Immune responses elicited by nematode parasite infections are characterized by T helper 2 (Th2) cell induction. The immunologic basis for changes in intestinal physiology accompanying nematode infection is poorly understood. This study examined whether worm expulsion and associated goblet cell hyperplasia and muscle contractility share a similar immune basis by shifting the response from Th2 to Th1 using interleukin-12 (IL-12) overexpression. We used a single administration of recombinant adenovirus vector expressing IL-12 (Ad5IL-12) in Trichinella spiralis-infected mice. Ad5IL-12 administered 1 day after infection prolonged worm survival and inhibited infection-induced muscle hypercontractility and goblet cell hyperplasia. This was correlated with upregulated interferon-γ (IFN-γ) expression and downregulated IL-13 expression in the muscularis externa layer. We also observed increased IFN-γ production and decreased IL-4 and IL-13 production from in vitro stimulated spleen and mesenteric lymph node cells of infected Ad5IL-12-treated mice. These results indicate that transfer and overexpression of the IL-12 gene during Th2-based nematode infection shifts the immune response toward Th1 and delays worm expulsion. Moreover, the immune response shift abrogated the physiological responses to infection, attenuating both muscle hypercontractility and goblet cell hyperplasia. These findings strongly indicate that worm expulsion, muscle hypercontractility, and goblet cell hyperplasia share a common immunologic basis and may be causally linked.
- Trichinella spiralis
- smooth muscle
- goblet cells
activation of the mucosal immune system of the gut during enteric infection results in altered intestinal physiology that contributes to host defense by evicting the infective agent from the gut (13). Changes include goblet cell hyperplasia and increased mucus secretion, increased ion water secretion, and enhanced intestinal propulsive activity. It is generally believed that these alterations in gut physiology are a direct result of immune activation rather than a nonspecific consequence of the mucosal inflammatory reaction to the infective agent.
Intestinal nematode infections, including Trichinella spiralis, generate a strong T helper 2 (Th2)-type response (15, 18, 25). Infection of mice or rats with T. spiralis results in intestinal muscle hypercontractility (51, 56) and goblet cell hyperplasia (21,32). A previous study from our laboratory (51) showed that mouse strains expelling the parasite rapidly (e.g., NIH Swiss) exhibit a greater degree of intestinal muscle contractility compared with that seen in mice expelling the parasite slowly, such as the B10.BR strain. It has also been shown (54) that changes in intestinal muscle function during this nematode infection are dependent on CD4+Th cells and major histocompatibility complex (MHC) class II molecules. Taken together, these observations suggest that intestinal muscle hypercontractility and worm expulsion share a common immunologic basis and may be causally related.
Infection of mice with T. spiralis also induces CD4+ T cell-mediated mucosal changes, which include intestinal goblet cell hyperplasia. Mucins from goblet cells may play an important role in the trapping and removal of intestinal worms from the gut (44). Both the quantitative changes in mucus production as well as the qualitative changes in mucins observed in nematode-infected mice and rats are T cell dependent (31, 34) and involve a Th2 response (32).
Interleukin-12 (IL-12) is a heterodimeric cytokine produced by antigen-presenting cells (primarily macrophages and B cells) and exerts immunoregulatory effects on T cells and natural killer (NK) cells (22, 50). IL-12 activates the differentiation of CD4+ Th cells toward Th1-associated responses by stimulating production of interferon-γ (IFN-γ) by Th and NK cells. In contrast, IL-12 inhibits Th2-type immune responses by suppressing T cell production of IL-4 (42). IL-12 has been shown to be an important factor in activating host protective immunity during infection by intracellular pathogens including Leishmania(39), Listeria (50), andChlamydia (29). Several studies (6,12) also suggested that IL-12 might be effective in the treatment of cancer and acquired immunodeficiency syndrome patients. In contrast, IL-12 inhibits the development of host protective immunity to nematode infections, including Trichuris muris andNippostrongylus brasiliensis infection (2, 18). Thus IL-12 is pivotal in the development of Th1 responses and in suppressing Th2 responses.
Most studies investigating the role of IL-12 have traditionally involved the systemic administration of recombinant IL-12 protein. This approach is costly and requires multiple injections due to the short half-life of the recombinant protein in vivo. The use of the recombinant adenovirus type 5 vector expressing IL-12 (Ad5IL-12) presents an attractive alternative approach, because these replication-deficient vectors are highly infectious and effectively express high levels of bioactive IL-12 both in vitro and in vivo (5). A single injection of Ad5IL-12 has been shown to trigger an effective Th1 response in BALB/c mice with Leishmania major, which usually induces a Th2-type response in this mouse strain (11).
In the present study, we investigated the effect of the overexpression of IL-12 on worm expulsion in T. spiralis-infected mice and on the accompanying changes in intestinal muscle contractility and goblet cell response. We reasoned that if the Th2 response is critical for both the successful expulsion of the parasite and the development of muscle and goblet cell changes, then suppression of the Th2 response using Ad5IL-12 should not only prolong the infection but also attenuate muscle and goblet cell changes.
We demonstrate for the first time that Ad5IL-12 effectively inhibits the development of intestinal muscle hypercontractility and goblet cell hyperplasia during T. spiralis infection and prolongs worm survival in the gut. This was accompanied by upregulation of the Th1 cytokine IFN-γ and downregulation of Th2 cytokine IL-13 in the muscularis externa of the intestine and increased IFN-γ production and decreased IL-4 and IL-13 production from in vitro stimulated spleen and mesenteric lymph node (MLN) cells.
MATERIALS AND METHODS
NIH Swiss mice were obtained from the National Cancer Institute (Frederick, MD), kept in sterilized, filter-topped cages, and fed autoclaved food in the animal facilities of McMaster University. Only 8- to 10-wk-old male mice were used. The protocols employed were in direct accordance with guidelines drafted by the McMaster University Animal Care Committee and the Canadian Council on the Use of Laboratory Animals.
T. spiralis infection.
The T. spiralis parasites used in the study originated in the Department of Zoology at the University of Toronto, and the colony was maintained through serial infections alternating between male Sprague-Dawley rats and male CD1 mice. The larvae were obtained from infected rodents 60–90 days postinfection (PI), using a modification (56) of the technique described by Castro and Fairbairn (8). For counting intestinal worm numbers, adult worms were recovered from mice after the intestine had been opened longitudinally, rinsed, and placed in Hanks' balanced salt solution for 3 h at 37°C. Worms were counted under a dissecting microscope.
Adenovirus vectors and treatment.
Construction and characterization of Ad5IL-12 have previously been described (5). Briefly, the Ad5IL-12 vector contains an expression cassette for the p35 subunit of IL-12 in the E1 region and the IL-12 p40 subunit in the E3 region. The control virus, DL70–3, is an adenovirus variant deleted in the E1 region (3). Each mouse was injected intraperitoneally with 2.5 × 108 or 5 × 108 plaque-forming units (pfu) of Ad5IL-12 or DL70–3.
Measurement of muscle contraction.
The preparation of the jejunal longitudinal muscle sections for muscle contractility experiments and the analysis of the carbachol-induced contraction have been described previously (51). Briefly, the jejunum was removed and placed in oxygenated (95% O2-5% CO2) Krebs solution, and 1-cm sections of whole gut were cut from the jejunum, beginning at the ligament of Treitz and proceeding distally. The lumen of each segment was flushed with Krebs buffer, before the insertion of short (2- to 3-mm) lengths of Silastic tubing (0.065 in. OD, 0.030 in. ID; Dow Corning, Midland, MI) into the open ends of the gut segments. Tubing was then tied in place with surgical silk. The insertion of the tubing was found to maintain patency of the gut segments over the course of the experiments. Segments were then hung in the longitudinal axis and attached at one end to a Grass FT03C force transducer (Quincy, MA), and responses were recorded on a Grass 7D polygraph. Tissues were equilibrated for 30 min at 37°C in Krebs buffer oxygenated with 95% O2-5% CO2 before starting the experiment. The previously identified optimal tension (400 mg) was then applied in carbachol dose-response experiments before the addition of the first dose of carbachol (51). Previous experiments (51) indicated that this was the optimal tension to determine the maximal responsiveness of both control and inflamed tissues. After the application of tension, gut segments were exposed to different concentrations of carbachol. After the maximal response to each dose was obtained, tissues were rinsed twice and equilibrated in fresh Krebs solution for 15 min before addition of the next dose. Contractile responses to carbachol were expressed as milligrams of tension per cross-sectional area, as described previously (51). For each mouse, the mean tension was calculated from at least three segments.
Detection of cytokines in muscle by RT-PCR.
Expression of mRNAs of IL-13 and IFN-γ in the jejunal muscle was investigated as described previously (36, 52). Briefly, after removal of the small intestine, the longitudinal muscle-myenteric plexus, including serosa, was stripped from the jejunum, beginning at the ligament of Trietz and proceeding 4 cm distally. Total cellular RNA was isolated based on a previously described guanidium isothiocyanate method (10). The concentration of RNA was determined by measuring absorbance at 260 nm, and its purity was confirmed using the ratio of absorbency at 260 nm to that at 280 nm. RNA was stored at −70°C until used for RT-PCR. mRNA was then reversed transcribed as described previously (38) to yield cDNA, and the cDNA was amplified by PCR using gene-specific primers.
Fifty-nanogram aliquots of cDNA (0.1 μg) were then mixed with 20 pmol each of the upstream primer 5′-TCT TGC TTG CCT TGG TGG TCT CGC-3′ and downstream primer 5′-GAT GGC ATT GCA ATT GGA GAT GTT G-3′ for IL-13 (38). IFN-γ was investigated using the primers 5′-CAT GGC TGT TTC TGG CTG TTA C-3′ and 5′-TCG GAT GAG CTC ATT GAA TGC-3′ as upstream and downstream primers, respectively (24). The housekeeping gene hypoxanthine phosphoribosyl transferase (HPRT) was used as the positive control, and to detect it, the upstream primer 5′-GTT GGA TAC AGG CCA GAC TTT GTT G-3′ and downstream primer 5′-GAT TCA ACT TGC GCT CAT CTT AGG C-3′ were used (48). PCR was performed in 50-μl volumes containing dNTP (200 μM), MgCl2 (1.5 mM), and 2.5 U Taq polymerase (GIBCO-BRL) with corresponding buffer and distilled water. Messages for IL-13, IFN-γ, and HPRT were coamplified using the following parameters: denaturation 94°C for 30 s, annealing at 55°C for 30 s, and extension at 72°C for 60 s. PCR products were loaded onto a 2.5% agarose gel and then visualized under ultraviolet light after ethidium bromide staining.
Production of cytokines from spleen and MLN.
Single-cell suspensions of spleen or MLN were prepared in RPMI 1640 containing 10% FCS, 5 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, 25 mM HEPES, and 0.05 mM 2-mercaptoethanol (all obtained from GIBCO-BRL). Cells (1 × 107) were incubated in the presence of 5 μg/ml of concanavalin A (ConA; Sigma). Culture supernatants were harvested after 24 h, and cytokine levels were determined.
Blood collection for IL-12 p40 measurement.
IL-12 is a 75-kDa heterodimeric cytokine composed of two disulfide-linked subunits, p35 and p40 (47). To investigate the level of IL-12 p40 in sera, we collected blood from the mouse retroorbital plexus at various time points after Ad5IL-12 administration.
IL-4, IL-13, and IFN-γ concentrations in the culture supernatants and IL-12 p40 level in sera were measured by enzyme immunoassay technique using commercially available kits purchased from R & D Systems (Minneapolis, MN).
The degree of small intestinal inflammation was monitored by assay of myeloperoxidase (MPO) activity. Samples of jejunum (50–100 mg) ∼2 cm distal to the ligament of Treitz were removed, snap frozen in liquid nitrogen, and stored at −70°C. MPO activity was measured using a previously described technique (58) and is reported as units of MPO per milligram of wet tissue. One unit of MPO is defined as the quantity of enzyme able to convert 1 μmol of hydrogen peroxide to water in 1 min at room temperature.
A segment of small intestine (1 cm in length) was taken at 10 cm from the pyloric sphincter, fixed in 10% neutral buffered Formalin (NBF), and processed using standard histological techniques. The NBF sections were stained with the periodic acid-Schiff technique for counting intestinal goblet cells. The number of goblet cells was expressed per 10 villus crypt units.
Data were analyzed using Student's t-test withP < 0.05 considered to be significant. All results are expressed as means ± SE.
In PBS-treated infected mice, almost all worms were expelled from the gut by day 14 PI. Administration of DL-70–3 preinfection or PI had no effect on worm expulsion. When Ad5-IL12 was administered 1 day before infection, >80% of worms remained in the gut by day 14 (Table 1). However, this was accompanied by a 50% mortality rate. Delayed worm expulsion was also seen when Ad5IL-12 was administered on days 1–3 PI and was maximal when the construct was given 1 day PI and decreased in magnitude in a time-dependent manner when given thereafter. In light of the mortality associated with administration of Ad5IL-12 before infection, all subsequent experiments involved administration of the construct on day 1 PI. As shown in Fig. 1, Ad5IL-12 administration delayed worm expulsion in a dose-dependent manner, and the maximal effective dose was 5 × 108 pfu, which was used in subsequent experiments.
Carbachol-stimulated muscle contraction was not altered by adding the TTX to the bath, therefore reflecting a direct action of the muscarinic agonist on smooth muscle (data not shown). As expected, in PBS-treated infected mice, intestinal muscle contractility increased significantly compared with that in noninfected controls. However, as shown in Fig.2, muscle hypercontractility was significantly attenuated in mice treated with a single injection of Ad5IL-12 during T. spiralis infection on day 10PI; carbachol-induced muscle contraction was significantly lower in mice treated with Ad5IL-12 compared with PBS- or DL70–3-treated mice. A similar attenuation was observed in Ad5IL-12-treated mice onday 7 PI (data not shown). The attenuation of muscle contractility in Ad5IL-12-treated infected mice was observed over a carbachol dose range of 0.1 mM to 1 μM (see Fig.3).
Goblet cell hyperplasia.
There was a significant increase in the number of goblet cells inT. spiralis-infected mice receiving PBS or DL70–3. As shown in Fig. 4, the development of intestinal goblet cell hyperplasia was inhibited inT. spiralis-infected mice that received Ad5IL-12, and this was evident on days 10 and 14 PI.
Expression of cytokines in muscularis externa.
We next investigated expression of cytokine genes in the muscularis externa by RT-PCR. IFN-γ and IL-13 were used as candidate Th1 and Th2 cytokines, respectively. The PCR products for IL-13 were strongly expressed in PBS- or DL70–3-treated infected mice on day 10 PI but were not detectable in uninfected control mice. Administration of Ad5IL-12 resulted in a downregulation of IL-13 expression and an upregulation of IFN-γ in infected mice (Fig.5).
In vitro cytokine production and IL-12 p40 level in serum.
Measurement of in vitro cytokine production by ConA-stimulated MLN revealed high amounts of both IL-4 and IL-13 in T. spiralis-infected mice treated with PBS or DL70–3. In contrast, the MLN production of IL-4 and IL-13 was significantly attenuated in T. spiralis-infected mice treated with Ad5IL-12 (Table 2). As expected, we documented a high level of IFN-γ production in mice that received Ad5IL-12 after T. spiralis infection. After Ad5IL-12 injection, serum IL-12 p40 levels were increased within 24 h and decreased gradually on days 6 and 10 PI (see Fig.6).
T. spiralis infection caused a significant increase in MPO activity on day 10 PI, as shown in Fig.7. There was a significant suppression of MPO activity after Ad5IL-12 treatment of infected mice. In contrast, there was no significant difference in MPO activity on day 10 PI between mice treated with control virus DL70–3 and PBS.
Previous studies have demonstrated that CD4+ Th cells are essential in host protection (15, 26) and Th2-type cells are implicated in the expulsion of worms from the nematode-infected gut (15, 19, 25). Our data confirm the critical role of the Th2 response by demonstrating that a single injection of Ad5IL-12 inhibits T. spiralis expulsion, contrasting with the multiple injections of recombinant IL-12 required to achieve this effect. The rise in serum IL-12 p40 levels within 24 h after Ad5IL-12 administration indicates that this construct is an efficient delivery vehicle for IL-12. Time course study for the administration of Ad5IL-12 revealed that the inhibition of worm expulsion was maximal when given 1 day after infection and decreased in magnitude in a time-dependent manner when given thereafter. Considering the high mortality associated with Ad5IL-12 administration before infection, we administered the construct 1 day after infection in the subsequent experiments.
We do not know the basis for the mortality associated with pretreatment of infected mice with Ad5-IL-12. No mortality was associated with administration of the control virus in T. spiralis-infected mice or with administration of Ad5IL-12 in noninfected mice. In addition, we did not observe mortality when Ad5IL-12 was given on different days PI beginning 24 h after the infection. Together, these findings exclude a systemic toxic effect of either the virus or IL-12 at the doses we used. We surmise that the administration of this vector before the infection permitted a more effective polarization to a Th1 response and a greater reduction in the protective Th2 response.
In addition to the delayed expulsion of worms, treatment with Ad5IL-12 attenuated the intestinal muscle hypercontractility that usually accompanies nematode infections, including N. brasiliensis(16), Heligosomoides polygyrus(23), and T. spiralis infections (51,56). We have reported that muscle hypercontractility observed during T. spiralis infection is, at least in part, immune mediated and attenuated in infected athymic (55) and CD4+- and MHC II-deficient mice (54). More recent work (36) has shown that muscle hypercontractility is also attenuated in mice deficient in signal transducer and activator of transcription factor 6 (Stat6), suggesting a role for Th2 cytokines. These observations, taken in conjunction with the fact that the magnitude of hypercontractility seen during T. spiralisinfection parallels the efficiency with which the host evicts the parasite from the gut (51), lead us to hypothesize that muscle hypercontractility and the process of worm expulsion are linked, most likely through a Th2 response. The expression of the Th2 cytokine IL-13 in the muscularis externa of untreated infected mice is consistent with this hypothesis. We believe that Th2 cytokines are produced by lymphocytes that invade the muscularis externa during the enteric phase of infection, as has been shown in T. spiralis-infected rats (14) and mice (54), as well as in humans with idiopathic inflammatory bowel disease (17). We believe that there is direct communication between the Th2 lymphocytes and smooth muscle cells in light of our (1, 4) preliminary findings in which exposure of isolated intestinal muscle cells to exogenous IL-4 increased the contractile response to carbachol whereas exogenous IFN-γ had no significant effect.
Evidence in support of this hypothesis is presented here with the demonstration that Ad5IL-12 administration retarded worm expulsion and attenuated the development of muscle hypercontactility in infected mice. Furthermore, this was accompanied by intense local expression of IFN-γ mRNA in the muscularis externa and downregulation of IL-13 mRNA in the tissue of AD5-IL-12-treated mice. In contrast, we observed strong expression of IL-13 mRNA in both PBS- and control virus-treated infected mice. In addition, measurement of in vitro cytokine production from MLN and spleen cells from T. spiralis-infected and Ad5IL-12-treated mice revealed an increase in IFN-γ production and decrease in IL-4 and IL-13 production. Taken together, we interpret these findings to indicate that the treatment of mice with Ad5IL-12 redirected the immune responses away from the protective Th2-type immunity toward a Th1 response, with an attendant attenuation of muscle hypercontractility.
IL-12 has been implicated in the modulation of airway smooth muscle hyperresponsiveness in asthma (28) and may influence the contractility of other cell types (7). However, our data do not support a direct role for IL-12 in the contractility of intestinal muscle. In noninfected mice, we saw no effect of Ad5IL-12 on carbachol-induced muscle contraction (Fig. 2). Furthermore, we do not believe that Ad5IL12 influenced muscle hypercontractility through the expression of IFN-γ in the tissue, because our (4) preliminary data have shown that this cytokine does not influence carbachol-induced muscle contraction. We consider the attenuation of muscle hypercontractility in Ad5IL-12 treated infected mice to reflect the downregulation of the Th2 response. The ability of Ad5IL-12 to attenuate rather than abolish the hypercontractility reflects the fact that there are both T cell-independent and -dependent components to muscle hypercontractility in this model (55) and IL-12 would be expected to affect only the latter.
Mucin production by goblet cells is considered to play an important role in host protection against intestinal nematode infection (31, 34, 44). Putative mechanisms underlying this role of mucin include the trapping of worms in the mucus layer and the inhibition of worm motility and feeding (44). In this study, we observed significantly lower numbers of goblet cells in Ad5IL-12-treated mice compared with those in PBS- or control virus-treated infected mice. An important role for IL-13 in the development of goblet cells has been reported (43) in mice infected with N. brasiliensis. Recently, we (35) have also reported a role for Stat6 activation in the regulation for the development of goblet cell hyperplasia during nematode infection. Because IL-12 activates the development of Th1-type immune response and inhibits the Th2-type response (42), it is likely that Ad5IL-12 inhibits the development of goblet cell hyperplasia by the same mechanism.
The contribution of other cell types and their products to the pathophysiology seen in this model should receive comment. Although intestinal eosinophilia and mastocytosis are features of T. spiralis infection (25), a protective role for these cells has not been identified in this nematode infection (15). We have also concluded that eosinophils are not the major contributor of intestinal muscle hypercontractility seen duringT. spiralis infection (36). Mast cells are considered important for successful worm expulsion (15,25). A recent study from our (53) laboratory revealed delayed worm expulsion and altered intestinal motor function in mast cell-deficient W/WV mice infected with T. spiralis. The altered motor function was due partly to the absence of c-kit as the altered function was not normalized after mast cell reconstitution by bone marrow grafting, suggesting a role for c-kit-dependent interstitial cells of Cajal in the motor function in gut. We have also shown that macrophages infiltrate the gut, including the neuromuscular layers during this infection in mice and that these cells are responsible for the changes seen in enteric nerves in this model (20). This effect is most likely mediated by proinflammatory cytokines such as IL-1β, IL-6, and tumor necrosis factor-α (TNF-α), which are known to suppress neurotransmitter release (41, 45). Furthermore, we have shown that these cytokines are expressed in the myenteric plexus ofT. spiralis-infected animals (33). Thus although macrophages are critical for the neural abnormalities seen in this model, T cells are the major cell type responsible for the hypercontractility of muscle and goblet cell hyperplasia.
Interestingly, in the present study we observed a lower level of MPO activity in infected mice treated with Ad5IL-12. MPO is an enzyme contained in the azurophilic granules of neutrophils and other myeloid cells and is commonly used as an index of neutrophil infiltration and acute inflammation (46). The precise mechanism by which IL-12 downregulates the MPO activity is not clear. Administration of recombinant IL-12 protein in Chlamydia psittaci inhibited the accumulation of neutrophils in the lungs, and this was associated with downregulation of several chemokines, including macrophage inflammatory protein-2, monocyte chemotactic protein-1, and TNF-α (29). The suppression of MPO activity by Ad5IL-12 duringT. spiralis infection may reflect an inhibitory effect of IL-12 on chemoattractants in the gut. As the increase in MPO duringT. spiralis infection was attenuated in athymic nude and Stat6-deficient mice (36, 55), it is also possible that the impaired increase in MPO seen at day 10 PI in Ad5IL-12-treated mice was a consequence of the suppressed Th2 response. We feel that the impaired MPO increase and the attenuation of muscle hypercontractility in Ad5Il-12-treated mice are not causally linked as previous studies (55, 57) have shown that changes in muscle are critically dependent on the presence of T cells rather than changes in MPO activity.
Our results are consistent with the hypothesis that it is the nature of the immune response that determines the adaptive host responses to enteric infection. Furthermore, by manipulating the host immune response, we are able to modulate the accompanying physiological changes, and this may have clinical relevance. Alterations in muscle contractility and other aspects of gut physiology occur in a variety of clinical settings associated with inflammation and immune activation (13). These include intestinal pseudoobstruction (9,37) as well irritable bowel syndrome (IBS) that occurs post enteric infection (27), following exacerbations of idiopathic inflammatory bowel disease (30, 40), and in some severe cases of idiopathic IBS (49). An understanding of the underlying immune mechanisms of altered physiology, coupled with the ability to modulate immune responses as depicted in this study, may yield new therapeutic opportunities in these conditions.
This study was supported by the Medical Research Council of Canada and the Canadian Institutes for Health Research (S. M. Collins and J. Gauldie).
Address for reprint requests and other correspondence: S. M. Collins, Rm. 4W8, Health Science Center, McMaster Univ., Hamilton, Ontario, Canada L8N 3Z5 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2001 the American Physiological Society