Establishment of MDCK/FX Cell for Efficient Replication of Influenza Viruses

AUTHORS

Hadi Fazel 1 , Shahla Shahsavandi 2 , Homayoon Mahravani 2 , Farhad Rezaei 1 , Jila Yavarian 1 , Nazanin Zahra Shafiiei-Jandaghi 1 , Talat Mokhtari Azad 1 , *

1 Department of Virology, School of Public Health, Tehran University of Medical Sciences, Tehran, Iran

2 Razi Vaccine and Serum Research Institute, Agricultural Research Education and Extension Organization, Karaj, Iran

How to Cite: Fazel H, Shahsavandi S, Mahravani H, Rezaei F, Yavarian J, et al. Establishment of MDCK/FX Cell for Efficient Replication of Influenza Viruses, Jundishapur J Microbiol. 2017 ; 10(5):e44891. doi: 10.5812/jjm.44891.

ARTICLE INFORMATION

Jundishapur Journal of Microbiology: 10 (5); e44891
Published Online: May 7, 2017
Article Type: Research Article
Received: December 19, 2016
Revised: March 5, 2017
Accepted: April 5, 2017
Crossmark

Crossmark

CHEKING

READ FULL TEXT
Abstract

Background: Replication of influenza virus to high titer is a prerequisite for successful cell-based vaccine production. Entry of virus into the cell depends on the cleavage of the hemagglutinin precursor (HA0) protein mediated by trypsin.

Objectives: The aim of the present study was to apply a technique to establish MDCK/FX manipulated cell, which may provide a new platform for developing influenza vaccine based on the cell culture approaches.

Methods: Chicken embryo FX expressed into the pCDNA3.1 vector was transfected into the MDCK cell line. The longevity of the generated cell and the viable cell density were evaluated for 17 passages prior to virus inoculation. Then, the ability of MDCK/FX cell for efficient replication of H9N2 influenza virus was evaluated by viral titration and quantitative RT-PCR.

Results: RT-PCR data revealed that FX was stably expressed in the cell after the subsequent passages without any change in the rate of culture’s confluency. Growth kinetic of H9N2 virus analysis demonstrated that MDCK/FX cell supported high-titer growth of the virus in which the viral titer is comparable to the virus grown in MDCK cells supplemented with TPCK-trypsin. Quantification of influenza infectious particles in the cell culture revealed the equivalents viral RNA copies and viral titers.

Conclusions: The results indicated a potential application for the MDCK/FX in influenza virus replication procedure and related studies.

Keywords

Influenza Virus Hemagglutinin MDCK Cell Factor X

Copyright © 2017, Ahvaz Jundishapur University of Medical Sciences. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/) which permits copy and redistribute the material just in noncommercial usages, provided the original work is properly cited.

1. Background

Vaccination is effective in preventing the spread of influenza viruses infections in human and avian populations. Production of egg-based influenza vaccines is still largely implemented in many countries. Despite the considerable advantages of these vaccines, a limited number of mammalian cell lines are approved for more rapid and larger scale cell culture-based vaccine production. As a good alternative substrate, Madin-Darby canine kidney (MDCK) cell line efficiently supports the growth of influenza viruses. However, the cell can be more flexible to supply vaccine production for use in possible influenza pandemic (1, 2). Either in embryonated egg or cell culture, fusion and entry of influenza virus is mediated by specific binding of hemagglutinin (HA) to sialic acid receptors and activation of HA protease (3, 4). The HA0 precursor proteolytically cleaved by intracellular host proteases makes up 2 structurally distinct globular head HA1 and stalk structure HA2 domains (5, 6). Infectivity of influenza viruses depends on forming fusogenic HA molecule and quality of HA0 precursor cleavage. The host proteases responsible for posttranslational cleavage of fusion-incompetent HA0 are expressed in respiratory tract or a limited number of tissue types.

Human and low pathogenic avian viruses have single arginine at HA cleavage site, which is activated by trypsin-like proteases secreted from epithelial cells of the lung or intestinal tissues (7-9). The hydrolysis of HA0 precursor is a necessary step for the viruses to be infective. The ability of the cells to internalize influenza virus is essential for the improvement of viral replication (10, 11). In laboratory trials, adding trypsin or TPCK supports the replication of influenza viruses in a variety of cell cultures. As a protease, the enzyme contributes to viral entry via cleaving HA0, but do not affect the ability of the cells to internalize the virus. However, the viruses grow efficiently on embryonated eggs in the absence of supplemental trypsin (10, 12). It has been found that Factor 10 (FX) is responsible for the cleavage of the fusion proteins of Sendai virus, Newcastle disease virus, and influenza virus (13-15). Considering the high efficiency of FX compared to trypsin on a molar basis and its specificity in cleavage reactions, the genetically manipulated BHK-21/FX cell was established (16). Then, the potential of the cell on virus replication kinetics was evaluated during influenza virus infection process. The results of the data analysis suggested that BHK-21/FX cell provides high-titer replication of low pathogenic influenza virus.

2. Objectives

The present study aimed at applying a technique to establish MDCK/FX manipulated cell, which may provide a new platform for developing influenza vaccine based on the cell culture approaches.

3. Methods

3.1. Transfection of FX Construct onto the MDCK Cell

The cDNA clone of the FX gene was constructed in pcDNA3.1 plasmid (Invitrogen, Carlsbad, CA) according to the previous study (16). In brief, the sequences of chicken FX were deposited in the NCBI GenBank database under accession numbers NM-205022, XM-015277827, and D00844 and were downloaded and aligned. The specific primers were designed based on the open reading frame of FX located in peptidase S1 domain (241-473 nt). MDCK cells were transfected with recombinant vector pcDNA3.1/FX using Lipofectamine 2000 (Life Technologies), and the transfected cells were selected and cultivated in the presence of 10% fetal bovine serum (FBS; Sigma Aldrich, Germany) and 800 mg/mL Geneticin (G418, Invitrogen). After 2 weeks, the survived cells were pooled, passaged 3 times in 1.5 mg of G418/mL, and frozen in aliquots. To verify the expression of FX in MDCK cell, the G418-resistant colonies were subjected to RT-PCR (iNtRON Biotechnology, Korea) using a specific primer set for amplification of FX. The manipulated cell was named MDCK/FX.

3.2. Growth Kinetic Screening of MDCK/FX Cell

The quality of MDCK/FX cell was evaluated by estimating the longevity of cell and the viable cell density in every 17 passages prior to virus inoculation. The cells with concentration of 1 × 106 cells/mL were grown in DMEM (Invitrogen), supplemented with 10% FBS and antibiotic solution, at 37°C with 5% CO2. The rate of culture confluency and the number of live cells were estimated.

3.3. Virus Infectivity Assay

At the 15th passage, the monolayers of MDCK/FX with 80% to 90% confluency were inoculated with H9N2 virus A/chicken/Iran SS8/2011 (17) at MOI of 0.01 PFU/cell. The cultures were incubated up to 72 hours post infection (hpi) at 37°C and controlled for cytopathic effect (CPE) appearance. Four different sets of MDCK culture flasks were inoculated in the presence and absence of supplemental 1 µg/mL TPCK-trypsin (Sigma Aldrich). The trial continued for 7 subsequent passages. The infectivity of influenza virus in each passage was determined using plaque assay and quantitative RT-PCR.

The mRNA level of viral matrix (M) gene from MDCK/FX inoculated cells was analyzed by one-step quantitative RT-PCR (SuperScript TMIII Platinum®Taq Mix, Invitrogen) at a defined interval from 8 hpi up to 48 hpi. The relative expressions were normalized using the housekeeping gene β-actin as internal control. Reactions were repeated 3 times for each sample, and standard deviations (SD) were calculated. The primer sequences (18) used in this study were as follow: MF: 5’-CTCATGGAATGGCTAAAGACA-3’ and MR: 5’-CGATCAADAATCCACAATATC-3’, β-actinF: 5’-TGCTGTGTTCCCATCTATCG-3’ and β-actinR: 5’-TTGGTGACAATACCGTGTTCA-3’.

3.4. Evaluation of Mutations in the Virus

The possible incidence of mutation in viral genome was determined by comparative sequence analysis. The full-length mucleotide sequences of HA genes of influenza virus at 3rd and 7th passages in MDCK/FX were determined and compared to the virus seed (17).

3.5. Statistical Analysis

Data were expressed as mean ± SD. Statistical correlation of data was checked for significance by ANOVA and Student’s t test.

4. Results

4.1. MDCK/FX Cell Establishment and Screening

The plasmid encoded FX was transfected to MDCK cells, and the cells were cultivated under antibiotic selection procedure. After 2 weeks, the survived antibiotic-resistant colonies were selected and subjected to RT-PCR to verify the expression of FX. Specific band at 233 bp lengths has confirmed the presence of FX gene in the cell at each passage.

4.2. Growth Kinetic Characterization of MDCK/FX Cell

No changes were observed in cell-cell adhesion and cell spreading behavior of MDCK/FX cell following transfection of FX construct. At each passage, the cell grew well and the mean cell density reached to 2 × 105 cells/cm2 after 72 hours of growth. Cell viability remained above 97% during the first day and declined to less than 80% by the 3rd day of incubation, indicating that expression of FX was not toxic to the MDCK cell. The MDCK/FX cell was not different from MDCK cell morphologically.

4.3. Potential Infectivity of Influenza Virus in MDCK/FX

The H9N2 virus replicated well in MDCK/FX and the marked CPEs including giant cell formation were observed at 24 hpi. The massive detaching from the culture flask were found in infected cell at 72 hpi compared to MDCK cell in the presence of supplemental trypsin (Figure 1). No virus growth was found in the absence of trypsin.

Cytopathogenecity of MDCK Cell (Left) and MDCK/FX (Right) Cell to Influenza Virus Infection (MOI 0.01) at 48 Hours After Inoculation (200 × Magnification)
Figure 1. Cytopathogenecity of MDCK Cell (Left) and MDCK/FX (Right) Cell to Influenza Virus Infection (MOI 0.01) at 48 Hours After Inoculation (200 × Magnification)

The virus titers peaked higher and earlier in MDCK/FX cell compared to MDCK plus TPCK-trypsin cell; however, no statistical differences were observed (P < 0.01). This clearly indicates that the cell permits the production of infectious virus particles (Figure 2A). The virus showed a modest delay in replication in MDCK cell with supplemental trypsin in which the mean peak titers of the virus in the cell reached after 48 hpi. The replication rate of the virus was monitored by detecting mRNA of M gene in quantitative RT-PCR up to 48 hpi (Figure 2B). In MDCK/FX cell, the numbers of M-expressed cells increased with time elapses due to protease activation of HA cleavage site which supports virus entry and replication. The virus infected-MDCK cells have exhibited the same panel in the presence of trypsin at 8 and 16 hours post infection, but significantly more M gene mRNA copies were found in MDCK/FX cell compared with MDCK cell by 24 hpi (P < 0.01).

Potential Infectivity of Influenza Virus in MDCK/FX Cell Compared to the MDCK Cells Supplemented With Trypsin. A) Virus Titration Assayed in the Subsequent Passages and B) the Relative Expressions of Viral M Gene Normalized using
Figure 2. Potential Infectivity of Influenza Virus in MDCK/FX Cell Compared to the MDCK Cells Supplemented With Trypsin. A) Virus Titration Assayed in the Subsequent Passages and B) the Relative Expressions of Viral M Gene Normalized using

4.4. Evaluation of Mutations in the Virus

Nucleotide sequences of HA gene of influenza virus replicated in MDCK/FX were determined at 3rd and 7th passages and the seed virus gene sequence was compared. Mutation in the nucleotide sequences of the viral gene that resulted in the change in amino acid codon was not detected. Sequence analyses of HA glycoprotein showed conservation of residues H183, L190, L226, Q227 and G228 in the receptor binding pocket and RSSR motif in the cleavage site.

5. Discussion

The early detection and isolation of influenza A viruses are very important to design effective control programs and prevent strategy against the infection. Conventionally, the viruses replicate in embryonated eggs and avian or mammalian-origin cells following bind to N-glycan structures of sialic acid (3, 18). It has been suggested that mammalian cell lines vary in the amount of viral particles that attach to the cells. In the present study, we successfully transferred a construct of FX gene to MDCK cell, then, we evaluated the growth kinetics of H9N2 virus on the manipulated MDCK/FX in the absence of TPCK-trypsin at 7 subsequent passages. Virus titration demonstrated that the cell supported high-titer growth of the virus in which the viral titer was comparable to the virus grown in MDCK cells supplemented with TPCK-trypsin. The virus showed a modest delay in replication in MDCK cell. Full-length amplification and sequencing patterns of the HA gene demonstrated that the H9N2 HA gene at 7 passages remained similar to the parent virus. These data indicate the ability of MDCK/FX as a permissive cell to replicate the low pathogenic influenza virus.

It has been reported that some influenza clinical isolates have low growth properties in MDCK cells line (19, 20). Thus, we modeled the embryonated egg cellular factors involved in the viral HA cleavage procedure for use in cell culture. Analysis of the N-glycan in embryonated chicken egg revealed the presence of both α 2, 3 (chorioallantoic) and α 2, 6 (amniotic) linkages, known to be important for efficient virus entry and internalization. During the life cycle of influenza virus, internalization of viral particles into target cells is the primary step for replication and production of high viral titers. Along with the specific receptor distribution of embryonated egg, the FX host cell ubiquitous protease has a role in promoting virus replication. This protease, as a determinant of viral tropisms, cleaves the major glycoprotein and contributes to the fusion activity. Enhancing the membrane fusion leads to increasing the binding rates and virion production.

Catalytic chain (FXa) of the virus activating protease is crucial for the substrate specificity (15), as it cleaves the virus precursor glycoproteins directly and facilitates the virus progeny production and spread (13). The duration of reproductive cycle of influenza virus and release of progeny was estimated to take 8 to 10 hours in MDCK cell and A549 cell (21-23). The progeny virus release time was monitored by estimation of viral M protein expression levels in MDCK/FX cells. Based on the quantitative RT-PCR data, the progeny viruses release in MDCK/FX at 8 hpi was similar to the MDCK cells supplemented with TPCK-trypsin.

Despite similarities of virus replication dynamic, difference in CPE appearance was observed between both cells. CPE appears in a short time in MDCK/FX cell compared to MDCK cells that can result from the effective cleavage of HA protein. The HA cleavage site of low pathogenic influenza virus is activated only in the respiratory and intestinal tract where trypsin-like enzymes are secreted (24). Addition of exogenous trypsin is required for the efficient replication of the viruses in vitro unless serine proteases are expressed in these cells. Analysis of replication kinetics of influenza viruses in different cell lines showed that the viral replication and faster progression of infection in LMH, HepG2, and A549 cells are higher than MDCK, HeLa and Vero cell lines because they are able to activate HA monobasic cleavage site and promote virus entry in the absence of trypsin (1, 11, 21).

It has been suggested that treatment of the cells with trypsin may cause an increase in cell death and a decrease in the virus titer. Moreover, FX is 20 times more effective than trypsin on a molar basis, and 8 times less volume of the factor than trypsin is required for the same trial. Trypsin causes large nonspecific loss of viruses’ infectivity under incubation conditions similar to FX. Thus, the factor is much more effective and specific in cleavage activation than in trypsin (13, 25). The potential of FX to cleave HA and promote replication of influenza viruses is confirmed by our study in which maximum viral titers were reached one day earlier in MDCK/FX with comparative MDCK cells.The impact of FX on H9N2 virus replication kinetics upon engineered expression in BHK-21 cell line has been evaluated (16).

In addition to activation of HA protease processing, type and density of receptors play an important role in virus replication (3, 5). Clinical isolates of human influenza viruses bind strongly to 6-linked sialic acid, whereas avian viruses bind to 3-linkage. The distribution of sialic acid on cell surfaces and affinity of the HA to the receptors are determinants of host tropism. Stable transfection with the SIAT1 gene increased the concentration of the sialyl α2, 6-Gal receptor on the surface of MDCK cells about twofold and decreased the same amounts of α2, 3-Gal than parent MDCK cells (26).

It seems that a low concentration of sialyl α2, 3-Gal receptors in conventional cell lines may lead to non-efficient replication of avian influenza isolates in SIAT-MDCK cell (27). Unlike this cell, amniotic cells of chicken embryonated eggs contain N-glycans with terminal mannose and sulfate residues capable of binding to human and avian influenza viruses. The presence of sulfated glycans may be a reason why human influenza viruses are efficiently replicated in chicken embryonated egg. Once bound, the host FX protease involves in HA priming and proteolytic cleavage. The enzymatically active FXa was detectable only in the allantoic and amniotic fluids (14, 15).

5.1. Conclusions

Using N-glycan structure and FX protease activity of chicken embryonated egg, we introduced a highly permissive engineered cell for influenza virus propagation. The MDCK/FX cell has 2 main features: uniform cell morphology and stable expression of FX. With the establishment of the cell, we will focus on analysis of N-glycans responsible for virus binding, evaluation of susceptibility of the cell to different virus infectious doses, and cultivation of more influenza viruses isolated from human and avian populations in future studies.

Footnote

References

  • 1.

    Feng SZ, Jiao PR, Qi WB, Fan HY, Liao M. Development and strategies of cell-culture technology for influenza vaccine. Appl Microbiol Biotechnol. 2011; 89(4) : 893 -902 [DOI][PubMed]

  • 2.

    Hussain AI, Cordeiro M, Sevilla E, Liu J. Comparison of egg and high yielding MDCK cell-derived live attenuated influenza virus for commercial production of trivalent influenza vaccine: in vitro cell susceptibility and influenza virus replication kinetics in permissive and semi-permissive cells. Vaccine. 2010; 28(22) : 3848 -55 [DOI][PubMed]

  • 3.

    Skehel JJ, Wiley DC. Receptor binding and membrane fusion in virus entry: the influenza hemagglutinin. Annu Rev Biochem. 2000; 69 : 531 -69 [DOI][PubMed]

  • 4.

    Bertram S, Glowacka I, Steffen I, Kuhl A, Pohlmann S. Novel insights into proteolytic cleavage of influenza virus hemagglutinin. Rev Med Virol. 2010; 20(5) : 298 -310 [DOI][PubMed]

  • 5.

    Bosch FX, Garten W, Klenk HD, Rott R. Proteolytic cleavage of influenza virus hemagglutinins: primary structure of the connecting peptide between HA1 and HA2 determines proteolytic cleavability and pathogenicity of Avian influenza viruses. Virology. 1981; 113(2) : 725 -35 [PubMed]

  • 6.

    Chen J, Lee KH, Steinhauer DA, Stevens DJ, Skehel JJ, Wiley DC. Structure of the hemagglutinin precursor cleavage site, a determinant of influenza pathogenicity and the origin of the labile conformation. Cell. 1998; 95(3) : 409 -17 [PubMed]

  • 7.

    Kido H, Okumura Y, Takahashi E, Pan HY, Wang S, Chida J, et al. Host envelope glycoprotein processing proteases are indispensable for entry into human cells by seasonal and highly pathogenic avian influenza viruses. J Mol Genet Med. 2008; 3(1) : 167 -75 [PubMed]

  • 8.

    Bottcher-Friebertshauser E, Freuer C, Sielaff F, Schmidt S, Eickmann M, Uhlendorff J, et al. Cleavage of influenza virus hemagglutinin by airway proteases TMPRSS2 and HAT differs in subcellular localization and susceptibility to protease inhibitors. J Virol. 2010; 84(11) : 5605 -14 [DOI][PubMed]

  • 9.

    Okumura Y, Takahashi E, Yano M, Ohuchi M, Daidoji T, Nakaya T, et al. Novel type II transmembrane serine proteases, MSPL and TMPRSS13, Proteolytically activate membrane fusion activity of the hemagglutinin of highly pathogenic avian influenza viruses and induce their multicycle replication. J Virol. 2010; 84(10) : 5089 -96 [DOI][PubMed]

  • 10.

    Shahsavandi S, Ebrahimi MM, Mohammadi A, Zarrin Lebas N. Impact of chicken-origin cells on adaptation of a low pathogenic influenza virus. Cytotechnology. 2013; 65(3) : 419 -24 [DOI][PubMed]

  • 11.

    Bottcher E, Matrosovich T, Beyerle M, Klenk HD, Garten W, Matrosovich M. Proteolytic activation of influenza viruses by serine proteases TMPRSS2 and HAT from human airway epithelium. J Virol. 2006; 80(19) : 9896 -8 [DOI][PubMed]

  • 12.

    Klenk HD, Rott R, Orlich M, Blodorn J. Activation of influenza A viruses by trypsin treatment. Virology. 1975; 68(2) : 426 -39 [PubMed]

  • 13.

    Gotoh B, Ogasawara T, Toyoda T, Inocencio NM, Hamaguchi M, Nagai Y. An endoprotease homologous to the blood clotting factor X as a determinant of viral tropism in chick embryo. EMBO J. 1990; 9(12) : 4189 -95 [PubMed]

  • 14.

    Gotoh B, Yamauchi F, Ogasawara T, Nagai Y. Isolation of factor Xa from chick embryo as the amniotic endoprotease responsible for paramyxovirus activation. FEBS Letters. 1992; 296(3) : 274 -8 [DOI]

  • 15.

    Ogasawara T, Gotoh B, Suzuki H, Asaka J, Shimokata K, Rott R, et al. Expression of factor X and its significance for the determination of paramyxovirus tropism in the chick embryo. EMBO J. 1992; 11(2) : 467 -72 [PubMed]

  • 16.

    Shahsavandi S, Ebrahimi MM, Masoudi S, Izadi H. Expression of factor X in BHK-21 cells promotes low pathogenic influenza viruses replication. Adv Virol. 2015; [DOI]

  • 17.

    Fazel H, Shahsavandi S, Masoudi S, Ebrahimi MM, Taghizadeh M. Evolutionary characterization of non-structural gene of H9N2 influenza viruses isolated from Asia during 2008-2012. Comp Clin Pathol. 2014; 23(3) : 523 -8 [DOI]

  • 18.

    Tombari W, Nsiri J, Larbi I, Guerin JL, Ghram A. Genetic evolution of low pathogenecity H9N2 avian influenza viruses in Tunisia: acquisition of new mutations. Virol J. 2011; 8 : 467 [DOI][PubMed]

  • 19.

    Asaoka N, Tanaka Y, Sakai T, Fujii Y, Ohuchi R, Ohuchi M. Low growth ability of recent influenza clinical isolates in MDCK cells is due to their low receptor binding affinities. Microbes Infect. 2006; 8(2) : 511 -9 [DOI][PubMed]

  • 20.

    Ilyushina NA, Ikizler MR, Kawaoka Y, Rudenko LG, Treanor JJ, Subbarao K, et al. Comparative study of influenza virus replication in MDCK cells and in primary cells derived from adenoids and airway epithelium. J Virol. 2012; 86(21) : 11725 -34 [DOI][PubMed]

  • 21.

    Venkateswarlu D, Perera L, Darden T, Pedersen LG. Structure and dynamics of zymogen human blood coagulation factor X. Biophys J. 2002; 82(3) : 1190 -206 [DOI][PubMed]

  • 22.

    Shahsavandi S, Ebrahimi MM, Sadeghi K, Mosavi SZ, Mohammadi A. Dose- and time-dependent apoptosis induced by avian H9N2 influenza virus in human cells. BioMed Res Inter. 2013; 2013 : 1 -7 [DOI]

  • 23.

    Frensing T, Kupke SY, Bachmann M, Fritzsche S, Gallo-Ramirez LE, Reichl U. Influenza virus intracellular replication dynamics, release kinetics, and particle morphology during propagation in MDCK cells. Appl Microbiol Biotechnol. 2016; 100(16) : 7181 -92 [DOI][PubMed]

  • 24.

    Gabriel G, Abram M, Keiner B, Wagner R, Klenk HD, Stech J. Differential polymerase activity in avian and mammalian cells determines host range of influenza virus. J Virol. 2007; 81(17) : 9601 -4 [DOI][PubMed]

  • 25.

    ZarrinLebas N, Shahsavandi S, Mohammadi A, Ebrahimi MM, Bakhshesh M. Replication efficiency of influenza A virus H9N2: a comparative analysis between different origin cell types. Jundishapur J Microbiol. 2013; 6(9)[DOI]

  • 26.

    Matrosovich M, Matrosovich T, Carr J, Roberts NA, Klenk HD. Overexpression of the alpha-2,6-sialyltransferase in MDCK cells increases influenza virus sensitivity to neuraminidase inhibitors. J Virol. 2003; 77(15) : 8418 -25 [PubMed]

  • 27.

    Oh DY, Barr IG, Mosse JA, Laurie KL. MDCK-SIAT1 cells show improved isolation rates for recent human influenza viruses compared to conventional MDCK cells. J Clin Microbiol. 2008; 46(7) : 2189 -94 [DOI][PubMed]

  • COMMENTS

    LEAVE A COMMENT HERE: