In vivo effect of inactivation of ribosome ... - Wiley Online Library

Blackwell Science, LtdOxford, UKMMIMolecular Microbiology0950-382XBlackwell Publishing Ltd, 2004? 200454410111021Original ArticleIn vivo inactivation of E. coli RRFG. Hirokawa et al.

Molecular Microbiology (2004) 54(4), 1011–1021

doi:10.1111/j.1365-2958.2004.04324.x

In vivo effect of inactivation of ribosome recycling factor – fate of ribosomes after unscheduled translation downstream of open reading frame Go Hirokawa,1,2 Hachiro Inokuchi,3 Hideko Kaji,4 Kazuei Igarashi1 and Akira Kaji2* 1 Department of Clinical Biochemistry, Graduate School of Pharmaceutical Sciences, Chiba University, Chiba 2608675, Japan. 2 Department of Microbiology, School of Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA. 3 Department of Biophysics, Graduate School of Science, Kyoto University, Kyoto 606-8502, Japan. 4 Department of Biochemistry and Molecular Pharmacology, Jefferson Medical College, Thomas Jefferson University, Philadelphia, PA 19107, USA. Summary The post-termination ribosomal complex is disassembled by ribosome recycling factor (RRF) and elongation factor G. Without RRF, the ribosome is not released from mRNA at the termination codon and reinitiates translation downstream. This is called unscheduled translation. Here, we show that at the non-permissive temperature of a temperaturesensitive RRF strain, RRF is lost quickly, and some ribosomes reach the 3¢¢ end of mRNA. However, instead of accumulating at the 3¢¢ end of mRNA, ribosomes are released as monosomes. Some ribosomes are transferred to transfer-messenger RNA from the 3¢¢ end of mRNA. The monosomes thus produced are able to translate synthetic homopolymer but not natural mRNA with leader and canonical initiation signal. The pellet containing ribosomes appears to be responsible for rapid but reversible inhibition of most but not all of protein synthesis in vivo closely followed by decrease of cellular RNA and DNA synthesis. Introduction At the protein synthesis termination step, nascent polypeptide is released from tRNA by release factor, forming the post-termination complex. This complex is then

Accepted 26 July, 2004. *For correspondence. E-mail [email protected]; Tel. (+1) 215 898 8828; Fax (+1) 215 573 2221.

© 2004 Blackwell Publishing Ltd

disassembled by ribosome recycling factor (RRF) and elongation factor G (EF-G) (see review Kaji et al., 2001). The disassembly of the post-termination complex is the actual final (fourth) step of protein biosynthesis. RRF is an essential protein for Escherichia coli (Janosi et al., 1994), and all prokaryotes examined so far have frr genes (coding for RRF). RRF homologues have also been found in eukaryotic cells, but all contain an N-terminal extension for localization and function only in the chloroplast or mitochondria (Kanai et al., 1998; Rolland et al., 1999; Teyssier et al., 2003). Over 50 lethal mutations, 12 temperature-sensitive mutations, their revertants, and suppressor mutations of RRF have been isolated (Janosi et al., 2000) and some of the mutations were confirmed (Fujiwara et al., 1999; 2001). RRF stimulates in vitro protein synthesis as much as sevenfold (Ryoji et al., 1981b; Pavlov et al., 1997). It also stimulates RNA polymerase in vitro, suggesting a possible role of RRF in RNA synthesis (Kung et al., 1977) as has been demonstrated with another protein synthesis factor, EF-Tu, which is involved in RNA synthesis (Blumenthal, 1979). Structural studies have revealed that RRF is a nearly perfect structural mimic of tRNA (Selmer et al., 1999; Yoshida et al., 2001). This was confirmed by other laboratories (Kim et al., 2000; Toyoda et al., 2000; Nakano et al., 2003). RRF occupies the A/P-site of ribosome (Hirokawa et al., 2002a; Lancaster et al., 2002), and orientation of the ribosome-bound RRF (Lancaster et al., 2002; Agrawal et al., 2004) showed that the elbow of RRF is found in an overlapping position at the junction of domains III, IV and V of EF-G, whereas RRF domain II occupies an overlapping position with domain IV in the GDP (guanosine diphosphate) state of EF-G (Agrawal et al., 2004). It has been suggested that RRF, like tRNA, is ‘moved’ on the ribosome by EF-G (Hirokawa et al., 2002b) before it is released from the ribosome (Kiel et al., 2003). In E. coli, it was shown that the inactivation of RRF during the growing phase causes a bacteriostatic effect, whereas during the lag phase it causes a bactericidal effect (Janosi et al., 1998). When RRF is absent, almost all the upstream ribosomes remain on the mRNA. These ribosomes remaining on the mRNA reinitiate unscheduled translation downstream from the termination codon (Ryoji

1012 G. Hirokawa et al.

It has been shown that RRF stimulates in vitro protein biosynthesis (Kung et al., 1977; Ryoji et al., 1981b; Shimizu et al., 2001) or in vitro oligopeptide synthesis (Freistroffer et al., 1997) four- to sevenfold. However, it has not been established that the primary role of RRF in vivo is in protein biosynthesis. To address this question, we utilized an E. coli tsRRF mutant, LJ14, that has a single amino acid substitution (Val 117 Asp) in RRF (Janosi et al., 1998), and studied incorporation of methionine into protein at the permissive and non-permissive temperatures. From the known function of RRF and kinetics of inactivation of pure tsRRF at the non-permissive temperature (Janosi et al., 1998), one

Fig. 1. Effects of RRF inactivation on synthesis of protein, RNA and DNA in vivo. LJ14 (A, C, E and G) and MC1061 (B, D, F and H) grown in minimal media at 28∞C were exposed to 42∞C at zero time and the rates of protein, RNA and DNA synthesis were measured by the incorporation of [35S]-methionine (250 mCi mmol-1, A and B), [3H]-uridine (26.3 mCi mmol-1, C and D) and [3H]-thymidine (39.5 mCi mmol-1, E and F) respectively. Methionine incorporation was measured by hot trichloroacetic acid (TCA)-insoluble counts, and uridine and thymidine incorporation were measured by cold TCA-insoluble counts. (G) and (H) represent cell densities at 540 nm. Filled symbols represent cells exposed to 42∞C and open symbols represent those kept at 28∞C. In (I) and (J), rates of synthesis [d(cpm)/dt] of protein (open circles) and RNA (filled circles) were calculated from A–D, and the ratios of the synthesis rates at 42∞C and 28∞C in LJ14 (I) or MC1061 (J) were plotted. Note that the rate of synthesis [d(cpm)/dt] means the increase of the incorporation into macromolecule per minute, and the ratio of the synthesis rate between two temperatures was calculated by dividing [d(cpm)/dt at 42∞C] by [d(cpm)/dt at 28∞C].

[35S]-methionine (cpm per 50 ml)

2x10 5

B

A

1x10 5

1x10 5

0

0

[3H]-uridine (cpm per 50 ml)

0

60

120

180

60

120

180

D

C

3x10 4

0

3x10 4

2x10 4

2x10 4

1x10 4

1x10 4 0

0 0

[3H]-thymidine (cpm per 50 ml)

Effect of in vivo inactivation of RRF on macromolecule synthesis

2x10 5

60

120

180

0

60

120

180

F

E

3x10 3

3x10 3

2x10 3

2x10 3

1x10 3

1x10 3 0

0 0 0.8 OD 540

Results

would not expect that instant cessation of in vivo protein synthesis. However, Fig. 1A shows that, methionine incorporation into protein in LJ14 was inhibited much more rapidly than expected after the temperature was shifted up to the non-permissive temperature (42∞C), whereas the incorporation in the wild-type strain (MC1061) was slightly accelerated (Fig. 1B). We then compared the results obtained in Fig. 1A with the temperature effect on RNA synthesis. Surprisingly, as shown in Fig. 1C and D, the temperature shift up to 42∞C caused just as rapid inhibition of RNA synthesis in LJ14 but not in the wild-type

60

120

180

G

0

60

120

180 0.8

H

0.4

0.4

0

0 0

60

120

180

0

60

120

180

Time (min)

d(cpm)/dt at 42∞C d(cpm)/dt at 28∞C

et al., 1981a,b; Janosi et al., 1998). Thus, ‘unscheduled’ polypeptides are produced by these ribosomes. Under these conditions, ribosomes can slide on mRNA up to 45 nucleotides downstream before beginning unscheduled translation. However, it was not clear whether the ribosomes engaged in the unscheduled translation actually reach the 3¢ end of mRNA. In this paper we show that, upon exposure of strain carrying temperature-sensitive RRF (tsRRF) to the non-permissive temperature in vivo, unexpected rapid decrease of protein synthesis takes place closely followed by inhibition of RNA synthesis. During this process, ribosomes reach the 3¢ end of mRNA and are released as monosomes. Some ribosomes are transferred to transfermessenger RNA (tmRNA). The pellet containing these ribosomes is still active for polyphenylalanine synthesis programmed by poly(U) (Nirenberg and Matthaei, 1961) but inactive for protein synthesis programmed by natural mRNA. This explains the rapid decrease but not complete cessation of protein synthesis upon loss of RRF in vivo.

I

2

J

2 1.5

1.5

1

1

0.5

0.5

0

0 0

20

40

60 0

20

40

60

Time (min) © 2004 Blackwell Publishing Ltd, Molecular Microbiology, 54, 1011–1021

In vivo inactivation of E. coli RRF 1013

MC1061 0

10

20

30

LJ14 60 120

0

10

20

30

60 120 min at 42°C

culture. It appeared from these figures that RNA synthesis was influenced as much as protein synthesis was by the shift of the culture to the non-permissive temperature. To analyse these data in detail, the rates [d(cpm)/dt] of protein and RNA synthesis at 42∞C and 28∞C were calculated from the data. We reasoned that exposure to 42∞C would first reduce the rate of protein synthesis followed by that of RNA synthesis if the primary function of RRF is in protein biosynthesis but not in RNA synthesis. This means that the ratio [d(cpm)/dt at 42∞C]/[d(cpm)/dt at 28∞C] is reduced first with protein synthesis. We calculated this ratio and plotted against the time after the temperature shift for LJ14 (Fig. 1I) and MC1061 (Fig. 1J). As shown in Fig. 1I, the ratio [d(cpm)/dt at 42∞C]/[d(cpm)/ dt at 28∞C] for protein synthesis (open circle) rapidly decreased whereas that for RNA synthesis (closed circle) showed slight increase followed by a decline upon inactivation of RRF in LJ14. This indicates that inactivation of RRF first inhibits protein biosynthesis followed by RNA synthesis inhibition. It is noted that, in the wild-type strain (Fig. 1J), no decrease in the ratio was observed. Actually, the ratio is higher at 42∞C than at 28∞C because more synthesis takes place at 42∞C. We conclude from Fig. 1I and J that the primary role of RRF in vivo is in protein synthesis. We also measured the in vivo incorporation of [3H]thymidine into DNA (Fig. 1E and F). Though much less and later than the inhibitory effect on protein and RNA synthesis, exposure to 42∞C inhibited DNA synthesis in LJ14 but not in wild-type cells. This is consistent with the observation that RRF inactivation during the growing phase causes a bacteriostatic effect (Janosi et al., 1998). It is noted in Fig. 1G and H that the optical density change was the slowest of the four parameters in response to the temperature shift. Mutant RRF V(117)D is rapidly lost in LJ14 upon temperature shift-up The rapid decrease of protein synthesis shown in the preceding section was unexpected and contradictory to very slow loss of the activity of the purified tsRRF at 42∞C observed in vitro (about 20% remained even after 60 min) (Janosi et al., 1998). We therefore examined the presence © 2004 Blackwell Publishing Ltd, Molecular Microbiology, 54, 1011–1021

Fig. 2. RRF is rapidly degraded in LJ14 at nonpermissive temperature. MC1061 and LJ14 were incubated in LB at 30∞C (permissive temperature) until OD540 reached 0.4, then exposed to 42∞C (non-permissive temperature for LJ14). At 10, 20, 30, 60 and 120 min after the temperature shift-up, cells were harvested and RRF present in the total proteins of 0.2 OD540 units of cell suspension was analysed by Western blotting. The sample at zero time was taken just before the temperature shift-up.

of RRF in LJ14 at the non-permissive temperature as shown in Fig. 2. We found that 10 min after the temperature shift-up, almost no RRF was present in LJ14. As wildtype RRF is relatively heat stable (Hirashima and Kaji, 1972), it is possible that the mutant tsRRF V(117)D is unfolded at high temperature and becomes a target of proteolytic degradation. The loss of RRF must be very efficient at the non-permissive temperature in LJ14 because the expression of RRF is elevated at the nonpermissive temperature (Teixeira-Gomes et al., 2000). We conclude that the rapid decrease of protein synthesis at the non-permissive temperature observed in Fig. 1 results from the rapid loss of RRF at the non-permissive temperature. It was noted in Fig. 2 that, even at the permissive temperature (30∞C), the amount of RRF was less in LJ14 than in the wild-type strain MC1061 (zero time in Fig. 2), although the growth rate of LJ14 was almost identical to that of MC1061 at the permissive temperature (Fig. 1G and H). This observation suggests that the cellular concentration of RRF is more than that required for normal growth. The fact that very small amount of RRF is present in LJ14 even at the permissive temperature explains the preceding finding that plant RRF homologue was toxic to LJ14 even at the permissive temperature whereas wild type was resistant to it (Rolland et al., 1999) because a large amount of RRF would overcome the toxic effect of plant RRF. Upon RRF inactivation, some ribosomes are transferred to tmRNA We then asked about the fate of ribosomes upon loss of RRF in vivo. It is known that in the absence of RRF, ribosomes are not released from natural post-termination complexes (Ogawa and Kaji, 1975) and reinitiate translation downstream of a termination codon (Ryoji et al., 1981a,b; Janosi et al., 1998). We call this ‘unscheduled translation’. During unscheduled translation, the ribosomes not only can change reading frames, but also reinitiate at random positions (Janosi et al., 1998). The unscheduled translation may be repeated depending on how frequently the ribosome encounters a termination codon in the 3¢ UTR. The last round of unscheduled trans-

B

LJ14 DssrA MC1061 DssrA / pKW24 / pKW24

LJ14

MC1061

(tsRRF, wtssrA) (wtRRF, wtssrA)

MW

0 30 60

(kDa)

