Low-Dose Radiation Sensitizes Human Nasal Epithelium to Viral Inflammation despite Preserved Structural Integrity

Abstract

Introduction

Patients receiving radiotherapy for nasopharyngeal carcinoma often experience prolonged nasal epithelial dysfunction and recurrent infections. However, the dose-dependent effects of radiation on nasal epithelial integrity and antiviral immune function remain incompletely defined.

Methods

Human nasal epithelial cells (hNECs) were differentiated at an air-liquid interface (ALI) and exposed to gamma irradiation (0, 1, 2, 4, 8, or 16 Gy) on ALI day 0 to define dose-dependent epithelial injury. DNA damage, epithelial cell number, and proliferative capacity were assessed after irradiation. To model post-irradiation viral challenge, cultures exposed to 1 Gy were infected with rhinovirus on ALI day 28, followed by assessment of ciliary beat frequency, transepithelial electrical resistance (TEER), immunofluorescence, cytospin analysis, and gene expression by RT-qPCR.

Results

Increasing radiation doses progressively exacerbated epithelial injury. Higher doses (8–16 Gy) caused marked epithelial loss and were not pursued for detailed phenotypic analyses. At 4 Gy, hNECs exhibited impaired ciliogenesis, impaired barrier integrity, and reduced proliferative capacity, whereas 2 Gy primarily caused structural abnormalities characterized by reduced acetylated α-tubulin and MUC5AC signals and an increased cell aspect ratio. In contrast, 1 Gy induced transient DNA damage while largely preserving epithelial structure by day 28. Upon rhinovirus infection, 1 Gy-exposed hNECs exhibited enhanced antiviral and inflammatory responses, including increased type III interferons, interferon-stimulated genes, inflammatory mediators, and intercellular adhesion molecule-1 (ICAM-1) expression.

Conclusion

These findings suggest that low-dose radiation may alter epithelial immune responsiveness even in the absence of overt structural damage, leading to amplified antiviral and inflammatory gene induction following rhinovirus challenge. Our results provide novel insights into how prior radiation exposure may reshape mucosal responses to viral infection and thereby contribute to persistent sinonasal inflammation after radiotherapy.

Introduction

Radiotherapy is the main treatment for nasopharyngeal carcinoma (NPC), but it often causes side-effects in nearby normal tissues, especially the sinonasal mucosa [1, 2]. Because the nasal cavity is close to the nasopharynx, the nasal epithelium is unavoidably exposed to incidental radiation during treatment [1, 3–5]. In patients undergoing radiotherapy for NPC, a high radiation dose (66–70 Gy in 33–35 fractions) is delivered to the tumor site; meanwhile, the sinonasal mucosa is inadvertently exposed to approximately 30 Gy (about 1 Gy per fraction) [2, 4]. Many studies have shown that radiotherapy can seriously damage the nasal epithelium, resulting in long-term disruption of epithelial barrier function, impaired basal progenitor cell maintenance with reduced regenerative capacity, and impaired mucociliary clearance [1, 3, 6–9]. These alterations are major contributors to the development of chronic rhinosinusitis and persistent upper airway inflammation observed in many NPC patients, even years after completing radiotherapy [1, 4].

While the harmful effects of high-dose radiation are well-known, the biological effects of low-dose radiation (especially doses ≤1 Gy) on the nasal mucosa remain poorly understood [10, 11]. Studies in the lung and digestive systems suggest that low-dose radiation can affect cytokine production, stem cell behavior, and innate immune responses [12–15]. These findings indicate that low radiation doses might change local host immunity through nonlethal mechanisms [16, 17]. This raises the possibility that, even in the absence of overt structural damage, low-dose radiation may alter subsequent epithelial responses to environmental or infectious challenges. However, it is still unknown whether similar effects occur in the nasal epithelium.

Rhinovirus (RV) is the most common respiratory virus causing upper respiratory tract infections [18–21]. It primarily enters nasal epithelial cells through the intercellular adhesion molecule-1 (ICAM-1) receptor [22, 23]. Because ICAM-1 serves not only as a major receptor for many RV strains but also as an epithelial inflammatory adhesion molecule, it represents a biologically relevant readout for examining how radiation may alter epithelial responsiveness to viral challenge and downstream host responses [23]. In individuals with a compromised nasal epithelial barrier, such as those with asthma, chronic rhinosinusitis, or radiation-induced epithelial damage, RV infection often leads to more intense and prolonged inflammation [18, 24]. Our previous research demonstrated that fully differentiated human nasal cells produce sustained interferon-stimulated gene responses to RV infection through the RIG-I pathway, highlighting the critical role of the baseline immune status of the epithelium in antiviral defense [25–28].

In this study, we used an air-liquid interface (ALI) culture model of human nasal epithelial cells (hNECs) to investigate the effect of low-dose radiation (1 Gy) on nasal epithelial structure and antiviral immune responses. We hypothesized that while the epithelium may remain morphologically intact following low-dose radiation, its immune response to viral infection may become heightened or dysregulated, potentially leading to prolonged inflammation. Specifically, we examined whether prior low-dose radiation alters epithelial antiviral signaling, inflammatory mediator production, and ICAM-1 expression in response to RV infection. This study aimed to elucidate the mechanisms underlying persistent nasal inflammation observed in some NPC patients after radiotherapy and to inform future therapeutic strategies.

Methods

Culture of hNECs

Primary hNECs were derived from nasal epithelial stem/progenitor cells obtained from adult donors undergoing elective septoplasty at Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical University. Written informed consent was obtained from all participants prior to study initiation. The study protocol was approved by the Ethics Committee of Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical University (Approval Nos. KY-Z-2021-397-02 and KY2024-140-02). The nasal epithelial stem/progenitor cells were expanded and seeded onto collagen-coated transwell inserts (Corning) and cultured under ALI conditions for 28 days to induce mucociliary differentiation, forming a pseudostratified epithelium with basal, ciliated, and secretory cells. Culture medium (PneumaCult-ALI, STEMCELL Technologies) was changed every other day.

Irradiation Protocol

On ALI day 0, the apical surfaces were exposed to gamma irradiation using a ⁶⁰Co source (Gamma Chamber 4000A, Board of Radiation & Isotope Technology, India) at a dose rate of 0.56 Gy/min. Experimental groups were irradiated at single doses of 1, 2, 4, 8, or 16 Gy, while the control group was sham-irradiated under identical conditions outside the radiation chamber. Cultures were maintained until ALI day 28 for downstream assays [3].

RV Infection

At ALI day 28, differentiated hNECs were apically infected with human RV 16 (HRV16, ATCC VR-283) at a multiplicity of infection (MOI) of 2.5. After 1 h of incubation at 33°C, the virus-containing inoculum was removed, and cells were maintained under ALI conditions. Samples were collected at 24 h post-infection for subsequent experiments.

Cytospin Preparation and Immunofluorescence Staining

For cytospin analysis, hNECs were dissociated using TrypLE Express (Gibco) and fixed with 4% paraformaldehyde. Cell suspensions (1–2 × 10⁴ cells/slide) were cytocentrifuged at 500 rpm for 5 min (Shandon Cytospin 4, Thermo Fisher Scientific). For transwell-based immunofluorescence, cultures were fixed and permeabilized with 0.1% Triton X-100, blocked with 10% goat serum, and incubated overnight with primary antibodies targeting γ-H2AX, ZO-1, p63, KRT5, FOXJ1, Ki67, MUC5AC, ICAM-1, F-actin, and acetylated α-tubulin (Ac-α-tubulin). The primary antibodies and dilution ratios used for immunofluorescence staining are summarized in Table 1. Alexa Fluor-conjugated secondary antibodies were applied the following day, and nuclei were counterstained with DAPI. Imaging was performed using a standard fluorescence microscope (Olympus IX51), and 3 fields per sample were captured for quantification.

Table 1.

Primary antibodies and dilution ratios used for immunofluorescence staining

Antibody	Host	Type	IF dilution	Catalog No.	Company
γ-H2AX	Mouse	Monoclonal	1:300	JBW301	Millipore
ZO-1	Rabbit	Polyclonal	1:400	HPA001636	Sigma
FOXJ1	Rabbit	Polyclonal	1:300	HPA005714	Sigma
MUC5AC	Rabbit	Polyclonal	1:500	sc-20118	Santa Cruz
Ki67	Rabbit	Monoclonal	1:400	ab16667	Abcam
KRT5	Rabbit	Monoclonal	1:600	ab52635	Abcam
p63	Mouse	Monoclonal	1:50	ab735	Abcam
acetylated α-tubulin	Mouse	Monoclonal	1:2,000	ab24610	Abcam
F-actin	Mouse	Monoclonal	1:400	ab205	Abcam
ICAM-1	Rabbit	Monoclonal	1:200	ab109361	Abcam

Transepithelial Electrical Resistance

Barrier integrity of ALI cultures was measured using an EVOM2 voltohmmeter (World Precision Instruments). Transepithelial electrical resistance (TEER) values (Ω·cm²) were calculated by subtracting the background resistance of blank inserts and multiplying by the surface area of the transwell.

