Parasite Immunology,2004, 26, 273–281
Blackwell Publishing, Ltd.
ORIGINAL ARTICLE
iNOS and nitrotyrosine in toxocaral hepatitis
Enhanced inducible nitric oxide synthase expression and nitrotyrosine
accumulation in experimental granulomatous hepatitis caused by
Toxocara canis
in mice
C.-K. FAN,1 Y.-H. LIN,2 C.-C. HUNG,3,4 S.-F. CHANG5 & K.-E. SU3
1Department of Parasitology, College of Medicine, Taipei Medical University, Taipei, Taiwan, ROC, 2Department of Pathology, College of
Medicine, Taipei Medical University, Taipei, Taiwan, ROC, 3Department of Parasitology, National Taiwan University College of Medicine,
Taipei 100, Taiwan, ROC, 4Division of Infectious Diseases, Department of Internal Medicine, National Taiwan University Hospital and
National Taiwan University College of Medicine Taipei, Taiwan, ROC and 5Graduate Institute of Cell and Molecular Biology, College of
Medicine, Taipei Medical University, Taipei, Taiwan, ROC
SUMMARY
The involvement of inducible nitric oxide synthase (iNOS) and nitrotyrosine (NT) in pathogenesis of toxocaral granulo-matous hepatitis (TGH) in a murine host was quantitatively determined by biochemical, parasitological, pathological, and immunohistochemical assessments in a 42-week investigation. Mice were sacrificed for serum collection and histological processing as well as acid–pepsin digestion of the liver in a larval recovery study. Significantly increased levels of total serum NO were found in the trial, indirectly suggesting iNOS activation in the liver. iNOS reactivity was predominantly observed in infiltrating leucocytes in lesions and normal and apocrine-like cholangiocytes; in contrast, hepatocytes and multinucleated giant cells showed negative cytoplasmic staining in TGH. Strong iNOS-like reactivity was also detected on the body wall of larvae. The locations of NT reactivity were nearly identical to those of iNOS expression; infiltrating leucocytes or cholangiocytes stained for iNOS were also stained for NT in TGH. Enhanced
iNOS expression, but not invading larvae (r = 0·256, P =
0·211), seemed to play a certain role in pathological damage in TGH due to a significant correlation between iNOS
expres-sion and serum alanine aminotransferase (ALT) levels (r =
0·593, P = 0·021) in the trial. Our present results indicate a
potential therapeutic strategy for treatment of GH caused by other nematodes through manipulation of iNOS expression.
Keywords granulomatous hepatitis, iNOS, mouse, NT,
Toxocara canis
INTRODUCTION
Humans acquire Toxocara canis infection by ingestion of embryonated eggs containing infective larvae. Major clinical consequences of prolonged migration of T. canis larvae in humans are visceral larva migrans (VLM) and ocular toxo-cariasis (OT) (1). Among visceral organs, the liver is the most often affected. When larvae gain access to the portal venous circulation and actively move through the liver, they leave behind a trail of tissue disorganization caused by granulomatous inflammation that may lead to granulomatous hepatitis (2). In the past 10 years, many cases of toxocaral granulomatous hepatitis (TGH) characterized with sympto-matic signs of fever, chills, abdominal pain, jaundice, advanced fibrosis, and bile duct destruction have been reported (3 – 8). In murine models, it has been widely recognized that the immune response of experimental toxocariasis is predomin-ately a Th2-type response (9 –11). Recently, somatic T. canis L2 antigens were reported to be capable of stimulating the production of nitric oxide (NO) by rat alveolar macrophages through up-regulation of inducible NO synthase (iNOS)
in vitro. Inhibition of the synthesis of NO triggered by iNOS
possibly lessened injury to the lungs of the infected rats as assessed by pathological studies (12,13); however, mech-anisms concerning how iNOS and /or NO exert their effects on tissue damage in vivo remained unclear. Nevertheless, it appeared that Th1-associated iNOS and NO might play a role in protection and /or pathogenesis of Th2-associated murine toxocariasis.
In mammalian cells, NO is formed from -arginine by the enzyme NO synthase (NOS, EC1·14·13·39). The family of NOS isoforms falls into two dominant categories, including a constitutive low-output form and a cytokine-inducible high-output form (iNOS). It has been recognized that CD4+ Th1-associated cytokines and bacterial products such as Correspondence: Kua-Eyre Su, Department of Parasitology,
National Taiwan University College of Medicine, 1 Jen-Ai Road, Section 1, Taipei 100, Taiwan, ROC (e-mail:
C.-K. Fan et al. Parasite Immunology
lipopolysaccharide (LPS) can activate iNOS to produce high levels of NO that may play a critical role in host defence or pathological damage to host tissues (14). Many studies have analysed the protective role of NO in host defence against Th1-inducing protozoan infections, e.g.
Plasmodium falciparum (14,15), as well as Th2-stimulating
helminthic infections, e.g. Schistosoma japonicum (16 –18). On the other hand, during inflammation, NO may cause increased nitrosative stress which leads to cytotoxic activity
in vivo. This is due to the possible rapid reaction of NO and
superoxide (SOX), which yields peroxynitrite (PT) (19), a potent oxidant that causes nitration and hydroxylation of tyrosine and tryptophan as well as DNA injury through yielding nitrotyrosine (NT ) as a stable end product after their interaction with cellular proteins (20,21). Thus, evidence of the in vivo generation and biological activity of PT can be shown by the presence of NT within cells from injured tissues (22). In other words, the detection of NT may serve as an indicator of reactive nitrogen intermediates (RNIs) generated from NO. It has been well documented that production of NO via iNOS up-regulation, especially PT, is implicated in various Th1-associated bacterial, viral, and autoimmune forms of hepatitis (23,24). Nevertheless, little is known concerning the role of the involvement of iNOS, NO, and NT in the pathogenesis of Th2-associated TGH.
The present study was conducted to investigate whether iNOS and /or NT expression may be involved in pathological damage in experimental TGH as determined by biochemical, parasitological, pathological, and immunohistochemical assessments from mice infected with T. canis.
MATERIALS AND METHODS
Parasites and the experiment protocol
Toxocara canis eggs were obtained from adult female worms,
and embryonated eggs were prepared as described elsewhere (11,25). Female ICR mice aged 6 – 8 weeks were obtained from the Centre for Experimental Animals, Academia Sinica, Taipei, Taiwan. Mice were housed in the animal facility of Taipei Medical University and maintained on commercial pellet food and water ad libitum. Viability of the T. canis embryonated eggs was assessed by the light stimulation method before use (25). Each mouse was infected with about
250 T. canis embryonated eggs in 100 µL of water by oral
intubation.
Infected mice were deeply anaesthetized with ether and killed by heart puncture at 1, 3, and 5 days and 1, 4, 8, 12, 16, 20, 24, 28, and 42 weeks post-infection (dpi or wpi), respectively. On each date, three infected mice and two age-matched uninfected mice were sacrificed for serum collection and liver processing. Serum was tested for the presence of
total nitrite derived from NO and alanine aminotransferase (ALT). Half of the liver from each mouse was processed for histological and immunohistochemical studies, the other half went through acid–pepsin digestion for larva recovery assay. All animal experiments were carried out in accordance with institutional Policies and Guidelines for the Care and
Use of Laboratory Animals, and all efforts were made to
minimize animal suffering.
Serum nitrite and ALT determinations
After reducing the nitrate to nitrite with nitrate reductase, nitrite concentrations were indirectly measured by a quanti-tative colorimetric assay using the Griess reagent system (Promega, Madison, WI, USA). This method had a sensi-tivity of up to 2·5 µ if low nitrite was used. Briefly, 50 µL of freshly prepared Griess reagent (1% sulphanilamide, 0·1% naphthylethylene diamide dihydrochloride, and 2·5% ortho-phosphoric acid) was added to 50-µL aliquots of serum and incubated at room temperature. Triplicate tests were run on each test sample. After 10 min, the absorbance at 540 nm was determined in individual wells with an automated spec-trophotometer (EIA reader model EL312e, Bio-Tek, Winooski, VT, USA). The nitrite concentration in samples was calcu-lated with reference to a sodium nitrite standard curve.
Serum ALT levels as an index of hepatocellular injury (26) of each mouse were measured using a serum multiple analyser (Johnson and Johnson, Ektachem DTSC II multi-analyser, New York, NY, USA).
Larva recovery
Larvae in the liver were recovered by the method of Bardon
et al. (27 ), with modifications. Briefly, half of the liver from
each infected mouse was individually ground in a Waring blender (Tatung, Taipei, Taiwan). Each sample was digested in 50 mL of a pepsin / HCl solution (pH 1–2, 10 000 IU, Sigma, Steinheim, Germany) for 3 h at 37°C. After the addi-tion of water, the digest was centrifuged (250 g, 10 min), and the larvae in the sediment were counted in a Petri dish at 100 × magnifications under an inverted microscope (Olympus, Tokyo, Japan).
Histopathology
The other half of the liver from each mouse was fixed separately in 10% neutral buffered formalin for at least 24 h and embedded in paraffin for pathological and immuno-histochemical studies. Liver sections, 5-µm in thickness, were processed by routine procedures and stained with haematoxylin–eosin for histological study. Pathological changes were examined under a light microscope.
Volume 26, Number 6/7, June/July 2004 iNOS and nitrotyrosine in toxocaral hepatitis
Immunohistochemical analysis
Liver sections were deparaffinized and rehydrated through descending ethanol gradients before further processing. Methods for immunohistochemical detection of iNOS and NT were as described by Morikawa et al. (28), with modifi-cations. Briefly, endogenous peroxidase activity was blocked by 3% hydrogen peroxide (Merck, Taufkirchen, Germany). The sections were submerged in 10 m sodium citrate buffer at pH 6·0, and heated in an 830-W microwave oven (Sun-pentown, Ciba, Japan) for at least 5 min to retrieve the antigens. To reduce background staining, an avidin / biotin blocking kit (SP2001, Vector, Burlingame, CA, USA) was employed to block endogenous avidin / biotin activity in liver tissue. To eliminate non-specific staining, Fc receptors were blocked with diluted normal goat serum (X0907, Dako, Carpinteria, CA, USA) for 30 min at room temperature in a humid chamber. Sections were then incubated for at least 12 h at 4°C with rabbit mouse iNOS polyclonal anti-body (catalogue no. 482728, EMD Biosciences, San Diego, CA, USA) or rabbit anti-mouse NT polyclonal antibody (catalogue no. 06284, Upstate, Lake Placid, NY, USA) diluted in phosphate-buffered saline (at 1 : 500 and 1 : 200, respectively); thereafter, sections were washed with 0·05% Tween 20 –Tris-HCl buffer three times for 5 min each. A set of immunohistochemical detection kits (K4003, Dako) was employed to detect iNOS- and NT-expressing cells, respectively, by incubating with the horseradish peroxidase-conjugated goat anti-rabbit antibody for 40 min at room temperature. The presence of peroxidase was detected with the chro-mogen 3,3-diaminobenzidine (DAB) ( K3468, Dako), which resulted in a brown colour. Sections were counterstained with Gill’s haematoxylin (H3401, Vector), dehydrated, and mounted with mounting medium ( H5000, Neomarkers, Fremont, CA, USA). In order to confirm the validity of the staining results, a human tonsil section was used as a itive control. To ascertain the specificity of the staining, pos-itive control sections were treated as above with omission of the primary antibody. In addition, human tonsil sections using normal rabbit serum as the primary antibody served as negative controls.
Quantification of iNOS- or NT-expressing cells by computerized image analysis
Images for analysis were captured using a digital camera ( Nikon, Coolpix 5000, Tokyo, Japan). The mean percentages (%) of immunoreactive cells to anti-iNOS or -NT antibodies on each specimen slide were assessed microscopically under a high-power field (HPF) at 400 × magnification by totally counting 30 – 45 fields of areas containing infiltrate per experimental group of infected mice or control group of
uninfected mice using an optical image analyser (Image-Pro Plus 4·5, Media Cybernetics, Silver Spring, MD, USA). Values were expressed as the means of immunoreactive cells present in 30 – 45 HPF areas ± the standard deviation of the mean (SD).
Statistical analysis
All data were processed using a statistical software system (SPSS, Chicago, IL, USA). For calculation of the signific-ance of differences between iNOS, NT expression, serum ALT, nitrite concentration, and larva recovery, Pearson’s coefficient (r) of correlation and Student’s t-test were used. For all statistical analysis; a P-value of < 0·05 was considered statistically significant.
RESULTS
Larva recovery from the liver of infected mice
The mean number of larvae recovered from the liver increased with time in the first 5 days of infection; it started from 8·2 ± 1·3 at 1 dpi to 17·2 ± 6·7 at 3 dpi, and reached a peak of 23·4 ± 6·4 at 5 dpi. It dropped drastically to 10·0 ± 1·9 at 1 wpi, was further reduced by half at 4 wpi, then fluctuated between 1·2 ± 0·4 and 0·6 ± 0·9 from 12 wpi onwards, till the experiment ended at 42 wpi ( Figure 1).
Serum nitrite levels
The serum nitrite level in infected mice at 1 dpi (87·1 ± 11·5 µ) was triple that of the uninfected control group (25·5 ± 2·3 µ, n = 24). It quickly reached a peak of 125·7 ± 8·6 µ at 3 dpi, and then decreased drastically to around 40 µ at 5 dpi. However, the levels remained fairly high (ranging from 53·7 ± 14·9 to 31·9 ± 2·3 µ) through the remaining sampling dates (Figure 1).
Hepatic injury assessed by serum ALT levels and pathological changes
Mean serum ALT level was unchanged at 1 dpi (21·7 ± 2·5 IU / L) as compared to that of the control group (24·6 ± 6·4 IU / L); it soon reached a peak (87·7 ± 21·3 IU / L) at 3 dpi. Similar to the situation of serum nitrite level, a drastic drop also occurred at 5 dpi (42·3 ± 11·8 IU / L). Thereafter, it fluctuated between 37·7 ± 11·9 IU / L and 54·4 ± 20·9 IU / L during 1–16 wpi, and then went back to normal between 20 and 24 wpi. However, it rebounded at 28 (60·3 ± 15·9 IU / L) and 42 wpi (70·0 ± 20·3 IU / L), respectively ( Figure 1).
The liver sections of uninfected and 1-day infected mice were normal. Granulomatous inflammatory response in the
C.-K. Fan et al. Parasite Immunology
hepatic portal areas and /or parenchyma with apparent infil-tration of leucocytes, including polymorphonuclear (PMN) granulocytes and lymphocytes, was observed at 3 dpi and 1 wpi. From 4 wpi onwards, the granulomatous lesions increased in size and were infiltrated by leucocytes and multinucleated giant cells, and the lesions gradually developed into an organized granuloma ( Figure 2a,b). With respect to the bile duct, intensive periportal infiltration by inflammatory cells (Figure 2c) was frequently found from 8 wpi onwards. Apocrine-like changes in bile duct epithelial cells (cholangiocytes) (Figure 2d) were also frequently (11 out of 15) observed in infected mice from 8 wpi onwards.
iNOS- and NT-expressing cells in the liver
Localization of iNOS reactivity revealed that most of the infiltrating inflammatory cells, primarily PMN granulocytes, whether amid the inflammatory areas ( Figure 3a) or at the rim of granuloma ( Figure 3b), were positive for iNOS during the entire period of the experiment, whereas multinucleated giant cells were negative for iNOS (Figure 3a). Interestingly, strong iNOS reactivity was also detected on the body wall of the larvae (Figure 3b). A positive signal of iNOS in normal (Figure 3c) and apocrine-like cholangiocytes ( Figure 3d) was also observed. Hepatocytes in normal parenchyma or inflamed lesions were always negative for iNOS either in the infected (Figure 3a–c) or uninfected mice. Also, negative controls did not show any signal of iNOS.
Quantitatively, only 2·1 ± 0·9% of the inflammatory cells expressed iNOS in control mice. The percentage of iNOS-positive inflammatory cells already increased significantly at 1 dpi (3·9% ± 1·3%, P < 0·05). It then increased drastically
and reached a peak of 37·3 ± 10·6% (P < 0·05) by 3 dpi. Thereafter the mean percentage of iNOS-expressing inflam-matory cells dropped to about half the peak value at 5 dpi and remained elevated (P < 0·05), ranging from 21·2 ± 3·1% to 6·0 ± 2·8%, till the end of the trial ( Figure 4).
Locations of NT reactivity were almost identical to those of iNOS expression; infiltrating inflammatory cells or cholan-giocytes were positive for both iNOS and NT (Figure 5a– d). However, NT staining was not found in The larvae per se but was intensively distributed in hepatic parenchyma close to the larvae (Figure 5b). Hepatocytes in normal parenchyma as well as in inflamed lesions were negative for NT either in the infected (Figure 5a– c) or control group of uninfected mice in the trial. Also, negative controls showed no NT signal.
Quantitatively, only 0·6 ± 0·3% of cells in liver sections were positive for NT in the control group. Not much change occurred at 1 dpi (1·0 ± 0·6%); the mean percentage of NT positive cells increased about 10-fold at 3 dpi (10·0 ± 4·3%, P < 0·05) and remained relatively high until 20 wpi (ranging from 5·9 ± 2·2% to 16·3 ± 4·9%). Thereafter, it decreased gradually but remained significantly elevated till the end of the experiment (Figure 4).
Statistical correlation analysis
Correlation analysis revealed no significant correlation between the number of recovered larvae and serum ALT levels (r = 0·256, P = 0·211), whereas significant correlation existed between iNOS expression and the following three parameters, i.e. serum ALT levels (r = 0·593, P = 0·021), serum nitrite levels (r = 0·521, P = 0·041), and NT expression (r = 0·551, P = 0·032).
Figure 1 Chronological changes in number of larvae recovered from the liver, and serum nitrite (µM) and alanine aminotransferase (IU/L) concentrations, in mice infected with Toxocara canis from day 1 to week 42 post-infection. Data are presented as mean ± SD. *,+P < 0·05 compared to respective control
Volume 26, Number 6/7, June/July 2004 iNOS and nitrotyrosine in toxocaral hepatitis
DISCUSSION
It has been well documented that protection against Th1-inducing infections is associated with the production of NO via iNOS up-regulation, which has also been implicated in many forms of Th1-mediated hepatitis (23,24). In murine models, the immune response of experimental toxocariasis has been clearly described to be predominately a Th2 response (9,10). In a previous study, we also confirmed the Th2-dominant type of immune response in murine toxo-cariasis, judged by elevated serum IgG1 antibody titres (an indicator of the Th2 type response) (11). Brunet et al. (29) demonstrated that the combined effect of NO and the Th2 response prevented severe hepatic damage during S. mansoni infection in C57BL/6 mice. Infection in NO-inhibited mice or IL-4–/– mice leads to more severe liver damage and reduced time to death. However, the roles of iNOS and NO in the pathology of toxocariasis needs to be evaluated.
The present study revealed that serum NO was elevated in infected mice, and that iNOS was significantly induced and the amount of NT (end product of PT) increased in the livers of T. canis-infected mice throughout the trial, as com-pared with normal livers of uninfected control mice. Since NT may have originated, among other sources, from NO-mediated nitration of cellular protein involving the forma-tion of PT, these findings suggest that PT might play a role in the progression of TGH. Our results showed a marked induction of iNOS expression in leucocytes, particularly of PMN granulocytes, suggesting that these cells may be the major source of large quantities of NO. Recent studies revealed that PMN granulocytes were able to express mRNA for iNOS and to secrete NO (30,31).
The reason for the re-elevation of serum ALT levels in mice infected with T. canis for longer than 28 weeks was not clear. It may not be due to re-invasion of a large number of larvae into the liver, because hepatic larva recovery remained Figure 2 Representative pathological changes in the liver from mice infected with Toxocara canis from day 1 to week 42 post-infection (dpi or wpi). (a) Inflammatory lesions characterized with PMN granulocytes (arrow head) and multinucleated giant cells (arrow) infiltration at 12 wpi. Bar = 5 µm. The upper and lower inset shows a PMN granulocyte and a multinucleated giant cell at higher magnification, respectively. (b) Inflammatory cells in the peripheral rim of a granuloma at 24 wpi. Bar = 5 µm. The lower inset shows a larva trapped in a granuloma at 16 wpi. (c) Intensive periportal infiltration by inflammatory cells at 8 wpi. Bar = 5 µm. (d) Cholangiocytes with apocrine-like changes (arrow) at 24 wpi. Bar = 5 µm.
C.-K. Fan et al. Parasite Immunology
Figure 3 Representative pictures showing immunohistochemical localization of inducible nitric oxide synthase (iNOS) in the liver sections of mice infected with Toxocara canis from day 1 to week 42 post-infection (dpi or wpi). (a) Lack of iNOS expression in multinucleated giant cell (small arrow) and hepatocytes (big arrow) in chronic granulomatous inflammation, with inflammatory cells (arrow head) showing iNOS expression at 12 wpi. Bar = 5 µm. The lower inset shows an iNOS-positive PMN granulocyte at a higher magnification. (b) iNOS-positive inflammatory cells (arrow head) at the rim of an organized granuloma, hepatocytes (arrow) were negative for iNOS expression at 16 wpi. Bar = 10 µm. The inset shows that positive iNOS activity was found on the body wall (arrow head) of larvae. (c) Normal cholangiocytes (arrow head) showing iNOS expression, but hepatocytes (arrow) were negative for iNOS at 12 wpi. Bar = 5 µm. (d) Intense iNOS expression in cholangiocytes with apocrine-like changes (arrow) at 16 wpi. Bar = 5 µm.
Figure 4 Chronological changes in percentage of iNOS and NT expression in the liver of mice infected with Toxocara canis from day 1 to week 42 post-infection. *,+P < 0·05 compared to respective control
Volume 26, Number 6/7, June/July 2004 iNOS and nitrotyrosine in toxocaral hepatitis
low and fairly constant after 12 wpi. In addition, the numbers of infiltrating inflammatory cells expressing iNOS at both dates were not exceedingly high. Furthermore, substantial experimental evidence suggested that iNOS was capable of preventing the initiation of apoptosis through NO-mediated inactivation of caspase, thus leading to prolonged inflam-mation and /or carcinogenesis (32). Whether persistent iNOS expression in infiltrating leucocytes might be an important factor contributing to prolonged inflammation, thus leading to liver injury of TGH, warrants further investigation.
Interestingly, T. canis larvae per se seemed to have iNOS-like activity in their hypodermis and muscle as observed in Figure 3(b). Although it is unclear why this iNOS-like activity occurred in the hypodermis and muscle of T. canis larvae,
NOS activity was reported to occur in the hypodermis and muscle of adult Ascaris suum (33). Whether the iNOS-like molecule in T. canis larvae is analogous to the NOS in adult A. suum needs further investigation. Cross-reactivity between mammalian and nematode enzymes or a non-specific bind-ing effect may also be possible. It is also unclear why neither multinucleated giant cells nor hepatocytes expressed iNOS and /or NT in murine TGH. One possible explanation is that multinucleated giant cells which evolve from macrophages seem to merely play a role in phagocytic removal of unwanted debris in inflamed sites instead of in modulation of the immune response, and those hepatocytes also play no important roles in defending against T. canis larval invasion of the liver.
Figure 5 Representative pictures showing immunohistochemical localization of nitrotyrosine (NT) in the liver sections of mice infected with Toxocara canis from day 1 to week 42 post-infection (dpi or wpi). (a) Lack of NT in multinucleated giant cell (small arrow) and hepatocytes (big arrow) in chronic granulomatous inflammation, whereas PMN granulocytes (arrow head) were positive for NT at 12 wpi. Bar = 5 µm. The lower inset shows a NT-positive PMN granulocyte at a higher magnification. (b) NT-positive inflammatory cells (arrow head) at the rim of an organized granuloma, with hepatocytes (arrow) showing no NT staining at 16 wpi. Bar = 10 µm. The inset shows that positive NT staining was found in hepatic parenchyma (arrow) close to the larva (arrow head) at 16 wpi. (c) Normal cholangiocytes (arrow head) were positive but hepatocytes (arrow) were negative for NT staining at 12 wpi. Bar = 5 µm. (d) Intense NT staining in cholangiocytes (arrow head) with apocrine-like changes, but hepatocytes (arrow) were negative for NT staining at 16 wpi. Bar = 5 µm.
C.-K. Fan et al. Parasite Immunology
It is noteworthy that normal or apocrine-like cholangiocytes exhibited significant iNOS and /or NT expression in TGH, although the latter showed more intense staining than the former. However, the physiopathological role of cholangio-cytes with apocrine-like changes, as well as iNOS and /or NT expression in these cholangiocytes in TGH, is unclear. It was proposed that iNOS- and /or NT-expressing cholangiocytes might be highly susceptible to the development of biliary cir-rhosis and /or cholangiocarcinoma due to the production of NO induced by iNOS in biliary epithelial cells; this has been postulated to contribute to epithelial cell cirrhosis and /or car-cinogenesis by causing damage to DNA and proteins (34,35). Espinoza et al. (12) observed that adult worm ES and L2 somatic antigens of T. canis were able to induce the in vitro production of NO by rat alveolar macrophages. They fur-ther observed that NO did not have a cytotoxic effect on T. canis larvae in vitro, and that in vivo production of NO induced by infection with T. canis in rats resulted in lung damage, whereas inhibition of iNOS decreased lung injury (13). However, they did not identify the NO-producing cells by immunohistochemical staining.
This is the first report to confirm the in situ iNOS and NT expression in leucocytes as well as cholangiocytes with normal and apocrine-like changes in murine TGH. It seems likely that leucocytes expressing iNOS may play a certain role in the pathological damage seen in TGH. Whether iNOS- and /or NT-expressing cholangiocytes of TGH are susceptible to the development of cirrhosis or cholangio-carcinoma remains to be further investigated.
ACKNOWLEDGEMENTS
The authors are grateful to the National Science Council of the ROC and Taipei Medical University for their support of this research (grants NSC 93-2314-B-038-065, TMU-88-Y05-A106 and TMU-90-Y05-A142). We thank Mr D. J. Kao and Mr C. W. Liao for expert technical assistance; the Taipei Municipal Institute for Animal Health is acknow-ledged for providing experimental stray dogs from which adult Toxocara canis worms were collected. Also, Mr D. Cham-berlin is acknowledged for critical revision of this paper.
REFERENCES
1 Kerr-Muir MG. Toxocara canis and human health. Br Med J 1994; 309: 5 – 6.
2 Kayes SG. Human toxocariasis and the visceral larva migrans syndrome: correlative immunopathology. Chem Immunol 1997; 66: 99 –124.
3 Ponder D, Marshall GS, Rabalais GP & Wood BP. Radiological case of the month: visceral larva migrans. Am J Dis Child 1991; 145: 699 –700.
4 Bhatia V & Sarin SK. Hepatic visceral larva migrans: evolution of the lesion, diagnosis, and role of high-dose albendazole therapy. Am J Gastroenterol 1994; 89: 624 – 627.
5 Kaushik SP, Hurwitz M, McDonald C & Pavli P. Toxocara canis infection and granulomatous hepatitis. Am J Gastroenterol 1997; 92: 1223 –1225.
6 Baldisserotto M, Conchin CF, Soares MG, Araujo MA & Kramer B. Ultrasound findings in children with toxocariasis: report on 18 cases. Pediat Radiol 1999; 29: 316 –319.
7 Hartleb M & Januszewski K. Severe hepatic involvement in visceral larva migrans. Eur J Gastroenterol Hepatol 2001; 13: 1245 –1249.
8 Kaplan KJ, Goodman ZD & Ishak KG. Eosinophilic granuloma of the liver: a characteristic lesion with relationship to visceral larva migrans. Am J Surg Pathol 2001; 25: 1316 –1321. 9 Smith HV. In Parasitic Nematodes: Antigens, Membranes and
Genes, ed. Kennedy MW. London: CRC Press; 1991: 116 –139. 10 Cuellar C, Fenoy S, del-Aguila C & Guillen L. Isotype-specific immune responses in murine experimental toxocariasis. Mem Inst Oswaldo Cruz 2001; 96: 549 –553.
11 Fan CK, Lin YH, Du WY & Su KE. Infectivity and pathogenicity of 14-month-cultured embryonated eggs of Toxocara canis in mice. Vet Parasitol 2003; 113: 145 –155.
12 Espinoza EY, Muro A, Martin MM, Casanueva P & Perez-Arellano JL. Toxocara canis antigens stimulate the pro-duction of nitric oxide and prostaglandin E2 by rat alveolar macrophages. Parasite Immunol 2002; 24: 311–319.
13 Espinoza EY, Perez-Arellano JL, Carranza C, Collia F & Muro A. In vivo inhibition of inducible nitric oxide synthase decreases lung injury induced by Toxocara canis in experimentally infected rats. Parasite Immunol 2002; 24: 511–520.
14 James SL. Role of nitric oxide in parasitic infections. Microbiol Rev 1995; 59: 533 –547.
15 Oswald IP, Wynn TA, Sher A & James SL. NO as an effector molecule of parasite killing: modulation of its synthesis by cytokines. Comp Biochem Physiol 1994; 108: 11–18.
16 Kanazawa T, Asahi H, Hata H, Mochida K, Kagei N & Stadecker MJ. Arginine-dependent generation of reactive nitrogen intermediates is instrumental in the in vitro killing of protoscoleces of Echinococcus multilocularis by activated macrophages. Parasite Immunol 1993; 15: 619 – 623.
17 Rajan TV, Porte P, Yates JA, Keefer L & Shultz LD. Role of nitric oxide in host defense against an extracellular, metazoan parasite, Brugia malayi. Infect Immun 1996; 64: 3351–3353. 18 Hirata M, Hirata K, Kage M, Zhang M, Hara T & Fukuma T.
Effect of nitric oxide synthase inhibition on Schistosoma japonicum egg-induced granuloma formation in the mouse liver. Parasite Immunol 2001; 23: 281–289.
19 Ischiropoulos H, Zhu L, Chen J, et al. Peroxynitrite-mediated tyrosine nitration catalyzed by superoxide dismutase. Arch Biochem Biophys 1992; 298: 431– 437.
20 Beckman JS. Oxidative damage and tyrosine nitration from peroxynitrite. Chem Res Toxicol 1996; 9: 836 – 844.
21 Szabo C & Ohshima H. DNA damage induced by peroxynitrite: subsequent biological effects. Nitric Oxide 1997; 1: 373 –385. 22 Haddad IY, Ischiropoulos H, Holm BA, Beckman JS, Baker JR
& Matalon S. Mechanisms of peroxynitrite-induced injury to pulmonary surfactants. Am J Physiol 1993; 265: L555 –L564. 23 Majano PL, Garcia-Monzon C, Lopez-Cabrera M, et al.
Induc-ible nitric oxide synthase expression in chronic viral hepatitis. Evidence for a virus-induced gene upregulation. J Clin Invest 1998; 101: 1343 –1352.
24 Li J & Billiar TR. Nitric oxide. IV. Determinants of nitric oxide protection and toxicity in liver. Am J Physiol 1999; 276: G1069 – G1073.
Volume 26, Number 6/7, June/July 2004 iNOS and nitrotyrosine in toxocaral hepatitis
25 O’Lorcain P. The effects of freezing on the viability of Toxocara canis and T. cati embryonated eggs. J Helminthol 1995; 69: 169 –171. 26 Burt AD & Day CP. In Pathology of the Liver, eds Burt AD, Portmann BC, Ishak KG, Scheuer PJ & Anthony PP. London: Churchill Livingstone; 2002: 67–106.
27 Bardon R, Cuellar C & Guillen JL. Larval distribution of Toxocara canis in BALB/C mice at nine weeks and one year post-inoculation. J Helminthol 1994; 68: 359 –360.
28 Morikawa A, Kato Y, Sugiyama T, et al. Role of nitric oxide in lipopolysaccharide-induced hepatic injury in -galactosamine-sensitized mice as an experimental endotoxic shock model. Infect Immun 1999; 67: 1018 –1024.
29 Brunet LR, Beall M, Dunne DW & Pearce EJ. Nitric oxide and the Th2 response combine to prevent severe hepatic damage during Schistosoma mansoni infection. J Immunol 1999; 163: 4976 – 4984.
30 Gagnon C, Leblond FA & Filep JG. Peroxynitrite production by human neutrophils, monocytes and lymphocytes challenged with lipopolysaccharide. FEBS Lett 1998; 431: 107–110.
31 Brennan ML, Wu W, Fu X, et al. A tale of two controversies. defining both the role of peroxidases in nitrotyrosine formation in vivo using eosinophil peroxidase and myeloperoxidase-deficient mice, and the nature of peroxidase-generated reactive nitrogen species. J Biol Chem 2002; 277: 17415 –17427.
32 Kim YM, Talanian RV & Billiar TR. Nitric oxide inhibits apoptosis by preventing increase in caspase-3-like activity via two distinct mechanisms. J Biol Chem 1997; 272: 31138 –31148. 33 Bowman JW, Winterrowd CA, Friedman AR, et al. Nitric oxide mediates the inhibitory effects of SDPNFLRFamide, a nematode FMRFamide-related neuropeptide, in Ascaris suum. J Neurophysiol 1995; 74: 1880 –1888.
34 Battista S, Bar F, Mengozzi G, et al. Evidence of an increased nitric oxide production in primary biliary cirrhosis. Am J Gastroenterol 2001; 96: 869 – 875.
35 Jaiswal M, LaRusso NF, Burgart LJ & Gores GJ. Inflammatory cytokines induce DNA damage and inhibit DNA repair in cholangiocarcinoma cells by a nitric oxide-dependent mechanism. Cancer Res 2000; 60: 184 –190.
