Effect of ammonium/nitrate ratio in nutrient solution on control of Fusarium wilt of tomato by Trichoderma asperellum T34

Authors

  • C. Borrero,

    Corresponding author
    1. Department of Ciencias Agroforestales, Escuela Técnica Superior de Ingeniería Agronómica, Universidad de Sevilla, Ctra Utrera km 1, E-41013 Sevilla
      E-mail: cborrero@us.es
    Search for more papers by this author
  • M.I. Trillas,

    1. Department of Biologia Vegetal, Facultat de Biologia, Universitat de Barcelona, Avda Diagonal 645, E-08028 Barcelona, Spain
    Search for more papers by this author
  • A. Delgado,

    1. Department of Ciencias Agroforestales, Escuela Técnica Superior de Ingeniería Agronómica, Universidad de Sevilla, Ctra Utrera km 1, E-41013 Sevilla
    Search for more papers by this author
  • M. Avilés

    1. Department of Ciencias Agroforestales, Escuela Técnica Superior de Ingeniería Agronómica, Universidad de Sevilla, Ctra Utrera km 1, E-41013 Sevilla
    Search for more papers by this author

E-mail: cborrero@us.es

Abstract

The behaviour of Fusarium oxysporum f.sp. lycopersici (Fol) and the effectiveness of the microbial control agent Trichoderma asperellum strain T34 were examined in hydroponically grown tomato plants under five ammonium/nitrate ratios. The results showed that disease severity was reduced by the action of T34 under increasing concentrations of ammonia. Furthermore, rhizosphere F. oxysporum populations decreased with T34 application. The presence of T34 augmented leaf nitrogen concentration in treatments infested with Fol. In addition, T34 application reduced iron concentration in tomato leaves at high ammonium/nitrate ratios and reduced the severity of Fusarium wilt at high iron and nitrogen leaf concentrations.

Introduction

Fusarium wilts, induced by pathogenic formae speciales of the soil-inhabiting fungus Fusarium oxysporum, cause severe losses in a wide variety of crops (Michielse & Rep, 2009). Maintaining control of the pH above 6 and fertilization practices that promote decreased nutrient availability to pathogenic fungi may result in a significant decrease in the incidence and severity of these diseases (Woltz & Jones, 1981). On the other hand, high nitrogen doses and high ammonium to nitrate ratios increase Fusarium wilt incidence and severity (Woltz & Jones, 1981; Jones et al., 1993). In this sense, nitrate applications may help to manage Fusarium wilt diseases in ornamental (carnation, chrysanthemum) and horticultural crops (cucumber, tomato, asparagus, pea, radish, etc.) (Huber & Thompson, 2007). However, the use of ammonium as a nitrogen source is necessary in hydroponic crops to prevent restrictions in micronutrient availability as a result of alkalinization caused by nitrates (Kosegarten et al., 1999, 2001). Indeed, Peterson (1980) stated that for inert and organic soilless media, optimum pH values are about 1 unit lower than the optimum pH characterized for mineral soils. Micronutrient availability also affects the incidence of disease, e.g. low iron levels in the growing medium reduced F. oxysporum growth and sporulation and were associated with suppression of Fusarium wilt of tomato (Woltz & Jones, 1981; Jones et al., 1993; Borrero et al., 2004).

Microbial control agents in the genus Trichoderma are used commercially to control plant diseases (Verma et al., 2007). Trichoderma asperellum strain T34, isolated from Fusarium-suppressive compost (Trillas & Cotxarrera, 2003), has been reported to control Fusarium wilt in tomato and carnation plants (Cotxarrera et al., 2002; Sant et al., 2010) and Rhizoctonia solani in cucumber plants (Trillas et al., 2006).

Effects of Trichoderma spp. on other microorganisms include mycoparasitism, antibiosis, and nutrient or space competition, as well as competition for key exudates from seeds that stimulate the germination of propagules of plant pathogenic fungi in soil. Furthermore, they inhibit or degrade pectinases and other enzymes that are essential for plant-pathogenic fungi and may also influence disease development by inducing resistance in and/or promoting growth of the plant host (Harman et al., 2004; Verma et al., 2007). Trichoderma spp. are considered avirulent plant symbionts. Some strains establish robust and long-lasting colonizations on root surfaces and penetrate into the epidermis and a few cells below this level (Harman et al., 2004). T34, when applied to roots, was also able to induce plant resistance and reduce proliferation of foliar pathogens (Segarra et al., 2007, 2009).

