PPBP gene as a biomarker for coronary heart disease risk in postmenopausal Thai women

Materials

Dulbecco’s phosphate-buffered saline (D-PBS) and TRIzol® reagent were purchased from Invitrogen (Carlsbad, CA, USA). Luna Universal One-Step Reaction mix (2×) was purchased from BioRad Laboratories Inc. (Hercules, CA, USA). Human CXCL7/PBP matched antibody pair kit (ab219537) and enzyme-linked immunosorbent assay (ELISA) accessory pack (ab210905) were purchased from Abcam (Cambridge, UK). Human HNP-1–3 ELISA reagents were purchased from Hycult Biotech (Uden, the Netherlands). All other reagents were from Sigma-Aldrich (St. Louis, MO, USA).

Study design and patient population

The patient flow and experimental design are summarized in Fig. 1. The study was conducted at the Faculty of Tropical Medicine, Mahidol University, Thailand. The study was approved by the Ethical Committees of the Faculty of Tropical Medicine (MUTM 2020-007-01, MUTM 2020-007-02) and Bhumibol Adulyadej Hospital (No. 43/64). All participants were informed of the study objectives and completed an informed consent form before enrollment. The sample size calculation from two independent means formula (Rosner, 2000) was made by assuming baseline of mean and standard deviation of all parameters obtained from our previous study (Maneerat et al., 2017). The calculated sample size of hyperlipidemia and coronary heart disease groups are 20 and 35 cases, respectively.

Patients

All the participants were unrelated postmenopausal women born to Thai parents. Seventy-eight patients were diagnosed, classified, and treated by a specialist (CM) at the Hospital for Tropical Diseases, Faculty of Tropical Medicine, and Bhumibol Adulyadej Hospital, Thailand. They were classified into three groups based on their clinical manifestations, according to the American College of Cardiology/American Heart Association criteria (2013) (Goff et al., 2014): 18 patients had high cholesterol levels (H group) (total cholesterol (TC), low-density lipoprotein (LDL), and high-density lipoprotein (HDL)), 39 patients had high cholesterol levels and received cholesterol-lowering drugs (HD group), and 21 patients were diagnosed with CHD (CHD group). No patients in the H and HD groups had any evidence of vital organ dysfunction. We also enrolled 12 healthy controls with no infections, underlying disease, or CVD risk factors (N group). No H patients or controls received any cholesterol- or blood pressure-lowering medication.

Blood sample collection and methods

Whole blood samples (10 ml) were collected once from healthy controls and patients into EDTA blood vacutainer tubes with serum clot activator before hyperlipidemia treatment (except HD group) or coronary bypass grafting. Plasma (2 ml) was immediately collected by centrifugation of whole blood samples and prepared for lipid analysis, and kept at −70 °C for the detection of plasma PPBP and HNP-1–3. Serum levels of lipids (TC, triglycerides (TG), HDL, and LDL) were measured.

Packed blood cells were resuspended in D-PBS (Carlsbad, CA, USA) and used to isolate mononuclear cells. Approximately 2 × 10⁶ peripheral blood mononuclear cells (PBMCs) in TRIzol (Invitrogen, Carlsbad, CA, USA) were kept at −70 °C for determination of PPBP and DEFA1/DEFA3 mRNA expression levels by quantitative reverse transcription-polymerase chain reaction (RT-qPCR) analysis, using specific primers. Gene expression was expressed as fold-change relative to the housekeeping β-actin gene (ACTB) (Kotepui et al., 2012; Maneerat et al., 2017).

Lipid test

Lipid markers including TC, TG, LDL, and HDL were analyzed enzymatically using kits (Randox Laboratories Ltd., Crumlin, UK) and a biochemistry analyzer (Architect CI 16200; Abbott Laboratories, Abbott Park, IL, USA).

Investigation of DEFA1/3 and PPBP mRNA expression by RT-qPCR

RT-qPCR was performed in duplicate. Each 10-µl RT-qPCR reaction contained 5 µl of Luna Universal One-Step Reaction mix (2×) (BioRad Laboratories, Hercules, CA, USA) mixed with 10 ng of RNA. Total RNA was isolated from 2 × 10⁶ PBMCs using TRIzol (Invitrogen, Carlsbad, CA, USA) and 0.4 µm of each set of forward and reverse primers. The primers were designed based on the DEFA1/DEFA3 genes (GenBank accession numbers NM_005217.3) (forward: 5′-TCCTTGCTGCCATTCTCCTG-3′ and reverse: 5′-TGCACGCTGGTATTCTGCAA-3′) (Li et al., 2014) and PPBP gene (GenBank accession numbers NM_002704.3) (forward: 5′-TTGTAGGCAGCAACTCACCC-3′ and reverse: 5′-TGCAAGGCATGAAGTGGTCT-3′ (Yeo et al., 2016). The expected PCR product sizes were 204 bp and 135 bp, respectively. The RT-qPCR reaction was run in a Bio-Rad CFX96 Real-time system (BioRad Laboratories, Hercules, CA, USA). The RT-qPCR conditions were as follows: 55 °C for 10 min and 95 °C for 1 min, followed by 40 cycles of denaturation at 95 °C for 10 s, extension at 60 °C for 30 s (+plate read), and melting curve analysis at 65 °C for 5 min. The expected PCR product size was 204 bp. ACTB primers (forward: 5′-TCACCCACACTGTGCCCATCTACGA-3′ and reverse: 5′-CAGCGGAACCGCTCATTGCCAATGG-3′) (Heid et al., 1996; Kotepui et al., 2012) were used to normalize the relative expression levels of DEFA1/DEFA3 and PPBP. The 2^−(ΔΔCt) method was used to quantify relative expression levels (Kotepui et al., 2012; Maneerat et al., 2017).

Determination of protein levels by ELISA

The plasma kept at −70 °C was thawed to room temperature (RT) and then centrifuged at 2,500 × g for 5 min. PPBP and HNP-1−3 protein levels were determined in the clear plasma.

Determination of plasma HNP-1–3 levels

Plasma HNP-1–3 concentrations were measured in triplicate using ELISA kits (Hycult Biotech, Uden, Netherlands), according to the manufacturer’s instructions. Briefly, 100 μl of each plasma from the three groups (N = 12, H = 18, and CHD = 21) was diluted 20-fold in sample dilution buffer, and standards were prepared at concentrations of 10,000–156 pg/ml. The diluted samples and two-fold serial diluted standard were transferred to microtiter wells coated with captured antibody to HNP-1–3 and incubated for 60 min at RT. Each well was then washed to remove unbound material and biotinylated tracer antibody to HNP-1–3 (100 μl) was added, followed by incubation for 60 min at RT. After washing to remove unbound material, streptavidin-peroxidase conjugate was added (100 μl/well) and incubated for 60 min at RT. Bound enzyme was detected by adding 100 μl of tetramethylbenzidine substrate to each well, and reactions were stopped by adding 100 μl of stop solution. The optical density was determined at 450 nm using an Ao Microplate Reader (Azure Biosystems, Dublin, CA, USA). Plasma levels of HNP-1–3 were calculated from the HNP-1–3 standard curves (Maneerat et al., 2016).

Determination of plasma PPBP levels

Plasma PPBP levels were measured in triplicate using a Human CXCL7/PBP matched antibody pair kit (ab219537) and ELISA accessory pack (ab210905) (Abcam, Cambridge, UK), according to the manufacturer’s instructions as previously described (Maneerat et al., 2016). Briefly, 100 μl of plasma from all groups, as above, were diluted 5,000-fold in sample dilution buffer, and standards at concentrations of 1,000–15.6 pg/ml were transferred to microtiter wells coated with 2 μg/ml of the specific captured antibody and incubated for 2 h at RT. Each well was then washed to remove unbound material and 0.5 μg/ml of biotinylated detector antibody was added, followed by incubation for 1 h at RT. After washing unbound material, 0.02 μg/ml of streptavidin-horseradish peroxidase conjugate was added and incubated for 1 h at RT. Bound enzyme was detected by adding 100 μl of tetramethylbenzidine substrate to each well, and reactions were stopped by adding 100 μl of stop solution. The optical density was measured at 450 nm using an Ao Microplate Reader (Azure Biosystems, Dublin, CA, USA). Plasma levels of PPBP were calculated from the PPBP standard curves (Maneerat et al., 2017).

