Artificial Sweeteners: A Wolf in Sheep’s Clothing?David A. Johnson, MD
March 22, 2016Not So Sweet?
Hello. I’m Dr David Johnson, professor of medicine and chief of gastroenterology at Eastern Virginia Medical School in Norfolk, Virginia.
I recently attended a conference at the Weizmann Institute of Science in Rehovot, Israel, and wanted to highlight some particularly fascinating information about noncaloric artificial sweeteners.
In use for over a century now, there are six commercially available artificial sweeteners in the United States. They are commonly used in a variety of foods, processed sodas, and other products that we routinely see on the shelf and ingest perhaps every day. Artificial sweeteners are frequently recommended to patients with prediabetes, diabetes, and obesity because they are noncaloric and don’t have an absorbable sugar.
Are they really good for us? The data on glycemic control in patients with diabetes or obesity are quite mixed. What I want to share with you today is why some of these artificial sweeteners are really not good and, in fact, may be bad for you.
A Fascinating Series of Experiments
Let’s look at a fascinating series of experiments from the Weizmann Institute of Science.
First, they took mice and gave them three artificial sweeteners: saccharin, sucralose (which is Splenda®), and aspartame. (These are the three most commonly used artificial sweeteners, but there are also three others.) They compared mice that were fed these sweeteners with mice that had routine chow feeds and glucose or sucrose, which are both absorbable sugars. At the end of the 4-week experiment, there was a profound effect on the glycemic control of the mice that were being fed the noncaloric artificial sweeteners; they had significant dysregulation of their glycemic control. Saccharin had the most pronounced effect, so they did another experiment using a lower dose of saccharin, which showed the same type of glycemic worsening in the mice.
Was this effect related to the microbiome? We know that sugars are potentially fermentable; noncaloric sugars can be presented to the gastrointestinal tract, and the microbiome can take advantage of them. The microbiome can convert them to things that may be fermented and then upregulate certain pathways by themselves, through degradation products, or they may have a prebiotic effect. In a prebiotic effect, the ingested substance may actually be toxic due to the metabolic waste products that the bacteria generate and may knock off some other bacteria or may favor some of the bad bacteria, creating what we call intestinal dysbiosis.
In the artificial sweetener model, the investigators at the Weizmann Institute of Science took the mice with artificial sweetener-induced glycemic intolerance and gave them antibiotics for 4 weeks. They were able to show that the antibiotics reversed this effect on glycemic intolerance. So, antibiotics directed at bacteria would actually correct glycemic control to normal. This suggests that the microflora were part and parcel to this, if not the crux of the problem.
They took this a step further and looked at some of the metabolic consequences, including what we call metagenomics, where pathways are upregulated by some of these bacterial changes. What they found was that there is a host of pathway upregulations. One pathway upregulation, in particular, among the saccharin-fed mice was an increase in the glycan degradation pathway. This is a pathway that has been strongly associated—not only in mice but also in humans—with diabetes and obesity. In addition to the glycan degradation pathway, other pathways in the saccharin-fed mice were upregulated, such as pathways that enhanced starch, sucrose, fructose, and mannose metabolism and folate, glycerolipid, and fatty acid biosynthesis. In contrast, the mice that didn’t receive the saccharin didn’t have these effects. These metagenomic upregulations are all well recognized as pathways that are expressed and enhanced in diabetes and obesity. What they demonstrated when they looked at the microbiome and the bacteria genomic profiling were increases in bacteria belonging to the Bacteroides genus and Clostridiales order and decreases in Lactobacillus reuteri among the saccharin-fed mice. Again, we can see that not only is there a dysbiosis, but there is a pattern that is reproducible across these studies.
Findings Relevant in Humans Too
Next, they tested these findings in humans, using a database with nutritional profiling in a large number of patients, with ongoing data collection. They identified 381 nondiabetics in their database, with about 44% males. They looked at associations with glycemic control and ingestion of noncaloric artificial sweeteners. They had a very dynamic way to look at dietary recall with a validated dietary history questionnaire.
When they looked at this and corrected for exposure to noncaloric artificial sweeteners, there were increases in things that you would expect from the mouse model. There were increases in hemoglobin A1c, more prediabetes, reduced glucose tolerance, impaired fasting glucose, and increased body weight and waist-to-hip ratios. So, there was this central obesity pattern seen in metabolic syndrome. These changes were all related to this exposure to noncaloric artificial sweeteners, and there seemed to be a dose-related effect. In other words, those people who used more of these noncaloric artificial sweeteners had even more pronounced effects.
Not to be stopped there, they took it a step further and looked at seven healthy, lean participants and fed them the US Food and Drug Administration acceptable daily intake of saccharin. They looked at glycemic effect as a response only to the saccharin ingestion, with standardized meals. Lo and behold, 4 out of the 7 participants actually developed impaired glucose tolerance, and their glycemic response relative to what they were at baseline was strikingly aberrant. In fact, the P value was <.001 as it relates to their change from baseline. So, 4 out of the 7 participants had profound glucose intolerance.
They also looked at the response in the microbiome, and they found that there were profound changes in these 4 participants. The 3 participants who didn’t have glucose intolerance also had a microbiome analysis, but it didn’t seem to differ from their baseline analysis. So something in these 4 participants really seemed to change. When these patients resumed their normal diet and stopped their saccharin intake, they returned to having normal glucose tolerance.
They then took the stool from these 7 participants, and they transposed this into mice. The germ-free mice, if they got the stool from the 4 participants with glucose intolerance, developed glucose intolerance as well. Therefore, this was a transmissible phenomenon using a fecal transplant.
A Profound Effect
Artificial sweeteners are frequently used around the world to try to decrease glucose exposure, increase glycemic control, and decrease the tendency for obesity. In fact, what we are seeing is that these artificial sweeteners actually have a profound effect on the metabolic consequences related to dysbiosis, despite being calorie-free. This dysbiosis drives a number of different pathways that can possibly increase risk of developing diabetes or exacerbation of glycemic control in patients with diabetes, and the same for obesity.
As we strive to try to improve these disease states, we actually may be making them worse. To conclude, buyer beware to patients with diabetes or obesity. These artificial sweeteners certainly may be a part, if not the crux, of the problem, and patients should discuss using these sweeteners with their physician. Physicians who recommend these sweeteners need to take a step back and really re-evaluate their recommendations, especially among their patients with diabetes and obesity. In fact, we may be dealing with a wolf in sheep’s clothing.