key: cord-0970753-991ob7ad
authors: Alrubayyi, Aljawharah; Gea-Mallorquí, Ester; Touizer, Emma; Hameiri-Bowen, Dan; Kopycinski, Jakub; Charlton, Bethany; Fisher-Pearson, Natasha; Muir, Luke; Rosa, Annachiara; Roustan, Chloë; Earl, Christopher; Cherepanov, Peter; Pellegrino, Pierre; Waters, Laura; Burns, Fiona; Kinloch, Sabine; Dong, Tao; Dorrell, Lucy; Rowland-Jones, Sarah; McCoy, Laura; Peppa, Dimitra
title: Characterization of humoral and SARS-CoV-2 specific T cell responses in people living with HIV
date: 2021-03-17
journal: Res Sq
DOI: 10.21203/rs.3.rs-309746/v1
sha: 116bdfbb6956d6deba54a9d96b47267c02c0cfb7
doc_id: 970753
cord_uid: 991ob7ad

There is an urgent need to understand the nature of immune responses against SARS-CoV-2, to inform risk-mitigation strategies for people living with HIV (PLWH). We show that the majority of PLWH, controlled on ART, mount a functional adaptive immune response to SARS-CoV-2. Humoral and SARS-CoV-2-specific T cell responses are comparable between HIV-positive and negative subjects and persist 5-7 months following predominately mild COVID-19 disease. T cell responses against Spike, Membrane and Nucleocapsid are the most prominent, with SARS-CoV-2-specific CD4 T cells outnumbering CD8 T cells. We further show that the overall magnitude of SARS-CoV-2-specific T cell responses relates to the size of the naive CD4 T cell pool and the CD4:CD8 ratio in PLWH, in whom disparate antibody and T cell responses are observed. These findings suggest that inadequate immune reconstitution on ART, could hinder immune responses to SARS-CoV-2 with implications for the individual management and vaccine effectiveness in PLWH.

The global outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), causing 39 COVID-19 disease, has resulted in an overall 3% case fatality rate, posing unprecedented 40 healthcare challenges around the world 1 ELISA was used to screen plasma samples for antibodies against the external Spike antigen, using 119 immobilized recombinant Spike S11-530 subunit protein (S1), and against immobilized full-length 120 internal Nucleoprotein (N) antigen to confirm prior infection as previously described 33,37,38 121 ( Fig.1a) . A sample absorbance greater than 4-fold above the average background of the assay was 122 regarded as positive, using a threshold established with pre-pandemic samples (Supplementary 123 Fig.1a , b) and as previously described 38 . The screening assay, followed by titer quantification 124 (based on an in-assay standard curve) 37 , demonstrated that 95.8% (23/24) of individuals from the 125 HIV positive group with prior laboratory confirmed COVID-19 and 30.43% (7/23) with suspected 126 disease, during the first wave of the pandemic, had measurable titers for SARS-CoV-2 S1 and N 127 sampled at a median 146 DPSO (DPSO range 46-232) and 181 DPSO (range 131-228), 128 respectively ( Fig.1a-c) . Similarly, in the HIV negative group with laboratory confirmed COVID-129 19 disease, 93.5% (29/31) had detectable SARS-CoV-2 antibodies to S1 and N at 146 DPSO (101-130 220), whereas none of the suspected/household contacts in this group (0/4) had quantifiable titers 131 (DPSO median 200; range 125-203) (Fig.1a-c) . S1 and N titers were found to be comparable 132 between the HIV positive and negative groups (Fig.1b, c) and correlated with one another, 133 although levels were heterogenous among donors as previously observed (Fig.1d) . 134

To determine whether the SARS-CoV-2 antibodies generated are able to inhibit SARS-CoV-2 136 infection, we employed a serum neutralisation assay with pseudotyped SARS-CoV-2, to calculate 137 the 50% inhibitory serum dilution (ID50) 28 . Overall, we detected similar neutralization levels 138 (Fig.1e ) and comparable profiles across the two study groups in terms of the number of individuals 139 with high potency, low potency or no neutralizing activity (<50 ID50) (Fig.1f) , which correlated 140 with anti-S1 IgG levels (Fig.1g) . A range of neutralizing antibodies (nAb) was detected in the 141 groups, with some samples exhibiting strong neutralization despite low S1 titers irrespective of 142 disease severity (Fig.1g) . 143

No association was observed between S1 binding titers, age and gender in the two groups 145 ( Supplementary Fig.1c) . A weak positive correlation was seen between neutralization levels and 146 age according to male gender in the HIV positive group, where subjects were older and females 147

were notably under-represented (Supplementary Fig.1d ). Neutralization levels did not correlate 148 with DPSO (Supplementary Fig.1e ) and were detectable up to 7 months post infection. No clear 149 association was observed according to ethnicity (Supplementary Fig.1f ). Together these results 150

show no significant differences in the IgG-specific antibody response to SARS-CoV-2 and 151 neutralization capacity according to HIV status after recovery from COVID-19 disease. These 152 findings should be considered in the context of this cohort in which the majority of cases were 153 mild and therefore may not reflect the full burden of disease associated with SARS-CoV-2 154 infection. 155

The presence of T helper 1 (Th1) immunity has been described in a number of studies investigating 158 T cell-specific immune responses to SARS-CoV-2 infection in various phases of the infection. We 159 therefore initially assessed global SARS-CoV-2 T cell frequencies by IFN-γ-ELISpot using 160 overlapping peptide (OLP) pools to detect T cell responses and cumulative frequencies directed 161 against defined immunogenic regions, including Spike, Nucleocapsid (N), membrane (M), 162

Envelope (Env) , and open reading frame (ORF)3a, ORF6, ORF7 and ORF8 (Fig.2a) . Out of the 163 30 HIV positive and 30 HIV negative individuals (including previously laboratory confirmed cases 164 and additional subjects found to be SARS-CoV-2 seropositive on screening), the majority of 165 donors in each group had a demonstrable cellular response directed predominately against Spike 166 and N/M. Responses to accessory peptide pools (ORFs) and the structural protein Env were less 167 frequent and significantly lower to other antigens observed, irrespective of HIV status ( Fig.2b and  168 Supplementary Fig.2a, b) . The overall magnitude of responses against Spike, M and N did not 169 differ significantly between the groups (Fig.2c) . In HIV positive donors the cumulative SARS-170

CoV-2 responses across all pools tested were lower in magnitude compared to that of T cells 171 directed against well-defined CD8 epitopes from Influenza, Epstein Barr Virus (EBV) and 172

Cytomegalovirus (CMV)-(FEC pools) tested in parallel within the same donors, but higher 173 compared to HIV-gag responses (Fig.2d) . By contrast, responses to FEC pools were comparable 174 in magnitude to the cumulative SARS-CoV-2-specific T cell responses detected in the HIV 175 negative donors, likely reflecting the lower CMV seropositivity in the HIV negative group 176 compared to the HIV positive group (54.28% CMV seropositive versus 97.87% CMV seropositive, 177 respectively) (Fig.2d ). In line with previous studies we observed a wide breadth and range of 178 cumulative SARS-CoV-2 T cell frequencies, with over 90% of donors in each group showing a 179 response (Fig.2e, f) 30, [39] [40] . However, the proportion of HIV positive and negative donors with T 180 cell responses to individual SARS-CoV-2 pools within given ranges varied, with a higher 181 percentage of HIV positive donors having low level responses (Fig.2g) . 182 183 Responses to Spike, M and N peptide pools were significantly higher in donors with confirmed 184 SARS-CoV-2 infection compared to subjects with no evidence of infection who displayed 185 relatively weak responses; small responses were also noted in a proportion of HIV positive subjects 186 with available pre-pandemic samples (Supplementary Fig.2c) . Additional work is required to 187 investigate potential cross-reactive components of these responses with other human 188 coronaviruses, as has been reported in other studies 31-32,39,41 . These data were derived from 189 cryopreserved samples, which may underestimate the magnitude of the detected responses 42 . 190 191 Given the considerable heterogeneity in the magnitude of the observed responses in both groups, 192

we related these to HIV parameters, age, gender and DPSO. We detected a positive correlation 193 between CD4:CD8 ratio and summed total responses to OLP pools against SARS-CoV-2 in HIV 194 positive subjects (r=0.3820, p=0.037) (Fig.2h) Fig.2d, e) . 196

These data suggest that, despite effective ART, incomplete immune reconstitution may potentially 197 impact on the magnitude of T cell responses to SARS-CoV-2. Previous observations have 198 demonstrated an association between SARS-CoV-2-specific T cells, age and gender, with T cell 199 immunity to Spike increasing with age and male gender in some studies 30 . Despite an older age 200 and male predominance in our HIV cohort we did not detect any association between ELISpot 201 responses to Spike and donor age (Supplementary Fig.2f, g) . There was no correlation between 202 DPSO and T cells directed either against Spike or total responses against SARS-CoV-2. These 203 responses were nonetheless detectable up to 232 DPSO (median 151 range 46-232) in HIV positive 204 subjects, and similarly in HIV negative donors (median 144; range 101-220) (Supplementary 205 Fig.2h, i) . Given that the majority of the donors, in both groups, experienced mild COVID-19 206 disease, any associations between the magnitude of responses and disease severity are limited. No 207 differences were observed in the magnitude of T cell responses according to ethnicity and gender, 208 irrespective of HIV status (Supplementary Fig.2j, k) . 209 210 T cell and antibody response complementarity 211 Next, we compared T cell responses, antibody levels and nAb responses in individual donors to 212 better understand any complementarity between humoral and cellular responses detected by IFN-213 γ-ELISpot. SARS-CoV-2-specific T cell responses correlated with antibody binding titers in the 214 HIV negative group (Fig.3a, c) . Although the majority of HIV positive subjects had detectable 215 antibody and T cell responses to SARS-CoV-2, the magnitude of the cellular immune responses 216 correlated weakly only with N IgG binding titers but not with S1 IgG binding titers (Fig.3b, d) . 217 We subsequently examined neutralization ID50 values for individual donors in relation to T cell 218 responses to individual SARS-CoV-2 antigen pools and summed responses. In HIV negative 219 donors a correlation was observed between T cell responses to Spike protein and ID50 (r= 0.4002, 220 p= 0.0315), with donors lacking a response to Spike generally maintaining low-frequency T cell 221 responses to other specificities. When cumulative responses were ranked by the magnitude of nAb 222 response, a single HIV negative donor with an ID50>1000 had no detectable SARS-CoV-2-223 specific T cells (Fig.3e) . In HIV positive donors no correlation was detected between 224 neutralization capacity and responses to individual SARS-CoV-2 peptides or pooled responses. A 225 single HIV positive donor (1/29) with undetectable neutralization activity, and another donor with 226 potent neutralization (>1000), had no measurable T cell response to any of the pools tested 227 (Fig.3f) . 228

Following the initial broad screening of the antiviral responses to SARS-CoV-2, intracellular 231 cytokine staining (ICS) was used to assess the composition and polyfunctionality of T cell 232 responses in a group of HIV positive (n=11) and HIV negative (n=12) donors with available PBMC 233 and detectable responses by IFN-γ-ELISpot. To determine the functional capacity of SARS-CoV-234 2-specific CD4 and CD8 T cells, we stimulated PBMCs with overlapping Spike, M and N (non-235 Spike) peptide pools, in addition to CMV pp65 and HIV gag peptides within the same individuals. 236

We focused on Spike, M and N as these antigens dominated responses detected by ELISpot. 237

Expression of the activation marker CD154 and production of IFN-γ, IL-2 and TNF-α were 238 measured as functional readouts (Fig.4a ). SARS-CoV-2-specific CD4 T cells directed against 239

Spike and non-Spike (M/N) predominantly expressed CD154 alone or in combination with IL-2, 240 TNF-α and IFN-γ, consistent with a Th1 profile, and these aggregated responses were comparable 241 between the groups (Fig.4a, b) . SARS-CoV-2-specific CD4 T cells exhibited polyfunctional 242 responses, with T cells expressing up to three cytokines (Fig.4c) . We detected no significant 243 differences in CD4 T cell responses, according to cytokine profile, to individual pools directed 244 against Spike, M and N in the two groups ( Fig.3c and Supplementary Fig.3a ). Aggregated CD4 245 T cell responses against all SARS-CoV-2 pools tested were higher compared to CMV-specific 246 responses and HIV-gag responses within the same donors (Supplementary Fig.3d, e) . 247 248 SARS-CoV-2-specific CD8 T cells largely expressed IFN-γ alone or in combination with TNF-α, 249 exhibiting a different cytokine profile to CD4 T cells as expected (Fig.4d) . A trend toward lower 250 mean aggregated CD8 T cell responses and polyfunctionality against Spike relative to non-Spike 251 was observed in HIV negative individuals ( Fig.4e-f ). Although SARS-CoV-2-specific CD8 T cell 252 responses did not differ significantly between the two groups, mean response frequency was lower 253 in HIV positive individuals against non-Spike pools (Fig.4e ). When we examined the individual 254 cytokine profile, depending on antigen specificity, IL-2 production was reduced in CD8 T cells 255 targeting non-Spike pools in HIV positive individuals compared to HIV negative donors 256 ( Supplementary Fig.3f-h) . The proportion of CD8 T cells specific for CMV was higher compared 257 to SARS-CoV-2-specific CD8 T cells irrespective of HIV status and similar to HIV-gag responses 258 ( Supplementary Fig.3i, j) . Notably, SARS-CoV-2-specific CD8 T cells against Spike and non-259

Spike pools were less frequent, with CD4 T cells similarly outnumbering CD8 T cells regardless Douek 1998), naïve T cell frequency was reduced in SARS-CoV-2 convalescent HIV positive 295 donors compared to HIV negative subjects. This was accompanied by higher proportions of 296 terminally differentiated effector memory (TEMRA: CD45RA + CCR7 − ) within the total CD8 T 297 cell population in HIV infected individuals, contributing to an altered representation of 298 naïve/memory T cells 50 (Fig.6c) . Notably, the percentage of naïve CD4 T cells correlated with the 299 CD4:CD8 ratio and SARS-CoV-2-specific T cell responses in HIV positive donors (Fig.6d, e) , 300 suggesting that the scarce availability of naïve CD4 T cells could influence the extent/magnitude 301 of the T cell response to SARS-CoV-2 infection. Recent data have demonstrated a link between 302 naïve CD4 T cells, age and COVID-19 disease severity in older individuals 18 . Whereas naïve CD8 303 and CD4 T cells correlated with age in HIV negative donors, this relationship between age and 304 naïve T cells was lost in HIV infected donors ( Fig.6f- CoV-2-specific parameters (Supplementary Fig.4a-d) . in the convalescent phase of COVID-19 disease has been described 55 . We detected elevated 319 frequencies of cTfh (CXCR5+PD-1+) HIV infected subjects compared to HIV infected donors, 320 however, no correlation was observed with antibody levels (Supplementary Fig.4e, f) 

Whole blood from all participants was collected in heparin-coated tubes and stored at room 470 temperature prior to processing. In brief, PBMCs were isolated by density gradient sedimentation. conditions. Responses that were found to be lower than two standard deviations of the sample 587 specific control were excluded. An additional threshold was set at > 5 SFU/10 6 PBMCs, and results 588 were excluded if positive control wells (PHA, FEC) were negative. 589 590 Intracellular cytokine stimulation (ICS) functional assay 591 ICS was performed as described previously 85 . Briefly, purified PBMCs were thawed and rested 592 overnight at 37 °C and 5% carbon dioxide in complete RPMI medium. After overnight rest, 593

PBMCs were stimulated for 6 h with 2µg/mL of SARS-CoV-2 peptide pools, Influenza, HIV-1 594 (DMSO) as a negative control in the presence of αCD28/αCD49d co-Stim antibodies (1 µg ml -1 ) 596

GolgiStop (containing Monensin, 2 µmol/L), GolgiPlug (containing brefeldin A, 10 µg ml -1 ) (BD 597 Biosciences) and anti-CD107α BV421 antibody (BD Biosciences). After stimulation, cells were 598 washed and stained with anti-CCR7 (BioLegend) for 30 min at 37 °C and then surface stained at 599 4°C for 20 min with different combinations of surface antibodies in the presence of fixable 600 live/dead stain (Invitrogen). Cells were then fixed and permeabilised (CytoFix/CytoPerm; BD 601 Biosciences) followed by intracellular cytokine with IFN-g APC, CD154 PE-Cy7 (BioLegend), 602 TNF-α FITC (BD Biosciences) and PerCP-eFluor 710 IL-2 (eBioscience). Samples were acquired 603 on a BD Fortessa X20 using BD FACSDiva8.0 (BD Bioscience) and data analysed using FlowJo 604 10 (TreeStar). The gates applied for the identification of virus-specific CD4 and CD8 T cells were 605 based on the double-positive populations for interferon-γ (IFN-g), Tumour necrosis factor (TNF-606 α), interleukin-2 (IL-2), and CD40 ligand (CD154). The total population of SARS-CoV-2 CD4 T 607 cells was calculated by summing the magnitude of CD154 + IFN-g + , CD154 + IL-2 + , and CD154 + 608 TNF-α + responses; SARS-CoV-2 CD8 T cells were defined as (IFN-g + TNF-α + , IFN-g + IL-2 + ). 609

Antibodies used in the ICS assay are listed in Supplementary Table 2 

OpenSAFELY platform

Risk factors for COVID-19 death in a population cohort study from the 868

Clinical infectious diseases : an official publication 869 of the Infectious Diseases Society of America

Immune deficiency is a risk factor for severe COVID-19 in people 871 living with HIV

Antigen-Specific Adaptive Immunity to SARS-CoV-2

Acute COVID-19 and Associations with Age and Disease Severity

Divergent SARS-CoV-2-specific T-and B-cell responses in severe but not 876 mild COVID-19 patients

Early induction of functional SARS-CoV-2-specific T cells associates with 879 rapid viral clearance and mild disease in COVID-19 patients

A brief history of T cell help to B cells

The multifaceted role of CD4(+) T cells in 884

CD8(+) T cell memory

Adaptive immunity to SARS-899

CoV-2 in cancer patients: The CAPTURE study

CD4 and CD8 Lymphocyte Counts as Surrogate Early Markers for 902 Progression in SARS-CoV-2 Pneumonia: A Prospective Study

Immunological memory to SARS-CoV-2 assessed for up to eight months 905 after infection. bioRxiv

Orthogonal SARS-CoV-2 Serological Assays Enable Surveillance of 907

Low-Prevalence Communities and Reveal Durable Humoral Immunity

Robust neutralizing antibodies to SARS-CoV-2 infection persist for 910 months

Longitudinal observation and decline of neutralizing antibody responses in 912 the three months following SARS-CoV-2 infection in humans

Convergent antibody responses to SARS-CoV-2 in convalescent 915 individuals

Discordant neutralizing antibody and T cell responses in 917 asymptomatic and mild SARS-CoV-2 infection

Robust T Cell Immunity in Convalescent Individuals with Asymptomatic 920 or Mild COVID-19

SARS-CoV-2-specific T cell immunity in cases of COVID-19 and SARS, 922 and uninfected controls

Preexisting and de novo humoral immunity to SARS-CoV-2 in humans

SARS-CoV-2-specific T cell responses and correlations with COVID-19 926 patient predisposition

HIV infection Worsens Age-Associated Defects in Antibody 929

Responses to Influenza Vaccine

One case of coronavirus disease

HIV with a low CD4(+) T-cell count

Broad and strong memory CD4(+) and CD8(+) T cells induced by SARS-944

CoV-2 in UK convalescent individuals following COVID-19

SARS-CoV-2-reactive T cells in healthy donors and patients with COVID-947 19

Loss of T cell responses following long-term cryopreservation

Persistent Cellular Immunity to SARS-CoV-2 Infection. bioRxiv

Marked T cell activation, senescence, exhaustion and skewing towards 953 TH17 in patients with COVID-19 pneumonia

Functional exhaustion of antiviral lymphocytes in COVID-19 patients

Spike-specific circulating T follicular helper cell and cross-neutralizing 979 antibody responses in COVID-19-convalescent individuals

Dysfunction in Aging and HIV-1 Infection

Evolution of Antibody Immunity to SARS-CoV-2. bioRxiv

DNA vaccine protection against SARS-CoV-2 in rhesus macaques

Safety and Immunogenicity of SARS-CoV-2 mRNA-1273 Vaccine 989 in Older Adults

Safety and immunogenicity of the ChAdOx1 nCoV-19 vaccine 992 against SARS-CoV-2: a preliminary report of a phase 1/2, single-blind, randomised 993 controlled trial

Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine. The 995

Antibody Status and Incidence of SARS-CoV-2 Infection in Health 997

Do antibody positive healthcare workers have lower SARS-CoV-2 infection rates 1003 than antibody negative healthcare workers? Large multi-centre prospective cohort study 1004 (the SIREN study)

Protective or Sterilising Immunity? Implications for Vaccine Strategies 1008 (2020)

SARS-CoV-2 T cell immunity: Specificity, function

Lack of peripheral memory B cell responses in recovered patients with 1013 severe acute respiratory syndrome: a six-year follow-up study

Opportunistic Infections Project Team of the Collaboration of Observational

CD4 cell count and the risk of AIDS or death in HIV-Infected adults on 1017 combination antiretroviral therapy with a suppressed viral load: a longitudinal cohort study 1018 from COHERE

Immune activation and inflammation in HIV-1 infection: causes 1020 and consequences

The CD4:CD8 ratio is associated with markers of age-associated 1022 disease in virally suppressed HIV-infected patients with immunological recovery

HIV-infected individuals with low CD4/CD8 ratio despite effective 1025 antiretroviral therapy exhibit altered T cell subsets, heightened CD8+ T cell activation, and 1026 increased risk of non-AIDS morbidity and mortality

Course and Clinical Significance of CD8+ T-Cell Counts in a Large 1029

Cohort of HIV-Infected Individuals

Increased risk of serious non-AIDS-related events in HIV-infected 1032 subjects on antiretroviral therapy associated with a low CD4/CD8 ratio

The 1036 CITIID-NIHR BioResource COVID-19 Collaboration

Neutralising antibodies drive 1040 Spike mediated SARS-CoV-2 evasion