CME Slides Forum - A.Y. Bagrov

A joint Congress by ERA-EDTA and ISN

INTESTINAL SIGNALS REGULATE RENAL SOLUTE EXCRETION

Rajiv Kumar, Rochester, USA

Chair: Nathan Levin, New York, USA

Stanley Shaldon, Fontvieille, Monaco

kumar

Dr R. Kumar
Division of Nephrology and Hypertension
Department of Medicine, Mayo Clinic
Rochester, MN, USA

I’d like to thank the organising committee for the opportunity to speak here. This is a little far a field from what I normally do but let me just tell you that there’s an interest I think in this regard because feedforward signalling may be important in the regulation of the concentrations and overall homeostasis of a number of solutes including sodium and potassium that are relevant to the control of blood pressure and also certainly relevant to the control of phosphate.

Just to review some of the key concepts of how homeostasis might be maintained, solute load that might change in response to the dietary intake of a given substance can result in a change in the concentration of that solute in the serum which in turn then causes a change in the renal excretion of that solute and then this feedback feeds back on the concentration of that particular solute bringing it back to normal. So this is classic feedback control. On the other hand the point that I’m going to try and make today is that feedforward control probably plays an important role in the regulation of sodium homeostasis, in the regulation of potassium homeostasis, as well as phosphate homeostasis.

The concept here is that the concentration of solute is sensed in the gastrointestinal tract which then in turn sends signals directly to the kidney which causes a change in the way in which the kidney handles that solute almost in anticipation of an increase in the concentration of that solute as time goes on.

Let me just go over some instances that relate to sodium and then phosphorous and potassium. Now the adaptation to changes in dietary solute load are mediated by signals and factors of intestinal origin in the case of sodium, also phosphate and potassium. Now it’s commonly assumed that when a dietary sodium load is changed, the renin-angiotensin-aldosterone access is the sole mediator of the observed changes in urinary sodium excretion. While it is true that on low sodium load when one takes a low sodium intake, there is a compensatory increase in aldosterone generation and excretion, it’s not so clear that when one takes a large amount of sodium that aldosterone is inhibited and is responsible for the homeostatic change that occurs with the increase in solute load.

So what’s the evidence to show that a dietary sodium load functions independent of aldosterone? Some of the earliest observations came actually in the mid ‘70s from Lennane and Peart and others and they showed that changes in dietary sodium were followed by rapid conservation of sodium by the kidneys. They speculated that the rapidity of the response was as a result of some sort of sensing that occurred in the gastrointestinal tract.

They did a very simple experiment in which they gave individuals 100 mmol of sodium either intravenously or orally. They found that there was a greater natriuresis after the administration of an oral salt load than there was following the administration of an equimolar amount of intravenous sodium.

They thought that this was consistent with the presence of an input receptor for sodium in the GI tract.

Now Robert Carey who worked with them at the time did a series of experiments in rabbits and later in humans in which he showed that following a low sodium diet that was the expected decrease in urinary sodium and also the expected increase in aldosterone excretion.

But when individuals or rabbits in this case adapted to a low sodium diet were administered either oral or IV sodium, they found, as shown here, that the animals receiving oral sodium had a much greater natriuresis than the rabbits receiving intravenous sodium, again consistent with the fact that there was some sort of a sensor in the GI tract.

When they looked at the cumulative amount of aldosterone excretion in animals given either an oral sodium load or an IV sodium load, they found that there was about the same amount of excretion suggesting that this entire process was independent of aldosterone.

Now they did further experiments in which they gave large amounts of parental aldosterone and found that this reflex was maintained even in the presence of super physiologic concentrations of aldosterone, again suggesting that aldosterone was not a key player in this regard.

Similar experiments done in humans show exactly the same thing. So if you could look here at the panel on the right, one can see that in the grey bars the amount of natriuresis that occurs over a period of time following an oral load of sodium is much greater than that observed following an equal amount if IV sodium.

This process also occurs completely in an identical fashion in normal subjects and in subjects with adrenocortical insufficiency, again demonstrating the lack of importance of aldosterone in this regard.

So, I think from these experiments it’s clear that sodium excretion after an oral salt load is independent of aldosterone, it occurs despite high aldosterone concentrations and the issue is, what is it that actually mediates this response? Are there any changes that occur in the context of hypertension or in the context of renal failure which is really what’s relevant to this audience?

So, there are a number of heat stable toxins that have been derived from E.coli that stimulate cyclic GMP secretion in various cells and that also inhibit sodium transport. Currie et al. at Monsanto using an assay that involved the generation of cyclic GMP were able to demonstrate that in rats and in rodents the factor responsible for the generation of cyclic GMP and the inhibition of sodium transport in various transporting epithelia was a peptide known as guanylin which has structural homologies to the heat stable E.coli enterotoxin that you know is responsible for diarrhoea in a number of cases in humans. So it seems that this peptide might be the mediator of this particular process.

When they examined the distribution of this particular peptide and by assaying cyclic GMP and extracts of various tissues, they found that the highest concentration of guanylin was found in the jejunum and the next highest in the kidney. Other tissues had very low concentrations of guanylin. So their hypothesis was that guanylin was indeed the mediator of natriuresis that was induced by an oral load of sodium. Subsequently another peptide of very similar structure was isolated from the kidney and this is known as uroguanylin and in essence it has the same biological properties as guanylin and the structure here of the 3 peptides: uroguanylin, the E.coli stable toxin as well as guanylin itself are shown in comparison. The important factor here is the presence of a number of cysteine residues that form disulfide bridges and result in a specific structure of these peptides.

Now the formal proof of the importance of uroguanylin in sodium homeostasis came about following the generation of an uroguanylin knockout mouse by Lorenz and colleagues. One can see here that in these particular mice there is less uroguanylin generated in the jejunum, as well as in the ileum than there is in wild type mice.

Moreover, these mice have hypertension so the inability to excrete a sodium load or the inability to generate cyclic GMP is associated with an increase in blood pressure in the knockout mice maintained on a variety of sodium containing diets, whether the animals are kept on a normal sodium, low sodium, high sodium diet, the hypertensive relative to wild type animals.

If one examines the ability of these animals to excrete a sodium load either orally or intravenously administered, one can see that in the animals that are the knockout animals shown here in the round solid circles, the amount of sodium excreted after an oral sodium load is diminished relative to the heterozygotes or wild type animals.

However, the amount of sodium excreted following an IV sodium load is essentially unchanged. So this tells you that in point of fact there is an important role of uroguanylin in adapting to a high sodium diet.

Now, this recent review by Qian et al. in Endocrinology in 2008 summarises what we think occurs with respect to sodium and sodium sensing in the gut. The enterochromaffin cells probably contain a sensor, the exact identity of which has not been determined as yet but I think is a very important point to focus on because if one can modulate the activity of this sensor, then clearly one can modulate the amount of uroguanylin formed. This uroguanylin is secreted in the form of a pro-peptide and inhibits the absorption of sodium within the intestine itself but is also transported via the blood stream, enters the kidney tubule where it also inhibits sodium transport in the proximal tubule. Moreover, uroguanylin is also formed within the kidney and there might be a separate independent signalling pathway that allows the intestine to communicate with the kidney and cause increased uroguanylin activity within the kidney independent of that that is arising in the intestine. So there could be two pathways that are involved.

So now what evidence is there that this occurs in other solutes as well? In the interest of time, I’ll just summarise the data with respect to phosphate and tell you that when one administers a high phosphate load to humans or to animals, there is a very rapid natriuresis. We showed that this process occurs within 5 or 10 minutes of the administration of a phosphate load. It is independent of parathyroid hormone since it occurs equally as well in parathyroidectomised animals. Without belabouring the point I will tell you that it is independent of phosphate load, it is mediated by something that is present within the intestinal mucosa itself. It is a circulating factor based on parabiosis experiments and the bottom line is that following a meal there is sensing of phosphate in the intestine that causes a rapid change in the fractional excretion of phosphate that is mediated by factors that we call intestinal phosphatonins.

Now, just in closing let me tell you that the same sort of phenomenon probably occurs in the case of potassium as well. Elegant experiments done by Lee and colleagues show that if you administer potassium via the stomach or intraportally or intravenously, one gets a caliuresis that’s about equivalent.

But when you give potassium along with a meal, the amount of caliuresis occurs is much greater than when you do not give a meal, suggesting that in conjunction with a meal oral potassium is able to be sensed by the GI tract and also mediates a change in the amount of potassium being excreted by the kidney.

So the bottom line is that the mechanism of feedforward control may be relevant for solute excretion that is relevant to blood pressure I think it’s important to investigate this in various models of hypertension and chronic renal failure to determine whether these particular pathways are altered and can be modulated. Thank you very much.

Question: The question is, do you think uroguanylin acts a direct inhibitor of sodium transporter mechanisms like a classical hormone which has its specific receptor coupled to some mechanisms of intracellular transduction?

Dr. Kumar: You know I’m terribly sorry, I think I can hear you but I don’t understand your question. Can we talk after the meeting and I’ll try and answer your question at that time?

Question: Is guanylin acting as a hormone or does it act directly on sodium reabsorption in the tubule?

Dr. Kumar: The evidence is that it is a hormone, you can measure it and one can actually show changes that occur in response to the sodium load. As I think I showed in the knockout animals, if you knock it out, that doesn’t equivocally establish that it’s a hormone but it does show that it is important in terms of generation of cyclic GMP and in terms of eliciting a natriuretic response but I think an indirect answer to your question if your colleague got the gist of it, it is a hormone, it does circulate and its levels do seem to change in response to sodium load. Does that answer your question? No alright.

Question: I have a question on your methodology. You quoted numbers of sodium very precisely and deducted your reasoning from it but my big concern is what was the accuracy of the average measurements that you were quoting in terms of the standard error of the mean of the sodium measurement itself, under what conditions was the sodium measured and what method did you use in the publications you were quoting?

Dr. Kumar: Well I can’t speak precisely about what Currie and company did and the standard deviations of sodium measurements but they’re pretty straightforward and usually the standard errors are pretty good.

Dr. Kumar: No I can’t, not with the sodium but I can be precise with respect to the phosphate we which measured. In those particular instances the change in the fractional excretion of phosphate was in the range of 20-30% which is a fairly large change if you consider that your basal fractional excretion of phosphate is around 15 or 20%.

Question: I’m very concerned about sodium measurements which are --- without a more precise definition of measurement particularly because of protein interference effects and accuracy. So I think when one looks at your sodium measurements, one should be a little bit more concerned about what method and not make too many generalisations just based on something that is extremely difficult to measure accurately.

Question: Rajiv, the schematic that you put up there suggested that this stuff is filtered and then acts from the luminal side of the kidney. Which diuretic is this an analogue of or alternatively which diuretic is the analogue of this?

Dr. Kumar: You know since it acts in the proximal tubule it probably is something analogous to acetazolamide since that’s predominantly proximal tubular in action. You know whether or not it acts in other segments of the nephron for example in the thick ascending limb or in the distal convoluted tubule I don’t know, I don’t think micropuncture studies have been done using uroguanylin but that’s an important issue that needs to be resolved.

Question: I would like to ask you, do you have an assay for measurement of the hormone you’re talking about in the cross-talk between the gut and the kidney? I think it must be essential to proof -- to see a hormone goes up and down when you -- with sodium.

Dr. Kumar: I understand that there are assays for uroguanylin. There is some controversy as to their accuracy and there’s also an issue as to what the precise circulating form of uroguanylin is, is it a pro-uroguanylin? From some of the data that I’ve reviewed it would seem that it is a pro-peptide that circulates that later on gets cleaved once it gets to the kidney. So I think there’s a little bit controversy as to the accuracy of uroguanylin peptide measurements. With respect to the phosphate, the phosphatonin we don’t know the identity of the materials so we don’t have an assay for it other than a bioassay. The same thing holds true for potassium.