**7. Effect of helminths on experimental autoimmunity**

#### **7.1 Experimental autoimmune encephalomyelitis (EAE)**

EAE in rodents has been widely used as an animal model of MS. In EAE, both Th1 and Th17 have been reported to be encephalitogenic and their relative importance depends on the mouse strain and MOG epitope used for immunization (El-behi et al., 2010). In addition, Th1 and Th17 have different encephalitogenic roles as demonstrated by pathological observations; e.g., distinct cellular infiltrates (macrophage predominant or neutrophil predominant) and distinct sites of inflammation (mainly in the spinal cord or mainly in the brain) (El-behi et al., 2010; Domingues et al., 2010). These observations reinforce the necessity for an antagonistic effect against both Th1 and Th17 for ideal immunomodulatory medicines.

Studies of therapeutic effects of helminths on EAE are summarized in Table 1. Antiencephalitogenic effects have been observed in all three groups of parasitic helminths; i.e. nematodes (*Heligmosomoides polygyrus* (Wilson et al., 2010), *Trichinella spiralis* (Gruden-Movsesijan et al., 2008), *Trichinella pseudospiralis* (Wu et al., 2010) ), trematodes (*Schistosoma japonicum* (Zheng et al., 2008), *Schistosoma mansoni* (Sewell et al., 2003; La Flamme et al., 2003), *Fasciola hepatica* (Walsh et al., 2009)) and cestodes (*Taenia crassiceps*) (Reyes et al., 2011). Among them, there are some distinct results in the effects of schistosomes. According to Sewell et al. (2003), the intra-peritoneal injection of schistosome eggs protected mice from EAE, whereas La Flamme et al. (2003), reported that a similar injection did not have protective effect. In general, a down-regulation of both Th17 and Th1 cytokine expression has been demonstrated (Walsh et al., 2009; Wu et al., 2010; Reyes et al., 2011) except in papers published before the emergence of the Th17 concept (Sewell et al., 2003; La Flamme et al., 2003). Regarding cellular involvement in the suppression, B cells highly expressing CD23 were shown to be responsible for EAE suppression (Wilson et al., 2010) in adoptive transfer experiments. In that study, B cells from IL-10-deficient mice as well as from wildtype mice conferred protection against EAE. The involvement of AAMΦ is also plausible, because AAMΦ markers are increased in the brain in *T. crassiceps*-infected EAE mice (Reyes et al., 2011). In addition, abrogation of schistosome-induced anti- encephalitogenic effects in STAT6-deficient mice (Sewell et al., 2003) might support the importance of AAM Φ.


↓:down-regulation, ↑:up-regulation

Table 1. Suppressive effect of parasitic helminths on EAE.

#### **7.2 Experimental T1D**

596 Autoimmune Disorders – Current Concepts and Advances from Bedside to Mechanistic Insights

the finding of a critical role for IL-23 in the pathogenesis of some experimental forms of autoimmune diseases (Cua et al., 2003; Murphy et al., 2003) and subsequent discovery of a pathogenic T cell subset producing IL-17 (Th17) (Langrish et al., 2005; Park et al., 2005) has led us to revisit the "Th1/Th2 paradigm". The simplified "Th1/Th2 paradigm" still explains many immunological phenomena, but there is accumulating evidence of the importance of Th17 in the pathogenesis of autoimmunity. By using IL-17-deficient mice, essential roles of IL-17 in the pathogenesis of autoimmune disease have been demonstrated directly (Nakae et al., 2003a, 2003b; Komiyama et al., 2003). Development of the Th17 lineage is antagonized by both Th1-related cytokines (IL-12 and IFN-γ) and a Th2 cytokine (IL-4) (Park et al., 2005; Nakae et al., 2007). Conversely, IL-23 and IL-17 negatively regulate Th1 differentiation (Nakae et al., 2007). Therefore, the balance of these Th subsets is much more complex than previously believed. Complicating the situation further is the recent finding that Th17 is not a stable subset and can be changed to the Th1 phenotype; i.e. plasticity of Th17 (Kurschus et al., 2010; Dong, 2011). The fate of Th17 cells depends on their surrounding environment (Dong, 2011; Lee et al., 2009) and Th17's pathogenic nature depends on the conversion to Th1 cells, in the case of experimental T1D (Martin-Orozco et al., 2009; Bending et al., 2009). The relative importance of each Th subset to the pathogenesis may differ with the disease model, however in most cases, the pathogenicity of Th1 and Th17 is still under debate. There is a report that a transcription factor, T-bet, is essential for the encephalitogenicity of T cells rather than cytokine production (Yang et al., 2009). In contrast, T-bet seems to be a negative regulator in experimental autoimmune myocariditis (Rangachari et al., 2006). Given the unstable nature of the Th17 subset and disease heterogeneity of individual patients, antagonism of both the Th1 and Th17 subsets would be a better choice for the successful suppression of autoimmune diseases. From this viewpoint, parasitic helminths may have

ideal immunomodulatory activities for treatment of autoimmunity.

EAE in rodents has been widely used as an animal model of MS. In EAE, both Th1 and Th17 have been reported to be encephalitogenic and their relative importance depends on the mouse strain and MOG epitope used for immunization (El-behi et al., 2010). In addition, Th1 and Th17 have different encephalitogenic roles as demonstrated by pathological observations; e.g., distinct cellular infiltrates (macrophage predominant or neutrophil predominant) and distinct sites of inflammation (mainly in the spinal cord or mainly in the brain) (El-behi et al., 2010; Domingues et al., 2010). These observations reinforce the necessity for an antagonistic effect against both Th1 and Th17 for ideal immunomodulatory

Studies of therapeutic effects of helminths on EAE are summarized in Table 1. Antiencephalitogenic effects have been observed in all three groups of parasitic helminths; i.e. nematodes (*Heligmosomoides polygyrus* (Wilson et al., 2010), *Trichinella spiralis* (Gruden-Movsesijan et al., 2008), *Trichinella pseudospiralis* (Wu et al., 2010) ), trematodes (*Schistosoma japonicum* (Zheng et al., 2008), *Schistosoma mansoni* (Sewell et al., 2003; La Flamme et al., 2003), *Fasciola hepatica* (Walsh et al., 2009)) and cestodes (*Taenia crassiceps*) (Reyes et al., 2011). Among them, there are some distinct results in the effects of schistosomes. According to Sewell et al. (2003), the intra-peritoneal injection of schistosome eggs protected mice from EAE, whereas La Flamme et al. (2003), reported that a similar injection did not have

**7. Effect of helminths on experimental autoimmunity 7.1 Experimental autoimmune encephalomyelitis (EAE)** 

medicines.

Non-obese diabetic (NOD) mice have been used widely as an animal model of T1D. Spontaneous destruction of pancreatic β-cells and subsequent hyperglycemia are observed in NOD mice. The pathogenesis of T1D in this mouse has been studied extensively, however, there is still considerable controversy over the relative contribution of Th1 (or IFNγ) and Th17 (or IL-17). Anti-IFN-γ treatment rendered NOD mice resistant to cyclophosphamide (CY)-accelerated diabetes (Debray-Sachs et al., 1991). In contrast, in IFNγ deficient NOD mice, neither insulitis nor diabetes was prevented although the onset was delayed (Hultgren et al., 1996; Gysemans et al., 2008). These findings suggest that IFN-γ is involved in, but not essential to, the pathogenesis. IFN-γ is not itself detrimental to β-cells, but induces apoptosis when acting with IL-1β or TNF-α (Gysemans et al., 2008). Regarding this pro-apoptotic effect, dual roles of IFN-γ in NOD mice have been indicated; i.e. IFN-γ induces β-cell destruction via STAT-1 but protects β-cells via IRF-1 (Gysemans et al., 2008). On the other hand, pathological roles of IL-17 have been also suggested in mice (Miljkovic et al., 2005: Emamaullee et al., 2009) and humans (Honkanen et al., 2010). Anti-IL-17 antibody prevented diabetes in NOD mice when administered around the time of onset (Emamaullee et al., 2009). At present, it is reasonable to conclude that both Th1 and Th17 play some role in diabetogenesis in NOD mice.

Parasitic Helminths as Potential Therapeutic Agents Against Autoimmune Disorders 599

by IL-17 and acts on osteoclast precursors to stimulate their differentiation into multinucleated osteoclasts (Okamoto & Takayanagi, 2010). An essential role for IL-17 in the pathogenesis of CIA has been demonstrated directly by using IL-17-deficient mice (Nakae et al., 2003a). IL-17 is an essential pathological factor also in other models of arthritis; e.g. IL-1 receptor antagonist-deficient mice (Nakae et al., 2003b), Ag-induced arthritis (AIA) (Irmler et al., 2007) and glucose 6-phosphate isomerase (GPI)– induced arthritis (Iwanami et al., 2008). Thus, these models could be considered Th17-dependent. In contrast, IFN-γ is now considered an ameliorating factor in CIA (Kelchtermans et al., 2009; Chu et al., 2007) and in AIA (Irmler et al., 2007). However, in proteoglycan-induced arthritis, IFN-γ (not IL-17) is responsible for the pathogenesis (Doodes et al., 2008). Taken together, models of autoimmune arthritis are mainly dependent on Th17 but exceptions do exist. In human RA, the pathological importance of TNF-α, IL-1β, IL-6 and IL-17 has been demonstrated directly by the striking efficacy of biological drugs targeting those cytokines (Nixon et al., 2007; Jones & Ding, 2010; Genovese et al., 2010). In addition, an anti-RANKL monoclonal antibody has been approved for osteoporosis and is now under clinical development for RA (Pageau,

