YAP regulates cell size and growth dynamics via non-cell autonomous mediators

Douaa Mugahid; Marian Kalocsay; Xili Liu; Jonathan Scott Gruver; Leonid Peshkin; Marc W Kirschner

doi:10.7554/eLife.53404

. 2020 Jan 9;9:e53404. doi: 10.7554/eLife.53404

YAP regulates cell size and growth dynamics via non-cell autonomous mediators

Douaa Mugahid ¹, Marian Kalocsay ¹, Xili Liu ¹, Jonathan Scott Gruver ¹, Leonid Peshkin ¹, Marc W Kirschner ^1,^✉

Editors: Duojia Pan², Marianne E Bronner³

PMCID: PMC6989120 PMID: 31913124

Abstract

The Hippo pathway regulates organ size, regeneration, and cell growth by controlling the stability of the transcription factor, YAP (Yorkie in Drosophila). When there is tissue damage, YAP is activated allowing the restoration of homeostatic tissue size. The exact signals by which YAP is activated are still not fully understood, but its activation is known to affect both cell size and cell number. Here we used cultured cells to examine the coordinated regulation of cell size and number under the control of YAP. Our experiments in isogenic HEK293 cells reveal that YAP can affect cell size and number by independent circuits. Some of these effects are cell autonomous, such as proliferation, while others are mediated by secreted signals. In particular CYR61, a known secreted YAP target, is a non-cell autonomous mediator of cell survival, while another unidentified secreted factor controls cell size.

Research organism: Human

Introduction

Organ and tissue size are tightly regulated; they scale with body size, and in several cases can be regenerated after partial loss (Penzo-Méndez and Stanger, 2015). The size and number of cells comprising an organ together determine overall size, and these can compensate for each other (Fankhauser, 1945). In some well-studied cases adult organ repair and regeneration involves de-differentiation, followed by proliferation and re-differentiation; while in others, such as skeletal and cardiac muscle, the dedifferentiation-re-differentiation process is skipped, and organs adapt to demand by increasing cell size; or to injury by ECM deposition (fibrosis). The mass of β-cells of the pancreas can increase by augmenting both cell size (hypertrophy) or cell number (hyperplasia) (Anzi et al., 2018; Ernst et al., 2011). Apoptosis can also play an important role in adult organ size homeostasis. For example, the liver enlarges significantly after phenobarbital treatment, but is restored to its normal size by apoptosis upon drug withdrawal (Yang and Xu, 2011). These examples demonstrate that organ size is not simply set during development and passively maintained into adulthood but is actively monitored to fit physiological demands.

The liver offers quite a dramatic example of size control. Although hepatocytes are largely quiescent in adults, the liver can regenerate completely after resection of up to 70% of its original mass (partial hepatectomy). This response is tri-phasic. It starts with an increase in the size (hypertrophy) of the remaining hepatocytes, followed by a period of cell proliferation producing daughter cells of normal size, and is terminated when the original liver mass is restored (Gilgenkrantz and Collin de l'Hortet, 2018). If, in the case of a partial hepatectomy, hepatocytes are blocked from proliferation, liver size can still be restored by hypertrophy of the remaining, often polyploid cells (Diril et al., 2012). This provides a powerful example of how tissue homeostasis can be maintained by a dynamic combination of cell growth and proliferation.

While the role of cell growth in proliferation-dependent liver size homeostasis has never been explicitly tested, it is tempting to speculate that after a partial hepatectomy quiescent hepatocytes re-enter the cell cycle only after increasing in size and that perturbations that impair growth under such conditions impair regeneration (Volarevic et al., 2000). Because of the interdependence of cell size and cell cycle control it is hard to distinguish whether a genetic perturbation that affects proliferation during regeneration primarily affects the cell cycle or cell growth machinery. Though cultured cells provide almost none of the tissue-level context for studying regeneration, they still robustly control cell size by compensatory changes in growth and division (Cadart et al., 2018; Ginzberg et al., 2018).

There is increasing evidence that the Hippo pathway plays a significant role in tissue size regulation (Dong et al., 2007) as well as regeneration (Lu et al., 2018). The pathway is comprised of a cascade of intracellular kinases, activated in response to cell-cell contact. These kinases ultimately phosphorylate and inhibit the downstream transcriptional regulators YAP and TAZ, by causing their sequestration, and often degradation in the cytoplasm (Zhao et al., 2010). Upon tissue injury, the upstream kinases become inactive and YAP/TAZ translocates into the nucleus, where it mediates transcriptional changes that facilitate organ regrowth (Lee et al., 2014). YAP/TAZ activity also depends on changes in the composition and physical properties of the extracellular environment, which is believed to be independent of Hippo activity (Aragona et al., 2013; Dupont et al., 2011; Kim and Gumbiner, 2015). Activation of YAP increases cell proliferation and inhibits apoptosis in vitro and in vivo (Dong et al., 2007; Xin et al., 2011; Zanconato et al., 2015). While the phenomena have been well characterized, the exact molecular mechanism by which YAP affects these different processes and/or their interdependence remains unclear.

In this report, we investigated the mechanisms by which YAP regulates cell number (through proliferation and death) and cell growth/size in mammalian cultured cells. The term ‘cell growth’ is commonly used in two very different contexts. It is important to distinguish these processes because ‘exponential growth’ has meant either an exponential increase of cell mass in each cell cycle or an exponential increase in the number of cells per unit time. In this paper we will use the term, ‘cell growth’ exclusively to describe changes in average cell size, and ‘population growth’ exclusively to describe changes in the number of cells.

Our results suggest that YAP affects both cell size and number through distinct mechanisms. Some of these effects appear to be cell-autonomous, but notably we find strong evidence that others are driven by diffusible, non-cell autonomous factors. In particular we show that CYR61, one of the extracellular-residing proteins that is regulated both on the mRNA and protein level in response to increased YAP activity, has no effect on cell size. However, CYR61 mediates a significant part of the YAP-dependent increase in cell number by limiting apoptosis in high density cultures. We conclude that YAP stimulates the production of a suite of extracellular and cell surface proteins that act in a paracrine and autocrine manner regulating both cell number and size.

Results

Overexpression of non-phosphorylatable YAP increases the rate of cell division, cell size, and maximum cell density in HEK293 populations

To study the effect of YAP on cell size and growth kinetics, we generated Dox-inducible, isogenic FlipinTrex-HEK293 cells expressing nuclear mCherry (n-mCherry) together with either nuclear GFP (nGFP) or a GFP-tagged phosphosite mutant of YAP (YAP5SA). Unlike wildtype (wt-) YAP, YAP5SA cannot be inhibited by the upstream kinases of the Hippo pathway and therefore, is constitutively active. The fact that the cell lines are isogenic, ensures that any observed changes in size or proliferation rate between the cell lines do not arise from the genomic location of the inserts, but rather to differences in biological activity of the expressed proteins. Conveniently the n-mCherry demarks the nucleus, enabling easy counting of cell number in low magnification images (Figure 1—figure supplement 1). The nuclear marker also serves another important function, as an indirect measure of cell size (Ginzberg et al., 2018). As shown in Figure 1—figure supplement 2, this correlation is very strong when cell size is measured by the highly accurate Quantitative Phase Microscopy (QPM). The Pearson correlation coefficient was ~0.9 when measured under various conditions. With this experimental design, we could conveniently track changes in cell number and cell size in a population over time.

We found that the population growth rate in these cells decreased as cell density increased, and eventually reached zero at high cell density (Figure 1A). This is a well-known phenomenon often referred to as density-dependent growth inhibition (Stoker and Rubin, 1967) or contact inhibition; and can be described by a logistic growth expression, where the net growth rate of a population is exponential at low density and decreases to a minimum at high density (Figure 1—figure supplement 3A–C) (Verhulst, 1845). In this phenomenological description, changes in the maximum cell number are denoted by Ymax, also known as the population carrying capacity; and the growth rate is described by k (with units of reciprocal time).

Figure 1. — (A) Example of population growth dynamics. The figure is a fit of data to the logistic growth equation in Figure 1—figure supplement 3A. Ymax is the carrying capacity of the population. (B) An example of changes in average nuclear area in a population of unsynchronized Flipin-Trex-293 cells over time after fitting to a plateau followed by exponential decay. While nuclear area is initially constant, it exponentially decays to a minimum as cell number increases. (C) Population growth curves of FlipinTrex-293 cells expressing constitutively active YAP (YAP5SA) or nuclear-GFP (nGFP). FlipinTrex-293 cells were seeded at ~20,000 cells/well on a 96-well plate (low density) and monitored over time. Ymax and k are both higher in 5SA cultures than nGFP cultures. In black, the average of 4 wells; in red, the fit to a logistic growth model; n = 4; mean ± SEM. (D) Changes in nuclear area over time in the same populations of cells in (C). Nuclear area is larger in YAP5SA cells vs. controls, but still decreases exponentially as cell density increases. In black, the average of 4 replicates; in red, the fit to a plateau followed by an exponential decay; n = 4; mean ± SEM. (E) Cell confluence is estimated by the relative area of the culture vessel covered by cells in bright field images. We observe no significant difference in the confluence of the YAP5SA vs nGFP cells. (F) Protein content was compared between populations of YAP5SA and nGFP ~20 hr after the same number of cells were seeded in 10 ml medium on 10 cm-dishes to represent either low density (~25% confluence) or high density (>90% confluence) conditions. All samples were simultaneously trypsinized, permeabilized, stained and measured on a LSRII flow cytometer at a concentration of 1 × 10⁶ cells/ml. A total of 10,000 cells were analyzed per condition. Protein content is higher in YAP5SA cells whether sparsely or densely seeded. (G) The population mean and SD of the data in (F).

Figure 1—figure supplement 1. — (A) Example of population growth dynamics. The figure is a fit of data to the logistic growth equation in Figure 1—figure supplement 3A. Ymax is the carrying capacity of the population. (B) An example of changes in average nuclear area in a population of unsynchronized Flipin-Trex-293 cells over time after fitting to a plateau followed by exponential decay. While nuclear area is initially constant, it exponentially decays to a minimum as cell number increases. (C) Population growth curves of FlipinTrex-293 cells expressing constitutively active YAP (YAP5SA) or nuclear-GFP (nGFP). FlipinTrex-293 cells were seeded at ~20,000 cells/well on a 96-well plate (low density) and monitored over time. Ymax and k are both higher in 5SA cultures than nGFP cultures. In black, the average of 4 wells; in red, the fit to a logistic growth model; n = 4; mean ± SEM. (D) Changes in nuclear area over time in the same populations of cells in (C). Nuclear area is larger in YAP5SA cells vs. controls, but still decreases exponentially as cell density increases. In black, the average of 4 replicates; in red, the fit to a plateau followed by an exponential decay; n = 4; mean ± SEM. (E) Cell confluence is estimated by the relative area of the culture vessel covered by cells in bright field images. We observe no significant difference in the confluence of the YAP5SA vs nGFP cells. (F) Protein content was compared between populations of YAP5SA and nGFP ~20 hr after the same number of cells were seeded in 10 ml medium on 10 cm-dishes to represent either low density (~25% confluence) or high density (>90% confluence) conditions. All samples were simultaneously trypsinized, permeabilized, stained and measured on a LSRII flow cytometer at a concentration of 1 × 10⁶ cells/ml. A total of 10,000 cells were analyzed per condition. Protein content is higher in YAP5SA cells whether sparsely or densely seeded. (G) The population mean and SD of the data in (F).

We initially thought that Ymax would only be affected by the area of the culture vessel (literally the carrying capacity), while k would be a function of the concentration of growth factors in the medium. Indeed, serum concentrations above 1% changed k (Figure 1—figure supplement 3D) without changing Ymax (Figure 1—figure supplement 3E). However, Ymax was lower at serum concentration <1% suggesting that the area of the culture vessel was not the only determinant of Ymax.

A more complete description should not consider proliferation as the sole determinant of cell number but also include the cell death rate. By measuring cell death directly by YOYO-1, a cell impermeable dye that fluoresces green when bound to dsDNA in dead cells, we could show that cell death rates are low in low-density cultures (Figure 1—figure supplement 4A,B). Hence at low density the net population growth is dominated by the cell division rate. At high density, when net population growth is zero, the death rate dramatically increases (Figure 1—figure supplement 4A). Thus extra- and intra- cellular signals appear to affect k and Ymax differently, with the former predominantly reflecting cell division rates and the latter, cell death rates.

Nuclear area in low density cultures (<~15,000 cells/well in a 96-well plate) is initially constant at ~120 μm² in control cells, after which it decreased about 33% to ~85 μm² at maximum density (Figure 1B). Assuming the nucleus is a perfect sphere, this change in area would correspond to a 50% decrease in volume. However, actual measurements of cell volume using Coulter Counter indicate a ~ 20% decrease between low density (25% confluent) and high density (100% confluent) cultures (Figure 1—figure supplement 4C). The closer correlation between nuclear area and volume than predicted by a spherical representation of the nucleus/cell is consistent with the strong correlation we observed between dry mass and nuclear area (Figure 1—figure supplement 2A).

The expression of YAP5SA increases both k and Ymax compared to controls (Figure 1C). Furthermore, at low cell density the area of the nucleus increases 17% in the YAP5SA cells. A change of such magnitude would translate to ~30% increase in volume, which we measured by Coulter Counter to be ~25% larger (Figure 1—figure supplement 5A). However, as the number of cells continued to grow, the nuclear area of YAP5SA cells also decreased to a minimum where the difference in size compared with controls was much reduced (~10%; Figure 1D). The onset of the exponential decrease was slightly delayed when cells overexpressed YAP5SA, though cultures appear to be of comparable density (Figure 1E).

To confirm that the changes in nuclear area reflected changes in biomass and not just changes in volume, we measured protein mass directly in individual cells using a covalent dye specific for lysine groups on the protein, which is a proxy for changes in size (Kafri et al., 2013). We found that average protein content per cell was ~30% higher in low-density populations expressing YAP5SA (Figure 1F,G) than in nGFP controls. We also observed that cells seeded at four times the density had ~25% less protein (Figure 1F,G). These measurements of protein mass reinforced both of the conclusions drawn from the changes in nuclear area: (1) YAP5SA cells are larger than control cells in low density cultures, and (2) cell size decreases at higher cell population density. We ruled out that the changes were merely due to an increase in the fraction of cells in S or G2/M by simultaneously measuring DNA and protein (Figure 1—figure supplement 5B–C). The latter findings were independently confirmed in three different isogenic clones using QPM on live, attached low-density cultures (Figure 1—figure supplement 5D). Based on our measurements there is a ~ 20% increase in dry cell mass in cells expressing YAP5SA. Thus, it is fair to say that the expression of YAP5SA increases cell size in all phases of the cell cycle in a highly reproducible manner independent of the choice of clone, staining errors or detachment of cells from their substrate. However, the exact magnitude of change in size varied somewhat depending on the measurement method of choice. Further experiments using non-isogenic HEK293 cells expressing GFP-tagged YAPWT vs. nGFP demonstrated that YAPWT overexpression only increased cell size in low density cultures but not in high density cultures, a condition where YAP is excluded from the nucleus and thus transcriptionally inactive (Figure 1—figure supplement 5E). We also confirmed that knocking down YAP by 80% in WT cells by constitutively expressing YAP-targeting shRNA (YAPKD cells; Figure 1—figure supplement 5F) decreased cell size (Figure 1—figure supplement 5G) as previously reported (Hansen et al., 2015). Through the remainder of the manuscript, we mainly focus on the most dramatic changes in size caused by YAP5SA in sub-confluent cultures.

