Fast-spreading SARS-CoV-2 variants: challenges to and new design strategies of COVID-19 vaccines

The COVID-19 pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is still threatening global health. According to the latest data, the number of diagnosed cases has exceeded 100 million. Comfortingly, experiences have been accumulated in preventing and treating COVID-19 through virological, immunological, epidemiological, and clinical investigations of this disease. Besides, the continuous advancement of different vaccines brings the dawn to defeat the epidemic. However, the emergence of fast-spreading SARS-CoV-2 mutant strains (B.1.1.7, B.1.351, and B.1.1.28.1) was reported at the end of 2020, causing concern to prevention and treatment of COVID-19. It is speculated that the emergence of the SARS-CoV-2 variants may portend a new phase of the pandemic.

The SARS-CoV-2 infects cells of the human through the binding of angiotensin-converting enzyme 2 (ACE2) by RBD of Spike protein (Fig. 1b). It seems that these key mutations affected the binding ability to ACE2. The variants discovered in the UK, South Africa, and Brazil have a substitution at position 501 of the spike protein (N501Y), which seems to enhance the binding ability to ACE2. Andersen et al. found that six amino acid residues of RBD are critical for the binding capacity of SARS-CoV-2 to ACE2 receptors, including L455, F486, Q493, S494, N501, and Y505. 17 Residues N501 interact with a salt bridge D38-K353 of ACE2. 18 This function contributes to increasing the binding ability to ACE2. 19 Qin et al. revealed that N501Y mutation potentially associated with the increased virulence in a mouse model. 18,20 Bloom's work also mentioned that the N501 site mutation of RBD could enhance affinity notably. 21 These preliminary pieces of evidence indicate that the N501Y mutation may increase transmissibility. 8,22 Besides, Kristian Andersen identified another notable feature of SARS-CoV-2 that the spike protein has a functional polybasic (Furin) cleavage site. Once the stability of spike protein declined due to cleavage by Furin proteases, it's possible to increase the binding ability to ACE2 receptor markedly. 23 Unfortunately, the B.1.1.7 strain emerged with P681H mutation near the protease cleavage site, threatening spike protein stability.
The E484K mutation is coincidently found in several variants, including B.1.351, B.1.1.28.1, B.1.525, and B.1.526. This mutation occurred at critical sites in the receptor-binding motif (RBM) of the RBD. As the central functional motif, RBM is relatively unconserved and directly affects the binding to the human ACE2 receptor. 24 The E484 interacts with the hotspot residue of human ACE2. Some evidence indicated that the E484K mutation might increase the immunological resistance of variants to neutralization of several monoclonal and human serum antibodies. Whelan et al. isolated 48 escape mutants by using a chimeric virus and 19 anti-RBD monoclonal antibodies. Subsequently, they used COVID-19 vaccine-elicited sera samples to detect the escape of mutants. All four mutants undergoing substitution at E484 are resistant to neutralization of human immune serum. 25 Bloom et al. had also observed that the mutant at E484 could significantly avoid recognition by polyclonal human serum antibodies. 26 Based on current data, scholars speculate that the emergence of the E484K mutation seems to have changed the antigenicity of SARS-CoV-2. Therefore, immune evasion is likely to occur in the novel strain B.1.351 that bears the E484K mutation.
Residues K417 ensures the normal binding affinity of coronavirus by forming a salt bridge with D30 of hACE2. The results of deep mutational scanning indicate that K417N/T mutation seems to have minimal impact on binding ability. 21 However, Qin et al. generated a mouse-adapted strain of SARS-CoV-2 (MASCp6), bearing both N501Y and K417N mutations, which showed 100% fatal rate to aged male mice. 27 This result perhaps reminds us to The major structure of SARS-CoV-2, including spike protein, membrane protein, envelope protein, nucleocapsid protein and RNA. As a transmembrane protein, angiotensin-converting enzyme 2 (ACE2) serves as the main entry point into cells for SARS-CoV-2 consider the infectivity and pathogenicity of 501Y.V2 for the aged population. Some studies found that L452R mutation weakens the binding ability of convalescent patients' antibodies and serum to spike protein. 26,28,29 The Q677 mutation has been detected in at least seven SARS-CoV-2 variants. However, there is no sufficient evidence to prove its impact on the pathogenicity of the variants. 30 In addition to the spike protein mutation, the Q27 stop mutation in the ORF8 region inactivates the ORF8 function by truncating the protein. A similar situation occurred in Singapore in March 2020. This variant named Δ382 has a deletion of 382 bp in the ORF8. It is found that Δ382 variants showed significantly higher replicative fitness in vitro, but the patients infected by this variant have no different viral load compared with that of wild type. 31 The emergence of multiple SARS-CoV-2 strains with ORF8 deletion worldwide indicates that ORF8 inactivation may be associated with the adaptive evolution of SARS-CoV-2. Outside of non-synonymous mutation, the HV 69-70 deletion has been detected in multiple lineages. It seems to facilitate the escape of the coronavirus from the host's immunological response. For example, the variant N439K contained HV 69-70 deletion showed partial immune evasion, the variant Y453F found in mink increased the binding ability to ACE2. 24 Although the evidence of antigen drift for SARS-CoV-2 is still insufficient, it is conceivable that the virus can acquire immunological resistance or other characteristics due to the accumulation of mutations. Coincidently, these mutations of new variants showed similar substitutions at the mutational sites. However, the emergence locations of the variants are geographically distant, indicating that the underlying mechanism by which the mutation is driven may share some similarities. It is, therefore, worthy of unveiling the biological function result from these mutations for the prevention and treatment of COVID-19.

CURRENT STATUS OF THE COVID-19 VACCINE DEVELOPMENT
Under the pressure of the COVID-19 pandemic, the speed of vaccine development and application is unprecedented. These vaccines can be divided into four categories: inactivated virus vaccine, nucleic acid vaccine, protein subunit vaccine and adenoviral vector-based vaccine. 32 More than 70 preclinical vaccines have been tested in animals, and 86 candidates have entered the clinical trials phase. But only 13 vaccines (Table 2) have either been approved for clinical application or released data from phase III clinical trials. 2 To date, more than 700 million doses of vaccination have been initiated in 115 countries worldwide, among which China and the United States vaccinated more than 100 million doses.
As the worldwide application of the COVID-19 vaccine, the side effects of vaccination have arisen the concern of society. According to the Centers for Disease Control and Prevention (CDC) and WHO, common adverse events after vaccination include headache, injection site pain, fatigue, dizziness, nausea, chills, pyrexia, etc. 33 In the US,~372 cases per million doses of mRNA vaccines (BNT162b2 or mRNA-1273) had been reported with non-serious adverse reactions. According to the UK safetymonitoring system, there are about 4000 adverse reactions per million doses of the ChAdOx1 vaccine (AZD1222). The Phase I/II clinical trial data of inactivated virus vaccine, including CoronaVac and two inactivated virus vaccines developed by the Sinopharm, showed that most of the adverse events were common side effects and none were serious. [33][34][35] To date, no death case has been reported directly attributable to the vaccination.
In brief, there is no doubt that the current vaccines are safe. However, concerns about the effectiveness of vaccines have also arisen with the emergence of variants. We still need more clinical data to monitor the effects of vaccines for a long time.   36,37 Moreover, Nussenzweig et al. investigated the antibody and memory B cell responses in 20 participants who received either mRNA-1273 vaccines or BNT162b2 vaccines. They found that the neutralizing activity of vaccine-elicited sera against pseudoviruses (E484K, N501Y, and K417N-E484K-N501Y combination) was reduced. 38 Another research also demonstrated that E484K mutant strain significantly reduced the neutralizing activity of human convalescent and post-vaccination sera. 39 Researchers used convalescent sera, vaccine-elicited sera (mRNA-1272 and NVX-CoV2373) and monoclonal antibodies to assess the neutralization phenotype of the pseudoviruses of 501Y.V1, 501Y.V2 and P.1. They observed a decrease in neutralizing activity. 7,40,41 However, the significant limitation of the current studies is that the engineered pseudovirus cannot fully present the biological properties of the authentic viruses.
In February 2021, an investigation reported that two approved vaccines (BBIBP-CorV and ZF2001) still have the protective efficacy to 501Y. V2 authentic virus, although neutralization titer of postvaccination sera against 501Y.V2 declined 1.6-fold. These data indicated that the 501Y.V2 showed more resistance to the vaccinee serum. 42 Sigal et al. also have found that plasma from convalescent patients infected with no-CoV variant (the variants usually showed the D614G mutation) have reduced neutralizing ability to 501Y.V2 variant, but plasma from convalescent patients infected with 501Y.V2 only showed moderate reduction of neutralizing to the no-COV variant. 43 Recently, Wang et al. assessed the immunological resistance of the variants to neutralization by using convalescent sera and sera from participants received inactivated-virus vaccines (BBIBP-CorV or CoronaVac). Their findings indicated that the neutralization of convalescent or BBIBP-CorV-elicited sera against B.1.1.7 variant reduced slightly, whereas the neutralization against B.1.351 reduced significantly. 44 The two variants showed more resistance to the CoronaVac-elicited serum than the wild-type virus. Several experiments have also been exerted to investigate the immunological resistance of variants to the neutralization of antibodies or sera. [45][46][47][48] The biotech firm Novavax recently disclosed the results of phase III clinical trials of NVX-CoV2373 for variants. The protective efficacy to 501Y. V1 (B.1.1.7) and 501Y. V2 (B.1.351) is apparently different. The effectiveness against 501Y.V1 is more than 85% and the efficacy against 501Y.V2 is less than 50%. 49 This finding indicated that SARS-CoV-2 variants also challenge recombinant protein vaccine.
In general, the available data have indicated that the variant of SARS-CoV-2 may have the ability to resist vaccine-induced immunity. These studies suggest that we should try to update the therapeutic strategy and vaccine design against the challenges from variants.

DESIGN STRATEGIES OF COVID-19 VACCINE AGAINST CHALLENGES FROM THE SARS-COV-2 VARIANTS
At present, most variants emerged locally and did not spread to other regions. Even if the variants partially escape the neutralization of antibodies elicited by the vaccination, theoretically it still cannot completely resist to the recognition of the existed antibodies since the variants share similarity of the antigenicity with the original virus. Therefore, organized and extensive vaccination by currently available vaccines is still necessary.
To fight against the challenges of SARS-CoV-2 variants, the development of vaccines effective to neutralize the variants is urgent. 50 The spike protein of SARS-CoV-2 is the most prevailing target for COVID-19 vaccine development. The emergence of variants with mutations in spike protein may disrupt some original vaccine development schedules. Although the spike protein structure of the variant might be changed, the designs of vaccines for the variants always target the spike protein. For the development of nucleic acid vaccine, protein subunit vaccine, and adenoviral vectorbased vaccine, it is relatively easy to update the vaccine antigen the same with that of the variant. In principle, the vaccine can be updated only by modifying the gene sequence of the spike protein.
However, the consequence of modification is to be investigated, especially for the safety, efficacy of the original virus and variants. Many institutions and pharmaceutical companies are currently focusing on the development of new vaccines for the SARS-CoV-2 variants. It is worth noting that three angles are critical for the design of the new vaccines. (i) design new vaccines against variants and vaccinate individuals based on initial vaccines to obtain fresh immunological memory (ii) Try to develop "multivalent vaccine" to gain immunity to multiple variants. (iii) obtain higher antibody titers by re-vaccinating the original vaccine. These investigations require a large amount of laboratory and clinical endeavors. Meanwhile, we need to closely monitor the genomic information of the virus to detect the mostly new variants. Other limitations should be broken, including insufficient vaccine manufacturing, transportation and preservation, no general guiding regulations, etc.
Coping with the life after COVID-19, we highlight several perspectives: firstly, to upgrade and develop vaccines promptly, we should continue to track the COVID-19 and detect the emergence of new variants. Secondly, no vaccine can be applied to all situations or cases. Therefore, diversified vaccine development and application are critical. Finally, we should break the barriers and promote global cooperation in research on the COVID-19. We need to share the data promptly to address the challenges of the future.