A mouse model for MERS coronavirus-induced acute respiratory distress syndrome

Abstract

Middle East respiratory syndrome coronavirus (MERS-CoV) is a novel virus that emerged in 2012, causing acute respiratory distress syndrome (ARDS), severe pneumonia-like symptoms and multi-organ failure, with a case fatality rate of ∼36%. Limited clinical studies indicate that humans infected with MERS-CoV exhibit pathology consistent with the late stages of ARDS, which is reminiscent of the disease observed in patients infected with severe acute respiratory syndrome coronavirus. Models of MERS-CoV-induced severe respiratory disease have been difficult to achieve, and small-animal models traditionally used to investigate viral pathogenesis (mouse, hamster, guinea-pig and ferret) are naturally resistant to MERS-CoV. Therefore, we used CRISPR-Cas9 gene editing to modify the mouse genome to encode two amino acids (positions 288 and 330) that match the human sequence in the dipeptidyl peptidase 4 receptor, making mice susceptible to MERS-CoV infection and replication. Serial MERS-CoV passage in these engineered mice was then used to generate a mouse-adapted virus that replicated efficiently within the lungs and evoked symptoms indicative of severe ARDS, including decreased survival, extreme weight loss, decreased pulmonary function, pulmonary haemorrhage and pathological signs indicative of end-stage lung disease. Importantly, therapeutic countermeasures comprising MERS-CoV neutralizing antibody treatment or a MERS-CoV spike protein vaccine protected the engineered mice against MERS-CoV-induced ARDS.

Figure 1. A CRISPR–Cas9 genetically engineered mouse model for MERS-CoV replication.

a, C57BL/6J mice were genetically engineered using CRISPR–Cas9 genomic editing to encode 288L and 330R in mDPP4 on one chromosome (heterozygous, 288/330^+/−) or on both chromosomes (homozygous, 288/330^+/+). b, Northern blot of mDPP4 mRNA expression. c, Immunohistochemistry (IHC) of mDPP4 protein in the lungs, brain and kidneys of individual C57BL/6J wild-type (WT), 288/330^+/− and 288/330^+/+ mice. d, Viral titres for MERS-CoV at 3 days post-infection from C57BL/6J WT, 288/330^+/− and 288/330^+/+ (all n = 4) mice infected with 5 × 10⁵ plaque-forming units (p.f.u.) of the indicated viruses. Bar graphs show means + s.d. Student's t-test was used to calculate P < 0.05 for comparisons of MERS-0 with each virus in 288/330^+/+ mice (a–c on graph) and 288/330^+/− mice (d–f on graph). e, Pathology of 288/330^+/+, 288/330^+/− and C57BL/6J WT mice at day 3 after infection with MERS-0. Lung tissue sections were stained to examine the pathology by haematoxylin and eosin staining (H&E), or were stained by IHC to detect nucleocapsid protein from MERS-0 infection. IHC and H&E pathology images are representative of at least three samples. Scale bars (c,e), 1 mm.

Figure 2. Mouse-adapted MERS-CoV causes fatal disease in 288/330^+/+ mice.

Mice were inoculated intranasally with 5 × 10⁶ p.f.u. a, Mortality of 288/330^+/− (n = 10) and 288/330^+/+ (n = 16) mice was monitored daily up to day 6 p.i. Data show the percentage of surviving mice. b, Mouse weights were measured daily up to day 6 p.i. for 288/330^+/+ mice infected with MERS-15 (n = 16) or MERS-0 (n = 10), 288/330^+/− mice infected with MERS-15 (n = 10) and C57BL/6J wild-type (WT) mice infected with MERS-15 (n = 7). Data are daily means of the percentage weight relative to day 0 ± s.d. c, Viral lung titres for MERS-CoV were determined at day 3 p.i. (288/330^+/− + MERS-15, n = 4; 288/330^+/+ + MERS-15, n = 5; 288/330^+/+ + MERS-0, n = 5; C57BL/6J WT + MERS-15, n = 4) and day 6 p.i. (288/330^+/− + MERS-15, n = 4; 288/330^+/+ + MERS-15, n = 4; 288/330^+/+ + MERS-0, n = 5; C57BL/6J WT + MERS-15, n = 3). The limit of detection (LOD) is indicated. Bars are means + s.d. d, Immunohistochemistry of lung sections for anti-MERS nucleocapsid at 3 days p.i. Results show 288/330^+/+ mice + MERS-15 (i), 288/330^+/− mice + MERS-15 (ii), 288/330^+/+ mice + MERS-0 (iii) and C57BL/6J WT mice + MERS-15 (iv). IHC images are representative of at least three samples. Scale bar (d), 1 mm.

Figure 3. Lung function in MERS-15-infected mice.

a,b, Respiratory function was monitored in live mice up to day 6 p.i. using whole-body plethysmography to measure Penh (a) and EF₅₀ (b) in 288/330^+/+ mice infected with MERS-15 (n = 9) or MERS-0 (n = 3), 288/330^+/− mice infected with MERS-15 (n = 4) and C57BL/6J wild-type (WT) mice infected with MERS-15 (n = 3). Data are daily means ± s.d. Student's t-test was used to compare 288/330^+/+ mice infected with MERS-15 and MERS-0 (*P < 0.05; **P < 0.01). c, Pathology of lungs from infected mice at day 3 p.i. demonstrating severe inflammation for 288/330^+/+ (i) and 288/330^+/− (ii) mice infected with MERS-15, and moderate inflammation for 288/330^+/+ mice infected with MERS-0 (iii) and C57BL/6J WT mice infected with MERS-15 (iv). d, Pathology at day 6 p.i. infected in mice. In 288/330^+/+ mice infected with MERS-15, there was severe inflammation and oedema in the large airways (i) and alveoli (ii), and hyaline membrane formation (iii). The 288/330^+/− mice infected with MERS-15 exhibited severe inflammation throughout the parenchyma (iv), hyaline membrane formation (v) and perivascular cuffing (vi). The 288/330^+/+ mice infected with MERS-0 (vii) and C57BL/6J WT mice infected with MERS-15 (viii) exhibited mild-to-moderate inflammation. Samples were stained with haematoxylin and eosin and are representative of at least three samples. Scale bars (c,d), 1 mm.

Figure 4. Clonal isolates of mouse-adapted MERS-CoV exhibit severe respiratory disease.

Mice were inoculated intranasally with 5 × 10⁶ p.f.u. a, The mortality of 288/330^+/+ mice infected with MERS-15 C1 (n = 7) or MERS-15 C2 (n = 11) was monitored daily up to day 6 p.i. Data reflect the percentage of surviving mice. b, Mouse weights were monitored daily for 288/330^+/+ mice infected with MERS-15 C1 (n = 7), MERS-15 C2 (n = 11) or MERS-0 (n = 6), and C57BL/6J wild-type (WT) mice infected with MERS-15 C2 (n = 6). Data are daily means ± s.d. c, Viral lung titres were determined at day 3 p.i. (288/330^+/+ + MERS-15 C1, n = 4; 288/330^+/+ + MERS-15 C2, n = 3; 288/330^+/+ + MERS-0, n = 3; C57BL/6J WT + MERS-15 C2, n = 3) and day 6 p.i. (288/330^+/+ + MERS-15 C1, n = 4; 288/330^+/+ + MERS-15 C2, n = 6; 288/330^+/+ + MERS-0, n = 3; C57BL/6J WT + MERS-15 C2, n = 3). The limit of detection (LOD) is indicated. Bars are means + s.d. d, IHC of lung sections at 3 days p.i. from 288/330^+/+ mice infected with MERS-15 C1 (i) or MERS-15 C2 (ii) and stained for nucleopcapsid. The pathology of the lungs from 288/330^+/+ mice infected with MERS-15 C2 was assessed by haematoxylin and eosin staining (H&E) at day 6 p.i. and demonstrated severe inflammation (iii), oedema (iv), hyaline membrane formation (v) and perivascular cuffing (vi). All H&E images are representative of at least three samples. Scale bar (d), 1 mm.

Figure 5. Human monoclonal antibody 3B11 protects mice from severe respiratory disease.

The 288/330^+/+ mice were intraperitoneally administered 250 µg of 3B11 human monoclonal antibody or isotype control antibody 12 h before challenge with 5 × 10⁶ p.f.u. MERS-15 C2. a, b, Mortality (a) and mouse weight (b) were monitored daily for mice receiving 3B11 (n = 12) or an isotype control (n = 12) up to day 6 p.i. Data show the percentage of surviving mice (a) or daily mean weight ± s.d. (b). c, Viral lung titres were determined at day 3 p.i. (n = 6 for both 3B11- and isotype control-treated mice) and day 6 p.i. (n = 6 for 3B11-treated mice; n = 3 for isotype control-treated mice). The limit of detection (LOD) is indicated. Bars are means + s.d. d, Lung function was assessed by Penh at 0, 3 and 6 days p.i. for mice receiving 3B11 (n = 6 at days 0, 3 and 6) or isotype control (n = 6 at days 0 and 3; n = 3 at day 6). Data are means + s.d. Student's t-test was used to compare mice receiving 3B11 with those receiving the isotype control at day 3 and day 6 p.i. (*P < 0.05; **P < 0.01).

Figure 6. Vaccination of 288/330^+/+ mice with a VRP delivering MERS-CoV spike protein protects mice from challenge with MERS-CoV.

The vaccination protocol is described in Methods. a,b, After a 5 × 10⁶ p.f.u. challenge, mortality (a) and mouse weight (b) were monitored daily for mice receiving GFP–VRP (n = 19) or spike–VRP (n = 19) vaccine up to day 6 p.i. Data reflect the percentage survival (a) or daily mean weight ± s.d. (b). c, Viral lung titres were determined on day 3 p.i. (n = 7 for spike–VRP and GFP–VRP) and day 6 p.i. (n = 12 for spike–VRP; n = 6 for GFP–VRP). The limit of detection (LOD) is indicated. Bars are means + s.d. d, Lung function was assessed by Penh at 0, 3 and 6 days p.i. for mice receiving GFP–VRP (n = 12 at days 0 and 3 p.i.; n = 6 at day 6 p.i.) or spike–VRP (n = 12 at all days p.i.). Data represent means + s.d. Student's t-test was used to compare mice receiving GFP–VRP with those receiving spike–VRP at days 3 and 6 p.i. (*P < 0.05; **P < 0.01).