Turf performance and physiological responses of native Poa species to summer stress in Northeast China

Experimental site and turf trials

The experiment was conducted over two years (2017–2018) on a black loam soil at the Northeast Agricultural University Horticulture Station in Harbin, Heilongjiang province of Northeast China (45°43′55″N, 126°43′21″E). The research area has a typical cold temperate, continental monsoon climate: hot and rainy summer, and cold and dry winter, with tropical cyclones in summer and autumn seasons. Mean temperatures and precipitation during the study period are listed in Fig. 1. Soil organic matter was 24.6 g/kg with a pH in water of 7.0. Turf was established in May 2016, seeded at 15 g/m² for complete grow-in to produce a mature turf as the trial was established.

Native species of P. pratensis, P. sibirica and P. sphondylodes (Abbreviated as Hpp, Hpsi and Hpsp, respectively) from Heilongjiang were used as materials, along with three USA commercial Kentucky bluegrass cultivars ‘Midnight’, ‘Moonlight’ and ‘BlueChip’ (Abbreviated as Mid, Moo and Blu, respectively). Natural habitats of ‘Hpp’, ‘Hpsi’ and ‘Hpsp’ are around 45°46′38″N, 126°38′54″E, 45°49′22″N, 127°52′17″E and 46°16′48″N, 124°35′54″E, respectively. Average turf quality (TQ) of ‘Mid’, ‘Moo’ and ‘Blu’ is 6.4, 5.5 and 3.6 reported earlier by Shortell et al. (2004). Cytogenetic analyses by FISH indicated differences in chromosomes and ploidy levels of the three native Poa species (Fig. 2).

The field trial was conducted using a randomized complete block design with four replicates, with each plot measured 1.2 m wide by 5 m. During the growing seasons, granular and liquid fertilizers were supplied at 2 g N/m² per month. Turfgrass was maintained to a height of 4 cm tall above ground.

Assessments and measurements

Seasonal measurements were made monthly from May to October in 2017 and 2018. Turf performance was evaluated under natural field conditions, while physiological indices were performed in laboratory on field-collected samples.

Turf quality, color and turf coverage

Turf quality was rated from 1 to 9, where 1 is very poor, 9 is excellent. Turf color was scaled from 1 to 9, where 1 is completely faded, 9 is the most freshly green. A score of 5 is the lowest acceptable value for both turf quality and color. Turf coverage as a percentage of plot areas was determined monthly.

Turfgrass growth rate and tiller numbers

The height of turf was recorded on 6th, 7th, 16th, 17th, 26th, 27th of each month throughout the experimental period. The height of grass canopy was measured at 10 sites per plot using a digital caliper (Guanglu, China); growth rate was calculated as the average of the 10 daily height increment (mm) from 4 cm to which it was mowed, two days earlier. Tiller numbers were counted at 6 sites per plot monthly and expressed as numbers of tillers per square centimeters.

Chromosome

Chromosome numbers in cells of root tips were detected and counted using the fluorescence in situ hybridization (FISH) method (Said et al., 2021). Briefly, 15 to 20 mm of the end of vigorously growing roots was excised using a scalpel and fixed in a fixative solution of ethanol: glacial acetic acid = 3: 1 (v/v) for 24 h, then transferred to an enzyme mixture (4% cellulase and 2% pectinase), hydrolyzed in a 37 °C water bath, washed with water and fixed, and placed on a glass slide and then dispersed in 45% acetic acid. After drying with a fixative, those well-dispersed slides were selected for FISH using 5S rDNA and 45S rDNA probes. Observations were finally performed with a fluorescence microscope (BX-61, Olympus, Japan), with images collected via a microscope camera (DS-Ri1, Nikon, Japan).

Chlorophyll and osmotic regulating substance

Chlorophyll (Chl) was extracted by soaking 0.1 g of fresh leaves in 10 mL of dimethyl sulfoxide for 72 h (Hiscox & Israelstam, 1979); its absorbance (at 646 nm and 663 nm) was recorded with a Micro plate spectrophotometer (Epoch, BioTek, USA).

Content of total soluble protein (TSP) was measured with the method of coomassie brilliant, blue G-250 (Bradford, 1976). Powders (2 g in wt) of fresh leaves with 6 ml distilled water were ground into homogenate, and stood in a centrifuge tube for 0.5–1.0 h. Extractions (0.1 ml) were then centrifuged at 4,000× g for 15 min at 4 °C, finally, mixed with G-250 (5 ml). Absorbance was read at 646 nm and 663 nm. The content of TSP was calculated according to the standard curve.

Free proline (Pro) content of leaves was determined using the acid ninhydrin method as described by Ábrahám et al. (2010) with some modifications. Fresh leaves (0.5 g) were ground in 10 ml of 3% sulphosalicylic acid and centrifuged for 10 min, supernatant (2 g) was added to 3 ml of freshly prepared acid-ninhydrin solution and 2 ml of glacial acetic acid. The mixture was incubated at 90 °C for 1 h and put in an ice bath to stop the reaction; and then, 5 ml of toluene was added to the extract mixture and stood for 20 min at 25 °C to separate toluene and water; finally, the toluene phase was collected to measure the absorbance at 520 nm.

Total soluble sugar (TSS) was measured as described previously (Wei et al., 2016) with some modifications. A 0.5 g of powder of fresh leaf was homogenized in 5 ml of 80% ethyl alcohol (v/v); the extraction was centrifuged at 6,000× g for 15 min at 4 °C. The separated supernatant was transferred to another test tube, with 12.5 ml of 80% ethyl alcohol (v/v) added. Then, 1 ml of the reaction liquid and 1 ml of 0.2% anthrone were mixed; the mixture was incubated at 100 °C for 10 min and put in an ice bath for 5 min to stop the reaction. TSS content was determined using a spectrophotometer at 620 nm.

Antioxidant enzymes activity and O₂^.−content

For the assay of antioxidant enzymes activity and level of superoxide anion content (O₂^.−), 0.5 g of fresh leaves was homogenized at 5–10 °C using 2 mL of 50 mM phosphate extraction buffer (pH 7.8) in an ice-cold mortar. The mixture was centrifuged at 12,000× g for 15 min at 4 °C to collect the supernatant for quantification of enzyme activities and superoxide anion (O₂^.−). The superoxide dismutase (SOD) activity was measured by recording the rate of p-nitro blue tetrazolium chloride (NBT) reduction in absorbance at 560 nm (Giannopolites & Ries, 1977). Superoxide anion content (O₂^.−) was measured by the hydroxylamine oxidation method and the absorbance was read at 530 nm (Tian, Gu & Zhu, 2003). The activity of peroxidases (POD) and catalase (CAT) were determined in absorbance at 240nm and 470 nm, respectively (Tian, Gu & Zhu, 2003).

Levels of glutathione (GSH), ascorbic acid (AsA), malondialdehyde (MDA) and H₂O₂

For glutathione (GSH), ascorbic acid (AsA) and malondialdehyde (MDA) contents, 0.1 g of fresh leaves was ground with 1 mL of 10% (w/v) trichloroacetic acid (TCA), centrifuged at 12,000× g at 4 °C for 20 min; then the supernatant was collected for assay. The content of AsA was measured following the method of Guo et al. (2013). 4 ml of NaH₂PO₄, 0.4 ml of 10% TCA, 0.4 ml of 44% H₃PO₄, 0.4 ml of 4% 2, 2-Bipyridine, and 0.2 ml of 3% FeCl₃ were added to 1 ml of supernatant. The mixture was heated at 37 °C for 1 h. Absorbance was measured at 525 nm. Content of GSH was determined according to the method of Griffith (1980), 1 ml of the supernatant was added with 2.4 ml of 0.1 mol L⁻¹ phosphate buffered saline (pH 7.7) and 0.2 ml of 4 mmol L⁻¹ 5, 5-dithiobium-(2-nitrobenzoic acid). Absorbance was determined at 412 nm. Content of MDA was determined by the thiobarbituric acid method (Ma et al., 2015), 2 ml of the supernatant was mixed with 2 ml of 0.6% TBA, the mixture was then heated at 95 °C for 30 min, quickly cooled, and then centrifuged at 10,000× g for 10 min. Absorbance of the supernatant was measured at 450 nm, 532 nm, and 600 nm to calculate the MDA content. The procedure outlined in the H₂O₂ Quantitative Assay Kit (BioVision, USA) was followed for the assay of H₂O₂ content at 415 nm.

Statistical analysis

Mean data over two years of 2017 and 2018 in projected turf performance and physiological indices were analyzed using ANOVA (SPSS v 18.0, Inc., Chicago IL). The Fisher’s protected least significant difference (LSD) at P < 0.05 was used to detect if significant differences existed among entries or among months in the experiment.

Turfgrass quality and color

Results of effects of seasonal temperature on turf quality (TQ) and color of the native and introduced bluegrasses are presented in Table 1. TQ of ‘Mid’ was statistically superior throughout the seasons but similar to that of native‘Hpp’ except during September.

TQ of ‘Moo’ was similar to that of ‘Mid’ during September and October. ‘Blue’ performed similar to native ‘Hpsi’ throughout the growing seasons. During May, September and October, the turf color of ‘Hpsp’ being darker green was significantly better to that of ‘Moo’, ‘Blu’ or ‘Hpsi’, but similar to that of ‘Mid’, ‘Moo’ and ‘Hpp’. Whilst TQ was the poorest, compared with other bluegrasses from June to September. During early spring seasons, the color of ‘Hpsp’ was significantly better than that of ‘Hpsi’ and ‘Blu’, but similar to that of ‘Moo’ and ‘Blu’ in October. In summer, the color of ‘Hpp’ (8.0 in June, 7.5 in July, and 6.8 in August) was similar to that of ‘Mid’ and ‘Moo’.

Turf coverage and growth rate

Analysis of variance exhibited significant variabilities in turf coverage and growth rate among accessions varying with changing temperatures (Table 2). Maximum turf coverage was observed in ‘Mid’, it was however statistically similar-to that in ‘Moo’ and ‘Hpp’. Turf coverage observed in ‘Blu’ was initially similar–to that in ‘Moo’, ‘Mid’ and ‘Hpp’ in the month of May but was lower than in ‘Mid’ in other subsequent months of the growing season. Although the turf coverage of ‘Hpsi’ performed lower than the rest except ‘Hpsp’, that was still acceptable. The lowest turf coverage was observed in ‘Hpsp’ from June to August. Generally, the turf coverage reduced in summer and then increased in autumn. The highest growth rate was observed in ‘Mid’ regardless of a rising seasonal temperature from early spring to summer, which was similar to that in native ‘Hpsi’ except September. The growth rate measured in ‘Hpp’ was similar to that in ‘Moo’ through the growing season, and significantly lower than that in ‘Mid’ and ‘Hpsi’ but generally higher than that in ‘Blu’. ‘Hpsp’ had the lowest growth rate during spring and autumn.

Number of tillers

The number of tillers were significantly different among entries in both years (Table 3). Maximum tillers were observed in ‘Mid’ regardless of changes of the seasonal temperatures. The tiller number of ‘Hpp’ was statistically similar to that of ‘Mid’ and ‘Moo’ for the whole growing period although slightly different in May and September. A similar seasonal trend in tiller numbers was observed in ‘Hpsi’ during both years, however, its tiller numbers were apparently lower than those of others except ‘Hpsp’ which had the lowest tiller numbers among all entries.

H₂O₂, O₂^.−, Chl and MDA contents

H₂O₂, O₂^.−, Chl and MDA contents among the turfgrass entries showed significant differences (Figs. 3A–3D). These parameters were observed to increase with rising temperatures from spring to summer, reaching a peak in August and then declining from October. Overall, H₂O₂ and O₂^.− contents of native accessions were significantly higher than those of introduced cultivars in spring and early summer. A maximum production of H₂O₂ and O₂^.− was observed in ‘Hpsp’ in May and June, which was surpassed by ‘Hpsi’ in summer and autumn. The lowest H₂O₂ and O₂^.− contents were observed in ‘Mid’. However, it was statistically similar to those in ‘Hpp’, ‘Moo’ or ‘Blu’, from May to July but notably lower in the rest of the experimental period (Jul.–Oct.). MDA content varied significantly among species with changing temperatures (Fig. 3D). The MDA concentration in all tested accessions was lower in spring, increasing in summer while declining in autumn. Among the entries, the maximum MDA was observed in ‘Mid’ in early spring, which was similar to ‘Hpp’. In summer and late autumn, the highest MDA concentration was noted for ‘Blu’ which was similar to that for ‘Moo’ and ‘Hpp’ while the lowest was observed in ‘Hpsi’. The Chl content in all tested accessions was also significantly affected by seasonal temperatures (Fig. 3C). The highest Chl content was observed in ‘Hpsp’ in early spring, which decreased in summer and recovered in autumn. Chl content in ‘Mid’ performed similar tothat in ‘Hpp’. Chl content in ‘Blu’ was distinctly lower in early spring than that in other entries while was similar to that in ‘Moo’ in summer and autumn. It was also noted that Chl content in ‘Moo’ and ‘Hpp’ was similar throughout the experiment.

SOD, CAT, POD, and GSH activity

SOD, CAT, POD, and GSH activities are presented in Fig. 4. The SOD activity showed an increasing trend with rising temperature from spring to summer and then declining from autumn (Fig. 4A). The highest SOD activity was observed in ‘Mid’ while the lowest was recorded in native ‘Hpsp’ regardless of the seasonal temperatures. The SOD activity in ‘Hpp’ was statistically similar to that in ‘Mid’ and ‘Moo’ in late summer and autumn-similar to that in ‘Moo’ and ‘Blu’ in early summer while higher than that in ‘Hpsi’ and ‘Hpsp’ for the experimental duration. The CAT activity varied significantly among the accessions, but, with no effect of seasonal temperatures (Fig. 4B). The highest CAT activity was observed in ‘Mid’ which was similar to that in ‘Hpsi’, and so were in ‘Moo’ and ‘Hpp’ across the seasons. The lowest CAT activity was observed in ‘Hpsp’ in spring and autumn. The POD activity was lower in spring while it increased with temperatures in summer (Fig. 4C). The Maximum POD ativity was observed in ‘Hpp’ in all seasons. The POD activity observed for ‘Mid’ was comparable to ‘Hpp’ for the experimental duration except in May, when it was slightly lower. ‘Hpsp’ showed the lowest POD activity followed by ‘Hpsi’. Similarly, ‘Hpp’ showed a GSH activity comparable to ‘Mid’ in early summer and autumn, which was clearly higher than ‘Hpsi’ and ‘Hpsp’ but similar to ‘Moo’ and ‘Blu’. A lower GSH activity was detected in ‘Hpsp’ regardless of seasonal temperatures (Fig. 4D).

TSP, TSS, Proline and AsA content

Significant variations (P < 0.05) in TSP, TSS, Proline and AsA content among Poa accessions under changing seasonal temperatures (Fig. 5). ‘Hpp’ had a TSP content similar to ‘Mid’ for most of the experimental period except late summer, when ‘Hpp’ and ‘Hpsi’ were higher than ‘Mid’ (Fig. 5A). ‘Moo’ performed similar to ‘Mid’ and ‘Hpp’ in spring and early summer, but lower in late summer and autumn, which were however similar to ‘Hpsp’. and ‘Blu’. The TSS content also increased with rising temperatures from spring to summer and then declined in autumn (Fig. 5B). ‘Mid’ produced the maximum TSS, which was similar to ‘Hpp’ throughout the experiment except late summer when ‘Hpp’ was distinctly greater than ‘Mid’. ‘Blu’ was comparable to ‘Moo’ and ‘Hpp’ across all the seasons. The lowest TSS production was observed in ‘Hpsi’. The accessions varied significantly in proline content with changing temperatures from spring to autumn (Fig. 5C). Proline concentration increased in ‘Mid’, ‘Moo’, and ‘Hpp’ more than other individual Poa entries. The performance among entries was mostly indistinguishable in spring and early summers but onward, ‘Hpsi’ and ‘Hpsp’ lost its ability to produce proline and showed the lowest proline concentration. The AsA content differed significantly among the accessions (Fig. 5D). An increase in AsA content was more prominent in ‘Mid’ followed by ‘Hpp’ from spring to summer. In spring, the AsA produced by ‘Hpp’, ‘Mid’, ‘Moo’ or ‘Blu’ was roughly the same, but from July to October, maximum AsA was detected in ‘Mid’ followed by ‘Hpp’, ‘Moo’ and ‘Blu’.The lowest AsA concentration was observed in ‘Hpsp’.