, even with good helix formers (Chow and Fasman,1974). We believe that the binding of procoat to leader peptidase is impaired with amino acids other than proline since large amounts of leader peptidase can promote cleavage in vitro (to the same extent as wild-type).
Recently, Laforet and Kendall (1991) showed that a proline was not required for the processing of an idealized leader peptide for alkaline phosphatase; cleavage occurred even when the proline 113 was changed to an alanine. They suggested that proline is normally important not for its affect on protein conformation but rather for its hydrophobicity. Therefore, proline substitutions with other amino acids block processing because these alterations change the hydrophobicity of this region. However, our studies here show that neither alanine, glycine, or tryptophan, which have comparable hydrophobicity (Eisenberg f iia l, 1984), can substitute for proline for processing. Moreover, glutamine, which is slowly processed in viv o , is signifcantly more polar than proline. Thus, we favor the idea that proline is required to alter the conformation of the polypeptide chain surrounding the cleavage site.
An helix breaking residue has also been shown to be important in the processing of eukaryotic preproteins. Nothwehr and Gordon (1989) introduced proline residues at various positions in the leader peptide of pre(Apro)apoA-ll and demonstrated that site of cleavage is optimal at a distance of 4-5 residues from the proline. This is precisely the typical distance of the the carboxyl-terminal region between the end of the apolar core domain and the cleavage site of eukaryotic signal peptides. The reason for this optimal distance is not known. However, as pointed out by Gordon and colleagues, the proline may demarcate the boundary between hydrophobic region of the membrane and the hydrophilic environment of the lumen of the ER. It possible then that removing the helix breaker shifts the membrane-spanning region of the leader peptide and therefore, 1 14 prevents leader peptidase from interacting with the cleavage site of the preprotein on the lumen surface of the membrane.
Fikes M a i (1990) recently reported the sequence requirements at -1 and -3 for efficient in vivo processing of the Escherichia coli maltose-binding protein by leader peptidase. In this study, they found that alanine, glycine, serine, cysteine and threonine were efficiently processed at -1, in good agreement with our results (apart from threonine). At -3, they observed rapid processing of alanine, glycine, serine, cysteine, threonine, valine, isoleucine, leucine and proline. These results are different than ours in several respects. In our study, the in vivo kinetics of processing of the leucine mutant were very slow, even though it was processed in vitro. In addition, the proline mutant was not processed in vivo. These results, showing that there are some variability in the sequence requirements of -1 and -3 for the various substrates, suggest that other factors, such as the substrate conformation, may also be important in processing by leader peptidase.
Our mutagenesis studies here prove that -1, -3 and -6 of the substrate procoat are important residues for leader peptidase binding and catalysis. However, this information is not sufficient to account for all the specificity of leader peptidase. This protease must recognize some specific conformation of the substrates since 115 it cleaves only precursor proteins, not any other proteins. Determination of how it does so should deepen our understanding of how leader peptidase achieves its difficult task. TABLE 6-1 List of Mutagenic Primers
Mutated codon Mutagenic primer
+1 ATCGTCACCCTCNNNAGCGAAAGACAG -1 GTCACCCTCAGCNNNGAAAGACAGCAT -2 ACCCTCAGCAGCNNNAGACAGCATCGG -3 CT C AGC AGCG AAN NNC AGC AT CGG AAC -4 AGCAGCGAAAGANNNCATCGGAACGAG -5 AGCGAAAGACAGNNNCGGAACGAGGGT -6 G AAAG ACAGC ATMMNAACG AGGGT AGC
Altered codons are underlined. 117
TABLE 6-2
In vivo and in vitro processing of procoat with alterations at -1.
Substitutions % Processed in vivo in vitro
Ala 100 +
Ser 91 +
Gly 82 n.d.
*Asn no
Thr 4
His 0
Pro 51 + /-
Leu 0 n.d.
* i le
Arg 0
Phe 0 n.d. 118
TABLE 6-3
In vivo and in vitro processing of procoat with an alteration at -3
Substitution % Processed in vivo in vitro
Ala 100 nd
Ser 91 +
Gly 100 +
Thr 39 +
Pro 0 -
Val 71 +
Lys 0 -
Gin 0 —
* Leu 27 +
Arg 0 n. 119
TABLE 6-4
In vivo and in vitro processing of procoat with alterations at -6
Substitutions % Processed
in vivo in vitro
Pro Yes Yes
Gly 0 Yes
Cys 0 Yes
His No Yes
Ala No n.d.
* Gin partially Yes
Leu 0 No
Glu No No
Phe No Yes
Arg 0 No
Trp 0 No 120
TABLE 6-5
In vivo and in vitro processing of procoat with alteration at +1
Substitution % Processed in vivo in vitro
Ala 100 +
Ser 100 +
Cys 16 n.d.
Thr 94 ?
Phe 75 n.d.
Val 92 +
Asp 100 +
Pro 0 _ 121
TABLE 6-6
Processing of Procoat with an. alteration at -2
Substitution % Processed in vivo in vitro
Asp 100 n.d.
Leu 100 +
Met 98 +
Glu 100 +
Pro 81 + 122
TABLE 6-7
Processing of Procoat with alterations at -4
Substitutions % Processed in vivo in vitro
Leu 100 Yes
Gly 100 +
Ser 100 n.d.
Gin 89 +
Asp 91 +
*Pro 0 -
Ala 100 +
Trp 100 n.d.
Glu 65 n.d.
Arg 48 +
His 100 n.d. 123
TABLE 6-8
Processing of procoat with an alteration at -5
in vivo in vitro
Ser 100 n.d.
Leu 100 n.d.
Val 100 n.d. lie 100 n.d.
Thr Yes n.d.
Arg 50 +
Phe 100 n.d.
Glu 100 +
Asp 100 +
Gin 95 + REFERENCES
Baker, R. K. and Lively, M. O. (1987) Biochem. 26, 8561-8567.
Bassford, P.J., Silhavy, T.J., and Beckwith, J.R. (1979) J. Bacteriol. 139, 19-31.
Bilgin, N., Lee, J. I., Zhu, H. Y., Dalbey, R. E., von Heijne, G. (1990)
Blobel, G. (1980) Proc. Natl. Acad. Sci. U. S. A. 77, 1496-1500.
Bochkareva, E. S., Lissin, N. M. and Girsohovich, A. S. (1988) Nature, 336, 254-257
Carlson, M. and Botstein, D. (1982) Cell 28, 145-154.
Chirico, W. J., Walters, M. G. and Blobel, G. (1988) Nature, 332,805- 810.
Chou, P. Y. and Fasman, G. D. (1974) Biochem. 13, 222-245.
Cohen, S. N., Chang, A. C. Y., Boyer, H. W. and Helling, R. B. (1973) Proc. Natl. Acad. Sci. U. S. A 70, 3240-3244.
Collier, D. N., Bankiatis, V., Weiss, J. B. and Bassford, P. J., Jr (1988), Cell, 53, 273-283.
Cover, W. H., Ryan, J. P., Bassford, P. J., Walsh, K. A., Bollinger, J. and Randall, L. L. (1987) J. Bacteriol., 169, 1794-1800.
Crooke, E., Brundage, L., Rice, M. and Wickner, W. (1988a) EMBO J., 7, 1831-1835.
124 Crooke, E., Guthrie, B., Lecker, S., Lill, R. and Wickner, W. (1988b) Cell, 54, 1003-1011.
Dalbey, R. E. (1990) Methods in Cell Biology, in press.
Dalbey, R. E. (1990) Trends Biochem. Sci., 15, 253-257.
Dalbey, R. E. and Wickner, W. (1985) J. Biol. Chem. 260, 15925- 15931.
Dalbey, R. E. and Wickner, W. (1986) J. Biol. Chem. 261, 13844- 13849.
Dalbey, R. E. and Wickner, W. (1987) Science 235, 783-787.
Dalbey, R. E., Kuhn, A. and Wickner, W. (1986) Microbiology- 1986, 234-237; ASM- Washington, D. C.
Dalbey, R.E and Wickner, W. (1988) J. Biol. Chem. 88, 404-408.
Dalbey, R.E, Kuhn, A. and Wickner, W. (1987) J. Biol. Chem. 262, 13241-13245.
Date, T. and Wickner, W. (1981) Proc. Natl. Acad. Sci. U. S. A. 78, 6106-6110.
Davis, N. G. and Model, P. (1985) Cell 41, 607-614.
Davis, N. G., Boeke, J. D. and Model, P. (1985) J. Mol. Biol. 181, 111 121.
Deshaies, R. J., Koch, B. D., Werner-Waxhburne, M., Craig, E. and Schekman, R. (1988) Nature, 332, 800-805.
Dierstein, R. and Wickner, W. (1986) EMBO J. 5, 427-431.
Eilers, M. and Schatz, G. (1986) Nature, 322, 228-232.
Ellis, J. (1987) Nature, 329, 378-379. 126 Emr, S.D., Schwartz, M. and Silhavy, T.J. (1978) Proc. Natl. Acad. Sci. U.S.A 75, 5802-5806.Engelman, D. M. and Steitz, T. A. (1981) Cell 23, 411-422.
Evans, E. A., Gilmore, R. and Blobel, G. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, 581-585.
Fikes, J. D., Barkocy-Gallagher, G. A., Klapper, D. G., and Bassford, P. J. Jr. (1990) J. Biol. Chem. 265, 3417-3423.
Hall, J. D., Gilmore, R. and Lamb, R. A. (1988) J. Cell Biol. 106, 1489- 1498.
Hemmingsen.S.M., Woolford, C., van der Vies, S.M., Tilly, K. Dennis, D.T., Georgopoulos, C.P., Hendrix, R. and Ellis, R.J. (1988) Nature, 333, 330-334.
Hendrix, R. W. (1979) J. Mol. Biol., 129, 375-392.
lino, T., Takahashi, M. and Sako, T. (1987) J. Biol. Chem. 262, 7412- 7417.
Innis, M. A., Tokunaga, M., Williams, M. E., Loranger, J. M, Chang, S. Y., Chang, S., Wu, H. C. (1984) Proc. Natl. Acad. Sci. U. S. A. 81, 3708-
Inouye, S., Soberon, X., Franceschini, T., Nakamura, K., Itakura, K., and Inouye, M. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 3438-3441.
Ito, K., Date, T. and Wickner, W. (1980) J. Biol. Chem. 262, 2123- 2130.
Jackson, R. C. and Blobel, G. (1980) Ann. N. Y. Acad. Sci. 343, 391- 404.
Koshland, D., Sauer, R. T. and Botstein, D. (1982) Cell 30, 903-914.
Kuhn, A. and Wickner, W. (1985) J. Biol. Chem. 260, 15914-15918.
Kuhn, A., Kreil, G. and Wickner, W. (1986) EMBO J. 5, 3681-3 127 Kuhn, A., Zhu, H-Y., and Dalbey, R. E. (1990) EMBO J. 9, 2385-2389.
Kumamoto, C. A. and Beckwith, J. (1985), J. Bacteriol., 163, 267-274.
Kumamoto, C.A. and Gannon, P.M. (1988) J. Biol. Chem., 263, 11554- 11558.
Laforet, G. A. and Kendail, D. A. (1991) J. Biol. Chem. 266, 1326- 1334.
Laws, J. K. and Dalbey, R. E. (1989) EMBO J. 8, 2095-2099.
Li, P., Beckwith, J. and Inouye, H. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 7685-7689.
Lively, M. O. and Walsh, K. A. (1983) J. Biol. Chem. 258, 9488-9495.
Maniatis. T., Fritsch, E. F. and Sambrook, J. (1982) Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N. Y.
Michaelis, S and Beckwith, J. (1982) Ann. Rev. Microbiol. 36, 435- 465.
Miller, J. H. (1972) Experiments in Molecular Genetics, Cold Spring Harbor Laboratory, Cold Spring Harbor, N. Y.
Mollay, C., Vilas, U. and Kreil, G. (1982) Proc. Natl. Acad. Sci. U. S. A. 79, 2260-2263.
Monier, S., Van Luc, P., Kreibich, G., Sabatini, D. D. and Adesnik, M. (1988) J. Cell Biology 107, 457-470.
Moore, K. E. and Miura, S. (1987) J. Biol. Chem. 262, 8806-8813.
Nilsson, I. and von Heijne, G. (1990) Cell 62, 1135-1141.
Osborn, M. J., Gander, J. E., Parisi, E., and Carson, J. (1972) J. Biol. Chem. 247, 3962-3972. Perlman, D. and Halvorson, H. O. (1983) J. Mol. Biol. 167, 391-409.
Randall, L. L. and Hardy, S. J. S. (1986) Cell 46, 921-928.
Rohrer, J. and Kuhn, A. (1990) Science 250, 1418-1421.
San Millan, J. L., Boyd, D., Dalbey, R. E., Wickner, W. and Beckwith, J. (1989) J. Bacteriol. 171, 5536-5541.
Sanger, F., Nicken, S., and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. U. S. A. 74, 5463-5467.
Schnaitman, C. A. (1970) J. Bacteriol., 104, 890-901
Spiess, M. and Handschin, C. (1987) EMBO J. 6, 2683-2691.
Spiess, M. and Lodish, H.F. (1986) Cell 44, 177-185. submitted.
Szczesna-Skopupa, E. Browne, N., Mead, D. and Kemper, B. (1988) Proc. Natl. Acad. Sci U. S. A. 85, 738-742.
Towbin, H., Staehlin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. U. S. A. 76, 4350-4354.
Verner, K. and Schatz, G. (1988) Science 241, 1307-1313.
Von Heijne, G., Wickner, W. and Dalbey, R. E. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 3363-3366. von Heijne, G. (1983) Eur. J. Biochem. 133, 17-21. von Heijne, G. (1985) J. Mol. Biol. 184, 99-105. von Heijne, G. (1989) Nature 341, 456-458. von Heijne, G. and Gavel, Y. (1988) Eur. J. Biochem 174, 671-678. von Heijne, G., Wickner, W. and Dalbey, R. E. (1988) Proc. Natl. Acad. Sci. U. S. A., 85, 3363-3366. 129 Walter, P., & Lingappa, V. (1986) Annu. Rev. Cell Biol. 2, 499-516.
Watts, C., Wickner, W. and Zimmermann, R. (1983) Proc Natl. Acad. Sci. U. S. A. 80, 2809-2813.
Weiss, J.B., Ray, P.H. and Bassford, P.J.Jr. (1988) proc. Natl. Acad. Sci. USA, 85, 8978-8982.
Wickner, W. and Lodish, H. F. (1985) Science 230, 400-407.
Wolfe, P. B. , Silver, P. and Wickner, W. (1982) J. Biol. Chem. 257, 7898-.
Wolfe, P. B. and Wickner, W. (1984) Cell 36, 1067-1072.
Wolfe, P. B., Rice, M. and Wickner, W. (1985) J. Biol. Chem. 260, 1836-1841.
Wolfe, P. B., Silver, P. and Wickner, W. (1982) J. Biol. Chem. 257, 7898-7902.
Wolfe, P. B., Wickner, W. and Goodman, J. (1983) J. Biol. Chem. 258, 12073-12080.
Yamana, K. and Mizushima, S. (1988) J. Biol. Chem. 263, 19690- 19696.
Yamane, K., Khihara, S. and Mizushima, S. (1987) J. Biol. Chem. 262, 2358-2362.
Zerial, M., Melancon, P., Schneider, C. and Garoff, H. (1986) EMBO J. 5, 1543-1550.
Zhu, H-Y and Dalbey, R. E. (1989) J. Biol. Chem. 264, 11833-11838.
Zoller, M. J. and Smith, M (1983) Methods Enzymol. 100, 468-500.
Zwizinski, C. and Wickner, W. (1980) J. Biol. Chem. 255, 7973-