Pummelo fruit extract (Citrus maxima) reduces oxidative damage in peripheral blood of obese patients

Abstract

Objectives

To verify the level of oxidation of biomolecules in the peripheral blood of obese patients before and after incubation with the pummelo pulp extract at 25 mg/mL.

Methods

This investigation included 241 individuals, distributed across the following groups: control (n = 69; 30 men and 39 women); pre-obesity (n = 78; 29 men and 49 women); level I obesity (n = 62; 20 men and 42 women); and level II or III obesity (n = 32; 11 men and 21 women).

Results

TBARS and MN levels were significantly altered by incubation with pummelo extract in both genders and in all the groups according to their obesity level. Another alteration in the antioxidant profile of the samples was a significant increase in vitamin C levels observed after incubation with pummelo extract in all experimental groups.

Conclusions

The aqueous extract of pummelo pulp showed in vitro biological activity against the oxidation of biomolecules, suggesting that its bioactive compounds may bring health benefits of obese patients.

Graphical Abstract

Graphical Abstract.

Introduction

Obesity is a chronic, multifactorial disease considered a global epidemic, responsible for systemic damage and linked to a range of comorbidities. These individuals experience a metabolic imbalance, partially linked to the consumption of high-calorie foods, resulting in the accumulation of excess adipose tissue and subsequent weight gain.¹ In Brazil, a survey conducted by the Ministry of Health in 2019 estimated that 55.7% of the population is overweight and that 20% already exhibit varying degrees of obesity. Furthermore, the study discloses that excess weight affects 57.8% of men and 54% of women.²

While obesity prevalence is greater among women, there exists a variation in the risk of developing: (1) cardiovascular diseases³; (2) type 2 diabetes⁴; and (3) insulin resistance.^5,6 All these pathologies are largely linked to the accumulation of abdominal fat, partially influenced by hormonal disparities.^5,7 This deposition of abdominal fat significantly contributes to chronic low-grade inflammation.⁷ In this context, the chronic low-grade inflammatory process exerts systemic effects including the activation of endothelial cells, infiltration of macrophages in adipose tissue, and notably, oxidative stress.^8,9

Treating obesity involves modifying dietary habits, incorporating physical activity, and, when necessary, utilizing medications tailored to each case. Given that this condition, much like numerous others, involves oxidative stress in its pathogenesis, leveraging antioxidant compounds presents a promising avenue for treatment. In light of this, multiple studies have centered on exploring the potential therapeutic benefits of botanicals and fruits, based on popular knowledge.^10,11

Fruits from the Citrus genus, which belong to the Rutaceae family, are widely consumed worldwide, and their potential therapeutic benefits have been linked to the existence of bioactive substances with antioxidant properties, including phenolic compounds (phenolic acids, coumarins, and flavonoids). Moreover, they are also rich in vitamins, fiber (pectin), and minerals.¹²Citrus maxima, commonly known as pummelo, originates from Southeast Asia and thrives in tropical regions. It stands as the largest among citrus fruits, characterized by its globular shape and 11–14 pear-shaped segments. The pulp of the fruit varies from white to pink, delivering a sweet-tart flavor.¹³ In a study by Nunes et al.,¹⁴ pummelo fruit pulp exhibited significant anti-inflammatory, antioxidant, and hepatoprotective effects in male Wistar rats with non-alcoholic fatty liver disease.¹⁴ Research involving pummelo fruits in human subjects is limited in the literature, and none have been conducted on obese patients. Consequently, the objective of this study was to assess the in vitro antioxidant capacity of pummelo pulp extract in the peripheral blood of obese patients and compare it with control individuals.

Material and methods

Participants

The study included 241 individuals of both sexes and without distinction of ethnicity. Whole blood samples from patients and controls were obtained by venipuncture at the Clinical Analysis Laboratory of the Health Department of the city of Uruguaiana/RS. In parallel, an interview was conducted with the study participants to obtain socio-demographic data and health profile. Waist circumference was measured and body weight obtained using appropriate equipment.

Were considered as inclusion criteria men or women, over 18 years old and regardless of ethnicity. Exclusion criteria: pregnant women, cancer patients, individuals under 18 years of age, smokers, with infectious-contagious and/or autoimmune diseases.

Patients were considered overweight, obese, grades 1, 2 and 3 according to the WHO classification (Table 1). The research protocol was authorized by the Human Research Ethics Committee, under no. 5.308.525 and the Informed Consent Form was obtained from each participant.

Table 1.

Classification of obesity according to World Health Organization.¹⁵

Body mass index (BMI / Kg/m2)	Classification
Under 18.5	Low weight
Between 18.6–24.9	Normal weight
Between 25–29.9	Overweight
Between 30–34.9	Grade I obesity
Between 35–39.9	Grade II obesity
Over 40	Grade III obesity

Pummelo material

Ripe fruits of pummelo were collected in a rural region in Uruguaiana, Brazil. The botanical identity of the plant specimen was performed out by Prof. Dr Renato Aquino Záchia, with the proof of this species being deposited in the SMDB Herbarium of the Federal University of Santa Maria (UFSM), Santa Maria, RS, Brazil (register number 16,569).

Extract preparation

The mesocarp and pulp of the fruit were utilized, excluding the seeds. A quantity of ten grams (10 g) was processed in a blender with 100 mL of distilled water. Following a freezing period of 48 h, the material underwent lyophilization, resulting in a powder that was stored at a temperature of −20 °C. Before application in experiments, the powder was reconstituted using distilled water. Previous analyses indicated that the major component within the fruit pulp sample was caffeic acid (3.71 mg/g of lyophilized extract), with the total phenolic content in the fruit pulp ranging from 13.14 to 42.88 mg of GAE/g. In the ABTS and DPPH assays, the fruit pulp (0.025 to 0.5 mg/mL) demonstrated inhibition percentages of 22.3% to 35.1% and 36.6% to 76.0%, respectively.¹⁴

Dosage information

Pummelo extract was resuspended in phosphate-buffered saline (PBS pH 7.4) to achieve the concentration (25 mg/mL). The experiments were conducted on the whole blood cells of subjects. The cells were exposed to chosen concentration of pummelo (1:1 v/v) for 3 h at 37 °C. We used of the aqueous extract of pummelo of 25 mg/mL corresponds to the ingestion of a quarter of the fruit in natura.

Oxidative damage analysis

Levels of thiobarbituric acid reactive species (TBARS)

Lipid peroxidation levels were measured using the TBARS method.¹⁶ For this, 200 μL of sample are conditioned with 1 mL of orthophosphoric acid (0.2 M), 500 μL of distilled water and 250 μL of trichloroacetic acid reagent (0.1 mol/L) for 120 min at a temperature of 95 °C. After cooling the sample, the reading was taken in a spectrophotometer, model SP-22, single beam (Biospectro) at 532 nm.

Plasma protein oxidation

To quantify the protein carbonylation, the technique, described by Levine et al.¹⁵, was performed. For this test, 100 μL of trichloroacetic acid (28%) was added to same amount to sample of plasma, causing the proteins present in the sample to precipitate, their breaking and marking was done using 1 mL of DNPH or 3% HCl, for the blank sample, which will have to be subtracted from the sample value at the end of the reading. Afterwards, the mixture was incubated for 90 min at 37 °C. Then, the interferents are washed using 1 mL of ethanol-ethyl acetate (1:1 v/v) and the proteins are resuspended using 200 μL of guanidine hydrochloride (6 M), prepared using 20 mM PBS buffer, the reading is performed in a SpectraMax M5 microplate reader (Molecular Devices^®) at 370 nm.¹⁵

Deoxyribonucleic acid (ADN) oxidation

To quantify ADN oxidation were used the micronucleus frequency (MN), described by (Schmid, 1975).¹⁷ Blood smears, hematological staining and subsequent microscopic analysis were performed. One hundred peripheral blood leukocytes were contacted and the MNs were observed and expressed in percentage.

