Cacao (Theobroma cacao L.) is a neotropical plant spe cies originated from the humid tropics of America, and thus presents great diversity in this zone (Enriquez, 2004; Motamayor et al. 2008). In Colombia, the cacao is cultivated in soils with different biological, physical, and chemical characteristics (Suarez et al., 2010). Most of the soils in the tropical areas present low nutrient availability, which results in low crop productivity (Brady and Weil, 2008; Meason et al., 2009; Osorio, 2011). This condition leads to add high doses of chemical, mineral, and organic fertil izers in order to improve yields, but the excessive use of these supplies can increase productions costs and generate negative environmental impacts (Brady and Weil, 2008; Xiao et al., 2008).

Arbuscular mycorrhizal fungi (AMF) are considered to be of great importance for adequate plant development, since they can improve water and nutrient uptake, particu larly phosphorus (P) (Osorio, 2011; Ramírez et al., 2013; Ramírez et al., 2014; Ramírez et al., 2015). This is relevant because in the tropics approximately 71% of plant species form arbuscular mycorrhizal association, 16% form other types of mycorrhizal associations, and only 13.4% are non-mycorrhizal (Coyne, 2000; Barea and Azcon, 2002). Among the first ones, it is well known the ability of cacao plants to form mycorrhizal associations and increase plant growth (Azizah-Chulan and Martin, 1992).

On the other hand, some bacteria from the Pseudomonas genus are known as plant growth promoters. They can colonize plant root systems and promote plant growth by (i) nutrient solubilization (e.g., P, Fe, among others), (ii) synthesizing hormonal regulators such as auxins and gibberellins (Gunes et al., 2015; Pii et al., 2015) and (iii) decrease phytopathogens populations in the soil and thus prevent plant diseases (Hallmann et al., 1997; Van Veen et al., 1997; Lugtenberg and Kamilova, 2009).

However, few researches exist about the presence of both types of these microorganisms in cacao plantations in contrasting environmental conditions, as those imposed by the TDF and TMF life zones, which certainly limits the application of these microorganisms at a commercial scale (Cuadros et al., 2011). For this reason, the present study had two objectives (i) identifying and quantifying AMF and strains of fluorescent Pseudomonas sp. in the rhizosphere of cacao plants in two agroecosystems of contrasting life zones: tropical dry forest (TDF) and tropical moist forest (TMF); and (ii) to evaluate the effectiveness of inoculation with these two groups of microorganisms on cacao plant growth in greenhouse.

The two agroecosystems exhibited a significant difference in soil parameters; for instance, soil pH in Maceo was extremely acidic (4.7), whereas in Santa Fe the soil pH was neutral (6.7) (Tab. 2). Consistently, the soil solution P in Maceo was very low (0.011 mg L^-1) and in Santa Fe was high (0.240 mg L^-1). Despite of these contrasts, not significant difference was detected in the foliar P concentration of cacao plants. Similarly, there was not significant difference in the CFU of Pseudomonas spp.

The soil test results are consistent with the typical soil fertility parameters commonly found in these two life zones. For instance, Maceo soils are very acidic and poor in available P as a result of leaching out of bases (Ca, Mg, K, Na) produced by a high rainfall regime and is rich in Al as a result of high weathering rates, which generate oxides of Al and Fe that held P in unavailable forms for plant roots. By contrast, Santa Fe soil has a neutral soil reaction and is rich in available P, this associated with low weathering, low rainfall and the consequent dominance of 2:1 clays (e.g., montmorillonite, vermiculite) that retain high amounts of exchangeable bases (Zapata, 2002).

Mycorrhizal colonization was significant different between both agroecosystems, in Maceo was 12.1% and in Santa Fe was 3.7%. Accordingly, the number of AMF spores was significantly higher in Maceo (16.2) than in Santa Fe (12.8).

Both agroecosystems showed similar AMF genera, but their relative distribution was different (Fig. 1). Although, Glomus genus is more abundant than the other genera, in Maceo it represented almost half AMF population, while in Santa Fe it was three quarters of the AMF population. Acaulospora genus was the second in abundance in both agroecosystems (9-24%), followed by Gigaspora (8-11%) and by the other two genera (Scutellospora, Entrophospora). These differences were satisfactorily reflected in Shanon Index (Tab. 2), which was significantly higher in Maceo (1.82) than in Santa Fe (1.30).

FIGURE 1

TABLE 2

Means followed by different letters indicate significant differences according to the Tukey test (P0.05).

The bacterium that was used as control R. solanacearum did not produce indole-acetic acid. In contrast, all Pseudomo nas isolates did produce it, but the concentration produce was variable among them (Fig. 2). The isolate P10 was the most active in this regard (79.9 µg mL^-1) followed by P9, P3 and P8 (48.9-59.3 µg mL^-1), and then by P1, P2, and P3 (29.5-38.9 µg mL^-1). The less effective isolate were P4, P5 and P6 (20.2-25.0 µg mL^-1).

FIGURE 2

The findings of this study coincide with those of Prieto et al. (2012) who found AMF species of Glomus, Scutellospora, Acaulospora and Gigaspora in cacao agroforestry systems placed in a TDF in Ecuador. However, these authors did not report Entrophospora genus. In a study by Cuenca and Meneses (1996) in Venezuela the genus most associated with cacao was Glomus, being G. etunicatum particularly prominent. According to Sieverding (1991), Glomus is one of the most competitive and effective AMF genus in with different botanical groups and different soil conditions. In this study, clearly Glomus is dominant in both agroecosys tems and in the case of any inoculation with AMF, species of Glomus should be considered. This was confirmed with the greenhouse results, in which Glomus was consistently effective to promote plant P uptake and growth (Zandavalli et al., 2004; Osorio 2011; Ramírez et al., 2013; Ramírez et al., 2014; Ramírez et al., 2015).

The MC values reported here coincide with those from Bolívar et al. (2009), who found MC between 2.8 and 13.3% under natural conditions. The former value might be considered low, since Cuenca et al. (1991) reported MC values up to 69%. Meanwhile, in agroforestry systems in a TMF life zone in Ecuador, Prieto et al. (2012) reported that MC did not surpass 3.5%, while values were as low as 0.9%. Perhaps, the high levels of soil available P found in Santa Fe agroecosytems may restrict the extent of mycor rhizal colonization (Peters and Habte, 2001), while the low levels of this in Maceo may promote a high mycorrhizal colonization. Cuenca et al. (1991) reported that in soils with low nutrient availability, cacao plants are highly dependent on AMF. This is because under these conditions, the plant requires this positive symbiosis for the uptake of this nu trient (Bonfante and Genre, 2010; Barea and Pozo, 2013).

Cuadros et al. (2011) reported that in field conditions MC depends largely on soil P content. López etal. (2007), Toro et al. (2008), and Azizah-Chulan and Martin (1992) reported that the recommended commercial dose of P fertilizers for cacao had negative effect on native AMF populations. They recommended adjusting the P fertilization management in order to promote the AMF symbiosis. On the other hand, Tena (2002) affirms that high levels of Ca and Mg (as those found in Santa Fe) may affect the AMF spores density and thus the symbiotic relationship with the host plant.

Despite the presence of fluorescent Pseudomonas sp., we did not find any promoting effects on cacao plants in the greenhouse experiment. It is not clear if in the two agro-ecosystems they had a role on plant nutrition or growth, it has been reported that they are biocontrol agents and plant growth promoters (Hallmann et al., 1997; Van Veen et al., 1997; Lugtenberg and Kamilova, 2009). Notice that in cacao plants at Maceo the foliar P was as good as in Santa Fe, in despite of that the former one was grown in a P deficient soil. It may be explained by the high levels of MC, the role of this bacterium in those systems should be investigated.

Uninoculated plants (control) showed a plant dry weigh of 150.9 g, which was significantly increased (+36%) only by Glomus inoculation. Other inoculation did not increase significantly the plant growth of cacao plant, included the co-inoculation G+P10 (Fig. 3A).

FIGURE 3

Control plants exhibited a foliar P concentration of 0.43%, this was significantly higher when the mycorrhizal fungi were inoculated (Fig. 3B). Inoculated plants with Acau-lospora had foliar P concentration of 0.67% and those inoculated with Glomus had a concentration even higher 0.92%, which represented relative increases of 56 and 114%, respectively. The inoculation with Pseudomonas sp. (P10) was ineffective to promote foliar P (0.51%). Moreover, the co-inoculation with P10 was not effective to increase the foliar P beyond that obtained with the AMF inoculation only. In this way, the foliar P concentration of plants treated with A+P10 was 0.73% (similar to A-inoculation) and with G+P10 was 0.86% (similar to G-inoculation).

Uninoculated plants and those inoculated with P10 did not have mycorrhizal colonization (Fig. 3C). In contrast, the plants inoculated with Acaulospora (without and with P10) showed a mycorrhizal colonization that ranged be tween 15.8-13.4%, respectively, which did not differ to each other. Otherwise, the inoculation with Glomus had values of mycorrhizal colonization significantly higher than the other with values that fluctuated between 32.5-37.9 %, without and with P10, respectively.

Inoculation of the IMC-67 clone with Glomus sp. resulted in higher MC values than those reported by Ballesteros et al. (2004) (ca. 20%) with native Glomus and Acaulospora. Similarly, Cuadros et al. (2011) reported MC values of 84% with both AMF genera. These results indicate that the MC of cacao roots by AMF inoculated is variable. From our perspective, this variable (MC) may be secondary, and it is necessary to focus on the effectiveness of AMF inocula tion on plant nutrient uptake and growth or even better on crop yield.

In this study it is clear that, at least for cacao seedlings, the use of AMF inoculation promote growth, as reported by Azizah-Chulan and Martin (1992). However, this effect seems to be species-dependent. Thus, species of Glomus must be included in a commercial inoculum intended to promote seedling growth and P uptake. The benefits ob tained with Glomus inoculation are quite relevant since the P foliar was twice higher with this than in control plants. On the other hand, the effect of Pseudomonas sp. P10 was null alone or in combination with AMF.

In both cacao agroecosystems, AMF and Pseudomonas were detected. The same AMF genera was found but in different relative distribution. The AMF genus Glomus was predominant in both agroecosystems. On the other hand, the number of CFU of Pseudomonas spp. was similar in both agroecosystems. In the greenhouse, cacao seedlings had higher levels of foliar P when inoculated with AMF (Acaulospora and Glomus) than uninoculated plants, but the effect was higher with Glomus species. Plant growth was only significantly promoted by Glomus inoculation and not by Acaulospora. The inoculation with the isolate P10 of Pseudomonas did not promote plant growth and P uptake in mycorrhiza-free or mycorhizal plants.