The Oxidation Hypothesis of Atherogenesis

For more than two decades, there has been continuing evidence of lipid oxidation playing a central role in atherogenesis. The oxidation hypothesis of atherogenesis has evolved to focus on specific pro-inflammatory oxidized phospholipids, which result from the oxidation of LDL phospholipids containing arachidonic acid, and which are recognized by the innate immune system in animals and humans. These oxidized phospholipids are largely generated by potent oxidants produced by the lipoxygenase and myeloperoxidase pathways. The failure of anti-oxidant vitamins to influence clinical outcomes may have many explanations including the inability of vitamin E to prevent the formation of these oxidized phospholipids and other lipid oxidation products of the myeloperoxidase pathway. Preliminary data suggest that the oxidation hypothesis of atherogenesis and the reverse cholesterol transport hypothesis of atherogenesis may have a common biological basis. The levels of specific oxidized lipids in plasma and lipoproteins, the levels of antibodies to these lipids, and the inflammatory/antiinflammatory properties of HDL may be useful markers of susceptibility to atherogenesis. ApoA-I and apoA-I mimetic peptides may promote both a reduction in oxidized lipids and enhance reverse cholesterol transport and therefore may have therapeutic potential. by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom LDL oxidation and artery wall cell cytotoxicity In 1979 Chisolm and colleagues (1) reported that oxidation of LDL was injurious to artery wall cells and suggested that low density lipoprotein (LDL) oxidation may be important in atherogenesis. Chisolm and colleagues also demonstrated that high density lipoprotein (HDL) inhibited the LDL-induced cytotoxicity (1). Over the ensuing two decades this group elucidated the basis for these observations and established the important role of oxidized cholesterol products especially cholesterol hydroperoxides (2). The search for mechanisms of LDL-induced foam cell formation It was also in 1979 when Brown and Goldstein reported that acetylated LDL but not native LDL was taken up by “scavenger receptors” instead of the LDL receptor resulting in cholesteryl ester accumulation in macrophages characteristic of foam cells (3). Since acetylation was not known to occur, following publication of this seminal manuscript there was a search for physiological ligands that would explain foam cell formation. Fogelman and Schecter and colleagues soon reported that malondialdehyde, an obligate product of the oxidation of arachidonic acid by the lipoxygenase pathways, could cause Schiff-base formation with the epsilon amino groups of apolipoprotein B lysines in LDL. The altered lipoprotein was recognized by macrophage scavenger receptors resulting in cholesteryl ester accumulation characteristic of foam cells (4). The following year Steinberg and colleagues demonstrated that cultured endothelial cells in a medium rich in metal ions and in the absence of the antioxidants present in serum could oxidize LDL to a form recognized by macrophage scavenger receptors (5). Three groups reported on the mechanism for this modification of LDL by artery wall cells in 1984 (6 by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom 8). In 1988 and 1989 manuscripts were published indicating that oxidatively modified LDL was present in the artery wall of animals and humans with atherosclerosis (9-11). Witztum, Steinberg, and Parthasarathy (12 –15) and Chisolm (16) emphasized that the cells of the artery wall transfer oxidative waste products into their membranes and secrete them into the subendothelial space. Parthasarathy (14) reported that the ability of the artery wall cells to oxidize LDL is directly related to their ability to "seed" the LDL with reactive oxygen species. Navab and colleagues (17,18) demonstrated in vitro, using human artery wall cell cocultures, that these cells were capable of creating microenvironments into which their oxidative waste could be secreted. At the same time, these microenvironments excluded aqueous antioxidants and allowed trapped LDL to undergo mild oxidation. What causes monocytes to enter the artery wall in the first place? Napoli and colleagues demonstrated that LDL accumulation and oxidation precede monocyte entry into the artery wall (21). Studying the arteries of human fetuses, they found that LDL was present and was oxidized prior to the entry of monocytes and that monocyte entry occurred at sites of oxidized LDL accumulation, strongly suggesting that LDL oxidation is a cause of monocyte entry and not just a by product of monocyte metabolism (21). Berliner and colleagues reported that mildly oxidized LDL, which was still recognized by the LDL receptor, stimulated human aortic endothelial cells to bind monocytes but not neutrophils, and caused artery wall cells to produce the potent monocyte chemoattractant, MCP-1, and the differentiation factor, M-CSF (22-28). Subsequently, it was found that the biological activity of this minimally modified LDL (MM-LDL) was due to the oxidation of LDL phospholipids, which contain arachidonic by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom acid in the sn-2 position (29-36). Direct measurement of these oxidized phospholipids in atherosclerotic lesions in animals suggested that they exist at concentrations that would be biologically active in vivo (31). Additionally, these oxidized phospholipids are generated in the membranes of cells that are under oxidative stress such as that caused by IL-1β (32) and are incorporated into apoptotic vesicles and blebs (37, 117). The oxidized phospholipids in MM-LDL are recognized by autoantibodies Witztum and colleagues demonstrated that an IgM monoclonal antibody isolated from apoE null mice, (EO6), 1) binds to epitopes in human and animal atherosclerotic lesions (10, 21), 2) binds to epitopes necessary for macrophage recognition of oxidized LDL (38), and 3) binds to apoptotic cells (39), also recognizes the oxidized phospholipids in MM-LDL (38). The epitope recognized by the EO6 antibody was found to be structurally and functionally identical to classic “natural” T15 anti-phospholipid antibodies that are of B-1 cell origin and are reported to provide protection from virulent pneumococcal infection (40). Subsequently it was shown that vaccination against Streptococcus pneumoniae protected LDL receptor null mice from atherosclerosis (41). Thus, the oxidized phospholipids found in MM-LDL and the antibodies that naturally occur to these oxidized phospholipids appear to be part of the innate immune system (42,43). Since sepsis is a major cause of fetal wastage and neonatal mortality, the presence of LDL-derived oxidized phospholipids that cause monocyte entry into the artery wall of the developing fetus and infant and the antibodies against these proinflammatory oxidized phospholipids, which also recognize bacteria and may provide a survival advantage. Indeed, Napoli and colleagues found that maternal hypercholesterolemia enhanced fatty streak formation in fetal aortas (21). If in fact, a by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom survival advantage is conferred by the presence of the LDL-derived pro-inflammatory oxidized phospholipids, the antibodies against these oxidized lipids, and the monocytes that are induced to migrate as a result of the presence of these oxidized lipids, and this survival advantage was enhanced by maternal hypercholesterolemia, there would be positive evolutionary pressure for hypercholesterolemia. This might explain why hypercholesterolemia is so prevalent in human populations. What are the mechanisms that cause the formation of the oxidized phospholipids found in MM-LDL? A role for cell-generated reactive oxygen species in mediating LDL oxidation was postulated for many years (12-16). The first direct proof in vivo of a role for the lipoxygenase pathway was provided by Funk and colleagues in apoE null mice that were made genetically deficient in 12/15 lipoxygenase and were found to have significantly less atherosclerosis as a result (44-46). Deletion of the 12/15 lipoxygenase gene in LDL receptor null mice (47) and in macrophages of mice that were both LDL receptor null and also deficient in the apolipoprotein B editing catalytic polypeptide-1 enzyme (48) resulted in decreased atherogenesis in both mouse models. Conversely, overexpression of the 12/15 lipoxygenase gene in the endothelium of LDL receptor null mice accelerated atherosclerosis (49). Moreover, Hedrick and colleagues generated transgenic mice on a C57BL/6J background that modestly overexpressed the murine 12/15 lipoxygenase gene. These mice had 2.5-fold elevations in the levels of 12(S)-HETE and a 2-fold increase in the expression of 12/15 lipoxygenase protein in vivo. These mice developed spontaneous aortic fatty streak lesions on a chow diet further indicating the importance of the lipoxygenase pathway in atherogenesis (50). Navab and colleagues found that human by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom artery wall cells required the 12/15 lipoxygenase gene in order to generate the oxidized phospholipids found in MM-LDL (51). Myeloperoxidase null mice had increased, not decreased atherosclerois (112), but mouse macrophages unlike human macrophages have little myeloperoxidase. Myeloperoxidase reaction products have been found in human atherosclerotic lesions (52) and there is increasing evidence that the myeloperoxidase pathway can generate proinflammatory oxidized lipids (34, 53-58). Another pathway that may be a potential source of reactive oxygen species for the generation of pro-inflammatory oxidized lipids is the NADPH oxidase pathway (59, 60). Figure 1 depicts the formation of the fatty acid hydroperoxides HPODE and HPETE (Panel A), the action of HPODE on an LDL-derived phospholipid, PAPC (Panel B) and the formation of three of the oxidized phospholipids in MM-LDL (Panel C). Challenges to the Oxidation Hypothesis of Atherogenesis The oxidation hypothesis of atherogenesis includes a large number of oxidized lipids. As indicated in the title of this manuscript, this review is focused on oxidized phospholipids and HDL. For a more broadly focused perspective, the reader is referred to a number of excellent recent reviews (34, 35, 43, 57, 70, 120). Most of the in vitro studies on LDL oxidation were carried out with media containing relatively high levels of free metal ions. Heinecke appropriately questioned the applicability of these in vitro studies conducted in media rich in free metal ions to the situation in vivo (61). As shown in the experiments described in Figure 2, a medium rich in free metal ions is not required for LDL oxidation or induction of monocyte chemotactic activity in the artery wall coculture system described by Navab et al (18). by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom Heinecke (61) and Stocker and colleagues (62) also questioned the oxidation hypothesis based on the findings by Chan and colleagues (63) that rabbits genetically engineered to specifically overexpress 15 lipoxygenase in their macrophages actually had less atherosclerosis. Additionally Stocker and colleagues treated Watanabe Heritable Hyperlipidemic (WHHL) rabbits with antioxidants (probucol and its metabolite bisphenol) resulting in decreased aortic cholesterol ester and triglyceride peroxides without a decrease in aortic atherosclerosis in the bisphenol treated animals but with a statistically significant 31% decrease in en face lesions in the probucol treated group (62). The findings of Chan and colleagues (63) need to be considered in the context of the studies of Zhao et al demonstrating that deficiency of the 12/15 lipoxygenase pathway in macrophages of mice that were LDL receptor null and also deficient in the apolipoprotein B editing catalytic polypeptide-1 enzyme resulted in significantly less atherosclerosis (48). Additionally, Harats et al (49) and Reilly et al (50) demonstrated that overexpression of 15 lipoxygenase in the endothelium of mice increased atherosclerosis. Conversely, Cyrus et al (44,46) demonstrated that disruption of the 12/15 lipoxygenase gene decreased atherosclerosis in apoE null mice and George et al showed similar results in LDL receptor null mice (47). Moreover, specific pharmacological inhibition of 15 lipoxygenase in rabbits significantly reduced atherosclerosis (64, 118). Thus, on balance the report by Chan and colleagues is not strong evidence against the hypothesis. With regard to the findings of Stocker and colleagues (62) that antioxidants decreased lesion cholesteryl ester and triglyceride peroxides without decreasing the by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom cellular and other extracellular components of atherosclerosis in WHHL rabbits, it must be noted that these investigators did not measure, directly or by immunologic means, the aortic content of the specific pro-inflammatory oxidized phospholipids (e.g. those recognized by the EO6 antibody). The data of Stocker and colleagues (62) would only speak against the oxidation hypothesis if they had shown that these specific proinflammatory oxidized phospholipids had been reduced without a reduction in aortic atherosclerosis, and they did not provide such evidence (62). Similarly, the failure of vitamin E and other oral antioxidants to reduce measures of atherosclerosis in human clinical trials (65-67) must also be considered in this light. None of these trials provided evidence that the levels of the specific pro-inflammatory oxidized phospholipids were decreased by the oral antioxidants tested (68,69). Indeed, these trials did not even show that administration of the antioxidants reduced lipid oxidation in the test subjects (68, 69). Moreover, FitzGerald and colleagues (119) demonstrated that doses of vitamin E as high as 2000 IU/d for 8 weeks in human volunteers failed to reduce the excretion of specific isoprostanes that are a marker of oxidized phospholipids. Heinecke (70) has indicated that vitamin E does not prevent lipid oxidation mediated by myeloperoxidase in vitro. Indeed, myeloperoxidase might actually convert an anti-oxidant to a pro-oxidant reaction (120). Navab et al (18) reported that vitamin E could not inhibit the oxidation of LDL nor prevent LDL-induced monocyte chemotactic activity in a coculture of human artery wall cells. As shown in Figure 3, vitamin E also does not inhibit the lipoxygenase product 13(S)-HPODE from oxidizing the LDL-derived arachidonic acid-containing phospholipid, PAPC. Thus, vitamin E is not able to prevent by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom the formation of the specific pro-inflammatory oxidized phospholipids (71). Navab et al (51) reported that 13(S)-HPODE was greater than 200 times more potent in oxidizing PAPC and forming these specific pro-inflammatory oxidized phospholipids than was hydrogen peroxide. An example of the danger of extrapolating clinical trial data comes from a review of the history of the acceptance of thrombosis as an important component of the clinical events associated with atherosclerosis. Before 1950, acute myocardial infarction was assumed to be due to coronary thrombosis. During the 1950s large-scale clinical trials used warfarin to prevent heart attack. The results were at best confusing. In this milieu of scientific confusion, a prominent pathologist suggested that thrombosis in myocardial infarction was a post-mortem artifact. As a result, in the1960s and 1970s acute myocardial infarction was thought to be due to supply-demand imbalance in the myocardium. It was thought that a chronic narrowing of the coronary arteries due to atherosclerosis limited blood flow, and that when demand increased without a concomitant increase in flow, myocardial infarction resulted. It was not until the advent of coronary arteriography during acute myocardial infarction and the introduction of potent thrombolytic agents and direct angioplasty for acute myocardial infarction in the 1980s and 1990s that the field came to clearly see the primary role that thrombosis plays in acute myocardial infarction. We now know that warfarin is an inadequate agent for preventing arterial thrombosis in which platelets play a predominant role. Concluding that lipid oxidation is not central to the development of atherosclerosis because of the failure of the antioxidants used to date in clinical trials could lead to an erroneous by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom conclusion similar to the false belief that thrombosis was not important in the genesis of myocardial infarction because warfarin was ineffective in initial clinical trials. Furthermore, genetic studies in humans and in mice are consistent with a role of lipid oxidation in atherogenesis. Paraoxonase1 (PON1) can prevent the formation of and destroy the pro-inflammatory oxidized phospholipids in vitro, and PON1 null mice have increased atherosclerosis (98, 99) whereas PON1 transgenic mice have decreased atherosclerosis (100). Also, over 20 epidemiologic studies have shown an association between polymorphisms of PON1 and heart disease in human populations (113). Additionally, studies modulating other antioxidant genes such as heme oxygenase-1 (114-116), suggest that lipid oxidation is a primary cause of inflammation in atherosclerosis. The Role of HDL in Atherogenesis In contrast to vitamin E, HDL, apoA-I and apoA-I mimetic peptides have been shown to prevent LDL oxidation in cell-free systems (1, 13,14) and in the artery wall coculture studies of Navab et al (20, 51). Moreover, HDL, apoA-I and apoA-I mimetics have been shown to decrease lesions and improve vascular reactivity in animal models of atherosclerosis (73-80), and in humans (81-83). The primary mechanism by which HDL and apoA-I and apoA-I mimetic peptides exert their beneficial effect has been presumed to be due to enhancement of reverse cholesterol transport (75, 84-86). However, apoA-I has also been shown to be capable of removing “seeding molecules” from LDL, thus preventing the oxidation of LDL-derived phospholipids to those which are thought to be responsible for the inflammatory response characteristic of atherosclerosis (20, 51). by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom Sevanian and colleagues noted that a subpopulation of freshly isolated LDL contains lipid hydroperoxides (87). Navab et al (20) found that freshly isolated LDL from normal individuals always contained small amounts of lipoxygenase pathway products (e.g. HPODE, HPETE). These were present even when blood was collected into tubes containing potent antioxidants. The levels of HPODE and HPETE did not increase during in vitro incubations in the presence of these antioxidants indicating that they were present in LDL in vivo (20). However, when the freshly isolated LDL was incubated with apoA-I in the presence of antioxidants and the LDL and the apoA-I were then rapidly separated, the LDL treated with apoA-I contained only about one-third to one-half as much HPODE and HPETE as was present initially (20). Prior to these incubations the apoA-I contained no detectable HPODE or HPETE, but after the incubation with LDL half to two-thirds of the HPODE and HPETE that had been present in the LDL was transferred to the apoA-I along with some cholesterol and phospholipid (20). The LDL treated with apoA-I was unable to generate lipid hydroperoxides, nor was it able to induce monocyte adherence or monocyte chemotactic activity when added to human artery wall cocultures (20). If the apoA-I that was incubated with the LDL was subjected to lipid extraction and the extracted lipids were added back to the LDL that had been treated with apoA-I, the reconstituted LDL was able to induce lipid hydroperoxide formation and induce monocyte adherence and monocyte chemotactic activity (20). Consistent with these properties of apoA-I, Bowry et al (88) reported that HDL is a major carrier of lipid hydroperoxides in humans. As shown in Figure 4, HDL appears to be the major carrier of lipid hydroperoxides in mice and the concentration of lipid hydroperoxides in HDL taken from the atherosclerosis susceptible C57BL/6J mice either by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom on a low-fat chow diet or on an atherogenic diet was significantly greater than the lipid hydroperoxide levels found in the HDL of the atherosclerosis resistant C3H/HeJ mice. Six hours after injection of human apoA-I (100 μg/mouse) into the tail veins of C57BL/6J mice their LDL was no longer able to induce lipid hydroperoxide formation or induce monocyte chemotactic activity in human artery wall cocultures (20). In contrast injection of human apoA-II (100 μg/mouse) or saline did not prevent LDLmediated induction of lipid hydroperoxide formation or LDL-induced monocyte chemotactic activity (20). Similarly, six hours after infusing apoA-I and phospholipid into healthy human volunteers, there was a dramatic decrease in the ability of their LDL to induce lipid hydroperoxide formation and monocyte chemotactic activity in the cocultures in six out of six subjects (20). Thus, apoA-I has the ability to remove HPODE and HPETE from human LDL and to dramatically reduce the inflammatory properties of LDL in both mice and humans (20). HDL Inflammatory Index The ability of apoA-I to promote reverse cholesterol transport and remove the “seeding molecules” from LDL that are necessary for the formation of the inflammatory LDL-derived oxidized phospholipids likely contributes to the inverse relationship between HDL-cholesterol levels and susceptibility to atherosclerosis. While the inverse relationship between HDL-cholesterol levels and risk for acute myocardial infarction is highly significant, the relationship is far from perfect. In the AFCAPS/TexCAPS study, subjects with “average” total cholesterol levels were followed for an average of 5.2 years. Of those given placebo, the event rate during the study was by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom 2.1%, 2.9%, and 3.4% for those with HDL-cholesterol levels of >40 mg/dL, 35-39 mg/dL, and <34 mg/dL, respectively. While the differences were highly significant, knowledge of the HDL-cholesterol level in predicting whether a specific individual would or would not have an event was clearly of limited use (90, 91). Similarly, in the original Framingham study, the incidence of coronary heart disease was compared with HDL-cholesterol levels (92). With minimal assumptions, one can calculate from the published data that 44% of the events occurred in men with HDLcholesterol levels of ≥ 40 mg/dL and 43% of the events occurred in women with HDLcholesterol levels ≥ 50 mg/dL (91). Since a significant number of coronary heart disease events occur in patients with normal LDL-cholesterol levels and normal HDL-cholesterol levels (91-93), there has been a continuing search for markers with better predictive value (91, 94). One of the first indications that HDL might be such a marker came from the studies of Van Lenten et al (95). They reported that the acute phase response in humans converted HDL from anti-inflammatory to pro-inflammatory (95). These studies (95) compared HDL taken from humans before and after elective surgery. Prior to surgery HDL was anti-inflammatory in a human artery wall cell coculture model, (i.e. HDL inhibited LDL oxidation and inhibited LDL-induced monocyte chemotactic activity). However, at the peak of the acute phase response, 3 days after surgery, HDL from the same patient was pro-inflammatory (promoted LDL oxidation and monocyte chemotactic activity in the human artery wall coculture). One week after surgery the HDL returned to an anti-inflammatory state (91, 95). These changes in HDL are consistent with a classic acute phase response. Gabay and Kushner (96) emphasized that the acute phase response by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom can become chronic. Navab et al demonstrated that such a chronic acute phase response was present in apoE null mice on a chow diet and in LDL receptor null mice on a high-fat diet and was accompanied by pro-inflammatory HDL (72). In addition to apoliporoteins such as apoA-I, HDL also contains enzymes that can prevent the formation or destroy the oxidized phospholipids that mediate the inflammatory response induced by MM-LDL (97). These enzymes include paraoxonase (98-100), platelet activating factor acetylhydrolase (101), lecithin cholesterol acyltransferase (102), and possibly glutathione peroxidase (97). In vivo, the absence of the HDL-associated enzyme paraoxonase was demonstrated to result in increased LDL oxidation and increased atherosclerosis (99). Van Lenten et al demonstrated that the acute phase response in rabbits and humans resulted in decreased activities of HDL-associated paraoxonase and platelet activating factor acetylhydrolase (95). An example of a chronic acute phase response in humans is the persistent elevation (highest tertile) of the positive acute phase reactant C reactive protein (CRP) in the absence of a detectable infection or other acute stress (94). Navab et al (71) reported that the inflammatory/anti-inflammatory properties of HDL from 27 normolipidemic coronary heart disease (CHD) patients clearly separated the patients from 31 ageand gender-matched controls. Each of the patients had at least a 50% narrowing of a coronary artery, none smoked, none were diabetic, none were taking hypolipidemic medications, and all had normal blood lipids as defined by total cholesterol <200 mg/dL, LDL-cholesterol < 130 mg/dL, triglycerides < 150 mg/dL, HDL-cholesterol >40 mg/dL for males, and >50 mg/dL for females. The patient HDL, in contrast to control HDL, was pro-inflammatory in the human artery wall cell coculture by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom model and in a cell-free assay (71). These patients had no evidence of an acute illness that could explain an acute phase response. Thus, Navab et al (71) postulated that the inflammatory properties of HDL in these patients represented a “chronic” acute phase response similar to that described by CRP levels in the top tertile of “normal” (94). Ansell et al (91) studied two additional patient groups. Group I comprised 26 patients who presented with stable CHD or CHD equivalents by NCEP ATP-III criteria (103) that were not yet on a statin or another hypolipidemic agent and whose physician recommended treatment with a statin (91). The inflammatory/anti-inflammatory properties of HDL from these patients was compared before and six weeks after starting statin therapy. A second group of patients were studied who presented with high HDLcholesterol levels and documented CHD (91). The inflammatory/anti-inflammatory properties of the HDL from both groups of patients were compared to age and gender matched healthy controls using an HDL inflammatory index (91). The HDL inflammatory index was measured in two ways. The first was a cell-based assay in which the monocyte chemotactic activity generated in a human artery wall cell coculture by a standard control LDL was determined in the absence and presence of the test HDL (91). The monocyte chemotactic activity determined in the absence of the test HDL was normalized to 1.0. The monocyte chemotactic activity obtained when the test HDL was added was then compared to the monocyte chemotactic activity in the absence of HDL. If the value after adding the test HDL was greater than that obtained without the test HDL (i.e. HDL inflammatory index >1.0) the test HDL was classified as pro-inflammatory. On the other hand, if the value obtained after addition of the test HDL was less than that without the test HDL (i.e.< 1.0) the HDL was classified as anti-inflammatory (91). by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom Ansell et al also used a cell-free assay in which PEIPC (see Figure 1C) was added to a cell-free system containing DCFH (similar to the system described in Figure 3) (91). PEIPC was chosen because this oxidized phospholipid accounts for more than 80% of the LDL-induced monocyte chemotactic activity in human artery wall cell cocultures (i.e. 80% of the monocyte chemotactic activity resulting from the addition of LDL to the cocultures is due to the formation of PEIPC). The fluorescent signal generated by PEIPC in the absence of the test HDL was normalized to 1.0. If the signal increased after addition of the test HDL (i.e >1.0) the test HDL was classified as pro-inflammatory. If the fluorescent signal decreased after addition of the test HDL (i.e.<1.0), then the HDL was classified as anti-inflammatory (91). As shown in Figure 5A the patients in Ansell’s Group I had pro-inflammatory HDL prior to statin therapy. After six weeks of simvastatin (40 mg/day) (91), their HDL was less pro-inflammatory but was still pro-inflammatory. In contrast, the healthy age and gender-matched controls had anti-inflammatory HDL. Figure 5B demonstrates similar results for the cell-free assay. Figure 6 shows the HDL lipid hydroperoxide content for Ansell’s Group I patients and controls. The lipid hydroperoxide content of the patient’s HDL was significantly greater than that of the healthy controls (Figure 6). After six weeks of simvastatin therapy there was a trend to lower levels of HDL lipid hydroperoxides but this did not reach significance (p=0.07) (91). Figure 7 demonstrates that the HDL from the Group II patients, 20 patients with HDL-cholesterol levels ranging between 84 mg/dL and 148 mg/dL, none of whom were taking a statin or other hypolipidemic medication and all of whom had documented CHD (Group II in reference 91), was pro-inflammatory. by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom In Ansell’s study the HDL inflammatory index separated patients from the healthy controls much better than HDL-cholesterol levels (91). In Group I, only 3 of 26 patients had abnormally low HDL-cholesterol levels prior to simvastatin therapy while 20 of 26 patients had an HDL inflammatory index ≥ 1.0 and all 26 patients had an HDL inflammatory index > 0.6, compared to all 26 of the controls whose values were <1.0 by the coculture assay (91). Twenty-four of the 26 controls had an HDL inflammatory index < 0.6 by the coculture assay (91). After six weeks of simvastatin therapy there was a highly significant reduction in the inflammatory properties of the HDL in Group I patients, but their HDL remained significantly more inflammatory than HDL from controls (91). In patient Group II, which was comprised of patients referred for study because of CHD and high HDL-cholesterol levels (95±14 mg/dL), only one patient had an elevated LDL-cholesterol level (>160 mg/dL), only two patients had elevated triglycerides (>150 mg/dL), and none were diabetic (91). None of these 20 patients were on a statin or other hypolipidemic agent. Eighteen of these 20 patients had an HDL inflammatory index ≥ 1.0 and only one had an HDL inflammatory index < 0.6, while all 20 of the controls had an HDL inflammatory index < 0.6 by the coculture assay (91). Since there were significant correlations between HDL-lipid hydroperoxides (LOOH), and the HDL inflammatory index determined in either the coculture or the cellfree assay (91), one might ask if the HDL inflammatory index is simply a measure of HDL-LOOH? If this were the case, adding normal HDL would always give a value > 1.0 in the cell-free assay since all of the controls had some LOOH in their HDL (91). However, of the 46 healthy control subjects studied by Ansell and colleagues (26 for by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom Group I and 20 for Group II) only one had an HDL inflammatory index >1.0 as determined by the cell-free assay. Navab et al (97) reported that the HDL inflammatory index measures the net action of a large number of factors in HDL. These factors include oxidized phospholipids, lipid hydroperoxides, paraoxonase activity, platelet activating factor acetyl hydrolase activity, lecithin cholesterol acyl transferase activity, possibly GSH peroxidase activity, apoA-I, apolipoprotein J, serum amyloid A, ceruloplasmin, antioxidant vitamins, and probably products such as nitrotyrosine, which can be generated by myeloperoxidase. Thus, the HDL inflammatory index likely represents the net effect of all of these factors. The LOOH content and the other factors in HDL are likely interdependent. The correlation of LOOH content with HDL that is less able to prevent the formation of the LDL-derived pro-inflammatory oxidized phospholipids (91) suggests that oxidation impairs HDL function. Macdonald et al (104) reported that oxidation of HDL per se does not make HDL less functional. Macdonald and colleagues (104) reported that tyrosyl radical oxidation of mouse HDL induced formation of apoAI-AII heterodimers (tyrHDL), which enhanced the ability of mouse HDL to deplete cultured fibroblasts of their regulatory pool of cholesterol. When apoE null mice were injected intraperitoneally twice weekly with 150 μg tyrHDL from age 10 to 18 weeks there was a maximum 2.3-fold increase in endogenous HDL cholesterol levels, which fell toward the end of the treatment period. Treatment with tyrHDL resulted in 37% less aortic lesion development than in control HDL-treated mice (P<0.001) and 67% less than in salineinjected animals (P<0.001) (104). Macdonald et al hypothesized that increasing the by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom mobilization of cellular cholesterol to apoAI could decrease atherosclerosis without necessarily causing sustained increases in circulating HDL cholesterol levels (104). In an accompanying editorial, Bergt, Oram, and Heinecke (105) suggested that cross-linked heterodimers of apoA-I and apoA-II in tyrosylated HDL may be responsible for its enhanced ability to remove cholesterol from lipid-laden cells (106). These authors (105) speculated that heterodimers may act more like lipid-free apolipoproteins, perhaps because of conformational changes in apoA-I that expose more amphipathic α-helices to ABCA1. They also noted that apoA-I mimetic peptides synthesized from D-amino acids appear to retard atherogenesis in hypercholesterolemic mice without affecting HDL levels and that in vitro studies have suggested that this effect is mediated in part by inhibition of LDL oxidation (77). Bergt, Oram, and Heinecke (105) suggested the possibility that such apoA-I mimetic peptides (77) might also promote reverse cholesterol transport but this had not been investigated (105). Additionally, Bergt, Oram, and Heinecke (105) commented that HDL appears to be the major carrier of lipid hydroperoxides in the plasma, and may transport oxidized cholesteryl esters to the liver for excretion (107). They concluded that HDL might also lower the risk for atherosclerosis by altering the structures or metabolic fates of oxidized lipids that would otherwise exert atherogenic effects (105). Is there a relationship between the inflammatory properties of HDL and the ability of HDL to mediate cellular cholesterol efflux? Grunfeld and colleagues (108) reported that HDL taken from Syrian hamsters in which an acute phase reaction was induced by LPS injection was less able to mediate cellular cholesterol efflux than was HDL taken from control animals. by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom Pussinen et al (109) reported that HDL-mediated cholesterol efflux significantly improved in patients with periodontitis who were PCR positive for Actinobacillus actinomycetemcomitan and whose CRP decreased (53.7% reduction, p=0.015) after treatment. Pussinen et al (109) concluded that periodontitis causes similar, but milder changes in HDL metabolism than those that occur during the acute phase response, and that periodontitis may diminish the anti-atherogenic potency of HDL. Reddy and colleagues (110) reported that LDL oxidation by human artery wall cells was controlled by the cholesterol content of the cells, which was determined in part by ABCA1 activity. Reddy et al proposed that the reverse cholesterol transport hypothesis of atherogenesis and the oxidation hypothesis of atherogenesis might be two sides of the same coin (110). The preliminary data shown in Figure 8 are consistent with the proposal by Reddy et al (110). A single oral administration of D-4F, an oral apoA-I mimetic peptide, to cynomologous monkeys at doses as low as 1.67 mg/kg, reduced plasma and lipoprotein lipid hydroperoxides two hours after the dose (Figure 8A). Two hours after a single oral administration of D-4F at 1.67 mg/Kg, the monkey HDL contained 118 picomoles of D-4F per mg HDL-cholesterol (Figure 8B). Concomitant with the fall in lipoprotein lipid hydroperoxides, there was a significant improvement in the monkey HDL antiinflammatory properties (Figure 8C) and a decrease in the ability of the monkey LDL to induce human artery wall cells to produce monocyte chemotactic activity (Figure 8D). Paralleling these changes was a dramatic increase in the ability of the monkey HDL to promote cholesterol efflux from human monocyte macrophages (Figure 8E). by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom Summary and Conclusions Scientific controversy on subjects such as the role of thrombosis in acute myocardial infarction and the oxidation hypothesis of atherogenesis may be taken in a broader human context. In speaking to the House of Commons on the 12th of November of 1940 on the occasion of the passing of Neville Chamberlain, Winston S. Churchill said, “At the lychgate we may all pass our own conduct and our own judgments under a searching review. It is not given to human beings, happily for them, for otherwise life would be intolerable, to foresee or to predict to any large extent the unfolding course of events. In one phase men seem to have been right, in another they seem to have been wrong. Then again, a few years later, when the perspective of time has lengthened, all stands in a different setting. There is a new proportion. There is another scale of values. History with its flickering lamp stumbles along the trail of the past, trying to reconstruct its scenes, to revive its echoes, and kindle with pale gleams the passion of former days. What is the worth of all this? The only guide to a man is his conscience; the only shield to his memory is the rectitude and sincerity of his actions. It is very imprudent to walk through life without this shield, because we are so often mocked by the failure of our hopes and the upsetting of our calculations; but with this shield, however the Fates may play, we march always in the ranks of honour.” There is continuing evidence of a role for lipid oxidation in atherogenesis. The oxidation hypothesis of atherogenesis has evolved to focus on specific pro-inflammatory oxidized phospholipids, which result from the oxidation of LDL phospholipids containing arachidonic acid, and which are recognized by the innate immune system in animals and humans. These oxidized phospholipids are generated via the lipoxygenase by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom and myeloperoxidase pathways. Among many possible reasons, the failure of antioxidant vitamins to influence clinical outcomes may have resulted because of the failure of vitamin E to prevent the formation of these pro-inflammatory oxidized phospholipids. Preliminary data suggest that the oxidation hypothesis of atherogenesis and the reverse cholesterol transport hypothesis of atherogenesis may be linked. Measurements of the HDL inflammatory index and measurements of the levels of oxidized lipids in lipoproteins and products of the myeloperoxidase pathway may prove to predict susceptibility to atherogenesis. ApoA-I and apoA-I mimetic peptides may reduce oxidized lipids and also improve reverse cholesterol transport and, therefore, may have therapeutic potential. by gest, on O cber 1, 2017 w w w .j.org D ow nladed fom Acknowledgments This work was supported in part by USPHS grants HL 30568, HL 34343, GCRC grant RR 00865, Mr. Donald Kahn, The Joseph F. Troy Research Foundation, Laubisch, Castera, and M.K. Grey Funds at UCLA. bygest,onOcber1,2017www.j.orgDownladedfom References1.Hessler JR, Robertson AL, Chisolm GM. LDL-induced cytotoxicity and its inhibition by HDL in human vascular smooth muscle and endothelial cells in culture.Atherosclerosis 1979; 32:213–229.2. Colles SM, Maxson JM, Carlson SG, Chisolm GM. Oxidized LDL-induced injuryand apoptosis in atherosclerosis. Potential roles for oxysterols. Trends Cardiovasc Med2001;11:131–138.3. Goldstein JL, Ho YK, Basu SK, Brown MS. Binding site on macrophages thatmediates uptake and degradation of acetylated low density lipoprotein, producingmassive cholesterol deposition. Proc Natl Acad Sci U S A. 1979; 76:333-337. 4. Fogelman AM, Shechter I, Seager J, Hokom M, Child JS, Edwards PA.Malondialdehyde alteration of low density lipoproteins leads to cholesteryl esteraccumulation in human monocyte-macrophages. Proc Natl Acad Sci U S A.1980;77:2214-2218. 5. Henricksen T, Mahoney EM, Steinberg D. Enhanced macrophage degradation of lowdensity lipoprotein previously incubated with cultured endothelial cells: recognition byreceptor for acetylated low density lipoproteins. Proc Natl Acad Sci U S A.1981;78:6499-6503. 6. Morel DW, DiCorleto PE, Chisolm GM. Endothelial and smooth muscle cells alterlow-density lipoprotein in vitro by free radical oxidation. Arteriosclerosis. 1984;4:357-