Statistical analysis

All data were determined for normality using the kolmogorov–Smirnov test. Most were non-normally distributed variables. Clinical data, plasma levels of PPBP and HNP-1–3 were reported as median (minimum, maximum). mRNA expression was presented as fold-change relative to mRNA in healthy controls. Parameters were compared between two and more than two groups by Mann–Whitney test and Kruskal–Wallis test, respectively. Correlations between parameters and clinical data were analyzed by the Spearman rank correlation test. The α level was set at <0.05 with a 95% confidence interval. All statistical analyses were performed using SPSS version 18 (SPSS, Chicago, IL, USA).

Characteristics and clinical manifestations of controls and patient groups

The general data and clinical manifestations of the controls and patients are shown in Table 1. Patients were classified based on their clinical manifestations according to the American College of Cardiology/American Heart Association criteria (2013) (Goff et al., 2014). Our criteria for patient selection was previously described (Maneerat et al., 2016). Especially, the H group had high cholesterol levels (TC, LDL, HDL), but no evidence of vital organ dysfunction. Before inclusion, we confirmed that participants in the N and H groups had not received any cholesterol- or blood pressure-lowering medication. CHD patients were about to undergo coronary bypass grafting. All CHD patients had been treated with statins since they were diagnosed and after surgery. None of the N or H patients had any infections or underlying diseases. As shown in Table 1, there was no significant difference in age between the N and H groups (p > 0.05), but patients in the CHD group were significantly older than the other groups (p < 0.0001). The differential complete blood counts (CBC) were similar in all three groups (p > 0.05) except number of platelets in H were significantly higher than CHD groups (p = 0.0115), and monocyte in CHD were significantly higher than the other groups (p < 0.0001). The TC (p < 0.0001), and LDL levels (p = 0.0003, and p < 0.0001) were significantly higher in the H compared with the N and CHD groups. The HD group included participants with hyperlipidemia who were taking lipid-lowering drugs daily. These women showed significant differences in most lipid parameters, including TC (p < 0.0001), TG (p = 0.0048), and LDL (p < 0.0001), compared with the H group. The significant difference in CBC between the H and HD groups included differences in white blood cell count (p = 0.0146), RBC (p = 0.0280), hematocrit (p = 0.0004), platelets (p = 0.0292), and neutrophils (p = 0.0147) (Table S1). These findings indicated that the clinical manifestations in the HD group were significantly affected by lipid-lowering treatment. The HD group was therefore excluded from subsequent analyses.

mRNA expression in PBMC extracts

The relative mRNA expression levels (mean two-fold changes) of PPBP and DEFA1/DEFA3 in the N, H, and CHD groups are shown in Fig. 2. PPBP mRNA levels were significantly higher in the CHD group (5.439, 0.173–25.02) compared with the N (0.8470, 0.265–3.506), p = 0.0018) and H groups (0.173, 0.830–7.260), p = 0.0009) (Fig. 2A). DEFA1/DEFA3 mRNA levels were not significantly different among N, H, and CHD groups (p > 0.05) (Fig. 2B).

Plasma levels of PPBP and HNP-1–3

Plasma levels of PPBP and HNP-1–3 proteins were determined by ELISA and compared among the N, H, and CHD groups (Fig. 3). PPBP levels were significantly higher in the H (1268.23, 583.579–1154.980 ng/ml, p < 0.0001) and CHD (867.965, 556.912–1492.70 ng/ml, p = 0.0004) groups compared with the N group (575.685, 68.842–944.632 ng/ml) (Fig. 3a). In contrast, there were no significant differences in HNP-1–3 levels among the patient and control groups (p > 0.05) (Fig. 3b).

Correlations between mRNA expression, plasma protein levels, and clinical manifestations

The correlations among characteristics/clinical manifestations, PPBP and DEFA1/DEFA3 mRNA expression, and plasma PPBP and HNP-1–3 levels are shown in Table S2. Plasma PPBP levels were significantly correlated with age (r = 0.3950, p = 0.0307), TC (r = 0.5794, p = 0.0008), LDL (r = 0.4923, p = 0.0057), and number of platelets (r = 0.4067, p = 0.0257). In contrast, plasma HNP-1–3 levels were not significantly associated with any clinical parameters. Neither DEFA1/DEFA3 mRNA nor plasma levels of HNP-1–3 were correlated with number of neutrophils (p > 0.05).