The non-cell autonomous nature of YAP-mediated changes in cell size

In isogenic Flipin-Trex cells, YAP5SA or nGFP expression is doxycycline (Dox)-inducible. Therefore, we would not expect that in the absence of Dox there would be any difference in cell size between uninduced YAP5SA, nGFP and parental cells. To our surprise in the absence of Dox the average cell mass in the three YAP5SA-expressing clones was significantly higher than in nGFP clones and the parental cells (9% vs. 20% when cells are cultured in Dox for 4 days; Figure 2A vs. Figure 1—figure supplement 5D). It is well known that Dox-inducible expression systems can be leaky, so one possible explanation for the larger size would be that the <1 % cells expressing YAP5SA in the absence of Dox could be affecting the behavior of the population as a whole. To test this possibility, we compared the size of parental FlipinTrex-293 cells (WT) when co-cultured together with cells expressing GFP-tagged YAP5SA or nGFP (Figure 2B). We found that when cultured in the same well WT cells were similar in size to the YAP5SA cells, both of which were ~30% larger than the nGFP cells and their co-cultured WTs (Figure 2C). Thus, although the expression of YAP5SA is obviously cell autonomous, its effect on size is shared with untransfected cells in the same well.

Figure 2. — (A) The average dry mass/cell ± S.D. in cultures of 3 isogenic clones of Flipin-Trex-293 cells engineered to express YAP5SA or nGFP was measured using quantitative phase microscopy (QPM). Cells had never been cultured in the presence of doxycycline before they were transferred to a Fibronectin-coated, glass-bottom 6-well plate at ~50,000 cells/well for the measurement. 30 min before the measurement Hoechst was added at a concentration of 2 μg/ml to allow DNA staining of live cells. The overall average per clone was calculated from >4,000 cells. The average size per cell cycle phase was determined from cells binned into the respective cell cycle phase depending on Hoechst-levels. Cell size is slightly higher in the population of uninduced 5SA cells vs. controls. (B) Experimental setup of data in (C). In short, cells expressing nuclear GFP (nGFP) or GFP-tagged YAP5SA (YAP phospho-mutant) cells were intermixed with wildtype (WT) cells at a ratio of 2:1 and grown for ~24 hr on 10 cm dishes before they were trypsinized and processed as single-cell suspensions at a concentration of 1 × 10^6 cells/ml. Total cell number was kept the same between both cultures, and all samples were simultaneously prepared for labelling using the SE-647 protein dye and analyzed on a LSRII. (C) Histograms of average protein content as estimated by SE-657 signal per cell. Cells were separated into two groups depending on whether or not they expressed GFP. Cc denotes the GFP-negative WT cells in the culture. GFP+ (red dotted line) and GFP- cells (black dotted line) in YAP5SA-co-cultures are larger than cells in the nGFP co-cultures. (D) Experimental set up for data in (E). WT cells were co-cultured with YAP5SA or nGFP cells on a 0.4 μm Transwell membrane in a 10 cm dish. The membrane allows the exchange of medium components but not physical contact. (E) WT cells from the setup in (D) were processed as explained above to allow SE-dye based quantification of their protein content. WT cells co-cultured with YAP5SA cells have higher protein content than those co-cultured with nGFP cells. (F) FlipinTrex-293 cells expressing n-mCherry were starved for 6 days in 0.5% FBS before they were treated with equal amounts of 5SA- or nGFP-conditioned medium. Medium conditioned by 5SA cells but not nGFP cells increases nuclear area.

Figure 2—figure supplement 1. — (A) The average dry mass/cell ± S.D. in cultures of 3 isogenic clones of Flipin-Trex-293 cells engineered to express YAP5SA or nGFP was measured using quantitative phase microscopy (QPM). Cells had never been cultured in the presence of doxycycline before they were transferred to a Fibronectin-coated, glass-bottom 6-well plate at ~50,000 cells/well for the measurement. 30 min before the measurement Hoechst was added at a concentration of 2 μg/ml to allow DNA staining of live cells. The overall average per clone was calculated from >4,000 cells. The average size per cell cycle phase was determined from cells binned into the respective cell cycle phase depending on Hoechst-levels. Cell size is slightly higher in the population of uninduced 5SA cells vs. controls. (B) Experimental setup of data in (C). In short, cells expressing nuclear GFP (nGFP) or GFP-tagged YAP5SA (YAP phospho-mutant) cells were intermixed with wildtype (WT) cells at a ratio of 2:1 and grown for ~24 hr on 10 cm dishes before they were trypsinized and processed as single-cell suspensions at a concentration of 1 × 10^6 cells/ml. Total cell number was kept the same between both cultures, and all samples were simultaneously prepared for labelling using the SE-647 protein dye and analyzed on a LSRII. (C) Histograms of average protein content as estimated by SE-657 signal per cell. Cells were separated into two groups depending on whether or not they expressed GFP. Cc denotes the GFP-negative WT cells in the culture. GFP+ (red dotted line) and GFP- cells (black dotted line) in YAP5SA-co-cultures are larger than cells in the nGFP co-cultures. (D) Experimental set up for data in (E). WT cells were co-cultured with YAP5SA or nGFP cells on a 0.4 μm Transwell membrane in a 10 cm dish. The membrane allows the exchange of medium components but not physical contact. (E) WT cells from the setup in (D) were processed as explained above to allow SE-dye based quantification of their protein content. WT cells co-cultured with YAP5SA cells have higher protein content than those co-cultured with nGFP cells. (F) FlipinTrex-293 cells expressing n-mCherry were starved for 6 days in 0.5% FBS before they were treated with equal amounts of 5SA- or nGFP-conditioned medium. Medium conditioned by 5SA cells but not nGFP cells increases nuclear area.

Non-cell autonomous effects could be mediated either by cell-cell contact, or by diffusible molecules (secreted into or depleted from culture medium). To investigate whether these effects required cell-cell contact, WT Flipin-Trex-293 cells were seeded on the bottom of a well, physically separated from a population of feeder cells expressing YAP5SA or nGFP in a Transwell chamber (Figure 2D). In this condition both cell types share and exchange components in the medium but do not physically contact each other. When cell size was measured by protein content, cells co-cultured with YAP5SA cells were ~50% larger compared with those co-cultured with nGFP controls (Figure 2E). This suggests that YAP’s non-cell autonomous effect on cell size is likely mediated by a soluble factor(s) produced by YAP5SA-expressing cells. Based on these results we would expect that conditioned medium from 5SA-expressing cells, when added to naïve cells, would result in a size increase, which it did (Figure 2F). Conditioned medium also increased the size of NIH 3T3 fibroblasts after a 48 hr treatment (Figure 2—figure supplement 1A,B).

The effect of YAP5SA on cell number is mediated both by cell autonomous and non-autonomous processes

To investigate whether YAP expression affects cell number as well as cell size in a non-autonomous manner, we co-cultured cells expressing n-mCherry ‘recipients’ with cells expressing either YAP5SA or nGFP only ‘feeders’ (Figure 3A). The two cell types were well-mixed keeping the total cell count constant while varying the ratio between the recipient and feeder cells. We found that recipient cells grew to a higher number (Ymax) when YAP5SA cells were in large excess (four to eight times as many), compared to being co-cultured with nGFP controls (Figure 3B,C). The population growth rate at low density (k) was not affected by the number of YAP5SA expressing cells (Figure 3D).

Figure 3. — (A) Experimental design for results in (**B–G**). Flipin-Trex-293 cells expressing nuclear GFP (nGFP) or GFP-tagged YAP5SA (YAP phospho-mutant) work as feeder cells for intermixed recipients expressing nuclear mCherry (n-mCherry). The total number of cells per well at seeding is kept constant at ~20,000 cells/well on a 96-well plate, though the ratios between cell types varied. Ymax and k are estimated based on counts from the recipient population only. (**B–D**) Increasing the fraction of YAP5SA to n-mCherry cells increases Ymax but does not affect k; *p-values* indicate the likelihood that the slope is non-zero; n = 5 wells; mean ± SEM. (B) Depicts raw data. (E) Population growth curves of YAP knockdowns (light blue) or empty vector controls (Ctrl; purple) when co-cultured with twice the number YAP5SA (dashed lines) or nGFP (solid line) cells; n = 6; mean ± SEM. Black lines represent the fit of the data to a logistic model. (F) Ymax is higher when recipient cells are co-cultured with YAP5SA feeders even if recipients have reduced levels of endogenous YAP (YAPKDs). Ymax is slightly lower than controls in YAPKDs. (G) K is lower in YAPKD recipients than ctrls, and is unaffected by feeder cell type. Meanwhile, co-culturing ctrls with YAP5SA feeders as depicted in (A) does increase k mildly. (H) Cells expressing n-mCherry were starved in 0.5% FBS for 6 days before they were treated with 5SA- or nGFP-conditioned medium. Treating cells with YAP5SA-conditioned medium increases Ymax in comparison to nGFP-conditioned medium; n = 4; mean ± SEM.

We suspected the difference between k and Ymax in the presence of YAP5SA was because its effect on k is cell autonomous, while its effect on Ymax is not. If this hypothesis were true, k would be lower in YAPKD cells irrespective of what cells they were co-cultured with. In contrast Ymax should increase in YAPKDs when co-cultured with YAP5SA feeders. Indeed, k was lower in YAPKDs vs controls, and co-culturing either cell type with YAP5SA cells had little effect on k, while significantly increasing Ymax (Figure 3E–G). These results are most easily explained if the effect of YAP on k was indeed cell autonomous, whereas its effect on Ymax was non-cell autonomous. Furthermore, we show that treating serum-starved FlipIn-Trex-293 or 3T3 cells with 5SA-conditioned medium increased the number of cells in dense cultures vs. the addition of conditioned medium derived from nGFP cells (Figure 3H and Figure 1—figure supplement 5D,E). These results also support a YAP-dependent, non-cell autonomous mode of regulating Ymax.

Many proteins affected by YAP5SA overexpression are secreted or membrane proteins

We have tested the notion that these non-cell autonomous effects are mediated by secreted proteins that are transcriptional targets of YAP. Using RNA-Seq and mass spectrometry we compared the transcriptional and corresponding protein expression changes in FlipinTrex-293 cells in response to YAP5SA expression (Figure 4A, Supplementary file 1). We concluded that extracellular proteins were a large portion of those expressed in response to YAP5SA (eight fold enrichment relative to all detected proteins, p-value<10⁻¹²; Figure 4B). While in some cases the change in RNA and protein levels were similar, in many other cases they were not. For example, CTGF, CYR61 and AMOTL2, which are canonical YAP targets, were elevated at the protein and mRNA levels, while we see an almost four-fold decrease in LOX and TNC proteins, with no change on the mRNA level (Figure 4A). CYR61 and CTGF are members of the CCN family of matricellular proteins, non-structural proteins that reside in the extracellular matrix and are thought to regulate cellular processes like migration, apoptosis, division and differentiation, involving the coordination of several signaling pathways (Holbourn et al., 2008). Though both genes have previously been used as robust transcriptional readouts for YAP activity (Dupont et al., 2011; Yimlamai et al., 2014), their contribution to the YAP-dependent changes in cell behavior has not been clarified. Specifically, their role in coordinating YAP-dependent cell proliferation, apoptosis and size control has not been explicitly studied. We also considered whether Amphiregulin, a secreted molecule related to EGF, that had previously been identified as a potential mediator of YAP-dependent non-cell autonomous growth in 3D cultures of MCF10 cells (Zhang et al., 2009) was involved in YAP-dependent size or proliferation control in our cells. In these experiments, Amphiregulin RNA was expressed at a very low level (FPKM <1 compared with FPKM >100 for CYR61); and was not identified by mass spectrometry. By contrast we identified 3 CYR61 and 6 CTGF peptides.

Figure 4. — (A) Scatter plot of changes in protein (y-axis) vs. mRNA levels (x-axis) upon YAP expression. Dots in purple indicate proteins have a log2[fold change] (log2FC) more than two standard deviations away from the mean. Dotted line indicates a 1:1 change in protein vs. mRNA levels. (B) The 10 most significantly enriched cellular compartments in which regulated proteins reside (according to fold enrichment relative to all identified proteins). Benjamini-Hochberg *adjusted p-values* are indicated in italics.

The YAP target, CYR61, affects cell number but not cell size

To test whether CYR61 and/or CTGF mediated any of the non-cell autonomous changes in cell size or number, we used neutralizing antibodies to inhibit their extracellular activity in co-cultures of n-mcherry recipient cells and YAP5SA cells. The identity of the recombinant proteins and the specificity of the antibodies we used were demonstrated by their mutual recognition in Figure 5—figure supplement 1. Neutralizing CTGF had a small effect on cell number, mildly decreasing k (~12%) while mildly increasing Ymax (~12%) of the n-mCherry cells in the co-cultures (Figure 5A–C). On the other hand, neutralizing CYR61 significantly decreased Ymax (up to ~50%) but increased k (up to ~200%) in a dose-dependent manner (Figure 5D–F). Our RNA-Seq data suggest that the amount of CYR61 (FPKM >100) transcribed in these cultures is greater than CTGF (FPKM <50). Consistent with the difference in expression levels, increasing the concentration of antibody had a much larger effects for anti-CYR61, as compared to CTGF (Figure 5A,D). These findings support a major role for CYR61 in the non-cell autonomous, YAP-mediated increase in Ymax. However, the increase in the rate of population growth observed at low cell density appears not to be mediated by CYR61, consistent with the measurements in Figure 3 that show that the dominant effects on k are cell autonomous.

Figure 5. — CTGF or CYR61 were neutralized by the addition of antibodies at the indicated concentration to the culture medium when cells were seeded. (**A–C**) In co-cultures of 6x YAP5SA: 1x nuclear mCherry expressing cells (n-mCherry), CTGF neutralization increases the n-mCherry cells’ carrying capacity (Ymax), while decreasing their rate of growth (k). (**D–F**) Neutralizing CYR61 in these cultures has opposing effects of larger magnitude compared to CTGF. (**G–I**) Treating cells expressing empty vector (WT) with 1 μg/ml of CYR61 increases Ymax and decreases k. (**J–L**) YAPKD cells respond to exogenous CYR61 with an increase in Ymax and a decrease in k. (**M–N**) Cell size is not affected by the presence of anti-Cyr61 or anti-CTGF. (**O, P**) Adding recombinant CYR61 at 1 μg/ml also has little effect on cell size in YAPKDs and controls. (B), (C), (E), (F), (H), (I), (K), (L) n = 5 wells; mean ± SEM. (**M–P**) n = 5 wells; mean ± SD).

Figure 5—figure supplement 1. — CTGF or CYR61 were neutralized by the addition of antibodies at the indicated concentration to the culture medium when cells were seeded. (**A–C**) In co-cultures of 6x YAP5SA: 1x nuclear mCherry expressing cells (n-mCherry), CTGF neutralization increases the n-mCherry cells’ carrying capacity (Ymax), while decreasing their rate of growth (k). (**D–F**) Neutralizing CYR61 in these cultures has opposing effects of larger magnitude compared to CTGF. (**G–I**) Treating cells expressing empty vector (WT) with 1 μg/ml of CYR61 increases Ymax and decreases k. (**J–L**) YAPKD cells respond to exogenous CYR61 with an increase in Ymax and a decrease in k. (**M–N**) Cell size is not affected by the presence of anti-Cyr61 or anti-CTGF. (**O, P**) Adding recombinant CYR61 at 1 μg/ml also has little effect on cell size in YAPKDs and controls. (B), (C), (E), (F), (H), (I), (K), (L) n = 5 wells; mean ± SEM. (**M–P**) n = 5 wells; mean ± SD).

To test whether YAP expression in the responding cells is required for the CYR61-dependent increase in Ymax, we knocked down YAP and added CYR61. As Shown in Figure 5G–L YAPKDs as well as controls responded by increasing Ymax by ~30% and 20% respectively, but decreased k by about 10% and 50% (Figure 5G–L), confirming that CYR61’s non-cell autonomous influence on cell number – specifically the carrying capacity (Ymax) – does not require active YAP in the responding cells. These results were therefore consistent with the co-culture experiments (Figure 3E–G). Finally, neither CTGF or CYR61 neutralization nor addition affected nuclear area (Figure 5M–P); hence some factors other than CTCF or CYR61 must be responsible for the non-autonomous effects of YAP on cell size.

Cyr61 increases Ymax by reducing apoptosis

Ymax is clearly different between nGFP and 5SA cells (Figure 1C), although the distribution of cells in the cell cycle suggests no difference in proliferation rates between both cell types at high density (Figure 6A). We hypothesized the striking effect of CYR61 on Ymax is most easily explained by an effect on cell death, which occurs at higher rates in high vs. low density cultures (Figure 1—figure supplement 1–4A,B). Consistent with that, in the presence of anti-Cyr61 antibody, there is a three-fold increase in cell death in cultures of Flipin-Trex-293 cells, which can be reduced to control levels upon addition of an anti-apoptotic peptide (Q-VD-OPH; Figure 6B). We also observed that cells treated with anti-CYR61 were mostly all dead by the end of 6 days (Figure 6C). Cells treated with CYR61 or IgG were almost 100% alive with many still cycling. We conclude that high levels of CYR61 reduce density-dependent apoptotic cell death in YAP5SA cells.

Figure 6. — (A) FlipinTrex-239 cells expressing YAP5SA (5SA) or nuclear GFP (nGFP) were seeded at a density of 10⁷ cells per 10 cm dish which is equivalent to the number of cells on a fully confluent dish. ~24 hr later, cells were trypsinized, treated with the cell-permeable DNA dye Hoechst 3342, and analyzed on a LSRII. The distribution of FlipinTrex-293 cells in the different cell cycle phases is comparable between these cultures. (B) Cells were seeded in the presence of 50 nM of the green fluorescent, non-cell permeable DNA dye YOYO-1; 10 μg/ml of anti-CYR61 or polyclonal rabbit IgG as a conrol; and 10 μM of the anti-apoptotic peptide Q-VD-OPH. Cell death increases as cell density increases as indicated by the increase in green fluorescent signal per well. Cell death is exacerbated in the presence of anti-Cyr61, and can be restored to control levels in the presence of the anti-apoptotic peptide Q-VD-OPH. (C) FlipinTrex-293 cells were cultured at high density cultures in the presence of YOYO-1 and treated with 10 μg/ml of anti-Cyr61/IgG or 1 μg/ml of recombinant CYR61 for 6 days. By that time, most of the cells treated with anti-CYR61 are mostly dead, while IgG controls and those treated with recombinant CYR61 are mostly alive.

Discussion

In the early studies of cell growth, the focus was on unicellular organisms or cells grown at low density. Studying cell growth at high density meant having to deal with complex interactions of the individual components, the space they occupy, and effects on the solvent or the medium, all of which are less confounding at low density. Nevertheless, interest in growth at high cell density was piqued when biologists made 2D cultures of animal cells on petri dishes. For example, chicken embryo fibroblasts grown in such a manner generate a monolayer and stop growing (Weinberg, 2007). It is unlikely that this was simply depletion of the medium, for when Rous Sarcoma Virus was added the quiescent cells were transformed and proliferated, overgrowing the monolayer and producing three dimensional ‘foci’ on top of the monolayer (Groupe and Manaker, 1956). There are two phenomena here to consider, the still unexplained tendency of cells to stop growing upon crowding and the tendency of a few infected or transformed cells to overcome whatever inhibitions were observed in the uninfected monolayers. With the passage of 50 years these phenomena have neither been completely resolved nor completely forgotten. The problem of cell interactions is still a serious topic in cancer (Kamińska et al., 2015) and more recently has emerged in many other biological settings, such as the behavior of cells in regeneration or in stem cell niches (Lane et al., 2014).

More interest in the reciprocal interactions of proliferating and resident cell populations was stirred by the discovery of the Hippo pathway, originally discovered in Drosophila, and later shown to be activated in high-density cultures. The major effector of the pathway is the transcription regulator YAP, which is formally considered an oncogene, because it stimulates growth (Dong et al., 2007). Over-activating YAP results in a breakdown of some of the usual barriers to growth in complex tissues and in high density cell cultures (Zhao et al., 2007). An interesting feature of the pathway is its sensitivity to alterations in the nature of the substratum, such as ECM composition and stiffness (Dupont et al., 2011; Kim and Gumbiner, 2015). It is still generally assumed that the aggressive behavior of the cells is governed in a cell autonomous manner – like the original Rous transformation mentioned above. Another, and more unique feature of YAP is its control of organ size. Organ size is determined by a tight coordination of cell death, proliferation and sometimes differentiation as well as cell size. Organ size control is best studied in vivo where the complexity of cell-cell and cell-matrix interactions are better maintained, but this complexity makes teasing apart underlying mechanisms challenging. Cultured cells can be used to model some of the aspects of organ size in the whole animal but lacks the more complex context. Nevertheless, as cultured systems become more complex, there is an expectation that mechanistic features of organ and tissue growth can be recognized and understood.

In this paper we have extended observations to cell autonomous and non-cell autonomous signaling downstream of YAP in cultured cells. In this system we have examined the growth of cells at different cell densities; the effects of cell proliferation and cell death; and the combined effects of proliferation and cell size. By such measurements we were able to gain a clearer understanding of YAP’s role in regulating these phenomena on a cellular as well as population level. We found that the growth rate of populations, k, at low density, was elevated significantly when cells expressed a constitutively active form of YAP, YAP5SA. compared to wild type (Figure 1B) cultures. The effect on k was cell autonomous and driven predominantly by a change in proliferation rates. This increase in proliferation was accompanied by an increase in the fraction of cells in S-G2-M phases of the cell cycle and may be explained by previous reports that there is a YAP-mediated increase in the cell cycle regulators CyclinE and Cyclin D, and consequently a higher rate of entry into S-phase (Dong et al., 2007; Shu and Deng, 2017; Tapon et al., 2002).

Even more obvious was YAP5SA’s effect on the number of cells at high density, where the carrying capacity, Ymax, increased about 35%. The effect of YAP on Ymax was demonstrably non-cell autonomous and mediated by CYR61, which increases Ymax even in the absence of endogenous YAP indicating that it is not simply an autocrine mediator. CYR61 is one of the strongest genes we find differentially expressed on both the protein and mRNA level. It is a common YAP target that is a member of the CCN group of matricellular proteins, known to be involved in tissue remodeling (Chen et al., 2001), inflammation and embryogenesis (Katsube et al., 2009; Latinkic et al., 2003). In our experiments, the effects of CYR61 on cell number at high density are largely a result of a reduction in apoptosis, while it seems to suppress proliferation at low density. Previously, CYR61 was shown to reduce apoptosis in MCF7 cells (Lin et al., 2004) and is one of the earliest induced genes in response to a partial hepatectomy, which also involves/requires YAP (Grijalva et al., 2014; Lu et al., 2018; Su et al., 2002). Only recently has CYR61 activity in hepatocytes been implicated in YAP-mediated liver fibrosis by activating macrophages (Mooring et al., 2020), suggesting that CYR61 is more than a YAP reporter in that context, and is of importance in mediating non-cell autonomous YAP activity in vivo. CYR61 was also reported to play a role in inhibiting hepatocyte proliferation in response to carcinogen-mediated injury (Chen et al., 2016). Altogether, this suggests that CYR61 plays various roles in cell culture as well as in vivo, which could depend on the cell type as well as the particular context in which CYR61 is activate. This is particularly interesting in the whole-organ context, where the coordination of interactions between cell types is often necessary and would be expected to depend on cell non-autonomous signaling. The present study involved only a single type of cell examined in two contexts as responders and signalers. Exploring the role of YAP and its extracellular targets in mixed cultures of different cell types could provide additional insight into its role in complex tissues.

Finally, cell size, also seems to be regulated non-cell autonomously upon YAP expression. We applied several modes of measurement: nuclear area, volume, protein content and dry mass, all of which are often used as metrics to describe cell size, and they all show that expressing YAP5SA increases cell size between ~17–30%. We believe the changes in nuclear area and dry mass (17% vs. 20% respectively) are the most reliable and are also well-correlated. We tested whether the two extracellular proteins most affected by YAP expression, CYR61 and CTGF, have an effect on WT cell size. Neither the depletion of either protein nor the addition of the recombinant proteins affected cells size. We did not find evidence that the classic size regulating growth factors IGF1/2 are produced to any appreciable degree by our cells. There are, however, many candidate secreted proteins that are stimulated by YAP and which should be tested in the future.

The YAP targets, CTGF and CYR61, both members of the CCN family of matricellular proteins, have opposing effects on cell number in co-cultures with YAP5SA expressing cells. The levels of CTGF are small compared with CYR61 (Figure 5) which we believe explains why the observed non-cell autonomous effects on population growth are mainly those mediated by CYR61. CCN proteins, unlike typical growth factors, are not known to bind to any receptor tyrosine kinases, but rather to integrins and other growth factors, thereby mediating the activity of those conventional growth factors (Leask and Abraham, 2006). With low serum in the medium (<1%), Ymax decreases (Figure 1—figure supplement 3E). Size is also affected by the level of serum (Figure 1—figure supplement 6B), suggesting that another matricellular factor might mediate YAP-dependent changes in cell size.

Work by Tian and colleagues (Tian et al., 2013) had previously demonstrated that a high molecular weight hyaluronic acid produced by naked mole rate fibroblasts contributes to early contact inhibition; the equivalent of a decrease in the population’s carrying capacity. This form of hyaluronic acid acts via the upstream Hippo component NF2, which inhibits YAP activity. Combined with our results on secreted factors, such findings suggest that extracellular cues are an important part of the regulation of both cell and organ (tissue) size. The non-cell autonomous nature of YAP signaling would allow local signals to affect remote cells of the same or different type as is expected to occur in regeneration. In other words, the activation of YAP in a small fraction of cells at the site of injury could through secreted signals engage a wider portion of the tissue or even a whole organ to participate in tissue repair. The receiving cells might also secrete signals as part of their response, further reinforcing the signals and allowing them to propagate farther than would be possible by diffusion alone.

Materials and methods

Cloning

pEGFP-C3-YAP5SA plasmids were generated from the pEGFP-C3-hYAP1 plasmid gifted from Marius Sudol (Addgene plasmid # 17843; RRID:Addgene_17843 Basu et al., 2003). YAP amino acids S61, S109, S127, S164, S381 were all mutated to Alanine using the QuikChange II XL Site-Directed Mutagenesis Kit (Agilent). The following nuclear localization signal atggatccgaagaaaaaacgtaaaggccgtatggatccgaagaaaaaacgtaaagg-ccgt was appended to the 3’-end of GFP in the pAcGFP1-C3 vector (Clontech) to ensure its nuclear localization.

Cell line generation

All cell lines were maintained in DMEM (Gibco, 10569010) supplemented with 10% FBS (Gibco), 1% P/S (Gibco) unless otherwise stated. Dox-inducible HEK293 cell lines of nGFP or GFP-tagged YAP5SA were generated by sub-cloning either construct into the pcDNA5/FRT expression vector (Invitrogen) then co-transfecting the generated plasmids with the pOG44 Flp-Recombinase Expression Vector to ensure recombination into the genome of Flipin-TRex −293 cells (Invitrogen). Selection with Hygromycin and single clone isolation was followed by a test for Zeocin-sensitivity to ensure proper integration into the genome. Cells were maintained in Doxycycline unless otherwise specified.

YAPsh and control constructs were generously provided by Taran Gujral (Gujral and Kirschner, 2017). After transfection of HEK293 cells (ATCC Cat# CRL-1573, RRID:CVCL_0045) using PEI-40 and antibiotic selection in Puromycin for two weeks, isolated single colonies were observed and were individually isolated. To label cells with nuclear localized mCherry we used Lentiviral infection. First nuclear-mCherry was excised from pBRY-nuclear mCherry-IRES-PURO (Addgene plasmid #52409, RRID:Addgene_52409) by digestion with EcoRI and NotI and ligated into lentiviral expression plasmid pLVX-Ef1α-N1-mCherry. Lentiviruses were then produced via Fugene six co-transfection of psPAX2 (Addgene plasmid #12260, RRID:Addgene_12260) and pMD2.G (Addgene plasmid #12259, RRID:Addgene_12259) into 293 T cells ((ATCC Cat# CRL-11268, RRID:CVCL_1926). The harvested lentivirus was then used to infect the desired cell line. One week after cells were infected with Lentivirus, they were FACS sorted for mCherry expression.

Transwell assays

Transwell assays were done using 75 mm Transwell plates with a 0.4 µm Pore Polycarbonate Membrane Insert (Corning). Recipient cells were cultured on the bottom of the plates while feeder cells were cultured on the insert for 72 hr.

Antibodies and recombinant proteins

Neutralization antibodies against CYR61 (mainly NOVUS biological NB100-356, RRID:AB_10000986 raised against a peptide of amino acids 150–250 of human CYR61, validated independently with Abnova PAB12526, RRID:AB_10550916 raised against a peptide of amino acids 250–350 of human CYR61) or CTGF (Peprotech 500-P252, RRID:AB_1268211) were used at the indicated concentrations in the respective figures. A polyclonal Rabbit IgG Antibody was used as a control (Protein tech; 30000–0-AP, RRID:AB_2819035). Recombinant proteins CTGF (Peprotech), CYR61 (Peprotech) and Insulin (Peprotech) were used at the concentrations and under the conditions indicated in the respective figure legends.

Conditioned medium production

FlipinTrex-cells were cultured in Nunc TripleFlasks with DMEM plus 10% FBS and 1% P/S until they reached 100% confluence. Cells were then washed in PBS and kept in medium with 0.5% FBS and Dox for 6 days. Medium was collected, spun down and the supernatant then filtered on a 0.2 μm membrane to remove cell debris. The medium was concentrated using 3 kD cut-off Centricon Plus-70 filter devices (Millipore). The concentrate was kept at −20°C until further use. For protein concentration estimation, the concentrate was diluted to within working range before a BCA assay (Thermo Scientific) was performed against a standard.

Incycyte-based image acquisition and automated image analysis for nuclear segmentation

Images from four different regions per well were acquired at regular intervals using a 10x lens (Nikon, MRH00101) from cells cultured on 96-well plates (Eppendorf) in an Incucyte Zoom apparatus installed in a regular tissue culture incubator. After image acquisition, the red signal was used for nuclear segmentation using Incucyte ZOOM’s automated image analysis software. Images from all wells and across the entire time course were segmented using the same parameters. The number of red objects was used to represent the total number of nuclei. The average area estimated from the segmented objects in the red channel represent the average nuclear area.

Quantitative phase microscopy (QPM)

50,000 Flipin-Trex-293 cells were seeded on a Fibronectin-coated (50 ug/ml) glass-bottom 6-well plate (Cellvis) ~18 hr before imaging. This ensured that cells were at a low-enough density to allow high-quality segmentation of ~4000 single cells per well. Cells were cultured with 2 μg/ml Hoechst for 30’ before imaging. Imaging and analysis was done as described in Liu et al. (2019). Briefly Cells were maintained at 36°C in 5% CO2. The phase images were taken using a 10x lens of N.A. 0.30 (Nikon, MRH10101) using the SID4BIO camera (Phasics) integrated into a Nikon Eclipse Ti microscope with a halogen lamp as the transmitted light source. A C-HGFI mercury lamp was used for fluorescence illumination. All the image processing was done using a custom MATLAB script executed on a high-performance computer cluster.

Data plotting and fitting

Data were imported into Prism7 (Graph Pad) where their visualization, fitting and statistical analysis was done.

Coulter counter measurement

Cell volume was measured on a Coulter Counter (Beckman Coulter). For the measurement cells were trypsinized to single cells then resuspended in 10 ml Isoton diluent buffer (Beckman Coulter).

Total protein/DNA staining

Cells were trypsinized to a single-cell suspension. 10^{^6} cells were counted per condition, thoroughly washed in PBS and fixed for 20 min in ice cold, 4% PFA solution. Cells were then washed in PBS and permeabilized using 0.5% Triton-X for 10 min at room temperature. After repeated washes, cells were kept on ice with a 10 μg/ml Hoechst 3342 and/or 0.4 μg/ml of Alexa Fluor 647 Succinimidyl-Ester (Invitrogen) in a light protected tube for 45 min before they were analyzed on an LSRII flow cytometer (BD Biosciences).

RNA-Sequencing and data analysis

Flipin-Trex-293 cells expressing Tet-inducible YAP5SA or nGFP were grown to confluence in medium supplemented with Tet-approved FBS (Takara) supplemented with 0 or 5 ng/ml of Dox. Total RNA was harvested and isolated using Qiagen’s RNeasy mini kit. Samples were further processed at Harvard’s Biopolymer facility as follows: After polyA enrichment of mRNA using Takara’s PrepX PolyA mRNA Isolation Kit, libraries were prepared using Illumina’s PrepX RNA-Seq Library reagent kit on the Apollo 324 system. Samples were pooled then single-end sequenced on the Illumina NextSeq 500 platform. Sequencing alignment was done using STAR (Dobin et al., 2013) and transcriptome assembly and differential expression was done using Cufflinks (Trapnell et al., 2012) on a computer cluster. Each comparison group contained at least two independent biological repeats. Data is deposited in GEO under accession number: GSE123296.

Mass spectrometry and data analysis

Cells were kept under similar conditions as those used for RNA-sequencing. Cellular protein lysate was collected and snap frozen in urea-free lysis buffer supplemented with protease and phosphatase inhibitors (Roche). The supernatant was collected and filtered on a low protein-binding PVDF 0.22 μm membrane to eliminate cell debris, then snap frozen with protease inhibitors (Roche) for further processing. Lysates were reduced with 5 mM DTT, alkylated with 15 mM N-ethylmaleimide for 30 min in the dark. Alkylation reactions were quenched with 50 mM freshly prepared DTT, and proteins precipitated by methanol/chloroform precipitation. Digests were carried out in 200 mM EPPS pH 8.5 in presence of 2% acetonitrile (v/v) with LysC (Wako, 2 mg/ml, used 1:75) for 3 hr at room temperature and the subsequent addition of trypsin (Promega #V5111, stock 1:75) over night at 37°C.

Missed cleavage rate was assayed from a small aliquot by mass spectrometry. For whole proteome analysis, digests were directly labeled with TMT reagents (Thermo Fisher Scientific). Labeling efficiency and TMT ratios were assayed by mass spectrometry, while labeling reactions were stored at −80°C. After quenching of TMT labeling reactions with hydroxylamine, TMT labeling reactions were mixed, solvent evaporated to near completion, and TMT labeled peptides purified and desalted by acidic reversed phase C₁₈ chromatography. Peptides were then fractionated by alkaline reversed phase chromatography into 96 fractions and combined into 24 samples. Before mass spectrometric analysis, peptides were desalted over Stage Tips (Rappsilber et al., 2003).

Data were collected by a MultiNotch SPS MS³ method (McAlister et al., 2014) using an Orbitrap Lumos mass spectrometer (Thermo Fisher Scientific) coupled to a Proxeon EASY-nLC 1000 liquid chromatography (LC) system (Thermo Fisher Scientific). The 100 μm inner diameter capillary column used was packed with C₁₈ resin (Accucore 2.6 μm, 150 Å, Thermo Fisher Scientific). Peptides of each fraction were separated over 3–5 hr acidic acetonitrile gradients by LC prior to mass spectrometry (MS) injection. The first sequence scan was an MS¹ spectrum (Orbitrap analysis; resolution 120,000; mass range 400–1400 Th). MS² analysis followed collision-induced dissociation (CID, CE = 35) with a maximum ion injection time of 150 ms and an isolation window of 0.7 Da. To obtain quantitative information, MS³ precursors were fragmented by high-energy collision-induced dissociation (HCD) and analyzed in the Orbitrap at a resolution of 50,000 at 200 Th.

MS³ injection time for phosphopeptides was 150 and 200 ms at a resolution of 50,000. Further details on LC and MS parameters and settings used were described recently (Paulo et al., 2016a).

Peptides were searched with a SEQUEST-based in-house software against a human database with a target decoy database strategy and a false discovery rate (FDR) of 1% set for peptide-spectrum matches following filtering by linear discriminant analysis (LDA) and a final collapsed protein-level FDR of 1%. Quantitative information on peptides was derived from MS³ scans. Quant tables were generated requiring an MS² isolation specificity of >75% for each peptide and a sum of TMT s/n of >0 over all channels for any given peptide and exported as TAB-separated raw text data. The peptide-level sums of TMT s/n signal were integrated into protein-level and converted into relative ratio with 90% confidence intervals using a recently developed method BACIQ - a Bayesian approach to confidence inference for quantitative proteomics. Briefly, this approach reconciles disagreement across multiple peptides and differences in absolute levels of peptide signal, reporting the ratios across conditions for a given protein, modeling quantitative proteomics measurement as a random variable distributed according to hierarchical Dirichlet-Multinomial probability distribution. This method accounts for the absolute level of the peptide signal in a way calibrated for a particular MS instrument. More specifically we used a charge conversion value of 1.7 as previously fitted by us for this instrument and mass resolution (Peshkin et al., 2019). Details of the TMT intensity quantification method and further search parameters applied were described in Paulo et al. (2016b).

Gene ontology analysis was done using DAVID (Huang et al., 2009) using all identified proteins as background.

Acknowledgements

We thank Mike Gage for his meticulous assistance with the preparation of the protein samples for mass spectrometry and his overall management of project resources. We thank Jodene Moore of the SysBio FACS facility for her indispensable help setting up FACS analyses on the facility machines. We also thank the BPF Next-Gen Sequencing Core Facility at Harvard Medical School for their expertise and instrument availability that supported this work. We are grateful to the Nikon Imaging Center at Harvard Medical School for sharing resources; and to the Research Computing Group for support with cluster computing for image processing, RNAseq data processing and data transfer. We are grateful for funding from HMS through the Dean’s Innovation grant. Finally, we are grateful to Shangqin Guo and Amaleah Hartman for the stimulating discussions and thoughtful critique of this paper.

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Contributor Information

Marc W Kirschner, Email: marc@hms.harvard.edu.

Duojia Pan, UT Southwestern Medical Center and HHMI, United States.

Marianne E Bronner, California Institute of Technology, United States.

Funding Information

This paper was supported by the following grants:

Harvard Medical School Dean's Innovation Grant to Marc W Kirschner.
National Institute of General Medical Sciences GM026875 to Marc W Kirschner.
National Institute of Child Health and Human Development HD091846 to Marc W Kirschner.

Additional information

Competing interests

No competing interests declared.

Author contributions

Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing - review and editing.

Formal analysis, Methodology, Writing - review and editing.

Methodology, Writing - review and editing.

Formal analysis, Software, Writing - review and editing.

Conceptualization, Funding acquisition, Writing - review and editing.

Additional files

Supplementary file 1. Table with protein and mRNA changes.

elife-53404-supp1.xlsx^{(588.4KB, xlsx)}

Transparent reporting form

elife-53404-transrepform.docx^{(246.2KB, docx)}

Data availability

Sequencing data has been deposited in GEO under accession code GSE123296.

The following dataset was generated:

Douaa Mugahid. 2018. Induction of YAP5SA-GFP vs nGFP expression in FlipinTrex293 cells. NCBI Gene Expression Omnibus. GSE123296

References

Anzi S, Stolovich-Rain M, Klochendler A, Fridlich O, Helman A, Paz-Sonnenfeld A, Avni-Magen N, Kaufman E, Ginzberg MB, Snider D, Ray S, Brecht M, Holmes MM, Meir K, Avivi A, Shams I, Berkowitz A, Shapiro AMJ, Glaser B, Ben-Sasson S, Kafri R, Dor Y. Postnatal exocrine pancreas growth by cellular hypertrophy correlates with a shorter lifespan in mammals. Developmental Cell. 2018;45:726–737. doi: 10.1016/j.devcel.2018.05.024. [DOI] [PubMed] [Google Scholar]
Aragona M, Panciera T, Manfrin A, Giulitti S, Michielin F, Elvassore N, Dupont S, Piccolo S. A mechanical checkpoint controls multicellular growth through YAP/TAZ regulation by actin-processing factors. Cell. 2013;154:1047–1059. doi: 10.1016/j.cell.2013.07.042. [DOI] [PubMed] [Google Scholar]
Basu S, Totty NF, Irwin MS, Sudol M, Downward J. Akt phosphorylates the Yes-associated protein, YAP, to induce interaction with 14-3-3 and attenuation of p73-mediated apoptosis. Molecular Cell. 2003;11:11–23. doi: 10.1016/S1097-2765(02)00776-1. [DOI] [PubMed] [Google Scholar]
Cadart C, Monnier S, Grilli J, Sáez PJ, Srivastava N, Attia R, Terriac E, Baum B, Cosentino-Lagomarsino M, Piel M. Size control in mammalian cells involves modulation of both growth rate and cell cycle duration. Nature Communications. 2018;9:e3275. doi: 10.1038/s41467-018-05393-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
Chen CC, Mo FE, Lau LF. The angiogenic factor Cyr61 activates a genetic program for wound healing in human skin fibroblasts. Journal of Biological Chemistry. 2001;276:47329–47337. doi: 10.1074/jbc.M107666200. [DOI] [PubMed] [Google Scholar]
Chen C-C, Kim K-H, Lau LF. The matricellular protein CCN1 suppresses hepatocarcinogenesis by inhibiting compensatory proliferation. Oncogene. 2016;35:1314–1323. doi: 10.1038/onc.2015.190. [DOI] [PMC free article] [PubMed] [Google Scholar]
Diril MK, Ratnacaram CK, Padmakumar VC, Du T, Wasser M, Coppola V, Tessarollo L, Kaldis P. Cyclin-dependent kinase 1 (Cdk1) is essential for cell division and suppression of DNA re-replication but not for liver regeneration. PNAS. 2012;109:3826–3831. doi: 10.1073/pnas.1115201109. [DOI] [PMC free article] [PubMed] [Google Scholar]
Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29:15–21. doi: 10.1093/bioinformatics/bts635. [DOI] [PMC free article] [PubMed] [Google Scholar]
Dong J, Feldmann G, Huang J, Wu S, Zhang N, Comerford SA, Gayyed MF, Anders RA, Maitra A, Pan D. Elucidation of a universal size-control mechanism in Drosophila and mammals. Cell. 2007;130:1120–1133. doi: 10.1016/j.cell.2007.07.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
Dupont S, Morsut L, Aragona M, Enzo E, Giulitti S, Cordenonsi M, Zanconato F, Le Digabel J, Forcato M, Bicciato S, Elvassore N, Piccolo S. Role of YAP/TAZ in mechanotransduction. Nature. 2011;474:179–183. doi: 10.1038/nature10137. [DOI] [PubMed] [Google Scholar]
Ernst S, Demirci C, Valle S, Velazquez-Garcia S, Garcia-Ocaña A. Mechanisms in the adaptation of maternal β-cells during pregnancy. Diabetes Management. 2011;1:239–248. doi: 10.2217/dmt.10.24. [DOI] [PMC free article] [PubMed] [Google Scholar]
Fankhauser G. Maintenance of normal structure in Heteroploid Salamander Larvae, through compensation of changes in cell size by adjustment of cell number and cell shape. Journal of Experimental Zoology. 1945;100:445–455. doi: 10.1002/jez.1401000310. [DOI] [PubMed] [Google Scholar]
Gilgenkrantz H, Collin de l'Hortet A. Understanding liver regeneration: from mechanisms to regenerative medicine. The American Journal of Pathology. 2018;188:1316–1327. doi: 10.1016/j.ajpath.2018.03.008. [DOI] [PubMed] [Google Scholar]
Ginzberg MB, Chang N, D'Souza H, Patel N, Kafri R, Kirschner MW. Cell size sensing in animal cells coordinates anabolic growth rates and cell cycle progression to maintain cell size uniformity. eLife. 2018;7:e26957. doi: 10.7554/eLife.26957. [DOI] [PMC free article] [PubMed] [Google Scholar]
Grijalva JL, Huizenga M, Mueller K, Rodriguez S, Brazzo J, Camargo F, Sadri-Vakili G, Vakili K. Dynamic alterations in hippo signaling pathway and YAP activation during liver regeneration. American Journal of Physiology-Gastrointestinal and Liver Physiology. 2014;307:G196–G204. doi: 10.1152/ajpgi.00077.2014. [DOI] [PubMed] [Google Scholar]
Groupe V, Manaker RA. Discrete foci of altered chicken embryo cells associated with rous sarcoma virus in tissue culture. Virology. 1956;2:838–840. doi: 10.1016/0042-6822(56)90064-2. [DOI] [PubMed] [Google Scholar]
Gujral TS, Kirschner MW. Hippo pathway mediates resistance to cytotoxic drugs. PNAS. 2017;114:E3729–E3738. doi: 10.1073/pnas.1703096114. [DOI] [PMC free article] [PubMed] [Google Scholar]
Hansen CG, Ng YL, Lam WL, Plouffe SW, Guan KL. The hippo pathway effectors YAP and TAZ promote cell growth by modulating amino acid signaling to mTORC1. Cell Research. 2015;25:1299–1313. doi: 10.1038/cr.2015.140. [DOI] [PMC free article] [PubMed] [Google Scholar]
Holbourn KP, Acharya KR, Perbal B. The CCN family of proteins: structure-function relationships. Trends in Biochemical Sciences. 2008;33:461–473. doi: 10.1016/j.tibs.2008.07.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
Huang daW, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nature Protocols. 2009;4:44–57. doi: 10.1038/nprot.2008.211. [DOI] [PubMed] [Google Scholar]
Kafri R, Levy J, Ginzberg MB, Oh S, Lahav G, Kirschner MW. Dynamics extracted from fixed cells reveal feedback linking cell growth to cell cycle. Nature. 2013;494:480–483. doi: 10.1038/nature11897. [DOI] [PMC free article] [PubMed] [Google Scholar]
Kamińska K, Szczylik C, Bielecka ZF, Bartnik E, Porta C, Lian F, Czarnecka AM. The role of the cell-cell interactions in Cancer progression. Journal of Cellular and Molecular Medicine. 2015;19:283–296. doi: 10.1111/jcmm.12408. [DOI] [PMC free article] [PubMed] [Google Scholar]
Katsube K, Sakamoto K, Tamamura Y, Yamaguchi A. Role of CCN, a vertebrate specific gene family, in development. Development, Growth & Differentiation. 2009;51:55–67. doi: 10.1111/j.1440-169X.2009.01077.x. [DOI] [PubMed] [Google Scholar]
Kim NG, Gumbiner BM. Adhesion to fibronectin regulates hippo signaling via the FAK-Src-PI3K pathway. The Journal of Cell Biology. 2015;210:503–515. doi: 10.1083/jcb.201501025. [DOI] [PMC free article] [PubMed] [Google Scholar]
Lane SW, Williams DA, Watt FM. Modulating the stem cell niche for tissue regeneration. Nature Biotechnology. 2014;32:795–803. doi: 10.1038/nbt.2978. [DOI] [PMC free article] [PubMed] [Google Scholar]
Latinkic BV, Mercurio S, Bennett B, Hirst EM, Xu Q, Lau LF, Mohun TJ, Smith JC. Xenopus Cyr61 regulates gastrulation movements and modulates wnt signalling. Development. 2003;130:2429–2441. doi: 10.1242/dev.00449. [DOI] [PubMed] [Google Scholar]
Leask A, Abraham DJ. All in the CCN family: essential matricellular signaling modulators emerge from the bunker. Journal of Cell Science. 2006;119:4803–4810. doi: 10.1242/jcs.03270. [DOI] [PubMed] [Google Scholar]
Lee MJ, Byun MR, Furutani-Seiki M, Hong JH, Jung HS. YAP and TAZ regulate skin wound healing. Journal of Investigative Dermatology. 2014;134:518–525. doi: 10.1038/jid.2013.339. [DOI] [PubMed] [Google Scholar]
Lin MT, Chang CC, Chen ST, Chang HL, Su JL, Chau YP, Kuo ML. Cyr61 expression confers resistance to apoptosis in breast Cancer MCF-7 cells by a mechanism of NF-kappaB-dependent XIAP up-regulation. Journal of Biological Chemistry. 2004;279:24015–24023. doi: 10.1074/jbc.M402305200. [DOI] [PubMed] [Google Scholar]
Liu X, Oh S, Kirschner MW. Optimization of quantitative phase microscopy reveals new features of the growth rate of adherent mammalian cells. bioRxiv. 2019 doi: 10.1101/631119. [DOI]
Lu L, Finegold MJ, Johnson RL. Hippo pathway coactivators yap and taz are required to coordinate mammalian liver regeneration. Experimental & Molecular Medicine. 2018;50:e423. doi: 10.1038/emm.2017.205. [DOI] [PMC free article] [PubMed] [Google Scholar]
McAlister GC, Nusinow DP, Jedrychowski MP, Wühr M, Huttlin EL, Erickson BK, Rad R, Haas W, Gygi SP. MultiNotch MS3 enables accurate, sensitive, and multiplexed detection of differential expression across Cancer cell line proteomes. Analytical Chemistry. 2014;86:7150–7158. doi: 10.1021/ac502040v. [DOI] [PMC free article] [PubMed] [Google Scholar]
Mooring M, Fowl BH, Lum SZC, Liu Y, Yao K, Softic S, Kirchner R, Bernstein A, Singhi AD, Jay DG. Hepatocyte stress increases expression of YAP and TAZ in hepatocytes to promote parenchymal inflammation and fibrosis. Hepatology. 2020 doi: 10.1002/hep.30928. [DOI] [PMC free article] [PubMed] [Google Scholar]
Paulo JA, O'Connell JD, Everley RA, O'Brien J, Gygi MA, Gygi SP. Quantitative mass spectrometry-based multiplexing compares the abundance of 5000 S. cerevisiae proteins across 10 carbon sources. Journal of Proteomics. 2016a;148:85–93. doi: 10.1016/j.jprot.2016.07.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
Paulo JA, O’Connell JD, Gygi SP. A Triple Knockout (TKO) Proteomics Standard for Diagnosing Ion Interference in Isobaric Labeling Experiments. Journal of The American Society for Mass Spectrometry. 2016b;27:1620–1625. doi: 10.1007/s13361-016-1434-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
Penzo-Méndez AI, Stanger BZ. Organ-Size regulation in mammals. Cold Spring Harbor Perspectives in Biology. 2015;7:a019240. doi: 10.1101/cshperspect.a019240. [DOI] [PMC free article] [PubMed] [Google Scholar]
Peshkin L, Gupta M, Ryazanova L, Wühr M. Bayesian confidence intervals for multiplexed proteomics integrate Ion-statistics with peptide quantification concordance. Molecular & Cellular Proteomics. 2019;18:2108–2120. doi: 10.1074/mcp.TIR119.001317. [DOI] [PMC free article] [PubMed] [Google Scholar]
Rappsilber J, Ishihama Y, Mann M. Stop and go extraction tips for matrix-assisted laser desorption/ionization, Nanoelectrospray, and LC/MS sample pretreatment in proteomics. Analytical Chemistry. 2003;75:663–670. doi: 10.1021/ac026117i. [DOI] [PubMed] [Google Scholar]
Shu Z, Deng W-M. Differential regulation of cyclin E by Yorkie-Scalloped signaling in organ development. G3: Genes|Genomes|Genetics. 2017;7:1049–1060. doi: 10.1534/g3.117.039065. [DOI] [PMC free article] [PubMed] [Google Scholar]
Stoker MG, Rubin H. Density dependent inhibition of cell growth in culture. Nature. 1967;215:171–172. doi: 10.1038/215171a0. [DOI] [PubMed] [Google Scholar]
Su AI, Guidotti LG, Pezacki JP, Chisari FV, Schultz PG. Gene expression during the priming phase of liver regeneration after partial hepatectomy in mice. PNAS. 2002;99:11181–11186. doi: 10.1073/pnas.122359899. [DOI] [PMC free article] [PubMed] [Google Scholar]
Tapon N, Harvey KF, Bell DW, Wahrer DC, Schiripo TA, Haber D, Hariharan IK. Salvador promotes both cell cycle exit and apoptosis in Drosophila and is mutated in human Cancer cell lines. Cell. 2002;110:467–478. doi: 10.1016/S0092-8674(02)00824-3. [DOI] [PubMed] [Google Scholar]
Tian X, Azpurua J, Hine C, Vaidya A, Myakishev-Rempel M, Ablaeva J, Mao Z, Nevo E, Gorbunova V, Seluanov A. High-molecular-mass hyaluronan mediates the Cancer resistance of the naked mole rat. Nature. 2013;499:346–349. doi: 10.1038/nature12234. [DOI] [PMC free article] [PubMed] [Google Scholar]
Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, Pimentel H, Salzberg SL, Rinn JL, Pachter L. Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and cufflinks. Nature Protocols. 2012;7:562–578. doi: 10.1038/nprot.2012.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
Verhulst PF. Recherches mathématiques sur la loi d’accroissement de la population. Nouv Mém Acad R. Sci B.-Lett. Brux. 1845:1–41. [Google Scholar]
Volarevic S, Stewart MJ, Ledermann B, Zilberman F, Terracciano L, Montini E, Grompe M, Kozma SC, Thomas G. Proliferation, but not growth, blocked by conditional deletion of 40S ribosomal protein S6. Science. 2000;288:2045–2047. doi: 10.1126/science.288.5473.2045. [DOI] [PubMed] [Google Scholar]
Weinberg RA. The Biology of Cancer. New York: Garland Science; 2007. [Google Scholar]
Xin M, Kim Y, Sutherland LB, Qi X, McAnally J, Schwartz RJ, Richardson JA, Bassel-Duby R, Olson EN. Regulation of insulin-like growth factor signaling by yap governs cardiomyocyte proliferation and embryonic heart size. Science Signaling. 2011;4:ra70. doi: 10.1126/scisignal.2002278. [DOI] [PMC free article] [PubMed] [Google Scholar]
Yang X, Xu T. Molecular mechanism of size control in development and human diseases. Cell Research. 2011;21:715–729. doi: 10.1038/cr.2011.63. [DOI] [PMC free article] [PubMed] [Google Scholar]
Yimlamai D, Christodoulou C, Galli GG, Yanger K, Pepe-Mooney B, Gurung B, Shrestha K, Cahan P, Stanger BZ, Camargo FD. Hippo pathway activity influences liver cell fate. Cell. 2014;157:1324–1338. doi: 10.1016/j.cell.2014.03.060. [DOI] [PMC free article] [PubMed] [Google Scholar]
Zanconato F, Forcato M, Battilana G, Azzolin L, Quaranta E, Bodega B, Rosato A, Bicciato S, Cordenonsi M, Piccolo S. Genome-wide association between YAP/TAZ/TEAD and AP-1 at enhancers drives oncogenic growth. Nature Cell Biology. 2015;17:1218–1227. doi: 10.1038/ncb3216. [DOI] [PMC free article] [PubMed] [Google Scholar]
Zhang J, Ji JY, Yu M, Overholtzer M, Smolen GA, Wang R, Brugge JS, Dyson NJ, Haber DA. YAP-dependent induction of amphiregulin identifies a non-cell-autonomous component of the hippo pathway. Nature Cell Biology. 2009;11:1444–1450. doi: 10.1038/ncb1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
Zhao B, Wei X, Li W, Udan RS, Yang Q, Kim J, Xie J, Ikenoue T, Yu J, Li L, Zheng P, Ye K, Chinnaiyan A, Halder G, Lai ZC, Guan KL. Inactivation of YAP oncoprotein by the hippo pathway is involved in cell contact inhibition and tissue growth control. Genes & Development. 2007;21:2747–2761. doi: 10.1101/gad.1602907. [DOI] [PMC free article] [PubMed] [Google Scholar]
Zhao B, Li L, Tumaneng K, Wang CY, Guan KL. A coordinated phosphorylation by lats and CK1 regulates YAP stability through SCF(beta-TRCP) Genes & Development. 2010;24:72–85. doi: 10.1101/gad.1843810. [DOI] [PMC free article] [PubMed] [Google Scholar]

eLife. doi: 10.7554/eLife.53404.sa1

Decision letter

Editor: Duojia Pan¹

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

The Hippo signaling pathway regulates organ size by controlling the activity of the transcription factor, YAP (Yorkie in Drosophila). Despite its well-established role in organ size regulation, how YAP coordinately regulates cell number and cell size remains unclear. In this study, the authors dissected the effect of YAP on cell proliferation, cell death and cell size by carefully measuring the growth of cultured mammalian cells at different densities. They show that YAP can regulate these parameters through independent circuits. While YAP stimulates cell proliferation in a cell-autonomous manner, it increases cell size and cell survival largely through diffusible, non-cell autonomous factors. In particular, their studies implicate CYR61, a canonical YAP target gene, as a YAP-induced secreted factor that limits apoptosis in high density cultures, while another unidentified secreted factor likely mediates YAP-dependent increase in cell size. Overall, this study provides a clearer understanding of YAP's role in regulating cell growth on a cellular as well as population level, and highlights the importance of non-cell autonomous signaling downstream of YAP.

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting your work entitled "YAP independently regulates cell size and population growth dynamics via non-cell autonomous mediators" for consideration by eLife. Your article has been reviewed by three peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by a Senior Editor. The reviewers have opted to remain anonymous.

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we regret to inform you that your work will not be considered further for publication in eLife. While the non-autonomous effect of YAP on cell size and cell number is potentially interesting, the reviewers have raised significant technical concerns, detailed below, that have dampened their enthusiasm.

Reviewer #1:

YAP is a potent oncogene and is known to regulate both cell growth and proliferation. However, the underlying circuitry by which YAP impacts cell size and cell number is not well understood. In this study, Mugahid, et al. report the interesting finding that in cultured cells, YAP can regulate cell size and cell number independently. They further implicate CYR61, a known secreted YAP target, as the major non-cell autonomous regulator of cell number. These findings are potentially interesting. However, I hope the authors can address the following points.

1) All the analyses presented in this paper have been based on a single cell line. This raises the question of whether the observations are unique to HEK293 cells. Since the authors imply their findings as a general mechanism, it is important to conduct similar analyses in at least one additional cell line.

2) The authors used blocking antibodies to examine the relative contribution of CYR61 versus CTGF in mediating the non-autonomous effect of YAP5SA. Have these antibodies been validated for their blocking activities? I cannot find this information in the main text or the Materials and methods section.

3) At times I find the paper difficult to follow logically. For example, early in the paper the authors stated that the cell size effect is best assessed at low cell density, leading me to assume that all follow-up analyses of cell size would be carried out at low cell density. However, some of the later studies appear to be done using high cell density (e.g., the co-culture experiment). To avoid confusions, it would be helpful for the authors to be explicit about how and why each experiment is done as the way it was described. Also, many figure legends are written in the style of conclusions, with little information on how the data are generated. Overall, I feel that the presentation can be improved with better rationales and interpretations.

Reviewer #2:

The authors investigate whether Hippo signaling affects organ size regulation by increasing cell size, cell proliferation, or a combination of both. In particular, the authors argue that YAP activation increases both cell size as well as total cell number at contact inhibition. They further find that the effect of YAP activation acts non-autonomously via a secreted factor CYR61. Since CYR61 has little effect on cell size, they conclude that YAP's effect on cell size and cell number act through separable mechanisms. This paper is potentially interesting, but I have serious technical concerns that significantly dampen my enthusiasm.

There is no indication that the sample size numbers are biological replicates. Much of the reported data appear to be averages of the same experiment conducted on the same clone. Sample sizes should indicate biological replicates not technical replicates (n = 1). This is of critical importance when measuring the cell size of clonal lines. For example, my lab has observed individual clonal lines exhibiting different size homeostasis despite being derived from the same parental line. Therefore, the size increase of nGFP expressing cells compared to WT cells could be due to clonal abnormalities. These experiments should be replicated using multiple clones, and statistically compared to WT cells without co-culture.

A major unaddressed technical problem is the variability of SE dye as well as nuclear area as quantitative measurements of cell size. In our hands, the SE signal depends on the density of cell suspensions at the time of reaction. Cells from the same population fixed and diluted to dense or sparse suspensions and reacted with the same molarity of SE dye yield dramatically different dye signals (denser suspension having lower dye signal). Though it has been shown that within a single sample preparation, SE signal correlate with cell size, it is not generalizable across samples. Therefore, it is crucial for the proper interpretation the size phenotypes that appropriate SE dye calibrations are done to show that the preparation-dependent effects do not confound the results. Similarly, while it has been shown for many cell types that nuclear area linearly scales with overall cell volume, it is unclear how the increase in cell density as contact inhibition occurs may change 3D nuclear morphology and therefore the nuclear area v. cell volume linear relationship. If this linear scaling factor changes during contact inhibition, nuclear area cannot be compared across the growth curve. Especially since the cell size effects under discussion are subtle (<30% change), more detailed controls are needed.

At many points, the figure legends are insufficient for understanding what experiments were performed. The manuscript does not indicate the conditions in which cells were collected for flow analysis (e.g. SE-dye stains). A major point of Figure 1 is that cell size is dependent on culture density, but the density of the cultures used is not explicitly denoted throughout Figures 2-5 or anywhere in the text or Materials and methods.

The authors show no relationship between YAP overactivity and k (Figure 2E), but do show a relationship between CYR61 neutralization and k (Figure 4F). The apparent contradiction between YAP not being necessary for regulating growth rate and CYR61 being sufficient to alter growth rate is not addressed in the manuscript and could be related to the technical concerns listed above.

The main experiments the authors use to argue for non-cell autonomous effects of YAP is via co-culture and neutralizing antibody experiments. Firstly, co-cultures results can be confounded by complex mechanical constraints, especially if the two cell types have different cell sizes, as shown in Figure 2. Secondly, the authors did not show the effect of the anti-CYR61 antibody on wild-type cells (Figure 4A, D), and therefore it is difficult to interpret whether its effects are specific. It would be important to show these controls. Lastly, a complementary experiment showing that conditioned media produced by YAP5A cells can lead to similar effect in increasing k or Ymax in recipient wild-type cells would argue that phenotypes observed in co-culture are indeed mediated through secreted ligands.

Reviewer #3:

The manuscript aims to provide insight into the role of YAP transcription factor in determining the size of organs. It reports the results of in vitro experiments with cultured HEK293 cells, which includes the identification of a certain secreted factor (a previously known target of YAP) as a mediator of cell proliferation, controlling the limiting density of a confluent cell culture. Other conclusions and interpretation are however quite equivocal. Overall, in my opinion, this manuscript is substantially below the standard of eLife.

The manuscript aims to disentangle regulation of cell size and cell number, but lacks quantitation to do it in a convincing way. Figure 1A-D suggests a simple picture: cell area is constant while cells proliferate before reaching confluence, but goes down after achieving confluence. Rate of proliferation goes to zero as cells reach minimal area. The authors use a logistic equation to describe growth and the resulting cell number with two parameters: one for the rate of growth and the other for the carrying capacity corresponding to the maximal cell density which in general depend on growth conditions and substrate. This is all reasonable. However, curiously, the manuscript fails to note that minimal cell area and the terminal density of cells (and hence the Ymax or "carrying capacity") are directly related (by inverse proportionality). Instead they write: "Since size at the later time points seems to be a complex product of initial size and the achieved Ymax, we focused on changes in nuclear area (cell size) in low density cultures."

There are two major points of confusion here. First: since in the confluent layer cell area and density are inversely proportional to each other, it is not clear what the authors mean by independent control of cell size and cell number. Second: physiology of non-confluent cells is typically quite different from the confluent ones: e.g. mesenchymal/epithelial, so one should not be surprised that factors controlling cell size at low density and the terminal density in confluency may be quite different. This muddies the water as far as the overarching issue of organ size control is concerned, since the paper strongly implies that cell size and the terminal density provide different channels of control. [[See for example the Abstract: "…is the major regulator of the non-cell autonomous increase in cell number, but does not affect cell size. The molecular identity of the non-cell autonomously acting mediator of cell size is yet to be identified."]

I suspect that confusion arises because literature on organ size determination often distinguishes between cell size control (equivalent to cell density) and cell number control with the latter referring to the total cell count in a variable size organ (such as Drosophila wing imaginal disc). In that case, cell area and total cell number are independent variables that together define the total area (or "size") of the organ. In contrast, since the in vitro well size is fixed, "Nuclei/Well" measurement performed by the authors report only the density. In the case of a fixed well area, cell size and number are not independent. The fundamental aspect of organ size determination – its global nature – is completely overlooked by the authors. Lacking confinement of the dish well, how do real organs know their size? This work doesn't give us a clue. The confusion involving cell size/cell density/cell number clouds the interpretation of the results and detracts also from the factual findings regarding YAP/secreted factors reported in the paper.

Section V focuses on the effect of substrate stiffness, which is known to affect the growth (and differentiation) of cell growing on them, both at low and high density. It is not surprising that substrate modulates all of the effects reported in the manuscript. What does this teach us about size determination? Relating the findings to the broader thought on role of mechanics in size determination, and what one should and should not expect to see with in vitro adherent cell cultures could avoid confusion. It doesn't help that the authors appear to be unaware of a number of studies addressing the role of mechanical interactions in organ size determination (in Drosophila) as well as contact inhibition (in the dish).

The Discussion section revisits 50 years of "contact inhibition" history while missing key ideas (and quantitative studies) on density dependent mobility and the effects of confinement, which would have helped with the logic of the study. Another area of study that is relevant, but overlooked, is the question of "cell competition" which is relevant both to contact inhibition and size determination. Secreted factors have long been suggested as a readouts of tissue mass. How does the secreted factor controlled by YAP relate to those ideas?

In addition to the overall logic of the work I have trouble with some important technical aspects.

1) Logistic fits shown in Figure 1—figure supplement 3C are quite poor at estimating saturation density Ymax. Since measurement of the latter is central to the analysis, I have very little confidence in its accuracy and validity. Furthermore, logistic description at best, kicks in upon confluence, which is reached at different times under different conditions (e.g. it should depend on cell size in the low density state), so an additional fitting parameter would seem to be in order. In either case, since the authors use cell size in a pre-confluent state, one would like to see a more careful/quantitative study of growth rate as a function of cell density (and cell size) in that pre-confluent phase of growth.

2) Do Figures 4A, D 5A, E; 6B, G present data or some sort of a fit (logistic?) to data. Error bars? (Error bars are also missing Figure 1—figure supplement 3. The Materials and methods section does not explain the method used for determining Nuclei/well measurement. Does it involve image segmentation? What are the errors? Could there be a systematic error increasing with increasing cell density?

3) The authors demonstrate the non-autonomous effect of YAP signaling by examining the dependence of Ymax on the fraction of YAP5SA cells in co-culture. Since the effectors are secreted/soluble factors, I would expect that their effect depends on the volume of the media. Does it?

eLife. 2020 Jan 9;9:e53404. doi: 10.7554/eLife.53404.sa2

Author response

[Editors’ note: what follows is the authors’ response to the first round of review.]

[…] While the non-autonomous effect of YAP on cell size and cell number is potentially interesting, the reviewers have raised significant technical concerns, detailed below, that have dampened their enthusiasm.

Reviewer #1:

YAP is a potent oncogene and is known to regulate both cell growth and proliferation. However, the underlying circuitry by which YAP impacts cell size and cell number is not well understood. In this study, Mugahid, et al. report the interesting finding that in cultured cells, YAP can regulate cell size and cell number independently. They further implicate CYR61, a known secreted YAP target, as the major non-cell autonomous regulator of cell number. These findings are potentially interesting. However, I hope the authors can address the following points.

1) All the analyses presented in this paper have been based on a single cell line. This raises the question of whether the observations are unique to HEK293 cells. Since the authors imply their findings as a general mechanism, it is important to conduct similar analyses in at least one additional cell line.

We have not established other isogenic cell lines to redo the analyses in their entirety. However, arguably the most important part of the study is to show that the non-autonomous response to YAP is not a special feature of HEK293 cells. We now show that 3T3 cells respond to YAP-induced secreted molecules by increasing their size and number when treated with YAP5SA-conditioned medium (Figure 2—figure supplement 1).

2) The authors used blocking antibodies to examine the relative contribution of CYR61 versus CTGF in mediating the non-autonomous effect of YAP5SA. Have these antibodies been validated for their blocking activities? I cannot find this information in the main text or the Materials and methods section.

The reviewer’s point is well-taken, as we had not included any such data in the original manuscript. We have now included supplementary data (Figure 5—figure supplement 1) which validate that: (1) recombinant CYR61 and CTGF are recognized by the respective antibodies and that (2) the same antibodies recognize the proteolytically cleaved (active) forms of CYR61 and CTGF in concentrates of conditioned medium from YAP5SA and nGFP-expressing cells. The data only shows that the antibody recognizes the correct protein but does not prove inactivation per say. As there seems to be no well-established assay for CYR61 activity at present, it was not possible to show direct inactivation. However, we do find the same results using a second antibody raised against a different epitope of the CYR61 protein making it unlikely that it’s simply an off-target effect (Figure 5—figure supplement 1I, HJ). We are further reassured that the results are CYR61- specific as addition of purified recombinant CYR61 to cells results in changes opposite to those seen with antibody treatment.

3) At times I find the paper difficult to follow logically. For example, early in the paper the authors stated that the cell size effect is best assessed at low cell density, leading me to assume that all follow-up analyses of cell size would be carried out at low cell density. However, some of the later studies appear to be done using high cell density (e.g., the co-culture experiment). To avoid confusions, it would be helpful for the authors to be explicit about how and why each experiment is done as the way it was described. Also, many figure legends are written in the style of conclusions, with little information on how the data are generated. Overall, I feel that the presentation can be improved with better rationales and interpretations.

We have included a more detailed explanation of each specific experiment both in the main text and in the figure legends. We realize that this was a general concern all reviewers shared. Regarding the particular comment on cell size measurements and density, most experiments are done at low density as mentioned in the main text, but we had mentioned there would be some exceptions. As a minimum number of cells is needed to effectively condition the culture medium over a certain course of time, co- cultures were an instance of the exception to the rule. In that case, the total number of cells was not just YAP5SA cells, and we observed that the non-cell autonomous effects on size were more prominent after longer culture times at which cell density is often higher than what we consider low density. We attribute that to the need for a longer time/higher number of cells before the soluble mediator accumulates to effective levels.

Reviewer #2:

The authors investigate whether Hippo signaling affects organ size regulation by increasing cell size, cell proliferation, or a combination of both. In particular, the authors argue that YAP activation increases both cell size as well as total cell number at contact inhibition. They further find that the effect of YAP activation acts non-autonomously via a secreted factor CYR61. Since CYR61 has little effect on cell size, they conclude that YAP's effect on cell size and cell number act through separable mechanisms. This paper is potentially interesting, but I have serious technical concerns that significantly dampen my enthusiasm.

There is no indication that the sample size numbers are biological replicates.

We assure the reviewer that the sample sizes reflect biological replicates when stated because these measurements were done on several independent live cell cultures in parallel. On the Incucyte we can measure the number of nuclei and their average area across multiple wells, and what we state as a biological replicate are the number of wells compared per group. We have always repeated the entire set of experiments at least once to confirm the robustness of the changes. In case of the SE-dye data which we reported in the first submission, we depicted histograms from a single population per condition and did not report any statistics. The data were a representative sample of experiments repeated at least twice using the same method and showed similar qualitative changes. For the QPM measurements, which we have now included and base our findings on, the replicates come from 3 different, yet isogenic clones.

We have clarified what n represent in the respective figure legends.

Much of the reported data appear to be averages of the same experiment conducted on the same clone. Sample sizes should indicate biological replicates not technical replicates (n = 1). This is of critical importance when measuring the cell size of clonal lines. For example, my lab has observed individual clonal lines exhibiting different size homeostasis despite being derived from the same parental line. Therefore, the size increase of nGFP expressing cells compared to WT cells could be due to clonal abnormalities. These experiments should be replicated using multiple clones, and statistically compared to WT cells without co-culture.

a) We of course agree. We addressed the issue of replicates in the prior point. However, the reviewer does raise a valid point on clonal differences, which are somewhat different from the issue of biological replicates. Having considered that issue ourselves, we made sure we used FlipinTrex-293 cells which allow the generation of isogenic, Dox-inducible cell lines for the bulk of our most informative experiments as indicated in the Materials and methods section. These are the cells we used for making comparisons between the 5SA vs. nGFP cell lines and anything depending on them. This eliminates the problem of the variable site of genomic integration of the plasmid in the different clones, which, as the reviewer points out, could make the comparisons less reliable if the cell lines were not isogenic. We have now explained the rationale for that choice more thoroughly in the main text and not just the Materials and methods. To further confirm the soundness of our choice, and because we take the reviewer’s concerns quite seriously considering the importance of these findings, we went back and measured the dry mass of 3 different 5SA and GFP clones and compared them to the parental population using quantitative phase microscopy (Figures 2A and Figure 1—figure supplement 5D). These measurements demonstrate that:i) The average dry mass of the different nGFP clones is 283.6 ± 12.3 pg, while the average of the 5SA clones is 316.1 ± 6.3 under the conditions we typically keep them.

i) The difference between nGFP-expressing clones and the parental population is 1.5% , and suggests that the ~ 10% difference observed using the protein dye measurements did not reflect true biological differences.

ii) The 5SA-expressing clones are ~20% larger than nGFP clones in the presence of Dox through all the phases of the cell cycle, which is statistically significant (p<0.001, two-way ANOVA) and represents a difference of 5 standard deviations.

iii) The 5SA-expressing clones are significantly larger than nGFP clones through all the phases of the cell cycle (bet. 4-9%, p < 0.01, two-way ANOVA) even in the absence of Dox, which we attribute to the production of a soluble mediator of cell size by the fraction of leaky cells in the 5SA cultures.

b) We do want to note that the HEK293 cells we used for the comparison between GFP-WTYAP vs. nGFP cell in Figure 1Fand the YAPsh vs. empty vector-expressing lines are not isogenic. Considering that the WT-YAP experiment is only a validation of the FlipinTrex data, we feel comfortable including the data. It could also serve as an example of an experiment on an independent clone. We are less concerned about the YAPsh cells and their controls, as in this case we are reducing protein expression. In summary concern with clonal differences does not apply to our key experiments.

c) The reviewer requested that we include representative data on size distribution from non-co- cultured nGFP and 5SA cells. We have now included that data as the population mean ± S.D. (Figure 1G). We also refer the reviewers’ attention to the data in Figure 1—figure supplement 5Dwhich depicts the mean dry mass of three different clones per condition and the S.D.

A major unaddressed technical problem is the variability of SE dye as well as nuclear area as quantitative measurements of cell size. In our hands, the SE signal depends on the density of cell suspensions at the time of reaction. Cells from the same population fixed and diluted to dense or sparse suspensions and reacted with the same molarity of SE dye yield dramatically different dye signals (denser suspension having lower dye signal). Though it has been shown that within a single sample preparation, SE signal correlate with cell size, it is not generalizable across samples. Therefore, it is crucial for the proper interpretation the size phenotypes that appropriate SE dye calibrations are done to show that the preparation-dependent effects do not confound the results. Similarly, while it has been shown for many cell types that nuclear area linearly scales with overall cell volume, it is unclear how the increase in cell density as contact inhibition occurs may change 3D nuclear morphology and therefore the nuclear area v. cell volume linear relationship. If this linear scaling factor changes during contact inhibition, nuclear area cannot be compared across the growth curve. Especially since the cell size effects under discussion are subtle (<30% change), more detailed controls are needed.

a) In retrospect this is the point that I am happiest that the reviewer forced us to reconsider. Our original data were inadequate and it took very different experiments to enforce the points we wished to make. The concern of the reviewer went to the inherent reliability of the dye measurement, which was justifiable. The dye can be quantitative but it is not an absolute measurement thus raising doubts about its validity across samples. Furthermore, it is not very precise (Cv of ~20%) Though we agree that it may be possible to find ways to make comparisons reliable if the experiment is set up carefully, it will always be suspect in the absence of internal controls. In fact, that is one of the main reasons we used nuclear area as a proxy for cell size through much of the paper, as it allows the comparison of replicates, and conditions over time with little worry about technical variability due to the staining process, etc. That said, in doing the SE-staining, all conditions represented in one figure were processed as identically as possible, meaning they were all seeded on the same day, trypsinized, fixed and permeabilized simultaneously at the same count of cells per experiment (1-2 x106 cell/ml), and stained and analyzed at the same time. Our conclusions are based on the reproducibility of the trends across at least 3 independent repeats showing the same qualitative differences (for example Figures 1E, Figure 1—figure supplement 6Bare independent experiments done in single cell type cultures). We used the Coulter counter, which measures yet another metric of cell size: cell volume. The data are included as Figure 1—figure supplement 5A in the new submission and shows that there is a ~ 25% increase in cell volume in cells expressing YAP5SA. In this method no sample processing is required post-trypsinization and cell count should have no effect on the readout. A limitation here is that volume may be less significant than dry mass in that volume change could be due to uptake of water. Furthermore, as is the case with nuclear area it does not allow us to differentiate between cells at different phases of the cell cycle. Therefore, we used a second method based on interferometry, Quantitative Phase Microscopy (QPM), which we have recently improved by various computational means. QPM has a Cv of 2%. We measured dry cell mass of 3 clones of the two cell lines in their attached form as mentioned in response to the prior point. Using Hoechst to stain the nuclei of live cells, we were able to bin the cells into their respective cell cycle phase. The QPM data shows that 5SA cells are ~ 20% greater than nGFP controls throughout the cell cycle (Figure 1—figure supplement 5D).

Please note how tight the distributions are with the SD in all cases equal to, at most, 2% of the mean.

b) We also agree with the reviewer that we cannot assume that scaling between cell size and nuclear area in high density cultures would be the same as at low density. To address the validity of nuclear area as a proxy for cell size we used QPM to measure the dry mass per cell and compared it to the cell’s nuclear area. We know that for this cell type, nuclear area and dry mass are highly correlated across the different measurements we have done (Pearson correlation coefficient ~0.9; Figure 1—figure supplement 2A). As we can only reliably quantify dry cell mass in low density cultures (as cell-cell adhesion interferes with accurate segmentation), we could not obtain such high-quality data in confluent cultures. We certainly do not claim that the relationship between cell size and nuclear area necessarily scales 1:1 across the entirety of the growth curve, i.e.: there might not be an exponential decrease in cell size in high density cultures as reflected by the nuclear area. Nevertheless we do see a decrease in cell size at high density as reflected in: (i) a decrease in protein content using the SE dye (Figures 1F, Figure 1—figure supplement 5E, G), (ii) a decrease in protein content as reflected by BCA protein quantification from the same number of cells from high, medium and low density cultures (Figure 1—figure supplement 4D) and (iii) a decrease in cell volume measured by Coulter Counter as shown in Figure 1—figure supplement 4C. With that in mind, we have focused most of our work in the later parts of the manuscript (beyond Figure 1) on measurements in sub-confluent cultures, where QPM gives very accurate information about the dry mass of the cell.

At many points, the figure legends are insufficient for understanding what experiments were performed. The manuscript does not indicate the conditions in which cells were collected for flow analysis (e.g. SE-dye stains). A major point of Figure 1 is that cell size is dependent on culture density, but the density of the cultures used is not explicitly denoted throughout Figures 2-5 or anywhere in the text or Materials and methods.

We have revisited the legends for clarification as suggested by several of the reviewers. Cell number, the culture vessel and the approximate corresponding density have been explicitly reported in each legend.

The authors show no relationship between YAP overactivity and k (Figure 2E), but do show a relationship between CYR61 neutralization and k (Figure 4F). The apparent contradiction between YAP not being necessary for regulating growth rate and CYR61 being sufficient to alter growth rate is not addressed in the manuscript and could be related to the technical concerns listed above.

We would like to point out that the apparent “discrepancy” between the effect of CYR61 and k has been confirmed in several independent experiments. More specifically we consistently see that (1) the addition of recombinant CYR61 decreases k (Figure 5G, J), (2) the addition of anti-CYR61 antibody increases k Figure 5A, D), while (3) YAP5SA expression increases k (Figure 1D) and (4) knocking down YAP decreases k (Figure 3E). Our new data also supports that (1) YAP expression increases Ymax in a CYR61-dependent manner, but (2) YAP’s strongest effects on k are cell autonomous. Given that YAP mediates the transcription of not only CYR61, which is secreted, but a multitude of intracellular proteins, it is not completely surprising that the effect of YAP on k is not CYR61-dependent. One possible explanation for seeing an effect on k at all when low- density cells (still not contact inhibited) are treated with recombinant CYR61 or anti-CYR61 antibodies is that the effective concentration of CYR61 in these cultures becomes less dependent of the amount accumulated in the medium due to cellular production. We have not exhaustively investigated all other possible explanations, but our data does lead us to conclude with relative confidence that CYR61 and YAP overexpression do not have the same effect on k. We have added our thoughts on this discrepancy to the manuscript.

The main experiments the authors use to argue for non-cell autonomous effects of YAP is via co-culture and neutralizing antibody experiments. Firstly, co-cultures results can be confounded by complex mechanical constraints, especially if the two cell types have different cell sizes, as shown in Figure 2. Secondly, the authors did not show the effect of the anti-CYR61 antibody on wild-type cells (Figure 4A, D), and therefore it is difficult to interpret whether its effects are specific. It would be important to show these controls. Lastly, a complementary experiment showing that conditioned media produced by YAP5A cells can lead to similar effect in increasing k or Ymax in recipient wild-type cells would argue that phenotypes observed in co-culture are indeed mediated through secreted ligands.

a) We agree that co-cultures can be confounded by mechanical constraints. It is something that we specifically considered, and it is not clear why this was not commented on. We did not stop with co- culture but used a Transwell assay where the cells are not in physical contact (Figure 2D, E). We further showed the same activity using medium conditioned by 5SA-expressing cells (Figure 3H), where there are no possible interactions between living cells, such as reciprocal signaling.

b) We have included a supplementary figure on the effect of CYR61 on pure mCherry-expressing WT/nGFP cells (Figure 2—figure supplement 1G, J), which indicate that neutralizing CYR61 in WT cells also decreases Ymax. Furthermore, the new data in Figure 6 is based on data from those WT cells which produce 8 times less CYR61 than YAP5SA cells. We also do not see any titratable effects on Ymax when cells are treated with a control polyclonal Rabbit IgG antibody.

c) The issue of specificity has been addressed in more detail in the response to point 2 of reviewer 1’s comments.

Reviewer #3:

The manuscript aims to provide insight into the role of YAP transcription factor in determining the size of organs. It reports the results of in vitro experiments with cultured HEK293 cells, which includes the identification of a certain secreted factor (a previously known target of YAP) as a mediator of cell proliferation, controlling the limiting density of a confluent cell culture. Other conclusions and interpretation are however quite equivocal. Overall, in my opinion, this manuscript is substantially below the standard of eLife.

The manuscript aims to disentangle regulation of cell size and cell number, but lacks quantitation to do it in a convincing way. Figure 1A-D suggests a simple picture: cell area is constant while cells proliferate before reaching confluence, but goes down after achieving confluence. Rate of proliferation goes to zero as cells reach minimal area. The authors use a logistic equation to describe growth and the resulting cell number with two parameters: one for the rate of growth and the other for the carrying capacity corresponding to the maximal cell density which in general depend on growth conditions and substrate. This is all reasonable. However, curiously, the manuscript fails to note that minimal cell area and the terminal density of cells (and hence the Ymax or "carrying capacity") are directly related (by inverse proportionality). Instead they write: "Since size at the later time points seems to be a complex product of initial size and the achieved Ymax, we focused on changes in nuclear area (cell size) in low density cultures."

There are two major points of confusion here. First: since in the confluent layer cell area and density are inversely proportional to each other, it is not clear what the authors mean by independent control of cell size and cell number. Second: physiology of non-confluent cells is typically quite different from the confluent ones: e.g. mesenchymal/epithelial, so one should not be surprised that factors controlling cell size at low density and the terminal density in confluency may be quite different. This muddies the water as far as the overarching issue of organ size control is concerned, since the paper strongly implies that cell size and the terminal density provide different channels of control. [[See for example the Abstract: "…is the major regulator of the non-cell autonomous increase in cell number, but does not affect cell size. The molecular identity of the non-cell autonomously acting mediator of cell size is yet to be identified."]

I suspect that confusion arises because literature on organ size determination often distinguishes between cell size control (equivalent to cell density) and cell number control with the latter referring to the total cell count in a variable size organ (such as Drosophila wing imaginal disc). In that case, cell area and total cell number are independent variables that together define the total area (or "size") of the organ. In contrast, since the in vitro well size is fixed, "Nuclei/Well" measurement performed by the authors report only the density. In the case of a fixed well area, cell size and number are not independent. The fundamental aspect of organ size determination – its global nature – is completely overlooked by the authors. Lacking confinement of the dish well, how do real organs know their size? This work doesn't give us a clue. The confusion involving cell size/cell density/cell number clouds the interpretation of the results and detracts also from the factual findings regarding YAP/secreted factors reported in the paper.

We of course agree with the reviewer that organ size control (with the potentially unlimited boundaries for growth), and the growth of a population on a dish are potentially determined by different factors. We now make clearer that this paper is not about organ size control, and the tissue culture system is not being used to draw one-to-one parallels to organ size regulation in vivo. That said, cells proliferate and die, and cells have a mean size. These can and do vary independently physiologically and upon perturbation. We have simply used the tissue culture system to learn about the effect of YAP on (1) cell size, (2) cell division and (3) death rates. As the balance of all three parameters affects organ growth, the findings could be relevant to how YAP regulates organ size but drawing any specific conclusion about organ size goes beyond our experiments. Our interpretation of our results in the context of organ size is at most speculative and it is up to future experiments to show how these contribute to growth and size regulation of organs. It is not clear to us whether the reviewer believes that we are confused (we are not) or that we might confuse others. We have cleared this up in the Discussion where we discuss more precisely what the relevance of our findings in tissue culture could be the control of organ size in what we hope is a non- dogmatic way. We have also differentiated between the in vivoand culture context in the Introduction for clarity.

Second, we would argue that while the area of a dish is finite, a higher number of cells does not necessarily mean cells need to become smaller. The area of the well limits the substrate-attached area of the cells, but cells can continue to grow in height. So, fitting more cells does not necessarily require them to be smaller in size. As the reviewer has speculated, it is tempting to think that cells stop proliferating simply because they have reached a minimum size, and as YAP5SA cells are larger initially they can proliferate more times before they stop dividing. However, that did not turn out to be the case, and this was not the independence referred to in the title. The “independence” in the title refers to what turned out to be the effect of CYR61 on the number of cells at high density by affecting the death rates, without affecting cell size. That independence between proliferation rates and cell size was further emphasized by the finding that the effect of YAP5SA on cell division rates (represented in k) seem to be dominated by cell autonomous effectors, while the size effects are non-cell autonomous. We appreciate that the reviewer might have been especially worried about confusion for others. We have endeavored throughout this paper to point out that we are talking about cell communication and cellular phenomena and not organ size while at the same time pointing out that there might be some implications of these studies in an in vivocircumstance.

The Discussion section revisits 50 years of "contact inhibition" history while missing key ideas (and quantitative studies) on density dependent mobility and the effects of confinement, which would have helped with the logic of the study. Another area of study that is relevant, but overlooked, is the question of "cell competition" which is relevant both to contact inhibition and size determination. Secreted factors have long been suggested as a readouts of tissue mass. How does the secreted factor controlled by YAP relate to those ideas?

Cell-cell competition is quite an interesting idea to explore, and we did not overlook it on purpose, but believe that our data does not provide more than loose speculation on it. As the new version of the manuscript also confirms that the effects we observe in co-culture can be mediated by conditioned medium, we believe competition might be playing a smaller role with respect to the parameters we studied here, but of course does not rule out its overall importance in other contexts. As to the role of secreted proteins in organ size regulation in vivo, etc., we do speculate broadly on that in the last paragraph or so of the Discussion but tried to keep the breadth of our speculations within reasonable length and within the immediate implication of the data we have.

In addition to the overall logic of the work I have trouble with some important technical aspects.

1) Logistic fits shown in Figure 1—figure supplement 3C are quite poor at estimating saturation density Ymax. Since measurement of the latter is central to the analysis, I have very little confidence in its accuracy and validity. Furthermore, logistic description at best, kicks in upon confluence, which is reached at different times under different conditions (e.g. it should depend on cell size in the low density state), so an additional fitting parameter would seem to be in order. In either case, since the authors use cell size in a pre-confluent state, one would like to see a more careful/quantitative study of growth rate as a function of cell density (and cell size) in that pre-confluent phase of growth.

While the reviewer might not find the data in Figure 1—figure supplement 3C) convincing (R2>0.99), the main explanation for the sup-exemplary fit is that we do not have enough data points in the later phases of the growth curve. We have included all the new data with the original and fitted data, all of which have more points in the plateau phase of growth and show a near-perfect fit (see Figures 1C, 2E, 3E, Figure 1—figure supplement 5G). We observe that 5SA cells, although larger in size in the low-density regime, start to decrease in size later than the WT cells. So if they were reaching confluence earlier, their response to contact-inhibition is mildly retarded, which wouldn’t be captured by the simple additional parameter the reviewer requests be added to a basic logistic model.

That being said, we have taken the reviewer’s comment seriously and used a model-agnostic method of estimating Ymax. In this case Ymax is simply the number of nuclei/well at which population growth plateaus (rate of change is zero; termed Ymax’). We estimate the rate of growth (k’) by fitting only the earlier time points for each data set to an exponential growth equation. Unsurprisingly, this analysis yields qualitatively similar results as k in the logistic function approximates to the exponential growth rate at low density by definition. The changes in maximum cell number (Ymax’) are also maintained. As Ymax’ and k’ do not change the conclusions of the manuscript in any way, we continue to use the logistic fit in the main figures.

2) Do Figures 4A, D 5A, E; 6B, G present data or some sort of a fit (logistic?) to data. Error bars? (Error bars are also missing Figure 1—figure supplement 3. The Materials and methods section does not explain the method used for determining Nuclei/well measurement. Does it involve image segmentation? What are the errors? Could there be a systematic error increasing with increasing cell density?

We had originally excluded error bars because we are actually just showing the fitted data. We have made a point of including the raw data with error bars and their fit in all new panels.

We mention in the Materials and methods section that nuclei were segmented using the image analysis software on the Incucyte server. While there might be an increase in segmentation error at higher density, it is off-set by an increase in the total number of nuclei per well. Overall, segmenting nuclei is relatively robust on the various platforms we have tried. We have also validated that there is a decrease in cell size with density using a coulter counter (cell volume), BCA estimation of total protein content from the same number of cells from cultures of different densities, as well as the SE-dye (reflecting total protein content; See response to reviewer 2).

3) The authors demonstrate the non-autonomous effect of YAP signaling by examining the dependence of Ymax on the fraction of YAP5SA cells in co-culture. Since the effectors are secreted/soluble factors, I would expect that their effect depends on the volume of the media. Does it?

We certainly see a decrease in the effect of YAP5SA-conditioned medium on cell number when diluted. See Author response image 1.

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Citations

Douaa Mugahid. 2018. Induction of YAP5SA-GFP vs nGFP expression in FlipinTrex293 cells. NCBI Gene Expression Omnibus. GSE123296

Supplementary Materials

Supplementary file 1. Table with protein and mRNA changes.

elife-53404-supp1.xlsx^{(588.4KB, xlsx)}

Transparent reporting form

elife-53404-transrepform.docx^{(246.2KB, docx)}

Data Availability Statement

Sequencing data has been deposited in GEO under accession code GSE123296.

The following dataset was generated:

Douaa Mugahid. 2018. Induction of YAP5SA-GFP vs nGFP expression in FlipinTrex293 cells. NCBI Gene Expression Omnibus. GSE123296

[bib1] Anzi S, Stolovich-Rain M, Klochendler A, Fridlich O, Helman A, Paz-Sonnenfeld A, Avni-Magen N, Kaufman E, Ginzberg MB, Snider D, Ray S, Brecht M, Holmes MM, Meir K, Avivi A, Shams I, Berkowitz A, Shapiro AMJ, Glaser B, Ben-Sasson S, Kafri R, Dor Y. Postnatal exocrine pancreas growth by cellular hypertrophy correlates with a shorter lifespan in mammals. Developmental Cell. 2018;45:726–737. doi: 10.1016/j.devcel.2018.05.024. [DOI] [PubMed] [Google Scholar]

[bib2] Aragona M, Panciera T, Manfrin A, Giulitti S, Michielin F, Elvassore N, Dupont S, Piccolo S. A mechanical checkpoint controls multicellular growth through YAP/TAZ regulation by actin-processing factors. Cell. 2013;154:1047–1059. doi: 10.1016/j.cell.2013.07.042. [DOI] [PubMed] [Google Scholar]

[bib3] Basu S, Totty NF, Irwin MS, Sudol M, Downward J. Akt phosphorylates the Yes-associated protein, YAP, to induce interaction with 14-3-3 and attenuation of p73-mediated apoptosis. Molecular Cell. 2003;11:11–23. doi: 10.1016/S1097-2765(02)00776-1. [DOI] [PubMed] [Google Scholar]

[bib4] Cadart C, Monnier S, Grilli J, Sáez PJ, Srivastava N, Attia R, Terriac E, Baum B, Cosentino-Lagomarsino M, Piel M. Size control in mammalian cells involves modulation of both growth rate and cell cycle duration. Nature Communications. 2018;9:e3275. doi: 10.1038/s41467-018-05393-0. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib5] Chen CC, Mo FE, Lau LF. The angiogenic factor Cyr61 activates a genetic program for wound healing in human skin fibroblasts. Journal of Biological Chemistry. 2001;276:47329–47337. doi: 10.1074/jbc.M107666200. [DOI] [PubMed] [Google Scholar]

[bib6] Chen C-C, Kim K-H, Lau LF. The matricellular protein CCN1 suppresses hepatocarcinogenesis by inhibiting compensatory proliferation. Oncogene. 2016;35:1314–1323. doi: 10.1038/onc.2015.190. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib7] Diril MK, Ratnacaram CK, Padmakumar VC, Du T, Wasser M, Coppola V, Tessarollo L, Kaldis P. Cyclin-dependent kinase 1 (Cdk1) is essential for cell division and suppression of DNA re-replication but not for liver regeneration. PNAS. 2012;109:3826–3831. doi: 10.1073/pnas.1115201109. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib8] Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29:15–21. doi: 10.1093/bioinformatics/bts635. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib9] Dong J, Feldmann G, Huang J, Wu S, Zhang N, Comerford SA, Gayyed MF, Anders RA, Maitra A, Pan D. Elucidation of a universal size-control mechanism in Drosophila and mammals. Cell. 2007;130:1120–1133. doi: 10.1016/j.cell.2007.07.019. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib10] Dupont S, Morsut L, Aragona M, Enzo E, Giulitti S, Cordenonsi M, Zanconato F, Le Digabel J, Forcato M, Bicciato S, Elvassore N, Piccolo S. Role of YAP/TAZ in mechanotransduction. Nature. 2011;474:179–183. doi: 10.1038/nature10137. [DOI] [PubMed] [Google Scholar]

[bib11] Ernst S, Demirci C, Valle S, Velazquez-Garcia S, Garcia-Ocaña A. Mechanisms in the adaptation of maternal β-cells during pregnancy. Diabetes Management. 2011;1:239–248. doi: 10.2217/dmt.10.24. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib12] Fankhauser G. Maintenance of normal structure in Heteroploid Salamander Larvae, through compensation of changes in cell size by adjustment of cell number and cell shape. Journal of Experimental Zoology. 1945;100:445–455. doi: 10.1002/jez.1401000310. [DOI] [PubMed] [Google Scholar]

[bib13] Gilgenkrantz H, Collin de l'Hortet A. Understanding liver regeneration: from mechanisms to regenerative medicine. The American Journal of Pathology. 2018;188:1316–1327. doi: 10.1016/j.ajpath.2018.03.008. [DOI] [PubMed] [Google Scholar]

[bib14] Ginzberg MB, Chang N, D'Souza H, Patel N, Kafri R, Kirschner MW. Cell size sensing in animal cells coordinates anabolic growth rates and cell cycle progression to maintain cell size uniformity. eLife. 2018;7:e26957. doi: 10.7554/eLife.26957. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib15] Grijalva JL, Huizenga M, Mueller K, Rodriguez S, Brazzo J, Camargo F, Sadri-Vakili G, Vakili K. Dynamic alterations in hippo signaling pathway and YAP activation during liver regeneration. American Journal of Physiology-Gastrointestinal and Liver Physiology. 2014;307:G196–G204. doi: 10.1152/ajpgi.00077.2014. [DOI] [PubMed] [Google Scholar]

[bib16] Groupe V, Manaker RA. Discrete foci of altered chicken embryo cells associated with rous sarcoma virus in tissue culture. Virology. 1956;2:838–840. doi: 10.1016/0042-6822(56)90064-2. [DOI] [PubMed] [Google Scholar]

[bib17] Gujral TS, Kirschner MW. Hippo pathway mediates resistance to cytotoxic drugs. PNAS. 2017;114:E3729–E3738. doi: 10.1073/pnas.1703096114. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib18] Hansen CG, Ng YL, Lam WL, Plouffe SW, Guan KL. The hippo pathway effectors YAP and TAZ promote cell growth by modulating amino acid signaling to mTORC1. Cell Research. 2015;25:1299–1313. doi: 10.1038/cr.2015.140. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib19] Holbourn KP, Acharya KR, Perbal B. The CCN family of proteins: structure-function relationships. Trends in Biochemical Sciences. 2008;33:461–473. doi: 10.1016/j.tibs.2008.07.006. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib20] Huang daW, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nature Protocols. 2009;4:44–57. doi: 10.1038/nprot.2008.211. [DOI] [PubMed] [Google Scholar]

[bib21] Kafri R, Levy J, Ginzberg MB, Oh S, Lahav G, Kirschner MW. Dynamics extracted from fixed cells reveal feedback linking cell growth to cell cycle. Nature. 2013;494:480–483. doi: 10.1038/nature11897. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib22] Kamińska K, Szczylik C, Bielecka ZF, Bartnik E, Porta C, Lian F, Czarnecka AM. The role of the cell-cell interactions in Cancer progression. Journal of Cellular and Molecular Medicine. 2015;19:283–296. doi: 10.1111/jcmm.12408. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib23] Katsube K, Sakamoto K, Tamamura Y, Yamaguchi A. Role of CCN, a vertebrate specific gene family, in development. Development, Growth & Differentiation. 2009;51:55–67. doi: 10.1111/j.1440-169X.2009.01077.x. [DOI] [PubMed] [Google Scholar]

[bib24] Kim NG, Gumbiner BM. Adhesion to fibronectin regulates hippo signaling via the FAK-Src-PI3K pathway. The Journal of Cell Biology. 2015;210:503–515. doi: 10.1083/jcb.201501025. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib25] Lane SW, Williams DA, Watt FM. Modulating the stem cell niche for tissue regeneration. Nature Biotechnology. 2014;32:795–803. doi: 10.1038/nbt.2978. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib26] Latinkic BV, Mercurio S, Bennett B, Hirst EM, Xu Q, Lau LF, Mohun TJ, Smith JC. Xenopus Cyr61 regulates gastrulation movements and modulates wnt signalling. Development. 2003;130:2429–2441. doi: 10.1242/dev.00449. [DOI] [PubMed] [Google Scholar]

[bib27] Leask A, Abraham DJ. All in the CCN family: essential matricellular signaling modulators emerge from the bunker. Journal of Cell Science. 2006;119:4803–4810. doi: 10.1242/jcs.03270. [DOI] [PubMed] [Google Scholar]

[bib28] Lee MJ, Byun MR, Furutani-Seiki M, Hong JH, Jung HS. YAP and TAZ regulate skin wound healing. Journal of Investigative Dermatology. 2014;134:518–525. doi: 10.1038/jid.2013.339. [DOI] [PubMed] [Google Scholar]

[bib29] Lin MT, Chang CC, Chen ST, Chang HL, Su JL, Chau YP, Kuo ML. Cyr61 expression confers resistance to apoptosis in breast Cancer MCF-7 cells by a mechanism of NF-kappaB-dependent XIAP up-regulation. Journal of Biological Chemistry. 2004;279:24015–24023. doi: 10.1074/jbc.M402305200. [DOI] [PubMed] [Google Scholar]

[bib30] Liu X, Oh S, Kirschner MW. Optimization of quantitative phase microscopy reveals new features of the growth rate of adherent mammalian cells. bioRxiv. 2019 doi: 10.1101/631119. [DOI]

[bib31] Lu L, Finegold MJ, Johnson RL. Hippo pathway coactivators yap and taz are required to coordinate mammalian liver regeneration. Experimental & Molecular Medicine. 2018;50:e423. doi: 10.1038/emm.2017.205. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib32] McAlister GC, Nusinow DP, Jedrychowski MP, Wühr M, Huttlin EL, Erickson BK, Rad R, Haas W, Gygi SP. MultiNotch MS3 enables accurate, sensitive, and multiplexed detection of differential expression across Cancer cell line proteomes. Analytical Chemistry. 2014;86:7150–7158. doi: 10.1021/ac502040v. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib33] Mooring M, Fowl BH, Lum SZC, Liu Y, Yao K, Softic S, Kirchner R, Bernstein A, Singhi AD, Jay DG. Hepatocyte stress increases expression of YAP and TAZ in hepatocytes to promote parenchymal inflammation and fibrosis. Hepatology. 2020 doi: 10.1002/hep.30928. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib34] Paulo JA, O'Connell JD, Everley RA, O'Brien J, Gygi MA, Gygi SP. Quantitative mass spectrometry-based multiplexing compares the abundance of 5000 S. cerevisiae proteins across 10 carbon sources. Journal of Proteomics. 2016a;148:85–93. doi: 10.1016/j.jprot.2016.07.005. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib35] Paulo JA, O’Connell JD, Gygi SP. A Triple Knockout (TKO) Proteomics Standard for Diagnosing Ion Interference in Isobaric Labeling Experiments. Journal of The American Society for Mass Spectrometry. 2016b;27:1620–1625. doi: 10.1007/s13361-016-1434-9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib36] Penzo-Méndez AI, Stanger BZ. Organ-Size regulation in mammals. Cold Spring Harbor Perspectives in Biology. 2015;7:a019240. doi: 10.1101/cshperspect.a019240. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib37] Peshkin L, Gupta M, Ryazanova L, Wühr M. Bayesian confidence intervals for multiplexed proteomics integrate Ion-statistics with peptide quantification concordance. Molecular & Cellular Proteomics. 2019;18:2108–2120. doi: 10.1074/mcp.TIR119.001317. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib38] Rappsilber J, Ishihama Y, Mann M. Stop and go extraction tips for matrix-assisted laser desorption/ionization, Nanoelectrospray, and LC/MS sample pretreatment in proteomics. Analytical Chemistry. 2003;75:663–670. doi: 10.1021/ac026117i. [DOI] [PubMed] [Google Scholar]

[bib39] Shu Z, Deng W-M. Differential regulation of cyclin E by Yorkie-Scalloped signaling in organ development. G3: Genes|Genomes|Genetics. 2017;7:1049–1060. doi: 10.1534/g3.117.039065. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib40] Stoker MG, Rubin H. Density dependent inhibition of cell growth in culture. Nature. 1967;215:171–172. doi: 10.1038/215171a0. [DOI] [PubMed] [Google Scholar]

[bib41] Su AI, Guidotti LG, Pezacki JP, Chisari FV, Schultz PG. Gene expression during the priming phase of liver regeneration after partial hepatectomy in mice. PNAS. 2002;99:11181–11186. doi: 10.1073/pnas.122359899. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib42] Tapon N, Harvey KF, Bell DW, Wahrer DC, Schiripo TA, Haber D, Hariharan IK. Salvador promotes both cell cycle exit and apoptosis in Drosophila and is mutated in human Cancer cell lines. Cell. 2002;110:467–478. doi: 10.1016/S0092-8674(02)00824-3. [DOI] [PubMed] [Google Scholar]

[bib43] Tian X, Azpurua J, Hine C, Vaidya A, Myakishev-Rempel M, Ablaeva J, Mao Z, Nevo E, Gorbunova V, Seluanov A. High-molecular-mass hyaluronan mediates the Cancer resistance of the naked mole rat. Nature. 2013;499:346–349. doi: 10.1038/nature12234. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib44] Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, Pimentel H, Salzberg SL, Rinn JL, Pachter L. Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and cufflinks. Nature Protocols. 2012;7:562–578. doi: 10.1038/nprot.2012.016. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib45] Verhulst PF. Recherches mathématiques sur la loi d’accroissement de la population. Nouv Mém Acad R. Sci B.-Lett. Brux. 1845:1–41. [Google Scholar]

[bib46] Volarevic S, Stewart MJ, Ledermann B, Zilberman F, Terracciano L, Montini E, Grompe M, Kozma SC, Thomas G. Proliferation, but not growth, blocked by conditional deletion of 40S ribosomal protein S6. Science. 2000;288:2045–2047. doi: 10.1126/science.288.5473.2045. [DOI] [PubMed] [Google Scholar]

[bib47] Weinberg RA. The Biology of Cancer. New York: Garland Science; 2007. [Google Scholar]

[bib48] Xin M, Kim Y, Sutherland LB, Qi X, McAnally J, Schwartz RJ, Richardson JA, Bassel-Duby R, Olson EN. Regulation of insulin-like growth factor signaling by yap governs cardiomyocyte proliferation and embryonic heart size. Science Signaling. 2011;4:ra70. doi: 10.1126/scisignal.2002278. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib49] Yang X, Xu T. Molecular mechanism of size control in development and human diseases. Cell Research. 2011;21:715–729. doi: 10.1038/cr.2011.63. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib50] Yimlamai D, Christodoulou C, Galli GG, Yanger K, Pepe-Mooney B, Gurung B, Shrestha K, Cahan P, Stanger BZ, Camargo FD. Hippo pathway activity influences liver cell fate. Cell. 2014;157:1324–1338. doi: 10.1016/j.cell.2014.03.060. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib51] Zanconato F, Forcato M, Battilana G, Azzolin L, Quaranta E, Bodega B, Rosato A, Bicciato S, Cordenonsi M, Piccolo S. Genome-wide association between YAP/TAZ/TEAD and AP-1 at enhancers drives oncogenic growth. Nature Cell Biology. 2015;17:1218–1227. doi: 10.1038/ncb3216. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib52] Zhang J, Ji JY, Yu M, Overholtzer M, Smolen GA, Wang R, Brugge JS, Dyson NJ, Haber DA. YAP-dependent induction of amphiregulin identifies a non-cell-autonomous component of the hippo pathway. Nature Cell Biology. 2009;11:1444–1450. doi: 10.1038/ncb1993. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib53] Zhao B, Wei X, Li W, Udan RS, Yang Q, Kim J, Xie J, Ikenoue T, Yu J, Li L, Zheng P, Ye K, Chinnaiyan A, Halder G, Lai ZC, Guan KL. Inactivation of YAP oncoprotein by the hippo pathway is involved in cell contact inhibition and tissue growth control. Genes & Development. 2007;21:2747–2761. doi: 10.1101/gad.1602907. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib54] Zhao B, Li L, Tumaneng K, Wang CY, Guan KL. A coordinated phosphorylation by lats and CK1 regulates YAP stability through SCF(beta-TRCP) Genes & Development. 2010;24:72–85. doi: 10.1101/gad.1843810. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

YAP regulates cell size and growth dynamics via non-cell autonomous mediators

Douaa Mugahid

Marian Kalocsay

Xili Liu

Jonathan Scott Gruver

Leonid Peshkin

Marc W Kirschner

Roles

Abstract

Introduction

Results

Overexpression of non-phosphorylatable YAP increases the rate of cell division, cell size, and maximum cell density in HEK293 populations

Figure 1. YAP5SA expression affects cell size and population growth dynamics.

Figure 1—figure supplement 1. Example of images acquired on and analyzed by the Incucyte ZOOM to obtain data about the number of nuclei and their average area.

Figure 1—figure supplement 2. Nuclear area is a good proxy for cell size in Flipin-Trex-293 cells.

Figure 1—figure supplement 3. Fitting changes in cell number over time using the Logistic growth equation.

Figure 1—figure supplement 4. Culture density affects cell death rates and cell size.

Figure 1—figure supplement 5. The effect of YAP5SA, YAPwt and YAPsh expression on cell size.

Figure 1—figure supplement 6. Serum is required for the 5SA-dependent changes in proliferation and size.

The non-cell autonomous nature of YAP-mediated changes in cell size

Figure 2. YAP5SA increases cell size non-autonomously.

Figure 2—figure supplement 1. The effect of 5SA-conditioned medium on 3T3 fibroblasts.

The effect of YAP5SA on cell number is mediated both by cell autonomous and non-autonomous processes

Figure 3. YAP affects the population growth rate (k) and carrying capacity (Ymax).

Many proteins affected by YAP5SA overexpression are secreted or membrane proteins

Figure 4. YAP-dependent changes in protein and mRNA expression levels.

The YAP target, CYR61, affects cell number but not cell size

Figure 5. The effects of CTGF and CYR61 on cell size and number.

Figure 5—figure supplement 1. Validation of anti-CYR61 and anti-CTGF antibodies.

Cyr61 increases Ymax by reducing apoptosis

Figure 6. CYR61 affects cell proliferation and death at high density.

Discussion

Materials and methods

Cloning

Cell line generation

Transwell assays

Antibodies and recombinant proteins

Conditioned medium production

Incycyte-based image acquisition and automated image analysis for nuclear segmentation

Quantitative phase microscopy (QPM)

Data plotting and fitting

Coulter counter measurement

Total protein/DNA staining

RNA-Sequencing and data analysis

Mass spectrometry and data analysis

Acknowledgements

Funding Statement

Contributor Information

Funding Information

Additional information

Competing interests

Author contributions

Additional files

Data availability

References

Decision letter

Roles

Author response

Author response image 1. 1 pt lines reflect the mean and 0.25 pt lines are ± SEM.

Associated Data

Data Citations

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases