Validation of in-vitro data by in-vivo evidence: the example of angiotensin (1–7)

Similar to most areas of biomedical investigation, the field of cardiovascular and hypertension research has greatly evolved over the past few decades. Whereas previously, most, if not all, research activity concentrated on experimental work in intact animals and clinical investigation, the availability and increased use of new technology has brought enormous changes. Techniques such as perfusion of isolated organs, tissue culture, molecular biology, genetic engineering, genomics, and so on, have opened novel avenues of investigation and thereby provided opportunities for enhancing our knowledge to an extent that would have been unimaginable just a few years earlier. This development has generated extraordinary progress in the field. In parallel with this evolution, the profile of the investigators in the field has changed, with a greater influx of young doctorates who are more interested in data derived from the use of technological advances than in understanding the integrative physiology. Research funding has followed this trend to the point where clinical and/or whole animal investigation is regarded as outdated and thus less worthy of financial support. The implication is that whatever mechanism can be characterized at the genetic and molecular level reserves no more secrets to the enlightened mind. In-vitro research (at most combined with acute experiments in intact animals, if at all) has become the standard of all high-level research, whereas careful and long-term in-vivo investigation has been slowly disappearing, which has resulted in a gradual loss of experimental and clinical research skills. Although this reductionist approach has brought much insight and progress, it has become almost completely blind to aspects such as the interaction of multiple mechanisms, hierarchy and quantitative contributions of different mechanisms in an integrated system, as well as the ever present multiple environmental factors. As a result, we are continuously confronted with a huge quantity of new scientific data gained mostly from in-vitro experiments using the most modern methodology, often accompanied by sweeping extrapolations as to their key role in cardiovascular physiology and/or the pathogenesis of hypertension, despite there being little knowledge about their actual contribution to sustained integrated cardiovascular regulation. Recent research involving the renin–angiotensin system illustrates many of these unresolved controversies. To name just a few: (i) renin of extra-renal origin and its contribution to left ventricular and vascular remodelling; (ii) the role of AT2-receptors; (iii) the role of the angiotensin-converting enzyme (ACE) type 2; (iv) the direct role of renin via renin-receptors; and, last but not least, (v) the role of angiotensin (1–7). This latter peptide, which appears to be the metabolic product of ACE2, has been the focus of much investigation, with interesting results having been obtained in recent years. However, as is so often the case nowadays, most of this reported evidence originates from in-vitro studies and/or acute in-vivo experiments using pharmacological doses. Even ignoring the complicated interactions that exist in an intact organism, the basic question of what concentration of an agonist can be expected in a given tissue under physiological conditions is almost impossible to answer. It is widely accepted that agonists and antagonists almost always have to be administered at what appear to be much higher doses in vitro than those used long-term in vivo to produce an effect. By itself, this does not question the value of in-vitro research, but it calls for validation of the results obtained in a long-term in-vivo experiment under more or less physiological conditions. The transition from in-vitro to long-term in-vivo investigation often reserves some unexpected and not necessarily hoped for surprises. An interesting example of this is reported in this issue of the journal by van der Wouden et al. [1]. These authors set out to study the role of angiotensin (1–7) in the renal vasculature of the rat. With this purpose: ‘isolated small renal arteries were studied in an arteriograph system by constructing concentration–response curves to angiotensin II (Ang II), without and with Ang(1–7). In isolated perfused kidneys (IPK), the response of Ang II on renal vascular resistance was measured without and with Ang(1–7)’. Under both experimental in-vitro conditions angiotensin (1–7) clearly attenuated the angiotensin II induced vasoconstriction, thus shifting the angiotensin II dose–response curve to the right. However, subsequently ‘The influence of Ang(1–7) on Ang II-induced glomerular afferent and efferent constriction was assessed with intra-vital microscopy in vivo under anaesthesia. In freely moving rats [the authors then] studied the effect of Ang(1–7) on Ang II-induced reduction of renal blood flow (RBF) with an electromagnetic flow-probe’. By contrast to the findings of the initial experiments performed in vitro, the authors could not demonstrate any significant effect of angiotensin(1–7) on the renal vasculature in vivo. In other words, for whatever reason, the results of the well-conducted in-vitro experiments could not be confirmed in the intact animal. The significance of the study by van der Wouden et al. [1] stems from the authors first reproducing in-vitro effects to confirm the previously reported actions of angiotensin (1–7) on the renal vasculature [2,3]. Thus, what they observed in vitro fits very well within the accepted scheme, and they cannot be criticized for reporting totally divergent results. However, the verification in vivo did not support the results obtained in vitro. What are the reasons for this discrepancy? There are many possible explanations, ranging from unrealistically high doses used in vitro producing pharmacological rather than physiological effects, all the way to counter-regulatory mechanisms being activated by angiotensin (1–7) in vivo that cannot come into play in the in-vitro situation, and there are numerous other possible interpretations of the results. Whatever the final outcome, the authors have made the effort of first establishing in vitro what is to be expected on the basis of earlier reports, and then verifying in vivo whether the general concept persists. This might serve as a model of how modern research in the cardiovascular field should be conducted. Clearly, the present research cannot provide the final answer to the question of whether angiotensin (1–7) plays an important physiological role in cardiovascular homeostasis or in the pathogenesis of hypertension. Nevertheless, it confirms previous reports that could not reproduce a sustained effect on blood pressure [4], or renal haemodynamics [5], despite the use of pharmacological doses (400 ng/kg per min) of angiotensin (1–7) producing plasma levels far exceeding hitherto measured levels of the hormone [4,5]. Undoubtedly, other studies will produce new results contradicting the present findings, and it is only with time that a common thread of evidence will evolve. However, additional new in-vitro data supporting the role of angiotensin (1–7) should not be considered sufficient to contradict the current in-vivo results, unless some basic flaw in these studies has been overlooked. The challenge remains to present solid and convincing long-term data demonstrating a vasoactive effect of physiological concentrations of angiotensin (1–7) in vivo. To date, this evidence is not available. The study by van der Wouden et al. [1] illustrates two important points. First, any ‘negative’ results are relevant. The well-known and widely discussed bias against publication of ‘negative’ results is scientifically wrong. A well-balanced opinion on a scientific issue requires access to all relevant information, both in support and against the proposed thesis. Second, the extrapolation from basic research data to clinically significant evidence must overcome considerable hurdles and is not acceptable until confirmed in an integrated and sustained environment in vivo. Is angiotensin (1–7) a physiologically important peptide regulating the renal and/or the systemic circulation? We still do not know the answer.