Epidermal cells help coordinate leukocyte migration during inflammation through fatty acid-fuelled matrix metalloproteinase production

Hall, Christopher J.; Boyle, Rachel H.; Sun, Xueying; Wicker, Sophie M.; Misa, June P; Krissansen, Geoffrey W.; Print, Cristin G.; Crosier, Kathryn E.; Crosier, Philip S.

doi:10.1038/ncomms4880

Article
Published: 23 May 2014

Epidermal cells help coordinate leukocyte migration during inflammation through fatty acid-fuelled matrix metalloproteinase production

Christopher J. Hall¹,
Rachel H. Boyle¹,
Xueying Sun¹,
Sophie M. Wicker¹,
June P Misa¹,
Geoffrey W. Krissansen¹,
Cristin G. Print¹,
Kathryn E. Crosier¹ &
…
Philip S. Crosier¹

Nature Communications volume 5, Article number: 3880 (2014) Cite this article

4936 Accesses
42 Citations
1 Altmetric
Metrics details

Subjects

Abstract

In addition to satisfying the metabolic demands of cells, mitochondrial metabolism helps regulate immune cell function. To date, such cell-intrinsic metabolic-immunologic cross-talk has only been described operating in cells of the immune system. Here we show that epidermal cells utilize fatty acid β-oxidation to fuel their contribution to the immune response during cutaneous inflammation. By live imaging metabolic and immunological processes within intact zebrafish embryos during cutaneous inflammation, we uncover a mechanism where elevated β-oxidation-fuelled mitochondria-derived reactive oxygen species within epidermal cells helps guide matrix metalloproteinase-driven leukocyte recruitment. This mechanism requires the activity of a zebrafish homologue of the mammalian mitochondrial enzyme, Immunoresponsive gene 1. This study describes the first example of metabolic reprogramming operating within a non-immune cell type to help control its contribution to the immune response. Targeting of this metabolic–immunologic interface within keratinocytes may prove useful in treating inflammatory dermatoses.

You have full access to this article via your institution.

Download PDF

Cell death as an architect of adult skin stem cell niches

Article Open access 22 April 2024

Sensory neuron-derived TAFA4 promotes macrophage tissue repair functions

Article 19 May 2021

Cellular and molecular mechanisms of skin wound healing

Article 25 March 2024

Introduction

Investigating how metabolism shapes the immune response is gaining significant attention due to the emerging connection between metabolic and inflammatory processes. Fuelled by the obesity epidemic, this new field of immunometabolism is underpinned by observations that inflammation is a hallmark of many chronic metabolic conditions^1,2. In addition to the activation of immune cells within metabolic tissues, functional heterogeneity of immune cells can be regulated, on many levels, by cell-intrinsic metabolic reprogramming^3,4,5. Whether non-immune cells also contribute to the immune response through coupling metabolism and immune function is, at present, unclear.

Mitochondria are emerging as key organizers of innate immune cell function⁶. One mechanism through which mitochondria can influence immune cell activity is through elevated production of mitochondrial reactive oxygen species (mROS). Although a major source of cellular ROS, mROS have traditionally been regarded as insignificant by-products of oxidative phosphorylation (OXPHOS). More recent studies have identified inflammation-responsive mechanisms that elevate mROS and shown that mROS can have direct antibacterial activity and influence the expression of immune-response genes^6,7,8. Whether non-immune cells also couple metabolic and inflammatory processes during their contribution to the immune response is unclear.

In addition to providing a physical barrier protecting the host from its environment, the skin is a critical component of the immune system⁹. The immune response during cutaneous inflammation relies on finely tuned communication between immune cell subsets and non-immune cells such as epidermal keratinocytes¹⁰. Dysregulation of this communication can lead to inadequate or persistent inflammation that characterizes many inflammatory dermatoses¹¹. In response to infection or injury keratinocytes produce matrix metalloproteinases (MMPs), a family of enzymes that collectively degrade all components of the extracellular matrix (ECM)^12,13. The recruitment of leukocytes from the bloodstream and their migration through the ECM to sites of inflammation is, in part, dependent upon MMP activity. It has been suggested that keratinocyte-derived MMP9 production facilitates migration of leukocytes into the inflamed epidermis during cutaneous inflammation¹³. Both the expression and activity of MMPs has been shown to be directly regulated by mROS¹⁴. Furthermore, tumour necrosis factor alpha (TNF-α)-driven MMP9 expression within human keratinocytes is ROS-dependent¹⁵. Collectively, these data suggest mROS production within keratinocytes may promote leukocyte migration through regulation of MMP activity. These results also imply a possible connection between the metabolic process of OXPHOS within epidermal cells and leukocyte recruitment.

The zebrafish is a well-established model to investigate innate immune cell function¹⁶. By 2 days post fertilization (dpf) zebrafish larvae are populated with neutrophil and macrophage lineages. Live imaging these innate immune cells within transgenic reporter lines has revealed remarkable conservation of function with their mammalian equivalents^17,18,19,20. Live imaging analyses have also given unique insights into leukocyte behaviour during wound-associated cutaneous inflammation^21,22,23,24. In contrast to the mammalian epidermis that consists of proliferating and differentiated keratinocytes within several strata, the epidermis of embryonic/larval zebrafish is relatively simple consisting of an outermost enveloping and inner epidermal basal layer^25,26. Despite this simple architecture, functional similarities exist between the zebrafish and mammalian epidermis, supporting the use of zebrafish as a model of skin diseases^26,27,28.

Exploiting the live imaging potential of the zebrafish system, previous work in our group has revealed activated macrophages utilize β-oxidation of fatty acids to fuel OXPHOS-coupled mROS that contributes to macrophage effector function⁷. This metabolic reprogramming of macrophages requires a mitochondrial enzyme encoded by immunoresponsive gene 1 (irg1), a vertebrate homologue of a bacterial enzyme necessary for fatty acid metabolism²⁹. Zebrafish possess 2 IRG1 homologues, irg1 and irg1-like (irg1l), most likely the result of a duplication event. A recent study has suggested a function for Irg1l during cutaneous inflammation³⁰. We predicted that Irg1l may couple metabolic and immunological processes, similar to its paralog Irg1. Using the zebrafish system to explore the function of Irg1l, we uncover a new Irg1l-dependent signalling pathway that connects metabolic and inflammatory processes within epidermal cells during inflammation. We demonstrate expression of irg1l within epidermal cells in response to live bacterial infection and wound-induced inflammation, where it contributes to leukocyte recruitment. In response to bacterial infection, epidermal cells enhance their production of mROS in a process dependent on Irg1l and fuelled by fatty acid β-oxidation. We further show that Irg1l and β-oxidation are necessary for infection- and wound-responsive mmp9 expression within epidermal cells that demonstrate elevated mitochondrial uptake of fatty acids. Finally, we demonstrate elevated abundance of IRG1 within mitochondria of keratinocytes in a mouse model of cutanous inflammation (atopic dermatitis) supporting the translation of our zebrafish findings to a mammalian setting. We propose a new mechanism where Irg1l-dependent β-oxidation-fuelled mROS production within epidermal cells contributes to MMP-9-driven leukocyte migration during inflammation. For the first time, this study connects inflammation-driven changes in epidermal cell metabolism with their contribution to the immune response. This suggests that metabolic reprogramming, as a mechanism to help orchestrate the immune response, is not restricted to immune cells.

Results

irg1l is expressed by epidermal cells following infection

We first identified zebrafish Irg1l (a homologue of mammalian IRG1) as a gene upregulated in response to live Salmonella enterica serovar Typhimurium (hereafter referred to as Salmonella) infection using the GeneChip Zebrafish Genome Array (Affymetrix) (Fig. 1a), consistent with a previous study³⁰. Quantitative (q) PCR analysis of irg1l expression following Salmonella infection (~\n600 colony-forming units (cfu) injected into the hindbrain ventricle of 2 dpf larvae) revealed elevated expression from 1 h post injection (hpi), when compared with PBS-injected controls (Fig. 1b). Infection-responsive expression of irg1l peaked at 1 day post injection (dpi) before returning to ‘steady-state’ levels by 3 dpi (Fig. 1b). Throughout this study, unless otherwise stated, an infection dose of 600 cfu of Salmonella was injected into the hindbrain ventricle of 2 dpf larvae to stimulate an antibacterial immune response. When compared with PBS-injected controls, Salmonella-injected larvae (2 hpi) possessed infection-responsive irg1l expression within cells covering the infection site as well as within nasal and neuromast epithelial cells (Fig. 1c,d). Constitutive expression of irg1l was detected within epithelial cells of the oropharyngeal region and cloaca (Fig. 1c). Injection of E. coli lipopolysaccharide (LPS) also stimulated similar early irg1l expression (Supplementary Fig. 1). Of note, LPS-stimulated irg1l expression did not persist as long as that induced by live Salmonella injection (Supplementary Fig. 1). By 1 dpi, infection-responsive expression of irg1l became largely restricted to neuromasts of the developing lateral line system (Supplementary Fig. 2a,b). Some infected larvae also possessed irg1l expression within the intestine and pronephric tubules (Supplementary Fig. 2c,d). In light of the well-recognized contribution that the epidermis/skin provides to a functional immune system, we investigated the role of Irg1l within the epidermis. Expression analysis revealed that epithelial cells with infection-responsive irg1l expression, surrounding the infection site, were also positive for the epidermal marker keratin 4 (krt4), as detected at 4 hpi (Fig. 1e). To visualize epidermal irg1l expression with respect to injected bacteria, we utilized GFP-expressing Salmonella (hereafter referred to as Sal-GFP)³¹. Injected Sal-GFP were observed in close proximity to (but never within) overlying irg1l⁺ epidermal cells (Fig. 1f). Transcriptome profiling has previously demonstrated expression of irg1l to be elevated in response to both intravenous injection of, and immersion in, the natural fish pathogen Edwardsiella tarda³⁰. In contrast to this study, we were not able to detect epidermal expression of irg1l following immersion in live Salmonella or LPS, as detected by whole-mount in situ hybridization (WMISH) analysis (Supplementary Fig. 2e–h). This suggests, in our infection model, host-derived endogenous signals contribute to infection-responsive expression of irg1l within epidermal cells.

