Galectin-3 Promotes Keratinocyte Migration by Regulating EGFR Recycling Through Alix in MVB

Galectin-3 promotes keratinocyte migration by regulating epidermal growth factor receptor recycling through Alix in multivesicular bodies

Wei Liu1, Daniel K. Hsu1, Huan-Yuan Chen1,2, Ri-Yao Yang1, Larissa N. Larsen1, Jiajing Li1, Roslyn R. Isseroff1, Fu-Tong Liu1,2

1. Department of Dermatology, University of California, Davis. California 95817, USA; 2.

Institute of Biomedical Sciences, Academia Sinica, Taipei 115, Taiwan, R.O.C.

Address correspondence to: Fu-Tong Liu, Department of Dermatology, University of California, Davis, School of Medicine, 3301 C Street, Suite 1400, Sacramento, CA 95816, USA. Tel: +1 916 734 6377; Fax: +1 916 442 5702; E-mail:

The Eepidermal growth factor (EGF)-receptor(EGFR)- mediated signaling pathway plays very important roles in cell survival1, migration2, and proliferation3, as well as oncogenesis4 and wound re-epithelialization2,3. Intracellular trafficking of EGFR is critical for maintaining EGFR surface expression and proper responses to ligand stimulation5. Galectin-3, a member of an animal lectin family, has been implicated in a number of physiological and pathological processes6. Although the effects of recombinant galectin-3 on a variety of cell types through their binding to cell surface glycans have been extensively studied7, the role of endogenous galectin-3 remains elusive. Through studiesof keratinocytes isolated from galectin-3-deficient mice, we establish that galectin-3 positively regulates keratinocyte migration and skin wound re-epithelialization. We link this pro-migratory function to a crucial role of galectin-3 in controlling intracellular EGFR trafficking and EGFR surface expression after EGF stimulation. In the absence of galectin-3, the surface levels of EGFR are dramatically reduced and the receptor accumulates diffusely in the cytoplasm.This is associated with reduced rates of both endocytosis and recycling. This novel function of galectin-3 is mediated through direct interaction with its binding partner Alix, which is a protein component of the endosomal sorting complex required for transport (ESCRT) machinery. Galectin-3 is localized in multivesicular bodies (MVBs) and regulates the number of MVBs. The expression of another receptor regulated by MVBs, beta-adrenergic receptor, is also controlled by galectin-3. Our results suggest that galectin-3 is a critical regulator of intracellular trafficking of growth factor receptors and potentially controls a large number of important cellular responses through this intracellular mechanism.

Galectin-3 belongs to an animal lectin family defined by consensus sequences as well as β-galactoside-binding activities. It has been implicated in a number of cellular processes, includingsuch as cell growth8, cell differentiation9, and apoptosis10. While a study of corneal wound repair implicated galectin-3 in the healing process and by extension, cell migration11, the mechanism by which galectin-3 affects migration is not clear. Keratinocyte migration is a crucial step during the process of skin wound re-epithelialization process12 and the EGF-EGFR-ERK signaling pathway plays an important role2. Besides controlling cell migration, the EGF-EGFRthis pathway has been implicated in numerous cellular processes, such as cell survival, cell proliferation, cell differentiation and cell stress. Dysregulation of this the pathway can be oncogenic in some cell types. The cell surface expression of EGFR is critically controlled by intracellular trafficking, in which multivesicular bodies (MVBs) are crucial for receptor down-regulation13. EGFR can be sorted into different compartments inside the cell through MVBs 14, recycled back to plasma membrane, or degraded when the MVBs fuse with lysosomes.

We observed that the migration speed of isolated gal3-/- compared to gal3+/+ mouse keratinocytes was significantly lowerthan that of gal3+/+ counterparts (Fig. 1a). The decreased motility of gal3-/- keratinocytes was further confirmed by their decreased ability to close an in vitro scratchwound created in a confluent monolayer(Fig. 1b, c). In vivo, skin wounds in gal3-/- mice demonstrated significantly delayed re-epithelialization compared to gal3+/+ mice (Fig. 1d, e). Galectin-3 was found to be translocated to the leading edge of cells at 30 min post scratch, where it colocalized with phosphotyrosine but not with actin (Supplementary Fig.1), suggesting that galectin-3 is involved in signal transduction prior to cell migration. We compared gal3+/+ and gal3-/- keratinocytesfor EGF-stimulated EGFR-ERK signaling15. At 5 and 10 min after EGF stimulation, gal3+/+keratinocytes demonstrated increased phosphorylation of both EGFR (at Tyrosine 1068 residue) and ERKas compared to gal3-/-keratinocytes(Fig.1f). The migration speed of gal3+/+ keratinocytes was significantly reduced by an EGFR tyrosine kinase-specific inhibitor PD15035 (Fig. 1g), whilebut that of gal3-/- keratinocytemigration was not unaffected by this inhibitor, suggesting that galectin-3 modulates keratinocyte migration through its effects on the EGFR response to EGF.

The cCell surface EGFR expression was compared between gal3+/+ and gal3-/-keratinocytes by flow cytometry (Fig.2a, b). The mean fluorescence intensity of EGFRin gal3-/-cellswas significantly lower than that in gal3+/+ cells(0.8 ± 0.04 vs. 5.5 ± 0.9, P<0.05).In cultured gal3+/+ keratinocytes, EGFR, as expected, is predominately a plasma membrane protein, as expected. Remarkably, in gal3-/- keratinocytes, EGFR was found to beaccumulated in the peri-nuclear area of cytoplasm, but not on the plasma membrane(Fig. 2c). In gal3+/+ mice, EGFR was localized to the basal-lateral domain in the cell-cell contact region of peri-wound epidermis. In significant contrast, in gal3-/- mice, EGFR was accumulated in the cytoplasm of keratinocytes but not on the plasma membrane (Fig. 2d). We then tested the effect of EGF on the surface expression pattern of EGFR between gal3+/+ and gal3-/- mouse keratinocytes. When cells were cultured in the absence of EGF (growth factor deprivation), EGFR was localized on the plasma membrane in both genotypes (Fig. 2e 0h), and its levels werewith comparable levelsbetween gal3+/+ and gal-/-cells (Fig. 2f lanes 3&4). After EGF stimulation for 6 hr, the receptor was present on the plasma membrane in gal3+/+ cells,but hadwas accumulated in the cytoplasm of gal3-/- keratinocytes (Fig.2e 6h), and itswith significantly higher levels in gal3-/- keratinocytes were significantly higher than gal3+/+ counterparts (Fig. 2f,g lanes 1&2).

(need to mark lane 1, 2, 3, 4 in Fig 2 f). EGFR in gal3-/- keratinocytes does not co-localize with PDI[GK1] (endoplasmic reticulum ER marker), ceramide (Golgi complex marker), Rab8 (post trans Golgi network vesicle marker), EEA1 (early endosome marker), LAMP1 (lysosome marker) or Rab7 (MVB marker), but diffusely accumulates in the peri-nuclear region of the cytoplasm (Supplementary Fig. 2).By immunoelectron microscopy (immunoEM), EGFR was detected in large clusters in gal3+/+ keratinocytes, but diffusely distributed in gal3-/- keratinocytes (Fig. 2h, i). This observation by immunoe EM confirms the immunofluorescence data, and suggests that galectin-3 is critical for the maintenance of surface expression and intracellular trafficking of EGFR in keratinocytes.

The observed differences in EGF-induced redistribution of the EGFR in the gal3+/+ and in gal3-/-keratinocytes suggests the presence of a defect in receptor endocytosis, recycling, or degradation in gal3-/- keratinocytes. We found EGF-induced endocytosis of EGFR to be decreased in gal3-/- keratinocytes relative to gal3+/+ cells (Fig. 3a), using either low (2ng/ml) or high (100ng/ml) (Fig. 3b) EGF stimulation. EGFR recycling was also deficient in the gal3-/- keratinocytes, as no recycling was detected in gal3-/- keratinocytes up to 6 hr (Fig. 3c), while in gal3+/+ keratinocytes the EGFR recycled back to the plasma membrane within 30 min. The observed lower EGFR surface expression in gal3-/- keratinocytes could be due to either reduced new protein synthesis or a lower rate of recycling. Pre-treatment of cells with the protein synthesis inhibitor, cycloheximide, for 4 hr prior to stimulation with EGF discriminated between these possibilities, and demonstrated that the decrease in surface EGFR levels in gal3-/-relative to gal3+/+keratinocytes was independent of protein synthesis(Fig. 3d).

In addition to defective endocytosis and recycling of the EGFR in gal3-/-cells, the contribution of receptor degradation was examined. In the absence of any inhibitors, EGFR in both gal3+/+ and gal3-/- keratinocytes was degraded at 2 h following stimulation with 50 ng/ml EGF (Fig. 3e, upper lane). In gal3+/+ keratinocytes, EGFR degradation was attenuated by the lysosome inhibitor, chloroquine, but not the proteasome inhibitor,MG132. (Fig. 3e middle and bottom lanes on the left), suggesting that EGF-induced EGFR degradation occurs mainly through the lysosomal pathway in gal3+/+ keratinocytes. However, in gal3-/- keratinocytes, significant EGFR degradation persisted in the presence of either inhibitor (Fig. 3e, right lanes), suggesting that both lysosomal and proteasomal degradation pathways wereare activated.

We studied EGFR intracellular trafficking by immunofluorescence (Supplementary Fig. 3). At 5min post EGF stimulation, EGFR was found in EEA-1 positive early endosomes in both genotypes and.Aat 30min post EGF stimulation, EGFR was found in Rab7-positive MVBs in gal3+/+ keratinocytes bur not. However, in gal3-/- keratinocytes, EGFR was not colocalized with Rab7. At 60 min post stimulation, EGFR was found in LAMP1-positive lysosomes in gal3+/+keratinocytes, but not in gal3-/- keratinocytes. Thus gal3-/- keratinocytes exhibit multiple defects that contribute to the cytoplasmic accumulation, rather than cell membrane localization, of EGFR due to aberrant degradation by the proteasomal pathway, decreased recycling, and impaired endocytosis.

We turned our attention to the mechanism by which absence of galectin-3 could impair endocytosis. Galectin-3 is known to bind to EGFR through lectin-carbohydrate interactions 16. Demetriou et al.proposed that galectin-3 forms lattices with Mgat5-modified glycoproteins on the plasma membrane, therebyretarding receptor endocytosis. Evidence, however, mitigates this mechanism of action in keratinocytes. First, galectin-3 is localized mainly in the cytoplasm of keratinocytes (Supplementary Fig. 4). Second, as mentioned above, the ligand-induced EGFR rate of endocytosis is lower in gal3-/-than gal3+/+ mouse keratinocytes compared to gal3+/+ counterparts(Fig. 3a). Finally,recombinant galectin-3 induced EGFR endocytosis in neonatal human keratinocytes (NHK) (Fig. 3f, g), rather than stabilizing EGFR on the plasma membrane. Thus, we propose that endogenous galectin-3 regulates EGFR surface expression through an intracellular, rather than extracellular mechanism.

We then studied the association of galectin-3 with MVBs, which are organelles known to connect all three aspects of intracellular EGFR trafficking: endocytosis, recycling and degradation. We also studied the role of Alix, an intracellular binding partner of galectin-317, and a protein component of the endosomal sorting complex required for transport (ESCRT) machinery 18,with demonstrated abilityknownto attenuate EGFR endocytosis 19 and shown to regulate membrane invagination in early endosomes and formation of MVBs in vitro20.We observed galectin-3 to be translocated to MVBs at 30 min after EGF stimulation and colocalize with Rab7, a marker of MVBs, by immunofluorescence microscopy (Fig. 4a). The translocation of galectin-3 was further confirmed biochemically by sucrose density gradientfractionation. After EGF stimulation, both galectin-3 and Alix coexistedin the same fraction with Rab7, and these proteins were also detected in a fraction corresponding to the nucleus (Fig. 4b, lower panels fraction 3 and10). By immunEMoelectron microscopy, galectin-3 was detected on the membrane of intralumenal vesicles of MVBs in gal3+/+ keratinocytes after EGF stimulation (Fig. 4c). Importantly, the number of MVBs per cell was significantly higher in gal3-/-than in gal3+/+ keratinocytes compared to gal3+/+ keratinocytes (Fig. 4d, P < 0.001), while the number of lamellar bodies, amarker of keratinocyte phenotype, was comparable in these two genotypes.No significant difference was observed in the size of MVBs, as well as the size and the number of intra-lumen vesicles, between the two genotypes (data not shown).Thus, galectin-3 is associated with MVBs, and functions in MVBs.

Previously, we identified Alix as a binding partner for galectin-3 through a yeast two-hybrid screening using a Jurkat cDNA library17.Herewe demonstrate that galectin-3 can be co-precipitated with Alix from cultured keratinocytes 30 minafter EGF stimulation (Fig. 4e). In addition, galectin-3 was colocalized with Alix 30 min after scratch in NHK monolayers at the leading edge andin the peri-nuclear region (Fig. 4g). We compared the levels of EGFR co-immunoprecipitated with Alix in gal3+/+ and gal3-/- keratinocytes treated with a membrane permeable chemical crosslinker (Fig. 4f). EGFR released from the plasma membrane during cell lysis can potentially be bound to galectin-3through lectin-carbohydrate interactionand appear to be associated with Alix, if the galectin-3 protein is complexed with Alix. To exclude this possibility we included 5 mM lactoseinto the lysis buffer, to block galectin-3-carbohydrate binding. In gal3+/+ keratinocytes, Alix was not associated with EGFR at time 0, but an increased association was noted 10 min after EGF stimulation. In contrast, there was a significant amount of EGFR associated with Alix at time 0 in gal3-/- keratinocytes and a decreased association between the two proteins 10 min after EGF stimulation. The results suggest that galectin-3 is critical for modulating Alix-EGFR association after EGF stimulation through direct binding to Alix. It should be mentioned that thus far, there is no evidence that galectin-3 binds to EGFR directly inside the cells.

We further tested the role of galectin-3 in the surface expression of other cell surface receptors. 2-aAdrenergic receptor (B2AR) is a G-protein -coupled receptor, which has been reported to be downregulated through MVB sorting21. We compared the localization of B2AR between gal3+/+ and gal3-/- keratinocytes. In gal3-/- keratinocytes, B2AR was also found to accumulate in the peri-nuclear region of the cytoplasm (Fig. 5a). This supports the theorynotion that galectin-3 regulates the intracellular trafficking of multiple growth factors receptors through regulation of MVB sorting.

The presence of glycans on cell surface receptors with which galectin-3 interact hasve led to the description of several extracellular functions. Our studies haveuncovered a new paradigm for intracellular regulation of cell surface expression of growth factor receptorsby galectin-3. Through this mechanism, galectin-3 has the potential to be a key player in a large number of cellular processes mediated by these receptors. Previous studies have established the role of this protein in cell survival, cell cycle progression, and neoplastic transformation4. In view of this and the established role of EGFR in carcinogenesis, galectin-3 has great potential as a novel therapeutic target for cancers. The discovery of a carbohydrate-independent function of galectin-3 at MVBs mayshould have a significant impact on our understanding of the roles of intracellular galectins.

Methods summary

Primary cell isolations, single cell migration, scratch assay, dorsal skin wounding, EGFR endocytosis, recycling and degradation experiments were performed as described in detail in Supplementary Information.

Full methods and any associated references are available in the online version of the paper at

1Yang, L. and Baker, N. E. Cell cycle withdrawal, progression, and cell survival regulation by EGFR and its effectors in the differentiating Drosophila eye. Dev Cell4(3), 359 (2003).

2Fang, K. S. et al. Epidermal growth factor receptor relocalization and kinase activity are necessary for directional migration of keratinocytes in DC electric fields. J Cell Sci112 ( Pt 12), 1967 (1999).

3Andl, C. D. et al., Epidermal growth factor receptor mediates increased cell proliferation, migration, and aggregation in esophageal keratinocytes in vitro and in vivo. J Biol Chem278 (3), 1824 (2003).

4Hynes, N. E. and Lane, H. A., ERBB receptors and cancer: the complexity of targeted inhibitors. Nat Rev Cancer5 (5), 341 (2005).

5Futter, C. E., Pearse, A., Hewlett, L. J., and Hopkins, C. R., Multivesicular endosomes containing internalized EGF-EGF receptor complexes mature and then fuse directly with lysosomes. J Cell Biol132 (6), 1011 (1996).

6Dumic, J., Dabelic, S., and Flogel, M., Galectin-3: an open-ended story. Biochim Biophys Acta1760 (4), 616 (2006).

7Ochieng, J., Furtak, V., and Lukyanov, P., Extracellular functions of galectin-3. Glycoconj J19 (7-9), 527 (2004).

8Ellerhorst, J. A., Stephens, L. C., Nguyen, T., and Xu, X. C., Effects of galectin-3 expression on growth and tumorigenicity of the prostate cancer cell line LNCaP. Prostate50 (1), 64 (2002).

9Hikita, C. et al., Induction of terminal differentiation in epithelial cells requires polymerization of hensin by galectin 3. J Cell Biol151 (6), 1235 (2000).

10Yang, Ri-Yao, Hsu, Daniel K., and Liu, Fu-Tong, Expression of galectin-3 modulates T-cell growth and apoptosis. Proceedings of the National Academy of Sciences93 (13), 6737 (1996).

11Cao, Zhiyi et al., Galectins-3 and -7, but not Galectin-1, Play a Role in Re-epithelialization of Wounds. J. Biol. Chem.277 (44), 42299 (2002).

12Santoro, M. M. and Gaudino, G., Cellular and molecular facets of keratinocyte reepithelization during wound healing. Exp Cell Res304 (1), 274 (2005).

13von Zastrow, M. and Sorkin, A., Signaling on the endocytic pathway. Curr Opin Cell Biol19 (4), 436 (2007); Zwang, Y. and Yarden, Y., Systems biology of growth factor-induced receptor endocytosis. Traffic10 (4), 349 (2009).

14Katzmann, D. J., Odorizzi, G., and Emr, S. D., Receptor downregulation and multivesicular-body sorting. Nat Rev Mol Cell Biol3 (12), 893 (2002).

15Shirakata, Y. et al., Heparin-binding EGF-like growth factor accelerates keratinocyte migration and skin wound healing. J Cell Sci118 (Pt 11), 2363 (2005).

16Lau, Ken S. et al., Complex N-Glycan Number and Degree of Branching Cooperate to Regulate Cell Proliferation and Differentiation. Cell129 (1), 123 (2007).

17Chen, H. Y. et al., Galectin-3 negatively regulates TCR-mediated CD4+ T-cell activation at the immunological synapse. Proc Natl Acad Sci U S A106 (34), 14496 (2009).

18Katoh, K. et al., The ALG-2-interacting protein Alix associates with CHMP4b, a human homologue of yeast Snf7 that is involved in multivesicular body sorting. J Biol Chem278 (40), 39104 (2003).

19Schmidt, M. H. et al., Alix/AIP1 antagonizes epidermal growth factor receptor downregulation by the Cbl-SETA/CIN85 complex. Mol Cell Biol24 (20), 8981 (2004).

20Falguieres, T. et al., In vitro budding of intralumenal vesicles into late endosomes is regulated by Alix and Tsg101. Mol Biol Cell19 (11), 4942 (2008).

21Shenoy, S. K., McDonald, P. H., Kohout, T. A., and Lefkowitz, R. J., Regulation of receptor fate by ubiquitination of activated beta 2-adrenergic receptor and beta-arrestin. Science294 (5545), 1307 (2001).

22Isseroff, R. R., Ziboh, V. A., Chapkin, R. S., and Martinez, D. T., Conversion of linoleic acid into arachidonic acid by cultured murine and human keratinocytes. J. Lipid Res.28 (11), 1342 (1987).

23Rheinwatd, James G. and Green, Howard, Serial cultivation of strains of human epidemal keratinocytes: the formation keratinizing colonies from single cells. Cell6 (3), 331 (1975).

24Isseroff, R. R., Martinez, D. T., and Ziboh, V. A., Alterations in fatty acid composition of murine keratinocytes with in vitro cultivation. J Invest Dermatol85 (2), 131 (1985).

25Chen, J., Hoffman, B. B., and Isseroff, R. R., Beta-adrenergic receptor activation inhibits keratinocyte migration via a cyclic adenosine monophosphate-independent mechanism. J Invest Dermatol119 (6), 1261 (2002).

26Haas, A. F. et al., Low-energy helium-neon laser irradiation increases the motility of cultured human keratinocytes. J Invest Dermatol94 (6), 822 (1990).

27Hsu, D. K. et al., Targeted disruption of the galectin-3 gene results in attenuated peritoneal inflammatory responses. Am J Pathol156 (3), 1073 (2000).

28Sivamani, R. K. et al., Stress-mediated increases in systemic and local epinephrine impair skin wound healing: potential new indication for beta blockers. PLoS Med6 (1), e12 (2009).

29Delacour, Delphine et al., Requirement for Galectin-3 in Apical Protein Sorting. Current Biology16 (4), 408 (2006).

30Yamaoka, A., Kuwabara, I., Frigeri, L. G., and Liu, F. T., A human lectin, galectin-3 (epsilon bp/Mac-2), stimulates superoxide production by neutrophils. J Immunol154 (7), 3479 (1995).

31Sigismund, S. et al., Clathrin-mediated internalization is essential for sustained EGFR signaling but dispensable for degradation. Dev Cell15 (2), 209 (2008).

32Shi, J. and Kandror, K. V., Sortilin is essential and sufficient for the formation of Glut4 storage vesicles in 3T3-L1 adipocytes. Dev Cell9 (1), 99 (2005).

Supplementary

Methods

All animal experiments were approved by University of California Davis Institutional Animal Care and Use Committee (IACUC) and followed the guidelines of the Animal Welfare Actand theHealth Research Extension Act. Experiments using human tissues were approved by University of California Davis Institutional Review Board.

Human Primary keratinocyte culture and Mouse primary keratinocyte culture

Human keratinocytes were isolated from neonatal foreskins asdescribed22 and cultured using a modified method of Rheinwald and Green 23. Cells were grown inkeratinocyte growth medium (0.06 mM Ca2+, KGM, Epilife, Invitrogen, Carlsbad, CA ) containing human keratinocyte growth supplement (0.2 ng/mL EGF, 5µg/mL insulin, 5 µg/mL transferrin, 0.18 µg/mLhydrocortisone, and 0.2% bovine pituitary extract, Invitrogen, Inc., Carlsbad, CA, USA) and antibiotics (100 U/mLpenicillin, 100 µg/mL streptomycin, and 0.25 µg/mLamphotericin, Gemini Bio-Products, Inc., Calabasas, CA, USA)at 37°C in a humidified atmosphere of 5% CO2. Normal human keratinocyte (NHK) culturesisolated from at least two different foreskins were used andexperiments were performed with cells between passage3–7.