Nuclear Speckles: Molecular Organization and Selective RNA Processing

1. One structure, two literatures

The bodies now called nuclear speckles were seen twice, sixty years apart in technique. Santiago Ramón y Cajal described translucent intranuclear clumps in 1910; Hewson Swift's electron microscopy in 1959 found the corresponding objects — clusters of electron-dense ribonucleoprotein granules in the interchromatin space [57]. J. Swanson Beck coined "speckles" in 1961 for the punctate nuclear pattern stained by autoimmune sera. The equivalence of the light-microscope speckle and the EM interchromatin granule cluster was established by immunoelectron microscopy in the 1980s: antibodies against snRNP Sm antigens decorate the 20–25 nm granules of IGCs [53] [16] [14], and after Fu & Maniatis [20] raised the monoclonal antibody SC35 against spliceosome assembly factors, immunogold mapping confirmed that the bright SC35 fluorescence domains are the IGCs, while the weaker diffuse nucleoplasmic signal corresponds to perichromatin fibrils [11] [31]. "Nuclear speckles," "splicing speckles," "SC35 domains," "splicing factor compartments" and "IGCs" in this review name the same structure.

A typical mammalian interphase nucleus carries 20–50 speckles of roughly 0.5–3 µm, irregular in outline, dispersed through the interchromatin compartment; number and size vary with cell type and transcriptional state [54] [21].

2. Ultrastructure

Granules and fibrils. Each IGC is an assemblage of individual RNP granules 20–25 nm in diameter, connected at intervals by thin fibrils into beaded, reticular sub-clusters separated by less electron-dense channels [39] [48]. Bernhard's EDTA regressive staining [3] — which bleaches DNA-containing chromatin while retaining contrast on RNPs — is what made this taxonomy visible: IGCs are DNA-free granule aggregations, distinct from the perichromatin fibrils (PFs) that emerge at condensed-chromatin borders [39]. Elemental mapping by electron spectroscopic imaging (ESI) and image-EELS shows the granules themselves are nucleic-acid-rich (high phosphorus), while the space between granules is occupied by a phosphorus-poor, protein-dense material [1] [26].

What the granule cluster is not doing. Perichromatin fibrils are the ultrastructural form of nascent transcripts: EM autoradiography of pulse-labelled RNA [15] and BrUTP microinjection with immunogold detection [64] [11] place nascent RNA, hyperphosphorylated RNA polymerase II, hnRNP proteins and splicing factors on PFs at the IGC periphery and along chromatin borders throughout the nucleoplasm. The IGC interior, by contrast, incorporates essentially no BrUTP over labelling windows of 4–90 minutes [11]: transcription does not occur inside the granule cluster. Adjacent to the IGC lies the interchromatin granule-associated zone (IGAZ), a fibrogranular region enriched in U1 but not U2 snRNP [11]; EM in situ hybridization independently places poly(A) RNA in both perichromatin fibrils and interchromatin granules [63].

Purification. IGCs survive biochemical isolation: Mintz et al. [36] purified them from mouse liver and identified ~75 enriched proteins; Saitoh et al. [48] extended the inventory to 146 proteins by mass spectrometry, 81% of them devoted to RNA metabolism and 54% directly to pre-mRNA splicing — the composition expected of a splicing-factor reservoir, a point that returns in §7.

3. The chromatin around the speckle

The speckle does not float in empty nucleoplasm; it sits in a structured euchromatic neighbourhood, and the first demonstration of that neighbourhood was ultrastructural. Before 1998, in vitro work on reconstituted templates supported a picture of active chromatin as extended 10-nm or 30-nm fibres dispersed through the nucleus. Hendzel, Kruhlak & Bazett-Jones [26] tested that picture in situ, combining antibodies to dynamically acetylated histones H3/H4 (turning over with a half-life of ~12 minutes) with confocal microscopy, TEM and ESI in mammalian fibroblasts. Three results from that study still organize the field:

