A mathematical model for the dependence of keratin aggregate formation on the quantity of mutant keratin expressed in EGFP-K14 R125P keratinocytes.

We examined keratin aggregate formation and the possible mechanisms involved. With this aim, we observed the effect that different ratios between mutant and wild-type keratins expressed in cultured keratinocytes may have on aggregate formation in vitro, as well as how keratin aggregate formation affects the mechanical properties of cells at the cell cortex. To this end we prepared clones with expression rates as close as possible to 25%, 50% and 100% of the EGFP-K14 proteins (either WT or R125P and V270M mutants). Our results showed that only in the case of the 25% EGFP-K14 R125P mutant significant differences could be seen. Namely, we observed in this case the largest accumulation of keratin aggregates and a significant reduction in cell stiffness. To gain insight into the possible mechanisms behind this observation, we extended our previous mathematical model of keratin dynamics by implementing a more complex reaction network that considers the coexistence of wild-type and mutant keratins in the cell. The new model, consisting of a set of coupled, non-linear, ordinary differential equations, allowed us to draw conclusions regarding the relative amounts of intermediate filaments and aggregates in cells, and suggested that aggregate formation by asymmetric binding between wild-type and mutant keratins could explain the data obtained on cells grown in culture.

Introduction

The intermediate filament (IF) cytoskeleton of epithelial cells consists of differentiation-specific sets of type I and type II keratin proteins, which establish a functional filament network that is crucial for the mechanical integrity of cells and tissues [1]. Over the last three decades, hundreds of keratin gene mutations have been discovered and linked to a large group of human diseases manifesting cell and tissue fragility [2].

K14p.Arg125, a residue located in the helix initiation motif of keratin 14 (K14), is the most frequently mutated residue in this protein. It is linked to over 30% of reported cases of severe epidermolysis bullosa simplex (EBS), formerly known as severe generalized EBS or EBS Dowling-Meara [3]. Several amino acid changes of this residue have been found in patient-derived cells: R125H, R125C, R125S, R125L and R125P. In vitro, they all cause some degree of keratin filament network perturbation, which also manifests in keratin aggregates formation [4-7]. Herrmann and colleagues analyzed the impact of the K14 R125H mutation on in vitro keratin filament polymerization and found that under certain conditions, K14 R125H monomers were still able to form functional keratin IFs when mixed with wild-type (WT) K5 at 1:1 ratio [8]. Thus, even if all K14 protein used were mutant, when mixed with normal K5 monomers, it would form rather normal looking keratin filaments. The authors concluded that keratin aggregation, which is often seen both in primary and immortalized mutant keratinocytes, might be the consequence of other mechanistic reasons besides disrupted filament assembly. Nevertheless, the mechanism of keratin aggregate formation is still elusive.

Another interesting aspect of keratin aggregates is that they are not visible in all severe EBS-patient derived keratinocytes and even less so in mild keratin mutants, but they do appear after exposure to some type of external stress. In particular, keratin aggregation is triggered by thermal or hypo-osmotic stress in vitro [5, 9]. Others have shown that mutant keratin aggregation may also be reduced if mutant keratinocytes are pre-treated with a chemical chaperone like trimethylamine N-oxide (TMAO) [10, 11]. TMAO and similar compounds act by stabilizing proteins in their native state, protecting them against thermal denaturation and aggregation. The presence of keratin aggregates has also been linked to increased MAPK activity, which place EBS patient-derived cells in a state of constant stress [9].

Previously, we have extended an existing mathematical model for keratin turnover in wild-type cells [12] to account for the emergence of keratin particles in mutant keratinocytes [13]. Our model includes the turnover between soluble, particulate and filamentous keratin forms. We assumed that keratin mutations cause a slowdown in the assembly of an intermediate keratin phase into filaments and in silico we have demonstrated this to be enough to cause the loss of keratin filaments and the appearance of keratin aggregates [13].

In the present study, we looked further into the topic of keratin aggregates formation and to the possible mechanisms involved. We addressed this by observing the effect that different ratios between mutant and WT keratin proteins expressed in cultured keratinocytes may have on aggregate formation in vitro, and how keratin aggregate formation affects the mechanical properties of cells at the cell cortex [14]. This is an important question, as although the vast majority of EBS cases are inherited in a dominant negative fashion, the actual ratio between WT and mutant keratin protein expressed in cells is not known, and this in turn may also affect the patient’s phenotype. Previous AFM and optical stretcher measurements of WT and keratin-free keratinocytes indicated that loss of keratin significantly decreased the stiffness of keratinocytes [15, 16]. However, the re-expression of substoichiometric amounts of K5/K14 normalized cell stiffness and rescued intercellular adhesion defects [15, 17]. When the same mutation we looked at in our study was introduced into mouse cells lacking the entire type I keratin cluster (K14 R125P in human is analogous to K14 R131P in mouse), the mutation had a big impact on cells by affecting IF network formation and by causing keratin aggregation [18]. In addition, this mutation in mouse cells had a larger impact on cell stiffness (causing a decrease) than the actual absence of all type I keratins.

In this study we used stably transfected keratinocyte cell lines we have previously engineered [19], by introducing EGFP labeled K14 constructs containing either the WT cDNA sequence, the R125P mutation, or the V270M mutation (the latter is situated in the L12 linker region of K14 and is linked to the mild, localized type of EBS [20]) into the background of a WT cell line (NEB1 cells). Here we analysed further the clones with expression rates as close as possible to 25%, 50% and 100% of the EGFP-K14 proteins (WT or mutant). At the same time, we took on a systems biology approach and proposed a reaction network that considers the coexistence of WT and mutant keratins in the cell. With this model we drew conclusions regarding the relative amounts of intermediate filaments and aggregates in the cell. The model consists of a set of coupled, non-linear, ordinary differential equations (ODEs) for which we found the equilibrium state. The model implemented in this work expands the work that was developed over the last two decades [12, 21, 22] on the keratin cycle of normal keratinocytes. Though in Gouveia et al (2020) [13] we modeled the R125P mutation, the interplay between WT and mutant keratins will be addressed for the first time in the present work [5].

Materials and methods

Cell lines and culture conditions

We previously engineered several isogenic cell lines by introducing and stably expressing an extra copy of EGFP labeled K14 WT, EGFP labeled K14 R125P and EGFP labeled K14 V270M constructs [19]. In brief, the human K14 wild-type cDNA (NM_000526) was originally cloned into the EGFP-C1 vector (Clontech, USA) and tested by transient transfection of control keratinocyte cells (NEB1). This vector was then used to prepare the mutant construct by introducing the K14 R125P mutation using the QuikChange Site-Directed Mutagenesis Kit (Stratagene, USA). To generate the stable cell lines used in this study, the K14 wild-type and mutant cDNAs were cut out of the EGFP-C1 constructs and re-cloned into the pLEGFP-C1 (Clontech, USA) retro-viral vector using the HindIII and BamHI restriction sites, and the resulting retro-viral wild-type and mutant constructs were used to transfect the NEB1 (control) cell line. After single cell cloning and antibiotic selection with G418, clones were tested by Western blotting for the expression level of the EGFP construct, and the ones that were expressing the EGFP K14 construct (WT and mutant) as close to 25%, 50% and 100% ratio with the endogenous K14 WT (tested by Western blot), were further expanded and used. As none of the 60 analyzed clones had the exact desired ratio, in the end we compared clones with an 18%, 45% and 99% content of EGFP K14 protein (WT or mutant). For easier reference, the clones will still be labeled and referred to in the paper as “25”, “50” and “100”.

Cells were cultured in RM+ medium: 75% DMEM (Dulbecco’s modified Eagle’s medium) plus 25% Ham’s F12 medium, with added 10% fetal calf serum and additional growth factors hydrocortisone (0.4 mgmL−1), cholera toxin (10−10 M), transferrin (5.0 mgmL−1), liothyronine (2 × 10−11 M), adenine (1.9 × 10−4 M) and insulin (5 mgmL−1). All cell lines were fibroblast feeder cell independent and were cultured at 37°C and 5% CO2. Twenty-four hours before measurements were performed with optical tweezers, cells were seeded in custom-made sample chambers. They were assembled from an acetone cleaned, uncoated, glass cover-slip to which a PDMS insert was attached using plasma.

In proteasome inhibition experiments, MG132 was dissolved in culture buffer at 1 μM concentration and cells were incubated for 2 hrs at 37°C and 5% CO2, after which images were taken on a Nikon TE2000 Eclipse inverted microscope.

Standard protein extraction and Western blotting

All growth media was removed from the culturing flasks and ice-cold RIPA buffer (Sigma-Aldrich, USA) complemented with protease inhibitors was added. Cells were carefully scraped from the bottom with a cell scraper. The cell suspension was collected and centrifuged for 10 min at 12500 rpm and at 4°C. The protein concentration was determined by Pierce™ BCA Protein Assay Kit (Thermo Fischer Scientific, USA). Equal amounts of protein (20 μg) were run on a 4–20% gradient polyacrylamide gel. Subsequently, proteins were transferred to a PVDF membrane. Membranes were blocked in 5% (w/v) skimmed milk in TBS-T (20 muM Tris, 150 mM NaCl, 0.02% Tween-20, pH 7.5), which was followed by overnight incubation in mouse primary antibody against keratin 14 (LL001, gift of prof. dr. E.B. Lane) at room temperature. After washing, membranes were incubated with an anti-mouse secondary antibody conjugated to horseradish peroxidase, for 1 hour at room temperature. Proteins were detected with a LAS-400 analyzer after 5 min incubation with the SuperSignal West Pico Plus Chemiluminescent Substrate (Thermo Fisher). Protein quantification was done using ImageJ software.

Preparation of soluble and insoluble protein fractions

Keratinocytes were grown in culture at 37°C, 5% CO2, using RM+ medium with 10% FBS (Sigma Aldrich, USA) in T25 flask. The medium was changed every other day till they reached 90% confluence. Cells were then washed twice with 2 ml DPBS (Gibco Thermo Fisher Scientific, USA,) and lysed in 120 μL RIPA extraction buffer (Thermo Fisher Scientific, USA) supplemented with protease inhibitors (Merck, Sigma Aldrich, USA), which were freshly added prior to use. Cells were carefully scraped from the bottom with a cell scraper and the collected extract was transferred to an Eppendorf tube (Eppendorf, Germany). This process was repeated once using another 80 μL of RIPA extraction buffer. Total cell lysates were then centrifuged at 12500 rpm and 4°C for 10 minutes. After centrifugation, the supernatant was transferred to new Eppendorf tube and marked as supernatant (S). The RIPA buffer extraction should result with the majority of keratins being solubilized and thus found in the supernatant. The pellet was then treated as cell extract and re-suspended in high salt buffer B (10 μL Tris-HCl, pH 7.6, 140 mM NaCl, 1.5 M KCl, 5 mM EDTA, 5 mM EGTA, 1% Triton X-100, with protease inhibitors, prepared as described in Werner et al. [23], homogenized with a pestle (Sigma Aldrich, USA), and centrifuged at 12500 rpm at 4°C for 10 minutes. The resulting supernatant, which should contain leftover keratin that was not solubized by RIPA buffer (e.g. aggregates. oligomeric particles of intermediate and large sizes), was transferred to a new Eppendorf tube and marked as pellet (P), the RIPA insoluble fraction. All protein extracts were stored at -80°C. This protocol is a modified version of previously published protocols. [23, 24].

Real time microscopy

Cover-slips with live cells were assembled into a perfusion open-closed chamber, a miniature climate box system suitable for cultivation and live cell imaging of eukaryotic cells (POC chamber, H. Saur, Germany). The chamber temperature was kept at 37°C during imaging by a heater. Images (512 x 512 pixels) were collected using a 100x 1.4 NA oil immersion objective on a Nikon inverted microscope TE2000 Eclipse equipped with a motorized Z stage (Applied Precision, USA) and linked to a Micromax CCD camera (Roper Scientific, USA), part of the DeltaVision imaging system (Applied Precision, USA). Time points were collected each 10 seconds during a period of 30 minutes. Images (S1 Raw images) were subsequently deconvolved and analyzed using softWorX software (Applied Precision, USA).

Optical tweezers and indentation experiments

Cell deformation experiments were performed on an Eclipse Ti inverted microscope (Nikon) equipped with laser tweezers (Tweez 250si, Aresis, Slovenia). The optical tweezers (OT) were set on constant optimal power (laser wavelength was 1064 nm). The laser beam was focused through a water immersion objective (60x, NA 1.00, Nikon) and the sample as well as the objective were heated to 37°C. A custom-made sample chamber containing adherent keratinocytes was mounted on the microscope stage and silica beads with diameter 5.06 μm (CS01N, Bangs Labs, Fishers, USA) were added to it. A bead was trapped by optical tweezers and positioned near a single, non-confluent cell, facing the cell side where the cell membrane was most vertical, i.e., the side without extensive lamellipodia. With the help of the piezo microscope stage (Nano-LPS-200, Mad City Labs, USA) the bead centre was positioned approximately 5 nm above the surface. By moving the stage with a constant velocity of 1 μms−1, the cell was pushed into the bead, and its position was monitored in real-time with a digital camera (Zyla 5.5, Andor, Ireland) at 50 fps and by using a custom-written Matlab program (Mathworks, USA). After the bead was pushed for 0.5 μm from the center of the optical trap, the stage was automatically retracted. The force was calculated from F(t) = −κΔx(t), where κ is the stiffness of the optical trap. The deformation is calculated from the known stage position and the bead displacement: ε(t) = x − Δx(t). In the case of OT, fitting of the Hertz-Sneddon model was unreliable because the point of contact was often hard to determine accurately. The cell stiffness was therefore determined as the slope of the linear part of the force-deformation curve. Measurements were performed on at least 35 cells per each cell line.

Proposed reaction network

In light of what was found in the experimental work presented below, we propose a new reaction network (see Fig 1) that demonstrates how the coexistence of WT and mutant keratin affects the intermediate filament network.

Fig 1

Graphical diagram of the proposed reaction network.

Keratins can be found in complex polymeric structures of different sizes. For the keratin polymerization dynamics to be mathematically treatable, and analogously to other keratin models [13, 25], we group the keratin configurations in few different states depending if it forms very small oligomers (soluble keratin), intermediate size oligomers (particulate keratin) or filaments. In the reaction network presented here, we consider 8 different states: 4 for each keratin species (WT and mutant). In the equations describing the reaction dynamics, S and S will represent the concentration of keratin in the soluble phase (W for wild type and M for mutant), P and P are the concentrations of keratin in the particulate phase, F and F the respective concentration in the filamentous phase, and A and A the respective amount of WT and mutant keratin that is in the form of aggregates. We consider that all depolymerization processes occur at the same rate k−. This assumption is valid if we consider that all the assembled keratin forms are deconstructed one unit at a time. is the rate at which WT or mutant keratin found in the soluble state forms particles, while is the rate at which these same particles assemble to form filaments. We propose that when two particles of different types, one WT and one mutant, assemble with reaction rate k, they will integrate a keratin aggregate. We use the parameter γ to reflect the asymmetric formation of aggregates from WT and mutant particles, which is not necessarily in a 1:1 ratio. A possible asymmetric mechanism of aggregate formation is by binding WT and mutant particles of different typical sizes. These aggregates are not static and they can lose keratin back to the soluble phase.

Mathematical model

The reaction network presented above in Fig 1 is modeled mathematically by the following system of ordinary differential equations (ODEs): These equations describe the temporal evolution of the concentrations of the different keratin phases. We assume all reactions can be modeled using first-order kinetics [13, 25] except for the formation of aggregates which depends on the product P P. Since we are only interested in the values of the concentrations when the keratin cycle reaches a dynamical equilibrium, we solve the system of equations until the value of all concentrations are stationary.

To simplify the analysis of the equations, we rewrite the ODEs in a non-dimensional form. We redefine the time variable as τ = k− t and replace the concentrations by their respective fractions (indicated in lowercase) of the total amount of keratin in the cell cytoplasm, K, which is assumed constant. Thus, we are left with the following reformulated system of equations: In these ODEs, the non-dimensional form of the reaction rates are denoted by , , , , λ, with and λ = k K/k−, where K is the concentration of all keratin types (WT and mutant) in the cell. Here we set K = 1 × 10−3 M [12]. The total fraction of mutant keratin in the cell will be denoted by χ = s + a + p + f.

Numerical methods

We then proceed to solve these ODEs numerically. In this case, explicit numerical methods such as the Runge-Kutta 4th order method are not adequate due to the stiffness of the equations, since the reaction rates are within a 4 orders of magnitude range. In these situations, the suitable choice is an implicit solver with an adaptive time-step. Here, we chose the LSODA initial-value problem integrator that combines the Adams-Bashforth explicit multistep method with an implicit Backwards Differentiation Formula of the second order [26]. LSODA also includes an adaptive time-step based on the values of the derivatives at each step [27]. The SciPy (Python 3.X.X) implementation of the LSODA algorithm was adopted to solve the equations due to its ease of use and high performance for small systems with a dense Jacobian [28]. The simulations were run until the stationary state. We considered that the stationary state is numerically reached when the absolute value of the time derivatives of all keratin fractions were below 1 × 10−10 s−1.

Results and discussion

Mutant keratin aggregates are dynamic

One characteristic of mutant keratin aggregates is that they are not static: they form, grow and disappear in the cytoplasm of keratin mutant cells [29-32]. Fig 2 presents steps in EGFP labeled keratin filament dynamics showing the formation and disappearance of keratin aggregates in cells expressing a severe mutation (EGFP-K14 R125P construct) at 50% ratio with the endogenous K14 WT protein. The arrows indicate some of the aggregates, which appear at the cell periphery (Fig 2A), grow in size by merging with other aggregates (Fig 2B), and then quickly disappear within minutes, (Fig 2C and 2D) on their inward trajectory towards the nucleus (see also S1 Raw images). At the same time, where keratin aggregates were previously visible in large numbers, the keratin filament network remodeled and filaments are now visible (Fig 2D).

Fig 2

Fluorescently labeled keratin aggregates are dynamic.

Steps in EGFP labeled keratin filament’s dynamics. (A) Aggregates, some marked by red arrows, appear at the cell periphery. (B) Aggregates merge and grow in size (arrows). (C) Larger aggregates disappear within minutes of their inward trajectory to the nucleus (arrows). (D) New filaments can be seen in areas where aggregates were visible previously.

Proteasome inhibition induces keratin aggregates accumulation in keratin mutant cells

It has been shown that the activity of the ubiquitine-proteasome system may affect disease severity through keratin aggregate turnover [33]. As in our live cell imaging experiments the EGFP-K14 protein containing keratin aggregates not only accumulated in the cytoplasm but also quickly disappeared, we decided to apply MG132, a proteasome inhibitor, to test whether the disappearance of EGFP labeled keratin aggregates is linked to proteasomal activity. In Fig 3 we show representative images of cells before and 2 hours after treatment. After the inhibitor was administered, aggregates grew in size and accumulated in cells (with about 10% of cells with aggregates before treatment vs. 99% of cells with aggregates after the 2hr treatment with MG132), demonstrating that indeed the EGFP K14 R125P keratin aggregates in our mutant keratin expressing cells are actively disposed of, and therefore this process is both dynamic and reversible.

Fig 3

Proteasome inhibition leads to the accumulation of fluorescently labeled keratin aggregates.

(A) EGFP-K14 R125P cells before treatment with MG132 proteasome inhibitor. Aggregates are visible only within some cells. Aggregates are visible only in about 10% of cells. (B) EGFP-K14 R125P cells after 2 hours of treatment with 1 μM MG132. Aggregates are present in 99% of cells, indicating that the proteasome system has a leading role in the turnover of these aggregates.

To look more closely at these aggregates, their possible dependence on the amount of mutant keratin expressed and how their accumulation may affect cell stiffness, we performed clonal selection of our stably transfected EGFP cell lines [19]. Fig 4 shows a Western blot with the ratios between the EGFP labeled K14 construct (WT or mutant) and the endogenous K14 WT protein in clones selected for further analysis. As shown, the selection of clones with closely matching content of K14 was only partially possible, so we decided to select clones that had the desired ratio between the EGFP labeled and endogenous K14 WT, as close as possible to 25%, 50% and 100%. For each type of clone we selected both the two mutants (EGFP-K14 R125P and the EGFP-K14 V270M expressing cells) and the appropriate control cell line expressing EGFP-K14 WT protein (Fig 4).

Fig 4

Western blot analysis of selected EGFP-K14 (WT or mutant) clones.

The upper band belongs to the fluorescently labeled K14 protein, while the lower band to the unlabeled endogenous K14 WT. The numbers 25%, 50% and 100% indicate the ratio between EGFP-K14 to endogenous K14 WT of the clones that were selected for further analysis.

Lower cortical stiffness correlates with the presence of keratin aggregates

We hypothesized that the presence of keratin aggregates affects cortical stiffness, and so cell stiffness of clones from Fig 4 was tested with optical tweezers. In order to obtain a reliable estimate of how the percentage of expressed mutant K14 protein may affect cell stiffness, we performed measurements on at least 35 different cells from each cell line. The results are statistically significant and are summarized in Fig 5. Surprisingly, the stiffness of cells where all K14 is derived from the EGFP-K14 R125P mutant construct (100% content, Fig 5A), did not significantly differ from the stiffness of the control cells expressing 100% EGFP-K14 WT, and their means lie within comparable SEM ranges. A similar result was also obtained for cells expressing 50% of the EGFP-K14 protein (Fig 5B), while a major difference in cell stiffness was detected for the clones where the EGFP-K14 R125P keratin is expressed at a 25% ratio to the endogenous K14 WT (Fig 5C). In this case the EGFP-K14 R125P cells were about 30% less stiff than the EGFP-K14 WT control ones. These results were surprising as a shift in cell stiffness was predominantly expected for cells expressing 100% and 50% EGFP-K14 R125P protein due to the high content of mutant keratin in these cells.

Fig 5

Correlation between cell stiffness and the expression rate of the EGFP-K14 constructs.

Graphs depict how the expression rates of the EGFP-K14 constructs affect cell stiffness in different EGFP-K14 cell lines. All 100% EGFP-K14 cell lines (A) and 50% EGFP-K14 cell lines (B) do not display any major difference in cell stiffness. In striking contrast to this, there is a significant difference in cell stiffness between the 25% EGFP-K14 cell lines (C). In particular, the 25% EGFP-K14 R125P cells are on average 30% softer than the corresponding control 25% EGFP-K14 WT cells.

We then also quantified the number of cells with keratin aggregates in these cell lines. The results are summarized in Fig 6. Lower cell stiffness did correlate with the presence of keratin aggregates, however the number of cells with keratin aggregates was inversely proportional to the expression level of EGFP-K14 R125P protein. Namely, aggregates were detected in the cytoplasm of up to 18% of keratinocytes with a 25% ratio of EGFP-K14 R125P, in comparison to 11% of cells with a 50% ratio of EGFP-K14 R125P and only 0.8% of cells expressing EGFP-K14 R125P protein at a 100% ratio. Representative images of the observed differences are shown in Fig 6B (100% EGFP-K14 R125P), C (50% EGFP-K14 R125P) and D (25% EGFP-K14 R125P).

Fig 6

Quantification of aggregates in cells of EGFP-K14 R125P cell lines.

(A) Summary of data, pointing out to the fact that the least cells with aggregates were seen in the 100% EGFP-K14 R125P cell line (0.8%), followed by the 50% one (10.8%), while the most cells with aggregates were seen in the 25% EGFP-K14 R125P cell line (18%). In the panels below are representative images (fields of view) of each of these cell lines (B-100%, C—50%, D—25%). The arrows point out cells that are representative for cells with aggregates.

Endogenous K14 WT is sequestered in the cell pellet

We further analyzed the total cell extract (RIPA extract) and remaining cell pellet (further treated/extracted by high salt buffer) fractions of our different EGFP-K14 cell lines on Western blot (Fig 7) by testing them for K14. This revealed that the pellet fraction of the 25% EGFP-K14 R125P mutant contains a large amount of endogenous K14 WT, and just a trace amount of the EGFP-K14R125P protein (Fig 7). At the same time, all the EGFP-K14 WT cell lines have little or no endogenous K14 present in the high salt extracts. We therefore reason that unlike the majority of WT and mutant keratin protein, which under RIPA buffer extraction normally ends up in the supernatant of a total cell extract (see also 100% and 50% clones in Fig 7), the large amount of endogenous K14 WT found in the cell pellet in the case of the 25% EGFP K14 R125P cell line may be due to it being sequestered within keratin aggregates, which may be of lower solubility than the filaments.

Fig 7

Western blot of K14 content in EGFP-K14 WT and R125P mutant cell lines in the total cell extract (RIPA supernatant) and the pellet fraction (after extraction with a high salt buffer).

The upper band belongs to the protein derived from the EGFP-K14 construct, while the lower band to the endogenous K14 WT. 25%, 100% and 50% indicate the different EGFP-K14 clones, while S marks the supernatant fraction and P the pellet fraction of the total cell extract(s). Surprisingly, the 25% EGFP-K14 R125P cell line, which also displayed the lowest cortical stiffness and highest number of cells with keratin aggregates, contains a significant amount of endogenous K14 WT in the pellet fraction (over an order of magnitude more than the EGFP-K14 R125P protein in the band above). All other cell lines, including the 50% and 100% EGFP-K14 R125P ones have either no or only trace amounts of both EGFP labelled and endogenous K14 in the pellet fraction.

As these results were unexpected we decided to use a systems biology approach to try to analyze this phenomenon in silico with an enhanced mathematical model. This model will not look into the variability in aggregate concentration within a population of cells with similar keratin mutant levels (which will be the focus of a future work), but it will explore possible aggregation mechanisms that allow for the existence of a peak in aggregation at low mutant keratin fractions, as observed in Figs 6 and 7.

Aggregates formed by asymmetric binding of WT and mutant keratins lead to a maximum in aggregate accumulation at small molar fractions of mutant keratins

The model introduced in this work presents a minimal mathematical description of keratin filament and keratin aggregate formation that considers asymmetric binding between WT and mutant keratins. The values used in the simulation for the reaction rates , , , , have either been measured experimentally or fitted to data by other mathematical models [12, 13] (see Table 1). According to these reaction rates, the formation of mutant filaments is a slower process than the formation of WT filaments. The reaction rate associated with the formation of keratin aggregates, λ, and the factor representing the asymmetry in binding between WT and mutant particles to form aggregates, γ, have not been measured experimentally. As such, we will explore how the different values for these two parameters affect the ratio between the fraction of keratin found in aggregates and in filaments. This ratio between non-soluble particulate keratin phases and keratin filaments will be denoted by Q = (a + a + p + p)/(f + f).

Table 1

Parameters used in the mathematical models (see also S1 Document).

Process	Parameter	Value (s−1)	Adimensional	Reference
Form WT particles	kWSP	1.0 × 10−3	λWSP=1	[12]
Form mutant particles	kMSP	1.0 × 10−3	λMSP=1	[13]
Form WT filaments	kWPF	1.0 × 10−1	λWPF=100	[12]
Form mutant filaments	kWPF	5.0 × 10−4	λMPF=0.5	[13]
Depolymerization	k −	1.0 × 10−3	1	[12]

In Fig 8A we present a typical plot in the stationary state of Q as a function of the total fraction of mutant keratin in the cell, χ = s + a + p + f. With the parameters in Table 1, we always observe almost no aggregate formation when there is no mutant keratin, since when χ = 0 we obtain Q = p/f ≈ 0. Similarly, for χ = 1, there is no aggregate formation and the fraction of particulate keratin in the cell, Q = p/f, is small since the mutant keratin is still able to form a filament network. For intermediate values of χ we will observe the accumulation of aggregates. For a particular value of mutant keratin in the cell, , the observable fraction of aggregates reaches its maximum value, Q.

Fig 8

Fraction of keratin in aggregates in the stationary state obtained by the simulation.

(A) Typical plot of the observable fraction of keratin aggregate in the cell in the stationary state, Q, as a function of the total fraction of mutant keratin in the cell, χ. (B) Molar fraction of mutant keratin that leads to the maximum observable fraction of keratin aggregates, , as a function of γ and λ. When the reaction rate to form aggregates is high and if γ ⪆ 3 we obtain the maximum in the observed aggregate fraction for mutant keratin below 50% of the total keratin in the cell. The full lines border the region where . (C) Maximum observable aggregate fraction, Q, as a function of γ and λ. Q increases sharply with λ. The full lines border the region where .

The values of and Q obtained in the simulation will depend on both γ and λ. In Fig 8B we plot the value of , the molar fraction of mutant keratin that leads to the maximum observable fraction of keratin aggregates, as a function of γ and λ.

For low λ, the maximum in the observed aggregate fraction is obtained when the mutant keratin is above 50% of the total keratin in the cell, independently on the value of γ. When the process of aggregate formation is slow, the amount of aggregates in the steady state will directly depend on the size of the pools of particulate WT and mutant keratins in the cell. And so, since the formation of mutant filaments is a slower process than the formation of WT filaments, the fraction of mutant keratin in the particulate phase is higher than the fraction of WT keratin in the particulate phase. Therefore, the most effective way to increase the fraction of keratin in the particulate phase in the cell is to increase the fraction of mutant keratin. Consequently, aggregate formation will peak at keratin mutant fraction above 50%.

On the other hand, if the reaction rate of aggregate formation is high, we obtain the maximum in the observed aggregate fraction for mutant keratin below 50% of the total keratin in the cell, if γ ⪆ 3. At high λ, aggregate formation is faster than filament formation. Therefore, if γ is large, adding WT keratin is very efficient in increasing the amount of aggregates in the cell, since it readily reacts with mutant keratin in a ratio of γ to 1. However, we observe that if γ ⪅ 1, is always above 50%.

There is a region of parameter space where the experimental results can be reproduced by the mathematical model

The experimental results of Fig 6 discussed above show that the maximum of keratin aggregates is measured for the 25% ratio EGFP-K14 R125P cell line. We identify in Fig 8B the range of parameters where lays between 15 and 30%. We observe that for values of λ in the order of ≈ 103 − 104 (corresponding to k on the order of 103 − 104 M−1s−1 and for γ>8 (including the order of magnitude of the ratio between WT and mutant prevalence in the pellet observed in the data of Fig 7), the maximum of aggregates is always obtained for a ratio of mutant keratin close to the one that leads to the maximum of aggregates in the experiments on cells. For smaller values of γ between 3 and 8, we still observe that a higher value of λ ≈ 104 − 106 (corresponding to k on the order of 104 − 106 M−1s−1) leads to a peak in aggregate formation for 15–30% ratio of mutant keratin.

Computational results predict that for cells with 25% EGFP-K14 R125P the amount of keratin in aggregates is one order of magnitude greater than the keratin filament network

In Fig 8C we plot the value of the maximum observable aggregate function, Q, as a function of γ and λ. Expectably, we obtain a larger accumulation of aggregates for larger λ. Strikingly, we observe that in the region of parameters with on the vicinity of 15 − 30% (bordered by the black lines in Fig 8C), and for γ > 10, the value Q is in the ≈ 10 − 50 range, depending on the value of λ. This means that if we look at the keratin distribution inside a 25% EGFP-K14 R125P cell, the model predicts there will be about one order of magnitude more keratin in aggregates than in filaments (see S1 Document).

Conclusion

Our study advances the knowledge in the field by analyzing the effect that different quantities of mutant keratin protein expressed have on cell stiffness and the presence of keratin aggregates. We show that the two processes are linked, as the larger the number of cells with aggregates, the lower the stiffness of the cells’ cortical area. Our data also indicates a link between the amount of mutant protein expressed and the presence of aggregates.

We also show that the process of keratin aggregation in EGFP-K14 R125P mutant keratin expressing cells is readily dealt with by the proteasome system, as the inhibition of the proteasome with MG132 causes accumulation of fluorescently labeled keratin aggregates. Thus, EGFP labeled keratin aggregates do not accumulate in cells in the same way as in some other human diseases of protein aggregation, like Alzheimer’s or Parkinson’s disease, where they lead to cell death [34]. Nevertheless, these keratin aggregates appear to be insoluble in denaturing extraction buffers and to be able to “trap” significant amounts of endogenous WT protein.

Another interesting aspect of the K14 R125P mutation is that although this is a severe mutation, in our experiments even an EGFP labeled K14 R125P protein is able to form functional filaments when it is present at concentrations close to 100% of total K14 keratin, which is reminiscent of the results that Herrmann and colleagues [8] obtained on purified keratin protein in in vitro polymerization experiments.

The fact that keratin aggregates do not spontaneously appear in all cases of severe keratin gene mutations but that exposure to some kind of physiological stress causing protein misfolding triggers their formation, suggests that the mechanism of aggregate development is linked to basic laws of folding kinetics and native state stability of (keratin) protein. Consequently, different (keratin) mutations may or may not affect protein folding in the same way. In general, mutations create new interaction patterns between amino acids in a protein folding sequence, which destabilizes it and manifests as slowing down of the folding reaction and the formation of transition states, including aggregates [35]. We have taken into account this in our mathematical model of keratin dynamics in cells expressing mutant keratin protein [13], and in this study we have extended this model further by examining different scenarios that are needed for aggregates to form. We then compared the results of mathematical modeling with the data we obtained in experiments on cells. Against our expectations, our in vitro experiments on keratinocytes expressing different amounts of EGFP labeled mutant and WT K14 protein showed a maximum of keratin aggregation in cells when the mutant K14 protein is expressed at concentrations around 25%, while the least aggregation is obtained at concentration over 50% of mutant vs. endogenous WT keratin. We would like to stress that keratin aggregates are not stable nor static structures like aggregates seen in some other protein aggregation diseases (e.g. Alzheimer). Instead, they are dynamic and constantly going through turnover, both produced and degraded by the cells at any given point in time (see S1 Raw images, where keratin aggregates slowly disappear and filaments are being formed instead). In addition, we have not performed any selection of cells during our stiffness measurements, but cells were measured at random, so the cell stiffness data we presented on the graphs in Fig 5 are average values of the measurements performed. The peek of aggregate concentration at low mutant keratin levels was reproduced in our mathematical model if we hypothesized that for keratin aggregates to form, an asymmetric binding of WT and mutant keratin is needed. The extensive exploration of minimal models where WT and mutant keratins bind at a 1:1 ratio predicted, opposite to what was observed experimentally, a maximum of agglomerate formation at mutant keratin fractions above 50% (see S1 Document). We also explored the hypothesis of a nucleating oligomer formed by both WT and mutant keratin that could grow by the addition of WT keratin. This second hypothesis also did not reproduce the experimental results (see S1 Document). Interestingly two previous studies on vimentin, an intermediate filament protein that builds homopolymeric filaments, bear similarities to our mathematical model’s explanation of the findings we obtained on the K14 R125P mutation. Namely, about 25% of mutant vimentin is enough to cause disruption of the endogenous filament network [36], while the influence of the same vimentin mutations on cataract formation in the eye lens of mice [37] showed that animals expressing less than 30% of mutant vimentin had in their tissues cytoplasmic vimentin inclusions that also contained endogenous WT vimentin.

Therefore, our mathematical model predicts that aggregates are the result of the asymmetric binding of WT and mutant keratins, with a ratio γ: 1, where γ > 1. This agrees with the imbalance between the two types of keratin recovered from the pellets. Finally, when the fraction of mutant keratin is such that the concentration of aggregates is maximum, the model predicts that the concentration of keratin in the aggregates is at least one order of magnitude larger than the amount of keratin in the filaments in the cell. Therefore, in these conditions, the keratin aggregates will be ubiquitous in the cell, just as observed. In a future study we will address quantitatively the variability observed in keratin aggregation at the cell level. Introducing variability in the reaction rates of the model will also permit to better explore quantitatively these measured amounts of keratin.

Parameter value estimation and alternative models.

(PDF)

Click here for additional data file.

Dynamics of keratin aggregates inside keratinocytes.

(MPG)

Click here for additional data file.

The original raw images of the two Western blots from Figs 4 and 7.

(PDF)

Click here for additional data file.

17 Aug 2021

PONE-D-21-22878

A Mathematical Model of the Dependence between Keratin Aggregate Formation and the Quantity of Mutant Keratin Expressed in EGFP-K14 R125P Keratinocytes

PLOS ONE

Dear Dr. Travasso,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Oct 01 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Jian Xu, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

3. PLOS ONE now requires that authors provide the original uncropped and unadjusted images underlying all blot or gel results reported in a submission’s figures or Supporting Information files. This policy and the journal’s other requirements for blot/gel reporting and figure preparation are described in detail at https://journals.plos.org/plosone/s/figures#loc-blot-and-gel-reporting-requirements and https://journals.plos.org/plosone/s/figures#loc-preparing-figures-from-image-files. When you submit your revised manuscript, please ensure that your figures adhere fully to these guidelines and provide the original underlying images for all blot or gel data reported in your submission. See the following link for instructions on providing the original image data: https://journals.plos.org/plosone/s/figures#loc-original-images-for-blots-and-gels.

In your cover letter, please note whether your blot/gel image data are in Supporting Information or posted at a public data repository, provide the repository URL if relevant, and provide specific details as to which raw blot/gel images, if any, are not available. Email us at plosone@plos.org if you have any questions.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The paper entitled ``A Mathematical Model of the Dependence between Keratin Aggregate Formation and the Quantity of Mutant Keratin Expressed in EGFP-K14 R125P Keratinocytes" presents a combination of experimental and computational results on the formation of keratin aggregates in keratinocytes.

The paper presents interesting results using a broad collection of methods from cell biology, biomechanics and biomathematics; however, some results are over-interpreted or all data are not presented or properly explained making some rationale hard to follow. The paper needs revisions before publication.

-Total amount of keratin (endogenous + EGFP K14): Have you verified the total amount of keratin in the different conditions (25$\\%$, 50$\\%$ and 100$\\%$)? Does any condition induce an increase/decrease of the total amount of keratin in cells?

- Line 136 (Definition of insoluble fraction (P)): Specify what does P refer to? filaments + particles + aggregates? endogenous only, EGFP K14 only, mix of endogenous and EGFP K14 material?

-Interpretation of $\\gamma$ : $\\gamma$ is used in the model to describe the asymmetric formation of aggregates from wild and mutant particles. From the equations of the model we have: if $\\gamma<1$ (resp. $\\gamma>1$), the per capita rate of formation of $A_W$ from $P_W$ is smaller (resp. larger) than the per capita rate of formation of $A_M$ from $P_M$ for the same concentrations of the other type of particles $P_M$ or $P_W$. The authors write ``$\\gamma$ is used to represent the difference in size between typical WT and mutant particles that are able to bind and form aggregates.''. This sentence is misleading. A simple sentence in the mathematical model section as " $\\gamma$ is used to represent the asymmetric formation of aggregates from wild and mutant particles" should be enough. Related to this issue, problem with the interpretation of results on page 11 and line 427.

- Define $K_{Total}$ in line 206.

- Define the total fraction of mutant keratin in cells $\\chi_M$ in line 212:$\\chi_M=s_M+p_M+A_M+f_M$ with $\\chi_W+\\chi_M=1$

- Results of models are given at stationary state. Specify the criterion used to detect the fact the solutions of the model have reached the steady state value. That is missing in section ``Numerical methods'' starting on line 213.

- Remove the sentence ``Due to the size of the system of equations and to its non-linearity, it is not possible to find the stationary state solution analytically''. By reducing the dimensional system to a 6-dimensional system (using the conservation of law on the total amount), some expression might be obtainable but an only numerical approach as followed in the paper is enough.

- Figure 4: Does the total amount of keratin in cells change when mutants are added?

- Figure 5: At least 35 cells are used to obtain the measurements of cell stiffness. Nothing is mentioned on how these cells are sampled (selected) for each condition. Does the samples used for stiffness measurements are representative of results shown in Figure 6?

- Figure 5: What about the stiffness of cells expressing only endogenous keratin? Does the addition of EGFP K14 construct change the cell stiffness?

- Figure 5: Specify the values referred by stars.

- Figure 5: Change the title of this figure. This title is not representative of the information displayed in the figure.

- Figure 6: Add the results for EGFP K14 WT as a control case.

- Line 283, "..however the quantity of aggregates did not correlate with the expression level of EGFP-K14 R125P protein." How do the authors get this conclusion? No information on the quantity of aggregates is provided, only information about the number of cells with aggregates is available. Rewrite this sentence.

- Figure 7: Add the results for EGFP K14 WT as a control. Usually IF proteins are found mostly assembled in the insoluble fraction. However, for all conditions in presence of mutant, the soluble fraction is larger than the insoluble one. (As previously mentioned, more details should be given on what is the insoluble fraction.)

- In line 307 "..due to it being sequestered within keratin aggregates, which may be of lower solubility than the filaments.". The authors interpret the results from Fig 7 by speculating that aggregates might have a lower solubility than filaments. However, the disassembly rate $k_{-}$ is assumed to be the same for all (particles, aggregates and filaments) in the model.

- Line 319: The definition of Q as the observable fraction of keratin aggregate in the cell cytoplasm is a little misleading. The equation defining Q is the ratio of the proportion of particulate material to that of filamentous material in cells.

- Line 320: Typo in the definition of Q.

- Add a reference to a figure to the sentence starting on line 351.

- On Line 361, give a proper reference to a figure or other instead of just "see above".

- Validation of the model by experiments (paragraph starting on line 349): Figure 6 deals with the number of cells with aggregates, Figure 7 deals with the soluble and insoluble fractions (the insoluble fraction is not clearly defined yet) of endogenous and EGFP K14. But no clear experimental data are shown about the quantity of aggregates in a cell. Add information or explanation or rewrite.

- Further validation of the model by experiments: From Figure 7, information on the total (endogenous + EGFP K14) soluble and insoluble fractions is available. This information is also tractable in the mathematical model. The model should also be able to predict these features too. Knowing the ratio between particulate and filamentous material (defined by Q) is interesting but this information could be accompanied by the distribution of keratin between the soluble and insoluble fractions.

- The experimental results and their interpretation highlight the importance of the solubilization process translated in the model with the term depending on $k_{-}$. A full sensitivity analysis should be carried out to determine the influence of other parameters (not only $\\lambda_{agg}$ and $\\gamma$).

- Line 376: ".. as the more keratin aggregates are present in cells, the lower the stiffness of the cells' cortical area." Similar remark as before, this statement seems to be an over-interpretation of results.

Minor remarks:

- Add a reference to the sentence starting on line 36.

- Typo or missing information on line 82: "(Clontech, , USA)"

- Typo on line 420

- The supplementary material should be polished.

Reviewer #2: The manuscript by Gouveia et. al. entitled “A Mathematical Model of the Dependence between Keratin Aggregate Formation and the Quantity of Mutant Keratin Expressed in EGFP-K14 R125P Keratinocytes” is interesting and well written. The authors investigate how cells with various proportions of mutant keratin and wild type keratin are different. They examine the number of cells exhibiting keratin aggregates, cell stiffness, and the amount of soluble vs insoluble keratin in the cells. Additionally, the authors developed a mathematical model to help explain the surprising experimental results. The paper should be published. I have several comments.

Pg 5 “gamma is used to represent the difference in size …” I think gamma gives the ratio of wt to mutant in the aggregation and stating that would be clearer.

Fig 5 What do the *** mean? It is not explained in the caption

Fig 6 The arrows are not mentioned in the caption

Pg 9 Five times more WT than mutant but only 18 percent of cells show aggregates. You should be able to approximate how much more WT is in the aggregates. Also is the mutant consistent with the 18 percent – the soluble is much darker than the pellet is it greater than 18 percent (since the pellet is more than just the aggregate)?

Fig 7 Why change the order in this figure to 25% 100 % 50%? Also the 100 % had the least aggregate but seems to have more pellet than the 50 %?

Pg 10 Why is Q not (aW+aM)/(fW+fM)? Why include the particulate keratin? It seems figure 8 would make more sense (or at least in the explanation) if the aggregate keratin was reported from the model as well as Q (or instead).

Pg 10 line 372 “magnitude more keratin in aggregates than in filaments” The model says Q should be that not the aggregates (Q is more than just the aggregates).

More explanation of the stiffness results should be included.

Minor issues:

Pg 6 line 203 concentrations is stationary change to concentrations are stationary

Pg 10 Q=(….)/(fM+fM) should be (…)/(fM+fW)

Pg 12 line 420 “model?s”

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

1 Nov 2021

Please check attached pdf.

Submitted filename: Reply.pdf

Click here for additional data file.

25 Nov 2021

A Mathematical Model for the Dependence of Keratin Aggregate Formation on the Quantity of Mutant Keratin Expressed in EGFP-K14 R125P Keratinocytes

PONE-D-21-22878R1

Dear Dr. Travasso,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Jian Xu, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have addressed all my questions and comments and clarified the confusing parts of the original material. I have no further requests.

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

14 Dec 2021

PONE-D-21-22878R1

A Mathematical Model for the Dependence of Keratin Aggregate Formation on the Quantity of Mutant Keratin Expressed in EGFP-K14 R125P Keratinocytes

Dear Dr. Travasso:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Jian Xu

Academic Editor

PLOS ONE