| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Institut für Physiologie und Experimentelle Pathophysiologie, and Institut für Klinische und Molekulare Virologie, Friedrich-Alexander-Universität Erlangen-Nürnberg, D-91054 Erlangen, Germany
1Correspondence: Institut für Physiologie und Experimentelle Pathophysiologie, Universitätsstr. 17, D-91054 Erlangen, Germany. E-mail: kress{at}physiologie1.uni-erlangen.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1- and/or
ß1-adrenergic receptors providing a model for the
pathophysiology of the severe pain associated with the acute
reactivation of VZV.—Kress, M., Fickenscher, H. Infection by human
varicella-zoster virus confers norepinephrine sensitivity to sensory
neurons from rat dorsal root ganglia.
Key Words: DRG cells • norepinephrine • postherpetic neuralgia • sensory neurons
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
.
After latency for many decades, the virus can be reactivated and cause
shingles (herpes zoster). This reactivation may result in severe
complications, including encephalitis or myelitis, with high mortality
or long-term morbidity. Zoster morbidity increases with age;
particularly in the elderly, severe zoster pain and postherpetic
neuralgia are frequent. Approximately 300,000 cases of shingles occur
annually in the U.S. in immunocompetent persons (2)
. The
incidence of shingles is even higher in immunodeficient patients in
whom complications are more severe (3)
. It is still
controversial as to what extent immunization with the VZV vaccine
strain OKA (reviewed in refs 4
, 5
) influences the rate of
herpes zoster and postherpetic neuralgia decades after the vaccination.
The potential of the vaccine to cause zoster and neuralgia is not well
characterized due to lack of long-term experience. Moreover, it is
unclear whether vaccine-induced immunity will be sufficient to prevent
infection with pathogenic wild-type isolates many years after
vaccination.
Aiming at a better understanding of zoster pain, we focused on
virus-induced changes in the nervous system. The persisting virus is
located in neurons but probably not in perineuronal satellite cells in
sensory ganglia (1
, 6
, 7)
. VZV replication in culture is
highly cell-associated and slow in comparison to herpes simplex virus
(HSV). The DNA sequence of a VZV laboratory strain carries 69 open
reading frames, most of which have homologous genes in HSV
(8)
. Functional changes in sensory neurons after HSV
infection have been described. HSV affected all classes of afferent
fibers and DRG neurons (9)
. After HSV infection, decreased
excitability and reduced action potential firing were observed that
were antagonized by acyclovir (10)
. Comparable data are
not available for VZV since the virus has a pronounced species
specificity. An in vitro model for latent VZV infection of
rat DRG neurons has been described (11)
, but the available
in vivo VZV infection models in rats and guinea pigs so far
have not been appropriate for studying acute VZV-induced changes in
neuron function (12
13
14
15
16)
. Only recently, however, a rat
behavioral model for VZV-induced allodynia and hyperalgesia was
developed in the rat that seems to closely resemble symptoms of severe
pain observed in humans (17)
.
To study functional changes in VZV-infected neurons underlying zoster pain, we introduced novel culture conditions allowing the efficient infection of rat DRG neurons. The cells show signs of lytic virus replication and express the viral immediate-early proteins IE62, IE63 and the late glycoprotein gE. In individual neurons, microfluorometric measurements allowed us to describe a gain-of-function after VZV infection with a de novo sensitivity to the pain-associated neurotransmitter norepinephrine. In contrast to a series of wild-type VZV isolates, the vaccine strain OKA yielded a strongly reduced norepinephrine sensitivity.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Briefly, lumbar DRG (thoracic segment T13 to lumbar
L5) were harvested from adult female Wistar rats (100 to 160 g)
from an inbred colony. After dissection, the ganglia were transferred
into Dulbecco’s modified Eagle medium (DMEM, Gibco, Karlsruhe,
Germany) supplemented with 50 µg/ml gentamicin (Sigma, Deisenhofen,
Germany). The connective tissue was removed and the ganglia were
treated with collagenase (0.28 U/ml in DMEM, 75 min; Roche
Biochemicals, Mannheim, Germany) and, after two washes, with trypsin
(25,000 U/ml in DMEM, 12 min; Sigma). After dissociation with a
fire-polished Pasteur pipette, the cells were sedimented and finally
resuspended in culture medium. After plating on glass coverslips coated
with poly-L-lysine (200 µg/ml, Sigma), the cells were
cultivated in serum-free TNB 100 medium (Biochrom, Berlin, Germany)
supplemented with penicillin/streptomycin (each 200 U/ml; Gibco),
L-glutamine (2 mM; Gibco), and nerve growth factor (mouse NGF 7S, 100
ng/ml; Alomone Labs, Tel Aviv, Israel) at 37°C in a humidified
atmosphere containing 5% CO2.
Virus cultures and infection of neurons by cocultivation
Primary human foreskin fibroblasts (passage 9 to 20) were
cultivated in the presence of 5% CO2 in DMEM
supplemented with heat-inactivated fetal bovine serum (10%),
L-glutamine (2 mM), and gentamicin (100 µg/ml, all from Gibco). The
fibroblasts were split by ratio 1:2 once a week. A series of VZV
strains was used: NIK, a clinical isolate shown before to latently
infect rats in vivo (kindly provided by Catherine
Sadzot-Delvaux, Liège, Belgium; ref 12
); HJO, a
clinical isolate from a zoster lesion of the ear; PJ, a clinical
isolate from a thoracic zoster lesion; vaccine strain OKA, derived from
the commercial vaccine material (Varilrix, SmithKline Beecham, Munich,
Germany); and parental strain OKA, kindly provided at low passage
number by Paul Kinchington (Pittsburgh, Pa.). The various virus strains
were passaged once a week on fresh fibroblasts
(0.5x105 cells on 25 cm2)
that had been trypsinized and split within 1 h before infection.
For this purpose, infected cultures with 30–50% cytopathogenic effect
were trypsinized and transferred to the fresh fibroblasts at the split
ratios 1:100, 1:50, 1:25, and 1:12.5. After 1 wk, either the 1:100 or
the 1:50 culture was used for the next passage.
For the infection of rat DRG neurons by coculture, infected cultures
with 30–50% cytopathogenic effect (typically the 1:12.5 or 1:25 split
cultures) were trypsinized on day 4 after splitting. The infected cells
(
0.5x105) were centrifuged, washed, and
resuspended in 2 ml medium. Freshly prepared rat DRG cells were plated
as a droplet (100 µl, 300 neuronal cells per culture, in
supplemented TNB 100) onto poly-L-lysine coated round
coverslips (in 24-well plates) or petri dishes (35 mm diameter with
glass bottom). Infected fibroblasts were added to the droplet in a
volume of 30 µl (7500 cells per culture). The microcultures were
kept in 130 µl for 4 h in humidified atmosphere at 37°C.
Supplemented serum-free TNB 100 medium was then added. After 48 h
cocultivation, fibroblasts were no longer detectable in the cultures.
Immunocytochemistry
After 2 days in culture, the cells were fixed for 15 min with
Zamboni’s fixative (150 ml saturated picric acid, 20 g
paraformaldehyde, and 850 ml phosphate buffer pH 7.4; ref
19
). The coverslips were stored at -20°C. For
immunocytochemical staining, the coverslips were warmed to room
temperature in phosphate-buffered saline (PBS) for 10 min. Some
coverslips were incubated with a directly labeled VZV-specific
monoclonal antibody for 30 min to detect viral antigens (nonspecified
VZV-specific fluorescein isothiocyanate-conjugated monoclonal
antibody; BioWhittaker, Walkersville, Md.). Indirect immunofluorescence
was performed to detect the viral proteins IE62, IE63, and gE. The
monoclonal 8616 was used for demonstrating IE62 and the monoclonal 8612
was applied for gE (both from Chemicon, Temecula, Calif.). A polyvalent
rabbit antiserum was used to demonstrate IE63 (kindly provided by
Catherine Sadzot-Delvaux, Liège, Belgium; ref 12
).
In the first step, the respective antibody was applied in the presence
of 10% fetal bovine serum, 0.5% Triton X-100, 1% normal goat serum,
and human immunoglobulins (Cohn’s fraction II, 2 mg/ml; Sigma) in PBS
for 24 h at 4°C. After three washes in saline containing 1%
normal goat serum, appropriate secondary antibodies coupled to
fluorescein isothiocyanate (1:50, goat anti-mouse) or to the red
fluorescent Cy3 (1:330; goat anti-rabbit, both from Dianova, Hamburg,
Germany) were applied in the presence of 1% normal goat serum and
human immunoglobulins (Cohn’s fraction II, 2 mg/ml) in PBS for 30 min
at room temperature. After three washes in PBS with 1% normal goat
serum and 0.5% Triton X-100, the coverslips were mounted on glass
slides with glycerol jelly (Merck, Darmstadt, Germany) and analyzed
with a fluorescence microscope (Leica, Heidelberg, Germany) equipped
with a video camera, frame grabber, and a software program for image
analysis, which was developed in-house and modified for this purpose.
Calcium measurements in isolated neurons from dorsal root ganglia
The external solution contained 140 mM NaCl, 5 mM KCl, 2 mM
CaCl2, 1 mM MgCl2, 10 mM
glucose, and 10 mM HEPES buffer pH 7.3. For stimulation, 1 µM
capsaicin (freshly diluted from a 3 mM stock solution in 98% ethanol)
and gamma amino butyric acid (GABA, 30 µM freshly diluted from a 30
mM stock solution in distilled water) were added to normal external
solution. Norepinephrine, phenylephrine, and isoproterenol were freshly
dissolved in external solution and used in final concentrations of 100
µM. For calcium measurements, 3 µM FURA-2/AM (Molecular Probes,
Leiden, The Netherlands) was added to the extracellular solution for 30
min. After 30 min of washout in normal extracellular solution, the
cultures were transferred to the recording microscope where only a
single cell was analyzed per culture. Background-corrected fluorescence
images were taken with a slow scan CCD camera system with fast
monochromator (PTI, New Jersey) coupled to an Axiovert microscope with
an x40 fluotar oil immersion objective (Zeiss, Oberkochen,
Germany). Fura-2 was excited at 340 and 380 nm wavelengths (
).
The fluorescence was collected at > 420 nm at a frequency of
1 Hz with equal exposure time of 200 ms.
[Ca2+]i was calculated as
previously published (20
, 21)
and the calibration
constants obtained in vitro were Rmin
= 0.44, Rmax = 8.0, and
Keff = 1.2 µM. For chemical stimulation, a fast
10-channel system with common outlet was used for drug application
(22)
. The magnetic valves were controlled manually from a
switchboard and the time constants for a full exchange of solution
were 150 ms.
Data analysis
For detailed statistical analysis, the CSS software package was
used (StatSoft, Tulsa, Okla.). All summarizing results are given as
means ± SE. For intraindividual data comparisons, the
Wilcoxon matched pairs test was calculated, if not stated otherwise,
and differences were considered significant at P <
0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
With the indirect immunofluorescence procedure using the anti-IE62
monoclonal antibody, 35% of the neurons infected with NIK were
stained (Fig. 1
; see Fig. 2
). All these cells were colabeled with a rabbit antiserum
against IE63, which stained a higher proportion of the neurons
(two-thirds of neurons, Fig. 1
). Similar percentages of labeled
neurons were observed in cultures infected with the wild-type isolate
HJO. There was no difference in staining rates or intensities between
different wild-type isolates and the vaccine strain OKA (Fig. 2
). The monoclonal antibody against the virus-specific glycoprotein gE
also stained approximately one-third of the neurons. The percentage of
immunolabeled neurons suggests that a minimum of one-third of the
neurons in the cocultures were infected by VZV after 2 days of
cocultivation.
|
|
Functional changes of DRG neurons after VZV cocultivation
VZV-infected neuron cultures were functionally tested in
comparison to noninfected cells. Only a single cell per culture dish
was analyzed. At 3 days after infection or with higher concentrations
of virus-infected fibroblasts, most neuron-like structures had lost all
sensitivity to the physiological stimuli used (GABA, capsaicin) and
therefore were considered nonfunctional or dead. Two days after
infection, many functions were unchanged in the infected neuron
cultures. Voltage-dependent currents were recorded in five neurons from
infected cultures using the whole-cell voltage-clamp configuration of
the patch-clamp technique (21
, 23)
. No pronounced
alteration in inward or outward currents was observed. Similarly,
neuronal responsiveness to GABA or capsaicin was unaltered in infected
cultures as compared to controls. In contrast, the baseline calcium
concentration which was 114 ± 6 nM in control neurons
(n=38) was significantly increased in neurons from cultures
infected with wild-type isolates or with the vaccine strain (136±5 nM
[NIK, n=77], 165±1 nM [HJO, n=48], 145±14
nM [PJ, n=11], 152±7 nM [OKA vaccine strain,
n=60], and 141±8 nM [OKA, parental strain,
n=31], all P<0.05).
As expected, most control neurons were insensitive to norepinephrine
application and only 1 of 38 control neurons responded to adrenergic
stimulation, with a peak rise in
[Ca2+]i of 54 nM. In
contrast, in cultures infected with wild-type isolates (NIK, HJO, PJ,
and parental OKA), 50% of small-sized capsaicin-sensitive neurons
were also sensitive to norepinephrine and this percentage was higher
than in controls or OKA vaccine strain-infected cultures (1 of 38
neurons and 17%,;F3> Fig. 3a
and Table 1
). The magnitude of the calcium increase was significantly higher
(ranging from 97±26 nM to 148±50 nM for wild-type isolates) than in
controls (1.5±1.5 nM) or in the OKA vaccine strain (15±5 nM, all
P
0.05, Fig. 3b
). Cultures infected with the
parental OKA strain showed intermediate sensitivity: although 50%
of neurons responded to norepinephrine, the average magnitude of the
responses was only 65 ± 16 nM, and this was not significantly
different from the OKA vaccine strain. Norepinephrine sensitivity was
predominantly induced in capsaicin-sensitive neurons. Only
exceptionally capsaicin-insensitive but GABA-sensitive neurons
responded to norepinephrine. In five neurons, repetitive stimulation
with norepinephrine was performed. Figure 4
demonstrates that the responses were prone to pronounced tachyphylaxis.
The third application yielded only about one-third of the response
magnitude determined for the first stimulation.
|
|
|
To prove that the norepinephrine sensitivity was induced in infected
neurons, five norepinephrine-sensitive neurons were incubated with a
fluorescent Cy3-coupled monoclonal antibody against the virus-specific
surface protein gE in the recording chamber for direct immunostaining
without fixation. However, none of the tested neurons carried
detectable amounts of gE on their surface. Under these conditions,
mainly dead neurons were labeled by surface staining with Cy3-anti-gE.
Therefore, in another series of norepinephrine sensitive neurons,
indirect immunocytochemistry for IE63 was performed after fixation in
the recording chamber. Three of the five neurons that were sensitive to
norepinephrine could be stained with the antibody against IE63, whereas
none of the five insensitive neurons was stained (Fig. 5
).
|
Adrenergic receptor subtypes involved in VZV-related norepinephrine
sensitivity
To determine which adrenergic receptor subtypes were functionally
expressed in the infected neuron cultures, the
a1 receptor agonist phenylephrine and/or the
ßb1 receptor agonist isoproterenol were tested,
since these receptor subtypes are known to be coupled to second
messenger pathways that cause rises in
[Ca2+]i on activation.
Norepinephrine-sensitive neurons (17 of 39) responded to phenylephrine
with an increase in
[Ca2+]i of 152 ± 26
nM and 12 of 33 exhibited a rise in
[Ca2+]i of 117 ± 30
nM in response to isoproterenol. Of 33 neurons tested with both
substances, five were sensitive to both agonists. This suggests that
42% (14 of 33) of the neurons from infected cultures express
functional 1 receptors, 21% express
functional ßb1 receptors, and 15% express
both receptors, whereas 36% were insensitive to both agonists (12
of 33). This lack of sensitivity may be explained by the tachyphylaxis
of the response, since the agonists were always applied after a
preceding norepinephrine stimulus.
To determine the source of the calcium increase, experiments were
performed in calcium-free extracellular solution. In most cells, the
calcium responses were still preserved in the absence of extracellular
calcium (Fig. 6
). This independence from extracellular calcium ions suggests that
calcium is released from intracellular stores, which is in line with
the known signal transduction pathways for the
1 receptor subtype and, according to recent
findings, for the ß1 receptor subtype also.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
50% of the capsaicin-sensitive neurons
infected with wild-type VZV isolates expressed functional
1- and/or
ß1-adrenergic receptors, leading to an increase
in [Ca2+]i on adrenergic
stimulation. Such changes were significantly less pronounced in
cultures infected with the OKA vaccine strain, suggesting that the
up-regulation of adrenergic receptor expression and the development of
zoster or postherpetic neuralgia could be substantially reduced in
patients who experience VZV reactivation and zoster symptoms caused by
the vaccine virus.
In our system of neuronal VZV infection, up to 50% of the neurons
expressed virus-specific proteins after 2 days of cocultivation. At
this time, neuronal cell death was not detectable and the infected
cells appeared fully viable, because they reacted appropriately to
various stimuli. The neuron cultures seemed to be permissive to the
virus: gE was expressed on the surface of a fraction of infected cells
and in a high percentage of cells after fixation and permeabilization.
Moreover, after 3 days or at higher virus concentrations, only a few
neurons survived. These neurons presented in an unhealthy condition and
the majority of the neuron-like structures responded to none of the
routinely used stimuli (i.e., they were insensitive to GABA,
norepinephrine, or capsaicin). This rapid loss of functional cells
together with the expression of virus antigens argues for virus-induced
lytic cell death, which could possibly be augmented by virus-induced
apoptosis (24)
. Therefore, the presented infection system
cannot be applied to study virus latency, but seems appropriate to
investigate effects from acute and productive infection. The observed
changes seem to be similar to the situation in the patient, where
reactivation and virus replication cause severe pain.
Infection with HSV, another -herpes virus, does not induce
comparable long-lasting forms of neuralgia or severe pain, but rather
is characterized by prodromal and lesion-related episodes of pain and
itching. In a cellular model, HSV infection has been shown to reduce
the excitability of sensory neurons and to cause a loss of
tetrodotoxin-sensitive action potentials developing during 5 to 15 h postinfection (10)
. In contrast to these findings, we
demonstrate that the infection with wild-type VZV strains causes a
gain-of-function and confers a de novo sensitivity to
adrenergic agonists to DRG neurons. Under normal conditions, this cell
type does not respond to norepinephrine. However, adrenergic effects
are characteristic of other neuropathic changes in nociceptive neurons:
adrenergic effects have also been reported in other models of
neuropathic pain, such as nerve ligation or axotomy
(25
26
27
28
29)
. In patients with postherpetic neuralgia, the
injection of adrenergic agonists induced pain and allodynia. Even when
postherpetic neuralgia had been present for years, injection of
adrenergic agonists into the skin increased the pain substantially,
most likely through direct activation of C-nociceptors
(30)
. In addition, treatment with sympatholytic drugs or
sympathectomy improved the severe pain in many cases (31)
.
These findings are in line with the present data showing that
VZV-infected DRG neurons in culture become sensitive to adrenergic
stimulation. Therefore, we hypothesized that an up-regulation of
adrenergic receptors could be the functional correlate for zoster pain.
The culture system was also used to investigate which subtypes of
adrenergic receptors are up-regulated in VZV-infected neurons. In the
past, the expression of receptor subtypes on DRG neurons had been
controversially discussed. In many studies, 2
receptors have been suggested to contribute to neuropathy
(28)
. However, they are coupled to second messenger
pathways, which decrease intracellular calcium levels. In the present
study, results with pharmacological agonists suggest that
1- and, to a lesser extent,
ß1 receptors are involved in the
norepinephrine-induced calcium release. These two receptor subtypes are
coupled to signal cascades, which are able to increase the
intracellular calcium concentration (32
, 33)
. The
preserved increase in
[Ca2+]i under
calcium-free conditions further supports an 1
receptor-mediated mechanism; this is also favored by the pronounced
tachyphylaxis of the responses, which may be due to depletion of
intracellular calcium stores. Pain and hyperalgesia are consistent with
rises in [Ca2+]i, and an
increase in intracellular calcium concentration plays a role especially
in heat sensitization of nociceptors and heat hyperalgesia
(23)
. Such heat hyperalgesia is one of the symptoms of
zoster/postherpetic neuralgia and may be explained by the present
findings (30)
.
An attenuated vaccine strain OKA was developed in order to prevent
primary infection leading to chicken pox (reviewed in ref
5
). In the present study, the OKA vaccine strain infected
the neuron cultures in a similar way as the wild-type isolates. There
was no obvious difference in the number of infected neurons between the
vaccine strain and wild-type isolates. Similar to other infectious
agents (34
, 35)
, VZV infection weakly mobilized
intracellular calcium in the cultivated neurons, irrespectively of the
virus isolate or strain tested. This is consistent with previous
reports revealing biological differences between vaccine OKA and other
VZV strains only in special contexts, such as defective accumulation of
gC or reduced skin tropism in SCID-hu mice (16
, 36)
.
However, healthy children who have received the OKA vaccine strain
develop immunity, usually without signs of disease and with a low rate
of minor side effects (4
, 5)
. The OKA vaccine strain
seemed to decrease the incidence of zoster and associated neuralgia
(5
, 37)
. Our data further argue in favor of the safety of
this live vaccine. The degree of sensitivity to adrenergic stimulation
was much higher in wild-type VZV isolates than in the OKA vaccine
strain. The lower number of norepinephrine-sensitive cells and the
smaller response magnitudes observed in cultures infected with the OKA
vaccine strain could presumably generate a lower incidence or intensity
of neuropathic pain upon reactivation of the vaccine virus.
Although the DNA sequence of a VZV laboratory strain is available
(8)
, the genome structures of the OKA vaccine or parental
strains or of wild-type isolates are largely unknown. Presently, there
is no hint for a genetic correlate of the viral phenotype causing
norepinephrine sensitivity after infection of sensory neurons. The
presented infection system supplies a unique tool to study mutant
viruses in order to reveal the genetic basis leading to the
up-regulation of adrenoreceptors. Such mechanisms may also have
implications for other types of neuropathic pain that involve
adrenoreceptors in nociceptive neurons.
|
ACKNOWLEDGMENTS |
|---|
Received for publication July 21, 2000.
Revision received November 7, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. Hermann, T. Schlereth, T. Vogt, and F. Birklein Clonidine induces nitric oxide- and prostaglandin-mediated vasodilation in healthy human skin J Appl Physiol, December 1, 2005; 99(6): 2266 - 2270. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Zahn, S. Leis, C. Schick, M. Schmelz, and F. Birklein No {alpha}-adrenoreceptor-induced C-fiber activation in healthy human skin J Appl Physiol, April 1, 2004; 96(4): 1380 - 1384. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
