| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 278, Issue 50, 50128-50135, December 12, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
¶
From the
Department of Biochemistry and Molecular Biology, University of Texas, Medical School and Graduate School of Biomedical Sciences, Houston, Texas 77030 and the Department of Microbiology and Immunology, University of Melbourne, Victoria 3010, Australia
Received for publication, September 4, 2003 , and in revised form, October 1, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
The most compelling evidence for a specific role for lipids in membrane protein topological organization is the requirement for phosphatidylethanolamine (PE)1 for the proper orientation of the 12 transmembrane domains (TMs) of the lactose permease LacY of Escherichia coli (3). Assembly in the absence of PE results in a topological inversion of the N-terminal six TMs and their associated extramembrane domains. PE is required in a late step of maturation for the proper folding of the periplasmic extramembrane domain (P7) linking TMs VII and VIII (4). Proper folding of this domain is required for active (but not facilitated) transport of LacY substrates (5). The native topological organization of at least the C6 cytoplasmic domain (connecting TMs VI and VII), the native conformation of the P7 domain, and the active transport function of LacY assembled in vivo in the absence of PE can be restored under the following conditions: post-assembly synthesis of PE in vivo (3), addition of PE to isolated membrane vesicles (4), or reconstitution of purified LacY into liposomes containing PE (6). Phosphatidylserine is the only physiological lipid that can substitute for PE. However, in vitro, phosphatidylcholine supports the proper topological organization of proteoliposomal LacY, but not transport or the native conformation of the P7 domain (6, 7). The minimum lipid requirement for the native conformation of the P7 domain and the function of LacY is a diacylphospholipid containing an ionizable amine head group organized in a bilayer (7).
The major objective of this study was to examine whether or not the topology of another polytopic membrane protein is reversibly affected by changes in phospholipid composition. To investigate the generality of the influence of specific lipids on the assembly of membrane proteins, we have focused on the high affinity phenylalanine permease PheP of E. coli. PheP and LacY are both electrochemical potential-driven transporters, but the former is a member of the amino acid/polyamine/organocation superfamily (8), whereas the latter belongs to the larger major facilitator superfamily (9). Members of these families are diverse in substrate specificity (sugars, amino acids, drugs, and other low molecular mass compounds), yet show significant homology in function, sequence, and structure as well as similarities in putative topogenic amino acid residues. Many of these transporters are characterized by a 12-TM topology with both N and C termini located in the cytoplasm (see Fig. 1). The C-terminal halves of these proteins strictly follow the positive inside rule (10) for their cytoplasmic domains. However, the N-terminal halves contain acidic amino acids in the cytoplasmic extramembrane domain. A net positive charge in these domains dictates cytoplasmic disposition, but conserved single negatively charged residues are topogenic factors favoring periplasmic location if they are present in the proximity of the aqueous-membrane interface (11–13).
|
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
DE3 lysogenization kit and plasmid pET-11a were from Novagen.
Bacterial Strains, Plasmids, and Growth Conditions—Strain AD93 (pss93::kanR) was used as the host strain. This strain cannot make PE and is not viable without either plasmid pDD72 (pssA+ camR and pSC101 temperature-sensitive replicon) or growth medium containing millimolar concentrations of MgCl2 (16). Strain AD93 (grown at 37 °C) lacks PE, and strain AD93/pDD72 (grown at 30 °C) retains the normal E. coli composition of phospholipids, including PE. Strain AA9256 (pss93::kanR ParaB-pssA+ 
araBA) (see Ref. 3 for description and characteristics of this strain) was used to regulate PE synthesis. Strain AA9256(T7) was made by integration of DE3 prophage (carrying the T7 RNA polymerase gene under lac operon control) into the chromosome of strain AA9256; the strain was used to express target genes under the control of the T7 promoter. The wild-type pheP gene and the pheP gene encoding PheP lacking cysteines (PhePCys–) were subcloned into plasmid pBR322 (EcoRI and SalI site) to produce plasmids pBR-pheP and pBRphePCys–, respectively. A series of pBR322-derived plasmids encoding derivatives of PhePCys– containing single Cys replacements in extramembrane domains were constructed by site-directed mutagenesis (17). The nomenclature for these plasmids is pBRpheP-X, e.g. pBRpheP-G13 has a G13C replacement (Fig. 1). pETpheP-X plasmids were constructed from pET-11a (carrying lacI) by cloning pheP-X genes into the BamHI and NdeI regions, thus placing the genes under tandem T7 promoter and lac operator control. Cells were grown in LB medium containing ampicillin (100 µg/ml) as required and 50 mM MgCl2. Expression of pssA under the control of the araB promoter was either induced by growth in the presence of 0.2% arabinose or repressed by growth in the presence of 2% glucose; to maintain controlled expression, cells were maintained in exponential growth phase.
Chemical Labeling of Cys Residues—A 50-ml culture of cells expressing a PheP-X derivative was grown in LB medium plus Mg2+ and ampicillin to mid-log phase (A600 = 0.5–0.6) and resuspended after harvesting to A600 = 25 in 1 ml of buffer A (100 mM K+-HEPES (pH 7.5), 250 mM sucrose, 50 mM MgCl2, and 0.1 mM KCl). Samples were incubated with 5 mM AMS for 30 min at 25 °C to block water-accessible Cys residues exposed to the periplasmic side of the cytoplasmic membrane. AMS was removed by two cycles of centrifugation and resuspension in buffer A. Cells (either with or without AMS pretreatment) were biotinylated by adding MPB to a final concentration of 100 µM, followed by incubation for 5 min at 25 °C. The reaction was quenched by addition of 
-mercaptoethanol to 20 mM, followed by two cycles of centrifugation and resuspension in buffer A containing 20 mM -mercaptoethanol. To expose Cys residues facing the cytoplasm to external solvent, cell suspensions were vortexed vigorously with 0.5% (v/v) toluene for 1 min and subjected to labeling as described above. Labeled cells were resuspended in 0.5 ml of 10 mM Tris-HCl (pH 8.0), 5 mM EDTA, and 5 mM -mercaptoethanol and then solubilized and immunoprecipitated as described below.
Immunoprecipitation and Western Blot Analysis—After MPB labeling, cells were lysed by addition of an equal volume (0.5 ml) of 0.2 M NaOH, vortexed, incubated for 15 min on ice, and then centrifuged at 20,800 x g for 15 min at 4 °C. The pellets were washed once with 4 M KCl and once with 10 mM Tris-HCl (pH 8.0) and then solubilized by resuspension in 50 µl of 50 mM Tris-HCl (pH 8.0), 1 mM EDTA, and 1% SDS, followed by vigorous vortexing for 30 min at room temperature and incubation for 10 min at 37 °C. Samples were diluted by adding 1.4 ml of cold 50 mM Tris-HCl (pH 8.0) containing 150 mM NaCl, 0.1 mM EDTA, and 0.1% Lubrol (immunoprecipitation buffer) and immunoprecipitated with anti-PheP polyclonal antibody overnight at 4 °C. The antibody complex was isolated by addition of 50 µl of a suspension of formalin-treated Staphylococcus aureus cells (Pansorbin) reconstituted in immunoprecipitation buffer according to the supplier's instructions. The samples were gently rocked at 4 °C for 60 min. After centrifugation, the pellet was washed once with immunoprecipitation buffer, twice with 1 ml of 50 mM Tris-HCl (pH 8.0) containing 1 M NaCl and 0.1% Lubrol, and once with 10 mM Tris-HCl (pH 8.0). The final precipitates were solubilized by resuspension in 30 µl of SDS sample buffer and incubated at 37 °C for 15 min. Samples were centrifuged, and the solubilized proteins were subjected to SDS-PAGE on 10–12% gels. The samples were transferred from the gels to nitrocellulose membranes as described previously (5). Avidin-horseradish peroxidase (1:50,000 dilution of a 2 mg/ml stock solution) and conventional ECL reagents were used to visualize biotinylated proteins according to the manufacturers' instructions. A Bio-Rad Fluor-S Max Multimager was used to record the results.
Other Methods—Transport of [3H]phenylalanine was measured in intact E. coli cells as previously described (18). Phospholipid composition of cell membranes was analyzed after 32PO4 labeling of cells as previously described (16) using chloroform/methanol/acetic acid (65: 25:9, v/v) as the chromatography solvent.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
-galactosidase (LacZ) (Fig. 3A), which contains many natural cysteine residues, demonstrated that the host cells used in this study were neither leaky nor permeable to MPB (Fig. 3B, second and fourth lanes) under the labeling conditions we chose for PheP unless the cells were first permeabilized with toluene (first and third lanes).
|
|
To further substantiate the misorganization of the N-terminal helical hairpin (NT-TM I-P1-TM II-C2) of PheP in PE-lacking E. coli cells, single Cys derivatives expressed in both cell types were pretreated with side-specific control reagents (toluene or AMS) before MPB labeling. In PE-containing cells, when samples were permeabilized with toluene prior to MPB treatment, no additional biotinylation occurred for the periplasmic domains, but extensive biotinylation of the cytoplasmic domains occurred (Fig. 5A, left two lanes). These results are in agreement with the putative location of single cysteines in the current topology map of PheP (Fig. 1) (24). In PE-lacking cells, the P1 domain was rendered accessible to MPB after toluene treatment, whereas the NT and C2 domains showed no increase in accessibility, consistent with an inversion in the topological orientation of the first two TMs. The remainder of the domains behaved the same as in PE-containing cells (Fig. 5A, right two lanes). AMS is a hydrophilic, membrane-impermeable, non-biotinylated sulfhydryl reagent (20). Periplasmic cysteines accessible to MPB in PE-containing and PE-lacking whole cells should be blocked from reaction with MPB if the cells are pretreated with AMS; however, cysteines that lie within the bilayer and that react with bilayer-partitioned MPB would not be protected by AMS. Fig. 5B shows that, in PE-lacking (but not PE-containing) cells, the NT and C2 domains were protected by AMS, as were the P3 and P5 domains in both cell types. This further establishes that, under the conditions used, MPB did not react with cysteines sequestered in either the membrane bilayer or the lumen.
|
-D-thiogalactopyranoside (IPTG) to the growth medium. Cells were grown first on glucose and IPTG to allow synthesis and membrane assembly of PheP in the near absence of PE. Then, glucose and IPTG were removed, and cells were grown on arabinose to allow new synthesis of PE in the absence of newly synthesized PheP. Cells were isolated and assayed for transport function and Cys accessibility. Parallel cultures were grown with 32PO4 to determine the respective phospholipid composition.
The level of PE was <10% of total phospholipids in cells grown on glucose with IPTG (Fig. 6A, left lane). PheP transport function was also reduced as expected (Fig. 6B). Treatment of intact cells expressing single Cys derivatives of PheP with MPB showed that topological misorganization of the first hairpin (NT-TM I-P1-TM II-C2) occurred (Fig. 6C), i.e. the NT, C2, and P3 domains were biotinylated with and without toluene treatment, indicating their periplasmic orientation, whereas the P1 domain was not accessible to MPB unless cells were first treated with toluene, indicating its cytoplasmic orientation. Switching to growth on arabinose in the absence of IPTG returned the phospholipid composition to normal (
75% PE) after 90 min (Fig. 6A, right lane). Simultaneously, PheP transport function was restored (Fig. 6B), and proper topological organization was detected (Fig. 6C), i.e. the NT and C2 domains were no longer biotinylated unless cells were pretreated with toluene, whereas the P1 and P3 domains were accessible to MPB to the same extent with and without toluene. These results strongly support a reversible topological organization of the N-terminal first helical hairpin of PheP that is determined by the membrane phospholipid composition post-insertion.
|
Length of TM III—The hydropathy plot (25) of the deduced amino acid sequence of PheP (26) suggests a longer TM III than was originally predicted by PhoA fusions (23). Because of hydrophobic amino acid residues lying within the boxed region of the C2 domain (Fig. 1), its physical characteristics (overall hydrophobicity (25) of G = –18 kcal/mol for membrane insertion of the extended TM III) strongly favor partitioning of this region into the membrane bilayer.
To examine this possibility, we determined the accessibility of single cysteine replacements (A91C, G96C, A99C, and S103C) to MPB in this region. Although sequence analysis verified the replacements and Western blot analysis showed that PheP was synthesized and assembled into the membrane, none of these cysteines was accessible to MPB even after permeabilization with toluene in either PE-containing or PE-lacking cells (data not shown). This result is consistent with the integral membrane location of this domain of PheP and supports an extended TM III of up to 34 amino acids. It is unlikely that all four cysteines extending over a 13-amino segment are sequestered in an extramembrane solvent-inaccessible secondary structure. In support of this hypothesis is the recent demonstration that changes at Tyr107 could suppress the mutant phenotype caused by substitutions at Gly35 (buried in the bilayer), implying that position 107 lies well within the membrane (28).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
PE also plays an active role in determining the topological organization of PheP both during initial membrane assembly and dynamically after membrane insertion has occurred. The PE-dependent topological organization of PheP was shown through differences in accessibility by membrane-impermeable sulfhydryl reagents to single Cys replacements in the extramembrane domains connecting TMs. The advantage of this technique is that modification is restricted to a single amino acid residue that is less likely to disturb the overall protein structure (29). This idea was also supported by the maintenance of PheP function in cells with natural phospholipid composition. Conclusions were based on the use of complementary approaches comparing accessibility in PE-containing versus PE-lacking cells, periplasmic domains versus cytoplasmic domains, sealed membranes versus toluene-permeabilized membranes, and reactive cysteines versus AMS-protected cysteines. The cytosolic protein LacZ was used as an internal permeability control to demonstrate that the sulfhydryl reagents employed were not membrane-permeable under the labeling conditions used. The labeling rationale proposed and introduced here employed whole cells and eliminated the need for the preparation of oriented and sealed membrane vesicles for the substituted cysteine accessibility assay.
Our results strongly support the complete inversion of the N-terminal first helical hairpin (NT-TM I-P1-TM II-C2), with retention of the normal topological organization of the remainder of PheP when assembled into PE-lacking membranes (Fig. 7). The C2 and P3 domains are located on opposite sides of the membrane in PE-containing cells (upper panel), whereas in PE-lacking cells, they are both located on the periplasmic side of the membrane (lower panel). Thus, a large structural rearrangement must occur in TM III, which connects the C2 and P3 domains. The most probable rearrangement would be a "U"-shaped mini-loop partially inserted into the periplasmic side of the membrane. Such U-shaped structures usually occur when hydrophobic segments are either too short to span the membrane (30) or too long to be fully embedded perpendicular to the lipid bilayer (31, 32). This structural arrangement is consistent with the inaccessibility of cysteines in the proposed extended N-terminal side of TM III and the lack of charged residues in the C-terminal half of the previously proposed C2 domain. This U-shaped loop would allow the C-terminal half of TM III to remain on the same side of the membrane as TMs IX and X, with which TM III appears to form a water-filled pore for substrate transport (24). TM III should have the potential or ability to support such a rearrangement because this region of PheP must be quite flexible to assist in moving substrate through the pore. Partial (but not extensive) rearrangement of TM III might affect both substrate binding (Km) and transport (Vmax), as was observed.
|
Several other transport proteins have highly flexible domains associated with transport function. The dynamic behavior of the functionally important TM II of the Na+/proline transporter PutP is due to conformational flexibility of the hydrophilic face of the cytoplasmic half of this domain (34). TM II of PutP also contains a high number of Gly residues, allowing for a bent rather than straight transmembrane helix. The highly conserved Gly63 may be important for the conformational flexibility of TM II. Similarly, the presence of Gly residues allows 9 of the 12 TMs of LacY, as verified by the crystal structure (35), to be bent and presumably flexible. The flexible hinge region of LacY that allows a topological inversion in the absence of PE occurs in the center of the protein (3), which is associated with substrate transport and is also a very flexible region of the protein (35, 36). A single amino acid substitution in TM V of LacY reduces flexibility and eliminates active transport, but not substrate binding (37). This mutant of LacY appears to be locked in an intermediate conformation that lies along the substrate transport pathway (35). Finally, a potential role has been proposed for the movement of TM3 of the plasma membrane Chinese hamster P-glycoprotein into and out of the bilayer during the transport cycle (38, 39).
Membrane protein topology is not determined solely by the hydrophobic transmembrane helices. Charged residues in short loops connecting the hydrophobic segments are potent topogenic determinants; the best known effect is codified in the so-called positive inside rule, which states that positively charged loops tend to remain on the cytoplasmic side of the membrane (40, 41). However, the NT and C2 domains of PheP contain acidic amino acids (Glu9, Asp10, Glu16, Glu76, Glu79, and Glu80) that are positioned near the membrane-cytoplasmic interface. Negatively charged amino acids, particularly near the interface (11–13), favor translocation of hydrophilic loops to the periplasmic side of the membrane. Thus, the NT and C2 domains may be in a metastable or "frustrated" state due to the opposing signals of positively and negatively charged amino acids, making them more prone to respond to changes in the surface charge of the membrane. The interaction between charged protein residues and a surface charged lipid head group is critical for the stability and maintenance of this state. Negative charges in the NT domain and the first half of the C2 domain may act as specific topogenic sequences that interact with and are anchored by the positive head group of PE. This interaction would be absent in PE-lacking cells. Also, PE deficiency may increase the negative charge density at the membrane surface, making membrane translocation of the C2 domain favorable (42). Increasing the anionic lipid content in a liposome system favors the translocation of weakly positively charged domains in a Sec-independent manner in opposition to the positive inside rule (43). PE has a smaller head group compared with phosphatidylglycerol and cardiolipin, the other two major phospholipids of the E. coli inner membrane, which, combined with a high negative charge density, would provide a more loosely packed membrane surface for membrane protein domain rearrangements (44).
Another factor that could, in principle, affect the interpretation of our in vivo experiments is the protonmotive force. The protonmotive force has been shown to be a determinant of membrane protein topology (45). Can reversible topological effects be caused by changes in the protonmotive force or be affected by protonmotive force? We can rule out the first possibility because the energetic parameters (both pH and 
) of PE-lacking cells were found to be comparable with those of PE-containing cells (46). However, it is likely that the driving force of the protonmotive force (positive outside) acting on the negatively charged residues within the NT and C2 domains (without PE as an anchorage determinant) can overcome the large positive G for translocating these domains. The high net hydrophobicity of the C2 domain plus the extended TM III may require interaction between PE and the acidic residues of the C2 domain for stable cytoplasmic orientation of the C2 domain.
Several aspects of this study may provide new insight into mechanisms that govern the proper topology of polytopic membrane proteins and/or the function of membrane proteins. The fact that both LacY and PheP undergo reversible dynamic reorientation for part of the protein modulated by a change in phospholipid composition suggests an intriguing possibility that TM segments or hairpins can move freely across the bilayer. Thus, the orientation of individual TMs may not be "fixed" during assembly, after native structure has been attained, or during transport catalytic cycles. Because PE-lacking cells are viable, only a subset of nonessential proteins appear to respond to changes in lipid environment. Thus far, the common lipid-responsive feature of LacY and PheP domains is a flexible region or hinge point between domains that allows for switching of topological organization within a folded compact structure. For PheP, this hinge point would be the extended TM III; and for LacY, the hinge point would be the insufficiently hydrophobic TM VII, which could exist stably outside of the membrane (3). These putative hinge points are also in a mobile and flexible region of the respective proteins associated with substrate translocation across the membrane. These properties may be shared by other secondary transporters (lysine-specific permease LysP, aromatic amino acid permease AroP, and high affinity Na+/proline transporter PutP), which are also dysfunctional in PE-lacking E. coli cells.2
The aberrant topology in each case is localized to the first half of PheP and LacY, with the remaining downstream TMs exhibiting a normal topological disposition. Therefore, TMs on either side of a flexible hinge region can organize independently of each other in response to lipid environment, whereas those proteins without such a hinge region either cannot assume different topologies or cannot fold and are degraded. Such topological flexibility may be important during the assembly of some membrane proteins, as evidenced by the transient exposure of the insufficiently hydrophobic TM10 of human erythrocyte band 3 (anion exchanger 1) to the lumen of the endoplasmic reticulum during assembly, followed by late membrane insertion after the remaining 11 TMs are properly inserted and the protein has partially folded (47). It is quite interesting that TM10 is thought to include an important site for anion transport.
For thermodynamic reasons, translocon-independent insertion is thought to occur via "helical hairpins" composed of interacting antiparallel helices and their short intervening polar loops (48). Formation of a helical hairpin may sufficiently increase the net hydrophobicity of the sequence to drive polar loop residues across the lipid bilayer. Pairing of two transmembrane segments in the E. coli tetracycline antiporter TetA(C) and sensor kinase protein KdpD was suggested to be essential for integration into the membrane (49, 50). The N-terminal hairpin of PheP appears to behave as an independent and very mobile insertion and translocation unit.
We previously cited several examples suggesting that lipid-assisted assembly involving PE might be a general phenomenon. The data presented here certainly extend this mechanism to amino acid secondary transporters of bacteria. In yeast, both the arginine (51) and tryptophan (52) permeases are defective when PE levels are lowered. The defect appears to be the inability to assemble the permeases in the Golgi apparatus into lipid rafts and to transport these complexes to the plasma membrane. Whether this is a defect in permease assembly or other membrane assembly/trafficking events dependent on PE is not clear. In Drosophila, sterol regulatory element-binding protein processing is also controlled by PE (27). Therefore, it is clear that PE plays a very specific role in the assembly and conformation of membrane proteins in both prokaryotes and eukaryotes.
|
FOOTNOTES |
|---|
¶ To whom correspondence should be addressed: Dept. of Biochemistry and Molecular Biology, University of Texas Medical School, 6431 Fannin St., Suite 6.200, P. O. Box 20708, Houston, TX 77030. Tel.: 713-500-6051; Fax: 713-500-0652; E-mail: William.Dowhan{at}uth.tmc.edu.
1 The abbreviations used are: PE, phosphatidylethanolamine; TM, transmembrane domain; AMS, 4-acetamido-4'-maleimidylstilbene-2,2'-disulfonic acid; MPB, (+)-biotinyl,3-maleimidopropionomidyl-3,6-dioxaoctanediamine; PhePCys–, cysteineless PheP; IPTG, isopropyl--D-thiogalactopyranoside.
2 M. Bogdanov and W. Dowhan, unpublished data.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Bogdanov, J. Xie, P. Heacock, and W. Dowhan To flip or not to flip: lipid-protein charge interactions are a determinant of final membrane protein topology J. Cell Biol., September 9, 2008; 182(5): 925 - 935. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Soto and G. M. Carman Regulation of the Saccharomyces cerevisiae CKI1-encoded Choline Kinase by Zinc Depletion J. Biol. Chem., April 11, 2008; 283(15): 10079 - 10088. |
) or strain AD93 (–PE; 
) carrying plasmid pBRpheP in either the absence (
) of proton uncoupler (50 µM carbonyl cyanide p-(trifluoromethoxyl)phenylhydrazone) as described under "Experimental Procedures." B, double-reciprocal plot of the initial velocity (V) of Phe uptake versus the concentration of Phe (S) for the above two strains. C, expression levels of PheP in PE-containing and PE-lacking cells. The indicated cells were isolated by centrifugation, and a cell-free extract was prepared by passage through a French press at 8000 p.s.i. The membrane fraction was isolated by centrifugation at 150,000 x g. Increasing amounts of membrane protein (12.5, 25, and 50 µg) from the above cells (third through eighth lanes) were subjected to SDS-PAGE, followed by Western blot analysis using PheP-specific antibody. Membrane protein (50 µg of protein) from AD93/pDD72 without (first lane) or with (second lane) plasmid pBR-pheP was treated in a similar manner. The level of the PheP-specific band at 42 kDa was normalized to that of the nonspecific band at 36 kDa. Chromosomally encoded PheP levels were too low to detect under the conditions of the experiment, as indicated in the second lane.
(LacZ–, 
