Gene Expression Technology

670 Pages·1991·11.759 MB·English

by  Goeddel D.V., Emr S.D., Gold L. (eds.)

Preface
The articles in this volume were assembled to enable the reader to
design effective strategies for the expression of cloned genes and cDNAs.
More than a compilation of papers describing the multitude of techniques
now available for expressing cloned genes, this volume provides a manual
that should prove useful for solving the majority of expression problems
one is likely to encounter.
The four major expression systems commonly available to most inves-
tigators are stressed: Escherichia coli, Bacillus subtilis, yeast, and mamma-
lian cells. Each of these systems has its advantages and disadvantages,
details of which are found in Chapter 1 and the strategic overviews for the
four major sections of the volume. The papers in each of these sections
provide many suggestions on how to proceed if initial expression levels are
not sufficient.
DAVID V. GOEDDEL
vx
Contributors to Volume 185
Article numbers are in parentheses following the names of contributors.
snoitail~A detsil are current.
SRAL N~lSMHARBA (13), Department of -DiB ANITSIRHC Y. CHEN (35, 37), Department of
molecular Chemistry, Genentech Inc., Cell Genetics, Genentech Inc., South San
South San Francisco, California 94080 ,ocsicnarF California 94080
PAULINA SABLAB (3), Departamento de Bio- ASSENAV MLOHSIHC (35, 37), Department of
logia Molecular, Centro de Investigacion Cell Genetics, Genentech Inc., South San
sobre Ingenieria Genbtica y Biotechnolo- ,ocsicnarF California 94080
gia, Universidad Nacional Autonoma de
A. MARK NAG1C (30), Laboratory of -celoM
Mexico, Cuernavaca, Morelos, Mexico
ular Genetics, National Institutes of
CLINTON E. BALLOU (36), Department of Health, Bethesda, Maryland 20892
Biochemistry and Molecular Biology, Uni-
MAILLIW REGNEVELC (25), Department of
versity of California at Berkeley, Berkeley,
Molecular Biology, lmmunex Corpora-
California 94720
tion, Seattle, Washington 98101
RAMA M. EJAGALEB (8), Division of Molec-
DORIS COlT (28), Chiton Corporation,
ular and Cell Biology, Lilly Research Lab-
Emeryville, California 94608
,seirotaro Indianapolis, Indiana 46285
ENNA LLEB (29), Zymo Genetics Inc., Seat- D. SNESUOC (27), Wellcome, Beckenham,
Kent, England
tle, Washington 98105
OCSICNARF RAVILOB (3), Departamento de HERMAN A. ED BOER (9), suealroG -arobaL
Biologia Molecular, Centro de Investiga- ,seirot University of Leiden, 2300 RA Lei-
cion sobre Ingenieria Genbtica y Biotech- den, The Netherlands
nologia, Universidad Nacional Autonoma DANIEL DIMAIO (45), Department of
de Mexico, Cuernavaca, Morelos, Mexico Human Genetics, Yale University School
DAVID BOTSTEIN (23), Genentech Inc., of Medicine, New Haven, Connecticut
South San Francisco, California 94080 06510
RALPH A. WAHSDARB (33), Department of SAMOHT F. EUHANOD (30), Department of
Biological Chemistry, College of Medicine, Biology, Indiana University, Bloomington,
University of California at Irvine, Irvine, Indiana 47405
California 92 717 WERDNA J. DORNER (44), Department of
YNOHTNA J. EKARB (34), Chiron -aroproC Molecular and Cellular Genetics, Genetics
tion, Emeryville, California 94608 Institute Inc., Cambridge, Massachusetts
SEMAJ R. HCAORB (22), Department of Biol- 02140
ogy, Princeton University, Princeton, New DLANOD J. OKNEBWOD (35), Department of
Jersey 08544 Molecular immunology, Genentech Inc.,
COLIN R. CAMPBELL (41), Department of South San Francisco, California 94080
Genetics, University of Illinois College of NHOJ W. FFRODNEBUD (6), Department of
Medicine, Chicago, Illinois 60612 Biology, Brookhaven National Laboratory,
DIVAD F. LEAHCIMRAC (16), Synergen Inc., ,notpU New York 11973
Boulder, Colorado 80303 NHOJ J. DUNN (6), Department of Biology,
A. SREBMAHC (27), Department of Biochem- Brookhaven National Laboratory, Upton,
istry, University of Oxford, Oxford OX1 New York 11973
3QU, England TTOCS D. RME (21), Division of Biology,
CHUNG NAN CHANG (35), Protein Design California Institute of Technology, Pasa-
Laboratories, Palo Alto, California 94304 dena, California 91125
xi
xii SROTUBIRTNOC OT EMULOV 185
TINA YRREVEHCTE (26), Department ofFer- Biological Sciences, Carnegie Mellon Uni-
mentation Research, Genentech Inc., ,ytisrev Pittsburgh, Pennsylvania 15213
South San Francisco, California 94080
RANDAL J. KAUFMAN (39, 42, 44), Depart-
DAVID V. GOEDDEL (1), Molecular Biology ment of Molecular and Cellular Genetics,
Department, Genentech Inc., South San Genetics Institute Inc., Cambridge, Mas-
,ocsicnarF California 94080 sachusetts 02140
LARRY GOLD (2, 7), Department of Molecu- YLREBMIK YASLEK (29), Department of -DiB
lar, Cellular and Developmental Biology, chemistry, University of British Columbia,
University of Colorado at Boulder, Vancouver, British Columbia V6K 157,
Boulder, Colorado 80309 Canada
NASUS NAMSETTOG ( 11 ,) Laboratory of Mo-
RICHARD L. KENDALL (33), Department of
lecular Biology, National Cancer Institute,
Biological Chemistry, College of Medicine,
National Institutes of Health, Bethesda,
University of California at lrvine, Irvine,
Maryland 20892
California 92717
JAY D. GRALLA (4), Department of Chemis-
WAYNE A. KEOWN (41), Department of -eG
try and Biochemistry and the Molecular
netics, University of Illinois College of
Biology Institute, University of California
Medicine, Chicago, Illinois 60612
at Los Angeles, West Los Angeles, -rofilaC
nia 90024 A. J. KINGSMAN (27), Department of Bio-
chemistry, University of Oxford, Oxford
SICNARF J. GRANT (29), Zymo Genetics
OX1 3QU, England
Inc., Seattle, Washington 98105
ROBERT HAMILTON (35), Department of S. M. NAMSGNaK (27), Department of -DiB
Fermentation Research and Process De- chemistry, University of Oxford, Oxford
OX1 3QU, England
velopment, Genentech Inc., South San
Francisco, California 94080 TADAHIKO KOHNO (16), Synergen Inc.,
Boulder, Colorado 80303
DENNIS J. HENNER (5, ,71 20), Department
of Genetics, Genentech Inc., South San MAILLIW J. KOHR (35), Department of Pro-
,ocsicnarF California 94080 tein Chemistry, Genentech Inc., South San
RONALD A. HITZEMAN (35, 37), Depart- ,ocsicnarF California 94080
ment of Cell Genetics, Genentech Inc., MICHAEL KRIEGLER (40), Cetus Corpora-
South San Francisco, California 94080 tion, Emeryville, California 94608
NYRHTAK J. HOFMANN (24), Department of RAJU S. ITAPALREHCUK (41), Department of
Virus and Cell Biology, Merck Sharp and Genetics, University of Illinois College of
Dohme Research Laboratories, West Medicine, Chicago, Illinois 60612
Point, Pennsylvania 19486
STUART F. J. LE GRICE (18), Department of
SEMAJ E. HOPPER (24), Department of Bio- Infectious Diseases, Case Western Reserve
logical Chemistry, The Milton S, Hershey University School of Medicine, Cleveland,
Medical Center, Pennsylvania State Uni- Ohio 44106
,ytisrev Hershey, Pennsylvania 17033
ARTHUR D. NOSNIVEL (38), Department of
BRUCE H. HORWITZ (45), Department of Cell Genetics, Genentech Inc., South San
Human Genetics, Yale University School ,ocsicnarF California 94080
of Medicine, New Haven, Connecticut
CntmG LIU (35), Department of Protein
06510
Chemistry, Genentech Inc., South San
ANNA S. HUI (9), Department of Pharmacol- Francisco, California 94080
ogy and Cell Biophysics, University of Cin-
HONG MA (23), Division of Biology, -rofilaC
cinnati, Cincinnati, Ohio 45267
nia Institute of Technology, Pasadena,
HTEBAZILE W. JONES (31), Department of California 91125
CONTRIBUTORS TO VOLUME 185 xiii
VIVIAN L. MACKAY (29), Zymo Genetics lecular and Cell Biology, Lilly Research
Inc., Seattle, Washington 98105 Laboratories, Indianapolis, Indiana 46285
JENNIE P. MATHER (43), Department of Cell RONALD G. SCrtONER (8), Division of Mo-
Culture Research and Development, Gen- lecular and Cell Biology, Lilly Research
entech Inc., South San Francisco, Califor- Laboratories, Indianapolis, Indiana 46285
nia 94080
LOREN D. SCHULTZ (24), Department of
SAMOT MOKS (12), Department of Biochem- Virus and Cell Biology, Merck Sharp &
istry, The Royal Institute of ,ygolonhceT Dohme Research Laboratories, West
S-100 44 Stockholm, Sweden Point, Pennsylvania 19486
ECNERWAL M. MYLIN (24), Department of ANDRZEJ Z. IKSWElZDELS (29), Zymo -eG
Biological Chemistry, The Milton S. Her- netics Inc., Seattle, Washington 98105
shey Medical Center, Pennsylvania State
SAMOHT J. YVAHLIS (15), Department of Bi-
,ytisrevinU Hershey, Pennsylvania 17033
ology, Princeton University, Princeton,
AHTNASAV NAGARAJAN (19), Central Re- New Jersey 08544
search and Development Department, E. L NANCY J. NOSPMIS (35, 37), Department of
du Pont de Nemours and Co., Wilmington, Cell Genetics, Genentech Inc., South San
Delaware 19880 ,ocsicnarF California 94080
BJORN NILSSON (13), Department of Bio- ARJUN SINGH (35), Department of Cell -eG
chemistry and Biotechnology, Royal Insti- netics, Genentech Inc., South San Fran-
tute of Technology, S-100 44 Stockholm, ,ocsic California 94080
Sweden
SAERDNA SOMMER (16), Biogrowth Inc.,
PETER O. OLINS (10), Biological Sciences Richmond, California 94806
Department, Monsanto Company, St. JOAN A. STADER (15), School of Basic Life
Louis, Missouri 63198 Sciences, University of Missouri, Kansas
ROGER PAI (35), Department of Product Re- City, Missouri 64110
,yrevoc Genentech Inc., South San Fran- C. A. YAWNATS (27), Institute of Molecular
,ocsic California 94080 Medicine, Headington, Oxford, England
AINIGRIV L. PRICE (25), Department of Mo- TIM SNRAETS (23), Department of Biochem-
lecular Biology, Immunex Corporation, istry and Biophysics, University of -rofilaC
Seattle, Washington 98101 nia, San Francisco, San Francisco, Cali-
fornia 94143
SrIAUKAT H. RANGWALA (10), Biological
Sciences Department, Monsanto Com- GARY D. OMROTS (7), Department of Molec-
pany, St. Louis, Missouri 63198 ,ralu Cellular and Developmental Biology,
University of Colorado at Boulder,
MARK E. RENZ (35), Department of -leveD
Boulder, Colorado 80309
opmental Biology, Genentech Inc., South
San Francisco, California 94080 F. WILLIAM STUDIER (6), Department of Bi-
ology, Brookhaven National Laboratory,
ALAN B. ROSE (22), Department of ,ygoloiB
,notpU New York 11973
Princeton University, Princeton, New Jer-
sey 08544 WAYNE E. TAYLOR (25), Department of
Chemistry and Biochemistry, California
ALAN H. GREBNESOR (6), Department of Bi-
State University, Fullerton, California
ology, Brookhaven National Laboratory,
92634
,notpU New York 11973
AICIRTAP NOSLO-PMAKET (28), Chiron Cor-
NEVETS GREBNESOR (28), Protos -aroproC poration, Emeryville, California 94608
tion, Emeryville, California 94608
ROnERT C. NOSPMOHT (16), Synergen lnc.,
ETTIGIRB E. RENOHCS (8), Division of Mo- Boulder, Colorado 80301
xiv CONTRIBUTORS TO VOLUME 185
SAIHTAM N~ILHU (12), Department of Bio- lac Chemistry, College of Medicine, Uni-
chemistry, The Royal Institute of -lonhceT versity of ainrofilaC at lrvine, lrvine, -ilaC
,ygo S-IO0 44 Stockholm, Sweden fornia 92717
K~ITH D. NOSNIKLIW (32), Department of LEINAD G. ARUSNAY (5, 14), Department of
,yrtsimehcoiB Emory University School of
Cell ,sciteneG Genentech Inc., South San
Medicine, Atlanta, aigroeG 22303
,ocsicnarF ainrofilaC 08049
M. WILSON (27), Delta Biotechnology Not-
tingham, England ILRAC YIP (29), Zymo Genetics Inc., Seattle,
notgnihsaW 50189
S1LRAM NOTGNIHTROW (25), Department of
Biochemistry, University of ,notgnihsaW NOTLE T. YOUNG (25), Department of Bio-
Seattle, Washington 59189 chemistry, University of Washington,
Rvo ADAMAY (33), Department of -igoloiB Seattle, Washington 59189
1 SYSTEMS FOR HETEROLOGOUS GENE EXPRESSION 3
1 Systems rof Heterologous Gene Expression
By DAVID .V GOEDDEL
This volume concentrates on the four major expression systems that
are commonly available to most investigators: Escherichia coli, Bacillus
subtilis, yeast, and mammalian cells. Each of these systems has its advan-
tages and disadvantages, some of which I briefly outline in this introduc-
tory article.
One expression system which is not covered in this volume is the
baculovirus system in insect cells since it has been primarily developed by
one group, and a detailed manual describing the system is available from
that laboratory.t There are also examples in the literature of heterologous
gene expression in a variety of other organisms, including Streptomyces, a
number of fungi, pseudomonads, and others. Although these systems
might become relatively more important in the future, we feel their utility
has not at present been demonstrated broadly enough to recommend their
routine use for heterologous protein production.
In general, the expression of each cDNA or gene presents its own
peculiar set of problems that must be overcome to achieve high-level
expression. The synthesis of foreign proteins is still largely empirical. There
is no set of hard-and-fast rules to follow. In fact, a particular protein is
almost as likely to be the exception as it is to follow any set of rules.
Keeping this caveat in mind, I will make some generalizations which I
hope will aid the reader in selecting an initial expression system.
Types of Proteins to Be Expressed
For the purpose of gene expression in heterologous cells, proteins can
be arbitrarily grouped into four broad classes. The first class covers small
(less than -80 amino acids) peptides. These are most easily expressed as
fusion proteins, usually in E. coll. The second class are polypeptides that
are normally secreted proteins (e.g., enzymes, cytokines, hormones) and
range in size from about 80 to 500 amino acids. This class of proteins is
often the most straightforward to express (in all four systems) and secretion
or direct expression should be considered the method of choice. In particu-
lar, direct expression in E. coli has proved extremely effective for the subset
of proteins in the 100-200 amino acid size range. A third class consists of
very large (greater than about 500 amino acids) secreted proteins and cell
i V. A. Lukow and M. D. Summers, Bio/Technology6, 47 (1988).
thgirypoC © 0991 by cimedacA Press, Inc.
METHODS IN ,YGOLOMYZNE VOL. 581 llA rights of reproduction in any form .devreser
4 INTRODUCTION 1
surface receptor proteins. Unless the protein of interest is of microbial
origin, one generally has the most success with this class using a mamma-
lian cell expression system. The fourth group encompasses all nonsecreted
proteins larger than -80 amino acids. Many proteins fall into this class;
however, relative to secreted proteins, much less work has been directed
toward overexpression of proteins in this category. Therefore, the selection
of an appropriate expression system should be based on the intended use
for the protein (see below).
For What Purpose Is the Expressed Protein Needed?
Probably the most common reason for expressing a new gene or cDNA
has been to verify that the correct sequence has, in fact, been isolated. If
verification is all that is required, and if the protein being expressed is from
a higher eukaryote, then transient expression in mammalian cells is the
preferred expression route. Transient expression is not only easy to per-
form, but also gives answers quickly and has a very high probability of
yielding a biologically active protein. Historically, E. coli has also been
used with great success for identification of mammalian proteins, often
yielding biologically active protein, and, nearly always, immunoreactive
material. For verification of microbial proteins, the generalization "make
yeast proteins in yeast and bacterial proteins in bacteria" can be made.
A relatively large amount of substantially pure protein is desirable for
in vivo biological experiments or structural determination. In this case, it is
worth taking the time and effort to evaluate more than one expression
system to find the optimal one for that protein. Bacteria, yeast, and mam-
malian cells have all been used to produce clinical grade proteins in large
amounts.
Escheriehia coli can be considered a good bet for the preparation of
proteins, or portions thereof, to be used for the generation of antibodies. It
is usually possible to express large quantities of antigen rapidly in E. coli,
even though the overexpressed protein may not be correctly folded. In
Section II of this volume there are articles on E. coli that outline, in
addition to the old and still reliable trp vector systems, various novel
methods for the direct and fusion expression of polypeptides suitable for
antibody preparation.
An increasingly common need for the expression of cloned genes is to
obtain mutagenized protein for structure-function experiments (see 45
in this volume). The selection of the appropriate expression system for
such studies is usually based on the ease of generating many defined
mutants in an assayable form. Therefore, E. coli is an obvious first choice.
1 SYSTEMS FOR HETEROLOGOUS GENE EXPRESSION 5
However, if E. coli does not yield protein suitable for analysis, other
systems must be examined. Obviously, if posttranslational modification
contributes to the properties being studied, or if one is looking for pheno-
typic changes in the host cells, the selection of an expression system should
mimic the natural source of the protein as closely as possible.
Escherichia coli Expression Systems
Considerable lore has accumulated over the years concerning E. coil
expression. It is often assumed that any protein can be produced in E. coli
as long as it is not too small, too large, too hydrophobic, and does not
contain too many cysteines. To some extent, these generalizations are
correct; if one wants to express Factor VIII or a complex mammalian cell
surface receptor containing 40 disulfides, E. coli is not the proper host.
However, E. coli is a suitable, and often desirable host, for a large number
of other expression needs. Escherichia coil systems for direct expression,
fusion protein expression, and secretion, each of which has its particular
advantages (and disadvantages), have been developed and refined. While it
might be difficult to predict a priori the relative difficulty of purification for
proteins produced by these three methods, there are usually other reasons
for selecting one approach over the others.
Fusion protein expression strategies ensure good translation initiation
and often permit one to overcome the instability problems that can be
encountered with small peptides. Furthermore, the high degree of certainty
of expression by the fusion approach makes this a preferred method for
generating immunogens. The fusion methods described in this volume also
have purification advantages built into them, including insolubility of the
fusion protein and purification protocols tailored to the fusion partner.
Direct expression is usually the ideal method for E. coli production of a
heterologous protein of 100-300 amino acids, assuming there is not an
inordinate number of cysteines.
The reducing environment present in E. coli does not permit cysteine-
rich proteins to form the disulfide bonds required for proper conformation.
This problem can sometimes be overcome by secretion into a more oxi-
dized environment. However, secretion from E. coli is still largely a hit-or-
miss proposition and is most likely to work with a naturally secreted
protein (as opposed to trying to engineer the secretion of an intraceUular
protein). There are many beneficial features to secretion: Problems with
unwanted N-terminal methionines are avoided, proteins tend to fold better
than during intracellular expression, and protein purification can often be
simplified.
6 NOITCUDORTNI 1
Bacillus subtilis and Yeast Expression Systems
Although B. subtilis and yeast are not usually thought of as first choices
for heterologous gene expression, there are advantages of each that make
them good candidates for specific expression needs. These advantages are
discussed by Henner 17 and Emr 21 elsewhere in this volume. In
particular, B. subtilis is a good choice for obtaining the secretion of a
prokaryotic protein. Yeast offer many important advantages of both pro-
karyotic and eukaryotic systems. Yeast grow rapidly, achieve high cell
densities, and are able to make posttranslational modifications that bacte-
rial systems are unable to perform. Many advances in yeast expression
have helped place yeast on a more equal footing with E. coli as an initial
choice for an expression system.
Mammalian Cell noisserpxE smetsyS
Progress in mammalian cell expression systems has been nothing short
of astonishing in the last few years. Transient expression systems, which
initially permitted the verification that a desired gene had been cloned,
have become increasingly popular in a wide range ofcDNA cloning strate-
gies. Furthermore, transient expression assays allow one to evaluate rapidly
the relative merits of different vectors for subsequent use in stable expres-
sion systems. Stable mammalian cell expression is more time-consuming
than any of the other systems described in this volume; however, stable
systems are essential for large-scale protein production in mammalian
cells. The time and effort required for selection, amplification, and growth
of stable cell lines are more than offset by the great versatility of mamma-
lian cells in producing all of the different classes of protein described above;
there are virtually no examples of proteins that cannot be made in at least
detectable levels in mammalian cells.
Synthetic DNA
It si important to emphasize the benefits that can be gained by taking
full advantage of the power of synthetic DNA. In making expression
constructions, one should not be constrained by the DNA sequence or
restriction sites present in the DNA to be expressed. Codon usage, poten-
tial RNA secondary structure, and ribosome-binding-site sequences can be
changed quite easily by the use of synthetic DNA, and the potential value
of such alterations at improving expression should not be underestimated.
It si worth considering as many variables as possible in designing an
optimal expression vector for the gene of interest. If the existing sequences
1 SYSTEMS FOR HETEROLOGOUS GENE EXPRESSION 7
cannot be easily manipulated, replace and/or alter with synthetic DNA as
much sequence as is necessary to achieve the "idealized" construction.
Future Developments
It is interesting to try to predict what surprises the next 5-10 years
might hold for the heterologous expression of cloned genes. It is likely that
there will be continued improvements made in the four major expression
systems covered by this volume. Perhaps microbial hosts will be engineered
so that they are able to perform specific mammalian posttranslational
modifications such as the addition of complex sugars. It is even conceiv-
able that each of the four host systems described here might be sufficient
for any expression need, thereby eliminating the necessity of having access
to more than one system. And finally, the recent demonstrations that
heterologous proteins can be produced in the milk of transgenic animals
makes it intriguing to speculate that even this technology might become an
important component of the repertoire of a typical gene expression labora-
tory.