© 1998 by European Society of Cardiology
Copyright © 1998, European Society of Cardiology
Of mice and men – the future of cardiovascular research in the molecular era
Max-Planck-Institute, Dept. of Experimental Cardiology, Benekestrasse 2, D-61231 Bad Nauheim, Germany
* Corresponding author. Tel.: +49 (6032) 705 402; Fax: +49 (6032) 705 419; E-mail: wschaper@kerckhoff.mpg.de
Received 22 January 1998; accepted 19 February 1998
KEYWORDS Genomic analysis; Integrative physiology
 |
1 Introduction |
|---|
 Top 1 Introduction 2 A look at...3 The importance of...4 ConclusionsReferences |
|---|
The advent of the molecular age has changed experimental cardiovascular research and continues to make these changes deeper and more and more irreversible by the day. Gradually over the last 12 years dogs and cats, the classical experimental animals in cardiovascular sciences, became unfashionable for a variety of reasons, some of them nonscientific (i.e., new legislation, expense, also to avoid clashes with animal rights activists etc.). They were replaced by rabbits and rats (less expensive, but not necessarily better suited) and, more recently, by transgenic mice. This description of the changing preferences for experimental animals shows that not all of these changes have their roots in the new opportunities of molecular approaches. That was to be expected because not all cardiovascular problems can be reduced to exercises in gene expression. Acute effects and most pharmacological problems are not necessarily solved utilizing the gene approach, although screening for new pharmacological agents uses more and more genetically altered cells and animals.
Gene expression studies are most appropriate when studying chronic adaptation of tissues to a change in the environment of cells or organs, like pressure overload, or the growth of new blood vessels or the adaptation to chronic hypoxia, to name only a few.
Molecular biology assumes its prominent place today because it promises a paradigm shift in modern medicine, namely that chronic degenerative diseases are amenable to causal treatment and that structural changes can be elicited by substituting failing or weakly expressed genes or by substituting mutated genes with the correct copy. It can be envisaged that atherosclerosis and its organ manifestations will be avoided by gene therapy and that already existing or threatening arterial occlusions can be compensated for by the stimulation of angiogenesis and arteriogenesis. Since these new methods will in all likelihood be less expensive than surgery and traditional pharmacotherapy, cardiovascular medicine will become socially more acceptable.
We need these changes in approach because traditional cardiovascular research is no longer on the A-list for support by politicians and granting agencies, not only in Europe but also in the USA. This has its roots probably in the enormous success that the more technical-oriented clinical cardiology had enjoyed over the last 20 years and which may have created the illusion that cardiovascular disease is under control and no longer in need of the rather generous funding that we were used to for the last three decades. That the problem is not under control is highlighted by the fact that these high-technology approaches treat only the manifestations and the symptoms of the disease but not its cause. This concentration on technical solutions may have delayed the molecular approach that was introduced much earlier in cancer research. Furthermore, this approach in spite of its success is taxing the health budget and will become unaffordable by a large percentage of the population.
|
2 A look at the problem |
|---|
|
Top1 Introduction2 A look at...3 The importance of...4 ConclusionsReferences |
|---|
The American Heart Association and agencies for life statistics have published data that highlight the size of the health problem and the associated costs. Currently 30 billion dollars are spent each year for cardiovascular drugs on the American market, and the combined European market will be of about the same size. Over 570 000 coronary bypass operations were performed in the US and only slightly fewer in the EU, i.e., about one million with such operations being performed at an average cost of 25 000 dollars each. About the same number of angioplasty procedures are being performed at a cost of 15 000 dollars each and about 6 million hospitalizations are necessary to treat cardiovascular disease. Drugs, surgery and hospitalizations together amount, for the USA alone, to about 100 billion dollars. A similar amount must be assumed for Europe. And all this to treat symptoms and consequences rather than the causes of the disease. So far the health burden of another organ manifestation of vascular disease, i.e., ‘brain attacks’ (strokes), has not been mentioned: in 1993, 150 000 patients died of stroke in the US and about 4 million stroke survivors are now living with lasting disabilities; the number of people surviving a stroke is about 600 000 patients per year. Similar estimates may be assumed for Europe [1]. It is clear that it is high time for a paradigm shift.
