 |
Introduction |
Oral
N-acetyl-L-cysteine (NAC) has been
used clinically as a remedy for chronic obstructive pulmonary disease
for decades, and it is generally believed that its beneficial effect is
due to a mucolytic activity. However, it has not been possible to demonstrate presence of the drug in epithelial lining fluid after oral
administration (Cotgreave et al., 1987
); this is not fully compatible
with its proposed in vivo mucolytic activity. Moreover, NAC is reported
to primarily reduce exacerbation rates in the disease (Boman et al.,
1983). Considering this, it is possible that NAC acts in chronic
obstructive pulmonary disease patients, at least in part, by enhancing
their host defense.
In support of such an idea, numerous reports show that NAC treatment in
vitro can modulate activities of lymphoid cells. Human T cell IL-4
synthesis and B cell IgE and IgG4 production are decreased by the
compound, whereas T cell IL-2 production is enhanced (Jeannin et al.,
1995). NAC inhibits apoptosis of T cell hybridomas (Jones et al., 1995)
and reverses down-regulation of IL-2 mRNA and IL-2 activity in human
lymphocytes (Flescher et al., 1994). In biochemical terms, NAC has been
generally considered to act as a precursor of glutathione (Dröge
et al., 1994; Jeannin et al., 1995; Jones et al., 1995; Yan et al.,
1995), as an antioxidant (Aruoma et al., 1989), and/or as a reductant
that modulates redox-sensitive transcription factors like AP-1 or
NF-
B or otherwise influences transcriptional events (Schreck et al.,
1991; Dröge et al., 1994; Yan et al., 1995; Xia et al., 1996).
The effects of NAC, seen at close to mM concentrations, conform to the
in vitro stimulatory effects of other low molecular weight thiols, like
2-mercaptoethanol and glutathione, on lymphoid cell function (Hiestand
and Strasser 1985a; Dröge et al., 1994; Jeannin et al., 1995).
Modulation of in vivo immune responses by NAC have also been reported
(Hiestand and Strasser, 1985b; Kinscherf et al., 1994; Jeannin et al.,
1995).
Reports have suggested that the efficacy of in vivo modulation by some
thiols relies on the presence of an intact disulfide bridge (Hiestand
and Strasser, 1985b; St Georgiev, 1988). Therefore, we examined the
effects of the disulfide dimer of NAC,
N,N'-diacetyl-L-cystine (DiNAC) in in vivo systems reflecting various immune responses in
rodents. One such system is the delayed type hypersensitivity (DTH)
reaction, an in vivo reflection of a cell-mediated immune response
(Askenase, 1992). The expression DTH is often used to denote reactions
induced by protein antigens as well as by contact sensitizers. However,
recent reports indicate that different subsets of T cells govern
classical DTH reactions to protein antigens and contact sensitivity
(CS) reactions (Grabbe and Schwarz, 1998). The present report deals
with effects of DiNAC in CS skin reactions induced by
4-ethoxymethylene-2-phenyl-2-oxazolin-5-one (oxazolone), fluorescein isothiocyanate (FITC), and 2,4-dinitrofluorobenzene (DNFB),
as well as in a DTH reaction to methylated BSA (mBSA) assessed
by foot pad swelling.
| |
Materials and Methods |
Synthesis and Sources of Compounds.
Compounds
N,N'-diacetyl-L-cystine
(DiNAC; referred to as D7042) and its
di-L-lysinium salt (referred to as D7193 - preferred for use over D7042 due to pharmaceutical reasons),
N-acetyl-L-cysteine, ethylene-2,2'-bis(dithio)bis(ethanol) (ADA 202-718),
L-homocystine, N,N'-diacetyl-L-homocystine,
N,N'-diacetyl-cystamine,
L-cystathionine, N,N'-diacetyl-L-cystathionine,
racemic lanthionine, and racemic N,N'-diacetyl
lanthionine, were synthesized by Astra Draco AB (Lund, Sweden) or Astra
APP (Södertälje, Sweden). Aminoguanidine bicarbonate salt,
L-buthionine-[S, R]-sulfoximine (BSO),
diethyldithiocarbamate (DDTC), DNFB, FITC-isomer 1, and mBSA were all
obtained from Sigma Chemical Co. (St. Louis, MO).
Bis-(2-hydroxyethyl)-disulfide (HEDS) was from Fluka AG (Buchs,
Switzerland), whereas oxazolone was obtained from either BDH Chemicals
Ltd. (Poole, Dorset, England) or from Sigma. Cyclosporin A (CSA;
Sandimmun) was obtained from Sandoz Pharma AG (Basel, Switzerland) and
rapamycin and
N
-nitro-L-arginine
methyl ester (L-NAME) were obtained from Alexis Corp. (San Diego, CA). Sodium pentobarbiturate (Mebumal) was purchased from Apoteksbolaget (Umeå, Sweden) and enflurane (Efrane) was obtained
from Abbott S.p.A. (Campoverde Lieti, Italy). Sources of
antibodies used in immunohistochemical experiments are given below.
Animals.
Male and female mice of the following strains were
obtained from Bomholtsgaard (Ry, Denmark) or Charles River Breeding
Laboratories (Kent, UK): BALB/c, CD-1, C57BL/6, CBA/J, NOD,
MRL/Mp and MRL-lpr/lpr. The mice were used at the weight of
18 to 20 g. Female Sprague-Dawley rats were from Möllegaard
(Ejby, Denmark); they were used at 150 g. The animals were caged
for at least 8 days after arrival before experiments were initiated.
Animals had free access to food (R3, Ewos, Södertälje,
Sweden) and water. The light period in the room was 12 h (6:00
AM-6:00 PM). Rabbits of New Zealand strain of both sexes were obtained
from HB Rabbit Farm, Lidköping, Sweden. All animal study designs
were approved by local ethics committees.
Induction and Assessment of CS to Oxazolone, DNFB, and FITC.
