Dopamine D2 receptor agonists abrogate neuroendocrine tumour angiogenesis to inhibit chemotherapy-refractory small cell lung cancer progression

Abstract

Small cell lung cancer (SCLC) is difficult to treat due to its aggressiveness, early metastasis, and rapid development of resistance to chemotherapeutic agents. Here, we show that treatment with a dopamine D₂ receptor (D₂R) agonist reduces tumour angiogenesis in multiple in vivo xenograft models of human SCLC, thereby reducing SCLC progression. An FDA-approved D₂R agonist, cabergoline, also sensitized chemotherapy-resistant SCLC tumours to cisplatin and etoposide in patient-derived xenograft models of acquired chemoresistance in mice. Ex vivo, D₂R agonist treatment decreased tumour angiogenesis through increased apoptosis of tumour-associated endothelial cells, creating a less favourable tumour microenvironment that limited cancer cell proliferation. In paired SCLC patient-derived specimens, D₂R was expressed by tumour-associated endothelial cells obtained before treatment, but D₂R was downregulated in SCLC tumours that had acquired chemoresistance. D₂R agonist treatment of chemotherapy-resistant specimens restored expression of D₂R. Activation of dopamine signalling is thus a new strategy for inhibiting angiogenesis in SCLC and potentially for combatting chemotherapy-refractory SCLC progression.

Fig. 1. Activation of D₂R signalling by quinpirole reduces tumour growth in a human small cell lung tumour xenograft model.

a Experimental timeline: SCID mice were orthotopically injected with luciferase-labelled human DMS-53 SCLC cells, then imaged for bioluminescence 7 days after the injection but before the start of treatment. Mice then received intraperitoneal injections of PBS vehicle (control group) or 10 mg/kg quinpirole (treatment group) every other day for 13 days, after which they were imaged again for bioluminescence. b Distribution of tumour growth rate across the vehicle- and quinpirole-treated groups. Each circle represents an individual mouse. Data are shown as mean ± SEM. A value of P ≤ 0.05 (two-way unpaired t-test) was considered significant. c Representative luciferase imaging from day 7 (before treatment with D₂R agonist) and day 21 (after treatment with D₂R agonist).

Fig. 2. D₂R agonist treatment reduces tumour growth and promotes apoptosis of tumour-associated endothelial cells in a chemonaïve human SCLC PDX model.

a–d NSG mice were subcutaneously injected with 5 × 10⁶ human SCLC cells originally derived from a chemonaïve patient (MSK-LX40) [63]. Mice were treated with either vehicle (10% DMSO in PBS) or cabergoline (5 mg/kg) daily for 14 days once mean tumour volume reached 100–200 mm³. Tumour growth was recorded by measuring tumour volume with callipers three times a week (a). Experiments were concluded before average tumour volume exceeded 1500 mm³. At the endpoint of the experiments, mice were euthanized, and xenografted tumours were harvested and weighed using a digital balance (b). Tumour volume was calculated from calliper measurements after extirpation (c). Each circle represents an individual mouse (b, c). Images of harvested tumours were acquired (d). e Co-immunofluorescence staining was performed on formalin-fixed, paraffin-embedded (FFPE) tissues harvested from NSG mice (n = 3 mice per group) treated with either vehicle or cabergoline to detect both the TUNEL⁺ and CD31⁺ cells in the tumour microenvironment. f The colocalization of TUNEL and CD31 staining was quantified by counting the number of double-positive cells (i.e., TUNEL⁺ and CD31⁺ cells) per CD31⁺ blood vessel. Each circle represents one visual field (≥5 visual fields were counted for each tissue). g Immunofluorescence staining was performed using a monoclonal Ki-67 antibody on FFPE tumour tissue specimens harvested from chemonaïve human SCLC tumour–bearing mice treated with vehicle (left) or 5 mg/kg cabergoline (right). h The proliferation of tumour cells within tissues harvested from cabergoline- and vehicle-treated mice (n = 3 for each group) was quantified by counting the number of Ki-67-positive cells per visual field ( ≥ 5 visual fields were counted for each tissue). Data are shown as mean ± SEM. A value of P ≤ 0.05 (two-way unpaired t-test) was considered significant. i, j HUVEC treated with vehicle (DMSO) or 100 μM cabergoline for 24 h were transferred to a 1% O₂ hypoxia chamber for an additional 6 h (i) or treated with hypoxia-inducer 150 μM CoCl₂ for an additional 24 h (j). Lysates were collected and immunoblotted for HIF1α and reference protein, GAPDH.

