Very-low-density lipoprotein mediates transcriptional regulation of aldosterone synthase in human adrenocortical cells through multiple signaling pathways
Sarama Saha · Stefan R. Bornstein · Juergen Graessler · Steffi Kopprasch
Received: 1 November 2011 /Accepted: 25 January 2012 /Published online: 14 February 2012 C Springer-Verlag 2012
Abstract Diabetic dyslipidemia is characterized by increased circulatory very-low-density lipoprotein (VLDL) levels. Aldosterone, apart from its role in fluid and electrolyte ho- meostasis, has also been implicated in insulin resistance and myocardial fibrosis. The impact of VLDL as a potential risk factor for aldosterone-mediated cardiovascular injury in dia- betes mellitus, however, remains to be investigated. We have therefore studied native and modified VLDL-mediated ste- roidogenesis and its underlying molecular mechanisms in human adrenocortical carcinoma cells, NCI H295R. Native VLDL (natVLDL), isolated from healthy volunteers, was subjected to in vitro modification with glucose (200 mmol/l) or sodium hypochlorite (1.5 mmol/l) for preparation of glyco- xidized and oxidized VLDL, respectively. VLDL treatment induced steroidogenesis in both a concentration- and time- dependent manner. Native and glycoxidized VLDL (50 µg/ ml) were almost two-fold more potent in adrenocortical aldo- sterone release than angiotensin II (100 nmol/l). These forms of VLDL significantly augmented transcriptional regulation of aldosterone synthase (Cyp11B2), partially through scaven- ger receptor class B type I, as evident from the effect of BLT-1. In contrast to glycoxidized VLDL, oxidized VLDL signifi- cantly attenuated the stimulatory effect of natVLDL on adre- nocortical hormone synthesis. Moreover, treatment with specific pharmacological inhibitors (H89, U0126, AG490) provided supporting evidence that VLDL, irrespective of modification, presumably recruited PKA, ERK1/2 and Jak-2 for steroid hormone release through modulation of Cyp11B2
mRNA level. In conclusion, this study demonstrates a novel insight into intracellular mechanism of VLDL-mediated aldo- sterone synthesis through transcriptional regulation of ste- roidogenic acute regulatory protein (StAR) and Cyp11B2 expression in human adrenocortical carcinoma cell line.
Keywords VLDL · Aldosterone . MAP kinase . Janus kinase · Aldosterone synthase
Introduction
Type 2 Diabetes Mellitus (T2D) is characterized by hyper- glycemia and insulin resistance and is known to be associ- ated with dyslipidemia, hypertension and central obesity (Brunzell et al. 2008). Diabetic dyslipidemia is associated with persistently high circulatory levels of very-low-density lipoprotein (VLDL), resulting from increased hepatic pro- duction as well as decreased hepatic uptake of VLDL (Adiels et al. 2008). VLDL, a triglyceride-rich lipoprotein, has been postulated as the principal determinant of low- density lipoprotein (LDL) particle size. The intricate asso- ciation of high circulatory VLDL and LDL levels together with low concentrations of high-density lipoprotein (HDL) has been considered to be a major risk factor (although not solely responsible) for cardiovascular disorders and contrib- utes to the pathology in metabolic syndrome (Adiels et al. 2008). Furthermore, VLDL is able to promote atherosclero- sis through the up-regulated expression of endothelial adhe- sion molecules and enhanced macrophage chemotaxis (Kannel and Vasan 2009).
VLDL is a heterogeneous quasispherical particle com- prised in approximately 10% proteins (Cushley and Okon 2002). Phosphatidyl inositol (PI)-3 kinase-mediated addi- tion of cytosolic triglyceride (TG) to apoprotein B (apo-B)
S. Saha () · S. R. Bornstein · J. Graessler . S. Kopprasch Department of Internal Medicine III, Carl Gustav Carus Medical School, Technical University of Dresden, Fetscherstraße 74,
leads to the formation of matured VLDL in the liver (Stillemark-Billton et al. 2005). In addition to TG, VLDL also contains free and esterified cholesterol as well as var- ious apoproteins, including apo-B100, apo-C and apo-E (Dinkel et al. 2006).
Other lipoproteins, HDL and LDL, constitute major po- tential sources of cholesterol for aldosterone biosynthesis in the adrenocortical zona glomerulosa cell layer, which is characterized by selective expression of aldosterone syn- thase (Capponi 2002). Aldosterone, an important compo- nent of the renin-angiotensin-aldosterone system (RAAS) is implicated in the cardiovascular homeostasis; it is also at- tributed to the development of cardiovascular injury by facilitating the formation of myocardial fibrosis and athero- sclerosis (Krug and Ehrhart-Bornstein 2008). Although the VLDL TG content is utilized for energy production, the impact of its cholesterol content in aldosterone biosynthesis has not been delineated.
