Selection and Evaluation of Phosphate-Solubilizing Fungal Consortia Inoculated into Three Varieties of Coffea arabica Under Greenhouse Conditions

Abstract

Phosphorus-solubilizing fungi represent a viable alternative to traditional fertilizers for use in coffee cultivation. The aim of this work was to select fungal consortia with a high phosphorus-solubilizing capacity for application to three varieties of coffee plants under greenhouse conditions. The research comprised three phases: Firstly, solubilizing strains were identified morphologically and molecularly. Secondly, compatibility tests were carried out to select combinations of phosphorus-solubilizing fungi. The selection of the consortia was evaluated based on their phosphorus-solubilizing capacity, and the consortia with the solubilizing activity were chosen for application to coffee plants. In the greenhouse phase, three coffee varieties were inoculated; the treatments involved single, dual, and triple inoculation, as well as a control without fungi. Five species were identified: Fusarium crassum, F. irregulare, Leptobacillium leptobactrum, Penicillium brevicompactum, and Trichoderma spirale, plus one strain of Absidia sp. The in vitro phase of the study revealed that 11 consortia demonstrated compatibility, and their phosphorus solubilization capacity and phosphatase activity were evaluated. As a result, four consortia with high phosphorus solubilization capacity were selected for inoculation on coffee plants. The greenhouse phase results showed that the three coffee varieties inoculated in consortia showed higher phosphorus availability in the substrate and significant growth.

1. Introduction

Coffee cultivation is of great social, economic, and environmental importance in Mexico. Currently, the country is ranked thirteenth in the world in terms of coffee production [1]. Veracruz is the second-largest producer in the country, mainly growing varieties of Coffea arabica L., such as Typica, Bourbon, Caturra, Garnica, Mundo Novo, Catuaí, and Pacamara [2]. In recent years, rust-resistant varieties such as Anacafe, Costa Rica, and Marsellesa have been introduced.

In Veracruz, coffee is mainly cultivated in soils of volcanic origin, with high levels of metal ions such as Al, Ca, and Fe [3]. These ions interfere with phosphorus (P) availability, making them limiting nutrients for coffee production. In coffee plantations with P deficiency, the application of phosphorus-solubilizing fungi (PSF) is an effective alternative to improve the nutrition of coffee plants.

Besides nitrogen (N), P is the second most important macronutrient for coffee plants, as it is involved in growth, flowering, production, and fruit ripening. P deficiency is addressed through the application of chemical fertilizers; however, their use can generate the risk of environmental contamination, and their inefficiency must also be considered, as only 10–15% of the fertilizer is absorbed by plants, while the rest is fixed in the soil [4].

In the soil, there are populations of microorganisms involved in the transformation of organic and inorganic compounds. The interactions among this group of functional organisms, such as nitrogen fixers, cellulolytic organisms, and phosphorus mobilizers and solubilizers, directly impact the growth and development of plant species [5].

PSF naturally recycle P from the soil through mineralization and solubilization, making it available to plants (Figure 1). Through the mineralization process, PSF produce phosphatase enzymes that dephosphorylate organic compounds present in the soil by breaking phosphoester or phosphoanhydride bonds. Among the diversity of phosphatases produced by PSF, acid and alkaline phosphatase enzymes are the most abundant in nature [6], with acid phosphatases being predominant in acidic soils such as the coffee-growing soils of Veracruz [7,8].

During solubilization, PSF release organic acids that chelate Fe, Al, and Ca metal ions, producing phosphates [9]. Additionally, the active functional groups of organic acids can also effectively chelate with metal cations (Ca²⁺, Fe³⁺, Al³⁺, etc.), thereby promoting the release of phosphorus. Several genera of PSF have been identified; among the genera with the highest phosphorus-solubilizing capacity are Aspergillus and Penicillium [10].

In general, the in vitro solubilization capacity of PSF has been evaluated for individual strains. However, the use of fungal consortia composed of two or more species, as opposed to the exclusive use of single strains, could optimize their solubilization efficiency. The positive effects of the individual inoculation of PSF on coffee plants, under both greenhouse and field conditions, has been documented in several studies [11,12,13,14]. However, few studies have used PSF consortia. Perea-Rojas et al. [15] performed both individual and consortia inoculations in climatic chambers, while Cisneros et al. [16] performed them in greenhouses.

The aim of this research was to select compatible consortia with two or more species that promote both increased P solubilization and plant development for three coffee varieties under greenhouse conditions.

2. Materials and Methods

2.1. Reactivation of PSF Strains

In March 2022, six strains were isolated from the soil of a coffee plantation at Finca Santa Rosa (19.39902 N; 96.99313 W). The strains were deposited in the strain collection of the Centro de Micología Aplicada at Universidad Veracruzana. To reactivate them, they were transferred to potato dextrose agar solid culture medium (PDA, BIOXON, Jalisco, Mexico) and incubated (Thermo Fisher Scientific, Waltham, MA, USA) at 25 °C for 15 days.

2.2. Molecular Identification of PSF Strains

DNA was extracted directly from the mycelia of pure cultures using the Wizard^® kit (Promega, Madison, WI, USA), according to the manufacturer’s instructions. Amplification was performed with ITS5/ITS4 primers by polymerase chain reaction, using MyTaq DNA polymerase. The PCR products were sequenced at Macrogen, Inc. (Seoul, Korea). The BLASTn algorithm was used to perform a genetic comparison of the obtained sequences with data contained in the National Center for Biotechnology Information GenBank [17].

2.3. Selecting Compatible Consortia

Once the strains were identified, compatible consortia were selected from among the six PSF strains (Trichoderma Y19, Absidia Y22, Penicillium Y32, Leptobacillium Y81, Fusarium Y85, and Fusarium Y91). The compatibility between strains was assessed using the mycelial inhibition test [18]. Fifteen dual combination treatments of the different strains were evaluated. The combinations were carried out on PDA culture medium in Petri dishes with a diameter of 90 mm, with active mycelium discs of each strain (5 mm in diameter) placed 40 mm apart (influenced mycelial growth). As a control, each strain was allowed to grow freely in a Petri dish in which a colonized disc and an uncolonized disc were placed (free mycelial growth). Three replicates were used for each treatment. The Petri dishes were incubated at 25 °C for 20 days. The percentage of mycelial inhibition was calculated with the following equation:

$$\%\mathrm{MI}={\displaystyle \frac{\mathrm{FMG}-\mathrm{IMG}}{\mathrm{FMG}}}\times 100$$

where

• % MI = percentage of mycelial inhibition;
• FMG = free mycelial growth (cm);
• IMG = influenced mycelial growth (cm).

For the selection of consortia, combinations with a mycelial inhibition percentage of less than 70% were chosen.

2.4. Phosphate-Solubilizing Capacity of Individual and Consortium Strains

The tricalcium phosphate solubilization capacity of the six PSF strains (individual and consortia) was evaluated in Sundara liquid culture medium at pH 6.8 [19]. Four discs of active mycelium, each 5 mm in diameter, were inoculated into 150 mL flasks with 50 mL of phosphate-revealing medium. Twenty treatments were established, with three replicates for each: six single-strain treatments, eleven double consortia, one triple consortium, and one quintuple consortium, plus the control (no fungi). The flasks were incubated at 25 °C for 10 days. Subsequently, the contents were filtered on Whatman^® 42 filter (Cytiva, Buckinghamshire, UK) paper, the pH was measured with a pH meter, and the soluble P content was determined using the ascorbic acid method [20]. Absorbance was read on a spectrophotometer (JENWAY 6305, Thermo Fisher Scientific, Waltham, MA, USA) at 880 nm, and the results were compared with a standard P curve and were expressed in mg/mL.

2.5. Acid Phosphatase Activity

The acid phosphatase activity of the strains was tested using the method of Tabatabai and Bremer [21]. For this, 900 μL of enzyme extract, 90 μL of 1 M acetate buffer (pH 5), and 10 μL of 15 mM p-nitrophenyl phosphate were mixed and incubated for 1 h at 37 °C. Absorbance was measured at 412 nm on a spectrophotometer. The data obtained were compared with a standard curve and were expressed in µg nitrophenol/min/mL.

2.6. Inoculation of PSF on Plants of Different Coffee Varieties

Four-month-old coffee plants of the Anacafe 14, Costa Rica 95, and Marsellesa varieties were used. The plants were provided by Finca Santa Rosa. They were transplanted into 30 × 20 cm bags, using a substrate of perlite, sand, and soil at a 1:1:4 ratio. The soil used came from the coffee plantation of a single farm, and a chemical analysis was carried out to determine the contents of total P, organic matter, organic C, N, and total C in the soil (

Table 1. Chemical properties of the soil at Finca Santa Rosa.

