The influences of fungal endophytes inoculation on the biochemical status of grape cells of different varieties in vitro

Abstract

The metabolic patterns of grape cells can be specifically shaped by different strains of dual-cultured fungal endophytes. In this work, a solid co-culture system was furtherly proposed to illustrate the different impacts of endophytic fungi on the biochemical status of grape cells of different varieties. By measuring the metabolic impacts of contact fungal endophytes on grape cells of the varieties ‘Rose honey’ (RH) and ‘Cabernet sauvignon’ (CS), we observed that most of the fungal strains used had promoting effects on grape cellular biochemistry parameters. Compared with the control, inoculation with most of the fungal strains increased the superoxide dismutase (SOD) and phenylalanine ammonia-lyase (PAL) activities as well as the total flavonoid (TF) and total phenolics (TPh) contents in both types of grape cells. Among the tested strains, RH34, RH49 and MDR36 had relatively stronger biochemical impacts on grape cells. More interestingly, in addition to the varietal specificity, a certain degree of fungal genus specificity was also observed during the metabolic interactions between fungal endophytes and grape cells, as fungal endophytes from the same genus tended to be clustered into the same group based on the affected biochemical traits. This work revealed the differential biochemical status effects of fungal endophytes on different varietal grape cells and raised the possibility of reshaping grape qualities by applying endophytes.

Introduction

Endophytes are microorganisms that reside within the tissues or organs of living plants without causing any visible manifestation of disease (Stone et al. 2000). Recent researchers tended to define the term ‘endophytes’ as referring only to the habitat of organisms and not to their functions, that is all microorganisms that colonize plant tissues internally for all or part of their lifetime (Cordovez et al. 2019). During a long period of coevolution, a mutualistic and symbiotic relationship is gradually established between endophytes and plants. According to the ‘balanced antagonism’ hypothesis, endophytes possess virulence factors that ensure self-resistance to the toxic metabolites of the host (Schulz and Boyle 2005; Schulz et al. 1999). This hypothesis suggests that asymptomatic colonization is a balance of antagonism between the plant and the endophyte. If endophytic virulence and plant defenses are balanced, the association remains asymptomatic and avirulent, and the endophyte manages to grow within its host without causing visible manifestations of infection or disease (Arnold 2007; Schulz and Boyle 2005; Souvik et al. 2012). However, the plant-endophyte interaction might be a much more complex and precisely controlled interaction than a simple equilibrium between virulence and defense. According to the plant-endophyte coevolution hypothesis (Ji et al. 2009), the host plant is able to synthesize a large number of stress-induced secondary metabolites to defend themselves against pathogens with the assistance of endophytes. During their long period of coevolution, the host plant provides nutrients and a habitat to support the survival of the endophytes. In turn, endophytes benefit their host plants by producing a number of bioactive constituents to increase adaptability to pathogens and abiotic stresses and benefit host growth (Kuldau and Bacon 2008; Marks and Clay 1996; Ownley et al. 2008; Rodriguez et al. 2009; Silvia et al. 2007). As a result, these hypotheses call for alternative and more detailed experimental designs to address the ecological significance of endophytes.

Grapevines are considered as one of the most important crops from an economic standpoint and are the one of the most cultivated fruit plants worldwide. Grapevines harbor diverse microorganisms including fungal endophytes that are partially contributed as ‘terroir’ which defined the qualities and characteristics of grapes and wines (Meinert 2006). During the whole vine lifespan, endophytes play significant roles as beneficial microorganisms or pathogens. The beneficial endophytes expect to be utilized in grapevine viticulture. Dwibedi and Saxena (2018) found some endophytic fungi isolated from the stem and leaf tissues of Vitis vinifera possessed the resveratrol-producing ability, belonging to genera Aspergillus, Botryosphaeria, Penicillium, Fusarium, Alternaria, Arcopilus, and Lasiodiplodia. Seven strains of endophytic fungi isolated from the grapevine variety of Cabernet sauvignon were deduced to have the capability of producing resveratrol, especially the fungus C2J6 with stable genetic properties producing resveratrol highly (Liu et al. 2016). These attested the special ability of the endophytes to produce the same or similar bioactive substances as the host plants. Thus, the interactions between fungal endophytes and grapevines deserve a thorough investigation. However, relationships between endophytes and their host plants, grapevines, are far from understood to date.

Our previous studies have demonstrated the differential impacts of endophytic fungal strains on the physiological and biochemical status of grape cells of variety CS in a dual culture system (endophytic fungi and cells were not physically in contact with each other) (Huang et al. 2017; Yang et al. 2018). In the present study however, endophytic fungi were inoculated directly onto grape cells from different varieties RH and CS. This will allow to further assess the biochemical impacts of the physically contacted fungal endophytes on grape cells and observe the inter/intra-cellular growth condition of fungal endophytes, which the grape cells will receive physical, biological and biochemical influences of the co-cultured fungi, as well as the effects of endophytic fungal strain on two varieties.

Materials and methods

Preparation of grape cells

Grape cells were induced from grape meristems (V. vinifera cv. Cabernet sauvignon and V. vinifera L.×V. labrusca L. Rose honey) as previously described (Hawker et al. 1973). Cells were subcultured for more than 5 generations on G5 solid medium supplemented with 30 g l⁻¹ sucrose, 0.2 mg l⁻¹ cytokinin, 0.1 mg l⁻¹ naphthylacetic acid and 7.5 g l⁻¹ agar in a growth chamber at 26°C.

Preparation of endophytic fungal strains

Endophytic fungal strains (Table 1) were all isolated from the leaves of the grape variety RH in local vineyards (Yunnan province, China) by the tissue patch method as previously described (Chen et al. 2020). The pure cultured fungal strains were identified using ITS DNA sequences (Ma et al. 2014). Total DNA of fungal isolates were firstly extracted with CTAB methods, and then ITS sequences were amplified with primer pairs ITS4 and ITS5 (White 1990). The nucleotide sequences of fungal strains have been deposited in GenBank under the accession numbers ON740926–ON740939. Before co-culture with grape cells, fungal strains were cultured on potato dextrose agar medium (peeled and diced potato 200 g, dextrose 20 g, agar 15 g and water 1 l) in Petri dishes (9 cm diameter) for one week.

Table 1. Morphological response of fungal endophytes and grape cells during co-culture.

Strain ID	Species	GenBank accession	Degree of callus oxidation		Degree of fungal mycellia colonization in or between cells		The weight gain of callus value (g·day−1)		The weight gain of callus RI (%)
			RH	CS	RH	CS	RH	CS	RH	CS
RH7	Epicoccum sp.	ON740926	heavy	moderate	few	few	0.4±0.2abcde	0.3±0.1abc	0.2	−29.7
RH12	Nigrospora sp.	ON740927	heavy	moderate	more	moderate	0.2±0.1e	0.2±0.1bc	−38.2	−58.4
RH32	Alternaria sp.	ON740928	heavy	moderate	moderate	few	0.2±0.1e	0.4±0.1ab	−38.4	−13.4
RH34	Trichothecium sp.	ON740929	heavy	moderate	more	more	0.3±0.1de	0.1±0.0c	−29.6	−68.2
RH36	Fusarium sp.	ON740930	heavy	moderate	few	few	0.5±0.2abcd	0.3±0.1abc	16.0	−39.9
RH44	Alternaria sp.	ON740931	heavy	heavy	few	moderate	0.4±0.1abcde	0.2±0.1c	11.5	−61.7
RH47	Fusarium sp.	ON740932	heavy	heavy	moderate	moderate	0.3±0.2cde	0.1±0.0c	−12.2	−76.4
RH48	Colletotrichum sp.	ON740933	normal	normal	moderate	moderate	0.5±0.1abc	0.4±0.1ab	30.4	−10.7
RH49	Fusarium sp.	ON740934	heavy	heavy	moderate	more	0.5±0.1abcd	0.2±0.1bc	18.8	−57.2
MDR1	Nigrospora sp.	ON740935	heavy	heavy	more	more	0.2±0.1e	0.2±0.0bc	−37.2	−52.1
MDR3	Fusarium sp.	ON740936	heavy	heavy	more	more	0.4±0.1abcde	0.2±0.1bc	0.7	−53.8
MDR4	Fusarium sp.	ON740937	heavy	heavy	moderate	moderate	0.4±0.1abcde	0.2±0.1bc	2.7	−51.4
MDR33	Colletotrichum sp.	ON740938	heavy	heavy	more	more	0.4±0.1abcde	0.3±0.1abc	−1.2	−39.1
MDR36	Colletotrichum sp.	ON740939	moderate	heavy	few	moderate	0.4±0.1bcde	0.2±0.0bc	−9.1	−50.8
Control			normal	normal	none	none	0.4±0.1abcde	0.4±0.1a

Values of physiological traits were indicated as ‘mean±standard error.’ Lowercase letters followed each value of the weight gain, indicating the significance of difference among inoculations. Only when no same letter appeared between inoculations, should be interpreted as significant difference. Strains RH7, RH12, RH32, RH34, RH44, RH48 were the same as those in the previous papers (Huang et al. 2017; Yang et al. 2018) .

Establishment of fungi-grape cell co-culture system

Sterilized callus medium (25 ml) was added to each sterilized Petri dish to generate solid culture plates. Approximately 2.0 g of prepared grape cells was weighed and transferred to the center of each Petri dish. The grape cells were cultured in the dark at 26°C for one week. For the fungi-grape cell co-culture, each endophytic fungus was fully suspended in 0.9% normal saline (final concentration was 2.5 g l⁻¹), and 4 µl (10 µg) of the suspension was added to the cells. Grape cells without any fungal inoculation were used as controls. The co-cultures and the controls were dark-cultured at 26°C for another 2 weeks. Every inoculation and the control contained 5 biological replicates. And fungal endophytes colonization in grape cells was observed by optical microscope.

Determination of the morphological traits of the grape cells

The oxidation degrees of the grape cells were judged according to the color of the cells: normal (no oxidation and light yellow); light oxidation (dark yellow); moderate oxidation (light brown) and heavy oxidation (dark brown). When different degrees of oxidation occurred in the replicates, the heaviest degree of oxidation was used to interpret the result of the inoculation.

Sample harvesting and measuring the weight gain of the grape cells

The grape cells were weighed at the beginning and end of the co-culture, the weights were recorded, and the daily mass increase (grams) was used to describe the cell weight gain. The grape cells were harvested after 2 weeks of co-culture, and fungal mycelia (some grape cells that were obviously covered with fungal mycelia) were carefully removed as much as possible with tweezers or a surgical knife. Then, the grape cells were ground into powder in liquid nitrogen and stored at −80°C for further assays. The assays for each trait were performed in triplicate.

Measurement of physiological and biochemical traits

Superoxide dismutase (SOD) activity was determined by the nitro-blue tetrazolium method as previously described (Dhindsa et al. 1981). Briefly, approximately 0.3 g frozen grape cell powder was weighed and added to 5 ml of 50 mM phosphate buffered saline (pH 7.8) containing 0.1% polyvinyl poly-pyrrolidone (PVPP), and the mixture was rapidly ground with liquid nitrogen. After centrifugation at 10,000 rpm at 4°C for 10 min, the supernatant was used for the assay. The mixture (6 ml) contained the supernatant (0.6 ml), 50 mM phosphate buffered saline (3 ml), 0.75 mM nitro-blue tetrazolium (0.6 ml), 0.13 mM methionine (0.6 ml), 0.02 mM riboflavin (0.6 ml), and 0.1 mM EDTA-Na₂ (0.6 ml). After incubation for 30 min under 4,000 lx light, the mixture was used to determine SOD activity with the absorbance at 560 nm.

The peroxidase (POD) assay was performed by the guaiacol method as described by Bergmeyer (1974). Frozen cell powder (approximately 0.3 g) was added to 5 ml of 200 mM PBS (pH 6.0) containing 0.1% PVPP, and the mixture was rapidly ground with liquid nitrogen. After centrifugation at 10,000 rpm at 4°C for 10 min, the supernatant was taken for the assay. The reaction mixture was prepared as follows (for 30 samples): 200 ml PBS (0.2 M, pH 6.0), 76 μl guaiacol (2-methoxyphenol), and 112 μl of 30% H₂O₂. The POD activity was assayed with the sample supernatant (0.2 ml) and the reaction mixture (6 ml) by the absorbance measuremen at 470 nm 5 times, once each minute.

The phenylalanine ammonia-lyase (PAL) activity in the enzyme extracts was assayed according to the method proposed by Carolyn et al. (1996), with some modifications as described by Xi et al. (2013). Frozen cell powder (approximately 0.3 g) was added to 5 ml of 50 mM boric acid buffer (pH 8.7) containing 0.1% PVPP, and the mixture was rapidly ground with liquid nitrogen. After centrifugation at 10,000 rpm at 4°C for 10 min, the crude enzyme (1 ml), 20 mM phenylalanine (1 ml) and 50 mM borate buffer (4 ml) were incubated for 60 min at 40°C in a water bath. The reaction was stopped by adding 200 μl hydrochloric acid (6 mM) and then was estimated by measuring A₂₉₀.

Total flavonoid (TF) was determined using the aluminum chloride colorimetric method with a minor modification (Jia et al. 1999). In a centrifuge tube, frozen cell powder (approximately 0.3 g) was added to 5 ml methanol (80%), ultrasonically extracted for 20 min, and then centrifuged at 10,000 rpm at 4°C for 10 min. The extract (2 ml) was mixed with 5% sodium nitrite (0.3 ml) and distilled water (4.4 ml). After 5 min, 10% aluminum chloride (0.3 ml) was added for 1 min, and 1 M sodium solution (2 ml) was then added to the solution. The absorbance was measured against a blank at 510 nm. The content was calculated by comparing the results with catechin (Sigma) as a standard.

Total phenolics (TPh) was determined using the Folin phenol colorimetric method (Asami et al. 2003). As TF determination, frozen cell powder (approximately 0.3 g) was added to 5 ml methanol (80%), ultrasonically extracted for 20 min, and then centrifuged at 10,000 rpm at 4°C for 10 min. The extracts (0.5 ml) were mixed for 5 min with distilled water (4.1 ml) and Folin phenol (0.4 ml), 7% sodium carbonate solution (4 ml) was added, and the mixture was incubated for 90 min at room temperature. The absorbance was measured at 750 nm and calculated by comparing the results with gallic acid (Sigma) as a standard.

Data analysis

The data were reported as the mean±standard error for multiple replicates and analyzed using SPSS Statistics 22.0 software for Windows. One-way analysis of variance (ANOVA) and Tukey’s HSD tests were used to determine the significance of the difference among the samples at the p≤0.05 level. Response indexes (RI) were used to describe the effects of the inoculation and were calculated by the formula: RI=(V_inoculation−V_control)/V_control. In the formula, V_inoculation is the mean value of a variable, and V_control is the value of the control for the corresponding variable. A positive RI indicates the promotion of an effect, and a negative RI indicates inhibition. Boxplots of RI data were generated by Origin Pro 8.0. A heat-map was drawn using Excel 2007, and squared Euclidean distance hierarchical clustering of the different endophytic fungal strains was generated based on biochemical parameters. PCA was performed with R software (Version 3.6.1).

Results

After 2 weeks of co-culture, the morphological, physiological and biochemical status of the RH and CS grape cells were significantly modified by some of the fungal endophytes compared with the control. Additionally, the impacts of fungal inoculation on RH and CS grape cells were generally different (Tables 1, 2). By optical microscope, fungal endophytes colonization in grape cells was observed (Figures 1, 2). Among these inoculations, the fungal mycelium of some inoculations could penetrate across cell walls and spread into the cells (RH12, RH34, RH44, RH49, MDR3 and MDR33), while most of the inoculations just colonized between cells. The colonization of fungal endophytes in RH and CS grape cells was varied from one fungal strain to another, but showed similar in the two varieties of grape cells. Fungal mycelium of strains RH34, RH49, MDR1, MDR3 and MDR33 reproduced more in CS cells, while strains RH7, RH32 and RH36 displayed less, and others showed moderate (Figure 1, Table 1). For RH cells, fungal mycelium of strains RH12, RH34, MDR1, MDR3, MDR33 reproduced more frequently, while strains RH7, RH36, RH44 and MDR36 displayed less and others showed moderate (Figure 2, Table 1).

Table 2. Biochemical values of grape cells after co-cultured with different strains of fungal endophytes and the different significances compared to control.

Strain ID	SOD (U/gfw)		POD (U/gfw)		PAL (U/gfw)		TF (mg/gfw)		TPh (mg/gfw)
	RH	CS	RH	CS	RH	CS	RH	CS	RH	CS
RH7	12.7±12.8	14.1±7.2	461.2±132.4*	1472.1±310.3	3.6±1.4	4.9±2.8	0.2±0.1	0.3±0.2	1.7±0.5	2.2±1.3
RH12	6.8±3.4	15.5±3.9	398.5±103.8**	1462.7±197.0	4.1±1.1	4.9±0.7	0.3±0.1	0.3±0.1	2.3±0.5*	2.2±0.5
RH32	8.4±1.2	18.3±8.5	489.2±223.5*	896.1±209.1	4.7±1.3	3.2±0.9	0.2±0.1	0.2±0.1	2.1±0.5	1.9±0.4
RH34	19.3±11.2**	23.7±8.0**	442.2±195.0**	1431.8±380.3	4.8±1.5	7.0±2.2*	0.4±0.1*	0.5±0.4**	2.6±0.3*	3.1±1.1**
RH36	9.5±8.0	11.2±6.4	535.9±124.5	1768.0±446.4**	7.1±3.4**	7.3±1.9*	0.4±0.1**	0.3±0.2	3.1±0.7**	2.3±0.8
RH44	7.6±1.3	10.3±7.0	488.0±170.5*	1832.5±407.1**	3.3±0.9	7.3±2.2*	0.2±0.1	0.4±0.2	1.7±0.9	2.6±1.1*
RH47	4.2±2.8	10.9±4.0	311.9±57.2**	1919.3±307.4**	4.4±1.9	6.3±2.2	0.2±0.1	0.3±0.1	2.2±0.8	2.4±0.5
RH48	3.3±3.1	4.7±6.1	958.3±154.6	846.0±244.9	2.8±2.2	2.8±0.2	0.2±0.1	0.0±0.0	1.4±0.6	0.8±0.2
RH49	6.5±5.3	15.9±2.8	435.9±155.9**	1974.6±677.4**	6.4±0.8**	6.1±2.0	0.4±0.1**	0.5±0.1*	3.0±0.7**	2.9±0.3*
MDR1	8.8±6.2	17.0±3.1	286.8±75.3**	878.2±156.4	5.7±2.5*	4.5±1.1	0.4±0.2**	0.3±0.2	2.7±1.0**	2.1±0.6
MDR3	8.5±3.4	30.7±2.4**	466.3±174.0*	1538.7±167.8	3.2±1.4	4.8±1.5	0.3±0.2	0.3±0.1	2.3±1.0	2.5±0.4*
MDR4	4.0±2.8	19.9±3.7*	346.8±62.8**	1395.5±89.1	3.1±0.9	5.6±2.2	0.2±0.0	0.3±0.2	2.00±0.5	2.5±0.8
MDR33	5.3±5.6	24.8±5.6**	326.5±80.9**	1128.6±169.3	3.2±0.9	5.9±0.9	0.2±0.0	0.5±0.1**	1.65±0.3	3.1±0.4**
MDR36	19.3±10.5**	17.5±2.4	594.0±310.8	1247.2±258.8	9.3±2.7**	6.2±0.9	0.7±0.2**	0.4±0.1*	5.4±1.3**	3.3±0.3**
Control	2.7±1.5	6.9±3.6	851.8±132.3	981.5±205.5	1.3±0.2	3.1±0.7	0.1±0.0	0.0±0.0	0.74±0.20	1.0±0.2

Values of biochemical traits were indicated as the mean±standard error with different significances marked as ‘*’or ‘**,’ compared to the control. *: significant difference at 5%, and **: significant difference at 1% (Tukey’s test).

Figure 1. Endophytes colonization in grape cells of CS after 14 days of co-culture. Bar=20 µm.

Figure 2. Endophytes colonization in grape cells of RH after 14 days of co-culture. Bar=20 µm.

On the other hand, co-culturing fungi with grape callus caused different degrees of oxidation (from normal to heavy), but the majority of the fungal strains used in this assay caused heavy oxidation (Supplementary Figure S1; Table 1). All strains caused similar degrees of oxidation in the two varieties of grape cells, except the fungal strains RH7, RH12, RH32, RH34 and RH36, which caused heavy oxidation in the RH grape cells and moderate oxidation in the CS grape cells. In contrast, the MDR36 strain caused heavy oxidation in the CS grape cells but moderate oxidation in the RH grape cells. On the other hand, due to the strength of the fungal invasion and the long time of co-culture (2 weeks), all the fungal strains used in this assay caused oxidation in the grape cells except strain RH48. Meanwhile, fungal endophyte inoculation greatly inhibited weight gain in CS grape cells; 9 of the total strains notably inhibited grape callus growth more than 50%, while RH32 and RH48 inhibited growth by only 13.4% and 10.7%, respectively (Table 1). In RH grape cells, half of the used fungal strains inhibited the grape callus growth, while the other half of the strains promoted and strains RH36, RH44, RH48 and RH49 promoted growth by more than 10%.

Of all the detected biochemical parameters, SOD, POD, PAL, TF, and TPh were the parameters that were highly variably influenced; however, their responses varied between grape cell varieties as well as among fungal strains (Table 2; Figures 3, 4). In general, during co-culture with fungal strains, all the parameters tended to be promoted except POD activity, which was positively influenced in CS cells but negatively influenced in RH cells (Figure 4A). Compared with the control, SOD activity was promoted in RH cells by strains RH7, RH34, and MDR36 more than 3–6-fold and promoted in CS cells by strains RH34, MDR3, and MDR33 more than 2–3-fold (Table 2). POD activity was greatly inhibited in RH cells except by RH48 but was only inhibited in CS cells by RH32, RH48, and MDR1. PAL and TPh were significantly promoted by all of the tested fungal strains in both varieties of grape cells, except by RH48 in CS cells, and the RH grape cells were influenced more variably than the CS cells (Table 2).

Figure 3. Heat map and clustering of biochemical changes of grape cells of RH and CS after co-cultured with different strains of fungal endophytes. Each row represents a fungal strain inoculation and each column represents a biochemical variable.

Figure 4. Boxplot of the impacts of inoculated fungal strains. (A) Biochemical traits in response to endophytic fungal inoculations. Red colored boxes represent the RH biochemical response ranges, and purple colored boxes represent the CS response ranges. (B) Impacts of fungal endophytic on grape cells biochemical traits of RH and CS. Red colored boxes represent the impacts on RH biochemical traits and purple colored boxes represent the impacts on CS biochemical traits.

According to the intensity of the biochemical changes displayed in the boxplots, the SOD, PAL, TF, and TPh of the RH cells as well as the SOD, TF, and TPh of the CS cells were more modified by endophyte inoculation than the other parameters (Figure 4A). Among these parameters, TF appeared to be the most sensitive parameter to endophytic inoculation. All of the strains used had significant effects on the content of TF, both in RH (0.9–7.4-fold) and CS (0.1–15.7-fold) grape cells (Figure 4A; Supplementary Table S1). Concerning the fungal endophytes used on the grape cells, most of the strains had strong impacts on biochemical parameters. RH34, RH49 and MDR36 significantly influenced 3–4 of the tested parameters, while strains such as RH7, RH12, RH32, and RH48 had relatively less influence on the grape cells (Figure 4B; Table 2).

Clustering analysis was performed based on the detected biochemical parameters of both grape varieties together, and all the fungal strains could be divided into 4 groups (Figure 3). Group I included 10 strains that was categorized as promoting or inhibiting the studied biochemical traits. The strains in this group included all of the tested strains from the genera Fusarium (5/5, five of the five used in this study), Nigrospora (2/2), Alternaria (2/2), and Epicoccum (1/1). RH34 (Trichothecium), MDR33 (Colletotrichum) and MDR36 (Colletotrichum) were clustered into groups II and III and were defined as having strong infective abilities. The grape biochemical traits were only slightly influenced or even depressed by another strain from the genus Colletotrichum, RH48.

PCA was performed to visualize the effects of endophytic fungal inoculation on the biochemical properties of grape cells, and had summarized the results of the research (Figure 5). The principal components PC1 and PC2 explained approximately 87.7% of the total variance. In the biplot, PC1 mainly separated the RH and CS cells, based largely on the contents of TPh, TF, PAL, and SOD. Among the parameters, TPh contributed the most to PC1, while SOD contributed the least. PC2 addressed the different effects on the grape cells, mainly related to the differences in POD activity (Figure 5). Thus, the fungal strains inoculated on the RH and CS cells were grouped into 2 clusters on the plot because of the different response of parameter POD. One cluster included all the strains inoculated on CS cells, and the other cluster included all the strains inoculated on RH cells except MDR36. The biplot provided a visual representation of the impacts of endophytic fungal inoculation on biochemical traits. For the RH cells, all of the strains contributed positively to SOD, PAL, TF and TPh compared to their levels in the control, while they all contributed negatively to POD, except RH48. For CS cells, all of the strains contributed positively to SOD, PAL, TF and TPh compared to their levels in the control except RH48, while they all contributed positively to POD except MDR1, RH48 and RH32. PCA plot also displayed the effect relevance and differences of certain fungal strains on grape variety RH or on CS. From the view of their distribution position on PC1, most of the tested strains such as RH12, RH32, RH48 and MDR1 produced similar biochemical changes on both the RH and CS grape cells, and compared with the control, fungal strain RH34, RH36 and MDR36 had stronger biochemical impacts on both the RH and CS grape cells than other strains.

Figure 5. PCA of the impacts of endophytic fungal inoculation on grape cells biochemical traits of RH and CS.

Discussion

The quantity and composition of grapes mainly depend on varieties, cultivation management, and environmental factors, such as drought, temperature, light/UV, plant hormones and environmental microbes (Arrizabalaga-Arriazu et al. 2020; Blanco-Ulate et al. 2015; Calderan et al. 2021; Guan et al. 2016; Pastore et al. 2017; Pilati et al. 2017; Yu et al. 2020; Zhou et al. 2018). Recent studies have covered the ability of some fungal endophytes to modulate the metabolism of host plants. Nevertheless, the mechanism of fungal endophytes to modify plant parameters remains poorly understood. It is hypothesized that the mechanisms underlying the metabolic impact of endophytes on a host include endophyte self-metabolizing, endophyte and host co-metabolizing, and signalling (Ludwig-Müller 2015). Obviously, the self-metabolizing and co-metabolizing pathways require physical contact between endophyte and host plant cells so that the endophyte can influence the plant metabolism. Previously, morphological, physiological and biochemical interactions between endophytic fungi and grape cells in vitro were demonstrated in a dual culture experiment (Huang et al. 2017; Yang et al. 2018), which the endophytic fungi and callus were not physically contacted, and diffusible signal molecules emitted from fungi were main stimulators that triggered the physiological responses in grape cells. In the present work, endophytic fungi were inoculated directly onto grape cells from two different varieties, RH and CS, for 2 weeks of co-culture, and the grape cells would receive physical, biological and biochemical influences of the co-cultured fungi. This paper therefore attempted to illustrate the possibility that the physiological and biochemical qualities and characteristics of grape cells were modified as well as to describe the specific interactions between fungal endophytes and grape cells.

In our study, all the tested RH and CS grape cells were subcultured for the same number of generations and had similar physiological status. Based on the above results, the degree of response to endophyte inoculation, i.e., the difference from the control cells, was greater in RH cells than in CS cells. And grape cells of different varieties showed different metabolic responses to endophytic fungal inoculation. Notably in our work, the endophytic fungal strains chosen for the inoculations were all isolated from the leaves of RH grapes. Thus, the impacts of all these fungal strains on RH and CS grape cells combined the effects of variety differences with some degree of host- and nonhost-originated from this variety difference. In a previous study, grape cells showed different growth effects in response to host- and nonhost-originated endophytes in dual culture (Peters et al. 1998). The growth of fungal endophytes was higher in dual culture with grape cells of the host than it was with cells of a nonhost (Sieber et al. 1990; Sirrenberg et al. 2010). Similarly, the same effect was found with a grass endophyte, illustrating that the growth of fungus was positively correlated with the host (Lu and Clay 1994). In our present results, regardless of the infection responses or metabolic patterns of the grape cells, it was intriguing that the responses to endophytic fungal inoculation displayed host specificity to some extent, which is consistent with previous research. However, all the endophytic fungal strains in the present study were isolated from the leaves of RH grapes. Further studies are needed to deduce host specificity in the strict sense by using endophytic fungal strains isolated from CS grapes.

Additionally, biochemical patterns shaped by strain-specific fungal endophytes were also observed in the grape cells in this work (Figure 3). In clustering the tested endophytic fungal strains based on the detected biochemical parameters of both varieties of grape cells, fungal strains from the same genus, such as Fusarium, Nigrospora, Alternaria and Colletotrichum, were likely to cluster together (Figure 3). These results suggest that there was a certain degree of genus specificity during the metabolic interactions between fungal endophytes and grape cells. Based on the results of this experiment, we observed that most of the tested endophytes tended to influence physio-biochemical profiles (SOD, PAL, TF, TPh) in grape cells of the two cultivars. According to many reports in literatures, plants lead to the generation of reactive oxygen species (ROS) or oxidative burst as a plant defense scheme, and endophytes defend themselves by producing a certain type of metabolites, that possesses antioxidants activities such as SOD and POD (Dubey et al. 2020; Saed-Moucheshi et al. 2014). However, the POD activity of RH grape cells was suppressed by the endophytes inoculation while increased with most endophytes inoculation in CS cells. The mechanisms of endophytes inoculation suppressed the POD activity in RH callus may be explained as the fungus has destroyed the defense system of RH cells, or ROS has burst but the cells were on the verge of death by oxidation, as the endophytes inoculation caused severely oxidation in the RH cells.

Therefore, the above results have shown that fungal endophytes may modulate their infective characteristics and the biochemical status of grape cells, and such responses can be genus-specific and different for different host varieties. This suggests that endophytic fungi may also work on the host at the individual plant level. As previous studies have shown, endophytes possess the ability to synthesize bioactive metabolites similar to those found in their host plants, which is acknowledged as the most promising source of bioactive compounds for agricultural, industrial and medicinal entities (Aly et al. 2010; Kharwar et al. 2011; Kusari and Spiteller 2012; Staniek et al. 2008; Strobel et al. 2004). In future research, we could select candidate fungal endophytes of certain genera to make guided changes to the quality of certain or different grape varieties.

Abbreviations

CS

Cabernet sauvignon

PAL

phenylalanine ammonia-lyase

PCA

principal component analysis

POD

peroxidase

RH

Rose honey

RI

response indexes

ROS

reactive oxygen species

SOD

superoxide dismutase

TF

total flavonoid

TPh

total phenolics