Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wice, B. M.
Right arrow Articles by Gordon, J. I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wice, B. M.
Right arrow Articles by Gordon, J. I.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Volume 270, Number 37, Issue of September 15, pp. 21907-21918, 1995
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
A Tetraspan Membrane Glycoprotein Produced in the Human Intestinal Epithelium and Liver That Can Regulate Cell Density-dependent Proliferation (*)

(Received for publication, May 23, 1995)

Burton M. Wice ,  Jeffrey I. Gordon (§)

From the Department of Molecular Biology and Pharmacology, Washington University School of Medicine, St. Louis, Missouri 63110

ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES

ABSTRACT

The human cell line HT-29 provides a model system for studying regulation of proliferation and differentiation in intestinal epithelial cell lineages: (i) HT-29 cells cultured in glucose resemble undifferentiated multipotent transit cells located in the lower half of intestinal crypts; (ii) proliferating HT-29 cells cultured in inosine resemble committed cells located in the upper half of the crypt; (iii) nonproliferating, confluent HT-29-inosine cells have features of differentiated enterocytes and goblet cells that overlie small intestinal villi. A cDNA library prepared from HT-29-inosine cells was screened with a series of subtracted cDNA probes to identify proteins that regulate proliferation/differentiation along the crypt-villus axis. A cDNA was recovered that encodes a 202-amino acid protein with four predicted membrane spanning domains and two potential sites for N-linked glycosylation. Levels of this new member of the superfamily of tetraspan membrane proteins (TMPs) increase dramatically as nondividing epithelial cells exit the proliferative compartment of the crypt-villus unit and migrate onto the villus. The protein is also produced in nondividing hepatocytes that have the greatest proliferative potential within liver acini. Three sets of observations indicate that in the appropriate cellular context, intestinal and liver (il)-TMP can mediate density-associated inhibition of proliferation. (i) Accumulation of il-TMP glycoforms precedes terminal differentiation of HT-29-inosine cells and occurs as they undergo density-dependent cessation of growth. il-TMP levels are lower and glycosylation less extensive in HT-29-glucose cells, which do not undergo growth arrest at confluence. (ii) HeLa cells normally do not produce il-TMP. Forced expression of il-TMP inhibits proliferation as cells approach confluence. The extent of il-TMP glycosylation in the transfected cells is similar to that observed in HT-29-inosine cells and greater than in HT-29-glucose cells. (iii) SW480 cells are derived from a human colon adenocarcinoma and do not express il-TMP. Like nontransfected HeLa cells, they do not stop dividing at confluence, whether grown in medium containing glucose or inosine. Expression of il-TMP has no effect on the growth properties of SW480 cells. The extent of il-TMP glycosylation in SW480-glucose cells is similar to that noted in HT-29-glucose cells, lending further support to the notion that il-TMP's activity is related to its state of N-glycosylation.

INTRODUCTION

The lifespan of an epithelial cell encompasses a series of decisions. The cell must decide when to proliferate, to commit to a specific cell lineage, to terminally differentiate, and/or to undergo a programmed death. In the small intestine, proliferation, differentiation, and death programs are expressed along a geographically well defined pathway that extends from the crypt of Lieberkühn to the villus' apical extrusion zone(1) . The mouse and human intestinal epithelium are continuously and rapidly renewed (2, 3) . Thus, the crypt-villus axis represents a model of perpetual development in mammals and provides an opportunity to examine the molecular mechanisms that regulate decision making in epithelial cells.

Most of what is known about intestinal epithelial renewal comes from studies of the mouse. The mouse small intestinal epithelium contains four principal terminally differentiated cell types. Absorptive enterocytes comprise >80% of its cells(4) . Members of the mucus-producing goblet and enteroendocrine cell lineages exhibit remarkable variations in their differentiation programs as a function of their location along the crypt-villus and duodenal-ileal axes (see, e.g., (5, 6, 7) ). Paneth cells secrete a variety of anti-microbial peptides, digestive enzymes, and growth factors(8) . These epithelial lineages arise from a multipotent stem cell located near the base of each crypt. Descendants of active multipotent stem cell(s) undergo several rounds of cell division in the mid-portion of each crypt(9) . Enterocytes, goblet, and enteroendocrine cells undergo terminal differentiation as they cease proliferating and migrate in vertical coherent bands from the crypt to the apex of a surrounding villus(10) . Cells are either phagocytosed or exfoliated into the intestinal lumen at the villus tip. Differentiation and removal is completed in 2-5 days depending upon the lineage and the location of crypt-villus units along the duodenal-to-ileal axis(2, 4, 11, 12, 13, 14) . Members of the Paneth cell lineage complete their differentiation program as they migrate downward to the base of each small intestine crypt where they reside for several weeks(15) .

Given the spatial complexities of this epithelium, most analyses of the regulation of its proliferation and differentiation programs have used in vivo models (see, e.g., Refs. 1, 16, and 17). However, in vitro models would be useful for initially identifying gene products that may regulate these processes. HT-29 cells are derived from a human colon adenocarcinoma (18) and represent one such model. If cultured in standard medium containing 25 mM glucose, they proliferate even after reaching confluence and do not produce proteins synthesized by terminally differentiated intestinal epithelial cells in vivo(19) . If HT-29 cells are cultured in the absence of glucose, using galactose (19) or inosine (20, 21) as the carbon source, they cease to proliferate once they reach confluence. If maintained in this confluent state for 10-14 days, most of the quiescent cells differentiate into enterocytes; they become polarized, form tight junctions, and elaborate an apical brush border membrane containing a variety of hydrolases(19, 21) . If glucose is added back to confluent cells that have been cultured for many generations in its absence, they will still terminally differentiate. This suggests that at some point during exponential growth in the absence of glucose, HT-29 cells make the decision to differentiate. Huet et al.(22) noted that 10% of differentiated HT-29 cells synthesize and secrete intestinal mucins, indicative of a goblet cell-like phenotype. Importantly, they found that a single clone of HT-29-glucose cells can give rise to progeny with enterocyte- and goblet-like phenotypes when switched to media containing carbon sources other than glucose. These results indicate that (i) HT-29 cells cultured in glucose have properties of undifferentiated multipotent transit cells located in the lower half of intestinal crypts, (ii) proliferating HT-29 cells cultured in inosine resemble committed cells located in the upper half of the crypt, and (iii) confluent HT-29-inosine cells have features of terminally differentiating villus-associated enterocytes and goblet cells. Thus, HT-29 cells cultured in media with different carbon sources and/or at different growth phases make decisions that may resemble decisions that occur along the (human) crypt-villus axis.

In this report we describe how subtracted cDNA probes were used to screen an HT-29-inosine cDNA library to identify an intestinal and liver tetraspan membrane protein (il-TMP). (^1)il-TMP can mediate density-dependent cell proliferation. Moreover, this function can be directly correlated with the extent of its N-glycosylation.

MATERIALS AND METHODS

Cell Culture

HT-29, SW480, and HeLa cells were cultured at 37 °C under an atmosphere of 95% air, 5% CO(2) in Dulbecco's modified Eagle's medium (DMEM; [glucose] = 25 mM) supplemented with 10% heat-inactivated fetal calf serum (FCS)(23) . ``Inosine'' cells were cultured for at least 20 generations in glucose-free DMEM supplemented with 2.5 mM inosine and 10% dialyzed FCS(21, 24) . Cells harvested during early to mid-log phase of growth were defined as ``proliferating.'' Confluence was defined as the point when cells first cover the entire bottom of a T-flask. Cells maintained in a confluent state for 10-14 days were designated ``postconfluent.''

Screening of a cDNA Library Prepared from Proliferating HT-29-inosine Cells

The protocols used for preparing, characterizing, and screening this library are described in (23) . Briefly, the library was prepared from poly(A) RNA isolated from proliferating HT-29-inosine cells. A P-labeled cDNA ``plus'' probe was generated from the same RNA preparation and subjected to subtractive hybridization with an excess of poly(A) RNA isolated from proliferating HeLa cells cultured in DMEM, 25 mM glucose. A ``minus'' cDNA probe was used to identify housekeeping sequences in the library. This probe was synthesized from proliferating HeLa-glucose cell RNA and subjected to subtractive hybridization with proliferating HT-29-inosine cell RNA. Phage that reacted with the ``plus'' probe AND failed to hybridize with the ``minus'' probe were characterized further.

Human Tissues

Tissues were obtained during elective surgical procedures and/or from adult organ donors according to guidelines and protocols approved by our University's Human Studies Committee. Some samples were obtained through the National Disease Research Interchange. All samples were snap-frozen and stored in liquid nitrogen.

To determine if the steady state concentration of il-TMP mRNA varies along the duodenal-colonic axis, segments of intestine were recovered at defined points along this axis from three organ donors (age = 32, 36, and 72 years). The entire bowel of each donor was sampled: i.e. duodenum, jejunum, ileum, and proximal, middle, and/or distal colon (n = 2 separate full thickness samples/segment/donor).

RNA Blot Hybridization Analyses

Total cellular RNA was extracted from cultured cells and from human tissues(23) , fractionated by formaldehyde-agarose gel electrophoresis, and transferred to nitrocellulose membranes. Blots were probed with P-labeled EcoRI/EcoRI il-TMP fragment or a 650-base pair NcoI/NcoI fragment containing the neomycin phosphotransferase gene from pControl/neo (see below). Hybridization and washing stringencies are described in (23) . Some blots contained a range of concentrations of purified in vitro transcribed il-TMP mRNA standards. Blots were scanned with a PhosphorImager (Molecular Dynamics). Only signals in the linear range of sensitivity were used for calculating il-TMP mRNA levels in samples of total cellular RNA.

Generation of Peptide-specific Antibodies

Two different peptides, representing a portion of each of the two putative extracellular domains of il-TMP, were produced using an Applied Biosystems model 430 synthesizer. Peptide A (GDYLNDEALWNKC) encompasses amino acids 134-146. Peptide B (GKVIDDNDHLSQEIC) corresponds to residues 33-46 and includes an additional C-terminal cysteine. Both peptides were conjugated to Keyhole Limpet hemocyanin via their C-terminal cysteine(25) . Each peptide-conjugate was used to immunize two New Zealand White rabbits. Each animal produced antibodies that recognize the corresponding unconjugated, immobilized peptide as determined by enzyme-linked immunosorbent assay. Antibodies were purified from serum by affinity chromatography using the ImmunoPure Ag/Ab Immobilization Kit 2 (Pierce).

Immunocytochemical Studies

Cells were grown on glass coverslips and washed in phosphate-buffered saline (PBS) before fixation. Frozen samples of human intestine or liver were embedded in O.C.T. compound (Miles), and 5-8-µm-thick sections were prepared. Cells or tissue sections were fixed in 100% methanol for 10 min at 0 °C and incubated with (i) affinity-purified rabbit il-TMP peptide-specific antibodies (final concentration = 0.5 µg/ml blocking buffer (blocking buffer = PBS, 1% (w/v) bovine serum albumin, 0.2% (w/v) nonfat powdered skim milk, 0.3% (v/v) Triton X-100) or (ii) mouse monoclonal antibody (mAb; HBB 3/775/42) raised against dipeptidylpeptidase IV (DPP-IV; (26) ; obtained from H.P. Hauri, University of Basel; ascites fluid diluted 1:100 in blocking buffer). Antigen-antibody complexes were visualized using indocarbocyanine (Cy3)-conjugated donkey anti-rabbit immunoglobulin (Ig) or fluorescein isothiocyanate (FITC)-conjugated donkey anti-mouse Ig (Jackson Immunoresearch Laboratories; diluted 1:1000 and 1:100, respectively). After removal of the secondary antibodies, nuclei were counterstained with bisbenzimine (Sigma).

Western Blot Analysis

Cultured cells were washed twice with ice-cold PBS and stored at -80 °C. Cells were thawed in RIPA buffer (PBS containing sodium deoxycholate (1%, w/v), Triton X-100 (1%, v/v), and SDS (0.1%, w/v)) supplemented with protease inhibitors (EDTA (10 mM), aprotinin (50 µg/ml; Sigma), leupeptin (50 µg/ml; Sigma), Pefabloc (Boehringer Mannheim; 500 µg/ml), and pepstatin A (10 µg/ml; Sigma)). Cellular DNA was sheared by passage through a 27-gauge needle. Insoluble material was removed by centrifugation for 5 min at 12,000 times g. SDS-polyacrylamide gel electrophoresis (SDS-PAGE) sample loading buffer was added to the lysates (final concentration of SDS = 2%, 2-mercaptoethanol = 5%; Tris-HCl = 50 mM, pH 6.8). For analysis of adult human tissue proteins, minced pieces of frozen tissue were lyophilized and then rehydrated in 20 ml of 2 times sample buffer plus protease inhibitors/mg dry weight tissue. The thawed, rehydrated tissue was then crushed with a glass rod, insoluble material was removed by centrifugation as above, and DNA was sheared by passage through a 27-gauge needle. To prevent aggregation of il-TMP, the final concentration of total cell protein in the sample loading buffer was kept below 300 µg/ml and the solution was not heated prior to SDS-PAGE(27) .

Separated proteins were transferred to polyvinylidene difluoride (PVDF) membranes (Amersham Corp.), and the membranes were probed with (i) affinity-purified, il-TMP peptide-specific antibodies (final concentration = 170 ng/ml of PBS containing gelatin (1%, w/v), Tween 20 (0.2%, v/v), and sodium azide (0.1%, w/v)), (ii) a mouse mAb directed against actin (C4; (28) ), or (iii) a mouse mAb raised against human sucrase-isomaltase (mAb HSI 9, a gift from Andrea Quaroni, Cornell University; cf. (29) ). il-TMP-antibody complexes were visualized with alkaline phosphatase-conjugated secondary antibodies using the Western Light(TM) kit (Tropix). Blots were stripped for reprobing with additional antibodies(30) .

Glycosidase Digestions of Postconfluent HT-29-inosine or -glucose Cell Lysates

Cleared cell lysates (3 mg of protein/ml of RIPA buffer) were diluted 10-fold in a solution containing 2% SDS, 5 mM dithiothreitol, plus the protease inhibitors listed above. The mixture was incubated for 10 min at room temperature. To sequester the SDS added for protein denaturation, the mixture was diluted 10-fold in glycosidase reaction buffer (see below) and incubated at 37 °C for 10 min before addition of glycosidase. Digestion with purified recombinant peptide N-glycosidase F (Boehringer Mannheim; 8 units/ml incubation) was performed in the presence of 20 mM sodium phosphate (pH 7.5), 5 mM dithiothreitol, 0.5% Triton X-100, plus the mixture of protease inhibitors. Digestion with purified recombinant Streptomyces plicatus endo-beta-N-acetylglucosaminidase H (endo H; Boehringer Mannheim; 5 milliunits/ml incubation) was accomplished using a similar buffer except that 10 mM sodium acetate (pH 5.5) was substituted for sodium phosphate. Enzymatic digestion was allowed to proceed for 2 h at 37 °C. Sample loading buffer was added, and the reaction products were fractionated by SDS-PAGE without prior heating. The separated proteins were transferred to PVDF membranes and processed as above.

Expression of il-TMP in Transfected HeLa and SW480 Cells

Construction of Dicistronic Expression Vectors

The vectors designed for these studies allow expression of two different gene products from a single mRNA transcript(31, 32) . An ``upstream'' reporter is synthesized following ribosome binding to the dicistronic mRNA's 5`-cap and subsequent scanning for the first Met codon with a favorable sequence for initiation of translation(33) . The ``downstream'' reporter is translated by ribosomes that bind to an internal ribosomal entry site (IRES) derived from the encephalomyocarditis viral genome(31) . All expression plasmids contained the EcoRI/XhoI fragment of pBluescript SK(+) plus a cytomegalovirus promoter (provided by John Majors, Washington University). pControl/neo was constructed by inserting two fragments into the vector: (i) an EcoRI/Rsr II fragment from pLZ1N(32) , which contains the encephalomyocarditis viral IRES plus the first portion of the neomycin phosphotransferase gene (Neo); and (ii) a Rsr II/SalI fragment that contains the remaining portion of Neo plus SV40 splice and polyadenylation sequences from pSV2neo(34) . An il-TMP expression vector, pil-TMP/neo, was constructed by inserting the EcoRI/EcoRI fragment of il-TMP DNA into the unique EcoRI site of pControl/neo.

Transfection and Selection

HeLa and SW480 cells were plated at a density of 1 times 10^5 in 60-mm tissue culture dishes (Costar) containing complete medium (DMEM/FCS) and refed the next day with 5 ml of fresh medium. After 1-2 h, calcium phosphate-DNA coprecipitates were added (10 µg plasmid DNA/dish). Cells were incubated for 6 h at 37 °C and then subjected to a 1-min shock with 15% dimethyl sulfoxide (prepared in HEPES-buffered saline; (25) ). Cells were washed with PBS, refed complete medium, incubated for another 36-48 h, and then (i) analyzed for il-TMP expression (defined in the text as ``transiently transfected cells'') or (ii) subjected to selection with G418 to obtain pools of stably transfected cells. Selection in G418 (Life Technologies, Inc.; 400 µg/ml complete medium) was allowed to proceed until cell death was no longer apparent (typically 2-3 weeks). Cells were trypsinized and replated every 2-3 days during G418 selection to minimize cell-cell contacts.

Analyzing the Effects of il-TMP on Growth of Stably Transfected Pools of HeLa or SW480 Cells

Stably transfected cells were plated at a density of 9 times 10^4 cells/25-cm^2 T-flask in DMEM, 10% FCS, G418. Cells were refed daily with fresh medium. Cells contained in separate T-flasks were washed with PBS 1, 2, 3, 4, 5, 6, 8, 10, 12, and 14 days after plating, and lysed in RIPA buffer. Lysates were analyzed for total protein concentration according to Peterson(35) . Duplicate flasks of pControl/neo- or pil-TMP/neo-transfected cells were analyzed at each time point in each of three independent transfection experiments.

Focus formation was assessed in parallel cultures of pControl/neo- and pil-TMP/neo-transfected HeLa cells. After 12-14 days of culture, cells were washed twice with ice-cold PBS and stained for 15 min with a solution of 25% isopropanol, 10% acetic acid, 0.025% Coomassie Blue. Flasks were rinsed with 25% isopropanol, 10% acetic acid, allowed to air-dry, and viewed under a dissecting microscope. For each independently transfected pool of cells, focus formation was surveyed in two to three separate experiments.

Immunoprecipitation of il-TMP from HT-29 and Stably Transfected HeLa or SW480 Cells

Cells were lysed at 0 °C in PBS containing 0.1% SDS, 5 mM dithiothreitol, and the protease inhibitors listed above. Insoluble material was removed by centrifugation at 12,000 times g for 5 min. The resulting supernatant was incubated for 10 min at room temperature to reveal peptide epitopes recognized by the il-TMP antibodies. To sequester the SDS used for this denaturation step, a solution of PBS, 20% Triton X-100, 20% sodium deoxycholate was added to the lysate (1 ml/20 ml of lysate) and the mixture incubated for another 10 min at room temperature. Affinity-purified antibodies raised against each of the two il-TMP peptides were then added (2.5 µg of a given antibody preparation/50 µg of total cellular protein) and the solution was incubated overnight at 4 °C. Protein A-Sepharose (Pharmacia Biotech Inc.) was introduced (50 µg/µg antibody) and the suspension incubated for 1 h at 4 °C. The Protein A-Sepharose, containing bound antibodies and il-TMP, was harvested by centrifugation and washed three times with RIPA buffer and once with PBS. Following addition of sample loading buffer, the suspension was incubated for 10 min at 37 °C and then subjected to SDS-PAGE. Immunoprecipitated proteins were transferred to PVDF membranes and il-TMP detected using the Western-Light(TM) immunoblotting kit.

RESULTS

il-TMP mRNA Is Detectable in Adult Human Small Intestine and Liver

As noted in the Introduction, proliferating HT-29-inosine cells are committed to differentiate after they reach confluence. We reasoned that a cDNA library prepared from these cells encodes proteins that participate in regulating proliferation, allocation to the enterocytic and goblet cell lineages, and/or the differentiation programs of these cells. When this library was screened with cDNA probes that had been subtracted using a protocol described in an earlier report(23) , 11 recombinant phage were recovered from 100,000 screened. These phage contained four unique cDNA inserts(23) . One of the cDNAs reacted with 1.0- and 1.4-kilobase mRNAs present in proliferating HT-29-inosine but not HeLa cells, thereby satisfying the initial selection criteria for the library screen. The two mRNA transcripts are also present in total cellular RNA isolated from adult human jejunum and liver. They are not detectable in adult human stomach, kidney, lung, skeletal muscle, heart, or placental RNAs or in RNAs prepared from human fibroblasts, peripheral blood macrophages, and splenic mononuclear cells (with or without mitogenic stimulation) (data not shown).

Surveys of different regions of the entire intestine from each of three adult human organ donors revealed a distinct duodenal-colonic gradient in the concentration of the two transcripts (Fig. 1). Highest levels were noted in the jejunum (7 pg/µg of total cellular RNA). Concentrations in the duodenum, ileum, and colon were 2-4-, 5-10-, and 10-40-fold lower, respectively, than in jejunum. The steady state level of the transcripts in jejunum was 10-20-fold higher than in liver.

Figure 1: il-TMP mRNA levels vary along the cephalocaudal axis of adult human intestine. RNA blots containing 5 µg of total cellular RNA/lane were probed with a full-length P-labeled il-TMP cDNA. All samples were derived from different intestinal segments from a single adult human organ donor. Note that the highest steady state levels of il-TMP mRNA are encountered in the jejunum. Similar results are observed with RNA samples prepared from two other organ donors.


Southern blots of human genomic DNA, digested to completion with EcoRI, BamHI, or HindIII, indicated that the cDNA was derived from a single copy gene (data not shown).

Fig. 2A presents the nucleotide sequence of the 1385-base pair cDNA. A polyadenylation signal (ATTAAA) is located 27 nucleotides upstream of its poly(A) tail. A second polyadenylation signal (AATAAA) spans nucleotides 953-958 and is likely to account for the shorter mRNA species observed in human intestine, liver, and cultured HT-29 cells (see below). The cDNA encodes a primary translation product of 202 amino acids with a calculated mass of 21,396 Da. There is no site near the N terminus likely to undergo cotranslational cleavage by signal peptidase(36) . Cysteine and glycine residues are clustered in three regions. The hexapeptide Cys-Cys-Gly-Cys-Cys-Gly appears in two of these regions (residues 74-79 and 192-197 in Fig. 2A).

Figure 2: Nucleotide and deduced amino acid sequences of il-TMP. Panel A, nucleotide and amino acid numbers are indicated to the left and right, respectively, of the sequences. Two potential polyadenylation signals are boxed. Two potential sites for N-linked glycosylation are circled. Cys-Cys-Gly motifs are underlinedtwice. The peptide sequences used for generation of polyclonal antibodies are underlinedonce. PanelB, Kyte-Doolittle hydropathy profile of il-TMP. Negative values indicate hydrophilic regions, while positive values indicate hydrophobic areas. Note the presence of four hydrophobic domains (graylines) and the presence of two predicted extracellular domains (boldblacklines). The peptide sequences used to raise polyclonal antibodies against il-TMP are indicated by the filledboxes within the blacklines labeled A and B. The upwardpointingarrows show the location of potential sites for N-linked glycosylation of il-TMP.


A hydropathy plot (37) predicts that the protein contains four hydrophobic domains, each sufficiently long to span cellular membranes (Fig. 2B). The hydrophilic region between the third and fourth hydrophobic domains has two potential sites for N-linked glycosylation (Asn and Asn; Fig. 2, A and B).

A search of nonredundant protein data bases using the BLAST network (38) revealed that the protein is 50% identical to L6, an antigen expressed on the surface of colon, lung, breast, and ovarian carcinoma cells(39, 40) . The protein also shows significant homology to members of a superfamily of tetraspan membrane proteins (TMPs; (41) ). Most TMPs are rich in cysteines and glycines, contain at least two copies of a Cys-Cys-Gly motif, and have four transmembrane domains. TMPs also contain a sequence of variable length and hydrophilicity positioned between the third and fourth hydrophobic domains that typically has sites for N-linked glycosylation. Based on these sequence comparisons, we named the 202 residue polypeptide intestinal and liver tetraspan membrane protein or il-TMP.

il-TMP Is Expressed at Highest Levels in Nonproliferating Villus-associated Epithelial Cells and Peri-portal Hepatocytes

To examine the relationship between il-TMP accumulation and the proliferative potential of intestinal and hepatic cell lineages, polyclonal antibodies were generated in rabbits against peptides corresponding to portions of il-TMP's two putative extracellular domains (Fig. 2B). Two rabbits were immunized with each peptide. Control Western blots of postconfluent HT-29-inosine cell lysates established that each of the four affinity-purified antibody preparations recognizes a protein of 21.5 kDa, corresponding to the calculated mass of the primary translation product of il-TMP mRNA. Each antibody preparation also reacts with 25-40-kDa N-linked glycoforms of il-TMP (see below). The pattern of reactivity was similar with all antibodies and was blocked by preincubation with the appropriate peptide (data not shown).

Frozen sections of adult human jejunum were incubated with each of the four antibody preparations. There is a dramatic increase in il-TMP levels at the crypt-villus junction (Fig. 3, A and B). High levels are maintained as nonproliferating, differentiated epithelial cells complete their migration to the villus tip (Fig. 3A). Less intense staining is present in the crypt (Fig. 3, A and B). Staining is limited in all cases to the apical borders of epithelial cells. No staining was detected with preimmune sera (data not shown) or if the antibodies were preincubated with the appropriate peptide (e.g.Fig. 3C). None of the antibodies reacted with any mesenchymal cell populations (Fig. 3, A and B). These findings suggest that if il-TMP affects proliferative status in crypt-villus units, it does so as a negative regulator.

Figure 3: Cellular distribution of il-TMP in adult human intestine and liver. PanelA, frozen section of adult human jejunum incubated with antisera raised against residues 134-146 of il-TMP (final dilution = 1:2000). Antigen-antibody complexes were visualized with Cy3-conjugated donkey anti-rabbit Ig. The arrow points to a crypt/villus junction. PanelB, higher power view of a crypt-villus junction located in the jejunum. The section was processed as in panelA. Note that the concentration of il-TMP in the apical membranes of crypt epithelial cells is lower than in the apical membranes of villus-associated enterocytes. PanelC, frozen section prepared from the same sample of human jejunum as used for panelsA and B. The primary antisera was preincubated with its peptide antigen (GDYLNDEALWNKC) (blocking control). PanelD, frozen section of adult human liver stained with affinity-purified antibodies raised against residues 134-146 of il-TMP followed by Cy3-conjugated donkey anti-rabbit Ig. Nuclei (blue) were revealed by counterstaining with bisbenzimine. The closedarrow points to immunoreactive il-TMP associated with the canalicular membranes of hepatocytes. The openarrow points to a cluster of il-TMP-positive cells whose identity has not been established. Shorter photographic exposure times revealed that il-TMP is concentrated in the plasma membrane of these cells. PanelE, frozen section of liver incubated with an affinity-purified preparation of rabbit antibodies to residues 134-146 of il-TMP and a mouse monoclonal antibody raised against human DPP-IV. Antigen-antibody complexes were visualized with Cy3-conjugated donkey anti-rabbit Ig and FITC-conjugated donkey anti-mouse Ig. Arrows point to the il-TMP-positive (orange) canalicular membrane of hepatocytes located in zone 1 of the acinus (zone 1 = periportal area). The arrowhead points to representative DPP-IV-positive hepatocytes located in zones 2 and 3 of the acinus. Immunoreactive DPP-IV (green) is associated with the cannicular membrane. PanelF, higher power view of the section shown in panelE. Note that DPP-IV (green) but not il-TMP (orange) is present in the apical border of bile duct epithelial cells (closedarrow; cf. Ref 42). il-TMP but not DPP-IV is expressed in the unidentified population of cells (e.g.openarrow). Bars = 25 µm.


Frozen sections were also prepared from intestinal segments harvested at various positions along the duodenal-ileal axis of a single organ donor. The cellular distribution and levels of il-TMP are similar in duodenal and jejunal crypt-villus units. The apical border of epithelial cells are also stained in ileal villi, but the intensity of staining is lower (data not shown). These cephalocaudal differences are consistent with the observed regional variations in steady state il-TMP mRNA concentrations (Fig. 1).

The structural and functional unit of the liver is the acinus. Each acinus can be divided into three zones. Zone 1 is defined as the periportal region. Zone 3 surrounds the central vein. Zone 2 is positioned between zones 1 and 3. Tritiated thymidine labeling studies suggest that stem cells located in zone 1 give rise to populations of hepatocytes that terminally differentiate as they slowly migrate toward the central vein(43) . Incubation of frozen sections of adult human liver with each of the four il-TMP antibody preparations revealed a ``chicken wire'' pattern of staining in zone 1 (Fig. 3, D-F). Blocking controls and studies with preimmune sera established the specificity of this reaction (data not shown). The chicken wire pattern of il-TMP staining is similar to that observed with DPP-IV, a protein known to be targeted to the canalicular membrane of ``mature'' hepatocytes positioned in zones 2 and 3 (Fig. 3E; cf.(42) and (44) ). There is an inverse relationship in the expression of these two proteins along the zone 1-3 axis: multilabel confocal microscopic studies revealed that il-TMP and DPP-IV are only coexpressed in a narrow band of hepatocytes positioned between zones 1 and 2 and that il-TMP is not present in zones 2 and 3 (data not shown). The fact that il-TMP is only expressed in the nondividing population of hepatocytes with highest proliferative potential is consistent with the notion that this protein could play a role in suppressing proliferation in the acinus as well as in crypt-villus units.

il-TMP Is Induced in HT-29-inosine Cells during the Period when They Undergo Cell Contact-associated Inhibition of Proliferation

The time course of il-TMP expression was evaluated in HT-29-inosine cells as a function of their proliferative status, density, and state of differentiation. Sucrase-isomaltase, alkaline phosphatase, aminopeptidase N, and dipeptidylpeptidase IV are apical brush border-associated hydrolases whose expression is induced several days after HT-29-inosine cells reach confluence. Levels gradually increase over the course of the next 7-14 days(19, 20, 21) . We used sucrase-isomaltase as a representative maker of terminally differentiated cells. Sucrase-isomaltase is first detected 9 days after plating. Steady state concentrations increase up to day 16, after which time they remain elevated (Fig. 4B). Actin concentrations remain constant up to 23 days after plating (Fig. 4B). The 21.5-kDa form of il-TMP is detectable 2 days after plating. Its concentration does not change during the log phase of growth (days 2-5), increases modestly (2-fold) as cells begin to undergo contact-associated inhibition of growth (days 6-9), and then stays constant in quiescent cells (days 11-23). The changes in concentration of the 25-40-kDa immunoreactive species parallel those of the 21.5-kDa form, although the magnitude of their increase during days 6-9 is considerably greater (Fig. 4B). These 25-40-kDa proteins represent N-linked il-TMP glycoforms, based on their susceptibility to cleavage by peptide N-glycosidase F (Fig. 5). Thus, a dramatic increase in il-TMP glycoforms occurs as HT-29-inosine cells undergo growth arrest and well before these cells produce detectable levels of sucrase-isomaltase.

Figure 4: il-TMP expression precedes differentiation in HT-29-inosine cells. PanelA, HT-29-inosine or HT-29-glucose cells were plated in 25-cm^2 T-flasks containing glucose-free DMEM supplemented with 2.5 mM inosine or DMEM containing 25 mM glucose, respectively. At the indicated times after plating, cells were recovered, lysed, and aliquots of the lysate assayed for protein content. The arrows point to the time when cells first become 100% confluent. The shadedboxes encompass the period of increasing steady state levels of il-TMP glycoforms. PanelsB and C, aliquots containing 5 µg of cellular protein were also analyzed by Western blotting. Triplicate protein blots were probed with affinity-purified rabbit antibodies to residues 134-146 of il-TMP, a mouse mAb raised against human sucrase-isomaltase, or a mouse mAb raised against actin. Bound antibodies were visualized using alkaline phosphatase-conjugated goat anti-rabbit or goat anti-mouse Ig. The arrows point to the day when cells first become 100% confluent. Note that although the steady-state levels of il-TMP glycoforms increase in postconfluent cells, longer exposures of the il-TMP blots indicate that the electrophoretic mobilities of the glycoforms do not change during the course of the experiment.


Figure 5: Glycosylation of il-TMP is different in HT-29-inosine and HT-glucose 29 cells. Lysates were prepared from confluent HT-29-inosine (Ino) or HT-29-glucose (Glc) cells and then incubated with (+) or without(-) peptide N-glycosidase F or endo H. Aliquots of the glycosidase reaction containing 0.6 µg of cellular protein were fractionated by SDS-PAGE and transferred to nitrocellulose membranes. The protein blots were probed with affinity-purified antibodies raised against residues 134-146 of il-TMP. Antigen-antibody complexes were detected with alkaline phosphatase-conjugated secondary antibodies.


The Pattern of il-TMP Expression in HT-29-glucose Cells Differs from That in HT-29-inosine Cells

HT-29-glucose cells proliferate more rapidly during log phase than HT-29-inosine cells (doubling time = 31 h compared to 41 h). Unlike HT-29-inosine cells, HT-29-glucose cells do not undergo growth arrest after achieving confluence (Fig. 4A). In addition, they do not express detectable amounts of sucrase-isomaltase either prior to, or after, reaching confluence, reflecting their failure to differentiate (Fig. 4C). In the representative experiment shown in Fig. 4(A and C), HT-29-glucose cells reached 100% confluence 4 days after plating. Cell density continued to increase as rapidly dividing cells piled up into multilayers (days 6-10). The increase in cell mass was sustained, albeit at a much slower rate, from day 10 to day 23 (Fig. 4A). There are three obvious differences in il-TMP expression in HT-29-glucose compared to HT-29-inosine cells. First, il-TMP is not induced in HT-29-glucose cells until 5 days after they have already piled up into a multilayer. Second, the level of all il-TMP species (nonglycosylated plus glycosylated forms) is lower in HT-29-glucose compared to HT-29-inosine cells at each time point surveyed (compare days 2-23 in Fig. 4, B and C). Third, the il-TMP glycoforms produced in confluent proliferating HT-29-glucose cells are quite distinct from the il-TMP glycoforms synthesized by confluent, nonproliferating HT-29-inosine cells (25-30 kDa versus 25-40 kDa; cf. Fig. 4(B and C) and 5).

il-TMP Is Less Extensively Glycosylated in HT-29-glucose Compared to HT-29-inosine and Nonproliferating Villus-associated Epithelial Cells

il-TMP from both HT-29-glucose and HT-29-inosine cells is sensitive to peptide N-glycosidase F treatment (Fig. 5), indicating that both synthesize the protein with N-linked sugars. Glycosylation is known to be altered in HT-29-glucose compared to HT-29-inosine cells due to impaired processing of high mannose forms of N-linked oligosaccharides(20, 45) . Therefore, HT-29-glucose and -inosine cell lysates were incubated with endo H, which only hydrolyzes N-linked high mannose and some hybrid oligosaccharides. The il-TMP present in postconfluent HT-29-glucose and HT-29-inosine cells is not digested by this glycosidase (Fig. 5). When the same blot was reprobed with Galantus nivalis agglutinin (specificity = Manalpha3Man epitopes; (46) ), we observed a shift in the mobilities of reactive glycoproteins and/or their disappearance (data not shown), indicating that the endo H was active.

Since most of the il-TMP in the small intestine is present in non-dividing villus-associated enterocytes, Western blot analysis of lysates prepared from full thickness samples of human small intestine should reflect the state of il-TMP glycosylation in these cells. Protein blots revealed that il-TMP levels along the cephalocaudal axis mimic the variations noted during our immunocytochemical and RNA analyses. The size distribution of il-TMP glycoforms in human duodenal, jejunal, and ileal extracts indicates that the protein is more extensively glycosylated than in HT-29-glucose cells (Fig. 6).

Figure 6: Glycosylation of il-TMP in villus-associated epithelial cells is similar to that in HT-29-inosine cells. Lysates were prepared from HT-29 cells cultured in glucose (Glc) or inosine (Ino) and from full thickness segments of adult human small intestine. Western blots of cellular proteins (4 µg/lane) were probed with affinity-purified antibodies to residues 134-146 of il-TMP.


Together these findings suggest that (i) il-TMP oligosaccharides do not accumulate as high mannose forms in either HT-29-glucose or -inosine cells, (ii) glycosylation in the proliferating glucose-treated cells is probably impaired at a step distal to the activity of mannosidase II, and (iii) glycosylation may influence il-TMP's ability to affect proliferation as cells approach confluence.

il-TMP Is Targeted to the Plasma Membrane in Both HT-29-inosine and HT-29-glucose Cells

il-TMP is present in highest concentrations at the plasma membrane of proliferating HT-29-inosine cells, postconfluent differentiated HT-29-inosine cells, and postconfluent undifferentiated, proliferating HT-29-glucose cells (Fig. 7, A-C). Thus, impaired glycosylation of il-TMP in HT-29-glucose cells does not appear to impede its transport to the cell surface. Moreover, any differences in the ability of il-TMP to affect proliferation in HT-29-glucose and HT-29-inosine cells cannot be simply ascribed to differences in its targeting to the plasma membrane.

Figure 7: Expression of il-TMP in cultured cell lines. PanelA, proliferating HT-29-inosine cells were harvested during log phase and stained with affinity-purified polyclonal antibodies to residues 134-146 of il-TMP plus FITC-conjugated donkey anti-rabbit Ig. Note the intense staining of the cell periphery as well as the punctate pattern of staining of the plasma membrane of these polarized cells. PanelB, postconfluent, differentiated HT-29 inosine cells fixed and stained with the same il-TMP antibody used in panelA plus Cy3-conjugated donkey anti-rabbit Ig. Note the evenly distributed, punctate staining of the entire cell surface. PanelC, postconfluent HT-29-glucose cells fixed and stained as in panelB. Note thesimilar patterns of staining in these nonpolarized cells and in the proliferating HT-29 inosine cells shown in panelA. PanelD, HeLa cells were transfected with pil-TMP/neo. Three days later, cells were fixed in methanol and stained for il-TMP using affinity-purified antibodies to residues 134-146 of il-TMP and Cy3-conjugated donkey anti-rabbit Ig (red). Nuclei were counterstained with bisbenzimine (blue). Note the high levels of il-TMP expression in a subset of cells. Bars = 25 µm. PanelsE and F, Coomassie-stained flasks containing a pool of HeLa cells stably transfected with pControl/neo (panelE) or pil-TMP/neo (panelF). Cells were stained 14 days after plating in glucose-containing medium and viewed under the dissecting microscope. The arrow in each panel points to a focus of cells. Comparison of panelsE and F reveals that focus number and size is greater in cells containing the il-TMP expression vector. Note that the magnification factor is 20-fold lower in panelsE and F compared to panelD.


Forced Expression of il-TMP Produces Cell Density-related Inhibition of HeLa Cell Proliferation

il-TMP was expressed in two established human epithelial cell lines that normally do not produce this protein so that we could determine directly whether it mediates cell density-related changes in proliferation. The vectors used for these studies permit production of two different gene products from a single dicistronic mRNA. One vector, pil-TMP/neo, contained il-TMP as the first open reading frame (ORF), followed by an IRES and a neomycin phosphotransferase ORF. The second vector, pControl/neo, contained IRES-neomycin phosphotransferase without the upstream il-TMP ORF.

We first isolated pools of transiently transfected HeLa cells to determine whether the vector could direct production il-TMP. Cells were fixed in methanol 72 h after transfection and stained for il-TMP with the affinity-purified antibody preparations. As expected, cells transfected with pControl/neo did not express the protein. Cultures transfected with pil-TMP/neo contained a subpopulation of cells with readily detectable amounts of il-TMP (Fig. 7D).

HeLa cells were then stably transfected with pil-TMP/neo or pControl/neo. We were concerned that if il-TMP plays a role in promoting cell contact-associated inhibition of growth, the selection for neomycin resistance would favor amplification of cells that either produced less il-TMP or were less sensitive to its putative growth inhibitory effects. Therefore, HeLa cells were maintained in a state where cell-cell contacts were minimized throughout the course of G418 selection, i.e. they were trypsinized and replated every 2-3 days until stably transfected pools were obtained. The data presented in Fig. 8(A and B) confirm transcription of integrated plasmid DNA in two independently derived pools of pControl/neo and pil-TMP/neo-HeLa cells. Three mRNA species are present (Fig. 8A). The 3.0-kilobase transcript encodes both il-TMP and neomycin phosphotransferase. The 1.0- and 1.4-kilobase transcripts presumably arise from use of either one of the two polyadenylation signals present in the 3`-untranslated region of il-TMP mRNA (Fig. 8C). The relative concentrations of the three species are similar in both cellular pools.

Figure 8: il-TMP mRNA levels in stably transfected HeLa cells. PanelsA and B, Northern blots containing RNAs prepared from two independently derived stably transfected pools of HeLa cells (10 µg of total cellular RNA/lane). Duplicate blots were probed with a full-length P-labeled il-TMP cDNA (panelA) or the neomycin phosphotransferase gene (Neo; panelB). PanelC, explanation of the origins of the multiple transcripts observed in pil-TMP/neo-HeLa cells shown in panelA. The structure of the expression vector is shown. The upwardpointingarrows indicate the positions of the two polyadenylation signals present in il-TMP. When either is utilized, the IRES/Neo portion of the RNA transcript is eliminated.


The growth rates of stably transfected pControl/neo- and pil-TMP/neo-HeLa cells were essentially identical at low density in medium containing glucose and G418 (doubling time = 26 h; Fig. 9A). pControl/neo cells achieve confluence 8 days after plating. They then continue to divide and form numerous multilayered foci (e.g.Fig. 7F). pil-TMP/neo cells do not reach confluence, even 12 days after plating. On day 12, virtually all pil-TMP/neo HeLa cells were distributed in clusters containing a single layer of cells rather than in multilayered foci (e.g.Fig. 7E) (n = 3 independent transfection experiments). These differences in cell number are reflected by differences in total cellular protein/T-flask/time point surveyed (Fig. 9A). Parallel measurements of total cellular protein/flask, DNA content/flask, or cell number/flask provide similar results when assessing HeLa cell growth(24) .

Figure 9: Glycoforms of il-TMP produced in transfected HeLa and SW480 cells. PanelA, stably transfected pools of HeLa cells containing pil-TMP/neo or pControl/neo were plated in 25 cm^2 T-flasks containing DMEM (25 mM glucose). All attached cells were harvested at the indicated times from each T-flask, and the amount of total cellular protein/flask was determined. Two flasks were assayed/time point. PanelB, pools of HeLa or SW480 cells stably transfected with pil-TMP/neo or pControl/neo were grown in DMEM (25 mM glucose). The transfected cells were harvested during late log phase (80% confluent). HT-29-inosine (Ino) or HT-29-glucose (Glc) cells were harvested during late log phase (not shown) or 10 days after reaching confluence and used as controls. Cells were lysed and il-TMP was immunoprecipitated from 60 µg HT-29-inosine and HT-29-glucose cell protein, 800 µg of HeLa cell protein, and 1200 µg of SW480 cell protein. Immunopurified proteins were fractionated by SDS/PAGE, transferred to PVDF membranes, and the protein blots probed with affinity-purified antibodies raised against residues 134-146 of il-TMP. Bound antibodies were detected using alkaline phosphatase-conjugated goat anti-rabbit Ig. The il-TMP glycoforms present in pil-TMP/neo-HeLa cells have a migration profile similar to those present in (i) nonproliferating postconfluent HT-29-inosine cells (lane1) and (ii) proliferating preconfluent HT-29-inosine cells (data not shown). The glycoforms present in pil-TMP/neo-SW480 cells have a migration profile similar to those produced in pre- and postconfluent HT-29-glucose cells (data not shown and lane4, respectively).


There is a 3-fold reduction in cell number produced by il-TMP 12 days after plating (Fig. 9A). This is likely to be an underestimate of the magnitude of the protein's effect on proliferation as cells reach confluence. If il-TMP functions to inhibit growth, any population of cells in the pool that produces low levels of il-TMP would have a proliferative advantage over a population that expresses relatively higher levels. The result would be a progressive increase in the fractional representation of the latter population and a ``masking'' of il-TMP's growth inhibitory properties. In addition, foci of pControl/neo HeLa cells detach from the T-flask and remain suspended in the medium. This fraction is not scored in our assay of cell density, which only considers cells that remain attached after refeeding.

The Extent of il-TMP Glycosylation Is Similar in HT-29-inosine and Transfected HeLa Cells

Immunoprecipitable il-TMP produced by stably transfected HeLa cells during the late log phase of growth in medium containing glucose and G418 has a pattern of migration during SDS-PAGE similar to that observed in postconfluent, nonproliferating HT-29-inosine cells (Fig. 9B) or in villus-associated epithelial cells (Fig. 6) and different from the pattern seen in postconfluent HT-29-glucose cells (Fig. 9B).

il-TMP Fails to Suppress Growth of Confluent SW480 Cells

The SW480 cell line was derived from a human colon adenocarcinoma(47) . il-TMP transcripts are not detectable in RNA prepared from SW480 cells, whether they are grown in the presence of glucose or inosine (data not shown). Unlike HT-29-inosine cells, SW480 cells cannot be induced to differentiate(48) . Like HT-29-glucose and HeLa cells, they proliferate even after achieving confluence. Pools of stably transfected SW480 cells were isolated using repeated trypsinization during G418 selection. Cells containing pControl/neo or pil-TMP/neo had similar growth rates after plating in glucose or inosine medium and continued to proliferate after reaching confluence (data not shown).

The Extent of il-TMP Glycosylation Is Similar in HT-29-glucose and Transfected SW480 Cells

The il-TMP glycoforms in SW480 cells have mobilities similar to the mobilities of il-TMP glycoforms in HT-29-glucose cells (Fig. 9B, compare lanes3 and 4). These glycoforms are distinct from those in pil-TMP/neo-HeLa cells cultured in the same medium (DMEM, 25 mM glucose) or in HT-29-inosine cells (Fig. 9B, lanes1 and 2). The pattern of il-TMP glycosylation in SW480 cells is consistent with the finding of Ogier-Denis et al.(45) that ``global'' N-linked glycosylation in this line resembles HT-29-glucose rather than HT-29-inosine cells.

DISCUSSION

Members of the TMP Superfamily Have Been Implicated in Regulating Proliferation and Adhesion

There are now 17 known members of the tetraspan membrane protein superfamily (reviewed in Refs. 41 and 49; also see (50) ). Alignments reveal 20-50% amino acid sequence identity between il-TMP and members of the superfamily (data not shown). As noted above, TMPs share several common structural features including four predicted transmembrane domains, the presence of several conserved cysteine-containing motifs, and sites for N-glycosylation in the extracellular domain located between the third and fourth transmembrane domains. Eleven TMPs have been shown to be N-glycosylated (e.g.(50) and (51) ).

Our results with transfected HeLa cells establish that il-TMP can participate in cell density-related inhibition of proliferation. A number of TMPs have been implicated in the regulation of proliferation. TAPA-1 is a cell surface protein expressed in many human cell lines (52) . Incubation of several lymphoma-derived lines with TAPA-1 mAbs inhibits their proliferation(52) . Antibodies to CD37, a glycoprotein produced by human B-lymphocytes, can have a positive or negative effect on proliferation, depending upon how the B-cells are stimulated(53) . mAbs to CD82 can inhibit the mitogenic activation of peripheral B-lymphocytes (54) or act synergistically with mAbs to CD3 to stimulate T-cell proliferation(55) . OX-44 (the rat ortholog of the human leukocyte TMP known as CD53) identifies the subset of thymocytes that proliferate in response to alloantigens and lectins(56) . mAbs to OX-44 induce proliferation of T-lymphocytes(57) . TI-1 is a transforming growth factor-beta-inducible TMP produced in mink lung epithelial cells whose expression changes with their state of proliferation(58) . CD9/DRAP 27, an integral membrane protein produced by a variety of epithelial and hematopoietic cell lineages, potentiates the juxtacrine growth factor activity of the membrane-anchored, heparin-binding epidermal growth factor-like precursor/diphtheria toxin receptor(59) .

Some TMPs also appear to be involved in mediating cell-cell and/or membrane-membrane adhesion. A CD9 mAb induces adhesion(60) . OX-44 associates with the cell adhesion receptor, CD2(57) . TAPA-1 mAbs induce cell-cell adhesion(61, 62, 63) . CD82 (KAI1) can suppress metastasis of human prostate cancer cells(64) . RDS(65) , also known as peripherin (66) , is located in the outer segment disc membranes of retinal photoreceptor cells. In mice, a truncation mutant of RDS produces abnormal development of the outer segment discs followed by slow degeneration of the photoreceptor and blindness (retinal degeneration slow; rds). Studies in transgenic animals indicate that this disease can be corrected by expressing the normal protein in rds photoreceptors(67) . Travis et al.(68) proposed that RDS/peripherin may function as an adhesion molecule that stabilizes outer segment discs through homophilic or heterophilic interactions with adjacent discs.

Experiments employing monoclonal antibodies to various TMPs suggest that members of this superfamily participate in cellular signal transduction cascades. For example, incubation of human platelets with mAbs against CD9 results in an increase in intracellular Ca(69, 70, 71) . Intracellular Ca also increases when U937 cells are exposed to CD82 mAbs(54) . OX-44 mAbs produce elevations in inositol phosphates, stimulate tyrosine phosphorylation of cellular proteins, and increase Ca in RNK-16 cells(57) . TAPA-1 mAbs also increase tyrosine phosphorylation(63) . These effects appear to require the participation of other cell surface proteins (e.g. Refs. 49, 57, and 62).

Although we have operationally defined il-TMP as a participant in the regulation of contact-associated inhibition of HeLa cell growth, its function in cultured HT-29-inosine cells, or in the human intestinal epithelium, may be as a mediator of cell-cell contacts (e.g. adhesion), or inter- or intracellular signal transduction cascades.

A Combinatorial Model for Regulating (il-)TMP Function

Our studies of il-TMP plus other reported analyses of TMPs suggest a combinatorial model for regulating their function. In this model, modulation of TMP production, modulation of the extent of TMP glycosylation, and/or modulation of expression of proteins that partner with a TMP, could contribute to defining the TMP's effect on a given cell lineage.

TMP Production

Three observations indicate that modulation of the level of expression of il-TMP influences its effect on proliferative status. (i) In HT-29-inosine cells, il-TMP levels increase as cells shift from a proliferative to quiescent state. (ii) In HT-29-glucose cells, il-TMP levels do not increase until after the cells have piled up into a multilayer and remain lower than in postconfluent, nonproliferating HT-29-inosine cells. (iii) Introduction of an il-TMP expression vector into HeLa cells is sufficient to enhance cell contact-associated inhibition of proliferation.

N-Glycosylation

Two observations suggest that N-glycosylation is required for il-TMP to express its inhibitory effects on growth. (i) The extent of il-TMP N-glycosylation in HeLa cells was similar to that present in the unrelated epithelial cell line, HT-29, even though they were grown in the presence of different carbon sources (HT-29 = inosine and pil-TMP-transfected HeLa = glucose). The extent of N-glycosylation was also similar to that present in nonproliferating villus-associated enterocytes. (ii) pil-TMP/neo-transfected SW480 cells contain levels of il-TMP comparable to those in pil-TMP/neo-HeLa cells when grown under identical conditions but the SW480 cells do not stop proliferating once they reach confluence. The extent of il-TMP glycosylation in SW480 cells is similar to that in HT-29-glucose rather than in HT-29-inosine cells.

N-Glycosylation of other TMPs appears to be important for expression of their biological functions. Treatment of isolated retinas with tunicamycin produces a phenotype similar to that observed in rds mice(72) . Some forms of retinitis pigmentosa in humans (which is similar to retinal degeneration slow) are associated with rds gene mutations(73, 74) . Most of these mutations occur in the glycosylated domain located between the protein's third and fourth transmembrane domains. These observations suggest that the structure of this extracellular domain and its state of glycosylation are critical for the homotypic or heterotypic interactions of RDS. The extent of N-glycosylation of CD82 fluctuates dramatically with human T-cell lymphotrophic virus, type I transformation or phytohemagglutinin activation of T-lymphocytes(51, 75) . Fukudome and co-workers have suggested that extensive N-glycosylation of CD82 may be important for reducing self-fusion of T-cells already infected with human T-cell lymphotrophic virus, type I by reducing cell-cell interactions.

The mechanism by which glycosylation affects TMP function is not known. The combinatorial model of (il-)TMP regulation may have to consider each glycoform as having unique biological properties. il-TMP contains two potential sites for N-glycosylation. HeLa cells offer an opportunity to assess the relative activities of wild type il-TMP versus mutants with Asn Gln and/or Asn Gln substitutions. However, results obtained in HT-29-glucose cells and in pil-TMP/neo-transfected SW480 cells suggest that the nature of the carbohydrate moiety rather than its mere presence or absence modulates il-TMP function. This latter notion can be assessed with the transfected HeLa and SW480 cell lines described in this paper. For example, pil-TMP/neo and pControl/neo-transfected HeLa cells can be treated with inhibitors of carbohydrate processing enzymes and the effects on the ability of the il-TMP pathway to modulate cell growth monitored. In addition, glycosyltransferases associated with oligosaccharide branching can be introduced into pil-TMP/neo- (and pControl/neo) SW480 cells to generate new patterns of (il-TMP) glycosylation and assess the effects on proliferation. This, of course, will require that the appropriate nucleotide sugars are available to the glycosyltransferase and that the carbohydrate moiety already present on il-TMP can serve as an acceptor site. In this sense, regulation of specific il-TMP glycoform production may need to be viewed as a composite of the regulation of il-TMP expression, (glycosyl)transferase expression, and expression of enzymes involved in sugar metabolism.

Protein Partners

The inability of il-TMP to produce cell contact-associated inhibition of HT-29-glucose and SW480 cell proliferation may also reflect their inability to produce proteins that are components of cell surface-associated il-TMP complexes in HT-29-inosine and HeLa cells. TMPs are known to partner with other TMPs or non-TMPs. TAPA-1 interacts with a 16-kDa protein known as LEU-13, forming a complex that appears to be important for signaling(61, 62) . CD37, CD53(76) , R2(77) , and TAPA-1 (all TMPs) and CD19, CD21, and HLA-DR antigens (all non-TMPs) can assemble into large, multiprotein complexes(78) . Since CD19/CD21 and TAPA-1/LEU-13 are expressed at different times during B-cell differentiation, the composition of these complexes varies and may produce different biological effects(62) .

Non-cell Autonomous Decision Making along the Crypt-Villus Axis

The precise role of il-TMP in establishing and maintaining cell-cell contacts and/or a growth-arrested state in crypt-villus units (or in the liver) requires that gain-of-function or loss-of-function experiments be performed in an in vivo model. Nonetheless, the concept that molecules like il-TMP participate in regulating the adhesive and proliferative status of intestinal epithelial cells is consistent with an emerging set of studies that highlight the importance of cell-cell and cell-matrix contacts in regulating their proliferation, differentiation, and death programs. Expression of a dominant negative N-cadherin mutant in villus-associated enterocytes of chimeric-transgenic mice disrupts cell-cell and cell-matrix contacts. This disruption is associated with loss of polarity and dedifferentiation plus precocious entry into a death program(17) . Results such as these raise questions about how much of the decision making that occurs along the crypt-villus axis is cell-autonomous and how much is dependent upon a molecular cross-talk between adjacent epithelial cells, the extracellular matrix, and components of underlying mesenchyme.

FOOTNOTES

*
This work was supported in part by Grant DK30292 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked ``advertisement'' in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank(TM)/EMBL Data Bank with accession number(s) U31449[GenBank].

§
To whom correspondence should be addressed: Dept. of Molecular Biology and Pharmacology, Campus Box 8103, Washington University School of Medicine, 660 S. Euclid Ave., St. Louis, MO 63110. Tel.: 314-362-7243; Fax: 314-362-7058; jgordon{at}pharmdec.wustl.edu.

(^1)
Abbreviations used include: il-TMP, intestinal and liver tetraspan membrane protein; DMEM, Dulbecco's modified Eagle's medium; FCS, fetal calf serum; PBS, phosphate-buffered saline; IRES, internal ribosome entry site; ORF, open reading frame; DPP-IV, dipeptidylpeptidase IV; PAGE, polyacrylamide gel electrophoresis; Ig, immunoglobulin; FITC, fluorescein isothiocyanate; Cy3, indocarbocyanine; Endo H, endo-beta-N-acetylglucosaminidase H; mAb, monoclonal antibody; PVDF, polyvinylidene difluoride.

ACKNOWLEDGEMENTS

We thank members of our laboratory and Dwight Towler for their many stimulating discussions.

REFERENCES

  1. Gordon, J. I., and Hermiston, M. L. (1994) Curr. Opin. Cell Biol. 6,795-803 [CrossRef][Medline] [Order article via Infotrieve]
  2. Wright, N. A., and Irwin, M. (1982) Cell Tissue Kinet. 15,595-609 [Medline] [Order article via Infotrieve]
  3. Kellett, M., Potten, C. S., and Rew, D. A. (1992) Epith. Cell Biol. 1,147-155 [Medline] [Order article via Infotrieve]
  4. Cheng, H., and Leblond, C. P. (1974) Am. J. Anat. 141,461-480 [CrossRef][Medline] [Order article via Infotrieve]
  5. Roth, K. A., Hertz, J. M., and Gordon, J. I. (1990) J. Cell Biol. 110,1791-1801 [Abstract/Free Full Text]
  6. Aiken, K. D., Kisslinger, J. A., and Roth, K. A. (1994) Dev. Dyn. 201,63-70 [Medline] [Order article via Infotrieve]
  7. Falk, P., Roth, K. A., and Gordon, J. I. (1994) Am. J. Physiol. 266,G987-G1003
  8. Bry, L., Falk, P., Huttner, K., Ouellette, A., Midtvedt, T., and Gordon, J. I. (1994) Proc. Natl. Acad. Sci. U. S. A. 91,10335-10339 [Abstract/Free Full Text]
  9. Potten, C. S., and Loeffler, M. (1990) Development 110,1001-1020 [Abstract/Free Full Text]
  10. Schmidt, G. H., Wilkinson, M. M., and Ponder, B. A. J. (1985) Cell 40,425-429 [CrossRef][Medline] [Order article via Infotrieve]
  11. Cheng, H. (1974) Am. J. Anat. 141,481-502 [CrossRef][Medline] [Order article via Infotrieve]
  12. Cheng, H., and Leblond, C. P. (1974) Am. J. Anat. 141,503-520 [CrossRef][Medline] [Order article via Infotrieve]
  13. Cheng, H., and Leblond, C. P. (1974) Am. J. Anat. 141,537-562 [CrossRef][Medline] [Order article via Infotrieve]
  14. Paulus, U., Loeffler, M., Zeidler, J., Owen, G., and Potten, C. S. (1993) J. Cell Sci. 106,473-484 [Abstract]
  15. Cheng, H. (1974) Am. J. Anat. 141,521-536 [CrossRef][Medline] [Order article via Infotrieve]
  16. Duluc, I., Freund, J.-N., Leberquier, C., and Kedinger, M. (1994) J. Cell Biol. 126,211-221 [Abstract/Free Full Text]
  17. Hermiston, M. L., and Gordon, J. I. (1995) J. Cell Biol. 129,489-506 [Abstract/Free Full Text]
  18. Fogh, J., and Trempe, G. (1975) in Human Tumor Cells in Vitro (Fogh, J., eds) pp. 115-159, Plenum Press, New York
  19. Pinto, M., Appay, M.-D., Simon-Assmann, P., Chevalier, G., Dracopoli, N., Fogh, J., and Zweibaum, A. (1982) Biol. Cell 44,193-196
  20. Trugnan, G., Rousset, M., Chantret, I., Barbat, A., and Zweibaum, A. (1987) J. Cell Biol. 104,1199-1205 [Abstract/Free Full Text]
  21. Wice, B. M., Trugnan, G., Pinto, M., Rousset, M., Chevalier, G., Dussaulx, E., Lacroix, B., and Zweibaum, A. (1985) J. Biol. Chem. 260,139-146 [Abstract/Free Full Text]
  22. Huet, C., Sahuquillo-Merino, C., Coudrier, E., and Louvard, D. (1987) J. Cell Biol. 105,345-357 [Abstract/Free Full Text]
  23. Wice, B. M., and Gordon, J. I. (1992) J. Cell Biol. 116,405-422 [Abstract/Free Full Text]
  24. Wice, B. M., Reitzer, L. J., and Kennell, D. (1981) J. Biol. Chem. 256,7812-7819 [Free Full Text]
  25. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K. S. (eds) (1989) Current Protocols in Molecular Biology , John Wiley & Sons, New York
  26. 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]
  27. Laemmli, U. K. (1970) Nature 227,680-685 [CrossRef][Medline] [Order article via Infotrieve]
  28. Lessard, J. L. (1988) Cell Motil. Cytoskel. 10,349-362 [CrossRef][Medline] [Order article via Infotrieve]
  29. Beaulieu, J.-F., Nichols, B., and Quaroni, A. (1989) J. Biol. Chem. 264,20000-20011 [Abstract/Free Full Text]
  30. Bollag, D. M., and Edelstein, S. J. (1991) Protein Methods , pp. 203, Wiley-Liss, New York
  31. Kaufman, R. J., Davies, M. V., Wasley, L. C., and Michnick, D. (1991) Nucleic Acids Res. 19,4485-4490 [Abstract/Free Full Text]
  32. Ghattas, I. R., Sanes, J. R., and Majors, J. E. (1991) Mol. Cell. Biol. 11,5848-5859 [Abstract/Free Full Text]
  33. Kozak, M. (1987) Nucleic Acids Res. 15,8125-8148 [Abstract/Free Full Text]
  34. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual , p. 545, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY
  35. Peterson, G. L. (1977) Anal. Biochem. 83,346-356 [CrossRef][Medline] [Order article via Infotrieve]
  36. Folz, R. J., and Gordon, J. I. (1987) Biochem. Biophys. Res. Commun. 146,870-877 [CrossRef][Medline] [Order article via Infotrieve]
  37. Kyte, J., and Doolittle, R. F. (1982) J. Mol. Biol. 157,105-132 [CrossRef][Medline] [Order article via Infotrieve]
  38. Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J. (1990) J. Mol. Biol. 215,403-410 [CrossRef][Medline] [Order article via Infotrieve]
  39. Hellström, I., Horn, D., Linsley, P., Brown, J. P., Brankovan, V., and Hellström, K. E. (1986) Cancer Res. 465,3917-3923
  40. Marken, J. S., Schieven, G. L., Hellström, I., Hellström, K. E., and Aruffo, A. (1992) Proc. Natl. Acad. Sci. U. S. A. 89,3503-3507 [Abstract/Free Full Text]
  41. Horejsí, V., and Vlcek, C. (1991) FEBS Lett. 288,1-4 [CrossRef][Medline] [Order article via Infotrieve]
  42. Hixson, D. C., Ponce, M. D. L., Allison, J. P., and Walborg, E. F., Jr. (1984) Exp. Cell Res. 152,402-414 [CrossRef][Medline] [Order article via Infotrieve]
  43. Sigal, S. H., Brill, S., Fiorino, A. S., and Reid, L. M. (1992) Am. J. Physiol. 263,G139-G148
  44. Matsumoto, Y., Bishop, G. A., and McCaughan, G. W. (1992) Hepatology 15,1048-1053 [Medline] [Order article via Infotrieve]
  45. Ogier-Denis, E., Codogno, P., Chantret, I., and Trugnan, G. (1988) J. Biol. Chem. 263,6031-6037 [Abstract/Free Full Text]
  46. Shibuya, N., Goldstein, I. J., Van Damme, E. J. M., and Peumans, W. J. (1988) J. Biol. Chem. 263,728-734 [Abstract/Free Full Text]
  47. Leibovitz, A., Stinson, J. C., McCombs, W. B., III, McCoy, C. E., Mazur, K. C., and Mabry, N. D. (1976) Cancer Res. 36,4562-4569 [Abstract/Free Full Text]
  48. Chantret, I., Barbat, A., Dussaulx, E., Brattain, M. G., and Zweibaum, A. (1988) Cancer Res. 48,1936-1942 [Abstract/Free Full Text]
  49. Fearon, D. T. (1993) Curr. Opin. Immunol. 5,341-348 [CrossRef][Medline] [Order article via Infotrieve]
  50. Yu, J., Lin, J.-H., Wu, X.-R., and Sun, T.-T. (1994) J. Cell Biol. 125,171-182 [Abstract/Free Full Text]
  51. Fukudome, K., Furuse, M., Imai, T., Nishimura, M., Takagi, S., Hinuma, Y., and Yoshie, O. (1992) J. Virol. 66,1394-1401 [Abstract/Free Full Text]
  52. Oren, R., Takahashi, S., Doss, C., Levy, R., and Levy, S. (1990) Mol. Cell. Biol. 10,4007-4015 [Abstract/Free Full Text]
  53. Ledbetter, J. A., Shu, G., and Clark, E. A. (1987) in Leucocyte Typing III: White Cell Differentiation Antigens , pp. 339-340, Oxford University Press, Oxford
  54. Gil, M. L., Vita, N., Lebel-Binay, S., Miloux, B., Chalon, P., Kaghad, M., Marchiol-Fournigault, C., Conjeaud, H., Caput, D., Ferrara, P., Fradelizi, D., and Qiullet-Mary, A. (1992) J. Immunol. 148,2826-2833 [Abstract]
  55. Nojima, Y., Hirose, T., Tachibana, K., Tanaka, T., Shi, L., Doshen, J., Freeman, G. J., Schlossman, S. F., and Morimoto, C. (1993) Cell Immunol. 152,249-260 [CrossRef][Medline] [Order article via Infotrieve]
  56. Paterson, D. J., Green, J. R., Jefferies, W. A., Puklavec, M., and Williams, A. F. (1987) J. Exp. Med. 165,1-13 [Abstract/Free Full Text]
  57. Bell, G. M., Seaman, W. E., Niemi, E. C., and Imboden, J. B. (1992) J. Exp. Med. 175,527-536 [Abstract/Free Full Text]
  58. Kallin, B., DeMartin, R., Etzold, T., Sorrentino, V., and Philipson, L. (1992) Mol. Cell. Biol. 11,5338-5345
  59. Higashiyama, S., Iwamoto, R., Gioshi, K., Raab, G., Taniguchi, N., Klagsbrun, M., and Mekada, E. (1995) J. Cell Biol. 128,929-938 [Abstract/Free Full Text]
  60. Masellis-Smith, A., Jensen, G. S., Seehafer, J. G., Slupsky, J. R., and Shaw, A. R. E. (1990) J. Immunol. 144,1607-1613 [Abstract]
  61. Takahashi, S., Doss, C., Levy, S., and Levy, R. (1990) J. Immunol. 145,2207-2213 [Abstract]
  62. Bradbury, L. E., Kansas, G. S., Levy, S., Evans, R. L., and Tedder, T. F. (1992) J. Immunol. 149,2841-2850 [Abstract]
  63. Schick, M. R., Nguyen, V. Q., and Levy, S. (1993) J. Immunol. 151,1918-1925 [Abstract]
  64. Dong, J.-T., Lamb, P. W., Rinker-Schaeffer, C. W., Vukanovic, J., Ichikawa, T., Isaacs, J. T., and Barrett, J. C. (1995) Science 268,884-886 [Abstract/Free Full Text]
  65. Travis, G. H., Brennan, M. B., Danielson, P. E., Kozak, C. A., and Sutcliffe, J. G. (1989) Nature 338,70-73 [CrossRef][Medline] [Order article via Infotrieve]
  66. Connell, G. J., and Molday, R. S. (1990) Biochemistry 29,4691-4698 [CrossRef][Medline] [Order article via Infotrieve]
  67. Travis, G. H., Groshan, K. R., Lloyd, M., and Bok, D. (1992) Neuron 9,113-119 [CrossRef][Medline] [Order article via Infotrieve]
  68. Travis, G. H., Sutcliffe, J. G., and Bok, D. (1991) Neuron 6,61-70 [CrossRef][Medline] [Order article via Infotrieve]
  69. Carroll, R. C., Worthington, R. E., and Boucheix, C. (1990) Biochem. J. 266,527-535 [Medline] [Order article via Infotrieve]
  70. Hato, T., Sumida, M., Yasukawa, M., Watanabe, A., Okuda, H., and Kobayashi, Y. (1990) Blood 75,1087-1091 [Abstract/Free Full Text]
  71. Ozaki, Y., Matsumoto, Y., Yatomi, Y., Higashihara, M., and Kume, S. (1991) Eur. J. Biochem. 199,347-354 [Medline] [Order article via Infotrieve]
  72. Ulshafer, R. J., Allen, C. B., and Fliesler, S. J. (1986) Invest. Ophthalmol. Vis. Sci. 27,1587-1594 [Abstract/Free Full Text]
  73. Kajiwara, K., Hahn, L. B., Mukai, S., Travis, G. H., Berson, E. L., and Dryja, T. P. (1991) Nature 354,480-483 [CrossRef][Medline] [Order article via Infotrieve]
  74. Wells, J., Wroblewski, J., Keen, J., Inglehearn, C., Jubb, C., Eckstein, A., Jay, M., Arden, G., Bhattacharya, S., Fitzke, F., and Bird, A. (1993) Nat. Genet. 3,213-218 [CrossRef][Medline] [Order article via Infotrieve]
  75. Imai, T., Fukudome, K., Takagi, S., Nagira, M., Furuse, M., Fukuhara, N., Nishimura, M., Hinuma, Y., and Yoshi, O. (1992) J. Immunol. 149,2879-2886 [Abstract]
  76. Amiot, M. (1990) J. Immunol. 145,4322-4325 [Abstract]
  77. Gaugitsch, H. W., Hofer, E., Huber, N. E., Schnabl, E., and Baumruker, T. (1991) Eur. J. Immunol. 21,377-383 [Medline] [Order article via Infotrieve]
  78. Angelisová, P., Hilgert, I., and Horejsí, V. (1994) Immunogenetics 39,249-256 [Medline] [Order article via Infotrieve]
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Cell Sci.Home page
T. Lekishvili, E. Fromm, M. Mujoomdar, and F. Berditchevski
The tumour-associated antigen L6 (L6-Ag) is recruited to the tetraspanin-enriched microdomains: implication for tumour cell motility
J. Cell Sci., March 1, 2008; 121(5): 685 - 694.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
J. Zhang, F. Li, X. Liu, L. Shen, J. Liu, J. Su, W. Zhang, Y. Deng, L. Wang, N. Liu, et al.
The Repression of Human Differentiation-related Gene NDRG2 Expression by Myc via Miz-1-dependent Interaction with the NDRG2 Core Promoter
J. Biol. Chem., December 22, 2006; 281(51): 39159 - 39168.
[Abstract] [Full Text] [PDF]

Home page
 GutHome page
B Diosdado, M C Wapenaar, L Franke, K J Duran, M J Goerres, M Hadithi, J B A Crusius, J W R Meijer, D J Duggan, C J J Mulder, et al.
A microarray screen for novel candidate genes in coeliac disease pathogenesis
Gut, July 1, 2004; 53(7): 944 - 951.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
L. Pascual-Le Tallec, E. Dulmet, X. Bertagna, and Y. de Keyzer
Identification of Genes Associated with the Corticotroph Phenotype in Bronchial Carcinoid Tumors
J. Clin. Endocrinol. Metab., November 1, 2002; 87(11): 5015 - 5022.
[Abstract] [Full Text] [PDF]

Home page
 J. Cell Sci.Home page
F. Berditchevski
Complexes of tetraspanins with integrins: more than meets the eye
J. Cell Sci., January 12, 2001; 114(23): 4143 - 4151.
[Abstract] [Full Text] [PDF]

Home page
 JCBHome page
M. Yanez-Mo, A. Alfranca, C. Cabanas, M. Marazuela, R. Tejedor, M. Angeles Ursa, L. K. Ashman, M. O. de Landazuri, and F. Sanchez-Madrid
Regulation of Endothelial Cell Motility by Complexes of Tetraspan Molecules CD81/TAPA-1 and CD151/PETA-3 with alpha 3beta 1 Integrin Localized at Endothelial Lateral Junctions
J. Cell Biol., May 4, 1998; 141(3): 791 - 804.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
I. Tachibana, J. Bodorova, F. Berditchevski, M. M. Zutter, and M. E. Hemler
NAG-2, a Novel Transmembrane-4 Superfamily (TM4SF) Protein That Complexes with Integrins and Other TM4SF Proteins
J. Biol. Chem., November 14, 1997; 272(46): 29181 - 29189.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
F. Berditchevski, K. F. Tolias, K. Wong, C. L. Carpenter, and M. E. Hemler
A Novel Link between Integrins, Transmembrane-4 Superfamily Proteins (CD63 and CD81), and Phosphatidylinositol 4-Kinase
J. Biol. Chem., January 31, 1997; 272(5): 2595 - 2598.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wice, B. M.
Right arrow Articles by Gordon, J. I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wice, B. M.
Right arrow Articles by Gordon, J. I.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 1995 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement