Previous Article | Next Article ![]()
Journal of Bacteriology, July 2007, p. 5393-5398, Vol. 189, No. 14
0021-9193/07/$08.00+0 doi:10.1128/JB.00228-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.


Division of Infectious Diseases, Department of Medicine, Massachusetts General Hospital/Harvard Medical School, Cambridge, Massachusetts
Received 11 February 2007/ Accepted 7 May 2007
|
|
|---|
|
|
|---|
YaeT (Omp85), a highly conserved protein (7, 8), is required for gram-negative outer membrane protein assembly (5, 29, 30). In cells that lack YaeT, membrane insertion, folding, and/or oligomerization of many outer membrane proteins, including LamB, OmpA, TolC, OmpF/C, phospholipase A, and PhoE in Escherichia coli and PorA, PorB, PilQ, FrpB, and OMPLA in Neisseria meningitidis, are defective (5, 23, 28-30). In addition, processing of the neisserial autotransporter IgA1 protease at the cell surface is impaired (28), consistent with a defect in its assembly in the outer membrane. Given these findings, we explored whether YaeT is required for the outer membrane assembly of autotransporters more generally.
YaeT is required for secretion of Shigella flexneri autotransporters IcsA and SepA. Since YaeT is essential in E. coli (29, 30) and N. meningitidis (28), we postulated that it would be essential in S. flexneri. Therefore, we generated a derivative of S. flexneri wild-type strain 2457T (14) in which YaeT could be depleted (Table 1). On solid medium, the depletion strain did not grow in the absence of arabinose because in this strain expression of yaeT is under the control of the arabinose promoter. In rich liquid medium, growth was measured under replete (0.2% arabinose) or depleting (0.1% fucose (19) conditions. Under depletion conditions, the depletion strain's growth rate rapidly dropped off, whereas under replete conditions, it grew as well as the wild-type strain under either condition (Fig. 1A). Thus, under these conditions, yaeT is essential for growth in S. flexneri. At 3 h of depletion, YaeT expression was less than 5% that of the wild-type or of the depletion strain grown under replete conditions (Fig. 1B). Of note, the disruption in yaeT did not have significant polar effects on hlpA (skp), the gene that lies immediately downstream of yaeT and that encodes the periplasmic chaperone Skp, since levels of Skp were unaltered (Fig. 1B).
|
View this table: [in a new window] |
TABLE 1. Bacterial strains and plasmids used in this study
|
![]() View larger version (48K): [in a new window] |
FIG. 1. Secretion of S. flexneri autotransporters IcsA and SepA is dependent on YaeT. (A) YaeT is essential for growth in S. flexneri. Growth curves of wild-type S. flexneri 2457T (squares) and the 2457T YaeT depletion strain (circles) under replete conditions (0.2% arabinose in LB; filled symbols) or depletion conditions (0.1% fucose in LB; open symbols) at 37°C. Time zero represents the start of growth in depleted versus replete medium; 180 min represents the point of back-dilution of cultures into the same medium. (B) Levels of YaeT and Skp in cell pellets of the wild-type or YaeT depletion strain after 3 h of growth under repletion or depletion conditions, with FtsZ as a loading control. Western blots were probed with YaeT, Skp, or FtsZ antibody. (C and D) Levels of IcsA (C) or SepA (D) in culture supernatants or cell pellets of these strains after growth in repletion or depletion medium for 5 h (3 plus 2 h) as described in the text. Western blots were probed with IcsA, SepA, or FtsZ antibody. The absence of FtsZ in the culture supernatants indicates the absence of cell lysis. The images are representative of those obtained for three or more independent experiments. IcsA' or SepA', proteolytically processed mature amino terminus of IcsA or SepA, respectively; IcsA or SepA, mature full-length IcsA or SepA, respectively; ara, 0.2% arabinose; fuc, 0.1% fucose. Molecular size markers are indicated in kilodaltons.
|
To examine the role of YaeT in the secretion of IcsA and SepA, we examined culture supernatants for proteolytically processed amino-terminal domains of protein that had been newly secreted after depletion of YaeT. Cells were grown for 3 h under conditions of depletion or repletion, washed to remove spent supernatant, resuspended in fresh medium, and grown an additional 2 h. Proteins that then accumulated in the culture supernatants of cells grown under depletion conditions represent proteins secreted after depletion of YaeT. The processed amino-terminal domains of IcsA and SepA were both dramatically reduced in the culture supernatants of YaeT-depleted cells compared to those of replete or wild-type cells (Fig. 1C and D), indicating that YaeT is required for proper expression and/or secretion of IcsA and SepA in S. flexneri.
YaeT depletion leads to DegP-dependent instability of unassembled IcsA. Given the described role for the periplasmic protease/chaperone DegP in the efficient secretion of IcsA (22) and in proteolytic degradation of misassembled outer membrane proteins and given the observation that depletion of YaeT leads to accumulation of some outer membrane proteins in the periplasm (29), we tested whether the observed reduction in levels of IcsA might be due to DegP-dependent degradation. We generated derivatives of strain 2457T in which either DegP alone or both DegP and YaeT could be depleted (Table 1). Under depletion conditions, the DegP depletion strain and DegP YaeT double depletion strain were defective for growth at 42°C (data not shown), as previously reported for E. coli degP mutants (15, 26). Under depletion conditions at 30°C, the growth rate of the DegP depletion strain was similar to that of the wild-type, but that of the double depletion strain dropped off more rapidly than that of the YaeT depletion strain (Fig. 2A), as has been noted previously in E. coli (29). The decrease in viability of cells lacking both YaeT and DegP compared with cells lacking YaeT alone demonstrates that in the absence of YaeT, DegP activity increases the time to cell death, indicating that DegP provides an advantage to these cells.
![]() View larger version (71K): [in a new window] |
FIG. 2. DegP-dependent instability of unassembled IcsA following depletion of YaeT in S. flexneri. (A) Growth drops off more rapidly with depletion of both DegP and YaeT than with depletion of YaeT alone. Growth curves of wild-type S. flexneri 2457T (squares), a YaeT depletion strain (circles), a DegP depletion strain (triangles), or a YaeT and DegP double depletion strain (diamonds) under replete conditions (filled symbols) or depletion conditions (open symbols) at 37°C. Growth was as described in the legend of Fig. 1. (B) Levels of IcsA in cell pellets of these strains after growth in repletion or depletion medium for 3 h (1.5 h plus 1.5 h) as described in the text. Western blots were probed with IcsA or FtsZ antibody. (C) Detection of IcsA on the surface of intact bacteria. Indirect immunofluorescence using antibody to IcsA (left panels) and phase-contrast (right panels) images are shown. Bar, 10 um. (D) Heat-modifiable mobility of IcsA. Western blots were probed with IcsA antibody. The two blots were run in parallel using samples that were prepared at the same time. Strains and growth conditions for panels C and D are as in panel B. uIcsA, unfolded IcsA; fIcsA, folded IcsA; IcsA', proteolytically processed mature amino terminus of IcsA; ara, 0.2% arabinose; fuc, 0.1% fucose. Molecular size markers are indicated in kilodaltons.
|
For these experiments, since cell lysis occurred after prolonged growth under depletion conditions, the time of depletion and of growth in fresh medium were reduced to 3.0 h total (1.5 h each); the reduction of YaeT levels at 1.5 h was comparable to that at 3 h (data not shown). For the double depletion strain, IcsA was not detected in the culture supernatant even under replete growth conditions (data not shown); this was surprising, since growth under this condition was not impaired (Fig. 2A), but it might reflect altered stoichiometry of DegP and YaeT, since in this strain, the two proteins are expressed from plasmids of different copy numbers (Table 1).
In the absence of YaeT, IcsA is not exposed at the cell surface. The decrease in processed amino-terminal domains of IcsA and SepA in culture supernatants could represent either a decrease in outer membrane assembly of the protein or a decrease in proteolytic processing per se. To distinguish between these possibilities, we tested whether the mature amino-terminal domain of IcsA was exposed on the surface of YaeT- and DegP-depleted cells, as it normally is on wild-type cells (10, 12). IcsA was detectable on significantly fewer YaeT- and DegP-depleted cells and YaeT-depleted cells and was qualitatively less bright on these cells than on wild-type or DegP depleted cells, indicating that YaeT is required for the translocation of IcsA to the cell surface (Fig. 2C and Table 2). It was not possible to distinguish whether the IcsA on the surface of some YaeT-depleted cells reflected residual full-length IcsA or proteolytically processed IcsA associated with these cell pellets (Fig. 2B). As discussed above, the reason for the relatively low frequency of IcsA secreted by the YaeT DegP depletion strain grown in arabinose is unclear.
|
View this table: [in a new window] |
TABLE 2. Detection of IcsA on the surface of intact S. flexneri cells
|
As has been observed for other beta-barrel proteins (27), when cell pellets of wild-type S. flexneri were heated to only 30°C, the major IcsA band migrated faster than when pellets were heated to 100°C (Fig. 2D, left panel, wild type). The faster-migrating band in the 30°C sample likely represents folded and outer membrane-assembled protein, and the slower-migrating band in the 100°C sample likely represents unfolded denatured protein. In contrast to the wild-type strain, upon depletion of both YaeT and DegP, the major band in the 30°C sample migrated at the same rate as unfolded denatured protein (Fig. 2D, right panel, YaeT DegP depletion), indicating that YaeT is required for folding of IcsA. This result is consistent with a requirement for YaeT for the assembly of IcsA in the outer membrane.
Taken together, the accessibility of IcsA to DegP in the periplasm in the absence of YaeT and the loss of the fast-migrating form in the absence of YaeT and DegP indicate that YaeT is required for the assembly of the IcsA beta barrel in the outer membrane. The requirement for YaeT for insertion of two other beta-barrel-containing outer membrane proteins, TolC and OmpA, has been observed (29); like IcsA, each of these accumulates in the periplasm in the absence of YaeT and DegP.
YaeT is required for the secretion of autotransporters from other gram-negative bacilli. To test whether autotransporters from other organisms also require YaeT for secretion, we assayed secretion of the diffusely adherent E. coli autotransporter AIDA-I and the Bordetella sp. autotransporter BrkA in an E. coli YaeT depletion background. Normally, upon its secretion, the proteolytically processed amino-terminal domains of AIDA-I and BrkA remain noncovalently associated with the bacterial surface and, hence, are present in cell pellets (3, 20). Following growth under YaeT depletion conditions, the levels of the proteolytically processed amino-terminal domains of AIDA-I or BrkA in cell pellets were reduced compared to those of wild-type cells or cells grown under replete conditions (Fig. 3A and B), indicating that, as for IcsA and SepA, YaeT is required for outer membrane assembly of AIDA-I and BrkA. Although some species specificity has been observed for YaeT recognition of outer membrane proteins (23), our findings indicate that Bordetella BrkA is sufficiently well recognized by E. coli YaeT for efficient assembly. The levels of full-length mature AIDA-I or BrkA did not increase (Fig. 3B and data not shown), suggesting that, as was observed for IcsA, in the absence of proper outer membrane assembly, these improperly processed autotransporters may be unstable. To determine whether the residual amount of processed BrkA associated with the pellet under fucose conditions represented protein that had been translocated to the cell surface prior to YaeT depletion, at 3 h of depletion we removed surface-associated proteins with proteinase K, as described previously (4), and then examined whether BrkA accumulated upon continued growth under depletion conditions. Proteinase K treatment removed the cell-associated proteolytically processed BrkA (Fig. 3C, left three lanes), without detectable degradation of periplasmic proteins (Fig. 3C, bottom panel). Upon subsequent growth under depletion conditions, no BrkA was associated with the cell pellets (Fig. 3C, right three lanes), indicating that in the absence of YaeT, no BrkA is translocated to the cell surface.
![]() View larger version (43K): [in a new window] |
FIG. 3. YaeT is required for the secretion of autotransporters from gram-negative bacilli other than S. flexneri. Levels of AIDA-I (A) or BrkA (B) in cell pellets of E. coli JCM158 or the isogenic YaeT depletion strain (JCM166) carrying an AIDA-I expression plasmid (p-aidA), a BrkA expression plasmid (p-brkA), or an empty vector after growth in repletion or depletion medium for 3 h. Western blots were probed with AIDA-I, BrkA, or FtsZ antibody. (C) Levels of BrkA associated with cell pellets of a YaeT depletion strain after treatment with proteinase K. Western blots were probed with BrkA or ß-lactamase antibody. All lanes are from the same blot. AIDA-I or BrkA, mature full-length AIDA-I or BrkA, respectively; AIDA-I' or BrkA', proteolytically processed mature amino terminus of AIDA-I or BrkA, respectively; BrkA*, proteinase K cleavage product of BrkA; PK, proteinase K concentration in µg/ml; ara, 0.2% arabinose; fuc, 0.1% fucose. Molecular size markers are indicated in kilodaltons.
|
This study was supported by NIH grants AI35817 (to M.B.G.) and AI061073 (to M.B.G.) and by AHA award 0325770T (to S.J.).
Published ahead of print on 18 May 2007. ![]()
Present address: Department of Periodontics, University of Washington, Seattle, WA 98195. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»