You are on page 1of 6

Plant Foods Hum Nutr

DOI 10.1007/s11130-016-0581-2

ORIGINAL PAPER

Evaluation of Total Flavonoids, Myricetin, and Quercetin


from Hovenia dulcis Thunb. As Inhibitors
of α-Amylase and α-Glucosidase
Yonghong Meng 1 & Anping Su 1 & Shuang Yuan 1 & Huaguo Zhao 2 & Siyuan Tan 1 &
Chingyuan Hu 3 & Hong Deng 1 & Yurong Guo 1

# Springer Science+Business Media New York 2016

Abstract This study was designed to investigate the inhibi- Keywords Flavonoids . T2DM . α-amylase .
tion effect and mechanism of total flavonoids, myricetin and α-glucosidase . Hovenia dulcis Thunb
quercetin extracted from Hovenia dulcis Thunb. on α-amylase
and α-glucosidase in order to explore the potential use of
Hovenia flavonoids in alleviating postprandial hyperglyce- Abbreviations
mia. The results demonstrate that total flavonoids, myricetin, ESI-MS The electrospray ionization mass
and quercetin were effective inhibitors of α-amylase with IC50 spectrometry
values of 32.8, 662 and 770 μg ml−1, respectively. And all H. dulcis Thunb. Hovenia dulcis Thunb.
three were effective inhibitors of α-glucosidase with IC50 NMR Nuclear magnetic resonance
values of 8, 3 and 32 μg ml−1, respectively. Enzyme kinetics T2DM Type 2 diabetes mellitus
tests and Lineweaver-Burk results showed the inhibition ef- TFs Total flavonoids
fects of total flavonoids, myricetin and quercrtin on α-amylase
were all reversible and competitive, and the effects on α-
glucosidase were all reversible but non-competitive. This Introduction
study revealed that Hovenia flavonoids, especially myricetin,
are effective and promising functional foods in alleviating Type 2 diabetes mellitus (T2DM) is a serious chronic meta-
type 2 diabetes mellitus. bolic disease which represents more than 90 % of all diabetes
patients [1]. Diet plays an important role in T2DM develop-
ment [2]. Regulating and retarding the activity of the major
Electronic supplementary material The online version of this article digestive enzymes, mainly pancreatic α-amylase and α-glu-
(doi:10.1007/s11130-016-0581-2) contains supplementary material, cosidase, has been reported to reduce the absorption rate of
which is available to authorized users. carbohydrates in the gastrointestinal tract, control or even pre-
vent T2DM and alleviate postprandial hyperglycemia [3].
* Yonghong Meng Pancreatic α-amylase and α-glucosidase are responsible for
mengyonghong@snnu.edu.cn
the hydrolysis of carbohydrates into disaccharides and mono-
saccharides, which are absorbed into the bloodstream from the
1
Shaanxi Engineering Lab for Food Green Processing and Security small intestine [4]. Thus, inhibition of these two enzymes
Control, College of Food Engineering and Nutritional Science, maybe an effective tool in alleviating hyperglycemia, and con-
Shaanxi Normal University, 620 West Chang’an Avenue, Chang’an,
Xi’an 710119, People’s Republic of China
sequently beneficial for T2DM patients.
2
Many studies have demonstrated that plant flavonoids have
Xi’an Healthful Biotechnology Co., Ltd., HangTuo Road, Chang’an,
Xi’an 710100, People’s Republic of China
the function of inhibiting enzymes to reduce blood sugar. For
3
example, Kim et al. [5] investigated the potential of sorghum
Department of Human Nutrition, Food and Animal Sciences, College
of Tropical Agriculture and Human Resources, University of Hawaii
in the development of effective anti-diabetic agents. Lin et al.
at Manoa, 1955 East West Road, AgSci 216, Honolulu, HI 96822, [6] revealed the flavonoids extracted from guava fruit signif-
USA icantly reduced the glucose and blood urea levels in the
Plant Foods Hum Nutr

diabetic mice model. And Dalar et al. [7] found that Then the extracts were concentrated under vacuum to a
E. bornmuelleri leaf extracts can modulate metabolism of liquid density range of 1.02–1.05 g ml−1, and then filtered
sugars through inhibition of the relevant digestive enzymes. for subsequent studies. D101 macroporous resin was applied
In this study, we were interested in finding more potent α- to separate the extracts with resin sample loading amount of
amylase and α-glucosidase inhibitors from Hovenia dulcis 125 mg g−1. The impurities were first removed using distilled
Thunb. (H. dulcis Thunb.), a natural plant, which belongs to a water and followed by 5–10 % (v/v) ethanol. The eluent was
small genus of Rhamnaceae family, being widely distributed in collected using 65 % (v/v) ethanol, concentrated by vacuum,
southern and southwestern China [8]. The ancient Chinese med- and then subjected to a silica gel chromatography column,
icine records showed that the H. dulcis Thunb. (known as the eluted by petroleum ether (PE): ethyl acetate (EA) system.
longevity fruit in China) has many beneficial properties, such as Myricetin fraction was collected using PE:EA = 5:2 (v/v),
antifebrile, detoxifying alcoholic intoxication, slaking thirst and quercetin fraction using PE:EA = 3:1 (v/v). The ultimate elu-
helping to produce saliva, strengthening the stomach, and pro- ate was concentrated and dried once again. The dried sample
moting digestion [9]. The main chemical constituents of was repeatedly recrystallized with 100 % methanol to obtain
H. dulcis Thunb. are flavonoids, including kaempferol, apigenin, highly purified myricetin and quercetin. These purified pow-
myricetin, quercetin, dihydromyricetin and anthraquinones em- ders were used in the subsequent experiments for the determi-
odin. etc. [8, 10]. We found the total flavonoids (TFs), myricetin nation of chemical and enzymology analyses.
and quercetin among Hovenia chemical ingredients are quality
inhibitors of α-amylase and α-glucosidase, and their inhibition Determination of Structures of Main Active Compositions
ratio and mechanisms were elucidated by examining their enzy- of TFs
matic dynamics. Our results provide new knowledge for devel-
oping Hovenia flavonoids as a high value-added functional food. The electrospray ionization mass spectrometry (ESI-MS)
spectra were measured on a Bruker Apex III FT ion cyclotron
resonance mass spectrometer in the positive ionization mode
Materials and Methods at a potential of 4.0 kV (on the discharge needle). The 1H
nuclear magnetic resonance (NMR) spectra were recorded
Materials and Reagents on a Bruker AMX 400 spectrometer from Bruker Biospin
(Rheinstetten, Germany) operating at a proton frequency of
Mature H. dulcis Thunb. seeds were harvested from Xunyang 400.13 MHz. The resulting spectra were manually phased and
County of Shaanxi province of China. D101 macroporous the baselines were corrected manually.
resin (5.0 i.d. ×100 cm) was purchased from Nankai
Hecheng S&T (Tianjin, China). The pancreatic α-amylase, Chemical Analysis of Myricetin and Quercetin
α-glucosidase, rutin and 4-nitrophenylα-D-glucopyranoside
(pNPG) were purchased from Sigma-Aldrich (St. Louis, Thermo U3000 HPLC system and C18 column
MO, USA). The acetic acid and acetonitrile were HPLC grade (4.6 × 250 mm, Venusil, USA) were used for the analysis.
and purchased from Tianjin Hedong Red Cliff Chemical The flow rate of mobile phase consisted of 0.1 % acetic acid
Reagent Factory (Tianjin, China). All other reagents were of in pure water and pure acetonitrile at the flow rate of
analytical grade. 1 ml min−1. Eluent program was 20 % pure acetonitrile for
3 min, to 38 % for 24 min, to 90 % for 1 min, and then held at
Extraction, Quantification and Isolation of Main Active 90 % for 5 min, to 20 % for 6 min. The injection volume was
Compositions from TFs 10 μL, and UV absorbance was measured at 368 nm [13].

Dried mature seeds of H. dulcis Thunb. were milled into fine Inhibition Activity and Mechanism Study
powder and screened through a 60-mesh sieve to ensure their
homogeneity. Powdered samples (10 g) were reflux extracted One unit of α-amylase is the amount of enzyme that catalyzes
with 100 ml of 75 % ethanol for three times (1 h each time). and reduces 1 μmol sugar per minute at pH 6.9 and 25 °C. The
After using rotary evaporator to remove ethanol from the ex- rate of α-amylase reaction was assayed using the method of
tract, the residual extract fraction was frozen to dryness [11]. Oboh et al. [14] with slight modification. Briefly, 0.5 ml of
TFs was assayed using the method of Xie et al. [12]. The 0.02 M sodium phosphate buffer (pH 6.9 with 0.006 M NaCl),
concentration of TFs in extract was expressed as mg of rutin 0.5 ml of 0.54 U ml−1 α-amylase solution, and 0.5 ml of 1 %
equivalents per gram dry weight of extract. The calibration starch solution in 0.02 M sodium phosphate buffer were
curve was determined using the following equation: mixed and incubated at 25 °C for 3 min; enzyme reaction
Y = 0.031 X + 0.0347, R2 = 0.9992. Where X is absorbance stopped using 1 ml of dinitrosalicylic acid color reagent,
value of sample, Y is sample concentration. cooled to room temperature, and change in absorbance at
Plant Foods Hum Nutr

Fig. 1 HPLC chromatogram of


TFs and standard myricetin and
quercetin. The structures of
standard myricetin and quercetin
are shown above their peaks

540 nm per minute was measured using a UV previously described. Nonlinear fitting was employed to cal-
spectrophotometer. culate the IC50 values.
One unit of α-glucosidase is the amount of enzyme that In order to study the inhibitory effects of Hovenia flavonoids
catalyzes the hydrolysis of 1 mmol substrate (pNPG) on α-amylase kinetics, we used various concentrations (0.6,
per minute at pH 6.8 and 37 °C. The rate of α- 0.8, 1, 1.2, 1.4, and 1.6 %) of starch solution, various inhibitors
glucosidase reaction was assayed according to Xi et al. (8 μg ml−1 TFs, 50 μg ml−1 myricetin and 50 μg ml−1 quercetin
[15] with slight modification. The 2 ml reaction system solution) in the presence of 0.54 U/ml α-amylase. And for α-
consisted of 35 U ml−1 α-glucosidase, 2.02 × 10−4 mol L−1 glucosidase, various concentrations (0.75, 1.0, 1.25, 1.5, 1.75,
substrate pNPG, and 0.1 mol L−1 sodium phosphate buffer and 2 μg ml−1) of substrate, and various inhibitors (0.4 μg ml−1
(pH 6.8). After incubated at 37 °C for 10 min, the change in TFs, 1 μg ml−1 myricetin and 10 μg ml−1 quercetin solution) in
absorbance at 405 nm per minute was measured using a UV the presence of 35 U/ml α-glucosidase were used. Then the
spectrophotometer. Lineweaver-Burk plots were used to determine the inhibition
When inhibition mode on α-amylase was tested, the final mode by calculating the inhibition constant through nonlinear
concentrations of 32 μg ml−1 TFs, 100 μg ml−1 myricetin and regression. IC50 value is the concentration of a test substance
200 μg ml−1 quercetin solution were separately pre-incubated required to achieve half maximal inhibition rate of a given
with 1 % starch solution and different concentrations of each reaction. The inhibition rate is calculated using the following
enzyme at 25 °C for 20 min. When inhibition mode on formula:
α-glucosidase was tested, the final concentrations of .
0.4 μg ml−1 TFs, 0.5 μg ml−1 myricetin and The inhibition rateð%Þ ¼ ðA−AI Þ A*100%
5 μg ml−1 quercetin solution were separately preincubat-
ed with 5 mmol L−1 pNPG different concentrations of A = absorbance of reaction system without inhibitor.
each enzyme at 37 °C for 20 min. There is no inhibitor in AI = absorbance of reaction system with addition of
control system. Then the enzyme reaction was started as inhibitor.

Table 1 Inhibitory effect of flavonoids from H. dulcis Thunb. on α-amylase and α-glucosidase

α-amylase α-glucosidase

Regression equation IC50 Regression equation IC50


(μg ml−1) (μg ml−1)

TFs y = 1.1*10−3 × 3–0.061 × 2 + 1.51× + 26.34 32.8 y = −2*10−5 × 4–0.0096 × 3–0.36 × 2 + 4.71× + 32.86 8
Myricetin y = 56.008 × 4–249.51 × 3 + 345.27 × 2–113.42× + 35.42 662 y = −2*107 × 3 + 134599 × 2 + 2919.7× + 40.09 3
Quercetin y = 178.78 × 3–276.45 × 2 + 154.35× + 13.28 770 y = 2*106 × 4-251237 × 3 + 6163.4 × 2 + 504.7× + 33.76 32

Bx^ represents the different inhibitor concentration, By^ represents the inhibition rate
Plant Foods Hum Nutr

Fig. 2 Kinetics curves of different inhibitors (TFs, myricetin, and quercetin) on α- amylase and α-glucosidase (a α-amylase; b α-glucosidase)

Competitive inhibition formula: Statistical Analysis


 
1 Km ½I  1 1 All values were expressed as mean ± standard deviation, and
¼ 1þ þ ð1Þ
V Vmax Ki ½S  Vmax each value was the mean of separate experiments in triplicate.

Non-competitive inhibition formula:


   
1 Km ½I  1 1 ½I  Results and Discussion
¼ 1þ þ 1þ ð2Þ
V Vmax Ki ½S  Vmax Ki
Isolation of Active Flavonoid Ingredients
Where V is the initial reaction rate; Vmax is the maximum and its Structural Analyses
reaction rate; Km is the Michaelis-Menten constant; Ki is the
inhibition constant. [S] is the concentration of substrate; [I] is From the MS and NMR data, two main pure compounds were
the inhibitor concentration. identified as myricetin and quercetin with the molecular

Fig. 3 Lineweaver-Burk plot of different inhibitors on α-amylase (a TFs; b myricetin; c quercetin) and on α-glucosidase (d TFs; e myricetin; f
quercetin)
Plant Foods Hum Nutr

Table 2 Related parameters on


inhibition kinetics of flavonoids Double reciprocal R2 Km Vmax (OD Ki Inhibition
from H. dulcis Thunb. on α- equation (mg ml−1) min−1) (μg ml−1) type
amylase
TFs y = 48.57× + 2.07 0.99 24.28 0.5 12 competitive
Myricetin y = 49.58× + 1.99 0.99 24.79 0.5 70 competitive
Quercetin y = 46.48× + 1.99 0.99 23.24 0.5 90 competitive
Control y = 29.63× + 2.01 0.99 14.68 0.5 / competitive

Bx^ represents the reciprocal of substrate concentration, By^ represents the reciprocal of rate of catalysis reaction
rate, B/^ represents no inhibition constant (Ki) value

weight of 317.02 and 301.03, respectively (MS and NMR because excessive inhibition of α-amylase could lead to ab-
spectrum were shown in Supplementary Fig.1-3). HPLC re- normal bacterial fermentation of undigested starch in the co-
sults (Fig.1) show that myricetin and quercetin are the major lon [17]. Interestingly, it is significant to note that myricetin
components in H. dulcis Thunb., representing 63.2 and had the most inhibition (IC50 = 3) for α-glucosidase but lower
12.1 % in TFs, respectively. inhibition (IC50 = 662) for α-amylase, which suggests that
myricetin most likely will produce less side effect. In addition,
IC50 of the TFs, Myricetin and Quercetin for α-Amylase other studies have shown that myricetin has various other
and α-Glucosidase functions without toxicity, which may alleviate complications
of T2DM [11, 18]. Taken together, we have demonstrated that
Table 1 shows the inhibition rate regression equation and IC50 myricetin, one of the promising components extracted from
of three tested substrates on α-amylase and α-glucosidase. R2 H. dulcis Thunb. is a promising compound to alleviate post-
for each regression equation was 0.99, indicating good fitting prandial hyperglycemia or to be used as health care products.
degree. TFs was the most effective inhibitor of α-amylase
followed by myricetin, and quercetin with calculated IC50 The Inhibition Mode and Inhibition Kinetics
values of 32.8, 662 and 770 μg ml −1 , respectively. of α-Amylase and α-Glucosidase
Myricetin, however, was the most effective inhibitor of α-
glucosidase followed by TFs, and quercetin with calculated Effect of three tested compounds on initial reaction rates was
IC50 values of 3, 8 and 32 μg ml−1, respectively. measured as described in the methods section. When there is a
Acarbose is a well-known and typical anti-diabetic drug reversible inhibitor in the reaction system, its rate curve goes
because its IC50 for α-glucosidase is 17.8 μg ml−1 [4]. In through the origin. From Fig. 2 we can see that all the rate
contrast, IC50 of myricetin (3 μg ml−1) for α-glucosidase lines went through the origin, so the inhibition type of TFs,
was six times lower than that for acarbose. And the IC50 of myricetin and quercetin for α-amylase and α-glucosidase was
TFs for α-glucosidase was 8 μg ml−1, its effectiveness was all reversible inhibition. Reversible inhibition mode doesn’t
also better than that of acarbose. Therefore, Hovenia flavo- have accumulation effect in vivo, thus it should be safer to
noids exerted a higher inhibition effect on α-glucosidase, be used as a functional food.
which means Hovenia flavonoids are potentially better alter- Lineweaver-Burk plots were generated to clarify inhibition
natives for T2DM patients than acarbose. In addition, tradi- mode. As shown in Fig. 3a, b, c, all inhibition kinetics curves
tional oral hypoglycemic drugs for diabetes have undesirable crossed y axis, so the inhibition type of TFs, myricetin and
side effects, such as liver toxicity, kidney damage and adverse quercetin for α-amylase was competitive. From the
gastrointestinal symptoms [5, 16]. It is worth noting that pre- Michaelis-Menten equation in Table 2, the value of the verti-
vious reports have demonstrated that mild α-amylase inhibi- cal axis intercept (1/Vmax) remained unchanged with all three
tory activity is useful to alleviate gastrointestinal malaise, compounds, indicating that the TFs, myricetin and quercetin

Table 3 Related parameters on


inhibition kinetics of flavonoids Double reciprocal R2 Km Vmax Ki Inhibition type
from H. dulcis Thunb. on α- equation (mg ml−1) (OD min−1) (μg ml−1)
glucosidase
TFs y = 57.32× + 40.98 0.99 1.4 0.02 0.22 Non-competitive
Myricetin y = 81.20× + 58.73 0.99 1.4 0.01 0.32 Non-competitive
Quercetin y = 57.07× + 40.37 0.99 1.4 0.02 50 Non-competitive
Control y = 21.16× + 14.21 0.99 1.4 0.07 / Non-competitive

Bx^ represents the reciprocal of substrate concentration, By^ represents the reciprocal of rate of catalysis reaction
rate, B/^ represents no inhibition constant (Ki) value
Plant Foods Hum Nutr

were competitive inhibitor for α-amylase. And the value of 3. Bhandari MR, Jong-Anurakkun N, Hong G, Kawabata J (2008) α-
glucosidase and α-amylase inhibitory activities of Nepalese medic-
the inhibition constant (Ki) was 12, 70 and 90 μg ml−1, re-
inal herb Pakhanbhed (Bergenia Ciliata, haw. Food Chem 106(1):
spectively. In Fig. 3d, e, f, all inhibition kinetics curves 247–252. doi:10.1016/j.foodchem.2007.05.077
crossed x axis, so the inhibition type of TFs, myricetin and 4. Deng YT, Lin-Shiau SY, Shyur LF, Lin JK (2015) Pu-erh tea poly-
quercetin for α-glucosidase was non-competitive. From the saccharides decrease blood sugar by inhibition of alpha-glucosidase
activity in vitro and in mice. Food Funct 6(5):1539–1546.
Michaelis-Menten equation in Table 3, the value of the trans-
doi:10.1039/c4fo01025f
verse axis intercept (Km) remained unchanged with all three 5. Kim J-S, Hyun TK, Kim M-J (2011) The inhibitory effects of eth-
compounds, indicating that TFs, myricetin and quercetin were anol extracts from sorghum, foxtail millet and proso millet on α-
non-competitive inhibitor for α-glucosidase. And the value of glucosidase and α-amylase activities. Food Chem 124(4):1647–
the inhibition constant (Ki) is 0.22, 0.32 and 50 μg ml−1, re- 1651. doi:10.1016/j.foodchem.2010.08.020
6. Lin CY, Yin MC (2012) Renal protective effects of extracts from
spectively. Because there are combined actions between in- guava fruit (Psidium guajava L.) in diabetic mice. Plant Foods Hum
hibitor and enzyme as well as between inhibitor and enzyme- Nutr 67(3):303–308. doi:10.1007/s11130-012-0294-0
substrate complex in the non-competitive enzyme reaction, 7. Dalar A, Turker M, Zabaras D, Konczak I (2014) Phenolic compo-
sition, antioxidant and enzyme inhibitory activities of Eryngium
the inhibition effect of non-competitive mode was stronger
bornmuelleri leaf. Plant Foods Hum Nutr 69(1):30–36.
than competitive mode. This explains why Hovenia flavo- doi:10.1007/s11130-013-0393-6
noids have a lower IC50 for α-glucosidase. 8. Wang M, Jiang C, Ma L, Zhang Z, Cao L, Liu J, Zeng X (2013)
In conclusion, Hovenia flavonoids offer a prospective al- Preparation, preliminary characterization and immunostimulatory
activity of polysaccharide fractions from the peduncles of
ternative for the management of postprandial hyperglycemia.
Hovenia dulcis. Food Chem 138(1):41–47. doi:10.1016/j.
Myricetin exerted promising inhibition effect on two major foodchem.2012.09.098
hydrolytic enzymes, α-amylase and α-glucosidase, and it ex- 9. Medicine NUoC (2006) Dictionary of Chinese herb medicine.
hibits similar reversible inhibition mechanism to that of Shanghai Scientific and Technologic Press, Shanghai
10. Park J-Y, Moon J-Y, Park S-D, Park W-H, Kim H, Kim J-E (2016)
acarbose. Furthermore, myricetin is more effective than
Fruits extracts of Hovenia dulcis Thunb. Suppresses
arcobase due to non-competitive inhibition on α-glucosidase. lipopolysaccharide-stimulated inflammatory responses through nu-
This study indicates that H. dulcis Thunb. may be used as a clear factor-kappaB pathway in raw 264.7 cells. Asian Pac J Trop
functional food with the purpose of T2DM alleviation as well Med 9(4):357–365. doi:10.1016/j.apjtm.2016.03.017
as to be utilized by the nutraceutical industry as a source of 11. Guo J, Meng Y, Zhao Y, Hu Y, Ren D, Yang X (2015) Myricetin
derived from Hovenia dulcis Thunb. Ameliorates vascular endothe-
myricetin. lial dysfunction and liver injury in high choline-fed mice. Food
Funct 6(5):1620–1634. doi:10.1039/c4fo01073f
Acknowledgments This research was supported by the overall plan- 12. Xie JH, Dong CJ, Nie SP, Li F, Wang ZJ, Shen MY, Xie MY (2015)
ning of Shaanxi science and technology (Project No. 2015KTCQ02-07), Extraction, chemical composition and antioxidant activity of flavo-
Doctoral Scientific Fund Project of the Ministry of Education of China noids from Cyclocarya paliurus (Batal.) Iljinskaja leaves. Food
(20130202120007), National Key Technology R&D Program Chem 186:97–105. doi:10.1016/j.foodchem.2014.06.106
(2015BAD16B02) and an earmarked fund for the China Agriculture 13. Park JS, Kim IS, Shaheed Ur R, Na CS, Yoo HH (2016) HPLC
Research System (CARS-28). Determination of Bioactive Flavonoids in Hovenia dulcis Fruit
Extracts. J Chromatogr Sci 54(2):130–135. doi:10.1093
/chromsci/bmv114
Compliance with Ethical Standards 14. Oboh G, Ademosun AO, Ayeni PO, Omojokun OS, Bello F (2014)
Comparative effect of quercetin and rutin on α-amylase, α-gluco-
Conflict of Interest All authors declare that they have no conflict of sidase, and some pro-oxidant-induced lipid peroxidation in rat pan-
interest. creas. Comp Clin Pathol 24(5):1103–1110. doi:10.1007/s00580-
014-2040-5
15. Peng X, Zhang G, Zeng L (2016) Inhibition of alpha-glucosidase by
vitamin D3 and the effect of vitamins B1 and B2. Food Funct 7(2):
References 982–991. doi:10.1039/c5fo00992h
16. Muller C, Dimitrov Y, Imhoff O, Richter S, Ott J, Krummel T,
Bazin-Kara D, Chantrel F, Hannedouche T, Cerrene, group Cs
1. Wilke T, Boettger B, Berg B, Groth A, Mueller S, Botteman M, Yu (2016) Oral antidiabetics use among diabetic type 2 patients with
S, Fuchs A, Maywald U (2015) Epidemiology of urinary tract in- chronic kidney disease. Do nephrologists take account of recom-
fections in type 2 diabetes mellitus patients: An analysis based on a mendations? J Diabetes Complicat 30(4):675–680. doi:10.1016/j.
large sample of 456,586 German T2DM patients. J Diabetes jdiacomp.2016.01.016
Complicat 29(8):1015–1023. doi:10.1016/j.jdiacomp.2015.08.021 17. Kwon Y, Vattem D, Shetty K (2006) Evaluation of clonal herbs of
2. Hernandez-Saavedra D, Mendoza-Sanchez M, Hernandez-Montiel Lamiaceae species for management of diabetes and hypertension.
HL, Guzman-Maldonado HS, Loarca-Pina GF, Salgado LM, Asia Pac J Clin Nutr 15(1):107–118
Reynoso-Camacho R (2013) Cooked common beans (Phaseolus 18. Nyambe-Silavwe H, Villa-Rodriguez JA, Ifie I, Holmes M, Aydin
vulgaris) protect against beta-cell damage in streptozotocin- E, Jensen JM, Williamson G (2015) Inhibition of human α-amylase
induced diabetic rats. Plant Foods Hum Nutr 68(2):207–212. by dietary polyphenols. J Funct Foods 19:723–732. doi:10.1016/j.
doi:10.1007/s11130-013-0353-1 jff.2015.10.003

You might also like