Sage Journals: Discover world-class research

Abstract

Purpose: Amyotrophic lateral sclerosis (ALS) is a lethal neurodegenerative condition, in which motor neurons start to degenerate due to the accumulation of protein aggregates in the neuron cytoplasm. The formation of aggregates causes neurotoxicity, facilitated by the N-terminal domain (NTD) of the transactive response DNA-binding protein-43 (TDP-43). Therapies used to treat ALS manage secondary symptoms, but do not stop the activity of the rogue NTD domain of TDP-43. Therefore, new drug candidates should be designed to deal efficiently with this disease by inhibiting the domains involved in the development of ALS. This study determined the chemical affinity of aromatic medicinal compounds with NTD. Screening of 1323 medicinal compounds was conducted with PYRX 0.9 software against NTD. Compounds obtained from this analysis were further used to predict absorption, distribution, metabolism, excretion, and toxic (ADMET) properties and their effect on major gene targets of ALS. Results: From 1300 + compounds, acetovanillone showed binding affinity for NTD and had good ADMET and drug likeness attributes. This compound reduced the expression of CXCL2, NOP56, and SOD1 genes implicated in ALS pathogenesis. Conclusion: These results concluded that acetovanillone is a candidate drug for in vitro and clinical studies into the exploitation of drugs within ALS therapeutics.

Keywords

Acetovanillone gene expression quinones medicinal compounds virtual screening

Introduction

Among different neurodegenerative disorders, amyotrophic lateral sclerosis (ALS) is far more malignant than the rest. In this disease, mechanical and respiratory functions start to deteriorate due to the degeneration of motor neurons. The accumulation of toxic protein in the cytoplasm of these neurons causes their degradation. According to the latest analytical data, the disease prevalence of ALS is highest in America (12: 1 00 000 individuals) followed by Europe (2: 1 00 000 individuals), and lowest in Asia (1:1 00 000 individuals).¹ There are 2 forms of ALS, classified based on environmental and genetic factors.^2-4 Sporadic ALS is a form of ALS which is believed to be idiopathic, but recent studies show that both environmental and genetic factors instigate disease development.^3,4 Familial ALS is linked to a genetic mutation inherited in the autosomal dominant pattern among families. More than 70%-80% of affected individuals rapidly develop symptoms, with a mean survival rate of 1 to 2 years if left untreated. In 15% of patients, symptom manifestation is rather sluggish, with a mean survival rate of 10 or more years.⁵ Ultimately, all patients affected with either version of ALS die due to respiratory complications.

Drugs are being developed to slow down disease progression, avoid complications, and preserve the integrity of motor neurons in affected patients. Two drug candidates, Edaravone and Riluzole, have been approved by different drug regulatory authorities. These drugs are designed to extend the mean survival rate, provide symptomatic treatment and intervene with heart rate; however, these compounds may also make individuals vulnerable to fungal infection. However, isolating major targets of ALS has never been the objective for designing efficient drugs, due to limited understanding of the molecular mechanism underlying this disease. Improvement in genomic sciences has revealed a promising new molecular target that is involved in catastrophic processes such as accumulation of protein aggregates and neurotoxicity in ALS caused by transactive response DNA-binding protein-43 (TDP-43).⁶

TDP-43 is a highly conserved 43 KDa eukaryotic protein involved in both DNA and RNA processing. Both RNA and DNA-binding properties are provided by different domains present in this protein. These domains are the N-terminal domain (NTD), which is equipped with a nuclear localization signal, the RNA-recognition motifs (RRM1 and RRM2), the nuclear export signal, and the C-terminal domain.⁷ Mutation in the C-terminal domain leads to the formation of cytoplasmic protein inclusions, whereas the NTD mediates the accumulation and aggregation of proteins, including TDP-43 itself, and increases toxicity.⁸ Unfortunately, the in-depth molecular process is still unknown. However, recent studies have highlighted that deletion of the NTD from TDP-43 substantially reduces the toxic protein content from the motor neurons of mouse models, and rejuvenates motor and cognitive functions.⁸

Therefore, the main target considered for our study is the NTD of TDP-43, which is implicated in causing self-aggregation, the accumulation of unprocessed proteins in the cytoplasm of neurons, and the facilitation of neuron degradation from toxicity. We targeted this domain with aromatic medicinal compounds, using a structure-based virtual screening strategy to obtain drug candidates. The effect of these compounds on gene expression was predicted in order to acquire good inhibitors for a robust treatment of ALS.

Results

Interaction Analysis

The docking interaction of 1323 ligands from AromaDb was tested against the NTD of TDP-43. The active site of the NTD was predicted and ligands were directed towards the active interface. The analysis of such a large library of small molecules was assisted by PYRX 0.9 virtual screening software. This software quickly determined the interaction of ligands with a receptor, and selected ligands based on chemical affinity with the active amino acids of the target receptor. From this evaluation, only 1 ligand acetovanillone (apocynin) showed affinity for the active site residues of the NTD receptor. Ligands showing no affinity were discarded from further analysis (Table 1). The pharmacophore modelling revealed that this ligand established hydrogen bonding with GLU17 of the NTD receptor via the hydroxyl group of its structure (Figure 1). Moreover, adjacent residues of the receptor formed hydrophobic interactions (π-alkyl bonding) with PRO19 through a benzene ring of acetovanillone.

Figure 1.

Illustration of binding mode of acetovanillone with N-terminal domain.

Table 1.

Summary of Chemical Interaction Between NTD of TDP-43 With Acetovanillone.

No	Receptors	Ligands	COACH identified active residues	Bonding between ligand and receptor	Binding energy (Kcal/mol)	Ligand efficiency (Kcal/mol)	Inhibition constant (Ki)	LigRMSD calculation (Å)
1	TDP-43 N-Terminal Domain (6T4B)	Compound CID: 2214 (Acetovanillone)	SER2, GLU17, PRO19, GLU21, SER29, THR30, GLN34	OH-GLU7 (Hydrogen Bond), Benzene Ring-PRO19 (Pi-Alkyl Bond)	-4.8	-0.4	31.10 mM	0.87

Abbreviations: NTD, N-terminal domain; TDP-43, transactive response DNA-binding protein-43.

The results from the virtual screening were further validated by re-performing the docking of the NTD with edaravone, using Autodock 4.2 and keeping the same docking parameters. Edaravone showed no hydrogen bonding with the NTD residues, except for π-anion bonding with GLU9 (Supplemental Figure 1). Unlike acetovanillone, edaravone cannot instigate any inhibitor response against this receptor. This is because chemical interaction with the active site residues causes functional changes in the receptor which disrupt its activity, and is evident in previous studies.^9,10 The main mechanism of adaravone is to scavenge reactive oxygen radicals and regulate inflammation and neuroprotection.¹¹ The validation proves that edaravone has no affinity for the NTD. Therefore, acetovanillone possesses some inhibitory potential to avert protein aggregation and toxicity in neurons, which is facilitated by the NTD of TDP-43.

Toxic Profile Analysis

The prediction of acute toxicity of acetovanillone was done using the GUSAR database. The algorithm of this database uses quantitative structure-activity relationship models to calculate the lethal dose-50 of a compound based on its chemical structure. The acute toxicity of acetovanillone for oral and intraperitoneal passage is 2,199,000 and 6 59,800 mg/kg respectively. It is a class 4 chemical as classified by the Organisation for Economic Co-operation and Development (OECD) project. Using Adver-Pred and ROSC-Pred, the side effects of this compound were determined, including cardiac failure and stomach and kidney damage upon overdose (see Table 2). These results were further validated in the literature. According to these studies, the LD₅₀ value of acetovanillone for oral passage is 9,000 mg/kg and more than 423 mg/kg for intraperitoneal passage.^12,13 Such a discrepancy in results indicates a serious error in the algorithm, which needs to be improved by incorporating more training set models.

Table 2.

Lethal Dosage Calculation of Acetovanillone.

No	Compounds	LD₅₀ value for oral route (mg/kg)	Reported LD₅₀ value for oral route (mg/kg)	LD₅₀ value for intraperitoneal route (mg/kg)	Reported LD₅₀ value for intraperitoneal route (mg/kg)	Chemical classification by OECD project	Adverse effects	Organ specific damage
1	Acetovanillone	2,199,000 mg/kg	> 9000 mg/kg	659,800	> 423	Class 4	Cardiac failure	Stomach, kidney

ADME of Screened Compounds

To determine the efficacy and safety of the compounds, a prediction of the absorption, distribution, metabolism and excretion (ADME) properties is of vital importance in order to procure regulatory approval for clinical and animal trials. Such predictions are facilitated by the pKCSM and SwissADME databases. Acetovanillone was analyzed for ADME properties as summarized in Table 3. It was observed that this compound has 93% gastrointestinal (GI) absorption, moderate water solubility, and moderate skin permeability. This compound has no affinity for p-glycoprotein and cytochrome P450 (CYP) variants. It conforms to the Lipinski rule of 5 drug ability criteria, and possesses a high bioavailability score. No toxicity was observed that might cause DNA mutation or cardiac issues.

Table 3.

ADME Pattern of Acetovanillone.

Properties	Compound
Properties	acetovanillone
Absorption
Human intestinal absorption	High
Water solubility (log mol/L)	−1.809 (Soluble)
Skin permeability (Log Kp)	−2.781 cm/s
Distribution
P-glycoprotein substrate	No
P-glycoprotein I inhibitor	No
P-glycoprotein II inhibitor	No
BBB permeability (log BB)	Yes (0.348)
CNS permeability (log PS)	Yes (-2.276)
Metabolism
CYP2D6 substrate	No
CYP3A4 substrate	No
CYP1A2 inhibitor	No
CYP2C19 inhibitor	No
CYP2C9 inhibitor	No
CYP2D6 inhibitor	No
CYP3A4 inhibitor	No
Excretion
Total clearance (log mL/min/kg)	0.63
Renal OCT2 substrate	No
Toxicity
AMES toxicity	No
Hepatotoxicity	No
hERG inhibition	No
Eye irritation	No
Carcinogenicity	No
Drug likeness and bioavailability score
Lipinski	Yes; 0 violation
Bioavailability score	0.55 (High)

Abbreviations: BBB: blood brain barrier, CNS: central nervous system, AMES: Salmonella typhimurium reverse mutation assay, HERG: human ether-à-go-go-related gene.

Gene Expression Analysis

The effect of acetovanillone was evaluated against the major gene targets of ALS. This was done using the in silico gene expression database DIGEP-Pred, which uses the structural formulae of the compounds to determine possible inhibitory effects on genes.¹⁴ Results showed that acetovanillone has downregulatory effect on CXCL2, producing a heightened inflammatory reaction in the CNS¹⁵; on NOP56, facilitating the degradation of motor neurons¹⁶; and on the superoxide dismutase-1 (SOD1) gene, as an oxidative stress inducer.¹⁷ All of these genes contribute to the pathogenesis of ALS. Besides downregulation, this compound promotes the expression of GCLM gene, an oxidative stress reducer,¹⁸ which is downregulated in ALS, as shown in Table 4. The downregulation of GCLM aggravates the ALS pathology by disrupting the activity of antioxidant proteins, resulting in mitochondrial impairment and degeneration of neurons.¹⁹ We validated our results with edaravone by checking its activity with the same genes in order to explain their action on ALS. Edaravone upregulates the activity of TMSB4X, which activates neuronal repair and provides protection for the central nervous system (CNS) and neurons.¹¹ This is the primary function of this drug: no other genetic activity was observed.

Table 4.

Alteration in Gene Expression Caused by Acetovanillone and Edaravone.

No	Compounds	% Inhibition/promotion	Gene sectors	Role/function in ALS pathology
1	Acetovanillone	79.4	CXCL2 ↓	Aberrant pro-inflammatory response
		76.6	NOP56 ↓	Motor neurons degeneration by altered RNA processing
		53.1	SOD1 ↓	Oxidative stress, protein misfolding, DNA damage
		65.3	GCLM ↑	Suppresses the activity of antioxidant proteins
2	Edaravone	71.7	TMSB4X ↑	Neuron growth and repair

Abbreviations: ALS, amyotrophic lateral sclerosis; SOD1, superoxide dismutase-1.

Discussion

Cytoplasmic TDP-43 aggregation and toxicity are mediated by the NTD of TDP-43. This domain promotes the accumulation of protein inclusions and aggregates in the cytoplasm of neurons, activating the activity of caspase-3, and leading to neurodegeneration. The main function of TDP-43 is to process pre-mRNA and certain proteins, as well as microRNA biogenesis and RNA metabolism. It is commonly found in the nucleus of neurons. Aberration in the NTD of TDP-43 thus causes an accumulation of unprocessed proteins and mRNAs, further exacerbating disease progression.^7,8

Different synthetic drugs are being developed to halt the progression of ALS by managing secondary conditions. These interventions provide temporary relief, but cannot slow down disease progression. Therefore, the main focus of drug development studies should be: (1) To downregulate the expression or inhibit the activity of mutated TDP-43 and reduce the proteinopathy. (2) To improve mitochondrial activity for increased expression of antioxidant proteins and prevention of reactive oxygen species formation. (3) Neuronal protection and repair. Approved therapeutics for ALS, such as edaravone, provide neuroprotection by scavenging reactive oxygen species, whereas the activity of other pathogenic targets such as SOD1 and other pro-inflammatory proteins is ignored.

Natural compounds, especially aromatic medicinal compounds, are precious commodities, and their utilization can be beneficial in unlocking potential therapeutic solutions for this disease. In medieval times, various neuropathologies were treated by the preparation of different aromatic medicinal plant extracts which improve cognitive and motor activities.²⁰ Chemical analysis of these extracts revealed bioactive aromatic compounds. These compounds were able to provide multiple pharmacological functions, useful for stunting the progression of ALS and promoting neurite growth.²¹

We tested aromatic medicinal compounds as potential drug candidates by procuring various phytochemicals and evaluating them against the NTD of TDP-43. Out of 1323 ligands, acetovanillone showed interaction with the NTD by forming hydrogen bonds with active residues. This binding disrupts the substrate processing site, leading to protein inhibition. This compound possesses neuroprotective attributes, and rejuvenates the injured brain by suppressing the activity of the nicotinamide adenine dinucleotide phosphate (NADPH) oxidase involved in aggravating brain injury. It also promotes neuronal and neurite growth and reduces reactive oxygen species and neurodegeneration.²²

Further absorption, distribution, metabolism, excretion, and toxicity (ADMET) studies revealed the safe nature of acetovanillone, but reports of acute toxicity from the GUSAR database showed substantial aberrancy in the lethal dose-50 when cross-validated with the in vitro experiments. This can be altered by incorporating more in vitro training set models into the database to improve accuracy. This compound has the tendency to penetrate inside the blood brain barrier (BBB) and CNS without hindrance, whereas most CNS drugs require external carriers for transport. We also evaluated and validated the effect of acetovanillone and edaravone on gene expression. Acetovanillone downregulated essential targets of ALS, whereas edaravone only upregulated the activity of TMSB4X. Thymosin-4 beta, also known as TMSB4X, activates a neuronal repair response, which is the main mechanism for the action of edaravone in ALS treatment.

Conclusion

Our study evaluated for the first time the role of acetovanillone with the NTD of TDP-43 and important ALS genes. Our results obtained from in silico analysis were also cross-validated by using edaravone to increase confidence in our study results and methodology. Our study opened new doors in in silico drug discovery and development by integrating genomics into the study of the effects of drugs on genes. Nonetheless, we are constantly evaluating different algorithms, and highlighting short-comings in their training sets in order to develop novel methods for the discovery of new compounds and their molecular mechanism of action, and also to shorten the time between in vitro and clinical trials analyses. Acetovanillone shows a probable interaction with the NTD, downregulates important gene targets in ALS, and possesses adequate ADMET attributes. Results obtained from this study must be used in different experimental studies in order to validate these effects on ALS.

Materials and Methods

Acquisition of Protein and Ligands

The N-terminus of TDP-43 was acquired from the PDB databank under the PDB ID: 6T4B and chosen as the receptor. The acquired protein was structurally refined and prepared with Modrefiner (Supplemental Table 1). The refined protein model was submitted to the COACH database²³ in order to predict active site residues in the protein, which might inhibit its function. The Aromadb database²⁴ was used to obtain medicinal compounds which were prepared to be utilized as ligands in this research. Edaravone was also used as a ligand, and was procured from PubChem to validate the methodology.

Screening Setup

PyRX 0.9 software was exploited for virtual screening analysis, and Autodock 4.2 was used to determine the chemical affinity between receptor and ligands. The grid-box configuration and spacing for docking interaction are presented in Supplemental Table 2. The NTD of TDP-43 was selected as a receptor in the software, and all medicinal ligands were targeted toward the active interface of the receptor prescribed by COACH (Supplemental Table 3).²³ The results from the screening analysis and edaravone were re-docked with NTD of TDP-43 with the same interaction parameter to validate the findings of this research.

ADMET Investigation

The pharmacokinetic and drug likeness of screened ligands were verified by procuring SMILES from PubChem (Supplemental Table 4). These SMILES were then used to determine the ADMET attributes of these ligands. The databases employed in this study were GUSAR,²⁵ Adver-Pred,²⁶ and ROSC-Pred²⁷ for toxic behavior prediction and pharmacokinetic and drug likeness of compounds with pKCSM²⁸ and SwissADME.²⁹

In Silico Gene Expression Studies

The screened ligand and edaravone SMILES were taken and deposited in the dialog box of DIGEP-Pred database.¹⁴ This database predicts ligand activity on various genes, and provides results in the downregulation and upregulation of genes by calculating Pa and Pi values. These values were converted into percentages, to make the data more understandable.

Supplemental Material

sj-docx-1-npx-10.1177_1934578X211030815 - Supplemental material for Exploring Aromatic Medicinal Compounds for the Treatment of Amyotrophic Lateral Sclerosis

Supplemental material, sj-docx-1-npx-10.1177_1934578X211030815 for Exploring Aromatic Medicinal Compounds for the Treatment of Amyotrophic Lateral Sclerosis by Fahad Hassan Shah and Song Ja Kim in Natural Product Communications

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Research Foundation of Korea (NRF) (grant number 2020R1I1A3069699).

ORCID iD

Song Ja Kim

Supplemental Material

Supplemental material for this article is available online.

References

Longinetti

Fang

. Epidemiology of amyotrophic lateral sclerosis: an update of recent literature. Curr Opin Neurol. 2019;32(5):771-776.

Karceski

. Understanding the different types of ALSAbout ALS. Neurology. 2020;94(8):e880-e883.

Tesauro

Bruschi

Filippini

, et al. Metal(loid)s role in the pathogenesis of amyotrophic lateral sclerosis: environmental, epidemiological, and genetic data. Environ Res. 2021;192:110292. doi:https://doi.org/10.1016/j.envres.2020.110292

Mathis

Goizet

Soulages

Vallat

J-M

Le Masson

. Genetics of amyotrophic lateral sclerosis: a review. J Neurol Sci. 2019;399:217-226. doi:https://doi.org/10.1016/j.jns.2019.02.030

Farace

Fenu

Lintas

, et al. Amyotrophic lateral sclerosis and lead: a systematic update. Neurotoxicology. 2020;81:80-88. doi:https://doi.org/10.1016/j.neuro.2020.09.003

Birsa

Bentham

Fratta

. Cytoplasmic functions of TDP-43 and FUS and their role in ALS. Seminars in Cell & Developmental Biology. 2020;99:193–201.

Prasad

Bharathi

Sivalingam

Girdhar

Patel

. Molecular mechanisms of TDP-43 misfolding and pathology in amyotrophic lateral sclerosis. Front Mol Neurosci. 2019;12:25.

Sasaguri

Chew

Y-F

, et al. The extreme N-terminus of TDP-43 mediates the cytoplasmic aggregation of TDP-43 and associated toxicity in vivo. Brain Res. 2016;1647:57-64. doi:https://doi.org/10.1016/j.brainres.2016.04.069

Salman

Shah

Idrees

, et al. Virtual screening of immunomodulatory medicinal compounds as promising anti-SARS-CoV-2 inhibitors. Future Virol. 2020;15(5):267-275. doi:10.2217/fvl-2020-0079

10.

Shah

Salman

Idrees

Akbar

. In silico study of thymohydroquinone interaction with blood–brain barrier disrupting proteins. Futur Sci OA. 2020:FSO632. doi:10.2144/fsoa-2020-0115

11.

Cruz

. Edaravone (radicava): a novel neuroprotective agent for the treatment of amyotrophic lateral sclerosis. P T. 2018;43(1):25-28.

12.

Zhifeng

Ruitao

Yang

, et al. The acute and subacute toxicity study of apocynin nitrone: a potential antioxidative agent. J Pharm Biomed Sci. 2016;06(02):76-80.

13.

Guillén

Manzanos

. Study of the components of a solid smoke flavouring preparation. Food Chem. 1996;55(3):251-257. doi:https://doi.org/10.1016/0308-8146(95)00126-3

14.

Lagunin

Ivanov

Rudik

Filimonov

Poroikov

. DIGEP-Pred: web service for in silico prediction of drug-induced gene expression profiles based on structural formula. Bioinformatics. 2013;29(16):2062-2063. doi:10.1093/bioinformatics/btt322

15.

Zhao

Beers

Hooten

, et al. Characterization of gene expression phenotype in amyotrophic lateral sclerosis monocytes. JAMA Neurol. 2017;74(6):677-685.

16.

Miyazaki

Yamashita

Morimoto

, et al. Early and selective reduction of NOP56 (Asidan) and RNA processing proteins in the motor neuron of ALS model mice. Neurol Res. 2013;35(7):744-754.

17.

Pansarasa

Bordoni

Diamanti

Sproviero

Gagliardi

Cereda

. SOD1 In amyotrophic lateral sclerosis: “ambivalent” behavior connected to the disease. Int J Mol Sci. 2018;19(5):1345.

18.

Killoy

Harlan

Pehar

Helke

Johnson

Vargas

. Decreased glutathione levels cause overt motor neuron degeneration in hSOD1WT over-expressing mice. Exp Neurol. 2018;302:129-135.

19.

Zuo

Zhou

, et al. TDP-43 aggregation induced by oxidative stress causes global mitochondrial imbalance in ALS. Nat Struct Mol Biol. 2021:28(2):132-142.

20.

Shah

STA

Salman

Idrees

Khan

. Therapeutic approaches of prominent medicinal plants for targetting Alzheimer’s disease. Zeitschrift fur Arznei-& Gewurzpflanzen. 2020;25(1):15-18.

21.

Koppula

Kim

I-S

Kumar

Kim

B-W

Choi

D-K

. The role of bioactive compounds on the promotion of neurite outgrowth. Molecules. 2012;17(6):6728-6753. doi:10.3390/molecules17066728

22.

. ‘ t Hart

Copray

Philippens

. Apocynin, a low molecular oral treatment for neurodegenerative disease. Biomed Res Int. 2014;2014:298020. doi:10.1155/2014/298020

23.

Yang

Roy

Zhang

. Protein–ligand binding site recognition using complementary binding-specific substructure comparison and sequence profile alignment. Bioinformatics. 2013;29(20):2588-2595. doi:10.1093/bioinformatics/btt447

24.

Kumar

Prakash

Tripathi

, et al. Aromadb: a database of medicinal and aromatic plant’s aroma molecules with phytochemistry and therapeutic potentials. Front Plant Sci. 2018;9:1081.

25.

Lagunin

Zakharov

Filimonov

Poroikov

. QSAR Modelling of rat acute toxicity on the basis of PASS prediction. Mol Inform. 2011;30(2−3):241-250.

26.

Ivanov

Lagunin

Rudik A

Filimonov

Poroikov V

. ADVERPred–Web service for prediction of adverse effects of drugs. J Chem Inf Model. 2018;58(1):8-11.

27.

Lagunin

Rudik

Druzhilovsky

Filimonov

Poroikov

. ROSC-Pred: web-service for rodent organ-specific carcinogenicity prediction. Bioinformatics. 2018;34(4):710-712.

28.

Pires DE

Blundell

Ascher

. pkCSM: predicting small-molecule pharmacokinetic and toxicity properties using graph-based signatures. J Med Chem. 2015;58(9):4066-4072.

29.

Daina

Michielin

Zoete

. SwissADME: a free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci Rep. 2017;7(1):42717. doi:10.1038/srep42717

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.85 MB