This article makes an important connection between ocular and gut response to infection and suggests there are mechanisms by which the innate immune system maintains a kind of short-term, dynamic and system-wide immunological memory. The authors use nucleic acid detection methods to claim the existence of a microbiome on the corneal surface. An important nuance to remember, however, as demonstrated by Fleiszig and colleagues in their meticulous characterization of the conjunctival and corneal surface, is that the corneal surface microbiome is limited to the conjunctiva and the cornea is microbe free in healthy animals. Fleiszig’s group establish this using a combination of cutting-edge imaging and bacterial culture-based approaches showing that the cornea is devoid of bacteria while the conjunctiva does harbor certain commensal live bacterial strains. In this study, Gadjeva’s group use 16S rRNA gene amplification to characterize the ocular surface microbiome, with the admittance that the actual numbers of live bacteria on the conjunctiva are very low, especially in comparison to the levels found in the gut. The authors present a hypothesis relating the ocular and gut microbiome to the development of ocular disease after bacterial challenge. This study elaborates upon a well-known phenomenon that germ free mice are more susceptible to infection after ocular bacterial challenge, and this finding is recapitulated in this work. The authors then explore the molecular mechanism underlying the protective effect of the ocular microbiota. Mass spectrometry was used to compare the protein profiles of the eye washes from germ free compared to normal mice, and showed decreased antimicrobial activity of germ free mice, including lower levels of IgA and complement. Interestingly, following bacterial challenge, germ free mice (which are more susceptible to ocular bacterial infection) have higher levels of proinflammatory cytokines IL-6, IL12p70, and KC and a trend for higher levels of proinflammatory TNF alpha, INF gamma and anti-inflammatory IL-10. Together, the molecular data supports the fact that germ free mice are less able to kill bacteria and display a higher degree of pathology due to an enhanced inflammatory response. Importantly, antibiotic treatment was found to reduce the resistance to bacterial challenge, which could be rescued by inoculating with a commensal bacterial strain, suggesting the immune defense is short-lived and maintained by the active and ongoing immune response to commensal bacteria. The authors demonstrate a defect in the antibacterial properties of neutrophils in germ free and antibiotic treated mice, and they demonstrate a protective role for IL1beta by treating with an IL1beta neutralizing antibody. These findings are of great significance, because the data suggests that mice with an active immune response (induced by dead bacteria stimulating the innate immune signaling molecules) are better protected against future infections. This missing piece helps make sense of the finding that gut microbiota can somehow offer protection against ocular bacterial challenge. This finding makes less sense if we consider the gut microbiome and the ocular microbiome as living and dynamic microbial communities having essentially equivalent effects on ocular immune response. The authors demonstrate how the gut and ocular microbial debris both contribute to ocular immunity through enhanced interleukin 1 beta (IL1beta) signaling. In fact, it is the bacterial nucleic acid and not the actively dividing, living bacteria that are the most likely to stimulate the toll like receptor signaling required to induce the secretion of interleukin 1 beta. Unlike an adaptive immune response, in which cellular memory is preserved in T cell receptors, B cell receptors and antibodies, the mechanism by which IL1beta enhances innate immunity within the entire organism is not understood. Together, the data suggests that the host response to prior bacterial encounters boosts health and prevents future bacterial infections. The role of interleukin 1 beta in ocular health is as yet not fully understood, but its importance appears in every study of trachoma pathogenesis, an ocular disease caused by Chlamydia trachomatis that begins as a conjunctival inflammation and can lead to severe keratitis induced by potentially many different types of commensal bacteria, fungi or even viruses. IL1beta levels are consistently found at very high levels in patients with late or end-stage trachoma. The data in this study suggests that the high levels of IL1beta found in late-stage trachoma patients is consistent with the high levels of IL1beta present in eyes that have been repeatedly microbial challenged, and perhaps the trachoma dataset should be viewed from this new perspective in which IL1beta is considered the remnant of a healthy immune response to repeated bacterial challenge, and not a sign of a damaged and diseased ocular surface that mediates tissue damage. Based on the results in this work, the enhanced levels of IL1beta may in fact confer a protective effect in these trachoma patients.
Pseudomonas aeruginosa, an opportunistic pathogen, is a leading cause of bacterial keratitis and the subject of many experimental models of keratitis pathogenesis. In this study, the steps involved in the proper resolution of inflammation in a mouse model of keratitis is elucidated, focusing on lipid mediators of inflammation. While eicosanoids such as prostaglandins, leukotrienes and thromboxanes initiate ocular inflammation, it is the lipoxins, resolvins, and protectins that are involved in the resolution of inflammation. The 15-lipoxygenase enzyme has been shown to be the present in the cornea and suggest it may be the primary corneal enzyme responsible for generating these pro-resolving lipids. Timely resolution of ocular inflammation is so critical to the eye especially, because of the threat to vision and the potentially dire consequences of vision loss to the quality and length of life. Part of the reason for the lack of studies on the lipid mediators of inflammation is the technical hurdles involved in performing lipid mass spectrometry. Importantly, lipid LC-MS/MS was an essential component of this study. The second essential aspect of this study is the use of two different mouse models of keratitis, one being susceptible (B6) and one being more resistant (BALB-c) to the development of keratitis following Pseudomonas aeruginosa infection via injection into the corneal stroma. The authors measured the levels of arachidonic acid (the pro-resolving cyclooxygenase substrate and the proinflammatory lipoxygenase substrate), a precursor to the lipid mediators of inflammation. Additionally, the authors also measure the levels of various metabolites of arachidonic acid. Importantly, the baseline levels of arachidonic acid are equivalent in both Pseudomonas aeruginosa infected B6 mice and Pseudomonas aeruginosa infected BALB/c mice. This provides sufficient evidence that any differences observed in the levels of down-stream arachidonic acid-derived molecules are solely the result of differences in the arachidonic acid metabolizing enzymes present and active in each mouse. As predicted, differences in metabolite levels were observed, including lipid metabolites that clearly distinguish the cyclooxygenase (COX)-2, 5-LOX and 12/15-LOX pathways. Susceptible B6 mice had higher levels of PGE2 and Thromboxane B2, both resulting from COX-2 activity, as well as higher levels of 5-HETE, the result of 5-LOX activity. BALB/c mice, on the other hand, had higher levels of 12-HETE and 15-HETE. Importantly, these differences could not be seen in the corneas of healthy, uninfected mice, indicating the importance of inducing an immune response in order to be able to observe and record the innate differences in the inflammatory responses of the two mice. Importantly, the amount of COX and LOX enzymes were also measured and found to align with the expected results based on both RNA and protein levels as measured by quantitative real time PCR and western blotting, respectively. Similar results were observed in DHA-derived metabolites. While the levels of DHA were similar in both mice after infection, the levels of the DHA derived metabolites differed. Given all of this interesting data, the reader is left feeling let down by the lack of follow-up in the remainder of the study. The authors elaborate upon these findings by characterizing a 15-LOX knockout mouse. The authors examine the bacterial load, disease scores, immune cell populations, and cytokine and chemokine profile of the 15-LOX mutant mice, showing there is an enhanced susceptibility to infection in these mice. There are difficulties in interpreting the 15-LOX knockout mouse data because of the potential downstream effects of this mutant that may also affect pathogenesis, even if cytokine and chemokine levels do not seem to play a major role, as suggested by the data. The authors do find a striking difference in the cell populations in 15-LOX -/- mutant mice, including significantly fewer neutrophils as compared to both B6 and BALB/c mice, while similar levels of macrophages were found. However, not all cell populations were thoroughly examined and the authors focused more on the levels of necrotic neutrophils and macrophages than on the details of the sub-populations of macrophages. Other studies have demonstrated roles for specific groups of cells; for example, CDR2- macrophages have been shown to have a role in the development of corneal tissue during neonatal development, and these cells also play a role in maintaining a proper immune response. It is unclear whether the knock-out mice are more susceptible to infection as an indirect result of the modified and unnatural levels of various cell types. The mouse model raises more questions than it answers. 15-LOX is supposed to produce pro-inflammatory lipid metabolites, and yet the subject of the study is the pro-resolving lipid mediators. One wonders what might happen to mice that are deficient in COX-2 and if those mice might be protected against infection, or if the pathways are less straightforward. Given the lack of understanding of the role of lipid mediators in resolving inflammation, a more informative study might have taken the approach of attempting to regulate the expression of each of these enzymes to try to recapitulate the phenotypes, to focus on the promoters and regulatory elements that dictate the levels of COX and LOX RNAs available, and to provide potential drug targets to enhance health.
As a preventative measure, newborns are given antibiotic eyedrops to protect against conjunctivitis, otherwise known as pink eye, often caused by the most prevalent sexually transmitted disease-causing bacteria, Chlamydia and Gonnorhea, among other bacteria. This practice makes sense given the high prevalence of undiagnosed and a-symptomatic reservoirs of these sexually transmitted diseases. However, the widespread use of antibiotics as prophylactics is a relatively new medical practice, and the long-term effects are not yet fully understood. In this paper, a very interesting connection is made between antibiotic use and eye development in neonates. The effect of continuous antibiotic use on the development of the cornea is studied in mice. While it might be expected to see some effect of antibiotic use on the development of the mouse immune system, a much more problematic phenotype is observed: corneal morphology, the formation of limbal blood vessels and the development of corneal nerves are all disrupted by the antibiotic treatment. The cornea is demonstrably thin and small and the number of dividing cells reduced by a large percentage in the mice that developed in the presence of an antibiotic treated mother. This has direct and irreversible consequences on vision in the neonate. A deformed cornea is, intriguingly, at least partially avoided if the antibiotic treated mouse is also given fecal transplant therapy. These important controls support the hypothesis that the gut microbiome directly regulates corneal tissue development, and is elaborated upon by studying the role of macrophages in this process. Importantly, other studies support a role for macrophages in tissue development but this is the first paper to demonstrate a role for macrophages in corneal development. The authors leverage their knowledge of the corneal macrophage population and ability to deplete corneal macrophages to show that depletion of CCR2- macrophages negatively affects corneal development and gene expression, while depletion of CCR2+ macrophages has no affect. The use of very relevant controls such as these strongly enhance the quality and believability of this study. Importantly, a phenotypic correlation was observed in the antibiotic treated mice between the presence of the CCR2- macrophage population and corneal deformities, including the observation that fecal transplant treated antibiotic treated mice had restored their CCR2- macrophage population. To confirm the role of CCR2- macrophages in corneal development, the authors not only showed a loss of these cells in antibiotic treated corneas, but also used RNASeq data to confirm the absence of macrophage regulating genes in the antibiotic treated corneas. The discovery of the importance of this particular sub-population of macrophages in corneal development is of particular significance to those interested in ocular surface biology. However, there are also broader implications of this study as it relates to public health practices. Further work is required to understand the threshold level of antibiotic exposure required to cause damage to the human fetus. While most healthy infants will only receive one ocular antibiotic treatment at birth, there are many populations that receive long-term antibiotic treatments, including children born with the genetic disease cystic fibrosis or other immune compromised children, as well as healthy children living in trachoma endemic regions, where prophylactic antibiotic treatment of infants and mothers is recommended by the World Health Organization. The era of antibiotics is clearly not going to last forever, given the prevalence of antibiotic resistant bacteria. Still, studies like this one show that even if we solve the antibiotic resistance problem, there are intrinsic dangers in the over-use of antibiotics, particularly for the developing fetus.
