Voet D., Voet Ju.G. Biochemistry

Подождите немного. Документ загружается.

The cryo-EM–based image of the yeast 80S ribosome

(Fig. 32-36), determined at 15 Å resolution by Andrej Sali,

Günter Blobel, and Frank, reveals that there is a high de-

gree of structural conservation between eukaryotic and

prokaryotic ribosomes. Although the yeast 40S subunit

(which consists of a 1798-nt 18S rRNA and 32 proteins)

contains an additional 256 nt of RNA and 11 proteins rela-

tive to the E. coli 30S subunit (Table 32-8; 15 of the E. coli

proteins are homologous to those of yeast), both exhibit a

similar division into head, neck, body, and platform (Fig.

32-36b vs Figs. 32-32a and 32-34b). Many of the differences

between these two small ribosomal subunits are accounted

for by the 40S subunit’s additional RNA and proteins, al-

though their homologous portions exhibit several distinct

conformational differences. Similarly, the yeast 60S subunit

(Fig. 32-35c; which consists of an aggregate of 3671 nt and

45 proteins) structurally resembles the considerably

smaller (Table 32-7) prokaryotic 50S subunit (Fig. 32-32b).

The yeast ribosome exhibits 16 intersubunit bridges, 12 of

which match the 12 that were observed in the X-ray struc-

ture of the T. thermophilus ribosome, a remarkable evolu-

tionary conservation that indicates the importance of these

bridges. Moreover, the tRNA that occupies the P site of the

yeast ribosome has a conformation that more closely re-

sembles that of the P-site tRNA in the T. thermophilus ri-

bosome than that of free tRNA

Phe

B. Polypeptide Synthesis: An Overview

Before we commence our detailed discussion of polypep-

tide synthesis, it will be helpful to outline some of its major

features.

a. Polypeptide Synthesis Proceeds from N-Terminus

to C-Terminus

The direction of ribosomal polypeptide synthesis was es-

tablished, in 1961 by Howard Dintzis, through radioactive

labeling experiments. He exposed reticulocytes that were

actively synthesizing hemoglobin to

H-labeled leucine for

times less than that required to make an entire polypeptide.

The extent to which the tryptic peptides from the soluble

(completed) hemoglobin molecules were labeled increased

Section 32-3. Ribosomes and Polypeptide Synthesis 1371

Figure 32-36 Cryo-EM–based image of the yeast 80S

ribosome at 15 Å resolution. (a) The ribosome shown in side

view analogous to Fig. 32-30 of the E. coli ribosome.The small

(40S) subunit is yellow, the large (60S) subunit is cyan, and the

tRNA that is bound in the ribosomal P site is green. Portions of

this ribosome that are not homologous to the RNA or proteins

of the E. coli ribosome are shown in gold for the small subunit

and magenta for the large subunit. (b) The computationally

isolated small subunit shown in interface view analogous to the

left panel of Fig. 32-32a. (c) The computationally isolated large

subunit shown in interface view analogous to the left panel of

Fig. 32-32b. [Courtesy of Joachim Frank, State University of New

York at Albany.]

See the Interactive Exercise.

(a) (b)

(c)

JWCL281_c32_1338-1428.qxd 8/4/10 4:45 PM Page 1371

with their proximity to the C-terminus (Fig. 32-37). Incom-

ing amino acids must therefore be appended to a growing

polypeptide’s C-terminus; that is, polypeptide synthesis pro-

ceeds from N-terminus to C-terminus.

b. Ribosomes Read mRNA in the 5¿S3¿ Direction

The direction in which the ribosome reads mRNAs was

determined through the use of a cell-free protein synthe-

sizing system in which the mRNA was poly(A) with a 3¿-

terminal C.

Such a system synthesizes a poly(Lys) that has a C-terminal

Asn.

This, together with the knowledge that AAA and AAC

code for Lys and Asn and the polarity of polypeptide syn-

thesis, indicates that the ribosome reads mRNA in the 5¿S

3¿ direction. Since mRNA is synthesized in the 5¿S3¿ di-

rection, this accounts for the observation that, in prokary-

otes, ribosomes initiate translation on nascent mRNAs

(Section 31-3).

c. Active Translation Occurs on Polyribosomes

Electron micrographs, as Rich discovered, reveal that ri-

bosomes engaged in protein synthesis are tandemly

arranged on mRNAs like beads on a string (Fig. 31-27) to

...

Lys Lys Lys Lys Lys Asn COO

5

3

AAAC

...

form assemblies known as polyribosomes (polysomes).

Polysomes arise because once an active ribosome has

cleared its initiation site on an mRNA, a second ribosome

can initiate at that site.

