Alternaria isolates used in this study are described in Table 1 and were selected from the culture collection of Dr. Carmit Ziv at the Volcani Institute. After single spore isolation, Alternaria spp. isolates were identified based on conidial morphology (Simmons, 2007) and DNA sequencing following previously used methods (Krasnow et al., 2024b). Strains were maintained on potato dextrose agar (19 g/L, Formedium, Swaffham, UK) at 20°C in constant darkness or in glycerol (15%) at −20°C or −80°C. Conidia for germination and infection experiments were harvested from 7-10-day-old cultures by adding sterile distilled water to a plate, dislodging conidia with a sterile spatula, filtering the suspension through 4 layers of cheese cloth to remove mycelia, and adjusting the conidial concentration to 10⁵/mL with a haemocytometer (Reichert HS-1483, Buffalo, NY, USA).

For all experiments, sweet red pepper ‘Lai Lai’ were grown according to standard practices in the region for fertilizer and insect management at the Zvi Research Station (31°59′49.0″N, 35°27′09.3″E). After harvest, at green (unripe mature) or red (ripe, <5% green) stage, the peppers were stored in commercial conditions of 7°C and 95-98% relative humidity (RH) at the Volcani Institute Department of Postharvest Science for a maximum of 1 week. Before inoculation, peppers were removed from cold storage, rinsed in sterile distilled water, and left to air dry on a laboratory bench.

To determine the effect of fruit ripening stage, isolate and temperature on fungal pathogenicity green and red fruit were inoculated in three locations at the shoulders of the fruit with 12 μL droplets of conidia suspension (10⁵/mL) and placed at 7°C (cold storage conditions) or 22°C under constant darkness. The fruit were wounded immediately prior to inoculation by inserting a sterile scalpel 1 mm into the fruit and turning 90° to make a small epidermal wound. Sporulation intensity and lesion size were recorded 14 days post inoculation (dpi). Sporulation was rated on a visual scale where; 0 = no sporulation, 1 = slight sporulation, and 2 = heavy sporulation. There were 3 inoculation points per fruit, five fruit per rep, and the experiment was conducted three times. At the end of the study, the fruit from 7°C were removed to 22°C for 7 days for further evaluation. Fruit were considered infected if a visible sunken lesion or mycelial growth was noted at the point of inoculation.

To determine the effect of fruit extracts on fungal growth, recently harvested (<3 days) sweet pepper were selected from storage. Fruit were washed in tap water and sterile distilled water, sectioned, and fruit pieces (250 g) were blended in distilled water (1:1 w/v) on ice and filtered through 4 layers of cheese cloth. The extract (50 mL) was dispensed into 6-125 mL acid-washed flasks, autoclaved, and a single 3-mm plug from the margin of an actively growing colony was added to each flask. After 7-day incubation in the dark at 7°C or 22°C the mycelia were filtered, washed three times with sterile water, dried at 60°C for 24 h, and weighed.

For testing conidial germination and hyphal growth on solid media fruit extracts were supplemented with 1.5% agar, autoclaved, and poured into 9 cm petri dishes. For conidial germination tests, three 12-μL droplets of conidia suspension (10⁵/mL) were streaked onto the solid media and germination rate (%) was recorded after 18 h in complete darkness at 7°C or 22°C. This experiment was conducted three times. To determine hyphal growth and colony morphology on extract media, agar plugs from 7-10-day culture was inoculated on solid fruit extract media and incubated for 7 days at 7°C or 22°C before evaluation.

To determine conidial germination and appressoria formation fruits were inoculated on unwounded fruit surface with a 12 μL drop of conidial suspension (10⁵/mL) and placed in clear plastic humid chambers (~100% RH). The fruit were incubated at ambient temperature (22 ± 2°C) under intermittent fluorescent light (12/12 h), to simulate retail conditions. After 24, 48, and 72 hours post inoculation (hpi), the epidermal layer was carefully removed with a scalpel and fixed in Carnoy’s fluid (acetic acid:ethanol, 1:3 v/v). Conidia were stained with toluene blue in lacto-phenol and the epidermal peel was mounted in glycerol:water on a cleaned microscope slide. Conidial germination and appressoria formation were counted for 50 conidia. Conidia were counted as germinated if the emerging germ tube was longer than 0.5 × the length of the conidial body. In the experiments, each treatment was tested with 4 fruits and 6 conidial droplets per fruit. A subset of fruit remained at 22°C after 72 h and were incubated for 5 days to determine if further infection developed.

For examination of infection characteristics of black mold, green and red pepper fruit were removed from cold storage within 1 week of harvest and inoculated with isolate 94CZ on wounded epidermal surface as described above. In order to describe internal infection processes (Jones et al., 1974), fruit were inoculated with 1 mL of conidial suspension injected into the fruit cavity with a sterile syringe. Control fruit received sterile distilled water. Fruit were opened, the diffusate poured off, and 1 × 1 × 2 mm tissue sections excised at 0, 6, 8, 12, 24, 48, 72, and 92 hpi and fixed in formalin, acetic acid, alcohol, and water (10:5:50:35 v/v/v/v), and dehydrated through a tertiary butyl alcohol series (Jensen, 1962). The sections were embedded in paraffin (melting point 52°C), and 12 μm sections were made using a rotary microtome (Leica RM2125 RTS, Nussloch, Germany). Sections were affixed to glass microscope slides and stained with safranin and fast green to differentiate cell walls, cellular components, cuticle, and fungal hyphae. For each time point, samples from 4 fruit were prepared and ~10 sections from each sample were observed. Photomicrographs were taken with a Leica microscope camera (Leica). The terminology for pepper cell structure, giant cells, and other organelles was used as described previously (Kim et al., 1999). Giant cells are the elongated, oversized cells present at the inner pericarp layer of sweet pepper cultivars.

Analytical standards of ALT, AME, and AOH were obtained from Fermentek (Jerusalem, Israel). These mycotoxins are commonly recovered from fruit infected by small spored Alternaria spp. and have known toxicity in mammals (Chen et al., 2021). To extract mycotoxins from infected peppers, pepper fruit were inoculated with the spore injection method as described above. The peppers were sliced open and the majority of healthy tissue, placenta, and seeds removed. Diseased pepper tissue was frozen at −80°C, lyophilized, and ground to a fine powder. A salts mixture of 4 g MgSO₄, 1 g NaCl, and 5 mL 1% formic acid in ethyl acetate (QuChERS method) was added to the samples, vortexed for 5 min and centrifuged for 10 min at ×8,586 g. Finally, the supernatant was evaporated to dryness under a nitrogen stream at 50°C. Dried samples were reconstituted in 100% methanol, filtered using a 0.22 μm PTFE membrane filter, and a 20 μL sample was injected into a reverse-phase UHPLC system (Waters ACQUITY ArcTM, FTN-R, Milford, MA, USA). The samples were analyzed in a linear gradient with a flow rate of 0.5 mL/min through a Kinetex 3.5 μm XB-C18 (150 × 4.5 mm) column (Phenomenex, Torrance, CA, USA). The gradient was as follows: 0.005% of TFA in water (A):0.005% of TFA in methanol (B):0.005% of TFA in acetonitrile (C), following the progression of 90% A:10% B (0-5 min); 30% A: 60% B: 10% C (6-12 min); 90% A:10% B (13-17 min) at 40°C. The Alternaria toxins peak curve was detected with a fluorescence detector (excitation 330 nm and emission 430 nm) and quantified by comparison with the calibration curves of standard mycotoxins to an accuracy of 0.01 μg/g. Standard stock was prepared at 1 mg/mL in 100% methanol and stored at −20°C. The working standard mycotoxins were prepared in methanol dilution with 1 g healthy uninfected pepper (red or green) that were spiked with different concentrations (0.05, 0.1, 0.5, 0.8 μg/g) of Alternaria mycotoxin mix (ALT, AME, and AOH) and analyzed for mycotoxin recovery (%).

Data were analyzed using JMP Pro Software version 16 (JMP, Cary, NC, USA). Differences among treatment means and conidial germination by fruit color were analyzed with analysis of variance (ANOVA) and Tukey’s honestly significant difference test (P = 0.05). Differences among treatments based on data taken as fruit infection for individual fruit were analyzed for each color. Residual data was assessed for normality and the Levine test for homogeneity was performed.

In order to determine the infection process of green vs. red pepper with Alternaria and to assess the relation of infection to the increasing incidence of postharvest black mold in peppers, we characterized the infection process. Initial symptoms on inoculated red fruit included small sunken brown to black lesions with fluffy white aerial mycelium or black, felty sporulation, regardless of temperature tested (Fig. 1A and B). Lesions continued to grow until significant portions of the fruit were infected. These symptoms were similar to those observed on diseased fruit collected from storage facilities (Supplementary Fig. 1). On green fruit, initial symptoms included small sunken brown to black lesions with slight white aerial mycelium or black sporulation. Lesions tended to stay small and confined, and did not expand rapidly (Fig. 1C and D).

Pathogen growth on peppers was significantly denser on red and green fruit at 22°C than 7°C (Table 2). The majority of inoculation points developed into lesions. The lesions were typical for black mold at both temperatures and included water-soaked lesions and gray mycelial growth followed by dark brown to black sporulation (Fig. 1A). At 7°C lesion development was relatively slow, and sporulation did not occur until approximately 10-14 dpi. Septate spores typical of Alternaria were present in the lesions that formed at both temperatures and the pathogen was successfully reisolated. The average incidence of infection for all isolates on red and green fruit at 7°C was 90% and 97%, and at 22°C was 100% and 88%, respectively (data not shown). All five isolates of Alternaria were pathogenic on both green and red pepper fruits in both temperatures. Significant growth and sporulation was evident on red fruit at 22°C, while at that same temperature growth on green fruit was evident but somewhat reduced. As expected, lesion size at 7°C was much smaller than at 22°C, although, in three out of the five isolates was similar on both green and red fruits (94CZ, 136CZ, and 307CZ). Sporulation was much less affected by maturity stage or temperature for these isolates. Isolate 135CZ was significantly slower in infecting green fruit and had limited sporulation on green fruit at 7°C. While different isolates showed variability in the development of black rot, all tested Alternaria were pathogenic on green fruit at ambient temperature as well as in cold conditions (Table 2). It is noteworthy that for all isolates tested, when infected red and green fruit were brought to 22°C after cold storage (7°C), lesions expanded rapidly. Red fruit became readily infected with secondary pathogens including B. cinerea or Penicillium expansum.

Fruit inoculated with the injection of conidia into the fruit at 22°C appeared healthy until ~72 hpi when a discolored region was observed on the outer fruit surface in the area of the infection. When red fruit were opened, the internal pericarp appeared water soaked at the point of infection 24 hpi and mycelial growth was visible by 48 hpi. By later stages of infection, sporulation on the inner pericarp was noted, and often the infection spread to placenta tissue and seeds that displayed sporulation. Green fruit also rotted, although, fungal hypha and sporulation developed slower than in red fruit (Supplementary Fig. 2). The lesions were moderate in size and did not expel liquid when pressed gently.

At suboptimal cold storage of peppers (2°C), lesion development was slow and sporulation was suppressed on both green and red fruits (Fig. 2A and B), although lesion formation was clear and evident already after 14 days. Notably, green peppers were much more susceptible to chilling injuries manifested as pitting and did not show resistance to Alternaria infection.

To compare the growth of Alternaria on extracts of green and red fruit, extracts were collected and used as a growth medium. Temperature had a significant effect on fungal growth on pepper extracts (P > 0.05) (Supplementary Table 1, Supplementary Fig. 3). At 7°C there was no difference in mycelial dry weight between fruit colors. At 22°C, there was significantly more mycelial dry wt. from red (mean 67 mg) than green (mean 42 mg) peppers (Supplementary Table 1). On solid media, there was a slight increase in growth on green than red pepper extracts at 7°C and 22°C, however, the effect of fruit color was not significant. However, sporulation was noticeably denser on red extracts than green (Supplementary Fig. 3). At 22°C, conidial germination was the same on red and green extract media by 24 h (100%). While at 7°C, germination on red and green extracts were 22% and 11%, respectively, at this time point (Supplementary Table 1).

To test the effect of fruit ripening stage on the initial stages of infection, A. alternata strain 94CZ was selected for further analysis. To characterize penetration of the pathogen into the fruit, conidial germination and appressoria formation were examined. Red and greed fruit were inoculated on unwounded surface and evaluated at different incubation temperatures (simulating cold storage and retail shelf life conditions). Conidial germination occurred at a similar rate on red and green peppers at the times and temperatures tested (Fig. 3A-C). Conidial germination was above 90% at all time points on both green and red peppers incubated at 22°C. At 2°C and 7°C, germination was much slower than at 22°C on both green and red fruit. At 7°C conidia germination reached 95% after 48 h, and germination was above 40% at 2°C for red and green fruit (Fig. 3A and B). Germ tubes were simple or branched. Germination became difficult to measure at 48 and 72 h due to heavy mycelial growth.

Appressoria were produced significantly more on red than on green fruit at 22°C, but was similar on both fruit types at 7°C or 2°C (Fig. 3D-F). On green fruit at 48 h more appressoria were produced at the warmer temperatures (Fig. 3D-F). The appearance of appressoria was similar on both fruit colors (Fig. 4A and B).

Collectively these experiments indicated that black mold caused by A. alternata could develop on mature unripe green fruit as well as on ripe red fruit at cold storage and retail temperatures. However, while pathogenicity on green fruit was somewhat delayed at retail temperature, when incubated at low temperatures, fruits at both ripening stages were similarly susceptible to the disease.

To better characterize the infection process of Alternaria on green and red fruit, a detailed microscopic evaluation was performed. When red fruit were inoculated on the wounded cuticle, infection appeared as a slightly sunken area by 24 hpi. Fruit inoculated with water remained unchanged (Fig. 4C and E). Conidia were observed germinating on the cuticle of red fruit (Fig. 4D) and germ tubes penetrated via the artificial wound. Mycelia grew intercellularly and accumulated beneath the wound site. In green fruit, the mycelia grew ~5 cell layers deep (Fig. 4F). Mycelia stained densely and a potential host response occurred near the epidermal layer, and nuclei appeared enlarged after inoculation. At later stages of infection in red fruit the cuticle layer did not pick up safranin stain, suggesting dissolution of the cuticle (Fig. 4D). The Alternaria isolated from these lesions resembled the original cultures.

Disease progression in fruit inoculated with injection of conidia was similar between ripeness stages, although, the rate and magnitude was slower in green fruit. For red fruit, penetration of the giant cells of the pericarp occurred by 8 hpi (Fig. 5B). Mycelia were observed growing intercellularly in the outer cell layers of the inner pericarp (Fig. 5C). By 12 and 24 hpi, ramification of hyphae through the first layer of cells was more apparent and mycelial growth was frequent in the giant cells (Fig. 5C). Cell breakdown was observed by 48 hpi and at 72 hpi complete cell dissolution at the infection site occurred (Fig. 5D). Nuclei in the pericarp and inner cell layer appeared swollen once infection progressed to these cells (Fig. 5B). At later stages of infection, complete dissolution of cells was observed (Fig. 5D). Vascular bundles in the pericarp (Fig. 6A) were affected by the pathogen with cell breakdown, disorganization and loss of structural integrity (Fig. 6B). Mycelia and occluding substances filled the vascular bundles. Darkening around the margin of the lesions was distinct on infected fruit. In green fruit, the giant cells of the inner pericarp of control sections appeared similar to those in red fruit (Fig. 5E). Mycelial growth was present but not as dense 24 hpi as in red fruit (Fig. 5F). In some sections from diseased green pepper, thickening of cell walls was observed in the pericarp layer, and cellular debris was noted (Fig. 5G). Cell wall disorganization and dissociation was apparent, but less noticeable in green fruit (Fig. 5G), and vascular bundles did not have noticeable cellular damage (Fig. 6C and D). The disruption of the cells was slower in green fruit, even at 72 hpi significant dissolution was not observed (Fig. 5H). However, heavy mycelial growth occurred in the outer cell layers at this point in the infection process (Fig. 5H).

To determine the effect of host fruit ripening stage on the mycotoxinogenic potential of A. alternata, mycotoxin analysis of ALT, AME, and AOH were performed for infected fruits. Each of the three mycotoxins was produced by all five isolates tested, although some variability was apparent between experiments (Table 3). The quantity of AOH varied greatly among replicates and fruit colors. In the first replicate of the experiment, the content of AOH in infected fruit was significant for red pepper while none of the isolates produced AOH in infected green fruit. Nevertheless, AOH was detected on green infected fruit in the other 2 replicates. For ALT, the range of mycotoxin recovered was 0-0.61 and 0-0.35 μg/g for replicate 1 and 2, respectively (Supplementary Table 2). In replicate 3, ALT was not recovered from red peppers. For green peppers, the range of AME content was 0.019-7.29 μg/g where detected (Supplementary Table 2). For red peppers, the range was 0.06-65.35 μg/g. Mycotoxin content varied in green compared to red fruit although, on average, the differences were not statistically significant (Supplementary Fig. 4). In general, mycotoxin content in infected fruit was affected by replicate and isolate, but not by ripening stage.