Plant growth–promoting traits of yeasts isolated from the tank bromeliad Vriesea minarum L.B. Smith and the effectiveness of Carlosrosaea vrieseae for promoting bromeliad growth

Abstract

Yeasts can play important roles in promoting plant growth; however, little information is available in this regard for yeasts in water of bromeliad tanks. Here, we characterize the ability of 79 yeast isolates from tank bromeliad Vriesea minarum, an endangered species, to solubilize phosphate, secrete siderophores, and synthesize indole-3-acetic acid (IAA). The results showed that 67.8% of all assayed yeast isolates mobilized inorganic phosphate; 40.0% secreted siderophores; and 89.9% synthetized IAA and IAA-like compounds. Among the species studied, Carlosrosaea vrieseae UFMG-CM-Y6724 is highlighted for producing IAA (76.1 μg mL⁻¹) and siderophores, and solubilizing phosphate. In addition, evaluation of the effects of filtrate containing IAA-like compounds produced by the C. vrieseae on the development and photosynthetic performance of V. minarum seedlings found it to improve seedling growth equal to that of commercial IAA. These results demonstrate that C. vrieseae can produce compounds with great potential for future use as biofertilizer agents.

Introduction

Bromeliaceae is one of the most morphologically, physiologically, and ecologically distinct clades of Neotropical angiosperms with 3403 described species [1–3] distributed among eight subfamilies [4]. Bromeliads are one of the most important, diverse, and conspicuous elements of Campo Rupestre (CR) vegetation in Brazil [5–8]. CR is a heterogeneous, rocky, mountaintop (>900 m asl), grassland vegetation mostly occurring in the Espinhaço Range, in southeastern Brazil [7, 8]. This species-rich vegetation is controlled by topography and pedo-environmental as well as micro-climatic conditions [9, 10]. Bromeliads nutritionally benefit from their tank phytotelm due to associations with free-living bacteria (e.g., diazotrophs, phototrophs, and decomposers), fungi (e.g., decomposers), invertebrates (e.g., insect detritivores, arthropod predators), and vertebrates [11]. Vriesea minarum L. B. Sm. is a rupicolous bromeliad of CR that is endemic to the Iron Quadrangle (IQ) and currently listed as endangered in the red list of Brazilian plant species [12, 13]. The species is capable of absorbing water and nutrients in its phytotelm through foliar trichomes (leaf-absorbing trichomes, hereafter – LATs), while its roots are reduced to the purely mechanical support function of attaching the plants to the substrate [14]. Gomes et al. [15] found a community of 36 species of yeasts in association with water from V. minarum tanks; the yeast species Kazachstania rupicola [16], Hannaella pagnoccae [17], Occultifur brasiliensis [15], Bullera vrieseae [18] reclassified by Liu et al. [19] as Carlosrosaea vrieseae, Kockovaella libkindii [20], and Pattersoniomyces tillandsiae [21] were described based on isolates obtained in this work. Other yeast species associated with bromeliads of tropical environments have also been described [17, 20, 22–29]; however, little is known about the functional diversity of yeasts associated with bromeliads.

The enzymatic capabilities of these microorganisms may have significant ecological functions for bromeliad life cycles. Among yeast isolates found in the water tank of V. minarum, 80% exhibited at least one enzymatic activity. Some of the enzymes produced by these yeasts may increase carbon and nitrogen availability for the microbial food web in tanks and facilitate nutrient uptake by bromeliads through the leaf surfaces of their tank [15]. The form in which nutrients are made available in tanks is determined by the source and the complex interactions among tank microbes, which transform them into forms that bromeliads can uptake [30]. There has been recent interest in exploiting beneficial microorganisms of bromeliad tanks with much of this research focusing on the use of particular species of bacterial for what is commonly referred to as plant growth promoting (PGP) [30–33]. Meanwhile, the role of other microbial groups, including yeasts, has received less attention.

A diverse range of yeasts exhibit plant growth–promoting traits, including phytohormone production [34–41]; phosphate solubilization [35, 42–45]; and siderophore production [44, 46]. Auxins, in particular indole-3-acetic acid (IAA), comprise a class of plant growth regulators known to stimulate both rapid (e.g., increases in cell elongation) and long-term (e.g., cell division and differentiation) responses in plants [47], including modifications to root morphology. Diverse soil microorganisms, including yeasts, can produce physiologically active quantities of auxins, which have pronounced effects on plant growth and development [44, 48–50]. In addition to phytohormone production, microbes can also enhance nutrient uptake through several mechanisms.

Plant growth–promoting yeasts (PGPYs) can solubilize inorganic phosphates releasing P and producing organic acids [33, 42, 44, 51, 52]. These acids can chelate the metal cation of phosphorus salts, thus increasing the bioavailability of this essential element to plant tissues, thereby preventing the precipitation of P [53]. Another positive effect that PGPYs have on plant nutrition is the production of siderophores [44, 54]. Siderophores are small, high-affinity metal-chelating compounds that, once secreted, bind insoluble iron ions to form siderophore-Fe complexes that are taken up either by the microorganism itself or by the plant [55]. In addition, siderophores provide a nutritional competitive advantage against pathogens by establishing biocontrol activity [50, 55–58].