Cryo-EM–based image reconstruction of E. coli

polysomes, by Ulrich Hartl and Wolfgang Baumeister, has

revealed that adjacent ribosomes are densely packed along

the mRNA and have staggered or pseudohelical arrange-

ments (Fig. 32-38). The mRNA is located along the inside

of these assemblies, the tRNA entrance sites are accessible,

and the polypeptide exit tunnel on each ribosome faces

the cytosol. Modeling suggests that such an arrangement

maximizes the distance between the nascent polypeptide

chains exiting neighboring ribosomes. This reduces the

ability of these incompletely folded polypeptides to non-

specifically aggregate and hence increases the yield of na-

tively folded protein.

d. Chain Elongation Occurs by the Linkage of the

Growing Polypeptide to the Incoming tRNA’s Amino

Acid Residue

During polypeptide synthesis, amino acid residues are

sequentially added to the C-terminus of the nascent, ribo-

somally bound polypeptide chain. If the growing polypep-

tide is released from the ribosome by treatment with high

salt concentrations, its C-terminal residue is invariably es-

terified to a tRNA molecule as a peptidyl–tRNA:

The nascent polypeptide must therefore grow by being

transferred from the peptidyl–tRNA to the incoming

aminoacyl–tRNA to form a peptidyl–tRNA with one more

O OH

–

CH R

n–1

CH R

Peptidyl–tRNA

Adenine

tRNA

1372 Chapter 32. Translation

Figure 32-37 Demonstration that polypeptide synthesis

proceeds from the N-terminus to the C-terminus. Rabbit

reticulocytes were incubated with [

H]leucine. The curves show

the distribution of [

H]Leu among the tryptic peptides from the

 subunit of soluble rabbit hemoglobin after the indicated

incubation times.The numbers on the horizontal axis are peptide

identifiers arranged from N-terminus to C-terminus. [After

Dintzis, H.M., Proc. Natl. Acad. Sci. 47, 255 (1961).]

13 24 1 17 3 9 18 12 27

Incubation time

60 min

7 min

4 min

Relative amount of

Peptide number, β chain

C-terminusN-terminus

JWCL281_c32_1338-1428.qxd 9/7/10 2:22 PM Page 1372

residue (Fig. 32-39). Apparently, the ribosome has at least

two tRNA-binding sites: the so-called peptidyl or P site,

which binds the peptidyl–tRNA, and the aminoacyl or A

site, which binds the incoming aminoacyl–tRNA (Fig. 32-

39). Consequently, after the formation of a peptide bond,

the newly deacylated P-site tRNA must be released and re-

placed by the newly formed peptidyl–tRNA from the A

site, thereby permitting a new round of peptide bond for-

mation. The finding by Knud Nierhaus that each ribosome

can bind up to three deacylated tRNAs but only two

aminoacyl–tRNAs indicates, however, that the ribosome

has a third tRNA-binding site, the exit or E site, which

transiently binds the outgoing deacylated tRNA. All three

sites, as we have seen (Fig. 32-34), extend over both riboso-

mal subunits.

The details of the chain elongation process are dis-

cussed in Section 32-3D. Chain initiation and chain termi-

nation, which are special processes, are examined in Sec-

tions 32-3C and 32-3F, respectively. In all of these sections

we shall first consider the process of interest in E. coli and

then compare it with the analogous eukaryotic activity.

C. Chain Initiation

a. fMet Is the N-Terminal Residue of Prokaryotic

Polypeptides

The first indication that the initiation of translation re-

quires a special codon, since identified as AUG (and, in

prokaryotes, occasionally GUG), was the observation that

almost half of the E. coli proteins begin with the otherwise

uncommon amino acid Met. This was followed by the dis-

covery of a peculiar form of Met–tRNA

Met

in which the

Met residue is N-formylated:

HC CHNH

S CH

O tRNA

Met

N-Formylmethionine–tRNA

Met

(fMet–tRNA

Met

)

Section 32-3. Ribosomes and Polypeptide Synthesis 1373

Figure 32-38 Cryo–EM–based image of an E. coli polysome.

The mRNA (which is mostly occluded) is represented by a red

line, the small ribosomal subunits are yellow, the large subunits

are blue-gray, and the red cones point to the polypeptide exit

tunnel on each large subunit.The assembly shown has a pseudo-

helical arrangement of ribosomes in which the center-to-center

distance between adjacent ribosomes averages ⬃230 Å.

Polysomes with somewhat different although equally densely

packed arrangements of ribosomes have also been observed.

[Courtesy of Ulrich Hartl and Wolfgang Baumeister, Max Planck

Institute of Biochemistry, Martinsreid, Germany.]

Figure 32-39 Ribosomal peptidyl transferase

reaction forming a peptide bond. The ribosome

catalyzes the nucleophilic attack of the amino

group of the aminoacyl–tRNA in the A site on

the peptidyl–tRNA ester in the P site, thereby

forming a new peptide bond and transferring

the nascent polypeptide to the A-site tRNA,

while displacing the P-site tRNA.

n–1

...

tRNA

(n)

P site A site P site A site

n+1

tRNA

(n+1)

tRNA

(n)

–1

...

tRNA

Peptidyl–tRNA Aminoacyl–tRNA Uncharged tRNA Peptidyl–tRNA

+1)(n

JWCL281_c32_1338-1428.qxd 8/4/10 4:45 PM Page 1373

–

araB

alE

lacI

lacZ

Q phage replicase

X174 phage A protein

R17 phage coat protein

Ribosomal S12

Ribosomal L10

trpE

trp leader

Initiation

codon

3 end of 16S rRNA

AUUCCUCCACUAG

–

3

5

The N-formylmethionine residue (fMet) already has an

amide bond and can therefore only be the N-terminal

residue of a polypeptide. In fact, polypeptides synthesized

in an E. coli–derived cell-free protein synthesizing system

always have a leading fMet residue. fMet must therefore be

E. coli’s initiating residue.

The tRNA that recognizes the initiation codon,

tRNA

Met

(Fig. 32-40), differs from the tRNA that carries

internal Met residues, tRNA

Met

, although they both recog-

nize the same codon. In E. coli, uncharged (deacylated)

tRNA

Met

is first aminoacylated with methionine by the same

MetRS that charges tRNA

Met

.The resulting Met–tRNA

Met

is specifically N-formylated to yield fMet–tRNA

Met

in an

enzymatic reaction that employs N

-formyltetrahydrofo-

late (Section 26-4D) as its formyl donor. The formylation

enzyme does not recognize Met–tRNA

Met

. The X-ray

structures of E. coli tRNA

Met

and yeast tRNA

Phe

(Fig.32-11b)

are largely similar but differ conformationally in their ac-

ceptor stems and anticodon loops. Perhaps these structural

differences permit tRNA

Met

to be distinguished from

tRNA

Met

in the reactions of chain initiation and elongation

(see Section 32-3D).

E. coli proteins are post-translationally modified by a de-

formylase, which hydrolytically deformylates their fMet

residue,and, in many proteins, by the subsequent removal of

the resulting N-terminal Met.This processing usually occurs

on the nascent polypeptide, which accounts for the obser-

vation that mature E. coli proteins all lack fMet.

b. Base Pairing between mRNA and the 16S rRNA

Helps Select the Translational Initiation Site

AUG codes for internal Met residues as well as the ini-

tiating Met residue of a polypeptide. Moreover, mRNAs

usually contain many AUGs (and GUGs) in different read-

ing frames. Clearly, a translational initiation site must be

specified by more than just an initiation codon. This occurs

in two ways: (1) the masking of AUGs that are not initia-

tion codons by mRNA secondary structure; and (2) inter-

actions between the mRNA and the 16S rRNA that select

the initiating AUG as we now discuss.

The 16S rRNA contains a pyrimidine-rich sequence at

its 3¿ end. This sequence, as John Shine and Lynn Dalgarno

pointed out in 1974, is partially complementary to a purine-

rich tract of 3 to 10 nucleotides, the Shine–Dalgarno

sequence, that is centered ⬃10 nucleotides upstream from

the start codon of nearly all known prokaryotic mRNAs

(Fig. 32-41). Base pairing interactions between an mRNA’s

1374 Chapter 32. Translation

Figure 32-40 Nucleotide sequence of E. coli shown

in cloverleaf form. The shaded boxes indicate the significant

differences between this initiator tRNA and noninitiator tRNAs

such as yeast tRNA

Ala

(Fig. 32-8). [After Woo, N.M., Roe, B.A.,

and Rich,A., Nature 286, 346 (1980).]

tRNA

Met

Figure 32-41 Some translational initiation sequences

recognized by E. coli ribosomes. The mRNAs are aligned

according to their initiation codons (blue shading).Their

Shine–Dalgarno sequences (red shading) are complementary,

counting G  U pairs, to a portion of the 16S rRNA’s 3¿ end

E. coli tRNA

Met

CGGCC

GUCGG

CGAG

GCUC

m G

17a

Unpaired bases

3 consecutive

•C base pairs

Extra base,

no D

(green shading; below). [After Steitz, J.A., in Chambliss, G.,

Craven, G.R., Davies, J., Davis, K., Kahan, L., and Nomura, M.

(Eds.), Ribosomes. Structure, Function and Genetics, pp. 481–482,

University Park Press (1979).]

