An orthopoxvirus-based vaccine reduces virus excretion after MERS-CoV infection in dromedary camels

Coronaviruses (CoVs) cause common colds in humans, but zoonotic transmissions occasionally introduce more pathogenic viruses into the human population. For example, the SARS-CoV caused the 2003 outbreak of severe acute respiratory syndrome (SARS). In 2012, a previously unknown virus, now named Middle East respiratory syndrome CoV (MERS-CoV), was isolated from the sputum of a 60-year-old Saudi Arabian man who suffered from acute pneumonia and subsequently died (1, 2). Several infection clusters have been reported over the past 3 years in the Middle East and also in South Korea, with ~35% of the reported human cases being fatal (3, 4). Dromedary camels (Camelus dromedarius) were suspected to be the reservoir host after neutralizing antibodies to MERS-CoV were detected in these animals (5–8). Subsequently, the virus detected in nasal swabs from these animals was found to be similar to that in human MERS cases associated with farms where the dromedaries were kept (9, 10). In addition, Chan et al. determined that MERS-CoV from dromedary camels replicates in human lung sections cultured ex vivo (11). More recent studies also provide serological evidence for camel-to-human transmission (12, 13). Because of its widespread presence in dromedary camels (14–16), zoonotic infections of MERS-CoV in humans will continue to occur. Therefore, strict implementation of quarantine and isolation measures, as well as the development of candidate vaccines and antivirals, is urgently needed.

The spike protein is considered to be a key component for vaccines against CoV infections. The identification of dipeptidyl peptidase 4 (DPP4) as the MERS-CoV receptor (17) has facilitated the subsequent characterization of the receptor binding domain in the S1 region of the MERS-CoV spike protein (18, 19). When tested as a vaccine in mice, full-length spike protein of MERS-CoV expressed by modified vaccinia virus Ankara (MVA-S) induced high levels of circulating antibodies that neutralize MERS-CoV and limited lower respiratory tract replication in animals transduced with the human receptor DPP4 and inoculated with MERS-CoV (20, 21). MVA, a highly attenuated strain of vaccinia virus, serves as one of the most advanced recombinant poxvirus vectors in preclinical and clinical trials for vaccines against infectious diseases and cancer. As a proof of principle, we tested the protective efficacy of a MVA–MERS-CoV candidate vaccine in dromedary camels.

In dromedary camels, MERS-CoV replication is mainly restricted to the upper respiratory tract (22). Therefore, we inoculated four dromedary camels twice at a 4-week interval, with 2 × 10⁸ plaque-forming units (PFU) MVA-S administered in both nostrils via a mucosal atomization device, to disperse the vaccine on the nasal epithelium, and 10⁸ PFU MVA-S delivered intramuscularly in the neck of each animal (23). Similarly, four control animals received nonrecombinant wild-type MVA (MVA-wt) (n = 2) or phosphate-buffered saline (PBS) (n = 2). All animals vaccinated with the MVA-S vaccine developed detectable serum neutralizing MERS-CoV–specific antibody titers (Fig. 1A). No MERS-CoV–specific antibodies were detected in sera of the PBS- or MVA-wt–immunized control animals. The specificity of the antibody response was confirmed by enzyme-linked immunosorbent assay, using recombinant S1 protein (fig. S1). In addition, low levels of MERS-CoV–neutralizing antibodies (virus neutralization titer of 1:20 to 1:40) were detected 3 weeks after the boost immunization in the nasal swabs of three animals (Fig. 1B). Because a MVA-vectored vaccine was used, antibodies neutralizing MVA were also detected (Fig. 1C); these antibodies cross-neutralized camelpox virus (Fig. 1D). Camelpox virus infections occur frequently in dromedaries and cause severe disease that can be prevented by vaccines based on attenuated camelpox viruses (24).

Three weeks after the boost immunization, all dromedary camels were inoculated with 10⁷ median tissue culture infectious dose (TCID₅₀) of MERS-CoV via the intranasal route, using a mucosal atomization device. Upon challenge, the animals showed only mild clinical signs, which were mainly limited to a relatively small rise in body temperature in control-vaccinated animals 1 day after challenge (fig. S2). In addition, some dry mucus was observed in one of the nostrils of most animals after day 4, but from days 8 to 10 onward, all control-vaccinated animals exhibited a runny nose that was not observed in MVA-S–vaccinated animals (Fig. 2, A and B). Previous studies have shown that both experimentally and naturally infected dromedary camels may show nasal discharge after MERS-CoV infection (15, 22). We next tested nasal respiratory tract samples for the presence of infectious virus. Whereas MERS-CoV was found at high titers in all four control-vaccinated animals, mean viral titers in the animals that received the MVA-S vaccine were significantly reduced (Fig. 2C). At 4 days post-inoculation (dpi), an increase in MERS-CoV RNA level was noted in the MVA-S–vaccinated animals (Fig. 2D). At 6 dpi, one of the MVA-S–vaccinated animals excreted low levels of infectious virus (10³ TCID₅₀/ml) (Fig. 2C). Sequencing of the spike gene of this virus showed no amino acid changes in the receptor binding domain (fig. S3), which suggests that this virus did not emerge as a result of escape from vaccine-induced antibodies (Fig. 2C). Rather, the observation that this animal had no detectable MERS-CoV antibody response in the nasal swab at time of challenge may indicate that, for unknown reasons, priming with the MVA-S vaccine was less effective in this animal compared with the other vaccinated animals. Antibodies to MERS-CoV rapidly increased 8 dpi in control-vaccinated animals (fig. S4), consistent with the absence of infectious virus in the nasal swabs at that time (Fig. 2C). Low levels of viral RNA, but no infectious virus, were detected in rectal swabs after MERS-CoV challenge (fig. S5), but not in any of the sera tested.

To analyze pathological changes and viral replication in organs of the animals, we euthanized two animals per group and performed necropsies at 4 and 14 dpi. Gross pathology showed no substantial changes in the organs of any of the animals. However, at 4 dpi MERS-CoV RNA transcripts were detected in several organs of the control-vaccinated animals (Fig. 3A), although infectious virus particles were restricted to noses and tracheas (Fig. 3B). In the absence of infectious MERS-CoV, relatively high levels of viral RNA have also been observed in tissues of experimentally infected rhesus macaques and rabbits (25, 26). In contrast, infectious MERS-CoV particles were found at low levels in the noses of animals that had received the MVA-S vaccine (Fig. 3B). At 14 dpi, only viral RNA was detected, mainly in control-vaccinated animals (fig. S6).

Differences in upper respiratory tract viral replication between vaccinated groups were confirmed by MERS-CoV in situ hybridization (ISH) and immunohistochemistry (IHC). At 4 dpi, only a few cells in the nasal epithelium of MVA-S–vaccinated dromedaries stained positive for MERS-CoV RNA by ISH, as compared with cells from control vaccinated animals (Fig. 4, A and B). Viral replication in the control-vaccinated animals was consistent with histopathological analyses showing multifocal moderate rhinitis with multifocal epithelial necrosis, as well as lymphocytic and neutrophilic exocytosis (Fig. 4C). In the nasal submucosa, we observed edema and infiltrates with lymphocytes, neutrophils, plasma cells, and macrophages. In the trachea and bronchi, we noted infiltration in the lamina propria, as well as a multifocal mild tracheitis and bronchitis with epithelial necrosis and lymphocytic and neutrophilic exocytosis. In the lymph nodes and the tonsils, we detected follicular hyperplasia. Marked MERS-CoV antigen expression in the nasal epithelium was associated with the nasal lesions (Fig. 4C). Through the use of ISH, the presence of MERS-CoV RNA in the nasal cavity was confirmed in cells similar to those that scored positive by IHC (Fig. 4C). Furthermore, a few epithelial cells in the trachea and bronchus and those covering the palatum molle—as well as large stellate cells (consistent with dendritic cells) in the lymphoid tissue of the palatum molle, tonsils, and tracheal and cervical lymph nodes—were found to be positive for viral antigen by IHC (fig. S7). In contrast, in MVA-S–vaccinated animals the rhinitis was accompanied by less submucosal edema with antigen expression in some nasal cells (Fig. 4C). Eosinophilic granulocytes were not observed in the lungs of MVA-S–vaccinated animals challenged with MERS-CoV. In one MVA-S–vaccinated animal, viral antigen expression was found in a few dendritic-like cells in the lymphoid tissue of the palatum molle, tonsils, and tracheal and cervical lymph nodes, as well as in the gut-associated lymphoid tissue of the duodenum (table S1). At 14 dpi, we observed multifocal mild rhinitis, tracheitis, and bronchitis and follicular hyperplasia in the lymphoid tissue of control- and MVA-S–vaccinated animals. In the lungs of almost all animals, we detected multifocal mild infiltration of neutrophils, histiocytes, and lymphocytes that was not associated with viral antigen expression. In the other extrarespiratory tissues examined, we found no substantial morphological changes or viral antigen expression. Overall, these results indicate that vaccination of dromedary camels with MVA-S induces protective immunity resulting in reduction of excreted infectious MERS-CoV, without evidence for antibody-dependent enhancement of viral replication, as seen in feline CoV infection (27). Given the potential transient nature of mucosal immune responses, follow-up studies are needed to determine the longevity of the responses induced by the MVA-S vaccine, with respect to protection as well as antibody-dependent enhancement of viral replication when antibody levels are waning. In addition, dosing of the vaccine and alternative methods of administration must be explored in more detail before this candidate vaccine will be useful in the field.

Protective immunity to CoVs is orchestrated by antibody and cellular immune responses. Investigations in mice have already provided evidence that inoculation with MERS-CoV spike protein–based candidate vaccines, monoclonal antibodies directed against the spike protein, or dromedary immune serum induces protective immunity against lower respiratory tract MERS-CoV infection (28–30). In dromedary camels, a DNA vaccine encoding the spike protein induced MERS-CoV neutralizing antibody responses that were similar to antibody levels in animals inoculated with the MVA-S vaccine, but no challenge experiments were performed (31). However, studies in the field also indicated that MERS-CoV–seropositive dromedaries may carry MERS-CoV viral RNA in their nasal excretions (8, 15, 16). Thus, sterilizing immunity may not be possible to achieve, as virus replicates in the upper respiratory tract even in the presence of specific antibodies, similarly to other respiratory viruses. Because dromedary camels do not show severe clinical signs upon MERS-CoV infection, vaccination of dromedaries should primarily aim to reduce virus excretion to prevent virus spreading. Young dromedaries excrete more infectious MERS-CoV than adults (8, 15, 16), so young animals should be vaccinated first. Our results reveal that MVA-S vaccination of young dromedary camels may significantly reduce infectious MERS-CoV excreted from the nose. Two major advantages of the orthopoxvirus-based vector used in our study include its capacity to induce protective immunity in the presence of preexisting (e.g., maternal) antibodies (32) and the observation that MVA-specific antibodies cross-neutralize camelpox virus, revealing the potential dual use of this candidate MERS-CoV vaccine in dromedaries. Dromedary camels vaccinated with conventional vaccinia virus showed no clinical signs upon challenge with camelpox virus, whereas control animals developed typical symptoms of generalized camelpox (33). The MVA-S vectored vaccine may also be tested for protection of humans at risk, such as health care workers and people in regular contact with camels.