SEARCH

SEARCH BY CITATION

Keywords:

  • AFLPs;
  • breeding;
  • Coffea;
  • marker-assisted selection;
  • resistance;
  • root-knot nematode

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Breeding for resistance to root-knot nematode Meloidogyne exigua in coffee may help in controlling this nematode, which causes substantial harvest losses throughout Latin America. Accessions of Coffea arabica, C. canephora and lines derived from the Timor Hybrid (wild C. arabica × C. canephora interspecific hybrid) were tested in a greenhouse for resistance to gall formation by counting individual galls on the roots of plants inoculated with second-stage juveniles of a M. exigua population from Costa Rica. The level of introgression from C. canephora was also assessed on 28 resistant and susceptible genotypes in a molecular study with amplified fragment length polymorphism (AFLP) markers. The frequency of resistant plants was very low or even nonexistent in C. arabica and very high in C. canephora with > 78% immune plants and 100% resistant plants. Several lines derived from the interspecific hybrid revealed a high level of resistance similar to that observed in the parent C. canephora species. A molecular study of Timor Hybrid-derived lines revealed high variability for the number of markers present in the study genotypes. A larger number of introgression markers was found in the group of resistant genotypes than in the susceptible genotypes. However, there were also lines with little introgression that had retained resistance. Based on genetic distances calculated from these markers, two groups of lines were revealed: those derived from Timor Hybrid CIFC1343, and those derived from Timor Hybrids CIFC832/1 and CIFC832/2, which may indicate the existence of different resistance genes in the two groups. The segregations observed in F1 and F2 progeny may be explained by at least a dominant gene. The lines derived from the Timor Hybrid are a worthwhile source of resistance to M. exigua that can be exploited to improve C. arabica with the help of molecular-assisted selection.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Root-knot nematodes of the genus Meloidogyne are frequently and abundantly found on Arabica coffee trees (Coffea arabica) in Central America (Campos et al., 1990; Anzueto et al., 1993). Recent studies confirmed the great diversity of potentially pathogenic species in the region (Hernández et al., 1996), of which M. exigua is the most common. This species is widespread in South America (Sosa-Moss, 1985; Campos et al., 1990; Leguizamon, 1990) and is dominant in Costa Rica (Flores & López, 1989). In Costa Rica, the attacks cause general weakening of the tree with an estimated drop in yields ranging from 10 to 20% (Bertrand et al. 1997). In Central America, all the cultivated varieties (namely, Typica, Bourbon, Caturra, Catuai, Costa Rica 95 and IHCAFE90) are susceptible.

C. arabica is characterized by a relatively low genetic diversity that is attributable to its allotetraploid origin, reproductive biology and evolution (Lashermes et al., 1999). In relation to resistance to M. exigua, Curi et al. (1970) has shown, with a Brazilian population, that most genotypes are susceptible. In contrast, resistance was discovered in the C. canephora species (Curi et al., 1970; Morera & López, 1987; Chaverri, 1987). Since the second half of the 20th century, most breeding programmes implemented worldwide (Brazil, Colombia, Kenya, Costa Rica, Honduras) have attempted to transfer characters of resistance to rust (Hemileia vastatrix), root-knot nematodes (Meloidogyne spp.), and coffee berry disease (Colletotrichum kahawae) into cultivars of C. arabica from the Timor Hybrid. This tree, which originated from the island of Timor (Bettencourt, 1973), is derived from a wild interspecific cross between C. arabica (2n = 2x = 44) and C. canephora (2n = 2x = 22). Several progeny have been used in breeding programmes: origins CIFC832/1, CIFC832/2 and CIFC1343. Those origins have been crossed with commercial varieties such as Caturra or Villa-Sarchi to produce Catimors or Sarchimors. The F1 hybrid is selfed and a pedigree programme is implemented over more than five generations. Based on this strategy, several cultivars have been released in Brazil (IAPAR 59), Colombia (Colombia cultivar) and Central America (IHCAFE 90 and Costa Rica 95) (Bertrand et al., 1999). Certain lines selected for their resistance to leaf rust are also resistant to M. exigua (Fazuoli et al., 1977; Fazuoli & Lordello, 1978; Bertrand et al. (1997); Gonçalvez & Pereira, 1998; Silvarolla et al., 1998). With the recent development of molecular markers of gene introgression from C. canephora (Lashermes et al. 2000), it is now possible to envisage genetic diversity studies, and the marking of resistance genes. The benefits obtained from marker-selection depend on several factors, such as the presence of independent single resistance genes and the degree of linkage between the markers and the target gene.

