ELSEVIER
Journal of Ethnopharmacology
journal homepage: www.elsevier.com/locate/jep
A
Journal of ETHNO- PHARMACOLOGY
Modulation of glucocorticoid, mineralocorticoid and androgen production in H295 cells by Trimesemine™M, a mesembrine-rich Sceletium extract
A.C. Swartª, C. Smith b,*
a Dept Biochemistry, Stellenbosch University, South Africa
b Dept Physiological Sciences, Stellenbosch University, South Africa
CrossMark
ARTICLE INFO
Article history: Received 29 July 2015
Received in revised form 10 November 2015 Accepted 17 November 2015 Available online 19 November 2015
Keywords:
Endocrine disrupting compound Cytochrome p450
Hypertension
Adrenal steroidogenesis Cortisol Sceletium tortuosum Alkaloid Anxiolytic Stress
ABSTRACT
Ethnopharmacological relevance: Stress-related illnesses rate among the most prevalent non-fatal dis- eases globally. With the global trend for consumer bias towards natural medicine, the Sceletium plant has become more prominent in the field of natural products. Although potentially useful effects of Sceletium tortuosum on the central nervous system have been reported, limited data is available on effects of the plant in the peripheral compartment.
Aim of the study: The current study aimed to elucidate the effect(s) of a Sceletium extract (TRI) rich in mesembrine (1% of plant extract w/w), on adrenal steroid biosynthesis.
Materials and methods: Steroidogenesis was assessed basally and in response to stimuli (forskolin, an- giotensin II, KCl), in human adrenocortical carcinoma cells (H295R). Steroid hormone levels were as- sessed using UPLC-MS/MS. UPLC-MS analyses of TRI identified major alkaloids A7-mesembrenone, mesembrenone and mesembrine.
Results: Highest dose TRI treatment (1 mg/ml, 34.5 µM mesembrine) increased pregnenolone and de- creased 16-hydroxyprogesterone levels (both P <0.00001) in forskolin-stimulated conditions only, suggesting CYP17 enzyme inhibition. This led to significant inhibition of forskolin-associated increases in cortisol levels at the highest dose (P < 0.001) and basal cortisol levels across all doses (P < 0.0001). In- dependently of forskolin, TRI inhibited androstenedione and testosterone production across all doses (both P <0.00001), suggesting inhibition of 3ßHSD and 17ßHSD respectively. TRI decreased both the angiotensin II- (P < 0.05) and forskolin-induced (P < 0.0001) increases in aldosterone production.
Conclusions: Our data suggest potentially beneficial effects of TRI in the context of stress and hy- pertension. These should be further investigated in a whole organism model, while the effects on the androgenic pathway should also be further elucidated.
@ 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Population longevity and a decline in quality of life are closely associated with high incidence of chronic non-fatal disease, which is placing an ever-increasing strain on public health sectors. In a recent report by the 2013 Global Burden of Disease Study Group, statistics indicated that global mortality rates were decreasing at a higher rate than the average number of years individuals were living with a disability. Significantly, major depressive disorder and anxiety disorders rated among the top ten causes of “years living with disability” in the 188 countries assessed in this study by Vos and GBD2013Collaborators (2015) It is in the clinical management
of these conditions that the use of mainstream drugs has become unaffordable for many and this, together with adverse side-effects, has led to the increased popularity of complementary and alter- native medicines. This general shift in consumer bias towards natural medicine has sparked considerable interest in plant-de- rived substances with the claimed potential to relieve stress, an- xiety and/or depression. However, the lack of scientific evidence on efficacy and safety regarding a wide spectrum of com- plementary therapies, led to the establishment of a National Centre for Complementary and Alternative Medicine at the US National Institute of Health (Goroll, 2014).
Although natural medicine may hold many potential benefits, natural products should not be exempt from comprehensive stu- dies of their effects on physiological systems, to provide scientific evidence for not only anecdotal claims of benefit but also potential undesirable effects (Offit, 2012). Exhaustive investigations into
* Corresponding author.
E-mail addresses: acswart@sun.ac.za (A.C. Swart), csmith@sun.ac.za (C. Smith).
Cholesterol
CYP11A1
Pregnenolone (PREG)
CYP17A1
17OH-Pregnenolone (17OH-PREG)
CYP17A1
Dehydroepiandrosterone (DHEA)
17ßHSD5
Androstenediol
16-OH Progesterone (16OH-PROG)
3ßHSD2
3ßHSD2
3ßHSD2
3ßHSD2
CYP17A1
Progesterone (PROG)
CYP17A1
17OH-Progesterone (17OH-PROG)
17ßHSD5
Androstenedione (A4)
1
Testosterone (T)
CYP21A2
CYP21A2
Deoxycorticosterone (DOC)
Deoxycortisol
CYP11B2
CYP11B1
CYP11B1
Corticosterone (CORT)
Cortisol
11-OH Androstenedione (11-OHA4)
CYP11B2
18-OH Corticosterone (18-OHCORT)
CYP11B2
Aldosterone (ALDO)
natural products at both molecular and cellular level under con- ditions reflecting both normal and pathological physiology would indicate potential beneficial effects as well as adverse effects not envisaged. Recent evidence of this are the deleterious cardiovas- cular and muscle breakdown effects now associated with the once popular appetite suppressing extracts from Hoodia gordonii (Blom et al., 2011; Smith and Krygsman, 2014a, 2014b).
One of the candidate plant genera which have been scientifi- cally investigated for the natural treatment for stress-related symptoms - and which have been known anecdotally for more than a century to alleviate stress - is Sceletium, and in particular Sceletium tortuosum (L.) N.E.Br, commonly known as Kanna or Kougoed, a member of the Mesembryanthemaceae subfamily in the family Aizoaceae. The anecdotal clinical effects associated with consumption of the plant material - e.g. analgesia, calmness, se- dation (Van Wyk and Wink, 2009) - strongly suggest both nervous system and endocrine system involvement. However, although some published data is available on the potential central me- chanisms of Sceletium, limited scientific data is available on the manner in which Sceletium may affect the endocrine system.
Recently, results from a study in humans suggested that a single dose of S. tortuosum (equivalent to 50 mg of dry plant ma- terial) decreases connectivity between the amygdala and hy- pothalamus, as assessed using MRI during a facial recognition task (Terburg et al., 2013). A limitation of the human study is that no physiological stress status assessment or psychiatric evaluations for anxiety and depression were performed in the student popu- lation used, although a similar population has recently been shown to have relatively high stress and anxiety levels (Smith et al., 2014). It is therefore difficult to assess whether these results pertain to a normal, stressed, anxious and/or depressed popula- tion. Also, in this study, the reported central effects were not as- sociated with clinical perception of changes in mood state and blood chemistry data was not reported, so that the clinical sig- nificance of the altered connectivity in the context of stress
remains debatable.
Prior to the aforementioned study, we reported on the effects of a similar dose of S. tortuosum extract as the one used in the human study (as well as a supraphysiological dose), in a rat model of restraint stress (Smith, 2011), focusing on both behavioural ef- fects and changes in blood hormone and cytokine levels. Our re- sults showed that a 17-day administration of S. tortuosum extract reduced basal corticosterone levels and decreased the stress-in- duced serum corticosterone response by ~ 20%, to result in a mild calming effect, as evidenced by a ~ 30% decrease in self-soothing behaviour.
From these studies, stress-related parameters appear to indeed be beneficially affected by S. tortuosum, as suggested by anecdotal evidence. The fact that in neither the human, nor the rat study, significant changes in mood or behavioural state was reported, may suggest that a peripheral mechanism of action may come into play to relieve stress, in addition to any central effects which may down-regulate the stress response at the level of perception. It is possible that Sceletium spp. may influence endocrine signalling pathways via the modulation of steroid receptor expression levels, or via interaction of these receptors with plant alkaloids which have been shown to act at receptor level (Djiogue et al., 2014; Dvorak et al., 2006; Sunden et al., 2015). However, the effects on corticosterone levels that were previously reported suggest that plant constituents may more likely be impacting steroid hormone levels directly by affecting adrenal steroid hormone biosynthesis and metabolism.
Glucocorticoids, cortisol and corticosterone, are both products of adrenal steroidogenesis, with cortisol considered to be the main stress hormone in humans while rats produce only corticosterone. Cholesterol is the precursor of all adrenal steroid hormones - glucocorticoids, mineralocorticoids and androgens. The biosynth- esis of these hormones are catalysed by the cytochrome P450 (P450) enzymes and the hydroxysteroid dehydrogenases (Fig. 1). Investigations into adrenal hormone biosynthesis undertaken in
H295R cells have been extensively reported (Miller and Auchus, 2011; Nielsen et al., 2012; Winther et al., 2013) and the cell model is commonly used to test biological compounds and endocrine disruptors (Komarnytsky et al., 2013; Matkovic et al., 2009) as it allows for the in vitro analysis of effects on the full complement of adrenal steroid hormones and steroidogenic enzymes. The current study therefore employed the H295R human adrenocortical car- cinoma cell model to investigate the effect of a proprietary Scele- tium extract on steroidogenesis. In addition, a particular strength of this study is that comprehensive LC-MS/MS analyses of steroid profiles presenting the full effect of S. tortuosum on adrenal ster- oidogenesis was performed, so that the current study was not limited to end-point analyses only.
2. Materials and methods
2.1. Materials
A lyophilised extract prepared from a proprietary hybrid (DV17) of S. tortuosum (L.) N.E. Br. and S. expansum (L.) L. Bolus (family Aizoaceae), Trimesemine™M (TRI), using a proprietary method (patent pending, refer to Supplementary online material 1 for details) was obtained from Botanical Resource Holdings Pty (Ltd) affiliate Verve Dynamics (Somerset West, South Africa)(Lot #BTRMA:001/024, manufacturing reference # DV SCIET:E 028/024 (24123). Please refer to Supplementary material 1 for the certifi- cate of analysis and quality control data). The extract is prepared by patented method (TRI is a more refined extract than was pre- viously used by our group (Smith, 2011), thus limiting confounding effects of other secondary plant metabolites. In our previous study the extract was prepared in a manner similar to a method recently reported to yield a relatively low mesembrine concentration (total alkaloid content between 0.35% and 0.45% w/w of extract, with mesembrine contributing less than 20% of total alkaloids) (Mur- bach et al., 2014). Of specific benefit to our present study, since mesembrine is the active component most commonly associated with anxiolytic or psychoactive effects (Harvey et al., 2011; Staf- ford et al., 2008), was the use of the mesembrine-rich preparation, TRI (total alkaloid content 3% w/w of extract, with mesembrine contributing more than 80% of total alkaloids). Ultra-high perfor- mance liquid chromatography-mass spectrometry (UPLC-MS) analyses of TRI was performed by the Stellenbosch University Central Analytical Facility in which 47-mesembrenone, mesem- brenone and mesembrine were identified as major constituents (Fig. 2). Mesembrine content of the TRI extract was determined by the same laboratory to be 1% of the extracted plant material (w/w). In the current study the TRI extract was subsequently assayed over the dose range of 0.0001-1 mg extract/ml, which contained 3.45 nM to 34.5 µM mesembrine.
DMEM/F12 and gentamicin were purchased from Invitrogen/ Gibco (Grand Island, New York, USA. Penicillin, streptomycin and trypsin-EDTA were obtained from Gibco BRL (Gaithersburg, MD, USA). Cosmic calf serum was supplied by HyClone(R), Thermo Scientific (South Logan, Utah, USA). MTT assay kits, angiotensin II (AngII) and forskolin were purchased from Sigma-Aldrich (St. Louis, MA, USA). Sigma-Aldrich and Steraloids (Newport, USA) supplied steroid standards. Deuterated cortisol (9,11,12,12-D4- cortisol) was purchased from Cambridge isotopes (Andover, MA, USA). The UPLC Kinetex PFP column was obtained from Phenom- enex (Torrance, CA, USA). All other chemicals were of the highest grade and supplied by trustworthy scientific supply houses.
2.2. Cell viability
Confluent cells were plated into 96-well culture plates (100 ul,
50000
mesembrine
CH3
CH3
CH3
CH3
CH3
CH3
40000
47-mesembrenone
30000
H3C
H3C
H
H3C
%
mesembrenone
20000
10000
0
L
0
1
2
3
4
5
6
8
10
Time (min)
4 000 cells) and incubated with high and low concentrations of TRI (1 mg/ml and 1 µg/ml) for a period of 48 hours at 5% CO2 and 37 ℃. At the end of the incubation period, cell viability was as- sessed using an MTT toxicology assay kit according to the manu- facturer’s instructions. Media containing the range of extract di- lutions were included. The extract had no deleterious effect on the H295R cell viability at any of the concentrations assayed - in fact, cells treated with TRI exhibited significantly better maintenance of cell viability (89 + 3% and 89 + 2% for cells treated with 0.0001 and 1 mg/ml Tri respectively) when compared to both control (79±3%, P<0.001) and forskolin-treated cells (76+3%, P < 0.001).
2.3. Steroid metabolism in H295R cells
H295R cells were cultured at 5% CO2 and 37 ℃ in growth medium (DMEM/F12 media, supplemented with L-glutamine, 15 mM HEPES, pyridoxine, 1.125 g NaHCO3/l, 1% penicillin strep- tomycin, 0.01% gentamicin) and 10% cosmic calf serum. Confluent cells were plated into 12 well plates (1 ml/well, 4 x 105 cells/ml) and incubated for 48 h. The medium was subsequently replaced with experimental medium (growth medium containing 0.1% cosmic calf serum) and cells were incubated for 12 h after which the appropriate treatments forskolin (10 M), AngII (10 nM), and KCl (16 mM) were added in experimental medium. Steroid meta- bolism was assayed in the presence of TRI by the addition of ex- tract (50 ul per well) to a final concentration of 0, 0.0001, 0.001, 0.01, 0.1 or 1 mg extract/ml, under both basal and stimulated conditions. After 48 h, media aliquots (500 ul) were removed and the steroids extracted by liquid/liquid extraction. Steroids were extracted using a 10:1 volume of dichloromethane to culture medium after the addition of D4-cortisol (15 ng), as an internal standard prior to extraction. The dichloromethane phase was dried under N2, and the dried residue resuspended in 150 ul methanol and analysed by UPLC-MS/MS analyses. Steroid quantification was performed as described in detail (Schloms et al., 2012).
A
1800
PREG
4500
Cortisol
OBasal Forskolin
1600
4000
1400
3500
Steroid (nM)
1200
#
Steroid (nM)
3000
1000
*
2500
800
2000
**
600
1500
*
400
1000
#
200
500
0
0
0.0001 0.001
0.01
0.1
0
1
0
0.0001 0.001
0.01
0.1
1
Sceletium (TRI) dose (mg/ml)
Sceletium (TRI) dose (mg/ml)
CYP17A1
CYP11B1
17OHPREG
3ßHSD2
CYP21A2
180
17OHPROG
4000
Deoxycortisol
160
**
3500
140
Steroid (nM)
3000
120
Steroid (nM)
2500
100
80
2000
*
60
1500
40
#
1000
20
500
0
0
0
0.0001
0.001
0.01
0.1
1
0
0.0001 0.001
0.01
0.1
1
Sceletium (TRI) dose (mg/ml)
Sceletium (TRI) dose (mg/ml)
CYP17A1
3ßHSD2
CYP17A1
60
PROG
120
16OHPROG
50
100
Steroid (nM)
40
Steroid (nM)
80
30
+
**
60
20
40
*
**
10
#
20
#
0
0
0
0.0001
0.001
0.01
0.1
1
0
0.0001
0.001
0.01
0.1
1
Sceletium (TRI) dose (mg/ml)
Sceletium (TRI) dose (mg/ml)
B
PREG
11OHA4
OBasal Forskolin
1800
600
1600
1400
500
Steroid (nM)
1200
Steroid (nM)
400
1000
800
#
300
600
200
**
400
200
100
±
T
$
¢
#
0
0
0
0.0001 0.001
0.01
0.1
1
0
0.0001
0.001
0.01
0.1
1
Sceletium (TRI) dose (mg/ml)
Sceletium (TRI) dose (mg/ml)
CYP17A1
CYP11B1
17OHPREG
CYP17A1
900
DHEA
3ßHSD2
1800
A4
1600
800
700
1400
Steroid (nM)
600
Steroid (nM)
1200
500
1000
400
800
300
600
200
#
400
+
T
T
T
100
200
0
0
0
0.0001
0.001
0.01
0.1
1
0
0.0001
0.001
0.01
0.1
1
Sceletium (TRI) dose (mg/ml)
Sceletium (TRI) dose (mg/ml)
17ßHSD5
70
Testosterone
60
Steroid (nM)
50
40
30
20
10
0
I
I
0
0.0001
0.001
0.01
0.1
1
Sceletium (TRI) dose (mg/ml)
C
PREG
40
OBasal Forskolin
1800
ALDO
35
1600
1400
30
Steroid (nM)
1200
Steroid (nM)
25
1000
*
20
800
#
15
**
600
10
400
200
5
#
0
0
0
0.0001
0.001
0.01
0.1
1
0
0.0001
0.001
0.01
0.1
1
Sceletium (TRI) dose (mg/ml)
Sceletium (TRI) dose (mg/ml)
3ßHSD2
CYP11B2
60
PROG
120
18OHCORT
**
50
100
Steroid (nM)
40
Steroid (nM)
80
30
+
60
20
40
*
10
#
20
#
0
0
0
0.0001
0.001
0.01
0.1
1
0
0.0001
0.001
0.01
0.1
1
Sceletium (TRI) dose (mg/ml)
Sceletium (TRI) dose (mg/ml)
CYP21A2
CYP11B2
CYP11B2
1000
DOC
5000
CORT
900
4500
800
4000
Steroid (nM)
700
3500
600
Steroid (nM)
3000
500
2500
400
2000
#
300
1500
200
1000
**
100
500
*
0
0
0
0.0001
0.001
0.01
0.1
1
0
0.0001
0.001
0.01
0.1
1
Sceletium (TRI) dose (mg/ml)
Sceletium (TRI) dose (mg/ml)
2.4. Statistical analyses
Data was analysed using Statistica v.12 (StatSoft Software). Two-way ANOVA was performed to assess main as well as
interaction effects of treatments (forskolin, TRI, AngII or KCl). This was followed by LSD post hoc testing. The Levene test for homo- geneity of variances was performed for all analyses. Where Leve- ne’s test rejected the hypothesis of homogeneity, a Games-Howell
post hoc test was performed to assess specific effect differences. Results are presented as means and standard deviations.