Chemical Analysis
pH	4.29
Total P (mg/kg)	634.46
Total nitrogen (%)	0.35
Organic matter (%)	6
Organic carbon (%)	3.48
Total carbon (%)	3.69

). These analyses were carried out at the Laboratory of Chemical Analysis of Soils, Water and Plants (LAQSAP) of the Instituto de Ecologia, A.C.

For the inoculation of PSF on coffee plants, inocula with a concentration of 1 × 10⁸ CFU/mL were prepared, as recommended by Souchie et al. [22]. The active mycelium grown on the PDA was used to prepare the inoculum for each strain. The spores were suspended in a water solution with Inex A (1%) and quantified in a Neubauer chamber to adjust the concentration to 1 × 10⁸ CFU/mL. The inoculum was applied directly to the substrate and the plant roots. For each coffee variety, eight treatments were applied, each with three replicates: three single-strain treatments, three dual consortia, one triple consortium, and a control (no fungus) (

Table 2. Experimental design of PSF treatments inoculated on coffee plants (Anacafe, Costa Rica and Marsellesa). From this point on, the colors represent the treatments: purple for the control, pink for the individual inoculations, blue for the dual consortia, and green for the triple consortium.

Treatments	PSF
C	No fungus
Y32	Penicillium 32
Y81	Leptobacillium 81
Y91	Fusarium 91
Y81+Y32	Leptobacillium 81 + Penicillium 32
Y91+Y32	Fusarium 91 + Penicillium 32
Y91+Y81	Fusarium 91 + Leptobacillium 81
Y32+Y81+Y91	Penicillium 32 + Leptobacillium 81 + Fusarium 91

). After 180 days in the greenhouse of the Centro de Investigación en Micología Aplicada, the following variables were evaluated: rhizosphere available P, root length, leaf dry biomass, leaf area, leaf P, height, and stem diameter.

2.7. Available P in Soil and Total Foliar P

The available P in the soil was evaluated according to the technique of Bray and Kurtz [23]. The P concentration was measured with a spectrophotometer at 880 nm. The data obtained were compared with a standard P curve and are expressed in mg/kg. The total foliar P content was measured using McKean’s technique [24] at the Laboratory of Chemical Analysis of Soils, Water, and Plants (LAQSAP) of INECOL A.C.

2.8. Inoculation Response Index

To evaluate the effects of PSF inoculation on the coffee plants, a response index (RI) was calculated for each measured growth variable (height, root length, stem diameter, area, and leaf dry biomass), following the formula of Plenchette et al. [25]:

RI = (inoculated plant-non-inoculated plant/non-inoculated plant) × 100

2.9. Statistical Analysis

The data were analyzed using one-way ANOVA and Tukey’s mean comparison test. The relationship between in vitro soluble P and the pH of the medium was estimated by linear regression. All analyses were performed at a significance level of p < 0.05, using Statistica 8.0 software.

3. Results

3.1. Identification of PSF Strains

Six strains belonging to the genera Fusarium (2), Leptobacillium (1), Penicillium (1), and Trichoderma (1) (Figure 2) were identified by comparing the sequences obtained from the mycelia with those from the GenBank database.

The morphological and molecular approach of five of the sequences showed a percentage of identity higher than 99% and an E value = 0. Using the ITS markers identified the strains as Fusarium crassum (Y85), F. irregulare (Y91), Leptobacillium leptobactrum (Y81), Pencillium brevicompactum (Y32), and Trichoderma spirale (Y19). Only the sequence of strain Y22 was of poor quality, so molecular identification with that sequence was not possible. This strain corresponds to the genus Absidia according to its morphological characteristics, such as sporangiophores arising from sporangiophores, rhizoids, pyriform sporangia with deliquescent walls, apophysis, and a septum below the sporangium (Figure 3) [26].

Table 3. Species identification of PSF strains using the BLASTn algorithm of the National Center for Biotechnology Information GenBank.

Key	Species	Max Score	Query Cover (%)	Identity Score (%)	E Value	GenBank Accession Numbers
Y22	Absidia sp.	-	-	-	-	-
Y91	F. irregulare	917	100	99.21	0.0	PV750149
Y85	Fusarium crassum	824	83	99.78	0.0	PV750148
Y81	Leptobacillium leptobactrum	974	97	99.44	0.0	PV750150
Y32	Penicillium brevicompactum	933	97	99.04	0.0	PV750151
Y19	Trichoderma spirale	1059	97	99.32	0.0	PV750152

shows the BLASTn data from the National Center for Biotechnology Information GenBank [17].

