10.4: Details of mRNA Processing in Eukaryotic nuclei

Eukaryotic mRNA primary transcripts undergo extensive processing, including splicing, capping and polyadenylation. The steps described here are considered in order of (sometimes overlapping!) occurrence. We begin with splicing—an mRNA phenomenon.

10.4.1. Spliceosomal Introns

The coding regions of bacterial genes are continuous. The discovery of eukaryotic split genes with introns and exons came as quite a surprise. It seemed incongruous for evolution to have stuck irrelevant DNA in the middle of coding DNA! For their discovery of split genes, Richard J. Roberts and Phillip A. Sharp shared the Nobel Prize for Physiology or Medicine in 1993. In fact, all but a few eukaryotic genes are split, and some have two or more introns separating bits of coding DNA, the exons. Figure 10.14 summarizes splicing to remove introns in pre-mRNAs and to splice exons (in the right order!) to make a mature, functional mRNA.

Figure 10.14: Eukaryotic precursor mRNAs are encoded by split genes containing noncoding introns and coding exons. After transcription of a pre-mRNA (1), splicing removes introns and ligates exons into a contiguous polypeptide coding sequence (2).

snRNPs are particles composed of RNA and ribonuclear proteins. They bind to specific sites in an mRNA and then direct a sequential series of cuts and ligations (the splicing) necessary to process the mRNAs. The process was reminiscent of the way in which movies were edited (see the photo of an early splicing “machine” at the top of this chapter), hence the term splicing to describe mRNA processing. Figure 10.15 below highlights the role of small ribonuclear proteins (snRNPs) in splicing.

Figure 10.15: Splicing a eukaryotic pre-mRNA involves association of the primary transcript (pre-mRNA) with five snRNPs (small nuclear RiboNuclear Proteins) that catalyze cleavage (hydrolysis) of the pre-RNA at 5’ and 3’ splice sites (bottom of image), to be followed by ligation of the 3’ to 5’ exon ends.

When snRNPs (often pronounced “snurps”) bind to a pair of splice sites flanking a pre-mRNA intron, they form a spliceosome, which completes the splicing, including the removal of intermediate lariat structures of the intron. The last step is to ligate exons into a continuous mRNA, with all its codons intact and in the right order—an impressive trick for some pre-mRNAs with as many as fifty introns! Spliceosome action is summarized in Figure 10.16 and in the “mRNA Splicing” animation at one of the following links.

Figure 10.16: Binding of snRNPs to a pre-mRNA leads brings 5’ and 3’ splice sites close to each other, catalyzing the hydrolysis at the sites and forming a lariat structure of the intron remnants.

194-2 The Discovery of Split Genes

195 mRNA Splicing

10.4.2 Specific Nuclear-Body Proteins Facilitate snRNP Assembly and Function

Recall the organization of nuclei facilitated by nuclear bodies. Nuclear speckles are associated regions of high mRNA transcription, processing, and splicing. (Check out the animation at Nuclear Bodies and Transcription for a 3D localization of markers of nuclear body activity.)

Cajal bodies (CBs) and Gems are nuclear bodies that are similar in size and have related functions in assembling spliceosomal snRNPs. Some splicing defects correlate with mutations in the coil protein associated with Cajal bodies; others correlate with mutations in SMN proteins normally associated with Gems. One hypothesis was that CBs and Gems interact in SnRNP and spliceosome assembly—but how? Consider the results of an experiment in which antibodies to coilin and the SMN protein were localized in undifferentiated and differentiated neuroblastoma cells (Figure 10.17).

Figure 10.17: Light (A,B) micrographs of undifferentiated (A) and differentiated (B) neuroblastoma cells show that Cajal bodies and Gems aggregate during differentiation. Fluorescent coilin and SMN antibodies localize separately in undifferentiated cells (C), but together after in differentiation (D). Since mutations in coilin and SMN proteins are associated with splicing defects, their co-localization after differentiation suggests that Cajal bodies and Gems have become involved in snRNP function.

Panels A and C above are undifferentiated cells in culture; panels B and D are cells that were stimulated to differentiate. In the fluorescence micrographs at the right, arrows point to fluorescent nuclear bodies. In panel C, the green, fluorescent antibodies to coilin (arrows) localize to CBs and the red fluorescent antibodies to SMN protein (arrowhead) binds to Gems—as expected. But in panel D, the two antibodies co-localize, suggesting that the CBs and Gems aggregate in the differentiated cells. This would explain the need for both coilin and SMN proteins to produce functional snRNPs.

The CBs and gems may aggregate in differentiated cells because of an observed increase in expression of the SMN protein, producing more Gems and driving the association with CBs. This and similar experiments demonstrate that different nuclear bodies do have specific functions. They are not random structural artifacts; rather, they have evolved to organize nuclear activities in time and space in ways that are essential to the cell.

10.4.3. Group I and Group II Self-Splicing Introns

While eukaryotic spliceosomal introns are spliced using snRNPs as described above, Group I or Group II introns are removed by different mechanisms.

Group I introns interrupt mRNA and tRNA genes in bacteria and in the genomes of mitochondria and chloroplasts. They are also occasionally found in bacteriophage genes, but rarely in nuclear genes (and then only in lower eukaryotes). Group I introns are self-splicing! Thus, they are themselves ribozymes. They do not require a spliceosome with its snRNPs or other proteins for splicing activity. Instead, they fold into a secondary stem-loop structure that positions catalytic nucleotides at appropriate splice sites to excise their own introns and to religate the exons.

Group II introns are found in chloroplast and mitochondrial rRNA, mRNA, tRNA and some bacterial mRNAs. These can be quite long. They form complex stem-loop tertiary structures and self-splice, at least in a test tube! However, Group II introns encode proteins required for their own splicing in vivo. Like spliceosomal introns, they form a lariat structure at an Aresidue branch site. All of this suggests that the mechanism of spliceosomal intron splicing evolved from that of Group II introns.

10.4.4. So Why Splicing?

The puzzle implied by this question is, of course, why do higher organisms have split genes in the first place? While the following discussion can apply to all splicing, it will reference mainly spliceosomal introns. Here are some answers to the question, “Why splicing?”

• Introns in nuclear genes are typically longer (much longer!) than exons. Because introns are noncoding (i.e., non-informational) making them large targets for mutation. In effect, most noncoding DNA—including introns—can buffer the ill effects of random mutations.
• You may recall that gene duplication on one chromosome and the loss of a copy from its homologue arise from unequal recombination (nonhomologous crossing over). It occurs when similar DNA sequences align during synapsis of meiosis. In an organism that inherits a chromosome with both gene copies, the duplicate can accumulate mutations if the other retains original function. The diverging gene then becomes part of a pool of selectable DNA, the grist of evolution. Descendants of organisms that inherit the duplicated genes have diversified the gene pool, again increasing the potential for evolution and species diversity.
• Unequal recombination can also occur between similar sequences (e.g., in introns) in the same or different genes, resulting in a sharing of exons between genes. After unequal recombination between the introns that flank an exon, one gene will acquire another exon while the other will lose it. Once again, if an organism retains a copy of the participating genes with original function, the organism can make the required protein and survive. Meanwhile, the gene with the extra exon may produce a similar protein, but one with a new structural domain and function.
• Just like a complete duplicate gene, one with a new exon that adds a new function to an old gene has been entered in the pool of selectable DNA. By creating proteins with different overall functions that nonetheless share at least one domain and one common function, the phenomenon of exon shuffling increases species diversity!

An example discussed earlier involves calcium-binding proteins that regulate many cellular processes. Structurally related calcium (\(\rm Ca^{++}\)) binding domains are common to many otherwise structurally and functionally unrelated proteins. Consider exon shuffling in the unequal crossover (nonhomologous recombination) shown in Figure 10.18.

Figure 10.18: Exon shuffling, the transfer or copying of an exon from one gene to another can occur by unequal recombination (cross-over) between different (nonhomologous) genes due to some sequence similarity within introns. The common sequences can align during synapsis in meiosis. If unequal crossing over occurs within introns or non-coding DNA (1) it creates a new gene, in this case one with three exons. Transcription and splicing of a pre-mRNA from this new gene produce a new protein.

In this example, regions of strong similarity exist in different (nonhomologous) introns in the same gene. These regions align during synapsis of meiosis. Unequal crossing over between the genes inserts exon C into one of the genes. The other gene loses the exon (not shown in the illustration).

In sum, introns are buffers against deleterious mutations, and they are equally valuable as potential targets for gene duplication and exon shuffling. This makes introns key players in creating genetic diversity, the hallmark of evolution.

196-2 Origin of Introns

197 Introns Evolution: What was selected here?

10.4.5 5-Prime Capping

Using GTP as a precursor, a methyl guanosine CAP added 5’-to-5’ to an mRNA functions, in part, to help mRNAs leave the nucleus and associate with ribosomes. The CAP is added to an exposed 5’ end, even as transcription and splicing are still in progress. A capping enzyme places a methylated guanosine residue at the 5’ end of the mature mRNA. Figure 10.19 shows the 5’ CAP structure (the check marks are 5’-3’ linked nucleotides).

Figure 10.19: Capping of mRNA results from catalysis of a 5′-to-5’ condensation reaction linking the 5’ end of an mRNA to a methylated guanidine triphosphate (\(\rm CH_3-GTP\)).

10.4.6 3-Prime Polyadenylation

A poly(A) polymerase catalyzes the addition of adenine monophosphates (AMPs) to the 3’ end of most eukaryotic mRNAs, even before any splicing is complete. Polyadenylation requires ATP and can add several hundred AMPs to the 3’ terminus of an mRNA. The enzyme binds to an -A-A-U-A-A- sequence near the 3’ end of an mRNA and starts adding the AMPs. Polyadenylation after transcription termination is illustrated in Figure 10.20.

Figure 10.20: mRNA polyadenylation is the addition of multiple adenine nucleotides to the 3’ end of an mRNA, catalyzed by the enzyme poly(A) polymerase. Other than histone mRNAs, most eukaryotic mRNAs are polyadenylated.

The result of polyadenylation is a 3’ poly(A) tail whose functions include assisting in the transit of mRNAs from the nucleus and regulating the half-life of mRNAs in the cytoplasm. The poly(A) tail shortens each time a ribosome finishes translating the mRNA.

198 mRNA 5' Capping and 3' Polyadenylation

CHALLENGE

Bacterial mRNAs have been found with very short poly(A) tails. Would you expect poly(A) tails on bacterial mRNAs to function the same way they do in eukaryotic cells? Explain your expectation.