<<

. 53
( 61 .)



>>

Edlted by S M Brown and A R MacLean Humana Press Inc, Totowa. NJ


345
Speck and Efstathiou
plaques produced by recombinant vtrus expressing 1acZ. The kinetics of
expression of 1acZ m neurons infected with recombinant HSV has been exam-
med during acute and latent phases of infection (2,3)
2.1.1. Method
1. Dissect ttssue (e g , sensoryganglia) from infected ammal and place in phos-
phate-buffered salme (PBS) at 4°C until ready to fix
2 Fix tissue m 2% paraformaldehyde, 0.5% glutaraldehyde in PBS at 4°C for 1 h m
the case of mouse ganglia (larger pteces of tissue may require longer ttmes m
fixative), and then wash out fixative with PBS
3 Permeabtlize ttssue by mcubatton at 4°C for 30 mm in 2 mMMgClz, 0.01% sodium
deoxycholate, and 0 02% Nomdet P-40 (Ftsons, Loughborough, UK)
4 Remove this mixture, and replace with the same solution contammg 5 mM
potassium ferrtcyamde-ferrocyanide plus X-gal (1 mg/mL) Incubate at 37T for
2-6 h, checking periodically using dissectmg microscope for the appearance of
blue-stained neurons
5 Wash out X-gal solution wtth several changes of PBS, and clear by mcubatton
for 2-12 h m 20% glycerol at 4°C Stained tissue can be vlsuahzed under a mtcro-
scope by placmg between two slides m a wet mount If desired, tissue can be
embedded m paraffin, and sections cut on a mtcrotome In our experience, the
blue color produced m the procedure above is not washed out during the embed-
dmg process
2.2. Detection In Situ of Viral Nucleic Acids
ISH ts a powerful technique that, vta hybridization of indicator molecules,
permits visualization of the cellular and possibly subcellular location of target
sequences. An increasing number of indicator molecules, radtoacttve and non-
radioactive, are available. Compared to radtoactive indicator molecules, non-
radioactive alternatives, such as digoxigemn, which show the presence of a
probe by a colored precipitate or by fluorescence, offer improved resolution at
the subcellular level (4) and long storage life, while avoiding difficulties asso-
ciated with radtoacttvtty. However, enumeration of target molecules from such
signals remains problemattc. Autoradiographic detection of radtoacttvtty pro-
vides a final readout m the form of silver grains, which can be counted, permtt-
ting an approximation of the number of target molecules to be made, based on
specific activity of the label and assumptions regarding efficiencies of hybrid-
ization and autoradiography. On this basis, we have estimated that mouse neu-
rons latently infected with HSV may each contam on the order of 1O4copies of
LAT (5). For excellent discusstons of the quantification of autoradiography,
see Rogers (6) and Gowans et al. (7)
As radioactive and nonradioactive systems, each have advantages. There-
fore, both are described in this chapter.
347
Assays for HSV Gene Expression
2.3. Fixation of Tissue
In our hands, the fixative of choice for tissue samples IS periodate-lysme-
paraformaldehyde (PLP; ref. 8) owing to its suitability for preservation of
nucleic acids and antigens (9-l I). However, if the preservation of nucleic acids
is the sole consideration, then other fixatives, such as 4% paraformaldehyde or
0.5% gluteraldehyde, each in PBS, may give satisfactory performance (see
Note 1). After impregnation with paraffin, tissue sections should be collected
onto slides treated with a suitable coating to ensure adheston of sections during
subsequent procedures. Glutaraldehyde-activated 3-aminopropyltriethoxy-si-
lane-coated slides (12) are adequate for this purpose.
2.4. Choice of Probe
Probes synthesized from recombinant DNA are recommended because they
are less likely to contain contaminating sequences than even highly purified
naturally dertved nucleic acids (13) RNA probes have a number of advan-
tages: first, being single stranded, there is less tendency to self-anneal, which
effectively mcreases probe concentration, thereby maximizing sensitivity (ZJ),
second, opposite sensetranscripts can be used as controls for nonspecific bind-
mg. These benefits are considered sufficient to justify the effort necessary to
subclone DNA sequences mto suitable plasmids, such as pBluescript@, which
contain bacteriophage T3 and T7 promoters.
2.5. Choice of Indicator Molecule
2.5.1. Radioactive Isotopes.
The ideal radioactive label would provide high resolution coupled with
high sensitivity at short exposure times. Five isotopes are in common use
m ISH: 3H, 1251, 35S, 33P, and 32P, In practice, none of them are ideal. A
detailed discusston of the decay characteristics of each isotope is beyond
the scope of this chapter (see ref. 6 or 24). However, the approximate path
lengths in nuclear emulsion of emissions from the above isotopes are 1, 2,
10-20, 15-20, and >20 p, respectively (6). 125I therefore provides much
higher resolution in ISH than does 32P, 33P, or 35S, while retaining suffi-
ciently high energy to permit exposure times similar to those used with 32P.
Exposure times with 3H are on the order of weeks or months, which may be
considered unacceptably long. Based on the preceding constderattons, 1251-
CTP is recommended for use m RNA probes. Despite the 60-d half-life of
1251, our hands, it is found that the signal-to-noise ratio is compromised
in
if label is used later than 1 mo after manufacture. A disadvantage of the use
of *251-CTP is that m our hands, there appear to be large batch-to-batch
variations m the signal-to-noise ratio obtamed.
Speck and Efstathiou
348
2.5.2. Digoxigenin: A Nonradioactive Indicator Molecule
In our hands, the nonradioactive indicator molecule digoxigenin (available
as digoxigenin-UTP for transcription reactions; Boehrmger Mannhelm,
Mannheim, Germany) yields excellent results in ISH applications, providing
excellent resolution at the subcellular level and giving sensitivity comparable
to that achieved with 1251. Preparations of this probe do not appear to suffer
loss of sensitivity after >2 yr of storage at -20°C nor have we experienced any
batch-to-batch vartations m the behavior of digoxigenm-UTP. Dtgoxigemn is
therefore the indicator molecule of choice, unless there is an absolute reqmre-
ment that the ISH signal be quantified.
2.6. Preparation of RNA Probes
2.6.1. Radloactwe Probes
RNA probes are prepared according to the method supplied by the makers
of the Riboprobe@ RNA labelmg kit (Promega, Madison, WI), and efficiency
of mcorporation IS measured by differential precipitation of RNA by trichloro-
acetic acid (25) In our experience, optimal performance of an RNA probe is
associated with >75% incorporation of available radiolabel into RNA tran-
scripts. Probes of 0.5-I 5 kb m length are used without hydrolysis The recom-
mended specific activity of ˜251-labeled RNA probes is 5 x lo8 dpm/pg, and
probes should be used at a concentration of 40 pg/pL (7)
2 6.2. Nonradioactive Probes
Incorporation of dtgoxigenin mto RNA probes IS carried out by following
the method given by the makers (Boehrmger Mannheim), which resembles
that used for incorporation of radioactively labeled nucleotrdes mto RNA, instead
using drgoxigenm-UTP. Optionally, a small amount (0.5 yL) of 32P-dCTP may
be included in the reaction mix, so that the efficiency of incorporation may be
measured by the trichloroacetic acid differential precipitation method (15).
2.7. Preparation of Tissue Sections for ISH
This method IS an adaptation of that of Gowans et al (7)
Dewax paraffin sections m xylene, and rehydrate gradually through graded etha-
1.
nol/water mixtures over a pertod of 1 h (see Note 4)
2 Fix in 0 1% glutaraldehyde in PBS for 30 min at 4™C, and wash in PBS (2 x 5 min)
3. Digest tissue secttons with protemase K m 20 rnMTris-HCl, pH 7.4,2 mA4CaC12
for 15 mm at 37°C. The opttmum concentratton of protemase K should be deter-
mined emptrtcally and may vary from batch to batch In our hands, a concentra-
tion of 100 yglmL gives strong autoradiographic signal wtth good preservation
of tissue morphology
349
Assays for HSV Gene Expression
4. Wash secttons in phosphate-buffered saline (PBS), refix m 0 1% glutaraldehyde
for 15 mm, and wash twice more in PBS
5 Acetylate secttons by immersmg in a freshly made solution of 0 25% acetic
anhydride m 0.1 A4 triethanolamine (pH 8 0) for 10 mm at room temperature with
gentle agitation (see Note 5)
6 Wash sections twtce in PBS, dehydrate gradually, and dry prior to application of
hybridization mtx
2.8. Hybridization
ISH follows the same general principles as filter hybridizatton, so that the
optimum temperature for hybridization is T,--25°C (T,, the temperature at which
50% of double-stranded nucleic acid hybrids are m liqurd dtssoclatedinto smgle-
stranded molecules, IS calculated as in Note 6), and the most stringent wash
should be at Tm---10°C. careful adherenceto thts guideline, we have found that
By
a posthybrldtzatron ribonuclease digestton step 1snot requned. The addttron of
50% formamrde to the reaction mixture has the effect of lowering the reaction
temperature to a level compatible with preservation of histologtcal detail.
The final hybridization mix contains 40 pg/pL RNA probe m 50% deiontzed
formamlde, 1X SSC, 100 mM Trts-HCl, pH 7.6, 10 mM Na*HPO+ 10 n&I
NaH2P04, 0.02% Ficoll, 0.12% polyvinyl pyrollidone, 500 ,ug/mL sheared dena-
tured salmon sperm DNA, 500 pg/mL yeast tRNA, 1.25 mg/mL nuclease-free
bovine serum albumin (BSA), 20 mM dithiothrettol (DTT), and 1 U/pL ribo-
nuclease inhibitor.
2.8.1. Method
1. Heat mix to 80°C for 5 min, and store on ice prior to addttton to slides
2 Apply hybridization mtx (3 pL) to sections, and cover with a 13-mm sterile sill-
conized cover slip, taking care not to trap air bubbles. For larger sections
increase the volume of probe mix in proportion to cover slip area
3 Seal cover slips with rubber cement (e g., repair adhesive 4051, PMD,
Bridlmgton, Yorkshire, UK), and incubate slides for 8-15 h at T,,,--25°C For
probes complementary to the LAT region of HSV (1 e , having a G + C content of
approx 75%), this temperature 1s 75°C.
2.9. Washing Procedure
1. Using tine-tipped forceps, remove rubber cement and cover slips, and immedt-
ately place slides in a large volume of 2X SSC at room temperature. If the probe
solution dries onto the sectron, this will be associated with a significant Increase
in background. Wash m 2X SSC for 1 h, and then twice in 0.1X SSC for 1 h at
room temperature.
2. High-strmgency wash. Wtth gentle agitation, wash slides for 20 mm at T,,,--10°C,
which for probes complementary to the LAT region of HSV is 75°C m 30%
deiomzed formamide, 0 1X SSC.
Speck and Efstathiou
350
3. Wash m 0.1X SSC for 30 mm at room temperature to remove traces of forma-
mide from sections.
4 Dehydrate slides through graded ethanol solutions, and air-dry prior to dipping m
nuclear emulsion
When using radioactive iodine as an indicator molecule, potassmm iodide
(100 tr&!) 1s included in all washes to reduce background.
2.10. Autoradiography and Staaining for Radioactive Labels
For a comprehensive dlscusslon of the propertles and uses of nuclear emul-
sions, see Rogers (6). We have found Amersham LM-1 emulsion sultable for
the detection of ˜251-labeled RNA probes. To minimize background grains, care
is required m the handling of nuclear emulsion, in particular, sudden changes
m conditions, such as pH or temperature, should be avoided.
A suitable quantity of Amersham LM-1 emulsion 1s mixed with an equal
volume of water m a darkroom equipped with safelight recommended by
Amersham and melted at 42-44”C for 30-40 min. The diluted emulsion is
carefully mixed, and slides are dipped vertically, set immediately on an ice-
cold flat metal plate for 30 min, dried for 1 h to overnight at room temperature,
and stored in a light-tight box at 4°C for 2-5 d. Exposed slides are developed
m Kodak D-19 developer for 4 mm at 23”C, with gentle agitation for 5 s at
30-s intervals, rinsed in 1% acetic acid stop bath for 30 s, and placed tn Ilford
Hypam rapid fixer (diluted 1:4) for 8 mm. Optimal development time should
be determined emptncally. Slides are then washed in tap water for 30 mm,
stained with rapid hematoxylm (7) for 3w5 s, and “blued” in 0.1X SSC for at
least 30 mm. Sections are dehydrated m graded ethanol solutions, stained with
eosm for 1 min, washed in ethanol (3 x 1 mm), washed in xylene (2 x 10 mm),
and “coverslipped” using DePex (BDH, Poole, UK) mountant.
2.7 7. Color Development: Digoxigenin ISH
This method 1s based on that described by the manufacturer (Boehrmger
Mannhelm).
1 Using coplin Jars, wash slides for 5 mm in buffer 1 (100 mA4 Tns-HCl, pH 7.5,
150 mMNaC1).
2 Incubate for 30 mm in blocking reagent (Boehrmger cat no. 1096- 176, 1% w/v m
buffer 1 This can be dissolved by heating with stm-mg on a hot plate for 30 mm)
3 Wipe slides to remove bulk of blocking reagent, and incubate sections with
antldlgoxigenm alkaline phosphatase conjugate, l/l000 in blocking solution, for
30 mm at 37°C in a humid box, about 50 pL/section
4. Wash 2 x 15 mm in buffer 1, rinse briefly in buffer 3 ( 100 mA4 Tns-HCl, pH 9 5,
100 mM NaCl, 50 mM MgCQ.
5. In fresh coplin jars, place slides m substrate solution (5-bromo-4-chloro-3-indolyl
351
Assays for HSV Gene Express/on
phosphate/Nitro blue tetrazolmm, made up accordmg to the manufacturer™s
instructions) for l-5 h m a dark box.
6. Wash several times m runnmg water and counterstain briefly ( 1&15 s) in hema-
toxylm wtth further washes m water for several minutes to ensure color differen-
tlation of counterstain.
7 Visualization of slides under wet mounting is sufficient to enable photography.
2.12. lmmunohistochemical Detection of Viral Antigens
The peroxidase-antrperoxidase method (16-18) usmg 3,3™-draminobenzene
(DAB; Life Technologies, Paisley, UK) as a substrate, when opttmized, gives
strong signal over HSV-infected cells with very low levels of background. DAB
also has the beneftt of forming a precipitate that is insoluble in the reagents used
m any subsequent ISH procedure. The primary antibodtes could be polyspecific,
for example, rabbit antiserum to HSV-infected cells (Dakopatts, Glostrup, Den-
mark), or monospecific, such as monoclonal antibody (MAb) 58-S (American
Type Culture Collectron HB8183) to ICP4. Binding of primary antisera 1s
detected by usmg as secondary antibody swine antirabbit or goat anttmouse
tmmunoglobulm as appropriate, followed by rabbit or mouse peroxtdase-
antiperoxidase conjugate, respectively (all from Dakopatts, Glostrup, Denmark).
We have used this method to detect a number of HSV-encoded proteins, mclud-
ing ICP8 and Vmw65, m addition to those detected as above (5).
1 Dewax slides m xylene, and hydrate through decreasing concentrations of etha-
nol, ending in Tris buffer (Tns-HCl, pH 7.5, 50 mM)
2 Immerse sections in Tris buffer containing 0 01% H,Oz for 10 mm at room tem-
perature, to block endogenous peroxidase activity.
3. Overlay sections with blocking solution (Tris buffer contammg 10% normal serum
of the species providmg the secondary antibody), and leave at 37°C in a humid
box for 30 mm (see Note 2).
4 Wash for 2 x 5 mm m Tris buffer.
5. Incubate secttons first with primary antibody, followed by secondary antibody,
and then peroxidase-antiperoxidase conjugate, with washes as m step 4 between
each mcubation. Antibodies and conjugate should be diluted m blocking solu-
tion, with the optimum dilutions determined empirically (see Note 3). Allow all
reactions to proceed for 30 min at 37°C m a humid box.
6. Wash sections as above, and then incubate slides in a solution of DAB (0 5 mg/mL,
containing 0.1% H202) for 2-5 min in the dark, with the actual time determined
empirically. If sections are not to be subsequently subjected to ISH, they may be
lightly counterstained with hematoxylm

2.13. Dual-Labeling
Simultaneous detection of nucleic acids and proteins on the same tissue sec-
tion (dual-labeling; refs. 7 and 19-21) enables the presence of viral mRNA to
352 Speck and Efs ta thiou

be correlated with the presence of virally encoded protems. Using dual-label-
ing for viral antigens and for LAT during acute infection with HSV in mouse
spinal ganglia, we have demonstrated that neurons with the characteristics of
latency (i.e., expressing LAT in the absence of antigens) are present from the
earliest detectable stages of acute infection in the nervous system (22)
In our experience, the dual-labeling method of Gowans et al. (7) (In which
immunohtstochemistry precedes radioactive ISH) has proven satisfactory. An
alternative approach rn whtch m nonradioactive ISH 1s carried out first has
been described (23) We have not attempted to quantify formally the loss of
sensrttvtty of ISH owing to precedtng tmmunohtstochemtstry. However, tn our
hands, there does not seem to be a marked reduction m signal strength. It should
be borne m mind that any dual-labeling procedure may lead to some loss of
senstttvtty tn each of the component assays.
As a precaution against degradation of RNA m sections, all tmmunohts-
tochemical reagents used should be sterilized.

<<

. 53
( 61 .)



>>