In hyperglycemic condition, LDL is susceptible to undergo oxidative and glycoxidative modification, which might con- tribute to the development of atherosclerosis in T2D individ- uals (Graessler et al. 2007). Furthermore, T2D individuals produce more VLDL1 particles (mature, triglyceride-rich VLDL) than nondiabetic controls (Adiels et al. 2006). Therefore, it can be hypothesized that VLDL plays an impor- tant role in modulating aldosterone release in hyperglycemic and oxidative stress conditions because of its enhanced circu- lating level and biochemical modification induced by the disease process itself. Moreover, in contrast to the native and modified LDL and HDL, biological characteristics of native and biochemically modified VLDL are not well defined. The present study sought to unravel the role of these VLDL in aldosterone biosynthesis and to present a potential link to aldosterone-mediated pathophysiological consequences in T2D.
The cell signaling cascade for aldosterone synthesis used by the primary mediators [such as angiotensin II (AngII), adrenocorticotropic hormone (ACTH) and potassium (K+)] has been well investigated. AngII and K+ involve protein kinase C (PKC) and calmodulin kinase II (CaMK II), re- spectively, for activation of steroid synthesis machinery. Conversely, ACTH accomplishes the same via cyclic aden- osine monophosphate (cAMP)-mediated protein kinase A (PKA) activation (Quinn and Williams 1988). HDL triggers multiple intracellular signal transduction cascades, involv- ing the activation of phospholipase C, PKC, mitogen- activated protein (MAP) kinase (Nofer et al. 2000), tyrosine kinase Src (Frias et al. 2009) and calmodulin kinase (Xing et al. 2011). LDL also recruits extracellular signal-regulated kinase (ERK1/2) for steroidogenesis (Ansurudeen et al. 2010).
This study demonstrates for the first time that native and modified VLDL are able to stimulate aldosterone synthase
via PKA, ERK1/2 and Janus kinase-2 (Jak-2) signaling cascades in human NCI H295R cells.
Materials and methods
Preparation of VLDL
Native VLDL (natVLDL) was isolated from overnight fast- ing normolipidemic and normoglycemic healthy volunteers using very fast density gradient ultracentrifugation as de- scribed previously (Pietzsch et al. 1995). After dilution with phosphate buffered saline (PBS) to 0.3 g/l protein, natVLDL solutions were placed in a dialysis tube and glycoxidized at 37°℃ with 200 mmol/l glucose containing 200 umol/ l EDTA and 1 mg/l NaN3 for 6 days (with change of glucose dialysis buffer every 48 h). After glucose treatment, lipo- protein preparations were extensively dialyzed against PBS. VLDL control preparations were treated in the same manner but without glucose. Oxidation of VLDL was carried out by adding sodium hypochlorite (1.5 mmol/l) to natVLDL at 37°℃ for 40 min. Preliminary experiments showed that the initial NaOCl content declined exponentially during oxida- tion and was no more detectable after 30 min.
Biochemical characterization of native and modified VLDL
The glucose- and hypochlorite-induced VLDL protein mod- ification was assessed by determination of fluorescent prod- ucts with an emission maximum at 430 nm when excited at 365 nm (Maeba et al. 1994) and by carbonyl formation according to Levine et al. (1990). Oxidative and glycoxida- tive changes in the VLDL lipid fractions were measured by accumulation of thiobarbituric acid-reactive substances (TBARS) as described by Yagi (1976).
Cell culture
NCI H295R cells were routinely propagated in DMEM (Sigma)/ F12 (Invitrogen) medium, supplemented with Insulin (66 umol/l), Hydrocortisone (10 umol/l), 17ß estra- diol (10 µmol/l), apo-transferrin (10 µg/ml), sodium selenite (30 µmol/l), 2% FCS (Biochrom) along with penicillin (100 units) and streptomycin (100 µg/ml) at 37℃ in a humidified atmosphere of 95% air with 5% CO2. Then, 80% of conflu- ent cells were treated either with one of the lipoprotein preparations such as natVLDL, glycoxidized VLDL (glycox VLDL), oxidized VLDL (oxVLDL), or AngII (100 nmol/l; Sigma Aldrich) or a respective vehicle such as PBS or dimethyl sufoxide (DMSO) for 24 h in serum-free media (SFM) with or without various inhibitors. Of note, H295R cells were always serum-starved for 24 h prior to the specific treatment with VLDL in the presence or absence of
pharmacological inhibitors. Aldosterone was measured in the conditioned media and cells were used for protein ex- traction and RNA isolation. The highly specific scavenger receptor class B type I (SR-BI) inhibitor BLT-1 (Block Lipid Transport-1, 2-Hexyl-1-cyclopentanone thiosemicarbazone) was obtained from Chembridge (San Diego, CA, USA). Other specific pharmacological inhibitors, such as Rp- CAMP, H89 (N-[2-(p-Bromocinnamylamino)ethyl]-5- isoquinolinesulfonamide·2HCl), U0126 (1,4-Diamino-2,3- dicyano-1,4-bis(2aminophenylthio)butadiene), SB203580 (4-(4-Fluorophenyl)-2-(4-methylsulfinylphenyl)-5-(4-pyr- idyl)1H-imidazole, HCl) and Janus kinase-2 (Jak-2) inhibi- tor, AG490 (N-Benzyl-3,4-dihydroxy-a-cyanocinnamide), were procured from Calbiochem (Darmstadt, Germany).
Aldosterone measurement
After treatment, aldosterone release in cell culture medium was determined in duplicate by radioimmunoassay (RIA) using Diagnostic Systems Laboratories kit.
Western blotting analysis
Following specific treatment, H295R cells were lysed with cell lytic M reagent (Sigma), containing 1% (v/v) protease inhibitor cocktail (Sigma-Aldrich). Protein concentration was quantified by bicinchoninic acid (BCA)TM protein assay kit (Pierce, USA). The denatured proteins were resolved by SDS-polyacrylamide gel electrophoresis (Laemmly 1970). The separated protein was then transferred onto a polyviny- lidene difluoride membrane (Roti-PVDF). After blocking with 5% BSA for 1 h, membranes were probed overnight with different antibodies for specific proteins such as phospho-extracellular signal regulated kinases (pERK1/2), ERK1/2 and phospho-signal transducer and activator of tran- scription (pSTAT3) (1:1,000; Cell Signaling Technologies, Germany). Normalization of protein loading was assessed using ß-actin antibody (1:1,000; Cell Signaling) as a control. The membranes were then reacted with peroxidase- conjugated secondary antibody (Bio Rad 1:6,000) and the signals were visualized by chemiluminescence reactions using Supersignal West Pico Chemiluminescent substrate kit (Pierce) and the Gene Genome bio-imager.
RNA isolation and quantitative real-time PCR
RNA isolation was carried out from treated H295R cells, using High Pure RNA Isolation kit (Roche). cDNA was prepared from 500 ng of total RNA by reverse transcription with M-MLV Reverse Transcriptase (Invitrogen) and oligo (dT) 12-18 primer (Invitrogen). Subsequently, real-time PCR was performed in order to quantify cDNA using LightCycler 480 SYBR Green I Master kit (Roche).
Primers used for amplification of the specific segments of respective cDNA are listed in Table 1. Differences in gene expression, expressed as fold of change, were calculated using the 2-4Act method where human 3-actin was used as a housekeeping gene.
Statistics
Statistical analysis was performed by one-way analysis of variance (ANOVA) using Statistical Package for Social Sciences (SPSS) followed by post hoc Dunnet’s T3 test in case of biochemical characterization of native and modified VLDL. Comparison between two groups was carried out by Student’s t test. A value less than 0.05 was considered to be statistically significant. All data are presented as mean ± standard error of mean (SEM).
Results
Biochemical characterization of native and modified VLDL
Biochemical characterization of native and modified lipo- protein preparations is summarized in Table 2. While pro- tein carbonyls are formed by a variety of oxidative mechanisms and y-glutamyl semialdehyde appears to be the major residue (Levine et al. 1990), the fluorescence increase at 365/430 nm is mainly due to modifications of the apolipoprotein lysine residues. Hypochlorite-induced oxidation of the VLDL protein components was character- ized by a pronounced 4.2-fold elevation of protein carbonyl levels whereas glucose modification resulted in a slight but significant elevation of fluorescence products and 1.7-fold increase in protein carbonyl content (p<0.05). In addition to protein modification, VLDL oxidation and glycoxidation were followed by significant lipid oxidation as reflected by the accumulation of TBARS levels (see Table 2).
VLDL mediates dose-dependent and time-dependent increase in aldosterone release
H295R cells were treated with different concentrations (1, 10 and 50 µg/ml) of VLDL preparations for 24 h. The supernatant was subsequently collected for measurement
| hß-actin-F | CCAACCGCGAGAAGATGA |
| h3-actin-R | CCAGAGGCGTACAGGGATAG |
| hCyp11B2-F | CCTGTTGAAGGCGGAACTGTCACTA |
| hCyp11B2-R | AAAGAGCGTCATCAGCAAGGGAAAC |
| hStAR-F | TGGCTGGAAGTCCCTCTAAGACCAA |
| hStAR-R | TTGCAGGCTTCCAGTAGGGATTCTC |
| Protein modification | Lipid modification TBARS (umol/g) | ||
|---|---|---|---|
| RF % (365/430 nm) | PC (umol/g) | ||
| natVLDL | 25.0±1.0 | 10.6±1.5 | 2.7±0.3 |
| ox VLDL | 29.4±4.0NS | 44.8±3.5 *** | 5.7±0.7* |
| glycox VLDL | 30.1±1.3* | 18.5±2.0* | 4.9±0.3 *** |
Data are means ± SEM of 6 to 12 lipoprotein preparations RF Relative fluorescence, PC protein carbonyl content, TBARS thio- barbituric acid-reactive substances
*p<0.05, *** p<0.001 as compared with natVLDL (univariate ANOVA with post hoc Dunnet-T3 test), NS not significant
of aldosterone by RIA. Figure 1a shows that natVLDL induced a dose-dependent increase in aldosterone release by 1.6-, 5.5- and 16-fold, respectively, compared to the basal aldosterone secretion from untreated cells (referred to as con- trol). Glycox VLDL (50 µg/ml) produced a 15-fold increase, whereas ox VLDL could induce only a 4-fold enhancement in comparison with controls. These results suggest that
a
800
Aldosterone (pmol/l)
700
1 µg/ml
600
10µg/ml
500
50µg/ml
400
300
§
200
100
0
C Angll
natVLDL
glycoxVLDL
OxVLDL
Aldosterone (pmol/l)
1000
800
C
natVLDL
600
glycoxVLDL
400
OxVLDL
200
0
1h
3h
6h
12h
24h
glycoxidative modification causes only a mild attenuation, whereas oxidative modification induces a pronounced reduc- tion of the stimulatory effect of native VLDL-mediated aldo- sterone secretion.
In order to study the kinetics of VLDL-mediated aldoste- rone release, H295R cells were incubated with 50 µg/ml of native and modified preparations of VLDL for different time periods from 1 to 24 h; the results are presented in Fig. 1b. All forms of VLDL produced a 24 h gradual increase in aldosterone release with a minimum significant measurable amount obtained after 1 h, implying that VLDL is able to induce both early as well as prolonged effects on adreno- cortical steroidogenesis.
VLDL produces its effect by transcriptional regulation of steroidogenic acute regulatory protein (StAR) and aldosterone synthase (Cyp11B2)
It has been established that LDL (Ansurudeen et al. 2010) and HDL, specifically HDL2 fraction (Xing et al. 2011), stimulate aldosterone production through increased expres- sion of aldosterone synthase (Cyp11B2). In order to study the impact of VLDL on the regulation of steroidogenic proteins at the molecular level, H295R cells were incubated with native and modified VLDL for 24 h in serum-free media (SFM). The cells were then utilized for RNA isola- tion, followed by quantification of mRNA abundance by real-time PCR. NatVLDL and glycox VLDL significantly increased the StAR mRNA levels (Fig. 2a) by 1.4- and 1.3- fold, respectively and the Cyp11B2 levels (Fig. 2b) by 18- fold and 15.5-fold, respectively. On the other hand, ox VLDL induced only a minor increase in Cyp11B2 mRNA expression (Fig. 2b). This result could, at least in part, explain the observed lower effect of oxVLDL on aldosterone release in comparison with native and glycoxi- dized VLDL-mediated steroid secretion.
VLDL induces aldosterone secretion and transcriptional regulation of aldosterone synthase through SR-BI
In order to investigate the role of SR-BI for selective deliv- ery of cholesterol from VLDL to adrenocortical cells for steroidogenesis, serum-starved H295R cells were stimulated with native or modified VLDL (50 µg/ml) with or without BLT-1 (16 µM) for 24 h in SFM. The conditioned medium was collected for aldosterone measurement. BLT-1 caused a significant inhibition of aldosterone secretion after co- incubation with native or glycoxidized VLDL by 70 and 67%, respectively (Fig. 3a). Conversely, BLT-1 could not produce any significant effect on ox VLDL-mediated hor- mone release, implying that oxidized VLDL-triggered cholesterol uptake depends mainly on alternative lipopro- tein receptors including the scavenger receptors SR-A,
a
1.6
1.4
*
*
T
*
¥
StAR mRNA (fold of control)
1.2
NS T
1
0.8
0.6
0.4
0.2
0
Angll
natVLDL
glycoxVLDL
OxVLDL
b
45
*
40
Cyp11B2 mRNA (fold of control)
35
30
25
20
*
*
15
10
5
*
0
Angll
nat VLDL
glycox VLDL
ox VLDL
MARCO, CD68, CD36, SR-BII and CLA-1 (Calvo et al. 1998).
Following 24 h stimulation with native and glycoxidized VLDL in the presence or absence of BLT-1 (16 umol/1), whole cell lysates were obtained for quantification of Cyp11B2 mRNA transcripts by real-time PCR. The present study shows that BLT-1 caused a significant reduction of native VLDL- and glycoxidized VLDL-mediated Cyp11B2 mRNA levels (Fig. 3b).
VLDL involves cAMP-dependent PKA for modulation of Cyp11B2 mRNA level and adrenocortical hormone release
ACTH is known to mediate its effect on adrenocortical aldosterone release by increasing cAMP level via activation of the adenylate cyclase system (Quinn and Williams 1988). In order to investigate the role of cAMP in VLDL-mediated aldosterone secretion, H295R cells were incubated with native and modified VLDL in the presence or absence of cAMP analogue Rp-cAMP (1 mmol/l) for 24 h and superna- tant was collected for aldosterone measurement. Figure 4a reveals that Rp-cAMP produced significant partial impair- ment of steroid release from all forms of VLDL. In order to determine VLDL-induced and cAMP-dependent downstream
a
600
Aldosterone (pmol/l)
500
T
without BLT-1
400
T
with BLT-1
300
200
**
**
T
I
100
NS
NS
0
C
natVLDL
glycoxVLDL
OxVLDL
b
25
Cyp11B2 mRNA (fold of control)
20
T
without BLT-1
T
15
with BLT-1
10
**
**
5
NS
0
C
natVLDL
glycoxVLDL
regulators of aldosterone release, experiments were performed with or without PKA inhibitor H89 (10 umol/l). Compared with Rp-cAMP, H89 induced similar responses (Fig. 4b). H89 also caused significant impairment of aldosterone synthase transcripts levels (Fig. 4c). This suggests the partial involve- ment of cAMP-PKA-dependent pathway for VLDL-mediated adrenocortical steroidogenesis.
Native and modified VLDL recruit extracellular signal-regulated kinase (ERK1/2) as a downstream regulator for adrenocortical hormone release
Several independent studies revealed that PKA recruit mitogen-activated protein kinase (MAP kinase) p38 and ERK1/2 as downstream kinase cascades to transmit extra- cellular signals for differential gene expression (Richards 2001; Schmitt and Stork 2002). In order to investigate the role of MAP kinases in VLDL-stimulated aldosterone re- lease, H295R cells were treated with various forms of
a
800
Aldosterone (pmol/l)
T
without Rp-CAMP
600
with Rp-CAMP
400
**
*
200
T **
T
0
C
natVLDL
glycoxVLDL
OxVLDL
b
600
Aldosterone (pmol/l)
500
without H89
T
T
400
with H89
**
300
T
**
T
200
T
*
100
T
*
NS
+
0
c
FSK
natVLDL
glycoxVLDL
oxVLDL
C
25
Cyp11B2 mRNA (fold of control)
20
without H89
T
with H89
15
*
*
10
5
NS
-
*
0
c
natVLDL
glycoxVLDL
OxVLDL
VLDL (50 µg/ml) in the presence or absence of MEK (the upstream regulator of ERK1/2) inhibitor U0126 (10 µmol/l) and p38 MAP kinase inhibitor SB203580 (10 umol/l) for 24 h. U0126 caused a significant reduction of natVLDL-, glcox VLDL- and oxVLDL-mediated adrenocortical steroid secretion by 72, 66 and 32%, respectively (Fig. 5a). On the other hand, p38 MAP kinase inhibitor SB203580 did not impair the VLDL-induced aldosterone release; instead it enhanced the hormone release (Fig. 5b). This suggests that p38 MAP kinase has some inhibitory effect on VLDL- mediated aldosterone release. This kind of inhibitory effect on signaling mechanisms has also been previously docu- mented (Birkenkamp et al. 2000; Zheng and Bollag 2003).
In order to investigate the role of MAP kinase in tran- scriptional regulation of aldosterone synthase, H295R cells were treated with all indicated forms of VLDL with or without U0126 (10 mol/l) for 24 h. The whole cell lysates were used for quantification of Cyp11B2 mRNA abundance by real-time PCR. As shown in Fig. 5c, U0126 caused an impairment of Cyp11B2 mRNA levels in response to native, glycoxidized and oxidized VLDL, indicating the involve- ment of ERK1/2 at the transcriptional regulation of VLDL- mediated aldosterone secretion.
Simultaneous immunoblot analysis (Fig. 5d) reveals that treatment with U0126 caused a complete inhibition of VLDL-induced ERK1/2 phosphorylation, while total intra- cellular ERK1/2 content remained unchanged. For kinetic study of VLDL-induced ERK phosphorylation, H295R cells were treated with native and modified VLDL for intervals ranging from 1 to 60 min and cell lysates were utilized for western blotting. Subsequently, the membrane was analyzed for ERK phosphorylation. As demonstrated in Fig. 5e, all forms of VLDL could induce phosphorylation within 1 min and the maximum response was observed after 5 min of treatment.
Jak-2/ STAT3 activation is required for VLDL-induced steroid hormone release
As incomplete inhibition with the MEK inhibitor U0126 was observed, we were prompted to explore other indepen- dent, divergent or convergent pathways potentially involved in VLDL-mediated steroidogenesis. Following incubation of all forms of VLDL-stimulated cells with or without the pharmacological specific Jak-2 inhibitor AG490 (50 µmol/l) for 24 h, supernatants were collected for aldosterone mea- surement. Whole cell lysates were utilized for RNA isola- tion followed by Cyp11B2 mRNA level quantification by real-time PCR. Figure 6a, b shows that Jak-2 inhibitor AG490 significantly impeded VLDL-induced aldosterone release, as well as Cyp11B2 mRNA levels. Both native (Fig. 6c) and modified forms (not shown) of VLDL induced phosphorylation of STAT3 within 1 min. The peak increase in STAT3 phosphorylation was observed at 30 min as com- pared to that of ERK at 5 min.
Discussion
Cholesterol is utilized as a precursor for the steroid hormone biosynthesis in all steroidogenic tissues including the adrenal gland. These specialized tissues prefer to use the exogenous lipoprotein-derived cholesterol over endogenous sources (de novo synthesis and stored lipid droplets; Kraemer 2007). Moreover, various animal species prefer to use different lip- oproteins as the primary source of cholesterol. For example,
C
a
800
25
Aldosterone (pmol/l)
☐ without U0126
with U0126
Cyp11B2 mRNA (fold of control)
20
☐ without U0126
600
15
with U0126
400
10
200
NS
5
NS
0
0
c
Angli
natVLDL
glycoxVLDL
oxVLDL
C
natVLDL
glycoxVLDL
OxVLDL
b
d
without SB203580
1000
PERK
Aldosterone (pmol/l)
with SB203580
800
ERK control natVLDL glycoxVLDL oxVLDL U0126
600
400
+
200
NS
0
C
natVLDL
glycoxVLDL
OxVLDL
e
PERK
natVLDL
C
1
5
10
15
30
60
ß actin
PERK glycoxVLDL
C
1
5
10
15
30
60
ß-actin
PERK OxVLDL
C
1
5
10
15
30
60
(time in min)
ß-actin
LDL cholesterol serves as a major source for the human adrenal gland (Higashijima et al. 1987), whereas steroidogen- ic cells in rats and mice preferentially use HDL cholesterol
immunoblot of VLDL-induced ERK1/2 phosphorylation with or with- out U0126 (10 umol/l) following treatment for 5 min and after strip- ping, the membrane was incubated with ERK1/2 antibody. e Immunoblot of the time course of ERK phosphorylation after incuba- tion with different forms of VLDL (50 µg/ml) for time periods of 1 to 60 min. Immunoblot data were normalized to ß actin protein level. These are representative of three independent experiments with two different VLDL preparations, showing similar tendency. C control, VLDL very-low-density lipoprotein, nat native, glycox glycoxidized, ox oxidized, AngII angiotensin II (as positive control), ERK extracel- lular signal-regulated kinase, Cyp11B2 aldosterone synthase, DMSO dimethyl sulfoxide
(Gwynne and Hess 1980). In addition to substrate delivery, LDL and HDL can act as signaling molecules to modulate aldosterone secretion through transcriptional regulation of
a
800
Aldosterone (pmol/l)
600
without AG490
with AG490
400
200
NS
0
c
Angli
natVLDL
glycoxVLDL
oxVLDL
b
25
Cyp11B2 mRNA (fold of control)
20
without AG490
15
with AG490
10
5
NS
0
C
natVLDL
glycoxVLDL
0xVLDL
C
pSTAT3-tyr natVLDL (time in min)
C
1
5
10
15
30
60
ß actin
Cyp11B2 (Ansurudeen et al. 2010; Xing et al. 2011). However, the role of VLDL in steroidogenesis and its under- lying mechanisms has not been thoroughly investigated, de- spite growing evidence that increased circulating VLDL levels are a core component of diabetic dyslipidemia. In the present study, we investigated the effect of native, glucose- modified and oxidized VLDL on adrenocortical aldosterone synthesis and the involvement of different intracellular signal- ing mechanisms in this process.
The early rate-limiting step of adrenocortical steroido- genesis involves the transport of cholesterol from the outer to the inner mitochondrial membrane by steroidogenic acute regulatory (StAR) (Li et al. 2003). The data from the present study show that only natVLDL and glycox VLDL caused significant induction of the StAR gene expression. It is possible that ox VLDL-mediated cholesterol transport depends primarily on alternative proteins, e.g., the benzodi- azepine receptor (also known as translocator protein) (Papadopoulos et al. 2006; Hu et al. 2010). The second
late rate-limiting step represents the conversion of deoxy- corticosterone to aldosterone by the mitochondrial enzyme aldosterone synthase (Xing et al. 2011). Our present data demonstrating the significant increase in Cyp11B2 mRNA by all forms of VLDL suggest that VLDL, like other secreta- gogues, is able to regulate aldosterone production in adrenal glands by aldosterone synthase gene expression. OxVLDL induced the modest effect in the Cyp11B2 expression, explain- ing the lowest efficiency of oxVLDL to mediate aldosterone release.
Interestingly, native and glycoxidized VLDL (50 µg/ml) were almost two-fold more potent than the most well-known secretagogue, AngII, in aldosterone release. Although VLDL was equally efficient as AngII in inducing the StAR gene expression, it was less potent in increasing the Cyp11B2 mRNA levels. This implies that post-transcriptional effects (translational and post-translational modifications) could be of more importance in VLDL- than AngII-mediated adreno- cortical steroid hormone release.
SR-BI selectively transports cholesterol for steroidogen- esis, as opposed to receptor-mediated internalization of the whole lipoprotein particle by the LDL receptor. It has pre- viously been shown that attenuation of SR-BI expression leads to the development of atherosclerosis resulting in cardiovascular injury and premature death (Braun et al. 2002; Nieland et al. 2004). Using the specific SR-BI inhib- itor BLT-1, we demonstrated that only native and glycoxi- dized VLDL were dependent on SR-BI receptor for cholesterol delivery to the adrenal gland; it is probable that cholesterol supply from oxVLDL depends on other classical or scavenger receptors. In the present study, we found that BLT-1 also decreased both native and glycoxidized VLDL- mediated Cyp11B2 mRNA levels, indicating the distinct involvement of SR-BI in transcriptional regulation of steroid hormone release.
The structural heterogeneity of VLDL may contribute to a variety of biological activities that are mediated by multi- ple signaling pathways. Our experiments with the pharma- cological agent H89 reveal that VLDL action on the aldosterone synthase transcription and the hormone secre- tion depends partially on PKA. Moreover, U0126 signifi- cantly decreased whereas SB203580 significantly increased VLDL-mediated aldosterone secretion in H295R cells. These findings suggest that VLDL exerts a reciprocal influ- ence on p38 and ERK1/2 MAP kinase activation and sub- sequent aldosteone synthesis. It also indicates that there may be some inhibitory factor suppressing the p38 MAP kinase activity.
