Fifty years of lipoprotein(a) – the Magical Mystery Tour continues


  • B Angelin

    Corresponding author
    1. Metabolism Unit, Center for Endocrinology, Metabolism and Diabetes, Department of Medicine, and Molecular Nutrition Unit, Center for Biosciences NOVUM, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden
    • Correspondence: Bo Angelin, Department of Endocrinology C2:94, Karolinska University Hospital Huddinge, Stockholm S-14186, Sweden.

      (fax: +46 8 711 0710; e-mail:

    Search for more papers by this author

This year marks not only 150 years since the launch of Journal of Internal Medicine (previously Acta Medica Scandinavica), but also 50 years since the discovery of lipoprotein (a) [Lp(a)], an enigmatic protein that has mystified medical scientists ever since. In this issue, we celebrate this event by publishing a stimulating review by Florian Kronenberg and Gerd Utermann, who have both long been involved in research to understand the role of this remarkable protein [1].

In 1963, the Norwegian physician Kaare Berg described a genetic variation in low-density lipoprotein (LDL) that was detected by heterologous antibodies from rabbits immunized with human LDL [2]. This variation, which he attributed to the presence of an antigen termed Lp(a), was initially considered to be simply one of numerous genetic polymorphisms described at that time.

The phase of trying to understand the potential role of Lp(a) actually has an interesting Scandinavian history. One particular perspective originated from early studies by Dahlén and colleagues in northern Sweden. In a number of reports published in this journal, they described a specific lipoprotein fraction in humans, pre-beta1-lipoprotein [3]. This protein was more frequently detected in subjects with early manifestations of cardiovascular disease or in their families. The demonstration, in a Scandinavian collaboration between Dahlén, Frick from Finland and Berg, that the pre-beta1-lipoprotein fraction was in fact Lp(a) was a major advance [4]. The association between high levels of Lp(a) and cardiovascular disease risk was rapidly confirmed in a number of cohort studies during the 1970s. Through detailed biochemical studies, it became clear that the Lp(a) particle consists of an apolipoprotein (apo)B-containing LDL particle with an attached molecule of apo (a) (Fig. 1), and that the large interindividual difference in the concentration of apo(a) represents a more complex genetic trait.

Figure 1.

Schematic model of Lp(a). The central LDL particle has a core of neutral lipids (triglycerides and cholesteryl esters) and a surface consisting of free cholesterol and phospholipids with an integrated apoB molecule. The attached apo(a) molecule has a variable number of kringle repeats that determines the size of the Lp(a) particle. The number of circulating Lp(a) particles is inversely related to their size.

The cloning of apo(a) in 1987 by McLean et al. [5] was a major breakthrough, and the fact that apo(a) has a very similar structure to plasminogen led to speculations that Lp(a) represents a link between plasma lipoproteins and risk of thrombosis. Knowledge of the variations in the molecular structure of the apo(a) gene was important for understanding the strong genetic regulation of Lp(a) concentration, and the inverse relation between Lp(a) particle size and plasma concentration [6].

Although Lp(a) level is stable in a given individual over time, it is difficult to obtain conclusive data from cohort studies because of the pronounced interindividual variation in concentration. Early studies had demonstrated an association between high Lp(a) levels and cardiovascular disease; however, this could not be confirmed in several prospective studies carried out in the 1990s. As discussed extensively in the review in this issue by Kronenberg and Utermann [1], the application of new methods of genetic epidemiology, including Mendelian randomization, has now re-established Lp(a) as the most powerful genetic risk factor for cardiovascular disease. Whether Lp(a) levels are high should therefore be included in the evaluation of high-risk individuals, as recently outlined in international guidelines [7]. However, because the availability of Lp(a)-lowering drugs is limited (see below), the most relevant use of this knowledge presently is to even more vigorously try to influence other risk factors in such individuals.

In his seminal review in 1989, ‘The mysteries of lipoprotein(a)’, Utermann discussed a number of important issues that still needed to be explored [6]. These included the following questions. What are the functions of Lp(a) and apo(a)? What is the mechanism of the strict genetic control of Lp(a) levels? Where is Lp(a) assembled? Is there a relationship between the metabolism of Lp(a) and that of LDL? And are atherogenic or thrombogenic properties the cause of the association between Lp(a) and cardiovascular disease? When reviewing the subject again 10 years later, Utermann found that despite some increased understanding, many of the most important questions remained unanswered [8].

Now, more than a further 10 years later, it is reassuring that some of these issues are finally being addressed [1], including the mechanisms of genetic regulation of Lp(a) variation and the importance of elevated Lp(a) as a risk factor for cardiovascular disease. However, a number of challenges remain; we still do not know the function of Lp(a) [1], or how it participates in the promotion of cardiovascular disease.

The finding that Lp(a) has developed twice during vertebrate evolution, as it is only present in hedgehogs and some primates, is intriguing. Furthermore, the fact that Lp(a) concentration increases during hibernation in hedgehogs may suggest that it has a stabilizing function at levels of very low body temperature and metabolic activity. In humans, Lp(a) is also part of the acute phase response to inflammation, and it has been speculated that it may be involved in wound healing. There are large variations in Lp(a) concentration between different ethnic groups also after adjustment for genotypic variation. High levels are observed in individuals with kidney disease, as well as in those with familial hypercholesterolaemia. Many hormones have pronounced effects on Lp(a) levels, which may indicate a more general role in the regulation of metabolic activity. Recent unexpected findings suggest that low Lp(a) levels may predispose towards type 2 diabetes, and that bile acids may influence Lp(a) production through the nuclear receptor FXR and circulating fibroblast growth factor 19 [1].

Although it is known that elevated Lp(a) levels signal an increased cardiovascular disease risk, relatively few treatment options are available at present. In clinical practice, nicotinic acid is the only drug that has a major effect on Lp(a), as was originally reported in this journal more than 20 years ago [9]. LDL or even Lp(a) apheresis may be considered as treatment for patients with very high plasma levels of Lp(a). Newer drugs that are presently in clinical trials, such as PCSK9 or CETP blockers, inhibitors of apoB production or thyromimetics, seem to have Lp(a)-lowering effects [1]. So far, the evidence that lowering of elevated Lp(a) levels can reduce cardiovascular disease risk is not strong, but trials to compare clinical outcomes using such treatments with matched LDL-lowering using statins should be possible within the not too distant future.

Lipoprotein (a) has thus been a challenge to medical science for 50 years, still keeping some of its mysteries [6]. Like the Magical Mystery Tour of the Beatles, who incidentally made their first LP recording exactly 50 years ago, the progress of understanding Lp(a) has followed an unforeseen route characterized by surprising events and unexpected revelations, incessantly providing new challenges to its fans – be they biochemists, geneticists, cardiologists, lipidologists or endocrinologists. For Lp(a), there is great promise that the tour will continue, providing new and unexpected discoveries and insights for many years to come.

Conflict of interest statement

No conflict of interest was declared.