Andras Hajnal, M.D. Ph.D. - US grants

In collaboration with the Waksman Institute for Molecular Genetics at Rutgers University, the DIMACs Center will run a research seminar in computational biology in July and August 1992. Topics will centeraround molecular genetics, and will include fragment assembly, artificial intelligence in molecular biology, informatics, pattern identification for genetic regulatory regions, DNA and protein structure predictions, the detection of distantly related proteins by sequence comparison, protein folding, and self-organization and adaptation of complex systems.

Although powerful tools for numerical, statistical, and algebraic computation are readily available and widely used, a comparable tool for combinatorial computing has not yet emerged. The lack of such a tool seriously impedes interaction between theory and practice in discrete mathematics. Building on shared experiences, a comprehensive software system for combinatorial computing that is efficient, portable, and extensible will be created. The core of the system will be an object-oriented library containing reusable implementations of state-of-the-art combinatorial algorithms and data structures. To facilitate experimentation, the workbench will have a powerful graphical user interface and a versatile command-interpreter based on a functional programming language. The tool is to be built and distributed as a self documenting or `Literate` program. This new software environment facilitates research in theoretical, experimental, and applied discrete mathematics.

ABSTRACT NCR-9527163 Hajnal, Andras Rutgers University MDC: Simulations of Integrated Communications Systems The rapidly developing synthesis of communications and computing technologies, the ongoing construction of immense digital data networks, and the global deregulation of telecommunications industries will have a tremendous impact on the national economy and infrastructure. This proposal brings together leading experts in network modeling, network design and analysis, and high performance discrete-event simulation to focus on the development and validation of high-performance modeling and simulation tools for broadband networks and wireless communication services. The project will apply these tools to cutting edge problems in network design and operations that cannot be answered otherwise. The goals of this project are to: (1) develop network description methodology, similar in spirit to VHDL (very high speed integrated circuit hardware design language), suitable for the description of current and future telecommulication networks (2) provide validated high-performance simulation software prototypes based on this methodology, and (3) demonstrate the use of the tools for network construction and operation problems that cannot be answered otherwise. An important aspect of this project involves education and training of graduate students and post-doctoral fellows by involvement in pressing industrial problems, and via sharing the high-performance modeling and simulation tools within the public domain. The multidisciplinary project team attacks the problem by covering all of its critical elements. The team includes applications scientists who will help shape the network description and simulation tools by their experience and knowledge of current research directions in networking, and who will use the tools to explore those directions: it includes network modeling experts who contribute mathematically valid network behavioral models that help to accelerate the simulations ; it includes experienced developers of high performance simulation tools. Most critically, all members will participate vigorously in the design of the network specification methodology, for that methodology must address the needs and concerns of each. We envision a role analogous to the academic and industrial research partnership which successfully fostered the rapid expansion of the computer-aided-design industry in microelectronics. Indeed, we employ some of the proven strategies critical to that expansion - standardized description methodologies and highly efficient simulation tools.

The proposal consists of research plans going in four main directions. In a past work, the investigator proved the strongest known theorem about true embedding partition relations. For every finite graph H and for every cardinal t there is a graph G "arrowing" H with t colors. Recent results give some hope of extending this result to countable target graphs and developing a "Nesetril-Rodl" type Ramsey theory for partition relations with infinitely many classes. The second direction is the investigation of polarized partition relations using a far reaching new method called "double ramification" developed jointly with Professor James Baumgartner of Dartmouth College. The third direction is a continuation of a successful collaboration with Professor Peter Komjath of the Eotvos Lorand University of Budapest, Hungary. Extensions of partition relations will be investigated- for the case in which noncomplete graphs are partitioned, a theory of simultaneous chromatic number of graphs and set systems will be developed. Theqe will involve forcing methods and new proof techniques. The fourth direction is a continuation of a long standing collaboration with Professor Istvan Juhasz of the Mathematical Institute of the Hungarian Academy of Sciences. Specific targets will be investigations of partitions destroying the topology of a space, and study of invariants of CCC structures. Combinatorial problems and methods have proved to be a powerful tool in inspiring new developments in set theory, and new metamathematical methods have led to spectacular advances in set theory and its applications to set theoretical topology. The project plans to enhance this fruitful cooperation.

DESCRIPTION (from applicant's abstract): Gustatory stimuli, particularly those that taste sweet to humans, initiate and maintain ingestive behavior. They also provide reward sufficient to support the learned behavior used by most animals to locate food. Reward mechanisms related to feeding (and other biological drives) have been linked to the mesolimbic dopamine (DA) system in general, and to the nucleus accumbens, specifically. Nevertheless, the function of accumbens DA in reward remains unclear. Taste provides a logical tool for examining central reward mechanisms. Sapid chemicals are quantified easily, they are transduced via a single, well-described sensory channel, and they can be disassociated from their nutritional consequences. Conversely, such testing of the central reward systems can shed light on a central conundrum of gustatory physiology -how the hedonic qualities of tastes can be reconciled with their purely sensory characteristics. Despite the substantial overlap between taste and reward, relatively little experimental attention has been focused on the interaction of these systems at the neural level. The present experiments are designed to: 1) investigate whether sapid sucrose activates the mesoaccumbens DA neurons in a concentration dependent manner during sham feeding; 2) evaluate the effect sucrose licking on NAc DA during chronic weight reduction, as a model of reward sensitization; and 3) examine the contributions of DI and D2 dopaminergic mechanisms of the NAc in the mediation of both the behavioral and neurochemical activation related to sucrose ingestion in both normal and underweight rats. Unlike previous sham-feeding studies that used systemic injections, dopaminergic drugs will be perfused locally in the NAc through reverse microdialysis. The long-term objective of such an investigation is to understand how food-related oral stimuli produce reward. This knowledge is directly relevant to understanding the control of normal ingestive behavior and to the maladies related to this essential behavior, such as obesity, anorexia nervosa, bulimia nervosa, and binge eating disorder.

DESCRIPTION (provided by applicant): Obesity causes or exacerbates many chronic illnesses, most notably non-insulin-dependent diabetes mellitus (NIDDM). Most obesity is caused by modest, but chronic overeating. Otsuka Long-vans Tokushima fatty (OLETF) rats, which lack the CCK-A receptor, are hyperphagic, obese, and gradually develop NIDDM. In OLETF rats, increased food intake is necessary for the development of obesity, suggesting that the NIDDM is secondary to the prediabetic hyperphagia. Thus, OLETF rats are reasonable model of the most prevalent form of NIDDM in humans. The underlying cause of the chronic hyperphagia in this strain is unknown and cannot be explained entirely by their peripheral satiation deficits. Rather, a dysfunction in central pathways critical to the control of meal size is the most likely contributor. In this project, OLETF rats are used to study the relationship between the hyperphagic behavioral phenotype and dopamine (DA) signaling within the central motivational system during the development of type-2 diabetes. We propose that altered dopaminergic functioning in the mesoaccumbens dopamine (DA) system contributes to the overeating in OLETF rats by increasing preference for the orosensory stimulatory effects of normally preferred foods. Behavioral, neurochemical and histological methods will be employed to challenge this hypothesis. The application has four specific aims: 1) to characterize the basic dopaminergic phenotype (basal and stimulated DA release and reuptake) of the OLETF rats at three ages, reflecting the development of diabetes; 2) to characterize hyperphagic behavioral phenotype by investigating nutrient preference functions in prediabetic OLETF rats based on their orosensory and postabsorbtive properties; 3) to assess the relationship between behavior and DA signaling by comparing the effects of sham-feeding of preferred sucrose or fat solutions between prediabetic OLETF and age- and body weight-matched non-mutant control (LETO) rats; 4) to address causality of the relationship by using chronic treatment of the psychostimulant methylphenidate to reverse preference for and intake of sucrose and fat, and to delay onset of diabetes in OLETF rats. These studies will help determine how plasticity in the dopaminergic system affects behavioral and metabolic factors related to hyperphagia and the development of dietary-induced NIDDM.

DESCRIPTION (provided by applicant): Gastric bypass surgery (GBS) is an exceptionally successful therapy for morbid obesity and type 2 diabetes. GBS patients typically lose 25-35% of total body weight, demonstrate improvements in medical co- morbidities, and sustained weight loss over fifteen years. Given the epidemic of obesity in the United States, an improved understanding of the mechanisms by which GBS causes and maintains weight loss represents an important area of research. Although GBS mechanically restricts food intake, it also appears to reduce appetite and the appeal of savory meals. However, it is unclear why the motivational system fails to drive patients to compensate for this massive weight loss with increased food intake and preference for palatable, calorie-dense foods - the normal homeostatic response. Elucidating this paradox would substantially improve our understanding of the regulatory mechanisms for eating and body weight. We are in a unique position to address this question based on an animal model of GBS demonstrating alterations in the central neural mechanisms regulating food reward functions developed in our laboratory. The current study proposes behavioral, pharmacological, neurochemical and histological studies in high energy/high fat diet- induced obesity rat models to test the hypothesis that GBS alters appetite and food preference functions resulting in changes to the food reward system. The experiments target the nucleus accumbens, a critical structure for reward, with focus on two major transmitters: dopamine and acetylcholine. We propose four specific aims to test different components of this hypothesis. The first aim will establish the behavioral effects addressing specific aspects of food reward (i.e., incentive, reinforcement and hedonic value), and compare dopamine involvement in these behaviors across weight reduction methods (i.e. caloric restriction vs. GBS). The second and third aims will mechanistically address the underlying dynamic and static (i.e. neuroadaptive) signaling mechanisms, respectively. Aim 4 will investigate if increased gut-brain peptide signaling contributes to improved food reward functions following GBS. We believe the proposed research has significant potential to impact patient care as it will improve our understanding of factors that could positively or negatively contribute to long-term weight maintenance and could elucidate new targets for developing less invasive treatments for obesity. Preliminary data suggest GBS beneficially impacts the regulation of appetite and food choice resulting in more dramatic, sustained weight loss than dieting. The current study examines how GBS, in contrast to dieting, influences the rewarding effects of palatable food in the brains of dietary obese rats. Information concerning essential changes in motivated behavior and underlying neural substrates produced by GBS could assist in the development of effective non-surgical approaches to obesity treatment.

