Gas Chromatography with Flame Ionization Detector (GC-FID)

Quantitative GC analyses were carried out using a Hewlett- Packard 5890 Series II GC apparatus equipped with FID and a non-polar DB-5 fused silica capillary column (30 m x 0.25 mm x 0.25 μm film thickness) (J & W Scientific). The oven temperature was programmed from 60 to 240 °C at a rate 3 °C min^-1 for integration purposes. Injector and detector temperatures were 260 °C. Hydrogen was used as the carrier gas at a flow rate of 1 L min^-1 and 30 p.s.i. inlet pressure in split mode (1:30). The injection volume was 0.5 µL of diluted solution (0.01) of oil in n-hexane. The amount of each compound was calculated from GC peak areas in the order of DB-5 column elution and expressed as a relative percentage of the total area of the chromatograms. Analyses were carried out in triplicate.

Gas Chromatography-Mass Spectrometry Analysis (GC-MS)

The GC-MS analysis was carried out using a Varian 220-MS IT GC system with a mass selective detector, mass spectrometer in EI 70 eV with a scan interval of 0.5 s and fragments from 40 to 550 Da. fitted with the same column and temperature program as that for the GC experiments, with the following parameters: carrier gas = helium; flow rate = 1 mL min^-1; split mode (1:30); injected volume = 1 µL of diluted solution (0.01) of oil in n-hexane. Identification of the components was based on GC retention indices with reference to a homologous series of C8-C40 n-alkanes calculated using the Van den Dool and Kratz equation and by computer matching against the mass spectral library of the GC/MS data system (NIST 98 and WILEY) and co-injection with authentic standards as well as other published mass spectra. Area percentages were obtained electronically from the GC-FID response without the use of an internal standard or correction factors.

The bioassay consisted of the treatments: distilled water (negative control); Roundup® (Glyphosate isopropylamine salt) as positive control (at 1, 10 and 20 µL mL^-1 distilled water), essential oil (1, 10 and 20 µL mL^-1) and distilled water + Tween® 20 (polyoxyethylene [20] sorbitan monolaurate), at 5% (v/v). Treatments were distributed in a completely randomized design with 30 replicates each. The synthetic herbicide Roundup® is recommended to apply at 2 - 4 L ha^-1 (10 - 20 µL mL^-1). E. globulus essential oil effects were evaluated on germination and radicle length in the previously mentioned species. The experiment was repeated during two periods.

One seed of each species was placed onto non-toxic paper (Whatman No. 1 filter paper) about 4 cm from the top edge, leaving a 3 - 4 cm gap on the sides; each seed was then covered with a second sheet of moist paper. The sheet of paper was rolled and placed upright in a deep-bottomed test tube, and 5 mL of each treatment to be evaluated was added (Gold, 2009). The test tubes were sealed with a paraffin film and placed in a growth chamber set at 22 °C ± 1 °C, 50% ± 1% RH, for 16:8 h light/dark cycles. After 7 days, the number of germinated seeds and radicle lengths were measured by unrolling the paper carefully to avoid tearing it or damaging the radicles of the young seedlings. Mortality was calculated based on seeds not germinated. The percentage of Radicle Growth Inhibition (RGI) was calculated by the difference between growth of Treatments (T) and the negative Control (C) using the following equation:

RGI = (C -T)/C) × 100

All measured variables were tested for normality (Shapiro- Wilk W test) and homoscedasticity (Bartlett test). Kruskal- Wallis non-parametric analysis of variance was utilized when data violated the normality assumption and could not be corrected employing a transformation. One-way analysis of variance (ANOVA) and Tukey test (p < 0.05) were also performed to detect possible differences among the treatments. Linear regression analysis was used for quantifying the relationship between the essential oil concentration and seed germination and radicle length of the three species studied. The 50% Lethal Concentration (LC50) (concentration causing 50% mortality compared to the control) was calculated for each treatment by probit analysis, based on mortality (seeds not germinated) obtained at each concentration of the samples (SAS, 2012).

GC-FID and GC-MS analysis demonstrated that the oil is mainly a mixture of monoterpenes, sesquiterpenes and alcohols. Dominant among the constituents were the monoterpenes, constituting approximately 97.41% of the oil (Table I). Major monoterpenes included the oxygenated monoterpene 1,8-cineole (77.91%), p-cymene (10.19%) and the monoterpene hydrocarbon α-pinene (4.33%).

Table I

Compound	RIa	LRb	%	Method of identification
α-thujene	924	923	0,15 ± 0,00	RI, MS
α-pinene	932	930	4,33 ± 0,09	RI, MS,CI
sabinene	969	970	0,52 ± 0,00	RI, MS
β-pinene	975	976	1,97 ± 0,03	RI, MS, CI
myrcene	988	989	0,62 ± 0,04	RI, MS
α-phellandrene	1002	1003	0,18 ± 0,00	RI, MS
p-cymene	1020	1021	10,19 ± 0,37	RI, MS, CI
1,8-cineole	1026	1025	77,91 ± 1,14	RI, MS, CI
Z-β-ocimene	1032	1034	0,72 ± 0,01	RI, MS
γ-terpinolene	1054	1055	0,17 ± 0,00	RI, MS
terpinolene	1086	1089	0,65 ± 0,00	RI, MS, CI
Total			97,41 ± 1,09

^a

Retention indices calculated from retention times in relation to those of a series of n-alkanes on a 30m DB-5 capillary column. bLinear retention indices from the literature. RI = retention index, MS = mass spectrum, CI = co-injection with authentic standards.

Previous studies have found that major constituents of Eucalyptus spp. essential oils are 1,8-cineole (49.07 to 83.59%) and α-pinene (1.27-26.35%) (Zhang, An, Wu, Liu & Stanton, 2012; Sebei, Sakouhi, Herchi, Khouja & Boukhchina, 2015; Morsi & Abdelmigid, 2016). 1,8-cineole is the most important compound of the genus Eucalyptus and is largely responsible for a variety of its pesticidal properties (Batish, Singh, Kohli & Kaur, 2008) and its antibacterial activity against Gram-positive and Gram- bacteria, related with the presence of this oxygenated compound (Ben et al., 2011). Different composition may be due to genetic variability between species of E. globulus, environmental factors and extraction procedures.

ANOVA values indicated that differences between treatments and between species were significant (p < 0.01) and species × treatments (p < 0.01). There was not a significant difference for each species between the treatments comprising distilled water + Tween® and distilled water.

L. sativa exhibited the highest number of seeds germinated in all treatments. With regard to A. fatua and A. hybridus, Glyphosate treatment at 20 μL mL-1 completely inhibited germination (Table II).

Table II

Treatment	Species	Average germination
Distilled water	L. sativa A. fatua A. hybridus	1.000 ± 0.00 a 0.633 ± 0.08 efgh 0.867 ± 0.06 abcd
E. globulus
1 µL mL-1	L. sativa A. fatua A. hybridus	0.933 ± 0.04 ab 0.733 ± 0.08 cdefg 0.433 ± 0.09 i
10 µL mL-1	L. sativa A. fatua A. hybridus	0.800 ± 0.07 bcde 0.533 ± 0.0 9 hi 0.233 ± 0.07 j
20 µL mL-1	L. sativa A. fatua A. hybridus	0.633 ± 0.08 efgh 0.433 ± 0.09 i 0.200 ± 0.07 jk
Glyphosate
1 µL mL-1	L. sativa A. fatua A. hybridus	0.600 ± 0.09 fghi 0.567 ± 0.07 ghi 0.033 ± 0.03 kl
10 µL mL-1	L. sativa A. fatua A. hybridus	0.767 ± 0.07 bcdef 0.200 ± 0.07 jk 0.000 ± 0.07 l
20 µL mL-1	L. sativa A. fatua A. hybridus	0.167 ± 0.06 jkl 0.000 ± 0.07 l 0.000 ± 0.00 l
Distilled water +Tween® 20	L. sativa A. fatua A. hybridus	1.000 ± 0.00 a 0.700 ± 0.08 defch 0.900 ± 0.05 abc

The values refer to means ± S.E. For a given measurement, mean values with the same letter in the same column are not significantly different (p < 0.05 Tukey test).

Essential oil treatments showed the highest average of germination in L. sativa, which has a middle-sized seed compared with A. fatua and A. hybridus.Batish et al. (2008) concluded that eucalyptus oil exhibits species-specific toxicity and that the toxic effect was more noticeable in small-seeded crops such as Amaranthus viridis compared with large-seeded Raphanus sativus. Treatment with essential oil at concentrations of 10 and 20 μL mL^-1 caused greatest inhibition (p < 0.01) of germination in A. hybridus, overcoming the inhibitory effect achieved in the species L. sativa and A. fatua (Table II). This indicates that A. hybridus is more sensitive to be inhibited in its germination by the essential oil in comparison with the other two species studied. This shows that A. hybridus is a species in which the phytotoxicity caused by the E. globulus essential oil occurs at low doses. Regarding the E. globulus essential oil LC50, the lowest concentration was A. hybridus (< 1 μL mL^-1), followed by the dose for A. fatua (11.78 μL mL^-1) and L. sativa (> 20 μL mL^-1).

Inhibition of germination caused by E. globulus essential oils and Glyphosate treatments on three seeds were highest for A. hybridus, followed by A. fatua and then by L. sativa. However, the inhibition of germination caused by Glyphosate was higher than the effect of essential oil on all concentrations. Batish, Setia, Singh & Kohli (2004) found that E. globulus essential oils caused inhibition in the germination and radicle growth of A. viridis, Cassia occidentalis, E. crus-galli, R. sativus, Triticum aestivum and Zea mays. This result suggests that E. globulus essential oils could possess a greater capacity to severely affect the germination, although the different plants species involved prevent direct comparisons. This could be related with volatile monoterpenes, eucalyptol (1,8-cineole), and camphor, which can induce nuclear abnormalities and increasing vacuole numbers (Nishida, Tamotsu, Nagata, Saito & Sakai, 2005; Pawlowski, Kaltchuk-Santos, Zini, Caramao & Soares, 2012). Batish et al. (2007) speculated that eucalypt oil inhibits the mitotic activity of growing cells; however, the mechanism of the inhibitory effect remains unknown. On the other hand, α-pinene, the third major component of E. globulus essential oil acts under at least two mechanisms: uncoupling of oxidative phosphorylation, and inhibition of electron transfer (Nishida et al., 2005) and allelochemicals might inhibit seed germination by suppressing the synthesis of gibberellins and indole acetic acid (Chu et al., 2014).

A partial correlation analysis between seed germination and oil concentration (positive correlation) suggested that the phytotoxicity effects of the essential oil of E. globulus might cause inhibition of seed germination.

The three species demonstrated highest radicle growth (p < 0.05) in treatments with distilled water and distilled water + Tween®, probably showing that division and/or the cell growth level radicle were normal (Table III).

Table III

Treatment	Species	Average radicle length (mm)	RGI (%)
Distilled water
	L. sativa A. fatua A. hybridus	3.000 ± 0.14 d 8.671 ± 0.30 b 2.436 ± 0.19 ef	0 0 0
E. globulus
1 µL mL-1	L. sativa A. fatua A. hybridus	0.500 ± 0.08 g 0.425 ± 0.04 h 0.177 ± 0.04 h	85.3 95.1 92.7
10 µL mL-1	L. sativa A. fatua A. hybridus	0.284 ± 0.02 h 0.189 ± 0.03 h 0.157 ± 0.02 h	90.5 97.8 93.6
20 µL mL-1	L. sativa A. fatua A. hybridus	0.177 ± 0.02 h 0.047 ± 0.02 h 0.233 ± 0.03 h	94.1 99.5 90.4
Glyphosate
1 µL mL-1	L. sativa A. fatua A. hybridus	1.313 ± 0.12 f 3.920 ± 0.65 c 0.200 ± 0.00 h	56.2 54.8 91.8
10 µL mL-1	L. sativa A. fatua A. hybridus	2.304 ± 0.27 f 0.267 ± 0.12 h 0.000 ± 0.00 i	23.2 96.9 100.0
20 µL mL-1	L. sativa A. fatua A. hybridus	0.290 ± 0.06 h 0.000 ± 0.00 i 0.000 ± 0.00 i	90.3 100.0 100.0
Distilled water + Tween®
	L. sativa A. fatua A. hybridus	3.101 ± 1.4 d 9.861 ± 0.41 a 2.501 ± 0.18 e	- - -

The values refer to means ± S.E. For a given measurement, mean values with the same letter in the same column are not significantly different (p < 0.05 Tukey test).

When comparing the effect of radicle growth among the three species evaluated, the species A. fatua exhibited highest growth (p < 0.05) in distilled water + Tween®, followed by the treatment with water (p < 0.01). L. sativa had highest radicle lenght at the 1 µL mL^-1 concentration of the essential oil compared with the other two species. No difference was observed in radicle growth and RGI among two other species at 10 and 20 μL mL^-1 concentrations of the essential oil. The RGI caused by the essential oil concentrations of 10 and 20 µL mL^-1 was statistically equal (p < 0.05) to that achieved by the Glyphosate concentration at 10 µL mL^-1 on A. fatua and 1 µL mL^-1 on A. hybridus and 20 µL mL^-1 on L. sativa. The phytotoxic effect of the essential oil is equal to that of Glyphosate, although at higher concentrations, shows that it has the potential to cause effects similar to those of the synthetic herbicide, but most likely without its adverse effects on the environment. A. fatua showed significantly higher radicle lenght (p < 0.05) at 1 µL mL^-1 concentration of Glyphosate compared with other seeds treated with different concentrations of Glyphosate and essential oil. This suggests that A. fatua resists the toxic effect of Glyphosate at its lowest concentration, exhibiting some type of resistance. This species has been reported as a herbicide-resistant weed (Heap, 2018). The increase in the concentration of E. globulus essential oil did not cause a significant increase in RGI. However, Rassaeifar et al. (2013), in working with E. globulus essential oil on seedling establishment of Amaranthus blitoides and Cynodon dactylon, found that when the concentration of the essential oil increases (0.00005 to 0.005 µL mL^-1), radicle length decreases.

Regression analysis showed that germination of A. hybridus had a positive linear correlation with increasing concentrations of essential oil (R2 = 0.83; p < 0.05). Likewise, previous investigation has shown that the essential oil of eucalyptus species increases the inhibition of germination by increasing the concentration of the essential oil (Rassaeifar et al., 2013; Vishwakarma & Mittal, 2014).

Treatments with essential oil demonstrated significantly greater inhibition (p ≤ 0.05) on germination and radicle growth when compared with control (Table II and III). At the highest concentration of the essential oil (20 μL mL^-1), the inhibition on germination was statistically the same as that caused by Glyphosate at (1 and 10 μL mL^-1). Once again, we can suppose that the use of the essential oil of E. globulus instead of the synthetic herbicide Glyphosate could have similar phytotoxic effects but less negative consequences on the environment. Increasing concentrations of the essential oil inhibited germination in a different manner, but not radicle growth. Batish et al. (2007), studying the effect of the essential oil of E. citriodora on Phalaris minor, found that this caused a greater effect on seedling growth than on germination. Based on this, studying the effect of foliar application of essential oil of E. globulus on herbicide resistant species is a topic that could be analyzed in future research. In the present study, the magnitude of the phytotoxic effect was observed both in radicle growth and germination. The essential oil at its three concentrations possesses the same radicle growth inhibitory effect as Glyphosate at its lowest concentration (RGI, 90.4 and 91.8, respectively).

In the species L. sativa, treatments showed significantly higher germination (p ≤ 0.01) in distilled water and distilled water + Tween® and essential oil at its lowest concentration (1 μL mL^-1) compared with the other treatments.

Glyphosate demonstrated highest inhibition of germination at 20 μL mL^-1. Treatments with essential oil at the three concentrations (1, 10, and 20 μL mL^-1) exhibited inhibition on germination that was statistically equal to that of Glyphosate at 10 μL mL^-1; the essential oil at such concentrations is as effective in inhibiting germination as Glyphosate (Table II). Yamagushi, Gusman & Vestena (2011) and de Souza & Cardoso (2013) found that aqueous extracts of E. globulus at increasing concentrations (100,000-1,000,000 μL mL^-1) occasioned the increase of inhibition of germination and of RGI of L. sativa. However, in the present study the concentrations were lower, and RGI was statistically equal at increasing concentrations.

Treatments with the essential oil (10 and 20 μL mL^-1) showed RGI that was statistically equal to that of Glyphosate at 20 μL mL^-1 and significantly higher than the controls (distilled water and distilled water + Tween®). On increasing the concentration of essential oil, inhibition of germination is increased. However, the RGI is statistically equal and is not affected by the increase in oil concentration from 10 to 20 μL mL^-1. Apparently, the emerged radicle of L. sativa exhibited Glyphosate resistance at its concentrations 1 and 10 μL mL^-1.