Cryptic exon splicing is a pathological RNA processing event in which normally repressed "cryptic" exonic sequences are aberrantly included in mature mRNA transcripts. In the context of neurodegeneration, cryptic exon inclusion has emerged as a central downstream consequence of TDP-43 nuclear depletion — the defining pathological feature of ALS and the majority of FTD cases. When TDP-43 is lost from the nucleus, it can no longer suppress the inclusion of cryptic exons in its target pre-mRNAs, leading to aberrant protein products, nonsense-mediated mRNA decay, and loss of essential proteins.
This mechanism has transformed our understanding of how TDP-43 proteinopathy causes neurodegeneration. Rather than a simple toxic gain-of-function from cytoplasmic TDP-43 aggregates, the loss of TDP-43's nuclear splicing function — and the resulting cryptic exon inclusion — appears to be a primary driver of neuronal dysfunction and death.
Cryptic exons are intronic sequences that contain potential splice sites but are not normally recognized by the spliceosome. They are "cryptic" because they are hidden from the splicing machinery under normal conditions. Several features distinguish cryptic exons:
- Weak splice sites: Cryptic exons have suboptimal splice donor and acceptor sequences that are normally insufficient for recognition
- Repressor-dependent silencing: They are kept silent by RNA-binding proteins (especially TDP-43) that bind nearby to block splice site recognition
- Evolutionary non-conservation: Most TDP-43-dependent cryptic exons are not conserved across species, suggesting they arose from random intronic mutations
- Pathological consequences: When included, cryptic exons typically introduce premature stop codons, causing nonsense-mediated decay (NMD) or truncated, non-functional proteins
TDP-43 (TAR DNA-binding protein 43) is a nuclear RNA-binding protein that regulates RNA splicing, stability, and transport. Its role in cryptic exon repression involves:
- UG-repeat binding: TDP-43 recognizes UG-rich sequences in introns flanking cryptic exons through its two RNA recognition motifs (RRM1 and RRM2)
- Spliceosome blocking: By binding to these UG-rich regions, TDP-43 sterically blocks the spliceosome from recognizing the cryptic splice sites
- Co-factor recruitment: TDP-43 recruits other splicing factors (hnRNPs, SR proteins) to reinforce cryptic exon skipping
- Genome-wide repression: TDP-43 represses hundreds of cryptic exons across the transcriptome
graph TD
A["Normal: TDP-43 in Nucleus"] --> B["TDP-43 Binds UG-rich<br/>Intronic Sequences"]
B --> C["Cryptic Exons REPRESSED"]
C --> D["Normal mRNA → Normal Protein"]
E["Disease: TDP-43 Mislocalizes<br/>to Cytoplasm"] --> F["Nuclear TDP-43 Depleted"]
F --> G["Cryptic Exons INCLUDED"]
G --> H["Premature Stop Codon"]
G --> I["Frameshift"]
G --> J["Novel Peptide Sequence"]
H --> K["Nonsense-Mediated Decay<br/>(mRNA Destroyed)"]
I --> K
J --> L["Truncated/Aberrant Protein"]
K --> M["Loss of Essential Protein"]
style A fill:#4CAF50,color:#fff
style E fill:#f44336,color:#fff
style M fill:#9C27B0,color:#fff
style L fill:#9C27B0,color:#fff
When TDP-43 mislocalizes to the cytoplasm (as occurs in ALS/FTD):
- Nuclear TDP-43 levels fall below the threshold needed for cryptic exon repression
- Cryptic exons are included in hundreds of target transcripts
- Most inclusions introduce premature stop codons → nonsense-mediated decay → protein loss
- Some inclusions produce aberrant proteins with novel, potentially toxic sequences
The most clinically significant TDP-43-dependent cryptic exon target is STMN2.
- Normal function: STMN2 (also called SCG10) is essential for axonal regeneration and maintenance in motor neurons
- Cryptic exon: TDP-43 loss leads to inclusion of a cryptic exon in STMN2 intron 1, introducing a premature polyadenylation signal
- Consequence: Truncated STMN2 mRNA that produces no functional protein
- Disease relevance: STMN2 protein is dramatically reduced in ALS/FTD patient motor neurons
- Therapeutic target: Restoring STMN2 expression (via ASOs that block the cryptic exon) is being pursued as a therapy for ALS
- Biomarker potential: Truncated STMN2 transcripts detectable in patient-derived neurons
A critical genetic modifier of ALS/FTD disease course.
- Normal function: UNC13A is essential for synaptic vesicle release and neurotransmission
- Cryptic exon: TDP-43 loss causes inclusion of a cryptic exon in UNC13A intron 20-21
- Genetic link: Common SNPs (rs12608932, rs12973192) in the UNC13A cryptic exon region are among the strongest ALS risk modifiers in GWAS
- Mechanism: These SNPs create stronger cryptic splice sites, making UNC13A more sensitive to TDP-43 loss
- Consequence: Loss of UNC13A protein impairs synaptic function
- Clinical impact: UNC13A risk variants associate with shorter survival in ALS patients
- Rho-GEF protein critical for dendritic spine morphology and synaptic plasticity
- TDP-43-dependent cryptic exon in KALRN leads to loss of functional protein
- May contribute to cognitive/synaptic dysfunction in FTD
- Essential for neuromuscular junction formation and maintenance
- Cryptic exon inclusion disrupts agrin protein production
- May contribute to NMJ dysfunction in ALS
- Splicing factor involved in pre-mRNA processing
- Loss amplifies splicing defects in a feed-forward manner
- Creates a cascade of splicing disruption
A remarkable feature of TDP-43-dependent cryptic exons is their lack of evolutionary conservation.
