Immunological evaluation of young unvaccinated patients with Turner syndrome after COVID-19

Abstract

Objectives

The X-chromosome contains the largest number of immune-related genes, which play a major role in COVID-19 symptomatology and susceptibility. Here, we had a unique opportunity to investigate, for the first time, COVID-19 outcomes in six unvaccinated young Brazilian patients with Turner syndrome (TS; 45, X0), including one case of critical illness in a child aged 10 years, to evaluate their immune response according to their genetic profile.

Methods

A serological analysis of humoral immune response against SARS-CoV-2, phenotypic characterization of antiviral responses in peripheral blood mononuclear cells after stimuli, and the production of cytotoxic cytokines of T lymphocytes and natural killer cells were performed in blood samples collected from the patients with TS during the convalescence period. Whole exome sequencing was also performed.

Results

Our volunteers with TS showed a delayed or insufficient humoral immune response to SARS-CoV-2 (particularly immunoglobulin G) and a decrease in interferon-γ production by cluster of differentiation (CD)4+ and CD8+ T lymphocytes after stimulation with toll-like receptors 7/8 agonists. In contrast, we observed a higher cytotoxic activity in the volunteers with TS than the volunteers without TS after phorbol myristate acetate/ionomycin stimulation, particularly granzyme B and perforin by CD8+ and natural killer cells. Interestingly, two volunteers with TS carry rare genetic variants in genes that regulate type I and III interferon immunity.

Conclusion

Following previous reports in the literature for other conditions, our data showed that patients with TS may have an impaired immune response against SARS-CoV-2. Furthermore, other medical conditions associated with TS could make them more vulnerable to COVID-19.

Introduction

The clinical disparities between sexes in SARS-CoV-2 infection have been intensively investigated around the world since the early stage of the COVID-19 pandemic [1]. Many studies have demonstrated that men have a higher risk of progressing to severe complications or lethal outcomes by COVID-19 than women, independent of age [2,3]. These sex-based differences could be associated with behavioral differences in protective measures adaption [4], as well as biological pathways encoded by X-chromosome genes related to immune responses against SARS-CoV-2 and other viruses [5], [6], [7], [8] and important aspects in COVID-19 physiopathology [9,10].

The number and activity of innate immune cells differ between the sexes [6]. Overall, the antigen-presenting cells from females are more efficient than antigen-presenting cells from males at the presenting peptides. Also, the neutrophils’ and macrophages’ phagocytic activities are higher in women than in men. In contrast, men have a higher proportion of natural killer (NK) cells than women. These cells are crucial in acute viral infection by recognizing the virus, promoting clearance, and preventing viral replication [11]. In turn, the differences in the adaptive lymphocyte subsets may be also related to greater antibody responses, higher basal immunoglobulin (Ig) levels, higher cluster of differentiation (CD)4+ T cells number, and upregulation of antiviral and proinflammatory genes in T cells, leading to better humoral and cellular responses of women against SARS-CoV-2 [12].

In addition, the sex hormone bias plays an important role in the differential immunity and physiopathology mechanisms of COVID-19. Many genes associated with immunity are activated by estradiol, which regulates cytokine levels, whereas androgens are immunosuppressive [13]. Estrogen also modulates the T regulatory cell quantity and protective effects in the respiratory and cardiovascular systems, both of which are significantly affected by SARS-CoV-2, justifying why women are less susceptible to severe COVID-19 [6,8]. The angiotensin-converting enzyme 2, directly involved in SARS-CoV-2 epithelial cells, is also encoded by X-chromosome and their expression is regulated by estradiol, which is also particularly important for better outcomes of COVID-19 [1].

Many genes that regulate innate and adaptive immune responses are in the X-chromosome. In this sense, the escape of X-linked genes in females confers a higher functional diversity in immune responses after random X-chromosome inactivation [7]. This phenomenon may be considered a protective factor for women against some infectious diseases because these individuals have stronger innate and adaptive responses against pathogens [12].

The toll-like receptors (TLRs) are pattern-recognition receptors involved in innate immune responses activating the cytokine-mediated pathways, including the interferon (IFN) family (IFN-α, IFN-β, and IFN-γ for viral infection), interleukin (IL)-1, and nuclear factor-κB (NFκB) [12], and play an important role in the sensing of SARS-CoV-2 infection [1]. The higher copy number of the TLR7 gene in women, because this gene is encoded on the X-chromosome, may enhance the innate immune responses in these individuals compared with healthy males, for example. Some genes for cytokine receptors and FOXP3 transcriptional factor, related to the innate response, are also encoded by the X-chromosome genes [7].

In women, Turner syndrome (TS) occurs due to the total or partial loss of the second X-chromosome [14]. The phenotype is mainly related to the loss of the small part of the X-chromosome portion that escapes the inactivation. The defective copy number of some X-chromosome-related genes leads to a series of clinically relevant characteristics, including dysregulation of Ig production [7] and T and B cell function [15].

