Blooming stages of Turkish hyssop (Hyssopus officinalis L.) affect essential oil composition

Abstract

Hyssop (Hyssopus officinalis L.), native to the Caucasus, North Western Iran, Turkish North Eastern Black Sea region, and Southern Anatolia, is a highly valued medicinal plant. The experiment was conducted to find the effect of harvesting at different blooming stages of the plant on fresh and dry herbage yield, dry leaf yield, essential oil content, and essential oil components. In total, twenty-nine components were identified in hyssop essential oil by GC/MS. Isopinocamphone was the dominating component (47.9–51.4%) in the all analysed oil samples. The results clearly demonstrated that oil contents are seriously affected by the environmental conditions and stage of blooming, with the highest oil yield and oil contents at the post-blooming stage.

Keywords:

• herbage and essential oil yield
• isopinocamphone
• ontogenetic variation

Introduction

Hyssop (Hyssopus officinalis L.), family Lamiaceae, is an important perennial culinary and medicinal plant. It is native to the Caucasus, North Western Iran, North Eastern Turkish Black Sea region, and Southern Anatolia (Tubives, 2007). It has erect branched stems up to 60 cm long covered with fine hairs at the tips, with narrow oblong 2–5 cm long leaves. It has small blue flowers borne on the upper part of the branches during summer. France is the major producer of hyssop; however, it is also cultivated in Italy, the former Yugoslavia, Bulgaria, Hungary, and Holland (Svab, 1992).

Hyssop, rich in volatile oil, flavonoids, tannins, and marrubin, has been used as a healing herb to alleviate digestive disorders, cure laryngitis, or accelerate wound healing in Turkish folk medicine for a long time. It relaxes peripheral blood vessels and promotes sweating. It is also used as an expectorant, carminative, anti-inflammatory, anticatarrhal, and antispasmodic in traditional medicine in many parts of the world. Hyssop oil finds its greatest use in flavouring alcoholic beverages, meat products, and seasonings (Svab, 1992; Akgul, 1993; Mitic & Dordevic, 2000; Jankovsky & Landa, 2002; Leung & Foster, 2003). It is further reported that certain fractions of hyssop (one being a polysaccharide designated as MAR-10) may inhibit the activity of the human immunodeficiency virus (HIV) (Kreis et al., 1990; Gollapudi et al., 1995).

There is a growing demand for plant-based medicines, health products, pharmaceuticals, food supplements, cosmetics etc. in the international market. Cultivation of medicinal and aromatic plants gives scope to improve the quality of drugs. Turkish farmers have established a worldwide reputation as cost-competitive suppliers of superior quality medicinal plants and spices. To stay viable, the Turkish agricultural sector must focus on the development of best agronomic practices for the production of medicinal plants of high importance. It is widely accepted that blooming stages affect essential oil content of many plants (e.g., Rohloff et al., 2000; Abd-El-Wahed & Gamal-El-Din, 2004; Deschamps et al., 2006). No data are available about hyssop; therefore, the present study aimed to determine the impacts of pre-, full, and post-blooming stages of hyssop on essential oil yield and oil contents.

Materials and methods

A local ecotype of hyssop widely found under natural covers of the North Eastern Turkish Black sea region and Southern Anatolia (Tubives, 2007) was obtained from the Department of Field Crops, Faculty of Agriculture, University of Ankara, Turkey. A seedbed nursery was raised in the experimental area of the Department of Field Crops, Dicle University, Diyarbakir (South Eastern Anatolia), Turkey to grow plantlets. Plantlets were transplanted to well-prepared field beds on 9 April 2003, when they had reached a height of 10–15 cm. For soil analysis of the experimental plot, a clean stainless steel probe was used to take 9 samples of 400 g each taken randomly at a depth of 0–40 cm. These were combined and mixed thoroughly in a plastic bucket. The samples were air-dried in a shady place before analysis to prevent alteration of the nutrient concentrations by soil micro-organisms, avoiding dust and foreign materials. One kg of this sample of mixed soil was placed in a soil sample bag and brought to the laboratory for analysis before the start of the experiment (Table I). The soil was analysed for total soil organic matter, total salts, soil saturation percentage, lime, phosphorus (P₂O₅), potassium (K₂O), soil pH, and electrical conductivity as per methods described by Klute (1986).

Table I. Soil conditions of the experimental site.

Organic matter (%)	1.16
Total salts (%)	0.16
Saturation percentage (%) with water	66.0
Lime (CaCO3) (%)	0.0
Phosphorus (P2O5) (kg ha^−1)	61.0
Potassium (K2O) (kg ha^−1)	0.0
Soil pH	7.45
Electrical conductivity (mmhoscm^−1)	3.8
Sampling depth (cm)	0–40

The experimental design consisted of randomised block design with three replications, in 3×1.35 m long lines 45 cm apart without irrigation or fertilisation. Weeding and thinning was done 30 days after plantation, by hand or as and when required. The plants were harvested with flower tops and leaves in the whole plots. The first year harvest was done on 18 June 2003 at the age of 2 months and 9 days at pre-blooming stage, on 27 June 2003 at the age of 2 months and 18 days at full blooming, and on 07 July 2003 at the age of 2 months and 28 days at post-blooming stage. The second-year harvests for pre-blooming, full blooming, and post-blooming stages were done on 20 May, 07 June, and 01 July 2004, at the age of 13 months 11 days, 13 months 28 days, and 14 months 22 days, respectively. Third-year harvests for pre-blooming, full blooming, and post-blooming were done on 31 May, 09 June, and 30 June 2005, at the age of 2 years 1 month 22 days, 2 years 2 months, and 2 years 2 months 21 days, respectively. Climatic conditions varied slightly between the experimental years, with mean January to July temperatures for 2003, 2004, and 2005 of 14.9, 14.8, and 15.1 °C, respectively. January to July mean long-term relative humidity was 56%, and that for 2003, 2004, and 2005 was 53, 51, and 45%, respectively. January to July mean long-term precipitation was 46.9 mm, with mean precipitation for 2003, 2004, and 2005 of 59.1, 49.8, and 37.2 mm, respectively. After removing border effects, the plants were cut at a height of 10 cm above soil and weighed to determine fresh herbage yield.

Dry leaf yield was determined after separating the leaves and stems in the dry herbage samples. Dry herbage yield was determined by drying fresh herbage samples from each plot in a cool, dry airy place for one week with a moisture content of 5.6–5.7% (w/w) measured according to AACC (1983) method 44–19. The roots were separated from the plants and the remaining parts were used for oil extraction. Essential oil percentage was measured volumetrically, by hydrodistillation using a Clevenger apparatus, in 20 g samples taken from each plot (%-v/w). The essential oil percentage of each plot was multiplied by dry leaf yield (L ha⁻¹) to determine essential oil yield.

Data on the mean fresh herbage yield, dry herbage yield, dry leaf yield, essential oil percentage, and essential oil components were analysed statistically using MSTAT-C (Michigan State University) computer program, and means were grouped using the least significant difference (LSD) test (p<0.05).

GC/MS conditions

Essential oils obtained by hydrodistillation of each plot sample were analysed by GS/MS. The analysis was performed using a Hewlett Packard 6890 N GC, equipped with HP-5 MS capillary column (30 m×0.25 µm) and HP 5973 mass-selective detector. For GC/MS detection an electron ionisation system with ionisation energy of 70 eV was used. Helium was used as carrier gas at a flow rate of 1 mL/min. Injector and MS transfer line temperatures were set at 220 and 290 °C, respectively. Column temperature was initially kept at 50 °C for 3 min, then gradually increased to 150 °C at a 3 °C min⁻¹ rate, held for 10 min, and finally raised to 250 °C at 10 °C min⁻¹. Diluted samples (1/100 in acetone, v/v) of 1.0 µL were injected automatically and in the split-less mode. Individual components were identified by spectrometric analyses using a computer library.

Results and discussion

The hyssop plants showed variable behaviour in all growth parameters during three years of experimentation due to the effects of temperature, relative humidity, and mean precipitation.

Ontogenetic variation and harvest time interaction affected fresh and dry herbage and dry leaf yield. The herbage (fresh and dry) and dry leaf yield were significantly affected by the changing relative humidity and precipitation from year to year and harvesting stage (p<0.05, Table II). The maximum herbage (fresh and dry) and dry leaf yield for each year were obtained during post-blooming and the minimum yield was obtained during the pre-blooming stage. On the whole, fresh and dry herbage yield and dry leaf yields in the second year were higher than those of the first and third year. Fresh and dry herbage yield increased from pre-blooming to post-blooming stage (Table II) due to an increase of vegetative cover of plants resulting in more utilisation of light for photosynthesis at the post-blooming stage (Ozguven & Tansi, 1998). Data related to fresh and dry herbage yield are compatible with those reported by Svoboda et al. (1993), who reported that the yield of fresh and dry leaves ranged between 5–32 tons ha⁻¹ and 0.67–3.26 tons ha⁻¹, respectively. Contrarily, Dzhumaev (1986) reported a lower per-hectare yield of 2–3 tons. Dry herbage yield also was higher than those reported by Tansi (1999) and Maksimovic et al. (1993), who reported average dry plant yield (hyssop) of 1980 kg ha⁻¹. These differences might have occurred primarily due to different genotypes used in the study or due to different ecological conditions (relative humidity, rainfall, temperature, etc.).

Table II. The effect of different harvesting stages on the fresh herbage yield, dry herbage yield and dry leaf yield of hyssop (the experimental design consisted of randomised block design with three replications). LSD = least significant difference.

	Fresh herbage yield (t ha ^−1)				Dry herbage yield (t ha ^−1)				Dry leaf yield (t ha^−1)
Harvest stages	2003	2004	2005	Mean	2003	2004	2005	Mean	2003	2004	2005	Mean
Pre-blooming	13.07e^1,2	31.83b	12.86d	19.25	3.96e*	9.15cd	4.14f	5.74	2.23d	4.09c	2.39d	2.91
Full blooming	29.91bc	35.09ab	19.93bc	28.31	8.05de	11.59b	7.20e	8.95	5.18b	5.28b	3.72c	4.72
Post-blooming	38.19a	34.26ab	25.79a	32.74	11.90ab	13.15a	10.25c	11.77	7.14a	6.83a	5.79b	6.59
Mean	27.06	33.73	19.53		7.97	11.29	7.20		4.85	5.40	3.97
LSD (0.05)		Harvest stages×year; 5.62				Harvest stages×year; 1.28				Harvest stages×year; 0.75
^1Means followed by the same letter are not significantly different according to LSD at 0.05 level.
^2 The experimental design consisted of randomised block design with three replications.

Jankovsky & Landa (2002) found that essential oil content is strongly influenced by variety, crop age, growth stage on the date of collection, climatic conditions, admixture of foreign plants, and extraction technology. Most of the researchers prefer dried samples (Venskutonis et al., 2005; Garg et al., 1999; Gorunovic et al., 1995) for distillation due to a number of reasons. Distillation from fresh or dry samples may affect the quality and quantity of the essential oil in aromatic plants, which may change during the drying process. Since the experimental fields were very far from the laboratories, it was feared that analysis of fresh samples might not give correct results; therefore, dried samples were preferred for oil content analysis. Our results showed wide variations in essential oil percentage and essential oil yield due to the effects of changing relative humidity and precipitation from year to year (p<0.05) with the highest yield during 2005.

The essential oil yield was also affected by the harvesting stage (p<0.05, Table III) such that young plants at the juvenile stage of pre-blooming showed lesser oil yield relative to the older plants at full blooming and post-blooming stages. The maximum oil percentage of 1.09% during 2005 at lower relative humidity and mean precipitation was significantly higher than the oil percentage of 2003 and 2004 with higher mean relative humidity and mean precipitation. The oil yield result is higher than those reported by Dzhumaev (1986). Svoboda et al. (1993) reported that essential oil concentration of hyssop varied depending upon the plant's flower colour. This comparison could not be made in this research, as the ecotype used in this study carried single coloured blue flowers. Dzhumaev et al. (1990), Svoboda et al. (1993), Mitic & Dordevic (2000), and Jankovsky & Landa (2002) report an essential oil percentage of hyssop in the range 0.34–1.4%, which is in agreement with the limits shown in this research and higher than those pointed by Maksimovic et al. (1993). They reported an oil yield of 0.23–0.38% in hyssop. These differences might have occurred primarily due to different genotypes, due to different ecological conditions or oil extraction techniques used in the respective experiments.

Table III. The effect of different harvesting stages on the essential oil content and essential oil yield (the experimental design consisted of randomised block design with three replications). LSD = least significant difference.

	Essential oil content (%)				Essential oil yield (L ha ^−1)
Harvest stages	2003	2004	2005	Mean	2003	2004	2005	Mean
Pre-blooming	0.82^1,2	0.40	0.95	0.74	18.5	20.6	22.3	20.5b
Full blooming	0.73	0.65	1.20	0.86	37.8	38.7	45.5	40.7a
Post-blooming	0.53	0.80	1.13	0.82	38.4	41.0	66.5	48.7a
Mean	0.69b*	0.62b	1.09a		31.6b	33.5b	44.8a
LSD (0.05)		Year: 0.15				Harvest stages: 10.6
^1Means followed by the same letter are not significantly different according to LSD at 0.05 level.
^2 The experimental design consisted of randomised block design with three replications.

Essential oil components

Twenty-nine components were identified in the oil of hyssop at different harvest stages (Table IV) compared with the previous findings with 18 components (Mitic & Dordevic, 2000). Marked differences in terms of harvest times were not observed in the oil composition of hyssop. Isopinocamphone, β-pinene, 4-carvomenthenol, γ-terpinene, carvacrol, and pinocarvone were found to be the main compounds of the oils. Isopinocamphone, β-pinene, 4-carvomenthenol, γ-terpinene, carvacrol, and pinocarvone together constitute 75% of total hyssop oil at pre-blooming, 81% at full blooming, and 78% at post-blooming stage, respectively.

Table IV. Essential oil composition of Hyssopus officinalis at different harvesting stages (the experimental design consisted of randomised block design with three replications).

	Pre-blooming				Full blooming				Post-blooming
Components	2003	2004	2005	Mean	2003	2004	2005	Mean	2003	2004	2005	Mean	Total Mean
4-Carvomenthenol	5.06	14.02	5.57	8.21	10.17	10.54	4.47	8.39	8.63	13.11	5.41	9.05	8.55
Camphene	0.14	–	0.10	0.08	–	0.19	0.15	0.11	0.13	–	0.10	0.08	0.09
Carvacrol	3.10	2.40	7.03	4.18	2.13	0.90	5.14	2.72	1.17	2.34	8.71	4.07	3.66
Caryophyllene oxide	0.20	0.50	0.08	0.26	0.23	0.31	0.07	0.20	0.20	–	0.11	0.10	0.19
cis-Sabinene hydrate	1.09	–	1.47	0.85	2.66	–	2.42	1.69	2.00	–	3.13	1.71	1.42
Elemol	0.96	5.59	0.56	2.37	0.82	1.29	0.25	0.79	0.49	1.06	0.45	0.67	1.28
Estragole	–	–	0.17	0.06	0.24	–	0.18	0.14	0.14	–	0.17	0.09	0.09
Eucalyptol	0.97	–	0.81	0.59	0.45	1.07	0.95	0.82	0.79	–	0.69	0.49	0.63
Germacrene-D	0.16	1.37	0.15	0.56	–	0.50	0.11	0.20	–	0.36	0.14	0.17	0.31
Isopinocamphone	53.87	45.04	44.92	47.94	56.68	47.05	40.92	48.22	53.54	55.17	45.55	51.42	49.19
Linalol	0.81	1.69	1.76	1.42	0.88	0.98	1.10	0.65	0.76	1.90	1.24	1.30	1.12
Myrcene	1.08	–	1.21	0.76	0.47	1.09	1.65	1.07	0.91	0.40	1.05	0.79	0.87
Myrtenol	1.77	5.00	1.09	2.62	2.12	1.66	0.92	1.57	1.46	2.02	0.97	1.48	1.89
Neryl acetate	0.09	0.40	0.20	0.23	0.41	–	0.23	0.21	0.25	0.61	0.38	0.41	0.28
p-Cymene	1.55	0.88	0.82	1.08	1.72	2.99	0.94	1.88	2.31	1.93	1.28	1.84	1.60
Pinocarvone	–	2.09	5.33	2.47	3.45	3.6	6.83	4.63	3.87	2.73	3.88	3.49	3.53
Sabinene	0.91	–	1.05	0.65	0.43	0.20	1.45	0.69	1.02	0.15	1.04	1.07	0.80
Terpinolene	0.75	–	0.57	0.44	0.55	0.71	0.57	0.61	0.79	–	0.49	0.43	0.49
Thymol	0.88	0.54	1.98	1.13	0.97	0.27	1.27	0.84	0.34	0.54	2.30	1.06	1.01
trans-Pinocarveol	0.82	0.43	–	0.42	0.38	0.46	–	0.25	0.35	0.45	–	0.27	0.31
α-Phellandrene	0.38	0.62	0.53	0.51	0.47	0.45	0.63	0.52	0.36	0.37	0.36	0.36	0.46
α-Pinene	0.74	–	0.72	0.49	0.20	0.86	1.07	0.71	0.52	0.21	0.51	0.41	0.54
α-Terpinene	1.64	0.23	1.50	1.12	1.13	1.56	1.94	1.54	1.85	0.46	1.24	1.18	1.28
α-Terpineol	0.53	1.69	0.62	0.95	0.81	–	0.54	0.45	0.71	1.20	0.64	0.85	0.75
α-Thujone	0.27	–	0.31	0.19	0.38	0.35	0.32	0.35	0.40	0.41	0.34	0.38	0.31
β-Phellandrene	0.95	–	0.95	0.63	0.57	1.29	0.75	0.87	1.17	0.35	0.90	0.81	0.77
β-Caryophyllene	0.19	0.75	0.11	0.35	–	0.32	0.08	0.13	–	0.29	0.10	0.13	0.20
β-Pinene	15.10	0.43	13.06	9.53	4.00	15.40	19.43	12.94	9.67	1.12	9.39	6.72	9.73
γ-Terpinene	2.93	1.96	4.62	3.17	5.58	3.78	3.84	4.40	5.04	2.85	3.77	3.88	3.82
Total	96.94	85.63	97.29	93.26	97.90	97.82	98.22	97.59	98.87	90.03	94.34	94.71	95.17

Results showed that hyssop oil has isopinocamphone as the major component (49%) and it increased gradually through the post-blooming stage. The other main component, β-pinene, increased from pre-blooming (9%) to full blooming (13%) stage, and decreased after the full blooming stage to 7%, in contrast to Nemeth (2005) who reported that β-pinene did not reduce during development stages. Between harvest stages, the highest isopinocamphone and β-pinene were obtained at full blooming stage during the first year for isopinocamphone (57%) and during the third year for β-pinene (19%), respectively (Table IV). However, Mitic & Dordevic (2000) found that the main component was isopinocamphone (45%), followed by pinocamphone (14%), germacrene-D-11-ol (6%), and elemol (6%). Veres et al. (1997) found that the oils from nine collections of H. officinalis grown from seed of different sources could be categorised depending upon their percentage composition of β-pinene, limonene, pinocamphone, and isopinocamphone. The oils were rich in isopinocamphone (5–50%), pinocamphone (3–50%), or contained β-pinene and limonene (1–60%). Our results are in accord with most of the previously published research except that reported by Gorunovic et al. (1995) and Vallejo et al. (1995). The main component in hyssop oil was identified as methyleugenol (38%) by Gorunovic et al. (1995), 1,8-cineole (53%) by Vallejo et al. (1995), and cis-pinocamphone (isopinocamphone) by Mitic & Dordevic (2000). Methyleugenol and 1,8-cineole were not detected in our sample. Conversely, the main component isopinocamphone in our oil was recorded at very much lower levels by both Gorunovic et al. (1995) and Vallejo et al. (1995), whereas Mitic & Dordevic (2000) reported that isopinocamphone constituted 45% of oils from Serbian hyssop ecotypes.

The results showed 57% isopinocamphone at full blooming during the first year, and this value is in agreement with previous reports (Mazzanti et al., 1998; Mitic & Dordevic 2000), This percentage is lower than that reported by Nemeth (2005) (61–74%). However, Garg et al. (1999) and Ozer et al. (2005) described pinocamphone as the major component. On the basis of their percentage in hyssop oil, major components obtained from this study are isopinocamphone and β-pinene, which are between limits of ISO standards 34–50% and 13–23%, respectively.

The components isopinocamphone, 4-carvomenthenol, and carvacrol showed variability during the growth cycles. Isopinocamphone and 4-carvomenthol increased from pre-blooming to post-blooming. The maximum content of pinocarvone (5%) was established at full blooming stage, otherwise the maximum γ-terpinene (4%) occurred at full blooming stage, but gradually reduced thereafter. It was shown that important differences occurred during these years in terms of oil components, which might be due to the biosynthesis of monoterpenes during this photoperiod.

A review of the published literature indicates that the oil composition of H. officinalis has large variations in the relative concentration of its major components (Mazzanti et al., 1998; Garg et al., 1999; Jankovsky & Landa, 2002; Ozer et al., 2005). In general, isopinocamphone and pinocamphone have mostly been reported in the essential oils of hyssop (Mitic & Dordevic, 2000). Our results are supported by the findings of Svoboda at al. (1993), Mitic & Dordevic (2000), Jankovsky & Landa (2002), and Mazzanti et al. (1998), as well as ISO standards.

Isopinocamphone and isopinocampheol constitutes about 50% of the oil. Hyssop oil's major components are very volatile and their relative proportion change under different environmental conditions (Svoboda et al., 1993). Nemeth (2005) reported that hyssop monoterpene composition increased at full blooming period during shoot development, the major component pinocamphone increased, and isopinocamphone decreased. In this study the highest isopinocamphone percentage was determined from the harvest at post-blooming stage.

In conclusion, harvesting stages had a significant effect on quantity and quality of hyssop; fresh and dry herbage yield and dry leaf yield of hyssop increased towards post-blooming stage. This suggests the importance of choosing a suitable harvesting stage to achieve the highest quality and quantity of essential oil.

Comparisons of this study with other studies also clearly reveal variations among essential oil composition of H. officinalis from different regions of the world. The maximum essential oil percentage of 1.09% and essential oil yield of 44.8 L ha⁻¹ were recorded during 2005 with the lowest fresh herbage yield of 19.5 t ha⁻¹, dry herbage yield of 7.2 t ha⁻¹, and dry leaf yield of 3.9 t ha⁻¹. These data suggest that hyssop plants do not behave similarly at different stages of growth. Moreover, the yield components and oil contents are seriously affected by environmental conditions such as temperature, precipitation, relative humidity, etc. at each stage of blooming. Therefore, it is strongly recommended that farmers should harvest hyssop at post-blooming stage for the highest oil yield.