The synergistic effects of direct PGP factors, such as plant nutritional supplementation and phytohormone production, and indirect factors, such as restricting pathogen colonization and nutrient scavenging by plant pathogens, are conducive to overall plant growth [44]. Meanwhile, in particular, the possible role of yeasts of water tanks as PGPYs that optimize nutrient availability and/or biochemical mechanisms underlying the nutritional process so as to improve conditions for bromeliad establishment and growth has yet to be addressed. Thus, we set out to study yeasts of the tank of the bromeliad V. minarum, which is a vulnerable species locally in Serra da Piedade [59] and an endangered species in the CR ecosystem of Brazil [13]. More specifically, we characterized 79 yeast isolates by testing their ability to solubilize phosphate, secrete siderophores, and synthesize IAA, a key phytohormone. In addition, we evaluated the effects of IAA produced by the yeast C. vrieseae on the development and photosynthetic performance of V. minarum seedlings. Altogether, the data reported here show that a functional approach can be used to identify yeast isolates with potential use for promoting plant growth as biofertilizer agents. The future use of these yeasts is likely to yield innovative environmentally friendly applications in the cultivation of bromeliads.

Materials and methods

Isolates

Tests were performed on a total of 79 yeast isolates representing 30 species obtained from tank water of V. minarum previously identified [15] and stored in the laboratory at − 80 °C. At least one and a maximum six isolates were tested for each species. The yeasts were grown by streaking YM agar plates (yeast extract-malt extract agar; 1.0% glucose, 0.5% peptone, 0.3% yeast extract, 0.3% malt extract, and 2.0% agar supplemented with 0.02% chloramphenicol), followed by incubation at 25 °C for 48 h prior to use. All experiments were performed with five replicates.

In vitro screening for inorganic phosphate solubilization

Inorganic phosphate solubilization was assayed following the method of Oliveira et al. [60]. Yeast isolates were inoculated onto plates with sterilized modified Pikovskaya’s medium containing Ca₃(PO₄)₂ (tri-calcium phosphate), an insoluble inorganic form of P, followed by incubation at 28 °C for 120 h. Formation of a clear zone around the colony indicated phosphate solubilization activity [61].

In vitro screening for siderophore production

Siderophore production was qualitatively determined by modified CAS agar assay [62]. Yeast (5 μL inoculum of a 48 h culture) was trickled on CAS agar medium and incubated for 120 h in the dark at 28 °C. A bottom agar plate of CAS-blue agar (10 mL of dye solution) was prepared [63] on which, once solidified, an overlay of modified Sabouraud agar (6 mL) was applied. Secretion of siderophores by yeast was visualized by a color change of the medium from blue to yellow/orange/pink as the produced siderophores, binding iron more tightly than the ferric complex of Chrome Azurol S, removed iron from the CAS agar medium.

In vitro screening for IAA production

Production of IAA was tested using the colorimetric method described by Gordon and Weber [64] with some modifications. Yeast isolates were prepared from 2-day-old cultures in YM agar; 200 μL of culture in sterile NaCl solution (0.85%) was used for inoculation (approximately 1.2 × 10⁹ CFU mL⁻¹). Flasks with 5 mL of GYMP broth (2% glucose, 0.5% yeast extract, 0.5% malt extract, 0.2% mono sodium phosphate) amended with 5 mM L-Tryptophan (Sigma) (Trp) and yeast suspension were incubated on an orbital shaker (model MA830, Marconi, Brazil) at 150 rpm at 28 °C in the dark for five days. Non-inoculated flasks served as controls. Cultures were centrifuged at 11,000×g for 15 min at 4 °C to pellet yeast and obtain the supernatant; 100 μL of Salkowski reagent (FeCl₃-HClO₄) was added to 100 μL of the supernatant. The absorbance at 530 nm was measured, after 30 min at room temperature and in the dark, using a UV-Vis varioskan spectrophotometer (Varioskan® Flash, Thermo Scientific). Similarly, a color gradient was also developed using standard solutions of IAA (Sigma-Aldrich) (10–120 μg mL⁻¹), and a standard curve was established; IAA and IAA-like compounds were expressed as IAA equivalents [34]. To determine whether the pathway for synthesizing of IAA through these yeasts is independent of Trp, we also analyzed IAA production in yeast cultures without Trp [36, 44].

Carlosrosaea vrieseae UFMG-CM-Y6724 cultivation and yeast filtrate acquisition (IAA and IAA-like)

Isolated yeasts were prepared from a 2-day-old culture in YM agar; 200 μL of culture in sterile NaCl solution (0.85%) was used for inoculation (approximately 1.2 × 10⁹ CFU mL⁻¹). Flasks with 5 mL of GYMP broth amended with 5 mM tryptophan (Sigma) and yeast suspension were incubated on an orbital shaker (model MA830, Marconi, Brazil) at 150 rpm at 28 °C in the dark for 5 days.

The supernatant and cell pellets were partitioned by centrifugation at 11,000×g at 4 °C for 15 min and the supernatant was filtrated through a 0.22-μm sterile membrane filter (Millipore®, Merck KGaA, Darmstadt, Germany). Simultaneously, 100 μL of Salkowski reagent (FeCl₃-HClO₄) was added to 100 μL of the supernatant. The absorbance at 530 nm was measured, after 30 min at room temperature and in the dark, using a UV-Vis varioskan spectrophotometer (Varioskan® Flash, Thermo Scientific). The filtrate with IAA and IAA-like compounds was quantified, expressed as IAA equivalents, and diluted for immediate use in culture medium for the growth of V. minarum seedlings.

Plant material and in vitro establishment

Seedlings of V. minarum L.B. Smith (Bromeliaceae) were used to understand the effects of the filtrate containing IAA produced by C. vrieseae UFMG-CM-Y6724 on plant growth and development, when added in the culture medium or in the tank region, due to the particular ability of this bromeliad of absorbing water and nutrients through LATs (14). Seeds of V. minarum were collected from mature infructescences from a naturally developed specimen at Serra da Piedade (19° 48′–19° 50′ S, 43° 39′–43° 42′ W), where the yeast was isolated, in the Southern Espinhaço Range in Southeast Brazil, under licensing from regulatory agencies.

In a horizontal laminar flow cabinet, the seeds were surface sterilized for 3 min in 70% ethanol and 15 min in 2% sodium hypochlorite plus 0.1% Tween-20, followed by four rinses in sterile distilled water. Seeds were then germinated on water agar (1.0%) supplemented with 20 μg mL⁻¹ benomyl [methyl 1-(butylcarbamoyl)-2-benzimidazolecarbamate, Sigma-Aldrich] and kept in growth chamber (model MA403, Marconi, Brazil) for 30 days with day/night alternating temperatures (25–20 ± 1 °C) and a 12-h photoperiod (fluorescent lamps providing 30 μmol m⁻² s⁻¹ photosynthetic photon flux density PPFD). The germinated seeds were then counted (more than 1000) and the seedlings transferred to ½ MS medium [65] with macro-nutrients at half strength and without sucrose but supplemented with 20 μg mL⁻¹ benomyl and gelified with 1.0% agar at pH 5.8 (before autoclaving). The seedlings were then placed in a growth chamber (model MA403, Marconi, Brazil) with alternating temperatures (25–20 ± 1 °C) and a 12-h photoperiod (fluorescent lamps providing 30 μmol m⁻² s⁻¹ PPFD).

The following treatments were established at 820 days post-germination: control—nine healthy seedlings placed individually in glass jars containing 50 mL ½ MS medium (pH was adjusted to 5.8 before autoclaving at 120 °C for 20 min) gelified with 1.0% agar and without sucrose and supplemented with 20 μg mL⁻¹ benomyl; Treatment #1—nine seedlings placed individually in glass jars containing 50 mL ½ MS medium gelified with 1.0% agar without sucrose and supplemented with 20 μg mL⁻¹ benomyl with the addition of 1.0 μM IAA (Sigma-Aldrich); Treatment #2—nine healthy seedlings inoculated with two drops of 1.0 μM IAA in the region of the tank; Treatment #3—nine seedlings placed on the same medium with the addition of 1.0 μM IAA-like compounds (expressed as IAA equivalents) from yeast filtrate; and Treatment #4—nine seedlings inoculated with two drops of 1.0 μM IAA-like compounds from yeast filtrate in the region of the tank. After inoculation in a horizontal laminar flow cabinet, the seedlings were placed in a growth chamber (model MA403, Marconi, Brazil) with alternating day/night temperatures (25–20 ± 1 °C) and a 12-h photoperiod (fluorescent lamps providing 30 μmol m⁻² s⁻¹ PPFD) for 6 months.

The effects of IAA-producing yeast on the growth parameters of V. minarum seedlings before and 6 months after the treatments were measured by fresh mass (FM), leaf number (LN), and lateral bud number (LB). The relative growth rate (RGR) was estimated using leaf number and the equation [66]: RGR = (ln (LN_f) – ln (LN_i))/t, where LN_f is final leaf number, LN_i is initial leaf number, and t is time after the application of each treatment, which is equal to 6 months. Percentage biomass (%) accumulated by seedlings after the application of the treatments was calculated using the equation: ((FW_f – FW_i)/FW_i)/100, where FW_f is final fresh weight and FW_i is initial fresh weight.

Chlorophyll pigments were extracted from three leaves of each seedling (five replicates) using dimethyl sulfoxide (DMSO), and chlorophyll a (Chla), chlorophyll b (Chlb), total chlorophyll (ChlT), and carotenoids (Ck) were quantified per fresh weight (FW), essentially as described by Wellburn [67]. Photosynthetic performance was assessed by measurements of chlorophyll a fluorescence for each individual seedling using a PAM chlorophyll fluorometer (MINI-PAM, Walz, Germany). Potential quantum yield of photosystem II (PSII) after 30 min of adaptation to dark [68] was calculated as F_v/F_m (F_v = F_m − F₀), where F_m is maximum fluorescence and F₀ is the basal level of fluorescence in dark-adapted leaves. Instantaneous light response curves for leaves were also produced with a modulated fluorescence meter (MINI-PAM, Walz). The parameters obtained from the instantaneous light response curves were determined over a 4-min period, with increasing levels of light in stages of 30 s each. A saturating light pulse was applied at the end of each light level stage to determine fluorescence parameters, which were used to estimate the maximum electron transport rate at light saturation (ETR_max); PPFD at 90% of ETR_max (PPFD_90%) according to Rascher et al. [69].

Statistical analyses

Assumptions for parametric analyses were tested using the Kolmogorov-Smirnov and Bartlett tests [70]. Significant differences in RGR, biomass percentage (%), pigment contents, F_v/F_m, ETR_max, and PFFD_90% were determined through one-way ANOVA. Significant differences (P < 0.05) between the treatments were evaluated using the multiple scale HSD Tukey test. All analyses were performed using Statistica 6.0 software.

Results and discussion

Yeast isolated from the tank of V. minarum with PGP activities

A total of 79 yeast isolates from the water tanks of V. minarum—representing 30 species, 17 basidiomycetous, and 13 ascomycetous—were tested for some PGP abilities. Of all the assayed yeast isolates, 67.8% (46 out of 68) mobilized inorganic phosphate from the substrates; 40.0% (30 out of 75) secreted siderophores; and 89.9% (71 out of 79) synthetized IAA and IAA-like compounds when grown in the presence of L-Tryptophan (Trp) (Table 1). Direct and indirect PGP abilities were displayed by isolates belonging to different groups (Table 1): a relative high percentage of the basidiomycetous yeast isolates were able to solubilize phosphate (97.4%), while the greatest number (17 out of 35, 48.6%) of siderophore producers were ascomycetous. Isolates positive for IAA and IAA-like molecules were similarly distributed among basidiomycetous (37/43, 86.0%) and ascomycetous (34/36, 94.4%) yeasts. All yeast isolates exhibited at least one of the PGP traits tested (Table 2), and numerous had more than one: for basidiomycetous yeast, 47.6% exhibited two and 28.6% exhibited three, while for ascomycetous yeast, 50.0% exhibited two and 27.8% exhibited three.

Table 1.

Plant growth-promoting (PGP) traits of yeast isolates for the Basidiomycetous and Ascomycetous groups

PGP traits		Groups		Total
		Basidiomycetous	Ascomycetous
Phosphate	Solubilizers	37	21	46
	Non-solubilizers	1	9	22
	Total	38	30	68
	% solubilizers	97.4	70.0	67.8
IAA	Producers	37	34	71
	Non-producers	6	2	8
	Total	43	36	79
	% producers	86.0	94.4	89.9
Siderophore	Producers	13	17	30
	Non-producers	27	18	45
	Total	40	35	75
	% producers	32.5	48.6	40.0

Table 2.

Plant growth–promoting (PGP) traits of yeasts isolated from the tank of Vriesea minarum: indole-3-acetic acid (IAA) production in the presence and absence of L-Tryptophan (L-Trp), phosphate solubilization (Pho.), and siderophore production (Sid.). IAA results are minimum and maximum values; solubilization index (SI) is the minimum and maximum diameter of the entire clear zone/diameter of the zone with yeast colonies; Sid. Results are numbers of isolates tested (in parenthesis) positive and negative for production. All isolates were deposited in the Collection of Microorganisms and Cells of Federal University of Minas Gerais (Universidade Federal de Minas Gerais - Coleção de Microrganismos e Células, UFMG-CM), Belo Horizonte, Minas Gerais, Brazil

Species (Isolates)	Number of isolates tested	IAA (μg mL−1) presence L-Trp	IAA (μg mL−1) absence L-Trp	Pho. (SI)	Sid.
Anomalomyces panici (UFMG-CM-Y6686, -Y6687, -Y6688)	3	17.6–77.7	0–10.5	1.19–1.41	+(1) –(2)
Aureobasidium pullulans (UFMG-CM-Y6689, -Y6690, -Y6691, -Y6692)	4	0–18.5	0	0	+(2) –(2)
Candida intermedia (UFMG-CM-Y6693)	1	34.1	1.26	1.43	–(1)
Candida melibiosica (UFMG-CM-Y6694, -Y6695, -Y6696)	3	21.7–41.0	0–1.7	1.25–1.33	+(3)
Candida membranifaciens (UFMG-CM-Y6697, -Y6698)	2	0–38.0	0–1.6	0–1.20	+(2)
Candida ubatubensis (UFMG-CM-Y6702)	1	60.8	9.1	1.13	+(1)
Carlosrosaea vrieseae (UFMG-CM-Y6722, -Y6723, -Y6724, -Y379, -Y380, -Y398)	6	0–76.1	0–12.2	1.24–1.47	+(6)
Colacogloea sp. (UFMG-CM-Y6382)	1	8.9	0	1.9	–(1)
Dioszegia sp. (UFMG-CM-Y6739)	1	9.6	1.9	1.11	–(1)
Fellomyces penicillatus (UFMG-CM-Y6740)	1	11.1	0	1.28	+(1)
Hannaella pagnoccae (UFMG-CM-Y175)	1	11.9	0	1.18	–(1)
Kazachstania rupicola (UFMG-CM-Y6703)	1	106.5	5.38	0	+(1)
Kockovaella libkindii (UFMG-CM-Y6053)	1	4.7	0	0	–(1)
Kodamaea ohmeri (UFMG-CM-Y6704, -Y6705, -Y6706)	3	11.8–61.6	0–4.6	0–1.38	+(1) –(2)
Meira sp. (UFMG-CM-Y6741, -Y6742, -Y6743)	3	0–15.2	0	1.33–1.37	+(2) –(1)
Metschnikowia koreensis (UFMG-CM-Y6707, -Y6708, -Y6709)	3	14.3–20.8	0	1.34–1.64	+(3)
Meyerozyma guilliermondii (UFMG-CM-Y6719, -Y6720, -Y6721)	3	12.8–23.4	0 – 1.3	0	+(2) –(1)
Meyerozyma sp. (UFMG-CM-Y6710, -Y6711, -Y6712)	3	11.0–35.6	0–1.9	1.21–1.38	+(1) –(2)
Myriangium sp. (UFMG-CM-Y6713, -Y6714, -Y6715, -Y6716, -Y6717, -Y6718)	6	0.9–73.6	0–5.6	0–1.22	–(6)
Occultifur brasiliensis (UFMG-CM-Y6744, -Y6745, -Y6746)	3	5.5–58.9	0	1.1–1.2	+(1) –(2)
Papiliotrema flavescens (UFMG-CM-Y6725, -Y6726, -Y6727, -Y6728)	5	0–11.2	0–1.8	0	+(4) –(1)
Papiliotrema laurentii (UFMG-CM-Y6729, -Y6730, -Y6731)	3	9.3–14.8	0–4.6	1.18–1.40	+(1) –(2)
Papiliotrema nemorosus (UFMG-CM-Y6732)	1	48.6	4.6	1.61	+(1)
Papiliotrema rajasthanensis (UFMG-CM-Y6736)	1	99.0	1.3	1.19	+(1)
Papiliotrema sp. (UFMG-CM-Y6737, -Y6738)	2	0–13.7	0	0 – 1.16	–(2)
Pattersoniomyces tillandsiae (UFMG-CM-Y6749)	1	17.2	0	1.75	–(1)
Rhodosporidium diabovatum (UFMG-CM-Y6750, -Y6751, -Y6752, -Y6753, -Y6754)	5	0–107.5	0–9.3	1.06–1.24	+(1) –(4)
Rhodotorula sp. (UFMG-CM-Y6755, -Y6756)	2	5.9–20.0	0	1.40–1.58	–(2)
Saturnispora sylvae (UFMG-CM-Y6699, -Y6700, -Y6701)	2	59.1–67.5	0–4.2	1.19–1.23	–(2)
Saitozyma podzolica (UFMG-CM-Y6733, -Y6734, -Y6735)	3	15.9–73.9	0–5.1	1.24–1.46	–(3)
Total isolates	79

The IAA concentration ranged from 0.75 to 12.21 μg mL⁻¹ when cultured in GYMP broth in the absence of exogenous Trp (Table 2). However, in the presence of Trp, IAA production increased to values ranging from 0.86 to 107.54 μg mL⁻¹, with two isolates, Kazachstania rupicola (106.5 μg mL⁻¹) and Rhodosporidium diabovatum (107.5 μg mL⁻¹), producing higher concentrations than the other yeasts tested (Table 2). Our results support the presence of a Trp-independent IAA biosynthetic pathway for some yeasts, even if the values are low, which is consistent with other studies [44]. The yeasts varied greatly in their IAA production efficiencies [34–37, 39, 44]. Some of the studied genera, such as Candida and Rhodotorula [48]; Hannaella, Papiliotrema (synonym Cryptococcus), and Rhodosporidium [38, 71]; and Aureobasidium, Meyerozyma, Pseudozyma, and Kazachstania [44], have been previously reported as IAA-producing at different levels, inter- and intraspecifically. Nevertheless, there have been no studies on the diversity of yeast with IAA-producing abilities in bromeliads of the CR ecosystem or even in Brazil for that matter.

Metschnikowia koreensis, Meyerozyma guilliermondii, Rhodotorula sp., and Saitozyma podzolica had the highest solubilization index (SI) for solubilizing inorganic P. Yeasts of the genus Rhodotorula have been previously cited as good phosphate solubilizers [43, 51]. We believed that phosphate-solubilizing yeast isolates could play an important role in providing supplemental P for V. minarum. The solubilization of phosphate-bearing inorganic materials by microorganisms would seem to be an attractive solution for fertilizing the soil, and several yeasts have already been identified for their ability to mobilize insoluble inorganic P, including other sources Ca, Fe, and rock phosphates [51, 53, 72–74].

Kazachstania rupicola and some isolates of the genera Anomalomyces, Aureobasidium, Candida, Carlosrosaea, Kodamaea, Meira, Metschnikowia, Meyerozyma, Occultifur, and Papiliotrema were positive for siderophore production (Table 2). Microbial siderophores are known to provide plants with metals, especially Fe, which favors growth when Fe bioavailability is low [75]. According to Ahmed and Holmström [55], microbial siderophores with high redox potential can be reduced to donate Fe (II) to the transport system of the plant or can chelate Fe from the soil and then undergo a ligand exchange with phytosiderophores. However, this mechanism depends on several parameters, including the stability constants and concentrations of both microbial and phytosiderophores, and the pH and redox conditions of the root environment [75]. In addition, the production of siderophores makes Fe unavailable for pathogen growth and thus is critical for promoting plant health [44]. The depletion of Fe results in pathogen failure, since Fe is essential for the growth of almost all living microorganisms because it acts as a catalyst in enzymatic processes, oxygen metabolism, electron transfer, and DNA and RNA synthesis [55, 76].

Tank bromeliads facilitate the availability and redistribution of nutrients through the aquatic microecosystems they form, in particular through litter decomposition in the tank [77]. Gomes et al. [15] demonstrated the production of enzymes by the same yeast species of the present study, including protease, xylanase, amylase, pectinase, and cellulase. These enzymes can increase carbon and nitrogen availability, and, because of the presence of phosphate solubilizer and siderophore producer yeasts, other nutrients, such as P, Ca, Mg, and Fe, can be made available for the microbial food web living the tank of V. minarum, as well as for bromeliad nutrition by LATs. Thus, due to the maintenance of diverse aquatic food webs, tank bromeliads can easily establish themselves in nutrient-poor habitats [11, 78], which may be very important for V. minarum because of its association with the nutrient-poor habitats of CR.

The role that yeasts play in plant growth is still not very clear, but studies suggest that they have potential as biological PGP agents [50, 58]. Among the species studied, we highlight C. vrieseae UFMG-CM-Y6724 because it produced more than 10 μg mL⁻¹ IAA in the absence of exogenous Trp and 76.1 μg mL⁻¹ IAA in its presence. In addition, this isolate solubilizes phosphate and produces siderophores. Plant growth depends on the production of different types of hormones and enzymes that directly influence plant growth and productivity. Besides IAA, cytokinin, and other hormones, plants require certain micro- and macro-nutrients, such as phosphate, potassium, zinc, manganese, and iron [50]. The potential role of C. vrieseae as biofertilizer for bromeliads will be discussed in the next session.

Effects of IAA-like producing C. vrieseae UFMG-CM-Y6724 yeast on bromeliad growth and development

Six months of cultivation of V. minarum seedlings in medium supplemented with IAA or IAA-like compounds produced by yeast resulted in an increase in relative growth rate (RGR) compared to the control (ANOVA, F = 2.689, P = 0.0487) (Table 3). The leaf number was also greater independent of the auxin type used to increase RGR. The supplementation of cultivation medium with IAA-like compounds from filtrates of C. vrieseae UFMG-CM-Y6724 (Treatment #3) accelerated lateral bud outgrowth (Table 3, Fig. 1). Unlike this treatment, the same exogenous concentration of IAA for other treatment types did not produce this response. No significant differences were found for the accumulation of biomass by V. minarum seedlings (Table 3).

Table 3.

Effect of IAA on relative growth rate (RGR, leaf number. months⁻¹), lateral bud number (LB), and percentage biomass (fresh weight, FW) accumulated in seedlings of Vriesea minarum: ½ MS medium without indole-3-acetic acid (IAA) (control); Treatment #1, 1.0 μM IAA; Treatment # 2, two drops of 1.0 μM IAA; Treatment #3, 1.0 μM auxin-like compounds (expressed as IAA equivalents) from filtrate of the yeast Carlosrosaea vrieseae UFMG-CM-Y6724; and Treatment #4, two drops of 1.0 μM IAA-like compounds from filtrate of the yeast Carlosrosaea vrieseae UFMG-CM-Y6724

Treatments	RGR	LB	FW (%)
Control	0.061 ± 0.01a	0.0	200.0 ± 31.8a
1	0.089 ± 0.03b	0.0	198.9 ± 52.3a
2	0.071 ± 0.02b	0.0	188.3 ± 52.7a
3	0.105 ± 0.034b	1.8 ± 0.5	188.8 ± 46.5a
4	0.091 ± 0.03b	0.0	141.6 ± 29.2a

Values indicate the mean ± standard deviation of independent experiments carried out with five replicates. Distinct letters indicate significant differences between treatments (Tukey; P < 0.05)

Fig. 1.

Morphological aspects of shoots of V. minarum seedlings in vitro. (A) ½ MS medium without indole-3-acetic acid (IAA) (control); (B) Treatment #1, 1.0 μM IAA; (C) Treatment # 2, two drops of 1.0 μM IAA; (D) Treatment #3, 1.0 μM auxin-like compounds (expressed as IAA equivalents) from filtrate of the yeast Carlosrosaea vrieseae UFMG-CM-Y6724; and (E) Treatment #4, two drops of 1.0 μM IAA-like compounds. Arrow = lateral bud

Many yeast genera and/or species found in the rhizosphere of other plant species are present in the tanks of V. minarum. Rhizosphere yeasts that produce IAA have been able to improve the growth of several crops, such as Sporobolomyces roseus promoting wheat yield by 16–30% [79]; a Rhodotorula sp. isolate increased tomato growth and fruit yield [80]; soil inoculation with Candida valida (= Pichia membranifaciens), Rhodotorula glutinis, and Trichosporon asahii, singly or in combination, has been reported to promote sugar beet growth [81]; isolates of Williopsis saturnus (= Cyberlindnera saturnus), an endophyte of maize and producer of IAA, promoted corn growth (shoot and root) in greenhouse trials [34]; Candida tropicalis (CtHY), also an IAA producer, promoted a 35% increase in root dry mass of rice [35]; and Torulaspora globosa, a rhizosphere yeast for the development of lettuce (cv. Crocantela) [40]. Although the roots of V. minarum purely function in mechanical support, and the LATs of the tank are responsible nutrient uptake in the adult phase, the roots can have an absorptive function in the early stages of development of the seedling, when the yeasts can play an important role as well.

No studies can be found involving the addition of yeast compounds to culture medium to evaluate their specificity related to growth promotion of bromeliad species. The effects of the compounds produced by the yeast C. vrieseae UFMG-CM-Y6724 on seedling growth of V. minarum evaluated in this study as a PGP were positive because the RGR of seedlings was remarkably higher than the control (Table 3). However, other mechanisms have yet to be evaluated, such as the production of other plant hormones (e.g., cytokinins – CK) and other secondary metabolites (e.g., ethylene gas) that may stimulate plant development. IAA, gibberellins (GA), and CK are considered secondary metabolites of fungi, being excreted by microorganisms near the end of the growth phase or during the stationary phase [82–84].

Bromeliad seedlings in medium with IAA-like compounds produced by yeast showed notable outgrowth of lateral buds. Van Dijck et al. [85] showed that in vitro propagation of two species of Aechmea could be achieved through lateral bud shoot growth in MS medium containing both cytokinin (6-benzylaminopurine, BAP) and IAA. Furthermore, BAP and IAA were found to synergistically stimulate the production of ethylene by the cultured plants. The stimulation of ethylene production was correlated with lateral bud outgrowth, but CK was concluded to be essential for lateral bud outgrowth of Aechmea in vitro. The increase of outgrowth of lateral buds is more related to the presence of cytokinins than IAA. The CK used most often to break the apical dominance of in vitro bromeliads is BAP [86, 87]. It is also known that cytokinins and auxins have a synergistic effect on the de novo synthesis of ACC synthase. ACC synthase controls the conversion of s-denosyl-methionine (SAM) to ACC, the precursor of ethylene [88, 89]. CK are organic molecules that promote plant growth through facilitated cellular division and growth and ethylene is synthesized from methionine in many tissues in response to stress [90]. Although the present study did not quantify the production of CK by the yeast C. vrieseae, other studies indicate that yeasts can produce several types of this phytohormone [84, 91]. Zeatin production, the most common CK and like other adenine-derived cytokinins, was detected among basidiomycetous and ascomycetous yeast species [84]. The environmental conditions of conventional in vitro culture consist of sealed containers with high relative humidity, reduced gas exchange, and artificial temperature and luminosity conditions [92]. High variation in CO₂ and ethylene accumulation may occur inside containers during the day [93, 94]. Another speculation is that supra optimal IAA concentrations stimulate ethylene production by enhancing ACC synthase activity [95].

The micro-environment inside sealed containers can cause physiological and anatomical disorders in plants [93, 96], including impaired stomatal functionality and changes in the performance of the photosynthetic apparatus [97–100]. The maximum quantum efficiency of PSII (F_v/F_m) was similar for all treatments in relation to the control (P > 0.05) with an average of 0.7 (Fig. 2), except for Treatment #2, clearly indicating an alteration in efficiency of the photosystems in the use of light (< 0.7). The analysis of light dependence curves indicated significant differences in ETR_max and PPFD_90% for Treatment #2 (Fig. 3), which had the lowest average absolute value of total chlorophyll (Table 4). Martins et al. [101] demonstrated that the suitable application of auxins (e.g., 1-naphthaleneacetic acid, NAA) can improve the performance of the photosynthetic apparatus of micropropagated plants. However, the in vitro condition can result in low photosynthetic activity of plants and is considered to be one of the major limiting factors of micropropagation efficiency [102]. This may also induce other physiological alterations and anatomical disorders in plants cultivated in vitro, such as decreased photosynthetic pigment content and poor leaf tissue function [92, 94, 103]. In the present study, the application of droplets of IAA in the bromeliad tank and not in the culture medium seemed to be prejudicial to photosynthesis with effects on growth. This result may be important for in vitro cultivation of this species and deserves to be well studied, since the use of a drop with yeast filtrate did not have this effect. In nature, adult plants absorb nutrients that are available in their tank through LATs and, probably, exogenous hormones produced by microorganisms.

Fig. 2.

Maximum quantum yield of PSII (Fv/Fm) of V. minarum seedlings immediately after their removal from culture medium after incubation for 6 months under alternating temperatures (25–20 ± 1 °C) and a 12-h photoperiod at 30 μmol m⁻² s⁻¹ PPF. (A) ½ MS medium without indole-3-acetic acid (IAA) (control); (B) Treatment #1, 1.0 μM IAA; (C) Treatment # 2, two drops of 1.0 μM IAA; (D) Treatment #3, 1.0 μM IAA-like compounds (expressed as IAA equivalents) from filtrate of the yeast Carlosrosaea vrieseae UFMG-CM-Y6724; and (E) Treatment #4, two drops of 1.0 μM IAA-like compounds. Each data bar represents a mean ± SE of nine seedlings. Different letters indicate significant differences between treatments (Tukey; P < 0.05)

Fig. 3.

Maximum apparent electron transport rates (ETR_max - black squares) and photosynthetic photon flux density at 90% of ETR_max (bars = PPFD_90%) of V. minarum seedlings immediately after their removal from culture medium after incubation for 6 months under alternating temperatures (25–20 ± 1 °C) and a 12-h photoperiod at 30 μmol m⁻² s⁻¹ PPF. (A) ½ MS medium without indole-3-acetic acid (IAA) (control); (B) Treatment #1, 1.0 μM IAA; (C) Treatment # 2, two drops of 1.0 μM IAA; (D) Treatment #3, 1.0 μM IAA-like compounds (expressed as IAA equivalents) from yeast filtrated of Carlosrosaea vrieseae UFMG-CM-Y6724; and (E) Treatment #4, two drops of 1.0 μM IAA-like compounds. Each data bar and square represent a mean ± SE of nine seedlings. Different letters indicate significant differences between treatments (Tukey; P < 0.05)

Table 4.

Effects of IAA on chlorophyll a, b, total chlorophyll (Chl a, Chl b, ChlT, respectively), and carotenoids (Ck) in seedlings of Vriesea minarum: ½ MS medium without indole-3-acetic acid (IAA) (control); Treatment #1, 1.0 μM IAA; Treatment #2, two drops of 1.0 μM IAA; Treatment #3, 1.0 μM auxin-like compounds (expressed as IAA equivalents) from filtrate of the yeast Carlosrosaea vrieseae UFMG-CM-Y6724; and Treatment #4, two drops of 1.0 μM IAA-like compounds from filtrate of the yeast Carlosrosaea vrieseae UFMG-CM-Y6724

Treatments	Chl b (μg g−1 FW)	Chl a (μg g−1 FW)	ChlT (μg g−1 FW)	Chl a/b	Ck (μg g−1 FW)	Ck/ChlT
Control	77.2 ± 19.1bc	141.5 ± 34.9bc	218.7 ± 54.0bc	1.83	12.0 ± 3.8bc	0.0542
1	154.7 ± 20.6a	283.5 ± 37.8a	438.2 ± 58.5a	1.83	23.8 ± 3.7ab	0.0543
2	94.1 ± 38.8b	172.4 ± 71.1b	266.5 ± 49.7b	1.83	19.1 ± 3.9b	0.0567
3	111.0 ± 24.7b	203.2 ± 45.1b	314.2 ± 69.8b	1.83	17.7 ± 3.6b	0.0566
4	121.6 ± 35.3b	222.6 ± 64.6b	344.2 ± 99.9b	1.83	17.2 ± 3.3b	0.0511

Values indicate the mean ± SD of five replicates. Distinct letters indicate significant differences between treatments (Tukey; P < 0.05)

The highest chlorophyll a content was observed in seedlings that were kept for 6 months in medium supplemented with 1.0 μM IAA—Treatment #1 (Table 4). Total chlorophyll content was 438.2 ± 58.5 μg g⁻¹ FW and carotenoids content was 23.8 ± 3.7 μg g⁻¹ FW for seedlings from this treatment, which was significantly higher than the auxin-free treatment. According to Mshelmbula et al. [104], the effect of IAA on chlorophyll content in leaves of rice under watering was increased synthesis. Some key components in the light signaling pathway, such as PIFs and HY5, connect light signals to the signaling pathways of multiple phytohormones, including ethylene, GA, and CK [105]. The present study found no significant differences in Chl a/b and Ck/ChlT ratios among treatments (Table 3). Thus, the present study showed that the application of IAA directly on the leaves of V. minarum seedlings, in contrast to being applied in the culture medium, affected photosystem II and the total chlorophyll content both significantly lower.

The present study emphasizes the high level and diversity of PGPY abilities in the tank of the bromeliad V. minarum. Kazachstania rupicola and R. diabovatum were the best IAA producers among the yeast species studied, while C. vrieseae UFMG-CM-Y6724 is highlighted for producing IAA in the absence and presence of exogenous Trp, solubilizing phosphate and producing siderophores. The addition of IAA and IAA-like compounds produced by C. vrieseae UFMG-CM-Y6724 equally improved the growth of V. minarum seedlings. The results of the present study demonstrate that the use of yeast filtrate containing auxin can improve the micropropagation of the studied bromeliad. The use of yeast in agriculture is currently an active area of research. The ecological benefits of yeast need to be explored and may provide a breakthrough for the bromeliad micropropagation and conservation sector. We believe that yeasts are vital for the establishment and development of bromeliads with ecological strategies like those of V. minarum. Understanding the diversity of microorganisms inhabiting bromeliad tanks has both ecological and economic importance since this information could be useful in the management and conservation of the bromeliads themselves.

Author contribution

All authors contributed to the study conception and design. Material preparation, data collection, and analyses were performed by Andrea R. Marques, Alessandra Abrão Resende, Alexandre Aparecido Duarte, and Ana Raquel O. Santos. The first draft of the manuscript was written by Andrea R. Marques, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

This research was supported by Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG) grant no CBB - APQ-02639-15.

Data Availability

The ecological datasets in the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval

Permissions for collection according to ICMBio instructions. Document 57088-1 was registered by AA Resende in SISBIO and document 097/2015 was registered by AA Resende in IEF/MG as authorization for collection of seeds samples for scientific activities.

Consent to participate

All authors agreed with the content and gave explicit consent to submit and obtained consent from the responsible authorities at the institute where the work has been carried out.

Consent for publication

All authors read and approved the final manuscript.

Conflict of interest

The authors declare no competing interests.