Introduction

Aging is a natural biological process characterized by the gradual decline of physiological functions across multiple tissues1,2. Several aging-related phenotypes serve as primary risk factors for various diseases including cancer, diabetes, neurodegeneration, cardiovascular diseases, and immune system diseases1,2. Progeroid syndromes, regarded as premature aging disease, encompass a group of rare genetic disorders that recapitulate multiple physiological aging-associated phenotypes during early development3,4. Notably, Hutchinson-Gilford progeria syndrome (HGPS) is a rare autosomal dominant genetic disorder primarily attributed to a point mutation in the LMNA gene (c.1824 C < T), resulting in an alternative splicing event and the production of a truncated Lamin A protein known as progerin5. HGPS patients suffer from a short lifespan, who often die from the heart diseases and strokes in childhood.

Several symptoms of progeria also happen to normal aging process. Clinical manifestations of HGPS include growth retardation, loss of hair, sclerotic skin, cardiovascular alteration, metabolic dysregulation, bone abnormalities, and inflammation5,6, most of which resembling accelerated aging phenotypes in childhood. On a cellular level, HGPS shares many cellular alterations with normal aging, including accumulated DNA damage, mitochondrial dysfunction, genomic instability, telomere aberrations, and loss of heterochromatin. Although HGPS displays some similarities with natural aging, it harbors some unique functional defects that are not always seen in the aged. For example, most HGPS patients develop severe cardiovascular diseases that cause patient death at young ages7,8. Moreover, the nervous system is normally unaffected in HGPS patients, while neural disorder, such as memory decline, gradually appears during aging9. Thus, HGPS serves as a valuable model for studies of not only the progeria syndromes but also aging and aging-related pathogenesis to some extent10.

Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has emerged as a global public health emergency of international concern11. Particularly, severe COVID-19 cases, characterized by pneumonia, respiratory failure, short of breath, septic shock, and multiple organ dysfunction, calls for more healthcare and medical resources12,13. Patients with certain comorbidities are at a high risk of progressing to severe COVID-19, and one of the main risk factors is aging14. Studies have shown that the severity and fatality rates are notably higher in elderly population as compared to the young, with over 73% deaths occurring in individuals over 6515. Despite identifying age-related changes such as low immune response activity contributing to increased susceptibility to infectious diseases among the aged, the specific aging-related phenotypes that contribute most to severe or critical COVID-19, and their underlying mechanisms, remain incompletely understood.

SARS-CoV-2 enters host cells by recognizing angiotensin-converting enzyme 2 (ACE2), a protein conserved across different species16. However, mice, the most used model for human diseases, are not susceptible to SARS-CoV-2 due to the low affinity of mouse ACE2 for the virus17,18. To address this inherent resistance, our previous studies focused on establishing a humanized ACE2 mouse model using stem cell-based genome editing and tetraploid compensation19. Additionally, mice carrying a point mutation (c.1827C > T, representing LMNA c.1824C > T in human) at exon 11 of Lmna gene mimic the clinical manifestations of human HGPS20. These HGPS mice display aging-related phenotypes within 3 months, significantly faster than natural aging20.

To investigate the response of HGPS mice to COVID-19 infection and compare the infection outcomes between the young, normal aging, and HGPS, in this study we generated an HGPS mouse model based on our hACE2 mouse and subjected these HGPS/hACE2 mice to viral challenge. Transcriptome analysis was conducted to monitor changes throughout the infection, and comparisons were made between young, aged, and HGPS mice in their response to SARS-CoV-2. We found that hACE2 mice carrying HGPS mutation show different viral replication dynamics compared to the young mice, which is similar to the aged mice. Transcriptome analysis identified common biological pathways affected across all animal groups, with HGPS mice also exhibiting metabolic dysregulation during viral infection. Furthermore, we demonstrated that the immune response was significantly impaired in aged and HGPS mice compared to young mice upon viral challenge, while aged mice show severe pulmonary hemorrhage in the respiratory tract. Overall, our study evaluates the potential pathological phenomena induced by SARS-CoV-2 infection in HGPS, offering insights for understanding progeria syndrome and informing the treatment of COVID-19 patients with premature aging.

