Soil from Serianthes Rhizosphere Influences Growth and Leaf Nutrient Content of Serianthes Plants

Abstract

Soils from the rhizosphere of perennial plants accumulate microorganisms that influence the growth of other plants. This microorganism biodiversity may be exploited by using these soils as an inoculum in new planting sites. Soils collected from the rhizosphere of mature Serianthes trees were subjected to treatments designed to reduce or increase microorganism populations, then were used in a series of five studies to grow Serianthes plants in container culture. Serianthes kanehirae and Serianthes grandiflora stem growth was 14–19% greater, leaf nitrogen was 40–46% greater, leaf phosphorus was 50–86% greater, and leaf potassium was 28–43% greater when grown in soils from Serianthes rhizosphere than in soils away from a Serianthes tree. Treating the Serianthes rhizosphere soils with sterilization or propiconazole fungicide reduced stem growth of S. grandiflora, S. kanehirae, and Serianthes nelsonii plants by 16–47% below that of untreated soils. The sterilization and fungicide treatments also consistently reduced phosphorus (48–50%) and potassium (12–21%) content of leaves when compared with untreated rhizosphere soil. Adding Rhizophagus irregularis inoculum to the sterilized soil reversed the reduction in S. grandiflora stem growth and leaf phosphorus content. These findings indicate that soils from the Serianthes rhizosphere contain beneficial microorganisms for Serianthes plant growth and leaf nutritional status and exploiting these soils as an inoculum for new planting sites may provide a net-positive influence on post-transplant growth and survival.

1. Introduction

The sessile nature of plants leads to an accumulation of soil factors that develop from plant–soil feedback (PSF) processes. PSFs include the buildup of biotic and abiotic factors in the soils beneath a plant that are negative or positive to subsequent growth of the same or other plants [1,2,3]. Positive PSF is often mediated by mutualisms, and mycorrhizas have been studied more often than other plant mutualists [4]. However, other soil-borne fungal communities may exert positive PSF [5,6]. With Fabaceae species, rhizobacteria populations can also be causal mechanisms of PSF [7]. The positive influences of these edaphic changes are described as niche facilitation [8].

The negative or positive effects of PSF are context-dependent, and numerous experimental approaches have been used to tease apart the various causal mechanisms. For example, one means by which invasive alien plants modify ecosystems is through PSF and changes in mycorrhizal diversity [9]. Experimental techniques which reduce soil microbial diversity or activity are useful for illuminating the effects of those microorganisms on soil processes and quality of subsequent plant growth [10]. Plant mutualisms with mycorrhizal fungi are of interest to plant scientists for their potential to promote access to mineral resources [11]. In addition to sterilization techniques which reduce the diversity of all soil-borne microorganisms, the fungicide propiconazole has been employed specifically to reduce mycorrhizal populations and efficacy [12]. The fungistatic nature of propiconazole blocks the biosynthesis of ergosterols that are required for the formation of fungal cell walls, so the growth of the fungus declines or ceases such that continued invasion of host tissues is minimized [13]. Experimental approaches exploiting propiconazole have reduced plant nutrient content [14].

Experimental techniques which increase soil microbial diversity are also useful for studying their influences on plant growth. For example, the arbuscular mycorrhiza Rhizophagus irregularis (Błaszk., Wubet, Renker, and Buscot) C. Walker and A. Schüßler is commonly used in scientific studies to determine the effects of arbuscular mycorrhiza on plant and soil health [15,16].

The use of these experimental approaches remains limited in the fields of horticulture and plant conservation. The Fabaceae species Serianthes grandiflora (Benth.) F. Müller is from the Philippines, Serianthes kanehirae Fosberg is from Palau and Yap, and Serianthes nelsonii Merr. is from the Mariana Islands [17,18]. These attractive trees are employed in urban horticulture as street trees within their native ranges. Serianthes nelsonii is also endangered [19,20], and the research needed to support recovery efforts is lacking [21]. To my knowledge, there have been no studies designed to determine the influence of Serianthes rhizosphere soil on Serianthes plant growth and nutrition.

The aim of this series of studies was to use field soil treatments that manipulated the soil microbiome to determine the influences on plant growth and leaf nutrient content of three Serianthes species. This work was considered preliminary and focused on the plant responses rather than directly on the microbiome responses. The first hypothesis is that Serianthes plant growth and the macronutrient content of leaves would be greater in soils obtained from Serianthes rhizosphere than in soils obtained far from a Serianthes tree. The second hypothesis is that Serianthes plant growth and leaf nutrition would be reduced by soil sterilization or fungicide treatments of Serianthes rhizosphere soils. The third hypothesis is that phosphorus (P) would be the leaf nutrient most affected by the soil treatments due to the influence of treatments on mycorrhizal activity. The results highlight the importance of understanding the mutualistic relationships that form in the Serianthes rhizosphere and how that knowledge may be used to improve the management of endangered trees.

2. Materials and Methods

Soils were collected from the surface 15 cm beneath Serianthes trees and used as a container medium to grow Serianthes seedlings in five Guam and Philippine studies. The precise localities of the source trees are not included due to conservation ethics. The source trees were located in northern Guam and northern Samar (Figure 1). The field soil was assumed to contain the rhizosphere microbiome of Serianthes roots and was passed through a 3-mm sieve to remove roots, rocks, seeds, and other debris. All containers employed in the studies were immersed in a 10% bleach solution for 10 min prior to use. The container medium used to produce the seedlings prior to initiating the experimental procedures was autoclaved prior to sowing the seeds. The seeds were soaked in a 10% bleach solution for 10 min before imbibing in water for one hour. The field soil was mixed with washed river sand as one part soil to two parts sand to improve drainage and to minimize the volume of rhizosphere soil required for the experiments. This sand was autoclaved prior to mixing. These methods ensured that the only source of living soil microorganisms during the experiments was the field soil.

