A nitroreductase responsive probe for early diagnosis of pulmonary fibrosis disease

Idiopathic pulmonary fibrosis (IPF) is a serious interstitial lung disease. However, the definitive diagnosis of IPF is impeded by the limited capabilities of current diagnostic methods, which may fail to capture the optimal timing for treatment. The main goal of this study is to determine the feasibility of a nitroreductase (NTR) responsive probe, 18F-NCRP, for early detection and deterioration monitoring of IPF. 18F-NCRP was obtained with high radiochemical purity (>95 %). BLM-injured mice were established by intratracheal instillation with bleomycin (BLM) and characterized through histological analysis. Longitudinal PET/CT imaging, biodistribution study and in vitro autoradiography were performed. The correlations between the uptake of 18F-NCRP and mean lung density (tested by CT), as well as histopathological characteristics were analyzed. In PET imaging study, 18F-NCRP exhibited promising efficacy in monitoring the progression of IPF, which was earlier than CT. The ratio of uptake in BLM-injured lung to control lung increased from 1.4-fold on D15 to 2.2-fold on D22. Biodistribution data showed a significant lung uptake of 18F-NCRP in BLM-injured mice. There was a strong positive correlation between the 18F-NCRP uptake in the BLM-injured lungs and the histopathological characteristics. Given that, 18F-NCRP PET imaging of NTR, a promising biomarker for investigating the underlying pathogenic mechanism of IPF, is attainable as well as desirable, which might lay the foundation for establishing an NTR-targeted imaging evaluation system of IPF.


Introduction
Idiopathic pulmonary fibrosis (IPF) is classified as the most severe form of idiopathic interstitial pneumonias, associated with a median survival rate of 2.5-3.5 years [1].Diagnosis as early as possible can allow patients to receive timely treatment and improve the decline in forced vital capacity [2,3].High-resolution computed tomography (HRCT) plays a principal role in clinical setting to assess IPF, and histological examination can be omitted when the key CT features were found [4].Regrettably, the features like honeycombing, reticular pattern, and traction bronchiectasis in lung images obtained by HRCT usually indicate an advanced stage of the disease [5,6].According to clinical trials reported, there is a significant rate of missed diagnosis and misdiagnosis, with evidence suggesting it to be as high as 30%-50 % [7].In cases where imaging results are atypical, lung biopsy is employed.However, some patients experience a poor prognosis, with high morbidity and mortality following the biopsy procedure [8].Consequently, there is an urgent need to develop specific biomarkers and highly sensitive, non-invasive imaging techniques for early IPF diagnosis, disease progression monitoring, and prediction of disease behavior, which should help facilitate personalized treatment strategies for patients.
Positron emission tomography (PET) imaging is a real-time, noninvasive, and quantitative approach that has potential to accurately measure the severity of lung diseases and identify foci of metabolic activity, thereby improving diagnostic accuracy [9][10][11].To date, as shown in Table 1, numerous radiotracers for IPF-targeted PET imaging have been reported, and several of them exhibit promising potential for diagnosis of IPF [12][13][14][15][16][17][18][19][20][21].The development of IPF-targeted radiotracers illustrates the necessity and urgency, and provides inspiration for the innovation of new biomarkers and probes.
Recent evidences indicated the involvement of redox imbalance in the progression of IPF.In a state of health, cells diligently maintain a delicate balance between oxidants and antioxidants.Redox imbalance refers to the disruption of this dynamic equilibrium within the organism.Overproduction of oxidants leads to oxidative harm of cellular biomolecules, triggering oxidative stress, while insufficient oxidant levels disrupt essential signaling pathways, causing a state of reductive stress.This reductive stress can lead to an increase in reductase enzymes, such as nitroreductase (NTR) [22,23].Several studies have shown that oxidative stress plays a crucial role in the pathogenesis of IPF [24].This redox imbalance can activate the body's antioxidant system, and the stress-sensitive transcription factors such as nuclear factor erythroid 2-related factor (NRF2), nuclear factor kappa B (nf-kappa B) will be activated to produce NAD(P)H to reduce oxidative modification [25].However, in the lungs of IPF, a hypoxic environment [26] inhibits the oxidation of NADH, leading to its accumulation and promoting the redox imbalance towards a state of reductive stress [27].Furthermore, hypoxia can also lead to the occurrence of reductive stress by triggering glycogen synthesis in hypoxic tissue regions.Specifically, in hypoxic area, NADPH was produced through the pentose phosphate pathway (PPP), and the synthesis of reduced glutathione was promoted, which eventually leads to the occurrence of reducing stress [28][29][30][31].In addition, complete nitroreduction by oxygen-sensitive NTR must also occur under hypoxic conditions, and this conversion catalyzes the nitroreduction of nitroaromatic compounds by the addition of a single electron using NAD(P)H as an electron donor, producing transient nitroanion radicals, and ultimately ends up with a primary amine.Under hypoxia, the reduction process converts the electron-withdrawing nitro group into the electron-donating amino group, leading to a significant change in electron density.This change can serve as a selective "switch" mechanism to activate inert compounds.Consequently, the unique characteristics of NTR, namely demonstrating the existence of IPF involving stress reduction, make it a promising biomarker for the visualization of IPF.
Despite decades of studies on IPF, the exact pathogenesis mechanism remains elusive.Several previous reports have shown a possible association between redox imbalance and IPF, with special attention to the important role of reductive stress.These studies have sparked our interest in exploring the link between NTR, a target associated with reductive stress, and IPF, while developing visualization methods to assist in the diagnosis and treatment of IPF.To achieve the objectives, we established an IPF murine model induced by bleomycin (BLM), and then performed longitudinal PET/CT imaging to demonstrate the value of early diagnosis of IPF.We developed a nitrogen mustard analogue, the NTR responsive probe 18 F-NCRP, as an imaging tool to assess disease activity.Moreover, this study conducted a series of in vitro and in vivo experiments to verify the correlation between NTR PET imaging and IPF disease progression, which might lay the foundation for the establishment of NTR-targeted imaging evaluation system for IPF.

General
All materials and reagents used in this study were purchased from commercial suppliers.[ 18 F]fluoride ion was produced by cyclone cyclotron (IBA, Belgium) in the Center for Molecular Imaging and Translational Medicine of Xiamen University.The hydroxyproline (HYP) assay kit (Acmec) and picro-sirius red stain kit were purchased from Shanghai Acmec Biochemical Co., Ltd.DAPI (Beyotime, P0131) was purchased from Shanghai Beyotime Biotechnology Co., Ltd.HIF-1α (Santa Cruz, sc-13515) was purchased from Xiamen Xincheng Biotechnology Co., Ltd.High performance liquid chromatography (HPLC) analysis was performed on Dionex Ultimate 3000 HPLC (Thermo Scientific, USA) equipped with an Elysia Raytest Gabi Star γ-radiation detector.Imaging studies were performed using an Inveon microPET/CT scanner (Siemens Medical Solutions Inc., USA).Autoradiography images were acquired by a storage phosphor imager (Cyclone Plus, PerkinElmer Instruments Inc., USA).Biodistribution data were measured by γ-counter (PerkinElmer, USA).HYP result was determined using a microplate reader (Thermo Scientific, USA).Tissue immunofluorescence was performed by Zeiss LSM 880+Airyscan microscope (Carl Zeiss, Germany).

Chemistry and radiochemistry
18 F-NCRP was synthesized according to previously published methods [32].Briefly, 1.85 GBq [ 18 F]fluoride ion was eluted from the preactivated QMA Carb cartridge (Waters light Sep-Pak, pretreated with 10 ml 0.5 M K 2 CO 3 and 10 ml H 2 O) with 1 ml of eluate (900 μl CH 3 CN and 100 μl H 2 O containing 11 mg Kryptofix 222 and 2 mg K 2 CO 3 ) into the vial.Then, it was evaporated three times with anhydrous CH 3 CN under 110 • C to remove water.Next, 2 mg of the precursor dissolved in 1 ml anhydrous DMSO was added into the [ 18 F]fluoride solution and then heated to 72 • C for 20 min.The crude product was diluted with 10 ml H 2 O and purified with a preactivated C18 (10 ml CH 3 OH and 10 ml H 2 O) and radio-HPLC (70 % CH 3 OH: 30 % H 2 O, 3 ml/min, UV = 254). 18F-NCRP was re-dissolved in saline with 5 % ethanol for further research.Radio-HPLC (70 % CH 3 OH: 30 % H 2 O, 1 ml/min, UV = 254) was used to determine the radiochemical purity (RCY).
The mixture was incubated at 37.5 • C for 4 h under the atmosphere of nitrogen.Samples were analyzed by radio-HPLC (70 % CH 3 OH: 30 % H 2 O, 1 ml/min).

Animal model
All experimental procedures and the use of animals were carried out in compliance with the guidelines of the Animal Care and Use Committee of the Laboratory Animal Center of Xiamen University.BLM (Beyotime, Shanghai, China) was used to induce pulmonary fibrosis in mice.10-week-old male C57BL/6J mice were given a single intratracheal injection of 2.5 mg/kg of BLM (dissolved in 50 μl 0.9 % normal saline, BLM-injured group) or 50 μl saline on day 0 (D0).PET imaging studies were conducted on day 8 (D8), day 15 (D15) and day 22 (D22).

Histochemical identification
Hydroxyproline content: A commercial kit was used to measure the levels of hydroxyproline content in the lung.In brief, lung tissues were digested with 6M HCl at 100 • C, followed by centrifugation, pH adjustment, volume determination, and color development.The microplate reader was adjusted to 560 nm for sample determination.
Histological study: Lung tissue specimens were fixed with 4 % paraformaldehyde overnight at 4-8 • C, then embedded with paraffin.The specimens were cut into 8-μm slices and then processed through xylene and graded ethanol (100 %, 95 %, 80 %, 75 % and 60 %) before being placed in double-distilled water.They were then stained with hematoxylin and eosin (H&E) or Sirius red, following the kit's instructions.Histologic images were acquired by the Leica DM4 B upright digital research microscopes (Leica).
Immunofluorescence: Lung tissues were frozen in optimal cutting temperature compound (O.C.T), stored at − 80 • C. The specimens were cut into 8-μm slices by the freezing microtome at − 20 • C. Tissue sections were fixed using 4 % paraformaldehyde for 10 min and washed three times in phosphate buffer saline (PBS).Nonspecific antibody binding was blocked by incubating the sections in 10 % goat serum for 30 min and washed in PBS three times.Subsequently, the tissue sections were incubated at 4 • C overnight with specific antibody of HIF-1α (mouse monoclonal, 1 : 50), and then incubated with Alexa fluor 488 anti-mouse IgG antibody as the secondary antibody (1 : 200) for 1 h.Cell nucleus was stained blue by antifade mounting medium (DAPI included).All the images were analyzed by ImageJ 7.0 software.

PET/CT imaging and autoradiography analysis
7.2 MBq of 18 F-NCRP (200 μl) was intravenously injected into mice (n = 5).All the mice underwent 5-min static PET scans and 10-min micro-CT scans at 30 and 60 min post injection (p.i.).During the scanning, mice were subjected to gas anesthesia to maintain spontaneous breathing.To monitor the uptake in lung during various stage, longitudinal PET/CT imaging was performed weekly from D8 to D22.Three mice were selected on D22 and dicumarol (0.4 mM, 50 μl) was intratracheally instilled 1 h before 18 F-NCRP injection to clear the NTR expression.The data were reconstructed using three-dimensional ordered-subset expectation-maximization (3D OSEM) algorithm.Regions of interest (ROI) were drawn on the decay-corrected whole body coronal images.All the mice were euthanized after PET/CT imaging on D22, and tissues were collected and exposed to obtain autoradiography images.

Biodistribution
Biodistribution of 18 F-NCRP was performed on D22.Each mouse was injected with 0.7 MBq (100 μl) of 18 F-NCRP via tail vein (n = 4).The BLM-injured mice were euthanized by decapitation at 30, 60, and 90 min p.i. Interested organs were collected, weighed, and radioactivity was counted using a γ-counter.The percentage of the injected dose per gram (%ID/g) of organs and tissues was calculated.The saline group served as the control.

Safety evaluation
The safety profile of 18 F-NCRP was explored by animal experiments.Mice were randomly divided into four groups: 37 MBq of 18 F-NCRP and same volume of saline for 3 male or 3 female mice per group, respectively.Weight was continuously monitored for 7 days.After that, the main organs of mice were collected for H&E staining to observe morphological changes, and the serum of each group was collected for blood tests to determine the liver and kidney function, including aspartate aminotransferase (AST), alanine aminotransferase (ALT), blood urea nitrogen (BUN) and creatinine (CR).

Statistical analysis
All data represent at least 4 independent mice, and quantitative data are expressed as the mean ± standard deviation (SD).Differences among groups were compared by a two-tailed Student's t-test.The Pearson correlation coefficient was computed between two quantitative variables (*p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001; ns, no significant difference).

NTR-specificity
18 F-NCRP was co-incubated with NTR in a hypoxic environment to verify responsiveness in vitro.In theory, the 18 F-NCRP could react with NTR, reducing the proportion of the 18 F-NCRP and simultaneously generating new reduction products.As expected, 68.96 % of 18 F-NCRP was reduced in the presence of NTR and NADPH, while only 16.71 % reduction was achieved with the introduction of dicumarol.In addition, as the concentration of 18 F-NCRP decreased, the fraction that underwent reduction also decreased.These results demonstrate that 18 F-NCRP specifically respond to NTR in vitro (Fig. 1).

Correlation analysis
Based on the above results, we did a correlation analysis between the BLM-injured lung uptake of 18 F-NCRP and lung mean density, HYP content, Sirius red positive area, HIF-1α positive area, Hypoxyprobe™-1 positive area and NTR content.As expected, a good correlation was observed between the 18 F-NCRP uptake of BLM-injured lung and CT data (r = 0.67, p < 0.0001) (Fig. 6A).Similarly, significant correlations were revealed between the radioactivity uptake and the HYP content (r = 0.85, p < 0.0001), and the quantitative data of Sirius red (r = 0.93, p < 0.0001) (Fig. 6B and C).Furthermore, results indicated a noteworthy positive association between the lung uptake of 18 F-NCRP and the expression levels of HIF-1α (r = 0.83, p < 0.0001), Hypoxyprobe-1 (r = 0.93, p < 0.0001), and NTR (r = 0.81, p < 0.01) (Fig. 6D-6F).

Safety of 18 F-NCRP
The curves of body weight changes in different groups of mice after administration of 18 F-NCRP (37 MBq) are shown in Fig. S9.Both the female and male mice injected with saline or 18 F-NCRP showed similar trends in body weight changes and all mice survived throughout the seven-day observation period.In addition, the results of blood routine tests showed that the indexes of liver and kidney function in all mice were within the normal range (Tables S2-S5), indicating that 18 F-NCRP did not cause damage to liver and kidney function.H&E staining of major organs showed no organic lesions in each group of mice (Fig. S10), indicating the lack of evident toxicological effects associated with 18 F-NCRP.

Discussion
The exact pathogenic mechanism of IPF remains elusive, with academic opinions diverging due to the complexity of signaling pathways and involvement of numerous cell types [33,34].As the most prevalent form of interstitial lung disease, IPF exhibits an incidence rate ranging from 4.6 to 17 cases per 100,000 people per year [35,36].Acute exacerbations occur in approximately 10 %-20 % of patients annually [37].The appearance of honeycombing, a key diagnostic feature on HRCT, typically signifies the end-stage of lung fibrosis.However, HRCT can be challenging when patients also have concurrent emphysema, potentially leading to delays in treatment initiation [38].Therefore, there is an urgent need to develop more sensitive and non-invasive approaches for the early detection of IPF.Furthermore, the identification of a sensitive and specific biomarker to characterize the disease endotype is of paramount importance.
In this study, the BLM-injured mice were characterized through histological analysis (Fig. 1).Consistent with previous research, we observed destruction of the alveolar structure and deposition of extracellular matrix, accompanied by a rise of HYP content from D8 to D22 [39].Compared with saline group, conspicuous red collagen fiber bundles were evident in the lung of BLM-injured mice.While much research on IPF has centered on redox imbalance, particularly focusing on signaling pathways involving NADPH oxidases, eosinophil peroxidase, mitochondrial electron transport chain, and myeloperoxidase, which can be activated to generate ROS [40,41], our study based on the expression of hypoxia signal in the IPF model, boldly demonstrated that the dynamic changes of redox imbalance, especially reductive stress, during the evolution of IPF disease could be monitored by PET imaging.HIF-1α emerges as the primary factor in the context of hypoxia, and our findings indicate that its expression increases with the severity of fibrosis, underscoring the presence of a hypoxic environment in the fibrotic lung.This observation is consistent with the research conducted by Tzouvelekis A et al. [42].Furthermore, we validated our hypothesis using 18 F-NCRP, which has been shown to specifically respond to NTR in this study.Our findings underscore the expression of NTR in the BLM-induced lung tissues (Fig. 5), which positively correlated with the lung uptake of 18 F-NCRP (Fig. 6).
As illustrated in Fig. 7, the primary objective of this study was to assess the potential of 18 F-NCRP for early detection and dynamic monitoring of the IPF progression.In NTR-rich lesions of lung fibrosis, the phenyl nitro group on 18 F-NCRP undergoes specific reduction to an electron-donating amino group, activating the nitrogen mustard group to cross-link with DNA, leading to the retention of radioactivity in the IPF region.Conversely, the selective "switch" mechanism remains "OFF" in normal organs.Consequently, PET imaging with 18 F-NCRP demonstrated promising results in the early detection of IPF.As shown in Fig. 2 and Fig. S6, PET imaging with 18 F-NCRP can detect IPF lesions earlier than CT, potentially extending the time window and allow for the implementation of more assertive therapeutic interventions.Moreover, the ratio of uptake in BLM-injured lung to control lung increased from 1.4-fold on D15 to 2.2-fold on D22, indicating that the 18 F-NCRP PET allows for continuous monitoring of IPF advancement.Importantly, a strong positive correlation was observed between the uptake of 18 F-NCRP in BLM-injured lungs and histopathological characteristics.This correlation underscores the potential utility of 18 F-NCRP PET imaging in evaluating the pathogenesis of IPF.
The ex vivo findings corroborated the imaging outcomes.In biodistribution (Fig. 4 and Table S1), the uptake of 18 F-NCRP in BLMinjured lung was 2.37 times higher than that in the control group at 30 min p.i. Notably, compared to the saline-treated group, there was a significant increase in the uptake of 18 F-NCRP in the heart and blood.One plausible explanation for this observation could be the development of chronic hypoxic pulmonary hypertension as a complication in the BLM-injured model [43,44].Chronic hypoxia-induced pulmonary vasoconstriction is a known mechanism underlying pulmonary hypertension development [45,46].Therefore, it is reasonable to infer an association between NTR expression and the manifestation of reductive stress in chronic hypoxic pulmonary hypertension.Consequently, 18 F-NCRP selectively accumulates in the hypoxic area where NTR is expressed.These findings underscore the specificity of 18 F-NCRP as a promising tool for the diagnosis and monitoring the progression of IPF, thus opening new avenues for exploring the pathogenic mechanism of IPF through the study of reductive stress.Moreover, it suggests the feasibility of targeting NTR for IPF PET imaging. 18F-NCRP, functioning as a radiotracer, offers real-time and noninvasive visualization through PET, enabling the detection of metabolic activity in lung foci in a gentle and prompt manner.However, a limitation of this study is that the fibrosis induced by BLM may not fully replicate the irreversible and complex course observed in IPF patients, and the underlying pathogenesis may differ slightly.In future investigations, more appropriate modeling methods will be explored to better understand the mechanisms of IPF in animal models.Concurrently, various animal modeling techniques will be explored to validate the efficacy of 18 F-NCRP.Furthermore, forthcoming research efforts will delve into exploring the correlation between 18 F-NCRP PET imaging and the therapeutic responses to drugs like nintedanib or pirfenidone, aiming to provide more precise guidance for personalized treatment [47].
In summary, 18 F-NCRP demonstrates promising performance in IPF detection.PET imaging reveals a notable lung uptake of 18 F-NCRP during the progressive stage of IPF.Compared to the traditional CT imaging, 18 F-NCRP PET imaging can detect IPF at an earlier stage.Moreover, the uptake of 18 F-NCRP in BLM-injured lungs exhibits a positive correlation with histopathological characteristics.Hence, 18 F-NCRP PET imaging of NTR, a promising biomarker for investigating the underlying pathogenic mechanism of IPF, is both achievable and desirable.

S
. Peng et al.

Fig. 7 .
Fig. 7. (A) Proposed mechanism of NTR-activated 18 F-NCRP in BLM-injured lung at various stages.(B) Schematics of the procedures and timelines of PET imaging for BLM-injured mice.Saline group served as the control.

Table 1
Overview of radiotracers for IPF PET imaging reported in the literatures.