JOURNAL OF BACTERIOLOGY, Dec. 1990, p. 7005-7010 0021-9193/90/127005-06$02.00/0 Copyright © 1990, American Society for Microbiology

Vol. 172, No. 12

Characterization of Cold-Sensitive secY Mutants of Escherichia coli TADASHI BABA,1 ANNICK JACQ,2t EDITH BRICKMAN,2 JON BECKWITH,2 TETSUYA TAURA,1 CHIHARU UEGUCHI,1 YOSHINORI AKIYAMA,1 AND KOREAKI ITO'* Institute for Virus Research, Kyoto University, Kyoto 606, Japan,' and Department of Microbiology and Molecular Genetics, Harvard Medical School, Boston, Massachusetts 021152 Received 19 June 1990/Accepted 2 October 1990

Mutations which cause poor growth at a low temperature, which affect aspects of protein secretion, and which map in or around secY (priA) were characterized. The prlU1012 mutant, previously shown to suppress a secA mutation, proved to have a wild-type secY gene, indicating that this mutation cannot be taken as genetic evidence for the secA-secY interaction. Two cold-sensitive mutants, the secY39 and secY40 mutants, which had been selected by their ability to enhance secA expression, contained single-amino-acid alterations in the same cytoplasmic domain of the SecY protein. Protein export in vivo was partially slowed down by the secY39 mutation at 37 to 39°C, and the retardation was immediately and strikingly enhanced upon exposure to nonpermissive temperatures (15 to 23°C). The rate of posttranslational translocation of the precursor to the OmpA protein (pro-OmpA protein) into wild-type membrane vesicles in vitro was only slightly affected by reaction temperatures ranging from 37 to 15°C, and about 65% of OmpA was eventually sequestered at both temperatures. Membrane vesicles from the secY39 mutant were much less active in supporting pro-OmpA translocation even at 3rC, at which about 20% sequestration was attained. At 15°C, the activity of the mutant membrane decreased further. The rapid temperature response in vivo and the impaired in vitro translocation activity at low temperatures with the secY39 mutant support the notion that SecY, a membrane-embedded secretion factor, participates in protein translocation across the bacterial cytoplasmic membrane.

The secY (priA) gene of Escherichia coli codes for an integral membrane protein that is essential for protein export from the cytosol to the periplasm or the outer membrane. Two classes of mutations in this gene have been characterized previously: priA mutations (10), suppressing signal sequence mutations, and temperature-sensitive secY mutations (15, 16, 25), causing pleiotropic defects in protein export. It was proposed that SecY or PrlA may provide a pathway in the membrane for protein translocation (1, 4, 5, 15). Although the sites of the sec Y and priA mutations have been localized at the nucleotide level (15, 22, 25, 26), characterization and mapping of additional secY mutations causing different phenotypes should be helpful in our understanding of the structure-function relationships of this multipathway (1) membrane protein. Recently, cold-sensitive (Cs) secretion mutants were isolated by selection based on increased expression of secA under conditions of impaired protein secretion, and some of the mutations were mapped in the secY region (21). In the work described here, we characterized two Cs sec Y mutants isolated by the abovedescribed selection as well as the priA1012 mutant isolated as an extragenic suppressor of a secA mutation (6). Sequence analysis showed that the former mutations alter amino acid residues located in the same cytoplasmic loop of SecY, whereas the latter mutation is outside the secY gene. One of the Cs mutants, the secY39 mutant, showed a very rapid response to a temperature shift, and protein secretion was slowed within 1 min upon exposure to nonpermissive temperatures (15 to 230C). Membrane vesicles prepared from wild-type and secY39 mutant cells were compared for their abilities to translocate the precursor to the OmpA protein (pro-OmpA protein) in vitro. On the basis of these analyses,

it is proposed that the sec Y39 mutation may affect some catalytic mechanism of the SecY protein. MATERIALS AND METHODS Bacterial strains. E. coli MC4100 (8) and MM118 (MC4100 rpsE priA1012 [6]) were described previously. CU164 (MC4100 secY39 zhd-33::TnlO) and CU165 (MC4100 secY40 zhd-33::TnlO) were constructed as follows. Strain PR478 (an MC4100 derivative carrying (1(secA': :lacZ)181(Hyb) and lysogenic for X pR9 [21]) was treated with ethyl methanesulfonate and screened for blue colonies on plates containing XG (5-bromo-4-chloro-3-indolyl-p-D-galactopyranoside) and PETG (phenyl-ethyl-thiogalactoside) as described previously (21). Two Cs mutants, named AJ39 and AJ40, thus obtained were shown to have mutations (named sec Y39 and secY40, respectively) linked to a transposon insertion near secY. These mutations were transferred to the fresh background of MC4100 by first introducing zhd-33::TnJO (25) into AJ39 and AJ40 and then transducing the Cs mutations together with TnJO into MC4100. Transductants that received secY39 and secY40 were named CU164 and CU165,

respectively. AD202 was an ompT::Kan (2) transductant of MC4100. AD208 was a zhd-33::TnlO secY39 transductant of AD202. These ompT::Kan derivatives were used for in vitro translocation experiments. Sequencing of the secY gene from the mutant strains. The chromosomal DNA (20) of CU164, CU165, or MM118 was digested with PstI, and 2.8-kb fragments, including those containing secY, were isolated and cloned into pUC118 (27). Clones carrying secY were identified by restriction analysis. Sequencing of the single-stranded DNA (27) was done by the chain termination method (23) with a Sequenase kit purchased from United States Biochemicals. Synthetic oligonucleotides (17-mer) of the wild-type sequences of the secY region with 5' ends at residues 5649, 5375, 5103, 4832, and 4561 of the antisense strand and at residues 4271, 4517, 4786,

* Corresponding author. t Present address: Institut de Microbiologie, Universite de ParisSud, Centre D'Orsay, 91405 Orsay Cedex, France.

7005

7006

J. BACTERIOL.

BABA ET AL.

5058, and 5325 of the sense strand (numbering according to Cerretti et al. [9]) were kindly provided by T. Sako and used as primers for the DNA polymerase reaction. In vivo pulse-chase. Cells were grown in M9 medium (19) containing 0.4% glycerol, 0.2% maltose, and 18 amino acids (20 lkg/ml) other than methionine and cysteine. Pulse-labeling with [35S]methionine (1,000 Ci/mmol; purchased from American Radiolabeled Chemicals) was done for 30 to 90 s at various temperatures. Chase with unlabeled methionine was initiated by adding 200 ,ug of L-methionine per ml. The labeled culture was directly treated with trichloroacetic acid and processed for immunoprecipitation and sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis as described previously (14). Antibodies used were those against maltose-binding protein (MBP), a periplasmic protein and OmpA protein (an outer membrane protein). The precursor and mature forms of these proteins were quantitated by scanning the autoradiograms with a Biomed laser scanning densitometer. Values were corrected for the presence of additional methionine residues in the signal sequences. In vitro translocation of pro-OmpA. An S140 extract was prepared from AD202 cells (MC4100 ompT::Kan) by the method of Zubay (32), with the exception that the second ultracentrifugation was done at 140,000 x g for 30 min. pro-OmpA was synthesized with plasmid pRD87 (13) DNA as the template and by the coupled transcription-translation reaction as described by Yamane et al. (30) with some minor modifications, including the addition of cyclic AMP (1 mM) for enhancement of lac promoter-controlled OmpA transcription. Cytoplasmic membranes were prepared from French press-disrupted AD202 (sec Y) and AD208 (secY39) cells, both of which were ompT defective, by published procedures (31). For the posttranslational translocation assay, the transcription-translation mixture was supplemented with chloramphenicol (100 ,ug/ml) and mixed with a 1/10 volume of membrane vesicles, the concentration of which had been adjusted to 3.4 mg of protein per ml, as determined with the Micro BCA protein assay reagent (Pierce) after solubilization with 1% SDS. After incubation at a temperature specified in each experiment, samples were withdrawn and chilled to 0°C, portions were treated with proteinase K (100 ,ug/ml) for 30 min, and the enzyme was inactivated with 1 mM phenylmethylsulfonyl fluoride. The protease-treated and untreated samples were mixed with an equal volume of 10% trichloroacetic acid, and precipitates were washed with acetone and subjected to SDS-polyacrylamide gel electrophoresis. The pro-OmpA and OmpA bands, the identities of which had been confirmed immunologically (data not shown), were quantitated with the densitometer.

RESULTS Cs mutations in the secY region. Riggs et al. (21) reported on the isolation of a series of mutations which caused increased expression of a secA-lacZ fusion gene. Such mutations were expected to cause some defect in the protein secretion mechanism, because the expression of secA is known to be elevated under conditions of impaired protein secretion. Some of these mutations caused Cs growth phenotypes, and a few of them were mapped in the sec Y locus of the chromosome (21). Additional Cs mutations in this region of the chromosome were obtained by mutagenizing cells of a strain carrying the secA-lacZ fusion gene with ethyl methanesulfonate and screening them for blue colonies on agar plates containing XG and PETG. PETG was included to partially inhibit

P-galactosidase and thereby to reduce the background color (21). The cells were then examined for linkage between the Cs mutation and a TnJO insertion near secY. Two mutants, Cs39 and Cs4O, which exhibited clear cold sensitivity as well as linkage to the transposon, were subjected to further examination. The mutations were introduced into the fresh background of MC4100 by joint P1 transduction with zhd-33::TnJO, and transductants that received the Cs mutation from Cs39 (CU164) or Cs4O (CU165) were used. The inability of CU164 and CU165 to grow at low temperatures (20 to 230C) was complemented by plasmid pKY6 (25) carrying secY' as the sole chromosomal gene. the 2.8-kb PstI fragment of the rplN-rpsM operons was excised from the chromosomal DNA, cloned, and sequenced for the sec Y region. The Cs39 and Cs4O mutants contained a single-basepair change, G C--A * T, at residues 5363 and 5380 of the rplN (spc) operon (numbering according to Cerretti et al. [9]). These mutations are hereafter called secY39 and secY40, respectively. The corresponding changes in the SecY protein product would be Arg-357 to His and Ala-363 to Ser for secY39 and secY40, respectively. These residues reside in the same putative cytoplasmic loop, cytoplasmic domain 5 (1), of SecY (Fig. 1). The prlA1012 mutant contains the wild-type secY gene. Brickman et al. (6) reported an extragenic suppressor mutation, prlA1012, which restored the growth and protein secretion defect of the secASl mutant at 37°C. The mutation was mapped close to the prlA3 mutation by P1 transduction. Although this mutation by itself does not cause any appreciable defect in protein secretion, it causes slow growth, especially at 30°C. We cloned and sequenced the secY gene from this mutant and found no deviation from the wild-type sequence. Thus, the mutation responsible for slow growth and suppression of secA51 should be in a gene closely linked to but distinct from sec Y (priA). Protein secretion in the secY39 mutant. The secY39 and secY40 mutants were examined for their ability to secrete MBP and OmpA by pulse-labeling of cells for 90 s at permissive (37 to 39°C) or nonpermissive (23°C) temperatures, immunoprecipitation, and SDS-polyacrylamide gel electrophoresis. Labeling of the precursor forms indicates a retardation in secretion (translocation) across the cytoplasmic membrane. The secY39 mutant markedly accumulated the precursor forms at 23°C (see below), but the secY40 mutant accumulated only minute amounts of the precursor forms, even after a prolonged incubation (up to 8 h) at the nonpermissive temperature (data not shown). Figure 2 shows the labeling patterns of MBP and OmpA in the secY39 mutant at various times before and after a shift from the permissive (37°C) to the nonpermissive (23°C) temperature. This mutant responded very rapidly to the temperature shift. As early as 1 min after the shift, both MBP and OmpA proteins were labeled predominantly as precursor forms (Fig. 2A, lane 2), and the proportion of the precursor forms increased only slightly after 1 to 2 h of exposure to the nonpermissive temperature (lanes 5 and 6). The pulse-chase kinetics of MBP and OmpA were then determined in wild-type and secY39 mutant cells growing at 39°C as well as after a shift down to 23 or 15°C. We included the temperature of 15°C for comparison with the in vitro results (see below). Cells were pulse-labeled with [35S] methionine for 30 s and chased with unlabeled methionine. Samples were withdrawn every 30 s and processed for immunoprecipitation and gel electrophoresis of OmpA and MBP. The intensities of the precursor and mature forms were measured, and the proportion of the latter was plotted -

COLD-SENSITIVE secY MUTANTS

VOL. 172, 1990

7007

COOH

secY39 (Argi--His)

secY40 (Ala,-Ser)

FIG. 1. Locations of the amino acid residues altered by the sec Y39 and sec Y40 mutations. Shown are the model (1) for the disposition of SecY in the membrane and the mutation sites determined by DNA sequencing.

against the chase length (Fig. 3). Wild-type cells at 390C exhibited rapid and complete secretion of both proteins immediately after their synthesis (Fig. 3A). Export of OmpA molecules in the wild-type cells was delayed at 23 and 15'C, taking, respectively, about 2 min (Fig. 3C, open circles) and 5 min (Fig. 3E, open circles) for about 90% completion. In contrast, export of MBP in the wild-type cells was only slightly affected by low temperatures; even at 15'C about 90% of MBP molecules were secreted after a 30-s chase (Fig. 3E, closed circles). The sec Y39 mutant was slow in exporting MBP and OmpA even at 390C, but almost all OmpA (Fig. 3B, open squares) as well as more than 80% of MBP (Fig. 3B, closed squares)

2

1

A MBP

(m

B

5

6

(PM 40mo-

sF

'z

t

_

Po

r4

OmpA(p

4

-awl

OmpA(P

MBP

3

Aamb..

.-.,

40mm* gmsm-

o

4_

__ m._ ._w

_

FIG. 2. Export of MBP and OmpA in the sec Y39 mutant. Strains CU164 (secY39) (A) and MC4100 (B) were grown at 37°C and then shifted to 23°C. At 0 (lane 1), 1 (lane 2), 2 (lane 3), 3 (lane 4), 60 (lane 5), and 120 (lane 6) min after the shift, portions of the cultures were pulse-labeled with [35S]methionine for 90 s, and MBP and OmpA polypeptides were immunoprecipitated. p, Precursor; m, mature.

molecules were secreted within 1 to 2 min. At 23 and 15'C, the sec Y39 mutant exhibited strikingly slow secretion of both proteins, and less than 50% of these protein molecules were secreted by 2.5 min (at 230C; Fig. 3D, open and closed squares) for 5 min (at 15°C; Fig. 3E, open and closed squares) of chase. At 39 and 23°C, the export of MBP was faster than that of OmpA in the wild-type cells, but the reverse was true in the secY39 mutant cells. We carried out similar pulse-chase experiments at 27 to 35°C and found that the rates of MBP and OmpA export at these temperatures were roughly comparable to those at 39°C (data not shown). Thus, it seems that there is some critical temperature between 27 and 23°C below which the secY39 mutant becomes restricted. In vitro translocation activity of membrane vesicles. The rapid response of the sec Y39 mutant to the temperature shift suggests that, in this mutant, temperature may affect some reaction parameter of translocation in which the altered gene product of secY participates. Such a notion supports the hypothesis that the SecY protein directly catalyzes the transmembrane movement of polypeptides. Thus, this mutant may provide a better system for studying the SecY requirement for protein translocation in vivo and in vitro than does the sec Y24 mutant, the secretion-defective phenotype of which is fully expressed only after about 2 h of growth at the nonpermissive high temperature (25). In vitro reproduction of the sec Y24 defect also requires either in vivo (3) or in vitro (12; Y. Akiyama, unpublished results) preincubation to inactivate the gene product. We carried out in vitro translocation of pro-OmpA at various temperatures by using inverted membrane vesicles prepared from wild-type and secY39 mutant cells that had been grown at 37°C. The bacterial strains used for the in vitro experiments all carried the ompT::Kan mutation to avoid possible cleavage of SecY by the OmpT protease after cell disruption (2). An immunoblotting experiment showed that membrane preparations from the wild-type and secY39

BABA ET AL.

7008

A

J. BACTERIOL.

0

a

-

B

(

1

0.8

2

3

4

5

6

7

8

A

0.6

-P -m

0.4

-P

0.2 [

-m I

EL..

I

BC

I

I

D

C

0

-p

0.84

-m

0) I-

0.6

c

co

E

OA

0

0.2 1

D

-p

-m

C

0 0.

0 0s

E

60

12

60

120

Prot.K Incubation

0.8

0 .1

0.6 . 0.41 0.2 a

60

120

180

240

300

Chase time (sec) FIG. 3. Effects of temperature on the export of MBP and OmpA in vivo. Cells of MC4100 (circles) and CU164 (secY39; squares) were pulse-labeled for 30 s and chased for the indicated periods at 390C (A and B), 230C (C and D), or 15'C (E). In the last two cases, pulse-labeling was initiated at 1 min (C and D) or 5 min (E) after the temperature shift. The ratios of the processed mature form to the precursor plus mature forms were determined for MBP (closed symbols) and OmpA (open symbols) following immunoprecipitation and SDS-polyacrylamide gel electrophoresis. mutant cells contained equivalent amounts of SecY (data not

shown). Figure 4 shows gel patterns of in vitro-synthesized OmpA molecules after posttranslational incubation with inverted cytoplasmic membrane vesicles and with or Without additional treatment with proteinase K to digest untranslocated materials. More detailed time courses were monitored in a different experiment (Fig. 5), in which the extents of translocation (proportions of proteinase K-resistant OmpA species) were plotted against the incubation time. The reaction of 370C with the wild-type membrane rapidly yielded processed and proteinase K-inaccessible (translocated) OmpA; the maximum level (about 65% translocation) was reached in 10 min (Fig. 5, open circles). Proteaseresistant materials were not observed without incubation (Fig. 4, lane 2) or when a membrane-disrupting detergent, Triton X-100, was present during the digestion (data not shown). Also, pro-OmpA remained totally unprocessed and proteinase K accessible after incubation at 370C for 80 min without added membranes (data not shown). In contrast to the above-described results, pro-OmpA translocation was significantly slower and much less complete when membranes from the secY39 mutant were used; after 80 min, the

-

L.

0

+

J-i

L.-. 1 _

-t.3. +

20

40

+

-

80

(min)

FIG. 4. Posttranslational translocation of pro-OmpA into inverted membrane vesicles in vitro. In vitro transcription-translation products directed by pRD87 DNA were incubated at 370C (A and B) or 15'C (C and D) with membrane vesicles for 0 (lanes 1 and 2), 20 (lanes 3 and 4), 40 (lanes 5 and 6), and 80 (lanes 7 and 8) min. Cytoplasmic membrane vesicles were isolated from AD202 (secY ompT::Kan) (A and C) and AD208 (secY39 ompT::Kan) (B and D). Odd-numbered samples were not treated with proteinase K (Prot. K) whereas even-numbered samples were treated with proteinase K before electrophoresis. p, Precursor; m, mature.

proteinase K-inaccessible fraction amounted to only about 20% (Fig. 4B and Fig. 5, open squares). Essentially similar results were obtained at 30 and 23TC (data not shown). In vitro translocation was carried out at an even lower temperature, 15'C. pro-OmpA translocation into the wildtype membrane was only slightly slower at 15'C than at 37TC, and the extent of translocation was unchanged (Fig. 4C and Fig. 5, closed circles). In contrast, the membrane vesicles from the secY39 mutant were extremely inactive in supporting pro-OmpA translocation at 15TC (Fig. 4D and Fig. 5, closed squares). In the above-described experiments, it was observed that whereas virtually all the processed molecules were sequestered in the reactions with the wildtype membranes (Fig. 4A and C), only less than half of the mature-sized product was sequestered into the proteaseinaccessible compartment in the reactions with the mutant membranes (Fig. 4B and D). DISCUSSION Mutations that map in sec Y or its vicinity and cause slow or lethal growth at low temperatures have been characterized. Although the mutation prlAl012 has been believed to be an allele of sec Y (priA) and has often been cited as genetic evidence for an interaction between the gene products of secA and secY, the present results indicate that it is in a gene different from sec Y. It is likely that the suppression of the secA mutation is due to a mutation in a nearby ribosomal protein gene. This idea is reasonable in light of the observations that some mutations in the protein synthesis

COLD-SENSITIVE secY MUTANTS

VOL. 172, 1990

C.) 0co

a) 00 0 04n

0.8

E

0.6

0 C 0 to I-

0.4

Cu c

_

0

SO

0 a 0

0.2

0f a.

7009

In vitro translocation of pro-OmpA into the wild-type membrane vesicles was quite efficient at both 37 and 15'C. In contrast, the sec Y39 membrane was less active even at 370C, and the activity was further reduced at 15'C. This temperature effect was reversible (data not shown), suggesting that some reaction parameter rather than the stability of the gene product was affected by the mutation. Some discrepancy was observed between the temperature profiles of OmpA translocation in vivo and in vitro. While in vivo translocation was significantly retarded at 230C, in vitro reproduction of cold sensitivity required a further lowering of the temperature down to 15'C. It was somewhat unexpected that in vitro translocation of pro-OmpA with the wild-type or the mutant membrane vesicles was not significantly affected by reaction temperatures ranging from 37 to 230C. The in vitro reaction may not faithfully reflect the in vivo parameters of translocation. For instance, some limiting factor in the in vitro system could determine the upper limit of translocation

speed.

20

40

60

80

Incubation time (mmn) FIG. 5. Effects of tennperature and the secY39 mutation on in vitro translocation. In an experiment similar to that shown in Fig. 4, posttranslational transloc ation was allowed for 0, 5, 10, 20, 40, and 80 min, and autoradiogra phic results were quantitated by densitometer tracings. Extents of translocation (ratios of proteinase K-inaccessible OmpA polypeptiides to total precursor plus mature forms of OmpA) were plotted agaiinst the reaction time. Symbols: 0, at 370C; 0, secY' at 15'(C; O, secY39 at 37°C; *, secY39 at 15°C.

secYo

machinery suppress th e protein secretion and growth defects caused by secA or sec: Ytemperature-sensitive mutations (17, 24). Thus, the hypotthesis that the SecA and the SecY proteins interact with each other, for which some biochemical evidence exists (1L1, 18), is yet to be supported genetically. Two Cs mutations, secY39 and secY40, were mapped in the same putative cyto plasmic domain of SecY. The temperature-sensitive sec Y2 4 mutation resides in the preceding cytoplasmic domain (25). These cytoplasmic domains of SecY might be import :ant in recognizing precursor proteins, in interacting with oth er secretion factors in the cytoplasm, and/or in actively trainslocating the precursor. The reason why the secY40 mutarnt, which is clearly Cs in growth, was only slightly defective in export of MBP and OmpA is not clear. It is possible t hat some envelope proteins that are crucial for bacterial siurvival are more strongly affected by secY40. Alternatively, SecY might have a growth-essential role, other than protei in translocation per se, that is affected by secY4O. We noted that ME3P and OmpA responded somewhat differently to the se(cY39 mutation or to a temperature downshift. The sec Y3 9 mutation affected MBP export more strongly than it affect ed OmpA export, whereas a lowered temperature (in wild-t ype cells) affected OmpA export more than MBP export. ThLese observations are consistent with the notion that differe nt proteins can be affected differently by an altered catalyst 4oftranslocation. The rapid response of the secY39 mutant to the temperature change is consistent with the notion that the e residue altered by sec Y39 is involved more directly in the cal atalytic function of SecY than are the residues altered by o1ther secY (temperature-sensitive) mutations.

The initial rate of pro-OmpA translocation into the wildtype membrane at 15'C was roughly half that at 370C. In the case of the sec Y39 membrane, the initial rate at 15'C was too low to measure accurately, but it was far below half the initial rate at 37°C (Fig. 5). The apparent accumulation of the processed but untranslocated molecules (Fig. 4B and D) could mean that the mutant membrane accumulates partially translocated transmembrane molecules. However, a simpler

explanation may be that, with the low translocation activity oftemtn'ebaevscls

h precursor rcro oeue of the mutant membrane vesicles, the molecules which remain outside the vesicles have a chance of being cleaved prematurely by the leader peptidase in some contaminating "right-side-out" vesicles. There has been some controversy in the literature as to the necessity of the SecY function in in vitro translocation reactions. Inactivation of translocation was observed either when membrane vesicles were prepared from the temperature-sensitive sec Y24 mutant grown at a high temperature (3) or when those prepared at a permissive temperature were preincubated at a high temperature (12). However, the addition of excess amounts of SecA to the inactivated sec Y24 membranes led to a recovery of translocation activity (11). Watanabe and Blobel (28) were able to demonstrate an inhibition of translocation by antibodies against SecY, but Watanabe et al. subsequently showed that reconstituted proteoliposomes lacking significant amounts of SecY were active in translocation (29). Brundage et al. (7) recently succeeded in fractionating solubilized membrane components and showed that the active fractions contained SecY and SecE as essential components for the reconstitution of SecA-dependent translocation activity. Although complete resolution of these conflicting results awaits further investigations, the present results with the secY39 mutant support and strengthen the hypothesis that SecY is indeed essential for protein translocation. Our interpretation that a reaction parameter of translocation is directly affected by secY39 needs further clarification, since the results obtained could also be explained by indirect conformational effects of the mutation. ACKNOWLEDGMENTS We thank T. Sako for the synthetic oligonucleotides used as the sequencing primers, U. Henning for pRD87, Y. Anraku for antiserum against the OmpA protein, and H. Yamagishi for providing a French press. This work was supported by grants to K.I. from the Ministry of Education, Science and Culture, Japan, and the Naito Foundation

7010

BABA ET AL.

J. BACTERIOL.

and to J.B. from the National Science Foundation and National Institutes of Health. A.J. was supported by the Massachusetts Division of the American Cancer Society. 17.

1. 2.

3. 4.

5. 6.

7.

8. 9.

10. 11.

12.

13.

14. 15. 16.

LITERATURE CITED Akiyama, Y., and K. Ito. 1987. Topology analysis of the SecY protein, an integral membrane protein involved in protein export in Escherichia coli. EMBO J. 6:3465-3470. Akiyama, Y., and K. Ito. 1990. SecY protein, a membrane embedded secretion factor of E. coli, is cleaved by the OmpT protease in vitro. Biochem. Biophys. Res. Commun. 167:711715. Bacallao, R., E. Crook, K. Shiba, W. Wickner, and K. Ito. 1986. The SecY protein can act post-translationally to promote bacterial protein export. J. Biol. Chem. 261:12907-12910. Bieker, K. L., and T. J. Silhavy. 1990. PrlA is important for the translocation of exported protein across the cytoplasmic membrane of Escherichia coli. Proc. Natl. Acad. Sci. USA 86:968972. Bieker, K. L., and T. J. Silhavy. 1990. PriA (SecY) and PrIG (SecE) interact directly and function sequentially during protein translocation in E. coli. Cell 61:833-842. Brickman, E. R., D. B. Oliver, J. L. Garwin, C. Kumamoto, and J. Beckwith. 1984. The use of extragenic suppressors to define genes involved in protein export in Escherichia coli. Mol. Gen. Genet. 196:24-27. Brundage, L., J. P. Hendrick, E. Schiebel, A. J. M. Driessen, and W. Wickner. 1990. The purified E. coli integral membrane protein SecY/E is sufficient for reconstitution of SecA-dependent precursor protein translocation. Cell 62:649-657. Casadaban, M. 1976. Transposition and fusion of the lac operon to selected promoters in E. coli using bacteriophages lambda and Mu. J. Mol. Biol. 104:541-555. Cerretti, D. P., D. Dean, G. R. Davis, D. M. Bedwell, and M. Nomura. 1983. The spc ribosomal protein operon of Escherichia coli: sequence and cotranscription of the ribosomal protein genes and a protein export gene. Nucleic Acids Res. 11:25992616. Emr, S. D., S. Hanley-Way, and T. J. Silhavy. 1981. Suppressor mutations that restore export of a protein with a defective signal sequence. Cell 23:79-88. Fandl, J. P., R. Cabelli, D. Oliver, and P. C. Tai. 1988. SecA suppresses the temperature-sensitive secY24 defect in protein translocation in Escherichia coli membrane vesicles. Proc. Natl. Acad. Sci. USA 85:8953-8957. Fandl, J. P., and P. C. Tai. 1987. Biochemical evidence for the sec Y24 defect in Escherichia coli protein translocation and its suppression by soluble cytoplasmic factors. Proc. Natl. Acad. Sci. USA 84:7448-7452. Freudl, R., H. Schwarz, M. Klose, N. R. Movva, and U. Henning. 1985. The nature of information, required for export and sorting, present within the outer membrane protein OmpA of Escherichia coli K-12. EMBO J. 4:3593-3598. Ito, K., P. J. Bassford, and J. Beckwith. 1981. Protein localization in E. coli: is there a common step in the secretion of periplasmic and outer-membrane proteins? Cell 24:707-717. Ito, K., Y. Hirota, and Y. Akiyama. 1989. Temperature-sensitive sec mutants of Escherichia coli: inhibition of protein export at the permissive temperature. J. Bacteriol. 171:1742-1743. Ito, K., M. Wittekind, M. Nomura, K. Shiba, T. Yura, A. Miura,

18.

19. 20.

21. 22.

23. 24. 25.

26.

27.

28.

29.

30.

and H. Nashimoto. 1983. A temperature-sensitive mutant of E. coli exhibiting slow processing of exported proteins. Cell 32: 789-797. Lee, C. A., and J. Beckwith. 1986. Suppression of growth and protein secretion defects in Escherichia coli secA mutants by decreasing protein synthesis. J. Bacteriol. 166:878-883. Lill, R., K. Cunningham, L. A. Brundage, K. Ito, D. Oliver, and W. Wickner. 1989. SecA protein hydrolyzes ATP and is an essential component of the protein translocation ATPase of Escherichia coli. EMBO J. 8:961-966. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Owen, R. J., and P. Borman. 1987. A rapid biochemical method for purifying high molecular weight bacterial chromosomal DNA for restriction enzyme analysis. Nucleic Acids Res. 15: 3631. Riggs, P. D., A. I. Derman, and J. Beckwith. 1988. A mutation affecting the regulation of a secA-lacZ fusion defines a new sec gene. Genetics 118:571-579. Sako, T., and T. lino. 1988. Distinct mutation sites in prlA suppressor mutant strains of Escherichia coli respond either to suppression of signal peptide mutations or to blockage of staphylokinase processing. J. Bacteriol. 170:5389-5391. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467. Shiba, K., K. Ito, and T. Yura. 1986. Suppressors of the sec Y24 mutation: identification and characterization of additional ssy genes in Escherichia coli. J. Bacteriol. 166:849-856. Shiba, K., K. Ito, T. Yura, and D. P. Cerretti. 1984. A defined mutation in the protein export gene within the spc ribosomal protein operon of Escherichia coli: isolation and characterization of a new temperature-sensitive secY mutant. EMBO J. 3:631-635. Stader, J., S. A. Benson, and T. J. Silhavy. 1986. Kinetic analysis of lamB mutants suggests the signal sequence plays multiple roles in protein export. J. Biol. Chem. 261:1507515080. Vieira, J., and J. Messing. 1987. Production of single-stranded plasmid DNA. Methods Enzymol. 153:3-11. Watanabe, M., and G. Blobel. 1989. Site-specific antibodies against PrlA (SecY) protein of Escherichia coli inhibit protein export by interfering with plasma membrane binding of preproteins. Proc. Natl. Acad. Sci. USA 86:1895-1899. Watanabe, M., C. V. Nicchitta, and G. Blobel. 1990. Reconstitution of protein translocation from detergent-solubilized Escherichia coli inverted vesicles: PrlA protein-deficient vesicles efficiently translocate precursor proteins. Proc. Natl. Acad. Sci. USA 87:1960-1964. Yamane, K., S. Ichihara, and S. Mizushima. 1987. In vitro translocation of protein across Escherichia coli membrane vesicles requires both the proton motive force and ATP. J. Biol.

Chem. 262:2358-2362. 31. Yamato, I., Y. Anraku, and K. Hirokawa. 1975. Cytoplasmic membrane vesicles of Escherichia coli: a simple method for preparing the cytoplasmic and outer membranes. J. Biochem. 77:705-718. 32. Zubay, G. 1973. In vitro synthesis of protein in microbial systems. Annu. Rev. Genet. 7:267-287.