This true experimental in vivo study was conducted in the Department of Anatomical Pathology, Faculty of Medicine, Universitas Indonesia, from September 2023 to September 2024. The tissue samples we used belonged to a previous study by Numlil Khaira Rusdi, “In Vivo Antimammary Tumor Effects of Soybean Extract with Targeted Lunasin (ET-Lun).”

Soybean of the Grobogan variety was bought from Indonesian Legumes and Tubers Crops Research Institute (Balitkabi), Malang, East Java. Lunasin extraction started by squeezing the seeds for 30 minutes at 100-150atm at 48.89°C. Then, the soybean was crushed to create a powder. The powder was liquefied using Phosphate Buffered Saline (PBS), five times the powder weight, at a pH of 7.4 for 60 minutes, followed by filtration using Whatman^TM 54. A vacuum rotary evaporator was used to create a thick extract from the filtrate. The lunasin concentration was determined using high-performance liquid chromatography (HPLC). A 100 mg extract of soybeans rich in lunasin was dissolved in 8 mL of distilled water. The solution was sonicated for 30 minutes, then distilled water was added until the concentration of the solution reached 10,000 ppm. Then, the solution was centrifuged for 30 minutes at 12,000 rpm, and the supernatant was collected. The filtrate was passed through a 0.22μm Millipore filter, resulting in a clear, colorless solution. The HPLC method was performed using acetonitrile as the mobile phase with a C18 column and a UV-Vis detector. The water had a ratio of 5:95, with a retention time of 35 minutes, a wavelength of 295nm, an injection volume of 20μL, and a mobile phase flow rate of 2mL/min. The extract with the solvent containing the highest solubility of lunasin was used in this study.¹³

Female SD rats aged 6 weeks were divided into 5 groups of 6 animals each: normal, negative control, positive control, lunasin, and combination. The sample size was determined using Federer’s formula, which calculated a minimum requirement of five rats per group, but six rats were used to account for potential mortality.¹³ DMBA was given intragastrically with a dose of 20 mg/kgBW twice a week until 11 total injections were administered to induce breast cancer in all groups except the normal group. Once the tumor volume reached 1-2 cm³, treatments were given for 8 weeks. The positive control group received tamoxifen 10 mg/kgBW, the lunasin group received soybean extract rich in lunasin, and the combination group received both tamoxifen and soybean extract rich in lunasin. The negative control group did not receive any treatment. Later, the rats were terminated, and tumor tissues were taken, processed, and embedded in paraffin blocks for immunohistochemical staining (IHC) following the protocol from Abcam.¹³

Tissue sample preparation involved several steps. First, fixation was performed by immersing breast cancer tissue sections in a formalin buffer solution (10% formalin in sodium acetate buffer, pH 7.0), followed by cutting the tissue to a thickness of 3-5 mm with a scalpel. Second, the tissue was dehydrated by soaking it in 96% ethanol for 30 minutes. This process was repeated five times, replacing the ethanol each time. Third, the tissue was cleared by immersing it for 15 minutes in xylene, followed by another 15 minutes in a second xylene solution. Finally, the tissue was placed in solid paraffin (melting point 60–70 °C) for 30minutes and then cooled in a freezer for 10 to 15minutes for impregnation and embedding. The paraffin block was stored at room temperature before being prepared for hematoxylin and eosin staining.¹³

Hematoxylin and eosin (HE) staining was performed to assess the breast cancer incidence in rats after DMBA induction. Tissue preparation of 4 μm was cut from the paraffin block and attached to slides. The samples were then deparaffinized in xylene I, II, and III for 5 minutes each. Next, the samples were rehydrated with 96% and 70% ethanol for 5 minutes each, followed by a rinse under running water for 5 minutes. The samples were stained in hematoxylin for 7 minutes, rinsed in running water for 10 minutes, and then dipped in saturated lithium carbonate for 1-2 minutes followed by another 5-minute rinse. If the blue color was insufficient, the samples were returned to hematoxylin for 2 minutes, rinsed, and then stained in eosin for 1-2 minutes. Dehydration was performed with 70%, 80%, and 96% ethanol, followed by clearing with xylene I, II, and III. Finally, a drop of Entellan, Sigma-Aldrich was added, and the samples were covered with a cover slip. The stained tissue samples were examined under a light microscope at 400× magnification. Positive HE staining was followed by immunohistochemical staining.¹³