Animals were sensitized (day 0) by a single epicutaneous application of
150 µL (mice), 400 µL (rats), or 1000 µL (rabbits) 3% oxazolone
solution in absolute ethanol-acetone (3:1) on the shaved thorax and
abdomen. Treatment with D7042, D7193, or other compounds was normally
initiated by oral feeding (gavage) of 10 ml/kg body weight of an
appropriate concentration of the compound immediately after
sensitization. Treatment continued once daily up to and including day
6. Control animals were given the corresponding amount of vehicle
(saline or water as indicated in the figure legends). Other treatment
regimens with D7042 or D7193 were examined as specified in the text. In
separate experiments, D7193 was administered in the drinking water
(prepared fresh each day) in free access, injected either i.v. in the
tail vein (20 µL) or i.p. (200 µL), or instilled intratracheally
under light Efrane anesthesia. Eight days after sensitization (i.e., on
day 7), mice were challenged on both sides of both ears by topical
application of a total of 20 µL of 1% oxazolone dissolved in peanut
oil; rats were challenged with 40 µL and rabbits with 400 µL of the
same solution. Ear thickness was measured before and 24 h after
challenge (in some experiments also at 48 h) using an Oditest
spring caliper handled manually (van Loveren et al., 1984) or coupled
to a computer-directed motor device. Challenges and measurements of ear
thickness were performed under light pentobarbital anesthesia. Most
experiments were performed under coded conditions.
The intensity of the CS reactions was expressed as increase in ear
thickness in mm, i.e., T24/48
-T0 mm, where T0 and
T24/48 represent the ear thickness before and 24 or 48 h after challenge, respectively. The figure recorded for
each animal and time point is the mean of the measurements on both ears
in individual tests. The results were expressed as the mean ± S.E.M from groups of 8 to 10 animals. Degree of significance for
differences between means of groups was obtained by Student's
two-tailed t test.
Mice were sensitized to DNFB or FITC, and challenges were performed as
described (Tang et al., 1996). The magnitude of the CS reactions
induced by DNFB and FITC were assessed as described above for
oxazolone-induced reactions.
Immunohistochemical Examinations.
Histological and
immunohistological examinations were performed in blind manner on coded
ears from BALB/c mice untreated, sensitized, and challenged with
oxazolone with or without treatment with D7193. The ears were cut off
at the bases and immediately snap-frozen in isopentane in liquid
nitrogen and stored at 
70°C. The frozen ears were processed for
routine histological examinations and for immunohistochemical
examinations using the ABC-technique according to principles described
previously (Scheynius et al., 1996). The frozen specimens were cut
through the center of the ear extending from the top to the base in a
cryostat. The cryostat sections, 6 µm thick, were acetone-fixed, and
incubated in 0.3% H2O2 in
phosphate-buffered saline for 15 min at room temperature to block
endogenous peroxidase. Sections were then treated with normal rabbit
serum to reduce nonspecific binding and with Avidin D solution and
biotin-solution blocking kit (Vector Laboratories, Inc., Burlingame,
CA). This was followed by incubation with rat anti-mouse monoclonal
antibodies (mAbs) (see below), biotinylated anti-rat IgG (Vector
Laboratories) and avidin-biotin-peroxidase complex (Dako A/S, Glostrup,
Denmark). The peroxidase reaction was developed with
3-amino-9-ethylcarbazole (Aldrich Chemical Co., Steinheim, Germany).
The sections were counterstained with Mayer's hematoxylin. Optimal
dilutions of the mAbs were determined in control experiments with
sections from normal mouse spleen and skin. Each ear specimen was also
processed for hematoxylin and eosin staining. The following rat
anti-mouse mAbs were used: anti-CD4 (L3T4, clone H129.19), anti-MHC
class II I-Ab, d (clone B21.2), anti-CD11a (the
-chain of LFA-1, clone FD441.8), anti-CD11b (the -chain of Mac-1,
C3biR, clone M1/70.15.11.5), anti-CD54 (ICAM-1, clone YNI/1.7.4), and
anti-CD44 (clone IRAWB14.4) all obtained from their hybridoma cells
expanded in vitro, anti-CD8 (Lyt2) obtained from Serotec Ltd.,
(Kidlington, Oxford, UK), and anti-CD106 (VCAM-1) from PharMingen (San
Diego, CA).
Dermal cell infiltration was assessed on hematoxylin and eosin-stained
sections on a semiquantitative scale from 0 to 3, where 0 = normal, 1 = small, 2 = moderate, and 3 = large dermal
cell infiltrates. The immunoperoxidase-stained sections were evaluated on a semiquantitative scale ranging from 0 to 3 in dermis, where 0 = no, 1 = few, 2 = moderate, and 3 = many positive
cells. The scale was adjusted for each antibody, so that the grading
"3" refers to the maximal number of positive cells within all
specimens. In epidermis, CD4+,
CD8+, LFA-1+, and
C3biR+ cells were counted at magnification 400×
in ten grids or five grids, respectively, and expressed on a 0 to 3 scale, where 0 = 0 to 1 positive cell, 1 = 2 to 5, 2 = 6 to 9, and 3 = 10 or more positive cells. Keratinocytes positive
for MHC class II antigens, ICAM-1, and CD44 were determined
semiquantitatively. At least two sections per antibody and specimen
were examined.
Multi-Probe RNase Protection Assay (RPA) Analysis of Cytokine
Profiles in CS Reaction Sites.
Four different groups of
mice were investigated for chemokine/cytokine expression; nonsensitized
mice treated either with vehicle or with 3 µmol/kg D7193 and mice
sensitized to and challenged with oxazolone that were treated either
with vehicle or with 3 µmol/kg D7193. Three individuals in each group
sacrificed 24 h after challenge were analyzed. RNA was prepared as
described (Chomczynski and Sacchi, 1987) from ears that had been cut
from sacrificed mice and snap-frozen in liquid nitrogen. The ears were
ground in liquid nitrogen and the resulting powder was suspended in
homogenization solution (4 M guanidium thiocyanate, 0.1 M Tris-Cl pH
7.5, 1% 
-mercaptoethanol). The slurry was sheared with a disposable
syringe using a 0.9-mm needle. Expression of mRNAs for eotaxin,
glyceraldehyde-phosphate dehydrogenase, IL-2, IL-4, IL-5, IL-6, IL-9,
IL-10, IL-13, IL-15, IFN-
, IP-10, L32, lymphotactin, MCP-1,
MIP-1, MIP-1, MIP-2, RANTES, and TCA-3 were examined with the
Riboquant multi-probe RPA system from PharMingen according to the
manufacturer's protocol.
In vitro transcription of linearized DNA templates was performed for
each probe in a total volume of 20 µL (10 µL
[-32P]UTP, 1 µL GACU pool, 2 µL DTT, 4 µL 5X transcription buffer, 1 µL RPA template set, 1 µL RNasin, 1 µL T7 RNA polymerase) and incubated at 37°C for 1 h. The
reaction was terminated by adding 2 µL of DNAse I and incubating at
37°C for an additional 30 min. Twenty six µL of 20 mM EDTA and 2 µL of yeast tRNA (2 mg/mL) were added, and the reaction was extracted
once with phenol/24:1 chloroform:isoamylalcohol (1:1) and once with
chloroform/isoamylalcohol alone. The aqueous phase was precipitated by
adding 50 µL of 4 M ammonium acetate and 250 µL of ice-cold
ethanol. After centrifugation, the pellet was dissolved in 50 µL of
the hybridization buffer provided in the kit. Incorporation of
radiolabel was quantified using a Bioscan Quick-count and the probe was
diluted to the recommended 3.0 × 105
cpm/µL. Ten µg of total RNA was used for each incubation together with approximately 5 × 105 cpm of probe in
total volume of 20 µL of hybridization buffer. The annealing reaction
mixture was incubated at 56°C overnight. RNase digestion was
performed by adding 100 µL of RNase mix (provided in the kit)
containing 192 pg/µL RNase A, 0.6 U/µL RNase T1, and RNase buffer
to the reaction. The mixture was incubated at 37°C for 45 min. To
stop the digestion, proteinase K was added together with yeast RNA as
carrier. The reaction mixture was extracted once with 130 µL
phenol/24:1 chloroform:isoamylalcohol (1:1). The aqueous phase was
removed to a new tube and precipitated with 120 µL of 4 M ammonium
acetate and 650 µL of ice-cold ethanol. Reactions were precipitated
at 20°C for 1 h; after centrifugation the pellet was dissolved
in 5 µL of formamide-loading buffer. Dissolved precipitates were
heated at 90°C for 3 min before resolution on 5%, 1X TBE, denaturing
polyacrylamide gels. Gels were transferred to Whatman paper and dried
in a Savant vacuum dryer. Dried gels were placed in a phospho-Imager
cassette (Molecular Dynamics) and exposed overnight. The phospho-screen
was scanned in a Molecular Dynamics Storm 860 scanner and analyzed with
Image-Quant software (Molecular Dynamics).
Footpad DTH Responses in Mice Sensitized to mBSA.
A DTH
response to mBSA was induced in BALB/c mice as described (Tarayre et
al., 1990). Briefly, mBSA emulsified in Freund's complete adjuvant was
injected intradermally on day 0 (a total of 0.25 mg of mBSA in a total
volume of 0.1 ml of emulsion was given at three injection sites).
Challenge was performed by injection on day 7 of 0.025 ml mBSA solution
(25 mg/mL) in the right hind paw; the left hind paw, injected with the
corresponding volume of the vehicle, served as a control. Twenty-four
hours later, the animals were deeply anesthetized and both hind paws
were cut. The DTH reaction was estimated from the difference in weight
between the two hind paws.
DTH Granuloma Reaction Induced by mBSA.
A s.c. DTH reaction
induced by mBSA resulting in a quantifiable chronic granulomatous
lesion (Dunn et al., 1989) was used to assess effects of treatment with
D7193 on a chronic cell-mediated inflammatory reaction. Briefly, mice
were sensitized to mBSA (a total of 25 mg of mBSA and 500 mg of Dextran
in a total volume of 0.1 ml of emulsion was given at a total of three
injection sites). The animals were challenged 3 weeks later by s.c.
implantation of Millipore filters (10 mm diameter) soaked in 25 µL of
mBSA solution (3 mg/mL). After 7 days, filter implants were dissected away from the lesions and weighed (wet/dry) for quantification of
inflammation. The results were expressed as the mean ± S.E.M from
groups of 8 to 10 animals. Degree of significance for differences between means of groups was obtained by Student's two-tailed
t test.
Pharmacological Modulation of the CS Skin Reactions to Oxazolone
in Mice
Mice sensitized to oxazolone as described
above were treated orally with D7193 or vehicle and simultaneously with
either freshly prepared NAC (3 µmol/kg or 30 µmol/kg orally), CSA
(30 µmol/kg orally), rapamycin (3 µmol/kg orally), aminoguanidine
(AMG; 813 µmol/kg i.p. once daily from day 2 before sensitization up
to and including the day of challenge, day 7),
L-NAME (115 µmol/kg orally by gavage from day 5 before
sensitization to day 6), or BSO (2 mmol/kg twice a day i.p. and 20 mM
in drinking water from day 5 before sensitization up to and including
the day of challenge, day 7). The results were expressed as the
mean ± S.E.M. from groups of 8 to 10 animals. Degree of
significance for differences between means of groups was obtained by
Student's two-tailed t test.
| |
Results |
Influence of DiNAC (D7042/D7193) on CS Skin Reactions to Oxazolone
in Mice.
The capacity of D7042 to modulate the CS skin reaction to
oxazolone is illustrated in Fig. 1, which
shows the results of measurements of ear thickness 24 h after
challenge in animals treated with D7042 compared with those from
animals treated with NAC. Treatment with both compounds causes a
dose-dependent increase of ear thickness; however, the log
dose-response relations indicate that DiNAC is 100 to 1000 times more
potent than NAC and twice as effective at the highest examined
comparable dose.
