Aging and Adaptation in Infectious Diseases II
Category: Application Area Application Area: Infectious Diseases
Date/Time: April 30, 2019 - May 3, 2019
Jean Carlson (UCSB)
Mercedes Pascual (Univ. Chicago)
Phil Arevalo (Univ. Chicago)
Sarah Cobey (Univ. Chicago)
Andrew P. Dobson (Princeton)
John Doyle (Caltech)
Katie Gostic (UCLA)
Andrea L. Graham (Princeton)
Qixin He (Univ. Chicago)
Eric Jones (UCSB)
Chris Kempes (SFI)
Micaela Martinez (Columbia Univ.)
Pamela Martinez (Harvard)
Alan Perelson (LANL/SFI)
André de Roos (Univ. Amsterdam)
David Schneider (Stanford)
Jiming Sheng (UCLA)
Marcos Viera (Univ. Chicago)
Add an Agenda Item[edit source]
Our working group aims to explore the role of aging and adaptation in infectious diseases operating over multiple organizational and temporal scales. General areas include immune system dynamics and age, host-pathogen co-adaptation in chronic vs. acute infections, pathogen antigenic diversity and endemism, effects of age on infectious diseases in human and non-human hosts. Overarching themes include memory, (co)adaptation, diversity, feedback, robustness and fragility. We are interested in aging as increasing fragility to infection, and in complex biological time as related to individual variation in disease progression and recovery. We are also interested in aging of the pathogen in terms of its ability to persist and withstand intervention efforts, and how this robustness is in turn related to pathogen (antigenic) diversity. In all these areas, the dynamic acquisition and loss of information through the immune system plays a central role at the individual and population levels. The goal of this second working group is to bring together a subset of the participants from the first meeting to develop collaborative research on the questions that emerged from our previous discussions . These questions include the interaction of the adaptive and innate immune system in the dynamics of infection, the role of early-childhood exposure (‘imprinting’) in later immune protection and in defining the temporal changes of the antigenic map, and the allometric scaling of the immune system dynamics with organism size.
Andrew P. Dobson (Princeton), Chris Kempes (SFI) - Session I: Immune System: Architecture and Dynamics[edit source]
A key lesson from allometric scaling perspectives has been that a variety of physiological processes and timescales systematically change with body size. These have important implications for interpreting a variety of ecological processes and for normalizing physiology across diverse organisms. Applying these concepts to infectious disease may, on the practical side, make it possible to scale interventions between organisms of very different size, and on scientific side, help us to systematize the ecology and evolution of hosts and parasites. In this talk we will discuss: 1) how various immune dynamics can be systematically scaled with body size and what implications this may have for organism physiology, 2) the time-scales of infection across diverse organisms, 3) the consequences of infection on lifetime reproduction across organisms of different size and across different broad taxonomic groups, and 4) the efficacy of vaccines and the timescales of immunity.
Jean Carlson (UCSB), Eric Jones (UCSB) - Session I: Immune System: Architecture and Dynamics[edit source]
The innate and adaptive components of the immune system do not age independently. We consider a mathematical model that couples these two complementary responses, and demonstrate a mechanism in which the progression of immunosenescence in the adaptive response leads to aging in the inflammatory response. We analyze the innate-adaptive interface of this coupled immune model by studying cytomegalovirus infection, a persistent infection that interacts with both components of the immune system.
Pamela Martinez (Harvard) - Session I: Immune System: Architecture and Dynamics[edit source]
Two kinds of immunity underlie pathogen competition for hosts as a function of history of exposure: specific immunity depending on memory of specific variants that have been seen before, and generalized immunity resulting only from the number of previous infections regardless of their antigenic identity. The role of immune selection remains unclear in rotavirus, the most common cause of diarrheal disease worldwide which contributes 40% of total hospitalizations in young children. With a process-based model that allows for both demographic and measurement noise, we analyzed over 10 years of monthly incidence data at the serotype level from Dhaka, Bangladesh. We specifically investigated the role of specific and generalized immunity in temporal patterns of antigenic diversity by fitting and comparing models that represent these different hypotheses, including consideration of variation in the two antigenic determinants on the surface of the virus. Our findings show that strong generalized immunity is needed to recover the serotype dynamics of rotavirus in Dhaka, a process dominated by exposure to VP4 (P-type), the outer layer protein that mediates cell attachment. Our results further indicate a role of specific immunity via the VP7 (outer capsid protein, G-type), whose effect is weak by comparison to that of VP4 but is nevertheless required to capture the epidemiological dynamics and antigenic diversity patterns of the virus.
