GAITing the GUT

The gastrointestinal tract contains one of the most abundant populations of macrophages in the body.¹ These cells play a critical role in the maintenance of gut homeostasis and defend the integrity of epithelial crypts and the mucosal membrane. Chronic disruption of gut homeostasis can cause sustained inflammation mediated by intestinal macrophages. Since inflammation plays a key role in ulcerative colitis (UC) and other forms of inflammatory bowel disease (IBD),^2–4 an improved understanding of how inflammation is regulated in the gut is essential for the development of effective therapies for such diseases.

Intestinal homeostasis is maintained by a population of resident macrophages that is sustained through recruitment of circulating monocytes (precursors of macrophages) as well as migration of macrophages from the lymphatic system to target mucosal tissues. These processes are driven by the concerted action of an array of chemokines and chemokine receptors.^{5, 6} In addition, the diverse microorganisms that densely populate the intestine continuously attract macrophages. Macrophages may encounter commensal microorganisms in the draining lymph nodes or during occasional compromise of the gut epithelial barrier.^{7, 8} Macrophages have diverse and plastic phenotypes and undergo a series of changes during their response, which is spatially and temporally regulated by multiple and often self-limiting signals. Several intermediate phenotypes between the so-called classically activated M1 (pro-inflammatory) and alternatively activated M2 (anti-inflammatory) phenotypes have been identified.^{9, 10} A surprising characteristic of intestinal macrophages is their ability to maintain a noninflammatory state. As shown by many studies, this noninflammatory state is maintained through downregulation of diverse molecules involved in the innate immune response of intestinal macrophages.¹¹ However, whether macrophages resolve intestinal inflammation by triggering a single mechanism that targets a cohort of inflammatory genes, or whether this is achieved through the concerted action of several mechanisms was previously unknown.

Our work published in Cellular & Molecular Immunology in 2016 identified a single physiological defense mechanism against inflammation that is active in intestinal macrophages.¹² Using a murine model of dextran sodium sulfate (DSS)-induced experimental colitis, we discovered a ribosomal protein L13a-dependent mechanism by which translation of inflammatory factors is suppressed in intestinal macrophages. This work built upon our previous discovery of a unique translation control mechanism mediated by a GAIT (interferon (IFN)-gamma-activated-inhibitor of translation) sequence element present in inflammation-associated mRNAs, such as IFN-γ-induced human ceruloplasmin (Cp) mRNA.¹³ As a part of the multiprotein RNA-binding GAIT complex, in macrophages, L13a directly blocks translation of a group of mRNAs encoding chemokine and chemokine receptors.¹⁴ This silencing relies on binding of the L13a-containing GAIT complex to the GAIT element present in the 3′ untranslated region (3′UTR) of its target mRNAs.¹⁵ The physiological significance of these observations was tested in our laboratory by creating mice with abrogated expression of L13a specifically in cells of myeloid origin, such as macrophages. Using the inflammation model of lipopolysaccharide (LPS)-induced endotoxemia¹⁶ and high-fat diet-induced atherosclerosis,¹⁷ we found that these macrophage-specific L13a-KO mice had more severe disease compared to control mice. Together, these studies suggest that defects in the L13a-dependent GAIT-mediated translational silencing pathway can contribute into the pathogenesis of many human diseases caused by chronic or overactive inflammation. The model shown in Fig. 1 depicts the role of this pathway in the pathogenesis of inflammatory diseases.

Fig. 1.

Silencing of a group of inflammatory proteins through a ribosomal protein L13a-dependent pathway provides an endogenous defense against many diseases

The “cytokine storm” generated by intestinal macrophages upon occasional or repeated exposure of these cells to commensal microorganisms plays an essential role in the pathogenesis of UC. Therefore, whether L13a-mediated translational silencing of a cohort of inflammatory proteins in macrophages could serve as a physiological defense mechanism against UC was investigated. To address this question, we compared the severity of disease in macrophage-specific L13a-KO mice and control mice fed DSS-containing water to induce UC. Disease severity was scored based on the clinical features of UC, including colorectal bleeding, bloody diarrhea, weight loss, shortening of the colon and death. We found that macrophage-specific deficiency of L13a significantly increased the severity of disease in this model. In the colons of KO mice, widespread and extensive damage to epithelial crypts, infiltration of macrophages and significant elevation of an array of inflammatory cytokines were observed. Moreover, consistent with the known contribution of penetration of commensal microorganisms through the damaged epithelium in UC, we detected live bacteria and significantly higher levels of endotoxin in the plasma of the KO mice than in the plasma of the controls.¹²

Although the above data are very intriguing, they do not directly explain the molecular basis of the increased severity of DSS-induced UC observed in macrophage-specific L13a-KO mice. To further test our hypothesis that abrogation of L13a-dependent translational silencing of GAIT target mRNAs determines disease severity, we conducted two sets of experiments. First, using luciferase reporter RNA with a natural GAIT element in its 3′-UTR and a rabbit reticulocyte lysate (RRL) in vitro translation system, we reconstituted GAIT-mediated translational silencing ex vivo by adding lysates prepared from the colons of DSS-fed macrophage-specific L13a-KO and control mice. Colon lysates from DSS-fed control mice activated translational silencing of the GAIT element-containing reporter RNA, but those from L13a-KO mice did not. Second, we tested endogenous GAIT-dependent translational silencing of three previously identified GAIT target mRNAs (CXCL13, CCL22, and CCR3) by comparing their polyribosomal association in colon lysates prepared from the DSS-fed control and L13a-KO mice. The polyribosomal abundance of all three mRNAs was substantially higher in the mice lacking expression of L13a in macrophages. Together, these results demonstrate that L13a-dependent GAIT element-mediated translational silencing serves as an endogenous defense mechanism against intestinal inflammation.¹²

Any disruption of the integrity of the intestinal epithelial barrier results in confrontation of intestinal macrophages with commensal bacteria. These conditions trigger an innate immune response through activation of pattern recognition receptors, such as toll-like receptors (TLRs),¹⁸ and nucleotide binding oligomerization domain (NOD)-like receptors,¹⁹ and subsequent synthesis of an array of inflammatory cytokines. Our work shows that activation of GAIT-mediated translational silencing in response to IFN-γ in the cellular model^13–15 and upon LPS¹⁶ and DSS¹² administration in animal models is critically important to restrain the expression of a cohort of inflammatory proteins. The presented evidence suggests that the onset of such silencing is programmed into the initial phase of the innate immune response. In addition, we identified features of GAIT elements that likely have roles in controlling the degree of silencing. We found that while no sequence conservation exists among the GAIT elements present in the mRNAs of different target chemokines and chemokine receptors, these elements share a conserved, characteristic hairpin RNA structure (as revealed by 2D-1H NMR spectroscopy of the GAIT elements of CCL22, CXCL13, and CCR4 mRNAs). Despite this similarity in secondary structure, the GAIT elements present in these mRNAs and in the Cp mRNA showed substantial differences in affinity towards the GAIT protein complex.²⁰ These differences are consistent with more than “all-or-none” control and may allow temporal silencing of targets depending on the microenvironment of the site of inflammation, which may subsequently allow fine-tuning of the resolution phase of inflammation.

The pathogenesis of IBD, UC, and Crohn’s disease is critically dependent on the concerted inputs of multiple signals from an array of inflammatory molecules. Therefore, targeting a single molecule is unlikely to achieve an effective therapeutic outcome. In contrast, our discovery of the L13a-dependent translational silencing mechanism, which impacts the synthesis of multiple inflammatory proteins, bears significant promise for the development of future clinical intervention strategies against these diseases. Our system for reconstitution of L13a-dependent translational silencing of GAIT element-containing reporter RNAs using colon lysates thus represents an invaluable model for drug discovery. This platform can be used to screen libraries of small molecules to identify compounds that promote silencing. The potential of the identified hits for human clinical trials could be further evaluated using a murine model of DSS-induced UC²¹ or organoid cultures from patients with UC.²²

Competing interests

The authors declare no competing interests.