Potential Interactions of Urate and Inorganic Phosphate with Xenobiotic Substrates and Physiological Substrates of Organic Anion Transporters

One thing I was going to mention is that an excessive intake of inorganic phosphate, alone or in combination with oral purine nucleotides, could conceivably interact with prescription or nonprescription drugs that are substrates of organic anion transporters (OAT's) or multidrug resistance (MDR) protein transporters. The main types of drug-drug interactions that are given attention in the literature are the interactions that involve the noncompetitive inhibition, induction, or competitive inhibition of cytochrome P450 enzymes. But another type of interaction that would be more difficult to predict or even measure could be the competition of two substrates for export, across the canalicular, or apical, membranes of biliary epithelial cells, into the bile. Urate and phosphate can compete for export into the blood or bile by OAT's on the plasma membranes of different cell types in the liver, and bilirubin (http://scholar.google.com/scholar?q=bilirubin+%22organic+anion%22&hl=en), bile acids (http://scholar.google.com/scholar?hl=en&q=%22bile+acids%22+%22organic+anion%22), and other compounds are also substrates of various OAT's. A lot of different drugs are also substrates of OAT's (http://scholar.google.com/scholar?q=drugs+transport+%22organic+anion%22&hl=en) and might compete with urate or phosphate or xanthine, for example, conceivably, for export into the bile. It's unlikely that these interactions would be significant, in my opinion, except at high or excessive dosages of uricogenic purines or inorganic phosphate or in people who have liver or kidney disease. As I've mentioned in past postings, however, some neuraminidase inhibitors and other drugs or metabolites of drugs that are excreted unchanged or otherwise eliminated primarily by renal excretion might interact more significantly with high dosages of oral purines or with excessive amounts of inorganic phosphate. The effect that could conceivably be problematic would be a slowing, in response to an increase in intracellular urate or phosphate, etc., of the rate of biliary or renal excretion of a given drug. That's one reason it's always necessary to discuss these things with one's doctor.

Nonetheless, urate has been used to treat various forms of liver disease in animal models [one example: Garcia-Ruiz et al., 2006: (http://www.ncbi.nlm.nih.gov/pubmed/16941682)], and researchers have shown that urate can protect against mitochondrial dysfunction induced by a wide variety of treatments that produce mitochondrial dysfunction by increasing peroxynitrite formation (http://scholar.google.com/scholar?hl=en&q=mitochondrial+peroxynitrite+urate+OR+uric). A lot of factors and disease states can increase peroxynitrite formation, and the "antioxidant" or "nitrosative-degradation-by-proxy," more accurately, effects of urate, along with its apparent capacity to decrease or directly inhibit PARP-1 activity, make it more useful than many other compounds or antioxidants, in my opinion. As I've discussed in past postings, it may well be advantageous for an antioxidant, such as urate, to not be regenerated. Nonetheless, urate can, for example, regenerate melatonin and guanosine radical species by apparently-nonenzymatic mechanisms (http://scholar.google.com/scholar?hl=en&q=melatonin+regeneration+urate). And, as far as the rest of this posting is concerned, there's evidence that hypophosphatemia and intracellular phosphate depletion in the liver may contribute to liver damage in some cases and disease states (see past postings). One of the most important considerations in the context of phosphate homeostasis is to be aware that, in my opinion, the "phosphate" contained in inositol hexakisphosphate and other phytate compounds, in cereal grains and "plant proteins," etc., is unlikely to provide much, if any, utilizable phosphate in humans [see here: (http://hardcorephysiologyfun.blogspot.com/2009/08/phytates-as-potentially-poor-sources-of.html); (http://hardcorephysiologyfun.blogspot.com/2009/07/phytates-inositol-hexaphosphate-and.html)]. As far as my own calculation of my "dietary phosphate" intake went, I didn't even bother to include a contribution of cereal-grain phosphate. I put a big "NOTH-THING" by the spot on the page for the mg phosphate derived from phytate-containing foods. But I can't make that determination or calculation for anyone except myself. If I had been in the business of obtaining "hocus-pocus-microbial-phytase-derived-phantom-phosphate" phosphate from foods, maybe I'd have listed an actual number. But anyway, as with any compound, bizzarely-high dosages could cause problems. In response to massive dosages of either uricogenic purines or inorganic phosphate, those problems could take the form of interactions with other OAT substrates.

Potential for Competition Among Phosphate, Uric Acid (Urate), and Antivirals Used to Treat Influenza for Transport by Organic Anion Transporters

The authors of this article [Yabuuchi et al., 1998: (http://jpet.aspetjournals.org/cgi/reprint/286/3/1391)(http://www.ncbi.nlm.nih.gov/pubmed/9732402?dopt=Abstract)] describe the capacity of the type I Na(+)/Pi cotransporter (NPT1), a sodium and inorganic phosphate (Pi) transporter, to transport either organic anions, including probenecid, or inorganic phosphate (Pi) out of the liver and into the blood. Yabuuchi et al. (1998) noted that probenecid can compete with Pi for transport by NPT1, and this could conceivably mean that a higher intake of Pi might inhibit the efflux of uric acid (urate, UA), an organic anion whose reabsorption by proximal tubule epithelial cells can be inhibited by probenecid (http://scholar.google.com/scholar?hl=en&q=urate+probenecid), from the liver or otherwise influence the efflux or uptake of urate or xanthine by cells in the liver or kidneys, etc. (http://scholar.google.com/scholar?hl=en&q=%22inorganic+phosphate%22+anion+transporter). It's also conceivable that increases in extracellular or, in a more likely event, intracellular Pi could slow the elimination of antiviral drugs used to treat influenza. For example, Oo et al. (2002) [Oo et al., 2002: (http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=127254&blobtype=pdf)(http://www.ncbi.nlm.nih.gov/pubmed/12019123)] noted that the active metabolites of some neuraminidase inhibitors are mostly excreted unchanged, such as through their uptake by the proximal tubule cells, from the peritubular capillaries, and efflux across the luminal (apical) membranes of proximal tubule epithelial cells into the tubular fluid. Karie et al. (2006) [Karie et al., 2006: (http://ndt.oxfordjournals.org/cgi/reprint/21/12/3606.pdf)(http://www.ncbi.nlm.nih.gov/pubmed/16799172)] noted that some neuraminidase inhibitors do not serve as substrates for cytochrome P450 oxidoreductases in the liver and do not inhibit those enzymes either, and that's a major reason that their active metabolites are mostly excreted unchanged by renal tubular excretion. Probenecid competes with some of these These interactions would probably not be very likely and would be most likely to occur, if at all, in people whose kidney function has already been diminished, as a result of age or other factors. This is a hastily-chosen article that describes the capacity of probenecid to inhibit the transport and, hence, renal excretion of some neuraminidase inhibitors or their active metabolites [(http://www.cdc.gov/Mmwr/preview/mmwrhtml/rr4814a1.htm); (http://scholar.google.com/scholar?q=probenecid+neuraminidase+inhibitor&hl=en)], and that basically means that oral purines or phosphate supplementation could conceivably slow the elimination of some neuraminidase inhibitors, and that wouldn't necessarily be desirable. It might sound good, and some people have proposed the use of probenecid to allow for the use of neuraminidase inhibitors at lower dosages (thereby allowing more people to be treated with antivirals, in the event of a "1970's-style shortage" of antivirals). But that could be a dangerous approach, given that the movement and constant efflux of some neuraminidase inhibitors is necessary to prevent the potentially problematic effects of their accumulation intracellularly, in cells in the liver or kidneys.

Thus, if one were taking an antiviral to treat an influenza infection and also taking some oral purine compound or source of inorganic phosphate (Pi), one might need to reduce the dosages of those or, as discussed by Karie et al. (2006), reduce the dosages of the antivirals. It would seem that reducing the dosage of the antiviral would not be the better approach, in theory, but one would obviously want to discuss this with one's doctor. Some of the major old M2 protein inhibitors, used as antivirals in the treatment of influenza, are derivatives of 1-aminoadamantane and are therefore also excreted unchanged. Aminoadamantane derivatives are apparently transported by organic cation transporters and would seem to not compete with UA or phosphate, but probenecid is a weak base and can sometimes inhibit the transport of substrates of organic cation transporters (http://scholar.google.com/scholar?hl=en&q=probenecid+aminoadamantane). There are strange ways in which substrates of organic cation transporters can influence the transport of other substrates (drugs or physiological compounds) of organic anion transporters [Khamdang et al., 2002: (http://jpet.aspetjournals.org/cgi/content/full/303/2/534)(http://www.ncbi.nlm.nih.gov/pubmed/12388633?dopt=Abstract)], maybe because they, like probenecid, are weak bases and could either be protonated or deprotonated or because they contain more than one ionizable group. There can be pH extremes and variations in the tubular fluid, for example, and there could be indirect interactions. An increase in the reabsorption of UA could, for example, be pH dependent and thereby produce an indirect, pH-sensitive reduction in the excretion of a drug that UA, by its binding to an efflux transporter intracellularly, in proximal tubule cells, and relative failure to serve as a substrate for transport by that transporter, competes with for transport, etc.

Another implication is that increases in the intracellular Pi concentration could reduce the loss of purine nucleotides both by inhibiting adenosine deaminase (and by activating adenosine kinase, arguably) and by reducing the efflux of cAMP or cGMP or other purine substrates of some organic anion transporters or multidrug resistance proteins that transport purines out of cells. This might mean that phosphate could, apart from its role in promoting normal purine salvage, serve as a dose-reducing agent for oral purines, such as ATP disodium, even in the absence of an influenza infection, obviously. But that's more theoretical, and these are just my opinions. Obviously, other medications, including but not limited to some antibiotics, are transported by organic anion transporters, too, and that's another reason one should discuss this type of thing with one's doctor.

Sugar Alcohols in Some ATP Preparations; Solvent Drag and Intestinal Absorption of Physiological Compounds by Passive Diffusion

I was just going to discuss the issue of the addition of mannitol vs. xylitol in some oral ATP disodium preparations. The traditional "dogma" is that oral mannitol is not absorbed to any significant extent, but Pappenheimer (1990) [Pappenheimer, 1990: (http://www.ncbi.nlm.nih.gov/pubmed/2116731?dopt=Abstract)] found that this was not the case, at least in animals. This could conceivably be important, given that the potentially-slow conversion of mannitol into fructose, by sorbitol dehydrogenase or other enzymes (Pappenheimer, 1990), in the liver could mean that its osmotic or "osmoregulatory" effects could become an issue for some susceptible individuals. Vernacchio et al. (2007) [Vernacchio et al., 2007: (http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1780176&blobtype=pdf)(http://www.ncbi.nlm.nih.gov/pubmed/17097152)] noted that unabsorbed xylitol could also potentially cause osmotic effects but that there's rapid adaptation to those effects. In any case, it's unlikely to be an issue, but my sense is that mannitol could potentially be more problematic than xylitol, and either unabsorbed or absorbed mannitol could conceivably be slightly problematic if taken at high dosages. Intravenous mannitol used to be used to treat cerebral edema and may still be used in some cases, but it tends to lead to derangements in osmolytes in the long term. This is not at all likely to happen in this context, but I'm just saying that much milder osmoregulatory changes might require a couple of days of adaptation, conceivably, with the ingestion of a sugar alcohol. Or there may be some maximal amount, particularly of mannitol, that any given individual can ingest without experiencing nausea or something like that. The amounts would probably be too small, though, to cause issues, but it's something to be aware of and to not confuse with the effects of ATP. The research suggests that, as noted by Vernacchio et al. (2007), there would be adaptation to any such effects in a few days. The magnitudes of the effects of xylitol vs. mannitol would also, obviously, depend on the amounts of each sugar alcohol in a given ATP preparation, and, without referring to any specific brands, some xylitol-containing ATP preparations seem to contain xylitol in combination with other sugars and might therefore contain a lower amount of "total sugar alcohols" than some other preparations might. This could potentially cause fewer side effects if a person were using higher dosages in the short term, for example. Researchers have used up to 4,000 or 4,200 mg per day of ATP in humans (actually, there's one study in which researchers used something like 6 grams per day, but it wasn't clear to me that it was actually absorbed or that the tablets actually broke apart). I doubt it would be necessary to use dosages that high (even as high as 4,000 mg/day), but the amounts of phosphate and also mannitol or other sugar alcohols could conceivably become factors that would warrant consideration at higher dosages. I doubt the amount of phosphate would be an issue at any dosage of ATP a person would take, but people who have kidney disease could conceivably want to specifically discuss that type of thing with their doctors. Obviously, one should always discuss these things with one's doctor.

It's interesting that Pappenheimer (1990) found that even glucose was absorbed, to a significant extent, by passive, paracellular diffusion (between the cells, rather than transport into the epithelial cells) and that solvent drag effects contributed to that mode of absorption. There's a general tendency to underestimate the contribution of passive diffusion to the intestinal absorption of compounds. I've seen research that provides evidence of the passive absorption of lots of different compounds, including guanosine, other nucleosides or nucleotides, reduced folates, etc. But the authors of a lot of articles talk about "brush-border transporters" and so on, and I've seen that the assumption of "brush-border" carrier-mediated transport can lead to erroneous pharmacology-related conclusions. In some cases, for example, researchers may argue that the rate of absorption is saturable or that only X amount of a compound can be absorbed at any given time, and those types of conclusions, based on the assumption of absorption by carrier-mediated transport, can tend to be invalid.

General Pharmacological Considerations

This article [Horter and Dressman, 2001: (http://www.ncbi.nlm.nih.gov/pubmed/11259834)] is really interesting, and the authors noted, on the last two pages of text, that the gastric luminal fluid volume can be only 20-30 mL in the fasted state (meaning that the person hasn't ingested anything for 12+ hours, though 14.5 hrs may be required for the stomach to completely empty) and that the USP procedures for evaluating tablet dissolution had been based, at the time the authors wrote the article, in 2001, on nonphysiological pH values (7.5 is not a pH value that's likely to be reached in the GI tract in many people) and surfactant concentrations. The authors also discussed the fact that the rate of dissolution is frequently the most important factor determining the rate of absorption (and, hence, the bioavailability, in many cases). The issue of the luminal fluid volume can be important in determining the rate of dissolution, and the authors noted, for example, that increases in the viscosity of the intestinal luminal fluid, such as in response to food intake, can slow the rate of dissolution and, hence, the rate of absorption. In general, if one wants to maximize the bioavailability of a physiological substance, one should take it on an empty stomach. One might want to spread the total daily dosage out across the day, but it's worthwhile to keep these types of things in mind.

Maximizing bioavailability is not likely to be very important for many supplements, especially if they're in capsule form, etc. For example, I don't think there's any need to try to maximize the bioavailability of encapsulated creatine monohydrate, given that a slight increase in bioavailability is not going to be very important. But, in the case of purine (and pyrimidine) nucleotides, for example, the half-life is extremely short, and the elevation of the concentration of the nucleotide or its metabolites (i.e. other purines) in the systemic circulation is extremely brief, following oral administration. There can be drastic changes in the bioavailabilities of nucleotides, in response to small changes in the rate of dissolution and absorption. Small changes in those parameters have the potential to produce large changes in the bioavailability.

In a related vein, there are still many reports, from articles in the literature and from other sources, of problems with the dissolution of supplements provided in tablet form. It's still a significant problem in the supplement industry, in my opinion. In this context, the issue is not just bioavailability but absorption. If a tablet doesn't dissolve, the absorption and bioavailability will be zero. Consumerlab.com has shown that some tablets essentially don't dissolve at all, and they suggest this complex method for telling if a tablet is going to dissolve (http://www.consumerlab.com/results/hometest.asp). I don't think that's necessary. If a tablet is going to dissolve properly, in my opinion, it should dissolve in a small glass of water in a few minutes. When tablets truly meet dissolution standards, they dissolve in a minute or less. Several years ago, I looked at a lot of reports from Consumerlab.com. They reported that some tablets couldn't be broken with a hammer, and I remember testing some tablets (by putting them in a glass of water) and finding that some of them required 2-3 hours to dissolve. That's obviously not acceptable. I think Consumerlab.com still has some free reports, but I'm not sure. I haven't looked at the site for a few years. A lot of tablets dissolve perfectly, but it's worthwhile to just put a tablet or softgel in a glass of tap water, in my opinion (if one is planning on taking the tablet). I'm saying that that's a way to test if one "sample" tablet of that particular product, from a particular manufacturer, is going to meet some rudimentary dissolution test. If it does, for example, then there's no need to think about it again. In general, though, the dissolution of capsules tends to be more reliable, in my opinion, than the dissolution of those other dosage forms.

Interactions of Phosphate Metabolism With Energy Metabolism and Adenosine Metabolism

These are some articles showing that phosphate availability can be an important factor that determines the rates of salvage of purine nucleotides and nucleosides, the adenylate charge, and the rate of deamination of adenosine to inosine [Matsumoto et al., 1979: (http://www.jbc.org/cgi/reprint/254/18/8956.pdf)(http://www.ncbi.nlm.nih.gov/pubmed/479172); Lockett et al., 1995: (http://www.ncbi.nlm.nih.gov/pubmed/8579734); (http://scholar.google.com/scholar?hl=en&q=energy+%22inorganic+phosphate%22+salvage+purine+OR+adenylate)]. Matsumoto et al. (1979) discussed the fact that inorganic phosphate [Pi, or PO4(3-)] normally inhibits adenosine monophosphate (AMP) deaminase activity, thereby preventing the catabolism of adenosine to inosine. This catabolism, however, can serve to maintain the energy charge, paradoxically, during the inhibition of energy metabolism (Matsumoto et al., 1979). But even when ATP levels are being maintained "normally," the sequestration or loss of intracellular phosphate tends to lead to the loss of adenosine nucleotides (reference 6, cited in 1979). Maj et al. (2000) found that the adenosine-induced inhibition of adenosine kinase (AK) activity, which is a major purine salvage enzyme in the brain and other tissues, decreases as the inorganic phosphate concentration increases. AK is sometimes viewed as being "bad" in the context of cerebral ischemia, and AK inhibitors can reduce brain damage due to ischemia by maintaining adenosine availability, etc. That's another reason that the provision of phosphate in the form of ATP disodium or another purine nucleotide might be advantageous. Phosphate depletion tends to produce a loss of adenosine (and, by extension, guanosine) nucleotides, and phosphate supplementation could have a mixture of beneficial and less-than-beneficial effects, particularly in the short term, on purine metabolism in the brain, for example. Increasing AK activity (meaning the phosphorylation of adenosine) without providing more exogenous adenosine could tend to decrease adenosine availability for cerebral blood flow autoregulation [Sciotti and Van Wylen, 1993: (http://www.ncbi.nlm.nih.gov/pubmed/8436611); (http://scholar.google.com/scholar?hl=en&q=%22adenosine+kinase%22+brain). It's partly because the concentrations of adenosine, both intracellularly (and extracellularly), are normally far lower than the Km of AK for adenosine. The same argument could be made in the case of the phosphate-mediated inhibition of AMP deaminase. Under conditions of low-level ischemia, as in a person with ATP depletion or purine nucleotide depletion (because of repeated cycles of ischemia or pronounced activation of the noradrenergic stress-response system in the brain), the degradation of AMP to IMP can, paradoxically, be "good," up to a point. In any case, there can even be strange short-term effects, in my opinion, of ATP disodium that could be explained, in part, by those paradoxical aspects of adenosine metabolism. Additionally, some few days may be required for changes in A1 adenosine receptor density or sensitivity to occur, even though extracellular adenosine levels are generally kept almost constant (in part by A1 adenosine receptor activation) [Andresen et al., 1999: (http://www.ncbi.nlm.nih.gov/pubmed/10490889)]. For example, A1 adenosine receptor antagonists tend to increase extracellular adenosine (Andresen et al., 1999) and can also have mood elevating effects or the like. Although the steady-state levels of extracellular adenosine and the sensitivities of adenosine receptors will generally adapt efficiently, in my opinion, to changes in stimulus-evoked increases in extracellular adenosine (exogenous ATP would be expected to primarily or almost exclusively augment stimulus-evoked extracellular adenosine concentrations and not steady-state extracellular adenosine levels), those adaptations could, in my opinion, require a day or two to take place. In my experience, there was some kind of threshold dosage, in the short term, above which there were no transient periods of somnolence or the like. I don't even know how I'd describe that type of thing, but my point is that there's some steady state that's reached and that there could be, in my opinion, potential for complex interactions with phosphate homeostasis. And the other point was that the use of sodium phosphate could disturb adenosine metabolism in the short term (and potentially the long term), even if one could not say that the effects are exclusively "bad."

Enhancing Safety and Minimizing Idiosyncratic Effects By Titration of Physiological Compounds

The author of this letter [Giesecke, 1990: (http://www.ncbi.nlm.nih.gov/pubmed/2316743)] discusses a strategy for taking small amounts of a compound, in this case fluoxetine hydrochloride, and thereby helping to avoid idiosyncratic hypersensitivities to the physiological effects of the compound. In this case, the doctor told the person to dissolve a 20-mg capsule of fluoxetine HCl into 100 mL of water or "apple juice" and then take 5 mL (i.e. ~1 tsp.) of the solution at a time (providing a 1-mg dose). The author contacted the manufacturer and was told that the solution of fluoxetine could be refrigerated and would exhibit a "shelf life" of 14 days. One wouldn't need to get into refrigeration, and, cumbersome as this approach might seem to be, it is one approach to maximizing safety in the use of over-the-counter nutrients or the like, even under a doctor's supervision. Assuming the nutrient is water-soluble, dissolving the nutrient in 100 or 500 mL, or whatever, of water will form a solution (meaning that the nutrient will be dissolved "homogeneously" throughout the liquid). Drinking a small amount of the water will then provide a fraction of a dosage form. It tends to be the case that there are physiological adaptations to some of the idiosyncratic and initial effects that "physiological" compounds or medications can have. The rationale for using that type of approach, even under a doctor's supervision, has to do with safety, in a general sense, and individual differences in physiology.

In the case of the article by Giesecke (1990), the idiosyncratic effect was that the drug had been producing prolonged periods of insomnia that had prevented the person from adjusting to the therapeutic effects of the medication. Fluoxetine, like many drugs and even nutrients, can produce initial effects that differ from the long-term effects. Belzung et al. (2001) [Belzung et al., 2001: (http://desco.univ-tours.fr/Psychobio%20Emotions/Public/CBarticle%2047.pdf)(http://www.ncbi.nlm.nih.gov/pubmed/11485052)] found that haloperidol antagonized this type of initial, stimulant-like effect that fluoxetine has been reported to have had in some people, and the effect of fluoxetine is thought to be due to either a transient action of fluoxetine as a dopamine-receptor agonist or some indirect effect that causes dopamine release [such as interaction with postsynaptic serotonin-1A (5-HT1A) receptors on ventral-tegmental-area (VTA) neurons receiving serotonergic inputs [Amargos-Bosch et al., 2004: (http://cercor.oxfordjournals.org/cgi/reprint/14/3/281)(http://www.ncbi.nlm.nih.gov/pubmed/14754868?dopt=Abstract)]. Belzung et al. (2001) actually discuss some articles that imply that fluoxetine could transiently decrease the firing rates of dopaminergic neurons in parts of the striatum and simultaneously increase the firing rates of neurons in the prefrontal cortex (PFC) that receive dopaminergic inputs from other parts of the striatum, and that's essentially the type of thing that one would expect to see in response to stimulant-like effects (dopamine release in the PFC and the high-Km, phosphorylated form of tyrosine hydroxylase, in dopaminergic neurons that project to the PFC, are less susceptible to feedback inhibition in response to ongoing dopaminergic stimulation than dopamine release, by neurons whose cell bodies are in the VTA, in the striatum is, etc.). Glutamatergic pyramidal neurons in the PFC project back to the VTA and dorsal raphe nuclei (containing the cell bodies of serotonergic neurons) and interact with serotonergic inputs to the VTA, etc. The idea is that there could be an initial, dopaminergic effect and a subsequent, usually-rapid downregulation of the responsiveness of the receptor or receptors that are initially activated by fluoxetine and that may be unusually sensitive in some people. It's possible that there could just be hypersensitivity, in the face of a fluoxetine-induced increase in extracellular serotonin, of 5-HT1A or 5-HT2 receptors on VTA neurons or pyramidal neurons in the PFC, etc.