Thursday, January 26, 2017

Senescence, cholesterol and adaptation

Senescence is an adaptive response to aging, failing mitochondria.

If the use of the word ‘adaptive’ seems counter intuitive, carrying as it does positive connotations, then related senescent adaptive responses include rising blood pressure and increased absorption of dietary cholesterol will seem heretical.


When I started my PhD on the aging of mitochondria now forty years ago we made a clear distinction between ‘aging’ and the onset of senescence. We had a colony of Wistar rats, a young adult rat was three months old, a middle-aged rat around 12-15-20  months and an old rat was 29-33 months. The latter were chosen when they showed signs of senescence, which as a rule of thumb was when they started to lose condition, moved around less, interacted less and lost muscle mass.  this state easily in humans the difference being that this onset would occur at over 70 years of age and not at 29 months! The maximum life span of our rats was 33-36 months whereas humans can live to more than 110 years.


Our assumption was that aging is a continuous process which starts (for us) in our twenties and is characterised by a steady deterioration which results in an elevated risk of malfunctions such as heart attacks, strokes and cancers but if these are avoided then barring accidents we make it to the senescent stage. With regard to mitochondria, we never found any mitochondrial changes pre-senescence and concluded that if they were there, they were too small for us to find. To my knowledge this is still the case today.
Organismal senescence I will argue is an adaptive response to the aging of mitochondria, albeit an apparently pointless adaptation as it does little more than prolong life-span. Below is why I think senescence is adaptive and the explanation involve sentropy.


From a simple thermodynamic point of view ‘we’ have a low entropy. For those unfamiliar with the concept of entropy it may suffice to say that for a system ( that is ‘us’ ) composed of countless particles arranged with such delicate, intricate, ordered complexity, comprises a highly improbable state of affairs maintained only by the continuous supply of energy from our surroundings. This supply comes from food and the ‘energy’ is liberated by mitochondria.  The energy that is used to do the work of building and maintaining our bodies is called Gibbs Free Energy (G) which is a compilation of simple heat energy (H) and entropy (S). It is formalised in the equation below:


G =H -TS system


Complex, highly structured tissues such as the brain and muscles would be the epitome of low entropy tissues and so would need a lot of Gibb’s Free Energy to maintain.  Conversely undifferentiated, amorphous tissues like adipose have relatively high entropy.


Miitochondria hang on to their capability to generate free energy via their membrane potentials to the very last, right up until the point they depolarise and trigger cell death. We also know that their capacity, or maximum throughput declines as they age.  So, at a certain point they will find that the organism’s  low-entropy ‘overhead’ will become unsupportable with regard to the free energy they can supply. What to do?


My guess is that the best response would be to reduce the commitment to low-entropy tissues first. That is, muscle and brain. Indeed the most obvious features of senescence is loss of brain volume and muscle mass ( but not necessarily fat for reasons mentioned above) . The ‘new’ old creature now is less ‘expensive’ to run and so should live longer at a cost of reduced mobility, power and cognitive ability.


If this were so, it  would explain the claimed increases in life-span under high-nutrient, very low calorie diets (CRON calorie restricted optimum nutrition 1500 Cals/day) which is well documented but baffling to explain.


I think the body is tricked into ‘thinking’ not that it is starving but that the energy production from mitochondria is failing and it is entering senescence. If so the notional senescent adaptive response would kick in. To make this happen organism-wide so to speak rather than a cellular phenomenon coordination is required. The starvation response pathway for example is well documented and is at least in part hormonally (leptin) regulated.


My guess is that  aging mitochondria produce less pregnenolone from cholesterol (the precursor to all steroid hormones ) which reduces the amount of oestrogen in circulation which in turn reduces its protective effect on mitochondria which then in its turn stimulates cell death (apoptosis) in vulnerable cells, particularly in the brain and muscles. CRON diets which are touted as life span extension diets are particularly low in cholesterol as well as in calories which could only  increase this signalling pathway. A CRON diet could simulate senescence in some respects and trigger the entropy-saving apoptotic response and prolong life span?


Inter-alia.


At the start of this post I alluded to other candidates for adaptive responses in old age.
Intriguingly the aging gut increases its absorption of dietary cholesterol very markedly as it ages, peaking in senescence. This looks to me more like an adaptive last ditch attempt to boost steroid synthesis rather than merely to give the oldies atherosclerosis! Recently it has been widely reported that high blood pressure in older people provides protection against dementia presumably by increasing blood and nutrient supply to the brain.