The aims of the present study were: (i) to verify that resistance sources to M. exigua are very rare or nonexistent in C. arabica and that C. canephora is resistant; (ii) to confirm that the resistance genes present in the lines derived from the interspecific Timor Hybrid are widespread and came from C. canephora; (iii) to examine the possibility of developing a marker-assisted selection programme based on molecular markers of gene introgression from C. canephora.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Nematode populations and inoculation

The nematode population came from a severely infested plot in which the coffee trees had average infestation levels of 1700 second-stage juveniles (J2) per gram of roots. The plot was located at the CICAFE experimental station (Barva de Heredia, Costa Rica) at 1100 m above sea level in a humid tropical climate with a 3–4 month dry season and a mean temperature of 22–23°C. The population used in this study was characterized from three samples by enzymatic electrophoresis for four enzyme systems (esterase, malate dehydrogenase, superoxide dismutase, and glutamate-oxaloacetate transaminase) (Hernández et al., 1996). The samples revealed the esterase phenotype VF1 of M. exigua (Esbenshade & Triantaphyllou, 1985) and the enzyme phenotypes N1, N1a and H1, respectively, for the malate dehydrogenase, superoxide dismutase, and glutamate-oxaloacetate transaminase systems.

Nematodes were extracted from roots by the mist technique (Hooper, 1970). Once the coffee seedlings reached the two cotyledon leaf stage, they were transferred to 250-mL pots containing a 2 : 1 mixture of disinfected soil and sand (pH 5·2–5·6). The inoculum was applied to the collar of each coffee seedling using a micropipette when the plants were between 2 and 3 months old (two pairs of leaves). Nematode inoculum concentration was determined from a 1-mL aliquot and 3000 ± 200 J2 were applied per pot. Observations were carried out 5 or 6 months after inoculation.

Experiments and plant material

The trials were carried out in a greenhouse in Costa Rica at a relative humidity of 70–90% and a mean temperature of 25°C. The objective of trial 1 was to confirm that the species C. arabica does not provide genes of resistance to M. exigua. The seeds came from Ethiopian accessions collected in the centre of origin of the species by Guillaumet & Halle, 1978) in 1966 or by the FAO in 1965 (FAO, 1968) and kept in the CATIE field genebank in Costa Rica. Twenty-nine accessions were assessed by taking at random the open-pollinated progeny of a single tree. In trial 2, the objective was to confirm that the species C. canephora is resistant to the nematode. The plant material came from C. canephora accessions (2n = 2x = 22). Ten accessions kept in the CATIE genebank and six accessions from a genebank in French Guiana were tested using seeds taken at random for each origin from a single tree in the collection. The objective of trial 3 was to confirm that resistance is widespread in lines derived from different hybridizations between accessions of Timor Hybrid populations (CIFC832/1, CIFC832/2, CIFC1343) and commercial cultivars (Caturra, Villa-Sarchi). A study on progeny segregation was carried out in trial 4, by studying one pure line (IAPAR59 in F8), one F1 progeny obtained by crossing one resistant plant of cv. IAPAR59 by the susceptible cv. CR95, and one F2 progeny obtained by self-crossing a resistant F1 hybrid, T5296 × T4759. The number of plants per family was in the range 13–25 for trials 1, 2 and 3 and 273–365 for trial 4. Cv. CR95, a Catimor line well-known for its susceptibility, was used as the susceptible control.

A field assessment was carried out on 28 trees that were more than 10 years old, representing 28 Catimor and Sarchimor lines, in generations F4 to F7. These trees were assessed on two occasions, in November 1996 and October 1999, in trials at the CICAFE experimental station, under conditions of severe infestation (800–4000 nematodes per gram of roots), using the protocols described by Bertrand et al. (1997). For each tree, the first 5–10 cm of soil were removed in a radius of ≈ 80 cm from the trunk. The unit observed was one secondary root around 30 cm in length with a diameter of 3–5 mm, with its root hairs. Each tree had four to six root units that could be observed in the first few centimetres of soil. Following these observations, the genotypes for which no galls were observed were scored as resistant (R), and those for which at least one gall was found were scored susceptible (S).

AFLP protocol

Leaf samples were taken from the 28 trees characterized for their resistance in the field and representing 28 introgressed lines. The amplified fragment length polymorphism (AFLP) procedure was performed as described in Lashermes et al. (2000). Aliquots of 500 ng genomic DNA were digested using two restriction enzymes, EcoRI and MseI. The linker-adapter and primer sequences were as described by Vos et al. (1995), with minor adaptation for coffee DNA. Pre-amplification was carried out with 1-bp extension primers. Selective amplification of restriction fragments was conducted using primers with three selective nucleotides. One of these primers (EcoRI primer) was end-labelled with γ-[33]-ATP using T4 polynucleotide kinase. A total of 42 AFLP primer combinations was used (sequences indicated in Lashermes et al. 2000). Amplification products were separated on 6% denaturing polyacrylamide gels. The gels were dried and exposed to Kodak Bio Max X-ray film. Amplification products were separated on 6% denaturing polyacrilamide gels. The gels were dried and exposed to Kodak Bio Max X-ray film.

Histological techniques

Histological sections were cut to observe plant reactions to the parasite. The samples were fixed in FAA (formaldehyde alcohol, acetic acid and distilled water) for 48 h. They were then dehydrated in a series of ethanol (70–100%), 1 h in each bath, embedded in Historesinô at 4°C overnight, and then moulded. Sections 3 mm thick were cut and stained with toluidine blue or naphthol blue black.

Analysis of data

Numbers of root galls were assessed using a six point gall index (GI) scale; the classes were GI = 0, 0 galls; GI = 1, 1–2 very small galls; GI = 2, 3–10 very small galls; GI = 3, 11–30 small and large galls; GI = 4, 31–100 small and large galls; GI = 5, over 100 small and large galls. It was considered that plants scoring 0, 1, 2 in this index were resistant (R) and those classed in 3, 4 and 5 were susceptible (S). Chi-squared tests were carried out when necessary, based on plant frequencies on the six-point scale, to compare groups of plants.

The AFLP amplification products were designated according to the restriction enzymes, the primer combination used and in order of decreasing fragment size. Only AFLP bands showing a clear polymorphism were scored as present (1) or absent (0). As previously described (Lashermes et al. 2000), the additional marker band (i.e. not detected in any of the accessions of C. arabica and observed in accessions of C. canephora) detected in the Timor Hybrid-derived genotypes analysed were considered as introgressed markers from C. canephora. Genetic distances (GD) between genotypes were estimated as follows:

inline image

where Nx is the number of introgressed bands in line x and not in line y, Ny is the number of introgressed bands in line y and not in line x, and Nxy is the number of introgressed bands in lines x and y. Analysis by the neighbour-joining tree construction method (Saitou & Nei, 1987) was performed with the treecom software package (Van de Peer & Wachter, 1994).

The number of markers for each Timor Hybrid group was statistically analysed by a one-way analysis of variance (anova) and compared using the least significant difference test (LSD) at P = 0·05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

C. arabica (trial 1)

None of the accessions of C. arabica tested were resistant. All had a majority of plants in classes 4 and 5 of the gall index. Of the 523 individuals tested, only nine (2%) had a gall index of < 3. That represented either resistant plants or plants for which inoculation was a failure. Some accessions appeared as susceptible as the control. It should be noted that 22 accessions had more than two plants in class 3, as opposed to none for the control. The histological sections cut according to plant class did not reveal any differences between the different classes of susceptible plants (GI = 3–5). The histological patterns were identical. The sectioned galls revealed several females deeply buried in the vascular bundles. These females were associated with large to very large egg masses. The giant syncytial cells were well formed and abundant.

C. canephora (trial 2)

All the C. canephora progenies tested proved to be 100% resistant. Of the set of 387 plants tested, 303 (78%) did not show any symptoms, 67 (17·3%) were in class 1 and 17 (4%) were in class 2. Sections cut from small galls in classes 1 and 2 showed that different situations were apparent, namely poor nematode development with a single female per gall and poorly developed egg masses or abnormalities due to incomplete development of the giant cells when compared with that observed in controls.

Lines derived from the Timor Hybrid (trial 3)

Evaluation for resistance to M. exigua of Catimor and Sarchimor lines, derived from the Timor Hybrid, revealed a highly contrasting response. Six lines were classed as resistant and two as susceptible. Among the resistant plants (193 plants), a majority were in class 0 (71%), and a smaller proportion of plants in class 1 or 2 (29%). In the two susceptible lines, there were a few plants in class 3, unlike the control, which had none. The histological sections revealed similar behaviour to those previously observed in C. canephora for the group of resistant plants or in C. arabica for the group of susceptible plants.

Relations between introgressed markers and resistance to M. exigua

A total of 113 introgressed marker bands (i.e. additional bands inherited from C. canephora) were observed among the Timor Hybrid-derived genotypes studied. The mean number of markers introgressed per genotype was 20·9, with extremes of one marker only for genotype T17929–37 introgression markers for genotype T17936 (Table 1). The genotypes divided into 12 susceptible and 16 resistant. A larger number of markers was found in the group of resistant genotypes than in the susceptible genotypes. Of the 16 resistant genotypes, 11 (i.e. 69%) had 20 or more markers, whereas among the susceptible plants, the proportion was 33·3%. The origins derived from Timor Hybrid CIFC832/1 had statistically fewer markers, on average, than the other two origins (Table 2). The number of markers was 13·0 for CIFC832/1, as opposed to 22·1 and 23·4 for origins CIFC832/2 and CIFC1343. Resistant genotypes were found among the three Timor Hybrid origins (Table 2). It should be noted that some genotypes had few introgression markers, but they were nonetheless resistant (T17925, T17930). The proportion of resistant genotypes was higher in the group of lines derived from Timor Hybrid CIFC1343, as shown in Table 2 (chi-squared = 5·61; P < 0·06).

Table 1.   Evaluation of 28 introgressed genotypes of C. arabica for resistance to a Meloidogyne exigua population of Costa Rica, by number of AFLP markers
No.CodeMarkers (n)Parent HdTaR/Sb
  • a

    Parent HdT, derived from the Timor Hybrid; 1, CIFC832/1; 2, CIFC832/2; 3, CIFC1343.

  • b

    R, resistant; S, susceptible.

1T1812221S
2T18123323S
3T18126233S
4T1812781S
5T18137252S
6T18139142S
7T17924323R
8T17925143R
9T17926283R
10T17927303R
11T17928263S
12T1792913S
13T17935163R
14T17936373R
15T17937333R
16T17940193R
17IAPAR cv212R
18CR 95 cv121S
19T1812181S
20T18130311R
21T5175171S
22T17930143R
23T17931303R
24T17933163R
25T18140202S
26T18141202R
27T5296272R
28T16784282R
Table 2.   Mean number of introgression markers and distribution of genotypes susceptible and resistant to the nematode M. exigua, depending on the origin of the Timor Hybrid, CIFC832/1, CIFC832/2 and CIFC1343.
Timor HybridMean number of markersaNumber of susceptible genotypesNumber of resistant genotypes
  • a

    Means followed by the same letter do not differ significantly (P = 0·05) according to the LSD test.

CIFC832/113·00 b51
CIFC832/222·14 ab34
CIFC134323·40 a411

The dendrogram (Fig. 1) calculated from the genetic distance estimated from the introgression markers illustrated the distances calculated between the genotypes derived from the three Timor Hybrid origins. The lines derived from origin CIFC1343 are grouped together except for T17929 which has only one introgression marker. The Sarchimors T18141 and T18140, which are very close genetically (same generation until F4) differed in their resistance reaction.

image

Figure 1.  Dendrogram of the introgressed Arabica genotypes generated by the neighbour-joining method using genetic distance based on introgression of AFLP marker bands. Genotypes were observed to be (R) resistant or (S) susceptible to M. exigua. The Timor Hybrid progenitor used in the introgression programme is indicated in parenthesis as 1, 2 and 3 for CIFC832/1, CIFC832/2 and CIFC1343, respectively.

Download figure to PowerPoint

Resistance trait segregation

A study of the reaction of one pure line, one F1 progeny and one F2 progeny is shown in Table 3. Plants derived from the fixed line were all resistant. Of the plants derived from the F1 progeny of a resistant plant crossed with a susceptible plant, there was only one susceptible plant out of the 274 plants tested. Among the F2 progenies, segregation was seen that was similar to a 3 : 1 or 12 : 4 type Mendelian model (chi-squared = 0·07, P > 0·93) indicating the existence of at least one dominant gene for resistance against M. exigua.

Table 3.   Evaluation of one pure line, and one F1 and F2 progeny of Coffea for resistance to an M. exigua population from Costa Rica. Resistance was based on a six-point scale assessed on the roots of each plant, evaluated 5–6 months after inoculation. Plants with gall indices (GI) of 0 to 2 were considered as resistant and those with gall indices (GI) of 3 to 5 were considered susceptible.
NameType of populationType of crossNumber of plants testedNumber of resistant plantsNumber of susceptible plants
IAPAR 59Pure line (F8)R × R273273
IAPAR59 × CR95F1 progenyR × S2742731
T5296 × T4759F2 progenySelf-crossing (R × S)36527689
CR95 (control)Pure line (F8)S × S2525

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The results revealed very different levels of resistance to M. exigua in the two species studied. In C. arabica, 98% of the plants tested proved to be susceptible and 100% of the C. canephora progenies were resistant. This result confirmed the observations by Curi et al. (1970) in Brazil. A low percentage of plants was found in class 3 for a large number of Ethiopian accessions of C. arabica, whereas there were none for the control. The resistance evaluation scale greatly influences the interpretation of the results. Presence of minor genes in the Ethiopian accessions could account for class differences. However, microscopic symptoms did not differ between class 3 and classes 4 or 5. It should be also noted that the control was a particularly early selected line, whose development in the nursery was vigorous. In contrast, the accessions of Ethiopian origin mostly gave rather weak trees, with relatively late development (unpublished data). The difference in vigour could explain the differences in reaction to the nematode. The more roots the plant had, the more the nematode developed.

All the C. canephora introductions were classed as resistant, with a very large proportion (78·3%) of plants in class 0 (immune). However, 17·3% and 3·6% of plants were found in classes 1 and 2, respectively. Microscopy revealed that classes 1 and 2 had varying symptoms reflected by the existence of a single female and giant cells developed to varying degrees. Detailed studies of host/nematode reactions should enable a more precise description to be made of the physiological bases of resistance. This would be particularly useful for finding out whether these various reactions are controlled by different genes. Plants from the variety Nemaya derived from the cross between T3561 (2–1) and T3751 (1–2) were 91% of symptom-free plants, suggesting low nematode invasion. These progenies have also been selected for their resistance to Meloidogyne spp. in El Salvador and to M. incognita in Guatemala and are distributed as rootstocks for cultivated C. arabica varieties (Bertrand et al., 1999).

Catimor and Sarchimor lines are frequently resistant to M. exigua, which confirms earlier results (Bertrand et al., 1997; Gonçalvez & Pereira, 1998; Silvarolla et al., 1998). The results obtained here showed that the level of resistance involved could be quite similar to that found in C. canephora. The Arabica parents of these lines do not reveal resistance to M. exigua, and supports the suggestion by Gonçalvez & Pereira (1998) and Silvarolla et al. (1998) that the resistance genes were introgressed from C. canephora and gave resistance to a large proportion of lines derived from the interspecific hybrid. However, among the 28 study genotypes, which came from the three Timor Hybrid origins, substantial variability in terms of introgression marker content was revealed. Some genotypes, such as T17929, had virtually no introgressed content. Moreover, this genotype is susceptible to leaf rust, Coffee Berry Disease and M. exigua (Bertrand et al. 1997). Despite a few noteworthy exceptions, such as genotype T18123, it would seem that the genotypes that had the largest number of introgression markers were also those that had the greatest chance of having retained the resistance genes. Conversely, for the genotypes with limited introgression, resistance was less frequent. However, it should be noted that resistant genotypes exist that revealed little introgression, which suggests that genes of resistance to M. exigua do not have strong linkages with the introgressed chromosome fragments from which they came. If it is considered that the study genotypes effectively represented the lines from which they were derived, it can be taken that all or part of the resistance genes would have been conserved in more than half of the lines. To the authors' knowledge, there has not been any rational selection of the lines for resistance to the nematode M. exigua. The loss or conservation of introgressed resistance genes is doubtless due more to the randomness of choices made by breeders and the number of generation during the pedigree selection process than to a rational choice by breeders. For example, genotypes T5175 and T8667, which represent cultivars IHCAFE90 and CR95, respectively, on which considerable selection pressures have been exerted (selection for productivity and resistance to leaf rust up to F7), revealed few introgression markers and were susceptible to M. exigua. On the other hand, resistance to M. exigua has been fixed in cultivar IAPAR59. In this case, the frequent existence of the nematode in the trials in Brazil would have favoured plants possessing such resistance. For lines derived from CIFC1343, most of which make up the multiline Colombia variety, conservation of resistance in most of the lines is largely due to the few generations of selection at the time of their distribution (Castillo & Moreno, 1986).

Based on genetic distances calculated from molecular markers, two groups of lines depending on the origins of the Timor Hybrid were revealed: firstly the group of lines derived from CIFC1343, and secondly the groups of lines derived from CIFC832/1 and CIFC832/2. Differences between the two groups may have been due to introgressed chromosome fragments of a different nature. In the latter case, notably when seeking resistance markers, the possibility will have to be considered that the resistance genes in CIFC1343 are different from those in CIFC832/1 and CIFC832/2. Genetic studies of host–parasite relations in cultivated plants with respect to resistance to Meloidogyne very often revealed that a limited number of genes are operating (Sidhu & Webster, 1981) and more recently the same has been shown by Roberts et al. (1996) in Vigna unguiculata, Wang & Goldman (1996) in carrot and Cap et al. (1993) in tomato. In coffee, Anzueto et al. (2000) also suggested the existence of only two complementary genes for resistance to M. incognita in Guatemala. The results presented here indicate the existence of at least one dominant gene for resistance to M. exigua. The high frequency of resistant lines suggest the possibility of either using lines derived from the Timor Hybrid directly as a variety, or intraspecific crossing programmes, notably with the Ethiopian accessions, which often reveal resistance to M. incognita (Anzueto et al. 2000) or to M. arabicida (unpublished data), so as to create vigorous and resistant F1 hybrids (Bertrand et al., 1999). It should be possible to detect DNA markers linked with qualitative resistance to M. exigua and one approach, namely bulk segregant analysis (Halden et al., 1997), is currently being pursued.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This work was supported in part by the European Union through the International Scientific Cooperation Programme (INCO-DC contract ERBIC18CT970181).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Anzueto F, Bertrand B, Sarah JL, Eskes AB, Decazy B, 2001. Resistance to Meloidogyne incognita in Ethiopian Coffea arabica origins: detection and study of resistance transmission. Euphytica 118, 18.DOI: 10.1023/A:1003712232325
  • Anzueto F, Eskes AB, Sarah JL, Decazy B, 1993. Résistance de quelques descendances de Coffea arabica et C. canephora vis-à-vis de deux populations de Meloidogyne spp originaires du Guatemala et du Brésil. Proceedings 15th International Scientific Colloquium on Coffee 1993. Paris, France: ASIC, 33848.
  • Bertrand B, Aguilar G, Bompard E, Rafin A, Anthony F, 1997. Comportement agronomique et résistance aux principaux déprédateurs des lignées de Sarchimor et Catimor au Costa Rica. Plantations Recherche Développement 4, 31221.
  • Bertrand B, Aguilar G, Santacreo R, Anzueto F, 1999. El mejoramiento genético en América Central. In: BertrandB, RapidelB, eds. Desafios de la Caficultura Centroamericana. San José, Costa Rica: IICA-CIRAD, 40556.
  • Bettencourt A, 1973. Consideraçoes Gerais Sobre O ‘Hibrido Do Timor’. Circular no. 31, Campinas, Brazil: Instituto Agronomico de Campinas.
  • Campos PV, Srivapalan P, Gnanapragasam NC, 1990. Nematode parasites of coffee, cocoa and tea. In:LucM, SikoraRA, BridgeJ, eds. Plant Parasitic Nematodes in Subtropical and Tropical Agriculture. Wallingford, UK: CAB International, 387430.
  • Cap GB, Roberts PA, Thomason IJ, 1993. Inheritance of heat-stable resistance to Meloidogyne incognita in Lycopersicon peruvianum and its relationship to the Mi gene. Theoretical and Applied Genetics 85, 77783.
  • Castillo Zapata J & Moreno G, 1986. La Variedad Colombia: Selección de un Cultivar Compuesto Resistente a la Roya Del Cafeto. Bogota, Colombia: Federación Nacional de cafeteros de Colombia.
  • Chaverri LFA, 1987. Evaluación de la Resistencia de Cinco Clones de Coffea canephora cv. Robusta Al ataque de dos Poblaciones de Meloidogyne exigua Goeldi, Thesis for Licenciado de la Escuela de Ciencias agrarias. Heredia, Costa Rica : Universidad Nacional.
  • Curi SM, Carvalho A, Moraes FP, Monaco LC, Arruda HV, 1970. Novas fontes de resistencia genética de Coffea no controle de nematoide do cafeeiro, Meloidogyne exigua. Biologico 36, 2935.
  • Esbenshade PR & Triantaphyllou AC, 1985. Use of enzyme phenotypes for identification of Meloidogyne species. Journal of Nematology 17, 620.
  • FAO, 1968. FAO Coffee Mission to Ethiopia. Rome, Italy: FAO.
  • Fazuoli LC & Lordello RA, 1978. Resistência de cafeeiros Híbrido do Timor a Meloidogyne exigua. Ciencia y Cultura, Suplemento, Brazil 30, 3.
  • Fazuoli LC, Monaco LC, Carvalho A, 1977. Resistência do cafeeiro a nematóides; I- Testes em progenies é híbridos para Meloidogyne exigua. Bragantia 39, 297307.
  • Flores L & López R, 1989. Caracterización morfológica del nematodo del cafeto Meloidogyne exigua (Nemata: Heteroderidae). Hembras y huevos. Turrialba 39, 35260.
  • Gonçalvez W & Pereira A, 1998. Resistência do cafeeiro a nematóides. IV- Reaçao de cafeeiros derivados do Híbrido de Timor a Meloidogyne exigua. Nematologia Brasileira 22, 3950.
  • Guillaumet JL & Halle F, 1978. Echantillonnage du matériel Coffea arabica récolté en Ethiopie. In: Charrier, A ed. Etude de la Structure et de la Variabilité Génétique des Caféiers, 14. Paris, France: IFCC, 138.
  • Halden C, Sall T, Olsson K, Nilsson NO, Hjerdin A, 1997. The use of bulked segregant analysis to accumulate RAPD markers near a locus for beet cyst nematode resistance in Beta vulgaris. Plant Breeding 116, 1822.
  • Hernández AM, Fargette V, Molinier H, Ramanason B, Decazy B, Sarah JL, 1996. Enzymatic characterization and reproductive fitness on coffee of root-knot nematode populations from Central America. Nematropica 26, 274.
  • Hooper DF, 1970. Laboratory methods for work with plant and soil nematodes. In: SoutheyJF, ed. MAFF Technical Bulletin No 2. London, UK: MAFF, 349.
  • Lashermes P, Andrzejewski S, Bertrand B, Combes MC, Dussert S, Graziosi G, Trouslot P, Anthony F, 2000. Molecular analysis of introgressive breeding in coffee (Coffea arabica L.). Theoretical and Applied Genetics 100, 13946.
  • Lashermes P, Combes MC, Robert J, Trouslot P, D'Hont A, Anthony F, Charrier A, 1999. Molecular characterisation and origin of the Coffea arabica L. genome. Molecular Genome and Genetics 261, 25966.
  • Leguizamon JE, 1990. Los nematodos del cafeto en Colombia y su control. 50 años de Cenicafé 1938–1988: Conferencias Conmemorativas. Caldas, Colombia: Cenicafé, 7380.
  • Morera GN & López R, 1987. Respuesta de seis lineas experimentales de Coffea spp. a la inoculación con Meloidogyne exigua. Nematropica 17, 1039.
  • Roberts PA, Matthews WC, Ehlers JD, 1996. New resistance to virulent root-knot nematodes linked to the Rk locus of cowpea. Crop Science 36, 88994.
  • Saitou N & Nei M, 1987. The neighbour-joining method: a new method for reconstructing phylognenic trees. Molecular Biology and Evolution 4, 40625.
  • Sidhu GS & Webster JM, 1981. The genetics of plant–nematode parasitic systems. Botanical Review 47, 387419.
  • Silvarolla BM, Gonçalvez W, Lima MA, 1998. Resistência do cafeeiro a nematóides. V. Reproduçao de Meloidogyne exigua em cafeeiros derivados da hibridaçao de Coffea arabica com C. canephora. Nematologia Brasileira 22, 519.
  • Sosa-Moss C, 1985. Report on the status of Meloidogyne research in Mexico, Central America and the Caribbean countries. In: SasserJN, CarterCC, eds. An Advanced Treatise On Meloidogyne, vol. 28, 32746.
  • Van de Peer Y & Wachter R, 1994. Treecom for windows: a software package for the construction and drawing of evolutionary trees for the Microsoft Windows environment. Computer Application Bioscience 10, 56970.
  • Vos P, Hogers R, Bleeker M, Reijans M, Van der Lee T, Hornes M, Frijters A, Pot J, Peleman J, Kuiper M, Zabeau M, 1995. AFLP: a new technique for DNA fingerprinting. Nucleic Acids Research 23, 440714.
  • Wang M & Goldman IL, 1996. Resistance to root-knot nematode (Meloidogyne hapla Chitwood) in carrot is controlled by two recessive genes. Journal of Heredity 87, 11923.
Footnotes