The Inﬂuences of Genotype and Year on Some Biologically Active Compounds in Honeysuckle Berries

: Berries of three Romanian Lonicera caerulea cultivars ‘Cera’, ‘Kami’, and ‘Loni’, grown at the Research Institute for Fruit Growing, Pitesti, Romania, were analyzed between 2020 and 2022 in terms of chemical composition. The study aimed to determine the concentrations of some compounds with antioxidant activity, highlight the most valuable cultivar, encourage the consumption of honeysuckle berries, and indirectly stimulate growers’ interest in this little-known species in Romania. Some phenolic compounds—lycopene, β -carotene, and vitamin C—were quantiﬁed. As a result of the study, the ‘Loni’ cultivar’s high total phenolic content, ﬂavonoids, anthocyanins, vitamin C, lycopene, chlorogenic and neochlorogenic acids, catechin, and rutin are to be noted. ‘Cera’ cultivar had the highest cryptochlorogenic acid content, and ‘Kami’ summarized the highest carotenoid level. These characteristics indicated that the three honeysuckle cultivars’ berries could have multiple uses, from fresh consumption, as part of a diet focused on maintaining human health, to being used as raw materials in the para-pharmaceutical industry, to obtain food supplements. The novelty characteristics and the nutritional value of its berries highlighted by this study have indicated that honeysuckle can become a crop of interest and proﬁtability.


Introduction
Lonicera caerulea, a member of the Caprifoliaceae family, is originally from the Holarctic temperate zone.It is relatively little known, although it has a history with deep roots and written evidence dating back to the pre-Linnaean period, when Clusius, in 1583, provided a description and a picture of a honeysuckle with blue fruits [1].This gives the honeysuckle an advantage for those consumers looking for unexperienced pleasurable tastes.Nevertheless, the financial success of a lonicera plantation depends on the market's preferences, and new food's acceptance is linked to the consumers' education, income, taste, and previous experience [2].The honeysuckle produces ovoid fruits that normally bulge in the center and are narrow at the ends, are dark blue, are covered by a cuticular wax layer, weigh 0.7-1.3g, and mature in stages [3].Their berries' sour taste is shaped mainly by the content of organic acids, and among the phenolic compounds, tannins, together with the iridoid glycosides and malic and citric acid esters, are responsible for the bitter sensation [4].Depending on the cultivar, the fruits can be sweet-sour, sweet-bitter, or sour-bitter.The honeysuckle is favored by its early presence on the market, as honeyberries are the first fruits to ripen, starting from the middle of May, depending on the cultivar and the climatic conditions [3].Through its composition, the honeysuckle is placed among the most valuable sources of antioxidants, and the species' great ability to adapt to different ecological conditions and their resistance to the very low temperatures of the winter period [3] reduce the risks of losing much of the fruit yield.In addition, under the conditions of a changing climate only in certain periods during the year and with a predisposition to climatic accidents in the spring [5][6][7], the productivity and even the survival of some species must be taken into account.To all this is added the fact that the involvement of honeysuckles in culture in different areas gives it the advantage of reaching consumers quickly, fresh, and without additional costs related to long-distance transport.In general, climate (along with pedological factors) is the determining factor in productivity increases or decreases and the quality of fruit agroecosystems.The relationship among the genetic background of fruit species, agricultural practices, and local environmental conditions represents production's quantitative and qualitative basis [8][9][10][11].Plants produce a wide range of secondary metabolites, and among them, phenolic compounds have been progressively synthesized during their evolution.It is appreciated that berries, including honeyberries, have higher contents of bioactive compounds.Many of these compounds have medical or socio-economic value, justifying the interest of the scientific world and the fruit industry [10][11][12][13][14][15].Lonicera caerulea shows high health potential and is a promising source of numerous bioactive compounds, mainly anthocyanins, phenolic acids, and flavonols [16].In several studies, phenolic compounds' protection against chronic diseases such as hypertension, diabetes, cardiovascular diseases, and atherosclerosis has been documented.Furthermore, phenolic compounds were shown to have beneficial effects on cognitive processes, ophthalmology conditions, and antibacterial activity, especially in kidney infections [12,13].Other reported properties of honeyberry fruit involve antimicrobial, anti-inflammatory, anti-atherosclerotic, and anticarcinogenic activities, which have been demonstrated in in vitro and some in vivo tests [4].Lonicera caerulea was introduced in Romania, at the Research Institute for Fruit Growing, Pitesti, Romania, in 1985.Three Romanian cultivars were obtained through free pollination, and two of them ('Loni' and 'Cera') were registered in 2004.The 'Kami' cultivar was registered seven years later, in 2011.Sumedrea et al. [3] described 'Loni' as a cultivar with medium-high vigor, erect and compact growth, and the ability to produce ovoid, dark-blue berries of 0.7-1 g, slightly covered in wax, with a sweet and sour taste.'Cera' produces fruits of about 0.9 g with an obovate shape.The cultivar has good resistance to frost, drought, diseases, and pests.'Kami' is a high-vigor cultivar with large fruits (1.0-1.3 g), a pleasant, almond-like taste, and resistance to drought and frost.Although it has very high adaptability to the unfavorable conditions of early spring (late frosts) or summer (dry periods), productivity-related data do not present honeysuckle as a very attractive crop.What could ensure the success of a lonicera plantation are the characteristics related to its berry biochemical composition that open opportunities for its exploitation.Starting from these considerations, this study was carried out over 2020-2022 to highlight the content of some bioactive compounds in berries of the three Lonicera caerulea cultivars bred in Romania, i.e., 'Cera', 'Kami', and 'Loni'.Additionally, this paper's concern is to highlight the most valuable Romanian honeysuckle cultivar to encourage this species' cultivation in farms.

