Molecular changes in hepatic metabolism in ZDSD rats–A new polygenic rodent model of obesity, metabolic syndrome, and diabetes

Abstract

In recent years, the prevalence of obesity, metabolic syndrome and type 2 diabetes is increasing dramatically. They share pathophysiological mechanisms and often lead to cardiovascular diseases. The ZDSD rat was suggested as a new animal model to study diabetes and the metabolic syndrome. In the current study, we have further characterized metabolic and hepatic gene expression changes in ZDSD rats. Immuno-histochemical staining of insulin and glucagon on pancreas sections of ZDSD and control SD rats revealed that ZDSD rats have severe damage to their islet structures as early as 15 weeks of age. Animals were followed till they were 26 weeks old, where they exhibited obesity, hypertension, hyperglycemia, dyslipidemia, insulin resistance and diabetes. We found that gene expressions involved in glucose metabolism, lipid metabolism and amino acid metabolism were changed significantly in ZDSD rats. Elevated levels of ER stress markers correlated with the dysregulation of hepatic lipid metabolism in ZDSD rats. Key proteins participating in unfolded protein response pathways were also upregulated and likely contribute to the pathogenesis of dyslipidemia and insulin resistance. Based on its intact leptin system, its insulin deficiency, as well as its timeline of disease development without diet manipulation, this insulin resistant, dyslipidemic, hypertensive, and diabetic rat represents an additional, unique polygenic animal model that could be very useful to study human diabetes.

1. Introduction

Metabolic syndrome (MetS) is a clustering of risk factors [1,2] that lead to an increased risk for cardiovascular disease (CVD, ~2-fold) [3,4], and increased susceptibility of developing type 2 diabetes (T2D, ~5-fold) [3,5]. It is also a cause and a consequence of nonalcoholic fatty liver disease (nonalcoholic steatohepatitis, NASH) [6–9]. MetS has also been implicated in the pathogenesis of several other clinical conditions, such as mental health disorders, stroke, obstructive sleep apnea (OSA), polycystic ovary syndrome (PCOS) and certain types of cancer [9]. The syndrome is characterized by multiple derangements that include insulin resistance with or without glucose intolerance, hypertension, dyslipidemia (i.e., hypertriglyceridemia, diminished level of high-density lipoprotein (HDL) cholesterol, raised small low-density lipoprotein (LDL) particles), low-grade inflammation, and a pro-thrombotic state [3]. Globally, the MetS is everywhere, with prevalence ranging from < 10% to 45–55% [10–13]. Approximately 33.4% of the US population is estimated to have MetS [14]. Furthermore, estimates of MetS prevalence in the US population varies considerably with ethnicity/race, sex, age and comorbidities [14]. Of serious concern is the increasing prevalence of MetS in children and young adults [15–17]; its prevalence is estimated to be in the range of 0–19.2% with a median prevalence of 3.3% [16]. Despite the plethora of abnormalities and unmanageable rise in health care cost, to date no single drug has been approved for treating MetS [16–22].

Thus, there is a pressing need for developing new efficacious drugs with good tolerability and safety profiles for the management of the individual components of MetS, but progress in this direction has been hindered by lack of availability of suitable animal models mimicking the MetS disease state in humans and associated clinical and metabolic complications. Although several rat and mouse models have been developed and extensively studied as representative for exploring the mechanisms of obesity, MetS, and T2D in humans [23–30], including two frequently used rat models, Zucker fatty fa/fa (ZF) [31] and ZF-derived Zucker diabetic fatty fa/fa (ZDF) [31,32], unfortunately, none of the existing rodent models can faithfully mimic all the aspects of the pathophysiology of human MetS. As an improvement, the Zucker diabetic Sprague–Dawley (ZDSD) rat has been developed as one of the most translatable inbred polygenic models for obesity, MetS, and T2D [33]. The primary advantage of the ZDSD model is that it more closely resembles the human condition by having an intact leptin pathway. The ZDSD rat was developed by crossing the ZDF rat (Lean^+/+) with the CD (SD) rat and selectively breeding for obesity and diabetes traits followed by inbreeding for > 35 generations [34].

Previous studies show that ZDSD rats develop T2D in a manner that mimics the development in humans [33,34]; pre-diabetes (8–16 weeks of age), through overt diabetes (> 16 weeks of age), to diabetic complications (24+ weeks of age). Diabetic complications include nephropathy, neuropathy, cardiovascular disease, delayed wound healing, fatty liver etc. The ZDSD rat also more closely mirrors the human metabolic syndrome phenotype than conventional rodent models, including increased body weight with abdominal fat, insulin resistance, dyslipidemia and hypertension (https://www.crownbio.com/metabolic-disease/diabetes/zdsd-rat). The objective of the current study was to more rigorously characterize metabolic endpoints, islet histology, as well as molecular changes, in ZDSD rats compared to age matched SD rats. In the current study, chow-fed SD and ZDSD rats were followed for 16 weeks starting at age 10 weeks. We found that ZDSD rats exhibited obesity with increased body fat, hyperglycemia, dyslipidemia, insulin resistance, and glucose intolerance. As disease progressed, ZDSD rats demonstrated elevated glucose levels and reductions in body weight. At the molecular level, genes involved in glucose metabolism, lipid metabolism and amino acid metabolism were changed dramatically in ZDSD rats compared with SD rats.

2. Methods and materials

2.1. Total RNA isolation and real-time qRT-PCR

Total RNA was extracted from liver tissue samples (~20 mg) using TRIzol reagent (Life Technologies, Grand Island, NY) and was reverse-transcribed (2 μg total RNA) using SuperScript IV (Life Technologies, Grand Island, NY) per the manufacturer’s instructions. Real-time PCR was performed using a FastStart Universal SYBR Green Master PCR Kit (Roche Applied Science, Indianapolis, IN) and an ABI Prism 7700 system (Applied Biosystems® Life Technologies, Grand Island, NY) per the manufacturer’s protocols. The amplification process time was as follows: 95 °C for 2 min; 40 cycles at 95 °C for 20 s, 60 °C for 20 s, 72 °C for 30 s, and a final 5-min extension at 72 °C. Results were normalized to the housekeeping gene Rplp0. The 2^−ΔΔCt method was used to calculate relative mRNA expression levels. The oligonucleotide primers and their sequences used in qRT-PCR are shown in Supplemental Table S1.

2.2. Western blot analysis

Liver samples (~20 mg) were homogenized in 200 μl Pierce RIPA buffer (Thermo Scientific/Pierce Biotechnology, Rockford, IL) with 1× Halt™ Protease and Phosphatase Inhibitor Cocktail (Thermo Scientific/Pierce Biotechnology, Rockford, IL). The supernatant fractions were obtained by centrifugation and then analyzed for protein concentration by a BCA™ Protein Assay Kit (Thermo Scientific/Pierce Biotechnology, Rockford, IL). 50 μg of total protein was subjected to 4–10% SDS-PAGE under denaturing conditions and transferred to PVDF membranes. Membranes were blocked with 5% non-fat milk in 1× TBST (tris-buffered saline containing 0.1% Tween 20) for 1 h and then incubated with primary antibody for 16 h at 4 °C. After three washes with 1× TBST, the membranes were incubated with IRDye 800CW goat anti-rabbit or IRDye 680LT goat anti-mouse secondary antibodies (LI-COR Biosciences, Lincoln, NE) for 1 h at room temperature. Proteins were detected with the Odyssey Infrared Imaging System (LI-COR Biosciences, Lincoln, NE). The antibodies used in this study are as follows: FABP1 (Cell Signaling Technology, Danvers, MA, #13368), CHOP (Cell Signaling Technology, Danvers, MA, #2895S), P-IRE1α (Novus Biologicals, Littleton, CO, NB100–2323), IRE1α (Novus Biologicals, Littleton, CO, NB100–2324), β-Actin (Cell Signaling Technology, Danvers, MA, #4970), P-ACC (Cell Signaling Technology, Danvers, MA, #3661), ACC (Cell Signaling Technology, Danvers, MA, #3662), GSK3 and P-GSK3 (Cell Signaling Technology, Danvers, MA, #9369).

2.3. Chemicals and reagents

Glucose, triglycerides, and total cholesterol kits were purchased from Stanbio Laboratories (Boerne, TX). Free cholesterol and non-esterified fatty acids (NEFA) measurement kits were purchased from FUJIFILM Wako Diagnostic U.S.A. Corporation (Mountain View, CA). All other reagents used were of analytical grade.

2.4. Animal studies

Six-week old male ZDSD/Pco rats were obtained from PreClinOmics, now Crown Bioscience - Indiana (Indianapolis, IN). Age-matched Sprague Dawely (SD) rats were obtained from Harlan, now Envigo (Indianapolis, IN). All animal experiments were performed per the procedures approved by the VA Palo Alto Health Care System Institutional Animal Care and Use Committee. Animals were housed in laboratory cages at 23 °C under a 12-hour light-dark cycle. Rats were maintained on a diet supplemented with baytril for one month as part of the facility quarantine program. Rats were changed over to Purina diet 5008 (Lab Diet, St. Louis, MO), and maintained on this diet for the rest of the study. This diet is formulated with high quality ingredients to assure minimal inherent biological variation in long-term studies, with 26.8% calories provided by protein, 16.7% calories provided by fat and 56.4% calories provided by carbohydrate. For detailed composition and nutrition values, please see http://www.labsupplytx.com/wp-content/uploads/2012/10/5008.pdf. The rats were weighed every 2 weeks, had blood drawn under fasting conditions every 4 weeks, and diet consumption was measured at 14, 22, and 26 weeks of age.

2.5. Glucose Tolerance Test (GTT) and Insulin Tolerance Test (ITT)

GTT and ITT were performed 7 days apart during the 17th week of age. ITT was performed again at 23 weeks of age, and GTT was performed at 26 weeks of age. For GTT, animals were fasted overnight, and glucose (1 g/kg body weight) was intraperitoneally (IP) injected. Blood was collected from the tail vein prior to injection and at 15, 30, 60, and 120 min post-injection to assess glucose clearance. For ITT, animals underwent a 4 h fast. Insulin (0.75 U/kg) was IP injected and blood collected prior to injection and at 15-, 30-, 45- and 60-min post-injection to assess insulin sensitivity. Blood glucose levels were determined immediately using a glucometer (Bayer Contour).

2.6. Quantification of serum metabolites

Blood was collected from the tail vein after a 4 h fast at the final week of the experimental period. Samples were centrifuged at 4000 g for 15 min at 4 °C and serum retained and stored at −80 °C. Serum glucose, triglycerides, total cholesterol, free cholesterol, and NEFA levels were determined with commercial assays per manufacturers’ protocols.

2.7. Histologic analysis of pancreatic samples

Pancreatic samples from 15 weeks old and 26 weeks old ZDSD and SD control rats were fixed in 10% Formalin, processed and sectioned. Sections were stained with hematoxylin and eosin as well as with anti-glucagon (brown) and anti-insulin (red) antibodies. Quantification of glucagon and insulin staining was performed on three areas of sections of each animal, and five animals in each group were analyzed.

2.8. Measurement of liver triglyceride content

Suitable aliquots of liver homogenates were extracted with a mixture of chloroform and methanol according to the procedure [35] and subsequently lipid samples were quantified for their triglyceride content using an enzymatic colorimetric assay kit supplied by Stanbio Laboratory per manufacturer’s protocol.

2.9. Blood pressure measurements

Blood pressure was measured using IITC Life Science Mouse Rat Blood Pressure (MRBP) Machine when the rats were 24 weeks of age. One week prior to blood pressure measurements, rats were placed in restrainers daily to acclimate the animals to the testing environment and minimize movement during the actual readings. On the day of the experiment, the MRBP machine is set up per the manufacturer’s instructions, setting the ambient temperature to 32 degrees Celsius. Each rat was placed into a pre-warmed restrainer and tail cuff sensor and were allowed to acclimate inside the chamber for 5 to 10 min prior to reading. A live run was selected, during which blood pressure was measured 5 times, and the best 3 readings were averaged.

2.10. Statistics

Data are reported as mean ± standard error of the mean (SEM). Comparisons were made using multiple student t-tests and student t-tests with Welch’s Correction using GraphPad Prism 7.0 (La Jolla, CA). P values of < 0.05 were considered statistically significant.

3. Results

3.1. Increased fat, body weight, plasma triglycerides, free fatty acids as well as spontaneous diabetes and insulin resistance are observed in ZDSD rats

At 10 weeks of age, there was no difference between the SD and ZDSD rats in regard to body weight. At 12 weeks of age and on, the ZDSD rats were significantly heavier than the SD rats (Table 1). Over time, ZDSD rats start becoming spontaneously diabetic, at which time weight loss and increases in blood glucose are seen [27]. At 22 weeks of age, some ZDSD rats started losing weight. Diet consumption was also measured at 14, 22, and 26 weeks of age (Fig. 1). The ZDSD rats ate approximately 7 g more diet per day than the SD rats during weeks 14 and 22, and approximately 9 g more per day at weeks 26, presumably because of rats becoming spontaneously diabetic.

Table 1.

Metabolic parameters.

Metabolites/parameters	10 weeks		14 weeks		18 weeks		22 weeks		26 weeks
	SD	ZDSD	SD	ZDSD	SD	ZDSD	SD	ZDSD	SD	ZDSD
Body weight (g)	302.8 ± 3.8	305.5 ± 10.8	380.0 ± 6.8	433.8 ± 9.2ǂ	421.9 ± 7.7	496.0 ± 14.99ǂ	459.1 ± 8.5	544.9 ± 12.8§	479.0 ± 10.8	539.5 ± 10.5†
Glucose (mg/dl)	135.4 ± 5.5	149.8 ± 3.3*	100.6 ± 3.2	111.0 ± 2.8*	142.3 ± 5.8	161.9 ± 5.5*	108.8 ± 3.3	153.3 ± 24.4	115.4 ± 7.4	194.8 ± 36.4
Triglyceride (mg/dl)	43.5 ± 3.1	94.6 ± 6.6§	61.0 ± 6.2	93.1 ± 12.3*	100.6 ± 8.6	175.0 ± 27.2*	70.3 ± 5.3	123.6 ± 4.3†	91.0 ± 11	146 ± 26
Total cholesterol (mg/dl)	73.0 ± 2.2	77.3 ± 2.6	64.8 ± 6.8	66.5 ± 3.5	68.6 ± 3.0	67.0 ± 2.1	88.3 ± 3.0	81.5 ± 3.8	100.8 ± 9.0	101.0 ± 7.9
Free cholesterol (mg/dl)	27.8 ± 0.5	30.1 ± 0.6†	21.0 ± 1.1	23.0 ± 1.0	30.1 ± 0.9	28.6 ± 0.8	32.9 ± 1.2	29.3 ± 0.8*	32.8 ± 2.2	27.4 ± 2.9
Cholesterol esters (mg/dl)	45.3 ± 2.1	47.3 ± 2.7	43.5 ± 5.8	43.4 ± 3.4	38.4 ± 2.3	38.5 ± 1.4	55.4 ± 2.85	52.5 ± 3.7	67.9 ± 7.3	73.8 ± 7.2
NEFA (mEq/L)	0.38 ± 0.03	0.54 ± 0.04†	0.45 ± 0.04	0.64 ± 0.04†	0.54 ± 0.05	0.66 ± 0.05	0.51 ± 0.04	0.76 ± 0.05†	0.51 ± 0.13	0.52 ± 0.22

^*P < 0.05.

^†P < 0.01.

^ǂP < 0.001.

^§P < 0.0001.

Fig. 1.

ZDZD rats exhibit higher diet consumption. Rats were fed Purina 5008 diet until they were 26 weeks of age. Diet consumption was monitored at 14, 22, and 26 weeks. Rats were individually housed and given a predetermined amount of food, which was measured again after 4 days. Results are mean ± SE (SD, n = 8; ZDSD, n = 8). Data were analyzed using t-tests, **** P < 0.0001.

Serum chemistry was measured in 4 h fasted rats at 10, 14, 18, 22, and 26 weeks of age (Table 1). At 10 weeks of age, glucose, triglycerides, free cholesterol, and NEFA were significantly elevated in the ZDSD rats. At 14 weeks, glucose, triglycerides, and NEFA were elevated. At 18 weeks, glucose and triglycerides were elevated. At 22 weeks of age, triglycerides and NEFA were elevated, while free cholesterol was significantly decreased. There was no difference in total cholesterol at any time point measured. However, at 22 weeks, the ZDSD rats started becoming spontaneously diabetic, one ZDSD rat had glucose levels over 300 mg/dl. At 26 weeks of age, one ZDSD rat had glucose levels over 200 mg/dl, while two ZDSD rats had glucose levels over 300 mg/dl, causing more variation in statistical analyses. Body composition determined by Dual-energy X-ray absorptiometry (DXA) was performed one day prior to termination. ZDSD rats had significantly increased Lean + BMC (bone mineral content), and Total Mass (Table 2). When grouping the ZDSD rats by non-diabetic and those who had lost weight due to hyperglycemia (glucose 125–249 mg/dl) and diabetes (glucose ≥250 mg/dl), the non-diabetic rats had significantly increased Fat, Lean + BMC, Total Mass, and Area (Table 2). The increase in total mass is consistent with the individual tissue findings. When normalized by body weight, the ZDSD rats had significantly increased liver, heart, peritoneal and retro peritoneal fat depots, while epidydimal, sub-cutaneous, and brown fat depots were unchanged (Table 2). We also examined the liver triglycerides, which showed similar levels in SD and ZDSD rats after normalization by liver weight (Table 2).

Table 2.

Dual-energy X-ray absorptiometry (DXA) analysis of body composition.

Parameters	SD (n = 8)	ZDSD (n = 8)
Fat (g)	57.41 ± 5.77	74.55 ± 7.59
Lean + BMC (g)	394.3 ± 7.31	419.6 ± 5.27*
Total Mass (g)	451.8 ± 10.97	494.1 ± 11.13*
% Fat	12.6 ± 1.02	14.9 ± 1.28
Area (cm2)	56.84 ± 6.54	69.16 ± 1.41
BMC (g)	10.09 ± 0.19	10.87 ± 0.17†
BMD (g/cm2)	0.159 ± 0.003	0.157 ± 0.002
Liver (% BW)	3.01 ± 0.05	3.47 ± 0.15*
Liver triglyceride (mg/g)	7.02 ± 0.49	6.63 ± 0.17
Heart (% BW)	0.32 ± 0.01	0.28 ± 0.01*
Epidydimal (% BW)	1.40 ± 0.11	1.31 ± 0.08
Peritoneal (% BW)	0.51 ± 0.07	1.02 ± 0.09ǂ
Retro peritoneal (% BW)	0.98 ± 0.10	1.65 ± 0.011*
Subcutaneous (% BW)	1.59 ± 0.22	1.86 ± 0.31
Brown (% BW)	0.13 ± 0.01	0.14 ± 0.03

^*P < 0.05.

^†P < 0.01.

^ǂP < 0.001.

To test insulin sensitivity, ITT was performed at 17 weeks and 23 weeks of age. At 17 weeks of age, ZDSD glucose levels were significantly higher prior to the insulin injection, and at 15, 30, and 45 min post injection (data not shown). At 23 weeks of age, ZDSD glucose levels were significantly higher prior to injection, and at 15 and 60 min post injection (Fig. 2A). When grouping the ZDSD rats by non-diabetic and those who had lost weight due to hyperglycemia, the ZDSD-diabetic animals had significantly higher glucose levels at every time point, while the non-diabetic animals were significantly increased prior to insulin injection and at 15 min post injection. However, only the ZDSD-diabetic group had increase area under the curve (Supplemental Fig. S1).

Fig. 2.

ZDSD rats exhibit impaired insulin and glucose tolerance. (A) Insulin Tolerance Test was performed at 23 weeks of age. (B) Glucose Tolerance Test was performed at 26 weeks of age. AUC was determined by the trapezoidal method. Results are mean ± SE (SD, n = 8; ZDSD, n = 8). Data were analyzed using t-tests, * P < 0.05; ** P < 0.01; ***P < 0.001; **** P < 0.0001. (C) Histologic analysis of hepatic and pancreatic samples from ZDSD and SD control rats. Hematoxylin and eosin staining of liver samples from 26 weeks old ZDSD (a) and SD (b) rats. Pancreas sections from 26 weeks old ZDSD (c), SD (d), and 15 weeks old ZDSD (e) rats were stained with anti-glucagon (brown) and anti-insulin (red) antibodies. Quantification of glucagon and insulin staining was performed on three areas of sections of each animal (f), and five animals in each group were analyzed. * P < 0.05; ** P < 0.01 for t-test between SD and ZDSD, ## P < 0.01 for t-test between SD and ZDSD young.

To test glucose tolerance, GTT was performed at 17 weeks and 26 weeks of age. At 17 weeks of age, the ZDSD rats had significantly higher glucose at 60- and 120-min post injection (data not shown), at 26 weeks of age, glucose levels were higher at every time point during GTT (Fig. 2B). Moreover, the area under the curve for the glucose response increased significantly in ZDSD rats. When grouping the ZDSD rats by non-diabetic and those who had lost weight due to hyperglycemia, the ZDSD-diabetic animals had significantly higher glucose levels at every time point post injection when compared to the SD group. However, the non-diabetic ZDSD group was still significantly higher from the SD group at every time point, with both ZDSD groups having increased AUC when compared to the SD group (Supplemental Fig. S1). These data suggest that ZDSD rats have impaired insulin sensitivity and glucose tolerance.

Histochemical analysis of liver and pancreas was conducted as shown in Fig. 2C. Although there was no significant difference in the liver histology between ZDSD and SD rats, the pancreas of ZDSD rats already showed significant destruction of islet structure as early as 15 weeks. Quantification analysis showed significant decrease in the area of cells with insulin staining as well as glucagon staining at both 15 weeks and 26 weeks of age in ZDSD animals.

At 24 weeks of age, the animals underwent blood pressure measurements. Compared to SD rats, ZDSD rats had increased systolic blood pressure (Fig. 3). However, there was no change when comparing mean or diastolic blood pressure measurements (Fig. 3).

Fig. 3.

ZDSD rats exhibit higher systolic blood pressure. Blood pressure measurements from rats at 24 weeks of age (Refer to the Methods and Materials for details). Results are mean ± SE. Data were analyzed using t-tests, **P < 0.01.

3.2. ZDSD rats exhibit up-regulated gene and/or protein expressions in lipid metabolism, amino acid metabolism, and glycolysis/TCA cycle

We evaluated mRNA levels of key enzymes/proteins and transcription factors involved in hepatic lipid metabolism, amino acid metabolism, and glycolysis. These studies were achieved by real-time qRT-PCR and results are shown in Figs. 4–10.

Fig. 4.

ZDSD rats exhibit higher expressions in fatty acid binding proteins and fatty acid oxidation genes. (A) CD36. (B) FABP1. (C) FABP4. (D) CPT1a. (E) CPT2. (F) ACAD1. Quantitative RT-PCR of RNA from livers of chow-fed SD rats and chow-fed ZDSD rats for 16 weeks. Results are mean ± SE of 3 independent RT-PCR experiments for each gene within each group (SD, n = 8; ZDSD, n = 8). Data are presented relative to Rplp0 in the same samples; relative mRNA levels were determined by qRT-PCR using the comparative C_T method. * P < 0.05; ** P < 0.01.

Fig. 10.

ZDSD rats exhibit higher protein expressions in ER stress, fatty acid binding protein and lipogenic enzymes. (A) Immunoblot of IRE1α, P-IRE1α, CHOP, FABP1 and ß-actin in hepatic extracts from SD and ZDSD rats. (B) Quantification of scanned data from Panel A. (C) Immunoblot of ACC, P-ACC, GSK-3β, P-GSK-3β and ß-actin in hepatic extracts from SD and ZDSD rats. (D) Quantification of scanned data from Panel C. After 16 weeks of chow diet feeding, animals were fasted for 4 h before tissue collection. Liver homogenates were immunoblotted for IRE1α, P-IRE1α, CHOP, FABP1, ACC, P-ACC, GSK-3β, P-GSK-3β proteins. β-actin was used as control in all western blots. Quantitative data presented in Panels B and D are means ± SE of 4–7 individual samples. * P < 0.05; ** P < 0.01; *** P < 0.001; **** P < 0.0001.

In Fig. 4A–C, the mRNA levels of key fatty acid binding/transport proteins, Cd36 (CD36/FAT), Fabp1 (L-FABP) and Fabp4 (A-FABP) were significantly increased (~8 fold, ~3 fold and ~3 fold, respectively) in ZDSD rats as compared to SD rats. In agreement, we also found an increase in the protein level of FABP1 in ZDSD (Fig. 10A, B). Next, we examined the expressions of key fatty acid oxidation genes and observed a significant increase (~3 fold) in Cpt1a (CPT1) mRNA level in ZDSD rats, while the expressions of Cpt2 (CPT II) and Acadl (LCAD) trended to be higher in ZDSD rats than SD rats, but did not reach statistical significance (Fig. 4D–F).

We also examined the expressions of several genes that are involved in ER stress and found that there were significant increases (~2.3 fold, ~1.4 fold and ~1.4 fold, respectively) in Ddit3 (CHOP), Hsp90b1 (GRP94) and Hspa5 (GRP78/BIP) mRNA levels in ZDSD rats, while Eif2a showed an increasing trend that did not reach statistical significance (Fig. 5A–D). With regards to phospholipid synthesis, we observed significant increases (~1.6 fold, ~1.9 fold, ~2.1 fold, ~2.3 fold, ~1.8 fold and ~1.2 fold, respectively) in the expressions of phospholipid synthase enzymes such as Ptdss1, Ptdss2, Pcyt1a, Pcyt2, Chpt1 and Pemt (Fig. 5E–J).

Fig. 5.

ZDSD rats exhibit higher expressions in ER stress and phospholipid synthase genes. (A) CHOP. (B) GRP94. (C) BIP. (D) Eif2a. (E) Ptdss1. (F) Ptdss2. (G) Pcyt1a. (H) Pcyt2. (I) Chpt1. (J) Pemt. Quantitative RT-PCR of RNA from livers of chow-fed SD rats and chow-fed ZDSD rats for 16 weeks. Results are mean ± SE of 3 independent RT-PCR experiments for each gene within each group (SD, n = 8; ZDSD, n = 8). Data are presented relative to Rplp0 in the same samples; relative mRNA levels were determined by qRT-PCR using the comparative C_T method. * P < 0.05; ** P < 0.01; *** P < 0.001.

In addition, we measured mRNA levels of genes that play important roles in lipogenesis and triglyceride synthesis. Scd1 (SCD1) and Dgat2 (ARAT) are two key enzymes that were up-regulated in ZDSD rats (~2.4 fold and ~3.2 fold) (Fig. 6A, B), while two other genes involved in fatty acid synthesis, Acaca (ACC1) and Fasn (FAS), showed similar mRNA expression levels in ZDSD and SD rats (Fig. 6C, D). In Fig. 6E–J, Slc16a7, Hmgcl, Plin4 and Gde1 expression levels were up-regulated in ZDSD rats (~1.7 fold, ~1.7 fold, ~1.8 fold and ~2.6 fold, respectively); likewise, we evaluated the expression of Mttp (MTP), a gene that participates in VLDL-TG assembly/secretion, and found that its expression level was increased significantly in ZDSD rats (~5.3 fold). Lpl, a lipid hydrolyzing enzyme, was found to be up-regulated in ZDSD rats (~3.2 fold).

Fig. 6.

ZDSD rats exhibit higher expressions in lipogenesis and other miscellaneous genes. (A) SCD1. (B) DGAT2. (C) ACC1. (D) FASN. (E) SLC16a7. (F) HMGCL. (G) MTTP. (H) GDE1. (I) Plin4. (J) LPL. Quantitative RT-PCR of RNA from livers of chow-fed SD rats and chow-fed ZDSD rats for 16 weeks. Results are mean ± SE of 3 independent RT-PCR experiments for each gene within each group (SD, n = 8; ZDSD, n = 8). Data are presented relative to Rplp0 in the same samples; relative mRNA levels were determined by qRT-PCR using the comparative C_T method. * P < 0.05; ** P < 0.01; *** P < 0.001; **** P < 0.0001.

In Fig. 7, we measured mRNA levels of a series of genes that are involved in glycolysis, which converts glucose into pyruvate. Among them, glucokinase facilitates the phosphorylation of glucose in the liver and the pancreatic β cell and, thus, contributes to the maintenance of blood glucose homeostasis. Glucokinase has three isoforms and we examined each one by designing different targeting primers. We found that glucokinase isoform 1 was down-regulated in ZDSD rats (~0.2 fold); glucokinase isoform 2 was up-regulated in ZDSD rats (~2.5 fold); and glucokinase isoform 3 was not changed (Fig. 7A–C). In liver, glucokinase isoform 2 is the major type. Consistent with its up-regulated expression, the expression level of Gckr (glucokinase regulator), which controls and regulates glucokinase activity, tended to be increased although not significantly (Fig. 7D). Gpi (PGI), Pfkl (PFK-L), Tpi1(TPI), and Pgk1(PGKA) showed significant increased mRNA expression levels in ZDSD rats (~1.8 fold, ~2.3 fold, ~1.6 fold and ~1.9 fold, respectively), while Hk1(HK1) and Pklr (PKL) were not changed in ZDSD rats (Fig. 7E–J).

Fig. 7.

ZDSD rats exhibit higher expressions in glycolysis genes. (A) Gck1. (B) Gck2. (C) Gck3. (D) Gckr. (E) HK1. (F) PGI. (G) Pfk1. (H) Tpi1. (I) Pgk1. (J) PK1. Quantitative RT-PCR of RNA from livers of chow-fed SD rats and chow-fed ZDSD rats for 16 weeks. Results are mean ± SE of 3 independent RT-PCR experiments for each gene within each group (SD, n = 8; ZDSD, n = 8). Data are presented relative to Rplp0 in the same samples; relative mRNA levels were determined by qRT-PCR using the comparative C_T method. ** P < 0.01; **** P < 0.0001.

In Fig. 8, we measured mRNA levels of four key genes that are involved in the tricarboxylic acid (TCA) cycle. The TCA cycle is a series of chemical reactions that release stored energy through the oxidation of acetyl-CoA derived from carbohydrates, fats, and proteins into carbon dioxide and ATP. We observed that Cs, Aco1, and Suclg1 mRNA levels are all increased in ZDSD rats (~1.8 fold, ~1.6 fold and ~1.6 fold, respectively), while Fh1 is not altered (Fig. 8A–D).

Fig. 8.

ZDSD rats exhibit higher expressions in TCA cycle genes. (A) Cs. (B) Aco1. (C) Fh1. (D) Suclg1. Quantitative RT-PCR of RNA from livers of chow-fed SD rats and chow-fed ZDSD rats for 16 weeks. Results are mean ± SE of 3 independent RT-PCR experiments for each gene within each group (SD, n = 8; ZDSD, n = 8). Data are presented relative to Rplp0 in the same samples; relative mRNA levels were determined by qRT-PCR using the comparative C_T method. ** P < 0.01; *** P < 0.001.

Moreover, we measured mRNA levels of eight genes involved in amino acid metabolism. Arg1, Tdo2, and Glud1 showed increases (~1.4 fold, ~1.4 fold and ~1.9 fold, respectively) in mRNA levels in ZDSD rats, whereas Gpt, Sds, Hal, Got1, Tat showed no significant changes in ZDSD rats as compared with SD rats (Fig. 9).

Fig. 9.

ZDSD rats exhibit higher expressions in amino acid metabolism genes. (A) ALT1. (B) SDH. (C) Hal. (D) AST1. (E) Arg1. (F) Tat. (G) Tdo2. (H) GDH1. Quantitative RT-PCR of RNA from livers of chow-fed SD rats and chow-fed ZDSD rats for 16 weeks. Results are mean ± SE of 3 independent RT-PCR experiments for each gene within each group (SD, n = 8; ZDSD, n = 8). Data are presented relative to Rplp0 in the same samples; relative mRNA levels were determined by qRT-PCR using the comparative C_T method. * P < 0.05; *** P < 0.001.

Since mRNA expression levels of key ER stress genes were increased in ZDSD rats, we measured the protein levels of two ER stress markers, CHOP (C/EBP homologous protein) and IRE1α. The unfolded protein response (UPR) is initiated after ER stress by three ER proteins: PERK (PKR-like ER kinase), IRE1 (Inositol Requiring 1) and ATF6 (Activating Transcription Factor 6). CHOP is regulated by the PERK-eIF2α-ATF4 pathway and has been shown to have a role in ER stress mediated apoptosis. As shown in Fig. 10A, ZDSD rats exhibit much higher protein levels of CHOP (~1.78 fold) compared to SD rats. Upon ER stress, IRE1 autophosphorylates to become active. The immunoblots demonstrated that both phosphorylated IRE1α (Ser724) and total IRE1α protein levels, after normalization to β-actin, were up-regulated in ZDSD rats by approximately 2.56-fold and 2.72-fold, respectively. However, the phosphorylation of IRE1α after normalization to total IRE1α did not exhibit an increase in ZDSD rats (Fig. 10A, B).

We also measured the total and phosphorylated levels of ACC and GSK3β. AMPK (AMP-activated protein kinase) is an enzyme that plays a role in cellular energy homeostasis. AMPK functions by phosphorylating key enzymes in metabolic pathways, as well as transcriptional factors. It phosphorylates ACC (acetyl-CoA carboxylase) on Ser79 (major phosphorylation site) and deactivates its activity. GSK3 (glycogen synthase kinase 3) is a Ser/Thr kinase which is rapidly phosphorylated at Ser9 by AKT, resulting in the inhibition of its kinase activity. In Fig. 10, both phosphorylated ACC (P-ACC) and phosphorylated GSK3β (P-GSK3β) showed significantly increased levels in ZDSD rats (Fig. 10C, D). Total ACC was also increased in ZDSD rats. However, when normalized to total protein, neither GSK3β nor ACC exhibited an increase in ZDSD rats (Fig. 10C, D).

4. Discussion

The present study was conducted to characterize metabolic endpoints, islet histology, as well as changes in molecular events connected with major metabolic pathways, in ZDSD rats compared to age matched SD rats. These studies were necessary given that the ZDSD model more closely resembles the human condition by having an intact leptin pathway, making it appear to be more suitable for performing preclinical studies (https://www.crownbio.com/metabolic-disease/diabetes/zdsd-rat) as compared to more commonly used spontaneously/congenital, diet induced, chemical-induced, and transgenic models [23–28,30].

A major drawback to the use of non ZDSD models is that none of these models fully mimics the pathophysiology of human MetS. Here we show that ZDSD rats exhibit impaired insulin and glucose tolerance, together with increased systolic blood pressure. Further analysis revealed that the islet structure in these rats, as a basis for impaired insulin secretion, is severely damaged as early as 15 weeks old. In addition, this study also demonstrated significant changes in hepatic mRNA and/or protein expression of key enzymes or regulatory proteins involved in fatty acid transport and mitochondrial fatty acid oxidation, lipogenesis, VLDL-TG turnover, glycolysis, TCA cycle, phospholipid biosynthesis, ER stress and amino acid metabolism using livers from 26-week-old ZDSD rats compared to 26-week-old control (SD) rats.

As noted above, the ZDSD rat is an inbred polygenic model for MetS, obesity, diabetes and diabetic complications. Unlike the ob/ob mouse, db/db mouse, ZF rat or ZDF rat, the ZDSD rat does not have a deficiency (mutation) of leptin or leptin receptor leading to the development of obesity and features of MetS and T2D, and, as a result, can fully duplicate the human disease condition. Thus, it is the most translatable rodent model for these disease conditions. In this model, like human disease, prediabetes starts around 8–16 weeks of age, which is accompanied by development of components of MetS, including insulin resistance, glucose intolerance, hyperlipidemia, systolic hypertension and inflammation, followed by progression to overt diabetes starting at >16 weeks of age [33,34,36]. We have confirmed these metabolic changes by performing GTT, ITT, blood chemistry, and blood pressure measurements. We did not find any evidence of steatosis, as this study was performed on chow-fed animals.

The qRT-PCR measurements indicated that the gene expression of proteins involved in hepatic fatty acid uptake and intracellular transport, such as CD36, Fabp1 and Fabp4 and Cpt1a, an enzyme involved in fatty acid β-oxidation, are altered in livers of ZDSD rats. The expression of other hepatic mitochondrial β-oxidation genes, such as Cpt2 and Acadl, however, is not affected in ZDSD rats. CD36 (also known as FA translocase) functions as an important mediator of cellular fatty acid uptake in skeletal muscle, cardiac muscle, adipose tissue and other tissue/cell types [37]. CD36 expression is relatively low in liver under normal physiological conditions, but can be highly upregulated under hyperlipidemic conditions and contributes to development of hepatic steatosis [38]. Our studies demonstrate that hepatic CD36 gene expression is upregulated in ZDSD rats. These studies are in agreement with previous studies showing that both hyperinsulinemia and diet-induced obesity lead to enhanced expression of CD36, which, in turn, contributes to hepatic insulin resistance, dyslipidemia and steatosis [39,40]. Likewise, mRNA expression of fatty acid binding protein 1/FABP1/L-FABP (Fabp1 gene) and fatty acid binding protein 4/FABP4/A-FAB (Fabp4 gene) is also, higher (3–4-fold) in ZDSD liver. We also observed increased protein levels of FABP1 in the liver of ZDSD rats. Previously, both FABP1 and FABP4 have been implicated in the pathogenesis of hepatic steatosis and NAFLD [41,42]. We observed that mRNA expression of hepatic Cpt1a, which encodes CPT1 protein, an important component of the “carnitine shuttle” which facilitates the transport of long chain fatty acids into the mitochondrial matrix, the site of β-oxidation [43,44], is selectively upregulated in ZDSD rats. In contrast, no changes, however, were noted in the gene expression of a representative enzyme, acyl-CoA dehydrogenase, long chain (Acadl gene), directly involved in mitochondria fatty acid oxidation [45]. Based on these observations, it appears that activation of the β-oxidation pathway alone is insufficient to overcome dyslipidemia.

Given that ZDSD rats develop diabetes with varying degrees of severity as they age and that ER stress [46–49] is implicated in the pathogenesis of diabetes [46,48], we compared the changes in key proteins, signaling molecules and transcription factors involved in ER stress and associated UPR and cell survival and death [46–49]. Of interest was the significant upregulation of hepatic mRNA levels of transcription factor CCAAT-enhancer-binding protein (C/EBP) homologous protein (CHOP), also known as DNA-inducible protein 153 (GADD153) (Ddit3 gene), heat shock protein 90beta, member 1 (glucose-regulated protein 94 [GRP94]) (Hsp90b1 gene) and heat shock protein family A (GRP78, BiP) (Hspa5 gene), in ZDSD rats as compared to SD rats. Here we observed that the hepatic expression of both proapoptotic CHOP and antiapoptotic GRP78/BiP proteins is upregulated in ZDSD rats. Although more investigation is needed, the increased expression of GRP78/BiP likely represents a compensatory mechanism to counteract the actions of CHOP. We also observed that the protein levels of total and phosphorylated forms of IRE1α (inositol-requiring enzyme 1α) increased significantly in the liver of ZDSD rats. The activated IRE1α causes unconventional processing of X-box-binding protein 1 (Xbp1) mRNA, yielding hybrid XBP1s that acts as a potent transcription factor to induce the expression of genes involved in protein folding and export from the ER, export and degradation of misfolded proteins, and lipid biosynthesis to improve ER stress [49–51]. Moreover, ER stress mediated activation of IRE1α is known to activate its downstream signaling kinase, c-Jun NH₂-terminal kinase (JNK), which, in turn, can lead to suppression of insulin receptor signaling by catalyzing serine phosphorylation of the insulin receptor substrate (IRS)-1, thus interfering with tyrosine phosphorylation of IRS-1 and the downstream insulin signaling cascade, culminating into insulin resistance [52,53]. Our findings that ZDSD rats are insulin resistant support this possibility. Taken together, these results provide evidence that ER stress is increased in the liver of ZDSD rats based on the upregulation of the expression/activity of several proteins that mediate ER stress or participate in the three integrated UPR-linked signaling pathways, the IRE1 pathway, the activating transcription factor 6 (ATF6) pathway, and the protein kinase RNA (PKR)-like ER kinase (PERK) pathway.

Many recent studies have revealed that ER stress and UPR activation play critical roles in the regulation of lipid homeostasis with ER-induced dysregulation of lipid metabolism leading to dyslipidemia [54–57]. We reasoned that since ZDSD mice had elevated levels of ER stress they might also exhibit altered hepatic lipid metabolism [54–57]. To address this possibility, we evaluated the mRNA/protein levels of key enzymes or proteins involved in lipogenesis [58], triglyceride synthesis [59,60], VLDL-TG assembly [61,62], phospholipid synthesis [54,63] and some miscellaneous proteins involved in lipid metabolism [58]. Our PCR measurements showed no changes in mRNA levels of two key enzymes involved in lipogenesis, ACC1 and FAS, but showed increased mRNA expression of SCD1 and DGAT2 enzymes, which are components of the TG synthesis pathway [58–60], as well as MTP, a key protein that facilitates hepatic VLDL-TG assembly [61,62]. Interestingly, total protein levels of ACC1 were increased, but no changes were noted regarding the ratio of pACC1/ACC1 between SD and ZDSD, suggesting that observed increases in ACC1 protein levels might be due to increased protein stability. The increases in SCD1, DGAT2, MTP and perilipin 4 or S3–12 protein (which plays a role in TG packing and biosynthesis of lipid droplets) mRNA levels in ZDSD livers may be in response to elevated levels of IRE1α. IRE1α-XBP1 is known to play a critical role in VLDL assembly and secretion, and IRE1α is required for optimum secretion of hepatic VLDL and LDL under conditions of excessive ER stress [57,64]. Finally, mRNA measurements of key enzymes involved in hepatic phospholipid synthesis except for phosphatidylethanolamine N-methyltransferase [54,63] are upregulated in ZDSD rats. This may be required for the optimum production of phosphatidylcholine (PC) and other phospholipids for efficient production and secretion of hepatic VLDL-TG during ER-stress.

Given that glycolysis converts glucose-derived excess carbohydrate to pyruvate and, subsequently, acetyl-CoA via the tricarboxylic acid cycle (TCA)/Krebs cycle and is converted to citrate substrate for lipogenesis, we measured mRNA changes in key enzymes involved in glycolysis [65], including insulin-sensitive glucokinase (Gck) [66], and in the TCA cycle [67]. Measurements indicated that mRNA levels of several hepatic glycolytic and TCA cycle enzymes are upregulated in ZDSD rats, suggesting the possibility that both glycolysis and the TCA cycle pathways are upregulated to meet excessive demand for citrate substrate for fatty acid synthesis and their utilization for VLDL-TG assembly. In addition, the mRNA expression of key liver enzymes of SD and ZDSD rats involved in amino acid catabolism [68] including several transaminases, which have the potential to provide keto acids (e.g., pyruvate, alpha-ketoglutarate) for the TCA cycle, surprisingly showed no changes in their mRNA levels between SD and ZDSD rats except for increased expression of arginase 1, tryptophan 2,3-dioxygenase and glutamate dehydrogenase 1 is upregulated in ZDSD livers, although the physiological significance of these increases is not apparent. Some caveats that should be taken into consideration are that these interpretations are based on gene expression analyses of various pathways, but enzymatic activities have not been measured; therefore, these interpretations might be considered speculative to some extent.

In summary, based on its intact leptin system, its insulin deficiency, as well as its timeline of disease development without diet manipulation, the insulin resistant, dyslipidemic, hypertensive, and diabetic ZDSD rat represents an additional, unique polygenic animal model that could be very useful to study human diabetes. We found that ZDSD rats exhibit elevated levels of ER-stress markers, which correlate with the dysregulation of hepatic lipid metabolism in these rats. mRNA expression of key apoptotic and antiapoptotic proteins that participate in UPR-linked signaling pathways, IRE1, ATF6, and PERK, are upregulated and likely to contribute to the pathogenesis of hepatic dyslipidemia and insulin resistance. More importantly, results from this study strongly indicate a dual role for the IRE1α-XBP1 pathway; increased activity of IRE1α may suppress insulin signaling and, thus, cause insulin resistance, whereas its increased levels may also help to improve dyslipidemia by promoting efficient clearance of hepatic VLDL-TG. Obviously, more experimental strategies will be needed in the future to sort out these possibilities. In addition, we found that expression of genes involved in hepatic glucose and amino acid metabolism are also significantly altered in ZDSD rats.

Abbreviations:

MetS

metabolic syndrome

T2D

type 2 diabetes

CVD

cardiovascular disease

ZDSD

Zucker diabetic Sprague–Dawley

NEFA

non-esterified fatty acids

GTT

glucose tolerance test

ITT

insulin tolerance test

BMC

bone mineral content

TCA

tricarboxylic acid

UPR

unfolded protein response

NAFLD

Nonalcoholic fatty liver disease

HFrD

high fructose diet

OMM

outer mitochondrial membrane

IMM

inner mitochondrial membrane