Efficacy of Terpenoid in Attenuating Aortic Atherosclerosis in Apolipoprotein-E Deficient Mice: A Meta-Analysis of Animal Studies

Background The apolipoprotein E knockout (ApoE -/-) mouse model is well established for the study of terpenoids in the prevention of atherosclerosis. Studies investigating the clinical benefit of terpenoids in humans are scarce. This systematic review and meta-analysis evaluated the effects of terpenoid administration on atherosclerotic lesion area in ApoE -/- mice. Methods A comprehensive literature search using PubMed, Embase, and the Cochrane Library databases was performed to identify studies that assessed the effects of terpenoids on atherosclerosis in ApoE -/- mice. The primary outcome was atherosclerotic lesion area, and study quality was estimated using SYRCLE's risk of bias tool. Results The meta-analysis included 25 studies. Overall, terpenoids significantly reduced atherosclerotic lesion area when compared to vehicle control (P<0.00001; SMD: -0.55; 95% CI: -0.72, -0.39). In terpenoid type and dose subgroup analyses, sesquiterpenoid (P=0.002; SMD -0.93; 95% CI: -1.52, -0.34), diterpenoid (P=0.01; SMD: -0.30; 95% CI: -0.54, -0.06), triterpenoid (P<0.00001; SMD: -0.66; 95% CI: -0.94, -0.39), tetraterpenoid (P<0.0001; SMD: -1.81; 95% CI: -2.70, -0.91), low dose (P=0.0001; SMD: -0.51; 95% CI: -0.76, -0.25), medium dose (P<0.0001; SMD: -0.48; 95% CI: -0.72, -0.24), and high dose (P=0.002; SMD: -1.07; 95% CI: -1.74, -0.40) significantly decreased atherosclerotic lesion area when compared to vehicle control. PROSPERO register number is CRD42019121176. Conclusion Sesquiterpenoid, diterpenoid, triterpenoid, and tetraterpenoid have potential as antiatherosclerotic agents with a wide range of doses. This systematic review provides a reference for research programs aimed at the development of terpenoid-based clinical drugs.


Introduction
Atherosclerosis is the main cause of cardiovascular disease, which is the leading cause of death globally [1,2]. Dyslipidemia and oxidative stress are relevant to the pathogenesis of atherosclerosis [3,4]. Therefore, statin-based lipid-modifying therapies, such as atorvastatin and rosuvastatin, are effective for lowering blood cholesterol levels and providing clinical benefits in patients with cardiovascular disease. However, the morbidity and mortality associated with atherosclerosis remain high [5], and there is an urgent unmet clinical need for novel prevention and treatment strategies [6].
In recent years, studies have shown that natural compounds, such as flavonoids, alkaloids, and terpenoids, attenuate atherosclerosis [7][8][9]. Terpenoids are a large and diverse class of naturally occurring organic chemicals that are similar to terpenes. Most terpenoids are multicyclic structures with oxygen-containing functional groups. Furthermore, terpenoids have a wide range of pharmacological effects, including antitumor, anti-inflammatory, antiatherosclerotic, and antimalarial activities [10][11][12]. The majority of studies on the antiatherosclerotic effects of terpenoids have focused on paclitaxel, [13,14] which is a natural diterpene, and consensus on the antiatherosclerotic effects of other terpenoids has not been reached.
Studies in animals allow for initial investigations on the safety and efficacy of new interventions and provide an important link between basic research and clinical trials. The apolipoprotein E knockout (ApoE -/-) mouse model spontaneously develops atherosclerotic plaques and is commonly used to mimic the pathophysiological process of atherosclerosis in humans [15,16]. The present systematic review and meta-analysis evaluated the effects of terpenoid administration on atherosclerotic lesion area in ApoE -/mice, in an effort to understand the clinical potential of terpenoids as antiatherosclerotic agents.

Materials and Methods
. . Reporting Standards. This systematic review complies with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The systematic review protocol was prepared using the SYRCLE format for animal intervention studies [17,18].
. . Search Strategy. An experienced information specialist (HL) searched the PubMed, Embase, and Cochrane Library databases from January 2001 to December 2018 using the keywords: "atherosclerosis," "atherogenesis," "apolipoproteins e," "apoe," "mice," and "terpenoid" and the following search strategies: (atherosclerosis OR atherogenesis) AND ("apolipoprotein * e" OR apoe) AND (mice OR mouse) AND (terpenoid OR hemiterpenoid OR monoterpenoid OR sesquiterpenoid OR diterpenoid OR sesterterpenoid OR triterpenoid OR tetraterpenoid OR polyterpenoid). The reference lists of included and review articles were manually searched to identify additional relevant studies. The search was performed on December 10, 2018, and was restricted to articles published in the English language.
. . Study Selection. Two reviewers (YZ and SS) independently examined the titles and abstracts of the articles identified by the literature search to select eligible studies. The full text of potentially relevant articles was retrieved and independently examined by two reviewers (YZ and SW) to determine whether these studies met the inclusion criteria. Disagreements on study selection were resolved by discussion and consensus.
. . Data Extraction. Two reviewers (HL and SW) independently extracted data from eligible studies, including the first author's name, publication year, age of mice, gender, diet, terpenoid dose, duration and route of treatment, control and treatment group sample sizes, location of the atherosclerotic lesion, stain used to assess the atherosclerotic lesion, and atherosclerotic lesion area. Data that were presented graphically in the original publications were extracted using Adobe Photoshop 7.0.
The primary outcome was atherosclerotic lesion area measured as a percentage or a numerical value.
Disagreements on data extraction were resolved by discussion and consensus.
. . Quality Assessment. Two investigators (SS and SW) independently assessed the quality of the included studies using SYRCLE's risk of bias tool, which contains domains evaluating sequence generation, baseline characteristics, allocation concealment, random housing, blinding, random and selective outcome assessments, incomplete outcomes data, and other sources of bias [44]. Publication bias was detected by visual inspection of funnel plots.
Disagreements on quality assessment were resolved by discussion and consensus.
. . Data Synthesis and Statistical Analysis. Statistical analyses were performed using Review Manager (RevMan Version 5.3 for Windows Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014). Standardized mean differences (SMD) with 95% confidence intervals (CI) were calculated to reflect the effects of terpenoids or vehicle control on atherosclerotic lesion area. A random-effects model was used to pool studies. Heterogeneity was determined as moderate (I ≥ 30%) or high (I ≥ 50%) using the inconsistency index.
Multiple independent groups in a study (e.g., different terpenoid doses) were considered separate datasets. In eleven studies [19,20,24,26,28,33,34,[38][39][40]42], multiple groups that tested different terpenoid doses were compared to a single control group. In order to avoid an artificial increase in sample size in the pooled analysis, the number of animals in the control group for each study was divided by the number of comparator groups.
Subanalyses were conducted to investigate the effects of sesquiterpenoid, diterpenoid, triterpenoid, and tetraterpenoid on the atherosclerotic lesion area.
Sensitivity analyses were conducted to determine whether the findings were robust. P<0.05 was considered statistically significant.
. . Study Characteristics. The characteristics of the 25 included studies are described in Table 1. These studies provided 59 datasets and involved 707 animals.
The duration of terpenoid treatment varied from four weeks to 24 weeks.
Terpenoid doses varied among different studies. It mainly ranged from 1 to 100 mg/kg/d. In addition, 500 and 2000 mg/kg/d were used in two studies [41,43].
. . Quality Assessment. Assessment of study quality is presented in Figure 2. A total of 19 (73.1%) studies were randomized, but the risks of bias due to allocation concealment and blinding were unclear. Sixteen studies had no missing outcomes data. The risk of selective outcomes reporting was unclear in nine studies. Across studies, the risk of bias from other sources was low.
Visual inspection of a funnel plot revealed substantial publication bias (Figure 3).

Discussion
The use of animal models provides a valuable approach to preclinical research, which informs treatment strategies for human diseases. Previous evidence from experiments in animals suggests that natural terpenoids have potential benefits for the treatment of atherosclerosis. However, parameters such as type and age of animal, sample size, housing conditions, and length of follow-up vary across studies. A synthesis and quantitative analysis of the data from animal models that accounts for these sources of heterogeneity may provide insight into the benefits of terpenoids as clinically desirable therapeutic agents in atherosclerosis. Therefore, we performed this systematic review and meta-analysis to evaluate the effects terpenoid administration on atherosclerotic lesion area in ApoE -/-mice. Findings showed that terpenoid administration significantly reduced aortic atherosclerosis lesion area compared to vehicle control.
To the authors' knowledge, this systematic review and meta-analysis is the first to evaluate the effects of terpenoids on atherosclerosis in ApoE -/-mice. Findings are expected to provide a scientific basis for clinical trials of terpenoids in cardiovascular diseases.
There was no heterogeneity between studies in the overall analysis, but there was a moderate degree of heterogeneity between studies in the analysis of sesquiterpenoid and high dose group. Potential sources of heterogeneity include age  and sex of mice and diet administered, each of which can influence the progression of atherosclerosis [1,61]. Furthermore, method of measurement can affect the assessment. Ten studies reported cross-sectional aortic lesion area, five studies reported longitudinal aortic lesion area, and ten studies reported both cross-sectional and longitudinal aortic lesion area.
Study Limitations. This meta-analysis was associated with several limitations. First, the relevance of our findings to humans is limited by species specific differences in lipoprotein metabolism and vascular physiology [62]. The ApoE -/-mouse model is well established for studying atherosclerosis, and the principal characteristics and progression of atherosclerosis in ApoE -/-mice and human subjects appear similar [15,16]; however, there are differences in pathogenesis. Specifically, the location of the atherosclerotic plaque may differ due to variations in heart rate, blood pressure, and hemodynamics. Atherosclerotic plaque builds up in the root of the aorta and in the brachiocephalic artery in ApoE -/-mice and the coronary artery, the carotid artery, the iliac artery, and the arteries of the lower limb in humans. Second, age is a risk factor in progression of atherosclerosis; therefore, due to differences in life cycle, the natural history of atherosclerotic disease in mice cannot be directly translated to humans. Third, patients usually present to the clinic with advanced atherosclerosis, but the included studies administered terpenoids to mice before disease had progressed and were therefore evaluating early prevention rather than benefit of treatment in advanced disease. Fourth, other animal models of atherosclerosis, such as LDLR -/-mice, rabbits, and hamsters, were not included in this meta-analysis. Fifth, only one parameter, atherosclerotic lesion area, was used to evaluate the effects of terpenoids on atherosclerosis. Other parameters, such as low density lipoprotein (LDL), high density lipoprotein (HDL), total cholesterol (TC), triglycerides (TG), and body mass index (BMI), were not considered. [63] Sixth, the sample sizes of some included studies were relatively small. Seventh, hemiterpenoid, monoterpenoid, and polyterpenoid were not assessed in the analysis. Further investigations in animal models using larger sample sizes are warranted to determine if terpenoids are beneficial for the treatment of atherosclerosis in humans.
The present meta-analysis revealed that terpenoid administration is effective for attenuating aortic atherosclerosis in ApoE-/-mice. In particular, sesquiterpenoid, diterpenoid, triterpenoid, and tetraterpenoid have a potential therapeutic effect with a wide range of doses. Large scale, prospective, and well-designed animal studies are needed to enhance our knowledge of the mechanism of terpenoids for the treatment of atherosclerosis. Randomized controlled trails in humans are required to confirm that terpenoids have clinical benefit as antiatherosclerotic agents.

Data Availability
The data used and/or analyzed during the current study are available from the corresponding author on reasonable request.