Experimental Site
The experiment was carried out between 2020 and 2022 in the Lonicera caerulea plantation of the Research Institute for Fruit Growing, Pitesti, Romania (R.I.F.G., 44 • 51 38 N 24 • 52 4 E, 285 m above sea level) on three Romanian honeysuckle cultivars ('Cera', 'Kami', and 'Loni').The plant material intended for planting was produced through in vitro micropropagation at the R.I.F.G.The establishment year of the honeysuckle plantation was 1991, the planting distances were 3 × 1.2 m, and the canopy shape was a bush.The soil presented the characteristics of the wet phreatic alluviosol protisols class and has a loamsandy granulometric composition, with a moderately acidic reaction.The data related to the climatic conditions in the period preceding the ripening of honeysuckle fruits, between

Sampling
Berries of the honeysuckle cultivars (samples of approximately 200 g per cultivar and harvest) were harvested at full maturity and visually assessed by the appearance of intense violet-blue coloring and the ease of the detachment of the berries from the plant.Three harvests were performed for each genotype, between the last week of May and the first week of June, and the samples were kept at −18 • C until the time of extract preparation (about 14 days).

Extraction Procedures
To obtain the methanolic extracts of the honeysuckle berries, 1 g frozen berry homogenate samples were treated with 10 mL of methanolic solution (8:2, v/v) and vortexed for 2 min at 3000 rpm.Afterward, the mixture was subjected to ultrasonic treatment (40 kHz) for 180 min, to extract polyphenols and 150 min, for the extraction of flavonoids.This step was followed by centrifugation for 15 min at 3000 rpm, and the resulting supernatant was filtered and used for analysis.For the HPLC analyses, the obtained methanolic extracts were subjected to evaporation at room temperature, until a constant mass was reached, resulting in a semi-solid consistency product, which was kept at −18 • C until further determination.Before the HPLC analysis, the semi-solid product was dissolved in concentrated methanol (99.8%) by ultrasonication at a temperature below 30 • C and filtered through Macherey-Nagel (MV) filters, with 0.20 µm pores.To obtain the methanolic extract necessary for the determination of monomeric anthocyanins, 1 g of frozen berry homogenate was added to a solution containing about 9 mL of methanol and 1 mL of 27% hydrochloric acid [17].The mixture was vortexed for 2 min at 3000 rpm and kept in the dark for 10 min, and then filtered.The filtrate was later used for the determination of monomeric anthocyanins.To obtain the aqueous extracts necessary for total tannin content determination, 1 g of frozen berry homogenate was treated with 10 mL of distilled water and subjected to vortex (2 min, 3000 rpm), followed by ultrasonication (30 min, 80 • C).The mixture was subsequently centrifuged (15 min, 3000 rpm), and the supernatant was used further for analyses.To obtain the extract necessary to determine the vitamin C content, a cold extraction was performed: 1 g of frozen berry homogenate was treated with 10 mL of 1% hydrochloric acid solution and subjected to vortex treatment for 2 min at 3000 rpm.After 10 min in the dark, the solution was filtered and used for determination.All extraction procedures were performed at a low temperature (4 • C, maintained by adding ice) and low light.For carotenoid extracts, a berry sample (1 g frozen berry homogenate) was added to a mixture of 25 mL of solvents (hexane: ethanol: acetone in a volume ratio of 2:1:1).The mixture was stirred for 30 min at 1500 rpm, 10 mL of distilled water was added, and stirring was continued for another 10 min.Afterward, the extraction was carried out by manual stirring every few hours during 72 h of rest in the dark at room temperature.

Determination of Total Phenolic Content
The determination of total content of phenolic compounds (TPC) was carried out by the spectrophotometric method, according to the methodology proposed by Cosmulescu et al. [18].The principle of the method is based on the formation of a blue compound, as a result of the reaction carried out in an alkaline environment between phospho-tungstic acid and polyphenols.The reaction mixture consisted of 0.5 mL of honeysuckle methanolic extract added to a 10 mL flask containing 7 mL of distilled water and 0.5 mL of Folin-Ciocalteu reagent.After 5 min of resting, 2 mL of 10% sodium carbonate solution was added.After another 60 min, the absorbance of the samples was measured and the concentration of polyphenols was estimated.Gallic acid monohydrate was used as a standard for calibration, and the total polyphenol content was expressed in mg gallic acid equivalent (GAE) 100 g −1 fresh weight (FW).

Determination of Total Tannin Content
To determine the total tannin content (TTC) in honeysuckle berries, the methodology proposed by Giura et al. [19] was followed.In a 10 mL flask containing honeysuckle aqueous extract and Folin-Ciocalteu reagent, a 10% sodium carbonate solution was added.Absorbances were read after 60 min at 760 nm wavelength.For calibration, gallic acid monohydrate was used as the standard compound.The total concentration of tannins was calculated and expressed in mg GAE 100 g −1 fresh weight (FW).

Determination of Total Flavonoid Content
For the quantification of flavonoids, the methodology proposed by AL-Ghudani and Hossain [20] was used.The principle of the method is based on the formation of a yelloworange compound through the reaction of flavonoids and aluminum chloride.To the initial reaction mixture, consisting of 1 mL of methanolic extract in a 10 mL volumetric flask containing 4 mL of distilled water and 0.3 mL of 5% sodium nitrite, after 5 min in the absence of light, 0.3 mL of 10% aluminum chloride was added.After standing for 5 min, a 2 mL solution of 1 M sodium hydroxide was added, and the sample was diluted with distilled water to a final volume of 10 mL.The absorbance of the solution was measured at 510 nm.For calibration, catechin hydrate was used as a standard, and the total flavonoid content (TFC) of the samples was expressed in mg catechin equivalent (CE) 100 g −1 fresh weight (FW).

Determination of Total Monomeric Anthocyanin Content
To determine the total monomeric anthocyanin content (TAC), a protocol similar to Anggraini et al. [17] was followed.The differential pH method is based on the reversible change in the color of a solution of monomeric anthocyanins with the change in pH.The difference in absorbance of the pigments at 520 nm is proportional to the concentration.The extracts were diluted with two different buffer solutions, potassium chloride 0.025 M (pH 1.0) and sodium acetate 0.4 M (pH 4.5), and the absorbance was read after 30 min at 520 and 700 nm.To calculate TAC, Anggraini et al.'s [17] formula was used.Cyanidin-3glucoside chloride was used as a standard, and the results are expressed in equivalents of cyanidin-3-glucoside (C3G) 100 g −1 fresh weight (FW).

Determination of Vitamin C Content
To quantify the ascorbic acid content, 10 mL of the extract (obtained according to the previously detailed procedure) was titrated with a 2,6-dichloroindophenol solution, according to Segura Campos et al. [21].The final titration point was considered to be the appearance of a pink-rose color persisting for more than 5 s.Ascorbic acid was used as a standard, and the results are expressed in mg 100 g −1 fresh weight (FW).

Determination of Lycopene and β-Carotene Content
The quantitative determination of carotenoids (lycopene and β-carotene) was carried out following the protocol proposed by Tudor-Radu et al. [22].The concentrations of two carotenoids in the specific extracts, expressed in mg 100 g −1 plant material, were calculated, using molar extinction coefficients of 184,900 M −1 cm −1 at 470 nm and 172,000 M −1 cm −1 at 503 nm for lycopene, and 108,427 M −1 cm −1 at 470 nm and 24,686 M −1 cm −1 at 503 nm for β-carotene in hexane.

Statistical Analysis
All analyses were performed in three replicates.The extracts necessary for the determination of TPC, TTC, TFC, TAC, and vitamin C were analyzed in triplicate in each year of the study.For carotenoids and the HPLC analysis (i.e., determination of the levels of chlorogenic, neochlorogenic, and cryptochlorogenic acids; and catechin, rutin, and isoquercetin), three analyzes were carried out, one for each year of study (essentially, the data from the three years of study were considered in this paper as three replicates), only considering the effect of the cultivar, but not the study year effect.Therefore, two-way ANOVA, followed by Duncan's multiple range test (p < 0.05), was used to study the cultivar and experimental year's effects on total phenolics, tannins, flavonoids, monomeric anthocyanins, and vitamin C. One-way ANOVA, followed by Duncan's multiple range test was used to study the cul-tivar's effects on lycopene; β-carotene; chlorogenic, neochlorogenic, and cryptochlorogenic acids; catechin; rutin; and isoquercetin (p < 0.05).
As shown in Table 4, except for TPC, the cultivar, climatic conditions of the experimental year, and their interaction had significant contributions to the variations in the fruit quality parameters (overall, the effect size-partial eta squared-ranged between 52.9 and 93.1%).TPC content of honeysuckle fruits varied significantly among cultivars (Table 4), from 790.20 mg GAE 100 g −1 ('Kami' cv.) to 1149.72 mg GAE 100 g −1 ('Loni' cv.), and the fluctuations recorded from one year to another (731.50mg GAE 100 g −1 in 2020 and 1156.94mg GAE 100 g −1 in 2022) had less significance (p = 0.077).There was a tendency for the three cultivars to accumulate higher contents of phenolic compounds between 2020 and 2022; therefore, the difference between 'Kami' and 'Cera' was higher in 2021 compared to 2020.Additionally, the most important increase in TPC during the study, 528.56 mg GAE 100 g −1 , was the one observed for the 'Cera' cv.On average, 'Cera' and 'Loni' cvs.presented similar levels of tannins (TTC) (Table 4), 584.03 and 609.04 mg GAE 100 g −1 , respectively.'Kami' lagged behind (412.18mg GAE 100 g −1 ).In general, tannins showed an increasing trend from 2020 (428.18mg GAE 100 g −1 ) to 2022 (645.84 mg GAE 100 g −1 ), and in this condition, the most important increase was recorded for 'Cera'-from 439.82 mg GAE 100 g −1 , in the first year, to 761.99 mg GAE 100 g −1 , in the last year.The most conservative was 'Kami', with a difference of only 111.52 mg GAE 100 g −1 .
The evolution of flavonoids (TFC) (Table 4) was similar to those of polyphenols and tannins: an increase from 2020 (416.60 mg CE 100 g −1 ) to 2022 (665.72 mg CE 100 g −1 ), although in this case, the differences between cultivars were higher.Thus, the 'Loni' cv.stood out for its high content of flavonoids (636.91 mg CE 100 g −1 ) and was followed by the 'Cera' cv., with an average of 534.85 mg CE 100 g −1 .The most important increase in TFC between the study years was recorded for 'Loni' cv., from 480.92 to 827.22 mg CE 100 g −1 , and the smallest difference was determined for 'Kami'.
As shown in Table 6, the correlation between lycopene and β-carotene was negative and insignificant, and the correlation established by β-carotene with catechin was negative and distinctly significant (r = −0.896**).In the case of lycopene, significant and positive correlations were established with chlorogenic acid (r = 0.876 **), neochlorogenic acid (r = 0.826 **), and catechin (r = 0.902 **); and the strongest positive correlation was with rutin (r = 0.967 ***).Except for cryptochlorogenic acid, which showed negative correlations with the other representatives of its group and with flavonoids, positive correlations with different degrees of significance were established between chlorogenic and neochlorogenic acids and the three flavonoids, of which, the strongest was between chlorogenic acid and rutin (r = 0.963 ***).Additionally, fruits with high antioxidant activity had high contents of neochlorogenic acid (r = 0.944 ***), chlorogenic acid (r = 0.827 **), isoquercetin (r = 0.828 **), and rutin (r = 0.727 *), and a low content of cryptochlorogenic acid (r = −0.989***).Regarding the levels of carotenoids of the three honeysuckle cultivars, it could be observed that the ratio between β-carotene and lycopene varied from 1.25 ('Loni' cv.) to 4.33 (Kami' cv.) (Table 7).The highest quantities of carotenoids (as a sum of lycopene and β-carotene) were recorded in the case of the 'Kami' cv., where the concentration of β-carotene was maximal (1.69 mg 100 g −1 ), and that of lycopene, which was medium (0.39 mg 100 g −1 ).'Loni' presented the highest concentration of lycopene (0.73 mg 100 g −1 ) but a minimum amount of β-carotene (0.91 mg 100 g −1 ).Furthermore, the highest variability in the concentration of the two carotenoids was observed for lycopene (42.76%,Table 5), almost 1.5 times higher compared to β-carotene.

Discussions
The results obtained in this study regarding TPC are higher than those reported by Senica et al. [23] for L. edulis (1612.61mg GAE 100 g −1 DW), and those from the more recent study of Česonien ė et al. [24]: 364-784.5 mg GAE 100 g −1 .More similar values were found by Chaovanalikit et al. [25], 427-1140 mg GAE 100 g −1 fresh fruit, and by Sic Žlabur et al. [26] (6.209 g GAE 100 g −1 DW).Moreover, less TPC variation compared to our study was reported by Shevchuk et al. [27]: from 444 to 1000 mg GAE 100 g −1 .The authors highlighted the stronger effect of the cultivar compared to the variability due to the year.This is contrary to our results, in which the effect of the cultivar was distinctly significant, and that of the year varied significantly (data also supported by partial eta squared values).
Tannins are found in large quantities in unripe fruits and are responsible for generating the astringent taste [16] resulting from the interaction between salivary proteins and (hydrolyzable) tannins [28].Regarding TTC, in a previous study, Mladin et al. [29] determined similar levels of tanned substances in honeysuckle berries, 0.192-0.429%GAE, but with higher variation limits.However, the authors reported that the tannin concentrations in the 'Loni' and 'Cera' cultivars were approximately two time lower as those found in the present study (0.254 and 0.301% GAE), which highlights the influences of the year and the climatic conditions.
The total flavonoid content was close to those reported by Rupasinghe et al. [30] and Raudon ė et al. [31].Rupasinghe et al. [30] reported for the berries of three honeysuckle cultivars ('Borealis', 'Indigo Gem', and 'Tundra') flavonoid concentrations of 594.43-699.29 mg quercetin equivalents (QE) 100 g −1 [30].Flavonoid concentrations between 20,000 µg g −1 DW and almost 50,000 µg g −1 DW for the cultivars 'Amphora', 'Indigo Gem', 'Leningradskij', 'Nimfa', and 'Tundra'-and also very low concentrations, below 10,000 µg g −1 DW in three other cultivars-were reported by Raudon ė et al. [31].Šic Žlabur et al. [26] presented similar flavonoid concentrations (2.825 g GAE 100 g −1 ) in fresh honeysuckle berries.Regarding the total content of monomeric anthocyanins, the study carried out by Auzanneau et al. [32] reported total anthocyanin levels of 8.4 and 41.1 mg C3G g −1 DW, a range in which the values found in our study fell, but without going down to the lower limit mentioned by authors.High variations in the anthocyanin content were found by Raudon ė et al. [31], in a study performed on eight cultivars of honeysuckle.The anthocyanin contents found in our study are comparable to those found by Raudon ė et al. [31], except for three cultivars mentioned by the authors, where low concentrations of phenolic compounds (below 10,000 µg g −1 DW) accounted for their low contents of anthocyanins.A lower anthocyanin content (116 and 593 mg 100 g −1 FW) in blue honeysuckle than that found in the present study was reported by Chaovanalikit al. [25].Other studies referred to anthocyanin contents similar to those reported by Raudon ė et al. [31], between 14.3 and 65 mg C3G g −1 DW [30,[33][34][35][36], a range of values that also includes the concentrations of the 'Cera', 'Kami', and 'Loni' cultivars.A recent analysis of 61 honeysuckle genotypes conducted by Fan et al. [37] indicated very high variability of the anthocyanin content of between 158.44 and 1751.44 mg 100 g −1 .De Silva and Rupasinghe [38] cited authors who reported anthocyanin contents varying between 68 and 649 mg 100 g −1 and found TAC varying between 39.2 and 294 mg C3G 100 g −1 in four haskap cultivars, depending on harvest dates.The vitamin C content in honeysuckle berries generally has lower values compared with our results, as reported by Jurikova et al. [39].
Comparing some cultivars and selections of honeysuckle with Morus nigra, Prunus tomentosa, and Amelanchier berries in terms of vitamin C (reported as mg 100 g −1 DW), Juríková et al. [39] found the following values: 67.66-186.61 for Lonicera, 40.46-96.80for Morus, 123.33 for Prunus, and 91.47-114.22 for Amelanchier.In further research, Juríková et al. [40] determined variations in the content of vitamin C in honeysuckle selections of between 9.71 and 46.67 mg 100 g −1 , an upper limit that approaches the minimum value reported in the present study.Other studies referred to slightly lower vitamin C contents of 14.55-53.58mg 100 g −1 [24], 17-25 mg 100 g −1 [30], or 17.7-35.6mg 100 g −1 [27].Shevchuk et al. [27] discussed the generally stronger variation in vitamin C content between cultivars rather than between study years, an aspect also found in our study.Šic Žlabur et al. [26] reported higher vitamin C concentrations of 5.348 g 100 g −1 in the cultivar 'Indigo Treat' grown in Croatia.Values closer to our results were also determined by Auzanneauet al. [32] in L. caerulea cultivars grown in Switzerland (1.78-4.21mg g −1 DW).
Other studies referred to the effects of abiotic factors and cited studies that indicated that, along with flavonoids, hydroxycinnamic acids have functions in the interaction of plants with environmental factors such as high-intensity light, extreme temperatures, heavy metals, water stress, etc. [8,10,16,44].In a recent study, Orsavová et al. [43] and Gołba et al. [16] reported significant variations in the profiles of compounds with antioxidant activity found in honeysuckle berries, depending on the cultivar and cultivation area, and also on the time of ripening.In our study, the strongest effects of the cultivar and the year of study were those related to TAC, followed by TFC and TPC, and the effect of the cultivar × year interaction was observed, especially in the variations of vitamin C content, followed by those of TAC.The greatest variabilities in TPC and TTC were recorded for 'Cera'.TFC varied intensively among the years of study for 'Loni', TAC for 'Cera' and 'Loni' cvs., and vitamin C for 'Kami' and 'Loni' cvs.

Conclusions
This study indicated that 100 g of fresh honeysuckle berries from 'Cera', 'Kami', and 'Loni' cultivars contained 519-1422 mg GAE of phenolics, 354.43-797.00mg GAE of tannins, 332.95-891.46mg CE of flavonoids, 289.89-853.84mg of C3G monomeric anthocyanins, 47.66-78.32mg of vitamin C, 0.27-0.76mg of lycopene, 0.83-1.73mg of β-carotene, 53.40-90.87mg of chlorogenic acid, 7.06-20.30mg of cryptochlorogenic acid, 4.17-10.46mg of neochlorogenic acid, 68.69-309.89mg of catechin, 14.40-30.52mg of rutin, and 1.95-3.08mg of isoquercetin.The consumption of fresh honeysuckle berries and the development of processed products, as a valuable source of health-promoting compounds, are of considerable interest and have received more and more attention in recent years due to their bioactive properties.The study of the influences of genotype and year on the phenolic content of three Romanian honeysuckle cultivars confirmed the usefulness of Lonicera caerulea species as a rich source of bioactive phenolic compounds with the potential to be used in food and pharmaceutical industries.For these reasons, greater efforts need to be made to provide new information on the characteristics of Lonicera caerulea species to promote its consumption and encourage its cultivation in farms.
* Means of three replicates per cultivar per year are presented.CV = variation coefficient.
*** The correlation is significant at the 0.001 level (two tailed).** The correlation is significant at the 0.01 level (two tailed).* Correlation is significant at the 0.05 level (two tailed). )