Wouldn’t it be strange if our medical efforts to reduce blood cholesterol and blood pressure reduced our chances of dying in return for producing fatty, prematurely aged, cognitively deficient creatures. Surely not?



Senescence is an adaptive response to aging, failing mitochondria. If the use of the word ‘adaptive’ seems counter intuitive carrying as it does positive connotations,  then related senescent adaptive responses include rising blood pressure and increased absorption of dietary cholesterol will seem heretical.


When I started my PhD on the aging of mitochondria now forty years ago we made a clear distinction between ‘aging’ and the onset of senescence. We had a colony of Wistar rats, a young adult rat was three months old, a middle-aged rat around 12-15-20  months and an old rat was 29-33 months. The latter were chosen when they showed signs of senescence, which as a rule of thumb was when they started to lose condition, moved around less, interacted less and lost muscle mass.  this state easily in humans the difference being that this onset would occur at over 70 years of age and not at 29 months! The maximum life span of our rats was 33-36 months whereas humans can live to more than 110 years.


Our assumption was that aging is a continuous process which starts (for us) in our twenties and is characterised by a steady deterioration which results in an elevated risk of malfunctions such as heart attacks, strokes and cancers but if these are avoided then barring accidents we make it to the senescent stage. With regard to mitochondria, we never found any mitochondrial changes pre-senescence and concluded that if they were there, they were too small for us to find. To my knowledge this is still the case today.
Organismal senescence I will argue is an adaptive response to the aging of mitochondria, albeit an apparently pointless adaptation as it does little more than prolong life-span. Below is why I think senescence is adaptive and the explanation involve sentropy.


From a simple thermodynamic point of view ‘we’ have a low entropy. For those unfamiliar with the concept of entropy it may suffice to say that for a system ( that is ‘us’ ) composed of countless particles arranged with such delicate, intricate, ordered complexity, comprises a highly improbable state of affairs maintained only by the continuous supply of energy from our surroundings. This supply comes from food and the ‘energy’ is liberated by mitochondria.  The energy that is used to do the work of building and maintaining our bodies is called Gibbs Free Energy (G) which is a compilation of simple heat energy (H) and entropy (S). It is formalised in the equation below:


G =H -TS system


Complex, highly structured tissues such as the brain and muscles would be the epitome of low entropy tissues and so would need a lot of Gibb’s Free Energy to maintain.  Conversely undifferentiated, amorphous tissues like adipose have relatively high entropy.


Miitochondria hang on to their capability to generate free energy via their membrane potentials to the very last, right up until the point they depolarise and trigger cell death. We also know that their capacity, or maximum throughput declines as they age.  So, at a certain point they will find that the organism’s  low-entropy ‘overhead’ will become unsupportable with regard to the free energy they can supply. What to do?


My guess is that the best response would be to reduce the commitment to low-entropy tissues first. That is, muscle and brain. Indeed the most obvious features of senescence is loss of brain volume and muscle mass ( but not necessarily fat for reasons mentioned above) . The ‘new’ old creature now is less ‘expensive’ to run and so should live longer at a cost of reduced mobility, power and cognitive ability.


If this were so, it  would explain the claimed increases in life-span under high-nutrient, very low calorie diets (CRON calorie restricted optimum nutrition 1500 Cals/day) which is well documented but baffling to explain.


I think the body is tricked into ‘thinking’ not that it is starving but that the energy production from mitochondria is failing and it is entering senescence. If so the notional senescent adaptive response would kick in. To make this happen organism-wide so to speak rather than a cellular phenomenon coordination is required. The starvation response pathway for example is well documented and is at least in part hormonally (leptin) regulated.


My guess is that  aging mitochondria produce less pregnenolone from cholesterol (the precursor to all steroid hormones ) which reduces the amount of oestrogen in circulation which in turn reduces its protective effect on mitochondria which then in its turn stimulates cell death (apoptosis) in vulnerable cells, particularly in the brain and muscles. CRON diets which are touted as life span extension diets are particularly low in cholesterol as well as in calories which could only  increase this signalling pathway. A CRON diet could simulate senescence in some respects and trigger the entropy-saving apoptotic response and prolong life span?


Inter-alia.


At the start of this post I alluded to other candidates for adaptive responses in old age.
Intriguingly the aging gut increases its absorption of dietary cholesterol very markedly as it ages, peaking in senescence. This looks to me more like an adaptive last ditch attempt to boost steroid synthesis rather than merely to give the oldies atherosclerosis! Recently it has been widely reported that high blood pressure in older people provides protection against dementia presumably by increasing blood and nutrient supply to the brain.


Wouldn’t it be strange if our medical efforts to reduce blood cholesterol and blood pressure reduced our chances of dying in return for producing fatty, prematurely aged, cognitively deficient creatures. Surely not?







l







Senescence is an adaptive response to aging, failing mitochondria. If the use of the word ‘adaptive’ seems counter intuitive carrying as it does positive connotations,  then related senescent adaptive responses include rising blood pressure and increased absorption of dietary cholesterol will seem heretical.


When I started my PhD on the aging of mitochondria now forty years ago we made a clear distinction between ‘aging’ and the onset of senescence. We had a colony of Wistar rats, a young adult rat was three months old, a middle-aged rat around 12-15-20  months and an old rat was 29-33 months. The latter were chosen when they showed signs of senescence, which as a rule of thumb was when they started to lose condition, moved around less, interacted less and lost muscle mass.  this state easily in humans the difference being that this onset would occur at over 70 years of age and not at 29 months! The maximum life span of our rats was 33-36 months whereas humans can live to more than 110 years.


Our assumption was that aging is a continuous process which starts (for us) in our twenties and is characterised by a steady deterioration which results in an elevated risk of malfunctions such as heart attacks, strokes and cancers but if these are avoided then barring accidents we make it to the senescent stage. With regard to mitochondria, we never found any mitochondrial changes pre-senescence and concluded that if they were there, they were too small for us to find. To my knowledge this is still the case today.
Organismal senescence I will argue is an adaptive response to the aging of mitochondria, albeit an apparently pointless adaptation as it does little more than prolong life-span. Below is why I think senescence is adaptive and the explanation involve sentropy.


From a simple thermodynamic point of view ‘we’ have a low entropy. For those unfamiliar with the concept of entropy it may suffice to say that for a system ( that is ‘us’ ) composed of countless particles arranged with such delicate, intricate, ordered complexity, comprises a highly improbable state of affairs maintained only by the continuous supply of energy from our surroundings. This supply comes from food and the ‘energy’ is liberated by mitochondria.  The energy that is used to do the work of building and maintaining our bodies is called Gibbs Free Energy (G) which is a compilation of simple heat energy (H) and entropy (S). It is formalised in the equation below:


G =H -TS system


Complex, highly structured tissues such as the brain and muscles would be the epitome of low entropy tissues and so would need a lot of Gibb’s Free Energy to maintain.  Conversely undifferentiated, amorphous tissues like adipose have relatively high entropy.


Miitochondria hang on to their capability to generate free energy via their membrane potentials to the very last, right up until the point they depolarise and trigger cell death. We also know that their capacity, or maximum throughput declines as they age.  So, at a certain point they will find that the organism’s  low-entropy ‘overhead’ will become unsupportable with regard to the free energy they can supply. What to do?


My guess is that the best response would be to reduce the commitment to low-entropy tissues first. That is, muscle and brain. Indeed the most obvious features of senescence is loss of brain volume and muscle mass ( but not necessarily fat for reasons mentioned above) . The ‘new’ old creature now is less ‘expensive’ to run and so should live longer at a cost of reduced mobility, power and cognitive ability.


If this were so, it  would explain the claimed increases in life-span under high-nutrient, very low calorie diets (CRON calorie restricted optimum nutrition 1500 Cals/day) which is well documented but baffling to explain.


I think the body is tricked into ‘thinking’ not that it is starving but that the energy production from mitochondria is failing and it is entering senescence. If so the notional senescent adaptive response would kick in. To make this happen organism-wide so to speak rather than a cellular phenomenon coordination is required. The starvation response pathway for example is well documented and is at least in part hormonally (leptin) regulated.


My guess is that  aging mitochondria produce less pregnenolone from cholesterol (the precursor to all steroid hormones ) which reduces the amount of oestrogen in circulation which in turn reduces its protective effect on mitochondria which then in its turn stimulates cell death (apoptosis) in vulnerable cells, particularly in the brain and muscles. CRON diets which are touted as life span extension diets are particularly low in cholesterol as well as in calories which could only  increase this signalling pathway. A CRON diet could simulate senescence in some respects and trigger the entropy-saving apoptotic response and prolong life span?


Inter-alia.


At the start of this post I alluded to other candidates for adaptive responses in old age.
Intriguingly the aging gut increases its absorption of dietary cholesterol very markedly as it ages, peaking in senescence. This looks to me more like an adaptive last ditch attempt to boost steroid synthesis rather than merely to give the oldies atherosclerosis! Recently it has been widely reported that high blood pressure in older people provides protection against dementia presumably by increasing blood and nutrient supply to the brain.


Wouldn’t it be strange if our medical efforts to reduce blood cholesterol and blood pressure reduced our chances of dying in return for producing fatty, prematurely aged, cognitively deficient creatures. Surely not?







l





Senescence is an adaptive response to aging, failing mitochondria. If the use of the word ‘adaptive’ seems counter intuitive carrying as it does positive connotations,  then related senescent adaptive responses include rising blood pressure and increased absorption of dietary cholesterol will seem heretical.


When I started my PhD on the aging of mitochondria now forty years ago we made a clear distinction between ‘aging’ and the onset of senescence. We had a colony of Wistar rats, a young adult rat was three months old, a middle-aged rat around 12-15-20  months and an old rat was 29-33 months. The latter were chosen when they showed signs of senescence, which as a rule of thumb was when they started to lose condition, moved around less, interacted less and lost muscle mass.  this state easily in humans the difference being that this onset would occur at over 70 years of age and not at 29 months! The maximum life span of our rats was 33-36 months whereas humans can live to more than 110 years.


Our assumption was that aging is a continuous process which starts (for us) in our twenties and is characterised by a steady deterioration which results in an elevated risk of malfunctions such as heart attacks, strokes and cancers but if these are avoided then barring accidents we make it to the senescent stage. With regard to mitochondria, we never found any mitochondrial changes pre-senescence and concluded that if they were there, they were too small for us to find. To my knowledge this is still the case today.
Organismal senescence I will argue is an adaptive response to the aging of mitochondria, albeit an apparently pointless adaptation as it does little more than prolong life-span. Below is why I think senescence is adaptive and the explanation involve sentropy.


From a simple thermodynamic point of view ‘we’ have a low entropy. For those unfamiliar with the concept of entropy it may suffice to say that for a system ( that is ‘us’ ) composed of countless particles arranged with such delicate, intricate, ordered complexity, comprises a highly improbable state of affairs maintained only by the continuous supply of energy from our surroundings. This supply comes from food and the ‘energy’ is liberated by mitochondria.  The energy that is used to do the work of building and maintaining our bodies is called Gibbs Free Energy (G) which is a compilation of simple heat energy (H) and entropy (S). It is formalised in the equation below:


G =H -TS system


Complex, highly structured tissues such as the brain and muscles would be the epitome of low entropy tissues and so would need a lot of Gibb’s Free Energy to maintain.  Conversely undifferentiated, amorphous tissues like adipose have relatively high entropy.


Miitochondria hang on to their capability to generate free energy via their membrane potentials to the very last, right up until the point they depolarise and trigger cell death. We also know that their capacity, or maximum throughput declines as they age.  So, at a certain point they will find that the organism’s  low-entropy ‘overhead’ will become unsupportable with regard to the free energy they can supply. What to do?


My guess is that the best response would be to reduce the commitment to low-entropy tissues first. That is, muscle and brain. Indeed the most obvious features of senescence is loss of brain volume and muscle mass ( but not necessarily fat for reasons mentioned above) . The ‘new’ old creature now is less ‘expensive’ to run and so should live longer at a cost of reduced mobility, power and cognitive ability.


If this were so, it  would explain the claimed increases in life-span under high-nutrient, very low calorie diets (CRON calorie restricted optimum nutrition 1500 Cals/day) which is well documented but baffling to explain.


I think the body is tricked into ‘thinking’ not that it is starving but that the energy production from mitochondria is failing and it is entering senescence. If so the notional senescent adaptive response would kick in. To make this happen organism-wide so to speak rather than a cellular phenomenon coordination is required. The starvation response pathway for example is well documented and is at least in part hormonally (leptin) regulated.


My guess is that  aging mitochondria produce less pregnenolone from cholesterol (the precursor to all steroid hormones ) which reduces the amount of oestrogen in circulation which in turn reduces its protective effect on mitochondria which then in its turn stimulates cell death (apoptosis) in vulnerable cells, particularly in the brain and muscles. CRON diets which are touted as life span extension diets are particularly low in cholesterol as well as in calories which could only  increase this signalling pathway. A CRON diet could simulate senescence in some respects and trigger the entropy-saving apoptotic response and prolong life span?


Inter-alia.


At the start of this post I alluded to other candidates for adaptive responses in old age.
Intriguingly the aging gut increases its absorption of dietary cholesterol very markedly as it ages, peaking in senescence. This looks to me more like an adaptive last ditch attempt to boost steroid synthesis rather than merely to give the oldies atherosclerosis! Recently it has been widely reported that high blood pressure in older people provides protection against dementia presumably by increasing blood and nutrient supply to the brain.


Wouldn’t it be strange if our medical efforts to reduce blood cholesterol and blood pressure reduced our chances of dying in return for producing fatty, prematurely aged, cognitively deficient creatures. Surely not?







l























l














Sunday, January 15, 2017

Eat Healthy, eat cholesterol


In previous posts I have talked about the role of cholesterol in the integrity of aging mitochondria’s membrane potential. All very technical, but in tune with the rise in the notion of supporting mitochondrial health to ward off aging. Below is an oblique take on the same subject from a dietary level.

This is a post about diet, specifically about a diet that is very much in favour as the way to combine good health with longevity. It is of course the ‘Mediterranean Diet’. I have genuinely never come across a single article with anything to say against it. If there were ever a consensus in the fashion-prone world of what to eat, this diet must be it.

So, what is a ‘Mediterranean Diet’? Obviously the is shorthand for ‘things folk eat around the countries bordering the Mediterranean sea’ and that we all agree are healthy. It’s credibility relies on a combination of the manifest healthy longevity of those native populations who adhere to it and it’s agreement with popular buzz-health-words such as omega-3, anti-oxidants and low-carb.

We may agree its components (other than liberal amounts of sunshine) are virgin olive oil,  fresh small(ish) fish, seafood, eggs, fresh fruit, vegetables, wild-fungi,  nuts and parsimonious portions of red meat and high GI carbs such as sugars and potato.  Any objectors to my list at this point? I think I am pretty safe so far and ‘afficionado dietistas’ will be able to dissect these foods minutely into a Sunday Supplement ‘who’s who’ of good things.

What follows is a list of food in order of how much cholesterol they contain or how cholesterogenic they are. This last term needs a line of explanation. Cholesterogenic refers to the promotion of formation of cholesterol by the body from building block molecules that are in chemical terms almost cholesterol. The two major ingredients that fall into this category are squalene and ergosterol.

Here then are the top ten:


Here are the top cholesterogenic ingredients:


I think the data makes it’s own point but for clarity a diet that we have called the Mediterranean Diet is probably one of the highest in cholesterol that you could come up with. Only a japanese-style diet could trump it...and yes they are famously long-lived and healthy too. Note also that the red-meats don’t even get into the top ten!

Finally if I were to make the claim that high-cholesterol diets are healthiest diets that you can eat, and anyone cared what I said or what evidence I have,  the post-truth world would scream and shout in protest. After all, this is a mere correlatio,n not a cause and effect mechanism being proposed. On the other hand the correlation of blood cholesterol levels and arterial disease has been treated with near reverence.

Tuesday, January 10, 2017

Cells that don't need mitochondria



Mitochondria age within a post mitotic (non-dividing) cell as a result of a cessation in their own renewal by replication and, when aged, are more likely to trigger apoptosis (cell death). Some mitochondria  ( for example in birds and bats) seem better ‘made’ than others and as a result the animals age more slowly than do other of similar size. Mitochondria the evidence shouts over and over are the arbitrators of apoptosis. The question is not how is this is well known, but why?

Multicellular creatures possess mitochondria for the simple reason that they simply could not produce the amounts of energy needed for movement,  nervous conduction and digestion without them.  To  liberate the energy to do this kind of work mitochondria combine oxygen with simple carbon compounds in a process called oxidative phosphorylation … and can do so at an impressive rate, enough to power huge brains and athletic bodies.  But the price to pay is mortality. Maybe this is a little ironic given that mitochondria and their hosts are descendants of immortal creatures. Truly a Faustian pact.