2.1. Home versus Away Soil Studies

2.1.1. Serianthes nelsonii Soil

The first experiment was conducted in Mangilao, Guam from 2 March 2015 until 20 July 2015 with S. kanehirae plants. The Home soil was collected on 1 March 2015 from the rhizosphere of a mature S. nelsonii tree in northern Guam. The Away soil was collected at the same time from the same habitat in a 50–70 m band surrounding the tree. This fresh soil was mixed one part soil to two parts washed river sand. For each of the two soil mixtures, four samples were collected to determine the nutrient content. Dry combustion was used to quantify carbon (C) and nitrogen (N) (FLASH EA1112 CHN Analyzer, ThermoFisher, Waltham, MA, USA). Available P was determined by the Olsen method [22], and available potassium (K) was quantified by inductively coupled plasma optical emission spectrometry (ICP-OES; Spectro Genesis; SPECTRO Analytical Instruments, Kleve, Germany) following digestion with diethylenetriaminepentaacetic acid.

The plant material consisted of S. kanehirae seedlings growing in tubes that were 5 cm in diameter and 12 cm deep. Pre-germinated seeds (sourced from Yap) were planted in the tubes in a 60%:40% peat:perlite medium and the seedlings were grown under a 50% shade screen on raised nursery benches. The height and basal stem diameter of each seedling were measured before bare-rooting and then re-planting in the experimental soils. The beginning plant size was 19 ± 4 cm in height (mean ± SE) and 2.9 ± 0.2 mm in stem diameter. The plants were arranged in a completely randomized design on a nursery bench under a 50% shade screen. They were positioned in a grid with 50-cm spacing. They were irrigated as needed, usually two to three times per week, and there were six replications. A graphical depiction of the experimental methods is shown in Figure 2.

Final measurements were made on 20 July 2015. Stem height and basal diameter were directly measured, and leaf number was counted. Growth in height and stem diameter was calculated from beginning and ending measurements. The three terminal fully-expanded leaves were removed for tissue analysis and were dried for 48 h at 75 °C in a forced draft oven, then milled to pass through a 20-mesh screen. Total C and N concentration was determined by dry combustion. The samples were digested by a microwave system with nitric acid and peroxide, then K and P were quantified by ICP-OES.

2.1.2. Serianthes grandiflora Soil

A second study was conducted in Angeles City, Philippines using soil harvested from the rhizosphere of a S. grandiflora tree from Barangay San Agustin, Lavezares, Samar. The soil was harvested on 15 November 2018 and the study was initiated on 19 November 2018. Home versus Away soil collection methods was as previously described. The bulk soil was amended with washed river sand (1:2) and sampled for C, N, P, and K analysis (n = 4).

The plant material consisted of S. grandiflora seedlings that were produced as previously described. The source of seeds was the tree from which soils were collected. The beginning plant size was 19 ± 3 cm in height and 2.8 ± 0.3 mm in stem diameter. The study was terminated on 12 March 2019. Growth and leaf elemental concentrations were determined as previously described, and there were six replications.

2.2. Soil Manipulation Studies

2.2.1. Serianthes kanehirae in Serianthes nelsonii Soil

The first soil manipulation experiment was conducted in Mangilao, Guam from 4 March 2015 until 3 August 2015 with S. kanehirae plants. The container substrate was soil collected on 1 March from the rhizosphere of a mature S. nelsonii tree in northern Guam. This fresh soil was mixed one part soil to two parts washed river sand. The soil mix was homogenized and then soil chemical traits were measured as described in Section 2.1.1.

The bulk soil was then separated into three equal parts to create the three treatments. The control treatment consisted of unamended fresh soil with the full spectrum of living rhizosphere microorganisms. The sterilized treatment was obtained by subjecting to moist heating at 121 °C for 30 min on three consecutive days [23,24]. The fungal reduction treatment was obtained with the use of the fungicide propiconazole applied at the rate of 0.025 g·m⁻² (Honor Guard PPZ, Control Solutions, Inc., Pasadena, TX, USA). This application was made after the 2.6-L containers with a surface area of 201 cm² had been filled with the soil mixture.

The plant material consisted of S. kanehirae seedlings sourced from Yap which were produced as previously described. The beginning plant size was 21 ± 4 cm in height and 3.1 ± 0.2 mm in stem diameter. The plants were grown under the same nursery conditions and the final measurements were as previously described. The study was terminated on 3 August 2015. In addition to the K and P that were quantified from the digested leaf tissue, calcium (Ca), magnesium (Mg), manganese (Mn), iron (Fe), zinc (Zn), boron (B), and copper (Cu) were also quantified by ICP-OES. There were six replications in a completely randomized design.

2.2.2. Three Serianthes Species in Serianthes grandiflora Soil

A second soil manipulation study was conducted in Angeles City, Philippines using soil harvested from the rhizosphere of a S. grandiflora tree from Barangay San Agustin, Lavezares, Samar. The soil was harvested on 15 November 2018 and the study was initiated on 19 November 2018. The bulk soil was amended with washed river sand (1:2), sampled for C, N, P, and K analysis (n = 4), then separated into three treatments as described in Section 2.2.1.

