Vaccines are being rolled out in many countries to contain the spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the agent responsible for the coronavirus disease 2019 (COVID-19). However, immune-evading mutations have been observed within several newly emerging variants of concern (VOCs), creating concern that population immunity may still be a long way off.
An interesting new study, released as a preprint on the medRxiv* server, describes the emergence of several such variants from a chronically immunosuppressed COVID-19 patient, who remained infected for over 170 days before clearing the virus.
The patient was a male transplant patient in his fifties who tested positive for SARS-CoV-2 by reverse transcription quantitative polymerase chain reaction (RT-qPCR) during the early phase of the COVID-19 pandemic in Germany.
The patient was put on tacrolimus, mycophenolate mofetil and prednisone to prevent graft rejection. The patient developed mildly symptomatic COVID-19, and tested positive for 145 days.
The daily swabs showed the virus was developing an array of substitutions and deletions of amino acids in the spike protein, which were partly resistant to neutralizing antibodies. Mycophenolate was removed from the patient’s protocol on day 126, soon after the titer of spike-positive immunoglobulin (Ig) G suddenly rose at day 123 days. The dose of prednisone was also increased.
The aim was to allow the body to mount a more effective antiviral response. On day 140 the patient remained RT PCR positive, and was put on remdesivir for 10 days. From day 149 to 189, all subsequent tests were negative, and the pathogen could no longer be isolated, indicating viral clearance.
High levels of spike-targeting IgG were also found from day 140, when the patient was readmitted, persisting even on day 175. During this period, the cycle thresholds (Ct) also steadily increased, showing that viral replication was decreasing.
Thus, the patient, who remained free of serious symptoms throughout the 25 weeks of infection, eventually cleared the virus due to an increase in neutralizing antibodies, and perhaps also due to the use of remdesivir, a nucleoside analog that inhibits viral replication.
How the patient isolates relate to other circulating strains
The viral genome sequences identified it as one of the strains circulating in Freiburg, Germany, during that period. However, it had several base substitutions in ORF1ab, the spike-encoding gene, ORF3, the membrane (M) and nucleoprotein (N) genes, relative to the Wuhan-Hu-1 genome.
Among the late isolates, one of them generated a broadly protective immune response in mice exposed to experimental infection with the original strain as well as the two variants found in the patient. These serve as a fingerprint and support the persistence of the same virus rather than repeated reinfections.
The typical rate of mutations is about 1-2 mutations a month, and this was confirmed in the present patient, with the relative stability of the viral genome over the early period of the infection.
From day 42, mutations began to accumulate, including the D614G substitution that is now globally dominant. Paired deletions and substitutions in the N-terminal domain (NTD) and the receptor-binding motif (RBM) were also found, namely, del141-144 with F490L and del244-247 with E484G, respectively.
The deletions affect neutralizing epitopes and may also impact the conformation of the NTD, while the amino acid substitutions may do the same in the RBM. Thus, the mutations arising in this patient share the same escape mutations as those from the UK, South Africa and Brazil.
Mutations impair viral fitness
Viral fitness appeared to drop between days 35 and 105, with high Ct values above 25. The day 14 and day 105 isolates were used as representatives of the isolates in these two periods. The latter showed impaired replication compared to the former.
When tested in mice, the day 14 isolate produced significant weight loss in the animals, only one survived, and almost all died. However, only two mice infected with the day 105 strain became severely ill. Thus, the acquired mutations reduced viral fitness.
Neutralizing activity against the virus was found only after day 123, when anti-spike IgG levels rose. The day 14 isolate was neutralized at up to 1:128 dilutions, but not the day 105 isolate, which resisted neutralization even at 1:32, indicating immune escape.
Neutralizing antibodies were also lower in titer in this patient compared to a convalescent COVID-19 patient with normal immune function. Moreover, the day 14 isolate was more susceptible to neutralization by convalescent sera as well as post-vaccination, with the BNT162b2 (Pfizer/BioNTech) mRNA vaccine, than the later one.
This was confirmed by the neutralization of pseudovirus entry into cells in culture. Day 105 was less effectively neutralized, shown by a more than eight-fold reduction in the titer at which 50% of neutralization was completed.
There was no difference in the neutralization of pseudoviruses expressing either day 14 spike, or the E484G or the F490L mutant spike, by convalescent serum. However, the two deletions reduced the efficacy of neutralization.
When exposed to post-vaccination serum, the E484G/del141-144 mutant spike was less well neutralized compared to either of these alone, indicating that possibly these are synergistic mutations.
The four mutations acquired in the NTD and RBD affect crucial epitopes, which were selected during the long course of infection in this immunocompromised patient, allowing it to escape from neutralizing antibodies.
Serum from mice who survived infection with either day 14 or day 105 isolates were collected at 21 days from infection or later. A wild-type infection survivor cohort was also used since only one mouse survived challenge with the day 14 strain, both having the same spike protein sequence.
Wild-type infection survivors had double the levels of virus-specific IgG compared to survivors of day 105 isolate infection. However, sera from the wild-type survivors had four-fold less neutralization capacity against the day 105 isolate than the day 14 isolate, showing partial immune evasion. Immune sera from mice challenged with day 105 isolate neutralized the corresponding strain more efficiently than the day 14 strain.
The currently circulating South African B.1.351 variant was more efficiently neutralized by day 105 serum than the UK B.1.1.7 variant. However, all convalescent mice survived the second challenge without any symptoms, indicating full protective immunity was induced by both day 14 and day 105 variants.
The researchers also found that the spike mutations found within the day 105 and day 140 isolates were selected because of humoral rather than cellular immune escape.
What are the implications?
The study thus supports the emergence of new variants that evade immunity in chronically immunosuppressed patients, as also reported with patients treated repeatedly with antibody cocktails and convalescent plasma. The variants in this study resemble the current UK, South African and Brazil variants, with escape mutations in the same spike region.
Secondly, it appears that mycophenolate mofetil should be discontinued in transplant patients with COVID-19 to enable effective spike-specific humoral responses.
Thirdly, the broad protection afforded by the immune sera from mice immunized with spike protein from the day 105 isolate may shape the design of future vaccines. Again, it appears that viral fitness is compromised by these mutations, depending on the type of host cell. This may also be related to an observed deletion in the furin cleavage site of the spike protein in the variants observed in this patient.
Such escape mutants could serve as initial seed for newly emerging variants with enhanced epidemic potential, especially if they overcome impaired viral fitness by further adaptation. Unexpectedly, the spike protein of this escape variant worked particularly well in immunization approaches and elicited broadly active neutralizing antibodies able to control SARS-CoV-2 variants of concern.”
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.