Breast cancer tissue in paraffin blocks was cut transversely at 3μm thickness using a microtome. The samples were placed on glass slides with a poly-L-lysine coating and heated in an oven at 60°C overnight. To remove paraffin, the samples were dipped in xylene three times for 3 minutes each, followed by dehydration in 100%, 95%, and 70% ethanol for 3minutes, 2minutes, and 1minute, respectively. The samples were then immersed in 0.01 M sodium citrate (pH 6.0) in a microwave for 5 minutes. After that, 3% hydrogen peroxide was applied for 5minutes at room temperature to eliminate endogenous peroxidase.¹⁴ The samples were incubated with p21 cyclin antibody for 2hours at room temperature in PBS using a humidity chamber, followed by overnight incubation at 4°C. A negative control, N-Universal, was used. The samples were then incubated with a secondary antibody at room temperature for 1 hour, followed by 30 minutes with HRP-conjugated streptavidin. To visualize the protein, 3.3’-diaminobenzidine (DAB) was applied for 10minutes at room temperature. Finally, the samples were counterstained with Harris hematoxylin, dehydrated, and mounted.¹⁵

P21 protein expression was assessed using a weighted histoscore (H-Score) based on the intensity of the streaks and the percentage of stained cells. P21 protein intensity was gathered using ImageJ and IHC Profiler. The intensity score measurement ranged from 0–3 (0=no protein expression; 1=weak expression; 2=moderate expression; 3=strong expression). The H-score was obtained by multiplying the percentage value of the intensity score = (% of cell with 0 x 1) + (% of cell with 1 x 2) + (% of cell with 2x3) + (% of cell with 3x4).¹⁶

H-Score data in percentages were analyzed using SPSS 27 with a confidence interval of 95%. Normality testing was carried out using a Shapiro-Wilk test, and homogeneity was assessed using Levene’s test. Normal and homogeneous data (P>0.05) were subsequently processed using one-way parametric ANOVA. Post hoc Tukey’s HSD test was conducted to examine possible differences between treatment groups.

Cytoplasm expresses p21 protein in brown, with the color intensity aligning with the level of protein expression. A strong brown color in the cytoplasm indicates a higher p21 expression, while cells with a lower p21 expression have a light brown to blue color. As can be seen in Figure 1, the normal group (A) has a high expression of p21 marked by a high intensity of brown coloring. The negative control group (B) has a low p21 protein expression, shown by blue coloring in the cytoplasm.

The average p21 protein expression, written as an H-Score from highest to lowest, was found between the normal, combination, lunasin, positive control, and negative control groups. The lunasin group, which received only soybean extract rich in lunasin, and the combination group, given both soybean extract rich in lunasin and tamoxifen, saw an increase in p21 protein expression compared to the negative control group (P<0.05). There is no statistically significant difference in p21 protein expression between the positive control and negative control groups (P>0.05). (Table 1)

In this study, the normal group, which did not receive any treatment, showed significantly (P<0.05) higher p21 protein expression compared to all other treatment groups, except for the combination group. The activation of nuclear factor kappa beta (NF-κβ) by DMBA caused a reduction in protein expression in the negative and positive controls, as well as the adjuvant groups.

NF-κβ is a transcription factor for the cellular myelocytomatosis oncogene (c-Myc), which acts as a transcriptional repressor of p21 protein.¹⁷As a CDK inhibitor, p21 functions in cell cycle arrest, and its reduced expression results in uncontrolled proliferation and failure of apoptosis, leading to breast cancer in DMBA-induced rats.¹⁸

Data analysis showed that the average p21 protein expression in the tamoxifen-treated group is higher than in the negative control group, but the difference is not statistically significant (P>0.05). Tamoxifen has been a central part of endocrine therapy for breast cancer since the 1970s.

This drug, a triphenylethylene derivative classified as a selective estrogen receptor modulator (SERM), is used to treat ER-positive breast cancer, which accounts for 70% of breast cancer cases. Tamoxifen acts as a partial estrogen antagonist, interfering with estrogen receptor (ER) function by competing with estrogen for receptor binding. The binding of tamoxifen to ER prevents oncogene activation by estrogen, inhibiting the estrogenic effects involved in cancer cell proliferation and growth.^13,19 Tamoxifen’s anti-proliferative effect is believed to occur through the activation of the JNK and p38 MAP kinase signalling pathways, which then activate Egr-1 via Elk-1. The transactivation of Egr-1 has been shown to induce p21 protein expression, promoting cell cycle arrest in the G0/G1 phase.^20,21 However, various studies report resistance of ER-positive breast cancer to tamoxifen due to overexpression of receptor tyrosine kinases (RTK), which activates the PI3K-PTEN/Akt/mTOR pathway, reducing tamoxifen’s effectiveness.²² This may explain the lack of a statistically significant difference in p21 protein expression between the positive and negative control groups. To address the resistance problem, combination therapy with tamoxifen and other drugs should be considered in breast cancer treatment.

P21 protein expression in the lunasin group was significantly higher than in the negative control group. This is consistent with the findings of a study by Szymczyk et al. which found upregulation of p21 protein in breast cancer cells exposed to lunasin.^23,24 The increase in p21 protein occurs through the disruption of integrin signalling.^23,25 Integrins are receptors found on various cells that play a key role in signaling between the cytoskeleton and the extracellular matrix.²⁶ The RGD domain of lunasin competes with the extracellular matrix for binding to integrins, thereby disrupting signaling pathways involved in cell adhesion, invasion, and interaction with extracellular matrix components.²⁷ This mechanism contributes to lunasin’s recently observed ability to inhibit tumor progression and metastasis.^23,28 The lunasin-integrin binding also leads to downregulation of FAK/Akt/ERK phosphorylation, preventing MDM2-p53 interaction and suppressing c-Myc expression.^25,27 As a result, ubiquitination of p21 protein does not occur, preventing mRNA destabilization and negative regulation of p21. The increase in p21 protein expression induces cell apoptosis and cell cycle arrest in the G1/S phase, allowing time for DNA repair before DNA synthesis.^29,30 The increased p21 protein expression following the administration of soy extract rich in lunasin is also thought to occur through modulation of histone acetylation and deacetylation dynamics, leading to increased expression of the p21 gene.^23,31

Additionally, lunasin can inhibit nuclear factor kappa β (NF-κβ) activation by increasing the expression of inhibitor of nuclear factor kappa β alpha (Iκβ-α), a protein that binds NF-κβ in the cytoplasm, preventing its translocation to the nucleus.^25,32,33 This results in reduced cell proliferation and the potential inhibition of cancer cell migration and invasion.³²

When comparing p21 protein expression between the positive control group and the lunasin group, we found that the latter showed higher expression, though the difference was not statistically significant. This may be due to the relatively small sample size, making it harder to detect true differences. Additionally, the high standard deviation observed suggests biological variability.³⁴ A larger sample size may be required to better distinguish lunasin from tamoxifen and confirm its impact on p21 expression.

A commonly used strategy in cancer therapy is the combination of two or more drugs to achieve a synergistic effect in cancer prevention and treatment.³⁵ In this study, data analysis showed that p21 protein expression in the combination group was significantly higher than in the negative control group. Additionally, the average p21 protein expression was also higher in the combination group compared to the positive control and lunasin groups, although the differences were not statistically significant. This may occur because tamoxifen and lunasin act through distinct signaling pathways to regulate p21 expression. Tamoxifen exerts its effect via Egr-1 activation, while lunasin functions through integrin signalling.²² Given that tamoxifen resistance is often driven by PI3K/Akt/mTOR pathway activation, lunasin’s ability to downregulate Akt suggests a role in overcoming this resistance.^22,25,27 By disrupting integrin signaling and inhibiting FAK/Akt/ERK phosphorylation, lunasin may help sustain p21 expression and counteract resistance mechanisms in breast cancer.³¹

Additionally, reduced p21 and p27 expression has been linked to increased tamoxifen resistance.³⁶ Since lunasin upregulates both proteins, its co-administration with tamoxifen may further help mitigate resistance and prolong treatment efficacy. Through these complementary mechanisms, ie, Akt downregulation and p21/p27 upregulation, lunasin may enhance tamoxifen’s therapeutic effects, potentially improving treatment outcomes.^20,25,36

The bioavailability of food-derived bioactive compounds depends on their stability during absorption, distribution, and metabolism.³⁷ However, therapeutic peptides, including lunasin, face challenges due to enzymatic degradation and limited gastrointestinal permeability.³⁸

In vitro studies show that lunasin undergoes significant degradation, although soybean-derived protease inhibitors, such as Bowman-Birk and Kunitz trypsin inhibitors, may enhance its stability and bioavailability.³⁷ Lunasin crosses the Caco-2 cell monolayer via paracellular diffusion, with 4.5% detected in human plasma and 30% absorbed in mice.^39,40 Once in circulation, it is distributed to tissues, crosses the blood-brain barrier, and enters cells despite its hydrophilic nature.⁴⁰ Its bioactivity depends on evading proteolysis, necessitating further stability assessments for therapeutic applications.⁴¹

Animal studies suggest that lunasin selectively targets cancer cells without adverse effects.⁴⁰ Treated mice showed no significant physiological or biochemical changes, including liver enzymes and creatinine levels.⁴² Despite crossing the blood-brain barrier, lunasin did not affect mouse behavior, although further neurological evaluations are needed.^41,43 Hsieh et al. found that lunasin inhibited breast cancer cell proliferation in a dose-dependent manner without affecting normal cell growth.³²

Rusdi et al. assessed the acute and sub-chronic toxicity of soybean seed extract and lunasin-targeted extract (ET-Lun). No mortality was observed at doses of 500, 2000, and 5000 mg/kgBW over 14 days. Sub-chronic toxicity testing showed no significant differences in serum glutamate oxaloacetate transaminase (SGOT) and serum glutamate pyruvate transaminase (SGPT) levels between control and ET-Lun-treated groups (250–750mg/kgBW). Histopathological analysis confirmed normal liver architecture, with no signs of hepatotoxicity.¹⁰ While lunasin shows promising results, large-scale human trials are needed to confirm its safety and efficacy in cancer treatment.