In the present study, we observed that native and modi- fied VLDL exhibited similar effects on ERK1/2 activation. The time course of ERK phosphorylation confirms previous findings of Sachinidis et al. (1999), who observed maximum effects of VLDL-mediated ERK phosphorylation in vascular
smooth muscle cells within 5 min. However, the activation persisted only for 30 min in contrast to 60 min in the present study, indicating a more sustained activation of the ERK 1/2 cascade that might be important for stress-induced adreno- cortical steroidogenesis in metabolic disorders for the main- tenance of fluid and electrolyte balance (Hoekstra et al. 2010). Of note, U0126 caused significant decrease of VLDL-mediated Cyp11B2 mRNA levels, whereas it in- duced a complete inhibition of VLDL-mediated ERK phos- phorylation. These results point to the existence of another parallel signaling pathway, which has been subsequently confirmed by results with the Jak family tyrosine kinase inhibitor AG490. Experiments with AG490 demonstrated that native and modified VLDL, in addition to ERK1/2 activation, also employed Jak-2 as a downstream target of aldosterone secretion and transcriptional regulation of aldo- sterone synthase.
In conclusion, VLDL is a lipoprotein that could play an important role in adrenocortical steroidogenesis. Glycoxidative modification did not cause significant attenuation of its ste- roidogenic potential, whereas oxidative modification signifi- cantly attenuated adrenocortical aldosterone synthesis. Native and glycoxidized (but not oxidized) VLDL preferentially uti- lized SR-BI for the selective transport of cholesterol to H295R cells, since BLT-1 significantly interfered with VLDL- mediated steroidogenesis. In addition to its function as choles- terol supplier, VLDL appeared to act as a signaling molecule through activation of PKA, ERK1/2 and Jak-2 for the tran- scriptional regulation of aldosterone synthase since the specific pharmacological inhibitors (H89, U0126 and AG490) reduced the VLDL-mediated steroid hormone release and Cyp11B2- stimulating effect.
Acknowledgement The authors would like to thank Martina Kohl, Sigrid Nitzsche and Eva Schubert for their excellent technical support and Kathy Eisenhofer for her careful reading of the manuscript. This work was supported by the Deutsche Forschungsgemeinschaft (KFO 252 to SRB).
References
Adiels M, Taskinen MR, Packard C, Caslake MJ, Soro-Paavonen A, Westerbacka J, Vehkavaara S, Häkkinen A, Olofsson SO, Yki- Järvinen H, Borén J (2006) Overproduction of large VLDL par- ticles is driven by increased liver fat content in man. Diabetologia 49:755-765
Adiels M, Olofsson SO, Taskinen MR, Borén J (2008) Overproduction of very low-density lipoproteins is the hallmark of the dyslipide- mia in the metabolic syndrome. Arterioscler Thromb Vasc Biol 28:1225-1236
Ansurudeen I, Pietzsch J, Graessler J, Ehrhart-Bornstein M, Saha S, Bornstein SR, Kopprasch S (2010) Modulation of adrenal aldo- sterone release by oxidative modification of low-density lipopro- tein. Am J Hypertens 23(10):1061-1068
Birkenkamp KU, Tuyt LML, Lummen C, Wierenga ATJ, Kruijer W, Vellenga W (2000) The p38 MAP kinase inhibitor SB203580 enhances nuclear factor-kappa B transcriptional activity by a non-specific effect upon the ERK pathway. Br J Pharmacol 131 (1):99-107
Braun A, Trigatti BL, Post MJ, Sato K, Simons M, Edelberg JM, Rosenberg RD, Schrenzel M, Krieger M (2002) Loss of SR-BI expression leads to the early onset of occlusive atherosclerotic coronary artery disease, spontaneous myocardial infarctions, se- vere cardiac dysfunction, and premature death in apolipoprotein E-deficient mice. Circ Res 90:270-276
Brunzell JD, Davidson M, Furberg CD, Goldberg RB, Howard BV, Stein JH, Witztum JL (2008) Lipoprotein management in patients with cardiometabolic risk. Diabetes Care 31(4):811-822
Calvo D, Gómez-Coronado D, Suárez Y, Lasunción MA, Vega MA (1998) Human CD36 is a high affinity receptor for the native lipoproteins HDL, LDL, and VLDL. J Lipid Res 39:777-788 Capponi AM (2002) Regulation of cholesterol supply for mineralocor- ticoid biosynthesis. Trends Endocrinol Metab 13(3):118-121 Cushley RJ, Okon M (2002) NMR studies of lipoprotein structure. Annu Rev Biophys Biomol Struct 31:177-206
Dinkel RE, Barrett PHR, Demant T, Parhofer KG (2006) In-vivo metabolism of VLDL-apolipoprotein-B, -CIII and -E in normoli- pidemic subjects. Nutr Metab Cardiovasc Dis 16:215-221
Frias MA, James RW, Wicht CG, Lang U (2009) Native and recon- stituted HDL activate Stat3 in ventricular cardiomyocytes via ERK1/2: role of sphingosine-1-phosphate. Cardiovasc Res 82:313-323
Graessler J, Pietzsch J, Westendorf T, Julius U, Bornstein SR, Kopprasch S (2007) Glycoxidised LDL isolated from subjects with impaired glucose tolerance increases CD36 and peroxisome proliferator-activator receptor y gene expression in macrophages. Diabetologia 50:1080-1088
Gwynne JT, Hess B (1980) The role of high density lipoproteins in rat adrenal cholesterol metabolism and steroidogenesis. J Biol Chem 255(22):10875-10883
Higashijima M, Nawata H, Kato K, Ibayashi H (1987) Studies on lipoprotein and adrenal steroidogenesis: I. Roles of low density lipoprotein- and high density lipoprotein-cholesterol in steroid production in cultured human adrenocortical cells. Endocrinol Jpn 34(5):635-645
Hoekstra M, Korporaal SJA, Li Z, Zhao Y, Eck MV, Berkel TJCV (2010) Plasma lipoproteins are required for both basal and stress- induced adrenal glucocorticoid synthesis and protection against endotoxemia in mice. Am J Physiol Endocrinol Metab 299(6): E1038-E1043
Hu J, Zhang Z, Shen WJ, Azhar S (2010) Cellular cholesterol delivery, intracellular processing and utilization for biosynthesis of steroid hormones. Nutr Metab 7(47):1-25
Kannel WB, Vasan RS (2009) Triglycerides as vascular risk factors: new epidemiologic insights. Curr Opin Cardiol 24(4):345-350 Kraemer FB (2007) Adrenal cholesterol utilization. Mol Cell Endocrinol 265-266:42-45
Krug AW, Ehrhart-Bornstein M (2008) Aldosterone and metabolic syndrome: is increased aldosterone in metabolic syndrome patients an additional risk factor? Hypertension 51:1252-1258
Laemmly UK (1970) Cleavage of structural proteins during the assem- bly of the head of bacteriophage T4. Nature 227:680-685
Levine RL, Garland D, Oliver CN, Amici A, Climent I, Lenz AG, Ahn BW, Shaltiel S, Stadtman ER (1990) Determination of carbonyl content in oxidatively modified proteins. Methods Enzymol 186:464-478
Li J, Feltzer RE, Dawson KL, Hudson EA, Clark BJ (2003) Janus Kinase 2 and calcium are required for angiotensinII-dependent activation of steroidogenic acute regulatory protein transcription in H295R hu- man adrenocortical cells. J Biol Chem 278(52):52355-52362
Maeba R, Shimasaki H, Ueta N (1994) Conformational changes in oxidized LDL recognized by mouse peritoneal macrophages. Biochim Biophys Acta 1215:79-86
Nieland TJF, Chroni A, Fitzgerald ML, Maliga Z, Zannis VI, Kirchhausen T, Krieger M (2004) Cross-inhibition of SR-BI- and ABCA1-mediated cholesterol transport by the small mole- cules BLT-4 and glyburide. J Lipid Res 45:1256-1265
Nofer JR, Fobker M, Höbbel G, Voss R, Wolinska I, Tepel M, Zidek W, Junker R, Seedorf U, von Eckardstein A, Assmann G, Walter M (2000) Activation of phosphatidylinositol-specific phospholipase C by HDL-associated lysosphingolipid: involvement in mitogenesis but not in cholesterol efflux. Biochemistry 39:15199-15207
Papadopoulos V, Baraldi M, Guilarte TR, Knudsen TB, Lacapere JJ, Lindemann P, Norenberg MD, Nutt D, Weizman A, Zhang MR, Gavish M (2006) Translocator protein (18kDa): new nomenclature for the peripheral-type benzodiazepine receptor based on its struc- ture and molecular’function. Trends Pharmacol Sci 27:402-409
Pietzsch J, Subat S, Nitzsche S, Leonhardt W, Schentke KU, Hanefeld M (1995) Very fast ultracentrifugation of serum lipoproteins: influence on lipoprotein separation and composition. Biochim Biophys Acta 1254:77-88
Quinn SJ, Williams GH (1988) Regulation of aldosterone secretion. Annu Rev Physiol 50:409-426
Richards JS (2001) New signaling pathways for hormones and cyclic adenosine 3’,5’-monophosphate action in endocrine cells. Mol Endocrinol 15(2):209-218
Sachinidis A, Kettenhofen R, Seewald S, Gouni-Berthold I, Schmitz U, Seul C, Ko Y, Vetter H (1999) Evidence that lipoproteins are carriers of bioactive factors. Arterioscler Thromb Vasc Biol 19 (10):2412-2421
Schmitt JM, Stork PJ (2002) Ga and GBy require distinct Src- dependent pathways to activate Rap1 and Ras. J Biol Chem 277:43024-43032
Stillemark-Billton P, Beck C, Borén J, Olofsson SO (2005) Relation of the size and intracellular sorting of apoB to the formation of VLDL 1 and VLDL 2. J Lipid Res 46:104-114
Xing Y, Cohen A, Rothblat G, Sankaranarayanan S, Weibel G, Royer L, Francone OL, Rainey WE (2011) Aldosterone production in human adrenocortical cells is stimulated by high-density lipoprotein 2 (HDL2) through increased expres- sion of aldosterone aynthase (CYP11B2). Endocrinology 152 (3):751-763
Yagi K (1976) A simple fluorometric assay for lipoperoxide in blood plasma. Biochem Med 15:212-216
Zheng X, Bollag WB (2003) AngII induces transient phospholipase D activity in the H295R glomerulosa cell model. Mol Cell Endocrinol 206(1-2):113-122