Development and validation of novel keloid-derived immortalized fibroblast cell lines

Keloids are a common connective tissue disorder with an ill-understood etiopathogenesis and no effective treatment. This is exacerbated because of the absence of an animal model. Patient-derived primary keloid cells are insufficient as they age through passaging and have a limited supply. Therefore, there is an unmet need for development of a cellular model that can consistently and faithfully represent keloid’s pathognomic features. In view of this, we developed keloid-derived immortalized fibroblast (KDIF) cell lines from primary keloid fibroblasts (PKF) by transfecting the human telomerase reverse transcriptase (hTERT) gene. The TERT gene encodes the catalytic subunit of the telomerase enzyme, which is responsible for maintaining the cellular replicative potential (cellular immortalization). Primary fibroblasts from keloid-specific lesional (peripheral, middle, and top) as well as extralesional sites were isolated and evaluated for cell line development and comparative cellular characteristics by employing qRT-PCR and immunofluorescence staining. Moreover, the immortalized behavior of KDIF cell lines was evaluated by comparing with cutaneous fibrosarcoma and dermatofibrosarcoma protuberans cell lines. Stable KDIF cell lines with elevated expression of hTERT exhibited the cellular characteristics of site-specific keloid fibroblasts. Histochemical staining for β-galactosidase revealed a significantly lower number of β-gal–positive cells in all three KDIF cell lines compared with that in PKFs. The cell growth curve pattern was studied over 10 passages for all three KDIF cell lines and was compared with the control groups. The results showed that all three KDIF cell lines grew significantly faster and obtained a fast growing characteristic as compared to primary keloid and normal fibroblasts. Phenotypic behavior in growth potential is an indication of hTERT-mediated immortalized transformation. Cell migration analysis revealed that the top and middle KDIF cell lines exhibited similar migration trend as site-specific PKFs. Notably, peripheral KDIF cell line showed significantly enhanced cell migration in comparison to the primary peripheral fibroblasts. All KDIF cell lines expressed Collagen I protein as a keloid-associated fibrotic marker. Functional testing with triamcinolone inhibited cell migration in KDIF. ATCC short tandem repeat profiling validated the KDIF as keloid representative cell line. In summary, we provide the first novel KDIF cell lines. These cell lines overcome the limitations related to primary cell passaging and tissue supply due to immortalized features and present an accessible and consistent experimental model for keloid research.


Introduction
Keloids are common fibroproliferative reticular dermal lesions of unknown etiopathogenesis, characterized clinically by an aggressive exophytic expansive growth into the surrounding skin with a high rate of recurrence post-therapy (1).Lack of relevant study models (including animal models as keloids only occur in human skin) (2) has been a challenging and rate-limiting issue for keloid research (3).Therefore, despite its inherent limitations compared with in vivo animal models, the use of in vitro monoculture studies using keloid-derived primary cultured fibroblasts has been invaluable in enabling studies in keloid pathobiology.
However, patient-derived primary cultured keloid fibroblasts are often scarce and insufficient as they age through passaging and have a limited supply.Additionally, keloid tissue is biologically heterogeneous with clinically distinct features within the lesion, including variable dermal cellularity, inflammatory infiltrate, and altered collagen I-III ratios (4), at margins (actively growing) compared with the center (dormant) of keloid (5), representing a variation at the cellular and molecular levels between central and peripheral keloid fibroblasts (6).This endotypic and phenotypic variation typically divides keloid into three distinct lesion-specific sites: (i) intralesional (keloid center), (ii) perilesional (keloid margin), and (iii) extralesional (normal appearing skin surrounding the lesion but distant to the margin) (7).
It has been established that in vitro disease-specific cell lines have lent themselves as an ideal option (a well-controlled system) to investigate the phenotypic and cellular characteristics of specific diseases.However, in vitro subculturing and maintenance cause aging (telomeric loss) in patient-derived primary cells (8).Significant telomeric loss was reported in keloids fibroblasts (9), resulting in a shorter lifespan and limiting the utility of keloid primary cells.Hence, induced overexpression of human telomerase reverse transcriptase (hTERT) has been envisioned as an attractive approach to extend lifespan of patient-derived primary cells, which is known as an "immortalized cell-line model approach."Induced immortalization bypasses cell events like replicative senescence and cellular crisis (10,11).Thus, these cell lines act as a gold standard as they remain genetically identical, accessible, and consistent (12).Fibroblast cells have been considered as potential candidates to develop immortalized cell line for keloid disease as they have been implicated as an important contributor to keloid pathobiology and subsequent tissue formation (13,14).
In view of this, we sectioned freshly obtained biopsies of keloid tissue into four specific sites: [i] top (papillary dermis of the intralesional center of keloid), (KT) [ii] middle (reticular dermis of the intralesional center of keloid) (KM), [iii] peripheral (margin of the keloid lesion) (KP), and [iv] extralesional tissue (KE) from the same donor (Figure 1C).Subsequently, these site-specific keloid tissue sections were used to isolate and develop keloid-derived immortalized fibroblast (KDIF) cell lines.The present study evaluated and compared the hTERT expression, immortalization, and characteristics/behavior (linked with immortalization) in primary keloid fibroblasts (PKFs) before and after genetic transfection (immortalization) (KDIF), having four distinct types of control groups.The control group included (i) normal fibroblasts (NFs) isolated from normal skin tissue, donated by a non-keloid-forming participant; (ii) extralesional keloid fibroblasts (KEs), as NFs isolated from normal (in appearance) skin adjacent to the keloid tissue, donated by a keloid participant; (iii) dermatofibrosarcoma protuberans (DFSP) fibroblasts, isolated from DFSP tissue; and (iv) fibrosarcoma fibroblasts (FSs) as established cancerous cell cultures, served as positive control for hTERT expression.Primary fibroblast cells from the PKF, KE, and NF groups are the representative of definite lifespan (nonimmortalization character), whereas primary DFSP tissue-derived fibroblast cells and FS cells are the representative of indefinite lifespan (immortalization behavior).Hence, the evaluation of immortalization behavior was crucial to define a stable immortalization in genetically transformed primary fibroblasts.Once immortalization behavior was developed and established (stable transfection), KDIF cell lines were authenticated through short tandem repeat (STR) profiling by Amercican Type Culture Collection (ATCC).To our knowledge, we provide the first authenticated and functionally validated immortalized cell line for keloid disease.

Materials and methods
Here, we provide the concise summary of the key methods and procedures employed in the current study.The detailed method and associated references are available in SI Appendix 1.

Tissue sampling and reference cell lines
Keloid participants (N = 3) were selected as donors for keloid lesional cutaneous tissue based on the study criteria.Keloid dermal tissue was dissected into four site-specific groups: (1) peripheral (margin of the keloid) (2), middle (deep reticular component of the center of keloid) (3), top (superficial papillary component of the center of keloid), and (4) keloid extralesional skin.Normal control skin tissue (non-keloid formers with no family history of keloid disease) samples were collected from healthy female participants (N = 3) during breast reduction surgery.Skin sarcoma tissue sample was donated from cases with a clinically and histologically confirmed DFSP condition (N = 2).All skin tissue samples were obtained from participants after receiving informed ethical consent.This study also included two fibroblastic sarcoma cell lines [Fsarc-01: HT-1080 (HT1080) (ATCC® CCL-121 ™ ) and Fsarc-02: HT-1080-Luc2 (ATCC® CCL-121-LUC2 ™ )] and one DFSP cell line (DFSP-CL) (ATCC-Hs 63T) (ATCC CRL-7043) purchased from ATCC (USA) as the reference cell lines for comparitive studies (Supplementary Table 1).All keloid, normal, and DFSP skin samples were processed for primary fibroblasts culturing by the collagenase method and cultured according to the standard protocol (detailed method is available in SI Appendix 1).

hTERT expression (pre-transfection)
The hTERT gene (quantitative real-time PCR) and protein expression [immunofluorescence (IF) staining and flow cytometry] were studied in PKFs before transfection and compared with the control groups.

hTERT plasmid transfection and generation of stable cell lines
An hTERT immortalized cell system was used to develop immortalized keloid-derived fibroblast cell line through the following steps.(1) All three groups of PKF cultures were transfected with plasmid DNA (pGRN145, MBA-141 ™ , ATCC) containing hTERT gene, separately.(2) The transiently transfected cell lines were treated with hygromycin B (HygB; an antibiotic marker for transfection), and resistant stable cell lines were selected and cloned.(3) Analysis of hTERT gene transfection at mRNA level was evaluated via quantitative real-time PCR (qRT-PCR), and protein expression was evaluated by flow cytometry.(4) The validation of cellular function of hTERT protein in cell senescence was assessed by senescence-associated ß-galactosidase histochemical staining assay in all three keloid-derived transformed fibroblast cell lines.

Defining cellular characteristics of keloid-derived immortalized fibroblasts
The stable transfected cell lines were evaluated for (a) cell viability, (b) cell growth curve, (c) cell cycle analysis, (d) cell migration, (e) cell invasion, (f) cellular senescence, and (g) hTERT protein expression via IF staining.The KDIF cell lines were validated by testing drug (triamcinolone) sensitivity on apoptosis, viability, and cell migration.

Statistical analysis
All experiments were conducted in triplicate, and the results were presented graphically as mean ± standard deviation (SD), 95% confidence interval, and percentage (%) when appropriate.The statistical analysis was carried out by Student's t-test, by using Microsoft Excel version 8.The data were also evaluated by applying two-way ANOVA and Tukey post-hoc test.Experiments were performed in triplicate and presented as mean ± S.D. The statistical significance level was set at *p < 0.05, **p < 0.01, and ***p < 0.001 (15).

Selection of stably transfected cell lines, propagation, and analysis
Post-transfection screening for stably transfected cells lines was carried out by selection of HygB (concentration ranges from 0 µg/mL tp 1,000 µg/mL)-resistant cells at 48-h antibiotic treatment, through the kill curve method and cell viability evaluated by employing 3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide (MTT) assay (Supplementary Figures S2A-C).The antibiotic-resistant clones were then used for studying the cellular characteristics of KDIF cell lines (Supplementary Table 2).
The cell growth curve pattern had been studied over 10 passages for all three KDIF cell lines and compared with control groups.The results showed that all three KDIF cell lines grew significantly faster and obtained a fast growing characteristic at passage 9 (Figure 3E) compared to primary keloid and NFs.This phenotypic behavior in growth potential indicated the result of spontaneous hTERTmediated immortalized transformation.Observations about cell viability and growth curve were further investigated by quantification of total DNA content at the G0, G1, S, and G2 phases of cell cycle in all three cell line's populations at 48-h and 12-day time points.Among all KDIF cell lines, PT-KT-045 cell line showed the highest percentage of cell population at the G1 phase (growth phase) (51.51 ± 0.23% p < 0.016) and at the S phase (36.15 ± 1.23% p < 0.002), at 48-h time point, that increased up to (83 ± 1.22% p < 0.001) the 12-day time point (Figure 3F), representing a fast growing cell line during early and late culture time points, which shows its active cell growth potential throughout the cell culturing.

Validation of keloid-derived immortalized fibroblast cell lines
The immortalization of cells, through overexpression of telomerase (hTERT), has the advantage of maintaining a stable cellular phenotype as the cells remain diploid, but primary cultures from independent isolations can vary.Their overall growth characteristic and cellular response can also be altered due to introduced changes in their genetic makeup during the process of immortalization (18).Therefore, it is necessary to investigate and compare the cellular response of KDIF cell lines to see whether these cell lines represent keloid cell behavior or characteristics that would further validate their candidature as a representative model.Therefore, we studied the protein expression of collagen I in all three KDIF cell lines via IF staining.All three KDIF cell lines showed an increased protein expression of collagen I (Figure 4).  to treat keloids (22).Hence, KDIF cell lines were subjected to functional analysis and drug response studies for verapamil and triamcinolone.KDIF cell lines showed sensitivity to triamcinolone (100 µM each), which decreased cell viability significantly up to 20 ± 1.3% (p < 0.02) in all KDIF cell lines compared to the control group (untreated) (Figure 5A).Further investigation using annexin V apoptotic marker through flow cytometry analysis showed an increased percentage of cell population, positive for pre-apoptotic signal (Annexin V), in response to verapamil and, particularly, triamcinolone drug treatment (100 µM each) in KDIF (PT-KP-045, 22.4 ± 1.2%, p < 0.001; PT-KT-045, 21.7 ± 0.6%, p < 0.003; and PT-KM-045, 18.8 ± 0.7%, p < 0.03) as well as PKF cell line group (KP, 14.3 ± 0.2%, p < 0.006; KT, 16.2 ± 0.04%, p < 0.01; and KM, 15.5 ± 0.3%, p < 0.025) compared to the vehicle group (Figure 5B).These results confirmed the cellular response of KDIF cell lines toward drug treatment and represent PKF cell line's drug sensitivity particularly for triamcinolone.Additionally, the effects of verapamil and triamcinolone (100 µM) on cell migration of KDIF and PKF cell lines were evaluated by employing an in vitro scratch assay at 24-h and 48-h time points.Triamcinolone significantly reduced the cell migration rate at both time points in all KDIF (PT-KP-045, 23.38 ± 4.50%, p < 0.0002; PT-KT-045, 13.64 ± 9.2%, p < 0.004; and PT-KM-045, 13.82 ± 2.11%, p < 0.009) as well as PKF cell lines (KP, 18.08 ± 6.9%, p < 0.001; KT, 11.84 ± 3.4%, p > 0.7; and KM, 15.40 ± 1.8%, p < 0.009) (Figure 5C), providing another supportive information about the representativeness of KDIF in terms of keloid behavior.

Discussion
Lack of a relevant study model and the absence of an animal model pose a challenge in studying keloid pathobiology.In addition, patientderived cultured PKFs are sub-optimal as they are heterogeneous in nature depending on lesional site of origin and have a limited lifespan as they age through passaging with a limited tissue supply.Thus, there is an unmet need to develop a robust and relevant cellular model that can faithfully represent keloid's pathognomic features.
Thus, we present, for the first time, KDIF cell lines from primary fibroblasts obtained from specific lesional sites (peripheral, middle, and top) within keloid tissue.We employed the strategy of overexpressing the hTERT gene to support the immortalized characteristic phenotype in keloid cell lines.For this purpose, specifically, we used the non-viral plasmid vector containing hTERT gene (pGRN145) as compared to available viral vectors (i.e., SV40), to eliminate any possibility of viral contamination in the genome of keloid fibroblast and to prevent the potential emergence of a cancerous phenotype in these cells.This approach was developed on the basis of the observation/knowledge that keloids exhibit a distinct nature compared to malignant tumors, which is clearly shown in our study by including two control groups of soft tissue carcinomas (1): DFSP and (2) FS.Through experiments utilizing qRT-PCR, flow cytometry, and IF techniques, we observed that the expression of the hTERT gene and protein in all three keloid site-specific primary fibroblasts (KT, KM, and KP) was lower when compared to DFSP and FS (as shown in Figures 1D-F).
Low hTERT expression in all PKF is responsible for limited lifespan in primary fibroblasts that can be counteracted by increased activity of telomerase, thus preserving telomere length and cellular functions (23).Hence, we envisioned this approach to extend the lifespan of patient-derived primary cells, which is known as the "immortalized cell line model approach" in order to induce overexpression of hTERT.Subsequently, we developed a KDIF cell line by introducing hTERT gene via transfection in all three site-specific PKF cells (peripheral, middle, and top keloid fibroblasts).Additionally, we demonstrated that stable KDIF cell lines showed enhanced expression of hTERT.The introduction of the hTERT gene is a widely used strategy to extend the lifespan of many cell types, and successful immortalization has, for instance, previously been reported in human retinal pigment epithelial cells (24).
Reconstitution of telomerase activity by induced expression of hTERT results in an immortal phenotype in various types of normal human cells, including fibroblasts.Despite transformation characteristics, it is unclear whether hTERT-immortalized cells are physiologically and biochemically the same as their normal counterparts (25).In view of the fact that continuous cell expansion always provides a selective advantage for rapid growth, the cellular phenotype can occasionally become biased because of the overgrowth of the most rapidly dividing cells, rather than the best differentiating cells (18).Therefore, we also evaluated the cellular as well as the functional characteristics of KDIF cell lines.Significantly improved cell viability and cell growth curve showed a fast growing trend in all three KDIF cell lines as a result of spontaneous hTERTmediated immortal transformation.Observations about cell viability and growth curve were further evaluated by studying the cell cycle phase, in all three cell lines at 48-h and 12-day time point by flow cytometry.PT-KP-045 cell line was noticed as actively growing cells at the late stage of culturing and surpassing senescence.Furthermore, it was observed that all KDIF cell lines exhibited decreased senescence activity.Moreover, KDIF cell lines exhibited significantly increased cell migration and invasion, owing to hTERT overexpression, which is known to promote cell migration (16,26,27).
It is well recognized that increased Collagen I protein expression is associated with dermal fibrosis (28).Previous studies reported a significantly elevated protein expression of collagen 1 in keloid fibroblasts particularly at the growing margins of keloid scars (4).In this study, IF microscopy showed the increased expression of collagen 1 protein in every KDIF cell line, with notably abundant presence in PT-KP-045 cell line.Even though our preliminary testing on collagen 1 expression appears limited; this was conducted in order to provide additional insight into functional evaluation of our preliminary findings.
Furthermore, the use of triamcinolone for treating the keloid fibroblasts has been an active area of research (29)(30)(31), as it has shown effectiveness in modulating fibroblast activity (32).In our functional testing of drug efficacy on immortalized keloid fibroblast cell line, we specifically investigated that triamcinolone treatmentinduced apoptosis affected viability and inhibited cell migration in KDIF cell lines as these are primary and core cellular characteristics identified during the fibroproliferative phenotypic development of keloids.These outcomes validate the responsiveness of KDIF cells to the drug and highlight the particular sensitivity of PKF cells to triamcinolone.This functional validation provides additional information about the representativeness of PKF as a suitable model for studying the cellular and therapeutic response of keloid fibroblasts in an in vitro research setting.
Moreover, human cell line identity and purity determination by "STR profiling" for all three KDIF cell lines established KDIF cell lines akin to standard cell lines exempted from genetic variation (genetically identical populations) and provided an unlimited cell population that overcomes the problem commonly encountered with limited supply of PKFs.In summary, these results demonstrate evident genetic, cellular, and biological alterations (hTERT genederived immortalization) in all KDIF cell lines as they represent keloid cellular behavior and characteristics, confirming their candidature as a suitable in vitro model for research into keloids.

FIGURE 1 (
FIGURE 1 (Continued) FIGURE 1 (Continued) (A) Study methodology flow diagram.KP, keloid peripheral fibroblasts; KT, keloid top fibroblasts; KM, keloid middle fibroblasts; KE, keloid extralesional skin; NF, normal skin fibroblast; DFSP, dermatofibrosarcoma protuberans; FS, fibroblastic sarcoma cell line.(B) Overall bird's-eye view of keloid tissue mass.(C) Cross-sectional view of keloid tissue mass showing four different anatomical locations in relation to the lesion: (i) top of the keloid skin includes the superficial papillary dermis (KT), (ii) center keloid tissue includes the deep reticular dermis as middle of the keloid (KM), (iii) margin as peripheral part of keloid skin (KP), and (iv) neighboring normal appearing skin as keloid extralesional skin (KE), marked on keloid tissue mass for sectioning and isolation of site-specific primary keloid fibroblasts (PKFs).hTERT gene expression analysis in primary keloid fibroblasts (KP, KT, KM, and KE) and control groups (NF, DFSP, and FS) via (D) qRT-PCR and (E) protein expression by flow cytometry, and results were presented graphically as the percentage of hTERT-FITC-positive cells and (F) Immunofluorescence staining with hTERT-FITC (green, nuclear signal) and 4′,6-Diamidino-2phenylindole dihydrochloride (DAPI) (blue, nuclear); scale bar, 20 mm.Experiments were performed in triplicate and analyzed by two-way ANOVA and significance levels set at *P < 0.05 and **P < 0.01.

2 Immortalized
FIGURE 2 Immortalized Keloid cell line validation post-transfection.(A) qRT-PCR demonstrated an increase in hTERT gene expression (fold increase) between respective primary keloid cell lines (control, blue) and post-transfected cell lines (transfected, red).(B) Immunofluorescence staining with hTERT-FITC (green, nuclear signal) and DAPI (blue, nuclear); scale bar, 20 µm.(C) b-Galactosidase-associated senescence expressed as the percentage of stained cells and presented graphically.Experiments were performed in triplicate and presented as mean ± S.D. Significance levels were set at *P < 0.05 and **P < 0.01.

3. 6
Functional validation: triamcinolone drug treatment decreased the cell viability and cell migration in KDIF and PKF cell lines Studies have shown that triamcinolone and verapamil affect cell viability and induce apoptosis in PKFs (19-21) and are used clinically FIGURE 3 Cellular characteristics of keloid-derived immortalized fibroblast cell lines.(A) Cell viability, (B) cell migration, (C) cell invasion, (D) images showing a number of invading cells (image magnification, ×10), (E) population doubling assay, and (F) cell cycle analysis via flow cytometry using propidium iodide (PI) staining.Experiments were performed in triplicate and presented as mean ± S.D.).Significance levels were set at *P < 0.05 and **P < 0.01.

FIGURE 4
FIGURE 4Immunofluorescence staining of increased collagen I protein expression in keloid cell lines compared with control.Cells were immunofluorescent labeled with anti-rabbit Alexa 488 (green, signal) marked with white arrows, and counterstained with DAPI (blue, nuclear signal) in transfected keloid fibroblasts group (PT-KP-045, PT-KT-045, and PT-KM-045) and DFSP, FS, and NF included as +ve control group.Scale bar set at 20 µm.

5
FIGURE 5 Functional validation of Keloid cell lines with drug testing.(A) Effect of verapamil and triamcinolone drug treatment (100 µM each) evaluated on cell viability of KDIF (PT-KT-045, PT-KM-045, and PT-KP-045) and percentage (%) cell viability was significantly decreased compared to vehicle group (untreated).(B) Effect of verapamil and triamcinolone drug treatment (100 µM each) evaluated on cellular apoptosis in KDIF, detected by staining with annexin V-conjugated FITC through flow cytometry.(C) Effects of verapamil and triamcinolone on cell migration represented graphically as the percentage (%) cell migration toward scratch zone at 24-h and 48-h time points.Experiments were performed in triplicate and presented as mean ± S.D.).Significance levels were set at *p < 0.05, **p < 0.01 and ***p < 0.001.