Genetics and Molecular Biology · Unit 4: From gene to protein, and its control

RNA Processing: Caps, Tails and Splicing

In the eukaryotic nucleus, the RNA that polymerase II makes is not yet a message. This lesson follows the three changes that make it one: a cap on the 5' end, a cut and a poly(A) tail at the 3' end, and splicing, which removes introns and joins exons. It tells how split genes were found in adenovirus 2 in 1977, and shows how to predict, from exon lengths alone, what a splice-site mutation does to an mRNA and its reading frame.

Watch first A Gene Is Longer Than Its Message · 3:10

The RNA that polymerase II copies from a eukaryotic gene is not the RNA that a ribosome reads.

Take an imagined human gene with four exons and three introns that runs to 5,137 nucleotides from the start of transcription to the end of its last exon. The finished mRNA carries 737 of them, plus a cap and a tail that the gene never encoded. This lesson is about the gap: what is added, what is cut out, how, and what goes wrong when one base at a cut site changes.

The idea: a primary transcript is not yet a message

What RNA polymerase II makes in transcription is the primary transcript, or pre-mRNA. Three changes turn it into mRNA.

  1. A cap is added to the 5’ end.
  2. The 3’ end is cut and given a poly(A) tail.
  3. Splicing removes the introns, the stretches of the gene that are not kept, and joins the exons, the stretches that are.

Think of an editor cutting a film while it is still being shot: none of the three steps waits for transcription to end. Capping happens once about 25 to 30 nucleotides of RNA have emerged from the polymerase, and splicing of early introns often begins while later parts of the gene are still being copied.

Bacterial mRNAs get no cap and carry no spliceosomal introns (a few bacterial genes hold self-splicing introns), and ribosomes translate them while they are still being made. The nuclear envelope gives eukaryotes time to process the message before it meets a ribosome.

Caps and tails: marking both ends

The cap is a guanosine methylated at position 7 (7-methylguanosine, m7G) joined to the first transcribed nucleotide 5’ to 5’, through a bridge of three phosphates, where a normal backbone links 3’ to 5’. Three enzymes build it in order:

  1. An RNA triphosphatase removes the outermost (gamma) phosphate from the RNA’s 5’ end.
  2. A guanylyltransferase transfers GMP from GTP onto that end, through an intermediate in which GMP is bound to a lysine of the enzyme.
  3. A methyltransferase adds a methyl group to N7 of the guanine, taking it from S-adenosylmethionine (see nucleotide metabolism for GTP and its relatives). This is cap 0. In many organisms a further methyl on the 2’-OH of the first transcribed nucleotide makes cap 1.

The cap blocks 5’-to-3’ exonucleases. In the nucleus a cap-binding complex holds it and helps the mRNA’s export. In the cytoplasm the translation factor eIF4E grips it and recruits the machinery that starts translation.

The tail starts with a cut. Near the end of the gene the RNA carries the signal AAUAAA, read in mammals by a complex called CPSF; cross-linking with purified human CPSF (Chan and colleagues, 2014) showed its subunits CPSF30 and Wdr33 contacting AAUAAA directly. Another complex, CstF, binds a GU-rich stretch downstream. The RNA is cut between them, about 10 to 30 nucleotides past AAUAAA, by the nuclease CPSF73. Poly(A) polymerase then adds adenosines to the new 3’ end without copying any DNA. The tail is not in the gene.

Newly made tails in mammalian cells are usually given as about 200 to 250 adenosines; yeast tails are shorter. Tails also shrink with time: in L4-stage C. elegans larvae, sequencing of individual tails found a median of 57. Ask of any tail length which organism, and whether new RNA or the whole population was measured.

With poly(A)-binding proteins, the tail helps protect the mRNA and promote translation. It is not universal. The replication-dependent histone mRNAs of animals are cut but not normally polyadenylated; their 3’ ends form a stem-loop instead.

The 1977 surprise: genes in pieces

Until 1977 a gene was expected to be a continuous stretch of DNA, as in E. coli. The evidence against came from adenovirus 2, a DNA virus that grows in human cells, and from two groups.

At MIT, Susan Berget, Claire Moore and Phillip Sharp purified hexon mRNA, the message for the virus’s main coat protein, and paired it with mapped pieces of viral DNA. At Cold Spring Harbor, Louise Chow, Richard Gelinas, Thomas Broker and Richard Roberts paired viral DNA with RNA from the polysomes of infected cells. Both viewed RNA-DNA hybrids in the electron microscope. Chow and colleagues made R loops, in which RNA displaces one DNA strand and pairs with the other; Berget and colleagues used single-stranded DNA fragments.