JWCL281_c32_1338-1428.qxd 8/4/10 4:45 PM Page 1374

Shine–Dalgarno sequence and the 16S rRNA apparently per-

mit the ribosome to select the proper initiation codon. Thus

ribosomes with mutationally altered anti-Shine–Dalgarno

sequences often have greatly reduced ability to recognize

natural mRNAs, although they efficiently translate mRNAs

whose Shine–Dalgarno sequences have been made comple-

mentary to the altered anti-Shine–Dalgarno sequences.

Moreover,treatment of ribosomes with the bactericidal pro-

tein colicin E3 (produced by E. coli strains carrying the E3

plasmid), which specifically cleaves a 49-nucleotide frag-

ment from the 3¿ terminus of 16S rRNA, yields ribosomes

that cannot initiate new polypeptide synthesis but can com-

plete the synthesis of a previously initiated chain.

The X-ray structure of the 70S ribosome reveals, in

agreement with Fig. 32-30, that a 27-nt segment of the

mRNA is wrapped in a groove that encircles the neck of

the 30S subunit (Fig. 32-42). The mRNA codons in the A

and P sites are exposed on the interface side of the 30S sub-

unit (as in Fig. 32-34b), whereas its 5¿ and 3¿ ends are bound

in tunnels composed of RNA and protein. The mRNA’s

Shine–Dalgarno sequence, which is located near its 5¿ end,

is base paired, as expected, with the 16S RNA’s anti-

Shine–Dalgarno sequence, which is situated close to the E

site.The resulting double helical segment is accommodated

in a cleft formed by both RNA and protein elements of the

16S subunit’s head, neck, and platform (Fig. 32-26).

c. Prokaryotic Initiation Is a Three-Stage Process

That Requires the Participation of Soluble Protein

Initiation Factors

See Guided Exploration 28: Translational initiation Intact ribo-

somes do not directly bind mRNA so as to initiate polypep-

tide synthesis. Rather, initiation is a complex process in

which the two ribosomal subunits and fMet–tRNA

Met

as-

semble on a properly aligned mRNA to form a complex that

is competent to commence chain elongation. This assembly

process also requires the participation of protein initiation

factors that are not permanently associated with the ribo-

some. Initiation in E. coli involves three initiation factors

designated IF-1, IF-2, and IF-3 (Table 32-9). Their exis-

tence was discovered when it was found that washing small

ribosomal subunits with 1M ammonium chloride solution,

Section 32-3. Ribosomes and Polypeptide Synthesis 1375

Figure 32-42 Path of mRNA through the ribosomal 30S

subunit as viewed from its interface side. The 16S RNA is cyan,

and the 27-nt mRNA is represented in worm form with its

A- and P-site codons orange and red, the Shine–Dalgarno helix

(which includes a segment of 16S RNA) magenta, and its

remaining segments yellow. The S3, S4, and S5 proteins are

green, the S7, S11, and S12 proteins are purple, and the

remaining ribosomal proteins have been omitted for clarity. The

S3, S4, and S5 proteins, which in part form the tunnel through

which the mRNA enters the ribosome, may function as a helicase

to remove secondary structure from the mRNA that would

otherwise interfere with tRNA binding. [Courtesy of Gloria

Culver, Iowa State University. Based on an X-ray structure by

Harry Noller, University of California at Santa Cruz. PDBid

1JGO.]

Table 32-9 The Soluble Protein Factors of E. coli

Protein Synthesis

Number of

Factor Residues

Function

Initiation Factors

IF-1 71 Assists IF-3 binding

IF-2 890 Binds initiator

tRNA and GTP

IF-3 180 Releases mRNA and

tRNA from recycled

30S subunit and aids

new mRNA binding

Elongation Factors

EF-Tu 393 Binds aminoacyl–tRNA

and GTP

EF-Ts 282 Displaces GDP from

EF-Tu

EF-G 703 Promotes translocation

through GTP binding

and hydrolysis

Release Factors

RF-1 360 Recognizes UAA and

UAG Stop codons

RF-2 365 Recognizes UAA and

UGA Stop codons

RF-3 528 Stimulates RF-1/RF-2

release via GTP

hydrolysis

RRF 185 Together with EF-G,

induces ribosomal

dissociation to small

and large subunits

All E. coli translational factors are monomeric proteins.

JWCL281_c32_1338-1428.qxd 8/4/10 4:45 PM Page 1375

IF-1

30S subunit

IF-2

GTP

IF-2

GTP

3′5′

IF-2

•

GTP

•

fMet – tRNA

Met

IF-2

GDP

AUG

5′ 3′

AUG

Shine–Dalgarno

sequence

30S Initiation complex

50S subunit

AUG

A site

P site

5′ 3′

70S Initiation complex

IF-1

+ P

IF-3

IF-1

IF-3

fMet

mRNA

which removes the initiation factors but not the “perma-

nent” ribosomal proteins, prevents initiation.

The initiation sequence in E. coli ribosomes has three

stages (Fig. 32-43):

1. On completing a cycle of polypeptide synthesis, the

30S and 50S subunits are separated (Section 32-3Fa). IF-3

then binds to the 30S subunit so as to prevent the reassocia-

tion of the 50S subunit. The X-ray structure of the 30S sub-

unit in complex with the C-terminal domain of IF-3 (which

by itself prevents the association of the 30S and 50S sub-

units), determined by Yonath and François Franceschi, indi-

cates that IF-3 binds to the upper end of the platform (Fig.

32-26) on its solvent (back) side. Hence IF-3 does not func-

tion by physically blocking the binding of the 50S subunit.

2. mRNA and IF-2 in a ternary complex with GTP and

that is accompanied by IF-1 subsequently

bind to the 30S subunit in either order. Hence,

recognition must not be mediated by a

codon–anticodon interaction. This interaction, neverthe-

less, helps bind to the ribosome. IF-1 binds

in the A site where it may function to prevent the inappro-

priate or premature binding of a tRNA. IF-3 also functions

in this stage of the initiation process: it destabilizes the

binding of tRNAs that lack the three G  C pairs in the an-

ticodon stem of (Fig. 32-40) and helps discrimi-

nate between matched and mismatched codon–anticodon

interactions.

3. Last, in a process that is preceded by IF-1 and IF-3

release, the 50S subunit joins the 30S initiation complex in

a manner that stimulates IF-2 to hydrolyze its bound GTP

to GDP  P

. This irreversible reaction conformationally

rearranges the 30S subunit and releases IF-2 for participa-

tion in further initiation reactions.

IF-2 is a member of the superfamily of regulatory GTPases

such as Ras and hence is a G protein (Section 19-2A). The

30S initiation complex therefore functions as its GAP

(GTPase-activating protein; Section 19-2Ca).

Initiation results in the formation of an 

mRNA  ribosome complex in which the occu-

pies the ribosome’s P site while its A site is poised to accept an

incoming aminoacyl–tRNA (an arrangement similar to that

at the conclusion of a round of elongation; Section 32-3D).

In fact, is the only tRNA that directly enters the P

site. All other tRNAs must do so via the A site during chain

elongation (Section 32-3D). This arrangement was estab-

lished through the use of the antibiotic puromycin as is dis-

cussed in Section 32-3Df.

d. Eukaryotic Initiation Is Far More Complicated

than That of Prokaryotes

Although translational initiation in eukaryotes superfi-

cially resembles that in prokaryotes, it is, in fact, a far more

complicated process. Whereas prokaryotic initiation only

requires the assistance of three monomeric initiation fac-

tors, that in eukaryotes involves the participation of at least

12 initiation factors (designated eIFn; “e” for eukaryotic)

that consist of at least 26 polypeptide chains. Eukaryotic

initiation occurs as follows (Fig. 32-44):

1. The process begins with the binding of eIF3 (which in

mammals consists of 13 different subunits) and eIF1A (a

monomer and homolog of bacterial IF-1) to the 40S sub-

unit in the inactive 80S ribosome (which had terminated

elongation in its previous elongation cycle) so that it re-

leases the 60S subunit.

2. The ternary complex of eIF2 (a heterotrimer), GTP,

and binds to the 40S ribosomal subunit ac-

companied by eIF1 (a monomer) to form the so-called 43S

Met–tRNA

Met

tRNA

Met

fMet–tRNA

Met

fMet–tRNA

Met

RNA

Met

fMet–tRNA

Met

fMet–tRNA

Met

fMet–tRNA

Met

1376 Chapter 32. Translation

Figure 32-43 Translational initiation pathway in E. coli. The

E site, which is unoccupied during this process, has been omitted

for clarity.

