Workshop 6: Disease models: Host-Pathogen Interactions

(June 21,2004 - June 25,2004 )

Organizers


Thomas Kepler
Biostatistics and Bioinformatics, Duke University
Denise Kirschner
Microbiology and Immunology, University of Michigan

In this workshop, we will focus on modeling the dynamics of pathogens interacting with the immune system. We will do this in three main areas: viral models, bacterial models, and parasitic models. We will then explore chemotherapy treatment strategies and the generation of drug resistant mutants. Treatment will be discussed on the last day of the week.

Large numbers of cells and organisms, and long time scales, make differential equations the appropriate and primary tool used to study these phenomenon. Delay equations, simulations, and stochastic models will also play a role.

Accepted Speakers

Seema Bajaria
Microbiology and Immunology, University of Michigan
Mary Carrington
Laboratory of Genomic Diversity, National Institutes of Health
Arturo Casadevall
Albert Einstein College of Medicine
John Chan
Albert Einstein University
Stewart Chang
Microbiology and Immunology, University of Michigan
Sandro Cinti
Infectious Diseases, University of Michigan
Miles Cloyd
Medical Branch, University of Texas
Victor Dirita
Microbiology and Immunology, University of Michigan
JoAnne Flynn
Molecular Genetics & Biochemistry, University of Pittsburgh
Zvi Grossman
Department of Physiology and Pharmacology, Tel-Aviv University
Amy Herring
Microbiology and Immunology, University of Michigan
Phil Hodgkin
Immunology, The Walter and Eliza Hall Institute of Medical Research
Garnett Kelsoe
Immunology, Duke University
Thomas Kepler
Biostatistics and Bioinformatics, Duke University
Denise Kirschner
Microbiology and Immunology, University of Michigan
Reinhard Laubenbacher
Virginia Bioinformatics Institute, Virginia Polytechnic Institute and State University
Jun Lu
Biostatistics and Bioinformatics, Duke University
Simeone Marino
Microbiology & Immunology, University of Michigan
Mark Miller
Physiology & Biophysics, University of California, Irvine
Mary O'Riordan
Microbiology and Immunology, University of Michigan
Sergei Pilyugin
Department of Mathematics, University of Florida
Andrew Quong
Biosecurity and NanoSciences, Lawrence Livermore National Laboratory
Todd Reinhart
Infectious Diseases & Microbiology, University of Pittsburgh
Jose Segovia-Juarez
Microbiology and Immunology, University of Michigan
Jaroslav Stark
Department of Mathematics, Imperial College
Robert Stengel
Engineering and Applied Science, Princeton University
Wai-Yuan Tan
Mathematical Sciences, University of Memphis
Georgia Tomaras
AIDS Research Center, Duke University Medical Center
John Tomfohr
Biostatistics and Bioinformatics, Duke University
Jorge Velasco-Hernandez
Department of Mathematics, Instituto Mexicano del Petroleo
John Ward
Department of Mathematical Sciences, Loughborough University
Ping Ye
Biomedical Engineering, Johns Hopkins University
Monday, June 21, 2004
Time Session
09:15 AM
10:00 AM
Mary Carrington - Patterns of KIR/HLA Association in HIV and HCV Infection

Natural killer (NK) cells are a unique group of lymphocytes involved in surveillance and killing of foreign or infected cells through a mechanism involving recognition of HLA molecules by an extremely diverse set of receptors on the NK cell surface. A major group of these receptors are the killer immunoglobin-like receptors (KIRs) that are encoded by genes mapping to chromosome 19q13.4. These receptors regulate inhibition and activation of NK cell responses through recognition of HLA class I molecules on target cells. Given their receptor-ligand relationship, we have hypothesized that KIR may be involved in many of the diseases for which an HLA influence has been identified. In this regard, we performed KIR genotyping in several AIDS cohorts and have shown that one activating KIR allele, KIR3DS1, in combination with the HLA-B alleles encoding its possible HLA ligand, HLA-Bw4-80Ile, results in delayed progression to AIDS, suggesting a synergistic interaction between the two loci. We have also performed KIR genotyping in a cohort of individuals with HCV infection. We show that genes encoding the inhibitory NK cell receptor KIR2DL3 and its HLA-Cw group1 ligand, which transmit relatively weak inhibitory signals, influence resolution of hepatitis C virus (HCV) infection. This effect was observed in Caucasians and African Americans with expected low infectious doses of HCV, but not in those with high-dose exposure, in whom the innate immune response is likely overwhelmed. Data from both of these infectious diseases indicate that net activation of NK cells is beneficial in the anti-viral immune response.

10:30 AM
11:15 AM
Todd Reinhart - Host/Pathogen Interactions at Multiple Levels during HIV-1 Infection and AIDS

The interactions between HIV-1 and the host occur at multiple levels and have generally been studied at all such levels. This spectrum spans from the associations of individual proteins, through cell/cell interactions, on up to the interactions of infected individuals within populations. Increased understanding of such interactions has led to intervention strategies - implemented or proposed - at each level including such diverse strategies as viral reverse transcriptase inhibitors and community outreach programs. Our interests have focused on the effects of HIV-1 infection on host function at the cellular and tissue levels, using infection of nonhuman primates with the related simian immunodeficiency virus (SIV) as a model system. Our ultimate goal is to identify new targets for therapeutic intervention, and we have approached this issue in two major ways. First, we have studied the effects of expression of a pathogenesis-associated viral protein named "Nef" on cell surface protein expression. Nef is not essential for viral replication in vitro, but contributes to high-level viral replication in vivo and the subsequent development of AIDS, although it has no known enzymatic function. In searching for cellular binding partners of Nef, we identified a critical component of the T cell signaling complex, TCR , as a Nef partner. SIV Nef bound TCR on two independent sites and reduced the surface expression levels of the TCR signaling complex, thereby likely inhibiting the ability of an infected cell to exert its function as an immune cell. Second, we have studied the effects of in vivo infection with SIV on the immune environments within lymphoid tissues. We have used gene expression profiling approaches such as cDNA arrays, in situ hybridization, and real-time RT-PCR to identify genes differentially expressed in tissues from rhesus macaques infected with pathogenic SIV. These approaches have revealed a number of alterations to immune environments in lymphoid, lung, and intestinal tissues during the progression of SIV-associated disease. The changes we have identified and which are likely to be important in this disease include: (1) up-regulation of chemokines, which control constitutive and inflammatory cell trafficking; (2) altered tissue compositions of dendritic cells, which are potent antigen presenting cells controlling the nature and strength of immune responses; and (3) up-regulation of members of a group of receptors within the innate immune system, Toll-like receptors, which control rapidly-induced inflammatory responses. These approaches have their respective strengths and limitations, but combined together can provide insight into the roles that these alterations to immune function play in the progression of disease, and thereby identify targets for new therapeutic intervention strategies.


Funding Sources: National Institutes of Health; National Heart, Lung, and Blood Institute.

11:45 AM
12:30 PM
Georgia Tomaras - Noncytolytic CD8+ T cell-Mediated Suppression and HIV-1 Escape

CD8+ T cells play an important role in controlling virus replication in both acute and chronic HIV infection. There is substantial interest in further deciphering the contributions of noncytolytic CD8+ T cells in reducing virus replication as part of the host's protective immune response. It is commonly known that HIV can escape potent immune pressures such as cytotoxic T lymphocytes and neutralizing antibodies. It has recently become evident that virus can escape from noncytolytic suppression illustrating the ability of this antiviral activity to exert significant immune pressure in vivo. The molecules involved in this antiviral response and their precise mechanisms remain elusive. Studies of HIV variants harboring escape mutations are likely to provide new insights into the identities of noncytolytic CD8+ suppression.

02:00 PM
02:30 PM
Leor Weinberger - Stochastic Fluctuations in HIV-1 Tat Transactivation May Lead to Proviral Latency

The human immunodeficiency virus type I (HIV-1) proviral latent reservoir is considered the most significant obstacle facing HIV-1 eradication from the patient. The exact mechanism by which this reservoir is established remains a topic of much research. Bacteriophage l is known to utilize stochastic molecular fluctuations (SMF) in viral protein levels (C1 and Cro) to influence its lifecycle decision between lytic and lysogenic states and recently SMF in yeast transcriptional and translational pathways have been observed to lead to clonal population variability, even in the absence of chromatin remodeling. SMF have yet to be demonstrated or implicated in higher eukaryotes or mammalian systems. Here we present the first evidence that an HIV-1 positive feedback regulatory pathway, implicated in the establishment of proviral latency (the HIV-1 Tat transactivation loop) may utilize such stochastic molecular fluctuations.


Work done in collaboration with Leor S. Weinberger, Adam P. Arkin, and David V. Schaffer

02:30 PM
03:00 PM
Seema Bajaria - CTL Action During HIV-1 is Determined via Interactions with Several Cell Types

During HIV-1 infection, interactions between immune cells and virus yield three distinct disease stages: high viral levels in acute infection, immune control in the chronic stage, and AIDS, when CD4+ T cells fall to extremely low levels. The immune system consists of many players that have key roles during infection. In particular, CD8+ T cells are important for killing of virally infected cells as well as inhibition of cellular infection and viral production. Activated CD8+ T cells, or cytotoxic T cells (CTLs) have unique functions during HIV-1, most of which are thought to be compromised during HIV-1 disease progression. Controversy exists regarding priming of CTLs, and our work attempts to address the dynamics occurring during HIV-1 infection. To explore the influence of CD8+ T cells as determinants in disease progression and issues relating to their priming and activation, we develop a two-compartment ordinary differential equation model describing cellular interactions that occur during HIV-1 infection. We track CD4+ T cells, CD8+ T cells, dendritic cells, infected cells, and virus, each circulating between blood and lymphatic tissues. Using parameter estimates from literature, we simulate commonly observed disease patterns. Our results indicate that CD4+ T cells as well as dendritic cells likely play a significant role in successful activation of CD8+ T cells into CTLs. Model simulations correlate with clinical data confirming a quantitative relationship between CD4+ T cells and CD8+ T-cell effectiveness.

03:30 PM
04:00 PM
Wai-Yuan Tan - Assessment of Treatment Effects on HIV Pathogenesis Under Haart by State Space Models

To monitor the progression of therapy in HIV-infected individuals treated with anti-viral drugs, it is critical to estimate and assess the efficiency of the drugs and to estimate the number of infectious and non-infectious HIV under treatment. In this paper we have developed a method to estimate these parameters and the state variables to assess effects of drugs on HIV pathogenesis. As an illustration, we have applied the method to some clinical and laboratory data of an AIDS patient treated with various anti-viral drugs. For this individual, we have estimated the numbers of infectious HIV virus and non-infectious HIV virus per $ml$ of blood over time, and the rates for measuring effects of drugs. These estimates show that the HAART protocol has effectively controlled the number of infectious HIV virus to below $400/ml$ copies although the total number of HIV copies was very high in some intervals.


Work done in collaboration with Wai-Yuan Tan, Ping Zhang, Xiao-Ping Xiong, and Pat Flynn.

04:00 PM
04:30 PM
Robert Stengel - Optimal Control of Disease Processes

Mathematical modeling of immune system response to pathogenic assault can greatly aid the management of infected individuals, but considerable development must occur before the biology and mathematics can be merged to provide tools for treatment. Given a model for a disease process, it remains to be seen how established methods of control system design can suggest therapeutic protocols that could be applied in a clinical setting.


We examine means for specifying optimal therapeutic protocols given a satisfactory immune system model. For illustration, the presentation is based on three models that represent humoral response to extracellular bacteria, cellular response to the human-immunodeficiency virus (HIV), and control of cancer by anti-tumor viruses. Therapy is based upon both open- and closed-loop optimal control strategies that take into account uncertainty in system models, measurements, and environment. We note that though observability of the dynamic state is a critical issue for closed-loop control, effective control can be provided by incomplete or "noisy" measurements. Immune system models possess classical attributes of dynamic stability or instability; for example, our humoral response model is stable, while the HIV model is not. In either case, effective therapies can be specified, although the HIV infection is never cured, requiring continued treatment to keep the condition in remission.

Tuesday, June 22, 2004
Time Session
09:00 AM
09:45 AM
Sandro Cinti - HIV/AIDS-2004

HIV/AIDS-2004

10:15 AM
11:00 AM
Miles Cloyd - Pathogenic Mechanism of HIV

HIV infection causes an acquired immunodeficiency, principally because it causes depletion of CD4 lymphocytes. The mechanism by which the virus depletes these cells however, is not clearly understood. Many studies demonstrate that uninfected CD4 cells are the cells predominantly depleted. The most plausible mechanism, which is backed by in vivo data, involves the consequences of HIV contact with resting CD4 lymphocytes, which are cells which do not support replication of HIV. HIV binding to and signaling through CD4 and chemokine receptor molecules on resting CD4 lymphocytes and other cell types (which extensively occurs as the rare productively infected cells migrate among other cells within the lymphoid tissues) induces up regulation L-selection and Fas. When these resting HIV-signaled CD4 cells return to the blood, they home very rapidly back to the peripheral lymph nodes and axial bone marrow. Thus, the disappearance of CD4 cells from the blood in HIV-infected individuals appears to be due to them leaving the blood. Approximately one third of these cells that are HIV induced to home to lymph nodes are subsequently induced into apoptosis during the process of trans-endothelial migration, during which they receive secondary signals through various homing receptors. These cells are not making HIV, which explain the observation that CD4 cells not making HIV are the predominate cells dying in the lymph nodes of HIV+ subjects. Mathematical modeling of this process yielded an accurate picture of the rate of depletion of CD4 cells over an eight to ten year period. This is unique mechanism of pathogenesis for a virus, and if correct, leads to the possibility that HIV might not cause depletion of CD4 lymphocytes if it used some other receptor to infect cells.

02:00 PM
02:30 PM
Jaroslav Stark - Avoiding Cytolysis: Evasion, Resistance and Repair

Cytotoxic killing of target cells by CD8 T cells and natural killer cells is one of the main mechanisms for the immune control of intracellular pathogens and tumours. The basis for this cell killing is the controlled release of perforin and granzyme granules at the immune cell-target cell interface, which together serve to puncture the cell membrane and activate the apoptotic program in the target cell. Both pathogens and tumours can sometimes evolve to escape immune killing. Research in this area has focused on mechanisms for evasion. However, an al-ternative strategy for a target cell is to boost defences rather than (or in addition to) avoiding detection. This has recently been highlighted by the discovery that certain tumour cells express a specific stoichiometric inhibitor of granzyme B known as PI-9. A variety of other cell types also express PI-9, including endothelial, mesothelial, and dendritic cells, as well as cytotoxic T cells themselves. We use a simple mathematical model to provide insight into the different roles of evasion and resistance in the evolution of escape mechanisms to avoid cytotoxicity. Finally, we suggest experiments to validate the hypotheses of the model, and discuss the implications for immunotherapy against intracellular pathogens and tumours.


Work done in collaboration with Jaroslav Stark, Cliburn Chan, Andrew J. T. George.

02:30 PM
03:00 PM
Zvi Grossman - Modeling the dynamics of HIV disease: simpler is not always better

The ability of simple mathematical models of cellular and viral dynamics in HIV infected people, and in SIV infected monkeys, to fit available kinetic data has often been over-interpreted. A common pitfall is a failure to properly differentiate between systemic and microscopic parameters, overlooking issues such as heterogeneity and non-uniformity, spatial as well as temporal, of the cellular and molecular constituents at the microscopic level. Another common pitfall is fitting a model to a limited set of data rather than to all the relevant available data. It is common also to make unjustified extrapolations from observed correlations to cause-and-effect relationships.


I shall overview the major outstanding issues in the area of HIV-infection dynamics and the attempts that have been made, with various degrees of success and acceptance, to use mathematical and conceptual modeling to help resolve these issues

03:30 PM
04:00 PM
Reinhard Laubenbacher - Interactive Visualization of Pathism Simulation Output Using Information-Rich Virtual Environments

In the life sciences, the development of rigorous models and databases of biological phenomena provides major benefits for biological research, drug design, and education. A grand goal in biology is to produce integrated information-rich biological databases that capture the complexity of reality. A common class of such databases can be characterized as integrating diverse information including: spatial representations of physical systems and phenomena, abstract data such as gene expression data and annotations, temporal dimension for time series, multiple levels of scale (from anatomical to cellular to molecular), and multiple runs of simulation output, and experimental results.


However, the current major shortcoming is the lack of effective user interfaces and visualizations for information-rich databases that enable biologists to gain insight. The true utility of the databases will come to fruition when biologists are able to explore and navigate them and relate effects between space, abstract data, and time across levels of scale. Current virtual environments and information visualizations lack the usability and support for such complex information-rich databases.


PathSim is an example of an information-rich model with associated databases. The main goal of PathSim is to model a variety of viral agents in human and animal hosts, from initial infection to viral clearance. PathSim allows an end-user to explore the physiology and dynamics of infections and immune system response. As an interface to this system, we are constructing and evaluating information-rich virtual environments (IRVEs) for the PathSim project. This interface framework can also be applied to other similar information-rich databases in the life sciences that share these characteristics.


An IRVE combines the capabilities of virtual environments and information visualization to support integrated exploration of both spatial and embedded abstract data. Biologists can view the simulated physical structures of the model in a 3D virtual environment, interact with visually embedded abstract data, navigate across levels of scale, choose data for display, and control simulation run management all within an integrated environment. For example, a user might decide to examine the effect of titer on the course of infection. Within the IRVE, the user deposits virions in the locations to be infected. After the simulation commences, the user revisits the IRVE to view signaling events initiated by virus deposition at the molecular level. Later, the user examines how fast the virus is spreading, killing cells, or recruiting immune cells to the vicinity. All activities are viewable in the virtual environment, with interactive links and data export to a suite of analytic tools also possible.


This work is discovering critical new methods for display and interaction in multi-scale IRVEs that are usable and useful for biologists. The system user interface operates on a wide range of hardware, from standard desktop displays to high-performance immersive CAVEs. The system will eventually be public and open for use in other applications.


Work done in collaboration with N.F. Polys, D. Bowman, K.A. Duca, R. Laubenbacher, and C. North.

04:00 PM
04:30 PM
Ping Ye - Lack of Good Correlation of Serum CC-chemokine Levels with HIV-1 Disease Stage and Response to Treatment

Three CC-chemokines, MIP-1α (CCL3), MIP-1β (CCL4), and RANTES (CCL5) are natural ligands for the HIV-1 co-receptor CCR5. To determine correlations between CC-chemokines and HIV-1 disease stage or response to treatment, we examined serum levels of MIP-1α, MIP-1β, and RANTES in sixty HIV-1 infected patients during eighteen months on highly active anti-retroviral therapy (HAART). Our results demonstrate that serum levels of MIP-1α and RANTES were elevated in HIV-1 infected individ-uals as compared with healthy controls. No significant difference has been found between four clinical stages of HIV-1 infection in serum levels of three CC-chemokines. Longitudinal HAART analyses re-veal there was a pronounced decline in serum MIP-1α levels over time. No difference in this decline was exhibited between HAART responders and non-responders. These findings indicate that production of MIP-1α and RANTES changes during HIV-1 infection and treatment; however, serum levels of CC-chemokines should not be used as a biomarker for HIV-1 disease stage or response to treatment.


Ping Ye, Powel Kazanjian, Steven L. Kunkel, and Denise E. Kirschner.

Wednesday, June 23, 2004
Time Session
10:30 AM
11:15 AM
Arturo Casadevall - The Damage-Framework of Microbial Pathogenesis and the Weapon Potential of a Microbe

In recognition of the need for a general theory of microbial pathogenesis that takes into account the contributions of both the host and the microbe we have developed the 'damage-response framework' of microbial pathogenesis. The 'damage-response' framework is based on three assumptions which we believe are universally agreed: 1) that microbial pathogenesis results from the interaction of two entities which are the microbe and the host; 2) that the host-relevant outcome of the interaction is the amount of damage incurred by the host as a consequence of the host-microbe interaction; and 3) that host damage can be initiated by the host response to a microbial determinant, a microbial determinant, or both. When damage is plotted as a function of the host response, a basic parabolic curve is generated whereby damage occurs primarily in situations of weak or strong immune responses. In most host-microbe interactions occurring in the setting of weak host responses, damage is primarily microbe mediated and/or the result of limitations of a critical aspect of the host response. In host-microbe interactions occurring in the setting of strong host responses, damage is primarily host-mediated. Considering various microbial disease syndromes and assuming that damage is a function of the host response allows the classification of host microbe interactions into six basic types. Based on knowledge of the disease syndromes that occur in patients, damage-response curves can be used to classify host-microbe interactions according to the amount of host damage as a function of the host immune response; whereas plotting damage as a function of time yields the basic states of microbial pathogenesis: pathogenesis: commensalisms, colonization, persistence and disease. The damage-response framework is useful because it considers all types of host-microbe interactions as continuous in the context of potentially quantifiable parameters. Furthermore, the damage-response framework simplifies the lexicon of microbial pathogenesis, predicts new types of microbial interactions and suggests new approaches to vaccine design. Using the principles of microbial pathogenesis we have also developed a formula for the weapon potential of a microbe, which allows one to compute this parameter from the principles of microbial pathogenesis.


Work done in collaboration with Arturo Casadevall and Liise-anne Pirofski.


References



  1. Casadevall, A., & Pirofski,,L. (1999). Host-pathogen interactions: redefining the basic concepts of virulence and pathogenicity. Infect. Immun., 67, 3703-3713.

  2. Casadevall, A., & Pirofski, L. (2000). Host-Pathogen Interactions: The basic concepts of microbial commensalism, colonization, infection, and disease. Infect. Immun., 68, 6511-6518.

  3. Casadevall, A., & Pirofski, L. (2001). Host-pathogen interactions: the attributes of virulence. J. Infect. Dis., 184, 337-344.

  4. Pirofski, L., & Casadevall, A. (2002). The meaning of microbial exposure, infection, colonisation, and disease in clinical practice. Lancet Infect. Dis., 2, 628.

  5. Casadevall, A., & Pirofski, L. (2003). The damage-response framework of microbial pathogenesis. Nature Microbiol. Rev., 1, 17-24.

  6. Casadevall, A., & Pirofski, L. (in press). The weapon potential of a microbe. Trends Microbiol.

11:45 AM
12:30 PM
Amy Herring - Mechanisms of Chronic Fungal Infection: The interplay between innate and adaptive immunity

Host defense against fungal infection is dependent on the interplay between the cells and early signals of innate immunity and T cells. The development of T1 cell-mediated immunity is essential for clearance of fungal pathogens. During states of chronic infection, the host response is often described as unresponsive, anergic, or dysregulated. However, the immunologic mechanisms underlying chronic fungal infections in otherwise healthy individuals remain unknown. We have developed several in vivo animal models to study the chronicity of Cryptococcus neoformans infection in the lung. In our first model, neutralization of TNF at the time of infection results in non-protective immunity to C. neoformans. The immune deviation induced by a transient TNF deficiency results in chronic inflammation and fungal infection. Our second model involves immunization of animals with C. neoformans-pulsed immature dendritic cells (imDC) prior to the pulmonary infection. imDC-recipient mice fail to clear the cryptococcal infection from the lungs and develop a non-protective immune response. Our studies suggest that interruption of critical early signals may underlie susceptibility to chronic fungal infection in immunocompetent individuals. We believe the timing of cytokine signals is more important than the simple presence of a cytokine during infection in determining whether a protective immune response develops. Our findings also demonstrate that immature dendritic cells pulsed with fungal antigen can promote non-protective immunity resulting in chronic pulmonary infection. The delayed induction of key early mediators due to virulence factors, immunotherapy, or secondary infections may promote fungal infection by inducing an ineffective cellular immune response (T2 vs. T1) and/or by immature dendritic cell induction of regulatory T cells.

02:00 PM
02:30 PM
Jorge Velasco-Hernandez - A Mathematical Approach to Modeling Structure and Conformation of Bacterial Consortia

I present work being developed at the IMP on three aspects of biofilm formation: spatial structure and its relation to coexistence of multispecies biofilms, the role of mutations in the exitence of colonial biofilms and the interaction between biofilms and the fluid environment in which they thrive. The results are preliminary and comments and criticisms are very wellcome.

02:30 PM
03:00 PM
John Ward - The Role of Quorum Sensing in Bacterial Wound Infections

Many species of bacteria incorporate a sophisticated cell-cell signalling mechanism, called quorum sensing (QS), to regulate their behaviour in a cell density dependent manner. Whilst infecting an open wound or burn, the opportunistic pathogen {it Pseudomonas aeruginosa} employs QS to initially subdue its virulence characteristics, "fooling" the immune response, whilst the population multiplies. Hence, when they do become virulent their greater numbers are more likely to overwhelm the immune system, leading to septicaemia and perhaps death. Here, the QS process involves the dimerising of a cell-signalling molecule (QSM) and a cognate protein, which enhances both QSM and virulence factor production; consequently, up regulation of virulence factor production is induced at high QSM concentrations, reflecting high population density.


We present a spatio-temporal model of bacterial growth and QS in an infected burn wound situation incorporating the known microbiology; the QS core of the model will be discussed in light of experimental work using liquid cultures, from which parameter estimates are obtained. Using asymptotic and numerical techniques the conflicting effects of QSM production in the infected regions and loss (via diffusion and degradation in the surrounding tissues) are studied. Regimes in which substantial up-regulation (and therefore virulence) can occur and on what timescale are determined in terms of the model parameters. Therapeutic implications will also be discussed.


Work done in collaboration with Adrian Koerber, John King, Paul Williams, Julie Croft, and Liz Sockett.

03:50 PM
04:20 PM
John Tomfohr - Pathway-level View of Gene Expression

Gene expression microarray experiments provide a snapshot of the expression levels of thousands of genes in a sample. The challenge is to interpret this data--for example, to identify key genes associated with some condition and to form hypotheses about their relation to that condition. While a large amount of work has focused on identifying significantly differently expressed individual genes, it is sometimes valuable to look at expression data at the level of groups of functionally related genes, such as those belonging to the same pathway or complex. This can reveal higher level features not as apparent from the variations in the individual genes alone. We present an approach to analyzing gene expression at a multi-gene level using a collection of about 400 predefined pathways and complexes. Gene expression levels are translated into pathway expression levels after a screening process that removes pathways for which the data show weak evidence of correlation between member genes. The method will be demonstrated using expression profiles from a study on diabetes and another on the immune response to inhaled LPS in mice.


Work done in collaboration with John Tomfohr, Jun Lu, and Thomas B. Kepler.

04:40 PM
05:10 PM
Sergei Pilyugin - Some Remarks on Backward Bifurcations and the Role of Coinfection in Multi-disease Dynamics

Some Remarks on Backward Bifurcations and the Role of Coinfection in Multi-disease Dynamics

Thursday, June 24, 2004
Time Session
09:00 AM
09:45 AM
JoAnne Flynn - The Immune Response to Mycobacterium Tuberculosis is Dynamic and Evolves from Primary to Chronic Infection

The immune response to Mycobacterium tuberculosis was studied in a murine model, focusing on the lungs and lung draining lymph nodes, over the course of 7 months. Bacterial numbers peak by week 4 post-infection and are then maintained at apparently steady levels in the lungs. The immune response, as measured by T cell infiltration and cytokine production, also peaks at 4 weeks post-infection, and then contracts. However, there are additional bursts of activation of the T cell response in the lungs over the next 6 months. Proliferation and apoptosis were also followed. The immune response may respond to small changes in bacterial numbers within the lungs by increasing to try to prevent reactivation. The effector functions of T cells were also studied. CD4 T cells are the primary producers of IFN-gamma early in infection, while CD8 T cells are cytotoxic for the first 6 weeks. In chronic infection, CD8 T cells are no longer cytotoxic, but begin to produce substantial amounts of IFN-g. The mechanisms responsible for modulation of CD8 T cell effector function remain unknown.

11:10 AM
11:55 AM
Mary O'Riordan - Bacterial and Host Response to Intracellular Infection

Bacterial and Host Response to Intracellular Infection

12:15 PM
01:00 PM
Mark Miller - Two-Photon Imaging of Lymphocyte Trafficking and Antigen Presentation

The adaptive immune response is initiated by physical contact between antigen-bearing dendritic cells (DCs) and antigen-specific naive T cells. These contacts occur deep within the highly specialized anatomy of secondary lymphoid tissues, such as the lymph nodes. A fundamental description of T cell priming dynamics in vivo has relevance to vaccine development and will advance our understanding of how immune responses are regulated during infection, cancer, and autoimmunity. Using two-photon microscopy we resolved the real-time behavior of endogenous DCs and CD4+ T cells in lymph node explants during a robust T cell response. Our results suggest that naive CD4+ T cells encounter DCs at random and not by following chemokine gradients emitted by DCs. In contrast, DCs enhance T cell repertoire scanning by vigorously deploying long agile dendrites, thereby increasing the available surface area for T cell interactions. Initial cognate T cell/DC interactions are remarkably dynamic and often involve serial contacts with DCs. As T cells activate they progress through several distinct phases of behavior from relatively immotile clusters containing a few cells to large groups of cells displaying dynamic swarming behavior. Quantitative analysis of our imaging data suggests that random motility is a natural property of lymphocytes and that stochastic, multi-agent based models may best describe lymphocyte trafficking and behavior in situ.


Work done in collaboration with Mark J. Miller, Ian Parker, and Michael D. Cahalan.

02:30 PM
03:00 PM
Simeone Marino - Dendritic Cell Trafficking and Antigen Presentation in the Human Immune Response to Mycobacterium Tuberculosis

Mycobacterium tuberculosis (Mtb) is an extraordinarily successful human pathogen, one of the major causes of death by infectious disease worldwide. A key issue for the study of tuberculosis is to understand why individuals infected with Mtb experience different clinical outcomes. To better understand the dynamics of Mtb infection and immunity, we coupled non-human primate (NHP) experiments with a mathematical model we previously developed that qualitatively and quantitatively captures important processes of cellular priming and activation. These processes occur between the lung and the nearest draining lymph node (DLN) where the key cells mediating this process are the dendritic cells (DC). We are able to reproduce typical disease progression scenarios including primary infection, latency or clearance.


The NHP experiments consist of bacteria and cell numbers from tissues of seventeen adult cynomolgus macaques (Macacca fascicularis) that were infected with M. tuberculosis strain Erdman (~25 CFU/animal via bronchoscope).


The main result of this work is that delays in either DC migration to the DLN or T-cell trafficking to the site of infection can alter the outcome of Mtb infection, defining progression to primary disease or latent infection and reactivated tuberculosis. Our results also support the idea that the development of a new generation of treatment against Mtb should optimally elicit a fast DC turnover at the site of infection, as well as strong activation of DCs for maximal antigen presentation and production of key cytokines. This will induce the most protective T cell response.


Simeone Marino, Santosh Pawar, Craig L. Fuller, Todd A. Reinhart, JoAnne L. Flynn, and Denise E. Kirschner.

03:00 PM
03:30 PM
Jose Segovia-Juarez - Understanding Control Mechanisms of TB Granuloma Formation with an Agent-based Model

Understanding Control Mechanisms of TB Granuloma Formation with an Agent-based Model

03:50 PM
04:20 PM
Stewart Chang - Why Does Mycobacterium Tuberculosis Use Multiple Mechanisms to Inhibit Antigen Presentation?

Why Does Mycobacterium Tuberculosis Use Multiple Mechanisms to Inhibit Antigen Presentation?

04:20 PM
04:50 PM
Jun Lu - Gene Expression Analysis of Host-Defense in Arabidopsis

This talk is about the mechanisms of plant host defense in Arabidopsis. I will give a background introduction on host defense systems in plants, and in particular, systematic acquired resistance (SAR). The activation of SAR by avirulent pathogens involves up-regulation of many pathogenesis-related genes, which further confer resistance to a broad spectrum of other pathogens. We use microarrays to identify genes involved in SAR, and to search for expression patterns that are distinct between virulent and avirulent pathogen infection.


Work done in collaboration with Jun Lu, John Tomfohr, Natalie Weaver, Dong Wang, Xiannian Dong, and Thomas Kepler.

Friday, June 25, 2004
Time Session
Name Email Affiliation
Assogba, Barnabe assogba.1@osu.edu Department of Veterinary Biosciences, The Ohio State University
Bajaria, Seema seemieb@umich.edu Microbiology and Immunology, University of Michigan
Best, Janet jbest@mbi.osu.edu Mathematics, The Ohio State University
Borisyuk, Alla borisyuk@mbi.osu.edu Mathematical Biosciences Institute, The Ohio State University
Burkhard, Mary Jo burkhard.19@osu.edu Department of Veterinary Biosciences, The Ohio State University
Carrington, Mary carringt@mail.ncifcrf.gov Laboratory of Genomic Diversity, National Institutes of Health
Casadevall, Arturo casadeva@aecom.yu.edu Albert Einstein College of Medicine
Chan, Cliburn c.chan@imperial.ac.uk Department of Immunology, Imperial College London
Chan, John jchan@aecom.yu.edu Albert Einstein University
Chang, Stewart stchang@umich.edu Microbiology and Immunology, University of Michigan
Cinti, Sandro scinti@umich.edu Infectious Diseases, University of Michigan
Cloyd, Miles mcloyd@utmb.edu Medical Branch, University of Texas
Cowell, Lindsay lgcowell@duke.edu Immunology, Duke University
Cracium, Gheorghe craciun@math.wisc.edu Dept. of Mathematics, University of Wisconsin-Madison
Danthi, Sanjay danthi.1@osu.edu Staff Scientist II, Genzyme Corporation
Day, Judy jday@mbi.osu.edu Department of Mathematics, University of Tennessee
De Leenheer, Patrick deleenhe@math.ufl.edu DIMACS Center, Rutgers University at New Brunswick
Dirita, Victor vdirita@umich.edu Microbiology and Immunology, University of Michigan
Dougherty, Daniel dpdoughe@mbi.osu.edu Mathematical Biosciences Institute, The Ohio State University
Flynn, JoAnne joanne@pitt.edu Molecular Genetics & Biochemistry, University of Pittsburgh
Goel, Pranay goelpra@helix.nih.gov NIDDK, Indian Institute of Science Education and Research
Grossman, Zvi lcgros@post.tau.ac.il Department of Physiology and Pharmacology, Tel-Aviv University
Guo, Jong-Sheng Mathematical Biosciences Institute, The Ohio State University
Guo, Yixin yixin@math.drexel.edu Department of Mathematics, The Ohio State University
Herring, Amy aherring@umich.edu Microbiology and Immunology, University of Michigan
Hodges, Andrew APHodges@manchester.edu DIMACS Center, Rutgers University at New Brunswick
Hodgkin, Phil hodgkin@wehi.edu.au Immunology, The Walter and Eliza Hall Institute of Medical Research
Jumper, Alicia Department of Mathematics, Oakwood University
Kelsoe, Garnett pulli014@mc.duke.edu Immunology, Duke University
Kepler, Thomas keple003@mc.duke.edu Biostatistics and Bioinformatics, Duke University
Kirschner, Denise kirschne@umich.edu Microbiology and Immunology, University of Michigan
Laubenbacher, Reinhard betsyw@vbi.vt.edu Virginia Bioinformatics Institute, Virginia Polytechnic Institute and State University
Lee, Kichol kduca@vbi.vt.edu Virginia Bioinformatics Institute, Virginia Polytechnic Institute and State University
Li, Michael mli@math.ualberta.ca Mathematical/Statistical Sciences, University of Alberta
Lim, Sookkyung limsk@math.uc.edu Department of Mathematical Sciences, University of Cincinnati
Lou, Yuan lou@math.ohio-state.edu Department of Mathematics, The Ohio State University
Lu, Jun lu000014@mc.duke.edu Biostatistics and Bioinformatics, Duke University
Macura, Natasa Natasa.Macura@Trinity.edu Department of Mathematics, Trinity University
Marino, Simeone simeonem@umich.edu Microbiology & Immunology, University of Michigan
Miller, Mark mmiller@sdsc.edu Physiology & Biophysics, University of California, Irvine
Nelson, George nelsong@ncifcrf.gov Bioinformatics/Biostatistics Group, SAIC/NCI-Frederick
O'Riordan, Mary oriordan@umich.edu Microbiology and Immunology, University of Michigan
O'Toole, Elizabeth otoole.26@osu.edu Department of Veterinary Clinical Sciences, The Ohio State University
Pilyugin, Sergei pilyugin@math.ufl.edu Department of Mathematics, University of Florida
Quong, Andrew aajq@sbcglobal.net Biosecurity and NanoSciences, Lawrence Livermore National Laboratory
Quong, Judy jnq@georgetown.edu Lombardi Cancer Center, Georgetown University
Ray, Christian jjray@umich.edu Microbiology and Immunology, University of Michigan
Reinhart, Todd reinhar@pitt.edu Infectious Diseases & Microbiology, University of Pittsburgh
Rejniak, Katarzyna rejniak@mbi.osu.edu Mathematical Biosciences Institute, The Ohio State University
Sandstede, Bjorn sandsted@math.ohio-state.edu Department of Mathematics, The Ohio State University
Schmid, Boris bvschmid@xsmail.com Theoretical Biology/Bioinformatics, Utrecht University
Segovia-Juarez, Jose jlsj@umich.edu Microbiology and Immunology, University of Michigan
Sheridan, John sheridan.1@osu.edu Molecular Virology, Immunology & Med Genetics, The Ohio State University
Sokhansanj, Bahrad sokhansanj@llnl.gov Computational Systems Biology Group, Lawrence Livermore National Laboratory
Stark, Jaroslav j.stark@imperial.ac.uk Department of Mathematics, Imperial College
Stengel, Robert stengel@princeton.edu Engineering and Applied Science, Princeton University
Tan, Wai-Yuan waitan@memphis.edu Mathematical Sciences, University of Memphis
Terman, David terman@math.ohio-state.edu Mathemathics Department, The Ohio State University
Tomaras, Georgia gdt@duke.edu AIDS Research Center, Duke University Medical Center
Tomfohr, John tomfohr@duke.edu Biostatistics and Bioinformatics, Duke University
Tsai, Chih-Chiang tsaijc@mbi.osu.edu Department of Mathematics, National Taiwan Normal University
Velasco-Hernandez, Jorge velascoj@imp.mx Department of Mathematics, Instituto Mexicano del Petroleo
Volpe, Joseph joseph.volpe@duke.edu Biostatistics and Bioinformatics, Duke University
Ward, John john.ward@lboro.ac.uk Department of Mathematical Sciences, Loughborough University
Wechselberger, Martin wm@mbi.osu.edu Mathematical Biosciences Insitute, The Ohio State University
Weinberger, Leor lsw@ucsd.edu Genomics, University of California, Berkeley
Wright, Geraldine wright.572@osu.edu School of Biology, Newcastle University
Ye, Ping pingye@bme.jhu.edu Biomedical Engineering, Johns Hopkins University
Yu, Yang yuyang@math.ohio-state.edu Department of Mathematics, The Ohio State University
CTL Action During HIV-1 is Determined via Interactions with Several Cell Types

During HIV-1 infection, interactions between immune cells and virus yield three distinct disease stages: high viral levels in acute infection, immune control in the chronic stage, and AIDS, when CD4+ T cells fall to extremely low levels. The immune system consists of many players that have key roles during infection. In particular, CD8+ T cells are important for killing of virally infected cells as well as inhibition of cellular infection and viral production. Activated CD8+ T cells, or cytotoxic T cells (CTLs) have unique functions during HIV-1, most of which are thought to be compromised during HIV-1 disease progression. Controversy exists regarding priming of CTLs, and our work attempts to address the dynamics occurring during HIV-1 infection. To explore the influence of CD8+ T cells as determinants in disease progression and issues relating to their priming and activation, we develop a two-compartment ordinary differential equation model describing cellular interactions that occur during HIV-1 infection. We track CD4+ T cells, CD8+ T cells, dendritic cells, infected cells, and virus, each circulating between blood and lymphatic tissues. Using parameter estimates from literature, we simulate commonly observed disease patterns. Our results indicate that CD4+ T cells as well as dendritic cells likely play a significant role in successful activation of CD8+ T cells into CTLs. Model simulations correlate with clinical data confirming a quantitative relationship between CD4+ T cells and CD8+ T-cell effectiveness.

Patterns of KIR/HLA Association in HIV and HCV Infection

Natural killer (NK) cells are a unique group of lymphocytes involved in surveillance and killing of foreign or infected cells through a mechanism involving recognition of HLA molecules by an extremely diverse set of receptors on the NK cell surface. A major group of these receptors are the killer immunoglobin-like receptors (KIRs) that are encoded by genes mapping to chromosome 19q13.4. These receptors regulate inhibition and activation of NK cell responses through recognition of HLA class I molecules on target cells. Given their receptor-ligand relationship, we have hypothesized that KIR may be involved in many of the diseases for which an HLA influence has been identified. In this regard, we performed KIR genotyping in several AIDS cohorts and have shown that one activating KIR allele, KIR3DS1, in combination with the HLA-B alleles encoding its possible HLA ligand, HLA-Bw4-80Ile, results in delayed progression to AIDS, suggesting a synergistic interaction between the two loci. We have also performed KIR genotyping in a cohort of individuals with HCV infection. We show that genes encoding the inhibitory NK cell receptor KIR2DL3 and its HLA-Cw group1 ligand, which transmit relatively weak inhibitory signals, influence resolution of hepatitis C virus (HCV) infection. This effect was observed in Caucasians and African Americans with expected low infectious doses of HCV, but not in those with high-dose exposure, in whom the innate immune response is likely overwhelmed. Data from both of these infectious diseases indicate that net activation of NK cells is beneficial in the anti-viral immune response.

The Damage-Framework of Microbial Pathogenesis and the Weapon Potential of a Microbe

In recognition of the need for a general theory of microbial pathogenesis that takes into account the contributions of both the host and the microbe we have developed the 'damage-response framework' of microbial pathogenesis. The 'damage-response' framework is based on three assumptions which we believe are universally agreed: 1) that microbial pathogenesis results from the interaction of two entities which are the microbe and the host; 2) that the host-relevant outcome of the interaction is the amount of damage incurred by the host as a consequence of the host-microbe interaction; and 3) that host damage can be initiated by the host response to a microbial determinant, a microbial determinant, or both. When damage is plotted as a function of the host response, a basic parabolic curve is generated whereby damage occurs primarily in situations of weak or strong immune responses. In most host-microbe interactions occurring in the setting of weak host responses, damage is primarily microbe mediated and/or the result of limitations of a critical aspect of the host response. In host-microbe interactions occurring in the setting of strong host responses, damage is primarily host-mediated. Considering various microbial disease syndromes and assuming that damage is a function of the host response allows the classification of host microbe interactions into six basic types. Based on knowledge of the disease syndromes that occur in patients, damage-response curves can be used to classify host-microbe interactions according to the amount of host damage as a function of the host immune response; whereas plotting damage as a function of time yields the basic states of microbial pathogenesis: pathogenesis: commensalisms, colonization, persistence and disease. The damage-response framework is useful because it considers all types of host-microbe interactions as continuous in the context of potentially quantifiable parameters. Furthermore, the damage-response framework simplifies the lexicon of microbial pathogenesis, predicts new types of microbial interactions and suggests new approaches to vaccine design. Using the principles of microbial pathogenesis we have also developed a formula for the weapon potential of a microbe, which allows one to compute this parameter from the principles of microbial pathogenesis.


Work done in collaboration with Arturo Casadevall and Liise-anne Pirofski.


References



  1. Casadevall, A., & Pirofski,,L. (1999). Host-pathogen interactions: redefining the basic concepts of virulence and pathogenicity. Infect. Immun., 67, 3703-3713.

  2. Casadevall, A., & Pirofski, L. (2000). Host-Pathogen Interactions: The basic concepts of microbial commensalism, colonization, infection, and disease. Infect. Immun., 68, 6511-6518.

  3. Casadevall, A., & Pirofski, L. (2001). Host-pathogen interactions: the attributes of virulence. J. Infect. Dis., 184, 337-344.

  4. Pirofski, L., & Casadevall, A. (2002). The meaning of microbial exposure, infection, colonisation, and disease in clinical practice. Lancet Infect. Dis., 2, 628.

  5. Casadevall, A., & Pirofski, L. (2003). The damage-response framework of microbial pathogenesis. Nature Microbiol. Rev., 1, 17-24.

  6. Casadevall, A., & Pirofski, L. (in press). The weapon potential of a microbe. Trends Microbiol.

Why Does Mycobacterium Tuberculosis Use Multiple Mechanisms to Inhibit Antigen Presentation?

Why Does Mycobacterium Tuberculosis Use Multiple Mechanisms to Inhibit Antigen Presentation?

HIV/AIDS-2004

HIV/AIDS-2004

Pathogenic Mechanism of HIV

HIV infection causes an acquired immunodeficiency, principally because it causes depletion of CD4 lymphocytes. The mechanism by which the virus depletes these cells however, is not clearly understood. Many studies demonstrate that uninfected CD4 cells are the cells predominantly depleted. The most plausible mechanism, which is backed by in vivo data, involves the consequences of HIV contact with resting CD4 lymphocytes, which are cells which do not support replication of HIV. HIV binding to and signaling through CD4 and chemokine receptor molecules on resting CD4 lymphocytes and other cell types (which extensively occurs as the rare productively infected cells migrate among other cells within the lymphoid tissues) induces up regulation L-selection and Fas. When these resting HIV-signaled CD4 cells return to the blood, they home very rapidly back to the peripheral lymph nodes and axial bone marrow. Thus, the disappearance of CD4 cells from the blood in HIV-infected individuals appears to be due to them leaving the blood. Approximately one third of these cells that are HIV induced to home to lymph nodes are subsequently induced into apoptosis during the process of trans-endothelial migration, during which they receive secondary signals through various homing receptors. These cells are not making HIV, which explain the observation that CD4 cells not making HIV are the predominate cells dying in the lymph nodes of HIV+ subjects. Mathematical modeling of this process yielded an accurate picture of the rate of depletion of CD4 cells over an eight to ten year period. This is unique mechanism of pathogenesis for a virus, and if correct, leads to the possibility that HIV might not cause depletion of CD4 lymphocytes if it used some other receptor to infect cells.

The Immune Response to Mycobacterium Tuberculosis is Dynamic and Evolves from Primary to Chronic Infection

The immune response to Mycobacterium tuberculosis was studied in a murine model, focusing on the lungs and lung draining lymph nodes, over the course of 7 months. Bacterial numbers peak by week 4 post-infection and are then maintained at apparently steady levels in the lungs. The immune response, as measured by T cell infiltration and cytokine production, also peaks at 4 weeks post-infection, and then contracts. However, there are additional bursts of activation of the T cell response in the lungs over the next 6 months. Proliferation and apoptosis were also followed. The immune response may respond to small changes in bacterial numbers within the lungs by increasing to try to prevent reactivation. The effector functions of T cells were also studied. CD4 T cells are the primary producers of IFN-gamma early in infection, while CD8 T cells are cytotoxic for the first 6 weeks. In chronic infection, CD8 T cells are no longer cytotoxic, but begin to produce substantial amounts of IFN-g. The mechanisms responsible for modulation of CD8 T cell effector function remain unknown.

Modeling the dynamics of HIV disease: simpler is not always better

The ability of simple mathematical models of cellular and viral dynamics in HIV infected people, and in SIV infected monkeys, to fit available kinetic data has often been over-interpreted. A common pitfall is a failure to properly differentiate between systemic and microscopic parameters, overlooking issues such as heterogeneity and non-uniformity, spatial as well as temporal, of the cellular and molecular constituents at the microscopic level. Another common pitfall is fitting a model to a limited set of data rather than to all the relevant available data. It is common also to make unjustified extrapolations from observed correlations to cause-and-effect relationships.


I shall overview the major outstanding issues in the area of HIV-infection dynamics and the attempts that have been made, with various degrees of success and acceptance, to use mathematical and conceptual modeling to help resolve these issues

Mechanisms of Chronic Fungal Infection: The interplay between innate and adaptive immunity

Host defense against fungal infection is dependent on the interplay between the cells and early signals of innate immunity and T cells. The development of T1 cell-mediated immunity is essential for clearance of fungal pathogens. During states of chronic infection, the host response is often described as unresponsive, anergic, or dysregulated. However, the immunologic mechanisms underlying chronic fungal infections in otherwise healthy individuals remain unknown. We have developed several in vivo animal models to study the chronicity of Cryptococcus neoformans infection in the lung. In our first model, neutralization of TNF at the time of infection results in non-protective immunity to C. neoformans. The immune deviation induced by a transient TNF deficiency results in chronic inflammation and fungal infection. Our second model involves immunization of animals with C. neoformans-pulsed immature dendritic cells (imDC) prior to the pulmonary infection. imDC-recipient mice fail to clear the cryptococcal infection from the lungs and develop a non-protective immune response. Our studies suggest that interruption of critical early signals may underlie susceptibility to chronic fungal infection in immunocompetent individuals. We believe the timing of cytokine signals is more important than the simple presence of a cytokine during infection in determining whether a protective immune response develops. Our findings also demonstrate that immature dendritic cells pulsed with fungal antigen can promote non-protective immunity resulting in chronic pulmonary infection. The delayed induction of key early mediators due to virulence factors, immunotherapy, or secondary infections may promote fungal infection by inducing an ineffective cellular immune response (T2 vs. T1) and/or by immature dendritic cell induction of regulatory T cells.

Interactive Visualization of Pathism Simulation Output Using Information-Rich Virtual Environments

In the life sciences, the development of rigorous models and databases of biological phenomena provides major benefits for biological research, drug design, and education. A grand goal in biology is to produce integrated information-rich biological databases that capture the complexity of reality. A common class of such databases can be characterized as integrating diverse information including: spatial representations of physical systems and phenomena, abstract data such as gene expression data and annotations, temporal dimension for time series, multiple levels of scale (from anatomical to cellular to molecular), and multiple runs of simulation output, and experimental results.


However, the current major shortcoming is the lack of effective user interfaces and visualizations for information-rich databases that enable biologists to gain insight. The true utility of the databases will come to fruition when biologists are able to explore and navigate them and relate effects between space, abstract data, and time across levels of scale. Current virtual environments and information visualizations lack the usability and support for such complex information-rich databases.


PathSim is an example of an information-rich model with associated databases. The main goal of PathSim is to model a variety of viral agents in human and animal hosts, from initial infection to viral clearance. PathSim allows an end-user to explore the physiology and dynamics of infections and immune system response. As an interface to this system, we are constructing and evaluating information-rich virtual environments (IRVEs) for the PathSim project. This interface framework can also be applied to other similar information-rich databases in the life sciences that share these characteristics.


An IRVE combines the capabilities of virtual environments and information visualization to support integrated exploration of both spatial and embedded abstract data. Biologists can view the simulated physical structures of the model in a 3D virtual environment, interact with visually embedded abstract data, navigate across levels of scale, choose data for display, and control simulation run management all within an integrated environment. For example, a user might decide to examine the effect of titer on the course of infection. Within the IRVE, the user deposits virions in the locations to be infected. After the simulation commences, the user revisits the IRVE to view signaling events initiated by virus deposition at the molecular level. Later, the user examines how fast the virus is spreading, killing cells, or recruiting immune cells to the vicinity. All activities are viewable in the virtual environment, with interactive links and data export to a suite of analytic tools also possible.


This work is discovering critical new methods for display and interaction in multi-scale IRVEs that are usable and useful for biologists. The system user interface operates on a wide range of hardware, from standard desktop displays to high-performance immersive CAVEs. The system will eventually be public and open for use in other applications.


Work done in collaboration with N.F. Polys, D. Bowman, K.A. Duca, R. Laubenbacher, and C. North.

Gene Expression Analysis of Host-Defense in Arabidopsis

This talk is about the mechanisms of plant host defense in Arabidopsis. I will give a background introduction on host defense systems in plants, and in particular, systematic acquired resistance (SAR). The activation of SAR by avirulent pathogens involves up-regulation of many pathogenesis-related genes, which further confer resistance to a broad spectrum of other pathogens. We use microarrays to identify genes involved in SAR, and to search for expression patterns that are distinct between virulent and avirulent pathogen infection.


Work done in collaboration with Jun Lu, John Tomfohr, Natalie Weaver, Dong Wang, Xiannian Dong, and Thomas Kepler.

Dendritic Cell Trafficking and Antigen Presentation in the Human Immune Response to Mycobacterium Tuberculosis

Mycobacterium tuberculosis (Mtb) is an extraordinarily successful human pathogen, one of the major causes of death by infectious disease worldwide. A key issue for the study of tuberculosis is to understand why individuals infected with Mtb experience different clinical outcomes. To better understand the dynamics of Mtb infection and immunity, we coupled non-human primate (NHP) experiments with a mathematical model we previously developed that qualitatively and quantitatively captures important processes of cellular priming and activation. These processes occur between the lung and the nearest draining lymph node (DLN) where the key cells mediating this process are the dendritic cells (DC). We are able to reproduce typical disease progression scenarios including primary infection, latency or clearance.


The NHP experiments consist of bacteria and cell numbers from tissues of seventeen adult cynomolgus macaques (Macacca fascicularis) that were infected with M. tuberculosis strain Erdman (~25 CFU/animal via bronchoscope).


The main result of this work is that delays in either DC migration to the DLN or T-cell trafficking to the site of infection can alter the outcome of Mtb infection, defining progression to primary disease or latent infection and reactivated tuberculosis. Our results also support the idea that the development of a new generation of treatment against Mtb should optimally elicit a fast DC turnover at the site of infection, as well as strong activation of DCs for maximal antigen presentation and production of key cytokines. This will induce the most protective T cell response.


Simeone Marino, Santosh Pawar, Craig L. Fuller, Todd A. Reinhart, JoAnne L. Flynn, and Denise E. Kirschner.

Two-Photon Imaging of Lymphocyte Trafficking and Antigen Presentation

The adaptive immune response is initiated by physical contact between antigen-bearing dendritic cells (DCs) and antigen-specific naive T cells. These contacts occur deep within the highly specialized anatomy of secondary lymphoid tissues, such as the lymph nodes. A fundamental description of T cell priming dynamics in vivo has relevance to vaccine development and will advance our understanding of how immune responses are regulated during infection, cancer, and autoimmunity. Using two-photon microscopy we resolved the real-time behavior of endogenous DCs and CD4+ T cells in lymph node explants during a robust T cell response. Our results suggest that naive CD4+ T cells encounter DCs at random and not by following chemokine gradients emitted by DCs. In contrast, DCs enhance T cell repertoire scanning by vigorously deploying long agile dendrites, thereby increasing the available surface area for T cell interactions. Initial cognate T cell/DC interactions are remarkably dynamic and often involve serial contacts with DCs. As T cells activate they progress through several distinct phases of behavior from relatively immotile clusters containing a few cells to large groups of cells displaying dynamic swarming behavior. Quantitative analysis of our imaging data suggests that random motility is a natural property of lymphocytes and that stochastic, multi-agent based models may best describe lymphocyte trafficking and behavior in situ.


Work done in collaboration with Mark J. Miller, Ian Parker, and Michael D. Cahalan.

Bacterial and Host Response to Intracellular Infection

Bacterial and Host Response to Intracellular Infection

Some Remarks on Backward Bifurcations and the Role of Coinfection in Multi-disease Dynamics

Some Remarks on Backward Bifurcations and the Role of Coinfection in Multi-disease Dynamics

Host/Pathogen Interactions at Multiple Levels during HIV-1 Infection and AIDS

The interactions between HIV-1 and the host occur at multiple levels and have generally been studied at all such levels. This spectrum spans from the associations of individual proteins, through cell/cell interactions, on up to the interactions of infected individuals within populations. Increased understanding of such interactions has led to intervention strategies - implemented or proposed - at each level including such diverse strategies as viral reverse transcriptase inhibitors and community outreach programs. Our interests have focused on the effects of HIV-1 infection on host function at the cellular and tissue levels, using infection of nonhuman primates with the related simian immunodeficiency virus (SIV) as a model system. Our ultimate goal is to identify new targets for therapeutic intervention, and we have approached this issue in two major ways. First, we have studied the effects of expression of a pathogenesis-associated viral protein named "Nef" on cell surface protein expression. Nef is not essential for viral replication in vitro, but contributes to high-level viral replication in vivo and the subsequent development of AIDS, although it has no known enzymatic function. In searching for cellular binding partners of Nef, we identified a critical component of the T cell signaling complex, TCR , as a Nef partner. SIV Nef bound TCR on two independent sites and reduced the surface expression levels of the TCR signaling complex, thereby likely inhibiting the ability of an infected cell to exert its function as an immune cell. Second, we have studied the effects of in vivo infection with SIV on the immune environments within lymphoid tissues. We have used gene expression profiling approaches such as cDNA arrays, in situ hybridization, and real-time RT-PCR to identify genes differentially expressed in tissues from rhesus macaques infected with pathogenic SIV. These approaches have revealed a number of alterations to immune environments in lymphoid, lung, and intestinal tissues during the progression of SIV-associated disease. The changes we have identified and which are likely to be important in this disease include: (1) up-regulation of chemokines, which control constitutive and inflammatory cell trafficking; (2) altered tissue compositions of dendritic cells, which are potent antigen presenting cells controlling the nature and strength of immune responses; and (3) up-regulation of members of a group of receptors within the innate immune system, Toll-like receptors, which control rapidly-induced inflammatory responses. These approaches have their respective strengths and limitations, but combined together can provide insight into the roles that these alterations to immune function play in the progression of disease, and thereby identify targets for new therapeutic intervention strategies.


Funding Sources: National Institutes of Health; National Heart, Lung, and Blood Institute.

Understanding Control Mechanisms of TB Granuloma Formation with an Agent-based Model

Understanding Control Mechanisms of TB Granuloma Formation with an Agent-based Model

Avoiding Cytolysis: Evasion, Resistance and Repair

Cytotoxic killing of target cells by CD8 T cells and natural killer cells is one of the main mechanisms for the immune control of intracellular pathogens and tumours. The basis for this cell killing is the controlled release of perforin and granzyme granules at the immune cell-target cell interface, which together serve to puncture the cell membrane and activate the apoptotic program in the target cell. Both pathogens and tumours can sometimes evolve to escape immune killing. Research in this area has focused on mechanisms for evasion. However, an al-ternative strategy for a target cell is to boost defences rather than (or in addition to) avoiding detection. This has recently been highlighted by the discovery that certain tumour cells express a specific stoichiometric inhibitor of granzyme B known as PI-9. A variety of other cell types also express PI-9, including endothelial, mesothelial, and dendritic cells, as well as cytotoxic T cells themselves. We use a simple mathematical model to provide insight into the different roles of evasion and resistance in the evolution of escape mechanisms to avoid cytotoxicity. Finally, we suggest experiments to validate the hypotheses of the model, and discuss the implications for immunotherapy against intracellular pathogens and tumours.


Work done in collaboration with Jaroslav Stark, Cliburn Chan, Andrew J. T. George.

Optimal Control of Disease Processes

Mathematical modeling of immune system response to pathogenic assault can greatly aid the management of infected individuals, but considerable development must occur before the biology and mathematics can be merged to provide tools for treatment. Given a model for a disease process, it remains to be seen how established methods of control system design can suggest therapeutic protocols that could be applied in a clinical setting.


We examine means for specifying optimal therapeutic protocols given a satisfactory immune system model. For illustration, the presentation is based on three models that represent humoral response to extracellular bacteria, cellular response to the human-immunodeficiency virus (HIV), and control of cancer by anti-tumor viruses. Therapy is based upon both open- and closed-loop optimal control strategies that take into account uncertainty in system models, measurements, and environment. We note that though observability of the dynamic state is a critical issue for closed-loop control, effective control can be provided by incomplete or "noisy" measurements. Immune system models possess classical attributes of dynamic stability or instability; for example, our humoral response model is stable, while the HIV model is not. In either case, effective therapies can be specified, although the HIV infection is never cured, requiring continued treatment to keep the condition in remission.

Assessment of Treatment Effects on HIV Pathogenesis Under Haart by State Space Models

To monitor the progression of therapy in HIV-infected individuals treated with anti-viral drugs, it is critical to estimate and assess the efficiency of the drugs and to estimate the number of infectious and non-infectious HIV under treatment. In this paper we have developed a method to estimate these parameters and the state variables to assess effects of drugs on HIV pathogenesis. As an illustration, we have applied the method to some clinical and laboratory data of an AIDS patient treated with various anti-viral drugs. For this individual, we have estimated the numbers of infectious HIV virus and non-infectious HIV virus per $ml$ of blood over time, and the rates for measuring effects of drugs. These estimates show that the HAART protocol has effectively controlled the number of infectious HIV virus to below $400/ml$ copies although the total number of HIV copies was very high in some intervals.


Work done in collaboration with Wai-Yuan Tan, Ping Zhang, Xiao-Ping Xiong, and Pat Flynn.

Noncytolytic CD8+ T cell-Mediated Suppression and HIV-1 Escape

CD8+ T cells play an important role in controlling virus replication in both acute and chronic HIV infection. There is substantial interest in further deciphering the contributions of noncytolytic CD8+ T cells in reducing virus replication as part of the host's protective immune response. It is commonly known that HIV can escape potent immune pressures such as cytotoxic T lymphocytes and neutralizing antibodies. It has recently become evident that virus can escape from noncytolytic suppression illustrating the ability of this antiviral activity to exert significant immune pressure in vivo. The molecules involved in this antiviral response and their precise mechanisms remain elusive. Studies of HIV variants harboring escape mutations are likely to provide new insights into the identities of noncytolytic CD8+ suppression.

Pathway-level View of Gene Expression

Gene expression microarray experiments provide a snapshot of the expression levels of thousands of genes in a sample. The challenge is to interpret this data--for example, to identify key genes associated with some condition and to form hypotheses about their relation to that condition. While a large amount of work has focused on identifying significantly differently expressed individual genes, it is sometimes valuable to look at expression data at the level of groups of functionally related genes, such as those belonging to the same pathway or complex. This can reveal higher level features not as apparent from the variations in the individual genes alone. We present an approach to analyzing gene expression at a multi-gene level using a collection of about 400 predefined pathways and complexes. Gene expression levels are translated into pathway expression levels after a screening process that removes pathways for which the data show weak evidence of correlation between member genes. The method will be demonstrated using expression profiles from a study on diabetes and another on the immune response to inhaled LPS in mice.


Work done in collaboration with John Tomfohr, Jun Lu, and Thomas B. Kepler.

A Mathematical Approach to Modeling Structure and Conformation of Bacterial Consortia

I present work being developed at the IMP on three aspects of biofilm formation: spatial structure and its relation to coexistence of multispecies biofilms, the role of mutations in the exitence of colonial biofilms and the interaction between biofilms and the fluid environment in which they thrive. The results are preliminary and comments and criticisms are very wellcome.

The Role of Quorum Sensing in Bacterial Wound Infections

Many species of bacteria incorporate a sophisticated cell-cell signalling mechanism, called quorum sensing (QS), to regulate their behaviour in a cell density dependent manner. Whilst infecting an open wound or burn, the opportunistic pathogen {it Pseudomonas aeruginosa} employs QS to initially subdue its virulence characteristics, "fooling" the immune response, whilst the population multiplies. Hence, when they do become virulent their greater numbers are more likely to overwhelm the immune system, leading to septicaemia and perhaps death. Here, the QS process involves the dimerising of a cell-signalling molecule (QSM) and a cognate protein, which enhances both QSM and virulence factor production; consequently, up regulation of virulence factor production is induced at high QSM concentrations, reflecting high population density.


We present a spatio-temporal model of bacterial growth and QS in an infected burn wound situation incorporating the known microbiology; the QS core of the model will be discussed in light of experimental work using liquid cultures, from which parameter estimates are obtained. Using asymptotic and numerical techniques the conflicting effects of QSM production in the infected regions and loss (via diffusion and degradation in the surrounding tissues) are studied. Regimes in which substantial up-regulation (and therefore virulence) can occur and on what timescale are determined in terms of the model parameters. Therapeutic implications will also be discussed.


Work done in collaboration with Adrian Koerber, John King, Paul Williams, Julie Croft, and Liz Sockett.

Stochastic Fluctuations in HIV-1 Tat Transactivation May Lead to Proviral Latency

The human immunodeficiency virus type I (HIV-1) proviral latent reservoir is considered the most significant obstacle facing HIV-1 eradication from the patient. The exact mechanism by which this reservoir is established remains a topic of much research. Bacteriophage l is known to utilize stochastic molecular fluctuations (SMF) in viral protein levels (C1 and Cro) to influence its lifecycle decision between lytic and lysogenic states and recently SMF in yeast transcriptional and translational pathways have been observed to lead to clonal population variability, even in the absence of chromatin remodeling. SMF have yet to be demonstrated or implicated in higher eukaryotes or mammalian systems. Here we present the first evidence that an HIV-1 positive feedback regulatory pathway, implicated in the establishment of proviral latency (the HIV-1 Tat transactivation loop) may utilize such stochastic molecular fluctuations.


Work done in collaboration with Leor S. Weinberger, Adam P. Arkin, and David V. Schaffer

Lack of Good Correlation of Serum CC-chemokine Levels with HIV-1 Disease Stage and Response to Treatment

Three CC-chemokines, MIP-1α (CCL3), MIP-1β (CCL4), and RANTES (CCL5) are natural ligands for the HIV-1 co-receptor CCR5. To determine correlations between CC-chemokines and HIV-1 disease stage or response to treatment, we examined serum levels of MIP-1α, MIP-1β, and RANTES in sixty HIV-1 infected patients during eighteen months on highly active anti-retroviral therapy (HAART). Our results demonstrate that serum levels of MIP-1α and RANTES were elevated in HIV-1 infected individ-uals as compared with healthy controls. No significant difference has been found between four clinical stages of HIV-1 infection in serum levels of three CC-chemokines. Longitudinal HAART analyses re-veal there was a pronounced decline in serum MIP-1α levels over time. No difference in this decline was exhibited between HAART responders and non-responders. These findings indicate that production of MIP-1α and RANTES changes during HIV-1 infection and treatment; however, serum levels of CC-chemokines should not be used as a biomarker for HIV-1 disease stage or response to treatment.


Ping Ye, Powel Kazanjian, Steven L. Kunkel, and Denise E. Kirschner.