59 replies to this topic

[niner]

Yes, Walle was speculative. However, the two rat studies you referenced may have only been discussing metabolites and not resveratrol. The first says "resveratrol" but then mention it is metabolized to aglycone and glucuronid. The second one definitely says it only measured the sulphate or glucuronate conjugates metabolites. So we don't know what the free resveratrol concentration is after oral ingestion by rats or mice. A newer study. Another None of them seem to mention "free resveratrol" in the blood of rats.

No, these studies discussed both free resveratrol, which is sometimes called the "aglycone", and the conjugates. In http://pmid.us/12065739 (Marier et al. J. Pharm. Exp. Ther. 2002) which you referenced as the second one, they dosed rats orally with 50mg/kg resveratrol and measured a Cmax in plasma of 6.6 uM for free resveratrol, (which in their notation is RESagl), and 105uM for the glucuronide. In Sale et al. ('the first') you can't tell from the abstract. I got the value of 32 uM from Anthony Loera, who has the full paper. The paper that you referenced as 'A newer study', http://pmid.us/16869992 (Abd El-Mohsen et al., Br. J. Nutr., 2006) looks at disposition of metabolites from a dose of tritiated resveratrol. They say "However, at 18 h the main form identified in liver, heart, lung and brain was native resveratrol aglycone, [free resveratrol] indicating that it is the main form retained in the tissues.", which is interesting, but the abstract doesn't give quantities for either the dose or the amount remaining in tissue, so it's hard to say what to make of it. In the last paper you linked, they use the phrase "resveratrol and its metabolites", indicating that "resveratrol" is the free or unconjugated or (a bit less accurately) aglycone form. There wasn't really any pertinent data there. So anyway, I think you are getting thrown off because you are not familiar with the jargon.

Back in 2002:
"Incubation with beta-glucuronidase and sulfatase to release free resveratrol was used to confirm the structures of these conjugates." ref 4 Makes me wonder if mice and men do the same "de-gluconation" and "de-sulfation", and if this is a well-known method of delivering compounds to tissue. The most abundant metabolites in humans, rats, and mice are trans-resveratrol-3-O-glucuronide and trans-resveratrol-3-sulfate. I suspect the body uses the modifications to determine friend from foe and to get them transferred to different locations in the body, through different membranes. Once there, there might be a mechanism for turning it back into free resveratrol. Suppose free resveratrol does not get inside cells as well. Then these studies with direct immersion in resveratrol may not be able to get as high a dose as we can get orally. I'm just saying...it's possible.

Both mice and men have enzymes that cleave glucuronides and sulfates, but the liver is not conjugating molecules from outside the body (xenobiotics) to direct them anywhere except the toilet. The liver is viewing these molecules as toxins and is highly evolved to get rid of them. It's entirely possible that a small percentage of the conjugates is making it through cell membranes, and perhaps even gets enzymatically cleaved back to the native form. Resveratrol itself is going to diffuse through membranes better than the conjugates by a good margin, because it is more lipophilic than the conjugate forms. We might be able to get some clue as to the extent of this from Abd El-Mohsen et al.

Note on comparative dosing: Here's an article saying the RESV dose/kg reported in media was very unfortunate http://www.ncbi.nlm....pubmed/17942826

These interspecies scaling or allometric methods are not very accurate. They can sometimes be off by over an order of magnitude! These methods are used before you put a drug in humans for the first time, when you don't have any blood data to go on. Now that we have plasma levels for resveratrol in humans and several other species, there is no point in using scaling methods. We know it doesn't work for resveratrol.

[zawy]

Glucuronidation is used to make a compound water-soluble and is used by steroid hormones and non-polar drugs for delivery, not just excretion. The drug or steroid is released by glucuronidase in lysosomes (external to mitochondria).

You said they found 6.6 uM for the AGL form and 105 uM GLU, a factor of 16 different, which is close to the factor of 23 found in humans. Even in the abstract they write:
" RES(AGL) ... exposure was approximately 46-fold lower than that of RES(GLU)"
"These results confirm that RES(AGL) is bioavailable and undergoes extensive first-pass glucuronidation"

Your second link:
(mice) "Resveratrol was metabolised to its sulphate or glucuronate conjugates" although the abstract doesn't say how long it took.

So I still haven't seen a reference that indicates rats have substantially more ACL/free-RESV in blood than humans.

Ref 5 seems to be the one that supports your viewpoint (where they found RES(AGL) in organs after 18 hr), but it's changing the argument from concentration in blood to concentration in tissue. That's an important change because blood concentration is relevant in comparing rat-to-human doses, especially since we have no data on human tissue concentration. Said another way, evidence for RESV-AGL in rat tissue is not evidence of lack of GLU et al forms in human tissue. So the theory that humans make use of the conjugated forms is unchallenged. ref 4 stated after looking at ref_5 and ref 3, "The major tissue metabolites were resveratrol-3,4'-disulfate, -3,4',5-trisulfate and β-D-glucuronide." I'm not sure where AGL form stands in that statement.

But remember ref_5 was supposed to find free resveratrol in organ tissue because the whole idea is that it is released from the glucuronate by glucuronidase once it's inside the cell. Since free-RESV is not water-soluble, can it even get into cells from the blood or is it always split from a GLU form once inside the cell? Do any in-vitro experiments use free-RESV straight in water, or do they all use ethanol or DMSO as a carrier?

Here's a good one ref 1:
(mice, rats, and humans) "Our results indicate that trans-resveratrol-3-O-glucuronide and trans-resveratrol-3-sulfate are the most abundant metabolites of resveratrol. Virtually no unconjugated resveratrol was detected in urine or serum samples, which might have implications regarding the significance of in vitro studies that used only unconjugated resveratrol."

Ref 2 is the third group of researchers i've seen who speculate "the ubiquitously existing human beta-glucuronidase could convert the metabolites back to resveratrol locally or systematically in vivo". They showed RESV > 30 uM was cytotoxic to human cells, but up to 300 uM was safe for the metabolites.

"Glucuronate-conjugated drugs are more easily cleared from the blood by the kidneys for excretion in the urine. The glucuronate-drug conjugation system can, however, lead to drug resistance; chronic exposure to certain drugs, such as barbiturates and AZT, leads to an increase in the synthesis of the UDP-glucuronyltransferases in the liver that are involved in glucuronate-drug conjugation. The increased levels of these hepatic enzymes result in a higher rate of drug clearance leading to a reduction in the effective dose of glucuronate-cleared drugs"

Edited by zawy, 17 January 2008 - 07:32 PM.

[niner]

Glucuronidation is used to make a compound water-soluble and is used by steroid hormones and non-polar drugs for delivery, not just excretion. The drug or steroid is released by glucuronidase in lysosomes (external to mitochondria).

There have been a few instances where synthetic drugs were designed to exploit the existence of glucuronidase, in cases that I'm aware of it's been done as part of a scheme to protect the drug from degradation prior to it getting into the cell. If a drug is extremely hydrophobic, there will be problems delivering it. The addition of a glucuronide group will reduce the hydrophobicity, and may bring a very hydrophobic drug into a more favorable range of hydrophobicity. In resveratrol's case, it is not that hydrophobic to begin with (logP ~3) and reducing it a lot is going to make it wash out of the body. As I said previously, xenobiotic metabolism is evolved to remove unnatural compounds from the body. It does this by making them more hydrophilic, typically by oxidation and/or conjugation, so that they wash out of the body.

You said they found 6.6 uM for the AGL form and 105 uM GLU, a factor of 16 different, which is close to the factor of 23 found in humans. Even in the abstract they write:
" RES(AGL) ... exposure was approximately 46-fold lower than that of RES(GLU)"
"These results confirm that RES(AGL) is bioavailable and undergoes extensive first-pass glucuronidation"
[...]
So I still haven't seen a reference that indicates rats have substantially more ACL/free-RESV in blood than humans.

My point is that the rodents have a higher plasma level of free resveratrol than humans do for a given mg/kg dose. If the interspecies scaling worked in this case, a mouse should have 1/12 the plasma level as a human, and a rat 1/6, for the same relative dose. We aren't seeing that. The ratio of free to conjugated resveratrol is an interesting point, suggesting that there's more to the difference between rodents and humans than just conjugation.

Ref 5 seems to be the one that supports your viewpoint (where they found RES(AGL) in organs after 18 hr), but it's changing the argument from concentration in blood to concentration in tissue. That's an important change because blood concentration is relevant in comparing rat-to-human doses, especially since we have no data on human tissue concentration. Said another way, evidence for RESV-AGL in rat tissue is not evidence of lack of GLU et al forms in human tissue. So the theory that humans make use of the conjugated forms is unchallenged. ref 4 stated after looking at ref_5 and ref 3, "The major tissue metabolites were resveratrol-3,4'-disulfate, -3,4',5-trisulfate and β-D-glucuronide." I'm not sure where AGL form stands in that statement.

What ref 5 shows is that the native resveratrol hangs out in tissues at some (possibly low) level, but that the conjugates have washed out, which is what you would expect. It would be hard to prove that conjugates were not present in human tissues short of a biopsy on a grad student, I suppose, but there is no evidence that they are there, nor, frankly, is there any reason to believe that they would be there other than in minuscule quantities. The reference to "tissue metabolites" in the Acta Pharm Sinica paper may have been to resveratrol that was metabolized in tissues other than the liver; hard to know exactly.

But remember ref_5 was supposed to find free resveratrol in organ tissue because the whole idea is that it is released from the glucuronate by glucuronidase once it's inside the cell. Since free-RESV is not water-soluble, can it even get into cells from the blood or is it always split from a GLU form once inside the cell? Do any in-vitro experiments use free-RESV straight in water, or do they all use ethanol or DMSO as a carrier?

Resveratrol is a lot more soluble in plasma than in water, because it can complex with albumin. The moderate hydrophobicity of resveratrol will allow it to diffuse through cell membranes relatively easily. There may also be active transport mechanisms that concentrate resveratrol in the tissues. We don't need to invoke intracellular deconjugation to explain how resveratrol gets into cells.

Here's a good one ref 1:
(mice, rats, and humans) "Our results indicate that trans-resveratrol-3-O-glucuronide and trans-resveratrol-3-sulfate are the most abundant metabolites of resveratrol. Virtually no unconjugated resveratrol was detected in urine or serum samples, which might have implications regarding the significance of in vitro studies that used only unconjugated resveratrol."

It was low dose, they were under the detection limit.

Ref 2 is the third group of researchers i've seen who speculate "the ubiquitously existing human beta-glucuronidase could convert the metabolites back to resveratrol locally or systematically in vivo". They showed RESV > 30 uM was cytotoxic to human cells, but up to 300 uM was safe for the metabolites.

Lots of people look at the very efficient conjugation and low levels of free resveratrol, and they speculate about active metabolites. It's all speculation. Albumin and diffusion are enough to explain free resveratrol getting into cells, not to mention the possible existence of active transport mechanisms.

"Glucuronate-conjugated drugs are more easily cleared from the blood by the kidneys for excretion in the urine. The glucuronate-drug conjugation system can, however, lead to drug resistance; chronic exposure to certain drugs, such as barbiturates and AZT, leads to an increase in the synthesis of the UDP-glucuronyltransferases in the liver that are involved in glucuronate-drug conjugation. The increased levels of these hepatic enzymes result in a higher rate of drug clearance leading to a reduction in the effective dose of glucuronate-cleared drugs"

Yup.

[zawy]

If the interspecies scaling worked in this case, a mouse should have 1/12 the plasma level as a human, and a rat 1/6, for the same relative dose. We aren't seeing that. The ratio of free to conjugated resveratrol is an interesting point, suggesting that there's more to the difference between rodents and humans than just conjugation.

Scaling implies the area under curve of plasma concentration vs. time, not the peak, should be 6 times less for the rat. However, 6 is not the right factor. As a previous post of mine explained, the FDA and EPA explicitly use the less accurate (kg rat/kg human)^1/3 because it gives a larger margin of safety for toxicology (drugs and pollutants). The 1/4 factor is known to be a more accurate method and is based on metabolic rate (Calorie consumption/kg/day) rather than strict surface area. This gives a factor of 3.8 for the 70 kg man verses 335 g rat.

In rat metabolism, the half life was 8 minutes calculated from the initial points and a "terminal" half life was 1.3 to 1.6 hours for free and total RESV when calculated from the last 3 datapoints. It's not clear what that means. 5 times the 1.5 hour half life should have at least 3% remaining, but this was true only for total RESV. Free-resv was lost/absorbed faster. I believe Boocock said 4 hours (probably free and total) in humans and Walle said 9 for total resv. The Walle article on humans also shows a second peak from 4 to 8 hours, but Walle didn't calculate a terminal half life for us. It took 72 hours for humans to lose what rats lost in 12 hours (total RESV). Humans had 11 times more total-RESV area under the curve (per mg/kg dose), 3 times more than scaling predicts. Supporting data: humans: 26 um*h/L at 0.357 mg/kg Oral (Walle et al), rats: were 328 at 50 mg/kg. Area under the curve in rats for free-RESV was 46 times lower than GLU. Free-Resv was 23 times less compared to both GLU and sulfate forms in humans (Boocock). One caveat: the rat dose was much higher than the Walle study (but not Boocock), which may bring some of the large difference back down to levels expected by scaling.

So humans have 11 times more total RESV and at least 22 times more free-RESV than mice when dosing is on a mg/kg basis. This is 3 and 6 times more RESV in the blood of humans after adjusting for scaling.

The rats they used weighed an average of 335 g which is the exact weight for the 3.8 / 6 scaling factors mentioned above.

Edited by zawy, 18 January 2008 - 07:29 PM.

[niner]

If the interspecies scaling worked in this case, a mouse should have 1/12 the plasma level as a human, and a rat 1/6, for the same relative dose. We aren't seeing that. The ratio of free to conjugated resveratrol is an interesting point, suggesting that there's more to the difference between rodents and humans than just conjugation.

Scaling implies the area under curve of plasma concentration vs. time, not the peak, should be 6 times less for the rat. However, 6 is not the right factor. As a previous post of mine explained, the FDA and EPA explicitly use the less accurate (kg rat/kg human)^1/3 because it gives a larger margin of safety for toxicology (drugs and pollutants). The 1/4 factor is known to be a more accurate method and is based on metabolic rate (Calorie consumption/kg/day) rather than strict surface area. This gives a factor of 3.8 for the 70 kg man verses 335 g rat.

In rat metabolism, the half life was 8 minutes calculated from the initial points and a "terminal" half life was 1.3 to 1.6 hours for free and total RESV when calculated from the last 3 datapoints. It's not clear what that means. 5 times the 1.5 hour half life should have at least 3% remaining, but this was true only for total RESV. Free-resv was lost/absorbed faster. I believe Boocock said 4 hours (probably free and total) in humans and Walle said 9 for total resv. The Walle article on humans also shows a second peak from 4 to 8 hours, but Walle didn't calculate a terminal half life for us. It took 72 hours for humans to lose what rats lost in 12 hours (total RESV). Humans had 11 times more total-RESV area under the curve (per mg/kg dose), 3 times more than scaling predicts. Supporting data: humans: 26 um*h/L at 0.357 mg/kg Oral (Walle et al), rats: were 328 at 50 mg/kg. Area under the curve in rats for free-RESV was 46 times lower than GLU. Free-Resv was 23 times less compared to both GLU and sulfate forms in humans (Boocock). One caveat: the rat dose was much higher than the Walle study (but not Boocock), which may bring some of the large difference back down to levels expected by scaling.

So humans have 11 times more total RESV and at least 22 times more free-RESV than mice when dosing is on a mg/kg basis. This is 3 and 6 times more RESV in the blood of humans after adjusting for scaling.

The rats they used weighed an average of 335 g which is the exact weight for the 3.8 / 6 scaling factors mentioned above.

Really, it's neither the Cmax nor the AUC we're interested in, but rather the length of time that concentrations are above the EC50 for SIRT1 deacetylation enhancement. It's not entirely clear how this is related to either Cmax or AUC, but I suspect it will track Cmax better. The plasma levels are a proxy for the levels in the cell, and a poor one at that. Unfortunately plasma levels are what we can measure. If you want to scale rats by 3.8 instead of the usual 6, that's fine. However, the pertinent comparison here would be free resveratrol AUC (not total resv) of both rats and humans at high doses. We have the value of 7.1 um*h/l for rats from Marier, but unless someone can get the full text of Boocock I don't think we can do much that's sensible. Where did you get an AUC for free resveratrol in humans? Walle didn't break it out from the total because he was using a radioactivity method rather than HPLC.

[zawy]

Where did you get an AUC for free resveratrol in humans? Walle didn't break it out from the total because he was using a radioactivity method rather than HPLC.

Boocock said AUC for free was "up to" 23 times less than GLU+Sulphate forms (unknown concentration..across the board?). Marier said free was 46 times lower than GLU in rats. So the ratio is twice as good in humans. Combine that with the total being 11 times higher in humans (Walle vs Marier) to get the "at least 22" factor. If there are other metabolites that are substantial in rats(sulphate?) the total/free ratio gets even higher, but it's cancelled out because the 11 factor would go lower, so the 22 remains. Walle was at much smaller doses than the other 2, so it's really stretching it, but there's some hope for comparability because Boocock got 500 ng/ml free-resv at a dose 200 times higher than Walle, so Boocock would predict Walle to get 2.5 ng/ml and Walle said < 5 ng/ml. Comparability between Boocock and Marier insn't bad since Boocock used up to 70 mg/kg and Marier used 50 mg/kg. The rats showed 6.6 ng/ml at 50 mg/kg which supposedly would be 9.2 ng/ml at 70 mg/kg. That's 539/9.2 = 60 times higher peak free-resv in humans than in rats on a mg/kg basis. The rat curves show free-resv to drop off much faster than GLU. Apparently, the human free-resv does not drop off nearly as fast relative to the total metabolites which is why this factor is even higher than the 22 factor of the AUC difference.

I do not insist that we use 3.8 for the scaling of a 335 g rat. The FDA and EPA insists that we use it to be accurate. The FDA and EPA insist that you use 6 if you consider RESV a toxin. Remember we divide by these factors. We talk in the inverse because 3.8 is easier to remember than 0.26. If resv is an "anti-toxin", you can argue that the FDA and EPA imply that we should have a safety factor going the other way and use 2.4 to divide by. That gives 0.17 as the scaling factor instead of .33 or .25. For the fat mice study, we have 22 mg/kg/day with 45 g mice: HED=22 x (45/70,000)^0.17 x 70 kg = 441 mg/day for humans. Using your Cmax theory, this means we should take 441/60 = 7.35 mg/day to get an equivalent dose....equal to a bottle of wine.

[maxwatt]

...snip...
... Using your Cmax theory, this means we should take 441/60 = 7.35 mg/day to get an equivalent dose....equal to a bottle of wine.

I'll drink to that. Do we have any figues for canines? I haven't found them.

[zawy]

I made a mistake: the 60 times less already has the metabolism factor in it because it was direct measurement. So niner's Cmax method would give 22 mg/kg rat / 60 *70 kg = 25 mg/day for a human. It also assumes the mice has similar metabolism to the rat. Also, 539 to get the 60 factor had a large standard deviation, around 350.

Niner and i are sure not going to say 25 mg/day is the optimum dose, as in vitro shows much greater benefit from much higher doses. The people who drink a bottle of wine a day have been shown to do almost as well as the fat mice, so it may not be such a bad estimate. But i'll stick with 22 times less (free-AUC) instead of 60 times less (Cmax), which gives 70 mg/day for comparision to the fat mice study.

To me it makes intuitive sense that humans would be 3 to 6 times better at utilizing a flavonoid (after adjusting for metabolism rate), but i can't think of why mice or rats would have evolved to consume a lower percentage of plants. If Cmax method is right, we are 60/3.8=16 times better. Maybe I'm thinking we have a more complex and interesting GI/liver/kidney system that can do more interesting things with complex compounds. It would be interesting to see if other flavonoids and complex phytonutrients follow a similar pattern.

Edited by zawy, 19 January 2008 - 04:57 PM.

[maxwatt]

....

To me it makes intuitive sense that humans would be 3 to 6 times better at utilizing a flavonoid (after adjusting for metabolism rate), but i can't think of why mice or rats would have evolved to consume a lower percentage of plants. If Cmax method is right, we are 60/3.8=16 times better. Maybe I'm thinking we have a more complex and interesting GI/liver/kidney system that can do more interesting things with complex compounds. It would be interesting to see if other flavonoids and complex phytonutrients follow a similar pattern.

Mice and rats tend to eat seeds rather than leaves and roots, where flavonoids would be more concentrated.
I still don't follow the logic that humans need less, not more resveratrol. Haven't the glucuronicated and sulfonated compounds of resveratrol been found inactive? You cannot assume they are functionally the same, or that they function as a delivery mechanism. My own experience with osteoarthritis and hallux rigidus indicate we need more, not less, than rodents for a comparable effect.

[zawy]

Looking back at Mohsen rat study we see total resv in tissue to be approximately 1/2 of the plasma concentration at both 2 and 18 hours. At 2 and 8 hr, about 0.8% and 0.35% of the oral dose was in tissue. At 18 hr, free-resv was the "main form" remaining in tissue. Let's say "main form" means 50%. Let's say only 25% of the 0.8% found at 2 hr was free-resv. So we have 0.2% free-resv in organs at 2 hr. However, Marier found free-resv in plasma to be 0.1% of the oral dose for the first 2 hours (averaged 3 uM * 228 g/mole * 8% blood/kg / 50mg/kg dose). Let's be skeptical and assume the 0.1% is not concentrating to get the 0.2% and also that metabolites are not being converted back to free form inside tissue. Mohsen found 1/2 as much total resv in tissue verses blood, so let's say only 0.05% of free-resv was in tissue at two hours. This is mixing Mohsen and Marier in a skeptical manner. Using 3.8, this should be marginally close 7.5 hour in humans. Being skeptical again, we can use my factor of 22 based on AUC instead of 60 based on Cmax when trying to convert to humans. So we have 1.1% of the oral dose is free-resv in organs at 7.5 hours. So 1 g dose with say 30 kg (30 L?) of organs allows humans to have at least 1.6 uM of free-resv in tissue at 7.5 hours. The most positive view would use 0.2% in tissue, Cmax 60 times factor for converting to humans, and 10 kg for organs to get 50 uM. If i didn't make any errors too big, and if the papers are solid, this should be a valid max and min.

One thing is that Mohsen says "plasma concentrations reached 1.7 % of the administered dose". I assumed above that he's comparing total mg in blood to total mg dose, and not doing something strange like concentration in blood as a pecent of concentration in body weight. I don't have that paper.

Edited by zawy, 19 January 2008 - 08:06 PM.

[maxwatt]

Looking back at Mohsen rat study we see total resv in tissue to be approximately 1/2 of the plasma concentration at both 2 and 18 hours. At 2 and 8 hr, about 0.8% and 0.35% of the oral dose was in tissue. At 18 hr, free-resv was the "main form" remaining in tissue. Let's say "main form" means 50%. Let's say only 25% of the 0.8% found at 2 hr was free-resv. So we have 0.2% free-resv in organs at 2 hr. This is feasible because Marier showed free-resv to be about 4% of AUC and 6% of Cmax for the first 2 hours. Using 3.8, this should be marginally close 7.5 hour in humans. Being skeptical, we can use my factor of 22 based on AUC instead of 60 based on Cmax when trying to convert to humans. So we have 4.4% of the oral dose is free-resv in organs at 7.5 hours. So 1 g dose with say 30 kg (30 L?) of organs allows humans to have at least 6 uM of free-resv in tissue at 7.5 hours.

The reasoning is plausible, but it has to be tested to be validated. Too may assumptions, data from disparate studies to be reliable at this point. My own experience is that I don't get certain effects that involve inhibition of NF-kappa B with less than 500 mg doses, and more is better.

[missminni]

The reasoning is plausible, but it has to be tested to be validated. Too may assumptions, data from disparate studies to be reliable at this point. My own experience is that I don't get certain effects that involve inhibition of NF-kappa B with less than 500 mg doses, and more is better.

I agree. So does my dad, and there's no study
saying it's harmful. In fact the contrary is true.

[niner]

Where did you get an AUC for free resveratrol in humans? Walle didn't break it out from the total because he was using a radioactivity method rather than HPLC.

Boocock said AUC for free was "up to" 23 times less than GLU+Sulphate forms (unknown concentration..across the board?). Marier said free was 46 times lower than GLU in rats. So the ratio is twice as good in humans. Combine that with the total being 11 times higher in humans (Walle vs Marier) to get the "at least 22" factor. If there are other metabolites that are substantial in rats(sulphate?) the total/free ratio gets even higher, but it's cancelled out because the 11 factor would go lower, so the 22 remains. Walle was at much smaller doses than the other 2, so it's really stretching it, but there's some hope for comparability because Boocock got 500 ng/ml free-resv at a dose 200 times higher than Walle, so Boocock would predict Walle to get 2.5 ng/ml and Walle said < 5 ng/ml. Comparability between Boocock and Marier insn't bad since Boocock used up to 70 mg/kg and Marier used 50 mg/kg. The rats showed 6.6 ng/ml at 50 mg/kg which supposedly would be 9.2 ng/ml at 70 mg/kg. That's 539/9.2 = 60 times higher peak free-resv in humans than in rats on a mg/kg basis. The rat curves show free-resv to drop off much faster than GLU. Apparently, the human free-resv does not drop off nearly as fast relative to the total metabolites which is why this factor is even higher than the 22 factor of the AUC difference.

Zawy, you made a huge mistake here. Marier reported a Cmax of 6.6 uM, not 6.6 ng/ml. 6.6 uM = 1505 ng/ml. This changes everything. Here are the comparisons:

Marier (rat) 6.6 uM/50 mg kg(-1) = 0.132
Boocock (human) 2.4 uM/71 mg kg(-1) = 0.0338

So the rats get 4 times the Cmax that humans do with a comparable oral dose. Based on Cmax, the interspecies scaling is off by a factor of 24.

I do not insist that we use 3.8 for the scaling of a 335 g rat. The FDA and EPA insists that we use it to be accurate. The FDA and EPA insist that you use 6 if you consider RESV a toxin. Remember we divide by these factors. We talk in the inverse because 3.8 is easier to remember than 0.26. If resv is an "anti-toxin", you can argue that the FDA and EPA imply that we should have a safety factor going the other way and use 2.4 to divide by. That gives 0.17 as the scaling factor instead of .33 or .25. For the fat mice study, we have 22 mg/kg/day with 45 g mice: HED=22 x (45/70,000)^0.17 x 70 kg = 441 mg/day for humans. Using your Cmax theory, this means we should take 441/60 = 7.35 mg/day to get an equivalent dose....equal to a bottle of wine.

I'll reiterate: HED calculations are not reliable. They are intended for use in situations when the drug has never gone into humans before. Once you have reliable human pharmacokinetic data, that's what you use. I wouldn't say the Cmax idea is "mine", it's just what most people talk about in early drug development. Actually, I prefer a modified AUC, but the data is harder to come by. Anyway, if you use the ratios of human and rat Cmax per mg/kg dose, to match the fat mice, we should be taking 22mg/kg * (0.132/0.0338) = 86 mg/kg. For a 70 kg human, that's 6 grams per day, or a lot of bottles of wine.

It's entirely reasonable to base these comparisons on AUC, but we need to use a measured AUC in humans at relatively similar doses. There are a ton of assumptions in the way you arrived at the factor of 22, chief among them the relative rate of clearance of metabolites between rats and humans. These could be substantially different. Otherwise it's pretty hard to swallow that the ratio of AUCs is 88 times as large as the ratio of Cmax's.

[zawy]

Thanks for the correction of the tragic mistake. Although I also compared Cmax to metabolic rate, i don't think we should do that, mainly because Cmax in rats is at 5 minutes or so compared to 1.5 hours in humans.

But metabolic rate should apply more to AUC by extending out the time of the curve. If it does we have 22 / 3.8 = 5.8 times more AUC free-resv in humans than in rats, after accounting for metabolism. Cmax of free-resv in rat was 16 times less than GLU form and both occured at about the same time. 2 to 8 hours later, it's 200 times less. If in humans the ratio of free-resv to total resv stays the same (no drop off of the ratio), it would cause about 200/16=13 times of a difference instead of 5.8. So humans also have a drop off of the ratio that is about half as bad as the rats. This assumes total resv in humans and rats follows the metabolic rate.

Why stick to the factor of 6 when 3.8 = mg/Calorie is the truth?

Edited by zawy, 20 January 2008 - 02:00 PM.

[zawy]

I finally got the Boocock study. 0.5 g free-resv absorbs almost twice as efficiently as 5 gram (AUC, Cav, and Cmax). 1 g absorbs just as well as 0.5 g. 2.6% of blood metabolites at 5g was free-resv as AUC. AUC human (high dose) is 18 AUC/(mg/kg) verses 31 AUC/(mg/kg) for rats. So my 22 is way off. Sorry

parameter 0.5 g, 1 g, 2.5 g, 5 g (coeff of variation)

Resveratrol

AUCinf (ng h/mL) 223.7* 544.8 (57.2) 786.5 (36.2) 1,319 (59.1)

Cmax (ng/mL) 72.6 (48.9) 117.0 (73.1) 268.0 (55.3) 538.8 (72.5)

Tmax (h) 0.833 (0.5-1.5) 0.759 (0.5-4.0) 1.375 (0.5-4.0) 1.500 (0.67-5.0)

Cav (ng/mL) 8.36 (57.8) 18.04 (51.6) 32.25 (43.0) 51.90 (80.7)

Half-life (h) 2.85* 8.87 (91.1) 4.22 (51.6) 8.52 (95.8)

CL/F (L/h) 2,235* 2,593 (65.1) 3,471 (29.9) 4,930 (50.0)

CLR (L/h) 1.177 (102.5) 0.696 (71.5) 0.656 (53.1) 1.443 (139.2)

V/F (liters) 9,198* 19,298 (54.3) 22,226 (67.3) 66,991 (112)

Glucuronide 1

AUCinf (ng h/mL) 1,919 (33.6) 3,059 (60.9) 5,664 (27.7) 9,923 (40.9)

Cmax (ng/mL) 404.6 (35.3) 473.6 (76.8) 874.4 (37.5) 1,285 (55.4)

Tmax (h) 2.00 (1.0-6.0) 2.250 (1.0-6.0) 2.375 (1.0-8.0) 2.00 (1.5-5.0)

Cav (ng/mL) 76.9 (37.2) 110.3 (56.1) 215.5 (43.5) 344.1 (51.5)

Half-life (h) 2.85 (48.6) 7.27 (93.9) 10.6 (92.9) 7.90 (39.1)

CL/F (L/h) 282.7 (27.3) 493.5 (74.7) 469.5 (25.7) 590.6 (45.2)

Glucuronide 2

AUCinf (ng h/mL) 1,287 (21.7) 2,589 (66.4) 4,320 (32.9) 8,546 (62.3)

Cmax (ng/mL) 369.5 (39.6) 672.6 (81.1) 1,626 (71.5) 1,735 (66.4)

Tmax (h) 1.500 (1.0-5.0) 1.750 (1.0-5.1) 2.000 (1.0-6.0) 2.520 (1.5-8.0)

Cav (ng/mL) 51.0 (27.6) 99.9 (66.2) 193.8 (39.3) 317.8 (65.6)

Half-life (h) 3.09 (69.8) 6.64 (92.1) 8.42 (88.9) 5.83 (51.2)

CL/F (L/h) 408.8 (26.7) 642.5 (83.0) 636.9 (32.6) 1,017 (94.6)

3-Sulfate

AUCinf (ng h/mL) 4,049 (26.6) 10,053 (73.2) 16,984 (41.7) 30,898 (46.1)

Cmax (ng/mL) 1,135 (25.7) 2,102 (81.3) 2,786 (27.2) 4,294 (48.0)

Tmax (h) 1.500 (1.0-5.0) 2.000 (1.0-5.0) 2.000 (1.0-5.2) 2.050 (1.0-6.0)

Cav (ng/mL) 172.0 (23.2) 402.6 (70.5) 597.0 (27.0) 1,089 (49.4)

Half-life (h) 3.21 (56.6) 4.51 (82.8) 11.5 (95.5) 7.71 (42.3)

CL/F (L/h) 131.2 (25.8) 151.8 (62.7) 171.2 (40.0) 207.8 (63.9)

Edited by zawy, 21 January 2008 - 01:14 PM.

[maxwatt]

Can you explain these numbers? I'm not following the shorthand notation.

[zawy]

Can you explain these numbers? I'm not following the shorthand notation.

At the top shows it. Format is "parameter 0.5 g, 1 g, 2.5 g, 5g" Numbers in parenthesis are the variations from the average for the 10 people. * is 1 sample.

[Hedgehog]

I finally got the Boocock study. 0.5 g free-resv absorbs almost twice as efficiently as 5 gram (AUC, Cav, and Cmax). 1 g absorbs just as well as 0.5 g. 2.6% of blood metabolites at 5g was free-resv as AUC. AUC human (high dose) is 18 AUC/(mg/kg) verses 31 AUC/(mg/kg) for rats. So my 22 is way off. Sorry

parameter 0.5 g, 1 g, 2.5 g, 5 g (coeff of variation)

Thats why I think SRT501 is only about 300-1000mg of resveratrol. I thought there is another paper saying the Resveratrols Cmax was about 25min? How many people where studied?

[maxwatt]

Can you explain these numbers? I'm not following the shorthand notation.

At the top shows it. Format is "parameter 0.5 g, 1 g, 2.5 g, 5g" Numbers in parenthesis are the variations from the average for the 10 people. * is 1 sample.

Thanks.

The Sirtris studies I've seen, they used 5 grams SRT501. I think that's 5 grams resveratrol in a dose of SRT501.

[zawy]

Thats why I think SRT501 is only about 300-1000mg of resveratrol. I thought there is another paper saying the Resveratrols Cmax was about 25min? How many people where studied?

Selected quotes, jammed together. I don't want to attach for copyright reasons.
Mechanistic experiments in vitro suggest tentatively that a concentration of at least 5 Amol/L resveratrol is required to elicit pharmacologic effects
relevant to chemoprevention (5-18). Ten subjects were entered at each dose level beginning at 1 g and then escalating sequentially to 2.5 and 5 g. After pharmacokinetic data were completed for the 5 g dose level, 10 subjects were studied at a dose level of 0.5 g. Blood samples were collected via a sited cannula into heparinized tubes before and after resveratrol administration at the following time intervals:
0.17, 0.33, 0.50, 0.67, 1, 1.5, 2, 2.5, 3, 4, 6, 8, 12, and 24 h.
The aim of this study was to determine whether single high oral doses of resveratrol are safe and yield systemic concentrations associated with chemopreventive activity in cells in vitro (5-18). Cmax concentrations of between 0.3 and 2.4 Amol/L, markedly below the resveratrol concentrations required in in vitro experiments to elicit pharmacologic effects associated with cancer chemoprevention (>5 Amol/L). resveratrol 3-sulfate, was present in the plasma at the 4 to 14 Amol/L
Cmax concentration range, and the more abundant of the two monoglucuronides was found at Cmax concentrations of 0.9 to 4.3 Amol/L. The poor bioavailability of resveratrol, as reflected by its clearance, apparent volume of distribution, and urinary excretion, is reminiscent of comparable data obtained in humans for other polyphenolic phytochemicals, exemplified by the green tea constituent epigallocatchin gallate (30). The low bioavailability of resveratrol across mice, rats, and humans has previously been reported (28, 31-33). In these studies, resveratrol was administered at doses which were considerably lower than those used here. The pharmacokinetic observations described here in humans are essentially consistent with data from experiments in rodents. In rodents, maximal resveratrol concentration was attained rapidly, 10 min post-dose or earlier (34), and parent compound and metabolic conjugates tended to be undetectable beyond 1 h post-dosing (25). The presence of a second peak of resveratrol in the plasma drug concentration-versustime profile observed here (Fig. 2) and the predominant amount of resveratrol compared with its metabolites in the feces are consistent with the hypothesis that resveratrol undergoes enterohepatic recirculation. This interpretation is congruent with results obtained in rats after oral resveratrol(35) and in human volunteers who received a low oral dose (25 mg) of 14C-labeled resveratrol (28). In the light of the amply documented antioncogenic properties of resveratrol in cells in vitro (5-18), its chemopreventive efficacy in rodent models (7, 9, 19-22) is thought to be mediated via the parent compound. The extensive sulfation and glucuronidation of resveratrol with consequent poor parent compound bioavailability, as described here and previously (28, 35), calls into question the role of parent resveratrol in the mediation of cancer chemopreventive efficacy, a notion which has been speculated on before (36). Whereas the pharmacologic properties of resveratrol conjugates are unknown, conjugated metabolites of naturally occurring flavonoids, polyphenols chemically resembling resveratrol, have been suggested to be responsible for, or contribute to, the pharmacologic activity of the parent molecule. For example, in vascular smooth muscle cells, quercetin 3-O-glucuronide inhibited both activity of c-Jun NH2-terminal kinase and binding of transcription factor activator protein-1 to DNA, as potently as its metabolic progenitor (37). Quercetin conjugates also seem to retain, at least in part, the antioxidant properties of the parent molecule (38, 39). The glucuronide of the flavonoid luteolin underwent h-glucuronidase–catalyzed deconjugation at sites of inflammation producing parent aglycon (40). The work described here suggests that systemic concentrations of total resveratrol conjugates achievable in the human biophase after oral resveratrol can reach the range of 10 to 20 uM, grossly estimated on the basis of the sum of concentrations of resveratrol 3-sulfate and resveratrol glucuronides, as described above, and of resveratrol disulfate and resveratrol glucuronide sulfate, which were detected but not quantitated. It seems conceivable that such concentrations engage biochemical mechanisms germane to cancer chemoprevention in tissues targeted for chemoprevention of malignancies either directly or via generation of resveratrol by deconjugation. These possibilities clearly warrant experimental verification. Furthermore, the potential role needs to be explored which metabolites may play in the recently described effects of resveratrol on energy homeostasis and aging in mice in vivo, effects probably mediated in part via activation of the senescence regulator SIRT1 (3, 4). In conclusion, the results presented here suggest that resveratrol undergoes avid metabolism in humans, which limits the availability of the parent molecule at organs remote from the site of absorption targeted for chemoprevention. It remains to be determined if repeated dosing schedules can achieve higher systemic concentrations of resveratrol than those observed here after a single dose, or whether sulfate and glucuronide metabolites, which are generated abundantly in the human biophase after resveratrol ingestion, possess efficacy in and of themselves.

Edited by zawy, 21 January 2008 - 06:15 PM.

[Hedgehog]

[quote name='zawy' date='21-Jan 2008, 04:59 AM' post='220037']

I finally got the Boocock study. 0.5 g free-resv absorbs almost twice as efficiently as 5 gram (AUC, Cav, and Cmax). 1 g absorbs just as well as 0.5 g. 2.6% of blood metabolites at 5g was free-resv as AUC. AUC human (high dose) is 18 AUC/(mg/kg) verses 31 AUC/(mg/kg) for rats. So my 22 is way off. Sorry

parameter 0.5 g, 1 g, 2.5 g, 5 g (coeff of variation)

So i'm i readint this right:

Cmax for resveratrol
0.5g is 72ng/mL
1g is 117.0ng/mL
2.5 is 268ng/mL
5g is 538.8ng/mL

Cmax (ng/mL) 72.6 (48.9) 117.0 (73.1) 268.0 (55.3) 538.8 (72.5)

[zawy]

So i'm i readint this right:

Yes, that's it

[Hedgehog]

So i'm i readint this right:

Yes, that's it

and (#'s) are the time points?

Cmax (ng/mL) 72.6 (48.9) 117.0 (73.1) 268.0 (55.3) 538.8 (72.5)

WOW this is good news for me. Because honestly I think/hope my numbers are really low compared to what they should have been. Basically my number results stayed at a constant 60-70ng/mL for the whole time. And I think this was because I overloaded the C18 column. So my graph was almost linear at that range for almost all time points. When you overload a column basically only X amount stays on it. So you will only get X amount off of the column. Because you are loading centrifuged blood plasma onto the C18 filter I assume it is easy to over load a column. You have a lot more then just resveratrol in blood plasma if you take a filter and put everything onto the filter I assume other particles could stick onto the filter and my sample would simply wash right off in some of the steps. So I guess if I used a larger C18 filter it might have improved my results and also the wash steps probably needed to be perfected. If you don't wash the filter correctly you lose a lot of your sample or compound of interest. This also happened because of my low recovery in the standards.

Since my results stayed linear I assumed that is what happened.

My dose was 402.132mg of resveratrol.

See attached image for one of my chromatographs

Attached Files

•  untitled.JPG   37.33KB   84 downloads

Edited by hedgehog_info, 21 January 2008 - 07:42 PM.

[niner]

Zawy, thanks for all your effort in getting this data!

I've added some rows for PK parameters divided by dose, and removed the SD's. Sorry about the horrid formatting.

Parameter 0.5 g, 1 g, 2.5 g, 5 g
Resveratrol
AUCinf (ng h/mL) 223.7 544.8 786.5 1319
AUC/gm (ng h /mL g) 447.4 544.8 314.6 264
Cmax (ng/mL) 72.6 117.0 268.0 538.8
Cmax/gm (ng/mL g) 145.2 117.0 107.2 107.8

Total Metabolites (Gluc1 + Gluc2 + 3-Sulf)
AUCinf (ng h/mL) 7255 15701 26968 49367
AUCinf/gm (ng h/mL g) 14510 15701 10787 9873
Cmax (ng/mL) 1909 3932 5286 7314
Cmax/gm (ng/mL g) 3818 3932 2115 1463

This represents new and improved evidence regarding "swamping" of gut metabolism. Previously a swamping effect was seen in a monolayer of CACO2 cells, an in vitro model for the gut. With this human data, it appears that there are two or more phenomena. Going from 0.5g to 1.0g, the AUC/gm for resveratrol increases by ~25%, while the metabolite AUC/gm increases by ~10%. This is consistent with swamping, though there may be other explanations. Going to higher doses, however, we see that the AUC/gm falls off significantly for both resveratrol and total metabolites. This is consistent with some sort of limiting factor in absorption, such as a solubility limit in the gut.

The metabolite AUC/gm dose is close enough to the resveratrol AUC/gm if normalized by the maximum in each case, that I suspect it's just reflecting the absorption phenomena. The Cmax for the metabolites shows a different behavior. At the highest dose, the metabolite Cmax/gm falls to 38% of the maximum. The resveratrol Cmax remains at 74% of the max at the highest dose. This suggests that the liver conjugative enzymes can be swamped, but only at the peak concentrations seen around the Cmax. This "self-swamping" by resveratrol occurs only briefly so it doesn't have much effect on the AUCs.

What advice can we draw from this regarding resveratrol dosing for maximum efficiency? From this data, I would say that for a small daily dose of less than ~1.5 grams, you should probably take it all at once. For larger doses, it may make sense to split them into two or three daily doses of at least 1 g each. This should maximize the total daily AUC, although the peak levels will be higher if it is all taken at once. This advice will probably change if the resveratrol is formulated in such a way as to improve absorption, e.g. SRT501. In fact, Sirtris has decided on the basis of their human trials that they can get by with once a day dosing. I suspect that the 501 formulation is dealing with the second factor that results in apparent absorption inhibition with plain resveratrol. I'm going to cross my fingers and hope that our EtOH/PEG|lecithin scheme is good enough, and continue to take my ~2.5 g in one dose.


More PK fun...
Glucuronide 1
AUCinf (ng h/mL) 1919 3059 5664 9923
AUC/dose (ng h /mL g) 3838 3059 2266 1985
Cmax (ng/mL) 404.6 473.6 874.4 1285
Cmax/dose (ng/mL g) 809 473.6 349.8 257

Glucuronide 2
AUCinf (ng h/mL) 1287 2589 4320 8546
AUC/dose (ng h /mL g) 2574 2589 1728 1709
Cmax (ng/mL) 369.5 672.6 1626 1735
Cmax/dose (ng/mL g) 739 672.6 650 347

3-Sulfate
AUCinf (ng h/mL) 4049 10053 16984 30898
AUCinf/dose (ng h/mL g) 8098 10053 6794 6180
Cmax (ng/mL) 1135 2102 2786 4294
Cmax/dose (ng/mL g) 2270 2102 1114 859

[ilanso]

Zawy, thanks for all your effort in getting this data!
....
Parameter 0.5 g, 1 g, 2.5 g, 5 g
Resveratrol
AUCinf (ng h/mL) 223.7 544.8 786.5 1319
AUC/gm (ng h /mL g) 447.4 544.8 314.6 264
Cmax (ng/mL) 72.6 117.0 268.0 538.8
Cmax/gm (ng/mL g) 145.2 117.0 107.2 107.8

Which means the best concentration peak we can hope for from 5g is just 2.36 mM (Rsv weighs 228 g / mol, so 1 mM is 228 ng/mL)?
But we get more bang for the buck (literally) from a 1/2g dose?

(mM=micromolar, s/b uM)

Edited by ilanso, 22 January 2008 - 06:32 PM.

[zawy]

Parameter 0.5 g, 1 g, 2.5 g, 5 g
Resveratrol
AUCinf (ng h/mL) 223.7 544.8 786.5 1319
AUC/gm (ng h /mL g) 447.4 544.8 314.6 264
Cmax (ng/mL) 72.6 117.0 268.0 538.8
Cmax/gm (ng/mL g) 145.2 117.0 107.2 107.8

Which means the best concentration peak we can hope for from 5g is just 2.36 mM (Rsv weighs 228 g / mol, so 1 mM is 228 ng/mL)?
But we get more bang for the buck (literally) from a 1/2g dose?

Yes, but for AUC, 1 g dose was more efficient, although the 0.5 g dose was just one sample point (probably has error). And don't forget the glucuronide forms may be just as important.

Attached Files

•  Clipboard01.jpg   15.16KB   14 downloads

[bixbyte]

An employee who works in the family business is dying of cancer, I considered telling him about resveratrol at a Christmas party my parents had for the management and he and I were both there, but I decided maybe it wasn't such a good idea because for all I know it could hasten his demise. So I feel guilty for not telling him and I would feel guilty if I did tell him and he died shortly afterward. I didn't get much of a chance to bring it up to him anyway, I mostly ate the food and listened to other people chit chat. Would you go as far as to recommend this stuff to someone who has cancer? If so, how would you bring it up to them?

Here is a medication that helps certain cancers:

http://www.thedcasit...rumID10/84.html

http://www.tiopoietine.info/Eng3.htm

Give them the link and be supportive.

[bixbyte]

Post #56 is this the RES AUC?


Thank YOU

Edited by bixbyte, 05 February 2008 - 07:45 AM.

[malbecman]

Wow, you guys have taken this to a whole new level, much beyond my basic understanding of pharmocokinetics. Thanks. Any chance we have enough animal and human data to do a PBPK model?

[health_nutty]

You guys are awesome!