Author response: A protein phosphatase network controls the temporal and spatial dynamics of differentiation commitment in human epidermis Article Swipe

MAKE IMPACT

Ajay Mishra , Bénédicte Oulès , Angela Oliveira Pisco , Tony Ly , Kifayathullah Liakath‐Ali , Gernot Walko , Priyalakshmi Viswanathan , Matthieu Tihy , Jagdeesh Nijjher , Sara-Jane Dunn , Angus I. Lamond , Fiona M. Watt · YOU? · · 2017 · Open Access · · DOI: https://doi.org/10.7554/elife.27356.036

Article Figures and data Abstract Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Epidermal homeostasis depends on a balance between stem cell renewal and terminal differentiation. The transition between the two cell states, termed commitment, is poorly understood. Here, we characterise commitment by integrating transcriptomic and proteomic data from disaggregated primary human keratinocytes held in suspension to induce differentiation. Cell detachment induces several protein phosphatases, five of which - DUSP6, PPTC7, PTPN1, PTPN13 and PPP3CA – promote differentiation by negatively regulating ERK MAPK and positively regulating AP1 transcription factors. Conversely, DUSP10 expression antagonises commitment. The phosphatases form a dynamic network of transient positive and negative interactions that change over time, with DUSP6 predominating at commitment. Boolean network modelling identifies a mandatory switch between two stable states (stem and differentiated) via an unstable (committed) state. Phosphatase expression is also spatially regulated in vivo and in vitro. We conclude that an auto-regulatory phosphatase network maintains epidermal homeostasis by controlling the onset and duration of commitment. https://doi.org/10.7554/eLife.27356.001 Introduction Commitment is a transient state during which a cell becomes restricted to a particular differentiated fate. Under physiological conditions, commitment is typically irreversible and involves selecting one differentiation pathway at the expense of others (Nimmo et al., 2015). While commitment is a well-defined concept in developmental biology, it is still poorly understood in the context of adult tissues (Simons and Clevers, 2011; Semrau and van Oudenaarden, 2015; Nimmo et al., 2015). This is because end-point analysis fails to capture dynamic changes in cell state, and rapid cell state transitions can depend on post-translational events, such as protein phosphorylation and dephosphorylation (Avraham and Yarden, 2011). We set out to examine commitment in human interfollicular epidermis, which is a multi-layered epithelium formed by keratinocytes and comprises the outer covering of the skin (Watt, 2014). The stem cell compartment lies in the basal layer, attached to an underlying basement membrane. Cells that leave the basal layer undergo a process of terminal differentiation as they move through the suprabasal layers. In the final stage of terminal differentiation, the cell nucleus and cytoplasmic organelles are lost and cells assemble an insoluble barrier, called the cornified envelope, which is formed of transglutaminase cross-linked proteins and lipids (Watt, 2014). We have previously shown that keratinocytes can commit to terminal differentiation at any phase of the cell cycle, and upon commitment they are refractory to extracellular matrix-mediated inhibition of differentiation (Adams and Watt, 1989). Although there are currently no markers of epidermal commitment, we have previously used suspension-induced differentiation of disaggregated human keratinocytes in methylcellulose-containing medium (Adams and Watt, 1989) to define its timing. Here, we use this simple experimental model to identify markers of commitment and discover why it is a transient state. Results Dynamic expression of protein phosphatases during suspension-induced differentiation Since keratinocytes increase in size as they differentiate (Adams and Watt, 1989), we enriched for undifferentiated cells by filtration prior to placing them in suspension for up to 12 hr (Figure 1a). By determining when cells recovered from suspension could no longer resume clonal growth on replating (Figure 1b; Figure 1—figure supplement 1a), we confirmed that there is a marked drop in colony forming ability between 4 and 8 hr. This correlates with an increase in the proportion of cells expressing the terminal differentiation markers involucrin (IVL) and transglutaminase 1 (TGM1) (Figure 1c,d; Figure 1—figure supplement 1b) and downregulation of genes that are expressed in the basal layer of the epidermis, including integrin α6 (ITGα6) and TP63 (Figure 1d) and the stem cell markers DLL1 and Lrig1 (Tan et al., 2013; Figure 1—figure supplement 1c). As expected, Ki67 expression was also reduced in suspension, reflecting the drop in proliferation upon differentiation (Figure 1—figure supplement 1c). Figure 1 with 1 supplement see all Download asset Open asset Genomic and proteomic analysis identifies protein dephosphorylation events that correlate with commitment. (a) Schematic of experimental design. (b) Colony formation by cells harvested from suspension at different times. Representative dishes are shown together with % colony formation (n = 2 independent experiments, n = 3 dishes per condition per experiment; p-values calculated by Tukey's multiple comparison test). (c) Cells isolated from suspension at different time points were labelled with anti-involucrin (IVL) antibody (green) and DAPI as nuclear