Friday, May 31, 2013

Science Round-Up Seconds: The Pro-Insulinogenic Effect of Artificial Sweeteners + Mechanisms & Consequences

Would having your coffee with splenda instead of sugar make this cookie even more hazardous for your glucose metabolism and what about your waistline?
If you've listened to yesterday's installment of the science Round-Up your are probably already in the know of the most important facts about the "pro-insulinogenic" effects of sucralose and how it is (a) neither sure what exactly is causing this increase in post-prandial insulin release, nor (b) whether this is the "bad thing" conventional wisdom would dictate it is.

If you've also read the corresponding press release from the Washington University in St. Louis, I've linked in yesterday's Facebook post on the matter, you will know that even the authors of the study are not yet sure about the real world implications of their results:
"The elevated insulin response could be a good thing, she pointed out, because it shows the person is able to make enough insulin to deal with spiking glucose levels. But it also might be bad because when people routinely secrete more insulin, they can become resistant to its effects, a path that leads to type 2 diabetes."
Before we are getting to those, let's briefly recap what exactly it was, the researchers did and what they observed: M. Yanina Pepino and her colleagues from the Washington University School of Medicine in St. Louis had recruited a group of people belonging to the rare species of obese subjects (BMI 42.3 ± 1.6 kg/m²) who (a) did not use non-nutritive sweeteners and were insulin sensitive. The subjects underwent  a 5-h modified oral glucose tolerance test on two separate occasions which was preceded by consuming either sucralose (experimental condition) or water (control condition) 10 min before the glucose load in a randomized crossover design in the course of which the researchers observed:
  • Figure 1: AUC (normalized for mean) and peak values of measured parameters of glucose metabolism (Pepino. 2013)
    20 ± 8% greater incremental increase in insulin area under the curve (AUC) (P < 0.03),
  • 22 ± 7% greater peak insulin secretion rate (P < 0.02), 
  • 7 ± 4% decrease in insulin clearance (P = 0.04), 
  • 23 ± 20% decrease in the calculated insulin sensitivity
  • all this occurred in the presence of a faster increase in blood glucose (remember this, it's going to be important)
  • aside from the increase in the incremental area under the curve (AUC) for insulin, there were no statistical differences between the AUC (=totally produced) measured serum markers for any of the parameters in figure 2 (top)
Almost as interesting as the things, the researchers did observe were yet changes they didn't observe, namely differences between conditions in active glucagon-like peptide 1, glucose-dependent insulinotropic polypeptide, glucagon incremental AUC, or indices of the sensitivity of the β-cell response to glucose. Why this is important? Well, actually you would expect increased GLP-1 and GIP levels, lowered glucagon (the hormone that will have your liver produce glucose) and an increased sensitivity of the b-cells to glucose (why else would there be more insulin floating around), but none of these effects was observed.

So what's the mechanism here, then?

Just like the scientists themselves state: "Although we found that sucralose affects the glucose and insulin response to glucose ingestion, we don’t know the mechanism responsible." (Pepino in an interview with the press guy from the University) and we don't know if these effects of sucralose / splenda are obesity specific or will occur only transiently and disappear upon continuous exposure.

As I mentioned on the show, there are yet a couple of possible mechanisms, all of which are in some form or another related to the sweet taste receptors, which are activated by all natural and artificial sweeteners and are expressed on our tongues, in our intestines and on the pancreas:
  • Figure 2: Illustration of the signaling cascade that's initiated by the activation or the sweet taste receptor (based on Lindemann. 2001)
    in vitro studies have already shown a C-AMP (regular) dependent increase in insulin release from pancreatic cells by sucralose (Nagakava. 2009); this was yet not observed in previous human studies, where mostly lean subjects ingested pure sucralose
  • the sweet taste receptor appears to have either a heterodimer structure or two distinct incarnations (as shown in the illustration to the right; see figure 2), of which sucralose could trigger only the 2nd pathway which may be insufficient to cause the depolarization of the pancreatic cells to trigger the consequent release of insulin, but it may well suffice to increase the depolarization and insulin release from the pancreas
From another artificial sweetener, namely Acesulfam-K we know that it can also increase the influx of glucose across the intestinal border by increasing GLUT-2 receptor expression (Zheng. 2013), so that this would a third - indirect effect on the insulin response that cannot be totally discarded.

In fact, the upregulation of intestinal glucose transport is the most likely explanation

If we take another look at the study outcomes, it is obvious that this increase in the transport of glucose across the intestinal border would in fact be the most likely mechanism to cause the (in this case appropriate) increase insulin response (more on that in the chapter on whether this is a good or bad thing). Previous studies in rodents also suggest that aside from the acute increase in GLUT-2 receptor expression, the chronic use of artificial sweetener has the ability to upregulate the expression of the "regular" Na-dependent glucose transporter and will thus have a persistent negative effect on what you could call the GI of everything you eat. Accordingly, the scientists speculate
"[...] that regular users of NNS [non-nutritive sweeteners] would have a higher glycemic response after an oral glucose tolerance test on the control day than irregular users and that the acute effects of sucralose intake would be blunted because differences between water and sucralose conditions would be smaller in regular than in irregular users of NNS." (Pepino. 2013)
In other words, chronic users won't be experiencing the effects that were observed in the study, but they will necessarily have a slightly increased insulin response to everything they eat (as long as they are healthy and not yet insulin resistance) irrespective of whether they consume it with or without non-nutritive (not just artificial) sweeteners.

Table 1: Sweetness, dose to stimulate the sweet taste receptor (EC50; based on Matsuda. 2011) and correlation of sweetness and EC-50 value.
In this context, it's certainly worth mentioning that it is unlikely that these effects are sucrolose specific. What is possible, though, is that different artificial sweeteners have more or less pronounced effects. From a 2011 study by Matsuda et al. we know that this is the case for their stimulatory effect on the sweet taste receptor and as my own juxtaposition + calulation of the correlation, which is missing a "minus" sign, in table 1 goes to show you, the "sweetness" measured in units of sucrose (normal sugar) is a relative reliable predictor of the degree to which the different artificial sweeteners activate the sweet taste receptor (not really surprising, is it?)

Being ~300x sweeter than sucrose stevia would by the way be somewhere between acesulfame K and saccharin Na and there is no reason to assume it would not have the same effects, only because it is "natural". I fact the observation Anton et al. made in a 2010 study, where the preingestion of stevia yielded a greater initial glucose spike and a correspondingly higher increase in 30min post-prandial insulin levels than aspartame (~37%) (Anton. 2010). These results clearly suggest that stevia is probably no exception to the rule (it could yet also be that aspartame is an exception to the rule, cf.  "Aspartame's Anti-Insulinogenic Effects During a Workout"; read more) - unfortunately the differences in the study design don't allow for a direct comparison of the Anton and the Pepino study.

And how bad is that?

If we don't really know what the mechanism is and who will be affected to which degree, do we at least know how bad the observed changes are? Unfortunately, the answer is "No", but the notion that any increase in insulin would be bad for you is clearly flawed.

As I hinted at in the show, one of the characteristic feature that renders Pima Indians susceptible to diabetes is the absence of an appropriate early spike in insulin (Lillioja. 1991). Corresponding evidence from other ethnicities (e.g. Kosaka.1996) confirms that the absence of this initial spike (early insulin response) is actually the first step people take on the "Royal Road to Diabesity", the underlying reasons are:
  • a failure of immediate suppression of hepatic glucose production, when exogenous glucose is available
    "[...] the impact on suppression of hepatic glucose production was dramatic, with the liver releasing glucose at a higher rate despite the presence of hyperglycemia and hyperinsulinemia. The alteration was a direct consequence of the specific defect because restoration of first-phase insulin secretion was followed by complete normalization of hepatic glucose production." (Del Prato. 2001)
  • the lack of the stimulatory effects of insulin on peripheral glucose uptake (Del Prato. 2003)
If you did not listen to the podcast yet, you are now probably asking yourselves: "Hold on, but does that not set you up to become obese?" The answer would certainly be yes, if you were insulin resistant, would not work out and would follow a hypercaloric diet with tons of carbs and fats.
Figure 3: Effects of insulin on glucose metabolism, glycogen storage (left) and fatty acid synthesis (right), time-resoled data (0, 2, 4, 8, 12h) on glycogen deposition in skeletal muscle in the inset b/w graph; the data (in mg per kg/min) was measured using an euglycemic clamp in the presence of high, but physiological insulin levels (Koopmans. 1998)
If that's not you, the actual effect of insulin on fatty acid synthesis and thus the amount of carbs that is stored as fat will be of a >20x lower magnitude than the effect insulin exerts on the storage of muscle glycogen - if you extrapolate the data from the first 20min of the euglycemic clamp data figure 3 is based on the ratio could be as high as 2,000x (in other words for each 1 unit of glucose being converted to fat, 2,000 units will be shuttled into the muscle).

"That's all bullocks? Insulin is bad *fullstop*"

If the above is what you still believe I may remind you that insulin is the natural solution to the "sugar problem", only when it seizes working trouble ensues. Moroever, I suppose some of you will be supplementing with pro-insulinogenic agents such as:
  • Figure 4: Data from 12 normal subjects using 5g or 10g of oral GABA (Cavagnini. 1982)
  • taurine
  • GABA
  • whey
  • EAAs
  • etc.
I guess it would be easy to extend the list of agents that will increase the insulin response and have still been shown to have beneficial effects on diabetes risk and body composition, but I am to lazy to do this now ;-)

"Where's my Splenda I want more Splenda!"

