The intimate interactions between hosts and parasites have been studied for many years . The ability of helminths, in particular T. muris to affect the immune environment within the host has a broad associated literature [23–25]. Genome wide approaches to understand the changes occurring in the host at a gene expression level have been utilised in many infectious models and diseases [26, 27]. Indeed microarray and RT-PCR analysis have been used to explore the responses of mouse strains infected with T. muris [14, 15]. However, these experiments use different mouse strains in order to compare the gene expression profiles associated with resistance or susceptibility. Here we use one mouse strain, C57BL/6, and different T. muris isolates to achieve susceptibility or resistant outcomes. In addition we look early in infection to identify changes in gene expression, that could later determine susceptibility or resistance of the host.
We demonstrate that the S isolate of T. muris, which is able to persist chronically in a C57BL/6 host, provokes a different gene expression profile in gut tissue compared to a C57BL/6 mouse infected with the E isolate, which is ultimately expelled. These changes are already apparent by d7 p.i. prior to E isolate expulsion. In particular, differences in 43 probe sets associated with GO groups - immune response, defence response, antigen presentation and response to stress were identified.
Interestingly the expression of genes from an S isolate infected gut resemble those seen in a naïve (uninfected) gut, whereas there was an elevation in the expression of the genes represented by the 43 probe sets during an E isolate infection. This data would imply that even at d7 p.i., the S isolate is dampening down expression of genes that would normally be associated with expulsion of the parasite. Microarray analyses on gut tissue from different strains of mice infected with the E isolate had previously identified differences in the expression of INDO and Ang4. As these two genes were also identified as being differentially expressed in this current study we investigated their expression further.
Indoleamine 2,3-dioxygenase plays an important part in the kynurenine pathway where it metabolises tryptophan which has been associated with controlling parasite growth in a number of infections [28–30]. IDO also has immunoregulatory roles [14, 31, 32], indeed its ability to dampen cell proliferation  may extend to the control of gut epithelial cell turnover and thus delayed or absent worm expulsion given the importance of epithelial cell turnover in the expulsion of T. muris . As the expression of IDO was elevated in gut tissue from both E and S isolate-infected mice its presence may reflect a less efficient ability, rather than a complete inability, to expel the parasite.
In contrast, Ang4 expression did differentiate between worm survival and expulsion with expression levels only elevated in E-infected mice. The angiogenins are a family of closely related proteins described as Paneth cell-derived and encoded for, in the mouse, by a gene cluster on chromosome 1 . They belong to the RNase superfamily  which includes eosinophil secretory granules proteins which are toxic to some gastrointestinal nematodes . Originally implicated in the growth of tumours, the normal physiological role of the angiogenins was unclear until the demonstration of a role for Paneth cell derived Ang4 as an endogenous anti-microbial protein central in epithelial host defence against gut-dwelling bacteria . Immunohistochemical staining of gut tissue in this study localised Ang4 production to the goblet cell, identifying the cell type as a novel cellular source in the large intestine. This interesting discovery expands on the already extensive literature on the importance of goblet cells in the expulsion of nematode infection, where increased numbers of goblet cells correlate with resistance [37, 38].
A goblet cell hyperplasia is reported in the context of many nematode infections including T. muris, T. spiralis, H. polygyrus and N. brasiliensis and this is thought to be under the control of a Th2 response [37–39]. Several goblet cell factors have been identified to play an important role in nematode infection including Relm-β and Muc2. Relm-β expression is linked with the production of Th2 cytokines and worm expulsion in T. muris, T. spiralis and N. brasiliensis . Thus effective worm expulsion may be achieved through combined effects of antimicrobial proteins including the angiogenins, Relm-β and indeed the intelectins [14, 41].
Goblet cell numbers did not differ quantitatively between E and S isolate infection until d21 p.i. However, the goblet cells were qualitatively strikingly different at d7 p.i. Thus the number of neutral goblet cells present in the gut of an S isolate infected C57BL/6 mouse are significantly higher than that seen in an E isolate infected mouse. By d21 the number of neutral goblet cells is similar for both isolates, if not higher in the E isolate infection. At this time point the majority of E isolate worms have been expelled.
The acidity of goblet cells is determined by the addition and removal of sugars to the mucins within the goblet cell . Transferases are required to do this and Table 2 highlights candidate transferases that may explain the differences seen in the quality of the goblet cells. Increased gene expression of the transferases is seen in the microarray data in the gut of an E isolate-infected mouse compared to an S isolate-infected gut. The increases in gene expression of the transferases listed could explain the increased acidity of the goblet cell mucins seen in an E isolate infection and may even underlie subsequent expulsion.