High-density lipoprotein sensor based on molecularly imprinted polymer
- 1.2k Downloads
Decreased blood level of high-density lipoprotein (HDL) is one of the essential criteria in diagnosing metabolic syndrome associated with the development of atherosclerosis and coronary heart disease. Herein, we report the synthesis of a molecularly imprinted polymer (MIP) that selectively binds HDL, namely, HDL-MIP, and thus serves as an artificial, biomimetic sensor layer. The optimized polymer contains methacrylic acid and N-vinylpyrrolidone in the ratio of 2:3, cross-linked with ethylene glycol dimethacrylate. On 10 MHz dual electrode quartz crystal microbalances (QCM), such HDL-MIP revealed dynamic detection range toward HDL standards in the clinically relevant ranges of 2–250 mg/dL HDL cholesterol (HDL-C) in 10 mM phosphate-buffered saline (PBS, pH = 7.4) without significant interference: low-density lipoprotein (LDL) yields 5% of the HDL signal, and both very-low-density lipoprotein (VLDL) and human serum albumin (HSA) yield 0%. The sensor reveals recovery rates between 94 and 104% at 95% confidence interval with precision of 2.3–7.7% and shows appreciable correlation (R 2 = 0.97) with enzymatic colorimetric assay, the standard in clinical tests. In contrast to the latter, it achieves rapid results (10 min) during one-step analysis without the need for sample preparation.
KeywordsHigh-density lipoprotein Molecularly imprinted polymer Quartz crystal microbalance Lipoprotein sensor
High-density lipoprotein (HDL) plays an essential role as antiatherogenic marker in the reverse cholesterol transport pathway . It does so by inhibiting oxidation of low-density lipoprotein (LDL) and hence preventing formation of oxidized LDL, which is a crucial atherogenic factor . Furthermore, HDL shows anti-inflammatory activity by inhibiting the production of inflammatory cytokines. Those induce expression of vascular cells and intracellular adhesion molecules on the coronary vascular . Therefore, decreased blood level of HDL means increased risk for developing metabolic syndrome and atherosclerosis and finally coronary heart disease (CHD) [4, 5]. Current clinical analysis approximates the serum concentration of actual HDL particles by determining the amount of cholesterol bound to HDL (HDL-C) in the serum  due to difficulties in measuring actual HDL particles in blood using any standard methods. Concentrations below 40 mg/dL HDL-C are related to high risk of CHD incidence, whereas values equal to or higher than 60 mg/dL indicate protection against CHD . In clinical analysis, HDL-C is analyzed by an enzymatic colorimetric assay that utilizes cholesterol esterase, cholesterol oxidase, and peroxidase coupled with UV-Vis photometry . This method requires sample pretreatment by precipitating all other serum proteins using reagents such as heparin , 50,000 Da dextran sulfate, phosphotungstic acid, polyethylene glycol , or dextran sulfate-coated iron particles  in the presence of divalent cations (e.g., Mg2+, Mn2+) [8, 9]. After centrifugation or magnetic separation, only HDL remains in the supernatant. However, increased blood levels of triglycerides or triglyceride-rich lipoproteins (e.g., very-low-density lipoprotein—VLDL) can interfere with precipitation and prevent sedimentation of aggregates. Therefore, supernatants may be contaminated with other lipoproteins leading to systematically too high results for HDL-C . Assay selectivity for HDL can be improved by adding polyethylene glycol beads coated with specific antibodies binding to serum apolipoprotein (Apo) B or C. Apo-B and Apo-C are present in several lipoproteins, namely, chylomicron, VLDL, intermediate-density lipoprotein (IDL), and LDL . The immune reaction hence eases precipitation. In a different approach, one can use a polyanion and synthetic polymer agents to block non-HDL lipoproteins before adding the colorimetric cholesterol reagents to determine HDL-C . Although the enzymatic HDL-C assay is inherently highly selective to cholesterol, it is limited by the abovementioned selectivity issues, as well as stability and high cost of antibodies and enzymes. Furthermore, it turned out that increased serum concentrations of triglycerides, bilirubin, ascorbic acid, free hemoglobin, and gamma-globulin, respectively, interfere .
As a consequence of such limitations, artificial recognition elements attract increasing interest in sensing. Molecularly imprinted polymers (MIPs) represent such biomimetic receptors showing appreciable selectivity, storage stability, resistance against biofouling, and reusability . MIPs contain functionalized cavities whose exact shape and surface chemistry is determined by self-organization between a growing polymer matrix and a template species, usually the target analyte . To date, a wide range of MIPs has been published covering small molecules as well as whole cells [14, 15]. Other bioanalytical applications of MIP-based sensors include detecting bio(macro)molecules, such as sugars , cholesterol , phospholipids , and proteins . Some MIPs have already been applied in clinical diagnosis and therapeutic monitoring, for instance to determine the concentration of human serum albumin (HSA) in serum , nicotine  or creatinine  in urine, and also in ABO blood group typing in whole blood . Recently, we reported the successful sensing of LDL with MIP-based quartz crystal microbalance (QCM) sensors (LDL-MIP sensor) directly in serum . Herein, we report the design of corresponding HDL-MIP sensors. The challenges for that were twofold: firstly, detection limits in the case of HDL have to be lower due to the lower clinically relevant threshold concentration (lower than 40 mg/dL for HDL-C, higher than 129 mg/dL for LDL-C). Secondly, LDL and HDL are both composed of similar types of components, namely, triglycerides and cholesteryl esters within the lipoprotein particle surrounded by a layer of amphipathic phospholipid, free cholesterol, and apolipoprotein. LDL and HDL differ by the ratios of these constituents and hence also slightly in diameter (21.5 ± 6.5 nm for HDL, 28.9 ± 9.2 nm for LDL) .
Materials and methods
Methacrylic acid (MAA), N-vinylpyrrolidone (VP), dimethyl sulfoxide (DMSO), potassium chloride (KCl), and agarose powder were purchased from VWR International (Vienna, Austria); N,N′-(1,2-dihydroxyethylene)bisacrylamide (DHEBA), 2,2′-azobis(isobutyronitrile) (AIBN), sodium bromide (NaBr), and Sudan Black B were from Sigma-Aldrich (Steinheim, Germany). Tris(hydroxymethyl)-aminomethane (Tris), ethylenediamine tetraacetic acid (EDTA), calcium chloride (CaCl2), magnesium chloride (MgCl2), urea (CH4N2O), and (D+)-glucose monohydrate were purchased from Merck (Darmstadt, Germany). Sodium chloride (NaCl) was obtained from Applichem (Darmstadt, Germany). Acetic acid was purchased from Carl Roth (Karlsruhe, Germany). 4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid (HEPES) was obtained from Alfa Aesar (Karlsruhe, Germany). HSA was purchased from Millipore (MA, USA). Brilliant gold paste (gold colloid, 12% gold content) was purchased from Heraeus, Germany. All reagents were of analytical or highest synthetic grade commercially available.
Conditions of ultracentrifugation for serum lipoprotein isolation
Ultracentrifuge rotor type 100 Ti, 80,000 rpm at 4 °C
Fractions obtained in the top layer
Lipoprotein density (g/mL)
Type of salt
Medium density (g/mL)
QCM transducer fabrication
Ten megahertz (MHz) QCM were fabricated by screen printing dual gold-electrode configuration onto commercially available AT-cut quartz plates (168 μm thick, 13.8 mm diameter; Great Microtama Industries, Surabaya, Indonesia) with a brilliant gold paste (Heraeus; 12%). Then, they were baked in the oven at 400 °C for 4 h. After measuring the resonance frequency and damping with an Agilent 8712ET network analyzer, QCM transducers with less than −5 dB damping were selected for further use .
Synthesis of HDL-MIP
Topographic images of HDL-MIP and NIP were recorded by atomic force microscopy (AFM) in contact mode. AFM was operated in air using a Bruker Instruments NanoScope VIII with a silicon nitride cantilever (ScanAsyst-air) at 0.5 Hz scan rate.
