Human evolutionary innovation for rapidly restoring glycogen, and link to cardiovascular disease?

Human evolutionary innovation for rapidly restoring glycogen, and link to cardiovascular disease?

We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

I'm a physicist, not a biologist, but I'm interested in human evolution and its link to the physiology of endurance sports. Circa August 2019, I read an article in the newspaper whose contents I'll summarize below from memory. I mentioned this to a colleague who teaches anatomy and physiology, and he said that it sounded not very credible. I told him I'd find the article and send him a link, but I wasn't able to find it again. I'm wondering if anyone can help me find the original scientific paper, or a popularization.

According to this account, humans have a mutation, not present in other primates, so that when we eat, our glycogen stores are replenished quite a bit faster -- maybe by a factor of 2 or something. Of course digestion is complicated, and the speed of digestion depends on the macronutrient and even on the subtype of that macronutrient (e.g., we probably process maltodextrin faster than whole-wheat bread), but for typical foods I think the idea was that a human would rebuild their liver and muscle glycogen reserves in a few hours, as oppose to twice that for other primates.

The article also said that this evolutionary innovation had a cost associated with it -- that it led to an increased susceptibility to certain types of cardiovascular disease, which our close relatives don't get as easily.


  • We (Homo genus) used a lot more glycogen as we sweated while we foraged.
  • Our efficient metabolism allowed us to replete our glycogen stores with seemingly little trade-off.
  • Those same metabolic genes are associated with obesity risk
  • Obesity is a risk factor for heart disease

Sweat uses glycogen

...and lots of sweating was beneficial for our species survival

I think what you are indirectly asking about, as unlikely as it seems, is sweat production.

It is known that humans have a large number of eccrine sweat glands compared to other animals, for which glycogen is the primary energy substrate. The ability to thermally vent via full-body eccrine sweat glands in this way, in conjunction with bipedalism, allowed foraging in the midday heat when predation is low. This is limited to the genus Homo (Leiberman, 2015), although it is not clear exactly which mutations are responsible as far as I can tell. Glycogen stores are more supported by their capillaries, and by extension have faster repletion, in primates in hotter drier climates and indeed in humans (Best & Kamilar, 2018). Now at this point, you are thinking "Aha! More capillaries mean more pressure on the cardiovascular system and more disease". I cannot find any evidence directly linking those two.

A note on glycogen production differences in our close relatives

The 1,4-alpha-glucan-branching enzyme indeed seems similar across close species. So either we have relatively larger livers to cope with this, or some complex genetic mechanism produces more of the protein. A quick BLAST alignment shows the identity of the 1,4-alpha-glucan-branching enzyme of chimpanzee species at >99% and E~0. The top results are shown below:

 Entry Organism Organism ID Info Status Q04446 Homo sapiens (Human) 9606 E-value: 0.0; Score: 3,814; Ident.: 100.0% reviewed A0A2R9CB94 Pan paniscus (Pygmy chimpanzee) (Bonobo) 9597 E-value: 0.0; Score: 3,803; Ident.: 99.6% unreviewed H2QMY2 Pan troglodytes (Chimpanzee) 9598 E-value: 0.0; Score: 3,803; Ident.: 99.6% unreviewed G3SDH8 Gorilla gorilla gorilla (Western lowland gorilla) 9595 E-value: 0.0; Score: 3,798; Ident.: 99.4% unreviewed A0A096NQ25 Papio anubis (Olive baboon) 9555 E-value: 0.0; Score: 3,729; Ident.: 97.6% unreviewed A0A2K6ANP2 Macaca nemestrina (Pig-tailed macaque) 9545 E-value: 0.0; Score: 3,723; Ident.: 97.3% unreviewed A0A0D9R0S5 Chlorocebus sabaeus (Green monkey) (Cercopithecus sabaeus) 60711 E-value: 0.0; Score: 3,718; Ident.: 97.4% unreviewed A0A2K5W1V0 Macaca fascicularis (Crab-eating macaque) (Cynomolgus monkey) 9541 E-value: 0.0; Score: 3,716; Ident.: 97.2% unreviewed A0A2I3GDY3 Nomascus leucogenys (Northern white-cheeked gibbon) (Hylobates leucogenys) 61853 E-value: 0.0; Score: 3,714; Ident.: 97.2% unreviewed A0A2K5I4Z7 Colobus angolensis palliatus (Peters' Angolan colobus) 336983 E-value: 0.0; Score: 3,713; Ident.: 96.7% unreviewed A0A1D5R8L3 Macaca mulatta (Rhesus macaque) 9544 E-value: 0.0; Score: 3,712; Ident.: 97.2% unreviewed A0A2K6KT87 Rhinopithecus bieti (Black snub-nosed monkey) (Pygathrix bieti) 61621 E-value: 0.0; Score: 3,705; Ident.: 96.9% unreviewed A0A2K6PB45 Rhinopithecus roxellana (Golden snub-nosed monkey) (Pygathrix roxellana) 61622 E-value: 0.0; Score: 3,699; Ident.: 96.7% unreviewed A0A2R9CB98 Pan paniscus (Pygmy chimpanzee) (Bonobo) 9597 E-value: 0.0; Score: 3,680; Ident.: 99.0% unreviewed A0A2I3T5K0 Pan troglodytes (Chimpanzee) 9598 E-value: 0.0; Score: 3,680; Ident.: 99.0% unreviewed A0A2I2ZFX5 Gorilla gorilla gorilla (Western lowland gorilla) 9595 E-value: 0.0; Score: 3,675; Ident.: 98.8% unreviewed A0A2K5QWD9 Cebus capucinus imitator 1737458 E-value: 0.0; Score: 3,665; Ident.: 96.0% unreviewed A0A2K6SM68 Saimiri boliviensis boliviensis (Bolivian squirrel monkey) 39432 E-value: 0.0; Score: 3,661; Ident.: 95.9% unreviewed F7FDF1 Callithrix jacchus (White-tufted-ear marmoset) 9483 E-value: 0.0; Score: 3,627; Ident.: 96.0% unreviewed A0A2K6ANP4 Macaca nemestrina (Pig-tailed macaque) 9545 E-value: 0.0; Score: 3,610; Ident.: 95.3% unreviewed A0A2I3MXY4 Papio anubis (Olive baboon) 9555 E-value: 0.0; Score: 3,607; Ident.: 96.9% unreviewed 

Although this is only one protein of a multi-protein metabolic pathway, I think there must be something much more complex about how our glycogen is repleted to deal with the higher demand than simple differences in proteins.

Increase metabolism and disease

Humans have benefitted from a "thrifty genotype" when it comes to metabolism (Neel, 1999). This, and the well-cited 1962 paper it is based on linking our faster metabolism to diabetes (Neel, 1962), may indeed be the paper you are looking for, but in my personal opinion, it is not very quantitative by today's standards and comes from the days before we had access to genomic data.

Although glycogen is only part of the story, it is generally accepted in the community that humans have an optimised metabolism. In terms of disease related to our metabolic evolution, GWAS showed that the genes that allow our increased metabolic activity is linked with obesity risk factors (Castillo et al., 2017).

To link that to your original question, obesity is linked with heart disease according to, well, everyone that has ever looked at it. Here is an NHS link to show it is a principal part of modern healthcare.

It may not be the exact answer you were expecting, so it will be fascinating if other papers corroborate this more directly.

Big brains need a faster metabolism

The scipop article you may have read could be from Science Daily, however this does not mention glycogen. This covers the Pontzer et al., 2016 paper which shows that we have a faster metabolism and this is linked with our brain size compared to our close primate relatives (another angle to the Leiberman 2015 paper discussed below). The scipop article goes on to talk about research linking this to an increased risk of heart disease, not mentioned in the original Nature article.

Human evolutionary innovation for rapidly restoring glycogen, and link to cardiovascular disease? - Biology

All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited.

Feature Papers represent the most advanced research with significant potential for high impact in the field. Feature Papers are submitted upon individual invitation or recommendation by the scientific editors and undergo peer review prior to publication.

The Feature Paper can be either an original research article, a substantial novel research study that often involves several techniques or approaches, or a comprehensive review paper with concise and precise updates on the latest progress in the field that systematically reviews the most exciting advances in scientific literature. This type of paper provides an outlook on future directions of research or possible applications.

Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to authors, or important in this field. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.

2. Types of intervention

Interventions can be classified into two broad categories: (1) preventive interventions are those that prevent disease from occurring and thus reduce the incidence (new cases) of disease, and (2) therapeutic interventions are those that treat, mitigate, or postpone the effects of disease, once it is under way, and thus reduce the case fatality rate or reduce the disability or morbidity associated with a disease. Some interventions may have both effects.

2.1. Preventive interventions

2.1.1. Vaccines

Vaccines are administered to individuals, usually before they have encountered the infectious agent against which the vaccine is targeted, in order to protect them when they are naturally exposed to the agent. Many are among the most cost-effective interventions, because, after a single dose or a series of doses of the vaccine, an individual may acquire long-term protection against the agent. They work by inducing a variety of immune mechanisms, through the humoral and/or cellular immune systems. The immunological responses and associated immunological memory induced by vaccination confer protection from later infections, though a booster vaccination may be necessary if the interval between the original vaccination and exposure to the agent is long. Most vaccines have to be administered before the infectious agent is encountered naturally, and thus field trials of such vaccines will involve the enrolment of healthy individuals and often involve infants or very young children—though the vaccine may be given at a later age if the age of natural infection is at later ages, for example, for most sexually transmitted infections (STIs), or if a new infectious agent, to which no one has been previously exposed, enters a community such as a new strain of influenza.

Not all vaccines are targeted at persons without previous exposure to the infectious agent. For example, there is substantial research to develop vaccines against parasitic diseases. The mode of action of some of these vaccines is to prevent parasitic proliferation within the host after invasion (and hence curtailment of disease), and some vaccines against vector-borne diseases are even targeted to prevent replication of the forms of the infection in the vector, so that onward transmission to humans is prevented.

For infectious diseases that affect both high-income countries (HICs) and LMICs, the first trials of new vaccines are usually conducted in HICs. This is because currently most new vaccines are developed and produced in HICs (though this situation is changing), and it is generally accepted that at least early clinical studies should be conducted in the country of vaccine manufacture. However, the results of trials in HICs may not be directly applicable to LMICs for a variety of reasons such as differing prevalences of other infections or of nutritional deficiencies, which might interfere with the mode of action of the vaccine. Thus, there will often be a need for further trials of the vaccine in LMICs, even if efficacy has been established in HICs. In addition, there has been increased focus in recent years on the development of vaccines against infectious agents that only, or almost only, occur in LMICs, such as malaria or visceral leishmaniasis, or where the overwhelming disease burden is in such countries, such as tuberculosis (TB) or HIV infection. For vaccines against these agents, the first major field trials to assess efficacy are likely to be conducted in LMICs.

2.1.2. Nutritional interventions

Food and nutrition are major determinants of human health and disease. Particularly in low-income countries and deprived populations in middle-income countries, under-nutrition remains a major cause of disease. Severe malnutrition, such as kwashiorkor or marasmus, is life-threatening, but milder forms of malnutrition are major risk factors that adversely influence the susceptibility to, and the outcome of, many infectious and other diseases, as well as cognitive development. In addition to calorie and protein deficiencies, specific deficiencies in micronutrients, such as iron, folate, zinc, iodine, and vitamin A, may be important determinants of severe diseases. Trials to address these problems may involve the regular provision of high-protein/calorie diets or supplementation to individuals with specific micronutrients, involving repeated visits to the same persons over several years, the frequency of administration depending on the nature of the supplement(s). Other trials, often with the intervention being applied at a community level, may involve food fortification (for example, iron, iodine, vitamin D) and experiments to change agricultural practices or eating or food preparation habits to increase the intake of particular micronutrients.

2.1.3. Maternal and neonatal interventions

A mother’s health and well-being during pregnancy and around the time of delivery, including access to appropriate care, are critical determinants of maternal mortality and neonatal and child health in the early years of life, and possibly for much longer. Preventive interventions before or during pregnancy include family planning, treatment of infections, such as syphilis and malaria, good nutrition, including micronutrients, good antenatal monitoring and care, and access to skilled care at the time of delivery and post-partum. Trials of maternal interventions may involve both community-based studies, with the early identification of pregnancies and the instigation of preventive interventions to avoid pregnancy complications, or may be hospital- or health centre-based, directed at improving the performance of the health system in caring for women during and after pregnancy and at the time of birth.

Interventions directed to the neonate are also important, such as exclusive breastfeeding and care practices, such as ‘kangaroo mother care’, a method of care of preterm infants, involving infants being carried, usually by the mother, with skin-to-skin contact.

2.1.4. Education and behaviour change

Some interventions directed at preventing disease are based solely upon changing human behaviour (for example, anti-smoking campaigns or campaigns to promote breastfeeding). Nearly all health interventions must have an associated educational component for their effective deployment, but the extent of educational effort required ranges from the provision of simple information (for example, when and where a clinic for immunization will be held) to efforts at increasing understanding (for example, of the importance of male circumcision for the prevention of HIV) and to attempts to change lifestyles (for example, diet or sexual habits). Education to increase knowledge and impart new skills may be necessary but is rarely sufficient to induce behaviour change. Individuals must also have the capacity, willingness, and motivation to act on the knowledge and to use the skills. The design and implementation of an educational intervention, and other 𠆌omplex’ interventions (Craig et al., 2008), will usually need to be researched through careful investigations in the community, using the kinds of methods discussed in Chapters 9 and 15.

Examples of educational components of disease control programmes include: ◆

educating children or mothers about the causes of the disease, such as diarrhoea, and how to prevent it

promoting adherence to long-term treatment such as for HIV infection or TB

developing effective participation in programmes that:

need broad coverage to maximize the effects of immunization or drug distribution

require people to recognize disease symptoms for early treatment

necessitate active co-operation in home improvements or insecticide programmes

involve direct action and responsibility in deploying vector, or intermediate host, traps

need community efforts for environmental improvements such as developing and maintaining improved water supplies or better disposal methods for faeces.

Organizing trials of behaviour change interventions are among the most challenging, and there are few examples illustrating the design of replicable interventions that achieve lasting behavioural change in the context of a trial. For example, changing tobacco smoking behaviour at a population level required decades of concerted, multifaceted campaigns. However, attempts to reduce diarrhoeal diseases and respiratory infections through the promotion of hand-washing with soap have produced encouraging results.

2.1.5. Environmental alterations

Alterations to the environment directed at reducing the transmission of infections are central to the control of many infectious diseases, particularly those that are transmitted through water, such as cholera, or through the faecal–oral route such as many gastrointestinal infections. Environmental interventions to reduce human faecal and urine contamination include latrine construction, provision of sewage systems, clean water supplies, and protected food storage. Other environmental interventions tackle indoor or outdoor air pollution or involve the disposal of contaminants such as pesticides or heavy metals. Many of these interventions require substantial educational efforts and lifestyle changes. They are also interventions that typically have to be applied to whole communities, rather than to individuals in a community, so that, in trials, the unit of randomization is the community or, in some instances, the household.

2.1.6. Vector and intermediate host control

Some major communicable diseases in developing countries depend on vector and intermediate hosts for their transmission. For different infections, the vectors include mosquitoes, tsetse flies, triatomine bugs, sandflies, ticks, and snails. There are a wide variety of control measures to reduce transmission of these infections through attacking the vectors or the reservoirs of infection. Most interventions require a good understanding of the vector or intermediate host, its life cycle, and the environmental conditions that it requires to propagate infections. Control measures may include the application of insecticides or larvicides, new or improved selective biological agents against disease vectors, engineering techniques for reducing vector habitats, community involvement in eliminating vector breeding sites and in deploying traps, housing and screening improvement for reducing human–vector contact, and strategies involving combinations of methods with, for example, the objective of reducing or delaying insecticide resistance. For many of these methods, intermediate process indicators, such as reduction in vector density, can be used for the assessment of impact, but it is often also necessary to determine the impact of the measures on the health status of the population. For example, for malaria, many different approaches to vector control have been used, based upon attacking the mosquito in various stages of its life cycle. These include control of breeding sites to reduce vector density by drainage and waterway engineering and application of specific larvicides and biological agents the use of mosquito netting, screens, and repellents for personal protection from bites aerosol distribution of insecticides to reduce adult mosquito densities and different approaches to killing adult mosquitoes, through either spraying residual insecticides, such as with dichlorodiphenyltrichloroethane (DDT), on the internal walls of houses where mosquitoes rest after a blood meal or through the use of insecticide-treated bed-nets (ITNs) that kill and/or repel mosquitoes seeking a blood meal. These different approaches require quite different study designs. Residual insecticide on the walls of houses offers relatively little direct protection to those in the treated household, as the mosquitoes take up the insecticide while resting after a blood meal. The protection is to those in other households whom these mosquitoes would have bitten for their next blood meal. To reduce transmission in high transmission areas, virtually all households in the neighbourhood must be sprayed. The higher the intensity of transmission, the more difficult it is to achieve sufficient coverage. The use of ITNs, developed as an intervention against malaria over the last two decades, leads to reductions in transmission, clinical disease, and overall childhood mortality. Trials of these kinds of intervention often involve communities, rather than individuals, as the unit of randomization. These trials are especially challenging to design, because some vectors, such as mosquitoes, may have a flight range that may lead to the 𠆌ontamination’ of intervention communities, with vectors coming in from outside of the community.

2.1.7. Drugs for the prevention of disease

Drugs or other interventions may be used for the prevention of infection (prophylaxis) or disease consequent on infection. An example of the former would be isoniazid prophylaxis to HIV-infected individuals to reduce their risk of TB, and of the latter, the treatment of HIV-infected individuals with antiretroviral drugs to slow the progression of their disease. Sometimes, the use of drugs for prophylaxis or to reduce disease progression does not involve individual diagnosis, but community or group diagnosis is needed to identify groups that should receive the treatment. For example, mass administration of anti-helminthic treatment to schoolchildren is sometimes administered in this way. Whether requiring specific diagnosis or not, therapeutic or preventive agents are usually taken on an individual basis, though sometimes agents can be distributed to everyone in a community through the water supply (for example, fluoride against dental caries) or in food (for example, historically, diethylcarbamazine for filariasis and chloroquine for malaria in medicated salt). Mass treatment of school-age children in areas highly endemic for the infection with an anti-schistosomal drug every year or two may be sufficient to virtually eliminate serious disease consequences of infection with Schistosoma mansoni.

Prophylaxis may be aimed at preventing or limiting infection, particularly in those at high risk for a limited period of time (for example, anti-malarials taken by those who are temporarily visiting malaria-endemic areas). The value of such an approach is limited by the duration of action of the agent (which determines the frequency with which it must be taken), by adverse reactions, and sometimes by the role of the intervention in stimulating the development of drug-resistant organisms. For some purposes, prophylaxis may be used by permanent residents of endemic areas (for example, anti-malarials in pregnancy).

Drugs also may be used prophylactically for treatment of preclinical infection (for example, during the incubation period before the onset of symptoms, as for the gambiense type of trypanosomiasis) or for treatment of subclinical infection (for example, ivermectin against onchocerciasis, and praziquantel against schistosomiasis).

Strategies for the use of such interventions include the mass treatment of entire populations or the targeted treatment of identifiable subgroups (such as school-age children) in areas where the infection is highly prevalent. Generally, such treatment is applied for the benefit of the individuals treated, but the objective may also be to reduce the transmission of the agent in the community more generally. When the prevalence is very high and the treatment is cheap, treating all those in a defined population may be more cost-effective than screening the whole population and then treating only those found infected.

2.1.8. Injury prevention

Injuries are major causes of death and disability, especially in LMICs. They disproportionately affect the young and have a large economic impact on society. For children and young people, road traffic accidents, drowning, fires, poisoning, interpersonal violence, and war are leading global causes of serious injuries, but often these are not considered ‘health problems’ and are not sufficiently integrated into public health thinking. Yet there are many potential interventions that might lead to reductions in deaths and disabilities from injuries, such as traffic calming or infrastructural changes to separate pedestrians from fast-moving vehicles to reduce motor vehicle injuries, and improving the security of water sources to reduce drowning accidents there is great need for more trials of interventions directed at reducing injuries.

2.2. Therapeutic interventions

2.2.1. Treatment of infectious diseases

The mechanism of action of a drug used for disease control will influence the design of field trials to evaluate its impact. Most drugs employed against infectious disease are used to kill or inhibit the replication or spread of the pathogen in the host. Strategies for disease control that use such agents may involve case detection (which requires an appropriate case definition and a diagnostic method), followed by treatment that is designed to reduce morbidity and mortality. Often, the public health success of this approach depends critically upon case finding, and, for diseases such as TB and leprosy, it depends also on case holding, i.e. being able to follow and treat each patient at regular intervals over sufficient time to eliminate the agent from the individual. Case finding and treatment may also reduce transmission of an agent if cases are the main reservoirs of infection, if case detection methods locate a high proportion of prevalent cases, and if the treatment is sufficiently effective.

2.2.2. Surgical and radiation treatment

RCTs of surgical and radiation treatments are usually done as clinical trials field trials of these interventions are relatively uncommon. However, procedures, such as cataract extraction or simple inguinal hernia repair, are examples of where field trials have been usefully undertaken. In general, the only distinctive feature that may set these apart, in terms of study design, from other field trials is the issue of 𠆋linding’ (see Chapter 11, Section 4). For some forms of surgery, ‘sham’ operations have been used in clinical studies and perhaps could be considered in field trials. In general, however, randomized trials of these procedures will have to be conducted without blinding.

2.2.3. Diagnostics to guide therapy

The efficient treatment of most diseases requires first that they be accurately diagnosed. Often the diagnosis is made on the basis of clinical symptoms and signs, but the imprecision of this method for many conditions is increasingly recognized. There is an urgent need for new, or improved, sensitive and specific diagnostic tests for many infectious and chronic diseases, that are both simple to use and cheap. For example, intervention strategies that depend upon case finding and treatment usually require suitable diagnostic tests. Specific studies may be necessary to measure the specificity, sensitivity, and predictive values of different diagnostic tests, as these properties will impact on the likely effectiveness of a case finding and treatment intervention. For example, the development and widespread introduction of rapid diagnostic tests for malaria, to replace microscopy or the presumptive treatment of fever, has been an important innovation in malaria control and has also focused attention on the need for improved diagnostic methods and appropriate treatment of non-malarial fevers.

Field trials to evaluate the performance characteristics of diagnostics are not discussed specifically in this book, other than in the context that they may be incorporated as part of an intervention strategy to improve the control of a specific disease. The design of studies to evaluate the properties of diagnostics has been discussed elsewhere (Peeling et al., 2010).

2.2.4. Control of chronic diseases

Chronic conditions may have an infectious aetiology (for example, HIV, TB) or may have environmental or other causes (for example, cardiovascular diseases and many cancers). Many chronic diseases, once diagnosed, may not be curable, but they can be controlled by a combination of education/behaviour change interventions, plus regular, often daily, use of pharmaceuticals. The nature of the clinical care required is often more complicated than required for acute conditions, such as diarrhoea and pneumonia, which, once diagnosed, usually require a single course of treatment. Interventions for chronic disease often must include screening of communities to identify cases assessment of each case for the stage of the disease and possible attendant complications that are likely to require a variety of laboratory tests and developing a long-term treatment and assessment plan. The treatment of such conditions often requires long-term monitoring, with a dependence on reliable laboratory results and a system to track the clinical and laboratory findings within a single individual over time. Trials of such interventions must often be conducted over several years, or even decades, to completely assess treatment efficacy.

2.3. Other forms of intervention

2.3.1. Legislation, legal action, taxation, and subsidies

Enforcement of anti-pollution laws, food labelling, and legal restrictions have an important role to play in public health. Behaviour may be strongly influenced by legal restrictions, and increasing prices through taxation have been shown to be effective in reducing tobacco and alcohol consumption, for example. However, it is difficult to design randomized trials of such interventions, because the interventions usually have to be implemented at the national level, making it very difficult to identify a suitable control group.

There has been increasing interest recently in providing various types of subsidies to individuals to change their health-related behaviour (often known as conditional cash transfers). Examples include incentives for children to remain in school, or to health care providers to provide services of at least a certain minimum quality (performance incentives). Some of these interventions have been evaluated through RCTs, and there is further scope for using such approaches.

2.3.2. Health systems interventions

Increasing recognition of the importance of interventions that operate at health systems level, such as policy implementation, financing, educational reform, and strengthening of leadership, management, and governance, has led to a variety of health sector training programmes, organization changes, decentralization and devolution, and various incentives and personnel policies. Most of these efforts have been introduced on a system-wide basis, with little thought about the value of rigorous assessment. But, with adequate planning, rigorous evaluation of these kinds of interventions should be possible through randomized trials, especially by making use of the ‘stepped wedge’ approach of a phased introduction of measures in different communities over a period of time (Brown and Lilford, 2006). Many health systems research studies may be considered as implementation research, and most could be considered as complex interventions, as discussed in Sections 2.3.3 and 2.3.4.

2.3.3. Implementation research

Within the context of field trials, implementation research does not aim to develop new interventions but focuses on optimizing the delivery of existing interventions that have previously been shown to be efficacious when implemented well. Implementation research explores the challenges of how best to implement research findings in the real world and how to contextualize interventions for specific settings. Hence, an example of an implementation research trial was one where a comparison was made of the costs and effectiveness of health workers delivering antiretroviral therapy to patients who attend a central clinic or hospital, compared with lay workers delivering the antiretrovirals to patients in their homes and only referring them to the clinic if they reported problems on a screening questionnaire (Jaffar et al., 2009).

A general reference on implementation research is Werner (2004).

2.3.4. Complex interventions

The design of a trial to evaluate the efficacy of a new vaccine or drug is relatively straightforward, in the sense that there are many past examples of such evaluations to draw upon when planning a new study. However, the evaluation of some interventions, such as the deployment of a new procedure in the health service or in public health practice, may involve consideration of several interacting components, including, for example, educational components and behavioural change. Such interventions pose special problems for evaluation, and these kinds of intervention have been called 𠆌omplex’. Many of the extra problems relate to the difficulty of standardizing the design and delivery of the interventions, their sensitivity to features of the local context, the organizational and logistical difficulty of applying experimental methods to service or policy change, and the length and complexity of the causal chains linking intervention with outcome.

In 2000, the UK Medical Research Council published a Framework for development and evaluation of RCTs for complex interventions to improve health to help researchers and research funders to recognize and adopt appropriate methods. These guidelines were updated and revised subsequently and can be downloaded from the Internet (<>).

Box 2.1 is reproduced from the guidelines and summarizes the steps in developing and evaluating trials involving complex interventions.

Box 2.1

The development𠄾valuation–implementation process.

During the last 450 million years, there have been at least five mass extinctions that have occurred due to a variety of causes, including changes in atmosphere gases, changing global temperatures, volcanic activity and an asteroid impact 1 . While often the focus is on those species that failed to survive, in many respects it is the survivors that deserve the most attention, for many of these animals have developed remarkable means of survival. Today, there are many examples of ‘extremophile’ species that can survive under remarkable situations, such as the Pompeii worm that can survive inferno (176°F) temperatures 2 , or the occellated icefish that lives in the Antarctic seas in the absence of red blood cells 3 , or the wood frog in northern Canada who freezes in winter, surviving because of the production of glycerol that acts as an antifreeze to allow slow circulation of blood in the freezing conditions 4 .

One of the most important means for survival is to have sufficient food and water, as well as the necessary minerals, electrolytes and nutrients to maintain muscle mass and body functions. It is also important to be able to adapt in conditions where oxygen levels may decrease. One means for doing this is to store caches of food in one’s den, but there is always the danger that the cache could be stolen, or that the den itself may become unsafe if discovered by predators. Thus, the ideal means for assuring survival is for the body itself to aid in the storage of food, water and other critical needs.

There appears to be a common mechanism by which many animals survive, and that it involves a unique metabolic pathway mediated by fructose, a simple sugar present in fruit 5 . Fructose is also produced in the body under conditions of stress. In turn, the metabolism of fructose uniquely activates processes that stimulate survival, and it works through specific hormones (such as vasopressin) as well as metabolic products (uric acid) to mediate its effects. Here, we provide a brief description of this central pathway that appears to have a key role in the evolution of species (Fig. 1).

Key facets of evolution

Evolution, by definition, is witnessed after the fact. Involving time implies that there is a huge difference when we anticipate short term and long term evolution of a biological entity. Furthermore, making predictions is hazardous, as evolution is myopic. It cannot have any grand design. However, in the long term, the very fact that an organism is still extant will highlight functions that allowed it to keep reproducing during that span of time. Things are different in the short term, with only a limited set of descent-related functions, possibly missing those that allow propagation in the distant future, for example. Moving to a new host from a host with which it has interacted for a long time, will suddenly expose the virus to an unfamiliar environment. Yet, it still follows the program of functions that allowed it to thrive in its usual host. In parallel, the new host is also the result of long-term evolution. While naïve for this specific invader, it has been shaped by natural selection that retained a variety of generic responses to react against that kind of invasion. In the case of viruses, natural innate immunity has been selected for functions that recognize the presence of viral features, as well as prevent, or at least control viral development (Nan et al., 2014 Chen et al., 2017 Hur, 2019 ). Many of these functions are shared by animal and even plant families. This is witnessed, for example, by the discovery – unexpected at the time – of the role of Toll receptors in Drosophila (Belvin and Anderson, 1996 ).

Evolution cannot decide beforehand whether a virus will be able to beget progeny in the long term In the short term, this implies that a successful virus will have maximized its descent without any direct feedback from the survival of its host, unless the host is killed before the virus had time to reproduce. Extreme virulence, with maximum killing efficiency is not sustainable in the long term, for want of hosts. This essential requirement directs us to look into functions expected to emerge as an epidemic unfolds. Unfortunately, understanding the various paths of evolution has led to a large number of simplifying hypotheses, shaped by anthropocentric views with an economic or moral flavour: the behaviour of a biological entity has been seen as « altruistic » or « selfish ». Mutations are perceived as « advantageous », « deleterious » or « neutral ». In the case of a viral pathogen, this is despite the fact that a great many processes leading to an active viral progeny cannot have pre-existing reasons to elude inconsistencies. For example, even in the absence of an identified functional consequence in a gene product, a nucleotide change in the genome sequence can affect metabolic organization of the host, fine tuning of the virus replicase, interference with innate immunity, temperature, modulation of functional tRNA availability, co-infection with another pathogen, and so forth. This has consequences for the multiplication of the virus, with constraints at all types of levels similar to those just listed. None of this is really « neutral ». Assuming neutrality or similar soft aspects is simply a means to describe processes we do not understand and to hide our ignorance. This is misleading if we hope to be able to anticipate at least some of the future of an epidemic, notwithstanding the inevitable role of ‘black swans’ in the way biological entities evolve (Taleb, 2008 ).

Whether a change in the virus genome has an effect will be seen in the course of time, with different possible outcomes depending whether the virus spread is surveyed in the short, medium or long term. What is important to us is not the model of evolution we would like to apply, but, rather, to make out the panel of features that may or will emerge as the pandemic expands, reaches a peak and calms down. In this respect, the present epidemic was highly predictable, and, in fact, predicted in many studies (Moya et al., 2004 Turinici and Danchin, 2007 Horby et al., 2013 ). Rather than use models to try and make predictions, we prefer here to try and identify the functions that enter into play during a viral infection, to see how they could help us anticipate at least some of the future of the epidemic.


New forms of human enhancement are increasingly coming to play due to technological development. If phenotypic and somatic interventions for human enhancement pose already significant ethical and societal challenges, germline heritable genetic intervention, require much broader and complex considerations at the level of the individual, society and human species as a whole. Germline interventions associated with modern technologies are capable of much more rapid, large-scale impacts and seem capable of radically altering the balance of humans with the environment. We know now that beside the role genes play on biological evolution and development, genetic interventions can induce multiple effects (pleiotropy) and complex epigenetics interactions among genotype, phenotype and ecology of a certain environment. As a result of the rapidity and scale with which such impact could be realized, it is essential for ethical and societal debates, as well as underlying scientific studies, to consider the unit of impact not only to the human body but also to human populations and their natural environment (systems biology). An important practicable distinction between ‘therapy’ and ‘enhancement’ may need to be drawn and effectively implemented in future regulations, although a distinct line between the two may be difficult to draw.

In the future if we do choose to genetically enhance human traits to levels unlikely to be achieved by human evolution, it would be crucial to consider if and how our understanding of humans and other organisms, including domesticated ones, enable us to better understand the implications of genetic interventions. In particular, effective regulation of genetic engineering may need to be based on a deep knowledge of the exact links between phenotype and genotype, as well the interaction of the human species with the environment and vice versa.

For a broader and consistent debate, it will be essential for technological, philosophical, ethical and policy discussions on human enhancement to consider the empirical evidence provided by evolutionary biology, developmental biology and other disciplines.


Mitochondria have emerged as a central factor in the pathogenesis and progression of heart failure, and other cardiovascular diseases, as well, but no therapies are available to treat mitochondrial dysfunction. The National Heart, Lung, and Blood Institute convened a group of leading experts in heart failure, cardiovascular diseases, and mitochondria research in August 2018. These experts reviewed the current state of science and identified key gaps and opportunities in basic, translational, and clinical research focusing on the potential of mitochondria-based therapeutic strategies in heart failure. The workshop provided short- and long-term recommendations for moving the field toward clinical strategies for the prevention and treatment of heart failure and cardiovascular diseases by using mitochondria-based approaches.

Cardiovascular disease (CVD) remains the number one killer in developed countries. Increases in life expectancy and improved treatment strategies for ischemic heart disease and myocardial infarction have led to a steady rise in heart failure (HF) prevalence. Despite the use of guideline-directed therapies, the morbidity and mortality of HF remain unacceptably high. There have been few new therapies for HF with reduced ejection fraction (HFrEF) in the past 20 years, and no compelling new therapies for HF with preserved ejection fraction (HFpEF). Novel approaches, orthogonal to traditional neurohormonal blockade, are thus greatly needed. Mitochondrial dysfunction and energy deficiency have been strongly implicated in the development of HF. 1–3 Targeting mitochondrial dysfunction in HF may provide novel approaches that are both hemodynamically favorable and complementary to current, somewhat limited approaches. To date, however, mitochondrial-targeted therapies have not succeeded in impacting this disease process. Phase III trials targeting the mitochondrial permeability transition pore with cyclosporine, for example, were disappointingly negative despite encouraging phase I/II results, as were large trials with antioxidants such as vitamin E. 4,5 A better understanding of mitochondria and their role in the pathobiology of HF, in conjunction with better tools for the delivery of mitochondria-targeted therapies and the monitoring of mitochondrial function in humans, are needed to translate this innovative treatment strategy.

Since the previous National Heart, Lung, and Blood Institute (NHLBI) mitochondria-focused workshop in 2007 titled, “Modeling Mitochondrial Dysfunction in Cardiovascular Disease,” 6 major advances have been made and substantial molecular information critical to our understanding has rapidly accumulated, bringing this classic discipline back to renewed attention. It is now recognized that mitochondria, traditionally viewed as the powerhouse of the cell, sense and respond to changes and stresses in the cellular environment and that they control critical cellular decision points. The recent workshop, “Unlocking the Secrets of Mitochondria: Path to a Cure in Heart Failure,” 7 held by the NHLBI on August 6 to 7, 2018, discussed major advances in mitochondrial science and identified key knowledge gaps in translating these advances to mitochondria-based therapies for HF. Here we report on the challenges and recommendations for 5 priority areas.

Multiplicity of Mitochondrial Functions in CVD and HF

The heart demands a substantial amount of energy relative to other organs. Mitochondria occupy approximately one-third of the volume of adult cardiomyocytes. 8 Oxidative metabolism in mitochondria provides the majority of energy consumed by the heart, and inability to generate and transfer energy has long been considered a key mechanism of contractile failure. 9,10 It is increasingly recognized, however, that mitochondrial function extends far beyond that of a power plant and includes important biological and regulatory roles such as redox balance, biosynthesis, reactive oxygen species (ROS) signaling, cell growth and death, ion homeostasis, protein quality control, and inflammation. 3,11–16 We are now beginning to appreciate that the pathogenic role of mitochondria in HF and CVD involves not only decreased ATP production but also a general maladaptation in the spectrum of its functions (Figure 1). These observations have enabled a large variety of new targets to be considered for mitochondria-based therapies. 17,18

Figure 1. Multiplicity of mitochondrial function. Mitochondria are known as the powerhouse of the cell. In addition to generating ATP, intermediate metabolism in the mitochondria produces metabolites for biosynthesis, protein modification, and signal transduction. Oxidative phosphorylation regulates the NAD(H) redox state and is coupled with the generation of reactive oxygen species (ROS) both can modulate and trigger posttranslational modifications. Mitochondrial metabolism is stimulated by Ca 2+ a lower Ca 2+ level impairs mitochondrial activity, whereas calcium overload can trigger the opening of the mitochondrial permeability transition pore (mPTP). The release of mitochondrial contents, such as cytochrome C, induces apoptosis, or the loss of membrane potential (a consequence of prolonged mPTP opening) causes ATP deprivation and necrosis. Leakage of damage-associated molecular patterns, such as mitochondrial DNA and peptides, or excessive ROS generation can cause inflammation that results in further tissue damage. Mitochondrial function is also regulated by biogenesis, fission, and fusion dynamics, and protein quality control via mitophagy. The transition of mitochondria from a powerhouse to a death engine involves a shift of the entire spectrum of functions. Modified with permission. 3

Research Gaps and Opportunities

Despite striking observations in preclinical studies, in large part involving bioengineered mouse models, the relative contribution of each unique biological function of mitochondria to the development of HF remains unclear. Moreover, little is known about the integration of mitochondrial bioenergetics with each role. Identification of novel therapeutic targets relies on further elucidation of mechanisms that link processes involved in oxidative metabolism (eg, fuel selection, energy production/transfer, and ROS generation/scavenging) with the numerous other functions of mitochondria. For example, decades of research have revealed that impaired myocardial energetics are accompanied by defects in substrate utilization, Krebs cycle flux, and oxidative phosphorylation. 1,19–24 Effective therapies for improving energy supply of the failing heart are lacking. 25–28 In addition, whether interventions targeting intermediary metabolism will be sufficient to overcome mitochondrial dysfunction and improve the outcome of HF is also an important unresolved question. Furthermore, little attention has been given to the nonenergetic roles of mitochondria in the failing heart, which likely play a critical role in pathological remodeling through proteomic and epigenomic modifications. 29,30

Preclinical studies demonstrate that mitochondrial Ca 2+ is a key regulator of energy metabolism and also a trigger of mitochondria-induced cell death via activating the mitochondrial permeability transition pore. 31 There is a paucity of information regarding the state of mitochondrial Ca 2+ dynamics in human HF, highlighting an important knowledge gap. It remains controversial whether mitochondria in the failing heart are Ca 2+ starved or overloaded. 14,32,33 From the therapeutic perspective, any extreme increase or decrease in mitochondrial Ca 2+ is likely to lead to negative outcomes, whereas more subtle interventions might be beneficial in the setting of HF or during ischemia-reperfusion injury. To address these gaps, we need to develop strategies that modulate mitochondrial Ca 2+ levels within the physiological range. Similarly, it is also unclear whether altered rates of mitophagy in HF are sufficient to maintain the balance between biogenesis and degradation and what form of mitophagy may be dysfunctional. Subtle interventions to alter these processes may be necessary to fine-tune mitochondrial quality to optimize function in HF. Our incomplete understanding of why the metabolic gene program is altered in HF and how to safely activate cardiac mitochondrial biogenesis to reverse defects in oxidative capacity is a significant barrier to therapeutic targeting of mitochondria at the present time.

Although mitochondrial dysfunction appears integral to HFrEF, the role of mitochondrial function in HFpEF is poorly defined. The lack of effectiveness of proven HFrEF treatments to improve outcomes in HFpEF might likely indicate that the mechanisms involved in these distinct forms of HF differ significantly. It is notable that the HFpEF population shares a number of characteristics with patient populations known to have impaired mitochondrial function (eg, older age and obesity). Thus, deciphering mitochondrial mechanisms in HFpEF is highly warranted.

The knowledge gaps identified at the workshop provide an excellent road map for future work, especially translational research on mitochondria-focused HF therapy. Research in the past decade has identified not only multiple pathological mechanisms but also a significant number of potential therapeutic targets. Moving these targets into therapy will require the collaborative efforts of biologists, engineers, and clinicians who (1) translate disease mechanisms to druggable targets, (2) devise effective strategies to engage the targets, and (3) develop monitoring systems to follow the biological outcome (see the “New Tools” and “Translation to Patients” sections). We expect that the effort will yield promising leads for clinical testing. Recognizing the multitude of mitochondrial mechanisms in HF should drive the focus of future investigations toward a balance of critical regulators of mitochondrial function, such as Ca 2+ , ROS, and redox state. We will continue to build on the concept that mitochondrial dysfunction in HF represents a spectrum shift rather than the loss of a single function (Figure 1). Future work will strive to restore homeostasis rather than manipulate individual functions. New discovery in mitochondrial biology lies in the integration of mitochondrial bioenergetics with its role in regulating cell fate.

Intra- and Intercellular Communication

It has become increasingly clear that mitochondria do not work in isolation. Communication within a mitochondrion, between mitochondria, and between mitochondria and other cellular organelles (eg, the nucleus and sarcoendoplasmic reticulum), and crosstalk of mitochondria across different cells or tissues, as well, are all increasingly recognized as important responses to environmental stressors. A number of studies have shown that mitochondrial metabolites, such as thioester (acyl-coenzyme A) and ROS, can directly modify the mitochondrial proteome, thus rapidly modulating mitochondrial activity in response to environmental changes. 34,35 Such regulatory circuits present potential opportunities for intervention because they couple mitochondrial metabolism with its sensitivity to stress. 11 Furthermore, communication between mitochondria and the nucleus through epigenetic modification is another powerful mode of determining cell fate. Defective oxidative phosphorylation significantly affects both acetylation and methylation processes by altering acetyl-coenzyme A metabolism and the methionine cycle, respectively, changes that can be reversed by restoration of the NADH redox state. 36 Moreover, α-ketoglutarate, a Krebs cycle intermediate, plays a specific role in cytosine demethylation reactions as a cofactor for the Tet family of dioxygenases in the nucleus. 37 In addition to the possibility that mitochondrial metabolism can alter the availability of epigenetic enzyme intermediates, emerging evidence indicates that mitochondrial genotype itself might influence nuclear DNA cytosine methylation patterns, as demonstrated in mitochondrial nuclear exchange experiments. 38 Mitochondrial nuclear exchange mice display differential DNA methylation patterns and gene expression between strains having identical nuclear DNA (nDNA) but different mitochondrial DNAs (mtDNAs). 39 It is interesting to note that this non-Mendelian control over gene expression was found to influence whole-body metabolism and susceptibility to adiposity upon high-fat feeding, 40 and the susceptibility to cardiac volume overload as well. 41 In humans, mtDNA mutations have been proposed to contribute to bioenergetic adaptation to dietary changes and are hypothesized to modulate disease susceptibility. 42–44 This concept suggests that the natural variation of mtDNA background, combined with the coevolution of its nucleus, significantly affects the cellular response to stimuli and contributes to individual variability in predisposition to cardiometabolic disease. Different mtDNA-nDNA combinations can affect the susceptibility to HF, 41,45 and it is possible that the underlying mitochondrial genomic response to meet environmental challenges may, in a retrograde fashion, regulate nuclear gene responses that further modulate disease susceptibility.

Research Gaps and Opportunities

The nature of mitochondria-originating signals predicts a network of changes that poses a challenge for the identification of specific mechanisms and targets. Although several emerging signaling mediators, such as mitochondria-derived metabolites, peptides, mtDNA, and ROS have been identified, little is known about their specific targets and functions in health and disease. There is no information on the mitochondrial metabolome in living cells. Direct evidence of how metabolites produced in mitochondria alter biochemical reactions in the nucleus remains elusive. In addition, we lack knowledge of site-specific quantitation of occupancy for each posttranslational modification (PTM) of mitochondrial or nuclear proteins. The field is still perplexed by the association between a robust biological effect and modifications of a relatively small fraction of proteins. Because a variety of molecules can modify the same amino acid residue, the additive and even synergistic (or antagonistic) effects among protein populations bearing different modifications on the same site should be considered in its totality. Likewise, the role of PTMs of mitochondrial proteins and their effects on protein-protein interactions is poorly understood.

To comprehend more fully how mtDNA influences the susceptibility of patients to CVD and HF, we need to gain a thorough understanding of mtDNA polymorphisms and mutations in the general population. Tissue differences in mitochondrial variants and their impact on the pathogenesis and progression of HF have not been addressed. Animal studies focusing on the causal relationship between mitochondria-nuclear genome interactions are currently lacking. Evidence derived from large sample sizes is needed to determine whether mitochondrial genotype can be used for stratification and treatment of patients who have HF.

The workshop recognized that further understanding of how mitochondria control nuclear functions through epigenetics, damage-sensing pathways, or simply by their maternal origin will substantially advance the field. The global impact of the vast number of changes that can occur through these mechanisms appears to be a problem to unravel by single-pathway analysis but can be tackled more readily through a systems approach. With emerging tools in hand or in development that integrate multiomic analysis with deep mitochondrial phenotyping, we have an enormous opportunity for discovery and innovation. Future studies will resolve the mechanism by which proteomes and metabolomes translate biological signals into functional outputs (Figure 2). For example, we will define how specific metabolites are altered by various stresses, which metabolites are secreted from the mitochondria and to the extracellular space, how the rate of secretion is modulated by mitochondrial activity, and how these metabolites signal other cells. We will determine the mechanisms by which PTMs alter mitochondrial phenotype throughnetworks and interactomes. Future therapeutics will be developed to target mechanisms for retrograde signaling to the nucleus or retrograde-anterograde crosstalk including novel pathways for modulating immunological responses, protein folding, endoplasmic reticular stress responses, and protein degradation processes.

Figure 2. Targeting mitochondrial functions in heart, lung, and blood health and disease. Mitochondria in multiple cell and organ types contribute to the pathogenesis of heart failure or its risk factors. A better understanding of basic mitochondrial biology including the multiplicity of functions played by mitochondria, their role in intra- and intercellular communication, and the relationship between mitochondrial genotype and phenotype is a high priority in developing mitochondria-based therapy. Other priority areas emphasized the need for improved tools and the development of new approaches for the study of mitochondrial function in humans and the development of strategies for the translation of basic observations to the clinic. Mt indicates mitochondrial PBMC, peripheral blood mononuclear cell and ROS, reactive oxygen species.

Phenotype and Genotype

The role of mitochondrial phenotype and genotype in determining the response to environmental stresses and the propensity for the development of HF is poorly understood. It is increasingly clear that mitochondrial function is heterogeneous across cell and tissue types, and from person to person. Mitochondrial genotype and phenotype are influenced by intrinsic factors such as genetics, race, age, and sex. Such heterogeneity may influence the response to environmental factors such as diet and contribute to the pathobiology of common risk factors for HF such as hypertension, diabetes mellitus, obesity, inflammation, and aging.

Preexistent natural genetic variation is a likely basis for differential risk among individuals. In this respect, mtDNA mutations have been proposed as the basis for bioenergetic adaptation to changes in diet and climate in prehistoric times, and today are hypothesized to contribute to cardiometabolic disease susceptibility. 42–44 Natural variation of mtDNA background, combined with the coevolution of the nDNA, may affect the cellular response to environmental stimuli and can account for the complexity of an individual’s predisposition to metabolic disease. 41 Thus, different mtDNA-nDNA combinations can affect susceptibility to HF. The concept of mitochondrial-nuclear genetic interaction, or mito-Mendelian genetics, may provide a means of central control for cellular function. Aging and sex are intrinsic determinants that may affect the mitochondrial phenotype. Aging, a major risk factor for HF, is associated with the degradation of nuclear and mitochondrial genetic integrity because of telomere shortening, a process that is opposed by telomerase reverse transcriptase. Mitochondrial telomerase reverse transcriptase is associated with beneficial and protective effects including improved metabolism, reduced ROS, and increased mtDNA integrity. 46 Sex differences have been noted with regard to mitochondria in the heart, 47,48 but the functional consequences are not clear. Estrogen and androgen receptors have been localized to mitochondria in various cell types, and a number of sex differences have been described with regard to mitochondrial efficiency, ROS production, antioxidant capacity, substrate preference, and Ca 2+ handling in the heart. Studies have also noted key sex-dependent differences in the expression and PTM of key mitochondrial proteins.

Mitochondrial dysfunction may play a role in the pathobiology of important risk factors for HF including hypertension, diabetes mellitus, obesity, inflammation, and aging (Figure 2). Mitochondria are a major source of ROS, such as superoxide and hydrogen peroxide, that lead to target organ damage, dysfunction, hypertrophy, and inflammation. Hypertension is associated with depletion and inactivation of the key mitochondrial deacetylase, sirtuin 3, which is involved in the regulation of key metabolic steps. Depletion of sirtuin 3 promotes development of hypertension and cardiac fibrosis. Diabetes mellitus and obesity promote a shift in mitochondrial phenotype in favor of fatty acid oxidation and attenuation of glucose oxidation. Some of the earliest consequences of enhanced fatty acid flux in the diabetic heart are increased mitochondrial uncoupling and generation of ROS, which result in decreased ATP synthesis despite the high energy demand. Initially adaptive mitochondrial responses may become maladaptive with depressed oxidative phosphorylation activity, reduced protein expression, impaired function of electron transport chain complexes, and the accumulation of damaged mitochondria. Cardiac magnetic resonance spectroscopy has shown that stores of high-energy creatine phosphate are decreased in association with oxidative modifications in electron transport chain proteins. 49,50 The mitochondrial phenotype in type 2 diabetes mellitus/obesity appears to differ from that in models of HFrEF, suggesting that diabetes mellitus and obesity may interact with other risk factors for HF. Inflammation is a common feature associated with hypertension, diabetes mellitus, obesity, and other risk factors for HF. Although the levels of inflammatory cytokines are elevated in HF and correlate with its severity and prognosis, the source(s) of circulating cytokines have not been identified. As with other solid organs, however, the heart possesses a modest population of resident macrophages and dendritic cells that provide an immune surveillance function to maintain cellular homeostasis. The ancestral prokaryotic features of mitochondria retain bacterial signatures, which when released after mitochondrial stress, may function as damage-associated molecular patterns to activate innate immunity that can amplify subsequent inflammation. 51 Cardiomyocytes and skeletal muscle cells may thus function as nonprofessional immune cells that respond to perturbed mitochondrial quality control by initiating intracellular immune surveillance programs to cause inflammation, thereby promoting cardiac and skeletal muscle dysfunction.

Research Gaps and Opportunities

Mitochondrial phenotyping and genotyping in humans and animal models of HF, in combination with large nuclear gene-sequence databases, are needed to address several questions: How does mitochondrial phenotype differ in HF of various causes? How do underlying nuclear and mitochondrial genotype, sex, and age affect mitochondrial phenotype and contribute to increased or decreased susceptibility to HF or HF risk factors such as hypertension, diabetes mellitus, obesity, aging, and inflammation? How do changes in mitochondrial bioenergetics mediated by functional mutations in the mtDNA and nDNA modulate cellular pathways and functions? How do the resulting changes in oxidant signaling and metabolite flux (eg, citric acid cycle) shape the host cell response to changes in the cellular microenvironment? To address these and related questions, it is important to develop a nuclear-mitochondrial genomic fingerprint in humans for comparing longitudinal data sets within cohorts. A long-term challenge is identification of the mechanisms by which different mtDNA-nDNA genetic variations alter metabolic systems that affect disease susceptibility. Likewise, it is important to determine how sex and age affect mitochondrial phenotype and genotype. What is the role of age-related telomere shortening in increasing the incidence of HF with age, and does telomerase reverse transcriptase activation represent a potentially useful strategy for reducing the adverse consequences of cardiovascular aging? How do sex-related differences in mitochondrial biology and function contribute to HF? Is there a link between mitochondrial dysfunction and inflammation in the pathophysiology of HF? The answers to these and related questions will expand our knowledge of disease mechanisms in HF and identify opportunities for precision medicine in its prevention and treatment.

New Tools

As in other disciplines, advances in the cutting-edge technologies of genomics, proteomics, transcriptomics, metabolomics, and epigenomics have been driving ground-breaking discoveries in the field of mitochondrial biology. Our understanding of mitochondrial physiology is further enhanced by the availability of compartment-specific sensors and the ability to assess mitochondrial respiration in intact cells. Despite increases in our knowledge, we lack the ability to discern and deeply phenotype mitochondria with respect to cardiac diseases. It is important to note that enabling tools and technologies for translating basic discoveries to clinical practice are urgently needed.

Research Gaps and Opportunities

First, we lack rigorous, quantitative information on genome, proteome, transcriptome, metabolome, interactome, and epigenome dynamics in different cells underlying cardiac phenotypes (both model systems and clinical cohorts, eg, Trans-Omics for Precision Medicine). Specific examples of critically important gaps include (1) understanding mtDNA sequences in renewable cell lines (eg, lymphoblastoid cell lines) as they relate to whole blood–derived DNA in the same individuals and how they are related to mitochondrial function and disease phenotypes (2) elucidating a comprehensive map of proteome PTMs (3) resolving the mechanism by which mitochondria communicate and signal to other compartments and (4) obtaining quantitative structural information defining the interface/crosstalk between protein-protein interactomes. Second, we lack computational tools and platforms for deep phenotyping, which include extracting information from data sets and discerning causality versus correlation. We need new machine learning–based strategies for data analysis that are capable of embracing the inherent complexities and volume of mitochondrial data sets. Third, we lack well-organized data sets and we are missing a well-curated mitochondrial molecular atlas integrating various omics, functional and clinical data sets (eg, mitochondrial knowledge graph). Existing data sets and tools are helpful but limited, and the relational organization of diverse data sets is obscure. As they currently stand, they are inadequate to support clinical translation and application of mitochondrial knowledge. Fourth, model systems are missing for integrated in vivo studies, such as large-animal models, reporter mice, and models for investigating sex differences. These tools are critical to bridging the test tube and computational results to clinical testing. Fifth, we need tools to assess mitochondrial function in patients with CVD. Because tissue sampling for mitochondrial analysis is at present impractical and largely impossible for HF, the current translational effort is seriously limited by the lack of surrogate readouts of tissue mitochondrial integrity or biomarkers of mitochondrial dysfunction. Last, only very limited tools/reagents are available to target mitochondria in the in vivo setting. Vehicles for delivering mitochondria-targeted cargo or methods to manipulate mtDNA remain a serious challenge in translational research.

These challenges set the stage for several exciting opportunities in mitochondrial research, illuminating the many ways that tools and technology may help to advance cardiovascular biology and medicine. Key action items were identified in the workshop to ultimately realize these opportunities. For the first time, we possess the tools to conduct deep clinical phenotyping by obtaining comprehensive, quantitative omics information for relevant phenotype via NHLBI’s Trans-Omics for Precision Medicine program. This will provide unparalleled characterization of genomic, proteomic, transcriptomic, metabolomic, and epigenetic information and will facilitate interactomic analysis. Further development of these tools and their capacities to resolve the different omic dimensions will enable the identification of statistically significant and biologically relevant changes in protein-protein, protein-metabolite, and metabolite-metabolite interactions to reveal molecular-level details underpinning mitochondrial dysfunction. 52–55 We can create model systems in which we systematically link mtDNA sequence information to functional phenotypes to understand the impact of genetic variations. Further investigations of mtDNA homoplasmic and heteroplasmic mutations 56 in different cell types will provide an important foundation for understanding the contributions of the mitochondrial genome to disease. With the use of powerful artificial intelligence and machine learning–based computational platforms, a large volume of information can be integrated into a relational molecular atlas through construction of a mitochondrial knowledge graph, providing a live demonstration of how omics/functional/clinical data sets can facilitate our understanding of HF pathogenesis. 57,58

The translation of mitochondria-based therapies will require new and improved methods to assess and modify mitochondrial function. The development and sharing of animal models that report mitochondrial function in an integrated setting will accelerate the translational effort. The development of biomarkers and imaging modalities that allow noninvasive and longitudinal assessment of mitochondrial function in patients is critical for moving discoveries to clinical testing and care. Biomarkers will also be used to identify patient subsets with different mitochondrial deficits that are most likely to respond to specific forms of therapy and to track target engagement with such interventions. In accordance with the diverse examples we provided, these tools require integration of analytical, biochemical, molecular, genetic, bioinformatics, and machine-learning approaches that measure and decode mitochondrial properties in vivo and in vitro. Innovations will advance research and therapeutic targeting of mitochondria.

Translation to Patients

Mitochondrial dysfunction has been implicated in the pathophysiology of HF of multiple etiologies and types (eg, HFrEF versus HFpEF). There has been significant progress in understanding the mechanisms of mitochondrial dysfunction and the identification of potential therapeutic targets. The successful translation of mitochondria-targeted therapy to patients with HF, however, remains elusive, with few clinical trials to date. Several promising approaches have been suggested by studies in animal models of HF including mitochondria-targeted antioxidants, interventions to restore NADH redox balance, elamipretide, coenzyme Q, resveratrol, and a variety of small molecules that may act by preventing mitochondrial sodium overload. Although each has had some level of success in preclinical models, very few have advanced to clinical trials in patients with HF. It is noteworthy that all of these potential therapies appear to act in ways that differ from currently used drugs for HF. The antioxidants decrease the effects of excessive ROS, elamipretide may stabilize cardiolipin, NAD + precursors and resveratrol may augment sirtuin activity, whereas ranolazine, sodium/glucose cotransporter 2 inhibitors, and CGP-37157 may act by correcting sodium/calcium balance in mitochondria. 59 Thus, the mechanism of action of mitochondria-targeted therapies, in general, is likely to be orthogonal to current therapies and thereby has the potential to exert complementary beneficial effects.

One of the few mitochondria-targeted therapies to be tested in humans is coenzyme Q, a lipid-soluble electron carrier that plays a central role in electron transport and ATP synthesis. Decreased levels of myocardial and plasma coenzyme Q have been observed in some studies of patients with HF, and meta-analyses have suggested a possible clinical benefit. 60,61 Q-SYMBIO (Coenzyme Q10 as adjunctive treatment of chronic heart failure: a randomized, double-blind, multicenter trial with focus on symptoms, biomarker status), a randomized trial of coenzyme Q for 2 years in 420 patients with New York Heart Association class III and IV HFrEF showed promising decreases in morbidity and mortality. 62 Elamipretide is a small peptide that targets the mitochondrial inner membrane where association with cardiolipin occurs, and it has been shown to lead to bioenergetic improvements in various models of mitochondrial dysfunction. In mice, elamipretide administered by osmotic pump for 4 weeks caused improvements in several aspects of adverse remodeling and cardiac function in models of HF caused by angiotensin infusion and pressure overload. 63,64 In dogs with embolization-induced HF, elamipretide administered by daily subcutaneous injection for 3 months led to improvements in myocardial hemodynamics and mitochondrial function. 65 Clinical experience with elamipretide in HF has so far been limited to 8 patients with HFrEF who received a single 4-hour infusion. 66 Elamipretide is also being tested in patients with diseases of mitochondrial dysfunction, including Barth syndrome, an ultrarare genetic condition caused by defective remodeling of mitochondrial cardiolipin that is characterized by dilated cardiomyopathy and skeletal myopathy.

Alterations in NADH and NADPH redox state have been observed in HF. In addition to the consequences for cellular redox regulation, NAD + is a cosubstrate for multiple enzymes, including sirtuin deacetylases, and thereby plays critical roles in PTMs of proteins by lysine that are important to multiple cellular metabolic processes and energy transduction. In murine HF models, the administration of NAD + or the NAD + precursors nicotinamide mononucleotide or nicotinamide riboside slowed the development of HF. 13,67,68 Nicotinamide riboside has entered clinical trials in systolic HF. Resveratrol is a naturally occurring polyphenol with pleotropic effects including antioxidant properties and activation of sirtuins. Preclinical studies have suggested that resveratrol and synthetic, related flavonoid derivatives exert beneficial effects in a variety of HF models, and at least one randomized clinical trial of resveratrol is under way in patients with HF.

A major issue in the translation of mitochondria-targeted therapies to humans is the difficulty inherent in assessing cardiac mitochondrial function in vivo. Positron emission tomography and magnetic resonance spectroscopy are the currently available noninvasive approaches for studying cardiac metabolism in vivo. The former has been primarily used to quantify substrate uptake and utilization and the latter for measuring cardiac high-energy phosphates and turnover with 31P magnetic resonance spectroscopy. These techniques have been applied in both animal models of HF and in humans with HF, leading to several important findings including that (1) cardiac substrate utilization is altered in HF, (2) cardiac high-energy phosphate pools and the rate of ATP turnover through the creatine kinase reaction are significantly depressed, (3) energetic changes often precede ventricular function and remodeling changes, and (4) depressed cardiac energetics independently predict clinical HF outcomes and mortality. An alternative or complementary approach to assessing mitochondrial function in humans is the measurement of bioenergetics/mitochondrial function by respirometry or extracellular flux analysis in human circulating platelets and leukocytes. 69–71 Observations in a variety of diseases including type 2 diabetes mellitus, pulmonary arterial hypertension, and asthma have led to the suggestion that mitochondrial alterations in circulating cells could be used as a biomarker of disease. It remains to be determined how and to what extent systemic mitochondrial function relates to mitochondrial function in target organs including the heart, skeletal muscle, and vasculature.

