Atorvastatin and Fenofibrate Exert Opposite Effects on the Vascularization and Characteristics of Visceral Adipose Tissue in New Zealand White Rabbits

Abstract

Statins may precipitate the onset of type 2 diabetes (T2D) in high-risk patients. In contrast, only the subset of individuals with insulin resistance and/or diabetes receives cardiovascular benefits with fibrates. In this context, previous observations from our laboratory suggested that atorvastatin induced an increase in visceral adipose tissue (VAT), whereas fenofibrate had the opposite effects in rabbits. Therefore, we determined the mass, morphology, and vascularization of VAT in New Zealand white rabbits (n = 6/group) that received 0.33 or 2.6 mg/kg/d of atorvastatin or fenofibrate, respectively, during 2 months. As expected, the cholesterol from the atorvastatin group was lower after treatment, while triglycerides decreased in the fenofibrate group. The mass of VAT from the fenofibrate group was 46% lower compared to the controls, meanwhile atorvastatin was associated with a larger diameter of adipocytes (+65%) than that of the control and fenofibrate groups. Fibroblast growth factor 2 (FGF2) gene expression was lower in the fenofibrate group than in the control group (−54%). By contrast, vascular endothelial growth factor A (VEGF-A) gene expression in fenofibrate-treated rabbits was 110% higher than in the control group. In agreement with the gene expression, the marker of angiogenesis platelet endothelial cell adhesion molecule 1 was slightly but significantly higher (+10%) in rabbits treated with fenofibrate than in controls, as determined by immunohistochemistry. These results suggest that fenofibrate is associated with a favorable remodeling of VAT, that is, reduced mass and increased vascularization in normolipemic rabbits; in contrast, atorvastatin induced a nonfavorable remodeling of VAT. These results may be related to the cardiovascular benefits of fenofibrate and the increased risk of T2D in high-risk patients induced by atorvastatin.

Keywords

hypoxia inflammation vasculogenesis dyslipidemia statins fibrates

Introduction

Visceral adipose tissue (VAT) maintains the control of energy balance by storing and mobilizing triglycerides. In addition, VAT secretes proteins such as pro- and anti-inflammatory cytokines, hormones, and growth factors that regulate physiological and pathological processes and participate in obesity-related diseases.¹ It is widely accepted that the excess of VAT is associated with dysfunctional lipid metabolism, systemic inflammation, and increase in the risk of insulin resistance (IR) and type 2 diabetes (T2D).^1,2

When the size of the adipocytes increases due to enhanced energy storage, the secretory pattern of adipose tissue is modified toward pro-inflammatory and pro-IR states.^2,3 In contrast, when the size increase of adipose tissue is accompanied by an appropriate vascularization, the metabolic complications are restrained.⁴ In this context, unpublished data from our laboratory showed that atorvastatin, at doses similar to that commonly prescribed in humans, was associated with an increased mass of VAT in rabbits. Conversely, this effect was not observed in rabbits treated with fenofibrate. Furthermore, previous studies have demonstrated that statins are able to decrease the plasma levels of vascular endothelial growth factor A (VEGF-A), one of the main angiogenic factors,⁵ while fibrates increase vascularization in cornea and retina cells.⁶ Consequently, it is likely that fibrates and statins have opposite effects on adipose tissue size and vascularization.

Therefore, the aim of this study was to characterize the VAT of New Zealand white rabbits treated with atorvastatin or fenofibrate. Besides the histological characteristics of VAT, we quantified the messenger RNA (mRNA) expression of 4 of the main genes involved in vascularization, VEGF-A, fibroblast growth factor 2 (FGF2), transforming growth factor-β (TGF-β), and tumor necrosis factor α (TNF-α), after the treatment with atorvastatin or fenofibrate. Our data suggested that fenofibrate was associated with a higher vascularization of VAT in normolipemic rabbits. The immunohistochemistry of platelet endothelial cell adhesion molecule 1 (PECAM-1) protein as endothelial biomarker further supported the possibility of an increased vascularity in VAT in rabbits treated with fenofibrate.

Methods

All the experiments were conducted according to the “Guide for the Care and Use of Laboratory Animals” (National Academy Press, Washington, DC, 2011) and were approved by the Instituto Nacional de Cardiología “Ignacio Chávez” Research Committee and Institutional Animal Care and Use Committee, No. 12-764-Atorvastatina. Atorvastatin and fenofibrate were a kind gift of Senosiain Laboratories, Mexico.

Animals

New Zealand male rabbits 2.3 to 3 kg received during 8 weeks a daily oral dose of 0.33 mg/kg of atorvastatin (n = 6) or 2.6 mg/kg of fenofibrate (n = 6), whereas the control group received only the vehicle as described previously.⁷ All animals received standard chow diet and ad libitum water and were maintained under a 12–12 hour light–dark cycle. Animal procedures were performed in accordance with the Scientific and Ethics Committee of the Instituto Nacional de Cardiología “Ignacio Chávez”

Biochemical Analysis

Ten-hour fasting blood samples were obtained from the marginal ear vein at the beginning and at the end of the treatment. Plasma was separated and used immediately to determine the biochemical parameters (cholesterol, high-density lipoprotein [HDL] cholesterol, triglycerides, and glucose) by enzymatic colorimetric methods (Randox Laboratories, Antrim, United Kingdom).

Measurement of Visceral Fat Mass

At the end of the treatment period, animals were euthanized by cervical dislocation, and the retroperitoneal VAT was carefully dissected by means of surgical instruments, gently rinsed with isotonic saline solution, wiped, and weighed in an analytical balance. Fragments of about 1 cm³ of adipose tissue were stored at −70°C in RNAlater Stabilization Solution (Life technologies, Carlsbad, California) or in 10% formaldehyde until use.

Quantitative Real-Time Polymerase Chain Reaction

Total RNA from 500 mg of VAT was isolated using Ribozol Plus RNA Purification Kit (Ameresco, Ohio), according to the manufacturer’s instructions. Isolated RNAs were incubated with RNase-free DNAse for 30 minutes at 37°C (Invitrogen, Life Technologies, Carlsbad, California). The enzyme was inactivated adding 25 mmol/L EDTA and heated at 75°C for 10 minutes. First-strand complementary DNA (cDNA) was synthesized using Superscript VILO cDNA Synthesis (Invitrogen, Life Technologies) and 500 ng of the DNA-free RNA according to the protocol provided by the manufacturer. Real-time quantitative polymerase chain reaction (qPCR) was performed in ABI 7300 Real-Time PCR System using 60 ng per assay. The TaqMan probes used were Oc03396227 (FGF2), Oc04176122 (TGF-β), Oc03397716 (TNF-α), and Oc03395999 (VEGFA). Gene expression level was calculated by standard curve method and normalized to GAPDH.

Histological and Immunohistochemical Analyses

Visceral adipose tissue biopsies were loaded into embedding cassettes and were fixed in 10% formalin solution for 24 hours. Afterward, the samples were dehydrated in a graded series of ethanol (80%, 96%, and 100%) and xylol and were embedded in paraffin. Three-micrometer paraffin-embedded sections were cut and placed on charged slides. The sections were dewaxed for 1 hour at 60°C and were soaked in xylol for 3 minutes. After xylol treatment, the biopsies were rinsed in ethanol 100%, 96%, and 80% for 3 minutes each. The tissue sections were then used for immunohistochemistry or stained.

The adipocyte size was determined by staining the slides with Harris hematoxylin for 5 minutes, decolorized in acid alcohol for 1 minute, and finally counterstained with eosin Y for 2 minutes. Slides were dehydrated with ethanol to different concentrations (80%, 96%, and 100%) and xylol for 3 minutes each and were mounted using a synthetic resin. Three noncontiguous slides by rabbit (n = 3) were randomly chosen for microscopic analyses; reported size corresponds to the mean of 180 to 200 cells by group.

For immunohistochemistry of PECAM-1, heat-induced antigen retrieval in slides was performed with 1 mmol/L EDTA buffer at pH 8.0 at 125°C for 10 seconds. Endogenous peroxidase activity was inhibited with 0.3% H₂O₂–methanol solution for 30 minutes.

The slides were incubated with anti-CD31 (PECAM-1) antibody for 24 hours at 4°C (1:200, ab199012; Abcam plc, Cambridge, Massachusetts) and revealed with an HRP-labeled goat antimouse immunoglobulin G (1:500, ab255719; Abcam plc) for 30 minutes using 3,3-diaminobenzidine tetrahydrochloride.

Three noncontiguous slides by rabbit (n = 3) were randomly chosen for microscopic analyses. Four microscopic fields per slide were acquired with Axio Scan.Z1 scanner (Zeiss, Germany), and PECAM-1 quantification was performed with ImageJ, IHC Profiler (NIH) program.

Statistical Analysis

Normally distributed variables determined with a Kolmogorov-Smirnov test were expressed as means ± standard error, and the comparisons among groups were performed using 1-way analysis of variance test. Other variables were compared by the Kruskal-Wallis test, and data were expressed as median and interquartile range. Significance level was considered when statistical test reached a P value <.05. Statistical analyses were performed with the Statistical Package for the Social Sciences (SPSS) version 21.0 (IBM, Armonk, New York)

Results

Adipose Tissue Mass and Adipocyte Size Determination

Table 1 shows the biochemical parameters before and at the end of the treatments. In agreement with the principal effects of atorvastatin and fenofibrate, plasma concentrations of total cholesterol decreased in atorvastatin-treated rabbits (−51%), while triglycerides levels decreased by 30% in the fenofibrate-treated group, both compared to basal conditions. Glucose and HDL-cholesterol remained unchanged after any of the 2 treatments (Table 1).

Table 1.

Plasma Biochemical Profiles of Experimental Animals.^a

mg/dL	Control, n = 6		Atorvastatin, n = 6		Fenofibrate, n = 6
mg/dL	Basal	Final	Basal	Final	Basal	Final
Cholesterol	48.93 (37.80-54.79)	45.83 (39.14-51.09)	53.63 (33.50-59.96)	25.93^b (22.04-38.20)	42.46 (39.60-50.73)	35.61 (20.24-48.08)
Triglycerides	62.20 (52.70-72.48)	61.81 (47.66-77.35)	66.44 (56.72-76.47)	74.68 (64.13-80.05)	72.11 (45.68-81.73)	49.93^b (33.21-67.09)
Glucose	114.46 (103.61-122.57)	112.86 (97.49-119.10)	115.26 (104.85-124.35)	109.45 (104.55-117.34)	118.52 (112.41-119.59)	112.82 (108.32-129.55)
HDL-cholesterol	28.73 (24.52-36.01)	23.54 (19.40-44.03)	31.05 (19.29-33.35)	24.59 (15.09-36.42)	28.08 (21.09-35.10)	19.89 (13.97-45.97)

Abbreviation: HDL, high-density lipoprotein

^aResults are expressed as median and interquartile range. Kruskal-Wallis test.

^b P < .05 versus basal conditions.

The weight gain during the 6 weeks of treatment was similar among control, atorvastatin, and fenofibrate groups (median [interquartile interval], 1.065 [0.615-1.211] kg, 1.063 [0.586-1.253] kg, and 0.900 [0.597-1.168] kg, respectively; P = .741). Even if any of the 2 drugs did not affect the weight gain of rabbits, the fenofibrate group had a lower VAT mass (−46%) than the control group at the end of the treatment (Figure 1). In contrast, the total mass of adipose tissue was comparable between atorvastatin and control groups. Since this observation could be associated with changes in the characteristics of the adipocytes, we analyzed the microscopic characteristics of visceral adipocytes. Although the total mass of adipose tissue was comparable between atorvastatin and control groups as mentioned earlier, the size of adipocytes was significantly higher (65%) in the former than in the latter (Figure 2).

Figure 1.

Measurement of visceral adipose tissue. Results are expressed as median and interquartile range. *P < .05 versus control group. N = 6 rabbits per group.

Figure 2.

Hematoxylin and eosin staining of visceral adipose tissue. A, Control group. B, Atorvastatin group. C, Fenofibrate group. Scale: 20 µm. D, Adipocyte size of visceral adipose tissue (µm). Results are expressed as mean ± standard error (SE). *P < .05 versus control group, **P < .05 versus fenofibrate group. n = 3 rabbits per group; 4 microscope fields were analyzed per slide, 3 slides per rabbit.

Expression of Genes Related to Angiogenesis in VAT

Results of qPCR (Figure 3) showed significantly higher levels of VEGF-A gene expression in the fenofibrate group than in the control group (110%). By contrast, expression of FGF2 was lower (−54%) in the fenofibrate group than in the control group (Figure 3).

Figure 3.

Gene expression in visceral adipose tissue. Results are expressed as median and interquartile range. *P < .05 versus control group. FGF2 indicates fibroblast growth factor 2; TGF-β, transforming growth factor-β; TNF-α, tumor necrosis factor α; VEGF-A, vascular endothelial growth factor A. n = 6 rabbits per group.

Compared to the control group, TGF-β had a tendency to higher levels of expression in VAT of the rabbits that received fenofibrate, whereas TNF-α tended to have lower levels in the atorvastatin group (Figure 3), but these differences did not reach statistical significance.

Quantitation of PECAM-1 as Marker of Vascularization in VAT

The increase in VEGF-A expression observed in the VAT of rabbits treated with fenofibrate suggested an enhanced angiogenic process. In order to explore this possibility, we performed an immunohistochemistry of PECAM-1 protein as endothelial biomarker in VAT obtained from rabbits treated with either atorvastatin or fenofibrate, and the results are shown in Figure 4. Rabbits that received fenofibrate showed slightly but significantly higher levels (10%) of PECAM-1 protein than controls, while rabbits treated with atorvastatin showed similar levels of this protein with respect to the control group (Figure 4).

Figure 4.

Immunohistochemistry of platelet endothelial cell adhesion molecule 1 (PECAM-1) protein in visceral adipose tissue. A, Control group. B, Atorvastatin group. C, Fenofibrate group. Scale: 20 µm. D, Quantification of positive level of PECAM-1. Results are expressed as mean ± standard error (SE). *P < .05 versus control group. n = 3 rabbits per group; 4 microscope fields were analyzed per rabbit. Arrows indicate the positive PECAM-1 zones (vessels) in the microphotographs.

Discussion

In this study, we demonstrated that fenofibrate induced a decrease in VAT mass, concomitantly with an increased VEGF-A gene expression associated with a discrete but significant augmentation of endothelial marker PECAM-1. By contrast, the treatment with atorvastatin led to an increase in the adipocyte size without a significant higher mass of total VAT. Furthermore, the augmented VEGF-A gene expression and endothelial marker PECAM-1 suggested that fenofibrate is able to induce vascularization on VAT.

Among the common animal models, we selected the rabbit because this species has been successfully used for lipid and lipoprotein studies since its metabolism is similar to that of humans.^7
–9 Some of the beneficial effects of statins and fibrates on different tissues have been attributed to their lipid-lowering properties^10
–12; the improvement in lipid profile, in its turn, may induce metabolic changes in several tissues including VAT. We used daily doses of atorvastatin and fenofibrate within the range of the commonly prescribed to patients. We have previously demonstrated that the doses used in this study are effective for lipid lowering and to induce important modification in HDL structure in dyslipidemic rabbits.⁷ By using these relative low amounts of atorvastatin or fenofibrate, particularly for experimental animals, and since we included only normolipemic animals, the lowering effects of the drugs were subtle (about 27 and 22 mg/dL decrease in cholesterol and triglycerides, respectively). Under these conditions, the observed changes in size and vascularization of VAT were mainly independent of plasma lipid level variations.

Our data demonstrated the presence of larger adipocytes in atorvastatin-treated rabbits than in the control group; concomitantly, the total VAT mass did not increase in rabbits that received atorvastatin when compared to controls. Together, these data suggest a different remodeling of VAT induced by atorvastatin, that is, larger cell size and lower number of constitutive cells. Accordingly, epicardial adipose tissue thickness was reduced in patients with T2D and post-menopausal women treated with atorvastatin, independently of lipid lowering.^13
–15

Visceral adipose tissue plays a central role in IR and T2D particularly due to adipocyte dysfunction, that is, decreased secretion of adiponectin and increased resistin secretion among other alterations.¹ It is accepted that adipocytes gradually become dysfunctional as a consequence of their size increase.² In this context, the size increase of adipocytes observed only after 4 weeks of atorvastatin treatment may contribute for a future adipocyte dysfunction. This idea is in agreement with recent reports that have associated statins intake with accelerated onset of T2D in high-risk patients.^16

–21 Hypoxia is one of the main factors that drive VAT to inflammation and cell dysfunction when adipocytes become larger²²; consequently, vascularization may counterbalance some of the noxious effects of cell size increase. Our data showed that atorvastatin does not enhance VAT gene expression of VEGF-A whose protein product is considered the main mediator of angiogenesis.²³ Accordingly, PECAM-1, a vital component of the endothelial cell intercellular junction, detected in VAT from atorvastatin-treated rabbits was similar to that of the control rabbits, indicating that vascularization remained unchanged after treatment. Taken together, these results suggest a decreased oxygen supply to VAT. In this context, previous reports indicate that hypoxia induce VAT dysfunction⁴; in its turn, VAT hypoxia is congruent with the early onset of T2D induced by statins in patients at high risk of diabetes.^16

–19 Further studies are needed to explore the potential statin-induced VAT dysfunction and its potential relationship with the incidence of T2D.

In contrast with atorvastatin, fenofibrate had positive effects on VAT, that is, total mass reduction, about twice the VEGF-A gene expression and a slight but significant increase in the endothelial marker, PECAM-1. Thus, our results demonstrated that fenofibrate enhances vascularization of VAT, possibly mediated by angiogenesis independent of decreased expression of FGF2.

Fenofibrate also induced lower adipose tissue mass without any significant change in either adipocyte size or body mass. These results are congruent with previous reports and demonstrated that a chronic treatment with fenofibrate is able to partially restrict the development of adiposity in mouse.²⁴ Concerning the vascularization of VAT in rabbits treated with fenofibrate, a more vascularized VAT may be related to a less dysfunctional tissue.⁴ Accordingly, fibrates reduce cardiovascular risk specifically in the subset of patients with IR and T2D.^25,26 Therefore, it is likely that such benefits could be related, at least in part, to an improvement in VAT functionality. More studies in humans are needed to validate this hypothesis.

The formation of a mature vascular network requires precise spatial and temporal regulation of a large number of angiogenic factors²⁷ including VEGF-A and FGF2; particularly, FGF2 is a mitogen that induces endothelial cell proliferation, migration, and pericyte attraction.^28,29 For this reason, we also analyzed the expression of FGF2 gene; the expression of this angiogenic factor was lower in VAT from rabbits treated with fenofibrate than in control rabbits. Previous reports demonstrated that exogenous FGF2 stimulates migration and proliferation of endothelial cells in vivo³⁰ and induces the development of large collateral vessels with adventitia.³¹ However, as overexpression of FGF2 does not lead to vascular defects,³² and normal vascularization is retained in KO mice,³³ its physiological relevance is still uncertain. Moreover, it should be emphasized that FGF2 gene expression was attenuated but not suppressed by fenofibrate, suggesting that protein is present in VAT to promote angiogenesis initiated by VEGF. In addition, other molecules could counterbalance a potential reduction in FGF2, since there is a high level of compensation among the growth factors mediating angiogenesis.^34,35

We recognize as a limitation the short term of this study, and we cannot rule out a long-term compensation of VAT, particularly vis-a-vis the nonfavorable effects of atorvastatin. However, if the acute proangiogenic effects of fenofibrate occur in other tissues, such acute effects could be of particular interest during myocardial infarction to limit the extent of the damaged tissue by inducing collateral vascularization.³⁵

Summary

Our data showed that atorvastatin and fenofibrate had opposite effects on rabbit VAT at ponderal doses similar to those used in humans; fenofibrate induced an increase in vascularization and a decrease in VAT mass, while atorvastatin promotes higher adipocyte size without any effect on the vascularization. Finally, it is necessary to sustain these results in human studies to reevaluate the use of both drugs in medical practice.

Footnotes

Authors’ Note

This work was done at the Instituto Nacional de Cardiología “Ignacio Chávez”, Mexico City, Mexico.

Acknowledgments

The authors are grateful to Rodrigo Velázquez Espejel and Armando Medina-Cruz for their technical support.

Author Contributions

Andrea Mondragón-García and María Luna-Luna contributed equally to this study. Andrea Mondragón-García contributed to design, contributed to acquisition, drafted the manuscript, and critically revised the manuscript. María Luna-Luna contributed to conception and design, contributed to acquisition, analysis, and interpretation, drafted the manuscript, and critically revised the manuscript. Cristobal Flores-Castillo contributed to acquisition and drafted the manuscript. Alberto Aranda-Fraustro and Elizabeth Carreón-Torres contributed to acquisition and critically revised the manuscript. Victoria López-Olmos and José Manuel Fragoso contributed to analysis and critically revised the manuscript. Gilberto Vargas-Alarcón and Óscar Pérez-Méndez contributed to interpretation and critically revised the manuscript. All authors gave final approval and agree to be accountable for all aspects of work ensuring integrity and accuracy.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by a CONACYT grant No. 233493. María Luna-Luna is a doctoral student from Programa de Doctorado en Ciencias Biomédicas de la Universidad Nacional Autónoma de México and received a fellowship from CONACYT, No. 408097.

ORCID iD

Cristobal Flores-Castillo

References

Luna-Luna

Medina-Urrutia

Vargas-Alarcón

Coss-Rovirosa

Vargas-Barrón

Pérez-Méndez

. Adipose tissue in metabolic syndrome: onset and progression of atherosclerosis. Arch Med Res. 2015;46(5):392–407.

Lafontan

. Adipose tissue and adipocyte dysregulation. Diabetes Metab. 2014;40(1):16–28.

Blüer

. Adipose tissue inflammation: a cause or consequence of obesity-related insulin resistance? Clin Sci. 2016;130(18):1603–1614.

Sun

Kusminski

Scherer

. Adipose tissue remodeling and obesity. J Clin Invest. 2011;121(6):2094–2101.

Sahebkar

Ponziani

Goitre

. Does statin therapy reduce plasma VEGF levels in humans? A systematic review and meta-analysis of randomized controlled trials. Metabolism. 2015;64(11):1466–1476.

Noonan

Jenkins

Keech

Wang

Lamoureux

. An update on the molecular actions of fenofibrate and its clinical effects on diabetic retinopathy and other microvascular end points in patients with diabetes. Diabetes. 2013;62(12):3968–3975.

Flores-Castillo

Zamora-Pérez

Carreón-Torres

, et al. Atorvastatin and fenofibrate combination induces the predominance of the large HDL subclasses and increased apo AI fractional catabolic rates in New Zealand white rabbits with exogenous hypercholesterolemia. Fundam Clin Pharmacol. 2015;29(4):362–370.

Martínez-Ramírez

Flores-Castillo

Sánchez-Lozada

, et al. Hyperuricemia is associated with increased apo AI fractional catabolic rates and dysfunctional HDL in New Zealand rabbits. Lipids. 2017;52(12):999–1006.

López-Olmos

Carreón-Torres

Luna-Luna

, et al. Increased HDL size and enhanced apo A-I catabolic rates are associated with doxorubicin-induced proteinuria in New Zealand white rabbits. Lipids. 2016;51(3):311–320.

10.

Egom

Hafeez

. Biochemistry of statins. Adv Clin Chem. 2016;73:127–168.

11.

Erol

. The role of fat tissue in the cholesterol lowering and the pleiotropic effects of statins—statins activate the generation of metabolically more capable adipocytes. Med Hypotheses. 2005;64(1):69–73.

12.

Staels

Dallongeville

Auwerx

Schoonjans

Leitersdorf

Fruchart

. Mechanism of action of fibrates on lipid and lipoprotein metabolism. Circulation. 1998;98(19):2088–2093.

13.

Park

J-H

Park

Kim

, et al. Effects of statins on the epicardial fat thickness in patients with coronary artery stenosis underwent percutaneous coronary intervention: comparison of atorvastatin with simvastatin/ezetimibe. J Cardiovasc Ultrasound. 2010;18(4):121–126.

14.

Alexopoulos

Melek

Arepalli

, et al. Effect of intensive versus moderate lipid-lowering therapy on epicardial adipose tissue in hyperlipidemic post-menopausal women: a substudy of the BELLES trial (Beyond Endorsed Lipid Lowering with EBT Scanning). J Am Coll Cardiol. 2013;61(19):1956–1961.

15.

González

Moreno-Villegas

González-Bris

Egido

Lorenzo

. Regulation of visceral and epicardial adipose tissue for preventing cardiovascular injuries associated to obesity and diabetes. Cardiovasc Diabetol. 2017;16(1):44.

16.

Sattar

Preiss

Murray

, et al. Statins and risk of incident diabetes: a collaborative meta-analysis of randomised statin trials. Lancet. 2010;375(9716):735–742.

17.

Mills

Chong

, et al. Efficacy and safety of statin treatment for cardiovascular disease: a network meta-analysis of 170,255 patients from 76 randomized trials. QJM. 2011;104(2):109–124.

18.

Culver

Ockene

Balasubramanian

, et al. Statin use and risk of diabetes mellitus in postmenopausal women in the Women’s Health Initiative. Arch Intern Med. 2012;172(2):144–152.

19.

Carter

Gomes

Camacho

Juurlink

Shah

Mamdani

. Risk of incident diabetes among patients treated with statin: population based study. BMJ. 2013;346:f2610.

20.

Wang

Liu

Chao

, et al. Statins, risk of diabetes, and implications on outcomes in the general population. J Am Coll Cardiol. 2012;60(14):1231–1238.

21.

Macedo

Douglas

Smeeth

Forbes

Ebrahim

. Statins and the risk of type 2 diabetes mellitus: cohort study using the UK clinical practice research datalink. BMC Cardiovasc Disord. 2014;14:85.

22.

Hosogai

Fukuhara

Oshima

, et al. Adipose tissue hypoxia in obesity and its impact on adipocytokine dysregulation. Diabetes. 2007;56(4):901–911.

23.

Zachary

. VEGF signaling: integration and multi-tasking in endothelial cell biology. Biochem Soc Trans. 2003;31(Pt 6):1171–1177.

24.

Guerre-Millo

Gervois

Raspé

, et al. Peroxisome proliferator-activated receptor alpha activators improve insulin sensitivity and reduce adiposity. J Biol Chem. 2000;275(22):16638–16642.

25.

Halcox

Misra

. Type 2 diabetes mellitus, metabolic syndrome, and mixed dyslipidemia: how similar, how different, and how to treat? Metab Syndr Relat Disord. 2015;13(1):1–21.

26.

Tenenbaum

Fisman

. Fibrates are an essential part of modern anti-dyslipidemia arsenal: spotlight on atherogenic dyslipidemia and residual risk reduction. Cardiovasc Diabetol. 2012;11:125.

27.

Bai

Zhou

Chen

Zhang

. Sequential delivery of VEGF, FGF-2 and PDGF from the polymeric system enhance HUVECs angiogenesis in vitro and CAM angiogenesis. Cell Immunol. 2018;323:19–32.

28.

Distler

Hirth

Kurowska

Gay

Distler

. Angiogenic and angiostatic factors in the molecular control of angiogenesis. Q J Nucl Med. 2003;47(3):149–161.

29.

Ruel

Laham

Parker

, et al. Long-term effects of surgical angiogenic therapy with fibroblast growth factor 2 protein. J Thorac Cardiovasc Surg. 2002;124(1):28–34.

30.

Ware

Simons

. Angiogenesis in ischemic heart disease. Nat Med. 1997;3(2):158–164.

31.

Scholz

Cai

Schaper

. Arteriogenesis, a new concept of vascular adaptation in occlusive disease. Angiogenesis. 2001;4(4):247–257.

32.

Fulgham

Widhalm

Martin

Coffin

. FGF-2 dependent angiogenesis is a latent phenotype in basic fibroblast growth factor transgenic mice. Endothelium. 1999;6(3):185–195.

33.

Miller

Ortega

Bashayan

Basch

Basilico

. Compensation by fibroblast growth factor 1 (FGF1) does not account for the mild phenotypic defects observed in FGF2 null mice. Mol Cell Biol. 2000;20(6):2260–2268.

34.

Khurana

Simons

. Insights from angiogenesis trials using fibroblast growth factor for advanced arteriosclerotic disease. Trends Cardiovasc Med. 2003;13(3):116–122.

35.

Beenken

Mohammadi

. The FGF family: biology, pathophysiology and therapy. Nat Rev Drug Discov. 2009;8(3):235–253.