Human peripheral blood mononuclear cells (PBMCs) were obtained from blood of healthy adult donors using a Ficoll-Plaque density gradient centrifugation Biocoll (Biochrom). Age of blood donors ranged between 18 and 35 years at the time of donation. PBMCs were used directly after isolation in transwell culture system. NK cells were isolated from PBMCs using negative selection kit EasySep human NK cell enrichment kit II (Catalog no. 19055, StemCell Technologies). Isolated NK cells were plated overnight before IFNα stimulation and degranulation assay; or used directly after isolation in transwell culture system. pDC were isolated from PBMCs using Plasmacytoid Dendritic Cell Isolation Kit II (Catalog no. 130-097-415, Miltenyi Biotec) and plated directly in transwell culture systems. All donors were recruited at the Leibniz Institute of Virology, Hamburg and provided written informed consent, studies were approved by the ethical committee of the Ärztekammer Hamburg (PV4780). Target cell line K562, a highly erythroleukemic cell line which does not express MHC class I molecules, was obtained from American Type Culture Collection (ATCC; Manassas, VA) and used for degranulation assays. K562 cell cultures were grown in RPMI-1640 supplemented with 10% heat-inactivated FBS and 1% Penicillin/Streptomycin (Sigma) at 37°C and 5% CO₂. K562 cells were regularly tested for mycoplasma contamination internally.

NK cells were cultured overnight (16h) with RPMI-1640 supplemented with 10% FBS, 1% Penicillin/Streptomycin (Sigma), 10 ng/ml IL-15 and 250 U/ml IL-2 (PeproTech) at a concentration of 2,5 × 10⁵ cells/ml. After, NK cells were pre-stimulated for 4h with either 10 IU/mL or 200 IU/mL of IFNα subtypes together with previously stated overnight medium culture conditions of IL-2 and IL-15. Recombinant human IFNβ (300-02BC, Preprotech); IFNα2 Human Interferon Alpha 2 (Alpha 2b) (11105–1), IFNα4 Human Interferon Alpha 4b (Alpha 4) (11180–1), IFNα7 Human Interferon Alpha J1 (Alpha 7) (11160–1), IFNα14 Human Interferon Alpha H2 (Alpha 14) (11145–1) were all provided by pbl assay science. IFN subtypes were resuspended according to the reported activity in human cells provided by the manufacturer. NK cells were then collected and used for degranulation assay and FACS staining.

Transwell co-culture consisted of 0,4 µm PET 24-well inserts (Thermo scientific/Sarstedt) in 24-well plates, which only allowed medium exchange and restricted NK cells to move to the lower chamber due to cell size restriction (data not shown). Transwell co-cultures were set by plating unstimulated or stimulated 1x10⁶ PBMCs or 4x10⁴ pDCs at the lower chamber, and 2x10⁵ isolated NK cells at the upper chamber for overnight (16h) stimulation with 1mL total volume per 24-well. PBMCs or pDCs were resuspended with TLR7/8 agonist CL097 (In vivogen) with final concentration of 1µg/mL; positive control IFNα14 at 200U/mL; unstimulated condition with medium alone. After overnight co-culture, NK cells were collected and used for FACS staining; culture medium supernatant was collected and used for IFNα quantification.

The frequency of degranulating NK cells was quantified by FACS staining, as previously described (31). Isolated NKs at 2,5x10⁵ cells/ml were stimulated with MHC devoid, K562 cells (ATCC), at an effector to target ratio of 1:5. Medium alone served as the negative control. Cells were stimulated with phorbol-12-myristate-13-acetate (PMA) (2.5 µg/ml) and Ionomycin (0.5 ug/ml) (Sigma) as a positive control. CD107a-BV510 antibody (BD Biosciences, San Jose, CA) was added directly to the tubes at 20 µl/ml. Cells were incubated for 1 h at 37°C in 5% CO₂ after which Brefeldin A (Sigma) was added at a final concentration of 10 µg/mL, as well as, Monensin (Golgi-Stop 554724, BD Biosciences) at a final concentration of 6 µg/mL and incubated for an additional 4h at 37°C in 5% CO₂ before FACS staining.

Cells were stained for surface markers for 20 min at room temperature in PBS (Sigma). Extracellular antibodies used were as follows: LIVE/DEAD Fixable Near-IR (Thermo Fischer Scientific; L10119), Anti-CD3-BUV737 (BD Biosciences; 612750), Anti-CD14-BUV737 (BD Biosciences; 612763); Anti-CD14-BUV395 (BD Biosciences; 563561); Anti-CD19-BUV737 (BD Biosciences; 564303), Anti-CD19-BUV395 (BD Biosciences; 563549); Anti-CD56-PerCP-Cy5.5 (Biolegend; 304626), Anti-CD16-Pe-Cy7 (Biolegend; 302016), Anti-CD107a-BV510 (Biolegend; 328632), Anti-CD69-BV650 (Biolegend; 310934). Samples were then washed and fixed using 4% formaldehyde or BD Cytofix/Cytoperm Kit (BD biosciences) according to manufacturer’s directions. Flow cytometry analysis was performed on a LSRII instrument (BD Biosciences). A total of 50,000 to 250,000 events were acquired and analyzed using FlowJo software. The analysis was performed on gated cells that fell within the lymphocyte population, stained as live cells using Live/Dead stain. Cells were then gated for negative CD3/CD14/CD19 expression to exclude other cell populations, and NK cells were identified using CD56/CD16 gating strategy. Gating strategies are shown in Supplementary Figure S1 . Within the NK cell population, we analyzed expression of CD107a and CD69 for each sample compared to their corresponding unstimulated control.

Concentration of IFNα was measured in duplicates from cell culture supernatant after overnight stimulation of PBMCs or pDC cultured in transwell culture system using Human IFN Alpha Multi-Subtype ELISA Kit (Invitrogen; 41105-1) following manufacture’s recommendations. Supernatants were stored at -80°C. Positive controls and stimulated supernatants were diluted 1/100 for appropriate cytokine detection, as recommended.

Graphs were generated and statistical analysis performed using GraphPad Prism 9 (GraphPad Software). Wilcoxon matched pairs signed rank test was used for comparison between two paired groups. Mann-Whitney test was employed for comparison between two unpaired groups. Two-way ANOVA test was used for comparison between two independent groups. Spearman correlation was used to assess the relationship between two sets of measurements from the same sample.

Previous studies have described that IFNα can induce NK cell function (21, 32) and that different IFNα subtypes mediate distinct antiviral immune functions (33–35). Here, we assessed the capacity of selected Type I IFN subtype proteins (IFNα2, IFNα4, IFNα7, IFNα14 and IFNβ) to induce activation and to promote degranulation of peripheral blood-derived NK cells by pre-incubating NK cells with different concentrations of IFN proteins prior to exposure to K562 cells. In line with previous studies using NK cells (36) and T cells (32, 37), different subtypes of Type I IFNs displayed different ability to induce NK cell activation, quantified by CD69 expression ( Figure 1A ), and degranulation, quantified by CD107a expression ( Figure 1B ). A representative example for CD69 expression ( Figure 1B ; right side) and CD107a expression ( Figure 1B ; right side) is illustrated; gating strategies are shown in Supplementary Figure S1 . Significantly stronger activation and degranulation by NK cells were observed at higher IFN concentrations (200 U/mL) compared to low IFN concentrations (10 U/mL), demonstrating a dose-dependent activation of NK cells. While all Type I IFN subtypes induced significant NK cell activation at low and high levels, significant degranulation was not induced after IFNα4 and IFNα7 stimulation at low concentrations. Pre-incubation with IFNα14 and IFNβ induced the highest induction of NK cell activation and degranulation, already at low concentrations. Furthermore, increases in NK cell activation (CD69) and degranulation (CD107) following activation with Type I IFN subtypes significantly correlated ( Figure 1C ; r=0.32; p<0.001). Taken together, these data demonstrate that different Type I IFN subtypes can differentially enhance functional response of activated NK cells against target cells.

Sex-differences in the production of Type I IFNs by PBMCs after TLR7/8-stimulation have been shown in multiple studies (11), and consistent with these previous studies we observed significantly higher levels of IFNα protein in supernatants (median 5.3-fold higher) following stimulation of PBMCs derived from females with the TLR7/8 agonist CL097 compared to male-derived PBMCs after 16h ( Figure 2A ; p=0.005). The production of IFNα is primarily attributed to pDC subpopulation within PBMCs (38). We therefore also stimulated isolated pDCs with the TLR7/8 agonist CL097, and observed comparable ranges of IFNα-production in supernatants ( Figure 2B ). As observed for PBMCs, female-derived pDCs produced significantly higher levels of IFNα (median 17-fold higher; p=0.008) than male-derived pDCs ( Figure 2B ). Using the transwell co-culture system depicted in Figure 2C , we subsequently aimed to determine the effects of higher production of IFNs after TLR7/8 stimulation by female PBMCs/pDCs on NK cell functions. Using NK cells derived from the same donor, we compared the effect of male- versus female-derived PBMCs, either non-stimulated or stimulated with CL097. Significantly higher NK cell activation levels, quantified by CD69 expression, were observed after NK cell co-incubation with CL097-stimulated PBMCs derived from females compared to males ( Figure 2D ; p=0.03 dotted line), showing that higher IFNα production from stimulated female-derived PBMCs can trigger higher NK cell activation. These findings were replicated using CL097-stimulated pDCs ( Figure 2E ; p=0.02 dotted line), thus demonstrating that pDCs are the primary source of IFNα production after TLR7/8-stimulation regulating NK cell activation ( Figure 2E ). Further comparison of CD69 expression between CD56^bright and CD56^dim NK cells subset showed that CD56^dim NK cells exhibit the highest fold change in response to activation ( Supplementary Figures S1B–D ), in line with previous studies (23). Taken together, female-derived pDCs from PBMCs are able to induce stronger NK cell activation compared to male-derived pDCs after TLR7/8 stimulation, demonstrating an extrinsic sex-specific effect resulting from higher IFNα production of female pDCs that can modulate NK cell functions.

Based on the recent description that sex-differences in UTX expression can result in stronger NK cell function in female mice (28), we aimed to investigate potential intrinsic sex-differences in the response of NK cells to Type I IFNs. Direct exposure to IFNα14 resulted in significantly enhanced activation of both male- and female-derived NK cells compared to their untreated condition ( Figure 3A , no PBMCs and Figure 3B , no pDCs). However, IFNα14-induced activation was higher in NK cells derived from females (15-fold increase) compared to NK cells derived from males (11-fold increase) ( Figure 3A , no PBMCs; p=0.005 dotted line), and a comparable stronger activation of NK cells derived from females (18- versus 10-fold increase; p=0.01) was observed after IFNα14-induced activation in Figure 3B (No pDCs; dotted line). We next used the above described transwell co-culture systems of NK cells with CL097-stimulated PBMCs, or CL097-stimulated pDCs, to investigate the combined effects of sex differences in Type I IFN production and in IFN-responsiveness of NK cells. Female-derived NK cells exhibited significantly enhanced activation compared to male-derived NK cells in co-culture with CL097-stimulated PBMCs ( Figure 3A , +PBMCs; p=0.01 dotted line within rectangle) or pDCs ( Figure 3B , +pDCs; p=0.008 dotted line within rectangle), independent of the sex of the PBMC/pDC donor. A subsequent integrated analysis combining both the sex of the NK cell donors and the PBMC/pDC donors as variables showed that strongest activation of NK cell was observed when CL097-stimulated PBMCs ( Figure 3C ; p=0.004 dotted line) or pDCs ( Figure 3D ; p=0.03 dotted line) were derived from females and incubated with NK cells derived from females. Taken together, these results demonstrate that NK cells derived from females are more strongly responding to Type I IFN-mediated activation than NK cells derived from males, and that this effect is strongest when co-incubated with IFNα-producing cells derived from females.