Since the U937 cells lack the 3-ketosteroid reductase activity needed for cholesterol synthesis, their growth is dependent on the exogenous cholesterol uptake. To determine to what extent the U937 cell proliferation is dependent on the presence of native LDL cholesterol in the culture medium, cells (2.0 x 105/ml) were seeded out in RPMI 1640 containing various amounts of LDL cholesterol. At cholesterol concentrations below 1 µg/ml , cell proliferation was similar to that observed with RPMI 1640 only (data not shown), while increasing concentrations of LDL cholesterol from 2 to 20 µg/ml caused a marked increase of cell growth (Study I Fig.1). Increments in LDL cholesterol above 20 µg/ml did not further increase U937 cell proliferation at this cell density.
In order to further clarify if the cholesterol delivery into cells is dependent on the efficient binding of LDL apoB to its receptors, we investigated the effects of methylated LDL and familial defective apolipoprotein B-100 (FDB) LDL (with known defective apoB binding affinity) on the cell proliferation. Methylated LDLs and LDLs from FDB patients were less effective in supporting cell proliferation compared with native LDL (Study I) .
This study demonstrated that LDL can transport cholesterol to cells for their growth via LDL receptor mediated pathway.
Small but significant amounts of E2 in esterified form could be quantitated in LDL and HDL after isolation from plasma incubation with 1,000 nmol/L or more of free E2 (Table 3).
Table 3. Content of free and esterified E2 in LDL and HDL following plasma incubation
LDL and HDL samples were isolated and gel-filtered following incubation of plasma with free E2 for 4 hours at 37°C. The separation and quantification of free and esterified E2 in LDL and HDL were carried out as described in the method section. Data are represented as the mean ± SD of five experiments with samples from different donors. *P< 0.05, **P< 0.01 vs. control (paired t test).
To demonstrate the possible role of LCAT in the esterification of E2, we incubated plasma with 100 µmol/L of unesterified E2 in the presence of 1.4 mmol/L of DTNB, a sulphydryl reagent also known to inhibit LCAT (Jauhiainen et al. 1989). The esterified E2 detected in LDL and HDL were reduced by approximately 97% (Study IV, Table 5). Small proportions of esterified E2 were also detected in HDL following incubation of isolated HDL (1 mg protein) with 3,000 nmol of unesterified E2in Celite dispersion (Study IV, Table 7).
The amounts of free and esterified isoflavones incorporated into LDL following in vitro incubation in Celite dispersion are summarized in Table 4 (Studies II, III). In addition, some unpublished data is included in this table. The mean molar ratios of incorporated free genistein and daidzein per mole of LDL were 0.07 and 0.22, respectively. Among esterified isoflavones, genistein and daidzein stearates were least incorporated, with molar ratios ranging from 0.06 to 0.43. Incorporation of both genistein and daidzein oleates into LDL was more effective, with molar ratios ranging between 2.43 and 7.87. The incorporation of daidzein dilinoleate was the most effective, with the molar ratio reaching 17.
Table 4. Incorporation of unesterified and esterified isoflavones into LDL in Celite dispersion
LDL (1mg protein) was incubated with 3,000 nmol of various isoflavones in Celite dispersion for 22 hours at 37°C. The isoflavones incorporated into the LDL particles were quantified by GC-MS using deuterated internal standards. Values are means ± SD of seven to eight independent experiments (from Studies II,III and other unpublished data). The isoflavone-ester described in this thesis is same as isoflavone-O-ester in Study III.
The incorporation of free or esterified E2 into LDL and HDL particles was analyzed following incubation in plasma. Only trace amounts of free E2 could be detected in gel-filtered LDL. However, small amounts of E2 in esterified form could be quantitated in LDL following addition of 1,000 nmol/L or more of free E2 to plasma (Table 3). Incubation of plasma with increasing amounts of E2 17-stearate resulted in increasing proportions of E2 in the ester fraction in LDL. With 100 nmol/L of E2 17-stearate significant incorporation was achieved compared with control (Study IV, Table 3).
Only negligible amounts of free E2were detected in HDL following incubation with plasma, but small amounts of esterified E2 could be demonstrated in this lipoprotein after addition of 1,000 nmol/L or more of free E2. Incubation with E2 17-stearate resulted in incorporation of E2 ester into HDL comparable to that in LDL following incubation with E2 17-stearate (Study IV, Table 4).
Substantial amounts of E2 esters were detected in LDL following incubation with E2 17-stearate and E2 17-oleate in Celite dispersion, but not with free E2 (Study IV, Table 6). The incorporation of E2 17-stearate and E2 17-oleate into LDL was proportional to the concentration added to the Celite dispersion, with values reaching 1.38 ± 0.26 molecules/LDL particle for 3,000 nmol of E2 17-stearate and 1.93 ± 0.36 molecules/LDL particle for 3,000 nmol of E2 17-oleate.
