I. sonorae, C. limetta, and B. media phytoextracts and their antidiabetic potential

Fitoextractos de I. sonorae, C. limetta y B. media y su potencial antidiabético

Original Article



Iza F. Pérez-Ramírez1 , Julio C. Rubio-Rodríguez1 , Luis M. Salgado2 , Ofelia Mora3 , María L. González- Dávalos3 , Rosalía Reynoso-Camacho1*

1 Facultad de Química, Universidad Autónoma de Querétaro, Cerro de las Campanas s/n, Las Campanas, Querétaro 76010, Querétaro. México.

2 Centro de Investigación en Ciencias Aplicadas y Tecnología Avanzada, Instituto Politécnico Nacional, Cerro Blanco 141, Colinas del Cimatario, Querétaro 76090, Querétaro. México.

3 Instituto de Neurobiología, Universidad Nacional Autónoma de México, Blvd. Juriquilla 3001, Juriquilla, Querétaro 76230, Querétaro. México.

ABSTRACT

In Mexican traditional medicine, several herbs are used for their potential antidiabetic effects. This study aimed to in- vestigate the mechanisms underlying the antihyperglycemic effects of I. sonorae, C. limetta, and B. media. Aqueous extracts of these herbs demonstrated a consistent reduction in post- prandial blood glucose levels, in healthy rats during a starch oral tolerance test. Notably, B. media and I. sonorae exhibited significant in vitro inhibitory effects against alpha-amylase activity (20.5 and 25.4 %, respectively), while B. media en- hanced glucose uptake in adipocytes by 4.0-fold, which was related to the overexpression of key genes involved in insulin signaling cascade including Glut4, Irs1, and Pi3k (2.9-, 2.6-, and 3.2-fold, respectively). Furthermore, multivariate analysis highlighted that hydroxybenzoic acid hexoside and feruloyl- quinic acid were linked to their alpha-amylase inhibitory ac- tivity, while 17 distinct polyphenols were associated with the insulin mimetic activity. These findings propose a potential application of these herbs in the development of functional beverages with promising anti-diabetic attributes.

Keywords: Herbs; traditional medicine; adipocytes; glucose uptake; digestive enzymes; polyphenols.


RESUMEN

En la medicina tradicional mexicana, diferentes hierbas son utilizadas por su potencial efecto antidiabético. Este estudio tuvo como objetivo investigar los mecanismos subyacentes a los efectos antihiperglucémicos de I. sonorae, C. limetta y B. media. Los extractos acuosos de estas hierbas demostraron una consistente reducción en los niveles postprandiales de glucosa en sangre durante una prueba de tolerancia oral al almidón. Destacando que, los extractos de B. media e I. sonorae mostraron un efecto inhibitorio significativo in vitro contra la actividad de alfa-amilasa (20.5 y 25.4 %, respectiva- mente), mientras que B. media aumentó 4.0 veces la interna- lización de glucosa en adipocitos, lo cual fue relacionado con la sobreexpresión de genes claves involucrados en la cascada de señalización de la insulina incluyendo Glut4, Irs1 y Pi3k (2.9, 2.6 y 3.2 veces, respectivamente). Además, el análisis

*Author for correspondence: Rosalía Reynoso-Camacho e.mail: rrcamachomx@yahoo.com.mx

Received: April 10, 2024

Accepted: September 12, 2024

Published: October 17, 2024

multivariado resaltó que el ácido hidroxibenzoico hexódio y el ácido feruloilquínico fueron asociado con la actividad inhibitoria de alfa-amilasa, mientras que 17 polifenoles fue- ron asociados con la actividad mimética a la insulina. Estos hallazgos proponen la aplicación potencial de estas hierbas en el desarrollo de bebidas funcionales con prometedoras propiedades antidiabéticas.

Palabras clave: Hierbas; medicina tradicional; adipocitos; internalización de glucosa; enzimas digestivas; polifenoles.


INTRODUCTION

Type 2 diabetes (T2D) is a prevalent metabolic disorder, affec- ting over 537 million people globally in 2021 according to the International Diabetes Federation (IDF). Hyperglycemia ari- ses due to impaired carbohydrate digestion, glucose uptake, and insulin signaling. Herbal extracts, rich in polyphenolic compounds, offer potential for managing hyperglycemia by inhibiting carbohydrate-digesting enzymes and improving insulin sensitivity (Tran et al., 2020).

According to Mexican traditional medicine, Citrus limetta roots, Ibervillea sonorae roots, and Briza media lea- ves are consumed for diabetes control. Several scientific reports have previously demonstrated their hypoglycemic potential. A methanol extract of C. limetta fruit peel reduced blood glucose levels in diabetic rats, which was associated with its polyphenolic composition and antioxidant capacity (Flores-Fernández et al., 2017). The juice and aqueous extract of I. sonorae roots decreased blood glucose in healthy mice, and its dichloromethane extract showed a similar effect in alloxan diabetic mice (Alarcón-Aguilar et al., 2005). Similarly, glucose levels were reduced when rats were fed with a high fat and fructose diet and were treated with aqueous extract of I. sonorae roots, which was partly associated with its fla- vonoid content (Rivera-Ramírez et al., 2011). Zapata-Bustos et al. (2014) reported that an aqueous extract of I. sonorae roots induces glucose uptake in insulin-sensitive and insulin- resistant adipocytes. However, the mechanisms associated with the anti-diabetic effect of I. sonorae and C. limetta has not been identified. Regarding B. media, its aqueous extract



1

Volume XXVI

DOI: 10.18633/biotecnia.v26.2308

Journal of biological and health sciences http://biotecnia.unison.mx

Universidad de Sonora

ISSN: 1665-1456

is used for diabetes treatment but there is no evidence of its effectiveness (Lane et al., 2006).

There is an increasing interest in the identification of polyphenol-rich sources for the development of functional beverages, which are the main application of the global po- lyphenol market, valued in in about 1.68 billion USD in 2022 (Grand View Research on polyphenols market). Therefore, the aim of this study was to identify the mechanisms asso- ciated with the hypoglycemic effect of C. limetta, B. media and I. sonorae aqueous extracts, via the evaluation of in vitro inhibitory activity against carbohydrate digestion enzymes and the modulation of key transcripts involved in the insulin signaling pathway in 3T3 L1 adipocytes, and to associate these beneficial health effects with their polyphenol profile through a chemometric analysis. These plants are analyzed in parallel due to their shared traditional use for diabetes management and their hypothesized similar hypoglycemic effects. By comparing these extracts, we aim to understand their distinct and potentially complementary mechanisms of action in controlling blood glucose levels, providing a com- prehensive evaluation of their therapeutic potential.


MATERIAL AND METHODS

Herbal material and preparation of herbal aqueous ex- tracts

Roots of Ibervillea sonorae and Citrus limetta roots, leaves and stems of Briza media were obtained from local markets in Querétaro, México, based on their traditional use in Mexican medicine for diabetes management. Extracts were prepared using two g of dried material in 100 mL of water and boiled for 30 min. This preparation method was chosen according to the recommendations of local herbalists to ensure traditional relevance. After boiling, all extracts were filtered using a 0.5 mm pore size filter and were stored at 4 ºC protected from the light.


Animal protocol experiments

Thirty male Wistar rats of 220 - 240 g were acquired from the Universidad Nacional Autónoma de México (Juriquilla, Querétaro, Mexico). The housing room was maintained at 23 ºC and 50 % relative humidity, under a 12 h light-dark cycle, animals were housed individually with free access to standard diet (La Rodent Diet 5001) and water. Experiments were performed in accordance with the Animal Care and Use Protocol (NOM-062-ZOO-1999), and protocol was approved by the Bioethics Committee, Faculty of Natural Sciences of the Universidad Autónoma de Querétaro (Querétaro, México; approval number: 5FCN2015).


Oral starch tolerance test

After one week of acclimatization, rats were fasted for 12 h and then randomly allocated in five groups of six rats each. The negative control group was administered water (vehicle) via intragastric cannulation, the positive control group was administered acarbose (10 mg/kg b.w.), and the treatment groups were administered each herbal extract: I. sonorae, C.

limetta, and B. media (6 mL/kg b.w.). This concentration of herbal extracts was based on previous studies conducted by our research group, which demonstrated effective results in similar experimental settings (Hernández-Saavedra et al., 2016). After 5 min, a starch solution (5 g/kg b.w.) was admi- nistered to all rats. Glucose was measured at 0, 15, 30, 60, 90, 120, 150, and 180 min., using blood obtained from the tail vein with an Accu-Chek System (Roche Diagnosis, Germany).


Evaluation of in vitro α-amylase and α-glucosidase enzy- mes inhibitory activity

Herbal extracts were evaluated against α-amylase and α-glucosidase according to the methods described by Apos- tolidis et al. (2007). Results were reported as percentage of inhibition.

Cell culture

3T3-L1 preadipocytes (1 x 104 cells) were cultured until con- fluence in Dulbecco’s modified Eagle’s medium (DMEM) with 10 % fetal bovine serum (FBS, Gibco by Life Technologies, New York, USA) and 1 % penicillin/streptomycin (Invitrogen). The differentiation of adipocytes was induced with 5 µg/mL insulin (Life Technologies), 0.25 µM dexamethasone (Sigma- Aldrich, Missouri, USA), 0.5 mM IBMX (Invitrogen), 2 % FBS, and 1 % penicillin/streptomycin in DMEM for 2 days. Then, the medium was replaced with DMEM with 5 µg/mL insulin, 2 % FBS, and 1 % penicillin/streptomycin for 1 day. Finally, the medium was replaced with DMEM with 2 % FBS and 1

% penicillin/streptomycin for 4 days. Mediums were changed every two days during adipocytes proliferation and differen- tiation.

In vitro glucose uptake assay

Differentiated cells were serum-starved for 4 h and were treated with 0.5 and 1.0 mg/mL of lyophilized herbal ex- tracts dissolved in 0.1 % dimethyl sulfoxide (DMSO, Sigma- Aldrich). This concentration of herbal extracts was based on previous studies conducted by our research group, which demonstrated effective results in similar experimental settings (Pérez-Ramírez et al., 2017). Negative control cells were incubated with 0.1 % DMSO, whereas positive control cells were incubated with 100 nM of insulin. Glucose uptake assays were performed as described previously by Urso et al. (1999). Cells were incubated at 37 ºC for 30 min. Glucose con- centration was determined using colorimetric-commercial kit (Spinreact, Girone, Spain). All experiments were assessed in triplicate.


Relative quantification of Glut4, Irs1, and Pi3k gene trans- cripts

As in glucose uptake assay, fully differentiated adipocytes were incubated at 37 ºC for 30 min with each treatment. Afterwards, cells were lysed with 1 mL of Trizol reagent (Ther- mo Fisher Scientific, Delaware, USA) for the isolation of total RNA according to the instructions of manufacturer. RNA was quantified with a NanoDrop 1000 spectrophotometer (Ther-

mo Fisher Scientific). RNA quality was assessed with 260/280 and 260-230 ratios, whereas RNA integrity was assessed by electrophoresis on a 1 % denaturing agarose gel.

The cDNA synthesis was carried out with 1 µg of total RNA, 4 µL of M-MLV 5X reaction buffer (Invitrogen, Ca, USA), 1 µL of 5 µM oligo dT (Invitrogen), 1 µL of 10 mM dNTP mix (Invitrogen), 1 µL (200 U) of M-MLV RT (Invitrogen), 3 µL of 0.1 M DTT (Invitrogen), 1 µL of RNase inhibitor (Invitrogen), and RNAse-free water (Sigma-Aldrich), reaching a final volume of 20 µL. The reaction was incubated at 42 °C for 50 min and was stopped by incubating at 70 °C for 15 min.

The RT-PCR reaction was carried out with Sybr Green qPCR master mix (Thermo Scientific) according to the ins- tructions of the manufacturer using a StepOne Real-Time PCR System (Applied Biosystems. California, USA). Target transcripts were amplified using the following primers: Glut4: sense 5’-TCATCAGGATAAACAGCAG-3’ and antisense 5’-TAC- TATTGTGTTCCTTTGC-3’, Pi3k: sense 5’-CATGTTCTGGAAACTT- CACCA-3’ and antisense 5’CCTGGGGAAACATAAACTTG-3’,

and Irs1: sense 5’-CCTCACAGTCTTCAGTGGCT-3’ and anti- sense 5’-ATAGTCCCCATTTCCTTTGC-3’. Gene expression was assessed in duplicate. Relative quantification was carried out with the 2-∆∆CT method (Livak and Schmittgen, 2001) using Actin (sense 5’- GCTACAGCTTCACCACCACA-3’ and antisense 5’- AGTTTCATGGATGCCACAGG-3’) and 18S (sense 5’- GGAGC- GATTTGTCTGGGTTA-3’ and antisense 5’- GTAGGGTAGGCA-

CACGCTGA-3’) as reference genes.

Polyphenol profile

The polyphenol profile was assessed by Ultra-Performance Li- quid Chromatography (UPLC) coupled to a Quadrupole-Time of Flight (QTOF) with an atmospheric pressure Electrospray Ionization (ESI) interphase (Vion, Waters Co, MA, USA). The column used was an Acquity BEH C18 (100 x 2.1 mm, 1.7 µm) using the conditions reported previously by Rodríguez-Gon- zalez et al. (2018). Data acquisition was carried out with UNIFI Scientific Information System (Waters Co). Polyphenols were identified by comparison of elemental composition with the exact mass of the pseudo molecular ion (error mass < 5 ppm) using Phenol-Explorer database (http://phenol-explorer.eu), analysis of fragmentation pattern and isotopic distribution. The high-resolution mass spectra of the main polyphenols identified in the herbal extracts are shown in Figures 1S-9S.

Data analysis

Statistical significance was evaluated using ANOVA followed by Dunnet’s test (p < 0.05). Identification of polyphenols linked to the extracts’ antidiabetic potential was assessed by Partial Least Square-Discriminant Analysis (PLS-DA) and Va- riable Importance in the Projection (VIP) vs Coefficient score plots. JMP software (v11.0, SAS Institute, NC, USA) facilitated all statistical analyses.

RESULTS

Effect of herbal extracts on glucose absorption

Blood glucose levels peaked at 30 min after ingesting 5 g/kg of starch under fasting conditions (Figure 1). C. limetta extract


Figure 1. Effect of I. sonorae, C. limetta, and B. media aqueous extracts on postprandial blood glucose after a starch load (5 g/kg b.w.) in healthy Wistar rats. Data are shown as mean values (n = 6) and error bars represent standard error. Different letters in the same experimentation time indicate significant (p < 0.05) differences by Tukey’s test.

Figura 1. Efecto de extractos acuosos de I. sonorae, C. limetta y B. media sobre la glucosa postprandial en sangre tras una carga de almidón (5 g/kg de peso corporal) en ratas Wistar sanas. Los datos son mostrados como media (n = 6) y las barras de error representan el error estándar. Letras diferentes en el mismo tiempo experimental indican diferencias significativas (p < 0.05) con la prueba de Tukey.

slightly reduced the hyperglycemic peak (0.12-fold), while B. media and I. sonorae decreased (p < 0.05) postprandial glu- cose levels by 0.34 - 0.36-fold. This hints a mild postprandial antihyperglycemic activity. In vitro tests demonstrated that I. sonorae and B. media had the highest inhibitory α-amylase activity (25.4 and 20.5 %, respectively) as compared to C. limetta (5.3 %) (Table 1). Acarbose exerted even greater inhibitory activity (88.1 %). For α-glucosidase, C. limetta and

B. media showed similar low inhibition (5.1 and 4.4 %, respec- tively), while I. sonorae lacked this effect.


Effect of herbal extracts on glucose uptake in 3T3-L1 adipocytes

All herbal extracts increased (p < 0.05) glucose uptake as compared to control cells (Figure 2). C. limetta and I. sonorae


Table 1. In vitro inhibitory activity of I. sonorae, C. limetta, and B. media

aqueous extracts against carbohydrates digestive enzymes.