As described in section 3, the global distribution and trends of RA do not match the hygiene hypothesis (Gabriel & Michaud, 2009; Shapira et al., 2010b). In addition, to our knowledge, there is no report of anti-RA effects of parasitic helminths in humans. Nonetheless, several investigators (including us) have found anti-arthritic effects of helminths or helminthderived products in rodents. Those studies are summarized in Table 3. We ourselves have reported suppressive effects of a blood fluke, *S. mansoni,* on mouse CIA (Osada et al., 2009). In that study, the *S. mansoni* infection reduced the severity of CIA as evaluated using scores of arthritis and numbers of arthritic paws. Histopathological observation revealed the prevention of bone destruction in the infected mice. According to an analysis of splenic cytokine production pattern, production of IL-17A, TNF-α and IFN-γ were down-regulated and that of IL-4 and IL-10 was up-regulated. The real-time PCR analysis of inflamed paws showed the striking augmentation of the gene expression of bone-absorptive proinflammatory mediators (IL-1β, IL-6 and RANKL) observed in non-infected arthritic mice to be abrogated in infected mice. *S. mansoni* is a gonochoristic worm and forms a pair in the portal vein. Egg deposition in the infected host organs is the major stimulus of Th2 polarization (Grzych et al., 1991) and other immunomodulatory events such as Treg induction (Taylor et al., 2006). In our experiments, the severity of CIA correlated inversely with the numbers of worm pairs, theoretically proportional to the number of eggs produced (Osada et al., 2009). In addition, in a time-course experiment, the splenic cytokine modulation (including down-modulation of IL-17 and TNF-α) was observed from 6 to 8 week post-infection, which corresponds to the beginning of egg deposition (unpublished observation). However, repeated intra-peritoneal injections of soluble egg antigen (SEA) did not protect mice from CIA and viable eggs lost their ability to suppress IL-17 production by freeze-thawing and subsequent crushing treatment (Osada et al., 2010). Therefore, a viable infection, which supplies viable eggs continuously, may be essential for the schistosomeinduced anti-arthritic effects. By plotting the relationship between infection intensity and severity levels, we have found that a single pair of worms was enough to abrogate the augmentation of pro-inflammatory mediators and approximately 4 pairs were enough to

suppress the onset of arthritis (Osada et al., 2009).

2009).

Along with NOD mice, inducible T1D models have been used for diabetes research. One of them is streptozotocin (STZ)-induced diabetes. STZ is a glucosamine- nitrosourea compound specifically toxic to pancreatic islet β-cells (Yamamoto et al., 1981). Multiple lowdose administrations of STZ cause immune mechanism-mediated β-cell destruction and diabetes in mice (Kantwerk-Funke et al., 1991). In this T1D model, pathogenic roles of both IL-12/IFN-γ axis (Herold et al., 1996; Müller et al., 2002; Gysemans et al., 2005; Cetkovic-Cvrlje & Uckun, 2005) and IL-23/IL-17 axis (Miljkovic et al., 2005, Mensah-Brown et al., 2006) have been suggested, as in NOD mice.


↓:down-regulation, ↑:up-regulation

Table 2. Suppressive effects of parasitic helminths on experimental T1D.

Studies of the therapeutic effects of helminths on experimental T1D models are summarized in Table 2. As in the case of EAE, all three groups of helminths (nematodes, trematodes and cestodes) have preventive effects against experimental T1D models. As Treg cells play an important regulatory role against diabetogenesis (Brode et al., 2006; Ott et al., 2005), some authors suggest that they are also responsible for the anti-diabetogenic effects of helminths (Hübner et al., 2009; Zaccone et al., 2009). However, other reports do not support the importance of Treg cells (Liu et al., 2009; Espinoza-Jiménez et al., 2010). This inconsistency probably suggests the presence of distinct suppressive mechanisms in each parasite species. In our recent study, intestinal helminth infection protected mice from diabetes induced by multiple low-doses of STZ in a STAT6-independent manner (unpublished observation). This finding suggests that immune polarization from Th1 to Th2 induced by helminths cannot explain the anti-diabetogenic effect. Further studies using gene-targeted mice are needed to elucidate the anti-diabetogenic mechanisms of helminths.

#### **7.3 Experimental autoimmune arthritis**

There are various models of autoimmune arthritis in rodents. Collagen-induced arthritis (CIA) is one of the most widely used classical models of RA (Stuart et al., 1982). In this model, anti-collagen antibodies are responsible for the destruction of cartilage (Terato et al., 1992). Regarding cytokines, TNF-α, IL-1β and IL-6 are crucial to the inflammatory bone destruction in autoimmune arthritis (Ferraccioli et al., 2010; Möller & Villiger, 2006). However, the most important factor in the bone destruction is receptor activator of NFκB ligand (RANKL). This cytokine is induced to express on osteoblasts and synovial fibroblasts

Along with NOD mice, inducible T1D models have been used for diabetes research. One of them is streptozotocin (STZ)-induced diabetes. STZ is a glucosamine- nitrosourea compound specifically toxic to pancreatic islet β-cells (Yamamoto et al., 1981). Multiple lowdose administrations of STZ cause immune mechanism-mediated β-cell destruction and diabetes in mice (Kantwerk-Funke et al., 1991). In this T1D model, pathogenic roles of both IL-12/IFN-γ axis (Herold et al., 1996; Müller et al., 2002; Gysemans et al., 2005; Cetkovic-Cvrlje & Uckun, 2005) and IL-23/IL-17 axis (Miljkovic et al., 2005, Mensah-Brown et al.,

T1D model Helminth Treatment Proposed mechanism Refs

Table 2. Suppressive effects of parasitic helminths on experimental T1D.

elucidate the anti-diabetogenic mechanisms of helminths.

**7.3 Experimental autoimmune arthritis** 

Spontaneous T1D in NOD mice *Litomosoides sigmodontis* Infection, Worm Ag IL-4↑, IL-5↑, Treg Hübner et al., 2009

Eggs, Egg Ag, Worm Ag i.p.

Studies of the therapeutic effects of helminths on experimental T1D models are summarized in Table 2. As in the case of EAE, all three groups of helminths (nematodes, trematodes and cestodes) have preventive effects against experimental T1D models. As Treg cells play an important regulatory role against diabetogenesis (Brode et al., 2006; Ott et al., 2005), some authors suggest that they are also responsible for the anti-diabetogenic effects of helminths (Hübner et al., 2009; Zaccone et al., 2009). However, other reports do not support the importance of Treg cells (Liu et al., 2009; Espinoza-Jiménez et al., 2010). This inconsistency probably suggests the presence of distinct suppressive mechanisms in each parasite species. In our recent study, intestinal helminth infection protected mice from diabetes induced by multiple low-doses of STZ in a STAT6-independent manner (unpublished observation). This finding suggests that immune polarization from Th1 to Th2 induced by helminths cannot explain the anti-diabetogenic effect. Further studies using gene-targeted mice are needed to

There are various models of autoimmune arthritis in rodents. Collagen-induced arthritis (CIA) is one of the most widely used classical models of RA (Stuart et al., 1982). In this model, anti-collagen antibodies are responsible for the destruction of cartilage (Terato et al., 1992). Regarding cytokines, TNF-α, IL-1β and IL-6 are crucial to the inflammatory bone destruction in autoimmune arthritis (Ferraccioli et al., 2010; Möller & Villiger, 2006). However, the most important factor in the bone destruction is receptor activator of NFκB ligand (RANKL). This cytokine is induced to express on osteoblasts and synovial fibroblasts

*Heligmosomoides polygyrus* Infection Independent of IL-10 and Treg Liu et al., 2009 *Trichinella spiralis* Infection IL-4↑, IL-10↑ Saunders et al., 2007 *Schistosoma mansoni* Infection / Egg i.p. Anti-insulin IgG↓ Cooke et al., 1999

*Schistosoma mansoni* Infection EL-Wakil et al., 2002

*Taenia crassiceps* Infection AAM Espinoza-Jiménez et al., 2010

NKT cells↑ Zaccone et al., 2003

Egg Ag i.p. Treg Zaccone et al., 2009

2006) have been suggested, as in NOD mice.

Streptozotocin-induced diabetes in mice

(Multiple low dose model)

↓:down-regulation, ↑:up-regulation

by IL-17 and acts on osteoclast precursors to stimulate their differentiation into multinucleated osteoclasts (Okamoto & Takayanagi, 2010). An essential role for IL-17 in the pathogenesis of CIA has been demonstrated directly by using IL-17-deficient mice (Nakae et al., 2003a). IL-17 is an essential pathological factor also in other models of arthritis; e.g. IL-1 receptor antagonist-deficient mice (Nakae et al., 2003b), Ag-induced arthritis (AIA) (Irmler et al., 2007) and glucose 6-phosphate isomerase (GPI)– induced arthritis (Iwanami et al., 2008). Thus, these models could be considered Th17-dependent. In contrast, IFN-γ is now considered an ameliorating factor in CIA (Kelchtermans et al., 2009; Chu et al., 2007) and in AIA (Irmler et al., 2007). However, in proteoglycan-induced arthritis, IFN-γ (not IL-17) is responsible for the pathogenesis (Doodes et al., 2008). Taken together, models of autoimmune arthritis are mainly dependent on Th17 but exceptions do exist. In human RA, the pathological importance of TNF-α, IL-1β, IL-6 and IL-17 has been demonstrated directly by the striking efficacy of biological drugs targeting those cytokines (Nixon et al., 2007; Jones & Ding, 2010; Genovese et al., 2010). In addition, an anti-RANKL monoclonal antibody has been approved for osteoporosis and is now under clinical development for RA (Pageau, 2009).

As described in section 3, the global distribution and trends of RA do not match the hygiene hypothesis (Gabriel & Michaud, 2009; Shapira et al., 2010b). In addition, to our knowledge, there is no report of anti-RA effects of parasitic helminths in humans. Nonetheless, several investigators (including us) have found anti-arthritic effects of helminths or helminthderived products in rodents. Those studies are summarized in Table 3. We ourselves have reported suppressive effects of a blood fluke, *S. mansoni,* on mouse CIA (Osada et al., 2009). In that study, the *S. mansoni* infection reduced the severity of CIA as evaluated using scores of arthritis and numbers of arthritic paws. Histopathological observation revealed the prevention of bone destruction in the infected mice. According to an analysis of splenic cytokine production pattern, production of IL-17A, TNF-α and IFN-γ were down-regulated and that of IL-4 and IL-10 was up-regulated. The real-time PCR analysis of inflamed paws showed the striking augmentation of the gene expression of bone-absorptive proinflammatory mediators (IL-1β, IL-6 and RANKL) observed in non-infected arthritic mice to be abrogated in infected mice. *S. mansoni* is a gonochoristic worm and forms a pair in the portal vein. Egg deposition in the infected host organs is the major stimulus of Th2 polarization (Grzych et al., 1991) and other immunomodulatory events such as Treg induction (Taylor et al., 2006). In our experiments, the severity of CIA correlated inversely with the numbers of worm pairs, theoretically proportional to the number of eggs produced (Osada et al., 2009). In addition, in a time-course experiment, the splenic cytokine modulation (including down-modulation of IL-17 and TNF-α) was observed from 6 to 8 week post-infection, which corresponds to the beginning of egg deposition (unpublished observation). However, repeated intra-peritoneal injections of soluble egg antigen (SEA) did not protect mice from CIA and viable eggs lost their ability to suppress IL-17 production by freeze-thawing and subsequent crushing treatment (Osada et al., 2010). Therefore, a viable infection, which supplies viable eggs continuously, may be essential for the schistosomeinduced anti-arthritic effects. By plotting the relationship between infection intensity and severity levels, we have found that a single pair of worms was enough to abrogate the augmentation of pro-inflammatory mediators and approximately 4 pairs were enough to suppress the onset of arthritis (Osada et al., 2009).

Parasitic Helminths as Potential Therapeutic Agents Against Autoimmune Disorders 601

Increases in both IL-17 and IL-12 mRNA in CD and UC patients has been reported (Nielsen et al., 2003). The most striking evidence of the involvement of IL-23/Th17 axis is the finding of Th17-related genes (including IL-23R gene) as susceptibility genes for CD (Brand, 2009). In contrast, Th2-related cytokines are dominant in UC (Sarra et al., 2010). Likewise, models of colitis are also composed of diseases with a distinct pathogenesis (Strober et al., 2002). Experimental colitis induced by the intra-rectal injection of hapten, such as TNBS or DNBS, resembles human CD and its development seems independent of IFN-γR (Camoglio et al., 2000) and dependent on IL-17R signaling (Zhang et al., 2006). Colitis can be also induced by causing dysfunction in epithelial cell barrier. Supply of drinking water containing dextran sulfate sodium (DSS) induces this type of colitis. In this model, T and B cells are dispensable (Dieleman et al., 1994) and macrophages seem to play both a pathological role (Bauer et al., 2010) and a regulatory role (Qualls et al., 2006; Smith et al., 2007). Considering predominant expression of Th2-related cytokines in the chronic phase, DSS-induced colitis resembles UC (Alex et al., 2009). Interestingly, neutralization of IL-17 aggravates the model (Ogawa et al., 2004), suggesting that IL-17 plays a protective role against DSS-induced colitis. By contrast, another study demonstrated that the colitis was alleviated in IL-17A deficient mice (Ito et al., 2008). The reason for this discrepancy is not clear, but might be due to different mouse

Using these models of colitis, a number of studies on helminth effects have been conducted (Table 4). Generally, helminths seem to ameliorate TNBS/DNBS- induced colitis. It is worth noting that even worms that usually cause intestinal pathology by laying a large number of eggs in the mesenteric vein (i.e. *Schistosoma* spp.) have anti-colitic effects (Moreels et al., 2004; Ruyssers et al., 2009; Elliott et al., 2003; Mo et al., 2007; Zhao et al., 2009; Bodammer et al., 2011; Smith et al., 2007). A down-regulation of Th1/Th17-related cytokines and upregulation of anti-inflammatory cytokines are observed in most studies. However, only a few papers have provided direct evidence of the involvement of certain cells or cytokines. For instance, anti-colitic effects of *S.mansoni* (Elliott et al., 2003) and *H.diminuta* (Hunter et al., 2005) were demonstrated to be STAT6-dependent. The latter authors also a showed the anti-colitic effect of the worms to be dependent on IL-10 and the presence of macrophages (Hunter et al., 2010). In contrast to the regularly observed helminths' ameliorating effects on Th1/Th17 dominant (TNBS/DNBS- induced) models, macrophage-mediated or Th2 dominant (DSS-induced or oxazolone- induced) colitis was not alleviated or rather worsened by some helminth infections (Table 4). This exacerbation seems due to the robust Th2-polarization by the helminths; e.g. *H. diminuta* infection worsened oxazolone-induced colitis via IL-5 induction and consequently caused eosinophila (Wang et al., 2010), demonstrated by a loss of exacerbation in eotaxin-deficient mice or anti-IL-5 antibodytreated mice. The finding that anti-colitic effects of schistosome male worms were lost in DSS-induced colitis when mice were infected with worms of mixed sex (Smith et al., 2007) may support the speculation, because the Th2-polarizing ability of schsitosome eggs (produced in mixed sex infection) is more potent than that of adult worms (Grzych et al., 1991). Weinstock's group has been studying anti-colitic mechanisms of the intestinal helminth *H.polygyrus* by using colitis in IL-10-deficient or TGF-βRII dominant negative (DN) mice (Elliott et al., 2004, 2008; Ince et al., 2009). According to the studies, the helminth's anticolitic effect depends not on IL-10 but on TGF-β signaling. This finding does not mean that host-derived TGF-β is necessary, because it has been recently shown that *H. polygyrus*derived "TGF-β-like molecules" mobilize host TGF-β signaling and induce subsequent Treg

strains used in their experiments as indicated by Ito et al.

cell development (McSorley et al., 2010; Grainger et al., 2010).


