Suppressor Scanning at Positions 177 and 236 in the Escherichia coli Lactose/H⁺ Cotransporter and Stereotypical Effects of Acidic Substituents That Suggest a Favored Orientation of Transmembrane Segments Relative to the Lipid Bilayer

Abstract

Acidic substituents for Ala-177 (helix 6) or Tyr-236 (helix 7) in LacY cause effects on sugar recognition and cosubstrate coupling that are stereotypical of neutral substituents. Thus, helices 6 and 7 are probably oriented to produce little side-chain contact with the low dielectric lipid bilayer at positions 177 and 236.

The Escherichia coli lactose permease (LacY) is a classical prototype for tightly coupled H⁺/substrate cotransport. LacY mutants which uncouple H⁺/galactoside cotransport are said to cause “slip” or to catalyze an “internal leak.” It is important to localize sites of uncoupling within the LacY tertiary structure in order to appreciate in structural terms the difference between active transport and facilitated diffusion. Several laboratories have been involved in the discovery (2, 3), characterization (1, 8–11, 15, 16), and theoretical treatment (12, 13) of uncoupling sites in LacY. Here we have revisited the LacY uncoupling sites at Ala-177 and Tyr-236, showing for the first time that these positions can accommodate acidic (i.e., charged) substituents which have pleiotropic effects on sugar recognition, growth rate, and cosubstrate coupling. The stereotypical effects of both neutral and charged substituents on the LacY phenotype suggest that a common perturbation of structure and mechanism is responsible for (i) broadening the sugar recognition spectrum (sucrose-positive phenotype) and (ii) increasing the magnitude of internal leaks involving either the carrier-proton (CH) complex (proton-leaky phenotype), or the carrier-sugar (CS) complex (sugar-leaky phenotype).

Stereotypical sugar recognition effects.

Twelve different amino acid substituents (His, Gly, Glu, Phe, Ala, Cys, Pro, Ser, Gln, Tyr, Lys, and Leu) at positions 177 and 236 were screened for the LacY sucrose-positive phenotype via site-directed amber suppression scanning. Strains appropriate for the screen were constructed by introducing an invertase-positive plasmid, pRAF-S11 (10), together with an appropriate LacY(Am) expression plasmid (called pSCK184Y), into an existing series of amber suppressor strains (7) that were negative for LacY and invertase (sucrase). The results indicated that five substituents (Glu, Phe, Cys, Gln, and Leu) at position 177 and eight substituents (His, Gly, Glu, Phe, Ala, Ser, Gln, and Lys) at position 236 were fermentation positive (red colonies) on MacConkey agar containing sucrose.

Follow-up studies with several LacY missense mutants (A177D, A177E, Y236D, and Y236E) showed that only the A177D and A177E mutants were LacY sucrose positive on MacConkey agar. Apparently, asparagine contamination (7) inherent to the glutamate suppressor strain (15% Asn and 85% Glu) accounted for the initial fermentation-positive result. Transport experiments (Table 1) confirmed that the A177D and A177E mutants transport [³H]sucrose significantly better than wild-type LacY, whereas the Y236D and Y236E mutants do not. Parallel experiments (Table 1) with [methyl-¹⁴C]methyl- 1-thio-β-d-galactopyranoside ([¹⁴C]TMG) showed that acidic substituents at position 236 cause a far greater defect in galactoside accumulation than acidic substitutions at position 177. The disparity could be accounted for by the distinct internal leaks exhibited by these mutants.

TABLE 1.

Sugar transport

LacY mutation	Sugar transport
	TMG uptake ratioa	Sucrose uptake ratiob
Wild type	85	0.32 ± 0.02
A177D	57.64 ± 0.91	0.77 ± 0.01
A177E	35.85 ± 0.32	0.86 ± 0.01
Y236D	1.58 ± 0.03	0.44 ± 0.02
Y236E	1.80 ± 0.01	0.48 ± 0.02

^aUptake ratio ([TMG]_in/[TMG]_out) ± standard error of the mean after a 30-s exposure to 0.1 mM TMG outside. 

^bUptake ratio ([sucrose]_in/[sucrose]_out) ± standard error of the mean after a 30-min exposure to 10 mM sucrose outside. 

Stereotypical growth-inhibiting effects.

Considerable growth-inhibiting physiological stress was induced by adding 1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) to the culture medium (M9 minimal salts with 0.2% glucose) for cells harboring plasmids that encode acidic substituents at position 177 (Fig. 1A) or 236 (Fig. 1B). In contrast, IPTG had no effect on the growth rates of cells expressing wild-type LacY. An accepted explanation in the case of neutral substituents (1) is that mutations increase the magnitude of internal leaks involving either the CH complex (proton-leaky phenotype), or the CS complex (sugar-leaky phenotype). In the absence of IPTG, the physiological impact (i.e., on the transmembrane pH gradient [ΔpH]) of such leaks would be minimized. In the case of acidic substituents at positions 177 and 236, LacY-dependent H⁺ leaks were also found.

FIG. 1.

Growth-inhibiting toxicity of LacY mutants that carry acidic substitutions at positions 177 and 236. Optical density was monitored in cultures with (open symbols) or without (solid symbols) 1 mM IPTG to induce plasmid-borne lacY expression. (A) Growth of cells with wild-type LacY (circles) or A177D (squares) or A177E (triangles) mutant LacY. (B) Growth of cells with wild-type LacY (circles) or Y236D (squares) or Y236E (triangles) mutant LacY.

Stereotypical internal leaks.

A diminished ΔpH—monitored by [¹⁴C]benzoate distribution (8)—was found in cells expressing LacY mutants with acidic substituents at positions 177 and 236 (Fig. 2), indicating an internal leak catalyzed by the CH complex. Parallel studies (Fig. 2) showed that thiodigalactoside (TDG) can either ameliorate (position 177) or exacerbate (position 236) the proton leak. The latter effect on position-236 mutants is catalyzed by a futile cycle involving transmembrane isomerization of leaky CH and leaky CS complexes, whereas in position-177 mutants, formation of a nonleaky CS complex improves ΔpH by decreasing the mole fraction of leaky CH complex (1). The direct effect of sugar back leakage (via CS) and the sugar-dependent dissipation of ΔpH provide a basis for understanding the profound galactoside accumulation defect observed in the position-236 mutants (Table 1).

FIG. 2.

Effect of TDG on the transmembrane ΔpH. For the indicated LacY mutants, the transmembrane ΔpH was monitored either in the presence or in the absence (control) of 1 mM TDG.

Stereotypical high-affinity TDG site.

Two important inferences about the acidic substituents may be drawn from the observation that TDG affects the ΔpH: (i) the proton leak is LacY dependent, and (ii) the LacY sugar binding site remains substantially intact—despite the observation that charged substituents cause stereotypic broadening of the ligand recognition repertoire. Indeed, the affinity for TDG is clearly quite high (Fig. 3). Together these results suggest that acidic substituents at positions 177 and 236 cause neither nonspecific proton leakiness via detergent-like effects on the membrane nor a nonspecific galactoside binding site denaturation that produces sucrose permeation.

FIG. 3.

Retention of high TDG affinity by LacY mutants carrying acidic substitutions at positions 177 and 236. The transmembrane ΔpH was monitored as a function of TDG concentration in E. coli expressing either wild-type LacY (•) or the A177E (▵), A177D (▴), Y236D (□), or Y236E (▪) LacY mutant.

Structural implications.

Since Ala and Tyr are relatively hydrophobic, one might have assumed a priori that charged residues would perform poorly as substituents at positions 177 and 236 in LacY—indeed, only neutral substituents had been studied previously (2, 5, 8, 10). Through site-directed suppressor scanning, this study explored a broad range of side-chain chemistries, revealing (i) that LacY tolerates acidic substituents at positions 177 and 236 and (ii) that the resulting phenotypes are virtually the same (5, 8, 10) as for neutral substituents (i.e., charges do not uniquely alter the protein structure). Thus, it is reasonable to suggest that in current models of the LacY tertiary structure (Fig. 4), the side chains at positions 177 and 236 ought to project away from the lipid and toward a more hydrophilic environment (higher dielectric medium)—where the high cost of burying an unpaired charge in hydrocarbon (6, 17) could be avoided. In terms of the model, such charge-stabilizing orientations of helices 6 and 7 suggest visually how substituents at positions 177 and 236 could indirectly perturb the transport channel based more on side-chain volume than on side-chain chemistry. This information is useful, since the precise orientation of transmembrane segments relative to the lipid bilayer remains to be elucidated.

FIG. 4.

A working model of the LacY tertiary structure. The solid ovals emphasize the positions of helix 6 (Ala-177) and helix 7 (Tyr-236) within the context of Brooker’s symmetrical lac permease model (4, 14). These helices are probably oriented so as to shield positions 177 and 236 from intimate contact with the lipid environment.