Nematicidal Streptomyces against Meloidogyne incognita by hydrosoluble macromolecules

Background: Although actinomycetes, which produce antibiotics, have been reported to produce nematicidal metabolites, macromolecules with nematicidal activities have rarely been described. Results: The culture ltrates of Streptomyces albogriseolus strain DN41 and S. micarius strain D153 presented 82.13 % and 86.96 % nematicidal activity against Meloidogyne incognita, respectively, after 12 h. A pot experiment highlighted that these culture ltrates were characterised by signicant ecacy. The biocontrol eciency of strain DN41 reached 61.68 % and that of strain D153 68.33 % after 30 days at a dose of 100 mL pot -1 . The active metabolites produced by the two strains were hydrosoluble and sensitive to heat, with molecular weights above 8000 Da for DN41 and of 1000–8000 Da for D153. However, these metabolites were negative to protein reaction. A chemical characteristic test indicated that these active metabolites were polar polymers that contained a large amount of sugar and phenolic hydroxyl compounds. Conclusions: These results revealed S. albogriseolus strain DN41 and S. micarius strain D153 as potential agents for the control of root-knot nematodes, as hydrosoluble macromolecules have good compatibility and are convenient for practical application.

After the extraction of the organic solution, the aqueous phase was characterised by the strongest nematicidal activity and the respective mortality generated by strains DN41 and D153 surpassed 98.5 % after 12 h. In contrast, the extracts from the ethyl acetate and n-butanol phases did not display signi cant nematicidal activity. Fig. 1

B, C Chemical characteristics of active metabolites
The active culture ltrates of strains DN41 and D153 were sensitive to heat and their activity were decreased with the increasing of temperature. Nematicidal activity was the highest at 37 °C, then decreased by 66 % for strain DN41 and 50 % for strain D153 at 47 °C, and was inhibited at 67 °C. (Fig. 2 A, B) The nematicidal activity of dialysate DN41 reached 98.00 % 12 h after dialysis with a molecular weight cut-off at 8000-10 000 Da, whereas no activity was detected for dialysate D153. In contrast, when performing a dialysis with a molecular weight cut-off at 1000 Da the nematicidal activity of dialysate D153 reached 94.99 % after 12 h. (Fig. 2 C, D) Notably, enzyme hydrolysis with proteinase-K did not signi cantly affect the dialysate of either strain. Furthermore, the reactions of ninhydrin biuret and ammonium sulphate precipitation were negative, which indicated that there were no amino acids, peptides, or proteins in the dialysates of both culture ltrates. Fehling's solution reacted with both dialysates and resulted in brick red precipitates, whereas Keller-Kiliani reactions were negative. This implied that the active substances contained reducing sugar but no α-deoxyglucose. Moreover, ferric chloride reactions were negative with the two dialysates, while potassium ferricyanide-ferric chloride tests were positive, which highlights that the dialysates contained aromatic amines and their reductive substances but no phenolics. Additionally, the production of lead acetate precipitates revealed there were tannins. However, the TLC detection of hydrolysed samples indicated there was no gallic acid nor catechinic acid. Likewise, phenolics, lactones, steroids, triterpenoids, and saponins were not detected in the two active dialysates. (Table 1) Finally, the organic elemental analysis determined that the molecular formula of the active metabolite produced by strain DN41 was (C 34 H 75 O 63 N) n , and that of the active metabolite associated with strain D153 was (C 7 H 13 O 9 N) n ( Table 2).

Pot experiment
The culture ltrates of the two strains were signi cantly more e cient against M. incognita than the negative control was after 30 days of treatment. Notably, the inhibition ration in per 1g soil of the positive control reached 96.87 %, and the highest inhibition ration of the strain DN41 and D153 was of 81.54 % at the dose of 180 mL pot -1 and the strain D153 was 97.56 % at the dose of 100 mL pot -1 , respectively ( Fig. 3 A). The inhibition ration in per 1g root of the positive control reached 77.83 %, and the highest inhibition ration of the strain DN41 was of 66.42 % at the dose of 180 mL pot -1 and the strain D153 was 77.79 % at the dose of 100 mL pot -1 , respectively (Fig. 3 B). The biocontrol e ciency of the positive control reached 68.33 %, and the highest biocontrol e ciency of strains DN41 and D153 was of 61.68 % and 68.33 %, respectively, at the dose of 100 mL pot -1 (Fig. 4). Furthermore, the culture ltrates of both strains did not damage the plants, and, after 30 days, promoted greater plant height than the negative control did (Fig. 5).

Discussion
Natural products produced by actinomycetes, especially Streptomyces spp., are important biological resources to control root-knot nematodes [7]. In recent years, there have been multiple studies on the control of root-knot nematodes through the secondary metabolites produced by Streptomycete [16]. Notably, the culture filtrates of S. avermitilis Manp [17] and S. hydrogenans DH16 [15] have been reported to cause 89.54 % and 90.00 % mortality against M.incognita, respectively, after 72 h. In our study, S. albogriseolus strain DN41 and S. micarius strain D153 were newly reported to have effect on M.incognita. Notably, the mortality generated by the culture ltrates of S. albogriseolus ND41 and S. micarius D153 has been reported to exceed 82 % after 12 h and 96 % after 48 h. This highlights that the two strains have a better effect on the control of M. incognita in vitro. In our pot experiment, the treatments with the culture ltrates of strains DN41 and D153 were characterised by a signi cant biocontrol efficiency, especially at the dose of 100 mL pot -1 , and their effects were close to the positive control avermectin.
The nematicidal metabolites of the two strains of actinomycetes evaluated in the present study both are compounds with a large polarity and are soluble in water but only slightly soluble in organic solvents. Their active substances are sensitive to heat and, hence, are inhibited at high temperatures. Through dialysis, we determined that the molecular weight of the active substance of strain A is above 8000-10 000 Da, while that of the active substance of strain B is relatively small, namely between 1000 and 8000 Da. We rst speculated that the active substances were proteins. However, after the addition of Protease-K, the activity of these substances was not affected. Furthermore, the reactions of biuret and ninhydrin, as well as the precipitation of ammonium sulphate, were all negative, which indicates that the active substances of the two strains of Actinomyces were heat-sensitive macromolecular compounds, but not proteins. Notably, these compounds contained a large amount of reducing sugars, but no α-deoxyglucose, lactone ring structure, steroids, nor triterpenes. Additionally, the metabolites did not contain phenolic compounds, but included aromatic amines and reductive substances. Based on the hydrolysis with potassium ferricyanide-ferric chloride reaction and lead acetate precipitation reaction, we inferred that the samples might be tannins, and gallic acid, catechin, and the hydrolysed samples would all be positive to the same chromogenic reaction. However, no gallic acid nor catechin was detected with TLC after hydrolysis. A possible explanation is that the large amount of phenolic hydroxyl in the structure of the active substance induced the positive result of the colour reaction of potassium ferricyanide-ferric chloride. Therefore, the nematicidal compounds may be polymers containing number of phenolic hydroxyl groups and sugars. Based on the elemental analysis, the molecular formula of the active substance of strain DN41 is (C 34 H 75 O 63 N) n , and that of the active substance of strain D153 is (C 7 H 13 O 9 N) n . Nonetheless, the structures of these macromolecular compounds are complex, and further investigation is required to clarify the bonding between the groups in these compounds.
To date, few hydrosoluble macromolecules with nematicidal activity have been reported. The water-soluble nucleoside antibiotic 1100-50, which was isolated from S. lavendulae SANK 64297, partly inhibited M. hapla J2s after 24 h [18]. Moreover, the actinomycete strain Snea253 produced amino oligosaccharides with a 98.21 % nematicidal activity against M.incognita 24 h after treatment with 1 mL of crude extracts [19]. In the present study, we discovered new hydrosoluble macromolecule polymers with a strong nematicidal activity, which contain number of sugars and phenolic hydroxyl compounds.

Conclusions
In conclusion, in our study, the two strains of S. albogriseolus ND41 and S. micarius D153 showed signi cant nematicidal activity against M. incognita, and signi cant biocontrol potentials in pot experiment , expecially at the dose of 100 mL pot -1 .
Moreover, the two strains showed no damage on the plant growth. We discovered new hydrosoluble macromolecule polymers with a strong nematicidal activity, which contain number of sugars and phenolic hydroxyl compounds. As hydrosoluble compounds have good compatibility and are convenient for application, the results of our study provide a baseline for the development of novel nematicides against root-knot nematodes.

Methods
Preparation of nematodes and actinomycetes M. incognita were obtained from infected tomato plants based on Kihika et al. [20]. Root pieces with galls were uprooted and washed with running water to remove the soil. Nematode eggs were selected, under a dissecting microscope, with a 300-mesh sieve in a culture plate containing sterile water, then placed in dark conditions at 27 °C ± 2 °C for 2-5 days. Hatched juveniles (J2s) were collected for subsequent nematicidal activity assay and pot experiment.
The actinomycetes S. albogriseolus DN41 and S. micarius D153 were isolated from soil samples obtained from tomato roots infested with M. incognita in Yunnan Province, China. The strains DN41 and D153 were cultured in sterile liquid MB medium at 28 °C and 180 rpm for 7 days [21]. The cultures were then ltered with a 0.22-μm membrane and the ltrates were stored at 4 °C for subsequent use.

Nematicidal activity assay
The nematicidal activity of culture ltrates, extracts, and dialysates against M. incognita were measured according to the modi ed method of Park et al. [22]. Freshly hatched J2s were added to a 16-well plate (200 J2s in 20 μL) and treated with 1 mL test samples. Sterile water served as control. Every treatment had three replicates and was repeated three times.

Extraction of active metabolites
The culture ltrates of the two Streptomyces strains were extracted with ethyl acetate rstly and ra nate phase were extracted with n-butanol sequentially. The ethyl acetate, n-butanol, and aqueous phase extracts were dissolved in methanol and diluted to a 5 % methanol aqueous solution. The nal concentration of each sample was adjusted to 1 mg mL -1 for the nematicidal activity test.
Culture ltrates (1 mL) with nematicidal activity were added to 2 mL centrifugal tubes, then heated 1 h in a water bath at 37 °C, 47 °C, 57 °C, 67 °C, 77 °C, 87 °C, and 97 °C, respectively. The nematicidal activity of each sample was tested after heating.
The culture ltrates were eluted with XAD-16 resin (Amberlite, USA) to remove non-polar compounds. Eluates were placed into pretreated dialysis bags (Spectrumlabs, USA) with a molecular weight cut-off at 1000 Da and 8000-10 000 Da, respectively, and dialyzed for 24-48 h. The dialysis buffer was replaced every 3-6 h. The dialysates were freeze-dried.
Parts of the dialysates were diluted with deionized water to a concentration of 1 mg mL -1 for the nematicidal activity test.

Determination of chemical characteristics
Each active dialysate solution (1 mL) was transferred to a 2-mL centrifugal tube, to which 1 μL of 20 mg mL -1 proteinase-K was added for hydrolysis, 2 h at 37 °C. The respective nematicidal activity of hydrolysates was then tested.
To identify the chemical components and characteristics of active dialysates, the following reactions were conducted according to the methods from Kong [23]: the ninhydrin reaction to identify amino acids, biuret reaction and ammonium sulfate precipitation test to identify peptides and proteins, Fehling's solution and Keller-Kiliani reaction to identify sugar, ferric chloride reaction to identify phenols, potassium ferricyanide-ferric chloride reaction to identify aromatic amine, lead acetate precipitation to identify tannins, hydroxamic acid iron test to identify lactones, thin layer chromatography (TLC) to compare the standards of gallic acid and catechinic acid, and with heated chloroform to identify carcass as well as triterpenoids and their glycosides. An organic elemental analyser (Vorio EL, Germany) was used to determinate the molecular formula of active metabolite substances.

Pot experiment
The pot experiment was performed in a controlled greenhouse environment following the modi ed method of Hu [24].
Each plastic pot was lled with 150 g of soil and 3-week-old Zhongza9 tomato seedlings (Jingdian Seeds Company, Yunnan, China) were transplanted. After a week, 500 newly hatched J2s were inoculated per pot. The respective culture ltrates of the two strains were inoculated to pots at different doses, namely 30 mL pot -1 , 100 mL pot -1 , and 180 mL pot -1 . The negative control consisted in plants grown in untreated soil, while the positive control corresponded to soil with 2.5 g pot -1 of 0.5 % avermectin. Each treatment had ve replicates and was repeated three times. Pots were cultured in a greenhouse at 27 ± 2 °C and watered daily.
After 30 days, plants were uprooted and 10g rhizosphere soil samples were collected from each pot. Juveniles were extracted by modi ed Baermann funnel technique from soil samples, roots with galls were rinsed with sterilized water and cut 5g root from each plant and placed in a culture plate containing sterile water. The nematode populations in per 1g soil and root were calculated under a dissecting microscope after setting aside for 24h at room temperature. The inhibition ration in per 1g soil and root, biocontrol e ciency and growth condition of plant were used to measure the biocontrol potentials of two strains.
Inhibition ration was estimated with the following formula: Inhibition ration (%) = (NP tr -NP c ) / NP c × 100 , where NP tr is the nematode populations in per 1g soil or root of the treatment group, NP c is the nematode populations in per 1g soil under control conditions. Biocontrol e ciency was estimated with the following formula: Biocontrol e ciency (%) = (GI c -GI t ) / GI c × 100, where GI c is the gall index under negative control conditions, and GI t is the gall index in the treatment group [25]. The gall index was measured according to a ten-point scale [26]. Moreover, the growth condition of plant was measured with plant height (cm), shoot weight (g), and root weight (g).

Statistical analysis
Data from the experiments were tested using a one-way analysis of variance, and signi cant differences were analyzed using SPSS software 26.0, the means of the treatments were compared with Duncan's multiple-range test (p < 0.05).    The effect of different temperature and dialysis on the nematicidal activity of two strains against M. incognita. A. response of S. albogriseolus ND41 to temperature; B. response of S. micarius D153 to temperature; C. nematicidal activity of dialysate from S. albogriseolus ND41; D. nematicidal activity of dialysate from S. micarius D153.

Figure 3
The inhibition ratio of two strains against M. incognita in soil and root at different doses. A. the inhibition ratio against M. incognita in soil.; B. the inhibition ratio against M. incognita in root. ND41-30, ND41-100 and ND41-180 were treated with culture ltrates of S. albogriseolus ND41 at the dose of 30mL pot-1, 100 mL pot-1 and 180 mL pot-1. D153-30, D153-100 and D153-180 were treated with culture ltrates of S. micarius D153 at the dose of 30mL pot-1, 100 mL pot-1 and 180 mL pot-1. PC were treated 2.5 g pot-1 of 0.5 % avermectin. The experiments were performed with three replicates and data was the average of replicates. The bar means standard deviation. * present the signi cant difference (p < 0.05).

Figure 5
Plant growth treated by two strains at different doses. ND41-30, ND41-100 and ND41-180 were treated with culture ltrates of S. albogriseolus ND41 at the dose of 30mL pot-1, 100 mL pot-1 and 180 mL pot-1. D153-30, D153-100 and D153-180 were treated with culture ltrates of S. micarius D153 at the dose of 30mL pot-1, 100 mL pot-1 and 180 mL pot-1. PC were treated 2.5 g pot-1 of 0.5 % avermectin. The experiment was performed with ve replicates and data was the average of replicates. The bar means standard deviation.