Immunological mechanisms of goblet cell hyperplasia and mucin production

Although infection with T. spiralis leads to an enteropathy characterized by goblet cell hyperplasia and an upregulation of Muc2, Muc3 and ITF production/expression, the precise mechanisms regulating intestinal goblet cell and mucin responses and their precise role in host defence remain to be determined. Goblet cell hyperplasia in nematode (including T. spiralis) infections has been suggested to be under immunological control (Miller and Nawa, Reference Miller and Nawa1979; Khan et al. Reference Khan, Abe, Ishikawa, Nawa and Yoshimura1995; Ishikawa et al. Reference Ishikawa, Wakelin and Mahida1997). T cells are critical in many immune responses, including those associated with intestinal nematode infection. It has been shown that the treatment of T. spiralis-infected mice with cyclosporin A (which profoundly inhibits cytokine production) causes a reduction in the goblet cell hyperplasia during the infection (Garside et al. Reference Garside, Grencis and Mowat1992), implying a role for immune mediators in the control of goblet cell response in this nematode infection. In N. brasilensis infection, it has also been shown that treatment of mice 1 day before infection with a single dose of anti-CD4 antibody significantly reduced both the amount of intestinal mucus produced and the spontaneous expulsion of worms (Khan et al. Reference Khan, Abe, Ishikawa, Nawa and Yoshimura1995). In addition, activation of mucosal T lymphocytes of duodenal biopsies with monoclonal anti-CD3 antibody significantly increased both radio-isotope labelled glucosamine incorporation into glycoproteins and the secretion of radio-isotope labelled glycoproteins in vitro (Crabtree et al. Reference Crabtree, Heatley, Trejdosiewicz and Losowsky1990). These observations clearly suggest an important role for T cells in infection-induced intestinal goblet cell hyperplasia and mucus production.

Among the distinct CD4⁺ T helper (Th) cell subsets (Mosmann and Coffman, Reference Mosmann and Coffman1989), the Th2 type cells are important in the protective immunity in the host against many intestinal nematode infections, including that of T. spiralis, and are characterized by the expression of cytokines, such as IL-4, IL-5, IL-9, and IL-13 (Grencis, Reference Grencis1997; Else and Finkelman, Reference Else and Finkelman1998). IL-4 and IL-13 share the alpha chain of the IL-4 receptor, and their binding to this receptor results in an activation of a signalling pathway involving the signal transducer and activator of transcription factor 6 (Stat6) through the phosphorylation by Janus kinases 1 and 3. Once activated, Stat6 proteins form homodimers, translocate to the nucleus and bind to promoter regions to regulate gene transcription. In T. spiralis infection in mice, Th2 cells play the predominant role in the development of intestinal goblet cell hyperplasia (Ishikawa et al. Reference Ishikawa, Wakelin and Mahida1997). Studies from our laboratory have shown a critical role for Stat6 in the development of goblet cell hyperplasia during this nematode infection (Khan et al. Reference Khan, Blennerhasset, Ma and Collins2001a). There was no significant difference in the number of goblet cells between non-infected Stat6-deficient and non-infected wild-type mice, suggesting that there is a baseline level of goblet cells in Stat6-deficient mice (which are not regulated by immune-dependent components) (Khan et al. Reference Khan, Blennerhasset, Ma and Collins2001a). Stat6-deficient mice infected with T. spiralis did not develop an infection-induced goblet cell hyperplasia as compared with wild-type infected mice (Fig. 1). In addition, a significant attenuation of T. spiralis infection-induced goblet cell hyperplasia was observed when the Th2 immune response was ‘shifted’ towards a Th1 immune response by IL-12 gene transfer (Khan et al. Reference Khan, Blennerhasset, Deng, Gauldie, Vallance and Collins2001b). Taken together, Th2 cytokines are strong candidates for the proliferation/differentiation of intestinal goblet cells during T. spiralis infection, although we have recently shown that IL-9 is not a major contributor, because anti-IL-9 antibody treatment has no significant effect on T. spiralis infection-induced goblet cell hyperplasia (Khan et al. Reference Khan, Richard, Akiho, Blennerhasset, Humphreys, Grencis, Van Snick and Collins2003).

Fig. 1.

