In vitro establishment of lytic and nonproductive infection by herpes simplex virus type 1 in three-dimensional keratinocyte culture.

Abstract

The F strain of herpes simplex virus type 1 (HSV-1) was tested for its ability to produce lytic or nonproductive infection in squamous epithelial cells cultured in a three-dimensional organotypic tissue culture. For the tissue culture, we used HaCat cells (immortalized skin keratinocytes) and normal fibroblasts derived from the skin. The cultures were infected with HSV-1 (5 PFU) either when the epithelial cells had grown as a monolayer with a confluence of 80% on the collagen fibroblast gel or 30 min after lifting of the epithelial cells into the air-liquid interface. The cultures were collected 1 week after inoculation. Typical cytopathic effects of HSV infection (ballooning and reticular degeneration with multinucleate giant cells) were seen only in those cultures in which the epithelial cells were infected before lifting. The presence of HSV was confirmed by DNA and RNA in situ hybridization and PCR. No morphological changes were found in cultures infected after lifting into the air-liquid interface. No infectious virus was recovered either from cells or culture supernatant. However, these cultures were positive for HSV DNA on PCR and showed expression of the LAT gene by in situ hybridization and Northern blot (RNA) hybridization. The present results indicate that both nonproductive and lytic HSV infection can be produced in vitro and the outcome of the infection depends on the time of viral inoculation in relation to epithelial maturation.

Full text

Page 1

JOURNAL OF VIROLOGY, Sept. 1996, p. 6524–6528
0022-538X/96/$04.00?0
Copyright ? 1996, American Society for Microbiology
Vol. 70, No. 9
In Vitro Establishment of Lytic and Nonproductive Infection by
Herpes Simplex Virus Type 1 in Three-Dimensional
Keratinocyte Culture
STINA SYRJA¨NEN,1,2* HANNAMARI MIKOLA,2,3MARJA NYKA¨NEN,2AND VEIJO HUKKANEN2,3
Department of Oral Pathology, Institute of Dentistry,1MediCity Research Laboratory, Faculty of Medicine,2and
Department of Virology, Faculty of Medicine,3University of Turku, Finland
Received 11 December 1995/Accepted 2 April 1996
The F strain of herpes simplex virus type 1 (HSV-1) was tested for its ability to produce lytic or nonpro-
ductive infection in squamous epithelial cells cultured in a three-dimensional organotypic tissue culture. For
the tissue culture, we used HaCat cells (immortalized skin keratinocytes) and normal fibroblasts derived from
the skin. The cultures were infected with HSV-1 (5 PFU) either when the epithelial cells had grown as a
monolayer with a confluence of 80% on the collagen fibroblast gel or 30 min after lifting of the epithelial cells
into the air-liquid interface. The cultures were collected 1 week after inoculation. Typical cytopathic effects of
HSV infection (ballooning and reticular degeneration with multinucleate giant cells) were seen only in those
cultures in which the epithelial cells were infected before lifting. The presence of HSV was confirmed by DNA
and RNA in situ hybridization and PCR. No morphological changes were found in cultures infected after lifting
into the air-liquid interface. No infectious virus was recovered either from cells or culture supernatant.
However, these cultures were positive for HSV DNA on PCR and showed expression of the LAT gene by in situ
hybridization and Northern blot (RNA) hybridization. The present results indicate that both nonproductive
and lytic HSV infection can be produced in vitro and the outcome of the infection depends on the time of viral
inoculation in relation to epithelial maturation.
Productive infection with herpes simplex virus (HSV) is
characterized by coordinately regulated and sequentially or-
dered expression of three classes of genes: the ?, ?, and ?
genes (9, 10). The cells lytically infected with HSV undergo a
series of structural alterations, with cell death as the final
outcome (18). Lytic infection with HSV involves an early shut-
off of the host cell metabolism, attributed to the ? class protein
vhs (virion-host shutoff) (6, 12), and a secondary shutoff, ap-
pearing late during the infection and requiring another viral
protein, ICP27 (5, 7). These shutoff functions contribute to the
fate of the infected cell. The typical changes observed in pro-
ductively HSV-infected cells include chromatin margination,
disaggregation of the nucleolus, and modification of the cellu-
lar membranes (18). The cytopathic effects observed at the
tissue level (i.e., in the infected epithelium) are a reticular
degeneration and ballooning of the cells as well as the appear-
ance of large irregular or multinucleated giant cells.
