Download Fitness Tools

Should you freak out about a small increase in body weight in a small-scale rodent study that is attributed to the consumption of saccharin in yogurt?

While epidemiological studies show that the consumption of products containing non-nutritive sweeteners (NNS) is associated with increased adiposity (Colditz. 1990; Fowler. 2008), type 2 diabetes mellitus (T2DM), metabolic syndrome and cardiovascular disease (Dhingra. 2007; Lutsey, Steffen. 2008). A mechanistic link between aspartame, sucralose, stevia & co and weight gain as well as its ill metabolic and cardiovascular consequences in humans is non-existent (learn more).

Rather than weight increases controlled human studies show that the consumption of artificially sweetened foods promote, not hinder the loss of body fat (Sørensen. 2014).
In animal models, though, the results have been more conflicting. While many studies show no effect of artificial sweetener consumption, the latest stud by Kelly Carraro Foletto and colleagues is not the first rodent study to suggest that non-nutritive sweeteners may also interfere in the regulation of compensatory appetite promoting weight gain (Davidson. 2011; Polyák. 2010; Rogers. 1988). This does yet not refute the findings of one of the latest meta-analysis of the effects of low-energy sweetener consumption on energy intake and body weight in man - a meta-analysis published in Nature's prestigious International Journal of Obesity that says...

Figure 1: The forest plots of the practically most relevant data of individual and combined effect sizes for sustained intervention studies comparing the effects on body weight of sweeteners versus sugar (upper panel) and versus water (lower panel) shows that not a single long(er) term study found negative effects - the exact opposite is the case. Even compared to water the use of low-energy sweeteners (artificial or not) lead to measurable, yet not always significant decreases in body weight in human trials (Rogers. 2015).

"that the balance of evidence indicates that use of LES [low or no energy sweeteners] in place of sugar, in children and adults, leads to reduced EI and BW, and possibly also when compared with water" (Rogers. 2015 | my emphasis).

And with respect to the often-cited "evidence" from animal and observational studies, the autors of the meta-analysis submit that...

"[...] the present review of a large and systematically identified body of evidence from human intervention studies, with varying designs, settings and populations (including children and adults, males and females, and lean, overweight and obese groups), provide no support for that view. The question then is whether those hypotheses should be rejected or whether, as seems unlikely, the relevant human intervention studies are consistently flawed in a way that leads, in most cases, to exactly the opposite outcome" (Rogers. 2015)

I do thus want to warn you: Do not overrate the already relatively small amount of extra-weight the rodents in saccharin group of Foletto's recent study gained (see Figure 2, left).

Figure 2: Cumulative weight gain and total cumulative energy intake of (only) 16 male Wistar reds fed diets that were supplemented with either saccharin-sweetened or non-sweetened yogurt added (Foletto. 2015)

In a previous study, Folleto et al. had already observed that saccharin can induce weight gain when compared with sucrose in Wistar rats despite similar total caloric intake. In their latest study they did not try to prove that this effect is independent of the rodents' energy intake and mediated by insulin-resistance and / or modified levels of leptin and PYY in the fasting state.
To this ends, the researchers randomly assigned 16 male Wistar rats to receive ~78kcal per week from either saccharin-sweetened (0.3% saccharin) yogurt or non- sweetened yogurt (0.5 kcal/g) in addition to chow (2.93 kcal/g) and water ad lib. For 14 weeks, Foletto, et al. measured the total food intake (from yogurt and chow) daily and the weight gain on a weekly basis (the results are plotted in Figure 2). Fasting leptin, glucose, insulin, PYY and HOMA-IR levels were measured only at the end of the 14-week study period, though.

Table 1: In view of the fact that any existing negative effects of dietary sweeteners may well be compound-specific. It is certainly worth noting that saccharin is no longer used in modern sweetener formulations of sodas (Wikipedia. 2015)

In spite of the already reported ~5% increase in cumulative weight gain over 14 weeks (p=0.027), the researchers found no differences in HOMA-IR (=insulin resistance), fasting leptin or PYY levels between groups that could mechanistically explain why the rodents who received saccharin sweetened yogurt gained more weight than their peers who received non-sweetened yogurts.
References:

• Foletto, Kelly Carraro, et al. "Sweet taste of saccharin induces weight gain without increasing caloric intake, not related to insulin-resistance in Wistar rats." Appetite (2015).
• Rogers, P. J., et al. "Does low-energy sweetener consumption affect energy intake and body weight? A systematic review, including meta-analyses, of the evidence from human and animal studies." International Journal of Obesity (2015).
• Sørensen, Lone B., et al. "Sucrose compared with artificial sweeteners: a clinical intervention study of effects on energy intake, appetite, and energy expenditure after 10 wk of supplementation in overweight subjects." The American journal of clinical nutrition (2014): ajcn-081554.

If you drink "old school" protein shakes with milk and eggs, you don't have to worry if casein or whey is better on hypercaloric diets ;-)

Creatine, whey and - at least optionally - casein are on the list of "must have" supplements for anyone who has been following a solid workout routine a decent high(er) protein diet for a year or longer. And even though this makes any news study on one of the three potentially interesting for a huge part of the waredownloadsoft readers, the results are often so unsurprising that they are hardly worth an individual waredownloadsoft article...

I mean, let's be honest: You wouldn't be excited if I wrote about study #1023 showing a practically irrelevant acute increase in protein synthesis, let alone totally meaningless increases in mTOR in response to the ingestion of whey protein, would you?

For today's installment of the short news, I will picked two studies that are at least not as simplistic (and useless) as the previously mentioned 1023rd whey protein study.
Study #1 sheds some light onto the intricate differences between the way(s) whey and casein may differently affect our energy balance, gut hormones, glucose metabolism, and taste preference (Pezeshki. 2015), while study #2 is actually a meta-analysis that makes a first attempt to actually quantify the average effect of creatine on leg strength (Lanhers. 2015).

• Whey or casein? The answer to this question probably depends on your goals - That's at least what the complex results of a similarly complex series of experiments from the Snyder Institute for Chronic Diseases at the University of Calgary appears to suggest (Pezeshki. 2015). In the corresponding study, the researchers determined the effects of dietary whey, casein, and a combination of the 2 on energy balance, hormones, glucose metabolism, and taste preference in rats.
  
  Yes, this does mean that we must not uncritically transfer the results to humans. It does not mean, however, that we cannot use the results of Pezeshki's study, at all. Much of what we know and have later proven in human studies about our metabolism has initially been observed in rodents. Doing the same experiments with humans would not just have been significantly more expensive, it would also have been thwarted by misreported energy intakes and a lack of dietary control. So, if you belong to the "it's a rodent study, so I ignore it, if the results don't fit my world-view" faction, you are free to fast forward to the Lanhers study. If not, here's the elevator pitch on the two experiments the scientists conducted:
  • The hormone / energy expenditure experiment: Obesity Prone CD (OP-CD) rats were fed a high-fat control diet (33% fat energy) for 8 wk, and then randomly assigned to 4 isocaloric dietary treatments (n = 12/group): the control treatment (CO; 14% protein energy from egg white), the whey treatment (WH; 26% whey + 14% egg white), the casein treatment (CA; 26% casein + 14% egg white), or the whey plus casein treatment (WHCA; 13% whey + 13% casein + 14% egg white) for 28 d. Measurements included food intake, energy expenditure, body composition, metabolic hormones, glucose tolerance and key tissue markers of glucose and energy metabolism.
  • The food intake / preference experiment: Naive Obesity Prone CD (OP-CD) rats (not the same rats as in experiment 1) were randomly assigned to 3 groups (n = 8/group). During an 8 d conditioning period, each group received on alternate days either the CO or WH, CO or CA, or CO or WHCA. Subsequently, preferences for the test diets were assessed on 2 consecutive days with food intake measurements at regular intervals.
  If you take a look at my compilation of the most relevant data in Figure 1, you will see significant improvements in glucose management in all groups (Figure 1 | G). You should yet also be able to see that the most significant increase in GLUT-4 expression and thus the muscles ability to take up glucose occurred in the casein and mixed protein groups. In the whey group, on the other hand, where the insulin levels were the highest the increase in GLUT4 receptor expression was negligible and the 60 minutes post value for glucose after the glucose tolerance est was higher, but not significantly higher than in either the mixed protein or casein group.
  

