Dystrophin (DMD) Frameshift Reading-Frame Rule and Exon-Skipping Therapy

Dystrophin (DMD), Monaco's Reading-Frame Rule, and Exon-Skipping Therapy represent a foundational paradigm in human molecular genetics and RNA-targeted drug discovery. Spanning 2.22 megabases (Mb) (~0.07% of the entire human genome) across 79 exons on the short arm of the X chromosome (ChrXp21.2–p21.1), the DMD gene is the largest known human gene: its primary pre-mRNA transcript is so vast that RNA Polymerase II (elongating at ~2.3 kb/min) requires roughly 16 hours to synthesize a single full-length molecule, while 99.4% of the locus consists of gigantic introns (such as Intron 44 at ~248 kb and Intron 1 at ~107 kb) spliced into a 14.0-kb mature muscle mRNA (Dp427m). Because of its immense physical footprint and intronic recombination hotspots, DMD exhibits one of the highest spontaneous mutation rates in humans (μ ≈ 1 × 10⁻⁴ per gene per gamete; ~33% of patients carry de novo mutations).

Yet mutations within this single X-linked gene produce two clinically divergent myopathies in hemizygous males: Duchenne muscular dystrophy (DMD, incidence ~1 in 3,500 to 1 in 5,000 live male births), a progressive necrotizing myopathy causing loss of ambulation by age 10–13 and cardiorespiratory failure in early adulthood; and Becker muscular dystrophy (BMD, incidence ~1 in 18,500 males), a milder allelic disorder in which patients often remain ambulatory into their fourth to seventh decades. In 1988, Anthony Monaco, Louis Kunkel, and colleagues resolved how multi-exon deletions of the DMD gene can be milder than single-exon deletions via Monaco's Reading-Frame Rule: multi-exon deletions that shift the open reading frame (out-of-frame) generate premature stop codons (PTCs), triggering nonsense-mediated mRNA decay (NMD) and complete absence of dystrophin (DMD), whereas even massive deletions (up to 46% of the coding sequence) that preserve the translational reading frame (in-frame) yield an internally shortened, partially functional dystrophin shock-absorber (BMD)—directly inspiring phosphorodiamidate morpholino oligomer (PMO) exon-skipping therapies (eteplirsen, golodirsen, viltolarsen, casimersen) and AAV micro-dystrophin gene therapy (delandistrogene moxeparvovec).

Four Structural Domains Bridging F-Actin to the Dystrophin-Associated Protein Complex

In striated skeletal and cardiac myocytes, the full-length muscle isoform of dystrophin (Dp427m, 3,685 amino acids, 427 kDa) localizes to the cytoplasmic face of the sarcolemma at specialized rib-like costameres, where it functions as a molecular spring connecting the intracellular contractile γ-actin cytoskeleton to the transmembrane dystrophin-associated protein complex (DAPC / dystroglycan–sarcoglycan complex) and the extracellular basal lamina (laminin-211 / merosin):

1. N-Terminal Actin-Binding Domain 1 (ABD1, Exons 1–8, residues 1–246): Composed of two tandem calponin-homology (CH1 and CH2) domains that bind directly to filamentous cortical F-actin around the Z-disc and M-band costameres.
2. Central Spectrin-Like Rod Domain (Exons 8–61, residues 247–3,040): Accounts for >75% of the polypeptide length, consisting of 24 triple-helical coiled-coil spectrin-like repeats (R1–R24, each ~109 amino acids folding into an antiparallel α-helical bundle) interspersed with four flexible proline-rich hinge regions (H1 before R1, H2 between R3–R4, H3 between R19–R20, and H4 after R24). Furthermore, repeats R11–R17 provide a secondary electrostatic actin-binding domain (ABD2), while R16–R17 form the obligatory docking site that anchors neuronal nitric oxide synthase (nNOSμ) to the sarcolemma to generate vasodilatory nitric oxide (NO) during muscle contraction (preventing functional exercise-induced ischemia).
3. Cysteine-Rich Domain (CR, Exons 62–69, residues 3,041–3,360): Contains a WW domain (embedded in H4), two EF-hand motifs, and a ZZ zinc-finger domain that together form the mandatory binding interface for the PPxY motif of transmembrane β-dystroglycan (DAG1) (linked extracellularly to α-dystroglycan and laminin-211, stabilized upstream at neuromuscular junctions by agrin).
4. C-Terminal Coiled-Coil Domain (CT, Exons 70–79, residues 3,361–3,685): Binds cytosolic α/β-dystrobrevin (DTNA) and α1/β1-syntrophins (SNTA1).

When dystrophin is absent (<1%–3% of normal levels in DMD), eccentric mechanical contraction tears the unbuttressed lipid bilayer of the sarcolemma, allowing unregulated influx of extracellular Ca²⁺ (via TRPC/stretch-activated channels), activating calpain proteases, mitochondrial permeability transition pores, and chronic necroinflammation that progressively replaces myofibers with fibroadipose tissue (creatine kinase CK elevated 20- to 100-fold, often 10,000–30,000 U/L).

Why Deletion Size Does Not Predict Clinical Severity

Across DMD and BMD patient registries, ~65% to 70% of all mutations are large multi-exon deletions (with ~10% multi-exon duplications and ~20% point nonsense/splice mutations) that cluster predominantly inside two intronic recombination hotspots: a minor proximal hotspot (Exons 2–20) and the major distal hotspot (Exons 45–55, centered on the 248-kb Intron 44), which accounts for nearly two-thirds of all deletions.

Monaco's Reading-Frame Rule (which holds true in >91% to 93% of all DMD/BMD deletions) explains why a patient lacking a single 176-bp exon (ΔExon 45) suffers severe DMD, whereas a patient with a deletion removing 11 consecutive exons (ΔExon 45–55, spanning over 500 kb of genomic DNA) has minimal or asymptomatic BMD:

1. Exon Border Phasing (Phase 0, Phase 1, Phase 2): Because each amino acid is encoded by a 3-nucleotide codon, an intron can interrupt the coding sequence either exactly between two codons (Phase 0), after the first nucleotide of a codon (Phase 1), or after the second nucleotide of a codon (Phase 2).
2. Frameshift (Out-of-Frame) Deletions (DMD): Exon 44 ends in Phase 2 (2 nucleotides into a codon), while Exon 45 (176 bp) ends in Phase 1 ((2 + 176) = 178; 178 mod 3 = 1) and Exon 46 starts after a Phase 1 intron. When Exon 45 alone is deleted, the spliceosome joins the Phase 2 3′ end of Exon 44 directly to the Phase 1 5′ start of Exon 46. This +1 / −2 nucleotide mismatch shifts every downstream codon in Exon 46, hitting a premature termination codon (TGA/TAA/TAG) within 16 codons and destroying the transcript via NMD.
3. In-Frame Deletions (BMD): Conversely, within the central spectrin-like rod domain (Exons 10–60), many exons encode modular triple-helical repeats whose boundaries share identical intron phases. For example, Exon 44 ends in Phase 2, and Exon 47 (Δ45–47), Exon 48 (Δ45–48), and Exon 55 (Δ45–55) also end in Phase 2. Therefore, deleting Exons 45–47, Exons 45–48, or Exons 45–55 joins a Phase 2 donor to a Phase 2 acceptor (total deleted nucleotides divisible by 3), preserving the downstream reading frame through Exons 62–79 so the essential Cysteine-Rich (CR) β-dystroglycan-binding domain is translated intact.

Phosphorodiamidate Morpholino Oligomers (PMOs) Converting DMD into BMD

The realization that ΔExon 45–55 and ΔExon 48–51 in-frame deletions produce functional Becker-like dystrophin inspired therapeutic exon skipping: using synthetic steric-blocking antisense oligonucleotides to mask an out-of-frame flanking exon during pre-mRNA alternative splicing, deliberately enlarging a frameshift deletion by one additional exon to restore the open reading frame:

• Mechanism of PMO Action: Unlike RNase H-recruiting gapmers, phosphorodiamidate morpholino oligomers (PMOs—eteplirsen, golodirsen, viltolarsen, casimersen) replace the ribose sugar with a 6-membered morpholino ring and the negatively charged phosphodiester linkage with an uncharged phosphorodiamidate group. Resistant to all cellular nucleases and devoid of innate Toll-like receptor (TLR) immune activation, a PMO hybridizes via Watson–Crick base pairing to an exonic splicing enhancer (ESE) inside the target exon (such as Exon 51, 233 bp). By sterically blocking SR (Ser-Arg) splicing factors (SRSF1/SRSF2) and U1/U2 snRNP recognition, the PMO causes the spliceosome to skip Exon 51. In a DMD boy with an out-of-frame ΔExon 48–50 deletion (Exon 47 ends in Phase 0; Exon 51 starts in Phase 2), skipping Exon 51 (233 bp) joins the Phase 0 end of Exon 47 directly to the Phase 0 start of Exon 52, generating an in-frame ΔExon 48–51 Becker-like transcript. Together, skipping Exon 51 (~13%), Exon 53 (~8%), and Exon 45 (~8%) is applicable to ~30% of all DMD patients.

Overcoming the 4.7-kb AAV Packaging Limit with Engineered Micro-Dystrophin

Because the full-length 14.0-kb DMD mRNA (and 11.0-kb coding sequence) is three times larger than the ~4.7-kb single-stranded DNA packaging capacity of adeno-associated virus (AAV), full-length DMD gene replacement is impossible in a single AAV capsid. However, studying a remarkably mild BMD patient who walked with a cane at age 61 despite an in-frame deletion of 46% of his DMD coding sequence (ΔExon 17–48) enabled Chamberlain, Mendell, and colleagues to engineer micro-dystrophin (4.38 kb including the muscle/cardiac-specific MHCK7 promoter/enhancer) delivered by the myotropic AAVrh74 capsid (delandistrogene moxeparvovec / Elevidys). By retaining only the essential N-terminal ABD1, hinges H1, H3, and H4, four spectrin repeats (R1–R3 and R24), and the C-terminal Cysteine-Rich (CR) β-dystroglycan-binding domain, Elevidys packages inside a single AAVrh74 virion and restores a 138-kDa sarcolemmal shock-absorber across skeletal and cardiac muscle regardless of the patient's underlying DMD mutation (excluding patients with Exon 8–9 deletions that trigger anti-transgene ABD1 T-cell immunity).

How can a patient missing 11 exons of the DMD gene (ΔExon 45–55) have mild Becker muscular dystrophy while a patient missing only 1 exon (ΔExon 45) has severe Duchenne muscular dystrophy?

According to Monaco's Reading-Frame Rule, the critical determinant of severity is not the number of deleted central rod repeats, but whether the deletion maintains the 3-nucleotide codon reading frame so the C-terminal β-dystroglycan-binding (Cysteine-Rich) domain (Exons 62–69) can still be translated. Exon 45 (176 bp) is not a multiple of 3 (176 mod 3 = 2), so deleting Exon 45 alone shifts the reading frame and causes a premature stop codon in Exon 46 (DMD). Conversely, deleting Exons 45–55 together removes an exact multiple of 3 nucleotides (Phase 2 to Phase 2), producing an internally shortened but doubly anchored dystrophin spring (BMD).

How does eteplirsen (Exon 51 skipping) restore dystrophin expression in a patient with a ΔExon 48–50 deletion?

In a patient with an out-of-frame ΔExon 48–50 deletion, Exon 47 ends cleanly between codons (Phase 0), but Exon 51 (233 bp) begins mid-codon (Phase 2), creating a frameshift and premature termination codon inside Exon 51. Eteplirsen is a 30-nucleotide phosphorodiamidate morpholino oligomer (PMO) that binds pre-mRNA inside Exon 51, hiding it from the spliceosome so Exon 47 splices directly to Exon 52 (which starts in Phase 0). This converts an out-of-frame Δ48–50 DMD transcript into an in-frame Δ48–51 Becker-like transcript.

Why does AAV gene therapy for Duchenne muscular dystrophy deliver a 138-kDa "micro-dystrophin" instead of the full 427-kDa dystrophin protein?

The full-length dystrophin coding sequence is 11.0 kb (14.0 kb mRNA), which far exceeds the strict ~4.7-kb physical genome packaging limit of an adeno-associated virus (AAV) capsid. Guided by mild Becker muscular dystrophy patients who naturally lack Exons 17–48, scientists removed 20 of the 24 redundant central spectrin-like repeats (R4–R23) and the distal CT tail while preserving the N-terminal actin-binding domain (ABD1), 4 spectrin repeats (R1–R3, R24) with 3 hinges (H1, H3, H4), and the Cysteine-Rich (CR) β-dystroglycan anchor, yielding a 138-kDa micro-dystrophin encoded by a compact ~3.6-kb cDNA.

1. Monaco, A. P., Bertelson, C. J., Liechti-Gallati, S., Moser, H., & Kunkel, L. M. (1988). "An explanation for the phenotypic differences between patients bearing partial deletions of the DMD locus." Genomics, 2(1), 90–95. https://doi.org/10.1016/0888-7543(88)90113-990113-9)
2. England, S. B., Nicholson, L. V., Johnson, M. A., et al. (1990). "Very mild muscular dystrophy associated with the deletion of 46% of dystrophin." Nature, 343(6254), 180–182. https://doi.org/10.1038/343180a0
3. Mendell, J. R., Rodino-Klapac, L. R., Sahenk, Z., et al. (2013). "Eteplirsen for the treatment of Duchenne muscular dystrophy." Annals of Neurology, 74(5), 637–647. https://doi.org/10.1002/ana.23982
4. Mendell, J. R., Sahenk, Z., Lehman, K., et al. (2020). "Assessment of systemic delivery of rAAVrh74.MHCK7.micro-dystrophin in children with Duchenne muscular dystrophy: a nonrandomized controlled trial." JAMA Neurology, 77(9), 1122–1131. https://doi.org/10.1001/jamaneurol.2020.1484