Who am I? (re-)Defining fibroblast identity and immunological function in the age of bioinformatics.

Abstract

Fibroblasts are the cells that hold us together, famed for their resilience and plasticity. However, the same capabilities that render these cells easy to work with also make them difficult to understand. In particular, the definitions we currently rely on as fibroblast biologists are imperfect and unsatisfying, including defining bona fide markers, gating and exclusion criteria, surface phenotypes shared with other cells, heterogeneity, as well as methods for extraction from tissues and genetic in vivo targeting. This, in turn, has made it difficult to study subset function and anatomy, which creates important knowledge gaps that hinder precise functional elucidation and translational efforts. In short, for too long, these cells have been anything but well-defined. All that is changing. Thanks to single-cell transcriptomic and in situ imaging technologies, innovative mouse models, and in-depth human explorations, the field is now entering a renaissance of understanding regarding fibroblast ontogeny, identity, and function. Nowhere is that clearer than in fibroblast immunology, where an explosion of information now places us within reach of a consensus nomenclature and organizing principles for the fibroblast lineage, and clinically precise targeting of fibroblasts in patients to achieve health-restoring or disease-preventing immune modulation. Long split into subject area siloes, the world’s foremost fibroblast immunology groups are now coming together as never before, pooling data, ideas, and resources to reach new levels of comprehension regarding shared networks, pathways, and targetable mechanisms of action. The impact of this renewed understanding of fibroblast function and heterogeneity is hard to overstate. It is being felt across diverse fields including immune system developmental biology, tissue biology and in inflammatory diseases such as inflammatory bowel disease, arthritis, fibrosis, and cancer. It is an exciting time to be a fibroblast immunologist. This volume of Immunological Reviews brings together a multi-dimensional body of work from authors with diverse interests, ranging from pre-clinical to translational, and from health to acute and chronic disease. We first focus on the role of fibroblasts in immunological homeostasis and tissue development, covering the effects of fibroblasts on B cell, T cell, and macrophage biology, before moving to immunological interactions in wound healing, chronic inflammatory disorders, COVID19, fibrosis and cancer. Fibroblasts are involved in a spectrum of outcomes, highlighting the importance of understanding the cues that drive these cells to respond and then either resolve or participate in chronic inflammatory processes that drive pathogenesis of so many intractable human diseases. Fibroblastic reticular cells (FRCs) of secondary lymphoid organs provide an in-depth, heavily explored example of fibroblasts directing and regulating immune homeostasis and responses. FRCs represents immunologically active fibroblasts from lymph nodes, spleen, and mucosal- or fat-associated lymphoid tissues, which direct adaptive and innate immune cells through the secretion of chemokines, cytokines, growth factors, and extracellular matrix components, weaving an array of architecturally distinct micro-niches. They provide the physical scaffolding required for leukocytes to migrate upon, as well as the chemical signals that keep them happy and located in the right zones, poised for immunological action. In this issue, Dr. Ludewig and colleagues1 and Dr. Purton and colleagues2 highlight the extraordinary advances made possible in this field through the strategic use of gene-targeting and lineage-tracing mouse models, and more recently by single-cell transcriptomics. An array of Cre-reporter mice and mice with inducible expression of Cre have, in particular, shone a light into the development and function of fibroblasts in the T zone, the B-cell follicles, marginal zone reticular cells, FDCs, pericytes, and fibroblasts at T-B zone borders, as well as uncovering their precise roles in B cell development in bone marrow and spleen. The development and application of such precise in vivo tools is a primary aim for almost every subsequent tissue and disease indication described in this issue. Single-cell work is currently helping to lay the groundwork that will enable these advances, which is a recurrent theme raised by the authors in this issue. Here, Ludewig et al also provide a thorough overview of considerations for single-cell dissection of fibroblast populations, to ensure analysis is robust and meaningful. Dr. Molofsky and colleagues3 provide a compelling overview of an under-appreciated fibroblast-immune interaction, describing the importance of perivascular cells in facilitating immunological interactions in lung, liver, adipose, central nervous system, and meninges. An emerging literature reveals that the adventitial niche, in which perivascular fibroblasts reside, is a vibrant immunological site containing macrophages, dendritic cells, mast cells, T cells (resident and trafficking), innate lymphoid cells, and lymphatics, with roles in steady-state immune function as well as the development of tertiary lymphoid tissues during inflammation. Another important stromal cell in this niche, the pericyte, stabilizes microvasculature and promotes leukocyte trafficking and maintenance through the production of well-known factors such as ICAM-1 and macrophage inhibitory factor. Bone marrow, splenic and thymic fibroblasts involved in B- and T-cell development are then brought into focus by Dr. Purton and colleagues,2 and Dr. Nitta and colleagues.4 A common theme is again the transformative nature of single-cell transcriptomics studies for our understanding of heterogeneity and division of labor among fibroblast subsets. In both fields, there is a clear need for innovative human studies to bring the relevance of mouse findings into focus. The thymic biology field has long emphasized the importance of epithelial cells in thymocyte development, but recent findings have uncovered a role for medullary fibroblasts in self-antigen expression driving immunological tolerance. A valuable resource paper in its own right, Nitta et al. also provide novel single-cell and pathway analyses as part of this review, together with clarifying markers for isolating various thymic fibroblast subsets by flow cytometry. The next section of this issue comprehensively discusses the potential relationship between macrophages and fibroblasts, which may imprint upon each other, yielding a range of medically important outcomes and opportunities for therapeutic intervention. With fascinating simplicity, two-step circuit-like interactions between fibroblasts and macrophages have come to prominence relatively recently, through a body of work here described by leading scientist in this field, Dr. Franklin.5 In vitro work defining organizational circuits between macrophages and fibroblasts is finding emerging support in vivo, notably in liver where Kupffer cell depletion drives upregulation of CSF1 in stellate cells, and in lymph nodes, where marginal reticular cells provide RANK signals to lymphatic endothelial cells, allowing correct localization of subcapsular sinus macrophages. These relationships, and others perhaps yet to be discovered, may have far-reaching effects on immune surveillance, infection clearance, and wound healing. Drs. Bellomo, Bajenoff and colleagues closely explore this process in the spleen, describing spatially separated macrophage populations that are specialized for important functions.6 These include the uptake of aged red blood cells and hemoglobin, the control of systemic viral and bacterial pathogens, and the clearance of apoptotic cells, known as efferocytosis, which is required to prevent autoimmunity. Different subsets of splenic fibroblasts control these processes by the provision of chemokines, cytokines, and growth and survival factors as well as physical interaction and support. As one example, in splenic spleen red pulp, cord fibroblasts expressing the WT1 transcription factor are intimately involved in maintaining red pulp macrophages responsible for uptake of aged red blood cells and hemoglobin. As another example, red pulp, but not white pulp macrophages are maintained through fibroblast-driven production of master regulator CSF1.6 Many questions remain regarding the precise identities and functions of fibroblast subsets within the spleen and other lymphoid tissues and their roles in maintaining macrophage subsets. Tissue repair tasks are some of the most important roles fibroblasts play. Their utility in fixing the countless insults inflicted upon the human body must be balanced against undesirable, and poorly treated outcomes such as tissue damage, fibrosis, and scarring. Tissue repair is a tightly regulated, robust and multi-faceted process involving hemostasis and inflammation, the proliferation of cells that produce connective tissue, ECM remodeling, and maturation and persistence of neo-stromal components. Using skin as a prototypic fibroblastic niche, Dr. Rinkevich and colleagues describe four key mechanisms for fibroblast-immune interactions in this process, including cytokine secretion, direct priming, ECM remodeling, and, fascinatingly, the mobilization of ECM niches to steer embedded immune cells toward wounds.7 Fibroblasts are central to maintaining and replenishing skin immunological niches and do so in a site- and subset-dependent manner, where distinct fibroblast subsets produce different chemoattractants, known to be involved in recruitment of specialized macrophage progenitors from circulation. The authors report that certain fibroblast subsets in skin are more prominently involved, particularly fascia fibroblasts. They describe how fibroblasts awaken in response to injury and begin secreting and remodeling ECM to stabilize and repair the tissue damage and express soluble and transmembrane factors that influence leukocyte behavior at the lesion. Well-known immunological pairings such as ICAM-1-ITB2 and CD40-CD40L are central to these processes. The mobilization of immunologically active tissue is a newly described finding where fibroblasts undergo collective migration to heal deep wounds, bringing along their associated mesh-like ECM and the immunological ecosystem residing within it, relocating leukocytes to the areas of the wound where they are required to participate in healing. Importantly, when the normal resolution of wound healing is incomplete, these processes become fundamentally dysregulated, yielding chronic inflammation characteristic of such diseases as scleroderma, keloid scarring, and Dupuytren’s disease, which feature pathological fibrosis, inflammation, and contraction.7 In an engaging review, Dr. Hinz and colleagues argue that fibroblast activation is a key commonality among a diversity of fibroblasts that have been identified through lineage tracing and single-cell transcriptomics.8 However, it is a process still poorly understood, despite the clear utility such understanding would bring to the design of new strategies to treat fibroblast-driven diseases. In this review, the macrophage-fibroblast interaction again comes into the spotlight as a type of immunological synapse that controls key decision points in healthy healing versus pathogenic tissue repair cascades.8 Like fibroblasts, macrophages show extraordinary plasticity in phenotype and function, and it is now well-known that they exist along a complex continuum that ranges from pro- to anti-inflammatory. Similarly, their capacity to engage in cross-talk with fibroblasts under inflammatory and damaged conditions yields diverse sequelae, from resolution-focused to pro-fibrotic. Growth factors and cytokines secreted from macrophages may control fibroblast maintenance, proliferation, activation, and ECM production. Conversely, fibroblasts may direct macrophage differentiation through changes in microenvironmental architecture, producing mechanical cues together with well-known growth factors such as M-CSF and GM-CSF. The authors go on to describe how chronic inflammation hijacks these processes in osteoarthritis and rheumatoid arthritis, yielding pain, joint damage, changes to ECM remodeling, and fibrosis.8 Rheumatologists have long reported an abnormal fibroblast expansion which correlated with inflammation severity in Rheumatoid Arthritis patients. Synovial fibroblasts actively drive tissue pathology by producing cytokines and chemokines that recruit leukocytes to the joint, driving a positive feedback loop that leads to pathological levels of inflammation and joint destruction. In this issue, Dr. Croft and colleagues describe the transformative nature of recent single-cell studies in not just proving the heterogeneity of fibroblast subsets in the joint but providing the necessary information required to infer separate functions, and providing inspiration for future targeting of the most pathogenic subsets while sparing those that are benign.9 This work is now driving the development of a new generation of anti-fibroblast therapeutics for rheumatologic disease. Similarly focused on rheumatic disease, in this case, Sjogren’s syndrome, Dr. Barone and colleagues provide an overview of the role of chronic inflammation in the development of tertiary lymphoid structures (TLS).10 These are clusters of T and B lymphocytes brought together in inflamed non-lymphoid tissues through the priming, proliferation, and maturation of local fibroblasts in response to defined cytokine cues, again highlighting their untapped potential as therapeutic targets in chronic inflammatory disease, supported by work in models of Sjogren’s syndrome, type-1 diabetes, experimental autoimmune encephalomyelitis, and influenza infection. The existence of a TLS-supporting fibroblast precursor is a tempting hypothesis that could prime certain tissues to support TLS-development. Defining key links between TLS formation and fibrosis, Dr Yanagita and colleagues take us to the kidney, where fibroblast-driven ectopic lymphoid tissue formation associates closely with renal damage during the stepwise progression of chronic kidney disease (CKD).11 Both aging and the development of CKD are marked by phenotypic changes to resident fibroblast populations. Contextual cues driving the acquisition of heterogeneous phenotypes, and subsequent spectrum of outcomes from regeneration to fibrosis is essential to understand in order to target pathogenic subtypes. The emerging single-cell landscape in this field is finally allowing the precise elucidation of differences between myofibroblasts, fibroblasts and pericytes in fibrotic kidneys, as well as the pathways that mark cell state transitions and predict acquisition of fibrotic potential. Cells that are transitioning to myofibroblasts first acquire a quiescent non-cycling phenotype before expressing key molecular drivers of renal fibrosis, such as TGFb.11 Findings such as these are finally yielding clear pictures of the precise cellular targets likely to yield clinical impact. Fibrosis can develop in many different organs when the wound healing response goes awry and connective tissue forms scars rather than returning to its healthy state. Fibrosis of the lung severely compromises airway integrity and respiratory function. Upon damage to the respiratory epithelium, airway epithelial cells, infiltrating immune cells and endothelial cells release stimuli that trigger the fibrogenic cascade. Fibroblasts respond to such factors by proliferating, migrating to the injured site, and differentiating into matrix secreting and contractile myofibroblasts. In addition to their well-established role in ECM deposition, myofibroblasts produce cytokines and chemokines that contribute to the wound healing response. In most healthy individuals, these activities are essential to repairing a tissue injury and closing the wound. Once the wound healing process is complete, the fibrotic reaction resolves and the tissue reverts to homeostasis. However, in pulmonary fibrosis, which can arise in many different respiratory challenges and interstitial lung diseases including viral infection and acute respiratory distress syndrome, the wound healing process ceases to terminate. As such, activated myofibroblasts are not replenished by healthy tissue fibroblasts and connective tissue does not return to a quiescent state. Rather, revolutions of the myofibroblasts and their fibrogenic activities persist, resulting in excessive deposition of collagen in lung parenchyma and progressive stiffening of the lung tissue. In a feed-forward manner, fibroblasts are activated by biophysical forces associated with more rigid tissue. Mechano-sensing drives revolutions of the myofibroblast cycle leading to deposition of additional matrix and further cross-linking and contraction. Collectively these events result in “overhealing” and development of interstitial scars. In organs like the lung, scarring leads to severe pulmonary fibrosis. Ultimately pulmonary function is compromised due to the increasingly hardened lung tissue. Few if any efficacious treatment options are available and patients often succumb to respiratory failure. Dr. Tatler and colleagues have made the heroic effort, amidst a fast-moving pandemic, to review our current understanding of COVID19 disease and the accompanying fibrotic changes of the lung that can be progressive and life-threatening.12 As observed in post-mortem tissue samples, SARS-CoV2 induces a number of airway alterations including damage to the epithelial barrier, recruitment of inflammatory cells and alterations to endothelial cells and vascular tone, and occlusion of the micro-circulatory system which collectively precipitate pulmonary fibrosis. Similar hallmarks of fibrotic disease have been observed in chest scans of COVID-19 patients. While research regarding the etiology of COVID-19 and related sequelae is rapidly evolving, the authors discuss factors that have been implicated in the initiation, severity, and persistence of fibrosis in COVID-19 disease including age, genetics, obesity, and metabolism.12 Fibrosis of the intestine, a common occurrence in inflammatory bowel disease and Crohn’s disease, in particular, can severely impede gastrointestinal health by causing strictures at different points along the small and large intestine and ultimately obstructing the bowel. As in pulmonary fibrosis, intestinal fibrosis is driven by excessive and persistent matrix deposition and stiffening by myofibroblasts. The precipitating events and cellular mechanisms that lead to intestinal fibrosis may differ somewhat from lung fibrosis; however, the exact etiology remains to be fully elucidated. Here, Dr. Rieder and colleagues propose how intestinal injury and microbial factors trigger the release of inflammatory and pro-fibrogenic factors that act on fibroblasts and adipocyte progenitors to ultimately drive the wound healing response.13 The roles of mesenteric adipocytes associated with creeping fat, mesenchymal subtypes from different sub-anatomical compartments of the intestine including fibroblasts, myofibroblasts, and smooth muscle cells, and various immune cells and inflammatory mediators in the development of debilitating fibrostenotic structuring in Crohn’s disease and other ileocolic diseases are discussed. In addition to a comprehensive discussion of the molecular mechanisms underlying intestinal fibrosis, the authors offer optimistic perspectives on development of clinical biomarkers and endpoints for future trials of anti-fibrotic therapeutics in inflammatory bowel diseases.13 Tumors can grow in almost any tissue and sub-anatomical location, often originating from a single epithelial cell once it incurs oncogenic mutations. Tumors divide rapidly and elaborate new factors, supporting their own growth via autocrine effects and conditioning the microenvironment to support their growth through paracrine circuits. Fibroblasts residing in proximity to tumors are bathed in and respond to tumor-derived factors, many adopting phenotypes and functions that differ from their healthy tissue counterparts. Programming of fibroblasts by tumor cells, which outnumber fibroblasts by orders of magnitude in large advanced tumors, brings about marked changes in this cell compartment that can directly support tumor cell growth or indirectly promote tumor growth by affecting immune and vascular cells. These distinguishing features have led to the identification and classification of multiple subtypes of “cancer-associated fibroblasts.” With the resurgence in emphasis on translational science, rapid technological innovation, and promise of modern cancer therapies, the holistic cellular landscapes of human and mouse tumors are being described at a remarkable tempo. Modern genomics, imaging, genetically engineered mouse models, and single-cell technologies have led to an explosion of data that is pushing our understanding of stromal cell interactions with immune cells in healthy tissues and many different inflammatory diseases ranging from chronic infection, cancer, and fibrosis far beyond the boundaries of previous decades. The field is frothy at the moment, with handfuls of scRNAseq studies, each describing multiple fibroblast subtypes with distinct features and markers and inferred functions and ontogenies for each cancer indication, stage, lesion type, and treatment. The necessary next step will be to assimilate the data, creating a holistic overview of the commonalities and differences in fibroblasts found across a range of tumor types. Several reviews on this issue seek to advance our understanding accordingly in this direction. Dr. Wang and colleagues describe sub-compartments of fibroblasts present in the tumor microenvironment, with three major subtypes of cancer-associated fibroblasts and two subtypes of healthy or normal tissue fibroblasts.14 Among the three subtypes of cancer-associated fibroblasts which exhibit “activated” phenotypes, two are defined as immunosuppressive and the other as pro-metastatic. They further explore the roles of cancer-associated fibroblasts in immune inflamed or hot tumors and excluded tumors in which CD8 T cells and other lymphocytes are sequestered in the stroma-rich region surrounding the tumor. In inflamed tumors, they surmise that cancer-associated fibroblasts recruit but also suppress the functions of CD8 T cells and dendritic cells via direct and indirect mechanisms. In immune-excluded tumors, they describe evidence that TGFb-activated myofibroblastic, cancer-associated fibroblasts impede anti-tumor immunity by impairing CD8 T cell cytotoxicity and migration, again through direct and indirect mechanisms.14 Dr. Mechta-Grigoriou and colleagues provide a balanced summary of the literature by describing the dichotomous effects of cancer-associated fibroblasts in promoting and suppressing tumor growth.15 On the one hand, cancer-associated fibroblasts can promote tumor growth via effects on the vasculature, extracellular matrix, tumor cell proliferation and invasiveness, tumor cell sensitivity to chemotherapeutics, and different immune cell types including CD8 T cells, CD4 T regulatory cells, and dendritic cells. On the other hand, contrasting evidence is also discussed regarding the role of cancer-associated fibroblasts in restraining the self-renewal and metastasis of tumor cells.15 The articles by Drs. Mechta-Grigoriou and Wang describe a variety of pre-clinical and clinical therapeutic approaches to directly modulate the pro-tumor effects of the cancer-associated fibroblast compartment.14, 15 In follicular lymphoma, a non-Hodgkins lymphoma, malignant B cells accumulate in lymph nodes, bone marrow, and other tissues. In these environments, the tumor cells are surrounded by immune cells as well as stromal elements that are reprogrammed by tumor-derived factors. Dr. Tarte and colleagues compare and contrast the cross-talk between B cells and fibroblastic stromal cells in lymphoid tissues of healthy subjects and in patients with follicular lymphoma.16 Mutations that lead to B-cell malignancy in follicular lymphoma upregulate factors that activate fibroblastic stromal cells and drive their differentiation into cancer-associated fibroblasts. Follicular lymphoma cancer-associated fibroblasts recruit and activate TAMs and T-follicular helper cells to B-cell niches within lymphoid tissues, which in turn, promote the survival of the lymphoma cells. Immunofluorescence images of follicular lymphoma depict dramatic remodeling of lymph node architecture including expanded B-cell follicles enswathed by dense collagen.16 Dr. Swarbrick and colleagues describe with textual and pictorial eloquence the applications and transformative impact of single-cell RNA sequencing on our understanding of stromal cell composition in human tumors and their extensive interactions with immune cells.17 The authors discuss the transcriptional, genomic, phenotypic, and functional hallmarks of three major subtypes of cancer-associated fibroblasts and provide a compelling roadmap for where the field is headed to elucidate the ontogeny and functional significance of fibroblasts, pericytes, and other stromal cells in different tumor types, locations, and therapeutic settings.17 Dr. Muller and colleagues use low- and high-powered lenses to bring relationships among fibroblast subtypes into focus and ambitiously attempt to superimpose fibroblasts from vastly different tissues and seemingly disparate indications into the same field of view.18 The authors discuss single fibroblast genomics studies across a variety of tissues and diseases to begin proposing consensus subtypes, highlight useful markers, identify anatomical and sub-anatomical hallmarks of the lineage, and explore stromal evolution and function in inflammatory diseases. Synthesizing recent discoveries and emerging theories from a rapidly growing field yielded a universal framework linking healthy tissue fibroblasts with their activated progeny as they emerge in perturbed tissues. Therapeutic approaches to modulate the immunological and pro-tumor functions of fibroblasts in cancer are also explored in this review. As Dr Hinz and colleagues state: Somebody has to do the work.8 Fibroblasts certainly do their share. Throughout the body, they indisputably play fundamental and long-unrecognized roles in immunological homeostasis and disease. As they yield their secrets after decades of careful study, it is now up to us to take on some of this work, to precisely elucidate fibroblast functions in health and disease, to bolster their actions where they help and inhibit them where they do not. In this issue of Immunological Reviews, we aimed to bring recent advances with relevance for human biology into prominence while identifying new routes of interest to facilitate future advances. We are in awe at the scope of work described in this issue, grateful for the time and effort of all authors, and excited to see the now-inevitable emergence of fibroblast-targeting therapeutics.