Veronica austriaca L. Extract and Arbutin Expand Mature Double TNF-α/IFN-γ Neutrophils in Murine Bone Marrow Pool

Plants from the Veronica genus are used across the world as traditional remedies. In the present study, extracts from the aerial part of the scarcely investigated Veronica austriaca L., collected from two habitats in Bulgaria—the Balkan Mountains (Vau-1) and the Rhodopi Mountains (Vau-2), were analyzed by nuclear magnetic resonance (NMR) spectroscopy. The secondary metabolite, arbutin, was identified as a major constituent in both extracts, and further quantified by high-performance liquid chromatography (HPLC), while catalpol, aucubin and verbascoside were detected at lower amounts. The effect of the extracts and of pure arbutin on the survival of neutrophils isolated from murine bone marrow (BM) were determined by colorimetric assay. The production of cytokines—tumor necrosis factor (TNF)-α and interferon (IFN)-γ was evaluated by flowcytometry. While Vau-1 inhibited neutrophil vitality in a dose-dependent manner, arbutin stimulated the survival of neutrophils at lower concentrations, and inhibited cell density at higher concentrations. The Vau-1 increased the level of intracellular TNF-α, while Vau-2 and arbutin failed to do so, and expanded the frequency of mature double TNF-α+/IFN-γhi neutrophils within the BM pool.


Identification of Metabolites in Veronica austriaca Extracts
In the present report, we describe the analysis of V. austriaca collected from two different habitats, with 4 biological replicates each. We applied 1 H NMR in combination with 2D NMR techniques (J-resolved, COSY, HSQC) of the methanolic extracts. In total, 13 individual compounds were assigned to the abundant signals, including amino acids, organic acids, iridoid glucosides and phenolic compounds (Table 1).  The major compound in extracts of both samples was identified as arbutin, a hydroquinone derivative-4-hydroxyphenyl-β-glucopyranoside. The most intensive doublets in 1 H NMR spectra at δ 7.00 and δ 6.79 ppm (d, 8.9 Hz), corresponding to AA'BB' system and a single doublet of anomeric proton of β-glucose moiety at δ 4.85 ppm (d, 7.9 Hz), were assigned to this phenolic glucoside and were confirmed by comparison with an authentic sample ( Figure 1, Table 1). The presence of catalpol, aucubin and verbascoside in extracts was confirmed accordingly. Furthermore, the presence of arbutin in both extracts was determined quantitatively by HPLC, and the content of 308.19 ± 20.67 µg/mg and 382.93 ± 28.11 µg/mg extract for Vau-1 and Vau-2 was established, respectively. Table 2 presents the content of arbutin in various concentrations of the extracts applied in our study.
The major compound in extracts of both samples was identified as arbutin, a hydroquinone derivative-4-hydroxyphenyl-β-glucopyranoside. The most intensive doublets in 1 H NMR spectra at δ 7.00 and δ 6.79 ppm (d, 8.9 Hz), corresponding to AA'BB' system and a single doublet of anomeric proton of β-glucose moiety at δ 4.85 ppm (d, 7.9 Hz), were assigned to this phenolic glucoside and were confirmed by comparison with an authentic sample ( Figure 1, Table 1). The presence of catalpol, aucubin and verbascoside in extracts was confirmed accordingly. Furthermore, the presence of arbutin in both extracts was determined quantitatively by HPLC, and the content of 308.19 ± 20.67 µg/mg and 382.93 ± 28.11 µg/mg extract for Vau-1 and Vau-2 was established, respectively. Table 2 presents the content of arbutin in various concentrations of the extracts applied in our study.   We examined the effect of the methanolic extract Vau-1, Vau-2 or pure arbutin (all dissolved in dimethyl sulfoxide (DMSO)) on survival of neutrophils isolated and purified from bone marrow (BM). We observed the decreased vitality of neutrophils cultured in the presence of 50 µg/mL Vau-1 or Vau-2 extracts in comparison to control culture ( Figure 2B). We noticed that the vehicle DMSO (0.3%) itself can have a marginal stimulatory effect on neutrophil vitality ( Figure 2B), hence, we further evaluated the activity of the extracts or arbutin in comparison to DMSO-treated cultures ( Figure 2C,D). Vau-1 induced a dose-dependent decrease in neutrophil survival ( Figure 2C). The treatment of neutrophils with Vau-1 led to a dose-dependent decrease in cell survival, with the lowest survival observed at the highest concentration of extract used, 50 µg/mL. Interestingly, pure arbutin at the same concentration also demonstrated an inhibitory effect, albeit less pronounced, on neutrophil survival, considering that 50 µg/mL Vau-1 extract contains 56.7 µM arbutin ( Figure 2B; Table 2). At concentrations ranging from 4 to 15 µg/mL (14.7 to 55.1 µM), arbutin increased neutrophil survival; at higher concentrations from 61-245 µg/mL (224 to 900 µM), it decreased cell density and cell vitality ( Figure 2D).

Effect of Vau-1/Vau-2 Extracts or Arbutin on TNF-α and IFN-γ Production in BM-Derived Neutrophils
Next, we delineated the production of the pro-inflammatory cytokines IFN-γ and TNF-α in BM-derived neutrophils, using flowcytometry and measuring mean fluorescence intensity (MFI) ( Figure 3A,B). The presence of DMSO (0.3%) had a stimulatory effect on IFN-γ production that was reduced by both extracts Vau-1 and Vau-2 at concentrations < 50 µg/mL (corresponding to 56.7 µM  Table 2) ( Figure 3A). Both Vau-1 and Vau-2 strongly inhibit the intracellular level of IFN-γ at a low concentration (0.025 µg/mL of the extracts contain 0.028 µM arbutin in Vau-1 and 0.035 µM arbutin in Vau-2, respectively; Table 2; Figure 3A). Arbutin had an opposite effect on intracellular IFN-γ by significantly inducing IFN-γ production when used at lower concentrations, ranging from 0.25 to 0.0025 µg/mL (corresponding to 0.9 to 0.009 µM, respectively) ( Figure 3A). These data suggest that the action of arbutin on IFN-γ production might be masked by other molecules present in the Vau-1 and Vau-2 extracts, or that arbutin specifically interfered with phorbol 12-myristate 13-acetate and ionomycin (PMA/Yon) signaling. Purified neutrophils were cultured in the presence of DMSO (0.3%) and decreasing concentrations of Vau-1, Vau-2 and arbutin for 36 h. Control cells were incubated with phosphate-buffered saline (cells). In the last 4 h of incubation, the neutrophils were stimulated with PMA/Yon in the presence of the Golgi inhibitor, monensim, in order to maximize cytokine accumulation. The neutrophils were then washed and stained for the neutrophil marker, Ly6G. The cells were then fixed, permeabilized and incubated with PE/Cy7 or APC/Cy7 conjugated antibodies against TNF-α and IFN-γ. After washing, gated Ly6G + cells were subjected to flow cytometry analysis for the intracellular production of cytokines. Data represent mean ± SD of sample from 2 experiments with neutrophils isolated and pooled from 7 mice and assayed in duplicate. P-values are shown for each group when compared to the group cultured with 0.3% DMSO; * p < 0.05; ** p < 0.01; *** p < 0.001, two-tailed Student t-test.

Effect of Vau-1/Vau-2 or Arbutin on the Pattern of Immature or Mature Neutrophils Expressing TNF-α and IFN-γ
The bone marrow contains neutrophils at various stages of maturation, which can be distinguished in mice by the density of Ly6G expression and cellular granularity ( Figure 4A). Ly6G hi neutrophils (cells expressing the marker at high levels) were defined as mature, while Ly6G low neutrophils (expressing the marker at low levels) were defined as an immature population ( Figure  4A). Ly6G + cells also showed variability in granularity content, when evaluated using the SSC-scatter on the flow cytometer. Immature cells were less granular and had a round shape, while In the last 4 h of incubation, the neutrophils were stimulated with PMA/Yon in the presence of the Golgi inhibitor, monensim, in order to maximize cytokine accumulation. The neutrophils were then washed and stained for the neutrophil marker, Ly6G. The cells were then fixed, permeabilized and incubated with PE/Cy7 or APC/Cy7 conjugated antibodies against TNF-α and IFN-γ. After washing, gated Ly6G + cells were subjected to flow cytometry analysis for the intracellular production of cytokines. Data represent mean ± SD of sample from 2 experiments with neutrophils isolated and pooled from 7 mice and assayed in duplicate. P-values are shown for each group when compared to the group cultured with 0.3% DMSO; * p < 0.05; ** p < 0.01; *** p < 0.001, two-tailed Student t-test. When we investigated the intracellular level of TNF-α ( Figure 3B), we found that Vau-1 stimulated, in a dose-dependent manner, this cytokine's production. Vau-2 and arbutin inhibited the cytoplasmic TNF-α in comparison to DMSO group, suggesting that other molecules in the extract might contribute to this activity. To eliminate the possibility that Vau-2 and arbutin induce the release of the cytokine, rather than its accumulation in the cytoplasm, the neutrophils were stimulated with PMA/Yon in the presence of monensim.

Effect of Vau-1/Vau-2 or Arbutin on the Pattern of Immature or Mature Neutrophils Expressing TNF-α and IFN-γ
The bone marrow contains neutrophils at various stages of maturation, which can be distinguished in mice by the density of Ly6G expression and cellular granularity ( Figure 4A). Ly6G hi neutrophils (cells expressing the marker at high levels) were defined as mature, while Ly6G low neutrophils (expressing the marker at low levels) were defined as an immature population ( Figure 4A). Ly6G + cells also showed variability in granularity content, when evaluated using the SSC-scatter on the flow cytometer. Immature cells were less granular and had a round shape, while mature neutrophils were identified as SSH hi ( Figure 4A), as they accumulated granules during differentiation. The total pool of BM-derived neutrophils contained sub-populations of cells with various shapes and maturity statuses ( Figure 4A,B). In addition, each sub-population showed a particular pattern of cytoplasmic IFN-γ and TNF-α levels ( Figure 5).
Molecules 2020, 25, x 7 of 16 mature neutrophils were identified as SSH hi ( Figure 4A), as they accumulated granules during differentiation. The total pool of BM-derived neutrophils contained sub-populations of cells with various shapes and maturity statuses ( Figure 4A and 4B). In addition, each sub-population showed a particular pattern of cytoplasmic IFN-γ and TNF-α levels ( Figure 5).   The vehicle DMSO (0.3%) increased the proportion of cells with an immature phenotype SSC hi Ly6G low and SSC low Ly6G low ( Figure 4B). Vau-1/Vau-2 and arbutin all increased in a dose-dependent manner the immature transient pool, defined as SSC hi Ly6G low (in blue), and decreased the mature transient pool, defined as SSC low Ly6G hi (in green; Figure 4B). Vau-2 and arbutin, but not Vau-1 (at low concentration) expanded the SSC low Ly6G low immature pool in comparison to the controls (in red; Figure 4B). The pool of SSC hi Ly6G hi mature cells was neither markedly affected by the extracts, nor expanded significantly by arbutin at the concentrations used (purple; Figure 4B).
In our previous experiments (Figure 3), the extracts and arbutin changed IFN-γ and TNF-α cytoplasmic levels, but it is not clear if this effect is characteristic for a particular stage of neutrophil maturation. In the Ly6G low state, neutrophils were single IFN-γ producers ( Figure 5A), double TNF-α/IFN-γ producers or negative for both cytokines (representative dot plots at Figure 5A). The transition from Ly6G low to Ly6G hi was related to the loss of single IFN-γ producing cells and progressive expansion of the double TNF-α/IFN-γ positive neutrophils. Thus, we focused on the frequency of those double cytokine producers as an additional indicator for neutrophil maturation. We also distinguished the pattern of low and high IFN-γ expression that varied within the Ly6G + sub-populations ( Figure 5A). In control groups (control or DMSO treated), low intracellular IFN-γ level was characteristic for mature transient and mature states rather than immature pools ( Figure 5A,B). The extracts and arbutin expanded in a similar manner the frequency TNF-α positive cells with lower IFN-γ expression (TNF-α + /IFN-γ low , Figure 5B), within the pool of mature transient state in comparison to control groups (in red in Figure 5B). Vau-1, at a low concentration (0.5 µg/mL), increased the proportion of TNF-α + /IFN-γ low cells within the immature pool (in blue, Figure 5B), while Vau-2 (0.5 µg/mL) and arbutin (from 0.25 to 24.5 µg/mL (0.9-90 µM)) increased the proportion of those double producers in the immature transient pools (in green, Figure 5B). Vau-2 and arbutin, but not Vau-1, decreased, in a dose-dependent manner, the frequency of TNF-α + /IFN-γ low producers within the mature pool (in purple, Figure 5B).
The frequency of double TNF-α + /IFN-γ hi ( Figure 5C) producers was higher in the extract-or arbutin-treated groups in comparison to controls. Vau-1 (0.5 µg/mL) further increased the proportion of those cells within the immature pool (in blue, Figure 5C), while Vau-2 and arbutin expanded, in a dose-dependent manner, the proportion of TNF-α + /IFN-γ hi neutrophils with mature phenotype (purple, Figure 5C).

Discussion
In this study we performed chemical analysis of methanolic extracts from V. austriaca collected from two different habitats. The 1 H NMR, in combination with 2D NMR spectroscopy, were used, and 13 individual compounds, including amino acids, organic acids, iridoid glucosides and phenolic compounds were identified. Among the secondary metabolites, the iridoid glucosides aucubin and catalpol, as well as the phenylethanoid glycoside-verbascoside found in both V. austriaca extracts are a characteristic feature for the Veronica plant species [2]. Previously, we observed that verbascoside acts as a potent modulator of neutrophil priming and activation, by altering various cellular functions, such as expression of the integrin CD11b and the chemokine CXCR2, and production of TNF-α and the degrading enzyme, MMP-9 [15]. At the molecular level, verbascoside can interfere with p38 signaling, while the compound aucubin can alter the metabolic AMP-activated protein kinase (AMPK) pathway, and can affect NRF2 activation [16]. The anti-inflammatory effect of catalpol was associated with an inhibition of neutrophil migration in a model of airway or paw inflammation [17]. Altogether, these studies suggested that the extracts of V. austriaca may exert potent biological activities, as they contain biologically active compounds, such as iridoid and phenolic glycosides.
We found that the major compound which comprised around 30% of the dry weight of the extracts was arbutin (in excess of 0.3 mg arbutin/mg extract). Although arbutin is widely distributed in the plant kingdom, its occurrence in Veronica spp. is not typical. To date, arbutin has been isolated and purified from V. turrilliana [18] and a commercial hybrid of Veronica [6], but has not been reported in V. austriaca [9]. The significant amount of arbutin in V. austriaca L. collected from two different habitats suggests that it was not produced coincidentally, due to different environmental conditions, for example. Arbutin-containing extracts of some plant species have, for centuries, been used in phytotherapy against urinary tract infection and skin hyperpigmentation, and have shown antioxidant, anti-inflammatory and antitumor activity [19]. Herein we observed for the first time that arbutin can affect neutrophil survival. The compound at concentrations ranging from 4 to 15 µg/mL (14.7 to 55.1 µM) increased neutrophil survival, while at higher concentrations from 61-245 µg/mL (224 to 900 µM), it decreased neutrophil density and vitality. Other studies in melanocytes have shown that arbutin is a potent tyrosinase inhibitor, with an IC 50 value of 1.09 mM [20]. In A375 human malignant melanoma cells, arbutin at high concentration (1 mM) up-regulated 88 genes and down-regulated 236 genes, including genes (AKT1, CLECSF7, FGFR3, and LRP6) controlling cell cycle progression [21], hence, suggesting that it may act similarly on neutrophils. However, the effect on cell vitality of arbutin at low concentrations was unexpected because the compound was inactive at concentrations ranging from 5 to 20 µM in several screening libraries for anti-cancer drugs, agonists of the farnesoid-X-receptor (FXR) signaling pathway and cell viability, and antagonists of the sonic hedgehog signaling (Shh) pathway [2]. By contrast, the Vau-1 extract decreased neutrophil survival in a dose-dependent manner. Similar cytotoxic activity of Veronica species has been shown in various cell lines, such as Hep-2 (human epidermoid carcinoma), RD (human rhabdomyosarcoma), and L-20B (transgenic murine L-cells), where IC 50 values were above 150 µM [22]. At low concentration, the methanolic extract of the aerial parts of Veronica spp. demonstrated cytotoxicity against colon (HF-6) and prostate (PC-3) human cancer cell lines [23]. Most of the authors linked the presence of iridoid and phenolic compounds in Veronica extracts (aucubin, catalpol, catalpol derivatives and verbascoside, among others) with its cytotoxic effect. Indeed, in our previous study, we found that verbascoside is an inhibitor of TLR2 and TLR4-mediated apoptosis of BM-derived neutrophils, although less potent than isoverbascoside. The enumeration of live cells in total BM cultures showed that, at high concentration (>160 µM), the compound reduced neutrophil numbers with 20-26%, and increased apoptosis up to 60% in death-sensitive conditions (Annexin V staining) [15]. Thus, we cannot exclude the possibility that the reduced vitality of neutrophils following treatment with the Vau-1 extract may be due to the presence of verbascoside.
Neutrophils produce pro-inflammatory cytokines like TNF-α and IFN-γ, that can be pre-stored in the cytoplasm and secreted upon stimulation (priming or activation). We observed that Vau-1 and Vau-2 inhibited strongly the intracellular level of IFN-γ at low concentrations, while arbutin had an opposite effect by inducing a significant increase in IFN-γ production. We suggest that the action of arbutin on the cytokines might be masked by other molecules present in the Vau-1 and Vau-2 extracts. It is also possible that the compounds interfered selectively with PMA/Yon-induced calcium signaling after stimulation for 4 h. The latter hypothesis is based on a study demonstrating that aqueous-acetone extracts of Veronica spp. (V. teucrium, V. jacquinii, and V. urticifola) have an inhibitory effect on calcium ionophore-stimulated platelets, resulting in the suppressed release of pro-inflammatory enzymes 12-lipoxygenase (12-LOX) and cyclooxygenase-1 (COX-1) and pro-inflammatory mediators, such as interleukin-8 (IL-8) and E-selectin [11].
Interestingly, while Vau-2 and arbutin inhibited TNF-α production, Vau-1 induced a dose-dependent increase in the cytokine's intracellular level. The presence of verbascoside might contribute to this effect. Several studies have shown that verbascoside can alter, via TAK1, the degree of NF-kB activation, phosphorylation and nuclear translocation [24]. This compound inhibited LPS-stimulated TNF-α release by intraperitoneal macrophages [25], the level of TNF-α in bronchoalveolar lavage fluid (BALF) in LPS-induced lung inflammation [26], and in liver in LPS-induced immunological liver injury [27]. We previously found that verbascoside at low concentrations (<16 µM) increased LPS-induced TNF-α production and frequency of TNF-α + BM-derived neutrophils via inhibition of p38 phosphorylation [15]. Similarly, here, we found that the Vau-1 extract, which contains less amounts of verbascoside than Vau-2, elevated the TNF-α intracellular level. It is also possible that aucubin, as well as the combination of different compounds in the extract, can act synergistically or antagonistically on TNF-α production. Another reason might be that TNF-α is already pre-stored in mature neutrophils and less cells synthesize TNF-α de novo. Thus, the effect of the compounds may require strong cell activation to inhibit cytokine production at the mRNA level [28]. With regard to the pure compound, arbutin, we found the inhibited generation of TNF-α in agreement with other reports using LPS-stimulated murine BV2 cells, where this effect was mediated by inhibited nuclear translocation and the transcriptional activity of NF-κB [29].
Neutrophils undergo differentiation and maturation in BM. The process of granulopoiesis is precisely regulated by mediators released in the periphery cytokines and chemokines, and by mediators secreted in the environment of the bone marrow niches. During differentiation, neutrophils acquire various types of secretory vesicles and granules, as well as store cytokines in the cytoplasm. Indeed, here we observe that the transition from Ly6G low to Ly6G hi neutrophils was associated with the loss of single IFN-γ producing cells and the progressive expansion of double TNF-α/IFN-γ positive neutrophils. At the terminal differentiation stage, neutrophils are mature and ready to express integrin and chemokine receptors in response to signals from the periphery, and in order to be mobilized into the blood. We found that Vau-1/Vau-2 and arbutin increased in a dose-dependent manner the immature transient pool defined as SSC hi Ly6G low , but they failed to dramatically alter the pool of mature cells. However, both Vau-2 and arbutin, but not Vau-1, expanded the immature pool. In addition, we observed that the Veronica extracts and arbutin changed IFN-γ and TNF-α cytoplasmic levels in cells at a particular stage of neutrophil maturation. Vau-1 at low concentration increased the proportion of TNF-α+IFN-γ low cells within the immature pool, while Vau-2 and arbutin elevated the proportion of those double producers in the immature transient pools, likely leading to the expansion of the proportion of TNF-α+IFN-γ hi neutrophils with a mature phenotype. It has been shown that, during differentiation and maturation, cells require the presence of reactive oxygen species (ROS), a particular level of nitric oxide (NO) and functional antioxidant enzymes and mechanisms. Indeed, ROS are considered as intracellular messengers that interact with specific receptors, signalling pathways including protein kinases, phosphatases, and transcription factors [30]. Recent studies showed that methanolic, ethanolic and aqueous extracts from a variety of Veronica species possess powerful radical scavenging activity against superoxide (SO), and nitric oxide (NO) radicals [31,32]. Hence, we further speculate that Vau-2 and arbutin affected maturation and frequency of TNF-α+IFN-γ hi mature neutrophils, by influencing the cellular antioxidant/oxidant mechanism, and perhaps by sustaining the activation of particular pathways and transcription factors. Arbutin may also be able to affect cell differentiation because it suppresses the tyrosinase elevation at the late stage of melanocyte differentiation [33]. It inhibits tyrosinase, an enzyme which can alter protein function by oxidation of the tyrosine residues, and starts further non-enzymatic reactions with other tyrosines, cysteines, lysines or histidines, at the same or a different molecule, resulting in inter-and/or intramolecular cross-links [34]. However, further investigations are needed to confirm the effect of arbutin and Veronica extracts on neutrophil differentiation and, therefore, to justify their application in acute inflammatory conditions or severe immunosuppression-both conditions with abrogated granulopoiesis and neutrophil maturation.

Plant Material
Aerial parts from V. austriaca were collected in their flowering period in July

Extraction Protocol and NMR Analysis
Ground aerial parts of V. austriaca were air dried and 50 mg of each of 4 biological replicates were homogenized with equal amounts of CD 3 OD (0.75 mL) and D 2 O (0.75 mL KH 2 PO 4 buffer, pH 6.0), containing 0.005% (w/v) trimethylsilyl propanoic acid (TSPA-d 4 ). After 20 min ultrasonication (35 kHz; UCI-50Raypa ® R. Espinar S.L., Barcelona, Spain), samples were centrifuged (14,000× g, 20 min), then, the supernatants were transferred to thin glass walled tubes (NMR tube; 5 mm) and finally analyzed at the NMR spectrometer as described in [35]. Briefly, proton ( 1 H) as well as 2D NMR spectra (J-resolved, COSY, HSQC), were recorded at 25 • C on an AVII+ 600 spectrometer (Bruker, Karlsruhe, Germany), operating at a proton NMR frequency of 600.01 MHz [35]. Deuterated methanol was used for internal lock. The resulting 1 H NMR spectra for each sample was further processed by referencing to the internal standard TSPA, phased and baseline corrected, by running TopSpin software (3.6.5, Bruker BioSpin Group). CD 3 OD and D 2 O from Deutero GmbH (Kastellaun, Germany) were used in the experiments.

Extraction Protocol for HPLC Analysis and Biological Tests
Air dried ground plant material (5 g) of each sample was homogenized with 125 mL of absolute CH 3 OH. After 60 min ultrasonication, the extract was filtered and evaporated to dryness in vacuo. The yields of the extract were 0.78 g (Vau-1) and 0.83 g (Vau-2).

HPLC Analysis of Arbutin Content in V. austriaca Extracts
The HPLC analyses were performed according to Rathi et al., 2019 [36], with some modifications, as described below. The quantification was conducted on Waters HPLC system, consisting of Waters 1525 Binary pumps with Waters 2487 Dual λ Absorbance Detector (Waters, Milford, MA, USA), equipped with a reverse-phase Kinetex ® C 18 , 100 Å (150 × 4.6 mm, 5 µm) core-shell column (Phenomenex, Torrance, CA, USA), operating at 26 • C. The mobile phase consisted of water (solvent A) and acetonitrile (solvent B), with a flow rate 0.5 mL/min and gradient elution as follows: 0-2 min 99% A; 2-6 min decreased to 40% A; 6-15 min gradually increased back to 99% A arbutin was detected at 285 nm. Six solutions of the arbutin, with different concentrations ranging from 25 to 200 µg/mL dissolved in methanol, were used to construct a linear calibration curve.

Animal Studies
Balb/c mice were purchased from Charles River Laboratories (Wilmington, MA, USA) and then bred in the Experimental Animals Facility at the Institute of Microbiology (Sofia, Bulgaria). Mice (female, 8-week-old, 18-20 g weight) were kept under standard conditions, fed with a laboratory diet (29% protein, 13% fat, 56% carbohydrates) and given water ad libitum, as described previously [13].

Statistical Analyses
A statistical analysis was accomplished by using InStat 3.0 (GraphPad Software, La Jolla, CA, USA). Data were presented as mean ± standard deviation (SD). The differences in the mean values between groups were analyzed with the two-tailed Student t-test. Differences were considered significant when p < 0.05.

Conclusions
In the present study, we observed, for the first time, that the methanolic extracts of V. austriaca collected from two different habitats contained arbutin as a major constituent of the secondary metabolites pool. The Vau-1 extract inhibited, in a dose-dependent manner, neutrophil vitality, increased the intracellular level of TNF-α, and the proportion of TNF-α+IFN-γ low cells within the immature pool, while Vau-2 extract inhibited IFN-γ and TNF-α production, but expanded the frequency of mature TNF-α+IFN-γ hi within the BM pool. Arbutin, at low concentrations, increased neutrophil survival, elevated the proportion of double TNF-α/IFN-γ producers in the immature transient pool, and increased the frequency of mature TNF-α+IFN-γ hi cells in the BM population. As Veronica plant species are widely used in traditional medicine, the identification of the plant metabolites contributing to their biological activity is of significant importance, and can eventually lead to the formulation of pharmaceutically relevant molecules.