The research report should be a detailed narrative explaining the function, biological processes, and localization of the gene product. Citations should be given for all claims.
You should prioritize authoritative reviews and primary scientific literature when conducting research. You can supplement
this with annotations you find in gene/protein databases, but these can be outdated or inaccurate.
We are specifically interested in the primary function of the gene - for enzymes, what reaction is catalyzed, and what is the substrate specificity? For transporters, what is the substrate? For structural proteins or adapters, what is the broader structural role? For signaling molecules, what is the role in the pathway.
We are interested in where in or outside the cell the gene product carries out its function.
We are also interested in the signaling or biochemical pathways in which the gene functions. We are less interested in broad pleiotropic effects, except where these elucidate the precise role.
Include evidence where possible. We are interested in both experimental evidence as well as inference from structure, evolution, or bioinformatic analysis. Precise studies should be prioritized over high-throughput, where available.
The requested identity is internally consistent: DNMT3L denotes human DNA (cytosine-5)-methyltransferase 3-like protein (UniProt Q9UJW3), not a different similarly named gene. Its literature-described ADD/PHD-like histone-reader region and C-terminal DNA-methyltransferase-like region agree with the supplied InterPro assignments. Crucially, however, DNMT3L is a catalytically inactive DNMT3-family regulatory protein, not an enzyme that independently methylates DNA. It couples recognition of chromatin lacking H3K4 methylation to recruitment, oligomerization, and stimulation of the active de novo methyltransferases DNMT3A/DNMT3A2 and, in some contexts, DNMT3B (jia2007structureofdnmt3a pages 1-2, noh2016readingbetweenthe pages 3-4, salle2006originofdna pages 35-39).
Its best-established biological function is in the nucleus of developing germ cells, where DNMT3L-containing chromatin complexes facilitate de novo CpG methylation at imprinting-control regions, dispersed repeats/retrotransposons, and additional genomic sequences. The decisive organismal evidence is predominantly from mice; direct endogenous human loss-of-function evidence remains limited. Accordingly, the core molecular annotation is high-confidence, whereas specific claims about human fertility or disease should be treated as conserved-function hypotheses rather than proven clinical relationships.
| Feature | Current conclusion | Key evidence | Confidence and limitations |
|---|---|---|---|
| Identity | DNMT3L, UniProt Q9UJW3, is the human DNA cytosine-5-methyltransferase 3-like protein. Its reported architecture agrees with the supplied Homo sapiens identity. | Mammalian DNMT3L is an ADD/PHD-like-domain-containing, methyltransferase-like paralogue of DNMT3A and DNMT3B (jia2007structureofdnmt3a pages 1-2, noh2016readingbetweenthe pages 15-18). | High for identity and family alignment. Much of the functional genetic evidence comes from mouse Dnmt3l rather than human Q9UJW3. |
| Catalytic status | DNMT3L is not an active DNA methyltransferase. It does not itself transfer methyl groups from SAM to cytosine. | Structural and biochemical evidence identifies a catalytically inactive methyltransferase-like domain that cannot bind AdoMet or AdoHcy normally (jia2007structureofdnmt3a pages 1-2, salle2006originofdna pages 35-39). | High. Describing DNMT3L as an enzyme without qualifying its catalytic inactivity is misleading. |
| ADD-domain chromatin recognition | The ADD/PHD-like domain recognizes the unmodified histone-H3 amino terminus, particularly H3K4me0. H3K4 mono-, di-, or trimethylation disrupts binding. | Histone-peptide and structural analyses demonstrate recognition of unmethylated H3K4 and discrimination against methylated H3K4 (noh2016readingbetweenthe pages 3-4, jia2007structureofdnmt3a pages 4-5, noh2016readingbetweenthe pages 15-18). | High for biochemical recognition. Genomic targeting additionally depends on transcription, chromatin context, DNA sequence, and partner proteins. |