3.2. Consortia Compatibility

From the 15 dual combinations, only 11 were selected, based on their inhibition percentages of 15.65–51.93%, as shown in

Table 4. Percentage inhibition of dual combinations between PSF strains at 20 days of evaluation.

Combinations	Inhibition Percentage (%)
Y91+Y19	100
Y91+Y22	38.68 1
Y91+Y32	15.65 1
Y91+Y85	51.93 1
Y91+Y81	50.2 1
Y85+Y19	100
Y85+Y22	46.5 1
Y85+Y32	34.28 1
Y85+Y81	45.28 1
Y32+Y19	100
Y32+Y22	50.17 1
Y32+Y81	45.49 1
Y81+Y19	100
Y81+Y22	49.14 1
Y19+Y22	20.11 1

¹ Dual combinations selected for their low inhibition percentage.

. The combination F. irregulare (Y91) × P. brevicompactum (Y32) showed the lowest inhibition percentage (15.65%) (Figure 4a), while the combination F. crassum (Y85) × F. irregulare (Y91) achieved 51.93% inhibition. For most of the dual combinations with T. spirale (Y19), the percentage of inhibition was 100%, except for the combination of T. spirale (Y19) × Absidia (Y22) (Figure 4b), for which the inhibition was 20.11%.

3.3. Available P In Vitro

Significant differences in soluble P values were observed among the strains grown in liquid medium (p = 0.001). In all of the treatments, the soluble P content was higher (25.4–172.66 mg/L) than in the control (6.64 mg/L) (p < 0.05). However, the soluble P content was significantly higher in most of the consortium treatments compared to the individual treatments (p < 0.05). The highest values were found in the dual consortia Y91+Y32, Y81+Y32, and Y91+Y81, and in the triple consortium Y32+Y81+Y91 (p < 0.05).

In relation to pH, significant differences were found between the treatments evaluated (p = 0.001). Ten days after inoculation, the pH for the control was 6.8 mg/L. For the PSF inoculation treatments, the pH varied from 5.52 to 2.28 mg/L. No relationship was detected between the soluble P content and the pH of the medium (p > 0.05) (Figure 5).

3.4. Quantification of Acid Phosphatase Activity

The analysis of variance revealed significant differences in acid phosphatase activity between the liquid cultures of the treatments evaluated (p = 0.001). The Y32+Y81 consortium obtained the highest acid phosphatase content (0.149 µg nitrophenol/min/mL) compared to the other treatments evaluated (p < 0.01) (Figure 6).

3.5. P Availability in the Rhizosphere

In the rhizosphere of the plants, the available P content differed according to variety. For Anacafe and Costa Rica, the P availability was higher for all treatments than for the control (p < 0.05) (Figure 7a,b). In the Marsellesa variety, the treatments Y91+Y32 and Y32+Y81+Y91 had the highest available P content (p < 0.05) (Figure 7c).

3.6. PSF Inoculation Response Index (RI)

In Anacafe (Figure 8), the dual treatment Y81+Y32 produced high RI values in leaf dry biomass (298.38%), leaf area (92.07%), and stem diameter (36%), while the triple consortium (Y32+Y81+Y91) promoted a high RI in total leaf P (9.56%) and plant height (87.2%).

In the variety Costa Rica (Figure 9), the highest RI values for root length (21.13%) and stem diameter (27.96%) were obtained with the single treatment Y91; the leaf dry biomass (205.20%), leaf area (249.51%), and height (23.69%) were optimized by the dual treatment Y91+Y32; and the triple consortium (Y32+Y81+Y91) produced the highest RI values for total leaf P (10.84%).

For Marsellesa (Figure 10), the single treatment Y81 produced the highest RI for leaf dry biomass (269.35%) and plant height (36.62%), while the dual treatment Y91+Y81 promoted higher RI values for root length (96.26%) and total leaf P (2.68%), and Y81+Y32 increased the RI value for leaf area (205.48%). Finally, the triple consortium (Y32+Y81+Y91) produced the highest RI for stem diameter (460%) (

Table 5. Effects of PSF inoculation on plant growth variables and RI in three coffee varieties 180 days after inoculation. The colors represent the treatments: pink for the individual inoculations, blue for the dual consortia, and green for the triple consortium.

Anacafe Variety (RI %)
Treatments	Root length	Leaf dry biomass	Leaf area	Leaf P	Height	Stem diameter
Y32	−19.93	220.65	68.57	4.28	65.52	23
Y81	−17.74	177.73	76.5	3.05	67.24	7
Y91	−18.34	249.8	79.89	−0.22	65.57	31
Y81+Y32	−17.23	298.38	92.07	3.6	65.53	36
Y91+Y81	−13.92	161.54	63.96	−14.31	65.51	19
Y91+Y32	−5.71	121.46	57.74	8.41	65.72	18
Y32+Y81+Y91	−5.11	177.73	89.66	9.56	82.7	23
Costa Rica Variety (RI %)
Treatments	Root length	Leaf dry biomass	Leaf area	Leaf P	Height	Stem diameter
Y32	−5.62	37.6	74.65	3.79	5.92	24.57
Y81	7.06	128.4	139.81	−10.43	3.95	27.96
Y91	21.13	119.2	154.17	5.77	9.2	22.03
Y81+Y32	11.28	125.6	174.6	6.56	7.9	12.71
Y91+Y81	11.28	40	95.55	4.87	13.58	24.57
Y91+Y32	−2.79	205.2	249.51	−0.82	23.69	26.27
Y32+Y81+Y91	4.23	94.8	134.27	10.84	6.59	4.23
Marsellesa Variety (RI %)
Treatments	Root length	Leaf dry biomass	Leaf area	Leaf P	Height	Stem diameter
Y32	60.42	237.69	101.88	−0.95	33.66	386.67
Y81	57.59	269.35	107.55	−3.48	36.62	436.67
Y91	50.96	128.14	112.13	2.38	28.17	370
Y81+Y32	67.95	227.14	205.48	−0.56	34.51	396.67
Y91+Y81	96.26	192.46	165.35	2.69	35.21	393.33
Y91+Y32	64.21	139.2	89.77	−10.3	25.35	406.67
Y32+Y81+Y91	62.29	148.24	139.93	−5.51	22.54	460

).

4. Discussion

Morphological and molecular identification determined the strains to be T. spirale (Y19), P. brevicompactum (Y32), L. leptobractum (Y81), F. crassum (Y85), and F. irregulare (Y91); the only strain identified on the basis of morphological characteristics was Y22, for the reasons previously discussed. All of the genera used in this study have been reported as phosphorus-solubilizing fungi.

Several studies have pointed out that the genera Trichoderma, Penicillium, and Leptobacillium promote plant development, increase tolerance to abiotic stresses, improve crop yields, and optimize nutrient availability [27,28]. Some species of Fusarium have been reported to be P solubilizers. Although this genus includes plant-pathogenic species associated with coffee wilt, the F. irregulare and F. crassum strains evaluated in this study have not been reported as pathogenic [29].

Most studies have evaluated the effects of bioinoculants formulated with individual strains, which might have limited efficacy, and have not considered the complex ecological and synergistic interactions that occur in diverse microbial communities [30].

There is currently growing interest in developing biofertilizers based on microbial consortia, which use functional diversity and cooperation between strains to improve the efficiency of P solubilization [31]. This involves selection methodologies based on compatibility between strains, such as the method of Rojas and Hormaza [18], which considers combinations with inhibition percentages of less than 70% as viable. In this study, the T. spirale strain, which inhibited most of the other strains, was discarded. The Trichoderma genus was documented to exhibit highly competitive behavior by interfering with the development of other microorganisms through competition for nutrients and space, along with a rapid growth rate [32].

The evaluation of P solubilization in vitro indicated that not all consortia increased the strains’ phosphate solubilization activity. The combinations with Absidia sp., although compatible, did not increase the soluble P content. This could be related to variability in the solubilizing efficiency of the strains, competition for nutrients in the culture medium, or even the consumption of available P by the microorganisms [33]. Although strain compatibility is a relevant criterion, this alone does not guarantee efficient synergy in P release; therefore, it is important to consider the compatibility and functionality of microbial consortia in studies for the purpose of formulating bioinoculants.

The dual and triple treatments involving F. crassum, F. irregulare, L. leptographium, and P. brevicompactum showed an increase in available P content compared to the individual treatments. The range of P solubilization was 165.05–172.06 mg/L, values notably higher than those reported by Moreno et al. [33], who evaluated combinations of HSP and bacteria, achieving soluble P concentrations of 39.5–76.5 mg/L.

