Transcription Profile Unveils the Cardioprotective Effect of Aspalathin against Lipid Toxicity

Aspalathin, a C-glucosyl dihydrochalcone, has previously been shown to protect cardiomyocytes against hyperglycemia-induced shifts in substrate preference and subsequent apoptosis. However, the precise gene regulatory network remains to be elucidated. To unravel the mechanism and provide insight into this supposition, the direct effect of aspalathin in an isolated cell-based system, without the influence of any variables, was tested using an H9c2 cardiomyocytes model. Cardiomyocytes were exposed to high glucose (33 mM) for 48 hours before post-treatment with or without aspalathin. Thereafter, RNA was extracted and RT2 PCR Profiler Arrays were used to profile the expression of 336 genes. Results showed that, 57 genes were differentially regulated in the high glucose or high glucose and aspalathin treated groups. STRING analysis revealed lipid metabolism and molecular transport as the biological processes altered after high glucose treatment, followed by inflammation and apoptosis. Aspalathin was able to modulate key regulators associated with lipid metabolism (Adipoq, Apob, Cd36, Cpt1, Pparγ, Srebf1/2, Scd1 and Vldlr), insulin resistance (Igf1, Akt1, Pde3 and Map2k1), inflammation (Il3, Il6, Jak2, Lepr, Socs3, and Tnf13) and apoptosis (Bcl2 and Chuk). Collectively, our results propose that aspalathin could reverse metabolic abnormalities by activating Adipoq while modulating the expression of Pparγ and Srebf1/2, decreasing inflammation via Il6/Jak2 pathway, which together with an observed increased expression of Bcl2 prevents myocardium apoptosis.


Introduction
During the last decade, there has been considerable interest in the use of plant-derived polyphenols as nutraceuticals to slow down the progression of metabolic diseases [1,2].Aspalathus linearis (commonly known as rooibos) is a rich source of plant polyphenols with known health promoting properties.In addition to reversing ischemia/reperfusion injury in the isolated perfused rat heart [3], rooibos has been shown to improve both lipid toxicity and oxidative stress in diabetic individuals at risk of developing cardiovascular disease [4,5].Furthermore, literature has indicated that polyphenols specific to rooibos may present strong ameliorative properties against diabetes mellitus and its associated complications [6][7][8].Of note, aspalathin, a C-glucosyl dihydrochalcone found uniquely in rooibos, has displayed an even greater potency to prevent diabetes-induced cardiovascular complications [9][10][11].For example, a fermented rooibos extract containing abundant levels of aspalathin protected primary isolated rat cardiomyocytes from experimentally induced oxidative stress and apoptosis [9].Moreover, recent data from our laboratory demonstrated that aspalathin protects cultured cardiomyocytes exposed to chronic hyperglycemia from altered substrate metabolism and subsequent cardiac dysfunction [10].
Altered substrate metabolism and reduced myocardial production of adenosine triphosphate (ATP) are main contributors to the development of left ventricular dysfunction, a characteristic of the diabetic cardiomyopathy (DCM).Apart from altered substrate metabolism, DCM has been associated with an enhanced intracellular lipid accumulation as observed in hearts obtained from leptinreceptor-deficient diabetic mice (Lepr db/db ) and diabetic patients with cardiomyopathy [12][13][14].The precise molecular mechanisms associated with the etiology of DCM remains to be fully elucidated.However, gene regulatory pathways of peroxisome proliferator-activated receptor alpha or gamma (Ppparα/γ) and sterol regulatory element-binding protein 1c (Srerbp1c) have been implicated as important molecular switches that regulate shifts in substrate preference and subsequent lipid accumulation in the heart of diabetic patients [15,16].Notably, in a study done by Marfella et al. [15], increased expression of PPARγ and Srebf1c was observed in pressure overload-hearts from diabetic patients.Similarly, in a study done by Son et al. [16], transgenic mice overexpressing PPARγ in the myocardium showed increased expression of fatty acid oxidation and lipid storage genes, inferring that hyperglycemia and impaired lipid accumulation play an important role in the development of DCM.It would therefore be of interest to use RT 2 PCR Profiler Arrays to predict the possible transcriptional mechanisms used by aspalathin to protect the myocardium against the development of high glucose-induced cardiomyopathy.Results obtained showed that aspalathin protected Lepr db/db mice against raised levels of total cholesterol, triglycerides and low-density lipoprotein (LDL) possibly by modulating genes associated with high glucose-induced lipid toxicity, fatty acid oxidation, insulin resistance, inflammation and apoptosis as demonstrated in H9c2 cardiomyocytes.

Effect aspalathin on lipid profile analysis
Effect of aspalathin on lipid profiles: Increased morbidity observed in type 2 diabetes individuals is primarily associated with heart failure, with lipid toxicity being the key pathological mechanism underlying the diabetic condition [5,10,12,14].The ability of aspalathin to improve lipoprotein clearance was assessed by performing, a total lipid profile on mice treated with two dose levels of aspalathin for six weeks (Table 1).The control received no aspalathin.The Lepr db/db_UC mice had increased body weight, FPG, insulin, HOMA-IR levels with associated elevated triglycerides, total cholesterol, LDL and HDL levels when compared to the Lepr db/+_UC group (Table 1).Aspalathin treatment, was able to dose dependently modulate this condition as the effect of the high dose was greater than either the low dose aspalathin or the metformin treated groups.However, similar to metformin, aspalathin treatment did not affect HDL cholesterol or reduce raised FBP levels.Inferring that aspalathin treatment improves serum lipid profiles and cardiac risk associated with an increased LDL (Table 1).However, it is known that Lepr db/db mice display a defective catabolism for the major apolipoprotein A-1 (Apoa1), leading to increased HDL and atherosclerosis protection [13].Similarly, the failure of aspalathin to lower FPG levels can be due to factors such as duration of treatment since previous studies have demonstrated its effectiveness when used at a shorter time interval [7,8].

In vitro effect of aspalathin
To confirm our in vivo findings and to decipher the molecular mechanism used by aspalathin to modulate substrate availability and improve hyperglycemia-associated lipid toxicity, gene expression profiling was performed on H9c2 cardiomyocytes exposed to high glucose.Of the 336 genes assessed, 57 genes (17%) were differentially expressed and of these, 45 and 12 genes (79% and 21%) were hyper-expressed in the high glucose and aspalathin treated groups, respectively (Table 2).STRING data analysis confirmed that aspalathin treatment largely improved the expression of genes involved in metabolic processes, identifying fatty acid and lipid metabolism as the top two regulatory processes (Fig 1).In vitro effect of aspalathin on fatty acid and lipid metabolism Data analysis identified 26 of the 57 (46%) differentially expressed genes to be involved in fatty acid and lipid metabolism (Table 2).STRING network analysis identified three clusters encoding interactive nodes representing genes associated with fatty acid/lipid transport, lipid metabolism and fatty acid metabolism (confidence score, 0.7) (Fig 1).

Increased supply of long-chain fatty acids (LCFAs):
Chronic hyperglycemia has been associated with an increased supply of circulating FFAs to cardiomyocytes [10,26].When this enhanced supply of FFAs exceed the rate of β-oxidation, myocardial triglyceride accumulation occurs, leading to lipotoxicity [14,15].Lipotoxicity subsequently leads to left ventricular dysfunction, a major characteristic feature of DCM.A study done by Drosatos et al. [25] showed that mice with cardiac-specific overexpression of acyl-CoA synthetase (Acsl) developed lipid toxicity and diastolic dysfunction, directly implicating LCFAs in the development of cardiac fibrosis and subsequent myocardial remodeling [25].Based on our dataset, STRING analysis identified a network of genes with nodes linking long-chain fatty acyl-CoA synthetase enzymes (Fig 1).These data confirm previous findings [25], where high glucose exposure resulted in increased mRNA expression of acyl-CoA synthetase long-chain family member 4 and 6 (Acls4, 5.6-fold and Acls6, 2.4-fold) as well as acyl-CoA synthetase medium-chain family member 3 and 4 (Acsm3, 14.2-fold and Acsm4, 6.1-fold) (Fig 1 and Table 2).Results obtained showed that aspalathin treatment was able to attenuate this effect.Altered lipid metabolism and increased cholesterol flux: Scientific evidence has shown that excessive cardiac lipid accumulation contributes to the development of cardiac dysfunction [10][11][12][13].Lipid toxicity in the myocardium has been associated with the transcriptional factor sterol regulatory element-binding protein 1/2 (Srebf1/2) and the transcriptional coactivator peroxisome proliferatoractivated receptor-gamma (PPARγ) [25,27].Srebf1/2 and PPARγ are important switches that regulate lipid accumulation and lipotoxicity.In a study done by Marfella et al. [15] on biopsies from diabetic patients with left ventricular dysfunction, increased cardiac lipid deposits were concomitant with enhanced mRNA expression of Srebf1/2, PPARγ as well as genes associated with accelerated βoxidation.Notably, in this study, elevated levels of circulating FFAs were observed after high glucose exposure.This was parallel to enhanced mRNA expression of Srebf1/2 (3.4 and 2.0-fold, respectively), as well as PPARγ (8.3-fold) (Fig 1 and Table 2).Increased expression of Srebf1 is further linked to the development of an artherogenic apolipoprotein profile, a characteristic of cardiac hypertrophy [29,30].Apolipoproteins play an important role in the regulation of lipoprotein metabolism, with the main function being the transport of triglycerides and cholesterol through the lymphatic and circulatory systems.[30].In particular, Apoa1 is the major protein component of HDL.HDL promotes efflux of cholesterol, phospholipids, and other lipophilic molecules from cells by an active process mediated by a cell-membrane transporter, ATP-binding cassette transporter (Abca1) [28,30].By contrast, apolipoprotein B (Apob) is the main apolipoprotein of chylomicrons and LDL.Elevated LDL levels have been associated with an increased risk of heart failure [28,30].In this study, we observed an increased mRNA expression of Apob (7.7-fold), apolipoprotein E (Apoe; 4.4-fold) and very lowdensity lipoprotein (Vldlr; 2.0-fold) that was decreased after aspalathin treatment (Fig1).Similar to increased HDL cholesterol in Lepr db/db mice, Apoa1 and its transporter Abca1 were increased by 4.2fold and 2.0-fold, respectively, after high glucose exposure.Aspalathin treatment reversed this effect.Of interest, increased Apob has been associated with upregulated expression of Cd36, the receptor for oxidized LDL.Furthermore, insulin sensitivity and improved lipotoxicity have also been linked to increased expression of the protein coding gene adiponectin, C1Q and collagen domain containing (Adipoq) [29].Adipoq is the gene that encodes for the protein adiponectin.Adiponectin influences insulin sensitivity and lipid metabolism and is primarily expressed in adipose tissue [31], but can also be found in vascular tissue [22,32].In this study, high glucose exposure resulted in the reduced expression of Adipoq (-6.0-fold), while aspalathin was able to reverse this effect.
Cardiac lipotoxicity caused by chronic hyperglycemia may lead to the development of cardiac fibrosis [30,32].We observed that aspalathin treatment was able to reverse lipid toxicity by modulating key regulatory genes, such as Adipoq PPARγ and Sreb1/2 (Fig 1).We propose that aspalathin can prevent lipid accumulation by increasing the expression of Adipoq.This result in decreased PPARγ and Srebf1/2 expression, leading to reduced lipid accumulation in H9c2 cardiomyocytes.This data were parallel to the observed reduction in systemic total cholesterol, triglycerides and low-density lipoprotein in Lepr db/db mice (Fig 1 and Table 2).

In vitro effect of aspalathin on the development of insulin resistance
It is well described that increased LCFAs result in the intramyocardial accumulation of diacylglycerol that activates protein kinase C (theta isoform), leading to the inhibition of insulin receptor substrate 1 and the development of an insulin resistant phenotype [10][11][12].The latter can have a profound effect on cardiac performance, as the heart is an insulin-responsive organ and hyperglycemia-induced insulin resistance decreases cardiac performance.Data analysis revealed that 14 (25%) of the 57 differentially expressed genes were associated with the development of hyperglycemic-induced insulin resistance (Table 2).Network mapping of the differential expressed genes identified two major interconnecting clusters (genes associated with protein kinase activity and development of insulin resistance) with serine/threonine-protein kinase homolog 1 (Akt1) and mitogen activated protein kinase (Mapk) being the major nodes of connection (confidence score, 0.  Insulin signaling and effect on myocardium: Akt1 is a pro-survival protein kinase that plays an important role in the regulation of various cellular functions, including metabolism (glucose and lipids), growth, migration, proliferation and cell survival [33].STRING network analysis identified Akt1 as the central node associated with 12 of the 14 differentially expressed genes.Akt1 constitutes an important node with diverse signaling cascades.In this study, a direct link was observed between phosphodiesterase 3B (Pde3b) and Akt1.Studies have demonstrated that cAMP-dependent protein kinase signaling is impaired in diabetes and is associated with cardiac dysfunction [34,35].Cyclic AMP (cAMP) plays a significant role in the thermogenic process as, well as potentiating glucosestimulated insulin release [34,35].In contrast, Pde3b is a negative regulator of cAMP and increased expression of this gene inhibits or diminishes the effect of cAMP [28].Pde3b is highly expressed in the myocardium and is known to decrease myocardial smooth muscle contractility [36].In studies done on isolated rat islets, inhibition of Pde3 was reported to improve insulin release [37].In the present study, high glucose exposure resulted in the upregulation of Pde3b by 2.1-fold (Fig 1 and Table 2).However, treatment with aspalathin was able to ablate this response, implying that aspalathin might act as a possible inhibitor of Pde3, a potential therapeutic target to protect the diabetic heart.Furthermore, Akt1 was found to be associated with various antioxidant genes including superoxide dismutase 2 (Sod2) and uncoupling protein 1 (Ucp1).Although predominantly found in the brown adipose tissue Ucp1 is also expressed in cardiac cells where, together with Sod2, it inhibits the damaging effects of reactive oxygen species (ROS) [38].Aspalathin was able to improve Sod2 (3.1fold) expression while the effect of high glucose on Ucp1 (-1.7-fold) expression was decreased (Fig 2).Interestingly, in a study done by Barreto et al. [39], where they investigated the mechanism of Ucp1 action on stress response, they observed that increased Ucp1 expression induced the upregulation of various antioxidant stress-response genes, such as Sod1, glutathione-S-transferase and glutathione peroxidase, concomitantly reducing ROS.Similarly, in a study by Dludla et al. [9], a rooibos extract containing 0.4% aspalathin and other antioxidant flavonoids significantly enhanced glutathione levels of cardiomyocytes isolated from diabetic rats exposed to ROS.Based on these findings, we speculate that the observed increased expression of Ucp1 expression could have been a compensatory mechanism used by the cells to ameliorate the adverse effects of increased FFA and hyperglycemiainduced ROS.
Impaired myocardial substrate preference, due to aberrant FFA levels and insulin resistance, activates a myriad of other maladaptive signaling pathways.This includes activation of insulin-like growth factor 1 (Igf1), a hormone similar in structure to insulin and known to be a physiological regulator of glucose transport and glycogen synthesis [40].Igf1 activates and phosphorylates Akt1 to attenuate the development of diabetes-induced myocardial apoptosis, by inhibiting tumor protein 53 [41].In this study, high glucose had no effect on the expression of Igf1, but decreased Akt1 mRNA expression (-2.2-fold) (Fig 2).Conversely, the tumor necrosis factor ligand superfamily member 6 (Faslg) that interacts with Akt1, has been shown to induce apoptosis through binding and inhibiting the pro-survival gene Cflar [42].In this study, high glucose treatment increased mRNA expression of Faslg (4.8-fold), while aspalathin ablated this response.Thus, our results infer that aspalathin protected cardiomyocytes exposed to chronic hyperglycemia against Faslg-induced apoptosis by activating the Igf-PI3K-Akt pro-survival pathway (Fig 2).Additionally, STRING data analysis showed that an interactive network was formed between Igf1, Akt1, serpin peptidase inhibitor, member 1/2 (Serpin1/2) and vascular endothelial growth factor A (Vegfa) (Fig 2).High glucose treatment resulted in a 2.0-fold increased expression of Vegfa, while reducing expression of Serpinb2 and Serpine1 by -2.7 and -46.1-fold, respectively (Fig 2 and Table 2).Similarly, Xue et al. [43] and Natarajan et al. [44] observed that chronic hyperglycemia upregulates Vegf and accelerates diabeticinduced cardiac fibrosis.They further speculate that the observed increase in high glucose-induced Vegf expression may be due to elevated inflammation consistent with a hyperglycemic state.The role of Serpine1 in cardiac fibrosis remains controversial however; Serpine1 deficient mice display cardiac fibrosis, enhanced inflammation, along with increased microvascular permeability and haemorrhage [45].This suggests that Serpine1 is cardio protective, and functions in maintaining normal microvasculature integrity.Hence, we would speculate that increased expression of Serpine1 might lead to cardio protection; however, this hypothesis requires further investigation as the regulation of Serpin 1 is controversial.

Protein kinases and mitochondrial function:
Damaged structural components of the heart, caused by insulin resistance, activate signaling pathways such as p38 mitogen-activated protein kinases (Mapk) [46].STRING network analysis identified three distinct nodes within this network, including dynamin 1-like (Dnm1l), Map2k1 and cAMP-dependent protein kinase catalytic subunit (Prkacb), that are involved in the process of mitophagy.Mitophagy is a process responsible for the removal of damaged mitochondria and defects in this process has been linked to cardiovascular dysfunction [47].Dnm1l is a mitochondrial protein that is involved in the apoptotic process and is crucial for maintaining mitochondrial function through increased Dnm1l-dependent mitophagy.Roy et al. [48] showed that mice lacking Dnm1l demonstrate impaired heart contraction with an associated reduction in mitochondrial fission, concluding that Dnm1l is critical for sustaining mitochondrial morphology and heart function.Interestingly, Dnm1l levels were increased 2.8-fold after aspalathin treatment.
Mitophagy induction can also be regulated/activated through the Prka-MTOR-ULK1 mediated signaling pathway [48].STRING analysis showed a strong interaction between protein kinase, AMPactivated, gamma 1 non-catalytic subunit (Prkag) and Mapk.Activation of Mapk is implicated in FFAinduced cardiotoxicity and subsequent cardiac failure [49].In this study we observed that high glucose resulted in the enhanced expression of Map2k1 (2.3-fold), with an associated decrease in the mitophagy activator Prkag1 expression (-2.6 fold).However, aspalathin treatment reversed this effect (Fig 2).Thus, we speculate that aspalathin might possibly protect the myocardium from chronic hyperglycemia through enhanced cAMP-mediated mitophagy, thereby alleviating FFA-induced MAPK toxicity and subsequent intracellular lipid accumulation in vitro.

In vitro effect of aspalathin on inflammation
In a diabetic state, a complex interplay between impaired cardiac substrate metabolism, insulin resistance and inflammation underlies the progression of the DCM.Furthermore, evidence is accumulating on the important role of inflammation in the development of cardiac hypertrophy and failure.Aberrant and/or prolonged suppression of cytokine signaling (Socs) proteins and JAKinduced signaling is detrimental and can give rise to a number of inflammatory pathologies known to affect cardiac function [50,51].Accordingly, 11 of the 57 (19%) differentially expressed genes were associated with cytokine signaling (Table 2).STRING analysis identified a network with five distinct nodes that include interleukins, Cd3e antigen (Cd3e), Socs3, and leptin receptor precursor (Lepr), with Janus kinase 2 (Jak2) being the predominant interactive node (Fig 3).

Leptin signaling:
Leptin is an adipokine predominantly expressed in adipose tissue, but it is also expressed in cardiac heart muscle cells [22].Accumulative evidence suggests that leptin impairs myocardial energy metabolism by favoring a complete reliance on FFAs as an energy source leading to cardiac hypertrophy [22,50].Leptin binding to the leptin receptor activates the JAK/Signal Transducers and Activators of Transcription (STAT) pathway.Activation of this pathway results in the translocation of STAT to the nucleus where it can activate Socs3, inhibiting leptin action, as well as insulin signaling [51].A meta-analysis of our data showed that Lepr expression was increased (7.2fold) after high glucose treatment.Interestingly, this increase was concomitant with an increase in the mRNA expression of both Jak2 (3.9-fold) and Socs3 (4.5-fold).This result is of interest, as increased Lepr expression is known to enhance Socs3 with an associated altered β-oxidation [51].
Cytokine signaling: In the diabetic heart, increased lipotoxicity and insulin resistance trigger the recruitment of macrophages and leukocytes before the release of pro-inflammatory cytokines such as interleukin 3 and 6 (Il3 and Il6), tumor necrosis factor superfamily (Tnf) and cluster of differentiation (Cd).Il6, Tnf and Cd cause the endothelial cells of blood vessels to express cellular adhesion molecules, such as Selectin E (Sele) and Cd44 antigen (Cd44), resulting in an acute localized cellular inflammation.Both Cd44 and Sele are cell-surface glycoproteins that mediate neutrophil, monocyte, lymphocyte and platelet rolling in the ventricular wall.These glycoproteins also play an important role in atherosclerotic lesion development and calcification of the lesion [52,53].Studies demonstrated that Cd44, as well as Sele null mice display atherosclerosis plaque formation [52,53].2).Based on our gene expression data, we propose that during chronic hyperglycemia, pro-inflammatory cytokines (Il3, Il6 and Tnf-α) are activated.This increased inflammatory cytokine response releases Sele and Cd44 from the storage granules in activated platelet and endothelial cells and recruits them to mediate the first step of leukocyte extravasation.Additionally, Il6 and Tnf activate the JAK/STAT and Map kinase pathways respectively, resulting in increased hypertrophy and apoptosis.Increased expression of Socs3 has been linked to inflammation-induced insulin resistance and enhanced JAK/STAT signaling.As observed in this study aspalathin treatment was able to reverse this effect (Fig 3) [54].

In vitro effect of aspalathin on apoptosis
Chronic hyperglycemia, insulin resistance and inflammation exacerbates oxidative stress and cardiomyocyte apoptosis [23,26].In this study, six of the 57 (11%) differentially expressed genes were associated with apoptosis signaling (Table 2).STRING data analysis identified one network with five interactive nodes, centralized around the B-cell lymphoma 2 gene (Bcl2).Genes identified in this network included Bcl2, Bcl2 binding component 3 (Bbc3), mitogen activated protein kinase 3 (Mapk3) and optic atrophy 1 homolog (Opa1) (Fig 4 and Table 2).The Bcl2 family of proteins are known to be strong modulators of apoptosis and can induce either proapoptosis or cell survival depending on the cell's fate [10,26].Two pro-survival genes reduced after high glucose exposure were identified (Bcl2 and Bcl2a1).Four pro-apoptotic proteins (Bbc3, Chuk, Mapk3 and Opa1) were identified (Fig 4 and Table 2).Activation of Mapk3 together with pro-apoptotic proteins such as Bbc3 has been associated with myocardial dysfunction [23,55], while upregulation of Opa1 has directly been linked to the reversal of mitochondrial apoptosis [23,55,56].Interestingly, aspalathin treatment at a high dose was able to ameliorate this effect by increasing the expression of survival genes.Thus, from this result we proposed that Bcl2 is the main regulatory network used by aspalathin to protect the myocardium against hyperglycemic-induced cell apoptosis.

Reagents and kits
Pure aspalathin (ca.98%, batch SZI-356-54) was synthesized by High Force Research (Durham, UK) according to a previously published protocol [17].H9c2 rat derived cardiomyoblasts were purchased from the European Collection of Cell Cultures (ECACC No. 8809294; Wiltshire, UK).Radioimmunoassay insulin kit was obtained from Linco Research (St.Charles, USA), halothane from Safeline Pharmaceuticals (Johannesburg, RSA), Dulbecco's Modified Eagle's Medium, penicillin, and streptomycin from Lonza (Verviers, Belgium), and fetal bovine serum sourced from Biochrom (Berlin, Germany).The RNeasy Mini Kit, RT 2 First Strand Kit, RT 2 Profiler PCR Arrays and RT 2 SYBR Green, qPCR Master Mix were obtained from Qiagen (Valencia, USA), while the Trizol reagent and Turbo DNase Kit were from ThermoFisher Scientific (Waltham, USA).All other consumables and reagents were purchased from Sigma-Aldrich Corp. (St.Louis, USA), unless otherwise specified.

Animal Ethics
Male C57BLKS/J homozygous Lepr db/db mice and their heterozygous leptin-receptor-deficient nondiabetic lean littermate controls Lepr db/+ were obtained from Jackson's Laboratories (Sacramento, USA) and housed at the Primate Unit and Delft Animal Centre (PUDAC) of the South African

Aspalathin treatment of mice
For this study, animal experiments were only performed to assess the effect of aspalathin on blood lipid profiles.H9c2 cells, as an isolated cell-based system, were used to directly test the effect of aspalathin on the myocardium without the influence of any variables that would be introduce when using an animal model.Following acclimatizing for one week, nine-week old Lepr db/db mice together with Lepr db/+ (n=6 per group) control mice were randomly divided into five groups.Mice were treated daily for six weeks through oral gavage with either a low (13 mg/kg/day) or high (130 mg/kg/day) aspalathin dose and compared to metformin (150 mg/kg/day).Treatment dosages were calculated based on a published study [8].Treatment groups included: i) Lepr db/+ untreated controls (Lepr db/+_UC ), ii) Lepr db/db untreated controls (Lepr db/db_UC ), iii) Lepr db/db treated with metformin (Lepr db/db_MET ), iv) Lepr db/db treated with low dose aspalathin (Lepr db/db_ASP_LD ) and v) Lepr db/db treated with high dose aspalathin (Lepr db/db_ASP_HD ).Aspalathin and metformin were dissolved in distilled water before orally administration at the same time (08:00-09:00) every day for six weeks, while untreated animals were given water in place of treatment.

Effect of aspalathin on homeostatic model assessment-insulin resistance (HOMA-IR)
Mice were fasted for 4 h before plasma glucose levels were measured on a weekly basis by tail pricks using an OneTouch Select handheld glucometer (LifeScan, Milpitas, USA).Fasting plasma insulin was determined using the Radioimmunoassay Kit, as per manufacturer's instruction.HOMA-IR was calculated using fasting plasma glucose levels (FPG) and insulin values, according to a previous described method [18].

Effect of aspalathin on lipid profiles
After the 6-week treatment period, 4 h fasted mice were weighed and anesthetized with halothane before blood was collected from the abdominal vena cava for subsequent lipid profile analysis.Blood was centrifuged at 4, 000 g at 4ºC for 15 min before the serum was removed and sent to PathCare Medical Diagnostic Laboratories (Cape Town, RSA) for total cholesterol, triglycerides, LDL, and high-density lipoprotein (HDL) analyses.

Cell culture and treatment of H9c2 cardiomyocytes with aspalathin for subsequent RT 2 PCR Profiler Array analysis
Embryonic ventricular rat heart-derived H9c2 cardiomyoblasts were cultured in supplemented Dulbecco's Modified Eagle's Medium (10% fetal bovine serum, 100 U/ml of penicillin, 100 µg/ml of streptomycin) for 48 h under standard tissue culture conditions (37°C in humidified air and 5% CO2).Confluent cells (60-80%) were seeded at a density of 2 x 10 4 in 6-well multi-plates.Thereafter, H9c2 cardiomyoblasts were differentiated into adult cardiomyocytes for 7 days using retinoic acid according to a previously described method [19].On day 8, H9c2 cells were exposed to 33 mM glucose for 48 h and then treated with or without aspalathin (1 µM) for an additional 6 h.Cells exposed to 5.5 mM glucose were used as a normal glucose control [10].

RT 2 -PCR array analysis of H9c2 cells treated with and without aspalathin
Total RNA from three biological experiments (each with three technical replicates) was extracted using Trizol reagent, according to a previously described protocol [10].RNA was purified using the RNeasy Mini Kit, while the Turbo DNase Kit was used to remove genomic DNA, as per manufacturer's instructions.RNA integrity was determined using an Agilent 2100 Bioanalyser (Agilent Technologies, Palo Alto, USA).RNA from three independent experiments were pooled and cDNA was synthesized from two microgram of pooled RNA using the RT 2 First Strand Kit, according to manufacturer's instructions.Rat Atherosclerosis (PARN-038ZA-2), Cytokine (PARN-011ZA-2), Fatty Acid Metabolism (PARN-007Z), and Insulin Resistance (PARN-156ZA-2) RT 2 Profiler PCR Arrays were used for mRNA profiling studies using ABI7500 (ThermoFisher Scientific, Waltham, USA).Analysis of PCR array data was done according to manufacturer's instructions, using a Microsoft Excel macros available from the manufacturer (http://pcrdataanalysis.sabiosciences.com/pcr/arrayanalysis.php).Each array contained five housekeeping genes (Actb, B2m, Gapdh, Gusb and Hsp90ab1) against which the sample data were normalized.The transcript level of each candidate gene was quantified according to the ΔΔCT method.Ct values > 35 were not included in the analysis and considered as negative.

Gene interaction and network analysis
Search Tool for the Retrieval of Interacting Genes/Proteins (STRING; http://string-db.org/)database was used for gene interaction and network analysis to represent information of known and anticipated gene interactions [20,21].

Statistical analysis
Data were expressed as the mean ± SEM. Results for in vitro experiments were expressed as the mean of 3 independent biological experiments with each experiment containing at least 3 technical replicates, unless otherwise stated.For in vivo experiments, each treatment group contained 6 mice.Statistical analysis was performed using GraphPad Prism software (GraphPad Software, Inc., La Jolla, USA).Comparisons between groups were performed using one-way multivariate ANOVA, followed by unpaired Student t-test, and a p-value of ≤ 0.05 was deemed as statistically significant.

Conclusions
Results obtained in this study lead us to propose that aspalathin could protect the myocardium against lipid toxicity in two ways.In the first instance, such treatment significantly improves the blood lipid profile to a far less damaging one therefore attenuating fuel substrate availability to the diabetic heart.It is likely that aspalathin mediates such effects by acting on targets in the liver, although this was not further investigated in the current study.Secondly, our data demonstrate that aspalathin also exerts direct, protective effects within the myocardium to protect the heart against excess fuel substrate availability.Here our data indicate that aspalathin may protect the myocardium against lipotoxicity and subsequent cell apoptosis by increasing the expression of Adipoq and subsequently decreasing the expression of Cd36 and Cpt1.Furthermore, aspalathin reduced lipid transport via Slc27a3/5, which together with reduced PPARγ and Srebf1 resulted in decreased total cholesterol and subsequent cell apoptosis.Moreover, we proposed that the reversal of lipid toxicity results in a decreased inflammatory response via the Il6/Jak/Stat pathway, which together with an observed increase in Bcl2 prevents myocardium cell apoptosis.Together, these findings propsed a probable mechanism by which aspalathin reverses metabolic abnormalities associated with the failing myocardium.However, this warrants further investigation.

Fig 1 .
Fig 1. Aspalathin prevented high glucose-induced impaired cardiac substrate metabolism by reducing the uptake and oxidation of free fatty acids.(A) Search Tool for the Retrieval of Interacting Genes (STRING) database confirmed a strong interaction between aspalathin treatment and genes associated with lipid transport, lipid and fatty acid metabolism, relevant to dysregulation of intracellular lipid accumulation and fatty acid oxidation.(B) Representative diagram of the proposed modulating regulatory mechanisms of aspalathin against increased lipid accumulation and oxidation.Adipoq: adiponectin, C1Q and collagen domain containing; Cd36: cluster of differentiation 36; Cpt1: carnitine palmitoyltransferase 1; Fabp3: fatty acid binding protein 3; FAO: fatty acid oxidation; FFAs: free fatty acids; Pparγ: peroxisome proliferator activated receptor gamma; Scd1: stearoyl-Coenzyme A desaturase 1; Srebf1/2: sterol regulatory element binding transcription factor 1/2; TAG: triacylglycerides.

Fig 3 .
Fig 3. Aspalathin prevented high glucose-induced inflammation.(A) Search Tool for the Retrieval of Interacting Genes (STRING) database confirmed a strong interaction between aspalathin treatment and genes associated with inflammation.(B) Representative diagram of the proposed protective mechanism of aspalathin against high glucose induced inflammation.Cd44: cluster of differentiation 44; Il3: interleukin 3; Il6: interleukin 6; Map2k1: mitogen-activated protein kinase kinase 1; Socs3: suppressor of cytokine signaling 3; Tnf: tumor necrosis factor; Vegfa: vascular endothelial growth factor A.

Fig 4 .
Fig 4. Aspalathin limited high glucose-induced apoptosis.(A) Search Tool for the Retrieval of Interacting Genes (STRING) database confirmed a strong interaction between genes associated with aspalathin treatment and regulation of high glucose induced apoptosis.

Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 November 2016 doi:10.20944/preprints201611.0094.v1
Peer-reviewed version available at Molecules 2017, 22, 219; doi:10.3390/molecules22020219MedicalResearch Council (SAMRC) in a controlled environment with a 12 h light/dark cycle in a temperature range of 23-25ºC (relative humidity: ∼50%).The mice received standard laboratory chow pellets (Afresh Vention, Cape Town, RSA) ad libitum and had free access to drinking water.The study was performed according to principles and guidelines of the South African Medical Research Council's Guidelines on Ethics for Medical Research: Use of Animals in Research and Training, 2004 (http://ww.mrc.ac.za/ethics/ethicsbook3.pdf) under the institutional ethical approval of the SAMRC (ECRA no.07/13) as well as Stellenbosch University Ethics Committee (SU-ACUM13-00021).