Neutralizing immunity against coronaviruses in Tanzanian health care workers

The ongoing vaccination efforts and exposure to endemic and emerging coronaviruses can shape the population's immunity against this group of viruses. In this study, we investigated neutralizing immunity against endemic and emerging coronaviruses in 200 Tanzanian frontline healthcare workers (HCWs). Despite low vaccination rates (19.5%), we found a high SARS-CoV-2 seroprevalence (94.0%), indicating high exposure in these HCWs. Next, we determined the neutralization capacity of antisera against human coronavirus NL63, and 229E, SARS-CoV-1, MERS-CoV and SARS-CoV-2 (including Omicron subvariants: BA.1, BQ.1.1 and XBB.1.5) using pseudovirus neutralization assay. We observed a broad range of neutralizing activity in HCWs, but no neutralization activity detected against MERS-CoV. We also observed a strong correlation between neutralizing antibody titers for SARS-CoV-2 and SARS-CoV-1, but not between other coronaviruses. Cross-neutralization titers against the newer Omicron subvariants, BQ.1.1 and XBB.1.5, was significantly reduced compared to BA.1 and BA.2 subvariants. On the other hand, the exposed vaccinated HCWs showed relatively higher median cross-neutralization titers against both the newer Omicron subvariants and SARS-CoV-1, but did not reach statistical significance. In summary, our findings suggest a broad range of neutralizing potency against coronaviruses in Tanzanian HCWs with detectable neutralizing immunity against SARS-CoV-1 resulting from SARS-CoV-2 exposure.


Participant's information, clinical and laboratory tests
Data on participants' demographics, clinical information were collected using a structured questionnaire through interviews with participants' demographic information including age, sex and departments at which they were working were collected.Clinical data such as history of symptoms (if any) relevant to COVID-19 during the month of January to June 2022, history of hospitalization, and risk information for severity of COVID-19 including cardiovascular diseases, renal and liver diseases, COPD and diabetes mellitus were collected.Laboratory test results for PCR test for SARS-CoV-2 was also collected.

Sample collection, handling and SARS-CoV-2 serological test
Whole blood samples were collected in EDTA vacutainer tube and sent to laboratory for immediate processing within 2-h of collection.Plasma was isolated by centrifugation then immediately stored in portions at −80 °C.We used STANDARD Q COVID-19 IgM/IgG plus antibody test kit (SD Biosensor) to detect SARS-CoV-2 IgG and IgM from plasma samples of study participants.This kit is a rapid immunochromatography test designed for the qualitative presumptive detection of anti nucleocapsid (N) and antispike -RBD specific SARS-CoV-2 IgM and IgG 17 .Briefly, 10 µL of plasma was added to the specimen well followed by 3 drops of buffer.Reading of the test results were done after 10-15 min as per manufacturer protocol.A sample was regarded as positive when the colored bands appeared on the result window corresponding to "C" Control line together with "G" and/or "M" Test line 18 .

SARS-CoV-2 ELISA
Anti-SARS-CoV-2 S-RBD protein Human IgG ELISA kit (Proteintech) was used for quantitative detection of anti-SARS-CoV-2 RBD IgG in plasma samples as per manufacturer's protocol.Briefly, plasma samples and controls were diluted to 1:100, and added in duplicate to microwells pre-coated with S-RBD recombinant protein.After 30 min incubation at room temperature, the wells were washed and a horseradish peroxidase conjugated anti-human IgG were added to each well and incubated.The presence of immune complex recombinant protein-human anti-S-RBD IgG antibody-HRP conjugated antibody was demonstrated after the addition of a chromogenic solution that initiated a color development reaction.The optical density reading obtained with a spectrophotometer set at 450/620 nm was proportional to the amounts of antibodies present in the specimen.The average of the duplicate readings for each standard and sample was subtracted from the average zero standard absorbance.Anti-S-RBD antibody concentrations were obtained from the best-fit standard curve determined by regression analysis using four-parameter logistic curve fit.

Coronaviruses pseudovirus production
Pseudoviruses bearing SARS-CoV-2 spike protein were produced as previously described 21 .Briefly, 293 T cells were seeded and co-transfected with 50 ng of HCoV spike-encoding plasmids and 2 µg of a previously prepared lentiviral backbone 22,23 , carrying a luciferase reporter gene and a HiBiT peptide tag, designated as pSG3 ΔENVΔNef -Luc2-IN/HiBit.After 48 h of incubation at 37 °C and 5% CO 2 , DNase was added to the cell supernatant to degrade any unutilized lentiviral plasmid and incubated for 30 min.The culture supernatant was clarified by spinning down at 400g for 5 min.Pseudovirus concentration was quantified based on HiBiT-generated luminescence and normalized by the corresponding level of p24 antigen as previously described 22 .The same procedure was followed to produce and quantify pseudoviruses bearing NL63, 229E, SARS-CoV-1 and MERS-CoV spikes.All pseudoviruses were stored at −80 °C until needed.

Plasma neutralization assay
Plasma was tested for SARS-CoV-2 spike neutralization as previously described 24,25 .In brief, heat inactivated plasma (56 °C for 45 min) was fivefold serially diluted on a 96-well plate from a starting dilution of 1:40.Pseudoviruses bearing SARS-CoV-2 spike were thawed and added at a concentration of 6 ng p24 antigen per well, and incubated.After 1 h incubation at 37 °C and 5% CO 2 , HOS cells stably expressing ACE2 and TMPRSS2 were added at a density of 10,000 cells/well and incubated for additional 48 h for chemiluminescence detection (ONE-Glo, Promega).Plasma neutralization was determined by calculating the percent reduction in luminescence between the test wells and pseudovirus only wells.Neutralizing titers were calculated on a dose-response curve fit with a non-linear function and expressed as NT 50 values.The same procedure was employed to quantify neutralizing titers against other HCoVs albeit with alterations in target cells and pseudovirus concentration.Specifically, target cells for NL63 and SARS-CoV-1 spike-bearing pseudovirus infection were HOS-ACE2/TMPRSS2 cells 23,26 , and those for MERS-CoV and 229E spike-bearing pseudovirus infection were HOS-DPP4/TMPRSS2 and HOS-ANPEP /TMPRSS2 cells, respectively.Optimal pseudovirus titers for neutralization were obtained by titration on relevant target cells where 6 ng were used for pseudoviruses bearing SARS-CoV-2, SAR-CoV-1 and MERS-CoV and 8 ng for HC0V-229E and 25 ng for HCoV-NL63.
To screen for non-specific neutralization activity, pseudoviruses expressing VSV-G protein instead of coronaviruses spike were used 27 .Samples demonstrating significant inhibition activity against VSV-G were categorized as exhibiting non-specific inhibition targeting the proviral backbone and were subsequently excluded from further analysis.

Neutralising activity against coronaviruses in HCWs samples
To quantify the neutralizing capacity of antisera from HCWs, we measured the dilution of antisera required to prevent 50% infectivity of pseudoviruses expressing spike proteins for SARS-CoV-2, SARS-CoV-1, MERS-CoV, and HCoV NL63 and 229E.Samples with neutralization titer (NT 50 ) of over 1 in 40 dilution were classed as neutralizing.For this analysis 14 sera samples showed non-specific inhibition activity and were excluded for the analysis.Of the remaining 186 plasma, nearly all samples (94%) showed a potent neutralization activity  1a).We next assessed whether vaccination status could impact SARS-CoV-2 neutralization in these highly exposed HCWs.We found that median neutralisation activity was significantly higher against SARS-CoV-2 wild-type spike (p = 0.0366), while it was modestly high, but not statistically significant, against SARS-CoV-1 (p = 0.0844) in the exposed vaccinated HCW sera.In contrast, there were no differences in neutralisation capacity between exposed vaccinated and exposed unvaccinated sera against the tested seasonal coronaviruses (Fig. 1b).
If infection with SARS-CoV-2 increased the neutralising antibody titer for other coronaviruses, we would expect a higher neutralization in SARS-COV-2 seropositive samples.When we compared neutralisation between serostatus groups, we found that seropositive group showed higher neutralising activity against SARS-CoV-1 while seronegative group showed no neutralisation activity (p = 0.0179).In contrast, both human coronaviruses NL43 and 229E showed no differences in neutralisation between SARS-CoV-2 seropositive and negative groups (Fig. 1c).We found a strong correlation between neutralisation of SARS-CoV-2 and SARS-CoV-1 suggesting cross-neutralizing immunity between the two coronaviruses.No correlation was observed between human seasonal coronaviruses and SARS-CoV-2 or SARS-CoV-1 (Fig. 1d).

Cross neutralising capacity against Omicron subvariants
During the data collection for this study, Tanzania and neighbouring countries were experiencing the end of the first wave of Omicron variant 16,28 .Therefore, we aimed to determine the cross-neutralizing immunity against the newer Omicron subvariants, i.e., BQ.1.1 and XBB.1.5,which were spreading in other parts of the world.A total of 68 plasma samples were analysed for neutralisation against the selected Omicron subvariants' spikes.Compared to the wild-type, Omicron subvariants showed increasing resistance to neutralization (p < 0.0001) (Fig. 2a).When compared to BA.1 and BA.2, the plasma from HCWs showed a significant decrease in neutralization activity against BQ.1.1 and XBB.1.5.
To assess whether vaccination status had impact on neutralizing immunity against the tested Omicron subvariants, the serum samples were grouped into two categories: exposed vaccinated and exposed unvaccinated groups.Across the subvariants, exposed vaccinated group showed higher neutralization titers compared to exposed unvaccinated groups.This trend was statistically significant in XBB.1.5subvariants (Fig. 2b), suggesting a tendency towards enhanced neutralizing immunity in HCWs who were exposed and received at least one dose of the COVID-19 vaccine.