The values in this study also exceed the values reported in other studies evaluating the in vitro P solubilization capacity of individual strains, such as A. niger with 93.5 mg/L [33], Fusarium with 30 mg/L, T. viride with 37 mg/L [34], and Paecilomyces carneus with 83.2 mg/L [35]. These data support the superior potential of microbial consortia over individual strains to improve P solubilization in culture media.

During P solubilization, organic acids are released that reduce the pH of the medium [5]. In our study, a decrease in pH was observed in the in vitro treatments, but no significant increase in available P was detected. This response may be related to the types of organic acids produced by the PSF. Jian et al. [36] reported that these microorganisms release various types of acids, including monobasic acids such as acetic acid; dibasic acids such as oxalic, succinic, malic, tartaric, and fumaric acids; and tribasic acids such as citric and isocitric acids. Nonetheless, it has been demonstrated that di- and tricarboxylic acids are more effective in the solubilization of phosphorus.

On the other hand, some enzymes, such as acid phosphatases and phytases, also play an important role in P availability [37]. In this investigation, acid phosphatase activity was detected in all treatments with PSF, which may indicate that the mineralization process takes place as part of the fungal life cycle. This is because hyphal senescence occurs, leading to the release of organic phosphorus that can be mineralized, resulting in the production of these enzymes [38]. In this research, the dual combination P. brevicompactum + L. leptobactrum showed the highest enzyme activity. Previous work has recorded the acid phosphatase production of the genus Penicillium [39], but there are no records yet for the genus Leptobacillium, and this paper is one of the first to document the enzymatic activity of acid phosphatases.

As previously mentioned, in the greenhouse, coffee plantation soil with a low P content (9.28–9.45 mg/kg), according to the classification of Sadeghian [40], was used. After inoculation, there was an increase in available P, ranging from 15.23 to 43.42%, in the coffee varieties studied, whereas in the control, the available P decreased, probably due to plant absorption and P fixation processes in the soil, as previously mentioned by Corrales et al. [41].

The effect of P availability on the substrate of coffee plants was influenced by the varieties evaluated and the combinations of PSF inoculated (single, dual, and triple). The differences in P solubilization between treatments and coffee varieties could have been influenced by root exudates, which determine responses to microbial interactions.

The inoculations enhanced the three coffee varieties’ leaf dry biomass, leaf area, height, and stem diameter. Three of the dual consortia and the triple consortium were more effective than the individual inoculations. It is important to note that the treatment responses were influenced by the coffee variety evaluated. The results obtained show the multiple mechanisms of action of PSF: not only their ability to increase P availability in the rhizosphere, but also their production of bioactive compounds that contribute to plant growth [42].

With respect to root length, only the low-growing varieties, such as Costa Rica and Marsellesa, showed a positive response to inoculation, while no effect was observed for the tall variety Anacafe. It is likely that tall varieties require more nutrients and more space for root development, so the use of pots may have influenced this variety’s root development, which could have resulted in its lack of response to inoculation. There is no current information relating root system development to variety size and nutrient requirements, so this is an area for further research in future coffee studies.

At present, soil fertilization relies heavily on the use of chemically synthesized inorganic fertilizers. However, inappropriate application can cause negative effects on human health and ecosystem balance. Therefore, biofertilizers emerge as a biotechnological alternative capable of improving soil fertility and increasing agricultural productivity within a framework of sustainable agriculture [43].

Currently, there are successful cases of the use of bioinoculants based on PSF and in various locations. For example, in Canada, since 1990, a biofertilizer based on Penicillium bilaiae has been registered for application in wheat cultivation, with positive results [44]. In Colombia, P. janthinellum is commercially available and is used in rice cultivation as a bioinoculant, with significant increases in production [45].

The present study identified three dual consortia and one triple consortium with high potential for application on coffee plants under field conditions. However, the beneficial effects on phosphorus availability and plant growth varied significantly depending on the PSF consortium and the variety evaluated, suggesting a complex and specific interaction between PSF and coffee varieties.

It is important to consider microorganism–plant compatibility as a key criterion in the design of more effective bioinoculants adapted to specific crops. One area to be developed is to determine the optimal doses of application, their long-term persistence, and the viability of the inoculum in the soil. Finally, the effective adoption of these technologies requires the design and implementation of comprehensive technology transfer strategies that include the training of coffee producers.