Directory UMM :Data Elmu:jurnal:A:Atherosclerosis:Vol153.Issue1.Nov2000:

(1)

Diabetic state induces lipid loading and altered expression and

secretion of lipoprotein lipase in human monocyte-derived

macrophages

Anca Dana Dobrian

_{, Vadim Lazar}

_{, Crina Sinescu}

_{, Dana Mincu}

_,

Maya Simionescu

_*

a_{Institute of Cellular Biology and Pathology}_,_‘_{Nicolae Simionescu}_’_,₈_B_._P_._{Hasdeu St}_.,₇₉₆₉₁_Bucharest_,_Romania b_{Clinic of Cardiology}_,_‘_D_._Bagdasar_’_{Emergency Hospital}_,_Bucharest_,_Romania

Received 7 December 1998; received in revised form 15 December 1999; accepted 14 January 2000

Abstract

Non-insulin-dependent diabetes mellitus (NIDDM) is frequently associated with macroangiopathies and coronary heart diseases. Lipoprotein lipase (LPL), an enzyme known to undergo significant functional alterations in diabetic state, is also a potential atherogenic protein. Since, to the best of our knowledge, there are no data concerning LPL secreted by macrophages of NIDDM patients we conducted a study to assess the expression and activity of LPL secreted by monocyte-derived macrophages from NIDDM patients with cardiovascular complications versus cardiovascular patients without diabetes (controls). Isolated cells from NIDDM patients, after 7 days in culture in the presence of 20% autologous serum, readily exhibit a foam cell phenotype, in contrast to the cells from controls. Macrophages were mainly loaded with triglycerides, whose cellular amount was well correlated to triglyceridemia of NIDDM subjects. Concomitantly, macrophages from NIDDM patients displayed a six-fold decrease of mRNA expression and a two-fold reduction of the activity of secreted LPL, as compared to control cells. These data suggest that in complicated diabetic state, macrophage loading leading to foam cell formation is accelerated, at least in part, due to a diminished expression and activity of LPL. These observations add and extend the data that may explain the occurrence of accelerated atherogenesis and of the atherosclerotic complications associated with diabetes. © 2000 Elsevier Science Ireland Ltd. All rights reserved.

Keywords:Atherosclerosis; Polymerase chain reaction; Foam cell; Non-insulin-dependent diabetes mellitus; Fluorometry

www.elsevier.com/locate/atherosclerosis

1. Introduction

Lipoprotein lipase (LPL), a hydrolytic enzyme lo-cated on the capillary endothelium, is a rate limiting factor for the catabolism of plasma triglyceride-rich lipoproteins [1,2]. Data in the literature indicate that in human subjects with type 2 diabetes mellitus, the LPL activity in post-heparin plasma is decreased [3,4], and that the polymorphism of LPL gene is associated with a high risk of coronary heart disease [5]. Besides the main location of the enzyme over the capillary beds, it was also detected within the arterial wall [6], especially in the regions affected by atherosclerotic plaque [7].

The atherogenicity of LPL of the arterial wall is rather controversial [8]. However, there are several reports that emphasise the potential atherogenic effects of LPL such as, enhanced LDL subendothelial retention via interaction with proteoglycans [9] increased lipid uptake by VLDL and LDL receptor on macrophages [10,11] as well as cytotoxic effects of their reaction products on cells of the arterial wall [12]. Studies conducted in vitro have demonstrated LPL secretion by the major cell types of the plaque, i.e. macrophages [13,14] and smooth muscle cells [15]. Consistent with the in vitro observations, O’Brien et al., using immunocytochem-istry and in situ hybridisation performed on human coronary atherosclerotic plaques, found that macrophage-derived foam cells are the major source of LPL [16]. It is also stated that both in murine peri-* Corresponding author. Tel.: +401-411-1145; fax: +

401-411-1143.

(2)

toneal macrophages [17] and in human macrophages isolated from atherosclerotic plaques [18] lipid loading is accompanied by a marked decrease in expression and activity of lipoprotein lipase. Accelerated atherosclero-sis is the main complication of diabetes. Also, diabetic state is known to induce metabolic activation in mono-cyte/macrophages [19] as well as enhanced phagocytic [20] and chemotactic [21] properties and increased cy-tokine production [19]. To the best of our knowledge there are no reported data either on LPL expression and secretion by monocyte-derived macrophages ob-tained from patients with non-insulin-dependent dia-betes mellitus or about the loading of these cells with lipids. Therefore, in the present study, we tested (a) the level of LPL mRNA expression and LPL activity secreted by monocyte/macrophages from patients with NIDDM versus non-diabetic subjects; and (b) the pos-sible correlation between LPL expression and activity and the lipid loading of these macrophages in culture.

2. Materials and methods

2.1. Patients

The subjects chosen for the study fall into two cate-gories, (1) non-insulin-dependent diabetics (NIDDM), diagnosed for more than 1 year, with associated is-chemic heart diseases (stable angina pectoris and his-tory of myocardial infarction) and; (2) patients suffering from similar ischemia but without diabetes mellitus (controls). A number of 30 NIDDM patients and 24 non-diabetics (controls) ageing 40 – 70 years were selected from ‘D. Bagdasar’ emergency clinic in Bucharest, with the personal patient consent. Patients were hospitalised for their coronary heart disease and not for diabetes. However, diabetic patients were diag-nosed for at least 1 year and their diabetes was gener-ally well controlled. For each patient the following parameters were assayed: serum cholesterol, HDL cholesterol, triglycerides, glycemia, blood pressure and body mass index. Coronary atherosclerosis in both groups of patients was documented by electrocardio-grams, exercise tolerance tests, and, in some cases, by coronary angiography. The patients in the two groups were age and sex matched. Data are expressed as mean values9S.D. and statistically analysed using the one-way ANOVA test for comparison between diabetic and non-diabetic patients.

2.2. Isolation and culture of monocyte-deri6_ed macrophages

Mononuclear cells were isolated from whole blood, collected on 3.8% Na2citrate using Hystopaque 1077,

according to [22]. Monocytes were subsequently

purified by centrifugation using a Percoll gradient [23]. After two washes with Hank’s Balanced Salt Solution, the cells were seeded (106 _{cells per well) in 24-well}

culture dishes in RPMI 1640 medium supplemented with 20% autologous serum, 100 U/ml penicillin and 100 mg/ml streptomycin. After 2 h incubation, the non-adherent cells were washed and the remaining cells were cultured for 6 days, in 5% CO2humidified atmosphere,

at 37°C, replacing the medium every 2 days. All the assays were performed on the 7th day, after a 20 h incubation of cells with fresh medium. The cell viability was evaluated using a LIVE/DEAD EukoLight Viabil-ity/Cytotoxicity Kit (Molecular Probes, Inc., Eugene, OR). Briefly, after washing with warm medium, and fixation using 2%p-formaldehyde, cells were incubated with both calcein (stain for living cells) and ethidium bromide homodimer (stain for death cells), according to the protocol indicated by the manufacturer. Finally, cells were counted under a Nikon fluorescence microscope.

2.3. Lipoprotein lipase acti6_{ity assays}

One milliliter medium (collected from four wells) was concentrated to a final volume of 100 ml using Cen-triPrep concentration tubes (cut-off 10 000 Da, Mil-lipore Intertech, Marlborough, MA). In some experiments, the cells were previously treated with hep-arin (4 Ul/ml), for 20 min, in order to release the LPL associated with the cell surface. The LPL activity was assayed by a non-radioactive method adapted from Nilsson – Ehle and Schotz [24]. Briefly, 100ml of concen-trated medium was incubated with 150ml of a substrate mixture, yielding a final concentration of 1.7 mM glyc-erol trioleate (corrected according to each individual plasma triglyceride values) and 0.5 mM lecithin, 1% BSA in 80 mM Tris – HCl, NaCl 150 mM, pH 8.2 and 50 ml heat inactivated human serum, in 500 ml total volume. In all cases, the triglyceride hydrolysis does not exceed 5 – 8% of the total amount of substrate in the mixture. Free fatty acids released in the reaction were measured using the method of Itaya [25]. A control consisting from RPMI-1640 culture medium with 20% autologous serum was treated similarly to the corre-sponding probe and the activity value obtained was subtracted from the latter. All measurements of LPL activity were performed within 2 h from medium collection.

2.4. Quantitati6_{e RT}_-_{PCR for LPL mRNA}

Total RNA was extracted from 104 _to ₁₀6

macrophages after 7 days in culture. Cells were scraped from the plates using a rubber policeman, than washed twice with serum-free medium and the total RNA was extracted using a GlassMax RNA Microisolation Spin

(3)

Cartridge System (GIBCO BRL, Vienna, Austria) ac-cording to the instructions provided by the manufac-turer. The method is a modification of the procedure described in [26], suitable for a quantity as low as 104

cells. The total RNA obtained had a A260/A280 ratio

between 1.8 and 2.0 and was immediately frozen and kept at −70°C until use. Quantitation of LPL mRNA was essentially performed according to [27]. Reverse transcription was accomplished for 1 h at 37°C using 100 – 300 ng total cellular RNA, 106_{copies of pAW109}

cRNA (Perkin-Elmer Corp., Norwalk, CT) as an inter-nal standard in a total volume of 20ml, containing 0.25 mg random hexamer primers, 0.5 mM dNTP, 50 U MMLV-RT-ase (RT-PCR kit purchased from Strata-gene Cloning Systems, LaJolla, CA). After reverse tran-scription, four serial 1:2 or 1:3 dilutions of the entire cDNA mixture were amplified using a PCR Core Kit from Boehringer Mannheim GMBH (Germany). The oligonucleotide primers for LPL used in the reaction (ordered from MWG (Germany)) were, 5%

-GAGATTTCTCTGTATGGCACC-3% ₍₅%_{primer) and}

5%_{-CTGCAAATGAGACACTTTCTC-3}% ₍₃% _primer).

Whenever the case, the 5%_{primers were labelled in the 5}%

end with [g-32P]ATP using polynucleotide kinase (Boehringer Mannheim GMBH, Germany). The PCR amplification protocol was carried out in a 100 ml volume, in 35 cycles involving denaturation at 95°, for 60 s, annealing at 60°, for 45 s and a final extension at 72°C, for 7 min, in a GeneAmp PCR System 2400 (Perkin-Elmer Corp., Norwalk, CT). PCR products were separated by electrophoresis on 3% Nusieve 3:1 agarose gels (FMS Bioproduct, Rockland, ME). Subse-quently to a 4 h migration, the PCR products were stained directly in the agarose gel using the fluorescent stain SYBR Green II, as in [28]. Quantitative analysis of the products was done by densitometry using a Shimadzu RF 5001 spectrometer and by plotting the arbitrary units of fluorescence (AUF) against the inter-nal standard concentration [27]. Alternatively, some of the probes were stained with ethidium bromide (Sigma Chemical Co., St. Louis, MO), the appropriate bands were excised from the gel and 32_{P was quantitated by}

Cerenkov counting.

2.5. Electron microscopy and morphometry

To visualise the loading of macrophages with lipids after 7 days in culture, the cells were washed three times with 100 mM phosphate buffer saline, pH 7.4 and fixed in the culture plates with a mixture containing 3% para-formaldehyde, 5% glutaraldehyde, 2% osmium te-traoxide in 1.1 M cacodylate buffer (pH 7.4) and saturated solution of ferrous citrate on ice [29]. After 40 min, cells were scraped from the plates and pelleted by centrifugation at 5000×g. After mordanting with 1% tannic acid for 10 min [29], the pelleted cells were

dehydrated with increasing concentrations of ethanol and embedded in Epon. Thin sections stained with uranyl acetate and lead citrate were examined with a Phillips 400 Electron Microscope. Quantitative mea-surements of lipid droplet number were made directly on electron micrographs of macrophages, in which the plane of the section was through the body of the cell. A number of 20 cells for each experimental condition were randomly selected (n=5 per group).

2.6. Lipid analysis

To asses the intracellular lipid content, the cells plated on four culture wells were washed three times with PBS and scraped with a rubber policeman in 1 ml total volume of water. After a 30 s sonication of the cell suspension, lipids were extracted in a mixture of chloro-form:methanol:water (2:2:1). The organic phase con-taining lipids was dried under nitrogen and the pellet was solubilised in a chloroform:methanol 2:1 (v/v) mix-ture. The amount of triglyceride and cholesterol (free and esterified) was measured using enzymatic kits pro-vided by Sigma (St. Louis, MO). Blanks and standard curves for each lipid class were run in parallel. Intracel-lular lipids were expressed as mg lipid per mg cell protein.

2.7. Other assays

Serum triglycerides, cholesterol and glycemia were assayed by using Sigma enzymatic kits. HDL choles-terol was measured following precipitation with na-trium phosphotungstate/MgCl2 as in [30]. LDL

cholesterol was calculated according to Friedewald for-mula. Cell protein was determined according to Brad-ford, as described in [31].

3. Experimental results

3.1. Characterisation of the patients

Non-insulin-dependent diabetic subjects with cardio-vascular complications (NIDDM) and patients with similar cardiovascular disease but without diabetes (controls) were sex- and age-matched. As shown in Table 1, NIDDM patients did not display severe obe-sity (BMIB33 kg/m2_{) or severe hypertriglyceridemia}

(serum triglycerides did not exceed 200 mg/dl) and there were no significant statistical differences in serum total cholesterol, triglycerides and HDL/LDL choles-terol between the patients in the two groups (P\0.05), except that the blood glucose was highly increased in NIDDM patients. In addition, all patients in the NIDDM group received hypoglycemiant sulfamides, but no insulin. With respect to cardiovascular disease,

(4)

patients in both groups presented stable angina pectoris (12 patients NIDDM, 15 controls), myocardial infarc-tion in history (six patients NIDDM, five controls) or both (12 patients NIDDM, 15 controls). Based on the clinical and biochemical data, we could assume that any significant difference in the results between NIDDM and controls may be mostly ascribed to the diabetic state.

3.2. Enzymatic acti6_{ity of LPL secreted by} macrophages isolated from NIDDM and controls

Monocyte-derived macrophages (MDM) from both NIDDM and control subjects were cultured in RPMI 1640 medium containing 20% autologous serum and the activity of LPL secreted by the cells was assayed in 20 h conditioned media, starting with day 1 of culture for 11 days. For cells belonging to both groups of patients, the results indicated a linear increase of the activity during the first 5 days, followed by a plateau for the next 4 days, and a slight decrease up to the 11th day, partly due to an increased cell mortality as assayed by LIVE/DEAD Cytotoxicity/Viability kit (data not shown). Also, over the 11-day period, the absolute values for LPL activity of diabetic macrophages were constantly lower than that found in control cells. Con-sequently, we performed all the assays for the activity of secreted LPL in the 20 h conditioned media at day 7 of culture. For each probe, LPL activity was assayed in parallel in the conditioned media and in an equal

Fig. 1. Enzymatic activity of LPL secreted by monocyte-derived macrophages isolated from NIDDM patients as compared to con-trols. Macorphages were cultured for 7 days in RPMI 1640 medium supplemented with 20% autologous serum. Medium was replaced every 2 days and finally 24 h before the assay. Conditioned medium was concentrated to a final 0.1 ml volume and the LPL activity was quantified using an artificial trioleinic substrate, for 1 h, at 37°C. FFA, free fatty acids; controls, patients with cardiovascular disease but without diabetes; NIDDM, patients with non-insulin-dependent diabetes mellitus and cardiovascular disease. P=0.00035 as deter-mined by one-way ANOVA test.

volume of RPMI with 20% autologous serum main-tained without cells in the same conditions, in order to subtract the intrinsic residual serum LPL activity and free fatty acids of the autologous serum used in the culture medium. To increase the reproducibility of the results, as a source of apo CII (for activation of LPL) a unique pool of heat-inactivated human serum, stored at −70°C, was used. During the 60-min assay the reaction was linear with respect to enzyme concentra-tion and time and in most cases the substrate hydrolysis was kept under 8%. LPL activity secreted by MDM separated from NIDDM was highly decreased (42%), as compared to controls (298.5940.7 vs. 691.19103.1 nmol FFA per mg cell protein/h, 37°C, P=0.00035) (Fig. 1). The values obtained for LPL activity in our experimental conditions were in the same range as those reported by other groups [14,32].

3.3. Expression of LPL mRNA secretion by human monocyte-deri6_{ed macrophages of NIDDM and} controls

The expression of mRNA for LPL in MDM was measured after total RNA extraction as described in Section 2. Each probe was reverse transcribed together Table 1

Clinical data and serum variables of subject groupsa

Variable NIDDM Controls Significance (n=28)

(n=31)

60.191.46

Age (years) 58.491.8 N.S.

Male gender 68.9 64.2 (%)

Body mass 29.491.1 26.892.4 N.S. index

(kg/m2₎

Glycemia 164.2912.3 88.493.9 B0.0001 (mg/dl)

Duration of 6.190.7 – diabetes

(years)

156.8912.9 148.4912.7 N.S. Triglycerides

(mg/dl)

199.496.95 N.S. Total 191.498.13

cholesterol (mg/dl)

3.3190.5 N.S. HDL/LDL 3.8590.6

cholesterol

a_{NIDDM, patients with non-insulin-dependent diabetes mellitus} and cardiovascular complications; controls, patients with cardiovas-cular disease only; values represent means9S.D.; N.S., a non-signifi-cant one-way ANOVA test.

(5)

with AW109 as an internal standard and the four dilutions of the resulting cDNAs were amplified by PCR for 35 cycles. In each case, the two amplification curves were parallel, which indicates that the amplifica-tion efficiencies were the same for both the target and the standard. This feature attests an accurate quantita-tive measurement of LPL mRNA expression. The reac-tion products from AW109 cRNA (300 bp) and the target mRNA for LPL (277 bp) were separated by 3% agarose gel electrophoresis. An example of PCR prod-ucts of LPL mRNA from MDM isolated from a car-diovascular patient (control) and the internal standard in serial 1:3 dilutions of the cDNA mixture is shown in Fig. 2. The products were visualised using SYBR Green I staining and quantitated by measuring the fluores-cence intensities of the resolved bands. The arbitrary units of fluorescence (AUF) for each band were plotted against the template concentration. The specificity of the reaction was validated by the lack of any detectable signal every time we included a control experiment in which either the reverse transcriptase was omitted or in which mRNA was replaced by water. Monocyte-derived macrophages from both NIDDM and control subjects were assayed for LPL mRNA expression on day 7 of culture. On average, the total number of cells subjected to RNA extraction was between 5×104 _and

5×105_{and the reverse transcribed RNA was} _{200 ng.}

In each experiment, 108 _{copies of AW109 cRNA were}

used. The quantity of LPL mRNA obtained from the amplification curves was normalised to the number of viable cells subjected to the total RNA extraction. Results showed a 6.5-fold decrease in mRNA expres-sion for LPL in MDM isolated from NIDDM as compared to control patients (Table 2). The data re-garding the LPL mRNA expression correlate well with those concerning the activity of secreted LPL in NIDDM and controls. The difference between the de-crease in LPL mRNA expression (85%) and in secreted

Table 2

LPL mRNA levels in monocyte-derived macrophages isolated from NIDDM and control patientsa

LPL mRNA (molecules per cell) Patient

NIDDM 15.595.42

101.7940.5 Controls

a_{Total RNA was isolated from monocyte-derived macrophages of} NIDDM or controls and cultured for 7 days in RPMI 1640 supple-mented with 20% autologous serum. LPL mRNA molecules per cell is calculated considering only viable cells counted after fluorescent staining with LIVE/DEAD Cytotoxicity kit. Data represent mean value 9S.D. of four experiments. The means are statistically differ-ent as judged by P=0.05, calculated using one-way ANOVA test. NIDDM, non-insulin-dependent diabetic patients with associated cardiovascular disease; controls, patients with cardiovascular disease only.

LPL activity (58%) in NIDDm versus control patients could be due, at least in part, to a different intracellular pool of LPL in the macrophages of the two groups. 3.4. Morphology and lipid accumulation in MDM from NIDDM and control patients

Freshly isolated human monocytes from both subject groups contained no detectable lipid inclusions and displayed similar morphological appearance (Fig. 3a and d). As early as day 4 in culture, MDM isolated from the NIDDM patients started to display numerous lipid droplets in the cytoplasm (30912), which in-creased steadily in number and by the 7th day, practi-cally all cells exhibited a typical foam cell morphology (with 90918 lipid droplets in the cytoplasm), with few organelles and a small, sometimes eccentric, nucleus (Fig. 3e and f). The MDM originating from control patients showed few, if any, intracellular lipid droplets after 4 days in culture (894) and on the 7th day, in some cells, a variable number of lipid inclusions were found (38917), a figure significantly lower when com-pared to NIDDM macrophages at the same time point (PB0.05). Extreme examples of control MDM are shown in Fig. 3b and c. After 11 days in culture the differences in lipid loading of cells from the two groups, were similar to those observed on day 7. However, though unlikely, we cannot rule out the possibility that eventually, after 2 – 3 weeks in culture, the macrophages isolated from controls would also exhibit a foam cell-like phenotype. To assess the lipid composition, the MDM were subjected to chloroform:methanol extrac-tion and the triglycerides and total cholesterol quantified. The results showed a significant triglyceride enrichment in NIDDM (8379110 mg per mg cell protein) versus controls (5009130 mg per mg cell protein) (Table 3). For total cholesterol, although the difference between the MDM of the two groups did not reach a statistical significance (P\0.05), the mean was Fig. 2. Quantitative analysis of LPL mRNA levels in

monocyte-derived macrophages of a patient with cardiovascular disease only (control); SYBR Green I staining of PCR products separated in 3% agarose gel. Lanes 1 – 5 indicate LPL mRNA PCR products from 1:3 dilutions of a sample containing 250 ng total RNA from macrophages and 106_{molecules of AW109 cRNA after amplification} for 35 cycles. Lane 6 is the 0.5 Kb molecular weight marker.

(6)

32% lower in NIDDM group (Table 3). In addition, results indicated that for both group, freshly isolated monocytes has lower values for the total cholesterol (114927 mg per mg cell protein) and triglycerides (270954 mg per mg cell protein), respectively. As ob-served from the above data, the main intracellular lipid pool is represented by triglycerides (especially in MDM from NIDDM group), which is in accordance with the results reported by others for MDM of healthy human subjects [33]. To test the influence of autologous serum on lipid loading of MDM, cross-experiments were per-formed, in which the serum from controls was used to culture cells of NIDDM and vice versa. Data about the intracellular cholesterol and triglycerides mass in macrophages from NIDDM and control subjects culti-vated in autologous serum are set out in Table 3. When the serum was switched between the cells isolated from the two groups, the results showed that MDM from NIDDM patients were still loaded with lipids in day 7 of culture (735997 mg triglyceride per mg cell protein and 237.5964 mg cholesterol per mg cell protein), while the MDM from controls displayed only a moder-ate lipid loading when grown in NIDDM serum-con-taining medium (385966 mg triglyceride per mg cell protein and 360937 mg cholesterol per mg cell protein). These data may suggest a different mechanism

of lipid uptake and/or metabolism in macrophages of NIDDM subjects as compared to controls. The poten-tial correlations between each of the total cholesterol and triglycerides concentrations in serum and their respective intracellular mass for each individual within the two subject groups were further investigated (Fig. 4). For the control group, there is a lack of correlation between the serum cholesterol and triglycerides and their intracellular counterparts (P=0.97,R=0.01 and P=0.65, R=0.13, respectively) (Fig. 4A and B). MDM of NIDDM patients also exhibit a weak correla-tion with respect to serum versus intracellular choles-terol (P=0.2,R=0.35) (Fig. 4D). In contradistinction, serum and intracellular triglycerides are very well corre-lated in macrophages of NIDDM group (P=0.0001, R=0.83) (Fig. 4C). These data bring further support for possible alteration in triglyceride metabolism of diabetic MDM. The prominent lipid loading of cells from diabetic patients is accompanied by a decrease in the expression and activity of LPL as described above. Fig. 5 shows a good correlation between the triglyceride loading of macrophages and the LPL activity in the cell medium. Both in control group (Fig. 5A) and in NIDDM group (Fig. 5B) higher LPL activity values corresponded to lower intracellular triglyceride amounts, leading to a progressive decrease of LPL

Fig. 3. Electron micrographs of monocytes and monocyte-derived macrophages from non-insulin diabetic patients with cardiovascular disease (NIDDM) and patients with cardiovascular disease only (controls). After isolation, monocytes from either controls (a) or NIDDM (d), were allowed to adhere for 2 h on 24-well culture plates in RPMI 1640 with 20% autologous serum and after extensive washings with serum free medium, cells were fixed for 20 min, scraped from the plates and processed for electron microscopy. Alternatively, monocyte-derived macrophages were cultured for 7 days and then processed (as above). In the cells isolated from controls various numbers of cytoplasmic lipid droplets (LD) are present (38917); two extreme cases are shown in (b and c). The 7 day cultured macrophages from two NIDDM patients, (e) and (f), have the typical morphological aspect of foam cells, i.e. the cell cytoplasm is filled with droplets. Bar, 1mm.

(7)

Table 3

Intracellular neutral lipid content of monocyte-derived macrophages isolated from NIDDM and controlsa

Controls (mg per mg cell protein) Significance

Lipid class NIDDM (mg per mg cell protein)

5009130

Triglycerides 8379110 P=0.05

Total cholesterol 287.5955 4209107 P=0.28

a_{Lipids were extracted from monocyte-derived macrophages after 7 days in culture, and quantitated using enzymatic kits. Values are given as} mean 9S.D. (n=13). The ‘P’ value is calculated using one-way ANOVA test. NIDDM, non-insulin-dependent diabetic patients with cardiovascular disease; controls, patients with cardiovascular disease and without diabetes.

activity while cells became loaded with lipids. For both groups the values were well corrected (r=0.657 for controls andr=0.763 for NIDDM) and ‘P’ is less than 0.03 as determined by ANOVA analysis.

4. Discussion

In patients with NIDDM, macrovascular complica-tions such as coronary artery and peripheral vascular disorders are the major health problems; in Caucasian patients with NIDDM their contribution to mortality overrides the impact of microvascular complications [34]. In order to emphasise the role of long lasting diabetic state on lipid metabolism of macrophages, we selected for this study one group of patients with non-insulin-dependent diabetes mellitus and ischemic heart diseases and one control group represented by patients with the same type of ischemia, but without diabetes. Our results indicated that the activity of LPL secreted by macrophages of NIDDM patients is 42% decreased as compared to controls. A similar diminu-tion in LPL activity is also reported for murine macrophages in streptozotocin-induced diabetes [35]. There are several possible reasons, which could account for the observed change in LPL activity. First, the high blood glucose concentration in NIDDM may serve as an alternative to free fatty acids as source of energy for the cell, leading to decreased synthesis and activity of LPL. This is also supported by the fact that, among all serum biochemical parameters that we have tested, fasting glycemia was the only statistically different one. Furthermore, our current experiments on murine peri-toneal macrophages cultured in RPMI supplemented with 0.1 and 0.2% glucose revealed a two-fold decrease in LPL activity in the latter condition. Another possible explanation is the activated state of monocytes in dia-betic subjects [36], which induces the secretion of sev-eral inflammatory cytokines such as IL-1, TNF or IFNg, that in turn, were shown to suppress secreted

LPL activity in human MDM [37] or mouse peritoneal macrophages [38]. Suppression of LPL activity in murine peritoneal macrophages loaded with triglyceride [17] or in macrophage-derived foam cells isolated from the arterial wall of human subjects [18] were also

re-ported. These data are in accordance to our results, in which MDM from NIDDM patients exhibit a de-creased secreted LPL activity concomitantly with a foam cell phenotype (after 7 days in culture) as com-pared to MDM from controls, which are only moder-ately loaded with triglycerides. Thus, a massive triglyceride loading may be partly responsible for the decreased LPL activity observed in MDM of NIDDM patients.

Our results also showed that LPL mRNA synthesis is six-fold reduced in MDM from NIDDM versus con-trols, while the activity is only two-fold decreased. A similar result is reported for LPL in adipose tissue of guinea pig during fasting [39]. The possible explana-tions may be the existence of a pool of inactive LPL stored in a high-mannose form within the endoplasmic reticulum [40] of control MDM or reduced stability of LPL mRNA in macrophages of NIDDM patients. A similar discrepancy was reported for J744 macrophages grown in high versus normal glucose medium, and the authors suggest that post-translational modifications may account for the difference observed [41]. Finally, we may speculate that due to a higher efficiency of the diabetic LPL than that of the normal enzyme in de-grading the substrate, the specific activity of the LPL from diabetics is actually more than two-fold decreased. The results presented here indicate that the NIDDM macrophages are prone to become loaded with lipids, mainly as triglycerides. Recent data [33] show that human monocyte derived macrophages from healthy subjects cultured in RPMI with 10% human serum, but not foetal calf serum, turn into triglyceride-rich foam cells, with no change in LPL activity. Also there is one report about an increase, rather than a decrease of LPL activity in macrophages isolated from diabetic patients compared to healthy controls [42]. In contrast to these results, we found a selective lipid loading of MDM accompanied by a decrease in LPL activity and expres-sion. This behaviour of macrophages obtained from NIDDM patients is in accordance with previous obser-vations [17,18] that indicated that macrophage derived-foam cells express lower LPL mRNA and activity. On the other hand, Mattsson et al. [18] showed that only the CD14 positive cells from human atheroma exhibit a decreased LPL activity and lipid loading; these cells are

(8)

capable of phagocytosis. Although scarce, there are some data in the literature to document an increased phagocytic capacity of human monocytes in diabetes [36]. Thus, we cannot rule out the possibility of a different number of CD14 positive cells among MDM population of NIDDN versus control patients. More-over, Behr et al. [43] have shown that a high degree of activation of monocytes/macrophages leads to a de-creased LPL activity.

Another observation concerns the strong correlation that exists between serum and intracellular triglyceride in MDM from NIDDM, but not in control patients. This suggests that NIDDM macrophages take up

triglyceride-rich lipoproteins via a non-down regulated pathway. One of these pathways could involve the VLDL receptor [44], recently demonstrated by in situ hybridisation and immunohistochemistry on macrophages and endothelium of human arteries [45]. The decreased LPL activity in NIDDM – MDM versus controls, together with a lower LPL affinity for diabetic VLDL, as reported in [46] may lead to a higher concen-tration of intact VLDL in the medium, and thus an increased number of ligands available for the VLDL receptor, that could eventually lead to the foam cell formation by different mechanisms. Furthermore, dia-betic patients have been noted to have elevations of

Fig. 4. Correlation between intracellular and serum triglycerides and total cholesterol in the two subject groups. After 7 days of culture with RPMI 1640 supplemented with 20% autologous serum, monocyte-derived macrophages were washed in warm medium, than scraped from the plates and total cholesterol and triglycerides determined after lipid extraction using enzymatic kits; serum lipids were quantitated similarly. A and B, patients with cardiovascular disease (controls,); C and D, non-insulin-dependent diabetic patients with cardiovascular complications (NIDDM, ).

(9)

Fig. 5. Correlation between intracellular triglycerides and LPL activity in the medium of 7 day cultured monocyte/macrophages isolated from (A), patients with cardiovascular disease (control,); and (B), non-insulin diabetic patients with cardiovascular complications (NIDDM,). Cell medium was used to determine LPL activity and cell lipids were extracted and quantitated as described in Section 2. Values are from 13 controls and 15 NIDDM.

abnormal triglyceride-rich lipoproteins that are en-riched in apoE [47] and are more avidly taken up by macrophages [48]. On the other hand, we may speculate that, due to an increased activation reported for dia-betic monocytes [37], the cells isolated from NIDDM patients had an increased capacity of oxidising lipo-proteins in serum-containing culture media that, in turn, are recognised by the macrophage scavenger re-ceptors, leading to fast lipid loading of the cells [49]. In support of this hypothesis, and of our data there is a recent report showing that inhibition of LPL expression in human monocyte-derived macrophages is dependent on LDL oxidation [50]. Another possible explanation for triglyceride loading of macrophages isolated from NIDDM patients is related to the apoE secreted by the cells [51]. A recent study demonstrated that apoE en-hances lipid uptake by macrophages in LPL deficiency [52]. On the other hand, Evans et al. [53] showed that extracellular lipolysis of VLDL subfractions by LPL is necessary for lipid loading of J744A.1 macrophages via a LDL receptor-mediated mechanism. Our data show-ing a decreased LPL secretion in diabetic macrophages may suggest though, that lipoproteins could be taken up with decreased efficiency by the cells via the LDL receptor. However, as a result of this decreased uptake, the local residence time of the lipoproteins would be prolonged, making modification (lipid peroxidation) and uptake by scavenger receptors more likely.

Collec-tively, these data may provide a plausible explanation as to why a reduced LPL activity in NIDDM derived macrophages may be able to potentiate triglycerides and, to a lesser extent, cholesterol loading of the cells. In conclusion, the results reported here indicate that, (i) monocyte-derived macrophages from NIDDM pa-tients became foam cells after 7 days in culture, whereas in the same conditions MDM from control patients are only moderately loaded with lipids; (ii) in diabetic patients the accumulation of triglycerides in macrophages is well correlated with triglyceridemia; and (iii) macrophage-derived foam cell formation coin-cides with a decreased LPL mRNA expression and enzymatic activity in these cells.

We can postulate that in diabetic state, the decreased synthesis of LPL may cause accumulation of triglyce-rides in macrophages that ultimately accelerates their transformation into foam cells thus contributing to the rapid formation of atheroma and the accelerated atherosclerosis characteristic for diabetic condition.

Acknowledgements

We gratefully acknowledge to Dr Georg Breier and Dr Werner Risau at Max Planck Institut fur Physiolo-gische und Klinische Forschung, Bad Nauheim, Ger-many for precious advice on RT-PCR technique. We

(10)

also thank loana Manolescu, Cristina Dobre, Elena Florea, Manda Misici, Nicoleta Mobre and Mihaela Schean for the qualified assistance. This work was supported by a grant from the Romanian Academy and an UNESCO I-Molecular and Cell Biology Network Research Grant, 1999.

References

[1] Eckel RH. Lipoprotein lipase. A multifunctional enzyme rele-vant to common metabolic diseases. New Engl J Med 1988;323:1000 – 6.

[2] Bensadoun A. Lipoprotein lipase. Annu Rev Nutr 1991;11:217 – 37.

[3] Howard BV. Lipoprotein metabolism in diabetes mellitus. J Lipid Res 1987;28:613 – 28.

[4] Stewart MW, Laker MF, Dyer RG, Game F, Mitcheson J, Winocour PH, Alberti KG. Lipoprotein compositional abnor-malities and insulin resistance in type 2 diabetic patients with mild hyperlipidemia. Arterioscler Thromb 1993;13:1046 – 52. [5] Ukkola O, Savolainen MJ, Salmela PI, von Dickhoff K,

Ke-saniemi YA. DNA polymorphisms at the lipoprotein lipase gene are associated with macroangiopathy in type 2 (non-insulin-de-pendent) diabetes mellitus. Atherosclerosis 1995;115:99 – 105. [6] Jonasson L, Bondjers G, Hansson GK. Lipoprotein lipase in

atherosclerosis: its presence in smooth muscle cells and absence from macrophages. J Lipid Res 1987;28:437 – 45.

[7] Yla-Herttuala S, Lipton BA, Rosenfeld ME, Goldberg IJ, Stein-berg D, Witztum JL. Macrophages and smooth muscle cells express lipoprotein lipase in human and rabbit atherosclerotic lesions. Proc Natl Acad Sci USA 1991;88:10143 – 7.

[8] Goldberg IJ. Lipoprotein lipase and lipolysis: central roles in lipoprotein metabolism and atherogenesis. J Lipid Res 1996;37:693 – 707.

[9] Rutledge JC, Goldberg IJ. Lipoprotein lipase (LpL) affects low density lipoprotein (LDL) flux through vascular tissue: evidence that LpL increases LDL accumulation in vascular tissue. J Lipid Res 1994;35:1152 – 60.

[10] Stein O, Friedman G, Chajek-Shaul T, Halperin G, Olivecrona T, Stein Y. Transfer of cholesteryl ilnoleyl ether from phos-phatidylcholine and phosphatidylethanolamine liposomes to cul-tured cells catalyzed by lipoprotein lipase. Biochim Biophys Acta 1983;750:306 – 16.

[11] Aviram M, Lund-Katz S, Phillips MC, Chait A. The influence of the triglyceride content of low density lipoprotein on the interac-tion of apolipoprotein B-100 with cells. J Biol Chem 1988;263:16842 – 8.

[12] Chung BH, Segrest JP, Smith K, Griffin FM, Brouillette CG. Lipolytic surface remnants of triglyceride-rich lipoproteins are cytotoxic to macrophages but not in the presence of high density lipoprotein. A possible mechanism of atherogenesis. J Clin Invest 1989;83:1363 – 74.

[13] Khoo JC, Mahoney EM, Witztum JL. Secretion of lipoprotein lipase by macrophages in culture. J Biol Chem 1981;256:7105 – 8. [14] Chait A, Iverius PH, Brunzell JD. Lipoprotein lipase secretion by human monocyte-derived macrophages. J Clin Invest 1982;69:490 – 3.

[15] Vance JE, Khoo JC, Steinberg D. Lipoprotein lipase in cultured pig aortic smooth muscle cells. Arteriosclerosis 1982;2:390 – 5. [16] O’Brien KD, Gordon D, Deeb S, Ferguson M, Chait A.

Lipo-protein lipase is synthesized by macrophage-derived foam cells in human coronary atherosclerotic plaques. J Clin Invest 1992;89:1544 – 50.

[17] Sofer O, Fainaru M, Schafer Z, Goldman R. Regulation of lipoprotein lipase secretion in murine macrophages during foam cell formation in vitro. Effect of triglyceride-rich lipoproteins. Arterioscier Thromb 1992;12:1458 – 66.

[18] Mattsson L, Johansson H, Ottosson M, Bondjers G, Wiklund O. Expression of lipoprotein lipase mRNA and secretion in macrophages isolated from human atherosclerotic aorta. J Clin Invest 1993;92:1759 – 65.

[19] Kitahara M, Eyre HJ, Lynch RE, Rallison ML, Hill HR. Metabolic activity of diabetic monocytes. Diabetes 1980;29:251 – 6.

[20] Katz S, Klein B, Elian I, Fishman P, Djaldetti M. Phagocytotic activity of monocytes from diabetic patients. Diabetes Care 1983;6:479 – 82.

[21] Geister C, Almdal T, Bennedsen J, Rhodes JM, Kolendorf K. Monocyte functions in diabetes mellitus. Acta Pathol Microbiol Immunol Scand [C] 1982;90:33 – 7.

[22] Boyum A. Isolation of lymphocytes, granulocytes and macrophages. Scand J Immunol Suppl 1976;5:9 – 15.

[23] Pertoft H, Johnsson A, Warmegard B, Seljelid R. Separation of human monocytes on density gradients of percoll. J Immunol Methods 1980;33:221 – 9.

[24] Nilsson-Ehle P, Schotz MC. A stable radioactive substrate emul-sion for assay of lipoprotein lipase. J Lipid Res 1976;17:536 – 41. [25] Itaya K. A more sensitive and stable colorimetric determination

of free fatty acids in blood. J Lipid Res 1977;18:663 – 5. [26] Chomczynski P, Sacchi N. Single step method of RNA isolation

by acid guanidinium thiocyanate – phenol – cholroform extrac-tion. Anal Biochem 1987;162:156 – 9.

[27] Wang AM, Doyle MV, Mark DF. Quantitation of mRNA by the polymerase chain reaction [published erratum appears in Proc. Natl. Acad. Sci. USA April, 87(7):2865]. Proc Natl Acad Sci USA 1989;86:9717 – 21.

[28] Schneeberger C, Speiser P, Kury F, Zeillinger R. Quantitative detection of reverse transcriptase-PCR products by means of a novel and sensitive DNA stain. PCR Methods Appl 1995;4:234 – 8.

[29] Simionescu N, Simionescu M. Galloylglucoses of low molecular weight as mordant in electron microscopy. II. The moiety and functional groups possibly involved in the mordanting effect. J Cell Biol 1976;70:622 – 33.

[30] Seigler L, Wu WT. Separation of serum-density lipoprotein for cholesterol determination: ultracentrifugation versus precipita-tion with sodium phosphotungstate and magnesium chloride. Clin Chem 1981;27:838 – 41.

[31] Bradford MM. A rapid and sensitive method the quantitation of microgram quantities of protein utilizing the principle protein-dye binding. Anal Biochem 1976;72:248 – 54.

[32] Ishibashi S, Yamada N, Shimano H, Takaku F, Akanuma Y, Murase T. Composition of very-low-density lipoproteins in non-insulin-dependent diabetes mellitus. Clin Chem 1989;35:808 – 12. [33] Garner B, Baoutina A, Dean RT, Jessup W. Regulation of serum-induced lipid accumulation in human monocyte-derived macrophages by interferon-g. Correlations with apolipoprotein E

production, lipoprotein lipase activity and LDL receptor-related protein expression. Atherosclerosis 1997;128:47 – 58.

[34] Krolewski AS, Warram JH. Joslin’s Diabetes Mellitus. Philadel-phia: Lea and Febiger, 1994:605 – 19.

[35] Behr SR, Kraemer FB. Insulin deficiency decreases lipoprotein lipase secretion by murine macrophages. Diabetes 1988;37:1076 – 81.

[36] Josefsen K, Nielsen H, Lorentzen S, Damsbo P, Buschard K. Circulating monocytes are activated in newly diagnosed type 1 diabetes mellitus patients. Clin Exp Immunol 1994;98:489 – 93. [37] Querfeld U, Ong JM, Prehn J, Carty J, Saffari B, Jordan SC,

Kern PA. Effects of cytokines on the production of lipoprotein lipase in cultured human macrophages. J Lipid Res 1990;31:1379 – 86.

(11)

[38] Friedman G, Chajek-Shaul T, Gallily R, Stein O, Shiloni E, Etienne J, Stein Y. Modulation of lipoprotein lipase activity in mouse peritoneal macrophages by recombinant human tumor necrosis factor. Biochim Biophys Acta 1988;963:201 – 7. [39] Semb H, Olivecrona T. The relation between glycosylation and

activity of guinea pig lipoprotein lipase. J Biol Chem 1989;264:4195 – 200.

[40] Ben-Zeev O, Doolittle MH, Davis RC, Elovson J, Schotz MC. Maturation of lipoprotein lipase. Expression of full catalytic activity requires glucose trimming but not translocation to the

cis-Golgi compartment. J Biol Chem 1992;267:6219 – 27. [41] Sartippour MR, Lambert A, Laframboise M, St-Jaques P,

Reinier G. Stimulatory effect of glucose on macrophage lipo-protein lipase expression and production. Diabetes 1998;47:431 – 8.

[42] Creedon GM, Sexton DM, Collins PB. Lipoprotein lipase ex-pression in human monocyte-derived macrophages. Biochem Soc Trans 1996;24:68S.

[43] Behr SR, Kraemer FB. Effects of activation on lipoprotein lipase secretion by macrophages. Evidence for autoregulation. J Exp Med 1986;164:1362 – 7.

[44] Takahashi S, Kawarabayasi Y, Nakai T, Sakai J, Yamamoto T. Rabbit very low density lipoprotein receptor: a low density lipoprotein receptor-like protein with distinct ligand specificity. Proc Natl Acad Sci USA 1992;89:9252 – 6.

[45] Multhaupt HA, Gafvels ME, Kariko K, Jin H, Arenas-Elliot C, Goldman BI, Strauss JF, 3rd, Angelin B, Warhol MJ, McCrae KR. Expression of very low density lipoprotein receptor in the vascular wall. Analysis of human tissues by in situ hybridization and immunohistochemistry. Am J Pathol 1996;148:1985 – 97.

[46] Saheki S, Hitsumoto Y, Murase M, Takeuchi N, Uchida K. In vitro degradation of very low density lipoprotein from diabetic patients by lipoprotein lipase. Clin Chim Acta 1993;217:105 – 14. [47] Weisweiler P, Schwandt P. Type 1 (insulin-dependent) versus type 2 (non-insulin-dependent diabetes mellitus: characterization of serum lipoprotein alterations. Eur J Clin Invest 1987;17:87 – 91.

[48] Kraemer FB, Chen YD, Lopez RD, Reaven GM. Effects of non-insulin-dependent diabetes mellitus on the uptake of very low density lipoproteins by thioglycolate-elicited mouse peri-toneal macrophages. J Clin Endocrinol Metab 1985;61:335 – 42. [49] Mazzone T, Lopez C, Bergstraesser L. Modification of very low density lipoproteins leads to macrophage scavenger receptor uptake and cholesteryl ester deposition. Arteriosclerosis 1987;7:191 – 6.

[50] Stengel D, Antonucel M, Geoua W, Dachei C, Lasnik P, Hourion D, Ninio E, Chapman MJ, Griglio S. Inhibition of LPL expression in human monocyte-derived macrophages is depen-dent on LDL oxidation state. Arterioscler Thromb Vasc Biol 1998;18:1172 – 80.

[51] Basu SK, Brown MS, Ho YK, Havel RJ, Goldstein JL. Mouse macrophages synthesize and secrete a protein resembling apolipoprotein E. Proc Natl Acad Sci USA 1981;78:7545 – 9. [52] Steinberg FM, Tsai EC, Brunzell JD, Chait A. ApoE enhances

lipid uptake by macrophages in lipoprotein lipase deficiency during pregnancy. J Lipid Res 1996;37:972 – 84.

[53] Evans AJ, Sawyez CG, Wolfe BM, Connelly PW, Maguire GF, Huff MW. Evidence that cholesteryl ester and triglyceride accu-mulation in J774 macrophages induced by very low density lipoprotein subfractions occurs by different mechanisms. J Lipid Res 1993;34:703 – 17.

(1)

32% lower in NIDDM group (Table 3). In addition,

results indicated that for both group, freshly isolated

monocytes has lower values for the total cholesterol

(114

9

27 mg per mg cell protein) and triglycerides

(270

954 mg per mg cell protein), respectively. As

ob-served from the above data, the main intracellular lipid

pool is represented by triglycerides (especially in MDM

from NIDDM group), which is in accordance with the

results reported by others for MDM of healthy human

subjects [33]. To test the influence of autologous serum

on lipid loading of MDM, cross-experiments were

per-formed, in which the serum from controls was used to

culture cells of NIDDM and vice versa. Data about the

intracellular cholesterol and triglycerides mass in

macrophages from NIDDM and control subjects

culti-vated in autologous serum are set out in Table 3. When

the serum was switched between the cells isolated from

the two groups, the results showed that MDM from

NIDDM patients were still loaded with lipids in day 7

of culture (735

997 mg triglyceride per mg cell protein

and 237.5

964 mg cholesterol per mg cell protein),

while the MDM from controls displayed only a

moder-ate lipid loading when grown in NIDDM

serum-con-taining medium (3859

66 mg triglyceride per mg cell

protein and 360

9

37 mg cholesterol per mg cell

protein). These data may suggest a different mechanism

of lipid uptake and/or metabolism in macrophages of

NIDDM subjects as compared to controls. The

poten-tial correlations between each of the total cholesterol

and triglycerides concentrations in serum and their

respective intracellular mass for each individual within

the two subject groups were further investigated (Fig.

4). For the control group, there is a lack of correlation

between the serum cholesterol and triglycerides and

their intracellular counterparts (P=

0.97,

R=

0.01 and

P

=0.65,

R

=0.13, respectively) (Fig. 4A and B).

MDM of NIDDM patients also exhibit a weak

correla-tion with respect to serum versus intracellular

choles-terol (P

=0.2,

R=0.35) (Fig. 4D). In contradistinction,

serum and intracellular triglycerides are very well

corre-lated in macrophages of NIDDM group (P

=0.0001,

R

=0.83) (Fig. 4C). These data bring further support

for possible alteration in triglyceride metabolism of

diabetic MDM. The prominent lipid loading of cells

from diabetic patients is accompanied by a decrease in

the expression and activity of LPL as described above.

Fig. 5 shows a good correlation between the triglyceride

loading of macrophages and the LPL activity in the cell

medium. Both in control group (Fig. 5A) and in

NIDDM group (Fig. 5B) higher LPL activity values

corresponded

to

lower

intracellular

triglyceride

amounts, leading to a progressive decrease of LPL

(2)

Table 3

Intracellular neutral lipid content of monocyte-derived macrophages isolated from NIDDM and controlsa

Controls (mg per mg cell protein) Significance Lipid class NIDDM (mg per mg cell protein)

5009130

Triglycerides 8379110 P=0.05

Total cholesterol 287.5955 4209107 P=0.28

a_{Lipids were extracted from monocyte-derived macrophages after 7 days in culture, and quantitated using enzymatic kits. Values are given as}

mean 9S.D. (n=13). The ‘P’ value is calculated using one-way ANOVA test. NIDDM, non-insulin-dependent diabetic patients with cardiovascular disease; controls, patients with cardiovascular disease and without diabetes.

activity while cells became loaded with lipids. For both

groups the values were well corrected (r

=0.657 for

controls and

r

=0.763 for NIDDM) and ‘P’ is less than

0.03 as determined by ANOVA analysis.

4. Discussion

In patients with NIDDM, macrovascular

complica-tions such as coronary artery and peripheral vascular

disorders are the major health problems; in Caucasian

patients with NIDDM their contribution to mortality

overrides the impact of microvascular complications

[34]. In order to emphasise the role of long lasting

diabetic state on lipid metabolism of macrophages, we

selected for this study one group of patients with

non-insulin-dependent diabetes mellitus and ischemic

heart diseases and one control group represented by

patients with the same type of ischemia, but without

diabetes. Our results indicated that the activity of LPL

secreted by macrophages of NIDDM patients is 42%

decreased as compared to controls. A similar

diminu-tion in LPL activity is also reported for murine

macrophages in streptozotocin-induced diabetes [35].

There are several possible reasons, which could account

for the observed change in LPL activity. First, the high

blood glucose concentration in NIDDM may serve as

an alternative to free fatty acids as source of energy for

the cell, leading to decreased synthesis and activity of

LPL. This is also supported by the fact that, among all

serum biochemical parameters that we have tested,

fasting glycemia was the only statistically different one.

Furthermore, our current experiments on murine

peri-toneal macrophages cultured in RPMI supplemented

with 0.1 and 0.2% glucose revealed a two-fold decrease

in LPL activity in the latter condition. Another possible

explanation is the activated state of monocytes in

dia-betic subjects [36], which induces the secretion of

sev-eral inflammatory cytokines such as IL-1, TNF or

IFN

, that in turn, were shown to suppress secreted

LPL activity in human MDM [37] or mouse peritoneal

macrophages [38]. Suppression of LPL activity in

murine peritoneal macrophages loaded with triglyceride

[17] or in macrophage-derived foam cells isolated from

the arterial wall of human subjects [18] were also

re-ported. These data are in accordance to our results, in

which MDM from NIDDM patients exhibit a

de-creased secreted LPL activity concomitantly with a

foam cell phenotype (after 7 days in culture) as

com-pared to MDM from controls, which are only

moder-ately loaded with triglycerides. Thus, a massive

triglyceride loading may be partly responsible for the

decreased LPL activity observed in MDM of NIDDM

patients.

Our results also showed that LPL mRNA synthesis is

six-fold reduced in MDM from NIDDM versus

con-trols, while the activity is only

two-fold decreased. A

guinea pig during fasting [39]. The possible

explana-tions may be the existence of a pool of inactive LPL

stored in a high-mannose form within the endoplasmic

reticulum [40] of control MDM or reduced stability of

LPL mRNA in macrophages of NIDDM patients. A

similar discrepancy was reported for J744 macrophages

grown in high versus normal glucose medium, and the

authors suggest that post-translational modifications

may account for the difference observed [41]. Finally,

we may speculate that due to a higher efficiency of the

diabetic LPL than that of the normal enzyme in

de-grading the substrate, the specific activity of the LPL

from diabetics is actually more than two-fold decreased.

The results presented here indicate that the NIDDM

macrophages are prone to become loaded with lipids,

mainly as triglycerides. Recent data [33] show that

human monocyte derived macrophages from healthy

subjects cultured in RPMI with 10% human serum, but

not foetal calf serum, turn into triglyceride-rich foam

cells, with no change in LPL activity. Also there is one

report about an increase, rather than a decrease of LPL

activity in macrophages isolated from diabetic patients

compared to healthy controls [42]. In contrast to these

results, we found a selective lipid loading of MDM

accompanied by a decrease in LPL activity and

expres-sion. This behaviour of macrophages obtained from

NIDDM patients is in accordance with previous

obser-vations [17,18] that indicated that macrophage

derived-foam cells express lower LPL mRNA and activity. On

the other hand, Mattsson et al. [18] showed that only

the CD14 positive cells from human atheroma exhibit a

decreased LPL activity and lipid loading; these cells are

(3)

capable of phagocytosis. Although scarce, there are

some data in the literature to document an increased

phagocytic capacity of human monocytes in diabetes

[36]. Thus, we cannot rule out the possibility of a

different number of CD14 positive cells among MDM

population of NIDDN versus control patients.

More-over, Behr et al. [43] have shown that a high degree of

activation of monocytes/macrophages leads to a

de-creased LPL activity.

Another observation concerns the strong correlation

that exists between serum and intracellular triglyceride

in MDM from NIDDM, but not in control patients.

This suggests that NIDDM macrophages take up

triglyceride-rich lipoproteins via a non-down regulated

pathway. One of these pathways could involve the

VLDL receptor [44], recently demonstrated by in situ

hybridisation

and

immunohistochemistry

on

macrophages and endothelium of human arteries [45].

The decreased LPL activity in NIDDM – MDM versus

controls, together with a lower LPL affinity for diabetic

VLDL, as reported in [46] may lead to a higher

concen-tration of intact VLDL in the medium, and thus an

increased number of ligands available for the VLDL

receptor, that could eventually lead to the foam cell

formation by different mechanisms. Furthermore,

dia-betic patients have been noted to have elevations of

(4)

abnormal triglyceride-rich lipoproteins that are

en-riched in apoE [47] and are more avidly taken up by

macrophages [48]. On the other hand, we may speculate

that, due to an increased activation reported for

dia-betic monocytes [37], the cells isolated from NIDDM

patients had an increased capacity of oxidising

lipo-proteins in serum-containing culture media that, in

turn, are recognised by the macrophage scavenger

re-ceptors, leading to fast lipid loading of the cells [49]. In

support of this hypothesis, and of our data there is a

recent report showing that inhibition of LPL expression

in human monocyte-derived macrophages is dependent

on LDL oxidation [50]. Another possible explanation

for triglyceride loading of macrophages isolated from

NIDDM patients is related to the apoE secreted by the

cells [51]. A recent study demonstrated that apoE

en-hances lipid uptake by macrophages in LPL deficiency

[52]. On the other hand, Evans et al. [53] showed that

extracellular lipolysis of VLDL subfractions by LPL is

necessary for lipid loading of J744A.1 macrophages via

a LDL receptor-mediated mechanism. Our data

show-ing a decreased LPL secretion in diabetic macrophages

may suggest though, that lipoproteins could be taken

up with decreased efficiency by the cells via the LDL

receptor. However, as a result of this decreased uptake,

the local residence time of the lipoproteins would be

prolonged, making modification (lipid peroxidation)

and uptake by scavenger receptors more likely.

Collec-tively, these data may provide a plausible explanation

as to why a reduced LPL activity in NIDDM derived

macrophages may be able to potentiate triglycerides

and, to a lesser extent, cholesterol loading of the cells.

In conclusion, the results reported here indicate that,

(i) monocyte-derived macrophages from NIDDM

pa-tients became foam cells after 7 days in culture, whereas

in the same conditions MDM from control patients are

only moderately loaded with lipids; (ii) in diabetic

patients

the

accumulation

of

triglycerides

in

macrophages is well correlated with triglyceridemia;

and (iii) macrophage-derived foam cell formation

coin-cides with a decreased LPL mRNA expression and

enzymatic activity in these cells.

We can postulate that in diabetic state, the decreased

synthesis of LPL may cause accumulation of

triglyce-rides in macrophages that ultimately accelerates their

transformation into foam cells thus contributing to the

rapid formation of atheroma and the accelerated

atherosclerosis characteristic for diabetic condition.

Acknowledgements

We gratefully acknowledge to Dr Georg Breier and

Dr Werner Risau at Max Planck Institut fur

Physiolo-gische und Klinische Forschung, Bad Nauheim,

Ger-many for precious advice on RT-PCR technique. We

(5)

also thank loana Manolescu, Cristina Dobre, Elena

Florea, Manda Misici, Nicoleta Mobre and Mihaela

Schean for the qualified assistance. This work was

supported by a grant from the Romanian Academy and

an UNESCO I-Molecular and Cell Biology Network

Research Grant, 1999.

References