| CommuniGate | Scientific Papers On IBDs.

The Argus | CommuniGate | The Chronic Crohns Campaign UK

Feedback

Do You Find This Crohns Site Useful ?

Microbial Symbiosis Factor Prevents Intestinal Inflamatory D....

ol453|29May2008|doi:10.1038/nature07008
ARTICLES

A microbial symbiosis factor prevents
intestinal inflammatory disease

Sarkis K. Mazmanian1*, June L. Round1* & Dennis L. Kasper2,3
Humans are colonized by multitudes of commensal organisms representing members of five of the six kingdoms of life; however, our gastrointestinal tract provides residence to both beneficial and potentially pathogenic microorganisms.

Imbalances in the composition of the bacterial microbiota, knownas dysbiosis, are postulated to be a major factor in human disorders such as inflammatory bowel disease. We report here that the prominent human symbiont Bacteroides fragilis protects animals from experimental colitis induced by Helicobacter hepaticus, a commensal bacterium with pathogenic potential.

This beneficial activity requires a single microbial molecule (polysaccharide A, PSA). In animals harbouring B. fragilis not expressing PSA, H. hepaticus colonization leads to disease and pro-inflammatory cytokine production in colonic tissues. Purified PSA administered to animals is required to suppress pro-inflammatory interleukin-17 production by intestinal immune cells and also inhibits in vitro reactions in cell cultures. Furthermore, PSA protects from inflammatory disease through a functional requirement for interleukin-10-producing CD41 T cells. These results show that molecules of the bacterial microbiota can mediate the critical balance between health and disease. Harnessing the immunomodulatory capacity of symbiosis factors such as PSA might potentially provide therapeutics for human inflammatory disorders on the basis of entirely novel biological principles.

Many human disorders seem to require a critical�and often
unknown�environmental component. Inflammatory bowel diseases (IBDs) such as Crohn�s disease and ulcerative colitis represent aberrant immune responses of the human gastrointestinal tract with adverse clinical outcomes1.

Abundant clinical and laboratory research has shown that, in IBD, commensal bacteria harboured within mammalian intestines are the targets of inflammatory responses1�4. Antibiotic treatment alleviates intestinal inflammation in humans and experiment prevents development of all animals5. Germ-free re-derivation of animals genetically susceptible to colitis intestinal inflammation6. Transfer of CD41 T-cell clones specific for bacterial antigens induces disease in recipient animals7.

In humans and other animals, inflammatory responses are apparently directed towards specific subsets of commensal organisms that have pathogenic potential but are not typically infectious pathogenic agents. All mammals harbour these species; why inflammation ensues only in those affected by IBD is unknown. Some investigators have predicted that, in addition to genetic factors, an imbalance in the normal microbiota without acquisition of an infectious organism is at least partially responsible for IBD8. Metagenomic studies have shown that entire classes of bacteria are lost or over-represented as part of the IBD process9. Perhaps in certain disorders where environmental factors are implicated, an imbalance between commensal bacteria with pathogenic potential (which we term pathobionts) and symbionts (commensal bacteria with beneficial potential) in the microbiota has a role in pathogenesis.

Humans maintain a lifelong association with innumerable commensal microorganisms that inhabit almost every environmentally exposed surface of the body. The gastrointestinal tract harbours .1014 microorganisms of,1,000 species10. Collectively, the intestinal microbiota represents a �forgotten organ� that can execute many physiological functions and thus may profoundly influence human biology. Germ-free animals, born and raised under sterile conditions, have defects in the development of intestinal tissues, show reduced vascular, nutritional and endocrine function, and are more susceptible to infection than conventionally colonized animals11,12. Both gastrointestinal and systemic immune responses are deficient in the absence of commensal microoroganisms13. Thus, mammals seem to depend on the microbiota to promote development and differentiation of host tissue14. Because of the complexity of the interactions of this diverse consortium of microorganisms with the mammalian host, the molecules responsible for host�microbe communication remain almost entirely unknown. As the microbiota has been implicated in disease, an understanding of these molecules may benefit human health15.

We have demonstrated that germ-free animals have defects in
CD41 T-cell development and that the human commensal bacterium Bacteroides fragilis corrects these deficiencies through the expression of PSA13. The precise immune-cell subset affected by PSA has not yet been identified. CD41 T cells of the mammalian immune system can be generally divided into a naive (�uneducated�) CD41CD45Rbhigh population and an antigen-experienced (�educated�) CD41CD45Rblow population16. We found that splenic cells from germ-free animals included a smaller proportion of CD41CD45Rblow T cells than those from age-matched conventional mice with a complete bacterial microbiota (Fig. 1a). We examined the ability of B. fragilis to correct deficiencies in the CD41CD45Rblow T-cell population. Mono-colonization of germ-free mice with wild-type B. fragilis alone restores the CD41CD45Rb profile to that found in animals with a complete bacterial microbiota (Fig. 1a; left panels). Notably, colonization with a mutant strain defective in the ability to produce PSA (B. fragilis DPSA) did not generate an expansion of the CD41CD45Rblow T-cell population (Fig. 1a; bottom right panel).

It is well established that the latter population possesses potent 1Division of Biology, California Institute of Technology, Pasadena, California 91125, USA. 2Channing Laboratory, Brigham & Women�s Hospital, Harvard Medical School, Boston,Massachusetts 02115, USA. 3Department of Microbiology and Molecular Genetics, Harvard Medical School, Boston, Massachusetts 02115, USA.
*These authors contributed equally to this work.

620 �2008 Nature Publishing Group

NATURE| Vol 453 | 29 May 2008 ARTICLES

anti-inflammatory properties and confers protection in animal mod-The pro-inflammatory cytokines tumour-necrosis factor-a (TNF-a), els of inflammation17. These results suggest that B. fragilis may have evolved a molecular strategy to mediate protection from inflammation during host�bacterial mutualism. Protection from colitis by B. fragilis We used the well-established CD41CD45Rb transfer model of experimental colitis18 to investigate whether B. fragilis colonization protects animals from inflammatory disease. In this model, pathogenic CD41CD45Rbhigh T cells are separated from protective CD41CD45Rblow cells and transferred into specific pathogen-free Rag2/2 mice. On cell transfer, mice were colonized with Helicobacter hepaticus8,19, a pathobiont that is a benign commensal in wild-type animals but an opportunistic pathogen causing colitis in immunocompromised mice. After 8 weeks, animals were killed and colitis was assessed using a standard scoring system20. The pathology scores show that H. hepaticus colonization and CD41CD45Rbhigh T-cell transfer are sufficient to induce severe colitis in Rag2/2 mice (Fig. 1b, first column), as previously reported19,21. Co-colonization with wild-type B. fragilis resulted in significant protection from disease (second column); conversely, co-colonization with B. fragilis DPSA does not offer protection (third column). Tissue damage in colitis is widely believed to result from production of inflammatory cytokines in response to commensal bacteria22. a interleukin (IL)-1b and IL-23 are central to disease initiation and progression in this experimental model of colitis23. Furthermore, amounts of these cytokines are elevated in patients with IBD24, and therapies neutralizing TNF-a have yielded promising results in clinical trials in patients with Crohn�s disease25.

