Comparison of efficacy and toxicity of traditional Chinese medicine (TCM) herbal mixture LQ and conventional chemotherapy on lung cancer metastasis and survival in mouse models

Abstract

Unlike Western medicine that generally uses purified compounds and aims to target a single molecule or pathway, traditional Chinese medicine (TCM) compositions usually comprise multiple herbs and components that are necessary for efficacy. Despite the very long-time and wide-spread use of TCM, there are very few direct comparisons of TCM and standard cytotoxic chemotherapy. In the present report, we compared the efficacy of the TCM herbal mixture LQ against lung cancer in mouse models with doxorubicin (DOX) and cyclophosphamide (CTX). LQ inhibited tumor size and weight measured directly as well as by fluorescent-protein imaging in subcutaneous, orthotopic, spontaneous experimental metastasis and angiogenesis mouse models of lung cancer. LQ was efficacious against primary and metastatic lung cancer without weight loss and organ toxicity. In contrast, CTX and DOX, although efficacious in the lung cancer models caused significant weight loss, and organ toxicity. LQ also had anti-angiogenic activity as observed in lung tumors growing in nestin-driven green fluorescent protein (ND-GFP) transgenic nude mice, which selectively express GFP in nascent blood vessels. Survival of tumor-bearing mice was also prolonged by LQ, comparable to DOX. In vitro, lung cancer cells were killed by LQ as observed by time-lapse imaging, comparable to cisplatinum. LQ was more potent to induce cell death on cancer cell lines than normal cell lines unlike cytotoxic chemotherapy. The results indicate that LQ has non-toxic efficacy against metastatic lung cancer.

Figure 1. Efficacy of LQ on tumor size, growth and weight in subcutaneous nude mouse tumor models.

For the H460 and A549 cell lines, each treatment group contained 6 mice and 10 mice were used for the untreated control groups. For the LLC cell line, each treatment group contained 8 mice and 16 mice were used for the untreated control group. After the subcutaneous tumors grew, the nude mice were given either CTX (30 mg/kg/day i.p.) for 7 days or PX (600 mg/kg/day p.o) for 10 days. PBS (po) was used in the control group. Mice were treated at 150, 300, and 600 mg/kg/day of LQ (po) for 10 days. Tumor size was measured twice a week. Tumor weight was measured at the endpoint. Statistical significance between groups was determined with the Student’s t-test. H460 (A&D), A549 (B&E), and LLC (C&F) had growth inhibition and tumor size inhibition after treatment with all agents. The tumor weight was also significantly inhibited by all agents (G) (p<0.01). After LQ treatment, lung tumor growth and tumor weight were significantly inhibited, compared to the untreated control group in all models (p<0.01). CTX and PX treatment resulted in a significant inhibition of tumor weight (P<0.01) compared to the control group for all cell lines. LQ had more efficacy than PX (p<0.05) on all cell lines (G). CTX induced loss of body weight (p<0.05), but not LQ or PX (H).

Figure 2. Toxicity of treatment on the renal cortex.

(A) Renal cortex of untreated mouse. (B) The renal cortex of mice treated with CTX demonstrated renal toxicity. Toxicity was mainly in the proximal tubular cells, including degeneration, obstruction, necrosis and swelling. (C) The renal cortex of mice treated with LQ had no toxicity.

Figure 3. Efficacy of LQ on tumor volume (A), tumor weight (B) and body weight (C) in subcutaneous mouse tumor models of H460-GFP.

Five mice were used in each group. After the subcutaneous tumors grew to 100 mm², the nude mice were given either DOX (7.5 mg/kg i.v.) (twice a week for 3 times). PBS (po), LQ (600 mg/kg daily po). Tumor size and body weight were measured every day. Tumor weight was measured at the endpoint. Statistical significance between groups was determined with the Student’s t-test. After LQ treatment, lung tumor growth and tumor weight were significantly inhibited (P<0.01), compared to the untreated control group. DOX treatment also resulted in a significant tumor growth and weight inhibition compared to the untreated controls (P<0.01). The DOX-treated mice had significant body-weight loss compared to the LQ-treated and untreated controls (P<0.01). *p<0.05; **p<0.01. Images of H460-GFP tumors in live mice from the subcutaneous nude-mouse model are shown for the different treatment groups, and untreated control (D); LQ (E); and DOX (F).

