Nonnutritive sweeteners and the gut microbiome Original paper

In adults without diabetes, two weeks of consuming saccharin or sucralose resulted in higher blood glucose responses, whereas consuming stevia or aspartame did not. All four nonnutritive sweeteners altered the gut microbiome. A subsequent mouse study using fecal transplants from the study participants suggested that glycemic responses were related to changes in the gut microbiome.

This Study Summary was published on November 7, 2022.

Background

Despite their approval for use by various governing bodies, nonnutritive sweeteners (NNS) remain controversial among health practitioners and the public. The most common concerns about NNS include their effects on glycemia, as people with diabetes or prediabetes often use them, and their effects on the microbiome. In fact, Study Summaries previously covered an in vitro study assessing the interactions between the gut microbiota and the intestinal epithelium.

The study under review[1] has amplified the concerns surrounding NNS, glycemia, and the gut microbiome. Are the concerns justified?

The study

This open-label randomized controlled trial assessed the effects of short-term NNS consumption on blood glucose levels. As secondary outcomes, the authors assessed the effects of NNS consumption on the gut and oral microbiome. The authors also conducted a mouse study using fecal transplants from the human participants to investigate if changes in glucose tolerance might be causally linked to changes in the microbiome.

A total of 131 adults without diabetes (BMI of 18–28) in Israel who had not consumed NNS in the previous 6 months were recruited, of whom 120 completed the trial and provided sufficient data for analysis. At the beginning of the trial, the participants reported to the study facility, where the study investigators collected blood samples and provided the participants with a continuous glucose monitor (CGM) that they wore throughout the study period. The participants also received an at-home kit to sample their microbiome via stool and oral samples and conduct glucose tolerance tests (GTTs) on prespecified days during the trial.

The rest of the study consisted of three phases:

Phase 1 (days 1–7) consisted of baseline measurements of metabolic, metabolomic, and microbial parameters.

For phase 2 (days 8–21), the participants were randomized to receive one of the following treatments:

  • Aspartame
  • Saccharin
  • Sucralose
  • Stevia
  • Glucose vehicle
  • No sweetener control (NSC)

The participants in the four NNS groups (aspartame, saccharin, sucralose, and stevia) consumed 6 commercial packets of sweetener dissolved in water per day by taking 2 packets in the morning, afternoon, and evening. Each packet contained glucose as a bulking agent, resulting in the intake of approximately 5 grams of glucose per day in each NNS group from the packets. The glucose vehicle group received 6 glucose packets per day to provide the same amount of glucose as the NNS conditions. The NSC group did not receive any sweetener packets. The participants were unblinded to each condition due to “the distinct flavor of each sweetener.” Consumption of each NNS was below its respective acceptable daily intake (ADI; see this summary for more information).

For phase 3 (days 21–28), the participants ceased sweetener intake. Throughout this phase, the investigators assessed additional metabolic, metabolomic, and microbial parameters to determine whether NNS-induced changes persisted after the cessation of NNS intake.

The primary outcome was blood glucose response to the GTTs conducted at home (9 GTTs were conducted per participant throughout the trial). The authors also conducted a coefficient of variance analysis on CGM data to assess variability in daily blood glucose and assessed changes in participants’ gut and oral microbiomes.

For the mouse study, the authors transplanted 7–9-week-old male mice with fecal samples collected either at baseline (day 1) or the last day of phase 2 (day 21) from 42 human participants. The selected participants were the four individuals in each group with the most potent GTT response (top responders) and the three participants in each group with the lowest GTT response (bottom responders). The authors conducted glucose tolerance tests on the mice, assessing glucose levels from the blood of the tail vein immediately before and 15, 30, 60, 90, and 120 minutes after a glucose feeding.

The results

Glycemic outcomes

The average incremental area under the curve (iAUC) for blood glucose following the GTTs during phases 1 and 2 was higher in the saccharin and sucralose groups than in the glucose vehicle or NSC groups. However, there were no differences in iAUC between groups when the authors included GTT data from all three phases. There were no differences in glucose iAUC between aspartame or stevia and either of the control groups.

Compared to baseline, average iAUC in the sucralose and saccharin groups increased during phase 2 before returning to near baseline levels in phase 3. There were no within-group changes in iAUC in any of the other four groups.

Glycemic response to various nonnutritive sweeteners

Variability in blood glucose was higher in the saccharin and stevia groups than in the NSC group. However, the authors noted that the NSC group had lower glucose variability than the glucose vehicle group, and none of the NNS groups differed in glucose variability from the glucose vehicle group.

Plasma insulin (assessed on days 0, 14, and 28) was higher on days 14 and 28 than on day 0 in the glucose vehicle group and higher on day 14 than day 0 in the stevia group, with no other changes throughout the study in any of the study groups.

Gut microbiome outcomes

Throughout the trial, changes in gut microbiome genera and species were significantly different in the sucralose and saccharin groups compared to the NSC group.

All four NNS groups experienced significant differences in microbiome functional parameters compared to the NSC group. Specifically, changes in the sucralose group were related to purine metabolism, changes in the saccharin group included pathways related to glycolysis and glucose degradation (e.g., the production of lactate during anaerobic respiration), changes in the aspartame group related to polyamines metabolism, and changes in the stevia related to fatty acid biosynthesis. No differences occurred between the glucose vehicle and NSC groups in gut microbiome composition or functional parameters.

Changes to gut microbiome functional pathways

Oral microbiome outcomes

In the stevia group, the relative abundance of metabolism-related pathways decreased from baseline to the second week of phase 2. In addition, changes in the relative abundances of six Streptococcus species occurred in the sucralose group, a reduced relative abundance of Fusobacterium occurred in the saccharin group, and a reduced relative abundance of Porphyromonasand Prevotella nanceiensis occurred in the aspartame group. There were no oral microbiome changes in the NSC or glucose vehicle groups.

Correlations between glycemic and microbiome variables

In the sucralose group, baseline abundances of three bacterial species (Eubacterium CAG-252, Dorea longicatena, and Oscillibacter ER4) correlated with GTT-iAUC in the second week of phase 2. In addition, baseline abundances of several purine biosynthesis pathways were positively associated with GTT-iAUC in the second week of phase 2, and baseline abundances of mixed acid fermentation and TCA cycle functional pathways were inversely correlated with GTT-iAUC in the second week of phase 2.

In the saccharin group, baseline levels of Prevotella copri and UMP biosynthesis were positively associated with GTT-iAUC in the second week of phase 2. In contrast, baseline levels of Bacteroides xylanisolvens were negatively correlated with GTT-iAUC in the second week of phase 2. Many of the pathways at baseline that were negatively correlated with GTT-iAUC in the second week of phase 2 were related to glycolysis and glycan degradation.

In the stevia group, baseline abundances of Prevotella stercorea and Prevotella copri were positively associated with GTT-iAUC in the second week of phase 2. Baseline abundances of Bacteroides coprophilus, Parabacteroides goldsteinii, and Lachnospira sp. UBA5912 were also positively associated with GTT-iAUC in the second week of phase 2.

In the aspartame group, baseline abundances of B. fragilis and B. acidifaciens were positively associated with GTT-iAUC in the second week of phase 2, whereas B. coprocola was inversely correlated with GTT-iAUC in the second week of phase 2.

Notably, many of the microbiome parameters at baseline associated with GTT-iAUC in the second week of phase 2 demonstrated changes throughout phase 2 and reverted to baseline after phase 3.

The top responders in the sucralose group had a greater increase in pathways related to glycolysis and the TCA cycle from baseline to the end of phase 2 than the bottom responders. The top responders in the stevia group experienced increases in pathways related to the urea cycle and its metabolites, whereas the bottom responders experienced increases in Akkermansia muciniphila, a species associated with metabolic health. The top responders in the saccharin group experienced increases in degradation of the cyclic amide caprolactam, which the authors noted suggests a possible potential degradation of saccharin. Biosynthesis of isoleucine also increased in top saccharin responders.

Rodent study

The GTT-AUC in the rodents that received fecal transplants from the top human responders at the end of phase 2 in each of the four NNS groups was greater than the GTT-AUC in the rodents that received fecal transplants from the same donors at baseline. In contrast, no difference was observed in the GTT-AUC between rodents who received fecal transplants from the top human responders at the end of phase 2 in the glucose vehicle or NSC groups compared to rodents that received fecal transplants from the same donors at baseline. In addition, rodents who received fecal transplants from the bottom responders in the saccharin group at the end of phase 2 had a higher glycemic response than the rodents who received fecal transplants from the same donors at baseline. In contrast, no differences in average glycemic response were observed for any of the other five groups when comparing rodents who received fecal transplants from bottom responders at the end of phase 2 and those that received baseline fecal transplants from the same donors.

Note

On the surface, the results of this study suggest the following:

  • Sucralose and saccharin intake increased the glycemic response to carbohydrates.
  • Sucralose and saccharin intake altered gut microbiome composition, and all four NNS alter gut microbiome functional pathways and oral microbiome composition.
  • Gut microbiome composition and functional parameters at baseline were associated with glycemic response after two weeks of NNS consumption.
  • Mice that received fecal transplants from participants in the NNS groups after two weeks of NNS consumption experienced a higher glycemic response to carbohydrates than mice that received fecal transplants from the same participants at baseline.

However, several limitations to the current study’s findings also need to be considered:

  • The human GTTs were conducted at home, without researcher supervision.
  • The participants were not blinded to the intervention.
  • The participants did not consume NNS regularly. In fact, the authors had to screen 1,375 people to enroll the sample and noted that most ineligible individuals regularly consumed nonnutritive sweeteners, often unknowingly. Thus, the participants may have recruited a group of participants with fairly unique dietary habits.
  • The sucralose and saccharin groups in the human study did not experience impaired glucose tolerance per se. An individual is considered to have impaired glucose tolerance if their blood glucose levels are between 141 mg/dL and 200 mg/dL two hours after a GTT. The current study assessed iAUC for blood glucose and noted statistically significant differences in the sucralose and saccharin groups compared to baseline and the control groups.
  • The effects of sucralose and saccharin on glycemic control in the human participants contrast a large body of literature indicating that NNSs do not affect postprandial glycemic responses.[2][3]
  • The changes in structural and functional microbiome parameters were not necessarily harmful. Many things affect the microbiome, and whether these changes were clinically significant is unclear.

Nevertheless, the results of this study do suggest that NNS are not physiologically inert. While a detrimental effect of NNS on glycemic control still appears unlikely, given the current body of literature on the topic, the clinical implications for the effects of NNS on the gut microbiome warrant further research.

The big picture

A substantial body of literature has assessed the potential for NNS to affect the microbiome.

The authors of a narrative review[4] published in August of 2022 noted that animal research suggests that NNS may alter the gut microbiome. However, they also noted that heterogeneity among studies makes interpretation difficult. In addition, they indicated that several studies suggest that saccharin, acesulfame K, and sucralose may deplete Akkermansia muciniphila in mice and that this depletion may be associated with impaired glucose tolerance. Consistent with the current study, they highlighted the possibility that NNS may affect the microbiome and physiology of some susceptible individuals but not others.

A narrative review[5] published in April of 2022 reviewed 6 in vitro studies, 14 in vivo animal studies, and 4 human studies. The authors noted that in vitro and animal studies suggest that high doses of saccharin and sucralose can modify the gut microbiome, whereas human studies have not demonstrated an effect of these sweeteners on the gut microbiome. The authors highlighted that human studies have used lower relative doses of saccharin and sucralose than in vitro or animal studies and that human studies have assessed short-term outcomes to date.

The authors of a 2020 narrative review[6] proposed that intake of NNS may alter the composition and function of the gut microbiome, potentially resulting in metabolic dysfunction. The review primarily focused on in vitro and animal studies, highlighting a potential link between NNS, sweet taste receptors, and gut microbiome alterations.

Every month we summarize over 150 of the most noteworthy health and nutrition studies. Other health categories related to this summary include:Try Examine+ for free to view the latest research in 25 health categories and the entire Study Summaries archive, access our Supplement Guides, and unlock the Examine Database. Plus, earn continuing education credits!

Get free weekly updates on what’s new at Examine.

This Study Summary was published on November 7, 2022.

References

  1. ^Suez J, Cohen Y, Valdés-Mas R, Mor U, Dori-Bachash M, Federici S, Zmora N, Leshem A, Heinemann M, Linevsky R, Zur M, Ben-Zeev Brik R, Bukimer A, Eliyahu-Miller S, Metz A, Fischbein R, Sharov O, Malitsky S, Itkin M, Stettner N, Harmelin A, Shapiro H, Stein-Thoeringer CK, Segal E, Elinav EPersonalized microbiome-driven effects of non-nutritive sweeteners on human glucose tolerance.Cell.(2022-Sep-01)
  2. ^Arno Greyling, Katherine M Appleton, Anne Raben, David J MelaAcute glycemic and insulinemic effects of low-energy sweeteners: a systematic review and meta-analysis of randomized controlled trialsAm J Clin Nutr.(2020 Oct 1)
  3. ^Nichol AD, Holle MJ, An RGlycemic impact of non-nutritive sweeteners: a systematic review and meta-analysis of randomized controlled trials.Eur J Clin Nutr.(2018-06)
  4. ^Richardson IL, Frese SANon-nutritive sweeteners and their impacts on the gut microbiome and host physiology.Front Nutr.(2022)
  5. ^Del Pozo S, Gómez-Martínez S, Díaz LE, Nova E, Urrialde R, Marcos APotential Effects of Sucralose and Saccharin on Gut Microbiota: A Review.Nutrients.(2022-Apr-18)
  6. ^Turner A, Veysey M, Keely S, Scarlett CJ, Lucock M, Beckett ELIntense Sweeteners, Taste Receptors and the Gut Microbiome: A Metabolic Health Perspective.Int J Environ Res Public Health.(2020-06-08)