Jonathan Edwards
Senior Member (Voting Rights)
So maybe it is an irrelevance? It may occur but may not actually be necessary.
Review Fluid transport in the brain, 2022, Rasmussen, Mestre, Nedergaard
- Thread starter SNT Gatchaman
- Start date
So maybe it is an irrelevance? It may occur but may not actually be necessary.
jnmaciuch
Senior Member (Voting Rights)
Changes is cell volume correspond to changes in water content which correspond to changes in osmolarity
[Edit: if passage through astrocytes is the main way through which any solutes actually enter and exit the brain parenchyma then it would be pretty consequential]
jnmaciuch
Senior Member (Voting Rights)
Unless each various system accounts for some proportion of necessary waste clearance and a 15% reduction does have negative effects when it can’t be adequately compensated by other methods (or not compensated without other negative consequences)
Jonathan Edwards
Senior Member (Voting Rights)
What worries me about all this is that putting in grants on amyloid in brains has been a multi-million dollar money spinner for a decade or more. There have been lots of examples in the past where mechanisms that are claimed to be desperately important have turned out not to explani anything much and when the clinical need shifted all the research got forgotten. There was a mass of stuff on bicarbonate handling in stomach in the days when treatment for ulcers was all about acidity. But when Helicobacter came along it all got forgotten. For years people thought that oedema in nephrotic syndrome was due to low albumin so patients on ITU got albumin infusions until an intelligent Canadian team showed it just made their renal failure worse. And it is easy enough to see that the oedema cannot be due to low albumin anyway.
It may be great fun to watch tiny shifts in fluid in brains during the night and get millions in grants to get rid of brain amyloid but I think it worth pausing to ask whether this is really anything more than an artery at an elbow wiggling a bit to left and right with the pulse for no reason other than it is an irrelevant byproduct of the system dynamics.
It may be great fun to watch tiny shifts in fluid in brains during the night and get millions in grants to get rid of brain amyloid but I think it worth pausing to ask whether this is really anything more than an artery at an elbow wiggling a bit to left and right with the pulse for no reason other than it is an irrelevant byproduct of the system dynamics.
Jonathan Edwards
Senior Member (Voting Rights)
Of what, of what... sorry but that makes no sense at all to me. The time frame would be entirely wrong. The pumping in these systems is supposed to be driven by hydrostatic shits over less than. second (slow waves etc.). Cell volume changes take minutes surely. And which compartment do you suggest is affected in what way to what advantage?
Without a detailed of compartments we get nowhere.
jnmaciuch
Senior Member (Voting Rights)
Well they are already there on every other blood vessel in the body so presumably they serve some purpose or otherwise aren’t detrimental enough for vessel function to have been acted upon by negative selection
jnmaciuch
Senior Member (Voting Rights)
Again, they might not be the driving mechanism of some pumping motion but would be able to establish an osmotic gradient. The question of how does need to be answered, I agree. I’m not arguing that these systems must be relevant, just that their functioning is not inherently implausible and should not be dismissed on those grounds
Jonathan Edwards
Senior Member (Voting Rights)
I would like to see some evidence for that though.
Amyloid proteins are largely a problem if they fold a particular way and stick to things. So washing with fluid wouldn't be that much use I suspect. You may wash out the free amyloid proteins but maybe they aren't the problem.
It is just that I have seen this sort of merry-go-round so many times before. A claim of some dramatically new function that gets forgotten as soon as it is no longer a grant puller.
Jonathan Edwards
Senior Member (Voting Rights)
Yes of course they will be for blood vessels, where water flux has always been understood to be important, but that is not what we were discussing.
jnmaciuch
Senior Member (Voting Rights)
Sure, me too, I’m pretty sure that is what the goal of research in that area is aiming for. It may be overhyped but I do think it is ultimately good to fund exploratory research into an unknown facet of brain anatomy.
it’s not?
The reason to talk about aquaporins at all is because the endothelium of brain blood vessels does not contain gaps between cells as it does in other parts of the body, so the only things that can come in or out are things that can pass through selective channels on the membranes of those endothelial cells. Which is a question of water flux. And then the layer of astrocytes presumably provides additional gatekeeping and regulation of the parenchymal space since astrocytes are particularly adapted to hoarding and releasing solutes dynamically.