If a gene were continuous, a message would pair with one unbroken stretch of DNA. The body of each message did. Its 5’ end did not. Chow and colleagues mapped a leader of 150 to 200 nucleotides, shared by many late mRNAs and copied from three separate regions of DNA, now called the tripartite leader. Berget and colleagues proposed that four separate segments may be joined to make hexon mRNA, which implied that RNA copied from the DNA between them is removed.

Split genes were soon found in animal cell genes, including globin genes. The 1993 Nobel Prize in Physiology or Medicine went to Roberts and Sharp for their discoveries of split genes. Introns are not junk: some carry sequences that control splicing or transcription.

The spliceosome’s chemistry

Most human introns of the major (U2-type) class follow the GU-AG rule: GU at the 5’ end, AG at the 3’ end. A third site lies inside the intron, usually a few dozen nucleotides upstream of the 3’ end: the branch point, which contains an adenosine. Between it and the AG lies a run of pyrimidines.

Splicing is two transesterifications, each swapping one phosphodiester bond for another. No bond is lost overall, so the chemistry itself needs no ATP; the spliceosome spends ATP on assembling and rearranging itself.

  1. The 2’-OH of the branch-point adenosine attacks the phosphate at the 5’ splice site. The upstream exon is freed, and the intron’s 5’ G is now joined to the branch A by a 2’-5’ bond, forming a loop with a tail: the lariat.
  2. The freed 3’-OH of the upstream exon attacks the phosphate at the 3’ splice site. The exons are joined and the lariat intron is released.

The spliceosome is built anew on each intron from five small nuclear ribonucleoproteins (snRNPs: U1, U2, U4, U5 and U6, each a small RNA with proteins) and many other proteins. U1 snRNA finds the 5’ splice site by base pairing. U2 pairs with the branch-point sequence, leaving the branch A bulged out. The U4/U6.U5 tri-snRNP then joins, U1 and U4 leave, and the helicase Brr2 unwinds U4 from U6 so that U6 can pair with U2 and the 5’ splice site. Cryo-EM structures of yeast and human spliceosomes show U6 snRNA positioning the catalytic metal ions, so the active site is largely RNA, held and rearranged by proteins.

That fits a 1982 result: Thomas Cech’s laboratory showed that the intron of Tetrahymena ribosomal RNA removes itself with no protein. This group I intron uses an external guanosine, not a branch A, as the first attacking group. Self-splicing group II introns, found in organelles and bacteria, form a lariat as the spliceosome does.

A second, minor spliceosome, built with U11, U12, U4atac and U6atac, removes a small class of introns, well under 1 percent of those in human genes.

One gene, several proteins, and what a splice mutation does

Because splicing chooses which pieces to join, one gene can make several mRNAs. Wang and colleagues (2008) classed the patterns they found in human transcriptomes into skipped exons, alternative 5’ or 3’ splice sites, mutually exclusive exons and retained introns, and estimated that 92 to 94 percent of human genes undergo alternative splicing. Regulation of these choices belongs to eukaryotic gene regulation.

The same logic predicts mutations. A change that weakens a splice site most often causes the exon to be skipped. Less often a cryptic splice site, a sequence not normally used, takes over, or the intron is kept. A cloned beta-thalassemia allele with a change at the GT of intron 2, expressed in HeLa cells, gave a main RNA that kept the first 47 nucleotides of intron 2 and a minor one that joined exon 1 to exon 3. Since 47 = 3 x 15 + 2, the first shifts the frame.

The worked problem uses one rule: a change in coding length keeps the downstream frame only if it is a multiple of three.

frame kept if (change in coding length) / 3 leaves remainder 0

Worked problem. An imagined human gene has four exons. Exon 1 is 150 nucleotides: a 60-nucleotide 5’ UTR, then 90 coding nucleotides that begin with the AUG. Exon 2 is 87, exon 3 is 100, and exon 4 is 400 (it holds the stop codon and the 3’ UTR). Introns 1, 2 and 3 are 900, 1,500 and 2,000. Ignore the cap and tail.

  1. Mature mRNA. 150 + 87 + 100 + 400 = 737 nucleotides. (Adding the introns, 900 + 1,500 + 2,000 = 4,400, gives the 5,137 of the pre-mRNA.)
  2. A 5’ splice-site mutation of intron 2 causes exon 2 to be skipped. In vertebrates an exon is often recognised as a unit by the sites at both its ends, so losing one site can drop the whole exon. Length: 737 - 87 = 650. Since 87 = 3 x 29, the frame is kept and the protein loses 29 amino acids.
  3. Exon 3 is skipped instead. Length: 737 - 100 = 637. Since 100 = 3 x 33 + 1, the frame shifts by one base from the exon 2 to exon 4 junction on, and a premature stop codon in exon 4 is likely, though not certain without the sequence. In mammalian cells, a premature stop in the last exon usually escapes nonsense-mediated decay that depends on exon junctions (covered in the genetic code and translation), so a truncated protein could be made.
  4. Intron 1 is retained. Length: 737 + 900 = 1,637. Such RNAs are often kept in the nucleus or degraded.
Splicing builder: what does the mature mRNA look like?⚙ original · interactive
What the spliceosome does

Rule to keep: the downstream frame is kept only if the change in coding length is a multiple of three. This is a sketch of the logic for an imagined gene, not a prediction for a real one; the sequence decides whether a stop codon appears.

Design problem

A patient variant changes intron 3 position +5 (the G of GU is +1) from G to A, in a gene expressed in muscle. Design a minigene experiment that tests whether the variant changes splicing, predict the band sizes for each outcome, and say what the result can and cannot tell you.

One way to do it

Clone exon 3, intron 3 and exon 4 into a splicing reporter plasmid between two vector exons (see hacking the cell), in two versions differing only at +5. Transfect each into a cell line, ideally muscle-derived, and run RT-PCR with primers in the vector exons.

With invented lengths (100 nucleotides of vector exon between the primers, exon 3 of 120, exon 4 of 150, intron 3 of 800): normal splicing gives 100 + 120 + 150 = 370 bp, skipping exon 3 gives 250, and retaining intron 3 gives 1,170. A nearby cryptic 5’ site gives a band slightly off 370 that only sequencing can resolve.

Controls: empty vector (100 bp); wild-type minigene; a known splice-null change, such as +1 G to A of the same intron; no reverse transcriptase, to rule out plasmid DNA as template. Sequence every band.

Limits: the minigene lacks the rest of the gene, and the cell line may lack splicing factors of muscle, so the result is evidence, not proof. Confirm in patient RNA if it is available. The assay describes splicing; it is not a clinical reading of the variant.

How we measure it

R-loop and RNA-DNA hybrid mapping by electron microscopy

Pair a purified RNA with its DNA under conditions that favour RNA-DNA hybrids, then view the molecules in the electron microscope. Paired stretches and the loops of unpaired DNA between them show which parts of the DNA are present in the RNA and which are left out.

Minigene splicing assay

Clone an exon with its flanking intron sequence, wild type or variant, into a reporter plasmid, put it into cultured cells, and read the spliced RNA by RT-PCR with primers in the flanking exons. The band sizes, confirmed by sequencing, tell you whether the exon was included, skipped, or joined at a new site.

RNA sequencing with splice-junction reads

Sequence short fragments of cDNA made from mRNA and map them to the genome. Reads that span two exons record which exons were joined, so the share of messages that include a given exon can be estimated across tissues.

Finish line

Test yourself

8 questions

Practice and MCAT-style, with explained answers.

Next · Lesson 18

The Genetic Code and Translation

Sixty-four codons, twenty amino acids, three stops: how was the dictionary read, and who reads it in the cell?

Continue →

← Lesson 16 · Transcription: From DNA to RNA

Yeast spliceosome after the first step (C* complex) — the snRNAs at the active site
Yeast spliceosome after the first step (C* complex) — the snRNAs at the active site ● LIVE · RCSB PDB 5MQ0
open on RCSB PDB ↗
Loading live structure 5MQ0 from RCSB PDB…

Find U2, U5 and U6 snRNA near the branch region. In this stalled complex the first step has happened and the second has not.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
Human U1 snRNP — the particle that reads the 5' splice site
Human U1 snRNP — the particle that reads the 5' splice site ● LIVE · RCSB PDB 3CW1
open on RCSB PDB ↗
Loading live structure 3CW1 from RCSB PDB…

Look for the 5' end of U1 snRNA, the stretch that base-pairs with the 5' splice site, and the Sm protein ring around the RNA.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
Cap-binding protein eIF4E holding 7-methyl-GDP — how the cap is read
Cap-binding protein eIF4E holding 7-methyl-GDP — how the cap is read ● LIVE · RCSB PDB 1EJ1
open on RCSB PDB ↗
Loading live structure 1EJ1 from RCSB PDB…

The methylated guanine lies stacked between two aromatic side chains. This grip on the cap is one way the cap helps start translation.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
Tetrahymena group I intron, P4-P6 domain — a folded RNA that helps catalyse its own splicing
Tetrahymena group I intron, P4-P6 domain — a folded RNA that helps catalyse its own splicing ● LIVE · RCSB PDB 1GID
open on RCSB PDB ↗
Loading live structure 1GID from RCSB PDB…

A piece of the self-splicing intron folds into packed helices much as a protein folds, which is part of how an RNA can hold a catalytic site.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
Phillip Sharp

Phillip Sharp

With Susan Berget and Claire Moore at MIT, mapped adenovirus 2 hexon mRNA against viral DNA and found that its 5' end comes from DNA segments far from the body of the gene; shared the 1993 Nobel Prize in Physiology or Medicine with Richard Roberts for the discovery of split genes.

In the mid-1970s Phillip Sharp's laboratory at the Massachusetts Institute of Technology was studying adenovirus 2, a virus whose late genes make large amounts of a few mRNAs, among them the mRNA for hexon, the main protein of the virus coat. Susan Berget, Claire Moore and Sharp purified hexon mRNA by gel electrophoresis and let it pair with mapped pieces of the viral DNA, then looked at the hybrids in the electron microscope. A gene copied straight through should give one continuous stretch of RNA paired with one stretch of DNA. The body of the hexon mRNA did pair that way, but its 5' end did not: those sequences came from places on the genome well upstream. In 1977 they reported in PNAS that four separate viral RNA segments may be joined to make the mature message. The paper gave the field a physical picture of a gene in pieces, and the question that followed was how the pieces are joined.

✓ verbatim Berget, Moore & Sharp, Proc. Natl. Acad. Sci. USA 74:3171 (1977), 'Spliced segments at the 5' terminus of adenovirus 2 late mRNA'; Nobel Prize in Physiology or Medicine 1993 facts page, nobelprize.org.
Richard Roberts

Richard Roberts

With Louise Chow, Richard Gelinas and Thomas Broker at Cold Spring Harbor Laboratory, mapped a leader of 150 to 200 nucleotides, built from three separated regions of DNA, on many late adenovirus 2 mRNAs; shared the 1993 Nobel Prize in Physiology or Medicine with Phillip Sharp.

At Cold Spring Harbor Laboratory, Richard Roberts's group was also working on the late mRNAs of adenovirus 2. Louise Chow, Richard Gelinas, Thomas Broker and Roberts used R-loop mapping: under the right conditions an RNA can displace one strand of a DNA duplex and pair with the other, leaving a loop that can be seen and measured in the electron microscope. They paired viral DNA with RNA taken from polysomes, the ribosome-loaded messages of infected cells. The main part of each message lined up with its gene, but the 5' end did not. A leader of 150 to 200 nucleotides was shared by many late mRNAs, and it came from three separate regions of the DNA. The paper, published in Cell in 1977, was titled "An amazing sequence arrangement at the 5' ends of adenovirus 2 messenger RNA." Read with the MIT paper, it showed two independent teams reaching the same arrangement by related methods, which made the result hard to dismiss as an artifact.

✓ verbatim Chow, Gelinas, Broker & Roberts, Cell 12:1 (1977), 'An amazing sequence arrangement at the 5' ends of adenovirus 2 messenger RNA'; Roberts, Nobel autobiography, nobelprize.org.
Joan Steitz

Joan Steitz

Her laboratory at Yale identified small nuclear ribonucleoproteins (snRNPs) and, with Michael Lerner and colleagues, proposed in 1980 that the U1 snRNP takes part in splicing, noting the complementarity of its 5' end to splice junctions, by base pairing with the 5' splice site.

