Cell Biology · Unit 2: Membranes and compartments

Nuclear Architecture and Membraneless Compartments

The nucleus has no internal membranes, yet its chromosomes keep to their own regions and its nucleoplasm holds bodies with different jobs. This lesson describes chromosome territories and the two Hi-C compartments, the nucleolus and its smaller neighbours, and the proposal that many such bodies form by phase separation. Afterwards you can state what phase separation predicts, work a mass-balance test on concentration data, and say why round shapes, FRAP and hexanediol do not settle the mechanism.

Watch first Change the Amount, Watch the Bodies · 2:58

A human nucleus is only several micrometres across and holds roughly two metres of DNA, but it has no membranes inside it, and it is still not a uniform soup. Chromosomes keep to their own regions. Ribosomes are assembled in one place, splicing factors gather in others. So what holds things apart when there are no walls? This lesson separates the parts of that answer that are well supported from the parts that are still argued over.

Chromosomes keep to territories, mostly

During mitosis each chromosome is a compact rod. Afterwards it partly unpacks, but it does not spread through the whole nucleus. When whole chromosomes are labelled with fluorescent DNA probes, a method called chromosome painting, each one fills a region of its own, a chromosome territory. The DNA is packed as chromatin, DNA wound on histone proteins, and how tightly it is packed affects whether its genes are used, a topic covered in that lesson.

Picture balls of yarn packed in a box. Each ball keeps its own space, but where two balls press together, loose loops of one poke into the other. Whether territories really work that way was hard to see, because a light microscope blurs everything below a few hundred nanometres, as explained in that lesson. In 2006 Miguel Branco and Ana Pombo cut fixed human lymphocytes, activated with phytohemagglutinin, into frozen sections about 150 nm thick before painting them. The thin section removes most of the blur from above and below. Neighbouring territories overlapped at their edges, and how much two chromosomes overlapped correlated with how often that pair takes part in translocations measured in human lymphocytes after ex vivo irradiation. That is a correlation, not proof that overlap causes any given translocation.

A second view comes from Hi-C, a sequencing method. Chromatin is cross-linked in the cell, cut, and the ends are joined so that segments lying close in space become joined in one DNA molecule. Sequencing millions of these junctions gives a map of which parts of the genome touch. The 2009 Hi-C maps of human cells, at a resolution of one megabase, showed that open and closed chromatin segregate into two genome-wide compartments, now called A (open, gene-rich, mostly active) and B (closed, mostly quiet). Closed chromatin often lies against the nuclear lamina, the protein mesh under the nuclear envelope described in that lesson. Remember that Hi-C averages over millions of cells: it reports how often two regions are near each other, not where they sit in one nucleus.

Bodies without membranes: the nucleolus and its neighbours

A nuclear body is a region of the nucleoplasm where particular molecules are concentrated, with no membrane around it. The largest is the nucleolus, whose main job is ribosome biogenesis. It forms around the repeated genes for ribosomal RNA and has three parts, like an assembly line read from the inside out. The fibrillar centre holds the rDNA and the RNA polymerase I machinery. The dense fibrillar component around it, rich in fibrillarin, is where the newly made pre-rRNA begins to be processed. The outer granular component, rich in nucleophosmin (NPM1), is where ribosomal subunits are assembled before export. In nucleoli of Xenopus laevis oocytes, the fibrillarin-rich and NPM1-rich layers behave as two liquids that coexist without mixing.

Smaller bodies sit in the nucleoplasm around it. You can learn them by what they hold.

  • Nuclear speckles are enriched in pre-mRNA splicing factors. Immunofluorescence usually shows about 20 to 50 per nucleus in mammalian cells, and their contents exchange with the surrounding nucleoplasm. In mouse embryonic stem cells, genes near speckles showed more co-transcriptional splicing than genes farther away, and in human HEK293T cells, tethering a reporter pre-mRNA to speckles increased its splicing. Splicing itself is covered in that lesson.
  • Cajal bodies are marked by the protein coilin and contain components of the small nuclear ribonucleoproteins (snRNPs) that splice pre-mRNA. In an engineered mammalian cell system, tethering Cajal-body components to chromatin was enough to start apparently functional Cajal bodies from scratch.
  • PML bodies are built from the PML protein. In cultured mammalian cells and mouse experiments, oxidation drove PML to multimerize and nucleate the bodies, which then enhanced stress-induced SUMO modification of partner proteins held inside them. The name comes from promyelocytic leukaemia, in which a fusion of PML with a retinoic acid receptor is associated with disrupted bodies.
  • Paraspeckles are built on a long noncoding RNA, NEAT1. Mice lacking Neat1 have no paraspeckles yet were reported viable and fertile in the laboratory. Later work found milder defects, in pregnancy and lactation, so the bodies are not needed for survival under those conditions, which is not the same as doing nothing.