0 30 60 min

0 30 60

94 68 45 32 17

18

15

14

8.2 6.4 2.6

lation would begin downstream of the last termination codon closest to the 3¢ end and ribosomes eventually may reach the 3¢ end of mRNA. In Fig. 3, this hypothesis was tested observing the peptide tagging by tmRNA at the 3¢ end of mRNA (Keiler et al., 1996; Felden et al., 1997). In this experiment, we utilized a plasmid, pKW24, encoding a mutant tmRNA in which the C-terminal six codons in the tag ORF were all replaced with histidine codons (the ORF is ANDEHHHHHH) (Roche and Sauer, 2001). As shown in Fig. 3, when RRF was inactivated (LJ14 DssrA/pKW24 at 42∞C), tagged short peptides appeared (indicated by dotted box). The size of the peptides observed matches well with the predicted size of the unscheduled products closest to the 3¢ end of mRNA. From the possible 3¢ UTR sequences (Ermolaeva et al., 2000), we predicted the size to be 0–60 amino acid residues (the average is about 20 residues). The tagged short peptides observed are probably not degradation products of large proteins tagged by tmRNA, because they are not observed in MC1061 DssrA/pKW24 at 42∞C. It is also noted in Fig. 3 that many polypeptides with higher molecular weights were tagged both in LJ14 and in wild-type cells. These tagged polypeptides did not diminish in quantity upon raising the temperature either in LJ14 or in MC1061 suggesting again no appreciable degradation at the non-permissive temperature. As pointed out above, certain tmRNA peptide-tagged proteins of higher molecular weight were observed even in wild-type cells. It is noted that we did not see these bands in the control experiment

0 30 60

His-tagged RRF

A

(control for the anti-serum specificity)

1014 G. Hirokawa et al. Fig. 3. Production of peptides tagged with tmRNA-coded peptides upon inactivation of RRF. A. In the strain used in this experiment, the ssrA gene has been disrupted in LJ14. This is called LJ14 DssrA. This strain could grow slowly at 42∞C but not at 45∞C (data not shown). The strain was transformed with a plasmid (pKW24) carrying His-tagged ssrA. Similarly, MC1061 DssrA/pKW24 was constructed. LJ14 DssrA/ pKW24 and MC1061 DssrA/pKW24 were incubated in LB at 30∞C (permissive temperature) until OD540 reached 0.4 and then exposed to 42∞C (semi-permissive temperature for LJ14 DssrA/pKW24). At 0, 30 and 60 min after the temperature shift-up, cells were harvested. Proteins fused with tmRNA-His6 tag in total proteins equivalent to 0.2 OD540 were analysed by Western blot using rabbit anti-serum against the His6 peptide tag. Bands indicated by arrows increased in LJ14 but not in MC1061 at 42∞C. The dotted box indicates tmRNA-His6-tagged short peptides. MW, molecular weight. B. Absence of His-tagged proteins in the control cells with wild-type ssrA. LJ14 and MC1061 with wild-type tmRNA were grown at 30∞C to OD540 of 0.4. The cultures were then exposed to 42∞C. Cell extracts were analysed as in (A). For the control of the His6 protein detection, Nterminal His6-tagged RRF (400 ng) was subjected to the same Western blot procedure (extreme right lane).

(Fig. 3B) where the cells did not contain pKW24 or deletion of tmRNA gene. Figure 3 shows that some of these tagged proteins (shown by arrows) increased in LJ14 DssrA/pKW24 at the non-permissive temperature but not in MC1061 DssrA/ pKW24. This suggests that in the absence of RRF, tagging goes up at the point where ribosomes are stalled. This observation is consistent with the notion that RRF, together with tmRNA, is preventing ribosomes stalling together with tmRNA. Ribosomes engaged in unscheduled translation do not accumulate at the 3¢ end Even after the ribosomes are transferred to tmRNA, they should still translate the 3¢ UTR of tmRNA and should reach 3¢ end of tmRNA in the absence of RRF. If another molecule of tmRNA is available, the process should repeat. If ribosome stays on mRNA or tmRNA, those ribosomes after the first ribosome reaching the 3¢ end will also be forced to stay on mRNA and accumulate on these mRNAs. This would result in an increase of polysomes and a decrease in monosomes. Whether or not ribosomes stay on mRNA after it reaches the end of mRNA has an important implication regarding the mechanism of cessation of protein synthesis upon loss of RRF. If the ribosomes would stay on mRNA at the 3¢ end, the cessation of protein synthesis may be explained simply by the loss of pool of free ribosomes for translation initiation. © 2004 Blackwell Publishing Ltd, Molecular Microbiology, 54, 1011–1021


B

A

254

A

Top

Bottom

Top

Bottom

Fig. 4. Polysomes decreased and 70S ribosomes increased in LJ14 at the non-permissive temperature. Chloramphenicol-treated cell extracts (2.0 A260U) of LJ14 grown at 28∞C (permissive temperature) (A) or exposed to 42∞C (non-permissive temperature) for 20 min (B) were isolated as described in Experimental procedures. Sedimentation behaviour of polysomes and monosomes in 15–30% linear sucrose density gradient was examined by ultracentrifugation (Beckman SW50.1, 40 000 r.p.m., 75 min, 4∞C). The absorption at A254 was monitored with the ISCO UA6 spectrophotometer. Sedimentation is from left to right.

We analysed the sedimentation behaviour of ribosomes of LJ14 at both permissive (28∞C) and non-permissive (42∞C) temperatures (Fig. 4). As can be seen from Fig. 4, within 20 min of temperature shift-up, polysomes decreased in LJ14. Upon loss of RRF, 70S ribosomes (monosomes) increased from 29.3% at 28∞C to 72.2% at 42∞C. The ribosomal sedimentation pattern remained this way even at 120 min after the temperature shift-up (data not shown). In MC1061, although polysome decreased at 42∞C, the loss of polysomes was not as much as with LJ14 (data not shown). We conclude from Fig. 4 that ribosomes do not remain on mRNA but fall off from the 3¢ end in the absence of RRF. Ribosomes of LJ14 at the non-permissive temperature are active for translation of synthetic polynucleotides but not natural mRNA As protein synthesis is drastically and quickly reduced (Fig. 1) upon loss of RRF (Fig. 2), the ribosomes which fell off from 3¢ end of ribosomes (Figs 3 and 4) must not translate canonical mRNA. In the experiment described in Fig. 5, LJ14 grown at 28∞C was exposed to 42∞C for 20 min or 1.5 h, and the S30 extract was isolated. In Fig. 5A, the S30 extract of LJ14 was examined for cellfree translation of the MS2 phage RNA. The activity of the extract kept at the permissive temperature was two or five times higher than the extract exposed to the nonpermissive temperature for 20 min or 1.5 h respectively. This suggests that the cell-free system reflects the in vivo situation. Surprisingly, however, addition of purified wildtype RRF to this system did not restore the activity to the © 2004 Blackwell Publishing Ltd, Molecular Microbiology, 54, 1011–1021

extract exposed to 42∞C for either the short (Fig. 5B) or the long (Fig. 5C) period. The loss of activity was not observed in the extract from the 42∞C culture of MC1061 (Fig. 5D), indicating that the lower activity observed in the heat-treated LJ14 extract resulted from the inactivation of tsRRF. We attempted to recover in vitro protein synthesis activity by adding purified IF3, which dissociates 70S ribosomes formed as a result of the disassembly reaction catalysed by RRF and EF-G (Hirokawa et al., 2002b). As shown in Fig. 5E, added IF3 did not restore the activity either. Next, we determined which fraction, soluble or ribosomal, was damaged because of the absence of RRF. In the experiment shown in Fig. 5F, we show that the damage caused by the absence of RRF is in the pellet containing ribosomes rather than in the soluble fraction. In contrast to the loss of activity with MS2 RNA, both the soluble and pellet fractions isolated from LJ14, which were exposed to the non-permissive temperature as well as kept at the permissive temperature, were active for polyphenylalanine synthesis programmed by poly(U) (Fig. 5G). The control extract from wild-type cells, under identical conditions, gave similar activity with poly(U) (Fig. 5H). This suggests that the elongation process is not affected. Rather, one or more of the processes involving initiation, termination and/or disassembly that are not involved with the poly(U) programmed system are affected by the inactivation of RRF. Recovery of cell growth after loss of RRF is slow but not unusual compared with the recovery of protein synthesis after inhibition by antibiotics As loss of RRF is bacteriostatic but not bactericidal (Janosi et al., 1998), damage caused by the loss of RRF must be repairable. We reasoned that the time-course of recovery of the cells from the loss of RRF may shed some light on the mechanism of loss of protein synthesis. With this in mind, the cultures were exposed to the nonpermissive temperature for 20 min or 1.5 h before the shift back to the permissive temperature. Regardless of the exposure period, it took about 1 h before the incorporation of methionine began again (Fig. 6A). This recovery time was the same even if the exposure period was increased up to 3 h (data not shown). Recovery of the optical density increase took longer (1.5–2 h; Fig. 6B) but, similarly, the length of exposure period did not influence the recovery time. One hour recovery time was apparently not unusual for recovery of protein synthesis as shown in Fig. 6C. Thus, similar or even longer recovery time was observed with the bacterial cells treated with tetracycline (Fig. 6C) (a typical protein synthesis inhibitor). It is clear from Fig. 6 that it took about 1.2 h after removal of tetracycline for

0

No template RNA

C LJ14 S30 of 28°C culture

80 S30 of 42°C for 1.5 h

60 40 20 0 0 10 20 30 Incubation at 30∞C (min)

No RRF added + RRF 0.04 µM + RRF 0.4 µM + RRF 4 µM No template RNA

[14C]-phe incorporation (% activity)

0

100

80 + IF3 3.3 µM No RRF added + IF3 7.5 µM + IF3 1.7 µM + IF3 15.3 µM

60 40 20 0 0 10 20 30 Incubation at 30∞C (min)

[35S]-met hot TCA-insoluble count (cpm per 6 ml reaction)

[35S]-met incorporation (% activity)

0 10 20 30 Incubation at 30∞C (min)

100

20

F LJ14

S30 of 28°C culture S30 of 42°C culture

No template (42°C) No template (28°C)

H MC1061

S30 of 28°C culture

100

No template (42°C) No template (28°C)

40


[14C]-phe incorporation (% activity)

20



40

0

S30 of 42°C f or 20min

S30 of 42°C f or 20 min

No RRF added + RRF 0.2 µM + RRF 0.4 µM + RRF 0.9 µM + RRF 1.8 µM

60

S30 (42°C, 90 min) S30 (28°C)

E LJ14 S30 of 28°C culture

80

20

60


B LJ14 100

40 - MS2 RNA


- MS2 RNA

0

S30 (28°C) S30 (42°C, 20 min) S30 (42°C, 90 min)


S30 (42°C, 90 min)

20

60

S30 of 28°C culture S30 of 42°C culture

- Poly(U)

S30 (42°C, 20 min)

40

80

+ Poly(U)

60

80

G LJ14

100

- Poly(U)

80

S30 (42°C, 90 min) S30 (28°C)

100

+ Poly(U)

S30 (28°C)

+ MS2 RNA

D MC1061

A LJ14

100

+ MS2 RNA


1016 G. Hirokawa et al.

No template RNA

80 60 40 20 0 0 10 20 30 Incubation at 30∞C (min)

Ribosome of 28°C mixed with the supernatant of 42°C

1.8 x 10 6 1.2 x 10 6

Ribosome of 42°C mixed with the supernatant of 28°C

6 x 10 5

No template RNA (for Ribo-28°C + S150-42°C)

0 0 10 20 30 Incubation at 30∞C (min)

No template RNA (for Ribo-42°C + S150-28°C)

Fig. 5. In vitro activities of extract of LJ14 exposed to non-permissive temperature. A. Translation of MS2 phage RNA by heat-treated LJ14 extract is much reduced. S30 extracts (0.6 A260U) of LJ14 grown at 28∞C (circles), exposed to 42∞C for 20 min (triangles) or exposed to 42∞C for 1.5 h (squares) were isolated, and examined for in vitro polypeptide synthesis programmed by MS2 phage RNA (filled symbols) at 30∞C. For controls, no template RNA was added (open symbols). Hot TCA-insoluble counts of [35S]methionine (1.2 mCi pmol-1) with S30 of 28∞C culture and MS2 RNA at 30 min (1.60 ¥ 106 cpm in 6 ml of total reaction mixture 30 ml) were defined as 100% and the relative activity was plotted against the incubation time. Three independently isolated extracts were tested, and the averages and SEM are shown. B and C. Addition of RRF does not restore the activity of heat-treated LJ14 S30 extract. Extracts (0.6 A260U) of LJ14 exposed to 42∞C for 20 min (B) or 1.5 h (C) were examined as in (A) except that various amounts of purified wild-type RRF were added to the extract. In (B), closed squares, 0.2 mM RRF; open triangles, 0.4 mM RRF; closed triangles, 0.9 mM RRF; open diamonds, 1.8 mM RRF; closed circles, no RRF was added. In (C), closed squares, 0.04 mM RRF; open triangles, 0.4 mM RRF; closed squares, 4 mM RRF; closed circles, no RRF was added. In (B) or (C), open circles represent S30 extract of LJ14 at 28∞C; open squares represent controls without MS2 phage RNA. Certain symbols are hidden by others because the values were very close. 100% = 1.73 ¥ 106 cpm (B, open circle at 30 min) or 1.84 ¥ 106 cpm (C, open circle at 30 min) in 6 ml of total reaction mixture 30 ml. D. In vitro translation of MS2 phage RNA with the extract of wild-type cells is not influenced by culture temperature. S30 extracts of MC1061 at 28∞C (circles) and 42∞C (1.5 h exposure, squares) were examined for MS2 phage RNA directed [35S]-methionine incorporation as in (A) except that the specific activity of [35S]-methionine was 0.08 mCi pmol-1. 100% = 1.25 ¥ 105 cpm (closed circle at 30 min) in 6 ml of total reaction mixture of 30 ml. Closed symbols, with template; open symbols, without template. E. Addition of IF3 does not restore the activity of the LJ14 S30 extract. The S30 extract (0.6 A260U) of LJ14 exposed to 42∞C for 20 min was examined as in (B) except that, instead of RRF, various amounts of IF3 were added to the extract. Closed squares, 1.7 mM IF3; open triangles, 3.3 mM IF3; closed triangles, 7.5 mM IF3; open diamonds, 15.3 mM IF3; closed circles, no IF3 was added. S30 extract of LJ14 at 28∞C is open circles. Open squares: MS2 phage RNA was omitted in the reaction with S30 of 42∞C and 15.3 mM IF3. Certain symbols are hidden by others. 100% = 1.44 ¥ 106 cpm in 6 ml of total reaction mixture 30 ml (open circle at 30 min). F. Ribosomal fractions are responsible for the inactivity of heat-treated LJ14 S30 extract. Extracts (0.6 A260U) of LJ14 grown at 28∞C or exposed to 42∞C for 1.5 h were centrifuged at 150 000 g for 2 h, and the ribosomal pellets and the post-ribosomal supernatants were separated. The pellet of extract from cells grown at 28∞C was mixed with the supernatant of that from cells exposed to 42∞C (circles). Likewise, the pellet of S30 from cells exposed to 42∞C was mixed with the supernatant of that grown at 28∞C (squares). In vitro protein synthesis programmed by the MS2 phage RNA was examined with these mixtures as in (A). Closed symbols represent incorporation of [35S]-methionine (1.2 mCi pmol-1) with MS2 phage RNA as the template, and open symbols represent incorporation without template RNA. Open circles are hidden by open squares. G. S30 extract of LJ14 from the 42∞C culture is fully active for polyphenylalanine synthesis. S30 extracts of LJ14 grown at 28∞C (circles) or exposed to 42∞C for 1.5 h (squares) were isolated and in vitro poly[14C]-phenylalanine (0.33 mCi nmol-1) synthesis programmed by poly(U) was examined with 0.6 A260U of the extract. Closed symbols represent polyphenylalanine formation and open symbols represent the control without poly(U). Closed circles and open circles are hidden by other symbols. 100% = 2894 cpm in 6 ml of total reaction mixture 30 ml (closed circle at 30 min). H. In vitro polyphenylalanine synthesis by the extract of wild-type cells. S30 extract of MC1061 at 28∞C (circles) and 42∞C (1.5 h exposure, squares) cultures were examined for poly(U) directed [14C]-phenylalanine incorporation as in (G). 100% = 5058 cpm in 6 ml of total reaction mixture 30 ml (closed circle at 30 min). Closed symbols, with template; open symbols, without template. © 2004 Blackwell Publishing Ltd, Molecular Microbiology, 54, 1011–1021