In similar studies with HDL, both E217-stearate and E2 17-oleate were efficiently incorporated into this lipoprotein, the amounts reaching 4,125 pmol/mg HDL protein for 3,000 nmol of E2 17-stearate added and 8,233 pmol/mg HDL protein for 3,000 nmol E2 17-oleate added (Study IV, Table 7). The incorporation ratio of E2 esters in HDL was higher than that in LDL following Celite incubation (Study IV, Table 6).
The incorporation of isoflavone or E2 esters did not change the LDL lipid or apoprotein mass composition (cholesterol, phospholipids, triglyceride, and apoB) as compared with control LDL that was processed in the same way except in the absence of isoflavones or E2. Incubation of LDL with isoflavones or estrogens did not result in detectable apoB-100 degradation or other changes as detected on 3% SDS-PAGE (data not shown).
The U937 cell culture method (Study I) was originally set up in order to investigate the effect of apoB mutations on the receptor binding of LDL. As this method also provides a possibility to investigate cellular effects of bioactive substances carried in LDL particles, we emphasized the system for studying effects of isoflavones.
The U937 cell proliferation was remarkably inhibited by free genistein and daidzein added directly to the cell culture medium in the presence of native LDL, although the lowest concentrations (0.1-0.5 µmol/L) of genistein induced a temporary increase in cell growth. The inhibitory effect was concentration dependent, with a 50% inhibitory concentration (IC50) of 16.0 µmol/L for genistein and 39.6 µmol/L for daidzein (Study II, Fig. 2).
The LDLs incorporated with small amounts of unesterified genistein, daidzein or any of the genistein stearates did not exhibit inhibitory effects on U937 cell proliferation. However, the LDLs containing some isoflavone mono- and dioleates and dilinoleate significantly reduced the cell proliferation by 36% (genistein 7-oleate), 43% (genistein 4´,7-dioleate), 33% (daidzein 7-oleate), 30% (daidzein 4´,7-dioleate), and 93% (daidzein 4´,7-dilinoleate), compared with control LDL (Study II, Fig. 3). The corresponding concentrations of these isoflavone esters in cell culture medium calculated from their content in LDL were 0.36, 0.42, 0.24, 0.22, and 1.36 µmol/L, respectively. Neither of the 4´-oleates of genistein and daidzein contained in LDLs inhibited cell proliferation although relatively greater amounts of them were incorporated into LDL (Study II, Table 1, Fig. 3).
The Cu2+-initiated oxidation of LDL was significantly inhibited in a concentration-dependent manner by adding free genistein or daidzein directly to the LDL oxidation mixture (Study III, Table 2). Incubation of LDL with unesterified isoflavones or genistein stearic acid esters did not result in prolongation of lag times compared with control LDL incubated in the absence of any isoflavone (Study III, Table 3). However, the LDLs containing 4´,7-dioleates of daidzein and genistein significantly prolonged lag times by 46% (P<0.05) and 202.1% (P<0.01), respectively (Study III, Table 3, Figs. 2,3). In addition, a small but significant increase in lag time (20.5%, P<0.01) was brought about by daidzein 7-oleate. A trend prolongation of lag time was induced by genistein 7-oleate but did not reach statistically significance. In contrast, neither of the 4´-oleates of daidzein and genistein incorporated into LDL could protect this lipoprotein against oxidation as judged by the lag time. There was no correlation between the concentrations of any of the isoflavones in LDL particles and lag times.
The Cu2+-induced LDL oxidation was significantly inhibited in a dose-dependent manner by adding free E2 directly to the LDL oxidation mixture (Study IV, Table 1). The minimum effective concentration of E2 for oxidation inhibition was 100 nmol/L, with 22% increase (P=0.024) in lag time compared with control.
LDL isolated after incubation of plasma with free E2 or E2 17-stearate (maximum concentration, 5000 nmol/L) did not prolong the lag time significantly compared with controls (Study IV, Table 3). However, significant lag time prolongations were observed for LDL (1 mg protein) incubated with 500 and 3000 nmol E2 17-stearate or E2 17-oleate but not with unesterified E2 in the Celite dispersion (Study IV, Table 6, Figs. 1A,B). There was a positive correlation between the lag time and E2 -17-stearate or E2 17-oleate concentration in LDL (r=0.979, P< 0.001, n=8).
Similarly, no oxidation resistance of HDL could be detected either after isolation of HDL from the same plasma incubation with estrogens as LDL (Study IV, Figs. 2A,B). However, prolongations of lag time were observed when greater amounts of E2 17-stearate and E2 17-oleate were incorporated into HDL using the Celite transfer system but not with unesterified E2 (Study IV, Figs. 2C,D).