DESCRIPTION (provided by applicant): Obesity occurs because individuals regularly ingest more calories than they expend. This takes place when the reward value of the food is greater than the sum of the satiety signals arising in the gut. Nevertheless, the neural mechanisms that assign hedonic value to taste remain unknown. The neural substrates for the decision to whether ingest or reject a food are complete in the brainstem. In rats, the brainstem contains the first and second central relays for both the gustatory and vagal visceral afferent systems, the motoneurons responsible for ingestive behavior, and sufficient integrative capacity to make this decision. Nevertheless, control of ingestion requires connections to and from the forebrain because disconnecting it from the brainstem eliminates voluntary eating. Writ large, the goal of this research is to understand how the forebrain interacts with these hindbrain mechanisms to bring about the smooth, nuanced control of feeding behavior that characterizes omnivores such as rats and humans. Using behavioral, neuroanatomical, neurochemical, and electrophysiological analysis, this project will test hypotheses about how the reward value of gustatory stimuli is elaborated in the brain. The first Aim is to further determine the specific ventral forebrain projections from the parabrachial nuclei that are critical for modulating dopamine release as a function of the rewarding (and aversive) properties of taste stimuli. Aim 2 addresses functional correlates of those connections by comparing dopamine release in the nucleus accumbens and the central nucleus of the amygdala -- putative nodes in the forebrain reward system -- during homeostatic and hedonic ingestion. Aim 3 focuses on gustatory neural activity in the pontine taste relays while varying both oral and postoral taste stimulation. The findings will aid in understanding basic regulatory mechanisms that fail to control intake in individuals susceptible to developing dietary obesity due to increased stimulation from augmented palatability of the modern diet.

? DESCRIPTION (provided by applicant): Obesity and its associated health consequences are among the main causes of preventable death. At present, Roux-en-Y gastric bypass (RYGB) surgery is a commonly performed and very effective method to achieve significant, long-term weight loss and to treat associated comorbidities. However, in contrast to improved food preferences and reduced food cravings following RYGB, clinical reports have revealed a concern for patients developing an increased risk for alcohol consumption. Whereas our research group and others have recently found increased alcohol preference and intake in dietary obese male rats that received RYGB suggesting a biological etiology (i.e., not confounded by psycho-social, and co-morbidity factors specific to humans), there is no clear evidence that these effects would eventually lead to increased risk for development of alcohol addiction. In addition, the ultimate neural mechanisms underlying how RYGB may increase and sustain motivation for alcohol use and potential contributing factors (such as postoperative dietary compliance, weight loss history and hormonal/metabolic improvements) warrant investigation. Moreover, no study has yet investigated alcohol effects in female rats despite the fact that >80% of RYGB patients are women. Thus, this high-risk high-gain, proof-of-concept R21 application will use high fat diet-induced obese, non-alcohol- preferring Sprague-Dawley female rats, believed to capture the most common environmental etiology and multigenic characteristics of obesity, and a proxy for the exclusion criteria of heavy drinkers by most bariatric surgery centers. The central hypothesis at test is that RYGB increases preference for and intake of alcohol based on its increased rewarding effects, and in turn, poses an increased risk for development of addiction. Regarding the underlying mechanism, we propose that this effect is due to alleviated obesity-related deficits in the brain dopamine systems due to yet unknown unique effects of surgery, i.e., independent of weight loss or post-surgical change of diets. Aim 1 will use a comprehensive battery of behavioral tests to investigate the hypothesis that RYGB increases alcohol-seeking and -taking behaviors and results in increased vulnerability to alcohol addiction independent of weight loss and change in diet. Functional changes in brain activity after RYGB vs. caloric restriction to conditioned alcohol cues will be assessed using in-vivo brain imaging (positron emission tomography, µPET). Aim 2 will extend investigations to underlying mechanisms by testing the effects of surgery on functional and static measures of dopamine signaling, and vice versa: the effects of dopamine receptor manipulations on alcohol-related behaviors. These studies are expected to provide initial data for a future R01 application with a focus on specific pathways and pharmacological targets upstream to the dopamine reward system. This pre-clinical translational study is of high impact in that it will help clinicians to make personalized postoperative treatment plans for patients wit increased risk of alcohol use disorder and to prevent development of addiction.