ObjectiveThe anti-α4β7 integrin antibody vedolizumab is administered at a fixed dose for the treatment of IBDs. This leads to a wide range of serum concentrations in patients and previous studies had suggested that highest exposure levels are associated with suboptimal clinical response. We aimed to determine the mechanisms underlying these non-linear exposure-efficacy characteristics of vedolizumab.DesignWe characterised over 500 samples from more than 300 subjects. We studied the binding of vedolizumab to T cells and investigated the functional consequences for dynamic adhesion, transmigration, gut homing and free binding sites in vivo. Employing single-cell RNA sequencing, we characterised α4β7 integrin-expressing T cell populations ‘resistant’ to vedolizumab and validated our findings in vitro and in samples from vedolizumab-treated patients with IBD. We also correlated our findings with a post-hoc analysis of the Gemini II and III studies.ResultsRegulatory T (TReg) cells exhibited a right-shifted vedolizumab binding profile compared with effector T (TEff) cells. Consistently, in a certain concentration range, the residual adhesion, transmigration, homing of and availability of functional α4β7 on TReg cells in vivo was higher than that of/on TEff cells. We identified a vedolizumab-‘resistant’ α4β7-expressing β1+PI16+ TReg cell subset with pronounced regulatory properties as the substrate for this effect. Our observations correlated with exposure-efficacy data from Gemini II and III trials.ConclusionCompletely blocking TEff cell trafficking with vedolizumab, while simultaneously permitting residual homing of powerful TReg cells in an optimal ‘therapeutic window’ based on target exposure levels might be a strategy to optimise treatment outcomes in patients with IBD.

The Gram-positive bacterium Enterococcus faecium is becoming increasingly prevalent as a cause of hospital-acquired, antibiotic-resistant infections. A fundamental part of research into E. faecium biology relies on the ability to generate targeted mutants but this process is currently labour-intensive and time-consuming, taking 4 to 5 weeks per mutant. In this report, we describe a method relying on the high recombination rates of E. faecium and the application of the Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR)-Cas9 genome editing tool to more efficiently generate targeted mutants in the E. faecium chromosome. Using this tool and the multi-drug resistant clinical E. faecium strain E745, we generated a deletion mutant in the lacL gene, which encodes the large subunit of the E. faeciumβ-galactosidase. Blue/white screening using 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-gal) could be used to distinguish between the wild-type and lacL deletion mutant. We also inserted two copies of gfp into the intrinsic E. faecium macrolide resistance gene msrC to generate stable green fluorescent cells. We conclude that CRISPR-Cas9 can be used to generate targeted genome modifications in E. faecium in 3 weeks, with limited hands-on time. This method can potentially be implemented in other Gram-positive bacteria with high intrinsic recombination rates.

Background Closing mucosal defects to reach mucosal healing is an important goal of therapy in inflammatory bowel disease (IBD). Among other cells, monocyte‐derived macrophages are centrally involved in such intestinal wound healing. We had previously demonstrated that the anti‐α4β7 integrin antibody vedolizumab blocks the recruitment of non‐classical monocytes as biased progenitors of wound healing macrophages to the gut and delays wound healing. However, although important for the interpretation of disappointing results in recent phase III trials in IBD, the effects of the anti‐β7 antibody etrolizumab on wound healing are unclear so far. Methods We analyzed the expression of etrolizumab targets on human and mouse monocyte subsets by flow cytometry and assessed their function in adhesion and homing assays. We explored wound‐associated monocyte recruitment dynamics with multiphoton microscopy and compared the effects of etrolizumab and vedolizumab surrogate (‐s) antibodies on experimental wound healing and wound‐associated macrophage abundance. Finally, we investigated wound healing macrophage signatures in the large intestinal transcriptome of patients with Crohn's disease treated with etrolizumab. Results Human and mouse non‐classical monocytes expressed more αEβ7 integrin than classical monocytes and were a target of etrolizumab‐s, which blocked non‐classical monocyte adhesion to MAdCAM‐1 and E‐Cadherin as well as gut homing in vivo. Intestinal wound healing was delayed on treatment with etrolizumab‐s along with a reduction of peri‐lesional wound healing macrophages. Wound healing macrophage signatures in the colon of patients with Crohn's disease were substantially down‐regulated on treatment with etrolizumab, but not with placebo. Conclusions Combined blockade of αEβ7 and α4β7 with etrolizumab seems to exceed the effect of anti‐α4β7 treatment on intestinal wound healing, which might help to inform further investigations to understand the recent observations in the etrolizumab phase III trial program.