Screening for Resistance to Coffee Leaf Rust

Abstract

Coffee leaf rust (CLR), caused by Hemileia vastatrix (Hv), is one of the main limiting factors of Arabica coffee production worldwide. Breeding for rust resistance is the most appropriate and sustainable strategy to control CLR. The characterization of coffee resistance to Hv, initiated in the 1930s in India, expanded with the creation of Coffee Rusts Research Center (CIFC) in 1955, in Portugal. Since then, the screening of coffee resistance to Hv races, from different geographical origins, has been carried out assisting breeding programmes of coffee growing countries and originating over 90% of the resistant varieties cultivated worldwide. However, the high adaptability of Hv has resulted in the gradual loss of resistance of some varieties. Thus, the characterization of new sources of resistance is crucial, also to face the recent epidemic resurgence of CLR across Latin America and the Caribbean.

Here, we provide a protocol for the screening of coffee resistance to Hv using different methods of inoculation on attached and detached leaves and on leaf disks. Information on environmental and pathogenicity factors that may affect the assessment of coffee resistance is also presented. This protocol allows the characterization of rust resistance on coffee mutants at laboratories, greenhouses, and field conditions.

1 Introduction

Coffee, the most important agricultural commodity, is crucial for the economy of more than 70 countries and is a livelihood source for between 12 and 25 million farmers worldwide (ICO 2019). The value of coffee exports amounted to USD 20 billion in 2017/18 being the revenue of the coffee industry estimated to surpass USD 200 billion (Samper et al. 2017; ICO 2019).

The two main cultivated coffee species, Coffea arabica (Arabica) and C. canephora (Robusta) account, on average, for 60% and 40%, respectively, of the world’s coffee production (ICO 2020).

Coffee leaf rust (CLR), caused by the biotrophic fungus Hemileia vastatrix Berkeley and Broome, is considered the main disease of Arabica coffee. Since the historical first burst of CLR in the nineteenth century that caused the eradication of coffee cultivation in Sri Lanka, the disease gained a worldwide distribution, becoming practically endemic in all regions where coffee is grown (Wellman 1957; McCook 2006; Silva et al. 2006; Talhinhas et al. 2017; Keith et al. 2021). The disease produces economic losses over USD 1 billion annually (Kahn 2019).

H. vastatrix is a hemicyclic fungus producing urediniospores, teliospores and basidiospores, but only the dikaryotic urediospores, which form the asexual part of the cycle, reinfect successively the leaves whenever environmental conditions are favourable (Talhinhas et al. 2017 and references therein). After urediospore germination and appressorium differentiation over stomata, the fungus penetrates and colonizes the mesophyll tissues inter-and intracellularly giving rise to sporulation about 21 days after inoculation (Silva et al. 1999 and references therein; Silva et al. 2006; Talhinhas et al. 2017).

Breeding for resistance has been the most appropriate and sustainable strategy to control crop diseases.

Plant resistance to pathogens has been grouped into two different categories (Vanderplank 1968): complete resistance conditioned by single genes with major effects and incomplete resistance conditioned by multiple genes with minor and additive effects. A variety of terms have been used to refer to this perceived dichotomy, vertical versus horizontal, major-genes versus minor-genes, oligogenic versus polygenic, qualitative versus quantitative, race-specific versus race non-specific, hypersensitive versus non-hypersensitive, narrow-spectrum versus broad-spectrum (Parlevliet and Zadocks 1977; Roelfs et al. 1992). This diversity of terms reflects the assumptions made by the respective authors, but it also adds an element of confusion to the literature because some terms are used in different ways by different authors (Poland et al. 2009). Here the term incomplete resistance is considered as any form of resistance which allows for at least some reproduction of a given pathogen isolate on a given host plant (Eskes 1983).

Complete resistance results in phenotypes that fit into discrete classes of resistant and susceptible individuals according to Mendelian ratios (qualitative resistance). On the other hand, incomplete resistance cannot be easily categorized into distinct groups but in a continuous distribution of susceptible and resistant individuals (quantitative resistance) (Corwin and Kliebenstein 2017).

The traditional recording system for complete resistance on coffee to rust was developed at Coffee Rusts Research Center (CIFC) by d’Oliveira (1954–57) and consists of the identification of eight lesion types grouped into 4 classes of resistance and susceptibility. The incomplete resistance can be measured by its components, like infection frequency, lesion size, latent period and sporulation intensity (Browning et al. 1977; Parlevliet 1979; Eskes 1983).

The first effective characterization of coffee resistance to CLR, in experimental bases, was initiated in the 1930s in India (Mayne 1932, 1942). This work was greatly developed and broadened with the creation of CIFC in 1955, in Portugal. Inheritance studies have demonstrated that coffee-rust interactions follow the gene-for-gene relationship of Flor (1971) within a race-specific resistance system (complete resistance), being the resistance of coffee plants conditioned at least by nine major dominant genes (S_H1- S_H9) singly or associated. Reversely, it was possible to infer 9 genes of virulence (v₁-v₉) on H. vastatrix (Noronha-Wagner and Bettencourt 1967; Bettencourt and Rodrigues 1988). More than 55 H. vastatrix physiological races, from different geographic origins, were also identified over 60 years of world surveys carried out at CIFC (Rodrigues et al. 1975; Várzea and Marques 2005; Silva et al. 2006, 2022; Talhinhas et al. 2017; CIFC records), which allowed the characterization of coffee germplasm to support breeding programmes at coffee research institutions.

For many years, selection for H. vastatrix resistance has been based on highly specific complete resistance derived from major introgressed genes from C. arabica (S_H1, S_H2, S_H4 and S_H5) as well as from diploid species such as C. canephora (S_H6 - S_H9) or C. liberica (S_H3). To date, some of the most widely used sources of resistance to CLR are the Timor hybrids – HDTs (natural C. arabica x C. canephora hybrids) characterized and supplied by CIFC to research institutions of coffee growing countries (Rodrigues et al. 1975; Bettencourt and Rodrigues 1988).

The recent loss of resistance in some HDT-derived varieties, due to the appearance of more virulent rust races (Várzea and Marques 2005; Silva et al. 2006, 2022; Prakash et al. 2010; Talhinhas et al. 2017, CIFC records), as well as the current epidemics in Latin America and Caribbean, highlights the importance of the discovery and characterization of new sources of resistance.

Based on the CIFC’s routine activities, we present a detailed protocol focused on the screening of complete resistance to CLR. A description of the qualitative scale used for the assessment of the reaction types and the environmental and pathogenicity factors that may affect this evaluation is also reported. The methods described here can be used in a greenhouse, laboratory or in field conditions and are useful for screening  coffee mutants for leaf rust resistance.

2 Materials

2.1 To Collect and Store Inoculum

• 1. Urediniospores.

• 2. Gelatin capsules (8.5 mm).

• 3. Desiccator.

• 4. Vaseline (to close the desiccator).

• 5. Sulfuric acid solution.

• 6. Petri dishes.

• 7. Refrigerator (4 °C).

• 8. Cryogenic vials.

• 9. Minus 80 °C freezer.

• 10. Liquid nitrogen containers.

• 11. Cryo-gloves.

• 12. Flexible polyethylene tubing.

• 13. Laboratory lamp.

• 14. Scissors and tweezers.

• 15. Ethanol for surface and tools sterilization.

2.2 Spores Viability

• 1. Slides and coverslips.

• 2. Tweezers.

• 3. Micropipettes (100 µl).

• 4. Scalpel blades.

• 5. Watch glasses.

• 6. Formaldehyde solution.

• 7. Transparent nail polish.

• 8. Cotton blue lactophenol staining.

• 9. Light microscope.

2.3 Inoculation

• 1. Coffee leaves.

• 2. Urediniospores.

• 3. Distilled water.

• 4. Scalpels and soft brushes.

• 5. Test tubes.

• 6. Vortex mixer.

• 7. Micropipettes (10 µl, 20 µl).

• 8. Manual or electric sprayer.

• 9. Plastic bags.

• 10. Petri dishes.

• 11. Trays, nylon sponge and glass plates.

• 12. Analytical balance.

• 13. Ethanol for surface and tools sterilization.

• 14. Tween 80 solution.

• 15. Room with controlled light and temperature (Phytotron).

• 16. Incubation chambers.

2.4 Phenotyping of Coffee-Rust Interactions

• 1. Qualitative scale of reaction types.

• 2. Magnifying lens (if needed).

3 Methods

3.1 Procedures to Collect and Store Inoculum

For disease resistance screening, urediniospores collected with gelatin capsules are used as inoculum. The urediniospores must be collected from well sporulated young lesions. Note that spores from lesions in fallen leaves lose their viability very soon.

If enough spores cannot be harvested in the field for reliable screening tests, we can increase this amount with inoculations on vars. Caturra, Catuaí, Mundo Novo, Typica, Bourbon, etc. (carrying the resistance gene S_H5, i.e., with susceptibility to all the rust races infecting Arabicas) in greenhouse conditions.

Storage of rust samples must be done using recently collected spores thus to ensure high viability.

Rust samples can be stored for short and long term: (i) urediospores in gelatin capsules placed above sulfuric acid solution in a desiccator (50% relative humidity) and kept in refrigerator (4 °C) should retain good viability for about 180 days; (ii) in a freezer, at −80 °C, the spores keep the viability for more than 15 years; (iii) in liquid nitrogen, at −196 °C, spores can be stored for more than 20 years with high viability (CIFC records).

After storage at a negative temperature, a heat shock treatment (40 °C for 10 min) is required to break dormancy of the urediniospores and to recover their germination ability (CIFC records).

3.2 Spores Germination Tests

Before each experiment, laboratory germination tests are recommended to check the spores' viability. The urediniospore germination may be evaluated in vitro (glass slides) or in vivo (leaf pieces), being the last more accurate.

The germination in vitro is evaluated by placing aliquots of 100 µl of the urediniospore suspension (prepared as described in 3.3.1.2.) in glass slides which are kept in a moist chamber during 16 h at 23 °C. After this time, the germination is stopped with an aqueous solution of 3% formaldehyde. The glass slides are then covered with cover slips and observed under the microscope and the percentage of germinated spores counted on a minimum of ten fields of 100 urediniospores each (Silva et al. 1985).

The germination in vivo is evaluated on leaf pieces (5cm²) cut from the previously inoculated leaves (see Sect. 3.3.), 16–24 h after inoculation (Silva et al. 1999). After let dried, the fragments are painted with transparent nail polish on the lower surface. About 24 h later, the dried nail polish (leaf replica) is removed with the help of tweezers and dipped into cotton blue lactophenol to stain the fungal structures (urediniospores, germ tubes and appressoria). The leaf replicas are placed in glass slides containing the same staining, covered with cover slips and observed under the microscope. With this technique it is possible to evaluate the rates of both the germinated urediniospores and the appressoria differentiated over stomata. Countings are made on a minimum of six microscope fields of 100 urediniospores each.

3.3 Inoculation Techniques

The screening of disease resistance is usually carried out by artificial inoculations by spreading fresh urediniospores on the lower surface of the coffee leaves with the help of a sterilized soft hairbrush or by using a urediniospore suspension on attached leaves in greenhouses or at field conditions, as well as on detached leaves and leaf disks at laboratory conditions.

Young, fully expanded leaves of the terminal node are used. In the day before inoculation, the plants are abundantly watered, and the turgid leaves are inoculated on the plant (leaves attached to the plant) or removed from the plant (detached leaves or leaf disks).

3.3.1 Attached Leaves

3.3.1.1 Brushing

This method can be used in attached leaves in the greenhouse or in the field. Following the routine procedure used at CIFC, fresh urediniospores of H. vastatrix (about 1 mg per pair of leaves) are placed with a scalpel on the lower surface of the leaf (see Fig. 1a) and then brushed gently with a camel’s hairbrush (see Fig. 1b). The inoculated leaves are sprayed with distilled water (see Fig. 1c) and the plants are placed for 24 h under darkness at room temperature (18 °C to 24 °C) in moist chambers, (see Fig. 1d) after which they are placed in the greenhouses.

Fig. 1

Inoculation of attached leaves using the brushing technique. H. vastatrix urediniospores on the scalpel (a) and then brushed on the lower surface of the leaf with a camel's hairbrush (b). Inoculated leaves are sprayed with distilled water (c) and placed in a moist chamber (d). For large plants, the inoculated leaves are enveloped in a humid plastic bag (e) and covered with newspaper sheets (f)

When the moist chambers are too small to allow the incubation of plants, the inoculated leaves after sprayed with distilled water are enveloped with a humid plastic bag during 24 h (see Fig. 1e). To avoid direct incidence of the sun rays, the plastic bags are covered with ordinary paper or newspaper sheets (see Fig. 1f) (D’Oliveira 1954–57). The same procedure can be used in field conditions.

3.3.1.2 Urediniospore Suspension

The lower surface of the leaves is inoculated with an urediniospore suspension, with a concentration of 0.8 to 1.2 mg ml⁻¹, using a manual or electric sprayer (see Fig. 2), in a greenhouse or field conditions. These suspensions are prepared by suspending urediniospores in distilled water (to which 1–2 drops of Tween 0.02% is previously added). In the absence of Tween the following procedure is suggested: a spore mass is placed in a small test tube; one drop of distilled water is then added and kneaded into the spore mass with the help of a vortex mixer. This process is repeated by adding one drop of water at a time until the moistened spore mass has the pasty consistency of heavy cream. At this point, the bottom of the test tube is placed in a vortex mixer for several minutes and the remaining volume of water required for the final suspension is added during the stirring process. This step degasses the spore surfaces, which improves spore viability and yields almost complete dispersal of the spores in water. Good but incomplete suspensions leave a film of unwetted spores on the water surface if the stirring is omitted. Spores in suspensions prepared by this method germinate normally (CIFC records).

Fig. 2

Inoculation of the lower surface of the coffee leaves with an urediniospore suspension using an electric sprayer

The upper inoculated leaves and part of the branch to which the leaves are attached are sprayed with distilled water and enveloped with a humid plastic bag. To avoid direct incidence of the sunrays, the plastic is covered with paper/newspaper sheets. The plastic bags are removed about 24 h after. Inoculations at field conditions are carried out in the late afternoon and the bags are removed early in the morning (Eskes 1989).

3.3.2 Detached Leaves

The leaves are placed with the abaxial surface upwards in trays lined on the bottom with a nylon sponge saturated with distilled water. Each leaf is inoculated with droplets (10–20 µl) from the urediniospores suspension (with a concentration of 250–500 spores/droplet). The droplets are deposited between the veins, using a micropipette (see Fig. 3). The trays are covered with glass plates and placed in the dark for 24 h at 22 ± 2 °C. After this time, the drops are dried out with small pieces of filter paper, and the trays, covered with glass plates, are placed under moderate light conditions (fluorescent or indirect daylight of 500–1,000 lx) with a photoperiod of 12 h under similar temperatures (Eskes 1983).

Fig. 3

Detached leaves inoculated with droplets of an urediniospore suspension

3.3.3 Leaf Disks

Leaf disks are cut with cork borers from 1 to 2 cm in diameter and placed in Petri dishes or in trays with the upper leaf side down, on a sponge saturated with tap water. The disks are inoculated with droplets, from 10 to 20 µl of urediniospores suspension (with a concentration of 250–500 spores/droplet). After inoculation the boxes are closed with glass lids and incubated in the dark in the same conditions and environment described above for the detached leaves (Eskes 1982a).

3.4 Phenotypic Scoring Method for Disease Resistance

At greenhouse conditions, with a range of temperatures from 18 to 24 °C, the reading of the reaction types takes place usually 30–35 days after the inoculations, by a qualitative scale developed at CIFC by D’Oliveira (1954–57). However, the time to score the reactions can be extended to 45 days or more in the following situations: at higher or lower temperatures during the colonization process and with low aggressiveness of the fungal isolates.

This recording system has been followed at CIFC for more than 60 years to identify complete resistance on Coffea spp to CLR and to characterize rust races.

Qualitative scale used at CIFC to score the reaction types on attached leaves (D’Oliveira 1954–57; Bettencourt and Rodrigues Jr. 1988) (see Fig. 4).

Fig. 4

Reaction types, according to the qualitative scale used at CIFC. Flecks visible when holding the leaf against the light (a); tumefactions (b); reaction 0 (c); reaction 1 (d); reaction 2 (e); reaction 3 (f); reaction 4 (g)

i = immune (no visible symptoms).

fl = Flecks: small chlorotic flecks at the penetration sites, well visible with a pocket lens or when holding the leaf against the light.

; = Necrotic spots, visible macroscopically at the penetration site or dispersed over the infected area.

t = Punctiform tumefactions, often associated with flecks.

0 = Chlorotic spots, more or less intense, in the infected area, sometimes associated with small necrotic areas, but without spore production.

1 = Rare sporulating sori, always very small, sometimes only visible with a pocket-lens, in areas which are mainly chlorotic, sometimes associated with necrosis.

2 = Small or medium-sized pustules, diffused but visible macroscopically, in areas with intense chlorosis.

3 = Medium-sized or large pustules, surrounded by chlorosis.

4 = Large sporulating pustules, without true hypersensitivity, but sometimes surrounded by a slight chlorotic halo (highly susceptible or compatible).

X = Heterogeneous reaction with urediosporic pustules very variable in size associated with resistant reaction types.

The reaction types i, fl, t and 0 are jointly referred to as resistant (R), 1 as moderately resistant (MR), 2 as moderately susceptible (MS), and 3 and 4 as susceptible (S).

Detached leaves and leaf disks are useful to identify very susceptible genotypes to rust. However, intermediate levels of resistance expressed by a low reaction type (reactions 1 and 2) on attached leaves, at greenhouse or field conditions, may not be observable in leaf disks or detached leaves. In this way, whenever lesions without sporulation are found on detached leaves and leaf disks, we suggest to repeat the inoculations on attached leaves.

4 Notes

• 1. To collect inoculum

  When collecting rust samples in plants, either in the field or in the greenhouse it is important to avoid the presence of mycoparasites like the fungus Lecanicillium lecanii (Zimm.) Zare and W. Gams, with the ability of reducing spore viability and disease severity (Vandermeer et al. 2010; James et al. 2016; CIFC records). The first evidence of L. lecanii in rust lesions is in the form of small white spots at the center of the rust sori. The spots gradually enlarge; the cotton-like, white colored mycelium of the mycoparasite covered the rust sori. The development of this mycoparasite is restricted to the rust infected leaf parts, but never grow to over the entire width of the rust lesions (CIFC records).

• 2. Inoculum

  When the virulence of rust local populations is not known, the source of inoculum to be used to detect resistance in coffee mutants should be gathered from the same plants or similar genotypes where they come from. If resistance is found in the first inoculations, the screening on coffee mutants should continue with inoculum collected from different coffee genotypes in different regions to try to get rust samples with higher spectra of virulence. In general, the origin and distribution of rust races follow the resistance genes present in coffee populations.

• 3. Factors influencing the infection process and the resistance symptoms

  • 3.1. Moisture

    The urediniospores do not germinate, even at high relative humidity, if the free water is absent. If the water dries off before penetration, then the process is inhibited (Nutman and Roberts 1963; Rayner 1972; CIFC records).

  • 3.2. Temperature

    • (i) The temperature, while the leaf surface is wet, is one of the most important factors that determine the amount of spore germination and penetration. The optimum temperatures for germination are 20 to 25 °C (CIFC records).

    • (ii) Extreme temperatures after inoculation causes some depressive effect on fungal colonization and sporulation, with the slight lower reaction types on susceptible plants and in extreme may kill the fungus inside the leaves (Montoya and Chaves 1974; Ribeiro et al. 1978; Silva et al. 1992). The small chlorotic lesions, developed in these conditions, are likely to be confused with resistant reactions (CIFC records).

    • (iii) The enlargement of lesions on leaves and the sporulation are limited by temperatures over 35 °C and lower than 10 °C.

  • 3.3. Light intensity and/or leaf age

    • (i) Leaves exposed to higher light intensities before inoculation show more lesions than those exposed to lower intensities (Eskes 1982b, 1983, 1989).

    • (ii) In screening tests, the light intensity should preferably be kept at medium levels before inoculation and medium to low levels after inoculation (Eskes 1982b, 1983, 1989, CIFC records).

    • (iii) Some derivatives of interspecific tetraploid hybrids (C. arabica x C. canephora) like Icatu and Timor Hybrid (HDT) show lower and even resistant reaction type lesions at a lower light intensity under greenhouse conditions (Marques and Bettencourt 1979; CIFC records).

    • (iv) Studies on the effect of leaf age and light intensity on CLR found higher resistance on young leaves growing in the shade, and lower resistance for old leaves exposed to sunlight (Eskes 1983).

• 4. Phenotypic scoring for disease resistance

  • 4.1 In the assessment of complete resistance, the use of susceptible controls is needed to exclude the possibility of escapes (inadequate exposure to the pathogen and/or extreme temperatures during the initial infection process).

  • 4.2 The reaction type “i” (Immunity = no macroscopically visible symptoms), which may appear to be very desirable, is rarely observed under CIFC greenhouse conditions. Care should be taken not to confuse this reaction type with escape. Each time immunity occurs confirmation with a new test with the same pathotype is needed.

  • 4.3 Coffee plants irradiated at a dose of 100 Gy exhibited several morphologic changes in leaves like shape, length and width, the color of young leaves, number of leaves per plant, etc., and distance from the cotyledon to the first node (Quintana et al. 2019). These changes may influence the expression of resistance to leaf rust. Irradiated coffee plants should be inoculated, if possible, on leaves of different ages.

  • 4.4 “Resistance” should be distinguished from “Tolerance” which is defined as the ability of a crop to maintain a high yield in the presence of disease, being a difficult characteristic to measure, and its component traits are generally undefined (Newton 2016). Note that tolerance should not be confused with incomplete resistance.

  • 4.5 Partial resistance characterized by a reduced rate of epidemic development despite a high- or susceptible-infection type (Parlevliet 1975) was never detected at CIFC greenhouse conditions.

• 5. Incomplete resistance

  The contribution of Albertus B. Eskes (1989 and references therein) for the characterization of incomplete resistance on coffee to CLR, using leaf disks and detached leaves, was of paramount importance and his works are a reference to those who intend to develop studies on this kind of resistance.

  • 5.1 The quantitative or incomplete resistance of a host genotype cannot be assessed in absolute terms; it is always a relative measure compared with that of a well-known standard genotype. The latter is often the most susceptible genotype available (Parlevliet 1989).

  • 5.2 The degree of incomplete resistance evaluated in a particular coffee genotype may be masked by different levels of H. vastatrix aggressiveness. Intermediate compatibility in host/pathogen interaction (low reaction type and long latent period) can be due to incomplete resistance of coffee plants/or lower aggressiveness (fitness) of rust.

  • 5.3 Components of incomplete resistance to CLR

    Infection frequency: Number of lesions per leaf or leaf area unit, or the percentage of disks with lesions.

    Latent period: The time from inoculation to spore production. Normally it is calculated as the time taken for 50% of the lesions to sporulate or the time between the inoculation and the formation of the first spores.

    Incubation period: The number of days between the inoculation and the appearance of the first chlorotic lesions per leaf or disk

    Proportion of sporulating lesions: Percentage of sporulated lesions in relation to the total number of lesions by leaf or disk.

    Sporulation intensity: The number of spores produced per sporulating lesion or per infected leaf area, over a certain time interval.

    Lesion size: Normally evaluated at the end of the experiment

  • 5.4 Relations amongst reaction types (RT) and components of incomplete resistance

    The majority of the components of incomplete resistance are a quantitative extension of the scale used for RT. These components, as well as the RT’s are related to the same basic criteria, like lesion size, sporulation intensity, and the occurrence of chlorosis or necrosis. The latent period is related to lesion size when fungal growth is slow, the sporulation will generally be delayed, and the lesions will be smaller. The reaction types “0” (chlorosis is without sporulation) or necrotic spots will reduce the sporulation intensity and/or the duration of sporulation (Eskes 1981).