The plant material consisted of S. grandiflora, S. kanehirae, and S. nelsonii seedlings. The S. grandiflora seeds were harvested from the tree from which the study’s soil was harvested, the S. kanehirae seeds were from Yap, and the S. nelsonii seeds were from Rota. The soil treatment and nursery methods were the same as previously described.

The experimental layout was a two-way factorial with three soil treatments and three species in a completely randomized design with six replications. The beginning plant size for S. grandiflora was 22 ± 5 cm in height and 3.1 ± 0.3 mm in stem diameter, for S. kanehirae was 21 ± 6 cm in height and 2.8 ± 0.2 mm in stem diameter, and for S. nelsonii was 20 ± 4 cm in height and 2.9 ± 0.2 mm in stem diameter. The study was terminated on 14 April 2019. Growth and leaf elemental concentrations were determined as previously described.

2.2.3. Serianthes grandiflora in Serianthes grandiflora Soil

A third soil manipulation study was conducted in the Angeles City research site using S. grandiflora seedlings grown in S. grandiflora soil. The soil was harvested on 1 March 2019 and the study was initiated on 5 March 2019. Growth of the S. grandiflora seedlings, preparation of the bulk soil, and sampling for chemical analysis (n = 4) were as previously described. The bulk soil was separated into five equal batches. Three of the batches were individually subjected to the three treatments described in Section 2.1.1. Additionally, the remaining two batches received commercial R. irregularis mycorrhiza inoculum (Great White Granular 1, Plant Revolution, Inc., Santa Ana, CA, USA). Product label instructions were followed with 142 g mixed in the medium for each 2.6-L container. One of these treatments consisted of the fresh soil mix, the other treatment consisted of the sterilized soil mix.

The beginning plant size was 17 ± 4 cm in height and 2.3 ± 0.2 mm in stem diameter. The plants were arranged in a completely randomized design and nursery conditions and methods were as previously described. The study was terminated on 30 July 2019. The growth traits and leaf chemical traits were measured as previously described. There were six replications. A graphical depiction of the experimental methods is shown in Figure 3.

2.3. Statistical Analysis

The results from each of the two Home versus Away Soil Studies were subjected to a t-test to determine the influence of soil source on differences in plant growth and leaf chemistry variables. For the soil manipulation studies, all response variables were subjected to analysis of variance using the PROC GLM procedure (SAS Institute, Cary, NC, USA). For Experiment 1, the S. kanehirae data were subjected to one-way ANOVA with three treatments. For Experiment 2, the factorial data were subjected to a two-way ANOVA with three species and three soil treatments serving as the two factors. For Experiment 3, the S. grandiflora data were subjected to a one-way ANOVA with five treatments. Means separations for significant response variables were conducted with Tukey’s HSD.

3. Results

3.1. Home versus Away Soil Studies

3.1.1. Serianthes nelsonii Soils

The Home soil mix contained more C, N, and K than the Away soil mix (

Table 1. Characteristics of soil and Serianthes kanehirae leaves from plants grown in Home versus Away soils. The Home soils were obtained from the rhizosphere of a Serianthes nelsonii tree in northern Guam. The Away soils were obtained from a 50–70 band surrounding the tree. Significance was determined by a t-test, n = 6.

Response Variable	Soil Home	Soil Away	p	Leaf Home	Leaf Away	p
Carbon 1	102.5 ± 9.4	70.3 ± 6.2	<0.001	452.2 ± 1.2	451.5 ± 1.5	0.305
Nitrogen 1	11.8 ± 0.6	5.5 ± 0.4	<0.001	18.5 ± 0.7	12.7 ± 0.5	<0.001
Phosphorus 2	36.2 ± 3.2	57.2 ± 3.4	0.001	2.4 ± 0.9	1.6 ± 0.9	0.008
Potassium 2	61.5 ± 2.3	44.5 ± 1.9	<0.001	12.4 ± 0.7	9.7 ± 0.6	<0.001

¹ Soil and leaf content in mg·g⁻¹; ² Soil content in µg·g⁻¹, leaf content in mg·g⁻¹.

). In contrast, the Away soil mix contained more P than the Home soil mix. The S. kanehirae plants grown in the Home soil mix exhibited 14% greater height growth (t = 4.487, p < 0.001) and 15% greater stem diameter growth (t = 3.226, p = 0.005) than plants grown in soils Away from the S. nelsonii tree (Figure 4). Leaf C content was not influenced by soil treatment, but leaf N was 46%, P was 50%, and K was 28% greater in the plants grown in Home soils than in the plants grown in Away soils (Table 1).

3.1.2. Serianthes grandiflora Soils

The Home soil mix contained more C, N, and K than the Away soil mix (

Table 2. Characteristics of soil and Serianthes grandiflora leaves from plants grown in Home versus Away soils. The Home soils were obtained from the rhizosphere of a Serianthes grandiflora tree in northern Samar, Philippines. The Away soils were obtained from a 50–70 band surrounding the tree. Significance was determined by a t-test, n = 6.

Response Variable	Soil Home	Soil Away	p	Leaf Home	Leaf Away	p
Carbon 1	115.5 ± 5.4	79.8 ± 3.3	<0.001	448.3 ± 2.2	444.9 ± 2.4	0.149
Nitrogen 1	15.5 ± 0.9	7.6 ± 0.6	0.002	16.4 ± 0.7	11.7 ± 0.5	<0.001
Phosphorus 2	37.8 ± 2.6	69.2 ± 3.4	<0.001	2.6 ± 0.9	1.4 ± 0.7	0.001
Potassium 2	65.3 ± 3.3	40.2 ± 1.9	<0.001	12.9 ± 0.7	9.0 ± 0.6	<0.001

¹ Soil and leaf content in mg·g⁻¹; ² Soil content in µg·g⁻¹, leaf content in mg·g⁻¹.

). The Away soil mix contained more P than the Home soil mix. The S. grandiflora plants grown in the Home soil mix exhibited 15% greater height growth (t = 3.957, p = 0.001) and 19% greater stem diameter growth (t = 2.914, p = 0.008) than plants grown in Away soils from the S. grandiflora tree (Figure 5). Leaf C content was not influenced by soil treatment, but leaf N was 40%, P was 86%, and K was 43% greater in the plants grown in Home soils than in the plants grown in Away soils (Table 2).

3.2. Soil Manipulation Studies

3.2.1. Serianthes kanehirae in Serianthes nelsonii Soil

The container medium exhibited C of 103 ± 8 mg·g⁻¹, N of 13 ± 2 mg·g⁻¹, P of 30 ± 3 µg·g⁻¹, and K of 58 ± 3 µg·g⁻¹. Every plant growth trait was influenced by the soil treatments, with the fresh soil supporting plant growth that exceeded that of the other two treatments, and the sterilization and fungicide treatments supporting plant growth that did not differ (

Table 3. The influence of growing Serianthes kanehirae plants in soil obtained from Serianthes nelsonii rhizosphere on growth in stem height (SHG), growth in stem diameter (SDG), maximum internode length (IL), and total leaf number (LN). Mean ± SE, n = 6.

Element	Control	Sterilized	Fungicide	f	p
SHG (mm)	648.5 ± 50.8 a 1	468.8 ± 32.2 b	483.2 ± 36.4 b	10.75	0.001
SDG (mm)	9.1 ± 0.3 a	7.3 ± 0.3 b	7.4 ± 0.3 b	13.44	<0.001
IL (mm)	36.8 ± 2.6 a	25.8 ± 1.9 b	24.8 ± 1.8 b	12.15	<0.001
LN	13 ± 2 a	10 ± 1 b	10 ± 1 b	8.07	0.004

¹ Means with the same letter within each row are not different according to Tukey’s HSD.

). Stem height growth was reduced by 27%, stem diameter growth was reduced by 19%, internode length was reduced by 31%, and leaf number was reduced by 39% for plants in treated soils compared with untreated Home soils. All three treatments generated plants that were healthy in appearance.

The influence of soil treatment on S. kanehirae leaf nutrient content was dependent on the element (

Table 4. The influence of growing Serianthes kanehirae plants in soil obtained from Serianthes nelsonii rhizosphere on leaf chemistry. Mean ± SE, n = 6.

Element	Control	Sterilized	Fungicide	f	p
Calcium 2	11.0 ± 0.3	11.1 ± 0.4	11.9 ± 0.6	1.29	0.305
Carbon 2	436 ± 5	434 ± 4	442 ± 5	0.84	0.451
Magnesium 2	2.9 ± 0.1 ab 1	2.7 ± 0.1 b	3.0 ± 0.1 a	5.27	0.019
Nitrogen 2	19.9 ± 1.2	15.9 ± 0.9	18.6 ± 1.1	3.31	0.060
Phosphorus 2	2.3 ± 0.1 a	1.2 ± 0.1 b	1.2 ± 0.1 b	193.34	<0.001
Potassium 2	12.7 ± 0.5 a	9.7 ± 0.5 b	10.4 ± 0.4 b	10.36	0.002
Boron 3	35.1 ± 0.5	36.3 ± 0.6	36.4 ± 0.8	1.75	0.208
Copper 3	3.3 ± 0.2	3.3 ± 0.2	3.3 ± 0.2	0	1.000
Iron 3	41.8 ± 1.7 b	53.7 ± 1.5 a	54.5 ± 1.4 a	30.18	<0.001
Manganese 3	24.4 ± 0.8	24.0 ± 0.7	23.8 ± 0.6	0.13	0.877
Zinc 3	23.2 ± 0.7 b	33.3 ± 1.0 a	32.3 ± 0.8 a	64.96	<0.001

¹ Means with the same letter within each row are not different according to Tukey’s HSD. ² (mg·g⁻¹). ³ (µg·g⁻¹).

). For the elements that were significantly different among the treatments, the pattern among the soil treatments was generally similar to that of the growth traits. Leaf P content was reduced by 48% and leaf K was reduced by 21% for plants in the treated soils when compared with the plants in the control soils. Leaf content of P, K, and all other nutrients did not differ between the plants in the sterilized and fungicide-treated soils. Leaf Fe was increased by 29% and leaf Zn was increased by 41%for plants in the treated soils compared with the control soils. The behavior of Mg was unique, with leaves of plants growing in the fungicide treatment exhibiting content that exceeded that of plants growing in the sterilized soil treatment, and leaves of plants growing in control soils exhibiting intermediate content. The leaf content of the remaining elements was not influenced by soil treatments.

3.2.2. Three Serianthes Species in Serianthes grandiflora Soil

The container medium contained C of 133 ± 10 mg·g⁻¹, N of 17 ± 3 mg·g⁻¹, P of 36 ± 3 µg·g⁻¹, and K of 66 ± 4 µg·g⁻¹. The growth and leaf element response variables were greatly influenced by soil treatment, with B, C, and Cu content being the only response variables that were not significantly influenced by treatment (

Table 5. Results from 3 species × 5 soil treatment factorial ANOVA. Four plant growth traits and leaf content of 11 elements.

Response Variable	Species f2,45	Species p	Treatment f2,45	Treatment p	S × T f4,45	S × T p
Leaf number	7.098	0.002	27.607	<0.001	0.828	0.515
Internode length	2.252	0.117	25.805	<0.001	0.491	0.742
Height growth	6.725	0.003	28.293	<0.001	1.170	0.337
Stem diameter growth	0.030	0.970	34.030	<0.001	0.303	0.874
Calcium	0.066	0.936	11.397	<0.001	0.319	0.864
Carbon	0.248	0.782	1.513	0.231	0.078	0.998
Magnesium	1.204	0.309	5.061	0.010	0.432	0.785
Nitrogen	0.362	0.698	21.436	<0.001	0.065	0.992
Phosphorus	0.185	0.831	610.215	<0.001	0.245	0.911
Potassium	0.263	0.769	15.320	<0.001	0.164	0.956
Boron	0.539	0.587	0.013	0.988	0.401	0.807
Copper	0.542	0.586	1.792	0.179	0.167	0.954
Iron	0.304	0.739	192.982	<0.001	0.076	0.989
Manganese	1.157	0.324	3.568	0.036	0.076	0.989
Zinc	0.899	0.414	130.775	<0.001	0.135	0.969

). In contrast, only leaf number and growth in height were influenced by species, with the remainder of the variables remaining uninfluenced by species. Moreover, no response variable was significant for the species × treatment interaction, indicating the three species behaved similarly in response to the three soil treatments.

The ending leaf number was not different for S. grandiflora (11 ± 1) and S. kanehirae (11 ± 1), but the ending leaf number for S. nelsonii (9 ± 1) was less than that of the other species. Similarly, the increase in height was not different for S. grandiflora (533 ± 26) and S. kanehirae (524 ± 24), but the increase in height for S. nelsonii (461 ± 19) was less than that of the other species.

The increase in height, increase in stem diameter, internode length, and leaf number did not differ between Serianthes plants that were grown in sterilized versus fungicide-treated soils (

Table 6. The influence of growing Serianthes grandiflora, Serianthes kanehirae, and Serianthes nelsonii plants in soil obtained from Serianthes grandiflora rhizosphere on plant growth traits. Mean ± SE, n = 18.

Treatment	Stem Height Growth (mm)	Stem Diameter Growth (mm)	Internode Length (mm)	Leaf Number
Control	601.2 ± 17.1 a 1	8.9 ± 0.2 a	33.6 ± 1.6 a	12 ± 2 a
Sterilized	458.9 ± 11.9 b	7.1 ± 0.1 b	24.4 ± 0.9 b	9 ± 1 b
Fungicide 2	460.4 ± 10.7 b	7.2 ± 0.2 b	24.0 ± 0.8 b	9 ± 1 b

¹ Means with the same letter within each column are not different according to Tukey’s HSD. ² Soil drench with propiconazole.

). In contrast, the plants that were grown in the control soils exhibited stem height growth that was 47% greater, stem diameter growth that was 20% greater, internode length that was 28% greater, and leaf number that was 25% greater than plants grown in the sterilized or fungicide-treated soils.

Leaf C content did not differ among the soil treatments, and the Serianthes plants produced leaves that contained 444 ± 4 mg·g⁻¹ C. The control plants produced leaves with N content that was 26% greater, P content that was 50% greater, K content that was 12% greater, and Ca content that was 9% greater than plants grown in sterilized or fungicide-treated soil (

Table 7. The influence of growing Serianthes grandiflora, Serianthes kanehirae, and Serianthes nelsonii plants in soil obtained from Serianthes grandiflora rhizosphere on leaf macronutrient content. Mean ± SE, n = 18.

Treatment	Nitrogen (mg·g−1)	Phosphorus (mg·g−1)	Potassium (mg·g−1)	Calcium (mg·g−1)	Magnesium (mg·g−1)
Control	22.6 ± 0.8 a 1	2.4 ± 0.1 a	13.9 ± 0.3 a	11.6 ± 0.2 a	2.9 ± 0.1 a
Sterilized	15.8 ± 0.7 b	1.2 ± 0.1 b	12.4 ± 0.4 b	10.5 ± 0.2 b	2.6 ± 0.1 b
Fungicide 2	17.6 ± 0.8 b	1.2 ± 0.1 b	12.0 ± 0.3 b	10.7 ± 0.3 b	2.9 ± 0.1 a

¹ Means with the same letter within each column are not different according to Tukey’s HSD. ² Soil drench with propiconazole.

). The leaf contents of these macronutrients did not differ between plants growing in sterilized versus fungicide-treated soils. Leaf Mg content of plants grown in sterilized soil was less than plants grown in control or fungicide-treated soil.

For the micronutrients, leaf B and Cu content did not differ among the soil treatments. The Serianthes plants contained leaves with 35.8 ± 0.3 µg·g⁻¹ B and 3.2 ± 0.1 µg·g⁻¹ Cu. The Fe content was 33% greater and Zn content was 44% greater for Serianthes leaves from plants grown in treated soils compared with control soils (

Table 8. The influence of growing Serianthes grandiflora, Serianthes kanehirae, and Serianthes nelsonii plants in soil obtained from Serianthes grandiflora rhizosphere on leaf micronutrient content. Mean ± SE, n = 18.

Treatment	Iron (µg·g−1)	Manganese (µg·g−1)	Zinc (µg·g−1)
Control	40.5 ± 1.2 b 1	24.4 ± 0.5 a	22.5 ± 0.5 b
Sterilized	54.3 ± 1.3 a	23.1 ± 0.4 ab	32.6 ± 0.6 a
Fungicide 2	53.2 ± 1.4 a	22.7 ± 0.4 b	32.2 ± 0.5 a

¹ Means with the same letter within each column are not different according to Tukey’s HSD. ² Soil drench with propiconazole.

). The leaf content of these micronutrients did not differ between the sterilized versus fungicide treatments. Leaf Mn content of plants grown in control soils exceeded that of plants grown in fungicide-treated soils. Leaf Mn content of plants grown in sterilized soils was intermediate.

3.2.3. Serianthes grandiflora in Serianthes grandiflora Soil

The container medium exhibited C of 125 ± 8 mg·g⁻¹, N of 18 ± 3 mg·g⁻¹, P of 38 ± 3 µg·g⁻¹, and K of 63 ± 4 µg·g⁻¹. The S. grandiflora plants exhibited growth traits that responded to soil sterilization and fungicide treatments in a manner similar to the previous studies. Plants growing in control soils exhibited stem height growth that was 19% greater, stem diameter growth that was 16% greater, internode length that was 18% greater, and leaf number that was 23% greater than plants growing in sterilized or fungicide-treated soils (

Table 9. The influence of growing Serianthes grandiflora plants in soil obtained from Serianthes grandiflora rhizosphere on plant growth traits. Mean ± SE, n = 6.

Treatment	Stem Height Growth (mm)	Stem Diameter Growth (mm)	Internode Length (mm)	Leaf Number
Control	547.2 ± 28.9 a 1	8.0 ± 0.4 a	29.2 ± 2.3 a	11 ± 1 a
Sterilized	452.7 ± 21.4 b	6.6 ± 0.3 b	23.7 ± 2.0 b	8 ± 1 b
Fungicide 2	428.7 ± 22.6 b	6.8 ± 0.4 b	24.0 ± 1.8 b	9 ± 1 ab
Control + myco 3	538.2 ± 23.9 a	7.6 ± 0.3 a	26.3 ± 1.9 ab	10 ± 1 a
Sterilized + myco 4	531.8 ± 30.1 a	8.1 ± 0.4 a	29.2 ± 2.5 a	10 ± 1 a
f4,25	3.974	4.839	4.171	2.925
p	0.013	0.005	0.010	0.041

¹ Means with the same letter within each column are not different according to Tukey’s HSD. ² Soil drench with propiconazole. ³ Control soil plus Rhizophagus irregularis inoculum. ⁴ Sterilized soil plus Rhizophagus irregularis inoculum.

). The addition of a commercial mycorrhiza inoculum to control or sterilized soils generated plant growth that was not different from that of control soils.

Leaf Ca content did not differ among the soil treatments (f_4,25 = 0.679, p = 0.613), and the S. grandiflora plants produced leaves that contained 13.3 ± 0.3 mg·g⁻¹ Ca. The plants grown in sterilized soil produced leaves with less C than the plants grown in the other four soils (

Table 10. The influence of growing Serianthes grandiflora plants in soil obtained from Serianthes grandiflora rhizosphere on leaf macronutrient content. Mean ± SE, n = 6.

Treatment	Carbon (mg·g−1)	Nitrogen (mg·g−1)	Phosphorus (mg·g−1)	Potassium (mg·g−1)	Magnesium (mg·g−1)
Control	453.5 ± 1.6 a 1	28.7 ± 1.3 a	2.5 ± 0.2 a	16.1 ± 0.4 a	2.8 ± 0.2 ab
Sterilized	449.7 ± 1.1 b	17.9 ± 1.2 b	1.4 ± 0.1 b	13.7 ± 0.3 b	2.6 ± 0.1 b
Fungicide 2	453.7 ± 1.8 a	18.1 ± 1.4 b	1.2 ± 0.1 b	13.8 ± 0.4 b	3.3 ± 0.3 a
Control + myco 3	453.7 ± 1.8 a	28.9 ± 1.4 a	2.4 ± 0.2 a	17.0 ± 0.6 a	2.8 ± 0.2 ab
Sterilized + myco 4	455.8 ± 1.5 a	17.3 ± 1.1 b	2.5 ± 0.2 a	13.4 ± 0.4 b	2.6 ± 0.1 b
f4,25	3.74	218.23	39.13	65.69	18.34
p	0.016	<0.001	<0.001	<0.001	<0.001

¹ Means with the same letter within each column are not different according to Tukey’s HSD. ² Soil drench with propiconazole. ³ Control soil plus Rhizophagus irregularis inoculum. ⁴ Sterilized soil plus Rhizophagus irregularis inoculum.

). The control plants exhibited N content that was 37% greater, P content that was 48% greater, and K content that was 15% greater than plants growing in sterilized or fungicide-treated soils. Adding a commercial mycorrhiza inoculum to the control soils did not affect N, P, or K content. In contrast, adding a commercial mycorrhiza inoculum to the sterilized soils increased leaf P content 79% above that of the sterilized soils. Leaf N or K content did not respond to the inoculum supplied to the sterilized soils. The patterns of leaf Mg content were unique among the soil treatments, as plants grown in fungicide-treated soils exhibited the greatest and plants grown in both sterilized soil treatments exhibited the least Mg.

