Used GLM 5.1 tool to compile list of protein machinery involved in the biogenisis and function of miRNA and lncRNA in corals. This will be used to put together an ncRNA machinery fasta file, to be used in conjunction with Steven/Hollie’s Machinery.fasta, to look for miRNA targets that are involved in epigenetic function. This is necessary as the Machinery.fasta file appears to exclude machinery related to ncRNAs, instead focusing on the “classic” epigenetic features (DNA methylation, chromatin modification, and histone variants and modifications)

Prompt used:

I am a molecular ecology scientist researching epigenetics in corals. Please compile a list of protein machinery that are involved in miRNA and lncRNA biogenisis and function. Provide all relevant sources that you utilize. Add a column that details the function/contribution of each protein/complex in the context of miRNA and lncRNA biogenisis and function.

Protein Machinery Involved in miRNA and lncRNA Biogenesis and Function

miRNA Biogenesis — Nuclear Step
miRNA Biogenesis — Nuclear Export
miRNA Biogenesis — Cytoplasmic Step
miRNA Function — RISC and Effector Machinery
miRNA Biogenesis — Accessory/Regulatory Proteins
lncRNA Biogenesis — Transcription & Processing
lncRNA Function — Chromatin-Interacting Machinery
lncRNA Function — Post-Transcriptional & Cytoplasmic
Shared Machinery Between miRNA and lncRNA Pathways
References

1. miRNA Biogenesis — Nuclear Step

1.1 The Microprocessor Complex

The Microprocessor is the core nuclear complex that cleaves primary miRNA transcripts (pri-miRNAs) into precursor miRNAs (pre-miRNAs). It is minimally composed of:

Protein	Gene	Function	Contribution to miRNA Pathway	Key References
DROSHA	DROSHA (formerly RNASEN)	Class 2 RNase III endonuclease; catalytic subunit that cleaves pri-miRNA at the base of the hairpin stem ~11 bp from the ssRNA-dsRNA junction, generating a ~70 nt pre-miRNA with a 2-nt 3′ overhang	Gatekeeper of canonical miRNA biogenesis. Acts at the first committed step — without DROSHA cleavage, no canonical miRNA can be produced. DROSHA determines the 5′ and 3′ boundaries of the mature miRNA by its precise cleavage position, directly defining the miRNA seed sequence and hence target specificity. Loss of DROSHA abolishes virtually all canonical miRNA production, leading to embryonic lethality and impaired stem cell differentiation. The 2-nt 3′ overhang it generates is the recognition signal for the next step (Exportin-5 and Dicer), coupling nuclear and cytoplasmic processing.	Lee et al. (2003) Nature 425:415–419; Han et al. (2004) Genes Dev. 18:3016–3027
DGCR8	DGCR8 (DiGeorge syndrome critical region 8; aka Pasha in Drosophila and C. elegans)	dsRNA-binding protein cofactor; dimerizes and stabilizes DROSHA; recognizes the ssRNA-dsRNA junction of pri-miRNA via its heme-binding domain and dsRBDs; essential for proper positioning of DROSHA on pri-miRNA substrates	Essential orientation factor for the Microprocessor. DGCR8 determines WHERE DROSHA cleaves by recognizing the junction between the single-stranded flanking RNA and the double-stranded stem of the pri-miRNA hairpin. Without DGCR8, DROSHA cannot accurately locate its cleavage site, leading to aberrant or absent pre-miRNA production. DGCR8 also stabilizes DROSHA protein (preventing its degradation) and heme binding to DGCR8 stimulates dimerization and processing activity. In an auto-regulatory feedback loop, DROSHA cleaves DGCR8 mRNA to control DGCR8 levels, ensuring homeostatic balance of Microprocessor output.	Gregory et al. (2004) Nature 432:235–240; Han et al. (2004) Genes Dev. 18:3016–3027; Nguyen et al. (2015) Cell 161:1374–1387; Partin et al. (2020) Nat. Commun. 8:1737

Structural notes: The functional Microprocessor is a heterotrimer of one DROSHA and two DGCR8 molecules. DGCR8 dimerization is stimulated by heme binding. DROSHA contains two RNase III domains (RIIID1 and RIIID2) that form the catalytic center, a central domain (CED), an arginine/serine-rich (R/S-rich) domain, and C-terminal dsRBDs. DGCR8 contains an N-terminal heme-binding domain, dsRBDs, and a C-terminal tail (CTT) that stabilizes DROSHA. DROSHA and DGCR8 engage in an auto-regulatory feedback loop: DROSHA cleaves DGCR8 mRNA to control DGCR8 levels, while DGCR8 stabilizes DROSHA protein (Han et al. 2009 Cell 136:75–84).

2. miRNA Biogenesis — Nuclear Export

Protein	Gene	Function	Contribution to miRNA Pathway	Key References
Exportin-5 (XPO5)	XPO5	Karyopherin-β family nuclear export receptor; recognizes the ~2-nt 3′ overhang and dsRNA stem of pre-miRNA; transports pre-miRNA from nucleus to cytoplasm in a RanGTP-dependent manner; also stabilizes pre-miRNA by protecting from nuclear degradation	Essential nuclear-cytoplasmic shuttle for miRNA maturation. XPO5 is the sole export receptor for pre-miRNAs and bridges the nuclear and cytoplasmic steps of miRNA biogenesis. Without XPO5, pre-miRNAs accumulate in the nucleus and are degraded by the nuclear exosome, causing a global collapse of mature miRNA levels. XPO5 also acts as a quality control checkpoint — it selectively exports RNAs with the correct 2-nt 3′ overhang and stem length, ensuring that only properly processed DROSHA products reach DICER. XPO5 binding additionally stabilizes pre-miRNA by shielding it from nuclear nucleases.	Lund et al. (2004) Science 303:95–98; Yi et al. (2003) Genes Dev. 17:3011–3016; Kim et al. (2024) Nat. Rev. Mol. Cell Biol. (re-evaluation of XPO5 roles)
RanGTP	RAN	GTPase; provides the energy and directionality for XPO5-mediated nuclear export	Energy provider for directional pre-miRNA export. RanGTP hydrolysis drives the conformational change in XPO5 that releases pre-miRNA into the cytoplasm. The RanGTP gradient (high nuclear, low cytoplasmic) ensures unidirectional transport — without it, pre-miRNA export stalls and miRNA biogenesis fails at the nuclear-cytoplasmic boundary.	Lund et al. (2004) Science 303:95–98

3. miRNA Biogenesis — Cytoplasmic Step

3.1 The Dicer Complex

Protein	Gene	Function	Contribution to miRNA Pathway	Key References
DICER	DICER1	Class 3 RNase III endonuclease; cleaves pre-miRNA at the terminal loop to generate a ~22 nt miRNA duplex; acts as a “molecular ruler” measuring distance from the 5′/3′ ends of the pre-miRNA; contains a PAZ domain (binds 3′ overhang), two RNase III domains, a helicase domain, and a dsRBD	Final processing enzyme that liberates the mature miRNA duplex. DICER performs the second and final endonucleolytic cleavage, converting the ~70 nt pre-miRNA hairpin into a ~22 nt miRNA duplex ready for RISC loading. DICER’s PAZ domain recognizes the 2-nt 3′ overhang left by DROSHA, and its RNase III domains cleave both strands at a measured distance from the ends — this “molecular ruler” mechanism ensures consistent miRNA length across the miRNA repertoire. DICER loss abolishes mature miRNA production from pre-miRNAs, causing developmental lethality, defective stem cell differentiation, and widespread derepression of miRNA target genes. DICER also partners with TRBP/AGO2 to form the RISC Loading Complex.	Hutvágner et al. (2001) Science 293:834–838; Ketting et al. (2001) EMBO J. 20:4234–4243
TRBP (TAR RNA-binding protein)	TARBP2	dsRNA-binding protein; cofactor of DICER; binds pre-miRNA and enhances DICER processing efficiency; contains three dsRBDs (two bind pre-miRNA, one interacts with DICER); part of the RISC Loading Complex (RLC)	DICER cofactor that optimizes pre-miRNA cleavage and bridges DICER to AGO. TRBP positions the pre-miRNA substrate for efficient and accurate DICER cleavage by binding the dsRNA stem via its dsRBDs. Critically, TRBP also serves as the molecular bridge between DICER and AGO2 in the RISC Loading Complex, directly coupling pre-miRNA cleavage with strand selection and AGO loading — ensuring that the miRNA duplex is efficiently transferred to its effector protein without release into the cytoplasm. Loss of TRBP reduces DICER processing efficiency and impairs RLC assembly, leading to decreased mature miRNA levels.	Chendrimada et al. (2005) Nature 436:740–744; Haase et al. (2005) EMBO J. 24:4089–4100
PACT (Protein activator of PKR)	PRKRA	dsRNA-binding protein; paralog of TRBP; associates with DICER to facilitate miRNA processing similarly to TRBP; contains dsRBDs	Parallel DICER cofactor with partially redundant function. PACT enhances DICER processing of pre-miRNAs similarly to TRBP but can also function independently in specific contexts. Unlike TRBP, PACT also activates PKR (protein kinase R) in response to dsRNA, linking miRNA biogenesis to innate immune signaling. Loss of PACT reduces miRNA production in certain cell types and impairs the cellular response to viral infection.	Lee et al. (2006) EMBO J. 25:522–532; Lee HY et al. (2013) Mol. Cell 51:735–747
Loqs-PB	loquacious (in Drosophila)	Drosophila homolog of TRBP/PACT; required for efficient pre-miRNA processing by Dicer-1	The invertebrate functional equivalent of TRBP/PACT. Loqs-PB is essential for Dicer-1 activity in flies; without it, pre-miRNA processing is severely impaired. Included here for comparative purposes when studying miRNA machinery in cnidarians/corals, where homologs of Loqs/TRBP/PACT are expected but may not yet be fully annotated.	Jouravleva et al. (2022) Mol. Cell 82:4049–4063

4. miRNA Function — RISC and Effector Machinery

4.1 Core RISC Components

Protein	Gene	Function	Contribution to miRNA Pathway	Key References
AGO1	AGO1	Argonaute protein; loads miRNA guide strand; mediates translational repression and mRNA destabilization	Effector protein that executes miRNA-guided gene silencing (non-slicing). AGO1 binds the mature miRNA guide strand and uses its sequence as a template to find complementary mRNA targets. Upon target recognition, AGO1 recruits GW182/TNRC6 and downstream effectors to repress translation and promote mRNA decay. Without AGO1, a significant fraction of miRNAs have no effector platform for silencing. AGO1 also stabilizes bound miRNAs — AGO loading protects miRNAs from degradation.	Meister et al. (2004) Mol. Cell 15:185–197
AGO2	AGO2	Argonaute protein; the only human AGO with endonucleolytic (“slicer”) activity for perfectly complementary targets; core component of RISC; binds miRNA guide strand via MID (5′ phosphate recognition) and PAZ (3′ end recognition) domains	The catalytic heart of miRNA-mediated silencing and the only AGO with slicer activity. AGO2 is the most studied Argonaute and performs two critical roles: (1) as the universal RISC effector, it uses the miRNA guide to find and bind target mRNAs, then recruits TNRC6 and deadenylase machinery for translational repression and mRNA decay; (2) uniquely among human AGOs, it can endonucleolytically cleave (“slice”) perfectly complementary targets (e.g., some viral RNAs or siRNA-directed cleavage). AGO2 also plays a role in miRNA biogenesis itself — it processes some noncanonical pre-miRNAs independently of DICER. Loss of AGO2 disrupts the majority of miRNA-mediated silencing and is embryonic lethal.	Liu et al. (2004) Science 305:1437–1441; Rivas et al. (2005) Nat. Struct. Mol. Biol. 12:340–349
AGO3	AGO3	Argonaute protein; miRNA-mediated gene silencing; lacks slicer activity	Effector for miRNA-guided silencing (non-slicing). AGO3 contributes to the overall pool of RISC complexes that execute miRNA-mediated translational repression and mRNA destabilization, but lacks the slicer activity of AGO2. Functional redundancy with AGO1 and AGO4 in standard miRNA silencing, but may have specialized roles in specific cell types or contexts.	Hauptmann et al. (2005) RNA 11:1580–1590
AGO4	AGO4	Argonaute protein; miRNA-mediated gene silencing; lacks slicer activity	Effector for miRNA-guided silencing (non-slicing). Similar to AGO1 and AGO3, AGO4 forms miRNA-loaded RISC complexes for target repression. AGO4 may have specialized nuclear functions, including roles in transcriptional gene silencing and heterochromatin formation in some contexts.	Hauptmann et al. (2005) RNA 11:1580–1590