JWCL281_c32_1338-1428.qxd 8/4/10 4:45 PM Page 1376

Section 32-3. Ribosomes and Polypeptide Synthesis 1377

80S Initiation complex

eIF1A

Inactive 80S ribosome

60S

Site

60S

40S

Met – tRNA

Met

43S Preinitiation complex

48S Preinitiation complex

mRNA

eIF1

elF2B

elF5

elF5B

GTP

GDP

GTP

+ P

GTP

elF4H

elF4B

elF4A

elF4G elF4F

elF4E

ATP

ADP + P

ATP

ADP + P

GTP 60S subunit

+ GDP

GTP

eIF3

AUG

Met

GTP

Met

Figure 32-44 Translational initiation

pathway in eukaryotes. Initiation factors

are represented by colored rectangles

where they are first implicated in the

pathway and by circles of the same color

thereafter.The higher order complexes

are hypothetical. Several of the initiation

factors (4A, 4B, 4E, 4G, and 4H) have

been left out of the 48S preinitiation

complex for clarity. [Based on a drawing

by Hershey, J.W.B. and Merrick,W.C., in

Sonenberg, N., Hershey, J.W.B., and

Mathews, M.B. (Eds.), Translational

Control of Gene Expression, Cold Spring

Harbor Laboratory Press (2000).]

JWCL281_c32_1338-1428.qxd 8/5/10 3:16 PM Page 1377

preinitiation complex. Here the subscript “i” on

distinguishes this eukaryotic initiator tRNA, whose ap-

pended Met residue is never N-formylated, from that of

prokaryotes; both species are, nevertheless, readily inter-

changeable in vitro.

3. Eukaryotic mRNAs lack the complementary se-

quences to bind to the 18S rRNA in the Shine–Dalgarno

manner. Rather, they have an entirely different mechanism

for recognizing the mRNA’s initiating AUG codon. Eukary-

otic mRNAs, nearly all of which have an m

G cap and a

poly(A) tail (Section 31-4Ab), are invariably monocistronic

and almost always initiate translation at their leading AUG.

This AUG, which occurs at the end of a 5¿-untranslated re-

gion of 50 to 70 nt, is embedded in the consensus sequence

GCCRCCAUGG, with changes in the purine (R) 3 nt be-

fore the AUG and the G immediately following it reducing

translational efficiency by ⬃10-fold each and with other

changes having much smaller effects. In addition, secondary

structure (stem–loops) in the mRNA upstream of the initi-

ation site may affect initiation efficiency.

The recognition of the initiation site begins by the bind-

ing of eIF4F to the m

G cap. eIF4F is a heterotrimeric

complex of eIF4E, eIF4G, and eIF4A (all monomers), in

tRNA

Met

which eIF4E (cap-binding protein) recognizes the

mRNA’s m

G cap and eIF4G serves as a scaffold to join

eIF4E with eIF4A. Both the X-ray and NMR structures of

eIF4E in complex with m

GDP, determined by Nahum

Sonenberg and Stephen Burley and by Sonenberg and

Gerhard Wagner, reveal that the protein binds the m

base by intercalating it between two highly conserved

Trp residues (Fig. 32-45a) in a region that is adjacent to a

positively charged cleft that forms the putative mRNA-

binding site (Fig. 32-45b). The m

G base is specifically rec-

ognized by hydrogen bonding to protein side chains in a

manner reminiscent of G  C base pairing. eIF4G also

binds poly(A)-binding protein (PABP; Section 31-4Ab)

bound to the mRNA’s poly(A) tail, thereby circularizing

the mRNA (not shown in Fig. 32-44). Although this ex-

plains the synergism between an mRNA’s m

A cap and its

poly(A) tail in stimulating translational initiation, the

function of this circle is unclear. However, an attractive

hypothesis is that it enables a ribosome that has finished

translating the mRNA to reinitiate translation without

having to disassemble and then reassemble. Another pos-

sibility is that it prevents the translation of incomplete

(broken) mRNAs. This circularization, as we have seen in

Section 31-4Av, also protects the mRNA from degradation

1378 Chapter 32. Translation

Figure 32-45 X-ray structure of murine eIF4E in complex

with the m

G cap analog m

GDP. (a) The m

GDP-binding site

with the m

GDP and the side chains that bind it drawn in ball-

and-stick form with the atoms of the m

GDP colored according

to type (C green, N dark yellow, O red, and P bright yellow) and

the protein side chains with which the m

GDP interacts drawn in

various colors. Hydrogen bonds, salt bridges, and van der Waals

interactions are represented by dashed lines and bridging water

molecules are drawn as black spheres.The m

G base is

intercalated between the indole rings of Trp 56 and Trp 102,

where it specifically interacts with protein side chains through

hydrogen bonds and van der Waals interactions. The GDP’s

phosphate groups interact directly and indirectly with three basic

side chains. (b) The solvent-accessible surface of eIF4E colored

according to its electrostatic potential (red negative, blue

positive, and white neutral) and viewed approximately as in Part

a. The m

GDP is drawn in ball-and-stick form colored as in Part

a. The mRNA presumably binds in the positively charged cleft

(yellow arrow) that is adjacent to the m

G binding site and which

passes between Lys 159 and Ser 209. [Courtesy of Nahum

Sonenberg, McGill University, Montréal, Québec, Canada.

PDBid 1EJ1.]

(a)

(b)

JWCL281_c32_1338-1428.qxd 8/4/10 4:45 PM Page 1378

by preventing the action of decapping enzyme until the

mRNA’s poly(A) tail has been shortened to the point that

it can no longer bind PABP.

4. eIF4B (an RRM-containing homodimer) and eIF4H

(a monomer) join the eIF4F–mRNA complex where they

stimulate the RNA helicase activity of eIF4A to unwind

the mRNA’s helical segments in an ATP-dependent

process. This presumably also strips away the proteins that

are bound to the mRNA (Section 31-4Au). eIF4A is the

prototype of the so-called DEAD-box family of proteins

(also known as DExD/H family proteins; Section 31-4Au),

which is named after one of the sequence motifs shared by

the diverse members of this family, all of which have

NTPase activity.

5. The eIF4F–mRNA–eIF4B–eIF4H complex joins the

43S preinitiation complex through a protein–protein inter-

action between eIF4G and the 40S subunit-bound eIF3.

This differs substantially from the corresponding prokary-

otic process (Fig. 32-43) in which the mRNA is bound to

the 30S ribosomal subunit via associations between RNA

molecules (involving the Shine–Dalgarno sequence and

the codon–anticodon interaction).

6. eIF5 (a monomer) joins the growing assembly. The

43S preinitiation complex then translocates along the

mRNA, an ATP-dependent process called scanning, until it

encounters the mRNA’s AUG initiation codon, which is

optimally in the sequence GCC(A/G)CCAUGG. This

yields the 48S preinitiation complex. The recognition of the

AUG occurs mainly through base pairing with the CUA

anticodon on the bound , as was demon-

strated by the observation that mutating this anticodon re-

sults in the recognition of the new cognate codon instead of

AUG. This explains why the initiator tRNA must bind to

the small subunit before the mRNA.

7. The formation of the 48S preinitiation complex in-

duces eIF2 to hydrolyze its bound GTP to GDP  P

, which

results in the release of all the initiation factors, thereby

leaving the in the small subunit’s P site. The

hydrolysis reaction is stimulated by eIF5, acting as a GAP

(Section 19-2Ca).

8. The 60S subunit then joins the mRNA-bound

–40S subunit complex in a GTPase reac-

tion mediated by eIF5B (a monomer and homolog of

bacterial IF-2), thereby yielding the 80S ribosomal initia-

tion complex. Thus eukaryotic translation initiation con-

sumes two GTPs versus one for prokaryotic initiation

(Fig. 32-43).

9. What remains is to recycle the eIF2  GDP complex

by exchanging its GDP for GTP. This reaction is mediated

by eIF2B (a heteropentamer), which therefore functions as

eIF2’s GEF (guanine nucleotide exchange factor; Section

19-2Ca).

Many eukaryotic initiation factors are subject to phospho-

rylation/dephosphorylation and are therefore likely to par-

Met–tRNA

Met

Met–tRNA

Met

Met–tRNA

Met

ticipate in the control of eukaryotic translation, a subject

we discuss in Section 32-4.

Although the initiation sites on most eukaryotic

mRNAs are identified by the above-described scanning

mechanism, a few mRNAs have an internal ribosome entry

site (IRES) to which the 40S subunit can directly bind in a

process reminiscent of prokaryotic initiation. However, lit-

tle is yet known about the mechanism of IRES-based ini-

tiation. Indeed, IRESs lack clearly identifiable consensus

sequences.

D. Chain Elongation

See Guided Exploration 29: Translational elongation Ribo-

somes elongate polypeptide chains in a three-stage reaction

cycle that adds amino acid residues to a growing polypep-

tide’s C-terminus (Fig. 32-46):

1. Decoding, in which the ribosome selects and binds an

aminoacyl–tRNA, whose anticodon is complementary to

the mRNA codon in the A site.

2. Transpeptidation, in which the peptidyl group on the

P-site tRNA is transferred to the aminoacyl group in the A

site through the formation of a peptide bond (Fig. 32-39).

3. Translocation, in which A-site and P-site tRNAs are

respectively transferred to the P site and E site accompa-

nied by their bound mRNA; that is, the mRNA, together

with its base paired tRNAs, is ratcheted through the ribo-

some by one codon.

Translational elongation, which occurs at a rate of 10 to 20

residues/s, involves the participation of several nonriboso-

mal proteins known as elongation factors (Table 32-9). We

describe these processes in the following paragraphs.

a. Decoding

In the decoding stage of the E. coli elongation cycle, a

binary complex of GTP with the elongation factor EF-Tu

(also called EF1A) combines with an aminoacyl–tRNA.The

resulting ternary complex binds to the ribosome, and,in a re-

action that hydrolyzes the GTP to GDP  P

, the amino-

acyl–tRNA is bound in a codon–anticodon complex to the

ribosomal A site and EF-Tu  GDP  P

is released. In the

remainder of this stage, the bound GDP is replaced by GTP

in a reaction mediated the elongation factor EF-Ts (also

called EF1B). EF-Tu,as are several other GTP-binding ribo-

somal factors, is a G-protein, and hence the ribosome func-

tions as its GAP and EF-Ts is its GEF.

Aminoacyl–tRNAs can bind to the ribosomal A site

without the mediation of EF-Tu but at a rate too slow to

support cell growth.The importance of EF-Tu is indicated

by the fact that it is the most abundant E. coli protein; it is

present in ⬃100,000 copies per cell (5% of the cell’s

protein), which is approximately the number of tRNA

molecules in the cell. Consequently, the cell’s entire com-

plement of aminoacyl–tRNAs is essentially sequestered by

EF-Tu.

Section 32-3. Ribosomes and Polypeptide Synthesis 1379

JWCL281_c32_1338-1428.qxd 8/19/10 10:05 PM Page 1379

b. EF-Tu Is Sterically Prevented from Binding

Initiator tRNA

The X-ray structure of the Phe–tRNA

Phe

 EF-Tu 

GMPPNP ternary complex (GMPPNP is a nonhydrolyz-

able GTP analog; Section 19-3Ch), determined by Brian

Clark and Jens Nyborg, reveals that these two macromole-

cules associate to form a corkscrew-shaped complex in

which the EF-Tu and the tRNA’s acceptor stem form a

knoblike handle and the tRNA’s anticodon helix forms the

screw (Fig. 32-47). The conformation of the tRNA

Phe

closely resembles that of the uncomplexed molecule (Fig.

32-11b). The EF-Tu folds into three distinct domains that

are connected by flexible peptides, rather like beads on a

string. The N-terminal domain 1, which binds guanine nu-

cleotides and catalyzes GTP hydrolysis, structurally resem-

bles other known G-proteins.

The two macromolecules associate rather tenuously via

three major regions: (1) the CCA—Phe segment at the 3¿

end of the Phe–tRNA

Phe

binds in a cleft between EF-Tu’s

domains 1 and 2 (the blue and green mainly helical domain

and the yellow  sheet domain in Fig. 32-47) that ends in a

pocket large enough to accommodate all amino acid

residues; (2) the 5¿-phosphate of the tRNA binds in a de-

pression at the junction of EF-Tu’s three domains; and (3)

one side of the tRNA’s TC stem contacts the exposed

main chain and side chains of EF-Tu’s C-terminal domain 3

(the orange and red  barrel–containing domain in Fig. 32-

47). The tight association of the aminoacyl group with EF-

Tu appears to greatly increase the affinity of EF-Tu for the

otherwise loosely bound tRNA, which explains why EF-Tu

does not bind uncharged elongator tRNAs.

EF-Tu binds neither formylated aminoacyl–tRNAs nor

unformylated , which is why the initiator

tRNA never reads internal AUG or GUG codons.The first

base pair of is mismatched (C  A; Fig. 32-40) and

hence this initiator tRNA has a 3¿ overhang of 5 nt vs 4 nt

for an elongator tRNA. It seems likely that this mismatch,

together with the formyl group, prevents

from binding to EF-Tu. Indeed, EF-Tu binds to E. coli

whose 5¿-terminal C residue has been deaminatedtRNA

Met

fMet–tRNA

Met

tRNA

Met

Met–tRNA

Met

1380 Chapter 32. Translation

Figure 32-46 Elongation cycle in E. coli ribosomes. The E

site, to which discharged tRNAs are transferred before being

released to solution, is not shown. Eukaryotic elongation follows

EF-G

5′ 3′

5′

5′ 3′

Nascent

polypeptide

A site

Empty

Translocation

GTP

+ GDP + P

tRNA

mRNA

Peptidyl–tRNA

Transpeptidation

P site

Decoding

A site

′

Aminoacyl–

tRNA

EF-Tu • GDP

EF-Tu

• EF-Ts

GTPEF-Ts

EF-Tu • GTP

Aminoacyl–tRNA

•

EF-Tu

• GTP

GDP

EF-Ts

P site

A site

Peptidyl–tRNA

fMet

Uncharged

tRNA

a similar cycle but EF-Tu and EF-Ts are replaced by a single

multisubunit protein, eEF1, and EF-G is replaced by eEF2.

JWCL281_c32_1338-1428.qxd 8/4/10 4:45 PM Page 1380