For the micronutrients, leaf Cu (f_4,25 = 0.685, p = 0.609) and Mn (f_4,25 = 0.613, p = 0.657) content did not differ among the soil treatments. The S. grandiflora plants contained leaves with 3.4 ± 0.2 µg·g⁻¹ Cu and 23.9 ± 0.5 µg·g⁻¹ Mn. Plants grown in control soils exhibited leaf B content that was 8% less, Fe content that was 17% less, and Zn content that was 29% less than plants grown in sterilized or fungicide-treated soils (

Table 11. The influence of growing Serianthes grandiflora plants in soil obtained from Serianthes grandiflora. rhizosphere on leaf micronutrient content. Mean ± SE, n = 6.

Treatment	Boron (µg·g−1)	Iron (µg·g−1)	Zinc (µg·g−1)
Control	34.2 ± 0.6 b 1	43.8 ± 1.8 b	22.2 ± 0.8 b
Sterilized	36.5 ± 0.6 a	52.8 ± 1.3 a	31.7 ± 1.1 a
Fungicide 2	37.5 ± 0.9 a	52.7 ± 1.4 a	30.5 ± 0.9 a
Control + myco 3	36.8 ± 0.9 a	42.7 ± 1.3 b	24.1 ± 1.1 b
Sterilized + myco 4	36.6 ± 0.6 a	51.3 ± 1.5 a	31.7 ± 1.1 a
f4,25	2.93	15.52	3.52
p	0.041	<0.001	<0.001

¹ Means with the same letter within each column are not different according to Tukey’s HSD. ² Soil drench with propiconazole. ³ Control soil plus Rhizophagus irregularis inoculum. ⁴ Sterilized soil plus Rhizophagus irregularis inoculum.

). The leaf content of these micronutrients did not differ between the sterilized and fungicide treatments. Additions of a commercial mycorrhiza inoculum to sterilized soils did not lead to a change in leaf B, Fe, or Zn content. In contrast, additions of commercial mycorrhiza inoculum to control soils increased leaf B content but did not influence Fe or Zn content.

4. Discussion

Perennial trees engage in PSFs that alter the rhizosphere soils over time. The results herein indicated some of the Serianthes PSF outcomes include soil microorganisms that promote Serianthes plant growth and leaf nutrition. The first hypothesis was that plant growth and macronutrient content of leaves would be greater when Serianthes plants were grown in Home soils versus Away soils. The hypothesis was confirmed, in that stem growth was 14–19% greater, leaf nitrogen was 40–46% greater, leaf phosphorus was 50–86% greater, and leaf potassium was 28–43% greater when grown in soils from Serianthes rhizosphere than in soils away from a Serianthes tree. The second hypothesis was that plant growth would be reduced by soil sterilization and fungicide treatment of Home soils, This hypothesis was confirmed with S. kanehirae plants growing in S. nelsonii soils, and three Serianthes species growing in S. grandiflora soils. The third hypothesis was that leaf P would be influenced by the soil treatments to a greater degree than the other nutrients. This hypothesis was confirmed in that P content was up to 86% greater in Home soils versus Away soils and the sterilization and fungicide treatments of Home soils reduced P content by up to 50% below that of control soils.

This new knowledge may inform ongoing conservation endeavors. For example, recovery efforts for the critically endangered S. nelsonii have been characterized by post-transplant mortality of saplings from the nursery to the forest, revealing the need for more research to improve post-transplant success [21]. Although the methods of the present study focused on plant responses and not directly on soil microorganism responses, the results indicated the biotic characteristics of Serianthes PSFs may be exploited by conservation practitioners by transferring Serianthes rhizosphere soil inoculum to transplant sites in an effort to improve plant growth and survival. This protocol conforms to the approach of using knowledge about facilitation to develop tools for ecosystem restoration [25,26]. Indeed, correcting historical failures in transplanting endangered plants for species recovery may require the development of a better understanding of the symbiotic relationships between the soil microbiome and the plant [27].

The level of Serianthes host specificity required for mycorrhizal efficacy cannot be understood without more detailed manipulative studies. However, the outcomes of this series of studies illuminated several relevant issues. First, the Away soils contained more P than the Home soils, yet the Serianthes plant leaves contained more P when grown in the Home soils. These results indicated the fresh Away soil microbiome did not benefit Serianthes plant access to soil P as much as the fresh Home soil microbiome. Second, the microorganism biodiversity in the S. nelsonii rhizosphere soils benefitted S. kanehirae plants, and the microorganism biodiversity in S. grandiflora soils benefitted S. kanehirae and S. nelsonii plants as much as it benefitted S. grandiflora plants. These outcomes indicated that host specificities of edaphic mutualists were not constrained to the species level and appeared to extend to the genus level. Third, inoculation of sterilized soils with the generalist arbuscular mycorrhiza R. irregularis fully reversed the reductions in S. grandiflora plant growth and leaf P that were caused by sterilization of Home soils. These findings indicated fidelity to native S. grandiflora rhizosphere microbiota inoculum was not mandatory in order to improve plant performance by mutualisms with a commonly used mycorrhiza taxon.