  Figure 1: 0-60 minute hormone, glucose, cytokine and gene response to glucose tolerance test in diet-induced obese rodents fed diets containing either low amounts of protein or high amounts from whey, casein or a combination of whey + casein (Pezeshki. 2015).

  Since you will usually hear that whey is the #1 glucose repartitioner this may seem odd. In view of the fact that insulin can only drive glucose into the cells if they actually "stretch out their glucose "suckers", i.e. the GLUT-4 receptors, it is yet only logical - and that despite the fact that we cannot (yet) explain the differential GLUT-4 response. 

• What remains to be seen, though, is whether this measurable, but not in all details statistically significant difference actually matters. If you look at the food intake and preference data in Figure 3. It would seem as if it didn't. I mean, if we apply "Taubes'ian" logic, here, the high(er insulin levels in the whey group should have made the rats overeat...
  

  Figure 2: Daily food intake during conditioning trials. Rats were fed, on alternate days, a control diet or whey (A), casein (B), or a combination of whey and casein (C). Mean hourly food intakes on 2 consecutive days of preference testing, with access to both control and one of the high protein diets (D, E, F | Pzeshki. 2015).

  ... a brief glimpse at the data in Figure 2, however, tells you that this was clearly not the case. It is thus not surprising that the final lean and fat mass percentages in Figure 3 did not differ significantly between the groups, even though some of you will notice a non-significant lean vs. fat mass advantage for the casein group. 
• Eventually, it would still appear unwarranted to use the study at hand to argue in favor of casein instead of whey protein as your protein supplement of choice. If you are yet already following my suggestion to mix the two (read even more), the results of the study at hand would be another reason to stick to that practice.
  
  Overall, we do yet have to remember that significant treatment effects in response to the increase in protein were observed in all groups. In that, it may be true that the casein group had the highest reduction in food intake (30-43% reduction) and appears to have ended up with a slightly better body composition, but hey, we are talking about non-significant differences in a study that was not just done on rodents, but on obese rodents ... and you know that the weight status obese vs. lean / athletic can make more of a difference in the way people / lab animals react to supplements than the difference between man and mouse.
• Creatine and lower limb strength - what to expect? You are probably asking yourself now: "Why on earth is Adel so interested in this meta-analysis?" Well, the answer is as simple as it is personal: The only body part of mine that appears to to have ever really benefited from creatine supplementation, strength-wise, are my legs. Accordingly, I was attracted to the Lanhers' meta-analysis, because I hoped to find some clues on why this may be the case.
  
  Now, I guess I don't give away too much, when I say that the paper didn't provide me with new insights in this respect. What the scientists' compilation and meta-analysis of the data did do, however, was to confirmation that I am not the only one whose "wheels" respond extremely favorably to creatine supplementation: One of the main results of Lanhers' meta-analysis of 60 (!) studies is after all that the effect of creatine supplementation on leg strength is not just statistically significant in almost every study they reviewed, it is also and maybe even more importantly and astonishingly independent of population characteristics, training protocols, and supplementary doses and duration (reread this - you don't have to superdose it!).
  
  In that, I find it particularly striking that even training for mostly aerobic team sports, like handball, for example, can trigger significant increases in squat performance if the players are consuming 3-5g of creatine before or after their training sessions (Note: In the corresponding study by Izquierdo, the handball players also saw significant increases in bench press performance - It's thus by no means as if creatine was a "leg strength booster", only!).
  

  Figure 4: Meta-analysis on maximal weight lifted (kg) in squat. CI confidence interval, M male, SD standard deviation, T0 baseline, T1 following supplementation (Lanhers. 2015).

  For the total workout volume (not shown in Figure 4), Lanhers et al. report almost identical effect sizes as the researchers determined them based on the absolute increase in squat performance in kg (see Figure 4); and just like the relative strength gains, the relative volume increases did not depend on the training status or age (subjects were 21-69 years old) of the subjects.
  