We examined inflammatory cytokine concentrations during disease by directly culturing intestinal tissues of T-cell recipient colonized animals26. Amounts of the pro-inflammatory cytokines TNF-a (Fig. 1c), IL-12p40 and IL-1b (Supplementary Fig. 1) are elevated in the colons of Rag2/2 recipient mice colonized with H. hepaticus, but not in sections of small intestine�a site that is not affected in this model. Consistent with the protection observed by pathophysiological analysis of experimental colitis, TNF-a production was not elevated when these animals were co-colonized with wild-type B. fragilis. T-cell transfer plus co-colonization with H. hepaticus and B. fragilis DPSA results in increased colonic cytokine production similar to that seen in Rag2/2 animals colonized with H. hepaticus alone. Moreover, purified splenic CD41 T cells from H. hepaticus-colonized animals demonstrate increased TNF-a production; this condition is corrected by colonization with wild-type B. fragilis but not with the PSA deletion strain (Supplementary Fig. 2).

Expression of IL-23 is critical in the cascade of events leading to experimental colitis27,28.We found that increases in Il23 expression by splenocytes after disease induction are completely suppressed by intestinal colonization with PSA-producing B. fragilis (Fig. 1d). Over the course of the experiments, amounts of H. hepaticus and B. fragilis colonization did not Conventional 37% 26% Germ-free b
C57BL/6. The inability to gain weight is a hallmark of colitis in this experi(WT) Cell count Cell count differ between groups; thus, protection is not the result of bacterial clearance (Supplementary Figs 3 and 4). Instead, B. fragilis has evolved a specific immunomodulatory molecule that orchestrates beneficial immune responses to prevent its host from developing colitis.

Purified PSA prevents gut pathology 3-2-1-0-CD45Rb 42% 29%
Colitis scoreB. fragilis B. fragilis .PSA C57BL/6 To determine whether PSA is sufficient for protection from disease in (WT) the absence of the intact organism, we purified PSA to homogeneity29 and administered it by gavage (orally) to Rag2/2 mice. We measured disease progression by various pathological and histological criteria. Colitis scores after CD41CD45Rbhigh T-cell transfer in the absence of Smallintestine Colon Il23 (relative units) c 400 d 70 H. hepaticus colonization indicate the development of very mild 350 60 colitis due to inflammation elicited by the animals� specific patho-TNF-a (pg ml�1) 300 50 gen-free microbiota (Fig. 2a; first column). However, H. hepaticus 250 2/2 animals that received CD41CD45Rbhigh 200 40 colonized Rag T-cell 150 30 transfers developed severe colitis (Fig. 2a; second column). Oral100 20 PSA administration almost completely protected animals against 10 H. hepaticus-induced colitis (Fig. 2a; third column), reducing disease 0 Smallintestine - Colon - Smallintestine Colon

Smallintestine Colon to levels of control animals which do not develop colitis (Fig. 2a; fourth column). Figure 1 | B. fragilis colonization requires PSA for protection from experimental colitis. a, Mono-association of germ-free mice with wild-type B. fragilis expands the proportions of CD41CD45Rblow T cells in a PSA-dependent manner (mean 6 s.d. for 3 experiments: conventional, 38.4% 6 2.2; germ-free, 26.7% 6 1.3; B. fragilis, 40.8% 6 3.1; B. fragilis DPSA, 28.8% 6 2.6). All cells were gated on CD41 splenocytes. b, Cocolonization with B. fragilis ameliorates disease onset, whereas cocolonization with B. fragilis DPSA is not protective (P 5 0.004, Mann�Whitney U-test). Combined data from two independent experiments are shown. c, ELISA of colon organ cultures demonstrates increased expression of pro-inflammatory cytokine TNF-a in diseased colons, with significant reductions in animals co-colonized with wild-type B. fragilis but not with B. fragilis DPSA. d, qPCR for Il23p19 was performed on splenocytes
and normalized to L32 expression. Error bars represent s.d. for triplicate samples in all cases. WT, wild type. mental setting4.
Wasting disease occurred in Rag2/2 animals after
transfer of CD41CD45Rbhigh cells and colonization with H. hepaticus (Fig. 2b; PBS 1 Hh). These animals also developed intestinal pathology and expressed pro-inflammatory cytokines (as described above).

Oral administration of PSA from the outset completely protected animals against H. hepaticus-mediated wasting disease (Fig.2b; PSA 1 Hh). Demonstrating that Helicobacter hepaticus provides
the necessary antigens for inflammation induction, no pathology was observed in uncolonized animals (Fig. 2b; PBS 2Hh) or in animals without cell transfer.

We examined histological sections of colons for inflammation
resulting in experimental colitis. Transfer of CD41CD45Rbhigh T
cells into H. hepaticus-colonized Rag2/2 mice resulted in onset of severe colitis, as evidenced by massive epithelial cell hyperplasia and gross thickening of the gut wall (Fig. 2c; second panel). Consistent with previous studies, the combination of CD41CD45Rbhigh T-cell transfer and H. hepaticus colonization resulted in leukocyte infiltration of affected tissues�a hallmark of inflammation and disease.
�2008 Nature Publishing Group

ARTICLES NATURE | Vol 453 | 29 May 2008

(Fig. 2c; second panel, bottom)19,21. Oral administration of PSA into H. hepaticus-colonized cell transfer recipients conferred complete protection against experimentally induced colonic hyperplasia (Fig. 2c; third panel). Furthermore, PSA-treated animals showed no leukocyte infiltration in colonic tissues (Fig. 2c; third panel, bottom), indicating protection against inflammation. Taken together, these results suggest that PSA prevents colitis and protects mice against the associated weight loss and inflammatory cell infiltration observed in diseased animals. Control of chemically induced inflammation Experimental colitis and human IBD result from an initial inflammatory response that�lacking repression�advances in an uncontrolled fashion and ultimately leads to intestinal pathology and disease. To elucidate how PSA affects these primary inflammatory responses, we used an animal model of chemically induced colonic inflammation. Rectal administration of trinitrobenzene sulphonic acid (TNBS) to wild-type mice mimics the initiation of colitis by P = 0.0002 a Colitis score
2 1 0 Rag�/� + PBS + Rag�/� + PBS + Rag�/� + PSA + Rag�/� + PBS + CD45Rbhigh CD45Rbhigh + CD45Rbhigh + H. hepaticus
H. hepaticus H. hepaticus b 27 eliciting inflammatory T-cell responses. Disease was induced by the administration of TNBS (vehicle was used as a negative control), and oral treatment of PSA was evaluated. TNBS-treated animals exhibited the greatest amount of weight loss, and were unable to recover as rapidly in comparison to either vehicle-treated or PSA-treated animals. (Fig. 3a). Histological analysis confirmed PSA protection of colonic tissues against the massive epithelial hyperplasia and loss of colonocyte organization seen after TNBS treatment (Fig. 3b).
Studies have shown that pathogenic T-helper (TH)17 cells, which produce IL-17, mediate the induction of TNBS experimental colitis30.

We found that Il17 expression was increased among purified CD4T cells from mesenteric lymph nodes (MLNs; Fig. 3c) of diseased animals but not from those receiving PSA treatment. The increased 1 expression of Tnfa among CD4T cells from MLNs of TNBS-treated animals was also reduced in PSA-treated groups (Fig. 3d).