4.2 GW182/TNRC6 Scaffolding Proteins

Protein	Gene	Function	Contribution to miRNA Pathway	Key References
TNRC6A (GW182)	TNRC6A	Scaffolding protein; binds AGO proteins via Gly-Trp (GW) repeats; recruits deadenylase complexes; bridges AGO-miRNA complex to mRNA decay machinery; essential for miRNA-mediated gene silencing	The critical bridge between miRNA-target recognition and mRNA destruction. TNRC6A is the essential scaffolding protein that converts AGO-miRNA binding from a simple recognition event into an active mRNA silencing event. When AGO-miRNA finds its target mRNA, TNRC6A is recruited via GW repeats and serves as a docking platform for: (1) CCR4-NOT deadenylase complex (initiating poly(A) tail removal), (2) PAN2-PAN3 (initial deadenylation), and (3) PABP (coordinating deadenylation with translation repression). Without TNRC6, AGO-miRNA can bind targets but cannot efficiently repress them — silencing is severely compromised.	Liu et al. (2005) Nat. Cell Biol. 7:1261–1266; Behm-Ansmant et al. (2006) Genes Dev. 20:1885–1898
TNRC6B	TNRC6B	Paralog of TNRC6A; binds AGO proteins; recruits deadenylase machinery for mRNA silencing	Redundant scaffolding bridge for miRNA silencing. TNRC6B performs the same scaffolding function as TNRC6A, providing functional redundancy. The three TNRC6 paralogs may have tissue-specific expression patterns or subtle differences in AGO binding preference, but all can independently mediate miRNA-directed mRNA decay and translational repression.	Baillat & Shiekhattar (2009) Mol. Cell. Biol. 29:4144–4155
TNRC6C	TNRC6C	Paralog of TNRC6A; binds AGO proteins; can recapitulate biphasic deadenylation for mRNA decay	Redundant scaffolding bridge for miRNA silencing. Same function as TNRC6A/B with demonstrated ability to independently drive the biphasic deadenylation (PAN2-PAN3 followed by CCR4-NOT) that commits miRNA-targeted mRNAs to decay.	Chen et al. (2009) Nat. Struct. Mol. Biol. 16:1208–1214; Lazzaretti et al. (2009) Mol. Cell. Biol. 29:6379–6392

4.3 Deadenylation Machinery (Recruited by TNRC6)

Protein/Complex	Gene(s)	Function	Contribution to miRNA Pathway	Key References
CCR4-NOT complex	CNOT1–CNOT11, CNOT7 (CAF1), CNOT6 (CCR4)	Major deadenylase complex; recruited by TNRC6; removes poly(A) tail leading to mRNA destabilization; CAF1 (CNOT7/CNOT8) and CCR4 (CNOT6/CNOT6L) are the catalytic deadenylase subunits	Primary executioner of miRNA-directed mRNA degradation. The CCR4-NOT complex is the major deadenylase recruited by TNRC6 and carries out the second (major) phase of poly(A) tail removal during miRNA-mediated mRNA decay. CCR4 and CAF1 are the catalytic subunits that progressively remove adenosine residues from the poly(A) tail. Deadenylation is the rate-limiting step in miRNA-mediated silencing — once the poly(A) tail is shortened, the mRNA becomes vulnerable to decapping and exonucleolytic degradation. CCR4-NOT also contributes to translational repression independent of deadenylation. Loss of CCR4-NOT severely impairs miRNA-mediated mRNA decay.	Behm-Ansmant et al. (2006) Genes Dev. 20:1885–1898; Braun et al. (2013) Adv. Exp. Med. Biol. 768:147–163
PAN2–PAN3 complex	PAN2, PAN3	Deadenylase complex; mediates the initial (first phase) poly(A) tail trimming; PAN2 is the catalytic subunit; PAN3 bridges to TNRC6	Initiator of miRNA-directed biphasic deadenylation. PAN2-PAN3 performs the first, rapid phase of poly(A) tail trimming before CCR4-NOT carries out the extensive removal. PAN3 directly interacts with TNRC6, linking the miRISC to the first deadenylation step. This biphasic mechanism (PAN2-PAN3 → CCR4-NOT) ensures efficient and processive poly(A) removal, accelerating mRNA decay beyond what would be achieved by either deadenylase alone.	Chen et al. (2009) Nat. Struct. Mol. Biol. 16:1208–1214
PABP (Poly(A)-binding protein)	PABPC1	Binds poly(A) tail; interacts with TNRC6 (via PAM2 motif); coordinates deadenylation	Coordinator of deadenylation and translational repression. PABP bound to the poly(A) tail must be displaced or reconfigured for deadenylases to access the tail. TNRC6’s PAM2 motif interacts with PABP, displacing it from the poly(A) tail and simultaneously recruiting deadenylases. PABP is also a positive regulator of translation; its displacement by TNRC6 contributes directly to translational repression. Thus, PABP serves as a nexus connecting miRNA-directed deadenylation and translation inhibition.	Zekri et al. (2013) Mol. Cell. Biol. 33:22–34
PARN (Poly(A)-specific ribonuclease)	PARN	Deadenylase; involved in context-specific miRISC-mediated decay (e.g., miR-125b-AGO2 RISC)	Context-specific deadenylase for select miRISC complexes. PARN provides an alternative deadenylation route in specific miRISC contexts (e.g., miR-125b-AGO2 RISC), expanding the repertoire of deadenylation mechanisms available to miRNA-mediated silencing. Its role appears more specialized than the CCR4-NOT/PAN2-PAN3 axis.	Narasimhan et al. (2025) Noncoding RNA 11(4):62

4.4 Decapping & 5′–3′ Decay Machinery

Protein/Complex	Gene	Function	Contribution to miRNA Pathway	Key References
DCP1–DCP2	DCP1A/B, DCP2	Decapping enzyme complex; removes the 5′ m7G cap after deadenylation, committing the mRNA to 5′→3′ exonucleolytic decay	The irreversible commitment step in miRNA-mediated mRNA destruction. After deadenylation removes the poly(A) tail, DCP2 (the catalytic subunit) removes the 5′ m7G cap, making the mRNA irreversibly committed to degradation. DCP1 stimulates DCP2 activity. Decapping is the critical “point of no return” — once the cap is removed, the mRNA cannot be rescued by re-polyadenylation. This step is downstream of and dependent upon miRNA-directed deadenylation.	Chen et al. (2009) Nat. Struct. Mol. Biol. 16:1208–1214; Wang et al. (2002) Proc. Natl Acad. Sci. USA 99:12663–12668
EDC3, EDC4 (Hedls)	EDC3, EDC4	Enhancers of decapping; facilitate DCP1–DCP2 activity	Decapping enhancers that accelerate miRNA-mediated mRNA decay. EDC3 and EDC4 stimulate DCP2 decapping activity and help assemble mRNA decay complexes in P-bodies, ensuring efficient miRNA-directed mRNA turnover.	Narasimhan et al. (2025) Noncoding RNA 11(4):62
XRN1	XRN1	5′→3′ exoribonuclease; degrades decapped mRNA; also degrades mature miRNAs in some contexts	Terminal exonuclease that degrades miRNA-targeted mRNAs after decapping. XRN1 carries out the final degradation of the mRNA body after DCP2 removes the cap — this is the last step in miRNA-directed mRNA destruction. XRN1 also plays a role in mature miRNA turnover (degradation) in some contexts, providing a mechanism for regulating miRNA abundance independently of biogenesis.	Chang et al. (2012) Nat. Struct. Mol. Biol. 19:572–579

4.5 RISC Loading Chaperones

Protein	Gene	Function	Contribution to miRNA Pathway	Key References
Hsc70/Hsp90 chaperone machinery	HSPA8 (Hsc70), HSP90AA1/HSP90AB1	ATP-dependent chaperone machinery; required for loading of small RNA duplexes onto AGO proteins; facilitates conformational changes in AGO during RISC assembly	Essential chaperones that open AGO for miRNA loading and activate RISC. AGO proteins in their apo (empty) state are in a “closed” conformation that cannot accept miRNA duplexes. Hsc70/Hsp90 use ATP hydrolysis to induce a conformational opening of AGO, allowing the miRNA duplex to enter. The chaperones then facilitate the retention of the guide strand and ejection of the passenger strand. Without this chaperone activity, RISC loading fails and miRNAs cannot be loaded into their effector proteins — silencing is abrogated even if miRNA duplexes are produced.	Nakanishi (2016) Wiley Interdiscip. Rev. RNA 7:637–660; Iwasaki et al. (2010) Cell 140:672–680

4.6 Translational Repression

Protein	Gene	Function	Contribution to miRNA Pathway	Key References
DDX6	DDX6 (RCK/p54)	DEAD-box RNA helicase; recruited by CCR4-NOT complex; promotes translational repression and P-body formation	Effector of miRNA-mediated translational repression. DDX6 is recruited by the CCR4-NOT complex upon miRNA target recognition and directly inhibits translation initiation by interfering with eIF4F assembly and promoting 48S complex disassembly. DDX6 also promotes the aggregation of silenced mRNPs into P-bodies (processing bodies), where miRNA-targeted mRNAs are stored or degraded. This makes DDX6 a key executor of translational silencing — the earliest detectable effect of miRNA action, occurring before mRNA decay.	Narasimhan et al. (2025) Noncoding RNA 11(4):62; Chen et al. (2014) Nat. Commun. 5:4722

5. miRNA Biogenesis — Accessory/Regulatory Proteins

These RNA-binding proteins (RBPs) modulate miRNA biogenesis by binding pri-miRNA or pre-miRNA terminal loops, altering structure, or interacting with the core processing machinery.

5.1 Microprocessor-Associated (Nuclear) Regulators

Protein	Gene	Function	Effect	Contribution to miRNA Pathway	Key References
DDX5 (p68)	DDX5	DEAD-box RNA helicase; Microprocessor auxiliary factor; facilitates pri-miRNA recognition and processing	Promotes	RNA helicase cofactor that broadens Microprocessor substrate range and efficiency. DDX5 associates with the Microprocessor and is required for efficient processing of a large subset of pri-miRNAs (94 of 266 examined in knockout studies). It likely remodels pri-miRNA secondary structure or resolves RNP complexes to make the cleavage site more accessible to DROSHA-DGCR8. Without DDX5, many pri-miRNAs are processed inefficiently, leading to reduced levels of a broad set of mature miRNAs. DDX5 thus acts as a global amplifier of Microprocessor activity.	Gregory et al. (2004) Nature 432:235–240; Shiohama et al. (2007) Nucleic Acids Res. 35:5899–5911
DDX17 (p72)	DDX17	DEAD-box RNA helicase; Microprocessor auxiliary factor; works cooperatively with p68; its sequestration by YAP downregulates Microprocessor activity	Promotes	RNA helicase cofactor that cooperates with DDX5 and links Microprocessor to signaling pathways. DDX17 works alongside DDX5 to promote pri-miRNA processing. Critically, DDX17 is sequestered by the Hippo pathway effector YAP in low cell-density conditions — this provides a direct signaling-Microprocessor link, allowing cellular context (e.g., cell density, proliferation status) to globally tune miRNA output. Without DDX5/DDX17, the Microprocessor processes many pri-miRNAs less efficiently, and this regulatory connection to signaling is lost.	Gregory et al. (2004) Nature 432:235–240; Newman et al. (2010) Genes Dev. 24:1086–1092
hnRNP A1	HNRNPA1	Heterogeneous nuclear ribonucleoprotein; binds pri-miR-18a terminal loop and facilitates processing by altering pri-miRNA structure	Promotes (context-dependent)	Structure-remodeling RBP that enables processing of specific pri-miRNAs. hnRNP A1 binds the terminal loop of pri-miR-18a and induces a conformational change in the hairpin structure that facilitates DROSHA recognition and cleavage. This is a context-dependent requirement — moving the miR-18a hairpin to a different flanking sequence context eliminates the need for hnRNP A1, suggesting that hnRNP A1 overcomes inhibitory structural features in the native context. This demonstrates that individual miRNAs can have dedicated processing cofactors that overcome structural barriers, adding a layer of specificity to miRNA biogenesis regulation.	Guil & Cáceres (2007) Nat. Struct. Mol. Biol. 14:701–707
KSRP (KH-type splicing regulatory protein)	KHSRP	KH-domain RBP; binds G-rich terminal loop sequences of a subset of pri/pre-miRNAs (let-7, miR-20, miR-26b, miR-106a, miR-21, miR-16); component of both Drosha and Dicer complexes; promotes recruitment of both	Promotes	Biphasic miRNA processing activator that operates at both nuclear and cytoplasmic steps. KSRP is unique among miRNA regulators in that it functions in BOTH the nucleus (promoting DROSHA processing) and the cytoplasm (promoting DICER processing). By binding the terminal loop of a specific subset of pri/pre-miRNAs, KSRP recruits the processing machinery and enhances cleavage efficiency at both steps. This dual-step promotion ensures robust production of a defined subset of miRNAs. KSRP antagonizes LIN28 binding to let-7 loops — where LIN28 blocks processing, KSRP promotes it — creating a competitive switch for let-7 biogenesis that determines stem cell vs. differentiated cell fates.	Trabucchi et al. (2009) Nature 459:1010–1014
LIN28A / LIN28B	LIN28A, LIN28B	RNA-binding protein with cold-shock domain and zinc knuckles; selectively binds let-7 family pri/pre-miRNA terminal loops; blocks Drosha and Dicer processing; recruits TUT4/ZCCHC11 for pre-let-7 uridylation and degradation	Inhibits (let-7 specific)	Master developmental switch that blocks let-7 miRNA production to maintain pluripotency. LIN28 proteins are the most selective miRNA regulators known — they specifically target the let-7 family, which promotes differentiation. LIN28 acts at TWO steps: (1) in the nucleus, binding to pri-let-7 terminal loops to block DROSHA processing; (2) in the cytoplasm, binding pre-let-7 to block DICER processing AND recruiting TUT4/TUT7 to add oligo-U tails that mark pre-let-7 for DIS3L2-mediated degradation. This multi-step blockade ensures near-complete suppression of let-7 in stem cells. The LIN28/let-7 axis is a bistable switch governing the pluripotent-to-differentiated state transition — high LIN28 maintains stemness, high let-7 drives differentiation. Misregulation of this axis drives tumorigenesis.	Viswanathan et al. (2008) Science 320:97–100; Newman et al. (2008) RNA 14:1539–1549; Heo et al. (2008) Mol. Cell 32:276–284
TDP-43	TARDBP	DNA/RNA-binding protein; stimulates pri- and pre-miRNA processing for miR-132, miR-143, and others	Promotes	RBP that promotes a specific subset of miRNAs at both processing steps. TDP-43 enhances both DROSHA and DICER processing of a subset of miRNAs (miR-132, miR-143, and others) by binding their precursors. TDP-43 loss leads to reduced levels of these miRNAs, linking neurodegenerative disease-associated TDP-43 pathology to miRNA dysregulation.	Kawahara & Mieda-Sato (2012) Nucleic Acids Res. 40:3415–3427
BRCA1	BRCA1	DNA-binding tumor suppressor; interacts with DROSHA; stimulates pri-miRNA processing	Promotes	Tumor suppressor that globally enhances pri-miRNA processing. BRCA1 physically interacts with DROSHA and promotes pri-miRNA processing, connecting DNA damage response and tumor suppression to miRNA biogenesis. Loss of BRCA1 leads to reduced mature miRNA levels, contributing to oncogenic derepression of miRNA targets. This provides a mechanism for the tumor suppressor function of BRCA1 beyond its DNA repair role.	Kawai et al. (2014) Oncogene 33:3907–3917
EWSR1	EWSR1	RNA-binding protein; auxiliary component of the larger Drosha complex	Facilitates	Auxiliary factor of the extended Microprocessor complex. EWSR1 is part of the larger DROSHA-associated complex but its precise mechanistic contribution remains less well defined than DDX5/DDX17. It may assist in substrate recognition or in coupling transcription to pri-miRNA processing.	Gregory et al. (2004) Nature 432:235–240
FUS	FUS	RNA-binding protein; auxiliary component of the larger Drosha complex	Facilitates	Auxiliary factor of the extended Microprocessor complex. Like EWSR1, FUS is part of the larger DROSHA-associated complex and may assist in substrate recognition or coupling of transcription with pri-miRNA processing. FUS mutations are linked to ALS, and miRNA biogenesis defects may contribute to disease pathology.	Gregory et al. (2004) Nature 432:235–240

5.2 Cytoplasmic Regulators

Protein	Gene	Function	Effect	Contribution to miRNA Pathway	Key References
TUT4/ZCCHC11	ZCCHC11	Terminal uridylyl transferase (TUTase); recruited by LIN28 to add oligo-U tail to pre-let-7, marking it for degradation by DIS3L2	Inhibits (let-7)	Effector of LIN28-mediated let-7 degradation via oligo-uridylation. TUT4 is the enzymatic effector that carries out LIN28’s instructions to destroy let-7 precursors. When LIN28 binds pre-let-7, it recruits TUT4, which adds an oligo-U tail (3′ uridylation) to the pre-miRNA. This oligo-U tail is recognized by the exonuclease DIS3L2, which degrades the uridylated pre-let-7. This is a critical switch — without TUT4, LIN28 cannot efficiently eliminate pre-let-7, and let-7 accumulates even in the presence of LIN28, driving premature differentiation.	Heo et al. (2009) Mol. Cell 32:276–284; Hagan et al. (2009) Nat. Cell Biol. 11:1152–1157
TUT7/ZCCHC6	ZCCHC6	TUTase; works with TUT4 in LIN28-mediated pre-let-7 uridylation	Inhibits (let-7)	Cooperative TUTase that enhances LIN28-mediated pre-let-7 uridylation. TUT7 works redundantly with TUT4 to uridylate pre-let-7. Double knockout of both TUT4 and TUT7 completely abrogates LIN28-mediated let-7 suppression, whereas single knockouts show partial effects, demonstrating their cooperative action in this pathway.	Thornton et al. (2012) RNA 18:1983–1993
DIS3L2	DIS3L2	3′→5′ exoribonuclease; degrades oligouridylated pre-let-7	Degrades	Terminal exonuclease that degrades LIN28/TUT4-marked pre-let-7. DIS3L2 specifically recognizes and degrades oligo-uridylated pre-miRNAs (the product of LIN28-TUT4 activity), completing the let-7 destruction pathway. DIS3L2 is unique among 3′→5′ exonucleases in its strong preference for uridylated substrates, making it the dedicated “cleanup” enzyme for the LIN28 pathway. Loss of DIS3L2 leads to accumulation of uridylated pre-let-7 fragments and is associated with Wilms tumor.	Chang et al. (2013) EMBO J. 32:2173–2185
ADAR1/ADAR2	ADAR1, ADAR2	Adenosine deaminases acting on RNA; edit adenosine to inosine in pri-/pre-miRNAs; editing can reduce processing efficiency or alter target specificity	Context-dependent	RNA editing enzymes that diversify or block miRNA function. A-to-I editing within pri-miRNA or pre-miRNA sequences can have two major consequences: (1) editing within the stem can disrupt dsRNA structure, reducing DROSHA or DICER processing efficiency (blocking miRNA production); (2) editing within the mature miRNA sequence can alter the seed region, changing the set of mRNA targets the miRNA recognizes (retargeting the miRNA). This creates a mechanism for generating miRNA isoforms (isomiRs) with different target specificities from the same genomic locus, adding a layer of regulatory diversity.	Yang et al. (2006) J. Biol. Chem. 281:15857–15864
MCPIP1/ZC3H12A	ZC3H12A	Endoribonuclease; cleaves pre-miRNAs to inhibit processing	Inhibits	Pre-miRNA nuclease that directly destroys miRNA precursors. MCPIP1 directly cleaves pre-miRNAs at the terminal loop, destroying them before they can be processed by DICER. This is a rare example of a dedicated anti-miRNA nuclease that acts independently of the LIN28/TUT4 pathway. MCPIP1 provides a mechanism for rapidly downregulating miRNA levels in response to inflammatory signals, linking innate immunity to miRNA regulation.	Suzuki et al. (2011) Nat. Immunol. 12:1156–1163
RBM3	RBM3	RNA-binding motif protein; stimulates pre-miRNA processing	Promotes	Cold-inducible RBP that enhances pre-miRNA processing. RBM3 promotes DICER processing of pre-miRNAs and is upregulated in response to mild hypothermia, providing a mechanism for temperature-dependent regulation of miRNA biogenesis. This is particularly relevant for coral biology, where temperature fluctuations directly impact miRNA regulation through as-yet poorly characterized mechanisms.	Dresios et al. (2005) J. Neurosci. 25:5418–5425
YB-1	YBX1	RNA-binding protein; regulates a subset of miRNAs at the Dicer step	Context-dependent	RBP that modulates DICER processing of a subset of pre-miRNAs. YB-1 binds pre-miRNAs and can either promote or inhibit DICER processing depending on the specific miRNA and cellular context. YB-1 is also a major component of mRNA mRNP particles and links miRNA biogenesis to broader RNA metabolism.	Wu et al. (2015) RNA Biol. 12:133–144
DAZL	DAZL	RNA-binding protein; promotes pre-miRNA processing in germ cells	Promotes	Germ cell-specific miRNA processing factor. DAZL promotes pre-miRNA processing specifically in germ cells, contributing to the specialized miRNA repertoire required for gametogenesis. This demonstrates tissue-specific regulation of miRNA biogenesis.	Yan et al. (2022) as cited in Hynes & Kakumani (2024)
Nibbler	Nibbler (fly) / PNPT1 (mammalian homolog)	3′→5′ exonuclease; trims mature miRNA 3′ ends (miRNA trimming/ismiR formation)	Modifies	Post-loading miRNA trimming enzyme that refines miRNA target specificity. After a mature miRNA is loaded into AGO, Nibbler can trim nucleotides from the 3′ end, generating shorter miRNA isoforms (isomiRs) with altered target specificity. This “fine-tuning” of miRNA length after RISC loading provides a mechanism for adjusting the miRNA targetome without changing miRNA expression levels.	Han et al. (2011) Mol. Cell 43:1024–1034

6. lncRNA Biogenesis — Transcription & Processing

lncRNAs share many biogenesis steps with mRNAs (Pol II transcription, 5′ capping, splicing, 3′ polyadenylation), but several features and processing pathways are distinct.

6.1 Core Transcription & Processing Machinery (Shared with mRNA)

Protein/Complex	Function	Contribution to lncRNA Pathway	Key References
RNA Polymerase II (Pol II)	Transcribes most lncRNA genes; lncRNA genes often display distinct Pol II CTD phosphorylation patterns compared to mRNA genes (more Ser5P, less Ser2P)	The sole polymerase for lncRNA transcription. Pol II transcribes virtually all lncRNA genes, but with distinct kinetic and CTD phosphorylation signatures compared to mRNA genes. lncRNA genes typically show more Ser5P (associated with transcription initiation and early elongation) and less Ser2P (associated with productive elongation and 3′ processing), reflecting their generally slower transcription, reduced processivity, and weaker coupling to downstream processing. These distinct Pol II dynamics directly contribute to the characteristic low abundance and nuclear retention of many lncRNAs.	Nojima & Proudfoot (2022) Nat. Rev. Mol. Cell Biol. 23:123–137
CPSF (Cleavage and Polyadenylation Specificity Factor: CPSF160/WDR33/CPSF30/CPSF73/CPSF100)	3′-end processing of polyadenylated lncRNAs; CPSF73 (encoded by CPSF3) is the endonuclease that cleaves at the poly(A) signal (AAUAAA); CPSF100 is its inactive homolog	3′ end formation for polyadenylated lncRNAs. CPSF recognizes the AAUAAA poly(A) signal and cleaves the nascent transcript via CPSF73 endonuclease activity, defining the 3′ end of polyadenylated lncRNAs. This is a critical decision point — CPSF cleavage also triggers transcription termination by Pol II, coupling RNA 3′ end formation with the end of transcription. Many lncRNA genes have weaker poly(A) signals or use alternative polyadenylation sites compared to mRNA genes, contributing to their reduced stability and nuclear retention.	Nojima & Proudfoot (2022) Nat. Rev. Mol. Cell Biol. 23:123–137; Sun et al. (2018) Proc. Natl Acad. Sci. USA 115:E1419–E1428
CstF (Cleavage Stimulation Factor: CstF50/CstF64/CstF77)	Binds U/GU-rich downstream elements; stimulates cleavage at 3′ end	Enhancer of 3′ cleavage for lncRNA processing. CstF binds downstream U/GU-rich elements and stimulates CPSF-mediated cleavage. The relative weakness of downstream elements in many lncRNA genes leads to less efficient CstF recruitment, contributing to reduced 3′ processing and the characteristic instability of many lncRNA transcripts.	Nojima & Proudfoot (2022) Nat. Rev. Mol. Cell Biol. 23:123–137
CFI/CFIm (Cleavage Factor I: CFIm25/NUDT21, CFIm59, CFIm68)	Binds UGUA upstream elements; enhancer-dependent activator of 3′ processing	Alternative polyadenylation regulator that determines lncRNA isoform identity. CFIm binds UGUA elements upstream of the poly(A) site and influences the choice of alternative polyadenylation (APA) sites. For lncRNAs like NEAT1, APA choice determines which isoform is produced (NEAT1_1 vs. NEAT1_2), with radically different functional consequences (NEAT1_1 is a short cytoplasmic lncRNA; NEAT1_2 is a long nuclear scaffold essential for paraspeckle formation). CFIm thus acts as a master switch for lncRNA isoform identity.	Nojima & Proudfoot (2022) Nat. Rev. Mol. Cell Biol. 23:123–137
PAP (Poly(A) Polymerase)	Adds poly(A) tail to 3′ end of cleaved lncRNA	Poly(A) tail addition for lncRNA stability and export. PAP adds the poly(A) tail after CPSF cleavage, which protects the lncRNA 3′ end from exonucleolytic degradation and facilitates nuclear export. Many non-polyadenylated lncRNAs use alternative 3′ end formation mechanisms (Integrator, RNase P, snoRNP capping) that bypass PAP entirely.	Standard
PABPN1 (Poly(A)-binding protein nuclear 1)	Nuclear poly(A)-binding protein; controls poly(A) tail length; prevents nuclear export of improperly processed transcripts	Nuclear poly(A) quality control for lncRNA export competence. PABPN1 binds the poly(A) tail in the nucleus and controls its length, serving as a quality control checkpoint — only properly polyadenylated lncRNAs are allowed to be exported. PABPN1 also prevents the nuclear export of unspliced or improperly processed transcripts, ensuring that defective lncRNAs are retained and degraded. Loss of PABPN1 leads to nuclear export of aberrant lncRNAs and mRNA processing defects.	Kwiatek et al. (2021) Nucleic Acids Res. 48:6889–6905
Spliceosome (U1, U2, U4/U5/U6 snRNPs + ~150 proteins)	Removes introns from lncRNA primary transcripts; many lncRNAs are weakly or partially spliced compared to mRNAs	Intron removal and nuclear retention decision-maker for lncRNAs. The spliceosome processes lncRNA introns, but many lncRNAs are spliced far less efficiently than mRNAs — they often retain introns, have fewer exons, and use weaker splice sites. This weak splicing is not merely defective processing; it is functionally important: (1) intron retention contributes to nuclear retention of many lncRNAs, (2) inefficient splicing targets many lncRNAs for nuclear degradation, keeping their abundance low, and (3) some lncRNA functions (e.g., intron-derived sno-lncRNAs) specifically require unspliced regions. The spliceosome thus acts as a gatekeeper determining whether a lncRNA transcript is fully processed, exported, and functional, or retained and degraded.	Matera & Wang (2014) Nat. Rev. Mol. Cell Biol. 15:108–121

6.2 lncRNA-Specific Processing Machinery

Protein/Complex	Gene(s)	Function	Contribution to lncRNA Pathway	Key References
Integrator complex (INTS1–INTS14)	INTS1–INTS14	14-subunit complex; endonucleolytic 3′ processing of non-polyadenylated lncRNAs, enhancer RNAs (eRNAs), and snRNAs; INTS11 is the catalytic β-CASP/MBL endonuclease (homolog of CPSF73); INTS9 is the inactive homolog (homolog of CPSF100); binds Pol II CTD; restrains paraspeckle assembly by promoting NEAT1_2 to NEAT1_1 isoform switching	The dedicated 3′ processing machinery for non-polyadenylated lncRNAs and the master regulator of lncRNA isoform switching. Integrator is the primary 3′ endonuclease for lncRNAs that lack poly(A) tails — it cleaves nascent transcripts bound to the Pol II CTD, generating mature 3′ ends without polyadenylation. This is essential for eRNAs, many snRNA-like lncRNAs, and critically, for controlling NEAT1 isoform balance: Integrator cleaves NEAT1_2 (the long paraspeckle-scaffolding isoform), converting it to NEAT1_1 (the short isoform), thereby restraining paraspeckle formation. Loss of Integrator leads to NEAT1_2 accumulation, uncontrolled paraspeckle assembly, and misregulation of paraspeckle-sequestered proteins and RNAs. Integrator thus acts as a critical switch for nuclear body homeostasis through lncRNA isoform control.	Baillat et al. (2005) Cell 123:265–276; Lai et al. (2015) Nature 525:399–403; Plant et al. (2020) Sci. Adv. 6:eaaz9072; Pfleiderer & Moresco (2022) Transcription 12:251–265
RNase P	RPPH1 and multiple subunits	Endoribonuclease; generates mature 3′ ends of some non-polyadenylated lncRNAs (e.g., MALAT1) by cleavage	Alternative 3′ end formation for non-polyadenylated lncRNAs. RNase P cleaves the 3′ end of specific non-polyadenylated lncRNAs like MALAT1, generating a stable 3′ end without a poly(A) tail. The cleavage product (the upstream fragment) is further stabilized by a triple-helix structure at its 3′ end that protects it from degradation. This noncanonical 3′ end processing allows MALAT1 to accumulate to very high levels in the nucleus without a poly(A) tail, demonstrating an alternative stability mechanism that is critical for the function of certain nuclear-retained lncRNAs.	Wilusz et al. (2008) Cell 135:919–932; Wu et al. (2017) Trends Genet. 33:55–66
snoRNP complexes (snoRNA + fibrillarin, NOP56, NOP58, etc.)	Multiple	Cap both ends of some intron-derived lncRNAs (sno-lncRNAs) or one end (5′ snoRNA-ended, 3′ polyadenylated SPAs); protect from exonucleolytic degradation	Noncanonical end-protection machinery for intron-derived lncRNAs. snoRNP complexes bind the terminal loops of excised introns that contain snoRNAs, protecting both ends from exonucleases and generating stable intron-derived lncRNAs (sno-lncRNAs). This creates an entirely distinct class of lncRNAs that are protected at their ends not by m7G caps and poly(A) tails, but by snoRNP protein-RNA complexes. Without this snoRNP capping, these intronic sequences would be rapidly degraded by the nuclear debranching and exonuclease machinery. This mechanism generates functionally important lncRNAs (e.g., those from the Prader-Willi syndrome region) from what would otherwise be discarded intronic waste.	Wu et al. (2017) Trends Genet. 33:55–66; Yin et al. (2012) Mol. Cell 47:399–410

6.3 Nuclear Export of lncRNAs

Protein	Gene	Function	Contribution to lncRNA Pathway	Key References
NXF1/TAP	NXF1	Major mRNA export receptor; exports polyadenylated lncRNAs	Primary export receptor for polyadenylated lncRNAs. NXF1 exports the majority of polyadenylated lncRNAs from nucleus to cytoplasm, the same pathway used by mRNAs. Many lncRNAs are poor NXF1 substrates due to weak splicing and low association with the TREX export complex, contributing to their nuclear retention. The efficiency of NXF1-mediated export is a major determinant of whether a lncRNA acts in the nucleus (if retained) or cytoplasm (if exported), directly shaping its functional role.	Nojima & Proudfoot (2022) Nat. Rev. Mol. Cell Biol. 23:123–137
CRM1/Exportin-1	XPO1	Exports non-polyadenylated snRNAs and some lncRNAs; works with PHAX adaptor	Alternative export receptor for non-polyadenylated lncRNAs. CRM1 exports non-polyadenylated lncRNAs and snRNAs via the PHAX adaptor, providing an export route independent of the NXF1/TREX pathway. This is essential for lncRNAs that lack poly(A) tails and cannot use the NXF1 pathway.	Matera & Wang (2014) Nat. Rev. Mol. Cell Biol. 15:108–121
PHAX	PHAX	Phosphorylated adaptor for snRNA export; links CBC to CRM1	Adaptor that bridges cap-bound lncRNAs to the CRM1 export receptor. PHAX connects the cap-binding complex (CBC) to CRM1, specifically for snRNA and non-polyadenylated lncRNA export. Without PHAX, these non-polyadenylated RNAs cannot exit the nucleus.	Matera & Wang (2014) Nat. Rev. Mol. Cell Biol. 15:108–121
CBC (Cap-binding complex: CBP80/CBP20)	NCBP1, NCBP2	Binds 5′ m7G cap; links to export adaptors	5′ cap sensor that initiates lncRNA export pathway selection. CBC binds the m7G cap of newly transcribed lncRNAs and recruits export adaptors (PHAX for CRM1-dependent export, or TREX components for NXF1-dependent export). CBC thus acts as the initial “decision point” that determines which export pathway a lncRNA will use, based on its processing state and associated factors.	Matera & Wang (2014) Nat. Rev. Mol. Cell Biol. 15:108–121
ARS2 (Arsenite resistance protein 2)	SRRT	Links CBC to export pathways; involved in snRNA and lncRNA fate decisions	Fate decision factor that routes lncRNAs toward export or degradation. ARS2 binds CBC and helps determine whether a capped lncRNA will be exported or targeted for nuclear degradation. ARS2 promotes the export of snRNA-like lncRNAs (via PHAX/CRM1) while directing other lncRNAs toward the nuclear exosome for degradation (via the NEXT complex). ARS2 is thus a key junction in lncRNA fate: export for function, or degradation for quality control.	Matera & Wang (2014) Nat. Rev. Mol. Cell Biol. 15:108–121

6.4 Nuclear Surveillance / Degradation

Protein/Complex	Function	Contribution to lncRNA Pathway	Key References
Nuclear exosome (EXOSC1–EXOSC10, DIS3, DIS3L)	3′→5′ exoribonuclease complex; degrades many nascent lncRNAs cotranscriptionally; many lncRNAs are rapidly degraded by the exosome and accumulate only in specific cell types	Primary quality control and regulatory degradation machinery for lncRNAs. The nuclear exosome is the major 3′→5′ exonuclease that degrades the majority of nascent lncRNAs cotranscriptionally. This is not merely quality control — it is a regulatory mechanism that keeps lncRNA expression extremely low by default. Most lncRNAs are transcribed but immediately degraded; only those that escape exosome surveillance (through protective structures, polyadenylation, or specific RNP assembly) accumulate to functional levels. The exosome thus acts as a gatekeeper that determines which lncRNAs achieve functional expression, making it one of the most important regulators of lncRNA steady-state levels.	Nojima & Proudfoot (2022) Nat. Rev. Mol. Cell Biol. 23:123–137; Davis & Ares (2006) Proc. Natl Acad. Sci. USA 103:3262–3267
NEXT complex (Nuclear EXosome Targeting; ZFC3H1, RBM7, MTR4)	Targets nascent PROMPTs and lncRNAs for exosome degradation	Adapter that delivers unstable lncRNAs to the nuclear exosome. NEXT recognizes and binds a class of short, unstable lncRNAs called PROMPTs (promoter upstream transcripts) and other nuclear-retained lncRNAs, and delivers them to the nuclear exosome via the MTR4 helicase (which unwinds RNA for exosome threading). Without NEXT, these lncRNAs accumulate aberrantly in the nucleus, potentially interfering with transcription and chromatin organization.	Lubas et al. (2015) EMBO J. 34:1399–1416
PAXT complex (PolyA tail exosome targeting; ZFC3H1, PAPD5, etc.)	Targets polyadenylated lncRNAs for exosome degradation	Adapter that delivers polyadenylated lncRNAs to the nuclear exosome. PAXT specifically targets polyadenylated lncRNAs for degradation, working in parallel with NEXT (which targets non-polyadenylated lncRNAs). Together, NEXT and PAXT ensure that both polyadenylated and non-polyadenylated lncRNAs are surveilled, with PAXT serving as the quality control pathway for those lncRNAs that have received a poly(A) tail but should not be expressed.	Meola et al. (2016) EMBO J. 35:1499–1514
NMD (Nonsense-mediated decay; UPF1, UPF2, UPF3, SMG1, SMG5, SMG6, SMG7)	Degrades lncRNAs with premature termination codons; some lncRNAs are preferentially subject to NMD	Cytoplasmic quality control that degrades aberrant or regulatory lncRNAs. NMD degrades lncRNAs that contain premature termination codons (PTCs), either as a quality control mechanism for aberrant transcripts or as a regulated process for controlling the levels of specific lncRNAs. Some lncRNAs are intentionally designed with PTCs as a regulatory mechanism — they are constantly produced and degraded, creating a “transcriptional pulse” that may serve cis-regulatory functions at their locus of origin even though the RNA itself is rapidly destroyed. NMD thus shapes both the quality and quantity of cytoplasmic lncRNAs.	Quinn & Chang (2016) Nat. Rev. Genet. 17:47–62

7. lncRNA Function — Chromatin-Interacting Machinery

This is where lncRNAs exert some of their most impactful epigenetic functions. Nuclear lncRNAs interact with chromatin-modifying complexes to direct their activity to specific genomic loci.

7.1 Polycomb Repressive Complex 2 (PRC2)

PRC2 is one of the most extensively studied lncRNA-interacting complexes. lncRNAs serve as “address codes” or “guides” that recruit PRC2 to specific genomic loci for H3K27 trimethylation and gene silencing.

Protein	Gene	Function in PRC2	lncRNA Interaction	Contribution to lncRNA Pathway	Key References
EZH2	EZH2	Catalytic subunit; H3K27 methyltransferase (SET domain); adds 1–3 methyl groups to K27 on histone H3; phosphorylation at T350 (CDK1/2) enhances lncRNA binding; preferentially binds G-quadruplex RNA	Binds HOTAIR, XIST, MALAT1, ANRIL, KCNQ1OT1, and many other lncRNAs	The catalytic engine of lncRNA-directed epigenetic silencing. EZH2 is the enzymatic subunit that PRC2 uses to deposit H3K27me3, the repressive histone mark. lncRNAs serve as “address codes” that direct EZH2/PRC2 to specific genomic loci — without lncRNA guidance, PRC2 lacks sequence-specific DNA binding and cannot achieve locus-specific silencing. HOTAIR recruits EZH2 to the HOXD locus; XIST recruits PRC2 to the inactive X chromosome; ANRIL recruits PRC2 to the CDKN2B locus. Critically, lncRNA binding can both recruit PRC2 to chromatin AND modulate its catalytic activity — some lncRNAs inhibit EZH2 methyltransferase activity until JARID2 relieves this inhibition, adding another layer of regulation. Phosphorylation of EZH2 at T350 enhances its lncRNA-binding capacity, linking cell cycle signaling to lncRNA-directed epigenetic silencing.	Zhao et al. (2008) Science 322:750–756; Kaneko et al. (2010) Genes Dev. 24:2015–2025; Davidovich & Cech (2015) RNA 21:2007–2022
EZH1	EZH1	Alternative catalytic subunit of PRC2; predominates in post-mitotic tissues; forms PRC2-EZH1 variant; MALAT1 interaction stabilizes PRC2-EZH1 on chromatin under oxidative stress	MALAT1 binds EED to promote PRC2-EZH1 assembly	Oxidative stress-responsive PRC2 variant directed by MALAT1. EZH1 replaces EZH2 as the catalytic subunit of PRC2 in post-mitotic and differentiated tissues. Under oxidative stress, MALAT1 lncRNA binds EED to promote shuttling of EED into the nucleus, facilitating assembly of the PRC2-EZH1 variant. This enables stress-responsive epigenetic reprogramming through H3K27me3 redistribution. In the context of coral biology, this PRC2-EZH1/MALAT1 mechanism may be particularly relevant for understanding how oxidative stress from thermal bleaching triggers epigenetic changes.	El Said et al. (2021) Cell Death Dis. 12:850
EED	EED	Core PRC2 subunit; WD40 repeat protein; recognizes and binds H3K27me3 (allosteric activation of PRC2); binds MALAT1 under oxidative stress	MALAT1	Allosteric activator of PRC2 that links lncRNA binding to catalytic stimulation. EED recognizes existing H3K27me3 marks and allosterically stimulates EZH2/EZH1 methyltransferase activity, creating a positive feedback loop for epigenetic spreading. EED also directly binds MALAT1 under oxidative stress, serving as the RNA-binding subunit that triggers PRC2-EZH1 nuclear import and assembly. EED thus plays a dual role: maintaining epigenetic memory through H3K27me3 recognition, and responding to lncRNA signals for de novo recruitment.	El Said et al. (2021) Cell Death Dis. 12:850
SUZ12	SUZ12	Core PRC2 subunit; zinc finger domain with RNA-binding role; essential for PRC2 structural integrity and histone methyltransferase activity	Binds lncRNAs; role in RNA binding demonstrated	Structural scaffold and RNA-binding subunit of PRC2. SUZ12 is essential for PRC2 structural integrity — without SUZ12, EZH2 is unstable and PRC2 cannot form. SUZ12’s zinc finger domain contributes to RNA binding, providing an additional lncRNA contact point beyond EZH2. SUZ12 thus serves as both the structural backbone of PRC2 and a participant in lncRNA-directed targeting, ensuring that the complex remains intact while being guided to specific loci.	Kanhere et al. (2010) Mol. Cell 38:675–687; Cifuentes-Rojas et al. (2014) RNA 20:598–611
RBBP4 (RbAp48)	RBBP4	PRC2 core subunit; WD40 repeat histone-binding protein; interacts with histone H3 tails	Part of PRC2-lncRNA complexes	Histone-binding subunit that anchors PRC2 to nucleosomes. RBBP4 directly binds histone H3 tails on nucleosomes, anchoring the PRC2 complex to chromatin. This histone contact is essential for PRC2 to engage its substrate (the H3K27 residue) for methylation. RBBP4 ensures that lncRNA-guided PRC2 is properly positioned on the nucleosome for catalytic activity.	Cao & Zhang (2004) Mol. Cell 15:57–67
AEBP2	AEBP2	PRC2 accessory subunit; zinc finger protein; stimulates PRC2 activity	Part of PRC2 holoenzyme	Activity-stimulating accessory factor for PRC2. AEBP2 enhances the methyltransferase activity of PRC2, particularly for the transition from mono- to tri-methylation of H3K27. AEBP2 is part of the PRC2 holoenzyme that is guided by lncRNAs to specific loci.	Cao & Zhang (2004) Curr. Biol. 14:331–336
JARID2	JARID2	PRC2 accessory factor; reduces PRC2-RNA interaction to promote PRC2-chromatin binding; relieves lncRNA-mediated inhibition of EZH2 catalytic activity	Modulates PRC2-lncRNA interactions	Switch factor that transitions PRC2 from RNA-bound to chromatin-bound state. JARID2 plays a sophisticated regulatory role: many lncRNAs bind PRC2 but inhibit EZH2 catalytic activity (acting as a “brake”). JARID2 reduces PRC2-RNA interaction, promoting transfer of PRC2 from lncRNA-bound to chromatin-bound state, and simultaneously relieving the lncRNA-mediated inhibition of EZH2 catalytic activity. This “handoff” mechanism — from lncRNA-guided (inactive) to chromatin-bound (active) — ensures that PRC2 deposits H3K27me3 only when properly positioned on target chromatin, preventing promiscuous silencing.	Hao et al. (2021) Int. J. Biol. Sci. 17:3268–3280
PHF1	PHF1	PRC2 accessory subunit; stimulates H3K27me3 activity	Part of PRC2-lncRNA complexes	Stimulatory accessory factor that enhances PRC2 processivity. PHF1 stimulates the processivity of PRC2, promoting the transition from H3K27me1 to H3K27me3. PHF1 is part of the PRC2 holoenzyme that can be directed by lncRNAs to specific loci.	Standard

7.2 SWI/SNF (BAF) Chromatin Remodeling Complex

lncRNAs direct the SWI/SNF complex to cell type-specific enhancers, controlling chromatin accessibility.

Protein	Function	lncRNA Interaction	Contribution to lncRNA Pathway	Key References
BRG1/SMARCA4	ATP-dependent catalytic subunit of SWI/SNF; remodels nucleosomes to increase DNA accessibility	lncRNAs preferentially bind SWI/SNF and direct it to enhancers	The ATP-driven motor of lncRNA-directed chromatin opening. BRG1 is the catalytic engine that uses ATP hydrolysis to slide, evict, or restructure nucleosomes, opening chromatin for transcription factor binding and gene activation. lncRNAs preferentially bind SWI/SNF and direct BRG1 to cell type-specific enhancers, providing the sequence specificity that SWI/SNF lacks on its own. This lncRNA-SWI/SNF-enahancer axis activates gene expression programs — in contrast to the lncRNA-PRC2 axis which silences them — making lncRNA-directed SWI/SNF the counterpart activation pathway to PRC2-mediated repression.	Oo et al. (2025) Nat. Commun. 16:131
BRM/SMARCA2	Alternative ATPase subunit of SWI/SNF	Part of lncRNA-directed SWI/SNF	Alternative catalytic subunit for lncRNA-directed remodeling. BRM can replace BRG1 as the ATPase subunit in certain SWI/SNF complexes, providing functional diversity in lncRNA-directed chromatin remodeling.	Oo et al. (2025) Nat. Commun. 16:131
BAF155/SMARCC1, BAF170/SMARCC2	Core SWI/SNF subunits; scaffolding	Part of lncRNA-directed SWI/SNF enhancer hub	Structural scaffolds of the lncRNA-SWI/SNF enhancer hub. BAF155 and BAF170 are core scaffolding subunits that maintain SWI/SNF complex integrity and mediate protein-protein interactions within the complex. They are part of the “enhancer hub” model where lncRNAs competitively recruit SWI/SNF to specific enhancers.	Oo et al. (2025) Nat. Commun. 16:131
BAF47/INI1/SMARCB1	Core SWI/SNF subunit; tumor suppressor	Part of SWI/SNF	Tumor suppressor subunit essential for SWI/SNF function. BAF47 is required for the full chromatin remodeling activity of SWI/SNF and is a potent tumor suppressor. Its loss (as in malignant rhabdoid tumors) abrogates SWI/SNF activity and may disrupt lncRNA-directed gene activation programs.	Standard

7.3 Histone Modification Complexes (Beyond PRC2)

Protein/Complex	Gene(s)	Function	lncRNA Interaction	Contribution to lncRNA Pathway	Key References
LSD1/KDM1A	KDM1A	Histone H3K4me1/2 demethylase; transcriptional repression; part of CoREST complex	HOTAIR binds both PRC2 (5′ domain) and LSD1 (3′ domain) to coordinate H3K27me3 and H3K4 demethylation for gene silencing	Dual-modification silencing effector recruited by HOTAIR to erase active marks. LSD1 removes activating H3K4me1/2 marks, working in concert with PRC2’s deposition of repressive H3K27me3. HOTAIR is a modular scaffold — its 5′ domain recruits PRC2 (for H3K27me3 silencing) and its 3′ domain recruits LSD1/CoREST (for H3K4 demethylation). This dual recruitment achieves a “double hit” on target genes: simultaneous removal of the active mark (H3K4me) and deposition of the repressive mark (H3K27me3), ensuring robust and stable epigenetic silencing. This demonstrates that a single lncRNA can coordinate multiple chromatin-modifying complexes for synergistic gene regulation.	Tsai et al. (2010) Science 329:689–693
CoREST complex (RCOR1, LSD1, HDAC1/2)	Multiple	Histone demethylation and deacetylation	HOTAIR bridges PRC2 and LSD1/CoREST	Tripartite silencing machine that erases active marks and removes acetylation. The CoREST complex (LSD1 + RCOR1 + HDAC1/2) couples H3K4 demethylation with histone deacetylation, providing a triple silencing effect when recruited by HOTAIR: (1) H3K4 demethylation by LSD1, (2) histone deacetylation by HDAC1/2, and (3) H3K27 trimethylation by PRC2. This demonstrates the extraordinary combinatorial power of lncRNA scaffolding — one lncRNA can orchestrate multiple layers of chromatin modification simultaneously.	Tsai et al. (2010) Science 329:689–693
MLL1/KMT2A	KMT2A	H3K4 methyltransferase; transcriptional activation	HOTTIP lncRNA recruits WDR5/MLL complex to activate HOXA genes	Gene activation engine directed by HOTTIP lncRNA. MLL1 deposits activating H3K4me3 marks at gene promoters. HOTTIP lncRNA binds WDR5 to recruit the MLL1 complex to specific HOXA loci, switching on gene expression — the functional opposite of HOTAIR-PRC2-mediated silencing. This demonstrates that lncRNAs direct both activating and repressing chromatin machinery, and the same histone modification (H3K4 methylation) can be targeted by lncRNAs for either activation (via MLL1) or removal (via LSD1).	Wang et al. (2011) Nature 467:340–344
WDR5	WDR5	Core component of MLL/COMPASS complex; H3K4 methylation	HOTTIP binds WDR5 to target MLL to HOXA loci	The lncRNA-docking subunit of the MLL activation complex. WDR5 is the direct binding target of HOTTIP lncRNA — WDR5’s WD40 repeat domain binds a specific motif in HOTTIP, and this interaction is essential for recruiting the entire MLL1 complex to target loci. WDR5 thus serves as the “receptor” that translates lncRNA localization information into chromatin modification (H3K4me3 deposition), making it a critical interface between lncRNA signals and epigenetic activation.	Wang et al. (2011) Nature 467:340–344
MOF/KAT8	KAT8	H4K16 acetyltransferase (MYST family); part of MSL complex	XIST and other lncRNAs interact with dosage compensation complexes	Histone acetyltransferase for lncRNA-directed chromatin opening. MOF deposits the activating H4K16ac mark, which opens chromatin by neutralizing the charge interaction between histone H4 and DNA. lncRNA-directed MOF recruitment (e.g., in dosage compensation complexes guided by XIST-related RNAs) provides a mechanism for locus-specific chromatin opening through acetylation.	Standard
DNMT1/3A/3B	DNMT1, DNMT3A, DNMT3B	DNA methyltransferases; CpG methylation	CDKN2B-AS1 lncRNA forms RNA-DNA triplexes to recruit DNMTs; lncRNAs guide de novo DNA methylation	DNA methylation machinery recruited by lncRNAs for stable gene silencing. lncRNAs recruit DNMTs to specific CpG islands to deposit DNA methylation, a highly stable epigenetic mark that persists through cell division. CDKN2B-AS1 forms RNA-DNA triplex structures at target gene promoters to recruit DNMTs, while other lncRNAs use protein intermediaries (e.g., PRC2-DNMT connections). DNA methylation directed by lncRNAs provides a more durable silencing mechanism than histone modifications alone, and may be particularly relevant for transgenerational epigenetic inheritance in corals.	Liu et al. (2025) Biomolecules 15:881
HDAC1/2/3	HDAC1, HDAC2, HDAC3	Histone deacetylases; remove acetyl marks for chromatin compaction	Various lncRNAs recruit HDACs to specific loci	Deacetylation machinery for lncRNA-directed chromatin compaction. HDACs remove activating acetyl marks (e.g., H3K27ac, H4K16ac), compacting chromatin and repressing transcription. lncRNA-directed HDAC recruitment (e.g., through the CoREST complex, or independently) provides a mechanism for reversing gene activation. Combined with PRC2-mediated H3K27me3 deposition, HDAC-mediated deacetylation achieves multi-layered epigenetic silencing.	Standard

7.4 Nuclear Organization Proteins

Protein	Gene	Function	lncRNA Interaction	Contribution to lncRNA Pathway	Key References
CTCF	CTCF	CCCTC-binding factor; insulator protein; 3D genome organizer	lncRNAs cooperate with CTCF to mediate genome topology	3D genome organizer that cooperates with lncRNAs for locus-specific chromatin architecture. CTCF defines topologically associating domain (TAD) boundaries and insulator elements, creating 3D chromatin loops that regulate enhancer-promoter interactions. lncRNAs cooperate with CTCF to mediate specific genome topology — for example, by stabilizing CTCF binding at specific loci or by bridging CTCF to cohesin. This lncRNA-CTCF partnership can dynamically reshape 3D genome architecture in response to cellular signals, controlling which genes are in contact with which enhancers.	Spandidos-Publications (2025) Int. J. Oncol. DOI:10.3892/ijo.2025.5735
Cohesin (SMC1A, SMC3, RAD21, SA1/SA2)	Multiple	3D genome architecture; loops; facilitates enhancer-promoter contacts	NUP153 modulates gene activation by coordinating CTCF and cohesin with lncRNAs	Loop extrusion machinery that works with lncRNAs to shape enhancer-promoter contacts. Cohesin extrudes chromatin loops that bring enhancers into contact with their target promoters. lncRNAs influence this process by guiding cohesin loading sites, stabilizing loop structures, or recruiting cohesin via CTCF. The lncRNA-Cohesin-CTCF axis is a major mechanism for dynamic 3D genome reorganization in response to signaling, and its misregulation disrupts gene expression programs.	Oo et al. (2025) Nat. Commun. 16:131
NONO/SFPQ (Paraspeckle proteins)	NONO, SFPQ	Nuclear paraspeckle assembly; bind NEAT1 lncRNA scaffold	Essential for paraspeckle biogenesis; bind NEAT1	Core structural proteins of NEAT1-dependent paraspeckles. NONO and SFPQ bind directly to NEAT1_2 lncRNA and are essential for the formation of paraspeckles — nuclear bodies that sequester specific proteins and RNAs to regulate their availability. Paraspeckles function as “molecular sponges” — by sequestering transcription factors, RNA-editing enzymes (ADAR1), and mRNAs, they indirectly regulate gene expression. NEAT1_2 is the essential architectural scaffold; without NONO/SFPQ binding, paraspeckles cannot form and their sequestered molecules are released into the nucleoplasm, altering gene expression.	Plant et al. (2020) Sci. Adv. 6:eaaz9072
PSPC1, PSF	PSPC1, SFPQ	Paraspeckle proteins	Bind NEAT1 for paraspeckle integrity	Additional structural proteins for paraspeckle integrity and function. PSPC1 is a paraspeckle component that binds NEAT1 and contributes to the structural integrity of these nuclear bodies. PSF (SFPQ) serves dual roles in both paraspeckle structure and in DNA repair/RNA processing, linking paraspeckle dynamics to genome maintenance.	Plant et al. (2020) Sci. Adv. 6:eaaz9072

8. lncRNA Function — Post-Transcriptional & Cytoplasmic

8.1 lncRNA-miRNA Interactions (ceRNA / Sponge Mechanism)

lncRNAs can act as competing endogenous RNAs (ceRNAs) or “sponges” that sequester miRNAs away from their mRNA targets. The proteins involved are the same RISC machinery described in Section 4 (AGO, TNRC6, etc.), but the lncRNA provides additional binding sites.

Protein	Function in lncRNA-miRNA crosstalk	Contribution to lncRNA Pathway	Key References
AGO1–4	miRNA-loaded AGO binds lncRNA ceRNA/sponge via miRNA seed pairing	Effector that executes the ceRNA/sponge interaction between lncRNAs and miRNAs. AGO proteins mediate the physical interaction between miRNAs and lncRNA ceRNAs — the miRNA guide strand, loaded in AGO, base-pairs with complementary sites in the lncRNA just as it would with an mRNA target. When a lncRNA contains multiple miRNA binding sites, it can “soak up” multiple AGO-miRNA complexes, effectively depleting the pool of free miRNA available to repress its true mRNA targets. This ceRNA mechanism allows lncRNAs to indirectly upregulate genes that would otherwise be silenced by miRNAs, creating a post-transcriptional regulatory network.	Singh et al. (2025) Mol. Biol. Rep. DOI:10.1007/s11033-025-10906-4
TNRC6A/B/C	Scaffold for miRISC recruited to lncRNA targets	Scaffolding that enables miRNA-mediated silencing on lncRNA ceRNA targets. TNRC6 proteins are recruited when AGO-miRNA binds lncRNA ceRNA sites, forming functional miRISC on the lncRNA. This means that the lncRNA ceRNA not only sequesters the miRNA but also engages the full silencing machinery (deadenylases, decapping enzymes), potentially leading to lncRNA degradation. The balance between miRNA sequestration (ceRNA function) and miRNA-directed lncRNA degradation determines whether the lncRNA acts as a sponge or as a miRNA target.	Standard

