<<

. 28
( 61 .)



>>

3. Increase the volume of the sample to 5 mL with PB, mix by mvertmg the tube
three times, then centrifuge for 20 mm at 27,000g in a Beckman SW50 1 rotor
4 Decant the supernatant fluid and leave the tube inverted on some absorbent paper
m a beaker to allow all the PB to dram from the pellet
5 Use distilled water to suspend the pellet and place a small drop of virus suspen-
sion onto a formvar-coated electron microscope grid. Remove excess fluid and
m-mediately add a drop of 1% PTA
After removal of excess stam, the grid is ready for examination by electron
microscopy (EM) (see Note 9)
3.3.2. Labeling of Subenvelope Components
To allow gold probes to bmd to subenvelope components of intact envel-
oped vlnons, the viral envelope must first be ruptured or removed entirely.
3 3.2.1. VIRAL CAPSIDS
Detergents such as Tnton-X-100 will cause dissociation of the membranous
viral envelope and allow the release of vn-al capslds.
1. Concentrate the vmons by centrifugatron.
2. Resuspend the virus pellet in 750 uL of Triton-X- 100 lysis buffer, vortex, leave
at 4°C for 1 h, then vortex again.
3. Centrifuge for 5 mm in a microfuge.
4. Decant the supernatant lysis buffer, and resuspend the pellet m approx
250 pLofPB
180 Stannard

5 Follow steps 2-5 as described for the labeling of envelope components m
Sectlon 3.3 1
3.3.2.2 TEGUMENT PROTEINS
Treatment of virus particles with Triton-X-100 will remove not only the
vlrion envelopes but the tegument proteins as well. To achieve labeling of the
tegument proteins (Fig. 2B), the envelope can be ruptured by nonchemical
means to allow the gold probes to penetrate*
1 First concentrate the Intact vlrlons by centrlfugatlon
2 Suspend the pellet in distilled water and place one drop of virus suspension on a
formvar-coated electron microscope grid Remove excess fluid and allow the grid
to dry completely (see Note 10)
3 Place a drop of dilute gold probe onto a piece of dental wax m a humidified Petri
dish. Float the grid face down on the gold and Incubate at 37°C for 2 h (see
Note 11)
4 Rinse the grid gently with dlstliied water and stam with 1% PTA, then examme
by transmission electron microscopy (see Note 12)

4. Notes
1 Large-sized probes bmd at lower frequency than those prepared with smaller gold
particles, probably as a result of sterlc hindrance It IS a useful practice to prepare
gold particles of different sizes so that labelmg with a specific hgand can be
tested usmg probes of a variety of sizes When performing double-labeling
experiments with two probes of different sizes, It IS mformatlve to repeat the test
and reverse the sizes of the gold that IS coupled to each hgand (see Fig 1)
2. When preparing gold sols, the beakers or flasks m which Solutions A and B are
contamed should be of a size that will allow rapid mlxmg of both solutions Keep
these containers covered during the heating process to avoid loss of moisture by
evaporation that could cause changes m concentration of the chemlcais
3. Monoclonal antibodies (MAbs) have dlstmct advantages over polyclonal antl-
bodies (PAbs) for use in direct lmmunogold studies The most obvious advantage
IS that of speclficlty, smce ail the lmmunoglobulm molecules are identical The
relatively low ratio of specific molecules m a PAb mixture can be Improved by
affinity purification prior to the production of probes A polycional preparation
contams antibodles with a wide range of lsoelectrlc points, and calculation of the
optimal pH for coupling to gold 1s difficult. A pH of 9 0 generally has been found
to be smtable for heterogeneous lmmunoglobulms (12) MAbs have a better
defined pZ value and therefore adsorb more efficiently to the gold particles at a
slightly lower pH.
4 As many proteins tend to aggregate if kept m low tome strength solutions, It IS
wise to choose condltlons that will allow dlaiysls to occur in the shortest possible
time. Large volumes of dlaiysls buffer and constant stlrrmg ˜111help to speed up
the process Protem precipitates that form during dlalysls should be removed by
181
Labelmg of Herpesvms Suspensions
centrifugatton of the protein sample before it 1scoupled to the gold. Thts precipi-
tated protein will usually redissolve in buffers of higher molarity and can be used
without serious loss of potency in other tests
5. Flocculation of small size colloidal particles is sometimes slow Allow 10-30
min after the addition of 10% NaCl before calculating the titration endpoint
When preparing probes, the amount of protein that is added should not greatly
exceed the mimmal stabihzmg concentration An excess of free ligand in the
final preparation will compete with the ligand attached to the gold and reduce the
binding efficiency of the probes.
6 Washing by centrtfugatton is improved by using swinging bucket-type rotors,
such as the Beckman SW50.1 The increased distance of centrrfugatron over fixed
angle rotors allows better separation of large and small particles. Gold probes
will pellet more rapidly from suspending fluids containing 0.5% BSA than from
1% BSA. If some of the gold remains in suspension after 45 mm at 27,OOOg,
remove the supernatant fluid to a new tube and repeat the centrifugatron of that
fraction. Increasing the time of centrifugation will result m increased amounts of
the unbound ligand in the pellet and thus decrease the efficiency of washmg
7. Optimal labeling occurs in the presence of excess probe This assures that the
maximum number of binding sites are labeled and decreases the chance of
crosslmkmg the vu-us particles mto mrrnune complexes. On the other hand, if too
much gold is added to the reaction mixture, tt will be difficult to separate the
labeled virrons from the unbound gold that ˜111hamper the detection of specific
binding m the final EM preparation If the volume of the reaction mixture is kept
small (less than 300 pL), the total amount of probe that needs to be added can be
kept to a mmtmum. After incubation, labeled vuions can be separated from the
unbound gold by first dllutmg the sample 15- to 20-fold and then applymg differ-
ential centrifugation to pellet the virus particles while leaving most of the free
gold in suspension If the amount of background gold is unacceptably high m the
first pellet, resuspend the pellet in at least 5 mL of PB, mix by mvertmg the tube
three times, and then repeat the differential centrifugation step.
8. When using whole IgG molecules on probes, the chance of nonspecific Fc-binding
is always a problem and should be monitored by including appropriate control
experiments with gold probes prepared with an unrelated antibody from the same
animal species Preincubation of the virus sample with nonimmune mununoglobu-
lm before exposure to the specific probe may help to ehmmate nonspecific Fc-
binding. Alternatively, the antibody can be digested with papain and the F(ab)˜,
portion isolated for couplmg to the gold, although this process is somewhat labori-
ous. Most monoclonal antibodies do not give anomalous Fc-binding, although
murme IgG of the subclass IgGl has been observed to react nonspecifically with
cytomegalovirus structural proteins (D. H. Hardie, personal communication)
Therefore, it 1s recommended that tests include control probes comprising
MAbs of the same subclass of IgG as those under mvestigatron.
9. Good resolution of fine ultrastructural components of the virion depends to a
large extent on successful negative staining The stam should not be too dense
182 Stannard

nor too concentrated at the perrphery of the virus particles otherwise both struc-
tural detail and small gold probes will be obscured Best staining is achieved
when a small amount of low-molecular-weight protein is present in the sample
This will aid the “spreadmg” of the stain on the EM grid and create a light grey
background that is optimal for visualization of the labeled viral components.
When using virus samples concentrated from cell culture flmds, sufficient low
molecular weight protein will be present in the pellet When mvesttgatmg gradi-
ent-purified virus samples, however, tt may be necessary to apply a coating of
0.01% BSA to the EM grid before adding the stain.
10. When virtons m a hypotonic solution are allowed to dry on a grid, the viral enve-
lopes usually rupture and roll back Thus is a quick and efficient means of expos-
mg subenvelope components which can then be labeled zizsitu (i.e., on the grid)
as described Note, however, that this technique does not allow accurate detec-
non of antigens on the surface of the envelope
11. Gold probes should be diluted m TBSA-azide. The BSA content of the diluent
will minimize unwanted background stammg It is important that the drop of
gold-protein complex does not dry out during mcubation, as this ˜111 result m
unacceptable background staining Ensure that the drop of diluted probe is not
too small, and that the Petri dish is adequately humtdified by placing a thick layer
of water-saturated filter paper on the base of the dish
12. Take care not to damage the support film when rmsmg the grad Hold the grid
carefully wtth fine-tipped forceps, place a piece of torn filter paper between the
tines of the forceps so that tt Just touches the edge of the grad, and repeatedly
apply small drops of water to the opposite edge of the grid Apply the negatrve
stain in the same way

References
1. Frens, G. (1973) Controlled nucleatton for the regulation of the particle size m
monodisperse gold suspenstons. Nat Phys Scz 241,20-22
2 Slot, J W and Geuze, H J (1985) A new method for preparing gold probes for
multiple-labeling cytochemistry Eur J Cell Biol 38, 87-93.
3 Stannard, L M., Fuller, A. O., and Spear, P G. (1987) Herpes simplex virus gly-
coprotems associated with different morphological entities proJectmg from the
virion envelope J Gen Vu-01 68, 715-725
4 Stannard, L M , Rider, J R., and Farrar, G. H. (1989) Morphology and distribution of
gp52 on extracellular human cytomegalovirus (HCMV) supports biochemical evi-
dence that it represents the HCMV glycoprotem B. J Gen Vzrol 70, 1553-l 560.
5. Stannard, L M. (1989) Pz-Microglobulm binds to the tegument of human
cytomegalovirus An immunogold study. J Gen Vu-01 70,2 179-2 184
6. Stannard, L. M. and Hardte, D. R. (1991) An Fc receptor for human immunoglobuhn
G 1slocated within the tegument of human cytomegalovirus J Vu-o1 65,341 l-341 5.
7 Hopley, J. F A and Doane, F W. (1985) Development of a sensitive protein A-gold
rmmunoelectron microscopy method for detectmg viral antigens m fluid specr-
mens J Vwol Methods 12, 135-147.
Labeling of Herpesvirus Suspensions 183
8 Robinson, E. N., MC Gee, 2. A., and Clemens, C M (1987) Probing the surface
of bacteria: use of gold sphere tmmunological probes. Mcrob Pathogen 2,159-l 69
9 KJeldsberg, E. (1986) Use of gold IgG complexes and human antisera for electron
microscope tdentification of hepatitis A vn-us and pohovuuses. J Vzrol Methods
13,207-2 14.
10 Hernandez, F , Riviera, P , and Hosaka, Y (1987) Homogeneous drstrtbution of
paramfluenza virus glycoprotems demonstrated by immunogold-labelmg and light
stammg with many1 acetate m electron microscopy. J. Vzrol Methods 15,273-277
11 Roth, J. (1983) The colloidal gold marker system for light and electron mrcro-
scopic cytochemtstty, m Technzques zn Immunocytochemlstry, vol 2 (Bullock, G
R and Petrusz, P , eds ), Academic, London, pp 217-284.
12. De Mey, J (1983) Collotdal gold probes in tmmunocytochemistry, in Immunocy-
tochemistry Practical Applzcations rn Pathology and Bzology (Polak, J M and
van Noorden, S , eds.), Wright, PSG, Bristol, London, pp 82-l 12.
12
Expression of HSV Proteins in Bacteria
Elizabeth A. McKie


1. Introduction
Expression of herpes simplex virus (HSV) polypeptides in bacterial expres-
sion systemshas provided a useful tool for the generation of large quantities of
specific viral proteins for use in both biochemical and functional analysis, and
as immunogens for antisera production. Proteins can be expressed either in the
full-length native form or as fusion proteins with affinity tails.
The PET system (1,2) is probably the most wtdely used for production of
native full-length herpes stmplex vnus proteins, and has been used success-
fully to generate large quantities of both the large and small subunits of ribo-
nucleotide reductase for functional analysis (3,4) In the PET system, the
protein-coding sequence of interest is cloned downstream of the T7 promoter
and gene 10 leader sequences. Protein expression IS generally induced by the
addition of IPTG to the growth media. The bacterta are lysed and the expressed
protein purified using conventional chromatography techmques.
Several systems for expression of proteins as fusions are available. These
are all based on a general theme where the protein of interest IS expressed as a
fusion with a peptide that can be purified using an affimty matrix. Generally, a
specific cleavage site is available to permit release of the expressed protein
from its fusion partner. In herpesvirology, one of the most commonly used is
the GST system, m which proteins are expressed as fusions with the glu-
tathione-S-transferase from Schistosomalapomcum (5). These fusion proteins
can be easily purified using glutathione Sepharose 4B, and then eluted under
mild conditions usmg glutathione. The GST portion can then be cleaved from
the fusion protein using site-specific proteases.
The XpressTM Expression and Purification System (Invitrogen, San Diego,
CA) produces recombinant proteins that are fused to a short leader peptide,
Methods m Molecular Medmne, Vol 10 Herpes Smplex Vws Protocols
From
Edted by S M Brown and A R MacLean Humana Press Inc , Tolowa, NJ


185
McKie
186
which is derived from the bacteriophage T7 gene 10 and contains SIX histidme
residues (the polyhistidme sequence). These histidine residues have a high
affinity for divalent cattons and bind readily to a mckel-chelatmg resin.
Recombinant proteins can be eluted from the resin using either low pH buffer
or by competttion with imidazole or histidine. Protem A vectors are also avail-
able, these use protein A as the fusion partner and IgG Sepharose 6FF for affin-
tty purtficatton. Recently, PinPomtTMvectors have been introduced by Promega
(Madison, WI). In this case,the DNA codmg for the protein of Interest is cloned
downstream of a sequence encoding a pepttde that is biotmylated m VIVO.The
expressed protein binds to monomeric avidm resin and can be eluted under
mtld denaturing conditions.
Other available vectors are specifically designed for the expression of soluble
protems. In one such system,proteins are expressed as fusions with the Escheri-
chza coli thioredoxm protem, which remams soluble even when expressed at
levels as high as40% of the total cellular protein. The fuston protein accumulates
at sites in E colt called adhesion zones, and 1s selectively released from the
bacterium by either osmottc shock or heat treatment. An enterokmase cleavage
signal 1s present between the thioredoxm gene and the cloned polypeptide to
allow release of the protem of interest followmg purification.
The protocols described in this section provide a general guide to the meth-
ods used for assessinglevels of expression and protem solubihty, and are apph-
cable to almost any of the commerctally available protein expresston systems.
Protocols for purificatton of GST fusion proteins are also given in this chapter.
However, for other specific affinity purification systems,readers should refer
to the manufacturer™s mstructions.
2. Materials
2.1. Expression Plasmids and Bacteria
A wide range of expression plasmids are commerctally available, and the
application for which the expressed protein is to be used will determine which
is most suitable. PET vectors that permit expresston of the full-length nattve
protein are really only necessary if the expressed protein is to be used for func-
tional studies, since they have the disadvantage that the recombinant protein
must be purified free from contammatmg bacterial proteins, which requires a
sound knowledge of protein purification techniques. For many other applica-
tions, such as antibody production, fusion proteins are more suitable, since
they are more easily purified to homogemety.
Manufacturer™s generally supply bacteria with their expression plasmids. For
PET vectors, the strain of choice is BL2 l(DE3), which has a deletion in both the
ampT and lon protease, thus reducing proteolysis of expressedfusion proteins. In
this strain, the T7 RNA polymerase 1sencoded on a lysogenic 3L bactertophage
187
Herpes Simplex in Bacteria
under lacUV5 control. Induction of the polymerase with IPTG allows controlled
expression of genes placed downstream of the T7 RNA polymerase binding site.
pGEX vectors have no specific host requirements for propagation of the
plasmids or for expression of fusion proteins, but it is best to use strains that
are protease deficient, e.g., Y 1090.
2.2. Bacterial Growth Media
1. LB Agar: In 900 mL dH*O, dissolve 10 g NaCl, 10 g tryptone, 5 g yeast extract,
and 20 g bacteriological agar, pH to 7 0, with 5N NaOH, and make volume up to
1L Aliquot and autoclave
2 LB. In 900 mL dHzO, dtssolve 10 g NaCl, 10 g tryptone, 5 g yeast extract, pH to
7.0, with 5N NaOH, and make volume up to 1 L. Ahquot and autoclave.

2.3. Chemicals and Reagents
1 Lysozyme (Sigma, St. Louis, MO) to a final concentration of 500 ug/mL
2 100 mg/mL Amprcillm (Sigma). filter-sterihzed
3 100 mMIPTG (Sigma). filter-sterilized.
4. 20% Trlton X- 100 in PBS

<<

. 28
( 61 .)



>>