Article Text

Positive effects of diet-induced microbiome modification on GDM in mice following human faecal transfer
  1. Sigal Frishman1,2,
  2. Meital Nuriel-Ohayon3,
  3. Sondra Turjeman3,
  4. Yishay Pinto3,
  5. Or Yariv4,
  6. Kinneret Tenenbaum-Gavish5,
  7. Yoav Peled1,
  8. Eran Poran4,
  9. Joseph Pardo1,
  10. Rony Chen1,
  11. Efrat Muller6,
  12. Elhanan Borenstein1,7,
  13. Moshe Hod1,
  14. Yoram Louzoun8,
  15. Betty Schwartz2,
  16. Eran Hadar9,
  17. Maria Carmen Collado10,
  18. Omry Koren3,11
  1. 1Tel Aviv University, Tel Aviv, Israel
  2. 2Faculty of Agriculture, Hebrew University, Rehovot, Israel
  3. 3Azrieli Faculty of Medicine, Bar-Ilan University, Safed, Israel
  4. 4Clalit Health Services, Tel Aviv, Israel
  5. 5Rabin Medical Center and the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
  6. 6The Blavatnik School of Computer Science, Tel Aviv University, Tel Aviv, Israel
  7. 7Santa Fe Institute, Santa Fe, New Mexico, USA
  8. 8Department of Mathematics, Bar-Ilan University, Ramat Gan, Israel
  9. 9Helen Schneider Hospital for Women, Rabin Medical Center, Petach-Tikva, Israel
  10. 10Biotechnology, Unit of Probiotic, IATA-CSIC, Valencia, Spain
  11. 11Kyung Hee University, Seoul, Korea (the Republic of)
  1. Correspondence to Dr Omry Koren, Azrieli Faculty of Medicine, Bar-Ilan University, Safed 1311502, Israel; omry.koren{at}biu.ac.il

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

We recently reported in Gut that the microbiome is unequivocally implicated in early gestational diabetes mellitus (GDM) aetiology, starting in the first trimester (T1),1 and other groups have shown continued microbiota dysbiosis in women with GDM in second trimester (T2) and third trimester (T3).2 In continuation of our T1 research, we now have data showing that dietary interventions, the preferred and primary treatment of GDM, are effective in part by altering the gut microbiota. To elucidate the causal role of the microbiome on GDM, we performed faecal microbiota transplant (FMT) of samples from age/body mass index-matched women with and without GDM (n=5 each, table 1 and online supplemental table 1) in T2 and in T3, following dietary intervention, to germ-free mice to elucidate microbiome-mediated effects of diet on GDM (figure 1A). Retrospective analysis of donor samples suggests different microbiota compositions between the groups, irrespective of trimester (figure 1B,C); no differentially abundant taxa were identified.

Supplemental material

Figure 1

GDM phenotype transfer experiments based on faecal microbiota transplants (FMTs) from women in T2 and T3 of pregnancy. (A) Experimental design. (B,C) Faecal microbiota characterisation of the FMT donor samples. (B) Pregnant donors had trends of different α diversity between GDM status with a trending interaction effect of trimester (Faith’s PD, p<0.075). (C) There was also a trend toward different β diversity (unweighted UniFrac, p=0.068). (D–G) Mouse results from FMT experiment. (D) There was no difference in fasting glucose levels (time 0), but glucose levels were significantly higher in mice receiving the GDM FMT from T2 at 30 min after injection. (E) Similarly, in T3 FMT, there was no difference in fasting glucose levels at time 0, but glucose levels were significantly higher in mice receiving the control FMT from women with GDM at 30 and 60 min after injection. Asterisks indicate a statistically significant difference (**p<0.01; ***p<0.001) as determined by two-tailed Student’s t-tests. (F) There were significant differences in β diversity following FMT of T2 GDM and control samples (unweighted UniFrac, day 7, p=0.006) but only a trend towards difference in the T3 samples (not shown, p=0.07), suggesting that dietary modification can correct microbial dysbiosis. (G) Differential abundance analysis identified 10 differentially expressed genera. The heatmap is coloured by log-fold change (LFC)-adjusted p value. Positive values in the trimester column identify taxa over-represented in the samples collected in T3. Positive values in the GDM status column represent taxa over-represented in samples from women with GDM. GDM, gestational diabetes mellitus; IPGTT, intraperitoneal glucose tolerance test; PD, phylogenetic diversity; T2, second trimester; T3, third trimester.

Table 1

Donor cohort characteristics

To examine the effect of FMT on GDM phenotype, intraperitoneal glucose tolerance tests were performed. Variation in glucose metabolism was observed between the groups in each of the trimesters (figure 1D,E). Among mice that received FMT from women in T2, before dietary intervention, we found significantly higher glucose levels in the GDM group 30 min after the glucose injection (t-test: p=0.032); however, in mice that received T3 FMT, post-dietary intervention, we found significantly higher glucose levels in the control group at two time points, 15 (p=0.001) and 30 (p=0.042) min after injection. When considering total area under the curve, a significant difference was found in T3 with the control group having a significantly larger area (p=0.03). Fasting glucose on the day of the challenge (day 21) and on day 14 post-FMT was not significantly different. The weights of mice between the groups did not differ either. Further, serum (post-sacrifice, non-fasting) levels of tumour necrosis factor-α, interferon-γ, interleukin (IL)-6, IL-10, IL-17α, insulin and leptin did not vary between the groups of mice that received transplants from women with or without GDM from the two trimesters (multiplexed ELISA; p>0.5; online supplemental table 2).

To understand the role of the microbiota and related metabolism, we characterised the faecal microbiota of FMT-recipient mice (as in Pinto et al1) and found significant differences in β diversity between the groups receiving transplants from T2 (day 7 post-FMT; figure 1F; p=0.006). For T3 sample transplanted to mice, we found only a trend towards compositional differences (day 7 post-FMT, p=0.07, not shown). Additionally, when combining samples taken on days 7 and 21 post-FMT (to increase sample size) for differential abundance analysis, we identified 10 taxa differentially represented among groups and trimesters (figure 1G). PICRUSt23 was used to predict MetaCyc metabolic pathway4 abundance from microbiota data. Six differentially enriched pathways were found between mice receiving FMT from women with and without GDM in T2 (online supplemental table 3). Protein (three pathways related to L-isoleucine biosynthesis and the L-valine biosynthesis), nucleotide (UMP biosynthesis) and coenzyme A (CoA) biosynthesis pathways were elevated in the control group and were then slightly reduced in T3 (within the control group, linear mixed-effect model p=0.01–0.15), though this trend was not significant after false discovery rate (FDR) correction. In contrast, in GDM-recipient mice, the CoA biosynthesis pathway abundance slightly increased in T3 compared with T2 (p=0.014), but this result, like the previous one, was not significant after FDR correction. Seven differentially enriched pathways were found between GDM-recipient mice and control-recipient mice in T3 (online supplemental table 3), all of which were elevated in the control group. These included polysaccharide (L-rhamnose) biosynthesis, ubiquinone biosynthesis, pyruvate fermentation and methylphosphonate degradation pathways.

Together, these findings provide causal evidence that not only is the microbiota continually implicated in GDM aetiology (T2 results), but that dietary intervention works, at least in part, via microbiota-mediated effects, as seen by improved glucose tolerance among the T3 GDM-recipient mice. In the future, deeper phenotyping should be used (insulin tolerance, basal glucose) and be performed longitudinally, including time points closer to the FMT. These findings, though, already begin to extend the story we told in our recently published paper1 and further highlight the importance of the microbiota, not only in GDM diagnosis but also in its control.

Data availability statement

Data are available in a public, open access repository. All sequencing data were submitted to EBI (project accession number ERP143097).

Ethics statements

Patient consent for publication

Ethics approval

Informed consent was obtained from all participants, in accordance with Clalit’s Institutional Review Board approvals (no. 0135-15-COM and no. 0263-15-RMC). Animal ethics approval was given by Azrieli Faculty of Medicine Animal Ethics Committee (approval number 33-04-2018).

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Twitter @OmryKoren

  • Contributors SF, BS, EH, MCC and OK conceived and designed the study. SF, OY, KT-G, YoP, EP, JP, RC, MH and EH collected the clinical samples and metadata. SF, MN-O, ST, YiP, EM, EB and YL conducted the experiments and analysed the data. SF, ST and OK wrote the manuscript. All authors read and approved the final manuscript.

  • Funding This study was funded by the Israeli Ministry of Innovation, Science & Technology (grant number 3-15521). OK and MCC acknowledge the support by Biostime Institute Nutrition & Care (BINC) research grant. OK is supported by the European Research Council Consolidator grant (grant agreement no. 101001355).

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; internally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.