Intestinal goblet cell response in Trichinella spiralis-infected mice. (A) Light micrograph of small intestinal section stained with periodic acid-Schiff (PAS) from C57BL/6 wild-type (Stat6^+/+) mice after T. spiralis infection on day 14 post-infection (p.i.). (B) Light micrograph of small intestinal section stained with PAS from T. spiralis-infected Stat6-deficient (Stat6^−/−) mice on day 14 p.i. (C) Light micrograph of small intestinal section stained with PAS from control adenovirus (Ad5DL70-3)-treated T. spiralis-infected mice on day 10 p.i. (D) Light micrograph of PAS-stained small intestinal section from Ad5IL-12-treated and T. spiralis-infected mice on day 10 p.i.

Goblet cells are the main source of ITF, which is considered to be associated with mucin to increase the viscosity of mucus gel. A hallmark of the differentiation of intestinal goblet cells is the early expression of ITF (Thim et al. Reference Thim, Madsen and Poulsen2002). Recently, it has been demonstrated that IL-4 and IL-13 upregulate ITF from mucin-producing HT-29 CL.16E and HT-29 cells via the Stat6 pathway (Blanchard et al. Reference Blanchard, Durual, Estienne, Bouzakri, Heim, Blin and Cuber2004). In addition, it has also been shown that the transcription of MUC2 in human colon cancer cells is stimulated by IL-4 and IL-13 via a mitogen-activated protein kinase pathway (Iwashita et al. Reference Iwashita, Sato, Sugaya, Takahashi, Sasaki and Abe2003). As IL-4 receptors have been shown to be present on intestinal epithelial cells (Reinecker and Podolsky, Reference Reinecker and Podolsky1995), cultured airway epithelial cells (Dabbagh et al. Reference Dabbagh, Takeyama, Lee, Ueki, Lausier and Nadel1999) and HT-29 CL.19E cells (Blanchard et al. Reference Blanchard, Durual, Estienne, Bouzakri, Heim, Blin and Cuber2004), it is reasonable to presume that Th2 cytokines like IL-4 and IL-13 play an important role in ITF and mucin production by intestinal goblet cells.

Role of goblet cells/mucins in host defence

Mucins from goblet cells might play an important role in host defence during nematode infection in the gut by dislodging the established worms, or trapping the worms in the mucus and inhibition of parasite motility and feeding, thereby excluding them from their niche and favouring their expulsion (Miller, Reference Miller1987; Rothwell, 1989). Mucins might also mimic epithelial cell glycoproteins and thus prevent the attachment of microbes to the epithelium, or these mucins could facilitate the incorporation of mediators from inflammatory cells or of host antibody, which may have a deleterious effect on invading agents. The involvement of mucus in the expulsion of parasitic nematodes has been the subject of many investigations in various systems. Histological and perfusion studies revealed that N. brasiliensis worms failed to penetrate through mucus in primed rats (Miller, Reference Miller1987). In the same model, it has been shown that the pre-treatment of the immune intestine with mustard oil, which strips off the mucus layer, or with cysteine/papain, which is a mucolytic, allowed the establishment of worms in the intestine (Miller and Huntley, Reference Miller and Huntley1982). The entrapment of worms in the mucus and a possible barrier effect of mucus on the establishment of infective larvae were also observed during T. spiralis infection (Miller, Reference Miller1987). Recently, we showed that the inhibition of T. spiralis expulsion in Stat6-deficient mice and in mice following a switch of the immune response toward a Th1 type by IL-12 gene transfer was associated with significant reduction in the number of intestinal goblet cells (Khan et al. Reference Khan, Blennerhasset, Ma and Collins2001a, Reference Khan, Blennerhasset, Deng, Gauldie, Vallance and Collinsb). We also demonstrated that the expulsion of intestinal worms was substantially delayed in CD40 ligand-deficient (CD40L^−/−) mice as compared with wild-type mice following T. spiralis infection (Khan et al. Reference Khan, Motomura, Blennerhassett, Kanbayashi, Varghese, El-Sharkawy, Gauldie and Collins2005). Consistent with the delayed worm expulsion, there was a significant impairment of intestinal goblet cell hyperplasia and Th2 cytokine responses in CD40L^−/− mice (Khan et al. Reference Khan, Motomura, Blennerhassett, Kanbayashi, Varghese, El-Sharkawy, Gauldie and Collins2005). Taken together, these observations suggest that Th2 cytokines regulate the development of goblet cell hyperplasia in the gut during nematode infection via Stat6 activation, and that the increased number of goblet cells plays an important role in host protective immunity against the infection.

