More is not always better: responses of the endemic plant Vellozia nanuzae to additional nutrients

ABSTRACT

Soil nutrients are one of the main drivers of plant species composition and distribution, mainly due to the role they play in plant survival and reproductive success. However, the nutritional requirements of plants inhabiting their native ecosystems are still poorly known. This is the case for most species of campo rupestre vegetation, which are found on infertile soils of quartzitic and ferruginous origin. The present study evaluated the effects of macronutrients and substrates on survival and growth of the micro-endemic Vellozia nanuzae (Velloziaceae). Plant mortality was about 95 % higher in soil enriched with macronutrients and almost 100 % in soils with added manure in the first 30 days. Individual plants from treatments with added nutrients had lower growth (53 % less) compared to individuals on natural substrates. In conclusion, Vellozia nanuzaesurvived and developed better on soil of its original habitat even though it is acidic and poor in nutrients. Our results show that more nutrients are not always better for the survival and development of native species that inhabit harsh ecosystems.

Keywords: campo rupestre; plant nutrition; rupestrian grasslands; soil nutrients; Velloziaceae

Introduction

Plant survival and development are closely related to soil nutrient availability (Chapin III 1980; Epstein & Bloom 2004). Many nutrients, such as N, P, K, Ca, Mg, S, B, Cu, Cl, Fe, Mn, Mo and Zn, are essential to tissue formation and hence plant metabolism and development (Uchida 2000; Razaq et al. 2017). On the other hand, when in excess in natural ecosystems, these nutrients can be harmful and lead to plant mortality (Uchida 2000; Echart & Cavalli-Molina 2001; Negreiros et al. 2014). Anthropogenic activities (Melnikova & Sasai 2020) and global climate change (Hunt et al. 2020) are important sources of increased soil fertilization (Miller & Miller 1998; Matias et al. 2011; Xu et al. 2019), a global phenomenon that may affect native species inhabiting low fertile habitats (Fernandes 2016a). Other soil parameters may interact with nutrients further changing their availability in the soil. For instance, increased acidity can reduce the availability of N, P, K, Ca, Mg, and S, and thus soil fertility (e.g., Arens 1958; Goodland & Ferri 1979; Schaefer et al. 2016). The physical attributes of soil, such as texture and granulometry, also play important roles in plant development and success. For instance, sandy soils contain a large amount of coarse particles and have lower cation exchange capacity, which result in better soil aeration but reduced water retention and nutrient concentrations (Rawls et al. 1991; Klein & Klein 2015). Altogether, soil attributes represent important filters that ultimately determine plant success and hence plant community assembly (see Wijesinghe et al. 2005; Negreiros et al. 2014).

Fertilization techniques used to reduce acidity, such as liming, and alterations to soil substrate have been widely used to achieve increased vigor and development of plants. Although widely used in agriculture and silviculture, such techniques are not always successful for every ecosystem, as many plant species do not respond to them in the same way. In fact, many plant species develop better on soils that are considered nutritionally poor, such as in many natural grasslands and savannic formations (e.g., Haridasan 1988; 2000; Negreiros et al. 2009). It is surprising that many ecosystems that are considered nutrient-poor, acidic and with high aluminum concentrations, such as campo rupestre (rupestrian grassland), possess high plant diversity and endemism (Conceição et al. 2016; Fernandes 2016a). Campo rupestre is an ancient grassy ecosystem found in areas above 900 ml where rocky outcrops prevail (e.g., Fernandes 2016a; Silveira et al. 2016). As these species grow on nutrient-impoverished soils and under harsh environmental conditions, the vegetation is predominantly sclerophyllous and herbaceous with scattered shrubs and trees (Negreiros et al. 2014; for details see Fernandes 2016a).

