Abstract
purpose. To study the effect of aminoguanidine (AMG), an inhibitor of nitric
oxide production, on the ocular infection of Balb/c mice with herpes
simplex virus (HSV) type 1 strain F and HSV-2 strain G.
methods. Animals were treated with different amounts of AMG (0.5, 0.1, and 0.05
mg/mouse) by topical application in the eye from postinfection (PI)
days −2 through +5, considering 0 the day of infection. At different
PI days, development of herpetic keratitis was evaluated in treated and
control mice.
results. Treated animals showed a dose-dependent increase in ocular disease
after viral infection, compared with control animals. Viral titers in
ocular washings were higher in AMG-treated mice (PI day 2, HSV-1: AMG
0.5 mg, 1.3 × 103 plaque-forming units (PFU)/ml;
control, 0. 22 × 102 PFU/ml, P <
0.025). At PI day 3, control corneas had only scattered inflammatory
cells, whereas those from treated animals showed a conspicuous
infiltrate consisting primarily of neutrophils. Viral titers were also
higher in brains of treated mice. These animals died earlier and in a
greater proportion than control animals (percentage of mortality, PI
day 12, HSV-1: AMG 0.5 mg, 40% ± 4%; control, 18% ± 3%, P < 0.05).
conclusions. These data indicate an inhibitory effect of nitric oxide on HSV ocular
infection.
Herpes simplex virus (HSV) is a pathogen that infects the
mucosal surfaces of the eye, mouth, and genitalia, causing ulcerative
lesions. In a primary infection, the virus quickly replicates in
peripheral tissues, enters nerve endings, and travels to sensory nerve
ganglia, where it remains in a latent state. HSV transcription during
latency is confined to the repeat regions of the viral genome. Once
latent in the ganglia, the virus appears to avoid detection by the
cells of the immune system.
1 2 The timing of appearance
and incidence of isolation of virus from the trigeminal ganglia
suggests that is the most likely source of reactivated virus to produce
recurrent ocular disease and shedding in the tear film. The most
probable route of spread of virus reactivated in the trigeminal neurons
is within the axons of ocular nerves.
3 Under conditions
such as stress or exposure to UV light, the virus may be reactivated
and cause recurrent lesions at mucosal surfaces or on the skin.
Such recurrence is particularly harmful in the case of ocular
infection, in which damage to the eye can lead to blindness. In some
cases, the virus may spread through the nerves to surrounding areas of
dermatome, resulting in zosteriform lesions, or, more rarely, to the
central nervous system, leading to encephalitis and death. Topical
corneal infection with HSV results in a chronic inflammatory response
of the corneal stroma called herpetic stromal keratitis (HSK) mediated
by CD4+ T-helper 1 (Th1) cells. Replicating virus
as well as viral mRNA is absent from the cornea at the time when
lesions are clinically apparent, but it has been demonstrated that an
initial period of viral replication is necessary to induce HSK.
The chronic inflammatory response that typifies HSK occurs at a
location where and time when viral antigens cannot be demonstrated.
Such observations could indicate that the T cells present in the cornea
may be activated to orchestrate the inflammatory event, not by reacting
with viral peptides, but perhaps in response to host-derived
determinants unmasked as a result of the virus infection.
4 In this way, Zhao et al.
5 identify one potential host
antigen that shares some homology between sequences in the
C
H3 region of IgG2a
b (amino
acids 292-308) and the HSV-1 encoded UL-6 peptide that contains
identical or similar amino acids at seven of eight sequential positions
that contribute to T-cell recognition. HSK could be induced in animals
genetically incapable of generating HSV antigen-specific
CD4
+ T cells, indicating that clinical HSK may
not require viral antigen recognition by CD4
+ T
cells and that T cells of irrelevant specificity can be recruited,
activated, and driven into effector function in the HSV-infected
cornea.
This is suggested to represent a bystander activation effect caused by
the presence of proinflammatory mediators that result from HSV
replication.
6 Histologically, HSK is characterized by
neovascularization, corneal swelling, and an intense infiltrate that is
pronounced in the stroma by 2 weeks after infection. Neutrophils are
the most prominent cell type seen, although monocytes and lymphocytes
are also present. Recruitment of cells from the blood to the site of
tissue injury is thought to depend on the synthesis of specific sets of
inflammatory mediators and the establishment of chemoattractant
gradients in the endothelium and in the extracellular
matrix.
7 The mechanisms involved in innate immunity may be
crucial in controlling primary infection with HSV in ocular mucosa thus
diminishing viral replication and consequent HSK.
Nitric oxide (NO) is a free radical gaseous molecule that is a mediator
of vital physiological functions including host
defense.
8 9 10 Many cell types are able to produce NO
through the enzymatic conversion of
l-arginine to
l-citrulline by nitric oxide synthase (NOS). Macrophages
are the best characterized sources of an inducible form of NOS (iNOS)
that is activated in response to microbial infections. Throughout the
past decade, NO has been shown to play an important role as a
first-line defense against various pathogens. More recently, different
investigators have reported antiviral activity of NO has against
several viruses such as HSV-1, ectromelia virus, vaccinia virus,
encephalomyocarditis virus, vesicular stomatitis virus, and Japanese
encephalitis virus.
11 12 13 14 15 16 17 18 19 20 21 22 23 24 The antiviral activity of NO, at
least in vitro, involves a blockade at the stage of DNA replication but
has no effect on early protein synthesis.
4
In this work we studied the effect of aminoguanidine (AMG), an iNOS
inhibitor of ocular infection with HSV-1 and -2 in Balb/c mice.
The iNOS inhibitor, aminoguanidine hemisulfate (Sigma, St.
Louis, MO), was directly dissolved in sterile phosphate-buffered saline
(PBS) at the indicated concentrations for each experiment.
Oligonucleotides primers used in this study were synthesized by Gibco
(Grand Island, NY).
Male Balb/c mice, 4 to 10 weeks old, were used for all
experiments. Mice were housed five per cage with sterile wood-chip
bedding and were provided with chow pellets and tap water ad libitum.
The animals quarters were maintained at 21°C to 24°C, and 40% to
60% humidity with a 12-hour light–dark cycle. All experimental
procedures conformed to the ARVO Statement for the Use of Animals in
Ophthalmic and Vision Research.
HSV-1 strain F and HSV-2 strain G were maintained at a low
passage in our laboratory. Pools of virus stocks were prepared in Vero
cells and stored at −70°C. Plaque-forming unit (PFU) assays were
performed on Vero cells grown in 16-mm tissue culture plates containing
24 wells. Virus dilutions (0.1 ml/well) were allowed to adsorb for 1
hour at 37°C, and then each well was overlaid with 1 ml of minimum
essential medium (MEM; Gibco) containing 1% methylcellulose (4000 cP).
After 3 days of incubation at 37°C in 5% CO2,
the plates were fixed with 10% aqueous formaldehyde and stained with
0.5% crystal violet, and the virus plaques were counted. Viral stocks
were adjusted at 107 PFU/ml.
Animals were anesthetized by intraperitoneal injection of 2 mg
ketamine hydrochloride (Ketalar; Parke Davis, Morris Plains, NJ) and
0.04 mg xylazine (Rompun; Mosby, Inc., St. Louis, MO) in 0.1 ml
of PBS. They were inoculated by scarification of the left cornea with a
26-gauge needle through a 5-μl drop of medium containing
105 PFU of HSV. Control mice were inoculated
similarly with a preparation of uninfected Vero cells prepared in the
same way as the viral inoculum (mock inoculum). Animals with eyes
accidentally perforated at the time of corneal inoculation were not
included in clinical or histopathologic studies.
Mice were separated in four experimental groups. Between days−
2 through +5 each mouse was anesthetized as previously described and
topically treated three times per day with AMG: 0.5 mg (group 1), 0.1
mg (group 2), 0.05 mg (group 3), and PBS (control). Mice in an
additional control group received 0.5 mg AMG but were mock infected.
All dilutions were prepared in PBS.
Mice were clinically evaluated daily for 2 weeks after corneal
inoculation and examined using a slit lamp biomicroscope. Cornea, iris,
and lids of animals were examined for signs of disease.
4 Criteria for keratitis included stromal opacity due to edema and
cellular infiltration, corneal neovascularization, and corneal
ulceration. Edema was considered positive when stromal opacity was
severe (iris not visible). Corneal neovascularization was considered
positive when eyes showed unequivocal signs of hyperemia. These signs
did not appear simultaneously but developed gradually after infection.
Animals showing at least two of these signs were considered positive
for illness. Infected animals also show blepharitis, but this symptom
was not considered for determining whether an animal was positive for
keratitis. Clinical evaluations were performed in a masked fashion. The
experiments were repeated three times to test the reproducibility of
the results.
Representative eyes were removed and placed in 10% buffered
neutral formalin, embedded in paraffin, stained with hematoxylin and
eosin, mounted (Permount; Fisher Scientific, Fairlawn, NJ), and covered
with a coverslip for microscopic examination. Five representative
sections of each eye were examined to evaluate histopathology and
inflammatory infiltrate characteristics.
To evaluate viral replication in ocular mucosa, eyes were washed
with 20 μl of medium at 12, 24, 48, and 72 hours after infection, and
virus was titrated on Vero cells. For determining virus in eye tissues,
groups of five mice were killed on postinfection (PI) days 1 through 7,
and their left eyes removed. Each eye was ground in 0.5 ml of medium
and then frozen and thawed three times to disrupt the cells. The
resultant cell-free suspensions were titrated on Vero cells by the PFU
assay.
Groups of five mice were killed on PI days 3 through 15 to obtain
tissue samples of brains. Tissues were excised and collected into 1 ml
MEM, minced with scissors, and homogenized using an electric blender.
After sonication in an ice-cold water bath for 1 minute and
centrifugation at 3000 rpm for 10 minutes, virus in supernatants was
evaluated by PFU assay.
Groups of five mice were killed on PI days 1 through 7,
trigeminal ganglia were dissected, rinsed in PBS, and blotted on tissue
paper to remove traces of blood. Individual trigeminal ganglia were
homogenized, and the cell pellet was used to detect viral DNA by
polymerase chain reaction (PCR) analysis. The pellet was washed twice
with PBS, and the trigeminal ganglia pellets were suspended in 100 μl
of Tris-EDTA containing 0.1% sodium dodecyl sulfide (SDS) and 100 μg
proteinase K per milliliter. The mixture was incubated at 55°C
for 16 hours. The DNA was extracted and PCR performed using
two primers (sense: 5′-TTTCTCCAGTGCTAGCTGAAGG-3′, antisense:
5′-TCAACTCGCAGACACGACTCG-3′) that generated a 283-bp product
corresponding to the
ICP27 viral gene.
25 Trigeminal ganglia DNA (3.5 μl) was combined with 1×
Taq buffer, 0.25-μM concentrations of each PCR primer, 100-μM
concentrations of each deoxyribonucleoside triphosphate, and 2.5 U of
Taq polymerase (Promega, Madison, WI) in a 50-μl reaction
volume and overlaid with mineral oil. PCR was performed in a
thermocycler (Eppendorf, Freemont, CA). The first cycle of PCR was at
95°C for 5 minutes, 52°C for 1 minute, and 72°C for 1 minute,
followed by 35 cycles of 94°C for 1 minute, 52°C for 1 minute, and
72°C for 1 minute. PCR products were resolved in 1.5% agarose gels
and were visualized by ethidium bromide staining.
To evaluate iNOS expression, groups of five mice were killed on
PI days 0, 2, 4, and 6; left corneas were collected; and RNA was
extracted with a commercial system (RNAgents Total Isolation System;
Promega) according to the manufacturer’s instructions. First-strand
cDNA was synthesized from 4 μg of total corneal RNA using an
oligo-dT15 primer and Moloney murine leukemia virus (MMLV) reverse
transcriptase (RT; Promega) in a 40-μl reaction volume. PCR on iNOS
cDNA was performed using two primers (sense: 5′-CTT CCG AAG TTT CTG GCA
GCA GCG-3′, antisense: 5′-GAG CCT GCT GGC TTT GGG CTC CTC-3′) that
generated a 487-bp product corresponding to the
iNOS gene.
26 Expression of β-actin mRNA in the
corneal tissue sample was used as a control for RNA isolation. The
first cycle of PCR was at 95°C for 5 minutes, 52°C for 1 minute,
and 72°C for 1 minute followed by 45 (iNOS) or 32 (β-actin) cycles
of 94°C for 1 minute, 52°C for 1 minute, and 72C for 1 minute. PCR
products were resolved in 1.5% agarose gels and were visualized by
ethidium bromide staining.
Statistical analyses were performed by computer using analysis
of variance (ANOVA; Statgraphics Plus for Windows, ver. 3.0;
Manugistics, Rockville, MD).
To determine whether HSV ocular infection induces iNOS expression,
RT-PCR analyses of infected corneas were performed on different PI
days. As shown in
Figure 1 , HSV-1 infection induced iNOS mRNA expression in corneas of infected
animals from PI day 2. Similar results were obtained with HSV-2
infection (data not shown). To study the role of NO in controlling HSV
ocular infection, we topically treated animals with different amounts
of AMG. This compound significantly altered the development of stromal
keratitis in the higher concentrations used. As shown in
Figure 2A , nearly 60% of mice infected with HSV-1 and treated with 0.5 mg AMG
showed signs of illness by PI day 4, whereas only 18% of the control
animals did so. Similar results were observed in HSV-2–infected mice
(Fig. 2B) . In this case, data obtained with 0.5 and 0.1 mg AMG where
very similar, and therefore only data obtained with 0.1 mg of AMG are
shown.
The progression of disease was exacerbated by AMG treatment in a
dose-dependent manner.
Tables 1 and 2 show that the different pathologic signs of ulcerative keratitis were
more frequent and appeared earlier in treated animals. These results
were further confirmed by microscopic examination of hematoxylin and
eosin–stained histopathologic eye sections obtained by standard
procedures. As shown in
Figure 3 , by day 3 PI, corneas from treated (0.5 mg) or control groups showed
slight extracellular edema and congestive blood vessels. Control
corneas
(Fig. 3A) had only scattered inflammatory cells, whereas those
from treated animals
(Fig. 3B) showed a conspicuous infiltrate
consisting primarily of neutrophils. Two days later, the inflammation
was established in corneas from nontreated and treated animals
(Figs. 3C 3D) , although it was more pronounced in the last group. Edema and
numerous dilated blood vessels were present in the stroma, and a few
plasmacytes (<5% of total inflammatory cells) were seen in the inner
side of the stroma. By PI day 7, there were no differences in corneas
from both groups, and first signs of ulceration were observed
(Figs. 3E 3F) .
Viral titers in eye washings obtained 24 hours after infection were
significantly higher in animals treated with 0.1 and 0.5 mg AMG (
Fig. 4A 4B ). This augmentation of virus titers in treated animals was
maintained during the following 2 days. In the same way, by PI day 1
larger viral titers were recovered from eye tissues of animals treated
with the higher AMG concentration
(Fig. 5) . In this experiment, diminution in viral titers from PI days 3 through
7 was less pronounced in treated animals. In contrast, significantly
more virus was recovered at PI day 8 in brains of treated animals.
HSV-1 levels (in PFU per milliliter) were as follows: AMG 0.5 mg,
1.46 ± 0.1 × 10
3 (
P < 0.05); 0.05 mg, 2.4 ± 0.5 × 10
2;
nonsignificant (NS); control, 1.3 ± 0.6 ×
10
2. HSV-2 levels were as follows: 0.1 mg,
2.3 ± 0.4 × 10
3 (
P <
0.05); 0.05 mg, 2.0 ± 0.6 × 10
2 (NS);
control, 2.3 ± 0.8 × 10
2.
In addition, PCR analysis of viral DNA in trigeminal ganglia revealed
that from PI day 3, 100% of the animals treated with the higher AMG
dose and infected with HSV-1 were positive for
ICP27 DNA,
whereas at PI day 3, 40% of the control animals showed the presence of
HSV DNA
(Fig. 6) . No samples were positive by PI day 1, whereas no differences were
observed between treated and control mice on PI days 5 and 7. Infected
mice that died on PI day 7 showed signs of encephalitis (ataxia,
ruffled fur, and paralysis). The higher doses of AMG significantly
augmented lethality of HSV infection. As shown in
Figure 7 , 40% of the animals infected with HSV-1 and treated with 0.5 mg died
by PI day 9, whereas values close to 10% were recorded in control
animals. Similar results were obtained with HSV-2. By PI day 12, 30%±
4% of the animals treated with 0.1 mg AMG had died, compared with
14% ± 2% in the control group (
P < 0.05). Finally,
animals receiving only 0.5 mg AMG but mock infected did not show any
sign of morbidity. No deaths were recorded in this group.
In humans, infection of the cornea with HSV results in a recurrent
immune-mediated inflammatory response, HSK, which is one of the most
common infectious cause of blindness in many countries. There seems to
be general agreement that HSK largely represents an immunopathologic
disease, and clinicians usually treat the lesions with
anti-inflammatory drugs along with or even without anti-herpesvirus
drugs. The mechanistic nature of HSK remains uncertain, but from
studies in experimental animals, evidence for a variety of mechanisms
has been forthcoming. These include toxic immune complexes,
delayed-type hypersensitivity, and lymphocyte
cytotoxicity.
3 27 Productive infection with HSV is crucial
for HSK manifestation. NO production by different cellular types in
response to viral infection or cytokines may be one of the mechanisms
of innate immunity involved in limiting viral replication after ocular
infection with HSV.
In this work, we investigated the effect of AMG, an iNOS inhibitor, on
the ocular infection of Balb/c mice that had been infected with HSV-1
or HSV-2. We observed an increase in viral titers in ocular washings,
eye tissue, and brains of treated animals. Viral DNA also was detected
earlier in trigeminal ganglia of treated mice. This correlates with the
observed augmentation in mortality and neuropathologic signs in the
treated groups. All these data agree with previous reports from other
investigators showing antiviral activity in vivo of NO against
different viruses,
8 9 15 17 20 thus indicating a role of
NO in the natural resistance against ocular infection with HSV. This
was supported by RT-PCR experiments showing iNOS mRNA induction in
corneas of infected mice. In treated animals, we also observed an
earlier appearance together with an aggravation of the symptoms
associated with HSK, such as blepharitis, edema, neovascularization,
and ulceration.
HSK in mice involves a chronic immune-mediated inflammatory response
that leads to total destruction of the corneal architecture and to
corneal perforation. The corneal inflammation characterized
predominantly by polymorphonuclear neutrophil (PMN) infiltration and T
cells belonging to the CD4
+ population is
regulated by the Th1 cytokine interleukin (IL)-2, tumor necrosis factor
(TNF)-α, and interferon (IFN)-γ.
28 The PMNs cause
progressive destruction of the corneal tissue, which appears to be
responsible for the blinding complications of HSV corneal infections in
humans. The Th2 cytokines IL-4 and IL-10 are not detected in cells that
infiltrate the HSV-infected corneas. IL-10 injection in the infected
cornea was shown to inhibit corneal inflammation.
29 A
replication-induced proinflammatory milieu in the cornea may be crucial
for the subsequent progression of HSK, possibly because of enhancement
of the expression of corneal agonists that drive HSK
manifestation.
30 Productive infection with HSV resulted in
rapid upregulation and sustained expression of chemokines such as
N51/KC, macrophage inflammatory protein (MIP)-1β, MIP-2, monocyte
chemotactic protein (MCP)-1, or such cytokines as IL-1, IL-6, IL-8,
IL-12, and TNF-α.
31 32 33 In particular, IL-8 has been
pointed out as an important chemokine during HSK.
34
In this work, we observed that AMG-treated animals showed increased PMN
infiltration in corneas at PI days 3 and 5 compared with
control animals. Recently, it has been reported that peroxynitrite,
formed by the reaction between NO and superoxide, regulates cytokine
function during inflammation. Peroxynitrite attenuates neutrophil and
monocyte chemotaxis induced by MIP-1α and IL-8 in a dose-dependent
manner, possibly by inhibition of chemokine binding to neutrophils and
monocytes.
35 36 Thus, we hypothesize that the observed
increase in the influx of PMNs to corneas of treated mice could be due
to inhibition of NO production by AMG treatment during HSV infection,
thus suppressing the NO-inhibitory effect on PMN chemotaxis. In the
same way, preliminary data from our laboratory obtained by the RT-PCR
technique indicate an earlier TNF-α response in corneas of
AMG-treated animals, although no differences were observed in IFN-γ
induction between treated and control animals (data not shown). These
data could be consistent with previous reports indicating an important
chemokine (MIP-2, IL-8)-inducing activity of TNF-α
37 38 39 and enhanced Th1 responses as a consequence of NO
impairment.
40 41 We are currently investigating this
issue.
Last, although we observed that inhibition of iNOS during the first
days of infection increased HSV ocular infection, according to Fuji et
al.,
42 intraperitoneal treatment of HSV-1 rats infected
intranasally with
N-monomethyl-
l arginine (
l-NMMA) from PI days 3 through 7
decrease neurologic symptoms and increased survival of treated animals.
They demonstrated that NO production is related to histopathologic
changes in the brain during infection. Thus, although NO production can
be beneficial as an antiviral effector against HSV and other viruses,
it also may be detrimental by contributing to disease during immune
responses, as previously reported.
43 44 Thus, to unravel
the role of NO in the natural resistance to HSV infections, it may be
crucial to determine the stage of viral disease in which this molecule
exerts its major effect.
The authors thank Juan Flo for valuable help in molecular biology.