High frequencies of alpha common cold coronavirus/SARS-CoV-2 cross-reactive functional CD4+ and CD8+ memory T cells are associated with protection from symptomatic and fatal SARS-CoV-2 infections in unvaccinated COVID-19 patients

Background Cross-reactive SARS-CoV-2-specific memory CD4+ and CD8+ T cells are present in up to 50% of unexposed, pre-pandemic, healthy individuals (UPPHIs). However, the characteristics of cross-reactive memory CD4+ and CD8+ T cells associated with subsequent protection of asymptomatic coronavirus disease 2019 (COVID-19) patients (i.e., unvaccinated individuals who never develop any COVID-19 symptoms despite being infected with SARS-CoV-2) remains to be fully elucidated. Methods This study compares the antigen specificity, frequency, phenotype, and function of cross-reactive memory CD4+ and CD8+ T cells between common cold coronaviruses (CCCs) and SARS-CoV-2. T-cell responses against genome-wide conserved epitopes were studied early in the disease course in a cohort of 147 unvaccinated COVID-19 patients who were divided into six groups based on the severity of their symptoms. Results Compared to severely ill COVID-19 patients and patients with fatal COVID-19 outcomes, the asymptomatic COVID-19 patients displayed significantly: (i) higher rates of co-infection with the 229E alpha species of CCCs (α-CCC-229E); (ii) higher frequencies of cross-reactive functional CD134+CD137+CD4+ and CD134+CD137+CD8+ T cells that cross-recognized conserved epitopes from α-CCCs and SARS-CoV-2 structural, non-structural, and accessory proteins; and (iii) lower frequencies of CCCs/SARS-CoV-2 cross-reactive exhausted PD-1+TIM3+TIGIT+CTLA4+CD4+ and PD-1+TIM3+TIGIT+CTLA4+CD8+ T cells, detected both ex vivo and in vitro. Conclusions These findings (i) support a crucial role of functional, poly-antigenic α-CCCs/SARS-CoV-2 cross-reactive memory CD4+ and CD8+ T cells, induced following previous CCCs seasonal exposures, in protection against subsequent severe COVID-19 disease and (ii) provide critical insights into developing broadly protective, multi-antigen, CD4+, and CD8+ T-cell-based, universal pan-Coronavirus vaccines capable of conferring cross-species protection.


Introduction
The coronavirus disease 2019 (COVID- 19) pandemic has created one of the largest global health crises in nearly a century (1)(2)(3).As of February 2024, the COVID-19 outbreak has affected over 700 million people worldwide, with the number of deaths directly related to severe symptomatic COVID-19 infections reaching 7 million worldwide (1,2,4).Some unvaccinated symptomatic COVID-19 patients produce severe symptoms that typically begin with mild upper respiratory syndrome but may further develop into severe respiratory distress and death, particularly in immunocompromised individuals and those with pre-existing co-morbidities (5)(6)(7)(8).In contrast, other unvaccinated individuals never develop any COVID-19 symptoms despite being infected with SARS-CoV-2 (5,9,10).The underlying mechanisms that lead to protection from symptomatic and fatal SARS-CoV-2 infection in unvaccinated COVID-19 patients remain to be fully elucidated.
In this study, we hypothesized that different clonal repertoire of CCCs/SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells are induced by previous exposures to seasonal alpha CCCs (i.e., a-CCC-229E and a-CCC-NL63) and beta CCCs (i.e., b-CCC-HKU1 and b-CCC-OC43) and that certain clones of T cells are associated with either protective or pathogenic outcomes in SARS-CoV-2 infection.We report that, compared with unvaccinated severely ill COVID-19 patients and unvaccinated patients with fatal COVID-19 outcomes, unvaccinated asymptomatic COVID-19 patients displayed significantly (i) higher rate of the a-CCC species 229E (a-CCC-229E); (ii) higher frequencies of functional memory CD134 + CD137 + CD4 + and CD134 + CD137 + CD8 + T cells directed toward cross-reactive a-CCCs/SARS-CoV-2 epitopes from structural, non-structural, and accessory proteins; and (iii) lower frequencies of cross-reactive exhausted PD-1 + TIM3 + TIGIT + CTLA4 + CD4 + and PD-1 + TIM3 + TIGIT + CTLA4 + CD8 + T cells.These findings (i) support the crucial role of functional, polyantigenic a-CCCs/SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells, induced following previous exposures to a-CCC species, in protection against subsequent severe disease caused by SARS-CoV-2 infection and (ii) provides a strong rationale for the development of broadly protective, T-cell-based, multi-antigen universal pan-Coronavirus vaccines.

Human study population cohort and HLA genotyping
Between July 2020 to November 2022, 600 patients were enrolled at the University of California Irvine Medical Center with various severity of COVID-19 disease under an approved Institutional Review Board-approved protocol (IRB No. 2020-5779).Written informed consent was obtained from participants before inclusion.SARS-CoV-2 positivity was defined by a positive RT-PCR on a respiratory tract sample.None of the patients enrolled in this study received any COVID-19 vaccine.

Symptomatic and asymptomatic COVID-19 patient stratification based on disease severity
Following patient discharge, they were divided into six groups depending on the severity of their symptoms and their intensive care unit (ICU) and intubation (mechanical ventilation) status by medical practitioners.The scoring criteria were as follows: severity 5, patients who died from COVID-19 complications; severity 4, infected COVID-19 patients with severe disease who were admitted to the intensive care unit (ICU) and required ventilation support; severity 3, infected COVID-19 patients with severe disease that required enrollment in ICU, but without ventilation support; severity 2, infected COVID-19 patients with moderate symptoms that involved a regular hospital admission; severity 1, infected COVID-19 patients with mild symptoms; and severity 0, infected individuals with no symptoms.Among the 147 COVID-19 patients, subjects with a severity score of 0 were defined as asymptomatic, and subjects with a severity score of 1-5 were defined as symptomatic.

Pre-pandemic healthy controls
Subsequently, we used 15 liquid-nitrogen frozen PBMCs samples (blood collected pre-COVID-19 in 2018) from HLA- Unvaccinated patients (n = 147) were scored on a scale of 0-5 based on the severity of COVID-19 symptoms, regular hospital admission, intensive care unit (ICU) admission and death (severity score).Severity scores 0: asymptomatic patients who had no symptoms despite being tested positive for SARS-CoV-2 (ASYMP).Patients who were SARS-CoV-2 infected and developed symptoms (SYMP) were divided into four categories.Severity 1: patients who were screened at the hospital for COVID-19 but did not stay for regular admission.Severity 2: patients who were screened at the hospital for COVID-19 and went to non-ICU regular admission to treat their symptoms.Severity 3: patients who went to intensive ICU.Severity 4: patients who went to ICU with life support (i.e., mechanical ventilation at any point during their stay).Severity 5: patients who died from direct COVID-19 complications.The parameters displayed in the table (demographic features, HLA genotyping, clinical parameters, and prevalence of comorbidities) represent the number and percentages of patients within each disease severity.For the age parameter, median values are shown for each disease severity along with ranges (between brackets).The time between the onset of symptoms and the blood draw is shown as day-average numbers.The total number of comorbidities is the average of the sum of each patient's comorbidities.