Figure 4. Efficacy of LQ on tumor volume, tumor weight (A) and body weight (B) in orthotopic H460-GFP lung cancer nude-mouse models.

Ten mice were used in each group. H460-GFP fragments from tumors grown subcutaneously (1 mm in diameter) were harvested and implanted by surgical orthotopic implantation (SOI) into the left lungs of nude mice. The nude mice were given either PBS (po); DOX (7.5 mg/kg i.v. twice a week for 3 times from day 7) or LQ (600 mg/kg/day po daily. from day 7). Body weights were measured every day. Tumor weights were measured at the endpoint. Statistical significance between groups was determined with the Student’s t-test. Lung tumor growth was significantly inhibited by LQ (p<0.01), without body weight loss. DOX inhibited lung tumor growth (p<0.001) but induced significant body weight loss compared to control mice (p<0.001). Open images of H460-GFP tumors in the orthotopic nude mouse model are shown for the different treatment groups and untreated control (C); LQ (D); and DOX (E).

Figure 5. Efficacy of LQ on experimental lung metastasis.

Six mice were used in each group. Individual images of excised lungs from the mice in the experimental metastasis model were obtained. LLC-RFP cancer cells (2×10⁶) were injected into the tail vein of nude mice or C57BL/6 mice. From day 2, the mice were given PBS (po) every day as the untreated control. Mice were treated with LQ (600 mg/kg/day po) for 10 days. The LQ-treated mice had significantly reduced lung weight (p<0.001) and reduced red fluorescence area in the lung (p<0.001) compared to the untreated PBS controls. A and B are images of the lungs. C–D are lung weights. E is the total red fluorescence area.

Figure 6. Survival curve of LQ-treated mice with H460-GFP orthotopic human lung cancer (A), and mice with LLC experimental lung metastasis (B).

Ten mice were used in each group. LQ significantly increased survival in both the orthotopic (p = 0.015) and experimental metastasis (p = 0.001) models. DOX also significantly increased (p<0.001) survival in both models.

Figure 7. Effect of LQ on tumor blood vessels.

LLC-RFP cells were grown in nestin-driven GFP (ND-GFP) transgenic nude mice in which nascent blood vessel expressed GFP. Three mice were used in each group. In the LQ-treated mice, tumor blood vessels appeared to be destroyed. The GFP florescence of blood vessels was quantitated for area, density, integrated optical density (IOD), length and counts. The measurements indicated that LQ decreased blood vessels by 50%.

Figure 8. Time-Lapse imaging of H460 dual-color cells expressing GFP in the nucleus and RFP in the cytoplasm treated with LQ and CDDP.

A and B) untreated control cells at 0 hour and 36 hours. C) CDDP-treated cells at 36 hours. D) time lapse imaging of LQ-treated H460 cells at 4, 24 and 72 hours. Cell proliferation was inhibited. Cell proliferation indicated by yellow arrows. Apoptotic cells indicated by green arrows. Blue and red arrows indicate changes in the same field.

Figure 9. Efficacy of LQ on proliferation of cancer and normal cells in vitro.

Cells were seeded in 96-well plates and incubated overnight. Cells were then exposed to LQ. H460, LLC and A549 were sensitive to LQ at 72 h (A). IC₅₀ values were between 0.36∼0.45 mg/ml. All experiments showed a dose-dependent inhibition by LQ. However, LQ was less active on the normal cell lines HUVEC and VERO (A). The cancer and normal cells were all sensitive to DOX and CDDP (B. C). However, HUVEC and VERO cells were more sensitive to paclitaxol than the cancer cells (D). The graphs show combined values from two independent experiments, with each data point repeated in triplicate.