However, mechanisms of disease control by Trichoderma spp. are not completely identified and are strain-specific (Vinale et al., 2008). Thus, studies of basic interactions between target pathogens and this microbial control agent are still necessary. Because of the significance of nitrogen rates and the chemical forms in which this nutrient is supplied to plants and microorganisms, it is also necessary to determine the effect of nitrogen management on the control of plant diseases by Trichoderma spp. in order to establish the best fertilizer management and the most effective disease control strategies. The aim of this study was to determine the role of the chemical forms in which nitrogen is supplied to hydroponic crops in the control of Fusarium wilt in tomato plants by T34. In these studies, the effectiveness of T34 for controlling Fusarium wilt of tomatoes grown on perlite was evaluated for different nutrient solutions with a range of ammonium/nitrate ratios.

Materials and methods

Inoculum preparations

Trichoderma asperellum strain T34 was recovered from stored cultures (silica gel 4°C) and grown in malt broth (10 g L−1) in a 10 L Biostat B fermenter (Braun Biotech-Sartorius). Perlite for T34 treatments was inoculated with T34 as described below, incubated for 14 days and fertilized with half-concentration Hoagland’s nutrient solution (Hoagland & Arnon, 1938) at 1, 4 and 7 days after inoculation. The final concentration of T34 was 5·5 × 105 conidia mL−1 perlite, determined by dilution plating on Trichoderma medium based on Chung & Hoitink (1990). The medium contained 3 g glucose, 1 g NH4NO3, 0·9 g KH2PO4, 0·2 g MgSO4·7H2O, 0·15 g KCl, 20 mg FeSO4·7H2O, 20 mg MnSO4·H2O, 20 mg ZnSO4·7H2O, 30 mg Rose bengal and 20 g agar per L distilled water. After autoclaving, 50 mg chloramphenicol, 50 mg streptomycin sulphate, 1 mL tergitol and 0·1 g PCNB (99% a.i.) were added.

The pathogen used in the experiment was a monosporic isolate of F. oxysporum f.sp. lycopersici (Fol) race 2, isolate FN2, which was grown for 7 days in AMAP culture medium (Borrero et al., 2009). This medium contained 10 g agar, 10 g malt extract, 2·0 g l-asparagine and 0·5 g Peter’s foliar feed 27-15-12 and micronutrients (Scotts) per L distilled water. Sterile water (5 mL) was added to each culture plate, and after that the surface of the culture was scraped with a sterile bent metallic rod. The suspension of Fol was transferred to MAP liquid culture (as described above, but without agar) (Borrero et al., 2009) and grown in continuous agitation (132 rpm) for 10 days at 25°C. After that, the liquid culture was filtered through two layers of cheesecloth with the help of a Büchner funnel and a vacuum pump. After filtration, the suspension was centrifuged at 2000 g for 15 min to concentrate the conidia in a pellet. The supernatant was discarded and the conidia were washed with sterile distilled water. This procedure was repeated three times. The concentration of conidia was determined with a haemocytometer according to the recommendations of French & Hebert (1980). Perlite was infested with Fol inoculum (105 conidia mL−1 plant growth medium). All perlite was watered to reach a 60:40 (w/w) mixture of water:perlite and mixed overnight with an overhead shaker (Heidolph Reax 20).

Pathogenicity experiments

The experiment had a completely randomized balanced design with five replications involving three factors: (i) ammonium to nitrate ratio (0, 0·07, 0·25, 0·5 and 1·5), (ii) infestation with Fol (infested and noninfested growing media) and (iii) inoculation with T34 (inoculated and noninoculated). The experiment was performed twice at different times. The composition of the nutrient solutions is shown in Table 1.

Table 1.   Composition of the five different nutritive solutions used in the experiments
Compound (mmol L−1)Ammonium/nitrate ratio
0·000·070·250·501·50
Ca(NO3)255553
KNO354200
NH4Cl01359
KCl01355
CaCl200002
NaCl18161280
KH2PO411111
MgSO422222
CuSO40·00160·00160·00160·00160·0016
H2MoO40·00230·00230·00230·00230·0023
H3BO30·0920·0920·0920·0920·092
ZnSO40·00250·00250·00250·00250·0025
Fe.EDDHA0·010·010·010·010·01
MnCl20·0180·0180·0180·0180·018

Tomato (Solanum lycopersicum cv. Roma) seeds were germinated in moistened perlite. After 10 days, plants with two to three leaves were transplanted into pots (330 mL) containing experimentally treated perlite. Four tomato plants were established in each pot, which served as the unit of replication. Pots were fertigated daily with 50 mL nutrient solution (pH 6) with different ammonium/nitrate ratios (Table 1); the total nitrogen rate was the same in all cases (Table 1). The experiment was performed in a growth chamber (27°C, with a photosynthetically active radiation intensity of 280 μE m−2 s−1 and a 16:8 h light:dark photoperiod).

Disease severity was monitored at 2 day intervals over 35 days and was scored on a symptom severity scale where: 0 =  symptomless plants; 1 = ≤50% of leaves chlorotic or wilted; 2 = >50% of leaves wilted but plants not dead; and 3 = dead plants. Mean disease severity per pot was calculated. Disease severity was expressed as the proportion of the maximum possible disease severity. The standardized area-under-the-disease-progress-curve (AUDPC) per pot was calculated by disease severity integrated between symptom onset and experiment end times and divided by the total duration (days) of the epidemic in each experiment. In this way, the two experiments performed, which had different epidemic durations, could be compared (Campbell & Madden, 1990). At the end of the experiments, mean values of the fresh weight of leaves and stems were recorded for plants in each pot (= 4). These values were considered as replicates for each observed variable.

Populations of Fol and T34

The densities of Fol and T. asperellum in the plant growth medium after each experiment were determined by dilution plating on two semiselective media: Komada’s medium and Trichoderma medium. Komada’s medium, based on Dhingra & Sinclair (1995), contained 2 g l-asparagine, 20 g d-galactose, 1 g K2HPO4, 0·5 g MgSO4·H2O, 0·5 g KCl, 50 mg Fe(EDTA) and 20 g agar per L distilled water. After autoclaving, a solution with 0·5 g oxgall, 500 mg streptomycin sulphate, 1 g NaB4O7.10H2O, 1·25 g PCNB and 20 mL water was adjusted to pH 3·8 with phosphoric acid and added to the culture medium. Samples were taken from the rhizosphere at the end of the two experiments. For each treatment, perlite (0·5–1 g) was suspended in 10 mL water agar (2 g L−1). The suspension was shaken and a 10-fold dilution series (from 10−1 to 10−4) was prepared with water agar (2 g L−1). Suspensions were pipetted onto three plates per dilution. Colony forming units (CFU) were counted 4 days after plating and expressed as CFU mL−1 perlite sampled. Fol FN2 colony morphology was distinctive enough to differentiate it from other Fusarium spp.

Nutrient status of plants

Total nitrogen in plant leaves (using 0·1 g dry weight) was determined with sulphamic acid modification of the Kjeldahl method, as described by Mulvaney (1996). Iron availability to plants can also be affected by nitrogen forms in the growing medium (Kosegarten et al., 2001). Therefore, the content of this nutrient in leaves was also determined. For this purpose, leaf samples (0·25 g dry weight) were incinerated at 550°C for 10 h, and the ashes dissolved in 10 mL HNO3 (1 m), filtered and stored at 4°C until analysis. Iron was determined by atomic absorption spectroscopy (Wright & Stuczynski, 1996). One sample per pot was analysed at the end of each of the two experiments.

Statistical analysis

All data were analysed with Statgraphics Plus (version 5·1; Statistical Graphics Corp., 2002). The effects of T34 inoculation, ammonium/nitrate ratio and experiments on response variables were tested using general linear models analysis (three-way anova where the quantitative factor was analysed using orthogonal polynomial contrasts). In preliminary analyses all interactions between factors studied were included in the model. Interaction terms found not to be significant were removed from the final models. For qualitative factors with a significant effect, means were compared by Tukey’s test: differences were considered significant for ( 0·05). T34 inoculation and ammonium/nitrate ratio were treated as fixed effects and experiment as a random effect. Data for AUDPC were transformed for analysis with the arcsine √x; data for Fol populations were transformed for analysis with ln (x) and data for Trichoderma populations were transformed with x0·8.

Relationships between AUDPC and leaf nitrogen and iron concentrations in Fol- infested treatments were analysed with regression analysis.

Results

Disease severity (AUDPC) of Fusarium wilt in tomato plants increased with increasing ammonium/nitrate ratios in the nutrient solution (Fig. 1). At ammonium/nitrate ratios higher than 0·07, disease progress was different for perlite inoculated with T34 than without T34 (Fig. 1), justifying the exclusion of the treatment without ammonium supply in the corresponding statistical analysis. Arcsine (inline image) increased linearly with higher ammonium/nitrate ratios. This linear relationship was significantly different with or without T34 inoculation (significant interaction Rt L × T34, Table 2). This interaction reveals that T34 reduces severity of Fusarium wilt of tomato with different efficiency depending on the proportion of different nitrogen forms in the solution (Fig. 1). The T34-noninoculated and Fol-noninfested controls and plants grown in Fol-noninfested but T34-inoculated perlite showed no symptoms of Fusarium wilt disease. The application of T34 in the plant growth medium did not produce any evidence of phytotoxicity in the tomato plants.

Figure 1.

 Disease severity standardized area-under-disease-progress-curve (AUDPC) for tomato plants cultured in perlite inoculated or not with Trichoderma asperellum T34 and infested with Fusarium oxysporum f.sp. lycopersici at different ammonium/nitrate ratios at the end of the experiment.

Table 2.   Effects of three factors: T34 treatment, ammonium/nitrate ratio and experiment for the standardized area-under-disease-progress-curve (AUDPC) for tomato plants cultured in perlite inoculated with Fusarium oxysporum f.sp. lycopersici after 35 days of experiment
FactorsVariable AUDPCa
F-ratioP-value
  1. aF ratios and significance levels of the three-way anova factors: T34 treatment, ammonium/nitrate ratio and experiment, and their interactions. Ammonium/nitrate ratio was analysed using orthogonal polynomial contrasts. Disease severity scale was from 0 (symptomless plants) to 3 (dead plants). Data for AUDPC were normalized for analysis with the arcsine (inline image), = 5.

  2. bT34 = Trichoderma asperellum T34 treatment.

  3. cRt = ammonium/nitrate ratio.

  4. d= linear response.

T34b49·52<0·0001
Rtc Ld153·19<0·0001
Experiment1·890·1731
T34 × Rt L15·590·0002

Rhizosphere Fol populations at the end of the experiments were higher without T34 than with T34 and showed maximum values at a 0·5 ammonium/nitrate ratio, with or without T34 (Table 3). These populations responded linearly and quadratically to increasing ammonium/nitrate ratio (Table 3). The regression analysis between these two variables showed a significant positive relationship between the size of the Fol population and the ammonium/nitrate ratio (R= 0·25; = 0·0258). By contrast, populations of T. asperellum were lower without infestation with Fol (Table 3); however, these populations increased with increasing ammonium/nitrate ratios (linear response). No significant interactions between studied factors were found in a preliminary analysis. Therefore, interactions were removed from the model. The mean Fol population in perlite inoculated with T34 decreased, while without T34 the Fol infestation doses at the beginning of experiments (105 conidia mL−1) were maintained. Conversely, the mean T34 population increased with and without Fol infestation in comparison to inoculation doses at the beginning of experiments (5·5 × 105 conidia mL−1), although this increase doubled with Fol infestation. This indicates that the F. oxysporum population was reduced by the presence of T34 and that T34 population growth was favoured by the presence of Fol.

Table 3. Fusarium oxysporum f.sp. lycopersici (Fol) and Trichoderma asperellum T34 populations in perlite in which tomato plants were cultured, and infested or not with Fol and inoculated or not with T34, at the end of the experiment. Ammonium/nitrate ratio mean values and standard errors are given
Ammonium/nitrate ratio (Rt)Populations of Fola (×104 CFU mL−1 perlite)Populations of T34a (× 106 CFU mL−1 perlite)
T34 treatment (T34)Fol infestation (F)
++
  1. aF ratios and significance levels of the three-way anova factors: T34 treatment or Fol infestation ammonium/nitrate ratio, and experiment. Ammonium/nitrate ratio was analysed using orthogonal polynomial contrasts. Disease severity scale was from 0 (symptomless plants) to 3 (dead plants). Data for Fol populations were transformed for analysis with ln (x) and data for Trichoderma populations were transformed with x0·8, = 4. Within a line, numbers followed by different letters are significantly different (< 0·05, Tukey’s test).

  2. bT34 = T34 treatment.

  3. cFol = Fol infestation.

  4. dRt = ammonium/nitrate ratio.

  5. e= linear response.

  6. f= quadratic response.

  7. gNI = not included. In preliminary analyses all terms (factors and interactions) were included in the model. Terms found not to be significant were removed from final models.

0·000·44 ± 0·237·33 ± 0·713·84 ± 3·833·49 ± 2·81
0·072·61 ± 1·633·87 ± 2·558·50 ± 4·371·70 ± 1·10
0·254·22 ± 0·695·49 ± 1·226·64 ± 0·822·05 ± 1·74
0·509·43 ± 6·3222·38 ± 6·598·28 ± 4·582·78 ± 2·15
1·506·91 ± 4·4417·81 ± 6·9312·37 ± 0·435·15 ± 3·47
T34 treatment4·72 ± 1·58 b11·38 ± 2·87 a  
Fol infestation  7·93 ± 1·45 a3·03 ± 0·90 b
anova resultsF-ratioP-valueF-ratioP-value
T34b9·460·0077  
Folc  9·630·0068
Rtd Le8·270·01154·670·0462
Rt Qf7·880·0133 NIg
Experiment0·340·56771·680·2127

Plant fresh weight for Fol-noninfested plants was reduced at increased ammonium/nitrate ratios (linear response) and T34 treatment had no significant effect on fresh weight (Table 4). The interaction between T34 presence and the ammonium/nitrate ratio linear response was significant for the treatments inoculated with Fol; at a lower ammonium/nitrate ratio, T34-inoculated plants showed reduced plant fresh weights. At a higher ammonium/nitrate ratio, T34-inoculated plants showed increased plant fresh weights and at an intermediate ammonium/nitrate ratio, T34-inoculated plants showed similar values of plant fresh weight to uninoculated plants (Table 4).

Table 4.   Fresh weight of tomato plants (aerial part) cultured in perlite infested or not with Fusarium oxysporum f.sp. lycopersici and inoculated or not with Trichoderma asperellum T34, at the end of the experiment. Fresh weight mean values and standard errors are given
Ammonium/nitrate ratio (Rt)Fresh weight of tomato plants (g)a
Fol infestation (F)
+
T34 treatmentT34 treatment
++
  1. aF ratios and significance levels of the three-way anova factors: T34 treatment, ammonium/nitrate ratio, and experiment, and their significant interactions, = 5. Ammonium/nitrate ratio was analysed using orthogonal polynomial contrasts.

  2. bT34 = T34 treatment.

  3. cRt = ammonium/nitrate ratio.

  4. d= linear response.

  5. eNI = not included. In preliminary analyses all terms (factors and interactions) were included in the model. Terms found not to be significant were removed from final models.

0·009·21 ± 0·7011·16 ± 0·6410·70 ± 0·4612·74 ± 0·39
0·079·70 ± 0·6711·93 ± 0·5211·87 ± 0·5214·29 ± 0·27
0·2510·47 ± 1·0210·42 ± 0·3912·10 ± 0·6413·07 ± 0·38
0·508·51 ± 0·499·37 ± 0·6010·77 ± 0·6811·53 ± 0·43
1·504·75 ± 0·443·08 ± 0·537·50 ± 0·498·54 ± 0·49
T34 treatment8·54 ± 0·429·32 ± 0·5010·65 ± 0·3412·11 ± 0·32
anova resultsF-ratioP-valueF-ratioP-value
T34b3·950·05012·000·3920
Rtc Ld160·18<0·0001142·53<0·0001
Experiment0·040·84831·620·4244
T34 × Rt L10·900·0014 NIe

Leaf nitrogen concentrations in treatments infested with Fol increased significantly with T34 application (Table 5). For these treatments, the nitrogen concentration increased with increasing ammonium/nitrate ratio (linear response). This indicates that the presence of T34 and the increase in the ratio augments nitrogen content. For these treatments, no significant interactions between studied factors were found in a preliminary analysis. Therefore, interactions were removed from the anova model. In treatments not inoculated with Fol, there was significant interaction between T34 application and the ammonium/nitrate ratio (Table 5), showing that, at a low ammonium/nitrate ratio, T34 was associated with higher values of nitrogen concentration. In contrast, at a high ammonium/nitrate ratio, T34 presented lower values of nitrogen concentration.

Table 5.   Total nitrogen and iron concentrations in leaves of tomato plants cultured in perlite infested or not with Fusarium oxysporum f.sp. lycopersici (Fol) and inoculated or not with Trichoderma asperellum T34, at the end of the experiment. Mean values for these elements and standard errors are given
Ammonium/nitrate ratio (Rt)Total nitrogena (mg g−1 leaf dry weight)Total irona (× 10−2 mg g−1 leaf dry weight)
Fol infestationFol infestation
++
T34 treatmentT34 treatmentT34 treatmentT34 treatment
++++
  1. aF ratios and significance levels of the three-way anova factors: T34 treatment, ammonium/nitrate ratio and experiment, and their significant interactions, = 5. Ammonium/nitrate ratio was analysed using orthogonal polynomial contrasts. For every two sets of data, within a line, numbers followed by different letters are significantly different (< 0·05, Tukey’s test).

  2. bT34 = T34 treatment.

  3. cRt = ammonium/nitrate ratio.

  4. d= linear response.

  5. e= quadratic response.

  6. fNI = not included. In preliminary analyses all terms (factors and interactions) were included in the model. Terms found not to be significant were removed from final models.

0·0046·95 ± 0·4644·62 ± 0·8545·33 ± 0·6544·92 ± 1·076·94 ± 0·336·58 ± 0·236·57 ± 0·247·17 ± 0·28
0·0746·28 ± 0·9044·27 ± 0·5445·32 ± 0·9643·97 ± 0·717·00 ± 0·276·99 ± 0·196·38 ± 0·227·54 ± 0·83
0·2547·58 ± 1·0947·17 ± 0·7446·54 ± 0·5947·80 ± 0·766·19 ± 0·266·82 ± 0·315·42 ± 0·316·19 ± 0·25
0·5050·54 ± 1·3650·85 ± 0·6547·54 ± 0·5552·11 ± 0·676·37 ± 0·399·62 ± 0·715·05 ± 0·196·94 ± 0·44
1·5062·52 ± 1·4659·53 ± 1·0658·35 ± 2·0063·66 ± 0·9113·06 ± 1·0017·34 ± 1·298·51 ± 0·3610·08 ± 0·30
T34 treatment50·68 ± 1·0 a49·18 ± 0·9 b48·62 ± 0·8 b50·49 ± 1·1 a7·91 ± 0·4 b9·47 ± 0·6 a6·38 ± 0·2 b7·5 ± 0·3 a
anova resultsF-ratioP-valueF-ratioP-valueF-ratioP-valueF-ratioP-value
T34b8·080·00559·130·003215·100·000216·130·0001
Rtc Ld369·69<0·0001372·7<0·000143·450·096148·41<0·0001
Rt Qe NIf NI15·540·0002 NI
Experiment3·320·07170·480·49210·720·54062·410·1242
T34 ×  Rt L NI13·040·000521·92<0·0001 NI

Iron concentration in leaves was significantly affected by the forms of nitrogen applied in the nutrient solution, reaching its peak with the highest ammonium/nitrate ratio (Table 5). In treatments infested with Fol, there was a significant interaction between T34 application and the ammonium/nitrate ratio, showing that, at a high ammonium/nitrate ratio, iron concentrations in leaves were lower with T34. For treatments not infested with Fol, T34 significantly reduced iron concentrations.

Positive correlations were found for both studied nutrients (nitrogen and iron) between their concentration and the disease severity caused by Fol (Figs 2 and 3). The equations obtained were: AUDPC = (−1·326 + 0·034 N concentration)2 and AUDPC = 1·209 + 0·419 ln(Fe concentration) for treatments without T34 application and AUDPC = −0·326 + 0·008 N concentration and AUDPC = 0·518 + 0·172 ln (Fe concentration) for treatments with T34 application. The higher the iron or nitrogen concentration, the greater severity, although this severity was reduced when T34 was present.

Figure 2.

 Correlation between standardized area-under-disease-progress-curve (AUDPC) and nitrogen concentration in leaves from tomato plants cultured in perlite infested with Fusarium oxysporum f.sp. lycopersici at different ammonium/nitrate ratios for fertigation, with the presence or absence of the microbial control agent Trichoderma asperellum T34, after 35 days of experiment.

Figure 3.

 Correlation between standardized area-under-disease-progress-curve (AUDPC) and iron concentration in leaves of tomato plants cultured in perlite infested with Fusarium oxysporum f.sp. lycopersici at different ammonium/nitrate ratios for fertigation, with the presence or absence of the microbial control agent Trichoderma asperellum T34, after 35 days of experiment.

Discussion

According to previous studies, incidence and severity of Fusarium wilt increases with increased ammonium/nitrate ratios (Woltz & Jones, 1981; Jones et al., 1993). In the present study, the microbial control agent T34 was effective in decreasing the severity of disease, except at the lowest ammonium to nitrate ratio tested (0·07), without a significant reduction in the fresh weight of tomato plants or in the total nitrogen content of plants inoculated with Fol. The reduction of Fol populations by T34 indicates interference between these two microorganisms (nutrient competition, antibiosis or mycoparasitism). This Fol root population reduction was also found with other Trichoderma spp. isolates (Adebayo & Ekpo, 2004). The greater T34 populations with Fol than without Fol indicate that this phytopathogen could be a substrate for T34 or that Fol can change the rhizospheric environment, improving it for T34. Fol infection causes xylem obstruction and therefore lower plant transpiration and water absorption (Beckman, 1987). Consequently, nutrient uptake is reduced as well. Thus, higher nutrient concentrations in the rhizosphere can favour T34. All this evidence is consistent with the hypothesis that T34 acts as a biological control agent of F. oxysporum, through competition and/or parasitism. These mechanisms have been described for other Trichoderma isolates (Vinale et al., 2008). Sivan & Chet (1989) demonstrated that competition for nutrients is the major mechanism used by Trichoderma harzianum (T-35) to control F. oxysporum f.sp. melonis. Moreover, Trichoderma has a strong capacity to mobilize and take up soil nutrients, thus making it more efficient and competitive than many other soil microbes (Benítez et al., 2004).

Populations of Fol and T34 were significantly affected by the ammonium/nitrate ratio. The maximum population of Fol was found at the ammonium/nitrate ratio of 0·5, regardless of the presence or absence of T34. In this sense, the correlation observed between Fol populations and ammonium/nitrate ratios indicates lower Fol population growth at higher ratios. These populations were higher without T34 than with T34. However, the maximum population of T34 was observed at the ratio 1·5, likewise regardless of the presence or absence of Fol. Although these data reveal different population dynamics for both fungi at increasing ammonium/nitrate ratios, they support the effectiveness of T34 in controlling Fol at the highest ammonium/nitrate ratio studied. It should be noted that the populations of T34 observed in perlite were much greater than those usually described in inoculated organic plant growth media (Trillas et al., 2006) where the presence of and interaction with autochthonous microorganisms limit the growth of T34 populations.

The increased iron concentration at the highest ammonium/nitrate ratio can be ascribed to a lower dry matter accumulation. The decreased iron concentration in leaves with T34 treatment revealed that T34 competes with the plant for this nutrient, which is consistent with previous works by de Santiago et al. (2009) and Segarra et al. (2010). The competition for iron is more evident at the highest ammonium/nitrate ratio, which promotes the greatest T34 population in the growing medium. Fol infestation did not result in a decreased iron concentration in plants, revealing that the pathogen has less efficient iron acquisition mechanisms than T34, which may also help to explain the control of this disease by T34 (Altomare et al., 1999; Verma et al., 2007).

The higher nitrogen and iron concentrations in T34-inoculated plants might indicate high availability for both nutrients, rendering no effect on fresh weight. Minor slopes from the correlations found between concentrations of both nutrients in plants and disease severity corresponded to treatments with T34. This indicates that T34 is more effective with high availability of both nutrients, for the studied availability interval.

In conclusion, T. asperellum T34 was found to be an effective microbial control agent against Fusarium wilt of tomato grown in perlite at different ratios of ammonium to nitrate applied as sources of nitrogen in the nutrient solution. Ammonium is needed in hydroponics to avoid restrictions in micronutrient supply (particularly iron) to plants, but its application exacerbates Fusarium wilt. The use of T34 as a microbial control agent may have important practical consequences in hydroponic cultures, allowing the use of ammonium sources without an excessive risk of increased Fusarium wilt severity.

Acknowledgements

This research was supported by grants from: Ministerio de Ciencia y Tecnología, Dirección General de Investigación (AGL2002-04313-C03-03); Ministerio de Educación y Ciencia (AGL2005-08137-C03-02); Consejería de Innovación, Ciencia y Empresa de la Junta de Andalucía (P06-AGR-02313); and Ministerio de Ciencia e Innovación (AGL2008-05414-C03-01/AGR) of Spain. We thank I. Ibáñez and S. Castillo for excellent technical assistance and J. Ordovás for critically reviewing the manuscript.

Ancillary