The argument is not that “water can’t enter or exit blood vessels in the brain at all so a separate system (glymphatics) is needed to account for water and solute movement.” It’s that the setup which evolved to enable selective permeability into and out of cerebral blood vessels also results in the formation of “glymphatic” channels, which may potentially have some additional functions and connections to other systems. Whether it’s particularly necessary for brain function is something that has only started to be assessed, but I think you may be searching for an answer to a problem that wasn’t actually put forth by this thread’s article. The main fluid dynamics would be similar as in other organs just with extra steps
Jonathan Edwards
Senior Member (Voting Rights)
I am not sure about that. If glucose and oxygen and carbon dioxide can whip through capillary and venular endothelium why not water? Oxygen and CO2 are both bigger than water and I am not aware of them having any special channels.
And I don't quite see what this extra gatekeeping is supposed to do if these astrocytes surround a channel full of CSF that is supposed to enter the parenchyma instead of the water from the capillaries. I can see an argument for having an integral enough channel wall to allow directional convective flow without it all spilling indiscriminately but you can do that with some type IV collagen and some GAG between some loosely connected glia. The aquaporins may be an important route but i cannot see that they have any great relevance to the purported function.
But if I remember rightly that was precisely the argument in the introduction.
I don't follow how having tighter control of protein flux at the capillary wall 'results' in having glymphatics.
Jonathan Edwards
Senior Member (Voting Rights)
Let me suggest a different analysis, based on the mechanisms Rod Levick and Charles Michel and I discussed in Oxford in the early 1980s in relation to the basics of the Starling equation, which turned out to be relevant to most of the other compartments.
Rod made the useful point that to maintain an oncotic gradient across vascular endothelium you have to have water flux out of the vessel. So Starling's equation can never 'balance'. If it balanced and there was no net efflux then the protein concentration outside the vessel would equilibrate with plasma if there was any permeability to protein, which there is.
In most tissues the oncotic gradient is maintained by a constant water efflux that washes away protein down lymphatics. An imbalance in Starling forces was never a problem. We all have a gross imbalance in our feet and unless we fall asleep on a long air flight they do not swell because increased tissue pressure generates reflex muscle twitching and lymph pumping, together with sponge squeezing of the matrix so that most of the time the ECM is actually squeezed dry and the lymphatics depend on a wicking process to access fluid. (Some time between age 70 and 90 the wicking process usually breaks down quite suddenly and you get puffy legs - something I had to point out to Rod because he had ignored that bit and although he trained in medicine in the year above me he never practiced and went into the the lab.)
Why do you need an oncotic gradient if Starling forces don't matter? The likely answer is that it is crucial to keep the levels of proteins like fibrinogen and prothrombin about five to ten times lower than plasma to avoid your ECM turning into clot the whole time (and likely similar arguments for IgG and complement) and for most other proteins that transport things in blood you don't want them clogging up lymphatics. If you do need more protein in ECF then you just secrete some histamine and bradykinin and open up the vessel to protein flux.
Turn now to brain. Neurons like to be in a really constant environment and it seems that to achieve this - and maybe to make really sure their synpses don't get clogged with fibrin that has to be replaced with collagen repair and metalloproteinase remodelling - the brain vasculature is designed to limit protein efflux even more. But there s a new problem. Probably for hydraulic protection purposes the brain is surrounded by a fluid at positive pressure. So water efflux under hydrostatic gradient is likely to be less than in other tissues.
I have not done the calculations - that would need Rod probably - but I think it quite likely that the brain has an oncotic problem to solve. If it wants to have a big oncotic gradient with rather minimal hydrostatic gradient across vessels then it is going to be sucked dry until its ECM protein content rises to nearer that of plasma, which defeats the point of having a low ECM protein.
There is only one way to solve this, maybe. That is to have a source of fluid of very low protein content, produced by rapid flux across a tight filter (at the choroid plexus), and to use that fluid to maintain a low protein content of brain ECF by flushing it through, without needing the gradient to be maintained by water efflux from capillaries. The rate of flux through ECM would not need to be high if the BBB was tight and very little protein leaked out, but it would need to constantly lower ECF oncotic pressure.
On this analysis I would argue that this glymphatic flux maybe has to be there for a purely oncotic reason, nothing to do with clearing waste. I can see that the routes proposed could do the job but to provide an explanation of how they work we need some sort of valve mechanism and a bit more detail on how the arterial channels differ from the venous ones. Maybe they go into this but I got bogged down with stuff that didn't add up.
Another thing i now wonder is whether the outgoing glymphatics drain into cahnnels accompanying venous sinuses that leave the cranium through a different route than that of arachnoid granulation reabsorption. Maybe they say that. If so, then one possibility is that the valve is actually the brain parenchyma itself with the venous side being an open outlet unconstrained by CSF pressure and the arterial side being a 'sponge wick' input under CSF pressure feed.
Rod made the useful point that to maintain an oncotic gradient across vascular endothelium you have to have water flux out of the vessel. So Starling's equation can never 'balance'. If it balanced and there was no net efflux then the protein concentration outside the vessel would equilibrate with plasma if there was any permeability to protein, which there is.
In most tissues the oncotic gradient is maintained by a constant water efflux that washes away protein down lymphatics. An imbalance in Starling forces was never a problem. We all have a gross imbalance in our feet and unless we fall asleep on a long air flight they do not swell because increased tissue pressure generates reflex muscle twitching and lymph pumping, together with sponge squeezing of the matrix so that most of the time the ECM is actually squeezed dry and the lymphatics depend on a wicking process to access fluid. (Some time between age 70 and 90 the wicking process usually breaks down quite suddenly and you get puffy legs - something I had to point out to Rod because he had ignored that bit and although he trained in medicine in the year above me he never practiced and went into the the lab.)
Why do you need an oncotic gradient if Starling forces don't matter? The likely answer is that it is crucial to keep the levels of proteins like fibrinogen and prothrombin about five to ten times lower than plasma to avoid your ECM turning into clot the whole time (and likely similar arguments for IgG and complement) and for most other proteins that transport things in blood you don't want them clogging up lymphatics. If you do need more protein in ECF then you just secrete some histamine and bradykinin and open up the vessel to protein flux.
Turn now to brain. Neurons like to be in a really constant environment and it seems that to achieve this - and maybe to make really sure their synpses don't get clogged with fibrin that has to be replaced with collagen repair and metalloproteinase remodelling - the brain vasculature is designed to limit protein efflux even more. But there s a new problem. Probably for hydraulic protection purposes the brain is surrounded by a fluid at positive pressure. So water efflux under hydrostatic gradient is likely to be less than in other tissues.
I have not done the calculations - that would need Rod probably - but I think it quite likely that the brain has an oncotic problem to solve. If it wants to have a big oncotic gradient with rather minimal hydrostatic gradient across vessels then it is going to be sucked dry until its ECM protein content rises to nearer that of plasma, which defeats the point of having a low ECM protein.
There is only one way to solve this, maybe. That is to have a source of fluid of very low protein content, produced by rapid flux across a tight filter (at the choroid plexus), and to use that fluid to maintain a low protein content of brain ECF by flushing it through, without needing the gradient to be maintained by water efflux from capillaries. The rate of flux through ECM would not need to be high if the BBB was tight and very little protein leaked out, but it would need to constantly lower ECF oncotic pressure.
On this analysis I would argue that this glymphatic flux maybe has to be there for a purely oncotic reason, nothing to do with clearing waste. I can see that the routes proposed could do the job but to provide an explanation of how they work we need some sort of valve mechanism and a bit more detail on how the arterial channels differ from the venous ones. Maybe they go into this but I got bogged down with stuff that didn't add up.
Another thing i now wonder is whether the outgoing glymphatics drain into cahnnels accompanying venous sinuses that leave the cranium through a different route than that of arachnoid granulation reabsorption. Maybe they say that. If so, then one possibility is that the valve is actually the brain parenchyma itself with the venous side being an open outlet unconstrained by CSF pressure and the arterial side being a 'sponge wick' input under CSF pressure feed.
Last edited:
jnmaciuch
Senior Member (Voting Rights)
I did say above that aquaporins offer selective passage to water and other small solutes. AQP1 is the main one with affinity for CO2 I’m remembering correctly.
And there are several hundred various ionic channels
Evolving an additional layer of astrocytes around blood vessels “results” in a gap space between them filled with fluid.
if so then it is contradicting itself and other findings on the field since solutes in the “glymphatic fluid” do come from the blood and some amyloids from brain do get chaperoned into the circulation
Jonathan Edwards
Senior Member (Voting Rights)
Which is probably what got me off to a bad start reading.
