CoChat
Sign Up
Cited by 14Birth, Development, and HealthDiet, Metabolism, and DiseaseObesity, Physical Activity, Diet

Transgenerational effects of obesogenic diets in rodents: A meta‐analysis

Open access
Anwer, Hamza, Morris, Margaret J., Noble, Daniel W. A., Nakagawa, Shinichi, Lagisz, Malgorzata·2022·Obesity Reviews
DOI 10.1111/obr.13342
Open this paper in CoChat — free accountOpen PDF

Abstract

Summary: Obesity is a major health condition that affects millions worldwide. There is an increased interest in understanding the adverse outcomes associated with obesogenic diets. A multitude of studies have investigated the transgenerational impacts of maternal and parental obesogenic diets on subsequent generations of offspring, but results have largely been mixed. We conducted a systematic review and meta‐analysis on rodent studies to elucidate how obesogenic diets impact the mean and variance of grand‐offspring traits. Our study focused on transgenerational effects (i.e., F2 and F3 generations) in one‐off and multigenerational exposure studies. From 33 included articles, we obtained 407 effect sizes representing pairwise comparisons of control and treatment grand‐offspring groups pertaining to measures of body weight, adiposity, glucose, insulin, leptin, and triglycerides. We found evidence that male and female grand‐offspring descended from grandparents exposed to an obesogenic diet displayed phenotypes consistent with metabolic syndrome, especially in cases where the obesogenic diet was continued across generations. Further, we found stronger evidence for the effects of grand‐maternal than grand‐paternal exposure on grand‐offspring traits. A high‐fat diet in one‐off exposure studies did not seem to impact phenotypic variation, whereas in multigenerational exposure studies it reduced variation in several traits.

AI snapshot

Written from the paper’s full text.

TL;DR

This systematic review and meta-analysis of 33 rodent studies found that grand-offspring of grandparents exposed to obesogenic diets show phenotypes consistent with metabolic syndrome, especially under multigenerational exposure. Effects were stronger for grand-maternal than grand-paternal exposure in multigenerational studies, while grand-offspring sex differences were not evident.

Key findings

  • From 33 articles, 407 effect sizes were obtained. One-off exposure grand-offspring had mean trait values 9% higher than controls, but this was non-significant (lnRR = 0.085, 95% CI −0.076 to 0.247, p = 0.301); multigenerational exposure grand-offspring had mean trait values 43% higher (lnRR = 0.358, CI 0.096 to 0.620, p = 0.007).
  • In multigenerational experiments, grand-maternal exposure had a significant effect on grand-offspring mean traits (51% increase, lnRR = 0.412, CI 0.140 to 0.683, p = 0.003), whereas grand-paternal exposure did not (16% increase, lnRR = 0.146, CI −0.183 to 0.475, p = 0.384; difference p = 0.038). One-off exposure showed no clear effect of exposed grandparent sex.
  • Adiposity was the most affected trait: 20% higher in one-off exposure (lnRR = 0.182, p = 0.023) and 121% higher in multigenerational exposure (lnRR = 0.794, p < 0.001). Multigenerational exposure also increased leptin by 139% (lnRR = 0.872, p < 0.001), fasting insulin by 58% (lnRR = 0.457, p = 0.005), and triglycerides by 41% (lnRR = 0.340, p = 0.008); body mass effect was small and non-significant (11%, p = 0.225).
  • Effects on trait variability were mostly small and non-significant, except multigenerational triglycerides, which had lower variability in treatment grand-offspring (lnCVR = −0.388, CI −0.756 to −0.020, p = 0.039). Higher total energy content of the obesogenic diet in multigenerational experiments predicted larger mean effects (slope = 0.188, p = 0.003).

Methods

  • Systematic review and meta-analysis following PRISMA; searched Scopus, Medline, ISI Web of Knowledge, Embase, plus grey literature, Google Scholar, and snowballing; initial search April 2018, updated May 2020. Included 33 articles on wild-type laboratory rodents where F0 were exposed to an obesogenic diet and F2/F3 grand-offspring traits were reported.
  • Extracted 407 effect sizes for body weight, adiposity, glucose, insulin, leptin, and triglycerides from 1164 treatment and 1090 control unique grand-offspring. Used lnRR for trait means and lnCVR for trait variability; ran multilevel meta-analytic and meta-regression models in R with moderators including exposure type, exposed grandparent sex, grand-offspring sex, and trait type, plus publication bias and additional diet/exposure analyses.