2.1 Where is the place for the integrative cardiovascular scientist?
While warming up to the exciting new paradigm of creating a biopharmacology which produces structural changes in the tissues via manipulations of the genome rather than the traditional acute effects of classical pharmacy, we have to take stock of our inventory and decide which of our animal models and experimental techniques are fit for the new age. Furthermore it has become obvious that the molecular biologist needs the cooperation of the cardiovascular scientists. Cells in culture as test systems cannot replace heart and blood vessels in vivo, new medicines and methods need verification in intact animals. This does not mean that physiologists are only useful when they confirm the molecular approach, the interesting scientific questions should originate at the bedside and in the experimental laboratory. This partnership between molecular biology and integrative physiology should be on equal terms. It should not be one of master and servant as had happened before. A famous Nobel laureate in Biochemistry, when asked about the source of his success with the isolation of neuropeptides, is quoted to have said: "I had good physiologists". But these remained largely unknown. The opposite is also true. "What do you do when you need the gene?" an integrative physiologist was asked: "I hire a slave!"
2.2 Are we using the right models?
It is clear that the right mammalian model for future cardiovascular studies is that which allows the manipulation of the genome. Considering expense, gestation times and technical feasibility (availability of embryonic stem cells, ability to see and reach the nucleus with a pipette) the mouse is almost the only alternative. The technical difficulties connected with performing experiments and measurements in the subminiature-sized murine cardiovascular system are enormous but the article by Doevendans and colleagues in this issue show that these difficulties can be overcome. Great hopes are focused on the availability of noninvasive techniques like echocardiography, nuclear magnetic resonance and infrared thermography but although these are promising techniques, they are also expensive and require extreme specialization.
However, this is truly at the cutting edge, because a recent literature survey shows that for three cardiovascular topics (heart failure, hypertrophy and ischemic preconditioning), mice are not yet an accepted model.
| |||||||||||||||||||||||||||||||||||||||||||||
(n=number of papers, survey covered 1992 to 1996).
We see from this survey that the most popular models are rats and rabbits, a choice which is dictated more by convenience than by science-based preference. We predict that this will radically change within a very short time.
|
3 The importance of size |
|---|
|
Top1 Introduction2 A look at...3 The importance of...4 ConclusionsReferences |
|---|
Animal size is an important, sometimes puzzling and often unnoticed variable in experimental research. It becomes important with the rapidly increasing popularity of the mouse as an experimental animal and the question arises whether results obtained in very small animals can be extrapolated to man. We have to bear in mind that cell size is practically invariant and that large and small animals differ only in cell number. Furthermore the duration of the cell cycle is also more or less constant, once G1 is entered. This means that the result of an experiment where cell number is concerned may differ greatly depending whether the experiment was carried out in a large or small animal: the same functional result requires many more cell divisions in large animals. Arteriogenesis (collateral artery growth) in a canine heart transforms a small preexisting arteriole of about 40 micrometers into a small artery of 800 micrometer [2, 3]. This is a 20-fold increase in diameter and a 50-fold increase in tissue mass. In the mouse a 20-micrometer vessel needs to be transformed only into a 100-micrometer artery which is only a 5-fold difference requiring fewer cell cycles and is completed much faster. It is therefore difficult to predict whether results obtained in mice are directly applicable to humans. The higher basic metabolism of small animals may exert a correcting influence: although vascular growth needs less time, ischemic tolerance may be less.
Another pertinent problem is infarct size. Even in an organ with no measurable collateral blood flow, as in the pig heart, infarct size is never 100% because diffusion of oxygen from the lateral borders, from the ventricular cavity and, in open chest experiments, from the epicardial surface salvages about 15% of the risk region [4]. Since the diffusion of oxygen in myocardial tissue is a physical constant it can be assumed that a similar absolute amount of tissue is reached in the mouse heart (10–4 smaller than that of the pig) by oxygen diffusion which may result in a relatively larger salvaged region. Again, the higher oxygen consumption in the mouse may exert a corrective influence in that the oxygen supply by diffusion will salvage fewer cells, but outcome still remains unpredictable and extrapolation to man risky.
The smaller volume of distribution of the body of a mouse creates also problems in drug dosage especially with topical application as in the treatment of tumors: the differences between local and systemic concentrations are much smaller in the mouse compared to larger animals.
3.1 What can we learn from structural genomics?
It is obvious that the predictability of experiments and its extrapolation to man is a function of the genetic kinship: the closer a mammal is related to man, the more relevant will be the experimental result. However this paradigm runs into difficulties in the realm of ethics and expense because we cannot do any large scale experiments on chimpanzees, our closest cousins. Unfortunately most other affordable mammals are more or less equally distant from man and it does not appear to be profitable to search for common genetic features and to base experiments on that. A point in case is the often repeated ‘similarity’ of the pig and human heart.
This similarity is based solely on the anatomical distribution of the right coronary artery that supplies the posterior wall of the left ventricle in most pigs and in a significant percentage of human hearts. In other aspects, like substrate preference or the ability to recruit smooth muscle for the assembly of new arteries, the porcine heart differs significantly from that of humans.
Although it is interesting to try a comparison of the genomes of classical experimental animals like dog, cat, pig, and guinea pig with that of man in order to determine degrees of kinship, the basis of comparison is too small. In spite of the enormous progress that was made in structural genomics, most of the known DNA sequences are based on the human and on the murine genomes. A search in the relevant databases showed that only 107 amino acid sequences are known from the dog and only 107 from the guinea pig, both popular experimental animals in cardiovascular research. This is less than 10% of the known murine sequences. There is hope though, because a new project has started to sequence the complete canine genome in order to tap the enormous resource of this species that varies so tremendously in physical shape and behaviour and is so similar with regard to human pathology that much can be learned from it about morphogenesis and about the genetics of degenerative diseases [5]. We have examined whether a more general comparison between humans and mice could be obtained. Therefore, a computer program was written which extracts all comparable amino acid sequences of humans and mice from the protein data bank SWISSPROT (11/96). This was accomplished by selecting all successive amino acid sequences from the mouse data subset and testing whether a corresponding amino acid sequence is available in the human data subset. The comparison of all selected amino acid pairs was performed in batch mode with the COMPARE command of the GCG software package (GCG, 1997) [6]. A total number of 1135 amino acid sequences was extracted and compared. The comparison of these sequences resulted in an overall homology of 85.7±13.6% and an identity of 82.5±15.7% between human and mouse proteins. These results may or may not be surprising. However, the interesting conclusion might be that the variety of species might be anchored only in the remaining 15% difference of the proteins of humans and mice.
The accumulated pairs of identical proteins of humans and mice can be ordered in a frequency distribution with respect to their percent identity. These results are presented in the following table:
|
It is obvious from this table that a large number of proteins exists with a very high identity reflecting an excellent conservation during species differentiation. The other tail of the distribution is represented by a number of proteins with an identity which is obviously very low to match the mean of 82.5%. Therefore, both sides of the distribution were investigated for some typical representatives.
Proteins with a high identity are homeobox proteins (hox) which are transcription factors responsible for the developmental regulatory system. Similarly, heat shock proteins have a high identity and also transcription factors which bind to promoters of acute phase genes. Additionally, there are proteins with a high identity which might be not so obvious: ATP-synthase alpha chain, cAMP-dependent protein kinase beta chain or bFGF together with its receptor. All these proteins are extremely well conserved during species differentiation. These results might also suggest some new functions of some proteins: the bFGF system was introduced generally as a growth factor system. However, its major function could be considered from a developmental point of view as a trophic system.
Proteins with a very weak identity between humans and mice are proteins from the prostaglandin system, ribosomal proteins and proteins of the T-cell receptor system. In contrast to the bFGF system, proteins of the transforming growth factor system (TGF) have a weak identity, which is poorly understood.
Another way of looking for relationships between species on the basis of structural homology of proteins is the construction of phylogenetic trees. This was performed by the following procedure: the amino acid sequence of two proteins was selected for the investigation, one with a well-conserved sequence [cytochrome oxidase c (cox1), Fig. 1] and the other with a lower overall homology [cellular tumour antigen protein 53 (p53), Fig. 2]. For these two sequences, the respective forms from different species were selected. The sequences were aligned with the PILEUP program of the GCG package. These aligned sequences served as input for the phylogenetic program PAUP which constructed the respective evolutionary trees. These trees demonstrate convincingly the lower homology of the p53 family with no relationship between the different species besides human and cerae (green monkey). In contrast, the phylogenetic program only separates rat and mouse from the other species with regard to sequence differences in cytochrome oxidase c.
|
|
These findings again show that practically all species with the exception of primates are equally distant from man and no argument can be phrased in favour of one or the other. This must be provisionally accepted in favour of the mouse model.
In summary then, we must conclude that at the present level of knowledge, genomic analysis (structural genomics) is not very helpful in selecting the ideal species for cardiovascular experiments. The possibility of manipulating the genome of the mouse appears therefore as the best warranty for progress, in spite of the technical difficulties and in spite of the difficulties regarding size.
3.2 Are ‘functional genomics’ giving us new clues?
In a few years, all interesting genes will be sequenced and the chapter of structural genomics will be closed.
Single gene expression studies have already now lost their luster and are relegated to the descriptive level. Gene control and the interaction between genes and the interaction between the environment at large and the genome will be the interesting problems of the future. The buzz word of this year will be ‘functional genomics’. A recent review in Science [7]tried to infuse meaning into this new phrase. The authors state that "functional genomics represents a new phase of genome analysis and will require creative thinking in developing innovative technologies that make use of the vast resources of structural genomics information. Functional genomics refers to the application of global experimental approaches to assess gene function by making use of the information and reagents provided by structural genomics. It is characterized by high throughput experimental methodologies combined with statistical and computational analysis of the results. Functional genomics promises to rapidly narrow the gap between sequence and function and to yield new insights into the behaviour of biological systems". Although this sounds still rather vague it highlights the enormous gap that exists between the far advanced structural side of genome research and its biological expression. Daniel Tosteson, Dean of the Harvard Medical School, and a physiologist by training, described the new situation as follows. "In the past we have had functions in search of sequences. In the future, pathology and physiology will become ‘functionators' for the sequences".
The authors of the Science article cite also a few technical possibilities for closing that gap. It is a revival of the time-honored mRNA expression study but with many (hundreds of) transcripts at the same time as a microarray differential display and on chips to read with confocal microscopic techniques etc.
This may be of enormous relevance to cardiovascular research because only a few diseases are monocausal and a mass screening of many genes in a given sample of tissue originating from pluricausal disease like atherosclerosis or hypertension or tissue that originates from a distinct experimental situation may be extremely revealing. This exactly fits in with our own experimental approach to study as many genes as possible under standardized experimental protocols [8]but was hitherto time consuming because it had to be done on a one by one basis employing classical hybridization techniques.
The reference to statistical and novel computational analysis of mass screening as cited above is extremely apt because as experimental biologists, we know that if more than three factors influence the outcome of an experiment, the power of prediction regresses rapidly toward zero. Large numbers of experiments are then the only answer. That these large numbers can only be achieved with high throughput is obvious. It is also obvious that these new possibilities will change the way of conducting research. It will among many other factors become more expensive and will rely on extreme specialization, even more than we are used to today. This may mean also more concentration to canters of giant funding and this may open the gap between countries and between specialized research institutions and the universities.
The Science article makes clear though that "functional genomics will not replace the time-honored use of genetics, biochemistry, cell biology and structural studies in gaining a detailed understanding of biological mechanisms" but the tenor of the article makes clear that these time-honored disciplines may no longer be at the cutting edge.
Another interesting implication that the author mentions is that the "traditional format of scientific publication cannot reflect the scope and depth of data produced. A great legacy of the structural genomics era is the philosophy and practice of the public release of data that we hope will carry over to the functional genomics age".
|
4 Conclusions |
|---|
|
Top1 Introduction2 A look at...3 The importance of...4 ConclusionsReferences |
|---|
Cardiovascular science is at a crossroad again. This is good and invigorating and a far cry from the doldrums of the mid-eighties where new ideas were scarce and where integrative cardiovascular science was steadily replaced (in many American universities even physically) by molecular sciences. Now a new relationship starts between molecular and integrative scientists because both realize that they cannot possible exist alone and isolated. The mouse (that roars) is the heraldic animal of that new union.
|
References |
|---|
|
Top1 Introduction2 A look at...3 The importance of...4 ConclusionsReferences |
|---|
- American Heart Association: Heart and Stroke Statistic. 1997.
- Schaper W, Schaper J, editors. Collateral circulation—heart, brain, kidney, limbs. Boston, Dordrecht, London: Kluwer Academic Publishers, 1993.
- Schaper W, editor. The collateral circulation of the heart. Amsterdam, London: Elsevier North Holland Publishing Company, 1971.
- Vogt A.M, Htun P, Arras M, Podzuweit T, Schaper W. Intramyocardial infusion of tool drugs for the study of molecular mechanisms in ischemic preconditioning. Basic Res Cardiol (1996) 91:389–400.[ISI][Medline]
- Federhoff N.E. Man and his dog. Science (1997) 278:205.
[Free Full Text] - (GCG) Genetics Computer Group. In: Wisconsin Package 9.0 ed. Madison Wisc.: 1997.
- Hieter P, Boguski M. Functional genomics: It's all how you read it. Science (1997) 278:601–602.
[Abstract/Free Full Text] - Knöll R, Zimmermann R, Arras M, Schaper W. Characterization of differentially expressed genes following brief cardiac ischemia. Biochem Biophys Res Commun (1996) 221:402–407.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||