Influence of serum amyloid A on the decrease of high density
lipoprotein-cholesterol in active sarcoidosis
Albert Salazar
a,*, Juan Man˜a
a, Concepcion Fiol
b, Isabel Hurtado
b,
Josep M. Argimon
c, Ramon Pujol
a, Xavier Pinto
aaInternal Medicine Ser6ice,Ciutat Sanita`ria de Bell6itge,Uni6ersity of Barcelona,C/Consell de Cent218,3°1a,08011Barcelona,Spain
bExperimental Research Unit,Ciutat Sanita`ria de Bell6itge,Uni6ersity of Barcelona,Barcelona,Spain cEpidemiological Unit,Ser6eis Sanitaris Costa de Ponent,Barcelona,Spain
Received 3 March 1999; received in revised form 28 October 1999; accepted 7 December 1999
Abstract
Objecti6e: We have previously observed low levels of high density lipoprotein (HDL) cholesterol in active sarcoidosis. The aim of this study was to analyze the role of serum amyloid A (SAA) on this lipid disorder.Methods: Eighty five untreated sarcoid patients, 40 with active disease and 45 with inactive disease, were recruited. Sarcoidosis activity was evaluated by means of clinical, chest X-ray, gallium-67 scan, serum angiotensin converting enzyme (peptidyl-dipeptidase A) values, and pulmonary function tests. Analysis of lipoprotein metabolism included: serum cholesterol, low density lipoprotein (LDL)-cholesterol, HDL-cholesterol, HDL2-cholesterol, HDL3-cholesterol, apolipoprotein A-I (apo A-I), apolipoprotein B (apo B), and triglyceride concentrations. Serum amyloid A protein and lecithin-cholesterol acyltransferase (LCAT) activity were measured.Results: In active sarcoidosis we found significantly reduced levels of HDL-cholesterol (1.1790.36 vs. 1.4490.39 mmol/l, P=0.002), HDL3-cholesterol (0.7890.23 vs. 1.0290.21 mmol/l, PB0.0001), and apo A-I (1.3690.29 vs. 1.6190.27 g/l, PB0.0001) and significantly increased levels of triglyceride (1.5190.64 vs. 1.0390.46 mmol/l, PB0.0001), and apo B (1.1490.25 vs. 0.9990.27 g/l, P=0.012) versus inactive sarcoidosis. Serum amyloid A concentrations were significantly increased in the patients with active disease (155.459154.01 mg/ml) compared to the inactive sarcoid patients (89.70965.36 mg/ml) (P=0.011). There were no significant differences in cholesterol, LDL-cholesterol, HDL2-cholesterol or LCAT values between groups. Multivariate logistic regression analysis showed that HDL-cholesterol (regression coefficientb= −1.96; S.E.=0.87; P=0.02) and SAA (regression coefficient b=0.01; S.E.=0.004; P=0.01) were the two variables independently associated with disease activity. Moreover, a significant negative correlation was observed between SAA levels and both HDL-cholesterol (r= −0.39; P=0.01) and apo A-I (r= −0.35; P=0.03) levels, in the active sarcoid group. Conversely, no correlation was found in the inactive sarcoid group. Conclusion: The low HDL-cholesterol and apo A-I concentrations seen in active sarcoid patients are associated with a significant increase of SAA levels. We suggest that the displacement of apo A-I by SAA on HDL accounts for the lower level of HDL-cholesterol seen in active sarcoidosis. © 2000 Elsevier Science Ireland Ltd. All rights reserved.
Keywords:Sarcoidosis; Cholesterol; Activity; Serum amyloid A
www.elsevier.com/locate/atherosclerosis
1. Introduction
Sarcoidosis is a multisystemic granulomatous disease of unknown etiology that is characterized by the prolif-eration of monocyte-macrophages and the formation of noncaseating epithelioid granulomas that most often affect the lymph nodes, lungs, eyes, skin, liver and spleen [1]. The activated reticuloendothelial system
could alter serum lipoprotein concentrations, as occurs with thermolysin-like metalloendopeptidase and an-giotensin converting enzyme [2]. Some authors have already described a relationship between diseases in-volving monocyte-macrophages and hypocholes-terolemia in patients with myeloproliferative disorders [3], Gaucher type I disease [4] and rheumatoid arthritis [5,6]. Kindman et al. [7] first published on the subject of lipid metabolism in sarcoid patients and found reduced serum values of cholesterol and HDL-cholesterol, though there was no comparison between clinically * Corresponding author. Tel.: +34-3-4516315.
active and inactive disease. We have already reported that the decrease in HDL-cholesterol that is observed in patients with sarcoidosis is limited to those with active disease [8]. However, the underlying mechanism of this phenomena has not yet been directly studied. Serum amyloid A protein is an acute-phase reactant of hepatic synthesis that may be transiently associated with HDL during acute inflammation and may potentially serve as a transient cholesterol-binding protein [9]. Serum amy-loid A may act either displacing apo A-I, which may result in increased catabolism of HDL, or inhibiting LCAT activity, which leads to low levels of esterified serum cholesterol [10,11]. To our knowledge, there is only one description in the medical literature about SAA in sarcoid patients that found elevated SAA lev-els, though there was neither comparison between ac-tive and inacac-tive disease nor study of lipid metabolism [12].
The aim of the present study was to analyze the role of SAA protein in the decrease of HDL-cholesterol observed in patients with active sarcoidosis.
2. Patients and methods
2.1. Patients
The diagnosis of sarcoidosis was made according to the following criteria: (1) a compatible clinical and radiological picture; (2) histological demonstration of noncaseating granulomas in one or more tissues with negative stains and cultures for mycobacteria and fungi,
or a positive Kveim test; and (3) exclusion of other granulomatous diseases. A total of 85 consecutive out-patients (ten men and 75 women) from a sarcoid clinic were included in the study. None of the patients was acutely ill, nor was receiving any steroids, B-blocking agents, thiazide diuretics, hormone replacement therapy or oral contraceptives at the time of the study. None of the patients was obese nor was on a hypolipidemic diet or losing weight. Alcohol intake in sarcoid patients was minimal. Patients were excluded if they had diabetes, hyperlipidemia or coronary artery disease, or if they had liver or kidney impairment or showed clinical signs of neoplastic disease or thyroid dysfunction, after un-dergoing a careful physical examination. As our pa-tients were not elderly and papa-tients with moderate or severe hyperlipidemia were excluded from the study, TSH was not routinary measured. All patients were matched for age, sex, and body mass index (Table 1) and met the clinical criteria just listed.
2.2. E6aluation of sarcoidosis acti6ity
The parameters used for sarcoidosis activity in this study were based on those reported by James and Jones Williams [13], and are described in Table 2. Minimal residual fibrotic radiological changes not associated with other findings were not considered as sign of active disease. In patients with doubtful or minor clinical, radiological or functional derangements, abnormalities of the serum angiotensin-converting enzyme and gal-lium scan helped to establish the presence of activity [14,15]. According to this definition, sarcoidosis pa-tients were classified into two different groups: 40 with active disease and 45 with inactive disease.
3. Serum samples and analytic procedure
Venous blood for lipoprotein analysis was drawn after a 12-h overnight fast, into blood-collection tubes without anticoagulant. After clotting, each specimen was centrifuged at 1200×g for 20 min, and stored at 4°C until analysis. High-density lipoprotein cholesterol and HDL3-cholesterol were obtained from serum by selective immunoprecipitation with polyethilenglycol solutions of different concentration and pH (IM-MUNO AG, Vienna, Austria). Cholesterol and triglyc-erides were determined by the enzymatic-colorimetric methods CHOD-PAP and GPO-PAP, respectively, comercialized by Boehringer Mannheim. The serum concentration of HDL2-cholesterol was calculated as the difference between the concentrations of HDL-cholesterol and HDL3-cholesterol. The serum concen-tration of LDL-cholesterol was calculated by using the Friedewald equation. Apolipoprotein A-I and apo B were measured by immunological turbidity test Table 1
Characteristics of patients with active and inactive sarcoidosis
Active sarcoidosis Inactive sarcoidosis (n=45,x9S.D.) (n=40,x9S.D.)
Age (years) 45.7915.7 44.3912.7 2:43 8:32
Sex
(male:female)
Body mass 27.795.9 29.795.1 index (kg/m2)
Table 2
Parameters used in the evaluation of sarcoidosis activity
Clinical features
Chest X-ray abnormalities
Decreased forced vital capacity and/or carbon monoxide–diffusing capacityB80% predicted
Increased serum calcium
Increased serum angiotensin-converting enzyme level\mean+2S.D. Increased gallium uptake
Table 3
Serum lipid, lipoprotein and apolipoprotein values
Inactive sarcoidosis (n=45,x 9S.D.)
Active sarcoidosis (n=40,x9S.D.) P
5.2991.13
5.4190.99 NS
Cholesterol (mmol/l)
1.0390.46
Triglyceride (mmol/l) 1.5190.64 B0.0001
3.4291.12
3.5390.85 NS
LDL-cholesterol (mmol/l)
1.4490.39 0.002
HDL-cholesterol (mmol/l) 1.1790.36
0.4290.21
0.3790.18 NS
HDL2-cholesterol (mmol/l)
1.0290.21
HDL3-cholesterol (mmol/l) 0.7890.23 B0.0001
1.6190.27
1.3690.29 B0.0001
Apolipoprotein A-I (g/l)
0.9990.27
Apolipoprotein B (g/l) 1.1490.25 0.012
(Boehringer Mannheim, Germany). Serum amyloid A was quantified from frozen samples by a solid phase sandwich enzyme-linked immuno-sorbent assay (ELISA) with a monoclonal antibody specific for SAA (Cytoscreen™, Biosource international, CA). The intra and inter-assay coefficient of variation were 4.9 and 7.8%, respectively. The analytical method used for pep-tidyl-dipeptidase A (serum angiotensin-converting en-zyme) was a spectrometric method (Sigma Diagnostics). To determine LCAT activity, the initial rate of choles-terol sterification was measured in human plasma. Briefly, to obtain plasma, blood was collected into tubes that contained heparin, cooled immediately in ice water and centrifuged at low speed with a refrigerated centrifuge. Each sample was separated in two parts, one of them processed immediately and the other was incu-bated at 37°C for 40 min with continuous shaking. Both parts were performed as follows (part I immedi-ately and part II after 40 min). One milliliter of LCAT reagent (Wak-Chemie, Medical GMBH, Germany) was added to 50ml of plasma, mixed, incubated for 25 min
at room temperature and their absorbances measured spectrophotometrically at 546 nm. The difference of both measured absorbances (OD part I-OD part II) was extrapolated into a calibration curve made with cholesterol standard (Wak-Chemie, Medical GMBH, Germany) at the same time. Results were expressed as nmol of cholesterol/h per ml plasma and was the mean of sextuplicate.
3.1. Statistical analysis
All calculations were performed using the SPSS-PC 6.0 software package. These programs were simple data description and comparison of two groups by t tests and the Mann – WhitneyU-test for variables that were not normally distributed. These tests were used to assess the univariate statistical association of demo-graphic, clinical and laboratory variables with disease activity. Logistic regression was used to assess the multivariate statistical associations of these variables between active sarcoidosis and inactive sarcoidosis groups. A stepwise method was used for the logistic
regression, including the following variables in the model: age, sex, body mass index, HDL-cholesterol, LDL-cholesterol, triglyceride, LCAT and SAA values. The correlations were calculated using the Pearson correlation coefficient. Results are presented as mean 9S.D.; PB0.05 were considered statistically significant.
4. Results
Table 3 shows the concentrations of serum choles-terol, triglyceride, LDL-cholescholes-terol, HDL-cholesterol and its subfractions HDL2-cholesterol and HDL3 -cholesterol, and apo A-I and apo B in the patients with untreated active and inactive sarcoidosis. Mean HDL-cholesterol serum concentration was 1.1790.36 mmol/l in the active sarcoid patients as compared with 1.449 0.39 mmol/l in the inactive disease group (P=0.002). Apo A-I concentrations were significantly reduced in the patients with active disease (1.3690.29 g/l) com-pared to the inactive disease group (1.6190.27 g/l) (PB0.0001). HDL3-cholesterol concentrations were significantly decreased in patients with active disease as compared with values in the inactive disease group (PB0.0001). Active sarcoid patients had higher levels of triglyceride and apo B concentrations than did inac-tive subjects (PB0.0001 and P=0.012, respectively). Serum concentrations of cholesterol, LDL-cholesterol, and HDL2-cholesterol were similar in both groups.
Table 4 shows the SAA concentrations and LCAT values. Serum amyloid A concentrations were signifi-cantly increased in the patients with active disease (155.459154.01 mg/ml) compared to the inactive sar-coid patients (89.70965.36 mg/ml) (P=0.011). The Mann – Whitney U-test confirmed the significance of the SAA values (P=0.004). There were no significant differences in LCAT values between groups.
other variables were associated with the activity of sarcoidosis.
HDL-cholesterol levels were negatively correlated with serum angiotensin-converting enzyme levels (r= −0.38;P=0.01) and SAA levels (r= −0.39;P=0.01) in patients with active sarcoidosis (Fig. 1), whereas no correlation between these variables was found in the inactive sarcoid group (r=0.06; P=0.65 and r= − 0.04; P=0.75, respectively). Moreover, the concentra-tions of the main protein fraction of HDL, the apo A-I, showed also a significant negative correlation with SAA levels in the active sarcoid group (r= −0.35;P=0.03), but no correlation was found in the inactive disease group (r= −0.02; P=0.87).
5. Discussion
This study reinforces the relationship between sar-coidosis activity and lipoprotein metabolism. The present data coupled with our previous results confirm-ing that the decrease in HDL-cholesterol serum concen-trations is significantly associated with disease activity [8]. HDL-cholesterol was again the only lipid value independently associated with disease activity in the
multivariate analysis. We also found, in keeping with our previous report, a significant negative correlation between serum angiotensin-converting enzyme levels, a well-known marker of inflammatory activity in sar-coidosis [16], and HDL-cholesterol levels in the active sarcoid group.
This pattern of depressed HDL-cholesterol and apo A-I has already been described in patients with a variety of systemic autoimmune diseases [5,17 – 20]. For instance, Lahita et al. [17] showed a pattern of dyslipo-proteinemia similar to our untreated active sarcoid patients in a group of patients with active systemic lupus erythematosus and demonstrated an association between anticardiolipin antibodies and low levels of HDL-cholesterol and apo A-I in patients who were not taking corticosteroids. Svenson et al. [21] suggested that it was the degree of inflammatory activity that deter-mined the lipoprotein alterations observed in their pa-tients with untreated chronic inflammatory arthritides. Other authors have demonstrated a significant inverse correlation between apo A-I levels and several inflam-matory parameters in a group of patients with active juvenile chronic arthritis [22].
The precise mechanism of the reduced serum HDL-cholesterol concentrations in sarcoid patients remains
Table 4
Serum amyloid A and lecithin-cholesterol acyltransferase values
Pa
Active sarcoidosis (n=40,x Inactive sarcoidosis (n=45,x 9S.D.) 9S.D.)
0.004 89.70965.36
Serum amyloid A (mg/ml) 155.459154.01
Lecithin-cholesterol acyltransferase (nmol/ml per h) 215.239122.13 238.769190.06 NS
aBy Mann–WhitneyU-test.
unknown. The factors that account for the decline in HDL-cholesterol and apo A-I have not been directly studied. The results of the current study show a signifi-cant increase of SAA levels in sarcoid patients classified as having inflammatory activity as compared with those having inactive disease. Although SAA levels in inactive sarcoid patients were higher than those described in healthy controls from other studies, we believe the magnitude of this increase is not high enough to inter-act with HDL metabolism. Elevated SAA levels were independently associated with disease activity. The presence of raised SAA levels has already been de-scribed in sarcoid patients, though a comparison be-tween subgroups of patients with active and inactive disease could not be performed because of the small sample size [12]. Serum amyloid A is an acute phase protein produced in response to cytokines synthesized by activated monocyte-macrophages [23,24].
This is the first report of the correlations among apo A-I, HDL-cholesterol and SAA levels in sarcoidosis. In our active sarcoid patients, we have demonstrated a significant inverse correlation between HDL-cholesterol and SAA levels. Moreover, a significant negative corre-lation was observed as well between apo A-I and SAA levels in the active sarcoid group. Alterations in serum lipids and their relationship to acute phase proteins have already been reported in other diseases. Memon et al. [25] found a significant negative correlation between SAA and HDL-cholesterol levels in lepromatous lep-rosy patients, and instead, there was no statistically significant correlation between C-reactive protein and HDL-cholesterol levels in these patients. Notable re-ductions in plasma HDL-cholesterol and apo A-I were observed in patients studied during the acute phase of Kawasaki disease. These changes were accompanied by the marked appearance of SAA protein in the plasma, associated with HDL3 particles. These values returned to normal after complete recovery [26]. It should be noted that in our study, HDL3-cholesterol concentra-tions were significantly decreased in patients with active disease. This result should be interpreted with caution because we used a selective precipitation method to obtain HDL and HDL3and, to our knowledge, we do not know whether the presence of SAA in the HDL could alter the interaction of HDL with the precipitat-ing reagent, interferprecipitat-ing in the precipitation of HDL2. Several studies have shown that, during the acute phase response, SAA is released from the liver and circulates in plasma incorporated into HDL particles, mainly of HDL3 density, with a concomitant decline in plasma HDL concentration [27 – 29]. Serum amyloid A displaces apo A-I from the HDL [30,31]. Van-Lenten et al. [32] reported a 73% decrease of apo A-I levels, as SAA levels in acute phase HDL increased, concomi-tantly. Alternatively, SAA may act inhibiting LCAT activity that leads to low levels of esterified serum
cholesterol and quantitative and qualitative changes in HDL2 and HDL3 particles [11]. In our patients there were no significant differences in LCAT values between groups, measured as the initial rate of cholesterol ester-ification with an exogen substrate. It may be possible that using lipoproteins of patients as substrate, an influence of SAA in LCAT activity would have been revealed.
The marked increase in plasma SAA level and its association with HDL (HDL – SAA) displaces apo A-I from the lipoprotein surface, with the free apoprotein likely to be rapidly removed from the plasma. Previous investigations have demonstrated that macrophages are capable of binding and endocytosing HDL – SAA [33]. Our findings emphasize that SAA plays a role in choles-terol metabolism during the course of inflammation and are in accordance with the suggested mechanism of SAA displacing apo A-I, which may result in increased catabolism of SAA-containing HDL particles.
Low HDL-cholesterol levels are a risk factor for atherosclerosis. Epidemiological studies have estab-lished a negative correlation between coronary artery disease (CAD) and levels of HDL. Other studies have observed that inflammatory responses may have an effect in the atherosclerotic process and some have suggested a potential pathophysiological link between SAA protein and the development of CAD [29,34 – 36]. Despite the fact that in patients with active sarcoidosis the HDL-cholesterol concentrations were low with ele-vated SAA levels, there have so far been no reports that indicate an increasing occurrence of atherosclerotic dis-ease in patients with sarcoidosis.
In summary, the decrease of HDL-cholesterol seen in patients with active sarcoidosis is independently associ-ated with a significant increase of SAA levels in these patients. We suggest that the displacement of apo A-I from the HDL particles by SAA accounts for the decline of HDL-cholesterol seen in patients with active sarcoidosis. Further studies are needed to investigate the composition of HDL particles from subjects with active and inactive sarcoidosis that could demonstrate that the rapid clearance of these particles from plasma is the final mechanism that lowers HDL-cholesterol.
References
[1] Thomas PD, Hunninghake GW. Current concepts of the patho-genesis of sarcoidosis. Am Rev Respir Dis 1987;135:747 – 60. [2] Almenoff JA, Teirstein AS. Clinical significance of serum
ther-molysin-like metalloendopeptidase and its relationship to serum angiotensin converting enzyme in sarcoidosis. Am J Med 1987;82:33 – 7.
[4] Le NA, Gibson JC, Rubenstein A, Grabowski GA, Ginsberg HN. Abnormalities in lipoprotein metabolism in Gaucher type I disease. Metabolism 1988;37:240 – 5.
[5] Svenson KLG, Lithell H, Hallgren R, Selinus I, Vessby B. Serum lipoprotein in active rheumatoid arthritis and other chronic inflammatory arthritides. I. Relativity to inflammatory activity. Arch Intern Med 1987;147:1912 – 6.
[6] Lorber M, Aviram M, Linn S, Scharf Y, Brook JG. Hypocholes-terolemia and abnormal high-density lipoprotein in rheumatoid arthritis. Br J Rheumatol 1985;24:250 – 5.
[7] Kindman LA, Gilbert HS, Almenoff JS, Ginsberg H, Fager-strom R, Teirstein AS. High-density lipoprotein cholesterol is reduced in patients with sarcoidosis. Am J Med 1989;86:376 – 8. [8] Salazar A, Man˜a´ J, Pinto´ X, Argimon JM, Castin˜eiras MJ, Fiol C, Pujol R. Low levels of high density lipoprotein cholesterol in patients with active sarcoidosis. Atherosclerosis 1998;136:133 – 7. [9] Liang JS, Sipe JD. Recombinant human serum amyloid A (apoSAAp) binds cholesterol and modulate cholesterol flux. J Lipid Res 1995;36:37 – 46.
[10] Kumon Y, Suehiro T, Ikeda Y, Yoshida K, Hashimoto K, Ohno F. Influence of serum amyloid A protein on high-density lipo-protein in chronic inflammatory disease. Clin Biochem 1993;26:505 – 11.
[11] Steinmetz A, Hocke G, Saı¨le R, Puchois P, Fruchart JC. Influ-ence of serum amyloid A on cholesterol esterification in human plasma. Biochim Biophys Acta 1989;1006:173 – 8.
[12] Rubinstein I, De Beer FC, Pras M, Knecht A, Baum GL. Serum amyloid A protein concentrations in sarcoidosis. Isr J Med Sci 1989;25:461 – 2.
[13] James DG, Jones Williams W. Criteria of activity. In: Smith LH, editor. Sarcoidosis and Other Granulomatous Disorders. Philadelphia, PA: Saunders, 1985:192 – 8.
[14] Man˜a´ J, Salazar A, Manresa F. Clinical factors predicting persistence of activity in sarcoidosis: A multivariate analysis of 193 cases. Respiration 1994;61:219 – 25.
[15] Man˜a´ J, Salazar A, Pujol R, Manresa F. Are the pulmonary function tests and the markers of activity helpful to establish the prognosis of sarcoidosis? Respiration 1996;63:298 – 303. [16] Rohatgi PK, Ryan JW, Lindeman P. Value of serial
measure-ment of serum angiotensin converting enzyme in the manage-ment of sarcoidosis. Am J Med 1981;70:44 – 50.
[17] Lahita RG, Rivkin E, Cavanagh I, Romano P. Low levels of total cholesterol, high-density lipoprotein, and apolipoprotein A1 in association with anticardiolipin antibodies in patients with systemic lupus erythematosus. Arthritis Rheum 1993;36:1566 – 74.
[18] Leong KH, Koh ET, Feng PH, Boey ML. Lipid profiles in patients with systemic lupus erythematosus. J Rheumatol 1994;21:1264 – 7.
[19] Mitamura T, Ohno S, Ariga H, Ohsaka T, Iwasaki N, Matsuda H, Matsumiya H. Lipoprotein cholesterol concentrations in pa-tients with Behc¸et’s disease. Clin Chim Acta 1988;175:277 – 84. [20] Newburger JW, Burns JC, Beiser AS, Loscalzo J. Altered lipid
profiles after Kawasaki syndrome. Circulation 1991;84:625 – 31. [21] Svenson KLG, Lithell H, Hallgren R, Vessby B. Serum
lipo-protein in active rheumatoid arthritis and other chronic inflam-matory arthritides. II. Effects of anti-inflaminflam-matory and disease
modifying drug treatment. Arch Intern Med 1987;147:1917 – 20. [22] Bakkaloglu A, kirel B, Ozen S, Saatci U, Topaloglu O, Besbas N. Plasma lipids and lipoproteins in juvenile chronic arthritis. Clin Rheumatol 1996;15:341 – 5.
[23] Malle E, Steinmetz A, Raynes JG. Serum amyloid A (SAA): an acute phase protein and apolipoprotein. Atherosclerosis 1993;102:131 – 46.
[24] Malle E, de Beer FC. Human serum amyloid A (SAA) protein: a prominent acute phase reactant for clinical practice. Eur J Clin Invest 1996;26:427 – 35.
[25] Memon RA, Hussain R, Raynes JG, Lateff A, Chiang TJ. Alterations in serum lipids in lepromatous leprosy patients with and without ENL reactions and their relationship to acute phase proteins. Int J Lepr Other Mycobact Dis 1996;64:115 – 22. [26] Cabana VG, Gidding SS, Getz GS, Chapman J, Shulman ST.
Serum amyloid A and high density lipoprotein participate in the acute phase response of Kawasaki disease. Pediatr Res 1997;42:651 – 5.
[27] Parks JS, Rudel LL. Alteration of high-density lipoprotein sub-fraction distribution with induction of serum amyloid A protein (SAA) in the nonhuman primate. J Lipid Res 1985;26:82 – 91. [28] Marhaug G, Dowton SB. Serum amyloid A: an acute phase
apolipoprotein and precursor of AA amyloid. Bailliere’s Clin Rheumatol 1994;8:553 – 73.
[29] Yamada T, Kakihara T, Kamishima T, Fukuda T, Kawai T. Both acute phase and constitutive serum amyloid A are present in atherosclerotic lesions. Pathol Int 1996;46:797 – 800. [30] Coetzee GA, Strachan AF, van der Westhuyzen DR, Hoppe
HC, Jeenah MS, de Beer FC. Serum amyloid A-containing human high density lipoprotein 3. J Biol Chem 1986;261:9644 – 51.
[31] Lindhorst E, Young D, Bagshaw W, Hyland M, Kisilevsky R. Acute inflammation, acute phase serum amyloid A and choles-terol metabolism in the mouse. Biochim Biophys Acta 1997;1339:143 – 54.
[32] Van Lenten BJ, Hama SY, de Beer FC, Stafforini DM, McIntyre TM, Prescott SM, La Du BN, Fogelman AM, Navab M. Anti-inflammatory HDL becomes pro-inflammatory during the acute phase response. Loss of protective effect of HDL against LDL oxidation in aortic wall cell cocultures. J Clin Invest 1995;96:2758 – 67.
[33] Rocken C, Kisilevsky R. Comparison of the binding and endo-cytosis of high-density lipoprotein from healthy (HDL) and inflamed (HDL – (SAA)) donors by murine macrophages of four different mouse strains. Virchows Arch 1998;432:547 – 55. [34] Fyfe AI, Rothenberg LS, de Beer FC, Cantor RM, Rotter JI,
Lusis AJ. Association between serum amyloid A proteins and coronary artery disease. Evidence from two distinct arterioscle-rotic processes. Circulation 1997;96:2914 – 9.
[35] Blum A, Vardinon N, Kaplan G, Laniado S, Yust I, Burk M, Miller H. Autoimmune and inflammatory responses may have an additive effect in postpercutaneous transluminal coronary angioplasty restenosis. Am J Cardiol 1998;81:339 – 41.
[36] Zimmermann J, Herrlinger S, Pruy A, Metzger T, Wanner C. Inflammation enhances cardiovascular risk and mortality in hemodialysis patients. Kidney Int 1999;55:648 – 58.