↓:down-regulation, ↑:up-regulation

Table 3. Suppressive effects of parasitic helminths on experimental arthritis.

In addition to the down-regulation of pro-inflammatory cytokines, pathogenic anti-collagen IgG levels were lowered in the infected mice. Since our publication, other investigators also have demonstrated therapeutic effects of schistosomes on arthritis models. He et al. (2010) showed that *S.japonicum* suppressed CIA. However, infection at 2 weeks before CIA induction resulted in an exacerbation of the disease, whereas infection at 7 weeks before induction prevented the disease. This is in contrast to our study in which infection at 2 weeks before induction reduced the severity of CIA. The reason of this discrepancy is not clear, but as speculated by He et al., the difference in parasite species (*S.mansoni* and *S.japonicum*) might have affected the outcome. Sun et al. (2010) demonstrated that a recombinant protein of *S. japonicum* (rSj16) ameliorated adjuvant-induced arthritis in rats. The authors observed a suppression of IL-12 production and augmentation of IL-10 production in rSj16-treated bone marrow-derived dendritic cells (BMDCs). A filarial nematode-derived phosphorylcholine- containing glycoprotein, ES-62, was proven to suppress ongoing CIA (McInnes et al., 2003). In vitro, ES-62 suppressed LPS-induced production of TNF-α and IL-6 from RA synovial cells. Moreover, ES-62 also inhibited TNF-α production in human T cells and macrophages. Regarding the involvement of regulatory cytokines and cells, Shi et al. (2011) reported that IL-10 from CD4+ cells of infected mice was important for anti-arthritic effects of the rat tapeworm *Hymenolepis diminuta*. They also found that the absence of IL-4Rα chain signaling in mice cancelled the anti-arthritic effects of the worm. These findings seem to suggest the importance of Th2-polarization by the parasite, however, we recently found that the Th2-polarizing intestinal parasite *H. polygyrus* did not protect mice from CIA but rather, exacerbated the disease (unpublished observation). The anti-arthritic mechanisms of helminths seem complicated, probably because of the heterogeneity of both experimental models and parasites.

#### **7.4 Experimental colitis as a model of IBD**

Human IBD includes heterogeneous inflammatory diseases of the intestines. For instance, in terms of T cell subset, CD had been considered a "Th1 disease", but at present, the importance of the Th17 subset in the pathogenesis of CD is emphasized (Sarra et al., 2010).

*Acanthocheilonema viteae* ES-62 i.p. IFN-↓, TNF-↓, IL-6↓, Anti-

*Schistosoma mansoni* Infection IL-17↓, TNF-↓, IL-6↓, RANKL

*Ascaris suum* Worm Ag i.p. Rocha et al., 2008

*Schistosoma japonicum* Infection IL-4↑, Anti-collagen IgG↓ He et al., 2010

*Ascaris suum* Worm Ag i.p./p.o. NO↓, IL-1↓ Rocha et al., 2008

*Schistosoma japonicum* rSj16 i.p. TNF-↓, IL-1↓,NO↓, IL-10↑ Sun et al., 2010

*Hymenolepis diminuta* Infection IL-10 from CD4+ cells Shi et al., 2011

In addition to the down-regulation of pro-inflammatory cytokines, pathogenic anti-collagen IgG levels were lowered in the infected mice. Since our publication, other investigators also have demonstrated therapeutic effects of schistosomes on arthritis models. He et al. (2010) showed that *S.japonicum* suppressed CIA. However, infection at 2 weeks before CIA induction resulted in an exacerbation of the disease, whereas infection at 7 weeks before induction prevented the disease. This is in contrast to our study in which infection at 2 weeks before induction reduced the severity of CIA. The reason of this discrepancy is not clear, but as speculated by He et al., the difference in parasite species (*S.mansoni* and *S.japonicum*) might have affected the outcome. Sun et al. (2010) demonstrated that a recombinant protein of *S. japonicum* (rSj16) ameliorated adjuvant-induced arthritis in rats. The authors observed a suppression of IL-12 production and augmentation of IL-10 production in rSj16-treated bone marrow-derived dendritic cells (BMDCs). A filarial nematode-derived phosphorylcholine- containing glycoprotein, ES-62, was proven to suppress ongoing CIA (McInnes et al., 2003). In vitro, ES-62 suppressed LPS-induced production of TNF-α and IL-6 from RA synovial cells. Moreover, ES-62 also inhibited TNF-α production in human T cells and macrophages. Regarding the involvement of regulatory cytokines and cells, Shi et al. (2011) reported that IL-10 from CD4+ cells of infected mice was important for anti-arthritic effects of the rat tapeworm *Hymenolepis diminuta*. They also found that the absence of IL-4Rα chain signaling in mice cancelled the anti-arthritic effects of the worm. These findings seem to suggest the importance of Th2-polarization by the parasite, however, we recently found that the Th2-polarizing intestinal parasite *H. polygyrus* did not protect mice from CIA but rather, exacerbated the disease (unpublished observation). The anti-arthritic mechanisms of helminths seem complicated, probably

collagen IgG↓

*Nippostrongylus brasiliensis* Infection Salinas-Carmona et al., 2009

↓, Anti-collagen IgG↓

McInnes et al., 2003

Osada et al., 2009

Arthritis model Helminth Treatment Proposed mechanism Refs

Table 3. Suppressive effects of parasitic helminths on experimental arthritis.

because of the heterogeneity of both experimental models and parasites.

