 |
Although DNA stores the information for protein synthesis and RNA carries out the instructions encoded in DNA, most biological activities are carried out by proteins. The accurate synthesis of proteins thus is critical to the proper functioning of cells and organisms. We saw in Chapter 3 that the linear order of amino acids in each protein determines its three-dimensional structure and activity. For this reason, assembly of amino acids in their correct order, as encoded in DNA, is the key to production of functional proteins.
Messenger RNA (mRNA) is translated into protein by the joint action of transfer RNA (tRNA) and the ribosome, which is composed of numerous proteins and two major ribosomal RNA (rRNA) molecules. [Adapted from A. J. F. Griffiths et al., 1993, An Introduction to Genetics Analysis, 5th ed., W. H. Freeman.]
):
Messenger RNA (mRNA) carries the genetic information copied from DNA in the form of a series of three-base code “words,” each of which specifies a particular amino acid.
Transfer RNA (tRNA) is the key to deciphering the code words in mRNA. Each type of amino acid has its own type of tRNA, which binds it and carries it to the growing end of a polypeptide chain if the next code word on mRNA calls for it. The correct tRNA with its attached amino acid is selected at each step because each specific tRNA molecule contains a three-base sequence that can base-pair with its complementary code word in the mRNA.
Ribosomal RNA (rRNA) associates with a set of proteins to form ribosomes. These complex structures, which physically move along an mRNA molecule, catalyze the assembly of amino acids into protein chains. They also bind tRNAs and various accessory molecules necessary for protein synthesis. Ribosomes are composed of a large and small subunit, each of which contains its own rRNA molecule or molecules.
Translation is the whole process by which the base sequence of an mRNA is used to order and to join the amino acids in a protein. The three types of RNA participate in this essential protein-synthesizing pathway in all cells; in fact, the development of the three distinct functions of RNA was probably the molecular key to the origin of life. How each RNA carries out its specific task is discussed in this section, while the biochemical events in protein synthesis and the required protein factors are described in the final section of the chapter.
RNA contains ribonucleotides of adenine, cytidine, guanine, and uracil; DNA contains deoxyribonucleotides of adenine, cytidine, guanine, and thymine. Because 4 nucleotides, taken individually, could represent only 4 of the 20 possible amino acids in coding the linear arrangement in proteins, a group of nucleotides is required to represent each amino acid. The code employed must be capable of specifying at least 20 words (i.e., amino acids).
If two nucleotides were used to code for one amino acid, then only 16 (or 42) different code words could be formed, which would be an insufficient number. However, if a group of three nucleotides is used for each code word, then 64 (or 43) code words can be formed. Any code using groups of three or more nucleotides will have more than enough units to encode 20 amino acids. Many such coding systems are mathematically possible. However, the actual genetic code used by cells is a triplet code, with every three nucleotides being “read” from a specified starting point in the mRNA. Each triplet is called a codon. Of the 64 possible codons in the genetic code, 61 specify individual amino acids and three are stop codons. Table 4-2 shows that most amino acids are encoded by more than one codon. Only two — methionine and tryptophan — have a single codon; at the other extreme, leucine, serine, and arginine are each specified by six different codons. The different codons for a given amino acid are said to be synonymous. The code itself is termed degenerate, which means that it contains redundancies.
Synthesis of all protein chains in prokaryotic and eukaryotic cells begins with the amino acid methionine. In most mRNAs, the start (initiator) codon specifying this aminoterminal methionine is AUG. In a few bacterial mRNAs, GUG is used as the initiator codon, and CUG occasionally is used as an initiator codon for methionine in eukaryotes. The three codons UAA, UGA, and UAG do not specify amino acids but constitute stop (terminator) signals that mark the carboxyl terminus of protein chains in almost all cells. The sequence of codons that runs from a specific start site to a terminating codon is called a reading frame. This precise linear array of ribonucleotides in groups of three in mRNA specifies the precise linear sequence of amino acids in a protein and also signals where synthesis of the protein chain starts and stops.
If translation of the mRNA sequence shown begins at two different upstream start sites (not shown), then two overlapping reading frames are possible; in this case, the codons are shifted one base to the right in the lower frame. As a result, different amino acids are encoded by the same nucleotide sequence. Many instances of such overlaps have been discovered in viral and cellular genes of prokaryotes and eukaryotes. It is theoretically possible for the mRNA to have a third reading frame.
). The vast majority of mRNAs,
however, can be read in only one frame because stop codons encountered in the
other two possible reading frames terminate translation before a functional
protein is produced. Another unusual coding arrangement occurs be- cause of
frameshifting. In this case the protein-synthesizing
machinery may read four nucleotides as one amino acid and then continue reading
triplets, or it may back up one base and read all succeeding triplets in the new
frame until termination of the chain occurs. These frameshifts are not common
events, but a few dozen such instances are known.
| Universal | Unusual | ||
|---|---|---|---|
| Codon | Code | Code | Occurrence* |
| UGA | Stop | Trp | Mycoplasma, Spiroplasma, mitochondria of many species |
| CUG | Leu | Thr | Mitochondria in yeasts |
| UAA, UAG | Stop | Gln | Acetabularia, Tetrahymena, Paramecium, etc. |
| UGA | Stop | Cys | Euplotes |
“Unusual code” is used in nuclear genes of the listed organisms and in mitochondrial genes as indicated.
SOURCE: S. Osawa et al., 1992, Microbiol. Rev. 56:229.
Having described the genetic code, we briefly recount how it was deciphered — one of the great triumphs of modern biochemistry. The underlying experimental work was carried out largely with cell-free bacterial extracts containing all the necessary components for protein synthesis except mRNA (i.e., tRNAs, ribosomes, amino acids, and the energy-rich nucleotides ATP and GTP).
Addition of such a synthetic mRNA to a bacterial extract that contained all the components necessary for protein synthesis except mRNA resulted in synthesis of polypeptides composed of a single type of amino acid as indicated. [See M. W. Nirenberg and J. H. Matthei, 1961, Proc. Nat’l. Acad. Sci. USA 47:1588.]
(a) When a synthetic mRNA with alternating A and C residues was added to a protein-synthesizing bacterial extract, the resulting polypeptide contained alternating threonine and histidine residues. This finding is compatible with the two alternative codon assignments shown. (Note that alternating residues yield the same sequence of triplets regardless of which reading frame is chosen.) (b) To determine which codon assignment shown in (a) is correct, a second mRNA consisting of AAC repeats was tested. This mRNA, which can be read in three frames, yielded the three types of polypeptides shown. Since only the ACA codon was common to both experiments, it must encode threonine; thus CAC must encode histidine in (a). The assignments AAC = asparagine (Asn) and CAA = glutamine (Gln) were derived from additional experiments. [See H. G. Korana, 1968, reprinted in Nobel Lectures: Physiology or Medicine (1963 – 1970), Elsevier (1973), p. 341.]
).
[Poly(G) did not work in this type of experiment because it assumes an unusable
stacked structure that is not translated well.] Next, synthetic mRNAs composed
of alternating bases were used. The results of these experiments not only
revealed more codons but also demonstrated that codons are three bases long. The
example of this approach illustrated in Figure
4-23 led to identification of ACA as the codon for threonine and CAC
for histidine. Similar experiments with many such mixed polynucleotides revealed
a substantial part of the genetic code.
Marshall Nirenberg and his collaborators prepared 20 ribosome-free bacterial extracts containing all possible aminoacyl-tRNAs (tRNAs with an amino acid attached). In each sample, a different amino acid was radioactively labeled (green); the other 19 amino acids were bound to tRNAs but were unlabeled. Aminoacyl-tRNAs and trinucleotides passed through a nitrocellulose filter (left), but ribosomes were retained by the filter (center) and would bind trinucleotides and their cognate tRNAs (right). Each possible trinucleotide was tested separately for its ability to attract a specific tRNA by adding it with ribosomes to samples from each of the 20 aminoacyl-tRNA mixtures. The sample was then filtered. If the added trinucleotide caused the radiolabeled aminoacyl-tRNA to bind to the ribosome, then radioactivity would be detected on the filter (a positive test); otherwise, the label would pass through the filter (a negative test). By synthesizing and testing all possible trinucleotides, the researchers were able to match all 20 amino acids with one or more codons (e.g., phenylalanine with UUU as shown here). [See M. W. Nirenberg and P. Leder, 1964, Science 145:1399.]
). In all, 61 of the 64 possible trinucleotides
were found to code for a specific amino acid; the trinucleotides UAA, UGA, and
UAG did not encode amino acids.
Although synthetic mRNAs were useful in deciphering the genetic code, in vitro protein synthesis from these mRNAs is very inefficient and yields polypeptides of variable size. Successful in vitro synthesis of a naturally occurring protein was achieved first when mRNA from bacteriophage F2 (a virus) was added to bacterial extracts, leading to formation of the coat, or capsid, protein (the “packaging” protein that covers the virus particle). Studies with such natural mRNAs established that AUG encodes methionine at the start of almost all proteins and is required for efficient initiation of protein synthesis, while the three trinucleotides (UAA, UGA, and UAG) that do not encode any amino acid act as stop codons, necessary for precise termination of synthesis.
The next step in understanding the flow of genetic information from DNA to protein was to determine how the nucleotide sequence of mRNA is converted into the amino acid sequence of protein. This decoding process requires two types of adapter molecules: tRNAs and enzymes called aminoacyl-tRNA synthetases. First we describe the role of tRNAs in decoding mRNA codons, and then examine how synthetases recognize tRNAs.
First, an aminoacyl-tRNA synthetase couples a specific amino acid to its corresponding tRNA. Second,a three-base sequence in the tRNA (the anticodon) base-pairs with a codon in the mRNA specifying the attached amino acid. If an error occurs in either step, the wrong amino acid may be incorporated into a polypeptide chain.
).
As studies on tRNA proceeded, 30 – 40 different tRNAs were identified in bacterial cells and as many as 50 – 100 in animal and plant cells. Thus the number of tRNAs in most cells is more than the number of amino acids found in proteins (20) and also differs from the number of codons in the genetic code (61). Consequently, many amino acids have more than one tRNA to which they can attach (explaining how there can be more tRNAs than amino acids); in addition, many tRNAs can attach to more than one codon (explaining how there can be more codons than tRNAs). As noted previously, most amino acids are encoded by more than one codon, requiring some tRNAs to recognize more than one codon.
(a) The primary structure of yeast alanine tRNA (tRNAAla), the first such sequence determined. This molecule is synthesized from the nucleotides A, C, G, and U, but some of the nucleotides, shown in red, are modified after synthesis: D = dihydrouridine, I = inosine, T = thymine, Ψ = pseudouridine, and m = methyl group. Although the exact sequence varies among tRNAs, they all fold into four base-paired stems and three loops. The partially unfolded molecule is commonly depicted as a cloverleaf. Dihydrouridine is nearly always present in the D loop; likewise, thymidylate, pseudouridylate, cytidylate, and guanylate are almost always present in the TΨCG loop. The triplet at the tip of the anticodon loop base-pairs with the corresponding codon in mRNA. Attachment of an amino acid to the acceptor arm yields an aminoacyl-tRNA. (b) Computergenerated three-dimensional model of the generalized backbone of all tRNAs. Note the L shape of the molecule. [Part (a) see R. W. Holly et al., 1965, Science 147:1462; part (b) from J. G. Arnez and D. Moras, 1997, Trends Biochem. Sci. 22:211.]
). The four stems are short double helices
stabilized by Watson-Crick base pairing; three of the four stems have loops
containing seven or eight bases at their ends, while the remaining, unlooped
stem contains the free 3′ and 5′ ends of the chain. Three
nucleotides termed the anticodon,
located at the center of one loop, can form base pairs with the three
complementary nucleotides forming a codon in mRNA. As discussed later, specific
aminoacyl-tRNA synthetases recognize the surface structure of each tRNA for a
specific amino acid and covalently attach the proper amino acid to the unlooped
amino acid acceptor stem. The 3′ end of all tRNAs
has the sequence CCA, which in most cases is added after synthesis and
processing of the tRNA are complete. Viewed in three dimensions, the folded tRNA
molecule has an L shape with the anticodon loop and acceptor stem forming the
ends of the two arms (Figure 4-26b).
).
), and a tRNA with G or U in the wobble position can
read two codons. Although A is theoretically possible in the wobble
position of the anticodon, it is almost never found in nature.
). For example, the phenylalanine codons UUU
and UUC (5′ → 3′) are both recognized by the tRNA
that has GAA (5′ → 3′) as the anticodon. In
fact, any two codons of the type NNPyr (N = any
base; Pyr = pyrimidine) encode a single amino
acid and are decoded by a single tRNA with G in the first (wobble) position of
the anticodon.
The heavy black lines indicate the bonds by which the nitrogenous bases attach to the 1′ carbon of ribose (C1).
). A tRNA with inosine
in the wobble position thus can recognize the corresponding mRNA codons with A,
C, or U in the third (wobble) position (see Figure 4-27). For this reason, inosine-containing tRNAs are heavily
employed in translation of the synonymous codons that specify a single amino
acid. For example, four of the six codons for leucine have a 3′ A, C,
or U (see Table 4-2); these four codons
are all recognized by the same tRNA (3′-GAI-5′), which has
inosine in the wobble position of the anticodon (and thus recognizes CUA, CUC,
and CUU), and uses a G·U pair in position 1 to recognize the UUA
codon.
Each of these enzymes recognizes one kind of amino acid and all the cognate tRNAs that recognize codons for that amino acid. The two-step aminoacylation reaction requires energy from the hydrolysis of ATP. The equilibrium of the overall reaction favors the indicated products because the pyrophosphate (PPi) released in step 1 is converted to inorganic phosphate (Pi) by a pyrophosphatase. The 3′ end of all tRNAs, to which the amino acid attaches, has the sequence CCA. Class I synthetases (purple) attach the amino acid to the 2′ hydroxyl of the terminal adenylate in tRNA; class II synthetases (green) attach the amino acid to the 3′ hydroxyl. (Ad = adenine; Cyt = cytosine.)
). Each of the 20 different synthetases recognizes
one amino acid and all its compatible, or
cognate, tRNAs. These coupling enzymes link an amino acid
to the free 2′ or 3′ hydroxyl of the adenosine at the
3′ terminus of tRNA molecules by a two-step ATP-requiring reaction
(Figure 4-29). About half the
aminoacyl-tRNA synthetases transfer the aminoacyl group to the 2′
hydroxyl of the terminal adenosine (class I), and about half to the 3′
hydroxyl (class II). In this reaction, the amino acid is linked to the tRNA by a
high-energy bond and thus is said to be activated. The energy
of this bond subsequently drives the formation of peptide bonds between adjacent
amino acids in a growing polypeptide chain. The equilibrium of the
aminoacylation reaction is driven further toward activation of the amino acid by
hydrolysis of the high-energy phosphoanhydride bond in pyrophosphate. The
overall reaction is
Shown here are the outlines of the three-dimensional structures of the two synthetases. The tRNA shown between them as a ribbon diagram will bind to either and is a slightly modified version of tRNAAsp. It is used as an illustration of common surface interactions between tRNA and class I and II enzymes. Sites on the opposite sides of this modified tRNAAsp make contacts with the two enzymes: the blue balls show contacts with the class II AspRS; those that make contact with class I ArgRS are indicated by yellow balls. The synthetases are shown positioned away from the tRNA for clarity, but the fit of the surfaces at close range is obvious. The ability of ArgRS to interact with the noncognate tRNAAsp is lost when residue G37 in tRNAAsp is methylated, a normal modification that occurs in vivo. However, the shape and binding sites of this modified tRNA are characteristic of class I and class II interactions with tRNAs. This molecular graphic picture was produced using the DRAWNA program. [Adapted from M. Sissler et al., 1997, Nucl. Acids Res. 25:4899; courtesy of R. Giegé.]
).
Because some amino acids are so similar structurally, aminoacyl-tRNA synthetases
sometimes make mistakes. These are corrected, however, by the enzymes
themselves, which check the fit in the binding pockets and facilitate
deacylation of any misacylated tRNAs. This crucial function helps guarantee that
a tRNA delivers the correct amino acid to the protein-synthesizing
machinery.
The ability of aminoacyl-tRNA synthetases to recognize their correct cognate tRNAs is just as important to the accurate translation of the genetic code as codon-anticodon pairing. Once a tRNA is loaded with an amino acid, codon-anticodon pairing directs the tRNA into the proper ribosome site; if the wrong amino acid is attached to the tRNA, an error in protein synthesis results.
As noted already, each aminoacyl-tRNA synthetase can aminoacylate all the different tRNAs whose anticodons correspond to the same amino acid. Therefore, all these cognate tRNAs must have a similar binding site, or “identity element,” that is recognized by the synthetase. One approach for studying the identity elements in tRNAs that are recognized by aminoacyl-tRNA synthetases is to produce synthetic genes that encode tRNAs with normal and various mutant sequences by techniques discussed in Chapter 7. The normal and mutant tRNAs produced from such synthetic genes then can be tested for their ability to bind purified synthetases.
Very probably no single structure or sequence completely determines a specific tRNA identity. However, some important structural features of several E. coli tRNAs that allow their cognate synthetases to recognize them are known. Perhaps the most logical identity element in a tRNA molecule is the anticodon itself. Experiments in which the anticodons of methionine tRNA (tRNAMet) and valine tRNA (tRNAVal) were interchanged showed that the anticodon is of major importance in determining the identity of these two tRNAs. In addition, x-ray crystallographic analysis of the complex between glutamine aminoacyl-tRNA synthetase (GlnRS) and glutamine tRNA (tRNAGln) showed that each of the anticodon bases neatly fits into a separate, specific “pocket” in the three-dimensional structure of GlnRS. Thus this synthetase specifically recognizes the correct anticodon.
The 67 known tRNA sequences in E. coli were compared by computer analysis. The conserved nucleotides in different tRNAs that recognize the same amino acid are shown as yellow circles in the left drawing, with the tRNA chain in blue. Increasing size indicates increasing conservation of a base at a given position. It is clear that nucleotides in the anticodon loop and in the acceptor stem are most often similar when a particular amino acid must be recognized. This appreciation is heightened by results shown in the right drawing. Here, nucleotides that have been experimentally demonstrated to have a role in identity (selection of an amino acid by an ARS-tRNA complex) are shown as yellow circles. In this case, the circle size indicates the relative frequency that a given position acts as an identity element. [From W. H. McClain, 1993, J. Mol. Biol. 234:257; also see L. D. H. Schulman and J. Abelson, 1988, Science 240:1590.]
). Figure 4-31 shows the extent of base
sequence conservation in E. coli tRNAs that become linked to
the same amino acid. Identity elements are found in several regions,
particularly the end of the acceptor arm. A simple case is presented by
tRNAAla: a single G·U base pair (G3·U70) in
the acceptor stem is necessary and sufficient for recognition of this tRNA by
its cognate aminoacyl-tRNA synthetase. Solution of the three-dimensional
structure of additional complexes between aminoacyl-tRNA synthetases and their
cognate tRNAs should provide a clear understanding of the rules governing the
recognition of tRNAs by specific synthetases.
If the many components that participate in translating mRNA had to interact in free solution, the likelihood of simultaneous collisions occurring would be so low that the rate of amino acid polymerization would be very slow. The efficiency of translation is greatly increased by the binding of the mRNA and the individual aminoacyl-tRNAs to the most abundant RNA-protein complex in the cell — the ribosome. This two-part machine directs the elongation of a polypeptide at a rate of three to five amino acids added per second. Small proteins of 100 – 200 amino acids are therefore made in a minute or less. On the other hand, it takes 2 to 3 hours to make the largest known protein, titin, which is found in muscle and contains 30,000 amino acid residues. The machine that accomplishes this task must be precise and persistent.
With the aid of the electron microscope, ribosomes were first discovered as discrete, rounded structures prominent in animal tissues secreting large amounts of protein; initially, however, they were not known to play a role in protein synthesis. Once reasonably pure ribosome preparations were obtained, radiolabeling experiments showed that radioactive amino acids first were incorporated into growing polypeptide chains associated with ribosomes before appearing in finished chains.
In all cells, each ribosome consists of a large and a small subunit. The two subunits contain rRNAs of different lengths, as well as a different set of proteins. All ribosomes contain two major rRNA molecules (dark red) — 23S and 16S rRNA in bacteria, 28S and 18S rRNA in eukaryotes — and one or two small RNAs (light red). The proteins are named L1, L2, etc., and S1, S2, etc., depending on whether they are found in the large or the small subunit.
). The ribosomal subunits
and the rRNA molecules are commonly designated in svedbergs (S), a measure of
the sedimentation rate of suspended particles centrifuged under standard
conditions (Chapter 3). The lengths
of the rRNA molecules, the quantity of proteins in each subunit, and
consequently the sizes of the subunits differ in prokaryotic and eukaryotic
cells. (The small and large rRNAs are about 1500 and 3000 nucleotides long in
bacteria and about 1800 and 5000 nucleotides long in humans.) Perhaps of more
interest than these differences are the great structural and functional
similarities among ribosomes from all species. This consistency is another
reflection of the common evolutionary origin of the most basic constituents of
living cells.
In general, the length and position of the stem-loops are very similar in all species, although the exact sequence varies from species to species. The most highly conserved regions are represented as red lines, and the numbered stem-loops unique to prokaryotes are preceded by a P. Eukaryotic small (18S) rRNAs exhibit a generally similar pattern of stem-loops, although, as with prokaryotes, a few are unique. [Adapted from E. Huysmans and R. DeWachter, 1987, Nucl. Acids Res. 14:73].
). An even larger number
of regularly positioned stem-loops have been demonstrated in large rRNAs. All
the ribosomal proteins have been identified and their sequences determined, and
many have been shown to bind specific regions of rRNA. It seems clear that the
fundamental protein-synthesizing machinery in all present-day cells arose only
once and has been modified about a common plan during evolution.
(a) This model shows the shapes of the large (blue) and small (yellow) subunits of the ribosome with three aminoacyl-tRNAs (pink, green, yellow) superimposed at the A, P, and E sites. The roles of these tRNA-binding sites during protein synthesis are discussed later. Chemical cross-linking experiments have demonstrated that the mRNA (orange beads) passes close to the anticodon loops of the tRNAs and that the nascent polypeptide chain is buried in the tunnel in the large ribosomal subunit that begins within 10 – 15 Å of the 3′ aminoacylated end of the tRNAs. The tunnel termination site on the ribosome surface has also been accurately mapped. (b) Large panel shows a field of 70S ribosomes. Small panels (left) show cryoelectron microscopy images of a single 70S ribosome, small (30S) subunit, and large (50S) subunit. Small panels (right) show computer-derived averages of many dozens of images in the same orientation. Cryoelectron microscopy is carried out on unstained samples of ribosomes or subunits flash frozen as “vitreous ices” (without ice crystals) in a very thin layer of water (Chapter 5). Individual images are analyzed by computer projections. [See R. K. Agrawal et al., Cell, in press; J. Frank, 1995, Nature 356:441; J. Frank et al., 1995, Biochem. Cell Biol. 73:757. Courtesy of J. Frank.]
). These studies not only have
determined the dimensions and overall shape of the ribosomal subunits, but also
have localized the positions of tRNAs bound to the ribosome during protein chain
elongation. Powerful chemical experiments have also helped unravel the complex
interactions between proteins and RNAs. In a technique called footprinting, for example, ribosomes
are treated with chemical reagents that modify single-stranded RNA unprotected
by binding either to protein or to another RNA. If the total sequence of the RNA
is known, then the location of the modified nucleotides can be located within
the molecule. (This technique, which is also useful for locating protein-binding
sites in DNA, is described in Chapter
10.) Thus the overall structure and function of ribosomes during
protein synthesis is finally, after 40 years, yielding to successful
experiments. How these results aid in understanding the specific steps in
protein synthesis is described in the next section.
The AUG codon for methionine is the most common start codon, specifying the amino acid at the NH2-terminus of a protein chain. Three codons function as stop codons and specify no amino acids.
A reading frame, the uninterrupted sequence of codons in mRNA from a specific start codon to a stop codon, is translated into the linear sequence of amino acids in a protein.
).
). The
anticodon can base-pair with its corresponding codon or codons in
mRNA.Because of nonstandard interactions, a tRNA may base-pair with more than one mRNA codon, and conversely, a particular codon may base-pair with multiple tRNAs.
). This reaction activates the amino acid, so it
can participate in peptide-bond formation.
). All
ribosomes are composed of a small and a large subunit. Each contains
numerous different proteins and one rRNA (small or large). The large
subunit also contains one accessory RNA (5S).
).
