Abstract Objectives The aim was to explore the function of the T-cell cytokine IFNγ for mesenchymal tissue remodelling in RA and to determine whether IFNγ signalling controls the invasive potential of fibroblast-like synoviocytes (FLS). Methods To assess architectural responses, FLS were cultured in three-dimensional micromasses. FLS motility was analysed in migration and invasion assays. Signalling events relevant to cellular motility were defined by western blots. Baricitinib and small interfering RNA pools were used to suppress Janus kinase (JAK) functions. Results Histological analyses of micromasses revealed unique effects of IFNγ on FLS shape and tissue organization. This was consistent with accelerated migration upon IFNγ stimulation. Given that cell shape and cell motility are under the control of the focal adhesion kinase (FAK), we next analysed its activity. Indeed, IFNγ stimulation induced the phosphorylation of FAK-Y925, a phosphosite implicated in FAK-mediated cell migration. Small interfering RNA knockdown of JAK2, but not JAK1, substantially abrogated FAK activation by IFNγ. Correspondingly, IFNγ-induced FAK activation and invasion of FLS was abrogated by the JAK inhibitor, baricitinib. Conclusion Our study contributes insight into the synovial response to IFNγ and reveals JAK2 as a potential therapeutic target for FLS-mediated joint destruction in arthritis, especially in RA. rheumatoid arthritis, synovium, cytokines and inflammatory mediators, molecular biology, cell receptor–ligand interaction, signalling and activation Rheumatology key messages Integrated into the deleterious synovial cytokine milieu, IFNγ promotes the invasive potential of rheumatoid fibroblast-like synoviocytes. Janus kinase 2 distinctly activates focal adhesion kinase-Y925F as part of the molecular machinery that drives IFNγ-induced invasion of fibroblast-like synoviocytes. Targeted inhibition of Janus kinases by baricitinib abrogates invasion by fibroblast-like synoviocytes that is directed by IFNγ. Introduction RA is a systemic inflammatory disease that primarily affects the synovium of diarthrodial joints. As a hallmark, fibroblast-like synoviocytes (FLS) reorganize to form an aggressive cell mass that extends into and destroys the articular cartilage and bone. The aberrant behaviour of RA FLS may be driven by the cytokine milieu that is produced by infiltrating immune cells , including T cells . Cooperation between T cells and FLS may contribute to persistent destructive synovitis. So far, however, few studies have explored the effects of T cells on FLS functions; FLS that were co-cultured with T cells demonstrated increased expression of co-stimulatory molecules  and released increased amounts of tissue-degrading enzymes  as well as pro-inflammatory chemokines/cytokines when compared with FLS cultured without T cells . Direct cell-to-cell interactions between FLS and T cells, as well as T-cell-derived cytokines might be responsible for these effects. Indeed, the Th-1 cytokine IFNγ , which is expressed in rheumatoid synovitis , was shown to modulate collagen and metalloproteinase synthesis in FLS [7, 8]. These studies indicate that IFNγ affects FLS functions. Its significance for inflammatory synovial tissue remodelling and FLS invasive potential, however, remains elusive. In this study, we explore the capacity of IFNγ and its downstream signalling machinery to induce synovial tissue remodelling and to modulate the invasive behaviour of FLS. Methods Isolation and culture of FLS With approval of the Ethics Committee of the Medical University of Vienna and according to the Declaration of Helsinki, FLS from 27 patients fulfilling the ACR/EULAR classification criteria for RA  were isolated and cultured as previously described . Demographic and clinical characteristics of the patients are shown in supplementary Table S1, available at Rheumatology Online. Primary antibodies, fluorescent dyes, cytokines and inhibitors Primary antibodies for western blot were as follows: from Cell Signaling, Danvers, MA, USA: focal adhesion kinase (FAK; clone D2R2E, dilution 1:1000), p-FAK (Y576/577; 1:1000), p-FAK (Y397; D20B1; 1:1000), p-FAK (Y925; 1:1000), nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha (IκBα) (44D4; 1:1000), Janus kinase (JAK1; 6G4; 1:1000), JAK2 (D2E12; 1:1000), Signal transducer and activator of transcription 1 (STAT1) (9H2; 1:1000) and p-STAT1 (58D6; 1:1000); and from Sigma-Aldrich, St. Louis, MO, USA: actin (1:1000). Primary antibodies were used as suggested by the manufacturer. Dyes for immunofluorescence staining were as follows: CellTracker Green CMFDA (Thermo Fisher, Waltham, MA, USA; 1:400), CellTracker Deep Red (Thermo Fisher; 1:500) and CellTracker Red CMTPX (Thermo Fisher; 1:400). Dyes were used as suggested by the manufacturer. Cytokines, from R&D Systems, Minneapolis, MN, USA, were as follows: IFNα, IFNβ, IFNγ and TNF. The concentration used for interferons was 100 U/ml, and for TNF 10 ng/ml. The inhibitor, baricitinib, was purchased from Selleckchem, Houston, TX, USA. Small interfering RNA experiments The FLS were transfected with 50 nM of small interfering RNA (siRNA) pools (JAK1: ON-TARGETplus SMARTpool L-003145-00-0005; JAK2: ON-TARGETplus SMARTpool L-003146-00-0005; ThermoScientific, Waltham, MA, USA) using Lipofectamine RNAiMAX (Invitrogen, Carlsbad, CA, USA) . Synovial micromass cultures Micromass organ cultures were constructed as described by Kiener et al. . After variable times of culture (7–28 days), the spheres were fixed with paraformaldehyde and embedded in paraffin. Paraffin-embedded sections were stained with Haematoxylin and Eosin. Pictures were taken with an Axioskop 2 microscope (Zeiss Oberkochen, Baden-Wurttemberg, Germany) equipped with a digital camera (Olympus, Shinjuku, Tokyo, Japan). For confocal live cell imaging, FLS were dyed with CellTracker Green (CMFDA), Red (CMTMR) and Deep Red. Imaging was done with a Leica TSC SP5 microscope (Leica, Wetzlar, Hesse, Germany). Images were processed with Imaris Bitplane software (Bitplane, Belfast, United Kingdom). Cell migration assay Polycarbonate membranes (24-well insert, pore size 8 μm; Sigma Aldrich, St. Louis, Missouri, USA) were coated with fibronectin (Fig. 1C; Life Technologies, Carlsbad, Carlifornia, USA) or fetal bovine serum (supplementary Fig. S1, available at Rheumatology Online; Hyclone by GE Healtcare (Little Chalfont, United Kingdom)). FLS were plated in the top chamber. FLS that migrated to the underside of the membrane were fixed with methanol, stained with Vectashield (Vector Laboratories, Burlingame, California, USA) and counted using a fluorescence microscope (Zeiss). Fig. 1 View largeDownload slide Effects of IFNγ on multicellular organization, cell shape and cell motility (A) Fibroblast-like synoviocytes (FLS36) cultured in extracellular matrix (ECM) were left untreated or stimulated with IFNγ or TNF for 7 days. Representative pictures of eight independent experiments with FLS from different donors are shown. (B) FLS from one donor (FLS27) were divided and labelled with different CellTracker dyes (Green, Red or Deep Red). FLS were then cultured in ECM as aforementioned. Pictures were taken on day 4. They are representative of six independent experiments with FLS from different donors. (C) FLS migration was assessed using modified Boyden chambers. FLS were left untreated or stimulated with IFNγ for 5 h. FLS from six donors were analysed by Student’s paired t-test. Fig. 1 View largeDownload slide Effects of IFNγ on multicellular organization, cell shape and cell motility (A) Fibroblast-like synoviocytes (FLS36) cultured in extracellular matrix (ECM) were left untreated or stimulated with IFNγ or TNF for 7 days. Representative pictures of eight independent experiments with FLS from different donors are shown. (B) FLS from one donor (FLS27) were divided and labelled with different CellTracker dyes (Green, Red or Deep Red). FLS were then cultured in ECM as aforementioned. Pictures were taken on day 4. They are representative of six independent experiments with FLS from different donors. (C) FLS migration was assessed using modified Boyden chambers. FLS were left untreated or stimulated with IFNγ for 5 h. FLS from six donors were analysed by Student’s paired t-test. Western blot FLS were lysed in radioimmunoprecipitation assay buffer supplemented with phosphatase and protease inhibitors (Roche, Basel, Switzerland). Proteins were separated by electrophoresis, followed by electrotransfer onto nitrocellulose membranes. After blocking, membranes were incubated with primary and horseradish peroxidase-conjugated secondary antibodies (Cell Signaling Technology, Danvers, MA, USA). Specific bands were detected with the ECL detection kit (Pierce, Waltham, Massachusetts, USA). Reblots were performed using ReBlot Plus Strong Stripping Solution (Millipore, Burlington, Massachusetts, USA). ImageJ (US National Institutes of Health, Bethesda, MD, USA) was used to quantify gel bands. Invasion assay The previously described matrix-associated transepithelial resistance invasion assay was used to assess FLS invasive capacity . Statistical analysis For parametrically distributed data, Student’s paired t-tests were used for comparing paired samples. If data were not normally distributed, Wilcoxon matched pairs non-parametric tests were applied instead. Student’s paired t-tests and Wilcoxon matched pairs non-parametric tests were performed using GraphPad Prism software (GraphPad, San Diego, CA, USA). To evaluate the invasive capacity of FLS over time depending on the type of stimulation (IFNγ, baricitinib, IFNγ + baricitinib; Fig. 2E), the generalized estimated equation analyses incorporated in the SPSS package 24 (IBM) were used. Fig. 2 View largeDownload slide Janus kinase inhibition results in diminished IFNγ-induced focal adhesion kinase-Y925 phorphorylation and invasion of fibroblast-like synoviocytes (A and B) Immunoblots of interferon- or TNFα-stimulated FLS. Representative immunoblots for six independent experiments with six different FLS cell lines are shown. (C) FLS transfected with non-targeting, JAK1 or JAK2 siRNA pools were left untreated or stimulated with IFNγ for 30 min. Blots are representative of six independent experiments with six different FLS cell lines. (D) DMSO- or baricitinib-pretreated FLS were stimulated with IFNγ for 30 min. Immunoblots are representative of six independent experiments. (E) Matrix-associated transepithelial resistance invasion assay. FLS were treated with DMSO, baricitinib (1000 nM), IFNγ or IFNγ+baricitinib. Mean (s.e.m.) of six independent experiments carried out in technical triplicates. Generalized estimating equation analyses. Bar, baricitinib; DMSO, dimethyl sulphoxide; FAK, focal adhesion kinase; FLS, fibroblast-like synoviocytes; JAK, janus kinase; siRNA, small interfering RNA. Fig. 2 View largeDownload slide Janus kinase inhibition results in diminished IFNγ-induced focal adhesion kinase-Y925 phorphorylation and invasion of fibroblast-like synoviocytes (A and B) Immunoblots of interferon- or TNFα-stimulated FLS. Representative immunoblots for six independent experiments with six different FLS cell lines are shown. (C) FLS transfected with non-targeting, JAK1 or JAK2 siRNA pools were left untreated or stimulated with IFNγ for 30 min. Blots are representative of six independent experiments with six different FLS cell lines. (D) DMSO- or baricitinib-pretreated FLS were stimulated with IFNγ for 30 min. Immunoblots are representative of six independent experiments. (E) Matrix-associated transepithelial resistance invasion assay. FLS were treated with DMSO, baricitinib (1000 nM), IFNγ or IFNγ+baricitinib. Mean (s.e.m.) of six independent experiments carried out in technical triplicates. Generalized estimating equation analyses. Bar, baricitinib; DMSO, dimethyl sulphoxide; FAK, focal adhesion kinase; FLS, fibroblast-like synoviocytes; JAK, janus kinase; siRNA, small interfering RNA. Results In RA, FLS reorganize and give rise to pathological pannus tissue that destroys the articular cartilage and bone. To explore whether IFNγ drives tissue remodelling in arthritis, we used a three-dimensional in vitro model (Fig. 1A and B) that was shown faithfully to recapitulate many in vivo functions of the synovial membrane . Consistent with previous data, lining formation with compacted cells at the tissue interface was observed for control micromasses (Fig. 1A). In contrast, IFNγ stimulation resulted in attenuated lining formation, with less tightly packed cells oriented in parallel to the surface of the sphere (Fig. 1A and supplementary Fig. S2A, available at Rheumatology Online). Similar to the control, the sublining FLS were scattered, with wide extracellular space between individual cells, but demonstrated a more elongated cellular phenotype, with long cellular extensions (Fig. 1B). In contrast, TNF induced pronounced aggregation of FLS in both the lining and the sublining area (Fig. 1A and B; supplementary Fig. S2, available at Rheumatology Online) . This indicates that IFNγ affects synovial tissue remodelling in a different manner from TNF. Cellular organization in tissues depends upon tight control of cellular motility. Recognizing distinct structural effects of IFNγ in synovial micromass cultures, we probed whether IFNγ regulates FLS motility. Indeed, migration assays revealed an increased migratory activity for IFNγ-stimulated compared with unstimulated FLS (Fig. 1C; supplementary Fig. S1, available at Rheumatology Online). Given that FAK controls cell migration by regulating focal adhesion turnover , we next determined FAK phosphorylation upon cytokine stimulation. Although phosphorylation of Y397 and Y576/Y577 was not affected, IFNγ increased FAK-Y925 phosphorylation (Fig. 2A; supplementary Fig. S3A, available at Rheumatology Online). IFNα/IFNβ and TNF stimulation resulted in the expected activation of the JAK-STAT1 or the nuclear factor-κB pathway, as shown by the phosphorylation of STAT1 or the degradation of IκBα, respectively, but neither type I IFNs (IFNα and IFNβ) nor TNF increased the phosphorylation of FAK (Fig. 2B; supplementary Fig. S3B, available at Rheumatology Online), indicating that Y925 phosphorylation of FAK is a distinct characteristic of IFNγ, but not of type I IFNs or TNF. IFNγ mediates its effects via JAKs . To address whether JAK1, JAK2 or both are involved in FAK activation, we silenced JAK expression with specific siRNA pools. Intriguingly, knockdown of JAK2, but not JAK1, consistently abrogated the IFNγ-induced phosphorylation of FAK-Y925 (Fig. 2C; supplementary Fig. S3C, available at Rheumatology Online). In line with these observations, baricitinib, a JAK1/JAK2 inhibitor that has recently been approved for the treatment of RA patients , decreased IFNγ-induced FAK-Y925 phosphorylation in a dose-dependent manner (Fig. 2D; supplementary Fig. S3D, available at Rheumatology Online). Based on our observations, we reasoned that baricitinib might limit the invasive potential of IFNγ-stimulated FLS. Using a recently described invasion assay , we observed an increased invasive capacity for IFNγ-stimulated FLS when compared with unstimulated FLS (control vs IFNγ, P < 0.0001; Fig. 2E). Moreover, inhibition of JAKs by baricitinib abrogated the IFNγ-induced invasive behaviour (IFNγ vs IFNγ + baricitinib, P = 0.005; Fig. 2E). Discussion Research into the mechanisms defining FLS activity in RA suggests that soluble mediators, such as cytokines secreted by leucocytes, play a key role. Notably, TNFα and IL1β are known to enhance FLS aggressiveness as well as pro-inflammatory cytokine and matrix metalloproteinase production . Herein, we report on a T-cell cytokine, namely IFNγ, that has hitherto not been directly linked to FLS migration and invasion. Using an in vitro model of the synovium , we observed IFNγ-induced synovial architectural changes that were strikingly different from that induced by TNF. These changes were accompanied by an increased migratory and invasive FLS phenotype and pronounced FAK activity. FAK is a tyrosine kinase that promotes normal as well as tumour cell migration and invasion. Recent evidence indicates that FAK is a determining factor for the aggressive behaviour of RA FLS. When compared with healthy synovial tissues, increased FAK activity was detected in rheumatoid synovitis . Moreover, epigenetic studies identified the FAK pathway as a hotspot for epigenetic alterations in RA FLS . The mechanisms that induce FAK activation include integrin-mediated FAK autophosphorylation at Y397 and Proto-oncogene tyrosine-protein kinase Src-mediated phosphorylation of the FAK kinase domain activation loop (Y576/Y577) . Confirming a previous observation , we found constitutively activated Y397 and Y576/Y577 in cultured RA FLS. The phosphorylation of both Y397 and Y576/Y577 was not affected by IFNγ or the other cytokines tested. Importantly, IFNγ stimulation distinctly promoted FAK phosphorylation at the less well-characterized phosphosite, Y925. The importance of Y925 for FAK-mediated cell migration and invasion was recently demonstrated by non-phosphorylatable and phosphomimetic mutants. FAK knockout mouse embryonic fibroblasts that expressed the non-phosphorylatable Y925F-FAK showed decreased focal adhesion turnover and reduced cell migration . In line with a reported association between JAK2 and FAK after growth hormone stimulation , our silencing experiments reveal that FAK activation by IFNγ is mediated by JAK2. This suggests that emerging RA therapeutics that target JAK2 would inhibit the herein reported IFNγ-JAK2-FAK signalling cascade and may, thus, prevent IFNγ-induced cartilage invasion by FLS. Indeed, we demonstrate that baricitinib inhibits the IFNγ-induced invasive capacity of FLS. As IFNγ is expressed in rheumatoid synovitis , this pathway might contribute to destructive events beyond those mediated by the injurious cytokines TNF and IL-6. In conclusion, here we contribute evidence that IFNγ promotes the migratory and invasive behaviour of FLS via the activation of JAK2 and FAK. This study provides a molecular basis for the structural efficacy of novel targeted RA therapeutics, such as baricitinib, and suggests that JAK inhibitors not only target immune cells but also the stromal compartment of RA. Funding: This research has received support from the ‘Medical Scientific Fund of the Mayor of the City of Vienna’. Disclosure statement: J.S.S. provides expert advice to AbbVie, Gilead, Lilly and Pfizer and has received grants for his institution from AbbVie. All other authors have declared no conflicts of interest. Supplementary data Supplementary data are available at Rheumatology online. References 1 Noss EH, Brenner MB. The role and therapeutic implications of fibroblast-like synoviocytes in inflammation and cartilage erosion in rheumatoid arthritis. Immunol Rev 2008; 223: 252– 70. http://dx.doi.org/10.1111/j.1600-065X.2008.00648.x Google Scholar CrossRef Search ADS PubMed 2 Steiner G, Tohidast-Akrad M, Witzmann G et al. Cytokine production by synovial T cells in rheumatoid arthritis. Rheumatology 1999; 38: 202– 13. http://dx.doi.org/10.1093/rheumatology/38.3.202 Google Scholar CrossRef Search ADS PubMed 3 Corrigall VM, Solau-Gervais E, Panayi GS. Lack of CD80 expression by fibroblast-like synoviocytes leading to anergy in T lymphocytes. Arthritis Rheum 2000; 43: 1606– 15. http://dx.doi.org/10.1002/1529-0131(200007)43:7>1606::AID-ANR26<3.0.CO;2-O Google Scholar CrossRef Search ADS PubMed 4 Burger D, Rezzonico R, Li JM et al. Imbalance between interstitial collagenase and tissue inhibitor of metalloproteinases 1 in synoviocytes and fibroblasts upon direct contact with stimulated T lymphocytes: involvement of membrane-associated cytokines. Arthritis Rheum 1998; 41: 1748– 59. http://dx.doi.org/10.1002/1529-0131(199810)41:10>1748::AID-ART7<3.0.CO;2-3 Google Scholar CrossRef Search ADS PubMed 5 Yamamura Y, Gupta R, Morita Y et al. Effector function of resting T cells: activation of synovial fibroblasts. J Immunol 2001; 166: 2270– 5. http://dx.doi.org/10.4049/jimmunol.166.4.2270 Google Scholar CrossRef Search ADS PubMed 6 Schroder K, Hertzog PJ, Ravasi T, Hume DA. Interferon-γ: an overview of signals, mechanisms and functions. J Leukoc Biol 2004; 75: 163– 89. Google Scholar CrossRef Search ADS PubMed 7 Stephenson ML, Krane SM, Amento EP, McCroskery PA, Byrne M. Immune interferon inhibits collagen synthesis by rheumatoid synovial cells associated with decreased levels of the procollagen mRNAs. FEBS Lett 1985; 180: 43– 50. Google Scholar CrossRef Search ADS PubMed 8 Alvaro-Gracia JM, Zvaifler NJ, Firestein GS. Cytokines in chronic inflammatory arthritis. V. Mutual antagonism between interferon-gamma and tumor necrosis factor-alpha on HLA-DR expression, proliferation, collagenase production, and granulocyte macrophage colony-stimulating factor production by rheumatoid arthritis synoviocytes. J Clin Invest 1990; 86: 1790– 8. Google Scholar CrossRef Search ADS PubMed 9 Aletaha D, Neogi T, Silman AJ et al. 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis Rheum 2010; 62: 2569– 81. http://dx.doi.org/10.1002/art.27584 Google Scholar CrossRef Search ADS PubMed 10 Kiener HP, Watts GF, Cui Y et al. Synovial fibroblasts self-direct multicellular lining architecture and synthetic function in three-dimensional organ culture. Arthritis Rheum 2010; 62: 742– 52. http://dx.doi.org/10.1002/art.27285 Google Scholar CrossRef Search ADS PubMed 11 Rosner M, Siegel N, Fuchs C et al. Efficient siRNA-mediated prolonged gene silencing in human amniotic fluid stem cells. Nat Protoc 2010; 5: 1081– 95. http://dx.doi.org/10.1038/nprot.2010.74 Google Scholar CrossRef Search ADS PubMed 12 Wunrau C, Schnaeker EM, Freyth K et al. Establishment of a matrix-associated transepithelial resistance invasion assay to precisely measure the invasive potential of synovial fibroblasts. Arthritis Rheum 2009; 60: 2606– 11. http://dx.doi.org/10.1002/art.24782 Google Scholar CrossRef Search ADS PubMed 13 Mitra SK, Hanson DA, Schlaepfer DD. Focal adhesion kinase: in command and control of cell motility. Nature reviews. Mol Cell Biol 2005; 6: 56– 68. 14 Genovese MC, Kremer J, Zamani O et al. Baricitinib in patients with refractory rheumatoid arthritis. New Engl J Med 2016; 374: 1243– 52. http://dx.doi.org/10.1056/NEJMoa1507247 Google Scholar CrossRef Search ADS PubMed 15 Shahrara S, Castro-Rueda HP, Haines GK, Koch AE. Differential expression of the FAK family kinases in rheumatoid arthritis and osteoarthritis synovial tissues. Arthritis Res Ther 2007; 9: R112. Google Scholar CrossRef Search ADS PubMed 16 Nakano K, Whitaker JW, Boyle DL, Wang W, Firestein GS. DNA methylome signature in rheumatoid arthritis. Ann Rheum Dis 2013; 72: 110– 7. Google Scholar CrossRef Search ADS PubMed 17 Stanford SM, Svensson MN, Sacchetti C et al. Receptor protein tyrosine phosphatase α-mediated enhancement of rheumatoid synovial fibroblast signaling and promotion of arthritis in mice. Arthritis Rheumatol 2016; 68: 359– 69. Google Scholar CrossRef Search ADS PubMed 18 Deramaudt TB, Dujardin D, Hamadi A et al. FAK phosphorylation at Tyr-925 regulates cross-talk between focal adhesion turnover and cell protrusion. Mol Biol Cell 2011; 22: 964– 75. http://dx.doi.org/10.1091/mbc.E10-08-0725 Google Scholar CrossRef Search ADS PubMed 19 Zhu T, Goh EL, Lobie PE. Growth hormone stimulates the tyrosine phosphorylation and association of p125 focal adhesion kinase (FAK) with JAK2. FAK is not required for STAT-mediated transcription. J Biol Chem 1998; 273: 10682– 9. Google Scholar CrossRef Search ADS PubMed © The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: email@example.com
Rheumatology – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.
All for just $49/month
Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.
Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.
It’s easy to organize your research with our built-in tools.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera