# HSEF v5.3 review: Microbiome-driven Obesity Hypothesis

> Sources: 68 primary reports; claims: 72; verified: 72; unverified: 0. Verification is bounded attribution. Access: 10 partial-body and 58 complete abstract/status examinations. Cutoff: 2026-10-05. No complete-paper audit or blinded calibration claimed.
> **Adversarial review:** Generated by Sol 6.1; adversarially reviewed by Claude Opus 5.5 (Anthropic) on 2026-10-06. 2 corrections applied (see ADVERSARIAL REVIEW AND APPROVAL). Status: **Approved as revised.**

## AT A GLANCE

**Microbes can causally alter adiposity in defined mice, and selected human interventions support a limited contribution to fat accumulation or weight maintenance. A general living microbial configuration that drives human obesity beyond diet and host physiology remains unestablished.**

| Level | Confidence | Causal role | Momentum | Stability | Divergence |
|---|---|---|---|---|---|
| Δ — First field-valid causal perturbation crossed | Moderate | reinforcing / stabilizing factor | Mechanistic only | Stable | Contested |

**Deciding experiment:** Randomize a prespecified human obesity population to a defined live strain/consortium, an otherwise matched inactive microbial preparation and placebo, under matched dietary exposure. Use a registered fat-mass primary ITT endpoint over6-12 months, separate maintenance from initial loss, and measure strain/function exposure, energy intake/loss, expenditure and candidate mediators longitudinally. Confirm the same function with subtraction/addition and rescue in two independent recipient facilities; replicate clinical effects in a separate cohort.

**Belief versus evidence:** The preregistered scope-matched human-contribution gate SC2 yields Δ / Moderate. Selected controlled human adiposity contrasts cross the in-vivo causal boundary, so a blanket Γ on mouse-to-human inference is not reapplied to the convergent human set. SC1 reaches scoped Ζ for animal transmissibility/rescue; SC3 remains Γ for the endpoint-substitution inference. Weakest-link sensitivity is Γ, primary-subclaim and scope-matched Δ. A stronger universal or primary human-driver reading would remain Γ; that is a separate burden, not a requirement silently added to the registered contribution gate.

**Best-supported rival:** Diet and host energy balance with microbial interaction. Controlled high-energy feeding with a randomized microbial product [26] separates additional microbial contribution from diet exposure alone; FMT-TRIM [55] shows engraftment can occur without fat or clamp benefit. Those instruments favor interaction and endpoint-specific modulation over either pure co-occurrence or a universal obesity configuration.

## COMPETING MECHANISTIC MODELS

**Rank 1: Diet and host energy balance with microbial interaction (Ε).** Diet supplies substrate and changes microbial ecology; microbial effects depend on energy exposure and host physiology. Diet intervention and matched-diet transfer contrasts, host receptors and diet/strain counterexperiments [2,26,30,34,38-39,51,60,64-65,67-68] map causal interactions. Diet does not erase microbial contribution: controlled diet plus VSL3 changes fat gain [26], and defined transfer/rescue alters animal adiposity.

**Rank 2: Metabolic-health modulation with partly independent adiposity (Ε).** Microbial interventions can modify insulin, bile or inflammation without a large or consistent body-fat effect. Clamp and antibiotic/FMT contrasts with adiposity nulls support endpoint separation [11,14,18-19,54-59,63]. Selected positive fat or weight-maintenance trials prevent an absolute no-adiposity reading [9,21,23-26,53,65].

**Rank 3: Microbial human adiposity contribution (Δ).** The registered contributory claim: specific microbial interventions can modify human fat accumulation under bounded conditions. Randomized common-diet fat-gain attenuation and selected adiposity contrasts, plus new primary weight maintenance [9,21,23-26,53], cross the human causal boundary. A universal primary configuration remains Γ: multiple fat/weight-primary nulls, diet dependence, nonliving products and incomplete specificity [5,16,20,22,55,58,61-62,65-66].

**Rank 4: Host physiology and reverse direction (Ε).** Host genes, pregnancy, medication and energy exposure can remodel microbes before the measured phenotype. Host-state mouse contrasts and longitudinal diet/drug observations provide scoped directionality [3-4,10,31,39,41,44,48,50-51,67]. These effects do not refute forward microbial causality; human observational direction remains partly unresolved.

**Rank 5: Homeostatic compensation (Ε).** Microbial calories and signalling can be offset or redirected by appetite, expenditure and absorptive adaptation. Cold-transfer adaptation and GPR43 substrate signalling demonstrate model compensation [67-68]; protective carbohydrate consumption adds an opposite energy-harvest route [49]. Human quantitative long-term compensation is not identified by these animal experiments; the model is compatible with some genuine fat effects.

The tournament does not alter the nominal level. Compatible pathways and flexible alternatives are not automatically proved by explaining observations.

## FULL REVIEW

### Registered claim and scope

Gut microbial configuration causally drives human fat accumulation through energy harvest, microbial products and endocrine signalling, beyond diet and host physiology.

The index specifies scope-matched human SC2 as the gate. Mouse transfer and human insulin-sensitivity effects are not substituted for human adiposity causation; added SC3 makes that endpoint-transfer claim explicit.

Selection: public-interest rank 10 from the editorial ranking compiled 2026-09-23; categorical prior verdicts were visible. Private fresh review; earlier documents are preserved.

### Evidence and causal interpretation

The nominated claim asks whether microbial configuration adds a causal contribution to human fat accumulation beyond shared diet and host influences. That is a lower burden than universal microbial necessity or primary initiation. The registered human SC2 is the gate. This review gives positive human adiposity experiments their credit without allowing mouse transfer or an insulin endpoint to replace human fat evidence.

The preclinical causal case is substantial. Conventionalization/Fiaf, obese-community transfer, cultured discordant-twin communities and Christensenella amendment change recipient adiposity [1-3,6,64]. Ridaura explicitly shows lean-community invasion and diet-dependent rescue [2]. These experiments address recipient causality; the human donors do not become randomized human causal recipients. Crude FMT remains Moderate specificity under rubric04, while defined rescue and function/pathway contrasts provide stronger narrowing.

Endocrine and absorptive routes include GPR41/PYY/transit, FXR/bile acids, epithelial NFIL3/HDAC3 and small-intestinal microbial adaptation [4,10,13,15,42]. Defined metabolite/strain studies show protective and obesogenic functions rather than one dysbiosis sign: Megamonas/iolG degrades myo-inositol [28], Turicibacter lipids restrain ceramides/fat uptake [12], and microbial BSH/bile acids connect ketogenic diet to calorie absorption [34]. Human portions of these studies are associations or dietary interventions, not randomized microbial mediators.

Original GPR43 sections show an important opposite direction [67]. Receptor effects disappear with germ-free conditions or antibiotics, reappear with conventionalization and are restored by acetate in a receptor-dependent way. KO and overexpression mice have different genetic backgrounds, and lifelong manipulation remains a limitation. The result supports context-specific energy regulation; it cannot justify a universal rule that more fermentation causes more fat. Cold-community transfer first causes browning/fat loss and later increases absorptive adaptation [68], while commensal carbohydrate consumption can reduce host-accessible energy [49].

A strong diet/host counterexample is Fleissner: germ-free C3Hmice on one high-fat formula gained more adiposity than conventional animals, with a different Western-diet result [60]. Host genotype, pregnancy, circadian pathways and medications can themselves alter communities [3-4,10,31,39,41,44,48,50-51]. These contrasts constrain universal microbial necessity. They do not erase forward microbial effects in the defined positive systems.

The clearest small human fat-accumulation instrument is VSL3 during common high-energy/high-fat feeding [26]. Twenty healthy young men were randomized to product or placebo; DXA fat gain was0.63versus1.29 kg over4 weeks. This supports additional human causal contribution within that dietary exposure, but the healthy non-obese cohort is outside the registered obese-population gate and does not alone carry SC2. It is short, small and abstract-only in this review, with a multistrain product and unverified functional engraftment; it cannot establish one endogenous human obesity configuration.

Other human formulation trials have mixed signals [21,23-24]. Ba8145 original Methods/Results show live-cell BMI and killed-cell waist-related benefits, but the MRI visceral-fat between-placebo contrasts are nonsignificant despite a within-baseline reduction;126 received intervention, and killed-arm fiber was higher. Its statistical supplement remains unread. BBr60 reports between-treatment weight/BMI benefit. L.rhamnosus plus prebiotics has an overall weight null but a treatment-by-sex interaction and female fat/weight benefit. The latter trials remain abstract-only, with primary hierarchy and multiplicity unresolved. These signals retain their measured endpoint credit, not universal total-fat efficacy.

B420 illustrates why endpoint hierarchy matters [22]. Original Methods confirm the fat-mass primary, modifiedITT209 with last-observation-carried-forward, Dunnett comparisons and a post-unblinding factorial PP analysis. ITT fat was null; the combination was positive among selected compliant participants, and the factorial signal was posthoc. The title-level conclusion therefore exceeds a clean primary-ITT interpretation. Reduced intake may be a legitimate causal route but does not identify community-mediated fat storage; zonulin correlation is not a validated barrier mechanism.

GutBugs gives a randomized fat-distribution signal despite BMI-SDS primary and total-fat percentage nulls [9]. Original Methods/Results confirm prespecified secondary DXA android/gynoid ratio, no clinical multiplicity adjustment, and posthoc metabolic-syndrome subgroup analysis. The2025 four-year paper [25] is the same trial, not independent replication:55/87 returned, allocations had been unblinded, and clinical p-values were nominal without imputation. Lower fat percentage and waist among returnees deserve attention, but attrition, intervening exposures and selection limit long-term causal attribution.

The2026 pasteurized-Akkermansia trial strengthens the human weight-maintenance case [53]. Ninety adults underwent low-energy-diet loss before randomized supplementation; reported24 week regain was1.2versus3.2 kg with primaryP=.012. Consensus and the publisher exposed the abstract, not Methods/Results or fat outcomes. Publisher ancillary statements disclose company leadership/employment/advisory positions and patent inventorship. This is a genuine reported primary clinical positive, with full allocation/attrition/multiplicity and total-fat details unresolved. A nonliving product may act directly on host signalling without living-community engraftment.

The2019Akk pilot had metabolic positives and nonsignificant weight/fat effects nearP=.09 [16]. Original sections confirm32 completers of 40, bioimpedance rather than DXA, baseline insulin imbalance and limited calorie/activity measurements. The abstract and main body phrase the fat comparator differently; the nonsignificant result is retained without inventing a significant fat contrast. Preliminary human volunteers in the2017mouse/protein paper overlap this programme [17]. The2025 live-Akk trial [5] found no whole-group weight/HbA1c difference but a low-baseline subgroup signal; its pre-specification, between-group interaction and multiplicity were not available.

Human FMT most consistently tests engraftment or glucose rather than adiposity. Lean-FMT clamp benefit is transient [11,54]; metabolic-syndrome donors can worsen clamp sensitivity [43]. Repeated FMT succeeds on an engraftment primary [19], while FMT-TRIM has original primary-clamp and fat/weight nulls despite donor uptake [55]. Its small sample excludes neither subtle benefit nor selected contexts, but shows that engagement alone is insufficient. The Mocanu abstract emphasizes a within-arm HOMA improvement [56]; without the original randomized between-arm comparison, significance in one arm and not another cannot be treated as a demonstrated differential effect. The responder follow-up [57] is the same cohort.

Human energy absorption has a causal perturbation signal: inpatient vancomycin increased primary stool calorie loss [38]. This is stronger than Jumpertz's diet-linked150 kcal association [59], but still short and not a sustained adiposity outcome. Another antibiotic RCT found no clinically relevant metabolic/adipocyte-size effect despite ecology changes [58], while an active-comparator trial found worsened insulin sensitivity [63]. Antibiotic pleiotropy and context explain why these do not establish one taxonomic energy-harvest rule.

Propionate delivery supplies a cautionary positive/null sequence [61-62]. Original2015Methods name mean weight and food intake as coprimary: mean weight differenceP=.099 and total-adipose difference null, despite fewer threshold-defined weight-gain events and a secondary intra-abdominal distribution signal. The2024iPREVENT trial randomized 270; original model included256 with any follow-up and226at12 months. Primary weight and secondary fat-mass contrasts were null. Adherence and low control gain constrain interpretation; they do not transform the trial into a primary success.

Rinott aFMT also joins an overall primary null with a diet-specific effect [65]. Original web Methods confirmITT90 with one imputation and planned diet/treatment interaction analysis; primary regainP=.28, green-Mediterranean interactionP=.03. Diet-stratified weight maintenance retains scoped credit, while the2023taxa analysis [66] is a selected follow-up of the same trial. Original protocol and supplement pre-specification were not independently audited.

Consensus supported discovery, missing-aspect checks and gaps, but did not supply an authoritative grade. Deep omitted the2026positive trial and labelled the Ridaura transfer as Nowotny2013 in places; exact PubMed identities governed citations. Its meter and search funnel were excluded from counts and scores. Pro supported partial Rinott details checked against the original, correctly bounded Mount to abstract access in its final table, and declared several requested papers unavailable despite earlier FullText tool labels. Direct originals filled the B420, GutBugs and propionate gaps. Methods therefore report specific checked aspects and access failures rather than general full-text validation.

Sixty-eight primary reports have complete abstract/status examinations and ten have selected original body-section examinations, including one web summary receipt for Rinott. Zero complete-paper/supplement audits are claimed. Primary clinical positive and null results, report families, ancillary commercial disclosures and unresolved corrections/access determine Moderate confidence. The Δ human-contribution judgement is neither universal primacy nor a recommendation to use these products clinically.

## HYPOTHESIS STATE EVALUATION FRAMEWORK (HSEF v5.3)

<!-- consensus-consistency-2026-10-09 -->
### Consensus provenance reconciliation — 9 October 2026

Case-specific Consensus discovery records were located and linked to this final edition. This supplements the review's provenance; it leaves the original assessed claim, grades and source counts intact. Provider summaries and full-text badges do not establish examination of original primary passages by this reviewer. Source access and incorporation retain the limits stated in the original methods and any dated research update.

Recorded evidence artifacts:

- `deliverables/hsef_v53_popularity_ranks_06_10_2026-10-05/microbiome-human-obesity/consensus/consensus-result.png`
- `deliverables/hsef_v53_popularity_ranks_06_10_2026-10-05/microbiome-human-obesity/consensus/deep-dom.txt`
- `deliverables/hsef_v53_popularity_ranks_06_10_2026-10-05/microbiome-human-obesity/consensus/deep-response.txt`
- `deliverables/hsef_v53_popularity_ranks_06_10_2026-10-05/microbiome-human-obesity/consensus/final-dom.txt`
- `deliverables/hsef_v53_popularity_ranks_06_10_2026-10-05/microbiome-human-obesity/consensus/fulltext-check-response.txt`

The located artifacts do not state a reliable source-comparison count. No count is inferred.

A dated adversarial approval record exists for the separately identified microbiome-human-obesity.review.md (2026-10-06; see edition-approval-audit.json). Its scope is that recorded edition. This supplement has no separate approval and no coverage is inferred from topic identity. No score or confidence change is made by this reconciliation.
<!-- end-consensus-consistency -->


### Subclaim matrix and aggregation

| Subclaim | Scope | Level | Decision |
|---|---|---|---|
| SC1 Microbial transfer transmits an adiposity phenotype in defined recipient mice. | specified donor, diet, housing and germ-free/conventional recipient conditions | Ζ | Transfer, cultured-community/strain amendment and rescue repeatedly move recipient adiposity, with defined functions and host-receptor tests addressing distinct errors [1-4,6-7,10,12-15,17,28,33-37,41-42,45-49,64,67-68]. The broad defined-recipient claim reaches Ζ across distinct mechanisms; this does not promote every individual strain or infer human causality. |
| SC2 The same microbial mechanism contributes meaningfully to human obesity. | humans with prespecified baseline obesity/metabolic status | Δ | The Δ crossing rests on a randomized human perturbation with a positive registered primary endpoint in the scoped population: the 2026 pasteurized-Akkermansia weight-maintenance trial [53]. Obese-population strain/formulation trial signals and the randomized FMT fat-distribution contrast add supporting human evidence [9,21,23-25], but GutBugs was null on its BMI-SDS primary and on total body-fat percentage, so its prespecified secondary android/gynoid result, obtained without multiplicity control, supports rather than supplies the crossing. VSL3 high-energy-diet fat-gain attenuation [26] supports human physiology but is outside the registered obese-recipient gate. The 2026 pasteurized-Akk primary weight-maintenance benefit is additional clinical evidence [53], but does not alone establish a living-community or total-fat mechanism. Primary nulls, selected analyses, incomplete mediator attribution and access prevent Ε generalization [5,16,20,22,54-62,65-66]. No universal necessity or primacy burden is imposed on this contributory subclaim. |
| SC3 Microbial effects on insulin sensitivity or inflammation establish that microbes control human body-fat accumulation. | human intervention populations; adiposity distinct from biomarkers | Γ | Glucose/insulin and inflammatory effects do not by themselves establish body-fat control. Mouse depletion, human engraftment and clamp interventions can change those readouts without corresponding fat changes [11,14,18-19,38,40,43,52,54-59,63]. Positive fat trials retain their own credit in SC2; metabolic efficacy is not dismissed because complete mediation is absent. |

Preregistered aggregation: scope-matched. The preregistered scope-matched human-contribution gate SC2 yields Δ / Moderate. Selected controlled human adiposity contrasts cross the in-vivo causal boundary, so a blanket Γ on mouse-to-human inference is not reapplied to the convergent human set. SC1 reaches scoped Ζ for animal transmissibility/rescue; SC3 remains Γ for the endpoint-substitution inference. Weakest-link sensitivity is Γ, primary-subclaim and scope-matched Δ. A stronger universal or primary human-driver reading would remain Γ; that is a separate burden, not a requirement silently added to the registered contribution gate. Sensitivities: weakest-link Γ; primary-subclaim Δ; scope-matched Δ.

### Decisive-perturbation inventory and causal tally

[1] Obese-donor community transfers greater fat accumulation and energy-harvest capacity to germ-free recipients. Crude communities and exceptional recipients do not identify a universal human obesity configuration or exclude diet interaction.

[2] Discordant-twin fecal communities and cultured collections transfer adiposity; lean-community invasion rescues in a diet-dependent manner. Human donors are not human causal recipients; saturated-fat/low-produce diet blocks rescue and donor/cage dependence limits generality.

[3] Human heritable Christensenellaceae associate with low BMI; cultured C.minuta amendment reduces recipient mouse weight gain. Causal attribution applies to mouse amendment; the human twin relation is observational and weight gain is not a human treatment endpoint.

[4] Microbial control of circadian NFIL3 links innate immune signalling to intestinal lipid absorption and body composition. A host pathway in mice does not demonstrate independent human adiposity causation; circadian and dietary conditions matter.

[6] Conventionalization increases body fat despite lower intake; Fiaf knockout experiments implicate host lipid-storage regulation. Germ-free development and host-genotype manipulation limit human inference; later diet/strain counterexperiments constrain universality.

[7] Emulsifier-associated communities transfer inflammation/metabolic phenotype; germ-free contrasts support a microbial contribution. Crude transfer has at most Moderate specificity; exposure may have additional host effects and linked erratum27144359 was not audited in full.

[9] GutBugs randomized 87 adolescents: BMI-SDS primary and total-fat percentage null; android/gynoid ratio lower at6/12/26 weeks. Original Methods use no clinical multiplicity adjustment; secondary fat distribution is not total fat loss, and metabolic-syndrome resolution is post hoc.

[10] Epithelial HDAC3 integrates microbial/circadian cues, induces Cd36/lipid uptake and promotes diet-induced obesity. Host transcriptional perturbation supports a mouse mechanism; human adiposity and effect magnitude were not established.

[11] Randomized lean versus autologous FMT improved six-week clamp insulin sensitivity in metabolic-syndrome men. Insulin sensitivity is not fat accumulation; crude FMT lacks strain attribution and the linked unspecified erratum requires separate audit.

[12] Turicibacter and purified bacterial lipids suppress host ceramides/fat uptake and obesity; dietary fat reduces colonization and continuous dosing restores benefit. The human metagenomic relation is observational; diet-dependent microbial protection does not establish a human pro-obesity configuration.

[13] Jejunal microbiota from high-fat donors increases recipient lipid absorption even on low-fat diet; conditioned media induce epithelial uptake genes. Recipient absorption and organoid activity are not a quantified human body-fat effect; crude transfer specificity remains limited.

[15] FXR genotype and germ-free/conventional contrasts change diet-induced adiposity; donor community transfers phenotype. Host receptor dependence and transfer do not locate an invariant microbial cause of human obesity.

[16] 2019 live/pasteurized Akk pilot improved selected metabolic measures;40enrolled32 completed; weight/fat changes were nonsignificant nearP=.09. Body fat used bioimpedance, not DXA; original text and abstract differ on the fat comparator, but neither demonstrates significant fat loss; calorie/activity detail and baseline insulin balance limited.

[17] Pasteurization and Amuc1100/TLR2-linked protein partly recapitulate metabolic/fat benefits in mice; preliminary human safety was reported. A bioactive nonliving component is not living-community engraftment; preliminary human volunteers belong to the later2019trial family.

[19] Repeated FMT in 61 obese T2D adults improved24 week donor engraftment, the primary endpoint; combined lifestyle arm had lipid/liver-stiffness changes. Engraftment-primary success is not adiposity success; baseline comparisons and cointervention cannot establish independent FMT fat loss.

[21] Ba8145 original Methods/Results report live-cell BMI and killed-cell waist-related between-placebo benefits;126 received intervention of 135 reported participants. MRI visceral-fat between-placebo contrasts are nonsignificant(P=.120/.086 for killed cells), despite significant within-baseline change; statistics/hierarchy in unread supplement and higher killed-arm fiber complicate interpretation; Akk correlation is not mediation.

[23] 75-person BBr60 plus common dietary guidance trial reports greater weight/BMI reduction and metabolic/taxa changes than placebo. Abstract-only fat outcomes, primary hierarchy and multiplicity remain unresolved; metabolomic correlation does not identify the mediator.

[24] L.rhamnosus formulation with prebiotics had an overall ITT weight null but treatment-by-sex interaction; women showed more weight/fat loss over24 weeks. Sex interaction retains scoped causal credit; formulation and restricted diet prevent isolating a strain-only mechanism, and sex-specific primary status remains to be audited.

[25] 2025GutBugs4 year follow-up found lower waist and body-fat percentage among 55returnees, with persistent donor strains and nonsignificant BMI difference. Same trial family; unblinded after26 weeks,37%lost, no missing-data imputation and unadjusted clinical multiplicity; intervening diet/medication unknown.

[26] 20healthy young men on a common high-fat1000 kcal-surplus diet gained less DXA fat and weight with VSL3 than placebo after4 weeks. Direct controlled fat-accumulation contrast supports a bounded human contribution; small/short study, multiple-strain product and unverified engraftment limit mechanism/generalization.

[28] 631obese374control cohort associates Megamonas with obesity; M.rupellensis and heterologous iolG expression promote mouse lipid absorption, with myo-inositol functional support. The human portion is observational and the primary body unavailable; gene/strain perturbations narrow a mouse mechanism without demonstrating human treatment mediation.

[33] PAHSA-treated community transfers insulin sensitivity on high-fat diet; Bt reduces adiposity in female mice, partly dependent on sex hormones. Effect differs by diet and sex; glucose improvement and female fat reduction must not be combined into universal human mediation; linked preprint is same study.

[34] Microbial BSH/strain and bile-acid interventions reduce mouse calorie absorption; human observational and small ketogenic-diet studies show concordant associations. Human diet intervention is not microbial-mediator randomization; bile-acid replacement does not alone prove the community configuration causes human fat gain.

[35] P.distasonis improves ob/ob and high-fat mouse phenotypes; succinate plus secondary bile acids recapitulate metabolic benefits. Model-specific rescue supports function rather than a universal abundance signature; isolated mediator effects on glucose are distinct from all fat endpoints.

[36] B.vulgatus/cholic acid and epithelial Tph1 contrasts implicate a microbe-metabolite-serotonin axis in lipid absorption and high-fat adiposity. Host epithelial and dietary dependencies are integral; human colonization and adiposity mediation are not established.

[37] 2026 low-protein diet browning defect in germ-free mice is rescued by defined consortia; bile-acid/FXR and ammonia/FGF21 axes are nonredundant. Browning genes and dietary compensation are not all clinical fat outcomes; human donor selection is not a human causal trial; linked preprint excluded as duplicate.

[38] Inpatient crossover feeding plus blinded vancomycin intervention met primary stool-energy-loss endpoints in 25 completers. Human energy-harvest perturbation is real bounded evidence, but broad antibiotic effects and short timing do not quantify sustained adiposity or a specific microbial mediator.

[40] Microbial imidazole propionate is enriched in T2D material and impairs mouse glucose/insulin signalling through p38gamma/p62/mTORC1. Glucose pathology is not adiposity; human metabolite and hepatic pathway associations do not independently establish human fat mediation.

[41] Third-trimester community differs from first trimester and transfers more adiposity/insulin insensitivity to germ-free mice. Normal host pregnancy remodels microbes upstream; adaptive physiology and crude cross-species transfer do not establish common human obesity initiation.

[42] Gpr41loss lowers adiposity in conventional and defined Bt/M.smithii-colonized mice despite similar chow intake, with transit/PYY/energy-harvest links. Host receptor, community and transit jointly determine effect; no quantified human fat contribution or universal SCFA direction follows.

[43] Donor-characteristic FMT contrast in 22metabolic-syndrome men found worsened clamp sensitivity with metabolic-syndrome donors and bile/immune changes. Two-week glucose effect is not fat accumulation; no neutral-placebo inference or defined mediator from crude donor material.

[45] Continuous LPS infusion increases fat/weight and liver insulin resistance; CD14mutants resist many LPS/high-fat phenotypes. Artificial infusion and host knockout support a defined inflammatory axis, not endogenous human exposure magnitude or all insulin resistance.

[46] Live Akkermansia reduces high-fat fat gain and barrier/inflammatory abnormalities; heat-killed preparation was ineffective in this study. Heat-killing is not equivalent to later pasteurization; formulation/context matter, and human inverse abundance is associative.

[47] Antibiotics lower endotoxemia, adiposity and inflammation in high-fat/obob mice; CD14deletion partly mimics the phenotype. Antibiotic pleiotropy and shared investigative programme constrain specificity; no human fat or endogenous LPS mediation estimate.

[48] Maternal microbial SCFAs signal through embryonic GPR41/43 and affect later offspring metabolic susceptibility. Developmental maternal mechanism is distinct from adult recipient obesity and not directly generalized to human fat accumulation.

[49] Acetylated cellulose limits body-mass gain through microbiota-dependent host-accessible sugar depletion and altered oxidation. Mouse prebiotic substrate effect is diet-dependent; bacterial carbohydrate consumption can protect rather than increase host energy harvest.

[52] Urinary hippurate associates with human diet/metabolic state; chronic infusion improves glucose tolerance/insulin secretion in high-fat mice. Host conjugation contributes to metabolite origin; glucose benefit is not direct human fat reduction or universal microbial mediation.

[53] 2026 pasteurized-Akk RCT 90 people reports primary24 week weight regain1.2vs3.2 kg(P=.012) after successful low-energy-diet loss. Abstract plus publisher ancillary statements only: fat results, attrition and multiplicity unavailable; nonliving product benefit does not demonstrate living-community causation; company/patent conflicts disclosed.

[54] Lean-donor FMT improves six-week insulin sensitivity in selected baseline communities, with no metabolic effect at18 weeks. Metabolic responder stratification is not prevalidated adiposity selection; transient clamp effect is not sustained fat control.

[56] 70-person FMT/fiber factorial reports six-week HOMA improvement in FMT-low-fermentable arm. Abstract chiefly reports within-armP=.02: significance versus other null arms is not a between-arm treatment effect; original contrasts and adiposity mediation unverified.

[61] Propionate ester reduces threshold-defined weight-gain events and secondary intra-abdominal-fat distribution; acute appetite/hormone perturbation works. Original24 week mean weight differenceP=.099 and total-adipose difference null;49/60 completers and32composition data; ester delivery is not randomized living community.

[63] Seven-day vancomycin versus amoxicillin changes microbes/bile acids and worsens peripheral clamp sensitivity in 20 men. Active-antibiotic comparator and short treatment do not isolate a microbial mediator or establish fat gain.

[64] Germ-free Western-diet protection is lost with Fiafdeletion; AMPK/fatty-acid oxidation supports complementary host mechanisms. Diet/strain counterstudy limits universal Fiaf or microbiota necessity; protection is not endogenous human causal weight evidence.

[65] Rinott aFMT had overall primary regain null(P=.28), but green-Mediterranean interaction(P=.03) and diet-specific benefit(P=.02); Methods confirmITT90 with one imputation. Diet-specific weight result is not uniform FMT or total-fat success; original protocol/supplement pre-specification not independently audited; crude transfer remains nonspecific.

[67] GPR43loss/OE effects disappear with germ-free/antibiotic conditions; conventionalization and acetate rescue restore receptor-dependent fat effects. Original strains differ between KO and OE backgrounds; lifelong manipulation and receptor/tissue specificity limit universal SCFA or human obesity inference.

[68] Cold-community transfer drives browning/fat loss followed by absorptive compensation; Akk co-transfer diminishes intestinal adaptation. Physiological demand, substrate and compensation change the sign/magnitude; a protective transfer is not a universal human fat-producing configuration.

Tally: 46 discriminating contrasts. Null and negative contrasts are discriminators, not positive threshold crossings. Reports and replications are not independent causal lines. Positive flags describe a scoped phenotype/mediator intervention, including animal or non-adiposity endpoints; they are not68 independent causal lines or68 human fat successes. FMT crude-community evidence is at most Moderate specificity even when the trial design is sound. Randomized-cohort follow-up retains a positive observed contrast with attrition/multiplicity limits.

### Cap audit, precedence and Β/Γ/Δ boundary

Rubric04 transfer and rescue thresholds bind the recipient-animal scope. Universal forced-transfer cap5 binds the inference from that experiment to endogenous humans, not a randomized human fat contrast. Defined mouse strains, host pathways and rescue converge on SC1Ζ; independently controlled human adiposity interventions support SC2Δ. They do not demonstrate the same molecule or community function in all humans, so Ε generalization is not met. A metabolically positive trial is not treated as a fat trial. No weakest-leg cap is reapplied to the whole case.

Gamma hard-rule audit: Association-only human microbiome signatures cannot establish causation. Forced transfer crosses the recipient-mouse boundary; it alone cannot establish endogenous human adiposity. Positive controlled human adiposity interventions receive separate causal credit; convergence is not recapped by an abstract-only or observational leg. Insulin sensitivity, engraftment and body weight cannot silently become total-fat outcomes.

SC1 field-valid definition: “
- Gnotobiotic or transfer experiment showing phenotype movement, at minimum
- Even without defined-consortium precision or robust rescue, a transfer paradigm in vivo is the minimum that crosses this threshold

Per **Cap precedence** above, a gnotobiotic or forced-transfer experiment crosses Δ for the recipient-animal subclaim it was actually run in, while universal cap 5 (seeding or forced-transfer paradigms cap at Γ) still blocks the inference from that result to endogenous human biology — score the two scopes separately rather than choosing between the rules.” Positive full-scope result: yes. Transfer, cultured-community/strain amendment and rescue repeatedly move recipient adiposity, with defined functions and host-receptor tests addressing distinct errors [1-4,6-7,10,12-15,17,28,33-37,41-42,45-49,64,67-68]. The broad defined-recipient claim reaches Ζ across distinct mechanisms; this does not promote every individual strain or infer human causality.

SC2 field-valid definition: “
- Gnotobiotic or transfer experiment showing phenotype movement, at minimum
- Even without defined-consortium precision or robust rescue, a transfer paradigm in vivo is the minimum that crosses this threshold

Per **Cap precedence** above, a gnotobiotic or forced-transfer experiment crosses Δ for the recipient-animal subclaim it was actually run in, while universal cap 5 (seeding or forced-transfer paradigms cap at Γ) still blocks the inference from that result to endogenous human biology — score the two scopes separately rather than choosing between the rules.” Positive full-scope result: yes. The Δ crossing rests on a randomized human perturbation with a positive registered primary endpoint in the scoped population: the 2026 pasteurized-Akkermansia weight-maintenance trial [53]. Obese-population strain/formulation trial signals and the randomized FMT fat-distribution contrast add supporting human evidence [9,21,23-25], but GutBugs was null on its BMI-SDS primary and on total body-fat percentage, so its prespecified secondary android/gynoid result, obtained without multiplicity control, supports rather than supplies the crossing. VSL3 high-energy-diet fat-gain attenuation [26] supports human physiology but is outside the registered obese-recipient gate. The 2026 pasteurized-Akk primary weight-maintenance benefit is additional clinical evidence [53], but does not alone establish a living-community or total-fat mechanism. Primary nulls, selected analyses, incomplete mediator attribution and access prevent Ε generalization [5,16,20,22,54-62,65-66]. No universal necessity or primacy burden is imposed on this contributory subclaim.

SC3 field-valid definition: “
- Gnotobiotic or transfer experiment showing phenotype movement, at minimum
- Even without defined-consortium precision or robust rescue, a transfer paradigm in vivo is the minimum that crosses this threshold

Per **Cap precedence** above, a gnotobiotic or forced-transfer experiment crosses Δ for the recipient-animal subclaim it was actually run in, while universal cap 5 (seeding or forced-transfer paradigms cap at Γ) still blocks the inference from that result to endogenous human biology — score the two scopes separately rather than choosing between the rules.” Positive full-scope result: no. Glucose/insulin and inflammatory effects do not by themselves establish body-fat control. Mouse depletion, human engraftment and clamp interventions can change those readouts without corresponding fat changes [11,14,18-19,38,40,43,52,54-59,63]. Positive fat trials retain their own credit in SC2; metabolic efficacy is not dismissed because complete mediation is absent.

### Upward threshold ladder

SC1 Δ: “Requires a transfer paradigm or causal perturbation in vivo.” Satisfied: yes. 

SC1 Ε: “Requires context-specific causal effect with defined limits. A transferable phenotype exists, but microbial identity or generalizability remains incomplete.” Satisfied: yes. 

SC1 Ζ: “Requires:
- Defined consortium or strain-level causal evidence
- Rescue logic
- Replication across at least two relevant models or cohorts” Satisfied: yes. 

SC1 Η: “Requires:
- Defined consortium or factor rescue in multiple contexts
- Narrowed mechanism
- Strong exclusion of diet, housing, batch, and handling confounds” Satisfied: no. To reach Η, reproduce narrowed strain/factor rescue across independent facilities with diet, cage, host background and handling controls; do not combine different pathways into one universal strain mechanism.

SC2 Δ: “Requires a transfer paradigm or causal perturbation in vivo.” Satisfied: yes. 

SC2 Ε: “Requires context-specific causal effect with defined limits. A transferable phenotype exists, but microbial identity or generalizability remains incomplete.” Satisfied: no. For Ε, confirm the bounded human adiposity effect in a prespecified population with fat-primary ITT analysis, validated functional exposure, dietary controls and reproducibility; use mediation and inactive-factor comparisons to narrow mechanism rather than as a prerequisite for giving randomized efficacy credit.

SC3 Δ: “Requires a transfer paradigm or causal perturbation in vivo.” Satisfied: no. In the same prespecified intervention show a positive fat endpoint, temporally ordered mediator engagement and a controlled mechanistic link, instead of substituting insulin or inflammation for fat.

Upper ladder: Defined strain/consortium rescue plus receptor/gene interventions across distinct mouse contexts carry SC1 to Ζ. Human randomized adiposity and primary weight-maintenance trials are field-valid clinical perturbations considered on their own terms; crude transfer, null primary results, endpoint differences and unverified functional mediation prevent a narrowed reproducible SC2Ζ mechanism. Natural experiments and dose/eradication equivalents were considered but no examined human obesity set meets that burden. Higher levels remain live; an unethical human obesity-induction experiment is not demanded.

Independence: SC1Ζ rests on multiple experimental error exclusions rather than article counts: transfer controls human-donor confounding, defined amendment narrows identity, and gene/receptor/metabolite rescue narrows function. Different pathways converge on modifiable recipient adiposity, not one universal strain. Human contribution is graded separately atΔ. GutBugs2020/2025, Mocanu2021/2024, Rinott2020online/2021issue plus2023secondary and preliminary2017/later2019Akk volunteers are report families. FMT studies have separate registrations; overlap elsewhere is unresolved rather than declared absent.

Recipient transfer and ecological rescue [2, 3]: Common recipient environments and lean/defined-strain amendment weaken donor-human lifestyle confounding; diet and cage invasion are explicit rather than assumed absent.

Defined microbial functions and host pathway instruments [28, 12, 67, 37]: Heterologous microbial gene, purified lipid, receptor/acetate rescue and defined-consortium pathway tests address different community-attribution errors; host background, lifelong manipulation and facility limits remain.

Controlled human adiposity and weight outcomes [26, 9, 53, 55, 62]: Randomized clinical exposure supplies human causal information while matched diet, engraftment-null and larger trial-null contrasts constrain generalization; distinct endpoints do not prove one identical pathway.

### Falsification and pitfalls

Target engagement without nominated outcome. FMT-TRIM engraftment occurred without a significant fat or weight effect [55]; synbiotic taxa changed without composition benefit [20]. Additional downgrade: False.

Removal does not prevent phenotype generally. One germ-free diet/strain setting developed more adiposity [60]; short human antibiotics changed ecology without clinically relevant adiposity effects [58]. Additional downgrade: False.

Larger longer test fails primary replication. iPREVENT did not reproduce a mean-weight prevention effect and had a fat-mass null [62], while the earlier trial itself had a mean-weight null [61]. Additional downgrade: False.

Mouse transfer treated as human causality. Human donors remain mouse recipient experiments [2-3,28].

Primary null replaced by selected success. B420PP/posthoc, whole-trial live-Akk null, and aFMT dietary interaction require distinct reporting [5,22,65-66].

Within-arm significance treated as interaction. Mocanu abstract emphasizes within-arm HOMA; original randomized contrast is not independently verified [56].

Body weight or fat distribution renamed total fat. GutBugs, propionate and pasteurized-Akk endpoints differ [9,25,53,61].

Nonliving product efficacy equated with living configuration. Heat-killed/pasteurized bacteria and purified protein/lipids can work without community engraftment [12,17,21,53].

Selected long-term returnees treated as complete randomized follow-up. GutBugs4 year is unblinded,63%recaptured and clinically multiplicity-unadjusted [25].

Microbial metabolite sign assumed universal. GPR41, GPR43, cold compensation and commensal carbohydrate consumption have distinct directions [42,49,67-68].

### Adversarial steelman: above, then below

The strongest case for Ε is a controlled human fat-gain experiment [26], sex- and diet-specific adiposity responses [21,24,65], repeated FMT fat-distribution signals including4 year fat percentage [9,25], and a new primary weight-maintenance success [53]. Defined mouse functions make a merely associative interpretation untenable.

The strongest case for Γ is that prominent FMT and fat-primary trials are null, positive clinical effects are often secondary or selected, and a pasteurized product does not prove live-community causation [5,16,20,22,55,58,61-62]. Mouse transfer alone cannot establish human scope.

Retain Δ for the prespecified human-contribution gate. Reject Γ as a blanket grade that would erase controlled human fat accumulation evidence; reject Ε for the broad mechanism because reproducible specific human adiposity attribution and generalization remain incomplete. Animal SC1Ζ is retained separately, while the insulin-to-fat inference SC3 remains Γ.

### Final score, causal direction and M/S/D

The preregistered scope-matched human-contribution gate SC2 yields Δ / Moderate. Selected controlled human adiposity contrasts cross the in-vivo causal boundary, so a blanket Γ on mouse-to-human inference is not reapplied to the convergent human set. SC1 reaches scoped Ζ for animal transmissibility/rescue; SC3 remains Γ for the endpoint-substitution inference. Weakest-link sensitivity is Γ, primary-subclaim and scope-matched Δ. A stronger universal or primary human-driver reading would remain Γ; that is a separate burden, not a requirement silently added to the registered contribution gate.

The human interventions can modify fat gain or weight maintenance under upstream diet/host conditions. They do not establish universal necessity, initiation and rate limitation together. A changed microbial marker may also be downstream of diet or host physiology, but the demonstrated intervention effects preclude classifying the entire mechanism as a passive epiphenomenon.

Momentum: Mechanistic only. The 2026 primary weight-maintenance trial [53] and 2024-2026 strain/gene/lipid/consortium studies [12,28,34,36-37,49] improve experimental specificity; new results include context and endpoint limits.

Stability: Stable. No verified pending trial is used to forecast a score change. Independently replicated fat-primary, strain-function and mediation trials could change the level; their results are not assumed.

Divergence: Contested; no single implied-level arithmetic. These dimensions do not change level or confidence.

## PROPOSED DECISIVE EXPERIMENTS

Randomize a prespecified human obesity population to a defined live strain/consortium, an otherwise matched inactive microbial preparation and placebo, under matched dietary exposure. Use a registered fat-mass primary ITT endpoint over6-12 months, separate maintenance from initial loss, and measure strain/function exposure, energy intake/loss, expenditure and candidate mediators longitudinally. Confirm the same function with subtraction/addition and rescue in two independent recipient facilities; replicate clinical effects in a separate cohort.

Readout: DXA total fat mass and validated visceral fat, with distinct body weight, energy absorption, clamp and hormone outcomes; predefined between-arm contrasts and mediator timing. Controls: Allocation concealment, blinding, full ITT/missing-data sensitivity, inactive-factor control, stable medications, stratified baseline microbiome/sex/diet, exposure/engraftment measures, clinical multiplicity and cage/diet-batch controls. Supporting: A prespecified clinically meaningful fat difference with confirmed microbial functional exposure and independently replicated causal pathway, beyond matched dietary exposure without assuming unchanged appetite is required. Falsifying: Adequate exposure with a precise fat-primary null across prespecified contexts, inactive product matching live efficacy without the nominated function, or loss of the effect after dietary/host and missing-data controls.

## RESEARCH TIMELINE

| Year | Milestone | Primary source | Interpretation |
|---|---|---|---|
| 2004 | Conventionalization and Fiaf-linked fat storage | [PMID15505215](https://pubmed.ncbi.nlm.nih.gov/15505215/) | Causal recipient model |
| 2006 | Obese-community energy-harvest transfer | [PMID17183312](https://pubmed.ncbi.nlm.nih.gov/17183312/) | Crude transfer; human scope remains separate |
| 2007 | LPS/CD14 adiposity instrument | [PMID17456850](https://pubmed.ncbi.nlm.nih.gov/17456850/) | Defined inflammatory route |
| 2010 | Germ-free high-fat counterexample | [PMID20441670](https://pubmed.ncbi.nlm.nih.gov/20441670/) | Diet and host boundary |
| 2012 | Lean FMT clamp benefit | [PMID22728514](https://pubmed.ncbi.nlm.nih.gov/22728514/) | Insulin sensitivity rather than fat |
| 2013 | Discordant twin cultured-community transfer | [PMID24009397](https://pubmed.ncbi.nlm.nih.gov/24009397/) | Diet-dependent rescue |
| 2013 | GPR43 microbial/acetate rescue | [PMID23652017](https://pubmed.ncbi.nlm.nih.gov/23652017/) | Protective substrate signalling |
| 2014 | Christensenella amendment | [PMID25417156](https://pubmed.ncbi.nlm.nih.gov/25417156/) | Defined mouse strain contribution |
| 2015 | Controlled human probiotic fat-gain attenuation | [PMID26466123](https://pubmed.ncbi.nlm.nih.gov/26466123/) | Small four-week high-energy-diet context |
| 2015 | Propionate appetite and fat-distribution trial | [PMID25500202](https://pubmed.ncbi.nlm.nih.gov/25500202/) | Mean weight and total adipose nulls retained |
| 2016 | B420 fat-primary trial | [PMID27810310](https://pubmed.ncbi.nlm.nih.gov/27810310/) | ITT null; PP/posthoc signals |
| 2019 | Live/pasteurized Akk pilot | [PMID31263284](https://pubmed.ncbi.nlm.nih.gov/31263284/) | Metabolic positives; adiposity nonsignificant |
| 2020 | FMT-TRIM | [PMID32150549](https://pubmed.ncbi.nlm.nih.gov/32150549/) | Engraftment without significant fat/clamp benefit |
| 2020 | GutBugs primary report | [PMID33346848](https://pubmed.ncbi.nlm.nih.gov/33346848/) | BMI/total fat null; distribution signal |
| 2020online/2021issue | Diet-modulated aFMT | [PMID32860791](https://pubmed.ncbi.nlm.nih.gov/32860791/) | Overall regain null; diet interaction |
| 2021 | FMT plus fiber | [PMID34226737](https://pubmed.ncbi.nlm.nih.gov/34226737/) | HOMA endpoint; within-arm contrast caution |
| 2024 | Megamonas/iolG lipid-absorption mechanism | [PMID38996548](https://pubmed.ncbi.nlm.nih.gov/38996548/) | Human association, causal mouse gene |
| 2024 | iPREVENT | [PMID39391015](https://pubmed.ncbi.nlm.nih.gov/39391015/) | Larger12 month weight/fat null |
| 2025 | Live-Akk baseline-level trial | [PMID39879980](https://pubmed.ncbi.nlm.nih.gov/39879980/) | Whole-trial null; subgroup signal |
| 2025 | GutBugs4 year follow-up | [PMID40877311](https://pubmed.ncbi.nlm.nih.gov/40877311/) | Positive fat percentage with attrition/unblinding |
| 2026 | Pasteurized-Akk maintenance trial | [PMID42120725](https://pubmed.ncbi.nlm.nih.gov/42120725/) | Primary weight-regain benefit; fat details unavailable |
| 2026 | Turicibacter lipid mechanism | [PMID41197631](https://pubmed.ncbi.nlm.nih.gov/41197631/) | Diet-dependent protective microbial function |
| 2026 | Defined consortia and adipose browning | [PMID41781619](https://pubmed.ncbi.nlm.nih.gov/41781619/) | Host/diet mechanistic context |

## KNOWLEDGE GAPS

Replicate fat-primary human outcomes with prespecified between-arm contrasts rather than secondary, compliance-selected or response-defined analyses.

Separate living-community changes from the activity of nonliving cells, purified proteins or bacterial lipids using matched inactive-component controls.

Validate baseline taxon/function treatment selectors prospectively with interaction tests; within-arm significance and posthoc responders do not establish selection.

Quantify human energy intake/loss, expenditure and adiposity over time to test compensation and distinguish mechanism from short-term stool calorimetry.

Replicate strain/substrate rescue across facilities with cage, vendor, diet batch, host background and handling controls; do not universalize SCFA direction.

Complete original Methods, protocols, supplements and correction checks for newly positive clinical trials, including2026Akk fat outcomes and allocation/attrition; audit commercial conflicts and shared cohorts.

Maintain long-term blinded follow-up with attrition sensitivity and multiplicity control; GutBugs4 year findings need independent replication.

## METHODS AND CONSENSUS CONTRIBUTION

Consensus was used through the signed-in browser for discovery, counterevidence, research-gap and timeline suggestions, and checks of specific indexed full-text passages. Extracted Methods/Results details were reconciled with original-paper passages where available. Provider screening counts, citation counts and meters were not used as primary counts or HSEF scores. This source log records access depth; it does not claim complete-paper or supplement review. The monthly API allowance was conserved: zero Consensus API calls in this batch.

## REVIEWER NOTES

68 primary reports, ten selected original partial bodies and zero complete-paper/supplement audits. Exact archived HSEF v5.3. The ranking/index prior was visible; limited unblinded calibration ordering is disclosed. Raw search candidates and secondary leads are not examinations; copied cohorts are not independent lines. Private review package, not published.

Search leads, reviews, protocols and correction notices do not pad the primary count. Downloaded XML does not earn examination credit. Linked notices remain attached; unexamined correction details limit inference. This is a coordinator source/claim consistency assessment, not independent blinded calibration.

## ADVERSARIAL REVIEW AND APPROVAL

**Authoring model:** Sol 6.1. **Adversarial reviewer:** Claude Opus 5.5 (Anthropic). **Date:** 2026-10-06.

**Scope of checks.** All 68 reference PMIDs re-verified against PubMed records (title, first author, journal, year, DOI, status flags): 68/68 resolve to the stated work; no retractions or expressions of concern; year differences are print-issue versus e-pub year, and two apparent title mismatches are PubMed dropping italicised gene or strain names (NFIL3, Muc-T). 20 load-bearing claims spot-checked against PubMed abstracts and the local examination records, including the VSL#3 fat-gain figures, the 2026 pasteurized-Akkermansia primary result, the GutBugs primary and secondary endpoints and its four-year follow-up, the B420 ITT-versus-per-protocol analysis, FMT-TRIM, the two propionate trials, the aFMT diet-interaction trial, the Akkermansia pilot and the 2025 baseline-dependent trial, the vancomycin stool-calorie study and the Jumpertz energy-harvest association. Scoring logic re-derived from `framework/hsef-hypothesis-review-v5/SKILL.md` and `references/hsef-framework.md`: cap order and cap precedence, the index-prescribed scope-matched aggregation, the Β/Γ/Δ boundary including its requirement that the qualifying perturbation's result be positive, the Ζ independence basis for SC1, and the Δ-versus-Ε distinction for an unmapped boundary.

| # | Severity | Location | Original → corrected | Evidence |
|---|---|---|---|---|
| 1 | Minor | Subclaim matrix SC2 rationale and the matching score-card field | "Obese-population strain/formulation trial signals and randomized FMT fat-distribution contrasts justify Δ contribution [9,21,23-25]" → the Δ crossing is attributed to the randomized trial with a positive registered primary endpoint in the scoped population (PMID 42120725), with the GutBugs result named as supporting evidence whose primary endpoint was null | Framework Step 9, item 5: "A study of exactly the right design that returns a null or contradicts the hypothesis does not cross Δ." PMID 33346848 reports no effect on its BMI-SDS primary or on total body-fat percentage; the android-to-gynoid ratio is a prespecified secondary without multiplicity adjustment. The review's own pitfall list already flags secondary-for-primary substitution. SC2 and the headline remain Δ because PMID 42120725 supplies a positive primary-endpoint human perturbation in the scoped population. |
| 2 | Minor | AT A GLANCE table; "Momentum:" line; `extended_scoring_MSD.momentum.value` | Momentum "Forward" → "Mechanistic only" | "Forward" is not a value on the framework's Momentum scale (Decisive work underway / Mechanistic only / Associational only / Little new work / Abandoned); the recorded artifact names completed specificity-improving work, and Stability states that no pending result is relied on. M/S/D never modify the level or confidence. |

**Rejected findings (checked, no change).** (a) Headline Δ — this is the highest level in the batch and was re-derived independently: the registered gate SC2 is scope-matched to the human claim, and a randomized human microbial perturbation with a positive registered primary weight outcome in adults with overweight or obesity crosses the field-valid boundary, while universal-configuration and primacy readings stay at Γ as the review states. (b) Δ rather than Ε for SC2 — the context boundary has not been mapped by internal negative controls, and under the framework's Step 10 rule untested generalization is Δ, not Ε. (c) SC1 at Ζ — transfer, defined-strain amendment and receptor or gene rescue exclude different systematic errors, which is the framework's independence test; the review also states that this does not transfer to human scope. (d) Treating the pasteurized organism as within the registered mechanism — the registered claim names microbial products and endocrine signalling, and the review separately records that a non-living product does not establish a living-community mechanism. (e) Citation years differing by one from the PubMed e-pub year (PMIDs 19043404, 26740296, 27892954, 31301451, 33785557, 24336217, 31147381, 25500202, 24316517, 32860791, 41197631 and others) — print/issue-year convention. (f) Trial figures checked against sources (0.63 vs 1.29 kg; 1.2 vs 3.2 kg, P = 0.012; ITT n = 209; 226 at 12 months; 55 of 87 returnees) all match.

**Residual limitations.** Access remains bounded: 10 partial-body and 58 abstract/status examinations with no complete-paper or supplement audit; several positive formulation trials and the 2026 maintenance trial were examined at abstract depth, so their endpoint hierarchies, multiplicity handling and total-fat outcomes remain unresolved, and commercial disclosures on the 2026 trial are recorded but not adjudicated. This pass checked citation integrity, attribution and framework conformance; it is not independent blinded peer review.

**Approval:** Following adversarial review and the corrections listed above, the revised document is approved. — Claude Opus 5.5 (Anthropic), 2026-10-06.

## REFERENCES

1. Turnbaugh PJ, Ley RE, Mahowald MA et al. An obesity-associated gut microbiome with increased capacity for energy harvest. Nature. 2006. [PMID 17183312](https://pubmed.ncbi.nlm.nih.gov/17183312/). DOI: 10.1038/nature05414. Access: abstract_metadata. Study: PMID17183312.

2. Ridaura VK, Faith JJ, Rey FE et al. Gut microbiota from twins discordant for obesity modulate metabolism in mice. Science (New York, N.Y.). 2013. [PMID 24009397](https://pubmed.ncbi.nlm.nih.gov/24009397/). DOI: 10.1126/science.1241214. Access: abstract_metadata. Study: PMID24009397.

3. Goodrich JK, Waters JL, Poole AC et al. Human genetics shape the gut microbiome. Cell. 2014. [PMID 25417156](https://pubmed.ncbi.nlm.nih.gov/25417156/). DOI: 10.1016/j.cell.2014.09.053. Access: abstract_metadata. Study: PMID25417156.

4. Wang Y, Kuang Z, Yu X et al. The intestinal microbiota regulates body composition through NFIL3 and the circadian clock. Science (New York, N.Y.). 2017. [PMID 28860383](https://pubmed.ncbi.nlm.nih.gov/28860383/). DOI: 10.1126/science.aan0677. Access: abstract_metadata. Study: PMID28860383.

5. Zhang Y, Liu R, Chen Y et al. Akkermansia muciniphila supplementation in patients with overweight/obese type 2 diabetes: Efficacy depends on its baseline levels in the gut. Cell metabolism. 2025. [PMID 39879980](https://pubmed.ncbi.nlm.nih.gov/39879980/). DOI: 10.1016/j.cmet.2024.12.010. Access: abstract_metadata. Study: Live-Akk-NCT04797442.

6. Bäckhed F, Ding H, Wang T et al. The gut microbiota as an environmental factor that regulates fat storage. Proceedings of the National Academy of Sciences of the United States of America. 2004. [PMID 15505215](https://pubmed.ncbi.nlm.nih.gov/15505215/). DOI: 10.1073/pnas.0407076101. Access: abstract_metadata. Study: PMID15505215.

7. Chassaing B, Koren O, Goodrich JK et al. Dietary emulsifiers impact the mouse gut microbiota promoting colitis and metabolic syndrome. Nature. 2015. [PMID 25731162](https://pubmed.ncbi.nlm.nih.gov/25731162/). DOI: 10.1038/nature14232. Access: abstract_metadata. Study: PMID25731162.

8. Turnbaugh PJ, Hamady M, Yatsunenko T et al. A core gut microbiome in obese and lean twins. Nature. 2009. [PMID 19043404](https://pubmed.ncbi.nlm.nih.gov/19043404/). DOI: 10.1038/nature07540. Access: abstract_metadata. Study: PMID19043404.

9. Leong KSW, Jayasinghe TN, Wilson BC et al. Effects of Fecal Microbiome Transfer in Adolescents With Obesity: The Gut Bugs Randomized Controlled Trial. JAMA network open. 2020. [PMID 33346848](https://pubmed.ncbi.nlm.nih.gov/33346848/). DOI: 10.1001/jamanetworkopen.2020.30415. Access: full_text_partial. Study: GutBugs-ACTRN12615001351505.

10. Kuang Z, Wang Y, Li Y et al. The intestinal microbiota programs diurnal rhythms in host metabolism through histone deacetylase 3. Science (New York, N.Y.). 2019. [PMID 31604271](https://pubmed.ncbi.nlm.nih.gov/31604271/). DOI: 10.1126/science.aaw3134. Access: abstract_metadata. Study: PMID31604271.

11. Vrieze A, Van Nood E, Holleman F et al. Transfer of intestinal microbiota from lean donors increases insulin sensitivity in individuals with metabolic syndrome. Gastroenterology. 2012. [PMID 22728514](https://pubmed.ncbi.nlm.nih.gov/22728514/). DOI: 10.1053/j.gastro.2012.06.031. Access: abstract_metadata. Study: Lean-FMT-NTR1776.

12. Klag K, Ott D, Tippetts TS et al. Dietary fat disrupts a commensal-host lipid network that promotes metabolic health. Cell metabolism. 2026. [PMID 41197631](https://pubmed.ncbi.nlm.nih.gov/41197631/). DOI: 10.1016/j.cmet.2025.10.007. Access: abstract_metadata. Study: PMID41197631.

13. Martinez-Guryn K, Hubert N, Frazier K et al. Small Intestine Microbiota Regulate Host Digestive and Absorptive Adaptive Responses to Dietary Lipids. Cell host & microbe. 2018. [PMID 29649441](https://pubmed.ncbi.nlm.nih.gov/29649441/). DOI: 10.1016/j.chom.2018.03.011. Access: abstract_metadata. Study: PMID29649441.

14. Zarrinpar A, Chaix A, Xu ZZ et al. Antibiotic-induced microbiome depletion alters metabolic homeostasis by affecting gut signaling and colonic metabolism. Nature communications. 2018. [PMID 30030441](https://pubmed.ncbi.nlm.nih.gov/30030441/). DOI: 10.1038/s41467-018-05336-9. Access: abstract_metadata. Study: PMID30030441.

15. Parséus A, Sommer N, Sommer F et al. Microbiota-induced obesity requires farnesoid X receptor. Gut. 2017. [PMID 26740296](https://pubmed.ncbi.nlm.nih.gov/26740296/). DOI: 10.1136/gutjnl-2015-310283. Access: abstract_metadata. Study: PMID26740296.

16. Depommier C, Everard A, Druart C et al. Supplementation with Akkermansia muciniphila in overweight and obese human volunteers: a proof-of-concept exploratory study. Nature medicine. 2019. [PMID 31263284](https://pubmed.ncbi.nlm.nih.gov/31263284/). DOI: 10.1038/s41591-019-0495-2. Access: full_text_partial. Study: Akk-NCT02637115.

17. Plovier H, Everard A, Druart C et al. A purified membrane protein from Akkermansia muciniphila or the pasteurized bacterium improves metabolism in obese and diabetic mice. Nature medicine. 2017. [PMID 27892954](https://pubmed.ncbi.nlm.nih.gov/27892954/). DOI: 10.1038/nm.4236. Access: abstract_metadata. Study: Akk-preliminary-human-overlaps-NCT02637115-plus-mouse.

18. Allegretti JR, Kassam Z, Mullish BH et al. Effects of Fecal Microbiota Transplantation With Oral Capsules in Obese Patients. Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association. 2020. [PMID 31301451](https://pubmed.ncbi.nlm.nih.gov/31301451/). DOI: 10.1016/j.cgh.2019.07.006. Access: abstract_metadata. Study: Allegretti-FMT-NCT02741518.

19. Ng SC, Xu Z, Mak JWY et al. Microbiota engraftment after faecal microbiota transplantation in obese subjects with type 2 diabetes: a 24-week, double-blind, randomised controlled trial. Gut. 2022. [PMID 33785557](https://pubmed.ncbi.nlm.nih.gov/33785557/). DOI: 10.1136/gutjnl-2020-323617. Access: abstract_metadata. Study: Repeated-FMT-NCT03127696.

20. Sergeev IN, Aljutaily T, Walton G et al. Effects of Synbiotic Supplement on Human Gut Microbiota, Body Composition and Weight Loss in Obesity. Nutrients. 2020. [PMID 31952249](https://pubmed.ncbi.nlm.nih.gov/31952249/). DOI: 10.3390/nu12010222. Access: abstract_metadata. Study: PMID31952249.

21. Pedret A, Valls RM, Calderón-Pérez L et al. Effects of daily consumption of the probiotic Bifidobacterium animalis subsp. lactis CECT 8145 on anthropometric adiposity biomarkers in abdominally obese subjects: a randomized controlled trial. International journal of obesity (2005). 2019. [PMID 30262813](https://pubmed.ncbi.nlm.nih.gov/30262813/). DOI: 10.1038/s41366-018-0220-0. Access: full_text_partial. Study: PMID30262813.

22. Stenman LK, Lehtinen MJ, Meland N et al. Probiotic With or Without Fiber Controls Body Fat Mass, Associated With Serum Zonulin, in Overweight and Obese Adults-Randomized Controlled Trial. EBioMedicine. 2016. [PMID 27810310](https://pubmed.ncbi.nlm.nih.gov/27810310/). DOI: 10.1016/j.ebiom.2016.10.036. Access: full_text_partial. Study: B420-NCT01978691.

23. Bai Z, Wu Y, Gao D et al. Gut Microbiome and Metabolome Alterations in Overweight or Obese Adult Population after Weight-Loss Bifidobacterium breve BBr60 Intervention: A Randomized Controlled Trial. International journal of molecular sciences. 2024. [PMID 39456659](https://pubmed.ncbi.nlm.nih.gov/39456659/). DOI: 10.3390/ijms252010871. Access: abstract_metadata. Study: PMID39456659.

24. Sanchez M, Darimont C, Drapeau V et al. Effect of Lactobacillus rhamnosus CGMCC1.3724 supplementation on weight loss and maintenance in obese men and women. The British journal of nutrition. 2014. [PMID 24299712](https://pubmed.ncbi.nlm.nih.gov/24299712/). DOI: 10.1017/S0007114513003875. Access: abstract_metadata. Study: PMID24299712.

25. Wilson BC, Zuppi M, Derraik JGB et al. Long-term health outcomes in adolescents with obesity treated with faecal microbiota transplantation: 4-year follow-up. Nature communications. 2025. [PMID 40877311](https://pubmed.ncbi.nlm.nih.gov/40877311/). DOI: 10.1038/s41467-025-62752-4. Access: full_text_partial. Study: GutBugs-ACTRN12615001351505.

26. Osterberg KL, Boutagy NE, McMillan RP et al. Probiotic supplementation attenuates increases in body mass and fat mass during high-fat diet in healthy young adults. Obesity (Silver Spring, Md.). 2015. [PMID 26466123](https://pubmed.ncbi.nlm.nih.gov/26466123/). DOI: 10.1002/oby.21230. Access: abstract_metadata. Study: PMID26466123.

27. Asnicar F, Berry SE, Valdes AM et al. Microbiome connections with host metabolism and habitual diet from 1,098 deeply phenotyped individuals. Nature medicine. 2021. [PMID 33432175](https://pubmed.ncbi.nlm.nih.gov/33432175/). DOI: 10.1038/s41591-020-01183-8. Access: abstract_metadata. Study: PMID33432175.

28. Wu C, Yang F, Zhong H et al. Obesity-enriched gut microbe degrades myo-inositol and promotes lipid absorption. Cell host & microbe. 2024. [PMID 38996548](https://pubmed.ncbi.nlm.nih.gov/38996548/). DOI: 10.1016/j.chom.2024.06.012. Access: abstract_metadata. Study: PMID38996548.

29. Vangay P, Johnson AJ, Ward TL et al. US Immigration Westernizes the Human Gut Microbiome. Cell. 2018. [PMID 30388453](https://pubmed.ncbi.nlm.nih.gov/30388453/). DOI: 10.1016/j.cell.2018.10.029. Access: abstract_metadata. Study: PMID30388453.

30. Meslier V, Laiola M, Roager HM et al. Mediterranean diet intervention in overweight and obese subjects lowers plasma cholesterol and causes changes in the gut microbiome and metabolome independently of energy intake. Gut. 2020. [PMID 32075887](https://pubmed.ncbi.nlm.nih.gov/32075887/). DOI: 10.1136/gutjnl-2019-320438. Access: abstract_metadata. Study: PMID32075887.

31. Rothschild D, Weissbrod O, Barkan E et al. Environment dominates over host genetics in shaping human gut microbiota. Nature. 2018. [PMID 29489753](https://pubmed.ncbi.nlm.nih.gov/29489753/). DOI: 10.1038/nature25973. Access: abstract_metadata. Study: PMID29489753.

32. Le Chatelier E, Nielsen T, Qin J et al. Richness of human gut microbiome correlates with metabolic markers. Nature. 2013. [PMID 23985870](https://pubmed.ncbi.nlm.nih.gov/23985870/). DOI: 10.1038/nature12506. Access: abstract_metadata. Study: PMID23985870.

33. Lee J, Wellenstein K, Rahnavard A et al. Beneficial metabolic effects of PAHSAs depend on the gut microbiota in diet-induced obese mice but not in chow-fed mice. Proceedings of the National Academy of Sciences of the United States of America. 2024. [PMID 38968121](https://pubmed.ncbi.nlm.nih.gov/38968121/). DOI: 10.1073/pnas.2318691121. Access: abstract_metadata. Study: PMID38968121.

34. Li X, Yang J, Zhou X et al. Ketogenic diet-induced bile acids protect against obesity through reduced calorie absorption. Nature metabolism. 2024. [PMID 38937659](https://pubmed.ncbi.nlm.nih.gov/38937659/). DOI: 10.1038/s42255-024-01072-1. Access: abstract_metadata. Study: PMID38937659.

35. Wang K, Liao M, Zhou N et al. Parabacteroides distasonis Alleviates Obesity and Metabolic Dysfunctions via Production of Succinate and Secondary Bile Acids. Cell reports. 2019. [PMID 30605678](https://pubmed.ncbi.nlm.nih.gov/30605678/). DOI: 10.1016/j.celrep.2018.12.028. Access: abstract_metadata. Study: PMID30605678.

36. Wen X, Feng X, Xin F et al. B. vulgatus ameliorates high-fat diet-induced obesity through modulating intestinal serotonin synthesis and lipid absorption in mice. Gut microbes. 2024. [PMID 39569932](https://pubmed.ncbi.nlm.nih.gov/39569932/). DOI: 10.1080/19490976.2024.2423040. Access: abstract_metadata. Study: PMID39569932.

37. Tanoue T, Nagayama M, Roochana AJA et al. Microbiota-mediated induction of beige adipocytes in response to dietary cues. Nature. 2026. [PMID 41781619](https://pubmed.ncbi.nlm.nih.gov/41781619/). DOI: 10.1038/s41586-026-10205-3. Access: abstract_metadata. Study: PMID41781619.

38. Basolo A, Hohenadel M, Ang QY et al. Effects of underfeeding and oral vancomycin on gut microbiome and nutrient absorption in humans. Nature medicine. 2020. [PMID 32235930](https://pubmed.ncbi.nlm.nih.gov/32235930/). DOI: 10.1038/s41591-020-0801-z. Access: abstract_metadata. Study: Absorption-NCT02037295.

39. David LA, Maurice CF, Carmody RN et al. Diet rapidly and reproducibly alters the human gut microbiome. Nature. 2014. [PMID 24336217](https://pubmed.ncbi.nlm.nih.gov/24336217/). DOI: 10.1038/nature12820. Access: abstract_metadata. Study: PMID24336217.

40. Koh A, Molinaro A, Ståhlman M et al. Microbially Produced Imidazole Propionate Impairs Insulin Signaling through mTORC1. Cell. 2018. [PMID 30401435](https://pubmed.ncbi.nlm.nih.gov/30401435/). DOI: 10.1016/j.cell.2018.09.055. Access: abstract_metadata. Study: PMID30401435.

41. Koren O, Goodrich JK, Cullender TC et al. Host remodeling of the gut microbiome and metabolic changes during pregnancy. Cell. 2012. [PMID 22863002](https://pubmed.ncbi.nlm.nih.gov/22863002/). DOI: 10.1016/j.cell.2012.07.008. Access: abstract_metadata. Study: PMID22863002.

42. Samuel BS, Shaito A, Motoike T et al. Effects of the gut microbiota on host adiposity are modulated by the short-chain fatty-acid binding G protein-coupled receptor, Gpr41. Proceedings of the National Academy of Sciences of the United States of America. 2008. [PMID 18931303](https://pubmed.ncbi.nlm.nih.gov/18931303/). DOI: 10.1073/pnas.0808567105. Access: abstract_metadata. Study: PMID18931303.

43. de Groot P, Scheithauer T, Bakker GJ et al. Donor metabolic characteristics drive effects of faecal microbiota transplantation on recipient insulin sensitivity, energy expenditure and intestinal transit time. Gut. 2020. [PMID 31147381](https://pubmed.ncbi.nlm.nih.gov/31147381/). DOI: 10.1136/gutjnl-2019-318320. Access: abstract_metadata. Study: Donor-FMT-NTR4327.

44. Ley RE, Bäckhed F, Turnbaugh P et al. Obesity alters gut microbial ecology. Proceedings of the National Academy of Sciences of the United States of America. 2005. [PMID 16033867](https://pubmed.ncbi.nlm.nih.gov/16033867/). DOI: 10.1073/pnas.0504978102. Access: abstract_metadata. Study: PMID16033867.

45. Cani PD, Amar J, Iglesias MA et al. Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes. 2007. [PMID 17456850](https://pubmed.ncbi.nlm.nih.gov/17456850/). DOI: 10.2337/db06-1491. Access: abstract_metadata. Study: PMID17456850.

46. Everard A, Belzer C, Geurts L et al. Cross-talk between Akkermansia muciniphila and intestinal epithelium controls diet-induced obesity. Proceedings of the National Academy of Sciences of the United States of America. 2013. [PMID 23671105](https://pubmed.ncbi.nlm.nih.gov/23671105/). DOI: 10.1073/pnas.1219451110. Access: abstract_metadata. Study: PMID23671105.

47. Cani PD, Bibiloni R, Knauf C et al. Changes in gut microbiota control metabolic endotoxemia-induced inflammation in high-fat diet-induced obesity and diabetes in mice. Diabetes. 2008. [PMID 18305141](https://pubmed.ncbi.nlm.nih.gov/18305141/). DOI: 10.2337/db07-1403. Access: abstract_metadata. Study: PMID18305141.

48. Kimura I, Miyamoto J, Ohue-Kitano R et al. Maternal gut microbiota in pregnancy influences offspring metabolic phenotype in mice. Science (New York, N.Y.). 2020. [PMID 32108090](https://pubmed.ncbi.nlm.nih.gov/32108090/). DOI: 10.1126/science.aaw8429. Access: abstract_metadata. Study: PMID32108090.

49. Takeuchi T, Miyauchi E, Nakanishi Y et al. Acetylated cellulose suppresses body mass gain through gut commensals consuming host-accessible carbohydrates. Cell metabolism. 2025. [PMID 40381616](https://pubmed.ncbi.nlm.nih.gov/40381616/). DOI: 10.1016/j.cmet.2025.04.013. Access: abstract_metadata. Study: PMID40381616.

50. Nagata N, Nishijima S, Miyoshi-Akiyama T et al. Population-level Metagenomics Uncovers Distinct Effects of Multiple Medications on the Human Gut Microbiome. Gastroenterology. 2022. [PMID 35788347](https://pubmed.ncbi.nlm.nih.gov/35788347/). DOI: 10.1053/j.gastro.2022.06.070. Access: abstract_metadata. Study: PMID35788347.

51. Xiao L, Sonne SB, Feng Q et al. High-fat feeding rather than obesity drives taxonomical and functional changes in the gut microbiota in mice. Microbiome. 2017. [PMID 28390422](https://pubmed.ncbi.nlm.nih.gov/28390422/). DOI: 10.1186/s40168-017-0258-6. Access: abstract_metadata. Study: PMID28390422.

52. Brial F, Chilloux J, Nielsen T et al. Human and preclinical studies of the host-gut microbiome co-metabolite hippurate as a marker and mediator of metabolic health. Gut. 2021. [PMID 33975870](https://pubmed.ncbi.nlm.nih.gov/33975870/). DOI: 10.1136/gutjnl-2020-323314. Access: abstract_metadata. Study: PMID33975870.

53. Mount S, Canfora EE, Jocken JW et al. Pasteurized Akkermansia muciniphila MucT for weight loss maintenance in people with overweight and obesity: a controlled randomized trial. Nature medicine. 2026. [PMID 42120725](https://pubmed.ncbi.nlm.nih.gov/42120725/). DOI: 10.1038/s41591-026-04394-7. Access: abstract_metadata. Study: Akk-maintenance-NCT05417360.

54. Kootte RS, Levin E, Salojärvi J et al. Improvement of Insulin Sensitivity after Lean Donor Feces in Metabolic Syndrome Is Driven by Baseline Intestinal Microbiota Composition. Cell metabolism. 2017. [PMID 28978426](https://pubmed.ncbi.nlm.nih.gov/28978426/). DOI: 10.1016/j.cmet.2017.09.008. Access: abstract_metadata. Study: PMID28978426.

55. Yu EW, Gao L, Stastka P et al. Fecal microbiota transplantation for the improvement of metabolism in obesity: The FMT-TRIM double-blind placebo-controlled pilot trial. PLoS medicine. 2020. [PMID 32150549](https://pubmed.ncbi.nlm.nih.gov/32150549/). DOI: 10.1371/journal.pmed.1003051. Access: full_text_partial. Study: FMT-TRIM-NCT02530385.

56. Mocanu V, Zhang Z, Deehan EC et al. Fecal microbial transplantation and fiber supplementation in patients with severe obesity and metabolic syndrome: a randomized double-blind, placebo-controlled phase 2 trial. Nature medicine. 2021. [PMID 34226737](https://pubmed.ncbi.nlm.nih.gov/34226737/). DOI: 10.1038/s41591-021-01399-2. Access: abstract_metadata. Study: Mocanu-NCT03477916.

57. Zhang Z, Mocanu V, Deehan EC et al. Recipient microbiome-related features predicting metabolic improvement following fecal microbiota transplantation in adults with severe obesity and metabolic syndrome: a secondary analysis of a phase 2 clinical trial. Gut microbes. 2024. [PMID 38685731](https://pubmed.ncbi.nlm.nih.gov/38685731/). DOI: 10.1080/19490976.2024.2345134. Access: abstract_metadata. Study: Mocanu-NCT03477916.

58. Reijnders D, Goossens GH, Hermes GD et al. Effects of Gut Microbiota Manipulation by Antibiotics on Host Metabolism in Obese Humans: A Randomized Double-Blind Placebo-Controlled Trial. Cell metabolism. 2016. [PMID 27411009](https://pubmed.ncbi.nlm.nih.gov/27411009/). DOI: 10.1016/j.cmet.2016.06.016. Access: abstract_metadata. Study: PMID27411009.

59. Jumpertz R, Le DS, Turnbaugh PJ et al. Energy-balance studies reveal associations between gut microbes, caloric load, and nutrient absorption in humans. The American journal of clinical nutrition. 2011. [PMID 21543530](https://pubmed.ncbi.nlm.nih.gov/21543530/). DOI: 10.3945/ajcn.110.010132. Access: abstract_metadata. Study: PMID21543530.

60. Fleissner CK, Huebel N, Abd El-Bary MM et al. Absence of intestinal microbiota does not protect mice from diet-induced obesity. The British journal of nutrition. 2010. [PMID 20441670](https://pubmed.ncbi.nlm.nih.gov/20441670/). DOI: 10.1017/S0007114510001303. Access: abstract_metadata. Study: PMID20441670.

61. Chambers ES, Viardot A, Psichas A et al. Effects of targeted delivery of propionate to the human colon on appetite regulation, body weight maintenance and adiposity in overweight adults. Gut. 2015. [PMID 25500202](https://pubmed.ncbi.nlm.nih.gov/25500202/). DOI: 10.1136/gutjnl-2014-307913. Access: full_text_partial. Study: Propionate-NCT00750438.

62. Pugh JE, Petropoulou K, Vasconcelos JC et al. Increase in colonic PRopionate as a method of prEVENTing weight gain over 12 months in adults aged 20-40 years (iPREVENT): a multi-centre, double-blind, randomised, parallel-group trial. EClinicalMedicine. 2024. [PMID 39391015](https://pubmed.ncbi.nlm.nih.gov/39391015/). DOI: 10.1016/j.eclinm.2024.102844. Access: full_text_partial. Study: iPREVENT-ISRCTN16299902.

63. Vrieze A, Out C, Fuentes S et al. Impact of oral vancomycin on gut microbiota, bile acid metabolism, and insulin sensitivity. Journal of hepatology. 2014. [PMID 24316517](https://pubmed.ncbi.nlm.nih.gov/24316517/). DOI: 10.1016/j.jhep.2013.11.034. Access: abstract_metadata. Study: Antibiotic-NTR2566.

64. Bäckhed F, Manchester JK, Semenkovich CF et al. Mechanisms underlying the resistance to diet-induced obesity in germ-free mice. Proceedings of the National Academy of Sciences of the United States of America. 2007. [PMID 17210919](https://pubmed.ncbi.nlm.nih.gov/17210919/). DOI: 10.1073/pnas.0605374104. Access: abstract_metadata. Study: PMID17210919.

65. Rinott E, Youngster I, Yaskolka Meir A et al. Effects of Diet-Modulated Autologous Fecal Microbiota Transplantation on Weight Regain. Gastroenterology. 2021. [PMID 32860791](https://pubmed.ncbi.nlm.nih.gov/32860791/). DOI: 10.1053/j.gastro.2020.08.041. Access: full_text_partial. Study: DIRECT-PLUS-NCT03020186.

66. Kamer O, Rinott E, Tsaban G et al. Successful weight regain attenuation by autologous fecal microbiota transplantation is associated with non-core gut microbiota changes during weight loss; randomized controlled trial. Gut microbes. 2023. [PMID 37796016](https://pubmed.ncbi.nlm.nih.gov/37796016/). DOI: 10.1080/19490976.2023.2264457. Access: abstract_metadata. Study: DIRECT-PLUS-NCT03020186.

67. Kimura I, Ozawa K, Inoue D et al. The gut microbiota suppresses insulin-mediated fat accumulation via the short-chain fatty acid receptor GPR43. Nature communications. 2013. [PMID 23652017](https://pubmed.ncbi.nlm.nih.gov/23652017/). DOI: 10.1038/ncomms2852. Access: full_text_partial. Study: PMID23652017.

68. Chevalier C, Stojanović O, Colin DJ et al. Gut Microbiota Orchestrates Energy Homeostasis during Cold. Cell. 2015. [PMID 26638070](https://pubmed.ncbi.nlm.nih.gov/26638070/). DOI: 10.1016/j.cell.2015.11.004. Access: abstract_metadata. Study: PMID26638070.

## Source-specific evidence cards

### [1] An obesity-associated gut microbiome with increased capacity for energy harvest.

**Shows:** Obese-donor community transfers greater fat accumulation and energy-harvest capacity to germ-free recipients.

**Does not show:** Crude communities and exceptional recipients do not identify a universal human obesity configuration or exclude diet interaction.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [2] Gut microbiota from twins discordant for obesity modulate metabolism in mice.

**Shows:** Discordant-twin fecal communities and cultured collections transfer adiposity; lean-community invasion rescues in a diet-dependent manner.

**Does not show:** Human donors are not human causal recipients; saturated-fat/low-produce diet blocks rescue and donor/cage dependence limits generality.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [3] Human genetics shape the gut microbiome.

**Shows:** Human heritable Christensenellaceae associate with low BMI; cultured C.minuta amendment reduces recipient mouse weight gain.

**Does not show:** Causal attribution applies to mouse amendment; the human twin relation is observational and weight gain is not a human treatment endpoint.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

### [4] The intestinal microbiota regulates body composition through NFIL3 and the circadian clock.

**Shows:** Microbial control of circadian NFIL3 links innate immune signalling to intestinal lipid absorption and body composition.

**Does not show:** A host pathway in mice does not demonstrate independent human adiposity causation; circadian and dietary conditions matter.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [5] Akkermansia muciniphila supplementation in patients with overweight/obese type 2 diabetes: Efficacy depends on its baseline levels in the gut.

**Shows:** The 58-person live-Akkermansia RCT found no whole-group between-treatment weight or HbA1c difference; low-baseline-Akk subgroup had colonization and fat/weight signals.

**Does not show:** Abstract-only subgroup pre-specification, interaction and multiplicity are unresolved; within-arm significance is not a between-arm interaction.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [6] The gut microbiota as an environmental factor that regulates fat storage.

**Shows:** Conventionalization increases body fat despite lower intake; Fiaf knockout experiments implicate host lipid-storage regulation.

**Does not show:** Germ-free development and host-genotype manipulation limit human inference; later diet/strain counterexperiments constrain universality.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [7] Dietary emulsifiers impact the mouse gut microbiota promoting colitis and metabolic syndrome.

**Shows:** Emulsifier-associated communities transfer inflammation/metabolic phenotype; germ-free contrasts support a microbial contribution.

**Does not show:** Crude transfer has at most Moderate specificity; exposure may have additional host effects and linked erratum27144359 was not audited in full.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID27144359. Subclaims: SC1.

### [8] A core gut microbiome in obese and lean twins.

**Shows:** 154 people show shared microbial gene functions and obesity-associated diversity, phylum and pathway differences.

**Does not show:** Families and diet share determinants; association does not establish direction, a diagnostic phylum ratio or a human fat-producing mechanism.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [9] Effects of Fecal Microbiome Transfer in Adolescents With Obesity: The Gut Bugs Randomized Controlled Trial.

**Shows:** GutBugs randomized 87 adolescents: BMI-SDS primary and total-fat percentage null; android/gynoid ratio lower at6/12/26 weeks.

**Does not show:** Original Methods use no clinical multiplicity adjustment; secondary fat distribution is not total fat loss, and metabolic-syndrome resolution is post hoc.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Methods:paragraph0, 1:Methods:paragraph1, 1:Methods:paragraph5, 1:Methods:paragraph6, 1:Methods:paragraph7, 1:Methods:paragraph9, 1:Methods:paragraph10, 2:Results:paragraph0, 2:Results:paragraph4, 2:Results:paragraph5, 2:Results:paragraph8. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [10] The intestinal microbiota programs diurnal rhythms in host metabolism through histone deacetylase 3.

**Shows:** Epithelial HDAC3 integrates microbial/circadian cues, induces Cd36/lipid uptake and promotes diet-induced obesity.

**Does not show:** Host transcriptional perturbation supports a mouse mechanism; human adiposity and effect magnitude were not established.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [11] Transfer of intestinal microbiota from lean donors increases insulin sensitivity in individuals with metabolic syndrome.

**Shows:** Randomized lean versus autologous FMT improved six-week clamp insulin sensitivity in metabolic-syndrome men.

**Does not show:** Insulin sensitivity is not fat accumulation; crude FMT lacks strain attribution and the linked unspecified erratum requires separate audit.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMIDnot supplied. Subclaims: SC2, SC3.

### [12] Dietary fat disrupts a commensal-host lipid network that promotes metabolic health.

**Shows:** Turicibacter and purified bacterial lipids suppress host ceramides/fat uptake and obesity; dietary fat reduces colonization and continuous dosing restores benefit.

**Does not show:** The human metagenomic relation is observational; diet-dependent microbial protection does not establish a human pro-obesity configuration.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

### [13] Small Intestine Microbiota Regulate Host Digestive and Absorptive Adaptive Responses to Dietary Lipids.

**Shows:** Jejunal microbiota from high-fat donors increases recipient lipid absorption even on low-fat diet; conditioned media induce epithelial uptake genes.

**Does not show:** Recipient absorption and organoid activity are not a quantified human body-fat effect; crude transfer specificity remains limited.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [14] Antibiotic-induced microbiome depletion alters metabolic homeostasis by affecting gut signaling and colonic metabolism.

**Shows:** Depletion improves glucose tolerance and insulin sensitivity without changing adiposity, alongside SCFA/bile changes.

**Does not show:** Broad antibiotics have host effects; glucose improvement explicitly cannot substitute for fat loss.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [15] Microbiota-induced obesity requires farnesoid X receptor.

**Shows:** FXR genotype and germ-free/conventional contrasts change diet-induced adiposity; donor community transfers phenotype.

**Does not show:** Host receptor dependence and transfer do not locate an invariant microbial cause of human obesity.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [16] Supplementation with Akkermansia muciniphila in overweight and obese human volunteers: a proof-of-concept exploratory study.

**Shows:** 2019 live/pasteurized Akk pilot improved selected metabolic measures;40enrolled32 completed; weight/fat changes were nonsignificant nearP=.09.

**Does not show:** Body fat used bioimpedance, not DXA; original text and abstract differ on the fat comparator, but neither demonstrates significant fat loss; calorie/activity detail and baseline insulin balance limited.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 0:Online Methods:paragraph0, 0:Online Methods:paragraph1, 0:Online Methods:paragraph2, 0:Online Methods:paragraph4, 0:Online Methods:paragraph5, 0:Online Methods:paragraph6, 0:Online Methods:paragraph13, 0:Online Methods:paragraph15, 7:Main text (unsectioned body paragraphs):paragraph1, 7:Main text (unsectioned body paragraphs):paragraph2, 7:Main text (unsectioned body paragraphs):paragraph3, 7:Main text (unsectioned body paragraphs):paragraph13, 7:Main text (unsectioned body paragraphs):paragraph14, 7:Main text (unsectioned body paragraphs):paragraph4, 7:Main text (unsectioned body paragraphs):paragraph5, 7:Main text (unsectioned body paragraphs):paragraph6, 7:Main text (unsectioned body paragraphs):paragraph7, 7:Main text (unsectioned body paragraphs):paragraph8, 7:Main text (unsectioned body paragraphs):paragraph9. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [17] A purified membrane protein from Akkermansia muciniphila or the pasteurized bacterium improves metabolism in obese and diabetic mice.

**Shows:** Pasteurization and Amuc1100/TLR2-linked protein partly recapitulate metabolic/fat benefits in mice; preliminary human safety was reported.

**Does not show:** A bioactive nonliving component is not living-community engraftment; preliminary human volunteers belong to the later2019trial family.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

### [18] Effects of Fecal Microbiota Transplantation With Oral Capsules in Obese Patients.

**Shows:** Obese metabolically uncompromised participants show donor-like engraftment/bile changes but no GLP1-area or BMI improvement.

**Does not show:** Safety was primary, small pilot; PubMed abstract has an implausible BMI-threshold typo, so exact eligibility needs original verification.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [19] Microbiota engraftment after faecal microbiota transplantation in obese subjects with type 2 diabetes: a 24-week, double-blind, randomised controlled trial.

**Shows:** Repeated FMT in 61 obese T2D adults improved24 week donor engraftment, the primary endpoint; combined lifestyle arm had lipid/liver-stiffness changes.

**Does not show:** Engraftment-primary success is not adiposity success; baseline comparisons and cointervention cannot establish independent FMT fat loss.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [20] Effects of Synbiotic Supplement on Human Gut Microbiota, Body Composition and Weight Loss in Obesity.

**Shows:** Twenty-person synbiotic plus reduced-energy diet pilot changed beneficial taxa but not between-arm body mass, fat mass or composition.

**Does not show:** Small precision and combined diet limit exclusion of modest effects; taxa improvement is not a successful obesity endpoint.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [21] Effects of daily consumption of the probiotic Bifidobacterium animalis subsp. lactis CECT 8145 on anthropometric adiposity biomarkers in abdominally obese subjects: a randomized controlled trial.

**Shows:** Ba8145 original Methods/Results report live-cell BMI and killed-cell waist-related between-placebo benefits;126 received intervention of 135 reported participants.

**Does not show:** MRI visceral-fat between-placebo contrasts are nonsignificant(P=.120/.086 for killed cells), despite significant within-baseline change; statistics/hierarchy in unread supplement and higher killed-arm fiber complicate interpretation; Akk correlation is not mediation.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Methods, 2:Results. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [22] Probiotic With or Without Fiber Controls Body Fat Mass, Associated With Serum Zonulin, in Overweight and Obese Adults-Randomized Controlled Trial.

**Shows:** B420factorial RCT randomized 225; ITT 209 fat-primary null; PP combination and post-unblinding factorial analyses positive.

**Does not show:** Original Methods confirm modified ITT/LOCF, Dunnett contrasts and posthoc PP factorial; selective compliance and reduced intake weaken direct microbial mediation; zonulin correlation is not barrier causation.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Materials and Methods:paragraph0, 1:Materials and Methods:paragraph5, 1:Materials and Methods:paragraph7, 1:Materials and Methods:paragraph15, 1:Materials and Methods:paragraph16, 1:Materials and Methods:paragraph17, 2:Results:paragraph0, 2:Results:paragraph1, 2:Results:paragraph4, 2:Results:paragraph5, 2:Results:paragraph7, 2:Results:paragraph10. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [23] Gut Microbiome and Metabolome Alterations in Overweight or Obese Adult Population after Weight-Loss Bifidobacterium breve BBr60 Intervention: A Randomized Controlled Trial.

**Shows:** 75-person BBr60 plus common dietary guidance trial reports greater weight/BMI reduction and metabolic/taxa changes than placebo.

**Does not show:** Abstract-only fat outcomes, primary hierarchy and multiplicity remain unresolved; metabolomic correlation does not identify the mediator.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [24] Effect of Lactobacillus rhamnosus CGMCC1.3724 supplementation on weight loss and maintenance in obese men and women.

**Shows:** L.rhamnosus formulation with prebiotics had an overall ITT weight null but treatment-by-sex interaction; women showed more weight/fat loss over24 weeks.

**Does not show:** Sex interaction retains scoped causal credit; formulation and restricted diet prevent isolating a strain-only mechanism, and sex-specific primary status remains to be audited.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [25] Long-term health outcomes in adolescents with obesity treated with faecal microbiota transplantation: 4-year follow-up.

**Shows:** 2025GutBugs4 year follow-up found lower waist and body-fat percentage among 55returnees, with persistent donor strains and nonsignificant BMI difference.

**Does not show:** Same trial family; unblinded after26 weeks,37%lost, no missing-data imputation and unadjusted clinical multiplicity; intervening diet/medication unknown.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph0, 1:Results:paragraph1, 1:Results:paragraph2, 1:Results:paragraph3, 1:Results:paragraph4, 1:Results:paragraph5, 1:Results:paragraph6, 1:Results:paragraph8, 1:Results:paragraph10, 1:Results:paragraph12, 1:Results:paragraph14, 1:Results:paragraph15, 2:Discussion:paragraph1, 2:Discussion:paragraph3, 3:Methods:paragraph0, 3:Methods:paragraph1, 3:Methods:paragraph2, 3:Methods:paragraph3, 3:Methods:paragraph17, 3:Methods:paragraph42, 3:Methods:paragraph43, 3:Methods:paragraph45, 3:Methods:paragraph46. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [26] Probiotic supplementation attenuates increases in body mass and fat mass during high-fat diet in healthy young adults.

**Shows:** 20healthy young men on a common high-fat1000 kcal-surplus diet gained less DXA fat and weight with VSL3 than placebo after4 weeks.

**Does not show:** Direct controlled fat-accumulation contrast supports a bounded human contribution; small/short study, multiple-strain product and unverified engraftment limit mechanism/generalization.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [27] Microbiome connections with host metabolism and habitual diet from 1,098 deeply phenotyped individuals.

**Shows:** PREDICT1 links microbiome, habitual diet and standardized-meal metabolic responses in 1098 participants with external biomarker replication.

**Does not show:** Same-meal responses and prediction are not microbiome randomization; public cohorts may recur elsewhere and cannot become independent causal replications.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [28] Obesity-enriched gut microbe degrades myo-inositol and promotes lipid absorption.

**Shows:** 631obese374control cohort associates Megamonas with obesity; M.rupellensis and heterologous iolG expression promote mouse lipid absorption, with myo-inositol functional support.

**Does not show:** The human portion is observational and the primary body unavailable; gene/strain perturbations narrow a mouse mechanism without demonstrating human treatment mediation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

### [29] US Immigration Westernizes the Human Gut Microbiome.

**Shows:** Immigration-associated microbial loss and Westernization track residence, generation and obesity, including19 people sampled before/after migration.

**Does not show:** Migration simultaneously changes food, environment and host exposures; longitudinal association does not isolate microbial causation of fat gain.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [30] Mediterranean diet intervention in overweight and obese subjects lowers plasma cholesterol and causes changes in the gut microbiome and metabolome independently of energy intake.

**Shows:** Isocaloric Mediterranean diet altered microbes/metabolites and lowered cholesterol, the primary endpoint, in 82adults.

**Does not show:** Diet was randomized, not microbes; microbial/insulin covariance does not establish adiposity mediation or microbiome-independent weight control.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [31] Environment dominates over host genetics in shaping human gut microbiota.

**Shows:** 1046 people show shared-household/diet/drug associations and modest host-genetic contribution to overall community variation; microbes improve phenotype prediction.

**Does not show:** Prediction and variance attribution do not establish direction, independence from all host physiology or causal fat accumulation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [32] Richness of human gut microbiome correlates with metabolic markers.

**Shows:** Low microbial gene richness associates with adiposity/metabolic inflammation and later weight gain in a Danish cohort.

**Does not show:** Temporal prediction reduces one timing ambiguity but does not randomize diet/host determinants or prove richness itself causal.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [33] Beneficial metabolic effects of PAHSAs depend on the gut microbiota in diet-induced obese mice but not in chow-fed mice.

**Shows:** PAHSA-treated community transfers insulin sensitivity on high-fat diet; Bt reduces adiposity in female mice, partly dependent on sex hormones.

**Does not show:** Effect differs by diet and sex; glucose improvement and female fat reduction must not be combined into universal human mediation; linked preprint is same study.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. UpdateOf: PMID37808673. Subclaims: SC1, SC3.

### [34] Ketogenic diet-induced bile acids protect against obesity through reduced calorie absorption.

**Shows:** Microbial BSH/strain and bile-acid interventions reduce mouse calorie absorption; human observational and small ketogenic-diet studies show concordant associations.

**Does not show:** Human diet intervention is not microbial-mediator randomization; bile-acid replacement does not alone prove the community configuration causes human fat gain.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

### [35] Parabacteroides distasonis Alleviates Obesity and Metabolic Dysfunctions via Production of Succinate and Secondary Bile Acids.

**Shows:** P.distasonis improves ob/ob and high-fat mouse phenotypes; succinate plus secondary bile acids recapitulate metabolic benefits.

**Does not show:** Model-specific rescue supports function rather than a universal abundance signature; isolated mediator effects on glucose are distinct from all fat endpoints.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [36] B. vulgatus ameliorates high-fat diet-induced obesity through modulating intestinal serotonin synthesis and lipid absorption in mice.

**Shows:** B.vulgatus/cholic acid and epithelial Tph1 contrasts implicate a microbe-metabolite-serotonin axis in lipid absorption and high-fat adiposity.

**Does not show:** Host epithelial and dietary dependencies are integral; human colonization and adiposity mediation are not established.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [37] Microbiota-mediated induction of beige adipocytes in response to dietary cues.

**Shows:** 2026 low-protein diet browning defect in germ-free mice is rescued by defined consortia; bile-acid/FXR and ammonia/FGF21 axes are nonredundant.

**Does not show:** Browning genes and dietary compensation are not all clinical fat outcomes; human donor selection is not a human causal trial; linked preprint excluded as duplicate.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. UpdateOf: PMID39678347. Subclaims: SC1.

### [38] Effects of underfeeding and oral vancomycin on gut microbiome and nutrient absorption in humans.

**Shows:** Inpatient crossover feeding plus blinded vancomycin intervention met primary stool-energy-loss endpoints in 25 completers.

**Does not show:** Human energy-harvest perturbation is real bounded evidence, but broad antibiotic effects and short timing do not quantify sustained adiposity or a specific microbial mediator.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [39] Diet rapidly and reproducibly alters the human gut microbiome.

**Shows:** Animal versus plant diets rapidly alter microbial composition and expression, including bile-tolerant and foodborne microbes.

**Does not show:** Establishes diet-to-community responsiveness, not community-to-fat causation; transient detection is not durable functional colonization.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [40] Microbially Produced Imidazole Propionate Impairs Insulin Signaling through mTORC1.

**Shows:** Microbial imidazole propionate is enriched in T2D material and impairs mouse glucose/insulin signalling through p38gamma/p62/mTORC1.

**Does not show:** Glucose pathology is not adiposity; human metabolite and hepatic pathway associations do not independently establish human fat mediation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [41] Host remodeling of the gut microbiome and metabolic changes during pregnancy.

**Shows:** Third-trimester community differs from first trimester and transfers more adiposity/insulin insensitivity to germ-free mice.

**Does not show:** Normal host pregnancy remodels microbes upstream; adaptive physiology and crude cross-species transfer do not establish common human obesity initiation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

### [42] Effects of the gut microbiota on host adiposity are modulated by the short-chain fatty-acid binding G protein-coupled receptor, Gpr41.

**Shows:** Gpr41loss lowers adiposity in conventional and defined Bt/M.smithii-colonized mice despite similar chow intake, with transit/PYY/energy-harvest links.

**Does not show:** Host receptor, community and transit jointly determine effect; no quantified human fat contribution or universal SCFA direction follows.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [43] Donor metabolic characteristics drive effects of faecal microbiota transplantation on recipient insulin sensitivity, energy expenditure and intestinal transit time.

**Shows:** Donor-characteristic FMT contrast in 22metabolic-syndrome men found worsened clamp sensitivity with metabolic-syndrome donors and bile/immune changes.

**Does not show:** Two-week glucose effect is not fat accumulation; no neutral-placebo inference or defined mediator from crude donor material.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [44] Obesity alters gut microbial ecology.

**Shows:** Ob/ob mice have altered Firmicutes/Bacteroidetes despite shared diet and family comparisons.

**Does not show:** Leptin genotype precedes phenotype and microbial ecology; this observational contrast cannot establish microbial direction or phylum-ratio necessity.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

### [45] Metabolic endotoxemia initiates obesity and insulin resistance.

**Shows:** Continuous LPS infusion increases fat/weight and liver insulin resistance; CD14mutants resist many LPS/high-fat phenotypes.

**Does not show:** Artificial infusion and host knockout support a defined inflammatory axis, not endogenous human exposure magnitude or all insulin resistance.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [46] Cross-talk between Akkermansia muciniphila and intestinal epithelium controls diet-induced obesity.

**Shows:** Live Akkermansia reduces high-fat fat gain and barrier/inflammatory abnormalities; heat-killed preparation was ineffective in this study.

**Does not show:** Heat-killing is not equivalent to later pasteurization; formulation/context matter, and human inverse abundance is associative.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [47] Changes in gut microbiota control metabolic endotoxemia-induced inflammation in high-fat diet-induced obesity and diabetes in mice.

**Shows:** Antibiotics lower endotoxemia, adiposity and inflammation in high-fat/obob mice; CD14deletion partly mimics the phenotype.

**Does not show:** Antibiotic pleiotropy and shared investigative programme constrain specificity; no human fat or endogenous LPS mediation estimate.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [48] Maternal gut microbiota in pregnancy influences offspring metabolic phenotype in mice.

**Shows:** Maternal microbial SCFAs signal through embryonic GPR41/43 and affect later offspring metabolic susceptibility.

**Does not show:** Developmental maternal mechanism is distinct from adult recipient obesity and not directly generalized to human fat accumulation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [49] Acetylated cellulose suppresses body mass gain through gut commensals consuming host-accessible carbohydrates.

**Shows:** Acetylated cellulose limits body-mass gain through microbiota-dependent host-accessible sugar depletion and altered oxidation.

**Does not show:** Mouse prebiotic substrate effect is diet-dependent; bacterial carbohydrate consumption can protect rather than increase host energy harvest.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [50] Population-level Metagenomics Uncovers Distinct Effects of Multiple Medications on the Human Gut Microbiome.

**Shows:** 4198 people and243repeat samples identify drug/polypharmacy-associated community functions with initiation/cessation tracking.

**Does not show:** Confounder evidence is not obesity mechanism; treatment and disease indication share causes, and linked36610810erratum not fully audited.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID36610810. Subclaims: SC2.

### [51] High-fat feeding rather than obesity drives taxonomical and functional changes in the gut microbiota in mice.

**Shows:** Diet and mouse strain explain substantial community shifts when high-fat obesity susceptibility is experimentally dissociated.

**Does not show:** Community changes need not track fat gain; indomethacin and genotype add effects and do not erase all microbial causality.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

### [52] Human and preclinical studies of the host-gut microbiome co-metabolite hippurate as a marker and mediator of metabolic health.

**Shows:** Urinary hippurate associates with human diet/metabolic state; chronic infusion improves glucose tolerance/insulin secretion in high-fat mice.

**Does not show:** Host conjugation contributes to metabolite origin; glucose benefit is not direct human fat reduction or universal microbial mediation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [53] Pasteurized Akkermansia muciniphila MucT for weight loss maintenance in people with overweight and obesity: a controlled randomized trial.

**Shows:** 2026 pasteurized-Akk RCT 90 people reports primary24 week weight regain1.2vs3.2 kg(P=.012) after successful low-energy-diet loss.

**Does not show:** Abstract plus publisher ancillary statements only: fat results, attrition and multiplicity unavailable; nonliving product benefit does not demonstrate living-community causation; company/patent conflicts disclosed.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [54] Improvement of Insulin Sensitivity after Lean Donor Feces in Metabolic Syndrome Is Driven by Baseline Intestinal Microbiota Composition.

**Shows:** Lean-donor FMT improves six-week insulin sensitivity in selected baseline communities, with no metabolic effect at18 weeks.

**Does not show:** Metabolic responder stratification is not prevalidated adiposity selection; transient clamp effect is not sustained fat control.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [55] Fecal microbiota transplantation for the improvement of metabolism in obesity: The FMT-TRIM double-blind placebo-controlled pilot trial.

**Shows:** FMT-TRIM randomized 24; primary clamp null and secondary weight/fat mass null despite variable donor engraftment.

**Does not show:** Original Methods power only for large clamp effects; pilot null is not equivalence, donor strength differs and16SASVs cannot uniquely resolve all strains.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Methods:paragraph0, 1:Methods:paragraph1, 1:Methods:paragraph3, 1:Methods:paragraph6, 1:Methods:paragraph7, 2:Results:paragraph0, 2:Results:paragraph6, 2:Results:paragraph13, 2:Results:paragraph14, 2:Results:paragraph15, 2:Results:paragraph16, 2:Results:paragraph17. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [56] Fecal microbial transplantation and fiber supplementation in patients with severe obesity and metabolic syndrome: a randomized double-blind, placebo-controlled phase 2 trial.

**Shows:** 70-person FMT/fiber factorial reports six-week HOMA improvement in FMT-low-fermentable arm.

**Does not show:** Abstract chiefly reports within-armP=.02: significance versus other null arms is not a between-arm treatment effect; original contrasts and adiposity mediation unverified.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [57] Recipient microbiome-related features predicting metabolic improvement following fecal microbiota transplantation in adults with severe obesity and metabolic syndrome: a secondary analysis of a phase 2 clinical trial.

**Shows:** Mocanu trial follow-up groups21responders8nonresponders; baseline taxa/engraftment correlate with HOMA response.

**Does not show:** Same family and response-defined subgroups; predictionAUC.70 is not external validation or new randomized human fat causation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [58] Effects of Gut Microbiota Manipulation by Antibiotics on Host Metabolism in Obese Humans: A Randomized Double-Blind Placebo-Controlled Trial.

**Shows:** 57-obese-prediabetic-men antibiotic RCT changes communities/SCFA/bile but not clinically relevant energy harvest, insulin sensitivity or adipocyte size through8 weeks.

**Does not show:** Short broad intervention cannot falsify all configurations; null adipocyte size is not all total-fat endpoints, and correction completeness is limited.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [59] Energy-balance studies reveal associations between gut microbes, caloric load, and nutrient absorption in humans.

**Shows:** Inpatient energy-balance study in 12lean9 obese links nutrient load, stool energy and Firmicutes/Bacteroidetes, including an approximately150 kcal association.

**Does not show:** Dietary load changes microbes and absorption together; microbial proportions were not randomized and150 kcal is not a demonstrated sustained fat-gain cause.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [60] Absence of intestinal microbiota does not protect mice from diet-induced obesity.

**Shows:** Germ-free C3Hmice gain more fat on one high-fat formula while Western diet behaves differently; Fiaf differences do not match universal protection.

**Does not show:** Diet/strain boundary challenges necessity and simple Fiaf universalization; does not negate positive mechanisms in other hosts or substrates.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

### [61] Effects of targeted delivery of propionate to the human colon on appetite regulation, body weight maintenance and adiposity in overweight adults.

**Shows:** Propionate ester reduces threshold-defined weight-gain events and secondary intra-abdominal-fat distribution; acute appetite/hormone perturbation works.

**Does not show:** Original24 week mean weight differenceP=.099 and total-adipose difference null;49/60 completers and32composition data; ester delivery is not randomized living community.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Methods:paragraph3, 1:Methods:paragraph4, 1:Methods:paragraph5, 1:Methods:paragraph7, 1:Methods:paragraph9, 2:Results:paragraph1, 2:Results:paragraph5, 2:Results:paragraph6, 2:Results:paragraph9, 2:Results:paragraph10, 2:Results:paragraph13. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [62] Increase in colonic PRopionate as a method of prEVENTing weight gain over 12 months in adults aged 20-40 years (iPREVENT): a multi-centre, double-blind, randomised, parallel-group trial.

**Shows:** iPREVENT270 randomized:12 month primary weight difference1.02kgCI-.37to2.41 null; secondary fat-mass difference.07kgCI-1.07to1.21 null.

**Does not show:** Original model used256 with any follow-up,226at12 months; adherence and low control weight gain constrain exposure/generalization, not permission to declare overall efficacy.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Methods:paragraph0, 1:Methods:paragraph1, 1:Methods:paragraph3, 1:Methods:paragraph4, 1:Methods:paragraph6, 1:Methods:paragraph8, 1:Methods:paragraph11, 1:Methods:paragraph12, 1:Methods:paragraph13, 1:Methods:paragraph22, 1:Methods:paragraph23, 1:Methods:paragraph28, 1:Methods:paragraph29, 1:Methods:paragraph30, 2:Results:paragraph0, 2:Results:paragraph4, 2:Results:paragraph5, 2:Results:paragraph10, 2:Results:paragraph14, 2:Results:paragraph15, 2:Results:paragraph16, 2:Results:paragraph18. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [63] Impact of oral vancomycin on gut microbiota, bile acid metabolism, and insulin sensitivity.

**Shows:** Seven-day vancomycin versus amoxicillin changes microbes/bile acids and worsens peripheral clamp sensitivity in 20 men.

**Does not show:** Active-antibiotic comparator and short treatment do not isolate a microbial mediator or establish fat gain.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [64] Mechanisms underlying the resistance to diet-induced obesity in germ-free mice.

**Shows:** Germ-free Western-diet protection is lost with Fiafdeletion; AMPK/fatty-acid oxidation supports complementary host mechanisms.

**Does not show:** Diet/strain counterstudy limits universal Fiaf or microbiota necessity; protection is not endogenous human causal weight evidence.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [65] Effects of Diet-Modulated Autologous Fecal Microbiota Transplantation on Weight Regain.

**Shows:** Rinott aFMT had overall primary regain null(P=.28), but green-Mediterranean interaction(P=.03) and diet-specific benefit(P=.02); Methods confirmITT90 with one imputation.

**Does not show:** Diet-specific weight result is not uniform FMT or total-fat success; original protocol/supplement pre-specification not independently audited; crude transfer remains nonspecific.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: PMC original author manuscript: Methods/Statistical analysis and Results/Weight regain, lines300-305,335-342,396-413. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [66] Successful weight regain attenuation by autologous fecal microbiota transplantation is associated with non-core gut microbiota changes during weight loss; randomized controlled trial.

**Shows:** DIRECT-PLUS secondary analysis associates low-core/high-noncore community change with aFMT weight maintenance.

**Does not show:** Same trial family, response-related multivariate subgroup and taxa selection; not independent efficacy replication or prospective validation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [67] The gut microbiota suppresses insulin-mediated fat accumulation via the short-chain fatty acid receptor GPR43.

**Shows:** GPR43loss/OE effects disappear with germ-free/antibiotic conditions; conventionalization and acetate rescue restore receptor-dependent fat effects.

**Does not show:** Original strains differ between KO and OE backgrounds; lifelong manipulation and receptor/tissue specificity limit universal SCFA or human obesity inference.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 0:Results:paragraph0, 0:Results:paragraph1, 0:Results:paragraph2, 0:Results:paragraph3, 0:Results:paragraph4, 0:Results:paragraph5, 0:Results:paragraph6, 0:Results:paragraph7, 0:Results:paragraph8, 1:Discussion:paragraph4, 1:Discussion:paragraph5, 1:Discussion:paragraph6, 2:Methods:paragraph0, 2:Methods:paragraph1, 9:Main text (unsectioned body paragraphs):paragraph2, 9:Main text (unsectioned body paragraphs):paragraph3. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [68] Gut Microbiota Orchestrates Energy Homeostasis during Cold.

**Shows:** Cold-community transfer drives browning/fat loss followed by absorptive compensation; Akk co-transfer diminishes intestinal adaptation.

**Does not show:** Physiological demand, substrate and compensation change the sign/magnitude; a protective transfer is not a universal human fat-producing configuration.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.