Latency is the hallmark of herpesvirus infections (18). HSV
establishes latent infection in sensory neurons (19). In the
sensory ganglion, the latent HSV genome resides in the nu-
cleus of the neuron, expressing only one of its genes, the gene
for latency-associated RNA (LAT) (22, 25). The LAT RNA is
predominantly localized in the neuronal nucleus, and it con-
sists of several overlapping RNA species with different termini
(4). Although LAT does not apparently code for a protein, it is
a widely used marker in the demonstration of latently infected
cells (19, 21).
A variety of in vitro models have been developed for the
study of HSV latency in different cell types (3, 13, 16, 17, 23).
However, the restriction of HSV replication by different treat-
ments of cultured cells may not accurately reflect the events
leading to establishment of HSV latency in the sensory neu-
rons of animals and humans (19).
In the present report, we describe a novel approach to the
study of HSV infection at the tissue level. We have used a
three-dimensional epithelial culture system (1) which pro-
motes the differentiation of human keratinocytes. These raft
cultures have been recently used in studies of other viruses,
particularly human papillomavirus (14). In HSV-infected or-
ganotypic raft cultures, we have observed cytopathic changes
identical to those found in the squamous epithelium in vivo,
including the formation of typical intraepithelial vesicles and
multinucleation. Moreover, depending on the time point of the
HSV infection relative to the confluence of the cell layers, a
nonproductive form of HSV infection develops in the culture.
Culture of HaCat cells and normal skin fibroblasts. The
HaCat cells were grown in Dulbecco’s modified Eagle’s me-
dium (DMEM) supplemented with 1% nonessential amino
acids, 2 mM L-glutamine, 50 ?g of streptomycin per ml, 100 U
of penicillin per ml, and 10% fetal calf serum. To establish the
fibroblast culture, a biopsy from uninfected skin was trans-
ported to the laboratory in DMEM containing 100 ?g of strep-
tomycin per ml, 200 U of penicillin per ml, and 0.5 ?g of
Fungizone (Gibco Laboratories, Grand Island, N.Y.) per ml.
The connective tissue was cut into small fragments of 1 mm3,
and explant cultures were started according to standard tech-
niques. The fibroblasts were grown in supplemented DMEM.
Skin fibroblasts derived from the explants were from their fifth
passage.
Preparation of reorganized collagen gel. Vitrogen 100 col-
lagen (Celtrix Pharmaceuticals, Inc., Santa Clara, Calif.) was
mixed with 10? DMEM and neutralized by 0.1 M NaOH to
pH 7.4 ? 0.2. Fibroblasts were suspended in the collagen
solution at a cell density of 280,000 cells per 0.7 ml, and this
suspension was plated in 16-mm-diameter tissue culture dishes
* Corresponding author. Mailing address: MediCity Research
Laboratory, Faculty of Medicine, Tykisto ¨katu 6, 20520 Turku, Finland.
Phone: 358 21 3338349. Fax: 358 21 3338399. Electronic mail address:
stina.syrjanen@utu.fi.
6524

Page 2

(Costar, Cambridge, Mass.). The collagen-fibroblast suspen-
sion was allowed to gel at 37?C for 1 h, after which the dishes
were filled with fresh Green’s medium. The medium consisted
of DMEM-Ham F12 (3:1), 10% fetal calf serum, 4 mM glu-
tamine, 5 ?g of insulin (Sigma, St. Louis, Mo.) per ml, 0.18 mM
adenine (Sigma), 0.4 ?g of hydrocortisone (Sigma) per ml, 0.1
nM cholera toxin (Sigma), and 5 ng of epidermal growth factor
(Boehringer Mannheim, Germany) per ml and was maintained
at 37?C in an atmosphere of 5% CO2in 90% relative humidity.
Medium was changed three times a week for 1 week.
Epidermal cell culture. HaCat cells, passage 15 (200,000
cells per well), were gently added onto the surface of the
fibroblast-collagen gels. The Green’s medium was changed ev-
ery second day. After 3 days, when the cells had reached
confluence, the cultures were lifted into the air-liquid interface
with a stainless steel grid. Epithelial cells were then allowed to
stratify for 7 days. One day before the lifting, cultures (with a
confluence of 80%) were infected with HSV type 1 (HSV-1).
Additional cultures were infected with HSV-1 30 min after
lifting, while the other uninfected cultures served as negative
controls.
HSV-1. The F strain of HSV-1 was obtained from the Amer-
ican Type Culture Collection (ATCC, Rockville, Md.) and
propagated in human foreskin fibroblasts. The infected cul-
tures were centrifuged at 4?C in a Sorvall Technospin R cen-
trifuge at 2,700 ? g for 10 min. The pellet was resuspended in
a small volume of phosphate-buffered saline, frozen and
thawed three times, and sonicated for 3 min at 40% duty,
continuous cycle. The infectious titer of the virus preparation
was determined by standard plaque assay. Virus was stored at
?70?C at a concentration of 5 ? 106PFU/ml. For each well, 5
PFU were used.
Fixation and preparation of the tissue cultures. All tissue
cultures were collected at day 18 (1 week after infection) and
divided into two parts. One part was fixed in buffered 10%
formalin for 24 h, while the other was snap frozen and kept at
?70?C for further PCR analysis. The formalin-fixed material
was embedded in paraffin and cut into 5-?m sections onto
organosilane-coated slides for hematoxylin and eosin (HE)
staining and for DNA and RNA in situ hybridizations (ISHs).
ISH for HSV-1 DNA. As a probe for HSV-1, a biotinylated
commercial probe was used (Enzo Diagnostics, N.Y.). The
probe is a mixture of two clones of HSV DNA sequences in the
BamHI site of pBR322, with insert sizes of 16.0 kb and 8.0 kb.
ISH was performed as previously described (24). Shortly after
deparaffinization and deproteinization with proteinase K
(Boehringer) (0.3 mg/ml) for 15 min at 37?C, the sections were
simultaneously denatured with hybridization mixture contain-
ing 2 ?g of biotinylated probe per ml in 2? SSC (1? SSC is
0.15 M NaCl plus 0.015 M sodium citrate), 50% formamide,
0.4 mg of herring sperm DNA per ml, and 10% dextran sulfate.
After overnight hybridization at 55?C, the sections were sub-
sequently washed with 2? SSC at room temperature, 0.2?
SSC at 60?C, and 2? SSC at room temperature. The hybrids
were labeled with streptavidin-alkaline phosphatase (Amer-
sham, Buckinghamshire, United Kingdom), and the complexes
were detected with nitroblue tetrazolium as chromogen and
5-chromo-4-chloro-3-indolylphosphate (BCIP) as substrate. A
biopsy specimen from the oral mucosa, previously shown to
contain HSV-1 DNA, served as a positive control. As negative
controls, all sections were treated similarly except that the
HSV-1 probe was omitted from the hybridization mixture.
ISH for LAT RNA. The digoxigenin (DIG)-labeled single-
stranded RNA probe for LAT RNA was a 560-nucleotide SP6
transcript of a plasmid containing the 0.5-kb HpaI-SalI sub-
fragment of the BamHI fragment B of HSV-1 DNA (gift from
Bernard Roizman, University of Chicago) cloned into a
pGEM-3Z vector (Promega, Madison, Wis.). The plasmid
DNA was linearized with EcoRI (Boehringer) and subjected to
transcription with SP6 polymerase and DIG-11-UTP (Boehr-
inger), as described previously (11). The probe for ?0 mRNA
was transcribed with T7 polymerase from the same plasmid
linearized with SalI. This RNA probe has a 230-nucleotide
sequence complementary to ?0 mRNA. The correct sizes of
the probes were verified by electrophoresis in a formaldehyde
gel. The ISH was done as described previously for DIG RNA
probes (11). It involved a prehybridization step with 300 ?g of
denatured salmon sperm DNA per ml and the use of mouse
brain nucleic acids (500 ?g/ml) as a blocking agent in the
hybridization buffer. The hybridization was performed for 18 h
at 45?C under sealed caps in a humidified incubator. The slides
were washed, as described in reference 20, without the use of
RNases during the washes. For detection of DIG-labeled hy-
brids, the sections were incubated with the alkaline phos-
phatase-conjugated anti-DIG antibody (Boehringer). The com-
plexes were detected with nitroblue tetrazolium–X-phosphate
color reagents (11). The slides were observed by using a
masked protocol. As positive controls, mouse trigeminal gan-
glia (TG) infected with HSV-1 were used. The negative con-
trols were sections from uninfected cultures and mice. The
specificity of the signals was tested with additional sections
with both RNase and DNase treatments. Slides were pre-
treated with RNase (100 ?g/ml; Sigma) or DNase (75 U/ml,
RNase-free DNase; Boehringer) for 30 min at 37?C before
hybridization. The slides were fixed again with modified Car-
noy’s fixative (11) for 2 h directly after RNase treatment.
Northern blot (RNA) analysis of LAT RNA. Preparations of
total RNA of the raft cultures were made with the TRIzol
reagent (Gibco BRL). Total RNA from latently infected [5
weeks postinfection, 106PFU of HSV-1(F) per eye] TG of
BALB/c mice was prepared by the guanidine thiocyanate-CsCl
centrifugation method. Twenty micrograms of total RNA or 15
?g of DNase-treated total RNA from raft cultures was run in
1% agarose-formaldehyde gel, stained with ethidium bromide,
and blotted on GeneScreen Plus (DuPont NEN) nylon mem-
brane. The filter was hybridized with a single-stranded RNA
probe, which was prepared by transcription from the same
template as the ISH LAT probe but with [32P]CTP (Dupont
NEN) as the labeled nucleotide. The hybridization took place
at 65?C in a solution containing 50% formamide, 3? SSC, 5?
Denhardt’s solution, 10% dextran sulfate, 1% sodium dodecyl
sulfate (SDS), and 250 ?g each of Escherichia coli transfer
RNA (Sigma) and single-stranded herring sperm DNA (Sig-
ma) per ml.
Extraction of DNA. For the PCR, DNA was extracted from
the frozen cultures by the method of Miller et al. (15). Shortly,
samples were lysed in 1 ml of 10 mM Tris (pH 8.3), 400 mM
NaCl, 1% SDS, 2 mM EDTA, and 0.3 mg of proteinase K per
ml overnight at 37?C. Proteins were precipitated by adding 300
?l of saturated NaCl. After centrifugation, supernatant was
removed and DNA was precipitated with ethanol.
PCR detection of HSV-1 DNA. The PCR test for HSV DNA
was modified from a procedure described for cerebrospinal
fluid specimens (2). The primers, detecting HSV-1 strains and
several HSV-2 strains (5?-ATC ACG GTA GCC CGG CCG
TGT GACA and 5?-CAT ACC GGA ACG CAC CAC ACAA)
and the internal hybridization probe (5?-TAC GAG GAG
GAG GGG TAT AAC AAA GTC TGT), were derived from
the glycoprotein D gene of HSV-1 (2). DNA extracts from
frozen cultures were tested in two dilutions each. The positive
controls were HSV-1 virions, strain F (ATCC), containing 1
PFU in each reaction. The dilutions were made from a high-
VOL. 70, 1996NOTES6525

Page 3

titer stock (109PFU/ml). Negative controls contained distilled
water. The working spaces for specimen handling, construction
of the reaction mixture, PCR cycling, and gel electrophoresis
were all physically separate from each other. The specimens
were pipetted with Finnpipette positive displacement pipet-
tors. The PCR contained 200 ?M each deoxynucleotide, 20
pmol of each primer, 1 U of polymerase (DynaZyme, Finn-
zymes, Espoo, Finland), and corresponding buffer provided
with the enzyme. The PCR steps were 95?C for 30 s, 55?C for
30 s, and 72?C for 60 s, for 41 cycles. The first cycle included an
incubation at 95?C for 5 min. The 72?C incubation in the last
cycle was extended by 4 min. The presence of a 221-bp reaction
product was observed by ethidium bromide staining after elec-
trophoresis in 2% agarose gels. The specificity of the product
was confirmed by hybridization to a DIG-labeled oligonucleo-
tide probe, which was detected with CSPD luminescent re-
agent (Boehringer) on X-ray film (8).
Detection of infectious HSV-1 by immunoperoxidase stain-
ing. The supernatants from and homogenates of the raft cul-
tures were cultivated on Vero cells on 12-well culture dishes
and studied for the presence of infectious HSV-1 by a previ-
ously described rapid detection system (26), which is based on
immunoperoxidase staining of the viral proteins after an over-
night culture, with a monoclonal antibody to HSV-1.
Morphology of the raft cultures. The HaCat cells grown in
the organotypic cell culture had the morphologic appearance
of a partially differentiated epithelium. Epithelial cells were
found to be organized into a thin sheet six to eight cell layers
thick. There was some attempt to form a basal cell layer, albeit
not one as regular as is normally found in vivo. A thin layer of
flattened superficial cells was seen, evidencing the terminal
differentiation (Fig. 1A). HSV established a productive infec-
tion of the epithelial cells only when the cells were infected
before lifting the epithelium at the air-liquid interface. This
lifting allows differentiation of the epithelium. Typical cyto-
pathic effects of a productive HSV infection were observed in
the HaCat cells grown in the organotypic cell culture (Fig. 1C
and D). Large irregular or multinucleate epithelial cells with
ballooning and reticular degeneration of the epithelium were
observed by light microscopy. Also, typical intraepithelial ves-
icles were detectable. When HSV was applied on the surface of
the lifted epithelium, no morphological changes indicating a
productive HSV infection were seen in these cultures, but the
morphology was similar to that of the uninfected control cells
(Fig. 1B).
HSV DNA and RNA in the raft cultures. The histopathologic
findings were confirmed by ISH for the detection of HSV-1
DNA with biotinylated probes. Cultures with the cytopathic
changes typical of HSV infection showed strong positivity for
HSV DNA by ISH (Fig. 2A and B). HSV DNA was seen in all
epithelial cells, but the most intense hybridization signals were
encountered in the multinucleate giant cells. Interestingly, no
HSV DNA positivity was found in the fibroblasts. The cultures
for which HSV was added on the surface of the lifted epithe-
lium remained negative for HSV DNA by ISH (Fig. 2C). How-
ever, PCR analysis of these cultures was repeatedly positive
(Fig. 3A and B). In the PCR runs, the low positive PCR
FIG. 1. (A) HaCat cells grown in the raft culture system. Epithelial cells are
organized into a thin sheet six to eight cell layers thick. A layer of flattened
superficial cells is present, evidencing the terminal differentiation. (B) Tissue
culture of HSV-1-infected HaCat cells. The virus was inoculated onto the epi-
thelium 30 min after lifting the culture into the air liquid interface. The mor-
phology was similar to that of the uninfected control cells. (C and D) HaCat cells
grown in the raft culture system and infected with HSV-1 at 1 day before the
lifting. Typical cytopathic effects of productive HSV infection with large irregular
or multinucleated epithelial cells, ballooning, and reticular degeneration of ep-
ithelium are seen. Typical intraepithelial vesicles are also present. HE staining.
Magnification ?250.
FIG. 2. ISH to detect HSV-1 DNA. (A and B) Strong signals are seen in all
epithelial cells showing cytopathic effects of productive infection. Panel A mag-
nification, ?250; panel B magnification, ?500. (C) No ISH signals are present in
nonproductive HSV-1 infection. (D) Uninfected control epithelium. Magnifica-
tion, ?250.
FIG. 3. PCR detection of HSV DNA in the raft cultures. (A) Ethidium
bromide staining of the specimens after electrophoresis in 2% agarose gel. (B)
Southern hybridization with an internal oligonucleotide probe, 3?-end-labeled
with DIG-dUTP. Detection on X-ray film was carried out by use of luminescent
CSPD reagent. Lanes: 1, distilled water; 2, a cerebrospinal fluid specimen neg-
ative for HSV-1; 3, uninfected HaCat raft culture; 4, productively infected HaCat
raft culture (20 ng of DNA extract); 5, nonproductively infected HaCat raft
culture (20 ng of DNA extract); 6, distilled water; and 7 and 8, 1 PFU of HSV-1
(F). M, molecular size markers. The arrowhead indicates the location of the
positive PCR product in the gel.
6526NOTESJ. VIROL.

Page 4

controls [1 PFU of HSV-1(F)] were constantly positive,
whereas the negative controls (cerebrospinal fluid specimens
negative for HSV and specimens containing only distilled wa-
ter) were all negative. The PCR specimens were total DNA
extracts of the cell cultures containing material from both the
connective tissue and the epithelium. The HSV-1 PCR-positive
but ISH-negative cultures (viral inoculation 30 min after lift-
ing) were further tested for presence of infectious virus in the
samples. Both the culture supernatants and the direct homo-
genates of the cultures were unable to produce any HSV in-
fection in Vero cells. The standard method of verifying HSV
latency in mouse TG by using the homogenates of incubated
tissue was not relevant in this study. The latency was further
analyzed by studying the expression of viral ?0 and LAT RNA
in the cultures with single-stranded, DIG-labeled RNA probes.