counterstain (blue). IVL-positive cells were counted using ImageJ (n = 3 independent cultures; more than 300 cells counted per condition. p-value calculated by two-tailed t-test). Scale bars: 50 μm (d) RT-qPCR quantification of ITGα6, TP63, IVL and TGM1 mRNA levels (relative to 18 s expression) (n = 3 independent cultures). (e) t-SNE plot of genome-wide transcript expression by keratinocytes placed in suspension for different times. The t-SNE algorithm takes a set of points in a high-dimensional space and finds a faithful representation of those points in a lower-dimensional space, typically the 2D plane. (f) Heatmap representing hierarchical clustering of differentially expressed proteins (p<0.05). (g–i) Dot plots correlating expression of significantly differentially expressed peptides (p<0.05) that change twofold relative to 0 hr in at least one of the time points, with their corresponding differentially expressed transcripts. Pearson correlations (r) are indicated. (j) Histogram of normalised SILAC ratios corresponding to high confidence phosphorylation sites that differ between 0 and 4 hr. (k) Scatter plot correlating log2 normalised SILAC ratios for total protein changes (y-axis) with log2 phospho-peptide ratios (x-axis) between 0 and 4 hr. (b, c) *p<0.05; **p<0.01; ns = non-significant). https://doi.org/10.7554/eLife.27356.002 To define commitment at the molecular level, we next collected keratinocytes after 4, 8 and 12 hr in suspension and performed genome-wide transcriptomics, using Illumina-based microarrays, and proteome-wide peptide analysis, by SILAC-Mass-Spectrometry (MS) (Figure 1e,f; Figure 1—figure supplement 1d–g; Supplementary file 1, 2). Keratinocytes collected immediately after trypsinisation served as the 0 hr control. When comparing the starting cell population (0 hr) with cells suspended for 4, 8 or 12 hr, t-SNE plots of differentially expressed genes (Figure 1e) and unsupervised hierarchical clustering of genes and proteins (Figure 1f; Figure 1—figure supplement 1d) indicated that the 4 hr samples clustered separately from the 8 and 12 hr samples. GO term enrichment analysis of differentially expressed transcripts or proteins showed enrichment of terms associated with epidermal differentiation at 8 and 12 hr (Figure 1—figure supplement 1e,g) (Sen et al., 2010; Mulder et al., 2012), which is consistent with the drop in clonogenicity seen at these time points. When we compared the significantly differentially expressed proteins (p-value<0.05) that changed ≥2 fold at one or more time points with their corresponding transcripts (Figure 1g–i), there was a moderately positive correlation at 8 and 12 hr (Pearson correlations of 0.51 and 0.68, respectively), consistent with the correlation between bulk mRNA and protein levels seen in previous studies of mammalian cells (Schwanhäusser et al., 2011; Ly et al., 2014). However, at 4 hr transcripts and proteins were only weakly correlated (Pearson correlation 0.19, Figure 1g). The poor correlation between protein and transcript levels at 4 hr suggested a potential role for post-transcriptional mechanisms in regulating commitment. To investigate this, we performed unbiased SILAC-MS-based phospho-proteomic analysis. SILAC-labelled peptides isolated from cells at 0, 4 and 8 hr time points were enriched for phosphopeptides using HILIC pre-fractionation and titanium dioxide affinity chromatography (Supplementary file 3). Over 3500 high confidence phosphorylation sites were identified with an Andromeda search score >= 30 and quantified at both the 4 and 8 hr time points. At 4 hr approximately two thirds of the changes involved dephosphorylation (Figure 1j), but these dephosphorylation events could not be attributed to decreases in protein abundance, as shown by the discordance between total protein abundance and changes in protein phosphorylation (Figure 1k). We next identified protein phosphatases that were differentially expressed between 0 and 4 hr. We intersected the proteomic and genomic datasets, revealing 47 differentially expressed phosphatases, 22 of which were upregulated (Figure 2a). Interrogation of published datasets revealed that these phosphatases were also dynamically expressed during calcium-induced stratification of human keratinocytes (Hopkin et al., 2012) and differentiation of reconstituted human epidermis (Lopez-Pajares et al., 2015) (Figure 2—figure supplement 1a,b). Figure 2 with 4 supplements see all Download asset Open asset Functional screen identifies candidate phosphatases that regulate commitment. (a) Heatmap showing differential expression at 4, 8 and 12 hr relative to 0 hr of phosphatases that are upregulated at 4 hr in the microarray dataset. (b) Effect of knocking down the 22 phosphatases identified in (a) on clonal growth of keratinocytes. Values plotted are average % clonal growth in n = 3 independent screens with n = 3 independent cultures per screen. Green: siSCR control. Red, blue: phosphatases with statistically significant effects on colony formation are shown (red: increase; blue: decrease). Grey: no statistically significant effect. (c, d) Effect of knockdowns on clonal growth after 0, 4, 8 or 12 hr in suspension. (c) Representative dishes. (d) Quantitation of mean % clonal growth ± SD (n = 3 independent samples). p-values generated by unpaired T-test. (e, f) RT-qPCR quantification of TP63 (e) and TGM1 (f) mRNA levels (relative to 18 s expression) in the same conditions as in (c). n = 3 independent transfections. (g) Epidermal reconstitution assay following knockdown of DUSP6, PTPN1, PPP3CA or PTPN13. n = 2 independent transfections. Top row shows representative H and E images. Epidermal thickness was quantified in multiple fields from eight sections per replicate ± SD relative to scrambled control (siSCR). Middle row shows staining for TP63 (pink) with DAPI nuclear counterstain (blue). % DAPI-labelled nuclei that were TP63+ was quantified in n = 2–3 fields per replicate. Bottom row shows staining for Ki67 (brown) with haematoxylin counterstain (blue). % Ki67+ nuclei was quantified in n = 3–6 fields per replicate. Error bars represent mean ± s.d. p-values were calculated using one-way ANOVA with Dunnett's multiple comparisons test (*p<0.05; **p<0.01; ****p<0.0005; ****p<0.0001; ns = non significant). https://doi.org/10.7554/eLife.27356.004 Identification of protein phosphatases that regulate commitment To examine the effects on keratinocyte self-renewal of knocking down each of the 22 phosphatases identified in the screen, we transfected primary human keratinocytes with SMART pool siRNAs and measured colony formation in culture (Figure 2b; Figure 2—figure supplement 2a). Live-cell imaging was used to monitor cell growth for 3 days post-transfection (Figure 2—figure supplement 2b). Knocking down seven of the phosphatases significantly increased clonal growth (Figure 2b), with five phosphatases – DUSP6, PPTC7, PTPN1, PTPN13 and PPP3CA - having the most pronounced effect (p-value<0.001). Silencing of these five phosphatases also increased colony size and delayed the loss of colony forming ability in suspension (Figure 2c,d; Figure 2—figure supplement 2c–d). Consistent with these findings, phosphatase knockdown delayed the decline in TP63 and increase in TGM1 levels during suspension-induced differentiation (Figure 2e,f; Supplementary file 4). Knocking down each phosphatase did not have a major effect on the growth rate of keratinocytes and live-cell imaging did not reveal any apoptosis (Figure 2—figure supplement 2e). Cumulatively, the effects on keratinocyte self-renewal and differentiation suggest that DUSP6, PPTC7, PTPN1, PTPN13 and PPP3CA are pro-commitment protein phosphatases. The effects of knocking down DUSP10 at 0 hr differed from those of the other five phosphatases. DUSP10 knockdown significantly reduced clonogenicity (p-value<0.001; Figure 2b; Figure 2—figure supplement 2a) and, in contrast to the other phosphatases, decreased the growth rate of keratinocytes (Figure 2—figure supplement 2b,e). In addition, whereas expression of DUSP6, PPTC7, PTPN1, PTPN13 and PPP3CA declined by 8 hr in suspension, DUSP10 expression remained high (Figure 2a). Taken together, their effects on keratinocyte self-renewal and differentiation suggest that all six are commitment-associated protein phosphatases, with DUSP10 differing from the other phosphatases in potentially antagonising commitment. To examine the effect of knocking down the pro-commitment phosphatases on the ability of keratinocytes to reconstitute a multi-layered epithelium, we seeded cells on de-epidermised human dermis and cultured them at the air-medium interface for three weeks (Figure 2g; Figure 2—figure supplement 2f,g). Knockdown of DUSP6, PTPN1, PPP3CA and PTPN13 did not prevent cells from undergoing terminal differentiation, as evidenced by suprabasal expression of involucrin and accumulation of cornified cells (Figure 2g; Figure 2—figure supplement 2f). However, the number of TP63-positive cells was increased, whether measured as a proportion of the total number of cells (Figure 2g) or per length of basement membrane (Figure 2—figure supplement 2g). Knockdown of DUSP6, PTPN1 or PPP3CA increased epidermal thickness without increasing the proportion of proliferative, Ki67-positive cells (Figure 2g; Figure 2—figure supplement 2g). Conversely, PTPN13 knockdown led to an increase in the percentage of Ki67-positive cells and a slight reduction in epidermal thickness (Figure 2g), which could reflect an increased rate of transit through the epidermal layers. In addition, Ki67 and TP63-positive cells were no longer confined to the basal cell layer of epidermis reconstituted following phosphatase knockdown but were also present throughout the viable suprabasal layers. To complement our studies with siRNAs, which achieve transient knockdown, we performed stable knockdown of the pro-commitment phosphatases with shRNA lentiviral vectors (two different shRNAs per target; Figure 2—figure supplement 3a) and performed epidermal reconstitution experiments (Figure 2—figure supplement 3b,c). We also developed a tool for automated, unbiased measurements of epidermal thickness (Figure 2—figure supplement 4, Source code 1). No apoptotic cells were observed in reconstituted epidermis, as evaluated by lack of Caspase-3 labelling. Consistent with the effects of siRNA-mediated knockdown, stable knockdown of DUSP6, PPTC7, PTPN1 and PPP3CA increased epidermal thickness, whereas PTPN13 knockdown did not (Figure 2—figure supplement 3b,c). We also confirmed expansion of the TP63+ compartment (Figure 2—figure supplement 3d). Virtually all Ki67+ cells co-expressed TP63 (Figure 2—figure supplement 3d,e). Furthermore, many suprabasal cells co-expressed TP63 and IVL (Figure 2—figure supplement 3f). Thus, on both transient and stable knockdown, the transition from the stem to the differentiated cell compartment was disturbed (Figure 2—figure supplement 3f). Regulation of ERK MAPK and AP1 transcription factors To identify the signalling networks affected by upregulation of phosphatases during commitment, we performed GO analysis of ranked peptides that were dephosphorylated at 4 hr. The top enriched pathways were ErbB1 signalling, adherens junctions, insulin signalling and MAPK signalling (Figure 3a). Several of the proteins we identified are components of more than one pathway (Figure 3b) and all have been reported previously to regulate epidermal differentiation (Connelly et al., 2010; Haase et al., 2001; Kolev et al., 2008; Trappmann et al., 2012; Scholl et al., 2007). In particular, constitutive activation of ERK delays suspension-induced differentiation (Haase et al., 2001). Figure 3 with 1 supplement see all Download asset Open asset Pro-commitment phosphatases regulate MAPK signalling and AP1 transcription factors. (a) Gene Ontology (GO) term enrichment analysis of ranked peptides dephosphorylated at 4 hr. (b) Venn diagram showing intersection of signalling pathways regulated at 4 hr. (c) Top 15 dephosphorylated peptide sites at 4 hr, showing ratio between change in phospho-peptides and change in total protein. Highlighted in orange are phosphorylations on MAPKs. (d, e) Representative western blot (d) and quantification (e) showing phospho-ERK1/2 and total ERK1/2 in cells harvested after 0, 4, 8 and 12 hr in suspension. Cyclophilin B: loading control. (f) Heatmap represents mRNA expression (relative to 18 s RNA) of AP1 transcription factor superfamily members during suspension-induced differentiation post-phosphatase knockdown (n = 3; values plotted are means of three independent transfections). See Supplementary file 6 for p-values generated by two-way ANOVA. (g–h) Representative western blot, with quantitation, of phospho-ERK1/2 and total ERK1/2 in suspended cells following DUSP10 knockdown. siSCR and loading controls are shown. (i) Heatmap showing mRNA expression (relative to 18 s mRNA) of JUN, FOS and MAF family members after DUSP6 and DUSP10 knockdown (values plotted are means of three independent transfections normalised against scrambled control; see Supplementary file 7 for p-values generated by two-way ANOVA). (j) Clonal growth (representative dishes and quantification) following doxycycline-induced over-expression of DUSP10, DUSP6 and mutant DUSP6C293S in primary keratinocytes (n = 3 independent cultures). p-values were calculated using one-way ANOVA with Dunn's multiple comparisons test (*p<0.05; ns = non significant). https://doi.org/10.7554/eLife.27356.009 We next ranked protein phosphorylation sites according to the log2-fold decrease at 4 hr, plotting the ratio between the change in phosphorylation sites and the change in total protein (Supplementary file 5). To specifically identify dephosphorylation events, we excluded from the ranking phosphorylations that remained constant while total protien abundance increased by more than 0.5 in log2. Consistent with the predicted dynamic interactions between signalling pathways (Figure 3b), phosphorylation sites on MAPK1 (ERK2) and MAPK3 (ERK1) were identified in the top 15 most decreased sites (Figure 3c). Other proteins in the top 15 included components or regulators of the cytoskeleton (FLNA), Rho signalling (DOCK5, ARHGEF16, CIT) and EGFR signalling (EPS8), again consistent with the GO terminology analysis. We performed western blotting to confirm that ERK1/2 activity was indeed modulated by suspension-induced terminal differentiation and by the candidate pro-commitment phosphatases (Figure 3d,e). As reported previously, the level of phosphorylated ERK1/2 diminished with time in suspension (Janes et al., 2009) (Figure 3d). siRNA-mediated knockdown of DUSP6, PPTC7, PTPN1, PTPN13 or PPP3CA resulted in higher levels of phosphorylated ERK1/2 relative to the scrambled control, both at 0 hr and at some later time points (Figure 3d,e). These effects are consistent with the known requirement for ERK MAPK activity to maintain keratinocytes in the stem cell compartment (Trappmann et al., 2012). Transcriptional regulation of epidermal differentiation is mediated by the Activator Protein 1 (AP1) family of transcription factors (Eckert et al., 2013), which is the main effector of the MAPK and ErbB signalling cascades (Chang and Karin, 2001). Quantification of the levels of AP1 transcripts during suspension-induced terminal differentiation revealed that different AP1 factors changed with different kinetics, as reported previously (Gandarillas and Watt, 1995) (Figure 3f). Notably, the level of several members of the MAF subfamily of AP1 factors (MAF, MAFB and MAFG) significantly increased during differentiation, consistent with recent evidence that they mediate the terminal differentiation program in human keratinocytes (Lopez-Pajares et al., 2015). In line with these observations, knockdown of individual pro-commitment phosphatases reduced the induction of MAF AP1 factors in suspension (Figure 3f; Supplementary file 6). These experiments are consistent with a model whereby induction of phosphatases in committed keratinocytes causes dephosphorylation of ERK MAPK and prevents the increase in expression of AP1 transcription factors that execute the terminal differentiation program. Given that DUSP10 knockdown differed from the other protein phosphatases in decreasing the clonogenicity (Figure 2b; Figure 2—figure supplement 2a) and growth rate of keratinocytes (Figure 2—figure supplement 2b,e), we examined the effect of DUSP10 knockdown on ERK1/2 activity and AP1 factor expression. Unlike knockdown of DUSP6, PPTC7, PTPN1, PTPN13 or PPP3CA, knockdown of DUSP10 (Figure 3—figure supplement 1a,b) did not increase ERK1/2 activity (Figure 3g,h). Consistent with its known regulation of p38 MAPK (Caunt and Keyse, 2013), DUSP10 knockdown increased phospho-p38 at 0 hr; however, this effect was not observed at later times in suspension (Figure 3—figure supplement 1c,d). When the effects of DUSP10 knockdown on AP1 transcription factors were compared with those of DUSP6, it was clear that these phosphatases differentially regulated several AP1 factors, including members of the JUN, FOS and MAF subfamilies (Figure 3i; Supplementary file 7). We confirmed the differing effects of DUSP10 and DUSP6 on cell fate by using either Cumate or Doxycycline inducible overexpression in human keratinocytes (Figure 3j; Figure 3—figure supplement overexpression of DUSP10 increased colony overexpression of DUSP6 reduced clonal growth (Figure negative mutant of DUSP6 phosphatase activity et al., 2009) no effect (Figure 3j; Figure 3—figure supplement protein phosphatase network as a switch to transition cells between the stem and differentiated cell We next examined whether the six phosphatases we identified could form a network commitment and differentiation. To we a Boolean network and a model in which all phosphatases, DUSP6, PPTC7, PTPN1, PPP3CA and DUSP10, were to either positively or negatively (Figure supplement 1a). this we the concept of an Abstract Boolean et al., which by or approximately different Boolean network (Figure supplement 1a). In with this modelling we expression levels as or their mean expression was higher or than the average of all genes (Supplementary file the set of all Boolean networks we to test whether there be a network consistent with the expression states over the differentiation time To this we the in the tool as by et al., We a set of experimental the different time points during suspension-induced differentiation (Figure 1a). These define whether each be on or at each time the differentiation and this we that a Boolean network was to the measured expression (Figure supplement 1a). As we were not to a network to model the expression states, we each individual time to identify whether the network differ over the time in suspension. To the experimental data to identify network we performed individual knockdowns of the phosphatases after 0, 4, 8 or 12 hr in suspension. we measured by RT-qPCR the effect that each individual knockdown on the mRNA level of any of the other phosphatases (Figure supplement 2a) at each time We used the of the to its or only it was statistically significant Supplementary file This led to the networks in Figure the in each phosphatase with time in suspension, relative to the 0 hr positive effects on expression and These networks reveal different interactions at at each of the time points, that the network during commitment and differentiation. Figure 4 with 3 supplements see all Download asset Open asset network of phosphatases controls commitment. (a) represent log2 in phosphatase mRNA expression relative to 0 hr (values plotted are means of three independent experiments normalised against siSCR (b) that are as on the Boolean network We three experimental expression changes during suspension-induced differentiation in the of as as and expression states at the indicated time cells in the of we a whereby the change the representative network in to achieve the expression the network that represents the at that for a that we did not a the can in the network or switch means the is to be at that means to be at that The is in Supplementary file (c) to the model of the Boolean in (b) are interactions calculated in while are interactions from Figure supplement See also Supplementary file 8 and (d) Heatmap represents mRNA expression (relative to 18 s mRNA) of individual phosphatases in cells in suspension for 12 hr with or (values plotted are the means of three independent experiments normalised against (e, f) of commitment as two in a of the state space for control cells or cells with or In the control both stem and differentiated cell states are stable while commitment is an unstable state. Since the 0 hr network is to the expression for at 12 hr, we that on the only stable state is the stem state. there is a mandatory switch from the 0 hr network but the 12 hr network be at any time we that commitment becomes a stable state while the stem and differentiated cell states are (g) of of human epidermis labelled with against commitment phosphatases or (green) and with DAPI (blue). The of each phosphatase relative to is also shown of primary keratinocytes cultured on and labelled with to DUSP10 and Scale bars: In of these we again a Boolean network to the of epidermal stem cell commitment. the expression levels (Supplementary file we to test whether the network for each time (Figure were consistent with the this we used an of called which the to as by et to whether the networks could the dynamic changes in expression by these as states on the commitment (Figure supplement We that the networks by the knockdown experiments could not the experimental This suggested that there were interactions between the phosphatases. To identify such we the effect that

Type

peer-review

Language

en

Landing Page

https://doi.org/10.7554/elife.27356.036

OA Status

gold

Related Works

10

OpenAlex ID

https://openalex.org/W2985098889

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W2985098889

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.7554/elife.27356.036

Digital Object Identifier

Title

Author response: A protein phosphatase network controls the temporal and spatial dynamics of differentiation commitment in human epidermis

Work title

Type

peer-review

OpenAlex work type

Language

en

Primary language

Publication year

2017

Year of publication

Publication date

2017-09-18

Full publication date if available

Authors

Ajay Mishra, Bénédicte Oulès, Angela Oliveira Pisco, Tony Ly, Kifayathullah Liakath‐Ali, Gernot Walko, Priyalakshmi Viswanathan, Matthieu Tihy, Jagdeesh Nijjher, Sara-Jane Dunn, Angus I. Lamond, Fiona M. Watt

List of authors in order

Landing page

https://doi.org/10.7554/elife.27356.036

Publisher landing page

Open access

Yes

Whether a free full text is available

OA status

gold

Open access status per OpenAlex

OA URL

https://doi.org/10.7554/elife.27356.036

Direct OA link when available

Concepts

Phosphatase, Biology, Cellular differentiation, Cell biology, Context (archaeology), Transcription factor, Genetics, Phosphorylation, Gene, Paleontology

Top concepts (fields/topics) attached by OpenAlex

Cited by

0

Total citation count in OpenAlex

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload