<<

. 20
( 61 .)



>>

Mono S chromatographies are performed at each pH. Flow-through fractions are
collected and bound proteins eluted with a O-1 M NaCl gradient developed over
20 mL. The Rl elution profile is analyzed by Western blotting. At neutral pH Rl
bound to the Mono Q column, but this is of limited use in purification as the
protein does not elute as a distinct peak and was detected in a number of fractions
Conner
126
across the salt gradient RI IS rapidly lost from solutton, either by degradation or
precipitation (see Note 1 l), at acrd or alkali pH
6. The binding of Rl to a variety of affinity matrices should be analyzed using a
small-scale batch procedure A variety of matrrces, mcludmg heparm aftigel,
phosphocellulose, hydroxylapatite, Ctbacron blue, reactive red and reactive yel-
low agaroses and ATP agarose are tested for Rl binding by incubation with the
35% (NH&SO, fraction (see Note 12). All mampulattons are performed m 1 5-mL
Eppendorf tubes. Matrices in powder form are swollen using HEPES buffer for 2
h and washed extensively before use Matrices provided as suspensions are also
washed extensively m HEPES buffer One hundred mtcrohters of a 50% slurry of
matrix IS incubated with 200 uL of (NH&SO, fraction for 30 mm at 25°C with
constant agrtatton. The supernatant was removed after a brief centrtfugatron
and retained Matrices are washed three times with 1 mL of HEPES buffer and
bound proteins were eluted by boiling m SDS-PAGE sample buffer Superna-
tants and bound proteins are analyzed by SDS-PAGE and Western blottmg
(Fig. 2) and, by comparison of these fractions, matrices useful m Rl purtficatton
can be identified. Rl binds to Cibacron blue and reactive red agaroses and hep-
arin affigel but shows no specific interactions wrth the other matrices Interactton
with a matrix 1s assumed to be spectfic when the amount of Rl in the bound
fraction is greater than the amount remaining m the supernatant Frve mtlllhter
columns of Ctbacron blue agarose and heparin affigel are tested for Rl bmdmg
and elutton condrttons assessed usmg O-l and 0-2M gradients of NaCl Protein
elution is analyzed by SDS-PAGE. A combination of these columns results m Rl
preparations of greater than 90% purity Crbacron blue chromatography should
be used first as RI eluted from this matrix as a broad peak between l-2M NaCl
Rl containing fractions, identified by SDS-PAGE, are dialyzed overnight against
HEPES buffer and applied to the heparm affigel column Rl elutes from this
column as a sharp peak at 150 mMNaC1 (see Note 13)
7. An alternative method IS used for Rl purification that utthzes the specific mter-
action of thus protein with R2, an interaction that 1s disrupted by a pepttde,
YAGAVVNDL, corresponding to the R2 mteractron sue (2) R2 IS rmmobrhzed,
at 1 mg/mL, to CNBr-activated sepharose and the matrix is incubated with the
35% (NH&SO4 RI fraction for 20 mm at 25°C Five mrlhhters of affinity matrix
is used and purrficatton performed by batch procedures in a 20-mL universal
The supernatant IS removed after a brief centrifugatron at 2500g and the matrix
washed three ttmes wtth 10 vol of HEPES buffer, with 2MNaCl Rl elutton IS
achieved by incubation of the matrtx, for 1 h at 25”C, m a 2-m solution of
peptrde Peptrde IS removed by gel filtration on a lOO-mL superose 12-FPLC
column This method of purrficatron IS suitable for producing small amounts
(l-2 mg) of purified Rl (see Note 14).

4. Notes
1. Lysozyme extractron of soluble proteins from E colz is improved if bacteria are
frozen overnight Protease inhibitors can be added at this point tf the protein of
127
Protein Purification




Fig. 2. Western blot, probed with Rl specific antiserum showing Rl interactions
with a variety of affinity matrices. Odd and even numbered lanes show supernatants
and bound proteins respectively. Lanes 1 and 2, Cibacron blue agarose, 3 and 4 reac-
tive red agarose, 5 and 6 heparin affigel, 7 and 8 reactive yellow agarose, 9 and 10
ATP agarose, .l 1 and 12 hydroxylapatite, and 13 and 14 phosphocellulose. Rl is indi-
cated by an arrow.