This article however is about cells that don’t really need such a high-rolling lifestyle, cells that live near anoxic lives within the multicellular furnaces of the mammal. How do they get on with their suicidal mitochondrial guests when really they don’t need them?

The answer to this question is both intriguing and consistent with what we know about mitochondria. It’s accepted that the multicellular life requires mitochondria to provide the energy to maintain the organism, but it is not the case that mitochondrial energy is needed to power cell division.  This means that in anoxic or near anoxic environments, within the organism, growth and reproduction carries on  without the need for the mitochondria’s power pack. Below are some interesting examples.

Chondrocytes are the cells that secrete the substrate for the cartilage matrix and, as is well known, the aging mammal suffers badly from deteriorating joint cartilage. Research shows that the chondrocytes are in quite anoxic environments and happily go about their work using only the energy from glycolysis, however the few mitochondria that they possess still apparently want their say over cell death. The aging and deteriorating chondrocyte has the signature profile of leaking and depolarising mitochondria known to be responsible for cell death in other tissues 1. All the pain and none of the gain for the cohabiting mitochondria is the phrase that comes to mind.

Next up are the white adipocytes better know as fat cells. These have very few mitochondria and in the centre of fat masses we see anoxic conditions and once again a reliance on glycolysis. Aging and senescent adipocytes are very much in the frame for the onset of insulin resistance with age and the onset of Type 2 diabetes. This is associated with increased oxidative stress2      ( a mitochondrial signature) but I can find no direct work on the role that mitochondria play ... but I have my suspicions!

Cancer cells are notoriously biased towards the anoxic life as made famous by the Warburg effect3, and indeed it was thought that their predisposition towards anaerobic glycolysis and away from  mitochondrial oxidative phosphorylation means that cancer cells do not need mitochondria.

It turns out that the picture is quite complicated. Indeed many cancer cells do ‘eat up’ excess mitochondria (mitophagy) and also inhibit their production of ATP but they also seem to ‘benefit’ from increased ROS (reactive oxygen species) production in that it has a mutagenic and hence potentially  transforming ( cancer producing) plus side … but also they seem to be able to do this whilst inhibiting the mitochondria’s apotoptic abilities4. This is truly intriguing and can be put this way: Cells which atavistically ‘revert’ to an immortal  anaerobic state actually leverage the ROS mutagenic potential of their malfunctioning mitochondrial symbionts while at the same time disabling their ability to destroy their host … amazing.

Finally we have eukaryote cells which have gone the whole hog and ditched mitochondria ... genes and all. Monocercomonoidesis5 a unicellular flagellate living in the near-anoxic gut biome which is the environment bounded by the gut in which our gut flora live.  Monocercomonoidesis is recently famous because it has been shown to have: no mitochondria; no genes for mitochondria in its nucleus and has borrowed a bacterium’s iron-sulfur complex to do a bit of essential chemistry that the mitochondria were doing.

It uses its ancestral glycolytic pathway and is doing just fine even though it appears to re-write the definition of an eukaryotic cell which is supposed to have mitochondria even if few and virtually vestigial.  Basically, it has ditched a couple of billion years of evolution. But why?

This whirlwind tour of the near anoxic world of the eukaryote cell has posed a few conundrums.
Firstly, one paradigm remains intact and that is that mitochondria-deplete cells are in effect  either unicellular ( eg Monocercomonoides or chondrocytes) or are in poorly vasculated undifferentiated tissues masses ( tumours or adipose tissue).

We also can see that mitochondria are ‘expensive’ and are not conserved or are poorly conserved in cells which don’t need them. It also looks like they are in whatever context regarded as lethal. The question is begged as to whether ditching mitochondria in some settings is a good idea or whether they perform a vital housekeeping function destroying post-mitotic cells that are past their ‘cell-by’ date, pun intended.

Whatever, the picture I am seeing more clearly than ever is one of an intracellular ecosystem, a miniature eurozone of complex interdependencies. Monocercomonoides has lost one of its prominent members, though we don’t know quite when or whether it has done it alone. Evolution can go ‘backwards’ as well as ‘forwards’