Abstract

BACKGROUND: Prolylcarboxypeptidase (PRCP) is a membrane-associated serine protease that regulates biologic peptides bradykinin, angiotensin II, α-melanocyte stimulating hormone, and activates prekallikein. PRCP and its variants have been associated with metabolic syndrome in males, hypertension in preeclampsia, resistance to ACE inhibitors, and embryonic angiogenesis. Recent studies on PRCPgt/gt mice that have 25% normal PRCP levels show that these animals are hypertensive, have higher risk for arterial thrombosis, increased vascular inflammation, and reduced angiogenic repair after ischemia and injury (Blood 117:3929, 2011; Blood 122:1522, 2013). The hypothesis of this investigation is that polymorphisms in PRCP are associated with cardiovascular disease (CVD).

METHODS: DNA from 2,243 subjects from the PEACE Trial (NEJM 351:2058, 2004) were genotyped at 2 PRCP SNPs (rs7104980, rs2298668) and 3 KLKB1 (prekallikrein) SNPs (rs4253252, rs3733402, rs3087505). These subjects were 82% female and had a history of (in decreasing frequency) angina (71%), angiographic coronary disease (66%), heart attack (58%), PTCA (44%), hypertension (42%), CABG (36%), diabetes (15%), stroke (4%), and TIA (3%). The association between single SNP alleles and different cardiovascular related phenotypes was assessed using logistic regression models. These models were adjusted for age, weight, gender, history of hypertension, and history of diabetes.

RESULTS: When the logistic regression model was adjusted for age, weight, and gender, SNP rs710980 in PRCP had a 21% increased odds [odds ratio=1.211; 95% CI=(1.008, 1.454)] of having a history of PTCA if carrying the G allele (frequency of 66%) as compared to not having the G allele (frequency 34%). Additionally, adjusting the logistic regression model for history of hypertension and diabetes did not alter the odds ratio for PTCA history. However when the logistic regression model included history of hypertension and diabetes, the odds of having a history of a MI if carrying the same allele also is increased by 21% [odds ratio=1.21; 95% CI=(1.001, 1.455)]. In summary, the rs710980 intronic PRCP SNP conferred risk for CVD, while the PRCP exonic SNP rs2298668 showed no relationship in this population. Alternatively, the exonic KLKB1 SNP rs3733402 of Apple Domain 2 where high molecular weight kininogen binds conferred reduced odds of 24% [odds ratio=0.76; 95% CI=(0.622, 0.928)] of having a history of angiographic coronary disease if carrying the G allele (frequency 67%) as compared to not having the G allele (frequency of 33%). The other two KLKB1 SNPs showed no associations.

DISCUSSION: These combined data on PRCP suggest that it is a risk factor for CVD. The present genetic data are consistent with biochemical, cell culture, and in vivo investigations showing that PRCP levels influence vascular biology, renal function, and metabolism. The extent of the genetic associations of PRCP polymorphisms with CVD may be resultant of the database interrogated. The PEACE trial consisted of individuals not severely ill with CVD. Its hypothesis to show that administration of ACE inhibitors leads to increased protection from mild CVD was not met. However, the present investigation showing several positive odds ratios with a certain PRCP SNP for CVD suggests that PRCP is a stronger risk factor for CVD than can be demonstrated in the PEACE subject population. PRCP should be interrogated in additional populations of subjects with CVD.

Funding Support: NIH HL052779-17, HL112666-2

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.