Influence of the Plant Extract Complex “Admax” on Global Gene Expression Levels
in Cultured Human Fibroblasts
Anatoly Antoshechkin, MD, Ph.D., Jose Olalde, M.S., Marina Antoshechkina, MD. Vladimir Briuzgin, M.D. Leonid Platinskiy, M.D.
ABSTRACT. Ethanol/water extracts from roots of Leuzea carthamoides Iljin, Rhodiola rosea L., Eleutherococcus senticosus Maxim, and from dry berries of Schizandra chinensis Baill. are known as adaptogenic remedies, which enhance physical endurance, counteract fatigue and re- store suppressed immunity. Molecular mechanisms underlying effects of the extracts are poorly understood. In this study, a combination of these four extracts called AdMaxTM (Nulab, Inc., Florida) was examined for its ability to influence gene expression levels in cultured human fibrob- lasts in vitro with the help of whole-genome Affymetrix oligonucleotide microarrays. We showed that AdMax treatment results in significant changes (at least 2 fold, p < .05) in expression of 67 genes that are in- volved in metabolism of protein, nucleic acids, lipid and carbohydrates, in regulation of transcription, protein and ion transport, response to stim- ulus and stress. Enhancing expression of the PANK2 gene is of special interest in connection with AdMax ability to enhance physical endurance and counteract fatigue. PANK2 encodes a mitochondrial enzyme pan- tothenate kinase 2, which provides coenzyme A biosynthesis and thereby plays crucial role in energy metabolism. Partial deficiency of PANK2 gene activity leads to pantothenate kinase-associated neurodegeneration. In this connection potential therapeutic use of AdMax in patients with neurodegenerative diseases is discussed.
KEYWORDS. Gene expression, energy metabolism, adaptogenic properties
INTRODUCTION
Leuzea carthamoides Iljin, Rhodiola rosea L., Eleutherococcus sentico- sus Maxim, and Schizandra chinensis Baill. are known as the most effective adaptogenic plants. Ethanol/water extracts from the plants are capable of increasing nonspecific resistance in human and animal organisms towards external and internal stress factors. This allows for the organism to adapt to stressful conditions (Panossian & Wagner, 2005). Studies of mechanisms of adaptive response of the organism to stress have revealed the crucial role of adaptive reactions in prevention of large number of widespread illnesses, including cardiovascular diseases, type 2 diabetes, depression, and others. These diseases are stress-induced and their prevention is deter- mined by nonspecific resistance to stress (Chrousos, 1998). Adaptogenic plant extracts also have an active role in the elimination of stress-induced physiological changes in the body, including physical and mental fatigue, elevation of endurance during intensive exercises, and the increase of ATP and muscle protein syntheses (De Bock, Eijnde, Ramaekers, & Hespel, 2004; Kholodova, Tugai, & Zimina, 1997).
Adaptogenic properties of ethanol/water extracts from roots of Leuzea, Rhodiola, Eleutherococcus, and from berries of Schizandra are deter- mined by their biologically active constituents. Not all of the constituents have been identified, but molecular structures and biological activities of the ones most important to adaptogenic action have been estab- lished. Characteristic biologically active compounds are: ecdysterone (20- hydroxyecdisone) for Leuzea root extract (Syrov & Kurmukov, 1967); salidroside and rosavin for Rhodiola extract (Ganzera, Yayla, & Khan, Antoshechkin et al 295, 2001); several eleutherosides for Eleutherococcus (Hikino, Takahashi, Otake, & Konno, 1986); and several schizandrins for Schizandra berry extract (Kochetkov, Khorlin, & Chizov, 1961).
These active compounds belong to different classes of organic molecules (sterols, glycosides, lignans) and produce distinctive biological effects in various tests. However, they possess lower effectiveness when isolated from the plant extract in comparison with the effectiveness of the total extract indicating synergistic action of the extract’s components. It is likely that some constituents of the plant extract act as adjuvant compounds, which enhance the activity of the compounds actually responsible for the biological effect. Several other mechanisms of synergistic interaction of plant extract constituents can determine biological effects of the extract (Gilbert & Alves, 2003).
Mechanisms of action of the adaptogenic plant extracts are poorly un- derstood. It has been suggested that active components of adaptogenic extracts may interact with cell surface hormone receptors or with cell membrane-localized members of various signaling pathways resulting in specific changes of gene expression (Lafont & Dinan, 2003).
To test this directly, we examined the ability of a complex of dry ethanol/water extracts from the four adaptogenic plants AdMax to in- fluence gene expression levels in cultured human fibroblasts using full- genome oligonucleotide microarrays.
MATERIALS AND METHODS The Extract Complex “AdMax”
AdMaxTM is a combination of lyophilized ethanol/water extracts from dry roots of Leuzea carthamoides Iljin, Rhodiola rosea L., Eleutherococcus senticosus Maxim, and from dry berries of Schizandra chinensis Baill. All these plants are wild growing in Southern and Eastern Siberia. The roots and berries of the corresponding plants were harvested, washed with water, dried and transported in dry state. Seventy percent ethanol/water extraction of the each plant material was performed with subsequent lyophilization, mixing and incapsulation of the obtained powders. The preparation is manufactured by Nulab, Inc., Clearwater, Florida 33765, USA. For AdMax standardization 70% ethanol/water extract of AdMax was analyzed by means of reverse phase gradient C 18 HPLC (Agilent 1100) coupled with ESI-MS (ThermoFinnigan LCQ). Ecdysterone (20-hydroxyecdysone) is characteristic for AdMax and can serve as its marker substance.
296 JOURNAL OF DIETARY SUPPLEMENTS
Cell Culture and Total RNA Isolation
Human fibroblast cell line MRC-5 was grown under standard conditions in minimum essential medium (MEM, Gibco) supplemented with 10% of fetal bovine serum (Gibco) at 37◦C. When cells were close to confluence, culture medium was replaced with the medium containing AdMax at con- centration of 3 μg/ml and cells were grown for additional 16 hr. Control cells were grown in parallel in identical medium without AdMax. Total RNA was isolated using Trizol (Invitrogen) reagent following manufac- turer’s instructions. Control and experimental cell cultures were grown in triplicates and processed independently producing six RNA samples, each of which was analyzed on an individual microarray chip.
Microarray Analysis and Data Processing
Microarray analysis was performed in accordance with standard pro- cedures recommended by Affymetrix, Inc using Affymetrix GeneChip Human Genome U133 Plus 2.0 arrays. Following hybridization and scan- ning, raw data in the form of image files were converted to gene expression values using Affymetrix GeneChip Operating Software (GCOS), which utilizes MAS 5.0 algorithm for data normalization, background subtrac- tion, estimation of nonspecific binding, calculation of detection p-values, and generation of “presence” calls. The software’s term “presence” call means high level of specific hybridization in the particular microarray spot and serves as quality control of the hybridization process. Two-tailed Stu- dent’s t-test assuming unequal sample variance was used to identify genes that displayed significant changes in the mean expression levels between control and treated samples of at least 2 fold with the t-test p-value less then 0.05. Only probes that were called “present” in at least two sam- ples were considered in the analysis to ensure robustness of our statistical approach.
RESULTS AND DISCUSSION
Results of qualitative HPLC/ESI-MS fingerprinting analysis of AdMax composition are presented in Figure 1. The chromatogram contains more than 80 prominent peaks, each of which characterizes the amount of corre- sponding compound in AdMax. The chromatogram as a whole can be used as AdMax fingerprint in combination with pattern recognition algorithms.
FIGURE 1. HPLC/ESI-MS registered base peak ion chromatogram of AdMax composition upon negative (−) ionization with two scan ranges (m/z 95-420 and 405-2000).
FIGURE 2. Major biological processes affected by AdMax treatment. Number of genes involved in a particular process is indicated.
Microarray analysis of more than 38,500 human genes identified 67 genes that changed their expression levels by at least 2 fold (P < 0.05) upon treatment with AdMax (see Appendix) indicating that AdMax acts as a potent modulator of activity of some human genes.
Genes shown to be affected by AdMax are involved in a variety of cellular processes including protein, nucleic acid, lipid and carbohydrate metabolism, regulation of transcription, protein and ion transport, response to stimulus and stress (Figure 2).
Seven AdMax-regulated genes have also been shown to participate in several well-established biological pathways (Table 1).
Data obtained in this study show that AdMax treatment modulates differ- entially expression of a number of genes in human fibroblasts and demon- strate the broad spectrum of the AdMax biological activity. We show that AdMax treatment most strongly affects genes involved in metabolic pro- cesses. Because of the interconnection of metabolic processes in the cell, the observed changes of expression levels could be either a direct conse- quence of AdMax interaction with the members of a particular pathway, or an indirect effect of compensatory interactions between pathways. Studies of gene expression is a fast growing field of functional genomics that is aiming at the detailed reconstruction of biological pathways from expres- sion data. It has already proven to provide invaluable insights into possible mechanisms of action of herbal preparations and its utility will certainly continue to grow in the future.
TABLE 1. Biological Pathways in Which Admax-Affected Genes Are Involved
Our data suggest that physiological effects of AdMax can be attributed, partially, to regulation of such genes as PANK2 and IGHG1. Expression of PANK2 increases 2.3 (p = .004) fold upon treatment with AdMax. PANK2 encodes a mitochondrial enzyme pantothenate kinase 2, which activates coenzyme A (CoA) biosynthesis. CoA and acetyl CoA play a key role in energy metabolism. Decrease in acetyl CoA supply to the citrate cycle leads to fatigue, decline of physical performance and reduction of muscle protein synthesis. Activation of CoA biosynthesis by AdMax is likely to counteract these manifestations of energy insufficiency (De Bock et al. 2004; Kurmukov, 1967; Syrov & Kurmukov, 1967). Decrease of pantothenate kinase 2 activity is also associated with degeneration of the neurons, which is probably caused by energy insufficiency in neurons as well (Westaway, Ching, Levinson, Gitschier, & Hayflick, 2006).
Our results also indicate that AdMax increases expression of IGHG1 by 2.0 fold (P = 0.011). IGHG1 controls the binding of immunoglobulines to antigens, providing a link to immune system response to infection. Activity of the immune system is suppressed by a number of factors, including physical stress and the action of various chemicals. Increase of IGHG1 expression may contribute to restoration of immune system activity, which has been observed upon treatment with AdMax and its constituents (Azizov, Seifulla, & Chubarova, 1997; Kormosh, Laktionov, & Antoshechkina, 2006).
Gene expression analyses generate important data that can be used to elucidate biological activity of pharmaceutical drugs and promote our understanding of molecular mechanisms of their action (Breitling, 2006). The use of microarray technology is especially important for objective registration of bioactivity of herbal preparations and evaluation of their effects on health. It has recently been suggested that gene expression analyses can be used for development of a screening system to determine the value of natural medicinal products (Katz, Harris, Lau, & Chau, 2006).
Because microarray analysis of gene expression is used recently for drug discovery, our data might also be considered from this point of view. As it has been shown that partial deficiency of PANK2 gene activity leads to pantothenate kinase-associated neurodegeneration (PKAN) (Westaway et al. 2006), Parkinson’s disease (Tretter, Sipos, & Adam-Vizi, 2004), and Alzheimer’s disease (Atamna & Frey, 2007). AdMax ability to enhance expression of PANK2 gene might be used to prevent or mitigate these diseases. To examine this supposition, we are planning to carry out a pilot preclinical trial of AdMax in patients with the initial manifestations of Parkinson’s disease.
ACKNOWLEDGMENT
The work was carried out in cooperation with Microarray Facility of Pennsylvania University, PA, USA.
REFERENCES
Atamna H, Frey WH, 2nd. Mechanisms of mitochondrial dysfunction and energy deficiency in Alzheimer’s disease. Mitochondrion. 2007;7:297–310.
Azizov AP, Seifulla RD, Chubarova AV. Effect of tincture of leuzea and leveton on humoral immunity of athletes. Exsp Klin Farmacol. 1997;60:47–48.
Breitling R. Biological microarray interpretation: The rules of engagement. Biochim Biophys Acta. 2006;1759:319–27.
Chrousos GP. Stressors, stress, and neuroendocrine integration of the adaptive re- sponse. Ann NY Acad Sci. 1998;851:311–35.
De Bock K, Eijnde BO, Ramaekers M, Hespel P. Acute Rhodiola rosea intake can im- prove endurance exercise performance. Int J Sport Nutr Exerc Metab. 2004;14:298– 307.
Antoshechkin et al. 301
Ganzera M, Yayla Y, Khan IA. Analysis of the marker compounds of Rhodiola rosea L. (golden root) by reversed phase high performance liquid chromatography. Chem Pharm Bull (Tokyo). 2001;49:465–67.
Gilbert B, Alves LF. Synergy in plant medicines. Curr Med Chem. 2003;10:13–20. Hikino H, Takahashi M, Otake K, Konno C. Isolation and hypoglycemic activity of eleutherans A,B,C, D, E, F, and G: Glycans of Eleutherococcus senticosus roots. J
Nat Prod. 1986;49:293–97. Katz S, Harris R, Lau JT, Chau A. The use of gene expression analysis and proteomic
database in the development of a screening system to determine the value of natural
medical products. Evid Based Complement Alternat Med. 2006;3:65–70. Kholodova IuD, Tugai VA, Zimina VP. Effect of vitamin D3 and 20-hydroxyecdysone on the content of ATP, creatine phosphate, carnosine and Ca2+ in skeletal muscles.
Ukr Biochim Zh. 1997;69:3–9. Kochetkov NK, Khorlin A, Chizov OS. Schizandrin—lignan of unusual structure.
Tetrahedron Lett. 1961;730–34. Kormosh N, Laktionov K, Antoshechkina M. Effect of combination of extract from
several plants on cell-mediated and humoral immunity of patients with advanced
ovarian cancer. Phytother Res. 2006;20:424–25. Lafont R, Dinan L. Practical uses for ecdysteroids in mammals including humans: an
update. J Insect Sci. 2003;3:7–35. Panossian A, Wagner H. Stimulating effect of adaptogens: An overview with particular
reference to their efficacy following single dose administration. Phytother Res.
2005;19:819–38. Syrov VN, Kurmukov AG. Anabolic activity of phytoecdysone — ecdysterone isolated
from Rhaponticum carthamoides (Will.) Iljin. Farmacol Toksikol. 1967;39:690–93. Tretter L, Sipos I, Adam-Vizi V. Initiation of neuronal damage by complex I deficiency
and oxidative stress in Parkinson’s disease. Neurochem Res. 2004;29:569–77. Westaway SK, Ching KN, Levinson B, Gitschier J, Hayflick SJ. Gene symbol: PANK2. Disease: Pantothenate kinase-associated neurodegeneration (PKAN). Hum Genet.
2006;119:671–72.
302 JOURNAL OF DIETARY SUPPLEMENTS
APPENDIX
AdMax-regulated genes, Affymetrix Probe Id, gene symbol, ratio of gene expression levels in AdMax-treated cells vs. control cells, t-test p- value, and gene title are indicated
Probe Id
1566709 at 235564 at
233613 x at
232413 at 235555 at
221973 at 243378 at
208798 x at 232570 s at 214163 at 233555 s at 235462 at
227148 at
210794 s at
201295 s at 1557193 at
212176 at 210425 x at 241863 x at 232461 at 1564200 at 214048 at 242335 at 228718 at 224565 at 213910 at 221920 s at 214657 s at
Gene Symbol
TNFAIP8
ZNF117 REXO2
PGBD3 —
LOC150759 —
GOLGA8A ADAM33 C1orf41 SULF2 CPEB2
PLEKHH2
MEG3 LOC647251 WSB1 PTPN2
C6orf111 GOLGA8B CCDC39 AHI1 LOC646324 MBD4 SLC25A37 ZNF44 TncRNA IGFBP7 SLC25A37 TncRNA
Ratio
0.191
0.210 0.274
0.300 0.308
0.314 0.316
0.323 0.335 0.366 0.380 0.380
0.393 0.397
0.398 0.405
0.405 0.406 0.411 0.412 0.412 0.413 0.423 0.425 0.429 0.443 0.444 0.449
t-test
0.002
0.002 0.020
0.007 0.008
0.004 0.025
0.000 0.033 0.005 0.033 0.009
0.002 0.003
0.010 0.030
0.003 0.012 0.005 0.039 0.039 0.009 0.015 0.002 0.002 0.015 0.034 0.005
Gene Title
Tumor necrosis factor, alpha-induced protein 8
Zinc finger protein 117 (HPF9) REX2, RNA exonuclease 2 homolog (S.
cerevisiae) PiggyBac transposable element derived 3 Transcribed locus, moderately similar to
NP 659411.1 Hypothetical protein LOC150759 Transcribed locus, moderately similar to
NP 060312.1 Golgi autoantigen, golgin subfamily a, 8A ADAM metallopeptidase domain 33 Chromosome 1 open reading frame 41 Sulfatase 2 Cytoplasmic polyadenylation element
binding protein 2 Pleckstrin homology domain containing,
family H member 2 Maternally expressed 3
WD repeat and SOCS box-containing 1 Protein tyrosine phosphatase, non-receptor
type 2 Chromosome 6 open reading frame 111 Golgi autoantigen, golgin subfamily a, 8B Coiled-coil domain containing 39 Abelson helper integration site 1 Hypothetical protein LOC646324 methyl-CpG binding domain protein 4 Solute carrier family 25, member 37 Zinc finger protein 44 Trophoblast-derived noncoding RNA Insulin-like growth factor binding protein 7 Solute carrier family 25, member 37 Trophoblast-derived noncoding RNA
Probe Id
221427 s at
226363 at
244517 x at 226334 s at
241345 at 213653 at 221139 s at 209006 s at 238089 at 1563496 at 213703 at 230149 at 229141 at 221421 s at
222074 at 243824 at
235408 x at 228023 x at 230774 at
226665 at
226404 at
214093 s at
227099 s at 1556597 a at 1559102 at 211693 at
220466 at 232786 at 216361 s at
242790 at
Gene Symbol
CCNL2 LOC643556
ABCC5
RNF146 AHSA2
ZNF265 METTL3 CSAD C1orf63 MAN2C1 LOC202460 LOC150759 — WDR33 ADAMTS12
UROD —
ZNF117 AMY2B ZADH1
AHSA2
RNPC2
FUBP1
LOC387763 LOC284513 — IGHG1
CCDC15 COG6 MYST3
SNF8
Ratio
0.449 0.449
0.451
0.453 0.454
0.459 0.459 0.462 0.462 0.463 0.465 0.466 0.470 0.475 0.477
0.478 0.480
0.482 0.484 0.486
0.486
0.489
0.492
0.495 0.498 0.499 2.003
2.157 2.180 2.193
2.226
t-test
0.000 0.000
0.003
0.005 0.008
0.018 0.002 0.009 0.035 0.007 0.038 0.026 0.006 0.014 0.006
0.019 0.031
0.003 0.009 0.028
0.002 0.004 0.008
0.031 0.033 0.026 0.011
0.003 0.009 0.009
0.027
Gene Title
Cyclin L2 Similar to Aurora kinase A-interacting
protein (AURKA-interacting protein) ATP-binding cassette, sub-family C
(CFTR/MRP), member 5 Ring finger protein 146 AHA1, activator of heat shock 90kDa
protein ATPase homolog 2 Zinc finger protein 265 Methyltransferase like 3 Cysteine sulfinic acid decarboxylase Chromosome 1 open reading frame 63 Mannosidase, alpha, class 2C, member 1 Hypothetical protein LOC202460 Hypothetical protein LOC150759 Transcribed locus
WD repeat domain 33 ADAM metallopeptidase with
thrombospondin type 1 motif, 12 Uroporphyrinogen decarboxylase MRNA; cDNA DKFZp779F2127 (from clone
DKFZp779F2127) Zinc finger protein 117 (HPF9) Amylase, alpha 2B (pancreatic) Zinc binding alcohol dehydrogenase,
domain containing 1 AHA1, activator of heat shock 90kDa
protein ATPase homolog 2 RNA-binding region (RNP1, RRM)
containing 2 Far upstream element (FUSE) binding
protein 1 Hypothetical LOC387763 Hypothetical protein LOC284513 CDNA clone IMAGE:4791593 Immunoglobulin heavy constant gamma 1
(G1m marker) Coiled-coil domain containing 15 component of oligomeric golgi complex 6 MYST histone acetyltransferase (monocytic
leukemia) 3 SNF8, ESCRT-II complex subunit, homolog
(S. cerevisiae)
Antoshechkin et al. 303
304
JOURNAL OF DIETARY SUPPLEMENTS
Probe Id
228966 at 243233 at
235944 at 231588 at 217288 at 216387 x at 244710 at 1558372 at 231680 at
Gene Symbol
PANK2 PAN3
HMCN1 PRCP FLJ13236 — FLJ32786 — LOC646590
Ratio
2.267 2.390
2.575 2.650 2.672 2.879 3.199 5.596 6.290
t-test
0.004 0.026
0.030 0.035 0.023 0.002 0.025 0.047 0.004
Gene Title
Pantothenate kinase 2 (Hallervorden-Spatz syndrome)
PAN3 polyA specific ribonuclease subunit homolog (S. cerevisiae)
Hemicentin 1 Prolylcarboxypeptidase (angiotensinase C) Hypothetical protein FLJ13236 — Hypothetical protein FLJ32786 CDNA FLJ34038 fis, clone FCBBF2005645 Hypothetical protein LOC646590
doi: 10.1080/19390210802414337