There is potentially a treasure trove of as yet undiscovered protective, health enhancing and antimicrobial compounds produced by the mucosal epithelium that naturally aid in acting as a barrier to infection, waiting to be discovered. Anti-microbial peptides including defensins, LL37 the more recently discovered keratin derived antimicrobial peptides are all secreted by the ocular surface epithelial cells and into the tear fluid. Glycoproteins known to play a role in barrier defense at the ocular surface are mostly the transmembrane constituents of the glycocalyx, including the mucins Muc 16, Muc 1 and Muc 4, as well as the secreted Muc 5 produced by goblet cells within the conjunctiva. Interestingly, in this paper a novel glycoprotein found on the ocular surface is demonstrated to have antimicrobial effects, including against the common ocular pathogen Pseudomonas aeruginosa. In this study, the glycoprotein DMBT1 is identified as a protective factor produced by ocular surface epithelial cells. The mechanism of action is well delineated, and shown to directly affect the twitching motility of the bacteria by affecting the type 4 pili. This phenotype was directly compared to a known twitching motility bacterial mutant, pilA, an important internal assay control for the measurements of changes in bacterial movement and motility. Human tear fluid was found to limit the twitching motility of bacteria, and in a dose-dependent manner, and this was shown to be eliminated by treatment with proteinase K. Importantly, lysozyme and lactoferrin, known tear components, did not affect twitching motility, and neither did artificial tears that are available for purchase over-the-counter. Exclusion chromatography and mass spectrometry identified DMBT1 as the anti-twitching factor in the active fraction, and purified DMBT1 recapitulated the phenotype. This important discovery demonstrates a novel protective mechanism of a glycoprotein that does not involve direct bacterial killing, but instead has a targeted impact on the ability for bacteria to traverse through multiple bacterial layers.
This review paper is highly under-read. It does an excellent job of synthesizing cellular immunology concepts and the less widely known pro-inflammatory and pro-resolution pathways of the arachidonic acid derived lipid mediators of inflammation. Specifically, after reading this review, the reader can appreciate the importance of macrophage recruitment to the ocular surface following the initiation of the inflammatory response for successful resolution of the inflammatory response. Neutrophils are the front line of defense in ocular inflammation, and can be found in very high numbers on the eye upon waking from sleep. When spent neutrophils are not properly removed by recruited macrophages, it leads to the recruitment of T cells and tissue destruction. This sequence of events can be applied to a number of different types of inflammation induced on the ocular surface and the common role for T cells in ocular surface pathology.
Microbiome studies have become popular over the last few years, and their prevalence seems to suggest a major role for microbial communities in influencing human health. The ongoing conversation between commensal microbes and the human body is an important aspect of human health, and the microbiome can be used both for disease diagnosis and treatment. Importantly, in this study, the microbiome of the ocular surface is convincingly demonstrated to be absent from the healthy corneal surface; however, it can be induced to appear in the absence of two known modulators of the immune response also known to be involved in the defense of the cornea against bacterial binding and epithelial barrier penetration. Two novel and complementary technologies were used to directly image live bacteria on the live ocular surface of mice. This approach is preferable to taking ocular swabs and culturing the microbial content because it minimizes the technical biases inherent in culturing. It is also preferable to polymerase chain reaction (PCR) based and other nucleic acid quantification-based approaches, because the viability of the microbial population identified using such techniques cannot be known. Importantly, a healthy immune response involves the direct killing of microbes, and the dead corpses of the microbes may remain long after the threat is gone. The first technology utilizes alkyne-functionalized D-alanine (alkDala), and since bacteria utilize D amino acids, but mammals do not, this amino acid is only incorporated into the metabolically active bacteria present and is absent from the host cells (this was demonstrated experimentally). The alkDala probe will be used as a substrate by bacterial peptidoglycan enzymes during the formation of the bacterial cell wall peptidoglycans. The probe can be fluorescently visualized following addition of an azide-fluorophore using copper-catalyzed click chemistry to create a covalent bond between the alkyne and the azide. In order to detect bacteria that lack peptidoglycan metabolism, a fluorescent in situ hybridization (FISH) approach was also used. FISH was performed against a universal 16S rRNA target. Imaging using these techniques revealed the absence of bacteria on the corneal surface. Importantly, bacteria was detected on the neighboring conjunctival surface, and a DMN-Tre labeling probe specific to Corynebacterium and Mycobacterium species revealed that many of the bacteria identified on the conjunctival surface were likely Corynebacterineae. As a follow-up experiment to demonstrate the validity of the approach, mouse eyes were inoculated with a known amount of Pseudomonas aeruginosa, Staphylococcus aureus and a murine eyelid commensal. The same imaging-based techniques were then performed, bacterial culturing was performed as a way to validate the imaging methodology, and since the assay involved a controlled inoculation with easily culturable bacteria. The inoculated bacteria were recovered from the ocular surface and successfully cultured (at lower levels than were in the initial inoculum), demonstrating the ability of the mouse cornea to clear bacterial challenge. This bacterial culture result has been recapitulated in all other similar studies in the literature. However, the novel aspect of this assay was the imaging aspect, showing that the cornea associated bacteria remain culturable. This new data shows that the bacterial clearance capability of the ocular surface is the result of actual bacterial clearance and not the induction of a non-metabolically active, persistent form of the inoculated bacteria. Intriguingly, this study also makes the discovery that metabolically active bacteria can be found on the corneal surface of IL-1R and MyD88 knockout mice. This finding is in line with previous publications out of the Fleiszig lab showing a role for both of these molecules in preventing the adhesion of exogenously applied bacteria. This is the first evidence to show that these knockout mice harbor a corneal microbiome even before experimentation begins. Both IL-1R and MyD88 are innate immunity molecules involved in mediating cytokine and chemokine secretion during TLR signaling. MyD88 is the universal adaptor protein for all TLR signaling. Colony PCR of 16S rRNA revealed the identity of the bacterial flora found on the IL-1R knockout mouse corneal surface, which was mostly Staphylococcus, Propionibacterium and Bacillus species. This data supports the hypothesis that immunocompromised eyes, but not healthy eyes, harbor a corneal microbiome. This makes sense, given the incredible importance of keeping the corneal region transparent to maintain vision. The presence of a microbiome on the corneal surface enhances the probability for tissue integrity loss, which could lead to entry of bacteria into the corneal stroma and the induction of an immune response, leading to regional corneal opacity (blindness). Of course, as the authors point out, this study did not include an analysis of the viral load, fungal communities or dormant (non-metabolically active) bacterial populations on the corneal surface. However, it does provide a convincing set of data to support the hypothesis that the healthy cornea does not have a microbiome, by demonstrating the absence of viable bacteria on the corneal surface. The state-of-the-art techniques employed here are a critical aspect of this paper and should be considered as complementary tools to the more widely used microbial culture, imaging and nucleic acid detection techniques when performing microbiome studies.
In this important paper, a novel mucosal epithelial cell antimicrobial defense mechanism is discovered and described. Dr. Tam and colleagues perform an elegant assay to identify unknown antimicrobial properties in tear fluid. While both keratinized and non-keratinized (mucosal) epithelial cells are well known to protect against infection, the molecular basis for these innate defense mechanisms are not well understood. Epithelial cells have been studied much more in terms of their role in detecting pathogens and recruiting professional phagocytes and adaptive immune cells. For example, ocular surface mucosal epithelial cells are able to prevent bacterial binding and traversal through the epithelial layer in healthy animals. This defense mechanism requires a disease state, contact lens wear, or damage in order for microbial penetration to occur. While the physiology of the entire ocular surface participates in this innate barrier defense mechanism, it also includes chemical defenses secreted by the ocular epithelial cells, such as defensins and LL-37, which are secreted into the tear fluid. Some of the known secreted antimicrobials are not constitutively expressed and require toll-like receptor (TLR) activation by microbial pattern associated molecular patterns (PAMPs). In this assay, a screen was performed to identify novel constitutively expressed antimicrobial proteins in telomerase-immortalized human corneal epithelial cells. Unlike other immortalized cell lines, these cells mimic primary cells in a number of ways including the ability to differentiate, desquamate and stratify into multilayers. These cultured cells were differentiated in calcium, then the lysate was fractionated and each fraction mixed with bacteria (Pseudomonas aeruginosa). The 3- 10 kDa fraction was shown to be bactericidal and was sent for mass spectrometry analysis of the constituents. Several keratin fragments were identified in this fraction, including fragments of keratins 5, 6A, 15 and 19. Of these, the keratin 6A derived fragments were present most consistently and in highest abundance. Importantly, the production of the keratin 6A fragments depended upon growing the cells in the presence of sufficient calcium. The reason for this was not elaborated upon but may indicate possible trigger pathways for the secretion of keratin 6A, whether related to cellular differentiation or other calcium dependent signaling pathways. A set of experiments using synthesized peptides followed. Specificity of the antimicrobial activity of the keratin 6A derived fragments, coined KDAMPs (keratin-derived anti-microbial peptides), especially the highly potent 19-mer fragment, was demonstrated by including a 19 amino acid long scrambled peptide control in each assay. A concentration of 200 micrograms per milliliter was able to reduce in vitro cellular infection rate and cytotoxicity more than 80%. Dose-dependent antipseudomonal activity was observed for the 19-mer, as well as multiple longer and shorter keratin 6A KDAMPs. The bactericidal activity of the 19-mer and other K6A-derived peptides against other bacteria was also tested, including Streptococcus pyogenes, Serratia marcescens, Staphylococcus aereus and Staphylococcus epidermidis. However, the killing activity depended upon the peptide, bacteria and concentration. Even so, all of the bacteria tested could be killed with at least one of the tested K6A-derived peptides, suggesting a reliable host survival strategy. Importantly, keratin 6A also demonstrated a role in the cytoprotective properties of cells. siRNA mediated knock down of keratin 6A in live mice via subconjunctival injection resulted in an enhanced rate of bacterial adherence to the epithelial surface, an important indicator of corneal susceptibility to infection. These antimicrobial peptides are interesting for reasons unrelated to their biological function as well and may be an interesting topic for research for protein chemists and structural biologists. The mechanism of action underlying the bactericidal properties of the K6A-derived peptides was explored using TAMRA tagging of the peptides to allow for fluorescence imaging, and demonstration of fluorescent labeling of the bacteria following incubation with the KDAMPs. Additionally, the fast killing action (between 5 and 15 minutes) of the peptides suggest a bacterial cell permeabilization based mechanism. The apparent mechanism of action is cell wall lysis, yet these peptides do not have the canonical protein folding motifs typically associated with cell lysis. Unlike other known antimicrobial peptides, these cytokeratin 6 derived antimicrobial peptides do not require an overall cationic charge or the presence of alpha helices, but instead require glycine residues. The exact method by which these peptides create pores in bacterial cell walls and retain their selectivity for microbes have yet to be elucidated, and may be novel. However, pore formation may not be the mechanism of bactericidal action of KDAMPs and the data presented leaves much room for testing alternative hypotheses.
Dry eye is a multi-factorial disease of the corneal surface that manifests as a problem with the tear film. In many patients this means decreased tear production, but in some people there can be more tear production than normal due to the over-compensation of the lacrimal gland, the gland responsible for secreting the salty fluid component of the tear fluid. In addition to the lacrimal gland, the meibomian glands may also be involved in the etiology of dry eye disease in some patients. The meibomian gland secretes the lipid component of tears that prevents the evaporation of tear fluid. In many patients, the meibomian gland dysfunction is the underlying reason for rapid evaporation of tear fluid resulting in dry eye. In addition to these glands, the goblet cells that reside within the mucosal epithelial lining called the conjunctiva are also critical in the maintenance of the tear composition, as the goblet cells secrete large quantities of mucins, protective glycoproteins (mucus). In this study, the results of a phase 3, double masked, placebo controlled, randomized multi-center clinical trial of the newest FDA approved treatment for dry eye disease, lifitegrast, is reported. Lifitegrast was designed to be a small molecule mimic of the LFA-1 binding domain of ICAM-1, and is therefore a competitive antagonist of LFA-1. Many patients with dry eye disease display a higher than normal expression of ICAM-1 on the surface of their endothelial cells and other cell types, which has the effect of binding LFA-1 on circulating lymphocytes. This binding effectively recruits T cells to the ocular surface. Therefore, lifitegrast is an immune cell target with the expected mechanism of decreasing the T cell population on the corneal surface during dry eye disease progression. In this clinical trial, 711 participants presenting with the type of dry eye disease that is associated with lack of tear fluid were treated with lifitegrast or placebo for twelve weeks. No serious ocular adverse events were reported. Improvements were seen in both the lifitegrast treated and placebo treated groups as early as two weeks into the study, with the lifitegrast treatment group showing significant improvement over placebo at all time points. Moreover, the improvements observed were both in terms of signs and symptoms of the disease. For example, the measures of corneal and conjunctival integrity revealed clear physiological improvements (more cells were alive and intact following treatment) on both surfaces. Direct measurements of the amount of tear fluid on the eye using the Schirmer tear test also showed significant improvements. These would be considered the signs and not the symptoms of the disease. The symptoms also improved, as would of course be expected given the improvement in the disease signs. The authors note that this is a limited clinical study in part because two groups were excluded from the study: those that had LASIK within a year of the study (dry eye is often a side effect if LASIK), and those who wear contact lenses. Since both of these groups of people are particularly susceptible to developing dry eye symptoms, yet also more likely to have other complications, future studies will need to address these more vulnerable populations with respect to the safety and efficacy of lifitegrast. To put this study into perspective, there were only four clinical studies of lifitegrast ophthalmic preparation for dry eye disease published prior to this 2017 work. While this study was only 84 days long, another potential limitation of the study, a prior clinical study (the SONATA study) lasted a year and was used to assess the safety of the treatment. Importantly, all of the clinical trials reported to date have had the same positive improvements in dry eye symptoms following treatment with lifitegrast. These trials are very reassuring and lifitegrast is an important addition to the dry eye treatment toolbox. Prior to the introduction of this drug, cyclosporine was the only immune modulatory therapy approved for dry eye disease; the other treatments all involved artificial tears, physiological interventions including heating and massaging the meibomian glands to unplug clogged glands, the use of autologous serum as an alternative tear drop, the use of ductal plugs to prevent the draining of tears (when appropriate), among other interventions. However, of these treatments, only cyclosporine and lifitegrast have shown an ability to reverse dry eye disease in some patients, rather than simply alleviating the symptoms. Both of these drugs target the T cell population that is higher in the dry eye ocular surface compared to the healthy ocular surface, highlighting the importance of the immune response in driving chronic dry eye.
