Hydroalcoholic Extracts and Cancer Cell Toxicity in In Vitro Barrier Models

Authors: Marcel Benadiba, Raphael Serruya, and Dr. Yehoshua Maor
Institution: Phytor Lab for Drug Development, The Hadassah Medical Center, Hebrew University Biotechnology Park (JBP), Jerusalem, Israel
Contact:
✉️ Corresponding author: Yehoshua Maor

Abstract

Modern chemotherapeutic compounds often damage healthy cells and diminish patient quality of life.
To improve cancer treatment selectivity, natural products remain a key source of potential therapeutic molecules.

In this study, hydroalcoholic extracts composed of Convolvulus arvensis, Boswellia serrata, and Curcuma longa—each in different ratios—were tested on human cancer cell lines using advanced in vitro barrier models.

The data revealed that specific extract combinations can significantly modulate cytotoxicity and selectivity in distinct cancer types, particularly in colon, pancreatic, and neuroblastoma cell lines.

Keywords: Gut-pancreatic axis, Blood-brain barrier, Curcuma longa, Convolvulus arvensis, Boswellia serrata
Abbreviations: TJ – Tight Junctions; GPA – Gut-Pancreatic Axis; BBB – Blood-Brain Barrier; TEER – Transendothelial Electrical Resistance

1. Introduction

Convolvulus arvensis (Bindweed)

A perennial medicinal plant used in traditional Islamic and Persian medicine for hepatic cleansing and fever reduction.
It has shown antioxidant, anticancer, and stimulatory activity. However, literature reveals limited scientific data about its anticancer mechanism.

Boswellia serrata (Frankincense)

Traditionally used for respiratory and inflammatory disorders, Boswellia exhibits antibacterial, antifungal, and anticarcinogenic effects.
Studies report efficacy against multiple cancers (prostate, pancreatic, breast, brain, leukemia, melanoma), yet its mechanism remains partly undefined.

Curcuma longa (Turmeric)

Curcumin, the main bioactive compound, has anti-inflammatory, antioxidant, anti-angiogenic, and apoptotic properties.
It affects disorders such as arthritis, diabetes, and several cancers.
However, its low bioavailability in vivo limits therapeutic application.

2. Materials and Methods

2.1. Preparation of Hydroalcoholic Extracts

• 20g of dried plant material macerated in 100 ml of 25% ethanol solution for one month.

• Extracts mixed in distinct ratios and labeled as #1, #2, #3, and #5.

• Further processed by ultrasonic dispersion at 40°C, 100 ml/min, and stored in a cool, dry place.

2.2. Cell Lines

• Aspc1: Human pancreatic cancer

• Caco-2: Human colon cancer

• SH-SY5Y: Neuroblastoma

• BJ cells: Normal epithelial fibroblasts

All cultured under standard sterile conditions in RPMI or DMEM medium supplemented with serum and antibiotics.

2.3. MTT Viability Assay

• Cells plated in 96-well plates at 1×10⁴ cells/well.

• Treated with concentrations from 0.05 to 7.5 mg/ml of extracts #2, #5, and mix 2/5 (1/3 #2 + 2/3 #5).

• Compared with Irinotecan and Carboplatin (15 µM each).

• Absorbance measured at 570 nm after 72 h incubation.

2.4. TEER Assay

Used to assess monolayer integrity across gut-pancreatic and BBB models.
Caco-2, HUVEC, Aspc1, and U87MG astrocyte cells were co-cultured on Transwell membranes.
Electrical resistance was measured with Millicell Ohm meters.

2.5. HPLC Profiling

High-performance liquid chromatography identified distinct chromatographic profiles for each extract.
Gradient: 50% methanol → 95% over 45 min; detection at 315 nm, column temp 37°C.

2.6. Statistical Analysis

All results analyzed with Student’s t-test (p < 0.05) for significance.

3. Results and Discussion

3.1. Gut–Pancreatic Model

• Tested concentrations: 1.8 – 0.05 mg/ml.

• EC₅₀ for:

  • #2: 0.9 mg/ml

  • #5 and Mix 2/5: 1.5 mg/ml

➡️ #2 was more potent than #5 and Mix 2/5 against Aspc1 (pancreatic cancer).
Both surpassed Irinotecan and Carboplatin, which reduced viability by only ~30%.

Mix 2/5 disrupted the Caco-2 gut monolayer, while #2 and #5 strengthened it (higher TEER).
Despite this, their small molecules still crossed the barrier and inhibited pancreatic cancer growth.

Figure 3 (page 10) shows TEER results confirming Mix 2/5’s ability to cross the gut barrier and target Aspc1 cells.

3.2. Selectivity to Normal Cells

BJ fibroblasts exhibited much lower toxicity, confirming selectivity for cancerous cells.
This indicates potential for colon-cancer-specific therapeutic application.

3.3. Blood–Brain Barrier (BBB) Model

Tested combinations:

• #1 + #2 (50/50)

• #1 + #5 (50/50)

• Mix 5/2 (inverted mix) = 2/3 #2 + 1/3 #5

Findings:

• Mix 5/2: Crossed the BBB and reduced SH-SY5Y neuroblastoma proliferation without damaging barrier integrity.

• #1 + #2 and #1 + #5: Increased TEER (tightened barrier), thus preventing molecule passage.

• 0.4 mg/ml concentration proved ideal for maintaining BBB structure while achieving selective cytotoxicity.

3.4. Mechanism of Action

Capsule forms (#2, #3, #5) were analyzed by Western blot for E-cadherin, β-catenin, and ERK1/2: