*Article* **Fermented Mangosteen (***Garcinia mangostana* **L.) Supplementation in the Prevention of HPV-Induced Cervical Cancer: From Mechanisms to Clinical Outcomes**

**Zaira Kharaeva 1, Pavel Trakhtman 2, Ilya Trakhtman 3, Chiara De Luca 4, Wolfgang Mayer 4, Jessie Chung 5, Galina Ibragimova <sup>6</sup> and Liudmila Korkina 3,6,\***


**Simple Summary:** Human papillomavirus (HPV) is connected with virtually all cases of cervical cancer. The viral infection-associated chronic inflammation, oxidative stress, and alterations in apoptosis have been considered as leading risk factors for carcinogenesis in humans. In an observational clinical study, we identified oxidative markers and the cervical/circulating ligands of TNF-alpha-induced apoptosis involved in HPV-associated cervical carcinogenesis. In the following clinical trial, 250 females infected with high-cancer-risk HPV16/18 (healthy and pre-cancerous) were recruited into a placebo-controlled clinical study of supplementation with fermented mangosteen (FM, 28g/day, daily) for three months. Our findings indicate that FM, and not a placebo, in combination with routine anti-viral therapy, could prevent, slow down, or even interrupt HPV-associated cervical carcinogenesis, mainly through the suppression of leukocyte recruitment into infected tissue, through anti-inflammatory effects, and through the restoration of nitric oxide metabolite-initiated TRAIL-dependent apoptosis.

**Abstract:** In the observational clinical study, we identified the oxidative markers of HPV-associated cervical carcinogenesis and the local/circulating ligands of TNF-alpha-induced apoptosis. Cervical biopsies of 196 females infected with low-cancer-risk HPV10/13 or high-cancer-risk HPV16/18 (healthy, pre-cancerous CIN I and CIN II, and CIN III carcinoma) were analysed for OH radical scavenging, catalase, GSH-peroxidase, myeloperoxidase (MPO), nitrate/nitrite, nitrotyrosine, and isoprostane. Ligands of TNF-alpha-dependent apoptosis (TNF-alpha, TRAIL, IL-2, and sFAS) were determined in cervical fluid, biopsies, and serum. Cervical MPO was highly enhanced, while nitrotyrosine decreased in CIN III. Local/circulating TRAIL was remarkably decreased, and higher-than-control serum TNF-alpha and IL-2 levels were found in the CIN I and CIN III groups. Then, 250 females infected with HPV16/18 (healthy and with CIN I and CIN II) were recruited into a placebo-controlled clinical study of supplementation with fermented mangosteen (FM, 28g/day, daily) for three months. Post-trial colposcopy revealed normal patterns in 100% of the FM group versus 62% of the placebo group. Inflammatory cells in cervical fluid were found in 21% of the FM group versus 40% of the placebo group. Locally, FM drastically diminished MPO and NO2/NO3, while it remarkably increased TRAIL. Additionally, FM supplementation normalised serum TRAIL, TNF-alpha, and IL-2.

**Keywords:** apoptosis; cervical carcinogenesis; fermented mangosteen; HPV; IL-2; myeloperoxidase; nitric oxide metabolites; sFAS; TRAIL; TNF-alpha

**Citation:** Kharaeva, Z.; Trakhtman, P.; Trakhtman, I.; De Luca, C.; Mayer, W.; Chung, J.; Ibragimova, G.; Korkina, L. Fermented Mangosteen (*Garcinia mangostana* L.) Supplementation in the Prevention of HPV-Induced Cervical Cancer: From Mechanisms to Clinical Outcomes. *Cancers* **2022**, *14*, 4707. https://doi.org/10.3390/ cancers14194707

Academic Editor: Lingzhi Wang

Received: 15 July 2022 Accepted: 21 September 2022 Published: 27 September 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

#### **1. Introduction**

Human papillomavirus (HPV) is connected with virtually all cases of cervical cancer. The viral infection-associated chronic inflammation has been considered as a leading risk factor for carcinogenesis in humans [1–3]. It has been estimated that 15–20% of cancers worldwide are attributable to virus, bacteria, or parasite infections causing chronic inflammation [4]. One of the characteristic features of chronic inflammation is the longlasting overproduction of reactive oxygen and nitrogen species (ROS and RNS, respectively) released by inflammatory, epithelial, and endothelial cells [5–7]. From one side, ROS and RNS play multiple essential roles in the innate immune response to infectious agents [8,9], as well as in metabolism and the elimination of mediators of inflammation [10]. On the other hand, their excess could negatively affect the normal structure and functions of host cells and tissues, thus creating conditions for endogenous mutagen formation [11] and/or carcinogenic transformations [11–13]. For example, oxidative and nitration/nitrosylation metabolites that are formed upon the interaction of DNA with ROS and RNS are either mutagenic [11] factors causing tumour progression [14] or can induce carcinogenesis through the activation of proto-oncogenes and the inactivation of tumour suppressor genes [15–17].

Oxidative imbalance has been clearly recognised as a promoting factor in HPV-initiated carcinogenesis [18]. For example, while the expression of thioredoxin reductase 2 and glutathione-S-transferase peaked, inducible nitric oxide synthase (iNOS) was progressively reduced in dysplastic and neoplastic cervical tissue. Applying a redox proteomic approach, De Marco et al. came to the conclusion that the HPV16 neoplastic progression of cervical cancer is associated with the oxidative modifications (carbonylation) of DNA and proteins involved in cell morphogenesis and terminal differentiation in dysplastic tissues. In contrast, cancer tissues were characterised by the selective reduction of carbonyl adducts on detoxifying/pro-survival proteins that reflected an improved control of oxidative alterations [18].

The cytokines of the tumour necrosis factor (TNF) ligand family, such as the cell surface death receptor (sFAS), TNF, and TNF-related apoptosis-inducing ligand (TRAIL), are wellknown regulators of apoptosis through their corresponding receptors [Ryu]. TRAIL was first identified as an inducer of p53-independent apoptosis in a variety of tumour cell lines and in cervical cancer patients (reviewed in [19]).

There are several evolving preventive strategies that could substantially reduce the burden from cervical carcinoma by the elimination of etiological factors and the inhibition of its development. First of all, they include HPV vaccines, diets enriched in fruits and vegetables [20–23], food supplements [23,24], and other chemopreventive agents [25–30]. Major molecular and cellular pathways for non-medicinal anti-cervical carcinoma preparations target the elimination/diminishing of the load of existing high-risk HPV infection [30–32], the restoration of normal non-viral microbiota [33] and redox balance [25,31,34–38], anti-inflammatory action [21,39], cell cycle influencers, and proapoptotic effects towards infected tumour transformed cells [40,41].

One of the most promising candidates, possessing several of the indicated activities, is fermented mangosteen (*Garcinia mangostana* Linn., Guttiferae), a popular botanical food supplement sold in large quantities around the world. Although well-known in ethnopharmacology since ages past, mangosteen-based products have been only recently studied for their phytochemical content and biological activities. Thus, the widely claimed cancer chemopreventive and anti-cancer effects of fermented mangosteen (FM) have been mainly attributed to the xanthons, tannins, and polyphenols of this health dietary additive [42–45]. These biological actives are mainly concentrated in the pod (synonyms: pericarps or rinds) [46] and seeds, non-edible parts of this delicious fruit usually referred to as "the queen of fruits". The blue and purple colours of the rape mangosteen skin are associated with anthocyanines, resveratrol, quercetin, rutin, and ellargic acid, all of them known for numerous biological activities, including chemopreventive anti-cancer effects [26–28,40,41,47,48]. The white colour of the seed placenta, characteristic of the

mangosteen pod, indicates the presence of beta-glucans and lignans with specific immunomodulating and anti-cancer actions, while the seed shells contain large amounts of lignins, which possess anti-cancer, anti-viral, and anti-microbial properties (reviewed in [49,50]). To increase the bioavailability of secondary plant metabolites encaged into the fibrous skin, mesopericarp, and seeds, they are subjected to a long-lasting controlled microbial and/or yeast fermentation.

Fermentation is the most ancient and the most natural way of plant food processing and preservation. During the fermentation (or external digestion) of plant parts, high molecular weight molecules, such as polysaccharides, glycolipids/glycoproteins, and nucleic acids of plant cell walls, seeds, peels, or pods, are decomposed to low molecular weight units (molecular moieties) that are readily absorbed through the gastrointestinal barrier and immediately available for a variety of metabolic processes [51]. Fermentation significantly enhances the cancer chemopreventive properties of non-fermented plants [52]. Moreover, fermented fruit preparations are enriched with the membrane fragments, lipoglycopeptide complexes, and exometabolites of fermenting yeasts and lactobacilli, probiotics that contribute to their immuno-modulating, nutritional, and physiological value [53,54]. Unfortunately, practically all research on the bioactivity of FM has been carried out in in vitro and ex vivo experiments, while clinical data are lacking.

The primary goal of the present study was the search for oxidative markers of HPV16 and HPV18-related cervical carcinogenesis and their possible connection with local and circulating ligands of TNF-alpha-induced apoptosis. In the first stage of the study, we found that the content of nitric oxide metabolites (nitrotyrosine and nitrates/nitrites) was significantly enhanced in the dysplastic cervical intraepithelial neoplasia (CIN) of different grades (CIN I and CIN II) groups, and the activity of MPO peaked in the invasive cervical cancer (CIN III) group.

Based on the results obtained, the placebo-controlled clinical trial on the clinical efficacy (the prevention and/or slowing down of carcinogenesis) of standardised FM gel, taken orally, was carried out. The secondary outcomes of the trial were oxidative and TNF-alpha-induced apoptosis ligand markers at systemic and cervical tissue levels in the practically healthy and the pre-cancerous CIN I and CIN II groups of female patients infected with HPV16/18, and in healthy females infected with low-risk HPV10/13.

#### **2. Materials and Methods**

#### *2.1. Ethics Statement*

All experiments with human material (blood, cervical biopsies, and cervical fluid) were carried out in accordance with the Helsinki Declaration, and the protocols of two clinical studies were reviewed and approved by the Ethical Committee of Berbekov's Kabardino-Balkar State Medical University, Nal'chik, Russia (Protocol No. 133-2/2018 of 21st November 2018). The recruited patients and the healthy age-matched female donors signed an informed consent form.

#### *2.2. Clinical Study Design and Recruitment Criteria*

Two clinical studies were conducted. The first was an observational trial aimed at identifying local tissue oxidative markers of papillomavirus (HPV)-connected cervical carcinogenesis and their possible connection with local and circulating ligands of TNF-alpha-induced apoptosis. During the period from January 2019 to December 2019, 196 women attending the Gynaecology Department of Kabardino-Balkar Berbekov's State University, Nal'chik, Russia, and presenting HPV infections of low (HPV10 and HPV13, n = 42; age range: 21-44 years) and high (HPV16 and HPV18, n = 154; for age range see Table 1) risk of cervical carcinogenesis, were recruited into the study after obtaining their informed consent. Fifteen age-matched females, not infected with HPV or other common viruses, such as cytomegalovirus (CMV) or herpes virus (HV1 and HV2), were invited to participate in the study as healthy controls. The demographic features, HPV infections, and

clinical diagnoses of the participants in the first stage of the clinical observational study are collected in Table 1.

**Table 1.** Demographic and clinical data of women—participants in the observational study on the oxidative markers and TNF-alpha apoptosis ligands in HPV infections associated with cervical carcinogenesis.


Women infected with HPV16 and HPV18 were diagnosed colposcopically and histologically on the basis of dysplastic lesions of different histological grades (n = 75; age range: 24–45 years; the pre-cancerous well-differentiated CIN I and moderately differentiated CIN II groups) or with clinically, histologically, and immunologically confirmed invasive cervical cancer (n = 45; age range: 28–53 years; the poorly differentiated CIN III group). The inclusion criteria were as follows: (a) patients older than 18 years; (b) not pregnant; (c) not infected with any other common viruses; (d) not bearing or have had any tumour apart from invasive cervical cancer in order to be included in the CIN III group; (e) not treated with any immune response-modulating or anti-viral drug for the last three months.

At the first visit, all participants were subjected to a full gynaecological examination. Cervical fluid containing ectopic cervical cells was collected and processed for cytological, immunological, and virological analyses. Biopsies were taken for standard histological evaluation, as well as for oxidative marker and apoptosis ligand assays. Of note, in the cases of dysplastic and neoplastic lesions, biopsies were taken from the area in a close vicinity (5 mm from the lesion border). All participants donated their venous blood for immunological analyses, as well. Any decision about the diagnostic procedures, treatment protocols, and follow-up period was based exclusively on the objective clinical picture, irrespective of any need of the observational trial.

Taking into account the outcome that several definite oxidative markers and TRAIL levels (in plasma and cervical fluid) corresponded to the severity of HPV-associated cervical dysplasia/neoplasia, we proceeded with a randomised, double-blind, placebo-controlled single-centre trial on the clinical efficacy and effects towards redox and apoptosis parameters of the FM food supplementation on HVP-infected patients with cervical dysplasia (CIN I and CIN II). A 5% honey solution in mineral water was used as the placebo. Females infected with HPV16 and HPV18 without clinical and colposcopic evidence of dysplasia (n = 152; age range: 25–45 years; practically healthy controls) and presenting clinical/colposcopic features of cervical dysplasia of different grades (CIN I or CIN II, n = 98; age range: 25–52 years) were invited to participate in the trial and signed informed consent forms. The inclusion criteria (a)–(c) and (e) were similar to those for the first observational clinical study. The criteria (d) were re-formulated as "not bearing or have had any tumour". One more criterium was added: (f) no antioxidant and mineral supplementation for at least three months before entry to the trial. During the period from January 2019 to November 2019, two hundred and fifty women were enrolled and randomly placed into four groups: Group 1—practically healthy, placebo (n = 70); Group 2—practically healthy, FM supplementation (n = 82); Group 3—CIN I and CIN II, placebo (n = 48); Group 4—CIN I and CIN II, FM supplementation (n = 50). All patients with clinical symptoms of dysplasia received surgical removal of lesions. All the participants were prescribed conventional anti-viral therapy per os and in vaginal lavages. The patients' demographic distribution and clinical data are summarised in Table 2.


**Table 2.** Demographic and clinical data of women—participants in placebo-controlled trial on the clinical efficacy, oxidative markers, and TNF-alpha ligand levels of fermented mangosteen (FM) supplementation.

In addition, patients from Groups 2 and 4 received FM supplementation as a standardised syrup (14 mL × 2 times a day at meal time for three months). The patients assigned to Groups 1 and 3 received 5% honey diluted in mineral water (14 mL × 2 times a day at mealtime for three months). FM was kindly provided free of charge by Carica Ltd., Manila, the Philippines.

As positive controls, the serum and cervical fluid of 30 healthy age-matching females without serum antibodies to HPV, herpes simplex types I and II, and cytomegalovirus (CMV), as well as free of HPV and CNV DNA in the cervical fluid, were used.

#### *2.3. Food Supplement in Question*

Standardised syrup of fermented mangosteen (FM) manufactured by Carica Ltd. (Manila, the Philippines), approved as a food supplement and distributed locally and abroad, was a kind gift of the manufacturer. The FM was produced from fruits collected from wild (non-cultivars and non-genetically modified) species of tropical *Garcinia mangostana* Linn. grown in remote, non-industrial parts of the Philippines. In brief, pods of rape fruit were separated from the edible flesh, washed, and mashed. The mash underwent the process of controlled fermentation by a food-quality culture of *Saccharomyces cerevisia* yeasts and *Lactobaccilus casei* within a 6-month period. Controlled enzymatic splitting was reached due to the strictly regulated temperature, oxygen, and nutrient supply, as well as frequent in-process quality tests. The fermentation process was stopped by adding the honey of wild bees (5%), followed by pasteurisation at 55 ◦C and filtration under pressure. The final FM product was a sour-sweet thick gel that was reddish-brown in colour. Post-fermentation analyses showed a high concentration of phenolics, fruit acids (pH 5.2), unsaturated fatty acids, sitosterols, macroelements, such as calcium, magnesium, and potassium, and microelements, such as copper, zinc, and iron. The ready FM product was free from pathogenic microorganisms and toxic elements, i.e., Sr, As, and Pb.

#### *2.4. Clinical Diagnosis*

All participants were subjected to a scrupulous gynaecological examination. The extended colposcopy with a magnification ×15 was carried out by a Letsegang colposcope (Germany) with video registration in order to identify and register the state of cervical epithelia. The cytological screening of cells that were collected from eczo- and endo-cervices by a spatula was performed by a conventional Papanicolaou smear test (Pap smear test).

#### *2.5. Viral Analyses*

Different types of HPV were determined by a type-specific PCR assay, using corresponding primers [55–57]. Total RNA was isolated using the GenElute Mammalian Total RNA kit (Sigma-Aldrich, Milan, Italy) and was reverse-transcribed using the iScript cDNA Synthesis kit (Bio-Rad, Hercules, CA, USA). cDNA was amplified with IQ SYBR green Supermix (Bio-Rad, Hercules, CA, USA), using the MiniOpticon Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA).

Herpes (HSV1 and HSV2) and cytomegalovirus (CMV) infections were assessed by real-time PCR, using the following primers: HSV1 fwd: CCT-TCG-AAC-AGC-TCC-TGG; rev: ATG-ACG-CCG-ATG-TAC-TTT-TTC-TT. HSV2 fwd: TCC-ATT-TTC-GTT-TTG-TGC-CGG; rev: ATG-ACG-CCG-ATG-TAC-TTT-TTC-TT. CMV fwd: ATG-ACG-CCG-ATG-TAC-TTT-TTC-T; rev: 5- -ACT-GGT-CAG-CCT-TGC-TTC-TAG-TCA-CC [Fax].

#### *2.6. Biological Material Sampling and Processing*

#### 2.6.1. Blood Sampling and Processing

Venous blood (6 mL) was taken into two hermetically sealed tubes and left for 1 h at room temperature for blood cell sedimentation. Serum was collected, aliquoted, and stored at −70 ◦C until the ELISA and PCR assays.

#### 2.6.2. Cervical Tissue Sampling and Processing

Biopsies were taken for standard histological pathological evaluation, as well as for oxidative marker and apoptosis ligand assays, in a colposcopy-guided manner. Of note, in the cases of dysplastic and neoplastic lesions, biopsies were taken from the area in a close vicinity (5 mm from the lesion border). The biopsies were placed in an ice-cold 0.1 M potassium phosphate buffer (pH 7.4) and thoroughly homogenised. The homogenates were centrifuged at 900× *g* and +4 ◦C for 40 min. The supernatants were collected and stored at −80 ◦C until they were analysed. The protein content in the supernatants was determined by Lowry's method, which is described elsewhere.

#### 2.6.3. Cervical Fluid Collecting and Processing

Fluid containing ectopic cervical cells was collected from the walls of the cervical channel after its opening by a spatula, was centrifuged at 230× *g* for 10 min, and then the supernatant was stored for immunological analyses. The cervical cell sediment was washed twice with a potassium phosphate buffer and used for differential counting with an automatic Coulter counter (Beckman Coulter Inc., High Wycombe, UK) and for the HPV type-specific PCR assay.

#### *2.7. Assays for Oxidative Markers*

#### 2.7.1. Enzymatic Activities

The activities of pro- and antioxidant enzymes were measured by spectrophotometric methods: for MPO, the absorbance at 560 nm was recorded after the reaction with orthodianizidine; for catalase, the absorbance at 240 nm was recorded during the reaction with hydrogen peroxide; and for glutathione peroxidase, the absorbance at 412 nm was measured after the reaction with GSH in the presence of t-butyl peroxide. Homogenates were added to the reaction mixture at a concentration of 1 mg protein per mL, and the results were expressed in % of the control mixtures without homogenates.

#### 2.7.2. Hydroxyl Radical Scavenging

Hydroxyl radical scavenging capacity was assessed by luminol-dependent chemiluminescence in a Fenton reaction. In brief, the reaction was initiated by the addition of 17.7 μM FeSO4 to 0.01 M potassium phosphate buffer containing 17.7 μM hydrogen peroxide and 0.2 mM luminol. The integral square under the chemiluminescence spike was calculated and plotted against the value obtained in the absence of homogenate. The data were expressed in arbitrary units.

#### 2.7.3. Nitrotyrosine Determination

The levels of nitrotyrosine were measured in accordance with methods described previously [58,59]. The supernatants were assayed for 3-nitrotyrosine using a Nitrotyrosine ELISA kit (Northwest/AMS Biotechnology, Manchester, UK), following the manufacturer's instructions. Briefly, the samples were incubated in the wells coated with a captured nitrotyrosine antibody and a biotinylated secondary tracer antibody. The addition of streptavidine-peroxidase, followed by tetra-methyl benzidine, resulted in colour development proportional to the nitrotyrosine levels, which were quantified spectrophotometrically at 450 nm. Total nitrate/nitrite levels were measured by Griess reaction using the analytical kit "Nitrate/nitrite Assay Kit Colorometric" (Sigma Co., Milan, Italy).

#### 2.7.4. Isoprostane Determination

The isoprostane content in the cervical tissue was determined as described in [60]. The homogenates were prepared on ice and the supernatants were diluted with distilled and de-ionised water. The solutions were acidified to below pH 4.0 and used for the further determination of free 15-F2t-IsoP by enzyme immunoassay (EIA) (Northwest NWLSSTM). Briefly, 15-F2t-IsoP in the samples or standards was allowed to compete with 15-isoprostane F2t conjugated to horseradish peroxidase (HRP) for binding to a polyclonal antibody specific for 15-isoprostane F2t coated on a micro plate. Subsequent tetra-methyl benzidine additions resulted in a blue colour development that was inversely proportional to the quantity of 15-isoprostane F2t in the original samples or standards. After the addition of a stop solution, the absorbance was read at 450 nm.
