Haptoglobin inhibits phospholipid transfer protein activity in hyperlipidemic human plasma
© Henderson et al; licensee BioMed Central Ltd. 2009
Received: 22 June 2009
Accepted: 23 July 2009
Published: 23 July 2009
Skip to main content
© Henderson et al; licensee BioMed Central Ltd. 2009
Received: 22 June 2009
Accepted: 23 July 2009
Published: 23 July 2009
Haptoglobin is a plasma protein that scavenges haemoglobin during haemolysis. Phospholipid Transfer Protein (PLTP) transfers lipids from Low Density Lipoproteins (LDL) to High Density Lipoproteins (HDL). PLTP is involved in the pathogenesis of atherosclerosis which causes coronary artery disease, the leading cause of death in North America. It has been shown that Apolipoprotein-A1 (Apo-A1) binds and regulates PLTP activity. Haptoglobin can also bind to Apo-A1, affecting the ability of Apo-A1 to induce enzymatic activities. Thus we hypothesize that haptoglobin inhibits PLTP activity. This work tested the effect of Haptoglobin and Apo-A1 addition on PLTP activity in human plasma samples. The results will contribute to our understanding of the role of haptoglobin on modulating reverse cholesterol transport.
We analyzed the PLTP activity and Apo-A1 and Haptoglobin content in six hyperlipidemic and six normolipidemic plasmas. We found that Apo-A1 levels are proportional to PLTP activity in hyperlipidemic (R2 = 0.66, p < 0.05) but not in normolipidemic human plasma. Haptoglobin levels and PLTP activity are inversely proportional in hyperlipidemic plasmas (R2 = 0.57, p > 0.05). When the PLTP activity was graphed versus the Hp/Apo-A1 ratio in hyperlipidemic plasma there was a significant correlation (R2 = 0.69, p < 0.05) suggesting that PLTP activity is affected by the combined effect of Apo-A1 and haptoglobin. When haptoglobin was added to individual hyperlipidemic plasma samples there was a dose dependent decrease in PLTP activity. In these samples we also found a negative correlation (-0.59, p < 0.05) between PLTP activity and Hp/Apo-A1. When we added an amount of haptoglobin equivalent to 100% of the basal levels, we found a 64 ± 23% decrease (p < 0.05) in PLTP activity compared to basal PLTP activity. We tested the hypothesis that additional Apo-A1 would induce PLTP activity. Interestingly we found a dose dependent decrease in PLTP activity upon Apo-A1 addition. When both Apo-A1 and Hpt were added to the plasma samples there was no further reduction in PLTP activity suggesting that they act through a common pathway.
These findings suggest an inhibitory effect of Haptoglobin over PLTP activity in hyperlipidemic plasma that may contribute to the regulation of reverse cholesterol transport.
Haptoglobin is an acute phase protein that scavenges haemoglobin released into the circulation . Haptoglobin, the plasma protein with highest binding affinity to haemoglobin, is mainly expressed in the liver . It plays an anti-oxidant role by binding free haemoglobin and forming a complex that is taken up by hepatocytes and macrophages . The human haptoglobin gene encompasses three alleles: Hp1F, Hp1S and Hp2 . The Hp2 allele is the fusion product of the Hp1F and Hp1S alleles. Haptoglobin presents as a dimer of two of these alleles which binds to one haemoglobin dimer . Haptoglobin expression is induced several fold in the event of inflammation triggered by infection, injury or cancer development [1, 5]. Haptoglobin has been shown to play an antioxidant/anti-inflammatory role, to contribute to neutrophil activation , to maintain reverse cholesterol transport  and to modulate the inhibition of cyclooxygenase and lipooxygenase , amongst other functions. In particular, haptoglobin has been shown to inhibit Lecithin-Cholesterol Acyltransferase (LCAT) in human ovarian follicular fluid . LCAT is involved in the removal of cholesterol excess from peripheral tissues . LCAT transfers an acyl chain from high density lipoprotein (HDL) lecithin to cellular cholesterol. This activity is stimulated by the presence of Apo-A1, the main protein constituent of HDL. Balestrieri et al  demonstrated that LCAT activity is negatively correlated with the Hp/Apo-A1 ratio in human follicular fluid. The mechanism of action of haptoglobin inhibition of LCAT activity has been described . The binding site of Haptoglobin on Apo-A1 has been mapped and it was demonstrated that the interaction of haptoglobin to Apo-A1 is independent to the binding of haptoglobin and haemoglobin. A peptide designed based on the sequence in Apo-A1 that putatively interacts with Haptoglobin was shown to restore LCAT activity inhibited by Hp demonstrating that the Apo-A1-Hp interaction is responsible for the inhibition of LCAT activity. Based on this evidence it has been speculated that haptoglobin may play a role in the inhibition of reverse cholesterol transport.
In the present study we investigated the effect of haptoglobin on the activity of another enzyme involved in reverse cholesterol transport, phospholipid transfer protein (PLTP). PLTP is a plasma protein that transfers phospholipids from triglyceride-rich lipoproteins such as very low-density lipoproteins (VLDL) and low-density lipoproteins (LDL) to high density lipoproteins (HDL) [12, 13]. PLTP occurs in plasma as two main forms: a high activity PLTP (HA-PLTP) and a low activity PLTP (LA-PLTP). HA-PLTP is associated with the majority of plasma PLTP activity. PLTP activity has been shown to be affected by its association to Apo-A1 [14, 15]. There is increasing evidence supporting the role of PLTP on atherosclerosis development . Moerland et al.,  showed in a transgenic mouse model of PLTP expression that an acutely increased PLTP expression resulted in a highly atherogenic lipoprotein profile. Shelly et al.,  found that the phospholipid transfer protein deficiency ameliorated diet-induced hypercholesterolemia and inflammation in mice. There is evidence that even a 10% reduction on PLTP activity can lead to a significant reduction of atherosclerosis progression , highlighting the role of PLTP on the development of cardiovascular disease.
In the present study we hypothesize that haptoglobin inhibits PLTP activity. This is based on the fact that PLTP activity is dependent on its binding to Apo-A1  and that haptoglobin has been shown to bind Apo-A1  and to inhibit LCAT activity . This work will further contribute to our understanding of the role of haptoglobin on modulating reverse cholesterol transport as well as the development of atherosclerosis.
Apolipoprotein A1 (μg/mL) and Haptoglobin (μg/mL) levels in hyperlipidemic (H1-H6) and normolipidemic (N1-N6) plasmas used in this study (mean ± SD).
3.9 ± 0.1
5897.7 ± 527
23.2 ± 1.9
5240.4 ± 110
275.7 ± 13
1245.7 ± 28
265 ± 26
4579.9 ± 259
185.8 ± 9
318.6 ± 14
167.3 ± 7
3891.1 ± 231
161.2 ± 4.9
500.5 ± 41
186.2 ± 16
6104.2 ± 134
201 ± 2.1
1480 ± 99
54.8 ± 3
4126.5 ± 124
269.9 ± 34
1322.7 ± 48
337.7 ± 28
3125.5 ± 48
Based on the model of LCAT inhibition by haptoglobin, we determined the relationship between PLTP activity and Ht/Apo-A1 ratio (Figure 2c) in hyperlipidemic plasma. A negative linear correlation was found (R2 = 0.69, p < 0.05) suggesting an inhibitory role of haptoglobin on PLTP activity in this group of plasmas. When a semi logarithmic non linear regression was used, we obtained a higher correlation (R2 = 0.998) than with the linear model.
Haptoglobin genotype has been shown to regulate reverse cholesterol transport in diabetes in vitro and in vivo . It has been proposed that an enhanced oxidative modification of serum lipoproteins (LDL and HDL) in individuals with the Hp2 genotype is an important determinant of accelerated atherosclerosis in these individuals . Interestingly, PLTP has been shown to efflux cholesterol Apo-A1 in murine macrophages .
Another mechanism by which haptoglobin may regulate reverse cholesterol transport is by inhibiting LCAT . Since this inhibition is mediated through the Hp-Apo-A1 interaction , we propose that other enzymatic activities regulated by Apo-A1 may be affected by haptoglobin levels. In particular, we are interested in PLTP which is an important enzyme involved in reverse cholesterol transport  and its activity has been shown to be dependent on its association with Apo-A1 . The positive correlation between PLTP activity and Apo-A1 levels in hyperlipidemic plasma persisted. However, we didn't observe this correlation in normolipidemic patients. A positive correlation was found between PLTP activity and Apo-A1 in plasma from patients with type 1 diabetes  which is consistent with our observations. These authors had previously demonstrated that patients with type 1 diabetes have a significantly elevated PLTP activity and that this activity is correlated with HDL levels . One of the differences between the two plasma groups that we used was the HDL content. We found a negative correlation (p = 0.018) between PLTP activity and HDL levels. This correlation was specific to HDL as it was not found with total cholesterol and triglycerides. Colhoun et al.,  showed differences between the correlation of PLTP activity and HDL particle size. PLTP activity negatively correlated with small HDL while it positively correlated with large HDL. Soro et al.,  also showed a negative correlation between HDL2 and PLTP activity. Since there is an association between reduced HDL particle size and hyperlipidemia , it is possible that in our hyperlipidemic patient group there is a higher small HDL/large HDL ratio than in normolipidemic controls and this contributes to a negative correlation between PLTP activity and HDL.
