Diabetic kidney disease (DKD) is a severe and common complications in diabetic patients, it brings serious economic burden on society both in Western and Eastern countries.1 Recent studies indicated that chronic kidney disease (CKD) induced by diabetes was more common than primary glomerulonephritis in China.2 It has been demonstrated that albuminuria, elevated blood pressure, metabolic abnormalities, excessive oxidative stress and mitochondrial dysfunction were vital pathogenic factors in DKD.3,4 Unfortunately, the detailed pathogenesis of DKD is still not fully understood, and the mainstay of current treatment for DKD including controlling blood glucose and blood pressure are not fully effective. Hence a better understanding of the DKD pathogenesis is urgently needed.

Recent studies showed that genetic factors damage was involved in the onset of DKD.5 Additionally, the susceptibility of DKD was associated with some single genes polymorphism (e.g. methylenetetrahydrofolate reductase and angiotensin converting enzyme).6 Angiotensin converting enzyme (ACE) gene contained 21kb base, it was located on 17q23 including 26 exons and 25 introns. Single nucleotide polymorphisms (SNPs) frequently occurs in the ACE gene, it has been identified 6 polymorphism markers of ACE, and Alu insertion/deletion (I/D) fragment in the 16th intron is the most investigated, ACE gene polymorphism could be divided into DD, ID, II genotype based on this I/D polymorphic marker locus.7 Some previous studies has found that ACE I/D polymorphism could influence the occurrence of diabetes-related renal damage.8 In addition, some pooled analysis as regards the impacting of ACE I/D gene polymorphism on DKD susceptibility has been completed.9,10 However, the pooled results were controversial and inconsistent. In this study, we further assess the potential impact of ACE I/D gene polymorphism on DKD through analyzing much more trials.

This pooled-analysis showed that the ACE I/D polymorphism was statistically associated with DKD susceptibility, it indicated that ACE D allele/DD genotype might be a risk factor for DKD. On the contrary, ACE II genotype might be a protective factor for DKD.

Genome-wide association studies (GWAS) research was frequently carried out to explore the relationship between various gene single nucleotide polymorphisms (SNPs) and an array of diseases. In such studies, HWE testing for each SNPs was often the first and quality control step. Those SNPs that did not pass the HWE tests were eliminated before moving on to the next step.89 On the other hand, in case-control genetic association studies, departures from HWE in controls have been associated with problems in the design, genotyping error or selection bias.90 In pooled analysis, checking HWE among controls was a good idea for included trials. Trikalinos et al. has demonstrated that exclusion of trials with departures from HWE may sometimes change the estimate pooled analysis result, they advocated that studies with departures from HWE should be excluded in pooled analysis.91 For this reason, in this pooled analysis, we carefully checked all selected trials and excluded these studies failing to meet the HWE test. In addition, the quality of included studies were generally at the medium level according to the Newcastle-Ottawa Scale (NOS), it indicated that the included studies met the criteria accepted for valid SNP-association studies.

Some previous pooled analysis concerning about the impact of ACE I/D gene polymorphism on DKD risk has been completed. In 2012, a meta analysis included 14,108 DKD cases and 12,472 controls from 63 published studies has been performed, it indicated that ACE I/D polymorphism was associated with DKD development in the Asian type 2 diabetes subjects.92 However, the genotype distributions of the control groups do not conform to HWE in some trials. And because of that, in order to gain a more credible pooled analysis result, we re-examined the related studies and included some other more high quality trials. In line with Wang et al., our study also found that the ACE I/D genotype was correlated with DKD risk in type 2 diabetes patients. Due to the fact that we included a plethora of studies in our analysis compared to the aforementioned analyses, we feel that our pooled results are more convincing.

ACE is a pivotal factor of the renin–angiotensin–aldosterone system (RAAS), it contains 26 exons and 25 introns located on 17q23. In 1990, ACE gene polymorphism was firstly described based on the insertion or deletion (I/D) of a 287bp Alu in the 16th intron.93 Whereafter, a series of ACE polymorphic genetic markers have been found (e.g. A240T, T93C, T594lC). Among these polymorphic marker, I/D polymorphism (rs4340) was the most investigated. On account of this I/D polymorphic marker, we could divide ACE gene polymorphism into DD homozygote, II homozygote and ID heterozygote. It has been demonstrated that ACE I/D polymorphism could affect ACE activity level both in plasma and various tissues.94 Additionally, a great number of previous studies have been carried out and verified the impact of ACE I/D gene polymorphism on various diabetes-related diseases.

DKD is a severe complication both in type 1 and type 2 diabetic patients, it damages about 40% of all diabetic subjects and is a crucial cause of chronic renal failure both in the Eastern and Western world. The pathogenesis of DKD is very complicated, it has been verified that various signaling pathways and molecular factors are activated during DKD, such molecular events include activation of systemic and local RAAS, generation of pro-inflammatory cytokines and excessive reactive oxygen species.3 In addition, recent GWAS studies demonstrated that DKD patients always suffer from genetic damage, and genetic factors are involved in the development of DKD, more critically, some specific gene SNPs might be associated with DKD susceptibility, thus it could provide remarkable clinical significance for preventing and early diagnosing of DKD through detailed illuminating the genetic mechanisms involved in DKD.

RAAS activation play a vital role in the occurrence and development of DKD, the RAAS is a pivotal regulator of renal arterial blood pressure by angiotensin II. However, conversion of low activity angiotensin I to high activity angiotensin II was relying on ACE. It has been showed that the ACE level is strongly correlated with ACE I/D polymorphism. Although the ACE I/D gene polymorphism is taken place in the non-coding gene region, the base insertion or deletion itself might alter the splicing process of the ACE precursor mRNA, then influence the stability of ACE mRNA, and ultimately affect the expression or stabilization of ACE. In situ hybridization for ACE mRNA on renal biopsy studies have found that the expression of ACE mRNA was increased in those subjects with the ACE DD genotype.95 Additionally, the serum ACE levels was also higher in the those individuals with D genotype than those with ID genotype or II genotype.93 And because of that, it was reasonable to consider that ACE I/D genetic variation was associated with the development of DKD. ACE D allele carriers had more higher ACE levels both in serum and kidney tissue, which lead to a more efficient activation of angiotensin II, and consequently resulted in the deterioration of DKD. In line with these, our pooled study further demonstrated that DKD risk was higher in those subjects with D allele than I allele carriers. We observed that the presence of II genotype offered a significant protective effect for DKD, whereas the presence of DD genotype conferred remarkable risk for DKD. The detailed mechanistic aspects that underlie the relationship between ACE I/D gene polymorphism and DKD was not completely clear. As mentioned earlier, the impact of ACE I/D gene polymorphism on DKD could be partially attributed to the effect of the ACE I/D polymorphic variant on the expression of the ACE gene. On the other hand, a recent study performed by Mahwish et al. found that ACE I/D genotypes was associated with dyslipidemia in diabetic patients, the DD genotype subgroup subjects were characterized by a significant higher levels of plasma triglycerides and total cholesteroln.96 In addition, the association of ACE I/D genotypes with atherosclerotic risk factors such as hypertension, dyslipidemia, and obesity in type 2 diabetic patients has been reported.97 Taken together, ACE DD genotype might result in the formation of diabetic renal lesions through elevating angiotensin II levels and a key contributor to dyslipidemia in a hyperglycemic environment further culminating in renal complications.

In this study, we found that the impact of ACE I/D gene polymorphism on DKD susceptibility was inconsonant in different types of diabetes and races. The pooled analysis showed that ACE I/D gene polymorphism was correlated with DKD susceptibility in Asian individuals, but there was no obvious correlation in Caucasian subjects. For another, we found that there was no correlation between them among 13 studies concerning type 1 diabetic patients, while ACE I/D gene polymorphism was correlated with the onset of DKD risk in type 2 diabetic patients. It indicates that ACE I/D genetic factors contribute more in patients with type 2 diabetes mellitus. Likewise, this inconsistency was also found in previous pooled analysis, Ng et al. found that ACE gene polymorphism was associated with DKD among type 2 diabetic Asians, while there was a reduced risk of DKD associated with the ACE I/D gene polymorphism among Caucasians with either type 1 or type 2 diabetes.98 Similarly, a pooled analysis performed by Wang et al. further found that the Asian group with T2DM showed a significant association. However, it failed to find any significant effects for different genetic models in T1DM and Caucasian subjects.92 Conversely, another pooled analysis performed by Xu et al. included 17 case-control studies in 2016 showed that ACE I/D polymorphism was correlated with DKD in the Asian groups with type 1 diabetes.9 While Fujisawa et al. found that the association was significant both in Asian populations and in Caucasian populations.10 Some reasons may account for the different results between Asians and Caucasians. Firstly, different lifestyle, environmental exposure, and different socioeconomic status may modify individual DKD susceptibility in different ethnic groups. Secondly, different genetic backgrounds in different racial subjects may influence genetic phenotypes.9 On the other hand, there are some other explanations for the predisposition to DKD in patients with type 2 DM. As mentioned above, the D allele of the ACE gene has been connected with higher ACE activity and increased level of angiotensin II. It has been found that increased angiotensin II could worsen insulin resistance and lipid metabolism disorders.99 In addition, both muscle capillary density and endogenous hepatic glucose production also could be affected by ACE I/D gene polymorphism.100

This study has several potential limitations. Firstly, most included trials were limited number and size. Second, there were evidences of public bias in this pooled study, in addition, the included trials were from various countries and races, which might decrease the reliability of this pooled analysis. Finally, our research was focused on ACE I/D genetic alteration, but previous studies have indicated that gene polymorphism in many other genes including Interleukin-6 -174G/C and angiotensinogen T174M gene polymorphism were correlated with DKD susceptibility,101,102 thus it can be argue that further pooled analysis concerning these genes SNPs are needed.

ACE I/D gene polymorphism is correlated with DKD risk in Asian, Chinese populations and type 2 diabetic individuals. ACE D allele and DD genotype is a risk factor for DKD. Conversely, ACE II genotype seems to be a protective factor of DKD. However, no correlation between ACE I/D gene polymorphism and the susceptibility of DKD was found in Caucasian or type 1 diabetic patients.

Shi-kun Yang, Wen-li Zeng, Fen-fen Chu analyzed the data for the manuscript and wrote the manuscript. Shi-kun Yang, Na Song, Wen-li Zeng performed the literature search. Fen-fen Chu, Shi-kun Yang edited the manuscript.

The data used to support the findings of this study are available from the first author and corresponding author upon request. This is a meta analysis using previous relevant published studies. There is no Human participants and/or Animals informed consent. None of the authors is in any condition that may represent a potential conflict of interest. The experiments were carried out according to the Ethics Review Committee of The Third Xiangya Hospital, Central South University.

This study was supported by the Hunan Provincial Health Commission Project (202103050178). Hunan Provincial Clinical medical technology innovation guide project (2020SK53601). The science and technology planning project of Hengyang City (2019jh011012).

The authors declare that they have no conflict of interest.