  In view of these results, there's really no good reason to doubt the scientists' conclusions that (a) "[c]reatine supplementation is effective in lower limb strength performance" and (b) works "for exercise with a duration of less than 3 min". If you also take into account that the average effect size for weight and volume are in the range of ~30% irrespective of the population characteristic, training protocols, and supplementary doses and duration, I hope that you can agree that there's also no reason to doubt my repeatedly voiced recommendation to "take your creatine bros" ... ah, and take it with sodium bicarbonate, obviously ;-)

• Izquierdo, M. I. K. E. L., et al. "Effects of creatine supplementation on muscle power, endurance, and sprint performance." Medicine and Science in Sports and Exercise 34.2 (2002): 332-343.
• Lanhers, Charlotte, et al. "Creatine Supplementation and Lower Limb Strength Performance: A Systematic Review and Meta-Analyses." Sports Medicine (2015): 1-10.
• Pezeshki, et al. "Dietary Whey and Casein Differentially Affect Energy Balance, Gut Hormones, Glucose Metabolism, and Taste Preference in Diet-Induced Obese Rats." The Journal of Nutrition. First published ahead of print August 26, 2015.

This is what it's all about. Real food does not come in form of "macros". It comes in form of complex food matrices that determine its effect on one's health - including one's glycemic health.

You will probably remember that I have touched on a specific aspect of the effects of and interactions between different macronutrients in what scientists often refer to as "food matrices" on the glycemic response to standardized glucose loads in previous articles like the famous "True or False?" article that dealt with the question: "Will Adding Fat to A Carby Meal Lower the Insulin Response?" (read it).

You don't remember this or any of the other articles? Well, in that case, I probably have to tell you again that the mere fact that the postprandial glucose are lower does not mean that a certain food or combination of certain macronutrients would increase your insulin sensitivity (adding fat to a high carbohydrate meal certainly doesn't do that, believe me).
While this is possible for certain supplements like berberine and other AMPK activators, it is mandatory that we differentiate the following three cases:

1. Reduced postprandial blood glucose levels in response to an increase in insulin sensitivity as it is triggered by exercise or AMPK activators,
2. Reduced postprandial blood glucose levels in response to increased insulin levels as they occur with the co-ingestion with whey protein and
3. Reduced postprandial blood glucose levels in response to increased insulin levels and a decreased rate of absorption of glucose as it is the case if you add fat to carbohydrates.

That was too fast? Too complicated? Or both? Never mind. The discussion of two recent articles from the Lund University and the Yong Loo Lin School of Medicine (see blue box) will hopefully help you understand the difference - I promise ;-)

Let's take a look at study design, results and implications

In their study, Wathik Alsalim and his colleagues from the Lund University in Sweden and the Consiglio Nazionale delle Ricerche in Italy investigated the integrative impact of macronutrients on postprandial glycemia, β-cell function, glucagon and incretin hormones in man. The subjects were male and female Caucasian subjects, aged 30-70 years and BMI 20-35 kg/m², without diabetes (normal fasting glucose and normal HbA1c) or with T2D without any pharmacological glucose-lowering therapy and HbA1c <60 mmol/mol (<7.8%).

"Exclusion criteria were liver disease, diabetic nephropathy, proliferative diabetic retinopathy, pregnancy or breast feeding, treatment with oral antidiabetic or insulin, previous myocardial infarction, coronary heart disease or angina pectoris, previous surgery on the gastrointestinal tract, larger surgical intervention the last 12 weeks or treatment with oral steroids or thiazide diuretics" (Alsalim. 2015).

Participants were studied at the Lund University's Clinical Research Center on four occasions in a randomized cross-over design, separated by at least four and maximally eight weeks. After overnight-fast (no food after 10pm), subjects were provided with antecubital vein catheter. After two baseline samples at 5, and 2 mins, they ingested in randomized order either one of the macronutrients alone

• glucose - 330kcal = 83g; Skåne University Hospital Pharmacy, Lund, Sweden,
• protein mixture - 110kcal = 30g; ISO WHEY protein consisting of milk and egg protein
• fat emulsion - 110kcal = 24ml; 50% long-chain triglycerides and 50% water;

The response to the individual macros was then compared to the ingestion of a 550kcal meal containing 330kcal (60%) from glucose, 110kcal (20%) from protein and 110kcal (20%) fat - a proportion of which the scientists say that it "was selected to represent a common meal with 60% carbohydrate, 20% protein and 20% fat" (Alsalim. 2015).

Figure 1: To allow for maximal control, the "meal" (macro composition on the right) was a shake (Alsalim. 2015).

Water was ingested at the same time of each load to standardize the ingested volume to 400ml; all ingestions were consumed within 5 min. Blood samples were taken throughout a 300min period after each challenge test.
Now, guess what? Yes, yes, and no ... the glucose meal was not the most insulinogenic one. Accordingly, we have to be careful not to mistake the scientists' conclusion that "[a]dding protein and fat macronutrients to glucose in a mixed meal diminishes glucose" (Alsalim. 2015) as evidence that fat and protein will reduce the potential obesogenic effect of carbohydrates. After all, the data in Figure 2 leaves no doubt that we are not talking about effect (a) from our list at the beginning of the article, i.e. a reduction in glucose excursions in response to "an increase in insulin sensitivity as it is triggered by exercise" (see introduction).

Figure 2: Glucose and insulin response as well as corresponding insulinogenic index and insulin clearance of the individual macros and the mixed meal - all data expressed relative to the values of 110kcal of glucose (Alsalim. 2015).

Rather than that, the reduced glucose AUC was a result of (c), i.e. an increase in insulin levels due to increased insulin production, as well as a decreased clearance of insulin and a decreased rate of absorption of glucose (-16%, -52%, -59% and -70% reduced glucose influx through the portal vein at 30, 45, 60 and 90 min after the meal) - and with the increased insulin levels you will also run an increased "risk" of fat storage (whether you gain weight will still depend on your overall energy intake not your insulin levels, though).

Figure 3: Insulin is not always the bad guy. A 2004 study by Hallschmid, et al. shows that it's centrally mediated effects in the brain can actually help men (but not women) lose weight effortlessly, when they administer it cyclically (only chronically elevated insulin levels are truly problematic) intranasally.

At this point, it may be necessary to point out that the increase in insulin in response to the co-ingestion of protein and fat does not imply that guzzling glucose on its own would be healthier than eating a mixed meal. In conjunction with the increases in the important (fat burning) satiety hormone GLP1, which were observed only in the diabetic subjects, though, these changes are yet far more beneficial for the average type II diabetic than they are for the lean athlete, who doesn't need the extra-insulin to maintain normal blood glucose levels and may even consider them an obstacle on his / her way to single-digit body fat levels.
References:

• Alsalim et al. "Mixed meal diminishes glucose excursion compared to glucose by several adaptive mechanisms in man." Diabetes, Obesity and Metabolism (2015): Accepted article.
• Collier G, O'Dea K. The effect of coingestion of fat on the glucose, insulin, and gastric inhibitory polypeptide responses to carbohydrate and protein. Am J Clin Nutr. 1983 Jun;37(6):941-4.
• Hallschmid, Manfred, et al. "Intranasal insulin reduces body fat in men but not in women." Diabetes 53.11 (2004): 3024-3029.
• Kaur B, Ranawana V, Teh AL, Henry CJ. "The Glycemic Potential of White and Red Rice Affected by Oil Type and Time of Addition." J Food Sci. (2015).
• Robertson, R. Paul, et al. "Glucose toxicity in β-cells: type 2 diabetes, good radicals gone bad, and the glutathione connection." Diabetes 52.3 (2003): 581-587.
• Steen, Eric, et al. "Impaired insulin and insulin-like growth factor expression and signaling mechanisms in Alzheimer's disease-is this type 3 diabetes?." Journal of Alzheimer's disease 7.1 (2005): 63-80.

No acute changes in blood glucose, but extreme changes in insulin levels. How can this trigger a reduction in glycated albumin - How's that possible?

You've read about gamma aminobutyric acid (GABA) at the waredownloadsoft before. While most people think of it mainly as a calming agent, though, waredownloadsoft readers know that it has the remarkable ability to heal the insulin producing β-cell in rodents by stimulating their replication, protecting them against apoptosis, and attenuating insulitis (Soltani. 2011; Tian. 2013; Prud'homme. 2014; Purwana. 2014). And while these favorable effects were first observed in mice, researchers are quite sure that they are valid in humans, too- that's also because said effects have been confirmed only recently by Tian et al. (2014) and Purwana et al. (2014) in vitro as well as in xenotransplanted human islets.
If you know that, the observations researchers from the Fudan University in Shanghai report in their latest study probably won't come as a surprise. In their investigation into the  pharmacokinetics and pharmacodynamics of GABA in healthy volunteers, the researchers found that the chronic provision of 2g of GABA three times per day triggers a highly significant decrease in glycated albumin (GA) levels, the intermediate measure of blood glucose control (in-between acute blood glucose measurements and HbA1c | Roohk. 2008), within only 7 days.

