Effect of Reoxygenation on Radioresistance of Chronically Hypoxic A549 Non-Small Cell Lung Cancer (NSCLC) Cells Following X-Ray and Carbon Ion Exposure

Abstract

Hypoxia-induced radioresistance in non-small cell lung cancer (NSCLC) hinders radiotherapy efficacy. Fractionated schedules exploit reoxygenation between fractions to reverse this resistance, but the effects of post-irradiation reoxygenation remain unclear and may depend on radiation quality. We investigated survival, cell cycle progression, cytokine secretion, and gene expression in hypoxic (1 % O₂) and reoxygenated A549 cells irradiated with X-rays or carbon ions. Colony-forming assays revealed an Oxygen Enhancement Ratio (OER) > 1 for both hypoxic and reoxygenated cells after X-rays, indicating persistent radioresistance; carbon ion OER ≈ 1 reflected oxygen-independent cytotoxicity. Hypoxia weakened radiation-induced G₂ arrest, and this was unaffected by reoxygenation. IL-6 secretion increased after X-rays and IL-8 after carbon ions exposure; both were enhanced under hypoxia and reoxygenation. RNA sequencing revealed that hypoxia induced a pro-survival, epithelial-to-mesenchymal transition (EMT)-promoting, and immune-evasive transcriptional program, which was largely reversed by reoxygenation but without increased clonogenic killing. These findings indicate that short-term reoxygenation after irradiation can normalize hypoxia-driven transcriptional changes yet does not restore radiosensitivity, supporting the advantage of high-linear energy transfer (LET) carbon ions for targeting resistant hypoxic NSCLC cells.

1. Introduction

Radiotherapy is used in treating over half of all cancer patients, and the response of solid tumors to radiation therapy is a key determinant of treatment efficacy in cancer management [1]. One of the most significant factors influencing radiosensitivity is the oxygenation status of tumor cells, as molecular oxygen in their nucleoplasm enhances the production of DNA-damaging reactive oxygen species (ROS) during irradiation, making it a potent radiosensitizer [2]. Tumor hypoxia, characterized by regions of low oxygen availability, poses a well-documented barrier to effective radiation therapy. As a result, hypoxic tumor cells often exhibit increased resistance to radiation-induced cell death compared to well-oxygenated cells, posing a major challenge in achieving optimal therapeutic outcomes [3].

Approximately two-thirds of X-ray-induced DNA damage occurs indirectly through ROS production. In contrast, high linear energy transfer (LET) radiation, such as carbon ions, deposits energy more densely along the tracks of radiation, leading to more direct DNA damage [4]. As a result, high-LET radiation is less dependent on ROS generation for its cytotoxic effects and may offer a therapeutic advantage in hypoxic tumors by circumventing the dependency on oxygen-mediated radiosensitization.

The Oxygen Enhancement Ratio (OER) quantifies the degree to which oxygen amplifies radiation-induced damage. It is defined as the ratio of the radiation dose needed to achieve a given biological effect under hypoxia versus normoxia [5]. For low-LET radiation, such as X-rays, the OER typically ranges from 2.5 to 3, indicating that hypoxic tumor cells may require up to three times the radiation dose to achieve the same level of cell killing as oxygenated cells. However, for high-LET radiation, the OER is lower due to the greater contribution of direct DNA damage mechanisms [6].

Beyond modulating radiosensitivity directly, hypoxia also induces a range of cellular adaptations that promote tumor cell survival, proliferation, invasion, and even metastasis, which can influence the outcome of radiotherapy. Under low-oxygen conditions, tumor cells activate hypoxia-inducible factors (HIFs), which regulate the expression of genes involved in angiogenesis, immune evasion, and metabolic reprogramming [7]. Immune evasion may protect hypoxic cells from detection and destruction by the innate and adaptive arms of the immune system [8]. Additionally, hypoxia promotes epithelial-to-mesenchymal transition (EMT), a process associated with increased invasiveness and resistance to apoptosis [9].

In tumor biology, reoxygenation refers to the process by which previously hypoxic tumor regions regain oxygen availability. This may occur following radiation treatment and represents a critical dynamic in modulating tumor radiosensitivity. For this reason, it serves as one of the rationales for fractionated radiotherapy, where the total radiation dose is delivered in multiple smaller fractions over time [10]. This approach allows for gradual reoxygenation of hypoxic tumor cells between fractions as radiation-induced cell death reduces overall metabolic demand and improves oxygen diffusion within the tumor microenvironment (TME). Consequently, previously radioresistant hypoxic cells may become more vulnerable to subsequent radiation doses [11].

However, reoxygenation has multifaceted effects on the TME. While restoring radiosensitivity via ROS generation, excessive ROS production may intensify oxidative stress, leading to DNA damage and genomic instability, which may further drive tumor progression and resistance to therapy [12]. Moreover, reoxygenation may influence cell cycle progression [13], alter cytokine secretion [14], and modulate key signaling pathways such as PI3K/AKT and MAPK, which are critical regulators of cell survival and proliferation [15,16,17]. While reoxygenation can enhance radiosensitivity, its long-term impact on tumor progression remains uncertain, warranting further investigation into its molecular consequences.

Reoxygenation prior to irradiation increases tumor radiosensitivity, and this is the key basis for fractionated radiotherapy. Due to the cellular adaptations to hypoxia described above, we hypothesized that reoxygenation after irradiation may also increase radiosensitivity by reversing the radioresistant phenotype induced by proliferative, EMT-related, and metabolic adaptations to hypoxia. Furthermore, since hypoxia influences radiosensitivity to X-rays to a higher extent than to carbon ions, we hypothesize that reoxygenation results in a stronger radiosensitization towards X-rays compared to carbon ions.

Previously, we reported that chronically hypoxic human NSCLC A549 cells (maintained at 1 % O₂) exhibit greater radioresistance to X-rays than their normoxic counterparts, whereas this effect is significantly reduced with carbon ion irradiation [18]. We also suggested that hypoxia-induced radioresistance may be linked to alterations in cell cycle dynamics and cytokine secretion [18,19,20].

In this study, we investigated the cellular response to reoxygenation following irradiation in chronically hypoxic A549 cells. Irradiation was performed using both X-rays and carbon ions. Specifically, we compared cells that were reoxygenated post-irradiation with those that remained under hypoxic conditions. We evaluated clonogenic survival, cell cycle distribution, cytokine secretion, and gene expression. By directly comparing continuously hypoxic and reoxygenated A549 cells, our study aims to elucidate the biological mechanisms underlying NSCLC cell survival and therapeutic response after reoxygenation.

2. Results

A549 cells were pre-incubated at 20 % O₂ (normoxia) and 1 % O₂ (hypoxia) for 48 h prior to X-ray or carbon ion exposure (Figure 1). Following irradiation, the hypoxic cells were either maintained at 1 % O₂ (continuous hypoxia) or 20 % O₂ (reoxygenation).

2.1. Minimal Impact of Reoxygenation on Clonogenic Survival of A549 Cells After X-Ray and Carbon Ion Exposure

Continuous hypoxia (1 % O₂ for 48 h) resulted in a higher surviving fraction of A549 cells following X-ray exposure compared to normoxia, as shown by colony formation assays (Figure 2a). This was reflected in a higher D₁₀ value under hypoxia (5.47 Gy vs. 4.16 Gy under normoxia;

Table A1. Dose required to reduce the surviving fraction to 10 % (D₁₀) of the pre-irradiation cell number after exposure of A549 cells to X-rays or carbon ions under various oxygen conditions.

Radiation Quality	Plating Time	D10 (Gy) (±SE)			p-Values
		Normoxia (N)	Hypoxia (H)	Reoxygenation (R: H→N)	Normoxia vs. Hypoxia	Normoxia vs. Reoxygenation	Reoxygenation vs. Hypoxia
X-rays	LP	4.16 ± 0.40	5.47 ± 0.27	5.52 ± 0.26	0.039	0.029	ns
Carbon ions	LP	1.89 ± 0.25	1.76 ± 0.38	1.79 ± 0.14	ns	ns	ns

D₁₀ was derived from the survival curves generated by colony-forming ability (CFA) assays (Figure 2); LP: late plating; ns: not significant.

and Figure 2c). Reoxygenation following X-ray exposure (D₁₀ = 5.52 Gy) did not significantly affect radioresistance relative to continuous hypoxia.

In contrast, after carbon ion irradiation, neither continuous hypoxia nor reoxygenation significantly altered clonogenic survival compared to normoxia. This is evident from overlapping survival curves (Figure 2b) and similar D₁₀ values (Table A1 and Figure 2c).

Relative Biological Effectiveness (RBE) of carbon ions relative to X-rays was consistently > 2 across all conditions (

Table A2. Relative Biological Effectiveness (RBE) based on D₁₀.

Radiation Quality	Type of Assay	RBE (±SE)			p-Values
		Normoxia (N)	Hypoxia (H)	Reoxygenation (R: H→N)	N vs. H	N vs. R	R vs. H
Carbon ions compared to X-rays	LP	2.27 ± 0.18	3.49 ± 0.46	3.13 ± 0.15	0.002	0.039	ns

LP: late plating; ns: not significant.

and Figure 2d). RBE was significantly greater under hypoxia and reoxygenation than under normoxia, though reoxygenation slightly reduced RBE compared to continuous hypoxia, without reaching statistical significance.

The Oxygen Enhancement Ratio (OER) calculated for X-rays was above 1 for both hypoxia and reoxygenation (OER = 1.34 and 1.35, respectively;

Table A3. Oxygen Enhancement Ratio (OER) based on D₁₀.

Radiation Quality	Type of Assay	OER (±SE)		p-Values
		Hypoxia (H)	Reoxygenation (R: H→N)	H vs. R
X-rays	Late plating	1.34 a ± 0.08	1.35 b ± 0.08	ns
Carbon ions	Late plating	0.96 ± 0.12	0.97 ± 0.07	ns

ns: not significant; ^a for X-rays vs. carbon ions exposure under hypoxia, p-value = 0.001; ^b for X-rays vs. carbon ions exposure after reoxygenation, p-value = 0.001.

and Figure 2e), indicating hypoxia-induced radioresistance. Reoxygenation after irradiation, however, did not significantly modify this effect. OER values for carbon ions were close to 1, indicating that the presence of oxygen did not alter cell killing by carbon ions.

2.2. Distinct G1 and G2 Phase Radiation Responses in Reoxygenated Compared to Continuously Hypoxic A549 Cells

Reoxygenated A549 cells did not show a significant decline in G1 population after X-ray exposure when compared to their respective unirradiated controls, unlike normoxic and continuously hypoxic cells (Figure 3 and Figure A1). In normoxic cells, the G1 fraction returned to baseline by 18 h post-irradiation, while in continuously hypoxic cells, it remained low even at 24 h.

After carbon ion irradiation, the G1 fraction decreased in all oxygenation conditions and remained suppressed at 24 h. However, in reoxygenated cells, the decline became significant only at 18 h post-irradiation—later than in normoxic and continuously hypoxic cells (12 h).

The G1 population recovered to control levels at 24 h in normoxic and reoxygenated cells (X-rays), but not in continuously hypoxic cells. No recovery was observed in any group after carbon ion exposure. Notably, the G1 decline was significantly greater in normoxic cells compared to hypoxic or reoxygenated cells (Figure 3a,d,g).

Irradiation led to an increase in G2 population across all oxygenation states (Figure 3c,f,i). After X-rays, the G2 fraction returned to baseline in normoxic and reoxygenated cells by 18 h but remained elevated in continuously hypoxic cells even at 24 h. After carbon ion exposure, G2 remained elevated at 24 h in all groups. The G2 increase was more pronounced in normoxic cells than in hypoxic or reoxygenated cells (Figure 3f,i).

2.3. Minimal Modulation of IL-6 and IL-8 Secretion by Reoxygenation in A549 Cells

Under normoxic conditions, irradiation did not significantly alter the secretion rate of IL-6, regardless of whether X-rays or carbon ions were used (Figure A2a,c). A similar pattern was observed in reoxygenated cells, whereas continuously hypoxic cells showed a transient increase in IL-6 secretion 6 h after X-ray exposure (Figure 4a), which was not sustained at 24 h (Figure 4c).

Notably, oxygenation status itself—independent of irradiation—had a stronger effect on IL-6 secretion. Both hypoxic and reoxygenated cells exhibited significantly higher IL-6 levels compared to normoxic cells at both 6 and 24 h, with or without radiation (Figure 4a,c).

In the case of IL-8, carbon ion exposure induced a marked increase in secretion at 6 h post-irradiation in hypoxic and reoxygenated cells (Figure 4b), an effect that was not observed at 24 h (Figure 4d and Figure A2b,d) and was absent under normoxia at all time points. X-rays, by contrast, did not significantly affect IL-8 secretion under any condition.

Interestingly, while IL-6 secretion was clearly influenced by oxygenation status, IL-8 secretion was more selectively responsive to carbon ion exposure under hypoxic and reoxygenated conditions, with no significant impact from oxygenation alone.

2.4. Differential Gene Expression in A549 Cells Under Different Oxygenation Conditions with and Without Irradiation

To investigate the biological relevance of differential gene expression under various experimental conditions, the results were stratified into seven comparison groups:

• Continuous hypoxia without irradiation (H0) vs. normoxia without irradiation (N0)
• Reoxygenation without irradiation (R0) vs. normoxia without irradiation (N0)
• Continuous hypoxia after irradiation with 8 Gy (H8) vs. normoxia after irradiation with 8 Gy (N8)
• Reoxygenation after irradiation with 8 Gy (R8) vs. normoxia after irradiation with 8 Gy (N8)
• Normoxia after irradiation with 8 Gy (N8) vs. normoxia without irradiation (N0)
• Hypoxia after irradiation with 8 Gy (H8) vs. hypoxia without irradiation (H0)
• Reoxygenation after irradiation with 8 Gy (R8) vs. reoxygenation without irradiation (R0)

The single most important finding from differential gene expression analysis was that hypoxia promoted an elaborate transcriptional program in A549 cells relative to normoxia, whereas reoxygenation largely reversed these effects. This is exemplified by volcano plots for all genes differentially regulated under hypoxia (Figure 5a) and after reoxygenation (Figure 5b).

2.4.1. Reoxygenation Reversed Hypoxia-Regulated Expression of Proliferation Genes in Irradiated and Unirradiated A549 Cells

Among the genes associated with the GSEA hallmark pathway for cell proliferation, continuous hypoxia (H0 vs. N0) led to the differential expression of 14 genes—7 upregulated and 7 downregulated compared to normoxic conditions (Figure 6a,b and

Table A4. Effect of oxygen conditions on expression of cellular proliferation genes in continuously hypoxic (H) A549 cells relative to normoxic (N) controls and in reoxygenated (R) A549 cells relative to normoxic (N) controls.

NCBI Gene Symbol	Log2 Fold Change		NCBI Gene Name	Log2 Fold Change
	H0 1 vs. N0	R0 vs. N0		H0 1 vs. N0	R0 vs. N0
BMPR2	1.12	0.16	IL1A	2.63	0.93
BNC1	1.15	0.31	IL1B	1.80	0.43
CD274	0.91	1.03	IL6	1.23	0.67
CHRM1	−3.02	−0.01	IL6R	−1.36	−0.36
CSF1	−1.08	−0.39	KITLG	−1.10	−0.19
CTF1	1.23	0.42	TNFSF9	−1.09	−0.10
EHF	−1.40	−0.01	VEGFA	1.00	0.19
FGF18	−1.05	−0.86

¹ H0 signifies unirradiated, continuous hypoxic (1 % O₂) A549 cells, N0 represents unirradiated normoxic cells, and R0 denotes unirradiated A549 cells that were reoxygenated after 48 h of hypoxia (1 % O₂). All cells were preincubated for 48 h in specified oxygen conditions, followed by a medium change. RNA isolation from cells was carried out 4 h after the medium change; ns: not significant; Statistically significant results are in bold; in red: upregulation, in blue: downregulation. The GSEA hallmark pathway for cell proliferation (M16210) was utilized to acquire proliferation-related genes.

). The upregulated genes included BMPR2, BNC1, CTF1, IL1A, IL1B, IL6, and VEGFA, many of which encode cytokines and growth factors implicated in promoting cell survival and inflammation under hypoxic stress. In contrast, CSF1, TNFSF9, KITLG, EHF, and CHRM1 were downregulated, potentially reflecting suppression of proliferative or differentiation signals under hypoxia. This differential transcriptional response was absent in reoxygenated cells (R0 vs. N0). However, reoxygenation of hypoxic cells enhanced expression of immune evasion promoting CD274 (PD-L1).

In irradiated A549 cells, hypoxia (H8 vs. N8, Figure 6c,d and

Table A5. Combined effect of oxygen concentration and irradiation on the regulation of proliferation-related genes in irradiated, continuously hypoxic A549 cells (H8) relative to irradiated normoxic A549 cells (N8), and in hypoxic A549 cells that were reoxygenated after irradiation (R8) relative to irradiated, normoxic cells (N8).

NCBI Gene Symbol	Log2 Fold Change
	H8 1 vs. N8		R8 vs. N8
	X-Rays	Carbon Ions	X-Rays	Carbon Ions
ADRA1D	−1.03	−1.19	−0.32	−0.78
ALOX12	−1.24	−0.30	−0.21	−0.06
BMPR2	1.25	1.15	0.04	0.29
BNC1	1.87	1.04	0.49	0.96
CD274	1.11	1.05	0.59	1.56
CHRM1	−5.36	−3.60	−0.99	−2.92
CTF1	1.81	1.09	1.49	−0.02
DLEC1	−1.19	0.10	−0.02	−0.25
ETS1	1.05	0.76	0.30	0.56
FABP6	0.13	1.17	−0.52	1.20
FIGF	−1.38	−0.79	−0.37	−0.23
IGFBP7	0.69	1.29	0.09	1.09
IL1A	2.81	3.46	1.13	2.40
IL1B	2.58	0.05	0.96	N/A
IL31RA	1.08	0.58	0.22	0.32
IL6	1.18	1.28	0.50	0.96
IL6R	−1.18	−1.05	−0.24	−0.51
LMO1	0.40	1.56	0.00	1.17
PTK2B	1.01	0.31	0.25	0.13
SPOCK1	1.05	1.06	0.30	0.81
SSTR5	−1.64	−1.36	0.04	−1.12
TNFSF9	−1.05	−0.95	−0.16	−0.58
VEGFA	1.09	1.04	0.10	−0.09

¹ H8, N8, and R8 signify continuously irradiated hypoxic (1 % O₂), normoxic, and reoxygenated A549 cells, respectively. All cells were preincubated for 48 h in specified oxygen conditions, followed by irradiation and a medium change. RNA isolation from cells was carried out 4 h after irradiation. Statistically significant results are in bold; in red: upregulation, in blue: downregulation. The GSEA hallmark pathway for cell proliferation (M16210) was utilized to acquire proliferation-related reference genes.

) induced significant differential expression of multiple genes involved in promotion of proliferation and inflammation. Eight genes—BMPR2, BNC1, CD274, CTF1, IL1A, IL6, SPOCK1, and VEGFA—were upregulated, while three—ADRA1D, IL6R, and SSTR5—were downregulated, independent of radiation quality. Reoxygenation largely reversed these transcriptional changes, though the extent varied with radiation type: IL1A remained upregulated after X-ray exposure, whereas CD274, FABP6, IGFBP7, IL1A, and LMO1 remained upregulated following carbon ion irradiation.

2.4.2. Reoxygenation Reversed Hypoxia-Regulated Expression of EMT Genes in Unirradiated and Irradiated A549 Cells

Compared to normoxic controls, continuous hypoxia (H0 vs. N0) resulted in differential upregulation of 16 EMT genes categorized in the GSEA hallmark pathway for EMT (Figure 7a,b and

Table A6. Effect of oxygen concentration on expression of EMT-related genes in continuously hypoxic (H) A549 cells relative to normoxic (N) controls and in reoxygenated (R) A549 cells relative to normoxic (N) controls.

NCBI Gene Symbol	Log2 Fold Change		NCBI Gene Name	Log2 Fold Change
	H0 1 vs. N0	R0 vs. N0		H0 1 vs. N0	R0 vs. N0
COL1A1	1.80	0.35	MYL9	1.18	0.67
COL5A1	1.43	0.54	PMEPA1	1.18	0.17
COL6A3	1.13	0.35	SERPINE1	1.64	0.43
IGFBP3	1.73	0.75	SNAI2	1.04	0.60
IL6	1.23	0.67	TAGLN	1.21	0.56
CXCL8	1.31	0.84	TGFBR3	−1.52	−0.32
INHBA	1.41	0.16	TIMP3	1.35	0.70
LAMC2	1.64	0.44	VEGFA	1.00	0.19
LRRC15	1.58	0.32

¹ H0 signifies unirradiated, continuous hypoxic (1 % O₂) A549 cells, N0 represents unirradiated normoxic cells, and R0 denotes unirradiated A549 cells that were reoxygenated after 48 h of hypoxia (1 % O₂). All cells were preincubated for 48 h in specified oxygen conditions, followed by a medium change. RNA isolation from cells was carried out 4 h after the medium change. Statistically significant results are in bold; in red: upregulation, in blue: downregulation. The GSEA hallmark pathway for epithelial–mesenchymal transition (M5930) was utilized to acquire EMT reference genes.

). These genes are known to be associated with extracellular matrix (ECM) remodeling, cytoskeletal dynamics, and modulation of cell adhesion to promote cell migration and invasiveness. Four hours of reoxygenation normalized their expression levels, making the transcriptomic EMT profile of reoxygenated cells resemble that of normoxic cells (R0 vs. N0) and reversing this effect.

In irradiated A549 cells, the comparison of hypoxic and normoxic conditions (H8 vs. N8) revealed upregulation of 16 EMT-related genes following X-ray exposure (Figure 7c,d and

Table A7. Combined effect of oxygen concentration and irradiation on the expression of EMT-related genes in irradiated, continuously hypoxic A549 cells (H8) relative to irradiated normoxic A549 cells (N8), and in hypoxic A549 cells that were reoxygenated after irradiation (R8) relative to irradiated, normoxic A549 cells (N8).

NCBI Gene Symbol	Log2 Fold Change
	H8 1 vs. N8		R8 vs. N8
	X-Rays	Carbon Ions	X-Rays	Carbon Ions
ABI3BP	0.41	1.21	0.11	1.09
CDH6	1.33	0.88	0.42	0.10
COL1A1	1.75	2.36	0.32	0.89
COL4A1	0.89	1.02	0.03	0.54
COL5A1	1.51	1.95	0.24	1.35
CCN2	1.30	1.18	-0.16	0.19
CTHRC1	0.11	1.37	-0.29	0.80
ENO2	0.56	1.08	0.01	0.51
FBLN2	0.40	1.07	-0.10	0.72
FSTL1	0.91	1.01	0.09	0.74
GADD45B	0.93	1.19	0.21	0.66
IGFBP3	1.36	2.82	0.32	1.95
IL32	0.72	1.40	0.12	1.45
IL6	1.18	1.28	0.50	0.96
CXCL8	1.49	1.41	0.65	1.27
INHBA	1.55	1.55	0.24	0.74
ITGA2	1.61	0.69	0.25	0.50
LAMC2	2.16	1.42	0.37	1.03
LOX	0.37	1.07	0.08	0.37
LOXL1	0.53	1.25	0.14	1.08
LOXL2	0.70	1.18	0.18	0.87
LRRC15	0.97	2.81	-0.08	1.55
MATN3	0.54	1.09	0.25	0.89
MYL9	0.73	1.87	0.12	1.59
PMEPA1	1.25	1.25	0.09	0.40
SCG2	−0.53	−0.99	−0.03	−1.07
SERPINE1	1.84	1.91	0.29	0.98
SNAI2	1.03	0.95	0.44	0.72
SPOCK1	1.05	1.06	0.30	0.81
TAGLN	1.15	2.02	0.04	1.71
TGFBI	0.88	1.36	0.02	0.81
TGFBR3	−1.19	−1.36	−0.17	−0.62
TPM1	0.44	1.22	−0.02	0.98
TIMP3	1.47	2.26	0.19	1.30
VEGFA	1.09	1.04	0.10	−0.09
WNT5A	−0.42	−1.10	0.06	−0.65

¹ H8, N8, and R8 signify continuously irradiated hypoxic (1 % O₂), normoxic, and reoxygenated A549 cells, respectively. All cells were preincubated for 48 h in specified oxygen conditions, followed by irradiation and a medium change. RNA isolation from cells was carried out 4 h after irradiation. Statistically significant results are in bold; in red: upregulation, in blue: downregulation. The GSEA hallmark pathway for epithelial–mesenchymal transition (M5930) was utilized to acquire EMT reference genes.

) and 30 genes following carbon ion exposure (Figure 7e,f and Table A7), with 13 genes commonly upregulated across both radiation qualities. Only one gene was downregulated after X-ray exposure, compared to two after carbon ion exposure. Reoxygenation (R8 vs. N8) completely reversed the hypoxia-induced EMT gene upregulation in X-ray–treated cells (Table A7); however, nine EMT-related genes remained upregulated in carbon ion–exposed cells despite reoxygenation.

2.4.3. Antiproliferative Signaling Post-Irradiation Is Observed Regardless of Oxygenation Status in A549 Cells

Radiation exposure increased transcription of ten genes associated with proliferation control in A549 cells after X-ray exposure when compared to unirradiated controls regardless of oxygenation status (Figure 8a and

Table A8. Effect of ionizing radiation on expression of proliferation-related genes by normoxic A549 cells (N), continuously hypoxic (1 % O₂) A549 cells (H), and hypoxic A549 cells reoxygenated after irradiation (R). Differential expression refers to irradiated (N8, H8, R8) versus unirradiated (N0, H0, R0) cells.

NCBI Gene Symbol	Log2 Fold Change
	N8 1 vs. N0		H8 vs. H0		R8 vs. R0
	X-Rays	Carbon Ions	X-Rays	Carbon Ions	X-Rays	Carbon Ions
BTG2	2.70	1.97	2.71	1.83	2.74	1.88
CDKN1A	2.63	2.61	2.62	2.38	2.56	2.50
IL1A	1.37	0.43	1.54	1.26	1.57	1.90
KITLG	1.01	1.36	1.38	1.46	1.06	1.27
MDM2	2.47	2.14	2.65	2.16	2.37	2.12
PGF	2.46	2.59	2.65	3.05	2.50	2.94
PLK1	−1.73	−1.59	−2.01	−1.61	−1.57	−1.65
PPM1D	2.03	1.47	1.90	1.23	1.91	1.25
SESN1	2.31	2.06	2.32	1.95	2.33	2.07
SYK	1.40	1.58	1.06	1.26	1.37	0.87
TNFSF9	1.53	1.39	1.57	1.52	1.47	0.91
SSTR5	1.38	1.62	0.64	1.16	1.64	0.72
BUB1	−0.95	−0.81	−1.04	−0.99	−0.79	−0.97
CD70	0.70	1.06	0.91	1.10	0.77	1.13
CDC25C	−1.03	−0.62	−1.03	−0.69	−0.77	−0.75
CDKN2C	−0.76	−0.98	−1.08	−0.66	−0.83	−0.88
DLEC1	0.11	0.47	−0.22	1.44	0.19	0.33
EBI3	0.60	1.04	0.81	1.45	0.16	1.02
ICOSLG	0.79	0.97	1.17	0.88	1.08	0.83
KIF2C	−1.27	−0.98	−1.23	−0.86	−1.16	−1.04
MKI67	−1.00	−0.68	−0.89	−0.71	−0.98	−0.72
MXD1	0.47	1.00	0.64	0.96	0.40	0.78
PDGFA	1.00	0.94	1.12	0.45	0.99	0.77
POU3F2	1.98	0.69	2.89	0.93	0.95	0.26
RACGAP1	−1.18	−0.93	−1.11	−0.86	−1.06	−0.82
SERTAD1	1.08	0.72	0.83	0.50	1.01	0.54
TPX2	−1.01	−0.87	−0.99	−0.87	−0.91	−0.78
TTK	−1.05	−0.74	−0.99	−0.74	−0.89	−0.65

¹ N8, H8 and R8 signify irradiated normoxic, continuously hypoxic (1 % O₂) and reoxygenated cells, respectively, while N0, H0 and R0 denote unirradiated normoxic cells, continuously hypoxic (1 % O₂) and reoxygenated cells A549 cells. All cells were preincubated for 48 h in specified oxygen conditions, followed by irradiation and a medium change. RNA isolation from cells was carried out 4 h after irradiation. Statistically significant results are in bold; in red: upregulation, in blue: downregulation. The GSEA hallmark pathway for cell proliferation (M16210) was utilized to acquire proliferation-related reference genes.

). The upregulated genes included CDKN1A (a cyclin-dependent kinase inhibitor that enforces cell cycle arrest), BTG2 and SESN1 (inhibitors of cell proliferation and stress response mediators), IL1A (a pro-inflammatory cytokine with antiproliferative effects), KITLG and PGF (growth factors involved in cell survival and angiogenesis), MDM2 (a p53-regulated negative feedback modulator), and PPM1D (a phosphatase involved in DNA damage recovery)—collectively reflecting activation of stress-adaptive and antiproliferative pathways. Additionally, PLK1, a key mitotic kinase, was downregulated in all conditions (Figure 8b and Table A8). This pattern suggested a net inhibitory effect on proliferation. The same trend was observed after carbon ion irradiation in normoxic and hypoxic cells (Figure 8c,d and Table A8); however, under reoxygenation following carbon ion exposure (R8 vs. R0), the transcriptional changes were not statistically significant.

2.4.4. Minimal Effect of Irradiation on EMT Regulation Regardless of Oxygenation Status in A549 Cells

Evaluation of differential expression of EMT-related genes in A549 cells showed that EMT was not a major component of A549 transcriptional response to irradiation under any oxygenation condition following X-rays or carbon ion exposure (Figure 9 and

Table A9. Effect of exposure to ionizing radiation on differential expression of EMT-related genes by normoxic A549 cells (N), continuously hypoxic (1 % O₂) A549 cells (H), and hypoxic A549 cells reoxygenated after irradiation(R). Differential expression is evaluated against respective unirradiated controls.

NCBI Gene Symbol	Log2 Fold Change
	N8 1 vs. N0		H8 vs. H0		R8 vs. R0
	X-Rays	Carbon Ions	X-Rays	Carbon Ions	X-Rays	Carbon Ions
ACTA2	0.98	1.55	1.74	2.23	0.99	2.17
FAS	2.06	2.31	2.75	2.85	2.16	2.64
GADD45A	1.11	0.77	0.95	0.55	1.02	0.91
PDGFRB	0.90	1.04	1.21	1.96	1.30	1.81

¹ N8, H8, and R8 signify irradiated normoxic, continuously hypoxic (1 % O₂), and reoxygenated cells, respectively, while N0, H0, and R0 denote unirradiated normoxic cells, continuously hypoxic (1 % O₂), and reoxygenated A549 cells. All cells were preincubated for 48 h in specified oxygen conditions, followed by irradiation and a medium change. RNA isolation from cells was carried out 4 h after irradiation. Statistically significant results are in bold; in red: upregulation. The GSEA hallmark pathway for epithelial–mesenchymal transition (M5930) was utilized to acquire EMT reference genes.

).

3. Discussion

In this study, we examined the effect of reoxygenation on radiosensitivity and radiation response of chronically hypoxic A549 cells following irradiation with either X-rays or carbon ions. Reoxygenation after irradiation did not significantly influence clonogenic survival, cell cycle arrest, or inflammatory cytokine secretion, but it reversed key hypoxia-induced transcriptional programs.

Reoxygenation prior to irradiation is a well-established strategy to enhance tumor radiosensitivity, primarily by increasing the proportion of oxygenated cells in the radiosensitive phase of the cycle and by facilitating fixation of DNA damage in the presence of molecular oxygen. Our work deliberately diverged from this paradigm to focus on reoxygenation after irradiation—a scenario that can occur clinically when cytotoxic effects of radiation reduce tumor cell density and metabolic demand, thereby improving oxygen availability in the residual tumor microenvironment.

Clinically, these results may imply that while interventions aimed at increasing oxygenation remain valuable before or during irradiation, such as hyperbaric oxygen therapy, carbogen breathing, or hypoxia-targeted agents, their benefit is likely diminished if implemented only after radiation exposure. In carbon-ion therapy, high-LET tracks intrinsically mitigate hypoxia-associated radioresistance; therefore, additional reoxygenation does not add meaningful benefit irrespective of its timing.

Although reversal of hypoxia-driven programs related to EMT, inflammation, proliferation, and glucose metabolism in our study did not translate into short-term changes in radiosensitivity to a single dose of X-rays or carbon ions, it may still shape long-term tumor behavior. EMT is linked to metastasis and to cancer-stem-cell–like phenotypes associated with treatment resistance and invasiveness; likewise, hypoxia-conditioned inflammatory and proliferative signals can foster angiogenesis/lymphangiogenesis and immune evasion [21]. This is why concurrent use of antiangiogenics like bevacizumab, antiproliferative agents such as Gefitinib and immune checkpoint inhibitors such as Pembrolizumab with radiotherapy has proven to be a promising therapeutic strategy in NSCLC [22]. However, we did not assess these outcomes; future work may test whether post-irradiation reoxygenation modulates metastasis, stemness, and subsequent therapy response in vivo.

3.1. Reoxygenation After Irradiation Has No Impact on Survival in A549 Cells

D₁₀ and OER values (Figure 2c,e) demonstrated that reoxygenation following X-ray exposure did not reverse hypoxia-induced radioresistance. Hypoxia-induced radioresistance to low-LET ionizing radiation like X-rays is well established [23] and has also been reported for NSCLC, including A549 cells [24]. Reoxygenation of hypoxic cells prior to irradiation can reverse radioresistance [25]. On the other hand, reoxygenation after irradiation has been reported to keep A549 cells radioresistant, particularly if treated concurrently with transforming growth factor β (TGF-β) [26].

On the other hand, D₁₀ and OER values demonstrated no statistically significant difference in cell survival following carbon ion irradiation across all oxygenation conditions. It is fairly well established that OER of high-LET ionizing radiation like carbon ions is low, which makes them a promising therapeutic approach to treat hypoxic tumors [27,28]. The efficacy of carbon ions in killing NSCLC cells, particularly A549, has also been previously reported [28,29]. However, we were unable to find any literature evaluating the effect of reoxygenation following irradiation on the survival of irradiated hypoxic cells, which may occur clinically in both acutely and chronically hypoxic cells during radiotherapy [30,31].

The high RBE of carbon ions across all oxygenation states (Figure 2d), combined with their capacity to overcome hypoxia-induced radioresistance irrespective of reoxygenation, underscores their potential as an effective treatment option for hypoxic NSCLC cells.

3.2. Reoxygenation After Irradiation Has No Impact on Cell Cycle Dynamics in A549 Cells

We show that reoxygenation post-irradiation had no significant impact on G2 arrest dynamics—whether attenuation or prolongation—in hypoxic A549 cells with either radiation quality (Figure 3). However, reoxygenated cells exhibited delayed or diminished G1-phase transition within the first 24 h compared with cells maintained under hypoxia or normoxia.

The slowing of the cell cycle under chronic hypoxia is well documented, with complete arrest occurring in many cell lines under anoxic conditions [32,33]. The attenuated G2/M block observed in both continuously hypoxic and reoxygenated cells in our study may reflect reduced proliferation rates limiting the number of cells reaching the G2/M checkpoint, rather than impaired checkpoint function. The smaller attenuation observed after carbon ion exposure could be due to their ability to induce more prolonged G2 arrest (Figure 3i vs. Figure 3f), allowing more time for cells to accumulate at the checkpoint. Our findings align with previous work in hTERT-immortalized retinal pigment epithelial (RPE-1) cells, where reoxygenation did not reverse hypoxia-induced quiescence and was suggested to contribute to radioresistance in both hypoxic and reoxygenated states [34].

The delayed G1 transition we observed in reoxygenated A549 cells may result from an additional slowing of the cell cycle following irradiation. While pre-irradiation reoxygenation is generally associated with enhanced cell cycle progression [33], the immediate post-irradiation impact remains poorly characterized. Notably, hypoxia-induced cell cycle inhibition may require up to 24 h to fully reverse in HeLa cells after reoxygenation, as shown using FUCCI analysis [13].

In summary, hypoxia-induced quiescence in A549 cells persisted after 24 h of reoxygenation, potentially contributing to the relative X-ray resistance of both hypoxic and reoxygenated cells by extending the available DNA repair time [35]. In contrast, this effect appears less relevant for carbon ions, likely due to their higher RBE [35,36]. This is consistent with the established principle that quiescent cells are more resistant to low-LET radiation, such as X-rays, because they predominantly reside in the radioresistant G1 phase, whereas high-LET radiation like carbon ions is less dependent on cell cycle phase [37].

3.3. Reoxygenation After Irradiation Has No Impact on IL-6 and IL-8 Secretion in A549 Cells

Reoxygenation after irradiation did not significantly alter IL-6 or IL-8 secretion in hypoxic A549 cells compared with continuously hypoxic cells, irrespective of radiation quality (Figure 4).

For IL-6, hypoxia increased secretion relative to normoxia, and this difference persisted at both 6 and 24 h after irradiation and reoxygenation (Figure 4a,c). IL-6 has been linked to enhanced cell survival and proliferation via EMT activation in both normal and cancerous cells, including NSCLC [38,39,40]. Elevated IL-6 secretion in both hypoxic and reoxygenated cells may therefore contribute to the increased radioresistance observed relative to normoxic controls. The selective enhancement of IL-6 secretion under hypoxia after X-ray exposure (Figure 4a) could further increase radioresistance in hypoxic cells in our X-ray survival experiments, an effect that was absent after carbon ion irradiation.

Carbon ion irradiation significantly increased IL-8 secretion in both hypoxic and reoxygenated cells. IL-8 secretion under hypoxia has been associated with tumor cell survival through EMT activation and immune evasion via PD-L1 upregulation [41,42]. Consistent with this, our RNA sequencing data showed increased CD274 expression (which encodes PD-L1) in irradiated hypoxic and reoxygenated cells compared with normoxic controls (Figure 6c,e). IL-8 is also induced under nutrient deprivation in A549 cells [43], suggesting that the elevated secretion observed here could promote survival under stress. However, despite higher IL-8 levels after carbon ion exposure in hypoxic or reoxygenated cells, our survival assays showed no corresponding advantage relative to normoxic counterparts.

3.4. Reoxygenation After Irradiation Reversed Hypoxia-Induced Transcriptional Changes

Gene expression analysis revealed that a 4 h reoxygenation period was sufficient to reverse hypoxia-induced transcriptional programs related to proliferation and EMT, with complete reversal after X-ray exposure but only partial reversal following carbon ion irradiation (Figure 7). These pathways were analyzed due to their established links to cell differentiation, cell cycle regulation, and the development of a radioresistant phenotype in hypoxic tumors.

While continuous hypoxia promoted pro-survival signaling and increased clonogenic survival after X-ray irradiation, the loss of this transcriptional profile upon reoxygenation after irradiation did not translate into reduced survival. This suggests that for radiosensitivity, the oxygen concentration and cellular status during irradiation were essential. The benefits of reoxygenation-induced transcriptional profile reversal might appear only with some time delay at a second irradiation, as exploited in clinical practice with fractionation of the total tumor dose.

For modalities such as carbon ion therapy, which inherently mitigate hypoxia-associated radioresistance through high-LET effects, this persistence is of less concern. The high RBE of carbon ions, combined with their capacity to bypass hypoxia-induced resistance irrespective of reoxygenation status, underscores their potential as an effective therapeutic option for hypoxic NSCLC tumors. Targeting such tumors with high-LET radiation may therefore circumvent the limitations of reoxygenation-based radiosensitization strategies and offer improved local control in clinically hypoxic disease.

3.4.1. Reoxygenation and Proliferative Signaling

Chronic hypoxia (H0 vs. N0) in A549 cells induced a transcriptional program enriched for pro-inflammatory cytokines and growth factors (IL1A, IL1B, IL6, CTF1, VEGFA, BMPR2, BNC1), many linked to survival, angiogenesis, and inflammatory signaling under hypoxic stress [44,45,46] (Figure 7a,b). Simultaneously, differentiation- and immune-activating signals such as CSF1, TNFSF9, KITLG, EHF, and CHRM1 were downregulated, potentially modulating tumor immune escape and proliferation in a context-dependent manner [47]. The overall pattern suggests that hypoxia promotes survival and immune evasion while reducing certain growth-inhibitory inputs. These changes were largely absent after reoxygenation, except for CD274 upregulation, consistent with enhanced immune evasion potential [48,49,50].

Irradiation (both X-rays and carbon ions) consistently upregulated cell cycle inhibitors and stress-response genes such as CDKN1A, BTG2, and SESN1 [51,52], alongside growth-promoting and survival-associated factors KITLG, PGF, MDM2, and PPM1D [53] (Figure 8). PLK1 was uniformly downregulated, indicating mitotic suppression. This mixed profile reflects concurrent anti-proliferative checkpoint activation and pro-survival signaling, which aligns with our clonogenic assay and cell cycle analysis findings of partial but not complete proliferation suppression post-irradiation.

In irradiated cells, hypoxia maintained elevated expression of multiple pro-survival and immune-evasive genes. including IL6, VEGFA, CD274, SPOCK1 [48,49,50,54,55,56,57] while suppressing genes encoding growth-inhibitory receptors ADRA1D and SSTR5 [58,59,60,61] (Figure 6). Such persistence of hypoxia-driven transcription may contribute to greater radioresistance with X-rays. The effects of carbon ions, in contrast, appeared less influenced by these hypoxia-linked gene programs, consistent with their ability to overcome hypoxia-induced resistance.

Reoxygenation largely reversed hypoxia-induced transcriptional patterns, particularly after X-rays, but key immune-evasive and survival-promoting genes (e.g., CD274, IL1A) remained elevated after carbon ion exposure. The clinical relevance of these differences has to be evaluated in further studies.

Overall, chronic hypoxia primes NSCLC cells via inflammatory, angiogenic, and immune-evasive transcriptional programs; radiation triggers both growth arrest and survival-promoting DDR signals; and reoxygenation after irradiation only partially mitigates hypoxia-induced resistance—least effectively after carbon ion irradiation.

3.4.2. Reoxygenation and EMT-Related Signaling

Chronic hypoxia (H0 vs. N0) in A549 cells induced a strong EMT-promoting transcriptional profile, with upregulation of 16 genes (Figure 7). Four hours of reoxygenation completely reversed this response, restoring expression to near-normoxic levels. This EMT-promoting signaling in hypoxia may contribute to radioresistance by sustaining migratory and proliferative advantages. Many hypoxia-upregulated genes, including COL1A1, COL5A1, COL6A3, LAMC2, and SERPINE1, are linked to enhanced migration, invasion, immune evasion, or treatment resistance in various cancers, including NSCLC [62,63,64,65,66]. Cytoskeletal regulators (MYL9, TAGLN) and transcription factors (SNAI2) promote EMT and cancer stemness [67,68,69], while PMEPA1, IL6, IL8, IGFBP3, INHBA, and VEGFA contribute to proliferation, angiogenesis, and immune modulation [70,71].

Following irradiation, EMT was not a dominant transcriptomic component overall (Figure 9); however, irradiated hypoxic cells (H8 vs. N8) exhibited notable EMT gene upregulation—16 genes after X-rays and 30 after carbon ions—with 13 genes common to both (Figure 7c and 7e). Reoxygenation fully reversed this after exposure to X-rays but only partially after irradiation with carbon ions, where nine EMT genes remained elevated, including IL32, LOXL1, LOXL2, MATN3, TGFBI, and WNT5A. These genes are associated with ECM remodeling, hypoxia/HIF-2 responsiveness, EMT induction, apoptosis inhibition, and radioresistance [72,73].

Overall, hypoxia primes A549 cells with a robust EMT transcriptional program that is largely reversible after reoxygenation in non-irradiated or X-ray–treated cells but is partially retained after carbon ion exposure. The persistence of EMT-related gene upregulation after carbon ions may contribute to sustained migratory, invasive, and treatment-resistant phenotypes despite restored oxygenation.

4. Materials and Methods

4.1. Cell Lines and Culture

The human lung adenocarcinoma cell line A549 (male, p53 wild type, KRAS-mutant) was procured from LGC Genomics (Berlin, Germany) [74]. Cells were maintained in 25 cm² or 80 cm² culture flasks (LABsolute, Th. Geyer GmbH, Renningen, Germany) using Alpha-Minimal Essential Medium (α-MEM; PAN Biotech, Aidenbach, Germany) supplemented with 10 % (v/v) dialyzed fetal bovine serum (FBS; PAN Biotech), 2 % (v/v) sterile glucose solution (0.94 mol/L), 1 % (v/v) penicillin (10,000 U/mL)/streptomycin (10 mg/mL) (PAN Biotech), 1 % (v/v) neomycin/bacitracin (Biochrom AG, Berlin, Germany), and 1 % (v/v) amphotericin B (250 µg/mL) (PAN Biotech). A seeding density of 5000 cells/cm² was selected to achieve approximately 30–40 % confluence after 48 h of incubation.

Routine mycoplasma screening was performed via PCR analysis of cell culture supernatants at the Leibniz-Institut DSMZ (Braunschweig, Germany), confirming absence of contamination. Cells were cultured at 37 °C in a humidified atmosphere, either under normoxic conditions (20 % O₂) using a CO₂ incubator (5 % CO₂; Heraeus HERAcell 150, Thermo Fisher Scientific, Karlsruhe, Germany) or under hypoxia (1 % O₂) in an InvivO₂ 400 hypoxia workstation (Baker Ruskinn, I&L Biosystems GmbH, Königswinter, Germany) flushed with 5 % CO₂, 1 % O₂, and 94 % N₂.

All procedures involving hypoxic cultures—including medium replacement, fixation, and cell lysis—were carried out within the hypoxia workstation. To ensure oxygen equilibration, media and reagents were degassed by pre-warming to 25 °C in a Sonorex Digiplus ultrasonic water bath (Bandelin, Berlin, Germany) operating at 35 kHz for 40 min, followed by equilibration inside the hypoxia chamber for an additional 40 min with loosened caps prior to use.

4.2. Irradiation

Following 48 h of incubation, A549 cultures were exposed to either X-rays or carbon ions (Figure 10). Prior to irradiation, culture flask caps were securely tightened. For hypoxic samples, flasks were placed inside airtight containers before being transferred out of the hypoxia workstation via the airlock. These containers were only opened briefly during the actual irradiation period. Immediately after exposure, flasks were returned to the airtight containers and transported promptly back to the hypoxia chamber.

To confirm that this handling process did not alter oxygen levels in the culture medium, dissolved oxygen measurements were performed in preliminary trials using a Seven2Go S9 dissolved oxygen meter (Mettler Toledo, Giessen, Germany). These measurements consistently verified that oxygen concentration remained stable during transport and irradiation.

X-ray irradiation (200 kV, 15 mA; LET: 0.3–3.0 keV/µm) was performed using an RS 225 X-ray cabinet (X-strahl, Ratingen, Germany) at the Institute of Aerospace Medicine, DLR, Germany. A dose rate of 1.0 Gy/min was achieved by positioning samples 450 mm from the X-ray source exit window. To remove low-energy photons, a 0.5 mm copper filter was installed at the exit port. Cells were irradiated in either 3 cm or 6 cm culture dishes or in 25 cm² or 80 cm² flasks, depending on the experimental setup. Dose rate and cumulative dose were continuously monitored during exposure using a TM30013 ionization chamber connected to a UNIDOS^webline dosimeter (PTW, Freiburg, Germany).

Carbon ion irradiation was conducted at the cyclotron facility of the Grand Accélérateur National d’Ions Lourds (GANIL), Caen, France, at a dose rate of 1 Gy/min. Cells were positioned in the plateau region of the Bragg curve to ensure a constant LET across the cell layer. A clinically relevant LET (~75 keV/µm) in water was obtained by reducing the carbon ion beam energy from 95 MeV/n to 35 MeV/n using a 16.9 mm-thick polymethyl methacrylate (PMMA) degrader, followed by further energy loss through the polystyrene base of the culture flask. This resulted in a final beam energy of 25.7 MeV/n and a calculated LET of 73 keV/µm in water. The residual range in water was 2550 µm, confirming irradiation within the plateau region. Heavy ion fluence (P/cm²) was used to calculate absorbed dose (Gy). Due to the horizontal beam geometry, flasks were kept upright during exposure and filled to the neck with culture medium to avoid cell desiccation.

After irradiation, the culture medium was replaced, and cells were incubated for experiment-specific time periods. Normoxic samples were maintained at 20 % O₂, while hypoxic samples were either kept at 1 % O₂ (continuous hypoxia) or switched to 20 % O₂ (reoxygenation).

4.3. Cell Survival Analysis Following Irradiation Under Normoxia and Hypoxia

Clonogenic survival was assessed using Puck’s colony-forming assay (CFA) to compare surviving fractions of A549 cells maintained under normoxia, continuous hypoxia, or reoxygenation (Figure 10) after receiving graded doses of X-rays (0, 0.5, 1, 2, 4, 6, and 8 Gy) or carbon ions (0, 0.5, 1, 2, and 4 Gy).

Cells were seeded in 25 cm² flasks and pre-incubated for 48 h under either normoxic (20 % O₂) or hypoxic (1 % O₂) conditions before irradiation (Section 4.2). Following irradiation or sham exposure, the culture medium was immediately replaced. Continuous hypoxia samples were returned to the hypoxia workstation, whereas normoxic samples were placed back in the standard incubator. For reoxygenation, a subset of hypoxic cells was transferred to normoxic conditions (20 % O₂).

Twenty-four hours post-irradiation, cells were trypsinized and replated in 6 cm Petri dishes (LABsolute, Th. Geyer GmbH, Renningen, Germany) for colony formation. The number of cells seeded was adjusted according to plating efficiency and the expected dose-dependent cell killing to yield approximately 75 colonies per dish. Normoxic and reoxygenated cultures were handled and incubated under standard conditions (20 % O₂) for two weeks. Hypoxic cultures were replated and grown within the hypoxia workstation at 1 % O₂ for the same duration. Colonies were fixed and stained for counting after the incubation period (Figure 10).

After two weeks, once the colonies became visible, they were fixed and stained with 5 mL of crystal violet (0.2 % w/v)–formaldehyde (3.5 %) staining solution per Petri dish for 20 min after removing culture medium from the Petri dishes. Stained colonies comprising over 50 cells were counted using a manual colony counter (Schuett count, Schuett-biotec, Göttingen, Germany). Survival fractions were determined by dividing the colony count by the number of cells that were seeded for each dose. Survival curves were generated for each oxygen condition and radiation quality by plotting the surviving fractions on a logarithmic scale as a function of dose on a linear scale. The single-hit multi-target model was used to perform regression analysis of experimental data. In addition, we derived D₁₀—the radiation dose that resulted in 10 % survival—from the modeled curves.

The Relative Biological Effectiveness (RBE) of carbon ions compared to X-rays under different oxygen conditions was calculated according to Equation (1):

$$\mathrm{R}\mathrm{B}\mathrm{E}={\displaystyle \frac{{\mathrm{D}}_{10 (\mathrm{X}-\mathrm{r}\mathrm{a}\mathrm{y}\mathrm{s})}}{{\mathrm{D}}_{10 (\mathrm{C}\mathrm{a}\mathrm{r}\mathrm{b}\mathrm{o}\mathrm{n} \mathrm{i}\mathrm{o}\mathrm{n}\mathrm{s})}}}$$

(1)

The influence of hypoxia and of reoxygenation after chronic hypoxia on cell survival was evaluated by calculating the Oxygen Enhancement Ratio (OER) using Equations (2) and (3), respectively:

$$\mathrm{H}\mathrm{y}\mathrm{p}\mathrm{o}\mathrm{x}\mathrm{i}\mathrm{a}:\mathrm{O}\mathrm{E}\mathrm{R}={\displaystyle \frac{{\mathrm{D}}_{10 (\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{u}\mathrm{o}\mathrm{u}\mathrm{s} \mathrm{H}\mathrm{y}\mathrm{p}\mathrm{o}\mathrm{x}\mathrm{i}\mathrm{a})}}{{\mathrm{D}}_{10 \left(\mathrm{N}\mathrm{o}\mathrm{r}\mathrm{m}\mathrm{o}\mathrm{x}\mathrm{i}\mathrm{a}\right)}}}$$

(2)

$$\mathrm{R}\mathrm{e}\mathrm{o}\mathrm{x}\mathrm{y}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}:\mathrm{O}\mathrm{E}\mathrm{R}={\displaystyle \frac{{\mathrm{D}}_{10 \left(\mathrm{R}\mathrm{e}\mathrm{o}\mathrm{x}\mathrm{y}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\right)}}{{\mathrm{D}}_{10 \left(\mathrm{N}\mathrm{o}\mathrm{r}\mathrm{m}\mathrm{o}\mathrm{x}\mathrm{i}\mathrm{a}\right)}}}$$

(3)

4.4. Analysis of Cell Cycle Response Following X-Ray Exposure Under Normoxia and Hypoxia

A549 cells were seeded in 6 cm Petri dishes at a density of 5000 cells/cm² and incubated for 48 h under either normoxic (20 % O₂) or hypoxic (1 % O₂) conditions. Cells were then exposed to either X-rays or carbon ions at a dose of 8 Gy. After irradiation, normoxic cultures were returned to 20 % O₂, while hypoxic cultures were either maintained at 1 % O₂ (continuous hypoxia) or shifted to 20 % O₂ (reoxygenation) (Figure 10).

At selected time points over the subsequent 24 h, cells were harvested by trypsin/EDTA detachment (1 mL) and fixed in 3.5 % formaldehyde. After 30 min, fixed cells were washed with phosphate-buffered saline (PBS) and stained with 4′,6-diamidino-2-phenylindole (DAPI; 500 ng/mL) containing Triton X-100 (3 µg/mL) in PBS. Samples were incubated for 30 min in the dark at room temperature.

DNA content was quantified using a Cytoflex S flow cytometer (Beckman Coulter, Indianapolis, IN, USA). DAPI was excited with a violet laser (405 nm), and fluorescence emission was collected in the PB450 channel. Cell cycle phase distribution for normoxic, continuously hypoxic, and reoxygenated samples was determined after applying gates on forward vs. side scatter and PB450 width vs. area plots to exclude debris and doublets. Histograms of single-cell fluorescence were analyzed using FloJo software (Version 10, BD Biosciences, San Jose, CA, USA) with the Dean–Jett–Fox mathematical model [75].

4.5. Quantification of Cytokine Secretion Following X-Ray Exposure Under Normoxia and Hypoxia

The concentrations of IL-6 and IL-8 in culture supernatants were determined using ELISA kits (Invitrogen, Thermo Fisher Scientific, Karlsruhe, Germany). Supernatants (3 mL) were collected at 6 h and 24 h after irradiation with either X-rays or carbon ions. Following collection, samples were placed in Eppendorf tubes and stored at −80 °C until analysis. Normoxic samples were re-incubated at 20 % O₂, whereas hypoxic samples were either maintained at 1 % O₂ (continuous hypoxia) or transferred to 20 % O₂ (reoxygenation) (Figure 10).

For normalization, A549 cell counts from each corresponding culture were obtained using the LUNA automated cell counter after detachment with 3 mL trypsin/EDTA.

Ninety-six–well ELISA plates (Corning™ Costar™ 9018, Kaiserslautern, Germany) were coated with primary capture antibodies (100 µL/well, 1:250 dilution in PBS) provided in the kit and incubated overnight at 4 °C. Wells were then blocked with 200 µL/well blocking buffer (1:5 dilution in deionized water) to prevent non-specific binding. Samples (100 µL/well) and serial dilutions of the supplied standard were added to the plates, followed by overnight incubation at 4 °C.

Detection antibodies (100 µL/well, 1:250 dilution in PBS) from the kit were then applied, followed by 1 h incubation at room temperature. For IL-6 detection, Streptavidin-HRP (100 µL/well, 1:100 dilution) was used; for IL-8 detection, Avidin-HRP (100 µL/well, 1:250 dilution) was applied. Plates were incubated for 30 min at room temperature before adding the substrate 3,3′,5,5′-Tetramethylbenzidine (TMB) (100 µL/well) for 15 min. The reaction was stopped with 100 µL/well of 2 N H₂SO₄.

All incubations were performed on a plate shaker, with five washes between steps using PBS containing 0.05 % Tween-20.

4.6. Gene Expression Analysis Following X-Ray Exposure Under Normoxia and Hypoxia

Global transcriptional profiles were assessed in A549 cells cultured under normoxia, continuous hypoxia, or reoxygenation following exposure to 8 Gy of X-rays or carbon ions (Figure 10). Four hours post-irradiation, the culture medium was completely removed, and cells were lysed in RLT buffer (Qiagen, Hilden, Germany) supplemented with β-mercaptoethanol (1:100; Sigma-Aldrich, St. Louis, MO, USA). Total RNA was extracted using the RNeasy Mini Kit (Qiagen) according to the manufacturer’s instructions.

RNA concentration and integrity were determined using the RNA 6000 Nano Assay (Agilent Technologies, Böblingen, Germany) on a Bioanalyzer (Agilent Technologies). Only samples with RNA Integrity Numbers (RIN) > 9.0 and yielding at least 3 µg total RNA (n = 4 biological replicates per condition) were processed further.

Samples were shipped on dry ice to GENEWIZ (Leipzig, Germany) for mRNA sequencing. Poly(A)-selected libraries were prepared and sequenced on the Illumina NovaSeq 6000 platform (paired-end, 2 × 150 bp; ~350 million read pairs per run). Reads were aligned to the Homo sapiens GRCh38 reference genome, and unique gene hit counts within exon regions were obtained. Differential expression analysis was performed in R using the DESeq2 package version 1.48.2 [76]. Genes were classified as differentially expressed if they met both criteria: adjusted p-value < 0.05 and absolute log₂ fold change > 1.

To investigate biological relevance, differential expression results were stratified into predefined comparison groups as shown in

Table 1. Comparison groups for evaluation of differential gene expression in A549 cells following modulation of oxygenation and irradiation status (X-rays/carbon ions).

Effect Being Studied	Abbreviation	Compared Groups
Effect of oxygenation status in unirradiated cells	H0 vs. N0	Continuous hypoxia without irradiation vs. normoxia without irradiation
	R0 vs. N0	Reoxygenation after 48 h of hypoxia without irradiation vs. normoxia without irradiation
Effect of oxygenation status in irradiated cells	H8 vs. N8	Hypoxia before and after irradiation with 8 Gy vs. normoxia before and after irradiation with 8 Gy
	R8 vs. N8	Reoxygenation after 48 h of hypoxia and irradiation with 8 Gy vs. normoxia before and after irradiation with 8 Gy
Effect of irradiation under different oxygen conditions	N8 vs. N0	Normoxia before and after irradiation with 8 Gy vs. normoxia without irradiation
	H8 vs. H0	Hypoxia before and after irradiation with 8 Gy vs. hypoxia without irradiation
	R8 vs. R0	Reoxygenation after 48 h of hypoxia and irradiation with 8 Gy vs. reoxygenation without irradiation

.

Within each comparison group, we focused on differential expression changes in curated gene panels associated with proliferation, epithelial–mesenchymal transition (EMT), and glycolysis. The EMT panel was defined using the GSEA hallmark gene set M5930 (200 genes), and proliferation using the GSEA hallmark pathway M16210 (512 genes).

4.7. Statistical Analysis

All experiments were performed at least three times (RNA Seq four times) at independent time points, with technical replicates varying by assay type: six for colony-forming assays, four for RNA sequencing, and three for cell cycle and cytokine analyses. Data processing, including calculation of means, standard deviations, and standard errors of the mean (SE), was carried out in Microsoft Excel 2016 (Microsoft Corporation, Redmond, WA, USA). Graph generation and statistical testing were performed using GraphPad Prism 9 (Dotmatics, Boston, MA, USA). For survival data obtained from colony-forming assays, multiple two-way unpaired t-tests were applied. Cell cycle and cytokine datasets were evaluated by two-way ANOVA. RNA sequencing data were analyzed using the Wald test to calculate p-values, with false discovery rate (FDR) adjustment via the Benjamini–Hochberg method to obtain adjusted p-values (padj).

5. Conclusions

Our findings reveal a nuanced interaction between chronic hypoxia, reoxygenation, and radiation quality in shaping the biological response of A549 NSCLC cells. High-LET carbon ions consistently overcame hypoxia-induced radioresistance, whereas X-rays retained reduced cytotoxicity under hypoxia that was not restored by immediate post-irradiation reoxygenation. Reoxygenation after irradiation did not enhance radiosensitivity, alter hypoxia-attenuated G2/M arrest, or modify inflammatory cytokine secretion, yet it substantially reversed hypoxia-driven transcriptional programs related to proliferation and epithelial–mesenchymal transition. The persistence of hypoxia-associated functional resistance despite transcriptional normalization suggests that it occurs too late restore radiosensitivity in chronically hypoxic tumors for an irradiation occurring just before reoxygenation, but the normalization might be beneficial in controlling invasive and metastatic behavior of the tumor cells and in reducing radioresistance when the next radiotherapy fraction is applied. Clinically, while oxygen-modifying strategies remain valuable when applied before or during irradiation, their benefit appears limited after radiation exposure. High-LET modalities such as carbon ion therapy may therefore provide a more robust approach for targeting hypoxic NSCLC subpopulations, independent of reoxygenation status.