Jun 05

The Big Deal About Lipoprotein(a)

A Mysterious Figure

While studying the lipid system in depth, lipoprotein(a) (pronounced lipoprotein little a; also called Lp(a)) was a particle that repeatedly came up in the study material in passing, although at the time I had no idea what it was. It was never something I had seen mentioned in the mainstream information on lipids, but the more I read about it the more I started to get the suspicion that lipoprotein(a) wasn’t exactly like other lipoproteins, like LDL and HDL. The general structure was the same as other lipoproteins – with a phospholipid “shell”, cholesterol being carried as its “cargo”, and proteins attached to the shell (called apolipoproteins) that allowed it to carry out certain functions. But, the research often described lipoprotein(a) as mysterious, or an enigma, and oft repeated was that its function was still largely unknown. At the same time, it was stated that it was an important risk factor for heart disease, and many papers discussed possible ways to lower it. I was left wondering if lipoprotein(a) was really just a particularly deadly particle causing damage wherever it went, or if there could be more to the story.

A Lipoprotein With a Tail

Lipoprotein(a) is a low density lipoprotein that is found in humans, old world apes, and the hedgehog.1 Lipoprotein(a), like LDL, contains a protein called apolipoprotein B (apoB) and Lp(a) is often described as “LDL-like”. This is because the structure of  lipoprotein(a) is very similar to LDL, but with one addition. Attached to the apoB there is another protein – apolipoprotein(a). Like other apolipoproteins, apo(a) is what allows Lp(a) to carry out different functions, but the structure of apolipoprotein(a) is vastly different from other apolipoprotein structures I had seen.

Instead of being incorporated into the shell of the lipoprotein as others are, it is instead attached to apoB at one end and wraps around Lp(a) like a long tail.Apo(a) also comes in different sizes, and its size is determined by genetic factors, based on how many copies of a protein (called a kringle) it has. The size of apo(a) is one of the determining factors for levels of lipoprotein(a) in the blood: the larger the apo(a) form (the longer the ‘tail’), the lower the genetic baseline of Lp(a), and likewise the shorter the ‘tail’, the higher the baseline Lp(a).

Risky Business

Beyond studies focusing on Lp(a) metabolism, structure, and function, many studies I saw were centered around lipoprotein(a) as a risk factor for heart disease. This is because people with cardiovascular disease typically have higher levels of lipoprotein(a)3, lipoprotein(a) appears to have some moderate predictive outcomes when it comes to cardiovascular disease4, and some studies show that having higher genetic levels of lipoprotein(a) is associated with increased risk, as well5 – although associated doesn’t necessarily mean causal. But, is the big picture so uncomplicated that lipoprotein(a) can be painted as a “risk” that we’re better off having as low as possible, as early as possible? The answer to this is quickly complicated if one looks at lipoprotein(a)’s association with all-cause mortality, cancer mortality, and risk for brain and airway bleeding as low levels are correlated to higher risk for all of them.6, 7 While genetic baselines do contribute a large deal to lipoprotein(a) levels in the blood, it isn’t the only factor involved.

Beyond Genes

Beyond genetic levels lies more clues…

Dietary changes, specifically low fat high carbohydrate diets can raise lipoprotein(a)8, and different protein sources can also impact levels.9 Additionally, Insulin-Like Growth Factor (IGF-1) lowers lipoprotein(a), although the mechanism isn’t known and may involve either increased clearance, or decreased production.10, 11 I found it quite interesting, as well, that interleukin-6 (IL-6; a protein used for inflammatory signalling) raises lipoprotein(a) levels in vitro12 which is likewise reflected in human models where the IL-6 receptor is blocked with drug therapy resulting in lower lipoprotein(a) levels.13 This fit with the speculation that lipoprotein(a) is an acute phase reactant similar to hs-CRP. In other words, lipoprotein(a) may go up from certain types of inflammation caused by damage or infection elsewhere in the body.

There is in fact some evidence for this, as seen in in vitro experiments14 and studies looking at patients during the acute phase response compared to controls.15 Higher levels of lipoprotein(a) are also found in those with conditions related to inflammation, such as lupus16 and rheumatoid arthritis17 although this may also be partially genetic.18 This role as an acute phase reactant – levels rising in response to specific inflammatory signalling – could partially explain why it is correlated with heart disease risk beyond genetically determined levels, as atherosclerosis is tied to inflammation and damage in the arteries as well.

Graph Source: doi:10.1161/ATVBAHA.107.145805

Lp(a) vs. Lp-Pla2

One way to separate the risk of lipoprotein(a) alone from its increased level during inflammatory states is to control for a risk factor that would indicate damage that might increase inflammation (and lipoprotein(a) by proxy) – such as oxidative damage. One study compared lipoprotein(a) levels with levels of Lipoprotein-Associated Phospholipase 2 (Lp-pla2). Lp-pla2 interacts with oxidized fats found on the phospholipid shells of lipoproteins when they’re damaged, removing them in order to protect the lipoprotein from further damage caused by oxidative byproducts. In this way, Lp-pla2 has antioxidant and protective functions, and high levels of lp-pla2 activity would be indicative of high levels of oxidative damage.19

When comparing people with high or low levels of lipoprotein(a) compared to high or low levels of lp-pla2, in those with high lipoprotein(a) but low levels of lp-pla2 the hazard ratio for increased cardiovascular risk was only 1.1 (that would be a 10% comparative increase, not especially significant). This was the same risk as having lipoprotein(a) in the lowest group but a mid-range level of lp-pla2. Meanwhile, those with high lipoprotein(a) and high lp-pla2 had a hazard ratio of 3.5, a 350% relative increase.20 In other words, if lipoprotein(a) was high, but signs of oxidative damage were low, so was risk for heart disease.

More Than Just a Marker

Beyond all the talk about hazard ratios, and risk, and all-cause mortality, though, there was one question that persisted while studying lipoprotein(a): What is it for? The other lipoproteins had clear uses outlined for distributing energy, or cellular repair, or managing immune reactions, but lipoprotein(a)’s use in the system remained elusive and poorly defined. It didn’t appear to transport energy, and although it was similar to LDL in shape, it has a lower affinity for the LDL receptor21, and thus likely couldn’t be used primarily by cells for repair via traditional means, either. Luckily, there have been a few possible hints about its use in the system, beyond as just a marker for risk.

For one, the structure of lipoprotein(a)’s “tail” – apo(a) – is similar to plasminogen22, which is used during injury repair. When an injury occurs, for example in an artery, platelet accumulation occurs and a protein called fibrin acts like a glue to bind it together, forming a scab-like structure over the wound to prevent bleeding.23 This scab is not just a bandage over a wound, but is actively involved in the healing process and is constantly changing through progression of the repair. One of these changes is mediated through plasminogen binding to fibrin, to break apart the “glue” (fibrin) holding the scab together in order to maintain proper structure, and ensure thrombosis does not occur. This dissolution process of plasminogen is called fibrinolysis.24

Balancing the Scales

Lp(a), along with plasminogen, may help maintain balance between clot production and dissolution.

Apo(a) appears to bind competitively to fibrin over plasminogen, blocking the fibrinolysis effects of plasminogen, and thus may contribute to decreased clot dissolution25, although this same mechanism may be useful in maintaining homeostasis during wound healing. Just like plasminogen and fibrin, lipoprotein(a) is found in healing tissue, but not in healthy tissue, at the same sites that fibrin is located, especially on the surface of the fibrous cap. It is speculated that lipoprotein(a) helps prevent excess fibrinolysis, which would result in bleeding and impaired repair, on the outside surface of the clot in order to aid with injury resolution.26 This use in clot strengthening, and inhibiting clot dissolution, through binding to fibrin may explain why higher levels of lipoprotein(a) are associated with lower levels of death related to brain and airway bleeding, as well, as increased fibrinolysis during a major bleeding event could be detrimental in terms of mortality outcomes.

Carrying a Heavy Burden

Beyond its involvement in wound repair, I discovered that lipoprotein(a) also has a few other key features. For one, it appears to be involved in the immune system similar to other lipoproteins. Infection by Hepatitis C, for example, is  inhibited via interaction with apolipoprotein(a) and this inhibition is proportional to the apo(a) size. In other words, the longer tails did a better job at inhibiting infection in vitro.27 The extent of lipoprotein(a)’s involvement in the immune system is likely still largely unknown, but this interaction does provide one example of the possibilities that may be uncovered in the future.

Viruses aren’t the only thing to attach themselves to apo(a), though. One of lipoprotein(a)’s most interesting aspects is its role as a preferential carrier for oxidized phospholipids. As discussed previously, phospholipids are what the membrane of cells are made of. When cells, or lipoproteins, become damaged they release these oxidized phospholipids (oxPL) to prevent further injury. What happens to these oxidized phospholipids? If Lp(a) is present, they preferentially accumulate on and bind to apo(a).28

OxLDL By Another Name…

Because lipoproteins can transfer their oxPL to Lp(a), or more accurately that they remove oxPL from their shell and Lp(a) picks it up, the levels of oxLDL and Lp(a) are very similar – almost the same.29 It isn’t that Lp(a) is the only lipoprotein to become oxidized, in fact it isn’t Lp(a) itself being oxidized that’s being picked up by these tests, but rather that apo(a) is carrying oxPL originating from other particles. There is a possibility that Lp(a) plays a role in the innate immune system, and picks up this oxPL in order to detoxify it, and further transfers byproducts from this process to other carriers to remove it from the system entirely. However, if oxidative stress is too high, the capacity of Lp(a) to handle this role may be impaired, and thus lead to increased risk of heart disease – hence why high Lp-pla2 activity modifies risk as it may be a marker of how much “workload” that Lp(a) has.30 This role as a detoxifier may also explain why, referring back to the Lp-pla2 paper, when comparing cardiovascular outcomes between those with no Lp(a) (and thus no oxLDL) and those in the next lowest quintile, the risk doubles. Peter of Hyperlipid has also speculated that the package of oxPL, carried by lipoprotein(a), may be useful in inducing apoptosis in cancer cells, and there is some evidence showing apo(a) inhibits tumor growth.31, 32

Riddle Wrapped in Mystery

There are no shortcuts to finding answers, or solving puzzles

To be sure, there is much we do not know about lipoprotein(a). Not only as far as risk in general, but also if it’s the lipoprotein(a) in itself that contributes to risk or if the context matters. We do have some hints, seemingly pointing towards context being key, and at the very least it seems that it is high lipoprotein(a) at a later stage of disease that may be influencing risk, as lipoprotein(a) isn’t associated with early thickening of the arteries, which contradicts the common idea that lipoprotein(a) is harmful to the arteries in itself.33 In addition, it appears that the oxidized phospholipid content contained on apo(a) is highly important when associated with extent of disease progression34, with one study stating that oxPL may “significantly contribute or primarily account for” the risk associated with lipoprotein(a) (emphasis mine).

Another question left unanswered is how much population based risk calculating studies are influenced by those with familial hypercholesterolemia as they also tend to have higher levels of lipoprotein(a) and are already at higher risk of developing cardiovascular disease and dying early.35 While the argument is made that this risk is from lipoprotein(a) or high cholesterol levels in general, it may also be that the difference in LDL receptor efficiency may also result in delayed or inefficient healing from arterial damage36 and thus higher need for the reparative aspects of lipoprotein(a). It is notable that in French centenarians, high lipoprotein(a) levels were quite prevalent37, and that in elderly populations there was no correlation between lipoprotein(a) levels and all-cause mortality in men38 – perhaps because of the lower concentration of those with familial hypercholesterolemia in elderly groups.

Beyond risk, there is much to learn about the functional role of lipoprotein(a), as well, and as study of it is far younger than the other lipoproteins we may have a long wait before we can shed light on what other influences it has, what other roles it may play, and what other mysteries it may contain. I am sure I will revisit lipoprotein(a) as we continue to learn about it, and I look forward to unraveling the mystery of such a unique lipoprotein. Until then, you can check out this quick recap to sum up lipoprotein(a) and context of risk in under 10 minutes in the short talk I did at Low Carb Breckenridge 2018:


Berglund, L. “Lipoprotein(a): An Elusive Cardiovascular Risk Factor.” Arteriosclerosis, Thrombosis, and Vascular Biology, vol. 24, no. 12, Dec. 2004, pp. 2219–26. Crossref, doi:10.1161/01.ATV.0000144010.55563.63.
Gaubatz, J. W., et al. “Human Plasma Lipoprotein [a]. Structural Properties.” The Journal of Biological Chemistry, vol. 258, no. 7, Apr. 1983, pp. 4582–89.
van Buuren, Frank, et al. “Incidence of Elevated Lipoprotein (a) Levels in a Large Cohort of Patients with Cardiovascular Disease.” Clinical Research in Cardiology Supplements, vol. 12, no. Suppl 1, Mar. 2017, pp. 55–59. PubMed, doi:10.1007/s11789-017-0087-y.
Kotani, Kazuhiko, et al. “Evidence-Based Assessment of Lipoprotein(a) as a Risk Biomarker for Cardiovascular Diseases – Some Answers and Still Many Questions.” Critical Reviews in Clinical Laboratory Sciences, vol. 53, no. 6, Nov. 2016, pp. 370–78. Crossref, doi:10.1080/10408363.2016.1188055.
Nordestgaard, Børge G., et al. “Lipoprotein(a) as a Cardiovascular Risk Factor: Current Status.” European Heart Journal, vol. 31, no. 23, Dec. 2010, pp. 2844–53. Crossref, doi:10.1093/eurheartj/ehq386.
Langsted, Anne, et al. “High Lipoprotein(a) and Low Risk of Major Bleeding in Brain and Airways in the General Population: A Mendelian Randomization Study.” Clinical Chemistry, vol. 63, no. 11, Nov. 2017, pp. 1714–23. Crossref, doi:10.1373/clinchem.2017.276931.
Sawabe, Motoji, et al. “Low Lipoprotein(a) Concentration Is Associated with Cancer and All-Cause Deaths: A Population-Based Cohort Study (the JMS Cohort Study).” PloS One, vol. 7, no. 4, 2012, p. e31954. PubMed, doi:10.1371/journal.pone.0031954.
Faghihnia, Nastaran, et al. “Changes in Lipoprotein(a), Oxidized Phospholipids, and LDL Subclasses with a Low-Fat High-Carbohydrate Diet.” Journal of Lipid Research, vol. 51, no. 11, Nov. 2010, pp. 3324–30. Crossref, doi:10.1194/jlr.M005769.
Nilausen, Karin, and Hans Meinertz. “Lipoprotein(a) and Dietary Proteins: Casein Lowers Lipoprotein(a) Concentrations as Compared with Soy Protein.” The American Journal of Clinical Nutrition, vol. 69, no. 3, Mar. 1999, pp. 419–25. Crossref, doi:10.1093/ajcn/69.3.419.
10 Laron, Z., et al. “Insulin-like Growth Factor-I Decreases Serum Lipoprotein(a) during Long-Term Treatment of Patients with Laron Syndrome.” Metabolism, vol. 45, no. 10, Oct. 1996, pp. 1263–66. Crossref, doi:10.1016/S0026-0495(96)90245-0.
11 Wang, Xing Li, et al. “Acute Effects of Insulin-like Growth Factor-1 and Recombinant Growth Hormone on Liprotein(a) Levels in Baboons.” Metabolism, vol. 51, no. 4, Apr. 2002, pp. 508–13. Crossref, doi:10.1053/meta.2002.31328.
12 Ramharack, R., et al. “Dominant Negative Effect of TGF- 1 and TNF- on Basal and IL-6 Induced Lipoprotein(a) and Apolipoprotein(a) MRNA Expression in Primary Monkey Hepatocyte Cultures.” Arteriosclerosis, Thrombosis, and Vascular Biology, vol. 18, no. 6, June 1998, pp. 984–90. Crossref, doi:10.1161/01.ATV.18.6.984.
13 García-Gómez, Carmen, et al. “Lipoprotein(a) Concentrations in Rheumatoid Arthritis on Biologic Therapy: Results from the CARdiovascular in RheuMAtology Study Project.” Journal of Clinical Lipidology, vol. 11, no. 3, May 2017, pp. 749-756.e3. Crossref, doi:10.1016/j.jacl.2017.02.018.
14 Noma, A., et al. “Lp(a): An Acute-Phase Reactant?” Chemistry and Physics of Lipids, vol. 67–68, Jan. 1994, pp. 411–17.
15 Min, W. K., et al. “Relation between Lipoprotein(a) Concentrations in Patients with Acute-Phase Response and Risk Analysis for Coronary Heart Disease.” Clinical Chemistry, vol. 43, no. 10, Oct. 1997, pp. 1891–95.
16 Borba, E. F., et al. “Lipoprotein(a) Levels in Systemic Lupus Erythematosus.” The Journal of Rheumatology, vol. 21, no. 2, Feb. 1994, pp. 220–23.
17  Dursunoğlu, Dursun, et al. “Lp(a) Lipoprotein and Lipids in Patients with Rheumatoid Arthritis: Serum Levels and Relationship to Inflammation.” Rheumatology International, vol. 25, no. 4, May 2005, pp. 241–45. Crossref, doi:10.1007/s00296-004-0438-0.
18 Asanuma, Yu, et al. “Serum Lipoprotein(a) and Apolipoprotein(a) Phenotypes in Patients with Rheumatoid Arthritis.” Arthritis & Rheumatism, vol. 42, no. 3, Mar. 1999, pp. 443–47. Crossref, doi:10.1002/1529-0131(199904)42:3<443::AID-ANR8>3.0.CO;2-Q.
19 Silva, Isis T., et al. “Antioxidant and Inflammatory Aspects of Lipoprotein-Associated Phospholipase A2 (Lp-PLA2 ): A Review.” Lipids in Health and Disease, vol. 10, no. 1, 2011, p. 170. Crossref, doi:10.1186/1476-511X-10-170.
20 Kiechl, S., et al. “Oxidized Phospholipids, Lipoprotein(a), Lipoprotein-Associated Phospholipase A2 Activity, and 10-Year Cardiovascular Outcomes: Prospective Results From the Bruneck Study.” Arteriosclerosis, Thrombosis, and Vascular Biology, vol. 27, no. 8, May 2007, pp. 1788–95. Crossref, doi:10.1161/ATVBAHA.107.145805.
21 Snyder, M. L., et al. “Binding and Degradation of Lipoprotein(a) and LDL by Primary Cultures of Human Hepatocytes. Comparison with Cultured Human Monocyte- Macrophages and Fibroblasts.” Arteriosclerosis, Thrombosis, and Vascular Biology, vol. 14, no. 5, May 1994, pp. 770–79. Crossref, doi:10.1161/01.ATV.14.5.770.