- Mouse and human TDP-43 perform the same function (cryptic exon repression)
- But the specific cryptic exons are different between species
- The STMN2 cryptic exon exists in humans but NOT in mice
- This means mouse models of TDP-43 pathology do not recapitulate the loss of STMN2
- Humanized mouse models (carrying the human STMN2 locus) have been developed to address this
- This species specificity has major implications for preclinical drug development
- Standard RNA-seq can detect cryptic exon inclusion by identifying novel splice junctions
- Short-read sequencing identifies junction reads spanning cryptic-canonical exon boundaries
- Long-read sequencing (PacBio, Oxford Nanopore) reveals full-length transcript isoforms containing cryptic exons
- Computational tools: MAJIQ, rMATS, LeafCutter designed for differential splicing analysis
- Targeted RT-PCR with primers spanning cryptic exons provides rapid, sensitive detection
- Quantitative RT-PCR measures the ratio of cryptic vs. normal transcripts
- Can be applied to patient-derived iPSC neurons and post-mortem tissue
- Antibodies against cryptic exon-encoded peptide sequences (e.g., truncated STMN2)
- Enables spatial mapping of cryptic exon inclusion in tissue sections
- Confirms correlation between TDP-43 mislocalization and cryptic exon inclusion at single-cell level
ASOs that block cryptic exon inclusion are a leading therapeutic approach.
- STMN2-targeted ASOs: Designed to bind the cryptic exon splice site in STMN2, preventing its inclusion and restoring normal STMN2 protein expression
- UNC13A-targeted ASOs: Block the cryptic exon in UNC13A to maintain synaptic function
- Splice-switching ASOs do not degrade the target mRNA — they redirect splicing
- Multiple ASO candidates in preclinical development for ALS
Rather than targeting individual cryptic exons, restoring nuclear TDP-43 could correct all downstream splicing defects.
- Nuclear import enhancers
- Reducing cytoplasmic TDP-43 aggregation
- Chaperone-based therapies to maintain TDP-43 solubility
- Challenge: once TDP-43 forms cytoplasmic aggregates, nuclear restoration may be difficult
- AAV-mediated delivery of STMN2 to motor neurons could bypass the cryptic exon problem
- CRISPR-based editing to strengthen the normal splice sites and weaken cryptic splice sites
- Challenges include delivery to motor neurons and long-term expression
While TDP-43-dependent cryptic exons are best characterized, analogous mechanisms may operate in other diseases:
FUS also functions as a splicing regulator, and FUS depletion leads to its own set of cryptic exon inclusions, relevant to FUS-ALS
hnRNPA1 and hnRNPA2B1 mutations cause multisystem proteinopathy; loss of these proteins may cause cryptic exon inclusion in their target transcripts
¶ PTBP1/PTBP2 and Neuronal Splicing
Polypyrimidine tract binding proteins regulate neuronal-specific splicing programs. Their dysfunction could contribute to cryptic exon events in neurodegeneration
¶ Clinical and Diagnostic Implications
- Biomarker development: Detection of truncated STMN2 or UNC13A transcripts as biomarkers for TDP-43 nuclear depletion
- Patient stratification: Genotyping UNC13A risk SNPs to identify patients who may benefit most from UNC13A-targeted therapies
- Disease staging: Measuring the extent of cryptic exon inclusion could indicate disease progression
- Companion diagnostics: Monitoring cryptic exon levels during ASO clinical trials as a pharmacodynamic readout
- How many cryptic exons does TDP-43 repress across the entire transcriptome, and which are the most pathologically consequential?
- Can targeting just 2-3 key cryptic exons (STMN2, UNC13A) provide meaningful therapeutic benefit, or must nuclear TDP-43 function be globally restored?
- How do cell type-specific transcriptomic differences affect cryptic exon vulnerability?
- What is the temporal sequence — does cryptic exon inclusion begin before or after clinical symptom onset?
- Can cryptic exon-derived peptides serve as neoantigens recognized by the immune system, contributing to neuroinflammation?
Recent publications highlighting key advances in this mechanism:
- Decoding TDP-43: the molecular chameleon of neurodegenerative diseases.
- An ANXA11 P93S variant dysregulates TDP-43 and causes corticobasal syndrome.
- Depletion of TDP-43 exacerbates tauopathy-dependent brain atrophy by sensitizing vulnerable neurons ...
- Abundant transcriptomic alterations in the human cerebellum of patients with a C9orf72 repeat expans...
- Understanding age-related pathologic changes in TDP-43 functions and the consequence on RNA splicing...