Professor Arlette Darfeuille-Michaud: The Discovery of Adherent-invasive Escherichia coli

Dr Arlette Darfeuille-Michaud’s research career began in the laboratory of Dr Bernard Joly at Auvergne University in Clermont-Ferrand, France, where she earned a PhD in 1987. Until her promotion to professor of microbiology and molecular biology at Auvergne University in 1994, she studied the pathogenesis of Escherichia coli strains associated with diarrhoeal diseases and Klebsiella pneumoniae strains involved in nosocomial infections. These studies provided important epidemiological information and identified novel pathogenic components, which could potentially be targeted for preventing or curing diseases caused by these pathogens. She was involved in the discovery of a number of bacterial adhesive factors such as the colonisation factor antigen CFA/III,¹ antigen 2230,² nonfimbrial adhesin A (NfaA),^3,4 and F17c fimbria⁵ in diarrhoea-associated E. coli strains, and CF29K^6,7 and KPF-28⁸ in K. pneumoniae strains. She helped identify a new extended-spectrum β-lactamase CTX-1 (in view of its strong activity against cefotaxime), which confers multiple-antibiotic resistance to K. pneumoniae strains.⁹ As part of these studies, she demonstrated that the human colon carcinoma cell line Caco-2 was suitable for detecting enterotoxigenic E. coli (ETEC) strain adhesion, establishing the first cell line model system to study ETEC adhesion to enterocytes.¹⁰ This model was subsequently utilised to show that a Lactobacillus acidophilus strain isolated from human stools effectively inhibited ETEC adhesion to intestinal epithelial cells (IEC).¹¹

Building on her vast expertise in intestinal diseases and E. coli, Dr Darfeuille-Michaud became interested in the role of bacteria in the pathogenesis of inflammatory bowel disease (IBD). Although numerous lines of evidence gathered from human studies and preclinical models indicated that bacteria played an important role in the development and perpetuation of IBD,¹² no IBD clinical isolates were available to dissect mechanisms of action. In 1998, she reported that many E. coli strains isolated from the ileal mucosa of patients with Crohn’s disease (CD) adhered to IEC, but were devoid of any known virulence factor-encoding genes harboured by typical pathogenic E. coli involved in acute gastrointestinal diseases.¹³ These E. coli strains thus belong to a new E. coli pathovar, which she later designated as ‘adherent-invasive E. coli (AIEC)’.¹⁴ The high prevalence of AIEC specifically in the ileal mucosa of patients with CD^15,16 was confirmed by a number of other independent investigators (for reference, see a recent review¹⁷).

This seminal observation marked a turning point in Dr Darfeuille-Michaud’s research career and the field of IBD research, leading to a line of investigation regarding the role of AIEC in IBD pathogenesis and mechanisms of action in the following years.

Dr Darfeuille-Michaud went on to characterise these AIEC and demonstrated their ability to adhere to and invade IEC through an actin microfilament- and microtubule-dependent process,¹⁴ to survive and replicate within macrophages without triggering cell death,¹⁸ and to induce the pro-inflammatory cytokine tumour necrosis factor [TNF]-α from infected macrophages¹⁸ which promotes intramacrophagic replication.¹⁹ Other AIEC properties include replication in mature phagolysosomes within macrophages,²⁰ induction of autophagic neutrophil cell death,²¹ and formation of a biofilm.²²

Using the AIEC reference strain LF82 that she isolated from a CD patient in 1998,¹³ Dr Darfeuille-Michaud identified the key pathogenic determinants of AIEC that mediate bacterial adhesion to and invasion of host cells, which include type 1 pili,²³ flagella,²⁴ and outer membrane vesicles (OMVs).²⁵ She also characterised the bacterial elements and regulatory networks modulating these virulence factors, for example lipoproteins, histone-like proteins, the master transcription regulator FlhD₂C₂, the second messenger cyclic dimeric GMP (c-di-GMP), and the σ^E pathway.^{25,26,27,28,29,30} In addition, she revealed the genetic components promoting AIEC intramacrophagic survival and replication^31,32 and demonstrated that long polar fimbriae mediate AIEC interaction with M cells and Peyer’s Patches.^30,33 The complete genome sequence of LF82 showed the presence of key virulence factors such as bacterial invasin encoded by ibeA, GipA which targets Peyer’s patches, and the pdu gene cluster which is implicated in gut colonisation and intramacrophagic replication.³⁴

Dr Darfeuille-Michaud also identified host factors permitting AIEC colonisation in the intestine. She demonstrated that carcinoembryonic antigen-related cell adhesion molecule 6 (CEACAM6) binds to AIEC type 1 pili³⁵ and the endoplasmic reticulum stress response factor Gp96 to AIEC OMVs,³⁶ both interactions being important for bacterial adhesion/invasion to IEC. Noticeably, she observed that CD patients often have high expression levels of CEACAM6 and Gp96 in ilea epithelia^35,36 and that transgenic mice (CEABAC10 mice) expressing ectopic human CEACAM6 developed severe colitis when colonised with LF82.³⁷

She showed that AIEC via their flagella were able to activate and upregulate toll-like receptor 5 (TLR5) and Nod-like receptor family member IPAF (ICE protease-activating factor), membrane and cytosolic receptors for flagellin, to potentiate an inflammatory response in DSS-treated mice.³⁸ AIEC are also capable of subverting host defence mechanisms, showing fascinating adaptive properties. For example, she reported that AIEC infection upregulated microRNA 30C and 130A, whose function inhibited autophagic response, thereby promoting AIEC intraepithelial survival.³⁹ AIEC infection induces the release of TNF-α from infected macrophages, which promotes CEACAM expression by IEC,³⁵ suggesting that AIEC can promote their own colonisation in the intestine. Other AIEC pathogenic traits that Dr Darfeuille-Michaud helped elucidate include AIEC binding to N-glycosylated chitinase 3-like-1 on IEC which exacerbates colitis in DSS-treated mice,⁴⁰ and their ability to induce the pore-forming tight junction protein Claudin-2 in IEC to disrupt the mucosal barrier.⁴¹

Interestingly, Dr Darfeuille-Michaud also detected AIEC in healthy individuals, although at a much lower number and frequency than in CD patients.^13,15,16 Thus AIEC likely elicit colitogenic effects only in combination with other CD-associated factors (host genetic susceptibility, environmental components, etc.). Indeed, she demonstrated that AIEC could exploit CD-associated host genetic factors to promote disease development. She showed that NOD2, ATG16L1, and IRGM function were essential to restrict intracellular replication of AIEC in infected human epithelial cells and macrophages.^42,43 In an upcoming issue of ECCO-JCC (Vazeille E, Buisson A, Bringer M-A, et al. Monocyte-derived macrophages from Crohn’s disease patients are impaired in the ability to control intracellular adherent-invasive Escherichia coli and exhibit disordered cytokine secretion profile. J Crohn’s Colitis. doi:10.1093/ecco-jcc/jjv053, in preparation) Dr Darfeuille-Michaud’s research group show that macrophages retrieved from CD patients, especially those with autophagy defects due to polymorphisms in NOD2 and ATGL1, are particularly impaired in their ability to control intracellular AIEC replication, which leads to heightened inflammatory responses. In addition, she previously reported that a synonymous SNP in IRGM, associated with CD in Europeans, resulted in the failure of microRNA miR-196 to regulate IRGM expression, which could compromise the control of AIEC intracellular replication.⁴⁴ These studies provided a molecular understanding of how innate immunity could shape host response to microbes and influence development of IBD.

Inhibition of AIEC adhesion and invasion to IEC could represent a preventive and therapeutic strategy for AIEC-mediated chronic gut inflammation. Dr Darfeuille-Michaud reported promising results using probiotics such as E. coli Nissle 1917⁴⁵ and Lactobacillus casei DN-114001,⁴⁶ and molecules targeting bacterial type 1 pili^47,48 in vitro. Most encouragingly, she recently demonstrated that Saccharomyces cerevisiae CNCM I-3856 inhibited LF82 adhesion to the brush border of CD enterocytes and reduced LF82-induced colitis in CEABAC10 mice.⁴⁹

Dr Darfeuille-Michaud was involved in the studies showing that a high fat/high sugar Western diet or a low methyl diet could promote AIEC gut colonisation and AIEC-induced intestinal inflammation in CEABAC10 mice.^50,51 Once again, this work underscores the concept that IBD results from interactions between environmental factors (eg diet), bacteria, and host genetics.

In summary, the seminal work of Dr Darfeuille-Michaud has led to the identification and characterisation of AIEC. The isolation of IBD clinical isolates allows the researchers to explore mechanisms by which microbes promote development of intestinal inflammation. Dr Darfeuille-Michaud’s impressive body of work has greatly advanced our understanding of CD pathogenesis and host-microbe interactions in the intestine. The field of IBD is indebted to Dr Darfeuille-Michaud’s pioneering work on AIEC. Although her passing has created an immense void in the scientific community, her research legacy is assured by the continuous work of numerous laboratories dedicated to the elucidation of dysregulated host-microbe interactions in the intestine, particularly the action of AIEC.

Conflict of interest statement

None declared.