| DNMT3A and DNMT3B cofactor function | DNMT3L binds and stimulates de novo methyltransferases, especially DNMT3A and DNMT3A2. DNMT3B stimulation is reported but is less comprehensively defined and may be context-dependent. | Protein interaction, interface mutagenesis, and activity experiments support DNMT3A stimulation; reviews also report DNMT3B activation (jia2007structureofdnmt3a pages 1-2, noh2016readingbetweenthe pages 3-4, jia2007structureofdnmt3a pages 5-10). | High for DNMT3A cooperation; moderate for the physiological importance and substrate scope of DNMT3B cooperation. |
| Complex architecture | The crystallized catalytic-region assembly is a 2:2 tetramer, DNMT3L–DNMT3A–DNMT3A–DNMT3L, containing two DNMT3A active sites. | The 2.9-Å structure measured approximately 160 × 60 × 50 Å. Both DNMT3A–DNMT3L and DNMT3A–DNMT3A interfaces were required for efficient catalysis (jia2007structureofdnmt3a pages 1-2, jia2007structureofdnmt3a pages 5-10). | High for the crystallized minimal complex. Full-length chromatin assemblies may be more dynamic or form larger oligomers. |
| Cellular site | DNMT3L functions in the nucleus at chromatin, where its ADD domain reads histone H3 and its C-terminal region recruits or activates DNMT3 enzymes acting on DNA. | Histone-tail recognition and DNMT3A–DNA complex structures place its molecular action at nucleosomal chromatin (noh2016readingbetweenthe pages 3-4, jia2007structureofdnmt3a pages 5-10, noh2016readingbetweenthe pages 15-18). | High as a mechanistic inference. The retrieved evidence did not provide a definitive human subnuclear-localization map. |
| Genomic imprint establishment | DNMT3L facilitates sex-specific germline methylation at imprinting control regions, including maternal imprints during oocyte growth and paternal marks at H19 and Dlk1–Gtl2 in male germ cells. | Genetic evidence supports DNMT3L-dependent maternal imprinting and loss of paternal-locus methylation in mutant mouse germ cells. Dlk1–Gtl2 methylation fell below 10% in one analysis (jia2007structureofdnmt3a pages 4-5, salle2006originofdna pages 69-75, salle2006originofdna pages 168-172). | High in mice; moderate for direct equivalence in human gametogenesis because comparable human loss-of-function evidence is limited. |
| Retrotransposon silencing | DNMT3L promotes methylation-dependent repression of dispersed repeats and retrotransposons, including LINE-1- and IAP-related sequences, in the male germ line. | Reviews and mutant-germ-cell studies report hypomethylation of repeat or retroviral elements in the absence of Dnmt3l (noh2016readingbetweenthe pages 3-4, jia2007structureofdnmt3a pages 4-5, salle2006originofdna pages 145-153). | High for a mouse germline role. Exact repeat-family targeting and direct effects in human germ cells remain incompletely established. |
| Germline and fertility phenotypes | Mouse Dnmt3l loss causes male infertility, germ-cell depletion, delayed or abnormal meiosis and synapsis, and female production of nonviable offspring because of defective oocyte methylation. | At 10 days postpartum, γH2AX-positive tubules numbered 28.2 ± 2.6 per 100 in mutants versus 74.9 ± 4.0 in controls; prenatal expression peaks during methylation acquisition (salle2006originofdna pages 69-75, salle2006originofdna pages 164-168, salle2006originofdna pages 161-164). | High for mouse knockout phenotypes. These outcomes are not demonstrated human clinical phenotypes. |
| Recent 2023–2024 context | Recent research primarily refines the broader germline-reprogramming and locus-targeting framework; it has not displaced the established model of DNMT3L as a noncatalytic H3K4me0 reader and DNMT3 cofactor. | The strongest direct mechanistic evidence remains foundational structural and mouse-genetic work; newer literature provides context rather than a new DNMT3L-specific mechanism (salle2006originofdna pages 69-75, salle2006originofdna pages 168-172, salle2006originofdna pages 87-91). | Moderate. Few 2023–2024 studies directly interrogated endogenous human DNMT3L protein function. |
| Translational status | DNMT3L remains principally a research target and mechanistic marker, not an established drug target, clinical diagnostic biomarker, or approved therapeutic target. | Open Targets reports several disease associations, but their low-to-moderate scores do not establish causality; no clinical implementation was identified (OpenTargets Search: -DNMT3L). | Low clinical maturity. Proposed links to infertility, cancer, cardiovascular disease, or inflammatory disease require functional and prospective human validation. |
Table: Compact evidence map for human DNMT3L (Q9UJW3), distinguishing established molecular mechanisms from mouse-derived biological functions and uncertain human translation.
The supplied target—gene symbol DNMT3L, organism Homo sapiens, UniProt Q9UJW3—matches the protein investigated in the DNMT3 literature. No conflicting gene with the same symbol was encountered. The protein is a paralogue of DNMT3A and DNMT3B and contains (i) an N-terminal cysteine-rich ADD domain, historically termed a PHD-like domain, and (ii) a C-terminal methyltransferase-like domain that mediates association with catalytic DNMT3 proteins. These features align with the supplied ADD, DNMT3_ADD/PHD, C5-methyltransferase, and SAM-dependent methyltransferase-superfamily annotations (jia2007structureofdnmt3a pages 1-2, noh2016readingbetweenthe pages 15-18).
The C-terminal annotation must not be interpreted as evidence of enzymatic activity. DNMT3L lacks critical residues needed for normal methyltransferase chemistry and cannot appropriately bind the methyl-donor/product ligands AdoMet/AdoHcy. It therefore has no established intrinsic DNA cytosine methyltransferase reaction (jia2007structureofdnmt3a pages 1-2, salle2006originofdna pages 35-39).
The reaction promoted by the DNMT3L-containing complex is transfer of a methyl group from S-adenosyl-L-methionine (SAM) to carbon 5 of cytosine, principally at CpG sites, producing 5-methylcytosine and S-adenosyl-L-homocysteine. The catalytic subunit is DNMT3A/DNMT3A2 or DNMT3B—not DNMT3L. DNMT3L’s primary function is therefore best defined as a chromatin-reading, assembly, and allosteric cofactor for de novo DNA methylation (jia2007structureofdnmt3a pages 1-2, noh2016readingbetweenthe pages 3-4).
Biochemical interaction and interface-mutagenesis experiments show that DNMT3L directly stimulates DNMT3A. Disruption of the DNMT3A–DNMT3L interface impairs oligomerization and stimulation, consistent with DNMT3L stabilizing a catalytically competent DNMT3A conformation. Interaction and stimulation of DNMT3B have also been reported, although physiological dependence and target scope are less completely defined than for DNMT3A/DNMT3A2 (jia2007structureofdnmt3a pages 1-2, noh2016readingbetweenthe pages 3-4, jia2007structureofdnmt3a pages 5-10).
DNMT3L’s ADD domain binds the extreme N-terminal tail of histone H3 and preferentially recognizes unmethylated lysine 4 (H3K4me0). Mono-, di-, or trimethylation of H3K4 disrupts binding, whereas H3K9 methylation is tolerated. DNMT3L can thus help couple de novo DNA methylation to chromatin lacking an active-promoter-associated H3K4 methyl mark (noh2016readingbetweenthe pages 3-4, jia2007structureofdnmt3a pages 4-5, noh2016readingbetweenthe pages 15-18).
This is chromatin-state recognition, not strict DNA-sequence specificity. Actual genomic targeting also depends on accessibility, transcription through target regions, nucleosome modifications, DNA sequence/context, and interactions with other germline factors. H3K4 methylation can protect regions from DNMT3L-associated methylation, while H3K4me0 creates a permissive—not necessarily sufficient—state (noh2016readingbetweenthe pages 3-4, jia2007structureofdnmt3a pages 4-5).
The landmark structure by Jia and colleagues was published in September 2007 in Nature (DOI/URL). The 2.9-Å structure of the minimal catalytic-region complex revealed a linear DNMT3L–DNMT3A–DNMT3A–DNMT3L tetramer, containing two DNMT3A catalytic sites. The assembly measured approximately 160 × 60 × 50 Å, and both the DNMT3A–DNMT3L and DNMT3A–DNMT3A interfaces were important for catalysis (jia2007structureofdnmt3a pages 1-2, jia2007structureofdnmt3a pages 5-10).
The paired DNMT3A active sites are about 40 Å apart. DNMT3A contacts approximately 9–12 bp per monomer, and methylation correlations occur at roughly 8–10-bp intervals; maternally imprinted differentially methylated regions were reported to have an average CpG periodicity near 9.5 bp. These observations support a model in which oligomer geometry contributes to periodic CpG methylation, although full-length chromatin complexes are likely more dynamic than the crystallized minimal assembly (jia2007structureofdnmt3a pages 5-10, jia2007structureofdnmt3a pages 2-4).
DNMT3L performs its established function inside the nucleus at chromatin. This assignment follows directly from its binding to nucleosomal histone H3 tails and assembly with DNMT3 enzymes that methylate genomic DNA. Its functional location is therefore chromatin associated rather than cytosolic, membrane bound, secreted, or extracellular (noh2016readingbetweenthe pages 3-4, jia2007structureofdnmt3a pages 5-10, noh2016readingbetweenthe pages 15-18).
The retrieved evidence did not provide a definitive, high-resolution map of endogenous human DNMT3L subnuclear compartments. Claims of localization to particular nuclear bodies should therefore not be inferred from the general chromatin mechanism alone.
Mammalian primordial germ cells first erase much of the inherited DNA-methylation landscape; male and female germ cells subsequently establish sex-specific methylation patterns. DNMT3L functions during the rebuilding phase by enabling de novo DNMT activity on appropriate chromatin. In growing mouse oocytes, it is required for normal acquisition of maternal methylation imprints. Dnmt3l-deficient females can complete meiosis but produce nonviable offspring because oocyte-derived methylation and imprinting are abnormal (salle2006originofdna pages 69-75, salle2006originofdna pages 87-91).
In the male mouse germ line, Dnmt3l loss compromises methylation at paternal imprinting regions. At H19, one assayed site lost approximately 85% of methylation and other sites about 40%; methylation at Dlk1–Gtl2 fell below 10% at tested sites. The maternally marked U2af1-rs1 region remained largely unmethylated, arguing for failed acquisition of appropriate paternal marks rather than indiscriminate ectopic methylation (salle2006originofdna pages 168-172).
DNMT3L-associated methylation represses dispersed repeats and retrotransposons in developing male germ cells. Dnmt3l-deficient mouse germ cells show hypomethylation of retroviral/repetitive sequences, including contexts involving LINE-1 and IAP elements. Failure of this pathway is associated with retrotransposon reactivation, chromosomal abnormalities, and germ-cell loss (noh2016readingbetweenthe pages 3-4, jia2007structureofdnmt3a pages 4-5, salle2006originofdna pages 145-153).
DNMT3L is not solely a repeat-targeting factor. Purified Dnmt3l-null primitive spermatogonia also showed severe hypomethylation at approximately 30 surveyed loci on chromosomes 4 and X, including unique nonrepetitive, non-CpG-island sequences. This supports a broad role in germline methylome establishment whose target selection is chromatin dependent (salle2006originofdna pages 145-153).
Mouse Dnmt3l expression peaks in prenatal gonocytes during the period of methylation acquisition and overlaps with Dnmt3a2. One developmental analysis reported an approximately 800-fold increase at embryonic day 15.5 relative to E13.5, with high expression persisting at E18.5; expression then declines during spermatogonial differentiation and pachynema (salle2006originofdna pages 161-164).
Dnmt3l-null male mice are infertile and develop germ-cell depletion, delayed meiotic entry, abnormal homologous-chromosome synapsis, and subsequent apoptosis. At postnatal day 10, γH2AX-positive tubules—used as an early meiotic-entry measure—numbered 28.2 ± 2.6 per 100 tubules in mutants versus 74.9 ± 4.0 in controls. Germ-cell numbers were already reduced by more than 50% at day 6 in one analysis, and proliferation at three weeks was approximately half of wild type (salle2006originofdna pages 172-176, salle2006originofdna pages 164-168, salle2006originofdna pages 161-164).
These phenotypes likely combine direct consequences of an incorrectly methylated germline genome with secondary effects of transposon activation and abnormal chromosome behavior. They should not be interpreted as evidence that DNMT3L is itself a structural synaptonemal-complex protein.
The conserved human protein architecture and in-vitro molecular mechanism are well supported. By contrast, most causal evidence connecting DNMT3L to imprint establishment, transposon repression, meiosis, and fertility derives from mouse knockout experiments. Human gametogenesis differs from rodent gametogenesis in developmental timing and spermatogonial-state organization; therefore, mouse phenotypes cannot automatically be presented as demonstrated human phenotypes.
The most defensible human annotation is that DNMT3L is a developmentally regulated epigenetic cofactor expected to participate in germline and pluripotent-cell de novo methylation. It is biologically plausible as a candidate modifier of infertility or imprinting defects, but the retrieved literature did not establish DNMT3L deficiency as a routine monogenic diagnosis in humans.
Current real-world use is mainly experimental:
Open Targets lists literature-derived DNMT3L associations with cardiovascular disorders, ventricular fibrillation, hepatocellular carcinoma, rheumatoid arthritis, and genetic primary adrenal insufficiency, but the reported aggregate scores are low to moderate—approximately 0.084–0.347—and do not establish causality, mechanism, or therapeutic tractability (OpenTargets Search: -DNMT3L).
No DNMT3L-directed approved therapy, validated companion diagnostic, or established clinical biomarker was identified. Broad nucleoside DNMT inhibitors act chiefly through catalytic DNMTs and are not DNMT3L-selective. Targeting DNMT3L could also risk disrupting germline imprinting and genome defense, making tissue specificity and reproductive safety major barriers.
A key conclusion of the recent-literature assessment is that the strongest DNMT3L-specific mechanistic evidence remains the foundational biochemical, structural, and mouse-genetic literature. Work in 2023–2024 has mainly refined the broader context of germline reprogramming, locus-specific imprint establishment, metabolic control, environmental sperm epigenomics, and species differences rather than overturning the core DNMT3L model.
Recent studies continue to place DNMT3L in the DNMT3A-dependent machinery relevant to paternal H19/Igf2 imprint establishment and germ-cell methylome rebuilding. However, newer locus studies emphasize that transcription through an imprinting-control region and local chromatin state help determine where the DNMT3 machinery acts; DNMT3L should therefore not be viewed as a stand-alone locus-address code. Likewise, recent human single-cell testis atlases add context about human spermatogonial heterogeneity but do not yet supply the equivalent of a causal human DNMT3L knockout experiment (salle2006originofdna pages 69-75, salle2006originofdna pages 168-172, salle2006originofdna pages 87-91).
An authoritative mechanistic review by Noh, Allis, and Li was published in ACS Chemical Biology in December 2016 (DOI/URL). It frames ADD-domain recognition as a molecular bridge between histone and DNA methylation and remains consistent with current understanding: H3K4me0 is permissive for DNMT3L engagement, while methylated H3K4 antagonizes it (noh2016readingbetweenthe pages 3-4, noh2016readingbetweenthe pages 15-18).
The most precise functional annotation is:
DNMT3L is a catalytically inactive, nuclear chromatin reader and regulatory subunit of de novo DNA-methylation complexes. Its ADD domain recognizes histone H3 tails lacking H3K4 methylation, while its C-terminal methyltransferase-like domain binds and stimulates DNMT3A/DNMT3A2—and contextually DNMT3B—to establish germline DNA methylation at imprinting-control regions, retrotransposons, and other chromatin targets.
This wording avoids three common errors: calling DNMT3L an active methyltransferase; assigning it an independent CpG substrate specificity; or presenting mouse infertility phenotypes as already proven human disease. Confidence is high for catalytic inactivity, H3K4me0 recognition, DNMT3A interaction, tetramer architecture, and mouse germline function; moderate for the extent of DNMT3B cooperation and direct conservation of every mouse target in humans; and low for current disease-target or therapeutic claims.
References
(jia2007structureofdnmt3a pages 1-2): Da Jia, Renata Z. Jurkowska, Xing Zhang, Albert Jeltsch, and Xiaodong Cheng. Structure of dnmt3a bound to dnmt3l suggests a model for de novo dna methylation. Nature, 449:248-251, Sep 2007. URL: https://doi.org/10.1038/nature06146, doi:10.1038/nature06146. This article has 1102 citations and is from a highest quality peer-reviewed journal.
(noh2016readingbetweenthe pages 3-4): Kyung-Min Noh, C. David Allis, and Haitao Li. Reading between the lines: “add”-ing histone and dna methylation marks toward a new epigenetic “sum”. ACS Chemical Biology, 11(3):554-563, Dec 2016. URL: https://doi.org/10.1021/acschembio.5b00830, doi:10.1021/acschembio.5b00830. This article has 29 citations and is from a domain leading peer-reviewed journal.
(salle2006originofdna pages 35-39): S La Salle. Origin of dna methylation patterns in the male germ line: roles of the novel dna methyltransferases in male germ cells. Unknown journal, 2006.
(noh2016readingbetweenthe pages 15-18): Kyung-Min Noh, C. David Allis, and Haitao Li. Reading between the lines: “add”-ing histone and dna methylation marks toward a new epigenetic “sum”. ACS Chemical Biology, 11(3):554-563, Dec 2016. URL: https://doi.org/10.1021/acschembio.5b00830, doi:10.1021/acschembio.5b00830. This article has 29 citations and is from a domain leading peer-reviewed journal.
(jia2007structureofdnmt3a pages 4-5): Da Jia, Renata Z. Jurkowska, Xing Zhang, Albert Jeltsch, and Xiaodong Cheng. Structure of dnmt3a bound to dnmt3l suggests a model for de novo dna methylation. Nature, 449:248-251, Sep 2007. URL: https://doi.org/10.1038/nature06146, doi:10.1038/nature06146. This article has 1102 citations and is from a highest quality peer-reviewed journal.
(jia2007structureofdnmt3a pages 5-10): Da Jia, Renata Z. Jurkowska, Xing Zhang, Albert Jeltsch, and Xiaodong Cheng. Structure of dnmt3a bound to dnmt3l suggests a model for de novo dna methylation. Nature, 449:248-251, Sep 2007. URL: https://doi.org/10.1038/nature06146, doi:10.1038/nature06146. This article has 1102 citations and is from a highest quality peer-reviewed journal.
(salle2006originofdna pages 69-75): S La Salle. Origin of dna methylation patterns in the male germ line: roles of the novel dna methyltransferases in male germ cells. Unknown journal, 2006.
(salle2006originofdna pages 168-172): S La Salle. Origin of dna methylation patterns in the male germ line: roles of the novel dna methyltransferases in male germ cells. Unknown journal, 2006.
(salle2006originofdna pages 145-153): S La Salle. Origin of dna methylation patterns in the male germ line: roles of the novel dna methyltransferases in male germ cells. Unknown journal, 2006.
(salle2006originofdna pages 164-168): S La Salle. Origin of dna methylation patterns in the male germ line: roles of the novel dna methyltransferases in male germ cells. Unknown journal, 2006.
(salle2006originofdna pages 161-164): S La Salle. Origin of dna methylation patterns in the male germ line: roles of the novel dna methyltransferases in male germ cells. Unknown journal, 2006.
(salle2006originofdna pages 87-91): S La Salle. Origin of dna methylation patterns in the male germ line: roles of the novel dna methyltransferases in male germ cells. Unknown journal, 2006.
(OpenTargets Search: -DNMT3L): Open Targets Query (-DNMT3L, 5 results). Buniello, A. et al. (2025). Open Targets Platform: facilitating therapeutic hypotheses building in drug discovery. Nucleic Acids Research.
(jia2007structureofdnmt3a pages 2-4): Da Jia, Renata Z. Jurkowska, Xing Zhang, Albert Jeltsch, and Xiaodong Cheng. Structure of dnmt3a bound to dnmt3l suggests a model for de novo dna methylation. Nature, 449:248-251, Sep 2007. URL: https://doi.org/10.1038/nature06146, doi:10.1038/nature06146. This article has 1102 citations and is from a highest quality peer-reviewed journal.
(salle2006originofdna pages 172-176): S La Salle. Origin of dna methylation patterns in the male germ line: roles of the novel dna methyltransferases in male germ cells. Unknown journal, 2006.