
==== Front
Gut
Gut
gutjnl
gut
Gut
0017-5749
1468-3288
BMJ Publishing Group BMA House, Tavistock Square, London, WC1H 9JR

gutjnl-2023-331456
10.1136/gutjnl-2023-331456
PostScript
1506
2312
LetterPositive effects of diet-induced microbiome modification on GDM in mice following human faecal transfer
Frishman Sigal 12
Nuriel-Ohayon Meital 3
Turjeman Sondra 3
http://orcid.org/0000-0002-5280-1040
Pinto Yishay 3
Yariv Or 4
Tenenbaum-Gavish Kinneret 5
Peled Yoav 1
Poran Eran 4
Pardo Joseph 1
Chen Rony 1
Muller Efrat 6
http://orcid.org/0000-0003-3002-0945
Borenstein Elhanan 17
Hod Moshe 1
Louzoun Yoram 8
Schwartz Betty 2
http://orcid.org/0000-0002-3101-2827
Hadar Eran 9
Collado Maria Carmen 10
http://orcid.org/0000-0002-7738-1337
Koren Omry 311
1 Tel Aviv University, Tel Aviv, Israel
2 Faculty of Agriculture, Hebrew University, Rehovot, Israel
3 Azrieli Faculty of Medicine, Bar-Ilan University, Safed, Israel
4 Clalit Health Services, Tel Aviv, Israel
5 Rabin Medical Center and the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
6 The Blavatnik School of Computer Science, Tel Aviv University, Tel Aviv, Israel
7 Santa Fe Institute, Santa Fe, New Mexico, USA
8 Department of Mathematics, Bar-Ilan University, Ramat Gan, Israel
9 Helen Schneider Hospital for Women, Rabin Medical Center, Petach-Tikva, Israel
10 Biotechnology, Unit of Probiotic, IATA-CSIC, Valencia, Spain
11 Kyung Hee University, Seoul, Korea (the Republic of)
Correspondence to Dr Omry Koren, Azrieli Faculty of Medicine, Bar-Ilan University, Safed 1311502, Israel; omry.koren@biu.ac.il
10 2024
5 1 2024
73 10 e17e17
01 11 2023
14 12 2023
© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.
2024
https://creativecommons.org/licenses/by-nc/4.0/ This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

INTESTINAL MICROBIOLOGY
DIABETES MELLITUS
DIET
http://dx.doi.org/10.13039/501100000781 European Research Council 101001355 Biostime Institute for Nutrition and Care - Geneva BINC Israeli Ministry of Innovation, Science & Technology 3-15521 special-featureunlocked
==== Body
pmcWe 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.10.1136/gutjnl-2023-331456.supp1 Supplementary data

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

	Control	GDM	P value	
Sample size*	7	6		
Body mass index, kg/m2	23.1 (22.5–25.0)	29.0 (21.9–35.3)	0.4	
Age, years	31 (30–36)	34 (32–37)	0.6	
Neonatal birth weight, g	3.39 (3.11–3.46)	3.16 (3.05–3.56)	>0.9	
Education, years			0.6	
 12	0 (0%)	1 (17%)		
 15	3 (50%)	2 (33%)		
 16	2 (33%)	1 (17%)		
 17	0 (0%)	2 (33%)		
 20	1 (17%)	0 (0%)		
 Unknown	1	0		
Vegetarian/vegan	1 (14%)	0 (0%)	>0.9	
Continuous variables are presented as median and (IQR). Categorical variables are presented as count (%). Mann-Whitney U tests and Χ2 tests were used to compare variables between the groups.

*Five donor samples per group per trimester were used. In a few cases, the same donor could not be used for each trimester. One sequenced GDM-T2 sample did not pass quality control and was removed from subsequent analyses. See online supplemental table 1.

GDM, gestational diabetes mellitus; T2, second trimester.

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 al 1) 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

Obtained.

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).

X @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.
==== Refs
References

1 Pinto Y , Frishman S , Turjeman S , et al . Gestational diabetes is driven by microbiota-induced inflammation months before diagnosis. Gut 2023;72 :918–28. 10.1136/gutjnl-2022-328406 36627187
2 Kuang Y-S , Lu J-H , Li S-H , et al . Connections between the human gut microbiome and gestational diabetes mellitus. Gigascience 2017;6 :1–12. 10.1093/gigascience/gix058
3 Douglas GM , Maffei VJ , Zaneveld JR , et al . PICRUSt2 for prediction of metagenome functions. Nat Biotechnol 2020;38 :685–8. 10.1038/s41587-020-0548-6 32483366
4 Karp PD , Riley M , Paley SM , et al . The MetaCyc database. Nucleic Acids Res 2002;30 :59–61. 10.1093/nar/30.1.59 11752254
