Phosphate (Pi) Sequestration by Fructose; Potential Effects of Changes in Pi Availability on the Mitochondrial Proton Gradient and on XDH Activity

This is one of the other articles that includes a discussion of the mechanisms by which fructose acutely increases plasma uridine and also urinary uridine excretion [Yamamoto et al., 1997: (http://www.ncbi.nlm.nih.gov/pubmed/9160822)], but Yamamoto et al. (1997) didn't show the decreases in plasma uridine, to levels below the baseline concentrations, that occur after the increases (see a recent posting). Yamamoto et al. (1997) also didn't address the mechanism by which the fructose-induced inorganic phosphate (Pi) sequestration leads to purine degradation, but a key mechanism is that the decrease in intracellular Pi disinhibits adenosine monophosphate (AMP) deaminase. AMP deaminase is normally inhibited by Pi. Yamamoto et al. (1997) cited a lot of interesting research, however. They suggested that the ethanol-induced (and, by less direct mechanisms, fructose-induced) increases in hypoxanthine and xanthine might have resulted from the elevations in the cytosolic NADH/NAD+ ratio that results from the metabolism of ethanol to acetaldehyde, given that NADH inhibits xanthine dehydrogenase activity. Fructose could also produce that effect, albeit to a lesser extent than ethanol. In addition to the ATP depletion that ultimately can occur through the disinhibition of AMP deaminase, resulting from fructose-induced Pi sequestration, Yamamoto et al. (1997) referred to the direct consumption of ATP in the fructokinase reaction that forms fructose-1-phosphate and thereby sequesters Pi [see also Phillips and Davies, 1985: (http://jp.physoc.org/content/520/3/909.full)(http://www.ncbi.nlm.nih.gov/pubmed/2992452)]. It's worth noting that fructose also depletes guanosine triphosphate (and guanosine nucleotides in general, as shown in multiple articles), partly because fructokinase activity is apparently GTP-dependent (Phillips and Davies, 1985). Fantastic. It depletes all the major nucleotide pools. Cytidine depletion would also be expected to occur (I'll bet there's some research showing that, too), given that cytidine is formed from uridine. But the point I was going to make is that changes in intracellular Pi could regulate xanthine dehydrogenase activity by buffering the intracellular pH, given that increases in the intracellular pH tend to activate phosphofructokinase and glycolytic activity overall. That increase in glycolysis would then increase the NADH/NAD+ ratio and reduce xanthine dehydrogenase activity, and that could conceivably allow for more salvage of hypoxanthine (and even xanthine, which can be salvaged to a minimal extent by a two-enzyme pathway). Yamamoto et al. (1997) cited research showing that lactate can decrease the rate of urinary uric acid excretion but apparently doesn't reduce the excretion of hypoxanthine or xanthine [the oxypurines that Yamamoto et al. (1997) are referring to]. Does Pi repletion increase or decrease ischemia-induced glycolytic activity? Pi repletion generally does increase the activities of glycolytic enzymes, in many of the articles I've seen, but it could also reduce the kinds of wild fluctuations in the intracellular pH that can occur during ischemia. The Pi-induced increases in glycolytic activity by allosteric mechanisms could increase the cytosolic NADH/NAD+ ratio [Zhou et al., 2005: (http://jp.physoc.org/content/569/3/925.full.pdf+html)(http://www.ncbi.nlm.nih.gov/pubmed/16223766?dopt=Abstract)] and inhibit xanthine dehydrogenase activity (meaning that, from a simplistic standpoint, that effect could decrease uric acid formation and enhance purine salvage, conceivably), and, in the absence of a high intake of a phosphate salt displaying an abnormal ratio of monobasic to dibasic orthophosphate (orthophosphate refers to [HPO4(2-) + H2PO4(-) + the less-than-1-% contribution of PO4(3-)]), Pi repletion can produce an alkalinizing effect that could also activate glycolysis and further reduce xanthine dehydrogenase activity. But it could also exert more of a neutral effect. Those are just speculative thoughts.

For that matter, I wonder if the alkalotic effects of excesses of Pi might abolish or decrease the mitochondrial proton gradient under some circumstances, by mimicking the effects of uncouplers. Pi could conceivably stimulate respiration by that mechanism [that commonly occurs as a compensatory response (http://scholar.google.com/scholar?hl=en&q=stimulate+uncoupler+mitochondrial+respiration)], and that could explain those articles I cited, in a past posting, showing that Pi can increase the postprandial metabolic rate in humans, etc. That could conceivably account for some of its supposed psychiatric or psychoactive effects, and the "pseudodepression" and other effects of Pi depletion could be due to the poor "regulation" of the mitochondrial membrane potential. There are all sorts of articles showing that the stimulation of respiration is associated with phosphate influx into mitochondria, and phosphate influx interacts with ADP-stimulated respiration, etc. The point is that the effects of different concentrations of intracellular or intramitochondrial Pi on respiration could conceivably be either "bad" or "good," depending on the way you look at the effects.

It would be interesting to see some in vivo research on the effects of Pi depletion or repletion on the exercise-induced loss of purine nucleotides, for example, because it could be a complex set of effects. It's interesting that Hellsten et al. (1999) [Hellsten et al., 1999: (http://jp.physoc.org/content/520/3/909.full)(http://www.ncbi.nlm.nih.gov/pubmed/10545153?dopt=Abstract)] argued that the initial effect of exercise had been to increase Pi availability, thereby inhibiting AMP deaminase activity, but that the decreases in intracellular pH that had subsequently occurred had activated AMP deaminase activity. It's interesting that an increase in the inhibition of AMP deaminase by Pi would tend to lead to a relative increase in adenosine availability, and some of that adenosine would presumably serve to increase blood flow to the exercising muscles. I wonder if that increase could lead to a greater loss of adenosine, however, or if the Pi-mediated inhibition of AMP deaminase activity (as in the endothelial cells in which much of the adenosine deaminase-mediated deamination of interstitial-fluid adenosine occurs) would mean that more adenosine could be released and then also salvaged. The intracellular and extracellular adenosine concentrations are not usually very different, and there's a slight, inwardly-directed, transmembrane adenosine gradient. Usually, one thinks of adenosine release as being a unidirectional process that's "coupled" to an increase in the degradation, by adenosine deaminase in endothelial cells, of the adenosine to inosine and hypoxanthine. But, presumably, that's not always going to be the case. It's interesting that uncouplers are used to increase extracellular adenosine concentrations [see the reference to "respiratory uncouplers" on the first page of Rubio et al., 1972: (http://www.ncbi.nlm.nih.gov/pubmed/5022662)], and my overall point is that excessive concentrations of intracellular Pi, to the extent that they are achievable, could conceivably have some adverse effects that would go beyond the well-known increases in the risk of calcification, etc.

Formation of ATP Derivatives of Intermediates in Cholesterol Biosynthesis: Potential Relevance to Energy Metabolism in the Brain, etc.

This article [Sillero et al., 2009: (http://www.ncbi.nlm.nih.gov/pubmed/19414000)] discuss the capacity of some bisphosphonates to inhibit farnesyl pyrophosphate synthetase, thereby inhibiting cholesterol biosynthesis and also the posttranslational prenylation of GTPases and other proteins. The authors also found that a variety of ligase enzymes, such as DNA ligases, apparently, cleave ATP and, as part of their catalytic mechanisms, attach the AMP derived from that cleavage to the ligase enzyme or to a cosubstrate, X (it's not clear what the cosubstrates are, and I don't feel like looking them up). This forms an E-X-AMP or E-AMP complex of AMP with the enzyme. Some of the ligase enzymes can then transfer the AMP to isopentenyl pyrophosphate (Iso-pp) or to a bisphosphonate and form Iso-pppAdenosine or "ATP derivatives" of bisphosphonates that may contribute to the induction of apoptosis in osteoclasts or monocyte-macrophage lineage cells, including osteoclast progenitor cells. One mechanism by which these ATP derivatives are thought to cause apoptosis is through the inhibition of the mitochondrial adenine nucleotide translocase transporter. Faust et al. (1980) [Faust et al., 1980: (http://www.jbc.org/cgi/reprint/255/14/6546.pdf)(http://www.ncbi.nlm.nih.gov/pubmed/7391033)] found that aminoacyl-tRNA synthetases could form delta2-isopentenyl-tRNA, and those tRNAs are similar to the ATP derivatives, discussed above, in the sense that the isopentenyl group is transfered to the N6 nitrogen atom of an adenosine that is part of the tRNA. Faust et al. (1980) claim that, when cholesterol levels are low in some cells, more mevalonate is diverted into cholesterol formation than into ubiquinone and isopentenyl derivatives (isopentenyladenosine is one commonly-discussed derivative, and I'm actually forgetting if that's the same thing as Iso-pppAdenosine or is different from that). I don't know that that's true, because supposedly the ubiquinone pathway, at least, is maintained very efficiently under almost all circumstances, as I recall. It's only when HMG-CoA reductase activity has become greatly decreased that ubiquinone formation decreases, as I remember. But Faust et al. (1980) nonetheless show that inhibiting HMG-CoA reductase can decrease mevalonate levels, as is well-known, and thereby reduce the formation of some of these endogenous isopentenyl-adenosine derivatives (either free Iso- derived species or tRNA-bound derivatives, etc.).

This could be relevant to the roles that decreases in the neuronal or astrocytic cholesterol contents may play in the etiologies of neurodegenerative or psychiatric conditions, in my view. If the cellular cholesterol concentration that is capable of influencing HMG-CoA reductase activity in a cell in the brain were to be too low to maintain low levels of these nucleotide derivatives of isoprenoids, it's conceivable that that could, in some way, influence energy metabolism. The isopentenyl adenosine derivatives might interfere with nucleotide transport or contribute to DNA damage, etc. I haven't thought about the details of this, but Sillero et al. (2009) discuss the very large capacity of the mevalonate-derived isoprenoids and other cholesterol-biosynthetic intermediates to sequester pyrophosphate, and that could compromise adenosine nucleotide salvage. Fructose, for example, can cause ATP depletion in the liver by sequestering phosphate in one of the fructose bisphosphates (I forget which one), etc. These are fairly crude suggestions, but it's an interesting area of research.

Note on "Fructose Panel"; Phosphate Sequestration by Fructose And Other Compounds

This article, describing the findings of a panel discussion in which people compared some of the effects of fructose and sucrose, is noteworthy:

http://www.webmd.com/food-recipes/news/20081211/high-fructose-corn-syrups-bad-rap-unfair

If one reads the article carefully, one can see that the panel is saying that fructose is no more likely to contribute to obesity than fructose. There is, first, the statement that sucrose, a sugar that is the predominant disaccharide in table sugar, is about 50 percent fructose by mass. Assuming no part of a dose of sucrose is absorbed as a disaccharide (and it looks like it isn't), then it's probably valid to assume that eating 100 grams of sucrose will produce plasma fructose and glucose levels that are the same as those that would be produced by eating 50 grams of free fructose and 50 grams of free glucose. I'm not sure that this would always be the case, given that the pharmacokinetics of sugars can differ among individuals, but let's suppose that it would be. Then there's the statement that high-fructose corn syrup is about 55 percent fructose, meaning that sucrose is no more likely than high-fructose corn syrup to produce obesity.

I don't, personally, agree with this statement, and, in my opinion, the issue is the amounts of fructose that people are more or less forced to eat (not the comparison to sucrose), given the large amounts of fructose that are in so many foods. High-fructose corn syrup is added to so many products that it's almost impossible, in my view, to avoid eating significant amounts of fructose. But the implication of the article is that fructose is metabolized in the same way as glucose, derived from "free" glucose or glucose from complex carbohydrates, is.

Fructose has been shown to produce metabolic changes that can be very different from the changes that glucose, given to the same individual animal or person, produces, and these differences can be particularly significant in animals or people who have ingested glucose in a slow-release form, such as some cereal grains might provide, as opposed to a quickly-released form. The article above mentions an article showing that fructose is more efficiently converted into triglycerides. This is likely to be the case, in my opinion, and there are hundreds if not thousands of research articles that have shown that to be the case. Fructose can also cause significant depletions of both adenine and guanine nucleotides from the liver and, when it's infused directly into the kidneys, from the kidneys as well. These effects of fructose have been known for many years.

I don't have time to post the references/links to all the research, but the study of metabolism doesn't, in my view, boil down to the adding up of calories in and calories out. The depletion from the liver of ATP, in particular, and purine nucleotides in general can be very significant in response to fructose infusion or ingestion. The mechanisms underlying two major fructose-induced metabolic effects--namely ATP depletion and increase in fatty acid synthesis--are fairly well-understood. Here's one article that discusses the mechanisms to some extent. The idea is that fructose enters the liver (or, in this case, the kidneys) en masse, sequesters phosphate (inorganic phosphate, Pi) by its conversion into monophosphorylated and diphosphorylated metabolites (mostly fructose-1-phosphate, in this article, but also glycerol-1-phosphate, as monophosphorylated metabolites), and, as a result of phosphate depletion, reduces the amount of phosphate that's available for maintaining ATP. ATP depletion per se leads to the loss of purine nucleotides, and then the nucleotides are converted into uric acid. There might be other mechanisms, and I'll try to read up on the topic. The uric acid that the nucleotides are converted into is, incidentally, very unlikely to be responsible for the metabolic syndrome, in many cases (I've discussed this in previous postings). I don't have time to go into the explanations, but here are the articles discussing the phosphate depletion mechanism as a mechanism by which fructose ingestion can lead to the depletion of ATP, GTP, and purines overall:

http://www.jbc.org/cgi/content/abstract/255/17/8239
Pubmed Unique ID: (http://www.ncbi.nlm.nih.gov/pubmed/6773936?dopt=Abstract)

This article, below, is actually really interesting and discusses the sequestration of phosphate, as fructose-1-phosphate, in response to fructose loading, or as glycerol-1-phosphate, in response to glycerol, as a mechanism that leads to adenine nucleotide depletion from the liver. That might be relevant for understanding the adverse effects, namely ATP depletion or a reduction in the ATP/(total adenine nucleotide) ratio (or in the so-called adenylate charge), that could potentially result from the excessive accumulation of exogenous purines in the liver (or from exogenous ribose, although ribose hasn't been shown to cause nearly the same degree of adenine nucleotide depletion that xylitol and fructose and glycerol produce). There would be more purines to buffer the phosphate-sequestration-induced losses, but the effect might occur. Here's that article, though, discussing phosphate sequestration, by fructose and other metabolic intermediates, as a mechanism for purine nucleotide depletion:

http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1896274
(pubmed unique id: http://www.ncbi.nlm.nih.gov/pubmed/17324122)