Graphical Abstract: Iodine (I)- and selenium (Se)-biofortified lettuce provides dietary fibres and micronutrients that are processed by the gut microbiome. Microbial fermentation of dietary fibres generates short-chain fatty acids (SCFAs), which act as key signalling metabolites linking diet to host epigenetic regulation. SCFAs inhibit histone deacetylase 3 (HDAC3), a corepressor that, in the absence of thyroid hormone (T3), is recruited by thyroid hormone receptors (THRA/THRB) to repress transcription of target genes, including murine DNMT3A. In parallel, iodine serves as an essential structural component of thyroid hormones (T4 and T3), while selenium is required for the activity of deiodinases (DIO1–3), with DIO1 and DIO2 converting thyroxine (T4) into biologically active T3. Through this dual input—microbiome-derived metabolites and micronutrient-dependent thyroid hormone activation—the diet–microbiome axis converges on chromatin-modifying pathways. The coordinated contribution of dietary iodine and selenium with microbiome-derived metabolites in shaping epigenetic states and maintaining intestinal and systemic homeostasis remains largely unexplored and constitutes the central mechanistic question addressed by the NUMIEP project. 

1.Iodine and selenium are essential micronutrients required for thyroid hormone (TH) synthesis, metabolic regulation, neurodevelopment, redox homeostasis and overall endocrine balance. Current public health recommendations to lower dietary salt intake (WHO, EFSA) have increased the need for salt-independent sources of organic-bound iodine and selenium. Agronomic biofortification of vegetables—particularly lettuce—has demonstrated high efficiency in increasing the accumulation and bioavailability of these micronutrients while maintaining plant quality and safety (Smoleń et al. 2016a,b).

2. Although multiple agronomic studies confirm the feasibility and safety of I+Se biofortification, virtually nothing is known about how such biofortified foods affect consumer physiology—particularly through interactions with the gut microbiome and its metabolite production. The gut microbiota metabolises organic-bound micronutrients and ferments lettuce-derived dietary fibres into short-chain fatty acids (SCFAs)—notably butyrate (C4), propionate (C3) and acetate (C2) (Krautkramer et al. 2017). Beyond SCFAs, the microbiota also generates additional metabolites such as indoles and phenolic derivatives, which function as signalling molecules; however, their physiological relevance in the context of biofortified functional foods remains largely unexplored (Kadayifci et al. 2018).

3. Diet-derived one-carbon nutrients—such as folates, vitamins B6, B9 and B12 —are well-established regulators of DNA methylation pathways (Kadayifci et al. 2018). In contrast, the extent to which microbiota-specific metabolites influence epigenetic regulation remains far less understood. Microbial metabolites including SCFAs modulate chromatin structure by influencing histone acetylation and DNA methylation (Krautkramer et al. 2017). Butyrate is a particularly potent modulator that inhibits histone deacetylases (HDACs), thereby increasing histone acetylation, enhancing chromatin accessibility, and regulating both global and locus-specific DNA methylation (Woo & Alenghat 2022; Silva et al. 2020). Furthermore, SCFAs modify the methylation potential of host cells by altering intracellular S-adenosylmethionine (SAM) and S-adenosylhomocysteine (SAH) levels—SAM activating DNMTs and SAH inhibiting them (Woo et al. 2022). However, how these epigenetic processes respond specifically to I+Se biofortified lettuce remains insufficiently investigated.

4. Thyroid-related genes—including Thyroid Hormone Receptors (TRs: THRB and THRA), as well as the selenium-dependent deiodinases DIO1 and DIO2—represent highly epigenetically sensitive loci (Wojcicka et al. 2014). THRB undergoes promoter-specific hypermethylation in pathological settings and functions as an environmentally responsive epigenetic hub capable of recruiting HAT or HDAC complexes depending on ligand availability  (Saponaro et al. 2020). Epigenome-wide association studies show that CpG methylation strongly correlates with circulating thyroid hormone levels (Weihs et al. 2023). Moreover, THRB can regulate DNA methyltransferase DNMT3A expression (Kyono et al. 2016) and be regulated by HDAC3 in a negative feedback look, linking thyroid hormone signalling directly to the environmentally responsive de novo DNA-methylation machinery, including their modulation by dietary factors (Kadayifci et al. 2018, Seebacher et al., 2024). Taken together, TH receptors remain engaged locally in epigenetic feedback loops with both DNMT3A and butyrate-inhibited HDAC3—a relationship that remains poorly defined.

5. Enzymes such as DNMT3A and HDACs link nutrient-derived signals to genome-wide methylation and chromatin structure, but their systemic modulation—particularly by microbiome-dependent metabolites—shows paradoxical outcomes, including concurrent global hypomethylation and locus-specific hypermethylation observed in pathological states (Festuccia et al., 2017). This epigenetic paradox suggests that environmental factors, including functional foods and microbiota-derived metabolites, influence chromatin organization in ways that remains insufficiently understood. 

6. Despite substantial progress in understanding the microbiome–epigenome axis, no study has investigated whether microbial metabolites produced during the digestion of I+Se biofortified lettuce can reprogram the local epigenetic state of key thyroid hormone–related genes (THRB, THRA, DIO1, DIO2, DIO3) or induce genome-wide epigenetic remodeling within the intestine. This constitutes a critical unresolved mechanism at the interface of nutritional epigenomics, microbiome biology, and thyroid–gut axis regulation. Resolving these mechanistic gaps is essential for understanding how diet shapes long-term health trajectories and contributes to either risk or protection against major civilization-related diseases—including cancer, inflammatory disorders, neurodegeneration, cardiovascular disease, and aging (Fransquet et al. 2019). For more details, please visit the References and our Team

The NUMIEP project is structured into seven tightly integrated work packages spanning the diet–microbiome–metabolome–epigenome–phenome continuum. WP1 (PL) generates chemically and nutritionally standardised iodine- and selenium-biofortified lettuce, providing reproducible dietary intervention material for all downstream studies. WP2 (PL) evaluates the systemic effects of I+Se lettuce in an in vivo mouse feeding model, with comprehensive biobanking of tissues, biofluids and fecal samples under harmonised SOPs. WP3 (CZ) uses human and mouse fecal material to perform in vitro microbial fermentation, microbiome profiling and metabolomic characterisation, producing defined SCFA-rich and non-SCFA metabolite-conditioned media. WP4 (CZ) applies these fermentation-derived metabolite fractions to human intestinal epithelial 2D and 3D models to characterise early epigenomic, transcriptomic and functional phenotypic responses. WP5 (PL) integrates microbiome, metabolome, epigenome and transcriptome datasets using multi-layer network-based approaches to reconstruct causal diet–microbiome–epigenome pathways and prioritise regulatory nodes. WP6 (PL & CZ) experimentally validates these nodes using targeted gain- and loss-of-function approaches in epithelial models to establish causal metabolite-driven mechanisms underlying chromatin remodelling and epigenetic regulation. WP7 (PL & CZ) ensures project coordination, FAIR data management, dissemination and publication, culminating in the NUMIEP Atlas—an integrated resource linking dietary biofortification to microbiome-driven epigenomic and phenotypic reprogramming.

Input material preparation (supports all downstream WPs). This WP produces chemically and nutritionally standardised batches of I+Se lettuce with validated micronutrient composition (I+Se speciation, stability, homogeneity). These batches constitute the core dietary intervention material for in vivo experiments and fermentation assays, ensuring reproducibility across PL–CZ laboratories. Milestone M1 (Mo 1–6): QC-validated, standardised I+Se and control lettuce batches ready for WPs 2-4.

Production. This WP integrates microbiome modulation and microbial metabolite profiling. Using control human and experimental mouse fecal samples, as well as reference butyrogenic bacterial cultures, it assesses how I+Se lettuce influences microbial community structure through 16S rRNA/NGS profiling and shotgun-assisted functional inference. In vitro fermentation generates SCFA-rich and non-SCFA metabolite fractions (indoles, phenolics, redox-active metabolites), which serve as defined stimuli for downstream epithelial and organoid assays in WP4. Milestone M3 (Mo 6–18): Fermentation-conditioned media, microbiome and metabolome datasets, and SCFA/non-SCFA metabolite profiles.

Production. This WP integrates microbiome modulation and microbial metabolite profiling. Using control human and experimental mouse fecal samples, as well as reference butyrogenic bacterial cultures, it assesses how I+Se lettuce influences microbial community structure through 16S rRNA/NGS profiling and shotgun-assisted functional inference. In vitro fermentation generates SCFA-rich and non-SCFA metabolite fractions (indoles, phenolics, redox-active metabolites), which serve as defined stimuli for downstream epithelial and organoid assays in WP4. Milestone M3 (Mo 6–18): Fermentation-conditioned media, microbiome and metabolome datasets, and SCFA/non-SCFA metabolite profiles.

Human colon 3D spheroids and 2D epithelial cultures (Caco-2, HT-29) are exposed to control and metabolite-rich fermentation fractions (conditioned media and controls) to characterise early chromatin, transcriptional and phenotypic responses. Toxicity, efficacy measurements as well as functional and structural phenotyping (eg. barrier integrity assays, oxidative-stress indicators) and morphometric measurements—will be performed so that phenotypic readouts can be interpreted as downstream reflections of epigenetic and transcriptional changes. Milestone M4 (Mo 8–24): Completed biobank of treated human cells & spheroids with annotated metadata, including genotypic and phenotypic readouts, for WPs 5–6.

This WP performs the full multi-omics analysis pipeline, including global and locus-specific DNA methylation profiling, chromatin-accessibility mapping (ATAC-seq), bulk and single-nucleus transcriptomics, microbiome data processing and integration of metabolomics from WP3. Datasets from WPs 2–4 are quality-controlled and integrated using network-based, multi-layer modelling to generate causal maps linking I+Se-driven microbial and metabolite shifts to epigenomic states, chromatin accessibility and TH- transcriptional regulation. These analyses yield a prioritised list of regulatory nodes for mechanistic validation in WP6. Milestone M5 (Mo 11–32): Completed microbiome 16S-NGS datasets, multi-omics (bulk RNAseq, snRNA-seq, snATAC-seq, global and locus-specific methylation) datasets, integrated mechanistic model, and a prioritised list of causal epigenetic and transcriptional nodes.

Using prioritised nodes from WP5, this WP performs targeted gain- and loss-of-function studies to validate key metabolite-driven mechanisms in epithelial models, including effects on cell-state transitions. Selected metabolite–gene–chromatin interactions are experimentally validated to establish causal pathways underlying the epigenetic paradox—global hypomethylation coupled with locus-specific hypermethylation—as in the Methods.  Milestone M6 (Mo 20–35): Experimentally validated metabolite-driven epigenomic mechanisms.

Ensures data harmonisation, metadata standardisation, and FAIR-compliant deposition of sequencing, metabolomic and epigenomic datasets. Oversees cross-border coordination, protocol integration, dissemination, training, and public communication via the NUMIEP platform. Milestone M7 (Mo 1–36): Published diet–microbiome–epigenome atlas (NUMIEP Atlas) and final integrated resource.  

Aneta Koronowicz et al. developed a health-oriented innovation at the University of Agriculture in Krakow within the Inkubator Innowacyjności 4.0 project. She created a dietary supplement based on lettuce biofortified with organic iodine compounds, which exhibits immunomodulatory properties and is intended to help prevent iodine deficiency. She propose this product as an alternative to traditional iodized table salt, offering potential health benefits such as reduced cardiovascular risk and support for the prevention and adjunctive treatment of COVID-19, due to its comprehensive nutritional and functional properties.

Prof. David Sinclair proposes that aging results from accumulated epigenetic noise caused by imperfect DNA repair. Stressors like fasting, intense exercise, or cold activate longevity pathways that enhance cellular maintenance. Molecules such as NMN may aid this process, while partial Yamanaka factor reprogramming has shown potential to reverse aging in animal models.  "Avoiding DNA damage, reducing calorie and protein intake, engaging in high-intensity training, and exposing the body to controlled cold or heat stress all activate repair pathways that help maintain a healthy epigenome."
Kim et al. 2023  (PMID:37463842) Host-microbiome interactions in nicotinamide mononucleotide (NMN) deamidation
Kane et al. 2024 (PMC11230277) Long-term NMN treatment increases lifespan and healthspan in mice in a sex dependent manner

The four Yamanaka factors—c: Oct4 (POU5F1), Sox2 (SOX2), Klf4 (KLF4), and c-Myc (MYC)—can partially reset the epigenome of adult cells to a more youthful state without erasing their cellular identity, and recent preliminary reports describe SB000 as a proprietary single-factor candidate for cellular rejuvenation, although its molecular identity has not been disclosed.
Kovatcheva et al. 2023
Aguirre  et al. 2023
Camillo et al. 2025

This video explains how DNA damage over time leads to “epigenetic drift,” which disrupts normal gene regulation and contributes to aging. It also describes how failures in DNA repair mechanisms accelerate this epigenetic deterioration, helping to explain why cellular function declines with age.
Borrego-Ruiz and Juan J Borrego 2024