Tumor cell-derived placental growth factor sensitizes antiangiogenic and antitumor effects of anti-VEGF drugs

Abstract

The role of placental growth factor (PlGF) in modulation of tumor angiogenesis and tumor growth remains an enigma. Furthermore, anti-PlGF therapy in tumor angiogenesis and tumor growth remains controversial in preclinical tumor models. Here we show that in both human and mouse tumors, PlGF induced the formation of dilated and normalized vascular networks that were hypersensitive to anti-VEGF and anti-VEGFR-2 therapy, leading to dormancy of a substantial number of avascular tumors. Loss-of-function using plgf shRNA in a human choriocarcinoma significantly accelerated tumor growth rates and acquired resistance to anti-VEGF drugs, whereas gain-of-function of PlGF in a mouse tumor increased anti-VEGF sensitivity. Further, we show that VEGFR-2 and VEGFR-1 blocking antibodies displayed opposing effects on tumor angiogenesis. VEGFR-1 blockade and genetic deletion of the tyrosine kinase domain of VEGFR-1 resulted in enhanced tumor angiogenesis. These findings demonstrate that tumor-derived PlGF negatively modulates tumor angiogenesis and tumor growth and may potentially serve as a predictive marker of anti-VEGF cancer therapy.

Fig. 1.

Hypersensitivity of PlGF-expressing human JE-3 choriocarcinoma to anti-VEGF and anti-VEGFR therapy. (A) Growth rates of vehicle- and VEGF blockade–treated JE-3 tumors. T/C, treatment vs. control. Data show both tumor volumes and tumor weights (6 mice per group). (Scale bar, 1 cm.) (B) Representative micrographs of CD31-positive tumor vessels in vehicle- and VEGF blockade–treated JE-3 tumors. Red, CD31-positive signal; green, NG2-positive signal. Arrowheads point to vessels. (Scale bar, 50 μm.) (C) Quantification of microvessel numbers, branch points, pericyte coverage, and vessel diameter in vehicle- and VEGF blockade–treated JE-3 tumors (7–14 samples per group). (D) Growth rates of vehicle-, VEGFR-1 blockade–, and VEGFR-2 blockade–treated JE-3 tumors. Data show both tumor volumes and tumor weights (6 mice per group). ns, not significant. (Scale bar, 1 cm.) (E) Representative micrographs of CD31-positive tumor vessels in vehicle-, VEGFR-1 blockade–, and VEGFR-2 blockade–treated JE-3 tumors. Red, CD31-positive signal; green, NG2-positive signal. Arrowheads point to vessels. (Scale bar, 50 μm.) (F) Quantification of microvessel numbers, branch points, pericyte coverage, and vessel diameter in vehicle-, VEGFR-1 blockade–, and VEGFR-2 blockade–treated JE-3 tumors (4–9 samples per group).

Fig. 2.

Growth rates, angiogenesis, and responses of plgf shRNA– and control shRNA transduced JE-3 tumors to anti-VEGF therapy. (A) Tumor growth rates and tumor angiogenesis in plgf shRNA and control shRNA transduced JE-3 tumors (6–7 mice per group). In immunohistological panels, CD31-positive signals are presented in red, NG2-positive signals are in green, and tumor cells are in blue. Arrowheads point to vessels. (Scale bar, 50 μm.) (B) Tumor growth rates, tumor weights, and morphology of vehicle- and VEGFR-2–treated plgf shRNA– and control shRNA–transduced JE-3 tumors (5–8 mice per group). Dashed lines compare T/C of the treated and nontreated groups when tumor sizes are equal in PlGF and control tumors. (Scale bar, 1 cm.) (C) Quantitative PCR analysis of plgf and vegf mRNA levels in plgf shRNA– and control shRNA–transduced JE-3 tumor cells (3 samples per group). (D) Representative micrographs of CD31-positive tumor vessels in vehicle- and VEGFR-2 blockade–treated plgf shRNA and control shRNA JE-3 tumors. Red, CD31-positive signal; green, NG2-positive signal. Arrowheads point to vessels. (Scale bar, 50 μm.) (E) Quantification of microvessel numbers, branch points, pericyte coverage, and vessel diameter in vehicle- and VEGFR-2 blockade–treated plgf shRNA and control shRNA JE-3 tumors (6–16 samples per group).

Fig. 3.

PlGF-expressing mouse tumors in response to anti-VEGF therapy. (A) Growth rates, tumor weights, and morphology of vehicle- or VEGF blockade–treated vector T241 and PlGF T241 tumor-bearing fibrosarcomas (6 mice per group). (Scale bar, 1 cm.) (B) Growth rates, tumor weights, and morphology of vehicle-, VEGFR-1 blockade–, or VEGFR-2 blockade–treated vector T241 and PlGF T241 tumor-bearing fibrosarcomas (5–11 mice per group). Dashed lines compare T/C of the treated and nontreated groups when tumor sizes are equal in PlGF and control tumors. (Scale bar, 1 cm.)

Fig. 4.

Microvasculatures in vehicle- and anti-VEGF–treated vector T241 and PlGF T241 tumors in WT and Flt-1 TK^−/− mice. (A) Representative micrographs of CD31-positive tumor vessels in vehicle-, VEGF blockade–, VEGFR-1 blockade–, and VEGFR-2 blockade–treated vector T241 or PlGF T241 tumors. Red, CD31-positive signal; green, NG2-positive signal. Arrowheads point to vessels. (Scale bar, 50 μm.) (B) Quantification of microvessel numbers, branch points, pericyte coverage, and vessel diameter in vehicle-, VEGF blockade–, VEGFR-1 blockade–, and VEGFR-2 blockade–treated vector T241 or PlGF T241 tumors (8–11 samples per group). (C) Representative micrographs of CD31-positive tumor vessels in vehicle- and sunitinib-treated vector T241 or PlGF T241 tumors. Red, CD31-positive signal; green, NG2-positive signal. Arrowheads point to vessels. (Scale bar, 50 μm.) (D) Quantification of microvessel numbers, branch points, pericyte coverage, and vessel diameter in vehicle- and sunitinib-treated vector T241 or PlGF T241 tumors (4–11 samples per group). (E) Representative micrographs of CD31-positive tumor vessels in vector T241 or PlGF-tumors in WT or Flt-1 TK^−/− mice. Red, CD31-positive signal; green, NG2-positive signal. Arrowheads point to vessels. (Scale bar, 50 μm.) (F) Quantification of microvessel numbers, branch points, pericyte coverage, and vessel diameter in vector T241 and PlGF T241 tumors in WT (3–4 mice per group) or Flt-1 TK^−/− mice (3–4 mice per group) (6–13 samples per group).