Modeling of adaptive immunity uncovers disease tolerance mechanisms

When an organism is challenged with a pathogen a cascade of events unfolds. The innate immune system rapidly mounts a preliminary nonspecific defense, while the acquired immune system slowly develops microbe-killing specialists. These responses cause inflammation, and along with the pathogen cause direct and indirect tissue damage, which anti-inflammatory mediators seek to temper. This interplay of systems is credited for maintaining homeostasis but may produce unexpected results such as disease tolerance. Tolerance is characterized by the persistence of pathogen and damage mitigation, where the relevant mechanisms are poorly understood. In this work we develop an ordinary differential equations model of the immune response to infection in order to identify key components in tolerance. Bifurcation analysis uncovers health, immune- and pathogen-mediated death clinical outcomes dependent on pathogen growth rate. We demonstrate that decreasing the inflammatory response to damage and increasing the strength of the immune system gives rise to a region in which limit cycles, or periodic solutions, are the only biological trajectories. We then describe areas of parameter space corresponding to disease tolerance by varying immune cell decay, pathogen removal, and lymphocyte proliferation rates.

Therapeutic efficacy of antimalarial drugs targeting DosRS signaling in Mycobacterium abscessus

A search for alternative Mycobacterium abscessus treatments led to our interest in the two-component regulator DosRS, which, in Mycobacterium tuberculosis, is required for the bacterium to establish a state of nonreplicating, drug-tolerant persistence in response to a variety of host stresses. We show here that the genetic disruption of dosRS impairs the adaptation of M. abscessus to hypoxia, resulting in decreased bacterial survival after oxygen depletion, reduced tolerance to a number of antibiotics in vitro and in vivo, and the inhibition of biofilm formation. We determined that three antimalarial drugs or drug candidates, artemisinin, OZ277, and OZ439, can target DosS-mediated hypoxic signaling in M. abscessus and recapitulate the phenotypic effects of genetically disrupting dosS. OZ439 displayed bactericidal activity comparable to standard-of-care antibiotics in chronically infected mice, in addition to potentiating the activity of antibiotics used in combination. The identification of antimalarial drugs as potent inhibitors and adjunct inhibitors of M. abscessus in vivo offers repurposing opportunities that could have an immediate impact in the clinic.

Characterization of Immune Cells From the Lungs of Patients With Chronic Non-Tuberculous Mycobacteria or Pseudomonas aeruginosa Infection

Little is known of the lung cellular immunophenotypes in patients with non-tuberculous mycobacterial lung disease (NTM-LD). Flow-cytometric analyses for the major myeloid and lymphoid cell subsets were performed in less- and more-diseased areas of surgically resected lungs from six patients with NTM-LD and two with Pseudomonas aeruginosa lung disease (PsA-LD). Lymphocytes, comprised mainly of NK cells, CD4⁺ and CD8⁺ T cells, and B cells, accounted for ~60% of all leukocytes, with greater prevalence of T and B cells in more-diseased areas. In contrast, fewer neutrophils were found with decreased number in more-diseased areas. Compared to NTM-LD, lung tissues from patients with PsA-LD demonstrated relatively lower numbers of T and B lymphocytes but similar numbers of NK cells. While this study demonstrated a large influx of lymphocytes into the lungs of patients with chronic NTM-LD, further analyses of their phenotypes are necessary to determine the significance of these findings.

Mycobacterium avium Infection in a C3HeB/FeJ Mouse Model

Infections caused by Mycobacterium avium complex (MAC) species are increasing worldwide, resulting in a serious public health problem. Patients with MAC lung disease face an arduous journey of a prolonged multidrug regimen that is often poorly tolerated and associated with relatively poor outcome. Identification of new animal models that demonstrate a similar pulmonary pathology as humans infected with MAC has the potential to significantly advance our understanding of nontuberculosis mycobacteria (NTM) pathogenesis as well as provide a tractable model for screening candidate compounds for therapy. One new mouse model is the C3HeB/FeJ which is similar to MAC patients in that these mice can form foci of necrosis in granulomas. In this study, we evaluated the ability of C3HeB/FeJ mice exposure to an aerosol infection of a rough strain of MAC 2285 to produce a progressive infection resulting in small necrotic foci during granuloma formation. C3HeB/FeJ mice were infected with MAC and demonstrated a progressive lung infection resulting in an increase in bacterial burden peaking around day 40, developed micronecrosis in granulomas and was associated with increased influx of CD4⁺ Th1, Th17, and Treg lymphocytes into the lungs. However, during chronic infection around day 50, the bacterial burden plateaued and was associated with the reduced influx of CD4⁺ Th1, Th17 cells, and increased numbers of Treg lymphocytes and necrotic foci during granuloma formation. These results suggest the C3HeB/FeJ MAC infection mouse model will be an important model to evaluate immune pathogenesis and compound efficacy.

Microhemorrhage is an early event in the pulmonary fibrotic disease of PECAM-1 deficient FVB/n mice

Platelet Endothelial Cell Adhesion Molecule 1 (PECAM-1) deficient mice in the FVB/n strain exhibit fatal chronic pulmonary fibrotic disease. The illness occurs in the absence of a detectable pro-inflammatory event. PECAM-1 is vital to the stability of vascular permeability, leukocyte extravasation, clotting of platelets, and clearance of apoptotic cells. We show here that the spontaneous development of fibrotic disease in PECAM-1 deficient FVB/n mice is characterized by early loss of vascular integrity in pulmonary capillaries, resulting in spontaneous microbleeds. Hemosiderin-positive macrophages were found in interstitial spaces and bronchoalveolar lavage (BAL) fluid in relatively healthy animals. We also observed a gradually increasing presence of hemosiderin-positive macrophages and fibrin deposition in the advanced stages of disease, corresponding to the accumulation of collagen, IL-10 expression, and myofibroblasts expressing alpha smooth muscle actin (SMA). Together with the growing evidence that pulmonary microbleeds and coagulation play an active part in human pulmonary fibrosis, this data further supports our hypothesis that PECAM-1 expression is necessary for vascular barrier function control and regulation of homeostasis specifically, in the pulmonary environment.

Iterative feature removal yields highly discriminative pathways

BACKGROUND

We introduce Iterative Feature Removal (IFR) as an unbiased approach for selecting features with diagnostic capacity from large data sets. The algorithm is based on recently developed tools in machine learning that are driven by sparse feature selection goals. When applied to genomic data, our method is designed to identify genes that can provide deeper insight into complex interactions while remaining directly connected to diagnostic utility. We contrast this approach with the search for a minimal best set of discriminative genes, which can provide only an incomplete picture of the biological complexity.

RESULTS

Microarray data sets typically contain far more features (genes) than samples. For this type of data, we demonstrate that there are many equivalently-predictive subsets of genes. We iteratively train a classifier using features identified via a sparse support vector machine. At each iteration, we remove all the features that were previously selected. We found that we could iterate many times before a sustained drop in accuracy occurs, with each iteration removing approximately 30 genes from consideration. The classification accuracy on test data remains essentially flat even as hundreds of top-genes are removed.Our method identifies sets of genes that are highly predictive, even when comprised of genes that individually are not. Through automated and manual analysis of the selected genes, we demonstrate that the selected features expose relevant pathways that other approaches would have missed.

CONCLUSIONS

Our results challenge the paradigm of using feature selection techniques to design parsimonious classifiers from microarray and similar high-dimensional, small-sample-size data sets. The fact that there are many subsets of genes that work equally well to classify the data provides a strong counter-result to the notion that there is a small number of "top genes" that should be used to build classifiers. In our results, the best classifiers were formed using genes with limited univariate power, thus illustrating that deeper mining of features using multivariate techniques is important.

Infection of central nervous system endothelial cells by cell-associated EHV-1

Infection with equine herpesvirus-1 (EHV-1) causes respiratory disease, late-term abortions and equine herpesvirus myeloencephalitis (EHM). Our understanding of EHM pathogenesis is limited except for the knowledge that EHV-1 infected, circulating peripheral blood mononuclear cells (PBMC) transport virus to the central nervous system vasculature causing endothelial cell infection leading to development of EHM. Our objective was to develop a model of CNS endothelial cell infection using EHV-1 infected, autologous PBMC. PBMCs, carotid artery and brain endothelial cells (EC) from 14 horses were harvested and grown to confluency. PBMC or ConA-stimulated PBMCs (ConA-PBMCs) were infected with EHV-1, and sedimented directly onto EC monolayers ('contact'), or placed in inserts on a porous membrane above the EC monolayer ('no contact'). Cells were cultured in medium with or without EHV-1 virus neutralizing antibody. Viral infection of ECs was detected by cytopathic effect. Both brain and carotid artery ECs became infected when cultured with EHV-1 infected PBMCs or ConA-PBMCs, either in direct contact or no contact: infection was higher in carotid artery than in brain ECs, and when using ConA-PBMCs compared to PBMCs. Virus neutralizing antibody eliminated infection of ECs in the no contact model only. This was consistent with cell-to-cell spread of EHV-1 infection from leucocytes to ECs, demonstrating the importance of this mode of infection in the presence of antibody, and the utility of this model for study of cellular interactions in EHV-1 infection of ECs.

Non-invasive diagnosis of early pulmonary disease in PECAM-deficient mice using infrared pulse oximetry

Pulse oximetry is a common tool for detecting reduced pulmonary function in human interstitial lung diseases. It has not previously been used in a mouse model of interstitial lung disease. Further, platelet endothelial cell adhesion molecule deficient mice rarely show symptoms until disease is advanced. Using blood oxygen saturation, different stages of disease could be identified in a non-invasive manner. These stages could be correlated to pathology. Collagen deposition, using Picrosirius Red, did correlate with blood oxygen saturation. These studies are the first to show the use of an infrared pulse oximetry system to analyze the progression of a fibrotic interstitial lung disease in a mouse model of the human diseases. Further, these studies show that an early alveolar damage/enlargement event precedes the fibrosis in this mouse model, a stage that represents the best targets for disease analysis and prevention. This stage does not have extensive collagen deposition. Most importantly, targeting this earliest stage of disease for therapeutic intervention may lead to novel treatment for human disease.

PECAM-independent thioglycollate peritonitis is associated with a locus on murine chromosome 2

Background

Previous studies have demonstrated that knockout or inhibition of Platelet/Endothelial Cell Adhesion Molecule (PECAM, CD31) in a number of murine strains results in impaired inflammatory responses, but that no such phenotype is seen in the C57BL/6 (B6) murine background.

Methodology/Principal Findings

We have undertaken a quantitative trait locus (QTL) mapping effort between FVB/n (FVB) and B6 mice deficient for PECAM to identify the gene or genes responsible for this unique feature of B6 mice. We have identified a locus on murine chromosome 2 at approximately 35.8 Mb that is strongly associated (LOD score = 9.0) with inflammatory responses in the absence of PECAM.

Conclusions/Significance

These data potentiate further study of the diapedesis machinery, as well as potential identification of new components of this machinery. As such, this study is an important step to better understanding the processes of inflammation.

The murine CD99-related molecule CD99-like 2 (CD99L2) is an adhesion molecule involved in the inflammatory response

CD99, a glycoprotein found on the surfaces of leukocytes and concentrated at the borders of endothelial cells, plays a major role in the migration of leukocytes across endothelial cells into sites of inflammation, and has other roles in thymocyte development. The human and mouse genomes encode only two proteins related to CD99. One of these, XGA, is a red blood cell surface antigen. The function of the other, CD99-like 2 (CD99L2), is not known. We cloned mouse CD99L2 and used CD99L2 isolated from transfected cells to raise specific antibodies. Similar to human CD99, CD99L2 was expressed at the borders between transfected cells as well as on mouse leukocytes and vascular endothelial cells in situ. Transfection of L cell fibroblasts with CD99L2 imparted to them the ability to adhere to each other in a divalent cation-dependent, homophilic manner. Anti-CD99L2 antibody blocked influx of neutrophils and monocytes into a site of inflammation in vivo.

Different susceptibilities of PECAM-deficient mouse strains to spontaneous idiopathic pneumonitis

Platelet Endothelial Cell Adhesion Molecule (PECAM) is an adhesion and signaling molecule used for leukocyte extravasation. We have generated two strains of PECAM-deficient mouse, one in the original C57BL/6 and a second by backcrossing nice generations into the FVB/n strain. The FVB/n strain has reduced responses in models of acute inflammation. We show here that this strain is also susceptible to a chronic pneumonia which leads to pulmonary fibrosis. In contrast, PECAM-deficient C57BL/6 mice do not develop this lung disease and have normal responses in acute models of inflammation. This demonstrates that PECAM-dependent and -independent mechanisms are found in both acute and chronic inflammation. Further, the PECAM-deficient FVB/n strain has many pathologic similarities to the human disease Idiopathic Pulmonary Fibrosis, suggesting that similar molecular mechanisms may play a role in human disease.

Platelet Endothelial Cell Adhesion Molecule Deficiency or Blockade Significantly Reduces Leukocyte Emigration in a Majority of Mouse Strains

PECAM is a molecule used specifically during the diapedesis step when neutrophils and monocytes leave the blood compartment. Anti-PECAM reagents, such as Abs and soluble fusion proteins, block diapedesis both in vivo and in vitro. However, the PECAM knockout mouse in C57BL/6 strain has no serious defects in most models of inflammation. We show in this study that the same PECAM knockout backcrossed into the FVB/n strain clearly has reduced leukocyte emigration in two models of inflammation. Furthermore, we show that anti-PECAM reagents can block leukocyte emigration in several other wild-type strains of mice like FVB/n, SJL, and the outbred strain Swiss Webster. This clearly shows that the C57BL/6 strain is uniquely able to compensate for the loss of PECAM function. Murine models of inflammatory disease that have been studied using C57BL/6 mice should be re-evaluated using FVB/n or other mouse strains to determine whether PECAM plays a role in those models.

Locomotion of monocytes on endothelium is a critical step during extravasation

Monocytes, like all leukocytes, undergo a series of sequential steps during extravasation from blood into tissues: tethering, rolling, adhesion and diapedesis. We have discovered an essential step, which we call locomotion, in which the monocyte moves from a site of firm adhesion to the nearest junction to begin diapedesis. Blocking CD11a-CD18 and CD11b-CD18 on human monocytes or adhesion molecules ICAM-1 and ICAM-2 on endothelial cells prevented the monocytes from reaching junctions. The blocked monocytes spun in circles as if they were unable to direct their movement despite being able to adhere and polarize normally. This step fills a gap in the paradigm of extravasation as a multistep process.

CD99 plays a major role in the migration of monocytes through endothelial junctions

CD99 is a heavily O-glycosylated 32-kD type I transmembrane protein that is expressed on most hematopoietic cells. We show here that CD99 is expressed on endothelial cells and is concentrated at the borders between confluent cells. We found that a monoclonal antibody to CD99, hec2, selectively inhibited diapedesis of monocytes across endothelial cells by >90%. Diapedesis involved the homophilic interaction of CD99 on monocytes with CD99 on endothelial junctions. CD99 functioned distally to the point at which platelet-endothelial cell adhesion molecule 1 (PECAM-1, also known as CD31), another adhesion molecule involved in transmigration, played its critical role. Confocal microscopy showed that anti-PECAM-1 arrested leukocytes on the apical surface of endothelium, whereas blocking CD99 arrested monocytes at a point where they were partially through the junction. Therefore, diapedesis, the forward migration of leukocytes through endothelial junctions, is regulated sequentially by two distinct molecules, PECAM-1 and CD99.

Pertussis toxin treatment in vivo reduces surface expression of the adhesion integrin leukocyte function antigen-1 (LFA-1)

Pertussis toxin treatment in macaques inhibits lymphocyte extravasation from the blood and leads to transient lymphocytosis and leukocytosis. We examined lymphocyte adhesion molecules known to be involved in the extravasation process to find possible mechanisms for the effects of pertussis toxin treatment. The two subunits of LFA-1, CD11a and CD18, showed decreased surface expression on lymphocytes from pertussis toxin treated animals compared to untreated animals. The adhesion molecule CD44, and the alpha subunit of the integrin VLA-4 (CD49d) were not decreased by pertussis toxin treatment. Lower surface expression of CD11a and CD18 was observed on all lymphocyte subsets and was correlated inversely with the extent of lymphocytosis. The magnitude of lymphocytosis after pertussis toxin treatment was higher in SIV-infected macaques than in uninfected animals. However, changes in LFA-1 levels were similar in both groups. These data show that LFA-1 surface levels are affected by pertussis toxin in vivo and this change may account in part, for the ability of pertussis toxin to induce lymphocytosis.

Transgenic Mice Expressing Different Levels of Soluble Platelet/Endothelial Cell Adhesion Molecule-IgG Display Distinct Inflammatory Phenotypes

Platelet/endothelial cell adhesion molecule-1 (PECAM-1, CD31), expressed on the surfaces of leukocytes and concentrated in the junctions between endothelial cells plays an important role in transendothelial migration of neutrophils and monocytes. Soluble recombinant PECAM-IgG injected i.v. into mice blocks acute leukocyte emigration by 80%. To study the role of PECAM in models of chronic inflammation, we generated transgenic mice constitutively expressing soluble full-length murine PECAM as an IgG chimera. Three founder lines expressed this transgene and constitutively secreted murine PECAM-IgG into the plasma where it was maintained at characteristic concentrations for each line. All mice had similar hematologic profiles to wild-type littermates and were healthy when maintained in the standard laboratory animal facility. Both the leukocytes and the endothelium of mice of all transgenic lines expressed the same levels of endogenous PECAM-1 as wild-type littermates. Similarly, there were no detectable differences in the expression of several other common leukocyte and endothelial cell adhesion molecules. Mice that produced moderate (10–20 μg/ml) concentrations of PECAM-IgG demonstrated a severely blunted acute inflammatory response, despite mobilizing appropriate numbers of circulating leukocytes. Surprisingly, mice that constitutively produced high (400–1000 μg/ml) concentrations of PECAM-IgG were unresponsive to its anti-inflammatory effects. This is the first demonstration that a soluble form of a cell adhesion molecule can be stably expressed and retain efficacy in vivo over prolonged periods. This approach is applicable to many other extracellular molecules. However, the plasma concentrations of such constitutively produced inhibitors may greatly influence the resulting phenotype.

Transgenic mice expressing different levels of soluble platelet/endothelial cell adhesion molecule-IgG display distinct inflammatory phenotypes

Platelet/endothelial cell adhesion molecule-1 (PECAM-1, CD31), expressed on the surfaces of leukocytes and concentrated in the junctions between endothelial cells plays an important role in transendothelial migration of neutrophils and monocytes. Soluble recombinant PECAM-IgG injected i.v. into mice blocks acute leukocyte emigration by 80%. To study the role of PECAM in models of chronic inflammation, we generated transgenic mice constitutively expressing soluble full-length murine PECAM as an IgG chimera. Three founder lines expressed this transgene and constitutively secreted murine PECAM-IgG into the plasma where it was maintained at characteristic concentrations for each line. All mice had similar hematologic profiles to wild-type littermates and were healthy when maintained in the standard laboratory animal facility. Both the leukocytes and the endothelium of mice of all transgenic lines expressed the same levels of endogenous PECAM-1 as wild-type littermates. Similarly, there were no detectable differences in the expression of several other common leukocyte and endothelial cell adhesion molecules. Mice that produced moderate (10-20 microg/ml) concentrations of PECAM-IgG demonstrated a severely blunted acute inflammatory response, despite mobilizing appropriate numbers of circulating leukocytes. Surprisingly, mice that constitutively produced high (400-1,000 microg/ml) concentrations of PECAM-IgG were unresponsive to its anti-inflammatory effects. This is the first demonstration that a soluble form of a cell adhesion molecule can be stably expressed and retain efficacy in vivo over prolonged periods. This approach is applicable to many other extracellular molecules. However, the plasma concentrations of such constitutively produced inhibitors may greatly influence the resulting phenotype.

Asymptomatic simian immunodeficiency virus infection decreases blood CD4(+) T cells by accumulating recirculating lymphocytes in the lymphoid tissues

Declining blood CD4(+) T-cell counts mark the progress of simian immunodeficiency virus (SIV) disease in macaques and model the consequences of untreated human immunodeficiency virus infection in humans. However, blood lymphocytes are only a fraction of the recirculating lymphocyte pool, and their numbers are affected by cell synthesis, cell depletion, and distribution among blood and lymphoid tissue compartments. Asymptomatic, SIV-infected macaques maintained constant and nearly normal numbers of recirculating lymphocytes despite the decline in CD4(+) T-cell counts. Substantial depletion was detected only when blood CD4(+) T-cell counts fell below 300/microliter. In asymptomatic animals, changes in CD4(+) T-cell distribution were more important than lymphocyte depletion for controlling the blood cell levels.

Dissemination of lymphocytic choriomeningitis virus from the gastric mucosa requires G protein-coupled signaling

The gastric mucosa is an important portal of entry for lymphocytic choriomeningitis virus (LCMV) infections. Within hours after intragastric (i.g.) inoculation, virus appears in the gastric epithelia, then in the mesenteric lymph nodes and spleen, and then in the liver and brain. By 72 h i.g.-inoculated virus is widely disseminated and equivalent to intravenous (i.v.) infection (S. K. Rai, B. K. Micales, M. S. Wu, D. S. Cheung, T. D. Pugh, G. E. Lyons, and M. S. Salvato. Am. J. Pathol. 151:633-639, 1997). Pretreatment of mice with a G protein inhibitor, pertussis toxin (PTx), delays LCMV dissemination after i.g., but not after i.v., inoculation. Delayed infection was confirmed by plaque assays, by reverse transcription-PCR, and by in situ hybridization. The differential PTx effect on i.v. and i.g. infections indicates that dissemination from the gastric mucosa requires signals transduced through heterotrimeric G protein complexes. PTx has no direct effect on LCMV replication, but it modulates integrin expression in part by blocking chemokine signals. LCMV infection of macrophages up-regulates CD11a, and PTx treatment counteracts this. PTx may prevent early LCMV dissemination by inhibiting the G protein-coupled chemotactic response of macrophages infected during the initial exposure, thus blocking systemic virus spread.