Transcriptional analysis of TNBS-treated colons demonstrated that the expression of both Il17 and Tnfa was highly elevated in diseased but not in PSA-protected animals (Fig. 3e, f). Therefore, PSA inhibits intestinal pathology and inflammation in a chemically induced model of experimental colitis. PSA induces production of IL-10 Protection from experimental colitis is engendered through antiinflammatory processes that prevent undesirable reactions against the intestinal microbiota23. Interleukin-10-deficient (Il102/2) animals develop colitis31. IL-10 is one of the most potent antiinflammatory cytokines and is required for protection in many animal models of inflammation 21,27,32. As assayed by quantitative real-time polymerase chain reaction (qPCR), transcriptional expression of Il10 within colons of PSA-treated mice was significantly higher than in control and TNBS-treated mice (Fig. 4a). IL-10 is produced by many cell types; however, CD41 T cells that express IL-10 have immunosuppressive activities that inhibit inflammation during experimental colitis33. We purified fresh CD41 T cells from MLNs of PSA-treated mice in which inflammation was reduced, and we found greatly elevated expression of the Il10 transcript (Fig. 4b). PBS � Hh We assessed whether PSA was sufficient to induce IL-10 in vitro; a PBS + Hh PSA + Hh No transfer * ** *** P = 0.011 P = 0.005 P = 0.0008 specific expression of IL-10 from culture supernatants, whereas coculture with B. fragilis DPSA induced significantly lower expression 21 of IL-10 (Supplementary Fig. 5). As PSA induces expression of IL-10 specific increase in IL-10 production occurred when bone-marrow1derived dendritic cells (BMDCs) and naive CD4T cells were treated 25 Body weight (grams) with purified PSA (Fig. 4c). When BMDCs and naive CD41 T cells were infected with H. hepaticus co-cultured with B. fragilis, we found 23 in vitro, we speculated whether this molecule is required for inhibition of inflammatory responses to H. hepaticus. We infected BMDC� T-cell co-cultures with live H. hepaticus and measured production of Figure 2 | Purified PSA protects against experimental colitis. a, Colonization with H. hepaticus drives the onset of severe colitis (second column). PSA-treated animals either do not develop colitis or develop only mild disease (third column) (P . 0.05 for dissimilar results, P , 0.01 for similar results, Kruskal�Wallis comparison of all groups; P 5 0.0002, two-tailed Mann�Whitney U-test). The data shown are the pooled results from duplicate experiments. b, Wasting disease in Rag22/2 animals results from transfer of CD41CD45Rbhigh T cells and colonization with H. hepaticus (PBS 1 Hh). PSA treatment by gavage protects animals against wasting (PSA 1 Hh). Two-factor analysis of variance (ANOVA) indicates that comparisons between all indicated groups (asterisks) are statistically significant. Error bars represent s.d. between four animals per group; experiments were performed in duplicate. c, Architecture of colonic sections from wild-type animals (first panel). CD41CD45Rbhigh T-cell transfer into H. hepaticus-colonized Rag22/2 mice resulted in severe colitis, as evidenced by massive epithelial hyperplasia and pronounced inflammation (second panel); the higher magnification below shows inflammatory cell infiltration into colonic tissues. Oral PSA treatment protects H. hepaticus-colonized animals (third panel) from colitis. Images in each row are the same magnification; original magnification 310 for top, 340 for bottom row. WT, wild type. 19 17 0 10 20 30 40 5060 Days after cell transfer Rag�/� + PBS + Rag�/� + PSA + CD45Rbhigh + CD45Rbhigh + c C57BL/6 (WT) H. hepaticus H. hepaticus

�2008 Nature Publishing Group

ARTICLES
and IL-17 by MLN cells; however, in the absence of IL-10 production in colonized animals, B. fragilis co-colonization did not reduce the concentration of these pro-inflammatory molecules (Fig. 5a and b, respectively). As expected, the absence of PSA had no effect. Again using the cell transfer model of colitis, we transferred CD41CD45Rbhigh T cells to H. hepaticus-colonized Rag2/2 animals.

Administration of an anti-IL-10 receptor antibody to mice (to block IL-10 signalling) during oral treatment with PSA abrogated protection from colitis (Fig. 5c). When Il102/2 animals were treated with TNBS in the presence or absence of PSA, weight and histology data (Supplementary Figs 7 and 8) indicate that IL-10 production is required for PSA-elicited reduction of intestinal immune responses. Our data suggest that PSA-mediated protection entails the generation and/or expansion of IL-10-producing CD41 T cells. To determine whetherIL-10 production by CD41 T cells is required for protection, we transferred CD41CD45Rbhigh T cells from Il102/2 donor mice into Rag2/2 recipients and then colonized the recipients with H. hepaticus. As expected, groups of mice receiving Il102/2 T cells along with H. hepaticus developed severe colitis (Fig. 5d; first column) and were not protected by PSA (Fig. 5d; second column). This result, supported by histological findings in colons, suggests that PSA induces protection from �previously pathogenic� CD41CD45Rbhigh T cells in an IL-10-dependent manner (Fig. 5e). Weight analysis at death shows that colitic PBS-and PSA-treated animals receiving Il102/2 CD41CD45Rbhigh T cells (unlike control NATURE| Vol 453 | 29 May 2008 the pro-inflammatory cytokine TNF-a. Addition of increasing amounts of the pathogenic commensal bacterium resulted in a dose-dependent increase in TNF-a concentration, as determined by enzyme-linked immunosorbent assay (ELISA) of culture supernatants (Fig. 4d; left three bars). Treatment of cells with purified PSA decreased TNF-a production in response to H. hepaticus (Fig. 4d; middle three bars). Most notably, co-incubation of cell cultures with H. hepaticus and PSA in the presence of a neutralizing IL-10 receptor antibody completely reversed this phenotypic effect and increased expression of TNF-a (Fig. 4d; right three bars). The results are similar for the related pro-inflammatory cytokine IL-1b (Supplementary Fig. 6). Thus, IL-10 produced in response to PSA is required for inhibition of inflammatory reactions in cell cultures.
IL-10-producing T cells suppress colitis.

We investigated the requirement for IL-10 in suppression of intestinal inflammation. Initially, Il102/2 animals were colonized with H. hepaticus alone or in combination with B. fragilis (either wild type or DPSA). We subsequently collected MLNs and re-stimulated cells in culture with soluble H. hepaticus antigens using an assay previously developed to measure antigen-specific responses to this pathogen27. H. hepaticus-colonization resulted in increased production of TNF-a ab 105 animals receiving no transferred cells) developed wasting disease 1 (Fig. 5f). Thus, IL-10 production by CD4T cells is required for PSA-mediated protection from experimental colitis. These results constitute the first reported evidence of a symbiotic bacterial molecule that networks with the immune system to coordinate anti inflammatory responses required for mammalian health. ab Percentage of initial body weight
TNBS + PSA TNBS + PBSControl 85*** Control P = 0.003 P =0.00780 100 95 95 TNBS + PBS TNBS + PSA 75 1234 200 20 Colon CD4+ MLN Il10 (relative units) Il10 (relative units) 160 120 80 16 12 8 Days after TNBS cd Il17 (relative units) Control CD4+ MLN TNBS + TNBS + Tnfa (relative units) 14 40 4 CD4+ MLN Control TNBS + TNBS + PBS PSA PBS PSA 8 6 12 0 0 Control TNBS + TNBS + Control TNBS + TNBS + 10 c 2,000 d 3,000 4 0 TNF-a (pg ml�1 ) ControlPSA LPSAnti-CD3 & 2 2,500 IL-10 (pg ml�1) 1,500 0 2,000 PBS PSA PBS PSA ef 1,000 1,500 1,000 Il17 (relative units) 250 200 150 100 Tnfa (relative units) 10,000 0 Anti-IL-10R Colon Colon 500 500 8,000 0 6,000 �� +++ +++ H. hepaticus: 4,000 PSA 2,000 (100 �g ml�1): Control TNBS + TNBS + Control TNBS + TNBS + �� +++ PBS PSA PBS PSA (20 �g ml�1): Figure 3 | Intestinal immune responses are modulated by a beneficial bacterial molecule. a, Oral PSA administration causes a statistically significant increase in body weight related to TNBS-treated PBS controls.