Immunological mechanisms of intestinal muscle hypercontractility

Infiltration of muscle layers by T lymphocytes during T. spiralis infection and the presence of a close communication between lymphocytes and muscle cells (as revealed by ultrastructural analysis) strongly suggest the possibility of a direct interaction between these cell types (Collins, Reference Collins1996). To investigate the role of T cells in intestinal muscle function, studies were performed using the T. spiralis model employing congenitally athymic mice, specific T cell subset-deficient (CD4⁺ cell or CD8⁺ cell-deficient) mice, and MHC II-deficient mice (Vallance et al. Reference Vallance, Croitoru and Collins1998, Reference Vallance, Galeazzi, Collins and Snider1999b). Infection-induced intestinal muscle hypercontractility is T cell-dependent, because changes were attenuated from infected athymic mice. The role of CD8⁺ T cells seems not to be significant, as there was no difference in the development of infection-induced muscle hypercontractility and in worm expulsion between CD8-deficient and wild-type mice (Vallance et al. Reference Vallance, Galeazzi, Collins and Snider1999b). However, CD4⁺ T cells appear to play a role, since CD4-deficient mice exhibited impaired muscle contractility; MHC II-deficient mice showed the greatest attenuation in the development of muscle contractility and also exhibited significant inhibition in worm expulsion (Vallance et al. Reference Vallance, Galeazzi, Collins and Snider1999b). The reconstitution of T cell function in athymic mice, by injecting splenic lymphocytes from euthymic mice prior to infection with T. spiralis, restored the profile of intestinal muscle contraction observed in euthymic mice and accelerated the expulsion of the parasite from the gut. To examine the role of CD4⁺ cells further, MHC II-deficient mice were reconstituted with purified CD4⁺ T cells and the effect on intestinal muscle contractility was assessed following infection. The results suggest partial restoration of muscle function during the infection, implying the importance of CD4⁺ T cells in enteric muscle function. Recently, we have shown that a Stat6-dependent Th2 immune response is critical for the development of intestinal muscle hypercontractility in T. spiralis infection (Khan et al. Reference Khan, Vallance, Blennerhasset, Deng, Verdue, Matthaei and Collins2001c) and have observed a significant reduction in infection-induced intestinal muscle contractility in Stat6^−/− mice as compared with wild-type mice. The attenuated muscle contractility was associated with delayed worm expulsion from the intestine. Importantly, we were also able to demonstrate the presence of IL-4 and IL-13 in the lamina muscularis externa during the primary infection with T. spiralis in wild-type mice, whereas there no transcription was detected for IL-4 or IL-13 in infected Stat6^−/− mice. It has been also shown that IL-4 and IL-13 enhances carbachol-induced contraction in dispersed smooth muscle cells isolated from the jejunum of mice (Akiho et al. Reference Akiho, Blennerhassett, Deng and Collins2002). These findings support the finding that Stat6 signalling promotes protective immunity against T. spiralis through mast cell- and T cell-dependent mechanisms (Urban et al. Reference Urban, Schopf, Morris, Orekhova, Madden, Betts, Gamble, Byrd, Donaldson, Else and Finkelman2000). These observations suggest that Th2 cytokines, IL-4 and IL-13, acting via Stat6, mediate the generation of infection-induced intestinal muscle hypercontractility. More recently, it has been reported that IL-4 and IL-13 play a key role in the generation of intestinal muscle contractility in mice infected with N. brasiliensis and Heligosomoides polygyrus (see Zhao et al. Reference Zhao, McDermott, Urban, Gause, Madden, Yeung, Morris, Finkelman and Shea-Donohue2003), further supporting our hypothesis that a Th2 response mediates muscle contractility in nematode infection.

To evaluate the role of Th1/Th2 immune responses in intestinal muscle function, muscle contractility was investigated following over-expression of IL-12 in T. spiralis-infected mice. IL-12 is a key cytokine in the differentiation of CD4⁺ Th cells toward Th1-associated responses by stimulating the production of IFN-γ by Th and natural killer cells (Manetti et al. Reference Manetti, Parronchi, Giudizi, Piccinni, Maggi, Trinchieri and Romagnani1993). It has been reasoned that if the Th2 response is critical for both the development of intestinal muscle hypercontractility and the expulsion of the worms, then a shift of the Th2 response toward a Th1 response should not only prolong the infection but also attenuate infection-induced muscle hypercontractility. Indeed, that is the case and a shift to Th1 response via an overexpression of IL-12 significantly altered intestinal muscle hypercontractility in this Th2-based, enteric infection (Khan et al. Reference Khan, Blennerhasset, Deng, Gauldie, Vallance and Collins2001b). In this study, IL-12 gene transfer using a single injection of the recombinant adenovirus vector expressing IL-12 (Ad5IL-12) into T. spiralis-infected mice effectively inhibited the development of infection-induced intestinal muscle hypercontractility and prolonged worm survival in the gut. This was accompanied by an upregulation of the Th1 cytokine IFN-γ and downregulation of Th2 cytokines IL-4 and IL-13 production. These findings clearly show that Th2 cytokines are prime candidates for the development of intestinal muscle hypercontractility in this infection. However, not all Th2 cytokines are critical, as it has been shown that IL-5 or IL-9 are not major contributors to the generation of T. spiralis-induced intestinal muscle hypercontractility (Vallance et al. Reference Vallance, Blennerhassett, Deng, Matthaei, Young and Collins1999a; Khan et al. Reference Khan, Richard, Akiho, Blennerhasset, Humphreys, Grencis, Van Snick and Collins2003). Deficiency in IL-5 or neutralization of IL-9 did not have a significant effect on muscle contraction in the host during this infection, which might be due to a dominant effect of IL-4 or IL-13.

To understand the initial events linking infection to the development of muscle hypercontractility, we recently examined the contribution of CD40-CD40L interaction in the development of intestinal muscle hypercontractility and in the Th2 response in CD40L-deficient (CD40L^−/−) mice infected with T. spiralis (Khan et al. Reference Khan, Motomura, Blennerhassett, Kanbayashi, Varghese, El-Sharkawy, Gauldie and Collins2005). The interaction between CD40 ligand (CD40L) on T cells and CD40 on antigen-presenting cells has been shown to regulate both humoral and cellular immune responses (Grewal and Flavell, Reference Grewal and Flavell1996). Expulsion of intestinal worms was substantially delayed in CD40L^−/− mice as compared with wild-type mice following T. spiralis infection (Khan et al. Reference Khan, Motomura, Blennerhassett, Kanbayashi, Varghese, El-Sharkawy, Gauldie and Collins2005). Consistent with delayed worm expulsion, there was a significant attenuation of intestinal muscle contractility in CD40L^−/− mice (Khan et al. Reference Khan, Motomura, Blennerhassett, Kanbayashi, Varghese, El-Sharkawy, Gauldie and Collins2005). Infected CD40L^−/− mice also exhibited marked impairment in the production of Th2 cytokines. These data suggest that the CD40-CD40 ligand interaction is an important, early event in the generation of a Th2 response and in the development of intestinal muscle hypercontractility in T. spiralis infection.

In addition to T cells, other cells such as mast cells, ICC and EE cells in the gut mucosa may play important roles in the development of intestinal muscle hypercontractility. Intestinal mucosal mastocytosis is a characteristic feature of nematode (including T. spiralis) infection (Grencis, Reference Grencis1997), and these cells are considered to have an important role in host protective immunity (Grencis et al. Reference Grencis, Else, Huntley and Nishikawa1993; Ha et al. Reference Ha, Reed and Croll1983). The activation of mast cells and release of their mediators have been implicated not only in the expulsion of the worms but also in the development of an altered physiology seen during the infection (Perdue and McKay, Reference Perdue and McKay1994; Goldhill et al. Reference Goldhill, Morris, Malizewski, Urban, Finkelman and Shea-Donohue1997; Lawrence et al. Reference Lawrence, Paterson, Higgins, MacDonald, Kennedy and Garside1998; Madden et al. Reference Madden, Young, Zhao, Gause, Finkelman, Katona, Urban and Shea-Donohue2004). Recently, an important role for mast cells has been demonstrated in epithelial cell function in T. spiralis infection, suggesting that the mast cells may contribute to worm expulsion in this infection by altering epithelial cell permeability (McDermott et al. Reference McDermott, Bartram, Knight, Miller, Garrod and Grencis2004). Studies of T. spiralis-infected c-kit-defective W/W^V mice, which are mast cell-deficient, have revealed an attenuation of infection-induced intestinal muscle hypercontractility (Vallance et al. Reference Vallance, Blennerhassett, Huizinga and Collins2001). Bone marrow grafts from congenic wild-type mice restored mast cells as well as other haematopoietic cell populations to W/W^V mice, but the attenuated muscle contractility persisted. A possible explanation for this observation was the demonstration that W/W^V mice also lack a population of c-kit-dependent ICC. In spite of the restoration of mast cells, bone marrow reconstitution in W/W^V failed to correct deficiencies in ICC, which might be a possible cause of the persistent altered muscle function in these mice (Vallance et al. Reference Vallance, Blennerhassett, Huizinga and Collins2001). ICC regulates myoelectrical slow wave activity and is important in the control of the propulsive motor activity of the small bowel (Huizinga et al. Reference Huizinga, Thunenberg, Kluppel, Malysz, Mikkelson and Bernstein1995). In the primary infection of mice with T. spiralis, structural damage has been observed in the network of ICC, which was associated with the appearance of ectopic pacemaker activity and the occurrence of retrograde peristalsis, implicating a major role for ICC in inflammation-induced gut motor disturbances (Der et al. Reference Der, Bercik, Donnelly, Jackson, Berezin, Collins and Huizinga2000). These findings were also evident in bone marrow-reconstituted W/W^V mice. Together, these results indicate that non-bone marrow-derived cells such as ICC, in cooperation with mast cells, play separate but integrated roles in the function of intestinal muscle in mice with T. spiralis infection.

EE cells are specialized cells of gut epithelium and are dispersed throughout the mucosa of the gut. These cells release various biologically active compounds, such as cholecystokinin (CCK), chromo granin A (CGA) and serotonin (5-hydroxytryptamine, 5-HT), in response to various chemical and mechanical stimuli (Gershon, Reference Gershon1999; Lundgren, Reference Lundgren2002; Sharkey and Mawe, Reference Sharkey and Mawe2002). The most well-characterized subset of EE cells are the enterochromaffin (EC) cells, which synthesize and release the biogenic amine 5-HT (Cetin et al. Reference Cetin, Kuhn, Kulaksiz, Adermann, Bargsten, Grube and Forssmann1994; Sharkey and Mawe, Reference Sharkey and Mawe2002). 5-HT is an important enteric mucosal signalling molecule influencing gut physiology (motor and secretory function), and is considered important in maintaining intestinal homeostasis (Sharkey and Mawe, Reference Sharkey and Mawe2002). EC cells are located in close proximity to the mucosal sensory nerve endings and interganglionic neurons, being synaptic on motor excitatory and inhibitory neurons. An upregulation of EC cell numbers has been observed in infections of nematodes, including N. brasiliensis, T. spiralis or Trichuris muris (see Farmer and Laniyonu, Reference Farmer and Laniyonu1984; Wheatcroft et al. Reference Wheatcroft, Wakelin, Smith, Mahoney, Mawe and Spiller2005; Wang et al. Reference Wang, Steeds, Motomura, Deng, Verma-Gandhu, El-Sharkawy, McLaughlin, Grencis and Khan2007). Studies using T cell receptor (β×δ)-deficient mice demonstrated a reduced increase in EC cells in the small intestine in T. spiralis infection, suggesting a role for T cells in the development of infection-induced EC cell hyperplasia in this infection (Wheatcroft et al. Reference Wheatcroft, Wakelin, Smith, Mahoney, Mawe and Spiller2005). Recent work from our laboratory has shown an important immuno-endocrine axis in the gut, in which secretory products from CD4⁺ T cells interact with EC cells to enhance 5-HT production in the gut via Th2-based mechanisms (Wang et al. Reference Wang, Steeds, Motomura, Deng, Verma-Gandhu, El-Sharkawy, McLaughlin, Grencis and Khan2007). We also observed that EC cell and 5-HT responses in inflammation induced by T. muris infection are influenced by Th1 or Th2 cytokine predominance (Motomura et al. Reference Motomura, Ghia, Wang, Akiho, El-Sharkawy, Collins, Wan, McLaughlin and Khan2008). In addition, we have demonstrated that the increase in EC cell numbers in T. muris infection correlates well with an upregulation of colonic muscle hypercontractility (Motomura et al. Reference Motomura, Verma-Gandhu, El-Sharkawy, McLaughlin, Grencis and Khan2004). Recently, the 5-HT2A receptor has been shown to contribute to enteric infection-induced intestinal muscle hypercontractility, which further supports a role for 5-HT in alteration of muscle function in the gut during infection and inflammation (Zhao et al. Reference Zhao, Urban, Morimoto, Elfrey, Madden, Finkelman and Shea-Donohue2006).

Role of intestinal muscle hypercontractility in host defence

Previous work from our laboratory has shown that the inflammation-induced intestinal muscle hypercontractility is prominent in the proximal part of the intestine, and that the distal parts such as the ileum and colon exhibit reduced muscle contractility (Grossi et al. Reference Grossi, McHugh and Collins1993). Therefore, it seems likely that the distribution of changes would create an aboral gradient in muscle tension generation during infection, enhancing the aboral propulsion of luminal contents. If such forces contribute to the host's defence against parasite infection, then one might expect to see a relationship between the magnitude of intestinal muscle hypercontractility and the ability of the host to evict the parasites from the gut. This is indeed the case, with strong responders to T. spiralis infection (such as NIH Swiss mice which rapidly expel the parasite) exhibiting a greater degree of muscle hypercontractility and slow responders (such as B10.BR mice which slowly expel the parasite) exhibiting a lesser degree of hypercontractility (Vallance et al. Reference Vallance, Blennerhassett and Collins1997). These observations support the proposal for an association between host defence and changes in muscle function, and prompted a closer examination of the underlying immune-mediated mechanisms involved in the alteration of intestinal muscle contractility. Studies involving mice deficient in T cells, CD4⁺ cells, MHC II, CD40L and Stat6, and IL-12 gene transfer-mediated switching of the immune response towards Th1 type clearly demonstrate that infection-induced immune-mediated intestinal muscle hypercontractility in T. spiralis infection is associated with efficient worm expulsion. An attenuation of intestinal muscle contractility is associated with an inhibition of worm expulsion in this model. Recently, it was shown that worm expulsion is delayed in smooth muscle cell IL-4R-alpha-deficient mice infected with N. brasiliensis, suggesting a role for IL-4R-alpha-responsive smooth muscle cells in host defence in nematode infections (Horsnell et al. Reference Horsnell, Cutler, Hoving, Mearns, Myburgh, Arendse, Finkelman, Owens, Erle and Brombacher2007). It seems likely that cytokines, specifically IL-4 and IL-13 from the Th2 type cells, effect the secondary messenger system of muscle cells by acting either on the calcium channels or on the muscarinic receptor, which subsequently alters the intestinal muscle contractility; this alteration of contractility is associated with enhanced parasite eviction. We have investigated a causal link between these responses by administering an opiate (buprenorphine) to attenuate muscle hypercontractility in T. spiralis-infected mice, and determined whether this delayed worm expulsion occurred without attenuating the Th2 response. The results showed a marked inhibition of worm expulsion and intestinal muscle hypercontractility (Fig. 2) without impairing the IL-4 response, supporting the hypothesis that altered motility plays an important role in host defence in this model (Khan et al. Reference Khan, Motomura, Verdu, Blennerhasset, El-Sharkawy, Verma-Gandhu and Collins2004). Taken together, these findings indicate clearly that the intestinal motor apparatus acts as an extension of the immune system, aiding in the expulsion of worms through increased propulsive activity.

Fig. 2.

Intestinal muscle contractility and worm expulsion after opiate treatment. C57BL/6 mice were infected with Trichinella spiralis and were treated with opiate (buprenorphine; BPN) subcutaneously. Control mice received sterile saline. (A) Contraction of intestinal muscle. Maximum tension generated by intestinal muscle taken from non-infected and from BPN- and saline-treated mice after infection. (B) Worm recovery from BPN- and saline-treated mice after T. spiralis infection on day 14 p.i. * Significantly different (P<0·05) between BPN-treated and saline-treated infected mice.