Abstract
Familial chylomicronemia syndrome (FCS) is a rare heritable disorder associated with severe hypertriglyceridemia and recurrent pancreatitis. Lipoprotein lipase deficiency and apolipoprotein C-II deficiency are two well-characterized autosomal recessive causes of FCS, and three other genes have been described to cause FCS. Because therapeutic approaches can vary according to the underlying etiology, it is important to establish the molecular etiology of FCS.
A man originally from North Africa was referred to the University of Pennsylvania Lipid Clinic for severe hypertriglyceridemia and recurrent pancreatitis, consistent with the clinical diagnosis of FCS. Molecular analyses of FCS-associated genes revealed a homozygous missense variant R72T in APOC2. Molecular modeling of the variant predicted that the apolipoprotein C-II R72T peptide has reduced lipid binding affinity. In vitro studies of the patient’s plasma confirmed the lack of functional apoC-II activity. Moreover, the apoC-II protein was undetectable in the patient’s plasma, quantitatively as well as qualitatively.
We identified a missense APOC2 variant causing apoC-II deficiency in a patient with severe hypertriglyceridemia and recurrent pancreatitis. Beyond dietary management and usual pharmacologic therapies, an apoC-II mimetic peptide may become an optional therapy in patients with apoC-II deficiency in the future.
Familial chylomicronemia syndrome (FCS) is a rare disorder presenting with severe hypertriglyceridemia (HTG) and recurrent pancreatitis (1, 2). Pancreatitis typically presents during childhood or adolescence, with triglyceride (TG) levels often exceeding 2000 mg/dL. The best-characterized molecular causes of FCS are lipoprotein lipase (LPL) (OMIM 238600) and apolipoprotein C-II (apoC-II) (OMIM 608083) deficiency due to biallelic loss-of-function mutations in the LPL and APOC2 genes, respectively (2). LPL is the primary enzyme hydrolyzing TGs in TG-rich lipoproteins, and apoC-II is its required cofactor. Thus, LPL and apoC-II deficiencies are clinically indistinguishable, requiring functional and/or molecular studies to differentiate them. LPL deficiency exhibits markedly impaired LPL activity in postheparin plasma, resisting correction by the addition of normal plasma or apoC-II protein, whereas impaired LPL activity from apoC-II deficiency corrects by the addition of normal plasma or apoC-II (1, 2).
In FCS, a fat-restricted diet is the mainstay of therapy, but, even with diet and adjunctive pharmacotherapy, pancreatitis often recurs (2). Importantly, for severe pancreatitis, apoC-II deficiency can be transiently managed with fresh frozen plasma (FFP), providing apoC-II (3). Similarly, LPL deficiency may be treated with the gene therapy alipogene tiparvovec, which has been approved in Europe (4) and is available experimentally in the United States (cf NCT02904772). Because each therapy only treats its corresponding defect, a patient without a genetic diagnosis would not be a candidate for either therapy. Given divergent therapies, the knowledge of molecular diagnosis of FCS can be life saving and highlights the importance of implementing genetic testing in clinical practice. Here, we present a case of FCS from apoC-II deficiency due to a homozygous APOC2 missense variant R72T and show that an apoC-II mimetic peptide can restore lipolytic activity in the patient’s plasma.
Materials and Methods
The subject was enrolled in a research study on the genetic etiology of dyslipidemias at the University of Pennsylvania Lipid Clinic. Participants whose phenotype suggests monogenic HTG and pancreatitis are recruited for genotyping. The patients provided their consent to medical records reviews, blood draw, lipid testing, and genetic analysis. The National Heart Lung and Blood Institute provided comparator specimens. Protocols were approved by their respective Institutional Review Boards.
Medical and family history
The patient, a 42-year-old man from North Africa, presented with HTG at age 30 with acute pancreatitis, the first of several yearly episodes. Typical TG ranged from 700 to 4000 mg/dL (maximum 7112 mg/dL) despite fat restriction and lipid-lowering medications (fenofibrate, 40 mg; omega-3 acid ethyl esters, six capsules; and atorvastatin, 80 mg daily). His mother and father are distantly related, with additional consanguinity on both sides. His brother, one sister, and two maternal uncles have HTG and pancreatitis (Supplemental Fig. 1).
Genetic analysis
DNA was extracted from whole blood on the QIAsymphony SP system (Qiagen, Hilden, Germany). Next-generation sequencing was used using the LipidSeq targeted panel (5), focusing on FCS-related genes. The identified variant was confirmed by Sanger sequencing.
ApoC-II cofactor study
Recombinant human LPL (rhLPL) was used in the apoC-II cofactor studies (2.85 µg/mL final concentration). The patient’s plasma, as well as a biochemically confirmed LPL-deficient plasma sample and a non–LPL-deficient HTG plasma sample, served as the source of TG substrates (8.0 µg TG per well) in 50 µL of phosphate-buffered saline (pH 7.4). An apoC-II mimetic peptide (C-II-a) was used to activate rhLPL, and the activity was monitored by released free fatty acids as described (6). FFAs generated were quantified enzymatically (Wako Chemicals, Richmond, VA) on a SpectraMax 384 Plus plate reader (Molecular Devices, Sunnyvale, CA).
Western blot analysis of ApoC-II
EDTA-plasma (3.5 mL) from the study patient, a biochemically confirmed LPL-deficient patient, a non–LPL-deficient patient with HTG, and a healthy volunteer, as well as purified apoC-II (50 ng) (A50302; Meridian Life Science, Inc., Memphis, TN) were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis on 4% to 12% NuPAGE gels (Life Technologies, Carlsbad, CA). Western blots were performed as described (7) with the following modifications: Immubilon Psq (Millipore, Billerica, MA) as the membrane and affinity-purified rabbit antihuman apoC-II (#K9711R; Meridian Life Sciences, Memphis, TN) as the primary antibody, diluted 1/8000 in Antibody Dilution Buffer (7).
Results and Discussion
Analysis of known FCS-associated genes (LPL, APOC2, APOA5, LMF1, and GPIHBP1) revealed a homozygous missense variant c.215G>C in APOC2, translating to p.R72T (p.R50T without the signal peptide) in apoC-II. Only about 2% of FCS arises from APOC2 defects (2), with <20 unique mutations reported in the Human Gene Mutation Database (www.hgmd.cf.ac.uk). Missense or nonsense mutations are the most common type among pathogenic APOC2 mutations (about 50%), and they are scattered throughout the protein. This particular variant has not been reported in the Human Gene Mutation Database or in the Exome Aggregation Consortium (n = 60,706 unrelated individuals) and is the only one identified within the lipoprotein binding domain.
Four different in silico analyses (PolyPhen2, PROVEAN, MutationTaster, and SIFT) predict that the arginine-to-threonine amino acid change in this variant is deleterious (Supplemental Table 1). The missense change occurs at the interface between the hydrophobic and hydrophilic face of the second amphipathic helix of apoC-II (Fig. 1). At this position, positively charged amino acids (e.g., arginine and lysine) are generally known to be critical in lipid binding (8). This accords with the calculated lower hydrophobic moment of the mutant helix and with its decreased helicity and more shallow depth of penetration through the phospholipid monolayer determined by molecular modeling (Fig. 1).
The structure modeling of C-II-a, a native and mutated apoC-II peptide. (A and B) Helical wheel plots of the second helix of native apoC-II (A) and of mutant apoC-II (B) drawn using Wheel.pl, v 1.4 2009-10-20 software. Charged and hydrophobic amino acids are colored in purple and light green, respectively. The identified variant R72T is indicated by an arrow in each figure. The arginine to threonine change occurs at the interface between the hydrophobic and hydrophilic face of an amphipathic helical portion of apoC-II, which altered the hydrophobic moment, a measure of the amphiphilicity, of the peptide, indicated by the number in the center of helix. (C and D) Modeling of binding of the second helix of native apoC-II (C) and of mutant apoC-II protein (D) to the phospholipid monolayer (membrane). The helical structures were created with PEPFOLD 3.0 software. Native whole protein structure was adopted from NCBI MMDB (ID: 28588) and mutated in silico using the Schrodinger Software package. The generated proteins were then introduced to the phospholipid monolayer using the OPM database and the PPM web server (http://opm.phar.umich.edu/server.php).
No enhanced lipolysis was observed in all plasma samples with the addition of C-II-a alone due to the low LPL level without prior heparin treatment. Adding exogenous rhLPL alone failed to enhance LPL lipolytic activity in the patient plasma, indicating a lack of functional apoC-II (Fig. 2). Adding C-II-a markedly increased lipolytic activity, thus confirming the absence of functional apoC-II that was explained by the lack of LPL activity. In contrast, adding exogenous rhLPL to the plasma from a confirmed LPL-deficient patient restored the lipolytic activity, and no further enhanced lipolysis was observed after adding C-II-a.
An apoC-II cofactor study using an rhLPL. Control plasma with high levels of TG (TG: 1146 mg/dL), apoC-II–deficient patient plasma (TG: 1711 mg/dL), and biochemically confirmed LPL deficiency plasma (TG: 1211 mg/dL) were incubated in vitro with rhLPL alone, with apoC-II mimetic peptide (C-II-a) alone, or with both rhLPL and C-II-a, and the production of free fatty acids generation was measured. The results represent the mean in triplicates ± SD. **P < 0.005; ***P < 0.0005.
The patient’s plasma apoC-II was below the detectable limit of 1 mg/dL quantitatively measured by an immunoturbidimetric assay (Kamiya Biochemical, Seattle, WA) on a Cobas C311 analyzer (Roche Diagnostics, Mannheim, Germany). This was confirmed by Western blotting, revealing no discernible band corresponding to apoC-II in the patient’s plasma (Supplemental Fig. 2). Because residues around the site of mutation are known to contribute to apoC-II/lipoprotein binding (9), these findings suggest the mutant apoC-II R72T is rapidly catabolized owing to poor lipoprotein binding, with subsequent renal clearance.
There are potential benefits to diagnosing apoC-II deficiency. First, the genotype decisively distinguishes it from LPL deficiency. This is critical to choosing appropriate therapy, especially because apoC-II deficiency can be managed emergently by providing functional apoC-II. Given severe HTG and life-threatening pancreatitis, infusing FFP can acutely supplement exogenous/functional apoC-II, reducing TGs to stabilize pancreatitis (3). This could be done in any hospital capable of providing FFP provided the accurate diagnosis is known. Regrettably, this is seldom the case, prompting undertreatment of such patients when they become critically ill from pancreatitis. Second, genotyping can be used efficiently to screen family members for apoC-II deficiency before suffering life-threatening symptoms. Our results show that an apoC-II mimetic peptide lowers TGs in apoC-II–deficient mice (6, 10). In our case, this peptide was used successfully to substitute the missing endogenous apoC-II protein in our patient’s plasma studies. This apoC-II mimetic peptide is under preclinical development and could eventually treat apoC-II–deficient patients as well as other forms of HTG with secondary apoC-II deficiency.
Abbreviations:
- apoC-II
apolipoprotein C-II
- FCS
familial chylomicronemia syndrome
- FFP
fresh frozen plasma
- HTG
hypertriglyceridemia
- LPL
lipoprotein lipase
- rhLPL
recombinant human lipoprotein lipase
- TG
trigyceride.
Acknowledgments
Disclosure Summary: A.D.M., D.J.R, E.D., F.B., L.M., M.E., M.U., N.S., R.L.D., R.A.H., S.D., and T.U.S. have nothing to declare. A.T.R., A.W., D.S., L.F., M.A., and R.D.S. are employed by the National Institutes of Health, which holds the patent for the ApoC-II mimetic peptide used in this study (U.S. Pat: 8,936,787), and A.T.R. and M.A. are inventors.
References
Brunzell J, Deeb SS. Familial lipoprotein lipase deficiency, Apo C-II deficiency, and hepatic lipase deficiency. In: Valle D, Scriver CR, Beaudet AL, Vogelstein B, eds. The Online Metabolic & Molecular Bases of Inherited Disease. New York, NY: McGraw-Hill; 2014. Chapter 117.
Freeman LA. Western blots. In: Freeman L, ed. Lipoproteins and Cardiovascular Disease: Methods and Protocols. New York, NY: Humana Press; 2013:369–385.
Author notes
These authors contributed equally to this work.
Address all correspondence and requests for reprints to: Richard L. Dunbar, MD, Division of Translational Medicine and Human Genetics, University of Pennsylvania, 3600 Spruce Street, Maloney Building 9th Floor, Room MO9010, Philadelphia, Pennsylvania 19104. E-mail: richard.dunbar@uphs.upenn.edu.