interest is susceptible to proteolytic degradation. Inhibitors can be tested indi-
vidually to determine those that prevent degradation or a cocktail of inhibitors,
specific for a wide variety of proteases, can be used. Advice on the selection and
use of protease inhibitors is available from suppliers, e.g., Boehringer-Mannheim.
2. Other methods of cell lysis exist including sonication and the use of a French
pressure cell. Lysozyme extraction, however, requires no additional equipment.
3. In some instances proteins overexpressed in E. coli form insoluble aggregates
that will be lost during centrifugation. Induction of protein expression in E. coli
at lower temperatures reduces the formation of these aggregates (4).
4. Lysozyme extraction also releases nucleic acids and the crude supematant may
be viscous. Nucleic acids can be precipitated at this point by streptomycin sulfate
(Sigma). To 5 mL of supematant, 0.3 mL of 5% streptomycin sulfate is added
and, after 20 min incubation on ice the precipitated nucleic acids are removed by
centrifugation at 18,OOOgfor 20 min. Proteins that interact with nucleic acids will
be precipitated by this method; protein-nucleic acid interactions usually can be
disrupted by the inclusion of lMNaC1, which must be removed, either by dialy-
sis or desalting (see Note 5) before continuing with the purification.
5. (NH&SO4 precipitation is a useful first step in a purification as it greatly reduces
the volume of the extract to be used in subsequent steps. Prior to any column
chromatography steps it may be necessary to dilute out residual (NH,)*S04 or
remove it by dialysis or desalting. Desalting is the method of choice as it can be
performed rapidly on a number of commercially available columns, however, the
volume applied per run is restricted to l-3 mL. Dialysis is generally performed
overnight at 4°C against a volume of buffer that is at least IO-fold greater than
that of the sample. Both dialysis and desalting invariably result in the loss of
some protein. Occasionally, (NH&SO4 precipitation can cause nonspecific
Conner
728
aggregatton of proteins and this can be prevented by the mclusion of detergents
(e g., 0 1% NP40 or 4 mM CHAPS) m the buffers.
6 The volume over which a pH or salt gradient is developed depends on a number
of factors, includmg the size of column and the elution profile of the protein of
Interest but it is recommended that a gradient IS developed over a mimmum of
five times the column volume. The size of fractions collected durmg elution from
the column also varies according to mdividual protocols but, m general, smaller
fractions improve protein peak resolution.
7 At pH 5 3, R2 did not bind to either Mono Q or Mono S columns and this was
mitially considered for use m purification However, several E colz proteins also
shared this property and a pH slightly above the pI was chosen At a pH slightly
below the PI, R2 would bmd weakly to a Mono S column,
8. Manufacturers of ion-exchange and other matrices provrde data on the protein
binding capacity of the resin Information on recommended runnmg conditions,
and buffers, for elution, regeneration, and cleaning are also provided In any sepa-
ration run less protem than the maximum capacity should be applied to the
column to avoid saturation of the matrix
9 Although R2 elutes as a distinct peak, the protem elution trails mto later frac-
tions. Inclusion of 5-10% glycerol in the buffers improved elution of R2 by
preventing this trailing.
10 Purity can be estimated by densitometry of SDS polyacrylamide gels or accu-
rately determined by amino acid analysis. It is also advisable to confirm the iden-
tity of the purified protem by either Western blotting using a number of specific
antisera or by N-terminal sequencing
11. Some proteins are known to precipitate at, or close to then isoelectrtc point and
this may have happened with Rl . Rapid degradation also was observed,
12. A large number of affimty matrices are available from suppliers such as Sigma,
Pharmacia, or Bio-Rad Lectm-affinity chromatography is a useful purification,
step if the protein is a glycoprotem
13. Hydrophobicity interaction chromatography was considered for Rl purification,
but the protein precipitated at the high salt concentrations was used to promote
interactions with the matrix. Concentrations of salts, such as (NH&Sob, NaCl,
or KzS04, in excess of 2M stabilize mteractions of hydrophobic regions of pro-
teins with hydrophobic groups, such as phenyl-, octyl, or butyl, immobihzed on
the matrix and bound proteins can be eluted by lowering the salt concentration.
The concentrations of (NH4)$04, NaCl, or K2S04 m Rl preparations were
adjusted to between 2 and 4A4 by the careful addition of finely ground salts and
the solution centrifuged at 13,000g for 10 mm Supernatants were analyzed by
SDS-PAGE and, m all cases, Rl was precipitated. At salt concentrations below
2M, Rl did not interact with the hydrophobic mteraction matrices tested HIC IS
a useful technique to consider followmg salt elution of proteins from another
matrix; the protem of Interest may bmd dtrectly to an HIC matrix at thts con-
centration or, the salt concentration can be increased directly to achieve
bmdmg.
129
Protein Purification
14 Afftmty chromatography, using specific protein-protem or protein-nucletc acids
interactions, 1san excellent method for producing small amounts of highly puri-
fied proteins. Strong mteractrons (such as antibody-antigen) may require extreme
conditions, such as htgh salt, denaturatton, or extremes of pH, to elute the bound
protein, which may be detrimental to both the immobtlized ligand and purified
protem The usefulness of these methods IS restricted by the availability of the
ltgand, the conditions requtred to recover the bound protein, and the ability to
regenerate the affinity matrrx, such constramts may limrt the use of these tech-
niques in large-scale purifications

References
1 Conner, J , Marsden, H., and Clements, J B (1994) Rrbonucleotide reductase of
herpesviruses. Rev Med Vzrol. 4,25-34
2 Dutta, B M , Frame, M C , Subak-Sharpe, J H , Clark, W N , and Marsden, H
S. (1986) Specific inhibition of herpes virus ribonucleotide reductase by synthetic
peptides. Nature (Lond) 321,439-441.
3. Studier, F W and Moffat, B A. (1986) Use of bacterrophage T7 polymerase to
direct high-level expression of cloned genes J A401 BJOZ 189, 113-130
4. Furlong, J., Meighan, M , Conner, J , Murray, J., and Clements, J. B (1992) Methods
for improved protein expression using PET vectors. Nucleic Acids Res 20,4668.
5 Furlong, J , Conner, J , McLauchlan, J , Lankinen, H , Gait, C , Marsden, H S ,
and Clements, J. B. (1991) The large subunit of herpes simplex vnus type 1 rtbo-
nucleotlde reductase expression m Eschertchza co11 and purtfication Vwologv
182,84685 1
6 Lankinen, H., McLauchlan, J , Wetr, M , Furlong, J., Conner, J., McGarrrty, A ,
Mrstry, A., Clements, J B., and Marsden, H. S (1991) Purrficatron and character-
izatton of the herpes stmplex virus type 1 rtbonucleottde reductase small subunit
following expression m Eschertchza co/c J Gen b-01. 72, 1383-I 392
7. Atherton, E , Gait, M. J , Sheppard, R. C , and Wtlhams, B J. (1979) The polya-
mide method of solid phase pepttde and ohgonucleottde synthesis Bloorg Chem
8,35 l-370
8. Sheppard, R. C (1983) Continuous flow methods m orgamc synthesis. Chem
Brzt. 19,402-4 13
Q

Expression and Purification
of Secreted Forms of HSV Glycoproteins
from Baculovirus-Infected Insect Cells
Sharon H. Willis, Charline Peng,
Manuel Ponce de Leon, Anthony V. Nicola,
Ann H. RUX, Gary H. Cohen, and Roselyn J. Eisenberg


1. Introduction
Herpes simplex virus (HSV) remains a major human pathogen worldwide
(2.5) causing cold sores, eye and genital mfections, blmdness, encephalttrs,
and neonatal mfectlons. Most adults have anttbodles against the oral form of
the virus HSV-1 (9), and a sigmftcant number are infected with the genital
form, HSV-2. Both serotypes establish lifelong latent infections and reacti-
vate perlodtcally to produce recurrent disease (25). After mfectron, virus-
encoded glycoprotems are expressed on all cellular membranes and are major
targets of the host™s immune response. The vlrion envelope contains 10 gly-
coprotems that are important for infection and pathogenesis of HSV-1 and
HSV-2. Because HSV contains so many glycoprotems, sorting out their func-
tions m virus entry remains a dtfficult task. Our approach has focused on
establishing structurefunction relationships of the individual glycoprotems
with particular emphasis on gC and gD. After many years of studymg the
properties of these protems m HSV-infected and plasmid-transfected mam-
malian cells, we have now begun to overexpress the proteins usmg a
baculovuus expression system.
The first mteractlon between HSV and a susceptible cell occurs when gC
binds to cell surface heparan sulfate proteoglycans (29). The second step of
virus entry involves the mteractlon of gD with a second cellular receptor (3,17),
possibly the mannose-g-phosphate receptor (Man6pR) (2). Followmg this
interaction, the virus envelope fuses with the host cell plasma membrane. Four
From Methods m Molecular Medmne, Vol 10 Herpes Smplex !hus Protocols
E&ted by S M Brown and A R MacLean Humana Press Inc , Totowa, NJ

131
Willis et al.
132

viral glycoprotems (gB, gD, gH, and gL) are important for this step, although
the precise mechanism is not understood
Immunological, biochemical, and genetic approaches have established
working models of gD structure (6,8,12,14,19,22), but a complete understand-
mg requues X-ray crystallographtc methods. Therefore, one of our long-term
goals ts to solve the 3D structure of gD and to learn how alterations in structure
relate to function. To obtain the large amounts of gD needed, a baculovn-us
system is used to express a truncated form of the glycoprotem (gD-1[306t])
that lacks the transmembrane region (TMR) and carboxyl terminus (27).
gD-l(306t) is therefore secreted into the culture supernatant but contains all
residues associated with its function Having large amounts of a soluble form
of purified gD has allowed us to analyze its structure on three different levels.
First, we have continued anttgemc mapping studies, and now we use btophyst-
cal techmques such as mass spectrometry and circular dichroism (CD) to fur-
ther probe the structure of gD Second, we have utilized scanning transmission
electron microscopy (STEM) and 2D electron crystallography to obtain details
of gD structure and a topography at the ultrastructural level. Thud, we are able
to crystallize gD- l(306t) (34) and efforts are now underway to obtain crystals
suitable for X-ray diffraction.
In addition to the structural studies, we are also analyzing the functional
properties of wild-type and variant forms of baculovtrus-expressed HSV gly-
coproteins. For example, experiments using the truncated proteins will enable
us to extend five observations about gD function:
1. gD Binds to cell surfaces (I 6);
2. gD Blocks HSV plaque formation (16),
3 gD Binds to the Man6pR (2);
4. gD Participates m fusion, and
5 gD Mediates supermfection interference (4)
Thus far, we have shown that gD-l(306t) binds to cells in a saturable man-
ner and also blocks HSV plaque formation (31).
The production of a soluble form of gC has allowed us to explore the mter-
action between gC and the cell surface m the absence of detergent. Truncated
forms of gC lacking the TMR and carboxyl regions were expressed m
baculovirus-infected cells (31). These proteins bind to mammalian cells, bmd
to mnnobtlized heparm, and block attachment of vu-us to cells. In addttron,
each of these proteins bind conformation-dependent monoclonal antibodies
(MAbs) and to the human complement component C3b, showing that the proper
conformation of gC 1smaintained when the truncated molecule is expressed in
insect cells (32). Thus, our experience with the baculovlrus expression system
Secreted HS V Glycopro teins 133
is that it opens up new areas of investigation by provldmg a relatively simple
means of obtaining large quantities of biologically active and structurally cor-
rect forms of HSV glycoprotems.
This chapter therefore IS devoted to our colleagues desiring the capacity to
obtain large amounts of protem for their studies. The term “large amounts” IS
relative to the mvestlgator We find that no matter how much protein is made,
it is quickly expended and easier ways of getting purified proteins m bulk are
constantly being sought. We hope that what we have learned ˜111 have an
impact on the ability of other labs to produce proteins usmg a baculovlrus
infection system. In this spmt, we present protocols based on our experiences,
recognizing that other published protocols exist for many of the techniques
we describe. Our goal IS to bring you as close as possible to our “benchside”
with both pleasant and unpleasant experiences from working our way
through the procedures for the isolation and purrflcation of baculovn-us-
expressed HSV glycoprotems
2. Materials and Equipment
The materials and equipment we use as well as where they can be purchased
are listed within the body of the text.
3. Methods
Our early attempts to express gD-l(306t) usmg the wild-type gD slgnal

<<

. 20
( 61 .)



>>