Vitamin C levels

The quantification of ascorbic acid was established via HPLC, according to the methodology developed by Karatepe in 2004.¹⁸ To a 100 μL portion of plasma, 100 μL of 0.1 M perchloric acid and 1 mL of distilled water were incorporated. Subsequently, the samples underwent centrifugation at a speed of 4,500 rpm for a duration of 5 min before being employed for HPLC analysis. A mobile phase composed of monobasic potassium phosphate (pH 3.6, 30 mM) and methanol at a proportion of 82.5:17.5 (v/v) was used. The flow rate was maintained at 1.2 mL/min. Employing a C18 column, the chromatograms were monitored at 250 nm, targeting the two specific analytes. The quantification of concentrations was executed using the principles of external standardization.

Data analysis

A paired t-test was conducted to assess differences in oxidative stress parameters within paired observations before and after pummelo extract incubation. The chi-square test was used to analyze associations among categorical data. A comprehensive multivariate approach was adopted, involving both Principal Component Analysis (PCA) and the construction of correlation networks. PCA aimed to elucidate patterns in the dataset, as well as relationships among all variables. Samples were clustered based on experimental treatment, obesity level, and disease presence. Correlation network analyses were constructed with Pearson r values to explore relationships among variables within each investigated group. All analyses were carried out using GraphPad Prism 9.2 or Origin 2022 software.

Results

The sociodemographic profile and lifestyle variables of the study participants are presented in Table 2. This investigation included 241 individuals, distributed across the following groups: control (n = 69; 30 men and 39 women); pre-obesity (n = 78; 29 men and 49 women); level I obesity (n = 62; 20 men and 42 women); and level II or III obesity (n = 32; 11 men and 21 women). The study observed a prevalence of women participants, with a majority identifying as white. Among the participants the majority were married or divorced. The level of education of the participants analysis showed that the majority had completed the elementary or high school, followed by completion a high degree course. It was also noted that the majority of individuals in the groups do not use vitamins and/or other dietary supplements, with a prevalence among participants of consuming >3 meals per day in control and pre-obesity groups, and ≤3 meals per day in others groups (P = 0.0506). With the exception of the control group, the remaining groups reported having elevated percentages of comorbidities possibly associated with degrees of obesity (P <0.0001). Concerning physical activity engagement, a predominance of physical inactivity was observed mainly across obese groups (P = 0.0130); conversely, water intake was evident in all groups. As expected, the measurement of abdominal circumference was associated with the obesity level (P < 0.0001).

Table 2.

Socio-demographic and lifestyle characteristics of study population according to the obesity level.

		Control (N = 69)	Pre-obesity (N = 78)	Level I (N = 62)	Level II or III (N = 32)	P
Age	Years	43.07 ± 19.43	52.11 ± 16.37	50.09 ± 15.89	48.59 ± 14.33
Gender	Men	30 (43.48)	29 (37.18)	20 (32.26)	11 (34.38)	0.5899
	Women	39 (56.52)	49 (62.82)	42 (67.74)	21 (65.62)
Color	White	41 (59.42)	43 (55.13)	30 (48.39)	17 (53.13)	0.6999
	Brown/black	29 (40.58)	35 (44.87)	32 (51.61)	15 (46.87)
Civil state	Single	33 (47.82)	28 (35.90)	26 (41.94)	11 (34.38)	0.4289
	Married / other	36 (52.18)	50 (64.10)	36 (58.06)	21 (65.62)
Schoolarity level	Primary	22 (31.88)	31 (39.74)	23 (37.10)	16 (50,00)	0.7103
	Secondary	29 (42.03)	27 (34.62)	24 (38.71)	11 (34.37)
	Tertiary	18 (26.09)	20 (25.64)	15 (24.19)	5 (15.63)
Vitamin	Yes	13 (18.84)	22 (28.21)	11 (17.74)	6 (18.75)	0.3885
	No	56 (81.16)	56 (71.79)	51 (82.26)	26 (81.25)
Food intake	≤3	24 (34.78)	35 (44.87)	36 (58.06)	17 (53.13)	0.0506
	>3	45 (65.22)	43 (55.13)	26 (41.94)	15 (46.88)
Diseases	Yes	23 (33.33)	48 (61.54)	45 (72.58)	24 (75,00)	<0.0001
	No	46 (66.67)	30 (38.46)	17 (27.42)	8 (25.00)
Physical activity	Yes	32 (46.38)	35 (44.87)	21 (33.87)	5 (15.63)	0.0130
	No	37 (53.62)	43 (55.13)	41 (66.13)	27 (84.37)
Water	Yes	60 (86.96)	71 (91.03)	60 (96.77)	31 (96.87)	0.1382
	No	9 (13.04)	7 (8.97)	2 (3.23)	1 (3.13)
Ethanol	Yes	30 (43.48)	26 (33.33)	21 (33.87)	16 (50.00)	0.2674
	No	39 (56.52)	52 (66.67)	41 (66.13)	16 (50.00)
Smoke	Yes	13 (18.84)	11 (14.10)	15 (24.19)	8 (25.00)	0.3989
	No	56 (81.16)	67 (85.90)	47 (75.81)	24 (75.00)
Circumference	Normal	36 (52.17)	4 (5.13)	0 (0)	0 (0)	< 0.0001
	Pre-obesity	33 (47.83)	74 (94.87)	62 (100)	32 (100)

