Alterations of epigenetic regulators and P53 mutations in murine mesenchymal stem cell cultures: A possible mechanism of spontaneous transformation

Abstract

BACKGROUND:

Recent studies demonstrated the involvement of mesenchymal stem/stromal cells (MSCs) in carcinogenesis, but the molecular mechanism behind this transformation is still obscured.

OBJECTIVE:

To screen both the expression levels of polycomb and trithorax epigenetic regulators and TrP53 mutations in early and late MSC culture passages in an attempt to decipher the mechanism of spontaneous transformation.

METHODS:

The study was conducted on early and late passages of MSC culture model from C57BL/6J mice. The expression profile of 84 epigenetic regulators was examined using RT2 profiler PCR array. TrP53 mutations in the DNA binding domain was screened. Codons, amino acids positions and the corresponding human variants were detected in P53 sequences.

RESULTS:

Sixty-two epigenetic regulators were dysregulated. Abnormalities were detected starting the third passage. Nine regulators were dysregulated in all passages. (C $>$ G) substitution P53 mutation was detected in passage 3 resulting in Ser152Arg substitution. Passages 6, 9, 12 and the last passage showed T $>$ C substitution resulting in Cys235Arg substitution. The last passage had T deletion and A insertion resulting in frame shift mutations changing the p.Phe286Ser and p.Asn103Lys respectively.

CONCLUSION:

In vitro expanded MSCs undergo transformation through alteration of epigenetic regulators which results in genomic instability and frequent P53 mutations.

Keywords

MSCs epigenetic regulators P53 mutation transformation C57BL/6J mice

1. Introduction

Mesenchymal stem cells (MSCs) are heterogeneous cell populations that serve as reserves for tissue regeneration in the presence of disease or injury. Although MSCs are found in various tissues, bone marrow-derived stem/stromal cells (BMSCs) have been most thoroughly investigated [1]. These cells were first isolated from bone marrow by Friedenstein in 1967 and have been described as plastic adherent cells with a fibroblast-like morphology [2]. In 2006, the International Society for Cellular Therapy (ISCT) established criteria for identifying unique populations of MSCs. They defined MSCs according to the following criteria: 1 – MSCs must be purified from the BM stromal population based on plastic adherence under standard culture conditions. 2 – MSCs must be positive for CD105, CD90, and CD73, express low levels of MHC-I, and be negative for MHC-II, CD11b, CD14, CD34, CD45, and CD31. 3 – MSCs must differentiate in vitro into osteocytes, chondrocytes, and adipocytes [3].

The cell-of-origin concept refers to the normal cell type that acquires the first cancer-promoting mutation and initiates tumor formation [4]. During the last years there has been overwhelming evidence that MSC is the cell-of-origin for many types of sarcomas [5, 6]. As these cells either transform directly into cancer stem cells or settle in the tumor stroma to facilitate the process of malignant transformation and increase the ability of cancer cells for aggressiveness. MSCs have been shown to induce Ewing sarcoma [7], osteosarcoma [8], chondrosarcoma [9], synovial sarcoma [10] and gastric cancer [11]. The transforming properties of MSCs and their possible role as sarcoma progenitors make these cells useful as a model for studying sarcomagenesis and progression.

Spontaneous transformation of both murine and human MSCs in long-term cultures, have been reported in many studies [12, 13, 14]. Houghton et al. showed that bone marrow derived MSCs home to the gastric mucosa during Helicobacter infection and differentiate abnormally as dysplastic cells, can initiate, contribute to and become gastric adenocarcinoma. The time can be dramatically decreased by introducing MSCs which have been repeatedly passaged in culture and allowed to spontaneously transform [11].

Although MSC involvement in the carcinogenesis has been investigated and verified that these cells are the origin of cancer stem cells, we have not yet discovered an effective treatment that selectively target cancer stem cells. This is because of our failure to determine the mechanism of MSC transformation and the lack of knowledge of the factors that control this mechanism.

Epigenetic changes are heritable changes in gene expression without alteration of the coding sequence of the gene. Epigenetic mechanisms include, DNA methylation, histone modification and noncoding RNA-mediated processes. In addition to the common hallmarks of cancer which includes sustaining proliferative signaling, evading growth suppressors, resisting cell death, enabling replicative immortality, inducing angiogenesis, and activating invasion and metastasis, these epigenetic changes have recently emerged as common hallmarks of many cancers [15].

Epigenetic regulators are a group of elements that regulate gene expression along the genome. They act antagonistically to either promote (Trithorax group TrxG) or repress (Polycomb group PcG) transcription through regulation of specific amino acid modifications in histones [16]. Dysregulation of these elements may lead to genome wide changes in gene expression which in turn, lead to cancer development [17]. Recently, research has focused on the relationship between the abnormalities in these regulators and many types of cancers. According to the cancer stem cell model, alterations in epigenetic mechanisms of stem cells lead to malignant transformation of these cells [17]. These early epigenetic events in stem cells are also the earliest events in cancer initiation. Epigenetic disruption in stem cells may give rise to a high-risk aberrant progenitor cell population that can undergo transformation upon gain of subsequent genetic mutations [17]. Until now, we have not been able to identify the factors that lead to these epigenetic alterations.

We followed up the expression profile of 84 epigenetic regulators and P53 mutations in early and late passages of bone marrow derived MSC model from C57BL/6J mice, in an attempt to decipher the mechanism of in vitro transformation of MSCs.

2. Materials and methods

2.1 Expansion of cryopreserved mouse MSC passages

We used already established MSCs and progressively passaged for more than a year. The cells were originally obtained from the bone marrow of male C57BL/6J mice. The cells were first established on July, 2004 [11] and have been well characterized on March 2006 [11], where surface marker expression have been characterized by fluorescent activated cell sorting (FACS), and lineage specific marker assessment was determined by immunohistochemistry or reverse transcription polymerase chain reaction (RT-PCR). At each passage, cultures were split, and one plate was held in culture at confluence for detection of foci, one plate was split into the next passage culture, and the remainder of the plates was frozen [18].

Cryopreserved C57BL/6J mouse MSCs from passages 3, 6, 9, 12, 28, A2, B2 and the last passage were removed from the liquid nitrogen. The cells were thawed and 1x Pbs (pH 7.2) was added then centrifuged at 1200 rpm for 10 minutes. The supernatant was removed, and the cells were re-suspended in1x phosphate buffer saline (pH 7.2) and centrifuged again for 10 minutes at 1200 rpm. The supernatant was removed, and the cells were re-suspended in 5 ml Dulbecco’s Modified Eagle Medium (DMEM) containing 10% fetal calf serum (FCS), L-Glutamine and penicillin streptomycin. The cells were left in CO2 incubator at 37 ${{}^{\circ}}$ C and then were subjected to RNA extraction.

2.2 Isolation and culture of control bone marrow MSCs from C57BL/6J mice

Control MSCs used for gene expression and P53 sequence analyses were isolated from the bone marrow of male C57BL/6J mice. The mice used were the same species sex and age from which MSCs used in the study were established. Bone marrow isolation from mice was performed in accordance with relevant guidelines and regulations of the University of Massachusetts Medical School. C57BL/6J mice were purchased from Jackson Laboratories (Bar Harbor, ME, USA) and housed in micro isolator cages in an SPF facility, given free access to chow and water.

At 8 weeks of age, male C57BL/6J mice were euthanized by CO ${}_{2}$ inhalation followed by cervical dislocation. Femur and tibia were removed, and marrow flushed from the marrow cavity. For isolation of MSCs, total marrow was placed in culture in different plates; adherent cells were collected at 4 days and expanded in DMEM supplemented with 2 mmol/L glutamine, 100 units/mL penicillin, 0.1 mg/mL streptomycin, and 10% fetal bovine serum (FBS) at 37 ${}^{\circ}$ C in 5% CO ${}_{2}$ until the cells reached confluent. Then the cells were collected and subjected to RNA extraction.

2.3 RNA extraction

Culture media from both control and MSc passages was removed and the cells were washed with 1x phosphate buffer saline (PH7.2). RNA from each passage was isolated using RNeasy plus kit from Qiagen. (Cat no. 74134). The integrity of the RNA samples was determined by 2% denaturing agarose gel electrophoresis. RNA was quantified and subjected to reverse transcription to cDNA.

2.4 cDNA synthesis

1121.3 ng of RNA per sample were used for cDNA synthesis. cDNA was synthesized using the RT2 First Strand Kit Qiagen (Cat no. 330401), 5 $\mu$ l of genomic DNA elimination buffer (GE) was added to the RNA and completed to 10 $\mu$ l. The reagents were mixed well and incubated at 42 ${{}^{\circ}}$ C for 5 minutes then chilled on ice for at least 1 minute. The RT cocktail was prepared by adding 4 $\mu$ l RT buffer (BC3), 1 $\mu$ l primer and external control mix (P2), 2 $\mu$ l RT enzyme mix (RE3) then completed to 10 $\mu$ l by adding 3 $\mu$ l of H2O. 10 $\mu$ l of RT cocktail were added to10 $\mu$ l genomic DNA elimination mixture. The cocktail was mixed well and incubated at 42 ${{}^{\circ}}$ C for 15 minutes. The reaction was stopped by heating at 95 ${{}^{\circ}}$ C for 5 minutes. 91 $\mu$ l of H2O was added to the synthesis reaction.

2.5 Real-time reverse transcription polymerase chain reaction for mouse polycomb and trithorax complexes

RT2 SYBR Green ROX qPCR Mastermix (Cat. no. 330520) was briefly centrifuged. In a clean tube 1350 $\mu$ l master mix was mixed to 102 $\mu$ l cDNA synthesis reaction. The mixture was completed to 2700 $\mu$ l by RNAse free water. Twenty five $\mu$ l of the PCR components mix were added to each well of the RT2 Profiler PCR Array (Cat. no. PAMM-506) then sealed with optical adhesive film. The array was centrifuged at 1000 g for 1 minute to remove air bubbles. PCR works was done using the Eppendorf Mastercycler ep realplex. The machine was set up according to the instruction manual. The PCR machine was programmed as follows: One cycle of heat activation at 95 ${{}^{\circ}}$ C for 10 minutes, followed by 40 PCR amplification cycles as follows: 95 ${{}^{\circ}}$ C for 15 seconds and 60 ${}^{\circ}$ C for 1 minutes. Finally, the melting curve was set for 1 cycle as follows: 95 ${{}^{\circ}}$ C for 15 seconds, 60 ${{}^{\circ}}$ C for 15 seconds, hold 20 minutes and finally 95 ${{}^{\circ}}$ C for 15 seconds.

After ending the PCR program, CT (threshold cycles) values were collected and the fold change for each gene was calculated using 2 ${}^{-\rm\Delta\Delta CT}$ method on the PCR array data analysis software. (http://www.sabiosciences.com/ pcrarraydataanalysis.php).

2.6 P53 amplification

PCR amplification for the DNA binding domain of Trp53 was studied on the RNA samples using one step RT-PCR kits Qiagen (cat. no. 210210). The following primers were used Trp53F(96–115), 5 ${}^{\prime}$ -AGATATCCTGCCATCACCTCACTG-3 ${}^{\prime}$ , and Trp53R (918–900) 5 ${}^{\prime}$ -GGGCAGCGCTCTCTTTGC-3 [18]. The PCR cycles were as follows, 1 cycle at 50 ${{}^{\circ}}$ C for 30 min. (reverse transcription), 1 cycle at 95 ${}^{\circ}$ C for 15 min. (enzyme activation), followed by 35 cycles of amplification as follows, denaturation at 94 ${{}^{\circ}}$ C for 30 seconds, annealing at 54 ${{}^{\circ}}$ C for 1 minute, extension at 72 ${{}^{\circ}}$ C for 1 minute and finally one cycle at 72 ${{}^{\circ}}$ C for 10 minutes. The PCR product (823 bp) was electrophoresed on 2% agarose and the bands were gel purified using gel purification kit Qiagen (cat. no. 28704) and sent to the sequence core facility of the University of Massachusetts Medical School.

Table 1a
The fold regulation of the 35 dysregulated epigenetic regulators in passage 3 (group1)

Gene symbol	Fold regulation
Arid1b	4.80
Ash1l	2.19
Ash2l	2.64
Asxl1	2.93
Asxl2	2.19
Bap1	2.40
Cbx2	2.66
Cbx3	4.18
Cbx5	9.46
Cbx8	2.10
Cxxc1	2.32
Dnmt3b	2.59
Epc1	2.15
Ezh2	4.23
Ino80	2.25
Ino80d	2.52
Kdm2b	2.09
Kdm5d	3.04
L3mbtl2	2.72
Larp7	2.41
Kmt2d	2.77
Kmt2c	2.64
Mov10	2.21
Pcgf2	3.25
Pcgf5	3.98
Phc1	4.01
Phf19	3.04
Rbbp5	3.19
Ring1	3.14
Rnf2	2.52
Smarca1	3.61
Smarcc1	2.95
Snai1	6.46
Usp11	3.61
Rnasel	$-$ 9.57

Gene symbols and fold regulation of the 35 dysregulated epigenetic regulators in passage 3 (group1). Thirty Four genes up and one downregulated.

Table 1b

The fold regulation of the 34 dysregulated epigenetic regulators in passage 6 (group2)

Gene symbol	Fold regulation
Arid1b	3.57
Ash1l	2.29
Asxl1	2.40
Bap1	2.26
Cbx2	2.34
Cbx3	2.98
Cbx5	11.52
Cbx8	2.39
Ezh2	2.11
Htt	2.12
Ino80	3.88
L3mbtl2	2.56
Mov10	2.71
Pcgf2	2.32
Pcgf5	2.47
Phc1	3.09
Phf19	10.03
Ppp1r8	3.29
Rbbp5	2.17
Rbp2	10.38
Smarca1	2.23
Smarca4	2.14
Smarcb1	2.17
Smarcc1	3.62
Snai1	3.93
Usp11	4.78
Aebp2	$-$ 2.08
Cbx7	$-$ 36.61
Dnmt3a	$-$ 2.16
Dnmt3l	$-$ 2.35
E2f6	$-$ 2.34
Kdm5d	$-$ 294.88
Rnasel	$-$ 12.33
Zbtb16	$-$ 11.50

Gene symbols and fold regulation of the 34 dysregulated epigenetic regulators in passage 6 (group2). Twenty six genes up and 8 downregulated.

Table 1c

The fold regulation of the 38 dysregulated epigenetic regulators in passage 9 (group3)

Gene symbol	Fold regulation
Arid1b	3.79
Ash1l	2.59
Asxl1	2.57
Cbx1	2.06
Cbx2	3.19
Cbx3	3.19
Cbx5	16.59
Cbx8	2.30
Dnmt3b	3.08
Ezh2	4.32
Htt	2.32
Ino80	3.21
Ino80d	2.19
Kdm2b	2.32
L3mbtl2	3.39
Larp7	2.62
Mov10	2.99
Pcgf2	2.97
Pcgf5	2.81
Phf19	15.69
Ppp1r8	5.07
Rbbp4	2.83
Rbbp5	2.72
Rbbp7	2.66
Rbp2	3.23
Smarca4	2.09
Smarcb1	2.27
Smarcc1	4.38
Snai1	4.35
Usp11	4.35
Cbx4	$-$ 2.77
Cbx7	$-$ 45.51
Dnmt3a	$-$ 2.62
Dnmt3l	$-$ 7.88
Ezh1	$-$ 2.56
Kdm5d	$-$ 221.01
Rnasel	$-$ 53.74
Zbtb16	$-$ 8.62

Gene symbols and fold regulation of the 38 dysregulated epigenetic regulators in passage 9 (group3). Thirty genes up and 8 downregulated.

Table 1d

The fold regulation of the 22 dysregulated epigenetic regulators in passage 28 (group4)

Gene symbol	Fold regulation
Arid1b	2.36
Cbx1	2.54
Cbx3	2.61
Cbx5	7.29
Ezh2	2.34
Mov10	2.88
Pcgf2	4.02
Pcgf5	2.09
Phc1	2.24
Ppp1r8	2.11
Smarca1	5.72
Smarcc1	2.37
Usp11	4.68
Cbx7	$-$ 3.85
Dnmt1	$-$ 2.45
Dnmt3a	$-$ 9.74
Dnmt3l	$-$ 3.59
Hltf	$-$ 3.64
Rnasel	$-$ 8.13
Smarcc2	$-$ 2.03
Zbtb16	$-$ 10.44

Gene symbols and fold regulation of the 22 dysregulated epigenetic regulators in passage 28 (group 4). Fourteen genes up and 8 downregulated.

Table 1e

The fold regulation of the 42 dysregulated epigenetic regulators in the last passage (group5)

Gene symbol	Fold regulation
Ash2l	3.90
Bap1	3.21
Cbx1	2.93
Cbx2	2.62
Cbx3	3.95
Cbx5	8.01
Cbx7	3.54
Cbx8	2.54
Cxxc1	2.06
Dnmt1	4.54
Dnmt3b	2.93
Dnmt3l	10.72
Eed	2.81
Ezh2	6.64
Ino80	2.47
Kdm2b	2.52
L3mbtl2	4.80
Larp7	4.09
Kmt2d	2.28
Mov10	7.58
Mtf2	6.07
Pcgf2	2.81
Pcgf5	2.37
Phc1	2.33
Phf19	6.97
Ppp1r8	4.32
Rbbp5	2.03
Rbp2	911.44
Sirt1	2.27
Smarca4	2.12
Smarca5	3.01
Smarcc1	3.74
Suz12	2.72
Trim27	2.61
Wdr5	2.03
Cbx4	$-$ 9.63
Dnmt3a	$-$ 4.22
Kdm5d	$-$ 122.62
Rnasel	$-$ 7.77
Scml2	$-$ 4.99
Smarca1	$-$ 25.78
Zbtb16	$-$ 9.84

Gene symbols and fold regulation of the 42 dysregulated epigenetic regulators in the last passage (group5). Thirty five genes up and 7 downregulated.

2.7 P53 sequence analysis

The sequences chromatograms were viewed by FinchTV program. TrP53 sequence results were analyzed using mouse blast search (http://genome.ucsc.edu/ cgi-bin/hgBlat?command=start). Wildtype and mutant TrP53 DNA codons, amino acids position and symbols were retrieved from Benchling (Biology Software).

3. Results

3.1 Most of polycomb and trithorax complexes showed abnormal expression in early and late MSC passages

We studied the fold change of a total of 84 epigenetic regulators including polycomb and trithorax elements in the control and 5 passages of MSCs, these passages are 3, 6, 9, 28 and the last passage (is a late MSC culture passage which has been in vitro cultured for more than a year, it was the last MSC passage number has been passaged during the time of the study). The control cells were MSCs cultured from the bone marrow of 8 week old male C57BL/6J mice. When the cells reached confluent, they were subjected to RNA extraction and analysis. We considered the genes are abnormally regulated if they have at least 2 fold changes (2 folds up or 2 folds down regulation from the control). Out of 84 genes, 62 were abnormally expressed (supplementary materials Fig. S1). We detected these abnormalities in all passages prior beginning from the 3rd passage. In passage 3, there were 34 genes up, and 1 downregulated (Table 1a, Fig. 1a). In passage 6, there were 26 up, and 8 downregulated genes (Table 1b, Fig. 1b). In passage 9, there were 30 up, and 8 downregulated genes (Table 1c, Fig. 1c). In passage 28, there were 14 up, and 8 downregulated genes (Table 1d, Fig. 1d). However, the last passage showed 35 up, and 7 downregulated genes (Table 1e, Fig. 1e).

Among the PcG and TrX elements, we found 9 regulators are abnormally regulated throughout the early and late passages and show significant dysregulation. These regulators are Lysine (K)-specific demethylase 5D (kdm5d), Retinol binding protein 2, cellular (Rbp2), Ribonuclease L (2 ${}^{\prime}$ , 5 ${}^{\prime}$ -oligoisoadenylate synthetase-dependent) (RnaseL), Zinc finger and BTB domain containing 16 (Zbtb16), Chromobox homolog 5 (cbx5), Chromobox homolog 7 (cbx7), DNA (cytosine-5-)-methyltransferase 3-like (Dnmt3I), PHD finger protein 19 (phf19) and SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily a, member 1 (smarca1). Dnmt3l and cbx7 were downregulated in passages 6, 9, 28 then upregulated in the last passage, smarca1 was upregulated in passages 3, 6, 28 and downregulated in the last passage, kdm5d was upregulated in the 3rd passage and downregulated in the remaining passages. However, Rbp2, cbx5, phf19 were upregulated in all passages and RnaseL and Zbtb16 were downregulated in all passages (Table 2, Fig. 2a and b).

Table 2
The fold regulation of the 9 abnormally expressed genes in all MSC passages

Gene	Passage3	Passage6	Passage9	Passage28	Last passage
Rbp2	$-$ 1.34	10.38	3.23	1.08	911.44
kdm5d	3.04	$-$ 294.88	$-$ 221.01	$-$ 1.83	$-$ 122.62
RnaseL	$-$ 9.57	$-$ 12.33	$-$ 53.74	$-$ 8.13	$-$ 7.77
Zbtb16	1.21	$-$ 11.50	$-$ 8.62	$-$ 10.44	$-$ 9.84
cbx5	9.46	11.52	16.59	7.29	8.01
cbx7	$-$ 1.31	$-$ 36.61	$-$ 45.51	$-$ 3.85	3.54
Dnmt3I	$-$ 1.08	$-$ 2.35	$-$ 7.88	$-$ 3.59	10.72
phf19	3.04	10.03	15.69	$-$ 1.45	6.97
smarca1	3.61	2.23	$-$ 1.78	5.72	$-$ 25.78

The fold regulation of RbP2, Kdm5d, RNASEL, Zbtb16, Cbx5, Cbx7, DNMT3L, Phf19 and smarc1a, in the 5 early and late MSC passages.

3.2 Demethylases were upregulated while methylases were downregulated in the early passages

Kdm5d was upregulated in passage 3 and downregulated in passages 6, 9, 28 and the last passage, while Rbp2 was upregulated in passages 6, 9 and the last passage. However, Chromobox Cbx7 was downregulated in passages 3, 6, 9, 28 and upregulated in the last passages while, dnmt3l was downregulated in passages 3, 6, 9, 28 and upregulated in the last passage (Table 2, Fig. 2a and b).

3.3 Substitution mutations were detected in the early and late while deletion and insertion mutations were appeared in the late MSC culture passages

We sequenced the DNA binding domain of the TrP53 gene in a total of 8 passages representing the early and the late passages. We sequenced passages 3, 6, 9 and 12 which represent the early passages and passages 28, A2, B2 (A2 and B2 are late transformed MSC lines isolated after 4 months from sub cloning of the original cell line) and the last passage which represent the late passages.

We did not find any mutation in the control passage. For passage 3, we detected (C $>$ G) substitution mutation (TrP53:c.456C $>$ G) (Fig. 3a) resulting in Ser152Arg substitution (supplementary Figure S2a) which corresponds in human to the synonymous variant (TP53):c.465C $>$ G (p. Thr155=).

Figure 1.

The scatter plot compares the normalized expression of every gene of the 84 epigenetic regulators between each passage and the control passage by plotting them against one another to visualize large gene expression changes. The central line indicates unchanged gene expression. The dotted lines indicate the selected fold regulation threshold. Data points beyond the dotted lines in the upper left and lower right sections meet the selected fold regulation threshold. (a) Scatter plot shows 35 dysregulated epigenetic regulators in passage 3 (group1), 34 abnormally up and 1 abnormally downregulated genes. (b) Scatter plot shows 34 dysregulated epigenetic regulators in passage 6 (group2), 26 abnormally up and 8 abnormally downregulated genes. (c) Scatter plot shows 38 dysregulated epigenetic regulators in passage 9 (group3), 30 abnormally up and 8 abnormally downregulated genes. (d) Scatter plot shows 22 dysregulated epigenetic regulators in passage 28 (group4), 14 abnormally up and 8 abnormally downregulated genes. (e) Scatter plot shows 42 dysregulated epigenetic regulators in the last passage (group 5), 35 abnormally up and 7 abnormally downregulated genes.

Figure 2.

Represents the fold changes of the 9 dysregulated genes in all the 5 early and late passages of the BM derived MSC culture model. (a) Following up the fold change of Kdm5d, RNASEL, Zbtb16, Cbx5, Cbx7, DNMT3L, Phf19 and smarc1a in the 5 early and late MSC passages. (b) Following up the fold change of Rbp2 gene in the 5 early and late MSC passages.

Figure 3.

Electropherogram of part of the TrP53 gene sequence for Passages 3, 6, 9, 12 and the last passage of the BM derived MSC culture model from C57BL/6 mice. (a) A is showing normal sequence. B is showing heterozygous c.456C $>$ G (p. Ser152Arg) mutation in passage 3. (b) A is showing normal sequence. B is showing homozygous c.703T $>$ C (p. Cys235Arg) mutation in passages 6, 9, 12 and the last passage. (c) A is showing normal sequence (TTC encodes for Phenylalanine in position 286). B is showing homozygous deletion of T (c.857delT) results in frame shift mutation changing Phenylalanine into Serine (TCC) i.e. (p.F286fs*56) in the last passage. (d) A is showing normal sequence (AAC encodes for Asparagine in position 103). B is showing homozygous insertion of A (TrP53:c.308-309insA) results in frame shift mutation changing asparagine into lysine (AAA) i.e. (p.N103fs*27) in the last passage.

For passages 6, 9, 12 and the last passage, we detected T $>$ C substitution mutation (TrP53:c.703T $>$ C) (Fig. 3b) resulting in Cys235Arg substitution (supplementary Figure S2b) which corresponds in human to the missense variant (TP53):c.712T $>$ C (p. Cys238Arg).

The last passage showed the presence of T deletion (TrP53):c.857delT) (Fig. 3c) which results in frame shift mutation changing the protein phenylalanine at position 286 into Serine (p.F286fs*56) (supplementary Fig. S2c) which corresponds in human to the frame shift mutation (TP53):c.866delT (p.L289fs*56).

The last passage also showed the presence of A insertion (TrP53: c.308-309insA) or (c.308dupA) (Fig. 3d) resulting in frame shift mutation changing the protein Asparagine at position 103 into Lysine (p.N103fs*27) (supplementary Fig. S2d) corresponding in human to the frameshift mutation (TP53):c.315insA (p.S106f s*43).

Passage 28 had G and T insertion, passage A2 had G insertion, Passage B2 had G insertion and A deletion (Data not shown).

4. Discussion

Several studies have demonstrated the involvement of MSCs in the carcinogenesis process. These studies have made several assumptions about the process of MSC transformation. Despite these assumptions, so far we could not decipher the mysteries of this process. The results of our study can contribute to the interpretation of how malignant transformation of MSCs. The present study has proven that there is an abnormality in the gene expression of most epigenetic regulators in MSC cultures, with different mutations in the P53 gene confirming the transformation of these cells. Recent studies have demonstrated a relationship between the disruption of epigenetic processes and carcinogenesis [17].

Although we have demonstrated an abnormality in the gene expression of 62 epigenetic regulators, 9 are dysregulated in all MSC passages. For the remaining 53 regulators, each was dysregulated in at least one passage. These 9 regulators were found to regulate the expression of many genes that are cornerstones in cancer pathways including Wnt receptor signaling pathway [19], EMT [20], IFN [21], P53 pathway [22], PTEN signaling pathway [23] and AKT-NF- $\kappa$ B-miR-21 pathway [24].

The presence of abnormalities in the gene expression of these 9 regulators in all passages indicates its important role in MSC oncogenesis and their possible role as drivers in this transformation. Other studies are clearly needed to study the functional characteristics and the therapeutic role of these abnormally expressed epigenetic regulators. This occurs by detecting the impact of targeting these factors on tumor formation in animal models for cancer.

We noticed that the demethylases were active and the methylases were inactive in the early passages, these findings may indicate the importance of demethylation in the early stages of MSC transformation. Many studies documented that demethylation is an early event in carcinogenesis [25, 26]. Also one of the functions of the DNA methyltransferase Dnmt3L is facilitating the methylation of cytosine residues, this is in addition to its ability to stimulate the catalytic activity of DNMT3 A and DNMT3B as well as its ability to bind to chromatin leading to anchoring the DNA methylation machinery onto chromatin DNA templates [27]. This is an important issue in the DNA proofreading mechanism where DNA polymerase can accurately differentiate between the old and the newly synthesized DNA strand through cytosine methylation of the old DNA strand. The downregulation of DNMT3L in the early passages of MSCs may induce a defect in cytosine methylation of these cells, so the DNA polymerase cannot accurately differentiate between the original and the newly synthesized DNA strand. This may lead to defect in the DNA proofreading mechanism and persistence of mutation. This may explain presence of spontaneous P53 mutation in the early and late MSC cultures. Moreover, presence of mutation due to a defect in the DNA proofreading mechanism shows that the mutation likely occurs randomly along the DNA, not only in the P53 gene.

This scenario seems logical, especially because the majority of the 9 dysregulated epigenetic regulators are responsible for methylation and demethylation. One of these 9 regulators is zbtb16 (PLZF), an important regulator induces methylation through activating DNMT1 [28]. This epigenetic regulator is down regulated in all MSC passages. Studies indicated that dysregulated zbtb16 leading to hypo-methylation resulting into activation of zbtb16 target genes including c-myc, c-kit and Crabp1 [28]. Interestingly, both zbtb16 and DNMT3L have a common target; it is the retrotransposon L1 where DNA methylation activity of these 2 regulators induces full silencing of L1 gene. It was found that cell stress inhibit the methylation activity of zbtb16 through translocation from the nucleus to the cytoplasm [29]. So, cell stress resulting from the in vitro expansion of MSCs may be responsible for inactivation of zbtb16 in our MSC culture model.

Presence of P53 mutations in the early and late passages of our MSCs culture model confirms the transformation of these cells. The detected mutations were found in different human tumors [30, 31]. Li et al. studied the P53 mutation in the DNA binding domain of these cells and found 5 point mutations in the late-passages, 3 of which were pathogenic [18]. In the present study, we followed up the progression of P53 mutation in the early and late passages and we found 4 types of mutations. In passage 3, the mutation was non-pathogenic in human and did not appear in the other passages. The other 3 types of mutations were pathogenic in human and one of which was frequent in all tested passages starting from passage 6. Previous studies found that more than half of human cancers carry loss of function mutations of P53 [32] and 95% of these mutations were detectable within the genomic region encoding the DNA-binding domain [33].

Interestingly, many studies found a correlation between P53 mutation and abnormal expression of cbx7 [34], DNMT3L [35], RNASEL [36], Zbtb16 [37] and smarca1 [38]. These elements are abnormally expressed in different types of malignancies and are associated with P53 inhibition.

MSCs used in our study were examined for tumor formation by injection in a C57BL/6J mice model for gastric cancer where tumor was formed after 18 months and by examining the resulting tumor, the injected MSCs were detected (11). Moreover, the authors found that they can decrease the time for tumor formation by introducing MSCs which have been repeatedly passaged in culture for more than a year. The results of the in vivo experiments clearly support our findings assuming the in vitro transformation of MSCs.

The present findings may suggest a model for in vitro transformation of MSCs where the stress on MSCs due to repeated replication in culture, may induce the epigenetic regulators for abnormal expression. The upregulation of demethylases and downregulation of methylases in the early passages may affect the DNA proofreading mechanism resulting into genomic instability and accordingly frequent P53 mutation. Persistence of P53 mutation in parallel to the abnormal expression of epigenetic regulators may guarantee the transformation of MSCs. By time, MSCs accumulate mutations enough to induce a malignant transformation.

Although we did this study on mice MSCs and this may explain the large number of mutations, but we must be aware that we are studying a group of factors that regulates the expression of genes. These epigenetic regulators have a common function in both human and mice. Abnormality in any of these regulators may lead to genomic instability resulting in malignant transformation. Regardless the number of mutations, abnormality in any of these regulators will have an adverse effect on gene expression and accordingly disturbance in many processes and pathways in both mice and human cells. Regarding the P53 mutations in our study, we found the same types of mutations in different human cancers. Therefore, we must take the utmost care and ensure the safety of MSCs before applying these cells in the treatment, especially if the cells used were passaged in culture.

5. Conclusions

From the present study, we may conclude that BM derived MSCs undergo transformation in culture toward malignancy. The cells are characterized by abnormal expression of 62 epigenetic regulators. Nine epigenetic regulators are dysregulated in all passages. Upregulation of demethylases and downregulation of methylases is characteristic in the early passages. P53 mutation is prominent in all passages. A single base substitution mutation is frequent in the early and late passages, while deletion and insertion mutations are detected in the late passages. Alterations of the epigenetic regulators in the early and late passages of BM derived mouse MSCs result in genomic instability and frequent P53 mutation and, accordingly, malignant transformation. Targeting the abnormally expressed epigenetic regulators in MSC cultures and studying the impact of targeting on in vivo tumor formation is clearly needed to examine the therapeutic role of these regulators.

Funding

Dr. Ayman Metwally received a 6-month post-doctoral fellowship from the Science and Technology Development Fund (STDF) Egypt, under the US-Egypt Junior Scientist Development Visit Grants to Dr. Houghton’s Laboratory at the University of Massachusetts Medical School, USA. The funding agency supported the travel and salary for Dr. Metwally during the fellowship period.

Author contributions

Conception: Ayman Metwally and JeanMarie Houghton

Interpretation or analysis of data: Ayman Metwally

Preparation of the manuscript: Ayman Metwally

Revision for important intellectual content: JeanMarie Houghton and Hanchen Li

Supervision: JeanMarie Houghton

Methodology: Ayman Metwally and Hanchen Li

Validation: Ayman Metwally and JeanMarie Houghton

Supplementary data

The supplementary files are available to download from https://dx-doi-org.web.bisu.edu.cn/10.3233/CBM-203121.

sj-docx-1-cbm-10.3233_CBM-203121.docx - Supplemental material

Supplemental material, sj-docx-1-cbm-10.3233_CBM-203121.docx

Footnotes

Acknowledgments

We thank Dr. Eman Bayomi, Department of Medical Molecular Genetic, National Research Center, Egypt for revising and editing the p53 sequence results.

References

G.I.

, Bone marrow-derived stem/stromal cells and adipose tissue-derived stem/stromal cells: Their comparative efficacies and synergistic effects, J Biomed Mater Res A 105(9) (2017), 2640–2648. doi: 10.1002/jbm.a.36089.

Kozlowska

Krawczenko

Futoma

Jurek

Rorat

Patrzalek

and Klimczak

, Similarities and differences between mesenchymal stem/progenitor cells derived from various human tissues, World J Stem Cells 11(6) (2019), 347–374. doi: 10.4252/wjsc.v11.i6.347.

Dominici

Le Blanc

Mueller

Slaper-Cortenbach

Marini

Krause

Deans

Keating

Prockop

and Horwitz

, Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement, Cytotherapy 8(4) (2006), 315–317. doi: 10.1080/14653240600855905.

Visvader

J.E.

, Cells of origin in cancer, Nature 469(7330) (2011), 314–322. doi: 10.1038/nature09781.

Rodriguez

Rubio

and Menendez

, Modeling sarcomagenesis using multipotent mesenchymal stem cells, Cell Research 22(1) (2012), 62–77. doi: 10.1038/cr.2011.157.

Mutsaers

A.J.

and Walkley

C.R.

, Cells of origin in osteosarcoma: Mesenchymal stem cells or osteoblast committed cells? Bone 62 (2012), 56–63. doi: 10.1016/j.bone.2014.02.003.

Riggi

Cironi

Provero

Suva

M.L.

Kaloulis

and Garcia-Echeverria

, Development of Ewing’s sarcoma from primary bone marrow-derived mesenchymal progenitor cells, Cancer Res (65) (2005), 11459–11468. doi: 10.1158/0008-5472.CAN-05-1696.

Mohseny

A.B.

Szuhai

Romeo

Buddingh

E.P.

Briaire-De

B.I.

and De

J.D.

, Osteosarcoma originates from mesenchymal stem cells in consequence of aneuploidization and genomic loss of Cdkn2, J Pathol 219 (2009), 294–305. doi: 10.1002/path.2603.

Boeuf

Kunz

Hennig

Lehner

Hogendoorn

and Bovee

J.A.

, Chondrogenic gene expression signature in mesenchymal stem cells is a clasXsifier of conventional central chondrosarcoma, J Pathol 216 (2008), 158–166. doi: 10.1002/path.2389.

10.

Naka

Takenaka

Araki

Miwa

Hashimoto

and Yoshioka

, Synovial sarcoma is a stem cell malignancy, Stem Cells 28 (2010), 1119–1131. doi: 10.1002/stem.452.

11.

Houghton

Stoicov

Nomura

Rogers

A.B.

Carlson

Cai

Fox

J.G.

Goldenring

J.R.

and Wang

T.C.

, Gastric cancer originating from bone marrow-derived cells, Science 306(5701) (2004), 1568–1571. doi: 10.1126/science.1099513.

12.

Miura

Padilla-Nash

H.M.

Molinolo

A.A.

Patel

Seo

Sonoyama

Zheng

J.J.

Baker

C.C.

Chen

Ried

and Shi

, Accumulated chromosomal instability in murine bone marrow mesenchymal stem cells leads to malignant transformation, Stem Cells 24 (2006), 1095–1103. doi: 10.1634/stemcells.2005-0403.

13.

Rubio

Garcia-Castro

Martin

M.C.

de la Fuente

Cigudosa

J.C.

Lloyd

A.C.

and Bernad

, Spontaneous human adult stem cell transformation, Cancer Res 65 (2005), 3035–3039. doi: 10.1158/0008-5472.CAN-04-4194.

14.

Meza-Zepeda

L.A.

Noer

Dahl

J.A.

Micci

Myklebost

and Collas

, High-resolution analysis of genetic stability of human adipose tissue stem cells cultured to senescence, J Cell Mol Med 12 (2008), 553–563. doi: 10.1111/j.1582-4934.2007.00146.x.

15.

Hanahan

and Weinberg

R.A.

, The hallmarks of cancer, Cell 100 (2000), 57–70. doi: 10.1016/s0092-8674(00)81683-9.

16.

Ringrose

and Paro

, Epigenetic regulation of cellular memory by the Polycomb and Trithorax group proteins, Annu Rev Genet 38 (2004), 413–443. doi: 10.1146/annurev.genet.38.072902.091907.

17.

Sharma

Kelly

T.K.

and Jones

P.A.

, Epigenetics in cancer, Carcinogenesis 31(1) (2010), 27–36. doi: 10.1093/carcin/bgp220.

18.

Fan

Kovi

R.C.

Moquin

Konz

Stoicov

Kurt-Jones

Grossman

S.R.

Lyle

Rogers

A.B.

Montrose

and Houghton

, Spontaneous expression of embryonic factors and p53 point mutations in aged mesenchymal stem cells: A model of age-related tumorigenesis in mice, Cancer Res 67(22) (2007), 10889–10898. doi: 10.1158/0008-5472.CAN-07-2665.

19.

Teng

Y.C.

Lee

C.F.

Y.S.

Chen

Y.R.

Hsiao

P.W.

Chan

M.Y.

Lin

F.M.

Huang

H.D.

Chen

Y.T.

Jeng

Y.M.

Hsu

C.H.

Yan

Tsai

M.D.

and Juan

L.J.

, Histone demethylase RBP2 promotes lung tumorigenesis and cancer metastasis, Cancer Res 73(15) (2013), 4711–4721. doi: 10.1158/0008-5472.CAN-12-3165.

20.

Zhou

Kannappan

Chen

Leng

Zhang

and Xuan

, RBP2 induces stem-like cancer cells by promoting EMT and is a prognostic marker for renal cell carcinoma, Exp Mol Med 48(6) (2016), e238. doi: 10.1038/emm.2016.37.

21.

Malathi

Paranjape

J.M.

Bulanova

Shim

, J.M. Guenther-Johnson and Faber

P.W.

, A transcriptional signaling pathway in the IFN system mediated by 2′-5′-oligoadenylate activation of RNase L, Proc Natl Acad Sci USA 102(41) (2005), 14533–14538. doi: 10.1073/pnas.0507551102.

22.

Choi

W.I.

Kim

M.Y.

Jeon

B.N.

Koh

D.I.

Yun

C.O.

Lee

C.E.

Kim

and Hur

M.W.

, Role of promyelocytic leukemia zinc finger (PLZF) in cell proliferation and cyclin-dependent kinase inhibitor 1A (p21WAF/CDKN1A) gene repression, J Biol Chem 289(27) (2014), 18625–18640. doi: 10.1074/jbc.M113.538751.

23.

Cao

Zhu

Zhou

Liu

Xuan

Yan

Zheng

Zhou

Chen

Huang

and Feng

, PLZF mediates the PTEN/AKT/FOXO3a signaling in suppression of prostate tumorigenesis, PLoS One 8(12) (2013), e77922. doi: 10.1371/journal.pone.0077922.

24.

S.J.

Zhao

L.Q.

Wang

X.F.

Z.H.

Hua

R.X.

Wan

C.H.

Zhang

J.Y.

Zhang

X.W.

Huang

M.Z.

Gan

Sun

H.L.

Dimri

G.P.

and Guo

W.J.

, CBX7 regulates stem cell-like properties of gastric cancer cells via p16 and AKT-NF-κB-miR-21 pathways, J Hematol Oncol 11(1) (2018), 17. doi: 10.1186/s13045-018-0562-z.

25.

Suter

C.M.

Martin

D.I.

and Ward

R.L.

, Hypomethylation of L1 retrotransposons in colorectal cancer and adjacent normal tissue, Int J Colorectal Dis 19(2) (2004), 95–101. doi: 10.1007/s00384-003-0539-3.

26.

Goelz

S.E.

Ogelstein

B.V.

Hamilton

S.R.

and Feinberg

A.P.

, Hypomethylation of DNA from benign and malignant human colon neoplasms, Science 228(4696) (1986), 187–190. doi: 10.1126/science.2579435.

27.

Wienholz

B.L.

Kareta

M.S.

Moarefi

A.H.

Gordon

C.A.

Ginno

P.A.

and Chédin

, DNMT3L modulates significant and distinct flanking sequence preference for DNA methylation by DNMT3A and DNMT3B in vivo, PLoS Genet 6(9) (2010), e1001106. doi: 10.1371/journal.pgen.1001106.

28.

Puszyk

Down

Grimwade

Chomienne

Oakey

R.J.

Solomon

and Guidez

, The epigenetic regulator PLZF represses L1 retrotransposition in germ and progenitor cells, EMBO J 32(13) (2013), 1941–1952. doi: 10.1038/emboj.2013.118.

29.

Doulatov

Notta

Rice

K.L.

Howell

Zelent

Licht

J.D.

and Dick

J.E.

, PLZF is a regulator of homeostatic and cytokine-induced myeloid development, Genes Dev 23(17) (2009), 2076–2087. doi: 10.1101/gad.1788109.

30.

Chang

M.T.

Asthana

Gao

S.P.

Lee

B.H.

Chapman

J.S.

Kandoth

Gao

Socci

N.D.

Solit

D.B.

Olshen

A.B.

Schultz

and Taylor

B.S.

, Identifying recurrent mutations in cancer reveals widespread lineage diversity and mutational specificity, Nat Biotechnol 34(2) (2016), 155–163. doi: 10.1038/nbt.3391.

31.

Ali

S.M.

Yao

Wang

Cheng

Schrock

A.B.

Chirn

G.W.

Chen

Gay

Elvin

J.A.

Suh

Miller

V.A.

Stephens

P.J.

Ross

J.S.

and Wang

, Comprehensive genomic profiling of different subtypes of nasopharyngeal carcinoma reveals similarities and differences to guide targeted therapy, Cancer 123(18) (2017), 3628–3637. doi: 10.1002/cncr.30781.

32.

Hollstein

Sidransky

Vogelstein

and Harris

C.C.

, P53 mutations in human cancers, Science 253(5015) (1991), 49–53. doi: 10.1126/science.1905840.

33.

Vousden

K.H.

and Lu

, Live or let die: The cell’s response to p53, Nat Rev Cancer 2(8) (2002), 594–604. doi: 10.1038/nrc864.

34.

Scott

C.L.

Gil

Hernando

Teruya-Feldstein

Narita

Martínez

Visakorpi

Cordon-Cardo

Peters

Beach

and Lowe

S.W.

, Role of the chromobox protein CBX7 in lymphomagenesis, PNAS 104(13) (2007), 5389–5394. doi: 10.1073/pnas.0608721104.

35.

Park

I.Y.

Sohn

B.H.

Choo

J.H.

Joe

C.O.

Seong

J.K.

Lee

Y.I.

and Chung

J.H.

, Deregulation of DNA methyltransferases and loss of parental methylation at the insulin-like growth factor II (Igf2)/H19 loci in p53 knockout mice prior to tumor development, J Cell Biochem 94(3) (2005), 585–596. doi: 10.1002/jcb.20263.

36.

Dong

J.T.

, Prevalent mutations in prostate cancer, J Cell Biochem 97(3) (2006), 433–447. doi: 10.1002/jcb.20696.

37.

Krug

Ganser

and Koeffler

H.P.

, Tumor suppressor genes in normal and malignant hematopoiesis, Oncogene 21(21) (2002), 3475–3495. doi: 10.1038/sj.onc.1205322.

38.

Shetzer

Kagan

Koifman

Sarig

Kogan-Sakin

Charni

Kaufman

Zapatka

Molchadsky

Rivlin

Dinowitz

Levin

Landan

Goldstein

Goldfinger

Pe’er

Radlwimmer

Lichter

Rotter

and Aloni-Grinstein

, The onset of p53 loss of heterozygosity is differentially induced in various stem cell types and may involve the loss of either allele, Cell Death Differ 21(9) (2014), 1419–1431. doi: 10.1038/cdd.2014.57.

Alterations of epigenetic regulators and P53 mutations in murine mesenchymal stem cell cultures: A possible mechanism of spontaneous transformation

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Materials and methods

2.1 Expansion of cryopreserved mouse MSC passages

2.2 Isolation and culture of control bone marrow MSCs from C57BL/6J mice

2.3 RNA extraction

2.4 cDNA synthesis

2.5 Real-time reverse transcription polymerase chain reaction for mouse polycomb and trithorax complexes

2.6 P53 amplification

Table 1a The fold regulation of the 35 dysregulated epigenetic regulators in passage 3 (group1)

3. Results

3.1 Most of polycomb and trithorax complexes showed abnormal expression in early and late MSC passages

Table 2 The fold regulation of the 9 abnormally expressed genes in all MSC passages

3.3 Substitution mutations were detected in the early and late while deletion and insertion mutations were appeared in the late MSC culture passages

5. Conclusions

Funding

Author contributions

Supplementary data

sj-docx-1-cbm-10.3233_CBM-203121.docx - Supplemental material

Footnotes

Acknowledgments

References

Table 1a
The fold regulation of the 35 dysregulated epigenetic regulators in passage 3 (group1)

Table 2
The fold regulation of the 9 abnormally expressed genes in all MSC passages