Therefore, TS studies can be a valuable source to understand the X-chromosome role in immunity against infectious diseases, including the SARS-CoV-2 infection. To the best of our knowledge, there are no reported studies on immune characterization in this specific cohort and COVID-19. Here, we present a retrospective study with six young Brazilian patients with TS who had symptomatic COVID-19, with the aim to evaluate their humoral and cellular immune responses according to their genetic profile.

Methods

Volunteers’ recruitment, blood collection, and samples processing

Based on a report in national media, in June 2020, we had contact with the first case of a girl aged 10 years with TS who recovered from severe COVID-19. Then, we ascertained five more cases of convalescent women with TS that lived in São Paulo, the most populated city in Brazil. The participants were selected based on their clinical history of SARS-CoV-2 infection and confirmatory tests presentation in the interviews at least 3 months after the COVID-19 episode. Peripheral blood samples were collected in the convalescent period between July 2020 and August 2021. None of them had been vaccinated. The clinical information of volunteers with TS is presented in Table 1 and the genotype and the copy number variation analysis are shown in Table 2 .

Table 1.

Demographical data of the turner syndrome volunteers and COVID-19 information.

ID	Age (years)	COVID-19 episode
		Recurrence	Severity	Symptoms
01	10	No	Severe	Initial symptoms: sore throat with low fever. Progressed into severe symptoms: shortness of breath and dizziness. Very brief intensive care unit stay.
02	39	Yes	Mild	Only sore throat and cough.
			Asymptomatic	-
03	26	No	Mild	Flu-like symptoms and diarrhea.
04	37	No	Mild	Flu-like symptoms and ageusia.
05	12	No	Mild	Flu-like symptoms.
06	6	No	Mild	Flu-like symptoms.

*Flu-like symptoms: fever, chills, headache, body aches, cough, sore throat, runny nose, fatigue.

Table 2.

Genotype and CNV analysis of the turner syndrome volunteers.

			Deletion limits (GRCh38/hg38)
Patient	Genotype	CNV analysis using NGS data	Start (bp)	End (bp)	Clinically relevant genes number
ID01	45,X (70%)	Complete deletion	1	156010416	242
	46,XXdel(Xp) (30%)(a)	Partial deletion	1	51745520	86
ID02	45,X(b)	Complete deletion	1	156010416	242
ID03	45,X(b)	Complete deletion	1	156010416	242
ID04	45,X (6%) / 46,XX (89%) / 47,XXX (4%) / 48,XXXX (1%)(b)	Deletion not detected	1	156010416	242
ID05	45,X(b)	Complete deletion	1	156010416	242
ID06	45,X(b)	Complete deletion	1	156010416	242
Gene content (clinically relevant genes)
Partial deletion	SHOX, CSF2RA, ARSL, NLGN4X, STS, ANOS1, TBL1X, GPR143, CLCN4, MID1, HCCS, AMELX, MSL3, FRMPD4,TLR7,TLR8, TRAPPC2, OFD1, GLRA2, FANCB, PIGA, AP1S2, NHS, CDKL5, RS1, PHKA2, ADGRG2, PDHA1, SH3KBP1, RPS6KA3, CNKSR2, SMPX, MBTPS2, SMS, PHEX, PTCHD1, KLHL15, EIF2S3, PDK3, POLA1, ARX, IL1RAPL1, NR0B1, GK, DMD, CFAP47, XK, CYBB, RPGR, OTC, TSPAN7, BCOR, ATP6AP2, USP9X, DDX3X, NYX, CASK, MAOA, NDP, KDM6A, SLC9A7, RP2, NDUFB11, RBM10, UBA1, SYN1, CFP, FTSJ1, PORCN, EBP, WAS, GATA1, HDAC6, PQBP1, SLC35A2, OTUD5, TFE3, WDR45, PRICKLE3, SYP, CACNA1F, CCDC22,FOXP3, USP27X, CLCN5, BMP15
Complete deletion	SHOX, CSF2RA, ARSL, NLGN4X, STS, ANOS1, TBL1X, GPR143, CLCN4, MID1, HCCS, AMELX, MSL3, FRMPD4,TLR7,TLR8, TRAPPC2, OFD1, GLRA2, FANCB, PIGA, AP1S2, NHS, CDKL5, RS1, PHKA2, ADGRG2, PDHA1, SH3KBP1, RPS6KA3, CNKSR2, SMPX, MBTPS2, SMS, PHEX, PTCHD1, KLHL15, EIF2S3, PDK3, POLA1, ARX, IL1RAPL1, NR0B1, GK, DMD, CFAP47, XK, CYBB, RPGR, OTC, TSPAN7, BCOR, ATP6AP2, USP9X, DDX3X, NYX, CASK, MAOA, NDP, KDM6A, SLC9A7, RP2, NDUFB11, RBM10, UBA1, SYN1, CFP, SSX1, FTSJ1, PORCN, EBP, WAS, GATA1, HDAC6, PQBP1, SLC35A2, OTUD5, TFE3, WDR45, PRICKLE3, SYP, CACNA1F, CCDC22,FOXP3, USP27X, CLCN5, BMP15, SSX2, KDM5C, IQSEC2, SMC1A, HSD17B10, HUWE1, PHF8, TSR2, FGD1, MAGED2, ALAS2, UBQLN2, ARHGEF9, AMER1, ZC4H2, LAS1L, MSN, AR, OPHN1, EFNB1, EDA, IGBP1, ARR3, KIF4A, DLG3, TEX11, IL2RG, MED12, NLGN3, GJB1, NONO, TAF1, OGT, GCNA, HDAC8, PHKA1, XIST, SLC16A2, RLIM, NEXMIF, ABCB7, FGF16, ATRX, MAGT1, COX7B, ATP7A, PGK1, TBX22, BRWD3, POU3F4, ZNF711, POF1B, CHM, DIAPH2, PCDH19, SRPX2, TIMM8A, BTK, GLA, HNRNPH2, GPRASP2, PLP1, SERPINA7, TBC1D8B, CLDN2, DNAAF6, PRPS1, MID2, COL4A6, COL4A5, IRS4, ACSL4, AMMECR1, CHRDL1, PAK3, DCX, ALG13, PLS3, STEEP1, UBE2A, UPF3B, RNF113A, NDUFA1, NKAP, LAMP2, CUL4B, C1GALT1C1, GLUD2, GRIA3, THOC2, XIAP, STAG2, SH2D1A, OCRL, XPNPEP2, SASH3, ZDHHC9, BCORL1, ELF4, AIFM1, IGSF1, FRMD7, HS6ST2, GPC4, GPC3, PHF6, HPRT1, SLC9A6, FHL1, CD40LG, RBMX, GPR101, ZIC3, FGF13, F9, ATP11C, SOX3, FMR1, AFF2, IDS, MAMLD1, MTM1, HMGB3, VMA21, GABRA3, NSDHL, BGN, ATP2B3, CCNQ, SLC6A8, BCAP31, ABCD1, SSR4, L1CAM, AVPR2, NAA10, HCFC1, MECP2, OPN1LW, OPN1MW, FLNA, EMD, RPL10, TAFAZZIN, ATP6AP1, GDI1, FAM50A, LAGE3, G6PD, IKBKG, DKC1, F8, RAB39B, CLIC2, TMLHE

CNV, copy number variation; NGS, next-generation sequencing; TLR, toll-like receptor.

^aInferred from NGS data

^bKaryotype.

Peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll-Paque density gradient centrifugation and stored in liquid nitrogen until use. Plasma was processed by centrifugation and kept frozen at -80°C until use. Six women, age-matched without TS who recovered from COVID-19 in the same conditions, were recruited as controls for the immunological assays, following the same procedures described above.

Genomic assays

Whole exome sequencing was performed in peripheral blood DNA with the Illumina NovaSeq platform at HUG-CELL facilities. The sequencing data were analyzed following bwa-mem, and GATK Best Practices workflow, quality control, and annotation were performed as previously described [16]. The genotypes of the six volunteers with TS were confirmed by whole exome sequencing analysis. The inference of genetic ancestry was performed by ADMIXTURE v1.36 [17] in the supervised analysis (k = 4), after filtering the markers for linkage disequilibrium (r2 = 0.1) using a 50Kb sliding window with 10kb steps, totaling 53,987 single-nucleotide polymorphisms. The samples from the 1000 Genomes Project [18] and the Human Genome Diversity Cell Line Panel [19] with over 95% inferred ancestry in a given group were used as the parent populations.

The genetic analysis was focused on the potential presence of rare variants related to inborn errors of type I IFN immunity genes as previously described [17,18]. In addition, we analyzed the genotypes and haplotypes of the human leukocyte antigen (HLA) cluster by using a customized HLA mapper (version 4) [20] to optimize read alignment along the major histocompatibility complex region. We obtained the complete exonic sequences of HLA-A, HLA-B, HLA-C, HLA-E, HLA-G, MICA, MICB, and HLA-DOB for each individual by converting the phased variant call format obtained in the previous step into complete sequences using vcfx transcript (www.castelli-lab.net/apps/vcfx). As previously described, we translated these sequences into protein sequences (the allotypes) [21].

Immunological assays

Humoral response analysis by serology assessment

The detection of IgA, IgM, and IgG against SARS-CoV-2 nucleocapsid protein (NP) and receptor-binding domain (RBD) were performed through enzyme-linked immunosorbent assay developed in-house by our research group, as described previously [22]. Serology was performed at least 60 days after the initial COVID-19 episode.

Phenotypic characterization of PBMCs

For phenotypic characterization by flow cytometry, PBMCs were incubated with a viability marker LIVE/DEAD Fixable Red Dead Cell Stain Kit (Invitrogen, Carlsbad, CA, USA) for 20 minutes and then incubated with surface antibodies in different panels for phenotypic characterization of T lymphocytes, B lymphocytes, and NK cells (Supplementary Table S1) for 30 minutes at room temperature. After this period, the samples were fixed with phosphate buffer solution (PBS) and 1% formalin for 15 minutes at room temperature. Then, the cells were washed and resuspended in PBS. The events were acquired on a Fortessa LSR flow cytometer (BD Biosciences). Data analysis was performed using the FlowJo™ software. The gating strategy used is illustrated in Supplementary Figure S1.

Phenotypic evaluation of PBMCs stimulated with TLRs 7/8 agonist and IFN-α

The PBMC samples were plated in 48-well plates (Costar, Cambridge, MA, USA) in RPMI culture medium (Gibco, Carlsbad, CA) containing 5% human AB serum (Sigma-Aldrich, St. Louis, MO, USA). The cells were stimulated with 2.5 ug/ml CL097 (Invivogen, Califórnia, EUA) or 10³ U/ml IFN-α (Bergisch Gladbach, Germany) and incubated in a 5% CO₂ oven at 37°C for 24 hours. Subsequently, the cells were collected, washed, and incubated with the viability marker LIVE/DEAD Fixable Red Dead Cell Stain Kit (Invitrogen, Carlsbad, CA, USA) for 20 minutes. Then, the cells were incubated with the surface antibodies for 30 minutes and fixed for 15 minutes with 1% formalin. The cells were washed and incubated for 30 minutes with antibodies for intracellular labeling, along with saponin 0.05% in PBS 1X. After washing, the cells were acquired on a Fortessa LSR flow cytometer (BD Biosciences). Data analysis was performed using the FlowJo™ software. The gating strategy used is illustrated in Supplementary Figure S2 and the antibodies used are shown in Supplementary Table S1.

Evaluation of the production of cytotoxic factors by T lymphocytes and NK cells

To evaluate the cytotoxic profile of CD4+ T lymphocytes, CD8+ T cells, and NK cells, the PBMCs were distributed in 48-well microplates (Costar, Cambridge, MA, USA) in RPMI culture medium (Gibco, Carlsbad, CA) containing 5% of human AB serum (Sigma-Aldrich, St. Louis, MO, USA). The cells were stimulated with 30 ng/ml phorbol myristate acetate (PMA) and 500 ng/ml ionomycin (Sigma-Aldrich) in the presence of antibody CD107a PE-Cy 5 (clone H4A3, BDBiosciences) and brefeldin A (10 μg/ml, Sigma-Aldrich) and incubated in 5% CO₂ at 37°C for 6 hours. Subsequently, the cells were incubated with viability marker LIVE/DEAD Fixable Red Dead Cell Stain Kit (Invitrogen, Carlsbad, CA, USA) for 20 minutes. Then, the cells were incubated with the surface antibodies for 30 minutes and fixed for 15 minutes with 1% formalin. The cells were washed and incubated for 30 minutes with antibodies for intracellular labeling, along with saponin 0.05% in PBS 1X. After this period, the cells were washed and resuspended in PBS.

The cells were acquired in a Fortessa LSR flow cytometer (BD Biosciences). Data analysis was performed using the FlowJo™ software. The gating strategy used is illustrated in Supplementary Figure S3 and the antibodies used are shown in Supplementary Table S1.

Measurement of cytokines by cytometric bead array

The levels of IL-2, IL-4, IL-5, IL-6, IL-10, IL-17A, IFN-γ, and tumor necrosis factor (TNF)-α plasma cytokines were quantified by the cytometric bead array method using a commercial kit (Flex Cytometric Bead Array Enhanced Sensitivity; BD Pharmingen, San Diego, CA, USA) according to the manufacturer's instructions. Data were acquired on the BD LSRFortessa™ Cell Analyzer (BD Biosciences, CA, USA) and analyzed on the FCAP Array Software v3.0 (BD Biosciences, CA, USA).

Statistical analysis

The data were analyzed using GraphPad Prism software (Version 8.0). Comparisons between groups were performed using the nonparametric Mann-Whitney test, and the same group was performed using the nonparametric Wilcoxon test. The level of significance was considered when P-value ≤0.05.

Results

Genetic analysis by whole exome sequencing

Genetic ancestry

The ancestry estimates showed that European composition is the predominant ancestry for five volunteers (more than 65%), except for ID 06 who has 64% of African ancestry, as presented in Supplementary Table S2, compared with the Brazilian average genetic ancestry [16].

Inborn errors of type I INF immunity genes

Two of the six volunteers with TS carry rare variants (with a mean combined annotation dependent depletion score > 20) in genes involved in the regulation of type I and III IFN immunity. In ID 01, we detected one very rare missense variant in the UNC93B1 gene (:NM_030930:exon6:c.G703A:p.A235T, one heterozygote among 152,194 alleles in gnomAD genomes v3.1.2) carried in the heterozygous state, which is predicted to be deleterious (SIFT and Polyphen) and may be linked to her increased risk of developing severe COVID-19. ID 04 carries another rare missense variant in the same gene (UNC93B1:NM_030930:exon11:c.G1777A:p.G593R), previously reported as a variant of uncertain significance on ClinVar. There are 31 heterozygotes of 150,320 alleles in the gnomAD genomes dataset (v3.1.2), most (N = 13) of which were from Latino ethnicity, suggesting that the variant of uncertain significance missense variant in ID 04 is common among Native American ancestry and unlikely to have a large monogenic effect.

HLA genes

Table 3 presents genotypes (two-field resolution) of some genes from the HLA region in the major histocompatibility complex complex. We highlighted the presence of alleles with clinical significance in the context of COVID-19 disease [19]. Interestingly, ID 01, who developed severe illness, carries HLA-E*01:01 alleles in homozygosity, which are associated with COVID-19 clinical severity [19].

Table 3.

HLA class I and II alleles of the turner syndrome volunteers.

ID	01	02	03	04	05	06
Gene	Allele 1, allele 2	Allele 1, allele 2	Allele 1, allele 2	Allele 1, allele 2	Allele 1, allele 2	Allele 1, allele 2
HLA-A	*01:01:01,*23:01:01	*03:01:01,*24:02:01	*01:01:01,*03:01:01	*24:02:01,*24:02:01	*01:01:01,*03:01:01	*02:01:01, *33:01:01
HLA-B	*14:01:01, *49:01:01	*15:01:01,*41:01:01	*08:01:01,*38:01:01	*15:08:01,*50:01:01	*07:02:01,*18:01:01	*14:02:01,*18:01:01
HLA-C	*08:02:01, *07:01:01	*17:01:07,*03:03:01	*12:03:01, *07:01	*03:03:01,*06:02:01	*05:01:01,*07:02:01	*08:04, *12:03:01
HLA-E	*01:01:01, *01:01:01	*01:03:02,*01:03:02	*01:01:01,*01:03:02	*01:03:02,*01:03:02	*01:03:02,*01:03:02	*01:01:01,*01:01:01
HLA-G	*01:04:04, *01:06:01	*01:01:01,*01:04:01	*01:01:01,*01:01:22	*01:04:01,*01:04:01	*01:01:01,*01:06:01	*01:01:01,*01:03:01
HLA-DRB1	*07:01:01, *07:01:01	*11:03:01,*13:02:01	*03:01:01,*03:01:01	*04:03, *14	*15:01:01,*15:01:01	*01:02, *01:02

HLA, human leukocyte antigen.

HLA haplotypes which were linked to high severity Covid-19 are in bold.

Immune responses analysis

Humoral response against SARS-CoV-2

Serological assays for SARS-CoV-2 IgA, IgM, and IgG were performed through enzyme-linked immunosorbent assay for the RBD of the spike protein and the NP at least 3 months after the initial COVID-19 diagnosis (Table 4 ). It is important to highlight that these immune assays were performed before the volunteers’ vaccination against COVID-19. Interestingly, three of the six girls with TS (ID 01, 02, and 05) did not present IgG antibodies against SARS-CoV-2.

Table 4.

Serological results of enzyme-linked immunosorbent assay for SARS-CoV-2 IgA, IgM, and IgG antibodie.

	ID	Relevant Information	Immunoglobulins
			IgA		IgM		IgG		Daysa
			NP	RBD	NP	RBD	NP	RBD
Turner syndrome patients	01	Critical Covid-19	-	+	+	+	-	-	145
	02	Reinfected	-	-	+	+	-	-	330 days after infection 60 days after reinfection
	03	-	-	-	+	+	-	+	195
	04	Mosaicism	-	+	-	+	+	+	360
	05	-	-	-	-	-	-	-	160
	06	-	-	-	-	-	+	+	135
Controls+	01	30 years old Mild COVID-19	-	+	+	+	+	-	90
	02	28 years old Mild COVID-19	-	-	+	-	+	+	90
	03	28 years old Mild COVID-19	-	-	+	+	-	+	90
	04	25 years old Mild COVID-19	-	+	+	-	+	+	125
	05	15 years old Mild COVID-19	-	-	-	-	+	+	105
	06	8 years old Mild COVID-19	-	-	-	+	+	+	95

Ig, immunoglobulin; NP, nucleocapsid protein; RBD, receptor-binding domain; +, women healthy donors.

^aTime (in days) between the Covid-19 episode (positive test) and the serology.

Characterization of lymphocytes and phenotypic evaluation of peripheral blood mononuclear cell stimulated with TLRs 7/8 agonist (CL097) and INF-α

To understand whether the absence of an X-chromosome could impair the antiviral responses induced by the TLR7 pathway, we stimulated the PBMC of volunteers with TS and control volunteers with CL097 (TLR7/8 agonist) or IFN-α, the main cytokine induced by this pathway during the antiviral response. Figure 1 shows the lymphocyte composition in the total blood before stimulation of the TLR7 pathway or IFN-α and the quantitative assessment of the proinflammatory cytokine production.

Figure 1.

Phenotypic characterization and evaluation of basal antiviral cellular responses after stimulation with CL097 (TLR7/TLR8 agonist) and IFN-α. (a) Phenotypic characterization of lymphocytes. (b) Lymphocytes’ basal cytokine production of IFN-γ, interleukin-10, and tumor necrosis factor-α after stimulation with CL097 and IFN-α.

(*) for P-values <0.05 and (**) for P-values <0.01.

i = CD4 T lymphocytes; ii = CD8 T lymphocytes; iii = NK cells and iv = B lymphocytes.

CD, cluster of differentiation; IFN, interferon; NK, natural killer; TS, Turner syndrome.

The frequencies of CD4+, T CD8+, B lymphocytes, and NK cells in total blood are shown in Figure 1a. Interestingly, a significantly increased percentage of NK cells in the blood of volunteers with TS in comparison with control individuals was observed (P <0.01). The frequencies of CD4+, CD8+ T cells, and the observed B lymphocyte proportions were similar between the volunteers with TS and controls.

The CD4+ T, CD8+ T, and B lymphocyte's production of IFN-γ, IL-10, and TNF cytokines after CL097 and IFN-α stimulus are shown in Figure 1b.

On the other hand, the production of the analyzed cytokines was higher in the TS group than in the control for NK cells and B lymphocytes.

Cytotoxic factors’ production by T lymphocytes and NK cells

Because the antiviral response induced by CL097 in volunteers with TS showed a tendency to be lower than the control group, we wanted to verify if the same occurred with the cytotoxic response in lymphocytes, given the importance of this response during viral infections.

The lymphocytes’ production of IFN-γ, CD107a, granzyme B, perforin, and TNF cytotoxic factors after PMA/ionomycin stimulus are shown in Figure 2 . Overall, volunteers with TS showed greater cytotoxic activity of peripheral blood lymphocytes than the controls, in both basal conditions and after being stimulated with PMA/ionomycin. Major significant increases were concerned with the production of granzyme B and perforin cytolytic granules by NK cells and CD8+ T lymphocytes (p-values P <0.01 for both cases) after PMA/ionomycin stimulus. Also, regarding the NK cell's cytotoxic activity, both the IFN-γ and TNF production at basal levels were higher in the volunteers with TS than in controls.

Figure 2.

Analysis of Basal and PMA/Iono-stimulated IFN-γ, CD107a, Granzyme B, perforin, and TNF-α cytotoxic factors’ profiles of T lymphocytes and NK cells of production of IFN-γ, CD107a in controls and volunteers with TS (a) CD4+ T lymphocytes; (b) CD8+ T lymphocytes and (c) NK cells.

(*) for P-values <0.05 and (**) for P-values <0.01.

CD, cluster of differentiation; IFN, interferon; NK, natural killer; TNF, tumor necrosis factor; TS, Turner syndrome.

Measurement of plasma cytokines by cytometric bead array

Quantifications of IL-2, IL-4, IL-5, IL-6, IL-10, IL-17A, IFN-γ, and TNF plasma levels measured by cytometric bead array are shown in Table 5 . No significant differences in these cytokines’ plasma levels were observed between the volunteers with TS and controls.

Table 5.

Plasma cytokines levels of the volunteers measured by cytometric bead array.

	Plasma cytokines
Group	IL-2	IL-4	IL-5	IL-6	IL-10	IL-17A	Interferon-γ	Tumor necrosis factor-α
Turner syndrome	13,96 土 3,15	5,25 土 1,49	5,98 土 2,43	9,90 土 3,24	9,44 土 4,78	3,41 土 1,84	4,28 土 1,65	8,63 土 4,47
Control	10,64 土 0,94	4,23 土 0,21	1,91 土 0,84	6,44 土 1,48	3,28 土 1,35	2,08 土 1,35	1,46 土 0,91	2,87 土 1,40

IL, interleukin.

Values were expressed by mean ± standard error.

Discussion

Based on our observations, the young volunteers with TS seem to have a delayed or insufficient humoral immune response to SARS-CoV-2, particularly IgG, (Table 4) and a decrease in IFN-γ production (known for having a potent antiviral activity) by CD4+ and CD8+ after stimulation with TLR7/TLR8 agonists (Figure 1). In contrast, we observed a higher cytotoxic activity in our volunteers with TS than the volunteers without TS at basal conditions (without stimulus; Figure 2). Also, two volunteers with TS carry rare genetic variants in genes that regulate type I and III IFN immunity, which might influence the symptomatic presentation of COVID-19 in these individuals.

Worldwide studies with different cohorts of unvaccinated individuals have demonstrated that SARS-CoV-2 IgG antibodies seroconversion is achieved within 5-7 days after COVID-19 symptom onset. The maximum seroconversion occurs in 3-6 weeks, and it is expected that healthy individuals, regardless of sex or age, produce it in response to the pathogen [23]. The concentrations of SARS-CoV-2-specific IgG remained high and stable for several months before declining [24]. One study, which enrolled unvaccinated infected patients, showed that both IgG antibodies against the SARS-CoV-2 NP and the RBD portion of the spike protein were detected in more than 85% of patients at 1, 3, and 6 months after diagnosis [25]. Our previous studies with unvaccinated individuals who had COVID-19 presented similar rates (>85%) [26,27]. In contrast, IgM antibodies peaked in the third week (>85%) and then began to decline faster after 60 days [28].

A poor humoral response against SARS-CoV-2 was observed in three of our volunteers with TS (ID 01, 02, and 05) who did not present IgG antibodies against it months after the infection and still presented specific IgM. Patient ID 02 did not present specific IgG even after reinfection but instead, IgM was detected after 60 days. These data seem to be supported by previous studies reporting that the humoral responses against pathogens are reduced in TS, with lower IgG/IgM antibodies production and variable levels of IgA in TS than with same-aged women with both X-chromosomes [20,29,30]. In addition, the negative IgG2m allotype in TS has been associated with impaired immune responses to vaccination against Haemophilus influenzae [31]. Another possible hypothesis for the deficient humoral response in TS can be associated with a reduced expression of the X-linked-gene UTX, related to T follicular helper cell function, which has been recently proven to be differentially methylated in patients with TS [2,15]. Interestingly, ID 04 with mosaic TS (X0; XX) is the only volunteer from this group who displayed IgA, IgM, and IgG antibodies against SARS-CoV-2 even 1 year after infection.

Regarding the immune cell subsets, the CD4+/CD8+ T cells ratio is commonly lower in TS than in women without TS, indicating a higher quantity of CD8+ T cells [1]. However, the phenotypic characterization showed a slight predominance of CD4+ T lymphocytes’ frequency compared with CD8+. Besides the frequency, it has also been reported that TS is related to T CD4+ regulatory cell dysregulation, which may be explained by the loss of one allele of the FOXP3 gene located in the X-chromosome and a higher ratio of the Th17 subset. Consequently, T regulatory cells from TS could not suppress the autologous T cells that can be involved in the autoimmunity disease [17]. Interestingly, we also observed a significant amount of NK cells in volunteers with TS, which is not expected in physiological conditions [32]. NK cells have direct cytotoxic activity against virus-infected cells by releasing cytotoxic granules containing perforin and granzymes.

The phenotypic evaluation of PBMC stimulated with the agonist of TLR7/TLR8 and IFN-α revealed that upon type I IFN stimulation, our volunteers with TS are not able to produce IFN-γ and they have a tendency for lower production of TNF by CD4+ and CD8+ than the control group. IFN-γ possesses a potent antiviral activity, whereas TNF appears to induce multiple antiviral mechanisms and synergize with IFN-γ in promoting antiviral activities [33]. Thus, it seems that our volunteers with TS have a weaker antiviral response than the volunteers without TS when stimulated with TLR7/ TLR8 agonist, which mainly can be explained by TLR7 lower gene-dose expression in TS [32].

Interestingly, TLR7 has been described to exert an important role in the protective type I IFN immunity. TLR7 is one of the crucial pattern-recognition receptors for SARS-CoV-2 single strain RNA and females may be protected to some extent against severe COVID-19 due to the biallelic TLR7 expression [8]. X-linked recessive TLR7 deficiency is associated with critical COVID-19 pneumonia in about 1.8% of male patients below the age of 60 years [34]. Moreover, the reduced type I IFN in patients with severe COVID-19 preceded clinical worsening [35], and the induction of type I response by relevant immune agonists, such as TLR7/8 agonist or TRL3, is perturbed in hospitalized patients [36]. Therefore, TLR7’s lower gene-dose expression in TS may be related to the differences found between this group and the control group.

However, in contrast with the antiviral response, we observed a compensatory enhancement of the cytotoxic activity by peripheral blood lymphocyte cells from volunteers with TS. Significant increases were observed in the production of granzyme B and perforin cytolytic granules produced specially by NK cells and CD8+ T lymphocytes after PMA/ionomycin stimulus. This interesting finding may explain why most volunteers with TS, even with weak antiviral and delayed humoral immune responses, did not have worse outcomes and could control the viral infection.

The immune dysregulation in TS could be explained by haploinsufficiency of some genes in the X-chromosome and the presence of some HLA alleles previously associated with critical COVID-19, along with hormonal changes throughout the age spectrum [20]. Previous studies have already compared the expression of X-linked genes related to immunity in females with TS and without TS showing important differences in innate and adaptive responses [37]. Along with the evidence of immune response dysregulation in TS, the medical conditions often seen in TS may predispose them to complications from COVID-19.

Besides well-known clinical features, such as short stature and gonadal failure, patients with TS tend to have various medical conditions, including heart, liver, and renal abnormalities; obesity; and hypertension [38,39]. They are also more susceptible to autoimmune diseases, such as Hashimoto thyroiditis, type 1 diabetes, and juvenile rheumatoid arthritis [40], having twice more risk as women without TS and four times more risk than men to develop autoimmune diseases [19].

The genetic analysis showed that two individuals with TS (ID 01 and 04) carry rare variants in the UNC93B1 gene, associated with impaired cellular interferon antiviral responses [37], including against the influenza virus [41], which may contribute to life-threatening COVID-19 pneumonia [17,18]. ID 01 displayed a severe form of COVID-19, whereas ID 04 had a mild form. In addition, ID 01 carries HLA-E*01:01 alleles in homozygosity, which were previously associated with COVID-19 clinical severity [19]. The genomic findings might influence the symptomatic presentation of COVID-19 in these individuals and the severity form described in the child aged 10 years (ID 01).

To the best of our knowledge, this is the first study evaluating the immunological and genetic profile of unvaccinated young women with TS who were infected by SARS-CoV-2 and developed symptomatic COVID-19. The data presented herein might be interesting to clinicians who take care of patients with TS and enhance our comprehension of the immunological response to SARS-CoV-2 infection by these individuals. The study limitations include the great difficulty in ascertaining such a rare cohort of unvaccinated women with TS who recovered from COVID-19. Besides, because this is a retrospective study, we could not quantify the SARS-CoV-2 viral load from the respiratory tract of participants, sequence the virus genome, and perform experiments at the acute phase of SARS-CoV-2 infection.

From future perspectives, it would be interesting to evaluate the COVID-19 vaccine response in patients with TS. Furthermore, because the endocrine and neuroendocrine systems greatly influence the immune system, there is a need to further study these systems in patients with TS, especially the role of growth hormone not only in COVID-19 but also in other infectious diseases.

Conclusion

Our analysis showed that the antiviral activity mediated by INF production is weaker in TS, which might be due to the presence of only one copy of the X-chromosome impacting the expression of immunity-related genes. This immune response may also overcome a possible deficiency of humoral immune responses in TS. However, they did not have negative disease outcomes due to the compensatory cytotoxic activity mainly mediated by NK cells.

Disclosures of competing interest

The authors have no competing interests to declare.

Funding

This work was supported by the São Paulo Research Foundation (grant numbers 2013/08028-1, 2014/50931-3, and 2020/09702-1), the National Council for Scientific and Technological Development (grant numbers 465355/2014-5 and 404134/2020-3), the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior-Brasil - Finance Code 001 and JBS S.A (grant number 69004). The funders were not involved in the study design, collection, analysis, interpretation of data, the writing of this case report, or the decision to submit it for publication. All the cited funders supported the conduction of the experiments equally.

Ethical approval

The study was approved by the committee for ethics in Research of the Institute of Biosciences at the University of São Paulo (CAAE 34786620.2.0000.5464 and 56349222.9.0000.5464) following the Declaration of Helsinki principles, ICH06 Good Clinical Practices, and Brazilian Health Regulatory Agency (ANVISA) resolution number 466 from 2012 that regulates research with humans.

Acknowledgments

The authors are extremely grateful for the participation and collaboration of the volunteers, the nurses for sample collection, the clinic that performed the blood tests, and the HUG-CELL technical team. The authors would also like to thank Brazilian Senator Mara Gabrilli for financial support and to Ruth R. Franco.

Author contributions

MVC and MVRS: data curation, investigation, formal analysis, and writing - original draft. SCGS and LMO: methodology, investigation, visualization, formal analysis, and writing - original draft. MSN: conceptualization, formal analysis, investigation, methodology, software, writing - review & editing. MOS and ECC: formal analysis, investigation, methodology, software, writing - review & editing. JYM: methodology and investigation. KSS: investigation, visualization, and writing - review & editing. ECN and MNS: writing - review & editing. MZ: conceptualization, funding acquisition, project administration, writing - original draft, writing - review & editing. All authors contributed to the article and approved the submitted version.

Data availability statement

The original contributions presented in the study are included in the article and supplementary material; further inquiries can be directed to the corresponding author.