Research Gaps and Opportunities

Although preclinical work suggests the potential benefit of mitochondria-targeted therapies in HF, it remains to be established whether improving mitochondrial function will result in improved clinical outcomes in patients. As highlighted in the prior sections, success in translation to patients will critically depend on continued progress in understanding the pathophysiologic role of mitochondria in HF and the development of effective mitochondria-targeted therapeutic agents. Clinical translation presents obstacles not faced in preclinical studies. Patients are heterogeneous with regard to a myriad of factors that can affect mitochondrial genotype and phenotype, and thus, the susceptibility to both metabolic challenges and risk factors for HF such as hypertension, diabetes mellitus, obesity, aging, and inflammation. In addition to this inherent variability from patient to patient, HF is a syndrome with numerous causes that have distinct and overlapping pathophysiologic mechanisms. Consequently, important patient- and cause-specific differences in mitochondrial phenotype may affect the success of a particular mitochondria-targeted intervention. Accordingly, it will be important that the relationship between HF pathophysiology and mitochondrial phenotype be clarified and considered in the design of clinical trials of mitochondria-targeted agents. This pathophysiologic heterogeneity may be particularly important in regard to patients with HFpEF versus HFrEF.

Another hurdle is the difficulty in measuring mitochondrial function in humans. Positron emission tomography and magnetic resonance spectroscopy are demanding technologies that are not widely available outside teaching and research institutions. 31P and 13C magnetic resonance spectroscopy have limited sensitivity and relatively low spatial resolution, and their demonstrated ability to provide sophisticated measures of cardiac metabolism in isolated hearts has not been fully translated to in vivo studies in humans. Better, more robust, noninvasive means to measure in vivo mitochondrial function and energy metabolism are critically needed to probe the links between impaired mitochondrial metabolism and HF development in patients, to identify subsets of patients most likely to respond to a particular therapy, to assess the effect of therapy on mitochondrial function, and to determine the relationship of changes in mitochondrial function to clinical parameters. Biomarkers and assays indicative of mitochondrial function in circulating cells could provide useful guidance in patient selection and assessment of therapeutic responses. Further studies are needed to determine the relations between such biomarkers and mitochondrial function in target organs such as heart and skeletal muscle. Last, as opportunities for clinical trials increase, it is important to develop a network-based approach that facilitates the use of standardized protocols, measurements, and strategies, including stratification on key factors such as disease etiology, age, race, and sex.

Summary and Recommendations

The 2-day workshop identified 5 priority areas for research to promote the development of mitochondria-directed therapies for HF. Several of the priorities addressed the need to better understand relevant aspects of basic mitochondrial biology including the multiplicity of functions played by mitochondria, their role in intra- and intercellular communication, and the relationship between mitochondrial genotype and phenotype (Figure 2). Other priority areas emphasized the need for improved tools for the study of mitochondrial function in humans and the development of strategies for the translation of basic observations to the clinic. Recommendations fell into 3 main areas.

Recommendation 1

Foster multidisciplinary and integrated systems–based approaches to define the mechanistic role of mitochondria in cardiovascular health and disease and to discover novel therapeutic opportunities.

Studies integrating the multiplicities of mitochondrial function extending beyond energy provision, such as mitochondria-nuclear communication, PTMs, ROS, and Ca 2+ homeostasis, and regulation of inflammatory responses, are critical to further development of the field and identification of new therapies.

The genotype-phenotype relationship warrants further investigation and should be included, such as assessment of the influence of age, sex, and race on the mitochondrial responses to stress.

It is highly recommended that future projects take a team science approach by bringing together investigators across disciplines, such as biologists, informaticians, engineers, epidemiologists, and clinicians.

Recommendation 2

Develop and share with the community novel reagents, tools, and knowledge for mitochondrial research in humans and in translationally relevant animal and cellular models, including imaging modalities, pharmacological agents, biomarkers, multiomics, and computational approaches.

Develop novel reagents and tools to not only provide multiomics information on mitochondria, but also generate knowledge maps that enable rapid translation of laboratory discoveries.

Technical innovations in modeling, intervening, and monitoring mitochondrial function in vitro and in vivo are particularly encouraged.

Biomarkers that identify specific patient populations who will benefit from mitochondria-based therapy are of high significance for clinical trials.

Recommendation 3

Foster translational approaches to mitochondrial therapies for rapid translation of novel therapies and discoveries into clinical trials.

Develop techniques to assess cardiac mitochondrial function in patients with HF, including identification of biomarkers, to identify mitochondrial phenotype, guide patient selection, and assess the effects of new therapeutic agents on mitochondrial function.

Perform proof-of-concept studies in relatively homogeneous patient populations selected with regard to key determinants of mitochondrial phenotype to assess the relationship between mitochondrial function and cardiac performance.

Use a network-based approach to clinical development that standardizes protocols, measurements, and strategies for patient selection.


The views expressed in this article are those of the authors and do not necessarily represent the views of the National Heart, Lung, and Blood Institute the National Institutes of Health or the US Department of Health and Human Services.

Sources of Funding

This workshop was supported by the National Heart, Lung, and Blood Institute (NHLBI), National Institutes of Health (NIH). Relevant grant support of authors is as listed in the following: NIH R01GM086688 (to Dr Bruce) NIH R01DK103750, NIH 1UL1TR001430, and American Heart Association 16GRNT27660006 (to Dr Bachscmid) NIHR01DK098646, NIH R01DK100826, and American Heart Association 16GRNT30990018 (to Dr Boudina) NIH RO1HL064750 and NIH NO1HV-28178 (to Dr Colucci) NIH R35HL135736, NIH R01HL128071, and the Harrington Discovery Institute (to Dr Dorn) NIH R01HL124116 (to Dr Dikalov) NIH P01HL112730, NIH R01HL132075, NIH R01HL144509 (to Dr Gottlieb) NIH R01HL128349, NIH R01HL058493 (to Dr Kelly) NIH R01HL136496 (to Dr Kohr) NIH R01HL52141 (to Dr Mochly-Rosen) NIH R01HL128071, NIH R01HL130861, NIH R01HL138475 (to Dr Kitsis) NIH R01HL049244, NIH R01HL132525, and NIH R01HL113057 (to Dr Lewandowski) NIH R01HL125695 (to Dr McClung) NIH R01HL137259 and NIH R01HL134821 (to Dr O’Rourke) NIH R21HL126209 and NIH P01HL092969 (to Dr O’Brien) NIH R01HL126952 (to Dr Park) NIH R01HL133003-01A1, NIH R01GM113816-01, and Hemophilia Center of Western Pennsylvania (to Dr Shiva) NIH R01HL122124, NIH R01HL093671, NIH R01HL137426, NIH R01HL142864, and NIH R01HL137266 (to Dr Sheu) NIH R01HL110349, NIH R01HL129510, NIH R01HL126209, and NIH R01HL142628 (to Dr Tian) NIH R01NS021328, NIH R01MH108592, NIH R01OD010944, and US Department of Defense W81XWH-16-1-0401 (to Dr Wallace) and NIH R01HL61912 and NIH R01HL63030 (to Dr Wong).


Dr Dorn is a founder of Mitochondria in Motion, Inc., a Saint Louis, Missouri–based biotech research and development company focused on enhancing mitochondrial trafficking and fitness in neurodegenerative diseases. Dr Kitsis is a co-founder of Aspida Therapeutics Inc. and consultant for Amaron Bio. Dr Vernon is the principal investigator on a clinical trial sponsored by Stealth Biotherapeutics, Newton, MA.


School of Medicine, Boston University School of Medicine, Boston, MA, USA

Renal Section, Department of Medicine, Boston University School of Medicine, Boston, MA, USA

Mohamed Hassan Kamel & Vipul C. Chitalia

Boston Veterans Affairs Healthcare System, Boston, MA, USA

Global Co-creation Lab, Institute of Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA, USA

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar


All authors contributed equally to writing and reviewing/editing before submission.

Corresponding author



Non-alcoholic fatty liver disease (NAFLD) is rapidly becoming a global health problem. Cardiovascular diseases (CVD) are the most common cause of mortality in NAFLD patients. NAFLD and CVD share several common risk factors including obesity, insulin resistance, and type 2 diabetes (T2D). Atherogenic dyslipidemia, characterized by plasma hypertriglyceridemia, increased small dense low-density lipoprotein (LDL) particles, and decreased high-density lipoprotein cholesterol (HDL-C) levels, is often observed in NAFLD patients.

Scope of review

In this review, we highlight recent epidemiological studies evaluating the link between NAFLD and CVD risk. We further focus on recent mechanistic insights into the links between NAFLD and altered lipoprotein metabolism. We also discuss current therapeutic strategies for NAFLD and their potential impact on NAFLD-associated CVD risk.

Major conclusions

Alterations in hepatic lipid and lipoprotein metabolism are major contributing factors to the increased CVD risk in NAFLD patients, and many promising NASH therapies in development also improve dyslipidemia in clinical trials.


Szabo, S., Tache, Y. & Somogyi, A. The legacy of Hans Selye and the origins of stress research: a retrospective 75 years after his landmark brief “letter” to the editor of Nature. Stress 15, 472–478 (2012).

Levine, S. Influence of psychological variables on the activity of the hypothalamic-pituitary-adrenal axis. Eur. J. Pharmacol. 405, 149–160 (2000).

Brown, S. A. & Azzi, A. Peripheral circadian oscillators in mammals. Handb. Exp. Pharmacol. 2013, 45–66 (2013).

Roenneberg, T. & Merrow, M. The circadian clock and human health. Curr. Biol. 26, R432–R443 (2016).

Bass, J. & Lazar, M. A. Circadian time signatures of fitness and disease. Science 354, 994–999 (2016).

Turek, F. W. Circadian neural rhythms in mammals. Annu. Rev. Physiol. 47, 49–64 (1985).

Skene, D. J. et al. Separation of circadian- and behavior-driven metabolite rhythms in humans provides a window on peripheral oscillators and metabolism. Proc. Natl Acad. Sci. USA 115, 7825–7830 (2018).

Buhr, E. D. & Takahashi, J. S. Molecular components of the mammalian circadian clock. Handb. Exp. Pharmacol. 217, 3–27 (2013).

Takahashi, J. S. Transcriptional architecture of the mammalian circadian clock. Nat. Rev. Genet. 18, 164–179 (2017).

Baron, K. G. & Reid, K. J. Circadian misalignment and health. Int. Rev. Psychiatry 26, 139–154 (2014).

Potter, G. D. et al. Circadian rhythm and sleep disruption: causes, metabolic consequences, and countermeasures. Endocr. Rev. 37, 584–608 (2016).

Zhang, R., Lahens, N. F., Ballance, H. I., Hughes, M. E. & Hogenesch, J. B. A circadian gene expression atlas in mammals: implications for biology and medicine. Proc. Natl Acad. Sci. USA 111, 16219–16224 (2014).

Smarr, B. L. & Schirmer, A. E. 3.4 million real-world learning management system logins reveal the majority of students experience social jet lag correlated with decreased performance. Sci. Rep. 8, 4793 (2018).

Gardner, M. et al. Dysregulation of the hypothalamic pituitary adrenal (HPA) axis and cognitive capability at older ages: individual participant meta-analysis of five cohorts. Sci. Rep. 9, 4555 (2019).

Selye, H. Stress and the general adaptation syndrome. BMJ 1, 1383–1392 (1950).

Russell, G. M. & Lightman, S. L. Can side effects of steroid treatments be minimized by the temporal aspects of delivery method? Expert Opin. Drug Saf. 13, 1501–1513 (2014).

Sorrells, S. F. & Sapolsky, R. M. An inflammatory review of glucocorticoid actions in the CNS. Brain Behav. Immun. 21, 259–272 (2007).

Busillo, J. M. & Cidlowski, J. A. The five Rs of glucocorticoid action during inflammation: ready, reinforce, repress, resolve, and restore. Trends Endocrinol. Metab. 24, 109–119 (2013).

McEwen, B. S. et al. The role of adrenocorticoids as modulators of immune function in health and disease: neural, endocrine and immune interactions. Brain Res. Rev. 23, 79–133 (1997).

Elenkov, I. J. & Chrousos, G. P. Stress system — organization, physiology and immunoregulation. Neuroimmunomodulation 13, 257–267 (2006).

Brinkmann, V. & Kristofic, C. Regulation by corticosteroids of Th1 and Th2 cytokine production in human CD4+ effector T cells generated from CD45RO- and CD45RO+ subsets. J. Immunol. 155, 3322–3328 (1995).

Wiegers, G. J. & Reul, J. M. Induction of cytokine receptors by glucocorticoids: functional and pathological significance. Trends Pharmacol. Sci. 19, 317–321 (1998).

Abraham, I. M., Meerlo, P. & Luiten, P. G. Concentration dependent actions of glucocorticoids on neuronal viability and survival. Dose Response 4, 38–54 (2006).

Plaschke, K., Muller, D. & Hoyer, S. Effect of adrenalectomy and corticosterone substitution on glucose and glycogen metabolism in rat brain. J. Neural Transm. 103, 89–100 (1996).

Belanoff, J. K., Gross, K., Yager, A. & Schatzberg, A. F. Corticosteroids and cognition. J. Psychiatr. Res. 35, 127–145 (2001).

Roozendaal, B. Stress and memory: opposing effects of glucocorticoids on memory consolidation and memory retrieval. Neurobiol. Learn. Mem. 78, 578–595 (2002).

Brown, E. S. Effects of glucocorticoids on mood, memory, and the hippocampus. Treatment and preventive therapy. Ann. NY Acad. Sci. 1179, 41–55 (2009).

de Kloet, E. R., Oitzl, M. S. & Joels, M. Stress and cognition: are corticosteroids good or bad guys? Trends Neurosci. 22, 422–426 (1999).

Decani, S., Federighi, V., Baruzzi, E., Sardella, A. & Lodi, G. Iatrogenic Cushing’s syndrome and topical steroid therapy: case series and review of the literature. J. Dermatolog. Treat. 25, 495–500 (2014).

Kenna, H. A., Poon, A. W., de los Angeles, C. P. & Koran, L. M. Psychiatric complications of treatment with corticosteroids: review with case report. Psychiatry Clin. Neurosci. 65, 549–560 (2011).

Buijs, R. M., Markman, M., Nunes-Cardoso, B., Hou, Y. X. & Shinn, S. Projections of the suprachiasmatic nucleus to stress-related areas in the rat hypothalamus: a light and electron microscopic study. J. Comp. Neurol. 335, 42–54 (1993).

Watts, A. G. & Swanson, L. W. Efferent projections of the suprachiasmatic nucleus: II. Studies using retrograde transport of fluorescent dyes and simultaneous peptide immunohistochemistry in the rat. J. Comp. Neurol. 258, 230–252 (1987).

Jacobson, L. Hypothalamic-pituitary-adrenocortical axis regulation. Endocrinol. Metab. Clin. North Am. 34, 271–292 (2005).

Herman, J. P., Ostrander, M. M., Mueller, N. K. & Figueiredo, H. Limbic system mechanisms of stress regulation: hypothalamo-pituitary-adrenocortical axis. Prog. Neuropsychopharmacol. Biol. Psychiatry 29, 1201–1213 (2005).

Dallman, M. F. et al. Corticosteroids and the control of function in the hypothalamo-pituitary-adrenal (HPA) axis. Ann. NY Acad. Sci. 746, 22–31 (1994).

Jasper, M. S. & Engeland, W. C. Splanchnic neural activity modulates ultradian and circadian rhythms in adrenocortical secretion in awake rats. Neuroendocrinology 59, 97–109 (1994).

Buijs, R. M. et al. Anatomical and functional demonstration of a multisynaptic suprachiasmatic nucleus adrenal (cortex) pathway. Eur. J. Neurosci. 11, 1535–1544 (1999).

Kiessling, S., Sollars, P. J. & Pickard, G. E. Light stimulates the mouse adrenal through a retinohypothalamic pathway independent of an effect on the clock in the suprachiasmatic nucleus. PLOS ONE 9, e92959 (2014).

Husse, J., Leliavski, A., Tsang, A. H., Oster, H. & Eichele, G. The light-dark cycle controls peripheral rhythmicity in mice with a genetically ablated suprachiasmatic nucleus clock. FASEB J. 28, 4950–4960 (2014).

Ishida, A. et al. Light activates the adrenal gland: timing of gene expression and glucocorticoid release. Cell Metab. 2, 297–307 (2005).

Oster, H. et al. The circadian rhythm of glucocorticoids is regulated by a gating mechanism residing in the adrenal cortical clock. Cell Metab. 4, 163–173 (2006).

Charmandari, E. et al. Peripheral CLOCK regulates target-tissue glucocorticoid receptor transcriptional activity in a circadian fashion in man. PLOS ONE 6, e25612 (2011).

Bailey, S. L. & Heitkemper, M. M. Circadian rhythmicity of cortisol and body temperature: morningness-eveningness effects. Chronobiol. Int. 18, 249–261 (2001).

Donner, N. C., Montoya, C. D., Lukkes, J. L. & Lowry, C. A. Chronic non-invasive corticosterone administration abolishes the diurnal pattern of tph2 expression. Psychoneuroendocrinology 37, 645–661 (2012).

Lightman, S. L. The neuroendocrinology of stress: a never ending story. J. Neuroendocrinol. 20, 880–884 (2008).

Nicolaides, N. C., Kyratzi, E., Lamprokostopoulou, A., Chrousos, G. P. & Charmandari, E. Stress, the stress system and the role of glucocorticoids. Neuroimmunomodulation 22, 6–19 (2015).

Gibbison, B. et al. Dynamic pituitary-adrenal interactions in response to cardiac surgery. Crit. Care Med. 43, 791–800 (2015).

Spiga, F. et al. Dynamic responses of the adrenal steroidogenic regulatory network. Proc. Natl Acad. Sci. USA 114, E6466–E6474 (2017).

Ma, X. M., Levy, A. & Lightman, S. L. Emergence of an isolated arginine vasopressin (AVP) response to stress after repeated restraint: a study of both AVP and corticotropin-releasing hormone messenger ribonucleic acid (RNA) and heteronuclear RNA. Endocrinology 138, 4351–4357 (1997).

Dallman, M. F. Stress update: adaptation of the hypothalamic-pituitary-adrenal axis to chronic stress. Trends. Endocrinol. Metab. 4, 62–69 (1993).

Henley, D. E. et al. Hypothalamic-pituitary-adrenal axis activation in obstructive sleep apnea: the effect of continuous positive airway pressure therapy. J. Clin. Endocrinol. Metab. 94, 4234–4242 (2009).

Boonen, E. et al. Reduced cortisol metabolism during critical illness. N. Engl. J. Med. 368, 1477–1488 (2013).

Peeters, B. et al. Adrenocortical function during prolonged critical illness and beyond: a prospective observational study. Intensive Care Med. 44, 1720–1729 (2018).

De Kloet, E. R., Vreugdenhil, E., Oitzl, M. S. & Joels, M. Brain corticosteroid receptor balance in health and disease. Endocr. Rev. 19, 269–301 (1998).

Reul, J. M. & de Kloet, E. R. Two receptor systems for corticosterone in rat brain: microdistribution and differential occupation. Endocrinology 117, 2505–2511 (1985).

Dallman, M. F. Fast glucocorticoid actions on brain: back to the future. Front. Neuroendocrinol. 26, 103–108 (2005).

Russell, G. M. et al. Rapid glucocorticoid receptor-mediated inhibition of hypothalamic-pituitary-adrenal ultradian activity in healthy males. J. Neurosci. 30, 6106–6115 (2010).

Lowenberg, M., Verhaar, A. P., van den Brink, G. R. & Hommes, D. W. Glucocorticoid signaling: a nongenomic mechanism for T cell immunosuppression. Trends Mol. Med. 13, 158–163 (2007).

Orchinik, M., Murray, T. F., Franklin, P. H. & Moore, F. L. Guanyl nucleotides modulate binding to steroid receptors in neuronal membranes. Proc. Natl Acad. Sci. USA 89, 3830–3834 (1992).

Orchinik, M., Murray, T. F. & Moore, F. L. A corticosteroid receptor in neuronal membranes. Science 252, 1848–1851 (1991).

Joels, M., Pasricha, N. & Karst, H. The interplay between rapid and slow corticosteroid actions in brain. Eur. J. Pharmacol. 719, 44–52 (2013).

Walker, J. J. et al. The origin of glucocorticoid hormone oscillations. PLOS Biol. 10, e1001341 (2012).

Patel, P. D. et al. Glucocorticoid and mineralocorticoid receptor mRNA expression in squirrel monkey brain. J. Psychiatr. Res. 34, 383–392 (2000).

Groeneweg, F. L., Karst, H., de Kloet, E. R. & Joels, M. Rapid non-genomic effects of corticosteroids and their role in the central stress response. J. Endocrinol. 209, 153–167 (2011).

de Kloet, E. R., Fitzsimons, C. P., Datson, N. A., Meijer, O. C. & Vreugdenhil, E. Glucocorticoid signaling and stress-related limbic susceptibility pathway: about receptors, transcription machinery and microRNA. Brain Res. 1293, 129–141 (2009).

Russell, G. M., Kalafatakis, K. & Lightman, S. L. The importance of biological oscillators for HPA activity and tissue glucocorticoid response: coordinating stress and neurobehavioural adaptation. J. Neuroendocrinol. 27, 378–388 (2015).

Lewis, J. G. et al. Plasma variation of corticosteroid-binding globulin and sex hormone-binding globulin. Horm. Metab. Res. 38, 241–245 (2006).

Lewis, J. G., Bagley, C. J., Elder, P. A., Bachmann, A. W. & Torpy, D. J. Plasma free cortisol fraction reflects levels of functioning corticosteroid-binding globulin. Clin. Chim. Acta 359, 189–194 (2005).

Hammond, G. L., Smith, C. L. & Underhill, D. A. Molecular studies of corticosteroid binding globulin structure, biosynthesis and function. J. Steroid Biochem. Mol. Biol. 40, 755–762 (1991).

Frairia, R. et al. Influence of naturally occurring and synthetic glucocorticoids on corticosteroid-binding globulin-steroid interaction in human peripheral plasma. Ann. NY Acad. Sci. 538, 287–303 (1988).

Cameron, A. et al. Temperature-responsive release of cortisol from its binding globulin: a protein thermocouple. J. Clin. Endocrinol. Metab. 95, 4689–4695 (2010).

Kyrou, I., Chrousos, G. P. & Tsigos, C. Stress, visceral obesity, and metabolic complications. Ann. NY Acad. Sci. 1083, 77–110 (2006).

Chapman, K., Holmes, M. & Seckl, J. 11beta-hydroxysteroid dehydrogenases: intracellular gate-keepers of tissue glucocorticoid action. Physiol. Rev. 93, 1139–1206 (2013).

Seckl, J. R. 11beta-hydroxysteroid dehydrogenases: changing glucocorticoid action. Curr. Opin. Pharmacol. 4, 597–602 (2004).

Verma, M. et al. 11beta-hydroxysteroid dehydrogenase-1 deficiency alters brain energy metabolism in acute systemic inflammation. Brain Behav. Immun. 69, 223–234 (2018).

Follenius, M., Simon, C., Brandenberger, G. & Lenzi, P. Ultradian plasma corticotropin and cortisol rhythms: time-series analyses. J. Endocrinol. Invest. 10, 261–266 (1987).

Hartmann, A., Veldhuis, J. D., Deuschle, M., Standhardt, H. & Heuser, I. Twenty-four hour cortisol release profiles in patients with Alzheimer’s and Parkinson’s disease compared to normal controls: ultradian secretory pulsatility and diurnal variation. Neurobiol. Aging 18, 285–289 (1997).

Rivest, R. W., Schulz, P., Lustenberger, S. & Sizonenko, P. C. Differences between circadian and ultradian organization of cortisol and melatonin rhythms during activity and rest. J. Clin. Endocrinol. Metab. 68, 721–729 (1989).

Waite, E. J. et al. Ultradian corticosterone secretion is maintained in the absence of circadian cues. Eur. J. Neurosci. 36, 3142–3150 (2012).

Ixart, G., Barbanel, G., Nouguier-Soule, J. & Assenmacher, I. A quantitative study of the pulsatile parameters of CRH-41 secretion in unanesthetized free-moving rats. Exp. Brain Res. 87, 153–158 (1991).

Spiga, F. et al. ACTH-dependent ultradian rhythm of corticosterone secretion. Endocrinology 152, 1448–1457 (2011).

Spiga, F., Liu, Y., Aguilera, G. & Lightman, S. L. Temporal effect of adrenocorticotrophic hormone on adrenal glucocorticoid steroidogenesis: involvement of the transducer of regulated cyclic AMP-response element-binding protein activity. J. Neuroendocrinol. 23, 136–142 (2011).

Lim, C. & Allada, R. Emerging roles for post-transcriptional regulation in circadian clocks. Nat. Neurosci. 16, 1544–1550 (2013).

Liston, C. et al. Circadian glucocorticoid oscillations promote learning-dependent synapse formation and maintenance. Nat. Neurosci. 16, 698–705 (2013).

Lightman, S. L. & Conway-Campbell, B. L. The crucial role of pulsatile activity of the HPA axis for continuous dynamic equilibration. Nat. Rev. Neurosci. 11, 710–718 (2010).

Stavreva, D. A. et al. Ultradian hormone stimulation induces glucocorticoid receptor-mediated pulses of gene transcription. Nat. Cell Biol. 11, 1093–1102 (2009).

Conway-Campbell, B. L., Pooley, J. R., Hager, G. L. & Lightman, S. L. Molecular dynamics of ultradian glucocorticoid receptor action. Mol. Cell. Endocrinol. 348, 383–393 (2012).

George, C. L., Lightman, S. L. & Biddie, S. C. Transcription factor interactions in genomic nuclear receptor function. Epigenomics 3, 471–485 (2011).

So, A. Y., Chaivorapol, C., Bolton, E. C., Li, H. & Yamamoto, K. R. Determinants of cell- and gene-specific transcriptional regulation by the glucocorticoid receptor. PLOS Genet. 3, e94 (2007).

Zalachoras, I., Houtman, R. & Meijer, O. C. Understanding stress-effects in the brain via transcriptional signal transduction pathways. Neuroscience 242, 97–109 (2013).

Sarabdjitsingh, R. A. et al. Stress responsiveness varies over the ultradian glucocorticoid cycle in a brain-region-specific manner. Endocrinology 151, 5369–5379 (2010).

Sarabdjitsingh, R. A. et al. Ultradian corticosterone pulses balance glutaminergic transmission and synaptic plasticity. Proc. Natl Acad. Sci. USA 111, 14265–14270 (2014).

Kalafatakis, K. et al. Ultradian rhythmicity of plasma cortisol is necessary for normal emotional and cognitive responses in man. Proc. Natl Acad. Sci. USA 115, E4091–E4100 (2018).

Gjerstad, J. K., Lightman, S. L. & Spiga, F. Role of glucocorticoid negative feedback in the regulation of HPA axis pulsatility. Stress 21, 403–416 (2018).

Bornstein, S. R., Engeland, W. C., Ehrhart-Bornstein, M. & Herman, J. P. Dissociation of ACTH and glucocorticoids. Trends Endocrinol. Metab. 19, 175–180 (2008).

Silverman, M. N., Miller, A. H., Biron, C. A. & Pearce, B. D. Characterization of an interleukin-6- and adrenocorticotropin-dependent, immune-to-adrenal pathway during viral infection. Endocrinology 145, 3580–3589 (2004).

Franchimont, D. et al. Adrenal cortical activation in murine colitis. Gastroenterology 119, 1560–1568 (2000).

Viblanc, V. A. et al. An integrative appraisal of the hormonal and metabolic changes induced by acute stress using king penguins as a model. Gen. Comp. Endocrinol. 269, 1–10 (2018).

Cruz-Topete, D. & Cidlowski, J. A. One hormone, two actions: anti- and pro-inflammatory effects of glucocorticoids. Neuroimmunomodulation 22, 20–32 (2015).

Biddie, S. C., Conway-Campbell, B. L. & Lightman, S. L. Dynamic regulation of glucocorticoid signalling in health and disease. Rheumatology 51, 403–412 (2012).

Miller, G. E., Cohen, S. & Ritchey, A. K. Chronic psychological stress and the regulation of pro-inflammatory cytokines: a glucocorticoid-resistance model. Health Psychol. 21, 531–541 (2002).

Oster, H. et al. The functional and clinical significance of the 24-hour rhythm of circulating glucocorticoids. Endocr. Rev. 38, 3–45 (2017).

Keller, M. et al. A circadian clock in macrophages controls inflammatory immune responses. Proc. Natl Acad. Sci. USA 106, 21407–21412 (2009).

Boivin, D. B. et al. Circadian clock genes oscillate in human peripheral blood mononuclear cells. Blood 102, 4143–4145 (2003).

Koo, J. W., Russo, S. J., Ferguson, D., Nestler, E. J. & Duman, R. S. Nuclear factor-kappaB is a critical mediator of stress-impaired neurogenesis and depressive behavior. Proc. Natl Acad. Sci. USA 107, 2669–2674 (2010).

Pace, T. W., Hu, F. & Miller, A. H. Cytokine-effects on glucocorticoid receptor function: relevance to glucocorticoid resistance and the pathophysiology and treatment of major depression. Brain Behav. Immun. 21, 9–19 (2007).

Pace, T. W. et al. Increased stress-induced inflammatory responses in male patients with major depression and increased early life stress. Am. J. Psychiatry 163, 1630–1633 (2006).

Cohen, S. et al. Chronic stress, glucocorticoid receptor resistance, inflammation, and disease risk. Proc. Natl Acad. Sci. USA 109, 5995–5999 (2012).

Spiegel, K., Leproult, R. & Van Cauter, E. Impact of sleep debt on metabolic and endocrine function. Lancet 354, 1435–1439 (1999).

Hauner, H., Schmid, P. & Pfeiffer, E. F. Glucocorticoids and insulin promote the differentiation of human adipocyte precursor cells into fat cells. J. Clin. Endocrinol. Metab. 64, 832–835 (1987).

Dallman, M. F. et al. Glucocorticoids, chronic stress, and obesity. Prog. Brain Res. 153, 75–105 (2006).

Tsigos, C. et al. Dose-dependent effects of recombinant human interleukin-6 on glucose regulation. J. Clin. Endocrinol. Metab. 82, 4167–4170 (1997).

McEwen, B. S. Sleep deprivation as a neurobiologic and physiologic stressor: allostasis and allostatic load. Metabolism 55, S20–S23 (2006).

Zhu, B., Shi, C., Park, C. G., Zhao, X. & Reutrakul, S. Effects of sleep restriction on metabolism-related parameters in healthy adults: a comprehensive review and meta-analysis of randomized controlled trials. Sleep Med. Rev. 45, 18–30 (2019).

Gavrila, A. et al. Diurnal and ultradian dynamics of serum adiponectin in healthy men: comparison with leptin, circulating soluble leptin receptor, and cortisol patterns. J. Clin. Endocrinol. Metab. 88, 2838–2843 (2003).

Knutson, K. L. & Van Cauter, E. Associations between sleep loss and increased risk of obesity and diabetes. Ann. NY Acad. Sci. 1129, 287–304 (2008).

Adam, T. C. & Epel, E. S. Stress, eating and the reward system. Physiol. Behav. 91, 449–458 (2007).

Young, E. A., Carlson, N. E. & Brown, M. B. Twenty-four-hour ACTH and cortisol pulsatility in depressed women. Neuropsychopharmacology 25, 267–276 (2001).

Heuser, I., Yassouridis, A. & Holsboer, F. The combined dexamethasone/CRH test: a refined laboratory test for psychiatric disorders. J. Psychiatr. Res. 28, 341–356 (1994).

Ising, M. et al. The combined dexamethasone/CRH test as a potential surrogate marker in depression. Prog. Neuropsychopharmacol. Biol. Psychiatry 29, 1085–1093 (2005).

Krishnan, V. & Nestler, E. J. The molecular neurobiology of depression. Nature 455, 894–902 (2008).

Miller, A. H. Depression and immunity: a role for T cells? Brain Behav. Immun. 24, 1–8 (2010).

Koo, J. W. & Duman, R. S. IL-1beta is an essential mediator of the antineurogenic and anhedonic effects of stress. Proc. Natl Acad. Sci. USA 105, 751–756 (2008).

Horowitz, M. A., Zunszain, P. A., Anacker, C., Musaelyan, K. & Pariante, C. M. Glucocorticoids and inflammation: a double-headed sword in depression? How do neuroendocrine and inflammatory pathways interact during stress to contribute to the pathogenesis of depression? Mod. Trends Pharmacopsychiatry 28, 127–143 (2013).

Munhoz, C. D. et al. Chronic unpredictable stress exacerbates lipopolysaccharide-induced activation of nuclear factor-kappaB in the frontal cortex and hippocampus via glucocorticoid secretion. J. Neurosci. 26, 3813–3820 (2006).

Pariante, C. M. Glucocorticoid receptor function in vitro in patients with major depression. Stress 7, 209–219 (2004).

Kenis, G. & Maes, M. Effects of antidepressants on the production of cytokines. Int. J. Neuropsychopharmacol. 5, 401–412 (2002).

Bjornsdottir, S. et al. Drug prescription patterns in patients with Addison’s disease: a Swedish population-based cohort study. J. Clin. Endocrinol. Metab. 98, 2009–2018 (2013).

Dunlop, D. Eighty-six cases of Addison’s disease. BMJ 2, 887–891 (1963).

Giordano, R. et al. Metabolic and cardiovascular profile in patients with Addison’s disease under conventional glucocorticoid replacement. J. Endocrinol. Invest. 32, 917–923 (2009).

Johannsson, G. et al. Adrenal insufficiency: review of clinical outcomes with current glucocorticoid replacement therapy. Clin. Endocrinol. 82, 2–11 (2015).

Lovas, K., Loge, J. H. & Husebye, E. S. Subjective health status in Norwegian patients with Addison’s disease. Clin. Endocrinol. 56, 581–588 (2002).

Feek, C. M. et al. Patterns of plasma cortisol and ACTH concentrations in patients with Addison’s disease treated with conventional corticosteroid replacement. Clin. Endocrinol. 14, 451–458 (1981).

Isidori, A. M. et al. Effect of once-daily, modified-release hydrocortisone versus standard glucocorticoid therapy on metabolism and innate immunity in patients with adrenal insufficiency (DREAM): a single-blind, randomised controlled trial. Lancet Diabetes Endocrinol. 6, 173–185 (2018).

Bancos, I. et al. Primary adrenal insufficiency is associated with impaired natural killer cell function: a potential link to increased mortality. Eur. J. Endocrinol. 176, 471–480 (2017).

Bjanesoy, T. E. et al. Altered DNA methylation profile in Norwegian patients with autoimmune Addison’s disease. Mol. Immunol. 59, 208–216 (2014).

Langenheim, J., Ventz, M., Hinz, A. & Quinkler, M. Modified-release prednisone decreases complaints and fatigue compared to standard prednisolone in patients with adrenal insufficiency. Horm. Metab. Res. 45, 96–101 (2013).

Mallappa, A. et al. A phase 2 study of Chronocort, a modified-release formulation of hydrocortisone, in the treatment of adults with classic congenital adrenal hyperplasia. J. Clin. Endocrinol. Metab. 100, 1137–1145 (2015).

Lovas, K. & Husebye, E. S. Continuous subcutaneous hydrocortisone infusion in Addison’s disease. Eur. J. Endocrinol. 157, 109–112 (2007).

Venneri, M. A. et al. Circadian rhythm of glucocorticoid administration entrains clock genes in immune cells: a DREAM trial ancillary study. J. Clin. Endocrinol. Metab. 103, 2998–3009 (2018).

Oksnes, M. et al. Continuous subcutaneous hydrocortisone infusion versus oral hydrocortisone replacement for treatment of Addison’s disease: a randomized clinical trial. J. Clin. Endocrinol. Metab. 99, 1665–1674 (2014).

Riedel, M., Wiese, A., Schurmeyer, T. H. & Brabant, G. Quality of life in patients with Addison’s disease: effects of different cortisol replacement modes. Exp. Clin. Endocrinol. 101, 106–111 (1993).

van Staa, T. P. et al. Use of oral corticosteroids in the United Kingdom. QJM 93, 105–111 (2000).

Overman, R. A., Yeh, J. Y. & Deal, C. L. Prevalence of oral glucocorticoid usage in the United States: a general population perspective. Arthritis Care Res. 65, 294–298 (2013).

Curtis, J. R. et al. Population-based assessment of adverse events associated with long-term glucocorticoid use. Arthritis Rheum. 55, 420–426 (2006).

McDonough, A. K., Curtis, J. R. & Saag, K. G. The epidemiology of glucocorticoid-associated adverse events. Curr. Opin. Rheumatol. 20, 131–137 (2008).

Leung, D. Y. & Bloom, J. W. Update on glucocorticoid action and resistance. J. Allergy Clin. Immunol. 111, 3–22 (2003).