Two-factor ANOVA values for all indicated groups (asterisks) are statistically significant. Error bars represent s.d. between four animals per group. b, Colons from TNBS plus PBS-treated groups show severe pathology, whereas those from TNBS and PSA-treated animals have histological architecture similar to that seen in untreated controls. The images shown are representative sections from animals in two independent experiments. Original magnification 320. c, d, qPCR of purified CD41 T cells from MLNs with Il17a-and Tnfa-specific primers demonstrates that oral PSA administration reduces Il17a (c) and Tnfa (d) expression during disease. e, f, Transcriptional expression of Il17a (e) and Tnfa (f) from homogenized colons. Error bars are from duplicate runs of three independent experiments (c�f). Figure 4 | PSA induces Il10 expression in TNBS-treated animals and inhibits TNF-a production in primary cultured cells through IL-10 production. a, b, Wild-type mice were treated with ethanol (control), TNBS, or TNBS and PSA. qPCR assay of colons (a) and CD41 T cells purified from MLNs (b) show elevated Il10 expression in response to PSA. c, Incubation of BMDC�T cell co-cultures with purified PSA specifically induces IL-10 production at concentrations comparable to those induced by LPS or by anti-CD3 and anti-CD28 antibodies. d, Infection of BMDC�T cell cocultures with increasing concentrations of H. hepaticus (multiplicity of infection: 0.1, 1.0 and 10, as depicted by triangles) results in increased TNF-a release. Treatment of infected cells with PSA reduces amount of TNF-a (middle three bars). Addition of an anti-IL-10 receptor antibody (anti-IL10R) alleviates suppression of inflammatory responses, resulting in increased TNF-a production (right three bars). Error bars show s.d. for triplicate samples in all cases.

�2008 Nature Publishing Group

ARTICLES NATURE | Vol 453 | 29 May 2008

Summary and implications According to the �hygiene hypothesis� put forth nearly two decades ago, reduced exposure to infections in early childhood�owing to diminishing family size and improvements in living standards and personal hygiene, for example�may increase the risk of allergic and autoimmune disease34. This concept is supported by epidemiological and clinical reports documenting increased incidences of IBD, colon cancer, asthma, type 1 diabetes and multiple sclerosis over the past 50 years in societies with improved medical care and hygiene (for example, Europe, the United States and Japan) but not in undeveloped countries35. However, the application of major interventions, including vaccination, sanitation, and antibacterial and antiviral therapies, often does not permit discrimination between a d 3 800 MLN 600 infectious and non-infectious microorganisms and has undoubtedly led to changes in human association with the microbial world as a whole.

The hygiene hypothesis does not address humanity�s primary relationship with bacteria: the harbouring of multitudes of microbial species during commensalism. Our studies show that symbiotic bacteria residing in the mammalian gastrointestinal tract produce molecules that mediate healthy immune responses and protect the host from inflammatory disease. We propose that the mammalian genome does not encode for all functions required for immunological development but rather that mammals depend on critical interactions with their microbiome (the collective genomes of the microbiota) for health.

As mammals have harboured their commensal partners for millennia, adaptive co-evolution has formed an inextricable bond between microbe and host36. Imbalances in the microbiota may contribute to some human diseases, and altered composition of the gut bacteria has been implicated in obesity37. We show that B. fragilis 2 protects its host from inflammatory disease caused by H. hepaticus in 0 H. hepaticusH. hepaticusB. fragilisH. hepaticus Control Colitis score TNF-a (pg ml�1) 400 an animal model of experimental colitis. The implication that intest 1 200 inal bacteria actively network with the host�s immune system highlights the importance of the composition of the microbiota for 0 overall health. If specific classes of bacteria have indeed evolved to promote the host�s health, then disease may well result from the absence of these organisms and their beneficial molecules (for example, as a result of improved hygiene). Inflammation resulting
Il10 �/�CD4+ CD45Rbhigh No transfer be from dysbiosis between symbionts and pathobionts may lay the 4,000 molecular foundations for many intestinal�and perhaps non-MLN Rag �/� + PBS 3,000 + H. hepaticus 2,000 Rag �/� + PSA 1,000 + H.hepaticus 0 IL-17 (pg ml�1) intestinal�diseases. The exploration of probiotics (bacteria such as Il10 �/�CD4+ lactobacilli and bifidobacteria that promote health) has thus far failed CD45Rbhigh to identify specific bacterial molecules or host mechanisms required for protection38. Here we present evidence that a single bacterial molecule can ameliorate inflammatory disease in animals. Our Rag �/� + PBS No transfer observations suggest that many other symbiosis factors�bacterial molecules that have evolved to promote human health�remain to be discovered. The finding that PSA from B. fragilis is a natural anti + H. hepaticus f inflammatory molecule that actively promotes mammalian health c 3 may provide aplatform for the development of therapies based on the fundamental relationship between humans and their beneficial microbial partner.

METHODSSUMMARY28 26 18 20 22 S S+ PSA Weight (grams) Colitis score 2 1 24 Three models of intestinal inflammation were used: (1) CD41CD45Rbhigh 16 T cells Il10 �/�CD4+ CD45Rbhigh No transfer Figure 5 | IL-10 is required for PSA-mediated protection from intestinal inflammation and experimental colitis. a, b, Il102/2 mice were left uncolonized (control) or were colonized with H. hepaticus (to induce inflammation) either alone or in combination with B. fragilis (wild-type or DPSA). MLNs from experimental groups were pooled and re-stimulated with soluble H. hepaticus antigen (5 mgml21) for 48 h. Secretion of the pro-inflammatory cytokines TNF-a (a) and IL-17A (b) was analysed by ELISA. Error bars show s.d. for triplicate samples. c, Colitis scores show that PSA protection requires IL-10 signalling, as treatment with anti-IL-10 receptor antibody (anti-IL-10R) abrogates the PSA-mediated protection. Data represent two independent experiments. d, PSA-mediated protection from disease requires IL-10-producing CD41 T cells. Treatment with PSA does not reduce colitis when CD41CD45Rbhigh T cells are transferred from Il102/2 mice. Control animals without cell transfer do not develop colitis. Results are shown for one representative trial of two independent experiments. e, Histological colon sections show that PSA does not protect animals from experimental colitis when CD41CD45Rbhigh T cells cannot produce IL-10. All images are from the same magnification. Original magnification 310. f, Mean body weights for groups of animals (n 5 4) when culled demonstrate that IL-10 is required for PSA-mediated protection from wasting.

Error bars show s.d. between 4 animals per group from one representative trial of two independent experiments.
were purified from the spleens of wild-type or Il102/2 donor mice by flow cytometry and transferred into Rag2/2 (C57BL/6) recipients as described; (2) TNBS colitis was induced by pre-sensitization of wild-type (C57BL/6) mice on the skin with a TNBS and acetone mix. Seven days after sensitization, 2.5% TNBS in ethanol was administered rectally; mice were killed 3�6 days later; and (3) Il102/2 mice were colonized (by oral gavage) with H. hepaticus alone or in combination with wild-type B. fragilis or B. fragilis DPSA. B. fragilis NCTC 9343 and H. hepaticus ATCC 51149 were obtained from the American Type Culture Collection. Cytokines from the spleens, colons, or MLNs were assayed by ELISA, qPCR, or flow cytometry. Colitis was assessed with tissue sections (fixed, paraffin embedded, sectioned onto a slide and stained with haematoxylin and eosin) and was scored by a pathologist in a blinded experimental set-up (R. T. Bronson) according to a standard scoring system: 0, no thickening of colonic tissues and no inflammation (infiltration of lymphocytes); 1, mild thickening of tissues but no inflammation; 2, mild thickening of tissues and mild inflammation; 3, severe thickening and severe inflammation. BMDCs were purified from femurs of mice after extraction and washing in PBS. Cells were cultured for 8 days in C-RPMI-10 in the presence of GM-CSF (20 ng ml21; Biosource). CD41 T cells were purified by negative selection over a magnetic column (Miltenyi or R&D Systems). Received 7 February; accepted 18 April 2008.

1. Poxton, I. R., Brown, R., Sawyerr, A. & Ferguson, A. Mucosa-associated bacterial flora of the human colon. J. Med. Microbiol. 46, 85�91 (1997).
2. Sellon, R. K. et al. Resident enteric bacteria are necessary for development of spontaneous colitis and immune system activation in interleukin-10-deficient mice. Infect. Immun. 66, 5224�5231 (1998). �2008 Nature Publishing Group

NATURE| Vol 453 | 29 May 2008 ARTICLES

3. Elson, C. O. Commensal bacteria as targets in Crohn�s disease. Gastroenterology 119, 254�257 (2000).
4. Sartor, R. B. Mechanisms of disease: pathogenesis of Crohn�s disease and ulcerative colitis. Nature Clin. Pract. Gastroenterol. Hepatol. 3, 390�407 (2006).
5. Videla, S. et al. Role of intestinal microflora in chronic inflammation and ulceration of the rat colon. Gut 35, 1090�1097 (1994).
6. Taurog, J. D. et al. The germfree state prevents development of gut and joint inflammatory disease in HLA-B27 transgenic rats. J. Exp. Med. 180, 2359�2364 (1994).
7. Kullberg, M. C. et al. Induction of colitis by a CD41 T cell clone specific for a bacterial epitope. Proc. Natl Acad. Sci. USA 100, 15830�15835 (2003).
8. O�Hara, A. M. & Shanahan, F. The gut flora as a forgotten organ. EMBO Rep. 7, 688�693 (2006).
9. Frank, D. N. et al. Molecular-phylogenetic characterization of microbial community imbalances in human inflammatory bowel diseases. Proc. Natl Acad. Sci. USA 104, 13780�13785 (2007).
10. Gill, S. R. et al. Metagenomic analysis of the human distal gut microbiome. Science 312, 1355�1359 (2006).
11. Ley, R. E., Peterson, D. A. & Gordon, J. I. Ecological and evolutionary forces shaping microbial diversity in the human intestine. Cell 124, 837�848 (2006).
12. Smith, K., McCoy, K. D. & Macpherson, A. J. Use of axenic animals in studying the adaptation of mammals to their commensal intestinal microbiota. Semin. Immunol. 19, 59�69 (2006).
13. Mazmanian, S. K., Liu, C. H., Tzianabos, A. O. & Kasper, D. L. An immunomodulatory molecule of symbiotic bacteria directs maturation of the host immune system. Cell 122, 107�118 (2005).
14. Pamer, E. G. Immune responses to commensal and environmental microbes. Nature Immunol. 8, 1173�1178 (2007).
15. Dethlefsen, L., McFall-Ngai, M. & Relman, D. A. An ecological and evolutionary perspective on human�microbe mutualism and disease. Nature 449, 811�818 (2007).
16. Bell, E. B. Function of CD4 T cell subsets in vivo: expression of CD45R isoforms. Semin. Immunol. 4, 43�50 (1992).
17. Izcue, A., Coombes, J. L. & Powrie, F. Regulatory T cells suppress systemic and mucosal immune activation to control intestinal inflammation. Immunol. Rev. 212, 256�271 (2006).
18. Maloy, K. J. et al. CD41CD251 TR cells suppress innate immune pathology through cytokine-dependent mechanisms. J. Exp. Med. 197, 111�119 (2003).
19. Cahill, R. J. et al. Inflammatory bowel disease: an immunity-mediated condition triggered by bacterial infection with Helicobacter hepaticus. Infect. Immun. 65, 3126�3131 (1997).
20. Scheinin, T., Butler, D. M., Salway, F., Scallon, B. & Feldmann, M. Validation of the interleukin-10 knockout mouse model of colitis: antitumour necrosis factor-antibodies suppress the progression of colitis. Clin. Exp. Immunol. 133, 38�43 (2003).
21. Kullberg, M. C. et al. Bacteria-triggered CD41 T regulatory cells suppress Helicobacter hepaticus-induced colitis. J. Exp. Med. 196, 505�515 (2002).
22. Bregenholt, S. Cells and cytokines in the pathogenesis of inflammatory bowel disease: new insights from mouse T cell transfer models. Exp. Clin. Immunogenet. 17, 115�129 (2000).
23. Powrie, F. & Maloy, K. J. Immunology. Regulating the regulators. Science 299, 1030�1031 (2003).
24. Xavier, R. & Podolsky, D. K. Commensal flora: wolf in sheep�s clothing. Gastroenterology 128, 1122�1126 (2005).
25. Rutgeerts, P. et al. Infliximab for induction and maintenance therapy for ulcerative colitis. N. Engl. J. Med. 353, 2462�2476 (2005).
26. Rakoff-Nahoum, S., Paglino, J., Eslami-Varzaneh, F., Edberg, S. & Medzhitov, R. Recognition of commensal microflora by toll-like receptors is required for intestinal homeostasis. Cell 118, 229�241 (2004).
27. Kullberg, M. C. et al. IL-23 plays a key role in Helicobacter hepaticus-induced T cell-dependent colitis. J. Exp. Med. 203, 2485�2494 (2006).
28. Hue, S. et al. Interleukin-23 drives innate and T cell-mediated intestinal inflammation. J. Exp. Med. 203, 2473�2483 (2006).
29. Tzianabos, A. O. et al. The capsular polysaccharide of Bacteroides fragilis comprises two ionically linked polysaccharides. J. Biol. Chem. 267, 18230�18235 (1992).
30. Elson, C. O. et al. Monoclonal anti-interleukin 23 reverses active colitis in a T cell-mediated model in mice. Gastroenterology 132, 2359�2370 (2007).
31. Kuhn, R., Lohler, J., Rennick, D., Rajewsky, K. & Muller, W. Interleukin-10-deficient mice develop chronic enterocolitis. Cell 75, 263�274 (1993).
32. Asseman, C., Mauze, S., Leach, M. W., Coffman, R. L. & Powrie, F. An essential role for interleukin 10 in the function of regulatory T cells that inhibit intestinal inflammation. J. Exp. Med. 190, 995�1004 (1999).
33. Groux, H. et al. A CD41 T-cell subset inhibits antigen-specific T-cell responses and prevents colitis. Nature 389, 737�742 (1997). 34. Strachan, D. P. Hay fever, hygiene, and household size. Br. Med. J. 299, 1259�1260 (1989).
35. Bach, J. F. The effect of infections on susceptibility to autoimmune and allergic diseases. N. Engl. J. Med. 347, 911�920 (2002).
36. Liu, C. H., Lee, S. M., Vanlare, J. M., Kasper, D. L. & Mazmanian, S. K. Regulation of surface architecture by symbiotic bacteria mediates host colonization. Proc. Natl Acad. Sci. USA 105, 3951�3956 (2008).
37. Turnbaugh, P. J. et al. An obesity-associated gut microbiome with increased capacity for energy harvest. Nature 444, 1027�1031 (2006).
38. Mazmanian, S. K. & Kasper, D. L. The love-hate relationship between bacterial polysaccharides and the host immune system. Nature Rev. Immunol. 6, 849�858 (2006).