Human IBD includes heterogeneous inflammatory diseases of the intestines. For instance, in terms of T cell subset, CD had been considered a "Th1 disease", but at present, the importance of the Th17 subset in the pathogenesis of CD is emphasized (Sarra et al., 2010).

**7.4 Experimental colitis as a model of IBD** 

*Heligmosomoides polygyrus* Infection

Collagen-induced arthritis (CIA) in mice

Zymosan-induced arthritis (ZYA) in mice/rats

Adjuvant-induced arthritis (AIA) in rats

Spontaneous arthritis in MRL/lpr mice

↓:down-regulation, ↑:up-regulation

FCA-induced monoarthritis in rats Increases in both IL-17 and IL-12 mRNA in CD and UC patients has been reported (Nielsen et al., 2003). The most striking evidence of the involvement of IL-23/Th17 axis is the finding of Th17-related genes (including IL-23R gene) as susceptibility genes for CD (Brand, 2009). In contrast, Th2-related cytokines are dominant in UC (Sarra et al., 2010). Likewise, models of colitis are also composed of diseases with a distinct pathogenesis (Strober et al., 2002). Experimental colitis induced by the intra-rectal injection of hapten, such as TNBS or DNBS, resembles human CD and its development seems independent of IFN-γR (Camoglio et al., 2000) and dependent on IL-17R signaling (Zhang et al., 2006). Colitis can be also induced by causing dysfunction in epithelial cell barrier. Supply of drinking water containing dextran sulfate sodium (DSS) induces this type of colitis. In this model, T and B cells are dispensable (Dieleman et al., 1994) and macrophages seem to play both a pathological role (Bauer et al., 2010) and a regulatory role (Qualls et al., 2006; Smith et al., 2007). Considering predominant expression of Th2-related cytokines in the chronic phase, DSS-induced colitis resembles UC (Alex et al., 2009). Interestingly, neutralization of IL-17 aggravates the model (Ogawa et al., 2004), suggesting that IL-17 plays a protective role against DSS-induced colitis. By contrast, another study demonstrated that the colitis was alleviated in IL-17A deficient mice (Ito et al., 2008). The reason for this discrepancy is not clear, but might be due to different mouse strains used in their experiments as indicated by Ito et al.

Using these models of colitis, a number of studies on helminth effects have been conducted (Table 4). Generally, helminths seem to ameliorate TNBS/DNBS- induced colitis. It is worth noting that even worms that usually cause intestinal pathology by laying a large number of eggs in the mesenteric vein (i.e. *Schistosoma* spp.) have anti-colitic effects (Moreels et al., 2004; Ruyssers et al., 2009; Elliott et al., 2003; Mo et al., 2007; Zhao et al., 2009; Bodammer et al., 2011; Smith et al., 2007). A down-regulation of Th1/Th17-related cytokines and upregulation of anti-inflammatory cytokines are observed in most studies. However, only a few papers have provided direct evidence of the involvement of certain cells or cytokines. For instance, anti-colitic effects of *S.mansoni* (Elliott et al., 2003) and *H.diminuta* (Hunter et al., 2005) were demonstrated to be STAT6-dependent. The latter authors also a showed the anti-colitic effect of the worms to be dependent on IL-10 and the presence of macrophages (Hunter et al., 2010). In contrast to the regularly observed helminths' ameliorating effects on Th1/Th17 dominant (TNBS/DNBS- induced) models, macrophage-mediated or Th2 dominant (DSS-induced or oxazolone- induced) colitis was not alleviated or rather worsened by some helminth infections (Table 4). This exacerbation seems due to the robust Th2-polarization by the helminths; e.g. *H. diminuta* infection worsened oxazolone-induced colitis via IL-5 induction and consequently caused eosinophila (Wang et al., 2010), demonstrated by a loss of exacerbation in eotaxin-deficient mice or anti-IL-5 antibodytreated mice. The finding that anti-colitic effects of schistosome male worms were lost in DSS-induced colitis when mice were infected with worms of mixed sex (Smith et al., 2007) may support the speculation, because the Th2-polarizing ability of schsitosome eggs (produced in mixed sex infection) is more potent than that of adult worms (Grzych et al., 1991). Weinstock's group has been studying anti-colitic mechanisms of the intestinal helminth *H.polygyrus* by using colitis in IL-10-deficient or TGF-βRII dominant negative (DN) mice (Elliott et al., 2004, 2008; Ince et al., 2009). According to the studies, the helminth's anticolitic effect depends not on IL-10 but on TGF-β signaling. This finding does not mean that host-derived TGF-β is necessary, because it has been recently shown that *H. polygyrus*derived "TGF-β-like molecules" mobilize host TGF-β signaling and induce subsequent Treg cell development (McSorley et al., 2010; Grainger et al., 2010).

Parasitic Helminths as Potential Therapeutic Agents Against Autoimmune Disorders 603

There are two ways of developing parasite-based biomedicines for clinical use. One approach is the direct applications of non-pathogenic/hypo-virulent viable helminths to patients, as introduced in section 8. In addition to TSO and hookworms, other hypo-virulent helminths could be considered for human application. However, before clinical trials, sufficient accumulation of epidemiological and experimental evidence of their therapeutic efficacy is required. Hypo-virulent intestinal nematodes (e.g. *Trichostrongylus* spp.), intestinal trematodes (e.g. *Metagonimus* sp.) and intestinal tapeworms (e.g. *Hymenolepis diminuta*) may become candidates for such studies in the future. It is also essential that the parasites can be maintained in domestic or experimental animals. This is because parasites that infect only humans cannot be maintained and expanded efficiently for clinical use. Another way of developing parasite-based biomedicines comes from the identification of effector molecules of parasites. Considerable numbers of immunomodulatory molecules have been identified from helminths (Harnett W & Harnett MM, 2010). The majority have shown therapeutic effects on experimental autoimmunity or allergy. Some investigators reported that viable parasites were superior to administration of the antigens of parasites (Hunter et al., 2010; Bodammer et al., 2011; Osada et al., 2010) in therapeutic efficacy. In addition, there is still considerable controversy over the roles of regulatory cells (e.g. Treg, Breg or AAMΦ) and regulatory cytokines (e.g. IL-4, IL-10, TGF-β) in helminth-induced immunomodulation. Therefore, further investigation is needed to elucidate the immunomodulatory mechanisms of viable parasite infections, and new findings obtained there should help to establish an optimal screening system for anti-autoimmune/anti-

Our studies introduced in this chapter were supported by Grants-in-aid for Scientific Research from the Japan Society for the Promotion of Science (nos. 20590435, 22590386) and Health and Labour Sciences Research Grants from the Ministry of Health, Labor and Welfare of Japan (H22-Kokui-Shitei-004). I am grateful to Prof. Tamotsu Kanazawa and other laboratory members for their valuable scientific advice. I also acknowledge the technical assistance of Ms. Yukari Horie and the clerical assistance of Ms. Tomoko Fujitomi.

Alex, P., Zachos, N.C., Nguyen, T., Gonzales, L., Chen, T.E., Conklin, L.S., Centola, M. & Li,

Aoyama, H., Hirata, T., Sakugawa, H., Watanabe, T., Miyagi, S., Maeshiro, T., Chinen, T.,

Ascherio, A. & Munger, K. L. (2010). 99th Dahlem conference on infection, inflammation and

DSS and TNBS-induced colitis. *Inflamm Bowel Dis.* 15 (3):341-352.

epidemiological evidence. *Clin Exp Immunol*. 160 (1):120-124.

infection. *Am J Trop Med Hyg.* 76 (5):972-976.

X. (2009). Distinct cytokine patterns identified from multiplex profiles of murine

Kawane, M., Zaha, O., Nakayoshi, T., Kinjo, F. & Fujita, J. (2007). An inverse relationship between autoimmune liver diseases and *Strongyloides stercoralis*

chronic inflammatory disorders: Epstein-Barr virus and multiple sclerosis:

**9. Concluding remarks** 

allergic substances from parasitic helminths.

**10. Acknowledgements** 

**11. References** 


Table 4. Effects of parasitic helminths on experimental colitis.

#### **8. Clinical trials of parasitic helminths against immunological disorders**

The administration of non-pathogenic or hypo-virulent parasitic worms could be considered for the treatment of immunological disorders. Several clinical trials using parasitic worms have been and are currently being conducted. Significant therapeutic effects have been confirmed in some of these studies. Weinstock's group conducted trials with *Trichuris suis* (porcine whipworm) ova (TSO) against CD (Summers et al., 2005a) and UC (Summers et al., 2005b), and demonstrated significant efficacy. TSO is also being tested for MS and promising results have been obtained in a phase I trial (Fleming et al., 2011). Regarding allergic disorders, Bager et al. (2010) found no therapeutic effect of TSO on allergic rhinitis. However, Summers et al. (2010) critically commented on the report that it was premature to conclude that TSO is ineffective on allergic rhinitis because the TSO treatment was too late and not sufficient. *Necator americanus* (Hookworm) is also under clinical trials for asthma (Feary et al., 2010) and CD (Croese et al., 2006). The advantage of this worm is its long life in the host (at least 6 years) and no need for repeated inoculation (Elliott and Weinstock, 2009). The parasite was well-tolerated without severe adverse effects on asthmatic patients, but a safe dosage of the parasites (10 infective larvae) did not show significant therapeutic efficacy (Feary et al., 2010).