There are several reasons to better understand the nutritional responses of plants of the campo rupestre ecosystem. First, knowledge of plant growth and development is of major importance for the restoration of this ecosystem, which has experienced high rates of habitat conversion and impacts from climate change (e.g., Fernandes et al. 2014; 2018; Fernandes 2016b; Pena et al. 2017). The restoration of campo rupestre is not an easy task due to the high climatic seasonality with a long period of water shortage and high soil temperatures, among other factors (see review in Fernandes et al. (2016). Another challenge is achieving restoration by seeding because the low viability of the seeds of many species requires that an extremely high quantity be used (Stradic et al. 2014; Dayrell et al. 2016; but see Negreiros et al. 2016; Gomes et al. 2015; 2018). Second, habitat conversion results in drastic changes to soil structure and quality (Morgan & Connolly 2013; Fernandes et al. 2016). For instance, the conversion of natural soils to cultivation can result in compaction, erosion and leaching of nutrients as wells as excessive fertilization, contamination by pesticides and acidification or alkalization (Hillel 2007). Moreover, another common land use in campo rupestre is mining, the activities of which cause deep and irreversible disruption to soil structure with the removal of the thin surface soil layer (topsoil) (e.g., Fernandes et al. 2014), along with its seed and root bank, further impacting plant colonization and seed germination (Fernandes et al. 2016). An opposite effect is that of road paving, which leads to the deposition of calcium, mostly along road margins (Barbosa et al. 2010). Calcium deposition leads to decreased aluminum toxicity, which produces habitat conditions that enable invasion by exotic ruderal species (Barbosas et al. 2010; Hilário et al. 2011). Third, the management and conservation of endemic plant species require special understanding and attention given the increasing threats to campo rupestre (e.g., Fernandes et al. 2018), a first step of which is understanding plant responses to nutrient availability.

The herbaceous species Vellozia nanuzae(Velloziaceae) is a micro-endemic of the campo rupestre ecosystem in Serra do Cipó, Minas Gerais, Brazil. The species is only found among a few rocky crevices (Mello-Silva 2015) in a limited geographic range. Given its restricted distribution, there is an urgent need to gain more knowledge about the physiology and propagation of the species so that proactive conservation strategies and adequate management can be deployed (Fernandes et al. 2007). Moreover, this species is also known to have a high potential for bioprospecting (Barbosa & Fernandes 2008). The leaves of V. nanuzae contain secondary compounds that have potential for cancer treatment, such as nanuzone and 11β-hydroxy-nanuzone (Pinto et al. 1988). The species also has potential for cosmetic use due to the resin produced by all of its organs (Maurya & Yadav 2005).

The aim of the present work was to compare the survival and growth of V. nanuzae seedlings in different substrates. To do so, soil from the natural habitat of the species was used with varying the levels of nutrient addition and proportions of other substrates. Two hypotheses were tested. Hypothesis I: Seedlings of V. nanuzae exposed to nutrient addition will exhibit lower survival and higher growth rates than seedlings exposed to soils with lower nutrient availability. The rationale behind this hypothesis is that species that are tolerant to nutritionally poor soils will survive better in soils of lower nutritional quality. Some plant species that live in nutritionally poor habitats have traits that retain and conserve nutrients, and thus ensure increased survival under such conditions (e.g., Coley et al. 1985; Grime & Campbell 1991; Arendt 1997; Warembourg & Estelrich 2001). In addition, Negreiros et al. (2014) found that some campo rupestre species have lower survival in soils with added nutrients, further providing support for this hypothesis. Nutritional increments generally stimulate increased plant growth; however, for plant species naturally inhabiting low fertility soils the response is less pronounced, if at all, because of their tolerance strategies, such as a stress resistance syndrome that confers slow growth (Chapin III et al. 1993; Aerts & Chapin 2000). Hypothesis II: seedlings of V. nanuzae grown in naturally occurring soil of campo rupestre will have a higher proportion of roots compared to aerial biomass. It is expected that plants tolerant of nutrient-poor soils will allocate more to roots due to limited nutrients (Chapin III 1980; Bloom et al. 1985; Wilson 1988; Mooney & Winner 1991; Moreira & Klink 2000; Hoffmann & Franco 2003). A common strategy for plant species of the campo rupestre ecosystem is strong investment in their underground structures due to strong pressures on aerial parts, such as fire and herbivory (see reviews in Fernandes 2016a; Pausas et al. 2018). Furthermore, roots are responsible for water absorption during the dry period, hence supporting the contention of high investment in the root system (Brum et al. 2017).

Materials and methods

Study area

The experiment was carried out in a greenhouse at Vellozia Reserve (19°17'46" S, 43°35'28" W; 1,200 m), Serra do Cipó, Minas Gerais, Brazil, between December 2016 and June 2017. The local climate is classified as mesothermic (Cwb), according to the classification of Köppen, which is characterized by dry winters and wet summers. The average annual rainfall is 1,500 mm and the average temperature varies between 17.4 and 19.8 ºC (see Madeira & Fernandes 1999).

Experimental design

Seeds of V. nanuzae L.B.Sm. & Ayensu were obtained through manual collection of ripe fruits in September 2016. The seeds for the experiment were sorted from those having signs of predation, pathogens or malformations. The seeds were then seeded in six 128-cell styrofoam trays (3 cm X 3 cm X 5.1 cm) with four or five seeds per cell. The substrate used for seed germination and seedling production was composed of equal parts of washed sand, vermiculite and vegetable soil (modified from Macedo et al. 1993).

After 120 days, 600 individuals of 0.5 to 1 cm in height (average height = 0.75 cm) were collected and transferred to plastic containers (8.5 cm in diameter and 18 cm in depth) of six treatments (n= 100 individuals per treatment): control (soil from the natural habitat of V. nanuzae); three treatments with the addition of different concentrations of nitrogen (N), phosphorus (P), and potassium (K) (NPK/100: 15 g/m³ of NPK + 20 g/m³ in dolomitic limestone; NPK/10: 150 g/m³ of NPK+ 200 g/m³ in dolomitic limestone; and NPK: 1500 g/m³ of NPK + 2000 g/m³ in dolomitic limestone); and two treatments with substrates containing different proportions of control soil (C), sand (S) and manure (M) (3:2:1 and 2:0:1, respectively) (see Tab. 1 for details). We used the manure marketed as Terral Esterco®. About 1.5 kg of soil was added to each container. The NPK used was 10-10-10 (10 % N, 10 % P2O5, 10 % K20), Vitaplan® brand. The amount of formulation of these macronutrients was based on the manufacturer's (Vitaplan) recommendations. The amount de NPK used for NPK/100 (15 g/m³) and NPK/10 (150 g/m³ of NPK) was, respectively, 100 and 10 less than those prepared for NPK (1500 g/m³). The campo rupestre soil used for the experiment was collected at points near where the species occurs in Vellozia Reserva. The soil was collected to a depth of 10-20 cm and within a 20 cm diameter around each point. After drying, the collected soils were sieved with a 5 mm mesh, to remove large fragments of gravel, and homogenized.

Table 1
Treatments with different concentration of nitrogen (N), phosphorus (P), and potassium (K) and proportions of substrates - native soil (control - C), sand (S) and manure (M) - used to test survival and growth of Vellozia nanuzae (Velloziaceae).

Treatments	Substrate	Liming	Fertilization
Control (C)	Campo Rupestre	No	No
NPK/100	(C)	Yes = 20 g/m3 in dolomitic limestone *.	15 g/m3 of NPK
NPK/10	(C)	Yes = 200 g/m3 in dolomitic limestone.	150 g/m3 of NPK
NPK	(C)	Yes = 2000 g/m3 in dolomitic limestone.	1500 g/m3of NPK
C+S+M	(C) + sand + manure (3:2:1).	No	No
C+M	(C) and manure (2:1).	No	No

• *Dolomitic limestone (PRNT 80 %, CaO 37 %, MgO 14 %, humidity 5 %).

Three soil samples from each treatment, chosen at random, were homogenized and sent for chemical and physical analyses prior to the experiment. The analyses were performed by Departamento de Solo, Universidade de Viçosa, Brazil. The method used was according to Donagema et al. (2011). The following parameters were measured for each sample: pH; organic matter, N, P, K, Ca^{2 +} and Mg²⁺ content; microelements S and B; potential acidity (H + Al); sum of bases (SB); cation exchange capacity at pH 7.0 (T); effective cation exchange capacity (t); base saturation index (V %); and remaining phosphorus (P-Rem). The results for each treatment are presented in Table 2.

