First record of Cladosporium oxysporum as a potential novel fungal hyperparasite of Melampsora medusae f. sp. deltoidae and screening of Populus deltoides clones against leaf rust

Abstract

Melampsora medusae f. sp. deltoidae is causing serious foliar rust disease on Populus deltoides clones in India. In the present study, a novel fungal hyperparasite on M. medusae has been reported. The hyperparasitic fungus was isolated from the uredeniospores of the rust fungi and identified as Cladosporium oxysporum by morphological characterization and DNA barcode technique based on the Internal Transcribed Spacer (ITS) region of nrDNA and beta-tubulin (TUB) gene region. Hyperparasitism was further confirmed through leaf assay and cavity slide methods. Leaf assay method showed no adverse effect of C. oxysporum on poplar leaves. However, the mean germination percentage of urediniospores was significantly decreased (p < 0.05) in the cavity slide method when a conidial suspension (1.5 × 10⁷ conidia per ml) of C. oxysporum was applied in different deposition sequences. Scanning and light microscopic observations were made to explore the mode of action of the hyperparasitism. The antagonistic fungus vividly showed three different types of antagonism mechanisms, including enzymatic, direct, and contact parasitism. Alternatively, by screening 25 high-yielding clones of P. deltoides, five clones (FRI-FS-83, FRI-FS-92, FRI-FS-140, FRI-AM-111, and D-121) were enlisted under highly resistant category. Present study revealed an antagonistic relationship between C. oxysporum and M. medusae, which could be an effective method of biocontrol in field plantations of poplar. Combining this biocontrol approach with the use of resistant host germplasm could be an environment friendly strategy for preventing foliar rust and increasing poplar productivity in northern India.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13205-023-03623-x.

Introduction

Poplar (Populus deltoides Bartr.) is a dioecious multipurpose tree species of high asset value which is widely used in agro-forestry programs in India (Kumar and Singh 2012; Przybysz and Przybysz 2013). Because of the vulnerability to several diseases, its health and resultant productivity are the major concerns for agro-forestry practitioners. In nursery, high incidence of pathogens causes 0.14–29.25% mortality (Singh et al. 2012). Poplar Leaf rust caused by Melampsora spp. is one of the most devastating diseases worldwide leading to premature defoliation and significant decline in biomass of poplar (Eberl et al. 2018; Wan et al. 2013). Around 80–100 species have been described under the genus Melampsora Castagne and of those, 9–32 have been associated with poplar trees (Hiratsuka 1982). So far in India, records of four species of Melampsora have been made on poplar trees: M. ciliata, M. populnea, M. rostrupii, and M. medusae (Gautam and Avasthi 2019).

As a management strategy, various chemical and cultural approaches are being applied to control the rust infection. However, chemicals are hazardous to the environment and less cost-effective in field application. Likewise, cultural approaches are also not feasible over large plantation areas affected by Melampsora rust. Thus, greater emphasis is being laid on developing environment friendly approaches that are effective in disease management with minimal cost. Naturally occurring antagonists play a vital role in defense against rust diseases in poplar.

The genus Cladosporium Link has the most promising species that can hyperparasitize rust pathogens (Moricca and Ragazzi 2008). The behavior and characteristics that drive mycoparasitism can be seen in Cladosporium, suggesting that these traits have been conserved and passed down throughout the evolutionary history. Mycoparasitism occurs when fungi feed on and parasitize other fungi, and this behavior has been observed in some species of Cladosporium. The ability to exploit other fungi for their nutrients is a trait that has been maintained over time, indicating the evolutionary importance of this behavior (Barge et al. 2022). Some of the most common species, such as C. tenuissimum, C. herbarum and C. oxysporum were extensively studied for their antagonism with M. larici-populina and M. ciliata (Sharma and Heather 1981; Sharma et al. 2002). Remarkably, C. aecidicola is the only species previously reported to be associated with M. medusae (Sharma and Heather 1980). Other studies have also highlighted the potential of Cladosporium species to act as mycoparasites on rust pathogens in crops (Gregersen and Smedegaard 1989; Christiansen and Smedegaard 1990; Zhan et al. 2014; Zhang et al. 2022), flowering plants (Barros et al. 1999; Torres et al. 2017) and conifers (Moricca et al. 2001).

Besides, deployment of resistant clones is also an alternative way to combat the prevalence of rust infection in field plantations (Isebrands et al. 2007). Breeding programs for host plant resistance rely on disease symptom severity from natural infection in the field. Assessing resistance under natural infection is always useful to practice because it is not always influenced by variable environmental conditions and uneven inoculum exposure. However, there is a need to screen poplar clones against major rust pathogens followed by their evaluation under different agro-climatic conditions and deployment in field plantations. Limited studies have been carried out for screening of P. deltoides clones against M. medusae in India (Dhiman 2022).

In view of this, the present study was conducted to identify and characterize a novel fungal hyperparasite of Melampsora rust in P. deltoides to be used as a potential biocontrol agent for its commercial application in future after field evaluation. In addition, efforts were also carried out to select the rust resistant clones of P. deltoides by screening the germplasm in natural conditions.

Materials and methods

Leaf symptomatology, isolation and microscopic characterization

Leaves with orange uredinia (rust infected leaves) of P. deltoides colonized by white dusty fungal mycelia were collected in October 2019 from Forest Research Institute (FRI), Dehradun, Uttarakhand, India (N30°20′; E78°00′). Isolation of the fungi associated with rust uredeniospores was done by cutting the infected areas of leaf into small pieces, surface sterilizing with 70% ethanol and transferring into PDA (Potato Dextrose Agar) media. The plate was then incubated for seven days at 25 °C.

For identification of the isolated fungi, morphological observations were studied using standard keys given by Wirsel et al. (2002) and Bensch et al. (2012). The size of conidiophores, conidia and ramoconidia were measured by averaging 50 measurements. Microscopic identification of Melampsora rust fungi urediniospores as well as paraphyses features, viz., shape, structure, length, width, and echinulation pattern were recorded. The size of urediniospores and paraphyses was examined under a light microscope. The features were further confirmed with published taxonomic descriptions of Melampsora (Spiers and Hopcroft 1994).

Molecular characterization

For each fungal species, DNA extraction was performed by using protocol given by Liu et al. (2000). Rust uredinia from infected leaves of poplar were scraped and the DNA was extracted. Extraction of DNA from antagonist, a small lump of mycelia from PDA culture was used. The DNA barcode regions, namely Internal Transcribed Spacer, ITS-1 and ITS-2 along with 5.8S rDNA, were amplified by universal primers ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) and ITS5 (5′-GGAAGTAAAAGTCGTAACAAGG-3′) (White et al. 1990) for Melampsora as well as Cladosporium. Along with ITS, an additional DNA barcode gene, i.e., beta-tubulin (TUB2), was used to obtain species level resolution of Cladosporium, which was amplified using TUB2 gene primers T1 (5′-AACATGCGTGAGATTGTAAGT-3′) and T22 (5′-TCTGGATGTTGTTGGGAATCC-3′) (Zalar et al. 2007). The polymerase chain reaction (PCR) was carried out under 50 µl reaction mixture containing 1X Taq buffer, 1.5 mM MgCl2, 0.05 mM of dNTPs, 0.4 mM primer and 5 unit of Taq DNA polymerase (GeNei™). The cyclic condition for PCR included an initial denaturation at 94 °C for 5 min, followed by 30 cycles of denaturation at 94 °C for 1 min, annealing at 55 °C for 1 min and extension at 72 °C for 1 min, followed by final extension at 72 °C for 8 min, and final hold at 4 °C.

The PCR products were purified and sequenced through Sanger’s dideoxy method by Biokart India Pvt. Ltd., Bengaluru, India. Raw sequence data were subjected to the quality check and the low-quality bases were manually edited using BioEdit sequencer alignment editor v.7.0.9 (Hall 1999). Further, the sequence reads generated with forward and reverse primers were assembled into a single contiguous sequence. The ITS gene sequences of rust fungi, and combined ITS and TUB2 gene sequences of Cladosporium were aligned with other homologous sequences available in GenBank database (retrieved from the NCBI), using the CLUSTALW program in MEGA X 10.1. Maximum Parsimony (MP) tree for Melampsora and Maximum Likelihood (ML) tree for Cladosporium were generated with 1,000 bootstrap replications.

Confirmation of hyperparasitism

The hyperparasitism was ascertained by two methods. The detail of the methodology and experiments is given as follows:

Leaf assay method: to confirm the parasitism between rust and the antagonist, moist chamber technique described by Torres et al. (2017) was applied. The fresh leaves and leaves infected with urediniospores were collected from glass house. The leaves were disinfected for 3 min with a 3% sodium hypochlorite solution, and then washed three times with autoclaved distilled water. There were three deposition sequences: (1) conidia of antagonist were applied to uredinia-covered leaves; (2) conidia of antagonist were applied on healthy leaves; and (3) control, rust infected leaves treated in the same way with autoclaved distilled water. Each treatment was applied on ten leaves. The pure culture of Cladosporium was prepared as a spore suspension (1.5 × 10⁷ conidia per ml) and inoculated on diseased and healthy leaves of poplar in Petri plates. The plates were then incubated at 24 ± 1 °C. After 96 h of incubation, percent parasitism was measured within a 5 cm² area by counting the number of uredinia. The significance of difference in mean percentage under various treatments was tested by Analysis of Variance (ANOVA). To test the normality of the data, Shapiro–Wilk test was used and homogeneity of variance was tested by Levene’s test. The level of significance was taken as 0.05. When the differences were significant, post hoc Duncan’s multiple range tests was used to test pair-wise differences. All statistical analysis was carried out using MS Excel and SYSTAT 13.2.

Cavity slide method: Freshly collected 5 mg urediniospores were inoculated with 25 µl conidial suspension (1.5 × 10⁷ conidia per ml) of the Cladosporium in a cavity slide as method standardized by Sharma et al. (2002). Four types of deposition methods were used to determine the germination of urediniospores (Table 1). Slides were observed under light microscope after 12 and 24 h. The experiment was carried out with four replications of each treatment. Germination percentage was calculated based on urediniospores count by means of light microscopy. Germination was defined as a germ tube length of up to half the spore diameter. To further observe the ultra-structure of parasitism, light and scanning electron microscopy were performed. For determining significance of difference between the treatments, ANOVA was performed followed by Duncan’s Multiple Range Test for significant effects.

Table 1.

Deposition sequences of urediniospores of Melampsora medusae and antagonist’s conidia to determine the antagonistic effect by cavity slide method

Treatment	Description
T-1	Conidia were inoculated first, followed by urediniospores after 1 h
T-2	Urediniospores were inoculated first, followed by conidia after 1 h
T-3	The urediniospore and conidia were subsequently placed on a cavity slide
T-4 (control)	Urediniospore with 25 µl sterile water, no antagonist

Screening of P. deltoides clones against rust fungi M. medusae in natural field condition

In the mid October 2019, the diseases severity was recorded on 25 high yielding poplar clones in germplasm bank established at FRI, Dehradun. Five ramets of each clone were monitored in nursery for the disease resistance under natural infection conditions for two consecutive years in 2019–2020. The clones were rated through disease severity scale using descriptive key as per Elena et al. (2014) with some modifications (Table 2). To evaluate whether observed differences (diseases severity ratings) were statistically significant, analysis was performed by location and year i.e., germplasm (2019) and nursery (2019 and 2020). The disease severity score was analyzed by non-parametric Kruskal–Wallis one way ANOVA at 5% level of significance (the effect of significant when p < 0.05). When the effect was significant, pair wise comparisons were made using Conover-Iman test. All the statistical analyses were performed using the SYSTAT 13.2 software. The percentage of infection index was calculated by using following formula:

$$\text{Infection index =}\sum \frac{a\times b}{N\times K}\times 100$$

where: a = values of no. of leaves with different level of infection, b = values of the scale (0–6), N = total number of leaves per plant and k = maximum scale value.

Table 2.

The Melampsora medusae Disease Rating Scale (DRS) used to evaluate the severity of the disease and the level of resistance or susceptibility of Populus deltoides clones

Scale	Description	Infection	Category
0	No symptoms on leaf	0	Immune
1	Little yellow pustules (less than 1% of leaf area infected)	0.1–5	Highly resistant (HR)
2	Few scattered pustules, cover abaxial surface of the leaves (1–5% of leaf area infected)	5.1–10	Resistant (R)
3	Small scale yellow pustules with moderate lesion on leaves (6–15% of leaf area infected)	10.1–20	Moderately resistant (MR)
4	Heavy pustules with moderate to extensive lesion and limited necrosis on leaves (16–33% of leaf area infected)	20.1–50	Moderately susceptible (MS)
5	Extensive lesion with extensive necrosis on leaves (34–54% of leaf area infected)	50.1–70	Susceptible (S)
6	Pustules with very extensive lesion on all plant part, yellowing and complete defoliation of leaves (55–100% of leaf area infected)	> 70	Highly susceptible (HS)

Results

Field symptoms of Melampsora rust fungi and Cladosporium spp. on P. deltoides

Initially, infection of rust fungi on poplar leaves produced yellow uredinia to form on the abaxial surface of the leaves. They were usually in the form of small and circular pustules. These pustules affected the upper part of the plant before gradually spreading to cover the entire leaf surface. Urediniospores were then formed in clusters with orange-yellow uredinia which covered both sides of the leaf surface (Fig. 1a). At the acme phase, dead uredinia were found, which turned yellow. The dark brown to black crusts patches were observed on the leaves resulting in chlorosis and necrosis. The necrotic spots became larger as the infection spreads contributing to early defoliation (Fig. 1b).

Fig. 1.

Field symptoms of Melampsora medusae and Cladosporium oxysporum on leaves of Populus deltoides: a rust uredinia (yellow) and dark brown to black necrotic patches on adaxial surface of a poplar leaf, b premature leaf shedding of P. deltoides clones due to a severe rust infection, c a pattern of hyperparasitism between C. oxysporum and rust uredinia as evident by cooccurrence of white, dusty, cottony growth of Cladosporium with orange rust urediniospores on abaxial leaf surface, d a net-like structure of C. oxysporum on the leaf with yellow, dead rust urediniospores

A remarkable pattern of hyperparasitism between Cladosporium and rust uredinia was observed during the field surveys, where Cladosporium was found to co-exist with rust uredinia on poplar leaves (Fig. 1c). Cladosporium formed a net like structure on the leaf, resembling a white-dusty, cottony growth. Also, the fungus hyphae spread their mycelia towards the host urediniospores (Fig. 1d). This pattern was the result of a hyperparasitism relationship between these two fungal species.

Morphological determination of Melampsora and Cladosporium spp

Based on the morphological characteristics, rust fungus was identified as Melampsora medusae. The echinulate of M. medusae urediniospores was uniformly distributed, ranging from half to three-fourth of the wall surface except for the smooth equatorial area, which is a key characteristic feature of the M. medusae (Fig. 2a). The echinulate of urediniospores was further characterized with round to flattened apex with truncate base. Rust urediniospores were golden yellow to pale yellow, and some of them appeared colorless (Fig. 2b). Paraphyses were capitate and had roughly spherical apex with uniformly thickened wall, and large lumen attached to it (Fig. 2c). Some paraphyses were oval to less commonly clavate in shape (Fig. 2d). The overall shape of the urediniospores was obovate, oval and pear-shaped.

Fig. 2.

Morphological characterization of Melampsora medusae: a stained uredeniospores of M. medusae showing round, pear shaped, and echinulate pattern, b golden yellow to pale yellow, flattened, oval shaped urediniospores with smooth equatorial patches, c–d oval shaped paraphyses with large lumen, e–f naturally infected leaf section showing the tightly encapsulated uredinium

The antagonistic fungus was isolated from the rust infected leaves of the poplar, which was recognized as Cladosporium oxysporum. Morphologically, the colonies of C. oxysporum were flat to low convex, grey-olivaceous to mainly olivaceous-grey with glabrous and wrinkled surface (Fig. 3a). The texture of colonies was velvety to floccose and fluffy with grey-olivaceous margins. Microscopic investigation revealed septate, branched and subhyline to pale olivaceous brown hyphae. The conidiophores were long, swollen, round about the stalk, unbranched, smooth or paler towards the apex. The nodose formation on conidiophores, a key feature, was not observed due to its rarity on PDA medium. Ramoconidia were rare while secondary ramoconidia were abundant (Fig. 3c). Conidia were numerous, aseptate, catenate and arranged in branched chain conformation. The conidia were smaller in size with smooth surface and limoniform or ellipsoid shape (Fig. 3d). The key characteristics of both the fungi were summarized in Table 3.

Fig. 3.

Morphological characterization of Cladosporium oxysporum: a–b front and back view of C. oxysporum culture on PDA media, c microscopic observation of dense secondary ramoconidia, d aseptate, smooth and limoniform-ellipsoid conidia of small size

Table 3.

Morphological characterization of Melampsora medusae and Cladosporium oxysporum

Melampsora medusae
Characteristics		Paraphysis head		Urediniospore
Shape		Capitate to clavate		Obovate or Oval
Apex shape		Capitate		Echinulate
Wall		Thickened at the equator		Uniformly thick
Size (µm)		19.55 ± 38.95 ± 55.12 × 11.17 ± 18.23 ± 26.39		18.1 ± 28.04 ± 48.31 × 12.16 ± 17.35 ± 21.94

Cladosporium oxysporum
Characteristics		Conidiophore		Ramoconidia		Conidia
Shape	Solitary, branched, straight to flexuous, nodose and subhyline to pale olivaceous- brown		0–1 ( – 2) septate, Straight, secondary ramoconidia were smooth, ellipsoid-ovoid to fusiform		Numerous, smooth, branched in chain, subglobose, obovoid, ellipsoid-ovoid to limoniform and septate
Size (µm)	49.28 ± 167.73 ± 339.23 × 0.85 ± 2.31 ± 5.33		6.9 ± 11.71 ± 18.06 × 1.07 ± 3.14 ± 5.53		1.55 ± 5.02 ± 8.23 × 1.27 ± 2.42 ± 4.32

In vivo antagonism assay

Uredinia-covered leaves inoculated with C. oxysporum had the same appearance as observed in diseased plant leaves in the field (Fig. 4a). Rust infected leaves were treated with distilled water, which resulted in a large amount of orange-colored uredinia (Fig. 4b). Furthermore, in the absence of uredinia, no C. oxysporum conidia or hyphae were found on the inoculated leaf surface after 96 days post inoculation (Fig. 4c). The Cladosporium isolate showed the significant parasitism when compared with that of the control (Fig. 4d).

Fig. 4.

Antagonism assay of Cladosporium oxysporum: a C. oxysporum spore suspension inoculated on rust infected poplar leaf (white dusty mycelial growth of Cladosporium appeared), b un-inoculated rust infected poplar leaf (no white dusty mycelia growth of Cladosporium appeared), c healthy poplar leaf (control) inoculated with spore suspension (no white dusty mycelia growth of Cladosporium appeared), d quantification of the parasitism percentage leaf area (5cm²) covered by uredinia of M. medusae, e effect of C. oxysporum on the germination of M. medusae at different deposition sequences on cavity slides. The deposition sequences were: T1, conidia inoculated first followed by urediniospores after 1 h; T2, urediniospores inoculated first followed by conidia after 1 h; T3, uredinia and conidia deposited simultaneously; and T4, rust urediniospores with water only (control). Values represent the average of 50 spore counts. Vertical bars are the standard errors of the mean

The cavity slide method with the all-deposition sequences C. oxysporum caused a significant reduction in mean germination percentage of urediniospores compared with the control at 12 and 24 h, respectively (Fig. 4e). The minimum germination of uredeniospores was observed when conidia were applied 1 h prior to urediniospores (T1) but the difference was non-significant when the treatments were reversed (T2). However, both the treatments were found more effective than the third treatment when both the fungi were simultaneously applied.

Hyperparasite mode of action under light and SEM microscopy

After germination, C. oxysporum started to form the intimate relationship with M. medusae (Fig. 5a). Moreover, it was observed that the protracted hyphae of parasitic fungus colonized or reached towards the rust urediniospores (Fig. 5b) and its germ tube (Fig. 5c), which affects the germination of urediniospores. Microscopic examination revealed two types of host cell wall penetration mechanisms, namely enzymatic and direct penetration. In enzymatic penetration, the spore of C. oxysporum attached to the surface of urediniospores (Fig. 5d), which leads to adhesion to the cell wall of the rust fungi, and subsequent rupture of the urediniospores (Fig. 5e). As a result, the urediniospores become sunken, hyaline, and are altered from their original form (Fig. 6a). The rust spore lost their cytoplasmic content which is an indication that enzymes have penetrated the spore, breaking down the cell wall and releasing the contents of the cell (Fig. 6b). Furthermore, the fungus has also demonstrated to act through direct penetration, where the mycoparasite hyphae enter the host through the formation of haustoria and without haustoria. When the haustorium was involved, a fine structure such as an infection cushion was observed (Fig. 6c, d). In addition, it has been observed that C. oxysporum conidiophores erupted straight from the urediniospores (Fig. 6e). A hyperparasitic fungus has also developed an appressoria-like structure to invade rust uredeniospores. These appressoria were darkly pigmented, or melanized (Fig. 6f). SEM microscopic observation also confirmed the contact hyperparasites mode of action via forming mechanical pressure on rust urediniospores (Fig. 7).

Fig. 5.

Light microscopy of several aspects of hyperparasitism of Cladosporium oxysporum on Melampsora medusae: a germinated spores of C. oxysporum showing intimate relationship with rust urediniospores, b germ tube of the hyperparasite near a group of ungerminated spore of rust fungi, c mycoparasite hyphae coiled around the germ tube of rust uredeniospores, d enzymatic penetration, conidia of C. oxysporum attached to the surface wall of urediniospores, e ruptured surface of a rust spore (arrow) due to enzymatic penetration mechanism. GS- germinated spores of C. oxysporum, GT- germ tube of M. medusae, HP- hyphae of C. oxysporum

Fig. 6.

Light microscopy of several aspects of hyperparasitism of Cladosporium oxysporum on Melampsora medusae: a hyaline, sunken and altered urediniospores after C. oxysporum attachment, b release of cytoplasmic content from urediniospores after adhesion of mycoparasite, c–d direct penetration of mycoparasite by forming infection cushions, e penetration of mycoparasite hyphae into host cells without forming haustoria, f the mycoparasite formed darkly-pigmented appressoria on the rust cell wall, g leaf section showing the C. oxysporum reproducing on rust urediniospores. IC infection cushion, AP appressorium

Fig. 7.

Scanning electron microscopy (SEM) of several aspects of parasitism of Cladosporium oxysporum on Melampsora medusae: a rust spores heavily parasitized by mycoparasite hyphae, b contact parasitism, antagonist produced physical pressure on host spore surfaces, c–d appressorium mediated parasitism between antagonism and rust spore, indicating that the germ tube of C. oxysporum have potential to adhere, collapsed, disintegrated and degraded the urediniospores. AP appressorium

Molecular identification and phylogenetic analysis

To support the morphological identification, these fungi were characterized through sequencing of DNA barcode loci, namely ITS and TUB2. The BLAST comparisons of the ITS region of rust pathogen showed 99.69% homology with M. medusae, which was also exemplified by the cluster analysis. The phylogenetic tree displayed five major clades in which M. medusae clade was well-distinguished from other related species with a strong bootstrap support (94%). Further, our isolate was grouped under the M. Medusae clade when rooted with M. yezoensis (Fig. 8). Thus, our rust pathogen has been identified as M. medusae based on both the morphological as well as molecular characteristics. The ITS sequence of M. medusae obtained in this study was submitted to the NCBI GenBank with accession number ON787810.

Fig. 8.

ITS sequence-based maximum parsimony tree depicting the phylogenetic relationship of the Melampsora medusae fungal isolate (bold with red color) with other Melampsora spp. The values at branches correspond the bootstrap support at 1000 replications

On the other hand, BLAST comparisons of ITS sequence of the antagonist fungi showed > 99% homology with members of the C. cladosporioides complex but could not be clearly distinguished in the phylogenetic tree constructed through maximum likelihood method. In order to obtain the species level discrimination, ITS sequence was further concatemerized with another DNA barcode loci (TUB2) and used for phylogenetic analysis. ML-tree based on combined data set displayed reliable topology where the C. cladosporioides complex was clearly distinguished from the C. herbarum complex with the boot strap support of 68%. As evident by the sub clustering in ML-tree, our isolate showed significant sequence similarity with C. oxysporum (Fig. 9) and clustered in a single subcluster with high boot strap support of (99%). Consequently, this antagonist fungal isolate was identified as C. oxysporum. The DNA sequences of both the regions of C. oxysporum have also been submitted in the NCBI GenBank with accession number ON787632 (ITS) and ON 981,381 (TUB2).

Fig. 9.

ITS and TUB-based maximum likelihood tree depicting the phylogenetic relationship of the Cladosporium oxysporum, a hyperparasitic fungus (bold with red color) isolated from urediniospore with other Cladosporium spp. The values at branches correspond the bootstrap support at 1000 replications

Screening of P. deltoides clones against M. medusae under field conditions

Based on the disease severity index, resistant and susceptible clones were identified, and stability of resistant trait was established by recording the data for two consecutive years under natural environment. Out of the 25 clones, four clones, namely FRI-FS-83, FRI-FS-92, FRI-FS-140, and FRI-AM-111, showing mild rust symptoms were designated in HR category (Table 4). Interestingly, the rust resistance of HR clones in the germplasm bank also remained consistent in the nursery conditions for the evaluation period of two consecutive years. Besides, a clone D-121 which was assigned under MR category in the germplasm showed higher resistance in the nursery. Contrarily, two clones, UDAI and OP-06, were designated as HS due to severe rust incidence recorded under same environment. However, none of the clones were found immune based on rating over the course of the year. Hence, the null hypothesis was rejected since the p value < 0.05 indicated a significant difference between the clones (Suppl. Table 1).

Table 4.

Comparative diseases severity variation of 25 poplar clones against Melampsora medusae rust infection in germplasm and nursery

Category	Germplasm, 2019	Nursery, 2019	Nursery, 2020
HR	FRI-FS-83, FRI-FS-92, FRI-FS-140, FRI-AM-111	FRI-FS-140, FRI-FS-83, FRI-FS-92, FRI-AM-111, D-121	FRI-FS-140, FRI-FS-83, FRI-FS-92, FRI-AM-111, D-121
R	FRI-FS-177, FRI-FS-25, G-48, FRI-FS-185, FRI-AM-15, KARANTI, OP-53	FRI-AM-77, FRI-FS-185	FRI-FS-185, FRI-FS-25, FRI-AM-15, BAHAR, G-48, FRI-AM-77,
MR	FRI-AM-77, D-121, FRI-AM-47	KARANTI, L-88	FRI-FS-177, OP-53, PIP-112
MS	BAHAR, FRI-AM-33, L-20084, PIP-112, FRI-AM-103	BAHAR, FRI-FS-25, G-48, OP-53, S7C15, FRI-AM-15, FRI-AM-33, FRI-FS-177	S7C15, KARANTI
S	FRI-AM-132, S7C15, OP-77	OP-77, FRI-AM-132, FRI-AM-47	FRI-AM-33, L-88, OP-77, FRI-AM-132, FRI-AM-47
HS	L-88, UDAI, OP-06	OP-06, FRI-AM-103, L20084, PIP-112, UDAI	OP-06, FRI-AM-103, L-20084, UDAI

Discussion

Rust infection symptomatology on leaf of the P. deltoides

The present study revealed that M. medusae was the key causal organism of the rust outbreak in Populus deltoides clones in India. Under natural environment conditions, rust infection is most likely to occur between mid-autumn and late winter due to humid and temperate climate conditions. According to EPPO report (2009), M. medusae was first reported in the eastern part of North America which subsequently was also recorded in other continents. Despite the heavy incidence, limited studies were carried out for the identification and management of rust infection of P. deltoides in India. For instance, Paul et al. (2004) has reported the association of M. medusae with poplar rust disease, but the detailed account of investigation is not available. In general, M. medusae required two hosts to complete their life cycle but the alternative host is not yet known in India. However, in the absence of secondary host the stage of urediniospore capable of completing their life cycles and passes directly from clones to clones rapidly (Smith and Charles 1998; Sinclair and Lyon 2005).

Morphological and phylogenetic analysis

Based on the morphology and molecular methods, the current work has made it feasible to accurately identify the fungal pathogen and its antagonist at species level, i.e., M. medusae and C. oxysporum, respectively. M. medusae could be distinguishable from others closely related species, such as M. populina, M. larici-populina and M. medusae-populina by uredinial pattern and ITS sequences. Echinulate pattern is considered as the key characteristics to distinguish M. medusae from the other related species (Spiers and Hopcroft 1994). However, the morphological parameters measured in this study were resembled with several Melampsora species (Newcombe et al. 2000; EPPO 2009; Vialle et al. 2011), which made it difficult to confirm the rust pathogen at species level. Further, it has been exemplified by sequence comparisons and phylogenetic analysis of ITS region, where the Melampsora isolate (ON787810) recovered in this study showed high level of similarity with M. medusae. As evident in the phylogenetic tree, our fungal isolate was clustered in a clade of M. medusae with high boot strap support (Fig. 8). In agreement to the Zheng et al. (2019), the present study also demonstrated that the Melampsora species could be conspicuously distinguished through the single gene approach based on ITS region when integrated with morphological characters.

Species level identification of Cladosporium is much challenging due to varied fungal phenotype exhibited under variable cultural and environmental conditions and proper taxonomic identification is neither feasible with morphological features nor using single universal gene, i.e., ITS. Therefore, multiple gene-based phylogenetic approaches were applied to resolve the species under Cladosporium complex (Dugan et al. 2004; Schubert et al. 2009; Bensch et al. 2010, 2018). Considering these facts, the morphological investigation was complemented with two gene approach, i.e., ITS and TUB2. BLAST comparisons showed maximum similarity with C. cucumerinum and C. oxysporum, which further resolved in phylogenic tree where the Cladosporium isolate recovered in this study showed a close association with the C. oxysporum with high boot strap support (Fig. 9). Higher similarity observed between both the Cladosporium species was also demonstrated in a earlier study using ITS and TUB sequences (Zalar et al. 2007). The distinction of both the species was further verified through the morphological features, such as shape, length of the conidia and conidiophores, and the growth pattern of the fungus on PDA media (Wirsel et al. 2002; Bensch et al. 2012).

Confirmation of hyperparasitism

During the surveys in germplasm bank, white dusty mycelia were observed to be prevalently occurring over the rust affected leaves contributing to decline in rust infection. It intrigued us to identify the fungus and establish the antagonism because similar observation was also made in P. nigra (Sharma and Heather 1978). Through the literature survey, it was found that Cladosporium spp. has been shown to have a close association with rust disease in plants (Salvatore Moricca and Assante 2005) and, in most cases, is reported to act as antagonist rather than the pathogenic (El-Dawy et al. 2021; Sheroze et al. 2002). Recently, Barge et al. (2022) have also identified 17 species of Cladosporium belonging to the C. cladosporioides and C. herbarum complexes as mycoparasites of Melampsora x Columbiana, a rust pathogen of P. trichocarpa.

Particularly, no evidence was found claiming that Cladosporium has caused any disease in Populus deltoides. In a study conducted on Populus ciliata and observed that Cladosporium humile had a pathogenic effect. Nevertheless, all clones of P. deltoides were shown to be impervious to C. humile (Sharma and Sharma 2001). The result of in-vitro leaf assay method from the present study also confirmed that C. oxysporum did not cause any disease development in P. deltoides (Fig. 4a). Furthermore, we found that, the leaves inoculated with conidial suspension of Cladosporium showed a significant decrease in the number of uredinial count of rust compared to those leaves which were not inoculated with conidial suspension, indicating an antagonistic relationship between C. oxysporum and M. medusae (Fig. 4d). A previous study found a similar association, showing that C. aecidiicola significantly reduced the viability of urediniospores of M. medusae (Sharma and Heather 1980). An important study conducted by Barge et al. (2022) used leaf disc assays to examine the potential of microparasitism as a mechanism of mitigating rust disease severity in P. trichocarpa, where it is shown that seven Cladosporium isolates (four from the C. cladosporioides and three from the C. herbarum complex) were capable of significantly decreasing the severity of the disease. Additionally, they concluded that C. cladosporioides was found to be slightly more effective than C. herbarum in conferring mycoparasitism. Interestingly in our finding Cladosporium isolate that showed reduction in rust urediniospores count also belong to C. cladosporioides complex (Fig. 9). Hence, it could be inferred that synthesis of secondary metabolites and enzymes particularly released by C. cladosporioides complex are more potent to reduce rust severity in poplar.

In the second experiment i.e., cavity slide method, it is demonstrated that, across all the different deposition methods, the presence of C. oxysporum reduced the rust spores in the sample as compared to the control. Our results indicated that the germination rate of urediniospore was lowest when the antagonist was introduced 1 h prior to the rust uredinia. However, no significant difference was observed when the treatment was reversed (T2) (Fig. 4e). Treatment 3 yielded higher germination than treatments 1 and 2, but the overall germination rate for all treatments was significantly lower than the control. This suggests that the addition of an antagonist prior to rust uredinia is effective in reducing rust germination. The addition of the antagonist may have inhibited the germination of the rust uredinia by blocking the sites where the spores attach to the host plant, thus preventing the germination of the rust uredinia. Additionally, the antagonist may have affected the germination of the rust uredinia by causing direct damage to the cell wall or membrane of the spores (as explained in more detail in the following section). Findings of the previous study were consistent with our results; Moricca et al. (2001) showed that C. tenuissimum was effective as an antagonistic fungus when its conidia were applied 1 h prior to the aeciospores of Cronartium flaccidum rust fungi. However, Assante et al. (2004) found that C. tenuissimum was not effective when applied simultaneously with uredeniospores of U. appendiculatus. In congruence, some earlier studies have also suggested that certain Cladosporium species have inhibitory effect on the germination of the Melampsora rust fungi (Sharma and Heather 1978, 1980; Omar and Heather 1979).

Hence, the Cladosporium isolate from the present study proves their aggressiveness towards rust uredeniospores to minimize the impact of M. medusae on P. deltoides clones. Despite the significant antagonistic potential, it needs to be evaluated in the field condition for utilizing as bio-control in forestry.

Hyperparasitism mode of action under light and SEM microscopy

In general, when the nutrition source is limited, mycoparasite start to colonize other associated fungi (Köhl et al. 2019). In present study, the mycoparasite ability of colonizing the rust uredeniospores and its germ tube was extremely high. Similar findings were also reported in C. uredinicola colonizing on Tranzschelia prunispinosae (Dolińska et al. 2011), and C. tenuissimum colonizing on Melampsora larici-populina (Sharma and Heather 1988). On contrary, Anderson et al. (2016) reported no coiling of C. uredinicola hyphae around P. araujiae rust of Araujia hortorum. Further, Light and SEM microscopic observations revealed three types of hyper-parasitism mechanism: enzymatic (Fig. 5d, e), direct (Fig. 6c–f), and contact parasitism via mechanical force (Fig. 7b). The changes in fungal cell structure from their original shape are often followed by enzymatic penetration (Yusuf et al. 2019; Torres et al. 2017; Mims and Richardson 2007) which was also evident in our study. In direct penetration, the hyperparasite straightly enters into the host uredeniospores by with or without forming special structure such as haustoria and infection cushion. Such type of mechanism known as biotropic mycoparasitism (Jeffries 1995). Though, in contact hyperparasitism, hyphae of the C. oxysporum produced physical pressure on host cell wall (Fig. 7b) (Tsuneda and Hiratsuka 1979).

Moreover, it is possible to anticipate that the mycoparasite hyphae would have degraded the spore surface by using an enzyme, and later by mechanical force, the hustoria would enter into the host cell wall. This hypothesis was further supported by a study of Edwards and Allen (1970) in which the infection mechanism of Erysiphe graminis f. sp. hordei was described in barley. Conclusively, the novel C. oxysporum isolate obtained from the current study showed a multiple mode of action of hyperparasitism signifying its aggressiveness. This indicates that their ability to target a host depends not only on a single mode, but rather, one or more modes of action are involved, that may change over time.

Screening of P. deltoides clones against M. medusae fungi

Though we have confirmed the biocontrol efficacy of C. oxysporum for rust disease in P. deltoides through in vitro experiments of hyperparasitism, a field validation is still required to prove the effectiveness and ruling out the negative impact over other plant species, animals and humans before recommending its commercial use. Alternatively, efforts were carried out to exploit the natural variability of rust resistance in poplar clones by screening under natural condition. As climatic conditions are changing rapidly, a rapid breakdown of resistance against pathogens is possible. Consequently, the behavior of plant species varies with the geographic distribution and climate conditions (Fasahat et al. 2015). Similar studies in susceptibility to rust in poplar clones were reported (Giorcelli et al. 1996; Pinon and Frey 1997; Dhillon and Sandhu 2020). Thus, in that case, there is a need to identify such clones that show stability and durability towards changing environmental conditions. In view of this, 25 commercially cultivated clones of P. deltoides were screened for the rust resistance through disease severity scale. This study has displayed a significant variability in rust resistance among poplar clones. For instance, four clones (FRI-FS-83, FRI-FS-92, FRI-FS-140, and FRI-AM-111) were found highly resistant in germplasm bank as well as nursery when evaluated for two consecutive years.

It was observed that D-121 was the sole clone that shifted from the MR to HR category when it was out planted in the nursery. Even though there is uncertainty with regards to disease resistance, it may still be put to better use since its performance was relatively better when planted outside of germplasm. The historical background of this particular clone was best in field with regards to growth, biomass nutrient uptake and carbon storage (Swamy and Mishra 2014; Swamy et al. 2006; Kumar and Singh 2012; Rawat et al. 2001). In addition, over the course of this study it was found that the clones UDAI and OP-06 maintained consistency in highly susceptible category. On the contrary, others reported that growth performance of the UDAI was finest in the nursery (Dhiman and Chander 2012).

Conclusion and future perspectives

In the present study, the rust pathogen M. medusae of P. deltoides was found to cause the most destructive disease as its infection leads to early defoliation, which has been investigated in detail for its symptomology. In order to gain a better understanding of the complete biology of the fungal pathogen and its history, such as how and from where it has spread and whether or not any alternative host exist in India, further research needs to be undertaken. Importantly, we have proved the antagonistic relationship of C. oxysporum with M. medusae, which could be further utilized as biocontrol agent in field plantations as it can be easily isolated from the host fungi. To the best of our knowledge, no reports were available showing hyper-parasitism of C. oxysporum over M. medusae in P. deltoides. However, this study also brings up the question of which came first, the hyperparasite or the rust. To understand the evolution of hyperparasite for realized virulence towards their host, studies should also be taken into account in the future to comprehend this perspective.

In consideration of the results obtained from screening of P. deltoides clones against rust fungi in natural infection condition, there is an opportunity for growers to select the stable clones that are resistant to rust diseases in challenging climatic conditions. By screening 25 clones, five clones namely FRI-FS-83, FRI-FS-92, FRI-FS-140, FRI-AM-111 and D-121 were found best against rust infection. Furthermore, these clones would be of great importance for plant breeders to develop new resistant clones to Melampsora species.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

KT and AP involved in conceptualization of the work, data generation and analysis, manuscript writing and interpretation, KT and PK involved in data collection, RKM involved in data analysis, compilation and manuscript writing; HSG and SB contributed in manuscript editing; RN performed the statistical analysis.

Funding

Not applicable.

Availability of data and materials

The sequence data generated in this study have been deposited in NCBI GenBank and are publicly available. All other data required to understand the manuscript are provided in original manuscript and supplementary material.

Code availability (software application or custom code)

Not applicable.

Declarations

Conflict of interest

On behalf of all authors, the corresponding author affirms that the manuscript has been neither published elsewhere nor submitted simultaneously for publication elsewhere. The article has been read by each author and that there is no conflict of interest.

Ethics approval

Not required.

Consent to participate

All authors agreed to participate for submission of the research to the journal.

Consent for publication

All the authors have approved the manuscript for submission and publication in journal.