J Biol Chem, Vol. 274, Issue 12, 8061-8067, March 19, 1999

Hierarchy of Sorting Signals in Chimeras of Intestinal Lactase-Phlorizin Hydrolase and the Influenza Virus Hemagglutinin^*

Ralf Jacob^§, Ute Preuss^§¶, Petra Panzer^¶, Marwan Alfalah, Stephanie Quack^¶, Mike G. Roth, Hussein Naim^**, and Hassan Y. Naim^¶

From the  Department of Physiological Chemistry, School of Veterinary Medicine Hannover, D-30559 Hannover, Germany, the ^¶ Protein Secretion Group, Heinrich Heine University of Düsseldorf, D-40225 Düsseldorf, Germany, the  Department of Biochemistry, University of Texas Southwestern Medical Center, Dallas, Texas 75235, and the ^** Institute of Molecular Biology Division I, University of Zürich, CH-8057 Zürich, Switzerland

	ABSTRACT

TOP ABSTRACT INTRODUCTION REFERENCES

Lactase-phlorizin hydrolase (LPH) is an apical protein in intestinal cells. The location of sorting signals in LPH was investigated by preparing a series of mutants that lacked the LPH cytoplasmic domain or had the cytoplasmic domain of LPH replaced by sequences that comprised basolateral targeting signals and overlapping internalization signals of various potency. These signals are mutants of the cytoplasmic domain of the influenza hemagglutinin (HA), which have been shown to be dominant in targeting HA to the basolateral membrane. The LPH-HA chimeras were expressed in Madin-Darby canine kidney (MDCK) and colon carcinoma (Caco-2) cells, and their transport to the cell surface was analyzed. All of the LPH mutants were targeted correctly to the apical membrane. Furthermore, the LPH-HA chimeras were internalized, indicating that the HA tails were available to interact with the cytoplasmic components of clathrin-coated pits. The introduction of a strong basolateral sorting signal into LPH was not sufficient to override the strong apical signals of the LPH external domain or transmembrane domains. These results show that basolateral sorting signals are not always dominant over apical sorting signals in proteins that contain each and suggest that sorting of basolateral from apical proteins occurs within a common compartment where competition for sorting signals can occur.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION REFERENCES

Polarized cells such as neurons and epithelial cells maintain separate plasma membrane domains, each with a distinct protein and lipid composition, through intracellular sorting mechanisms that recognize classes of proteins and deliver them into separate vesicles for transport to the correct surface domain (1, 2). Sorting to the correct membrane is essential for the proteins to exhibit their biological functions, whereas missorting often results in pathological conditions (3, 4). The recognition event responsible for sorting has been under intense investigation for two decades, and a number of peptide sequences capable of specifying transport to the basolateral surface of epithelial cells (5-11), or cell body of neurons (12-16), have been characterized. All of these signals are located in the cytoplasmic domains of transmembrane glycoproteins. In addition to basolateral signals, three types of signals for sorting proteins to the apical surface of epithelial cells, or axon of neurons, are known. Glycolipid anchors direct proteins to the apical surface of several types of epithelial cells (17, 18), apparently by associating in the trans-Golgi network (19, 20) with detergent-insoluble membrane domains enriched in glycosphingolipids and cholesterol (21). Oligosaccharides on some secreted proteins appear to specify apical transport (22), although this mechanism does not apply to all secreted proteins (23-26).

For many transmembrane glycoproteins, deletion of cytoplasmic sequences containing a basolateral sorting signal results in efficient transport of the protein to the apical surface, rather than the random transport expected for the deletion of specific sorting information (10, 27-30). For other proteins, deletion of cytoplasmic sequences caused randomized transport, proving that transport to the apical surface does not occur by default (9, 31). In the reverse approach, introducing basolateral sorting signals into the cytoplasmic domain of the influenza hemagglutinin (HA)¹ was shown to have a dominant effect over apical sorting information (8, 31, 32) that has been recently localized to the transmembrane domain (8). These observations implied that some proteins carry apical sorting information that is recessive to cytoplasmic basolateral sorting signals. Basolateral signals could dominate over apical signals simply by being recognized earlier in the biosynthetic pathway, or sorting could occur in a common compartment where basolateral signals might bind tighter to the sorting machinery than apical signals. To investigate these questions, we attached a series of basolateral sorting signals to the strictly polarized membrane protein of small intestinal epithelial cells, lactase-phlorizin hydrolase (LPH, EC 3.2.1.23-3.2.1.62), and determined their effect on the sorting of LPH.

LPH, an integral type I membrane glycoprotein, is 1927 amino acids long containing a membrane anchor of 19 contiguous hydrophobic amino acids and a cytoplasmic domain of 26 amino acids. It is synthesized as a precursor with apparent molecular masses of 215 and 230 kDa, representing the mannose-rich (pro-LPH_h) and complex (pro-LPH_c) glycosylated forms. Maturation of LPH involves proteolytic cleavage after complex glycosylation of the precursor to yield the brush-border form of 160 kDa (33-37). LPH is targeted strictly to the apical membrane of intestinal epithelial cells and Madin-Darby canine kidney (MDCK) cells (38). To investigate the position and relative strength of the apical sorting signal of LPH, sorting of a tailless LPH mutant (LPH_ct) (39) and chimeric proteins made by fusing LPH external and transmembrane sequences to the short, 12-amino acid-long cytoplasmic domain of several HA mutants was studied in MDCK cells. Wild type HA lacks basolateral sorting signals and is transported to the apical surface (40), but point mutations in the cytoplasmic domain of HA were identified that created both internalization signals and dominant basolateral sorting signals (8, 32, 41-43). In contrast to their function in HA, these basolateral sorting signals did not affect strict apical delivery of LPH. However, the chimeric LPH proteins gained the internalization capacity similar to those of the HA counterparts.

	EXPERIMENTAL PROCEDURES

Construction of cDNA Encoding Mutant LPH_ct and Chimeras of Intestinal LPH and Influenza Virus HA-- Standard recombinant DNA techniques were employed according to Sambrook et al. (44). The LPH mutant lacking the cytoplasmic tail, LPH_ct, has been reported previously (39). cDNAs encoding the external and transmembrane domain of LPH (amino acids 1-1901) (45) and the short, 12-amino acid-long, cytoplasmic domain of HA (amino acids 536-547) (36) (denoted as LPH-HA) were generated by polymerase chain reaction SOEing (46). The polymerase chain reaction product encoding LPH-HA_wt was cloned into the expression vector pJB20 (32). A similar strategy was utilized to construct the LPH-HA chimeras which contain a single mutation (LPH-HA_Y543) or a double mutation (LPH-HA_Y543/Y546 and LPH-HA_Y543/R546) in the HA cytoplasmic domain. The sequence of LPH_ct and each LPH-HA chimera was determined by sequencing with a Sequenase kit according to the instructions of the manufacturer (U. S. Biochemical Corp.).

Transfection and Generation of Stable Cell Lines-- MDCK cells and COS-1 cells were maintained in Dulbecco's modified Eagle's medium (DMEM, Life Technologies, Inc., Eggenstein/Germany) supplemented with 10% fetal bovine serum, 2 mM glutamine, 50 units/ml penicillin, and streptomycin at 37 °C in a 5% CO₂ atmosphere. Cells were transfected with 5 µg of the appropriate recombinant DNA using DEAE-dextran (for COS-1 cells) as described (47) or Polybrene (for MDCK cells) (38). Stably transfected MDCK cells were selected in the presence of 0.25 mg/ml active G418 (Life Technologies, Inc.), and after 18-23 days, surviving colonies were isolated with cloning rings. Stable transformants expressing LPH_ct or LPH-HA chimeras were screened by immunoprecipitation and by immunofluorescence staining. Expression of LPH-HA chimeras in intestinal Caco-2 cells was preformed transiently on membrane filters using the calcium phosphate procedure (48). Here, the cells were grown to confluency and the corresponding DNA was added at 2 µg/ml. For higher transfection efficiency on filters, the cells were treated prior to transfection with trypsin to dissociate the cells and to achieve an optimal exposure of cells to DNA. Three days post-transfection the cells were processed for cell surface immunoprecipitation with mAb anti-LPH (see below) after biosynthetic labeling for 18 h. We found that a 3-day period was sufficient for the cell layer to achieve complete polarity. This was biochemically assessed by cell surface immunoprecipitation of sucrase-isomaltase, which is targeted in Caco-2 cells to the apical membrane. Infection of Caco-2 cells with HA cDNA was performed as described by Naim and Roth (42) for MDCK cells.

Biosynthetic Labeling of Cells, Immunoprecipitation, and SDS-PAGE-- Metabolic labeling of MDCK cells grown on filters or plated in six-well culture dishes was performed as described previously (38). MDCK clones expressing LPH_ct or LPH-HA chimeras were labeled for 1 h with 100 µCi of [³⁵S]methionine (10 mCi/ml L-[³⁵S]Redivue^TM Pro-mix^TM, Amersham, Braunschweig/Germany) and chased for different times with unlabeled methionine. Caco-2 cells expressing transiently transfected LPH-HA_wt or LPH-HA_Y543/F546 were labeled continuously for 18 h to ensure a maximum labeling of the expressed recombinant proteins. Caco-2 cells infected with HA_wt or HA_Y543/F546 were continuously labeled for 2 h. Cell lysates were immunoprecipitated with mouse mAb anti-LPH (from hybridoma HBB 1/900/34/74) (34) as described by Naim et al. (47), and cell surface antigens were immunoprecipitated from intact cells on filters by addition of anti-LPH or anti-HA (42) antibody to either the apical or basolateral compartments. The immunoprecipitates were analyzed by SDS-PAGE according to the method of Laemmli (49). After electrophoresis the gels were fixed, soaked in 16% salicylic acid for signal amplification, and subjected to fluorography.

Detergent Extractability of LPH and Sucrase-Isomaltase-- MDCK-ML cells expressing LPH were biosynthetically labeled for 1 h with [³⁵S]methionine and chased over several time points. The cells were solubilized in the cold for 2 h with 1% Triton X-100 in 25 mM Tris-HCl, pH 8.0, 50 mM NaCl. The detergent extracts were centrifuged, and the supernatant was immunoprecipitated with mAb anti-LPH. The pellet was dissolved by boiling in 1% SDS for 10 min. Thereafter 10-fold volume of buffer containing 1% Triton X-100 was added. These extracts were centrifuged and the supernatant was immunoprecipitated with a mixture of two monoclonal antibodies, MLac 6 and MLac 10, that recognize denatured and native forms of LPH (39). A similar experimental procedure was followed to assess the detergent solubility of intestinal brush-border sucrase-isomaltase by using the colon carcinoma Caco-2 cells. Immunoprecipitation of native and denatured forms of sucrase-isomaltase was performed with a mixture of monoclonal antibodies (34). The immunoprecipitates were analyzed by SDS-PAGE on 5% or 6% gels.

Internalization Assays-- COS-1 cells transiently transfected with pJB20 encoding LPH_ct or the LPH-HA chimeras were grown in duplicate on coverslips. 48 h post-transfection, cells were rinsed in ice-cold DMEM and placed on ice. The cells were incubated for 2 h with mAb anti-LPH diluted 1:200 in 5% bovine serum albumin/PBS. Unbound antibody was removed by three washes with DMEM containing 10% fetal calf serum. Control samples were retained at 4 °C and for a second set of samples the temperature was raised to 37 °C with DMEM for 10 min in a circulating water bath. The cells were chilled on ice, rinsed in ice-cold PBS, and then fixed with 2% paraformaldehyde for 20 min at room temperature. After two washes with PBS, the cells were incubated with fluorescein isothiocyanate-conjugated goat anti-mouse antibody (Boehringer Mannheim, Mannheim, Germany), diluted 1:100 in 5% bovine serum albumin/PBS for 30 min at room temperature. Intracellular localization of proteins was assessed in transfected cells that were permeabilized with 0.1% Triton X-100 after fixing. The cells were examined with an Axioplan fluorescence microscope (Zeiss, Oberkochen, Germany) equipped with a 100× immersion objective.

	RESULTS

The Cytoplasmic Tail of LPH Is Not Required for Its Sorting to the Apical Membrane-- To evaluate the role of the cytoplasmic tail of LPH in its transport and sorting, we constructed a mutant cDNA lacking the entire sequence encoding the cytoplasmic tail of LPH (denoted pro-LPH_ct, Table I) and found that in COS-1 cells this deletion did not affect the transport-competence of the molecule (39). Biosynthetic processing of LPH_ct in a MDCK cell line continuously expressing the protein was similar to that of wild type pro-LPH in intestinal and MDCK cells (35, 38). The first detectable biosynthetic form, the 215-kDa mannose-rich pro-LPH_ct species, chased into the complex glycosylated 230-kDa pro-LPH_-ct polypeptide after 4-6 h, and a cleaved form of LPH_ct (approximate apparent molecular mass 160 kDa) appeared (Fig. 1). The cleaved 160-kDa form is the tailless analogue of LPH previously characterized in intestinal biopsy specimens (35), in transfected MDCK and CHO cells (38, 50), and will be therefore denoted LPH_ct.

View larger version (31K): [in this window] [in a new window]   Fig. 1.   Transport kinetics of LPHct in MDCK cells. MDCK cells stably expressing LPHct were pulse-labeled for 1 h with [35S]methionine and chased for the indicated times with 2.5 mM unlabeled methionine. LPHct was purified by immunoprecipitation and analyzed by electrophoresis on a 6% SDS gel and fluorography.

The kinetics of appearance of pro-LPH_ct in the apical or basolateral domains were investigated by cell surface immunoprecipitation of pro-LPH_ct and its derivative LPH_ct from cells grown on transparent polyester membrane filters as described previously for wild type LPH (38). Fig. 2 shows that the complex glycosylated 230-kDa pro-LPH_ct precursor and LPH_ct appeared after 4 h of chase at the apical surface. The intensity of the bands isolated from the apical domain became stronger at the 6-h chase point. No significant bands corresponding to these two LPH species were detected at the basolateral surface. Together, the pulse-chase and sorting analyses indicate that intracellular processing and targeting of pro-LPH_ct in MDCK cells is similar to its wild type pro-LPH counterpart and that the cytosolic portion of pro-LPH is devoid of apical sorting signals.

View larger version (78K): [in this window] [in a new window]   Fig. 2.   Polarized delivery of newly synthesized LPHct to the surface of MDCK cells. MDCK cells stably transfected with LPHct were grown on filters, pulse-labeled with [35S]methionine for 1 h, and chased in medium with 2.5 mM unlabeled methionine for the indicated times. LPHct was immunoprecipitated either from the apical (a) or the basolateral (b) surface and analyzed as in Fig. 1.

Apical Transport of LPH Is Independent of Its Association with Sphingolipid-Cholesterol Rafts-- A number of apically sorted proteins, such as influenza virus neuraminidase, HA, and some intestinal proteins, have been shown to be selectively associated with sphingolipid-cholesterol rafts (21, 51, 52). One of the characteristics of these protein-membrane structures is their insolubility in detergents such as Triton X-100 at 4 °C. We therefore examined whether pro-LPH is associated with rafts in the polarized cell line MDCK-ML that expresses pro-LPH (38). Cells were pulse-labeled with [³⁵S]methionine for 1 h and chased for various time points. The cells were extracted with Triton X-100, and the detergent solubility of pro-LPH biosynthetic forms versus insolubility was examined. Fig. 3 shows two representative chase time points. The 215-kDa mannose-rich pro-LPH appeared in the supernatant fraction (denoted S) at the earliest chase time point (1 h pulse, 0 h chase), and the pellet (P) was devoid of this form. The complex glycosylated 230-kDa pro-LPH was also found exclusively in the supernatant after 4 h of chase together with the mannose-rich 215-kDa species. Similar results were obtained with chase points earlier and later than 4 h. The absence of pro-LPH in the detergent insoluble fraction (P) indicates that pro-LPH is not associated with sphingolipid-cholesterol rafts. These data agree with previous observations that pro-LPH was only found in the detergent-soluble form in biosynthetically labeled explants from the pig small intestine (52). By contrast to pro-LPH, the 245-kDa complex glycosylated mature form of another brush-border protein, sucrase-isomaltase (34, 53), could be found in the Triton X-100 insoluble pellet (P) after 3 h of chase in biosynthetically labeled intestinal Caco-2 cells (Fig. 3). The mannose-rich form (210-kDa), on the other hand, was found only in the supernatant (Fig. 3). This result indicates that the 245-kDa mature form of sucrase-isomaltase is associated with sphingolipid-cholesterol rafts and suggests a role of these structures in the targeting of this glycoprotein to the apical membrane, but not in the sorting of LPH.

View larger version (44K): [in this window] [in a new window]   Fig. 3.   LPH is not associated with sphingolipid-cholesterol rafts. MDCK-ML and intestinal Caco-2 cells expressing LPH and sucrase-isomaltase (SI), respectively, were biosynthetically labeled for 1 h with [35S]methionine and chased over several time points. Triton X-100 detergent extracts were centrifuged, and the supernatants were immunoprecipitated with mAb anti-LPH or mAb anti-sucrase-isomaltase. The cellular pellets were extracted by boiling in 1% SDS followed by dilution with 10-fold volume of buffer containing 1% Triton X-100. These extracts were centrifuged, and the supernatants were immunoprecipitated with antibodies that recognize denatured and native forms of LPH or sucrase-isomaltase. The immunoprecipitates were analyzed by SDS-PAGE on 6% (LPH) or 5% (sucrase-isomaltase) gels and fluorography. Representative chase time points are shown.

The Chimeric LPH-HA Mutants-- The high fidelity of sorting of LPH (more than 90% was apically targeted) and its cleaved product (exclusively apically located) strongly suggest that this protein contains strong apical sorting signals. The strength and efficiency of the basolateral sorting machinery has lead to the notion that basolateral signals are dominant over apical signals in MDCK cells. Consistent with this concept, HA mutants constructed to contain basolateral signals were sorted to the basolateral rather than the apical membrane (8). To examine the strength of the potential apical sorting signal within the LPH molecule, the same cytosolic sequences that completely reversed the polarity of HA were fused to the ectodomain and transmembrane domain of pro-LPH (the chimeras are indicated LPH-HA). The cytosolic tails of the HA mutants used in this study were derived from mutant HA_Y543, HA_Y543/F546, and HA_Y543/R546. The corresponding chimeras are referred to as LPH-HA_Y543, LPH-HA_Y543/F546, and LPH-HA_Y543/R546. As a control, the pro-LPH tail was replaced by the cytosolic tail of wild type HA (LPH-HA_wt). These chimeric proteins were stably expressed in MDCK cells, and the biosynthesis, processing, transport, and sorting of the chimeric proteins were investigated.

LPH-HA Chimeras Containing Basolateral Sorting Signals Are Sorted to the Apical Membrane-- Exchanging the cytosolic tails of pro-LPH with mutants of the HA tail had no significant effects on the biosynthesis, processing, and transport rate of pro-LPH. In a fashion similar to wild type pro-LPH (38), the mutants were processed from the mannose-rich 215-kDa species to the complex glycosylated mature 230-kDa and proteolytically cleaved to the 160-kDa LPH analogues (Fig. 4A). To examine whether the transplanted cytosolic tails in pro-LPH had affected its sorting to the apical in a fashion similar to the effects observed with mutant HA molecules, monolayers of MDCK cells expressing LPH-HA_Y543, LPH-HA_Y543/F546, LPH-HA_Y543/R546, or LPH-HA_wt were grown on filters and were pulse-labeled with [³⁵S]methionine for 1 h and chased for several intervals. Cell surface immunoprecipitation with mAb anti-LPH was performed followed by SDS-PAGE. Fig. 4B demonstrates that all the biosynthetic forms of the LPH chimeras, i.e. the uncleaved complex glycosylated pro-LPH and the corresponding cleaved LPH analogues, appeared exclusively in the apical membrane. Very little, if any, was found at the basolateral membrane. Even the chimera containing the double mutation (Tyr⁵⁴³/Phe⁵⁴⁶) in the cytosolic domain was not effective in the context of the pro-LPH species. By contrast, this mutation completely reversed the sorting of HA from an apically to a basolaterally sorted molecule (8). Since the proportion of the labeled species located in the basolateral membrane was minor throughout the chase periods and because this proportion did not change in relation to the apical proportion, we conclude that the transport of the chimeras to the apical surface was direct, as it was the case with wild type pro-LPH (38). We demonstrate that not even a minor effect on the polarized sorting of pro-LPH to the apical membrane could be discerned when basolateral signals in the cytosolic tail of HA were examined in the pro-LPH species.

View larger version (56K): [in this window] [in a new window]   Fig. 4.   A, transport kinetics of LPH-HA chimeras. MDCK cells expressing the chimeras LPH-HAwt, LPH-HAY543, LPH-HAY543/F546, or LPH-HAY543/R546 were biosynthetically labeled for 1 h with [35S]methionine and were chased over a period of 6 h. Samples were analyzed by SDS-PAGE on 6% slab gels and fluorography. B, polarized expression of LPH chimeras in MDCK cells. Monolayers of MDCK cells expressing LPH-HAwt, LPH-HAY543, LPH-HAY543/F546, or LPH-HAY543/R546 were grown on filters. Six days after confluence, the cells were labeled with [35S]methionine for 1 h and chased for the indicated times. Chimeras were immunoprecipitated either from the apical (a) or basolateral (b) surface with mAb anti-LPH and analyzed by SDS-PAGE on 6% gels and fluorography. C, polarized expression of LPH chimeras in Caco-2 cells. LPH-HAwt and LPH-HAY543/F546 were transiently expressed in Caco-2 cells. Three days after transfection and confluence, the cells were continuously labeled with [35S]methionine for 18 h. Chimeras were immunoprecipitated either from the apical (a) or basolateral (b) surface with mAb anti-LPH and analyzed by SDS-PAGE on 6% gels and fluorography. D, polarized expression of wild type and mutant HA forms in Caco-2 cells. HAwt and HAY543/F546 were expressed by infection of Caco-2 cells grown on membrane filters 6 day after confluence. The cells were continuously labeled with [35S]methionine for 2 h. HAwt and HAY543/F546 were immunoprecipitated either from the apical (a) or basolateral (b) surface with anti-HA and analyzed by SDS-PAGE on 12.5% gels and fluorography.

To examine whether a similar or a different sorting behavior of these mutants occurs in an enterocytic cell line, we used colon carcinoma Caco-2 cells. In these cells endogenous or recombinant LPH is processed in a similar fashion to its counterparts in the intestinal mucosa and is targeted to the apical membrane (54). For this purpose the LPH-HA_wt and LPH-HA_Y543/F546 chimeras as well as the full-length HA cDNA analogues, i.e. HA_wt and HA_Y543/F546, were expressed in Caco-2 cells. Fig. 4C demonstrates that the LPH-HA chimeras examined, LPH-HA_wt and LPH-HA_Y543/F546, behaved in a fashion similar to their analogues in MDCK cells. Here, pro-LPH (230-kDa) as well as the cleaved 160-kDa LPH species were sorted to the apical membrane. Likewise, the biosynthesis and sorting of HA_wt and HA_Y543/Y546 Caco-2 cells were essentially similar to their counterparts in MDCK cells (Fig. 5D). Thus, HA_wt was predominantly found at the apical and HA_Y543/F546 at the basolateral membrane. In essence, the data demonstrate that the sorting pathways for the LPH-HA chimeras and also for the HA molecule are similar in MDCK and Caco-2 cells.

View larger version (101K): [in this window] [in a new window]   Fig. 5.   Internalization of LPHct and LPH-HA chimeras. Duplicates of COS-1 cells expressing LPHct (A), LPH-HAwt (B), LPH-HAY543 (C), LPH-HAY543/F546 (D), and LPH-HAY543/F546 (E) were treated with mAb anti-LPH antibody at 0 °C. One set of transfected cells was kept on ice as controls (indicated as 0 min), whereas the other set of cells was chased for 10 min at 37 °C to induce internalization (indicated as 10 min). All cells were fixed with 2% paraformaldehyde and either permeabilized with Triton X-100 (denoted as internal, b and d) or non-permeabilized (denoted as surface, a and c). The cells were stained with a second fluorescein isothiocyanate-conjugated antibody to determine the location of anti-LPH antibodies bound to the expressed proteins.

Internalization of LPH-HA Chimeras Is Dependent on Mutations in the Cytoplasmic Domain-- Two possibilities could explain the apical sorting of the LPH-HA chimeras. The apical sorting signal of pro-LPH might be dominant over the basolateral signals or the introduced signals might not function in the context of pro-LPH and were not accessible to the sorting machinery. One way to examine the latter possibility is to determine whether these sequences function as internalization signals in the context of the pro-LPH sequences, as they do in the context of HA (55). Normally, LPH does not contain an internalization signal; it escapes coated pits and remains exclusively at the cell surface. To determine whether the LPH-HA mutants were internalized, expression plasmids encoding each protein were transfected into COS-1 cells. Wild type pro-LPH and the tailless pro-LPH_ct mutant were included as controls, since neither should be internalized. Transfected COS-1 cells were grown on coverslips for 48 h post-transfection and were assayed for endocytosis by indirect immunofluorescence procedures (Fig. 5). Cells expressing either tailless pro-LPH_ct or LPH-HA_wt that were incubated with monoclonal anti-LPH antibody at 0 °C (Fig. 5A, a) or at 37 °C to allow endocytosis to proceed, had predominantly bright staining at the cell surface. By contrast, all three chimeric proteins containing internalization signals were internalized. Most notably, cells expressing the LPH chimeras with HA tails containing the double mutations, Tyr⁵⁴³/Phe⁵⁴⁶ and Tyr⁵⁴³/Arg⁵⁴⁶ revealed punctate fluorescence staining after incubation at 37 °C (Fig. 5, D and E, panel d in each case) but fluorescence that was restricted to the cell surface at 0 °C (Fig. 5, D and E, panel a in each case). LPH-HA_Y543 revealed a punctate pattern to a lesser extent (Fig. 5C), as would be expected from its slower rate of internalization (43, 55). These data demonstrate that the cytoplasmic sequences of the LPH-HA chimeras were available to be bound by components of the internalization apparatus and therefore must be exposed to the cytosol.

	DISCUSSION

In MDCK cells basolateral signals have been observed to dominate over apical signals when both are present in the same protein. Many basolateral proteins become apical proteins when the cytoplasmic basolateral sorting signal is mutated or removed. Basolateral sorting signals, when transferred to an apical protein, direct the chimeric protein to the basolateral surface. A single exception to this generality has been reported recently. When a portion of the ectodomain of the normally apically expressed neurotropin receptor is removed, the mutated protein is routed to the basolateral surface (56). However, it is not yet clear whether the neurotrophin receptor contains a recessive basolateral signal or whether the deletion in the ectodomain causes conformational changes in the protein that expose a cryptic basolateral signal that is dominant when available to interact with the sorting machinery. To address the question of whether basolateral signals are always dominant over apical signals, we made a series of mutants of LPH that contain overlapping basolateral and internalization signals that differed in the efficiency with which each specified either internalization or basolateral sorting (8, 42). Because these signals had dual function, we could determine whether the signals were available to interact with cellular sorting machinery by monitoring the capacity of the internalization signals to allow endocytosis of LPH, which normally lacks that capacity. Our results show clearly that a series of basolateral sorting signals capable of directing the influenza virus HA to the basolateral surface could not redirect the LPH, although the internalization signals in those sequences did cause LPH to be internalized. These results demonstrate that basolateral sorting signals are not always dominant over apical signals. This eliminates the possibility that basolateral signals are simply recognized earlier in the biosynthetic pathway and suggests that sorting is determined by the relative affinity of various sorting signals for the sorting machinery.

The nature of the signal or signals in LPH that allow it to be sorted so efficiently to the apical surface of epithelial cells remains to be determined. In contrast to basolateral signals, which all appear to be short amino acid motifs located in the cytosolic domain, three quite different features have been proposed to be important in apical targeting of transmembrane and secreted proteins in polarized epithelial cells (18, 21, 57). Recent data indicate that oligosaccharides can mediate protein sorting to the apical membrane. For instance, engineering two N-linked glycosylation sites into the normally unglycosylated growth hormone leads to a polarized targeting to the apical membrane of the otherwise unsorted protein (22). Another type of glycosylation, O-glycosylation, may constitute a targeting signal to the apical membrane (56). As yet unidentified apical sorting signals have been proposed to reside in the ectodomain of a number of proteins, such as intestinal brush-border proteins (58), but whether these signals depend upon glycosylation or not is currently unclear. Finally, protein association with sphingolipid-cholesterol rafts has been proposed as a potential mechanism in targeting proteins to the apical surface (22). This association occurs in the trans-Golgi network and results in detergent-insoluble membranes. Three examples are known of apical signals residing in transmembrane segments of proteins that associate with detergent insoluble membranes (59-61), although detergent insolubility has been shown not to be sufficient for apical sorting in one of these cases (60).

For LPH, a number of observations strongly support the idea that apical sorting signals are located in a specific portion of the ectodomain extending from Ala⁸⁶⁹ to Ile¹⁶⁴⁶. The LPH precursor, pro-LPH, is cleaved intracellularly to LPH (Ser²⁰-Arg⁸⁶⁸) and LPH, which is targeted to the brush-border membrane. Cleavage is not required for sorting, since the uncleaved pro-LPH precursor is also sorted almost exclusively to the apical membrane (38). The LPH is apparently not necessary for sorting pro-LPH since LPH expressed individually in MDCK cells is correctly sorted in a fashion similar to wild type pro-LPH (50). Deletion of 236 amino acids in the homologous region IV (45) of the ectodomain that is juxtaposed to the membrane generates a transport-competent and correctly sorted mutant protein (62). Importantly, the deletion of this stretch of 236 amino acids eliminates four potential N-glycosylation sites and the entire O-glycosylated domain of pro-LPH. Since these substantial changes in the glycosylation pattern of this mutant do not affect its sorting behavior, it is likely that neither N- nor O-linked glycosylation is directly responsible for the sorting of LPH. The cytosolic portion of LPH does not contain targeting signals, since a tailless LPH mutant is sorted to the apical membrane with similar fidelity to wild type LPH. Finally, the membrane anchoring domain of LPH does not have the characteristics of other transmembrane domains that contain apical sorting signals. LPH is completely soluble in Triton X-100 at 4 °C and is therefore not associated with lipid rafts (this study and Ref. 52). As there are other apically sorted proteins that could not be demonstrated to associate with detergent-insoluble membrane domains (52, 63), a separate mechanism for sorting these proteins into a parallel pathway to the apical surface is possible, or these proteins might associate with detergent-insoluble membranes with an affinity too weak to be observed experimentally.

Currently, we do not understand why the LPH has stronger apical sorting signals than does the HA. At least 90% of wild type LPH expressed in MDCK cells is delivered to the apical membrane compared with 75% apical sorting of HA expressed in continuous MDCK cell lines (11, 60). Since there are a number of different ways in which proteins can interact with apical sorting machinery, it is possible that apical sorting of LPH is regulated by several signals, which then act cooperatively to ensure strong association with the apical sorting machinery.

	ACKNOWLEDGEMENTS

We thank Dr. Martin Billeter, Institute of Molecular Biology, University of Zürich for critical review of the manuscript and for supporting Hussein Naim throughout this work. We thank Dr. Hans-Peter Hauri, Biozentrum, University of Basel, Dr. Erwin Sterchi, Institute of Biochemistry and Molecular Biology, University of Bern, and Dr. Dallas Swallow, Medical Research Council, London for generous gifts of monoclonal anti-LPH antibodies.

	FOOTNOTES

^* This work was supported by Grant Na 331/1-1 form the Deutsche Forschungsgemeinschaft (DFG), Bonn/Germany (to H. Y. N.), NATO Collaborative Research Grant CRG 931341 (to H. Y. N. and M. G. R.), and in part by Grant GM37547 from the National Institutes of Health (to M. G. R.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ The first two authors contributed equally to this work.

To whom correspondence should be addressed: Dept. of Physiological Chemistry, School of Veterinary Medicine Hannover, Bünteweg 17, D-30559 Hannover, Germany. Tel.: 49-511-9538780; Fax: 49-511-9538585; E-mail: hnaim{at}biochemie.tiho-hannover.de.

	ABBREVIATIONS

The abbreviations used are: HA, influenza virus hemagglutinin; LPH, lactase-phlorizin hydrolase; MDCK cells, Madin-Darby canine kidney cells; PAGE, polyacrylamide gel electrophoresis; mAb, monoclonal antibody; PBS, phosphate-buffered saline; DMEM, Dulbecco's modified Eagle's medium.

	REFERENCES

TOP ABSTRACT INTRODUCTION REFERENCES

1. Nelson, W. J. (1992) Science 258, 948-955[Abstract/Free Full Text]
2. Wandinger-Ness, A., Bennett, M. K., Antony, C., and Simons, K. (1990) J. Cell Biol. 111, 987-1000[Abstract/Free Full Text]
3. Fransen, J. A., Hauri, H. P., Ginsel, L. A., and Naim, H. Y. (1991) J. Cell Biol. 115, 45-57[Abstract/Free Full Text]
4. Moolenaar, C. E. C., Ouwendijk, J., Wittpoth, M., Wisselaar, H. A., Hauri, H. P., Ginsel, L A., Naim, H. Y., and Fransen, J. A. M. (1997) J. Cell Sci. 110, 557-567[Abstract]
5. Aroeti, B., Kosen, P. A., Kuntz, I. D., Cohen, F. E., and Mostov, K. E. (1993) J. Cell Biol. 123, 1149-1160[Abstract/Free Full Text]
6. Geffen, I., Fuhrer, C., Leitinger, B., Weiss, M., Huggel, K., Griffiths, G., and Spiess, M. (1993) J. Biol. Chem. 268, 20772-20777[Abstract/Free Full Text]
7. Honing, S., and Hunziker, W. (1995) J. Cell Biol. 128, 321-332[Abstract/Free Full Text]
8. Lin, S., Naim, H. Y., and Roth, M. G. (1997) J. Biol. Chem. 272, 26300-26305[Abstract/Free Full Text]
9. Odorizzi, G., and Trowbridge, I. S. (1997) J. Cell Biol. 137, 1255-1264[Abstract/Free Full Text]
10. Prill, V., Lehmann, L., von Figura, K., and Peters, C. (1993) EMBO J. 12, 2181-2193[Medline] [Order article via Infotrieve]
11. Thomas, D. C., Brewer, C. B., and Roth, M. G. (1993) J. Biol. Chem. 268, 3313-3320[Abstract/Free Full Text]
12. de Hoop, M., von Poser, C., Lange, C., Ikonen, E., Hunziker, W., and Dotti, C. G. (1995) J. Cell Biol. 130, 1447-1459[Abstract/Free Full Text]
13. Dotti, C. G., and Simons, K. (1990) Cell 62, 63-72[CrossRef][Medline] [Order article via Infotrieve]
14. Dotti, C. G., Parton, R. G., and Simons, K. (1991) Nature 349, 158-161[CrossRef][Medline] [Order article via Infotrieve]
15. Haass, C., Koo, E. H., Capell, A., Teplow, D. B., and Selkoe, D. J. (1995) J. Cell Biol. 128, 537-547[Abstract/Free Full Text]
16. Le Gall, A. H., Powell, S. K., Yeaman, C. A., and Rodriguez-Boulan, E. (1997) J. Biol. Chem. 272, 4559-4567[Abstract/Free Full Text]
17. Lisanti, M. P., Sargiacomo, M., Graeve, L., Saltiel, A., and Rodriguez-Boulan, E. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 9557-9561[Abstract/Free Full Text]
18. Brown, D. A., Crise, B., and Rose, J. K. (1989) Science 245, 1499-1501[Abstract/Free Full Text]
19. Le Bivic, A., Sambuy, Y., Mostov, K., and Rodriguez-Boulan, E. (1990) J. Cell Biol. 110, 1533-1539[Abstract/Free Full Text]
20. Matlin, K. S., and Simons, K. (1984) J. Cell Biol. 99, 2131-2139[Abstract/Free Full Text]
21. Simons, K., and Ikonen, E. (1997) Nature 387, 569-572[CrossRef][Medline] [Order article via Infotrieve]
22. Scheiffele, P., Peranen, J., and Simons, K. (1995) Nature 378, 96-98[CrossRef][Medline] [Order article via Infotrieve]
23. Soole, K. L., Hall, J., Jepson, M. A., Hazlewood, G. P., Gilbert, H. J., and Hirst, B. H. (1992) J. Cell Sci. 102, 495-504[Abstract/Free Full Text]
24. Gonzalez, A., Nicovani, S., and Juica, F. (1993) J. Biol. Chem. 268, 6662-6667[Abstract/Free Full Text]
25. Ragno, P., Estreicher, A., Gos, A., Wohlwend, A., Belin, D., and Vassalli, J. D. (1992) Exp. Cell Res. 203, 236-243[CrossRef][Medline] [Order article via Infotrieve]
26. Ullrich, O., Mann, K., Haase, W., and Koch-Brandt, C. (1991) J. Biol. Chem. 266, 3518-3525[Abstract/Free Full Text]
27. Mostov, K. E., de Bruyn Kops, A., and Deitcher, D. L. (1986) Cell 47, 359-364[CrossRef][Medline] [Order article via Infotrieve]
28. Hunziker, W., Harter, C., Matter, K., and Mellman, I. (1991) Cell 66, 907-920[CrossRef][Medline] [Order article via Infotrieve]
29. Hobert, M., and Carlin, C. (1995) J. Cell. Physiol. 162, 434-446[CrossRef][Medline] [Order article via Infotrieve]
30. Ball, J. M., Mulligan, M. J., and Compans, R. W. (1997) AIDS Res. Hum. Retroviruses 13, 665-675[Medline] [Order article via Infotrieve]
31. Thomas, D. C., and Roth, M. G. (1994) J. Biol. Chem. 269, 15732-15739[Abstract/Free Full Text]
32. Brewer, C. B., Thomas, D., and Roth, M. G. (1991) J. Cell Biol. 114, 413-421[Abstract/Free Full Text]
33. Danielsen, E. M., Skovbjerg, H., Noren, O., and Sjöström, H. (1984) Biochem. Biophys. Res. Commun. 122, 82-90[CrossRef][Medline] [Order article via Infotrieve]
34. Hauri, H. P., Sterchi, E. E., Bienz, D., Fransen, J. A. M., and Marxer, A. (1985) J. Cell Biol. 101, 838-851[Abstract/Free Full Text]
35. Naim, H. Y., Sterchi, E. E., and Lentze, M. J. (1987) Biochem. J. 241, 427-434[Medline] [Order article via Infotrieve]
36. Naim, H., Amarneh, B., Ktistakis, N. T., and Roth, M. G. (1992) J. Virol. 66, 7585-7588[Abstract/Free Full Text]
37. Lottaz, D., Oberholzer, T., Bähler, P., Semenza, G., and Sterchi, E. (1992) FEBS Lett. 313, 270-276[CrossRef][Medline] [Order article via Infotrieve]
38. Jacob, R., Brewer, C., Fransen, J. A. M., and Naim, H. Y. (1994) J. Biol. Chem. 269, 2712-2721[Abstract/Free Full Text]
39. Naim, H. Y., and Naim, H. (1996) Eur. J. Cell Biol. 70, 198-208[Medline] [Order article via Infotrieve]
40. Roth, M. G., Gething, M. J., Sambrook, J., Giusti, L., Davis, A., Nayak, D., and Compans, R. W. (1983) Cell 33, 435-443[CrossRef][Medline] [Order article via Infotrieve]
41. Lazarovits, J., and Roth, M. G. (1988) Cell 53, 743-752[CrossRef][Medline] [Order article via Infotrieve]
42. Naim, H., and Roth, M. G. (1994) J. Biol. Chem. 269, 3928-3933[Abstract/Free Full Text]
43. Naim, H., Dodds, D. T., Brewer, C. B., and Roth, M. G. (1995) J. Cell Biol. 129, 1241-1250[Abstract/Free Full Text]
44. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd Ed., Cold Spring Harbor Press, Cold Spring Harbor, NY
45. Mantei, N., Villa, M., Enzler, T., Wacker, H., Bol, W., James, P., Hunziker, W., and Semenza, G. (1988) EMBO J. 7, 2705-2713[Medline] [Order article via Infotrieve]
46. Horton, R. M., Cai, Z. L., Ho, S. N., and Pease, L. R. (1990) BioTechniques 8, 528-535[Medline] [Order article via Infotrieve]
47. Naim, H. Y., Lacey, S. W., Sambrook, J. F., and Gething, M. J. H. (1991) J. Biol. Chem. 266, 12313-12320[Abstract/Free Full Text]
48. Low, S. H., Wong, S. H., Tang, B. L., Subramaniam, N., and Hong, W. (1991) J. Biol. Chem. 266, 13391-13396[Abstract/Free Full Text]
49. Laemmli, U. K. (1970) Nature 227, 680-685[CrossRef][Medline] [Order article via Infotrieve]
50. Jacob, R., Radebach, I., Wüthrich, M., Grünberg, J., Sterchi, E. E., and Naim, H. Y. (1996) Eur. J. Biochem. 236, 789-795[Medline] [Order article via Infotrieve]
51. Rodriguez-Boulan, E., and Powell, S. K. (1992) Annu. Rev. Cell Biol. 8, 395-427[CrossRef]
52. Danielsen, E. M. (1995) Biochemistry 34, 1596-1605[CrossRef][Medline] [Order article via Infotrieve]
53. Naim, H. Y., Sterchi, E. E., and Lentze, M. J. (1988) J. Biol. Chem. 263, 7242-7253[Abstract/Free Full Text]
54. Ouwendijk, J., Peters, W. J. M., van de Vorstenbosch, R. A., Ginsel, L. A., Naim, H. Y., and Fransen, J. A. M. (1998) J. Biol. Chem. 273, 6650-6655[Abstract/Free Full Text]
55. Naim, H., and Roth, M. G. (1994) Methods Cell Biol. 43, 113-136
56. Yeaman, C., Le Gall, A. H., Baldwin, A. N., Monlauzeur, L., Le Bivic, A., and Rodriguez-Boulan, E. (1997) J. Cell Biol. 139, 929-940[Abstract/Free Full Text]
57. Lisanti, M. P., Caras, I. W., Davitz, M. A., and Rodriguez-Boulan, E. (1989) J. Cell Biol. 109, 2145-2156[Abstract/Free Full Text]
58. Alonso, M. A., Fan, L., and Alarcon, B. (1997) J. Biol. Chem. 272, 30748-30752[Abstract/Free Full Text]
59. Kundu, A., Avalos, R. T., Sanderson, C. M., and Nayak, D. P. (1996) J. Virol. 70, 6508-6515[Abstract]
60. Lin, S., Naim, H. Y., Rodreguez, A. C., and Roth, M. G. (1998) J. Cell Biol. 142, 51-57[Abstract/Free Full Text]
61. Huang, X. F., Compans, R. W., Chen, S., Lamb, R. A., and Arvan, P. (1997) J. Biol. Chem. 272, 27598-27604[Abstract/Free Full Text]
62. Panzer, P., Preuss, U., Joberty, G., and Naim, H. Y. (1998) J. Biol. Chem. 273, 13861-13869[Abstract/Free Full Text]
63. Tienari, P. J., De Strooper, B., Ikonen, E., Simons, M., Weidemann, A., Czech, C., Hartmann, T., Ida, N., Multhaup, G., Masters, C. L., Van Leuven, F., Beyreuther, K., and Dotti, C. G. (1996) EMBO J. 15, 5218-5229[Medline] [Order article via Infotrieve]