Once genes were known to be in pieces, something had to find the joints. Joan Steitz's laboratory at Yale had a tool for the search: antibodies from patients with the autoimmune disease systemic lupus erythematosus. Some of these patients make antibodies against their own nuclear particles, and Michael Lerner and Steitz used them to pull down small nuclear RNAs bound to proteins, the snRNPs. In 1980 Lerner, Boyle, Mount, Wolin and Steitz, in a paper titled as a question, asked in Nature whether snRNPs are involved in splicing. They pointed out that the 5' end of U1 snRNA can base-pair with the conserved sequence found at the 5' ends of introns, and proposed that snRNPs take part in splicing. Later work from many laboratories confirmed that U1 recognizes the 5' splice site by base pairing and that U2, U4, U5 and U6 snRNPs join it in the spliceosome. The idea that small RNAs, not only proteins, read the intron's signals shaped the field that followed.

✓ verbatim Lerner, Boyle, Mount, Wolin & Steitz, Nature 283:220 (1980), 'Are snRNPs involved in splicing?'
Thomas Cech

Thomas Cech

His laboratory showed that the intron of Tetrahymena ribosomal RNA splices itself out with no protein, using a guanosine nucleotide as a cofactor; this was evidence that RNA can be a catalyst. He shared the 1989 Nobel Prize in Chemistry with Sidney Altman.

Thomas Cech's laboratory at the University of Colorado was studying how the ciliate Tetrahymena removes an intron from its ribosomal RNA precursor. The working assumption was that a protein enzyme did the cutting and joining, so the group set out to find it. The intron came out even in control reactions meant to lack the enzyme. To rule out a protein carried along from the cell, Kruger, Grabowski, Zaug and colleagues made the precursor RNA in a test tube by transcribing cloned DNA, then added only salts, magnesium and a guanosine nucleotide. The intron still excised itself and closed into a circle. Their 1982 paper in Cell called this self-splicing. Because the RNA made no protein and had met none from Tetrahymena, the reaction had to be carried out by the RNA itself. The result, with Sidney Altman's work on ribonuclease P, opened the study of RNA enzymes and gave weight to the idea that RNA may sit at the catalytic heart of the spliceosome too.

✓ verbatim Kruger, Grabowski, Zaug, Sands, Gottschling & Cech, Cell 31:147 (1982), 'Self-splicing RNA: autoexcision and autocyclization of the ribosomal RNA intervening sequence of Tetrahymena'; Nobel Prize in Chemistry 1989 facts page, nobelprize.org.
Test yourself
Question 1 of 8 Score 0 / 0

How is the 7-methylguanosine of an mRNA cap joined to the first transcribed nucleotide?
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

  • Yeast spliceosome after the first step (C* complex) — the snRNAs at the active site — RCSB PDB 5MQ0 · coordinates CC0 · Cryo-EM, Saccharomyces cerevisiae; Fica et al., Nature 2017.
  • Human U1 snRNP — the particle that reads the 5' splice site — RCSB PDB 3CW1 · coordinates CC0 · X-ray crystal structure; Pomeranz Krummel et al., Nature 2009.
  • Cap-binding protein eIF4E holding 7-methyl-GDP — how the cap is read — RCSB PDB 1EJ1 · coordinates CC0 · X-ray crystal structure, mouse eIF4E; Marcotrigiano et al., Cell 1997.
  • Tetrahymena group I intron, P4-P6 domain — a folded RNA that helps catalyse its own splicing — RCSB PDB 1GID · coordinates CC0 · X-ray crystal structure; Cate et al., Science 1996.

Simulations

  • Splicing builder: what does the mature mRNA look like? · Original interactive — Biochemistrypedia (Deppmann Lab).

Scientist profiles & quotes

  • Phillip Sharp — Berget, Moore & Sharp, Proc. Natl. Acad. Sci. USA 74:3171 (1977), 'Spliced segments at the 5' terminus of adenovirus 2 late mRNA'; Nobel Prize in Physiology or Medicine 1993 facts page, nobelprize.org.
  • Richard Roberts — Chow, Gelinas, Broker & Roberts, Cell 12:1 (1977), 'An amazing sequence arrangement at the 5' ends of adenovirus 2 messenger RNA'; Roberts, Nobel autobiography, nobelprize.org.
  • Joan Steitz — Lerner, Boyle, Mount, Wolin & Steitz, Nature 283:220 (1980), 'Are snRNPs involved in splicing?'
  • Thomas Cech — Kruger, Grabowski, Zaug, Sands, Gottschling & Cech, Cell 31:147 (1982), 'Self-splicing RNA: autoexcision and autocyclization of the ribosomal RNA intervening sequence of Tetrahymena'; Nobel Prize in Chemistry 1989 facts page, nobelprize.org.

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

Animated explainer

A Gene Is Longer Than Its Message · 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

A gene is like a recipe with extra pages of nonsense stuck in the middle. The cell first copies the whole thing, nonsense and all. Then it cuts out the extra pages, glues the real pages together, puts a cover on the front and a long tail on the back. Only then does the recipe leave the nucleus to be cooked into a protein.
High school

In cells with a nucleus, the first RNA copy of a gene is longer than the final message. Parts called introns are cut out and the parts called exons are joined, in a step called splicing. The cell also adds a special cap to the front end and a tail of many A's to the back end. The cap and tail protect the message and help it leave the nucleus and be read by ribosomes. Because a cell can join exons in different ways, one gene can make more than one protein.
College

RNA polymerase II transcripts are processed while they are made. When the RNA is about 25 to 30 nucleotides long, a triphosphatase, a guanylyltransferase and a methyltransferase add a 7-methylguanosine joined 5' to 5' by a triphosphate. Downstream of AAUAAA, CPSF and CstF direct a cut, and poly(A) polymerase adds a tail without a template. The spliceosome, built from U1, U2, U4, U5 and U6 snRNPs and many proteins, removes introns in two transesterifications: the branch-point adenosine's 2'-OH attacks the 5' splice site, making a lariat, then the freed exon's 3'-OH attacks the 3' splice site. Alternative splicing (skipped exons, alternative 5' or 3' sites, mutually exclusive exons, retained introns) lets one gene make several mRNAs, and splice-site mutations can cause skipping, cryptic-site use or retention.
Graduate

The spliceosome is assembled anew on each intron and remodelled by RNA helicases at each stage; Brr2 unwinds U4 from U6 to activate it. Structures of yeast and human spliceosomes show U6 snRNA, paired with U2 and the intron, positioning catalytic metal ions, so the core chemistry resembles that of self-splicing group II introns, with proteins as scaffold and remodellers. In the yeast C* complex, the branch helix is rotated about 75 degrees relative to complex C, which makes room for the 3' splice site in the second step. At the 3' end, a reconstituted mammalian system of 14 polypeptides was sufficient for cleavage of an SV40 late RNA, and in that system cleavage required RBBP6 acting through an interface with the CPSF73 nuclease. Processing is coupled to transcription and export: in HEK293T cells export factors were loaded mainly on spliced but uncleaved RNAs, and depleting Chtop shifted poly(A)-site choice. An open question is how, in a given tissue, the many weak signals of exon definition, splicing enhancers and silencers, and elongation rate add up to the measured inclusion level of one exon.
An amazing title

Scientific titles are usually flat. The 1977 Cell paper from Richard Roberts's group at Cold Spring Harbor was called "An amazing sequence arrangement at the 5' ends of adenovirus 2 messenger RNA." The arrangement was a leader of 150 to 200 nucleotides, shared by many late viral mRNAs and built from three regions of DNA that lie apart on the genome. The MIT group of Phillip Sharp reported a matching result for hexon mRNA in PNAS the same year. The word in the title was a fair summary of how odd the finding looked.

— Chow, Gelinas, Broker & Roberts, Cell 12:1 (1977).
Forty-seven extra nucleotides

A beta-globin allele cloned from a fetus diagnosed with beta-zero thalassemia had a single base change: the GT at the start of intron 2 was altered. When Tom Maniatis's laboratory expressed the cloned allele in HeLa cells, it gave two abnormal RNAs. The main one kept the first 47 nucleotides of intron 2, because a cryptic 5' splice site inside the intron was used. A smaller amount joined exon 1 straight to exon 3. One base change, two wrong messages: an early and clear case of how a splice-site mutation acts.

— Treisman, Proudfoot, Shander & Maniatis, Cell (1982), doi 10.1016/0092-8674(82)90452-4.
38,016 ways to read one fly gene

The Drosophila gene Dscam encodes a cell-surface protein used by nerve cells. Several of its exons come as clusters of alternatives, of which each mRNA keeps only one. Multiplying the number of choices in each cluster gave Dietmar Schmucker and colleagues 38,016 possible mRNAs from a single gene. That is a count of what the gene could make, not of the isoforms found in any one cell. Even so, it shows how far mutually exclusive exons can stretch one stretch of DNA.

— Schmucker et al., Cell 101:671 (2000).

 ↑ Top