Download Retention of herpes simplex virus DNA sequences in the nuclei of

Journal
of General Virology (1997), 78, 867–871. Printed in Great Britain
...............................................................................................................................................................................................................
SHORT COMMUNICATION
Retention of herpes simplex virus DNA sequences in the nuclei
of mouse footpad keratinocytes after recovery from primary
infection
A. Simmons, R. Bowden and B. Slobedman
Infectious Diseases Laboratories, Institute of Medical and Veterinary Science, Frome Road, Adelaide, SA 5000, Australia
Persistence of herpes simplex virus (HSV) DNA in
mouse footpad keratinocytes was studied by nonisotopic in situ hybridization. HSV DNA was retained
in keratinocyte nuclei for more than 2 weeks after
disappearance of infectious virus and viral antigens.
The anatomical location of viral DNA became more
superficial with increasing time post-infection, reflecting the migration of cells from the basal layer of
the epidermis towards the stratum granulosum.
Latency-associated transcripts (LATs) were not detected in footpad cells at any of the times studied. In
contrast, LATs were detected readily in the nuclei of
lumbar ganglionic neurons innervating HSV DNA
positive footpads. It was concluded that, after
termination of productive infection in the skin, HSV
DNA persists transiently in keratinocyte nuclei, in
the absence of abundant latency-associated transcription. An implication of these data is that
detection of HSV DNA in the skin may reflect recent,
but not necessarily current, cutaneous virus replication.
Between recurrent episodes of herpes simplex virus (HSV)
infection, viral DNA is sequestered in a latent, non-replicating
state, in the nuclei of primary sensory neurons (reviewed by
Stevens, 1989). During latency, viral RNA transcription is
restricted to the repeated (diploid) region of the HSV genome
and, despite intensive study, the functions of latency-associated
transcripts (LATs) are currently unknown.
In addition to classic latency in neurons, there is experimental evidence suggesting that HSV can persist outside
the nervous system, notably in the eyes and skin of
experimentally infected animals (Scriba, 1977 ; Hill et al., 1980 ;
Openshaw, 1983 ; Cook et al., 1987 ; Clements & Subak-Sharpe,
1988 ; Claoue! et al., 1990). However, at the present time, the
Author for correspondence : Tony Simmons.
Fax ­61 8 222 3538. e-mail tsimmons!immuno.imvs.sa.gov.au
existence and biological significance of peripheral latency
remain objects of debate, partly because this phenomenon is
poorly characterized in molecular terms. A pertinent question
is, what cell-types could harbour HSV genomes in the skin?
Keratinocytes, which are the predominant epidermal cell-type,
are the main cutaneous site of productive HSV infection but, as
sites of classic latency, are not ideal candidates owing to their
rapid turnover rate. Nonetheless, non-productive HSV infection has been described in three-dimensional keratinocyte
cultures (Syrja$ nen et al., 1996), in which persistence of viral
DNA was restricted to cells infected after induction of terminal
differentiation. It is reasonable, therefore, that HSV DNA
might persist transiently, in a non-replicating state, in epidermal keratinocytes in vivo, following an episode of productive infection.
In the current study, this hypothesis was addressed by
testing mouse footpad cells for the presence, in situ, of HSV
antigens and HSV DNA, at various times after virus inoculation. Viral DNA was retained in keratinocyte nuclei for
approximately 2 weeks after recovery from productive
infection. Initially, cells containing HSV DNA were distributed
throughout all structural layers of the epidermis but, at later
times [24 days post-infection (p.i.)], viral DNA sequences were
confined to the stratum granulosum, the most superficial layer
of nucleated cells. LATs were not detected in HSV antigen
negative–DNA positive footpad cells but, in contrast, in
lumbar ganglia innervating the same footpads, LATs were
detected readily in neuronal nuclei.
Initially, the model system was characterized by testing
footpads for the presence of infectious virus at daily intervals,
from 5 to 10 days p.i. A well characterized, virulent isolate of
HSV type 1, strain SC16 (Hill et al., 1975), was used throughout.
Virus was grown and titrated in Vero cells and stored at
®70 °C until required. Adult female BALB}c mice (Specific
pathogen free facility, Animal Resource Centre, Perth, Western
Australia) were lightly anaesthetized and left hind footpads
were scratched 15 times with a 27 gauge needle, through a
20 µl drop of virus suspension containing 1±0¬10' p.f.u. To
quantify infectious virus in footpads, tissues were homogenized in 1 ml cell culture maintenance medium and tenfold
dilutions were tested for the presence of infectious virus, using
0001-4436 # 1997 SGM
IGH
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:01:14
A. Simmons and others
Fig. 1. Photomicrographs of PLP fixed spine/left hindlimb tissue sections (5 µm), stained with haematoxylin and eosin (H & E),
showing : (a) a heavily infiltrated, acutely infected footpad (arrowed) and lumbar ganglia (e.g. L3 ringed) innervating the site of
IGI
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:01:14
Retention of HSV DNA in mouse footpad cells
a standard plaque assay (Russell, 1962). At 5 and 6 days p.i. the
mean levels of virus recovered from groups of 3 mice were
8¬10' and 1¬10% p.f.u. per footpad, respectively. Infectious
virus was not detected (! 10 p.f.u. per footpad) on or after day
7 and consequently it was concluded that productive infection
is terminated between 6 and 7 days after inoculation of
10' p.f.u. SC16 into the footpads of BALB}c mice.
Immunohistochemistry and in situ hybridization (ISH) were
used to determine the anatomical and cellular locations of viral
antigen and nucleic acids, respectively, at various times p.i.
(Fig. 1 a–i). Groups of 3 mice were killed at daily intervals from
2 to 10 days p.i. (day 0) and further groups were killed on days
24 and 36. To facilitate simultaneous analysis of peripheral and
neural tissues from each animal, a block of tissue containing the
thoraco-lumbar spine and left hindlimb was fixed in paraformaldehyde–lysine–periodate (PLP) for 16–24 h, decalcified for
3–6 weeks in phosphate-buffered EDTA solution (70 g}l,
pH 7±4) and embedded in paraffin wax. Coronal sections (5µm)
were collected onto glutaraldehyde activated 3-aminopropyltriethoxysilane (Sigma) coated slides.
To detect HSV antigens, tissue sections were sequentially
reacted with rabbit anti-HSV infected cells, swine anti-rabbit Ig
and rabbit peroxidase–anti-peroxidase complex (all reagents
from Dakopatts). All reactions were allowed to proceed for
30 min at 37 °C with a 10 min wash in Tris–HCl (pH 7±4)
between steps. Bound antibody was detected with 3,3«diaminobenzidine (0±5 mg}ml containing 0±1 % H O ), and
# #
sections were lightly counterstained with haematoxylin. Viral
antigens were detected in epidermal cells up to and including
6 days p.i. (Fig. 1 d, day 5) and in lumbar dorsal root ganglia
from days 2 to 6 inclusively (not illustrated). In line with
clearance of infectious virus, HSV antigens were not detected
on or after day 7 (Fig. 1 e, day 9). Acute inflammation was a
prominent clinical and histological feature of infected footpads
(Fig. 1 a, arrowed).
To detect HSV nucleic acids in situ, strand-specific riboprobes complementary to the HSV-1 LAT region were
generated from pSLAT-2, which comprises a 1±6 kb PstI–HpaI
subfragment of SC16 BamHI B, cloned into Bluescribe M13expression vector (Stratagene). Transcription from the T7
promoter generated probes complementary to LATs. Nonisotopic (digoxigenin-labelled) ISH was done as described
previously (Arthur et al., 1993) with the following modifications : (i) when required, sections were treated for 1 h at
37 °C with RNase (500 µg}ml) or RNase followed by DNase
(1 U}µl in 40 mM Tris–HCl pH 7±5, 6 mM MgCl ) ; (ii) DNA
#
targets were denatured, after sealing the hybridization mixture
in place (50 µl}section), by heating slides for 5 min on a metal
tray floating in boiling water. In regions of zosteriform spread,
HSV nucleic acid sequences were shown, consistently, to
persist in keratinocyte nuclei (Fig. 1 f, day 9) as well as in areas
adjacent to the site into which virus was scarified (Fig. 1 g, day
9). Probes did not hybridize to uninfected tissue (not
illustrated). pSLAT2 probes generated from the T7 and T3
promoters stained keratinocytes with equal intensity, from
which it was concluded that the HSV nucleic acid detected in
keratinocytes was primarily DNA rather than LATs. This
conclusion was confirmed in tissue removed 9 and 24 days p.i.
by treating sections with nucleases prior to hybridization ;
staining of sections by T7 transcripts of pSLAT2, which are
anti-sense to LATs, was not significantly reduced in intensity
by RNase but staining was obliterated by RNase plus DNase
(not illustrated).
On days 9 and 10, cells containing HSV DNA were
abundant in all layers of the epidermis (Fig. 1 f, day 9). In
contrast, on day 24, HSV DNA was confined, in all mice
examined, to keratinocyte nuclei in the stratum granulosum
(Fig. 1 h), which is the most superficial layer of nucleated cells
(Fig. 1 c) ; HSV DNA was not detected in any mice killed on
day 36. Although LATs could not be detected in footpads 24
days p.i., LATs were detected readily in the same animals, in
lumbar dorsal root ganglia (Fig. 1 i).
The conclusion that HSV persists in the skin in a nonreplicating state for up to 18 days after recovery from acute
infection is based on the observation that digoxigenin-labelled
riboprobes detected HSV DNA in the nuclei of epidermal cells
24 days p.i., whereas viral antigens and infectious virus were
detected only up to 6 days p.i. Only one cell-type, the
keratinocyte, has a distribution and abundance compatible with
the DNA staining pattern observed in the footpads of infected
animals. In particular, on day 24, HSV DNA was confined to
cells of the superficial stratum granulosum, which comprises
exclusively terminally differentiated keratinocytes, from which
it was concluded that HSV DNA persists in keratinocyte
inoculation ; (b) H & E stained uninfected footpad (¬100) demonstrating the overall structure of the skin, in particular the
boundary between the epidermis (E) and dermis (D) (arrowed) ; (c) H & E stained uninfected footpad epidermis (¬400)
illustrating the densely packed nuclei of cells in the basal layer (B) and the flattened nuclei of superficial keratinocytes in the
stratum granulosum (S). Occasional melanocytes (M) can be distinguished morphologically in the basal layer ; (d) footpad skin
(¬100), 5 days p.i., stained for HSV antigens [black areas throughout epidermis ; epidermal–dermal boundary (arrowed)] ;
(e) footpad skin (¬100), 9 days p.i., stained for HSV antigens [not detected ; epidermal–dermal boundary (arrowed)] ;
(f) footpad skin (¬100), from the same mouse as panel (e), 9 days p.i., stained by non-isotopic ISH for HSV DNA [black
areas, corresponding to keratinocyte nuclei throughout epidermis ; epidermal–dermal boundary (arrowed)] ; (g) footpad
epidermis (¬400), 9 days p.i., stained by ISH for HSV DNA, showing a cluster of DNA positive keratinocytes nuclei (open
arrow), adjacent to a scab overlying the healed inoculation site (closed arrow) ; (h) footpad epidermis (¬400), 24 days p.i.,
stained by ISH for HSV DNA [black keratinocyte nuclei in stratum granulosum are HSV DNA positive ; epidermal–dermal
boundary (arrowed)] ; (i) lumbar dorsal root ganglion (L3, ¬400), stained by ISH for HSV LATs [black neuronal nuclei
(arrowhead)]. Sections shown in panels (d)–(i) were lightly counterstained with haematoxylin.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:01:14
IGJ
A. Simmons and others
nuclei. The change in the anatomical distribution of staining
between days 9 and 24, and the lack of staining on day 36,
suggested progression of DNA sequences from the basal
epidermis towards the stratum granulosum over a 2–3 week
period. This most probably reflects the migration of keratinocytes from the basal layer of the epidermis towards the surface
of the stratum granulosum, which occurs over a similar timeframe during the formation of keratinized epithelium (Halprin,
1972). No evidence of virus persistence in melanocytes of the
basal layer was found, despite their longevity and neural crest
origin.
There are profound differences between the persistence of
HSV in footpad keratinocytes, as detailed here, and neuronal
latency. Firstly, detection of HSV DNA in latently infected
neurons by current in situ hybridization protocols has been
problematic, whereas persistence of viral DNA in keratinocyte
nuclei was detected readily. Secondly, major LATs are
abundant in latently infected neurons, but transcripts from the
LAT locus were not detected, by in situ hybridization, in
keratinocytes. However, it is noted that LATs may have been
found in murine corneas in the absence of detectable virus
replication (Abghari et al., 1992). In addition, major LATs have
been detected in non-productively infected keratinocyte raft
cultures by in situ hybridization but, in stark contrast with
latently infected neurons in vivo, keratinocyte LATs were
located in the cytoplasms rather than nuclei of infected cells
(Syrja$ nen et al., 1996). The significance of the latter observation
is unknown.
In recent years there have been significant advances in our
understanding of the natural history of herpes simplex virus,
and prominent among these is the discovery that, after
recovery from primary infection, virus is shed frequently from
peripheral sites, such as the genital tract, in the absence of
symptoms (Koutsky et al., 1990). Asymptomatic virus shedding
is thought to be an important mode of transmission of HSV,
prompting the application of sensitive molecular biological
techniques, such as PCR, to the further study and diagnosis of
HSV infection (Cone et al., 1991). However, while the presence
of infectious virus is an unambiguous marker of recurrence, the
significance of finding HSV DNA sequences in peripheral
tissues is less clear. For instance, an obvious but important
implication of our data is that the presence of HSV DNA
sequences in peripheral tissues does not necessarily reflect the
presence of infectious virus. In addition, progression of DNA
positive keratinocytes towards the stratum granulosum suggests that HSV DNA persists at the periphery for a limited
time (ca. 2 weeks) after an episode of productive infection. If
this were found to be the case in humans, then detection of
viral DNA is a useful indicator of recent, but not necessarily
current, virus shedding. Detailed characterization of the
temporal relationship between virus shedding and the presence
of HSV DNA in human peripheral tissues, for instance in
humans with genital herpes, will clarify this issue.
Previous studies (e.g. Hill et al., 1980 ; Clements & Subak-
IHA
Sharpe, 1988) have reported evidence of potential reactivation
of HSV infection in mouse skin removed months after recovery
from primary infection. Preliminary experiments in our laboratory (not shown) have failed to recover HSV from mouse
footpads, removed as early as 14 days p.i., which still contain
detectable amounts of viral DNA in keratinocyte nuclei.
Consequently, it remains to be shown whether the DNA
sequences that persist transiently in keratinocyte nuclei are a
potential source of reactivatable infection.
The authors thank S. Efstathiou for supplying plasmid pSLAT2, and
P. Speck for valuable advice and discussion. This work was supported by
grants 92}0918 and 95}1098 from the National Health and Medical
Research Council of Australia.
References
Abghari, S. Z., Stulting, R. D. & Petrash, J. M. (1992). Detection of
herpes simplex virus type 1 latency-associated transcripts in corneal cells
of inbred mice by in situ hybridization. Cornea 11, 433–438.
Arthur, J., Efstathiou, S. & Simmons, A. (1993). Intranuclear foci
containing low abundance herpes simplex virus latency-associated
transcripts visualized by non-isotopic in situ hybridization. Journal of
General Virology 74, 1363–1370.
Claoue! , C. M. P., Hodges, T. J., Darville, J. M., Hill, T. J., Blyth, W. A. &
Easty, D. L. (1990). Possible latent infection with herpes simplex virus
in the mouse eye. Journal of General Virology 71, 2385–2390.
Clements, G. B. & Subak-Sharpe, J. H. (1988). Herpes simplex virus
type 2 establishes latency in the mouse footpad. Journal of General
Virology 69, 375–383.
Cook, S. D., Batra, S. K. & Brown, S. M. (1987). Recovery of herpes
simplex virus from the corneas of experimentally infected rabbits. Journal
of General Virology 68, 2013–2017.
Cone, R. W., Hobson, A. C., Palmer, J., Remington, M. & Corey, L.
(1991). Extended duration of herpes simplex virus DNA in genital
lesions detected by the polymerase chain reaction. Journal of Infectious
Diseases 164, 757–760.
Halprin, K. N. (1972). Epidermal ‘ turnover time ’ : a re-examination.
British Journal of Dermatology 86, 14–19.
Hill, T. J., Field, H. J. & Blyth, W. A. (1975). Acute and recurrent
infection with herpes simplex virus in the mouse : a model for studying
latency and recurrent disease. Journal of General Virology 28,
341–353.
Hill, T. J., Harbour, D. A. & Blyth, W. A. (1980). Isolation of herpes
simplex virus from the skin of clinically normal mice during latent
infection. Journal of General Virology 47, 205–207.
Koutsky, L. A., Ashley, R. L., Holmes, K. K., Stevens, C. E., Critchlow,
C. W., Kiviat, N., Lipinski, C. M., lner-Hansen, P. & Corey, L. (1990).
The frequency of unrecognized type 2 herpes simplex virus infection
among women : implications for the control of genital herpes. Sexually
Transmitted Diseases 17, 90–94.
Openshaw, H. (1983). Latency of herpes simplex virus in ocular tissue
of mice. Infection and Immunity 39, 960.
Russell, W. C. (1962). A sensitive and precise plaque assay for herpes
virus. Nature 195, 1028–1029.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:01:14
Retention of HSV DNA in mouse footpad cells
Scriba, M. (1977). Extraneural localisation of herpes simplex virus in
latently infected guinea pigs. Nature 267, 529–531.
Stevens J. G. (1989). Human herpesviruses : a consideration of the latent
state. Microbiological Reviews 53, 318–332.
Syrja$ nen, S., Mikola, H., Nyka$ nen, M. & Hukkanen, V. (1996). In vitro
establishment of lytic and nonproductive infection by herpes simplex
virus type 1 in three-dimensional keratinocyte culture. Journal of Virology
70, 6524–6528.
Received 27 August 1996 ; Accepted 4 December 1996
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:01:14
IHB