Binding of monoclonal antibody 4B1 to homologs of the lactose permease of Escherichia coli

Protein Science (1997), 61503-1510. Cambridge University Press. Printed in the USA. Copyright 0 1997 The Protein Society JIANZHONG SUN, STATHIS FRILLINGOS,' AND H. RONALD KABACK Howard Hughes Medical Institute. Departments of Physiology and Microbiology & Molecular Genetics, Molecular Biology Institute, University of California Los Angeles, Los Angeles, California 90095-1662 (RECEIVED January 21, 1997; ACCEPTED March 31, 1997) Abstract The conformationally sensitive epitope for monoclonal antibody (mAb) 4B1, which uncouples lactose from H + trans- location in the lactose permease of Escherichia coli, is localized in the periplasmic loop between helices VII and VI11 (loop VII/VIII) on one face of a short helical segment (Sun J, et al., 1996, Biochemistry 35:990-998). Comparison of sequences in the region corresponding to loop VII/VIlI in members of Cluster 5 of the Major Facilitator Superfamily (MFS), which includes five homologous oligosaccharide/H+ symporters, reveals interesting variations. 4B1 binds to the Citrobucter freundii lactose permease or E. coli raffinose permease with resultant inhibition of transport activity. Because E. coli raffinose permease contains a Pro residue at position 254 rather than Gly, it is unlikely that the mAb recognizes the peptide backbone at thisposition. Consistently, E. coli lactose permease with Pro inplace of Gly254 also binds 4B1. In contrast, 4B1 binding is not observed with either Klebsiella pneumoniue lactose permease or E. coli sucrose permease. When the epitope is transferred from E. coli lactose permease (residues 245-259) to the sucrose permease, the modified protein binds 4B1, but the mAb has no significant effect on sucrose transport. The studies provide further evidence that the 4B1 epitope is restricted to loop Vn/VIII, and that 4B1 binding induces a highly specific conformational change that uncouples substrate and Hf translocation. Keywords: bioenergetics; conformational epitope; oligosaccharide/H+ ; symporters; transport The lactose (lac) permease of Escherichia coli is a paradigm for secondary transport proteins from archaea to the mammalian cen- tral nervous system (reviewed in Kaback, 1983, 1989, 1992, 1996; Poolman & Konings, 1993). All available evidence indicates that the permease is composed of 12 a-helical rods that traverse the membrane in zigzag fashion with the N- and C-termini on the cytoplasmic face (Fig. 1). Based on site-directed excimer fluorescence (Jung et al., 1993), site-directed mutagenesis, second-site suppressor, and chemical res- i, cue studies [King et al., 1991; Lee et al., 1992; Sahin-Tbth et a. 1992; Dunten et al., 1993; Sahin-T6th & Kaback, 1993; Frillingos & Kaback, 1996a; see Lee al., 1996, in addition), a helix packing et model of the C-terminal half of the permease was formulated. The Reprint requests to: HHMI/UCLA 5-748 MacDonald Research Labs, Box 951662,LosAngeles,California90095-1662;e-mail: ronaldka hhmi.ucla.edu. 'Present address: Laboratory of Biological Chemistry, University of Ian- nina Medical School, Iannina GR-45 110, Greece. Abbreviations: BSA, bovine serum albumin; C-less permease, functional E. coli lactose permease devoidof Cys residues; E. coli sucrose permease/ loop VII/VIKL, E. cofi sucrose permease with loop VIlfVIU fromthe E. coli l a c permease; EDTA, ethylenediaminetetraceticacid; IPTG. isopro- KP,, pyl 1-thio-P,D-galactopyranoside; potassium phosphate; lac, lactose; mAb, monoclonal antibody; MFS, Major Facilitator Superfamily; PMS, phenazine methosulfate; RSO, right-side-out. model has been confirmed and extended by engineering divalent metal-binding sites (Jung et al., 1995; He et al., 1995a, 1995b), site-directed chemical cleavage(Wu et al., 1995), site-directed spin labeling and thiol crosslinking (Wu et al., 1996; Wu & Ka- back, 1996), and by the demonstration that monoclonal antibody 4B11 binds to the last two cytoplasmic loops (Sun et al., 1997). Another mAb designated 4Bl blocks deprotonation of the per- mease and uncouples lactose and H + translocation by binding to a conformational epitope on periplasmic surface of lac permease (Carrasco et al., 1984a; Herzlinger et al., 1984). Further studies (Sun et al., 1996) demonstrate that the 4B1 epitope is located in the periplasmic loop VII/VIIl on one face of a short a-helical seg- ment. Thus, Cys-scanning mutagenesis demonstrates that Phe247 is the primary epitope determinant, and sulfhydryl modification of single-Cysmutants shows that Phe250andGly254 are also important. Although 4B 1 binding uncouples lactose from H+ translocation, none of the residues in loop VII/VIII is important for activity (Frillingos et al., 1994; Sun et al., 1996). On the other hand, mAb 4B1decreasesthe apparent pK,of an Aspresiduein place of Glu325 (Frillingos & Kaback, 1996b), and alters the reactivity of single Cys-replacement mutants in the C-terminal half of the per- mease [e.g., V238C (helix VU), V331C (helix X), and single Cys 355 (helix XI)] with N-ethylmaleimide (Frillingos et al., 1997). These results and the observation that avidin binding to a biotin- 1503 1504 J. Sun et al. f C O 0 " Loop VIIiVIII Fig. 1. Secondary structure model of E. coli lac permease. The 12 hydrophobic transmembrane helices are depicted as rectangles. The sequence of loop VII/VIII is indicated and the single-letter amino acid code is used. The irreplaceable residues in the permease are emboldened. ' h o charge pairs (D237 with K358 and D240 with K319) in transmembrane helices VII, X, and XI are indicated. ylated Cys residue in loop VII/VIIIhas no effect on transport (Sun et al., 1996) suggest that uncoupling by 4B1 is due to a torsional effect induced by binding, which alters the pKJs) of residues that play a direct role in the mechanism. In this study, 4B1 binding and its effect on activity are examined with the members of Cluster 5 of the Major Facilitator Superfamily (Marger & Saier, 1993; Lee et al., 1994). The sequence of loop VlI/VIII (residues 245-259) of the E. coli lac permease shares varying degrees of homology with each of the four related sym- porters, ranging from 100% identity with the C. freundii l a c per- mease to 33% identity with the E. coli sucrose permease (Table 1). The mAb binds to the C. freundii lac permease or the E. coli Table 1. Comparison of the sequence of loop VIVvIIIl {from Am245to Arg259) of E. coli lac permease with the corresponding regions of lac permease from C. freundii and K.pneumoniae M5a1, E. coli rafSinose permease, a d E. coli sucrose permeasea Transporter Sequence (from Am245 to Arg2.59 in E. coli lac permease) * E. coli lacy C. freundii lacy K. pneumoniae lacy E. coli rafB E. coli cscB * * N FFTSF FATGE QGTR N FFTSF FATGE QGTR N F F k g FF s s p qr G T e i FFeSF FrTpq a G i k v F y a g l F e s h d vGTR 'One-letter amino acid code is used. The main epitope residues in E. coli lacy are marked with asterisk. Residues identical to those in E. coli l a c permease are capitalized. raffinose permease and inhibits transport activity. In contrast, 4B 1 neither binds nor inhibits transport with the K. pneumoniae lac permease or the E. coli sucrose permease. When the epitope is transferred from E. coli lac permease to the sucrose permease, 4B 1 binding is observed, but the mAb has no effect on activity. Results mAb 4B1 binding Because 4B 1 binds to an epitope in loop VII/VIIIof lac permease that shares varying degrees of homology with the corresponding region in five related oligosaccharide/H+ transporters in MFS Cluster 5 (see Table l), the homologs were examined for mAb 4B 1 binding. Spheroplasts prepared from IPTG-induced C.freundii bind 4B1 in a manner comparable to that of spheroplasts containing E. coli lac permease (Fig. 2A). Spheroplasts with E. coli raffinose permease bind to a lesser extent, and no 4B1 binding whatsoever is observed with spheroplasts containing K. pneumoniae lac per- mease or E. coli sucrose permease (Fig. 4A). Although the sequence of loop VII/VIII in C. freundii lac per- mease is identical to that of E. coli lac permease and 4B1 binding is anticipated, only about 50% identity is observed in the loop between E. coli lac permease and eitherE. coli raffinose permease, which binds 4B 1, or K. pneumoniae lac permease, which does not (Table 1; Fig. 2A). In particular, although two of the epitope de- terminants in E. coli lac permease (Phe247 and Phe250) are con- served, Gly254 is replaced with Pro in both cases. Mutant G254P in E. coli lac permease was constructed, found to be. expressed at control levels in themembrane and binds mAb 4Bl about 45% as well as C-less permease (Fig. 2B). Two residues (Ser249 and Thr253) in the sequence containing the epitope are conserved in raffinose permease, but not in K. pneumoniae lac permease where Ser249 is replaced with Gly and Thr253 is replaced with Ser. To 1505 mAb 4BI binding to lac permease homologs .............. E. cofilpRU600/rafB~i:~i:i~i:i:i:i:i:~:i:i:~~i:i:i::~ A - -3 E. C O ~ ~ / ~ R E S I O / ~ ~ C Y ..... K. pneumoniae (+IPTG) -3 K. pneumoniae (-IPTG) -:I - E. coli/pNOT41/ l a c y -~iiiii-iii.iiiiiii:i:iii~i:~:~ C. freundii (+IpTG) -: C . freundii (-IPTG) -2 E. colilpT7-5 -3 - ..................................... ....................................... .......................................... E. coli/pT7-5/lacY - ............................................ ................ ..... ~. ......................................... 400 I I I 20 60 4B1 Binding (CPM, X1000) B 20 0 40 60 80 100 120 4B1 Specific Binding (% C-less) Fig. 2. Binding of mAb 4B1 to homologs of E. coli lac permease and its mutants. A: Binding to E. coli TI84 spheroplasts harboring given plasmids or to spheroplasts of C. freundii or K. pneumoniae M5al. E. coli TI84 cells were grown and induced with IPTG. C.freundii and K.pneumoniae M5al were grownin the absence or presence of0.5 mM IPTG, indicated. Spheroplastswere prepared as and incubated with 4B1 followed by ['251]proteinA as described in Materials and methods. 4B1 binding is expressed as radioactivity (['251]proteinA) bound to spheroplast preparations containing equal amounts of protein. The amount of permease expressed in every preparation could not be quantitated due to the lack of appropriate antibodies against C. freundii and K. pneumoniae lac permeases or the E. coli raffinose permease. However, the negative controls (-IPTG) bind only about 11% of positive control (+IPTG). The four groups represent spheroplasts expressing E. coli rafB, K . pneumoniae, C. freundii, and E. coli lacy, respectively, and the negative controls. The K. pneumoniae and C.freundii permeases were also expressed in E. coli (E. coli/pRESlO/ZacY and E. coli/pNOT4lAacY, respectively). B: Binding to spheroplasts expressing mutants of E. coli lac permease. The amount of each mutant was quantitated as described in Materials and methods, and 4B1 specific binding is expressed as a percentage of that observed with C-less lac permease. All data were corrected for 4B1 binding tospheroplastsharboring plasmidpV-5 without a lacy insert. The results represent the average of two independent experiments. examine the role the two residues in more detail, mutants S249G E. coli lac permease and E. coli sucrose permease, and no con- of and T253S were constructed in E. coli lac permease (Fig. 2B). servation of Phe247, Phe250, or Gly254, which comprise the epi- Both mutants are expressed in the membrane to control levels, and tope (Sun et al., 1996) is observed in sucrose permease (Table 1). S249G binds 4B1 aboutas well as the control, while T253S binds Therefore, the finding that sucrose permease does not bind 4B1 is about 70% as well. Finally, only 3 1 % identity is observed between expected. 1506 J. Sun et al. Active transport C. freundii lac permease Active transport catalyzed by E. coli lac permease is markedly inhibited as a result of 4B1 binding (Carrasco et al., 1984a; Sun et al., 1996), and the four irreplaceable residues in E. coli lac permease (GIu269, Arg302, His322, and Glu325) are conserved in the homologous symporters. Therefore, the effect of 4B1 on active transport was tested in the homologous members of Cluster 5.4B1 inhibits active transport of either melibiose (Fig. 3A) or lactose (not shown) by C. freundii lac permease in RSO membrane ves- icles to an extent comparable to that observed with E. coli lac permease (ca. 80%).In addition, mAb 4B1 inhibits active transport of either raffinose (Fig. 3B) or melibiose (not shown) by E. coli raffinose permease, but to a lesser extent (ca. 50%), a finding consistent with the observation that 4B1 binding to raffinose per- mease is decreased relative to E. coli lac permease or C. freundii lac permease (Fig. 2A). As expected, noinhibition of transport by 4B1 is observed with K. pneumoniue lac permease or E. coli sucrose permease, neither of which binds the mAb (data not shown). A To examine whether the mAb 4B 1 epitope can be transferred from E. coli lac permease to a homolog that does not bind the mAb, loop VII/VIII in E. coli sucrose permease was replaced with the cor- responding region (residues 245-259) of E. coli lac permease to construct E. coli sucrose permease/loopVII/VIII. In addition, sucrosepermease/loopVII/VIIIwas engineered to contain the C-terminal dodecapeptide of E. coli lac permease to allow quan- titation by Western blotting (Frillingoset al., 1995; Sahin-Tbth et al., 1995). E. coli sucrose permease/loop VII/VIII is expressed in the membrane about 50% as well as wild-type sucrose permease and catalyzes sucrose transport. Relative to E. coli lac permease, the engineered sucrose permease mutant, binds 4B1 about 70% as well (Fig. 4A). However, 4B1 binding does not inhibit sucrose transport in RSO membrane vesicles containing chimeric sucrose permease to any significant extent (Fig. 4B). Discussion The studies reported here provide further characterization of the epitope for mAb 4B 1, which binds to loop VII/VIII of the E. coli lac permease and uncouples lactose from H+ translocation. The C. freundii lac permease exhibits 70% identity with E. coli lac per- mease and binds 4B1. Moreover, activity is inhibited by 4B1 to approximately the same extent as observed with E. coli lac per- mease. Similarly, E. coli raffinose permease has 56%identity with E. coli lac permease and also binds 4B1, but both binding and the inhibitory effect of the mAb are appropriately decreased relative to E. coli lac permease. In contrast, neither the K. pneumoniue lac permease nor the E. coli sucrose permease binds 4B1 nor is the mAb inhibitory, although the two transporters exhibit about 60% and 3 1 identity, respectively, with E. coli lac permease. Finally, % when the epitope is transferred from E. coli lac permease to the sucrose permease, although 4B1 binding is observed, the mAb has no significant effect on activity. As shown by Sun et al. (1996), Phe247, Phe250, and Gly254 comprise the 4B 1 epitope, and these residues probably fall on one face of a short helical segment. Furthermore, thiol modification studies on single-Cys replacement mutants inthe short helical segment show that modification of certain Cys residues on the 30 0 Grafing the 4BI epitope into the sucrose permease 60 90 120 Time (sec) E. coli raffinose permease 25 B 20 15 10 5 0 0 30 60 90 1 20 Time (sec) Fig. 3. Effect of mAb 4B 1 on active transport of C. freundii lac permease and E. coli raffinose permease. RSO vesicles from C. freundii or E. coli TI84 expressing E. coli raffinose permease were preparedand assayed either with no further treatment or incubated at 25 'T for 30 min with 0.7 mgjmL of 4B 1. [3H]Melibiose or [3H]raffinose transport in the presence of ascorbate and PMS was assayed under oxygen as described in Materials and methods. A: Effect of mAb 4B 1 on melibiose transport in RSO mem- brane vesicles from induced C. freundii. Open square, no additions; open circle, 4B1 added. B: Effect of mAb 4B1 on raffinose transportin RSO membrane vesicles from E. coli T184 harboring plasmid with the E. coli ram gene. Open square, no additions; open circle, 4B1 added. 1507 mAb 4BI binding to lac permease homologs I ................................ ................................. ................................ ................................. .. pSP7~/cscB/(loopVII/VIII)................................ k -I ............................................................ .............................. .............................. .............................. ................................ ................................. ................................ ................................. ................................ pSP72IcscB 0 25 50 75 100 4B1 Specific Binding (% C-less) 0 30 60 90 Time (sec) Fig. 4. A: Binding of mAb 4B1 to E. coli TI84 spheroplasts expressing E. coli sucrose permease/loop VII/VIII. Cells were grown and induced with 0.5 mM IPTG. Spheroplasts were prepared and incubated with 4B 1 and ['*'I]protein A as described in Materials and methods. The amount of permease expressed in each instance was quantitated by immunoblotting. B: Effect of mAb 4B1 on sucrose transport by E. coli sucrose permease/loop VII/VIII. RSO vesicles were prepared, and ['4C(U)]sucrose transport was assayed in the presence of ascorbate and PMS under oxygen as described in Materials and methods. Open square, no additions; open circle, 2 mg/mL 4B1 added; open triangle, 4 mg/mL 4B1 added. opposite face leads to increased 4B1 binding. Because a Gly res- idue is found at position 254, it was postulated that in addition to two helical turns, the epitope might contain a &turn. E. coli raf- finose permease exhibits about 50% identity within loop VII/VIII and also binds 4B1 about 50% as well as E. coli lac permease. More specifically, Phe247 and Phe250 are conserved, but Gly254 is replaced with a Pro residue. In addition, 4B1 binding by G254P E. coli lac permease is comparable to that observed with E coli raffinose permease. Taken as a whole, the results are consistent with the argument that 4B1 recognizes a structural feature at po- sition 254 rather than the peptide backbone, although it cannot be stated with certainty that there is a &turn at this position (Wilmot & Thornton, 1988). It is also noteworthy that Thr248 and Ala252, which are on the opposite face of the putative short helix, are re- Ar placed with Glu andg , respectively, in E. coli raffinose permease. Although Phe247 and Phe250 are conserved in K.pneumoniae lac permease, there are differences in five residues in the epitope relative to E. coli lac permease: Thr248, Ser249, Ala252, Thr253, and Gly254 are replaced with Lys, Gly, Ser, Ser, and Pro, respec- tively(Table 1). E. coli raffinose permease, which binds 4B1, contains Glu248 and Arg252. Thus, position 248 or 252 cannot play an essential role in 4B1 recognition. S249G E. coli lac per- measebinds 4B1 as well as control, the T253S mutant binds about 70% as well, and mutant G254P binds about 50% as well. 1508 J. Sun et al. Therefore, it is apparent that a change in eachof these three res- generously provided by T.H. Wllson and E Ausubel, respectively. idues is insufficient for abrogation of 4B1 recognition and thatis it Cells harboring plasmids encoding the C.freundii or K. pneumo- the combined effectof the alterations that is probably responsible. niae lac permease were also generously provided T.H. Wilson. by The E. coli sucrose permease, which exhibits only % identity Plasmidencodingthe E. coli raffinoseoperonwasgenerously 3 1 with E. coli lac permease, does not bind 4B1 nor does the mAb contributed by R. Schmitt.Plasmidencodingsucrosepermease with the E. coli lac permease C-terminus as an epitope tag was inhibit activity. Clearly, this can be attributed to the lack of con- servation of theresiduesthatdetermine4B1recognition(i.e., constructed as described (Sahin-T6th et al., 1995). Phe247, Phe250, and Gly254). The only residues in the epitope [l-'4C]Lactose, [ ( U - ~ ~ S J ~ A T P ['251]proteinAwerefrom and , that are conserved in sucrose permease are Phe246 and Phe251 Amersham. ['4C(U)]Sucrose and[3H]rafXnosewerepurchased (Table 1). Remarkably, however, when residues 245-259 from E. from DuPont NEN. [3H]Melibiose was a generous gift from Gkrard Leblanc. Deoxyoligonucleotides were synthesized on an Applied coli lac permease are transferred into the corresponding region of Biosystem 391DNA synthesizer.All restriction endonucleases, T4 sucrose permease, the chimera binds 4B 1 aboutas well as lac 70% DNA ligase, Taq DNA polymerase were from New England Bio- permease. Therefore, the 4B1 recognition site is clearly restricted labs. DNA Sequenase was from United States Biochemical. Rabbit to this sequence of amino acid residues in loop VII/VIII of lac polyclonal antiserum against C-terminus permease (Carrasco of lac permease. On the other hand, binding of 4B 1 to sucrose permease/ et al., 1984b) was prepared Babco. mAb 4B1 was purified from by loop VlI/VIII has no significant effect on activity. ascites fluidby Protein A-Sepharose affinity chromatography (Sun Although three of the four irreplaceable residues in E. coli lac et al., 1996). All other materials were reagent grade and obtained permease [Arg302 (helix IX), His322 (helix X) and Glu325(helix from commercial sources. X)], aswellasafunctionallyinteractingchargepair[Asp240 (helix VII) and Lys319 (helix X)], are conserved in the members of Cluster 5, Glu269is not conserved in E. coli sucrose permease. Mutant construction Rather, a Val residue is in the corresponding position in helix VIII of sucrose permease, and a Glu residue is at position 272 (Bock- Residues Asn245 to Arg259 (encoded by&C TTT TTT &A mann et al., 1992) (i.e., on the same face of helix VIII as 269 but - TTC TTC E A ACA GAA AGT U A GGA ACG CGC) in loop VII/VIII of E. coli lac permease were used to replace to the one turn removed toward the cytoplasmicface of the membrane). homologous region (GTC TTT TAT GCA GGT TTA TTC GAA Furthermore,Asp237(helix VII) andLys358(helix XI) in lac TCA CAC GAT GTA GGA ACG CGC)in E. coli sucrose perme- permease are replaced with Asn234 and Ser356, respectively, in sucrose permease. When an Asp and a Lys residue are introduced ase. The mutant was constructed by site-specific mutagenesis of cscB in plasmid pSW2 (Sahin-Toth