TBARS levels were significantly altered by incubation with pummelo extract in both genders (Fig. 1a–d for women and Fig. 1e–h for males) and across all levels of obesity or participants with normal weight. Regarding TBARS, all paired comparisons between before and after pummelo extract incubation presented remarkably low p-values and are depicted in Fig. 1. These results underscore the potential of pummelo extract in mitigating lipid damage. Likewise, the micronuclei count was significantly lower after pummelo extract incubation in women and men in all the experimental groups defined by gender and obesity levels (Fig. 2). Despite these considerable differences in TBARS levels and micronuclei counts, incubation with pummelo extract had a minor effect on carbonyl levels, for both genders and obesity groups (Fig. 3). Another alteration in the antioxidant profile of the samples was, a significant increase in vitamin C levels observed after incubation with pummelo extract in all experimental groups (P < 0.0001), as shown in Fig. 4.

Fig. 1.

TBARS values before (N) and after (P) the plasma in vitro incubation with pummelo extract (25 mg/mL) in woman (a–d) and men (e–h) with different obesity levels, being control group (a and e), preobesity (b and f), level I obesity (c and g) and level II or level III obesity (d and h). The data were presented as means with error bars representing the standard deviation.

Fig. 2.

Micronuclei counting before (N) and after (P) the plasma in vitro incubation with pummelo extract (25 mg/mL) in woman (a–d) and men (e–h) with different obesity levels, being control group (a and e), preobesity (b and f), level I obesity (c and g) and level II or level III obesity (d and h). The data were presented as means with error bars representing the standard deviation.

Fig. 3.

Carbonyl values before (N) and after (P) the plasma in vitro incubation with pummelo extract (25 mg/mL) in woman (a–d) and men (e–h) with different obesity levels, being control group (a and e), preobesity (b and f), level I obesity (c and g) and level II or level III obesity (d and h). The data were presented as means with error bars representing the standard deviation.

Fig. 4.

Ascorbic acid values before (N) and after (P) the plasma in vitro incubation with pummelo extract (25 mg/mL) in woman (a–d) and men (e–h) with different obesity levels, being control group (a and e), preobesity (b and f), level I obesity (c and g) and level II or level III obesity (d and h). The data were presented as means with error bars representing the standard deviation.

Another analysis that can be conducted on Figs 1 and 2, is that a noticeable increase micronuclei counts can be identified as the obesity level increases, and regarding TBARS a significant growth occurs between the pre-obesity and level I obesity in groups N. TBARS levels remained relatively constant (around 40 nM) when the pummelo extract was present, demonstrating their capacity to counteract the lipid damage caused by obesity-associated oxidative stress.

The correlations between oxidative stress and anthropometric variables within the groups were investigated by constructing a correlation network. A correlation network visually represents the relationships between variables, where nodes denote variables and edges colors indicate the strength and direction of their correlations. Validating the correlation profile, a robust positive correlation between abdominal circumference (6) and BMI (7) was observed in most of the groups. An interesting observation can be made in Fig. 5d, g, k, and o, where strong positive correlations were noted between micronuclei count (4) and both abdominal circumference (6) and BMI (7). These positive correlations were marked by dark-colored triangles (vertices in circles 4, 6 and 7) connecting the circles representing these three variables. In contrast, these correlations were either absent or weak in groups with lower levels of obesity, regardless of treatment with pummelo extract, showing the obesity role in micronuclei counting. A strong inverse correlation, widely evident in nearly all of the groups, was observed between ascorbic acid (3) and carbonyl protein oxidation (2). The most intense correlation between these two variables was in plasma obtained from women with Level II and Level III obesity prior to pummelo incubation, with a Pearson coefficient of −0.6381 (P = 0.00186), as illustrated in Fig. 5d. This finding highlights the capacity of ascorbic acid to mitigate protein damage by oxidative stress.

Fig. 5.

Correlation network of antioxidant and anthropometric profiles across investigated groups. Figures (a–g) were created using data obtained from women groups, and figures (h–o) were constructed based on data collected from men participants. Figures (a–d) and (h–k) correspond to the groups before pummelo fruit incubation (normal group), while figures (d–g) and (l–o) correspond to analyses after pummelo extract incubation. Obesity levels are arranged from right to left: Figures (a), (d), (h), and (l) correspond to the control group; figures (b), (e), (i), and (m) correspond to the pre-obesity level; figures (c), (f), (j), and (n) correspond to level I obesity; and figures (d), (g), (k), and (o) correspond to level II and level III obesity. Variables related to oxidative stress are represented by blue circles: TBARS (1), carbonyl (2), ascorbic acid (3), and micronuclei count (4); age and anthropometric variables are represented by green circles: Age (5), abdominal circumference (6), and BMI (7). Low Pearson r values between −0.20 and +0.20 were omitted.

Aiming to understand the relations among anthropometric variables and antioxidant measurements, was conducted a principal component analysis (PCA). This analysis produces plots that reduces data complexity by transforming variables into principal components (PC1 and PC2), which are linear combinations of the original variables. The original eight variables (micronuclei, TBARS, carbonyl protein, ascorbic acid, height, weight and abdominal circumference) were reduced to two principal components, namely P1 and PC2, which together explained 53.56% of the total variance. The investigated samples were then clustered based on three categorical variables: experimental group (Fig. 6a), obesity level (Fig. 6b) and presence of disease (Fig. 6c). The clustering pattern, as depicted in Fig. 6a, revealed that samples incubated with pumelo extract (P group) exhibited higher PC2 values, while control samples (C group) showed lower PC2 values. This pattern was attributed to lower TBARS values and higher vitamin C levels in the samples incubated with pumelo extract, as observed in the vector plot shown in Fig. 6d. Another distinct clustering pattern found is based on obesity level, mainly influenced by PC1 values. In Fig. 6b, the points on the left side of the plot correspond to patients with normal weight (in light green), followed in increasing order by patients with pre-obesity, level I obesity, and level II or III obesity (in dark green, blue and red). The variables with the most significant effect on PC1 values were weight, abdominal circumference, and micronuclei count. Furthermore, Fig. 6c illustrates another clustering pattern along the x-axis, where the color scale indicates the presence of disease. The largest of the green points are predominantly located on the right side of the plot (Fig 6c), highlighting the association between obesity and disease development.

Fig. 6.

Principal component analysis was performed on anthropometric variables and antioxidant measurements. PC1 and PC2 scores were plotted in (a), (b), and (c), with the categorical variables being the experimental group, the obesity level, and the presence of disease, respectively. In (d), the vector plot is presented, showing the correlations among all variables and their roles in clustering. C: Control group; P: Pummelo group; P: Pre-obesity; I: Level I; II/III: Lever II or III; Y: Yes; N: no.

To better understand the relationship among all investigated variables, the vector plot was constructed (Fig. 6d). May be observed in the vector plot a strong negative correlation between vitamin C and TBARS levels. Additionally, a strong positive correlation exists between micronuclei count and obesity-linked variables, such as abdominal circumference and weight.

Discussion

It may be highlighted that for the first time, this investigation reports the in vitro effects of pummelo aqueous extract on peripheral blood from patients with different obesity levels. Furthermore, our findings showed an impressive effect of pummelo extract on in vitro oxidative stress parameters.

Obesity is a chronic and multifactorial disease that reached pandemic levels in the last century. It is currently well known for its role in systemic damage and the development of comorbidities. Individuals with obesity experience a metabolic imbalance due to the consumption of high-energy foods, which leads to weight gain and an increase in adipose tissue.¹ In Brazil, according to a survey by the Ministry of Health, 55.7% of the population are above the normal weight, and 20.0% show some level of obesity. Additionally, the same institution reported that excess weight is found in 57.8% of men and 54.0% of women. As for obesity, it is present in 20.7% of women and 18.7% of men.²

Obesity produces a low-level chronic inflammatory process with systemic effects, which leads to the development of other pathologies, such as type II diabetes, cardiovascular diseases, hepatic steatosis, and others.¹⁹ In this context, oxidative stress plays a role in the progression of symptoms and aggravating factors by increasing biomolecule oxidation and the production of a redox imbalance.²⁰

In this investigation, a higher prevalence of women was identified, probably due to increased engagement with health services. An investigation conducted by other authors²¹ identified that patient care is mainly taken on by women, and a lower number of health services are designed for the men population, which could explain the lower utilization of health services by men. The low participation of men in seeking health services is also reported by other studies.^22,23 Data from IBGE showed that in 2009, there were 100 women for every 95 men, and changes in this ratio may reflect higher men mortality rates. The survey also reported that women use health services more frequently than the men population, such differences are determinants of consumption of services between sexes.²⁴

Another result of our investigation was an association between the presence of diseases and the measurement of abdominal circumference. A study conducted by other authors²⁵ also showed that abdominal circumference is a risk factor for the development of diseases, particularly cardiovascular diseases. Due to this aspect, our investigation emphasizes the importance of obesity prevention, aiming to preserve health.

Obesity is also involved with oxidative stress. This relationship may also be an origin of the development of several pathologies. In this study the in vitro concentration used of the aqueous extract of pummelo of 25 mg/mL corresponds to the ingestion of a quarter of the fruit in natura in humans. In this sense, the oxidation of biomolecules was determined before and after incubation with pummelo extract. One of the parameters investigated was protein carbonylation, and the experimental treatment with pummelo extract showed no effect on it. Furthermore, pummelo aqueous extract presented no genotoxic or mutagenic effects on peripheral human leukocytes, leading to a decrease in micronuclei count as a consequence of pummelo extract incubation. An investigation conducted by Park and colleagues¹¹ explored the activity of Citrus unshiu bark (a type of Asian pummelo) in the bone marrow of rats and did not identify changes in micronucleated cells compared to the experimental control. Thus, our results, and data from the literature, indicate the safety of pummelo regarding genotoxic and mutagenic effects.

Regarding TBARS levels, blood incubated with pummelo extract showed a significant decrease in all groups according to obesity levels and gender. Most importantly, pummelo extract was able to prevent the increase in TBARS levels produced by obesity. To our knowledge, this is the first report of TBARS levels in obese patients. Still with human patients, a clinical trial by Johnston (2003), showed that the ingestion approximately 1 glass of orange juice, had the ability to reduce lipid peroxidation by TBARS.²⁶ Silva et al.²⁷ identified that rats fed a hypercholesterolemic diet + Japanese orange extract (Fortunella japonica) had lower levels of TBARS compared to rats that ate the same diet without the orange extract.²⁷ Another study in rats showed malondialdehyde (MDA) modulation, but with a CCl₄ intoxication model, showing that pummelo peel extract was able to reduce lipid peroxidation and increase the activity of tissue antioxidant defenses.^28,29

These findings may be attributed to the antioxidants present in pumelo extract. Caffeic acid, recently identified as one of the main antioxidants in pumelo extract,¹⁴ is a widely distributed compound in nature; however, studies on plasma data remain scarce.³⁰ Nonetheless, the antioxidant activities of caffeic acid in metabolic syndrome have been extensively investigated. It has been shown that caffeic acid maintains a stable structure after the cleavage of its hydroxyl bonds, enabling it to eliminate radicals involved in lipid peroxidation or inhibit their actions.^31,32 Additionally, caffeic acid is associated with metabolic syndrome through the increased expression of leptin, GLUT3, Akt, and PI3K receptors, the sensitization of cells to insulin receptors, and the regulation of adipogenesis via the reduction of PPARγ expression in experimental models treated with caffeic acid.^31,33

With respect to the levels of ascorbic acid, an increase was observed in all groups, which was already expected since pummelo aqueous extract is a source of ascorbic acid, demonstrating its antioxidant activity. A study by Anmol et al.,³⁴ which conducted phytochemical analysis of Citrus grandis, a fruit from the same family as pummelo, showed its antioxidant capacity due to the high concentration of compounds such as ascorbic acid.³⁴ A study performed by Nunes et al.¹⁴ revealed that the aqueous extract of pummelo pulp possesses high antioxidant activity through the DPPH assay, ABTS radical, and high content of polyphenols. HPLC analysis identified two major compounds, caffeic acid and catechin, both of which have well-documented strong antioxidant activity.¹⁴ Thus, the present study reinforces the antioxidant potential of pummelo fruit pulp.

Within the peroxisome proliferator-activated receptor (PPAR) family, PPARα plays a crucial role in enhancing fatty acid oxidation. Lee et al.³⁵ demonstrated that ascorbic acid supplementation increases the mRNA levels of PPARα and related enzymes in the adipose tissue of mice.³⁵ Furthermore, Johnston et al.³⁶ found that individuals with adequate plasma levels of ascorbic acid experience increased lipid oxidation during moderate exercise compared to those deficient in this vitamin.³⁶

The PCA analysis produced well-defined clusters of the samples based on obesity level, incubation with pummelo extract, and disease presence. These patterns suggests that the selected variables play a role in the clustering and enable us to identify the variables with the most significant impact on clustering. In other investigations, PCA was employed to cluster serum samples from patients undergoing different durations of chemotherapy for cancer³⁷ or exhibiting active Toxoplasma gondii infections.³⁸ These investigations also unveiled the influence of antioxidant markers on the clustering pattern and demonstrated the correlations among them in a simple and comprehensive manner.

Aqueous extract of pummelo pulp showed in vitro biological activity against the oxidation of biomolecules, suggesting that its bioactive compounds may bring health benefits of obese patients. Another significant finding is the absence of genotoxicity and mutagenicity at the investigated concentrations, showing the safety of the pummelo ingestion. Additionally, in vitro incubation with pummelo substantially altered the antioxidant profile of patients’ plasma, an effect that could be demonstrated using multivariate statistical models. The findings reported here highlight that the aqueous extract of pummelo has great potential to reduce biomolecule oxidation. Due to its promising antioxidant activity, further investigations will be conducted to better explore the commercial potential of pummelo aqueous extract in new pharmaceutical formulations development. Preventing the progression of diseases where oxidative stress is a key factor is the main benefit of the advances achieved by our investigation.

Author contributions

Conceptualization, V.T.N.; V.M.; C.C.D.; J.C.E.P. Methodology: V.T.N.; E.G.C.; G.E.S.; L.S.S.; C.B.P.; S.M.M.S.; Formal analysis: I.L.G.; Writing—original draft preparation, V.T.N.; I.L.G.; V.M.; Writing—review and editing, V.T.N.; I.L.G.; V.M.; C.C.D. All authors have read and agreed to the published version of the manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflict of interest statement: The authors declare that they have no known competing financial interests or personal relationships that could have.

Data availability

Data will be made available upon request.

Ethical approval

The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee (project identification code: 5.308.525).