This review paper of medicinal anti-cancer plants used in Sri Lanka is helpful because it pinpoints the molecular targets of several popular plant medicines. While many more plant medicines from around the world (estimated at more than 70,000) and in Sri Lanka have been used to treat cancer, this review demonstrates the wide range of molecular targets of the different plant compounds. Since more than half of the current chemotherapy repertoire comes from plant compounds, it is imperative for biopharma to take another look at the known anti-cancer properties of plants. As is reviewed here, the earliest chemotherapies relied on cytoskeletal modifiers from the plant kingdom, including infamous microtubule-binding vinblastine and paclitaxel, both of which prevent cell division but lack specificity to the relatively rapidly dividing tumor cells. Two other early plant-derived chemotherapies include the topoisomerase inhibitors camptothecin and etoposide. The obvious toxicity that results from the non-specific cell killing mechanisms of action of these chemotherapies should be appreciated. Sri Lankan plant medicines (used by about 70% of the country) introduce us to the fact that ginger is anti-angiogenic and anti-inflammatory by suppressing tumor necrosis factor alpha; turmeric is a widely studied spice known to limit or stop tumor growth, and also has anti-inflammatory properties including downregulating COX-2, an arachidonic acid derived lipid mediator of inflammation. These phenomenal health promoting qualities of ginger and turmeric are very much unlike the toxic earlier chemotherapies described above. The next eight plants described in this article are less well known around the world but have all demonstrated cytotoxic effects against tumor cells using a wide variety of mechanisms. These mechanisms include manipulation of the recruitment of specific inflammatory cell populations, prevention of the secretion of extra-cellular matrix degrading enzymes (matrix metalloproteinases), and the initiation of apoptosis, with each plant using unique molecular strategies to induce apoptosis in the tumor cells. There are a number of reasons this study and those like it are important, including the fact that Sri Lankans are more likely to use and have access to these plant medicines over mainstream pharmaceuticals. Moreover, these traditional plant medicines use multiple complementary mechanisms of therapeutic action which can be utilized to a greater degree in the development of new pharmaceuticals and in the study of cancer progression.
This 2018 study by Chen and colleagues is especially important to many different niche interest groups, including those interested in studying Pseudomonas aeruginosa (PA) infection mechanisms, those studying the ocular inflammatory response, and anyone interested in keratitis etiology, because it ties together several loose ends. The focus of this paper is STING, a molecular readout of cytoplasmic DNA. The activation of STING following PA infection implies several things, including the fact that some bacteria must have entered the cytoplasm (PA is typically found protected within intracellular niche environments), and that an anti-inflammatory host innate immune response is initiated by the infected cell. The latter has implications for the ocular surface in particular, because the inflammatory response of the cornea is much more sensitive than other organs and body systems. A slight imbalance in the initiation or resolution of inflammation in the cornea will lead to blindness, and blindness due to keratitis can only be addressed through corneal transplantation. However, corneal transplantation often cannot be considered when the ocular surface has an active bacterial infection, for obvious reasons. Hence, keratitis as a result of ocular surface infection with PA will often lead to irreversible blindness. This is the first paper to establish that Pseudomonas aeruginosa activates STING.
PA is both an extracellular and intracellular bacterium. It is well known for its ability to thrive outside of cells, where it can form deadly biofilms. However, it can also enter cells and become an intracellular bacterium. Inside the cell, PA initially is phagocytosed and lives within a membrane-bound organelle that is related to the endosome-lysosome. Virulent PA is capable of evading impending death within the lysosome using a number of as yet not fully elucidated mechanisms, but involving the type III secreted factor Exoenzyme S. After escaping the endo-lysosomal pathway, PA lives and grows within the bleb-niche environment, separated from the host cell cytoplasm. The Fleiszig lab has published extensively on this topic. Given the general knowledge that PA, once it becomes intracellular, can activate STING, implies that at least some of the PA population enters the cytoplasm. This is novel information that needs to be further analyzed in order to understand where it fits into the existing knowledgebase. It is especially important to determine whether all intracellular PA spend part of their intracellular lifecycle within the host cytoplasm, or whether only a sub-population of intracellular PA become cytoplasmic. If it is the later, this knowledge would aid in further elucidation of appropriate PA drug and therapy targets, because it implies that some PA are more virulent than others. If all intracellular PA become cytoplasmic at some point, then it is important to establish when that occurs since it goes against the current understanding.
It is well established that the ocular surface will respond to PA infection with a very pro-inflammatory cytokine and chemokine secretory response as a result of toll-like receptor (TLR) signaling pathways being activated by bacterial pathogen associated molecular patterns (PAMPs). Typically, this immune response does not occur in healthy animals because of the anatomy and physiology of the cornea. When bacteria encounter the corneal surface, they are frequently washed away by tear film, killed by secreted antimicrobial compounds in the tears, and trapped within secreted mucins. Binding of bacteria to the corneal surface superficial epithelial cell layer is prevented by the presence of the glycocalyx, a complex organized matrix of glycoproteins (transmembrane mucins) and associated molecules. If bacteria do manage to penetrate the epithelial layers, they are not likely to pass through the underlying basement membrane, because of its small, approximately 0.2 micron pore size. If bacteria do not pass through the basement membrane and enter the stroma, no inflammatory response is initiated. Entry of bacteria into the stroma causes corneal opacity (blindness). For these reasons, researchers studying keratitis, including the authors of this study, will employ the scratch model of disease, which entails making an incision through the entire epithelium and basement membrane, and inoculating PA directly into the stroma to elicit an inflammatory response and thereby initiating varying degrees of keratitis.