**Figure 1: Irg1l is expressed within *krt4*⁺ epidermal cells following *Salmonella* infection.**

These results demonstrate that irg1l is rapidly expressed within epidermal cells in response to bacterial infection.

Macrophages are required for epidermal irg1l expression

We have previously shown macrophages respond, and functionally contribute, to the immune response in our hindbrain infection model³¹. We next investigated whether host-derived signals necessary for epidermal irg1l expression were macrophage-dependent.

We took advantage of the ability of spleen focus-forming virus proviral integration oncogene (Spi1)- and granulocyte colony stimulating factor receptor (Gcsfr)-targeting morpholino oligonucleotides (MOs) to deplete larvae of both the macrophage and neutrophil lineages, to first assess the myeloid-lineage contribution to infection-responsive epidermal irg1l expression^31,32. Larvae depleted of both Spi1 and Gcsfr demonstrated a decrease in epidermal irg1l expression following infection (Fig. 2a). To differentiate whether this myeloid-lineage contribution to infection-responsive irg1l expression was at the level of the macrophage or neutrophil, we next evaluated irg1l expression within interferon regulatory factor 8 (Irf8)-depleted larvae following infection. The transcription factor Irf8 has been shown to regulate macrophage versus neutrophil lineage commitment such that Irf8-depleted larvae possess an expanded neutrophil compartment and depleted numbers of macrophages³³. Similar to larvae depleted of both Spi1 and Gcsfr, Irf8-depleted larvae demonstrated a decrease in epidermal irg1l expression following infection (Fig. 2a). These results suggest macrophages contribute to infection-responsive epidermal irg1l expression. To visualize macrophages in relation to injected bacteria and epidermal irg1l expression, we exploited the macrophage-lineage marking Tg(mpeg1:Gal4/UAS:Kaede) reporter line¹⁷. Analysis of irg1l expression within Tg(mpeg1:Gal4/UAS:Kaede) larvae following Sal-GFP injection showed macrophage expressed 1 (Mpeg1)⁺ macrophages and bacteria in close proximity to overlying irg1l⁺ epidermal cells, as assessed at 4 hpi (Fig. 2b). Live imaging infected Tg(mpeg1:mCherry)¹⁷ larvae demonstrated that these macrophages were highly phagocytic toward the injected Sal-GFP (Fig. 2c).

**Figure 2: Infection-responsive epidermal expression of *irg1l* is dependent upon macrophages and Myd88.**

A well-characterized mechanism through which innate immune cells detect bacteria is via the Toll-like receptor (TLR) class of pattern recognition receptors³⁴. Cell surface and intracellular TLRs detect specific pathogen-associated molecular patterns, stimulating signalling cascades that couple pathogen detection to downstream immune cell functions, including cytokine production. TLR signalling is largely dependent on the adaptor molecule myeloid differentiation primary response gene 88 (MYD88). We have previously shown that innate immune cells in zebrafish express TLR adaptor molecules, including myd88 (ref. 18). In addition, zebrafish MyD88 has been demonstrated to contribute to pathogen clearance following infection³⁵. To assess whether MyD88-dependent TLR signalling is required for irg1l expression within epidermal cells following infection, we analysed irg1l expression within infected MyD88-depleted larvae. Epidermal expression of irg1l was not detected within infected MyD88-depleted larvae (Fig. 2d). Analysis of myd88 expression within infected larvae revealed expression within putative macrophages in the midbrain/hindbrain region (Fig. 2e). Of note, expression was not detected within the overlying epidermis.

These results suggest infection-responsive expression of irg1l within the epidermis requires macrophage detection of bacteria through a MyD88-dependent mechanism.

GC signalling suppresses epidermal irg1l expression

Glucocorticoids (GCs) are a class of steroid hormone that largely suppress the immune response. Through binding the glucocorticoid receptor (GR), GCs upregulate anti-inflammatory, while repressing pro-inflammatory, gene expression. GCs have been shown to possess conserved anti-inflammatory activity in larval zebrafish (Supplementary Fig. 3a–c)³⁶. We next investigated whether the infection-responsive expression of irg1l within epidermal cells was sensitive to the anti-inflammatory activity of GC/GR signalling.

To assess the effect of GCs on epidermal irg1l expression we co-injected Salmonella with dexamethasone, hydrocortisone and prednisolone and analysed irg1l expression at 4 hpi. In contrast to Salmonella alone, epidermal irg1l expression was not detected in co-injected larvae (Fig. 3a), suggesting GCs suppress infection-responsive expression of irg1l within epidermal cells. To evaluate the requirement for GR signalling during GC-mediated irg1l suppression, we co-injected Salmonella with the GR antagonist RU-486 and infected larvae depleted of the zebrafish ortholog of nuclear receptor subfamily 3, group C, member 1 (NR3C1) (encoding the GR), then assessed irg1l expression. When compared with control MO-injected infected larvae, RU-486-treated and Nr3c1-depleted larvae demonstrated elevated infection-responsive epidermal irg1l expression (Fig. 3b). In zebrafish, endogenous GCs are produced by the inter-renal tissue, which functions as an adrenal gland-equivalent. Inter-renal tissue and its steroidogenic activity can be ablated through knockdown of nuclear receptor subfamily 5, group A, member 1a (Nr5a1a) (Supplementary Fig. 3d,e)³⁷. A similar increase in epidermal irg1l expression following infection was observed within larvae depleted of Nr5a1a (Supplementary Fig. 3f). Immunofluorescence detection of Nr3c1 within infected larvae revealed Nr3c1 protein localized to Mpeg1⁺ macrophages within, and irg1l-expressing epidermal cells overlying, the midbrain/hindbrain infection region (Fig. 3c,d). The greater sensitivity of this tyramide signal amplification-based fluorescence detection method revealed irg1l expression extended throughout the trunk within Nr3c1⁺ epidermal cells (Supplementary Fig. 4). Of note, Nr3c1 also colocalized with other irg1l-expressing epithelial tissues including the neuromasts, cloaca, nasal epithelium and the oropharngeal region (Supplementary Fig. 4).

**Figure 3: GC/GR signalling suppresses infection-responsive expression of *irg1l* in epidermal cells.**

Together, these results support a model where infection-responsive epidermal expression of irg1l is suppressed by the anti-inflammatory activity of GCs, most likely at the level of GR/Nr3c1⁺ macrophages and/or epidermal cells.

Irg1l contributes to bacterial clearance

To examine whether Irg1l contributes to the clearance of injected bacteria, we monitored bacterial persistence within Irg1l-depleted larvae infected with Sal-GFP. When compared with control MO-injected larvae, Irg1-depleted larvae demonstrated a significant increase in bacterial persistence, as assessed by live fluorescence microscopy (Fig. 4a,b and Supplementary Fig. 5) and recovered bacteria (Fig. 4c). This defective bacterial clearance resulted in decreased survival post-infection (Fig. 4d). These data support a functional role for Irg1l during clearance of injected bacteria.

**Figure 4: Irg1l-depleted larvae demonstrate diminished bacterial clearance and reduced survival following infection.**

Irg1l contributes to epidermal mmp9 expression

A feature of cutaneous inflammation associated with infection is the infiltration of innate immune cells, including monocytes/macrophages and neutrophils, into the dermis and epidermis. Central to the recruitment of leukocytes is the ECM-degrading activity of the MMP family of proteases^13,38. Similar to mammals, bacterial infection can elevate the expression of mmp9 within the zebrafish larval epidermis (Supplementary Fig. 6), where it contributes to macrophage recruitment³⁹. We next investigated whether infection-responsive expression of irg1l within the epidermis contributes towards mmp9 expression and the recruitment of macrophages to the infection site.

Fluorescent expression analysis confirmed irg1l and mmp9 expression colocalized within epidermal cells following infection (Fig. 5a,b). qPCR analysis of mmp9 expression within control MO-injected and Irg1l-depleted larvae at 1, 2 and 4 hpi revealed that Irg1l contributed to infection-responsive mmp9 expression (Fig. 5c). This result was supported when assessing mmp9 expression within infected control MO and Irg1l-depleted larvae (Fig. 5d,e). Irg1l-depleted Tg(mpeg1:EGFP) larvae also demonstrated reduced macrophage recruitment to the infection site, when compared with control MO-injected Tg(mpeg1:EGFP) larvae, despite maintaining unaltered (whole larvae) numbers of macrophages (Fig. 5f,g and Supplementary Fig. 7).

**Figure 5: Irg1l contributes to infection-responsive *mmp9* expression within epidermal cells and the recruitment of macrophages to the infection site.**

These results suggest that Irg1l contributes to the recruitment of macrophages to the infection site through regulating epidermal mmp9 expression.

Irg1l regulates epidermal mmp9 expression following wounding

In addition to infection, cutaneous inflammation associated with wound healing, allergy and/or trauma also requires the timely recruitment of specific leukocyte subsets. Employing a well-characterized tail transection model of acute wound-associated inflammation (Fig. 6a)^17,40, we examined whether irg1l was expressed in response to wounding and, if so, whether it helped coordinate the inflammatory phase of wound healing.

**Figure 6: Irg1l contributes to epidermal *mmp9* expression and the recruitment of neutrophils to wounded tissues.**

Similar to that induced by bacterial injection, irg1l was expressed within epidermal cells at the wound site by 1 hour post wounding (hpw) (Supplementary Fig. 8a,b). Expression analysis confirmed that irg1l-expressing epidermal cells also expressed mmp9 (Fig. 6b). Irg1l-depleted larvae demonstrated a reduction in wound-induced epidermal mmp9 expression, when compared with control larvae, suggesting Irg1l is required, at least in part, for wound-induced epidermal mmp9 expression (Fig. 6c,d). As ‘first responders’ during inflammation, neutrophils are the first leukocytes to arrive at wounded tissues. Live imaging of wounded transgenic zebrafish larvae possessing fluorescent neutrophils has provided unique insights into the migratory kinetics of these innate immune cells^23,24,41. By live imaging Tg(lyz:EGFP) larvae we assessed whether Irg1l-regulated mmp9 expression contributed to the recruitment of lysozyme (Lyz)⁺ neutrophils to wounded tissue. To first evaluate whether MMP9 contributes to neutrophil recruitment in our model, we assessed neutrophil recruitment to tail wounds within MMP9-depleted larvae and larvae treated with the pan-MMP inhibitor GM6001 (ref. 42). In contrast to control larvae, MMP9-depleted and GM6001-treated larvae possessed fewer neutrophils at the wound region (Fig. 6e,f and Supplementary Fig. 9), despite maintaining typical (whole-larva) numbers of neutrophils (Supplementary Fig. 10). A similar reduction was also observed within Irg1-depleted larvae (Fig. 6g), despite maintaining typical (whole-larvae) numbers of neutrophils (Supplementary Fig. 10). Tracking individual cell trajectories within wounded Tg(lyz:EGFP) larvae revealed that neutrophils migrated with significantly reduced velocities, displacement rates and meandering indices within Irg1-depleted larvae, when compared with control larvae (Fig. 6h–l). Of interest, when examining migration paths of neutrophils towards wounded tissue, most neutrophils migrated ventral to somitic tissue from the caudal hematopoietic tissue (a larval hematopoietic site). Those neutrophils that migrated more medially were observed to migrate superficially in close proximity to the irg1l/mmp9-expressing epidermis rather than through deeper somitic tissues (Fig. 6m,n).

Together, these data suggest that, in addition to infection-driven cutaneous inflammation, Irg1l also contributes to the timely recruitment of innate immune cells to wounded tissues through regulating mmp9 expression within epidermal cells.

Mouse epidermal cells upregulate IRG1 during inflammation

Atopic dermatitis is an increasingly common allergic skin disease characterized by chronically relapsing cutaneous inflammation. Skin wash samples from atopic dermatitis lesions possess elevated levels of MMP9 and MMP activity, raising the possibility of a functional role for MMP9 during atopic dermatitis-associated inflammation⁴³. Elevated expression of IRG1 (the mammalian homologue of zebrafish Irg1l) has only been reported in the uterine epithelium during implantation, within immune-stimulated monocytes/macrophages (where it localizes to the mitochondria) and ‘neurotoxic’ microglia^29,44,45,46. We sought to explore whether IRG1 expression was also elevated within epidermal cells in a mouse model of atopic dermatitis.

Using an ovalbumin-mediated epicutaneous sensitization model of atopic dermatitis, we assessed IRG1 abundance within the inflamed skin by immunohistochemistry. Strong staining for IRG1 was detected in the thickened epidermis of ovalbumin-sensitized mice (employing two different IRG1 antibodies), when compared with control (PBS-sensitized) skin biopsies (Fig. 7a,b and Supplementary Fig. 11a,b). Immunofluorescence revealed IRG1 was present within the mitochondria of epidermal cells (Fig. 7c,d).

**Figure 7: Murine IRG1 is detected within epidermal cells in a mouse model of atopic dermatitis and localizes to the mitochondria.**

These results confirm elevated expression of IRG1 within the inflamed epidermis in a mouse model of atopic dermatitis and support a conserved role for mammalian IRG1 during cutaneous inflammation.

Irg1l regulates β-oxidation-driven mROS production

Beta-oxidation is the metabolic process where fatty acids are catabolized in the mitochondria to generate ATP through the OXPHOS pathway. As a by-product of OXPHOS, mitochondria produce ROS, in the form of superoxide, as electrons leak from the electron transport chain⁴⁷. Studies have revealed that ROS can function as a signalling molecule to regulate the transcription and functional activity of various inflammatory mediators, including MMPs^14,15,48. We have previously shown that the paralog of Irg1l, Irg1, is required within the mitochondria of activated macrophages for enhanced utilization of fatty acids that fuels mROS production⁷. We investigated whether Irg1l-dependent epidermal expression of mmp9 could be explained by a functional role for Irg1l during β-oxidation-driven mROS production.

To first assess whether epidermal cells increased production of mROS following infection, we took advantage of the mitochondria-specific MitoTracker and ROS-sensitive CellROX fluorescent probes. Live imaging ROS within epidermal cells of larvae co-injected with Salmonella, MitoTracker and CellROX into the hindbrain ventricle showed that ROS was restricted to mitochondria (Fig. 8a,b). Quantifying overlapping CellROX/MitoTracker fluorescence signal within epidermal cells overlying the midbrain/hindbrain infection site (measured as total fluorescence intensity of colocalized signal) revealed elevated mROS production, when compared with PBS-injected controls (Fig. 8c). Of note, similar analysis in the presence of the mitochondrial OXPHOS inhibitor rotenone abolished this infection-driven augmentation in ROS production, confirming the ROS detected was coupled to mitochondrial OXPHOS (Supplementary Fig. 12). Similar analysis within Irg1l-depleted larvae (Fig. 8c and Supplementary Fig. 13a) and larvae treated with an inhibitor of fatty acid β-oxidation, etomoxir (Fig. 8d and Supplementary Fig. 13b), demonstrated that this elevated mROS production was Irg1l- and β-oxidation-dependent. Examining mmp9 expression and macrophage recruitment within infected etomoxir-treated larvae revealed β-oxidation of fatty acids contributed to infection-responsive epidermal expression of mmp9 and macrophage recruitment (Fig. 8e,f and Supplementary Fig. 14a,b). Of note, etomoxir-treated larvae maintained unaltered (whole-larva) numbers of macrophages (Supplementary Fig. 7).

**Figure 8: Irg1l regulates β-oxidation-driven mROS production within epidermal cells that elevate mitochondrial uptake of fatty acids following infection.**

These data suggest that, in response to infection, Irg1l helps regulate β-oxidation of fatty acids within the mitochondria of epidermal cells, resulting in elevated mROS-driven mmp9 expression that contributes to macrophage recruitment.

Infection enhances mitochondrial uptake of fatty acids

We have previously live-imaged the subcellular trafficking of the fluorescent fatty acid analogue BODIPY FL C₁₆ to show that mitochondria of activated macrophages enhance their uptake of fatty acids to fuel OXPHOS-coupled mROS production⁷. We have further demonstrated that this infection-responsive process requires the activity of the zebrafish paralog of Irg1l, Irg1. Our model for Irg1l function within epidermal cells predicts similar enhanced uptake of fatty acids by the mitochondria to fuel mROS production and subsequent MMP activity. To investigate whether epidermal cells enhance their uptake of fatty acids in response to infection, we investigated the subcellular distribution of BODIPY FL C₁₆ within epidermal cells overlying the midbrain/hindbrain infection site.

Immersion of larvae within BODIPY FL C₁₆ and MitoTracker probe following injection of Salmonella into the hindbrain ventricle revealed mitochondrial uptake of the fluorescent fatty acid analogue within overlying epidermal cells, as detected at 4 hpi (Fig. 8g,h). Of note, although BODIPY FL C₁₆ largely colocalized to the mitochondria, in some instances fluorescent signal was detected outside of the mitochondria, albeit in close proximity (Fig. 8g). This likely reflects the dynamic process of fatty acid transport to the mitochondria. Further supporting the mitochondrial uptake of BODIPY FL C₁₆, mitochondrial localization within epidermal cells following infection was disrupted following treatment with etomoxir (Supplementary Fig. 15a,b). Etomoxir functions as an inhibitor of carnitine palmitoyltransferase-1 (ref. 49), a rate-limiting enzyme essential for the transport of fatty acids across the outer mitochondrial membrane. Quantifying the total fluorescence intensity of BODIPY FL C₁₆ within these cells showed enhanced mitochondrial uptake following infection, when compared with PBS-injected controls (Fig. 8i–k).

These data suggest that, following infection, elevated uptake and β-oxidation of fatty acids within the mitochondria of epidermal cells fuels mROS production through an Irg1l-dependent mechanism. Enhanced mROS-driven mmp9 expression then contributes to macrophage recruitment.

β-oxidation contributes to MMP production following wounding

We next investigated whether similar enhanced fatty acid β-oxidation was driving Irg1l-dependent mmp9 expression and neutrophil migration within our wound-induced model of cutaneous inflammation.

Similar to Irg1l-depleted larvae (Fig. 6c,d,g), etomoxir treatment resulted in a marked reduction in wound-induced epithelial mmp9 expression (Fig. 9a,b) and a significant reduction in neutrophil recruitment to tail fin wounds (Fig. 9c), despite maintaining unaltered whole-larvae numbers of Lyz⁺ neutrophils (Supplementary Fig. 10). We next exploited the live imaging potential of the MMPSense FAST fluorescent probe that produces fluorescence following cleavage by MMPs⁵⁰, to live image MMP activity within tail fin wounds. Measuring the total fluorescence signal of MMPSense within a defined wound region (150 μm from the wound border), at 4 hpw, we detected wound-specific MMP activity that was both MMP9- and Irg1l-dependent (Fig. 9d–g). This fluorescent signal at the wound region was also significantly diminished in the presence of the pan-MMP inhibitor GM6001 and etomoxir (Fig. 9h–k). Zebrafish leukocytes express both mmp9 and mmp13 (refs 36, 51), suggesting that they are a source of MMP activity. To confirm that the primary source of wound-induced MMP we were detecting and measuring using the MMPSense probe was epidermal, we quantified the percentage of total fluorescence intensity that was neutrophil- and macrophage-specific. Live imaging and quantification of MMPSense fluorescent signal within Tg(lyz:EGFP) and Tg(mpeg1:EGFP) larvae revealed that only 0.012±0.027% (n=11 larvae) and 2.05±2.26% (n=11 larvae) of the total fluorescence intensity within the wound region colocalized with Lyz⁺ neutrophils and Mpeg1⁺ macrophages, respectively (Supplementary Fig. 16).

**Figure 9: β-oxidation and Irg1l contribute to MMP production by epidermal cells that elevate mitochondrial uptake of fatty acids following wounding.**

Wounding enhances mitochondrial uptake of fatty acids

To evaluate whether the mitochondria of epidermal cells enhanced their uptake of fatty acids following wounding, similar to that observed following infection (Fig. 8g–k), we live-imaged the subcellular trafficking of BODIPY FL C₁₆ within epidermal cells following wounding. When compared with an equivalent anatomical position within the tail of non-wounded larvae, epidermal cells of wounded larvae demonstrated enhanced mitochondrial uptake of BODIPY FL C₁₆ (Fig. 9l–p) within the wound region (150 μm from the wound border). Of note, mitochondrial uptake of BODIPY FL C₁₆ was again inhibited following etomoxir treatment (Supplementary Fig. 15c,d).

Similar to our infection-induced inflammation model, these results suggest that elevated uptake and β-oxidation of fatty acids within the mitochondria of epidermal cells following wounding fuels MMP production through an Irg1l-dependent mechanism. Enhanced MMP activity then contributes to the timely recruitment of neutrophils to the wound region.

Discussion

By exploiting the live-imaging potential of embryonic zebrafish to dissect the function of Irg1l, we have uncovered a new mechanism by which epidermal cells help direct the recruitment of leukocytes during inflammation. Using an infection-induced model of cutaneous inflammation we show that infection-responsive expression of irg1l within epidermal cells is MyD88-dependent and GC-sensitive. We predict that resident MyD88⁺ macrophages stimulate epidermal irg1l expression via secreted cytokines. Although strongest in epidermal cells overlying the infection site, epidermal expression of irg1l (and mmp9) was detected throughout the entire larvae following infection, suggesting that this macrophage-derived signal acts systemically. A previous study has shown zebrafish Cxcl8 (IL8), secreted from an unidentified cellular source at the infection site, acts systemically to trigger the rapid recruitment of leukocytes from the caudal hematopoietic tissue⁵². Investigating whether a similar systemically acting chemokine/chemokine receptor partnership operates in our model to drive epidermal expression of irg1l is a focus of future research. Of interest, infection-induced upregulation of irg1l expression within GR/Nr3c1⁺ oropharyngeal tissues was not suppressed following GC treatment, suggesting that the sensitivity of irg1l expression to GC-mediated suppression is tissue-specific. We provide evidence that Irg1l is necessary for the clearance of injected bacteria through helping regulate enhanced fatty acid β-oxidation to fuel mROS-driven mmp9 expression within epidermal cells, which helps direct further macrophage recruitment (Fig. 10). We further reveal that a similar Irg1l-dependent immunometabolic mechanism operates within epidermal cells during wound-induced inflammation to help orchestrate the timely recruitment of neutrophils (Fig. 10). In addition to uncovering a new immunometabolic function for epidermal cells during inflammation, this study represents the first whole-animal live-imaging investigation of epidermal cell metabolism during inflammation.

**Figure 10: Proposed model describing function of Irg1l within epidermal cells following infection and wounding.**

Zebrafish possess co-orthologues for IRG1 (irg1 and irg1l). Our previous study has established a role for zebrafish Irg1 in coupling metabolism with immune cell function within macrophages⁷. We showed that irg1 was specifically expressed within immune-stimulated macrophages where the encoded enzyme was required for enhanced β-oxidation of fatty acids to fuel OXPHOS-coupled mROS production. Here we demonstrate a similar role for Irg1l during enhanced utilization of fatty acids to fuel mROS production within epidermal cells during infection- and wound-induced inflammation. To date, mammalian IRG1 expression has been reported in activated macrophages, neurotoxic microglia and within the uterine epithelium^44,46,53. IRG1 has been shown to localize to the mitochondria using fluorescent fusion proteins and cell fractionation experiments⁴⁵. Functional experiments further support a mitochondrial association^7,54. Employing a mouse model of atopic dermatitis and antibodies against human IRG1, we have shown elevated levels of IRG1, as detected by immunofluorescence, within the inflamed skin of ovalbumin-sensitized mice, where it associates with the mitochondria of keratinocytes. Our model predicts IRG1 is required for enhanced β-oxidation of fatty acids within keratinocytes during inflammation. In all eukaryotic organisms, β-oxidation occurs in mitochondria and peroxisomes⁵⁵. It is emerging that, in response to inflammatory stress, mitochondria are highly dynamic and can form intimate physical interactions with other organelles (including peroxisomes and mitochondrial-associated endoplasmic reticulum membranes)⁵⁶. These interactions are believed to form signalling scaffolds that help coordinate immune cell function. Identifying how IRG1 associates with the mitochondria in our mouse dermatitis model, by generating mouse-specific IRG1 antibodies and performing immunogold labelling/transmission electron microscopy, is a focus of future research. Our results support a conserved immunometabolic role for IRG1 within mammalian epidermal cells during cutaneous inflammation and suggest that the function of mammalian IRG1 is subfunctionally partitioned between Irg1 and Irg1l in zebrafish. Subfunctional partitioning is common with duplicated genes resulting from the pre-teleost genome duplication. Occupying separate loci, duplicate genes are independently mutable and can acquire mutations within regulatory elements that help direct tissue-specific expression. This can lead to paralogs with tissue-specific functions that cumulatively recapitulate that of the ancestral gene⁵⁷. Unlike ray-finned fish and reptiles, mammals possess one IRG1 orthologue ( http://www.ensembl.org/Danio_rerio/Gene/Compara_Tree?g=ENSDARG00000069844;r=9:22866973-22871390), suggesting that degenerative mutations in coding and/or regulatory regions (nonfunctionalization) have contributed to the loss of one of the gene duplicates in this lineage, a fate that accounts for most duplicated genes⁵⁸.

Our model predicts that, during cutaneous inflammation, elevated β-oxidation of fatty acids within epidermal cells contributes to the recruitment of leukocytes through mROS-driven augmentation of mmp9 expression. MMPs are well-recognized mediators of inflammation and innate immune cell activity⁵⁹. By directly modulating chemokine gradients and degrading ECM components, MMPs shape the outcome of inflammation by helping direct the recruitment of innate immune effector cells (including neutrophils and monocytes/macrophages) to sites of infection or injury⁵⁹. During cutaneous inflammation, keratinocytes are a major source of MMP activity (including that provided by MMP-9) that, in addition to regulating re-epithelialization, are believed to contribute to leukocyte recruitment^13,59. Supporting a connection between mitochondrial redox signalling and MMP activity, evidence suggests that mROS-sensitive signalling pathways direct both the expression and activity of MMP family members¹⁴. Furthermore, heat shock-induced MMP-9 expression within a human keratinocyte cell line is, in part, regulated by mROS⁶⁰, and TNF-α-stimulated human keratinocytes enhance production of MMP-9 through a ROS/NF-κB-dependent mechanism¹⁵. Our study reveals the contribution of mROS-driven MMP activity towards leukocyte recruitment is fuelled by metabolic reprogramming in keratinocytes towards elevated fatty acid β-oxidation.

Supporting a connection between elevated β-oxidation and the inflammatory response during cutaneous inflammation, the expression of enzymes associated with fatty acid oxidation are elevated within the skin of patients with the inflammatory dermatosis psoriasis^61,62. It has also been suggested that epidermal fatty acid-binding protein contributes to the pathogenesis of psoriasis through activation of the NF-κB signalling pathway⁶³, a powerful transcriptional regulator of pro-inflammatory genes. Furthermore, etomoxir-mediated β-oxidation inhibition is beneficial in the treatment of psoriasis⁶⁴. We provide a mechanistic link connecting enhanced fatty acid β-oxidation within epidermal cells to their immune function that acts to direct leukocyte recruitment. The precise mechanism through which Irg1l/IRG1 participates during β-oxidation-fuelled augmentation of mROS is the focus of future work.

Metabolic reprogramming as a strategy to regulate the immune response is known to operate within immune cells^3,4. To our knowledge, this study describes the first example of metabolic reprogramming being employed within a non-immune cell type to help shape its contribution to the immune response. Pharmacologic targeting of this metabolic–immunological interface within keratinocytes may prove efficacious in treating inflammatory dermatoses.

Methods

Zebrafish maintenance

Zebrafish (Danio rerio) embryos were obtained from natural spawnings and raised at 28 °C in E3 Medium supplemented with 0.003% phenylthiourea (PTU) to inhibit pigmentation. The Tg(mpeg1:Gal4-VP16)^gl24; Tg(UAS-E1b:Kaede)^s1999t (herein referred to as Tg(mpeg1:Gal4/UAS:Kaede)), Tg(mpeg1:EGFP)^gl22, Tg(mpeg1:mCherry)^gl23, Tg(lyz:EGFP)^NZ117 and Tg(ef1-α: DsRED2) transgenic reporter lines were used in this study. Research was conducted with approval from the University of Auckland Animal Ethics Committee.

Mice

Six to 8-week-old female C57BL/6 mice were bred and housed in the Animal Resource Unit, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. The mice were kept in an air-conditioned room with controlled humidity, temperature and a 12 h light/dark cycle. All experiments were conducted under a protocol approved by the University of Auckland Animal Ethics Committee.

Live bacterial and LPS injection of zebrafish larvae

Zebafish larvae were injected with non-fluorescent wild-type or GFP-expressing S. enterica serovar Typhimurium into the hindbrain ventricle as previously described³¹. In brief, anaesthetized 2 dpf larvae were mounted in 3% methyl cellulose (in E3 media) and injected with 1 nl of bacterial injection preparation (~\n600 cfu/nl in filter-sterilized PBS) into the hindbrain ventricle. Immediately before and after injecting a cohort of larvae, an injection bolus was injected into filter-sterilized PBS (in duplicate) that was plated at 1:10 and 1:100 dilutions to quantify and validate infection dose. LPS (E. coli K12, InvivoGen) was similarly injected (0.8 mg ml⁻¹, 1 nl), as previously described⁷.

Microarray and analysis

Total RNA was extracted from dissected trunks of infected and PBS control-injected larvae at 2 dpi (injected at 2 dpf) using TRIzol reagent (Invitrogen) and purified using Qiagen RNeasy columns as per the manufacturer’s instructions. Hybridization to Affymetrix GeneChip Zebrafish Genome arrays and data acquisition were performed at the Centre for Genomics and Proteomics, School of Biological Sciences, University of Auckland. Data were normalized using the RMA method and statistical analysis of differential expression between control and infected larval zebrafish trunks was performed using linear models (LIMMA).

Whole-mount in situ hybridization

WMISH was performed using digoxigenin and fluorescein-labelled riboprobes (Roche) and BM purple (Roche) colour precipitation or Tyramide Signal Amplification (Invitrogen) for fluorescent WMISH analysis, as previously described^31,65.

RT–PCR

Total RNA was extracted from embryos and larvae using TRIzol reagent (Invitrogen) as per the manufacturer’s instructions. Superscript III reverse transcriptase (Invitrogen)-generated cDNAs were used to produce gene-specific amplicons using primer pairs designed to span at least 1 intron to control for contaminating genomic DNA. To assess Irg1l and MMP9 MO specificities the following (forward/reverse) primer pairs were used, 5′-TCCAGATCCTGACCACAAC-3′/5′-TTATTGCAGTTGGGCAAGC-3′ and 5′-CTGATCTGATCTGAGCACTTGC-3′/5′-GCTGACGGACAAATATGCCTAAC-3′ respectively. Primers to ef1-α (5′-AAGGAGAAGATCGACCGTCG-3′ and 5′-TGATTGAGGGAAATTCACTTGG-3′) were used as an internal control. To generate irg1l-, krt4- and mmp9-targeting riboprobes the following primer pairs were used: irg1l, 5′-TCCAGATCCTGACCACAAC-3′ and 5′-TTATTGCAGTTGGGCAAGC-3′; krt4, 5′-ATGGGCTTTGTTGCGGCTCCTATC-3′ and 5′-TGAGTCGTAAGAAGGTTCAGAAGGG-3′; mmp9, 5′-CGACTCAAGATGAGACTTGGAG-3′ and 5′-GCTGACGGACAAATATGCCTAAC-3′. Riboprobes to detect myd88 were as previously described¹⁸.

MO injection

MOs (Gene Tools, Philomath, OR, USA) were resuspended in sterile water and injected in 1 × Danieau’s solution. Efficacious doses for all MOs were determined empirically. RT–PCR was used to determine MO specificity where splice-blocking MOs were used. Sequences of MOs (and doses) are as follows: control MO, 5′-CCTCTTACCTCAGTTACAATTTATA-3′ (1.5 pmole); Nr3c1 MO (splice-blocking), 5′-GCCAGAGATATATGGAATACCTTCA-3′ (0.5 pmole)⁷; Irg1l MO (splice-blocking), 5′-AACACATTAGGCAGTACTTACTGCT-3′ (1.5 pmole); Nr5a1a MO (translation-blocking), 5′-AATCCTCATCTGCTCTGAAGTC-3′ (0.5 pmole)³⁷; Irf8 MO (splice-blocking), 5′-AATGTTTCGCTTACTTTGAAAATGG-3′ (1.5 pmole)³³; MyD88 MO (splice-blocking), 5′-GGTTAAACACTGACCCTGTGGATC-3′ (1.0 pmole)⁷; Gcsrf MO (splice-blocking), 5′-ATTCAAGCACATACTCACTTCCATT-3′ (0.5 pmole)³¹; Spi1 MO (translation-blocking), 5′-GATATACTGATACTCCATTGGTGGT-3′ (1.0 pmole)⁶⁶; MMP9 MO (splice-blocking), 5′-ATACAAAATACTCGACTCACCTGCT-3′ (0.25 pmole).

Confocal imaging

Live embryos and larvae were mounted for confocal imaging, as previously described⁶⁷. All live confocal imaging was performed on an Olympus FV1000 FluoView laser scanning confocal microscope equipped with an incubation chamber. Confocal imaging of fluorescent WMISH was performed using a Nikon C1 Eclipse confocal microscope. Images collected by confocal microscopy were processed and analysed using Volocity 6.1.1 (Perkin Elmer). To detect ROS production and mitochondria, the CellROX Deep Red Reagent (Invitrogen C10422, 1 nl at 50 μM, absorption/emission maxima of ~\n644/665 nm) and MitoTracker Red CMXRos (Invitrogen M7512, 1 nl at 20 μM, absorption/emission maxima of ~\n579/599 nm) fluorescent dyes were injected into the hindbrain ventricle, respectively. To live image BODIPY FL C₁₆ and mitochondria larvae were immersed in 50 μM BODIPY FL C₁₆ (Invitrogen, D-3821, imaged using absorption/emission maxima as used to detect Alexa Fluor 488 (~\n495/519 nm)) and 1 μM MitoTracker Deep Red FM (Invitrogen M22426, absorption/emission maxima of ~\n644/665 nm), respectively, for 2 h immediately following infection or tail transection, washed for 30 min in PTU-treated E3 medium before mounting and imaging. To live image MMP activity, larvae were immersed in 40 μM MMPSense 645 FAST (Perkin Elmer, NEV10100, absorption/emission maxima of ~\n649/666 nm) for 2 h immediately following tail transection, washed for 30 min in PTU-treated E3 medium before mounting and imaging. When quantifying mROS, MMP activity and BODIPY FL C₁₆ uptake using fluorescence intensity measurements, conservative laser excitation settings were used to ensure that we did not create/measure a saturated fluorescence signal. See Supplementary Methods for detailed descriptions of confocal imaging/analysis when used for neutrophil tracking and quantification of macrophage numbers at infection site, mROS production, BODIPY FL C₁₆ fluorescence and MMPSense FAST fluorescence.

Chemical treatments

The following chemical agonists and antagonists were used. No abnormal development or toxicity was observed for all compounds used at the indicated working concentrations and routes of delivery. The GR antagonist RU-486 (Sigma, M8046) was delivered by immersion (2 h pre-treatment before infection, 250 μM 1% DMSO). The β-oxidation inhibitor etomoxir (Sigma, E1905) was delivered by immersion (1–2 h pre-treatment before infection or wounding, 100 μM for infection and 10 μM for wounding experiments, 1% DMSO). The OXPHOS inhibitor rotenone (Sigma, R8875) was delivered by immersion (1–2 h pre-treatment before infection, 1 μM 1% DMSO). The MMP inhibitor GM6001 (Sigma, M5939) was delivered by immersion (1–2 h pre-treatment before wounding, 200 μM 1% DMSO). The GR agonists dexamethasone (Sigma), prednisolone (Sigma) and hydrocortisone (Sigma) were injected into the hindbrain ventricle (1 nl at 0.67 mg ml⁻¹) with PBS or Salmonella. Anti-inflammatory activities of dexamethasone, prednisolone and hydrocortisone (delivered by immersion at ~\n5 μM, 1% DMSO) were assessed by their ability to inhibit neutrophil migration using an acute tail-wounding assay with 3 dpf Tg(lyz:EGFP) larvae.

Quantitative PCR

qPCR was performed with Platinum SYBR Green qPCR SuperMix-UDG with ROX (Invitrogen) using an ABI PRISM 7900HT Fast sequence detection system (Applied Biosystems) as previously described⁶⁸. Gene-specific oligonucleotides were designed using Primer Express software to detect irg1l (5′-CACTAGATGTGGCAGAGCGT-3′ and 5′-CCCAACTCCAATGCTGTCTA-3′), mmp9 (5′-TGGGCACCTGCTCGTTGA-3′ and 5′-TTGGAGATGACCGCCTGC-3′) and ef1-α (5′-TGCCTTCGTCCCAATTTCAG-3′ and 5′-TACCCTCCTTGCGCTCAATC-3′). All reactions were performed in biological triplicate and technical quadruplicate and a melting curve analysis carried out to confirm reaction specificity. Analysis used the comparative CT method with ef1-α as an internal control.

Individual CFU counts

Individual CFU counts of larvae infected with Sal-GFP was performed as previously described⁶⁹. Larval homogenates were plated in a dilution series on LB agar plates supplemented with 25 μg ml⁻¹ kanamycin.

Acute tail wounding

An acute tail fin wound was generated by transecting the tail fin of anaesthetized 2 dpf larvae using a sterile scalpel. Care was taken to transect tails at exactly the same anatomic location (posterior-most edge of notochord).

Flow cytometry

Flow cytometry was performed as previously described¹⁸. In brief, larvae were rinsed in calcium-free Ringer’s solution for 15 min before removal of yolks by repeated passage through a 200 μl pipette tip. Larvae were then digested in 0.25% trypsin-EDTA (Gibco) for 2 h at 28.5 °C with manual dissociation every 10 min. The digestion was inhibited by the addition of CaCl₂ to a concentration of 1 mM and 5% FBS. The dissociated cells were centrifuged for 5 min at 3,000 r.p.m. before resuspension in ice-cold 0.9 × PBS, 5% FBS. Cells were passed several times through 40-μm cell strainers (BD Falcon) before sorting. Flow cytometry was performed using a Becton Dickinson LSRII flow cytometer.

Immunofluorescence and immunohistochemistry

Immunohistochemical staining of paraffin sections was performed using the Vectastain elite ABC kit (Vector Laboratories, PK-6100) as per the manufacturer’s instructions. Immunofluorescence was carried out as previously described¹⁸. Primary and secondary antibody combinations used were the following: chicken anti-GFP (1:500/Abcam, ab13970) with goat anti-chicken Alexa Fluor 488 (1:500/Invitrogen); rabbit anti-Kaede (1:500/MBL, PM012) with goat anti-rabbit Alexa Fluor 633 or goat anti-rabbit Alexa Fluor 488 (1:500/Invitrogen); rabbit anti-Human GR (P-20) (1:75/Santa Cruz Biotechnology Inc., sc-1002) with goat anti-rabbit Alexa Fluor 546 (1:500/Invitrogen); rabbit anti-human IRG1 (1:300/N-20, sc-84191 and C-18, sc-84189, Santa Cruz Biotech.) with goat anti-rabbit Alexa Fluor 488 (1:500/Invitrogen) or biotinylated horse anti-rabbit IgG; mouse anti-mitochondria antibody (1:500/Abcam, ab3298) with goat anti-mouse Alexa Fluor 568 (1:500/Invitrogen).

Ovalbumin-mediated epicutaneous sensitization model of AD

The protocol for inducing atopic dermatitis was as previously described⁷⁰. In brief, mice were shaved, and the dorsal skin tape-stripped six times under anaesthesia. Ovalbumin (100 μg in 100 μl of PBS) or PBS (as a negative control) was placed on a patch of gauze (1 × 1 cm), which was secured to the skin site with a transparent bio-occlusive dressing for a 1-week period and then removed. Each mouse received a total of three 1-week exposures to a patch separated by 2-week patch-free intervals. Mice were monitored daily to ensure that patches remained in place throughout the sensitization period. Mice were killed 24 h after removal of the third patch and skin tissue samples were collected. Skin tissues were fixed in 10% formalin, embedded into paraffin and sectioned.

Statistical analysis

Statistical analyses were performed using Prism 5.0 (GraphPad Software, Inc.). Statistical significance was assessed using unpaired, two-tailed t-tests and all data presented in scatter plots show means with s.d. P values <0.05 were considered statistically significant.

Additional information

Accession codes: Microarray data used in this study have been deposited in the Gene Expression Omnibus, accession number GSE56365.

How to cite this article: Hall, C. J. et al. Epidermal cells help coordinate leukocyte migration during inflammation through fatty acid-fuelled matrix metalloproteinase production. Nat. Commun. 5:3880 doi: 10.1038/ncomms4880 (2014).

Accession codes

Accessions

Gene Expression Omnibus

GSE56365

References

Odegaard, J. I. & Chawla, A. The immune system as a sensor of the metabolic state. Immunity 38, 644–654 (2013).
Article CAS PubMed PubMed Central Google Scholar
Mathis, D. & Shoelson, S. E. Immunometabolism: an emerging frontier. Nat. Rev. Immunol. 11, 81–83 (2011).
Article CAS PubMed PubMed Central Google Scholar
Pearce, E. L. & Pearce, E. J. Metabolic pathways in immune cell activation and quiescence. Immunity 38, 633–643 (2013).
Article CAS PubMed PubMed Central Google Scholar
Biswas, S. K. & Mantovani, A. Orchestration of metabolism by macrophages. Cell Metab. 15, 432–437 (2012).
Article CAS PubMed Google Scholar
Chawla, A., Nguyen, K. D. & Goh, Y. P. Macrophage-mediated inflammation in metabolic disease. Nat. Rev. Immunol. 11, 738–749 (2011).
Article CAS PubMed PubMed Central Google Scholar
West, A. P., Shadel, G. S. & Ghosh, S. Mitochondria in innate immune responses. Nat. Rev. Immunol. 11, 389–402 (2011).
Article CAS PubMed PubMed Central Google Scholar
Hall, C. J. et al. Immunoresponsive gene 1 augments bactericidal activity of macrophage-lineage cells by regulating beta-oxidation-dependent mitochondrial ROS production. Cell Metab. 18, 265–278 (2013).
Article CAS PubMed Google Scholar
West, A. P. et al. TLR signalling augments macrophage bactericidal activity through mitochondrial ROS. Nature 472, 476–480 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Proksch, E., Brandner, J. M. & Jensen, J.-M. The skin: an indispensable barrier. Exp. Dermatol. 17, 1063–1072 (2008).
Article PubMed Google Scholar
Kupper, T. S. Immune and inflammatory processes in cutaneous tissues. Mechanisms and speculations. J. Clin. Invest. 86, 1783–1789 (1990).
Article CAS PubMed PubMed Central Google Scholar
Boguniewicz, M. & Leung, D. Y. Atopic dermatitis: a disease of altered skin barrier and immune dysregulation. Immunol. Rev. 242, 233–246 (2011).
Article CAS PubMed PubMed Central Google Scholar
Pilcher, B. K. et al. Role of matrix metalloproteinases and their inhibition in cutaneous wound healing and allergic contact hypersensitivity. Ann. NY Acad. Sci. 878, 12–24 (1999).
Article ADS CAS PubMed Google Scholar
Purwar, R., Kraus, M., Werfel, T. & Wittmann, M. Modulation of keratinocyte-derived MMP-9 by IL-13: a possible role for the pathogenesis of epidermal inflammation. J. Invest. Dermatol. 128, 59–66 (2008).
Article CAS PubMed Google Scholar
Nelson, K. K. & Melendez, J. A. Mitochondrial redox control of matrix metalloproteinases. Free Radic. Biol. Med. 37, 768–784 (2004).
Article CAS PubMed Google Scholar
Song, H. Y. et al. Suppression of TNF-alpha-induced MMP-9 expression by a cell-permeable superoxide dismutase in keratinocytes. BMB Rep. 44, 462–467 (2011).
Article CAS PubMed Google Scholar
Trede, N. S., Langenau, D. M., Traver, D., Look, A. T. & Zon, L. I. The use of zebrafish to understand immunity. Immunity 20, 367–379 (2004).
Article CAS PubMed Google Scholar
Ellett, F., Pase, L., Hayman, J. W., Andrianopoulos, A. & Lieschke, G. J. mpeg1 promoter transgenes direct macrophage-lineage expression in zebrafish. Blood 117, e49–e56 (2011).
Article CAS PubMed PubMed Central Google Scholar
Hall, C. et al. Transgenic zebrafish reporter lines reveal conserved Toll-like receptor signaling potential in embryonic myeloid leukocytes and adult immune cell lineages. J. Leukoc. Biol. 85, 751–765 (2009).
Article CAS PubMed Google Scholar
Hall, C., Flores, M. V., Storm, T., Crosier, K. & Crosier, P. The zebrafish lysozyme C promoter drives myeloid-specific expression in transgenic fish. BMC. Dev. Biol. 7, 42 (2007).
Article PubMed PubMed Central Google Scholar
Renshaw, S. A. et al. A transgenic zebrafish model of neutrophilic inflammation. Blood 108, 3976–3978 (2006).
Article CAS PubMed Google Scholar
Mathias, J. R. et al. Resolution of inflammation by retrograde chemotaxis of neutrophils in transgenic zebrafish. J. Leukoc. Biol. 80, 1281–1288 (2006).
Article CAS PubMed Google Scholar
Mathias, J. R. et al. Live imaging of chronic inflammation caused by mutation of zebrafish Hai1. J. Cell Sci. 120, 3372–3383 (2007).
Article CAS PubMed Google Scholar
Niethammer, P., Grabher, C., Look, A. T. & Mitchison, T. J. A tissue-scale gradient of hydrogen peroxide mediates rapid wound detection in zebrafish. Nature 459, 996–999 (2009).
Article ADS CAS PubMed PubMed Central Google Scholar
Yoo, S. K., Starnes, T. W., Deng, Q. & Huttenlocher, A. Lyn is a redox sensor that mediates leukocyte wound attraction in vivo. Nature 480, 109–112 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Le Guellec, D., Morvan-Dubois, G. & Sire, J. Y. Skin development in bony fish with particular emphasis on collagen deposition in the dermis of the zebrafish (Danio rerio). Int. J. Dev. Biol. 48, 217–231 (2004).
Article CAS PubMed Google Scholar
Chang, W. J. & Hwang, P. P. Development of zebrafish epidermis. Birth Defects Res. C Embryo Today 93, 205–214 (2011).
Article CAS PubMed Google Scholar
Li, Q. & Uitto, J. Zebrafish as a model system to study heritable skin diseases. Methods Mol. Biol. 961, 411–424 (2013).
Article CAS PubMed Google Scholar
Li, Q., Frank, M., Thisse, C. I., Thisse, B. V. & Uitto, J. Zebrafish: a model system to study heritable skin diseases. J. Invest. Dermatol. 131, 565–571 (2011).
Article CAS PubMed Google Scholar
Chen, B., Zhang, D. & Pollard, J. W. Progesterone regulation of the mammalian ortholog of methylcitrate dehydratase (immune response gene 1) in the uterine epithelium during implantation through the protein kinase C pathway. Mol. Endocrinol. 17, 2340–2354 (2003).
Article CAS PubMed Google Scholar
van Soest, J. J. et al. Comparison of static immersion and intravenous injection systems for exposure of zebrafish embryos to the natural pathogen Edwardsiella tarda. BMC Immunol. 12, 58 (2011).
Article CAS PubMed PubMed Central Google Scholar
Hall, C. J. et al. Infection-responsive expansion of the hematopoietic stem and progenitor cell compartment in zebrafish is dependent upon inducible nitric oxide. Cell Stem Cell 10, 198–209 (2012).
Article CAS PubMed Google Scholar
Su, F. et al. Differential regulation of primitive myelopoiesis in the zebrafish by Spi-1/Pu.1 and C/ebp1. Zebrafish 4, 187–199 (2007).
Article CAS PubMed Google Scholar
Li, L., Jin, H., Xu, J., Shi, Y. & Wen, Z. Irf8 regulates macrophage versus neutrophil fate during zebrafish primitive myelopoiesis. Blood 117, 1359–1369 (2011).
Article CAS PubMed Google Scholar
Trinchieri, G. & Sher, A. Cooperation of Toll-like receptor signals in innate immune defence. Nat. Rev. Immunol. 7, 179–190 (2007).
Article CAS PubMed Google Scholar
van der Sar, A. M. et al. MyD88 innate immune function in a zebrafish embryo infection model. Infect. Immun. 74, 2436–2441 (2006).
Article CAS PubMed PubMed Central Google Scholar
Zhang, Y. et al. In vivo interstitial migration of primitive macrophages mediated by JNK-matrix metalloproteinase 13 signaling in response to acute injury. J. Immunol. 181, 2155–2164 (2008).
Article CAS PubMed Google Scholar
Liu, Y. W. Interrenal organogenesis in the zebrafish model. Organogenesis 3, 44–48 (2007).
Article PubMed PubMed Central Google Scholar
Van Lint, P. & Libert, C. Chemokine and cytokine processing by matrix metalloproteinases and its effect on leukocyte migration and inflammation. J. Leukoc. Biol. 82, 1375–1381 (2007).
Article CAS PubMed Google Scholar
Volkman, H. E. et al. Tuberculous granuloma induction via interaction of a bacterial secreted protein with host epithelium. Science 327, 466–469 (2010).
Article ADS CAS PubMed Google Scholar
Gray, C. et al. Simultaneous intravital imaging of macrophage and neutrophil behaviour during inflammation using a novel transgenic zebrafish. Thromb. Haemost. 105, 811–819 (2011).
Article CAS PubMed Google Scholar
Pase, L. et al. Neutrophil-delivered myeloperoxidase dampens the hydrogen peroxide burst after tissue wounding in zebrafish. Curr. Biol. 22, 1818–1824 (2012).
Article CAS PubMed Google Scholar
Bai, S. et al. Matrix metalloproteinase expression and function during fin regeneration in zebrafish: analysis of MT1-MMP, MMP2 and TIMP2. Matrix Biol. 24, 247–260 (2005).
Article CAS PubMed Google Scholar
Harper, J. I. et al. A study of matrix metalloproteinase expression and activity in atopic dermatitis using a novel skin wash sampling assay for functional biomarker analysis. Br. J. Dermatol. 162, 397–403 (2010).
Article CAS PubMed Google Scholar
Li, H. et al. Different neurotropic pathogens elicit neurotoxic CCR9- or neurosupportive CXCR3-expressing microglia. J. Immunol. 177, 3644–3656 (2006).
Article CAS PubMed Google Scholar
Degrandi, D., Hoffmann, R., Beuter-Gunia, C. & Pfeffer, K. The proinflammatory cytokine-induced IRG1 protein associates with mitochondria. J. Interferon Cytokine Res. 29, 55–67 (2009).
Article CAS PubMed Google Scholar
Lee, C. G., Jenkins, N. A., Gilbert, D. J., Copeland, N. G. & O’Brien, W. E. Cloning and analysis of gene regulation of a novel LPS-inducible cDNA. Immunogenetics 41, 263–270 (1995).
Article CAS PubMed Google Scholar
Sena, L. A. & Chandel, N. S. Physiological roles of mitochondrial reactive oxygen species. Mol. Cell 48, 158–167 (2012).
Article CAS PubMed PubMed Central Google Scholar
Svineng, G., Ravuri, C., Rikardsen, O., Huseby, N. E. & Winberg, J. O. The role of reactive oxygen species in integrin and matrix metalloproteinase expression and function. Connect. Tissue Res. 49, 197–202 (2008).
Article CAS PubMed Google Scholar
Vickers, A. E. Characterization of hepatic mitochondrial injury induced by fatty acid oxidation inhibitors. Toxicol. Pathol. 37, 78–88 (2009).
Article CAS PubMed Google Scholar
Barber, P. A., Rushforth, D., Agrawal, S. & Tuor, U. I. Infrared optical imaging of matrix metalloproteinases (MMPs) up regulation following ischemia reperfusion is ameliorated by hypothermia. BMC Neurosci. 13, 76 (2012).
Article CAS PubMed PubMed Central Google Scholar
Yoong, S. et al. Characterization of the zebrafish matrix metalloproteinase 9 gene and its developmental expression pattern. Gene. Expr. Patterns 7, 39–46 (2007).
Article CAS PubMed Google Scholar
Deng, Q. et al. Localized bacterial infection induces systemic activation of neutrophils through Cxcr2 signaling in zebrafish. J. Leukoc. Biol. 93, 761–769 (2013).
Article CAS PubMed PubMed Central Google Scholar
Cheon, Y. P., Xu, X., Bagchi, M. K. & Bagchi, I. C. Immune-responsive gene 1 is a novel target of progesterone receptor and plays a critical role during implantation in the mouse. Endocrinology 144, 5623–5630 (2003).
Article CAS PubMed Google Scholar
Michelucci, A. et al. Immune-responsive gene 1 protein links metabolism to immunity by catalyzing itaconic acid production. Proc. Natl Acad. Sci. USA 110, 7820–7825 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Lodhi, I. J. & Semenkovich, C. F. Peroxisomes: a nexus for lipid metabolism and cellular signaling. Cell Metab. 19, 380–392 (2014).
Article CAS PubMed PubMed Central Google Scholar
Horner, S. M., Liu, H. M., Park, H. S., Briley, J. & Gale, M. Jr. Mitochondrial-associated endoplasmic reticulum membranes (MAM) form innate immune synapses and are targeted by hepatitis C virus. Proc. Natl Acad. Sci. USA 108, 14590–14595 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Postlethwait, J., Amores, A., Cresko, W., Singer, A. & Yan, Y. L. Subfunction partitioning, the teleost radiation and the annotation of the human genome. Trends Genet. 20, 481–490 (2004).
Article CAS PubMed Google Scholar
Lynch, M. & Conery, J. S. The evolutionary fate and consequences of duplicate genes. Science 290, 1151–1155 (2000).
Article ADS CAS PubMed Google Scholar
Parks, W. C., Wilson, C. L. & Lopez-Boado, Y. S. Matrix metalloproteinases as modulators of inflammation and innate immunity. Nat. Rev. Immunol. 4, 617–629 (2004).
Article CAS PubMed Google Scholar
Shin, M. H. et al. Reactive oxygen species produced by NADPH oxidase, xanthine oxidase, and mitochondrial electron transport system mediate heat shock-induced MMP-1 and MMP-9 expression. Free Radic. Biol. Med. 44, 635–645 (2008).
Article CAS PubMed Google Scholar
Madsen, P., Rasmussen, H. H., Leffers, H., Honore, B. & Celis, J. E. Molecular cloning and expression of a novel keratinocyte protein (psoriasis-associated fatty acid-binding protein [PA-FABP]) that is highly up-regulated in psoriatic skin and that shares similarity to fatty acid-binding proteins. J. Invest. Dermatol. 99, 299–305 (1992).
Article CAS PubMed Google Scholar
Masouye, I., Saurat, J. H. & Siegenthaler, G. Epidermal fatty-acid-binding protein in psoriasis, basal and squamous cell carcinomas: an immunohistological study. Dermatology 192, 208–213 (1996).
Article CAS PubMed Google Scholar
Ogawa, E. et al. Epidermal FABP (FABP5) regulates keratinocyte differentiation by 13(S)-HODE-mediated activation of the NF-kappaB signaling pathway. J. Invest. Dermatol. 131, 604–612 (2011).
Article CAS PubMed Google Scholar
Caspary, F. et al. A new therapeutic approach to treat psoriasis by inhibition of fatty acid oxidation by Etomoxir. Br. J. Dermatol. 153, 937–944 (2005).
Article CAS PubMed Google Scholar
Jowett, T. & Lettice, L. Whole-mount in situ hybridizations on zebrafish embryos using a mixture of digoxigenin- and fluorescein-labelled probes. Trends Genet. 10, 73–74 (1994).
Article CAS PubMed Google Scholar
Rhodes, J. et al. Interplay of pu.1 and gata1 determines myelo-erythroid progenitor cell fate in zebrafish. Dev. Cell 8, 97–108 (2005).
Article CAS PubMed Google Scholar
Hall, C., Flores, M. V., Crosier, K. & Crosier, P. Live cell imaging of zebrafish leukocytes. Methods Mol. Biol. 546, 255–271 (2009).
Article CAS PubMed Google Scholar
Oehlers, S. H. et al. A chemical enterocolitis model in zebrafish larvae that is dependent on microbiota and responsive to pharmacological agents. Dev. Dyn. 240, 288–298 (2011).
Article CAS PubMed Google Scholar
Oehlers, S. H. et al. The inflammatory bowel disease (IBD) susceptibility genes NOD1 and NOD2 have conserved anti-bacterial roles in zebrafish. Dis. Model Mech. 4, 832–841 (2011).
Article CAS PubMed PubMed Central Google Scholar
Sun, X., Zhang, J., Macgibbon, A. K., Black, P. & Krissansen, G. W. Bovine milk fat enriched in conjugated linoleic and vaccenic acids attenuates allergic dermatitis in mice. Clin. Exp. Allergy 41, 729–738 (2011).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

We thank Annie Chien and Alisha Malik for excellent technical assistance and Alhad Mahagaonkar for expert management of our zebrafish facility. We thank the Biomedical Imaging Research Unit, School of Medical Sciences, University of Auckland, for expert assistance in fluorescence and confocal microscopy and Stephen Edgar for flow cytometry support. We also thank Alan Davidson for supplying the cyp11a1 probe template and Graham Lieschke for providing the mpeg1 transgenic reporter lines. Funding for this work was provided by a grant awarded to P.S.C from the Ministry of Business, Innovation and Employment, New Zealand.

Author information

Authors and Affiliations

Department of Molecular Medicine and Pathology, School of Medical Sciences, University of Auckland, Auckland, 1023, New Zealand
Christopher J. Hall, Rachel H. Boyle, Xueying Sun, Sophie M. Wicker, June P Misa, Geoffrey W. Krissansen, Cristin G. Print, Kathryn E. Crosier & Philip S. Crosier

Authors

Christopher J. Hall
View author publications
You can also search for this author in PubMed Google Scholar
Rachel H. Boyle
View author publications
You can also search for this author in PubMed Google Scholar
Xueying Sun
View author publications
You can also search for this author in PubMed Google Scholar
Sophie M. Wicker
View author publications
You can also search for this author in PubMed Google Scholar
June P Misa
View author publications
You can also search for this author in PubMed Google Scholar
Geoffrey W. Krissansen
View author publications
You can also search for this author in PubMed Google Scholar
Cristin G. Print
View author publications
You can also search for this author in PubMed Google Scholar
Kathryn E. Crosier
View author publications
You can also search for this author in PubMed Google Scholar
Philip S. Crosier
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

C.J.H., R.H.B., X.S., S.M.W., J.P.M, G.W.K and C.G.P performed experiments and helped analyse the data. C.J.H., K.E.C and P.S.C. wrote the manuscript. All authors approved of the final version.

Corresponding author

Correspondence to Philip S. Crosier.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Supplementary information

Supplementary Information

Supplementary Figures 1-16 and Supplementary Methods (PDF 1463 kb)

Rights and permissions

Reprints and permissions

About this article

Cite this article

Hall, C., Boyle, R., Sun, X. et al. Epidermal cells help coordinate leukocyte migration during inflammation through fatty acid-fuelled matrix metalloproteinase production. Nat Commun 5, 3880 (2014). https://doi.org/10.1038/ncomms4880

Download citation

Received: 06 November 2013
Accepted: 11 April 2014
Published: 23 May 2014
DOI: https://doi.org/10.1038/ncomms4880

This article is cited by

Itaconate inhibits TET DNA dioxygenases to dampen inflammatory responses
- Lei-Lei Chen
- Carmen Morcelle
- Dan Ye
Nature Cell Biology (2022)
Role of IRG1 in Regulating Pro-inflammatory and Pro-labor Mediators in Human Myometrium
- Ratana Lim
- Martha Lappas
Reproductive Sciences (2020)
The innate immune cell response to bacterial infection in larval zebrafish is light-regulated
- Lucia Y. Du
- Hannah Darroch
- Christopher J. Hall
Scientific Reports (2017)
Stearoyl-CoA desaturase-1 mediated cell apoptosis in colorectal cancer by promoting ceramide synthesis
- Ling Chen
- Jie Ren
- Yan Qiu
Scientific Reports (2016)
Mitochondrial metabolism, reactive oxygen species, and macrophage function-fishing for insights
- Christopher J. Hall
- Leslie E. Sanderson
- Philip S. Crosier
Journal of Molecular Medicine (2014)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.