[35S]-methionine (cpm per 30 ml)

A

Kept at 28°C Exposed to 42°C for 20 min

8 x 10 4

Exposed to 42°C for 1.5 hrs

4 x 10 4

Kept at 42°C after temp. shift

0 –60

0

60 120 180 240 Time (min)

B

Kept at 28°C

OD 540

0.8

Exposed to 42°C for 30 min

0.6

Exposed to 42°C for 2 h

0.4

Exposed to 42°C for 3 h Kept at 42°C after temp. shift

0.2 0 –240 –120 0 120 240 360 Time (min)

C

No Tet was added

1 OD 540

0.8 Exposed to Tet f or 1.5 h then Tet was washed out.

0.6 0.4

Tet was added

0.2 0 –120

0 120 240 Time (min)

Fig. 6. Recovery from the inactivation of RRF or tetracycline treatment. A and B. LJ14 grown in minimal media at 28∞C was exposed to 42∞C at zero time for 20 (closed circles in A), 30 (open triangles in B), 90 (closed triangles in A), 120 (closed upside-down triangles in B) or 180 (closed diamonds in B) min. Then, the culture temperature was shifted back to 28∞C (indicated by the arrows). Open circles represent cells kept at 28∞C and open squares represent those kept at 42∞C after the temperature shift. (A) and (B) represent the incorporation of [35S]-methionine and cell density at 540 nm (OD540) respectively. The specific activity of [35S]-methionine used in (A) was 41.5 mCi mmol-1. C. Recovery from tetracycline (Tet) treatment. MC1061 was grown in minimal media at 28∞C until the OD540 reaches 0.2 (zero time), then tetracycline (30 mg ml-1) was added (closed circles and open triangles). After 1.5 h, cells represented by closed circles were washed and resuspended in the fresh minimal media and the recovery of the cell growth was observed. As controls, tetracycline was not added at zero time (open circles), or added tetracycline was not washed out (open triangles).

cells to resume growth. We conclude that in vivo recovery period of protein synthesis after loss of RRF is not unusual. Discussion In the preceding paper describing tsRRF, we showed that it took as long as 60 min at 45∞C to inactivate the purified tsRRF from LJ14 (Janosi et al., 1998). In the present communication we show that, in parallel with the rapid cessation of viable counts, protein synthesis decreased very rapidly (Fig. 1A). These observations can now be understood in light of the present finding that tsRRF disappears rapidly in cells upon exposure to the non© 2004 Blackwell Publishing Ltd, Molecular Microbiology, 54, 1011–1021

permissive temperature (Fig. 2) possibly because of the action of proteases recognizing unfolded protein (Wickner et al., 1999). Partial unfolding is probably not reflected on the in vitro activity but could make RRF susceptible to proteolysis in vivo. Protein synthesis decreased before the almost equally rapid cessation of transcription which follows a transient stimulatory effect (Fig. 1I). This unexpected rapid decrease of labelled methionine incorporation did not result from loss of amino acid uptake at the non-permissive temperature because proteins coded for by leaderless mRNA was synthesized almost normally at the non-permissive temperature (Moll et al., 2004). We then asked the question where ribosomes are during this rapid decrease of protein synthesis. We suggest that, on the basis of the following considerations, ribosomes engaged in unscheduled translation are released from the 3¢ end of mRNA and stay as 70S ribosomes without mRNA. First, upon loss of RRF at the nonpermissive temperature, we observed the appearance of small peptides (4–10 kDa, corresponding approximately to 40–100 residues) containing tmRNA encoded tagged peptide (Fig. 3). From the size of these oligopeptides they must represent the 3¢ end of UTR (see Results section pertaining to Fig. 3). Therefore, these data suggest that some ribosomes are transferred to tmRNA from the 3¢ end of mRNA. Second, at 42∞C, most of the ribosomes in LJ14 are monosomes (Fig. 4). The results indicate that ribosomes do not accumulate on mRNA or tmRNA. The 3¢ UTR region of tmRNA is 242 nucleotides (Accession No. D12501). The minimum distance between ribosomes in E. coli polysome is estimated to be about 40 nucleotides from the length of nucleotides covered by the ribosome (Steitz, 1969) to 100 nucleotides estimated from the frequency of initiation per second (Sorensen and Pedersen, 1991) and the speed of peptide formation (about 13 peptide bonds per second) (Sorensen et al., 1989). Therefore, we would have observed increase of trisomes or tetrasomes if the release of ribosomes from the 3¢ end of tmRNA is inhibited. Our interpretation that ribosomes fall off the 3¢ end of mRNA is not contradictory to the practice of ribosome display (Mattheakis et al., 1994; Hanes and Pluckthun, 1997) which depends on the concept that translating ribosomes do not fall off from the 3¢ end because their conditions are not physiological. Although recent report claims that ribosomes are stalled at the 3¢ end of truncated mRNA (Gilbert et al., 2004), no direct evidence was presented. Our data presented in Fig. 3 suggest that many tmRNAtagged proteins are made. This is consistent with the findings by others that numerous proteins are tagged (Karzai et al., 1999; Abo et al., 2000; Roche and Sauer,

