<<

. 43
( 61 .)



>>

280

reaching the pmna, it 1soften difficult to see discrete lesions, such as isolated
vesicles, under the dlssectmg microscope. In this case, confluent areas of lesion
m the epidermis rapidly reach a scab stage.
3.3.2. Other indicators of Recurrent Infect/on in the Sk/n or Eye
3.3 2 1 ISOLATION OF VIRUS FROM SITES OF RECURRENT INFECTION
In the ear model, swab skin lesions or take small punch biopsies of skin.
In the eye model, the tear film can be sampled by placing 20 pL of tissue
culture medium onto the eye (surface tension ˜111hold this in place). Using a
plpet tip, irrigate and aspirate the fluid 10 times, bemg careful to avoid touch-
mg the cornea. This procedure 1s done on surgically anesthetized mice and
therefore can be repeated on subsequent occasions. More rigorous searches for
virus require the mouse to be kllled so that ear or eye homogenates or scrapings
can be taken. Material collected from such samples 1s maculated onto tissue
culture cells sensitive to HSV-1, e.g., Vero cells, and these are observed for cpe
3.3.2.2. DETECTION OF ANTIGENS
As further evidence of recurrent Infection in the skin or cornea, viral antl-
gens can be sought. Obviously this 1sapplicable to sections of the pmna or eye
but, without the tedious procedure of senal sectioning, it 1seasyto miss small foci
of mfectlon. To overcome this problem, we have developed methods for exam-
mmg preparations of whole tissue sheetsfrom the cornea or the pmna (Z9-22)
3.4. Defection of Virus in Ganglia
Clearly, a number of different techniques such as virus lsolatlon in culture,
immunocytochemistry, polymerase chain reaction (22) can be used to detect
the virus m ganglia at different times after the primary and/or reactivated
infection. Since all these require removal of the ganglion from the animal, we
describe here the necessary dissection procedures and also techniques we have
found useful to detect latent infection by explant culture All instruments that
come into contact with the ganglia should be sterile. Immediately before dls-
sectlon mice are terminally anesthetized.
3.4.1. Removal of the Trigeminal Ganglia (Eye Model)
1 Cut the heart through an mctslon m the rib cage to produce as much exsangu-
mation as possible
2. Pin the mouse m a prone position onto a cork board
3 Remove the skin from the cranium, remove the top of the cranium, and carefully
lift away the brain to expose the trigemmal ganglia lying at the base of the cranial
cavity. Pm the head securely to the cork board.
4 Remove the ganglta by careful dissection with small bow sprung (Noyes) SCIS-
Models of HSV in the Mouse 281

sors and pointed stainless steel watchmakers forceps (no. 5) while viewing the
procedure through a dissectmg microscope. A no I 1 scalpel can be used to assist
in the dissection. Cut the ophthalmic/maxillary and mandibular nerves as close
as possible to then points of entry into their foramma in the base of the skull. A
final cut IS made on the perrpheral side of the dorsal root entry zone. The nerve
changes from an opaque white on the perrpheral srde to a more translucent whrte
in the CNS side (see Note 17).

Cervical Ganglia (Eye Model)
3.4.2. Superior
1 See step 1 in Section 3 4 1.
2. Pin the mouse m a supme position on a cork board. Make a midline incision from
the lower chm to the thorax and remove the thymus gland
3. Move the trachea to the rrght for drssectton of the left ganglion or to left for the
other ganglion and pm the trachea in place.
4. The vagus nerve IS a good landmark, locate this and then find the carotid vessels
which run parallel to it. The ganglion lies at the bifurcation of the carotid artery
and can be exposed by blunt drssectron (see Note 18).
3.4.3. Removal of the Cervical Ganglia (Ear Model)
1. Make a complete circular incision m the skin Just below the rtb cage and pull the
skin back over the head.
2. Cut the heart through an incision m the rib cage to produce as much exsangu-
ination as possrble
3 Pin mouse in a prone posmon to a cork board. Remove the top of the cramum, the
brain and the dorsal half of the vertebral column m the cervical region. Remove
the exposed spinal cord (see Note 19)
4. The relevant cervical ganglia (see Note 20) are removed from their sockets in the
vertebrae using the instruments described earlier. A fine seeker may also be help-
ful for easing the ganglia out of then “sockets ”
3.4.4. Explant Cultures for the Demonstration of Latent Infection
Two methods are used widely to detect latent infection in gangha removed
from experimentally infected animals. In one, the ganglia or fragments of gan-
glia are cocultured in small flasks with monolayers of a cell line such as Vero,
which is sensitive to HSV. These cultures can be maintained for several weeks,
with changing of the medium, and the monolayers examined for evidence of
cpe. In the secondmethod, the ganglion tissue is maintained for 5 or 6 d in 0.5-mL
tissue culture growth medium at 35°C. The tissue is then ground m a tissue
grinder and 5OqL samples are placed m triplicate onto monolayers of Vero
cells. After 2 d, the monolayers are fixed and examined for vn-us cpe.
In our experience there is little difference in sensitivity between these meth-
ods, and, since the method involving the 5- or 6-d incubation gives more rapid
results, it is generally preferred (see Note 2 1).
Hill and Shimeld

3.5. Blood Samples
When the eye or ear models are used for some purposes, the taking of blood
samples may be required at different times followmg the primary Infection or
followmg induction of recurrent infection, In the ear model, either of the two
commonly used methods, from the tall vem or the retroorbttal sinus, may be
used (see Note 22). However, in the eye model, sampling should be restricted
to the tall vein; 400-600 pL of blood will provide 5C150 uL of serum.

4. Notes
1 HSV- 1 stram SC 16 was first isolated m 1974 from a factal leston (16) This was
ortgmally chosen as a stram of medium virulence when compared with other
isolates in mace Smce m the rabbit ocular model, HSV- 1 strain McKrae (which
was first isolated from an eye lesion) (23) consistently has given a high incidence
of both spontaneous and induced recurrent mfectton, this strain was chosen for
the mouse eye model
Clearly, strains such as KOS should be avoided, since this strain produced no
recurrent eye infection m rabbtts (24) and no recurrent disease m the mouse ear
model (7). There have been very few compartsons of different strains with respect
to their ability to produce recurrent mfectton (25).
With the zostenfonn model, even relatively avirulent strains, such as KOS will
give a high incidence of zostenform spread from the inoculation site to the pmna (7).
Hence, m contrast to the model of true recurrent mfectton, m this model such strams
may have advantages smce few tf any ammals are lost from lethal CNS disease
It 1s difficult to make general statements about the dose of virus since many
other factors such as the strain, age, and sex of mouse will affect the outcome at
any partrcular dose
2 Greater resistance to death followmg prtmary infection with HSV m mice, at
least after cutaneous or mtraperttoneal mfectton, appears to be linked to ptgmen-
tation However, we have found the reverse to be true when the cornea was the
inoculation site. For example, a dose of 1O3PFU of HSV- 1 strain McKrae killed
17/30 (57%) S-wk-old CB57 BL/J female mice compared to only l/19 (5%) NIH
mice of a similar sex and age. Hence, in studies of recurrent mfectton it would
seem sensible to use a non pigmented strain. Moreover, lesions of the cornea,
iris, and retina and features such as erythema in the skin, are more easily visible
m nonptgmented animals.
In a compartson of five different albino strains m the ear model, BALB/c,
gave a low incidence of recurrent disease (26) There was little difference between
the other strains but the inbred strain NIH gave a consistently good incidence of
recurrent disease and therefore was used m our subsequent work and by others
who have used the eye model (27,28)
Since, on the whole, males are more susceptible to infection with a wide vari-
ety of infectious agents, it might be argued that using males might help to maxi-
mize recurrent disease However, the propensity of males to fight can, by leading
Models of HSV in the Mouse 283

to skm damage, seriously interfere with detection of recurrent skm lesions or m
severe cases necessitate killing the mice for humane reasons. Moreover, follow-
ing primary infection without the cover of immune serum, particularly in the
cornea, the greater susceptibility of males can lead to unacceptable losses from
severe CNS disease For example, with Gwk-old NIH mice inoculated with 1O4
pfu of HSV-1 strain McKrae the mortality rates were 37% for males compared to
15% for females Hence, for routine purposes, to avoid these problems females
are used for ear, eye, and zosterlform models (although the eye model was ongl-
nally established m male mice) However, for particular purposes, e g , expen-
ments on local immune responses followmg ocular infection, males may have an
advantage since local production of IgA is higher in the male lacrimal gland (29)
Finally, It has been known for many years that the resistance of mice to infection
with HSV increases significantly with age In the first development of the ear model,
4-wk-old animals were first used as a means of ensuring a high susceptib&y to both
primary and therefore, It was hoped, recurrent infection. In the eye model, 4-wk-old
mice even when infected under cover of immune serum have an unacceptably high
mortality rate (44%) that has reduced to 5% by 8 wk old. For this reason and because
they are immunologically mature, older mice are used in the ocular model.
On various occasions, bitter experience, m particular with subclmlcal mfectlons
with mouse hepatitis virus and/or infection with the common pm worm, has
taught us the absolute necessity for using SPF animals m these models Animals
with such mfectlons show a significant increase m resistance to primary mfec-
tion, a reduction m the incidence of latent infection m sensory ganglia, and a
concomitant reduction in the incidence of recurrent infection followmg a suitable
reactivating stimulus
Surgical anesthesia with sodium pentobarbitone gives a narrow safety margin
and 1snot to be recommended.
The doses for surgical anesthesia are: Midazolam hydrochloride 0 024 mg/lO g
and Fentanyl citrate 0.0015 mg/lO g and fluanisone 0.048 mg/lO g or a combma-
tlon of 0.1 mg/lO g xylazme and 1 mg/lO g ketamine. The first combmatlon has
two disadvantages, the two ingredients cannot be mixed and therefore have to be
delivered as two separate mJectlons and the most consistent result ts obtamed
following subcutaneous delivery (which 1s technically more demanding than
intraperitoneal delivery). However, the recovery rate with this combmatlon 1s
extremely good. It should be noted that sterile water should be used to dilute
these drugs since some of the components will precipitate m saline These doses
may need adjustment for different strains and ages of mice. It is important to
keep anesthetized animals warm during the recovery period
We have seen the followmg abnormahtles in uninfected mice: cornea1 vascular-
lzatlon of varying degrees, cataract, supraorbltal abscesses,and band-shaped stromal
opacities The incidence of such abnormalities varies with the strain of mouse.
If a primary eye infection IS required without passive lmmunizatlon of the mice it
is necessary to mmimlze the risk of a secondary bacterial infection by treating the
eyes daily from d 1 to 10 with antiblotlc eye cream
Hill and Shimeld