Table 2
Chemical and physical properties of the soil of the six treatments used to test survival and growth of Vellozia nanuzae. Three treatments had the addition of different concentrations of nitrogen (N), phosphorus (P), and potassium (K): 1) NPK/100 (15 g/m³ of NPK + 20 g/m³ in dolomitic limestone); 2) NPK/10 (150 g/m³ of NPK+ 200 g/m³ in dolomitic limestone); and 3) NPK (1500 g/m³ of NPK + 2000 g/m³ in dolomitic limestone). Three treatments had different proportions of substrates - native soil (control - C), sand (S) and manure (M): 1) C; 2) C+S+M; 3) 2C:1M

Parameters	Control	NPK/100	NPK/10	NPK	C+S+M	C+M
pH (H2O)	5.17	6.38	5.94	6.72	7.26	7.22
N (dag/kg)	0.08	0.05	0.05	0.07	0.1	0.29
P (mg/dm3 )	0.3	0.4	1.6	29.3	72.4	142
K (mg/dm3)	11	35	68	367	536	1.17
Ca2+ (c molc/dm3)	0.56	2.32	2.78	12.6	5.06	7.39
Mg2+ (c molc/dm3)	0.04	0.18	0.21	0.54	1.19	2.16
H + Al (c molc/dm3)	1.7	2	1.7	1.7	0.8	1.7
SB (c molc /dm3)	0.63	2.59	3.16	14.1	7.62	12.5
t (c molc/dm3)	0.63	2.59	3.16	14.1	7.62	12.5
T (c molc/dm3)	2.33	4.59	4.86	15.8	8.42	14.2
V %	27	56.4	65	89.3	90.5	88.1
P-Rem (mg/L)	25.7	23	27	26.1	43	43.9
Coarse sand (kg/kg)	0.19	0.21	0.19	0.19	0.50	0.19
Fine sand (kg/kg)	0.33	0.28	0.30	0.31	0.19	0.29
Silt (kg/kg)	0.35	0.35	0.36	0.36	0.20	0.35
Clay (kg/kg)	0.12	0.14	0.13	0.13	0.09	0.15

• Footnote: pH in water, KCl and CaCl 1:2.5 ratio; N extracted by sulphurous digestion, Kjeldhal distillation; P and K extracted by Extractor Mehlich 1; Ca^{2 +} and Mg^{2 +}, extractor 1 mol / L KCl ; H + Al, extractor 0.5 mol / L calcium acetate pH 7.0.

The containers containing the substrates and seedlings were arranged in a completely randomized design of six treatments × 100 replicates for 600 individual seedlings. The seedlings were grown in a greenhouse covered with shading material (50 %) and irrigated by micro-sprinklers for one minute twice a day. The containers possessed holes allowing free drainage of water. Manual weeding was performed weekly to eliminate invasive plants. Individuals of each treatment were evaluated once a month for five months (150 days) after transplantation.

Assessment of survival and growth

To evaluate plant survival according to treatment, the numbers of surviving and dead individuals were quantified after 30, 60, 90, 120 and 150 days. To analyze Hypothesis I, in relation to survival, a regression analysis was performed following a Weibull survival distribution with censored data (Pinder III et al. 1978; Crawley 2012), where the response variable was time to death of each plant and the explanatory variable was the nutrient addition treatment, with five levels: Control (C), NPK/100, NPK/10, NPK, C+S+M, and C+M. Individuals that remained alive until the end of the experiment (150 days) were recorded and used in the analysis as censored data to inform the final model. The trendlines were not fitted to the dataset in the chart; we prefered to use the raw data, showing how mortality happened over time. To analyze the growth performance in each treatment was evaluated over time by recording rosette diameter (cm) and the total number of leaves every 30 days. Rosette diameter was measured using a digital caliper (accuracy of 0.01 mm) while the number of leaves was counted with the aid of a manual counter. The relative growth rate (RGR) of seedling diameter was calculated folowing Hunt 1982: RGR = (lnD_final - lnD_initial) / (t_final - t_initial) where, D = diameter; and t = time. In order to determine relative growth rate (RGR) of seedling diameter, a linear model was constructed for each plant, and the slope used as a proxy for growth rate in each treatment. In this way, a single value was generated for each replicate (plant), which was used as the response variable in a Generalized Linear Model (GLM)-ANOVA with treatment as the explanatory variable. Models were then subjected to contrast analysis, with levels of treatment without significant differences (P > 0.05) being lumped. Only treatments where survival rates were higher than 50 % were compared in the RGR analysis to achieve a minimum balanced model design.

To analyze Hypothesis II, production of dry biomass (roots and shoots) was recorded for each individual survivor through destructive collection after the last measurement of the experiment. The roots were carefully washed with water on a 2 mm mesh sieve to remove particles of substrate adhered to them. The production of dry biomass was determined by bagging each component collected from the plants. The bagged components were then frozen and taken to the laboratory for oven drying at 70 °C until reaching constant weight (weighed with an analytical balance at 0.001 g precision), according to Chiariello et al. (1989). To evaluate biomass partitioning between plant organs (root and shoot), the following parameters were calculated for each plant using root and shoot biomass values, as carried out by Sainju et al. 2017: Root/ Shoot ratio (RSR) = MR/ MC; where: MR = root dry mass and MC = shoot dry mass . Data were analyzed by GLM-ANOVA with treatment being the explanatory variable. Models were submitted to contrast analysis, with levels of treatment without significant differences (P > 0.05) being grouped. Here we only compared the RSR of treatments where survival rates were higher than 50 %.

All statistical analyses were performed using R Statistical Software (R Development Core Team 2014).

Results

Soil properties

All of the properties related to the chemistry and texture of soil varied among treatments (Tab. 2). The campo rupestre soil (control) was the most acidic and nutritionally poor. We noted that the addition of NPK and manure led to an increase in pH and macro and micronutrients. The pH was about 29 % and 39 % higher for the control (pH = 5.17) than for the NPK (pH=6.72) and 2C:1M (pH= 7.22) treatments, respectively. The average amount of nitrogen for the control was 0.08 dag/kg and was similar to the three treatments of NPK (NPK/100, NPK/10, NPK). However, nitrogen content was 3.6 times higher for 2C:1M (0.29 dag/kg) than for the control (0.08 dag/kg). Mean phosphorus content was 97-fold higher for the NPK treatment (29.3 mg/dm³) and about 473 times higher (142 mg/dm³) for the 2C:1M treatment than for the control (0.3 mg/dm³). The average potassium content was higher for treatments C+S+M (536 mg/dm³) and NPK (367 mg/dm³) than for the control (11 mg/dm³). Mean magnesium content was 0.04 cmolc/dm³ for the control and 14 % higher for the NPK treatment and 54 % for the 2C:1M treatment. Calcium level was highest for the NPK treatment (12.6 cmolc/dm³) and lowest for the control (0.56 cmolc/dm³). The highest values for cation exchange capacity (14.1 cmol/dm³) were for the NPK treatment. Coarse sand (0.19-0.21 kg/kg), fine sand (0.28-0.33kg/kg), silt (0.35-0.36 kg/kg), and clay (0.12-0.15 kg/kg) were similar among control, NPK/100, NPK/10, NPK, and 2C:1M treatments. The treatment C+S+M had the highest values for coarse sand content (0.5 kg/kg) and the lowest values for fine sand (0.19 kg/kg), silt (0.20 kg/kg), and clay (0.09 kg/kg).

Seedling survival experiment

The survival of V. nanuzae seedlings was significantly lower in all treatments relative to the control treatment (native soil; 𝜒²= 235.41, DF=5, P<0.001), while the Tukey post-hoc analysis revealed that only manure (P=0.43) and high levels of NPK addition (P=0.72) did not differ in their influence on plant survival (Fig. 1). In fact, while only one individual in the control treatment died during the experiment, there was an average of 41 % mortality for individuals receiving treatments of low levels of NPK addition (i.e., NPK/100 and NPK/10); 85 % mortality for manure treatments, and 95 % mortality for the NPK treatment. Due to the extremely high mortality rates in CM, CMS and NPK treatments, they were not used in further growth analysis, since survival is the most limiting variable for a plant and such unbalanced design could bias the analysis.

Figure 1
Proportion of surviving plants of V. nanuzae under treatments of different concentrations of macronutrients and substrate composition over time (days). Three treatments had the addition of different concentrations of nitrogen (N), phosphorus (P), and potassium (K): 1) NPK/100 (15 g/m³ of NPK + 20 g/m³ in dolomitic limestone ); 2) NPK/10 (150 g/m³ of NPK+ 200 g/m³ in dolomitic limestone); 3) NPK (1500 g/m³of NPK + 2000 g/m³ in dolomitic limestone). Three treatments had different proportions of substrates (soil native (control - C), sand (S) and manure (M): 1) C; 2) 2C:1M; 3) 3C:2S:1M.

Seedling growth performance

Regarding plant growth, V. nanuzae individuals growing in the control treatment exhibited significantly greater increases in diameter than individuals grown in all other treatments within the 150 days of observation (F_2,213=16.56, P<0.001, Figs. 2 , 3A). Control plants (native soil) exhibited a two-times greater increase in diameter than plants grown in NPK enriched soils (NPK/100 and NPK/10). We noticed that the individuals of the CM treatment languished after 60 days with decreasing diameter.

Figure 2
Vellozia nanuzae L.B.Sm. & Ayensu (Velloziaceae), an endemic species of the campo rupestre ecosystem. Individuals of V. nanuzae that remained alive until the end of the experiment (150 days) selected randomly under different concentrations of macronutrients and substrate composition. Three treatments had the addition of different concentrations of nitrogen (N), phosphorus (P), and potassium (K): 1) NPK/100 (15 g/m³ of NPK + 20 g/m³ in dolomitic limestone ); 2) NPK/10 (150 g/m³ of NPK+ 200 g/m³ in dolomitic limestone); 3) NPK (1500 g/m³of NPK + 2000 g/m³ in dolomitic limestone).Three treatments had different proportions of substrates (soil native (control - C), sand (S) and manure (M): 1) C; 2) 2C:1M; 3) 3C:2S:1M.

Figure 3
A. Average (± standard error) relative growth rate (RGR) of seedling diameter (cm/day) of V. nanuzae under treatments of different concentrations of macronutrients and substrate composition. B. Average (± standard error) root-to-shoot biomass ratios of individuals of V. nanuzae under different concentrations of macronutrients and substrate composition. Three treatments had the addition of different concentration of nitrogen (N), phosphorus (P), and potassium (K): 1) NPK/100 (15 g/m³ of NPK + 20 g/m³ in dolomitic limestone ); 2) NPK/10 (150 g/m³ of NPK+ 200 g/m³ in dolomitic limestone); 3) NPK (1500 g/m³of NPK + 2000 g/m³ in dolomitic limestone).Three treatments had different proportions of substrates (soil native (control - C), sand (S) and manure (M): 1) C; 2) 2C:1M; 3) 3C:2S:1M.

The number of leaves per plant did not differ significantly among treatments (F_2,213=0.9, P=0.40).

The root/shoot biomass ratio was also significantly higher for individuals of V. nanuzae grown in the control treatment (F_2,168=37.19, P<0.001, Fig. 3B) compared to individuals grown in nutritionally enriched soils and soils with different substrates.

Discussion

Seedlings of V. nanuzae grown in soil from their natural habitat had significantly higher survival and growth rates than seedlings grown in soils enriched with nutrients. These results corroborate Hypothesis I - native plants that naturally inhabit nutrient poor soils will have lower survival rates when cultivated in nutrient enriched soils than when grown in soil with low nutritional availability. Campo rupestre soil is nutritionally poor (see review by Schaefer et al. 2016). Even so, the growth and survival of V. nanuzae plants in this natural substrate were greater, indicating that the species is highly tolerant to their restrictive habitat. In fact, individuals that developed in nutritionally enriched substrates exhibited a drastic reduction in survival, reinforcing the hypothesis of tolerance to the nutrient-poor environment (see Tab. 2) of the campo rupestre ecosystem. It is also important to highlight changes in soil Ca+2 content and pH under the different treatments (Tab. 2). High acidity did not seem to be a limiting factor of seedling development and survival for V. nanuzae. The direct negative effects of high pH or a high Ca concentration in the soil, and inability to use the phosphorus pool of the soil, also suggest that this species presents a calcifuge behavior (Tyler & Olsson 1993; Lee 1999). In this case, calcium acts as secondary messenger, which requires the maintenance of a low concentration of this nutrient (Lee 1999). However, more biochemical and physiological experiments are required to better elucidate this behavior. In another nutritional experiment, Negreiros et al. (2014) reported that Baccharis dracunculifolia, while showing increased growth in soils of higher nutritional quality, had lower survival rates. Although the tolerance of some plant species to nutritionally poor soils has already been demonstrated for some harsh ecosystems (e.g., Haridasan 1988; 2000), the mechanisms are mostly unknown and hence in need of further investigation.

