| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TECHNOLOGY DEVELOPMENT |
aThe Arnold and Mabel Beckman Center for Transposon Research, Gene Therapy Program, Institute of Human Genetics, Department of Genetics, Cell Biology and Development and
bDepartment of Medicine, Stem Cell Institute University of Minnesota, Minneapolis, Minnesota, USA
Key Words. Human embryonic stem cells • Sleeping Beauty • Nonviral integration • Stable expression • Bioluminescence imaging
Correspondence: Dan S. Kaufman, M.D., Ph.D., Stem Cell Institute and Department of Medicine, University of Minnesota, Translational Research Facility, 2001 6th St SE, Mail Code 2873, Minneapolis, Minnesota 55455, USA. Telephone: 612-624-0922; Fax: 612-624-2436; e-mail: kaufm020{at}umn.edu
Received on January 10, 2007;
accepted for publication on July 25, 2007.
First published online in STEM CELLS EXPRESS August 2, 2007.
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
Disclosure of potential conflicts of interest is found at the end of this article.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
Lentivirus vectors pseudotyped with vesicular stomatitis virus G protein have been shown to be effective for mediating transgene expression in 20%–80% of human ES cells after transduction at high multiplicity without loss of pluripotency [9–14]. For stable genetic modification of human ES cells using lentiviral vectors, a single expression cassette is often used in combination with selection for the encoded gene product [9, 10]. Persistent transgene expression has been improved by including a scaffold attachment region in the vector design [11]. Still, the most effective vectors incorporate a selectable marker regulated by the same (or by a separate) promoter that regulates transcription of a second transgene of interest [12, 13]. Recently, lentiviral vectors pseudotyped by envelope proteins from either the gibbon ape leukemia virus or the RD114 feline endogenous virus have been shown to effectively transduce human ES cells without transducing cocultured feeder cells (mouse embryonic fibroblasts [MEFs]) [14]. Even so, lentiviral vectors display an integration profile that favors transcriptionally active genes [15]. This integration profile can potentially interrupt or activate gene expression and may complicate biological interpretation or clinical translation of these transduced cells.
Genetic manipulation of human ES cells using nonviral strategies would afford significant benefits. Although conditions have been established for stable gene transfer following plasmid-based delivery into mouse ES cells, only limited success has been achieved using nonviral vectors in human ES cells [16–20]. Protocols designed to achieve stable expression have relied upon random integration of plasmid followed by drug selection for introduction of new sequences into human ES cells. This process is inefficient, as it relies on random double-strand break-mediated recombination. Thus, one alternative for increased frequency of nonviral integration is to add components of an integrating DNA transposon into the plasmid design.
Sleeping Beauty (SB) is a synthetic member of the Tc/1-mariner superfamily of transposons, reconstructed from dormant elements harbored in the salmonid fish genome [21]. The SB transposon system can mediate stable, high-level gene expression in human cells without the use of a virus. SB consists of two components: the transposon and transposase. The transposase (in this case, SB) mediates integration of the transposon, a mobile element encoding a cargo sequence flanked on both sides by inverted terminal repeats that harbor binding sites for the catalytic enzyme (SB). Stable expression results when SB inserts gene sequences into vertebrate chromosomes at a TA target dinucleotide through a cut-and-paste mechanism. This system has been used to engineer a variety of vertebrate cell types, including lung and liver, one-cell mouse embryos, mouse ES cells, mouse and human fibroblasts [22, 23], and primary human peripheral blood leukocytes [24], where stable gene transfer efficiencies ranged from approximately 1% to approximately 10% of the cells transfected.
In this study, we demonstrated that the SB transposon system provides an efficient method to achieve stable genetic modification of human ES cells. Furthermore, these transgenic cells retained the ability to differentiate into hematopoietic cells in vitro and into teratomas in vivo while maintaining effective transgene expression. We conclude that the SB transposon system provides an effective approach for genetic manipulation of human ES cells.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
promoter/intron sequence (CLP) (Invivogen). The construct confers resistance to Zeo in both mammalian and bacterial cells, providing a means for genomic recovery of transposon integrants. pKT2/CaL encodes the firefly luc sequence (Promega, Madison, WI, http://www.promega.com) under control of a chimeric cytomegalovirus enhancer/chicken β-actin promoter/chicken β-globin intron sequence (CAGS) [26]. pKT2/CLP-Luc (Fig. 2A) was generated by replacing the CAGS promoter in pKT2/CaL with the CLP regulatory sequence. The plasmid template for in vitro transcription of Sleeping Beauty transposase messenger RNA (rU-SB11-U) (Fig. 2A) and transposase expression plasmid (human phosphoglycerate kinase promoter-transposase [PGK-SB11]) (Figs. 1A, 2A) have been described previously [27]. Plasmid DNA was prepared using an Endofree Maxi Prep Kit (Qiagen, Valencia, CA, http://www.qiagen.com) to remove bacterial lipopolysaccharide.
|
|
Flow Cytometry
Colonies of human ES cells were harvested from MEF cocultures and prepared for flow cytometric analysis. Single-cell suspensions of undifferentiated human ES cells were stained with phycoerythrin (PE)-conjugated isotype control immunoglobin or specific antibodies against human SSEA-4 (R&D Systems). Live cells were identified and gated by exclusion of 7-aminoactinomycin D. All analyses were performed using a FACSCalibur (Becton, Dickinson and Company, San Jose, CA, http://www.bd.com) and FlowJo analysis software (Tristar Inc., San Carlos, CA).
Recovery and Sequence Analysis of Transposon Integration Sites
Zeocin-resistant colonies generated by nucleofection of human ES cells with pKT2/ZOG and transposase-encoding DNA were pooled and genomic DNA was isolated in accordance with conditions established by Gentra Systems (Minneapolis, www.gentra.com). One microgram of total genomic DNA was digested with BamHI, EcoRI, or XbaI, and recovery of transposon:chromosome junctions was performed as previously described [26]. Briefly, digested genomic DNA was ligated under dilute conditions (500 µl) with T4 DNA ligase (New England Biolabs, Ipswich, MA, http://www.neb.com), precipitated with 100% isopropanol and washed with 70% ethanol before resuspending in 10 µl of sterile H2O. Two microliters of reconstituted DNA was electroporated into DH10B electrocompetent Escherichia coli (Promega), allowing bacterial cells to recover in SOC medium before plating on Luria/Bertani agar containing 50 µg/ml zeocin (Invivogen). Zeocin-resistant colonies were counterselected on agar supplemented with 50 µg/ml kanamycin. Plasmid DNA was isolated from ZeoR/KanS colonies and sequenced using T2 inverted terminal repeat (ITR)-specific primers; ITR right (5'-ccactgggaatgtgatgaaag-3'); ITR left (5'-gacttgtgtcatgcacaaagtag-3').
Luciferase Assays and Bioluminescence Imaging
For stable transfection, human ES cells were nucleofected with a source of transposase and luc-encoding transposons. Luc activity of transfected human ES cells was determined by bioluminescence imaging or by enzymatic analysis of total cell lysates. For bioluminescence imaging, transfected human ES cells maintained in culture on mitomycin C-treated MEFs were analyzed for luc activity by applying D-luciferin substrate (Xenogen, Alameda, CA, http://www.xenogen.com), 12.5 µl of 25 mg/ml solution in phosphate-buffered saline (PBS), to each well, and luc activity was detected as emitted light 5 minutes later by exposure to an intensified, charge-coupled device camera (Series 100; Xenogen) for an additional 1 minute. For enzymatic assays, cells were harvested and lysed in 100–200 µl of 1x cell lysis buffer (Promega), and 20 µl of the lysate was assayed for luc enzyme activity on a Lumat LB 9507 series tube luminometer (Berthold Technologies, Oak Ridge, TN, http://www.bertholdtech.com). In each case, luc activity was normalized for light acquisition time or total number of cells.
To monitor luc expression during teratoma formation, 5–6 x 106 undifferentiated human ES cells engineered for stable expression of luc with SB were intramuscularly injected into the hind leg of immunodeficient SCID/Beige mice (Charles River Laboratories, Wilmington, MA, http://www.criver.com). At indicated times after ES cell injection, animals were injected i.p. with luciferin substrate (120 µl of 25.0 mg/ml in PBS). At 15–20 minutes postinjection, anesthetized mice were imaged for 3 minutes using the Xenogen imaging system (Series 100) as described [28]. Transgenic human ES cells differentiated into hematopoietic cells and cultured in methylcellulose were analyzed for luc activity by applying luciferin substrate (25 µl of 25 mg/ml solution of D-luciferin in 400 µl of DMEM/F12 supplemented with 5% fetal bovine serum) to each plate, and luc activity was detected as emitted light 5 minutes later with an exposure time of 5 minutes. For in vitro and in vivo experiments, raw imaging values were recorded as photons of light emitted per second per cm2.
Differentiation Assays
Differentiation of human ES cells to hematopoietic cells was carried out as previously described [1, 29]. Briefly, ES cells engineered with SB transposons encoding luc were cultured for 9–20 days on S17 mouse bone marrow stromal cells or by embryoid body formation on low-adherence plates to induce hematopoietic differentiation. Colony-forming assays were performed as previously described [29].
Immunohistochemical Assays
To test the undifferentiated state of human ES cells engineered for stable luc expression, cells that expressed luc for at least 3 months of continuous culture were seeded onto chamber slides (Gibco-BRL, Carlsbad, CA, http://www.gibcobrl.com) and allowed to expand for 3–4 days. ES colonies were fixed with 4% paraformaldehyde and permeabilized with PBS-Tween solution. Fish skin gelatin (0.4%) was added and incubated for 30 minutes to block nonspecific staining. The ES colonies were incubated with anti-luc (1:100) (Novus Biologicals, Inc., Littleton, CO, http://www.novusbio.com) at 4°C overnight and visualized with fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG (1:200; Jackson Immunoresearch Laboratories, West Grove, PA, http://www.jacksonimmuno.com). Goat serum (20%) was used for another 30 minutes of blocking before incubation with PE-conjugated anti-SSEA-4 (R&D Systems). ProLong Gold antifade reagent with 4,6-diamidino-2-phenylindole (Invitrogen) was used for nuclear staining. Unmodified ES cells and isotype controls were included to identify nonspecific staining or antibody cross-reactivity. Expression of SSEA-4 and luc was examined by indirect immunofluorescence analysis, whereas GFP was evaluated by direct fluorescence imaging in the FITC channel using an Axiovert 200m fluorescence microscope (Carl Zeiss, Jena, Germany, http://www.zeiss.com).
To examine luc expression in teratomas, tumors were excised and immediately embedded in Tissue-Tek OCT Compound (Fisher Scientific International, Hanover Park, IL, http://www.fisherscientific.com) on dry ice and stored at –80°C. Frozen sections (10 µm) were prepared using a cryostat microtome (IMEB Inc., San Marcos, CA, http://www.imebinc.com) and fixed in cold acetone for 15 minutes. After washing with PBS and blocking with 20% donkey serum for 3–5 hours, luc staining was carried out as described above. Cy3-conjugated donkey anti-mouse IgG(H+L) was used as secondary antibody (Jackson Immunoresearch Laboratories). The tissues were examined using a Carl Zeiss Axiovert 200m fluorescence microscope.
Southern Blot Analysis
Clonal human ES cell colonies were prepared with 2 x 106 cells treated with 0.25% trypsin (Mediatech) to produce a single cell suspension. Trypsin was inactivated with 10% serum containing medium and passed through a 70-µm filter. Cells were washed with Dulbecco's phosphate-buffered saline, resuspended in ES cell medium, and then added to one well of a six-well plate with MEFs. Seven days later, individual colonies were hand-picked and transferred to individual wells of four-well plates. Clonal colonies were then grown up and genomic DNA was isolated using the Puregene cell kit (Qiagen). Polymerase chain reaction primers were designed from the luc gene of pKT2/mCaL to make an 813-bp probe. DNA from each clone was digested with BamHI to cleave the vector twice, once inside and once outside the transposon. Digests were run on an 0.8% agarose gel and transferred to a nitrocellulose membrane. The DNA was hybridized overnight with 32P-labeled probe.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
To determine the extent to which sustained GFP expression was due to transposition rather than random integration, 1–2 x 106 human ES cells selected for zeocin resistance were extracted for genomic DNA, which was then digested with BamHI, EcoRI, or XbaI (enzymes that do not cut the sequence encoded by the transposon or the plasmid-encoded kanamycin resistance gene). A plasmid rescue procedure was then used to recover transposon integrants along with both chromosomal flanking sequences. Using this technique, we found that 98% of bacterial colonies (294 of 300) containing the ZeoR gene also lacked the donor plasmid resistance gene (i.e., they were KanS), suggesting a low frequency (2%) of random integration or maintenance of the transposon-encoded plasmid as an episome. Direct sequencing of several plasmids isolated from ZeoR/KanS recombinants revealed integration of transposon sequences into human chromosomes and duplication of a target TA dinucleotide (Fig. 1D). We screened 10 of these transposon:chromosome junction sequences for homologies by BLAST search of the Ensembl database. These searches confirmed that stable gene expression observed in human ES cells was due to transposition.
Generation of Human ES Cells That Stably Express Luciferase
To determine the stability of expression from integrated SB cargo sequences, we used firefly luciferase (luc) for sensitive bioluminescence imaging [30] of gene expression following differentiation of human ES cells both in vitro and in vivo. The luc transposons were designed to be transcriptionally regulated by promoters containing (pKT2/CaL) or depleted of CpG dinucleotides (pKT2/CLP-Luc) and were delivered either unsupplemented, supplemented with plasmid DNA (pPGK-SB11) as a source of SB, or supplemented with in vitro-transcribed SB transposase-encoding messenger RNA (rU-SB11-U) (Fig. 2A). Both sources of transposase have been shown to effectively mediate transposon insertion in cultured human fibroblasts [27]. We tested the use of mRNA because integration and/or continued expression of SB transposase-encoding sequences could be problematic if this led to remobilization of integrated transposons and subsequent genotoxicity.
Human ES cells were nucleofected with luc-encoding transposons (10 µg) and either no SB transposase, or rU-SB11-U or pPGK-SB11 (5 µg) as sources of SB transposase. Luc expression in undifferentiated cells was measured by bioluminescence imaging (Fig. 2). Only human ES cells cotransfected with both transposon and transposase-encoding sequences exhibited luc expression after 1 week (Fig. 2B). The luc-encoding construct did not contain a selectable marker, so to enrich for luc-expressing cells needed for subsequent in vivo testing, after 4 weeks of continuous culture, luc+ colonies were manually isolated and further expanded. After 80 days, stable expression of luc was observed in human ES cells cotransfected with either a DNA or RNA transposase-encoding sequence (Fig. 2B). Stable expression of luc from both the pKT2/CaL and pKT2/CLP-Luc vectors was demonstrated for more than 5 months. We further evaluated these cell populations for coexpression of luc plus either SSEA-4, TRA-1–60, or Oct4 by immunohistochemistry (IHC) (Fig. 2C–2E). Colocalized staining (yellow) was predominantly visible in undifferentiated colonies. Furthermore, flow cytometric analysis of these cells also clearly shows uniform expression of SSEA-4, TRA-1–60, and Oct4 to demonstrate the IHC results are representative of the full population (Fig. 2C–2E). Similar results were obtained by flow cytometric analysis of human ES cells engineered for stable expression of GFP while maintaining SSEA-4 expression (Fig. 1B). Stable genetic modification mediated by the SB transposon system was thus not detrimental to self-renewal of undifferentiated human ES cells. In addition, Southern blot analysis was conducted on individual clones to quantify the number of transposon insertions after SB-mediated luc gene transfer into human ES cells (Fig. 2F). Here, we demonstrate that colonies derived from individual human ES cells had anywhere from 1 insertion (Fig. 2F, lane 2) to 10 or more insertions (Fig. 2F, lanes 13–15). Also, for some clonal human ES cells that we isolated, luc expression was not observed. As expected, transgene insertion on the Southern blot was not observed in these clones (Fig. 2F, lanes 6–8, 11, and 12). Notably, the reproducible number of transgene insertions demonstrated in several clonal populations suggests that a relatively limited number of human ES cells initially had stable transgene integration.
Stable Expression After In Vivo Differentiation of Human ES Cells
Human ES cells are able to differentiate into all three germ layers (endoderm, mesoderm, and ectoderm). To determine the stability of gene expression after differentiation in vivo, human ES cells engineered to express luc under transcriptional control of the CpG-free promoter were injected into the hind leg muscles of three immunodeficient mice (SCID/Beige), and the engrafted teratomas were monitored for expression of luc by in vivo bioluminescence imaging at the indicated time points (Fig. 3A) over a period of 3–4 months. Images obtained for a representative animal demonstrate the typical kinetics of luc activity during the course of teratoma formation (Fig. 3B). On the day that the cells were injected, all mice exhibited localized luc activity detected as emitted light in the area of the leg muscle. This activity subsided over the first 3–4 weeks prior to engraftment. Teratoma formation was evident by the increasing level of luc expression over time. Histological analysis of teratomas 12–14 weeks after injection demonstrated tissues representing all three germ layers (Fig. 3C). Differentiated tissues included neuron (ectoderm), cartilage (mesoderm), and glandular epithelium (endoderm). To demonstrate that the luc signal originated from teratoma tissue, we assessed luc protein within the tumor by immunofluorescence, staining with isotype control (Fig. 3D) or a luc-specific antibody (Fig. 3E). This histological analysis demonstrated that tissues of all three germ layers were luc+ (Fig. 3).
|
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
In general, human ES cells have proven less amenable to genetic engineering than mouse ES cells. Lentivirus vectors have been most successfully used for achieving stable gene expression in human ES cells during continuous passage and following differentiation in vitro [10–14] and in vivo [9, 10, 12]. However, recent studies indicate that lentivirus-based vectors preferentially integrate into actively transcribed genes, including proto-oncogenes and signaling genes [15]. This suggests a higher predisposition toward uncontrolled cellular functions and tumorigenesis that could be problematic for developmental studies or for clinical application. In contrast, SB transposons exhibit a random pattern of integration [32–34], showing less tendency to integrate into transcribed genes or transcriptional regulatory regions compared with all randomly integrating vector systems studied to date [35]. Although only a limited number of integration sites were sequenced here, as expected, all integration coincided with duplication of a TA dinucleotide, and none of the recovered sequences were found in exons (Fig. 1D). This suggests that SB-mediated gene transfer will be less prone to insertional mutagenic events. Our results also demonstrate the effectiveness of RNA as an alternative to DNA as a source of transposase, reducing the likelihood of transposon-encoding sequences becoming integrated and thus preserving genome integrity.
Although plasmids alone have been used for genetic modification of human ES cells, this method typically yields only one stable clone for every 1–5 x 105 cells treated [16, 18–20], as the process relies predominantly on random double-strand break-mediated recombination. In fact, the efficiency of random integration in cultured human fibroblasts is frequently at least 2 orders of magnitude lower than the levels achieved by SB-mediated transposition [23]. We showed that transgene expression (GFP or luc) was maintained only in cells that were also provided with a source of transposase, indicating that random integration of the transposon encoding plasmid alone was inefficient. We found that approximately 5% of cells exhibit transgene expression at 7 days post-transfection using a GFP transgene plus SB transposase (Fig. 1B). Moreover, this efficacy increased to more than 80% stable GFP+ cells when drug selection was applied to facilitate expansion of engineered cells, compared with only 0.5% GFP+ cells without drug selection and without SB (Fig. 1B; data not shown).
Previous testing of SB-mediated transposition in ES cells has been limited to studies in the mouse system. SB activity was demonstrated by excision [36] and by excision coupled with reintegration [37] of a transposon located at a specific chromosomal locus when SB transposase was provided in trans. Methylation of transposon DNA enhanced excision from a specific chromosomal locus upon codelivery of both the transposon- and transposase-encoding sequences into mouse ES cells using a gene trap-type transposon lacking an internal expression cassette [36]. Here we demonstrate, in human ES cells, use of the SB transposon system to mediate stable gene expression when a transposon expression vector and transposase vector are delivered in trans, suggesting that this approach will be very effective to achieve stable gene expression in other ES cell systems (such as mouse ES cells) as well.
Our results imply that any nonviral integrating system that functions by a cut-and-paste mechanism may potentially be used to engineer human ES cells with efficiencies that are improved over random integration. These methods would include, but are not limited to, Xenopus Frog Prince [38], cabbage looper moth piggyBac [39], and Medaka fish Tol2 [40, 41], all of which have been shown to integrate into cultured human cells or mouse ES cells. In particular, piggyBac and Tol2 may be less restrictive than SB with respect to cargo size and relative transposase expression level [42], although these transposons have yet to be tested in human ES cells. In addition, any number of the site-directed phage integrases [43–45] that have been shown to mediate integration into mammalian genomes could also be used to engineer human ES cells with the potential for much greater integration site specificity and flexibility in cargo size. The availability and activity of nonviral vectors such as SB for the genetic modification of human ES cells thus offers broad opportunities for the study of human development and the generation of new therapeutic approaches for tissue regeneration and cell transplantation therapy.
|
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. S. Woll, B. Grzywacz, X. Tian, R. K. Marcus, D. A. Knorr, M. R. Verneris, and D. S. Kaufman Human embryonic stem cells differentiate into a homogeneous population of natural killer cells with potent in vivo antitumor activity Blood, June 11, 2009; 113(24): 6094 - 6101. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. I. Orban, A. Apati, A. Nemeth, N. Varga, V. Krizsik, A. Schamberger, K. Szebenyi, Z. Erdei, G. Varady, E. Karaszi, et al. Applying a "Double-Feature" Promoter to Identify Cardiomyocytes Differentiated from Human Embryonic Stem Cells Following Transposon-Based Gene Delivery Stem Cells, May 1, 2009; 27(5): 1077 - 1087. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Irion, M.C. Nostro, S.J. Kattman, and G.M. Keller Directed Differentiation of Pluripotent Stem Cells: From Developmental Biology to Therapeutic Applications Cold Spring Harb Symp Quant Biol, March 27, 2009; (2009) sqb.2008.73.065v1. [Abstract] [PDF]
|
|
|
|
|
|
C. Unger, E. Karner, A. Treschow, B. Stellan, U. Felldin, H. Concha, M. Wendel, O. Hovatta, A. Aints, L. Ahrlund-Richter, et al. Lentiviral-Mediated HoxB4 Expression in Human Embryonic Stem Cells Initiates Early Hematopoiesis in a Dose-Dependent Manner but Does Not Promote Myeloid Differentiation Stem Cells, October 1, 2008; 26(10): 2455 - 2466. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| STEM CELLS | THE ONCOLOGIST | CME | ALPHAMED PRESS JOURNALS |
