Immunogenicity of Candidate MERS-CoV DNA Vaccines Based on the Spike Protein

Abstract

MERS-coronavirus is a novel zoonotic pathogen which spread rapidly to >25 countries since 2012. Its apparent endemicity and the wide spread of its reservoir host (dromedary camels) in the Arabian Peninsula highlight the ongoing public health threat of this virus. Therefore, development of effective prophylactic vaccine needs to be urgently explored given that there are no approved prophylactics or therapeutics for humans or animals to date. Different vaccine candidates have been investigated but serious safety concerns remain over protein or full-length spike (S) protein-based vaccines. Here, we investigated the immunogenicity of naked DNA vaccines expressing different fragments of MERS-CoV S protein in mice. We found that plasmids expressing full-length (pS) or S1-subunit (pS1) could induce significant levels of S1-specific antibodies (Abs) but with distinct IgG isotype patterns. Specifically, pS1 immunization elicited a balanced Th1/Th2 response and generally higher levels of all IgG isotypes compared to pS vaccination. Interestingly, only mice immunized with pS1 demonstrated significant S1-specific cellular immune response. Importantly, both constructs induced cross-neutralizing Abs against multiple strains of human and camel origins. These results indicate that vaccines expressing S1-subunit of the MERS-CoV S protein could represent a potential vaccine candidate without the possible safety concerns associated with full-length protein-based vaccines.

Full text

1
Scientific RepoRts | 7:44875 | DOI: 10.1038/srep44875
www.nature.com/scientificreports
Immunogenicity of Candidate
MERS-CoV DNA Vaccines Based
on the Spike Protein
Sawsan S. Al-amri1, Ayman T. Abbas1,2, Loai A. Siddiq1, Abrar Alghamdi1,
Mohammad A. Sanki3, Muhanna K. Al-Muhanna4, Rowa Y. Alhabbab1,5, Esam I. Azhar1,5,
Xuguang Li6 & Anwar M. Hashem1,7
MERS-coronavirus is a novel zoonotic pathogen which spread rapidly to >25 countries since 2012.
Its apparent endemicity and the wide spread of its reservoir host (dromedary camels) in the Arabian
Peninsula highlight the ongoing public health threat of this virus. Therefore, development of eective
prophylactic vaccine needs to be urgently explored given that there are no approved prophylactics or
therapeutics for humans or animals to date. Dierent vaccine candidates have been investigated but
serious safety concerns remain over protein or full-length spike (S) protein-based vaccines. Here, we
investigated the immunogenicity of naked DNA vaccines expressing dierent fragments of MERS-CoV
S protein in mice. We found that plasmids expressing full-length (pS) or S1-subunit (pS1) could induce
signicant levels of S1-specic antibodies (Abs) but with distinct IgG isotype patterns. Specically,
pS1 immunization elicited a balanced Th1/Th2 response and generally higher levels of all IgG isotypes
compared to pS vaccination. Interestingly, only mice immunized with pS1 demonstrated signicant S1-
specic cellular immune response. Importantly, both constructs induced cross-neutralizing Abs against
multiple strains of human and camel origins. These results indicate that vaccines expressing S1-subunit
of the MERS-CoV S protein could represent a potential vaccine candidate without the possible safety
concerns associated with full-length protein-based vaccines.
Middle East respiratory syndrome coronavirus (MERS-CoV) is an emerging zoonotic pathogen recovered rst
from a fatal human case in Saudi Arabia in 20121 and continued to infect almost 1800 people in over 25 countries.
Saudi Arabia has reported the largest number of cases so far with cases continuing to increase. e virus causes
severe respiratory infection associated with fever, cough, acute pneumonia, shortness of breath, systemic infection
and occasional multi-organ failure in infected individuals leading to death in 35–40% of the cases2–4. Such a severe
disease usually occurs in immunocompromised patients, individuals with comorbidities and the elderly1,4–6. Most
of the reported MERS cases are linked to hospital outbreaks and family clusters due to close contact with infected
patients4,7–10. However, accumulating epidemiological data show high prevalence of MERS-CoV in dromedary
camels from several Arabian and African countries, suggesting that dromedaries might be the reservoir hosts of
this virus4,11–15. e continued endemicity of MERS-CoV in the Arabian Peninsula and the associated high death
rate clearly represent a public health concern with potential global spread as observed in the recent outbreak in
South Korea10. is is further complicated by the lack of prophylactic or therapeutic measures, underscoring the
importance of preparedness research against this potential pandemic virus.
Several supportive therapies and antivirals were proposed and examined for the treatment of MERS-CoV
infections16–20. However, most of these strategies were based on the experience gained during the severe acute
respiratory syndrome (SARS) outbreak or from MERS-CoV in vitro studies and require further preclinical and
1Special Infectious Agents Unit, King Fahd Medical Research Center, King Abdulaziz University, Jeddah, Saudi Arabia.
2Biotechnology Research Laboratories, Gastroenterology Surgery Center, Mansoura University, Mansoura, Egypt.
3Hematology Laboratory, King Abdulaziz University Hospital, Jeddah, Saudi Arabia. 4Materials Science Research
Institute, National Nanotechnology Center, King Abdulaziz City for Science and Technology (KACST), Riyadh,
Saudi Arabia. 5Department of Medical Laboratory Technology, Faculty of Applied Medical Sciences, King Abdulaziz
University, Jeddah, Saudi Arabia. 6Center for Vaccine Evaluation; Biologics and Genetic Therapies Directorate; Health
Canada; Ottawa, Ontario, Canada. 7Department of Medical Microbiology and Parasitology, Faculty of Medicine,
King Abdulaziz University, Jeddah, Saudi Arabia. Correspondence and requests for materials should be addressed to
A.M.H. (email: amhashem@kau.edu.sa)
Received: 13 September 2016
Accepted: 15 February 2017
Published: 23 March 2017
OPEN

www.nature.com/scientificreports/
2
Scientific RepoRts | 7:44875 | DOI: 10.1038/srep44875
clinical evaluation. e ideal strategy to rapidly control existing and potential outbreaks of MERS-CoV is to gen-
erate a safe and eective vaccine at least to target high-risk groups or animal hosts. e ability of more than 60%
of the infected patients to recover, clear the virus and develop immunity suggest that a vaccine based on the viral
components such as the spike (S) glycoprotein could be a suitable vaccine candidate. is is further supported by
the isolation of several human neutralizing antibodies (nAbs) against the MERS-CoV S protein and their ability
to neutralize and block viral entry and/or cell-cell spread at very low concentrations, and sometimes to confer
prophylactic and therapeutic protection in animal models21–27.
MERS-CoV S glycoprotein is composed of 2 subunits; the receptor binding domain (RBD) containing subunit
(S1) and the fusion machinery subunit (S2)28. Several vaccines candidates based on full-length or truncated S pro-
tein were developed and investigated including DNA vaccines29,30, viral vectored vaccines31–35, nanoparticle-based
vaccine36, whole inactivated MERS-CoV vaccine (WIV)37, as well as the S or RBD protein-based subunit vac-
cines29,38–42. While these experimental vaccines can induce protective response in animals, SARS-CoV vaccine
development and a recent MERS-CoV report37 suggest that there might be serious safety concerns associated
with the use of full length S protein as vaccine candidate including immunopathology and disease enhance-
ment43–48. ese concerns were proposed to be due to inductions of 2- skewed immune response and/or anti-S
non-neutralizing Abs.
DNA vaccines represent a promising vaccine development approach due to their easy production on a large
scale in a timely manner and well-established procedures for quality control. In addition, DNA vaccines can elicit
1-biased immune response in contrast to the protein-based subunit vaccines. However, all MERS-CoV DNA
vaccines reported so far were aimed at expressing full-length protein, which could induce adverse reactions. In
this study, we determined the immunogenicity and potential protective eects of MERS-CoV naked DNA vac-
cines expressing dierent length of S protein.
Materials and Methods
Cell line and MERS-CoV viruses. African Green monkey kidney-derived Vero E6 cells (ATCC #1568)
were grown in Dulbecco’s modied Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS),
1% penicillin/streptomycin, and 10 mM HEPES (pH 7.2) and maintained in a humidied 5% CO2 incubator
at 37 °C. MERS-CoV strains used in this study included a human isolate (MERS-CoV/Hu/Taif/SA/2015) and
two camel isolates (MERS-CoV/Camel/Taif/SA/31/2016 and MERS-CoV/Camel/Taif/SA/39/2016). MERS-CoV
viruses were isolated, passaged and titrated by TCID50 in Vero E6 cells as previously described49. All tested isolates
were at passage no. 2. All experiments involving live virus were conducted in our Biosafety level 3 facility follow-
ing the recommended safety precautions and measures.
DNA constructs. Four DNA vaccine candidates were generated as shown in Fig.1a. Full length
MERS-CoV S gene from MERS-CoV-Jeddah-human-1 isolate (GenBank accession number: KF958702)
was codon optimized for efficient mammalian expression and synthesized by Bio S & T (Montreal,
Canada). The coding sequence was then subcloned into the mammalian expression vector pcDNA3.1
under the control of the cytomegalovirus immediate-early promoter generating pS construct. The sec-
ond construct (pS1) expressing S1 domain (aa 1–747) was produced by cloning corresponding coding
region by PCR using Phusion High-Fidelity PCR Kit (Life Technologies) from pS plasmid into pcDNA3.1
vector using the following forward 5′ -GATCGCGGC CGCGCCACCATGATCCAC-3′ and reverse
5′ -GATCGGTACCTTACAGAATGAAAAAGACGC-3′ primers. Similarly, pS∆ TM expressing truncated S pro-
tein (aa 1–1295) without the transmembrane domain and pS∆ CD expressing truncated S protein (aa 1–1318) with-
out the cytoplasmic domain were generated by PCR subcloning into pcDNA3.1 vector using the above-mentioned
forward primer and the following reverse primers; 5′ -GATCGGTACCTTACCACTTGTTGTAGTATG-3′ and
5′ -GATCGGTACCTTACAGAATGAAAAAGACGC-3′ , respectively. All constructs were cloned between NotI
Figure 1. MERS-CoV Spike DNA vaccines. (a) Schematic representation of the generated DNA vaccine
constructs. Four constructs were generated including one expressing full length S protein (pS) and three other
constructs expressing truncated S protein with deleted cytoplasmic domain (pS∆ CD), deleted transmembrane
domain (pS∆ TM) or deleted S2 subunit (pS1). Numbers indicate amino acids. SP: signal peptide; RBD:
receptor-binding domain; TM: transmembrane domain; CD: cytoplasmic domain. (b) In vitro protein
expression in cell culture. Vero E6 cells with 80–90% conuency were transfected with the DNA constructs; 48 h
later, cell lysates were collected; protein expression was subsequently conrmed by western blot using anti-S1
polyclonal Abs. Arrows indicate band with expected molecular weight. (c) Time-line of immunization regimen.

www.nature.com/scientificreports/
3
Scientific RepoRts | 7:44875 | DOI: 10.1038/srep44875
and KpnI restriction sites in pcDNA3.1 vector using the T4 DNA ligase. All constructs were conrmed by restric-
tion digestion and sequencing. Bulk endotoxin-free preparations of all four constructs as well as the empty con-
trol plasmid (pcDNA) were prepared for animal studies using a plasmid Giga purication kit (Qiagen).
In vitro Protein expression. Prior to animal experiments, protein expression from all DNA constructs
was conrmed in vitro in Vero E6 cells (Fig.1b). Briey, 80–90% conuent Vero E6 cells in 6-well plates were
transiently transfected with 1 μ g of each DNA construct (pS, pS∆ CD, pS∆ TM, pS1, or pcDNA) using FuGENE 6
reagent (Roche) according to manufacturer’s instructions, followed by incubation at 37 °C in a 5% CO2 incubator
for 48 h. Transfected cells were then washed twice with phosphate-buered saline (PBS) and lysed with cell lysis
buer as previously described50, and subjected to western blot analysis for protein expression using rabbit anti-S1
Abs (Sino Biological). Western blot analysis conrmed that all gene products show bands at expected molecular
weights. Notably, the large band that is observed in all blots in Fig.1b is due to non-specic binding as it was also
detected in an un-transfected cell control (data not shown).
Animal Studies. Six- to 8-week-old female BALB/c mice were obtained from the core facility in King Fahd
Medical Research Center (KFMRC), King Abdulaziz University (KAU). All animal experiments were conducted
in accordance with institutional guidelines and the approval of the Animal Care and Use Committee at KFMRC.
Mice were divided into ve experimental groups (5 mice in each group) and immunized on days 0, 14 and 28
with three doses of 100 μ g of each construct dissolved in 100 μ l PBS. Mice were immunized intramuscularly with
two injections (50 μ l each) divided between the two thighs. ree weeks aer the last doses (day 49), mice were
euthanized and blood as well as spleens were collected for immune response analysis.
ELISA. e end-point titers of anti-S1 total IgG Ab as well as IgG1, IgG2a and IgG2b isotypes from immu-
nized mice were determined by ELISA as described previously29,50 with minor modications. Briey, 96-well
plates (EU Immulon 2 HB, ermo Scientic) were coated with the MERS-CoV S1 protein (Sino Biological) at
2 μ g/ml in PBS at 4 °C overnight. Plates were then washed 6 times with PBS containing 0.05% Tween-20 (PBS-T),
followed by blocking with 5% skim milk in PBS-T for 1 h at 37 °C. Aer washing, plates were incubated with a
2-fold serial dilution of mouse sera starting from 1:100 and incubated for 1 h at 37 °C. en, plates were washed
and incubated with peroxidase-conjugated rabbit anti-mouse IgG, IgG1, IgG2a or IgG2b secondary Abs (Jackson
Immunoresearch Laboratories) at concentrations recommended by the supplier and incubated for additional 1 h
at 37 °C. Aer extensive washing, Tetramethylbenzidine (TMB) substrate (KPL) was added for 30 min for color-
imetric development and the reaction was stopped with 0.16 M sulfuric acid. Absorbance was read spectropho-
tometrically at 450 nm. End-point titers were determined and expressed as the reciprocals of the nal detectable
dilution with a cut-o dened as the mean of pre-bleed samples plus three SD.
Viral microneutralization assay. Microneutralization (MNT) assay was performed as previously
described29,30. Briey, two-fold serial dilutions of heat-inactivated sera prepared in DMEM starting from a 1:5
dilution were incubated with equal volume of DMEM containing 200 TCID50 of MERS-CoV for 1 h at 37 °C
in a 5% CO2 incubator. e virus-serum mixture was then transferred on conuent Vero E6 cell monolayers in
96-well plates (four wells were used per dilution) and incubated at 37 °C in a 5% CO2 incubator. Cytopathic eect
(CPE) was observed on days 3 to determine nAb titer. e nAb titer for each sample is reported as the reciprocal
of the highest dilution that completely protected cells from CPE in 50% of the wells (MNT50).
CD8+ T cell intracellular cytokine staining (ICS). Memory CD8+ T cell IFN-γ responses were evalu-
ated at 3 weeks aer last immunizations as previously described50. Briey, single-cell suspensions of splenocytes
were prepared from individual mice in each group. Spleens from mice were collected in 10 ml of RPMI 1640
supplemented with 10% FBS and smashed between frosted ends of two glass slides. Processed splenocytes were
then ltered through 45-μ m nylon lters and centrifuged at 800 g for 10 min. Red blood cells were then lysed by
adding 5 ml of ammonium-chloride-potassium (ACK) lysis buer (Life Technologies) for 5 min at room tem-
perature, and equal volume of PBS was then added. Cells were centrifuged again and pellets were resuspended
in RPMI 1640 at a concentration of 1 × 107 cells/ml. Splenocytes were then added to a 96-well plate (1 × 106 per
well) and re-stimulated with 5 μ g/ml of several synthetic S1 MHC class I–restricted peptides including S291
(KYYSIIPHSI), S319 (QPLTFLLDF), S448 (YPLSMKSDL), S498 (SYINKCSRL), S647 (NYYCLRACV), S703
(TYGPLQTPV), which were synthesized by GenScript as previously described32. e stimulation was conducted
by incubation for 6 h at 37 °C and 5% CO2 in the presence of Protein Transport Inhibitor Cocktail (brefeldin A)
(BD Biosciences) according to the manufacturer’s instructions. Stimulated cells were then washed in FACS buer
and stained with LIVE/DEAD Fixable Violet Dead Cell Stain Kit (Invitrogen) and anti-mouse CD8α –FITC
antibody (clone 53–6.7; eBiosciences). e cells were then washed with FACS buer, xed and permeabilized
with Cytox/Cytoperm Solution (BD Biosciences) according to the manufacturer’s instructions, and labeled with
anti–mouse IFN-γ -APC-Cy7 antibody (clone XMG1.2; BD Biosciences). All data were acquired on a BD FACS
Calibur ow cytometer and analysis was completed with Flow Jo, Version 8.8.4 (Tree Star Inc.). Results for IFN-γ
producing CD8+ T cells were calculated as percentage of live CD8+ T cells aer subtracting the values obtained
from no peptide controls from each sample.
Data analysis. Statistical analysis was conducted using one-way ANOVA. Bonferroni post-test was used
to adjust for multiple comparisons between the dierent groups. All statistical analysis was conducted using
GraphPad Prism soware (San Diego, CA). P values < 0.05 were considered signicant.

www.nature.com/scientificreports/
4
Scientific RepoRts | 7:44875 | DOI: 10.1038/srep44875
Results
S1-subunit DNA vaccine induces high levels of anti-S1 Abs in mice. In order to evaluate the immu-
nogenicity of our DNA vaccine candidates, we immunized mice i.m. with three doses of the generated naked
DNA constructs (Fig.1c). To this end, evaluation of Ab levels aer one or two doses of naked DNA resulted in no
or barely detectable response in all groups consistent with previous report29, and thus we only analyzed responses
aer the last dose. As shown in Fig.2a, only mice immunized with pS and pS1 but not pS∆ TM generated signif-
icant levels of systemic S1-specic IgG compared to control group immunized with pcDNA vector. Interestingly,
pS1 elicited signicantly higher levels of S1-specic total IgG compared to pS immunized mice. It is of note that
DNA construct expressing truncated S protein without the cytoplasmic domain (pS∆ CD) failed to induce detect-
able Abs in initial pilot studies; it was not tested further.
Dierential induction of S1-specic IgG isotypes by Spike-based DNA vaccines. We next exam-
ined the dierences in S1-specic Ab isotypes in the sera of immunized mice in order to determine the quality
of the humoral response induced by the dierent DNA constructs. As shown in Fig.2b–e, immunization with
pS DNA vaccine mainly elicited IgG2a and IgG2b with signicantly lower levels of IgG1 isotype, indicating a
1-biased response (IgG2a/IgG1 ratio of > 1.5). As expected, empty vector control (pcDNA) and pS∆ TM failed
to produce any anti-S1 IgG isotype. On the other hand, plasmid DNA expressing S1 subunit (pS1) induced a
balanced 1/2 response (IgG2a/IgG1 ratio of ~1.0) with S1-specic Abs from all isotypes. While IgG2a and
IgG2b levels induced by pS1 were signicantly higher compared to pS∆ TM and empty vector control (pcDNA),
no signicant dierence was observed in the levels of these two isotypes between pS and pS1-vaccinated groups.
In contrast, level of S1-specic IgG1 Abs elicited by pS1 vaccine was signicantly elevated compared to all groups
including pS group. Collectively, compared with the full length S protein, these data suggest that S1 subunit deliv-
ered by DNA vector elicited stronger antibody responses and equal ratio of IgG2a/IgG1 whereas the full length S
protein induced a 1-skewed immune response.
S1-expressing DNA vaccine elicits significant level of IFN-γ response. Having observed the
1-skewed response in pS-immunized mice compared to the pS1 group, we decided to evaluate S1-specic
memory CD8+ T cell responses by ICS. Remarkably, immunization of mice with pS vaccine did not elicit any
signicant levels of IFN-γ compared to control group (pcDNA) aer re-stimulation with S291 peptide (Fig.3).
On the other hand, re-stimulation of CD8+ T cells from pS1-vaccinated animals induced signicantly higher
levels of IFN-γ compared to all other groups, suggesting that immune-focusing by using S1-based vaccine
could not only enhance Ab response but also cell-mediated responses. e inability of pS immunogen to induce
S1-specic CD8+ T cells IFN-γ was consistent with the overall weaker response compared to pS1-vaccinated
group. Interestingly, re-stimulation with several other peptides within the S1 subunit as previously described32
failed to elicit any IFN-γ from all groups (SupplementaryFigure1).
Spike-based DNA immunization elicited cross-neutralizing MERS-CoV Abs against human and
camel isolates. As our DNA vaccine constructs were made using coding sequence from a 2013 isolate that
directly transmitted form infected camel to a human, it was important to test their cross-neutralization activ-
ity against recent isolates. To this end, antisera from immunized mice were tested against human and camel
MERS-CoV isolates from 2015 and 2016. As shown in Fig.4, sera collected from mice immunized with DNA
expressing full-length S protein or S1 subunit were found to have comparable nAb titers against the human and
the camel isolates. ese ndings clearly show that S protein is a very promising vaccine target as it induced nAbs
against human and camel MERS-CoV strains isolated in 2015 (MERS-CoV Human/1390) and 2016 (MERS-CoV
Camel/31 and MERS-CoV Camel/39).
Figure 2. Humoral immune response induced by MERS-CoV Spike DNA vaccines. Circulating MERS-
CoV S1-specic Abs were determined at 3 weeks post 2nd boost. End-point titers are shown for (a) total IgG,
(b) IgG1, (c) IgG2a and (d) IgG2b isotypes. (e) IgG1/IgG2a ratio was calculated at 3 weeks aer 2nd boosting
to determine the type of immune response (2 versus 1) induced by the various constructs. BALB/c mice
were i.m. immunized with 100 μ g of each construct dissolved in 100 μ l PBS on days 0, 14 and 28. A control
group was immunized with empty pcDNA vector. Data are shown as mean titer ± s.d. from one experiment
out of two independent experiments, with n = 5 mice per treatment group in each experiment. ****P < 0.0001,
***P < 0.001, **P < 0.01 and *P < 0.05 (one-way ANOVA with Bonferroni post-test).

www.nature.com/scientificreports/
5
Scientific RepoRts | 7:44875 | DOI: 10.1038/srep44875
Discussion
e rapid spread and high mortality rate of MERS-CoV infections in several countries of the Arabian Peninsula
present a daunting challenge to the international community; the large zoonotic reservoir host of MERS-CoV
makes it dicult to eliminate the source of transmission. While public health measures are critical to contain
MERS-CoV spread and proven to be eective in limiting outbreaks, development of safe and preventive vaccine
is urgently needed.
Several groups have investigated various vaccine platforms to combat MERS-CoV29–42. Most of these experi-
mental vaccines were based on MERS-CoV full-length or truncated versions of the spike protein; these prototype
vaccines were found to have induced high levels of nAbs and sometimes conferred protection against MERS-CoV
challenge in several animal models. However, several previous SARS-CoV vaccine studies have also shown that
there might be some safety concerns associated with the use of WIV43, truncated S subunit/protein vaccines44 or
vectored vaccines expressing full-length S protein45. ese concerns included inammatory and immunopatho-
logical eects such as eosinophilic inltration of the lungs as well as Ab-mediated disease enhancement (ADE) in
immunized animals upon viral challenge. It is believed that induction of 2-polarized immune response and/or
non-neutralizing Abs against epitopes within the S protein (i.e. outside the neutralizing-epitope rich RBD or S1
subunit) are the reason for the observed immunopathology and disease enhancement in vaccinated animals46–48,
suggesting that use of S1 subunit over full-length S protein could be a safer option for vaccine development.
Furthermore, a recent report revealed that MERS-CoV vaccines might be associated with similar type of
immunopathologies especially upon induction of 2-skewed response or use of full-length or truncated S pro-
tein37. is could be a hurdle facing vaccine candidates expressing non-neutralizing epitopes such as the ones
based on full-length S protein29. erefore, immune focusing by using RBD or S1 subunit could represent an
attractive approach for safe and eective MERS vaccine. Indeed, several versions of RBD subunit vaccine were
tested38–42 and showed very promising results even upon immunization with very low dose51. However, protein
Figure 3. MERS-CoV Spike-specic memory CD8+ T cell responses. Immunized BALB/c mice were
sacriced at 3 weeks aer 2nd boosting and splenocytes were isolated and re-stimulated ex vivo with synthetic
S1 peptides for IFN-γ measurement by ICS. Live CD8+ T cells were stained for intracellular IFN-γ . (a) Flow
cytometry plots are representatives from one out of two independent experiments. (b) Bar graph represents
frequencies of IFN-γ memory CD8+ T cells. Data are shown as mean ± s.d from one experiment out of two
independent experiments, with n = 3 mice per treatment group in each experiment. **P < 0.01 (one way
ANOVA with Bonferroni post-test).
Figure 4. MERS-CoV Spike DNA vaccine induced nAbs. Neutralization titers were determined as the highest
serum dilutions from each individual mouse that completely protected Vero E6 cells in at least 50% of the wells
(MNT50). Titers are shown as means from 5 mice per group ± s.d from one experiment out of two independent
experiments.

www.nature.com/scientificreports/
6
Scientific RepoRts | 7:44875 | DOI: 10.1038/srep44875
subunit vaccines were found to induce skewed 2 response. erefore, more studies are needed to develop a safe
and approved adjuvant to elicit 1-skewed response29,47,48.
MERS-CoV DNA vaccines can induce 1-biased immune response even though multiple injections are usu-
ally required due to their low immunogenicity especially in large animals29,30. Up to date, only two studies have
investigated MERS-CoV DNA vaccines by utilizing full-length S protein, which is the primary target of immune
response in the host. To dissect the antigenic domains of the S protein, we examined the immunogenicity of
naked DNA vaccines expressing several versions of MERS-CoV S protein in mice. We found that pS-immunized
group elicited signicant IgG2a and IgG2b titers (1-skewed response) with very subtle S1-specic CD8+ IFN-γ
response. On the other hand, pS1-immunization generated markedly increased levels of all IgG isotypes in a bal-
anced 1/2 response along with low but signicantly elevated CD8+ IFN-γ response compared to pS group.
While further animal studies are required to determine whether induction of balanced 1/2 or 1-biased
immune response could aid in the development of safer MERS-CoV vaccine, S1-based vaccines could be a safer
option compared to the full-length S-based vaccines.
It is of note that the ELISA plates used for the measurement of binding IgG isotypes were coated with S1
recombinant protein (Fig.2), therefore, there might be more Abs in the pS vaccinated animals targeting epitopes
outside the S1 subunit (i.e. S2 subunit) that were never detected in our analysis. In addition, the observation
that both pS1 and pS induced similar nAb titers (Fig.4) suggests that full-length S protein harbors neutralizing
epitopes outside the S1 region as previously reported29. e induction of high levels of IgG1 by pS1 vaccine and
consequent balanced 1/2 response could probably be explained by the secretion of S1 subunit especially that
this immunogen contains the signal peptide without the cell membrane anchoring domains compared to pS vac-
cine. While additional studies are required to conrm this, we have previously shown that internal viral proteins
such as the inuenza nucleoprotein could be partially secreted and alter the immune response phenotype when
fused to a secretion signal50. Furthermore, the weak or undetectable response in pS∆ TM and pS∆ CD immunized
mice is noteworthy especially that Wang et al., showed that MERS-CoV DNA vaccine expressing S∆ TM induced
limited response in mice even aer electroporation29. Although this response could be due to misfolded protein
and rapid degradation of the antigen, or low expression level of these truncated spike proteins given that expres-
sion of S∆ TM gave low production yields from transfected HEK 293 as previously described29, similar vaccines
have been shown to be very eective in mice in the case of SARS-CoV52.
The finding that immune sera from both pS and pS1 immunized mice could cross-neutralize recent
human and camel eld isolates is critical. Most previous studies utilized strains such as Jordan-N3 (GenBank
ID: KC776174.1) and EMC/2012 (GenBank ID: JX869059.2) in live virus neutralization assay. ese viruses
were isolated in 2012; they may or may not be same as the currently circulating strains, given that strains
used here showed 5–7 and 1–2 amino acid changes in comparison to Jordan-N3 and EMC/2012, respectively
(SupplementaryFigure2). Furthermore, several other studies have used pseudovirus neutralization assay to test
contemporary strains which may not replicate the actual neutralization breadth against live viruses29,30. It is of
note that the similar levels of nAb titers in both pS and pS1 groups reported here appear to be dierent from
the observation by others, who found signicantly higher levels of nAbs induced by a DNA vaccine expressing
full-length S protein compared to that expressing S1 protein29. However, this discrepancy in results remains to be
fully understood but is likely due to the dierence in experimental conditions. Specically, Wang et al. utilized
electroporation with DNA immunization and pseudovirus neutralization assays to determine nAbs whereas we
used naked DNA vaccines and live virus neutralization assays.
Taken together, our study suggests the DNA vaccine expressing S1 subunit could represent a promising can-
didate vaccine against MERS-CoV while minimizing the risk of the immunopathologies associated with the use
of full S protein and 2 response. However, more studies are clearly required to enhance the immunogenicity of
naked DNA vaccine and to examine the safety of this prototype vaccine.
References
1. Zai, A. M., van Boheemen, S., Bestebroer, T. M., Osterhaus, A. D. & Fouchier, . A. Isolation of a novel coronavirus from a man
with pneumonia in Saudi Arabia. N. Engl. J. Med. 367, 1814–1820, doi: 10.1056/NEJMoa1211721 (2012).
2. Bermingham, A. et al. Severe respiratory illness caused by a novel coronavirus, in a patient transferred to the United ingdom from
the Middle East, September 2012. Euro. Surveill. 17, 20290 (2012).
3. Pebody, . G. et al. e United ingdom public health response to an imported laboratory conrmed case of a novel coronavirus in
September 2012. Euro. Surveill. 17, 20292 (2012).
4. Macay, I. M. & Arden, . E. MES coronavirus: diagnostics, epidemiology and transmission. Virol. J. 12, 222, doi: 10.1186/s12985-
015-0439-5 (2015).
5. Alraddadi, B. M. et al. is Factors for Primary Middle East espiratory Syndrome Coronavirus Illness in Humans, Saudi Arabia,
2014. Emerg. Infect. Dis. 22, 49–55, doi: 10.3201/eid2201.151340 (2016).
6. Lessler, J. et al. Estimating the Severity and Subclinical Burden of Middle East espiratory Syndrome Coronavirus Infection in the
ingdom of Saudi Arabia. Am. J. Epidemiol. 183, 657–663, doi: 10.1093/aje/wv452 (2016).
7. Assiri, A. et al. Hospital outbrea of Middle East respiratory syndrome coronavirus. N. Engl. J. Med. 369, 407–416, doi: 10.1056/
NEJMoa1306742 (2013).
8. Guery, B. et al. Clinical features and viral diagnosis of two cases of infection with Middle East espiratory Syndrome coronavirus: a
report of nosocomial transmission. Lancet. 381, 2265–2272, doi: 10.1016/S0140-6736(13)60982-4 (2013).
9. Memish, Z. A., Zumla, A. I., Al-Haeem, . F., Al-abeeah, A. A. & Stephens, G. M. Family cluster of Middle East respiratory
syndrome coronavirus infections. N. Engl. J. Med. 368, 2487–2494, doi: 10.1056/NEJMoa1303729 (2013).
10. Par, S. H. et al. Outbreas of Middle East espiratory Syndrome in Two Hospitals Initiated by a Single Patient in Daejeon, South
orea. Infect. Chemother. 48, 99–107, doi: 10.3947/ic.2016.48.2.99 (2016).
11. Hemida, M. G. et al. Middle East espiratory Syndrome (MES) coronavirus seroprevalence in domestic livestoc in Saudi Arabia,
2010 to 2013. Euro. Surveill. 18, 20659 (2013).
12. Alagaili, A. N. et al. Middle East respiratory syndrome coronavirus infection in dromedary camels in Saudi Arabia. MBio. 5,
e00884–14, doi: 10.1128/mBio.00884-14 (2014).
13. Azhar, E. I. et al. Evidence for camel-to-human transmission of MES coronavirus. N. Engl. J. Med. 370, 2499–2505, doi: 10.1056/
NEJMoa1401505 (2014).

www.nature.com/scientificreports/
7
Scientific RepoRts | 7:44875 | DOI: 10.1038/srep44875
14. Müller, M. A. et al. MES coronavirus neutralizing antibodies in camels, Eastern Africa, 1983-1997. Emerg. Infect. Dis. 20,
2093–2095, doi: 10.3201/eid2012.141026 (2014).
15. Corman, V. M. et al. Antibodies against MES coronavirus in dromedary camels, enya, 1992-2013. Emerg. Infect. Dis. 20,
1319–1322, doi: 10.3201/eid2008.140596 (2014).
16. Falzarano, D. et al. Treatment with interferon-α 2b and ribavirin improves outcome in MES-CoV-infected rhesus macaques. Nat.
Med. 19, 1313–1317, doi: 10.1038/nm.3362 (2013).
17. Hart, B. J. et al. Interferon-β and mycophenolic acid are potent inhibitors of Middle East respiratory syndrome coronavirus in cell-
based assays. J. Gen. Virol. 95, 571–577, doi: 10.1099/vir.0.061911-0 (2014).
18. Omrani, A. S. et al. ibavirin and interferon alfa-2a for severe Middle East respiratory syndrome coronavirus infection: a
retrospective cohort study. Lancet. Infect. Dis. 14, 1090–1095, doi: 10.1016/S1473-3099(14)70920-X (2014).
19. de Wilde, A. H. et al. MES-coronavirus replication induces severe in vitro cytopathology and is strongly inhibited by cyclosporin
A or interferon-α treatment. J. Gen. Virol. 94, 1749–1760, doi: 10.1099/vir.0.052910-0 (2013).
20. Dyall, J. et al. epurposing of clinically developed drugs for treatment of Middle East respiratory syndrome coronavirus infection.
Antimicrob. Agents. Chemother. 58, 4885–4893, doi: 10.1128/AAC.03036-14 (2014).
21. Ying, T. et al. Exceptionally potent neutralization of Middle East respiratory syndrome coronavirus by human monoclonal
antibodies. J. Vir ol. 88, 7796–7805, doi: 10.1128/JVI.00912-14 (2014).
22. Jiang, L. et al. Potent neutralization of MES-CoV by human neutralizing monoclonal antibodies to the viral spie glycoprotein. Sci.
Transl. Med. 6, 234ra59, doi: 10.1126/scitranslmed.3008140 (2014).
23. Du, L. et al. A conformation-dependent neutralizing monoclonal antibody specically targeting receptor-binding domain in Middle
East respiratory syndrome coronavirus spie protein. J. Viro l. 88, 7045–7053, doi: 10.1128/JVI.00433-14 (2014).
24. Tang, X. C. et al. Identication of human neutralizing antibodies against MES-CoV and their role in virus adaptive evolution. Proc.
Natl. Acad. Sci. USA 111, E2018–E2026, doi: 10.1073/pnas.1402074111 (2014).
25. Corti, D. et al. Prophylactic and postexposure ecacy of a potent human monoclonal antibody against MES coronavirus. Proc.
Natl. Acad. Sci. USA 112, 10473–10478, doi: 10.1073/pnas.1510199112 (2015).
26. Pascal, . E. et al. Pre- and postexposure ecacy of fully human antibodies against Spie protein in a novel humanized mouse model
of MES-CoV infection. Proc. Natl. Acad. Sci. USA 112, 8738–8743, doi: 10.1073/pnas.1510830112 (2015).
27. Agrawal, A. S. et al. Passive Transfer of A Germline-lie Neutralizing Human Monoclonal Antibody Protects Transgenic Mice
Against Lethal Middle East espiratory Syndrome Coronavirus Infection. Sci. ep. 6, 31629, doi: 10.1038/srep31629 (2016).
28. Lu, G. et al. Molecular basis of binding between novel human coronavirus MES-CoV and its receptor CD26. Nature. 500,
227–2231, doi: 10.1038/nature12328 (2013).
29. Wang, L. et al. Evaluation of candidate vaccine approaches for MES-CoV. Nat. C ommun. 6, 7712, doi: 10.1038/ncomms8712 (2015).
30. Muthumani, . et al. A synthetic consensus anti-spie protein DNA vaccine induces protective immunity against Middle East
respiratory syndrome coronavirus in nonhuman primates. Sci. Transl. Med. 7, 301ra132, doi: 10.1126/scitranslmed.aac7462 (2015).
31. im, E. et al. Immunogenicity of an adenoviral-based Middle East espiratory Syndrome coronavirus vaccine in BALB/c mice.
Vaccine. 32, 5975–5982, doi: 10.1016/j.vaccine.2014.08.058 (2014).
32. Zhao, J. et al. apid generation of a mouse model for Middle East respiratory syndrome. Proc. Natl. Acad. Sci. USA 111, 4970–4975,
doi: 10.1073/pnas.1323279111 (2014).
33. Haagmans, B. L. et al. An orthopoxvirus-based vaccine reduces virus excretion aer MES-CoV infection in dromedary camels.
Science. 351, 77–81, doi: 10.1126/science.aad1283 (2016).
34. Guo, X. et al. Systemic and mucosal immunity in mice elicited by a single immunization with human adenovirus type 5 or 41 vector-
based vaccines carrying the spie protein of Middle East respiratory syndrome coronavirus. Immunology. 145, 476–484, doi:
10.1111/imm.12462 (2015).
35. Song, F. et al. Middle East respiratory syndrome coronavirus spie protein delivered by modied vaccinia virus Anara eciently
induces virus-neutralizing antibodies. J. Vir ol. 87, 11950–11954, doi: 10.1128/JVI.01672-13 (2013).
36. Coleman, C. M. et al. Puried coronavirus spie protein nanoparticles induce coronavirus neutralizing antibodies in mice. Vaccine.
32, 3169–3174, doi: 10.1016/j.vaccine.2014.04.016 (2014).
37. Agrawal, A. S. et al. Immunization with inactivated Middle East espiratory Syndrome coronavirus vaccine leads to lung
immunopathology on challenge with live virus. Hum. Vaccin. Immunother. 7, 1–6, doi: 10.1080/21645515.2016.1177688 (2016).
38. Du, L. et al. A truncated receptor-binding domain of MES-CoV spie protein potently inhibits MES-CoV infection and induces
strong neutralizing antibody responses: implication for developing therapeutics and vaccines. PLoS One 8, e81587, doi: 10.1371/
journal.pone.0081587 (2013).
39. Yang, Y. et al. e amino acids 736–761 of the MES-CoV spie protein induce neutralizing antibodies: implications for the
development of vaccines and antiviral agents. Viral. Immunol. 27, 543–550, doi: 10.1089/vim.2014.0080 (2014).
40. Mou, H. et al. e receptor binding domain of the new Middle East respiratory syndrome coronavirus maps to a 231-residue region
in the spie protein that eciently elicits neutralizing antibodies. J. Vir ol. 87, 9379–9383, doi: 10.1128/JVI.01277-13 (2013).
41. Ma, C. et al. Searching for an ideal vaccine candidate among different MES coronavirus receptor-binding fragments—the
importance of immunofocusing in subunit vaccine design. Vaccine. 32, 6170–6176, doi: 10.1016/j.vaccine.2014.08.086 (2014).
42. Lan, J. et al. Tailoring subunit vaccine immunity with adjuvant combinations and delivery routes using the Middle East respiratory
coronavirus (MES-CoV) receptor-binding domain as an antigen. PLoS One 9, e112602, doi: 10.1371/journal.pone.0112602 (2014).
43. Deming, D. et al. Vaccine ecacy in senescent mice challenged with recombinant SAS-CoV bearing epidemic and zoonotic spie
variants. PLoS. Med. 3, e525, doi: 10.1371/journal.pmed.0030525 (2006).
44. Honda-Oubo, Y. et al. Severe acute respiratory syndrome-associated coronavirus vaccines formulated with delta inulin adjuvants
provide enhanced protection while ameliorating lung eosinophilic immunopathology. J. Viro l. 89, 2995–3007, doi: 10.1128/
JVI.02980-14 (2015).
45. Iwata-Yoshiawa, N. et al. Eects of Toll-lie receptor stimulation on eosinophilic inltration in lungs of BALB/c mice immunized
with UV-inactivated severe acute respiratory syndrome-related coronavirus vaccine. J. Vir ol. 88, 8597–8614, doi: 10.1128/JVI.00983-
14 (2014).
46. Weingartl, H. et al. Immunization with modied vaccinia virus Anara-based recombinant vaccine against severe acute respiratory
syndrome is associated with enhanced hepatitis in ferrets. J. Virol. 78, 12672–12676 (2004).
47. Tseng, C. T. et al. Immunization with SAS coronavirus vaccines leads to pulmonary immunopathology on challenge with the SAS
virus. PLoS. One. 7, e35421, doi: 10.1371/journal.pone.0035421 (2012).
48. Jaume, M. et al. SAS CoV subunit vaccine: antibody-mediated neutralisation and enhancement. Hong. ong. Med. J. 18, Suppl 2
31–6 (2012).
49. Coleman, C. M. & Frieman, M. B. Growth and Quantication of MES-CoV Infection. Curr. Protoc. Microbiol. 37, 15E.2.1–9, doi:
10.1002/9780471729259.mc15e02s37 (2015).
50. Hashem, A. M. et al. CD40 ligand preferentially modulates immune response and enhances protection against inuenza virus. J.
Immunol. 193, 722–734, doi: 10.4049/jimmunol.1300093 (2014).
51. Tang, J. et al. Optimization of antigen dose for a receptor-binding domain-based subunit vaccine against MES coronavirus. Hum.
Vaccin. Immunother. 11, 1244–1250, doi: 10.1080/21645515.2015.1021527 (2015).
52. Yang, Z. Y. et al. A DNA vaccine induces SAS coronavirus neutralization and protective immunity in mice. Nature. 428, 561–564
(2004).

www.nature.com/scientificreports/
8
Scientific RepoRts | 7:44875 | DOI: 10.1038/srep44875
Acknowledgements
is project was funded by a research endowment fund from King Abdulaziz University Scientic Endowment
(WAQF), Jeddah, Saudi Arabia under grant No. (WAQF-1435–7B) to AMH.
Author Contributions
A.T.A., M.K.A., X.L. and A.M.H. designed the study; S.S.A., A.T.A., L.A.S., A.A., M.A.S., R.Y.A. and A.M.H.
conducted the experiments; A.M.H. analyzed data and prepared the gures; A.M.H., X.L. and E.I.A. wrote the
manuscript. All authors read and approved the nal manuscript.
Additional Information
Supplementary information accompanies this paper at http://www.nature.com/srep
Competing Interests: e authors declare no competing nancial interests.
How to cite this article: Al-amri, S. S. et al. Immunogenicity of Candidate MERS-CoV DNA Vaccines Based on
the Spike Protein. Sci. Rep. 7, 44875; doi: 10.1038/srep44875 (2017).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
is work is licensed under a Creative Commons Attribution 4.0 International License. e images
or other third party material in this article are included in the article’s Creative Commons license,
unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license,
users will need to obtain permission from the license holder to reproduce the material. To view a copy of this
license, visit http://creativecommons.org/licenses/by/4.0/
© e Author(s) 2017