Enhanced Susceptibility of Galectin-1 Deficient Mice to Experimental Colitis

Galectin-1 is a β-galactoside-binding lectin, ubiquitously expressed in stromal, epithelial, and different subsets of immune cells. Galectin-1 is the prototype member of the galectin family which shares specificity with β-galactoside containing proteins and lipids. Immunomodulatory functions have been ascribed to endogenous galectin-1 due to its induction of T cell apoptosis, inhibitory effects of neutrophils and T cell trafficking. Several studies have demonstrated that administration of recombinant galectin-1 suppressed experimental colitis by modulating adaptive immune responses altering the fate and phenotype of T cells. However, the role of endogenous galectin-1 in intestinal inflammation is poorly defined. In the present study, the well-characterized acute dextran sulfate sodium (DSS)-induced model of ulcerative colitis was used to study the function of endogenous galectin-1 during the development of intestinal inflammation. We found that galectin-1 deficient mice (Lgals1−/− mice) displayed a more severe intestinal inflammation, characterized by significantly elevated clinical scores, than their wild type counterparts. The mechanisms underlying the enhanced inflammatory response in colitic Lgals1−/− mice involved an altered Th17/Th1 profile of effector CD4+ T cells. Furthermore, increased frequencies of Foxp3+CD4+ regulatory T cells in colon lamina propria in Lgals1−/− mice were found. Strikingly, the exacerbated intestinal inflammatory response observed in Lgals1− / − mice was alleviated by adoptive transfer of wild type Foxp3+CD4+ regulatory T cells at induction of colitis. Altogether, these data highlight the importance of endogenous galectin-1 as a novel determinant in regulating T cell reactivity during the development of intestinal inflammation.


INTRODUCTION
Galectins are among the best-characterized glycan-binding proteins that play key roles in multiple immune cell processes (1). They formed a family of 15 mammalian galactoside-binding proteins that exhibit a consensus amino acid sequence in their carbohydrate recognition domains (CRDs) (2). Among these, galectin-1, which is considered as a prototype member of the galectin family and is a homodimer of 14.5 kDa that plays pivotal roles in modulating acute and chronic inflammatory responses. Galectin-1 is ubiquitously expressed by stromal, epithelial, and endothelial cells from different tissues. Within the immune system myeloid cells like macrophages, as well as subsets of B and T cells including regulatory T cells (Tregs), have been described to express galectin-1 (3)(4)(5).
Perillo et al. first demonstrated that expression of galectin-1 in the thymic stroma contributed to deletion of autoreactive immature thymocytes. In addition, recombinant galectin-1 induced apoptosis of activated human T cells and human leukemia T cell lines (6,7). Since then, galectin-1 has been described as controlling T cell survival by binding with CD45, CD43, CD7, and even with CD69, all well-known glycosylated membrane receptors, triggering their clustering and polarization in microdomains that induced signaling and possibly transcriptional reprogramming that still remained undefined (7)(8)(9)(10)(11). In transcriptomic and proteomic analyses, we found that galectin-1 was significantly expressed and secreted by mouse and human activated Treg cells. The in vitro studies showed that the immunosuppressive activity of Treg cells was blocked by the use of anti-galectin-1 neutralizing antibodies. In the spleen, although the Treg frequencies in Lgals1 −/− mice were similar to wild type mice, their suppressive potential ex vivo was compromised (5). Rabinovich and collaborators have extensively demonstrated that galectin-1 can specifically control the proliferation of Th17, Th1, and cytotoxic CD8 + effector T cell responses but not on naïve, Th2 or regulatory T cells, thanks to an activation-specific repertoire of glycans expressed in their cell membrane glycoproteins that are essential for galectin-1 specific binding (12,13).
In addition to this, acute inflammatory responses can be modulated by recombinant galectin-1 through different mechanisms such as inhibition of neutrophil adhesion and replacement (14,15) and by modulating maturation and tissue migration of monocytes, dendritic cells, and macrophages (16,17). Effector T cell responses are modulated by galectin-1 through an array of mechanisms including control of T cell activation, differentiation, secretion of cytokines, and induction of apoptosis (6,(18)(19)(20)(21)(22). In recent years, the essential role of recombinant galectin-1 in the control of inflammation has been widely demonstrated in experimental studies for treatment of various diseases, including autoimmune encephalomyelitis (12,13), arthritis (23,24), hepatitis (25), autoimmune diabetes (26,27), and ulcerative colitis (28). In addition to this, mice deficient in the Lgals1 gene displayed increased susceptibility to develop autoimmune-like diseases such as autoimmune encephalomyelitis, arthritis, and lupus (1). Recently, a spontaneous inflammatory process in salivary glands that resembles Sjögren's syndrome has been described in aged Lgals1 −/− mice (29). Nevertheless, pieces of evidence of the regulatory effects of galectin-1 in in vivo models of intestinal inflammation are still scarce and the mechanisms underlying the immunosuppressive effects of endogenous galectin-1 have not been fully elucidated (1,30).
In the present study, the acute dextran sulfate sodium (DSS)induced model of colitis was used to study the function of endogenous galectin-1 during the development of intestinal inflammation. Mice deficient in the Lgals1 gene displayed an increase susceptibility to developing colitis. Colitic Lgals1 −/− mice exhibited an altered Th17/Th1 profile of effector CD4 + T cells and Foxp3 + CD4 + regulatory T cells (Treg cells) in mesenteric lymph nodes (mLNs) and colon lamina propria (cLP) in colitic Lgals1 −/− mice. Interestingly, the exacerbated intestinal inflammatory response observed in Lgals1 −/− mice was alleviated by adoptive transfer of wild type Foxp3 + CD4 + regulatory T cells at induction of colitis. Altogether, these findings contribute to delineate the mechanisms underlying the immunosuppressive role of endogenous galectin-1 during intestinal inflammation.

Mice
B6.Cg-Foxp3 tm2Tch /J (Foxp3 EGFP+ mice) and B6.Cg-Lgals1 tm1Rob /J (Lgals1 −/− mice) were purchased from the Jackson laboratory (stock numbers 006772 and 006337, respectively). Foxp3 EGFP+ mice were backcrossed for up to seven generations with Lgals1 −/− mice to generate Lgals1 −/− mice expressing the EGFP gene under the endogenous Foxp3 transcription factor promoter (Foxp3 EGFP+ Lgals1 −/− mice) on C57BL/6J background. All experiments were performed using these two mouse strains. For regulatory T cell adoptive transfer experiments, C57BL/6J (stock number 000664) and B6.Cg-Lgals1 tm1Rob /J were used as donors. All experiments were performed in accordance with the corresponding regulations regarding experimental animal welfare (RD 223/1998 and Directive 2010/63/EU protocols). The experimental protocol was reviewed and approved by the ethics committee for animal research at the CIEMAT and Comunidad de Madrid (based on the RD 53/2013).

Colitis Induction
Experimental colitis was induced by the oral administration of dextran sulfate sodium (DSS) (36,000-50,000 MW; MP Biochemicals; Irvine, CA, USA) in drinking water for 7 days ad libitum. The degree of intestinal inflammation achieved was induced according to the concentration of DSS used (1.25% DSS; w/v). Body weight and health status of mice were monitored daily and the degree of intestinal inflammation induced was scored according to the following scale (1): Body weight loss with respect to day 0 expressed as percentage (0: no loss; 1: 0-5%; 2: 5-10%; 3: 10-20%, 4: >20% loss of weight and 5: no survival) (2); stool consistency (0: normal stools; 1: loose stools; 2: watery diarrhea; 3: watery diarrhea with blood and 4: no survival) and (3) general activity (0: normal; 1-2: moderate activity; 3: null activity and 4: no survival). Transfer   CD4 + T cells were isolated from spleens using Mojosort Mouse CD4  T Cell isolation kit according to the manufacturer's instructions (ref.  480033; BioLegend, San Diego, CA). Regulatory T cells (Treg cells) were enriched from total CD4 + T cells by cell sorting based on CD25 high CD127 low/− antibody expression using a BD Influx Cell Sorter (BD Biosciences). The purity of the sorted T cell populations was above 96%. For Treg cells' adoptive transfer experiments, one million CD25 high CD127 low/− CD4 + Treg cells isolated from wild type C57BL/6J or B6.Cg-Lgals1 tm1Rob /J mice Lgals1 −/− mice were infused intraperitoneally into Foxp3 EGFP+ Lgals1 −/− mice at day 0, following 1.25% DSS administration in drinking water ad libitum for seven days.

Flow Cytometry
Cells from spleen (SPL) and mesenteric lymph nodes (mLNs) were mechanically dissociated with a 70-mm strainer.

Histology
Colon tissues were fixed in formaldehyde for 24 h, washed in phosphate-buffered saline (PBS), and embedded in paraffin. For histological evaluation hematoxylin and eosin (H&E) colon cross-sections were performed in order to analyze tissue structure, cell infiltrations, edema, and general tissue damage.

Semiquantitative Real-Time PCR
RNA from mLNs and colon biopsies was isolated using TriReagent method (Sigma Aldrich). RNA, 2 µg, was reverse transcribed using a High Capacity cDNA Reverse Transcription Kit (Life Technologies). Relative gene expression was determined by semiquantitative RT-PCR analysis using TaqMan Gene Expression Master Mix (Applied Biosystems, Waltham, MA, USA) ( Table 2). Each sample was normalized with b-actin using the formula 2−DCt. Amplification was performed in a 7900HT Fast Real-Time PCR System apparatus (Applied Biosystems).

Cytokines in Plasma
Peripheral blood was collected on the day of sacrifice. The samples were centrifuged at 12,000 rpm for 10 min, and plasma was stored at −20°C. Cytokine concentrations were measured using a mouse LEGENDplex multiplex assay (LEGENDplex by Biolegend, 8999 BioLegend Way, San Diego, CA 92121) according to manufacturer's instructions. Samples were acquired on a BD LSRFortessa cell analyzer (BD Biosciences).

Statistical Analysis
Results are expressed as mean, and error bars represent standard error of the mean (SEM). Normal distribution was analyzed by Shapiro-Wilk tests. Then, experimental groups were statistically evaluated using the non-parametric unpaired Mann-Whitney U test. 'P' values that were considered statistically significant were p <0.05. (*; # p < 0.05, **; ##p < 0.01, ***p < 0.001 and ****p < 0.0001). Survival curves were represented using Kaplan-Meier curves, and differences in experimental group survival were calculated by the long-rank (Mantel-Cox) test. Statistical analysis was performed with GraphPad Prism 9.

Absence of Endogenous Galectin-1 Leads to Increased Severity of DSS-Induced Colitis
To investigate whether endogenous galectin-1 plays a role in experimental colitis, wild type (WT) and Lgals1 −/− mice were

Gene
GenBank accession no. Assay ID (TaqMan ® ) treated with DSS in drinking water for 7 days ad libitum. A significantly increased severity of DSS-induced colitis was observed in Lgals1 −/− mice as shown by the increased body weight loss and increased disease activity index (DAI) of these mice with respect to WT mice ( Figures 1A, B). In addition, survival of Lgals1 −/− mice was dramatically reduced as compared to their WT counterparts ( Figure 1C). Colon length was also significantly reduced in colitic Lgals1 −/− mice with respect to colitic WT mice ( Figure 1D). These results were further confirmed by histological evaluation of H&Estained colon cross-sections where a more severe colitis characterized by increased edema, denser transmural leukocyte infiltration, and increased mucosal erosion was observed in DSStreated Lgals1 −/− mice with respect to colitic WT mice ( Figure 1E). Plasma levels of interleukin 6 (IL-6) were also significantly increased in colitic Lgals1 −/− mice compared to colitic WT mice ( Figure 1F). The enhanced susceptibility of Lgals1 −/− mice to developing DSS-induced colitis with respect to WT mice suggests that endogenous galectin-1 modulates intestinal inflammatory responses.

Increased Plasticity of Th17 Cells Toward Th1 Profile in DSS-Induced Colitic
To analyze the Th17/Th1 profile in DSS colitic WT and Lgals1 −/− mice, the expression of mRNA levels of RORgt and T-Bet transcription factors was determined by quantitative real-time PCR in mLNs and colon. As shown in Figure 2A, the RORgt/T-Bet ratio was significantly increased in colitic Lgals1 −/− mice with respect to colitic WT mice in both tissues suggesting that the exacerbated inflammatory response observed in colitic Lgals1 −/− mice is mainly due to an enhanced induction of Th17 cell responses. No differences were determined in the Foxp3/RORgt ratio between colitic WT and Lgals1 −/− mice although a tendency to increase in the colon in colitic Lgals1 −/− mice was noticed ( Figure 2B). In contrast, colitic Lgals1 −/− mice had increased Foxp3/T-Bet ratio ( Figure 2C) together with increased expression of IL10 ( Figure 2D) both in mLNs and colon.
These results indicate that endogenous galectin-1 ameliorates intestinal inflammatory responses mainly by targeting Th17 cell responses in the gut. However, the higher levels of Foxp3 and IL10 mRNAs observed in colitic Lgals1 −/− mice were surprising since these results inversely correlated with the exacerbated inflammatory status of the colitic Lgals1 −/− mice with respect to colitic WT mice ( Figure 1). Effector CD4 + T cells responses were further analyzed in colitic mice by flow cytometry. No clear differences were measured in the frequencies and total number of RORgt + CD4 + T cells between colitic WT mice and colitic Lgals1 −/− mice in mLNs. However, the absolute number of RORgt + CD4 + T cells was significantly reduced in cLP in colitic Lgals1 −/− mice with respect to the WT colitic mice ( Figure 2E). On the other hand, a significant increase in the percentage and absolute number of T-Bet + CD4 + T cells was observed in mLNs in colitic Lgals1 −/− mice compared to colitic WT mice ( Figures 2F, H). Colitic WT mice had a significant increase in the RORgt + /T-Bet + balance within CD4 + T cells when the absolute number of effector RORgt + CD4 + T cells and T-Bet + CD4 + T cells in mLNs and cLP was considered ( Figure 2G). These results may suggest that the exacerbated intestinal inflammatory responses observed in colitic Lgals1 −/− mice occurred as a consequence of enhanced plasticity of responder Th17 cells towards a Th1 pathogenic profile in the intestine. We next examined whether DSS-induced gut inflammation altered the frequency of total CD4 + T cells as well as Foxp3 + CD4 + Treg cells in different tissues. To do this, leukocytes isolated from peripheral blood (PB), spleen, mLNs, and cLP were analyzed by flow cytometry in healthy and DSS-colitic mice. No differences were observed in the frequency of CD4 + T cells in WT and Lgals1 −/− mice in healthy and DSS inflamed mice. Interestingly, increased frequencies of Foxp3 + CD4 + Treg cells were observed in cLP in healthy Lgals1 −/− mice with respect to healthy WT mice ( Figures 3A, B). DSS-induced colitis was associated with increased percentages of Foxp3 + CD4 + Treg cells in cLP in both WT mice and Lgals1 −/− mice with respect to their control non-inflamed counterparts ( Figures 3A, B right; D). No differences in Foxp3 frecuencias were found in other organs, such as mLNs between both genotypes ( Figure 3C). Conversely, no significant differences were found in the, frequencies of IL10 + CD4 + T cells in colitic WT and Lgals1 −/− mice in mLNs or in cLP ( Figures 4A, B). The Foxp3 + CD4 + Treg cells (Tregs)/ Foxp3 − CD4 + Teffector cells (Teff) ratio was increased in both WT and Lgals1 −/− mice in PB and cLPs as a consequence of DSSinduced intestinal inflammation ( Figure 3E). These differences were significantly higher in cLP in colitic Lgals1 −/− mice with respect to colitic WT mice. mLNs and cLP GFP-Foxp3 + CD4 + Treg cells from colitic Lgals1 −/− mice exhibited increased proliferative rate as measured by the expression of Ki67 (Figures 4C, D). This was in contrast to Teff cells from both colitic WT and Lgals1 −/− mice that showed similar proliferation rate ( Figures 4C, D). Taken together, these data suggest that the intestinal inflammation induced by treatment with DSS promotes Foxp3 + CD4 + Treg cell proliferation mainly in mLNs and cLP where the inflammatory insult took place. The proliferative response to the inflammatory challenge was significantly increased in Foxp3 + CD4 + Treg cells in colitic Lgals1 −/− mice compared to colitic WT Foxp3 + CD4 + Treg cells thus leading to an increase in the Tregs/Teff ratio in cLP in colitic Lgals1 −/− mice. Despite this, Lgals1 −/− mice were unable to mount an acute intestinal inflammatory challenge in the gut ( Figure 1). These data suggest a reduced in vivo functionality of Foxp3 + CD4 + Treg cells in colitic Lgals1 −/− mice in response to an acute intestinal inflammation challenge.
Enhanced Conversion of Foxp3 + CD4 + Regulatory T Cells to T-Bet + Foxp3 + CD4 + Treg Cell in the Absence of Endogenous Galectin-1 During the

