Putting molecular disease research into physiological context

Abstract

Researchers seek novel ways to promote whole animal physiology and link molecular studies of disease with systems biology. Nicole Garbarini investigates.

In a 2002 lecture to the American Physiological Society (APS) Respiration Section, former APS president Norman Staub, Sr. presents, as his first slide, two images. The first image is labeled ‘Molecular Biology’ and features a gleaming silver bullet train speeding into the distance. The second is labeled ‘Clinical Medicine’, and depicts a rickety wooden caboose standing alone. Staub uses this simplified metaphor and visual aid to comment on the state of the union – or, perhaps more accurately, disunion – between clinical medicine and molecular biology (Staub, 2002). Given the exponential increase in molecular biology advances, Staub posits the question, “Why has the potential for applications of molecular biology in clinical medicine not been realized?” He points to whole animal physiology as the discipline that connects clinical medicine with molecular biology. He suggests that basic science’s inefficient translation to the clinic, results from its limited ability to shift its findings from a restricted and controlled experimental environment to the complexity of a complete organism.

The decline of whole animal physiology from its heyday in the mid 20^th century is a trend that has been noted by many researchers. The neglect of systems biology and whole animal physiology has been linked to a wide variety of dead ends in scientific research. These include missing clinically important phenotypes in transgenic or knockout mouse models, which may be subtle or difficult to detect for the average biomedical researcher untrained in animal physiology, as well as wasted effort and expense incurred during novel drug development. Of the millions of dollars spent developing potentially effective drug therapies, a large expense comes from the waste incurred when novel compounds that pass tests in laboratory animal models fail in clinical trials, perhaps due to a limited understanding of the physiological relationship between model organisms and human patients.

Several initiatives are addressing this gap in the research process. David Wasserman, Professor of Molecular Physiology and Biophysics at Vanderbilt University, is involved with bringing an understanding of whole animal physiology back to the forefront of research. Wasserman serves as director of the Mouse Metabolic Phenotyping Center (MMPC) at Vanderbilt, one of seven MMPC facilities in the country funded by the National Institutes of Health (NIH). These MMPC centers were established by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) in 2001, to provide researchers with an expanded array of information about their mouse models by using standardized testing and diagnostic methods to expedite characterization of mouse metabolic phenotypes.

The rat has been the traditional model organism for studies of physiology; however, the genetic tractability of the mouse and subsequent availability of reagents make mice tremendously popular for research (Fig. 1). As genetic and molecular manipulation techniques in the mouse continue to expand, the limitations in the physiologic tests that are necessary to identify mouse phenotypes have become more evident. “The focus of physiology departments over the last few decades has not been on physiology; it has been on cell biology, gene transcription and such similar topics,” Wasserman commented. Thus, Wasserman and colleagues are shifting the focus to physiology by applying their animal physiology expertise to the mouse. They have scaled down tests designed for larger organisms, making them appropriate for this smaller animal model. Their efforts have resulted in quality measurements of factors related to diabetes and diabetic complications. Researchers can send their own mice to the MMPC core for testing, which is advantageous since many metabolic tests are both difficult and expensive to set up in a laboratory from scratch. Wasserman and his colleagues are also working to establish a database for mouse models of diabetes. Additionally, they engage in outreach efforts by teaching a yearly course on insulin clamp techniques in mice, so that interested researchers can establish this popular test in their own laboratories.

Fig. 1. The genetic tractability of the mouse makes it popular for research.

Courtesy of Andrew Salinger and Monica Justice.

Beyond metabolic phenotype data, additional information is available to scientists as the result of large-scale mouse phenotyping centers and phenotype databases. The Mouse Phenome Database hosted at Jackson Laboratories (http://www.jax.org/phenome) provides protocols and information on normals for different mouse strains, as well as an extensive bank of phenotypic information compiled from multiple sources. A large-scale mouse program developed in the RIKEN institute research centers in Japan hopes to standardize phenotypic analysis of mice following N-ethyl-N-nitrosourea (ENU) mutagenesis (http://www.brc.riken.jp/lab/jmc/en/).

Perhaps the most advanced effort in providing a standardized phenotyping resource to a large community is the European Mouse Disease Clinic (EUMODIC). EUMODIC is a European Commission-funded consortium of 18 research centers in eight countries, which will undertake a coordinated effort to phenotype 650 mutant mouse lines. The battery of tests being performed by the EUMODIC researchers was generated from an earlier project, which ran from 2002 to 2006, known as the European Union Mouse Research for Public Health and Industrial Applications, or EUMORPHIA. This four-year project resulted in the development of standardized protocols to comprehensively study mouse phenotypes, and the protocols were published in a web repository known as EMPReSS (European Mouse Phenotyping Resource for Standardised Screens, http://empress.har.mrc.ac.uk/). This project also established a ‘phenome’ database dubbed Europhenome (http://www.europhenome.org/).

The development of standard operating procedures for mouse phenotyping is also being taken advantage of by investigators, who are seeking to generate genetically manipulated mice at their home institutions. Monica Justice, at Baylor College of Medicine, is helping to set up a phenotyping core at Baylor, patterned largely after the protocols and initiatives of the EUMODICS group. The core will feature a range of tests including analyses of bone density and dysmorphology; clinical chemistry and blood work; slit lamp and ophthalmoscope exams of the eyes; and tests of physical attributes, behavior, strength, and locomotion. Justice notes that the idea of standardization is promoted very strongly by the European Mouse Disease Clinic, in that researchers must demonstrate that their tests of control strains meet the norms established by the consortium. Such a requirement adds quality control to laboratory assays. Justice describes how, because human laboratory tests include checks for quality control, this is needed in animal research, and how it is now beginning to be implemented.

Training courses for scientists are yet another way to reintroduce whole animal physiology and systems biology to cell and molecular biologists. As an example, an NIH-sponsored course entitled ‘Models and Technologies for Defining Phenotype’ was hosted at Wake Forest University in 2006; the goal of the course was to train researchers in defining phenotypes and in the tests relevant to transitioning their basic scientific findings into translational research (Penn et al., 2007). Such courses serve as training modules, and bring physiologists and molecular biologists together to facilitate communication and collaboration. Indeed, collaborative efforts between molecular biology and physiology researchers can contribute greatly to increasing our overall understanding of whole animal physiology and systems biology.

Collaborative efforts between scientific researchers and veterinary pathologists is another avenue that is being examined for its potential to enable a more comprehensive understanding of animal physiology. A National Academies report published in 2005 carefully documented the need for increased participation by veterinary doctors in biomedical research (Committee on the National Needs for Research in Veterinary Science & NRC, 2005). The report emphasizes the training of Doctors of Veterinary Medicine (DVMs) for work in biomedical research, and highlights the usefulness of increased collaborations between veterinary pathologists and scientific researchers to increase the understanding of, and care for, animal models of disease and of animals used as test subjects in the laboratory. Michael Lairmore, Chair of the Department of Veterinary Biosciences at Ohio State University, commented that veterinarian training on its own makes DVMs prime candidates for understanding differences and abnormalities in the anatomy and physiology of animal models of disease. Expert understanding of animal physiology is valuable for recognizing changes that occur in animal models of disease, as well as in evaluating the response of models used for drug discovery. He notes, however, that additional training beyond the standard comparative physiology curriculum is needed at the post-DVM level. “They [veterinarians] then become more focused in an area, such as pathology, and can go much greater in depth,” Lairmore commented. Many universities and research centers are now recognizing the value of hiring their own highly trained DVMs to work alongside their scientists to better understand genetic and disease models.

Recognition of the importance of whole animal physiology is now leading scientists to re-examine biological indicators using the new tools generated by advances in genetics, and cell and molecular biology. One example of this is the study of biomarkers, which are objectively quantifiable factors that signify the presence of a pathological process without relying on symptom information or a pharmacologic response to therapy. Biomarkers can include a variety of factors, such as certain gene transcripts, protein measurements, or metabolic intermediates. Robert McBurney, Senior Vice President of Research and Development, and Chief Scientific Officer at BG Medicine, studies biomarkers and develops bioinformatics tools in order to aid efficacious drug discovery and development. “In drug discovery, animals don’t usually exhibit the same sort of symptoms [as humans] but they may have the same molecular underpinnings in a disease model.” McBurney commented, “We want to return to a more holistic or systems-based analysis, but the only way that we can really do that is if we focus not on symptoms, but molecular information.” Thus, McBurney and colleagues are collaborating with the FDA and several pharmaceutical companies to discover, and develop the use of, biomarkers for liver toxicity in order to screen for drugs that pass pre-clinical tests in the laboratory, but fail in the clinic due to liver toxicity issues. Such biomarker studies could enable scientists to unite their newfound knowledge of molecular biology, biochemistry and genetics and explain the implications of this knowledge in animals or humans at the systems biology level. McBurney notes, “This is a return to whole animal physiology and pharmacology, but with much more granular, molecular measurements… We are moving from symptom-based diseases to molecular-based diseases.”

What makes a return to investigations of whole animal physiology particularly ideal now?

Robert McBurney’s view is that this union of molecular biology and systems biology is largely driven by advances in technology and information, such as computing power and sequencing of the human genome. Additionally, he suggests that rising healthcare costs are yet another reason why disease investigation and drug discovery need to be optimized. He agrees that increased understanding of biology and physiology on the level of the whole organism will help this to become a reality.

Likewise, Monica Justice agrees that the renewed interest in whole animal physiology is due to our current position in the timeline of progress, and is built on the advances of the last few decades. “Now is the time that we’ve developed the technologies to look at whole animal physiology,” Justice remarked, “We have all this molecular information at our fingertips and people are finding out that you can’t study something in isolation. It happens in the context of the whole complex organism.”