This review is more recent and much easier to read that the one in this thread. I think it answers your question to that last paragraph with:
along with a number of other potential efflux routes.
Under your proposal what is the direction of travel of the low protein ECM fluid? How would the fluid be forced through the brain parenchyma and not just released into the CFS? What happens to the waste products that can't cross the blood brain barrier? Can all of these waste products be removed by gila?
The other question that needs to be answered is why the brain produces so much turnover of CSF. If the entire quantity of CSF is turned over 3-5 times a day surely the fluid must be doing more than just suspending and protecting the brain. Perhaps your proposal answer that question, but it is clear to me why the glymphatic system would require such a volume of CSF production.
Jonathan Edwards
Senior Member (Voting Rights)
Yes, as I get old my attention span fails me so I may not have read enough of what is out there but I think an oncotic problem model may make sense of all these things.
Why, if CSF washes the brain via tubes around arteries then why is it produced deep in the bowels of the hemispheres in the lateral ventricles? Why isn't the choroid plexus next to the circle of Willis?
I think the key is not to assume you can ascribe a function to any particular local flux in the system without understanding the overall system function. Local flux may occur as a by-product of functionally important things going on in other ways (just as the wiggling brachial artery is).
My guess is that just like other tissues (but with a twist) the brain maintains different protein levels in different compartments by water flux using a squeezed sponge and valve system. I may be wrong but my guess is that CSF is produced deep inside hemispheres to provide a source of low 'oncolarity' fluid. That protein-poor fluid will be constantly soaked up by brain parenchyma. The flux will be diffusive rather than convective, or perhaps microconvective, through a 3D mesh as in a domestic sponge. Soaking up will also occur on the cortical surfaces from CSF around the brain but the peculiar inner ventricular system reflects the need for access to deeper tissue.
In addition to this surface uptake there may well be convective flux along arterial perivascular spaces to get more fluid into the parenchyma but as I see it this route may have no special importance and the microarchitecture may simply reflect needs of tissue integrity.
The other key event would be efflux of fluid that has transited into parenchyma via perivenous 'glymphatics' to channels exiting along with venous sinuses. The efflux will occur as long as the brain sponge is rhythmically squeezed, even if to a tiny extent. It will then absorb more low protein CSF as above and parenchymal ECF will stay at low oncolarity.
The basic reason for having this system I suspect is electrical. Brain neurons work using signals involving local potential differences at synapses provoking currents in the form of dendritic spikes and axonal action potentials. The system requires a pericellular ion sink. (Unlike printed circuits where conducting paths are printed on electrical insulating material, nerves are embedded in a conducting 'earth' phase.) Together with a need for synapses to operate separately I suspect this means that brain neurons have to sit in a very wet matrix as compared to other tissues. Indeed, if you look at brain histology the cells seem to be floating about at random rather than all packed tightly. (Especially so for grey matter.)
The brain then faces a problem that if most water flux across blood vessels occurs in the venules rather than capillaries (which I think is true for most tissues) then there is a major risk of the brain being sucked dry if the oncotic gradient is high and the intravenular pressure may actually be lower than intracranial pressure. Certainly by the time blood has reached veins the pressure will be lower than CSF because the internal jugular is normally collapsed - i.e. below atmospheric pressure.
The anatomy of the CSF and choroid plexus would solve this problem admirably by providing a source of protein poor water to suck up from within and without. The old idea of CSF flowing just from choroid through foramens to arachnoid granulation never made any sense and eveyone should have realised that. But on a squeezed sponge model it makes perfect sense. The anatomy is a bit bizarre maybe but I think that just reflects the need to solve a range of mechanical and chemical problems with a single system. It does not give much of a role to arachnoid granulations but maybe they are just there to even out fluctuations in pressure when, for instance, internal jugular pressure changes markedly with physical activities.
None of this says anything about clearing 'waste'. It may have that additional purpose but I remain sceptical. Sure, proteins will be washed out into lymphatics but that may be purely epiphenomenal and unneeded. If some protein does get across BBB or cells chuck out proteins for some reason that are not wanted then removing those will add to the oncotic story but my guess is that the system is set up to solve the oncotic problem. That entails maintaining a low protein level but whether any proteins are really 'waste' that cannot be cleaned up by microglia with receptors for post-translational modifications is maybe moot?
Last edited: