Experiment location

The experiment was carried out in a tunnel-type greenhouse with a white plastic cover, which provided 30 % shading, and with an anti-aphid mesh on the side walls, belonging to the Faculty of Agricultural Sciences of the Autonomous University of the State of Morelos, Chamilpa, Cuernavaca, Morelos, Mexico (18° 58’ 51’’ NL and 99° 13’ 55’’ WL, at 1,866 m a. s. l.). Environmental conditions in the greenhouse were monitored with a model U12 datalogger (Hobo®, Massachusetts, USA): average temperature of 25 °C, average relative humidity of 55 % and average solar radiation of 514 µmol·m^-2·s^-1.

Treatments evaluated

The treatments evaluated were: 1) without microorganism inoculation (control), 2) Rhizophagus intraradices, 3) Funneliformis mosseae, 4) Cerro del Metate AMF consortium, 5) Azospirillum brasilense, 6) R. intraradices + A. brasilense and 7) without microorganism inoculation and irrigated with Steiner's nutrient solution at 100 % concentration.

Treatments 1 to 6 were irrigated daily by hand with 0.5 L of ^{Steiner's nutrient solution (Steiner, 1984}) at 50 % during the first 21 days after transplanting (dat), and irrigated with 1.5 L of nutrient solution during the period from 22 to 50 dat, at which time the plants were harvested. Treatment 7 was used to compare the effect of nutrient concentration and inoculations with the microorganisms on plant growth.

Experimental unit and design

The experimental unit was a pot containing one basil plant. A completely randomized design with six replications was used.

Microbiological plant material and sowing of basil plants

The 'Sweet Nufar' basil cultivar (SunLine AG Inc, USA) was used. Seeds were sown on March 29, 2019 in 200-cavity polystyrene trays with Sunshine^® Mix 3 germination substrate, previously sterilized in an autoclave (cv 300, AESA) for 15 min at 120 °C (0.98 to 1.05 kg·cm^-2 steam pressure), according to the methodology described by ^{Zulueta-Rodríguez et al. (2016}). Azospirullum brasilense strain Cd was provided by the Biotechnology Center of the Autonomous University of the State of Morelos, and the AMF by the Phytopathology Laboratory of the Center for Research and Assistance in Technology and Design of the State of Jalisco (CIATEJ). The Funneliformis mosseae inoculum was isolated from the Cerro del Metate consortium, the Rizhophagus intraradices strain was isolated from the Micorriza INIFAP product, and the Cerro del Metate consortium was isolated from the Cerro del Metate mountain, located in Michoacán, from the rhizosphere of Agave cupreata (^{Trinidad-Cruz, Quiñones-Aguilar, Hernández-Cuevas, López-Pérez, & Rincón-Enríquez, 2017}).

Inoculation of basil plants with arbuscular mycorrhizal fungi

Transplanting was performed on May 4, 2019 (38 days after sowing [das]). The basil seedlings were placed in 5-L plastic pots containing tezontle with granulometry ranging from 1 to 5 mm in diameter, previously autoclaved for 15 min at 120 °C (0.98 to 1.05 kg·cm^-2 steam pressure). Treatments were established at transplanting; the AMF were applied at a dose of 150 spores per plant directly to the bare root. The fungi were found in river sand as an inoculation medium.

Inoculation of basil plants with A. brasilense

Basil plants were inoculated with A. brasilense strain Cd at a concentration of 1.04 x 109 colony-forming units (CFU) per gram of dry substrate (^{Bashan, 1986}). CFU counting was performed according to the methodology described by ^{Peña-Cortés, Peña-Cortés, and Gonzalo-Moreno (2011}). The pots had 3 kg of tezontle, and the inoculations were carried out every two weeks from May 14, 2019 until harvest at 50 dat.

Plant growth variables

Harvesting was carried out at 50 dat and the following variables were determined: plant height (PH, cm), basal stem diameter (SD, mm), fresh aerial biomass weight (FBW, g), dry aerial biomass weight (DBW, g) and leaf area (LA, cm²).

Decreased basil growth due to a reduction in the nutrient concentration in the nutrient solution

The percentage reduction of the characters plant height, fresh biomass weight and leaf area was determined by decreasing the concentration of the nutrient solution from 100 % (treatment 7) to 50 % (treatments 1 to 6) using the following formula:

where Ȳ ₇ is the mean of treatment seven (100 % nutrient solution concentration) and Y _ij is the observation of the i-th treatment in its j-th replication (50 % nutrient solution concentration).

Subsequently, Tukey's multiple comparison test (P ≤ 0.05) was performed.

Macronutrient concentration

The aerial dry biomass (leaves and stems) of the six replicates was ground with an electric mill (004108-013-000, Osterizer®, USA). Subsequently, the concentration of N (by the microkjeldahl method), P (by colorimetry), K (by flamometry), and Ca and Mg (both by atomic absorption spectrophotometry) were determined (^{Cruz-Álvarez et al., 2020}).

Mycorrhizal colonization and number of spores

As an indicator of the efficiency and functionality of the inocula in basil plants, the percentage of mycorrhizal colonization (PMC) by fungal structure was determined at 50 dat. For this, roots from each replicate were collected at the end of the experiment and fixed in FAA (formaldehyde: acetic acid: ethanol, 2:1:10:7); subsequently, the roots were cleared and stained (^{Phillips & Hayman, 1970}). The percentage of mycorrhizal colonization was evaluated in 30 1-cm root segments, according to the method described by ^{McGonigle, Miller, Evans, Fairchild, and Swan (1990}). PMC was determined using a microscope (model k7, Zeiss, Germany).

The number of spores in 100 g of dry soil was determined from extraction by the wet sieving and decanting method (^{Gerdemann & Nicolson, 1963}). Spore counting was performed visually with a stereo microscope (VE-S5C, Velab™). Wet sieving was observed in a gridded Petri dish (90 mm) and the number of spores was recorded.

Statistical analysis

Data expressed as percentages were transformed with the arcsine square root function. Normality and homogeneity of variance were checked using the Levene and Kolmogorov-Smirnov test. Subsequently, analysis of variance and Tukey's multiple comparison test (P ≤ 0.05) were performed using Statistical Analysis System version 9.1 software (^{SAS Institute Inc., 2004}).

Growth of basil inoculated with AMF and A. brasilense

When considering the treatments with 50 % nutrient solution (treatments 1 to 6) and comparing them with the control (without microorganisms) (Table 1), only A. brasilense significantly improved the development of 'Sweet Nufar' basil plants, by increasing PH, SD, FBW, DBW and LA (P ≤ 0.05), by 15, 18, 28, 27 and 24 %, respectively. The application of Rizhophagus intraradices and R. intraradices + A. brasilense increased FBW and LA (P ≤ 0.05), while Funneliformis mosseae only improved SD and LA (P ≤ 0.05). The Cerro del Mate consortium did not show advantages over the control without microorganisms.

^{Cruz-Romero, Barrios-Díaz, Rodríguez-Mendoza, Espinoza-Victoria, and Tirado-Torres (2016}) reported similar effects on growth when inoculating broccoli plants with A. brasilense; the increases were 11 % in SD, 38 % in LA, and 29 % in DBW. Likewise, it has been documented that A. brasilense increases nitrogen assimilation in species such as maize and wheat (^{Mariani-Zeffa et al., 2019}; ^{Shintate-Galindo et al., 2020}), which favors plant development, even when fertilization levels are limited. The above coincides with the results of the present study, since a 15 % increase in nitrogen concentration was observed in the tissues of the plants inoculated with A. brasilense with respect to the control (Table 2). In this regard, ^{Pereira-Coelho et al. (2021}) note that A. brasilense forms symbiosis with some plants and increases their water and nutrient assimilation capacity, which improves the growth and yield of agricultural crops.

Inoculation of R. intraradices improved FBW by 7 % and LA by 22 %, while F. mosseae increased SD by 12 % and LA by 17 % compared to the control (Table 1). Favorable effects from the use of AMF have also been reported in tomato (^{Bona et al., 2017}) and broad bean (^{Abd-Alla, El-Enany, Nafady, Khalaf, & Morsy, 2014}). This implies that the species R. intraradices and F. mosseae are also an option to improve basil crop development, although their effect is less than that achieved with A. brasilense.

The combination of R. intraradices + A. brasilense increased FBW by 24 % and LA by 24 % compared to the control. This coincides with the findings reported by ^{Mujica-Pérez, Medina-Carmona, and Rodríguez-Guerra (2017}), who point out that the inoculation of growth-promoting bacteria in combination with AMF generates a synergistic effect that is reflected in better plant development.

Basil growth due to reducing the nutrient solution concentration

When comparing the effect of nutrient concentration (100 and 50 %) on crop development, inoculation with A. brasilense had the smallest decrease in basil growth (P ≤ 0.05) due to the effect of nutrient reduction on the variables PH, FBW and LA; however, its response was similar to the decrease in growth with AMF inoculation and the combination of R. intraradices + A. brasilense (Figure 1). The greatest decreases in growth occurred when no beneficial microorganism was inoculated. Specifically, plants without microorganisms had a decrease of 15.6 % in PH, 49.9 % in FBW and 46.6 % in LA. This indicates that inoculation with growth-promoting rhizobacteria has the ability to improve basil plant development and increase PH and FBW, determining variables in the growth of this crop (^{Chiquito-Contreras et al., 2018}).

Figure 1 Decreased growth in 'Sweet Nufar' basil grown with 50 % nutrient solution compared to 100 % nutrient solution treatment, 50 days after transplanting. Each bar represents the average of six replicates ± standard error. R.+ A.: R. intraradices + A. brasilense. Bars with the same letter between each response variable do not differ statistically (Tukey, P ≤ 0.05). 

Macronutrient concentration

The use of beneficial microorganisms improved the concentration of nutrients in the basil plant tissue, mainly N and K (P ≤ 0.05), compared to the control without microorganisms (Table 2). The N concentration increased (P ≤ 0.05) in plants inoculated with R. intraradices (17 %), A. brasilense (16 %) and with the combination of both microorganisms (15 %). Regarding P, only A. brasilense increased (P ≤ 0.05) its concentration in plant tissues, by 11 %, while K concentration had no significant changes due to the effect of microorganism inoculation. The highest Mg concentrations occurred with the inoculation of R. intraradices and F. mosseae, with significant increases (P ≤ 0.05) of 18 and 20 %, respectively, compared to the control. It has been reported that the use of beneficial microorganisms such as rhizobacteria and AMF allow plants to absorb greater amounts of essential nutrients (^{Díaz-Hernández et al., 2018}). In the case of Ca concentration, there were no significant effects from the inoculation of microorganisms.

Mycorrhizal colonization and number of AMF spores

Figure 2A shows a longitudinal root section of the control, where no AMF structure was observed. In the colonization of mycorrhizal fungi R. intraradices (Figure 2B), F. mosseae (Figure 2C) and the Cerro del Metate consortium (Figure 2D) on basil roots, the presence of some characteristic AMF structures, such as intraradical hyphae, vesicles and arbuscules, was observed (Figures 2E, 2F and 2G). The functionality of mycorrhiza on plant nutrition is associated with the degree of maturity of the mycorrhiza and with the conditions in which it develops (^{Lara-Pérez et al., 2014}).

Figure 2 Longitudinal sections of 'Sweet Nufar' basil roots colonized with arbuscular mycorrhizal fungi (AMF): A) control (cv = vascular cylinder), B) Rhizopagus intraradices, C) Funneliformis mosseae, D) Cerro del Metate AMF consortium, E) intraradical hyphae (hi), F) vesicles (v), and G) arbuscules (a) and base of the arbuscule (ba). 

AMF inocula colonized plant roots, as the percentage of mycorrhizal colonization by hyphae was the highest (35.9 - 37.7 %), followed by arbuscules (13.0 - 14.8 %); however, there were no significant differences (P ≤ 0.05) between them (Figure 3). Regarding vesicles, the range was 4 - 8 %, and differences (P ≤ 0.05) were found between R. intraradices and the rest of the AMF. These results differ with those reported by ^{Elhindi, El-Din, and Elgorban (2017}), who found a PMC between 60.6 and 71.1 % in 'Nano Compatt' basil, while Zulueta et al. (2016) reported a PMC between 75 and 87 % in 'Nufar' basil. These differences can be attributed to the growing conditions, the use of different substrates, the inoculated AMF and the exposure time of the plants to the inoculum, since in those studies the duration of the experiments was 70 and 120 days, respectively.

Figure 3 Mycorrhizal colonization on 'Sweet Nufar' basil roots inoculated with arbuscular mycorrhizal fungi (AMF) 50 days after transplanting. Each bar represents the average of six replicates ± standard error. R. + A.: Rizhophagus intraradices + Azospirillum brasilense. Bars with the same letter between each response variable do not differ statistically (Tukey, P ≤ 0.05).  

With the co-inoculation of R. intraradices + A. brasilense there was no positive effect on the PMC of AMF, which contrasts with the findings reported by ^{Meng et al. (2015}) in maize and soybean, since the interaction of AMF and growth-promoting bacteria tends to increase the population of both individuals in the rhizosphere and, consequently, increases in PMC can occur. In addition, it was confirmed that the control and the treatment with A. brasilense presented 0 % colonization, since these treatments did not contain AMF.

Another indicator to determine mycorrhizal efficiency is the number of spores present in the substrate where the plants developed. At 50 dat, no differences (P ≤ 0.05) were found in the number of spores, where values ranged from 240 to 272 spores per 100 g of substrate. These values are considered low compared to those reported by ^{Quiñones-Aguilar, López-Pérez, Hernández-Acosta, Ferrera-Cerrato, and Rincón-Enríquez (2014}), who in papaya cultivation found 1,000 to 4,500 spores after 120 days of inoculation on organic substrates.

It is important to note that AMF sporulation is a survival mechanism (^{Johnson, 1984}), which could be influenced by the type of substrate. ^{Corbera, Paneque, Calaña, and Morales (2008}) point out that the physical and biological properties of the substrate are determining factors in the proper development of AMF, as well as the age of the plant (^{Agüero-Fernández et al., 2016}). In the present study, tezontle was used as growth medium and, since it is an inert substrate, it could have interfered in the sporulation process; however, that same inert characteristic guaranteed that the results obtained are due to the effect of the microorganisms and their interaction with the basil plants.