Abstract
BACKGROUND:
Iron catalyzes the generation of reactive oxygen species (ROS) as part of the innate antimicrobial defense. During sepsis, the dysregulated systemic inflammatory response to infection, iron homeostasis becomes disrupted, generating an excess of ROS causing damage to tissues. This can be potentially suppressed using iron chelators that selectively bind iron to prevent its participation in ROS-related inflammatory reactions.
OBJECTIVE:
We hypothesize that administration of DIBI, a novel iron-chelator, attenuates the dysregulated systemic immune response and reduces tissue damage in experimental endotoxemia.
METHODS:
Five groups of animals (n = 5–10) were included in this study: control, untreated endotoxemia, and endotoxemia animals treated with either DIBI-A, MAHMP, or DIBI-B. Intravital microscopy was performed on the intestine of anesthesized mice to observe leukocyte endothelial interactions and evaluate the intestinal microcirculation.
RESULTS:
Treatment of endotoxemic mice with DIBI-B reduced the number of adhering leukocytes in submucosal collecting (V1) venules by 68%. DIBI-B, MAHMP, and DIBI-A were able to restore functional capillary density (FCD) in the intestinal muscle layer by 74%, 44%, and 11%, respectively.
CONCLUSIONS:
DIBI-B reduces leukocyte recruitment and improves FCD in experimental endotoxemia, outperforming other chelators tested. These findings suggest a potential role for DIBI-B as a candidate drug for sepsis treatment.
Keywords
Introduction
In infection, iron utilization is a double-edged sword: on one hand, generation of reactive oxygen species (ROS) as part of the innate immune response is iron-dependent. On the other hand, the host attempts to restrict circulating iron levels to reduce the growth of invading microbes [1]. However, in cases of systemic inflammation, such as sepsis, hyper-activation of the immune system becomes apparent [2, 3]. There is evidence to suggest that during this dysregulated, hyper-inflammatory response, iron catalyzes an excessive production of ROS, which results in cellular damage of multiple organs and tissues [4, 5]. Under physiological conditions, free iron is bound to transferrin, which keeps iron soluble and non-toxic. In sepsis, these mechanisms becomes saturated and iron scavenging agents (chelators) may be able to provide compartmentalized sinks for unbound iron to keep it from participating in ROS-related inflammatory reactions and prevent its bioavailability to pathogens [6, 7].
There are a variety of chelating agents available to bind excess iron, and they can vary significantly in their ability to bind iron based on their metal ion affinity, cell permeability, and molecular weight. The novel iron chelator, DIBI, belongs to the synthetic hydroxypyridinone-class of chelators. Unlike classical iron chelators, such as desferrioxamine, representing bacterial derived siderophores to bind iron, hydroxypyridinones can be chemically modified to selectively target iron and conceal it from bacteria, while remaining non-toxic and biocompatible [8, 9]. The goal of our study was to evaluate leukocyte activation and tissue damage in experimental endotoxemia as a model of systemic inflammation in sepsis using this new generation of synthetic iron-chelating polymers.
Materials and methods
Animals
All experimental procedures were performed following the guidelines and standards outlined by the Canadian Council on Animal Care and were approved by the University Committee on Laboratory Animals at Dalhousie University. Six to eight-week-old male C57BL/6 mice (20–30 g body weight) were used in experiments and housed in the Carleton Animal Care Facility, Faculty of Medicine, Dalhousie University, Halifax, NS, Canada.
Experimental groups
Five groups of animals (n = 10 / group) were established for the study: control animals (CON), untreated endotoxemia induced by 5 mg/kg lipopolysaccharide (LPS), endotoxemia animals treated with either 10 mg/kg DIBI-A (LPS+DIBI-A), 40 mg/kg MAHMP (LPS+MAHMP), or 10 mg/kg DIBI-B (LPS+DIBI-B). Pilot experiments did not show any impact of acute iron chelation in healthy animals, therefore treatment only groups were not required. Iron chelators were provided by Chelation Partners Inc., Halifax, NS, Canada and administered 15 min following LPS bolus administration.
Anesthesia and surgery
Animals were anesthetized by intraperitoneal (i.p.) injection of sodium pentobarbital (90 mg/kg). Depth of anesthesia was assessed by animal’s response to toe pinch every 15 minutes during the procedure and supplemental doses of sodium pentobarbital were delivered IP when needed (9 mg/kg).
Prior to surgery, each animal’s neck and abdomen were shaved and disinfected with 70% ethyl alcohol. Cannulation of the right jugular vein was performed using a scalpel for skin incision, followed by blunt dissection until vein was completely clear of all surrounding tissue. The vein was tied off, and using micro scissors, a small incision was made in the vein for the catheter to be placed. Once in the vein, the catheter was secured in place using string and the wound was covered. During procedure, animal’s temperatures were measured using a rectal probe and maintained at 37±0.5°C with a heating pad.
Intravital microscopy
Intravital microscopy (IVM) was performed 120 min following LPS challenge. Thirty minutes prior to IVM, fluorescence dyes were administered intravenously (1.5 ml/kg 0.05% rhodamine 6G; 1 ml/kg 5% fluorescein isothiocyanate albumin; Sigma-Aldrich, Oakville, ON, Canada). IVM of the terminal ileum was performed as described previously [7]. Briefly, immediately after the fluorescent dyes were administered, the abdomen was opened and a section of terminal ileum was placed on a customized stage [10]. The hanging drop method [11, 12] was used to maintain physiological conditions. Leukocyte recruitment (numbers of rolling/adhering cells) was observed in collecting (V1) and postcapillary (V3) venules of the intestinal submucosa. Functional capillary density (FCD) was measured in the intestinal muscle layers and mucosal villi. FCD was quantified by measuring the lengths of perfused capillaries per area. Six 30 second videos were obtained for each section and analyzed offline using ImageJ (NIH, US) in a blinded fashion.
Histology
Samples from the small intestine not affected by IVM were collected from mice immediately after euthanasia. The intestine was separated from the mesentery, and fixated in 10% phosphate-buffered formalin for subsequent histological preparation and hematoxylin and eosin staining. Intestinal tissue was examined for mucosal lesions and scored according to Chiu et al. [13].
Statistical analysis
Results were expressed as mean±SEM. Normal distribution was confirmed by Kolmogorov-Smirnov test and one-way ANOVA was used to analyze data (GraphPad Prism 6.0, GraphPad Software, La Jolla, CA, USA). Differences at P < 0.05 were considered statistically significant.
Results
Adherent leukocytes
Leukocyte adhesion was significantly increased in submucosal V1 and V3 venules two hours after administration of LPS compared to control animals (Fig. 1A and B). The number of firmly adhering leukocytes in the V1 venules increased from 103±23 in control animals to 362±39 cells/mm2 in the untreated LPS group (mean±SEM). Treatment with iron chelators DIBI-A, MAHMP, and DIBI-B reduced the number of adhering leukocytes by 7%, 29%, and 68%, respectively. A 62% reduction of adhering leukocytes can also be seen in V3 venules after treatment with DIBI-B, while no apparent reductions can be seen with treatments of DIBI-A and MAHMP.
Effect of iron chelation on leukocyte adhesion in intestinal submucosal venules. Number of adherent leukocytes (cells/mm2) in V1 venules (A) and V3 venules (B); n = 10 per group. Each bar shows mean±standard error of mean. *P < 0.05 versus control.
The numbers of rolling leukocytes in the untreated LPS group were significantly reduced compared to controls in V1 and V3 venules (Fig. 2A and B). Administration of LPS in V1 venules reduced rolling leukocytes from 192±33 in control animals to 5±1 cells/min. This reduction was not significantly reversed in any of the iron chelator treatment groups.
Effect of iron chelation on leukocyte rolling in intestinal submucosal venules. Number of temporary adhering leukocytes (cells/min) in V1 venules (A) and V3 venules (V3). n = 10 per group. Each bar shows mean±standard error of mean. *P < 0.05 versus control.
Functional capillary density was significantly decreased in the muscular layer and mucosal villi following two hours of endotoxemia in the untreated LPS group compared to control animals (Fig. 3A and B). In endotoxemic animals treated with DIBI B, the FCD increased from 54±4 (untreated LPS animals) to 94±14 cm/cm2 (mean±SEM) in the muscular layer - a 74% improvement. FCD was also improved by 11% and 44% with treatment of DIBI A and MAHMP, respectively.
Functional capillary density within the intestinal muscle layer (A) and the intestinal mucosa (B). Calculated as total length of perfused capillaries within a rectangle field (cm/cm2). n = 10 mice per group. Each bar shows mean±standard error of mean. *P < 0.05 versus control.
In the mucosal villi, FCD was improved to the greatest extent in animals treated with MAHMP, which had a 145% improvement compared to untreated LPS animals.
Using the grading method introduced by Chiu et al. (12), histological samples were evaluated and assigned scores (Fig. 4). The highest degree of mucosal damage was observed in the untreated LPS group (score of 2; Fig. 4C). Administration of iron chelator treatments in endotoxemic animals showed reductions in mucosal damage, with the greatest reduction of damage seen in MAHMP and DIBI-B treatment groups (grade 0; Fig. 4A) and minor improvements with DIBI-A (grade 1; Fig. 4B).
Histopathology of intestinal mucosa. (A) Grade 0: normal mucosal villi (control, DIBI-B, MAHMP). (B) Grade 1: development of subepithelial Gruenhagen’s spaces near the apex of the villi (DIBI-A animal). (C) Grade 2: expansion of subepithelial spaces near the apex of the villi (LPS animal). Scores executed per Chiu et al. (12).
Administration of novel iron chelators resulted in reductions of leukocyte activation within the intestinal microcirculation and improved histology in experimental sepsis induced by LPS. The impact of DIBI-B on leukocyte activation was significantly more pronounced than with treatment of MAHMP or DIBI-A. Functional capillary density was improved in the intestinal microvasculature in experimental endotoxemia with administration of all iron chelators tested. In the muscular layer, DIBI-B was most effective, while MAHMP was the most effective treatment in the mucosal villi.
During endotoxemia we observed a massive increase in the number of adhering leukocytes in untreated animals. This was mirrored by a decrease in the number of rolling leukocytes. We measured a significant decrease in the number of adhering leukocytes in V1 and V3 venules in endotoxemic animals with DIBI-B treatment, however, these results were not mirrored by an increase of rolling leukocytes. Ingress of leukocytes to the site of inflammation requires multiple steps and involves the activation of adhesion molecules on both leukocytes and endothelial cells. Selectins and integrins are important adhesion molecules involved in the initial steps of leukocyte recruitment [14, 15]. Pro-inflammatory signals, including the presence of oxygen free radicals induce the expression of adhesion molecules, which triggers leukocyte rolling and subsequent adhesion [16, 17]. Therefore, reducing the formation of ROS by iron chelation may reduce leukocyte recruitment.
A recent study published by our lab studied the effects of DIBI-A on various measurements of inflammation with and without antibiotic co-treatment using a model of colon ascendens stent peritonitis (CASP) induced poly-bacterial sepsis [7]. This study revealed a significant reduction in leukocyte recruitment by DIBI-A with and without antibiotic co-treatment. An evident improvement of FCD was also shown in this study with DIBI-A restoring the FCD in CASP-induced sepsis to levels observed in sham control groups. Our new results using the endotoxemia model are in agreement with this previous study suggesting that the effects in bacterial sepsis are at least in part related to anti-inflammatory mechanisms rather than antibiotic effects.
Capillary perfusion within the microcirculation is a crucial function during instances of systemic inflammation such as sepsis, as poor perfusion can lead to tissue hypoxia, organ failure and death [18–20]. FCD therefore represents a valuable biomarker of the severity of inflammation. During acute inflammation there are two main changes in the microvasculature that characterize the degree of the inflammatory response and are responsible for a decrease in FCD; an increase in leukocyte endothelial cell interactions, and increased permeability of the microvasculature due to endothelial damage, resulting in edema and compression of the microcirculation [21]. In our model of endotoxemia we were able to confirm increased leukocyte-endothelial interactions in the microvasculature. Treatment of endotoxemic mice with DIBI-B and MAHMP restored perfusion in the microvasculature, while correspondingly reducing the degree of leukocyte recruitment. The cascade of signaling pathways caused by leukocyte-endothelial interactions and subsequent leukocyte adhesion are necessary to ultimately alter the microvascular barrier function, thus enabling leukocyte transmigration to sites of inflammation [22, 23]. Our results are in agreement with this, depicting that reducing the frequency of leukocyte-endothelial interactions also increases microvascular functional perfusion.
Due to the short duration of our endotoxemia model, the histological changes during the observation time were minor. The greatest measure of tissue damage was seen in endotoxemic mice without treatment. Administration of DIBI-B or MAHMP prevented tissue damage during the experimental period to control levels while administration of DIBI-A reduced some damage caused by endotoxins but not to the degree that other treatments did. These observations agree with our other measurements of immune activation where DIBI-B and MAHMP were more effective at attenuating the immune response than DIBI-A.
DIBI is a member of the hydroxypyridinones class of iron chelators, containing a six-membered aromatic ring, with a hydroxyl and ketone functionality. The positions of the ketone and hydroxyl functional groups can be varied to tailor the agent’s iron affinity, hydrophilicity, metabolic stability and functionality to develop the most effective chelators [24]. The iron chelators used in this experiment included DIBI-A (DIBI-029A, original test substance), DIBI-B (DIBI-R12, newest lead substance), and MAHMP, the chelating monomer for DIBI-A and DIBI-B. MAHMP by itself may be small enough to be cell membrane permeable, and therefore capable of chelating intracellular iron. As DIBI is made up of 9 MAHMP residues on a polymer backbone, it is likely too large to work intracellularly and can only bind extracellular iron, which might suggest greater efficacy of MAHMP. However, excessive cellular uptake of iron scavenging agents can result in toxicity [25] making MAHMP a potential problematic candidate for clinical use. As MAHMP is a monomer of DIBI, a higher dose was used that was comparable to DIBI.
Chelators currently being used clinically, such as deferoxamine, are made of siderophores derived from microorganisms as their lead chelating compounds [26]. Siderophores are high-affinity, iron-chelating ligands naturally excreted my microorganisms to compete with the body’s natural iron reserve [27]. The main limitation of using siderophores as iron chelators is that since these agents are derived from bacteria, pathogenic bacteria can synthesize proteins to access iron bound in exogenous siderophores (xenosiderophores) to stimulate their own growth [28]. There are numerous accounts of deferoxamine-treated thalassemic patients developing sudden onset of Yersinia enterocolitica septicemia, likely due to the virulence enhancing effect of deferoxamine [29]. This limitation can be overcome by using synthetic iron chelators, as shown in a study by Leisic et al., which compared the potential of deferiprone (similar chemistry to MAHMP) and deferoxamine to cause the spread of Y. enterocoliticia. It was demonstrated that deferiprone does not have the same virulence-enhancing effect observed with deferoxamine.
DIBI has been assessed against classical iron chelators in numerous other studies. A study by Coombs et al. investigated the effects of iron chelation on the growth of human and murine mammary carcinomas and fibrosarcomas. In this study DIBI’s iron chelating abilities were tested against deferiprone, deferoxamine, and MAHMP. It was found that DIBI had the greatest iron chelating efficacy, inhibiting cell growth by DIBI-mediated iron withdrawal [30]. Holbein and Mira de Orduna [31] also compared DIBI to a variety of chelating agents by studying the effect of iron chelation on microbial growth of well-known opportunistic pathogens Candida albicans and Candida vini. The results of this study showed that DIBI could provide almost complete inhibition of the growth of C. albicans and C. vini over a 4-day incubation period, while deferrioxamine could not inhibit the growth of either pathogen and deferiprone only slightly retarded the growth of C. albicans [31]. DIBI’s inhibitory effects were also shown to be reversible by addition of iron, demonstrating that DIBI is iron-specific and a more potent iron scavenger than other clinically available chelators.
There are numerous afflictions that could benefit from iron chelation therapy. Some of which include sepsis, where iron chelators can provide compartmentalized sinks for unbound iron, reducing its participation in ROS and preventing its bioavailability to pathogens [6]. Iron chelation therapy may also be useful in treating infections with low susceptibility/resistance to antibiotics [32, 33]. The development of new iron chelators is highly relevant as iron chelators currently approved for clinical use have adverse side effects, short half-life, high cost [34], and may provide microbes with an iron source by means of siderophores [27]. We show here that DIBI-B reduces leukocyte recruitment, improves FCD, and prevents mucosal damage in experimental endotoxemia, outperforming other iron chelators tested. DIBI-B’s ability to selectively bind iron while remaining non-toxic and biocompatible makes it a candidate drug for further investigation as an iron chelator with clinical potential.