Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Sep 27;7(39):63138-63157.
doi: 10.18632/oncotarget.11257.

Hypoxia enhances the malignant nature of bladder cancer cells and concomitantly antagonizes protein O-glycosylation extension

Andreia Peixoto  1   2   3   4 , Elisabete Fernandes  1   3   4   5 , Cristiana Gaiteiro  1 , Luís Lima  1   3   6 , Rita Azevedo  1   4 , Janine Soares  1 , Sofia Cotton  1 , Beatriz Parreira  1 , Manuel Neves  1   4 , Teresina Amaro  7 , Ana Tavares  1   7 , Filipe Teixeira  8 , Carlos Palmeira  1   9 , Maria Rangel  10 , André M N Silva  11 , Celso A Reis  3   4   6   12 , Lúcio Lara Santos  1   9   13 , Maria José Oliveira  2   3 , José Alexandre Ferreira  1   3   4   6   14
Affiliations

Hypoxia enhances the malignant nature of bladder cancer cells and concomitantly antagonizes protein O-glycosylation extension

Andreia Peixoto et al. Oncotarget. .

Abstract

Invasive bladder tumours express the cell-surface Sialyl-Tn (STn) antigen, which stems from a premature stop in protein O-glycosylation. The STn antigen favours invasion, immune escape, and possibly chemotherapy resistance, making it attractive for target therapeutics. However, the events leading to such deregulation in protein glycosylation are mostly unknown. Since hypoxia is a salient feature of advanced stage tumours, we searched into how it influences bladder cancer cells glycophenotype, with emphasis on STn expression. Therefore, three bladder cancer cell lines with distinct genetic and molecular backgrounds (T24, 5637 and HT1376) were submitted to hypoxia. To disclose HIF-1α-mediated events, experiments were also conducted in the presence of Deferoxamine Mesilate (Dfx), an inhibitor of HIF-1α proteasomal degradation. In both conditions all cell lines overexpressed HIF-1α and its transcriptionally-regulated protein CA-IX. This was accompanied by increased lactate biosynthesis, denoting a shift toward anaerobic metabolism. Concomitantly, T24 and 5637 cells acquired a more motile phenotype, consistent with their more mesenchymal characteristics. Moreover, hypoxia promoted STn antigen overexpression in all cell lines and enhanced the migration and invasion of those presenting more mesenchymal characteristics, in an HIF-1α-dependent manner. These effects were reversed by reoxygenation, demonstrating that oxygen affects O-glycan extension. Glycoproteomics studies highlighted that STn was mainly present in integrins and cadherins, suggesting a possible role for this glycan in adhesion, cell motility and invasion. The association between HIF-1α and STn overexpressions and tumour invasion was further confirmed in bladder cancer patient samples. In conclusion, STn overexpression may, in part, result from a HIF-1α mediated cell-survival strategy to adapt to the hypoxic challenge, favouring cell invasion. In addition, targeting STn-expressing glycoproteins may offer potential to treat tumour hypoxic niches harbouring more malignant cells.

Keywords: bladder cancer; glycosylation; hypoxia; invasion; sialyl-Tn.

PubMed Disclaimer

Conflict of interest statement

CONFLICTS OF INTEREST

The authors declare no potential conflicts of interests.

Figures

Figure 1
Figure 1. Representation of membrane proteins O-glycosylation (Ser/Thr residues) in advanced stage bladder tumours, with emphasis on STn biosynthesis
Protein glycosylation is a highly regulated process of critical importance for protein stability and function. Newly synthesized proteins are O-glycosylated in the Golgi apparatus by ppGalNAcTs. These enzymes are responsible by the addition of a GalNAc moiety to Ser/Thr residues, originating the Tn antigen (GalNAc-O-Ser/Thr-protein backbone), the simplest O-glycan. In healthy and/or less malignant cells, O-glycans are extended through the sequential addition of other sugars first by C1GalT1 and its chaperone Cosmc and then other glycosyltransferases. This culminates in highly complex, heterogeneous and elongated glycans often terminated by ABO or Lewis blood group related antigens (not represented). In cancer cells, the Tn antigen is immediately sialylated by ST6GalNAc-I, originating the STn antigen (Neu5Ac-GalNAc-O-Ser/Thr-protein backbone), thereby inhibiting further chain elongation. The expression of STn at the cell surface influences cell-cell adhesion and cancer cell recognition, favouring motility, invasion, immune escape and possibly altering intracellular signalling.
Figure 2
Figure 2. Molecular and morphological changes induced by hypoxia in bladder cancer cell lines
(A) Variations in HIF-1α in bladder cancer cells (T24, 5637, HT1376) exposed to hypoxia (0.1% O2; 24 h for T24 and 5637 and 6 h for HT1376) and Dfx in relation to normoxia (21.0% O2). All cell lines significantly increased the levels of HIF-1α in hypoxia and Dfx when compared to normoxia. (B) Lactate production in bladder cancer cells under normoxia, hypoxia and Dfx exposure. Growth under hypoxia or in the presence of Dfx promoted a significant increase in lactate production, which was more pronounced for Dfx conditions. These observations denote a shift towards anaerobic metabolism. (C) Effect of hypoxia on bladder cancer cell morphology and actin/tubulin cytoskeleton organization. F-actin and tubulin staining was performed in PFA fixed cells after exposure to normoxic, hypoxic and DFX conditions for 24 h (T24 and 5637) or 6 h (HT1376). F-actin was stained with Phalloidin-FITC (green), while α-tubulin was stained with a monoclonal antibody following incubation with an AlexaFluor594 secondary antibody (red). In turn, nuclei were counterstained with DAPI (blue). Scale bar represents 10 μm. Normoxic T24 cells form aggregated islands with evident actin staining at cell-cell adhesion sites (arrowheads). Membranes of outer cells present reduced projections and some actin stress fibbers are visible (arrows). Under hypoxic conditions, these cells present dissociated and scattered cell aggregates with pronounced and abundant actin filopodia (arrows) present at cell periphery. In comparison to normoxic conditions, less stress fibbers are visible. When exposed to DFX, T24 cells present an intermediate cell morphology between normoxia and hypoxia. Cell aggregates are not as cohesive as in normoxia and not as scattered as in hypoxia. Cells show reduced cell filopodia and exhibit abundant stress fibbers (arrows) co-localized with α-tubulin-rich focal adhesions (arrows). Under normoxia, 5637 cells present loosed islands with fewer cell-cell contact points (arrows) than the other cell lines. Short filopodia is seen at the cell periphery (arrows). In hypoxic conditions, 5637 cells appear as isolated cells with heterogeneous morphology. Pronounced, abundant and long actin filopodia (arrows) is present at cell periphery and no stress fibbers are visible. DFX exposure of these cells results in an intermediate cell morphology between normoxia and hypoxia. Cell islands seem to disperse and adhesion contacts are mainly established through cellular projections (arrows). Cell periphery exhibits lamellipodia and long filopodia. In both normoxic and hypoxic conditions, HT1376 cells form small cell islands, establishing cell contact through minor cell projections. Cell-cell adhesion areas are less enriched in actin filaments and no stress fibbers are visible. Under DFX exposure, cell islands are more cohesive with intense actin staining at cell-cell adhesion contacts. Cell periphery does not present filopodia nor lamellipodia. In resume, exposure to hypoxia drives cells to accumulate hypoxia marker HIF-1α, shift towards an anaerobic metabolism and to significant morphological alterations towards a less cohesive cell phenotype. These alterations are significantly more pronounced under Dfx exposure, highlighting the role of HIF-1α in the establishment of molecular alterations associated with hypoxia. Graphs represent average value of three independent experiments, *p < 0.05; **p < 0.01; ***p < 0.001 (Student's T-test).
Figure 3
Figure 3
(A) STn expression in bladder cancer cell lines (T24, 5637, HT1376) in normoxia (21.0% O2), hypoxia (0.1% O2; 24 h for T24 and 5637 and 6 h for HT1376), Dfx and Hypoxia followed by reoxygenation (0.1 + 21.0% O2). Briefly, approximately 104 cells were probed with the anti-STn TKH2 mouse monoclonal antibody and analysed by flow cytometry. Negative controls included cells treated with α-neuraminidase (NeuAse), responsible by removing the sialic acid from the STn antigen, thereby impairing recognition by the antibody. An IgG1 isotype was also used as negative control. Accordingly, exposure to hypoxia and Dfx induced a significant increase in STn expression (left FACS and bottom Graph). The specificity of TKH2 monoclonal antibody to STn was confirmed by the significant decrease in the mean fluorescence intensity presented by cells treated with neuraminidase (right FACS). STn increase in Dfx-treated cells suggests these events may be directly and/or indirectly regulated by HIF-1α. The reoxygenation of hypoxic cells restored the basal STn levels, demonstrating the key role played by O2 levels in STn expression. (B) STn-expressing glycoproteins determined by western blot in normoxia, hypoxia and Dfx for T24 cell line. The western blots highlight that the STn antigen is mostly present in proteins with a molecular weight above 50 kDa. The relative levels of STn are presented bellow for each lane and are the average value of three independent experiments. For each cell line, hypoxia and Dfx-exposed cells presented statistically higher levels of STn in comparison to normoxia when compared to the reference protein (p < 0.05 for hypoxia and Dfx), thus in agreement with flow cytometry analysis. Similar tendencies were observed for the other cell lines (data now shown). The loss of signals after NeuAse treatment confirms the specificity of antibody recognition. (C) Total STn levels by slotblot in T24, 5637 and HT1376 cells in normoxia, hypoxia and Dfx. A marked increased in STn is observed in hypoxia and Dfx for all cell lines. Again, the specificity of the signals was confirmed by NeuAse treatment. *p < 0.05; **p < 0.01; ***p < 0.001 (Student's T-test).
Figure 4
Figure 4
(A) Number of putative STn-expressing glycoproteins identified by VVA lectin affinity-nanoLC tandem ESI-MS in normoxic and hypoxic cells. Briefly, proteins isolated from T24, 5637 and HT1376 cell lines in normoxia and hypoxia were digested with α-neuraminidase, which removes the sialic acid from STn exposing the Tn antigen. The samples were then enriched for Tn-expressing glycoproteins by VVA affinity chromatography and identified by nanoLC tandem ESI-MS. Since the Tn antigen was not detected in the original samples (before neuraminidase treatment) by both western blot and flow cytometry, it is likely that the glycoproteins identified by this approach may yield the STn antigen. Accordingly, a higher number of putative STn-expressing glycoproteins were present in hypoxic cells in comparison to normoxia. Interestingly, MUC16 was the only glycoprotein common to all cells and conditions. (B) Main signalling pathways associated with STn-expressing glycoproteins in hypoxia. Over 35% of the identified glycoproteins were associated with integrin or cadherin signalling pathways, as highlighted in the Figure. Other key oncogenic pathways also included Wnt signalling and angiogenesis. The STn antigen was also found in glycoproteins involved in other 36 molecular functions, highlighting the broad span of biological roles of STn expressing glycoproteins.
Figure 5
Figure 5. Hypoxia enhances the invasion and migration capability of bladder cancer cells in a STn-dependent manner
(A) Invasive capability of bladder cancer cells (T24, 5637, HT1376) under normoxia, hypoxia, and hypoxia + anti-STn moAb TKH2. T24 and 5637 display invasive properties under normoxia, whereas HT1376 showed only residual invasive capability. Exposure to hypoxia significantly increased invasion in T24 and 5637 cell lines. Hypoxic HT1376 cells did not present an invasive phenotype. Exposure to anti-STn moAb TKH2 monoclonal antibody promoted a dramatic decrease in invasion in all cell lines. More importantly, the invasive capability of cells exposed to TKH2 was significantly lower when compared to cells grown in normoxia, where STn is also present, even though in lower levels than in hypoxia. Therefore, we envisage that the anti-STn antibody may inhibit/target cancer cells led to express de novo the STn antigen in hypoxia, as well as clones that originally presented the glycan. These results suggest that this antibody may be used to selectively inhibit invasion. (B) Exposure to hypoxia increased the migratory capacity of the cells in relation to normoxia, in a STn-dependent manner. T24 and 5637 cell lines increased their migratory capacity under hypoxia, which was significantly inhibited by exposure to TKH2, further suggesting that this antibody may be used to selectively inhibit migration. Noteworthy, since invasion assays were done using wells with defined gaps, t = 0 is the same for normoxia and hypoxia. (C) MMP activity in hypoxic cells in relation to normoxia determined by gelatinase zymography. Representative gelatinase zymogram of bladder cancer cells show two bands at 72 and 92 kDa corresponding to active forms of MMP-2 and MMP-9, respectively (upper Figure). Residual amounts of inactive pro-MMPs could also be observed. Accordingly, no significant alterations in MMP-2 and -9 activity could be observed (lower Graph) suggesting that MMP-2 and -9 activity is not a key event underlying the higher invasive capability developed by bladder cancer cells under hypoxia. Graphs represent average value of three different replicates and ***p < 0.001; **p < 0.01; *p < 0.05.
Figure 6
Figure 6. STn overexpression co-localizes with high nuclear HIF-1α expression areas
HIF-1α presents mainly a cytoplasmic expression and strong diffused expression troughout the cell when nuclear staining is present. (A) Advanced stage bladder tumour showing low levels of nuclear HIF-1α (low hypoxic tumour; 200× magnification). (B) Advanced stage bladder tumour showing high levels of nuclear HIF-1α (high hypoxic tumour; 200× magnification). (C) Advanced stage bladder tumour area showing high STn expression (100× magnification). (D) STn-positive advanced stage bladder tumour area with high nuclear HIF-1α expression (100× magnification).
Figure 7
Figure 7. Schematic representation of advanced stage hypoxic tumours showing increased STn expression
Accordingly, the Figure highlights the existence of a subpopulation of hypoxic cells showing high nuclear HIF-1α expression and also STn expression. These cells are endowed with the capability of invading the basal and muscle layers contributing to poor outcome.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, Parkin DM, Forman D, Bray F. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015;136:E359–386. - PubMed
    1. Stenzl A, Cowan NC, De Santis M, Kuczyk MA, Merseburger AS, Ribal MJ, Sherif A, Witjes JA, European Association of U Treatment of muscle-invasive and metastatic bladder cancer: update of the EAU guidelines. Eur Urol. 2011;59:1009–1018. - PubMed
    1. Brocato J, Chervona Y, Costa M. Molecular responses to hypoxia-inducible factor 1alpha and beyond. Mol Pharmacol. 2014;85:651–657. - PMC - PubMed
    1. Greer SN, Metcalf JL, Wang Y, Ohh M. The updated biology of hypoxia-inducible factor. EMBO J. 2012;31:2448–2460. - PMC - PubMed
    1. Wilson WR, Hay MP. Targeting hypoxia in cancer therapy. Nat Rev Cancer. 2011;11:393–410. - PubMed

MeSH terms