<<

. 27
( 61 .)



>>

novel nickel-chelating lipid. J Struct Btol 113, 117-123
19. Scopes, R (1987) Protein PurEfication Prtnctples and Practtce Springer-Verlag,
New York.
20 Mann, M. and Wilm, M. (1995) Electrospray mass spectrometry for protein char-
acterization. Trends Btochem Set 20, 2 19-224
172 Luisl, Anderson, and Hope

21 Grisshammer, R. and Nagat, K. (1994) Purification of over-produced proteins
from E co11cells, m DNA Clonwg 2 Expresson Systems A Practzcal Approach
(Glover, D M. and Hanes, B D., eds.), IRL, Oxford, UK.
22. Lapthorn, A J , Harris, D. C , LittleJohn, A., Lustbader, J W , Canfield, R. E ,
Machm, K. J., Morgan, F. J and Isaacs, N W (1994) Crystal structure of human
chormomc gonadotropm Nature 369,455-46 I
23 Bentley, G. A , Boulot, G , KarJalamen, K , and and Marmzza, R A (1995) Crystal
structure of the p chain of a T-cell antigen receptor Sczence 267, 1985-I 987
24 ManJunath, P and Sairam, M. R (1982) J Blol Chem 257,7 109-7 115
25 Pieles, U , Zurcher, W., Schar, M , and Moser, H E (1993) Matrix-assistedlaser-
desorption-iomzation time-of-flight mass spectrometry. a powerful tool for the
mass and sequence analysts of natural and modified oligonucleotides Nuclezc
Acids Res 21, 3191-3196.
26 Brooksbank, R (1994) PhD thesis Umversity of Cambridge
21 Everett, R. D., Barlow, P., Milner, A, LU˜SI, B., Orr, A, Hope, G , and Lyon, D
(1993) A novel arrangement of zmc-bmdmg residues and secondary structure m the
C,HC, motif of an alpha herpes virus protein family. J MO/ Bzol 234, 1038-1047
28 Bratg, K , Otwmoski, Z , Hegde, R., Boisvert, D C , Joachimiak, A, Horwich, A
L , and Sigler, P B (1994) The crystal structure of the bacterial chaperonm GroEL
at 2 8 A resolution. Nature 371, 578
29. Oubridge, C , Ito, N., Evans, P R , Teo, C.-H , and Nagai, IS. (1994) Crystal struc-
ture at 1 92 A resolution of the RNA-binding domain of the Ul A spltceosomal
protein complexed with an RNA hairpin Nature 372, 432-438
30 Jacobo-Molma, A , Dmg, J., Nanm, R G., Clark, A D , Lu, X , Tantillo, C., Wil-
liams, R. L., Kame, G., Ferris, A. L., Clark, P., Hizi, A , Hughes, S H , and Arnold,
E (1993) Crystal structure of human immunodefictency vtrus type 1 reverse tran-
scriptase complexed with double stranded DNA at 3.0 A resolution shows bent
DNA Proc. Nat1 Acad Scz USA 90,632O
31 Rnu, J M , Stanfield, R. L , Stura, E A , Salmas, P A., Profy, A T , and Wilson,
I A (1993) Crystal structure of a human immunodeficiency vnus type 1 neutral-
izing antibody, 50 1, m complex with its V3 loop peptide antigen Proc Nat1
Acad See USA 90,6325-6329
32 Prongay, A. J., Smith, T J , Rossman, M G , Ehrlich, L S , Carter, C A , and
McClure, J. (1990) Preparation and crystallization of a human mununodeficiency
virus p24-Fab complex Proc Nat1 Acad Scz USA 87,9980-9984
33. Tormo, J , Fita, I., Kanzler, O., and Blaas, D. (1990) Crystallization and prehmi-
nary X-ray diffraction studies of the Fab fragment of a neutralizing monoclonal
antibody directed against human rhmovuus serotype 2. J Bzol. Chem. 265,
16,799-16,800
11
Direct lmmunogold Labeling
of Herpesvirus Suspensions
Linda M. Stannard


1. Introduction
Gold particles in collotdal suspension are partrculariy well surted as markers
for Immune electron mrcroscopy. Their extreme electron opacity ensures that
they are detected with accuracy even at particle sizes of less than 3 nm. Gold
spheres can be made easily and mexpenstvely by reduction of gold chloride
with mild acid and heat (Z), and particles can be prepared in a variety of sizes
by varying the nature of the reducing agents (2).
An important feature of the colloidal particles is that they have negatrve charges
on then surfaces, and consequently bmd raptdly and spontaneously to proteins
at a pH close to, or slightly above, the rsoelectrrc point (pr> of the protein. This
irreversible electrostattcbond with protein stabilizesthe sol and prevents precipita-
tion of the gold particles m the presence of electrolytes. Becauseproteins (such as
lectms, nnrnunoglobulms, enzymes,staphylococcal protein A, and so on) can be
stably coupled to colloidal gold without sigrnticant loss of their brologrcal proper-
tres, gold-protein complexes form excellent probes for the ultrastructural tdenttfi-
cation of viral antigens or sitesof brologrcal actrvity.
Herpes vtrions contam multiple structural proteins and much stall has to
be learned about their physical arrangement in the vlrton at various stages
of morphogenesis, as well as about then brological functions m vivo. The
use of immunogold labeling and negative stain electron mrcroscopy makes
It possible not only to identify specific proteins within individual virus par-
ticles at an ultrastructural level but, moreover, to observe the mteracttons
and affinities of these proteins when they are m a natural conformation.
Gold-tagged monoclonal antibodies have been used to Identify envelope
glycoproteins on herpes simplex virions (3) and on human cytomegalovi-
From Methods /n Molecular Medune, Vol. 10 Herpes Smplex Vws Proioco/s
Edlted by S M Brown and A R MacLean Humana Press Inc , Totowa, NJ

173
Stannard




Fig. 1. By direct immunogold labeling of herpes simplex virions with specific
probes for gB (A), or gD (B) it is possible to detect morphological differences between
these two types of glycoprotein spikes in the virion envelope. The gB spikes are long
and relatively rigid, whereas gD spikes are shorter and more slender. Virions in (C)
and (D) have been double-labeled with monoclonal antibodies to gB and gD on differ-
ent sizes of gold particles. In C, anti-gB is coupled to the larger gold particles, and in
D the anti-gB is attached to the smaller gold particles. Note how the smaller sized
probes allow better visualization of individual glycoprotein spikes. The envelopes of
the virus particles in A and C are intact, but in B and D the envelopes have been
damaged sufficiently to allow penetration of the negative stain so that the nucleo-
capsids can be seen. Bar = 100 nm.


rus particles (4) (see Figs. 1 and 2). In addition, immunogold labeling has
demonstrated that both P,-microglobulin and the Fc portion of human IgG bind
to one of the tegument proteins of human cytomegalovirus (.5,6).
175
Labeling of Herpesvirus Suspensions




Fig. 2. Immunogold labeling of human cytomegalovirus (HCMV) particles has been
used to identify the location of two different viral proteins, one on the envelope and
the other in the tegument. (A) Gold particles coupled to a monoclonal antibody to the
major envelope glycoprotein (gB) were mixed with a suspension of HCMV particles.
The probes bind strongly to the virion envelope. (B) This virus particle was allowed to
dry on the EM grid, then floated on a droplet of colloidal gold coupled to p2-
microglobulin. The virion envelope has collapsed, allowing the Pz-microglobulin to
bind to its affinity protein that is situated in the tegument. (C) Tegument proteins of
virions in suspension can be labeled only if the envelope has ruptured. In this case the
virus particle has been double-labeled with anti-gB (larger gold) and P2-microglobulin
(smaller gold). Bar = 100 nm.

Indirect immunogold staining techniques have been used in the study of a
variety of viruses and bacteria by making use of colloidal gold coupled to either
protein A (7,8) or an anti-IgG (9, IO) to tag unlabeled IgG that have bound to
the virion in a primary reaction. However, by linking a specific antibody directly
to the gold particles, it is possible to obtain probes that are much more sensitive and
specific, and which can be used in a simple one-step labeling protocol. A further
176 Stannard

advantage of direct labehng ISthat it facilitates multiple labeling. By using gold par-
ticles of different sizes,each coupled to a drfferent monoclonal antibody (or altema-
tive hgand), the distribution and relationship of more than one antigen can be
demonstrated on a single virion (Figs. lC,D and 2C). This tecbmque is especially
rewarding for herpesvtrus parttcles becauseof their comparatively large size and
becauseof the comprehensive range of monoclonal antibodies that ISavailable.
The optimal parameters for the preparation and application of gold-protein
complexes have been extensively studied and reported m comprehensive
review articles (11,12). However, this chapter descrtbes a sample and effective
protocol to produce sensitive and specific immunoprobes that can be used with
minimal effort, and will allow for innovative versatility m the detailed exami-
nation of multiple facets of herpesvn-us architecture.
2. Materials
2.1. Preparation of Colloidal Gold Particles
All solutions must be prepared wtth double-drsttlled or, preferably, deion-
ized water. Filtration through a 0.22˜pm pore size Millipore filter is recom-
mended to ensure that traces of organic substancesare removed from the water.
It IS important to ensure that all glassware, magnetic fleas, and so on are scru-
pulously cleaned and rinsed with distilled water. Note: Extraneous ions present
during the preparative stage will cause the gold to flocculate.
Colloidal gold particles cl 5 nm m diameter aremade by reduction of chloroauric
acid (gold chloride) with amixture of two reducing agents,sodium citrate, and tanmc
acid. The sizeof the particles is determined by the amount of tanmc acid used in the
reduction mixture; the more tanmc acid, the smaller the sizeof the colloidal particles.
Chloroauric acid IS supphed m a crystalline form and IS extremely hydro-
philic. Therefore, it is advisable to dissolve the entire contents of the vial as
soon as it is opened and to store the gold chloride preparation as a 4% aqueous
solution. If kept at 4°C m the dark, it should last for many months.
Not all sources of tanmc acid are equally good for production of homoge-
neous gold ˜01s.Successful results can be obtained with tanrnc acrd supplied
by Mallmckrodt Inc. USA, as code no. 8835.
To make a 50-mL colloid:
1 Solution A (1% HA&l& Deionized water, 39.875 mL, 4% chloroauric acid,
0.125 mL.
2. Solution B (reduction mixture): 1% trisodmm citrate, 2 mL (freshly made); 1%
tanmc acid, O-2.5 mL; 0 015M K2C03, O-2.5 mL (equivalent volume to tanmc
acrd used), microfiltered deionizedwater up to 10mL.
It is important that the Na-citrate solution for the reducing mixture (Solution B) be
made fresh Just before use. Choosea volume of 1% tanmc acid that will yield gold
Labehng of Herpesvirus Suspensions 177

partrclesof the desired size.This volume may vary in the handsof different research-
ers, but a rough guide of quantrtres1sas follows: 25 uL tannic acid for IO-nm gold
particles; 50-l 00 pL tanmc actd for 6-8 nm gold, 1.5 mL tannic acid for 3 mn gold.
2.2. Preparation of Specific Probes
The followmg soluttons are required:
1 0.2 mM borax buffer, pH 9 0. For convenience, make a 10X stock solutton that
can be stored at 4°C unttl needed. Dtssolve 0.762 g of drsodmm tetraborate
(Na,B40, 10H20) m 1 L of distrlled water, and adjust the pH to 9 0 with 0 1N
HCl This stock solutron must be dtluted IO-fold before use.
2 0.2M K&O,
3. 0.2M H,PO,
4 10% NaCl
5 30% (100 vol) Hydrogen peroxide.
6 10% Bovine serum albumin (BSA) in dtsttlled water. Adjust the pH to 9 0 with
NaOH, and microfilter using a 0 22-pm filter
7 TBSA-azide. 0 02MTrrs-HCI, pH 8.2, m 0 15MNaC1, containing 0 5% BSA and
0 02M NaNs
2.3. Labeling of Virus Particles with Gold Probes
2.3.1. Viral Envelope Components
1 Large numbers of vnus particles can be recovered from the supernatant culture
fluids of infected cell monolayers that exhrbrt obvrous cytopathrc effects
2 Washing dtluent: 0,06M phosphate buffer, pH 7 2 (PB)
3. Negative stain for electron microscopy: 1% phosphotungsttc acid, pH 6.2 (PTA)
2.3.2. Viral Capsids
Trrton-X- 100 1ys1.sbuffer: Prepare a solutton of 1% Triton-X- 100 rn 10 mA4
Tris-HCl, pH 7.5, containing 1 rnA4 CaCl,, 0.194 NaCl, and 2 mA4 phenyl-
methylsulfonyl fluoride (PMSF), which ISa protease inhibitor. Note that PMSF
IS not readily soluble in aqueous solutions and therefore rt should first be drs-
solved m a small volume of ethanol such that the final concentratron of ethanol
m the lysrs buffer is 1%
3. Methods
3.1. Preparation of Colloidal Gold Particles
1 Decide on the stze of colloidal gold particles required, and prepare the reducmg
solution (Solution B) accordmgly (see Note 1)
2. Preheat both Solutron A and Solution B to 60°C by placing them m a 60°C water bath,
3. Place a magnetic flea m Solution A and stir on a magnetic stirrer heated to mam-
tam the temperature of 60°C
178 Stannard

4. Add Solution B to Solution A while rapidly mixing. The color of the gold mix-
ture will changefrom blue to red, indicating that reduction 1scomplete Continue
stirring at 60°C for 20 mm, then bring to boiling point and boil gently for approx
2 mm (see Note 2)
After coolmg, a sample may be examined m the electron microscope to
monitor the size and homogeneity of the colloidal particles.
Gold sols should be stored in sterile bottles at 4°C In the dark. They can be
kept for many months (even years) without deterloratlon
3.2. Preparation of Specific Probes
1 The protein llgand that 1sto be coupled to the gold collold should be as pure as
possible (see Note 3) You will require 20@-500 VL of protein at a concentration
of approx 1 mg/mL
2 Dialyze the protein sample against 0 2 mA4 borax buffer to reduce the lomc con-
centration of the protein solution (see Note 4).
3. Measure the pH of the gold collold by applying one or two drops of the suspen-
sion to a strip of good quality pH indicator paper (Particles of colloidal gold are
negatively charged and will clog the pores of the electrode of a conventional pH
meter ) Use 0 2MK2C03 to adJust the pH of the gold sol to be shghtly more basic
than the pl of the protein that will be used for the probe For human or rabbtt IgG,
a pH of approx 9.0 IS satisfactory. For monoclonal antibodies, pH 7.5 is prefer-
able. If necessary, the pH of the gold may be lowered usmg 0.2M H3P04
4 Prepare a mmltltration to determine the protein concentration that is needed to sta-
bilize the gold sol. In glass tubes make doubling dilutions of protein m lO+L
amounts of 0.2 mM borax buffer Add 100 pL of gold (at appropriate pH ) to each
tube, vortex, let stand for 2 min, then add 10 pL of 10% NaCl and vortex again. Salt
˜111cause unstabilized gold to flocculate, which results m a color change from red
to blue When collolds have been prepared using substantial amounts of tanmc
acid, approx 10 pL of 30% hydrogen peroxide also should be added together with
the NaCl This will serve to bleach the brownish color caused by polymers of tan-
mc acid, which might otherwise obscure the titration end point The mmlmal stabl-
llzmg concentratzon of protein 1sindicated by the highest dilution at which the gold
remains red. Protein should be used at this concentration (or if desired, increase by
about 10%) for the preparation of gold probes (see Note 5).
5 To prepare the probe, add 1 vol of protein (appropriately diluted m borax buffer) to
10 vol of colloidal gold at the required pH, vortex, and allow to stand for 2 min Test
a small ahquot of the protem-gold mixture to ensure that the gold has been stabl-
lized and remains red after the addition of 10% NaCl Then add 10% BSA solution to
give a final concentration 1%BSA. The BSA actsasan extrastabilizer.
of
6 Centrifuge the gold-protein complex at 48,000g (20,pOO rpm m a Beckman
SW50 1 rotor) for 1 h, and resuspend the sedimented gold particles m 5-10 mL
TBSA-azide Very small gold particles (3 nm) will require a longer time of cen-
tnfugation, approx 2 h, to sediment all the gold. Particles of 10 nm or more will
labeling of Herpeswrus Suspensions
be sedimented for 45 mm at 27,000g. Colloidal gold centrifuges to a soft pool
rather than a pellet Any underlymg hard pellet represents aggregated gold and
therefore should be discarded
7. The centrtfugation step should be repeated to wash away free unbound protem
that might compete with the probe during labeling experiments, and also any
remaining tannic acid that might have an adverse effect on the proteins bound to
the gold particles (see Note 6)
8. After washmg, suspend the gold-protein complex m a small volume of TBSA-
azide-about one-tenth of the mitral volume of gold used in step 4 Store probes
in tightly sealed containers at 4°C and do not expose them to light unduly
3.3. Labeling of Virus Particles
for Electron Microscopy
3.3 1. Labehng of Viral Envelope Components
1. Concentrate the vmons by centrifugation and resuspend the pellet m approx 250
pL of PB.
2 Add 20-30 uL of the gold probe, mix well (color of the mtxture should be very
pale pmk) and incubate for 2 h at 37°C or longer at room temperature (see Note
7) All mnnunogold tests should mclude approprtate controls (see Note 8)

<<

. 27
( 61 .)



>>