1018 G. Hirokawa et al. 2001) even in the middle of mRNA (Keiler et al., 1996; Withey and Friedman, 2002). When ribosomes pause on mRNA either in the middle of an ORF (at a rare codonlike AGA) or at certain termination codons, they can be transferred to tmRNA (Roche and Sauer, 1999; 2001; Hayes et al., 2002a,b). The ribosomes at rare codons or at the termination codons are known to be stalled (Jin et al., 2002) and these stalled ribosomes are rescued by tmRNA. Close examination of Fig. 3 shows that, upon loss of RRF, the number of his-tagged proteins remained the same but the amount of tagging (density of each band) increased with some proteins (bands with the arrows). We interpret this increase of density to suggest that loss of RRF increases the chance of the rescue by tmRNA indicating that RRF may compete with tmRNA for the rescue of stalled ribosomes. It is known that peptidyl-tRNA is released often from ribosomes at rare codons (OlivaresTrejo et al., 2003) or at the termination codons (Cruz-Vera et al., 2003). In vitro experiment showed that RRF can release peptidyl-tRNA from ribosomes (Ryoji, 1981; Heurgue-Hamard et al., 1998) to relieve stalled ribosomes. Taken together, RRF appears to share the function of tmRNA to relieve stalled ribosomes (Roche and Sauer, 1999; 2001; Hayes et al., 2002a,b). In support of this concept, Hayes et al. (2002a) reported that overexpression of RRF caused a small decrease of tmRNA tagging at the in frame stop codon of ybeL mRNA. We have demonstrated that in vivo decrease of protein synthesis in the absence of RRF results from the pellet fraction including ribosomes. The inhibition must be related to dissociation of ribosomes into subunits for next round of translation because poly(U)-dependent polyphenylalanine synthesis which does not require dissociation was normal. This is consistent with our separate observation that leaderless mRNA-dependent synthesis of protein, which also does not require ribosomal subunits dissociation, was relatively intact in vivo in LJ14 at the non-permissive temperature (Moll et al., 2004). The recovery time from the damage caused by the loss of RRF is the same as that from the recovery from a simple inhibition by antibiotics (Fig. 6). It appears therefore that the damage may be relatively simple and closely related to the mechanism of ribosome dissociation.

Experimental procedures Bacterial strains and plasmids MC1061 [F– wild-type frr araD139 (ara-leu)7679 (lacIPOZYA)X74 galU galK hsdR2 mrBl rpsL(Smr)] and LJ14 (MC1061, temperature-sensitive frr) are from Janosi et al. (1998). MC1061 DssrA and LJ14 DssrA were constructed by P1 phage transduction from W3110 DssrA (Komine et al., 1994). Deletion of the ssrA gene (gene coding for tmRNA)

was verified by polymerase chain reaction (PCR) amplification of genomic DNA. Plasmid pKW24 (Roche and Sauer, 2001), which encodes His6-mutant tmRNA, was kindly supplied by Dr R.T. Sauer (Massachusetts Institute of Technology).

Methionine, uridine and thymidine incorporation in vivo LJ14 and MC1061 were grown overnight at 28∞C in M9 minimal media (Miller, 1972) supplemented with 0.2% (w/ v) glucose, 1 mg ml-1 thiamine and 50 mg ml-1 each of the 19 amino acids (minus methionine). Cells were diluted to 0.01 OD540 and divided into eight 1.7 ml aliquots (named A through H). They were grown at 28∞C for 3 h (LJ14) or 2.5 h (MC1061) until OD540 reached 0.03. Then, mixtures of methionine (20 mM), uridine (300 mM) and thymidine (200 mM) were added. To tubes A and B, [35S]-labelled methionine (250 mCi mmol-1) was added (8.5 mCi per 1.7 ml culture), to tubes C and D [3H]-labelled uridine (26.3 mCi mmol-1) was added (13.4 mCi per 1.7 ml culture), and to tubes E and F [3H]-labelled thymidine (39.5 mCi mmol-1) was added (13.4 mCi per 1.7 ml culture). In tubes G and H, methionine, uridine and thymidine were unlabelled. Cold (C, D, E and F) and hot (A and B) trichloroacetic acid (TCA)-insoluble radioactivity of the 50 ml culture aliquot was determined by the filter disk technique (Mans and Novelli, 1960), and the optical density at 540 nm was measured in tubes G and H. In the experiment described in Fig. 6A and B, LJ14 was grown in minimal media (described above) until OD540 reached 0.25. Then [35S]-methionine (41.5 mCi mmol-1) was added to a final concentration of 200 mM (12.5 mCi per 1.5 ml of culture). Hot TCA-insoluble radioactivity of 30 ml culture aliquot and optical density of the culture at 540 nm were determined.

Western blot Cells in 1 ml of 0.2 OD540 culture were harvested and the total protein was precipitated by 10% TCA. Extracts were loaded on 13.5% SDS-PAGE, blotted onto PVDF membrane and RRF was detected by rabbit anti-serum against E. coli RRF (1:15 000 dilutions). To detect tmRNA-His6-tagged proteins, proteins were loaded on a 10–20% gradient SDS-PAGE, blotted onto PVDF membrane and detected by partially purified rabbit antibody against His6-tag (Affinity Bioreagents).

Ribosome profile analysis Cells were treated with 100 mg ml-1 chloramphenicol (CM) for 30 s then quickly chilled. Cell extract was prepared essentially as described by Flessel et al. (1967) with the following modifications. Cell pellets were suspended in 4.5 ml of a buffer (500 mM sucrose, 100 mM Tris-Cl pH 8.0, 100 mM NaCl) containing 100 mg ml-1 CM. EDTA (10 mM) and lysozyme (100 mg ml-1) were then added. After incubation on ice for 10 min, MgSO4 (final 10 mM) was added and centrifuged for 10 min to collect spheroplasts. The pellet was suspended in 450 ml of a buffer [10 mM Tris-Cl pH 7.6, 10 mM MgSO4, 50 mM NH4Cl, 1 mM dithiothreitol (DTT)] containing 50 mg ml-1 DNase I and 100 mg ml-1 CM. Then, 50 ml of 5% © 2004 Blackwell Publishing Ltd, Molecular Microbiology, 54, 1011–1021

In vivo inactivation of E. coli RRF 1019 Brij58 were added to disrupt the cells. Cell debris was removed by centrifugation for 10 min at 12 000 g. The supernatant was layered onto 4.5 ml of a 15–30% linear sucrose gradient in a buffer containing 10 mM Tris-Cl pH 7.6, 10 mM MgSO4, 50 mM NH4Cl and 0.5 mM DTT. The sucrose gradients were centrifuged at 195 000 g (40 000 r.p.m., Beckman SW 50.1) for 75 min. To analyse the ribosomal sedimentation pattern, the A254 was measured by an ISCO UA-6 spectrophotometer.

Isolation of S30 extract Cells (1 l, mid-log growing phase culture, 0.6–1.0 OD540) were quickly chilled and harvested by centrifugation at 5000 g for 10 min at 4∞C. The cell pellet was washed in 40 ml of TrisMg buffer [10 mM Tris-Cl pH 7.6, 10 mM Mg(OAc)2]. The cell pellet was weighed and disrupted by grinding with a 2.5-fold weight of alumina (SIGMA) at 4∞C. The mixture was suspended in a 1.5-fold weight of Tris-Mg buffer containing 5 mg ml-1 DNase I and centrifuged at 17 000 g for 10 min at 4∞C. To the supernatant, b-ME was added to a final concentration at 6 mM. The extract was centrifuged at 30 000 g for 30 min and the supernatant was dialysed against S30 buffer (10 mM Tris-Cl pH 7.6, 10 mM Mg(OAc)2, 30 mM KCl, 6 mM b-ME). For preparation of S30 extract of LJ14 exposed to 42∞C for 20 min (see Fig. 5), culture fluid (500 ml) was heated in 25 tubes (20 ml each) to raise the culture temperature effectively, then combined for further isolation of the extract.

