Fructose-Fed Induced Metabolic Syndrome Model in Cynomolgus Monkeys

The incidences of obese and metabolic syndrome have risen dramatically in recent decades. The pathological conditions of metabolic syndrome greatly increase the risk factors of cardiovascular disease, type 2 diabetes and NASH. The over consumption of high-fructose in beverage has been considered may play a role in the causes. To better understand the association of excessive fructose intake and pathophysiological progressing for obesity, type 2 diabetes and NASH, we assessed anthropometrics and metabolic parameters in a total of 71 cynomolgus monkeys after the consumption of high fructose diet within a 12 months period, including but not limited to body weight, body fat, HbA1C, liver enzymes. All the assessment values showed as Mean ± SD, and have compared from baseline of pre-feeding of high fructose diet to the 12 months of post feeding. The result showed that about 13% monkeys were overt T2DM, 32% monkeys became obese with an impaired IVGTT profile and 31% monkeys obesity animals at last, many features of the metabolic syndrome such glucose and insulin impairment were existed in related cohorts. We demonstrated that dietary of high fructose induced metabolic syndrome in cynomolgus monkeys that provides a good translational animal model system to investigate obesity and metabolic syndrome pathogenesis, prevention and treatment.


Introduction
Obesity, a complex health issues, is epidemic continually rising in recent decades with an estimated figure of 14 billion worldwide [1]. It has been a predominantly healthy problem in westernized societies [2][3][4], for example in the US, 31% of the population is obese at 2003 [5]. In addition, many Asian, South American and African nations have also reported that obesity rates are increasing rapidly during the last decade [2,[6][7][8][9]. Obesity develops because of a mismatch between energy intake and expenditure that results from behavior (feeding behavior and time spent active) and physiology (resting metabolism and expenditure when active). Both of these traits are affected by environmental and genetic factors [10].
The term "metabolic syndrome" dates back to at least the late 1950s, but came into common usage in the late 1970s [11,12]. It can be initialed by obesity and developed with chronic disorders of insulin resistance, hypertension, impaired glucose tolerance, hyperinsulinemia and dyslipidemia characterized by elevated triglyceride and low HDL concentrations [13]. In 2000, approximately 32% of U.S. adults had the metabolic syndrome. In more recent years that figure has climbed to 34% [14][15][16].
Obesity and metabolic syndrome, both chronic illnesses lead to a consequent of several health problems, such as diabetes, dislipidemia and cardiovascular diseases, and the metabolic syndrome has been associated with a plethora of cancers including breast, pancreatic, colon and liver cancer [17][18][19].
Take no account of genetic factors, the dramatic increase in the numbers of obese and metabolic syndrome in Western societies reflects mostly changing environmental factors and is linked to reduced activity and perhaps also increased food intake.
Excessive caloric intake has been related to high-fat foods and diets high both in simple sugars such as sucrose and in high-fructose corn syrup (HFCS) as a source of fructose [20][21][22]. HFCS is now manufactured and used in many countries throughout the world [23], HFCS now represent 40% of all added caloric sweeteners [24], and it is the major source of caloric sweeteners in soft drinks and many other sweetened beverages and is also included in baked goods, canned fruits, jams and jellies, and dairy products in the United States [25]. Furthermore, the rise in the consumption of high-fructose corn syrup in beverages has paralleled the rise in the prevalence of obesity and the metabolic syndrome [26][27][28]. Meta-analyses have also suggested that the consumption of sugar sweetened beverages is related to the risk of diabetes, the metabolic syndrome, and cardiovascular disease [29].
Besides of the epidemiologic investigations, many studies of the etiology, patho-physiological progress and treatment to obesity and metabolic syndrome have been undertaken. Recently several articles report that the consumption of fructose sweetened beverages increases visceral adipose deposition and triglycerides levels [30,31], produces dyslipidemia, and decreases glucose tolerance/insulin sensitivity [32]. HFCS made by enzymatic isomerization of glucose to fructose was introduced as HFCS-42 (42% fructose) and HFCS-55 (55% fructose) in 1967 and 1977, respectively. It is one of the major fructose resources, and about two-thirds of all HFCS consumed in the United States are in beverages [26]. Unlike pathway of glucose metabolism, fructose is only metabolized by liver. The effects of high fructose diet to the liver has been also investigated, and reported that hepatic injury has induced by dietary fructose in a study with calorically controlled primates [33].
Animal models were useful tool to explore the inherent and/or extrinsic inducement of the disease, to understand disease development, progression and related complication, to evaluate the target drug efficacy and so on. Most of metabolic syndrome models use rodent because of their small size, short generation interval, easy availability and economic considerations, they can be obtained either spontaneously or induced Journal of V Journal of Veterinary Science & eterinary Science & T Technology echnology seasonal fruits and vegetables were provided daily in accordance with the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) regulations and guidelines [40].

Anthropometrics assessment
Body weight, body fat percentage, waistline and Body Mass Index (BMI): Body weight, body fat percentage, waistline which circled navel and body mass index (BMI, body weight/crown-rump length 2, kg/m 2 ) were assessed at the time of pre diet-induction as baseline, then every three months after diet induction. Total body fat composition was determined by using dual-energy x-ray absorptiometry (DEXA, HOLOGIC Discovery), a sensitive technique for measuring body fat, which has been previously validated for the determination of body composition in monkeys [41]. The percentage level (%) total mass of fat divided by total body mass (see below formula).

%Fat=100 × fat mass/(fat mass+lean mass)
Clinical pathology panel: Blood samples were collected as the same pattern described above for anthropometrics assessments. The whole blood sample (K 2 -EDTA anticoagulation) was collected for analysis of hematology and plasma collection. The hematological assessments were analyzed by using an automatic analyzer (ADVIA 2120, Siemens).
The plasma sample have obtained by centrifugation of whole blood sample with K 2 -EDTA anticoagulation, and stored with an aliquot of whole blood sample at -70°C freezer until future analyze of glycemic indices.
Serum sample was also collected at same time and used for measuring clinical chemistry panel by using an automatic analyzer (HITACHI 7180, Hitachi High-Tech Science Systems Corporation). Clinical chemistry is including but not limit total cholesterol, triglyceride, non-esterified fatty acid (NEFA), liver and renal function parameter.

Intravenous Glucose Tolerance Tests (IVGTTs):
IVGTTs were performed quarterly. The animals were fasted for overnight 16 hrs and anesthetized with Zoletil 50 (Virbac S.A.) at about 3 mg/kg intramuscularly (i.m.) and then supplemental Zoletil 50 with half dosage of initial dose level was given intramuscularly (i.m.) if needed. The cephalic and/or saphenous veins were cannulated separately for glucose infusion and blood collection. Two baseline blood samples were collected into BD tube contain K 2 -EDTA at -5, and 0 minutes. Subsequently, 300 mg/kg of 50% dextrose was intravenously infused within a 30 second period. Additional blood samples were collected at 3, 5, 10, 20, 30, and 60 minutes after glucose infusion. Blood samples were gently inverted several times and immediately placed on wet ice prior to centrifugation at 2-8°C and 3000 g for 10 minutes to obtain plasma. Plasmas were stored at -70°C or lower until analysis.

Stratification of study cohorts:
The animals have been stratified into the study cohorts of lean, obese, insulin resistance and T2DM by chemicals or dietary or surgical manipulations or transgenic/knockout and/or by combination thereof. However, nonrodent models are urgently needed as a valuable supplement to rodents which have better patho-physiological metabolism similarities to humans. Nonhuman primates have been increasingly used in biomedical research as they are more anatomically and physiologically homologous to humans as compared to other animal models (e.g., rats, pigs) and share many other characteristics with humans. They are opportunistic omnivores which is same as human. Currently, studies related to obesity/diabetes research are moving toward using non-human primates (NHPs) [34][35][36].
Cynomolgus monkeys belong to old world monkeys, are sexually dimorphic with respect to size. Adult males are considerably larger than females, weighing 5-9 kg (11-20 lb) compared to the 3-6 kg (6.6-13.2 lb) of females [37]. The typical life span of the cynomolgus monkey ranges from 25-30 years [38], they are susceptible to age-related pathologies commonly observed in humans, such as obesity, and diabetes (and complications therefrom, including diabetic neuropathy and retinopathy) [39]. Along with the genome has been sequenced, making it's possible to compare gene code with human. In a word, the cynomolgus monkey has the third-largest range of any primate species, behind only humans and rhesus macaques, is an ideal species using in metabolic syndrome animal model.
In this article, the anthropometrics, and clinical pathological parameters, as well as the glycemic and lipidemia profiles data have been collected and presented from a total of 71 cynomolgus monkeys in a longitudinal study of experiment tried to mimic the etiology of human metabolic syndrome, to explore the link of high fructose diet with metabolic syndrome in large animal. Assessments have analyzed from a baseline of pre-feeding, and 3, 6, 9 to 12 months of post-feeding high-fructose diet, and compared to baseline from each post-feeding assessment.

Animal and husbandry information
The 61 males and 10 females of adults cynomolgus monkeys were obtained from JinGang Inc., Hainan providence, China and used in this study. They aged from 11-15 years and were kept in door cages, in paired housing at the vivarium facility of and accessibility of cage have accredited by AAALAC. Environment enrichment including mirror and different functional toys are provided for each animal. The room(s) were controlled and monitored for relative humidity (targeted mean range 40% to 70%) and temperature (targeted mean range 18 to 26°C) with 10 to 20 air changes/hour. The rooms were on a 12 hours light/dark cycle except when interruptions are necessitated by study activities. Animal using protocols for all the animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) (WuXi AppTec, com., Suzhou, Jiangsu province, The People's Republic of China).

Diet information
A commercial monkey chow diet (Lab Diets 5C48 is equivalent to LabDiets certified Primate Diet 5048, modified for import into China, Advance Protocol Old World Primate; PMI, St. Louis, MO, USA) was provided ad libitum to all the monkeys. This diet is a complete life cycle diet that provides 30% energy as protein, 12% energy as fat (ether extract), and 57% energy as carbohydrate. In addition, a 500 ml of fruitflavored (Archer Daniels Midland Company) 15% fructose sweetened beverage (75 g of fructose) were also provided ad libitum. Reverses osmosis water was available to all animals, ad libitum. Enriched with based on their initial assessments of anthropometrics, clinical path parameters and glycemic indices prior to high-fructose induction. The stratification of each cohort was described as Table 1.
The classified values of hemoglobin A1c percentage and body weight for stratification were referring to glycemic indices in cynomolgus described in chapter 14, nonhuman primates in biomedical research with the modification, volume 2 by Jan Wagner [42], and adapted from Cawthon Lang [37].

Statistical analysis:
Values are presented as the mean ± SD. A two-tailed p value <0.05 was considered to be statistically significant. The AUC for glucose and insulin, the clearance of glucose during the IVGTTs was calculated by using the linear-log trapezoidal rule [43] Phoenix WinNonlin software (version 6.2.1, Pharsight, Mountain View, CA).

Results
As Table 1 shown, there were 54 lean monkeys, 12 obese monkeys, 5 pre-diabetes and none of diabetic monkeys at pre-feeding of high fructose diet baseline time, after 12 months of high fructose consumption, 17 monkeys have remained in lean cohort, the monkey numbers in obese cohort has increased to 22 from 12. 23 pre-diabetes animals and 9 T2DM animals were overt. In other word, about 13% of the study cohort animals were developed T2DM, and 23 animals in pre-diabetes status which is probably in the process of going into T2DM cohort. In a nutshell, almost all animals developed components of the metabolic syndrome. The detail information is listed in Table 2.

Body weight, body fat, waistline and Body Mass Index (BMI)
All the monkeys initially gained weight significantly, from 8.07 ± 1.05 kg to 9.16 ± 2.23 kg, on the high-fructose diet in the first 3 months, and then the mean body weight slightly increased and get peak in 6 months (9.27 ± 2.29 kg), subsequently, the body weight were almost unchanged in the following months, but on the whole, the body weight increased by 13% at 12 months (p<0.001). Animal body weight profile in 12-month is shown in Figure 1a.
The initial body fat percentage (%) was 28.5 ± 8.21%, and then increased to 36.7 ± 8.82%, 38.3 ± 9.52% at 3 months and 6 months respectively, but in 12 months, the body fat percentage was slightly decreased to 37.6 ± 9.11% (Figure 1b). The waistline and BMI were 40.23 ± 6.89 cm and 38.62 ± 7.24 kg/m 2 at the beginning, those two parameters were also increased significantly in the first few months, which were reached the peak 48.20 ± 8.82 cm and 44.43 ± 8.64 kg/ m 2 in 6 months respectively. Whereafter, both two parameters were decreased a bit at the end of the measurement (Figure 1c and 1d). Overall, the profile of body weight, body fat percentage, waistline and BMI have similar curve, the increased percentage at 12 months of body fat percentage, waistline and BMI were 32% (p<0.001), 14% (p<0.001), 13% (p<0.001).
In the last measurement at 12 months, the body weights were highly correlated with their BMIs (r 2 =0.89, Figure 2c), and waistline (r 2 =0.54, Figure 2b). The body fat percentage and body weights were also correlated well (r 2 =0.21, Figure 2a), but not well with the former two values. And the correlation between body weights and HbA1c was the least one among the four analytical items (r 2 =0.01, Figure 2d).
As for kidney functional parameters, the mean urea concentration increased (+24%, P<0.001) at 12 months. Nevertheless, the creatinine concentration mainly decreased (-6%, P<0.001, Table 4). The hematology parameters did not show any obvious changing during the study (data not show).

Cohort
Hemoglobin    In order to make longitudinal comparison of metabolic relative parameter in different cohorts during the study period, the 71 animals at baseline and 6 months were also divided into 4 cohorts, by working backward from the data on 12 months based upon our hierarchical criteria. The mean fasting glucose concentrations were increased significantly (105.1 ± 53.6 mg/dL at 12 months versus 75.3 ± 11.4 mg/ dL at baseline). The Glucose AUC 0-60 min in all cohorts increased more or less, and the amount of increase in T2DM cohort was highest. In addition, the glucose clearance gradually decreased in all cohorts, and the decreasing ratio was also highest in diabetes cohort. Except lean cohort, the mean HbA1c had different increased ratios, without doubt increased most significantly in T2DM cohort. There is no obvious changing in mean fasting insulin concentrations in this comparison at different cohorts, but insulin AUC 0-60 min increased in all cohort except T2DM cohort ( Table 6). Figure 5. We can find that the glucose and insulin concentration curves did not change in lean cohort (Figures 4a and 5a), and the glucose concentrations increased in other three study cohorts, particularly in T2DM cohort (Figure 4b-4d). In obesity and pre-diabetes cohorts, the insulin concentrations both increased, especially in pre-diabetes cohort at 12 months, the insulin concentration has a significantly elevation, which match the insulin resistant characteristic (Figure 5b and 5c). As for diabetes cohort (Figure 5d), the insulin concentration decreased, indicated that the pancreas islet would not secrete enough insulin to help regulating the glucose metabolism, so the glucose concentrations did not return to normal at 60 minutes during IVGTT procedure.

The glucose and insulin concentration profiles in different cohort shown in Figures 4 and 5 during 1 year induction. And the insulin concentration profiles presents in
The horizontal comparison of glucose and insulin concentration profiles in IVGTT at 12 months presents in Figure 6, beyond doubt, the level of glucose concentrations were highest in T2DM cohort, however, the insulin concentrations were only higher than lean cohort. Both glucose and insulin concentrations were lowest in lean cohort among four study cohorts. In pre-diabetes cohort, the glucose concentration is slightly higher than obesity cohort, for insulin concentrations, mean concentrations at several time points were higher than obesity cohort, on the contrary to most other time points. Fructosamine concentrations in pre-diabetes were higher than lean and obesity cohorts, but lower than T2DM cohort (Table 6).

Discussion
In this investigation, only 10 females of 71 animals were used, both sexes can develop age-related metabolic syndrome diseases. Females have their highest birth rates around 10 years of age and completely stop bearing young by age 24 [44], all used female animals were age 11-15 years which were in that range. But 10 animals were not enough to make statistics analysis in this long-term study, based on the consideration, the data of 71 animals was analyzed as a whole.
Beverages sweetened with HFCS may have a link with the epidemic of obesity [45][46][47], HFCS provide more energy intake that lead to weight gain, and also obesity rate rising. According to the laws of thermodynamics, which state that energy can neither be created nor destroyed, the imbalance between intake and expenditure requires that we also have the capacity to temporarily store energy, and fat is preferred to be energy storage, since it is much denser than carbohydrate and also does not require large amounts of water for storage, therefore either food intake being too high, expenditure being too low (through low resting metabolic rate and/or activity expenditures), or a combination of both [10]. In our experiment, animals were housed in-door that lead to less activity and provided fructose-sweetened beverages, all those mimic human behavior that changing in eating habits and activity patterns. The number of obese and metabolic syndrome monkeys was changed gradually, the obese, pre-diabetes and T2DM monkeys were up to 30, 16, and 3 respectively at 6 months, and at the end of the study, there was only remaining 17 lean monkeys, it's no weird that the obesity animals decreased to 22, because pre-diabetes and diabetes monkeys increased to 23 and 9 (Table 2), the disease is in progressing. We can also found that mean body weight, body fat percentage, waistline and BMI changed significantly in the first 3 months, and then almost reach a plateau, however all four parameters increased significantly at 12 months versus baseline. In one study, rhesus monkey body weights were detected monthly, the changing profile is similar with our study [48].
For decades, BMI is as an indirect measure to define obese, WHO accepted that people with a BMI index >25 but <30 are said to be overweight, and people with an index >30 are defined as obese, but there are several well-recognized problems with this index. Firstly, it does not reflect body fatness changes very well when a person is also changing his or her height over time, it's not suitable for adolescent. Secondly, body builders and some athletes, who have developed large amounts of muscle tissue, may also be misclassified as obese, but the obese people carry around excessive amounts of body fat [10]. If we apply BMI to animal, there is another challenge that the monkey is not standing animal, using four limbs to climb, that make it difficult to define the height of the animal. As technological development, body fat percentage level, which is the direct indicator for obese, can be measured by DEXA, it is seemed to be better index of obesity in this primate species when compared with other obesity markers such as body weight and morphometry [49]. In one study, it suggested that normal female animals with %Fat of 25-35, boundary female animals had %Fat had of 35 -40, and obese female animals' %Fat was over 40 [50]. According to the criterion of American Council on Exercise [51], the percentage body fat of woman is higher than man in all groups, and woman with 40% plus and man with 30% plus body fat percentage are considerate as obese. So we assumed that body fat percentage which was over 40% in female monkeys and 35% in male monkeys would be classified as obese in this study.
Hemoglobin A1c a form of hemoglobin that is measured primarily   to identify the three-month average plasma glucose concentration. The recommendations of the American Diabetes Association (ADA), International Diabetes Federation, and World Health Organization [52][53][54], diagnosis of diabetes mellitus should be based on either an oral glucose tolerance test (OGTT) or hemoglobin A1c (HbA1c) findings, and HbA1c ≥ 6.5% (48 mmol/mol) will be diagnosed to diabetes. The HbA1c has several advantages compared with the fasting plasma glucose and OGTT, including greater convenience (fasting not required), greater preanalytical stability, and less day-to-day perturbations during stress and illness [53]. There is a positive correlation between glucose and HbA1c, when blood glucose levels are high, more glucose binds to hemoglobin in the red blood cells and the higher the glycated hemoglobin. Experimental evidence clearly shows that normal fasting serum glucose concentrations for monkeys are about 30 mg/dL lower than for normal humans [34,55], by that analogy, HbA1c is also lower in non-human primate, we conceived that HbA1c ≥ 5.5 is T2DM, less than or equal to 5.5 and great than 4.5 is pre-diabetes. But with caution to use HbA1c as cut point. On the one hand, it is important to take age and race/ethnicity into consideration, for example it's not suitable for using in children and adolescents [56,57]. On the other hand, several diseases or condition can influence the HbA1c value, lower HbA1c values are found in conditions such as recent transfusion and increased erythropoiesis secondary to hemolysis or blood loss, chronic kidney disease (CKD), anemia [58][59][60], to the contrary, asplenia and Iron deficiency anemia will lead a higher HbA1c values [61,62]. We also take above concerns into account, the hematology results were in normal range in adult monkeys [63]. Several articles report that triglycerides level were increased as the consequences of ingesting sugar-sweetened beverage [30][31][32]61], in our study, triglycerides level increased significantly after 1 year of beverage intake. Studies indicate that fructose consumption and have been associated with fatty liver [64,65] and hepatic fibrosis [66] in people, higher exposures to dietary fructose result in the development of hepatic lipidosis in nonhuman primates when consumed ad libitum for periods equivalent to ≥ 1 human year and ALT, AST, GGT increased after 6 weeks of high fructose diet exposure [67], the detected biomarkers (NEFA, TCHO, ALT, AST, GGT in Tables 3 and  4) changing trend in our study also reveal this possibility.
In a manner analogous to the determination of glycated hemoglobin, fructosamine testing determines the fraction of total serum proteins that have undergone glycation (the glycated serum proteins). Except glycated lipoprotein and glycated globulin, the main component of fructosamine is glycated albumin, albumin is the most abundant protein in blood, fructosamine levels typically reflect albumin glycation, albumin has a half-life of approximately 20 days, the plasma fructosamine concentration reflects relatively recent (2-3 week) changes in blood glucose [68]. More and more studies are being conducted fructosamine as a glycemic control indicators [69][70][71], the data showed that fructosamine concentrations (Table 6) in T2DM cohorts were higher than any other cohorts, and in pre-diabetes cohort, the level was far less than T2DM cohort but slightly higher than lean and obese cohorts. Fructosamine is possible to evaluate shortterm glycemic control, but we need to know that the diseases such hyperthyroidism and hypothyroidism [71,72] and nephritic syndrome [73] can lead to inaccurate result, because of the impact of protein (albumin) metabolism.
The association of obesity with type 2 diabetes is the ability of obesity to engender insulin resistance, insulin resistance plays an essential role in the development of the metabolic syndrome and T2DM [19,74]. The mechanisms of obesity-associated insulin resistance may include endocrine (fatty acids, adipokines and other adipocyte factors), inflammatory, neuronal pathways, and cell-intrinsic mechanisms (oxidative stress, ectopic fat storage, mitochondrial dysfunction and ER stress) [75]. Factors thought to contribute to insulin resistance include diet, exercise, smoking, stress, gene and age [76]. Insulin resistance usually connotes resistance to the effects of insulin on glucose uptake, metabolism, or storage. Insulin resistance in obesity and type 2 diabetes is manifested by decreased insulin-stimulated glucose transport and metabolism in adipocytes and skeletal muscle and by impaired suppression of hepatic glucose output [77][78][79]. Intravenous glucose tolerance test (IVGTT) is one of the best established methods     of measuring insulin resistance [80], this test is widely used in monkeys [81] for the decades. In this article, we performed IVGTT to evaluate the glucose tolerance and insulin sensitivity and beta cell function. The data show that glucose tolerance and insulin tolerance were impaired in the pre-diabetes and T2DM cohorts ( Table 6; Figures 4-6) at 6 months and 12 months respectively, the impairment degree was exacerbated as time went by. The inability of the β-cells to produce sufficient insulin in a condition of hyperglycemia is what characterizes the transition from insulin resistance to T2DM [19], the study present that hyperglycemia and hypoinsulimia are as characters in T2DM cohort, while in prediabetes cohort, hyperglycemia and hyperinsulimia are observed in the monkeys.

Conclusion
More and more evidence show that there is a link increasing high-fructose consumption result in obese and metabolic syndrome in human. We demonstrated that, consumption of a high-fructose diet in cynomolgus monkeys during 1 year period produces a series of diseases that have similar diseases symptoms of human, who show a metabolic progression from obese, insulin resistance and impaired glucose tolerance to T2DM, making the NHP mode is valuable for studying obesity and metabolic syndrome pathogenesis, prevention and treatment. The rapid metabolic changes occur also provide the possibility to evaluate these processes in long-term compliance with dietary and pharmaceutical interventions studies that cannot practically be performed in human.