The findings reported herein augment ecological knowledge that has resulted from recent Serianthes research. First, soil nutrient differences in Home vs. Away soils have been previously reported for S. nelsonii [28]. The results of the present study confirm these findings and expand this knowledge to include S. grandiflora trees. Second, based on leaf tissue stoichiometry, P is the most limiting nutrient for in situ S. nelsonii in Guam [29]. The ability to use native microbiota to partly mitigate this deficiency should be pursued in conservation planning. Application of fertilizer to foliage as a dilute aerosol [30] and use of P-only granular fertilizer [31] may also be factored into Serianthes nutrient management protocols. Third, the use of mefenoxam fungicide has been shown to increase the longevity of in situ S. nelsonii seedlings, indicating that soil-borne pathogens from the rhizosphere may result in negative biotic PSF [32]. In contrast, the results reported here revealed positive biotic PSF components of Serianthes PSFs. A net balance among negative and positive biotic components of the Serianthes rhizosphere will require more research to fully understand all context-dependent relationships among the soil microbiota taxa. Fourth, the release of C and N from S. nelsonii litter was relatively rapid in previous reports, indicating S. nelsonii may provide unique ecosystem services that include rapid nutrient turnover to create spatial heterogeneity of resources [33]. This previous study used non-Serianthes soils for the litter incubations, and the results of the current study indicate more litter decomposition studies are needed with Home soils to more fully understand the role of the home field advantage [34] in Serianthes litter decomposition behaviors.

The sterilization and fungicide treatments of Home soils reduced leaf P as predicted, but also influenced other leaf nutrient relations. First, the sterilization and fungicide treatments consistently increased leaf Fe and Zn content. Homeostasis among P and these two micronutrients is well-known [35], and the increases in Fe and Zn may have been associated with the reductions in leaf P content that resulted from the sterilization and fungicide treatments. Second, leaf N and K were reduced by Home soil sterilization or fungicide treatments in some but not all of the studies. Indeed, many plant species exhibit greater access to macronutrients other than P when supported by mycorrhizal mutualisms [36,37,38,39]. Mycorrhizal fungal communities may access soil N from organic sources [40], and the limited mycorrhiza activity in the sterilized and fungicide-treated soils may have caused the reduction in leaf N via this mechanism. The Guam study did not exhibit reduced leaf N for the sterilized and fungicide treatments, but the Philippine study did. These responses may indicate that the Guam soils contain fewer organic forms of N than the Philippine soils, or that the mycorrhiza species in the S. nelsonii rhizosphere are less efficient in accessing organic N than the species in the S. grandiflora rhizosphere. Indeed, mycorrhizal identity can influence the C, N, and P of host plants [41]. Third, the influence of soil treatments on leaf Mg was more diverse among the studies than on other leaf nutrients. For Serianthes, leaf Mg relationships may be complex and independent of soil microbiota abundance and activity.

More context-dependent research is needed to better understand the Serianthes soil microbiota and how to exploit these organisms for conservation efforts. First, the use of commercial mycorrhizal inoculants to enhance the growth of managed crops has become of interest as a form of biofertilizer [42]. The richness of commercial or harvested mycorrhizal inoculum, the volume of inoculum, and differences in native versus commercial sources of mycorrhiza may exert profound influences on crop response [43,44]. The present study has shown that the commercial mycorrhiza R. irregularis was highly effective in reversing the reductions in S. grandiflora plant growth and leaf P content that were caused by sterilization of Home soil. The use of a commercial mycorrhiza inoculum is advantageous in that it carries no risk of inadvertent inoculation of new planting sites with soil-borne Serianthes rhizosphere pathogens. However, the federal permits for handling endangered species in the United States Territory of Guam do not allow the use of non-native organisms such as commercial mycorrhiza for plants that are grown for transplanting to in situ plantings. This issue has been discussed in the context of using non-native rootstocks for grafting clonal S. nelsonii plants [45]. Many S. nelsonii trees may be planted in urban or peri-urban settings for the purpose of seed production, and in these venues, there is no need to maintain strict restrictions on the use of non-native beneficial organisms to improve S. nelsonii population recovery efforts. Second, the use of native microbiota from S. nelsonii rhizosphere soil is hindered by a lack of knowledge, so more research is needed to tease apart the nuanced positive and negative PSFs that may be exploited for improved species recovery efforts. The federal permits for handling endangered species in the United States Territory of Guam contain restrictions on the use of field soil in container nurseries. This needed research cannot proceed unless those restrictions are modified such that competent researchers are allowed to answer critical questions during the nursery phase. Third, the mycorrhiza literature is replete with context-dependent examples that should be the focus of future research. My study design included homogeneous 50% shade provided by a commercial shade screen and individual containers for each replication such that non-self-roots were never engaged by the experimental plants. Mycorrhizal contributions to host plants are influenced by proximity to and identity of companion plants or quantity and quality of incident light [46,47,48]. Therefore, more research is needed, especially within in situ settings where every individual plant is exposed to heterogeneous light conditions, and interaction with non-self-roots is unavoidable. Fourth, there is a limit to the volume of soil that can be harvested from the rhizosphere of individual trees. A dose-response study is needed to determine the smallest volume of soil that would generate the benefits from the use of rhizosphere soils as microbiota inoculum.

5. Conclusions

A series of studies in Guam and the Philippines has shown that soils from the rhizosphere of mature Serianthes trees improve Serianthes plant growth and increase leaf nutritional status. Stem growth was up to 19% greater, leaf nitrogen was up to 46% greater, leaf phosphorus was up to 86% greater, and leaf potassium was up to 43% greater when seedlings were grown in soils from a Serianthes rhizosphere than in soils away from a Serianthes tree. Therefore, rhizosphere soils may be used as an inoculum in new planting sites as a means of improving Serianthes transplant growth and survival. This knowledge may increase the conservation successes of endangered trees. The ongoing research needs to include balancing the positive and negative outcomes by determining methods to reduce negative PSF factors such as soil-borne pathogens or allelochemicals.