IgG Immune Complexes Inhibit Naïve T Cell Proliferation and Suppress Effector Function in Cytotoxic T Cells

Elevated levels of circulating immune complexes are associated with autoimmunity and with worse prognoses in cancer. Here, we examined the effects of well-defined, soluble immune complexes (ICs) on human peripheral T cells. We demonstrate that IgG-ICs inhibit the proliferation and differentiation of a subset of naïve T cells but stimulate the division of another naïve-like T cell subset. Phenotypic analysis by multi-parameter flow cytometry and RNA-Seq were used to characterize the inhibited and stimulated T cells revealing that the inhibited subset presented immature features resembling those of recent thymic emigrants and non-activated naïve T cells, whereas the stimulated subset exhibited transcriptional features indicative of a more differentiated, early memory progenitor with a naïve-like phenotype. Furthermore, we show that while IgG1-ICs do not profoundly inhibit the proliferation of memory T cells, IgG1-ICs suppress the production of granzyme-β and perforin in cytotoxic memory T cells. Our findings reveal how ICs can link humoral immunity and T cell function.

or negative controls (blue) show that IgG1-ICs inhibit naïve T cell proliferation (Generation 1) and stimulate a subset of dividing progeny. The latter phenomenon is observed as early as after 2-3 divisions and consistently in later divisions (Generation 6+). Negative controls include Fc5 IgG1-ICs, monomeric IgG1, TNP-BSA only, PBS, and/or other negative controls (Table S1). Mean values are plotted. Error bars represent the standard error of the mean (SEM) for each group. p-values are calculated using unpaired two-tailed t-tests adjusting for multiple comparisons using the Sidak-Holm correction. Consistent standard deviation is not assumed unless an experiment pertains to a single-well assay. Asterisks * , ** , and *** indicate p-values < 0.05, p-values < 0.01, and p-values < 0.0001 respectively. "n.s." stands for statistically not significant. θi (or 100-Փi) is % division (or 100-% undivided) for a particular T cell generation (i). For instance, θ1 is equal to (100-Փ1) where Փ1 is % undivided. Փ1 can be calculated using the following equation: where D is the total number of observed divisions (number of peaks -1) and X1 is the frequency of the 1 st peak gated in FlowJo. To calculate Փ2, the same equation is used but "X1" would need to be recalculated by excluding generation 1 before gating. To calculate Փ3, the same equation is used but "X1" would need to be recalculated by excluding generations 1 and 2 before gating-so on and so forth. A template excel sheet that automatically calculates and plots θi (or Փi) vs. is provided as a supplemental attachment. Alternatively, % undivided (Փ1) and other proliferation parameters can be calculated by the proliferation plugin in newer versions of FlowJo.
(b-c) Bar graph of percent undivided T cells (Փ1) for naïve and memory CD4+ and CD8+ T cells. The graph presents the data aggregate across all independent experiments, donors, and negative controls. p-values were calculated using one-way ANOVA. Asterisks **** indicate p-values < 0.0001. "n.s." stands for statistically not significant. Figure S4 | Only WT IgG Immune Complexes, but not IgA1 or IgE Immune Complexes, Inhibit Naïve T Cell Proliferation Division of CellTrace-stained CD8+ or CD4+ T cells activated with Dynabeads and incubated with WT IgG1-ICs (red), Fc5 IgG1-ICs (dark blue), IgA1-ICs (purple), IgE-ICs (dark orange), or other negative controls. Results for Donors 062 and Donor 060 are also presented in Figure 2e and are shown here overlaid with control ICs and next to all the negative controls utilized in these experiments.

Figure S5 | CCR6-CXCR3-Naïve CD4+ T cells are Profoundly Inhibited by WT IgG Immune
Complexes Naïve CD4+ T cells activated with Dynabeads and incubated with IgG1-ICs or negative controls are cultured

Figure S6 | IgG Immune Complexes Do Not Change Overall Culture Viability
The viability of naïve, memory, and total T cell cultures that are either non-stimulated or activated with anti-CD3/anti-CD28 Dynabeads in the presence of various controls (e.g. Fc5 IgG1-ICs, WT IgG1-ICs, or monomeric IgG1). The bars represent the mean and the error bars represent the standard error of the mean (SEM).

Figure S8 Continued | Other Markers Included in t-SNE Analyses and Gene Expression Data Complementing t-SNE
Analyses t-distributed Stochastic Neighbor Embedding (tSNE) analysis of multi-color flow cytometry panels of naïve T cells activated with Dynabeads and incubated with (WT) IgG1-ICs or negative controls (Fc5 IgG1-ICs, monomeric IgG1, PBS only, etc.). Naïve T cells were sampled equally across all controls and donors (N=3 donors; 2 naïve CD8+ and 1 naïve CD4+). Red and blue arrows point to clusters of T cells treated with IgG1-ICs and Fc5 IgG1-ICs, respectively. Generations 6+ (Gen6+) represents T cell populations that have divided at least 5 times, whereas Generation 1 (Gen1) represents undivided T cells. In addition to markers shown in main Figure 3, (a) Panel A examines markers associated with activation, exhaustion, and/or anergy. Gen1 T cells originating from IgG1-IC-treated or control-treated wells are shown in dark and light pink, respectively. Gen6 T cells originating from IgG1-IC-treated or control-treated wells are shown in dark and light grey, respectively. Heat-map t-SNE plots are shown for (i) CellTrace (ii) simultaneous stain for CCR7, CD62L, and CD28 (i.e. CCR7/CD62L/CD28 lump gate) (iii) CD57 (iv) simultaneous stain for CX3CR1 and KLRG1 (i.e. CX3CR1/KLRG1 lump gate) and (v) BTLA; (b) Panel B examines markers associated with differentiation status. Gen1 T cells originating from IgG1-IC-treated or control-treated wells are shown in dark and light purple, respectively. Gen6 T cells originating from IgG1-IC-treated or control-treated wells are shown in dark and light grey, respectively. Heat-map t-SNE plots are shown for select markers: (i) CellTrace (ii) CD127 (IL-2Rβ) (iii) CD45RO and (iv) CD28. (c) Heat-map t-SNE plots of non-stimulated naïve T cell are shown for all markers in Panels A and B. (e) Naïve CD8+ T cells activated with Dynabeads and incubated with either IgG1-ICs (red) or Fc5 IgG1-ICs (blue) were studied. However, IgG1-IC-treated T cells were first sorted to enrich for T cell populations stimulated by IgG1-ICs (roughly corresponding to Gen6+) and T cell populations inhibited by IgG1-ICs (Gen1). Fc5 IgG1-IC-treated T cells (control) are not sorted. Differential expression of genes of markers used in t-SNE analyses are shown to aid interpreting results associated with "lump gates" (e.g. TIGIT/TIM-3/LAG-3 on the APC channel). Genes are considered differentially expressed if the adjusted p-value < 0.05 and |fold change| is ≥ 1.5. (e) Univariate histograms for select markers from Panels A and B are shown. The black and fuchsia histograms represent Gen6+ and Gen1 T cells in all samples analyzed by t-SNE. Unfilled histograms pertain to non-stimulated T cells. The scatter dot plot shows the ratio of CD45RA:CD45RO expression (based on average MFI) for Gen6+ T cells that were incubated with IgG1-ICs (red) or controls (blue). Values below the dashed lines are negative (a common, unavoidable consequence of compensation). In this case, CD45RA expression for Donor 057 was much lower (negative CD45RA MFI) than CD45RO. Error bars represent the standard error of the mean (SEM) for each group. p-values were calculated using unpaired two-tailed t-tests using the Sidak-Holm correction for multiple comparisons. Asterisks * , ** , and *** indicate p-values < 0.05, p-values < 0.01, and p-values < 0.0001, respectively.

Figure S9 Conitued | Gene Expression of Select Receptors Including Lectins and Canonical/Non-Canonical Fc Receptors
Transcripts per million (TPM) for various genes (including various C-type lectins, Sialic-acidbinding immunoglobulin-type lectins, Fc receptors, and Complement receptors) for various resting and activated lymphocyte subsets. RNA-Seq raw data for all "Resting" subsets is obtained from Ranzani et al 149 . The median (line), mean (dot), minimum/maximum(whiskers), and interquartile range (box) are shown. The dotted and dashed lines represent the TPM=1 (expressed) and TPM=0.5 (very low expression) thresholds.

Figure S10 | FcγR Protein Expression is Observed Intracellularly and on the Cell Surface of Various T Cell
Subsets (a,b) Various CD8+ and CD4+ T cell subsets were stained with anti-CD32 mAb (colored histograms) or isotype control (filled grey). Cells were either stained extracellularly (surface stain) or both intracellularly and extracellularly. FMOs are shown as unfilled histograms. Displayed percentages correspond to % positive events set relative to unstained cells. Note that "surface stain" only panels from Donor 090 are also presented in Figure  6 but are shown here again alongside surface+intracellular stains for comparison.   ( Table S1 | Immune Complexes and Other Experimental Controls Utilized in Functional Experiments Immune complexes were prepared by pre-incubating TNP-BSA and anti-TNP antibodies. Monomeric controls were generated by pre-incubating TNP-BSA and Isotype control antibodies or WT BSA and Anti-TNP antibodies. Antigen only controls pertain to TNP-BSA or WT-BSA and PBS (no antibodies). Expected binding to FcRs is based on ELISA results ( Figure  S1) in the relevant range of treatment concentration (1-50 μg/mL). Details pertaining to amounts and incubation time are provided (Methods). proliferation experiments across T cell subsets and human donors. p-values are calculated using unpaired, two-tailed t-tests adjusting for multiple comparisons using the Sidak-Holm correction. Consistent standard deviation is not assumed unless an experiment pertains to a singlewell assay (values with no standard deviations). The mean ± standard deviation and range of percent differences between IgG1-IC-treated T cells and controls are also shown.

Markers Associated with Activation, Exhaustion, and Anergy
Differentiation-Associated Markers

Table S4 | Differentially Expressed or Up/Downregulated Genes Mentioned In-Text
DESeq2 was used to analyze differentially expressed genes (DEGs); genes were considered differentially expressed if the adjusted p-value < 0.05 and the |fold change (FC)| is ≥ 1.5 (i.e. the |Log 2 (FC)| is ≥ 0.58496). Non-DEGs were considered upregulated/downregulated if either the adjusted or nominal p-value < 0.10. Grey blocks represent genes whose differences were not statistically significant (i.e. both nominal and adjusted p-value > 0.10)

Table S5 | Average Transcripts Per Million (TPMs) for Genes Mentioned In-Text
Note that TPMs are not well-suited for comparisons across samples (DESeq2 analysis is used for differential expression across samples) only within genes of a sample. TPM<0.5 indicates weak or no expression. TPM > 1 indicates expression.  Figure 5, GSEA analyses reveal other potentially meaningful enrichments. Gene set enrichment analysis (GSEA) identifies pathways that correlate to the enrichment of genes across the InhT, StimT, and Fc5T groups. "Rest" represents any two of these groups. The FDR q-value and the normalized enrichment score (NES) are tabulated.

OT-I/II T Cell In Vitro Activation
Spleens were excised from OT-I and OT-II mice. T cells were then magnetically isolated from processed splenocytes (STEMCELL 19851). T cells were cultured at 1M/mL in complete medium (10ng/mL rμ-IL2) and activated with anti-mouse CD3/CD28 dynabeads (Thermo). Four days later, activated T cells were washed and processed for further analysis.
Approximately 1M sorted OT1 and OT2 T cells were processed for RNA extraction (Qiagen). 580 ng of OT1 total RNA and 310 ng of OT2 total RNA were processed to amplify cDNA using pre-designed primers and SuperScript® III One-Step RT-PCR System with Platinum® Taq DNA Polymerase (Thermo). The cDNA production step was set at 55⁰C for 35 minutes. That was followed by DNA denaturation (120s, 94⁰C) and 32 PCR cycles-denaturation (15s, 94⁰C), annealing (30s, 55⁰C), extension (35s, 68⁰C). Employed primer sequences and expected band sizes are provided (Table S3).

FcγR Expression by Flow Cytometry
Anti-mouse CD32b (clone AT130-2) and its matching IgG2a isotype control (clone eBM2a) were purchased from eBioscience. Anti-mouse CD16/CD32 (clone 93) was purchased from Biolegend. Isotype control brightness and final concentrations were matched to that of FcγR-staining mAbs before staining.
Where indicated, an FcR-blocking peptide solution was used (Innovex Biosciences). Activated OT T cells were stained for FcγRIIb with anti-CD32b mAb PE (Clone AT130-2) or isotype (clone eBM2a) and analyzed by FACS. To ensure that observed fluorescence shifts were indeed due to FcγRIIb binding, aliquots of anti-CD32b were pre-incubated with either soluble his-tagged-murine FcγRIIb, irrelevant soluble his-tagged-murine FcγRI, or DPBS. The soluble receptors (2-3 mg/mL) were added in molar excess (~50 μg HIS-FcγR in ~20uL DPBS) to the staining antibody (~2 μg in ~10uL) and incubated on ice for at least 30 minutes. Using those stocks, equal molar amounts of antibody were used to stain cells before FACS analysis. Cells were pre-blocked with rat serum for 20 minutes on ice. Antibody staining was done in 20% rat serum DPBS. Recombinant HIS-tagged FcγRs were produced in Expi293F cells, as described earlier, and purified by Nickel-NTA column chromatography (Thermo).

Supplementary Attachments
Supplementary Attachment 1. Proliferation Analysis Template. See separate excel file.