Development of Colitis
Numerous studies have demonstrated that the differentiation program of Foxp3 + CD4 + Treg cells is not fixed and in vivo a fraction of Treg cells can differentiate into Th17-and Th1-like cells during the course of Th17 and Th1 inflammatory responses, respectively. Therefore, the expression of RORgt and T-Bet were analyzed by flow cytometry within the Foxp3 + CD4 + Treg cells in colitic WT and colitic Lgals1 −/− mice. No differences were observed in the frequencies of RORgt + Foxp3 + CD4 + Treg cells in colitic WT and colitic Lgals1 −/− mice. However, the absolute number of RORgt + Foxp3 + CD4 + Treg cells in cLP in colitic Lgals1 −/− mice was significantly reduced with respect to colitic WT mice ( Figure 5A). Strikingly, the percentage of T-Bet + Foxp3 + CD4 + Treg cells were significantly increased in mLNs and cLP in colitic Lgals1 −/− mice compared to colitic WT mice (Figures 5B, D). As a consequence, the RORgt/T-Bet ratio within the Treg cells were significantly increased in colitic WT mice with respect to colitic Lgals1 −/− mice ( Figure 5C). These results suggest that an adequate expression of endogenous galectin-1 prevents the conversion of Foxp3 + CD4 + Treg cells into T-Bet-expressing Foxp3 + CD4 + T cells thus modulating more effectively acute inflammatory responses in vivo.  Figure 6, WT/Tregs-treated colitic Lgals1 −/− mice showed a similar decreased in body weight to colitic WT mice and was significantly reduced with respect to colitic Lgals1 −/− mice and KO/Tregs-treated colitic Lgals1 −/− mice ( Figure 6A). The disease activity index of colitic WT and WT/Tregs-treated colitic Lgals1 −/− mice were significantly reduced compared to untreated colitic Lgals1 −/− mice and KO/Tregs-treated colitic Lgals1 −/− mice ( Figure 6B). Similarly to colitic WT mice, all WT/Tregs and KO/Tregs-treated colitic Lgals1 −/− mice survived upon treatment with DSS. The significant differences seen in colon length between colitic WT mice and colitic Lgals1 −/− mice were not observed in colitic WT/Tregs treated colitic Lgals1 −/− mice. This is in contrast to the colon lengths of KO/Tregs-treated colitic Lgals1 −/− mice that were similar to untreated colitic Lgals1 −/− mice ( Figure 6C). Interestingly, WT/Tregs-treated colitic Lgals1 −/− mice showed similar histopathology signs of intestinal inflammation to colitic WT mice ( Figure 1E) with reduced leukocyte infiltration, decreased edema, and mucosal erosion with respect to colitic Lgals1 −/− mice. This is in contrast to KO/Tregs-treated colitic Lgals1 −/− mice where a reduced leukocyte infiltration with respect to colitic Lgals1 −/− mice was noticed ( Figure 6D). The expression of mRNA levels of Foxp3, RORgt and T-Bet transcription factors and IL10 was determined by quantitative real-time PCR in mLNs and colon. In mLNs, the RORgt/T-Bet ratios were significantly increased in colitic Lgals1 −/− mice with respect to colitic WT mice and colitic Lgals1 −/− mice treated with either WT/Tregs or KO/Tregs. However, in colon the RORgt/T-Bet ratios were similar in untreated and WT and KO/Tregstreated colitic Lgals1 −/− mice and significantly higher than in colitic WT mice ( Figure 6E). Interestingly, the increased Foxp3/ T-Bet and Foxp3/RORgt ratios in mLNs and colon in colitic Lgals1 −/− mice were significantly reduced upon adoptive transfer of WT/Tregs in colitic Lgals1 −/− mice but not with KO/Treg cells. for WT and, n = 6-13 for Lgals1 −/− mice per group, respectively). Data are representative of three independent experiments. Statistical analysis was done using Mann-Whitney U test. Data are shown as mean ± SEM, *p < 0.05; **p < 0.01, and ****p < 0.0001.
In mLNs the high expression levels of IL10 observed in colitic Lgals1 −/− mice were significantly reduced upon transfer of WT/ Tregs and KO/Tregs and similar to the expression levels in colitic WT mice. In the colons in KO/Tregs-treated colitic Lgals1 −/− mice the expression of IL10 was significantly increased with respect to colitic WT mice and WT/Tregs-treated colitic Lgals1 −/− mice, whereas no differences were determined when compared to colitic Lgals1 −/− mice ( Figure 6E).
Adoptive transfer of WT/Tregs was sufficient to modulate the enhanced severity of DSS-induced colitis in observed in Lgals1 −/− mice in contrast to what was observed in colitic Lgals1 −/− mice treated with KO/Treg cells.

DISCUSSION
The intestine is a distinct microenvironment with a rather unique physiological functions since it is constantly exposed to diverse environmental challenges. Host adaptation in the gut is considered an active process that involves critical interactions between different subsets of immune cells and tissue cells. Foxp3 + CD4 + regulatory T cells are instrumental in enforcing homeostasis at the intestinal barrier surface as illustrated by the development of intestinal inflammation in diseases caused by a primary deficiency in the number or function of Treg cells in the gut (31).
The role of galectin-1 in inflammation has been studied in several in vitro and in vivo experimental models (23,(32)(33)(34)(35). Exogenous administration of galectin-1 affects the viability, proliferation, and Th17 and Th1 responses of non-malignant T cells involved in the progression of cutaneous T-cell lymphoma (36). In homeostasis, Lgals1 −/− mice develop normally and do not exhibit any evident alteration compared to galectin-1 sufficient mice (37).
In this study, the acute dextran sulfate sodium (DSS)-induced model of colitis was used to study the function of endogenous galectin-1 during the development of intestinal inflammation. Mice deficient in the Lgals1 gene displayed an increase susceptibility to developing colitis. We analysed total CD4 + T cells and Foxp3 + CD4 + T cells in different tissues in steady state and during DSS-induced acute inflammation. Our data demonstrate that total CD4 + T cell infiltration in spleen, mLNs and cLP were similar in both mouse genotypes. Interestingly, in the absence of galectin-1 increased frequencies of Foxp3 + CD4 + Treg cells were noticed both in non-inflamed and DSS colitic Lgals1 −/− mice compared to WT mice in cLP. Furthermore, no differences were observed in the frequencies of IL10 + CD4 + T cells between both mouse genotypes thus ruling out involvement differences in peripherally induced regulatory CD4 + T cells. These results were consistent with a previous study where the frequency of IL10-expressing Treg cells in DSS-treated mice was similar between mouse strains with different susceptibilities to colitis development (38).
The Treg/Teff balance in colitic mice was significantly increased in cLP in colitic Lgals1 −/− mice. However, Lgals1 −/− mice exhibited an exacerbated inflammatory response to DSSinduced colitis as shown by the enhanced total body losses and disease activity indexes that ultimately led to significantly reduced survival of colitic Lgals1 −/− mice compared to WT mice. It is of interest that the proliferative rate of CD4 + effector T cells was similar in colitic WT and Lgals1 −/− mice as evidenced by Ki67 staining. Conversely, in mLNs and cLP the proliferation rate of Foxp3 + CD4 + Treg cells in colitic Lgals1 −/− mice was higher than in colitic WT mice. Our results are in agreement with a previous study conducted in a similar model of acute colitis induced by DSS where the authors demonstrated that Treg cells were required to limit the severity of intestinal inflammation. This was accompanied by an increased number of Treg cells in cLP with higher proliferation rate in both mLNs and in cLP than in effector T cells (39). In IBD patients, increased numbers of Tregs have been found in the inflamed areas of the intestine (40,41) further confirming that intestinal inflammation is associated with increased frequencies of Treg cells in the inflamed intestine.
RORgt/T-Bet ratios in mLNs and colon in colitic Lgals1 −/− mice were increased with respect to colitic WT mice when assayed by RT-PCR. These results were consistent with studies conducted by Toscano et al. were glycosylation-dependent immune modulatory circuits governed by galectin-1 specifically acted on Th17 and Th1 effector T cells (13). As a consequence, no differences were measured in the Foxp3/RORgt ratios in mLNs and cLP in both mouse genotypes, mostly due to the increased in the percentages of Foxp3 + CD4 + Treg cells in mLNs and cLP in Lgals1 −/− mice. Additionally, relative IL10 mRNA expression levels were higher in colitic Lgals1 −/− mice than in colitic WT mice although by flow cytometry the frequencies of IL10 + CD4 + T cells were similar in WT and Lgals1 −/− mice. It could be that an exacerbated IL-10 mediated innate immune response is triggered in colitic Lgals1 −/− mice to counteract the severe inflammation in the intestine.
RORgt/T-Bet ratios within the CD4 + T cells in mLNs and cLP were significantly reduced in colitic Lgals1 −/− mice with respect to colitic WT mice when assayed by flow cytometry. Hence, in the absence of endogenous expression of galectin-1 the increased pathogenic profile of colitic Lgals1 −/− mice compared to WT mice directly correlated with increased frequencies and number of T-Bet + CD4 + T cells in mLNs and cLP leading to a shift in the RORgt/T-Bet balance in colitic Lgals1 −/− mice.
Although considered finally differentiated, numerous studies have shown that Treg cells have certain functional plasticity by acquiring the expression of the master transcription factors that define the T helper cell subsets they are modulating (42,43). This is a very dynamic process that is also accompanied by coexpression of other master transcription factors (42,44). Previous studies have demonstrated that expression of T-Bet by Treg cells is required for adequate modulation of Th1 cell responses. Koch et al. first observed that in an inflammatory environment characterized by high levels of IFNg a subset of T-Bet + Foxp3 + Treg cells emerged. The IFNg-induced T-Bet + Foxp3 + Treg cells upregulated the expression of CXCR3 allowing the preferential migration of the induced CXCR3 + T-Bet + Foxp3 + Treg cells to tissues where a Th1 active response was taking place (44). In our study we show the accumulation of T-Bet + Foxp3 + CD4 + Treg cells in mLNs and cLP in colitic Lgals1 −/− mice. RORgt/T-Bet balance within the Foxp3 + CD4 + Treg cells was significantly reduced in colitic Lgals1 −/− mice as compared to colitic WT mice. Interestingly, Di Giovangiulio et al. recently demonstrated that Th1-like Treg cells sustained intestinal inflammation since Treg-specific T-Bet conditional ablation knockout mice developed milder DSSinduced colitis than their wild type counterparts. The authors claimed that the T-Bet + Foxp3 + CD4 + Treg cells in fact acted as enhancers of Th1 cell responses during the early phases of the inflammation (38). In agreement with this observation, miR146adeficient IFNg + Tregs induced increased Th1 effector responses compared to wild type mice (45). Furthermore, in a lethal model of T. gondii infection, Oldenhove at. al. demonstrated that T-Bet + IFNg + Foxp3 + CD4 + Treg cells were responsible for the exacerbated Th1 immune response in the infected mice (46). Future studies are warranted aiming at fully defining the functionality of in vivo converted T-Bet + Foxp3 + CD4 + Treg cells under certain inflammatory conditions in the gut.
There is an increasing body of evidence on the therapeutic potential of adoptively transferring healthy Treg cells into patients with a wide range of conditions, including IBD and other autoimmune diseases, in an attempt to shift the balance of active inflammation toward a more tolerogenic microenvironment. Cell therapy approaches with Treg cells are thought particularly useful in those pathophysiological conditions where a link of intestinal inflammation is associated with a decline in Treg functionality.
Altogether, these data support the immunomodulatory function of endogenous galectin-1 in DSS-induced colitis. Enhanced Th1 effector T cell responses together with increased frequencies of T-Bet + Foxp3 + Treg cells in colitic Lgals1 −/− mice were involved in the pathophysiological severe phenotype of colitic Lgals1 −/− mice. Adoptive transfer of wild type CD25 high CD127 low/− Foxp3 + CD4 + Treg cells efficiently ameliorated the exacerbated inflammatory response of colitic Lgals1 −/− mice whereas galectin-1 null Treg cells were unable to ameliorate the exacerbated inflammatory intestinal inflammation in colitic Lgals1 −/− mice further supporting the pivotal role played by galectin-1 of CD25 high CD127 low/− Foxp3 + CD4 + Treg cells in modulating the severity of DSS-induced colitis.
Although the precise mechanisms by which mature Tregs utilize galectin-1 during DSS inflammatory challenge are unclear, these data highlight the importance of endogenous galectin-1 as a novel determinant in regulating T cell reactivity during the development of intestinal inflammation. Further studies are warranted to gain more comprehensive understandings in the biological function of galectin-1 in the development, fate and stability of Treg cells in the intestine.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors without undue reservation.

ETHICS STATEMENT
The animal study was reviewed and approved by Ethics committee for animal research at the CIEMAT and Comunidad de Madrid.