et al., 1995) by using a two- into sucrose permease at positions 234 and 356, respectively, the double mutant exhibits high activity. However, unlike the situation stage polymerase chain reaction (PCR;Ho et al., 1989). Mutants S249G,T253S,andG254Pin E. coli lacpermeasewerecon- in E. coli lac permease where reversal of Asp237 and Lys358 has structed in plasmid pT7-5 with cassette lacy encoding C-less per- little effecton activity (Dunten et al., 1993), reversal the charge of pair in sucrose permease double mutant abolishes activity (Frillingos mease (van Iwaarden et al., 1991) by one-stage PCR. Mutations were verified by sequencing the lengthof the PCR-generated seg- et al., 1995). Therefore, although the lac and sucrose permeases are ment using dideoxynucleotide termination and synthetic sequenc- related and appear to need an Asp-Lys salt bridge between heli- ing primers (Sanger et al., 1977) after alkali denaturation (Hattori ces VII and for optimal insertion into the membrane XI (Frillingos & Sakaki, 1986). et al., 1995), the two proteins exhibit distinct differences. Materials and methods Growth of bacteria E. coli T184 [lacIiOiZ - Y - ( A ) , rspL, met -, thr-, recA, hsdM, E. coli T184 (lacZ-Y -) transformedwithplasmidencodinga rou- given transporter were grown aerobically at "C in Luria-Bertani hsdWF', lac POiZD"*(Y +A')I (Teather et al., 1980) was 37 tinely used for transformation. Other cell strains and plasmids broth with ampicillin (100 pg)mL) and streptomycin (10 pg)mL) are except for E. coli T184 expressing the raffinose permease, which described in Table 2. C.freundii and K. pneumoniae M5al were Table 2. Bacterial strains and plasmidsa Strain Genotype C. freundii lacy+ K. pneumoniae M5al lac[ + l a d 'lacy Plasmid E. coli lacy+, amp' pT75llacY E. coli CSCB'. amp' pSW2/cscB E. coli rajR'A+B' D+, pRU6OO/rafB Cm' pNOT.ll/lacY C. freundii lacy', amp' pRE5lO/lacY K. pneumoniae lacy', amp' Source Okazaki et al., 1994 Yorgey Ausabel and E P. van Iwaarden et al., 1991 Sahin-Toth et al., 1995 Aslanidis and Schmitt, 1990 Lee et al., 1994 McMorrow et al., 1988 + 'lad, lac repressor; lucZ, P-galactosidase; lacy, lac permease; CSCB, chromosomally encoded sucrose permease; ra@, raffinose repressor; rafA. a-galactosidase; r a p , raffinose permease; r a p , sucrose hydrolase; amp, ampicillin; Cm, chloramphenicol. 1509 m4b 481 binding to lac permease homologs was grown in the presence of chloramphenicol (34 pg/mL) and streptomycin (10 pg/mL). C. freundii and K.pneumoniae M5al were grown in Luria-Bertani broth without antibiotics. Overnight cultures were diluted 10-fold and allowed to grow for two hours before induction with 0.5 mM IPTG. After additional growth for two hours, cells were harvested by centrifugation. Preparation of spheroplasts and 4Bl binding Spheroplasts were prepared by lysozyme-EDTA treatment as de- scribed (Sun et al., 1996). Aliquots [0.5 mL containing 0.3 m g / d protein in 100 mM potassium phosphate (KPi;pH 7.5)/0.5 M sucrose/5% bovine serum albumin (BSA)] were mixed with 5 p L of purified 4B1 (5 mgfml), incubated at room temperature for one hour, centrifuged, washed once in incubation buffer without BSA, and resuspended in 0.4 mL of incubation buffer. Two microliters of [Iz5I]protein A (30 mCi/mg; 100 mCi/mL) were added, and in- cubation was continued for 45 min. The spheroplasts were then centrifuged, washed once by centrifugation, and resuspended to 50 p L in incubation buffer without BSA. Bound radioactivity was measured by liquid scintillation spectrometry using Scintsafem Eiono 1 cocktail buffer. Preparation of right-side-out (RSO) membrane vesicles RSO membrane vesicles were prepared by osmotic lysis of sphe- roplasts prepared with lysozyme and EDTA (Kaback, 1971; Short et al., 1975). Transport assays Active transport in RSO membrane vesicles was assayed under oxygen in the presence of 20 mM potassium ascorbate and 0.2 mM phenazine methosulfate (PMS) (Konings et al., 1971; Kaback, 1974). Membrane vesicles were suspended in 100mM KP, (pH 7.5)/ 10 mM MgSO, buffer to a concentration of 2 mg/mL total protein, and [I4C] lactose (10 mCi/mmol), [3H] melibiose (40mCi/mmol) or [3H] raffinose (50 mCi/mmol) were used at a final concentra- tion of 0 4 mM as indicated. . Quantitation o lac and sucrose permease f Spheroplast samples containing E. coli lac or sucrose permease (with the lac permease C-terminus asan epitope; Sahin-Tbth et al., 1995) weresubjected to sodium dodecylsulfatepolyacrylamide gel electrophoresis (SDS-PAGE) (Newman et al., 1981), and immu- noblot analyses were carried out by using rabbit polyclonal anti- body against the c-terminus of lac permease (Carrasco et al., 1984b). The amount of permease was quantitated with PhosphorImager Model 425F (Molecular Dynamics)as described (Sun et al., 1996). Protein determination Protein concentrations were determined as described (Peterson, 1977) with BSA as standard. Acknowledgments We thank T.H. Wilson, F. Ausubel, and R. Schmittforprovidingcell strains and/or plasmids and Kerstin Stempel for synthesizing oligodeoxy- nucleotides. References Aslandis C , Schmitt, R. 1990. Regulatory elements of the raffinose operon: Nucleotide sequences of operator and repressor genes. J Bacteriol I72:2 176- 2180. Bockmann I, Heuel H, Lengeler JW. 1992. Characterization of a chromosomally encoded, non-PTS metabolic pathway for sucrose utilization in E. coli EC3132. Mol Gen Genet 23522-32. Viitanen P,Herzlinger D, Kaback HR. 1984a. Monoclonal antibod- Carrasco N, ies against the lac carrier protein from Escherichia coli. 1. Functional stud- ies. Biochemistry 23:3681-3687. Carrasco N, Herzlinger D, Mitchell R, DeChiara S, Danho W, Gabriel TF, Kaback HR. 1984b. Intramolecular dislocation of the COOH terminus of the lac carrier protein in reconstituted proteoliposomes. Pmc Natl Acad Sci USA 81:4672-4676. Dunten RL, Sahin-Toth M, Kaback HR. 1993. Role of the charge pair formed by aspartic acid 237 and lysine 358 in the lactose permease of Escherichia coli. Biochemistry 32:3139-3145. Frillingos S, Kaback HR. 1996a. Chemical rescue of Asp237-+Ala and Lys358jAla mutants in the lactose permease of Escherichia coli. Biochem- istry 35:13363-13367. Frillingos S, Kaback HR. 1996b. Monoclonal antibody 4B1 alters the pK, of a carboxylic acid at position 325 (helix X) of the lactose permease of Esch- erichia coli. Biochemistry 3510166-10171. Frillingos S, Sahin-T6th M, Lengeler JW, Kaback HR. 1995. Helix packing in the sucrose permease of Escherichia coli: Properties of engineered charge pairs between helices VI1 and XI.Biochemistry 34:9368-9373. Frillingos S, Sahin-Toth M, Persson B, Kaback HR. 1994. Cysteine-scanning mutagenesis of putative helix VI1in the lactose permease of Escherichia coli. Biochemistry 3333074-8081. Frillingos S . Wu J, Venkatesan P, KabackHR. 1997. Binding of ligand or monoclonal antibody AB1 induces discrete changes in the lactose permease of E. coli. Biochem, in press. Hattori M, Sakaki Y. 1986. Dideoxy sequencing method using denatured plas- mid templates. Anal Biochem 152:1291-1297. s HeMM,Voss J, Hubbell WL, KabackHR. 1995a. U e of designed metal binding sites to study helix proximity in the lactose permease of Escherichia coli. 2. proximity of helix IX (Arg302) with helix X (His322 and Glu325). Biochemistry 3415667-15670. 1995b. Use of designed metal- He MM,Voss I, Hubbell WL,KabackHR. binding sites to study helix proximity in the lactose permease of Escherichia coli: I. Proximity of helix VI1(Asp237 and Asp240) withhelices X (Lys319) and XI (Lys358). Biochemistry 34:15661-15666. Herzlinger D, Viitanen P, Carrasco N, Kaback HR. 1984.Monoclonal antibodies against the lac carrier protein from Escherichia coli. 2. Binding studies with membrane vesicles and proteoliposomes reconstituted with purified lac car- rier protein. Biochemistry 23:3688-3693. Ha SN, Hunt HD, Horton R M , F’ullen JK, Pease LR. 1989. Site-directed mu- tagenesis by overlap extension using the polymerase chain reaction. Gene 7751-59. Use of site-directed Jung K, Jung H, Wu J, E v e GG, KabackHR.1993. fluorescence labeling to sfudy proximity relationships in the lactose perme- ase of Escherichia coli. Biochemistry 32: 12273-12278. Jung K,Voss J. HeM, Hubbell WL, Kaback HR. 1995. Engineering a metal binding site within a polytopic membrane protein, the lactose permease of Escherichia coli. Biochemistry 34:6272-6277. Kaback HR. 1971. Bacterial membranes. Methods Enqmol XXIl99-120. Kaback HR. 1974. Transport in isolated bacterial membrane vesicles. Methods Enzymol 31:698-709. Kaback HR. 1983. The lac carrier protein in Escherichia coli: From membrane to molecule. J Membr Biol 7695-1 12. Kaback HR. 1989. Molecular biology of active transport: From membranes to molecules to mechanism. Harvey Lect 83:77-103. lactose permease of Kaback HR. 1992. In and out and up and down with the Escherichia coli. In: Friedlander M, Mueckler M, eds. International review o cytology, Vol. 137A. New York Academic Press. pp 97-125. f Kaback HR. 1996. The lactose permease of Escherichia coli: Past, present and future. In: Konings WN, Kaback HR, Lolkema JS, eds. Handbook of bio- logical physics II: Transport in prokaryotic and eukaryotic organisms. Am- sterdam: Elsevier. pp 203-227. King SC, Hansen CL, Wilson TH. 1991. The interaction between aspartic acid 237 and lysine 358 in the lactose carrier of Escherichia coli. Biochem Biophys Acta 1062:177-186. Konings WN, Barnes E Jr, Kaback HR. 1971. Mechanisms of active transport in isolated membrane vesicles. 2. The coupling of reduced phenazine meth- osulfate to the concentrative uptake of P-galactosides and amino acids. J B i d Chem 2465857-5861. 1510 Lee JL, Hwang PP, Hansen C, Wilson TH. 1992. Possible salt bridges between transmembrane a-helices of the lactose carrier of Escherichia coli. J Biol Chem 26720758-20764. Lee J, Okazaki N, Tsuchiya T, Wilson TH. 1994. Cloning and sequencing of the gene for the lactose carrier of Citrobacter freundii. Biochem Biophys Res Commun 203:1882-1888. Lee 11, Varela MF, Wilson T H . 1996. Physiological evidence for an interaction between Glu-325 and His-322 inthe lactose carrier of Escherichia coli. Biochim Biophys Acfa 12781 11-1 18. McMorrow I, Chin DT, Fiebig K, Pierce JL, Wilson DM, Reeve ECR, Wilson T H . 1988. The lactose carrier of Klebsiella pneumoniae M5al: The physi- ology of transport and nucleotide sequence ofthe lac Y gene. Biochim Biophys Acfa 945:315-323. Newman MJ, Foster DL, Wilson T H , Kaback HR. 1981. Purification and re- constitution of functional lactose carrier from Escherichia coli. J Biol Chem 256 11804-1 1808. Okazaki N, Tsuda M, Wilson TH, Tsuchiya T.1994. Characterization of the Biol Pharm Bull 17794-797. lactose transportsystem in Cifmbacferfreundii. Peterson GL. 1977. A simplification of the protein assay method of Lowry et ai. which is more generally applicable. Anal Biochem 83:346-356. Poolman B, Konings WN. 1993. Secondary solute transport in bacteria. Biochim Biophys Acta 118315-39. Sahin-Tbth M, Dunten RL, Gonzalez A, Kaback HR. 1992. Functional inter- actions between putative intramembrane charged residues in the lactose permease of Escherichia coli. Proc Natl Acad Sci USA 89:10547-10551. Sahin-Tbth M, Frillingos S, Lengeler J W , Kaback HR. 1995. Active transport by the CSCBgene product in Escherichia coli K-12. Biochem Biophys Res Commun 208:1116-1130. Sahin-T6th M, Kaback HR. 1993. Properties of interacting aspartic acidand lysine residues inthe lactose permease of Escherichia coli. Biochemistry 32:10027-10035. J. Sun et al. Sanger F, Nicklen S, Coulsen A R . 1977. DNAsequencing with chain-terminating inhibitors. Pmc Nafl Acad Sci USA 745463-5467. Short SA, Kaback H R , Kahn LD. 1975. Localization of D-lactate dehydrog- enase in native and reconstituted Escherichia coli membrane vesicles. J Biol Chem 2504291-4296. Sun J, Li J, Carrasco N, Kaback HR. 1997. The last two cytoplasmic loops in the lactose permease of Escherichia coli comprise a discontinuous epitope for a monoclonal antibody. Biochemistry 36:274-280. Sun J, Wu J, Carrasco N, Kaback HR. 1996. Identification of the epitope for monoclonal antibody 4B1 which uncouples lactose and proton translocation in the lactose permease of Escherichia coli. Biochemisfry 35:990-998. Teather RM, Bramhall J, Riede I, Wright JK, Furst M, Aichele G, Wilhelm V, et al. 1980. Lactose carrier protein of Escherichia coli. Structure and ex- pression of plasmids carrying the Y-gene of the lac operon. Eur J Biochem 108:223-231. van Iwaarden PR, Pastore IC, Konings WN, Kaback HR. 1991. Construction of a functional lactose permease devoid of cysteine residues. Biochemistry 30:9595-9600. Wilmot CM, Thornton JM. 1988. Analysis and prediction of the different types of &turn in proteins. J Mol Biol203:221-232. Wu J, Kaback HR. 1996. A general method for determining helix packing in membrane proteins in situ: Helices I and I1 are close to helix VI1in the lactose permease of Escherichia coli. Proc Natl Acad Sci USA 93:14498- 14502. Wu J, Perrin D, Sigman D, Kaback H. 1995. Helix packing of lactose permease in Escherichia coli studied by site-directed chemical cleavage. Proc Natl Acad Sci USA 92:9186-9190. Wu J, Voss J, Hubbell WL, Kaback HR. 1996. Site-directed spin labeling and chemical crosslinking demonstrate that helix V is close to helices VI1 and VI11 in the lactose permease of Escherichia coli. Proc Natl Acad Sci USA 93:10123-10127.