The HaCaT cell line (spontaneously immortalized non-carcinogenic human keratinocytes) is widely considered a model for studying the function of normal human keratinocytes [1, 2]. However, stratification program and differentiation marker expression are abnormal in the HaCaT line keratinocytes [3].
It is known that there are two allelles of the TP53 gene the HaCaT keratinocyte genome that comprise two gain-of-function (GOF) mutations acquired via spontaneous immortalization (H179Y and R282Q) [4]. MutTP53 exerts significant effects on the increase in proliferation rate and cell growth in the HaCaT cell line and has more than 7000 DNA binding sites. Furthermore, protein functions associated with apoptosis triggering are preserved [5]. It is known that mutant р53 functions in at least two ways: it affects the function of p63/p73 inhibiting their binding to DNA [6] or binds to new DNA sites via interaction with other transcription factors (NF-Y, E2F1, NF-KB, VDR, p63) [7, 8].

Among fundamental physiological characteristics of the HaCaT cells, the most interesting are as follows: proliferation, migration, stratification, and creation of three-dimensional organotypic structures.
We have shown previously that alterations in mutp53 activity in the HaCaT cells result in the altered expression of differentiation markers (such as caspase 14, involucrin and transglutaminase 1) and are associated with altered expression of p63 [9].

Understanding the features of the НaCaT cell physiology and underlying mechanisms is essential for assessment of limitations associated with using such cells as model ones. Moreover, studying cells containing mutTP53 makes it possible to obtain new data for carcinogenesis studies [10, 11].

The study was aimed to assess the features of alterations in migration capabilities of the mutTР53 knockout HaCaT cells.

METHODS

Cell lines and culture conditions

The HaCaT cell line was acquired from the cell culture collection of the German Cancer Research Center (DKFZ, Heidelberg; Germany). The cell culture of the mutTР53 knockout (dp53) HaCaT keratinocytes was obtained earlier [13]. The cells were cultured at 37 °С with 5% СО₂ in the DMEM/F12 medium (1:1, Gibco; USA) supplemented with 1% GlutaMAX (Thermo Fisher Scientific; USA), penicillin/streptomycin solution with a concentration of 100 U/mL and 100 μg/mL, respectively (Gibco; USA), and 10% fetal bovine serum (Dia-M; Russia).
The cell culture of normal human keratinocytes (NHK, pool of five donors) was acquired from Perspektiva SPA LLC (Novosibirsk, Russia). The cells were cultured at 37 оС with 5% СО₂ in the Keratinocyte SFM medium (Gibco; USA) containing 1% GlutaMAX (Gibco; USA), 1% antibiotic/antifungal (Gibco; USA), 50 μg/mL of BPE (Gibco; USA), 10 ng/mL of EGF (Gibco; USA).
The cells were grown in culture flasks with a surface area of 25 cm² or in Petri dishes with a diameter of 60 mm (Corning; USA). The medium was replaced with the fresh one every other day.

Cell cycle analysis

The cell cycle was assessed based on the Edu incorporation (Click-iT Plus EdU Alexa Fluor 488; ThermoFisher, USA) and the amount of DNA estimated by staining with Propidium Iodide (PI) (Sigma; USA). The cells were seeded into wells of the six-well plate and grown before reaching 50% confluency. Then Edu was added, and the cells were stained in accordance with the manufactirer's guidelines. After that the cells were incubated with RNase solution (100 μg/mL) and PI (1 μg/mL) for 30 min. Detection was performed using the ZE5 flow cytometer (Bio-Rad; USA) and the Everest 2.4.0.1365 software (Bio-Rad; USA).

The strongly positive Edu staining was typical for cells in S phase. The strongly positive PI staining was typical for cells with the doubled DNA content being through the G2/M phase of cell cycle.

Comparison of cell proliferation rates

The cell proliferation rates of the wild type (WT) HaCaT cells and mutTP53 knockout (dp53) cells were compared using CytoTrace Red CMTPX (AAT Bioquest; USA). The cells were seeded into wells of the six-well plate and grown before reaching 50% confluency. Then the cells were stained with CellTracer (according to the manufacturer's guidelines) and cultured for 24 h. Detection was performed using the ZE5 flow cytometer (Bio-Rad; USA) and the Everest 2.4.0.1365 software (Bio-Rad; USA). The decrease in fluorscence intensity (dilution of the label) reflects the number of cell divisions.

Scratch test: assessment of the cell monolayer defect closure rate

To perfrom the scratch test, 50,000 of the wild type (WT) HaCaT cells, mutTP53 knockout (dp53) HaCaT cells or normal human keratinocytes (NHK) were added to the wells of the 24-well plate and pre-incubated in the complete culture medium at 37 °С with 5% СО₂ until the monolayer was formed in the well. Then standardized damage to the monolayer was caused using the plastic scraper, the cells were washed with the DPBS solution and grown in the complete culture medium for three days. Each well was photohraphed along the entire length of the scratch every day. The images were processed using the skimage library [12]. The area not occupied by cells (defect area) was calculated 0, 24, 48, and 72 h after scratching, and the ratio compared to the baseline defect area (relative defect area) was defined. The experiment was run in triplicate.

Proteomics data

The analysis of the earlier acquired proteomics data [13] deposited in the ProteomeXchange Consortium (available from http://proteomecentral.proteomexchange.org/cgi/ GetDataset?ID=PXD030700) was carried out using the MaxQuant software (v1.6.3.4). Three biological replicates per line (wild type HaCaT cells or mutTP53 knockout HaCaT cells) were analyzed in three technical replicates.

Estimation of gene expression

RNA was extracted using the RNeasy Kit (QIAGEN; USA) in accordance with the manufacturer's protocol. The amount of RNA obtained was measured with the NanoDrop 2000c spectrophotometer (Thermo Scientific; USA). The reaction of reverse transcription was carried out using the MMLV RT kit (Evrogen; Russia) according to the standard protocol by adding 1 μg of RNA at a time to the reaction. qPCR was performed uding qPCRmix-HS SYBR+LowROX (Evrogen; Russia). The reaction was repeated three times for every gene and sample. GAPDH was used as a reference gene. The primers used are provided in table.

Generation of multilayered skin equivalent

To generate the organotypic model, wild type (WT) HaCaT cells, mutTP53 knockout HaCaT cells or normal human keratinocytes (NHK) were grown in the Millipore PIHP01250 polycarbonate cell culture inserts with the pore size of 0.4 μm used in a 24-well plate. The membrane inserts were placed in the plate wells, after that 700,000 HaCaT cells (wild type or mutTP53 knockout cells) were added to the apical compartment and 700 μL of the Submerse medium (EpiLife (Gibco; USA) containing 1% GlutaMAX (Gibco; USA), 1% antibiotic-antimycotic (Gibco; USA), 1% HKGS (Gibco; USA), 10 ng/mL of EGF (Gibco; USA), and 1.5 mmol of CaCl2 (Sigma; USA)) were added to the basolateral compartment. The cells were incubated for 24 h (37 °C, 5% СО₂) to ensure cell attachment and proliferation activation. After 24 h the medium was completely removed from the apical compartment, while in the basolateral compartment it was replaced with 450 μL of the growth medium at the interface of phases (ALI-medium: Submerse + 10 ng/mL of KGF (Gibco; USA) and 1 mmol of L-ascorbic acid (Sigma; USA)). The cells were grown for 10 days, and the growth medium was replaced by fresh medium every day.

Immunofluorescence staining

The cell models obtained were fixed in 4% formalin and processed by standard histology methods: dehydrated in the series of alcohols with the increasing concentrations, embedded in paraffin blocks, and sliced using microtome.
To perform immunofluorescence staining, the cells were incubated with primary antibodies against KRT5 (ab52635; Abcam, USA), CK10 (ab9025; Abcam, USA) and Alexa Fluor 488-conjugated (ab150105; Abcam, USA) or Texas Red-conjugated (ab6793; Abcam, USA) secondary antibodies. All the specimens were stained simultaneously using the same reagent kit (dilutions of antibodies and buffers).

Estimation of electrical resistance

Transepithelial electrical resistance (TEER) of the skin models generated using cells of various types was defined with the EVOM voltmeter (World Precision Instruments, Inc.; USA). For that growth medium in the basal compartment was replaced with 0.9% NaCl and 300 μL of 0.9% NaCl were added to the apical compartment of the membrane insert. After that electrodes were placed in the apical and basolateral compartments of the membrane insert, and electrical resistance (TEER) of the membrane insert with the cultured cells (skin equivalent) was measured.

Data analysis

The results obtained were processed using R programming language for statistical analysis [14]. The intergroup differences between the studied parameter values were defined using the t-test with the Benjamini-Hochberg procedure for multiple testing correction. The differences were considered significant at p < 0.05. The data are presented as M ± m.

RESULTS

Assessment of the actively proliferating cells based on the Edu (analogue of Brdu) incoporation in DNA showed that the wild type HaCaT proliferated more actively than the mutTP53 knockout cells. In S phase there were 45.6 ± 3.2% wild type cells (HaCaT WT) on average, while the average number of mutTP53 knockout cells (HaCaT dp53) was 34.1 ± 2.9% (fig. 1). These data were also confirmed by the results of the CellTracer label dilution assay performed after 24 h of cell growth.

The scratch test showed that the defect closure rates of the NHK and WT HaCaT cell lines were almost equal. On day three there were 54% ± 9% and 66% ± 5% of the baseline defect areas for NHK and WT HaCaT, respectively. At the same time the mutTP53 knockout HaCaT cells showed a significant increase in the defect closure rate: on day three there were 12% ± 3% (p < 0.05 compared to the corresponding group of WT HaCaT) of the baseline defect area (fig. 2А, B).

A comparative panoramic proteome study of the wild type and mutp53 knockout HaCaT cells was previously conducted [13]. When performing data analysis a total of 2080 proteins were identified based on two or more peptides (potential contaminants, proteins identified based on one site, and false-positives were excluded). Among the proteins identified, 27 proteins showed differences in expression between the assessed lines (FDR < 0.05). Alterations in expression of LAMC2 and LAMB3 proteins were the most interesting: the expression was significantly higher in the mutTP53 knockout cells (4.95 and 4.58 times higher, respectively).

The data of proteomics study focused on LAMC2 were confirmed by real-time PCR (fig. 2): the LAMC2 expression was significantly higher in the mutTP53 knockout cells (9.96 ± 1.92, p < 0.05). However, there were no significant differences in the LAMB3 expression between the lines assessed.

When growing 3D cultures, all three lines formed multilayered skin equivalents (:media_3А). However, only normal human keratinocytes (NHK) showed stratification typical for normal human epidermal cells and formed the stratum corneum.

The well-defined stratum corneum was observed in the organotypic model of epidermis formed by normal human keratinocytes (NHK).

There were no clear cell stratification and stratum corneum formation in the organotypic models of epidermis formed by the wild type (WT) HaCaT keratinocytes and mutTP53 knockout HaCaT cells. KRT5 expression was observed in all layers, and the expression levels were comparable (fig. 3А).

When measuring transepithelial electrical resistance (TEER) in the organotypic model of epidermis formed by normal keratinocytes, electrical resistance increased throughout the observation period.

Organotypic models of epidermis formed by various cell types were different in terms of both maximum transepithelial electrical resistance values and the dynamic changes in transepithelial electrical resistance observed within five days (fig. 3B).

The highest transepithelial electrical resistance (TEER) value was found in the model of epidermis formed by normal keratinocytes (NHK): 5712 ± 146 ohm·cm2. In the models formed by the wild type HaCaT keratinocytes and mutTP53 knockout HaCaT cells, the TEER values were almost five times lower: 964 ± 82 and 1088 ± 91, respectively (p < 0.05 compared to NHK).

Transepithelial electrical resistance increased faster between days two and four with subsequent getting to plateau in the organotypic model of epidermis formed by normal human keratinocytes. The values of this parameter in the model of epidermis formed by the wild type HaCaT keratinocytes and mutTP53 HaCaT cells were lower compared to the model formed by NHK and got to plateau by day two.

The increase in transepithelial potential is associated with the formation of intercellular contacts and stratification of epithelium followed by the stratum corneum formation [15] that is observed in the model of epidermis formed by normal keratinocytes and is absent in the models formed by HaCaT keratinocytes, regardless of the mutTP53 knockout (fig. 3А).

DISCUSSION

It is known that mutant p53 in the HaCaT cells is associated with the increased cell growth and proliferation rates, which has been confirmed experimentally in the previously reported studies [5]. It is also associated with the cells' inability to normally differentiate and form the full-fledged stratum corneum.

According to the comparative proteomics study of primary human keratinocytes and HaCaT keratinocytes, one of the most noticeable differences is the expression of laminin-5 subunits, the α3, β3, and γ2 chains (LAMA3, LAMB3, LAMC2), that is significantly decreased in the HaCaT cells [16]. When discussing the findings, the authors of the above paper assume that this feature determines the abberrant nature of the HaCaT cells' differentiation program.

Laminins, being the main component of extracellular matrix, play an important role in cell adhesion, differentiation, and signal transmission [17]. The laminin-5 complex is essential for stabilization of dermoepidermal junction via binding to integrins α3β1, α6β4 and type VII collagen. Laminin-5 also plays a vital role in cell migration. It has been shown that downregulation of the gene encoding laminin-5 in keratinocytes results in reduced keratinocyte adhesion and impaired epidermal regeneration [18]. However, the increased expression of LAMB3 and LAMC2 subunits is often associated with cancer. As a result, tumor cells are characterized by higher migration rate and invasiveness [19].

Our findings show that mutТР53 knockout in the HaCaT cells does result in the increased expression of LAMB3 and LAMC2 (the data of cell proteomics study). However, it is worth mentioning that the scratch test has revealed no significant differences in the defect closure rates between normal keratinocytes and the wild type HaCaT cells on day three after scratching, while the defect closure rate of the mutTP53 knockout cells is significantly higher. High defect closure rate shown by the mutТР53 knockout HaCaT cells in the scratch test should be associated with the cells' migration activity, but not with their proliferation rate, since in these cells it is significanly lower than in the wild type HaCaT cells.

Apparently, mutations in ТР53 have no significant effect on the gene's ability to regulate migration, while gene inactivation (knockout) results in the significantly increased cells' ability to migrate, including compared to normal human keratinocytes. As a result, the mutTP53 knockout HaCaT cells acquire the features of cells with pro-carcinogenic phenotype.

It should be also noted that the elevated expression of laminin-5 resulting from the mutTP53 knockout has shown no effect on the HaCaT keratinocytes' ability to form the full-fledged stratified epithelium in the organotypic model of epidermis. This is manifested in the lack of the stratum corneum formation and low transepithelial electrical resistance (TEER) comparable to that of the wild type cells.

Thus, our findings confirm the regulatory role of mutant p53 in the HaCaT cells in terms of laminin-5 expression, but make it impossible to associate low expression levels with abberrant differentiation program in these cells.

CONCLUSIONS

Mutant TP53 in the HaCaT cells is a negative regulator of laminin-5 expression. The mutTP53 gene knockout results in the increased HaCaT cells' migration activity, but does not promote normalization of differentiation program. The data obtained supplement the information about the functional features of mutant TP53 in the НaСaТ cells.