Boston, Massachusetts, USA. Summary ... Infirmary, 243 Charles St, Boston, MA 0 2 1 14 , USA. ... NJ). isoproterenol HCl (Winthrop-Beron. Lab, New York.
Eye (1991) 5, 627-635
The Role of Cyclic Nucleotide Mediators in Latency and Reactivation of HSV-l Infected Neuroblastoma Cells ALEJANDRO RODRIGUEZ, MAITE SAINZ DE LA MAZA, JOHN MISSRY, C. STEPHEN FOSTER Boston, Massachusetts, USA
Summary The mechanisms that control herpes simplex virus type
1 latency and reactivation
are still poorly understood. We developed an in vitro murine neuroblastoma cell HSV-infected, acyclovir suppressed model to study the influence of different cyclic nucleotide mediators on the latency and reactivation of HSV-l. A positive cDNA 'in
situ' hybridisation for HSV genome was used to prove the establishment of a viral host cell nuclear relationship. An ABC-immunoperoxidase reaction to cell surface HSV mature glycoproteins was also performed to determine the time of viral reactiv ation with formation of mature virions. Supernates of cultured cells were placed on Vero cells for confirmation of reactivation by classic cytopathic effect. Theophylline
(SO [!g/ml) and dibutyril-cAMP (0.1, 0.5, 1 mg/ml) produced the most pronounced 150%. Epinephrine (10, 20 [!g/ml) had an intermediate effect on accelerating viral reactivation; and verapamil (20, SO [!g/ml), theophylline and epinephrine at lower doses had a smaller effect. Car bamylcholine (10 [!g/ml) prolonged the time to viral reactivation by 100%, 36 hours compared to control time of 18 hours. Insulin (0.1,0.5,1 mg/ml) also prolonged HSV
response, accelerating HSV reactivation time by
'latency' by six hours. Exogenous dibutyril-cGMP and carbamylcholine at lower concentrations did not have an effect on viral reactivation. These findings suggest that there is a relationship between changes of intracellular concentrations of cyclic nucleotides and HSV latency and reactivation.
Latently infected trigeminal ganglia have been thought to be the source of virus for recurrent herpetic keratitis in humans, a major cause of corneal blindness in developed countries. 1 Current antiviral therapy has proved ineffective in preventing the estab lishment of latency. The molecular details of the relationship between virus and host cell DNA during the latent state have not been
definitely elucidated. 2•3 An alternative thera peutic strategy to viral DNA elimination from latently infected ganglia might include efforts to indentify and manipulate factors involved in the control of mechanisms responsible for the induction of and reactivation from the latent state. A variety of physiological (fever, trauma, sunburn, menses, stress) and experi mental (epinephrine, prostaglandins, ultra-
From: Hilles Immunology Laboratory, Massachuesetts Eye and Ear Infirmary, Harvard Medical School, Boston, MA. Supported by the Susan Morse Hilles Fund. Correspondence to: C. Stephen Foster, MD. , Hilles Immunology Laboratory, Massachusetts Eye and Ear Infirmary, 243 Charles St, Boston, MA 0 2 1 14 , USA.
628
ALEJANDRO RODRIGUEZ ET AL
violet light) stimuli can be associated with HSV-l reactivation in patients and in experi mental models respectively.4- 1 II These 'trig
gers' may influence, through the action of first order messengers such as catecholamines
Park, NJ) were used to grow and maintain the cell lines in an incubator at 37°C and 5'X, CO, during the experiments.
Mediators: Mediators that increase intracel
and/'Jf arachidonic acid metabolites, cyclic adenosine monophosphate (cAMP) or cyclic
lular levels of cAMP: N6-2'-O-dibutyrilade
trations within latently infected neurons. We
epinephrine (Elkins-Sinn. Inc.. Cherry Hill,
nosine 3',5' cyclic'monophosphate sodium
guanosine monophosphate (cGMP) concen
salt
have already shown in an in vivo and in vitro murine trigeminal ganglion latency model
NJ).
cAMP and cGMP may influence maintenance
maceuticals. Whippany. NJ). Mediators that
(purity
=
97%) (Sigma, St Louis, MO),
isoproterenol
HCl
(Winthrop-Beron
Lab, New York. NY). theophylline (Elkins
that the relative proportions of intracellular
Sinn. Inc.) and verapamil HCI (Knoll Phar
of (elevated cGMP with no change in cAMP
increase intracellular levels of cGMP: N2,2'
latency.II
phate sodium salt (purity-96°/c)
levels) or reactivation from (elevated cAMP) In
order
to
study
further
the
O-dibutyrilguanosine 3"5' cyclic monophos (Sigma), car
relationship between intracellular neuronal
bamylcholine (Alcon Inc., Humacao. Puerto
studied the effects of putative mediators of
(Eli Lilly and Co .. Indianapolis. IN). Other
cyclic nucleotide levels and HSV latency we
Rico) and regular recombinant-DNA insulin
intracellular cyclic nucleotide levels on an
mediators used: sodium cromoglycate (Fisons
infected, acyclovir-suppressed model using
sodium phosphate (Elkins-Sinn. Inc.), cyclos
detection) and avidin-biotin immunoperoxi
Hanover,
in vitro murine neuroblastoma cell HSV
cDNA 'in situ'
hybridisation (viral DNA
dase (mature virion detection) techniques to determine the time of viral reactivation (posi
tive cDNA 'in situ' hybridisation, positive
ABC
immunoperoxidase)
from
'latency'
(positive cDNA in situ' hybridisation, nega '
tive ABC immunoperoxidase) for each medi
ator. These findings may have implications for clinical ocular HSV recurrent infections.
Corp.,
Bedford.
MA).
dexamethasone
porin A (Sandoz. Pharmaceutical Corp., East 2 NJ). 1
Experiment design: Neuroblastoma cells were trypsinised and plated into 6-well, 9.62 cm2
flat bottom tissue culture plates (Becton
Dickinson 10.5
x
contammg
Labware)
a
22 mm Thermanox tissue culture plas
tic coverslip (Miles Lab Inc .. Naperville, II). Each experiment was performed with its own
set of controls and was replicated at least four
Material and Methods
times to demonstrate the reproducibility of
Cell lines: C-1300 murine neuroblastoma cells
the results.
ville, MD) were cultured using Nutrient Mix
HSV-J infection of neuroblastoma cells: After
(American tissue Culture Collection, Rock
ture F-12 with L-glutamine media (Gibco Laboratories,
Grand
Island,
NY)
sup
plemented with 5% qualified fetal bovine
three days of cell incubation of 37°C, 5% CO2, the cultures were infected with 1 mllwell of HSV-l KOS strain (titer
=
2
x
10K PFU/ml,
serum (FBS) heated and inactivated at 60°C
from Dr David Knipe.
5,000 mcglml of streptomycin (Gibco Labs),
ent Mixture FO-12 to an MOl of 2. for one
Inc., McLean VA), and 250 mg of gentamicin
and the wells were washed and rinsed three
for
one
hour,
5,000 U/ml
of
penicillin,
250 mcglml of fungizone (Squibb Flow Labs,
sulfate (Gibco Labs). Vero cells from monkey
kidney (CCL8!, ATCC) were cultured with
Minimum Essential Medium (MEM) with
Earle's salts (Gibco Labs), 5% FBS and peni cillin-streptomycin-fungizone-gentamicin sol ution
as above. 75
cm' sterile cell culture
flasks (Becton-Dickinson Labware, Lincoln
Harvard Medical
School, Boston. MA) diluted with FBS Nutri
hour. The viral suspension was then removed
times with FBS Nutrient Mixture F-12 fol
lowed by a post-infection incubation period of six hours with 5% FBS Nutrient Mixture F-12.
Slippression of HSV-J with acyclovir: After
the
post-infection
HSV-l
incubation
period, the
infected neuroblastoma cells were
THE ROLE OF CYCLIC 'JCCLEOTlDE MEDIATORS IN LATENCY AND REACTIVATION OF HSV-J LATENCY
treated with 10 [lg/ml of acyclovir (ACV ) (Burroughs-Welcome,
Res. Triangle Park,
NC) and 1 [lg/ml of 7S-Nerve Growth Factor (7S-NGF)
(Collaborative
Research
629
applied for 10 minutes after removal of the
glass coverslip. The coverslips were rinsed with wash buffer for 10 seconds and the detec
Inc.,
tion reagent (avidinbiotinylated horseradish
induce suppression of HSV-l . U At the end of
coverslips were rinsed with wash buffer, coun
Bedford, MA) everyday for four days to day four, the cells were washed and rinsed three times with FBS Nutrient Mixture F-12 to remove the ACVnS-NGF solution.
ation time: Cyclic nucleotide mediators at dif
doses
HSV-infected,
were
then
added
to
terstained with fast green FCF and mounted with gelvatol for light microscopy.
ABC immllnoperaxidase staining: The other
Time-course for detection of HSV-J reactiv
ferent
peroxidase) was applied for 10 minutes. The
the
half of the coverslip was also rinsed in PBS (pH 7.2) and fixed with acetone as above.
Coverslips were incubated with a 5% normal
ACV-suppressed neuroblas
goat serum solutions for 30 minutes. A rabbit
(unreported) were performed to determine
Corp., Santa Barbara, CA) was diluted to
consisting of the removal of the supernate and.
microfilter
tivation on Vero cellsl� and immunohisto
cells for 15 minutes on a rotating table. A
time course was 4i1 to 72 hours depending on
centrifugation with 2800 rpm for 10 minutes at
Controls: These consisted of neuroblastoma
room temperature in a moisture chamber.
without
with PBS (pH 7.2) for 10 minutes. HP2
toma cells. Preliminary dose-response studi.es the mediator toxicity limits. A time-course a set of coverslips every two hours for cocul
chemical staining respectively was done. The the type of mediator used.
1: 100 with
1%
BSA, filtered with a 0.2 [lm
(Gelman
Sciences
Inc.,
Ann
Arbor, MI) and absorbed with neuroblastoma
further dilution to 1:1600 with 1% BSA and
25°C was followed by adding it to the cover
slips for 45 minutes (primary incubation) at
cells on coverslip without HSV infection and ACVnS-NGF
anti-HSV-l antibody (IgG anti-HSV , Dako
(positive-control);
and a set of coverslips HSV-infected, ACV
The coverslips were then rinsed three times
0.3'};" was added to each coverslip for 30 min
culture
utes. A rabbit ABC Vectastain kit (Vector
alone, which were withdrawn along with the
used to follow the staining. The secondary
(model control).
coverslip; one drop of goat anti-rabbit [gG
suppressed,
incubated
with
fresh
media and nerve growth factor (1 �lg/ml) mediator coverslips during every time-course
eDNA
'in
situ'
Laboratories [nc., Burlingame, CA) was then
incubation was performed after rinsing the
biotinylated antibody was applied for 45 min hybridisation:
Withdrawn
coverslips were left to dry at room temper
utes, followed by a rinse with PBS. The ter
tiary
45
minute
incubation
was
the
avidin-biotinylated
Diagnostics, Inc., New York, NY ). Therma
aminoethycarbazole substrate for 20 minutes.
with an HSV cDNA probe assay (ENZO
peroxidase
with
ature for at least two days and then stained
per
formed before reacting the coverslips with the
nox coverslips were cut into two equal pieces;
Coverslips were rinsed with acetate buffer
stain; the other was rinsed with PBS (pH 7 .2)
glacial acetic acid was applied before coun
one was used for the ABC immunoperoxidase and fixed with 100% acetone for 10 minutes.
One drop of the DNA probe reagent was then
applied to each Thermanox tissue culture
coverslip and a glass coverslip was then placed
(pH 5.0) and fixed with 4% formaline; 1%
terstaining
with
Gill's
hematoxylin
and
mounting with gelvatol for light microscopy.
Cocultivation of supernates on Vera cells: All
to cover the preparation which was then
supernates from the time-course (mediator
utes. After removing the coverslip from the
and placed on confluent monolayers of Vero
placed onto a 92°C heating block for 3-4 min
block, hybridisation proceeded at room tem
perature for 20 minutes in a moisture cham ber.
The
post -hybridisation
reagent
was
samples and controls) coverslips were filtered
cells and observed daily for 10 days, using an
inverted light microscope for cytopathic effect (CPE).
630
Fig. 1.
ALEJANDRO RODRIGUEZ ET AL.
Positive eDNA 'in situ'
hybridisation for HSV in HSV- 1 infected neuroblastoma cells (4 0x)
Statistics: Differences in time for HSV-l reac tivation (positive ABC immunoperoxidase staining) for each mediator compared to con trol were analysed using the Analysis of Vari ance test (p