8.2 lncRNA-Protein Interactions in Cytoplasm

Protein	Gene	Function	Contribution to lncRNA Pathway	Key References
YBX1	YBX1	RNA-binding protein; interacts with lncRNAs to regulate mRNA stability and translation	lncRNA cofactor that links cytoplasmic lncRNAs to mRNA fate regulation. YBX1 interacts with cytoplasmic lncRNAs and serves as an effector for lncRNA-mediated regulation of mRNA stability and translation. lncRNAs can guide YBX1 to specific mRNA targets or modulate its RNA-binding activity, providing a mechanism for sequence-specific mRNA regulation by lncRNAs in the cytoplasm.	Liu et al. (2025) Biomolecules 15:881
HNRNPK	HNRNPK	Heterogeneous nuclear ribonucleoprotein K; scaffolding for lncRNA-mediated regulation of metabolic enzymes (e.g., IDH2)	Scaffolding RBP that assembles lncRNA-metabolic enzyme complexes. HNRNPK mediates the interaction between lncRNAs and metabolic enzymes, enabling lncRNA-directed regulation of cellular metabolism. For example, LINC00571 uses HNRNPK as a scaffold to assemble a complex with ILF2 and IDH2, regulating TCA cycle metabolism. This demonstrates a cytoplasmic lncRNA mechanism where the lncRNA scaffolds protein complexes to regulate enzymatic reactions, extending lncRNA function beyond gene expression regulation into direct metabolic control.	Xi et al. (2024) J. Exp. Clin. Cancer Res. 43:22
HNRNPA2B1	HNRNPA2B1	RBP; interacts with DGCR8; mediates m6A-dependent pri-miRNA processing; also binds lncRNAs	Dual-function RBP at the intersection of m6A modification, miRNA biogenesis, and lncRNA function. HNRNPA2B1 recognizes m6A-modified RNAs and mediates m6A-dependent pri-miRNA processing by recruiting DGCR8, directly linking RNA modification to miRNA biogenesis. It also binds lncRNAs, potentially recruiting them to the Microprocessor or other processing machinery. This makes HNRNPA2B1 a key integrator of epitranscriptomic (m6A) modification with both miRNA and lncRNA regulation.	Alarcón et al. (2015) Nat. Commun. 6:7216
E3 ubiquitin ligases (DZIP3, MEX3B, TRIM25)	Multiple	lncRNAs scaffold E3 ligase-substrate interactions to promote ubiquitination and degradation of target proteins (e.g., HOTAIR-DZIP3-MEX3B; LINC00955-TRIM25-Sp1)	lncRNA-scaffolded protein ubiquitination machinery for targeted protein degradation. lncRNAs can function as molecular scaffolds that physically bring E3 ubiquitin ligases into proximity with their substrate proteins, dramatically enhancing ubiquitination and proteasomal degradation. HOTAIR scaffolds DZIP3 and MEX3B to promote degradation of their targets; LINC00955 scaffolds TRIM25 and Sp1 to enhance Sp1 ubiquitination. This represents a fundamentally different lncRNA mechanism — not regulating gene expression at the RNA level, but directly controlling protein stability through the ubiquitin-proteasome system. This “scaffold” function of lncRNAs in post-translational modification is an emerging area with broad implications for cellular regulation.	Yoon et al. (2013) Nat. Commun. 4:2939; Shin et al. (2025) Exp. Mol. Med. 57:1641–1650
BRD3/BRD4	BRD3, BRD4	Bromodomain proteins; lncRNA DIGIT and BRD3 form phase-separated condensates	Bromodomain proteins that form lncRNA-driven phase-separated condensates for transcriptional activation. BRD3 and BRD4 are readers of acetylated histones that promote transcription. The lncRNA DIGIT binds BRD3 to form phase-separated condensates at specific genomic loci, creating concentrated zones of transcriptional activity. This demonstrates that lncRNAs can organize 3D nuclear compartments through liquid-liquid phase separation (LLPS), bringing together transcription factors and chromatin regulators in dynamic, membraneless organelles that activate gene expression programs.	Daneshvar et al. (2020) eLife 9:e55258

9. Shared Machinery Between miRNA and lncRNA Pathways

Protein/Complex	Role in miRNA Pathway	Role in lncRNA Pathway	Contribution to Both Pathways
DROSHA	Cleaves pri-miRNAs	Cleaves some lncRNA hairpins (e.g., DGCR8 mRNA, some lncRNA-derived miRNA precursors); noncanonical processing of lncRNAs	Dual-function endonuclease at the apex of both miRNA and lncRNA processing. DROSHA’s primary role is cleaving pri-miRNAs, but it also processes certain lncRNA hairpin structures and regulates lncRNA levels through noncanonical cleavage. The Microprocessor can cleave hairpin structures within lncRNAs to produce miRNAs (as in miRNA-host lncRNAs) or to degrade the lncRNA itself. This creates a regulatory intersection: lncRNA levels can be controlled by DROSHA, and some lncRNAs are both miRNA precursors and functional lncRNA molecules.
DICER	Cleaves pre-miRNAs	Some lncRNA-derived small RNAs are Dicer-dependent; Dicer processes some lncRNA substrates	Shared processing enzyme for small RNA production from both miRNA and lncRNA precursors. DICER processes pre-miRNAs into mature miRNA duplexes, but also processes certain lncRNA-derived hairpins that produce small RNAs. Some lncRNAs contain hairpin structures that are Dicer substrates, generating small RNAs with regulatory functions. This shared processing creates functional overlap between the miRNA and lncRNA pathways.
DGCR8	Cofactor for DROSHA	Recognizes lncRNA secondary structures for Microprocessor; DGCR8 mRNA is itself a lncRNA-like substrate	Junction protein that bridges miRNA processing and lncRNA regulation via the Microprocessor feedback loop. DGCR8 is essential for DROSHA activity in miRNA processing, but DGCR8 mRNA is itself a substrate for DROSHA cleavage (an auto-regulatory feedback loop). Since DGCR8 mRNA has lncRNA-like features, this feedback loop sits at the intersection of miRNA and lncRNA regulation.
RNA Pol II	Transcribes pri-miRNAs and miRNA host genes	Transcribes lncRNA genes	The universal transcription engine for both pri-miRNAs and lncRNAs. Pol II transcribes both pri-miRNA genes (including miRNAs embedded in lncRNA host genes) and independent lncRNA genes. Distinct Pol II CTD phosphorylation patterns differentiate the transcription of miRNA host genes vs. lncRNA genes vs. mRNA genes, coupling transcription kinetics to downstream RNA processing fate.
Spliceosome	Processes intronic miRNAs (mirtrons bypass this)	Processes lncRNA introns	Shared splicing machinery with divergent efficiency for miRNA hosts vs. lncRNAs. The spliceosome processes both miRNA host gene introns (which may contain miRNAs) and lncRNA introns. However, lncRNAs are typically spliced far less efficiently than mRNAs, which has consequences for both pathways: retained introns in lncRNAs contribute to nuclear retention, and mirtrons (intronic miRNAs that bypass Drosha) depend on the spliceosome for their biogenesis.
Exportin-5	Exports pre-miRNA	Exports some lncRNA intermediates	Shared nuclear export receptor with distinct substrate preferences. XPO5 primarily exports pre-miRNAs but can also export some structured lncRNA intermediates. The 2-nt 3′ overhang and stem length requirements for XPO5 recognition overlap with features found in certain lncRNA structures.
AGO proteins	Core of miRISC	Bind some lncRNAs; lncRNAs can be AGO targets (cleaved or repressed)	Shared effector proteins with opposing consequences for miRNA vs. lncRNA function. AGO proteins are the core of miRISC (executing miRNA-mediated silencing), but they also bind lncRNAs as targets — lncRNAs with miRNA complementarity can be cleaved or repressed by AGO. Additionally, in the ceRNA mechanism, AGO-miRNA binding to lncRNA sponges depletes miRNA from mRNA targets, effectively converting AGO from a silencer to a vehicle for lncRNA-mediated gene activation.
KSRP	Promotes Drosha and Dicer processing	Regulates RNA stability including lncRNA	Biphasic processing activator for miRNAs with emerging roles in lncRNA stability. KSRP promotes both nuclear and cytoplasmic miRNA processing steps, and also regulates the stability of select lncRNAs, suggesting broader roles in RNA metabolism that span both pathways.
LIN28	Blocks let-7 processing	Interacts with lncRNAs in pluripotency networks	Developmental switch protein that integrates let-7 miRNA suppression with lncRNA pluripotency networks. LIN28 specifically blocks let-7 miRNA biogenesis, but also interacts with lncRNAs in pluripotency networks, suggesting it may coordinate miRNA and lncRNA regulation during the stem cell-to-differentiated cell transition.
hnRNP proteins	Regulate miRNA processing	Bind and regulate lncRNA splicing, stability, and function	Broad-spectrum RNA regulators that operate across both miRNA and lncRNA pathways. hnRNP proteins (A1, A2/B1, K, etc.) regulate miRNA processing (e.g., hnRNP A1 promotes pri-miR-18a processing; hnRNP A2/B1 interacts with DGCR8) and also regulate lncRNA splicing, stability, and function. Their broad RNA-binding activity makes them common mediators of cross-talk between the two pathways.
ADAR	Edits miRNA precursors	Edits lncRNA transcripts (e.g., in nuclear paraspeckles)	Shared RNA editing enzyme that diversifies both miRNAs and lncRNAs. ADAR edits adenosine to inosine in both miRNA precursors (altering processing efficiency or target specificity) and lncRNAs (e.g., creating inosine-containing NEAT1 that is recognized by paraspeckle proteins). This shared editing creates regulatory diversity in both pathways.
Nuclear exosome	Degrades aberrant miRNA intermediates	Primary degradation pathway for many lncRNAs	Shared quality control and regulatory degradation machinery. The nuclear exosome degrades aberrant miRNA processing intermediates and is the primary degradation pathway for most nascent lncRNAs. Exosome activity thus determines steady-state levels of both miRNA precursors and lncRNAs, making it a master regulator of non-coding RNA abundance.
CPSF/CstF	3′ processing of pri-miRNAs in some contexts	3′ processing of polyadenylated lncRNAs	Shared 3′ end processing machinery that defines the 3′ termini of both pri-miRNAs and lncRNAs. CPSF/CstF process the 3′ ends of polyadenylated lncRNAs and also contribute to 3′ processing of some pri-miRNAs. Alternative polyadenylation choice (mediated by CPSF/CstF/CFIm) determines both lncRNA isoform identity and pri-miRNA processing efficiency.

10. FINAL LIST

These are the genes I will need to obtain FASTAs for, to use in a BLAST search on our coral genomes. Note that I chose to exclude genes identified in 6.1, which play core transcription roles relevant for mRNA as well, and would thus be unable to confidently ascribe to ncRNA regulation.

Also, note a few manual changes from the agentic summary above, to account for coral miRNAs’ plant-like features:

Coral-Specific Dicer Partners (replacing TRBP/PACT)

Gene	Protein	Relevance
HYL1La	HYL1-like a	Plant-like Dicer partner; essential in corals; NO bilaterian equivalent
HYL1Lb	HYL1-like b	Second HYL1 paralog; present in corals

🪸 Coral-Specific miRNA Methylation (absent from bilaterian miRNA pathway)

Gene	Protein	Relevance
HEN1	HEN1 methyltransferase	2′-O-methylates miRNA 3′ ends in corals; essential for miRNA stability; plant-like feature

🪸 Coral-Specific: RdRP Amplification

Gene	Protein	Relevance
RdRP (multiple)	RNA-dependent RNA polymerase	*8 candidates in S. pistillata; enables transitive RNAi — ABSENT from most bilaterians; plant/C. elegans-like*

Protein	Gene
DROSHA	DROSHA (formerly RNASEN)
DGCR8	DGCR8 (DiGeorge syndrome critical region 8; aka Pasha in Drosophila and C. elegans)
Exportin-5 (XPO5)	XPO5
RanGTP	RAN
DICER	DICER1
TRBP (TAR RNA-binding protein)	TARBP2
PACT (Protein activator of PKR)	PRKRA
Loqs-PB	loquacious (in Drosophila)
AGO1	AGO1
AGO2	AGO2
AGO3	AGO3
AGO4	AGO4
TNRC6A (GW182)	TNRC6A
TNRC6B	TNRC6B
TNRC6C	TNRC6C
CCR4-NOT complex	CNOT1–CNOT11, CNOT7 (CAF1), CNOT6 (CCR4)
PAN2–PAN3 complex	PAN2, PAN3
PABP (Poly(A)-binding protein)	PABPC1
PARN (Poly(A)-specific ribonuclease)	PARN
DCP1–DCP2	DCP1A/B, DCP2
EDC3, EDC4 (Hedls)	EDC3, EDC4
XRN1	XRN1
Hsc70/Hsp90 chaperone machinery	HSPA8 (Hsc70), HSP90AA1/HSP90AB1
DDX6	DDX6 (RCK/p54)
DDX5 (p68)	DDX5
DDX17 (p72)	DDX17
hnRNP A1	HNRNPA1
KSRP (KH-type splicing regulatory protein)	KHSRP
LIN28A / LIN28B	LIN28A, LIN28B
TDP-43	TARDBP
BRCA1	BRCA1
EWSR1	EWSR1
FUS	FUS
TUT4/ZCCHC11	ZCCHC11
TUT7/ZCCHC6	ZCCHC6
DIS3L2	DIS3L2
ADAR1/ADAR2	ADAR1, ADAR2
MCPIP1/ZC3H12A	ZC3H12A
RBM3	RBM3
YB-1	YBX1
DAZL	DAZL
Nibbler	Nibbler (fly) / PNPT1 (mammalian homolog)
Integrator complex (INTS1–INTS14)	INTS1–INTS14
RNase P	RPPH1 and multiple subunits
snoRNP complexes (snoRNA + fibrillarin, NOP56, NOP58, etc.)	Multiple
NXF1/TAP	NXF1
CRM1/Exportin-1	XPO1
PHAX	PHAX
CBC (Cap-binding complex: CBP80/CBP20)	NCBP1, NCBP2
ARS2 (Arsenite resistance protein 2)	SRRT
Nuclear exosome (EXOSC1–EXOSC10, DIS3, DIS3L)	EXOSC1–EXOSC10, DIS3, DIS3L
NEXT complex (Nuclear EXosome Targeting; ZFC3H1, RBM7, MTR4)	ZFC3H1, RBM7, MTR4
PAXT complex (PolyA tail exosome targeting; ZFC3H1, PAPD5, etc.)	ZFC3H1, PAPD5
NMD (Nonsense-mediated decay; UPF1, UPF2, UPF3, SMG1, SMG5, SMG6, SMG7)	UPF1, UPF2, UPF3, SMG1, SMG5, SMG6, SMG7
EZH2	EZH2
EZH1	EZH1
EED	EED
SUZ12	SUZ12
RBBP4 (RbAp48)	RBBP4
AEBP2	AEBP2
JARID2	JARID2
PHF1	PHF1
BRG1/SMARCA4	BRG1/SMARCA4
BRM/SMARCA2	BRM/SMARCA2
BAF155/SMARCC1, BAF170/SMARCC2	BAF155/SMARCC1, BAF170/SMARCC2
BAF47/INI1/SMARCB1	BAF47/INI1/SMARCB1
LSD1/KDM1A	LSD1/KDM1A
CoREST complex (RCOR1, LSD1, HDAC1/2)	RCOR1, LSD1, HDAC1/2
MLL1/KMT2A	MLL1/KMT2A
WDR5	WDR5
MOF/KAT8	MOF/KAT8
DNMT1/3A/3B	DNMT1/3A/3B
HDAC1/2/3	HDAC1/2/3
CTCF	CTCF
Cohesin (SMC1A, SMC3, RAD21, SA1/SA2)	SMC1A, SMC3, RAD21, SA1/SA2
NONO/SFPQ (Paraspeckle proteins)	NONO, SFPQ
PSPC1, PSF	PSPC1, SFPQ
AGO1–4	AGO1–4
TNRC6A/B/C	TNRC6A/B/C
YBX1	YBX1
HNRNPK	HNRNPK
HNRNPA2B1	HNRNPA2B1
E3 ubiquitin ligases (DZIP3, MEX3B, TRIM25)	DZIP3, MEX3B, TRIM25
BRD3/BRD4	BRD3, BRD4

11. References

Primary Reviews Cited

Kim H, Lee YY, Kim VN. (2024) The biogenesis and regulation of animal microRNAs. Nature Reviews Molecular Cell Biology. https://doi.org/10.1038/s41580-024-00805-0
Ha M, Kim VN. (2014) Regulation of microRNA biogenesis. Nature Reviews Molecular Cell Biology 15:509–524.
Michlewski G, Cáceres JF. (2019) Post-transcriptional control of miRNA biogenesis. RNA 25:1–16.
Hynes C, Kakumani PK. (2024) Regulatory role of RNA-binding proteins in microRNA biogenesis. Frontiers in Molecular Biosciences 11:1374843. https://doi.org/10.3389/fmolb.2024.1374843
Finnegan EF, Pasquinelli AE. (2013) MicroRNA biogenesis: regulating the regulators. Critical Reviews in Biochemistry and Molecular Biology 48:51–68.
Newman MA, Hammond SM. (2010) Emerging paradigms of regulated microRNA processing. Genes & Development 24:1086–1092.
Narasimhan S et al. (2025) Partners in silencing: decoding the mammalian Argonaute interactome. Noncoding RNA 11(4):62. https://doi.org/10.3390/ncrna11040062
Mauro M et al. (2023) The multiplicity of Argonaute complexes in mammalian cells. J Pharmacol Exp Ther 384:1–9.
Nakanishi K. (2016) Anatomy of RISC: how do small RNAs and chaperones activate Argonaute proteins? Wiley Interdiscip. Rev. RNA 7:637–660.
Chen CY et al. (2009) Ago–TNRC6 triggers microRNA-mediated decay by promoting two deadenylation steps. Nature Structural & Molecular Biology 16:1208–1214.
Chong MMW, Leong KW. (2025) Regulation of the microprocessor by post-translational modifications. Frontiers in Cell and Developmental Biology 13:1721087. https://doi.org/10.3389/fcell.2025.1721087
Quinn JJ, Chang HY. (2016) Unique features of long non-coding RNA biogenesis and function. Nature Reviews Genetics 17:47–62.
Nojima T, Proudfoot NJ. (2022) Mechanisms of lncRNA biogenesis as revealed by nascent transcriptomics. Nature Reviews Molecular Cell Biology 23:123–137.
Ferrer J, Dimitrova N. (2024) Transcription regulation by long non-coding RNAs: mechanisms and disease relevance. Nature Reviews Molecular Cell Biology 25:396–415. https://doi.org/10.1038/s41580-023-00694-9
Davidovich C, Cech TR. (2015) The recruitment of chromatin modifiers by long noncoding RNAs: lessons from PRC2. RNA 21:2007–2022.
Wozniak M, Czyz M. (2023) lncRNAs-EZH2 interaction as promising therapeutic target in cutaneous melanoma. Frontiers in Molecular Biosciences 10:1170026.
Hao A et al. (2021) Emerging roles of lncRNAs in the EZH2-regulated oncogenic network. International Journal of Biological Sciences 17:3268–3280.
Oo JA, Warwick T, Pálfi K et al. (2025) Long non-coding RNAs direct the SWI/SNF complex to cell type-specific enhancers. Nature Communications 16:131. https://doi.org/10.1038/s41467-024-55539-6
Liu K et al. (2025) LncRNA–protein interactions: a key to deciphering lncRNA mechanisms. Biomolecules 15:881. https://doi.org/10.3390/biom15060881
Shin CH et al. (2025) Long noncoding RNAs regulating enzymatic reactions in cancer. Experimental & Molecular Medicine 57:1641–1650. https://doi.org/10.1038/s12276-025-01464-7
Singh M et al. (2025) Emerging landscape of lncRNA-miRNA interactions as architects of gene expression patterns. Molecular Biology Reports. https://doi.org/10.1007/s11033-025-10906-4
Wu H et al. (2017) The diversity of long noncoding RNAs and their generation. Trends in Genetics 33:55–66.
Pfleiderer MM, Moresco JJ. (2022) Emerging insights into the function and structure of the Integrator complex. Transcription 12:251–265.
Plant N et al. (2020) Integrator restrains paraspeckles assembly by promoting isoform switching of the lncRNA NEAT1. Science Advances 6:eaaz9072.
Matera AG, Wang Z. (2014) A day in the life of the spliceosome. Nature Reviews Molecular Cell Biology 15:108–121.

Key Primary Research Articles

Lee Y et al. (2003) The nuclear RNase III Drosha initiates microRNA processing. Nature 425:415–419.
Gregory RI et al. (2004) The Microprocessor complex mediates the genesis of microRNAs. Nature 432:235–240.
Han J et al. (2004) The Drosha-DGCR8 complex in primary miRNA processing. Genes & Development 18:3016–3027.
Nguyen TA et al. (2015) Functional anatomy of the human Microprocessor. Cell 161:1374–1387.
Lund E et al. (2004) Nuclear export of microRNA precursors. Science 303:95–98.
Hutvágner G et al. (2001) A cellular function for the RNA-interference enzyme Dicer in the maturation of the let-7 small temporal RNA. Science 293:834–838.
Chendrimada TP et al. (2005) TRBP recruits the Dicer complex to Ago2 for microRNA processing and gene silencing. Nature 436:740–744.
Trabucchi M et al. (2009) The RNA-binding protein KSRP promotes the biogenesis of a subset of microRNAs. Nature 459:1010–1014.
Viswanathan SR et al. (2008) Selective blockade of microRNA processing by Lin28. Science 320:97–100.
Heo I et al. (2008) Lin28 mediates the terminal uridylation of let-7 precursor microRNA. Molecular Cell 32:276–284.
Zhao J et al. (2008) Polycomb proteins targeted by a short repeat RNA to the mouse X chromosome. Science 322:750–756.
Tsai MC et al. (2010) Long noncoding RNA as modular scaffold of histone modification complexes. Science 329:689–693.
Wang KC et al. (2011) A long noncoding RNA maintains active chromatin to coordinate homeotic gene expression. Nature 467:340–344.
Baillat D et al. (2005) Integrator, a multiprotein mediator of small nuclear RNA processing, associates with the C-terminal repeat of RNA polymerase II. Cell 123:265–276.
Wilusz JE et al. (2008) 3′ end processing of a long nuclear-retained noncoding RNA yields a tRNA-like cytoplasmic RNA. Cell 135:919–932.
Guil S, Cáceres JF. (2007) The multifunctional RNA-binding protein hnRNP A1 is required for processing of miR-18a. Nature Structural & Molecular Biology 14:701–707.
El Said NH et al. (2021) Malat-1-PRC2-EZH1 interaction supports adaptive oxidative stress response in adult tissues. Cell Death & Disease 12:850.
Baillat D, Shiekhattar R. (2009) Functional dissection of the human TNRC6 (GW182-related) family of proteins. Molecular and Cellular Biology 29:4144–4155.
Yoon JH et al. (2013) Scaffold function of long non-coding RNA HOTAIR in protein ubiquitination. Nature Communications 4:2939.
Kwon SC et al. (2016) Structure of human DROSHA. Cell 164:81–90.
Partin AC et al. (2020) Cryo-EM structures of human Drosha and DGCR8 in complex with primary microRNA. Molecular Cell 78:411–422.

This document was compiled from peer-reviewed literature. All protein names follow standard human gene/protein nomenclature. Homologs in non-model organisms (including corals) may have different names or may not yet be annotated — see notes below for coral-specific considerations.

ADDENDUM: Notes for Coral Epigenetics Research

Considerations for Non-Model Organism Work

Homolog identification: Many of the proteins listed above have conserved homologs in cnidarians. Key databases for coral homolog searches include:
- NCBI RefSeq for Acropora, Nematostella, Hydra
- Ensembl Metazoa
- Coral-specific genome portals (e.g., Reef Genomics: http://reefgenomics.org)
Cnidarian-specific observations:
- The core miRNA machinery (Drosha, Dicer, DGCR8, AGO) is conserved in cnidarians (Grimson et al. 2008 Nature 453:1193–1197; Moran et al. 2014 BMC Genomics 15:433)
- PRC2 components (EZH2, EED, SUZ12) have been identified in Nematostella vectensis, suggesting lncRNA-chromatin modification pathways exist in cnidarians
- LIN28 homologs are present in cnidarians and may regulate let-7 similarly
Environmental epigenetics relevance: In corals, miRNA and lncRNA pathways are likely involved in:
- Thermal stress response and acclimatization
- Symbiosis establishment/maintenance with Symbiodiniaceae
- Epigenetic memory and transgenerational plasticity
- DNA methylation changes under ocean acidification
Experimental approaches: When studying these pathways in corals, consider:
- RNA-seq with small RNA and long RNA fractions
- CLIP-seq (e.g., AGO-CLIP) for identifying miRNA targets
- ChIRP/CHART for identifying lncRNA-chromatin interactions
- ATAC-seq combined with lncRNA profiling for chromatin accessibility

Deep Dive Expression: compiled ncRNA machinery list