Supplementary Information is linked to the online version of the paper at www.nature.com/nature.

Acknowledgements We thank R. T. Bronson for discussions about histopathology; members of the Mazmanian laboratory for critical comments throughout the course of the work; and J. McCoy for editorial expertise. S.K.M. acknowledges a fellowship from the Helen Hay Whitney Foundation; J.L.R. acknowledges support from the Jane Coffin Childs Memorial Fund. This work was supported by funding from the NIH/NIAID (R01 AI039576) to D.L.K., and by grants from the Searle Scholars Program, the Damon Runyon Cancer Research Foundation, and the Crohn�s and Colitis Foundation of America to S.K.M. Author Contributions S.K.M., J.L.R. and D.L.K. designed the research; S.K.M. and J.L.R. performed the research; S.K.M., J.L.R. and D.L.K. analysed the data and wrote the paper. Author Information Reprints and permissions information is available at www.nature.com/reprints. Correspondence and requests for materials should be addressed to S.K.M. (sarkis@caltech.edu) or D.L.K. (dennis_kasper@hms.harvard.edu).
�2008 Nature Publishing Group.

PDF Of This Available from tccc.uk@btinternet.com

Supplementary Info Link Here :

Supplementary Info Link

Soothing Intestinal Sugars - IBD & Cancer.

Soothing intestinal sugars

Marika C. Kullberg

The gut is a new frontier in microbiology, offering many opportunities for innovative investigation. The finding of one such study is that intestinal inflammation in mice can be tamed by bacterial sugars.

The human �gut flora� consists of between 300 and 1,000 microbial species, and some 1014 microorganisms in total (about ten times the number of cells of the human body). We usually live in harmony with these microbes, and would be less healthy without them. For example, they synthesize essential vitamins and amino acids, and also degrade otherwise indigestible plant material, as well as certain drugs and pollutants.

On page 620 of this issue, Mazmanian et al.1 report that Bacteroides fragilis, a common bacterium of the lower gastrointestinal tract in mammals, can prevent intestinal inflammation in mice. Specifically, the authors show that polysaccharide A (PSA) of B. fragilis prevents gut inflammation induced by another bac terium, Helicobacter hepaticus, or by the chemical compound TNBS (2,4,6-trinitrobenzene sulphonic acid).

This is an exciting finding, not least given that the incidence of human intestinal inflammation and inflammatory bowel disease has increased steadily in the Western world since the early 1950s. These conditions, which include Crohn�s disease and ulcerative colitis, are believed to stem in part from inappropriate immune responses to the gut microbiota2.

In the healthy intestine, immune balance is regulated by different types of white blood cells called CD4+ T lymphocytes. These cells include CD4+ effector T lymphocytes (which help us fight pathogens by secreting various immune mediators called cytokines) and CD4+ Dendritic cell a b c

Activated dendritic cell + PSA + PSA

Hh Hh Hh

IL-23, IL-12, IL-6, TNF-a, IL-1�

Inflammatory TH1/TH17 CD4+ effector T cell

Immunosuppressive CD4+ regulatory T cell

CD4+ T cell IFN-. IL-17. TNF-a, IL-1�

.IL-10, TNF-a, IL-1� .IL-10 IL-10

Sub-optimal effector T-cell activation

regulatory T lymphocytes (which, through their production of the cytokines IL-10 and TGF-�, dampen the effector T cells when their action is no longer needed). When the balance between these types of T cell is disturbed, the immune response goes awry and intestinal inflammation occurs2.

Evidence in support of the theory that bacteria trigger gut inflammation came from the discovery that mutations in NOD2, a host immune-cell receptor involved in detecting bacterial peptidoglycan, are associated with an increased risk of Crohn�s disease3,4. Furthermore, treatment with broad-spectrum antibiotics or probiotics (beneficial microbial species5) can improve health in patients with inflammatory bowel disease. Probiotics might inhibit the growth or invasion of pathogenic bacteria, or strengthen the gut-wall barrier.

They may also stimulate the production of IgA (antibodies that are secreted into the gut lumen), and of IL-10 and TGF-� (ref. 5). Mazmanian et al.1 used two experimental approaches. In the first, colitis was induced in mutant (Rag�/�) T-cell-deficient mice by infecting the animals with H. hepaticus and giving them CD4+ effector T cells. These effector cells start to respond to H. hepaticus, but in the absence of counterbalancing host regulatory T cells intestinal inflammation develops within a few weeks. The cytokine IL-23 plays a key role in this inflammatory response, and the disease is associated with a TH1/TH17 effector T-cell response to H. hepaticus6 (Fig. 1a). The second approach involves a chemical-induced colitis, in which the administration of TNBS to normal mice leads to acute inflammation within a few days.

In their new paper, Mazmanian et al.1 demonstrate that giving mice B. fragilis at the same time as the colitis-inducing agents improves the animals� health. A mutant of B. fragilis that lacks PSA could not prevent inflammation, implicating PSA in helping to maintain immune balance. Finally, colitis did not develop when purified B. fragilis PSA was administered together with H. hepaticus and effector T cells to the Rag�/� mice, or when this polysaccharide was given to the TNBS recipients. Together, these results provide direct evidence that PSA prevents intestinal inflammation in both model systems.

Bacteroides fragilis synthesizes at least eight distinct surface polysaccharides as part of its capsule, or coat7, PSA being the most abundant.

Although CD4+ T cells normally recognize and respond to peptide fragments of proteins, PSA can be taken up by so-called antigen-presenting cells, such as dendritic cells, and presented to CD4+ T cells, resulting in T-cell activation7.

Mazmanian et al.1 show that IL-10 secreted by T cells is essential for PSA to protect against colitis. But it is not yet clear whether PSA has induced �true� IL-10-secreting regulatory T cells in the mice protected from colitis.

Thus, although the authors demonstrate that PSA cannot prevent colitis when T cells cannot produce IL-10, this could be due to a more pathogenic nature of such IL-10- deficient effector T cells.

What about the molecular mechanism by which PSA prevents H. hepaticus-induced colitis ? One possibility is that this sugar acts on dendritic cells, thereby altering their capacity to trigger an efficient effector-T-cell response (Fig. 1b). Support for this hypothesis comes from studies showing that T cells isolated from PSA-treated animals are hyporesponsive8.

Alternatively, is PSA inducing IL-10-secreting regulatory T cells (Fig. 1c)? Treatment with filamentous haemagglutinin (FHA) from Bordetella pertussis, the causative agent of whooping cough, can protect Rag�/� mice from colitis induced by CD4+ effector T cells9. T cells isolated from these FHA-treated mice produced IL-10 in vitro following polyclonal stimulation9, supporting the model shown in Figure 1c. It will be interesting to see whether the same observation (that IL-10 is produced by T cells isolated from disease-protected mice) applies in the H. hepaticus colitis model in which PSA prevents intestinal inflammation. Moreover, do CD4+ T cells isolated from disease-free mice infected with B. fragilis plus H. hepaticus respond to B. fragilis and/or H. hepaticus antigens ? If so, what cytokines do such bacterium-specific T cells produce ?

Has PSA skewed the T-cell immune response to H. hepaticus antigens away from a pro-inflammatory TH1/TH17-type response (Fig. 1a) towards IL-10 secretion (Fig. 1c) ?

There is also the question of whether PSA can act against H. hepaticus-induced colitis not only when administered from the start of an experiment, but also when used to treat established disease. The long-term aim, of course, is to develop drugs to cure intestinal inflammation in humans. In this regard, treatment of patients suffering from inflammatory bowel disease with live eggs from the porcine whipworm, Trichuris suis, has produced promising results10. These effects are believed to be due to the induction by T. suis of regulatory T cells and factors such as IL-10, TGF-� and prostaglandin E2 that help maintain immune balance10.

One view is that the apparent boons of modern life � antibiotics, vaccines and improved sanitation � have reduced the incidence of parasitic worms and other microbes, and therefore also of disease-protective molecules such as �T. suis-like antigens� or PSA. That in turn may have altered the way our immune system responds to challenges, leading to the increased incidence of inflammatory diseases.

Figure 1 | Intestinal inflammation, and how B. fragilis PSA may prevent it in mice1. a, In mice that develop colitis, dendritic cells activated by H. hepaticus (Hh) produce pro-inflammatory cytokines (such as IL-23, IL-12, IL-6, TNF-a and IL-1�). They also present Hh antigen to CD4+ T cells that then differentiate into inflammatory TH1/TH17 CD4+ effector T cells; these cells are specific for Hh antigen, and produce the inflammatory cytokines IFN-. and IL-17 (ref. 6). b, c, Dendritic cells jointly stimulated with Hh and B. fragilis PSA are less responsive, with reduced secretion of the pro-inflammatory cytokines TNF-a and IL-1�, and enhanced production of anti-inflammatory IL-10 (ref. 1). They present both Hh antigen and PSA to CD4+ T cells, and may either (b) produce only sub-optimal effector T-cell activation or (c) induce the differentiation of IL-10-producing CD4+ regulatory T cells. The antigen specificity (Hh antigen and/or PSA) of such regulatory T cells is unknown.

603 NATURE|Vol 453|29 May 2008 NEWS & VIEWS

NEWS & VIEWS NATURE|Vol 453|29 May 2008

Ever-improving molecular techniques are providing tantalizing glimpses of the gut ecosystem11. With the launch of the Human Microbiome Project12, which plans to characterize the human microbiota and analyse its role in human health and disease, we are set to see considerable advances in understanding how host�microbial interactions may affect human health. When such information will translate into new therapeutic approaches is, however, anyone�s guess. �

Marika C. Kullberg is in the Immunology and Infection Unit, Department of Biology, University of York, and The Hull York Medical School, PO Box 373, York YO10 5YW, UK. e-mail: mk512@york.ac.uk

1. Mazmanian, S. K., Round, J. L. & Kasper, D. L. Nature 453, 620�625 (2008).

2. Coombes, J. L., Robinson, N. J., Maloy, K. J., Uhlig, H. H. & Powrie, F. Immunol. Rev. 204, 184�194 (2005).

3. Hugot, J.-P. et al. Nature 411, 599�603 (2001).

4. Ogura, Y. et al. Nature 411, 603�606 (2001).

5. Sartor, R. B. Curr. Opin. Gastroenterol. 21, 44�50 (2004).

6. Kullberg, M. C. et al. J. Exp. Med. 203, 2485�2494

(2006).

7. Mazmanian, S. K. & Kasper, D. L. Nature Rev. Immunol. 6, 849�858 (2006).

8. Stingele, F. et al. J. Immunol. 172, 1483�1490 (2004).

9. Braat, H. et al. Gut 56, 351�357 (2007).

10. Elliott, D. E., Summers, R. W. & Weinstock, J. V. Int. J. Parasitol. 37, 457�464 (2007).

11. Dethlefsen, L., McFall-Ngai, M. & Relman, D. A. Nature 449, 811�818 (2007).

12. http://nihroadmap.nih.gov/hmp

See Editorial, page 563.

CANCER

Whispering sweet somethings

Thea Tlsty

That genetic mutations contribute to cancer is undisputed. What now emerges is that a cancer cell�s microenvironment has a much stronger hand in the course a cancer takes than previously thought.

The goal of personalized medicine is to tailor a treatment to a specific cellular target at the appropriate stage of a disease, thus �defusing� the disease process. Cancer is an example of the way in which multifaceted approaches to attaining this goal are emerging. We have come to appreciate that a tumour is a collection of diverse cells � cells carrying cancer-causing mutations and the cells of its immediate microenvironment - that act in concert towards disease progression1,2. Three studies3�5 illustrate how these cells collude, and focus on the contribution of non-tumour cells. Within tissues, epithelial cells are supported by a connective framework called the stroma, which itself consists of specific cells, such as fibroblasts, endothelial cells and immune cells, as well as the extracellular matrix. Physiological processes occurring in this compartment, for example the development of new blood vessels in response to oxygen shortage, and host immune responses, could dictate cancer progression.

Writing in Nature Medicine, Finak et al.3 set out to examine how gene-expression profiles in cells of the stroma are affected by cancer. Comparing morphologically normal and tumour stroma from the breast tissue of patients with breast cancer, they identify gene-expression patterns that are strongly associated with a specific outcome of the disease and that could be used as predictors of disease progression.

One specific predictor, a group of 26 genes that the authors call the stroma-derived prognostic predictor (SDPP), stratifies the risk of breast-cancer progression using molecular markers that are independent of � but add power to � both standard clinical prognostic factors, such as the presence or absence of tumour cells in adjacent lymph nodes, and the more recently described6 predictors based on gene expression. SDPP identifies stromal subtypes that have gene-expression profiles relating to a good or poor outcome of breast cancer.

The clinical significance of work such as that of Finak et al. is twofold. First, discerning the subtleties of cell�cell interactions within the microenvironment of a malignant lesion (a localized, disease-associated change in a tissue) will indicate which particular therapy might be most effective for the specific biology of that tumour. Second, such insights could provide targets for developing new therapies. Finak et al. find that SDPP is not affected by treatment, suggesting that existing therapies do not target host responses that affect SDPP genes.

Reporting in Proceedings of the National Academy of Sciences, Postovit et al.4 use a contemporary approach to address the question of the stromal contribution to cancer malignancy. In this exciting study, the authors use an in vitro three-dimensional model that exposes cancer cells to the microenvironment to which human embryonic stem cells are normally exposed; they were hoping to identify conditions in the stroma that suppress malignant characteristics of cancer cells.

Stromal cells surrounding embryonic stem cells secrete a protein factor called Lefty, which inhibits the Nodal protein. Nodal, which during embryonic development prevents stem-cell differentiation, is abnormally expressed in human tumour cells, causing malignancy7. Postovit and colleagues found that metastatic tumour cells do not express Lefty.

Their results strongly support stromal regulation of malignancy and indicate that Lefty has a suppressive effect on cancer cells. The authors� work also suggests that factors secreted by the tumour stroma, and their derivatives, could be used as treatments to �reprogramme� the differentiation of malignant cells, suppressing tumour development and growth.

Although modulating tumour properties in invasive cancers � as discussed in the Finak and Postovit papers3,4 � could reduce the associated morbidity and mortality, early diagnosis and prevention are even more effective means of preventing cancer-associated death.

To address the clinical problems of cancer at these early stages, understanding the molecular processes underlying cancer initiation and progression is crucial. A paper by Hu et al.5 published in Cancer Cell addresses the mechanism of breast-cancer transition from a localized (in situ) lesion to an invasive form.

The authors used a cell line that, when injected into mice, mimics aspects of an early, non-malignant form of human breast cancer called ductal carcinoma in situ (DCIS)8, by forming non-invasive lesions in the animals� mammary gland. They next studied the role of myoepithelial cells in these lesions in suppressing the transition of DCIS to malignancy.

(Myoepithelial cells separate the basement membrane of the duct from the epithelial cells that face the duct lumen.)

Hu and colleagues� functional analysis of cell-type-specific gene expression identified several pathways that could be essential for interactions between stromal fibroblast cells and myoepithelial cells in controlling the integrity of a tissue�s basement membrane. These pathways, which modulate myoepithelial-cell differentiation, are mediated by essential signalling molecules such as TGF-�, Hedgehog, cell-adhesion molecules and the gene transcription factor p63.

Malfunction of these signalling pathways leads to the loss of myoepithelial cells and subsequent invasion of the basement membrane by their adjoining epithelial cells, which respond to signalsoriginating from fibroblasts. Determining whether the loss of myoepithelial cells is a cause or a consequence of the transition from in situ disease to invasive cancer will help to dictate therapeutic strategies.

Myoepithelial cells secrete a protein called maspin, which inhibits degradation of the extracellular matrix, an event thought to be essential for the transition from in situ cancer to an invasive form9. Moreover, this crucial tumour-suppressor protein is postulated to affect tissue invasion, programmed cell death and blood-vessel development. Hu and colleagues 5 identified several extracellular-matrix metalloproteins that are implicated in cancer transition to the invasive state, but maspin is not one of them. Perhaps distinct subtypes of pre-malignant tumours use different pathways for the transition. Recent characterization of of material into the stratosphere, blocking out sunlight: Krakatoa (Indonesia) in 1883, Santa Maria (Guatemala) in 1902, Agung (Indonesia) in 1963�64, El Chich�n (Mexico) in 1982, and Pinatubo (Philippines) in 1991.

These jumps exist in both the land and ocean data. But one shift remained a puzzle: a significant drop in SSTs from 1945 to 1946 that was not replicated in the land data. This shift is also present, but not as obvious, in the un filtered data. In an effort to explain this change, Thompson et al. looked to the metadata - in particular, to the provenance of SST measurements from around 1945.

Figure 1 hints at the explanation; it shows the total number of SST observations from the various national temperature archives. What�s striking is that both the relative fractions and the total numbers of observations vary considerably from year to year. These changes pervade the record, but unsurprisingly the two World Wars (1914�19 and 1939�45) represent the most significant shifts, both in source and in the total number of observations. And here, Thompson et al. suggest, we have the clue to the jump exposed in 1945: whereas during the preceding war years, 80% of measurements came from ships flying the US flag, a resumption of UK measurements at the end of the war saw their proportion jump to some 50%. At that time, unlike their American counterparts who took engine-intake measurements, the British relied primarily on uninsulated-bucket measurements.

So, what are the implications ? Most immediately, a further correction to the global temperature series will be necessary, albeit of a magnitude yet to be assessed. There are many wider ramifications to consider, but one should be handled directly: should we doubt the rise in global mean temperatures during the twentieth century as a result of this or other hidden, and as yet undiscovered, biases in the record ?

The answer is no. According to the filtering of natural variabilities that Thompson and colleagues have done, the only major discontinuity in SSTs is the one in 1945 (although other, insignificant shifts are dotted through the record). The shift from upwards-biased engine-intake measurements to downwards-biased bucket measurements demands a correction; naively speaking, temperatures between 1942 and 1945 would shift downwards by a magnitude of, say, 0.3 �C. Global warming would remain a reality � it would just be a bit more than previously thought.

How does this anticipated correction ripple through to climate models? Global mean surface temperatures are the most widely used data for evaluating the predictive capabilities of models on continental and larger scales6�8. They are also crucial for evaluating two other principal uncertainties in climate predictions: the factors forcing climate change (primarily, levels of aerosol particles in the atmosphere) during the twentieth century and thus in the future6�11; and the rate of heat uptake from the atmosphere to the ocean (Fig. 1 in ref. 7).

The SST adjustment around 1945 is likely to have far-reaching implications for modelling in this period. One particularly striking example can be found in a figure prominently displayed on page 11 of the 18-page �Summary for Policymakers�12 of the IPCC�s Fourth Assessment Report, produced last year. In this, the observed decadal mean temperatures of the 1940s � those that contain the anomalous 3�4year interval dominated by (upwardly biased) US engine-room-intake measurements � are the only ones to lie above model predictions.

Although we don�t know exactly how the temperature record prior to 1946 will be affected by the awaited correction, it is a safe bet that temperatures in this particular decade will be lower.

The 1940s just happen to fall at the end of what seemed to be a warming trend from the 1910s that was followed by a weak decline in global temperatures into the 1960s. Climate modellers have explained the warming as a response to natural forcings13, and the cooling as due to an increase in tropospheric aerosols, principally sulphates, as a result of increased economic activity in the decades following the Second World War. This temporarily offset the effects of man-made warming. Data analysts, on the other hand, have considered the maximum in the 1940s to be the expression of a natural fluctuation14. In light of the new finding1, each interpretation will need to be reconsidered � the first of many implications that will need to be explored. �

Chris E. Forest is in the Joint Program on the Science and Policy of Global Change, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, USA. Richard W. Reynolds is in the US National Climatic Data Center, National Oceanic and Atmospheric Administration, 151 Patton Avenue, Asheville, North Carolina 28801, USA.

e-mails: ceforest@mit.edu;

richard.w.reynolds@noaa.gov

1. Thompson, D. W. J., Kennedy, J. J., Wallace, J. M. & Jones, P. D. Nature 453, 646�649 (2008).

2. www.ipcc.ch/ipccreports/assessments-reports.htm

3. www.climatescience.gov/Library/sap/sap1-1/default.php

4. Kent, E. C. & Taylor, P. K. J. Atmos. Ocean. Technol. 23, 464�475 (2006).

5. Kent, E. C. & Kaplan, A. J. Atmos. Ocean. Technol. 23, 487�500 (2006).

6. Forest, C. E. et al. Science 295, 113�117 (2002).

7. Stott, P. A. & Forest, C. E. Phil. Trans. R. Soc. A 365, 2029�2052 (2007).

8. Knutti, R., Stocker, T. F., Joos, F. & Plattner, G.-K.

Clim. Dynam. 21, 257�272 (2003).

9. Andronova, N. G. & Schlesinger, M. E. J. Geophys. Res. 106, 22605�22612 (2001).

10. Hegerl, G. C. et al. in Climate Change 2007: The Physical Science Basis (eds Solomon, S. et al.) 663�746 (Cambridge Univ. Press, 2007).

11. Meehl, G. A. et al. in Climate Change 2007: The Physical Science Basis (eds Solomon, S. et al.) 747�846 (Cambridge Univ. Press, 2007).

12. www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1spm.pdf (2007).

13. Stott, P. A. et al. Science 290, 2133�2137 (2000).

14. Schlesinger, M. E. & Ramankutty, N. Nature 367, 723�726 (1994).

15. Woodruff, S. D., Slutz, R. J., Jenne, R. L. & Steurer, P. M. Bull. Am. Meteorol. Soc. 68, 1239�1250 (1987).

16. Kent, E. et al. Bull. Am. Meteorol. Soc. 88, 559�564 (2007).

17. Slutz, R. J. et al. Comprehensive Ocean�Atmosphere Data Set;

Release 1 (NOAA Environmental Research Laboratories,

Climate Research Program, Boulder, CO, 1985).

IMMUNOLOGY

PDF On This Available From: tccc.uk@btinternet.com

This page has been visited times.

Email page

Feedback

Home