1 Laboratory Research Program, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20824
2 Clinical Research Program, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20824
The National Heart, Lung, and Blood Institute "Symposium on Phenotyping: Mouse Cardiovascular Function and Development" was held on October 1011, 2002, at the Natcher Conference Center on the NIH campus in Bethesda, MD. In laying out the program for the meeting, we sought to bring together investigators with diverse background and expertise. Our goal was to encourage intellectual exchange between researchers from a wide range of disciplines, to shed new light on the complexities of mouse cardiovascular phenotyping. We were aided in our quest by an outstanding Program Committee which included Ken Chien, Axel Haase, Joanne Ingwall, Michael Parmacek, Roger Markwald, and Antoon Moorman. To this end, four topical sessions were assembled: 1) Imaging Modalities, 2) Functional Assessments, 3) Mouse Models of Human Cardiovascular Disease, and 4) Genomic Approaches. The included topics ranged from MRI, micro-PET, and ultrasound to telemetry, electrophysiology, and optical mapping of the conduction system. Speakers also described exercise modeling in the mouse, atherosclerosis, and the making of mouse models of DiGeorge syndrome. In the Genomics Approaches session, the use of laser capture microscopy, microarray expression analyses, and ENU mutagenesis was discussed. In addition, after each of the four sessions, an open forum for discussion was held on various topics including Imaging Modalities, The Problem of Anesthesia, Mouse vs. Other Animal Models, and Expression Profiling and Phenotype Databases. A poster session with 63 poster presentations covered a wide range of topics that mirrored the oral presentations.
On the whole, speakers presented a great deal of very exciting and cutting-edge science. They provided a very broad view of the present and future directions of mouse cardiovascular research. It is evident from the oral and poster presentations that great strides have been made in mouse cardiovascular phenotyping and that many new tools and mouse models are now available to study cardiovascular function and disease. Since in-depth reviews from many of the speakers appear in current (and subsequent) issues of Physiological Genomics, we will direct our comments mostly to the focused discussions held at the meeting. These discussions provided a palpable sense of the excitement in the field and also made evident issues that are of concern to investigators.
Participants in the Imaging Modalities discussion were in consensus that much already has been accomplished with technical advancements in various imaging modalities to facilitate mouse cardiovascular phenotyping. The talks and poster presentations provided beautiful examples were of the effective use of ECG-gated MRI, micro-PET, and ultrasound to analyze cardiovascular structure and function in mice. However, researchers expressed concerns regarding the availability and cost of accessing such technology. Specific costs included equipment acquisition and the availability of trained staff to run such facilities. An equally problematic issue is the cost of using such facilities, which can render the technology out of reach for many laboratories. There was disagreement as to whether there were sufficient centers to service demands and whether these were geographically well distributed. In discussing future directions, much enthusiasm was expressed regarding the power of molecular imaging, but it was generally agreed that further advancement in this technology is needed. A further complication is the chemistry required to support this imaging modality.
In The Problem with Anesthesia discussion, the general agreement was that "less is better." Michael Parmacek (University of Pennsylvania), who led this discussion, nicely summarized data from the existing literature showing the varying effects of different anesthetics on mouse cardiovascular function (Table 1). These snapshots serve to highlight the increasing concerns of the research community regarding studies carried out using different anesthetic protocols. It was generally agreed that this issue needs more careful consideration in the future. Concerns were also raised on awake protocols, that is, protocols in which mice are not anesthetized. Apparently, the incorporation of a "training" period for mice to acclimate to awake protocols may not eliminate a cardiovascular response; thus measurements obtained with tail-cuff measurements are said to differ from those obtained from telemetry.
|
Near the close of the meeting, a discussion was held on the need for establishing databases on gene expression profiles and cardiovascular phenotypes. Given the large number of mouse models already generated and the likelihood of many more to come, there is a real need for searchable databases that will allow ready access to this large body of information. This will most certainly propel the field forward, perhaps accelerating new discoveries. In the genomic field, sequence, SNP, and linkage databases have been invaluable in propelling new discoveries at a pace hitherto unimaginable. However, the situation with cardiovascular phenotyping is clearly more complicated. After all, it is not merely a matter of 4 nucleotide bases and 20 amino acids, which is complicated enough. In approaching this task of databasing cardiovascular phenotypes, there are four central questions that need to be addressed. First, what type of data should be included in such databases? There is such a variety of data being generated, and the research community must rank the potential value of different types of information. Some examples of data that may be included are microarray gene expression profiles, developmental gene expression patterns in three-dimensional reconstructed hearts, and physiological measurements of cardiac function. A second issue intimately intertwined with the first is the need to standardize experimental protocols and analysis platforms so that data from different laboratories may be compared and merged. The third question is how such data are to be organized. This will undoubtedly depend in large part on how these databases will be used. Finally, there is the question of which institutions and resources may be brought to bear for this enormous task. Having a "curator" is essential for the long-term viability of such databasing efforts. These are clearly complicated issues that will need further discussion by the larger cardiovascular research community. The enormity of the task suggests that what is needed is not one database, but the assembly of a series of separate but linked databases, each individually focused on different types of information. Seigo Izumo showed very impressive progress made by his group and other members of the Program for Genomic Applications (11 multi-institutional programs funded by the NHLBI) on setting up searchable databases for archiving data collected from a variety of mouse models exhibiting cardiovascular defects and disease. Overall, all agreed that the construction of cardiovascular phenotype databases is very worthwhile, although the logistics remain to be addressed in the future.
In summary, this meeting provided a forum for considering a wide range of topics and brought together a unique mix of investigators that would not often be found at the same meeting. In closing, we would like to address briefly the question some of our colleagues raised: "Why focus the meeting on the mouse exclusively?" We appreciate the fact that the mouse, given its small size and differences in physiology, can be problematic as a model for human physiology. Nevertheless, mouse models have become a central focus of intense research activity in the cardiovascular field, with new transgenic, chemically mutagenized, and disease models being generated at a rapidly increasing pace. By focusing on mice, we hope to explore issues relating to mouse modeling of cardiovascular development and disease more thoroughly. Nevertheless, it was clear that even with 17 scheduled talks and 63 poster presentations, many areas of mouse cardiovascular phenotyping were still left uncovered. We encourage those who did not attend the meeting, as well as those who did, to read the reviews in this and subsequent issues of Physiological Genomics. These reviews provide more in-depth consideration of the methods and models available for analyzing cardiovascular function and development. No doubt these reviews will also help point to the future directions of this rapidly growing field.
FOOTNOTES
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).