View larger version (74K):
[in this window]
[in a new window]
|
Fig. 1.
Effects of treatment with
N,N'-diacetyl-L-cystine
(D7042) or NAC, both given by oral gavage once daily during days 0 to
6, on the CS response to oxazolone in BALB/c mice sensitized to
oxazolone on day 0 and challenged on day 7. The response was measured
as the increase in ear thickness (mean ear swelling) ± S.E.M. 24 h after challenge (day 8). The response of control animals that
received saline is illustrated by the open left column.
n = 10-15, **p <.01,
***p <.001.
|
|
Separate experiments showed that the capacity of D7042 to enhance the
CS reaction at either 24 or 48 h after challenge did not differ if
treatment was terminated on day 6 as described above or if it continued
until evaluation (i.e., treatment for days 0-9, challenge day 7, evaluation day 8 and day 9; data not shown).
Experiments comparing the effects of D7042 (salt-free form of DiNAC) to
those of D7193 (the lysine salt form of DiNAC) at 0.03 or 3 µmol/kg
did not reveal any difference in activity between the two preparations.
Doses of lysine corresponding to those present in the administered
doses of D7193 did not affect the CS reaction (data not shown).
The effect of D7193 is highly reproducible from experiment to
experiment. Analysis of data from experiments performed during a period
of 6 years shows that the group mean increase in ear swelling 24 h after challenge in oxazolone-sensitized control animals is 0.09 mm (n = 55) compared with a group mean
increase of 0.16 mm in animals treated with 3 µmol/kg D7193
(n = 55; p <.001).
Separate kinetic experiments showed that a minute degree of ear
swelling seen already at 2 h after challenge (the early phase of
the CS reaction; cf. Askenase, 1992) is not enhanced in animals treated
with D7193 (results not shown).
Dose-response assessments of the effects of D7193 after oral, i.p.,
i.v., or intratracheal administration (Fig.
2) reveal that 0.03 µmol/kg as well as
3 µmol/kg of the compound stimulated the CS reaction to a similar
degree irrespective of route of administration whether the reactions
were read 24 h or 48 h (not shown) after challenge. In these
experiments, even the 0.0003 µmol/kg dose of D7193 induced a
significant enhancement of the CS reaction. Mice given D7042 in the
drinking water displayed increased CS reactivity of the same degree as
those given the compound by gavage treatment once daily (results not
shown).
View larger version (61K):
[in this window]
[in a new window]
|
Fig. 2.
The influence of the route for administration on
effect of D7193. The compound or the vehicle (saline) was given to
BALB/c mice sensitized to oxazolone on day 0 at the indicated dose
either p.o., i.v., i.p., or intratracheally (it) as indicated once
daily for days 0 to 6. The response to challenge with oxazolone on day
7 was measured as increase in ear thickness 24 h after challenge
(day 8). Controls received saline. n = 10, *p <.05, ***p <.001.
|
|
Optimal enhancement of the CS response to oxazolone by D7193 required
that treatment was performed during the whole sensitization period,
although animals treated for shorter periods of time (i.e., for days
0-4 or days 4-6) also express significantly enhanced CS responses
(Fig. 3). Treatment every second day is
as effective as daily treatment, but treatment each third day is not
effective (results not shown).
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 3.
The influence of treatment regimen on the capacity of
D7193 to enhance the CS response to oxazolone in sensitized BALB/c
mice. The experimental design followed the standard procedure except
for the treatment periods. The figures given below the
x-axis refer to the time period of oral treatment once
daily with 3 µmol/kg D7193. Day 0 refers to the time of
sensitization. n = 10, * p <.05, ***
p <.001.
|
|
We examined whether the effect of treatment with D7193 on the CS skin
reaction in oxazolone-sensitized animals 24 h after challenge was
strain-dependent. As shown in Fig. 4,
mouse strains differ in their CS response to oxazolone. However, the
relative degree of enhancement induced by D7193 treatment was similar
in the strains examined whether they can be considered a TH1 phenotype (CD-1, C57BL/6, CBA/J), a TH2 phenotype (BALB/c) based on responses to
Leishmania infection (see Bogdan et al., 1993), or carries an autoimmune phenotype (MRL-lpr/lpr, MRL-Mp or NOD).
View larger version (58K):
[in this window]
[in a new window]
|
Fig. 4.
The effect of oral treatment with 0.03 µmol/kg or
3.0 µmol/kg of D7193 day 0 to 6 on the CS response to oxazolone
24 h after challenge in animals of different mouse strains
sensitized to oxazolone on day 0. n = 10 for each
group, *p <.05, **p <.01,
***p <.001.
|
|
Modulation of Effects of D7193 on CS Skin Reactions to Oxazolone in
Mice by Selected Agents.
To assess the influence of NAC on the
effect of D7193, CS reactions in BALB/c were assessed in mice given
either D7193 at the indicated doses, freshly prepared NAC at
the indicated doses, or both agents (administered separately by gavage
once daily according to the standard schedule treatment). The results
show that simultaneous treatment with NAC reduces the enhancement of
the D7193-induced CS reaction to levels that are even lower than those
recorded in animals treated with NAC alone (Fig.
5A). Oxazolone-sensitized mice treated
with the glutathione-depleting agent BSO (2 mmol/kg twice a day i.p.
and 20 mM in drinking water (Leeuwenburgh and Ji, 1995) express
markedly increased CS reactions, which are even further augmented into
very large reactions by simultaneous D7193 treatment (Fig. 5B).
View larger version (35K):
[in this window]
[in a new window]
|
Fig. 5.
A, the effect of oral treatment with 0.03 µmol/kg
or 3.0 µmol/kg of D7193 separately, 3 µmol/kg or 30 µmol/kg of
NAC separately, or the indicated dose of both compounds simultaneously
on days 0 to 6 on the CS response to oxazolone in sensitized BALB/c
mice (n = 10). Controls received saline.
**p <.05, ***p <.001 (experiments with
single drug treatment compared to controls),
#p <.05, ##p
<.01, ###p <.001 (animals treated with NAC
compared to animals treated with a combination of NAC and D7193). B,
the effect of treatment with BSO [2 mmol/kg twice a day i.p. and 20 mM
in drinking water from day 5 before sensitization up to and including
the day of challenge (day 7)], D7193 (3.0 µmol/kg) day 0 to 6 or a
combination of the two drug regimens on the CS response to oxazolone in
sensitized BALB/c mice (n = 10).
***p <.001 (experiments with single drug treatment
compared to controls), ###p <.001 (animals
treated with D7193 compared to animals treated with a combination of
BSO and D7193).
|
|
The immunosuppressive agent CSA at a dose of 30 µmol/kg slightly
reduces the CS reaction to oxazolone when given by gavage once daily
from days 0 to 6 to BALB/c mice sensitized to oxazolone (Fig.
6A). In animals treated with CSA in this
way, simultaneous treatment with 3 µmol/kg D7193 leads to a partly
reduced CS reaction compared with that seen in animals treated with
D7193 alone (Fig. 6A). Oxazolone-sensitized mice treated with rapamycin
to inhibit the p70 S6 kinase pathway (Proud, 1996) from days 0 to 6 do
not display significantly reduced CS reactions. Simultaneous treatment with rapamycin and D7193 leads to effects similar to those seen with
treatment with CSA; rapamycin blocks the enhancement recorded with the
low dose of D7193 and partly blocks that induced by treatment with 3 µmol/kg D7193 (results not shown). Oxazolone-sensitized mice treated
with the nitric oxide (NO) synthase inhibitor AMG (Tracey et
al., 1995; Brenner et al., 1997) display slightly reduced CS reactions.
Still, the CS reactions are enhanced in mice simultaneously treated
with AMG and D7193 (Fig. 6B). In one of two experiments, animals
simultaneously treated with AMG and D7193 expressed a significantly
enhanced CS reaction compared with that in animals treated with D7193
alone (data not shown). Oxazolone-sensitized mice treated orally with
another NO synthase inhibitor, L-NAME, at a dose which is
effective in colonic inflammation in rats (Kiss et al., 1997)-a
species less sensitive to L-NAME than mice (Tracey et al.,
1995)-displayed slightly increased CS reactions; despite this,
simultaneous treatment with L-NAME tended to reduce the enhancement of the CS reaction brought about by D7193 treatment (Fig.
6C).
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 6.
A, the effect of oral treatment with CSA (30 µmol/kg), D7193 (0.03 or 3.0 µmol/kg), or a combination of the
drugs (given simultaneously but separately) on days 0 to 6 on the CS
response to oxazolone in sensitized BALB/c mice (n = 10). **p <.01, ***p <.001
(experiments with drug treatment compared to controls),
##p <.01, ###p
<.001 (animals treated with given dose of D7193 compared to animals
treated with a combination of CSA and the corresponding dose of D7193).
B, the effect of treatment by gavage with AMG [100 mg/kg (813 µmol/kg) i.p. once daily from day 2 before sensitization up to and
including the day of challenge (day 7)], D7193 (0.03 or 3.0 µmol/kg), or a combination of the drugs (given simultaneously but
separately) day 0 to 6 on the CS response to oxazolone in sensitized
BALB/c mice (n = 10). **p <.01,
***p <.001 (experiments with drug treatment compared to
controls). C, the effect of treatment by gavage with L-NAME
[115 µmol/kg once daily from day 5 before sensitization up to the
day for challenge (day 6)], D7193 (0.03 or 3.0 µmol/kg days 0-6),
or a combination of the drugs (given simultaneously but separately as
indicated) on the CS response to oxazolone in sensitized BALB/c mice
(n = 10). *p <.05,
**p <.01.
|
|
Histological and Immunohistochemical Assessment of the Enhancement
of the Skin CS Reaction Induced by D7193 in BALB/c Mice.
Histological and immunohistochemical examinations were performed on ear
specimens from six normal nontreated mice, six mice sensitized to and
challenged with oxazolone, and eight mice sensitized to and challenged
with oxazolone that were also treated with 3.0 µmol/kg D7193 orally
for 7 days. Hematoxylin and eosin staining of ear sections showed that
in three animals treated with D7193 there was marked cell infiltration
and edema. In the rest of the animals in this group, moderate to low
levels of cell infiltration were observed; this effect was also seen in
sensitized, challenged, but nontreated animals (Fig.
7A). In D7193-treated mice, the number of
CD8+ cells was increased compared with similarly
sensitized and challenged, but untreated mice, with the most pronounced
difference in epidermis compared to dermis (Fig. 7, B and C,
respectively). Representative examples showing increases in epidermal
CD8+ cells in D7193-treated versus nontreated
animals are shown in Fig. 8, A and B. No
marked differences in numbers of CD4+ cells in
epidermis or in dermis were observed between the animals treated or not
treated with D7193. MHC class II+ and
ICAM-1+ keratinocytes were more frequently
observed in D7193-treated mice compared with sensitized, challenged,
and untreated animals (data not shown). There was a slight increase of
LFA-1+, ICAM-1+, and
VCAM-1+ cells in dermis in the D7193-treated compared with
untreated animals. Other examined cell surface markers were not
affected by the D7193 treatment as revealed by immunohistochemical
assessment.
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 7.
Histological and immunohistochemical evaluation of
ears obtained from normal nontreated ( ) BALB/c mice
(n = 6) and mice sensitized to and challenged with
oxazolone with ( , n = 8) or without ( ,
n = 6) treatment with D7193 3.0 µmol/kg orally
for 7 days. Hematoxylin and eosin staining (A), presence of epidermal
CD8+ cells (B), and presence of CD8+ cells (C)
in dermis. For explanation of evaluation scores see Materials
and Methods.
|
|
View larger version (87K):
[in this window]
[in a new window]
|
Fig. 8.
