Mitochondrial alteration in malignantly transformed human small airway epithelial cells induced by alpha particles

Abstract

Human small airway epithelial cells (SAECs) immortalized with human telomerase reverse transcriptase (h-TERT) were exposed to either a single or multiple doses of α particles. Irradiated cells showed a dose-dependent cytotoxicity and progressive neoplastic transformation phenotype. These included an increase in saturation density of growth, a greater resistance to PALA, faster anchorage-independent growth, reinforced cell invasion and c-Myc expression. In addition, the transformed cells formed progressively growing tumors upon inoculation into athymic nude mice. Specifically, α-irradiation induced damage to both mitochondrial DNA (mtDNA) and mitochondrial functions in transformed cells as evidenced by increased mtDNA copy number and common deletion, decreased oxidative phosphorylation (OXPHOS) activity as measured by cytochrome C oxidase (COX) activity and oxygen consumption. There was a linear correlation between mtDNA copy number, common deletion, COX activity and cellular transformation represented by soft agar colony formation and c-Myc expression. These results suggest that mitochondria are associated with neoplastic transformation of SAEC cells induced by α particles, and that the oncogenesis process may depend not only on the genomes inside the nucleus, but also on the mitochondrial DNA outside the nucleus.

Introduction

Epidemiological and animal studies have shown that exposure to high concentration of radon is associated with an increased risk of lung cancer both in uranium miners and in experimental animals.^1,2 There is evidence that immortalized human bronchial epithelial cells can be malignantly transformed by a single dose of α particles.³ Since the discovery of X-rays more than a hundred years ago, it has usually been assumed that nuclear DNA (nDNA) is the direct target of radiation, in which oncogenes or tumor suppressors are altered to initiate carcinogenesis. Recent studies, however, have revealed that mitochondrial DNA (mtDNA) may also participates in the tumorigenic process.⁴ mtDNA has been shown to be more susceptible than nuclear DNA upon exposure to either radiation or other environmental carcinogens due to lack of protective histone and a less efficient repairing system.⁵ Mitochondria are cytoplasmic organelles with a variety of functions such as generation of ATP through respiration and oxidative phosphorylation (OXPHOS), production of reactive oxygen species (ROS), initiation and execution of apoptosis.⁶ Compared with glycolysis, mitochondrial respiration and OXPHOS are more efficient pathways in generating ATP from glucose in normal human cells. In contrast, a shift in glucose metabolism from OXPHOS to glycolysis (the Warburg effect) has frequently been observed in cancer cells due to defective OXPHOS which could potentially be applied as a biochemical hallmark of tumors.^7–9

In some previous studies, human epithelial cell lines used for transformation assays were generally set up by incorporating virus or certain gene segments to immortalize them, which is an essential pre-requisite for neoplastic transformation.^10–12 However, viral oncoproteins such as the large T in SV40 virus and E6/E7 in papillomarvirus are genomically unstable and intracellular signaling pathways preclude many normal gene functions. Thus, human telomerase reverse-transcriptase (h-TERT)-immortalized human small airway epithelial cells (SAECs) were created as a better alternative model to study intracellular molecular pathways related to cell transformation.¹³ In the present study, we examined various alterations in mtDNA and mitochondrial functions induced by α particles to see if these changes correlate with malignant cell transformation.

Materials and Methods

Cell culture and irradiation

h-TERT immortalized human SAEC cells established in one of the co-authors’ laboratory as described¹³ were used in this study. Cells were cultured in SAGM medium supplemented with various growth factors supplied by the manufacturer (SAEM, SAGM Single Quots, Lonza,Walkersville, MD, USA) and maintained at 37°C in a humidified 5% CO₂ incubator.

For irradiation, exponentially growing SAEC cells were plated in specially designed stainless steel rings with a 6 μm mylar bottom at a density of 5×10⁵ cells per dish one day before irradiation. For facilitating cell attachment, the mylar surface was coated with fibronectin (Invitrogen, Carlsbad, CA, USA) for 2 hrs before plating cells. Cells were irradiated with graded doses of 120 keV/μm ⁴He ions accelerated in the 4 MeV Singletron accelerator at Columbia University Radiological Research Accelerator Facilities (RARAF) as described previously.³ These high energy particles have a LET value comparable to α-particles emitted by radon daughter products. To monitor the efficiency of α-particles in transforming human SAEC cells, the cells were irradiated with either a single exposure or with multiple doses of up to 4 consecutive exposures with an interval of two weeks between irradiations. The irradiation doses were 0, 0.2, 0.4, 0.8, 1.0 and 2.0 Gy, respectively, in either single or multiple treatments. On an average, 4 to 6 dishes of cultures were used in each dose per experiment. After irradiation, part of cells from each irradiation dose were sub-cultured immediately to determine survival fractions while the rest of cells were pooled separately and expanded in cultures to assay for transformed phenotypes and/or to prepare for additional irradiation treatment.

Single dose cytotoxicity and survival fraction

Following irradiation, cells were trypsinized and resuspended in SAGM medium. The cell suspension was centrifuged and the resultant pellet was resuspended in SAGM. The number of cells was counted with an electronic cell counter (Coulter Electronic, Hialeah, FL, USA) and plated into 60 mm diameter dishes at a density that 50–60 viable cells would survive the irradiation treatment to form colonies. Following a 12–14 day incubation period, the dish bottoms were fixed and stained with crystal violet (Sigma, St. Louis, MO, USA) and the number of colonies was counted manually. Survival fraction was calculated as the ratio of plating efficiency in irradiated groups to that of normal control SAEC cells.

PALA resistance and cell proliferation

To determine the effects of PALA (N-phosphonoacetyl-L-aspartate, a cancer therapy drug, from the Drug and Synthesis Branch, Division of Cancer Treatment, National Cancer Institute, USA) on colony growth of the irradiated population, 2,000 exponentially growing cells from each group were plated into 100 mm diameter dishes with medium supplemented with 100 mM PALA (5 times LD₅₀ to the SAEC cells¹⁴). At the end of the 14 day incubation period, after aspiration of the medium, the culture dishes were fixed with paraformaldehyde and stained with crystal violet. The number of colonies was counted and PALA resistance was defined as the ratio of plating efficiency of cells in medium with PALA to that in medium without PALA.

To determine the growth kinetics of putative transformed SAEC cells post-irradiation, plating efficiency, cell growth rate and saturation density were examined. Plating efficiencies of cells were examined at passages 0, 10 and 40 post-irradiation. For growth curves, cells from various exponentially growing cultures were replated into 24 60 mm dishes with 1 × 10⁵ cells per dish in SAGM medium. At the time points studied (1, 2, 3, 5, 7, 9, 11 and 14 day), triplicate dishes from each group were trypsinized and the total number of cells per dish was counted. Growth curve and saturation density were analyzed with cell numbers and were indicated as indices of early transformed phenotypes upon α-irradiation as described³.

Anchorage-independent growth and invasion assay

To test anchorage-independent growth (soft agar growth capacity), control and irradiated SAEC cells at four month post-irradiation were trypsinized and replated at a density of 2 × 10³ cells in 2 ml of 0.35% agarose (Invitrogen, Carlsbad, CA,USA) over a 0.7% agar base in 60 mm dishes. Medium was replenished every three days and colonies with more than 50 cells were scored after 4 weeks in culture and the size of colonies was measured under microscope.

For invasion assay, the Matrigel invasion chamber was used following instructions provided by the manufacturer (BD BioCoat Matrigel invasion chamber, BD Biosciences, Bedford, MA, USA). Briefly, 8 mm-diameter filters (8 μm pore) of cell culture inserts coated with matrigel were placed in 24-well culture plates. Exponentially growing cells were trypsinized and resuspended in SAGM medium without supplemental growth factors. After counting, the cells were added to the upper compartment of the chamber (1 × 10⁵ cells/chamber). Growth factors were added in the medium in the lower chamber as chemoattractants. After incubation for 23 hrs at 37°C in a 5% CO₂ incubator, cells on the upper surface of the filter were removed by wiping with a cotton swab, and cells that had traversed the matrigel and attached to the lower surface of the filter were kept for studying. The filters were fixed and stained with crystal violet and cells on the lower surface of the membrane were counted under a light microscope and the average value was calculated. Experiments were repeated 3 times with 3 chambers per group.

c-Myc and p53 expression

c-Myc and p53 protein expression were examined by Western blot. Cellular proteins were extracted by lysing cells in extraction buffer (50 mM Tris-HCl, pH8, 150 mM NaCl, 1% NP-40, 0.1% sodium dodecyl sulfate, and 1 mM phenylmethylsulfonyl fluoride). The protein concentration was determined by Bio-Rad protein assay (Bio-Rad, Hercules, CA, USA). Equal amounts of protein (30 μg) were fractionated by electrophoresis in SDS-polyacrylamide gel. The proteins were subsequently transferred to PVDF membranes under semi-dry conditions. Antibodies against c-Myc and p53 (Cell Signaling Technology, Danvers, MA, USA) were applied to probe the membranes, respectively. The secondary antibodies (anti-rabbit or anti-mouse) (Amersham Bioscience, Piscataway, NJ, USA) were conjugated to horseradish peroxidase. Signals were detected using the ECL system (Amersham Bioscience, Piscataway, NJ, USA).

In vivo tumorigenicity assay

Four-week old male nu/nu mice from Harlan Laboratory (Indianapolis, IN, USA) were given a whole-body, 4 Gy dose of γ-irradiation. 24 hrs later, each animal was injected subcutaneously at two different sites with 5 × 10⁶ cells resuspended in 0.2 ml PBS per site, one at the interscapular area and the other at the lower back. The mice were divided into 4 groups of 5 animals each: control SAEC, 0.4 Gy-M (0.4 Gy multiple group), 0.8 Gy-S (0.8 Gy Single group) and 0.8 Gy-M (0.8 Gy multiple group). Animals were maintained under sterile conditions for 6 months and palpated for tumor appearance once a week. Animals were sacrificed as soon as the tumor nodules attained 0.8–1 cm in size and pathological section and HE stain of the xenografts were performed to determine their origin.

Mitochondrial copy number determination

Mitochondrial copy number was determined by real-time PCR using SYBR Green detection on an Applied Biosystems 7300 Real-time PCR System (Applied Biosystem, Carlsbad, CA, USA). 12S rRNA encoded by mtDNA (forward primer AGAACACTACGAGCCACAGC, reverse primer ACTTGCGCTTACTTTGTAGCC) and 18S rRNA (forward primer GGAGTATGGTTGCAAAGCTG, reverse primer CGCTCCACCAACTAAGAACG)/GAPDH (forward primer TACTGGTGTCTTCAC-CACCA, reverse primer CAGGATGCATTGCTGACAATC) encoded by nDNA were amplified.

Total cell DNA was extracted with DNAZol assay kit (Invitrogen, Carlsbad, CA, USA) and quantified by spectrometry. 50 ng DNA was used as template for the PCR amplifications. All reactions were done in triplicates. Real-time PCR conditions were set as follows: 50°C 2 min, 95°C 10 min followed by 40 cycles at 95°C for 30 s, 55°C for 30 s, and 72°C for 30 s. Relative quantification of mtDNA/nDNA ratio was defined as mitochondrial copy number and determined by comparative threshold cycle (C_T) method described previously.¹⁵

Mitochondrial common deletion determination

The human mtDNA 4977-bp common deletion was determined as previously described with modifications.¹⁵ The quantification was performed by a two-round nested PCR with two sets of primers that flank the deletion junction. With a short PCR extension time, the primers preferentially amplify those mtDNAs that contain the 4977-bp deletion, but not the wild type mtDNAs. A 504-bp fragment (Primers were 8252S (CCCGTATTTACCCTATAGCAC), 13734AS (AATCCTGCGAATAG-GCTTCCGGCTG)) was amplified from the first round regular PCR and then served as the template for the second round real-time PCR, which amplified a 104-bp fragment (Primers were 8417S (CCTTACACTATTCCTCATCACCCA), 13499AS (CCTGTGAGGAAAGGTATTCCTGCTA)). The first round PCR was carried out in a 25 μl reaction volume with a Bio-Rad DNA engine Pelfier Thermal Cycler using a high-fidelity polymerase (Ex Taq, Takara, Japan). The PCR conditions were as follows: 94°C for 5 min followed by 30 cycles at 95°C for 30 s, 60°C for 30 s, and 68°C for 30 s and then a final extension at 72°C for 7 min. The second real-time PCR was carried out with SYBR Green using the Applied Biosystems 7300 Real-time PCR System (Applied Biosystems, Carlsbad, CA, USA). All reactions were done in triplicates. The Real-time PCR conditions were as follows: 50°C for 2 min, 95°C for 10 min followed by 40 cycles at 95°C for 15 s, 58°C for 30 s, and 72°C for 30 s.

Lymphocyte DNA from a patient homoplastic for the 4977-bp common deletion was used as a positive control to create a quantification standard curve. The same amount of mtDNA (quantified by 12S rRNA with real-time PCR) from each sample and positive control were subjected to the two-round PCR. Results from the transformed SAEC and control cells were compared with the standard curve and the incidence of common deletion was calculated based on the standard curve, and represented as “fold increase” over the normal controls. The PCR products from each sample were purified and sequenced for further confirming the results.

Cytochrome C Oxidase (COX) assay

The COX activity was assayed by biochemistry kit (Sigma, St. Louis, MO, USA). Briefly, exponentially growing cells were trypsinized and washed with pre-cold PBS, and the pellets were resuspended in reaction lysis buffer at 4°C for 15 min and mitochondria were isolated by gradient centrifugation twice. COX activity was assessed by measuring the oxidation of reduced cytochrome C at 550 nm over 60 seconds (nmol oxidized cytochrome C per min per mg protein). The enzyme activities were calculated with: Units/ml= (ΔA/min × dil × 1.1)/((vol of enzyme) × 21.84) and then normalized to per mg protein of cell lysates.

For COX expression, Western blot was employed to measure COX1 and COX4. Same style of protein extraction and SDS electrophoresis were described previously. Antibodies against COX1 and COX4 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) were applied to probe the membranes. The secondary antibodies were conjugated to horseradish peroxidase and ECL assay was used to detect the signals (Amersham Bioscience, Piscataway, NJ, USA).

Oxygen Consumption assay

Oxygen consumption in intact cells was assayed as described previously.¹⁵ Briefly, (5–10) ×10⁶ cells were trypsinzed and washed in PBS and then resuspended in 1.5 ml DMEM lacking glucose, and oxygen consumption was measured over a 3 min period at 37°C in a Hansatech Clark’s oxygen electrode unit. Oxygen consumption was calculated by the formula: fmol/cell/min = (0.217 (μmol/mL) × 1.5 mL × A/min (mV/min)/(“100” - “0”)mV × cell numbers).

Membrane potential measurement

Membrane potential (ΔΨm) was detected by flow cytometry using JC-1 (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl-benzimidazolocarbocyanine iodide) dye (Sigma, St. Louis, MO, USA). Any event that dissipates the mitochondrial membrane potential prevents the accumulation of the JC-1 dye in the mitochondria and thus, the dye is dispersed throughout the whole cell leading to a shift from red (J-aggregates) to green fluorescence (JC-1 monomers). Briefly, (5–10) × 10⁵ cells were collected and stained with JC-1 at 37°C for 20 min and washed with PBS twice, then resuspended in JC-1 staining buffer and the JC-1 fluorescence was detected in FL2 channel of flow cytometry and defined as membrane potential.

Statistical and Correlation analysis

Data were presented as mean ± standard deviations. Comparisons between irradiation groups and controls were made by student’s t-test and one-way ANOVA. A p value of 0.05 or less between groups was considered to be significant.

Correlation between mitochondrial alteration and cell transformation was analyzed by SPSS17.0 statistical software. Soft agar colony formation and c-Myc expression were treated as independent variables, with dependent variables of mitochondrial copy number, common deletion, COX activity, oxygen consumption and membrane potential. p<0.05 was considered to be significant between mitochondrial alteration and cell transformation.

Results

Cell growth kinetics and PALA resistance

The h-TERT immortalized SAEC cells grew as a contact-inhibited monolayer with a population doubling time of ~24 hrs and plating efficiency of 10–15%. Upon exposure to graded doses of α-particles, plating efficiency decreased gradually and cells exhibited a reverse linear dose-response survival curve from 1.0 in control SAEC cells to 0.12 in 2.0 Gy group as shown in Fig. 1a.

Figure 1.

Plating efficiencies of the SAEC cells at different subcultured times (P0: Cells within 30min after irradiation, P10: passage 10 after first irradiation and P40: passage 40 after first irradiation) of single treatment (S) and multiple treatments (M) were compared. In P0, cells showed a progressive decrease in P.E. with increasing radiation dose which was largely restored to normal level by P10. In contrast, the P.E. of the various irradiated groups increased to about 30~40% by passage 40. The highest plating efficiency of 40.3% occurred in the 0.8 Gy single treatment group and 42% in the multiple treatment group (Fig. 1b). In addition, the saturation density in the 0.8 Gy multiple group increased to 10.8 × 10⁵/cm² compared to 5.2 × 10⁵/cm² in the control SAEC cells (Fig. 1c).

Resistance to the chemotherapeutic drug PALA, derived as a result of amplification of the CAD gene (carbamyl phosphate synthetase/aspartate transcarbamylase/dihydroorotase) is often used as an index of genomic instability. Three months (about passage 30) after first irradiation, the irradiated cells exhibited a much higher frequency of (1–2) × 10⁻² PALA resistance compared with that of (4–7) × 10⁻⁵ found in control cells (Fig. 1d), indicating a 10³ fold increase in the activity of the CAD gene.

Anchorage-independent growth and tumorigenicity

Using anchorage-independent growth assay and by four-month post-irradiation, all irradiated cells were found to produce agar-positive colonies irrespective of dose. Cells irradiated with 0.8 Gy dose of α-particles yielded the highest percentage of anchorage independent growth of 10% (Fig. 2a). In contrast, control SAEC cells showed no anchorage-independent growth, indicating that irradiated SAEC cells have undergone transformed changes during the four month of incubation after the radiation treatment.

Figure 2.

In vitro invasion assay is often used as a surrogate marker of tumor cell metastasis. After cell transformation, the invasion ability of irradiated SAEC cells exposed to multiple treatments with either 0.4 or 0.8 Gy together with an anchorage independent clone isolated from soft agar (CFS) seeded with 0.8 Gy-M were determined, As shown in Fig. 2b, the number of cells traversed the transwell matrigel membrane increased from 50 cells in the control group to 250–400 cells in the transformed groups, representing a 4–7 fold increase of invasion ability.

The expression of c-Myc and p53 proteins, determined by Western blots, was shown in Fig. 2c. Expression of c-Myc protein was up-regulated in the transformed cells irradiated with either a single or multiple doses of 0.4 Gy and 0.8 Gy of α-particles. In contrast, expression of p53 protein was found to be down-regulated in all the irradiated groups relative to control (Fig. 2c).

Tumorigenicity in immunosuppressed animals is often used as the ultimate test for malignant transformation. Among the three groups of irradiated cells tested, tumor nodules were found in 3 out of 5 nude mice from only the 0.8 Gy-M group at the injection sites 4 months after inoculation. Histological observation confirmed the xenograft to be epithelial and tumor origin, and transformed cells could be seen penetrating into the muscles in the HE stained histological section as shown in Fig. 2d.

Changes in mitochondrial copy number and common deletion

After transformation, mtDNA changes in transformed cells of 0.4 Gy and 0.8 Gy, single and multiple groups were measured to quantify the mitochondrial copy number and common deletion. Mitochondrial copy number was represented by the ratio of 12S rRNA/18S rRNA or 12S rRNA/GAPDH. The two internal standards showed similar results. Compared with control SAEC cells, the transformed cells from 0.4 Gy and 0.8 Gy groups showed an increase of 1.4–2.9 folds in mitochondrial copy number as shown in Fig. 3(a and b) and Table 1. As for common deletion measurement, according to the linear curve derived from the positive control (Fig. 3c), the rate of common deletion in the transformed cells increased 2.6–4.9 times compared with control SAEC cells (Fig. 3d and Table 1).

Figure 3.

Table 1.

Mitochondrial DNA and function changes in transformed SAECs

Groups	Copy number change		Common deletion change	COX activity (mUnits/mg protein)	Oxygen consumption (fmol/cell/min)	Membrane potential (Transformed/Control)
	12S/18S	12S/GAPDH
Control	1	1	1	784 ± 91.3	1.43 ± 0.18 (1.43 ± 0.18)#	77.6 ± 5.3
0.4 Gy-S	1.42 ± 0.48	1.78 ± 0.29*	2.6 ± 0.66	616 ± 170.7	1.69 ± 0.53 (0.92 ± 0.25)	62.4 ± 4.6*
0.4 Gy-M	1.95 ± 0.56*	2.32 ± 0.62*	3.95 ± 0.5*	336 ± 158.7*	1.87 ± 0.40 (0.81 ± 0.16)*	52.3 ± 8.1*
0.8 Gy-S	2.01 ± 0.49*	2.01 ± 0.4*	3.58 ± 0.89	299 ± 64.7*	1.98 ± 0.62 (0.99 ± 0.30)	54.2 ± 8.9*
0.8 Gy-M	2.12 ± 0.36*	2.92 ± 0.29*	4.9 ± 1.45*	224 ± 112*	2.8 ± 0.28 (0.93 ± 0.09)*	50.8 ± 1.7*

^*:Compared with control group, p<0.05.

^#:Oxygen consumption after adjusting with mitochondrial copy number. Data were from 3 independent experiments and mean ± SEM were showed.

Changes in mitochondrial functions

In addition to mitochondrial DNA alteration, changes in mitochondrial functions were also examined including COX activity, oxygen consumption and membrane potential in the well characterized transformed 0.4 and 0.8 Gy groups. The results showed a gradual decrease in the whole COX activity from 784 ± 91.3 mUnits/mg protein in control cells to 224 ± 112 mUnits/mg protein in transformed 0.8 Gy-M cells (Fig. 4a and Table 1), and a decrease in the expression of COX1 and COX4 as shown in Fig. 4b. The decrease in COX enzyme activity was found to be proportional to the radiation dose and number of treatments. Relative to the control, the percentage decrease in COX enzymatic activities for 0.4 Gy-S, 0.4 Gy-M, 0.8 Gy-S and 0.8 Gy-M was 78%, 43%, 38% and 28%, respectively.

Figure 4.

An increase in overall oxygen consumption was found in all the transformed cells with 2.8 ± 0.28 fmol/cell/min in 0.8 Gy-M group compared to 1.43 ± 0.18 fmol/cell/min in control cells. However, oxygen consumption depends on mitochondrial content. Upon adjusting for mitochondrial copy number, there was a proportional decrease in oxygen consumption in the transformed cells. Relative to the control, the percentage decrease in oxygen consumption, after adjusting for mitochondrial copy number, for 0.4 Gy-S, 0.4 Gy-M, 0.8 Gy-S and 0.8 Gy-M was 64%, 57%, 69% and 65%, respectively (Fig. 4c and Table 1).

Red fluorescence, which represents an intact mitochondrial membrane potential, decreased in the transformed cells. Software analysis also revealed depressed membrane potential from 77.6 ± 5.3 in control cells to 50.8 ± 1.7 in 0.8 Gy-M transformed cells (Fig. 4d and Table 1).

Correlation between mitochondrial alteration and cell transformation

Correlation between mitochondrial alteration and cell transformation was analyzed by SPSS17.0 software. Soft agar colony formation and c-Myc expression were used as independent variables. Analysis of the data shown in Table 2 demonstrated a linear correlation between mitochondrial copy number, common deletion, COX activity and cell soft agar colony formation/c-Myc expression, indicating an association of mitochondrial alteration with cell transformation.

Table 2.

Correlation coefficient between mitochondrial function/DNA change and soft agar colony formation/c-Myc expression

Indexes	Correlation coefficient of R2
	Soft agar colony formation	c-Myc expression
12S rRNA/18S rRNA	0.791*	0.768
12S rRNA/GAPDH	0.759	0.901*
Common deletion	0.800*	0.875*
COX activity	0.784*	0.794*
Oxygen consumption	0.484	0.384
Membrane potential	0.766	0.711

^*: Coefficient correlation statistical analysis p<0.05.

Discussion

Alpha particles have been shown to induce neoplastic transformation of mammalian epithelial cells with similar potential as some chemical carcinogens such as the tobacco nitrosamine NNK and arsenical compounds.^3,14,16,17 Our results confirmed that either a single or multiple doses of α-particles could initiate transformation of the immortalized SAEC cells, resulting in morphological changes and PALA resistance, as well as increased plating efficiency and colony formation in soft agar, ultimately led to malignant changes in our animal model.

Assessed by survival fraction, an acute cytotoxicity of α-particles on SAEC cells was characterized by a reverse linear dose-response shortly after irradiation, a similar result on other epithelial cells such as BEP2D which was reported in previous studies.³ In the long term subculture after irradiation, SAEC cells became transformed as indicated by increased plating efficiency and saturation density. These changes have been viewed as a phenotype of early transformation, and represent a status of accelerating proliferation and loss of contact inhibition in cell growth. On the other hand, anchorage-independent (soft agar) growth and invasive migration have been linked to late stage transformation that can often be demonstrated in vivo.¹⁸ In our experiment, irradiated cells of the 0.8 Gy group exhibited the highest rates of anchorage independent growth and invasive migration as compared with other groups. Correspondingly, this group also formed tumors in 3 out of 5 mice inoculated, the ultimate test of malignancy.

At the molecular level, expression of two proteins chosen in this study that are regarded to play an important role in cellular transformation showed an up-regulation of the oncogene c-Myc and a down-regulation of the tumor suppressor gene p53. Elevated expression of c-Myc, c-Ras, and c-Fos have frequently been detected during malignant transformation in many cell types.^19,20 p53 plays a key role in maintaining genome integrity and accuracy of chromosome segregation. Over expression of mutant p53 or loss of function of wild type p53 protein has been found in ~70% of human cancers.²¹ The change in the expression of c-Myc and p53 in this study was therefore conceivable and correlated with the neoplastic phenotypes of the irradiated SAEC cells.

PALA was used to measure the frequency of gene amplification in mammalian cells,²² by preventing pyrimidine biosynthesis through inhibition of the trifunctional enzyme CAD (carbamyl phosphate synthetase/aspartate transcarbamylase/dihydroorotase) in the synthesis of UMP, which could induce DNA mutation, chromosomal breaks and apoptosis.^23,24 A much higher activity of PALA resistance of the irradiated cells in this study suggested that an abnormal gene amplification and emergence of genomic instability among the α particle irradiated cells.

Recent reports suggest that mitochondria play an important role in the regulation of α particle-induced bystander effect.^25,26 Mitochondria are semi-autonomous organelles capable of independent gene transcription, translation and protein expression, mainly function in the process of cell energy metabolism, free radical generation and cell apoptosis. Unlike the nuclear genome, mitochondrial DNA (mtDNA) is devoid of histone protection and effective repair system, and therefore is vulnerable to carcinogenic attack.⁵ Damage to mtDNA may lead to mutation, integration and instability of the nuclear genome. Exposure to the same stressing factor, the mutation frequency in mtDNA would be 10 times higher than that of nuclear genome, and a small change in the sequence could affect the whole structure and subsequently resulting in the major function changes of mitochondria.⁷ Somatic mutation and damage to mtDNA can result in impairment of the oxidative phosphorylation (OXPHOS) system and enhance ROS production, which in turn accelerates the rate of DNA mutation.^27,28 It is likely that cancer arises not only from mutations in the genome inside the nucleus, but the process is facilitated by mitochondrial DNA mutations outside the nucleus. Changes in the number and sequence of mtDNA, as well as mitochondrial functions may play an important role in the initiation and maintenance of cancer cells.

A single cell contains hundreds of mitochondrion, each of them contains multiple copies of mitochondrial genome. Alterations in copy number could be induced by stressing factors and subsequently impair mitochondrial functions. Many studies reported that mitochondrial gene expression in tumor cells was up-regulated mostly due to an adaptive reaction to environmental stress in order to supply an increased energy demand.^28–30 Mutation in D-loop region of mtDNA may cause accelerated mtDNA replication and transcription, and long-term accumulation of increased copy number may lead to tumorigenesis. ³⁰ In our experiment, mtDNA copy number and common deletion were both increased in transformed SAEC cells of 0.4 Gy and 0.8 Gy groups. It was reported that α irradiation can induce breaks of mtDNA segment³¹ by displaying an increase in mitochondrial common deletion, which may include fragments encoding important genes such as 5 tRNA, cytochrome oxidase complex I and ATP synthase in the respiratory chain. Due to the defect in mitochondrial respiratory enzymes, the increased mitochondrial copy number may act as a salvage response to maintain the supply to the energy demand.

The effect of α irradiation on OXPHOS system was evidenced by a decrease of COX activity as well as a reduction of oxygen consumption and the dissipation of the mitochondrial electrochemical potential gradient (Membrane potential, ΔΨm) in the transformed SAEC cells, the disturbed mitochondrial respiration and ATP generation which were proposed as Warburg effect in tumor cells.³² Oxygen consumption among irradiated/transformed cells decreased after adjustment for mitochondrial copy number implying a defective mitochondrial function which is consistent with decreased cytochrome c oxidative activity. In terms of percentage change relative to the control, the decrease in COX activities among irradiated/transformed cells was higher than for oxygen consumption level. Complex IV activity does not strictly correlate with the levels of oxygen consumption. This is perhaps due to enhanced ROS production and the subsequent increase in mtDNA copy number, which has been associated with the degradation of complex IV subunits.^33,34

It remains unclear how mitochondrial alteration is associated with cellular transformation. In mammalian cells, mitochondria are one of the major sources of ROS production, which has an important role in diverse events such as cellular proliferation, differentiation and migration.^35,36 ROS production can be of endogenous (mitochondrial and P450 metabolism, inflammatory cell activation, etc.) as well as of exogenous (radiation, ozone, etc.) origin. As indicated in our previous experiment, oxidative stress in mammalian cells with depleted mitochondrial DNA significantly decreased compared to cells with normal mitochondrial DNA after alpha irradiation or asbestos,^25,37 and targeted cytoplasmic irradiation could induce oxidative DNA damages and reactive nitrogen species (RNS) in A_L cells,³⁸ suggesting that oxidative stress is mostly generated from the damaged mitochondria induced by alpha particles. Since the mitochondrial respiratory chain (electron transport complexes) is a major source of ROS generation in the cells, the vulnerability of the mitochondrial DNA to ROS-mediated damage appears to be a mechanism underlying ROS stress in cancer cells. Growing evidence suggests that cancer cells exhibit an enforced intrinsic ROS stress, and when ROS generation exceeds the cell’s ability to metabolize and detoxify them, a state of “oxidative stress” emerges.^39–41 The carcinogenicity of oxidative stress is primarily attributed to the genotoxicity of ROS in diverse cellular processes.⁴² For example, hydroxyl radicals can react with pyrimidines and/or purines as well as chromatin proteins, resulting in base modifications and genomic instability respectively, all of which can cause alterations in gene expression.³⁵ DNA alterations like strand breaks, base modifications and DNA-protein cross linkages, are all strongly implicated in the initiation stage of carcinogenesis. A recent study has observed a direct relationship among oxidative stress, DNA damage and elevated frequency of p53 mutations in human dysplastic colon and in colorectal carcinoma.⁴³ Thus, ROS-mediated DNA damage is likely plays an important role in the initiation of carcinogenesis as well as in malignant transformation and may represent a major contributor in the pathogenesis of human carcinogenesis. ^44,45 It is therefore possible that mitochondrial ROS production resulted from α-irradiation may alter mitochondrial capacity and induce mtDNA mutation that in turn promote the subsequent cell transformation.

The novelty and impact statements of the paper are as follows:

1. The cell lines used in this study is human telomerase reverse-transcriptase (h-TERT)-immortalized human small airway epithelial cells (SAECs) which created as a better alternative model to study intracellular molecular pathways related to cell transformation.

2. This paper studied mitochondrial alteration in malignantly transformed human SAEC cells induced by alpha particles which provide nontraditional insight of carcinogenesis of alpha irradiation.

Even a single dose of radiation in the form of alpha particles can bring on cancer by damaging DNA. Originally, it was assumed that the particles caused cancer by mutating nuclear DNA, but some recent studies have showed that they may wreak their chaos on mitochondrial DNA, as well. In this study, the authors investigated the effect of alpha particles on immortalized small airway epithelial cells. They immortalized these cells using human telomerase reverse transcriptase, a better model for studying intracellular molecular pathways, which can be altered by viral methods of immortalizing cells. Cells that had been transformed by exposure to alpha particles showed damage to mitochondrial DNA, suggesting there may be more paths to cancer than previously thought.