Figure 1: Acute glucose (top) and insulin (bottom) response after single and repeated administration of 2g of GABA (left). Chronic effects of repeated dose-administration of GABA on glycated albumin (right | Li. 2015).

If you scrutinize the data in Figure 1, though, there are a few questions we still need to find answers to. Obviously, the acute administration of GABA (Figure 1, left) lead to significant increases in insulin - both, under either fasting (1.6-fold, single dose; 2.0-fold, repeated dose; p < 0.01) or fed conditions (1.4-fold, single dose; 1.6-fold, repeated dose; p < 0.01).
Usually, this 1.4-fold or 1.6-fold increase in insulin should trigger a significant (at least transient) decrease in blood glucose. Since the latter wasn't the case in either the fasted or the fed tests the scientists conducted on the twelve subjects, who participated in the open-labeled, three-period trial, it appears more than counter-intuitive that the chronic administration of GABA which does not accumulate in the body and was found to be almost completely absorbed in 60 minutes and to have a half-life of 5h still lead to an GA decrease of approximately 11-12%.

Now this obviously confirms that GABA, due to its ability to increase islet hormonal secretion, has potential therapeutic benefits for diabetes, what the study does not tell us, though, is whether the lack of immediate effects on blood glucose levels can, as the scientists suspect, "in part be attributed to GABA-induced counter regulatory mechanisms, especially elevated glucagon" (Li. 2015) which rose so that the insulin-to-glucagon ratio remained unchanged. Yet while the latter could explain why the subjects did not become hypoglycemic in the face of increased insulin levels, the lack of certainty with respect to the underlying mechanisms makes the study results difficult to interpret.
References:

• Li, Junfeng, et al. "Study of GABA in Healthy Volunteers: Pharmacokinetics and Pharmacodynamics." Frontiers in Pharmacology 6 (2015): 260.
• Prud’homme, Gérald J., et al. "GABA protects human islet cells against the deleterious effects of immunosuppressive drugs and exerts immunoinhibitory effects alone." Transplantation 96.7 (2013): 616-623.
• Prud’homme, Gérald J., et al. "GABA protects pancreatic beta cells against apoptosis by increasing SIRT1 expression and activity." Biochemical and biophysical research communications 452.3 (2014): 649-654.
• Purwana, Indri, et al. "GABA promotes human β-cell proliferation and modulates glucose homeostasis." Diabetes 63.12 (2014): 4197-4205.
• Roohk, H. Vernon, and Asad R. Zaidi. "A review of glycated albumin as an intermediate glycation index for controlling diabetes." Journal of diabetes science and technology 2.6 (2008): 1114-1121.
• Soltani, Nepton, et al. "GABA exerts protective and regenerative effects on islet beta cells and reverses diabetes." Proceedings of the National Academy of Sciences 108.28 (2011): 11692-11697.
• Tian, Jide, et al. "γ-Aminobutyric acid inhibits T cell autoimmunity and the development of inflammatory responses in a mouse type 1 diabetes model." The Journal of Immunology 173.8 (2004): 5298-5304.
• Tian, Jide, et al. "Oral GABA treatment downregulates inflammatory responses in a mouse model of rheumatoid arthritis." Autoimmunity 44.6 (2011): 465-470.
• Tian, Jide, et al. "γ-Aminobutyric acid regulates both the survival and replication of human β-cells." diabetes 62.11 (2013): 3760-3765.
• Tian, Jide, et al. "Combined Therapy With GABA and Proinsulin/Alum Acts Synergistically to Restore Long-term Normoglycemia by Modulating T-Cell Autoimmunity and Promoting β-Cell Replication in Newly Diabetic NOD Mice." Diabetes 63.9 (2014): 3128-3134.