The HSCCC instrument was an OptiChrome-300 PLUS high-speed counter-current chromatograph system (Counter-current Technology, Jiangyin, China). The apparatus had a rotational speed of 400–1,200 rpm, with a centrifugal force of up to 110 g, and was equipped with two polytetrafluoroethylene (PTFE) multilayer coils (210 m × 1.6 mm id) with a total capacity of 300 mL and a 30 mL manual sample loop (Wang et al., 2019). The solvent was pumped into the column with a CF-80 pump (Counter-current Technology, Jiangyin, China). The effluent was detected at 318 nm with a UV2000D ultraviolet detector (Sanotac Scientific Instruments, Shanghai, China). An EasyChrom-1,000 chromatography workstation (Hanbon Science and Technology, Huaian, China) was used to record the chromatogram.

Analytical HPLC was performed using a Shimadzu LC-20A system (Shimadzu, Kyoto, Japan) equipped with two Shimadzu 6 AD pumps, an analytical Agilent ZORBAX SB-C18 column (4.6 × 250 mm, 5 μm), and an SPD-M20A diode array detector with UV detection at wavelengths of 190–500 nm. High-resolution time-of-flight mass spectrometry (HR-TOF-MS) for the isolated compound was performed using a Triple TOF 5600+ instrument (AB SCIEX, America). The nuclear magnetic resonance (NMR) spectra ¹H NMR and ¹³C NMR were recorded using a BRUKER AVANCE 3 instrument at 400 MHz and 100 MHz, respectively. The compound was dissolved in dimethyl sulfoxide-d₆ (DMSO-d₆) with 0.03% tetramethylsilane (TMS) as the inner standard.

All organic solvents used for the preparation of crude extract, column chromatography, and HSCCC separation were of analytical grade (Sinopharm Chemical Reagent, Shanghai, China). For HPLC, the mobile phase acetonitrile was HPLC grade (TEDIA, Anhui, China) and the water was purified using a Milli-Q water purification system (Millipore, United States).

The bacterium strain L. enzymogenes OH11 used in this study was originally isolated from the rhizosphere of a green pepper plant in our laboratory and deposited at the Center of General Microbiological Culture Collection (CGMCC No. 1978) in China (Tang et al., 2018a,b; Tang et al., 2019). The plant pathogens used in the bioactivity assay, including Fusarium graminearum, Colletotrichum fructicola, Alternaria alternata, Botryosphaeria dothidea, and Valsa pyri, were preserved in our laboratory (Zhao et al., 2019; Li et al., 2021).

The HSAF crude extract was prepared using the selected fermentation conditions and the macroporous resin adsorption/desorption method according to our previous study (Tang et al., 2018a,b; Tang et al., 2021). Briefly, the L. enzymogenes OH11 strain activated on a solid Luria-Bertani (LB) medium was inoculated into an LB liquid medium and incubated at 28°C for 12 h with shaking at 180 rpm as the seed culture. Then, the seed culture was inoculated (2% v/v) into the defined SGC medium (8.00 g/L soybean flour, 7.89 g/L glucose, and 0.72 g/L calcium ion) and incubated for 60 h. Next, the macroporous resin NKA (15 g/L) was added into the fermentation broth and the culture resumed shaking for another 12 h, and most of the HSAF produced in the broth was adsorbed into the resin. The sample solutions were filtered using gauze, and the resin was washed with ethanol (shaken for 2 h and repeated three times) to desorb HSAF.

To improve the HSAF resolution in the following separation procedure, the desorption solution was pretreated to remove impurities. The ethanol desorption eluent was placed under an ultraviolet germicidal lamp (UV wavelength at 254 nm) for 6 h, then it was concentrated using a rotary evaporator. The brown gum of HSAF extract was subjected to Sephadex LH-20 gel column chromatography and eluted with dichloromethane-methanol = 1:1. The fractions containing HSAF were combined and evaporated to dryness.

The selection of a suitable immiscible two-phase solvent system for the target compound is the most important step in HSCCC, and it is estimated to account for approximately 90% of the entire work (Ito, 2005). In general, the two-phase solvent system was selected according to the partition coefficient (K value) of the target component. The K value for the target compound was determined as follows: a small amount of crude sample (10 mg) was added into a test tube containing approximately 20 mL of a two-phase solvent system. After shaking vigorously, the contents were mixed thoroughly. Then the equilibration was established and the two phases were analyzed separately by HPLC. The peak area of the target compound in the upper phase was recorded as A_U, and that in the lower phase was recorded as A_L. The K value was calculated according to the equation K = A_U/A_L. For complex compounds that had two or more peaks, the separation factor (α = K2/K1, where K2 > K1) between them was also calculated.

In this study, the two-phase solvent system containing n-hexane–ethyl acetatemethanol–water (3:5:4:5 v/v/v/v) was selected for HSCCC separation. The solvent mixture was thoroughly shaken in a separation funnel and the two phases separated quickly. After setting overnight, the two phases were separated into two bottles, and an acidic modifier (0.1% trifluoroacetic acid, TFA) was added to the lower phase. The total volume of the two-phase solvent system was 1,020 mL, and the volume of the upper and lower phases was approximately 400 mL and 620 mL, respectively. The two phases were then treated with an ultrasonic cleaner to remove the bubbles. The 100 mg HSAF crude sample was dissolved in a 10 mL solvent mixture consisting of equal volumes of both the upper and lower phases, and the sample solution was ready for loading.

In the HSCCC procedure, the upper phase acted as the stationary phase and was pumped into the coiled column at a flow rate of 20 mL/min. After the column filled with the stationary phase, the lower phase acted as the mobile phase and was pumped into the column at a flow rate of 2.0 mL/min with the head-to-tail elution model, and at the same time, the apparatus was rotated at 1000 rpm and the temperature was kept at 25°C. After the mobile phase was seen flowing out, a hydrodynamic equilibrium was established. Then, 10 mL of the sample solution was injected into the system. The effluent was monitored with a UV detector at 318 nm and the fractions were manually collected according to the profiles of the chromatographic peaks. The target fractions were evaporated using an electric fan under the air stream to partially remove the organic solvent. The obtained precipitate was collected and evaporated using the rotary evaporator at a water-bath temperature of 50°C under reduced pressure, and the remaining solvents including water and TFA were removed. The remaining products were dissolved in methanol for HPLC analysis.

Qualitative and quantitative analysis of HSAF in the crude extract or in the HSCCC fraction was performed using the Shimadzu LC-20A HPLC system with an isocratic elution mode. The mobile phase was water–acetonitrile (45:55, both containing 0.04% TFA, v/v) at a total flow rate of 1.0 mL/min. The effluent was monitored by a UV detector at 318 nm. The pure standard of HSAF was obtained using the method of macroporous resin adsorption/desorption combined with preparative HPLC as in the previous study (Tang et al., 2018a,b; Tang et al., 2021). The standard compound was diluted into 5, 10, 50, 100, 200, 400, and 500 μg/mL with methanol and injected into HPLC to establish the linear regression equation. The concentration of HSAF in the analyzed solution was quantitatively determined by linear regression using the equation Y = 30,885X − 29,077 (R² = 0.9920), where X is the concentration of HSAF (μg/mL) and Y is the absorption peak area of HSAF. The purity of HSAF was calculated as (the weight of HSAF/the weight of sample) × 100%, and the recovery rate of HSAF in the HSCCC procedure was calculated as (the weight of HSAF obtained after HSCCC/the weight of HSAF before HSCCC) × 100%.

The chemical structure of the isolated compound was confirmed by HR-TOF-MS, ¹H NMR, and ¹³C NMR. The measured spectroscopic data were compared with the reported spectroscopic data (Graupner et al., 1997; Li et al., 2012). In addition, the antifungal activity of the isolated compound was determined using the methods described in our previous studies (Zhao et al., 2019; Li et al., 2021). Briefly, the compound was diluted with DMSO to create different gradient solutions and 0.5 mL of each solution was added into 100 mL of half-melted (approximately 45°C) PDA medium to yield a final concentration of 0.1, 0.2, 0.5, 1, 2, and 5 μg/mL. The same volume of DMSO was added as the control. Then, the center of the solidified medium was inoculated with a 5 mm mycelial plug that was cut from the edge of a fresh fungal colony. The plates were incubated at 28°C for 5 days and there were three replicates in the experiment. The inhibition rates for the gradient solutions were calculated, and the EC₅₀ values (the concentration inhibiting 50% of mycelial growth) were calculated using probit-log analysis. Three independent experiments were conducted for the bioactivity assays, and the EC₅₀ value was expressed as “mean ± standard deviation.” As a control HSAF, the pure standard compound of HSAF was prepared by prep-HPLC in our laboratory (Tang et al., 2021), and it was tested against the same five plant pathogens.

The biocontrol bacterial strain Lysobacter enzymogenes OH11 (CGMCC No. 1978) was demonstrated to produce a high yield of HSAF in our laboratory (Tang et al., 2018a,b; Zhao et al., 2021). In this study, L. enzymogenes OH11 strain was inoculated in the defined SGC medium and cultured for 60 h, and then HSAF was effectively adsorbed onto the macroporous adsorption resin NKA. The resin was washed with ethanol to obtain the initial crude extract. After evaporating to dryness, the initial crude extract was obtained as a dark brown gum, and the purity of HSAF in the crude extract was approximately 28.6%. As a result, there were many analogous compounds and impurities in the crude extract. In particular, many compounds were analogous to HSAF such as the main interferent alteramide B (ATB) (Figure 1) (Li et al., 2012; Xu et al., 2015; Ding et al., 2016a,b), and both HSAF and ATB have similar structures and close retention times in HPLC. In the preliminary experiment, the two compounds were usually difficult to separate using prep-HPLC or HSCCC. In order to obtain a better resolution for HSAF in the following procedure, pre-treatment for the initial crude extract was necessary.

In our previous research (Tang et al., 2021), we found that after exposing the fermentation broth to high-intensity fluorescent light for 2 days, most of the ATB was degraded, which could help acquire high purity of HSAF. However, this illumination time was slightly longer, and the effect of different kinds of light on the ATB degradation was unknown. In this study, the ethanol desorption solutions eluted from resin NKA were illuminated with three different light sources, including white light (wavelength at 400–760 nm), ultraviolet germicidal light (wavelength at 254 nm), and black light (wavelength at 365 nm). Furthermore, the ethanol solutions were analyzed by HPLC. The results in Figure 2 show that, after being illuminated with white light for 2 days, the peaks of HSAF and ATB in the ethanol solution were almost the same as those in the unilluminated solution. While under ultraviolet light at 254 nm, the peak of ATB disappeared and HSAF was still present. When exposed to black light at 365 nm, both the ATB and HSAF peaks disappeared. These results showed that ultraviolet light (254 nm) was more effective than white light in eliminating the interference of ATB. Moreover, black light at 365 nm was harmful to both ATB and HSAF. This may be a result of the 5/5/6-tricyclic system being destroyed under black light, as this system is critical to the stability of polycyclic tetramate macrolactam compounds (Huffman et al., 2010; Tang et al., 2021).

Furthermore, different times of ultraviolet light exposure were evaluated, and the results are shown in Figure 3. As can be seen from the HPLC results, the peak of ATB decreased with the illumination time setting from 0 h to 8 h, and we found that an illumination time of 6 h was enough for the ATB degradation. Based on these results, illumination with ultraviolet light for 6 h was proposed for eliminating the interference of ATB in the pretreatment of crude extract. According to the peak area measured by HPLC, the content of HSAF decreased by approximately 15.7% after exposure to ultraviolet light. Compared with our previous research, the illumination time was shortened from 2 days to 6 h.

In order to remove the impurities that could not be detected by the HPLC UV detector, column chromatography on Sephadex LH-20 was necessary. Using an organic phase (such as dichloromethane-methanol = 1:1 herein) as the isocratic elution solvent on the Sephadex LH-20 column can prevent the irreversible adsorption of the samples, and this method has been widely used for the separation and purification of natural products (Mottaghipisheh and Iriti, 2020). In this study, after illumination with ultraviolet light for 6 h, the HSAF extract was repeatedly loaded on the column using dichloromethane-methanol = 1:1 as the elution solvent. The effluent fractions containing HSAF were combined and concentrated. After pretreatment of the crude extract with UV light and column chromatography, the purity of HSAF in the extract reached approximately 45.8%, and this was favorable to the next HSCCC procedure.