John Doyle (Caltech) - Session I: Immune System: Architecture and Dynamics[edit source]
Effective layered architectures such as in brains and organisms seamlessly integrate high level goal and decision making and planning with fast lower level sensing, reflex, and action and facilitate learning, adaptation, augmentation (tools), and teamwork, while maintaining internal homeostasis. This is all despite the severe demands such actions can put on the whole body’s physiology, and despite being implemented in highly energy efficient hardware that has distributed, sparse, quantized, noisy, delayed, and saturating sensing, communications, computing, and actuation. Similar layering extends downward into the cellular level, out into ecological and social systems, and many aspects of this convergent evolution will increasingly dominate our most advanced technologies. Simple demos using audience’s brains can highlight universal laws and architectures and their relevance to tech, bio, neuro, med, and social networks. This suggests conjectures about senescence, and tradeoffs in the evolution of cancer, wound healing, degenerative diseases, auto-immunity, parasitism, and social organization, and potential animal models to explore these tradeoffs.
With this motivation, we’ll sketch progress on a new unified theory of complex networks that integrates communications, control, and computation with applications to cyberphysical systems as well as neuroscience and biology. Though based on completely different constraints arising from different environments, functions, and hardware, such systems face universal tradeoffs (laws) in dimensions such as efficiency, robustness, security, speed, flexibility, and evolvability. And successful systems share remarkable universals in architecture, including layering, localization, and diversity sweet spots, to effectively manage these tradeoffs, as well as universal fragilities, particularly to infectious hijacking.
I have videos of some introductory material and posted it in my public dropbox folder:
Some new neuro stuff (with videos) is in the subfolder 1.New_CDS141\2.2.UCSDneuro
There are lots of videos and papers on biology and medicine (and lots of tech) in the subfolder 0.Intro2Research.
Given the limited time and that I’m an extreme outlier, I’ll try to post videos (need to organize them) of some additional background material on aging, cancer, immune systems, wound healing, microbiome, etc that give some background on our approach to these topics.
Some videos/slides relevant to this meeting are in the subfolder: 1.New_CDS141\4.2.AgingSFI
Mercedes Pascual (Univ. Chicago), Qixin He (Univ. Chicago) - Session II: Immune System: Aging and Heterogeneity[edit source]
In several important pathogens, high prevalence occurs under widespread but incomplete immunity. This is the case for Plasmodium falciparum in highly-endemic regions of Africa, where asymptomatic infection occurs in individuals of all ages despite repeated infection. This large reservoir of infection constitutes the main challenge for elimination efforts and is enabled not only by the existing antigenic diversity of the pathogen, but also by the constant turnover of new variants. With an agent-based model of malaria and some analytical considerations, we present a novel threshold in transmission intensity that concerns the ability of the pathogen to diversify locally. We discuss how this aspect of the complex eco-evolutionary dynamics of transmission can be exploited for intervention efforts. We raise the open question of whether traditional epidemiological models that incorporate host age can be extended to capture this threshold.
Katie Gostic (UCLA) - Session II: Immune System: Aging and Heterogeneity[edit source]
Immune imprinting to the influenza viruses encountered in childhood strongly shapes lifelong risk. These findings raise questions about the expansion of antibody repertoires over time. Does the first influenza exposure, or the first few, have the greatest influence on lifelong immunological trajectories? Why have older cohorts evidently failed to develop strong memory against strains that emerged later in in their lifetimes? What are the costs and benefits of childhood imprinting for effective protection later in adulthood?