One of the limitations of our study is the sample size. However the correlations that we observed in the basal state were maintained even when exogenous Hp and Apo-A1 were added to the system. Nevertheless, Salvatore et al.,  found a correlation between the cholesteryl ester/cholesterol ratio (a measure of LCAT activity) and Hp/[Apo E +Apo-A1] ratio in a small number of multiple sclerosis patients (n = 9).
When we compared haptoglobin levels with PLTP activity, the correlation was insignificant (p = 0.08). However, when we changed our analysis to compare PLTP activity with Hp/Apo-A1 ratio, we did find a negative correlation suggesting a) that haptoglobin may inhibit PLTP activity and b) that Apo-A1 levels may affect this inhibitory interaction.
We further confirmed the Hp inhibitory effect by adding exogenous haptoglobin to the plasma samples. In hyperlipidemic plasma, PLTP activity was inhibited by haptoglobin addition in a dose-dependent way. This effect was not seen in normolipidemic plasma. This effect could also be related to the differences in low activity and high activity PLTP in hyperlipidemic and normolipidemic plasma and the differential association of Apo-A1 to these two forms of PLTP . The fact that PLTP activity is negatively correlated with the Hp/Apo-A1 ratio in hyperlipidemic plasma has not been previously reported.
PLTP activity was inhibited by Haptoglobin and Apo-A1 addition. Haptoglobin, HDL and PLTP activity correlation data suggests the potential to use haptoglobin as a biomarker for the development of atherosclerosis as well as a tool to understand the role of PLTP activity and haptoglobin levels in reverse cholesterol and atherosclerosis.
PLTP Activity Assay Kit's were obtained from Roar Biomedical (New York, NY, USA). Purified Hpt (at least 95% pure by SDS-PAGE) was purchased from Calbiochem (San Diego, CA). Apolipoprotein A1 was purchased from Sigma-Aldrich (St Louis, MI).
Twelve different human plasma samples (purchased from Bioreclamation [East Meadow, NY, USA]) were obtained from donors representing both normolipidemic plasma (N = 6) and hyperlipidemic plasma (N = 6) based on the standards set by the Ministry of Health and Welfare of Japan (cholesterol <220 mg/dl and triglycerides <150 mg/dl) . The cholesterol and triglyceride content of each one of these twelve samples has been reported previously .
Haptoglobin levels were determined using a two site Human Haptoglobin ELISA kit from ICL (Newberg, OR) as per the manufacturer instructions. Briefly, the control, standard and patient samples were added to the wells which had previously adsorbed the anti-Hp antibodies. The unbound proteins were removed by washing, and then anti-Hp antibodies conjugated to horseradish peroxidase were added. These enzyme-labeled antibodies form complexes with the previously bound plasma Hp. Following washings, a chromogenic substrate was added and absorbance was read at 450 nm. The concentration of Hp was determined using a standard curve of purified Hp.
The human Apo-A1 EIA kit was purchased from Cayman Chemical Company (Ann Arbor, MI). Briefly each well of the plate provided with the kit was coated with an Apo-A1 specific antibody. When the samples and controls were added to the wells any Apo-A1 would bind to these antibodies. After washings a new anti Apo-A1 antibody was added to detect the captured Apo-A1. Washings were followed by the addition of horseradish peroxidase conjugate that will recognize the complex. Upon washings and addition of a chromogenic substrate, the reaction was stopped with acid and absorbance was read at 450 nm. The intensity of the color is proportional to the concentration of Apo-A1 which was determined using a standard curve.
Each plasma sample was tested for PLTP activity using an activity assay that measures in vitro phospholipid transfer activity (Roar Biomedical, New York, NY). The PLTP assay was carried as per manufacturer instructions as described previously . Initially, the PLTP activity test was studied for different concentrations of plasma protein and it was decided to be 25 μg as the PLTP activity was linear in this range. PLTP activity was determined for plasma samples (25 μg). Controls were added in the form of a picomol standard to quantify PLTP activity and a blank control. Plates also measured plasma activity as a single entity in the PLTP kit. Measurements were taken for 6 different plasma samples in the hyperlipidemic range and 6 different plasma samples in the normolipidemic range. These samples were done in duplicate on each 96-well test plate and each test plate was repeated at least twice for each plasma sample.
The groups tested in this study were compared against each other by applying a repeated measure analysis of variance (ANOVA) test and blocking results in set plasmas to account for base PLTP activity variance. Statistical differences in the data was considered significant if the p value found was < 0.05. Data added for each plasma measure consisted of an N = 6 of which each experiment had at least a replicate value of 2.
The strength and direction of a linear relationship between two random variables was measured by the Pearson's coefficient of correlation as determined using SigmaStat™.
Phospholipid transfer protein
High density lipoprotein
Low density Lipoprotein.
K.M.W and C.G.L received funding from the Canadian Institutes of Health Research. The authors wish to thank Dr P. Gershkovich who critically read our manuscript before submission.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.