Tabla 1. Actividad inhibitoria in vitro de extractos acuosos de I. sonorae, C. limetta y B. media contra enzimas de digestión de carbohidratos.

Herbal extracts

-Amylase inhibition

-Glucosidase inhibition

I. sonorae

25.4 ± 0.3b

ND

C. limetta

5.3 ± 0.5

5.1 ± 0.5b

B. media

20.5 ± 2.0c

4.4 ± 0.4b

Acarbose

88.1 ± 6.3a

81.9 ± 6.6a

Values are expressed as maximum percentage of inhibition (%). Data are expressed as mean values ± standard deviation (n = 3). Means within a column followed by the same letter are not significantly different (p < 0.05) by Tukey’s test. ND: not detected.Los valores están expresados como porcentaje máximo de inhibición (%). Los datos son presentados como media ± desviación estándar (n = 3). Las medias dentro de una columna seguida de la misma letra no son significativamente diferentes (p < 0.05) por

la prueba de Tukey. ND: no detectado.


Figure 2. Effect of I. sonorae, C. limetta, and B. media aqueous extracts on glucose uptake in 3T3 L1 adipocytes. Data are showed as mean values (n = 3) and error bars represent standard error. Different letters indicate significant differences (p < 0.05) by Tukey’s test.

Figura 2. Efecto de extractos acuosos de I. sonorae, C. limetta y B. media sobre la internalización de glucosa en adipocitos 3T3 L1. Los datos son mostrados como media (n = 3) y las barras de error representan el error estándar. Letras diferentes indican diferencias significativas (p < 0.05) por la prueba de Tukey.

led to lower extracellular glucose levels (0.22-0.38-fold) as compared to the control group. Remarkably, B. media signi- ficantly reduced extracellular glucose levels (0.68-0.76-fold) at both concentrations, comparable to effect of insulin at the highest B. media concentration (1.0 mg/mL). B. media extract induced significant overexpression of Glut4, Irs1, and Pi3k ge- nes (2.27-2.85, 1.98 - 2.60, and 2.51 - 3.19-fold, respectively) as compared to control cells, showcasing a dose-response relationship (Figure 3).


Polyphenol profile of herbal extracts

A comprehensive analysis of the polyphenol profiles was conducted (Table 2). I. sonorae had 31 identified phenolic compounds, including eight hydroxybenzoic acids, twelve hydroxycinnamic acids, three flavanones, six flavonols, and two hydroxycoumarins. C. limetta roots extract had 39 phe- nolic compounds, including eleven hydroxybenzoic acids, twenty hydroxycinnamic acids, one flavanol, four flavanones, two flavonols, and one hydroxycoumarin. B. media contained 28 identified polyphenols, including twelve hydroxybenzoic acids, ten hydroxycinnamic acids, one flavanol, four flavano- nes, three flavonols, and two hydroxycoumarins. Multivariate analyses linked phenolic compounds to the previously des- cribed antidiabetic potential. I. sonorae and B. media extracts showed the highest in vitro α-amylase inhibitory activity (Table 1), which was linked mainly to hydroxybenzoic acid hexoside (PA_5), feruloylquinic acid (PA_24), and benzoic acid (PA_10) (Figure 4). Accordingly, these compounds were most abundant in these extracts (Table 2).

Eighteen polyphenols were associated with reduced glucose uptake (Figure 5): dihydroxybenzoic acid isomer I (PA_4), hydroxybenzoic acid isomer II (PA_7), rosmarinic acid (PA_36), protocatechuic acid (PA_8), dihydroxybenzoic acid

isomer II (PA_9), caffeoylquinic acid isomer II (PA_21), hydro- xybenzoic acid isomer I (PA_3), ellagic acid (PA_29), quercetin (F_17), naringin malonate (F_7), kaempferol hexoside- rhamnoside (F_13), narirutin (F_3), kaempferol sophoroside (F_12), naringenin (F_8), esculetin (C_3), hydroxycoumarin (C_4), and psoralen (C_1). These polyphenols were found in greater amount in B. media (Table 2), which exerted the greatest effect on glucose uptake in adipocytes (Figure 2). Additionally, eighteen polyphenols were linked to Glut4, Irs1, and Pi3k gene overexpression. Seventeen of these were also associated with an increase in glucose uptake of B. media (Fig. 5B), while sinapoylquinic acid isomer II (PA_32) was linked solely to gene overexpression in the insulin cascade pathway, not increased glucose uptake.


DISCUSSION

In summary, our study deepens into the anti-diabetic po- tential of three traditionally used herbal extracts (I. sonorae, C. limetta, and B. media) to elucidate their mechanisms of action and the associated polyphenols. Our findings align with the accumulated evidence of the role of polyphenols in metabolic disorders, particularly diabetes, beyond their an- tioxidant and anti-inflammatory properties. We adopted an in vitro multi-pronged approach to assess their mechanisms, focusing on their impact on glucose absorption and uptake. Regarding glucose intestinal absorption, all extracts significantly mitigated postprandial hyperglycemia, with B. media and I. sonorae extracts exhibiting the most pronoun- ced effects. This anti-hyperglycemic activity was particularly linked to their a-amylase inhibitory activity, an enzyme pivo- tal in starch digestion. In this regard, it has been previously reported that C. limetta peel extract inhibits α-amylase activi-

ty (Padilla-Camberos et al., 2014).

Notably, polyphenols such as hydroxybenzoic acid hexoside, benzoic acid, and feruloylquinic acid were key contributors to these effects, which can inhibit competitively or allosterically these digestive enzymes (Şöhretoğlu et al., 2023). Accordingly, it has been previously reported that an extract of phenolic compounds from green coffee beans, rich in feuroylquinic acids and caffeoylquinic acids, inhibited α-amylase activity (Cheng et al., 2019).

The second facet of our research, glucose uptake in 3T3 L1 adipocytes, produced interesting outcomes. B. media emerged as a potent inducer of glucose uptake, exerting a similar effect to insulin. These results suggest the insulin- mimetic potential of B. media, which was associated with the over-expression of Glut4, Irs1, and Pi3k genes, all critical elements in the insulin signaling cascade. This enhanced glucose uptake and expression of key genes in adipocytes is particularly valuable in the context of insulin resistance, a common characteristic of type 2 diabetes mellitus. Zapata- Bustos et al. (2014) reported that the extract of I. sonorae stimulates the glucose uptake in 3T3-F442A and 3T3-L1 adi- pocytes in a concentration-dependent manner, whereas this is the first study that reports this hypoglycemic mechanism for C. limetta roots and the B. media plant.

Table 2. Polyphenol profile of I. sonorae, C. limetta, and B. media aqueous extracts by UPLC-QTOF MSE.

Tabla 2. Perfil de polifenoles de extractos acuosos de I. sonorae, C. limetta y B. media por UPLC-QTOF MSE.

Code

Polyphenols

Rt (min)

Molecular Formula

Expected mass (Da)

Observed mass (Da)

Mass error (ppm)

I. sonorae

C. limetta

B. media

PA_1

Hydroxybenzoic acids Hydroxybenzoic acid isomer I Protocatechuic acid hexoside Vanillic acid

Dihydroxybenzoic acid isomer I Hydroxybenzoic acid hexoside Gallic acid ethyl ester Hydroxybenzoic acid isomer II Protocatechuic acidDihydroxybenzoic acid isomer II Benzoic acid

Syringic acid

Hydroxybenzoic acid isomer III Hydroxycinnamic acids

Caffeoyl tartaric acid Cinnamic acidCaffeoylquinic acid isomer I Caffeic acid ethyl ester

Coumaroylquinic acid isomer I Ferulic acid

Ferulic acid hexoside Coumaroyl hexose Caffeoylquinic acid isomer II Coumaroyl glycolic acid Ellagic acid hexoside Feruloylquinic acid Coumaric acidSinapoylquinic acid isomer I Ellagic acid pentoside

Coumaroylquinic acid isomer II Ellagic acid

Isoferulic acid Dicaffeoylquinic acid isomer I Sinapoylquinic acid isomer II Dicaffeoylquinic acid isomer II Caffeoylquinic acid isomer III

Dicaffeoylquinic acid isomer III Rosmarinic acid

Flavanols

(Epi)-catechin hexose Flavanones Neoeriocitrin Narirutin

NaringinHesperidinEriodictyol Naringin malonate Naringenin

Flavonols Myricetin rutinoside Rhamnetin hexoside Myricetin hexoside

Kaempferol sophoroside Kaempferol hexoside-rhamnosi- de

Quercetin hexoside

Quercetin malonyl-hexoside Kaempferol malonyl-hexoside Quercetin

Furanocoumarins Psoralen

Hydroxycoumarins Esculin

Esculetin Hydroxycoumarin

1.22

C7H6O3

138.0317

138.0312

-3.3747

1293.3

1772.8

620.4

PA_2

1.63

C13H16O9

316.0794

316.0790

-1.4622

21512.3

347697.0

5984.6

PA_3

1.70

C8H8O4

168.0423

168.0417

-3.2769

ND

26039.5

51512.0

PA_4

1.94

C7H6O4

154.0266

154.0266

-0.1933

668.5

446.8

20571.0

PA_5

2.20

C13H16O8

300.0845

300.0834

-3.6896

3536.4

487.7

3027.3

PA_6

2.36

C9H10O5

198.0528

198.0523

-2.5838

ND

96098.1

20600.6

PA_7

3.13

C7H6O3

138.0317

138.0315

-1.4297

1614.8

966.4

87775.5

PA_8

3.32

C7H6O4

154.0266

154.0259

-4.5348

853.3

444.1

1993.8

PA_9

4.21

C7H6O4

154.0266

154.0259

-4.7078

ND

ND

506.4

PA_10

4.27

C7H6O2

122.0358

122.0368

0.3424

2721.0

829.1

1300.0

PA_11

4.48

C9H10O5

198.0528

198.0523

-2.5853

ND

16507.5

11358.3

PA_12

6.67

C7H6O3

138.0317

138.0316

-0.3344

489.8

3217.6

1763.7

PA_13

1.08

C13H12O9

312.0481

312.0475

-1.9704

ND

938.1

ND

PA_14

1.45

C9H8O2

148.0524

148.0519

-3.6684

3911.7

ND

ND

PA_15

2.57

C16H18O9

354.0951

354.0954

0.8135

6010.1

450.2

ND

PA_16

3.52

C11H12O4

208.0736

208.0731

-2.1574

ND

2734.5

ND

PA_17

3.68

C16H18O8

338.1002

338.1005

1.0962

ND

747.8

ND

PA_18

3.68

C10H10O4

194.0579

194.0576

-1.8077

ND

2984.7

ND

PA_19

3.69

C16H20O9

356.1107

356.1107

-0.1713

ND

8275.5

394.7

PA_20

4.09

C15H18O8

326.1002

326.0996

-1.7866

1237.7

1070.8

ND

PA_21

4.17

C16H18O9

354.0951

354.0943

-2.0983

1451.5

999.5

10313.2

PA_22

4.45

C11H10O5

222.0528

222.0523

-2.4526

ND

879.9

5151.3

PA_23

4.73

C20H16O13

464.0591

464.0602

2.3819

ND

690.2

ND

PA_24

5.12

C17H20O9

368.1107

368.1103

-1.1118

1715.0

ND

1050.0

PA_25

5.18

C9H8O3

164.0473

164.0472

-0.8242

613.7

1563.4

780.1

PA_26

5.23

C18H22O10

398.1213

398.1200

-3.3438

ND

446.0

ND

PA_27

5.32

C19H14O12

434.0485

434.0492

1.4832

ND

2870.2

ND

PA_28

5.34

C16H18O8

338.1002

338.0999

-0.7703

ND

502.8

369.2

PA_29

5.63

C14H6O8

302.0063

302.0067

1.2807

ND

1194.1

1477.4

PA_30

5.73

C10H10O4

194.0579

194.0576

-1.8164

539.7

4217.5

2327.3

PA_31

6.29

C25H24O12

516.1268

516.1257

-2.0080

72712.9

2850.5

ND

PA_32

6.36

C18H22O10

398.1213

398.1231

4.6479

ND

1554.8

1459.7

PA_33

6.39

C25H24O12

516.1268

516.1260

-1.5250

28039.8

1634.2

2519.1

PA_34

6.40

C16H18O9

354.0951

354.0947

-0.9726

4404.7

ND

ND

PA_35

6.73

C25H24O12

516.1268

516.1272

0.8322

75354.8

2355.7

ND

PA_36

6.82

C18H16O8

360.0845

360.0841

-1.1864

586.0

ND

1831112.9

F_1

5.20

C21H24O11

452.1319

452.1328

2.0200

ND

1731.3

678.3

F_2

4.50

C27H32O15

596.1741

596.1736

-0.8151

679.8

802.3

ND

F_3

6.37

C27H32O14

580.1792

580.1788

-0.7378

ND

ND

5223.6

F_4

6.57

C27H32O14

580.1792

580.1796

0.6403

ND

190414.1

ND

F_5

6.76

C28H34O15

610.1898

610.1892

-0.9442

ND

662410.3

1141.0

F_6

7.73

C15H12O6

288.0634

288.0636

0.5717

191.1

ND

ND

F_7

8.03

C36H44O22

828.2324

828.2318

-0.7540

ND

ND

788.0

F_8

8.68

C15H12O5

272.0685

272.0686

0.3929

816.7

724.7

917.0

F_9

5.17

C27H30O17

626.1483

626.1494

1.6816

358.1

ND

ND

F_10

5.25

C22H22O12

478.0747

478.0757

2.1079

391.5

ND

ND

F_11

5.27

C21H20O13

480.0904

480.0905

0.1751

2583.4

3125.6

ND

F_12

5.72

C27H30O16

610.1534

610.1535

0.1290

ND

ND

5046.0

F_13

5.86

C27H30O15

594.1585

594.1605

3.4310

ND

ND

1449.4

F_14

5.90

C21H20O12

464.0955

464.0953

-0.4036

16498.3

735.0

ND

F_15

6.18

C24H22O15

550.0959

550.0948

-2.0132

246.8

ND

ND

F_16

6.75

C24H22O14

534.1010

534.0992

-3.2217

402.2

ND

ND

F_17

7.90

C15H10O7

302.0427

302.0422

-1.5475

ND

ND

431.4

C_1

6.92

C11H6O3

186.0317

186.0320

1.3986

ND

ND

29060.3

C_2

3.20

C15H16O9

340.0794

340.0786

-2.5525

1690.6

1427.8

ND

C_3

4.06

C9H6O4

178.0266

178.0260

-3.3713

403.6

ND

26338.5

C_4

6.82

C9H6O3

162.0317

162.0318

0.7271

ND

ND

112663.3

Results are expressed as arbitrary units. Data are shown as mean values (n = 3). Rt: retention time; ND: not detected; PA: phenolic acids; F: flavonoids; C: coumarins. Identification confirmed with commercial standards.

Los resultados están expresados como unidades arbitrarias. Los datos son mostrados como media (n = 3). Rt: tiempo de retención; ND: no detectado; PA: ácidos fenólicos; F: flavonoides; C: cumarinas. Identificación confirmada con estándares comerciales.


Figure 3. Effect of I. sonorae, C. limetta, and B. media aqueous extracts on Glut4, Irs1, and Pi3k relative expression in 3T3 L1 adipocytes. Relative expression was estimated in comparison with negative control cells. Data are showed as mean values (n = 3) and error bars represent standard error. Different letters for each gene indicate significant differences (p < 0.05) by Tukey’s test. Glut4: glucose transporter type 4, Irs1: insulin receptor substrate-1, Pi3k: phosphatidylinositol 3-kinase.

Figura 3. Efecto de extractos acuosos de I. sonorae, C. limetta y B. media sobre la expresión relativa de Glut4, Irs1 y Pi3k en adipocitos 3T3 L1. La expresión relativa fue estimada en comparación con las células de control negativo. Los datos son mostrados como media (n = 3) y las barras de error representan el error estándar. Letras diferentes para cada gen indican diferencias significativas (p < 0.05) por la prueba de Tukey. Glut4: transportador de glucosa tipo 4, Irs1: sustrato del receptor de insulina 1; Pi3k: fosfatidilinositol 3-quinasa.


Figure 4. Association between the polyphenol profile of I. sonorae, C. limetta, and B. media aqueous extracts and their antidiabetic potential assessed by α-amylase inhibitory activity. VIP: variable importance in the projection.

Figura 4. Asociación entre el perfil de polifenoles de extractos acuosos de I. sonorae, C. limetta y B. media y su potencial antidiabético determinado por la actividad inhibitoria de α-amilasa. VIP: importancia de la variable para la proyección.

Our detailed multivariate analysis suggested several polyphenols associated with the observed effects, some of which have been previously reported to impact glucose uptake and insulin signaling. For instance, it has been repor- ted that rosmarinic acid (Vlavcheski et al., 2017), naringin (Dayarathne et al., 2021), ellagic acid (Kábelová et al., 2021) and kaempferol (Moore et al., 2023) increase glucose uptake


Figure 5. Association between the polyphenol profile of I. sonorae, C. limetta, and B. media aqueous extracts and their antidiabetic potential assessed in 3T3 L1 adipocytes by glucose uptake (A) and the relative expression of Glut4, Irs1, and Pi3k (B). Glut4: glucose transporter type 4, Irs1: insulin receptor substrate-1, Pi3k: phosphatidylinositol 3-kinase, VIP: variable importance in the projection.

Figura 5. Asociación entre el perfil de polifenoles de extractos acuosos de I. sonorae, C. limetta y B. media y su potencial antidiabético determinado por la internalización de glucosa en adipocitos 3T3 L1 (A) y la expresión relativa de Glut4, Irs1 y Pi3k (B). Glut4: transportador de glucosa tipo 4, Irs1: sustrato del receptor de insulina 1; Pi3k: fosfatidilinositol 3-quinasa; VIP: importancia de la variable para la proyección.

in skeletal muscle cells and adipocytes. Protocatechuic acid is an hydroxybenzoic acid which stimulates insulin signaling pathways in myocytes, hepatocytes and adipocytes. Specifi- cally, this phenolic acid significantly increased the expression and protein level of IRS-1 and GLUT4 in insulin resistant adi- pocytes (Shakoor et al., 2023).

Numerous studies have reported the ability of polyphe- nols to increase insulin-dependent glucose uptake via GLUT4 activation by up-regulating Pi3k expression (Williamson and Sheedy, 2020). Chlorogenic acid and quercetin promote glucose uptake by increasing Glut4 mRNA levels and its translocation to cell membrane (Gannon et al., 2014). Raciti et al. (2018) evaluated a citrus extract rich in hesperidin and narirutin in adipocytes 3T3-L1 and reported increased Glut4 mRNA levels; however, the administration of the isolated compounds did not show the same effect, suggesting a

synergistic mechanism. Altogether, these results further credence to the notion that polyphenols play a pivotal role in the mechanisms behind the anti-diabetic effects of herbal extracts.


CONCLUSIONS

This study identifies the mechanism underlying the potential glucose-regulating properties of B. media and I. sonorae, suggesting their viability as anti-diabetic agents in tradi- tional medicine or as ingredients for functional beverages. Furthermore, the identification of polyphenols associated with these effects adds to the growing evidence highlighting the potential of these natural compounds in mitigating metabolic disorders, offering a promising opportunity for the development of novel therapeutic strategies. Further investigations are warranted to validate the impact of these herbal extracts on insulin resistance and glucose intolerance through intervention studies.


ACKNOWLEDGMENTS

We thank the Chemistry Faculty of the Universidad Autóno- ma de Querétaro for providing facilities for the UPLC-QTOF analysis (CONACyT project number: INFR-15-255182). This work was supported by Fondos Mixtos de Fomento a la Investigación Científica y Tecnológica CONACYT-Querétaro (Grant no. 2012-CO1-193469).


CONFLICTS OF INTEREST

The authors declare no actual or potential conflict of inter- ests, including financial, personal or relationship with other organizations.


REFERENCES

Alarcón-Aguilar, F.J., Calzada-Bermejo, F., Hernandez-Galicia, E., Ruiz-Angeles, C. and Roman-Ramos, R. 2005. Acute and chronic hypoglycemic effect of Ibervillea sonorae root extracts-II. Journal of Ethnopharmacology. 97: 447-452.

Apostolidis, E., Kwon, Y.I. and Shetty, K. 2007 Inhibitory potential of herb, fruit, and fungal-enriched cheese against key enzymes linked to type 2 diabetes and hypertension. Innovative Food Science & Emerging Technologies. 8: 46-54. Cheng, K., Wenjiang, D., Long, Y., Zhao, J., Hu, R., Zhang, Y. and Zhu, K. 2019. Evaluation of the impact of different drying methods on the phenolic compounds, antioxidant activity, and in vitro digestion of green coffee beans. Food Science &

Nutrition. 7: 1084-1095.

Dayarathne, L.A., Ranaweera, S.S., Natraj, P., Rajan, P., Lee, Y.J. and Han, C.H. 2021. The effects of naringenin and naringin on the glucose uptake and AMPK phosphorylation in high glucose treated HepG2 cells. Journal of Veterinary Science, 22: e92.

Flores-Fernández, J., Barragán-Álvarez, C., Díaz-Martínez, N., Villanueva-Rodríguez, S. and Padilla-Camberos, E. 2017. In vitro and in vivo postprandial glycemic activity of Citrus limetta peel flour. Pharmacognosy Magazine. 13: 613-616.

Gannon, N., Conn, C. and Vaughan, R. 2014. Dietary stimulators of GLUT4 expression and translocation in skeletal muscle: a

mini‐review. Molecular Nutrition & Food Research. 59: 48-64. Hernández-Saavedra, D., Pérez-Ramírez, I.F., Ramos-Gómez, M., Mendoza-Díaz, S., Loarca-Pina, G. and Reynoso-Camacho,

R. 2016. Phytochemical characterization and effect of Calendula officinalis, Hypericum perforatum, and Salvia officinalis infusions on obesity-associated cardiovascular risk. Medicinal Chemistry Research. 25: 163-172.

Kábelová, A., Malínská, H., Marková, I., Oliyarnyk, O., Chylíková,

B. and Šeda, O. 2021. Ellagic acid affects metabolic and transcriptomic profiles and attenuates features of metabolic syndrome in adult male rats. Nutrients. 13: 804.

Lane, J., Strich, H., Taylor, A., Timmermann, B., Malone, D., Teufel-Shone, N., Drummond, R., Woosley, R., Pereira, E. and Martinez, A. 2006. Use of herbal remedies by diabetic Hispanic women in the southwestern United States. Phytotherapy Research. 20: 250-255.

Livak, K.J. and Schmittgen, T.D. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2-∆∆Ct method. Methods. 25: 402-408.

Moore, W.T., Luo, J. and Liu, D. 2023. Kaempferol improves glucose uptake in skeletal muscle via an AMPK-dependent mechanism. Food Science and Human Wellness. 12: 2087- 2094.

Padilla-Camberos, E., Lazcano-Díaz, E., Flores-Fernandez, J.M., Owolabi, M.S., Allen, K. and Villanueva-Rodríguez, S. 2014. Evaluation of the inhibition of carbohydrate hydrolyzing enzymes, the antioxidant activity, and the polyphenolic content of Citrus limetta peel extract. The Scientific World Journal. 2014: 121760.

Pérez-Ramírez, I.F., González-Dávalos, M.L., Mora, O., Gallegos- Corona, M.A. and Reynoso-Camacho, R. 2017. Effect of Ocimum sanctum and Crataegus pubescens aqueous extracts on obesity, inflammation, and glucose metabolism. Journal of Functional Foods. 35: 24-31.

Raciti, G.A., Fiory, F., Campitelli, M., Desiderio, A., Spinelli, R., Longo, M., Nigro, C. and Miele, C. 2018. Citrus aurantium L. dry extracts promote C/ebpβ expression and improve adipocyte differentiation in 3T3-L1 cells. PloS One. 13: e0193704.

Rivera-Ramírez, F., Escalona-Cardoso, G.N., Garduño-Siciliano, L., Galaviz-Hernández, C. and Paniagua-Castro, N. 2011. Antiobesity and hypoglycaemic effects of aqueous extract of Ibervillea sonorae in mice fed a high-fat diet with fructose. Journal of Biotechnology and Biomedicine. 2011: 968984.

Rodríguez-González, S., Pérez-Ramírez, I.F., Castaño-Tostado, E., Amaya-Llano, S., Rodríguez-García, M.E., Reynoso-Camacho,

R. 2018. Improvement of physico-chemical properties and phenolic compounds bioavailability by concentrating dietary fiber of peach (Prunus persica) juice by-product. Journal of the Science of Food and Agriculture. 98: 3109.

Shakoor, H., Hussein, H., Al-Hassan, N., Alketbi, M., Kizhakkayil,

J. and Platat, C. 2023. The muscle-conditioned medium containing protocatechuic acid improves insulin resistance by modulating muscle communication with liver and adipose tissue. International Journal of Molecular Sciences. 24: 9490.

Şöhretoğlu, D., Renda, G., Arroo, R., Xiao, J. and Sari, S. 2023. Advances in the natural α‐glucosidase inhibitors. eFood. 4: e112.

Urso, B., Cope, D., Kallo, H., Hayward, A., Whitehead, J., O’Rahilly,

S. and Siddle, K. 1999. Differences in signaling properties of the cytoplasmic domains of the insulin receptor and insulin-

like growth factor receptor in 3T3-L1 adipocytes. Journal of Biological Chemistry. 274: 30864-30873.

Tran, N., Pham, B. and Le, L. 2020. Bioactive compounds in anti-diabetic plants: from herbal medicine to modern drug discovery. Biology. 9: 252.

Vlavcheski, F., Naimi, M., Murphy, B., Hudlicky, T. and Tsiani,

E. 2017. Rosmarinic acid, a rosemary extract polyphenol, increases skeletal muscle cell glucose uptake and activates AMPK. Molecules, 22: 1669.

Williamson, G. and Sheedy, K. 2020. Effects of polyphenols on insulin resistance. Nutrients. 12: 3135.

Zapata-Bustos, R., Alonso-Castro, A., Gómez-Sánchez, M. and Salazar-Olivo, L. 2014. Ibervillea sonorae (Cucurbitaceae) induces the glucose uptake in human adipocytes by activating a PI3K-independent pathway. Journal of Ethnopharmacology. 28: 546-52.