1. 1. Acetylated chromatin rings the IGC. Dynamically acetylated histones concentrate in discrete submicron foci at the periphery of hnRNA-accumulation sites — the IGCs — not uniformly through the nucleoplasm.
2. 2. The active chromatin is folded, not extended. ESI phosphorus/nitrogen mapping found no substantial pool of isolated 10-nm or 30-nm fibres; the perispeckle euchromatin is compacted into higher-order chromonema fibres, consistent with the folded chromonema model of Belmont & Bruce [2], with maximally decondensed loops reaching toward the IGC border.
3. 3. A DNA-depleted intermediary zone intervenes. Between the IGC surface and the acetylated chromonema lies a protein-rich, DNA-poor zone — the natural candidate location for factors that tether specific loci to the speckle periphery.

Genome-scale methods later recovered the same geography at kilobase resolution. Gene-dense, early-replicating R-bands preferentially decondense around SC35 domains while gene-poor G-bands of equal size stay segregated (Shopland et al. [50], under its correct title: clustering of multiple specific genes and gene-rich R-bands around SC-35 domains). SON TSA-seq, which measures genomic distance from speckles with a diffusing tyramide "cytological ruler," maps deterministic chromosome trajectories from lamina-associated domains to sites within <0.5 µm of speckles; the speckle-proximal genome is GC-rich, hyperacetylated and highly expressed [9]. And the association is functional, not merely spatial: in the genome-wide analysis of Bhat et al. [4], loci in the top 5% of speckle proximity showed ~3-fold higher co-transcriptional splicing efficiency than the most distant 5% (16.1% vs 5.5%), and experimentally tethering a reporter pre-mRNA to a speckle — rather than to the lamina — raised its splicing efficiency, making proximity causal, not correlative.

The speckle–chromatin link is also physical. Biochemical probing shows speckle proteins are tightly chromatin-associated — extractable only at high salt, comparably to histones — and limited DNase I nicking of chromatin collapses speckles into a dispersed or aggregated distribution without disturbing other nuclear structures [46]. The granule clusters are therefore best pictured as tethered to the surrounding chromatin scaffold, not as free-floating droplets — a fact §5's dynamics must accommodate.

4. Composition and the SON/SRRM2 scaffold

The SC35 correction. For three decades "SC35" named both an antibody and, supposedly, its antigen, the SR protein SRSF2. Ilik et al. [27] showed by quantitative IP–mass spectrometry of endogenous SC35 targets that the antibody's principal antigen is not SRSF2 but SRRM2 (~300 kDa), whose intrinsically disordered region carries the epitope; the "SC35 core" of thirty years of microscopy was mostly SRRM2. The same study, with Sharma et al. [49], established the scaffold: SON (~260 kDa) and SRRM2 are jointly essential. SON depletion alone collapses speckles into rounded/toroidal remnants; SRRM2 truncation alone leaves SON-nucleated speckles; depleting SON where SRRM2's disordered C-terminus is deleted dissolves speckles essentially completely [27]. Both proteins carry the extensive low-complexity/RS-rich disordered regions typical of condensate scaffolds, expanded in vertebrate evolution alongside GC-rich isochores.

Core and shell. Speckles are internally layered, not homogeneous. Hall et al. [25] found poly(A)+ RNA occupying a broader territory than the SC35 signal — a core/shell geometry at light-microscope resolution. Super-resolution work converges on the same picture from independent directions: SON and SRRM2 concentrate in dense sub-speckle cores, while MALAT1, U2 snRNP and pre-mRNA substrates distribute more peripherally [27]; quantitative super-resolution mapping independently places SON/SC35 centrally and MALAT1/snRNAs at the periphery [18]; and SRRM2 phase separation drives the assembly of these subcompartments [69]. The working model used through this review: a SON/SRRM2 scaffold core, and a peripheral shell where speckle-associated transcripts are handled.