T-cell epitopes screening, selection, and peptide synthesis
CCCs/SARS-CoV-2 cross-reactive peptide epitopes from 12 SARS-CoV-2 proteins, including 27   -15 , and N 388-403 ) that we formerly identified were selected as we previously described (1) (Table 2 and Supplementary Table 1).We used the Epitope Conservancy Analysis tool to compute the degree of identity of CD8 + and CD4 + T-cell epitopes within a given protein sequence of SARS-CoV-2 set at 100% identity level (1) (Table 2 and Supplementary Tables 1, 2).Peptides were synthesized (21st Century Biochemicals, Inc., Marlborough, MA), and the purity of peptides determined by both reversed-phase high-performance liquid chromatography and mass spectroscopy was over 95%.

Blood differential test
Total white blood cell (WBC) count and lymphocyte count per microliter of blood were performed by the clinicians at the University of California Irvine Medical Center, using a CellaVision ™ DM96 automated microscope.Monolayer smears were prepared from anticoagulated blood and stained using the May Grunwald Giemsa (MGG) technique.Subsequently, slides were loaded onto the DM96 magazines and scanned using a ×10 objective focused on nucleated cells to record their exact position.Images were obtained using the ×100 oil objective and analyzed by artificial neural network (ANN).

Peripheral blood mononuclear cells isolation and T-cell stimulation
Peripheral blood mononuclear cells (PBMCs) from COVID-19 patients were isolated from the blood using Ficoll (GE Healthcare) density gradient media and transferred into 96-well plates at a concentration of 2.5 × 10 6 viable cells per ml in 200 µl (0.5 × 10 6 cells per well) of RPMI-1640 media (Hyclone) supplemented with 10% (v/v) FBS (HyClone), sodium pyruvate (Lonza), L-glutamine, nonessential amino acids, and antibiotics (Corning).A fraction of the blood was kept separated to perform HLA genotyping of the patients and select only the HLA-A*02:01 and/or DRB1*01:01 positive individuals (Supplementary Figure 1).Fresh peripheral blood mononuclear cells (PBMCs) were used in this study, as they generally have higher viability and functionality compared to frozen PBMCs.Freezing and thawing can lead to cell damage and loss of T-cell functionality, which may affect the accuracy and reliability of experimental results.Frozen PBMCs may exhibit altered activation status compared to fresh cells.Cryopreservation can induce stress responses in cells, leading to changes in their activation state and potentially affecting immune response assays.In the context of COVID-19 research, where precise characterization of immune responses is crucial for understanding disease pathogenesis, vaccine development, and treatment strategies, using fresh PBMCs ensures the accuracy and reliability of experimental results.A side-by-side comparison of frozen and fresh PBMCs and pre-pandemic healthy control PBMCs yielded no significant difference.PBMCs were stimulated with 10 µg/ml of each one of the 43 individual CCCs/SARS-CoV-2 cross-reactive peptide epitopes (27 CD8 + T-cell peptides and 16 CD4 + T-cell peptides) and incubated in a humidified chamber with 5% CO 2 at 37°C (Supplementary Figure 2A).Post-incubation, cells were stained by flow cytometry analysis or transferred onto IFN-g ELISpot plates.The same isolation protocol was followed for HD samples obtained in 2018.Ficoll was kept frozen in liquid nitrogen in FBS DMSO 10%; after thawing, HD PBMCs were stimulated similarly for the IFN-g ELISpot technique.
All ELISpot reagents were filtered through a 0.22-µm filter.Wells of 96-well Multiscreen HTS Plates (Millipore, Billerica, MA) were pre-wet with 30% ethanol for 60 s and then coated with 100 µl primary anti-IFN-g antibody solution (10 µg/ml of 1-D1K coating antibody from Mabtech, Cincinnati, OH) OVN at 4°C.After washing, the plate was blocked with 200 µl of RPMI media plus 10% (v/v) FBS for 2 h at room temperature to prevent nonspecific binding.After 24 h, following the blockade, the peptide-stimulated cells from the patient's PBMCs (0.5 × 10 6 cells/well) were transferred into the ELISpot-coated plates.PHA-stimulated or nonstimulated cells (DMSO) were used as positive or negative controls of T-cell activation, respectively.Upon incubation in a humidified chamber with 5% CO 2 at 37°C for an additional 48 h, cells were next washed using PBS and PBS-Tween 0.02% solution.Next, 100 µl of biotinylated secondary anti-IFN-g antibody (1 µg/ml, clone 7-B6-1, Mabtech) in blocking buffer (PBS 0.5% FBS) was added to each well.Following a 2-h incubation followed by washing, wells were incubated with 100 µl of HRP-conjugated streptavidin (1:1,000) for 1 h at room temperature.Lastly, wells were incubated for 15-30 min with 100 µl of TMB detection reagent at room temperature, and spots were counted both manually and by an automated ELISpot reader counter (ImmunoSpot Reader, Cellular Technology, Shaker Heights, OH).

TaqMan quantitative polymerase reaction assay for the detection of CCC species in UPPHI and in COVID-19 patients
To detect common cold coronavirus co-infection in COVID-19 patients, Taqman PCR assays were performed on a total of 85 patients distributed into each different category of disease severity (9 ASYMP, 6 patients of category 1, 32 patients of category 2, 9 patients of category 3, 15 patients of category 4, and 14 patients of category 5).Nucleic acid was first extracted from each nasopharyngeal swab sample using Purelink Viral RNA/DNA mini kit (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer's instructions.Subsequently, extracted RNA samples were quantified using Qubit and BioAnalyzer.cDNA was synthesized from 10 mL of RNA eluate using random hexamer primers and SuperScript II Reverse Transcriptase (Applied Biosystems, Waltham, MA).The subsequent RT-PCR screening of the enrolled subjects for the four CCCs was performed using specific sets of primers and probes (55).
CCC-229E, CCC-OC43, and CCC-NL63 RT-PCR assays were performed using the following conditions: 50°C for 15 min followed by denaturation at 95°C for 2 min, 40 cycles of PCR performed at 95°C for 8 s, extending and collecting a fluorescence signal at 60°C for 34 s (56).For CCC-HKU1, the amplification conditions were 48°C for 15 min, followed by 40 cycles of 94°C for 15 s and 60°C for 15 s.For each virus, when the Ct-value generated was <35, the specimen was considered positive.When the Ct-value was relatively high (35 ≤ Ct < 40), the specimen was retested twice and considered positive if the Ct-value of any retest was <35 (57).

Identity and similarity analysis of CCCs/ SARS-CoV-2 cross-reactive epitopes
To assess the % identity (%id) of CCCs/SARS-CoV-2 crossreactive CD4 + and CD8 + T-cell peptide epitopes, we first identified the best matching CCCs peptide across the CCCs proteomes (Table 2).The full CCCs proteomes sequences were obtained from the National Center for Biotechnology Information (NCBI) GenBank [MH940245.1 (CCC-HUK1), MN306053.1 (CCC-OC43), KX179500.1 (CCC-NL63), and MN306046.1 (CCC-229E)].We processed this in the following three steps.(1) Corresponding CCCs peptides were determined after protein sequence alignments of all four homologous CCCs proteins plus the SARS-CoV-2 related one using various multiple sequences alignments (MSA) algorithms ran in JALVIEW, MEGA11, and M-coffee software's (i.e., ClustalO, Kalign3, and M-coffee-the latter computing alignments by combining a collection of multiple alignments from a library constituted with the following algorithms: T-Coffee, PCMA, MAFFT, ClustalW, Dialigntx, POA, MUSCLE, and Probcons).Furthermore, we confirmed our results with global and local pairwise alignments (Needle and Water algorithms ran in Biopython) performed to confirm the results.In case of obtaining different results with the various algorithms, the epitope sequence with the highest BLOSUM62-sum score compared to the SARS-CoV-2 epitope set as reference was selected (Table 2 and Supplementary Tables 1, 2).We calculated the % of identity and similarity score S s with its related SARS-CoV-2 epitope, for each of these CCCs peptides (Supplementary Tables 1-3).The peptide similarity score S s calculation is based on the method reported by Sune Frankild et al. (58) and the BLOSUM62 m a t r i x t o c a l c u l a t e a B L O S U M 6 2 s u m ( u s i n g t h e Bio.SubsMat.MatrixInfo package in Biopython) between a pair of peptides (peptide "x" from SARS-CoV-2 and "y" from one CCC) and compared their similarity.0 ≤ S s ≤ 1: the closest S s is to 1, the highest is the potential for T-cell cross-reactivity response toward the related pair of peptides (58).We used a threshold of S s ≥0.8 to discriminate between highly similar and non-similar peptides.(2) Then, we examined if other parts of each of the CCCs proteome (without restricting our search only to peptides present in CCCs homologous proteins) could contain better matching peptides than the CCCs peptides reported in Supplementary Tables 1-3.First, for each one of our 16 CD4 + and 27 CD8 + SARS-CoV-2 epitopes, we spanned the entire proteome of each CCCs using the Epitope Conservancy Tool (ECT: http://tools.iedb.org/conservancy/-with a conservancy threshold of 20%).All the CCCs peptides from the top query (i.e., with the highest % of identity) were reported for every four CCCs in Supplementary Tables 1-3.Second, among these returned top queries (peptides with the same highest % identity), we picked the one with the highest similarity score S s (bolded in Supplementary Tables 1-3-right column).( 3) We compared this peptide with the one previously found in Supplementary Table 1 based on MSA.When both methods returned the same peptide (from the same protein), we kept it (peptides highlighted in beige in Supplementary Tables 1-3).When both matching peptides (using the two different methods) were found to be different, we compared (i) %id MSA with %id ECT and (ii) S s MSA with S s ECT .If %id MSA ≤ % id ECT but S s MSA ≥ S s ECT , we kept the CCCs peptide found following the MSA method; however, if %id MSA ≤ %id ECT and S s MSA < S s ECT , we then picked the CCC peptide found using the ECT instead of the one found using MSA (peptides not highlighted in Supplementary Tables 1-3).
Using the %id and the calculated similarity score with the SARS-CoV-2 epitopes, all related CCCs' best-matching peptides are reported in Supplementary Tables 1-3.They were then evaluated based on their potential to induce a cross-reactive Tcell response (Supplementary Tables 1-3): (0), CCC best matching peptide with low to no potential to induce a cross-reactive response toward the corresponding SARS-CoV-2 epitope and vice versa (%id with the corresponding SARS-CoV-2 epitope < 67% and similarity score S s < 0.8); (0.5), CCC best matching peptide that may induce a cross-reactive response (%id with the corresponding SARS-CoV-2 epitope ≥ 67% OR similarity score S s ≥ 0.8); and (1), CCC bestmatching peptide is very likely to induce a cross-reactive response (%id ≥ 67% and S s ≥ 0.8).

Identification of potential cross-reactive peptides in non-CCC human pathogens and vaccines
We took advantage of the database generated by Pedro A. Reche (59).Queries to find matching peptides with our SARS-CoV-2derived CD4 + and CD8 + epitopes were performed from the data gathered; only peptides sharing a %id ≥ 67% with our corresponding SARS-CoV-2 epitope were selected.The corresponding similarity score S s was calculated.

Statistical analyses
To assess the linear negative relationship between COVID-19 severity and the magnitude of each SARS-CoV-2 epitope-specific Tcell response, correlation analysis using GraphPad Prism version 8 (La Jolla, CA) was performed to calculate Pearson correlation coefficients (R), coefficient of determination (R 2 ), and associated p-value (correlation statistically significant for p ≤ 0.05).The slope (S) of the best-fitted line (dotted line) was calculated in Prism by linear regression analysis.The same statistical analysis was performed to compare the cross-reactive pre-existing T-cell response in unexposed pre-pandemic healthy individuals (UPPHI) with the slope S (magnitude of the correlation between this epitope-specific T-cell response in SARS-CoV-2-infected patients and the protection against severe COVID-19).Absolute WBCs and lymphocyte cell numbers (per µL of blood, measured through BDT), corresponding lymphocytes percentages/ratio, flow cytometry data measuring CD3 + /CD8 + /CD4 + cell percentages and the percentages detailing the magnitude (Tetramer + T cell %), and the quality (% of PD1 + /TIGIT + , CTLA-4 + /TIM3 + or AIMs + cells) of the CD4 + and CD8 + SARS-CoV-2-specific T cells, were compared across groups and categories of disease severity by one-way ANOVA multiple tests.ELISpot SFCs data were compared by Student's t-tests.Data are expressed as the mean ± SD. Results were considered statistically significant at p ≤ 0.05.To evaluate whether the differences in frequencies of RT-PCR positivity to the four CCCs across categories of disease severity were significant, we used the Chi-squared test or Fisher's exact test.

Higher magnitudes of common cold coronavirus/SARS-CoV-2 cross-reactive CD4 + T-cell responses detected in unvaccinated asymptomatic COVID-19 patients
We first compared SARS-CoV-2-specific CD4 + T-cell responses in unvaccinated asymptomatic COVID-19 patients (those individuals who never develop any COVID-19 symptoms despite being infected with SARS-CoV-2) to unvaccinated symptomatic (those patients who developed severe to fatal COVID-19 symptoms) (Figure 1).We used 16 recently identified HLA-DR-restricted CD4 + T-cell epitopes that are highly conserved between human SARS-CoVs and CCCs (1).We enrolled 92 unvaccinated HLA-DRB1*01:01 + COVID-19 patients, who were genotyped using PCR (Supplementary Figure 1) and divided into six groups, based on the level of severity of their COVID-19 symptoms (from severity 5 to severity 0, assessed at discharge).Clinical and demographic characteristics of this cohort of COVID-19 patients are detailed in Table 1.Fresh PBMCs were isolated from these COVID-19 patients, on average within 4.8 days after reporting a first COVID-19 symptom or a first PCR-positive test (Table 1).PBMCs were then stimulated in vitro for 72 h using each of the 16 CD4 + T-cell peptide epitopes, as detailed in Materials and methods and illustrated in Supplementary Figure 2. The frequency of responding IFN-g-producing CD4 + T cells specific to individual epitopes was quantified, in each of the six groups of COVID-19 patients, using ELISpot assay (i.e., number of IFN-g-spot forming CD4 + T cells or "SFCs") (Figure 1).A positive IFN-g-producing CD4 + T-cell responses was determined as the mean SFCs > 50 per 0.5 × 10 6 PBMCs fixed as threshold.
Taken together, these results demonstrate that, like SARS-CoV-2specific CD4 + T cells, an overall higher magnitude of CCCs/SARS-CoV- Frequencies of white blood cells, lymphocytes, and CD3 + /CD4 + /CD8 + T cells in the blood of unvaccinated COVID-19 patients with various degrees of disease severity.(A) numbers of white blood cells (WBCs) and total lymphocytes per µl of blood (left two panels) and percentages and ratios of total lymphocytes among WBCs (right two panels) measured ex vivo by blood differential test (BDT) in unvaccinated COVID-19 patients with various degrees of disease severity (n = 147).(B) Averages/means of numbers and frequencies of CD3 + T cells and (C) of total CD4 + , and CD8 + T cells measured by flow cytometry from COVID-19 patients' PBMCs with various severity scores after 72 h of stimulation with a pool of 16 CD4 + and 27 CD8 + CCCs/SARS-CoV-2 cross-reactive epitope peptides.The right panels show representative dot plots from patients with disease severity scores from 0 to 5. Data are expressed as the mean ±SD.Results are representative of two independent experiments and were considered statistically significant at p ≤ 0.05 (one-way ANOVA).We next determined whether the low magnitudes of CCCs/ SARS-CoV-2 cross-reactive CD4 + and CD8 + T-cell responses detected in unvaccinated severely ill and fatal COVID-19 patients was a result of an overall deficit in the frequencies of total CD4 + and CD8 + T cells.Using a blood differential test (BDT), we compared the absolute numbers of white blood cells (WBCs) and bloodderived lymphocytes, vivo, in the unvaccinated COVID-19 patients (Figure 3A).
A significant increase in the numbers of WBCs was detected in unvaccinated COVID-19 patients with fatal outcomes, (i.e., patients with severity 5, ∼1.5to ∼2.6-fold) when compared with all the remaining five groups of unvaccinated COVID-19 patients (i.e., patients with severity 0, 1, 2, 3, and 4; p ≤ 0.02, Figure 3A-left panel).However, significantly lower absolute numbers of total lymphocytes were detected in the blood of unvaccinated COVID-19 patients with fatal outcomes (i.e., patients with severity 5) compared to unvaccinated COVID-19 patients with mild disease (i.e., patients with severity 1 and 2: ∼1.9to ∼2.3-fold decrease-p < 0.02) or to asymptomatic patients with no disease (i.e., patients with severity 0: ∼3.3-fold decrease-p < 0.0001) (Figure 3A-second panel from left).As a result, the more severe the disease, the lower the percentage of blood-derived lymphocytes within WBCs (Figure 3A-third panel from left), and the lower the ratio of lymphocyte/WBCs (Figure 3A-fourth panel from left).
Overall, these results indicate that unvaccinated severely ill COVID-19 patients and unvaccinated COVID-19 patients with fatal outcomes not only had a general leukocytosis but also lymphopenia, which developed as early as 4.8 days after reporting their first symptoms or their first PCR-positive test.Furthermore, we found a significant CD3 + T-cell lymphopenia positively associated with the onset of severe disease in unvaccinated COVID-19 patients (Figure 3B).On average, the two groups of unvaccinated severely ill COVID-19 patients and unvaccinated COVID-19 patients with fatal outcomes (i.e., patients with severity 3, 4, and 5) had a ∼1.9-fold decrease in absolute number of CD3 + T cells compared to three groups of unvaccinated asymptomatic COVID-19 patients with low to no severe disease (i.e., patients with severity 0, 1, and 2, Figure 3B, p < 0.001).Similarly, the numbers of total CD4 + and CD8 + T cells within CD3 + -gated cells were reduced early in the two groups of unvaccinated severely ill COVID-19 patients and unvaccinated COVID-19 patients with fatal outcomes (i.e., patients with severity 3, 4, and 5) compared to the three groups of unvaccinated asymptomatic COVID-19 patients with low to no severe disease (Figure 3C-left column graph).
Taken together, our findings demonstrate that, compared to asymptomatic COVID-19 patients who presented with little to no disease, the severely ill patients and patients with fatal COVID-19 outcomes showed the following: (i) a broad and early lymphopenia (and leukocytosis), (ii) a decrease of bulk CD3 + T-cell lymphocytes number (equally affecting CD4 + and CD8 + T cells), and (iii) a reduction in CD4 + and CD8 + T cells specific to highly conserved CCCs/SARS-CoV-2 cross-reactive epitopes from structural, nonstructural, and accessory protein antigens.
Unvaccinated severely ill COVID-19 patients present high frequencies of phenotypically and functionally exhausted CCCs/SARS-CoV-2 cross-reactive CD4 + and CD8 + T cells, detected both ex vivo and in vitro We next compared the phenotype and function of CD4 + and CD8 + T cells specific to CCCs/SARS-CoV-2 cross-reactive epitopes in unvaccinated asymptomatic COVID-19 patients, with little to no disease, versus the unvaccinated severely ill COVID-19 patients and the unvaccinated COVID-19 patients with fatal outcomes.
As expected, no differences were observed in phenotypic and functional exhaustion of EBV BMLF-1 280-288 -specific CD8 + T cells across the six groups of COVID-19 patients with various disease severities (Supplementary Figure 3B), suggesting that the exhaustion of CD4 + and CD8 + T cells in severely ill COVID-19 patients and to patients with fatal COVID-19 outcomes was specific to CCCs/SARS-CoV-2 cross-reactive epitopes.
Altogether, these results (i) indicate that phenotypic and functional exhaustion of CD4 + and CD8 + T cells, detected both ex vivo and in vitro, specific to highly conserved and CCCs/SARS-CoV-2 cross-reactive epitopes from both structural and nonstructural antigens was associated with symptomatic and fatal infections in unvaccinated COVID-19 patients and (ii) suggest the importance of functional CCCs/SARS-CoV-2 cross-reactive CD4 + and CD8 + T cells, directed toward structural, nonstructural, and accessory protein antigens, for protection against symptomatic and fatal infections in unvaccinated COVID-19 patients.

Higher rates of co-infection with alpha common cold coronavirus 229E present unvaccinated asymptomatic COVID-19 patients
We next compared the co-infection with each of the four main and seasonal a and b CCCs (i.e., a-CCC-NL63, a-CCC-229E, b- CCC-HKU1, and b-CCC-OC43) in a cohort of 85 unvaccinated COVID-19 patients divided into six groups based on the severity of COVID-19 symptoms, as above (i.e., patients with severity 5 to severity 0, Figures 7A, B).Using RT-PCR performed on nasopharyngeal swab samples, we found co-infection with the a-CCC species to be more common with significantly higher rates in the asymptomatic COVID-19 patients (i.e.unvaccinated naturally protected from severe symptoms) compared to severely ill COVID-19 patients and to unvaccinated patients with fatal outcomes (i.e., unvaccinated that were not naturally protected from severe symptoms) (Figure 7A-right panel; ∼2.6-fold increase in groups 1-2-3 versus groups 4-5-6 of disease severity; p = 0.0418 calculated with Fisher's exact test).Co-infection with the CoV-229E a-CCC species was more common with significantly higher rates in the unvaccinated asymptomatic COVID-19 patients compared to unvaccinated severely ill COVID-19 patients and unvaccinated patients with fatal outcomes (Figure 7B, right panels: ∼4.2-fold increase between unvaccinated asymptomatic COVID-19 patients and unvaccinated severely ill COVID-19 patients (i.e., patients with severity of 4-5-6; p = 0.0223).However, there was no significant difference in the rates of co-infection with b-CCC species (nor with any of the four CCC species) across all six groups of COVID-19 patients with various severity symptoms (Figure 7A, central and left panels, and Figure 7B, left two panels).As illustrated in Figure 8, these results indicate that (i) compared to severely ill COVID-19 patients and patients with fatal COVID-19 outcomes, the asymptomatic COVID-19 patients presented significantly higher rates of co-infection with the a-CCC species, and with the 229E of a-CCCs, in particular and (ii) suggest that co-infection with the a species of CCCs (particularly the 229E species of a-CCCs, but not the b species) was associated with the natural protection from symptomatic and fatal infections in unvaccinated COVID-19 patients with yet-to-be-determined mechanisms(s).High frequencies of a-CCCs/SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells are associated with natural protection from symptomatic and fatal infections in unvaccinated COVID-19 patients Next, we determined whether (i) the higher rates of co-infection with a-CCC species observed in the unvaccinated asymptomatic COVID-19 patients were associated with high frequencies of CCCs/ SARS-CoV-2 cross-reactive CD4 + and CD8 + T cells detected in these asymptomatic COVID-19 groups and (ii) the high frequencies of a-CCCs/SARS-CoV-2 cross-reactive epitope-specific CD4 + and CD8 + T cells were associated with fewer symptoms observed in unvaccinated COVID-19 patients.To this end, we determined the percentage of unvaccinated asymptomatic COVID-19 patients, unvaccinated severely ill COVID-19, and unvaccinated patients with fatal outcomes who presented significant IFN-g + CD4 + and IFN-g + CD8 + T-cell responses (i.e., IFN-g -ELISpot SFCs > 50) specific to a-CCCs/SARS-CoV-2 cross-reactive epitopes.
Taken together, these results demonstrate that, compared to low proportions of severely ill COVID-19 patients and patients with fatal outcomes, significant proportions of both unvaccinated asymptomatic COVID-19 patients and unexposed pre-pandemic healthy individuals (UPPHI) presented significant a-CCCs/SARS-CoV-2 strong cross-reactive CD4 + and CD8 + T-cell responses.These findings suggest a crucial role of functional a-CCCs/SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells, induced following previous a-CCC seasonal exposures, in protection against subsequent severe symptomatic SARS-CoV-2 infection, as illustrated in Figure 8.
Cross-reactive CD4 + and CD8 + T-cell epitopes from a-CCCs and SARS-CoV-2 that present high similarity and identity are associated with natural protection from symptomatic and fatal infections in unvaccinated COVID-19 patients Using both the Multiple Sequences Alignments (MSA) and the Epitope Conservancy Tool (ECT) algorithms and software, we determined the identity (%id) and the similarity scores (S s ) of cross-reactive CD4 + and CD8 + T-cell epitopes, between the four major CCC species (a-hCCC-NL63, a-hCCC-229E, and b-hCCC-HKU1, b-hCCC-OC43), on the one hand, and SARS-CoV-2, on the other hand, as described in Materials and methods (58), (Table 2 and Supplementary Tables 1-3).
Next, we determined whether the CCCs/SARS-CoV-2-crossreactive epitopes were cross-recognized preferentially by the CD4 + and CD8 + T cells from either unvaccinated asymptomatic COVID-19 patients, or unvaccinated severely ill COVID-19 patients and unvaccinated patients with fatal outcomes (Supplementary Table 4).No significant differences were detected when the slopes S of the SARS-CoV-2-specific CD4 + and CD8 + T-cell responses were applied towards epitopes that have no significant identity nor similarity to epitopes from the four CCCs.Significant differences were detected when the slopes S of the SARS-CoV-2-specific CD4 + and CD8 + T-cell responses were applied to epitopes that have significant identity and/or similarity to epitopes from at least one of the four CCCs (Supplementary Table 4).In contrast, SARS-CoV-2 CD4 + or CD8 + T cells cross-recognizing epitopes that are highly identical and similar exclusively in b-CCC species, but not in a-CCC species (i.e., epitopes ORF1ab 5019-5033 and ORF1ab 3013-3021 ), presented a significantly lower slope S (p = 0.04) (Supplementary Table 4).The ORF1ab 5019-5033 and ORF1ab 3013-3021 epitopes have slopes S close to 0 among all epitopes (Supplementary Table 4).These data indicated that (i) CCCs/SARS-CoV-2-cross-reactive CD4 + or CD8 + T-cell epitopes that share high identity and similarity exclusively with the a-CCC species were crossrecognized mainly by CD4 + or CD8 + T cells from asymptomatic COVID -19 patients; (ii) in contrast, the CCCs/SARS-CoV-2-crossreactive CD4 + or CD8 + T cell epitopes that share high identity and similarity exclusively with the b-CCC species were cross-recognized mainly by CD4 + or CD8 + T cells from severely ill symptomatic patients; and (iii) compared to severely ill COVID-19 patients and patients with fatal outcomes, the asymptomatic COVID-19 patients presented significantly higher frequencies of a-CCCs/SARS-CoV-2 cross-reactive CD4 + and CD8 + T cells.The findings suggest a crucial role of functional, poly-antigenic a-CCCs/SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells, induced following previous a-CCC seasonal exposures, in protection against subsequent severe symptomatic SARS-CoV-2 infection.

Discussion
Characterizing the underlying T-cell mechanisms associated with protection against COVID-19 severity in unvaccinated asymptomatic patients is a challenging task today, since most individuals have received at least one dose of COVID-19 vaccine (39).Only 15.2% of adults in the United States are unvaccinated (37,38).This study is one of the few to comprehensively characterize the cross-reactive memory CD4 + and CD8 + T cells in unvaccinated symptomatic and asymptomatic COVID-19 patients.We compared the antigen specificity, frequency, phenotype, and function of CCCs/SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells, cross-recognizing genome-wide conserved epitopes in a cohort of 147 unvaccinated COVID-19 patients, divided into six groups based on the severity of their symptoms.The findings demonstrate several relationships between antigen-specific T-cell responses and disease outcome.Specifically, severely ill symptomatic COVID-19 patients who required admission to intensive care units (ICUs) and patients with fatal COVID-19 outcomes, versus unvaccinated asymptomatic COVID-19 patients, displayed significantly (i) higher rates of co-infection with the 229E alpha species of CCCs (a-CCC-229E); (ii) higher frequencies of a-CCCs/SARS-CoV-2 cross-reactive functional memory CD134 + CD137 + CD4 + and CD134 + CD137 + CD8 + T cells, directed toward conserved epitopes from structural, non-structural, and accessory SARS-CoV-2 proteins; and (iii) lower frequencies of CCCs/SARS-CoV-2 cross-reactive and exhausted PD-1 + TIM3 + TIGIT + CTLA4 + CD4 + and PD-1 + TIM3 + TIGIT + CTLA4 + CD8 + T cells.These observations (i) support a crucial role for functional, poly-antigenic a-CCCs/SARS-CoV-2 crossreactive memory CD4 + and CD8 + T cells, induced following previous a-CCC seasonal exposures, in protection against subsequent severe symptomatic SARS-CoV-2 infection and (ii) provide critical insights into developing broadly protective, multiantigen, CD4 + and CD8 + T-cell-based, universal pan-Coronavirus vaccines capable of conferring cross-species protection.
The present comprehensive study of cross-reactive SARS-CoV-2 epitope-specific CD4 + and CD8 + T cells suggests that prepandemic exposure to seasonal a-CCC species, but not to b-CCC species, may have conferred protection from symptomatic COVID-19 infections by an as-yet-to-be-determined mechanism(s).It is likely that pre-existing CCCs/SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells, induced in UPPHI by seasonal a-CCC species, cross-recognized protective SARS-CoV-2 epitopes.These data are consistent with previous studies showing that high levels of CCCs immunity in convalescent patients are associated with improved survival in COVID-19 patients (60,61).
Because severely ill patients preferentially developed higher frequencies of co-infection with b-CCC species and higher frequencies of pre-existing b-CCCs/SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells, T-cell exhaustion may be related to prior exposure to seasonal b-species of CCCs.The present study has comprehensively characterized CCCs/ SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells in blood samples from over 140 unvaccinated symptomatic and asymptomatic COVID-19 patients.However, there remain several gaps in our understanding.First, the study of CCCs/SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells in unvaccinated symptomatic and asymptomatic COVID-19 patients has not been adjusted retrospectively to previous CCCs infections, due to the lack of pre-COVID-19 samples.At this point, the vast majority of adults in the United States have been infected and/or received at least one dose of the COVID-19 vaccine (37,38); thus, going forward, characterizing pre-COVID-19 cross-reactive memory CD4 + and CD8 + T cells in unvaccinated COVID 19 patients will be very difficult (39).Second, the study did not follow up with the COVID-19 patients at later times points after convalescence; hence, the reported CCCs/SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cell characteristics are reflective of their status shortly after exposure to SARS-CoV-2 or during the symptomatic disease.Although we assessed SARS-CoV-2-specific CD4 + and CD8 + T cell responses at an early stage of the disease (blood sampled on average 5 days after the appearance of the first reported symptoms), the precise timing of the patient's first exposure to SARS-CoV-2 is not known.Third, since the T-cell responses reported in this study were assessed in the peripheral blood, this may not reflect tissue-resident CD4 + and CD8 + T cells in the lungs and the brain.The reduced number of functional CCCs/SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells detected in the peripheral blood of symptomatic COVID-19 patients may be due to T-cell redistribution to other organs, such as the lungs and the brain.The asymptomatic infections in unvaccinated COVID-19 patients might be attributed to homing and redistribution of high numbers of functional CCCs/SARS-CoV-2 cross-reactive CD4 + and CD8 + T cells into the lungs of unvaccinated asymptomatic COVID-19 patients, rather than in peripheral blood.In this context, we recently reported that high frequencies of functional lung-resident memory CD4 + and CD8 + T cells contributed to protection against COVID-19-like symptoms and death caused by SARS-CoV-2 infection in a mouse model (2).Thus, future studies should investigate tissue-resident CD4 + and CD8 + T cells in the lungs to determine whether their frequency and function correlate with protection from symptomatic and fatal infections in unvaccinated COVID-19 patients.Finally, while the study enrolled 600 patients overall, the study compared the antigen specificity, frequency, phenotype, and function of common cold coronaviruses (CCCs) and SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells, targeting genome-wide conserved epitopes in a cohort of 147 unvaccinated COVID-19 patients screened for two HLA types, HLA-DRB1*01:01 and HLA-A*02:01.Thus, future studies are being conducted to assess T cells from other HLA types.Nevertheless, our results are consistent with the hypothesis that the early presence of high numbers of functional a-CCCs/SARS-CoV-2 cross-reactive CD4 + and CD8 + T cells targeting multiple antigens was associated with protection from symptomatic and fatal SARS-CoV-2 infections in unvaccinated COVID-19 patients (99).
This report also confirms previous reports that (i) early and broad lymphopenia positively correlated with COVID-19 disease severity and mortality (86, 100-102); (ii) broad leukocytosis combined with T cell lymphopenia was present in severe COVID-19 patients and extended those findings by demonstrating that the observed T-cell lymphopenia was particularly prevalent for SARS-CoV-2-specific T cells (86,100); and (iii) a significant age-dependent and comorbidityassociated susceptibility to COVID-19 disease, with patients over 60 years of age, and those with pre-existing diabetic and hypertension comorbidities being the most susceptible to severe COVID-19 disease (13,20).
In conclusion, the present comprehensive analysis of specific and cross-reactive SARS-CoV-2 epitope-specific T cells reveals clear relationships between T-cell responses and disease outcomes in unvaccinated COVID-19 patients.Compared to severely ill COVID-19 patients and patients with fatal COVID-19 outcomes, the asymptomatic COVID-19 patients presented high rates of coinfection with the a-CCC species and more functional and less exhausted a-CCCs/SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells, targeting structural, non-structural, and accessory proteins.The findings suggest functional, poly-antigenic a-CCCs/ SARS-CoV-2 cross-reactive memory CD4 + and CD8 + T cells, induced following CCCs repetitive exposures, are contributing factors in reducing the severity of SARS-CoV-2 infection, as illustrated in Figure 8.Most of the >10 billion doses of firstgeneration COVID-19 vaccines are based on the Spike antigen alone (103, 104) and function mainly by inducing neutralizing antibodies (105).Because the Spike protein has undergone a substantial number of mutations with each successive viral variant, these first-generation subunit vaccines are susceptible to immune evasion by new variants and subvariants, such as XBB.1.5,EG.5 (Eris), and HV.1 sub-variants of Omicron (71,72).To overcome this critical limitation, the next generation of COVID-19 vaccines should also target other highly conserved structural and non-structural SARS-CoV-2 antigens capable of inducing protection by cross-reactive CD4 + and CD8 + T cells (1,106).Herein, the findings of this report provide a roadmap for developing next-generation a-CCCs/SARS-CoV-2 cross-reactive CD4 + and CD8 + T cell-based, multi-antigen, pan-Coronavirus vaccines capable of conferring cross-species protection.

5 *
To assess (for each individual SARS-CoV-2 epitope) the magnitude of the correlation between the breadth of this epitope-specific T-cell response and the protection against severe COVID-19 Matching CCCs peptides were chosen after combining both MSA and ECT analysis (see Materials and methods).Each panel represents the alignment of epitopes from SARS-CoV-2 and the four main and seasonal a and b species of common cold coronavirus (CCCs) (i.e., a-CCC-NL63, a-CCC-229E, b-CCC-HKU1, and b-CCC-OC43).The SARS-CoV-2 peptide sequence is set as 100% identity.The amino acids color code was generated with Gecos software (https://gecos.biotite-python.org)using the following parameters: geckos -matrix BLOSUM62 -min 60 -max 75 -f.The distance between two amino acids in the substitution matrix (BLOSUM62) corresponds to the perceptual visual differences in the color scheme.Similarity scores (S S ) based on such matrix are a good predictive measure of potential cross-reactivity (along with % of peptide identity).S S ≥ 0.80 and %id ≥ 67% are in red.Identity percentages, similarity scores, conservation, and consensus sequences are indicated in each panel.For each SARS-CoV-2 epitope, the significance (p < 0.05) of each correlation is also indicated, along with the magnitude of the T-cell cross-reactive response measured by IFN-g ELISpots in HD individuals.
2 cross-reactive CD4 + T-cell responses present in unvaccinated asymptomatic COVID-19 patients.In contrast, a lower magnitude of CCCs/SARS-CoV-2 cross-reactive CD4 + T-cell responses were detected in unvaccinated severely ill COVID-19 patients and to patients with fatal COVID-19 outcomes and (ii) suggest a crucial role of CCCs/SARS-CoV-2 crossreactive CD4 + T cells, directed towards structural, non-structural, and accessory protein antigens, in protection from symptomatic and fatal Infections in unvaccinated COVID-19 patients.Higher magnitudes of CD8 + T-cell responses to common cold coronavirus/ SARS-CoV-2 cross-reactive epitopes detected in unvaccinated asymptomatic COVID-19 patients We next compared the CCCs/SARS-CoV-2 cross-reactive CD8 + T-cell responses in unvaccinated asymptomatic individuals vs.

1 IFN
FIGURE 1 IFN-g-producing CD4 + T-cell responses to CCCs/SARS-CoV-2 cross-reactive epitopes in unvaccinated COVID-19 patients with various degrees of disease severity.PBMCs from HLA-DRB1*01:01-positive COVID-19 patients (n = 92 are HLA-DRB1*01:01-positive out of 600 tested) were isolated and stimulated for a total of 72 h with 10 µg/ml of each of the previously identified 16 CCCs/SARS-CoV-2 cross-reactive CD4 + T cell epitope peptides.The number of IFN-g-producing CD4 + T cells was quantified in each of the 92 patients using ELISpot assay.(A) Average/mean numbers (± SD) of IFN-g-spot forming cells (SFCs) after CD4 + T-cell peptide-stimulation detected in each of the 92 COVID-19 patients divided into six groups based on disease severity scored 0-5, as described in Materials and methods, and as identified by six columns on a grayscale (black columns = severity 5, to white columns = severity 0) is shown.Dotted lines represent an arbitrary threshold set as a cutoff of the positive response.A mean SFC between 25 and 50 SFCs corresponds to a medium/intermediate response, whereas a strong response is defined for mean SFCs > 50 per 0.5 × 10 6 stimulated PBMCs.(B) Correlation between the overall number of IFN-g-producing CD4 + T cells induced by each of the 16 CCCs/SARS-CoV-2 cross-reactive CD4 + T-cell epitope peptides in each of the six groups of COVID-19 patients with various disease severity.The coefficient of determination (R 2 ) is calculated from the Pearson correlation coefficients (R).The associated p-value and the slope (S) of the best-fitted line (dotted line) calculated by linear regression analysis are indicated.The gray-hatched boxes in the correlation graphs extend from the 25th to 75th percentiles (hinges of the plots) with the median represented as a horizontal line in each box and the extremity of the vertical bars showing the minimum and maximum values.(C) Representative spots images of the IFN-g-spot forming cells (SFCs) induced by each of the 16 CCCs/SARS-CoV-2 cross-reactive CD4 + T cell epitope peptides in three representative patients, each falling into one of three groups of disease category: the unvaccinated asymptomatic COVID-19 patients (ASYMP, severity score 0), unvaccinated COVID-19 patients who developed mild to moderate disease (severity scores 1 and 2) and unvaccinated severely ill COVID-19 patients and unvaccinated patients with fatal COVID-19 outcomes (severity scores 3-5).PHA was used as a positive control of T-cell activation.Unstimulated negative control SFCs (DMSO-no peptide stimulation) were subtracted from the SFC counts of peptides-stimulated cells.Results are representative of two independent experiments and were considered statistically significant at p ≤ 0.05.

2 IFN
FIGURE 2 IFN-g-producing CD8 + T-cell responses to CCCs/SARS-CoV-2 cross-reactive epitopes in unvaccinated COVID-19 patients with various degrees of disease severity.PBMCs from HLA-A*02:01-positive COVID-19 patients (n = 71) were isolated and stimulated for a total of 72 h with 10 µg/ml of each of the previously identified 27 CCCs/SARS-CoV-2 cross-reactive CD8 + T-cell epitope peptides.The number of IFN-g-producing CD8 + T cells was quantified in each of the 71 patients using ELISpot assay.Panel (A) shows the average/mean numbers (± SD) of IFN-g-spot forming cells (SFCs) after CD8 + T-cell peptide stimulation detected in each of the 71 COVID-19 patients divided into six groups based on disease severity scored 0-5, as described in Materials and methods, and as identified by six columns on a grayscale (Black columns = severity 5, to white columns = severity 0).Dotted lines represent an arbitrary threshold set as a cutoff of the positive response.A mean SFCs between 25 and 50 SFCs corresponds to a medium/intermediate response, whereas a strong response is defined for mean SFCs > 50 per 0.5 × 10 6 stimulated PBMCs.(B) Correlation between the overall number of IFN-g-producing CD8 + T cells induced by each of the 27 CCCs/SARS-CoV-2 cross-reactive CD8 + T-cell epitope peptides in each of the six groups of COVID-19 patients with various disease severity.The coefficient of determination (R 2 ) is calculated from the Pearson correlation coefficients (R).The associated p-value and the slope (S) of the best-fitted line (dotted line) calculated by linear regression analysis are indicated.The gray-hatched boxes in the correlation graphs extend from the 25th to 75th percentiles (hinges of the plots) with the median represented as a horizontal line in each box and the extremity of the vertical bars showing the minimum and maximum values.(C) Representative spots images of the IFN-g-spot forming cells (SFCs) induced by each of the 27 CCCs/SARS-CoV-2 cross-reactive CD8 + epitope peptides in three representative patients, each falling into one of three groups of disease category: the unvaccinated asymptomatic COVID-19 patients (ASYMP, severity score 0), unvaccinated COVID-19 patients who developed mild to moderate disease (severity scores 1 and 2), and unvaccinated severely ill COVID-19 patients and unvaccinated patients with fatal COVID-19 outcomes, (severity scores 3-5).PHA was used as a positive control of T-cell activation.Unstimulated negative control SFCs (DMSO-no peptide stimulation) were subtracted from the SFC counts of peptides-stimulated cells.Results are representative of two independent experiments and were considered statistically significant at p ≤ 0.05.

Frequencies
of CCCs/SARS-CoV-2 cross-reactive CD4 + and CD8 + T cells in unvaccinated COVID-19 patients with various degrees of disease severity.PBMCs from HLA-DRB1*01:01-positive (n = 92) (A) or HLA-A*02:01-positive (n = 71) (B) unvaccinated COVID-19 patients with various degrees of disease severity were isolated and stimulated for 72 h with 10 mg/ml of indicated CCCs/SARS-CoV-2 cross-reactive CD4 + and CD8 + epitope peptides.The induced CD4 + and CD8 + T cells were then stained and analyzed by flow cytometry.The indicated epitope peptides were chosen among the CCCs/SARS-CoV-2 cross-reactive 16 CD4 + and 27 CD8 + epitope peptides based on tetramer availability.Panel (A) shows representative dot plots (left panels) and average frequencies of CCCs/SARS-CoV-2 cross-reactive CD4 + T cells (right panel) detected in three representatives COVID-19 patients, each falling into one of three groups of disease category: the unvaccinated asymptomatic COVID-19 patients (ASYMP, severity score 0), unvaccinated COVID-19 patients who developed mild to moderate disease (severity scores 1 and 2), and unvaccinated severely ill COVID-19 patients and unvaccinated patients with fatal COVID-19 outcomes (severity scores 3-5).Panel (B) shows representative dot plots (left panels) and average frequencies of CCCs/SARS-CoV-2 cross-reactive CD8 + T cells (right panel) detected in three representatives of COVID-19 patients and in panel (A).Data are expressed as the mean ± SD. Results are representative of two independent experiments and were considered statistically significant at p ≤ 0.05 (one-way ANOVA).

5
FIGURE 5 Co-expression of exhaustion and activation markers on CCCs/SARS-CoV-2 cross-reactive CD4 + T cells from unvaccinated COVID-19 patients with various degrees of disease severity.PBMCs from HLA-DRB1*01:01-positive unvaccinated COVID-19 patients with various degrees of disease severity were isolated and stimulated for 72 h with 10 mg/ml of five CCCs/SARS-CoV-2 cross-reactive CD4 + T-cell epitope peptides.The induced CD4 + T cells were then stained and analyzed by flow cytometry for the frequency of tetramer-specific CD4 + cells co-expressing exhaustion and activation markers.Panel (A) shows representative dot plots (upper panels) and average (lower panels) frequencies of CCCs/SARS-CoV-2 cross-reactive CD4 + T cells expressing exhaustion markers PD1/TIGIT and TIM-3/CTLA-4 detected in three representative groups of unvaccinated COVID-19 patients with various degrees of disease severity.Panel (B) shows representative dot plots (upper panels) and average (lower panels) frequencies of CCCs/ SARS-CoV-2 cross-reactive CD4 + T cells expressing activation markers (AIMs) CD134/CD137 detected in three representative groups of unvaccinated COVID-19 patients with various degrees of disease severity.Results are representative of two independent experiments, and data are expressed as the mean ± SD and were considered statistically significant at p ≤ 0.05 calculated using one-way ANOVA.

6
FIGURE 6 Co-expression of exhaustion and activation markers on CCCs/SARS-CoV-2 cross-reactive CD8 + T cells from unvaccinated COVID-19 patients with various degrees of disease severity.PBMCs from HLA-A*02:01-positive unvaccinated COVID-19 patients with various degrees of disease severity were isolated and stimulated for 72 h with 10 mg/ml of five CCCs/SARS-CoV-2 cross-reactive CD8 + T-cell epitope peptides.The induced CD8 + T cells were then stained and analyzed by flow cytometry for the frequency of tetramer-specific CD8 + cells co-expressing exhaustion and activation markers.Panel (A) shows representative dot plots (upper panels) and average frequencies of CCCs/SARS-CoV-2 cross-reactive CD8 + T cells (lower panel) expressing exhaustion markers PD1/TIGIT and TIM-3/CTLA-4 detected in three representative groups of unvaccinated COVID-19 patients with various degrees of disease severity.Panel (B) shows representative dot plots (upper panels) and average frequencies of CCCs/SARS-CoV-2 cross-reactive CD8 + T cells (lower panel) expressing activation markers (AIMs) CD134/CD137 detected in three representative groups of unvaccinated COVID-19 patients with various degrees of disease severity.Results are representative of two independent experiments, and data are expressed as the mean ± SD and were considered statistically significant at p ≤ 0.05 (one-way ANOVA).

7
FIGURE 7 Rates (frequency) of co-infection with seasonal common cold coronavirus species a-CCC-NL63, a-CCC-229E, b-CCC HKU1, and b-CCC-OC43 in unvaccinated COVID-19 patients with various degrees of disease severity.Four major human common cold coronaviruses species, CCC-HKU1, CCC-OC43, CCC-229E, and CCC-NL63, were detected using RT-PCR in the nasopharyngeal swabs of COVID-19 patients (n = 85, first column) who developed various disease severity.Panel (A) shows all four a-CCCs and b-CCCs species (left panel), b-CCC species alone (middle panel), and a-CCC species alone (right panel), detected in unvaccinated severely ill COVID-19 patients and unvaccinated patients with fatal COVID-19 outcomes (severity scores 3-4-5) vs. unvaccinated COVID-19 patients who developed no, mild, and moderate disease (severity score 1-2-3).(B) The rate (%) of co-infection with each one of the four major species, CCC-HKU1, CCC-OC43, CCC-229E, and CCC-NL63, detected in unvaccinated severely ill COVID-19 patients and unvaccinated patients with fatal COVID-19 outcomes (severity scores 3-4-5), in unvaccinated COVID-19 patients who developed mild to moderate disease (severity score 1-2), and in unvaccinated asymptomatic COVID-19 patients (severity score 0).The p-values calculated using the Chi-squared test compare the rate (%) of co-infection with each CCC species between unvaccinated COVID-19 patients with various degrees of disease severity.Results are representative of two independent experiments, and data are expressed as the mean ± SD and were considered statistically significant at p ≤ 0.05 calculated using Fisher's exact test.

FIGURE 8
FIGURE 8 Illustration showing higher frequencies of common cold coronavirus/SARS-CoV-2 cross-reactive CD4 + and CD8 + T cells detected in unvaccinated asymptomatic COVID-19 patients is associated with higher rates of co-infection with alpha common cold coronavirus strain 229E (a-CCC-229E).The first row shows increasing copies of a-CCC-229E detected in unvaccinated asymptomatic COVID-19 patients compared to unvaccinated symptomatic COVID-19 patients.The middle row shows increasing numbers of common cold coronavirus/SARS-CoV-2 cross-reactive CD4 + and CD8 + memory T cells detected in unvaccinated asymptomatic COVID-19 patients compared to unvaccinated symptomatic COVID-19 patients.The bottom row shows symptoms detected in unvaccinated COVID-19 patients with symptoms increasing from severity 0 in asymptomatic COVID-19 patients (left) to severity 5 in COVID-19 patients with fatal COVID-19 outcomes (right) as detailed in Materials and methods.".

TABLE 1
Demographic, age, HLA-genotyping, clinical parameters, and prevalence of comorbidities in unvaccinated COVID-19 patients with various degrees of disease severity.