Introduction
Common variable immunodeficiency disorders (CVID) are one of the most common clinically important primary immune deficiencies. CVID encompasses a group of heterogeneous primary antibody failure syndromes characterized by hypogammaglobulinemia associated with reduced or absent specific antibody production [
1,
2]. CVID patients commonly suffer from recurrent infections of the gastrointestinal and upper respiratory tracts [
3,
4], and inflammatory conditions and autoimmunity are also frequent in CVID patients [
4‐
6]. Immune dysregulation in CVID might be secondary to defects of B-cell differentiation and function, chronic and reduced T cell function, and altered NK cells and dendritic cells function, with chronic microbial translocation possibly contributing to the systemic immune activation and altered homeostasis of lymphocytic and myeloid lineages [
7‐
9]. Currently, immunoglobulin replacement therapy (IgRT), either intravenous (IVIg) or subcutaneous (SCIg), is a first-line therapy to prevent infections and aminorate all these immune alteration in CVID patients [
10,
11].
IVIg is a preparation of highly purified polyclonal poly-specific IgG isolated from plasma of thousands of healthy donors and widely used for the treatment of primary and secondary immunodeficiencies, as well as autoimmune and inflammatory disorders (reviewed in [
12]). Besides their antibody replacement effect, IVIg exerts immunoregulatory and anti-inflammatory actions on innate and adaptive immune cells [
13‐
17], and various non-mutually exclusive mechanisms have been proposed [
17‐
21] to explain its clinical effectiveness [
18‐
24]. The high number of donors used to prepare IVIg is particularly relevant to its anti-inflammatory effects, given the extensive pool of unique IgG antibody repertoire and anti-carbohydrate repertoire [
25]. The ample anti-inflammatory activity of IVIg has recently attracted attention to its potential therapeutic use for COVID-19 [
26‐
29], COVID-19-related Kawasaki disease [
30‐
32], and for the thrombotic thrombocytopenia observed after ChAdOx1 nCov-19 vaccination [
33‐
35]. We have previously demonstrated that IVIg skews macrophage polarization through FcγR-dependent mechanisms [
36] and that IVIg promotes tolerance towards inflammatory stimuli [
37]. However, extrapolation of the prophylactically administration in animal models of disease falls short in providing definitive answers about its mode of action in humans in vivo, where IVIg is commonly used in numerous other therapeutic strategies [
21].
To address the mechanisms of action of IVIg in vivo, we have now determined the phenotypic, transcriptomic, and functional profile of peripheral blood mononuclear cells (PBMC) from CVID patients prior and after IVIg infusion. Our results indicate that IVIg triggers the acquisition of an anti-inflammatory profile in PBMCs and monocytes, reduces the number of inflammatory circulating monocytes, and enhances the proportion of CD14+ monocytes whose phenotype and suppressor activity is compatible with that of myeloid-derived suppressor cells (MDSC). Our results indicate that monocytes are primary targets for the anti-inflammatory and inmmunosupressive effects of IVIg in CVID patients in vivo.
Materials and Methods
Patients and Clinical Samples
We studied a cohort of 11 CVID patients (age range of 20–70 years; mean age: 48.3 ± 13.8 years) followed at the Department of Clinical Immunology at the Hospital Clínico San Carlos (Madrid, Spain). CVID patients were diagnosed according to the classification of European Society of Immune Deficiencies (ESID) and the Pan-American Group for Immune Deficiency (PAGID) [
38,
39]. Given the inherent heterogeneity in CVID manifestations, we took into account the classification of clinical phenotypes proposed by Chapel et al. [
40], and our cohort included patients with “no-disease-related” complications (“infections only” phenotype) (
n = 5) and patients with inflammatory/autoimmune/lymphoproliferative complications (“inflammatory” phenotype) (
n = 6) (Supplementary Table
1). The study protocol was approved by the Ethics Committee of Hospital Clínico San Carlos (Madrid, Spain), and all subjects provided signed informed consent (Project 19/284-E).
All CVID patients were in a stable state with no apparent acute infection, and received IgRT as part of their routine treatment. The mean cumulative monthly dosage of IgRT was 400 mg/kg, with an infusion time from 4 to 6 h established according to the individual patient´s tolerability. The mean IVIg dose administered at the time of blood sampling was 28.7 ± 3.9 g. None of the 11 patients were taking steroids or other immunosuppressive or immunomodulatory drugs at the time of the study or the previous 6 months. Blood was obtained from CVID patients both before and after (6 h) receiving IgRT infusion, and PBMC were isolated over a Lymphoprep (Nycomed Pharma, Oslo, Norway) gradient according to standard procedures. Monocytes, T lymphocytes, and B lymphocytes were purified from PBMC by magnetic cell sorting using CD14, CD3, and CD20 immunomagnetic beads (Miltenyi Biotech, Bergisch Gladbach, Germany), respectively.
Microarray Analysis
Global gene expression analysis was performed on RNA obtained from PBMC, monocytes, T lymphocytes, and B lymphocytes isolated immediately before or after IVIg therapy of four independent patients. RNA isolation, microarray analysis (whole human genome microarray, Agilent Technologies, Palo Alto, CA), and statistical treatment of microarray data were performed following previously described procedures [
41‐
43]. Microarray data were deposited in the Gene Expression Omnibus (
http://www.ncbi.nlm.nih.gov/geo/) under accession nos. GSE133835 (PBMCs), GSE133907 (CD14 cells), GSE158576 (CD20 cells), and GSE158573 (CD3 cells). For Gene Set Enrichment Analysis (GSEA) (
http://software.broadinstitute.org/gsea/index.jsp) [
44], the gene sets available at the website, as well as previously defined gene sets, were used.
Phenotypic Analysis of Monocyte
Whole blood samples were collected before and after IVIg administration. Immediately after collection, blood sample was incubated at room temperature for 20 min with the indicated fluorescently tagged monoclonal antibodies. Following RBC lysis (RT, 15 min) using FACS Lysing solution (Becton Dickinson), cells were washed twice and analyzed on a FACS Canto II (Becton Dickinson) flow cytometer. For simultaneous surface and intracellular staining, cell surface antigen staining was performed first, and cells were later resuspended in Buffer Perm/Wash 1 × solution, treated with fixation and permeabilization solution (4 °C, 30 min in the dark) and subjected to intracellular staining. Monocyte subpopulations were phenotypically identified by a 8-color flow cytometry single platform assay using anti-CD14-APC Cy7, CD16-FITC, CX3CR1-PerCP Cy5, HLA-DR-BV510, CD86-PE, CCR5-BV421, CCR2-APC, and TNF-PE mAbs (BD, Becton- Dickinson Biosciences, Franklin Lakes, NJ).
T Cell Suppression Assay and Cytokine Secretion
Human peripheral blood CD4+ lymphocytes were isolated from CVID patients using magnetic cell sorting with anti-CD4 microbeads (Miltenyi Biotec), resuspended in RMPI 5% human AB serum (Sigma-Aldrich), and added into flatbottom 96-well plates (105 cells/well) that had been coated overnight with anti-human CD3 (10 μg/ml, BD Biosciences) and anti-human CD28 (1 μg/ml, BD Biosciences). Then, CD14+ cells isolated from CVID patients (both before and after IVIg infusion) were resuspended in RMPI 5% human AB serum, and co-cultured with CD4+ lymphocytes at the indicated ratios. After 48 h, [3H]thymidine was added (1 uCi/well, Perkin Elmer) during the last 20 h of coculture and thymidine incorporation was determined using a MicroBeta2 2450 Microplate Counter. Cell culture supernatants from the suppression assay were collected after 48 h and IFN-γ levels determined by ELISA (PBL Assay Science) following the protocol supplied by the manufacturers.
Statistical Analysis
Unless otherwise indicated and for comparisons of means, statistical analysis was performed using the Student t test, and a p value < 0.05 was considered significant (*p < 0.05; **p < 0.01; ***p < 0.001).
Discussion
CVID is the most frequently diagnosed primary immunodeficiency. Baseline inflammatory complications, autoimmune diseases, and lymphoproliferation are common in CVID patients [
4,
66‐
68], defining specific clinical phenotypes due to dysfunctional immune responses besides those seen upon recurrent infections [
40]. Although the mechanisms underlying CVID-associated immune dysregulation remain largely unclear [
68], previous reports have shown increased microbial translocation and systemic myeloid cell activation in CVID patients [
69‐
72], whose chronic monocyte activation appears related to persistence of T cell activation and the inflammatory and lymphoproliferative complications [
68,
73]. IVIg therapy is currently the treatment of choice for CVID, and we have previously shown that IVIg modifies the phenotype and function of myeloid cells in vitro and in vivo [
36,
37]. Although it is accepted that the main therapeutic benefit of IVIg in CVID patients is the presence of pathogen-specific antibodies [
74], IVIg-mediated cellular re-programming might also contribute to improve the control of infections in CVID patients [
10,
11,
75]. We now report that IVIg treatment of CVID patients provokes the acquisition of an anti-inflammatory profile in PBMC and monocytes, and that IVIg enhances the percentage of CD14
+ monocytes with a transcriptional, phenotypic, and functional profile compatible with those of myeloid-derived suppressor cells (MDSC). In parallel, IVIg infusion led to a marked reduction of the intermediate and non-classic monocyte subsets at the transcriptomic and phenotypic levels, a finding that is in agreement with the phenotypic effects of IVIg described by Cavaliere et al. in CVID patients [
54]. Our results indicate that monocytes are preferential IVIg targets in vivo, and that the IVIg-mediated changes in the relative levels of monocyte subsets might contribute to the anti-inflammatory and immunosuppressive effects of IVIg in CVID. Of note, no difference was observed between CVID patients with either “infections only” or “inflammatory” phenotype. However, given the size of the analyzed cohort, these results warrant further validation in an independent and larger cohort.
Compared to healthy individuals, CVID patients have been recently found to exhibit higher levels of low-density neutrophils, whose phenotype and suppressive activity is consistent with granulocytic MDSC [
70] that might contribute to the immune dysregulation in CVID. To our knowledge, the present report provides the first evidence for an IVIg-mediated increase in blood M-MDSC in CVID patients in vivo. We have also analyzed the levels of PMN-MDSC (CD15 + CD11b + CD33 + HLA-DRlow) in IVIg-treated CVID patients, but the results were inconclusive, and a larger cohort of CVID patients is required to clarify this issue in future studies. Our findings are in line with a previous report describing an increase in CD33
+/CD11b
+/HLA-DR
− MDSC in immune thrombocytopenia (ITP) patients treated with both IVIg and dexamethasone after 6 days [
76], and also agree with an enhancement of CD33
+/CD11b
+/HLA-DR
− cells in spleen cells from ITP patients exposed in vitro to IVIg for 90 h [
77]. Our results, however, indicate that IVIg enhances monocytic MDSC levels in peripheral blood as soon as 6 h after infusion, indicating an acute effect. MDSC appear to originate mainly from an emergency myelopoiesis [
78], and their enrichment might be due to monocyte reprogramming into an immunosuppressive state, early release of bone marrow immature myeloid cells into the circulation (emergency myelopoiesis), or a combination of both mechanisms [
79]. Whether the IVIg-induced increase in monocytic MDSC in CVID reflects a re-programming of peripheral blood monocytic cells or is secondary to release of bone marrow progenitors has yet to be addressed. Although the latter cannot be ruled out, since IVIg-treated CVID patients do not show elevated monocyte counts, and considering that IVIg re-program monocytes and macrophages in vitro [
36,
37], it is reasonable to assume that IVIg directly shapes peripheral blood monocytes at the transcriptional and phenotypic level in vivo in CVID patients. In any event, and regardless of its origin, the immunosuppressive character of IVIg-induced monocytic MDSC might help in protecting the host from the extensive tissue damage caused by the excessive monocyte activation usually observed in CVID [
73]. Moreover, and since MDSC also appear to increase immune surveillance and innate immune responses [
79], IVIg-induced MDSC might also contribute to maintain immune homeostasis and improve antimicrobial activities in CVID patients. Of note, we have not seen any differences in IVIg-induced MDSC increases between patients with distinct duration of IVIg treatment. However, considering the limited cohort we have analyzed, future studies should assess prospectively whether the duration of IVIg treatment has any effect on the IVIg-induced changes in MDSC population that we now report.
Regarding the IVIg-mediated decrease in the intermediate and non-classical monocyte subsets in CVID, seen at the phenotypic and transcriptional levels, our findings support the idea that IVIg can correct the imbalance of monocytes subsets seen in CVID patients, which exhibit increased levels of CD16
+ monocytes [
54,
73]. Indeed, our results corroborate previous findings on the ability of IVIg to diminish the number of non-classical monocytes in after 4 h in CVID patients [
54,
80‐
82], an effect that appears to be transient [
80] and has been also observed in patients with Kawasaki disease [
83]. Therefore, considering that CD16
+ monocytes exhibit more pro-inflammatory ability than classical monocytes [
50] and give rise to macrophages with a more pro-inflammatory gene profile [
84], the expression of genes preferentially expressed in intermediate/non-classical monocytes (Fig.
2D) could be used as molecular markers for immediate/early responses to IVIg infusion.
An additional consequence of the IVIg-mediated M-MDSC increase in CVID patients is its potential involvement in the generation of regulatory T lymphocytes (T
reg). Numerous reports have now established that IVIg enhances suppressive T
reg [
85‐
88], an effect observed in immune thrombocytopenia [
89], Guillain-Barré syndrome [
90], Myasthenia Gravis [
91], allergic airways disease [
92], Kawasaki disease [
93], and experimental autoimmune encephalomyelitis [
94]. In fact, T
reg expansion has been proposed as a biomarker to predict clinical response to IVIg therapy [
95], and is thought to be one of the mechanisms by which IVIg restores homeostasis in patients with autoimmune and systemic inflammatory disorders. Since MDSC promote T
reg expansion [
96‐
100] and recruitment [
101], the IVIg-induced increase in peripheral blood MDSC that we have observed in CVID has additional implications, and might be a primary step in the immunosuppressive ability of IVIg. While the global IVIg-induced anti-inflammatory effects may be beneficial to restrain chronic immune activation and inflammation in the complex interplay of factors involved in oncogenesis in CVID population [
102], a major concern of our data is the potential deleterious effects of IVIg on patients with established cancer. This latter aspect deserves further focused exploration.
In summary, we report that IVIg infusion has an immediate effect on the transcriptome, phenotype and function of peripheral blood monocytes in CVID patients, and that these IVIg-induced changes are compatible with IVIg promoting the acquisition of M-MDSC-like properties upon infusion. These results warrant further analysis of potential similar IVIg effects in other diseases, especially considering the ample immunomodulatory actions of MDSC and the large number of disorders that are currently treated with IVIg.
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.