Cell-free translation programmed by MS2 RNA or poly(U) S30 (0.6 A260U) extract was incubated with 100 mM each of 19 amino acids (minus methionine), 20 pmol of [35S]-methionine [1.2 mCi pmol-1 (Fig. 5A–C and E) or 0.08 mCi pmol-1 (Fig. 5D)] and 12 ml of S30 premixture (Promega) in a total of 30 ml in the presence or absence of 2.4 mg of MS2 phage RNA (Boehringer Mannheim) at 30∞C. Hot TCA-insoluble radioactivity was measured. For poly(U)programmed formation of polyphenylalanine (Fig. 5G and H), S30 (0.6 A260U) was incubated with 30 mM [14C]phenylalanine (0.33 mCi nmol-1), 1.4 mg ml-1 tRNAmix (Sigma), 10 mM phosphoenol-pyruvate, 16.7 mg ml-1 PEP kinase, 3.4 mM ATP and 1.6 mM GTP in 30 ml buffer (25 mM Tris-Cl pH 7.6, 14.7 mM Mg(OAc)2, 54 mM NH4Cl, 18 mM KCl, 0.6 mM DTT, 2.8 mM b-ME) in the presence or absence of 10 mg of poly(U) at 30∞C. Hot TCA-insoluble radioactivity was measured by the filter disk technique.

Acknowledgements We thank Dr Robert T. Sauer of the Massachusetts Institute of Technology for providing the tmRNA-His6 expressing plasmid (pKW24), Dr Maria D. Ermolaeva of the Institute of Genomic Research (TIGR) for providing putative E. coli Rhoindependent terminator sequences, and Dr V. Samuel Raj of University of Pennsylvania for providing purified IF3. We also thank Dr Akikazu Hirashima of Yakult Pharmaceutical, Dr Yoshio Inokuchi of Teikyo University for helpful discussions, and Dr Michael C. Kiel of University of Pennsylvania for helpful discussions and for critical reading of the manuscript. This work was supported in part by a NIH Grant GM60429 © 2004 Blackwell Publishing Ltd, Molecular Microbiology, 54, 1011–1021

(to A.K.), Nippon Paint Research Fund (to H.K.), grant-in-aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology, Japan (to K.I.), and grant-in-aid for Japan Society for the Promotion of Science (JSPS) Fellows (to G.H.).

References Abo, T., Inada, T., Ogawa, K., and Aiba, H. (2000) SsrAmediated tagging and proteolysis of LacI and its role in the regulation of lac operon. EMBO J 19: 3762–3769. Agrawal, R.K., Sharma, M.R., Kiel, M.C., Hirokawa, G., Booth, T.M., Spahn, C.M.T., et al. (2004) Visualization of ribosome-recycling factor on the Escherichia coli 70S ribosome: functional implications. Proc Natl Acad Sci USA 101: 8900–8905. Blumenthal, T. (1979) Qb RNA replicase and protein synthesis elongation factors EF-Tu and EF-Ts. Methods Enzymol LX: 628–638. Cruz-Vera, L.R., Hernandez-Ramon, E., Perez-Zamorano, B., and Guarneros, G. (2003) The rate of peptidyl-tRNA dissociation from the ribosome during minigene expression depends on the nature of the last decoding interaction. J Biol Chem 278: 26065–26070. Ermolaeva, M.D., Khalak, H.G., White, O., Smith, H.O., and Salzberg, S.L. (2000) Prediction of transcription terminators in bacterial genomes. J Mol Biol 301: 27–33. Felden, B., Himeno, H., Muto, A., McCutcheon, J.P., Atkins, J.F., and Gesteland, R.F. (1997) Probing the structure of the Escherichia coli 10Sa RNA (tmRNA). RNA 3: 89–103. Flessel, C.P., Ralph, P., and Rich, A. (1967) Polyribosomes of growing bacteria. Science 158: 658–660. Freistroffer, D.V., Pavlov, M.Y., MacDougall, J., Buckingham, R.H., and Ehrenberg, M. (1997) Release factor RF3 in E. coli accelerates the dissociation of release factors RF1 and RF2 from the ribosome in a GTP-dependent manner. EMBO J 16: 4126–4133. Fujiwara, T., Ito, K., Nakayashiki, T., and Nakamura, Y. (1999) Amber mutations in ribosome recycling factors of Escherichia coli and Thermus thermophilus: evidence for C-terminal modulator element. FEBS Lett 447: 297–302. Fujiwara, T., Ito, K., and Nakamura, Y. (2001) Functional mapping of ribosome-contact sites in the ribosome recycling factor: a structural view from a tRNA mimic. RNA 7: 64–70. Gilbert, R.J., Fucini, P., Connell, S., Fuller, S.D., Nierhaus, K.H., Robinson, C.V., et al. (2004) Three-dimensional structures of translating ribosomes by cryo-EM. Mol Cell 14: 57–66. Hanes, J., and Pluckthun, A. (1997) In vitro selection and evolution of functional proteins by using ribosome display. Proc Natl Acad Sci USA 94: 4937–4942. Hayes, C.S., Bose, B., and Sauer, R.T. (2002a) Proline residues at the C-terminus of nascent chains induce SsrAtagging during translation termination. J Biol Chem 277: 33825–33832. Hayes, C.S., Bose, B., and Sauer, R.T. (2002b) Stop codons preceded by rare arginine codons are efficient determinants of SsrA tagging in Escherichia coli. Proc Natl Acad Sci USA 99: 3440–3445. Heurgue-Hamard, V., Karimi, R., Mora, L., MacDougall, J.,

1020 G. Hirokawa et al. Leboeuf, C., Grentzmann, G., et al. (1998) Ribosome release factor RF4 and termination factor RF3 are involved in dissociation of peptidyl-tRNA from the ribosome. EMBO J 17: 808–816. Hirashima, A., and Kaji, A. (1972) Factor-dependent release of ribosomes from messenger RNA – requirement for two heat-stable factors. J Mol Biol 65: 43–58. Hirokawa, G., Kiel, M.C., Muto, A., Kawai, G., Igarshi, K., Kaji, H., and Kaji, A. (2002a) Binding of ribosome recycling factor to ribosomes – comparison with tRNA. J Biol Chem 277: 35847–35852. Hirokawa, G., Kiel, M.C., Muto, A., Selmer, M., Raj, V.S., Liljas, A., et al. (2002b) Post-termination complex disassembly by ribosome recycling factor, a functional tRNA mimic. EMBO J 21: 2272–2281. Janosi, L., Shimizu, I., and Kaji, A. (1994) Ribosome recycling factor (ribosome releasing factor) is essential for bacterial growth. Proc Natl Acad Sci USA 91: 4249–4253. Janosi, L., Mottagui-Tabar, S., Isaksson, L.A., Sekine, Y., Ohtsubo, E., Zhang, S., et al. (1998) Evidence for in vivo ribosome recycling, the fourth step in protein biosynthesis. EMBO J 17: 1141–1151. Janosi, L., Mori, H., Sekine, Y., Abragan, J., Janosi, R., Hirokawa, G., and Kaji, A. (2000) Mutations influencing the frr gene coding for ribosome recycling factor (RRF). J Mol Biol 295: 815–829. Jin, H., Bjornsson, A., and Isaksson, L.A. (2002) Cis control of gene expression in E. coli by ribosome queuing at an inefficient translational stop signal. EMBO J 21: 4357– 4367. Kaji, A., Kiel, M.C., Hirokawa, G., Muto, A., Inokuchi, Y., and Kaji, H. (2001) The fourth step of protein synthesis: disassembly of the post termination complex is catalyzed by elongation factor G and ribosome recycling factor, a nearperfect mimic of tRNA. Cold Spring Harb Symp Quant Biol 66: 515–529. Kanai, T., Takeshita, S., Atomi, H., Umemura, K., Ueda, M., and Tanaka, A. (1998) A regulatory factor, Fil1p, involved in derepression of the isocitrate lyase gene in Saccharomyces cerevisiae – a possible mitochondrial protein necessary for protein synthesis in mitochondria. Eur J Biochem 256: 212–220. Karzai, A.W., Susskind, M.M., and Sauer, R.T. (1999) SmpB, a unique RNA-binding protein essential for the peptide-tagging activity of SsrA (tmRNA). EMBO J 18: 3793–3799. Keiler, K.C., Waller, P.R.H., and Sauer, R.T. (1996) Role of a peptide tagging system in degradation of proteins synthesized from damaged messenger RNA. Science 271: 990–993. Kiel, M.C., Raj, V.S., Kaji, H., and Kaji, A. (2003) Release of ribosome-bound ribosome recycling factor by elongation factor G. J Biol Chem 278: 48041–48050. Kim, K.K., Min, K., and Suh, S.W. (2000) Crystal structure of the ribosome recycling factor from Escherichia coli. EMBO J 19: 2362–2370. Komine, Y., Kitabatake, M., Yokogawa, T., Nishikawa, K., and Inokuchi, H. (1994) A tRNA-like structure is present in 10Sa RNA, a small stable RNA from Escherichia coli. Proc Natl Acad Sci USA 91: 9223–9227. Kung, H.-F., Treadwell, B.V., Spears, C., Tai, P.-C., and

Weissbach, H. (1977) DNA-directed synthesis in vitro of bgalactosidase: requirement for a ribosome release factor. Proc Natl Acad Sci USA 74: 3217–3221. Lancaster, L., Kiel, M.C., Kaji, A., and Noller, H.F. (2002) Orientation of ribosome recycling factor in the ribosome from directed hydroxyl radical probing. Cell 111: 129– 140. Mans, R.J., and Novelli, G.D. (1960) A convenient, rapid and sensitive method for measuring the incorporation of radioactive amino acids into protein. Biochem Biophys Res Commun 3: 540–543. Mattheakis, L.C., Bhatt, R.R., and Dower, W.J. (1994) An in vitro polysome display system for identifying ligands from very large peptide libraries. Proc Natl Acad Sci USA 91: 9022–9026. Miller, J.H. (1972) Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Moll, I., Hirokawa, G., Kiel, M.C., Kaji, A., and Blasi, U. (2004) Translation initiation with 70S ribosomes: an alternative pathway for leaderless mRNAs. Nucleic Acids Res 32: 3354–3363. Nakano, H., Yoshida, T., Uchiyama, S., Kawachi, M., Matsuo, H., Kato, T., et al. (2003) Structure and binding mode of a ribosome recycling factor (RRF) from mesophilic bacterium. J Biol Chem 278: 3427–3436. Nirenberg, M.W., and Matthaei, J.H. (1961) The dependence of cell-free protein synthesis in E. coli upon naturally occuring or synthetic template RNA. Pros Natl Acad Sci USA 47: 1588–1602. Ogawa, K., and Kaji, A. (1975) Ribosome run through of the termination codon in the absence of the ribosome releasing factor. Biochim Biophys Acta 402: 288–296. Olivares-Trejo, J.J., Bueno-Martinez, J.G., Guarneros, G., and Hernandez-Sanchez, J. (2003) The pair of arginine codons AGA AGG close to the initiation codon of the lambda int gene inhibits cell growth and protein synthesis by accumulating peptidyl-tRNAArg4. Mol Microbiol 49: 1043–1049. Pavolov, M.Y., Freistroffer, D.V., MacDougall, J., Buckingham, R.H., and Ehrenberg, M. (1997) Fast recycling of Escherichia coli ribosomes requires both ribosome recycling factor (RRF) and release factor RF3. EMBO J 16: 4134–4141. Roche, E.D., and Sauer, R.T. (1999) SsrA-mediated peptide tagging caused by rare codons and tRNA scarcity. EMBO J 18: 4579–4589. Roche, E.D., and Sauer, R.T. (2001) Identification of endogenous SsrA-tagged proteins reveals tagging at positions corresponding to stop codons. J Biol Chem 276: 28509– 28515. Rolland, N., Janosi, L., Block, M.A., Shuda, A., Teyssier, E., Miege, C., et al. (1999) Plant ribosome recycling factor homologue is a chloroplastic protein and is bactericidal in Escherichia coli carrying temperature-sensitive ribosome recycling factor. Proc Natl Acad Sci USA 96: 5464–5469. Ryoji, M. (1981) Studies on the roles of ribosome releasing factor of Escherichia coli. PhD Thesis. University of Pennsylvania School of Medicine. Ryoji, M., Berland, R., and Kaji, A. (1981a) Reinitiation of © 2004 Blackwell Publishing Ltd, Molecular Microbiology, 54, 1011–1021

In vivo inactivation of E. coli RRF 1021 translation from the triplet next to the amber termination codon in the absence of ribosome-releasing factor. Proc Natl Acad Sci USA 78: 5973–5977. Ryoji, M., Karpen, J.W., and Kaji, A. (1981b) Further characterization of ribosome releasing factor and evidence that it prevents ribosomes from reading through a termination codon. J Biol Chem 256: 5798–5801. Selmer, M., Al-Karadaghi, S., Hirokawa, G., Kaji, A., and Liljas, A. (1999) Crystal structure of Thermotoga maritima ribosome recycling factor: a tRNA mimic. Science 286: 2349–2352. Shimizu, Y., Inoue, A., Tomari, Y., Suzuki, T., Yokogawa, T., Nishikawa, K., and Ueda, T. (2001) Cell-free translation reconstituted with purified components. Nat Biotechnol 19: 751–755. Sorensen, M.A., and Pedersen, S. (1991) Absolute in vivo translation rates of individual codons in Escherichia coli. The two glutamic acid codons GAA and GAG are translated with a threefold difference in rate. J Mol Biol 222: 265–280. Sorensen, M.A., Kurland, C.G., and Pedersen, S. (1989) Codon usage determines translation rate in Escherichia coli. J Mol Biol 207: 365–377. Steitz, J.A. (1969) Polypeptide chain initiation: nucleotide

© 2004 Blackwell Publishing Ltd, Molecular Microbiology, 54, 1011–1021

sequences of the three ribomosal binding sites in bacteriophage R17 RNA. Nature 224: 957–964. Teixeira-Gomes, A.P., Cloeckaert, A., and Zygmunt, M.S. (2000) Characterization of heat, oxidative, and acid stress responses in Brucella melitensis. Infect Immun 68: 2954– 2961. Teyssier, E., Hirokawa, G., Tretiakova, A., Jameson, B., Kaji, A., and Kaji, H. (2003) Temperature sensitive mutation in yeast mitochondrial ribosome recycling factor (RRF). Nucleic Acid Res 31: 4218–4226. Toyoda, T., Tin, O.F., Ito, K., Fujiwara, T., Kumasaka, T., Yamamoto, M., et al. (2000) Crystal structure combined with genetic analysis of the Thermus thermophilus ribosome recycling factor shows that a flexible hinge may act as a functional switch. RNA 6: 1432–1444. Wickner, S., Maurizi, M.R., and Gottesman, S. (1999) Posttranslational quality control: folding, refolding, and degrading proteins. Science 286: 1888–1893. Withey, J.H., and Friedman, D.I. (2002) The biological roles of trans-translation. Curr Opin Microbiol 5: 154–159. Yoshida, T., Uchiyama, S., Nakano, H., Kashimori, H., Kijima, H., Ohshima, T., et al. (2001) Solution structure of the ribosome recycling factor from Aquifex aeolicus. Biochemistry 40: 2387–2396.