BGPT: Paper Review: DLX5 and DLX6 Expression Is Biallelic and Not Modulated by MeCP2 Deficiency

Explore by Goal

Quick Answer Copied

Verdict (skeptical, evidence-based)

This paper directly tests (and does not support) the claim that DLX5/DLX6 are imprinted and that MeCP2 deficiency “relaxes imprinting” to increase their expression in Rett syndrome models. Using allele-specific expression assays in human somatic cell hybrids, human lymphoblastoid cell lines, and human brain (plus mouse qRT-PCR in Mecp2-mutant forebrain), the authors find biallelic expression for DLX5 and DLX6 and no consistent MeCP2-dependent change in Dlx5/Dlx6 expression, while imprinting for PEG3 and PEG10 is preserved.

Long Answer

Paper Review: DLX5 and DLX6 Expression Is Biallelic and Not Modulated by MeCP2 Deficiency

Citation: Schu et al., 2007 (Am J Hum Genet). DOI: 10.1086/520063

Claim being tested: whether DLX5/DLX6 are imprinted in humans and whether MeCP2 deficiency causes “loss/relaxation of imprinting” leading to increased DLX5/DLX6 expression in Rett syndrome models.

1) Evidence map (what the authors measured)

Human somatic cell hybrids (SCH): allele-specific RT-PCR distinguishing maternal vs paternal retained human chr7 copies for DLX5 and DLX6. The authors report biallelic expression in SCHs retaining chr7 of either parental origin.
Human lymphoblastoid cell lines (LCLs) + human brain: allele-specific RT-PCR using DLX5 and newly discovered DLX6 SNPs. The authors report biallelic expression for DLX5 and (for the available heterozygotes) DLX6, with no affected-vs-control difference in their comparisons.
Positive control for “loss of imprinting”: PEG3 and PEG10 (imprinted genes) in MeCP2-mutant human tissues. The authors report that PEG3 and PEG10 remain strictly monoallelic in MeCP2-deficient tissues.
Mouse forebrain (expression-level test): qRT-PCR in male Mecp2-mutant forebrain (7–9 wk) vs wild-type littermates. The authors report that Dlx5 and Dlx6 expression is highly variable and lacks consistent upregulation in Mecp2-null mutants.

2) Visual summary: allelic expression calls (from their Table 3)

Interpretation: DLX5 and DLX6 are called biallelic in all contexts where the authors could make allele-specific calls; PEG3/PEG10 are called monoallelic, functioning as a positive control that MeCP2 deficiency does not necessarily “break” imprinting maintenance in the same samples.

3) Methodological skepticism checklist (where mistakes can happen)

Primer specificity & allele discrimination: The paper emphasizes primer design to avoid mouse ortholog amplification and to enable human allele discrimination by SNPs. This is central because allele-specific RT-PCR can fail silently when mismatches are insufficient or when amplification efficiency differs by allele.
Low expression / PCR-cycle artifacts: The paper reports that initial increases in PCR cycle number (for LCLs) led to nonreproducible apparent monoallelic calls, likely reflecting low transcript abundance and technical artifacts. They report that handling workflow (avoiding freeze-thaw) was needed to obtain reproducible biallelic calls.
Restriction-based “imbalance” interpretation: The authors argue that an earlier claim of maternal-preferential expression in mice could be confounded by restriction-enzyme fragment visualization/heteroduplex formation and unequal handling of fragment sizes. (This is a critique of interpretation logic rather than of direct measurements in this paper.)
Biological variability vs systematic effect: For the mouse expression test, they emphasize substantial biological variability across individual animals/litters and report that combining data did not yield significant consistent differences for Dlx5/Dlx6.

4) Visual summary: “MeCP2 deficiency effects” as reported

Since the full numeric fold-changes per mouse are not fully reproduced in the provided text dump, the most evidence-faithful visualization is the study’s presence/absence of a consistent effect (as the authors state). The plot below encodes their stated qualitative outcome: no consistent upregulation in mutants for Dlx5/Dlx6, contrasted with preserved monoallelic imprinting for PEG3/PEG10.

5) Critical appraisal (what’s strong vs what remains uncertain)

Strengths

Replication-focused design: the paper is explicitly positioned as an attempt to reproduce a prior key claim linking MeCP2 to DLX5/DLX6 imprinting and overexpression.
Orthogonal assay layers: allele-specific transcript calls (DLX5/DLX6) plus separate expression-level qRT-PCR in mouse forebrain, plus imprinting controls (PEG3/PEG10).
Mechanistic coherence check: the authors argue that if MeCP2 deficiency caused “relaxed imprinting” at DLX5/DLX6, imprinted maintenance should be broken for those loci while imprinted control genes might not be affected; instead they see biallelic DLX5/DLX6 and preserved PEG imprinting.

Limitations / blind spots (science-only)

Allele-specific calling depends on expression detectability, especially in LCLs where the authors note low copy number and PCR-workflow sensitivity. While they address this technically, residual risk remains that allele dropout could occur in other laboratories or at other developmental stages.
Power for “consistent upregulation” in mouse forebrain: the authors stress biological noise and report no consistent difference across their combined dataset, but absence of evidence is still bounded by sample size and effect size.
DLX6 imprinting assessment is constrained by available heterozygotes: the authors report biallelic DLX6 expression using polymorphisms that enable discrimination, but the informative allele-specific sample count is limited in their presented contexts.
Cell model generality: SCHs and LCLs are not brain tissue; however, the paper includes human brain cortex, which strengthens translational relevance.

What would most strongly disprove these conclusions?

Demonstrating monoallelic DLX5 or DLX6 expression in human brain with robust allele discrimination, across independent cohorts and without allele dropout artifacts.
Showing a reproducible, locus-specific MeCP2-dependent change in DLX5/DLX6 expression (or allele-specific loss of imprinting) that cannot be explained by workflow, primer, or allele quantification artifacts, while imprinting controls remain intact.

6) Author review links (bespoke)

Feedback:

Updated: April 03, 2026