QCM were mounted in a custom-made cell connected to the oscillator circuit following a previously described protocol , which is also shown in Fig. 1. A typical measurement comprised of several steps: first, 180 μL of 10 mM phosphate-buffered saline (PBS, pH = 7.4) was injected into the measuring cell to obtain baseline signal. Afterward, the cell was flushed with 180 μL standard HDL-C solutions (3.12–350 mg/dL) in 10 mM PBS, respectively. All measurements were carried out in stopped flow until the signal reached its equilibrium state. Afterward, we washed the cell with 10% aqueous solution of acetic acid, followed by 0.1% SDS solution, and finally deionized water (10 min each at a flow rate of 0.46 mL/min) .
HDL-MIP sensor characterization
HDL-MIP sensors were characterized in terms of limit of detection, accuracy, precision, analytical sensitivity, and selectivity. Accuracy of HDL-MIP sensors was examined by recovery tests at clinically “low” and “normal” HDL-C concentrations. Test samples were prepared by adding different volumes of standard HDL-C solution at a concentration of 300 mg/dL (namely, 10, 40, and 80 μL) to 200 μL of 20 mg/dL HDL-C solution to reach final concentrations at 33.33, 66.66, and 100 mg/dL HDL-C, respectively. The corresponding frequency shifts were compared to the values expected from calibration of the sensor. For testing reproducibility, the sensor responses of HDL-C standard solutions containing 20, 50, and 100 mg/dL, respectively, were recorded three times each. For determining selectivity, the HDL-MIP sensor was exposed to standard solutions of possible interfering species at high concentrations that can be found in human serum, namely, 150 mg/dL LDL-C, 80 mg/dL VLDL-C, and 1000 mg/dL HSA, respectively.
Clinical sample measurement
Different volumes of standard HDL-C solution at a concentration of 385 mg/dL (namely, 10, 30, 50, and 60 μL) were spiked to different volumes of human serum with known HDL-C concentration c = 63 mg/dL (namely, 390, 370, 350, and 340 μL) to reach a final volume of 400 μL. All sera were diluted by mixing 1 part serum and 1 part PBS to reduce matrix effects prior to measurement .
As we did not have access to clinical samples with low HDL-C concentrations, those standards were prepared in “artificial serum” (AS). It contained 0.1% HSA, 4.5 mM KCl, 5 mM CaCl2, 4.7 mM (D+)-glucose monohydrate, 2.5 mM urea, 145 mM NaCl, and 1.6 mM MgCl2 in 200 mM HEPES buffer (pH = 7.4) . Then, HDL-free “artificial” serum was prepared by adding 100 mg/dL of LDL-C standard and 20 mg/dL VLDL-C standard. Different volumes of a standard HDL-C at a concentration of 385 mg/dL (namely, 5, 10, 20, 30, and 40 μL) were spiked to different volumes of HDL-free artificial serum (namely, 395, 390, 380, 370, and 360 μL) to reach a final volume of 400 μL at the concentrations of 4.8, 9.6, 19.3, 28.9, and 38.5 mg/dL HDL-C, respectively. Two types of assay matrixes, namely, 10 mM PBS and diluted HDL-free artificial serum (1 part HDL-free artificial serum plus 1 part 10 mM PBS), were utilized to achieve baseline signal. All spiked sera were diluted with 10 mM PBS by 1:2 prior to sensor measurements.
Results and discussion
Optimizing HDL-MIP synthesis
HDL-MIP sensor characterization
Methods for HDL-C determination
Methods for HDL-C determination
Detection range (mg/dL)
Detection time (min)
QCM-based HDL-MIP sensor
Homogeneous enzymatic colorimetric assay
QCM-based HDL immunosensor
Fiber-optic-based HDL immunosensor
Mean (mg/dL) (n = 3)
Validation of HDL-MIP sensor data
The HDL-MIP-based QCM sensor presented here is able to selectively detect HDL in the clinically relevant concentration range, both in (diluted) serum and in buffers. In contrast to existing clinical standard techniques for determining HDL, this sensor does not require any sample pretreatment other than diluting it. It hence represents a reagentless sensing technology and leads to reduced assay complexity and time of measurement. Furthermore, the signals are inherently reversible, making the system potentially useful for long-term measurements in clinical monitoring.
Open access funding provided by University of Vienna. This project was supported by the Royal Thai Government through a Scholarship granted by the Office of the Higher Education Commission (Grant#04/2556), Thailand. We also gratefully acknowledge the Faculty of Medical Technology and the Faculty of Pharmaceutical Sciences, Prince of Songkla University, for blood collection and ultracentrifugation.
Compliance with ethical standards
All experiments are based on a single sample of donated blood given by a consenting, informed, healthy volunteer at Hat Yai University Hospital. It was part of surplus production not used as blood preservation until clinical expiry date and provided to us for isolating lipoproteins and serum standards, which is in line with regulations and does not require formal additional consent to the study by the ethics committee.
Conflict of interest
The authors declare that they have no conflict of interest.
- 5.Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive summary of the third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA. 2001;285(19):2486–97.CrossRefGoogle Scholar
- 6.Brunzell JD, Davidson M, Furberg CD, Goldberg RB, Howard BV, Stein JH, et al. Lipoprotein management in patients with cardiometabolic risk: consensus conference report from the American Diabetes Association and the American College of Cardiology Foundation. J Am Coll Cardiol. 2008;51(15):1512–24.CrossRefGoogle Scholar
- 7.Warnick GR, Nauck M, Rifai N. Evolution of methods for measurement of HDL-cholesterol: from ultracentrifugation to homogeneous assays. Clin Chem. 2001;47(9):1579–96.Google Scholar
- 8.Warnick GR, Albers JJ. A comprehensive evaluation of the heparin-manganese precipitation procedure for estimating high-density lipoprotein cholesterol. J Lipid Res. 1978;19(1):65–76.Google Scholar
- 9.Kimberly MM, Leary ET, Cole TG, Waymack PP. Selection, validation, standardization, and performance of a designated comparison method for HDL-cholesterol for use in the cholesterol reference method laboratory network. Clin Chem. 1999;45(10):1803–12.Google Scholar
- 10.Nauck M, März W, Haas B, Wieland H. Homogeneous assay for direct determination of high-density lipoprotein cholesterol evaluated. Clin Chem. 1996;42(3):424–9.Google Scholar
- 11.Harris N, Galpchian V, Thomas J, Iannotti E, Law T, Rifai N. Three generations of high-density lipoprotein cholesterol assays compared with ultracentrifugation/dextran sulfate–Mg2+ method. Clin Chem. 1997;43(5):816–23.Google Scholar
- 18.Jang R, Kim KH, Zaidi SA, Cheong WJ, Moon MH. Analysis of phospholipids using an open-tubular capillary column with a monolithic layer of molecularly imprinted polymer in capillary electrochromatography-electrospray ionization-tandem mass spectrometry. Electrophoresis. 2011;32(16):2167–73.CrossRefGoogle Scholar
- 29.Cox RA, García-Palmieri M. Cholesterol, triglycerides, and associated lipoproteins. In: Walker HK, Hall WD, Hurst JW, editors. Clinical methods: the history, physical, and laboratory examinations. 3rd ed. Boston: Butterworths; 1990. p. 153–60.Google Scholar
- 35.Choi JW, Park JH, Lee WC, Oh BK, Min JH, Lee WH. Fluorescence immunoassay of HDL and LDL using protein A LB film. J Microbiol Biotechnol. 2001;11(6):979–85.Google Scholar
- 36.Warnick GR, Wood PD. National Cholesterol Education Program recommendations for measurement of high-density lipoprotein cholesterol: executive summary. The National Cholesterol Education Program working group on lipoprotein measurement. Clin Chem. 1995;41(10):1427–33.Google Scholar
- 38.Chicea D, Chicea R, Chicea LM. HSA particle size characterization by AFM. Rom Rep in Phys. 2013;65:178–85.Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.