Immunohistochemical staining of CD8+
cells in representative examples of ear specimens from BALB/c mice
sensitized to and challenged with oxazolone with (A) or without (B)
D7193 treatment orally for 7 days. Magnification 370×.
|
|
RPA Analysis of Chemokine/Cytokine Profiles in CS Reaction Sites
24 h after Challenge in Ears from BALB/c Mice Sensitized to
Oxazolone
In ear tissue obtained from BALB/c mice
sensitized to oxazolone, several chemokines/cytokines were found to be
induced 24 h after challenge (Fig. 9
A and B). Thus, IL-4, IL-10, IL-13, IL-6, lymphotactin, RANTES,
MIP-1, MIP-1, MIP-2, MCP-1, and low levels of IFN- can be
detected in the two groups of mice sensitized to and challenged with
oxazolone. IL-5, IL-9, and TCA-3 could not be detected. Expression of
eotaxin, IL-2, and IL-15 was detected in tissue from control animals as
well as in tissue from sensitized animals. There were no significant
changes in chemokine/cytokine expression pattern between mice that were
treated with D7193 or with vehicle whether or not they were sensitized to and challenged with oxazolone.
View larger version (63K):
[in this window]
[in a new window]
|
Fig. 9.
Multi probe RNase protection on total RNA from whole
mouse ears from nonsensitized mice treated with vehicle or D7193, or
mice sensitized to and challenged with oxazolone which were also
treated with vehicle or D7193. Flanking lanes contain undigested probe.
For experimental details see Materials and Methods.
|
|
Structure Activity Studies and Effects of Reference Compounds.
The effects of the disulfides ADA 202-718 and HEDS, previously
reported to act as immunostimulants (Hiestand and Strasser, 1985a,b)
and that of the immunomodulator DDTC, as well as those of some other
compounds closely related in structure to DiNAC, were also examined in
the oxazolone-induced CS reaction in BALB/c mice. Effects were
expressed relative to those of DiNAC determined on the same test
occasions. The data for the three compounds mentioned (Table
1) show that DiNAC is as effective as
DDTC and ADA 202-718 and slightly more effective than HEDS. Effects
similar to those of DDTC were also recorded with its disulfide dimer
disulfiram (data not shown). A lack of effect (when examined at 0.03 or
3 µmol/kg/day) was recorded for some compounds closely related to DiNAC, e.g., L-homocystine,
N,N'-diacetyl-L-homocystine,
cystamine, N,N-diacetyl-L-cystathionine,
and N,N'-diacetyl lanthionine, whereas a
borderline activity was recorded for
N,N'-diacetyl-L-cystamine. A low, albeit significant, effect was recorded for the
D-form of the DiNAC, i.e.,
N,N'-diacetyl-D-cystine
(efficacy reduced to 50% of that of D7193 at 0.03 as well as 3 µmol/kg/day).
View this table:
[in this window]
[in a new window]
|
TABLE 1
Effects of selected compounds on the CS reaction in BALB/c mice
Figures given are effects (T24-T0 or
T48-T0 each based on 10 individual animals) as ratios
to those of DiNAC determined at the same test occasion (i.e., percent
increase of examined compound divided by percent increase of D7193).
|
|
Effects of D7193 on CS Reactions to Oxazolone in Rats and
Rabbits.
The effect of D7193 treatment on oxazolone-induced CS
reactions was also examined in Sprague-Dawley rats and in New Zealand White rabbits. CS responses in Sprague-Dawley rats were augmented by
treatment with DiNAC according to a similar dose-response relationship as that recorded in mice (results not shown). A similar situation was
observed in outbred New Zealand White rabbits, although a tendency to a
bell-shaped dose-response curve with slightly reduced responses at the
highest dose when compared to those at lower doses was observed in this
species (results not shown).
Effects of D7193 on CS Reactions Induced by FITC and DNFB in BALB/c
Mice.
In BALB/c mice, FITC induces a TH2 type response, whereas
DNFB induces a TH1 type response (Tang et al., 1996). Mice were sensitized to FITC and to DNFB, and challenges were performed with the
corresponding agent in groups of animals treated with different daily
oral doses of D7193 or the corresponding vehicle. Interestingly, in
three of three experiments, D7193 enhanced the response to FITC in
FITC-sensitized animals in a dose-dependent manner but markedly reduced
the response to DNFB in DNFB-sensitized animals (results from
representative experiments are shown in Fig.
10, A and B, respectively). DiNAC is
less potent in augmenting the CS reaction in the FITC system (Fig. 10A)
than in the oxazolone system (Fig. 1). Similar effects were seen when
Sprague-Dawley rats were sensitized to and challenged with DNFB or FITC
with D7193 (results not shown).
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 10.
The effect of treatment with D7193 on CS responses
24 h after challenge to FITC (A) and DNFB (B) in BALB/c mice
sensitized to FITC and DNFB, respectively, on day 0. Treatment was
performed by gavage once daily from day 0 to day 6, challenge was
performed on day 7, and the CS reaction was measured on day 8. Results
are shown from one representative experiment of three performed for
each sensitizer (n = 10). *p
<.05, **p <.01, ***p <.001.
|
|
Effects of D7193 on Footpads DTH Reactions Induced by mBSA in
Mice.
Treatment with D7193 reduces footpads'
DTH reactions induced by mBSA in mBSA-sensitized mice in a
dose-dependent manner (Fig. 11).
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 11.
The effect of treatment with D7193 on DTH responses
(footpad swelling) to mBSA in BALB/c mice. Treatment was performed by
gavage once daily from day 0 to day 6. Challenge was performed day 7 and the DTH reaction was measured day 8 (24 h postchallenge) as
described in Materials and Methods.
n = 10, *p <.05.
|
|
Effects of D7193 Treatment on a DTH Granuloma Reaction Induced by
mBSA.
A DTH reaction to mBSA induced by the soaked filter paper
method (Dunn et al., 1989) results in a quantifiable chronic
granulomatous lesion. Treatment with 3 µmol/kg D7193 from days 0 to
21 reduced the development of this chronic cell-mediated inflammatory
reaction (Fig. 12); treatment with 0.03 µmol/kg did not have a effect (data not shown). Treatment during days
0 to 6 did not influence the DTH granuloma reaction, whereas treatment
during days 21 to 28 was as effective as treatment during days 0 to 21 (results not shown).
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 12.
The effect of treatment with D7193 on development of
a granulomatous lesion to mBSA in BALB/c mice. Mice were sensitized to
mBSA together with Freund's complete adjuvant. The animals were
challenged 3 weeks later (day 21) by s.c. implantation of Millipore
filters soaked in mBSA solution. After 1 week (day 28), lesions were
dissected away from the Millipore filter implants site, dried, and
weighed (wet/dry) for quantification of inflammation. Treatment with
D7193 (3.0 µmol/kg) was performed by gavage once daily from day 0 to
day 21. n = 10, ##p
<.05 (compared to no sensitization and no challenge),
**p <.05 (compared to vehicle treatment).
|
|
| |
Discussion |
The present data indicate that DiNAC modulates CS and DTH
responses in rodents. The cellular and molecular targets of DiNAC are
not defined. Their identification is hampered by the lack of effect of
the compound at relevant concentrations in in vitro systems hitherto
examined (results not shown). The reason for this is not clarified;
possible explanations include: 1) that the appropriate in vitro system
has not yet been examined or 2) that D7193 acts as an oxidant (see
below) and that the high oxygen tension in vitro compared to that in
vivo perhaps masks effects of DiNAC in the former situation.
Does DiNAC Act as an Immunomodulator?
A CS reaction resulting
from challenge of appropriately sensitized rodents reflects a response
mediated by antigen-presenting cells and T cell subsets (Askenase,
1992; Grabbe and Schwarz, 1998). Antigen presentation in the skin is
largely performed by Langerhans cells; early-acting CS-initiating T
cells and mast cells are also required in the afferent phase of the
response. Removal of epidermal Langerhans cells by steroid application
enhances the effector phase, suggesting that these cells
may exert a down-regulatory rather than a stimulatory role in the
efferent phase (see Grabbe and Schwarz, 1998). A compound that affects
only Langerhans cells would thus be expected to influence early and
late phases of the CS reaction in opposite directions. The stimulatory
effect observed with D7193, whether treatment is performed early or
late, would suggest that it does not act solely on Langerhans cells.
With respect to T cells, DTH responses to protein antigens are mediated
mainly by CD4+ cells, whereas CS reactions rely
on CD8+ effector cells (Grabbe and Schwarz,
1998). The present observation that D7193 increases the number of
CD8+ cells (but does not affect
CD4+ cells) in oxazolone-induced CS reactions
suggests that CD8+ T cells are direct or indirect
targets of the compound.
The response to oxazolone in BALB/c mice is characterized by 1) IgG2a
but not IgE antibody production and 2) IL-2 and IFN- production but
low levels of IL-4 and IL-10 generation by mitogen-stimulated lymph
node cells (Dearman et al., 1995; Dieli et al., 1997) suggesting a TH1
type of response to this sensitizer. In contrast, with respect to its
response to Leishmania, BALB/c mice are considered to have a
TH2 phenotype (Bogdan et al., 1993). The relative effect of D7193 in
oxazolone-sensitized mice is neither more nor less expressed in the
BALB/c strain than in other strains, which, in contrast to BALB/c based
on their responses to Leishmania, represent primarily the
TH1 phenotype mice (Bogdan et al., 1993). However, using the observation that FITC induces a TH2 response in BALB/c mice whereas DNFB induces a TH1 response (Tang et al., 1996), we observed that D7193
augments the CS reaction in the former system, but reduces it in the
latter. This finding suggests that the compound may influence TH1- and
TH2-mediated responses differentially in different systems. Although
additional experiments are needed to clarify the effect of D7193 in the
context of a TH1/TH2 concept, the contrasting effects of D7193 in the
FITC and DNFB systems are unlikely to be explained by nonimmunological effects.
Does D7193 Act on Cellular Adhesion or Homing?
The development
of optimal CS responses relies on several types of adhesion molecules
(see Grabbe and Schwarz, 1998). For example, blocking LFA-1 - ICAM-1
interactions during the afferent phase induces a state of
antigen-specific nonresponsiveness (Scheynius et al., 1996), and CS
reactions to DNCB are impaired in mice deficient in ICAM-1 (Sligh et
al., 1993). Although expression of selectins was not examined in the
present experiments, there was only a slight increase in
ICAM-1+ and LFA-1+ cells in ear specimens from
D7193 versus vehicle-treated animals. This suggests that D7193 does not
primarily influence the LFA-1 - ICAM-1 interaction.
The type I transmembrane glycoprotein CD44 supports extravasation of
circulating lymphocytes into lymphoid organs and is necessary for
optimal CS responses (Camp et al., 1993). The present data do not
reveal any change in CD44 expression with D7193 treatment.
Does D7193 Treatment Influence the Expression of
Chemokines/Cytokines?
Although there is a nonspecific
up-regulation of TNF- and IFN- in tissue exposed to irritants as
compared with a more specific increase in IL-1, MIP-2, IP-10, and
MHC class II signals early in the afferent phase of allergen-specific
CS reactions (Enk and Katz, 1992), TNF- and IFN- are main
effector cytokines in the latter (Gautam et al., 1994; Grabbe and
Schwarz, 1998). Previous data also suggest that IL-12 drives the CS
reaction, whereas IL-10 reduces it (see Grabbe and Schwarz, 1998). The
role of IL-4 apparently differs with the CS system examined; IL-4 does
not influence the CS response to oxazolone but apparently augments that
to picryl chloride (see Thomson et al., 1993; Asherson et al., 1996;
Grabbe and Schwarz, 1998). The present experiments detected expression of RANTES, MIP-1, MIP-1, MIP-2, MCP-1, IL-4, IL-6, and IL-10 in
ear tissue from animals sensitized to and challenged with oxazolone. However, mRNA expression levels for none of the examined
chemokine/cytokines were altered by D7193 treatment.
Precedents for the Activity of D7193?
Low molecular thiols
enhance various forms of the immune response in vivo and in vitro but
may also reduce such responses. Disulfides like HEDS and ADA 202-718
augment allogenic responses and IFN- production in mixed lymphocyte
reactions, and potentiate primary and secondary humoral immune
responses in vivo (Hiestand and Strasser, 1985a,b; Kinscherf et al.,
1994). Bell-shaped concentration-response relations in some tests for
ADA 202-718 suggest a complex mode of action (Hiestand and Strasser,
1985b). Thus, the effect of DiNAC in the CS reaction has functional
precedents. A stringent structure-activity relation was disclosed in
the oxazolone system as exemplified by 1) a lack of effect recorded for
some compounds closely related to DiNAC, e.g.,
L-homocystine,
N,N'-diacetyl-L-homocystine, cystamine,
N,N'-diacetyl-L-cystathionine,
and N,N'-diacetyl lanthionine, 2) a
borderline activity recorded for
N,N'-diacetyl-L-cystamine, and 3) a low effect recorded for the D form of the DiNAC. These results, as well as the dose relation for D7193 compared with that of
NAC (Fig. 1) showing that D7193 hardly acts as a prodrug of the latter,
and the potency of D7193, together suggest that DiNAC acts as an
oxidant at stereochemically defined site(s) of some specific target protein(s).
Possible Molecular Targets of D7193?
There are several
possible targets of DiNAC, some of which are presently being examined.
Thus, DiNAC may interfere with oxidoreductases like thioredoxin,
glutaredoxin, or protein disulfide isomerase or their corresponding
oxidoreductase reductases. The importance of these systems for immune
responses in general is underlined by findings that thioredoxin
modulates the production of a number of cytokines in vitro (Schenk et
al., 1996), is a growth factor for T cells inducing expression of the
-chain of the IL-2 receptor (Tagaya et al., 1989), and modulates
activities of transcription factors such as AP-1, NF-B, and TCF-1
(Schreck et al., 1991; Dröge et al., 1994; Schenk et al., 1996).
In this context, it is interesting to note that the contact sensitizer
1-chloro-2,4-dinitrobenzene (DNCB), a glutathione-depleting agent which
also affects T cell signal-transduction pathways (Kavanagh et al.,
1993), is an effective inhibitor of thioredoxin reductase (Arner et
al., 1995). However, the contrasting effects of D7193 on the
DNFB-induced CS reaction (inhibition) and on the BSO-enhanced CS
reactivity to oxazolone (further enhancement) suggest that influence on
glutathione levels is not the single major effect mechanism of D7193.
The inducible NO synthase inhibitor AMG (Tracey et al., 1995) reduces
autoimmune manifestations in an EAE model and induces a shift in
cytokine profiles from TH1 dominance to a TH2 type (Brenner et al.,
1997). The reduction of the oxazolone-induced CS reaction resulting
from AMG treatment, and its reversal by simultaneous treatment with
D7193, would suggest that D7193 may mimic the action of NO, i.e., by
acting as an oxidizing agent. A nonselective inhibitor of NO synthase,
L-NAME, did not affect the oxazolone-induced reaction and
did not affect the D7193-induced enhancement of the oxazolone CS
reaction. More detailed studies are needed to clarify the basis for
these findings.
Disulfides like HEDS and ADA 202-718 restore a depressed DTH response
in CSA-treated mice (Hiestand and Strasser, 1985b; St Georgiev, 1988).
We observed that the D7193-induced enhancement of the CS reaction to
oxazolone was reduced by simultaneous CSA or rapamycin treatment.
Because the CSA-induced inhibition of calcineurin leads to a block in T
cell IL-2 production whereas rapamycin-induced immune suppression
targets IL-2-induced T cell activation, the present findings suggest
that neither IL-2 production nor IL-2 action are sole targets of D7193.
As will be reported elsewhere, D7193 at appropriate doses in vivo also
markedly prolongs the lifespan of MRL-lpr/lpr mice (B.S. et
al., unpublished data), reduces development of atherosclerosis in
rabbit and mice model systems (K. Pettersson et al., unpublished data),
and reduces early and blocks late airway reactions in
allergen-challenged sheep (B. Abraham et al., unpublished data).
The immunomodulatory activity of D7193 in vivo is thus not restricted
to the CS reaction.
In conclusion, DiNAC is a potent and effective enhancer of the CS
reaction to oxazolone in rodents. In contrast, it reduces the CS
reaction to DNFB and the DTH reaction to mBSA. From the present data,
it may also be relevant to ask whether the immunological effects
attributed to NAC in vivo may be due, at least partly, to the activity
of a contaminating disulfide dimer molecule, DiNAC, with oxidant activity.
ADA 202-718, ethylene-2,2'-bis(dithio)-bis(ethanol);
AMG, aminoguanidine;
BSO, L-buthionine-[S, R]-sulfoximine;
CS, contact
(hyper)sensitivity;
CSA, cyclosporin A;
DDTC, diethyldithiocarbamate;
DiNAC, N,N'-diacetyl-L-cystine;
DNFB, 2,4-dinitrofluorobenzene;
DTH, delayed type hypersensitivity;
FITC, fluorescein isothiocyanate;
HEDS, bis-(2-hydroxyethyl)-disulfide;
mAbs, monoclonal antibodies;
mBSA, methylated BSA;
L-NAME, N-nitro-L-arginine methyl
ester;
NAC, N-acetyl-L-cysteine;
NO, nitric
oxide;
Oxazolone, 4-ethoxymethylene-2-phenyl-2-oxazolin-5-one;
RPA, RNase protection assay.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
the
Disulfide Dimer of N-acetylcysteine
Top
Abstract
Introduction
+ T lymphocytes and IL-2-producing CD4+