8 The dose of anti-HSV- 1 rabbit serum is higher than was used origmally since m
experiments where a range of doses were tested the higher dose was found to
give statistically significantly more mice that shed vnus on reactivatton than am-
mals given the origmal dose. Normally we use rabbit serum for passive immum-
zation; however, we and others (27,28) also have used human serum and have
found no dtfference between the two. There 1s considerable variation between
mice m the levels of anti-HSV-1 neutrahzmg antibodies acquired followmg
passive mnnunizatton Attempts to decrease these variations by delivermg the anti-
viral serum (appropriately diluted) via the tail vem were successful. However, this
procedure was not adopted, since it failed to increase the incidence of mice that
reactivated and was far more time-consummg than mtraperitoneal immumzation.
9. In our experience, application of virus to the eye m the absence of cornea1 trauma
produces blepharms but no cornea1 disease and a low incidence of latency even
when young mice are Inoculated with high doses of vnus (30).
Only gentle pressure is required to damage the cornea1 epithelium and slightly
greater pressure will extend the moculatlon site to Include the cornea1 stroma
Inoculatmg the cornea1 stroma alone is best done using an inJection into this tissue
Excessive pressure during scarification ˜111 result m puncture of the cornea
and collapse of the anterior chamber, and if this occurs the mouse should be killed
In the United Kingdom, for humane reasons, maculation of vnus is restricted
to one eye, we recommend this restrtctton should be adopted elsewhere.
10 Rather than subcutaneous maculation, vuus can be Inoculated mto the pmna by
scartfication of the upper surface of the pinna through a small volume (10 pL) of
virus suspension. However, the thinness of the pmna makes this difficult to do
uniformly without cutting through the pmna and/or causing bleeding.
11. If the mice are Just placed under the lamp they will tend to close then eyes and
most of the cornea ˜111 not be irradiated. If only irradiation of the cornea is
required this can be achieved by proptosmg the eye through a hole m a piece of
cardboard of sufficient size to expose the cornea but shield the lids
12. UV is a known sttmulant to recurrent disease m humans The optimum dose of
UV is a matter for discussion, in our studies we have tested a range of doses and
have found that treatment of the eye for 90 s with a lamp emitting a peak of 4 02
mJ/cm2 results m the maximum incidence of viral shedding In contrast, others
also using our model suggest that a lower dose increases viral shedding, how-
ever, they provide no statistical evidence to support this claim (27) In our eye
model, the incidence of reactivation, as judged by shedding of vu-us m the tear
film, varies from 25 to 60%; the addition of systemic steroids to the UV nradia-
tton (4,5) does not appear to Increase this incidence However, these drugs do
appear to increase the severity of recurrent ocular disease and to mcrease the
duration of shedding of vu-us
13 In the ear model, successful induction of recurrent disease on the pmna has only
been achieved by local treatments to this site These have included UV light (31),
local inJections of prostaglandms (31), application of irritants such as retmom
acid and xylene (32), stripping the epidermis with cellophane tape (17) Systemic
285
Models of HSVin the Mouse
treatments, m particular immunosuppression, produced insignificant results (33).
It is noteworthy that some stimuli like DMSO, may produce gangliomc reactiva-
tion but little or no recurrent disease (32). Of these various local stimuli, the most
widely applied has been physical trauma induced by cellophane tape stripping of
the skin A technique originally used to study the effects of mild physical trauma
on factors such as epidermal cell mitosis (34)
14. Although eyes can be examined under a dissectmg microscope only gross dis-
ease will be discernible; for detailed examination a slit-lamp microscope is
essential. Unanesthetized animals can be viewed but for a more rigorous examma-
non anesthesia, or at least sedation, is essential
Two types of slit-lamp illumination are useful. Direct illumination with a nar-
row beam directed obliquely into the eye that provides an optical section of the
trssues under examination. With this method the shape and thickness of the cor-
nea can be assessed and any irregularities in the cornea1 epithelium, such as small
lesions, are easily detected. Retro illumination with a wide slit beam directed
straight mto the eye IS a useful method for slit-lamp exammation of albino mice;
the eye, lids and surroundmg skm can be scrutnnzed. Generally the highest power
will be required to examine mouse eyes
Before examining experimental eyes the operator should be completely famil-
iar with normal eyes. Particular attention should be paid to the shallowness of the
anterior chamber and the position of the iris vessels that can be confused easily
with cornea1 vessels by the inexperienced.
The followmg tissues can be examined for signs of damage or disease. cornea,
limbal vessels, anterior chamber, pupil, iris, lens (not fully seen with an undilated
pupil), lids, and surrounding skin The conjunctiva is only partially accessible to
exammation since it is not possible to evert the mouse eyelids.
Fluorescem stain can be used to identify the bed of ulcers but requires observa-
tion with blue light to see fluorescence. Rose Bengal can be used to stain the mar-
gins of ulcers. One to two microliters of stain should be placed on the cornea, the
eyelids opened and closed several times to spread the stain evenly and the animal
examined immediately. If necessary, any excess stain can be removed by rapidly
washing the eye with sterile saline. The normal adult NIH mouse cornea, unlike the
human cornea, shows fine punctate staining with Rose Bengal; and, in addition, this
stain should be used with caution since tt has antiviral properties (3536).
Using retro illummation, the retina and vitreous can be examined. In an anes-
thetized mouse, place one drop of cyclopentolate hydrochloride on the eye; after
approx 10 min the pupil will be dilated. A drop of carboxymethyl cellulose is
placed on the eye and a glass cover slip is pressed gently onto the cornea to
neutralize the refraction of the cornea. By moving the mouse, almost the entire
retina can be brought mto focus. Examination of the mouse retma by direct
ophthalmoscopy does not appear to be feasible with a human ophthalmoscope.
Use of phenylephrine hydrochloride as a mydriatic rapidly causes the lens to
become opaque; usually this clears by the next day.
In the eye model, it is important to be able to distmguish damage to the cornea
Hill and Shimeld
286
caused by UV n-radiation and that produced by viral disease, especially the early
stages of disease, which may occur concurrently with damage from W Control

<<

. 43
( 61 .)



>>