This study demonstrated that STING protein production is enhanced in mouse eyes inoculated with PA using the scratch model of keratitis, using flow cytometry to determine that it is mostly the newly infiltrating macrophage population that expressed STING following PA infection. siRNA of STING in live mice showed a striking enhancement in the amount of pro-inflammatory cytokine expression (at the mRNA level) and an increase in the bacterial load. The reverse was observed as well; activation of STING via ocular (subconjunctival) injection of cGMP showed the opposite phenotype. These results were recapitulated in immortalized cell lines. The authors go on to use the cell culture assays to show that STING knockdown increases the amount of nuclear NFkappaB in PA infected cells, while cGMP stimulation decreases it. Further data, including an observation that iNOS expression was decreased in STING knockdown cells, all support the conclusions of the authors that the activation of STING induced by infection with Pseudomonas aeruginosa results in enhanced bacterial killing and an anti-inflammatory response.
It remains yet to be determined to what degree the anti-inflammatory response induced by STING contributes to PA infection progression, and whether this knowledge can be leveraged to minimize the destructive order of events following PA infection. Importantly, answering the questions about the intracellular lifecycle of PA outlined above may shed light on this question. Regardless, this work is an important contribution to our understanding of keratitis etiology.
Longitudinal studies in trachoma research are a rarity, and this study following Tanzanian children for four years every three months by Ramadhani and colleagues is exceptionally important. In addition to providing essential longitudinal data, this study also provides an interesting dataset to better understand what leads to scar tissue formation in certain individuals. A full 23% of the study population (103/448 participants) presented with scarring progression. This large population with scar tissue progression lends more credence to the study. The data shows a strong correlation between the frequency of presentation with papillary inflammation (TP) and scarring progression. The data further shows that bacterial load was equivalent between individuals with and without scarring. Given the presence of an association between presence of Chlamydia trachomatis and papillary inflammation, but the lack of association between bacterial load and scarring progression, the data as a whole strongly supports the hypothesis that the host response to Chlamydial infection is what drives disease, and that the host response gets slowly worse with multiple infections. The data clearly shows that increased frequency of conjunctival papillary inflammation is strongly associated with scarring progression. This information, while not exhaustive in explaining the etiology of disease progression, is certainly useful for health care providers who are empowered by the knowledge that interfering with papillary inflammation through prevention and/or treatment is likely to help prevent irreversible scar tissue formation which will eventually lead to irreversible blindness.
Azithromycin, an antibiotic belonging to the macrolide family, is the World Health Organization- recommended treatment for trachoma, an eye disease caused by infection with ocular strains of Chlamydia trachomatis. While azithromycin is known to be efficacious against the bacterium, trachoma progression is a more complicated phenomenon that may not be driven by active infection. Trachoma is characterized by the development of scar tissue within the conjunctiva. Almost all epidemiological studies show that some individuals progress to scar tissue formation even in the absence of active infection. Some studies suggest scar tissue formation can be driven by the presence of commensal bacteria that may further exacerbate the inflammatory processes that began with the initial infection with Chlamydia trachomatis. Moreover, there is a clear correlation between the number of times an individual has been infected and the likelihood of scar tissue formation. In this study by Ramadhani and colleagues, an important question is addressed that has received less attention in the literature: what is the effect of azithromycin treatment on the immune response? Answering this question would help aid our understanding of what drives trachoma progression, and to what degree the bacteria play a role. The authors examined changes in infection, clinical signs of trachoma and host immune responses in the same individuals before and after azithromycin treatment. They found that infection rates dropped after azithromycin treatment, but rose again after 6 months, suggesting re-emergence of infection. Clinical disease was strongly correlated with active infection. In line with other studies, proinflammatory cytokine and chemokine production was associated with active infection, while Th17 associated cytokines and matrix metalloproteinases (involved in tissue destruction) were associated with inflammation that persists after infection is cleared. However, this study could not tease apart immune factors associated with infection versus immune factors associated with inflammatory persistence after infection clearance because the variables were inextricably linked. A more informative study would have removed infection from the equation and examined the effect of azithromycin treatments on individuals with signs of trachoma but without infection. Hopefully, this study will stimulate interest for future studies that will more directly address the role of azithromycin in immune modulation of the trachomatous ocular surface.
There are three biovars of Chlamydia trachomatis: the ocular strains which cause trachoma, the urogenital strains responsible for upper urogenital tract disease, and the lymphogranuloma strains. Both the ocular and urogenital strains can infect the ocular surface and both cause follicle formation on the conjunctiva. The difference between the two is the urogenital strains cause an acute conjunctivitis while only the ocular strains cause the blinding infectious disease called trachoma. For this reason, there are not many epidemiological studies of acute conjunctivitis caused by urogenital strains of Chlamydia trachomatis. In this interesting study, 245 conjunctival specimens were screened for Chlamydia trachomatis, and all of the Chlamydia strains belonged to the urogenital biovar (strains D-K). Given that all of the samples were from Budapest, it is expected that no trachoma strains of Chlamydia trachomatis were found in the data set. If the samples had been taken from a trachoma endemic region such as Ethiopia, for example, it is likely that trachoma strains would have been prevalent. In this study, PCR-RFLP genotyping was used to identify the strains. C. trachomatis was found in 53 cases of adult conjunctivitis. This high level of C. trachomatis in conjunctivitis samples raises an important public health concern for health care providers, and is an important reminder that, even though it is considered a benign conjunctivitis as compared to trachoma, urogenital strains of Chlamydia trachomatis do contribute significantly to cases of conjunctivitis in sexually active populations. As the authors point out, it may be that conjunctivitis is the only sign or symptom of disease in some of these patients, and may be an indicator of the STI.
This paper by DeWitt and colleagues will have broad appeal to anyone studying immunology, immuno-oncology, infectious disease and autoimmunity. Several biotech companies now offer T-cell receptor (TCR) sequencing assays and one of the biggest goals of TCR sequencing is to utilize a better understanding of the immune repertoire for the identification of unknown antigens, to create a library of public TCRs and to ultimately associate TCR responses with disease states for predictive analytics and T cell therapy development. In this study, a strong host cell HLA dependency is identified as a key mediator of public TCR profiles, defined as TCRs that are identical in more than one individual. The authors identify 80 million unique TCR beta chain sequences in 666 individuals. Of those, 11 million are found in more than one person. This shows a huge range of TCR sequences within an individual and a population, as expected, given the very high number of possible sequences arising from V(D)J recombination events. Intriguingly, clustering public TCRs against HLA sequence results revealed a huge drop in the number of groups of interest. Specifically, the majority of the TCR clusters were strongly associated with at least one HLA allele and the number of HLA-restricted TCR clusters encompassed very specific disease-associated groups such as EBV-epitope responsive TCRs, for example. Together, the data presented supports the hypothesis and data in Emerson et al. (2017) that TCR beta chain occurrences within a population are strongly associated with specific HLA alleles. This serves as further evidence that MHC proteins (which are encoded for by HLA alleles) control TCR binding and reactivity to a specific antigen. T cell selection and proliferation as a result of antigen presentation via MHC binding, appears to be highly driven by the unique MHC sequence. In fact, the data suggests HLA alleles may be more predictive of immune repertoire and immune response than antigen prevalence, at least in some cases, although this hypothesis remains to be tested in the context of different disease states.
Chlamydial virulence factors are known to modulate Rab proteins directly or indirectly, and this paper provides another example of this overarching theme. Rab proteins regulate intracellular trafficking and since Chlamydia trachomatis is an obligate intracellular bacterium, its ability to modulate the vesicular maturation pathways by interacting with Rab proteins allows the bacterium to survive within a modified host endosome (phagocytic vacuole) called the inclusion. Faris and colleagues present a compelling case for the underlying mechanistic basis for this interaction in this Cell Reports article. The Chlamydial virulence factor CT229 was identified as an Inc protein in earlier screens, meaning that it is a transmembrane protein found in the Chlamydial inclusion membrane and therefore thought to play a role in bacterial virulence and modification of the host cell. CT229 was thought to be a Rab protein modifier because of its ability to bind Rab proteins. In this study, CT229 binding to ?? Rab proteins is validated and the authors demonstrate CT229 dependent recruitment of these Rabs to the Chlamydial inclusion by fluorescence microscopy of over-expressing fluorescently tagged Rab proteins. There are more than 60 human Rab proteins and the authors show that CT229 does not display universal Rab protein binding. Since Rab proteins must be GTP bound to be active and membrane bound, the authors use GTP- and GDP- bound Rab mutant mimics to validate the inclusion colocalization data with well-designed co-immunoprecipitation assays. However, the authors only compare GTP bound (activated) versus GDP-bound mimic mutant Rabs 1, 4 and 35. Therefore, the reader is less convinced of the role of CT229 in binding and recruiting other Rabs to the Chlamydial inclusion. The authors go on to show that Rab 4 and 35 are necessary for chlamydial growth and replication and each have at least one effector protein that is recruited to the inclusion in a CT229-dependent manner. Expression of full length CT229 and mutants in yeast revealed that the inherent toxicity of CT229 in yeast is dependent upon SNARE like domain of CT229. The SNARE like domain of CT229 was further demonstrated to play a role in Chlamydial replication rates and recruitment of Rabs 1, 4 and 35 to the Chlamydial inclusion within HeLa cells. A yeast suppressor screen suggested CT229 toxicity could be rescued by clathrin over-expression and suggested the involvement of a number of clathrin-dependent vesicle trafficking proteins in rescue of CT229 toxicity in yeast. This led to the hypothesis that CT229 disturbs normal clathrin-dependent vesicle transport pathways. Transferrin and M6PR are both known to be involved in Chlamydial toxicity as demonstrated in previous studies, and both are transported within clathrin-dependent vesicles. The authors therefore test the hypothesis that Rab 4 and 35 are involved in transporting transferring to the inclusion in HeLa cells, and demonstrate that this is the case. In summary, this work demonstrates that during the course of infection, the obligate intracellular bacterium Chlamydia trachomatis utilizes the CT229 virulence factor to interfere with the normal clathrin-dependent vesicle trafficking pathway and redirect these vesicles to the intracellular chlamydial inclusion. The result of this interference is the ability of CT229 to provide Chlamydia living inside the inclusion with more nutrients to facilitate bacterial growth and division within the inclusion.
Article