Ciliary Beat Frequency

Ciliary beat frequency was recorded using the Sisson-Ammons Video Analysis (SAVA) system under an inverted phase-contrast microscope (×200 magnification). After washing with 1× PBS at 37°C, ciliary motion was monitored at 23 ± 0.5°C. Frequencies were derived from three distinct fields per sample and averaged across a 3-min recording period.

RNA Extraction and Quantitative Real-Time PCR

Total RNA was extracted from cells using the mirVana miRNA Isolation Kit (Thermo Fisher Scientific) according to the manufacturer’s protocol. For cDNA synthesis, 1 µg RNA was reverse-transcribed using qScript cDNA SuperMix (QuantaBio). Quantitative real-time PCR (RT-qPCR) was performed using SYBR Green assays on a CFX Connect system (Bio-Rad), and gene expression was normalized to RPL13A housekeeping gene using the 2^−ΔΔCt method. The primer sequences used for RT-qPCR are listed in Table 2. Viral RNA levels were measured by RT-qPCR using RV-specific primers. Ct values were converted to (40 − Ct), where 40 represents the maximum cycle number.

Table 2.

Primer sequences used for RT-qPCR

Target gene	Forward (5′-3′)	Reverse (5′-3′)
RPL13A	GTC​TGA​AGC​CTA​CAA​GAA​AAG	TGT​CAA​TTT​TCT​CTC​CAC​G
KRT5	TGG​ACA​CTT​CAA​GAC​AAA​G	ATG​TAG​GCA​GCA​TCT​CAC​TC
TP63	CAG​CCT​ATT​TGT​AGT​TCC​AG	CAG​TCC​TAC​TTC​ACT​CAA​TC
MKI67	GAC​AGA​GGT​TCT​CAA​GGA​GAG	AAC​AAC​TCA​GAT​TTG​CCT​GC
IFNA1	ATC​TGG​TCC​AAC​ATG​AAA​AC	GGG​TGA​GAG​TCT​TTG​AAA​TG
IFNB1	ATT​CTA​ACT​GCA​ACT​CTT​GC	GAG​ACT​TGT​AGA​AAG​GAA​GG
IFNG	TCG​GTA​ACT​GAC​TTG​AAT​GTC​CA	TCG​CTT​CCC​TGT​TTT​AGC​TGC
IFNL1	GTT​GAG​TCA​CTG​GAA​GAG​AG	AAC​CTG​CAG​TTT​TTC​AGC​TG
IFNL2	ACA​TAG​CCA​GTG​TAG​AGA​GC	GAC​TCT​TCT​GAA​GGC​TTT​GG
IFNL3	TAA​GAG​GGC​CAA​AGA​TGC​CTT	CTG​GTC​CAA​GAC​ATC​CCC​C
DDX58 (RIG-I)	GGT​ATA​GGT​TAC​AGG​CAT​TTT​C	TTG​TTT​ACA​TGT​GTT​TGG​GC
MX1	ATG​AGG​TGG​CAG​CAA​TCA​G	CAG​GCT​TTG​ATC​TCT​GCT​GC
IFIH1 (MDA5)	GAT​TAG​GTG​TGA​TAC​CCA​AC	GTC​TGA​CAA​TTG​AAC​ACC​AGC
ISG15	AGA​TAC​CCA​GAG​GAG​ATG​C	TGT​TTA​TTC​CCT​GCC​AGT​GC
OAS1	GGA​GGT​TGC​AGT​GCC​AAC​GAA​G	TGG​AAG​GGA​GGC​AGG​GCA​TAA​C
IRF7	TCTTCTTCCAAGAGCTGG	CTA​TCC​AGG​GAA​GAC​ACA​C
ICAM1	ACCATCTACAGCTTTCCG	TCACACTTCACTGTCACC
RV-S1	GCACTCTGTTTCCC	CGGACACCCAAAGTAG
IFNA2	CCT​GAT​GAA​GGA​GGA​CTC​CAT	AAA​AAG​GTG​AGC​TGG​CAT​ACG
IFIT1	TTG​ATG​ACG​ATG​AAA​TGC​CTG​A	CAG​GTC​ACC​AGA​CTC​CTC​AC
IRF3	ACA​TCT​CCA​ACA​GCC​AGC​CTA​T	AGT​CCA​TGT​CCT​CCA​CCA​AGT​C
IL8	CAC​CGG​AAG​GAA​CCA​TCT​CA	AGAGCCACGGCCAGCTT
TSLP	CCC​TTC​ACT​CCC​CGA​CAA​AA	CCT​GAG​TAC​CGT​CAT​TTC​TCT​CA
CXCL10	TCT​GAA​TCC​AGA​ATC​GAA​GG	CTC​TGT​GTG​GTC​CAT​CCT​TG
IL6	GTAGCCGCCCCACACAGA	CAT​GTC​TCC​TTT​CTC​AGG​GCT​G

Statistical Analysis

All statistical analyses were performed using GraphPad Prism 10.1.2. Statistical comparisons were based on pairwise analyses. Normality was assessed prior to statistical testing. For normally distributed data, paired or unpaired Student’s t tests were used as appropriate, and Welch’s correction was applied when variances were unequal. For non-normally distributed data, Wilcoxon matched-pairs signed-rank tests or Mann-Whitney U tests were used where appropriate. When multiple pairwise comparisons were performed, p values were adjusted for multiple testing. p < 0.05 was considered statistically significant.

Results

Dose-Dependent Irradiation Induces DNA Damage, while Higher Doses Impair Epithelial Expansion and Proliferation

Figure 1 shows the experimental design and the dose-dependent effects of gamma radiation on hNECs in ALI culture. As illustrated in Figure 1a, hNECs were exposed to different doses of gamma radiation and then maintained in culture until day 28. Representative images at ALI day 28 showed progressively aggravated morphological disruption with increasing radiation dose, particularly in the higher-dose groups (Fig. 1b). Consistently, quantification of epithelial cell number at day 28 revealed a dose-dependent decline, with a significant reduction observed from 2 Gy onward (Fig. 1c). Higher doses (8–16 Gy) caused marked epithelial loss and were therefore not included in subsequent detailed phenotypic analyses. For γ-H2AX analysis, samples were collected at 1, 2, and 4 h after irradiation on the day of exposure, whereas all other analyses were performed using cultures harvested on ALI day 28. Radiation-induced DNA damage was assessed by examining γ-H2AX foci at these early time-points. In the 1 Gy group, γ-H2AX foci peaked at 1 h and largely resolved by 4 h, indicating a transient DNA damage response. In contrast, the 2 Gy and 4 Gy groups exhibited more abundant and persistent γ-H2AX foci, consistent with delayed resolution of radiation-induced DNA damage at higher doses (Fig. 1d, e). We next evaluated proliferative capacity at day 28 by Ki67 immunostaining and MKI67 mRNA expression. The proportion of Ki67-positive cells was largely preserved at 1 Gy and 2 Gy, but was significantly reduced at 4 Gy (Fig. 1f, g). Similarly, MKI67 mRNA expression was maintained at 1 Gy and 2 Gy, with a downward trend observed at 4 Gy (Fig. 1h). Together, these findings indicate that 1 Gy irradiation induces a transient DNA damage response without markedly compromising proliferative capacity, whereas higher doses progressively impair epithelial growth and proliferation.

Fig. 1.

Dose-dependent effects of gamma radiation on epithelial morphology, cell number, DNA damage, and proliferation in hNECs. a Experimental scheme: hNECs were differentiated at an ALI, exposed to gamma radiation (0, 1, 2, 4, 8, and 16 Gy) on ALI day 0, and analyzed as indicated. b Bright-field images showing epithelial morphology at ALI day 28 after irradiation (n = 3). c Cell counts at day 28 across doses. Data are mean ± SD (n = 5). p values are indicated in the figure. d Immunofluorescence staining of γ-H2AX (red) with nuclear counterstain (DAPI, blue) at 1, 2, and 4 h after irradiation with 0, 1, 2, or 4 Gy (representative images). e Quantification of γ-H2AX excess foci per cell after 1, 2, and 4 Gy. Data are mean ± SD (n = 5). p values are indicated in the figure. f Immunofluorescence staining of Ki67 (red) with nuclear counterstain (DAPI, blue) at day 28 after irradiation with 0, 1, 2, or 4 Gy (representative images). g Quantification of Ki67-positive cells at day 28. Data are mean ± SD (n = 4). p values are indicated in the figure. h RT-qPCR analysis of MKI67 mRNA expression at day 28, normalized to RPL13A. Data are mean ± SD (n = 3). p values are indicated in the figure.

Higher Radiation Doses Reduce Basal Progenitor Cell Markers and Impair Mucociliary Differentiation

To investigate the impact of gamma radiation on basal progenitor cells in the nasal epithelium, we assessed the expression of KRT5 and p63 in hNECs at day 28 after exposure to 0, 1, 2, or 4 Gy. Immunofluorescence staining (Fig. 2a) showed reduced KRT5 and p63 signals after irradiation, with the most marked decrease observed in the 4 Gy group. Quantitative analysis further demonstrated a significant reduction in the proportion of p63⁺KRT5⁺ basal cells at 4 Gy, whereas no significant change was observed at 1 or 2 Gy (Fig. 2b). The RT-qPCR analysis showed that KRT5 mRNA levels were not significantly altered across groups, whereas TP63 expression showed a downward trend at 4 Gy but did not reach statistical significance (Fig. 2c, d). These findings indicate that higher radiation doses compromise the maintenance of basal progenitor cell markers in hNECs.

Fig. 2.

Gamma radiation reduces basal cell markers at day 28. a Cytospin immunofluorescence staining for KRT5 (red), p63 (green), and nuclei (DAPI, blue) at day 28 after 0, 1, 2, or 4 Gy radiation. b Quantification of p63⁺KRT5⁺ basal cells. Data are mean ± SD (n = 4). p values are indicated in the figure. c, d RT-qPCR analysis of KRT5 (c) and TP63 (d) mRNA levels normalized to RPL13A. Data are mean ± SD (n = 3). p values are indicated in the figure.

We next examined whether gamma irradiation affected epithelial differentiation at day 28 post-irradiation. To assess ciliated cell differentiation, we examined the expression of FOXJ1, a transcription factor critical for motile ciliogenesis, together with Ac-α-tubulin, a structural marker of cilia. Immunofluorescence staining showed fewer FOXJ1-positive cells with increasing radiation dose (Fig. 3a), and quantification revealed reduced proportions of FOXJ1⁺ cells at 2 Gy and 4 Gy, with statistical significance observed only at 4 Gy (Fig. 3b). To evaluate functional consequences, we measured ciliary beat frequency (CBF). Ciliary function was preserved at 1 and 2 Gy but significantly impaired at 4 Gy (Fig. 3c). Additionally, we assessed differentiation of secretory and ciliated cells by immunofluorescence staining for MUC5AC (goblet cells) and Ac-α-tubulin (ciliated cells). Representative images in Figure 3d showed reduced MUC5AC and Ac-α-tubulin staining at higher radiation doses. Quantification of the stained area (Fig. 3e, f) revealed a significant decrease in both MUC5AC-positive area and Ac-α-tubulin-positive area at 2 Gy and 4 Gy, whereas no significant change was observed at 1 Gy.

Fig. 3.

Gamma radiation impairs ciliated and secretory cell differentiation and ciliary function. a Immunofluorescence staining for FOXJ1 (red), Ac-α-tubulin (green), and nuclei (DAPI, blue) at day 28 after 0, 1, 2, or 4 Gy radiation. b Quantification of FOXJ1⁺ cells. Data are mean ± SD (n = 4). p values are indicated in the figure. c CBF measurement and quantification. Data are mean ± SD (n = 4). p values are indicated in the figure. d Immunofluorescence staining for MUC5AC (red), Ac-α-tubulin (green), and nuclei (DAPI, blue) at day 28 after 0, 1, 2, or 4 Gy radiation. e, f Quantification of MUC5AC⁺ and Ac-α-tubulin⁺ areas at day 28 after irradiation. Data are mean ± SD (n = 4). p values are indicated in the figure.

Together, these data indicate that gamma radiation at 4 Gy compromises basal progenitor cell marker maintenance, whereas radiation at 2 Gy and 4 Gy disrupts mucociliary differentiation, with overt ciliary functional impairment observed at 4 Gy. By contrast, the effects of 1 Gy are limited under these conditions.

Higher Radiation Doses Disrupt Epithelial Barrier Integrity and Alter Cell Morphology

Figure 4 illustrates the dose-dependent effects of gamma radiation on epithelial barrier function and cell morphology at day 28 post-irradiation. Tight junction organization and cytoskeletal architecture were examined by immunofluorescence staining for ZO-1 and F-actin, respectively.

Fig. 4.

Gamma radiation disrupts cell junctional organization, alters cellular morphology, and impairs epithelial barrier function at higher doses. a Immunofluorescence staining of ZO-1 (red) and F-actin (green) and merged images at day 28 after 0, 1, 2, or 4 Gy radiation. b Cell shape analysis quantified by aspect ratio (higher values indicate elongated morphology). Data are mean ± SD (n = 5). p values are indicated in the figure. c TEER at day 28. Data are mean ± SD (n = 5). p values are indicated in the figure.

As shown in Figure 4a, ZO-1 staining appeared continuous and well-defined in the control and 1 Gy groups, whereas junctional irregularity became apparent at 2 Gy and was more pronounced at 4 Gy. Similarly, F-actin organization was progressively altered with increasing dose, accompanied by cytoskeletal disorganization and elongated cell morphology at higher doses. Quantitative analysis of cell aspect ratio (Fig. 4b) showed no significant change at 1 Gy compared with the control but significant increases at both 2 Gy and 4 Gy, indicating increased cell elongation following higher-dose irradiation.

We further evaluated epithelial barrier integrity using TEER measurements. As shown in Figure 4c, TEER values decreased in a dose-dependent manner, with significant reductions at 2 Gy and 4 Gy, but not at 1 Gy, compared with the control. Collectively, these results demonstrate that while 1 Gy irradiation has a limited impact, higher radiation doses progressively impair epithelial junctional organization, disrupt barrier function, and alter cell morphology.

Prior 1 Gy Gamma Irradiation Modulates Selected Type III Interferon-Associated and Antiviral Gene Responses to RV Infection

Figure 5 illustrates the experimental setup and transcriptional responses of various interferons and immune-related genes following RV infection with or without prior exposure to 1 Gy gamma radiation. As shown in Figure 5a, ALI cultures were irradiated at day 0 and infected with RV on day 28. Gene expression levels were assessed at 24 h post-infection.

Fig. 5.

Gamma radiation with 1 Gy enhances type III interferon-associated and antiviral gene responses to RV infection. a Experimental scheme: hNECs were irradiated (0 or 1 Gy) on ALI day 0, infected with RV on day 28, and analyzed at 24 h post-infection (hpi). b–d RT-qPCR analysis of IFNL1 (b), IFNL2 (c), and IFNL3 (d) mRNA expression. Data are mean ± SD (n = 3). p values are indicated in the figure. e–h RT-qPCR analysis of DDX58 (RIG-I) (e), IFIH1 (MDA5) (f), IFIT1 (g), and MX1 (h) mRNA expression. Data are mean ± SD (n = 3). p values are indicated in the figure. i–l RT-qPCR analysis of inflammatory mediators (IL8, TSLP, CXCL10, and IL6) mRNA expression. Data are mean ± SD (n = 3). p values are indicated in the figure.

As shown in online supplementary Figure S1 (for all online suppl. material, see https://doi.org/10.1159/000552427), the expression of type I interferons (IFNA1, IFNA2, and IFNB1) and IFNG did not differ significantly among groups. In contrast, IFNL1 and IFNL2 expression was significantly increased in the IR1+RV group compared with both the individual IR1 and RV groups (Fig. 5b, c). IFNL3 expression reached statistical significance only in the RV group (Fig. 5d). These results indicate that 1 Gy radiation is associated with a selective enhancement of RV-induced-type III interferon responses, particularly IFNL1 and IFNL2.

We further examined key interferon-stimulated genes and antiviral mediators. As shown in Figure 5e–h, DDX58, IFIH1, IFIT1, and MX1 were induced by RV infection. Compared with the IR1 group, the IR1+RV group showed significantly higher expression of DDX58, IFIH1, IFIT1, and MX1. Among these, IFIT1 and MX1 were also significantly higher in the IR1+RV group than in the RV group, whereas DDX58 and IFIH1 did not differ significantly between these two groups. In online supplementary Figure S2, ISG15 and OAS1 were upregulated in response to RV infection, but prior 1 Gy radiation did not further enhance their expression compared with the RV group. By contrast, IRF3 and IRF7 showed no significant differences among groups. Together, these findings indicate that prior 1 Gy radiation selectively enhances RV-induced antiviral gene responses, particularly IFIT1 and MX1.

We also analyzed mRNA expression of pro-inflammatory cytokines. As shown in Figure 5j and l, TSLP and IL6 were significantly upregulated in the IR1+RV group compared with the IR1 group, but not compared with the RV group, with IL8 and CXCL10 showing similar trends (Fig. 5i, k). Overall, these findings indicate that 1 Gy irradiation modulates the epithelial transcriptional response to RV infection by selectively enhancing type III interferon-associated, antiviral, and inflammatory gene responses.

Prior 1 Gy Irradiation Increases ICAM-1 Expression and Is Associated with a Trend toward Higher Viral RNA Levels following RV Infection

Figure 6 shows ICAM-1 expression and viral RNA levels following RV infection with or without prior 1 Gy irradiation. Representative immunofluorescence images (Fig. 6a) showed the strongest ICAM-1 staining in the IR1+RV group. Quantification of mean fluorescence intensity (Fig. 6b) confirmed a stepwise increase in ICAM-1 signal across the MOCK, IR1, RV, and IR1+RV groups, with the IR1+RV group significantly higher than the other three groups, and the RV group higher than the MOCK and IR1 groups. The RT-qPCR analysis (Fig. 6c) showed a similar upward trend in ICAM1 mRNA expression, with the highest level in the IR1+RV group; however, statistical significance was observed only when the IR1+RV group was compared with the MOCK group. Viral RNA levels (Fig. 6d) were significantly increased in both the RV and IR1+RV groups compared with the MOCK and IR1 groups, whereas viral RNA levels in the IR1+RV group showed a relatively higher trend compared with the RV group. Taken together, these results indicate that prior 1 Gy irradiation enhances ICAM-1 expression in response to RV infection, while viral RNA levels also show an increasing trend.

Fig. 6.

Prior 1 Gy radiation increases ICAM-1 expression and is associated with a trend toward higher viral RNA levels following RV infection. a Representative immunofluorescence images of ICAM-1 (red) with nuclear counterstain (DAPI, blue) in the MOCK, IR1 (1 Gy), RV, and IR1+RV groups. b Quantification of mean ICAM-1 fluorescence intensity in the four groups (mean ± SD; n = 3). p values are indicated in the figure. c RT-qPCR quantification of ICAM1 mRNA expression normalized to RPL13A (mean ± SD; n = 3). p values are indicated in the figure. d RT-qPCR analysis of viral RNA levels, expressed as (40 − Ct) values (mean ± SD; n = 3). p values are indicated in the figure. Control denotes the non-irradiated (0 Gy) group in non-infected experiments. MOCK denotes the mock-infected control in infection experiments. Gene expression was normalized to RPL13A. Data are shown as mean ± SD (n = 3). p values are indicated in the figure. IR1, 1 Gy irradiation alone; RV, RV infection alone; IR1+RV, 1 Gy irradiation followed by RV infection.

Discussion

This study systematically evaluated the dose-dependent effects of gamma radiation on hNECs, with a particular focus on epithelial regeneration, mucociliary differentiation, barrier integrity, and antiviral immune responsiveness. Using a well-established ALI model, we found that hNECs exhibited a clear dose-dependent response to radiation. Exposure to 1 Gy induced transient DNA damage with limited long-term structural disruption, whereas 2–4 Gy resulted in persistent DNA damage accompanied by impaired basal cell maintenance, defective mucociliary differentiation, and reduced epithelial barrier integrity. Importantly, the most notable finding of this study is that low-dose radiation is not biologically inert. Even when the overall epithelial architecture was largely preserved, prior exposure to 1 Gy still altered the subsequent response to RV infection, and amplified antiviral and inflammatory signaling.

The structural abnormalities observed at higher doses were consistent with defective recovery from radiation-induced DNA damage [3, 29–31]. The γ-H2AX analysis showed that DNA damage induced by 1 Gy appeared early and resolved rapidly, suggesting relatively efficient repair. In contrast, 2 Gy and 4 Gy led to persistent γ-H2AX foci with delayed resolution, indicating sustained DNA damage and impaired repair capacity. These findings were consistent with the subsequent phenotypic defects observed in the higher-dose groups, including reduced proliferative capacity, loss of basal progenitor cell markers, impaired ciliogenesis, and disruption of epithelial barrier function. Taken together, these data support the interpretation that higher-dose radiation directly compromises epithelial maintenance and regeneration, thereby resulting in more pronounced structural and functional injury.

In contrast, 1 Gy irradiation produced a distinct phenotype that is better characterized as epithelial “priming” rather than overt epithelial injury. Although 1 Gy did not cause persistent structural disruption by day 28, pre-irradiated cultures exhibited enhanced responses to subsequent RV challenge, including increased expression of selected type III interferons, antiviral genes, inflammatory mediators, and ICAM-1. These findings suggest that prior low-dose radiation may sensitize the epithelium to subsequent viral challenge, thereby amplifying antiviral and inflammatory responses.

Mechanistically, this primed state is more likely to reflect selective remodeling of epithelial antiviral signaling rather than broad activation of all interferon pathways. Type I interferons were not significantly altered, whereas IFNL1 and IFNL2 were preferentially enhanced, suggesting a selective augmentation of type III interferon-associated mucosal antiviral responses. Given that type III interferon responses are primarily epithelial-associated, this pattern is consistent with an epithelial-intrinsic response to prior radiation-induced stress [32, 33]. In parallel, ICAM-1 was included in this study as a relevant readout because it occupies a key position in the interaction between RV and the host epithelium [28, 34, 35]. ICAM-1 is not only an important entry receptor for many RV strains and therefore highly relevant in the nasal epithelium, a primary target tissue of RV infection, but also an important adhesion molecule reflecting epithelial activation and mucosal inflammatory responses, with roles in leukocyte recruitment and local inflammatory amplification [36–41]. Compared with the virus-only and irradiation-only groups, ICAM-1 expression was most strongly increased in the IR1+RV group, suggesting that low-dose radiation altered the manner in which the epithelium responded to subsequent viral challenge. Taken together, the selective enhancement of type III interferon-associated responses, the upregulation of ICAM-1, and the increased induction of downstream antiviral and inflammatory genes suggest that low-dose radiation can alter epithelial immune response patterns even in the absence of overt tissue injury.

From a clinical perspective, these findings may have important implications. In patients receiving radiotherapy for nasopharyngeal or sinonasal malignancies, the sinonasal mucosa may still receive incidental low-dose radiation even with modern conformal techniques, and repeated fractionated exposure may cumulatively impair epithelial repair and host defense [1, 4, 10]. Our data suggest two distinct consequences: higher doses may directly cause structural epithelial injury, whereas lower incidental doses may leave the epithelium largely morphologically preserved while functionally sensitizing it to subsequent viral infection/exposure. This “radiation plus viral infection” model may help explain why some patients experience persistent sinonasal inflammation or delayed mucosal recovery after radiotherapy [1, 42]. Notably, the relevance of low-dose radiation is not limited to oncology. Low-dose radiotherapy has been used or explored for selected non-malignant inflammatory or degenerative conditions, and during the COVID-19 pandemic, whole-lung low-dose radiation was also investigated as an immunomodulatory strategy in non-cancer patients, although clinical results were inconsistent [43–45]. Although these studies did not directly address the sinonasal mucosa or upper respiratory viral infection, they support the broader biological plausibility that low-dose ionizing radiation can modulate local inflammatory and immune responses outside cancer therapy [43–45]. Accordingly, our findings raise a hypothesis that warrants further investigation: in selected non-cancer patients with prior mucosal radiation, subsequent common respiratory viral infections may trigger stronger or more prolonged mucosal inflammatory responses. This possibility merits further study and may have implications for infection prevention, early recognition of viral exacerbations, and supportive management of mucosal inflammation in such patients.

Several limitations of this study should be acknowledged. First, although the ALI model is physiologically relevant, it cannot fully recapitulate the complexity of the in vivo nasal mucosal environment, including immune cell interactions, the microbiota, and vascular factors [46]. Second, our analysis was limited to 28 days after irradiation and therefore could not address longer-term remodeling, fibrosis, or recovery trajectories. Third, although we assessed representative markers of differentiation, proliferation, barrier function, and antiviral responsiveness, a more comprehensive characterization of radiation-induced epithelial reprogramming will require systematic transcriptomic and/or proteomic analyses. Fourth, our assessment of viral burden was limited to measurement of viral RNA levels rather than direct infectivity or replication assays [47]. Finally, we did not examine potential interventions that might mitigate these effects, such as antiviral strategies, anti-inflammatory therapy, or approaches aimed at preserving epithelial repair capacity.

Acknowledgment

We thank the Department of Physics, National University of Singapore for technical assistance and for providing access to the gamma irradiator.

Statement of Ethics

The study protocol was approved by the Ethics Committee of Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical University (Approval Nos. KY-Z-2021-397-02 and KY2024-140-02). Written informed consent was obtained from all participants prior to sample collection.

Conflict of Interest Statement

De-Yun Wang was a member of the journal’s Editorial Board at the time of submission. The remaining authors declare no conflicts of interest.

Funding Sources

This research was funded by the National Natural Science Foundation of China (Nos. 82371113 and 82171104, to Qianhui Qiu), the China Postdoctoral Science Foundation (No. 2024M750594, to Suizi Zhou), and the National Medical Research Council, Singapore (No. NMRC/CIRG/1458/2016, to De-Yun Wang).

Author Contributions

Jing Liu: writing – review and editing. De-Yun Wang and Qianhui Qiu: conceptualization and supervision. Yitong Liu, Qianmin Chen, and Ce-Belle Chen: investigation and methodology. Yitong Liu, Suizi Zhou, Yueying Yang, Hsiao-Hui Ong, Keshuang Wang, Zhiqing He, Chubo Xie, Yuanming Wang, Shuping Jiang, and Yang Peng: formal analysis, writing – original draft, and writing – review and editing. Vincent T. Chow and Mark Thong: resources, clinical sample collection, and writing – review and editing. All authors have read and approved the final version of the manuscript.

Funding Statement

This research was funded by the National Natural Science Foundation of China (Nos. 82371113 and 82171104, to Qianhui Qiu), the China Postdoctoral Science Foundation (No. 2024M750594, to Suizi Zhou), and the National Medical Research Council, Singapore (No. NMRC/CIRG/1458/2016, to De-Yun Wang).

Data Availability Statement

All data generated or analyzed during this study are included in this published article and its online supplementary materials. Further details are available from the corresponding authors upon reasonable request.