Sarah Cobey (Univ. Chicago), Phil Arevalo (Univ. Chicago), Marcos Viera (Univ. Chicago) - Session II: Immune System: Aging and Heterogeneity[edit source]
Antigenic mapping is an important technique used measure and visualize differences between viruses, but understanding how these maps change given immune system variation between individuals remains challenging. Do the immune systems of different individuals see the same virus differently? How does this perspective change as individuals age and experience different sets of viruses in different orders? We briefly touch on some approaches we've taken to answer these questions and suggest how exposure history might affect both individual immune responses and epidemiological patterns.
Micaela Martinez (Columbia Univ.) - Session II: Immune System: Aging and Heterogeneity[edit source]
I will be summarizing, and thematically integrating, two areas of research presented at the first SFI working group meeting. First, I will discuss the arrow of time, from birth to death, as seen through the lens of the immune system: from the development of the infant immune system through immunosenescence and the end of life. Second, I will review the internalization of time via two mechanisms: biological clocks and the "jamming" of immunological memory. Lastly, I will propose a conceptual framework for integrating these themes as spiral trajectories of endogenous immunity and vulnerability.
Andrea L. Graham (Princeton) - Session IV: Short Talks for Late Arrival[edit source]
Research on laboratory mice has provided much of what we know about the fundamental biology of the mammalian immune system. Yet because so few life-long experiments have been conducted on mice, we know remarkably little about immunosenescence in “the model mammal.” Classic work on Biozzi mice is an important exception. I will describe some of the experiments and key insights of the work of Biozzi and colleagues in the 1960s-1980s, especially on links between antibody responsiveness and organismal longevity.
David Schneider (Stanford) - Session IV: Short Talks for Late Arrival[edit source]
We have a data set that follows mice suffering from malaria from start to finish. We’ve looked at the microbiota, circulating immune cells, cytokines and metabolites to produce a time series that follows about 800 variables. We can map many of these, like the metabolites, onto function based networks that were worked out decades ago. We can also make networks de novo based only on the data. We added variation to this system by measuring these parameters across 8 different mouse strains that show extreme variation in their survival as well as testing aged mice for one strain. The problem I now face is showing how these networks vary over strain space and age in a way that helps the viewer understand the biology behind these changes. Should we be modeling the trajectory of the infections through interesting phase spaces? Should we be observing how the networks change over time and genetic space, and how should we do that?
Shenshen Wang (UCLA), Jiming Sheng (UCLA) - Session IV: Short Talks for Late Arrival[edit source]
In this collaborative project, we seek to understand various observations on immunosenescence, such as the breakdown of innate-adaptive collaboration and increasing variability of individual performance later in life. Earlier theoretical works have yielded much insight on the capacity of innate and adaptive immunity separately, yet these models with only one arm of the defense cannot explain the observed inflamm-aging (aging with inflammation). By considering the crosstalk between innate and adaptive responses, we build an integrative model that shows promising results consistent with experiments. Our preliminary findings highlight potential determinants of individual fates as well as the timing of inflammation dominance.
Post-meeting Summary by Organizer[edit source]
The group made progress toward four different areas:
Innate/Adaptive collaboration: The innate and adaptive components of the immune system do not age independently. We have now developed a mathematical model that couples these two complementary responses, and have demonstrated a mechanism by which the progression of immunosenescence in the adaptive response leads to aging in the inflammatory response. We began analysis of the innate-adaptive interface of this coupled immune model by studying cytomegalovirus infection, a persistent infection that interacts with both components of the immune system. Advances and discussions during the meeting included: improvements to the model equations; visualization of the history of system states in phase plane, for comparison to experimental data and potential warning signs of failure (death); consideration of the (‘shape space’) structure of antigenic variation in an ensemble of pathogens; and possible extensions to the epidemiological level (for a population of hosts).
Imprinting of acquired immune memory: consequences for maps of antigenic variation : Many pathogens consist of multiple strains that are perceived differently by the immune system. This variation can be represented concisely in “antigenic maps”, where distances between strains can be used to quickly estimate immunity to a new strain in hosts previously infected with another. However, the data and mathematical tools used to construct antigenic maps have so far ignored that hosts with different infection histories often perceive pathogens differently. For example, on exposure to a new strain of influenza, the immune system prefers to deploy memory responses that target familiar parts of the virus, instead of generating de novo responses that target unfamiliar antigenic structures. Thus, across a series of lifetime exposures, immune memory can become hyper-focused on the few parts of the virus that have not changed since childhood. And because influenza evolves antigenically over time, individuals born during different eras can develop immune repertoires that become focused on different parts of a given virus. Existing antigenic maps do not take this sort of historical contingency into account, and therefore fail to reflect population structure in how different birth cohorts perceive specific antigenic mutations. So far, at least two examples have been documented in which existing antigenic maps perceived a novel strain as similar to previously circulating ones when in fact mutations in the new strain caused antigenic escape (and unusually large numbers of cases) within some birth cohorts whose immune memory had failed. So far, cohort-specific immune escape has only been identified post-hoc, in response to outbreaks that have caused unusual age-specific impacts. By developing new theory, we aim to pave the way toward new antigenic mapping methods, with the eventual goal of predicting cohort-specific antigenic escape before it happens, and guiding preemptive vaccine updates. We are developing a simple model to compare a simulated virus’s “true” antigenic phenotype (based on viral structure, represented in shape space), with its “perceived” antigenic phenotype (a function of antigenic structure and host immune history). We will compare our new, history-dependent antigenic models with history-naïve models to ask, how often and how egregiously maps that do or do not consider immune history disagree. We will ask whether immune memory follows specific paths toward failure (antigenic escape), and whether specific paths to failure are particularly likely to remain undetected in history-naive antigenic maps.
Allometrically scaled immune responses. Body size scaling has a significant history at SFI; we have been using insights from these approaches to develop simple models for the immune systems of vertebrates that might be used from mice (the standard lab model), through sheep, humans, cows and horses to elephants. (We will most certainly take some digressions into bats and sea birds). Our approach combines the development of simplified very general models of an immune system, with literature surveys that examine how different components of the immune system function in hosts with different body sizes. Some key questions we are addressing are: (1) How does the trade off between Type I , cellular, and Type II acquired Immunity change for hosts with different life expectancy? If you only live for a short time, why should you develop any form of acquired immunological memory? Why not treat everything as a chronic infection? An alternate way to ask this question is to examine how the component cells of the immune system change in size in hosts with different body sizes. Are T-cells and B-cells always the same size or do they increase in size in larger bodied hosts as these individual components need to live longer to conserve memory? Or do longer lived hosts simply produce more of them?
Endemism and eco-evolutionary dynamics of pathogens with extensive antigenic diversity: We are motivated here by malaria and other important infectious diseases of humans and animals that achieve endemism in regions of high transmission, and are characterized by high prevalence with a large fraction of asymptomatic, but still infectious, individuals of all ages. Immunity built from repeated exposure as hosts age remains only partial and protects against severe disease but not infection. These pathogens rely on extreme antigenic variation typically encoded by multigene families to evade the immune system. Extreme antigenic variation underlies the extensive reservoir of transmission present across the host population, which makes Plasmodium falciparum and other similar pathogens so resilient to intervention/elimination efforts in endemic regions. In work that predates the SFI meetings, we have been developing theory for the eco-evolutionary dynamics of P. falciparum from the perspective of one important multigene family (known as var) encoding the major antigen of the blood stage of infection. At SFI, we have discussed analytical and computational results on a new threshold in transmission intensity whose crossing sharply modifies the ability of the pathogen to sustain antigenic diversity. At this second meeting, we specifically formulated a novel epidemiological model (as a system of partial differential equations) that considers host age and pathogen genetic diversity in synthetic form. This formulation is intended as an abstraction of the epidemiology in the high-dimensional agent-based model of malaria and var genes. It goes beyond extensions of the known SIR (Susceptible-Infected-Recovered) models that represent partial immunity but do not account for pathogen diversity. Initial bifurcation results show the possibility of a threshold response to parameter changes. We are addressing whether the PDE model can capture major features of the more complex agent-based one. This work involved Andre de Roos who was visiting SFI at the time of the meeting.
Reference Materials by Presenting Attendees[edit source]
Andrea L. Graham (Princeton)[edit source]
General Meeting Reference Material[edit source]
|Title||Author name||Source name||Year||Citation count From Scopus. Refreshed every 5 days.||Page views||Related file|
|Waning immunity.||Science (New York, N.Y.)||2019||0||2|