Limitations

  • Stated: effect sizes were unevenly distributed across questions (66% one-off exposure, 74% grand-maternal exposure, most traits besides body weight <17%); most studies did not report F0 outcomes; treatment grand-offspring sample sizes ranged from 3 to 34 (mean 13, median 10); approximately 80% of effect sizes came from F2, limiting separation of true transgenerational effects from developmental programming; moderators explained only R² 0–0.37; funnel plot asymmetry was indicated only for the multigenerational lnRR dataset.
  • Evident from the text: heterogeneity was high for mean effects (I² = 95.7% one-off, 99.2% multigenerational), and in multigenerational studies three-quarters of animals were fed a high-fat diet around trait measurement, so direct grand-offspring diet effects could contribute to large differences and mask purely transgenerational effects.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1111/obr.13342

Cites

136 of 136 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 38 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    A high-fat diet in one-off exposure studies did not seem to impact phenotypic variation.

    “A highfat diet in one-off exposure studies did not seem to impact phenotypic variation, whereas in multigenerational exposure studies it reduced variation in several traits.”
    Quote in the paperSubstantial claim
  • Findingown data

    In multigenerational exposure studies, a high-fat diet reduced variation in several traits.

    “A highfat diet in one-off exposure studies did not seem to impact phenotypic variation, whereas in multigenerational exposure studies it reduced variation in several traits.”
    Quote in the paperSubstantial claim
  • Findingown data

    Exposed grandparents from both one-off and multigenerational datasets were, on average, 14.9% heavier than their non-exposed counterparts (lnRR = 0.139, CI = 0.062 to 0.216).

    “Exposed grandparents from both one-off and multigenerational datasets were, on average, 14.9% heavier than their non-exposed counterparts (lnRR = 0.139, CI = 0.062 to 0.216; Table S17, Table S18).”
    Quote in the paperNot a substantial claim
  • Findingown data

    In multigenerational exposure experiments, the effect of grand-maternal exposure on grand-offspring mean trait values was significantly larger than that of grand-paternal exposure (lnRR difference = -0.266, CI = -0.517 to -0.015, p = 0.038).

    “In contrast, grand-paternal exposed lines were associated with a small and statistically nonsignificant effect on mean trait values of grand-offspring (grandfathers: increase of 16%, lnRR = 0.146, CI = -0.183 to 0.475, p = 0.384; Figure 3b) compared to more marked effects from maternal exposure lines (lnRR difference = -0.266,CI = -0.517 to -0.015, p = 0.038).”
    Quote in the paperSubstantial claim
  • Findingown data

    Grand-offspring in multigenerational exposure lines had mean trait values 43% larger than control grand-offspring.

    “First, we have shown that grand-offspring in multigenerational exposure lines had mean trait values 43% larger than control grand-offspring.”
    • Also stated in: § One-off vs. Multigenerational exposure · plain
      “In contrast, grand-offspring traits of treatment animals from Multigenerational experiments were 43% higher than their control counterparts.”
    Quote in the paperSubstantial claim
  • May combine two claims. 33
  • Classified as probably not a substantial claim. 54
  • The quote only partly states this claim. 1

  • May combine two claims.

    Findingown data

    For one-off exposures, effects on both granddaughter and grandson mean trait values were small and statistically not different from zero or each other (granddaughters: 10%, lnRR = 0.091, CI = -0.056 to 0.238, p = 0.226; grandsons: 11%, lnRR = 0.103, CI = -0.044 to 0.250, p = 0.170).

    “For one-off exposures, the effects on both granddaughter and grandson mean trait values were small and statistically not different from zero, or each other (granddaughters: 10%, lnRR = 0.091, CI = -0.056 to 0.238, p = 0.226; grandsons: 11%, lnRR = 0.103, CI = -0.044 to 0.250, p = 0.170; Figure 4a; Table S7).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The full meta-regression model with sex of exposed grandparents, sex of measured grand-offspring, and measured trait type jointly explained 13% of variation in effects on mean trait values in one-off exposure data, 38% in multigenerational data, and 7.9% of effects on trait variability in both datasets.

    “Although there was still a great deal of model uncertainty, these three moderators jointly explained 13% of the variation in the effects on mean trait values in one-off exposure data, 38% in multigenerational data, and 7.9% of the effects on trait variability, in both datasets.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    Transgenerational data came mainly from F2 grand-offspring (325 effect sizes from 30 studies) as opposed to F3 great grand-offspring (82 effect sizes from 10 studies).

    “Transgenerational data came mainly from F2 grand-offspring (325 effect sizes from 30 studies), as opposed to F3 great grand-offspring (82 effect sizes from 10 studies).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Grand-offspring traits of treatment animals from one-off experiments tended to be 9% higher than control grand-offspring, although this overall effect was statistically non-significant.

    “Grand-offspring traits of treatment animals from One-off experiments tended to be 9% higher than control grand-offspring, although this overall effect was statistically non-significant.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Model selection analysis indicated that trait type was the most influential moderator of average effect sizes on the trait means in one-off data, and for both trait means and variabilities in multigenerational data.

    “Model selection analysis indicated that trait type was the most influential moderator of average effect sizes on the trait means in one-off data, and for both trait means and variabilities in multigenerational data (Table S10).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    From 33 included articles, 407 effect sizes were obtained representing pairwise comparisons of control and treatment grand-offspring groups for body weight, adiposity, glucose, insulin, leptin, and triglycerides.

    “From the 33 included articles (Table 1), we obtained 407 effect sizes representing pairwise comparisons of control and treatment groups pertaining to body weight, adiposity, glucose, insulin, leptin, and triglycerides in grand-offspring generations F2 and F3.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Predicted differences between grand-offspring at similar age (around 100 days old) would be around 7% in one-off experiments, and around 16.9% after multigenerational experiments.

    “For comparison, predicted differences between their grand-offspring at similar age (around 100 days old) would be around 7% in one-off experiments, and around 16.9% after multigenerational experiments (but note very wide confidence and prediction intervals; Table S17, Table S18).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    After multigenerational exposure, the effect on average body mass was small and not statistically significant (11%, lnRR = 0.100, CI = -0.062 to 0.262, p = 0.225).

    “Surprisingly, the effect on average body mass was small and not statistically significant (11%, lnRR = 0.100, CI = -0.062 to 0.262, p = 0.225).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    There were 23 articles with one-off exposure type (272 effect sizes) and 15 articles with multigenerational exposure experiments (135 effect sizes).

    “We found 23 articles with one-off exposure type (272 effect sizes) and 15 articles with multigenerational exposure experiments (135 effect sizes; some articles include both types of exposure; see Figure 1).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In one-off exposure studies, grand-offspring descended from exposed grandparents tended to have overall mean trait values 9% higher than control counterparts, but this was statistically non-significant (lnRR = 0.085, CI = -0.076 to 0.247, p = 0.301).

    “Although statistically non-significant, grand-offspring descended from exposed grandparents in one-off exposure studies tended to have overall mean trait values 9% higher than their control counterparts (lnRR = 0.085, CI = -0.076 to 0.247, p = 0.301; Figure 2a; Table S4).”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: One-off and multigenerational experiment data represent different biological processes and questions.

  • May combine two claims.

    Findingown data

    For multigenerational exposures, effects on granddaughter and grandson mean trait values were large and statistically different from zero (granddaughters: 45%, lnRR = 0.373, CI = 0.087 to 0.658, p = 0.010; grandsons: 39%, lnRR = 0.331, CI = 0.044 to 0.618, p = 0.024).

    “For multigenerational exposures, the effects on granddaughter and grandson mean trait values were large and statistically different from zero (granddaughters: 45%, lnRR = 0.373, CI = 0.087 to 0.658, p = 0.010; grandsons: 39%, lnRR = 0.331, CI = 0.044 to 0.618, p = 0.024; Figure 4b; Table S7).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In multigenerational exposure experiments, grand-offspring descended from exposed grandparents had mean trait values 43% higher than control counterparts, and this effect was statistically significant (lnRR = 0.358, CI = 0.096 to 0.620, p = 0.007).

    “Grand-offspring descended from exposed grandparents in multigenerational exposure experiments, however, had mean trait values 43% higher than control counterparts.This effect was statistically significant (lnRR = 0.358, CI = 0.096 to 0.620, p = 0.007; Table S4).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: One-off and multigenerational experiment data represent different biological processes and questions.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingnone given

    Leptin, fasting insulin, and fasting glucose exhibited a similar trend that inter-subject variability appears lower in the treatment group.

    “Notably, leptin, fasting insulin, and fasting glucose exhibited a similar trend that inter-subject variability appears lower in the treatment group.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The average effects on trait variability for both types of exposure were small and statistically non-significant (one-off: increase of 3%, lnCVR = 0.033, CI = -0.134 to 0.200, p = 0.698; multigenerational: decrease of 7%, lnCVR = -0.074, CI = -0.282 to 0.134, p = 0.486).

    “The average effects on trait variability for both types of exposure were small and statistically non-significant (one-off data: increase of 3%, lnCVR = 0.033, CI = -0.134 to 0.200, p = 0.698; multigenerational data: decrease of 7%, lnCVR = -0.074,CI = -0.282 to 0.134, p = 0.486; Figure 2b; Table S4; lnCVR difference = -0.024,CI = -0.173 to 0.124, p = 0.750, R 2 = 0.001; Table S5).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: One-off and multigenerational experiment data represent different biological processes and questions.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    Many studies measured glucose in blood.

    “many studies measured glucose in blood”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Findingown data

    27 effect sizes comparing body weight of exposed to non-exposed grandparents of the same sex were obtained.

    “We obtained 27 effect sizes comparing body weight of exposed to non-exposed grandparents of the same sex.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the multigenerational lnRR dataset, there was a positive relationship between effect size and sqrt(VlnRR) (slope 1.605, 95% CI 0.510 to 2.700, p=0.004).

    “sqrt(VlnRR) 1.605 0.510 2.700 0.004”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Findingown data

    F0 body weight around the end of exposure to obesogenic diets was collected from 17 studies that reported this information.

    “We collected F0 body weight around the end of exposure to obesogenic diets from 17 studies that reported this information.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    The Lineage_HFD variable was not used during analyses due to the unbalanced distribution of data points across levels of this factor.

    “we had a new Lineage_HFD variable, which represents both breeding designs and exposure transmission line, but was not used during analyses due to the unbalanced distribution of data points across levels of this factor.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    The ages of F0 body weight measurements were centered around 100 days of age, when most of the exposures finished (at or after F0 mating).

    “The ages were centered around 100 days of age, when most of the exposures finished (at or after F0 mating).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    For some trait categories it was not possible to bring all measurements to the same units.

    “for some of the trait categories it was not possible to bring all the measurements to the same units”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingnone given

    In some cases, authors reported the incorrect statistics name (e.g., SE instead of SD).

    “In some cases, we deemed authors to have reported the incorrect statistics name (e.g., SE instead of SD).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the one-off dataset with lnRR as the effect size, the meta-regression intercept for the effect of scaled relative protein content of obesogenic diets was 0.085 (95% CI -0.074 to 0.244, p = 0.295).

    “One-off lnRR 0.001 Intercept 0.085 -0.074 0.244 0.295 Slope 0.006 -0.017 0.028 0.629”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). The quote only partly states this claim.

    Findingown data

    In the model for multigenerational grandoffspring body weight (lnRR), the slope for age at measurement was 0.001 (95% CI 0.000 to 0.001, p < 0.001).

    “MG grandoffspring body weight lnRR 0.062 Intercept 0.060 -0.080 0.201 0.401 Slope 0.001 0.000 0.001 0.000”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the model for grandparents' body weight (lnRR), the slope for age at measurement was 0.001 (95% CI 0.001 to 0.002, p < 0.001).

    “Grandparents body weight lnRR 0.220 Intercept 0.009 -0.093 0.110 0.870 Slope 0.001 0.001 0.002 0.000”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the one-off dataset with lnRR as the effect size, the slope for scaled relative protein content of obesogenic diets was -0.008 (95% CI -0.029 to 0.012, p = 0.413).

    “One-off lnRR 0.002 Intercept 0.083 -0.075 0.241 0.301 Slope -0.008 -0.029 0.012 0.413”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the multigenerational dataset with lnCVR as the effect size, the meta-regression intercept for the effect of scaled relative protein content of obesogenic diets was -0.073 (95% CI -0.288 to 0.143, p = 0.508).

    “Multigenerational lnCVR 0.000 Intercept -0.073 -0.288 0.143 0.508 Slope -0.003 -0.127 0.121 0.964”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the multigenerational dataset with lnRR as the effect size, the slope for scaled relative protein content of obesogenic diets was 0.063 (95% CI -0.069 to 0.194, p = 0.351).

    “Multigenerational lnRR 0.017 Intercept 0.354 0.085 0.623 0.010 Slope 0.063 -0.069 0.194 0.351”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the one-off dataset with lnCVR as the effect size, the slope for scaled relative protein content of obesogenic diets was -0.053 (95% CI -0.162 to 0.055, p = 0.336).

    “One-off lnCVR 0.015 Intercept 0.015 -0.145 0.174 0.856 Slope -0.053 -0.162 0.055 0.336”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the one-off dataset with lnRR as the effect size, the slope for scaled relative protein content of obesogenic diets was 0.006 (95% CI -0.017 to 0.028, p = 0.629).

    “One-off lnRR 0.001 Intercept 0.085 -0.074 0.244 0.295 Slope 0.006 -0.017 0.028 0.629”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the multigenerational dataset with lnCVR as the effect size, the slope for scaled relative protein content of obesogenic diets was -0.003 (95% CI -0.127 to 0.121, p = 0.964).

    “Multigenerational lnCVR 0.000 Intercept -0.073 -0.288 0.143 0.508 Slope -0.003 -0.127 0.121 0.964”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    For the multigenerational lnCVR dataset, the top model included trait as a moderator (AICc = 247.99, weight = 0.30).

    “Multigenerational lnCVR Trait 11 -111.92247.99 0.00 0.30”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    For the one-off lnCVR dataset, the top model was intercept-only (AICc = 426.84, weight = 0.53).

    “One-off lnCVR Intercept-only (no moderators) 4 -209.35426.84 0.00 0.53”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the model for multigenerational grandoffspring body weight (lnRR), the meta-regression intercept for the effect of age at measurement was 0.060 (95% CI -0.080 to 0.201, p = 0.401).

    “MG grandoffspring body weight lnRR 0.062 Intercept 0.060 -0.080 0.201 0.401 Slope 0.001 0.000 0.001 0.000”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the one-off dataset with lnRR as the effect size, the meta-regression intercept for the effect of scaled relative protein content of obesogenic diets was 0.083 (95% CI -0.075 to 0.241, p = 0.301).

    “One-off lnRR 0.002 Intercept 0.083 -0.075 0.241 0.301 Slope -0.008 -0.029 0.012 0.413”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the model for grandparents' body weight (lnRR), the meta-regression intercept for the effect of age at measurement was 0.009 (95% CI -0.093 to 0.110, p = 0.870).

    “Grandparents body weight lnRR 0.220 Intercept 0.009 -0.093 0.110 0.870 Slope 0.001 0.001 0.002 0.000”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the model for one-off grandoffspring body weight (lnRR), the meta-regression intercept for the effect of age at measurement was 0.056 (95% CI -0.006 to 0.118, p = 0.076).

    “OF grandoffspring body weight lnRR 0.006 Intercept 0.056 -0.006 0.118 0.076 Slope 0.000 0.000 0.000 0.272”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the multigenerational dataset with lnRR as the effect size, the slope for scaled relative protein content of obesogenic diets was 0.068 (95% CI -0.077 to 0.213, p = 0.361).

    “Multigenerational lnRR 0.020 Intercept 0.367 0.102 0.632 0.007 Slope 0.068 -0.077 0.213 0.361”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the multigenerational dataset with lnCVR as the effect size, the slope for scaled relative protein content of obesogenic diets was -0.040 (95% CI -0.184 to 0.104, p = 0.585).

    “Multigenerational lnCVR 0.006 Intercept -0.073 -0.292 0.147 0.517 Slope -0.040 -0.184 0.104 0.585”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the one-off dataset with lnCVR as the effect size, the meta-regression intercept for the effect of scaled relative protein content of obesogenic diets was 0.015 (95% CI -0.145 to 0.174, p = 0.856).

    “One-off lnCVR 0.015 Intercept 0.015 -0.145 0.174 0.856 Slope -0.053 -0.162 0.055 0.336”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    For the one-off lnRR dataset, the top model included trait as a moderator (AICc = -277.83, weight = 0.69).

    “One-off lnRR Trait 11 150.42 -277.830.00 0.69”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the model for one-off grandoffspring body weight (lnRR), the slope for age at measurement was 0.000 (95% CI 0.000 to 0.000, p = 0.272).

    “OF grandoffspring body weight lnRR 0.006 Intercept 0.056 -0.006 0.118 0.076 Slope 0.000 0.000 0.000 0.272”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the one-off dataset with lnCVR as the effect size, the slope for scaled relative protein content of obesogenic diets was -0.008 (95% CI -0.123 to 0.108, p = 0.897).

    “One-off lnCVR 0.000 Intercept 0.036 -0.138 0.210 0.684 Slope -0.008 -0.123 0.108 0.897”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the multigenerational dataset with lnCVR as the effect size, the meta-regression intercept for the effect of scaled relative protein content of obesogenic diets was -0.073 (95% CI -0.292 to 0.147, p = 0.517).

    “Multigenerational lnCVR 0.006 Intercept -0.073 -0.292 0.147 0.517 Slope -0.040 -0.184 0.104 0.585”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the one-off dataset with lnCVR as the effect size, the meta-regression intercept for the effect of scaled relative protein content of obesogenic diets was 0.036 (95% CI -0.138 to 0.210, p = 0.684).

    “One-off lnCVR 0.000 Intercept 0.036 -0.138 0.210 0.684 Slope -0.008 -0.123 0.108 0.897”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The composition of high-fat diets used can vary widely among studies.

    “The composition of high-fat diets used can vary widely among studies 35 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [35]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    Both maternal and paternal effects can be carried beyond the F1 generation and may involve epigenetic mechanisms.

    “Further, both maternal and paternal effects can be carried beyond the F1 (offspring) generation and this may involve epigenetic mechanisms, which work to alter the phenotype through changes in gene expression rather than eliciting changes to the DNA sequence itself [23][24][25][26][27] .”
    Quote in the paperSubstantial claim0 of 5 sources checked
    • cites [23]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [24]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [25]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [26]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [27]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    A variance-covariance matrix was used to account for non-independence from multiple treatment groups compared against the same control group.

    “To account for this non-independence, we calculated a variance-covariance matrix (equations (19.18) and (19.19) in Olkin and Glesser 64 ) and used it as a variance component in meta-analytic and meta-regression models.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [64]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Two researchers independently screened all records.

    “Two researchers (HA and AA, and then HA and ML for the search update) independently screened all records to locate potentially relevant studies.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The grand-offspring of exposed grandparents display phenotypes consistent with the metabolic syndrome, especially if the effect has been exacerbated by multigenerational exposure to obesogenic diets.

    “Overall, we have demonstrated that the grand-offspring of exposed grandparents display phenotypes consistent with the metabolic syndrome, especially if the effect has been exacerbated by multigenerational exposure to obesogenic diets.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    In the dataset, approximately 80% of effect sizes were from F2 grand-offspring and the remaining 20% from F3 grand-offspring: 82 effect sizes overall, including 58 from one-off exposure experiments (only 44 from great-grand-maternal exposure).

    “In our dataset, approximately 80% of effect sizes were from F2 grand-offspring.The remaining 20% of effect sizes is from F3 grandoffspring: 82 effect sizes overall, including 58 from one-off exposure experiments (only 44 from great-grand-maternal exposure).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    It has been suggested that one sex may be more sensitive than the other due to the role of sex chromosomes and factors acting during development, such as ontogenetic changes in gene expression.

    “Although mechanisms underlying sex-specific effects remain poorly understood, it has been suggested that one sex may be more sensitive than the other due to the role of sex chromosomes 79 and factors acting during development, such as ontogenetic changes in gene expression 80,81 .”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [79]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [80]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [81]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Interpretationreasoning

    The lack of a statistically significant grandparental sex effect in one-off exposure experiments is in line with the fact that effect sizes are both small and highly heterogeneous, and there are relatively few studies with grand-paternal exposures.

    “The lack of a statistically significant grandparental sex effect in one-off exposure experiments is in line with the fact that effect sizes are both small and highly heterogeneous, and there are relatively few studies with grand-paternal exposures (5 out of 18 articles with one-off exposure).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    Publication bias was assessed by visually inspecting funnel plots, modified multilevel Egger's regression, and time-trend meta-regression.

    “We assessed publication bias in the one-off and multigenerational datasets by visually inspecting funnel plots for asymmetry 68 in the distribution of the residuals of effect sizes (sensu 62 ) from a full multivariate model with all three fixed effects.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [68]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    A protocol of all methods was registered on the Open Science Framework before searches.

    “Before searches, we registered a protocol of all our methods on the Open Science Framework (https://osf.io/sg6wj/register/565fb3678c5e4a66b5582f67).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    The systematic review and meta-analysis followed PRISMA guidelines.

    “We followed The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines 53 in reporting our systematic review and meta-analysis.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [53]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Leptin is one of the major players involved in energy homeostasis; it is produced by adipocytes and has a central role in the regulation of food intake and body weight.

    “Leptin is one of the major players involved in energy homeostasis.It is produced by adipocytes and has a central role in the regulation of food intake and body weight 85 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [85]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Work in flies showed that high-fat diets increase phenotypic variation.

    “While work in flies showed that high-fat diets increase phenotypic variation 91 , we found that triglyceride levels varied significantly less in treatment than in the control grand-offspring (32% difference).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [91]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    A ceiling-effect scenario was previously proposed to explain the lack of increase in glucose concentrations of rodents under a high-fat diet.

    “Such a scenario was previously proposed to explain the lack of increase in glucose concentrations of rodents under a high-fat diet 92 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [92]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim. May combine two claims.

    Premisereasoning

    The effect sizes used are standardized, so they are unitless and should not be affected by original measurement units.

    “The effect sizes we used are standardized, so they are unitless and should not be affected by original measurement units.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    The F2 generation may already be at an adverse risk of developing metabolic disease, and a minimum F3 generation is required to detect true transgenerational effects.

    “As such, the F2 generation may already be at an adverse risk of developing metabolic disease, and a minimum F3 generation is required to detect true transgenerational effects.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    The minimum protein to support adequate growth and reproduction in mice is ~16% in rats and ~14%, by weight.

    “The minimum protein to support adequate growth and reproduction in mice is ~16% in rats and ~14%, by weight 93 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [93]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    After independent screening, decision conflicts were crosschecked and discussed, and final inclusion/exclusion decisions were made via consensus.

    “After independent screens, each record with a decision conflict was crosschecked and discussed, and a final decision to include or exclude was made via consensus.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Transgenerational transfer of dietary effects is weak but durable.

    “We also found no difference between F2-and F3generation grand-offspring in the average magnitude of effects (for both one-off and multigenerational studies), indicating weak, but durable, transgenerational transfer of dietary effects.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Two researchers (HA, AA) independently screened all records.

    “Two researchers (HA, AA) independently screened all records.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Low protein intake may decrease thermogenesis and satiety, as well as lead to an increase in subsequent energy intake.

    “This finding is surprising given the expectation that low protein intake may decrease thermogenesis and satiety, as well as lead to an increase in subsequent energy intake 94 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [94]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    Foods high in energy have a robust and significant effect on satiety and satiation, thereby facilitating overconsumption of fat, leading to obese phenotypes.

    “Foods high in energy have been shown to have a robust and significant effect on satiety and satiation, thereby facilitating overconsumption of fat, leading to obese phenotypes 96 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [96]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Individuals with obesity not only have elevated leptin levels but also develop leptin resistance, where increased circulating leptin does not reduce appetite and body mass.

    “However, individuals with obesity not only have elevated leptin levels but also develop 'leptin resistance', a complex phenomenon, where increased circulating leptin does not reduce appetite and body mass 86,87 .”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [86]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [87]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S2 (checkable).

    Premisecitation

    Truly transgenerational effects are those that are found in generations that were not exposed to the factor that triggered the change in phenotype.

    “Thirdly, truly transgenerational effects are those that are found in generations that were not exposed to the factor that triggered the change in phenotype 29 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [29]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Interpretationcitation

    The surprisingly small effect on body weights coupled with large effects on physiology in multigenerational experiments is in line with the concept of normal weight obesity, which highlights the need to stratify risk based on underlying metabolic factors, such as adiposity, and metabolic changes, rather than body weight alone.

    “The surprisingly small effect on body weights coupled with large effects on physiology in multigenerational experiments is in line with the concept of 'normal weight obesity', which highlights the need to stratify risk based on underlying metabolic factors, such as adiposity, and metabolic changes, rather than body weight alone 90 .”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [90]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    When an F0 father is exposed, their germline (future F1) is exposed, but not F2 germ cells; thus, in grand-paternal exposure studies, F2 is sufficient for detecting true transgenerational effects.

    “When an F0 father is exposed, their germline (future F1) is exposed, but not F2 germ cells.Thus, in grand-paternal exposure studies, F2 is sufficient for detecting true transgenerational effects 29,31 .”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [29]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [31]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationcitation

    Funnel plot asymmetry may stem from true unexplained biological heterogeneity, not publication bias.

    “Funnel plot asymmetry may stem from true unexplained biological heterogeneity, not publication bias 69 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [69]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    When an F0 mother is exposed to an obesogenic diet, the developing offspring (F1) is directly affected in utero and during lactation, as well as F2 germ cells inside these developing F1 offspring.

    “When an F0 mother is exposed to an obesogenic diet, the developing offspring (F1) is directly affected in utero and during lactation, as well as F2 germ cells inside these developing F1 offspring 30 .”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [30]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    The analysis used a Study_ID variable instead of Paper_ID to account for multiple data points from the same or similar experiments.

    “we created a Study_ID variable, which was used instead of Paper_ID in the analyses.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Protein content is not a major factor in regulating energy intake and causing an obesogenic response (i.e., adiposity), rather dietary fat is.

    “However, our result is consistent with a study showing that protein content is not a major factor in regulating energy intake and causing an obesogenic response (i.e., adiposity), rather dietary fat is 95 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [95]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim.

    Premisecitation

    Increased insulin is required to maintain normal glucose tolerance, also known as compensatory hyperinsulinemia.

    “Increased insulin is required to maintain normal glucose tolerance, also known as compensatory hyperinsulinemia 88 .”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [88]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    A protocol of all methods was registered on the Open Science Framework.

    “We registered a protocol of all our methods on the Open Science Framework (https://osf.io/sg6wj/register/565fb3678c5e4a66b5582f67).”
    Quote in the paperNot a substantial claim
77 of 77

  • The paper relies on this without support.

    There is very little work synthesizing studies on transgenerational effects of obesogenic diets.

    “Despite overwhelming evidence suggesting detrimental impacts associated with obesogenic diets, there is very little work synthesizing studies on transgenerational effects.”

  • The paper relies on this without support.

    Results of studies on transgenerational impacts of maternal and parental obesogenic diets on subsequent generations have largely been mixed.

    “A multitude of studies have investigated the transgenerational impacts of maternal and parental obesogenic diets on subsequent generations of offspring, but results have largely been mixed.”

  • The paper relies on this without support.

    One-off exposure experiments expose subtle transgenerational epigenetic changes, while multigenerational exposure experiments show cumulative effects of continuous exposure across generations so that epigenetic influences can interact with direct effects of obesogenic diets on development.

    “Specifically, one-off exposure experiments expose subtle transgenerational epigenetic changes, while multigenerational exposure experiments show cumulative effects of continuous exposure across generations so that epigenetic influences can interact with direct effects of obesogenic diets on development.”

  • The paper relies on this without support.

    One-off and multigenerational experiment data represent different biological processes and questions.

    “We performed analyses separately for one-off and multigenerational experiment data because they represent different biological processes and questions.”

  • The paper relies on this without support.

    In mammals, mothers can pass nutritional insults to offspring during gestation and lactation, while fathers influence offspring only via sperm and seminal fluid during mating.

    “In mammals, mothers can easily pass on nutritional insults to their offspring during gestation and lactation.Fathers can influence their offspring only via sperm and seminal fluid during mating.”

  • The paper relies on this without support.

    The meta-analysis is the first to examine differences in variance between obesogenic and control group diets.

    “To our knowledge, our meta-analysis is the first to examine differences in variance between obesogenic and control group diets.”

  • The paper relies on this without support.

    Body weight data from the same cohort at the time point closest to the adiposity measurement is appropriate for recalculating fat mass from proportion of body weight.

    “For adiposity data presented as proportion of body weight, we recalculated grams / milligrams of fat using associated body weight data for the same cohort of animals at the time point closest to the adiposity measurement.”

  • The paper relies on this without support.

    Total diet energy (kcal/g) and percentage of energy from fat are the key indicators of the obesogenic potential of the used diets.

    “For statistical analyses, we considered total diet energy (kcal / g), percentage of energy from fat as the key indicators of the obesogenic potential of the used diets.”
8 of 8

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Exposure-type difference is called non-significant despite a CI excluding zero

    Major

    The reported 95% confidence interval for the exposure-type difference runs from −0.256 to −0.118 and therefore excludes zero. That contradicts the statement that the difference was statistically non-significant.

  • Full-model explained variance disagrees with Table S9

    Major

    The text says the three moderators jointly explained 38% of variation in multigenerational mean-trait effects. Table S9 reports R2 = 0.096, or 9.6%, for that full model.

  • Table S16 assigns more F3 effects to one-off studies than exist overall

    Major

    Table S16 lists 325 F3 effect sizes for the one-off dataset, whereas the dataset description says there were only 82 F3 effect sizes across all studies. The table also reverses the overall F2 and F3 counts.

  • Variability contrast is labeled lnRR instead of lnCVR

    Minor

    The results call the grandparent-sex contrast in trait variability an lnRR difference. Table S6 identifies the same estimate and interval as an lnCVR contrast.

  • Discussion calls all one-off trait effects non-significant

    Minor

    The discussion says all one-off effects were statistically non-significant, but the results report significant one-off effects for adiposity (p = 0.023) and leptin (p = 0.022).

5 of 5

Cites

Flagged first, then by citations

Research works printed but not linked: 2

  1. Team Rs, 2020, RStudio: integrated development for R
  2. Sutton AJ, 2009, Publication bias, Handb Res Synth Meta‐Analysis, 2, 435

Cited by

Sorted by citations

Related

Authors
Anwer, H., Morris, M. J., Noble, D. W. A., Nakagawa, S., & Lagisz, M.
Affiliations
  • UNSW Sydney (AU)
  • Australian National University (AU)
ORCID
Venue
Obesity Reviews
Publisher
Wiley
Published
2021-09-30
Volume
23
Issue
1
Pages
e13342-e13342
ISSN
  • 1467-7881 (ISSN-L)
  • 1467-789X
DOI
10.1111/obr.13342
OpenAlex
W3204084905
Semantic Scholar
60e328daa4340043005b714c28171b0013719b93
Cited by
14
References
133
Citations by year
2026: 3 · 2025: 1 · 2024: 2 · 2023: 4 · 2022: 4
FWCI
2.27
Type
journal article
Language
en
Open access
Licence
Locations
  • Obesity Reviews (publishedVersion)
  • PubMed (publishedVersion)
  • EcoEvoRxiv (submittedVersion, cc-by-nd)
  • ANU Open Research (Australian National University) (submittedVersion)
  • EcoEvoRxiv (submittedVersion)
Funders
Australian Research Council (DP180100818)
Topics
  • Health Sciences › Medicine › Pediatrics, Perinatology and Child Health › Birth, Development, and Health
  • Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Diet, Metabolism, and Disease
  • Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Obesity, Physical Activity, Diet
Keywords
obesogenic diet, transgenerational effects, maternal diet, paternal diet, high-fat diet, metabolic syndrome, body weight, adiposity, glucose, insulin, leptin, triglycerides, phenotypic variability, multigenerational exposure, rodent models
MeSH terms
Animals, Female, Obesity, Pregnancy, Prenatal Exposure Delayed Effects, Rodent Diseases, Rodentia, Metabolic Syndrome, Diet, High-Fat

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-23.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-23.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review