Adenosine down-regulates the surface expression of dipeptidyl peptidase IV on HT-29 human colorectal carcinoma cells: implications for cancer cell behavior

Abstract

Dipeptidyl peptidase IV (DPPIV) is a multifunctional cell-surface protein that, as well as having dipeptidase activity, is the major binding protein for adenosine deaminase (ADA) and also binds extracellular matrix proteins such as fibronectin and collagen. It typically reduces the activity of chemokines and other peptide mediators as a result of its enzymatic activity. DPPIV is aberrantly expressed in many cancers, and decreased expression has been linked to increases in invasion and metastasis. We asked whether adenosine, a purine nucleoside that is present at increased levels in the hypoxic tumor microenvironment, might affect the expression of DPPIV at the cell surface. Treatment with a single dose of adenosine produced an initial transient (1 to 4 hours) modest (approximately 10%) increase in DPPIV, followed by a more profound (approximately 40%) depression of DPPIV protein expression at the surface of HT-29 human colon carcinoma cells, with a maximal decline being reached after 48 hours, and persisting for at least a week with daily exposure to adenosine. This down-regulation ofDPPIV occurred at adenosine concentrations comparable to those present within the extracellular fluid of colorectal tumors growing in vivo, and was not elicited by inosine or guanosine. Neither cellular uptake of adenosine nor its phosphorylation was necessary for the down-regulation of DPPIV. The decrease in DPPIV protein at the cell surface was paralleled by decreases in DPPIV enzyme activity, binding of ADA, and the ability of the cells to bind to and migrate on cellular fibronectin. Adenosine, at concentrations that exist within solid tumors, therefore acts at the surface of colorectal carcinoma cells to decrease levels and activities of DPPIV. This down-regulation of DPPIV may increase the sensitivity of cancer cells to the tumor-promoting effects of adenosine and their response to chemokines and the extracellular matrix, facilitating their expansion and metastasis.

Figure 1

Measurement of cell-surface DPPIV on HT-29 cells by indirect radioantibody binding assay. Confluent monolayer cultures of HT-29 cells were incubated with either anti-DPPIV mAb (total binding, •) or isotype control (nonspecific binding, ○), followed by ¹²⁵I-labeled anti-mouse Ig, F(ab′)₂ fragment as described in Materials and Methods. Points are mean values, n = 4. Standard errors fall within the symbols. The third curve (▪) shows the specific binding (total binding − nonspecific binding).

Figure 2

Adenosine modulation of DPPIV on HT-29 colorectal carcinoma cells. a and b: DPPIV protein on cell monolayers. The surface expression of DPPIV on HT-29 cells was measured after incubation in the absence or presence of adenosine (300 μmol/L) for the times indicated. The data are mean values ± SE, n = 3. *, Significant change because of adenosine, P < 0.05. **, P < 0.01. c: DPPIV protein on isolated cells. HT-29 cells were incubated for 48 hours in the absence (i) or presence (ii) of 300 μmol/L of adenosine and released by trypsinization. Cytofluorimetric profiles for HT-29 cells stained with isotype control mAb (open peaks) and with DPPIV-specific mAb (shaded peaks) are shown.

Figure 3

Failure of ADA loading to eliminate the decline in DPPIV because of adenosine. HT-29 cells were incubated for 48 hours in the absence (open bars) or presence (hatched bars) of adenosine (300 μmol/L), washed, and further incubated for 60 minutes at 4°C with 10 μg/ml of ADA beforethe binding assay for DPPIV. The data are mean values ± SE (n = 4). **, Significant reduction by adenosine, P < 0.01.

Figure 4

Dose dependence of adenosine down-regulation of DPPIV expression on HT-29 cells. a: Single dosing. HT-29 cells were incubated at 37°C with adenosine at the indicated concentrations for 48 hours and surface expression of DPPIV was then measured. b: Multiple dosing. HT-29 cells were incubated at 37°C with the indicated concentrations of adenosine (μmol/L) and number of doses (in parentheses) throughout the 48-hour treatment period. The data are mean values ± SE (n = 3). *, Significant reduction by adenosine, P < 0.05. **, P < 0.01.

Figure 5

Lack of effect of inosine and guanosine on the surface expression of DPPIV. HT-29 cells were incubated at 37°C with medium alone or in the presence of 100 μmol/L of adenosine, inosine, or guanosine. Forty-eight hours later cell-surface DPPIV was measured. The data are mean values ± SE (n = 4). **, Significant reduction by adenosine, P < 0.01.

Figure 6

Lack of requirement for adenosine uptake or adenosine phosphorylation for down-regulation of DPPIV. HT-29 cells were pretreated for 30 minutes with control vehicle or 10 μmol/L of dilazep (a); 1 μmol/L NBTI, 1 μmol/L dipyridamole (DP), or 1 μmol/L NBTI and DP (b); or 1 μmol/L iodotubercidin (IodoT) (c). The cells were then incubated with medium alone (open bars) or adenosine at 10 μmol/L (hatched bars) or 30 μmol/L (shaded bars), plus 2.5 μmol/L coformycin. Forty-eight hours later surface expression of DPPIV was measured. The data are mean values ± SE (n = 3). *, Significant reduction by adenosine, P < 0.05. **, P < 0.01.

Figure 7

Persistent depression of DPPIV protein and dipeptidase activity with long-term exposure to adenosine. HT-29 cultures were incubated in control media (open bars) or with adenosine (100 μmol/L) added daily (hatched bars). The cultures were then assayed for the presence of cell-surface DPPIV protein (a) or its associated dipeptidase activity (b). The data are mean values ± SE (n = 4). *, Significant reduction by adenosine, P < 0.05. **, P < 0.01.

Figure 8

The adenosine-induced down-regulation of DPPIV is accompanied by a decline in the ability of the cells to bind exogenous ADA. HT-29 cells were exposed for 48 hours to control conditions or with a single addition of 300 μmol/L of adenosine. Parallel cultures were then assayed for the presence of cell-surface DPPIV (a), or for their ability to bind exogenous ADA (b). The data are mean values ± SE (n = 3). **, Significant reduction by adenosine, P < 0.01.

Figure 9

The adenosine-induced down-regulation of DPPIV is accompanied by a decline in the ability of the cells to bind to cFN (a), and exhibit normal motility on cFN (b). HT-29 cells were treated with adenosine (300 μmol/L) or control vehicle as indicated, and tested for their ability to adhere to cFN (a) and to exhibit normal motility on a cFN substratum (b). Data are mean values ± SE with n = 4 (a) or n = 5 (b). **, Significant reduction by adenosine, P < 0.01.

Figure 10

Possible scheme for the involvement of adenosine-mediated DPPIV down-regulation in altering cancer cell behavior. a: Enhancement of the tumor-promoting functions of adenosine through reduced local levels of bound ecto-ADA. Adenosine levels in hypoxic tumors are high. Down-regulation of DPPIV leads to decreased ecto-ADA, adding to the rise in adenosine levels and therefore promoting events that facilitate tumor expansion. b: Implications of DPPIV down-regulation for other DPPIV functions. Down-regulation of DPPIV also leads to decreased dipeptidase activity and reduced binding to extracellular matrix (ECM) proteins. Increased chemokine activity and reduced ECM adherence, together with the effects of adenosine, may further act to enhance the spread of tumor cells.