制药与进化

[编者的话]

制药与进化能有什么关系呢？及其探索进化机制是发现致病机理、寻找药物靶点的重要手段。这方面的研究也是生物信息学家感兴趣的一个方面，您想了解更多吗？请看下面这篇文章。

Abstract

The therapeutic effects of a drug are the result of its action on a specific molecular target combined with the counterregulatory responses of the organism that aim to restore the affected parameters to previous values. The regulatory systems that control the steady state have evolved under natural selection, but because natural selection is a slow process and cannot foresee future developments, these systems are not always adjusted optimally to current conditions. Their activation can therefore reduce or prevent the desired effects of a drug and can even produce adverse effects. Recent examples from the treatment of obesity and arterial hypertension provide new insight to the contribution of such ancient counterregulatory responses to the therapeutic effects of modern pharmacotherapy. This widens the scope of the rapidly growing field of evolutionary medicine, because it demonstrates that evolutionary principles apply not only to the pathogenesis of human diseases, but also to their treatment.

The regulatory systems that control biological functions in all organisms have developed as a result of natural selection. Because natural selection is a slow process and cannot foresee future developments, the adaptation of an organism to a changing environment occurs with a certain delay [1]. The current regulatory systems in humans, particularly those related to body fluid volumes and energy stores, probably still reflect evolutionary adaptation to a pre-industrial environment, and are therefore not adjusted optimally to present circumstances. Moreover, if a control system evolved to protect a certain parameter against environmental challenges that were common in the past, it might be redundant today, because the nature and intensity of environmental challenges have changed substantially [2,3] . Natural selection also favours individual reproductive success rather than longevity. Because a main goal of modern pharmacotherapy (see Glossary) is prolonging life, it is unsurprising that ancient regulatory systems aiming to improve reproductive fitness counteract modern therapeutic interventions that aim to reduce mortality.

Integrative response of the organism to drugs

Pharmacological considerations

Most of the drugs currently available act on a known target and have a well-defined mechanism of action at both the molecular and cellular level [4]. However, their final therapeutic profile in a patient depends on many additional factors, such as genetically determined individual differences in responsiveness, which influence the pharmacodynamics of a drug, or in drug metabolism, which influence its pharmacokinetics [5,6] . An important, but often overlooked component that determines the long-term pharmacological profile of a drug is the reaction of the organism's regulatory systems, such as neural or hormonal control circuits [7], to the perturbation caused by the action of the drug. Such compensatory responses tend to restore the steady state that existed before the application of the drug, even if that state was pathological. Thus, the organism responds to a drug in the same way that it responds to any environmental stimulus that disturbs its steady state. These responses reduce substantially or abolish totally a possible therapeutic effect and can even induce adverse events.

Because the regulation of biological functions is the result of the evolutionary adaptation of the organism to its environment, a better understanding of the principles of evolution and its medical consequences is needed to analyse the overall reaction of an organism to a drug. This growing field, also known as 'evolutionary medicine' [8], has made remarkable progress in recent years and could have considerable influence on the clinical use of drugs.

Evolutionary implications for pharmacotherapy

Drugs affect the function of an organism by acting on either effector systems (e.g. cellular structures that mediate a biological function, such as channels, transporters or enzymes) or regulatory systems (e.g. components of a humoral or a neural network, such as a hormone or a transmitter). If a drug acts on an effector system, the counterregulatory responses are usually stronger than if it acts on a regulatory system. The therapeutic effect of a drug is therefore an integral response, which combines the molecular action of the drug and the modifying influences of physiological regulation. The set point and range of these regulatory systems has developed during evolution under the influence of natural selection. Because natural selection cannot foresee future developments, these systems are not always adjusted optimally to the present situation.

The rational use of existing drugs and the development of new ones should therefore be guided by the following considerations: (1) drugs should be more efficient and better tolerated if they provoke fewer counterregulatory responses; (2) drugs inhibiting regulatory systems should be advantageous, because they prevent counterregulation; and (3) the suitability of regulatory systems as drug targets depends on their evolutionary origin and their functional importance in the present environment.

The notion that evolution has an influence on the efficacy and tolerability of clinically used drugs can best be illustrated by taking diseases of civilization, such as obesity and arterial hypertension, as examples. These polygenic diseases are common in industrialized countries and are characterized by the maladaptation and dysfunction of control systems that were essential for survival in the pre-industrialized era but that are less important under present conditions.

