Acute Effects of Breakfast Fruits Meal Sequence and Postprandial Exercise on the Blood Glucose Level and DPP4 Activity among Type 2 Diabetes Mellitus Patients: A Pilot Study

Objectives Type 2 diabetes mellitus (T2DM) is a major global public health issue. Diet and physical exercise are modifiable factors that influence the glycaemic status of patients with T2DM. We aimed to investigate the acute effects of breakfast fruits meal sequence and postprandial exercise on the blood glucose level and dipeptidyl peptidase 4 (DPP4) activity among type 2 diabetes mellitus patients. Methods A randomized pilot study recruited patients with T2DM who attended two primary health care centres in Tasikmadu District, Karanganyar Regency, and Kartasura District, Sukoharjo Regency, Central Java, Indonesia, from July to October 2016. Eligible patients (4 men and 32 women) were randomly divided into four treatment groups. Venous blood samples were analyzed for fasting and one-hour postprandial blood glucose (FBG and 1 h PPG) levels and DPP4 activity. Blood glucose levels were measured using a routine hexokinase method, and serum DPP4 activity was determined spectrophotometrically after incubation with the Gly-Pro-p-nitroanilide substrate. Results Fruits last meal decreased FBG level whilst fruits first meal did not significantly decrease 1 h PPG level. Both treatments had no acute effects on DPP4 activity but the addition of postprandial exercise helped lower DPP4 activity. Fruit last and first meals showed significant opposite effects on mean changes of FBG level (p < 0.05). Conclusions This preliminary report of fruits meal sequence is potentially involved in acute regulation of blood glucose levels and that it might be independent of DPP4 activity in Indonesian patients with T2DM. Moreover, postprandial exercise may be an important intervention for T2DM through the mediation of DPP4 but has no acute effects on the regulation of blood glucose levels. Further studies are required to investigate whether or not different types of fruits and longer treatment intervals can affect blood glucose levels and DPP4 activity differently. This study also gives an insight into the feasibility of conducting food order modification with or without the combination of postprandial exercise in a primary health setting for our next studies.


Introduction
Type 2 diabetes mellitus (T2DM) remains a major public health problem and is estimated to afict hundreds of millions of people worldwide. Te trend has changed dramatically in recent decades and the number of patients with T2DM will increase signifcantly, reaching 578 million by 2030 [1]. Glycaemic control plays an important role in the prevention of T2DM complications. Elevated FBG and 2 h postprandial glucose levels are correlated with increased diabetic vascular complications [2]. Increased 1 h PPG levels are also associated with increased diabetic complications and mortality [3].
Dietary modifcation and physical exercise have been recommended as the frst lines of T2DM management to minimise the risk of developing T2DM complications [4,5]. For example, increased vegetable consumption decreases diabetes complications, such as diabetic retinopathy [6,7]. Additionally, modifcation of meal sequence, such as vegetable consumption before carbohydrates caused a decrease in glycaemic parameters [8][9][10][11]. Meanwhile, fruit consumption is also predicted to improve glycaemic parameters in patients with T2DM because the fbre and antioxidant contents of fruits are similar to those of vegetables [12]. Unfortunately, the investigation of modifed meal sequences involving fruits is still limited.
In general, recent evidence has indicated that regular physical activity strongly correlates with a reduction in T2DM development [13]. Te American Diabetes Association recommends lifestyle changes, including physical exercise and activity, as an important part of glucose management for people with prediabetes and diabetes. Physical exercise improves insulin release and sensitivity in patients with T2DM [14] by upregulating the expression of glucose transporter type 4 (GLUT-4) [5].
General diet recommendations have not specifcally defned on when fruit should be eaten [15,16]. Additionally, previous studies of meal sequences on the impact of fruits on T2DM management are still limited. Tis indicates that meal sequences are generally neglected and focus only on the number of fruits that need to be consumed. Although the recommendation of physical exercise for T2DM is available [14], the study on the potential role of postprandial exercise which can be integrated into T2DM management, is also limited. Terefore, to fll this research gap, herein we aimed to compare diferent fruit meal sequences with or without a combination of postprandial exercise (PE) on blood glucose levels and DPP4 activity in Indonesian patients with T2DM.

Study Design, Area, and Period.
Tis is a follow-up from two previous reports of the same preliminary study [17,18]. We have previously reported that after 7 days of treatment, consuming fruit frst as the meal (FF) signifcantly reduced fasting DPP4 activity in patients with T2DM compared to the consumption of fruit as the last meal (FL) [17]. However, we found no signifcant decrease in 30 min postprandial DPP4 activity after 7 days of treatment. In another report, we found that 2 min of physical exercise (jumping jacks) signifcantly decreased blood glucose levels in patients with diabetes [18]. Previously, it was also known that the physical exercise of jumping jack has health impacts through modifcations of endothelial cell function and glucose control [19,20].
Terefore, the current report included all of those groups with the addition of a new group who consumed FF, followed by 2 min of PE. We described new data from FBG and DPP4 of 0 and 60 minutes. However, due to the overlapping with the previous pilot study, the characteristic data of participants are partly in accordance with (and with permission to) the previous reports [17,18].
In this present study, we conducted a randomized pilot study. Te fruit-treatment study was performed at two primary health care centres in Tasikmadu District, Karanganyar Regency and Kartasura District, Sukoharjo Regency, Central Java, Indonesia, from July to October 2016. During this period, we were able to recruit 37 research participants but one participant from the FF group was excluded due to signifcantly incomplete data.

Te Selection of Research Participants.
Te research participants were selected using the following criteria: diagnosed with T2DM, aged 45-65 years old, and able to ingest fresh fruits. We excluded pregnant women, patients who were taking DPP4 inhibitors (sitagliptin, vildagliptin, and similar drugs), and patients with chronic diseases such as heart, renal, and hepatic diseases.

Ethical Clearance.
Te study protocol was approved by the Research Ethical Committee, Faculty of Medicine, Universitas Sebelas Maret, Public Hospital Dr. Moewardi, Surakarta (No: 502/VI/HREC/2017). Informed consent was obtained from all selected research participants to draw peripheral venous blood and to recall their food consumption within 24 h using questionnaires. All data were kept confdential.

Protocol of the Study.
Before starting the study, all participants received healthy nutrition education for T2DM. Tey had to consume an equal proportion of breakfast for the 7 days of treatment. On day 1 and 7, we provided breakfast, the list of meals, and fruits for the participants, such as bananas, oranges, and papaya. On days 1 and 7, participants consumed 50 g bananas whereas from days 2-6 participants consumed papaya or orange, which were equivalent to 50 g banana. Research participants consumed the meals and fruits based on their study groups in front of investigators on days 1 and 7. Te composition of the meals consumed on day 2 to day 6 was equally restricted to a specifc composition that represented food treatments on day 1 and day 7 during breakfast and was closely monitored by the investigators every day. A postprandial exercise of jumping jacks was performed for 2 min after having breakfast.
Te food composition consisted of carbohydrates (mixed brown and white rice), protein and fat sources (tofu, egg, and meat), and mixed vegetables. Te proportion of each macronutrient was determined according to the classical diet composition for diabetic patients: 55% carbohydrates, 25% fat, and 20% protein. Vegetable consumption was based on American Diabetes Recommendation [21].
Te research participants were randomly divided into four groups: (i) consumed FL without 2 min postprandial exercise, (ii) consumed FF without 2 min postprandial exercise, (iii) consumed FL, followed by 2 min postprandial exercise (FL + PE), and (iv) consumed FF, followed by 2 min postprandial exercise (FF + PE), respectively. Te protocol of the study was retrospectively registered in the ISRCTN registry on 13 th May 2022 with the trial registration number 13920339 and is publicly available at https://doi.org/10.1186/ ISRCTN13920339.

Characteristics of Research Participants.
Basic characteristics of research participants were obtained from their medical records at the two primary health cares. Parts of the characteristics (FL and FF) are in accordance and with the permission of our previous studies [17,18]. Te glucose control level was assessed by 2 h postprandial glucose level which is the best indicator for glucose control in the absence of HbA1C [22].

Nutritional Intake.
To assess the efects of treatment, we measured the participants' daily intake of macro and micronutrients. Te data were acquired from every research participant with regard to quantity, size, food type, composition, and food processing using a 24 h food recall questionnaire on three diferent days and converted into daily intake values of macro-and micronutrients using the free NutriSurvey software (https://www.nutrisurvey.de/), which has been translated in Indonesian.

Blood Glucose Level.
Blood glucose levels were measured on day 1 and day 7 of the 7-day treatment. Venous blood samples were used to measure FBG and 1 h PPG levels, which were examined using a routine hexokinase method in the clinical laboratory of two primary health cares.

Statistical Analysis.
All numerical data are presented as mean ± standard deviation, and categorical data are presented as numbers and percentages. SPSS version 20.0, for Windows (SPSS, Inc., Chicago, IL, USA), was used to analyse the statistical signifcance of our data. Normality and homogeneity tests were performed before comparing the averages of blood glucose levels and DPP4 activity among the groups. A comparison of the basic characteristics of the research participants was performed using the chi-square test while the paired Student's t test was performed to analyse the time-dependent diference of the parametric data. Te Wilcoxon test was used to analyse non-parametric data. An independent Student's t test was used to analyse the timeindependent diference of the parametric data while the Mann-Whitney test was used for non-parametric data.

Characteristics of Research Participants.
From 36 selected research participants, Table 1 indicates their general characteristics, of which the FL and FF groups consisted of 10 and 8 T2DM patients, respectively. Te mean ages of the four groups were relatively similar but the FL group had the youngest mean age (53.50 ± 7.28 years), compared to the treatment groups (55.88 ± 4.19; 54.11 ± 6.07 and 55.56 ± 3.54 years). Females were more dominant than males in all four groups (>85%). Te mean duration of T2DM in FF (7.31 ± 3.43 years) and FF + PE (5.56 + 3.60 years) groups was longer than that of in FL (4.65 + 3.48 years) and FL + PE (3.39 + 3.71 years) groups but it was not signifcantly different. Based on anthropometric parameters, most research participants were overweight and obese (65% or more) except in the FF group (50% normal weight and 25% overweight or obese).

Nutritional Intakes of Research Participants.
We measured the nutritional intake of every research participant using a food recall approach. We observed that pretreatment nutritional intake varied among the groups. However, after meal sequence treatment during breakfast, along with a standardized diet in two other daily meals, equal nutritional intake was achieved after 7 days among the four groups, especially in dietary carbohydrate and fbre intake, which were our main interests.
Within 7 days of treatment ( Figure 1) there were no signifcant diferences in the nutritional intake of energy, carbohydrates, fbres, proteins, and lipids among groups. However, protein intake on day 1 was signifcantly diferent between the groups.  (Figure 2(a)).

Fruits Last Meal and Directly Followed by 2 min Postprandial Exercise Decreased 1h PPG Levels after 7 Days of
Treatment. In addition to lowering FBG levels, we found that FL + PE signifcantly lowered 1 h PPG levels after 7 days of treatment (Figure 2(b)). Meanwhile, no signifcant difference was observed between the FL and FL + PE groups in terms of 1 h PPG level (298.60 ± 109.53 vs. 287.89 ± 109.44; p � 0.952) after 7 days of treatment.    (Figure 2(c)).

Te Addition of Postprandial Exercise Lowered DPP4
Activity. FL + PE and FF + PE groups had signifcantly decreased 1 h DPP4 activity after 7 days of treatment (Figure 2(d)). We also observed a signifcant decrease in DPP4 activity in the FL + PE group but not in the FF + PE group. In contrast, fasting and 1 h DPP4 activity were relatively stable in both FL and FF groups.

Fruits First Meal Showed Signifcant Opposite Efects on FBG Levels Compared to Fruits Last Meal.
To further analyse the efects of each treatment in detail, we investigated the change of mean from each variable over 7 days of treatment (  to the FL group (24.80 ± 93.28 vs −68.26 ± 42.11; p � 0.022). However, we did not see the same decrease of 1 h PPG in all groups.

Discussion
Herein, this randomized pilot study showed the efects of a fruit-carbohydrate meal sequence combined with 2 min of postprandial exercise on blood glucose levels and DPP4 activity in patients with T2DM. Our study was conducted at Karanganyar and Sukoharjo Regencies, in Central Java, Indonesia, which is populated by approximately 800,000 residents. Karanganyar and Sukoharjo Regencies were located ca. 120 and 90 km, respectively, from Semarang, the largest capital city of Central Java. Tus, it refected rural population areas in the southeastern part of Central Java, and most residents work in the agriculture sector. However, the overweight to obesity class of BMI was predominant in our groups of research participants, suggesting that the profle of our rural T2DM research participants was similar to that of urban Indonesian patients with T2DM [24]. One week before starting the study, we assessed the daily nutritional intake of our research participants and our fndings indicated variation among the groups (Figure 1). When we started our breakfast treatment and educated them to follow T2DM standard diets in two other daily meals, we were able to achieve similar nutritional intakes after 7 days except for protein intake, which was similar to that on day 1. Terefore, we excluded the possibility that signifcant differences in nutritional intake among the groups infuenced our results.
Our data indicated that the FBG levels in patients with T2DM who consumed FL without postprandial exercise were signifcantly reduced compared to other groups. Furthermore, a signifcant reduction in the FBG level was only observed in the FL + PE group after 7 days of treatment but not in the FF + PE group. Terefore, this suggests that consumption of fruit, irrespective of meal sequence and additional postprandial exercise, independently had a delayed efect on FBG levels.
Te present research participants consumed bananas in their modifed diet, which had high-fbre content [25]. Recent studies have shown that consuming high-fbre vegetables before carbohydrates can delay gastric emptying [8][9][10][11] and therefore inhibit carbohydrate absorption. Subsequently, it has a critical impact on the prevention of T2DM progression [26,27].
When we investigated 1h PPG levels, only FL + PE showed a signifcant reduction after 7 days of treatment. However, no signifcant change in 1 h PPG after 7 days was observed in all groups. In addition, FL + PE showed a signifcant reduction in DPP4 activity compared to FL. Altogether, these fndings suggest that postprandial exercise without meal sequence modifcation regulates DPP4 activity without afecting the 1 h PPG level.
Te efect of physical exercise on the improvement of FBG levels has been reported before [28], in which physical exercise decreased FBG levels among participants with a higher baseline FBG level. Physical activity can control blood glucose levels by promoting glucose consumption in skeletal muscles [29] and glucose uptake by GLUT-4 in the muscle cell membrane [5]. A diferent research group found that physical exercise decreased DPP4 activity in patients with metabolic syndrome [30]. Based on our study, even though there was no additional reduction of DPP4 activity, FF + PE showed a higher reduction of both fasting and 1 h DPP4 activity compared to FL + PE, which indicates that the combination of fruit at last meal with physical exercise has a greater acute regulation on DPP4 activity.
Tis study gives a preliminary insight into the role of fruits meal sequences and postprandial physical exercise for T2DM management, which regulates an acute phase of glucose levels and is probably independent of DPP4 activity. From this pilot study, we will further investigate a randomized controlled trial (RCT) with a bigger sample and diferent choices of fruits. A separate RCT will be conducted to investigate the impact of postprandial physical exercise in T2DM patients.
Our study has some limitations due to sample size and basic characteristics, which were less refective of the rural population and patients with T2DM as well as overweight or obesity. Te 7-day period of our investigation was also relatively short. Terefore, we could not distinguish diferent responses that may have occurred to the same treatment of fruits order modifcation and physical exercise, based on ethnic groups or their food habits and conducting a research study in diferent areas in Indonesia.

Conclusions
Tis preliminary report of fruits meal sequence is potentially involved in acute regulation of blood glucose levels and it might be independent of DPP4 activity in Indonesian patients with T2DM. Moreover, postprandial exercise may be an important intervention for T2DM through the mediation of DPP4 but has no acute efects on the regulation of blood glucose levels. Further studies are required to investigate whether or not diferent types of fruits and longer treatment intervals can afect blood glucose levels and DPP4 activity diferently. Tis study also gives an insight into the feasibility of conducting food order modifcation with or without a combination of postprandial exercise in a primary health setting for our next studies.

Data Availability
All raw and analyzed data are kept by Yohanes Cakrapradipta Wibowo and we will provide data on request by emailing him at wibowo.yohanec@gmail.com.

Conflicts of Interest
Te authors that they have declare no conficts of interest.