Expression of Helix pomatia Lectin Binding Glycoproteins in Women

Nov 11, 2008 - Another approach has been the enrichment of certain glyco- proteins using lectin ... cancer, but some of the healthy women did not know...
1 downloads 0 Views 3MB Size
Download PDF
Expression of Helix pomatia Lectin Binding Glycoproteins in Women with Breast Cancer in Relationship to Their Blood Group Phenotypes Charlotte Welinder,* Bo Jansson, Mårten Ferno ¨ , Håkan Olsson, and Bo Baldetorp Department of Oncology, Lund University, SE-221 85, Lund, Sweden Received June 17, 2008

Aberrant glycosylation occurs in essentially all types of human cancers. A difference in glycopattern of proteins will result in a change of function of the proteins. The lectin from Helix pomatia (HPA) recognizes N-acetylgalactosaminylated glycoproteins and very consistent results over the increased binding of HPA in tissue sections are associated with metastasis progression and poor patient prognosis in a range of human adenocarcinomas. The induced modification of protein function after changed glycosylation is unknown, and as a part in characterizing the glycoproteins carrying the specific carbohydrates, we analyzed the major HPA binding proteins in sera from healthy women, women with primary breast cancer with no metastasis (bcmet-), and women with metastasizing breast cancer (bcmet+) using lectin affinity chromatography and lectin blotting. The binding ligands were further identified using mass spectrometry (MALDI-TOF MS) to confirm the captured glycoproteins. The major HPA binding proteins in serum were found to be IgA1, complement factor C3, von Willebrand factor (vWF), alpha2-macroglobulin and IgM. This set of antigens is a panel of candidates for useful HPA related biomarkers in sera, but our results also emphasize the fact that the blood group phenotypes are of most importance when using the lectin HPA in recognition of cancer biomarkers in sera and plasma. The results emphasize that interpretation of an individual change in the glycosylation pattern of a specific tumor marker always needs to be analyzed in its right context. This study shows that the blood group phenotypes can have a major impact on the results when analyzing HPA lectin binding. Keywords: Helix pomatia lectin • HPA • GalNAc glycoproteins • breast cancer • blood group • serum

Introduction Glycosylation is one of the most common post-translational modifications of proteins and approximately 50% of all plasma proteins are glycosylated.1 The modification affects the folding, stability and biological function of the proteins.2 It has even been suggested that glycosylation could block phosphorylation and thus be directly involved in activation and regulation of signaling pathways.3 The lectin from Helix pomatia (HPA) recognizes N- acetylgalactosaminylated glycoproteins and has on several occasions been reported to be associated with metastasis behavior and bad prognosis of breast cancer, melanoma and other carcinomas.4,5 For several years, oncologists have sought to identify serum markers that would reveal the presence of specific cancers to be used for prognosis, monitoring drug effects and defining treatment options. In consequence of the high complexity and the high dynamic range in sera, efforts are made to remove high-abundance proteins like albumin and immunoglobulins, which are 10 orders of magnitude greater than the lowabundance proteins. The depletion of high-abundance proteins can be made using adsorption of immobilized dyes,6 immu* To whom correspondence should be addressed. Charlotte Welinder, Ph.D., Department of Oncology, SE-221 85 Lund, Sweden. E-mail: [email protected]. Phone: +46 46 17 85 35. Fax: +46 46 14 73 27.

782 Journal of Proteome Research 2009, 8, 782–786 Published on Web 11/11/2008

noaffinity extraction7 or with commercial antibody based kits. Another approach has been the enrichment of certain glycoproteins using lectin affinity chromatography.8-12 In this study, serum from healthy women, women with breast cancer, bcmet- and bcmet+, were characterized by the lectin HPA. Glycoproteins were detected either by using lectinblotting or enriching by using lectin affinity chromatography and separated using one-dimensional gel electrophoresis (1DPAGE) followed by identification using MALDI-TOF MS. The HPA recognizes a heterogeneous range of glycoconjugates, including, among others, glycoproteins bearing the Tn epitope (N-acetylgalactosamine-O-Ser/Thr) and blood group A antigens. Therefore, sera from most of the individuals with A and AB phenotype revealed a more intense staining pattern than individuals with O and B phenotype. As expected, the procedure showed differences between patients with different blood groups, stressing the facts that both secretor status and blood group is essential in the background information when screening for biomarkers. Far Western blot analysis of 1D-PAGE separated sera indicated differences between individual sera that were not possible to detect in the affinity purified material indicating a difference in sensitivity and/or specificity between the different methods. The intensity of staining between individual bands in far Western blot also indicated far Western blot to be more of a quantitative method for determination of HPA binding compared to the affinity procedure. The HPA 10.1021/pr800444b CCC: $40.75

 2009 American Chemical Society

Expression of H. pomatia Lectin Binding Glycoproteins

letters

affinity chromatography combined with 1D-PAGE followed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) increased the sensitivity of the screening, avoiding the presence of major serum proteins and easily identifying known glycoproteins in sera. Major bands in sera after HPA affinity chromatography appear to be IgA1, C3, alpha 2 macroglobulin, vWF and IgM.

methanol for 5 min and destained with 50% methanol, 10% acetic acid or incubated with biotin conjugated lectin. Lectin Blotting. The membrane was blocked in 1% bovine serum albumin in TTBS (20 mM Tris, 150 mM NaCl, 0.2% Tween 20, pH 7.5) overnight at 4 °C. The membrane was incubated with biotinylated HPA (5 µg/mL) in 1% BSA in TTBS for 3 h at room temperature. The membrane was then washed briefly twice and followed by 3 × 5 min washes in TTBS. The membrane was incubated with horseradish peroxidase (HRP)conjugated streptavidin (GE HealthCare) diluted 1:2000 in 1% BSA in TTBS for 1 h followed by two brief washes and 3 × 5 min washes. The membrane was developed using autoradiography film and the enhanced chemiluminescence (ECL) system (Amersham Bioscience). Peptide Mass Fingerprinting. The bands of interest were cut out from the PVDF membrane which had been stained with Coomassie. The membrane pieces were washed with 0.5 mL of Milli-Q water for 30 min followed by four washes of 0.5 mL of 40% acetonitrile (ACN) in 25 mM ammonium bicarbonate for 20 min each until the Coomassie stain was removed. Then, the membranes were washed with 0.5 mL of Milli-Q water for 30 min and blocked with 0.5% PVP-40 in 100 mM acetic acid for 30 min at 37 °C. The membranes were washed 5 × 0.5 mL Milli-Q followed by two washes with 50 mM ammonium bicarbonate. Proteins were degraded into characteristic fragments with trypsin (sequencing grade, Promega) in 25 mM ammonium bicarbonate overnight at 37 °C. Digestion was terminated by addition of 20 µL of 2% trifluoroacetic acid, which also extracted the peptides from the membrane. After 2 h at room temperature, the peptides were purified from the digestion buffer using C18 Ziptips (Millipore). Briefly, the solid phase was conditioned using 2 × 10 µL 50% ACN, 0.1% TFA in Milli-Q water. The organic solvent was washed away by two washes of 10 µL of 0.1% TFA. The samples were aspirated and dispensed 30 times followed by two washes of 0.1% TFA to remove salts and unbound material. The purified peptides were eluted directly onto the sample target (Anchorchip target, Bruker Daltonik) where 0.7 µL of matrix, 2.5-dihydroxybenzoic acid (3 mg/mL in 30% ACN) had been added. The MALDI-TOF MS instrument was a Bruker Reflex III instrument (Bruker Daltonik, Bremen). The instrument was equipped with a delayed extraction ion source, utilizing a nitrogen laser at 337 nm and was operated in reflector mode at an accelerating voltage of 20 kV. A total of 160-210 single shot spectra were accumulated from each sample. The XMASS 5.0 and MS Biotools software packages provided by the manufactures were used for data processing. Known auto proteolysis products from the trypsin were used for internal calibration. Database Searching. For protein identification, human protein sequences in the Swiss-Prot database were searched using the Mascot Software (Matrix Science Ltd.). Parameters specified in the search were taxa, Homo sapiens; missed cleavages, 1; peptide mass tolerance, (0.1 Da; fixed modification, carbamidomethyl (C); variable modification, none.

Experimental Section Serum Samples. A total of 47 serum samples were collected from postmenopausal women under controlled conditions at the University Hospital in Lund. All samples were processed after collection for a maximum of 2 h prior to freezing at -80 °C. The study was based on samples from 16 healthy women (median age of 55.7, range 47-60), 16 patients with bcmet- (median age of 60.7, range 53-67), and 15 patient with bcmet+ (median age of 60.4, range 52-74). The blood group phenotypes were known for all the women with breast cancer, but some of the healthy women did not know their blood group phenotype. The study was approved by the Lund University research ethics committee. Affinity Chromatography. The HPA-agarose (EY Laboratories, Inc., San Mateo, CA), 100 µL gel, was loaded into a Spin Filter column 0.22 µM Cellulose Acetate (Agilent Technologies) and equilibrated with 10 mM Tris-HCl, pH 8.0, and 150 mM NaCl. Serum, 100 µL, was loaded to the column in a total volume of 0.5 mL for 2 h in 10 mM Tris-HCl, pH 8.0, and 150 mM NaCl and in the presence of 0.1% NP40, final concentration. The beads were centrifuged at 500g for 5 min and the unbound fraction was removed. The beads were washed 10 times with 0.5 mL of 10 mM Tris-HCl, pH 8.0, 150 mM NaCl, and 0.1% NP40. The beads were then mixed with 100 µL of 0.1 M N-acetyl galactosamine in 10 mM Tris-HCl, pH 8.0, and 150 mM for 30 min to release glycoproteins adsorbed onto the beads. This step was repeated once. The eluted fractions were subjected to acetone precipitation; 800 µL of ice-cold 100% acetone was added to the 200 µL fraction, incubated on ice for 15 min, and centrifuged at 10 000g for 30 min. The supernatant was removed and the pellets were allowed to air-dry. The pellets were dissolved in 40 µL of SDS-sample buffer pH 8.5 (247 mM Tris, 2% SDS, 0.51 mM EDTA, 16.5 mM DTT, 10% glycerol, 0.025% (w/v) bromophenol blue). 1D-PAGE. Whole sera was diluted 1/10 in distilled water and 5 µL was mixed with 5 µL 2× SDS-sample buffer and heated at 60 °C for 10 min. The samples were allowed to cool and 5 µL of 0.5 M iodoacetamide was added. The samples were applied onto Criterion XT Precast Gel 4-12% Bis-Tris (13.3 × 8.7 × 0.1 cm, BioRad) and the separation was carried out using Criterion gel-apparatus (BioRad) at 200 V for 60 min using MOPS as running buffer. Twenty-five microliters of the enriched glycoprotein from sera were heated and iodoacetamide was added as above. The gel electrophoresis was performed using precasted 4-12% NuPAGE (8 × 8 cm × 0.1 cm, Invitrogen). Electrophoresis was carried out in an XCell Mini-cell apparatus (Invitrogen) at 200 V for 40 min using MOPS as running buffer. The gels were either stained with silver13 or transferred to PVDF membrane. Blotting. After electrophoresis, the gel was soaked in blotting buffer (25 mM Tris-Base, 193 mM glycine and 20% ethanol) for 15 min and then transferred onto PVDF membrane (Invitrogen) for 1 h at constant voltage of 30 V. The membrane was either stained with 0.1% R-350 Coomassie Brilliant Blue in 50%

Results SDS-PAGE Analysis of Whole Sera. Sera, 0.5 µL, from 16 healthy women, 16 bcmet+ and 15 bcmet- were separated using 1D-PAGE followed by far Western blot. The sera from different women with breast cancer and healthy women displayed marked differences in HPA-binding characteristics (Figure 1). Individual patterns seemed to correlate with patient Journal of Proteome Research • Vol. 8, No. 2, 2009 783

letters

Welinder et al.

Figure 1. Whole sera (0.5 µL), from healthy women (H1-16), bcmet+ (M1-14, 16) and bcmet- (N1-16), separated by 1D-PAGE followed by HPA-staining. Phenotypes of blood group listed above the figure.

Figure 2. Affinity chromatography enriched glycoproteins, separated by 1D-PAGE and transferred to PVDF membrane, from sera of different blood group phenotype. The protein pattern was visualized using Coomassie staining. Lanes 1-3, healthy women; lanes 4-5, bcmet+; and lanes 7-9, bcmet-. Phenotypes of blood group listed above the figure.

ABO blood group, but individual differences were also seen for example H5, H9 and H10. Enrichment of Serum Glycoproteins. Glycoproteins were enriched using lectin affinity chromatography from different blood groups phenotypes. The proteins were separated using 1D-PAGE and then transferred to PVDF membrane. The membrane was stained with Coomassie. The enriched glycoproteins pattern (Figure 2) showed high similarities between individuals with the same ABO blood group regardless of disease status. Blood group 0 and B had equal pattern which differed from the pattern of AB. According to the blood group phenotypes, a selection was made using only women with blood group 0+ to reduce the pattern inconsequently of the blood group phenotype. Three serum samples from each group, healthy women, bcmet+ and bcmet-, were applied to lectin affinity chromatography for enrichment of glycoproteins. After enrichment, proteins were separated using 1D-PAGE and stained with silver (Figure 3A). A small fraction of the enriched glycoproteins equivalent to 5 µL of serum were subjected to far Western blot staining with biotinylated-HPA (Figure 3B). No remarkably differences were seen between women with breast cancer and healthy women neither in the protein pattern nor in the HPA-staining pattern. 784

Journal of Proteome Research • Vol. 8, No. 2, 2009

The same experiment was also performed using only women with blood group A. The separated proteins were transferred to PVDF membranes and stained with Coomassie (Figure 4A) and with biotinylated-HPA (Figure 4B). Both the protein and the HPA-binding pattern were almost the same with minor differences seen; however, there was no discrepancy between the three groups. Identification of Glycoproteins in Sera. Affinity chromatography enriched glycoproteins from women with blood group A were identified from the Coomassie stained PVDF membrane. Some of the sera (M4, N1, N4, N7, H12, H14, H15, M10, N11, N13 and N14) had an intense low molecular double band in the 50 kDa region (Figure 1). All those samples were from women with blood group A or AB. These bands could not be observed in the affinity enriched samples from the same women (H14, H15 and M4) (Figure 4). In all sera tested in Figure 2, there is an increase in the number of bands compared to the far Western blot experiment, (all O, B or AB sera) indicating a larger sensitivity of the affinity technique compared to far Western blot. In a similar way, Figures 3 and 4 also show an increase in the number of HPA binding proteins with molecular weight above the IgA1 heavy chain (apparent MW of 55 kDa). The weak stain of M1 in Figure 1 is partly repeated in Figure 4, but the variation in intensity of the far Western blot pattern in Figure 1 is partly lost in the pattern of the affinity purified glycoproteins.

Discussion HPA has been extensively used in cancer research as an indication of poor patient prognosis. There are convincing data showing the correlation between the formation of lymph node metastasis and the binding of HPA in breast cancer.14,15 High binding of HPA reflects the increased expression of a heterogeneous range of O-linked N-acetylglucosamine (GalNAc) proteins including among others also the blood group A antigen and the Tn antigen, and is a consequence of changes in the expression range of glycotransferases16 or an elevated increase Golgi pH leading to a redistribution of the glycotransferases in the Golgi apparatus.17 In this study, we investigated the expression of HPA-binding proteins in sera from healthy women, bcmet- and bcmet+. When whole sera were analyzed for HPA-binding by protein separation using 1D-PAGE followed by far Western blot, marked differences in HPA-binding pattern were obtained (Figure 1). The HPA-binding seemed to correlate with the ABO blood

Expression of H. pomatia Lectin Binding Glycoproteins

letters

Figure 3. Affinity chromatography enriched glycoproteins separated by 1D-PAGE from sera of blood group O phenotype. The protein pattern was visualized by silver staining (A) and the glycoprotein pattern with HPA-staining (B). Lanes 1-3, healthy women; lanes 4 and 5, bcmet+; and lanes 7-9, bcmet-.

Figure 4. Affinity chromatography enriched glycoproteins, separated by 1D-PAGE and transferred to PVDF membrane, from sera of blood group A phenotype. The protein pattern was visualized using Coomassie staining (A) and the glycoprotein pattern with HPAstaining (B). Lanes 1-3, healthy women; lanes 4 and 5, bcmet+; and lanes 7-9, bcmet-.

groups. The HPA-binding was more intensive for individuals with blood group A and AB, which reflects the fact that HPA apart from binding GalNAc also binds to the blood group A antigen. There were also interindividual differences for individuals belonging to the same blood group in the HPA-binding, irrespective of disease state. However, suppose that we at this stage had pooled our samples (pooling H1-8, M1-8 and N1-8, from Figure 1). Pooling of samples is a common procedure for reducing the time and costs of analysis. In this special case, the result would have been totally misleading, showing that women with breast cancer had a more intensive HPA-binding pattern compared to healthy women, as a consequence of the fact that HPA recognizes the blood group A antigen and the majority of the healthy group belonged to blood group O and the majority of the cases belonged to A and AB in the disease groups. The interindividual differences from sera from women with the same blood group were less pronounced after the glycoproteins had been enriched using HPA-affinity chromatography. Minor differences in the patterns were seen between the women, but no evident differences between the disease state and healthy women (Figures 2-4). This could be a consequence of the limited resolving power of protein separation using 1DPAGE.

However, a major difference was observed comparing the whole sera stained by biotinylated-HPA (Figure 1) and the enriched glycoproteins (Figure 2-4). The whole sera showed a larger variation in the HPA-staining between the different blood group phenotypes than the enriched glycoproteins did. An explanation could be that the affinity chromatography steps drastically reduce albumin level in the sample and that the more diluted glycoproteins are concentrated and the glycoproteins become more apparent on the 1D-PAGE. Some protein bands are seen on the blots with whole sera in the mass range of 37-50 kDa (Figure 1) but are missing after the enrichment (Figure 2-4). This phenomenon could be explained by low binding to these antigens in solution or to a different binding specificity for the reagents, biotinylated-HPA versus HPAagarose. The lectin has also more than one binding site and the avidity gain in multiple binding points to a narrow concentrated antigen band in the blot is advantageous compared to affinity for the antigen in solution. In this study, HPA-binding glycoproteins in sera were identified as IgA1, Alpha-2-macroglobulin, complement C3, ig mu heavy chain disease protein (MUCB), Ig mu chain C region (MUC), von Willebrand factor, and serrotransferrin. IgA1 has been found to bind to HPA in breast cancer tissue.12 IgA has also been reported to be increased in breast carcinoma Journal of Proteome Research • Vol. 8, No. 2, 2009 785

letters patients and the presence in tumor could be the result of local production18 or increased uptake.19 IgA1 was one of the dominating proteins found in the far Western blot, present in all sera of different ABO blood groups. In our experiment, there was no clear difference in the expression pattern of IgA1 HPA binding in sera from healthy women, bcmet-, and bcmet+. However, there was a large individual difference in the far Western blot experiment of whole sera indicating quantitative differences. Since the bottom line is to find functional differences originating from differences in glycosylating pattern, it is not only important to hunt for differences, but also to map the Glycome. Our main task is to analyze proteins from the sera binding HPA and correlating the degree of metastasis and outcome of disease. There is a great potential of lectin affinity chromatography for the reduction of the complexity in sera. The lectin capture simultaneously concentrates classes of glycoproteins and eliminates some of the major blood protein, such as albumin. The resolution of 1D-PAGE is still too low to be able to detect enough differences in sera and a second dimension of separation is probably necessary to reveal suitable biomarkers. Our results emphasize the fact that the blood group phenotypes are of most importance when using the lectin HPA or the use of other blood group-recognizing lectins (e.g., Dolichos biflorus (DBA), Vicia villosa (VVA), Phaseolus lunatus (LBA) and Glycine max (SBA) recognizing blood group A-antigen, and Griffonia simplicifolia I agglutinin B4 (GS-I B4) and Ulex europaeus I (UEA-I) recognizing blood group B- and Oantigens, respectively) in the recognition of cancer biomarkers in sera and plasma. This is especially true when samples from different individuals are pooled and analyzed using the blood group-recognizing lectins. We believe that a more detailed mapping of all HPA binding proteins have a potential to reveal useful biomarkers for human adenocarcinomas.

Acknowledgment. This study was supported by grants from Mrs. Berta Kamprad Foundation, the Gunnar Nilsson Cancer Foundation, the Franke and Margareta Bergquist Foundation, the Swedish Cancer Society, the Swedish Research Council, Governmental Funding of Clinical Re-

786

Journal of Proteome Research • Vol. 8, No. 2, 2009

Welinder et al. search within Nation Health Service, the Lund University Hospital Foundations and the Royal Physiographical Society in Lund. Senior oncologist Lotta Lundgren is greatly acknowledged for supporting with clinical patient data.

Supporting Information Available: Supporting Information Table 1 describing the patient cohort; Supporting Information Table 2 summarizing the peptide mass fingerprinting identification. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Van den Steen, P.; Rudd, P. M.; Dwek, R. A.; Opdenakker, G. Crit. Rev. Biochem. Mol. Biol. 1998, 33 (3), 151–208. (2) Mitra, N.; Sinha, S.; Ramya, T. N.; Surolia, A. Trends Biochem. Sci. 2006, 31 (3), 156–63. (3) Comer, F. I.; Hart, G. W. J. Biol. Chem. 2000, 275 (38), 29179–82. (4) Mitchell, B. S.; Schumacher, U. Histol. Histopathol. 1999, 14 (1), 217–26. (5) Brooks, S. A. Histol. Histopathol. 2000, 15 (1), 143–58. (6) Ahmed, N.; Barker, G.; Oliva, K.; Garfin, D.; Talmadge, K.; Georgiou, H.; Quinn, M.; Rice, G. Proteomics 2003, 3 (10), 1980–7. (7) Wang, Y. Y.; Cheng, P.; Chan, D. W. Proteomics 2003, 3 (3), 243–8. (8) Dam, T. K.; Roy, R.; Das, S. K.; Oscarson, S.; Brewer, C. F. J. Biol. Chem. 2000, 275 (19), 14223–30. (9) Hirabayashi, J.; Arata, Y.; Kasai, K. Proteomics 2001, 1 (2), 295– 303. (10) Lis, H.; Sharon, N. Chem. Rev. 1998, 98 (2), 637–674. (11) Drake, R. R.; Schwegler, E. E.; Malik, G.; Diaz, J.; Block, T.; Mehta, A.; Semmes, O. J. Mol. Cell. Proteomics 2006, 5 (10), 1957–67. (12) Streets, A. J.; Brooks, S. A.; Dwek, M. V.; Leathem, A. J. Clin. Chim. Acta 1996, 254 (1), 47–61. (13) Shevchenko, A.; Wilm, M.; Vorm, O.; Mann, M. Anal. Chem. 1996, 68 (5), 850–8. (14) Brooks, S. A.; Leathem, A. J. Lancet 1991, 338 (8759), 71–4. (15) Noguchi, M.; Thomas, M.; Kitagawa, H.; Kinishita, K.; Kinami, S.; Takamura, H.; Miyazaki, I.; Mizukami, Y. Breast Cancer Res. Treat. 1993, 26 (1), 67–75. (16) Brooks, S. A.; Carter, T. M.; Bennett, E. P.; Clausen, H.; Mandel, U. Acta Histochem. 2007, 109 (4), 273–84. (17) Cardone, R. A.; Casavola, V.; Reshkin, S. J. Nat. Rev. Cancer 2005, 5 (10), 786–95. (18) Geng, L. Y.; Shi, Z. Z.; Dong, Q.; Cai, X. H.; Zhang, Y. M.; Cao, W.; Peng, J. P.; Fang, Y. M.; Zheng, L.; Zheng, S. World J. Gastroenterol. 2007, 13 (16), 2305–11. (19) Syre, G.; Sehn, M. Virchows Arch. 1981, 393, 315–20.

PR800444B