Discussion
The recent SARS-CoV-2 pandemic and previous outbreaks highlight the health threat posed by coronaviruses.While the WHO has declared the end of the COVID-19 emergency, our understanding on how the current pandemic and previous exposures to endemic and zoonotic coronaviruses have shaped our immunity against this family of viruses remain to be studied.In our study, we investigated the heterologous nature of neutralizing antibodies against coronaviruses in Tanzanian HCWs who during the height of the pandemic, represented one of the most exposed groups to SARS-CoV-2 infection 15 .We found that nearly all HCWs had developed a robust neutralizing immunity against SARS-CoV-2 spike, with indications of cross-neutralization activity against SARS-CoV-1 but not against the tested HCoVs.Similar to other studies [29][30][31] , hybrid immunity in this highly SARS-CoV-2 exposed group was associated with enhanced cross-neutralization activity of the elicited neutralizing immunity.
Overall the HCWs' sera demonstrated potent pseudovirus neutralization activity against the wild-type SARS-CoV-2 spike.This activity was strongly correlated with RBD binding antibody titers.Exposure to human and emerging coronaviruses may diversify adaptive immunity against SARS-CoV-2 1, [32][33][34][35] .In this study, our analysis of the relationship between neutralization activity against SARS-CoV-2 and that of coronaviruses (SARS-CoV-1 and HCoV-NL63) 36 , which share ACE2 as their cellular receptor, showed a strong association only with SARS-CoV-1 neutralizing activity.Although all sera showed potent neutralization activity against HCoV-NL63, we did not find any association between this activity and that of SARS-CoV-2, suggesting a lack of elicitation of heterologous neutralizing activity against SARS-CoV-2 from exposure to HCoV-NL63.Additionally, over two-thirds of sera samples in our study showed neutralizing activity against another alpha coronavirus, HCoV-229E, but this activity was not correlated with that against SARS-CoV-2.These findings are in line with the observations that exposure to human alphacoronaviruses is not associated with elicitation of cross-neutralizing immunity against SARS-CoV-2 37,38 .
On the flip side, one of our goal was to clarify whether the ongoing exposure to SARS-CoV-2 could boost immune responses to previously encountered coronaviruses or induce heterologous immunity against other coronaviruses.We demonstrate that there was a strong correlation between SARS-CoV-2 RBD binding antibody titers and neutralization activity against SARS-CoV-1 which is expected, given the substantial relatedness and sequence similarity, particularly within the RBD of both viruses 39 .Additionally, seronegative samples for SARS-CoV-2 showed no neutralization activity against SARS-CoV-1, suggesting that the observed neutralizing activity could be secondary to cross-neutralization activity from SARS-CoV-2-induced antibodies.However, pre-pandemic samples from African cohorts have shown the ability, albeit mild, to neutralize SARS-CoV-1 spike 2 , thus the observed activity against SARS-CoV-1 may not be entirely explained by cross-neutralization by SARS-CoV-2-induced antibodies.Nonetheless, our data demonstrate the potential for heterologous neutralizing immunity against SARS-CoV-1 that could be elicited following coronavirus infections.In a study by Alkhalifah  41 .In our study, none of the sera showed any neutralization activity against this coronavirus at a dilution of 1:40, suggesting a lack of exposure or elicitation of heterologous immunity to this virus in our study population.The absence of differences in neutralization activity against NL63 and 229E between SARS-CoV-2 serostatus groups, as observed in our study, could suggest the lack of reactivation of memory HCoVs B cells, contrary to that reported in other studies 42 .However, a specially designed study is needed to clarify this observation.
The constantly emerging SARS-CoV-2 variants have demonstrated a significant ability to evade humoral immunity elicited by vaccination or previous infection with older variants 43,44 .Our study revealed that HCWs sera had a significantly reduced neutralization capacity against the newer Omicron subvariants XBB1.5 and BQ1.1.However, our data suggest that a higher neutralization capacity against the wildtype SARS-CoV-2 is also associated with higher neutralization activity against emerging variants.Since the vaccination status in our study population was associated with higher neutralization titers, our results suggest that even in these highly SARS-CoV-2-exposed HCWs, there may be some benefits from vaccination with generic vaccine in improving the level of protective antibody titers against newer variants.
Our study has some limitations.First, it was a cross-sectional study involving HCWs in Tanzania, making it impossible to determine the timing of infection and the specific SARS-CoV-2 variants involved.Our panel of coronavirus spikes, tested for their neutralization, did not include all human coronaviruses, consequently, the extent of heterologous humoral immunity elicited could not be fully elucidated.Moreover, our study solely focused on neutralizing immunity against coronaviruses, and therefore, we cannot exclude the potential presence of heterologous T-cell immunity that might have been elicited following exposure to different coronaviruses.Nevertheless, our study contributes to the documentation of neutralizing immunity against coronaviruses in an understudied population whose potential exposure to other coronaviruses remains largely undescribed.