Fig. 3. Treatment with D₂R agonist and chemotherapy overcomes chemoresistance in human SCLC PDX models.

Five million human tumour cells harvested from chemoresistant SCLC PDXs (MSK-LX40R (a), JHU-LX108R (b), and JHU-LX33R (c)) were subcutaneously injected into the right flanks of NSG mice. When tumour volume reached 200 mm³ in size, mice were randomly divided into treatment groups such that each group had similar mean tumour volumes. Mice were treated by intraperitoneal injection of 5 mg/kg cisplatin on day 1 and 8 mg/kg etoposide on days 1, 2, and 3 with or without 5 mg/kg cabergoline daily for the indicated times. Tumour volume was measured 2–3 times per week until the final tumour reached either five times the initial tumour volume (a) or 2000 mm³ in size (b, c). d–f At the endpoint, the extirpated tumours were weighed on a digital scale. g–i The final volume of the tumours from euthanized mice was measured using a digital calliper. j–l Harvested tumours from each group were arranged randomly and photographed. Each circle represents a tumour harvested from an individual mouse. Data are shown as mean ± SEM. A value of P ≤ 0.05 obtained with two-way ANOVA followed by Sidak’s multiple test (a–c) or two-way unpaired t-test (d–i) was considered significant.

Fig. 4. D₂R agonist treatment together with chemotherapy decreases tumour cell proliferation and promotes endothelial apoptosis in a chemoresistant SCLC PDX model.

a FFPE tumour tissues (n = 3 per group) derived from mice treated with either chemotherapy alone or both cabergoline and chemotherapy were used for a co-immunofluorescence study to detect double-positive TUNEL⁺ and CD31⁺ cells in the tumour microenvironment. b To measure D₂R agonist-induced apoptosis in tumour-associated blood vessels, colocalization of TUNEL and CD31 staining was quantified in FFPE tumour tissues (n = 3 per group) harvested from mice bearing MSK-LX40R SCLC PDXs. Each circle represents one visual field, and five visual fields were counted for each tissue. c Immunofluorescence using monoclonal Ki-67 antibody was performed on FFPE tumour tissues (n = 5 per group) obtained from a chemoresistant human SCLC PDX model (MSK-LX40R) treated with 5 mg/kg cisplatin on day 1 and 8 mg/kg etoposide on days 1, 2, and 3 with or without 5 mg/kg cabergoline daily. Scale bar, 50 µm. d. The number of Ki-67-positive cells from the experiment illustrated in (c) was quantified. Each circle represents one visual field, and five visual fields were counted for each tissue. Data are shown as mean ± SEM. A value of P ≤ 0.05 was considered significant, two-way unpaired t-test.

Fig. 5. Activation of D₂R signalling in the SCLC tumour microenvironment promotes apoptosis of SCLC organoids and contributes to an enhanced CD8⁺ T cell response.

a, b Conditioned medium from human endothelial cells treated with a D₂R agonist promotes apoptosis of SCLC organoids. a The SCLC PDX JHU-LX33R was suspended in Matrigel and plated at a density of 1 × 10⁵ cells in 50 µL. Complete growth medium was added to the PDXs, which were grown in a humidified chamber at 37 °C supplied with 5% CO₂. After 14 days, the Matrigel domes were dissolved by adding dispase (1 U/mL) containing cold complete medium, followed by TrypLE, then the cells were replated at a density of 10⁴ cells per well of a 96-well plate in Matrigel and topped off with 100 µL conditioned medium harvested from HUVEC that had been stably transduced with plasmids encoding control LacZ-specific shRNA (shControl CM) or D₂R-specific shRNA (shD₂R CM) and subsequently treated with a D₂R agonist (quinpirole). b Conditioned medium collected from shControl HUVEC treated with the D₂R agonist quinpirole (50 µM) increased the caspase-3-mediated apoptotic response in the SCLC chemotherapy-resistant organoid model. c, d Treating human SCLC PDX organoids with D₂R agonist quinpirole contributes to an enhanced CD8⁺ T cell response in the SCLC immune microenvironment. c JHU-LX33R SCLC PDX organoids were treated with D₂R agonist quinpirole (50 µM) for 72 h, and conditioned medium from the organoids was collected and placed on human CD8⁺ T cells. d Human CD8⁺ T cells were cultured in growth medium containing 100 IU/mL IL-2 (IL-2: negative control) or a T cell activation cocktail of 100 IU/mL IL-2, CD3 antibody, and CD28 antibody (Activation: activation cocktail only; Vehicle CM: activation cocktail and vehicle-treated SCLC PDX conditioned medium; Quinpirole CM: activation cocktail and 50 µM quinpirole-treated SCLC PDX conditioned medium; Quinpirole: activation cocktail and 50 µM quinpirole). The human T cells were subjected to flow cytometry using fluorophore conjugated antibodies to detect granzyme B and results have been plotted in a bar graph. Micrograph (right of the bar graph) showing the gating strategy for identification of CD8⁺ T cells expressing Granzyme B. Two independent biological repeats and 4 technical repeats were performed. One-way ANOVA followed by Tukey’s multiple comparisons test was performed.

Fig. 6. D₂R is expressed by tumour-associated endothelial cells derived from SCLC patients.

a, b Immunohistochemistry was performed using a monoclonal D₂R antibody on FFPE human SCLC-A tissue samples biopsied from each patient before the start of the chemotherapy (i.e., chemonaïve) and following progressive disease after chemotherapy (i.e., chemoresistant). Representative images from paired chemonaïve (a) and chemoresistant (b) SCLC tumour specimens used to assess D₂R expression in tumour-associated endothelial cells before and after the chemotherapy. c Each tissue was scored for the percentage of endothelial cells that were D₂R-positive, and each circle on the graph represents an individual tissue (n = 9). Data are shown as mean ± SEM. A value of P ≤ 0.05 (two-way unpaired t-test) was considered significant.

Fig. 7. Endothelial D₂R expression is increased by D₂R agonist treatment in a chemoresistant SCLC PDX model.

a–c NSG mice were subcutaneously implanted with 5 × 10⁶ cells obtained from chemoresistant (MSK-LX40R) human SCLC PDXs. Mice were randomly divided into four groups: 1) vehicle; 2) cisplatin/etoposide; 3) cabergoline; and 4) cabergoline & cisplatin/etoposide. When mean tumour volume reached 200 mm³, mice were intraperitoneally administered vehicle (10% DMSO in PBS); 5 mg/kg cisplatin on day 1 and 8 mg/kg etoposide on days 1, 2, and 3, with or without 5 mg/kg cabergoline; or 5 mg/kg cabergoline alone five times a week. Tumour volume was measured three times a week until the vehicle-treated final tumour volume reached 2000 mm³ in size (a). At the endpoint, tumours were harvested from euthanized mice and photographs were taken to visualize gross morphology (b). Weight of the extirpated tumours was recorded using a digital scale (c). d Co-immunofluorescence staining was performed on FFPE tumour tissues (n = 3 per group) using primary antibodies against CD31 and D₂R. The number of endothelial cells expressing D₂R was quantified by counting the number of double-positive cells (i.e., D₂R⁺ and CD31⁺) divided by the CD31⁺ blood vessels present in a single visual field (≥5 visual fields were counted for each tissue). Data are shown as mean ± SEM. A value of P ≤ 0.05 (two-way unpaired t-test) was considered significant.