Serum-stable RNA aptamers to urokinase-type plasminogen activator blocking receptor binding

Abstract

The serine proteinase urokinase-type plasminogen activator (uPA) is widely recognized as a potential target for anticancer therapy. Its association with cell surfaces through the uPA receptor (uPAR) is central to its function and plays an important role in cancer invasion and metastasis. In the current study, we used systematic evolution of ligands by exponential enrichment (SELEX) to select serum-stable 2'-fluoro-pyrimidine-modified RNA aptamers specifically targeting human uPA and blocking the interaction to its receptor at low nanomolar concentrations. In agreement with the inhibitory function of the aptamers, binding was found to be dependent on the presence of the growth factor domain of uPA, which mediates uPAR binding. One of the most potent uPA aptamers, upanap-12, was analyzed in more detail and could be reduced significantly in size without severe loss of its inhibitory activity. Finally, we show that the uPA-scavenging effect of the aptamers can reduce uPAR-dependent endocytosis of the uPA-PAI-1 complex and cell-surface associated plasminogen activation in cell culture experiments. uPA-scavenging 2'-fluoro-pyrimidine-modified RNA aptamers represent a novel promising principle for interfering with the pathological functions of the uPA system.

FIGURE 1.

Upanap-12 inhibits binding of uPA to uPAR. (A) SPR sensorgram showing the capture of 2 nM human uPA (injected from seconds 60 to 120) on a sensor surface with immobilized human uPAR in the presence of increasing concentrations of upanap-12. (B) The reported uPA capture level after injection in each case was plotted as a function of the upanap-12 concentration relative to the response of 2 nM uPA alone. A sequence-unrelated 2′-F-Y control RNA was analyzed for comparison. For IC₅₀-value, standard deviations, and number of independent experiments, see Table 1.

FIGURE 2.

The binding of upanap-12 to uPA requires the GFD. The relative association of upanap-12 to different variants of uPA is indicated: The inactive zymogen form of uPA (pro-uPA) active uPA (uPA), uPA without the growth factor domain (ΔGFD), and uPA without the entire ATF (also called low molecular weight uPA; LMW-uPA). The domain structure of the different variants is shown at each bar, while the inset illustrates pro-uPA with domain designations: (GFD) growth factor domain, (KD) kringle domain, and (SPD) serine protease domain. uPA variants were captured on an SPR sensor surface containing immobilized polyclonal anti-uPA antibody to the following levels: pro-uPA (215–280 RU), uPA (210–260 RU), ΔGFD-uPA (210–230 RU), and LMW-uPA (140–160 RU). The binding level of 50 nM upanap-12 to each captured variant was subsequently recorded. Using the molecular weight of the different variants, the binding level of aptamers per mole of captured variant was calculated and presented relative to the pro-uPA result. Open bars represent mean values and standard deviations derived from three independent experiments.

FIGURE 3.

Upanap-12 can inhibit cell binding of uPA. A total of 3 pM of ¹²⁵I-ATF (the GFD and KD of uPA) and different concentrations of upanap-12 were incubated with U937 cells overnight at 4°C, followed by γ-counting of cell pellets and supernatants. The ratio of cell-bound ¹²⁵I-ATF to free ¹²⁵I-ATF was plotted as a function of the concentration (in log scale) of upanap-12 (▪) or a sequence-unrelated 2′-F-Y RNA control (□) only analyzed at the highest concentration. Shown are mean result and standard deviations derived from three independent experiments.

FIGURE 4.

Full-length upanap-12 and truncation variants. (A) Conserved elements of the secondary structures predicted for upanap-12, upanap-21, upanap-25, upanap-71, and upanap-79, featuring the two stems, the asymmetrical loop, and the hairpin loop. Duplex regions are highlighted in gray, X indicates any base (A, G, C, or U), Py indicates a pyrimidine (C or U), and Pu indicates a purine (G or A). (B–D) The predicted secondary structures of (B) full-length upanap-12, (C) upanap-12.49, and (D) upanap-12.33. Note that the 5′- and 3′-ends for upanap-12.49 and upanap-12.33 have been mutated compared with the full-length sequence to allow efficient production of transcripts by T7 RNA polymerase.

FIGURE 5.

uPA binding aptamers can inhibit uPAR- and VLDLR-dependent endocytosis of uPA–PAI-1 complex in U937 cells. ¹²⁵I-labeled uPA–PAI-1 complexes were incubated with U937 cells for 3 h at 37°C in the presence of the indicated concentrations of upanap-12.49, a sequence-unrelated control 2′-F-Y RNA, or uPA. After incubation, the uPAR- and VLDLR-mediated degradation of the uPA–PAI-1 complex was determined by trichloroacetic acid precipitation and γ-counting. Only intact ¹²⁵I-uPA–PAI-1 protein can be precipitated by trichloroacetic acid, while smaller peptide fragments or amino acids will stay in solution. Shown are mean values and standard deviations derived from three independent experiments.

FIGURE 6.

The effect of upanap-12.49 on cell surface-asociated plasminogen activation. (A) Illustration of the experimental setup. Upon binding of pro-uPA to uPAR-expressing U937 cells, pro-uPA is activated and converts cell surface-associated plasminogen to plasmin, providing the cell surface with locally confined proteolytic activity. The presence of α₂-antiplasmin, a fast inhibitor of plasmin activity in solution, but not on the cell surface, ensures that the observed cleavage of a fluorogenic plasmin substrate is a result of cell surface-associated plasmin activity only. (B) Cells, plasminogen and α₂-antiplasmin were mixed and preincubated for 15 min at 37°C before the addition of pro-uPA and the plasmin fluorogenic substrate, either in the presence or absence of upanap-12.49, a sequence-unrelated control 2′-F-Y RNA, or suPAR. Cell surface-associated plasmin activity was measured by fluorescence emission at 480 nm, resulting from the cleaved fluorogenic substrate after subtraction of uPA-independent background cleavage (given as arbitrary units). Data points are mean values and standard deviations derived from three independent experiments.