3. Results
The influence of TRI on adrenal steroid hormone production was determined in H295R cells under basal conditions as well as in forskolin-stimulated cells. We previously reported that forskolin increased steroid production in the three respective pathways, thus increasing steroid output (Schloms et al., 2012). Current re- sults corroborate our previous findings, with forskolin increasing basal steroid production 2.3-fold. The effect of TRI on the up- stream reactions of the steroidogenic pathways was evident in the dose-dependent effects on the levels of PREG across all doses tested (ANOVA main effect of TRI dose, P < 0.00001; Fig. 3a) al- though post hoc testing returned significantly increased levels only for the 1 mg/ml TRI only under basal conditions. These data showing an increase in PREG in the presence of increasing TRI concentrations is suggestive of CYP17 inhibition, as is also de- monstrated by the concomitant decrease in 16OHPROG (ANOVA main effect of Tri dose, P < 0.00001; Fig. 3a), a CYP17 product that has been reported not to be metabolised further in the adrenal. Assessing the effect of TRI on 17OHPROG production by CYP17A1 is hampered since the steroid is also metabolised by CYP21A2 - under basal conditions these levels were significantly decreased when compared to control for all doses of TRI (P < 0.005). The effect of TRI on 17OHPROG levels under stimulated conditions were however not as significant as that detected under basal conditions.
Upstream of 17OHPROG in the glucocorticoid pathway (Fig. 3a) TRI increased basal PROG in a dose-dependent manner (ANOVA main effect, P < 0.0001), with lower doses seeming to exert a re- latively mild up-regulatory effect (also refer to Supplementary online material 1, basal), while in contrast, the 1 mg/ml dose de- creased PROG significantly under both basal and forskolin-stimu- lated conditions (P < 0.0001 and P < 0.001 respectively), suggest- ing major inhibition of the 3ßHSD-dependent conversion of PREG to PROG only at this very high dose, while the lower doses have opposing modulatory activity. Downstream analyses showed that both basal 17OHPROG and cortisol levels were significantly4 in- hibited (P <0.005) by all TRI doses when compared to control. Similar decreases were seen in the levels of the cortisol precursor, deoxycortisol (P=0.08 for 1 mg/ml), although these changes did not reach statistical significance, due to relatively higher variability in data for this parameter. Decreased deoxycortisol production under forskolin-stimulated conditions was detected, with sig- nificant decreases from the control for all TRI concentrations (all at least P < 0.001). Cortisol production on the other hand, was only inhibited significantly (P < 0.0001) when compared to the control in the presence of the highest TRI concentration.
Further downstream in the androgen pathway (Fig. 3b), DHEA levels were not markedly influenced in the presence of TRI, while A4 and T were significantly reduced under both basal and stimu- lated conditions over the entire concentration range (ANOVA main dose-effect of TRI, both P < 0.00001), suggesting inhibition of both 3ßHSD and 17ßHSD activity respectively. The production of 11OHA4 was reduced across all doses under basal conditions, while stimulated steroid production was reduced significantly at only 1 mg/ml, indicating that CYP11B1 activity was only inhibited at higher concentrations of the extract under stimulated condi- tions (as was also observed in the case of forskolin-stimulated cortisol production).
In the mineralocorticoid pathway (Fig. 3c), treatment with TRI decreased forskolin-stimulated production of ALDO only at the highest concentration (1 mg/ml; P < 0.0001 when compared to all
other doses and control), while no effect was evident under basal conditions. Upstream, similar inhibitory effects were observed for the ALDO precursors DOC and CORT, the production of which was also inhibited significantly at 1 mg/ml TRI only, under both basal and forskolin-stimulated conditions (all P <0.001). In contrast, production of 18OHCORT - although also not affected under basal conditions - was significantly inhibited after treatment with all concentrations of TRI (P < 0.01 at least). It is important to note that in the conversion of CORT to ALDO, the intermediate 18OHCORT remains bound to the enzyme and that the steroids produced also reflect conversion by CYP11B1 of DOC to CORT.
ALDO production in the adrenal is significantly lower than that of the other adrenal hormones, with basal production being 300- fold lower than that of cortisol (600-fold lower with ACTH sti- mulation) (Nakamura et al., 2011). Therefore, in order to further investigate the effects of TRI on mineralocorticoid production, cells in our model were subsequently also stimulated with AngII and potassium chloride (KCI) - which both stimulate the production of mineralocorticoids (Bird et al., 1995) - and steroid production assayed in the presence and absence of TRI at 0.01 mg/ml con- taining 0.345 µM mesembrine.
When turning to the results from these experiments using AngII and KCl stimulation, analysis of steroid metabolites in the mineralocorticoid pathway showed that both AngII and KCl sti- mulated the production of ALDO (3-fold and 10-fold respectively, with P=0.06 and P < 0.01 respectively, Fig. 4). This increase may be explained by increased CYP11B2 activity, i.e. the conversion of DOC (which was significantly decreased in response to both AngII and KCl) to increase the shunt towards ALDO, as reflected in the increased levels of CORT and 18OHCORT from basal (Fig.4). As for ALDO, the effect of AngII stimulation on 180HCORT ( > 10-fold increase on average) did not achieve statistical significance, while KCL stimulation achieved significance at level of all parameters. The effects of TRI were evident only in the significantly decreased levels of ALDO (P <0.01 when compared to control) under AngII- stimulated conditions. Although DOC levels were not decreased significantly in the presence of TRI, the lowered levels do suggest inhibition of CYP21. The analyses also showed that PREG levels were significantly lower upon stimulation with both AngII and KCl when compared to basal levels, which may be attributed to the increased production of steroids in the glucocorticoid and andro- gen pathways, in which deoxycortisol and cortisol increased sig- nificantly as well as that of 11OHA4 (Supplementary online ma- terial 2).
AngII decreased basal PREG (102 +37 vs. 201 ± 62 nmol/l; P=0.001, Fig. 4) and 16OHPROG levels (55 + 12 vs. 83 + 13 nmol/l, NS, Supplementary online material 2) but in the presence of TRI, PREG levels recovered, increasing again by ~ 60% (167 + 25 nmol/ 1; P=0.01, Fig. 4) to levels comparable to basal (untreated) level, while 16OHPROG levels decreased by a further ~30% (41 ±2 nmol/l; NS, Supplementary online material 2), once again indicating inhibition of CYP17A1 by TRI, similar to changes de- tected in forskolin-stimulated cells in the presence of TRI at the same concentration. The decreased production of 17OHPROG un- der basal (P <0.0001) and stimulated (P <0.0001 and P <0.05) conditions in the presence of TRI also reflects our findings above.
In the glucocorticoid pathway (Supplementary online material 2), both basal and stimulated deoxycortisol (P <0.01 and P < 0.0001 respectively) and cortisol (P <0.05 and P <0.00001 respectively) production was reduced significantly, again similar to changes reported for the forskolin-stimulation experiment. In the androgen pathway (Supplementary online material 2), consistent decreases in both basal and stimulated DHEA (~40 and ~25% respectively, NS), A4 ( ~65 and ~30%, NS), 110HA4 ( ~55 and ~ 40%, the latter P < 0.01) and T ( ~65 and ~ 60%, both P=0.01) production was evident (Supplementary online material 2), which
PREG
ALDO
300
**
**
30
250
T
T
25
*
Steroid (nM)
200
**
Steroid (nM)
20
T
T
150
*
*
T
15
100
T
10
T
50
5
**
0
0
Basal
Angll
KCI
Basal
Angli
KCI
Untreated
Treated
Untreated
Treated
3ßHSD2
CYP11B2
PROG
18OHCORT
60
*
35
T
T
30
**
**
50
T
Steroid (nM)
25
T
T
T
Steroid (nM)
40
20
T
-
30
15
10
20
5
10
T
0
-
Basal
Angll
KCI
0
Basal
Angli
KCI
Untreated
Treated
Untreated
Treated
CYP21A2
CYP11B2
CYP11B2
DOC
CORT
700
500
600
*
T
400
*
T
T
*
T
Steroid (nM)
500
T
*
Steroid (nM)
T
400
T
300
300
200
T
T
200
100
100
0
0
Basal
Angli
KCI
Basal
Angll
KCI
Untreated
Treated
Untreated
Treated
is comparable to our findings for the 1 mg/ml dose in the for- skolin-stimulation model.
The most significant effects of KCl on basal steroid production were evident in the mineralocorticoid pathway in which the production of ALDO increased significantly (10-fold) together with its precursor metabolites, CORT and 18OHCORT (Fig. 4). Although exposure of the cells to KCl did not result in an increase in the overall production of steroids, in the glucocorticoid and androgen pathways (Supplementary online material 2) KCl resulted in a 2-fold increase in cortisol production and a 4.6-fold increase in 11OHA4 production. In the presence of TRI, PREG levels increased 2-fold and together with the decreased production of 11OHA4, these data indicated the inhibitory effect on CYP17A1 by TRI under these conditions as well. The inhibitory effects of TRI in the mi- neralocorticoid and glucocorticoid pathways were most evident in the levels of KCl-stimulated end products, with a 1.4- to 1.6-fold reduction in the levels of cortisol (P<0.00001), 110HA4 (P <0.001) and T (P <0.05).
4. Discussion
The H295R adrenal cell model is a human adrenocortical car- cinoma cell line which is widely used in investigations into the effects of compounds modulating adrenal steroid hormone pro- duction in response to specific stimuli. In addition to the cells expressing all the steroidogenic enzymes and being capable of producing all the adrenal steroid hormones, they can also be sti- mulated to mimic specific cellular responses. In this study we stimulated the cells with forskolin, which mediates its effect via CAMP and protein kinase A, in order to mimic the stress response since these cells are less responsive to ACTH (Bird et al., 1995). In addition, we also stimulated the cells with Ang II and KCl, which both specifically stimulate ALDO production-Ang II via protein kinase C and KCI via Ca2+ signalling pathways only (Bird et al., 1995). Our subsequent determination of not only end-products but also that of steroid hormone profiles by means of UPLC-MS/MS, allowed the comprehensive analyses of the influence of TRI on steroid hormone levels.
Using this model, we have been able to demonstrate several potentially beneficial effects of Sceletium extract. Firstly, in further support of the previously suggested anti-stress effect (Smith, 2011), we have shown direct peripheral inhibition of both corti- costerone and cortisol biosynthesis in the presence of TRI, both basally and in a simulated stressed condition, providing evidence that the extract mediates its effect at least in part, independently of central control mechanisms. Secondly, the extract was not as- sociated with decreased production of DHEA - a known antagonist of glucocorticoids. These two results provide scientific evidence for the largely anecdotal claims of anti-stress and anxiolytic effects of Sceletium. Thirdly, another result that increases the desirability of this plant medicine in the treatment of stress-related ailments is its inhibitory effect on aldosterone biosynthesis, and specifically under stimulated conditions only. This suggests that the plant may have application in hypertensive patients, without the associated risk of hypotension that normally accompanies drugs usually prescribed for this purpose.
In terms of specific mechanism(s) of action of the particular Sceletium extract assessed, analyses of effects of the extract under all conditions suggest that it inhibited CYP17A1, as was reflected in the levels of PREG and 16OHPROG. This enzyme catalyses hor- mone biosynthesis at the branch point of adrenal steroidogenesis, and together with 36HSD channels steroid metabolites into the three respective pathways. The perturbation of these two enzymes will thus directly influence the steroid shunt in these pathways, with changes in the steroid levels affecting the androgen,
mineralo- and glucocorticoid end products. However, since the increase in PREG levels may also be attributed to increased avail- ability of cholesterol and its conversion to PREG, which was not a focus in this study, the influence of TRI on CYP17A1 requires fur- ther investigation via this avenue. Furthermore, while PROG levels increased in the presence of TRI concentrations up to 0.1 mg/ml, indicating that at these concentrations 3ßHSD was not being in- hibited, 17OHPREG levels remained below the limit of quantifica- tion, while DHEA levels did not change significantly, all of which would seem to imply that the lyase activity of CYP17A1 was not influenced to the same degree as the hydroxylase activity of CYP17A1. Interestingly, while DHEA and DHEA-S levels did not change significantly in the presence of lower concentrations of TRI under any of the stimulated conditions assayed, both basal and forskolin-stimulated DHEA-S production increased 12-fold in the presence of 1 mg/ml TRI (ANOVA treatment effect, P < 0.0001), suggesting possible stimulation of adrenal sulfotransferase (SULT) activity. SULT2A1 catalyses the sulfation at the 30-hydroxyl group of DHEA, which decreases the bioactivity of this hormone.
Although these data indicate that the TRI-associated inhibition of androgen biosynthesis may be attributed to effects on CYP17A1 as well as 30HSD, evident in the significant reduction of A4 at all concentrations assayed, this aspect requires further studies. In- vestigations into the effect on enzymes in isolated cell systems expressing specific enzymes only will allow assays to be conducted away from competing enzymes. In the glucocorticoid pathway the increased PROG levels in the presence of forskolin may be attrib- uted to an increase in available substrate (PREG) and while 3ßHSD is characterised by different affinities towards its three substrates, influenced by cytochrome b5, the inhibitory effects of TRI may be attributed to any of these factors. In addition, the conversion of 17OHPROG to deoxycortisol by CYP21A1 was also inhibited at all concentrations, under both basal and forskolin-stimulated condi- tions. Analyses of CYP21A1 products, DOC and deoxycortisol, does indicate an inhibitory effect on the enzyme. As in the case of 3ßHSD, CYP21A1 also catalyses more than one reaction which, together with changes in upstream metabolite concentrations impacting on substrate supply and the presence of competing enzymes, makes analyses of the full effects of TRI on these ster- oidogenic enzymes more complex. Nevertheless, our data suggests that TRI may have a targeted mode of action against specific en- zymes. These desirable effects observed should be further defined and optimised for future therapeutic effects in suitable target populations.
In contrast to upstream enzymes, mechanisms by which effects on downstream enzymes are achieved, are more clear-cut. For example, the data suggest that both CYP11B1 and CYP11B2 are affected by TRI treatment, with inhibition most apparent at the highest concentration of the compound. This is evidenced by its modulating effect on the production of specific end products cat- alysed by these enzymes - cortisol, ALDO and 110HA4 -under basal as well as forskolin and Ang II stimulated conditions.
Current data show inhibition of cortisol and CORT at the highest TRI concentration, while T production was inhibited at all TRI concentrations assessed. This effect on T production in H295R cells indicate an inhibitory effect on 17ßHSD and may account for the decreased circulating T levels detected in our previous study in which male Wistar rats had been subjected to chronic restraint stress (Smith, 2011). It is important to note that while the adrenal is not the primary site for T biosynthesis in males, the decreased T levels in H295R cells are indicative of the inhibition of 17ßHSD, over the assayed TRI concentration range. The adrenals express low basal levels of 17ßHSD type 5 (Rege et al., 2013) which ef- fectively catalyses the conversion of A4 to T. As explained, the ef- fect on T biosynthesis as assessed here cannot be extrapolated to an expected whole body result. However, potential inhibition of T
biosynthesis should be investigated in an in vivo model, in order to prevent undesirable side-effects in especially older consumers.
With respect to the dosage used, the extract in this study has been comprehensively analysed. Not only was the extract characterised in terms of total alkaloid content, but also in terms of mesembrine content specifically. Given the large variation between different pro- prietary plant preparations from almost identical source material across many indigenous species (Smith and Krygsman, 2014b), it is important that as much as possible analytical data is provided with plant material, to enable integration of results in the literature. In contrast to recent reports on low-mesembrine containing extracts (Murbach et al., 2014; Nell et al., 2013) - which have thus far failed to show any concrete evidence of anti-stress effects in in vivo models - we report several desirable effects in this context in a Sceletium compound with high mesembrine content.
A search of the peer-reviewed literature revealed that me- sembrine in its isolated form has only once been assessed for physiological effects. Recently, the acute effects of a Sceletium ex- tract (100 mg/kg, containing 1.5% mesembrine), an alkaloid en- riched fraction of Sceletium (20 mg/kg, containing 11.8% me- sembrine) and pure mesembrine (20 mg/kg) were compared in an in vivo rodent model (Loria et al., 2014). According to doses re- ported, the mesembrine content administered in this way was 1.5 mg/kg, 2.4 mg/kg and 20 mg/kg for the three treatments re- spectively. Results reported differed substantially between treat- ments: only the alkaloid fraction was shown to have anti-de- pressant effect (although a similar but non-significant result was reported for the extract), while the pure mesembrine clearly had no effect in the forced swim model. In the elevated plus maze - a model for anxiety testing - the extract had no effect, while results for both the fraction and pure mesembrine resembled that of the anxiolytic pharmaceutical control. However, due to small sample sizes and large variability for this parameter, these results are not conclusive. Taken together, these results at best suggest that dif- ferent doses of mesembrine may have differential effects. In ad- dition, different effects may be facilitated by different alkaloid ratios in any particular extract. A puzzling finding of the study by Loria and colleagues was that none of the treatments seemed to have psychoactive properties, which is not in accordance with the plethora of anecdotal evidence, lending further support for the latter notion.
The other alkaloids in Sceletium have been chemically eluci- dated rather comprehensively (Jin, 2013). We have shown specific effects of TRI on steroid hormone production, suggesting that the alkaloids influence the catalytic activity of steroidogenic enzymes, but it should be noted that our study was confined to a closed system. Both mesembrine and mesembrenone have previously been shown to be metabolised by inducible P450 phase I enzymes in rats and in human microsomes (Meyer et al., 2015). The con- tribution of the in vivo transport, uptake and metabolism of TRI compounds, either attenuating or potentiating the effects of these alkaloids in in vivo sytems, cannot thus be ignored. The fact that the action of TRI alkaloids may possibly result in the induced ex- pression of the aforementioned enzymes, would further modulate circulating steroid hormone levels due to increased peripheral catabolismas reported in similar models (Jones et al., 2000; Ko- halmy et al., 2007; Meyer et al., 2015; Moilanen et al., 2007; Monostory and Dvorak, 2011).
Adrenal steroid hormones play vital roles in development and in the maintenance of physiological homoeostasis, and modulation of basal steroid levels. The perturbation of their peripheral meta- bolism and perhaps regulation by alkaloids may thus not in all cases lead to desirable outcomes. We have previously shown polyphenolic compounds to modulate adrenal P450 enzymes both in terms of selective inhibition of enzymes and specific steroid substrates in the context of Rooibos (Aspalathus linearis) (Schloms
and Swart, 2014). These data contributed significantly to our un- derstanding of the mechanisms by which the product exerts its reported physiological effects. To date, similar investigations into the interactions between different TRI constituents have not been conducted. Sceletium extracts are already commercially available globally, although effects on human physiological systems have not been fully elucidated. Specific in vivo or in vitro analyses on biological effects of these alkaloids - either in isolation or specified combinations and across a range of concentrations - are indicated to further characterise the actives contained in Sceletium.
5. Conclusion
In summary, our data indicate dose-dependent modulation of steroidogenic enzymes and adrenal steroid production at cellular level, which can be attributed to direct peripheral effects of the high-mesembrine containing TRI extract. The inhibitory effects of TRI on CYP17A1 may be directly responsible for our previously reported inhibition of CORT and testosterone production in rats subjected to restraint stress. Our data clearly shows that Sceletium modulates glucocorticoid production, as well as that of ALDO un- der conditions of simulated stress, and may thus have applications in the management of stress and anxiety as well as in aiding the treatment of hypertension. The potential clinical significance of the decreased adrenal androgen production should be further in- vestigated in the testes, which is the major androgen production site. In addition, effects on adrenal steroidogenesis reported in this study should be further investigated in a whole organism model to unravel molecular mechanisms of action and identify mesembrine derivatives or other Sceletium alkaloids to which beneficial and/or adverse actions could be ascribed. Use of models which allow assessment of peripheral effects in conjunction with central effects previously reported, would elucidate the potential of this ethno- medicine in the context of stress and as well as hypertension. Furthermore, a better understanding of the mechanism(s) by which TRI exerts peripheral effects would not only provide a sound physiological and biochemical foundation for potential clinical studies but may also bring to the fore possible interactions between mesembrine alkaloids and mainstream drugs.
Author contributions
Both authors contributed equally to the experimental work and preparation of this manuscript.
Acknowledgements
TrimesemineTM was a kind gift from Mr. Richard Davies (Verve Dynamics, Somerset West, South Africa) and Prof William E. Rainey (University of Michigan) kindly donated the H295R cell line. The authors would like to acknowledge Prof M Kidd (Stel- lenbosch Centre for Statistical Consultation) for assistance with statistical analysis of data. This project was funded by grants awarded to CS and ACS by the South African National Research Foundation (grant number CSUR13091742153).
Appendix A. Supplementary material
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.jep.2015.11.033.
References
Bird, I.M., Mathis, J.M., Mason, J.I., Rainey, W.E., 1995. Ca(2+)-regulated expression of steroid hydroxylases in H295R human adrenocortical cells. Endocrinology 136, 5677-5684.
Blom, W.A., Abrahamse, S.L., Bradford, R., Duchateau, G.S., Theis, W., Orsi, A., Ward, C.L., Mela, D.J., 2011. Effects of 15-d repeated consumption of Hoodia gordonii purified extract on safety, ad libitum energy intake, and body weight in healthy, overweight women: a randomized controlled trial. Am. J. Clin. Nutr. 94, 1171-1181.
Djiogue, S., Halabalaki, M., Njamen, D., Kretzschmar, G., Lambrinidis, G., Hoepping, J., Raffaelli, F.M., Mikros, E., Skaltsounis, A.L., Vollmer, G., 2014. Erythroidine alkaloids: a novel class of phytoestrogens. Planta Med. 80, 861-869.
Dvorak, Z., Vrzal, R., Maurel, P., Ulrichova, J., 2006. Differential effects of selected natural compounds with anti-inflammatory activity on the glucocorticoid re- ceptor and NF-kappaB in Hela cells. Chem. Biol. Interact. 159, 117-128.
Goroll, A.H., 2014. Moving toward evidence-based complementary care. JAMA In- tern Med. 174, 368-369.
Harvey, A.L., Young, L.C., Viljoen, A.M., Gericke, N.P., 2011. Pharmacological actions of the South African medicinal and functional food plant Sceletium tortuosum and its principal alkaloids. J. Ethnopharmacol. 137, 1124-1129.
Jin, Z., 2013. Amaryllidaceae and Sceletium alkaloids. Nat. Prod. Rep. 30, 849-868.
Jones, S.A., Moore, L.B., Shenk, J.L., Wisely, G.B., Hamilton, G.A., McKee, D.D., Tom- kinson, N.C., LeCluyse, E.L., Lambert, M.H., Willson, T.M., Kliewer, S.A., Moore, J. T., 2000. The pregnane X receptor: a promiscuous xenobiotic receptor that has diverged during evolution. Mol. Endocrinol. 14, 27-39.
Kohalmy, K., Tamasi, V., Kobori, L., Sarvary, E., Pascussi, J.M., Porrogi, P., Rozman, D., Prough, R.A., Meyer, U.A., Monostory, K., 2007. Dehydroepiandrosterone in- duces human CYP2B6 through the constitutive androstane receptor. Drug Metab. Dispos. 35, 1495-1501.
Komarnytsky, S., Esposito, D., Poulev, A., Raskin, I., 2013. Pregnane glycosides in- terfere with steroidogenic enzymes to down-regulate corticosteroid production in human adrenocortical H295R cells. J. Cell. Physiol. 228, 1120-1126.
Loria, M.J., Ali, Z., Abe, N., Sufka, K.J., Khan, I.A., 2014. Effects of Sceletium tortuosum in rats. J. Ethnopharmacol. 155, 731-735.
Matkovic, U., Pacenti, M., Trevisan, M., Palu, G., Barzon, L., 2009. Investigation on human adrenocortical cell response to adenovirus and adenoviral vector in- fection. J. Cell. Physiol. 220, 45-57.
Meyer, G.M., Wink, C.S., Zapp, J., Maurer, H.H., 2015. GC-MS, LC-MS(n), LC-high resolution-MS(n), and NMR studies on the metabolism and toxicological de- tection of mesembrine and mesembrenone, the main alkaloids of the legal high “Kanna” isolated from Sceletium tortuosum. Anal. Bioanal. Chem. 407, 761-778. Miller, W.L., Auchus, R.J., 2011. The molecular biology, biochemistry, and physiology of human steroidogenesis and its disorders. Endocr. Rev. 32, 81-151.
Moilanen, A.M., Hakkola, J., Vaarala, M.H., Kauppila, S., Hirvikoski, P., Vuoristo, J.T., Edwards, R.J., Paavonen, T.K., 2007. Characterization of androgen-regulated expression of CYP3A5 in human prostate. Carcinogenesis 28, 916-921. Monostory, K., Dvorak, Z., 2011. Steroid regulation of drug-metabolizing cyto- chromes P450. Curr. Drug Metab. 12, 154-172.
Murbach, T.S., Hirka, G., Szakonyine, I.P., Gericke, N., Endres, J.R., 2014. A tox- icological safety assessment of a standardized extract of Sceletium tortuosum (Zembrin(R)) in rats. Food Chem. Toxicol. 74, 190-199.
Nakamura, Y., Satoh, F., Morimoto, R., Kudo, M., Takase, K., Gomez-Sanchez, C.E.,
Honma, S., Okuyama, M., Yamashita, K., Rainey, W.E., Sasano, H., Ito, S., 2011. 18- oxocortisol measurement in adrenal vein sampling as a biomarker for sub- classifying primary aldosteronism. J. Clin. Endocrinol. Metab. 96, E1272-E1278.
Nell, H., Siebert, M., Chellan, P., Gericke, N., 2013. A randomized, double-blind, parallel-group, placebo-controlled trial of Extract Sceletium tortuosum (Zem- brin) in healthy adults. J. Altern. Complement. Med. 19, 898-904.
Nielsen, F.K., Hansen, C.H., Fey, J.A., Hansen, M., Jacobsen, N.W., Halling-Sorensen, B., Bjorklund, E., Styrishave, B., 2012. H295R cells as a model for steroidogenic disruption: a broader perspective using simultaneous chemical analysis of 7 key steroid hormones. Toxicol. Vitr. 26, 343-350.
Offit, P.A., 2012. Studying complementary and alternative therapies. JAMA 307, 1803-1804.
Rege, J., Nakamura, Y., Satoh, F., Morimoto, R., Kennedy, M.R., Layman, L.C., Honma, S., Sasano, H., Rainey, W.E., 2013. Liquid chromatography-tandem mass spec- trometry analysis of human adrenal vein 19-carbon steroids before and after ACTH stimulation. J. Clin. Endocrinol. Metab. 98, 1182-1188.
Schloms, L., Storbeck, K.H., Swart, P., Gelderblom, W.C., Swart, A.C., 2012. The in- fluence of Aspalathus linearis (Rooibos) and dihydrochalcones on adrenal steroidogenesis: quantification of steroid intermediates and end products in H295R cells. J. Steroid Biochem. Mol. Biol. 128, 128-138.
Schloms, L., Swart, A.C., 2014. Rooibos flavonoids inhibit the activity of key adrenal steroidogenic enzymes, modulating steroid hormone levels in H295R cells. Molecules 19, 3681-3695.
Smith, C., 2011. The effects of Sceletium tortuosum in an in vivo model of psy- chological stress. J. Ethnopharmacol. 133, 31-36.
Smith, C., Krygsman, A., 2014a. Hoodia gordonii extract targets both adipose and muscle tissue to achieve weight loss in rats. J. Ethnopharmacol. 155, 1284-1290.
Smith, C., Krygsman, A., 2014b. Hoodia gordonii: to eat, or not to eat. J. Ethno- pharmacol. 155, 987-991.
Smith, C., Viljoen, J.T., McGeachie, L., 2014. African drumming: a holistic approach to reducing stress and improving cardiovascular health. J. Cardiovasc. Med. 15, 41-446.
Stafford, G.I., Pedersen, M.E., van Staden, J., Jager, A.K., 2008. Review on plants with CNS-effects used in traditional South African medicine against mental diseases. J. Ethnopharmacol. 119, 513-537.
Sunden, H., Holland, M.C., Poutiainen, P.K., Jaaskelainen, T., Pulkkinen, J.T., Palvimo, J.J., Olsson, R., 2015. Synthesis and biological evaluation of second-generation tropanol-based androgen receptor modulators. J. Med. Chem. 58, 1569-1574.
Terburg, D., Syal, S., Rosenberger, L.A., Heany, S., Phillips, N., Gericke, N., Stein, D.J., van Honk, J., 2013. Acute effects of Sceletium tortuosum (Zembrin), a dual 5-HT reuptake and PDE4 inhibitor, in the human amygdala and its connection to the hypothalamus. Neuropsychopharmacology 38, 2708-2716.
Van Wyk, B.E., Wink, M., 2009. Medicinal plants of the world, 4th ed. Briza, Pretoria, p. 292.
Vos, T., 2015C. GBD2013Collaborators, 2015. Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990- 2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 385, 117-171.
Winther, C.S., Nielsen, F.K., Hansen, M., Styrishave, B., 2013. Corticosteroid pro- duction in H295R cells during exposure to 3 endocrine disrupters analyzed with LC-MS/MS. Int. J. Toxicol. 32, 219-227.