HSDM Theses and Dissertations

Abstract: Background: alphav (av) integrin is the most common integrin alpha subunit, it is expressed in different cells and involved in many cellular process including adhesion, migration, survival and growth factor signaling. alphav on immune cells has a role in promoting T regulatory (Treg) cells production and differentiation of T helper 17 (Th17) through activating TGF-beta as shown in conditional knockout of alphav in immune cells. Osteopontin (OPN) is a ligand for alphav and involved in modulation of inflammation through its interaction with alphav in immune cells. OPN has been shown to have a protective role in endodontic infection, so we predict alphav to have similar role. The aim of the study is to establish the role of αv integrin in the immune response to endodontic polymicrobial infection, using an established mouse model of endodontic infection Methods: alphavfl/fl mice were mated with mice expressing Cre under the control of the Tie2 promoter, which is expressed in all cells of the hematopoietic lineage as well as in endothelial cells. These mice were subjected to pulp exposure and infection with a mixture of four endodontic pathogens. Bone loss was assessed 0, 7, 14, and 21 days after infection using micro CT. Inflammation and neutrophils enumeration in periapical area evaluated using immunohistochemistry, also osteoclast presence in periapical area examined with TRAP staining. Microarray and qPCR used to investigate cellular and cytokines activity changes. Blood plasma collected from mice during sacrificing was evaluated using ELISA for level of IgG’s and antibody activity against the specific bacteria used for the infection. Results: Lack of alphav expression was confirmed on neutrophils by FACS. Twenty-one days after pulp exposure and infection with oral pathogens, 6 of 16 av-tie2cre KO mice developed swelling and abscesses in the mandible, while no abscesses were observed in control mice. However, the periapical bone resorption was similar in the two groups. Strikingly, ectopic bone formation on the buccal and lingual aspects of the mandibular surface was observed only in the av-tie2cre animals, suggesting a severe disruption of bone turnover caused by the inflammatory process in these animals. Neutrophils showed decrease in number in periapical area and microarray and the observe disseminating infection suggested a B cells defect either in function or migration. Conclusions: Together, these results demonstrate a critical role for the alphav integrin in the suppression of inflammation during the process of endodontic infection.

Introduction: Periradicular periodontitis is an inflammatory disease caused by perpetual microbial infection in the root canal system which, if left untreated, results in the destruction of tooth supporting tissues. Tenacious infection of virulent bacteria can lead to continuous activation of host immune response which results in excess production of proinflammatory mediators, including the bone resorption factor RANKL. To date, attempts to minimize the destruction of tooth supporting tissues have included laser therapy and phototherapy, but these approaches have been confined to treating and monitoring the exogenous etiology, i.e., bacterial infection. Therefore, to improve the diagnosis and treatment of periradicular periodontitis, new technologies are needed to control host cell biological responses and evaluate the level of inflammation in the root canal system and periradicular area. Objectives: Intracellular generated adenosine triphosphate (ATP) and its metabolite, adenosine (ADO), act as major energy currencies in the cell system. Only recently has it been reported that extracellular ATP (eATP) released from activated neutrophils elicits proinflammatory responses, while ADO, which is generated via enzymatic conversion of ATP by cell surface-expressed ecto-5′-nucleotidase (CD73) and Ectonucleoside triphosphate diphosphohydrolase-1 (NTPDase1/CD39), functions as anti-inflammatory mediator. Accordingly, this study investigated the effects of CD73 and ADO on osteoimmunological bone resorption in periradicular periodontitis. Methods: The possible anti-bone loss role of ADO was examined in vitro by reacting RANKL-stimulated RAW264.7 osteoclast precursors with ADO receptor agonist 5′-(N-Ethylcarboxamido) adenosine (NECA). More specifically, the effects of NECA on RANKL-induced osteoclastogenesis were evaluated by monitoring TRAP+ multinucleated cell number and pit formation. Since activated T cells are the major cellular source of RANKL in periradicular periodontitis, the effects of NECA on proliferation and sRANKL production by TCR/CD28-stimulated T cells were examined in vitro. The therapeutic efficacy of NECA on the pathogenic bone loss induced in periradicular periodontitis was evaluated in a pulp exposure-induced periradicular periodontitis model using C57BL/6 wild-type (WT) mice. On the other hand, the impact of ADO loss-of-function was compared between CD73-/-Foxp3-GFP+/+ mice and WT mice, both with experimentally induced periradicular lesion. Periradicular tissues were isolated from sacrificed mice, and the mRNAs of inflammation-associated and bone remodeling-related genes, amount of ATP and size of periradicular lesions were monitored up to 3 weeks using qPCR and µCT. Expressions of CD73 and other proteins in the periradicular tissues were examined by immunohistochemistry and immunofluorescence staining. Results: As determined from the in vitro experiments, NECA significantly suppressed not only RANKL-mediated osteoclastogenesis from RAW264.7 cells (P<0.05), but also significantly downregulated proliferation and RANKL production from TCR/CD28-activated T cells (P<0.05), suggesting that the elicited signal could downregulate RANKL-induced osteoclastogenesis. As determined from the in vivo experiments, the amount of ATP and expression of CD73 mRNA were significantly increased in the mouse periradicular lesion in a time-dependent manner. NECA-treated wild-type (WT) mice demonstrated significantly elevated mRNAs for CD39, CD73, alkaline phosphatase, and RUNX2, while showing decreased expressions of RANKL and IL1-β mRNAs, accompanied by diminished size of periradicular lesion, compared to nondrug-treated control WT mice experimentally induced with periradicular lesion. Finally, the amount of ATP and mRNAs for RANKL, IL1-β, IL-6, and TNF-α, as well as the size of periradicular lesion, were significantly larger (P<0.05) in CD73-/-Foxp3-GFP+/+ compared to control WT-mice. Conclusion: By its catalytic action to convert ATP to ADO, increased CD73, by its catalytic action to convert ATP to ADO, in the periradicular lesion may play a role in downregulating inflammatory bone loss, suggesting the CD73/ADO axis as a novel therapeutic target for periradicular periodontitis, leading to the development of a therapeutic regime which, beyond controlling bacterial infection, can control host cell biological responses and, hence, the level of inflammation in the affected root apex.

Background: A periapical lesion is an infection-induced inflammation around the apex of the tooth caused by bacterial invasion and ultimately results in the destruction of bone in the area. To date, the relationship between pathogens and the host immune system in the development of periapical lesions have been extensively investigated. However, a detailed understanding of the host factors that determine successful periapical wound healing is currently lacking. The long-term goal of the present study is to identify the molecular network that regulates wound healing and tissue regeneration in dentoalveolar infections in order to develop rational therapeutic approaches. Although a rodent model of periapical lesion was well established in Forsyth, adequate models for periapical wound healing are currently unavailable. Since the nature of apical periodontitis is infection-induced inflammation, the aim of this study was to systematically examine the effect of azithromycin treatment compared to ampicillin as a wound-healing treatment approach in mouse periapical lesions. Azithromycin is a macrolide antibiotic that specifically accumulates in macrophages exhibiting an anti-inflammatory effect via alteration of macrophage phenotype from pro-inflammatory M1 to anti-inflammatory M2 macrophages.

Material and Methods: Dental pulps of mandibular first molars in C57BL/6J mice (6 weeks of age) were surgically exposed and inoculated with a cocktail of common human endodontic pathogens, which are sensitive to employed antibiotics. Mice received daily doses of azithromycin (8 mg/kg), ampicillin (20 mg/kg) or phosphate-buffered saline (PBS) from day 10 to 20 post infection. No root canal treatments were provided. Mandibles were isolated on days 10 (disease baseline) and 21. Right hemi mandibles were subjected to micro-computed tomography (μCT) to determine the extent of periapical bone loss and subsequent histological and immunohistochemistry analysis. Left hemi mandibles were used for RNA extraction for gene expression analysis through quantitative RT-PCR. Moreover, primary peritoneal macrophages stimulated with LPS and treated with either azithromycin or ampicillin were evaluated for pro-inflammatory cytokines expression (using ELISA) and nitric oxide production (iNOS assay). The effect of azithromycin in the level of NF-kB gene expression was assessed through luciferase reporter assay using Luciferase Stable RAW 264.7 cells.

Results: The μCT results indicated that azithromycin treatment suppressed the extent of periapical bone loss compared to PBS treatment with a significant difference (p<0.05) in the average size of periapical lesions. In addition, the extent of bone loss on Day 21 in azithromycin group was significantly “recovered” compared with the baseline disease group (p<0.05). In histological observations, azithromycin-treated animals showed less neutrophil infiltration and a polarization of M2 macrophages in periapical lesions, while PBS treatment resulted in moderate neutrophil infiltration and an M1/M2-mixed macrophage profile. mRNA expression of pro-inflammatory cytokine IL-6 and chemokine CXCL2, as well as colony stimulating factor 2 (CSF2) had a significant down regulation in the  azithromycin group when compared to baseline disease (p<0.05). Up regulation of IL-4-mRNA was also observed. In vitro experiments confirmed the down regulation of IL-1α and IL-1β (p>0.05) and nitric oxide production (p<0.05) in LPS stimulated peritoneal macrophages treated with azithromycin. Finally, the luciferase activity was decreased when cells were stimulated with azithromycin 50uM. 

Conclusions: These results taken together suggest that the suppression of periapical bone loss by azithromycin appears to be dependent on its immunomodulatory properties. This report is the first finding on pro-resolving M2 macrophage polarization during periapical wound healing. This research can shed light to the development of a mouse model of periapical wound healing. In addition, this dual role of azithromycin, microbicidal and anti-inflammatory effects, could pose as a great adjuvant on the root canal therapy being highly translational.

Unresolved inflammation is a key factor linking metabolic dysregulation and the immune system in type 2 diabetes. Successful regulation of acute inflammation requires biosynthesis of specialized pro-resolving lipid mediators, such as resolvin E1 (RvE1), and activation of cognate g-protein coupled receptors (GPCR). RvE1 binds to BLT-1 on neutrophils and ERV-1/ChemR23 on monocyte/macrophages. The mechanism by which innate immune cells in metabolic diseases fail to respond to endogenous pro-resolution signals is not well understood. The goal of the studies presented was to investigate the expression and function of ERV-1 receptor on neutrophils from type 2 diabetes subjects. Results demonstrate that neutrophils from type 2 diabetic subjects expressed significantly increased ERV-1 receptor on their cell surface that was functional, as determined by phosphorylation of the signalling protein, ribosomal S6 (rS6). Healthy neutrophils express functional BLT-1 and low levels of minimally functional ERV-1. Stimulation with TNFα or LPS increased expression of ERV1 on neutrophils of both healthy and diabetic subjects. Increased expression of ERV1 was reversed by pre- and post- treatment with RvE1. TNFα or LPS stimulation increased phospho-rS6 significantly more in type 2 diabetic neutrophils than healthy subjects. Receptor antagonism experiments revealed that the phospho-rS6 increase was mediated by BLT-1 in healthy subject neutrophils and ERV1 in diabetic subject neutrophils. Diabetic neutrophil ERV-1 over expression was reversed by RvE1 treatment in a dose response manner (1-10nM). Analysis of the bioactive lipid metabolome revealed higher expression of inflammatory mediators in type 2 diabetes serum than healthy control with slightly reduced levels of RvE1 in type 2 diabetics (by ELISA). Neutrophil phagocytosis of pathogenic bacteria and inert bioparticles was impaired in type 2 diabetes. RvE1, signalling through ERV-1 but not BLT-1, partially rescued deficient phagocytosis by type 2 diabetic neutrophils with a clear shift of the dose response requiring increased RvE1. The dose of RvE1 required to activate resolution signals in type 2 diabetic neutrophils is significantly higher than healthy controls. Overall the findings in this dissertation further our understanding of the importance of expression and function of key resolution receptors in health versus disease and dysregulation of inflammation in type 2 diabetes.

Objectives: Graft versus host disease (GVHD), infections and disease relapse continue to be major complications of allogeneic hematopoietic cell transplantation (HCT). B cell abnormalities have been associated with these complications, and the objective of these studies was to identify and quantify abnormalities of B cell recovery after allogeneic HCT. Methods: Flow cytometry was used to analyze peripheral blood from healthy donors (HD) and samples obtained at 1, 2, 3, 6, 9, 12, 18, 24, 30, and 36 months post HCT. A panel of 8 fluorophore-conjugated monoclonal antibodies specific for B cell surface markers was used to identify levels of total B cells, antigen naïve, antigen experienced, transitional B cells, naïve B cells, IgD memory, pre germinal, post germinal, and plasmablasts. Results: 224 male patients and 175 female patients (total N=399) with a median age of 58 years (range 19-74) were included in this study. 308 patients received reduced intensity and 91 received myeloablative conditioning for allogeneic HCT in the management of various hematological malignancies. The 2-year overall survival was 65% and progression free survival was 55%. The absolute number of total B cells in patients compared to HD remained low for 9 months (118.8 x 106/L vs 204 x 106/L, P < 0.02) and normal levels were reached by month 12. Recovering cells were primarily antigen naïve B cells and the frequency of antigen experienced B cells remained below normal when compared to HD for all time points post HCT (P < 0.0001). Within the antigen naïve compartment, the frequency of transitional B cells remained elevated and naïve B cells remained below normal until normal levels were reached by month 9. Within the antigen experienced compartment, the frequency of pre germinal cells was elevated compared to HD from 6 months (4.8% vs 0.70%; P < 0.0002) post HCT. Plasmablast frequencies were elevated compared to HD from 6 months (7.3% vs 1.1%; P < 0.0009) post HCT. IgD memory cells did not reach normal frequency for most time points. Post germinal cells did not reach normal frequencies from 6 to 12 months post HCT. Conclusion: Disturbances in B cell maturation after HCT persist for prolonged periods despite recovery of normal numbers of circulating B cells. Naïve and transitional B cells that predominate have low affinity BCRs and have been implicated in cGVHD. Reduced numbers of high affinity antigen experienced B cells likely predispose patients to infectious complications post HCT.

Objective: The goal of this project was to determine the impact of local inflammation on changes in the subgingival biofilm composition in ligature-induced periodontitis in rats using the specialized pro-resolving mediator (SPM), resolvin E1 (RvE1). Materials and Methods: The impact of RvE1 on the microbiota of ligature-induced periodontitis was assessed in two separate experiments; treatment of established periodontitis and prevention of ligature-induced periodontitis. In the treatment study, eighteen rats were separated into four groups comprising no ligature, ligature alone (no treatment), ligature with topical RvE1 treatment (ligature+RvE1) and, ligature with topical vehicle treatment (ligature + Vehicle). 3-0 silk ligatures were tied around maxillary second molars bilaterally for three weeks to induce disease. After three weeks, the treatment phase began with the application of RvE1 or vehicle (ethanol) every other day for an additional three weeks. Subgingival plaque samples were collected every four days throughout the experiment. The composition of the subgingival microbiota was initially screened by checkerboard DNA-DNA hybridization using probes on 40 subgingival species. Definitive, unbiased characterization of the subgingival microbiota was accomplished with next-generation sequencing using the Illumina MiSeq® platform. Six rats were sacrificed on Days 1, 21 and 42 and maxillae were dissected to collect samples for gingival RNA extraction, bone morphometric measurements, and histomorphometric analysis. Local tissue gene expression (Cxcl-1, Ptgs2, Nos2) was detected using qRT-PCR. Tissue specimens were prepared for histology and stained with H&E and tartrate resistant acid phosphatase (TRAP). In the prevention study, sixteen rats were separated into four groups (no ligature, ligature + RvE1 (0.1µg/µl), ligature + RvE1 (0.5 µg/µl), ligature + Vehicle). 5-0 silk ligatures were placed around maxillary second molars bilaterally to induce disease. At the time of ligature placement, animals received assigned treatment thrice weekly (M, W, F) for four weeks. Subgingival plaque samples were collected every four days (M and F). Four rats were sacrificed at baseline (Day 1) and the vehicle and two treatment groups (four each) were sacrificed at day 28 and samples processed as described above. The two-group comparisons were assessed by Student’s t-test. The multiple-group comparison was assessed by one-way ANOVA and post hoc tests. Results: In the first study (treatment), topical application of RvE1 significantly reversed the bone loss associated with periodontitis compared to the vehicle. RvE1 application significantly reduced the expression of Cxcl1 and osteoclast density compared to the vehicle application. In the prevention study, RvE1 treatment significantly prevented the bone loss during the disease progression. RvE1 application significantly reduced the expression of Ptgs2, Nos2 compared to the vehicle application. Osteoclast density and inflammatory cell infiltration in the RvE1 groups were significantly lower than these in the Vehicle group. The cell counts of bacterial species gradually increased and the subgingival microbiota shifted during the disease progression. In the treatment study, RvE1 treatment significantly reduced cell counts compared to the vehicle application at the end of treatment phase. The shift of subgingival microbiota was limited by the RvE1 treatment. In the prevention study, the taxonomic composition and diversity of subgingival microbiota was controlled by the RvE1 application. The change of subgingival microbiota appeared to be associated with the state of inflammation in the periodontal environment. Conclusion: Resolvin E1 treatment of existing ligature-induced periodontitis significantly regenerates lost alveolar bone and prevents alveolar bone loss. Resolvin E1 treatment limits microbial shifts and reduces total bacterial load by inhibiting inflammation of local environment in experimental periodontitis.

Objectives The aim of this thesis was to study the biosynthesis of Maresin 1 (MaR1) in human macrophage and to investigate the potential defect of this pathway in patients with localized aggressive periodontitis as well as the role of MaR1 in regulating phagocyte functions and leukocyte-platelet aggregation using LAP as a disease model. Methods Human recombinant enzyme (12-lipoxygenase and soluble epoxide hydrolyse) and synthetic intermediates were used to investigate the maresin pathway by LC/MS/MS-based lipid mediator (LM) metabololipidomics. Human macrophage 12-LOX sequence was crosschecked with platelet-type 12-LOX. Enzyme expression levels in human myeloid mononuclear lineage were determined with flow cytometry. Peripheral blood was collected from LAP patients and age/gender-matched healthy controls (HC). Neutrophils and monocytes were isolated, and macrophages were obtained from monocytes with GM-CSF incubation. LAP and HC macrophage 12-LOX expression levels were monitored by flow cytometry and endogenous maresin metabolite levels were investigated using LM metabololipidomics. Periodontal pathogens including Porphyromonas gingivalis (P. g) and Aggregatibacter actinomycetemcotimans (A. a) were incubated with neutrophils and macrophages to investigate phagocyte functions. Phagocytosis and intracellular anti-microbial reactive oxygen species (ROS) production were determined using fluorescence plate reader. Bacterial killing by macrophages and in whole blood was evaluated by assessing residual colony forming units following incubation with bacteria. Results We demonstrated that human macrophage 12-LOX is identical with platelet-type 12-LOX and is upregulated from monocyte to different mononuclear lineage. 12-LOX converts DHA to 14S-HpDHA, which is a key intermediate in MaR1 biosynthesis. In addition, soluble epoxide hydrolyze (sEH) is responsible for the further production of novel anti-inflammatory and pro-resolving Maresin 2 (MaR2; 13, 14S, di-HDHA). Since 12-LOX is the key-initiating enzyme, investigation begin with macrophages from LAP patients, which have lower 12-LOX expression (~30% lower) and reduced MaR1 (87.8±50 vs. 239.1±32 pg/ 106 cells) levels compared to macrophages from healthy controls (HC). Functionally, LAP macrophages gave impaired phagocytosis (~40% lower) and killing of periodontal pathogens including Porphyromonas gingivalis and Aggregatibacter actinomycetemcomitans. In addition, neutrophils from LAP patients also displayed diminished kinetics (~30% slower) and reduced maximal phagocytosis (~20% lower) of these pathogens. Exogenous MaR1 rescued the impairment with LAP phagocytes enhancing phagocytosis (31-65% increase, 1nM), intracellular anti-microbial reactive oxygen species production (26-71% increase, 1nM) and bacterial killing of these periodontal pathogens in vitro. Additionally, MaR1 retained biological action and enhanced bacterial killing in the whole blood ex vivo from HC and LAP patients (22-37% reduction of bacterial titers, 10nM). MaR1 counter-regulated leukocyte-platelet aggregation against two stimulus (platelet activating factor (PAF) and P. gingivalis) and regulated the expression levels of surface adhesion molecules (CD18 and P-selectin). Patients with LAP had elevated leukocyte-platelet aggregation (40% higher compared to HC) in the whole blood that could be reduced by MaR1 in a dose-dependent manner (40-60% reduction of the excess aggregates, 1-100nM). Conclusions These results indicate that 12-LOX is the key enzyme in maresin biosynthetic pathway. LAP macrophages, had lower 12-LOX expression levels and dysregulated maresin biosynthesis. Reduced phagocyte functions were established with both LAP macrophages and neutrophils; exogenous MaR1 rescued this impairment with LAP patients and also enhanced the functions of HC phagocytes. Additionally, MaR1 enhanced bacterial killing and reduced elevated leukocyte-platelet aggregates in LAP whole blood. Together these results suggest that therapeutics targeting maresin pathway may have clinical application treating LAP and oral diseases associated with infection and uncontrolled inflammation.