In this study, we demonstrated the seroprevalence of four endemic HCoVs in the prepandemic COVID-19 sera of Filipino children admitted to hospital with severe respiratory illness and evaluated the SARS-CoV-2 cross-reactivity and neutralization capability of these antibodies. Over 90% of samples from newborns (< 2 months old) had antibodies against all four tested HCoVs. We found a decrease in the frequency of seropositive samples in infants aged between 3 and 11 months old. Over this age, the frequencies increased, starting at 9 to 12 months of age, until a plateau was reached at around 29 months of age. Approximately 22% of the samples showed some cross-reactivity against the SARSCoV-2 spike protein, but hardly any neutralization capacity was detected.
Several studies have reported seroprevalence to endemic HCoVs (alpha- and beta-coronaviruses) among children and adults in temperate areas such as the Netherlands, the United States, and China14–16, 24. These previous studies showed that the presence of these antibodies in sera was age-dependent in adults and in a small number of samples from children. Seroprevalence studies of HCoVs using large numbers of samples from children, especially those under 1 year old, are limited16, and there are no reports that analyze the sera of children from tropical areas. To address this gap in the research, we studied more than 150 serum samples from Filipino children aged under 1 year old. We detected maternal antibodies against all four tested HCoVs in most of the 2-month-old individuals. We noted a decrease in seropositivity to HCoVs in the samples taken from children aged 3–8 months old due to the disappearance of maternal antibodies; the results indicated that, after this age, individuals start to produce their antibodies against alpha- and beta-coronaviruses, reaching a plateau at 24–29 and 12–17 months of age for alpha- and beta-coronaviruses, respectively. Antibodies against alpha-coronaviruses have been reported to be present in the sera of 50–60% of newborns (0–2 months old) in the United States14. The frequencies of seropositive individuals have been found to be lower around 5 months of age, and were reported to increase after that age, until a plateau is reached at 2.5–3.5 years of age14. However, other studies have indicated that the plateau of seropositivity to HCoVs is achieved at around 5 years of age15,16. Recently, a Canadian study reported that anti-spike antibodies against the four endemic HCoVs were acquired by 10 years of age and were stable thereafter25. Our study, like the United States study, identified high levels of seropositivity in newborns and the production of antibodies against the four HCoVs at an earlier age compared to the other studies. Thus, our findings suggest that HCoVs are more prevalent in this tropical area than in other template areas.
HCoV spike proteins exhibit less than 35% homology1. Spike proteins are useful as specific antigens for antibody detection16. Regarding the correlation coefficient analysis performed to test the cross-reactivity, interestingly, the highest correlation coefficient with SARS-CoV-2 was found for HCoV-NL63, an alpha-coronavirus, but not for HCoV-HKU1 or HCoVOC43, which are beta-coronaviruses like SARS-CoV-2. The receptors of SARSCoV-2 and HCoV-NL63 are known to be angiotensin-converting enzyme 2 receptors5,26. Recently, it was reported that HCoV-NL63 and SARS-CoV-2 share epitopes that are recognized by the humoral response and that the levels of HCoVNL63 neutralizing antibodies increase after SARS-CoV-2 infection or vaccination27. Thus, our results are in line with this report and suggest the highest correlation coefficient between HCoV-NL63 and SARS-CoV-2. In contrast, the lowest correlation coefficient with SARS-CoV-2 was for HCoV-OC43. This may be due to the high antibody titer against HCoV-OC43, which would have resulted in a lower correlation coefficient.
Several studies have reported that specimens from the COVID-19 prepandemic-period were cross-reactive with SARS-CoV-2, but showed low neutralization capability 20,28. This study also showed same results as those of previous studies. This cause of cross-reactivity to SARS-CoV-2 is reported to be due to the recognition of the S2 region of HCoV spike proteins19,29, while the neutralizing capacity is achieved by the binding to the receptor-binding domain region30. Therefore, antibodies that showed cross-reactivity against SARS-CoV-2 in this study may bind to the S2 region. However, the impact of HCoV and humoral immunity on SARS-CoV-2 has been suggested. SARS-CoV-2 vaccination induces antibodies not only against SARS-CoV-2 but also against HCoVs31. In addition, neutralizing monoclonal antibodies against SARS-CoV-2 have been established from B cells derived from HCoVs-antibody-positive individuals32. Recently, it was reported that prior HCoV-OC43 S2 region immunity primes neutralizing antibody responses to otherwise sub-immunogenic SARS-CoV-2 S gene exposure and promotes S2 antibody responses in mice33. This study also showed that more than 50% of inhibition was only detected in one sample, but some samples showed 40% inhibition against SARS-CoV-2. These antibodies may increase neutralization capability by SARS-CoV-2 infection. Therefore, some antibodies induced by HCoV infection may have a neutralizing function against various coronaviruses which infect humans.
Our study analyzed samples from less than 5 years of children who cause severe pneumonia for seroprevalence to HCoVs and cross-reactivity and neutralization against SARS-CoV-2. However, we did not investigate the first infection and seasonality to HCoVs compare healthy children, adult and/or elderly population, or perform molecular analyses. To further understand the association between age, cross-reactivity, and HCoV genomes, future studies should not only be extended to adult samples but also include HCoV genomic analyses.
In conclusion, this study identified the high seropositivity to four endemic HCoVs among children in the Philippines. Most infants < 2 months old had maternal antibodies against four HCoVs, which disappeared after 3 months. Seroconversion by production of new antibodies was found to occur between 1 and 3 years of age. Although 20% of the samples collected from children before the COVID-19 pandemic showed crossreactivity against the SARS-CoV-2 spike protein, these antibodies had low neutralization capacity.