MALAT1: resident, not scaffold — with a live dispute attached. The lncRNA MALAT1 is the most abundant speckle-associated RNA and localizes to the shell [60]. Whether it matters is genuinely contested. Acute depletion in human cells alters SR-protein phosphorylation, splicing-factor recruitment and alternative splicing [60] [61]; antisense targeting suppresses metastasis in xenografts [24]. But constitutive Malat1-knockout mice from independent laboratories are viable and developmentally normal, with speckles indistinguishable in number, size, distribution and SR-protein content [41] [68], whose residual phenotype is cis-regulation of neighbouring loci. The contrast with NEAT1 — the obligatory paraspeckle scaffold [40] — is the point: MALAT1 behaves as a regulatory client of speckles, not a structural component, and its strong cell-culture phenotypes appear context-dependent. The review takes no side beyond that statement of the record.

5. Dynamics

5.1 Continuous exchange, constrained motion

Live-cell imaging ended the static picture. GFP-tagged ASF/SF2 (SRSF1) in living cells shows speckles whose interiors are still but whose edges are not: peripheral extensions bud off and move toward newly activated transcription sites [38]. FRAP/FLIP measurements quantify the exchange. Phair & Misteli [43] found rapid, ATP-independent recovery — ~70% of pre-bleach intensity within 60 s, a mobile fraction near 90% — and FLIP showed the speckle pool continuously recycles through the nucleoplasm. Kruhlak et al. [30], measuring the same fusion protein against free GFP controls, add the essential correction: ASF moves up to ~100-fold more slowly than free diffusion everywhere in the nucleus, in speckles and nucleoplasm alike, recovering in bleached speckles in step with the surrounding nucleoplasm rather than from a fast free pool. Their interpretation — splicing factors undergo frequent transient binding to immobile sites throughout the nucleoplasm, and speckles are the steady-state accumulations where those sites are densest — reconciles the two datasets and fits the chromatin-tethering evidence of §3: the components exchange continuously, but nothing in the system is simply diffusing free.

Whole speckles, correspondingly, are positionally confined within their local chromatin neighbourhood [30], and splicing/transcription factors redistribute between speckles and transcription sites as transcriptional activity changes [67].

5.2 Phosphorylation-tuned cohesion

Speckle integrity is a phosphorylation state. RS-domain phosphorylation governs whether SR proteins self-associate and stay in the scaffold or release to engage nascent transcripts and the polymerase II CTD. The kinase SRPK1, whose activity peaks at mitosis, phosphorylates RS domains and disassembles speckles [23]. The CLK-family kinase Clk/Sty phosphorylates the same domains and, overexpressed, disperses speckles entirely [12]; SRPK1 and CLK have complementary substrate specificities across the SR family [13]. Protein phosphatase 1 pushes the other way, retaining factors in speckles [37]. Both extremes are non-functional — hyper- and hypo-phosphorylation each inhibit splicing [44] — so cohesion is a tuned, continuously cycled state rather than an assembly that is simply on or off. The tuning continues to acquire new components: polyphosphate maintains speckle stability by inhibiting CLK3, adding a metabolite input to the phosphorylation switchboard [32].

5.3 Mitosis: dissolution, mitotic granule clusters, staged return

Speckles do not survive mitosis as speckles. As cells enter prophase — with mitotic kinase activity rising, including the SRPK1 peak — speckle components disperse through the mitotic cytoplasm [62] [19]. A subpopulation re-aggregates into mitotic interchromatin granules (MIGs): clusters of the same 20–25 nm granules, carrying snRNPs and SR proteins but no DNA and no hnRNPs [33]. Interchromatin granules tracked across the cycle by EM persist through division in this dispersed/granular form and reappear in daughter nuclei in late telophase [59].

MIGs are not imported whole. Ferreira et al. [19] and Prasanth et al. [45] established a staged return after the nuclear envelope and lamina reassemble: initiation-form polymerase II (Ser5-phosphorylated), TFIIE and CstF-64 enter first, with splicing factors held in the cytoplasm; hnRNP C follows; splicing factors and snRNPs enter within the next ~10 minutes as the MIGs shrink; elongation-form polymerase II (Ser2-phosphorylated) and a subpopulation of SRRM2 are last, arriving in late telophase/early G1 as bulk transcription restarts. Reassembly is therefore a scheduled sequence, not a single import event — and speckle copy number is correspondingly regulated (returning to the 20–50 range) rather than set by coalescence.

5.4 Transcriptional arrest, splicing inhibition, and fusion

Stop transcription and speckles change shape within minutes: perichromatin fibrils vanish, budding ceases, and the factors retract into fewer, larger, smoothly rounded speckles [8] [55] [38] [67]. Directly inhibiting splicing produces the same endpoint [42], and the coupling runs both ways — dissolving IGCs experimentally degrades the coordination of transcription with splicing [47]. Individual molecules nonetheless keep exchanging through the rounded speckles [30]. The enlargement is not passive swelling: after transcriptional inhibition, speckles move directionally over long range and fuse, which is how number falls and size grows [28].

5.5 The inter-speckle network

The newest dynamics result changes the unit of analysis from the single speckle to the network. Live-cell imaging of endogenous SON — wide-field, PALM, photoactivation and single-molecule tracking — shows SON-enriched, multiphase bridges forming and reforming between adjacent speckles across the interchromatin space; labelled SON transfers through them over micron distances by diffusion plus advective flow, in an ATP- and transcription-dependent manner. During recovery from transcriptional inhibition, transfer through the reforming connections increases, directed toward sites of renewed transcription (Kim et al. [29], EMBO J; DOI 10.1038/s44318-026-00935-5). Speckles thus exchange material not only with the nucleoplasm (§5.1) and by whole-body fusion (§5.4), but directly with each other through transient channels — a finding that reframes the storage-versus-site debate of §7, since a "depot" that ships material along dedicated connections to where transcription restarts is doing logistics, not warehousing.

6. The perispeckle checkpoint for intron-retained RNAs

The shell of the speckle is also a holding compartment. A 2026 study defines the speckle periphery as a spatially and temporally regulated compartment for long-lived intron-retained RNAs: transcripts whose detained introns keep them parked at the boundary, with the compartment resolving during mitosis [5]. This gives the core–shell anatomy of §4 a kinetic function — the periphery as checkpoint, releasing or processing held transcripts on signal or at cell division.

7. Storage depot or splicing site? An unresolved argument

This is the field's oldest functional dispute, and this review presents it as open.

For the depot. The IGC interior contains no DNA and no nascent transcription (§2); the classic ultrastructural sites of transcription and co-transcriptional splicing are perichromatin fibrils, which occur throughout the nucleoplasm, mostly far from any speckle [16] [17] [11]. Genes can be transcribed and spliced entirely in the interdomain space — dystrophin is the standard exhibit [51]. When transcription or splicing stops, factors drain back into the speckles (§5.4), as expected of a reservoir; and the purified proteome is that of a stocked supply depot [36] [48]. On this view speckles store, assemble and phosphorylation-cycle the splicing machinery, which works elsewhere [56].

For the active hub. Active genes do not distribute randomly: fibronectin, collagen Iα1 and β-actin associate with SC35 domains at high frequency while inactive genes do not, and COL1A1 transcripts form polarized tracks — gene at the domain edge, intron spliced at the periphery, spliced RNA extending into the domain [65] [66]. Spliced mRNAs transit into the domain before export, and a disease-expanded transcript that cannot enter is retained abnormally [52]. Post-transcriptional spliceosomes — the minority of splicing completed after release from the gene — are physically retained in speckles until splicing completes [22], and released transcripts traffic to splicing-factor reservoirs [35]. Genome-wide, proximity to a speckle triples co-transcriptional splicing efficiency, and forced proximity causes the increase [4]. On this view the speckle periphery, at minimum, is a working surface, not a shelf.

Synthesis attempts. The interfacial model [34] proposes that the boundary itself is the device: SR proteins partition inward, hnRNPs stay out, so an exon-rich/intron-rich transcript sits amphiphilically at the interface with its splice sites presented to the machinery concentrated there. The condensate framing more broadly — speckles as phase-separated assemblies of SON/SRRM2 disordered regions — explains the rapid exchange and phosphorylation sensitivity of §5. But the ultrastructure resists a simple liquid-droplet reading: native speckles are irregular and fenestrated, rounding into spheres only when transcription, splicing or tethering is disrupted; they are granular-fibrillar networks, not homogeneous fluids; and they are physically tied to chromatin [46] and to fixed nucleoplasmic binding sites [30]. The position this review ends on is the one the evidence currently supports: speckles are multiphase, chromatin-tethered RNP networks whose scaffold core concentrates the machinery, whose periphery is where speckle-associated transcripts are processed and checked, and whose material economy — exchange, fusion, and now direct inter-speckle transfer — is organized around the transcription occurring at their borders. Depot and workshop are the same building, seen from different sections.

8. Open questions

• Granule-level confirmation of the core–shell model: SON/SRRM2 sub-cores rest primarily on super-resolution fluorescence; direct EM/ESI demonstration at 20–25 nm granule scale is still thin — ironically, the modality that defined the IGC is now the missing one.
• What the intermediary zone (§3) contains, and how specific loci are tethered and released.
• The rules of the inter-speckle network (§5.5): which speckles connect, what cargo selection operates at bridges, and how transfer is routed to restarting transcription sites.
• Whether the perispeckle detained-intron checkpoint (§6) is a general feature of gene expression control or a specialized pathway for a transcript subset.
• The material state, precisely: which measurable properties of SON/SRRM2 assemblies (viscosity, viscoelasticity, ageing) predict the functional readouts — exchange rates, fusion behaviour, dissolution thresholds — that the descriptive literature reports.

References

1. [1] Abholhassani-Dadras S, Vazquez-Nin GH, Echeverria OM, Fakan S (1996). Image-EELS for in situ estimation of the phosphorus content of RNP granules. Journal of Microscopy 183:215–222. DOI 10.1046/j.1365-2818.1996.950653.x
2. [2] Belmont AS, Bruce K (1994). Visualization of G1 chromosomes: a folded, twisted, supercoiled chromonema model of interphase chromatid structure. Journal of Cell Biology 127:287–302.
3. [3] Bernhard W (1969). A new staining procedure for electron microscopical cytology. Journal of Ultrastructure Research 27:250–265.
4. [4] Bhat P et al. (2024). Genome organization around nuclear speckles drives mRNA splicing efficiency. Nature 629:1165–1173. DOI 10.1038/s41586-024-07429-6
5. [5] Biayna J et al. (2026). The periphery of nuclear speckles defines a spatially and temporally regulated compartment of long-lived intron-retained RNAs that resolves during mitosis. Nature Cell Biology 28:1857–1874. DOI 10.1038/s41556-026-02040-5
6. [6] Carmo-Fonseca M et al. (1991a). Mammalian nuclei contain foci which are highly enriched in components of the pre-mRNA splicing machinery. EMBO Journal 10:195–206.
7. [7] Carmo-Fonseca M et al. (1991b). In vivo detection of snRNP-rich organelles in the nuclei of mammalian cells. EMBO Journal 10:1863–1873.
8. [8] Carmo-Fonseca M, Pepperkok R, Carvalho MT, Lamond AI (1992). Transcription-dependent colocalization of the U1, U2, U4/U6, and U5 snRNPs in coiled bodies. Journal of Cell Biology 117:1–14.
9. [9] Chen Y et al. (2018). Mapping 3D genome organization relative to nuclear compartments using TSA-Seq as a cytological ruler. Journal of Cell Biology 217:4025–4048.
10. [10] Clevenger CV, Epstein AL (1984). Identification of a nuclear protein component of interchromatin granules using a monoclonal antibody and immunogold electron microscopy. Experimental Cell Research 151:194–207.
11. [11] Cmarko D et al. (1999). Ultrastructural analysis of transcription and splicing in the cell nucleus after bromo-UTP microinjection. Molecular Biology of the Cell 10:211–223.
12. [12] Colwill K et al. (1996a). The Clk/Sty protein kinase phosphorylates SR splicing factors and regulates their intranuclear distribution. EMBO Journal 15:265–275.
13. [13] Colwill K et al. (1996b). SRPK1 and Clk/Sty protein kinases show distinct substrate specificities for serine/arginine-rich splicing factors. Journal of Biological Chemistry 271:24569–24575.
14. [14] Eliceiri GL, Ryerse JS (1984). Detection of intranuclear clusters of Sm antigens with monoclonal anti-Sm antibodies by immunoelectron microscopy. Journal of Cellular Physiology 121:449–451.
15. [15] Fakan S, Puvion E, Spohr G (1976). Localization and characterization of newly synthesized nuclear RNA in isolated rat hepatocytes. Experimental Cell Research 99:155–164.
16. [16] Fakan S, Leser G, Martin TE (1984). Ultrastructural distribution of nuclear ribonucleoproteins as visualized by immunocytochemistry on thin sections. Journal of Cell Biology 98:358–363.
17. [17] Fakan S, Leser G, Martin TE (1986). Immunoelectron microscope visualization of nuclear ribonucleoprotein antigens within spread transcription complexes. Journal of Cell Biology 103:1153–1157.
18. [18] Fei J et al. (2017). Quantitative analysis of multilayer organization of proteins and RNA in nuclear speckles at super resolution. Journal of Cell Science 130:4180–4192. DOI 10.1242/jcs.206854
19. [19] Ferreira JA, Carmo-Fonseca M, Lamond AI (1994). Differential interaction of splicing snRNPs with coiled bodies and interchromatin granules during mitosis and assembly of daughter cell nuclei. Journal of Cell Biology 126:11–23.
20. [20] Fu XD, Maniatis T (1990). Factor required for mammalian spliceosome assembly is localized to discrete regions in the nucleus. Nature 343:437–441.
21. [21] Galganski L, Urbanek MO, Krzyzosiak WJ (2017). Nuclear speckles: molecular organization, biological function and role in disease. Nucleic Acids Research 45:10350–10368.
22. [22] Girard C et al. (2012). Post-transcriptional spliceosomes are retained in nuclear speckles until splicing completion. Nature Communications 3:994. DOI 10.1038/ncomms1998
23. [23] Gui JF, Lane WS, Fu XD (1994). A serine kinase regulates intracellular localization of splicing factors in the cell cycle. Nature 369:678–682.
24. [24] Gutschner T et al. (2013). The noncoding RNA MALAT1 is a critical regulator of the metastasis phenotype of lung cancer cells. Cancer Research 73:1180–1189.
25. [25] Hall LL, Smith KP, Byron M, Lawrence JB (2006). Molecular anatomy of a speckle. Anatomical Record Part A 288:664–675.
26. [26] Hendzel MJ, Kruhlak MJ, Bazett-Jones DP (1998). Organization of highly acetylated chromatin around sites of heterogeneous nuclear RNA accumulation. Molecular Biology of the Cell 9:2491–2507. PMID 9725908. DOI 10.1091/mbc.9.9.2491
27. [27] Ilik İA et al. (2020). SON and SRRM2 are essential for nuclear speckle formation. eLife 9:e60579.
28. [28] Kim J et al. (2019). Nuclear speckle fusion via long-range directional motion regulates speckle morphology after transcriptional inhibition. Journal of Cell Science 132:jcs226563. DOI 10.1242/jcs.226563
29. [29] Kim J, Hernandez Gonzalez GA, Chivukula Venkata N, Copik AJ, Han KY, Belmont AS (2026). Non-random inter-speckle material transfer mediated by dynamic multiphase interchromatin connections. The EMBO Journal. DOI 10.1038/s44318-026-00935-5
30. [30] Kruhlak MJ et al. (2000). Reduced mobility of the alternate splicing factor (ASF) through the nucleoplasm and steady state speckle compartments. Journal of Cell Biology 150:41–51.
31. [31] Lamond AI, Spector DL (2003). Nuclear speckles: a model for nuclear organelles. Nature Reviews Molecular Cell Biology 4:605–612.
32. [32] Lázaro B et al. (2026). Nuclear speckle dynamics are controlled by polyphosphate inhibition of CLK proteins. Nucleic Acids Research 54:gkag309. DOI 10.1093/nar/gkag309
33. [33] Leser GP, Fakan S, Martin TE (1989). Ultrastructural distribution of ribonucleoprotein complexes during mitosis. snRNP antigens are contained in mitotic granule clusters. European Journal of Cell Biology 50:376–389.
34. [34] Liao SE, Regev O (2021). Splicing at the phase-separated nuclear speckle interface: a model. Nucleic Acids Research 49:636–645. DOI 10.1093/nar/gkaa1209
35. [35] Melčák I et al. (2000). Nuclear pre-mRNA compartmentalization: trafficking of released transcripts to splicing factor reservoirs. Molecular Biology of the Cell 11:497–510.
36. [36] Mintz PJ et al. (1999). Purification and biochemical characterization of interchromatin granule clusters. EMBO Journal 18:4308–4320.
37. [37] Misteli T, Spector DL (1996). Serine/threonine phosphatase 1 modulates the subnuclear distribution of pre-mRNA splicing factors. Molecular Biology of the Cell 7:1559–1572.
38. [38] Misteli T, Cáceres JF, Spector DL (1997). The dynamics of a pre-mRNA splicing factor in living cells. Nature 387:523–527.
39. [39] Monneron A, Bernhard W (1969). Fine structural organization of the interphase nucleus in some mammalian cells. Journal of Ultrastructure Research 27:266–288.
40. [40] Nakagawa S, Naganuma T, Shioi G, Hirose T (2011). Paraspeckles are subpopulation-specific nuclear bodies that are not essential in mice. Journal of Cell Biology 193:31–39.
41. [41] Nakagawa S et al. (2012). Malat1 is not an essential component of nuclear speckles in mice. RNA 18:1487–1499.
42. [42] O'Keefe RT, Mayeda A, Sadowski CL, Krainer AR, Spector DL (1994). Disruption of pre-mRNA splicing in vivo results in reorganization of splicing factors. Journal of Cell Biology 124:249–260.
43. [43] Phair RD, Misteli T (2000). High mobility of proteins in the mammalian cell nucleus. Nature 404:604–609.
44. [44] Prasad J, Colwill K, Pawson T, Manley JL (1999). The protein kinase Clk/Sty directly modulates SR protein activity: both hyper- and hypophosphorylation inhibit splicing. Molecular and Cellular Biology 19:6991–7000.
45. [45] Prasanth KV, Sacco-Bubulya PA, Prasanth SG, Spector DL (2003). Sequential entry of components of the gene expression machinery into daughter nuclei. Molecular Biology of the Cell 14:1043–1057.
46. [46] Raina K, Rao BJ (2022). Mammalian nuclear speckles exhibit stable association with chromatin: a biochemical study. Nucleus 13:58–73. DOI 10.1080/19491034.2021.2024948
47. [47] Sacco-Bubulya P, Spector DL (2002). Disassembly of interchromatin granule clusters alters the coordination of transcription and pre-mRNA splicing. Journal of Cell Biology 156:425–436.
48. [48] Saitoh N et al. (2004). Proteomic analysis of interchromatin granule clusters. Molecular Biology of the Cell 15:3876–3890.
49. [49] Sharma A et al. (2010). Son is essential for nuclear speckle organization and cell cycle progression. Molecular Biology of the Cell 21:650–663.
50. [50] Shopland LS, Johnson CV, Byron M, McNeil J, Lawrence JB (2003). Clustering of multiple specific genes and gene-rich R-bands around SC-35 domains: evidence for local euchromatic neighborhoods. Journal of Cell Biology 162:981–990. DOI 10.1083/jcb.200303131
51. [51] Smith KP, Moen PT, Wydner KL, Coleman JR, Lawrence JB (1999). Processing of endogenous pre-mRNAs in association with SC-35 domains is gene specific. Journal of Cell Biology 144:617–629.
52. [52] Smith KP, Byron M, Johnson C, Xing Y, Lawrence JB (2007). Defining early steps in mRNA transport: mutant mRNA in myotonic dystrophy type I is blocked at entry into SC-35 domains. Journal of Cell Biology 178:951–964.
53. [53] Spector DL, Schrier WH, Busch H (1983). Immunoelectron microscopic localization of snRNPs. Biology of the Cell 49:1–10.
54. [54] Spector DL, Lark G, Huang S (1992). Differences in snRNP localization between transformed and nontransformed cells. Molecular Biology of the Cell 3:555–569.
55. [55] Spector DL (1993). Macromolecular domains within the cell nucleus. Annual Review of Cell Biology 9:265–315.
56. [56] Spector DL, Lamond AI (2011). Nuclear speckles. Cold Spring Harbor Perspectives in Biology 3:a000646.
57. [57] Swift H (1959). Studies on nuclear fine structure. Brookhaven Symposia in Biology 12:134–152.
58. [58] Thiry M (1993). Differential location of nucleic acids within interchromatin granule clusters. European Journal of Cell Biology 62:259–269.
59. [59] Thiry M (1995). Behavior of interchromatin granules during the cell cycle. European Journal of Cell Biology 68:14–24.
60. [60] Tripathi V et al. (2010). The nuclear-retained noncoding RNA MALAT1 regulates alternative splicing by modulating SR splicing factor phosphorylation. Molecular Cell 39:925–938.
61. [61] Tripathi V et al. (2013). Long noncoding RNA MALAT1 controls cell cycle progression by regulating the expression of oncogenic transcription factor B-MYB. PLoS Genetics 9:e1003368.
62. [62] Verheijen R et al. (1986). Distribution of the 70K U1 RNA-associated protein during interphase and mitosis. Correlation with other U RNP particles and proteins of the nuclear matrix. Journal of Cell Science 86:173–190.
63. [63] Visa N, Puvion-Dutilleul F, Harper F et al. (1993). Intranuclear distribution of poly(A) RNA determined by electron microscope in situ hybridization. Experimental Cell Research 208:19–34.
64. [64] Wansink DG et al. (1993). Fluorescent labeling of nascent RNA reveals transcription by RNA polymerase II in domains scattered throughout the nucleus. Journal of Cell Biology 122:283–293.
65. [65] Xing Y, Johnson CV, Dobner PR, Lawrence JB (1993). Higher level organization of individual gene transcription and RNA splicing. Science 259:1326–1330.
66. [66] Xing Y, Johnson CV, Moen PT Jr, McNeil JA, Lawrence JB (1995). Nonrandom gene organization: structural arrangements of specific pre-mRNA transcription and splicing with SC-35 domains. Journal of Cell Biology 131:1635–1647.
67. [67] Zeng C, Kim E, Warren SL et al. (1997). Dynamic relocation of transcription and splicing factors dependent upon transcriptional activity. EMBO Journal 16:1401–1412. DOI 10.1093/emboj/16.6.1401
68. [68] Zhang B et al. (2012). The lncRNA Malat1 is dispensable for mouse development but its transcription plays a cis-regulatory role in the adult. Cell Reports 2:111–123.
69. [69] Zhang M, Gu Z, Guo S et al. (2024). SRRM2 phase separation drives assembly of nuclear speckle subcompartments. Cell Reports 43:113827. DOI 10.1016/j.celrep.2024.113827