| COMMENTARIES |
Apolipoprotien(a) Isoforms and Lipoprotein(a) as Risk Factors for Atherosclerosis
Katerina Tosheska, M.D., M.Sc.1, Danica Labudovic, M.D., Ph.D. 1, Sonja Alabakovska, M.D., Ph.D. 1, Mirko Spiroski, M.D., Ph.D.2, and Bojana Todorova. M.D., Ph.D. 1, 1Institute of Medical and Experimental Biochemistry, Faculty of Medicine, University Ss. "Kiril and Metodij," Skopje, Republic of Macedonia, 2Institute of Immunobiology and Human Genetics, Faculty of Medicine, University Ss. "Kiril and Metodij," Skopje, Republic of Macedonia
Please address correspondence to:
Katerina Tosheska -Trajkovska, M.D., M.Sc.
Assistant of Biochemistry
Institute of Medical and Experimental Biochemistry
Faculty of Medicine
50 Divizija 6, 1000 Skopje, Republic of Macedonia
Tel: ++38923217303
Fax: ++38923230431
E-mail: tosheskatrajkovska@gmail.com
Introduction
A detailed study of specific lipoprotein parameters is necessary in individuals at high risk of coronary heart disease (CHD). During the last two decades, attention has been focused on identification of new risk factors (such as genetic risk factors) in these individuals. One potential risk factor that has been extensively studied is lipoprotein(a) [Lp(a)]. It is well documented, that Lp(a) accumulates in the arterial wall at the sites of atherosclerotic lesions [1]. The localization of Lp(a) within the arterial wall suggests a direct causative role of Lp(a) in the initiation or progression of atherosclerosis. The proatherogenic and prothrombotic effects of Lp(a) are related to its interaction with macrophages of the arterial wall and to the structural similarity of apo(a) and plasminogen, leading to deposition of Lp(a) in atherosclerotic lesions and interference with the fibrinolytic system [1-3].
Polymorphisms of Apolipoprotein (a)
Lp(a) contains a unique protein, apolipoprotein(a) [apo(a)], linked by a disulfide bond to apolipoprotein B100 of LDL [4,5]. The apo(a) gene locus on chromosome 6q26-27 (LPA) has been identified as the major quantitative trait locus (QTL) for Lp(a) concentrations [6,7]. The apo(a) gene resides within 50 kb of the plasminogen gene and shares a high degree of sequence identity with its neighbor [apo(a) contains a 5’ signal sequence, a 3’ plasminogen (PLG) like protease domain and ten different types of Kringle IV domains (KIV-1 to KIV-10)] [8].
Small apo(a) isoforms (< 22 KIV repeats) are positively associated with premature development of atherosclerosis, particularly when accompanied with high plasma Lp(a) concentrations (> 25-30 mg/dL) [11-12].
Since up to 90% of the variance in Lp(a) levels is attributable to the apo(a) locus [10,13], the possibility that polymorphisms of the apo(a) gene, other than size, could contribute to the increase of Lp(a) levels, must be considered. These include a pentanucleotide (TTTTA) repeat polymorphism in the promoter, and sequence variation in coding and non-coding regions of the gene, including a C/T polymorphism at position +93 (relative to the transcription start site [14-16].
Findings from Our Studies
In our study of healthy Macedonian adults [17], we determined allele frequencies of apo(a) isoforms separated by SDS-PAGE and immunoblotting. Despite the fact that this technique is not sensitive enough to detect low levels of apo(a) protein, results of our study showed that the frequency distribution of apo(a) isoforms fit the expectation of the Hardy-Weinberg equilibrium. Our data indicate that each apo(a) isoform we detected by our method was specified by a corresponding allele at the apo(a) locus. The distribution of alleles in our subjects was skewed towards alleles encoding large apo(a) isoforms associated with low Lp(a) levels. Study of healthy Macedonian children [18] confirmed this finding. Large apo(a) isoforms (S4 and S3 from single banded and S4S3 from double-banded) were most prevalent in both studies [17,18]. Our results obtained in healthy adults and children indicate that apo(a) isoforms are primarily genetically determined, and are not affected by plasma lipid and apolipoprotein concentration. Highly skewed distribution of plasma Lp(a) levels can be accounted for by high frequencies of alleles encoding large isoforms and a null allele.
The Importance of Lp(a) and Apo(a) Isoform Determination in Diabetic Children
Additional risk factors, such as genetic risk factors, may favor the increased cardiovascular morbidity and mortality observed in diabetic children [19,20]. Among the genetic risk factors, Lp(a) and apo(a) polymorphism has been evaluated. The results of Kronenber et al. [21] suggest that apo(a) may represents a genetic risk factor for the development of type 1 diabetes; this is in keeping with studies that demonstrated linkage of type 1 diabetes mellitus to chromosome 6q27 [22], which corresponds to the region where the apo(a) gene locus has been mapped [23].
As the process of atherosclerosis starts in childhood, the risk factors should be assessed and prevention of atherosclerosis should start in childhood. Studies analyzing the impact of Lp(a) on the development of CAD in diabetic patients gave controversial findings [24-27] which might be result of the insufficient number of included subjects in case-control studies, lack of information on apo(a) phenotype and urinary albumin excretion.
Studies considering apo(a) isoforms expose the problem that many isoforms are represented only in low numbers due to the existence of more than 30 alleles. One way to overcome the problem is to combine isoforms of similar size. Most of the researchers divide apo(a) isoforms into two subgroups according to the size of smaller apo(a) isoforms [28-30]. The LMW group usually includes all subjects with at least one apo(a) isoform with less than 22 KIV repeats; the HMW group comprised all subjects having only isoforms with more than 22 KIV repeats.
We compared apo(a) isoforms and Lp(a) concentration in 60 diabetic children and 100 non-diabetic children; all were without family history of atherosclerosis and none had detectable albuminuria [31]. Our results demonstrate an increased prevalence of single-banded than of double-banded isoforms in diabetic children. We were not able to find a significant difference in the percentage of LMW apo(a) isoforms between diabetics and the controls. HMW apo(a) isoforms were associated with decreased Lp(a) concentrations in both diabetic and non-diabetic subjects.
Our observations do not support increased Lp(a) concentrations in young normoalbuminuric IDDM subjects. These results indicate that apo(a) isoforms and plasma Lp(a) concentration are primarily genetically determined, and are not affected by plasma lipid and apoprotein concentrations.
But, the identification of diabetic children with high Lp(a) levels and/or apo(a) isoforms of low MW may be very important in clinical practice. In these subjects the genetic predisposition for CAD due to apo(a) gene is added to the cardiovascular risk related to diabetes and associated cardiovascular risk factors. Although there is no practical method for lowering Lp(a) concentrations at this time, these patients could be more intensively treated with respect to reduction of modifiable risk factors for cardiovascular disease and diabetes complications.
In summary, evaluation of both Lp(a) and apo(a) isoforms provides a more complete characterization of the risk for cardiovascular changes linked to the apo(a) gene than the evaluation of Lp(a) levels alone.
References