Molecular Biology: Chapter 7 — RNA processing

7.1 The cap #

7.1.1 An inverted guanosine added during initiation #

Pol II transcripts are capped at about 20–30 nucleotides, because capping enzymes bind the serine-5-phosphorylated CTD. A triphosphatase removes the γ phosphate. A guanylyltransferase adds GMP from GTP in a 5′–5′ linkage that ordinary exonucleases do not recognise. A methyltransferase methylates N7 of that guanine, giving cap 0. Animals often add a 2′-O-methyl on the first transcribed nucleotide as well.

In the nucleus the cap-binding complex, CBP80 and CBP20, binds the cap. In the cytosol, eIF4E binds it for initiation (Chapter 8). Pol I and pol III products do not receive this structure.

7.2 Splicing #

The study of Splicing provides fundamental principles governing cellular architecture, molecular interactions, and physiological regulation.

🔬 Key Illustrated Scientific Figure

Full-Screen Zoom

Click to enlarge on this page

7.2.1 GU–AG, the spliceosome, and the lariat #

Most nuclear pre-mRNA introns begin with GU and end with AG, with a polypyrimidine tract and a branch-point adenosine upstream of the AG. A minority use a different set of sites and the U12-dependent spliceosome. GU–AG is the rule, not a law without exceptions.

The spliceosome contains five snRNPs — U1, U2, U4, U5, and U6 — plus many proteins. U1 pairs with the 5′ splice site. U2 pairs at the branch point so the branch adenosine is bulged, helped by U2AF. The U4/U6/U5 tri-snRNP joins, then rearrangements release U1 and U4, and U6 pairs with U2 and the 5′ site. The two chemical steps do not consume ATP. First the 2′-OH of the branch adenosine attacks the 5′ splice site, freeing the upstream exon and making the 2′–5′ branched lariat. Then the 3′-OH of that exon attacks the 3′ splice site, joining the exons and releasing the lariat, which is debranched and degraded.

7.2.2 Alternative splicing #

One pre-mRNA can give several mRNAs: a cassette exon included or skipped, mutually exclusive exons, alternative 5′ or 3′ sites, or a retained intron. SR proteins generally promote nearby sites through exonic enhancers. The chemistry does not change.

FGFR2 is a clear case. Two mutually exclusive exons, IIIb and IIIc, encode different pieces of the ligand-binding domain, one used in epithelial cells and the other in mesenchymal cells, and ligand preference changes with the exon. Alternative splicing is one reason about 20,000 genes yield more polypeptides than that count. Not every detected isoform is abundant or functional.

7.2.3 Group I and group II introns, as a contrast #

Some introns splice without a spliceosome. Group I introns, of which the Tetrahymena rRNA intron is the historical case, use the 3′-OH of a free guanosine as the first nucleophile. There is no branch point and no lariat. Group II introns use an internal branch adenosine and a lariat.

🔬 Key Illustrated Scientific Figure

Full-Screen Zoom

Click to enlarge on this page

Nuclear pre-mRNA splicing is spliceosomal. Group II splicing is the chemical relative and the likely source of both spliceosomal introns and the snRNAs. Neither pathway removes the introns of a human pre-mRNA.

7.3 The 3′ end #

7.3.1 Cleavage and polyadenylation #

The 3′ end of a typical mRNA is cut and tailed, not stopped by a bacterial hairpin. In mammals AAUAAA lies about 10–30 nucleotides upstream of the cut, and a U-rich or GU-rich element lies downstream. CPSF recognises AAUAAA. The tail is about 200 nucleotides in mammals and about 70–80 in yeast, with nuclear poly(A)-binding protein helping to set the length. In the cytosol, cytoplasmic poly(A)-binding protein protects the tail and binds eIF4G, circularising the mRNA.

Serine-2 phosphorylation of the CTD recruits this machinery. After the cut, an exonuclease chases the polymerase that has run on, so pol II termination is a consequence of cleavage. Replication-dependent histone mRNAs end in a stem-loop, processed with the U7 snRNP, and are not polyadenylated.

7.4 RNA editing #

7.4.1 ApoB C-to-U, and a different system that exists #

Editing changes a transcript relative to its gene. In the intestine, the deaminase APOBEC1, with its cofactors, changes one cytidine of apolipoprotein B mRNA to uridine. The codon CAA, glutamine, becomes UAA, stop. Translation ends at about 48% of the full chain, producing ApoB-48 for chylomicrons. The liver does not edit that site and makes ApoB-100 for VLDL and LDL. A mooring sequence in the RNA positions the enzyme. The gene is not mutated, and no splice site is moved. Editing is not splicing.

A different and much more extensive editing occurs in trypanosome mitochondria, where guide RNAs direct insertion and deletion of uridines. It is named so that apoB is not taken to be the only meaning of the word, and it is not developed here. A separate mammalian activity, ADAR, deaminates adenosine to inosine in double-stranded RNA. Inosine pairs as guanosine. That reaction is editing too, and it is not the apoB reaction.

7.5 Export #

7.5.1 A finished messenger, then the pore #

An mRNA leaves as a particle whose proteins report completed processing. The cap-binding complex marks the 5′ end, poly(A)-binding protein the 3′ end, and successful splicing deposits an exon-junction complex about 20–24 nucleotides upstream of each exon–exon join. The receptor NXF1–NXT1 hands this particle to the pore. Bulk mRNA export does not use the Ran gradient that moves most proteins.

Unspliced pre-mRNA is normally retained and, if never finished, degraded. The exon-junction complex has a second job in the cytosol. A stop codon more than about 50 nucleotides upstream of the last exon–exon junction is read as premature, because a junction complex still sits downstream, and nonsense-mediated decay destroys the mRNA. Many nonsense and frameshift alleles therefore yield no stable mRNA and no truncated protein.

7.6 Key terms, summary, and self-check #

Key terms. Cap; snRNP; branch point; lariat; alternative splicing; CPSF; ApoB-48; exon-junction complex; nonsense-mediated decay.

Summary. The cap is an inverted, N7-methylated guanosine added co-transcriptionally to pol II RNA. The spliceosome removes GU–AG introns by two transesterifications and releases a lariat; alternative splicing, as at FGFR2, is the same chemistry with a different site choice. Group II introns share the lariat and not the spliceosome; group I introns use free guanosine and make no lariat. Cleavage downstream of AAUAAA and template-independent polyadenylation finish most mRNAs; histone mRNAs use a stem-loop instead. ApoB editing is one C-to-U change that creates a stop in the intestine. Only a capped, spliced, polyadenylated particle is a normal export substrate, and a stop left upstream of a junction complex marks the mRNA for decay.

Self-check.

1. What is unusual about the cap linkage, and which CTD mark recruits the capping enzymes?

2. Which hydroxyl makes the first attack in splicing, and what is the lariat?

3. How does a group II intron resemble spliceosomal splicing, and how does a group I intron differ?

4. Which single-base change produces ApoB-48, in which tissue, and why is it not a splice?

5. Why does a stop more than about 50 nucleotides upstream of the last exon junction destabilise an mRNA?

—

—

—