Safety and effects of acetylated and butyrylated high-amylose maize starch on youths recently diagnosed with type 1 diabetes: A pilot study

Abstract

Studies have indicated differences in gut microbial composition in people with type 1 diabetes (T1D) compared with healthy controls.¹ These include reduced taxa associated with fermentation of dietary fibres to produce short-chain fatty acids (SCFAs).¹ The gut microbiome can be altered using high-amylose maize starch (HAMS), a well-tolerated source of dietary fibre. Following colonic bacterial fermentation, acetylated and butyrylated HAMS (HAMS-AB) releases large amounts of SCFAs,² preventing T1D development in mouse models and, among those with established T1D, resulting in anti-inflammatory and immunomodulatory effects.^{2, 3}

The primary outcome of this pilot study was the assessment of the safety of HAMS-AB and its effect on the gut microbiome in people with recently diagnosed T1D. We hypothesized that HAMS-AB consumption would be safe in adolescents recently diagnosed with T1D and that it would result in changes in the gut microbiome composition compared with those not consuming HAMS-AB. Secondary outcomes included HAMS-AB's effects on stool SCFAs, glycaemia and β-cell function and mucosal-associated invariant T (MAIT) cell frequency and function. Post hoc exploratory analysis of circulating metabolites was also performed.

The full study protocol has been previously published.⁴ The study was registered with ClinicalTrials.gov under NCT04114357, and ethical approval was obtained at Indiana University (protocol number 1908640459). Briefly, after consent was obtained from parents/legal guardians and assent from participants, participants were randomised to start with either HAMS-AB and the standard recommended diabetes diet guidelines at home for 4 weeks or just the recommended diabetes diet for 4 weeks, with a 4-week washout period and then a crossover to the other arm for 4 weeks (12-week study period) (Figure S1 and Table). We used a crossover design to allow for assessment of HAMS-AB efficacy through comparison of individuals with themselves as their own controls. We used the recommended diabetes diet guidelines for participants, which is the standard of care, as the control for comparison. Briefly, individuals were counselled on the recommended total energy intake to maintain a healthy body mass index (BMI). In accordance with the 2018 International Society for Pediatric and Adolescent Diabetes (ISPAD) guidelines,⁵ we recommend the following macronutrient distribution for participants' three main meals: carbohydrate intake should approximate 45%–50% of energy; fat, <35% of energy (saturated fat <10%) and protein, 15%–20% of energy. Participants in this study (as is also standard practice in our clinic) were counselled on the glycaemic index of different foods as well as the general recommended fibre intake. Diet intake was assessed using the Automated Self-Administered 24-h (ASA24) dietary assessment method.⁶

Key inclusion criteria included the following: children (aged 11–17 years), BMI <85% for age and sex and T1D duration of 4–36 months. Additional criteria included a random C-peptide >0.17 nmol/L measured during the screening visit and being able to consume the test dose of HAMS-AB.⁴ Those who did not meet inclusion criteria or failed the screening visit were excluded from the study.

Stool collections were performed at home as has been previously described.⁷ Briefly, stool sample kits (consisting of gloves, a Zymo faeces catcher, RNA/DNA shield faecal collection tubes with preservative [for DNA sequencing] and without preservation [for SCFA analysis] and freezer packs) were shipped to participants, who were asked to collect a stool sample at home within 1–3 days prior to each study visit, except for the screening visit. Participants then hand delivered the samples during the research visit. Samples were taken within 1 h from delivery and placed in a −80°C freezer and stored there until analysis.

Participants were instructed to consume HAMS-AB orally with food, such as apple sauce or oatmeal, in two divided doses at breakfast and dinner at a total daily dose to be calculated as has been previously described for children: 10 g plus 1 g per year of age daily.⁸

Recruitment was from July 2020 to December 2022. Twelve participants were enrolled; seven finished the study. Three withdrew prior to consumption of HAMS-AB; anxiety around blood draws, family stress and struggling to follow the diabetes diet were the reasons reported. The other two did not tolerate HAMS-AB; one developed gagging with attempted consumption, and the second developed nausea. Symptoms resolved with HAMS-AB discontinuation (Figure S2).

Data from the remaining seven individuals were considered sufficient to proceed to a phase Ib trial, thus resulting in closure of this phase Ia trial. Table S2 shows the baseline characteristics of the seven participants who completed.

In this phase Ia clinical trial, we examined the safety of HAMS-AB consumption in youths recently diagnosed with T1D and its effects on the gut microbiome, metabolites, immune markers and glycaemia. We saw an acceptable safety profile of HAMS-AB, with no serious adverse events (SAEs). Most AEs were mild/moderate, all resolved before the end of the study period. We saw changes in the gut microbiome composition, metabolite profile and immune markers associated with HAMS-AB consumption. Therefore, our findings suggest the potential for HAMS-AB use in T1D management and infer disease-modifying effects, thus establishing the premise for further testing HAMS-AB effects in a larger sample size with more data collection.

HAMS-AB consumption led to an increased relative abundance of Bifidobacterium and Parabacteroides at the genus level. We also saw an increased relative abundance of B. longum and P. distasonis. Bifidobacteria are generally fermenters and SCFA producers that are typically decreased in T1D.¹ Meanwhile, P. distasonis are described as lower in individuals with a high-risk genotype for T1D.¹⁰ We also examined a possible carry-over effect and found a significant treatment effect on Parabacteroides and Bifidobacterium.

Following HAMS-AB consumption, there was a trend towards a significant increase for butyrate. We saw a downregulation of geraniol degradation and lipoic acid metabolism functional pathways. Geraniol is an acyclic monoterpene alcohol with well-known anti-inflammatory and antimicrobial properties.¹¹ Therefore, reduced degradation suggests persistence of its anti-inflammatory effects. Meanwhile, lipoic acid metabolism has been shown to be enriched in those with long-standing diabetes and nephropathy.¹² However, when comparing post-treatment periods, differences in these pathways were not seen.

Metabolomics analysis revealed an increase in metabolites associated with the gut microbiome, glycaemia and energy homeostasis. Hippurate increased post HAMS-AB and is a microbial metabolite associated with increased gut bacterial diversity and improved glycaemia.¹³ l-Glutamic acid is an important intermediate in metabolism and has been touted with potential for glycaemic control.¹⁴ Dihydroxyquinoline has protective and homeostatic effects on the intestinal tract by suppressing inflammation.¹⁵ Meanwhile, tryptophan, partially produced by the gut microbiome, is associated with reduced inflammation and regulating energy homeostasis.¹⁶

MAIT cells are innate-like T cells that are involved in the mucosal immune response.⁹ They are thought to play a key role in maintenance of gut integrity, thereby potentially providing a link between the gut microbiome changes and autoimmunity. We saw a reduction in the activation state of MAIT cells marked by reduced CCR6+, CD25+, PD1+ and BCL2⁻Granzyme B+ MAIT cells when comparing before with after HAMS-AB. The latter was also significantly reduced when comparing post-HAMS-AB with post-diet alone. This reduction in activation state is closely linked to a reduced inflammatory response and promotion of a more regulated immune profile.

This study has strengths. Several measures and metabolic markers were assessed, which is informative in evaluating effect size and for informing a fully powered study. The study used a cross-over design where individuals were their own controls, thus minimizing potential covariates that may affect results. Additionally, despite the dropout rate, several measures indicate a positive effect of HAMS-AB. Further, HAMS is a natural supplement that may be favoured by many patients. Limitations include the small sample size and short duration of intake as well as lack of a placebo for comparison. However, findings are hypothesis generating, and indeed, a larger phase Ib trial to assess these effects is underway (NCT06057454). Further, two individuals did not tolerate HAMS-AB, which stresses the individual differences in tolerance to dietary agents and the need for personalized treatment approaches. Despite this, we believe that our initial data are encouraging and have accomplished the goal of assessing safety and of providing an effect-size estimate.

Heba M. Ismail conceived the study and drafted and edited the manuscript. Nur A. Hasan and Michael Netherland Jr performed the bioinformatics analysis. Carmella Evans-Molina, Linda A. DiMeglio, Nur A. Hasan, Michael Netherland Jr and Jianyun Liu contributed to the study design, critically reviewed the manuscript and approved the final version. All have consented to the manuscript publication.

This study received support from the National Institutes of Health and National Center for Advancing Translational Sciences, Clinical and Translational Sciences Award, Grant Numbers KL2TR002530 (A. Carroll, PI) and UL1TR002529 (A. Shekhar, PI). We also acknowledge support from the Board of Directors of the Indiana University Health Values Fund for Research Award and the Indiana Clinical and Translational Sciences Institute funded, in part, by Grant U54TR002529 from the National Institutes of Health, National Center for Advancing Translational Sciences, Clinical and Translational Sciences Award; the Indiana Clinical and Translational Sciences Institute funded, in part, by Award Number ULITR002529 from the National Institutes of Health, National Center for Advancing Translational Sciences, Clinical and Translational Sciences Award; the Pilot and Feasibility Grant from the Indiana Center for Diabetes and Metabolic Diseases (P30DK097512); the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health under Award Number K23DK129799; the Doris Duke Charitable Foundation through the COVID-19 Fund to Retain Clinical Scientists Collaborative Grant Program (Grant 2021258) and The John Templeton Foundation (Grant 62288). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or other funding agencies.

The authors declare no conflict of interest.