*2.11. Determination of Resistance/Level of Innate Immunity of PBLs*

Resistance/innate immunity was determined by infection of leukocytes (1 × 106 cells/mL) ex vivo with a VSV dose of 100 TCID50. After 40 min of adsorption at room temperature (rt), the virus was washed out three times with RPMI medium with 2% FBS, and the cells were suspended in 1 mL of RPMI 2% FBS for investigations of the influence of EGb on PBLs resistance/innate immunity and cytokine production. A sample of the infected cells was kept at 4 ◦C and served as a control of the starting level of the virus. The rest of the cells were divided to two parts, i.e., VSV-infected and uninfected, and next were treated with 150 μg/mL EGb and incubated at 37 ◦C for 24 h. After that, time samples of the medium above the cells were collected and titrated in L929 cells. Viral titer was expressed in TCID50. Resistance of PBLs to VSV infection was assessed as follows: a VSV titer ≥ 4 log TCID50 was considered as a lack of resistance (deficiency in innate immunity), a titer of 2–3 log indicated partial resistance, and a titer of 0–1 log indicated complete resistance to VSV infection (high level of innate immunity).

#### *2.12. Cytokine Measurement*

The levels of IL-1β, IL-10, IL-15, IFN-α, IFN-γ, and TNF-α in supernatants from uninfected and VSV-infected leukocytes were detected using enzyme-linked immunosorbent assays (BD OptEIA TM human IL-1β, IL-10, IL-15, IFN-γ,TNF-α ELISA set, BD Biosciences; IFN-α Human ELISA Kit, Thermo Fisher, Carlsbad, CA, USA). The optical density was measured at 450 nm with λ correction 570 nm using a Multiskan RC spectrophotometric reader (Thermo Labsystems, Philadelphia, PA, USA). Cytokine concentrations were expressed in pg/mL.

#### *2.13. RNA Isolation and Real-Time PCR*

Real-time PCR was used to investigate mRNA expression. Total RNA (obtained from uninfected and VSV-infected cells treated with EGb 150 μg/mL) was extracted with the Relia Prep™ RNA Cell Miniprep System kit (Promega, Madison, WI, USA), and reverse transcription was performed using the High Capacity cDNA Reverse Transcription kit (Applied Biosystem, Thermo-fisher Scientific, Carlsbad, CA, USA) according to the manufacturer's instructions. Reaction was performed in T3000 Thermocycler (Biometra, Göttingen, Germany). Reaction volume was 20 μL, and final cDNA product was kept in −20 ◦C for a few weeks preceding quantitative PCR. Expression of interferon regulatory factor 3 and 7 (*IRF-3*, *IRF-7*), tetherin (*BST2*), suppressor of cytokine signaling 1 and 3 (*SOCS1, SOCS3*), nuclear factor NF-kappa-B p105 and p65 subunit (*NFKB1*, *p65* alias *RELA*), and interferon-induced GTP-binding protein Mx1 (*MxA*) was studied by quantitative PCR (qPCR) using Taq DNA Polymerase with Sybr Green I dye. Data were normalized to endogenous reference gene *18S*, which was confirmed to be stable across the groups. Reaction was performed using SG qPCR Master Mix (EURx, Gda ´nsk, Poland) in combination with appropriate primers shown in Table 1. Furthermore, primer specificity was confirmed empirically after qPCR by performing melting curve analysis of every sample. All experiments were performed using Light Cycler 480 II instrument (Roche Diagnostics GmbH, Basel, Switzerland). Every reaction was performed two times and differences in quantification cycle between both repeats were negligible (±0.2). Quantification cycles (Cq) for every reaction were calculated automatically by the Light Cycler 480 II instrument software (Roche Diagnostics GmbH, Basel, Switzerland). Relative changes in mRNA expression were calculated using the delta Cq method with the healthy control as the comparator.


**Table 1.** Primers used for real-time PCR.

#### *2.14. Statistical Analysis*

To capture the central tendency (average) of the data, the Hodges–Lehman estimator—pseudo median—was used (in the following part called *Median*). Confidence intervals (CI95) at significance level *α* = 0.05 were used to measure precision of estimation and testing some statistical hypothesis. CI95s were estimated with bootstrap method. Statistic *Sn* = *med*- *med xi* – *xj* ; *<sup>j</sup>* <sup>=</sup> 1... *<sup>n</sup>* was used as robust measure of variability [19] as well as minimal and maximal observations. *Sn* is typical difference between two randomly sampled observations. Delta Δ is Hodges–Lehman estimator of the shift parameter between distributions of two independent populations. Delta Δ is median of differences between all pairs of observations, where one observation belongs to group A, and the second observation in the pair belongs to group B. Delta is shift parameter used also in Wilcoxon rank-sum test to comparison two independent populations (also known as Mann–Whitney test). Delta, CI95 for delta, and *p*-values of the delta were estimated numerically with bootstrap method because of a small sample sizes. In the case of investigation of an influence of EGb treatment on the level of innate immunity of AD patients and controls, an analysis of variance (ANOVA) was used. In the case of ANOVA, statistical effect size for every variable in the model was measured with partial eta-squared defined as *η*<sup>2</sup> = *SSvariable SSvariable*+*SSerror* , where *SSvariable* and *SSerror* is sum of squares in ANOVA for effect of the variable and sum of squares for the experimental error, respectively. Interpretation of the *η*<sup>2</sup> was typical, after Cohen.

For every investigated *i* − *th* individual from AD patients and from controls, the effect of EGb treatment was calculated as *di* = *loge level a f ter treatment level be f ore treatment* for all cytokine production and for all gene expression. An average effect in the group was estimated as pseudo median (*Median*) with Hodges–Lehman estimator (*HL*)f central tendency, so *EGb ef f ect* = *HL*(*d*).