Results

Generation of HGPS/hACE2 mouse model

To generate a mouse strain carrying HGPS variants, we utilized CRISPR based base-editing system to introduce a homozygous c.1827 C < T (p.Gly609Gly) mutation in exon 11 of mouse Lmna gene, equivalent to the c.1824 C < T (p.Gly608Gly) mutation found in human LMNA, in mouse embryonic stem cells (mESCs) with humanized ACE221 (Fig. 1A and Supplementary Fig. 1A). Karyotyping and immunofluorescent staining of Lmna mutant mESCs confirmed correct chromosome arrangements and pluripotency (Supplementary Fig. 1C,D). To avoid the breeding issues and ensure an adequate supply of mice for infection studies, we employed tetraploid complementation using our modified mESCs, and generated humanized ACE2 mice carrying homozygous HGPS variants (Fig. 1A,C). These mice exhibit an alternative splicing event observed in HGPS patients (Supplementary Fig. 1B), resulting in the absence of Lamin A protein and the presence of the spliced Lamin protein known as progerin (Fig. 1B). Similar to HGPS patients, HGPS hACE2 mice display small body size, reduced body weight, and shortened life span (Fig. 1C–E), indicating a premature aging phenotype in these mice.

Fig. 1
figure 1

Characterization of HGPS/hACE2 mice. (A) Schematic diagram of HGPS/hACE2 mouse generation strategy using mES editing combined with tetraploid complementation techniques. (B) Western blot analysis of lung tissues from hACE2 and HGPS/hACE2 mice, showing the presence of Lamin A/C in wild type lung and progerin in HGPS lungs. Histone 3 was used as a loading control. (C) Representative photographs of 2-month-old unedited, heterozygous, and homozygous Lmna c.1827C < T mutant hACE2 mice. (D) Survival curve plot showing the life span of wild type, heterozygous and homozygous Lmna mutant mice. D = days (E) Boxplots illustrating the body weight distribution among young, aged, and HGPS Balb/c mice. (F) Volcano plot showing transcriptome changes in the lungs between hACE2 and HGPS/hACE2 mice. Significantly differentially expressed genes are highlighted in green (upregulated in HGPS) and red (upregulated in wild type). (G) Dropblot of GO analysis results derived from differentially expressed gene sets in the lungs of 2-month-old hACE2 and HGPS/hACE2 mice.

Full size image

Tissues such as bone, heart, spleen, and thymus exhibit varying levels of defects in HGPS patients, while not much is known for the lung5. To further investigate the impact of HGPS on the respiratory system in HGPS mice, we carried out transcriptome analysis of lung tissues from 2-month-old wild type and HGPS/hACE2 mice (Fig. 1F). Consistent with previous studies, we observed up-regulation of genes involved in P53-mediated DNA damage pathway including Gadd45a, Gadd45b, Gadd45g, Atf3, Btg2, and Cdkn1a (Supplementary Fig. 1E). Gene Ontology (GO) analysis revealed that immune-related pathways were more enriched in normal hACE2 mice compared to the HGPS ones, indicating a diminished immune response in HGPS mice (Fig. 1G). Conversely, multiple developmental pathways such as epidermis, vasculature, bone marrow, and adipose development were more enriched in HGPS compared to normal hACE2 mice, suggesting a premature developmental stage of lung in HGPS mice (Fig. 1G). Typical marker genes of distinct cell types in lungs show only subtle changes in HGPS/hACE2 compared hACE2 mice, indicating a relatively normal lung function in HGPS/hACE2 mice (Supplementary Fig. 1F). Overall, we have successfully generated a hACE2 mouse model harboring the HGPS mutation, which displays phenotypic similarities to human accelerated aging.

Viral replication and host response to primary infection with SARS-CoV-2

To investigate infection outcomes of HGPS mice, we examined the pathogenesis of SARS-CoV-2 in 2-month-old HGPS/hACE2 mice infected with the wild-type strain WIV04. In parallel, age-matched young hACE2 mice (2 months old) and aged hACE2 mice (over 14 months old) were included for virus challenging. Mice received intranasal inoculation with 1 × 105 50% tissue culture infective dose (TCID50) of the virus and lung samples were collected at 1, 3, 5, and 7 days post-infection (dpi) for virus replication measurement, clinical response examination, and transcriptome analysis (Fig. 2A). Although young, aged, and HGPS mice exhibited distinct expression levels of hACE2 in lung, transcriptome analysis of SARS-CoV-2 revealed that aged and HGPS mice displayed more expression of SARS-CoV-2 genes in lung including E, M, N, S, and ORF genes in lung at 1 dpi, which then decreased at 3 dpi (Fig. 2B and Supplementary Fig. 2A,C). In contrast, SARS-CoV-2 genes were detected at 1 dpi in the lungs of young mice, reaching the peaks at 3 dpi (Fig. 2B and Supplementary Fig. 2A). Immunohistochemistry (IHC) for SARS-CoV-2 N protein indicates predominant infection of airway epithelial cells with some viral presence in alveolar cells (Supplementary Fig. 2B). Consistent with the transcriptomic changes, immunofluorescence staining revealed a gradually decline of SARS-CoV-2 N protein in the lungs of the mice during the infection progression (Supplementary Fig. 2B). Interestingly, while infected young mice exhibited comparable N gene transcription levels, the protein was significantly less abundant compared to infected aged and HGPS mice (Supplementary Fig. 2B), suggesting efficient viral clearance in the young mice. We observed mild decreased body weight in infected young mice compared to the mock-infected ones, while no changes were detected between infected and uninfected aged mice (Fig. 2C). HGPS mice exhibited a slight loss of body weight in both infected and uninfected groups (Fig. 2C). In sum, these findings suggest that SARS-CoV-2 may replicate differentially in the lungs of young, aged, and HGPS/hACE2 mice, likely resulting in varied responses upon viral infection among different mouse groups.

Fig. 2
figure 2

SARS-CoV-2 infection and replication dynamics in young, aged, and HGPS hACE2 mice. (A) Experimental scheme of viral infection. Young, aged, and HGPS mice were intranasally infected with 1 × 105 TCID50. Lung samples were collected at 1, 3, 5, 7 days post-infection (dpi) for RNA and H&E. Mouse body weight were measured for up to 7 days. (B) Expression levels of viral genes (ORF1ab, ORF7a, ORF8, and ORF10 genes) in the lungs of infected young, aged, and HGPS mice at 1, 3, 5, and 7 dpi. Data was shown as normalized expression representing normalized read counts from RNA-seq data. Error bars, SEM from at least 2 biological replicates. Student t-test, *P < 0.05, **P < 0.01. (C) body weight changes of young, aged, and HGPS mice upon viral infection throughout the whole experiment.

Full size image

Aged hACE2 mice displayed more severe pathological phenotypes induced by SARS-CoV-2 infection

To further evaluate the outcomes of viral infection in different mouse group, we conducted histopathological analysis which illustrated distinct degrees of lung damage among the infected young, aged and HGPS/hACE2 mice. In uninfected control animals, slight inflammation was observed in aged and HGPS mice, characterized by a minor immune cell infiltration, which was not detected in the young mice (Fig. 3A,E,I). At 1 dpi, mild changes were observed in the lung tissues of young mice, including the presence of lymphocyte and macrophage infiltration in some alveolar spaces, along with slight thickened alveolar walls (Fig. 3B). Severe pneumonia developed at 3 dpi in the young mice, characterized by multifocal lesions, pulmonary hemorrhage, and massive infiltration with increased number of mixed inflammatory cells at peri-vascular regions (Fig. 3C). This phenotype became less severe at 5 dpi, with fewer inflammatory cell infiltration and milder thickened alveolar walls compared to those observed at 3 dpi (Fig. 3D). In consistence with young mice, aged mice displayed hemorrhage and slight thickened alveolar walls at peri-bronchial and peri-vascular regions at 1 dpi, which became more pronounced at 3 dpi (Fig. 3F–G and Supplementary Fig. 2D). By 5 dpi, we found that two out of three aged mice displayed severe pulmonary hemorrhage in the bronchus and pulmonary alveoli, along with inflammatory cell infiltration and fibrin exudation, suggesting a severe COVID-19 phenotype (Fig. 3H and Supplementary Fig. 2D). Still, all infected mice survived viral infection till euthanized for sample collection at 7 dpi. Notably, these two mice showing severe pulmonary hemorrhage phenotype were in poor condition when being euthanized for sample collection. In contrast, HGPS/hACE2 mice did not display a severe phenotype throughout the infection, showing only a slightly increased number of immune cells at peri-bronchial and peri-vascular areas at 1, 3 and 5 dpi (Fig. 3J–L), suggesting less severe phenotypes in there HGPS mice. Altogether, our histopathological findings demonstrate distinct pathological features induced by SARS-CoV-2 infection between different animal groups, highlighting a severe phenotype in aged mice compared to young hACE2 and HGPS/hACE2 mice.

Fig. 3
figure 3

Pathological changes in young, aged, and HGPS hACE2 mice after SARS-CoV-2 infection. (A–D) H&E staining of the lungs of young hACE2 mice. Mock-infected mice show no lesions. After viral infection, immune cell infiltration (1, 3, and 5 dpi, green arrow), hemorrhage (3 and 5 dpi, yellow arrow), and thickened alveolar walls (3 and 5 dpi, red arrow) were observed at different time points. (E–H) H&E staining of the lungs of aged hACE2 mice. Mock-infected mice show slight immune cell infiltration (green arrow). Upon viral infection, immune cell infiltration (1 dpi, green arrow), heavy pulmonary hemorrhage (1, 3, and 5 dpi, yellow arrow), and thickened alveolar walls (1, 3, and 5 dpi, red arrow) were observed at different time points. (I–L) H&E staining of the lungs of HGPS hACE2 mice. Mock-infected mice show slight immune cell infiltration (green arrow). After viral infection, immune cell infiltration (1, 3, and 5 dpi, green arrow) and thickened alveolar walls (1, 3, and 5 dpi, red arrow) were observed at different time points. Black scale bar = 200 μm, and green scale bar = 50 μm.

Full size image

Gene expression dynamics from different groups with viral infection

To comprehensively understand how SARS-CoV-2 induces pathological differences among different mouse groups, we conducted transcriptomic analysis to investigate the dynamics of gene expression profiles in the lungs of young, aged, and HGPS hACE2 mice throughout the infection. We classified high variable genes into several clusters based on their expression patterns and calculate the averaged expressing trajectories for each cluster. Subsequently, we performed Gene Ontology (GO) analysis to elucidate the potential affected biological and molecular pathways for each cluster. Although viral gene expression peaked differentially in the lungs of young compared to aged and HGPS/hACE2 mice, we observed upregulated genes involved in a common response to virus and activation of innate immune upon viral infection transiently at 3 dpi in cluster 1 across all the infected groups (Fig. 4A,D, and Supplementary Fig. 3A). In young and aged groups, mice showed increased levels of genes in B cell-mediated immunity as well as immunoglobulin production pathways in the lungs at 5 dpi, accompanied by upregulation of genes involved in muscle development in cluster 2 (Fig. 4B,D, and Supplementary Fig. 3B). This indicates an immediate humoral immune response following viral exposure, which further triggers muscle contraction of lung smooth muscle. Interestingly, this activation occurred earlier in HGPS mouse lungs at 1 dpi (Fig. 4B), suggesting a distinct immune response to SARS-CoV-2 in HGPS. Viral infection has been reported to cause cilia loss from ciliated cells, resulting in cilia dysfunction in respiratory epithelium22,23,24. In agreement with previous findings, we observed a slight downregulation of genes in cluster 3 involved in cilium assembly and movement in young and HGPS mice, which was more pronounced in aged group (Fig. 4C,D and Supplementary Fig. 3C). This was followed by increased expression of those genes across all groups starting at 3 dpi, suggesting a restoration of function in fluid movement and mucus clearance in the respiratory airway (Fig. 4C,D and Supplementary Fig. 3C).

Fig. 4
figure 4

Transcriptome changing dynamics of different mouse group upon viral infection. (A–C) Soft clustering of variable genes of young, aged, and HGPS mice after viral infection. (D) Gene Ontology analysis of genes sets corresponding to the gene clusters from (A–C). (E) Soft clustering of variable genes with unique patterns of young (cluster 4) and (cluster 4) aged mice on the left, and corresponding GO analysis results from each cluster on the right. (F) Gene clusters of HGPS group with different trends on the left, together with the GO analysis results on the right.

Full size image

In addition to the common functional enrichments shared by the three groups, we also identified gene clusters with distinct dynamic patterns unique to each mouse group. For instance, cluster 4 in young mice declined linearly throughout the entire infection and was strongly enriched for genes associated with cell junction assembly, extracellular matrix organization, and cell–matrix adhesion (Fig. 4E). Moreover, Notch/vascular genes exhibited a rapid decline rapidly in aged mice at 1 dpi, after which a more gradual decline prevailed, indicating functional defects in vascular and endothelial cells in the lungs of aged mice (Fig. 4E). In infected HGPS mice, genes encoding sulfur, glycoprotein, and liposaccharide metabolic process pathways were featured in cluster 4, exhibiting immediate increase after infection till 3 dpi, followed by a drop towards the end of infection (Fig. 4F). However, cluster 5 contained genes related to lipid transport and exocytosis regulation pathways who reached their lowest expression level at 3 dpi, followed by a pronounced increase till 7 dpi, showing an opposite trend compared to cluster 4 (Fig. 4F). These findings suggest a likely SARS-CoV2-induced metabolic dysfunction in infected HGPS lungs.

Transcriptome comparison reveal strong immune response in young mice upon viral infection

Next, we focused on infection-induced transcriptomic changes and compared the transcription differences between different mouse groups at each time point post infection. Differential gene expression analysis, together with hierarchical clustering, revealed specifically expressed gene clusters from each group, and GO analysis was applied to these gene sets. Overall, we observed a comparable number of genes between young and aged group at 1, 3, and 5 dpi, while HGPS group displayed subtle changes at 1 and 3 dpi (Fig. 5A–C). Consistent with our histopathological analysis, expression of genes enriched in virus defense, interferon-beta, and innate immune response pathways was significantly higher in the lungs of young mice throughout the infection, suggesting a more active immune response upon viral infection compared to the aged and HGPS mice (Fig. 5D). In the aged mice, immune pathways such as immune cell migration, immunoglobulin production, humoral immune response, and cell chemotaxis represented feature genes in the clusters at 1 and 3 dpi (Fig. 5E). Additionally, we found pathways related to hemorrhage, including wound healing, blood coagulation, and erythrocyte development, were significantly enriched in aged mice, further confirming the pulmonary hemorrhage observed in the lungs of aged mice in the histopathological study (Fig. 5E and Supplementary Fig. 4). We did not obverse a significant pathway enriched in HGPS mice at 1 and 3 dpi, likely due to the low numbers of unique genes (36 genes for 1 dpi and 53 genes for 3 dpi) in this group. Whereas at 5 dpi, 332 genes were more activated in the lungs of HGPS mice. These genes primarily belong to pathways associated with entrainment of circadian clock, regulation of blood circulation, and hormone transport. This suggests a likely dysregulation of circadian rhythms in HGPS mice, although the enrichment in this pathway was less pronounced (Fig. 5F). Altogether, our results demonstrated that young mice, which have a normal immune system to protect against virus, display the highest immune response activation upon SARS-CoV-2 infection compared to aged and HGPS mice. Conversely, viral infection only causes mild pathological phenomena in HGPS mice, while aged mice exhibit lung hemorrhage when challenged by SARS-CoV-2, which could contribute to severe COVID-19.

Fig. 5
figure 5

Transcriptome comparison between different mouse group at 1, 3, and 5 dpi. (A–C) Heatmaps illustrating significantly differentially expressed genes from different mouse groups at 1 (A), 3 (B), and 5 (C) dpi. Expression levels are shown as Z-score format. (D–F) GO analysis of genes more active in young group at 1, 3, and 5 dpi (D), more active in HGPS group at 5 dpi (F), and unique to aged group at 1, 3, and 5 dpi (E). Young mice are enriched for genes in virus response and innate immune response pathways. Aged mice show enrichment of immune cell migration as well as blood coagulation pathways. Genes unique to HGPS mice are involved in circadian rhythm pathway.

Full size image

Discussion

COVID-19 pandemic has profoundly threatened normal life and health of people worldwide, particularly impacting older adults who are more susceptible to SARS-CoV-2 infection. Therefore, understanding the causal relationship between aging and severe COVID-19 is of great significance for improving preventive measures and therapeutic strategies for the elderly population. In this study, we have investigated the pathological consequences of SARS-CoV-2 infection in a progeria syndrome, HGPS, using a mouse model. We systematically compared the transcriptome landscape of lung tissues from young, aged, and HGPS/hACE2 mice in relation to SARS-CoV-2 infection. Firstly, our transcriptome analysis indicated that innate interferon response and virus defense pathway were more robustly activated in young compared to the aged and HGPS/hACE2 mice, consistent with the findings from other mouse models25,26. Aging is known to be associated with a functional decline in immune response, likely due to long-term chronic inflammation, as also evidenced by our histopathological results (Fig. 3E)27. Therefore, this observation further supports the low immune response in the aged and HGPS.

Interestingly, while aged mice displayed high rival replication at 1 dpi, they subsequently developed a severe pulmonary hemorrhage in the lungs at 5 dpi, mirroring observations in patients infected with SARS-CoV-2 who display abnormal coagulation profiles28,29. Severe COVID-19 has been reported from multiple mouse models. Jiang et al. used a transgenic mouse model with hACE2 expression driven by human FOXJ1 promoter to study SARS-CoV-2. Body weight loss, heavy lung damage and death were observed in these mice upon viral infection30. Dong used an K18-hACE2 mouse model which recapitulates severe COVID-19 with a low viral dose (2 × 103 PFU)31. Kenneth et al. and Jiang et al. applied mouse adopted coronavirus on wild type mice and demonstrated severe COVID-19 in the aged mice32,33. Here, using a different humanized ACE2 strategy, we displayed heavy hemorrhage in our aged hACE2 mice after coronavirus infection, consistent with clinical manifestations seen in COVID-19 patients34. Genes involved in cilium assembly and regeneration were activated at 3 dpi from all three groups, indicating a potential restoration of cilial function and clearance mechanisms. Thus, our study suggests that, alongside hypersecretion of mucus gel leading to alveolar and bronchial blockage35, pulmonary hemorrhage may also contribute to severe functional defect in lung. Although, our histological results partially confirmed transcriptomic dynamics from different infected mouse groups, we also notice that in general transcriptome changes do not always reflect functional consequences, meaning that the relevant functional studies would be beneficial to further support our findings.

Aging, together with other aging-related diseases, are identified to be major risk factors for severe COVID1914. Since HGPS share several phenotypes with accelerated aging especially atherosclerosis, osteoporosis, and cardiovascular diseases5,36,37, one would assume a severe COVID19-like phenotype from HGPS after viral infection. However, beyond our expectation, HGPS/hACE2 mice only experienced mild pathological outcomes compared to the young and aged mice, indicating that Progeria syndrome may not be a risk factor of severe COVID-19. A few studies investigating the association between genetic variants and COVID-19 susceptibility/severity, identified genes related to cytokines and viral receptors as risk-genes for severe COVID-1938,39,40, whereas progeria variants are not among them. Besides the mutation in LMNA gene, the deficiency of ZMPSTE24, an enzyme to initiate the maturation of Lamin A protein, could induce HGPS syndrome, as well41. It has been reported that ZMPSTE24 function downstream of interferon-induced antiviral pathways, defending against a broad spectrum of enveloped virus42,43. Therefore, using a ZMPSTE24-deficiency progeria syndrome model for COVID-19 study may be worth further investigation41,44.

Recently a poster abstract at The Progeria Research Foundation 11th International Scientific Workshop revealed that, mortality from COVID-19 in the progeria community was twice as high as in the general population despite fewer COVID-19 infections, indicating a severe COVID-19 phenotype from progeria patients. We reason that this discrepancy could be, firstly, likely due to the different genetic background or immune response of HGPS between human and mouse. Second, we also realized SARS-CoV-2 infection on mice with humanized ACE2 receptor may not fully phenocopy the complexity of natural viral infection in mice or COVID-19 in human45. Therefore, using a HGPS mouse model with a different background and a mouse adapted strain of SARS-CoV-2 for infection could represent the natural viral infection procedure and may give a different infection outcome. Additionally, we also noticed that the other uninfected HGPS hACE2 mice, which are from the same batch of tetraploid complementation as the infected ones, were all died nearly at the end of infection date, suggesting that these infected mice were likely at the end of their lives as well. This could also explain the slight decline of body weight from both uninfected and infect HGPS/hACE2 mice. Thus, viral infection with an early stage of HGPS mice, e.g. at 1 or 1.5 month ago, may worth testing for severe COVID-19 as well as accelerated aging, since several studies reported that COVID-19 is associated with cellular senescence and accelerated epigenetic aging46,47,48. Notably, HGPS mutations often lead to nuclear membrane deformations and subsequent loss of heterochromatin49. Therefore, investigating the specific epigenetic modifications altered in viral-infected HGPS mice warrants further attention and research.

Overall, our study assesses the potential pathological effects triggered by SARS-CoV-2 infection in HGPS, providing valuable insights for understanding progeria syndrome and guiding treatment for SARS-CoV-2 and other variants infected patients with premature aging.

Materials and methods

Ethics statement

Mouse studies were carried out in an animal biosafety level 3 (ABSL3) facility at Wuhan Institute of Virology, Chinese Academy of Sciences (CAS). All the animal experiments in this study were approved by The Institutional Animal Care and Use in GIBH and the Institutional Review Board of the Wuhan Institute of Virology, CAS. All the procedures involving mice is complied with all relevant ethical regulations (Experimental approvement number, IACUC 2020120). All animal experiments were performed in accordance with ARRIVE guidelines.

Cell culture

Balb/c mouse embryonic stem cells (ESCs) were derived from dpc 3.5 mouse embryos, and were cultured on feeder cells in Knockout DMEM (Gibco, 10829018) supplemented with 1000 U/mL LIF (Novoprotein, C690), 10% KSR (Gibco, 10828028), NEAA (Gibco, 11140076), Glutamax (Gibco, 35050079), Sodium pyruvate (Gibco, 11360070), beta-mecaptomethanol (sigma, M3148), MEK inhibitor PD0325901 (Holzel Biotech, DC1056, 1 mM) and GSK3 inhibitor CHIR99021 (Holzel Biotech, DC1023, 3 mM).

Generation of Lmna mutant hACE2 mESCs

Lmna c.1827 C < T variant was introduced into a huminzed ACE2 mES cell line generated from our previous study with slight modification19. Briefly, the codon-optimized hACE2 gene was inserted into mouse genome following the same strategy as before19. To target Lmna c.1827, sgRNA (5′-aggagatggatccgcccacc-3′) along with linear targeting donor were transfected into hACE2 mESCs by electroporation. Single colonies were picked 2 days after transfection, and genomic DNA were extracted using DirectPCR Lysis Reagent (VIAGEN, 102-T). Exon 11 of Lmna was amplified using primer set (Forward primer: 5′-agtcagtcccaaactcgctg-3′; Reverse primer: 5′-caagagggactgcaaggagg-3′) and the c.1827 C < T variant was proved by sanger sequencing.

Generation of mouse models (tetraploid complementation)

Mouse tetraploid embryos used for tetraploid complementation were prepared as reported before19. In detail, mESCs cultured at Day 2 were trypsinized until small clumps of cells (15–20 cells per clumps) were seen under the microscope, and then were transferred into microdrops of KSOM medium with 10% FCS under mineral oil. Each clump was placed in a depression in the microdrop. Meanwhile, batches of 30–50 embryos are incubated briefly in acidified Tyrode's solution to dissolve the zona pellucida. Next, two embryos are placed per ES clump for aggregation. All aggregates were incubated overnight at 37C, 5% CO2. After 24 h of culture, eleven to fifteen embryos are transferred into one uterine horn of a 2.5 dpc pseudopregnant female mouse. Mature CD-1 females are used as pseudopregnant foster mothers with a weight of about 30 g.

Mice infection

The SARS-CoV-2 (IVCAS 6.7512) was prepared as reported before30. Male Balb/c mice with different genetic backgrounds (hACE2 and hACE2 plus Lmna c. 1827C < T) and different ages (2 months ago for young and HGPS hACE2 mice, more than 14 months ago for aged hACE2 mice) were treated with tribromoethanol (Avertin, 250 mg/kg) and intranasally infected with 1 × 105 TCID50 SARS-CoV-2 in 50 μL DMEM per mouse. The uninfected control mice were inoculated with DMEM only. Mice with different genetic backgrounds were randomly assigned to mock, 1, 3, 5, 7 dpi groups. Mice showing undetectable viral RNA in lung were regarded as a failed infection and were removed from the study. Mice were weighted and observed for clinical signs daily across the infection. 3 mice were euthanized using isoflurane followed by cervical dislocation at 1, 3, 5, 7 dpi per group. Lung tissues were collected for RNA and Hematoxylin and Eosin (H&E) staining. More details can be found in Fig. 2A and Table 1.

Table 1 An overview of viral infection setup.
Full size table

H&E and IHC staining

Lung samples were fixed with 4% paraformaldehyde, followed by paraffin embedment. Fixed lung tissues were cut into 3.5 μm sections for H&E staining and immunohistochemistry for the detection of SARS-CoV-2 N protein following the standard protocol as described before30. Briefly, For IHC, slides were deparaffinized and rehydrated, followed by heat-induced antigen retrieval using EDTA pH 8.0 in a microwave oven for 15 min. After washing with PBS/0.02% Triton X-100, the slides were blocked with 5% BSA at room temperature for 1 h. Then the slides were first incubated with a primary antibody (rabbit anti-RP3-CoV N protein polyclonal antibody, 1:500, made in-house), followed by a secondary incubation of Cy3-conjugated goat-anti-rabbit IgG (Abcam, ab6939) at 1:200 dilution. After washing with PBS, slides were stained with DAPI (Beyotime) at 1:100 dilution. The images were collected using a Pannoramic MIDI system (3DHISTECH, Budapest), and visualized using CaseViewer (2.4).

Transcriptome analysis

RNA was isolated from homogenized lung tissue using TRizol with standard protocol. For ployA based mRNA-seq, 1 μg total RNAs were used for library construction. mRNA was enriched and purified using Library Preparation VAHTSTM mRNA Capture Beads (Vazyme, NR401-01). Next, purified mRNAs were fragmented, followed by cDNA synthesis, second-strand synthesis, end repair, adaptor ligation and amplification using VAHTS Universal V8 RNA-seq Library Prep Kit for Illumina (Vazyme, NR605-01) and VAHTS DNA Clean Beads. Libraries were sequenced for on average 20 million pair-end reads.

The RNA-seq data processing was performed as described below. To analyze the transcriptome changes, raw reads were first trimmed to remove the adapter contamination, and then aligned to the mouse 10 mm genome reference using STAR (2.7.10a) with default settings50. FeatureCounts (2.0.1) was used for read assignment with the following parameters, -T 5 -g gene_name -p51. DEseq2 (1.42.0) was used for data normalization, differential expression analysis, and data visualization52. Genes with P value < 0.05 and absolute value of log2(fold change) > 1 were considered as differentially expressed genes and used for GO analysis. Package ggplot2 (3.4.4) was applied for making the volcano plots53. Time-series analysis was done using Mfuzz package in R (3.18)54 with detectable genes (more than 10 reads were assigned from all the samples). Heatmaps were generated using pheatmap package (1.0.12) in R (RRID:SCR_016418). All data are submitted to GEO under accession GSE264189.

Western blotting

Mouse lung tissues were snap-frozen in liquid N2 and then homogenized in clod Cell lysis buffer (20 mM Tris–HCl, pH 7.5, containing 150 mM NaCl, 2 mM DTT, 50% triglyceride, 100 mM EDTA, 1% SDS, 1% NP40 and 1% Triton X-100) supplemented with Protease inhibitor cocktail (Roche, 4693132001). Total tissue extracts were separated on a 12.5% SDS-PAGE gel (EpiZyme, PG113), and transferred to a PVDF membrane (Millipore, IPVH00010). The following primary antibodies were used: anti-LAMINA/C (CST, 4777S, 1:1000), anti-progerin (abcam, ab66587, 1:1000), Histone 3 (abcam, ab18521, 1:5000). Original scan of the immunoblots were shown in supplementary File 1 with molecular mass markers and indicated cropped area. The immunoblot was precut into pieces for different hybridization.