Before you order a 20kg batch of sucralose from China, now, I do yet still want to remind you of the fact that you can also find arguments in favor of the potential detrimental effects this hitherto non-understood +20% increase in insulin response during an OGGT could have:
  • We don’t know about long-term consequences. Will the increase remain or will we “depend” on the artificial sweetener to get an adequate insulin spike in the future (remember the study participants were non-users before)? 
  • In the same vein, we don't know what the consequences of the increased intestinal glucose absorption in response to chronic use  (if it is actually present in humans) will be.
  • Moreover, although previously published studies don't support this, you could develop an even more pronounced sweet tooth and totally mess the self-regulatory mechanism for food intake that is skewed in the "average Westerner", anyways.
  • Importantly, the effects are probably different in the obese + insulin intolerant for them the spike in insulin will have little benefits and tons of downsides (the beneficial effect cited above wrt to the glucose disposal in muscle are probably irrelevant), so that exactly those people who are targeted by artificially sweetened foods could see negative effects which may not be present in lean individuals.
  • And lastly, the general beneficial effects of insulin on glycogen repletion depend on glycogen depletion! In other words, if you don’t work out you will not benefit to the same degree, simply because there is no place to put the glucose (it should be said that this is a general problem, which is not specific to dietary sweeteners, but "having room" for glycogen to be stored is a major determinant of whether insulin is rather good or rather bad) 
And if all that is not convincing enough, just remind yourself that we do not even know what exactly is going on here.

Bottom line: It does therefore not appear to be indicated to change whatever has been working for you in the past. I can guarantee that you are NOT stagnating because you use a sweetened whey protein or BCAA product. The evidence simply is not there or as Renwick et al. put it in their review, there is "no consistent evidence that low-energy sweeteners increase appetite or subsequent food intake, cause insulin release or affect blood pressure in normal subjects." (Renwick. 2010).

Finally, I guess, I don't have to mention this, but still: I will keep you posted on any future research.

  • Anton SD, Martin CK, Han H, Coulon S, Cefalu WT, Geiselman P, Williamson DA. Effects of stevia, aspartame, and sucrose on food intake, satiety, and postprandial glucose and insulin levels. Appetite. 2010 Aug;55(1):37-43.  
  • Del Prato S, Tiengo A. The importance of first-phase insulin secretion: implications for the therapy of type 2 diabetes mellitus. Diabetes Metab Res Rev. 2001 May-Jun;17(3):164-74.  
  • Del Prato S. Loss of early insulin secretion leads to postprandial hyperglycaemia. Diabetologia. 2003 Mar;46 Suppl 1:M2-8. 
  • Koopmans SJ, Mandarino L, DeFronzo RA. Time course of insulin action on tissue-specific intracellular glucose metabolism in normal rats. Am J Physiol. 1998 Apr;274(4 Pt 1):E642-50.
  • Kosaka K, Kuzuya T, Hagura R, Yoshinaga H. Insulin response to oral glucose load is consistently decreased in established non-insulin-dependent diabetes mellitus: the usefulness of decreased early insulin response as a predictor of non-insulin-dependent diabetes mellitus. Diabet Med. 1996 Sep;13(9 Suppl 6):S109-19. 
  • Lillioja S, Nyomba BL, Saad MF, Ferraro R, Castillo C, Bennett PH, Bogardus C. Exaggerated early insulin release and insulin resistance in a diabetes-prone population: a metabolic comparison of Pima Indians and Caucasians. J Clin Endocrinol Metab. 1991 Oct;73(4):866-76. 
  • Masuda K, Koizumi A, Nakajima K, Tanaka T, Abe K, Misaka T, Ishiguro M. Characterization of the modes of binding between human sweet taste receptor and low-molecular-weight sweet compounds. PLoS One.
  • Nakagawa Y, Nagasawa M, Yamada S, Hara A, Mogami H, Nikolaev VO, Lohse MJ, Shigemura N, Ninomiya Y, Kojima I. Sweet taste receptor expressed in pancreatic beta-cells activates the calcium and cyclic AMP signaling systems and stimulates insulin secretion. PLoS One. 2009;4(4):e5106.
  • Pepino MY, Tiemann CD, Patterson BW, Wice BM, Klein S. Sucralose Affects Glycemic and Hormonal Responses to an Oral Glucose Load. Diabetes Care. 2013 Apr 30.
  • Renwick AG, Molinary SV. Sweet-taste receptors, low-energy sweeteners, glucose absorption and insulin release. Br J Nutr. 2010 Nov;104(10):1415-20.
  • Zheng Y, Sarr MG. Effect of the artificial sweetener, acesulfame potassium, a sweet taste receptor agonist, on glucose uptake in small intestinal cell lines. J Gastrointest Surg. 2013 Jan;17(1):153-8.


  1. ....cheers for all the hard work adel....brilliant show.

  2. As I've not found any, are there any sites that exclude discussion of health non necessities?

  3. Thanks for the review and insights. Suspect this may be useful to those on VLC/ketogenic diets, especially if there is difficulty entering/maintaining ketosis. Would be very interesting to know what effect these agents may have on insulin resistant subjects not ingesting high carb loads.

  4. You said on Jaminet's site that you used to use Nutrition Data (.com) until you found an error. I've also found (huge) errors there, so might you rate some sites?