These facts are well supported: the bodies exist, they exchange molecules with the nucleoplasm, and several of them help with a job. How they form is a different question.

The phase-separation hypothesis and what it predicts

A biomolecular condensate is a membraneless body that concentrates proteins and nucleic acids. One proposed way to make one is liquid-liquid phase separation: above a certain concentration, a solution of molecules demixes into a dense phase and a dilute phase, as salad dressing separates into oil and vinegar. For proteins the driving force is thought to be multivalency, molecules with many weak binding sites (repeated domains, disordered regions, RNA) that stick to many partners at once. These are the weak, reversible bonds described in that lesson. The idea took hold after Brangwynne, Hyman and colleagues reported in 2009 that P granules in the one-cell embryo of the worm C. elegans fuse, flow and drip like liquid drops.

Phase separation is one way to make a local high concentration, not the only one. A protein can bind clustered sites, many copies of a DNA or RNA sequence in one place, and pile up there with no demixing. A protein can also bridge chromatin, cross-linking distant segments so the fibre collapses into a dense globule. All three make a bright spot in a microscope. They differ in what happens when you change the amount of protein.

Phase separation predicts three things. There is a threshold, the saturation concentration, below which no bodies form. Above it, the dilute phase is buffered: extra protein goes into the bodies, which grow, while the concentration outside stays at the threshold. And there is a boundary, so molecules should mix inside a body faster than they cross its edge. A simple mass balance turns the second prediction into numbers.

Condensate lever rule: how bodies grow as the protein goes up⚙ original · interactive
Set the slider to the lesson’s cases
Dilute phase
Dense phase
Bodies fill (φ)

An idealized model, in this sketch: one dense phase at 50 μM, a saturation concentration of 1 μM, a dilute phase held at that value, and made-up data. The number and places of the dots are arbitrary, only their total area means anything. Binding to clustered sites or bridging chromatin can also make bright spots, and a protein that fails this test may still phase separate as one part of a mixture of several components. A protein that passes has not been shown to need the body for any function. Keep the rule φ = (c_total − c_sat) / (c_dense − c_sat).

c_total = c_sat × (1 − φ) + c_dense × φ, so φ = (c_total − c_sat) / (c_dense − c_sat)

Here φ is the fraction of nuclear volume taken up by the dense phase.

Worked problem (made-up data, for illustration). A GFP-tagged nuclear protein is expressed in a cell line at total nuclear concentrations of 2, 4 and 8 μM. Assume one dense phase at 50 μM, a saturation concentration of 1 μM, and a dilute phase held at that value.

  1. The denominator is c_dense − c_sat = 50 − 1 = 49 μM.
  2. At 2 μM, φ = (2 − 1)/49 = 1/49 = 0.020, so bodies fill 2.0% of the nucleus. At 4 μM, φ = 3/49 = 6.1%. At 8 μM, φ = 7/49 = 14.3%.
  3. So the model predicts that the bodies grow sevenfold in volume between 2 and 8 μM while the nucleoplasm stays near 1 μM.
  4. Now compare two proteins. For protein A, the measured nucleoplasm is 1.0, 1.0 and 1.1 μM: it fits. For protein B it is 1.5, 2.8 and 5.9 μM. Divide by the totals: 1.5/2 = 0.75, 2.8/4 = 0.70, 5.9/8 = 0.74. Protein B keeps a roughly constant fraction outside the bodies, which fits partitioning onto binding sites better.

Treat the result with care. A protein that fails this test may still phase separate as one part of a mixture of several components, which need not buffer each one. A protein that passes has not been shown to need the body for any function.

Evidence that is not enough: round shapes, fusion, FRAP, hexanediol

Many early claims rested on three observations, and none of them separates the models. Round shapes and fusion arise for any soft body pulled by surface forces, and clusters on chromatin can look the same. FRAP, recovery of fluorescence after photobleaching (see that lesson), measures turnover: molecules that bind and unbind fixed sites also refill a bleached spot. And 1,6-hexanediol, an alcohol thought to weaken hydrophobic contacts, at 5% for 5 minutes strongly suppressed chromatin motion in live HeLa cells. A body that dissolves in it has not been shown to be a liquid.

Heterochromatin shows why the system matters. Purified Drosophila HP1a demixed in the test tube, and HP1a foci in early Drosophila embryos looked liquid-like, with mature domains proposed to contain liquid and stable parts. In mouse fibroblasts, by contrast, chromocentres behaved like collapsed chromatin globules rather than HP1 liquid pools. Purified mouse HP1α mixed with DNA turned turbid at midpoints of 45 μM untagged and 23 μM with a GFP tag. Those are test-tube numbers, not saturation concentrations in a cell, and they show that the tag alone can shift the result. Neither study speaks for all heterochromatin.

The threshold test has its own catch. In engineered Corelet experiments in living cells, local capture of a protein on a seed triggered phase separation below the global concentration needed for it, so a body on chromatin can form where the average level looks too low.

What is contested, then, is whether phase separation is the mechanism for a given body in a given living cell, and whether its material state matters for its job.

Design problem

A laboratory reports that protein X forms round nuclear foci that fuse, recover after photobleaching and dissolve in 1,6-hexanediol. It concludes that X drives a phase separation needed for transcription. What is wrong with the argument, and what would you do instead?

One way to do it

None of the three observations is diagnostic. Roundness, fusion and FRAP recovery also fit binding to clustered sites or chromatin bridging. Hexanediol at 5% for 5 minutes suppresses chromatin motion in live HeLa cells. In that system, Cajal-body foci fell from 2.98 to 0.84 per cell, a drop of (2.98 − 0.84)/2.98 = 2.14/2.98 = 71.8%, yet that drop says nothing about mechanism.

Instead, tag X at its endogenous locus so its level is normal. Raise and lower that level and measure concentrations inside and outside the foci, testing for a threshold and a buffered dilute phase as in the worked problem. Use half-body FRAP to ask whether X mixes inside a focus faster than it crosses the edge. Make a mutant that keeps DNA binding but loses its multivalent region, and see whether foci and transcription change together. Finally, test transcription separately from assembly, with a control that moves the same genes to or from the foci by another route, as Bhat and colleagues did for splicing at speckles. Only then can you ask whether the foci themselves, rather than protein X alone, are needed.

How we measure it

Cryo-FISH with chromosome paints

Fixed cells are cut into thin frozen sections, about 150 nm in the Branco and Pombo work, and whole chromosomes are labelled with fluorescent DNA probes. The thin section removes most out-of-focus light, so overlap between two territories can be measured rather than blurred.

Hi-C

Chromatin is cross-linked in the cell, cut, and re-joined so that segments close in space are ligated together, then the junctions are sequenced. The result is a genome-wide map of contact frequencies averaged over many cells, from which open and closed compartments can be read.

Half-body FRAP

Fluorescence is bleached in one half of a nuclear body and recovery is followed in both halves and in the surrounding nucleoplasm. Whether molecules refill the bleached half mainly from the other half or from outside tells internal mixing apart from exchange across the edge.

Concentration titration with endogenous tagging

The protein is tagged at its own genomic locus, then its level is raised or lowered, for example by inducible expression or induced degradation. Measuring concentrations inside and outside the bodies at each level tests whether there is a threshold and a buffered dilute phase.

Finish line

Test yourself

8 questions

Practice and MCAT-style, with explained answers.

Next · Lesson 11

Vesicle Budding, Targeting and Fusion

A yeast mutant keeps making secretory enzymes at 37 °C but cannot release them. Where do they go?

Continue →

← Lesson 9 · Protein Quality Control and Organelle Turnover

Nucleosome core particle — the unit of chromatin that territories and compartments are made of
Nucleosome core particle — the unit of chromatin that territories and compartments are made of ● LIVE · RCSB PDB 1KX5
open on RCSB PDB ↗
Loading live structure 1KX5 from RCSB PDB…

About 147 base pairs of DNA wrap almost twice around a core of eight histones. Look at the histone tails that stick out past the DNA: their modifications help decide whether a region sits in open or closed chromatin.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
HP1 chromodomain bound to a methylated histone H3 tail — the reader of closed chromatin
HP1 chromodomain bound to a methylated histone H3 tail — the reader of closed chromatin ● LIVE · RCSB PDB 1KNE
open on RCSB PDB ↗
Loading live structure 1KNE from RCSB PDB…

The H3 tail lies in a groove of the chromodomain, and the methylated lysine sits in a cage of aromatic side chains. HP1 binds chromatin through this contact, which is why its foci are central to the argument over how heterochromatin is organized.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
1,6-Hexanediol — the alcohol used to dissolve nuclear bodies, and its confound
1,6-Hexanediol — the alcohol used to dissolve nuclear bodies, and its confound ● LIVE · PubChem CID 12374
open on PubChem ↗
Loading live structure 12374 from PubChem…

A six-carbon chain with a hydroxyl group at each end. It is thought to weaken weak hydrophobic contacts, but at 5% it also suppresses chromatin motion in live HeLa cells, so a body that dissolves in it has not been shown to be a liquid.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
Clifford Brangwynne

Clifford Brangwynne

With Anthony Hyman and colleagues, reported in 2009 that P granules in the one-cell embryo of the worm C. elegans behave as liquid droplets, which gave cell biologists a physical language for bodies without membranes.

P granules are clumps of RNA and protein that end up in the cells that will form the germ line of the worm Caenorhabditis elegans. In a 2009 paper from Anthony Hyman's laboratory at the Max Planck Institute of Molecular Cell Biology and Genetics in Dresden, Clifford Brangwynne and colleagues watched these granules in the one-cell embryo. The granules were round, they fused when they met, and they flowed and dripped as liquid drops do. Their components also exchanged with the surrounding cytoplasm. The paper argued that the granules gather at the rear of the embryo because they condense preferentially at the back, rather than being carried there. Physicists were among the authors, and the study described the granules in terms of viscosity and surface tension. Its claim was about P granules in one embryo. The broader idea, that many bodies in many cells form the same way, came afterwards and is still being tested, but the paper gave the field a set of physical questions to ask of any body without a membrane.

✓ verbatim Brangwynne et al., Science 324:1729-1732 (2009), Germline P granules are liquid droplets that localize by controlled dissolution/condensation.
Marina Feric

Marina Feric

With Clifford Brangwynne and colleagues, showed that the layers of the nucleolus in Xenopus oocytes behave as coexisting liquid phases, and that purified nucleolar proteins reproduce the layering.

A nucleolus has layers, and Marina Feric asked why. Working at Princeton University with Clifford Brangwynne and collaborators, she studied the large nucleoli in oocytes of the frog Xenopus laevis. In these nucleoli, a dense fibrillar component rich in fibrillarin (FIB1) sits inside a granular component rich in nucleophosmin (NPM1). The two layers behaved as separate liquids that coexist without mixing, much as oil and water do. With purified proteins the team rebuilt the arrangement in the test tube, and coarse-grained computer simulations reproduced it. Differences in surface tension between the two phases helped decide which one ended up inside. The 2016 paper in Cell gave a physical account of layering for these oocyte nucleoli and for purified components. Whether every nucleolus in every cell type is built the same way is a separate question, but the study showed that a layered structure with no membranes can, in principle, arise from the physics of the molecules themselves.

✓ verbatim Feric et al., Cell 165:1686-1697 (2016), Coexisting liquid phases underlie nucleolar subcompartments.
Ana Pombo

Ana Pombo

With Miguel Branco, used cryo-FISH on thin sections of activated human lymphocytes to show that neighbouring chromosome territories intermingle at their edges, and that the extent of intermingling correlates with translocation frequency.

By the early 2000s, painting whole chromosomes with fluorescent probes had shown that each interphase chromosome occupies its own region of the nucleus. What happened at the edges was harder to see. In a whole nucleus, light microscopy blurs every boundary, so territories that touch, keep apart or mix can look alike. Miguel Branco and Ana Pombo cut fixed human lymphocytes, activated with phytohemagglutinin, into cryosections about 150 nm thick and painted pairs of chromosomes in them. A thin section removes most of the blur from above and below. The territories overlapped at their edges, so they are not sealed boxes. The amount of overlap between pairs of chromosomes also correlated with how often those pairs exchange pieces in translocations measured in human lymphocytes after ex vivo irradiation. A correlation does not show the cause of any one translocation, but the 2006 paper in PLoS Biology replaced the picture of separate compartments with territories that meet and interpenetrate, and gave a physical reason to expect some chromosome pairs to form translocations more often than others.

✓ verbatim Branco & Pombo, PLoS Biology 4:e138 (2006), Intermingling of chromosome territories in interphase suggests role in translocations and transcription-dependent associations.
Erez Lieberman Aiden

Erez Lieberman Aiden

With Job Dekker, Eric Lander and colleagues, developed Hi-C, which maps which parts of the genome lie close together, and found that human chromatin segregates into two genome-wide compartments, one open and one closed.

Microscopes show where a few labelled stretches of DNA sit in a nucleus. They cannot show every pair of loci at once. In 2009 Erez Lieberman Aiden, Nynke van Berkum and colleagues, working with the laboratories of Job Dekker and Eric Lander, reported a method that could. In Hi-C, cells are fixed so that DNA segments lying near one another are cross-linked, the DNA is cut, and the cut ends are joined so that neighbours in space become joined in one molecule. Massively parallel sequencing then reads millions of these junctions, each one a record of two pieces of genome that were close. At a resolution of one megabase, the human maps showed that open and closed chromatin segregate into two genome-wide compartments, which the paper labelled A and B. The data come from millions of cells at once, so they give contact frequencies across a population rather than a picture of one nucleus. The method made the folding of whole genomes something that could be measured, compared between cell types and tested with mutants.

✓ verbatim Lieberman-Aiden et al., Science 326:289-293 (2009), Comprehensive mapping of long-range interactions reveals folding principles of the human genome.
Test yourself
Question 1 of 8 Score 0 / 0

Which statement best describes chromosome territories in activated human lymphocytes, as seen by cryo-FISH on thin sections?
Sources & Integrity

Every structure, slide, simulation, and clip on this page is listed below with the source it came from. None of the 3D structures is drawn by hand or by AI: each one loads by accession ID from RCSB PDB or PubChem. This panel is generated from the lesson's own data file, so what it lists is what the page loads.

Molecular structures live data

  • Nucleosome core particle — the unit of chromatin that territories and compartments are made of — RCSB PDB 1KX5 · coordinates CC0 · Nucleosome core particle with 147 bp of DNA, X-ray crystal structure at 1.9 Å (Davey et al., J. Mol. Biol., 2002)
  • HP1 chromodomain bound to a methylated histone H3 tail — the reader of closed chromatin — RCSB PDB 1KNE · coordinates CC0 · Drosophila HP1 chromodomain with an H3 tail peptide trimethylated on lysine 9, X-ray crystal structure (Jacobs & Khorasanizadeh, Science, 2002)
  • 1,6-Hexanediol — the alcohol used to dissolve nuclear bodies, and its confound — PubChem CID 12374 · PubChem compound record for 1,6-hexanediol

Simulations

  • Condensate lever rule · Original interactive — Biochemistrypedia (Deppmann Lab).

Scientist profiles & quotes

  • Clifford Brangwynne — Brangwynne et al., Science 324:1729-1732 (2009), Germline P granules are liquid droplets that localize by controlled dissolution/condensation.
  • Marina Feric — Feric et al., Cell 165:1686-1697 (2016), Coexisting liquid phases underlie nucleolar subcompartments.
  • Ana Pombo — Branco & Pombo, PLoS Biology 4:e138 (2006), Intermingling of chromosome territories in interphase suggests role in translocations and transcription-dependent associations.
  • Erez Lieberman Aiden — Lieberman-Aiden et al., Science 326:289-293 (2009), Comprehensive mapping of long-range interactions reveals folding principles of the human genome.

Quotes are verbatim from the cited source. Profile text is original to this site and was checked against the sources cited.

Animated explainer

Change the Amount, Watch the Bodies · narration: an AI narrator voice (Gemini, Charon) reading narration written from this lesson's text · animation: AI-generated in code (Manim) by Claude, a schematic with no drawn molecular structures, plots computed from their equations · captions from the narration script.

Images & figures

This lesson carries no raster figures. Every structure is rendered live from RCSB PDB or PubChem, and every plot is drawn from its equation in the browser.

integrity Machines check the mechanics. The build confirms that every local file exists, that every question's answer key points at a real choice, and that any AI-generated scientific figure names the person who checked it; a weekly job re-resolves every accession ID against its database. Questions are original; MCAT-style items are written for this site and are not real AAMC questions. Banned sources (the Tymoczko textbook, publisher decks, copyrighted figures) are not used.

Like I'm five

Inside the middle of a cell there are no walls, but things still stay in their own places. Each long string of DNA keeps to its own patch, like balls of yarn in a box. Some helpers also gather into little blobs, a bit like drops of oil in salad dressing. Scientists are still checking which blobs really are like oil drops and which are held together some other way.
High school

The nucleus has no membranes inside it, yet it is organized. Each chromosome occupies its own region, called a territory, though neighbouring territories mix at their edges. Open, active DNA and closed, quiet DNA tend to sit in different places. The nucleus also holds bodies without membranes: the nucleolus, where ribosomes are built, and smaller bodies that store splicing factors or other proteins. One idea is that some of these bodies form the way oil separates from vinegar, when enough of their molecules stick weakly to one another. That idea makes testable predictions, and many experiments that seemed to support it turn out to fit other explanations too.
College

Interphase chromosomes occupy territories that intermingle at their edges, shown by cryo-FISH on thin sections of activated human lymphocytes, and Hi-C contact maps sort chromatin into an open A compartment and a closed B compartment. Membraneless nuclear bodies include the nucleolus (fibrillar centre, dense fibrillar component and granular component, organized around ribosome biogenesis), nuclear speckles (splicing factors), Cajal bodies (coilin and snRNP components), PML bodies (PML, SUMO and partner proteins) and paraspeckles (built on the lncRNA NEAT1). A biomolecular condensate is a membraneless body that concentrates molecules, and one proposed mechanism is liquid-liquid phase separation driven by multivalent weak interactions. Phase separation predicts a saturation concentration, a dilute phase buffered at that concentration as bodies grow, and a boundary across which exchange is slower than mixing inside. Binding to clustered sites and polymer bridging can also make round, recovering, hexanediol-sensitive foci.
Graduate

In a simple two-phase model, mass balance gives c_total = c_sat(1 − φ) + c_dense(φ), so the dense-phase volume fraction φ rises linearly with total concentration while the dilute phase stays at c_sat. A nucleoplasmic concentration that instead tracks the total as a constant fraction fits partitioning onto binding sites that are not saturated. Neither result is decisive: multicomponent systems need not buffer one component, and local nucleation on chromatin can trigger separation below the global threshold, as engineered Corelet experiments showed. Half-body FRAP separates internal mixing from exchange across an edge, and in mouse fibroblasts it favoured collapsed chromatin globules for chromocentres, whereas HP1a foci in early Drosophila embryos looked liquid-like. Hexanediol at 5% for 5 minutes suppresses chromatin motion in live HeLa cells, which removes it as a specific test. The open question is functional: even where a body is shown to be a separate phase, does its material state, liquid rather than gel or bound cluster, matter for what it does? Directed recruitment of a reporter pre-mRNA to speckles raised its splicing in HEK293T cells, which shows that location matters, but it does not show that phase separation is the reason.
The solvent that also froze chromatin

For years, 1,6-hexanediol was a standard way to ask whether a body was a liquid: if it dissolved, the body was taken to be held together by weak contacts. Itoh and colleagues measured what the alcohol does to chromatin itself. Tracking single nucleosomes in live human cells, they found that 5% hexanediol for 5 minutes strongly suppressed chromatin motion in HeLa cells. In the same treatment, coilin-labelled Cajal-body foci fell from a mean of 2.98 to 0.84 per cell. A reagent that changes chromatin this much cannot tell you which mechanism held a body together.

— Itoh et al., Life Science Alliance 4:e202001005 (2021), Figure 1C and live single-nucleosome imaging.
A knockout with no paraspeckles and no obvious problem

Paraspeckles are built on the long noncoding RNA NEAT1. When Nakagawa, Hirose and colleagues deleted Neat1 in mice, the paraspeckles were gone, yet the mice were viable and fertile under laboratory conditions. Later studies found that some knockout females fail to establish pregnancy and that mammary gland development is impaired. In a survey of adult tissues, prominent paraspeckles turned up only in the subpopulations of cells that strongly express the long NEAT1_2 isoform. The result is a useful warning in both directions: a structure can be real and well defined without being needed in a comfortable cage, and its job may show up only under particular conditions or in particular cells.

— Nakagawa et al., J. Cell Biol. 193:31-39 (2011).

 ↑ Top