Background
The number of SARS-CoV-2 infections worldwide continues to rise rapidly, as the recently emerged Omicron variant shows enhanced transmissibility compared to that of the previous variants [
1‐
4]. Although a number of vaccines have been widely administered in many countries [
5], some patients still develop severe symptoms leading to increase in hospitalizations and number of deaths. Traditional antiviral drugs, such as hydroxychloroquine [
6] and lopinavir/ritonavir [
7], have shown unsatisfactory anti-SARS-CoV-2 effects in clinical trials. However, in addition to symptomatic treatments, such as hemodynamic support and ventilation support, a number of novel SARS-CoV-2 antiviral drugs, including remdesivir [
8], molnupiravir [
9], and Pfizer's oral antiviral drug Paxlovid [
9], have recently received Emergency Use Authorization from the FDA. Relative risk of hospitalization or death was reduced by 30% in a group treated with molnupiravir, according to the results of a Phase III clinical trial [
10]. In addition, interim data from Pfizer's Phase II clinical trial showed that, among participants treated within 3 days of the onset of COVID-19 symptoms, the Paxlovid group had an 89% reduction in the risk of COVID-19-related hospitalization or death from any cause compared to that of the placebo group [
11,
12].
Remdesivir and molnupiravir are nucleotide analogues, which integrate into nascent viral RNA strands, resulting in premature termination of RNA replication [
9,
13]. Paxlovid consists of two components, with PF-07321332 (nirmatrelvir) being the main active component that can inhibit the activity of precursors of SARS-CoV-2 main protease, thus inhibiting the synthesis of virus-related proteins and viral replication. Nirmatrelvir is primarily metabolized by the CYP3A4 enzyme while ritonavir is an inhibitor of both HIV-1 protease and CYP3A4, thus ensuring the maintenance of sufficient blood concentrations of nirmatrelvir [
9]. Recommended dosage of Paxlovid by FDA was 300 mg nirmatrelvir and 100 mg ritonavir twice a day for five consecutive days [
14]. Paxlovid, a popular antiviral drug used to treat SARS-CoV-2, has been reported to cause adverse events, such as taste disorders, diarrhea, hypertension, and myalgia, in clinical trials [
14]. Given the inhibition of CYP3A4 by ritonavir, drug interactions may be another source of serious adverse reactions [
15,
16].
Recently, increasing attention has been paid to the effects of SARS-CoV-2 infection on bone metabolism [
17,
18]. However, to date, no study has reported whether the use of these novel antivirals will initiate or accelerate the progression of degenerative diseases of the musculoskeletal system. Osteoarthritis, one of the most common degenerative joint disease, is the leading cause of disability worldwide [
19]. It is characterized by the degeneration of articular cartilage, osteophyte formation, and osteosclerosis of subchondral bone and synovitis [
20,
21]. Chondrocytes are responsible for secreting extracellular matrix, which comprises the articular surface. Aberrant mechanical conditions, inflammation in the microenvironment, and senescence are the main causes of chondrocyte degeneration and extracellular matrix degradation [
22,
23]. In recent years, the role of endoplasmic reticulum stress (ER stress) has emerged in the pathogenesis of osteoarthritis [
24]. When exposed to conditions such as hypoxia or nutrient deprivation, unfolded proteins will accumulate in the ER and the unfolded protein response (UPR) is activated to restore ER homeostasis. However, continuous and uncontrollable UPR also initiates cell death signaling and abolishes tissue homeostasis locally [
25].
Ferroptosis is a form of cell death characterized by iron-dependent lipid peroxidation. A balance of labile iron ion concentrations, reactive oxygen species (ROS), and antioxidant enzymes regulates the production and elimination of peroxidized phospholipids [
26,
27]. Oxidative stress and subsequent ferroptosis have been shown to be significantly involved in the development of osteoarthritis, and many therapies targeting ferroptosis and oxidative stress have achieved satisfactory outcomes in patients with osteoarthritis [
28,
29]. Nuclear factor erythroid 2-related factor 2 (Nrf2) is the key regulator for the maintenance of redox homeostasis. Recent studies have revealed its connection with glucose and glutamine metabolism. Nrf2 activation redirects glucose and glutamine into anabolic pathways, which contributes to tumor proliferation and resistance to chemotherapies [
30,
31]. In addition, redox signaling coordinates with ER stress to determine the cell fate since increased ROS levels can aggravate ER stress. Activating transcription factor 4 (Atf4), a downstream transcription factor of ER stress, can further strengthen the transcription activity of Nrf2 and increase the expression of antioxidant enzymes [
32]. ER stress is composed of three independent signaling pathways: Ire1α-Xbp1s, Perk-eIF2α-Atf4, and Atf6 pathways.
At present, whether the interplay of redox signaling and ER stress participates in the effects of Paxlovid on chondrogenic differentiation and progression of osteoarthritis remains unknown. Therefore, this study aimed to investigate the effects of Paxlovid, current FDA approved anti-SARS-CoV-2 drug, on chondrogenic differentiation in vitro and osteoarthritis progression in vivo. The findings are expected to attract increased attention to the occurrence of osteoarthritis in patients with COVID-19 treated with Paxlovid.
Methods
Animals and surgical procedures
C57BL/6 J mice were purchased from the animal center of Shanghai Ninth People’s Hospital, and animal experimental plans were approved by the ethics committee of the hospital (SH9H-2022-A95-1). Twenty-four 12-week-old male mice were divided randomly into four treatment groups: Sham, Paxlovid, DMM (destabilization of the medial meniscus), and DMM + Paxlovid. For mice receiving the DMM surgery, we anesthetized them with isoflurane. Following exposure of the right knee joint, we bluntly dissected the fat pad and transected the medial meniscotibial ligament to enable destabilization of the medial meniscus. 150 mg/kg of nirmatrelvir and 50 mg/kg of ritonavir dissolved in 10% dimethyl sulfoxide (DMSO) and 90% corn oil were injected intraperitoneally every day in the third week after DMM surgery in Paxlovid and DMM + Paxlovid treated groups. The Sham and DMM groups were intraperitoneally administered the same volume of phosphate buffered saline (PBS). Mice were sacrificed four weeks after surgery, and both knee joints were collected and fixed in 75% ethanol after being treated for 48 h in 4% paraformaldehyde (PFA) for future micro-computed tomography (micro-CT) and histologic analysis.
Micro-CT scanning
After fixation, right knee joints of the mice were scanned using a micro-CT scanner (Skyscan 1072; Skyscan, Belgium) at a resolution of 10.5 μm with 55 k Vp source and 145 μ Amp current. Imaging profiles were analyzed and remodeled in CT-An and CT-Vox software (Bruker, Germany) to analyze the subchondral bones and osteophytes.
Histology and immunofluorescence staining
For histologic and immunofluorescence analyses, the prefixed knee joints were decalcified in 10% ethylenediaminetetraacetic acid for 2 weeks, then embedded into paraffin. Serial tissue sectioning of 5 μm thickness in a sagittal plane was performed and stained withand safranine O/fast green to evaluate cartilage degeneration and clefts. OARSI score was evaluated according to the standards of Sophocleous [
33].
For immunofluorescence staining, the knee joint sections were de-paraffinized in xylene and standard alcohol gradients and then washed with PBS three times. Then, the sections were incubated with an antigen retrieval buffer for 15 min and washed with PBS three times. Then, the sections were incubated with an anti-fluorescence quencher for 5 min and blocked with Ultra V block for 15 min at room temperature to block nonspecific antibody binding sites. Anti-Col2a1 (ab34712, 1:200) and anti-p21 (ab188224, 1:1000) primary antibodies were purchased from Abcam, United Kingdom, while anti-Ddit3 (15,204-1-AP, 1:200) and anti-Sod1 (10,269-1-AP, 1:200) primary antibodies were purchased from Proteintech, United States. The sections were incubated with the corresponding antibodies at 4℃ overnight and incubated with corresponding secondary antibodies the next day. The sections were then washed with PBS and the nucleus was stained with DAPI. Images were obtained randomly using a Zeiss DM4000B microscope, and the ImageJ software was used to carry out quantitative analysis by counting the positive cells in each visual field.
Cell culture and reagents
We isolated murine chondrocytes from 3-week-old C57BL/6 J mice. After dissecting the femoral head of the mouse, we carefully isolated a layer of cartilage from the femoral head and spliced it into small pieces. After digestion in 0.25% trypsin for 30 min, these cartilage pieces were digested in 0.2% collagenase type 2 for 4 h in a cell incubator. The next day, the cells were centrifuged, and chondrocytes were cultured in Dulbecco’s modified eagle medium (DMEM) with 4.5 g glucose/L, 10% fetal bovine serum (FBS), and 1% penicillin–streptomycin (Gibco, Thermo Fisher Scientific, Waltham, MA, United States).
We purchased the immortalized mouse chondrocyte cell line ATDC5 from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The ATDC5 cells were cultured in DMEM with 4.5 g glucose/L, 5% FBS, and 1% penicillin–streptomycin. Cells were incubated in a humid environment at 37℃ with 5% CO2. Paxlovid, GSK2606414, and N-acetylcysteine (NAC) were purchased from MCE (New Jersey, United States) and dissolved in DMSO. We ensured that the final DMSO concentration in the medium used for cell culture was less than 0.1%.
Micromass culture
Micromass culture was used to evaluate the collagen secretion ability of chondrocytes. After digestion, centrifugation, and suspension, a 1.5 × 107 per mL ATDC5 cell suspension was prepared, and 10 μL of the suspension was added to the center of each well in a 24-well culture plate. After 2 h in the cell incubator, 1 mL of DMEM with 5% FBS and insulin-transferrin-selenium (ITS) (Gibco, Thermo Fisher Scientific, Waltham, MA, United States) was added to each well. Medium was added gently to avoid deformation of the micromass. Medium was replaced every other day for 7 days. After fixation in 4% PFA for 10 min, alcian blue and toluidine blue dyes were added to stain the extracellular matrix in the micromass.
Cell proliferation and cytotoxicity
The effect of Paxlovid on the proliferation of chondrocytes was determined using the Cell Counting Kit 8 (CCK8). Chondrocytes were seeded into 96-well culture plate at a density of 6,000 ATDC5 cells per well and cultured with 100 μL DMEM supplemented with 5% FBS and 1% penicillin–streptomycin. Medium was replaced on the second day, and different concentrations of Paxlovid were added at 24, 48, and 72 h. At each timepoint, 100 μL of the medium containing 10% CCK8 buffer (Dojindo, Kumamoto, Japan) was added to the wells and incubated in a cell incubator for 2 h to avoid light exposure. Absorbance was read at the 450 nm wavelength using an Infinite M200 pro multimode microplate reader (Tecan Life Sciences, Switzerland).
Cell cycle analysis
The cell cycle stage of the ATDC5 cells was analyzed using flow cytometry. After being cultured with different concentrations of Paxlovid for 24 h, the ATDC5 cells were digested, centrifuged, and suspended with 1 mL PBS three times. Pre-cooled 75% ethanol was used to fix the cells at 4 ℃ overnight. After fixation, the cells were centrifuged and suspended with 1 mL PBS three times the next day. Then, 5 μg/mL DAPI (Beyotime, Nanjing, China) was used to stain the cells placed on ice and protected from light exposure for 15 min. The cell cycle phase profile was collected on a BD Fortessa system (BD Bioscience, USA) and analyzed using FlowJo 10 (BD Bioscience, USA) to calculate the proportion of cells in different phases.
Senescence-associated-β-galactosidase (SA-β-gal) staining
Cellular senescence was detected using SA-β-gal staining. Murine chondrocytes were cultured with different concentrations of Paxlovid for 7 days in a 24-well-plate. After fixation with 4% PFA for 10 min, a SA-β-gal staining kit (Beyotime, Nanjing, China) was used to stain the senescent cells at 37 ℃ overnight. Images were taken randomly with a Zeiss microscope, and the ImageJ software was used for quantitative analysis.
ROS detection
Intracellular ROS was detected using a DCFH-DA probe (1:1000; S0033S; Beyotime Biotechnology, China). Briefly, the ATDC5 cells were plated in confocal dishes (Cellvis, CA, USA), then cultured with different concentrations of Paxlovid for 24 h. Then, the cells were incubated in serum-free culture medium containing 10 μM DCFH-DA for 20 min at 37 ℃. After that, the cells were stained with DAPI for nuclear fluorescence, then washed with PBS three times. Fluorescent images were captured via confocal microscopy (Zeiss DM4000B, Leica). The ImageJ software was used for semi-quantitative analysis of ROS-positive cells.
FerroOrange staining
A FerroOrange kit (F374, Dojindo, Shanghai, China) was used to detect intracellular Fe2+. After the Paxlovid treatment, the ATDC5 cells were washed with PBS and stained with DAPI for nuclear fluorescence, then washed with PBS three times. Then, the cells were incubated with FerroOrange working solution (1 μmol/L) for 30 min. Fluorescent images were captured via confocal microscopy (Zeiss DM4000B, Leica).
Protein extraction and Western Blot analysis
After incubation with different concentrations of Paxlovid for 24 h in a six-well plate, 200 μL of the RIPA lysis buffer (Beyotime, Nanjing, China) with a 1% cocktail of protease and phosphatase inhibitors (Thermo Fisher Scientific, Waltham, MA, United States) was added to each well containing the treated cells. The whole lysis process, performed on ice, was completed in 30 min. After, the cells were centrifuged at 12,000 × g for 15 min, the bicinchoninic acid assay was used to detect the protein concentration of the supernatant. 5X SDS-sample loading buffer was then added to the supernatant, and the protein was boiled at 99℃ for 10 min.
For western blot analysis, 20 μg protein per well was loaded onto SDS-PAGE gels and then transferred to a polyvinylidene fluoride membrane (Millipore, Merck, Germany). The membrane was blocked using 5% non-fat milk for 1 h at room temperature to block nonspecific antigen sites. After, the membrane was rinsed with tris-buffered saline and Tween 20 three times, it was incubated with the corresponding primary antibody overnight at 4 ℃. The membrane was incubated with secondary fluorescence antibodies for 2 h at room temperature, and the membrane was visualized using the Odyssey V3.0 image scanner (Li-COR. Inc., Lincoln, NE, USA). Quantitative measurements were made by measuring the gray value of each strip. Anti-Sox9 (ab185966, 1:1000), anti-CollagenII (ab34712, 1:1000), anti-p16 (ab211541, 1:2000), and anti-p21 (ab188224, 1:1000) primary antibodies were purchased from Abcam. Anti-β-actin (4970, 1:2000), anti-Xbp1s (40,435, 1:1000), anti-Atf4 (11,815, 1:1000), anti-Bip (3177, 1:1000), anti-Elf2α (9722, 1:1000), anti-phospho-Elf2α (3398, 1:1000), anti-Ddit3 (2895, 1:1000), and anti-Ire1α (3294, 1:1000) primary antibodies were purchased from CST, Boston, United States. Anti-Ho1 (10,701-1-AP, 1:1000) and anti-Sod1 (10,269-1-AP, 1:1000) primary antibodies were purchased from Proteintech. The secondary antibody was anti-rabbit IgG (H + L; DyLight™ 800 4 × PEG conjugate; Abcam).
Quantitative real-time PCR (qRT-PCR)
For qRT-PCR, total RNA was extracted from the treated ATDC5 cells using a total RNA kit (R6812-01HP, Omega Bio-tek Ink., Norcross, GA, United State). 1000 ng of cDNA was prepared according to the protocols provided by the cDNA synthesis kit (Takara, Shiga, Japan) and diluted to 200 μL. A total (10 μL) of 3.2 μL ddH
2O, 1 μL cDNA, 0.4 μL upstream primer, 0.4 μL downstream primer, and 5 μL SYBR premix were mixed and added to each well of a 384-well plate. qRT-PCR was performed using an ABI 7500 Sequencing Detection System (Applied Biosystems, Foster City, CA, USA). All primer sequences for the target genes are listed in Table
1. Gene expression levels were calculated using the 2
−△△CT method, and data were presented as fold increase compared to controls.
Table 1
Primers used in the qRT-PCR assay
Gapdh | Mus musculus | GGCAAGTTCAACGGCACAG | CGCCAGTAGACTCCACGACAT |
Col2a1 | Mus musculus | GCTACACTCAAGTCACTGAACAACCA | TCAATCCAGTAGTCTCCGCTCTTCC |
Sox9 | Mus musculus | CGTGGACATCGGTGAACTGAG | GGTGCTGCTGATGCCGTAAC |
P16 | Mus musculus | GGTCACACGACTGGGCGATT | GCACCGTAGTTGAGCAGAAGAG |
P21 | Mus musculus | GCCTGGTTCCTTGCCACTTCTT | ATTACGGTTGAGTCCTAACTGCCATC |
P53 | Mus musculus | CTCCAGCTACCTGAAGACCAAGAAG | GCAGAGACCTGACAACTATCAACCTAT |
Atf3 | Mus musculus | GGTCGCACTGACTTCTGAGG | CTCTGGCCGTTCTCTGGA |
Atf4 | Mus musculus | TGGCGAGTGTAAGGAGCTAGAAA | TCTTCCCCCTTGCCTTACG |
Atf6 | Mus musculus | AGAGGCAGCACACGCATTCC | TGATGGTCAGCAGGAGCAGAGA |
Chop | Mus musculus | CATACACCACCACACCTGAAAG | CCGTTTCCTAGTTCTTCCTTGC |
Ire1α | Mus musculus | CTGGCGAGAAGCAGCAGACTT | CCACCACAGGAGAGGCATAGTT |
Bip | Mus musculus | CCTCATCGGACGCACTTGGAAT | GCTTGTCGCTGGGCATCATTG |
Xbp1s | Mus musculus | GAGTCCGCAGCAGGTG | GTGTCAGAGTCCATGGGA |
Fgf21 | Mus musculus | CAGGGGTCATTCAAATCCTG | AGGAATCCTGCTTGGTCTTG |
RNA sequencing
The ATDC5 cells (3 × 10
5) were seeded in a six-well-plate and stimulated with Paxlovid (none in control group and 60 μM in Paxlovid group) for 24 h. Total RNA was extracted from the cells using a total RNA kit (R6812-01HP, Omega Bio-tek Ink., Norcross, GA, United State), according to the manufacturer’s protocol and then analyzed by RNA sequencing performed by Wuhan Huada Gene Technology Co., Ltd. (China). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis, Gene ontology (GO) enrichment analysis, gene set enrichment analysis (GSEA), and heat maps using mRNA relative expression as transcripts per kilobase million were performed using the Mybgi platform (Wuhan Huada Gene Technology,
https://mybgi.bgi.com/tech/ login).
Statistical analysis
Statistical analyses were conducted using SPSS version 19.0 for Windows (SPSS Inc., Chicago, IL.). All data are representative of more than three independent experiments unless otherwise indicated. For data with a normal distribution, comparison between two groups was analyzed using unpaired Student’s t tests. Comparisons between three or more groups was analyzed using one-way ANOVAs and Student-Newman-Keuls post hoc tests. For ranked data, Mann–Whitney and Kruskal–Wallis tests were used to compare between two groups or multiple groups. All statistical charts were designed using GraphPad Prism 8 (GraphPad Software Inc, San Diego, CA, USA). A P value < 0.05 was considered significant.
Discussion
At present, there are two main options under consideration for antiviral therapy for SARS-CoV-2 [
13]. The first neutralizes the RBD area of the spike protein using monoclonal antibodies to inhibit it from binding to the ACE2 receptor [
34], and the other involves small molecule drugs that interfere with virus replication. Recent research has found that more than 30 mutations exist in the spike protein of Omicron BA.1, which makes it capable of evading most anti-RBD neutralizing antibodies [
35]. However, small molecule drugs, such as Paxlovid and molnupiravir, are still effective against Omicron BA.2. Kawaoka et al. [
1] showed that nirmatrelvir and molnupiravir could inhibit the proliferation of Omicron BA.2 in the lower respiratory tract of hamsters. Considering the current prevalence of the Omicron mutant, humans are expected to coexist with Omicron and receive antiviral small molecule drugs for the foreseeable future. Therefore, it is of great significance to explore the long-term effects of SARS-CoV-2 infection and antiviral drugs on not only the respiratory system but also all the other systems.
Many studies have focused on the impact of Omicron infection on musculoskeletal diseases [
17,
18]. Inflammation can influence bone remodeling, and researchers have reported that chronic obstructive pulmonary disease, asthma, and cystic fibrosis can disrupt bone metabolism and lead to pathological bone loss [
36,
37]. In an established golden Syrian hamster model infected with SARS-CoV-2, significant multifocal loss of the bone trabeculae was found in the long bone and lumbar spine only 4 days after infection. This bone loss may be caused by the accumulation of circulating proinflammatory cytokines and activated osteoclast differentiation. Gao et al. [
17] found that SARS-CoV-2 can also directly infect bone marrow macrophages through non-canonical receptor Nrp1 rather than through ACE2, which in turn inhibits their differentiation into osteoclasts. These studies suggest that SARS-CoV-2 has the potential to cause complications that so far have been neglected, suggesting that we will need to follow-up on patients with COVID-19 for a longer duration to determine the impact of the virus on target organs outside the respiratory system.
Till date, no literature has reported whether novel small molecule antiviral drugs pose similar side effects on the musculoskeletal system. To the best of our knowledge for the first time our study showed that Paxlovid can trigger ER stress and oxidative stress in chondrocytes, which gives rise to chondrocyte senescence and hampers chondrogenic differentiation and extracellular matrix secretion. Upon Paxlovid treatment, disrupted redox homeostasis and compensatory elevation of antioxidants coordinate with ER stress to accelerate the degeneration and senescence of chondrocytes. Our in vivo experiments also showed that in a mechanical instability model, the use of Paxlovid can aggravate the cleft of cartilage and calcified tissue generation. These results suggest that patients receiving Paxlovid treatment also need long-term follow-up to determine its impact on the development of osteoarthritis.
Drug safety currently focuses on the use of Paxlovid in special populations and on drug interactions, and previous studies [
14] have shown that Paxlovid has no teratogenic effect on newborns. To the best of our knowledge for the first time our study investigated osteoarthritis as a side effect of anti-SARS-CoV-2 drugs. Our data revealed that 10 μM nirmatrelvir with 3.33 μM ritonavir was enough to evidently affect the gene expression associated with cartilage differentiation, senescence, and ER stress after a short 24 h exposure. After long-term exposure for 7 days, 2.5 μM nirmatrelvir with 0.67 μM ritonavir significantly reduced the level of collagen secreted by micromass cultured chondrocytes, which indicates that Paxlovid can interfere with chondrogenic differentiation in a concentration dependent manner. According to the present the pharmacokinetic data of Paxlovid, the maximal plasma concentration of nirmatrelvir is 2.21 μg/mL and area-under-the-plasma concentration versus time curve is 23.01 μg
.hr/mL after a single dose of 300 mg nirmatrelvir and 100 mg ritonavir in healthy individuals [
14,
38]. This is consistent with the effective concentration observed in our study. However, the knee cartilage is nourished by synovial fluid in the joint capsule and the local concentration of Paxlovid in synovial fluid is unknown. Therefore, further studies should explore the local drug concentration in patients. In addition, it is reasonable to speculate that Paxlovid also accelerates cell senescence and ER stress in other tissues considering its systemic circulation. A cohort of patients treated with Paxlovid ought to be established to monitor its long-term effects on the other systems, such as the cardiovascular system.
In vitro results showed that after a 7-day exposure, low concentration of Paxlovid was sufficient to reduce matrix protein secretion by chondrocytes. At present, Pfizer recommends Paxlovid twice a day for 5 consecutive days after infection [
14]. Even if the symptoms are not significantly relieved, the medication duration is not recommended to be prolonged. To determine whether such medication frequency and duration have similar destructive effects on knee cartilage, the duration and frequency of Paxlovid administration recommended by Pfizer was used for mice following DMM surgery. Histological analysis supported that the administration of Paxlovid for 1 week accelerates the degeneration of cartilage. It is thus suggested that patients with COVID-19, especially those with knee osteoarthritis, may face a high risk of disease progression and further deterioration.
This study has some limitations. Firstly, there is no study reporting the Paxlovid concentration present locally in synovial fluid of the knee joint, which makes it difficult to determine whether the actual Paxlovid concentration would be sufficient to exert an effect on chondrocytes. Secondly, the ideal in vivo model would have both SARS-CoV-2 infection and osteoarthritis with Paxlovid treatment. In contrast, our animal model received Paxlovid treatment without SARS-CoV-2 infection. Animal models simulating the actual condition should be used to determine whether SARS-CoV-2 infection and Paxlovid treatment have a synergistic effect on the occurrence and development of arthritis. Finally, this study only addressed the phenomenon of Paxlovid triggering cartilage senescence and degeneration through ER stress, oxidative stress, and downstream ferroptosis, aggravating the development of arthritis, but we did not explore the specific molecular mechanism behind this aggravation. Further research is needed to elucidate the mechanistic relationships between Paxlovid, ER stress, and oxidative stress. Recently, several researches have demonstrated that Nrf2 also regulates glutamine metabolism reprograming, which contributes to cancer malignant phenotypes such as resistance to chemotherapy. Whether metabolism reprogramming also plays a role in Paxlovid-induced oxidative stress seems to be a promising theme to study [
39,
40].
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.