The probes were validated with paraffin sections of BALB/c
mouse TG which were either uninfected or represented acute
(4 days postinfection) or latent (4 weeks postinfection) phases
of HSV-1(F) infection (data not shown). The hybridization
protocol favored the detection of RNA instead of viral DNA,
since the tissue samples were not heat denatured before hy-
bridization. In control settings, specimens were subjected to
treatment with RNase or DNase before hybridization. The
uninfected organotypic cultures were regarded as being nega-
tive for both LAT and ?0 RNA expression (Fig. 4A and B).
The cultures with the cytopathic effects of HSV showed strong
ISH signals with both probes (Fig. 4C and D), partially reflect-
ing the reaction of the probes with the high quantities of viral
DNA, since all the reactions were not removable by RNase
pretreatment (data not shown). The cultures infected after
lifting (without any cytopathic signs of HSV infection) reacted
with the probe for LAT RNA, but were clearly negative for ?0
mRNA (Fig. 5A and B). The LAT signal was intracytoplasmic
rather than nuclear, and it disappeared after pretreatment with
RNase but not with DNase (Fig. 5C and D). The LAT signal
could not be localized to any particular layer of the cultured
epithelium. The fibroblasts in the collagen matrix could not be
assessed because of high background reaction, but in the DNA
ISH, the fibroblasts remained definitively negative (Fig. 2).
Northern blot analysis showed that the 2.0-kb LAT RNA spe-
cies is expressed in both lytic and nonproductive forms of
infection (Fig. 6). Both the 1.5-kb and 2.0-kb LAT RNAs were
detected in the mouse TG specimen.
In the past, a variety of in vitro models have been developed
for the study of HSV latency in different cell types (3, 13, 16,
17, 23). However, the restriction of HSV replication by differ-
ent treatments of cultured cells is not similar to that occurring
during establishment of HSV latency in the sensory neurons of
animals and humans (19). To further elucidate the complex
mechanisms of HSV latency, new models capable of better
mimicking the in vivo events are urgently needed. In the
present report, we describe for the first time an in vitro model
to produce a lytic or nonproductive HSV infection in squa-
mous epithelial (HaCat) cells grown in an organotypic raft
culture. In the cultures infected before the total confluency and
lifting of the cells to the air-liquid interface, i.e., before differ-
entiation of the cell layers, a lytic HSV infection could be
FIG. 4. Uninfected organotypic cultures were negative for both LAT and ?0
RNA expression (A and B, respectively). Both LAT (C) and ?0 RNA (D) probes
reacted strongly with cultures infected before lifting at the air-liquid interface, as
indicated by ISH. Magnification, ?250.
FIG. 5. Cultures infected after lifting at the air-liquid interface showed no
cytopathic effect and reacted with the probe for LAT RNA (A) but were clearly
negative for ?0 mRNA (B). The LAT signal was cytoplasmic rather than nuclear,
and it was removed by pretreatment with RNase (C) but not with DNase (D).
Magnification, ?250.
FIG. 6. Northern blot analysis of the raft cultures for LAT RNA. Twenty
micrograms (lanes 1 to 3 and 7) or 15 ?g (lanes 4 to 6) of total RNA was
extracted from uninfected raft cultures (lanes 1 and 4), nonproductively infected
cultures (lanes 2 and 5), productively infected cultures (lanes 3 and 6), or TG of
latently infected BALB/c mice (lane 7) and run in a 1% agarose-formaldehyde
gel, blotted on nylon filter, and hybridized with a32P-labeled single-stranded
RNA probe for LAT. The RNA specimens in lanes 4 to 6 were treated with
DNase prior to electrophoresis. The arrowhead indicates the location of the
2.0-kb LAT RNA species. In the mouse TG specimen the 1.5-kb LAT can also
be detected.
VOL. 70, 1996 NOTES6527

Page 5

established with typical cytopathic changes and the formation
of characteristic intraepithelial vesicles in vitro. Application of
HSV on the lifted epithelium led to a nonproductive infection
characterized by the lack of infectious virus in the culture
supernatant and in the direct homogenate of the culture. PCR
for HSV DNA revealed the presence of HSV DNA in this
culture, and ISH for LAT RNA showed a cytoplasmic pattern
of hybridization. The infection of fibroblasts in the collagen
matrix was unlikely, however, since the ISH for HSV DNA was
negative in the fibroblasts. Similarly, it was apparent that the
positive PCR result could not be due to seed viruses only,
attached on the culture surfaces, since the LAT hybridization
was evident in all epithelial layers of the culture without inten-
sified signals restricted to the external surface. The 2.0-kb LAT
RNA species was detectable by Northern blot analysis in non-
productive and lytic forms of infection (Fig. 6). This inevitably
shows that in the nonproductive infection HSV gene expres-
sion takes place, although in a restricted manner. Further study
of HSV gene expression in this model is in progress. The PCR
tests were validated with respect to contamination by including
numerous HSV DNA-negative control specimens in the runs,
which invariably remained PCR negative. Interestingly, LAT
RNA was detected in the cytoplasm of the epithelial cells,
instead of the nucleus. In the mouse models, LAT RNA is
predominantly localized in the neuronal nucleus (4, 11, 25).
Our present approach may provide an applicable model for
in vitro studies of latent HSV infection once knowledge of the
state of the virus in the raft culture is further established. Thus
far, some of the models for in vitro latency have included
treatments of the host cells with antiviral agents and interferon
(3, 17). The use of a double mutant HSV with deficient ?
trans-inducing factor and ?0 (23) has yielded an interesting
state of infection in fibroblasts. However, in our culture model,
the HSV-1 is the prototype F strain and the form of infection
is modified exclusively by the timing of virus application rela-
tive to the epithelial differentiation in the culture. Further
analysis of the state of the viral DNA and gene expression,
which will finally establish the value of our model for in vitro
HSV latency studies, is in progress.
We are grateful to Bernard Roizman, University of Chicago, for the
HSV clone and for the protocols for producing the stock virus.
This study was supported by grants from the Medical Research
Council of the Academy of Finland.
REFERENCES
1. Asselineau, D., and M. Prunieras. 1984. Reconstruction of simplified skin:
control of fabrication. Br. J. Dermatol. Suppl. 111:219–222.
2. Aurelius, E., B. Johansson, B. Sko ¨ldenberg, Å. Staland, and M. Forsgren.
1991. Rapid diagnosis of herpes simplex encephalitis by nested polymerase
chain reaction assay of cerebrospinal fluid. Lancet 337:189–192.
3. Biswal, N., A. Patel, and S. Max. 1988. Regulation of viral and cellular genes
in a human neuroblastoma cell line latently infected with herpes simplex
virus type 2. Mol. Brain Res. 3:95–106.
4. Dobson, A. T., F. Sederati, G. Devi-Rao, W. M. Flanagan, M. J. Farrell, J. G.
Stevens, E. K. Wagner, and L. T. Feldman. 1989. Identification of the
latency-associated transcript promoter by expression of rabbit beta-globin
mRNA in mouse sensory nerve ganglia latently infected with a recombinant
herpes simplex virus. J. Virol. 63:3844–3851.
5. Fenwick, M. L., and J. Clark. 1982. Early and delayed shut-off of host protein
synthesis in cells infected with herpes simplex virus. J. Gen. Virol. 61:121–
125.
6. Fenwick, M., L. S. Morse, and B. Roizman. 1979. Anatomy of herpes simplex
virus DNA. XI. Apparent clustering of functions effecting rapid inhibition of
host DNA and protein synthesis. J. Virol. 29:825–827.
7. Hardwicke, M. A., and R. Sandri-Goldin. 1994. The herpes simplex virus
regulatory protein ICP27 contributes to the decrease in cellular mRNA
levels during infection. J. Virol. 68:4797–4810.
8. Heino, P., V. Hukkanen, and P. Arstila. 1994. Digoxigenin labeled probes
and their use in the laboratory diagnosis of virus infections, p. 101–112. In E.
Kurstak, R. G. Marusyk, F. A. Murphy, and M. H. V. van Regenmortel (ed.),
Applied virology research, vol. 3. Plenum Medical Book Co., New York.
9. Honess, R. W., and B. Roizman. 1974. Regulation of herpesvirus macromo-
lecular synthesis. I. Cascade regulation of the synthesis of three groups of
viral proteins. J. Virol. 14:8–19.
10. Honess, R. W., and B. Roizman. 1975. Regulation of herpesvirus macromo-
lecular synthesis: sequential transition of polypeptide synthesis requires
functional viral polypeptides. Proc. Natl. Acad. Sci. USA 72:1276–1280.
11. Hukkanen, V., P. Heino, A. Sears, and B. Roizman. 1990. Detection of
herpes simplex virus latency-associated RNA in mouse trigeminal ganglia by
in situ hybridization using nonradioactive digoxigenin-labeled DNA and
RNA probes. Methods Mol. Cell. Biol. 2:70–81.
12. Kwong, A. D., J. A. Kruper, and N. Frenkel. 1988. Herpes simplex virus
virion host shutoff function. J. Virol. 62:912–921.
13. Levine, M., A. Goldin, and J. Glorioso. 1980. Persistence of herpes simplex
virus genes in cells of neuronal origin. J. Virol. 35:203–210.
14. Meyers, C., and L. A. Laimins. 1992. In vitro model systems for the study of
HPV-induced neoplasias. Papillomavirus Rep. 3:1–3.
15. Miller, S. A., D. D. Dykes, and H. F. Polesky. 1988. A simple salting out
procedure for extracting DNA from human nuclear cells. Nucleic Acids Res.
16:1215.
16. Nilheden, E., S. Jeansson, and A. Vahlne. 1985. Herpes simplex virus latency
in a hyperresistant clone of mouse neuroblastoma (C1300) cells. Arch. Virol.
83:319–325.
17. Rapp, F. 1984. Experimental studies of latency in vitro by herpes simplex
viruses. Prog. Med. Virol. 30:29–43.
18. Roizman, B. 1990. Herpesviridae: a brief introduction, p. 1787–1793. In B. N.
Fields, D. M. Knipe, R. M. Chanock, M. S. Hirsch, J. L. Melnick, T. P.
Monath, and B. Roizman (ed.), Virology, 2nd ed. Raven Press, New York.
19. Roizman, B., and A. Sears. 1990. Herpes simplex viruses and their replica-
tion, p. 1795–1841. In B. N. Fields, D. M. Knipe, R. M. Chanock, M. S.
Hirsch, J. L. Melnick, T. P. Monath, and B. Roizman (ed.), Virology, 2nd ed.
Raven Press, New York.
20. Sandberg, M., and E. Vuorio. 1987. Localization of types I, II and III
collagen mRNAs in developing human skeletal tissues by in situ hybridiza-
tion. J. Cell Biol. 104:1077–1084.
21. Sears, A., V. Hukkanen, M. Labow, A. J. Levine, and B. Roizman. 1991.
Expression of the herpes simplex virus 1 ? transinducing factor (VP16) does
not induce reactivation of latent virus or prevent the establishment of latency
in mice. J. Virol. 65:2929–2935.
22. Stevens, J. G., E. K. Wagner, G. B. Devi-Rao, M. L. Cook, and L. Feldman.
1987. RNA complementary to a herpesvirus alpha gene mRNA is predom-
inant in latently infected neurons. Science 235:1056–1059.
23. Stuart-Jamieson, D. R., L. H. Robinson, J. I. Daksis, M. J. Nicholl, and
C. M. Preston. 1995. Quiescent viral genomes in human fibroblasts after
infection with herpes simplex virus type 1 Vmw65 mutants. J. Gen. Virol.
76:1417–1431.
24. Syrja ¨nen, S., P. Partanen, R. Ma ˜ntyja ¨rvi, and K. Syrja ¨nen. 1988. Sensitivity
of in situ hybridization techniques using biotin and35S-labeled human pap-
illomavirus (HPV) DNA probes. J. Virol. Methods 19:225–238.
25. Wagner, E. K., G. B. Devi-Rao, L. T. Feldman, A. T. Dobson, Y. F. Zhang,
W. M. Flanagan, and J. G. Stevens. 1988. Physical characterization of the
herpes simplex virus latency-associated transcript in neurons. J. Virol. 62:
1194–1202.
26. Ziegler, T., M. Waris, M. Rautiainen, and P. Arstila. 1988. Herpes simplex
virus detection by macroscopic reading after overnight incubation and im-
munoperoxidase staining. J. Clin. Microbiol. 26:2013–2017.
6528NOTESJ. VIROL.