Dysregulation of energy homeostasis: obesity

Evolution and energy homeostasis

The development of body fat stores depends on the energy balance, which results from the difference between energy intake and expenditure. When energy intake exceeds expenditure over a prolonged period, even a slight daily energy gain results in obesity [9,10] . Evolution has favored energy reserves, because they are a prerequisite for survival and individual reproductive success. Because energy deposits should not impair physical fitness, triglycerides evolved as a very condensed energy store [11]. For hunters–gatherers, this meant that energy reserves were not costly in terms of their weight (in the form of glycogen, the weight load would be more than four times higher). For the modern obese patient who wants to lose weight, it means that numerous calories have to be avoided by dieting or spent by physical exercise to achieve a modest weight reduction.

Efficient ways have also evolved to protect energy stores. The thrifty gene hypothesis [12,13] postulates that specific sets of genes, which optimize energy utilization and storage, prepared our ancestors for 'feast and famine' by efficiently protecting energy reserves when food supply was low and by replenishing them rapidly, when food supply was high [14,15] . These systems respond whenever caloric intake is reduced, be that during starvation or deliberate dieting. For the hunters–gatherers, this meant that energy reserves could be restored rapidly after a period of famine. For the modern obese patient, this means an increased risk of relapse after a period of therapeutic dieting.

These considerations suggest that natural selection resulted in an increased susceptibility for the accumulation of body fat under conditions of unlimited food supply. Indeed, obesity is very common in industrialized countries and its prevalence in the USA amounts to ~25% of the adult population; the continuous trend for increasing overweight and obesity in children and adolescents is particularly alarming. Obesity is associated frequently with other diseases, such as arterial hypertension and non-insulin-dependent diabetes mellitus, which makes it a major health issue [16]. There is thus a great medical need for effective and well-tolerated anti-obesity drugs [17].

Important role of leptin in energy balance

Energy balance is regulated through complex interactions of factors in the body and the brain among which the hormone leptin plays an essential role [18,19] ( Fig. 1). Leptin is produced by white adipocytes and is secreted into the blood in proportion to the size of all fat deposits in the body. Stimulation of its receptors in the hypothalamus, the centre in the brain for energy homeostasis regulation, initiates a signaling cascade that ultimately affects various functions, including feeding behaviour, thermogenesis and fertility. This became evident from mutant obese rodent strains in which genetic leptin deficiency or leptin receptor defects caused an increased food intake, reduced energy expenditure, and impaired sexual maturation and reproductive function. Chronic administration of exogenous leptin to mice with leptin deficiency led to normalization of these parameters [18,19] .

Fig. 1.
Leptin and energy balance. Leptin is encoded by the Ob gene, which is expressed in white adipose tissue. Leptin is similar in structure to cytokines, which are a heterogeneous group of endogenous, bioactive peptides released mainly from inflammatory tissue and cells of the immune system. It is secreted into the blood and crosses the blood–brain barrier probably via a carrier system, which has a limited maximum transport rate. In the hypothalamus, leptin acts on specific receptors of the class I cytokine receptor subtype. Through stimulatory or inhibitory effects on downstream effector systems (e.g. -melanocyte-stimulating hormone, neuropeptide Y, etc.) leptin reduces appetite, but also influences various endocrine systems, such as adrenocortical and gonadal hormones [9,46] . In the long term, decreased food intake leads to a reduction in adipose tissue and, consequently, a diminished production of leptin, which then increases food intake.

Limited therapeutic efficacy of leptin

Leptin has been proposed to prevent obesity because it acts in a feedback loop between peripheral fat deposits and the brain: increased body fat leads to elevated leptin levels that, in turn, reduce feeding and, at least in rodents, increase energy expenditure. However, that circulating leptin levels are elevated in obese human patients [20] is difficult to reconcile with the notion that leptin limits weight gain. One explanation is that a persistent increase in leptin levels leads to leptin resistance, which limits the efficacy of this feedback loop [19].

Soon after the first demonstration of leptin deficiency and leptin receptor dysfunction in mice, very rare comparable mutations were observed in humans and were found to be associated with severe obesity [21]. Chronic daily administration of recombinant human (rh) leptin to a patient with congenital leptin deficiency restored leptin plasma concentrations to normal values, leading to a progressive decrease in body weight and the correction of delayed sexual maturation [22].

By contrast, the results of the first clinical trial with rh leptin in obese patients without leptin deficiency were much less promising [23,24] . To achieve a significant weight reduction, high doses, which produced plasma levels ~20 times above the baseline (placebo treatment), had to be administered. Even then, the average weight loss was <10% of initial weight and the individual responses were highly variable (from +10% to -20%).

Leptin and evolution

The concept of leptin as a starvation rather than an adiposity signal, as first suggested by Ahima et al. [25], could help to explain these observations. These authors claim that a fall in leptin levels as a result of starvation causes an integrated response, for example stimulation of appetite, decrease in energy expenditure (at least in rodents), and suppression of the reproductive and thyroid axes. Reduced plasma leptin levels thus tell the brain to increase food intake and to save energy by a temporary reduction of nonvital functions, such as reproduction. If leptin exerted its physiological role as a starvation signal at low plasma concentrations, it is understandable that its main actions occur below the normal plasma levels of ~10–20 ng ml-1 ( Fig. 2). Thus, leptin substitution treatment in which leptin plasma concentrations are restored to normal [22] appears to have a better efficacy than does pharmacotherapy with leptin, in which plasma leptin concentrations are further increased from an elevated baseline in obese patients [23].

Fig. 2.
Role of leptin in evolution and civilization. There is a negative correlation between hunger and plasma leptin (solid line) and a positive correlation between body fat mass and plasma leptin (dashed line). The dotted lines indicate the set point under normal conditions. The leptin–food intake relationship can be viewed as a concentration-response curve. The left part represents the treatment of a leptin-deficient patient with exogenous recombinant human (rh) leptin, in whom plasma leptin was raised from undetectable values to levels in the normal range (~10–20 ng ml-1), resulting in a substantial reduction of pre-treatment food intake [22]. The right part of the curve represents the results from the first clinical trial with rh leptin in lean and obese humans with normal or high plasma leptin. Only doses of leptin that raised plasma leptin to values 20 times higher than baseline showed a relatively modest and highly variable response [23]. The difference between leptin substitution of patients with a lack of endogenous leptin and leptin pharmacotherapy in patients with increased endogenous leptin corresponds to the difference in the role of leptin during the pre- and post-industrial period. In the pre-industrial (paleolithic) period, leptin worked as a starvation signal at maximum gain, whereas in the post-industrial period it works as an adiposity signal at minimum gain.

This interpretation is consistent with the general hypothesis that, during evolution, humans developed stronger mechanisms to defend themselves against weight loss rather than gain [15]. Leptin is an example of a regulatory system that potentially served an important function during evolution when food supply was limited, but one that does not have much therapeutic value under the present conditions, when food supply is plentiful. Conversely, it is highly likely that a fall in leptin levels after successful weight loss could raise appetite, thus making it difficult for the patient to maintain the reduced weight.

Dysregulation of blood pressure: arterial hypertension

Evolution and arterial blood pressure

The transition from a marine to a terrestrial environment required the maintenance of an iso-osmotic internal environment. The amount of sodium (Na+), the main extracellular cation, determines the size of the extracellular space and, as part of it, the intravascular volume. To keep extracellular volume constant in the presence of a limited Na+ supply, strong regulatory systems evolved to control renal Na+ reabsorption [26,27] . Intravascular volume or plasma volume is an important component in the regulation of blood pressure, because it influences the pumping function of the heart (i.e. cardiac output) via cardiac filling pressures. Together with total peripheral vascular resistance, which is mainly a function of arteriolar tone, cardiac output determines blood pressure (cardiac output × total peripheral resistance = blood pressure) ( Box 1).

Box 1

View Box

The precise regulation of arterial blood pressure is a prerequisite for the maintenance of the functions of peripheral organs [28]. The basic components of the cardiovascular system, vascular tone and blood volume, are controlled by various hormonal and neural mechanisms, the most important being the renin–angiotensin–aldosterone system (RAAS) and the sympathetic nervous system (SNS) [7] ( Box 1). After hemorrhage, for instance, volume retention is too slow and a rapid vasoconstrictor response is needed. A temporary reduction in blood flow to the periphery (e.g. skin and muscle) maintains blood supply to vital organs (i.e. heart and brain) [7] ( Box 2). However, counterregulatory mechanisms that mediate life-saving responses in the short term can turn into risk factors if they are activated persistently as a result of dysregulation of the cardiovascular control mechanisms. A persistent increase in vasoconstrictor tone or a chronic expansion of extracellular fluid volume can lead to arterial hypertension.

Box 2

View Box

Approximately 15% of the adult population in the USA have elevated blood pressure. In ~95% of all hypertensive patients, the causal mechanisms are unknown, and the disease is called essential or primary hypertension. Either increased body fluid volumes resulting from defects in renal salt excretion or an increased vasoconstrictor tone are the main causes. If hypertension is untreated, it leads to cardiovascular complications, such as stroke, myocardial infarction and heart failure. Because essential hypertension is incurable, the patients usually depend on life-long pharmacotherapy [29,30] .

Pharmacotherapy of hypertension

The development of antihypertensive agents over the past decades is one of the great success stories in drug discovery. In addition to providing effective and well-tolerated treatment for patients, the progress in the pharmacotherapy of hypertension has increased enormously our understanding of cardiovascular regulation [31]. Here, we compare one of the first available classes of antihypertensive drugs, diuretics [32], with one of the latest additions, angiotensin II (Ang II) receptor blockers (ARBs) [33]. Their diverse profiles can be ascribed to the fact that diuretics act on an effector mechanism and provoke counterregulatory responses, whereas ARBs block a regulatory system that was important in the past but is no longer vital under the present circumstances. Diuretics and blockers of the Ang receptor subtype 1 (AT1) are used widely to treat hypertension and heart failure [33,34] .

Diuretics

Diuretics inhibit the reabsorption of Na+ in the renal tubules by blocking specific steps in transcellular Na+ transport. This results in an increased renal Na+ excretion, which eventually leads to Na+ depletion and a reduction in body fluid volume, including plasma volume. This in turn leads to a decrease in cardiac output and a fall in blood pressure [32] ( Box 1). These changes are sensed by pressure and volume-sensitive structures (baroreceptors), which activate control systems such as the RAAS and SNS to restore pressure and volume [7]. Whereas the SNS acts only transiently and adapts within hours or days to a new basal level (set point), the RAAS remains activated as long as a volume deficit persists, even if that takes weeks or months [35]. Thus, the net result of long-term diuretic treatment is determined by at least two factors: on the one hand by the drug-induced volume loss and on the other hand by the vasoconstriction and volume retention mediated by Ang II and aldosterone. The balance between these two factors can differ from patient to patient. This is of clinical relevance in those patients who respond with a particularly strong increase in aldosterone secretion that is high enough to compensate fully for the diuretic-induced volume loss. Their 'overresponse' to this volume loss makes such patients 'nonresponders' to the antihypertensive effect of diuretics [36] ( Box 2). The exact percentage of these patients in the general population and the genetic basis for their overresponse to diuretics has yet to be determined.

Inhibitors of the RAAS

Inhibitors of angiotensin-converting enzyme (ACE) were the first drugs used clinically to block the RAAS [34]. Their excellent efficacy and good tolerability supported the notion that the RAAS was a suitable target for antihypertensive agents. However, because ACE has additional functions, such as the breakdown of other peptides (e.g. kinins) to inactive fragments, inhibitors of this enzyme influence systems other than the RAAS. By contrast, ARBs act only on the RAAS. They have an antihypertensive efficacy equal to that of ACE inhibitors and, probably as a consequence of their higher selectivity, fewer side effects. In fact, the frequency of adverse events during treatment with ARBs is similar to that seen after placebo treatment [33].

Antihypertensive therapy and evolution

The explanation for the better antihypertensive profile of ARBs as compared with diuretics appears to be straightforward. Instead of acting directly on an effector mechanism, as diuretics do, and thereby provoking counterregulation, ARBs block an important regulatory system, thereby inducing a pharmacological response whilst simultaneously preventing a compensatory response. However, such an explanation raises the question of how a regulatory system can be simultaneously important and dispensable. Evolution could offer a possible answer.

If the environment in the pre-industrialized world is compared with that of today, it is evident that the conservation of Na+ is no longer a vital defence mechanism [37] and so it can be asked whether the RAAS is still necessary under the present conditions of salt supply [38]. It is now much more important to excrete excess Na+, which comes from processed food or is added during meals, by a suppression of salt-conserving systems, such as the RAAS. There might be a subgroup of the population that is unable to suppress this system sufficiently in the presence of a Na+ load. Such patients should benefit particularly from treatment with ARBs and the pharmacological blockade of the RAAS. Conversely, when the RAAS is stimulated after diuretic treatment, the addition of an ARB will eliminate the counterregulatory response and restore or augment the antihypertensive efficacy of the diuretic. In fact, combinations of diuretics with blockers of the RAAS are among the most commonly prescribed antihypertensive drugs.

Evolutionary influences on drug actions: a general concept?

Other examples of evolutionary pharmacology

The examples from the pharmacotherapy of diseases of civilization show the hidden role of evolution in the integrated therapeutic response to drugs. Are these examples just exceptions or do they suggest a general concept? There are many more cases of the interplay between evolution and pharmacotherapy.

Efficient cholesterol carrier molecules, such as lipoprotein (a), were probably beneficial for our ancestors who had limited access to dietary cholesterol. Today, high plasma cholesterol levels are a major therapeutic problem and lipoprotein (a) is currently considered to be a risk factor for atherosclerotic vascular disease and thus a possible target for pharmacotherapy [39]. Immunological and inflammatory processes devoted to the neutralization of harmful agents in childhood and adolescence could be detrimental later in life by accelerating the progress of aging [40]. Stimulating host-defence mechanisms and prolonging life expectancy could therefore be conflicting objectives of pharmacotherapy. It is likely that more examples can be found by analysing the complex feedback loops of endocrine systems and metabolic pathways, particularly in the context of reproductive function [41,42] . Such an analysis not only helps to explain therapeutic failures but can also lead to successful new pharmacological approaches.

Compensatory mechanisms as drug targets: problems turned into opportunities

A recent example from cardiovascular pharmacology demonstrates that compensatory mechanisms are not only to be considered as limiting factors for drug response but, on the contrary, can also be appropriate targets for pharmacotherapy. In congestive heart failure, when, as a consequence of ischemic or inflammatory myocardial disease, cardiac output is no longer adequate to maintain organ perfusion, the organism reacts in the same way as after blood loss. The SNS and the RAAS are activated and the resulting cardiac stimulation, vasoconstriction and Na+ retention raise blood pressure at the expense of an additional workload for the already failing heart [43].

In this situation, inhibition of the SNS by -adrenergic blockers was considered an absolute contraindication, because it was expected to lower the contractile force of the heart [44]. However, in several well-controlled clinical studies, low doses of -adrenergic blockers had a beneficial effect, most probably by breaking the vicious circle of compensatory stimulation of the SNS and overloading of the heart. That a traditional contraindication can turn into a new treatment option might seem surprising and has consequently been used as an example of 'paradoxical pharmacology' [45]. However, it shows that a better understanding of regulatory principles and compensatory phenomena offers better treatment opportunities.

Implications for pharmacotherapy

Such theoretical considerations should have practical consequences for experimental and clinical pharmacology. In modern antihypertensive therapy, many of these postulates are already part of clinical practice. Numerous drugs with different mechanisms of action are available and suppression of counterregulatory responses can be achieved by their rational use as single agents or in combination. Pharmacotherapy of obesity, which only recently received appropriate scientific and medical attention, is at an early stage and could therefore be an area in which evolutionary considerations could be applied to the discovery, development and clinical use of new drugs.

Conclusions

What can be done to integrate evolutionary thinking into the use of currently available drugs and into the development of new drugs?

First, it should be mandatory that in vivo pharmacology obtains a more prominent place in the preclinical validation of therapeutic concepts, because experimental studies in whole animals are prerequisite for the evaluation of compensatory responses of the organism to a pharmacological intervention.

Second, computer modeling of complex regulatory networks could make it possible to perform a quantitative analysis of feedback loops and long-term effects in silico. Even though such systems are still far from enabling us to predict pharmacological effects, they can be used to integrate experimental results from different sources into a unifying concept.

Third, an improved understanding of the genetic predisposition to diseases of civilization that are triggered or aggravated by environmental factors will help to provide individual treatments. However, because diseases, such as arterial hypertension and obesity, have a polygenic background, such an approach will take considerable effort.

Finally, evolutionary medicine should not only focus on the pathophysiology of diseases but should also investigate the implications of evolutionary thinking for pharmacotherapy.