The results of the present study are valuable for the conservation and management of species of the campo rupestre ecosystem, considering that innumerable activities have a direct influence on the nutritional quality of its soils. For example, Barbosa et al. (2010) demonstrated that the paving of roads in the campo rupestre ecosystem increased the calcium content and decreased the aluminum content of the soil along road margins. They also showed that this habitat disturbance resulted in a completely different plant community, often with a high number of ruderal and invasive species. The same has been observed as the result of the installation of pasture systems (e.g., Vendramini et al. 2007; Schaefer et al. 2016). Therefore, environmental restoration projects designed for the campo rupestre ecosystem need to take into consideration the nutritional requirements of plants and their temporal and functional dynamics (Fernandes et al. 2016).

In addition to increased survival rates, plants of V. nanuzae grown in soil from the campo rupestre ecosystem (i.e., control plants) also had higher average growth rates than those grown in enriched soils, hence providing support for Hypothesis I. These results indicate that nutritional enrichment of nutritionally poor environments can result in toxic conditions for native species, and thus trigger adverse effects at the ecosystem level (Fernandes & Price 1991). Although other species of the campo rupestre ecosystem had an opposite response with higher growth rates with nutritional enrichment (Calliandra fasciculata, Negreiros et al. 2009; Collaea cipoensis, Negreiros et al. 2009; Baccharis dracunculifolia, Negreiros et al. 2014), these species are widespread (C. fasciculata: Barneby 1998), autochorous (C. cipoensis: Gélvez‐Zúñiga et al. 2018), or pioneer (B. dracunculifolia: Müller et al. 2007) species. Clearly, the number of species studied so far does not allow us yet to propose a general trend for the response of campo rupestre species. Future habitat-wide studies that focus on plant responses to nutrition will aid in the construction of a more robust hypothesis for campo rupestre species. Regardless, we argue that disruption of soil stability and quality lead to strong changes in soil nutritional conditions that harm the survival and development of campo rupestre plant species. Soil conservation measures for areas close to populations of V. nanuzae could represent an important strategy for ensuring its survival and minimizing the chances for extinction given its rarity in the landscape.

The number of leaves has been shown to be a conservative trait (see Cui et al. 2018). The control plants and those subjected to different nutritional treatments had similar numbers of leaves, despite the physicochemical differences among the soils. This plant trait was maintained in soils with nutritional enrichment, while relative growth rates and biomass were reduced.

Plants of V. nanuzae grown under conditions of low soil nutrients and high toxicity had greater root development than those grown with enriched soil, further supporting Hypothesis II. This hypothesis predicted that seedlings of V. nanuzae grown in naturally occurring campo rupestre soil will have a higher proportion of roots compared to aerial biomass. Soil acidity and high aluminum concentration often result in reduced root biomass (Echart & Cavalli-Molina 2001; Caires et al. 2008); this was not, however, observed in V. nanuzae. The higher proportion of biomass in the roots of V. nanuzae can be interpreted as a response or tolerance to low nutrient availability (Lilienfein et al. 2001), water stress (Hoffmann & Franco 2003) and/or frequent fire (Castro & Kauffman 1998), all of which are important environmental filters in the campo rupestre ecosystem (Negreiros et al. 2014; see reviews in Fernandes 2016a). The greater root development of V. nanuzae would favor the nutrient uptake and assimilation necessary for growth in depleted soils (Berbara et al. 2006; Oliveira et al. 2016). Moreover, in nutrient-impoverished soils of campo rupestre, belowground specializations (rhizosheaths, mycorrhizas, and dark septate fungi) can be relevant traits for plant survival (Coutinho et al. 2015; Oliveira et al. 2015; Abrahão et al. 2019). Abrahão et al. (2019) verified that among 27 species of Velloziaceae, most (16 species) possessed rhizosheaths with long root hairs in soils poorest in P, which allows habitat specialization among rock‐dwelling and soil‐dwelling and favors a greater uptake of soil resources. Relationships between plants and soil are also not unidirectional. Increased nutrients can stimulate or favor the antagonistic soil biota and suppress protagonists (Zandt et al. 2019). Although recent advances have been made in understanding plant-microorganism associations in campo rupestre (review in Oki et al. 2016), future studies that elucidate the relationship between nutrition and microorganism diversity and its effects on growth and development of V. nanuzae are needed. In conclusion, Vellozia nanuzaeis tolerant to nutrient poor soil and develops better in soil from its original habitat, which is characterized as acidic and nutrient poor. Our results show that more nutrients are not always better for survival and development of native species that inhabit harsh ecosystems. Furthermore, the findings presented here reinforce the importance of conserving soil in order to conserve the species. There are hundreds of other species of the megadiverse campo rupestre ecosystem (see Fernandes 2016c) that may possess similar or even stronger or more finely tuned tolerance to soil poverty and toxicity, and hence are equally under threat from habitat disturbance and climate change. Given the present lack of governance and plans to lead effective conservation and monitoring of the campo rupestre ecosystem, and the lack of investment in Brazilian biodiversity and conservation (Fernandes et al. 2018), extinctions of plant species, such as V. nanuzae, are imminent.

Acknowledgements

We thank Arthur Moura, Bárbara Silveira, Catarina Freitas, Cecília Loureiro, Letícia Viana, Vanessa Gomes and Walisson Siqueira for assistance during this work; the editor and reviewers for their contributions; Vellozia Reserve for logistical support; and the research funding agencies CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior), CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico), FAPEMIG (Fundação de Amparo à Pesquisa do Estado de Minas Gerais), Vale and Anglo American. RS is thankful to CNPq (305739/2019-0), P&D Aneel-Cemig, PROECOS, GT599.

References

• Abrahão A, Costa PdB, Lambers H, et al 2019. Soil types select for plants with matching nutrient-acquisition and -use traits in hyperdiverse and severely nutrient-impoverished campos rupestres and cerrado in Central Brazil. Journal of Ecology 107: 1302-1316.
• Aerts R, Chapin FS. 2000. The mineral nutrition of wild plants revisited: a re-evaluation of processes and patterns. Advances in Ecological Research 30: 1-67.
• Arendt JD. 1997. Adaptive intrinsic growth rates: an integration across taxa. The Quarterly Review of Biology 72: 149-177.
• Arens K. 1958. O cerrado como vegetação oligotrófico. Boletim da Faculdade de Filosofia, Ciências e Letras, Universidade de São Paulo Botânica 15:57-77.
• Barbosa NPU, Fernandes GW. 2008. A destruição do jardim. Scientific American Brasil 79: 82-84.
• Barbosa NPU, Fernandes GW, Carneiro MA, Júnior LA. 2010. Distribution of non-native invasive species and soil properties in proximity to paved roads and unpaved roads in a quartzitic mountainous grassland of southeastern Brazil (rupestrian fields). Biological Invasions 12: 3745-3755.
• Barneby RC. 1998. Silk Tree, Guanacaste, Monkey's earring. A generic system for the synandrous Mimosaceae of the Americas. Memoirs of the New York Botanical Garden 74: 1-223.
• Berbara RL, Souza FA, Fonseca HMAC. 2006. III-Fungos micorrízicos arbusculares: muito além da nutrição. In:Fernandes MS. (ed.) Nutrição mineral de plantas. Viçosa, Sociedade Brasileira de Ciência do Solo. p.74-85.
• Bloom AJ, Chapin FS, Mooney HA. 1985. Resource limitation in plants - an economic analogy. Annual Review of Ecology and Systematics 16: 363-392.
• Brum M, Teodoro GS, Abrahão A, Oliveira RS. 2017. Coordination of rooting depth and leaf hydraulic traits defines drought-related strategies in the campos rupestres, a tropical montane biodiversity hotspot. Plant and Soil 420: 467-480.
• Caires EF, Garbuio FJ, Churka S, Barth G, Corrêa JCL. 2008. Effects of soil acidity amelioration by surface liming on no-till corn, soybean, and wheat root growth and yield. European Journal of Agronomy 28: 57-64.
• Castro EA, Kauffman JB. 1998. Ecosystem structure in the Brazilian Cerrado: a vegetation gradient of aboveground biomass, root mass and consumption by fire. Journal of Tropical Ecology 14: 263-283.
• Chapin III FS. 1980. The mineral nutrition of wild plants. Annual Review of Ecology and Systematics 11: 233-260.
• Chapin III FS, Autumn K, Pugnaire F. 1993. Evolution of suites of traits in response to environmental stress. The American Naturalist 142: 78-92.
• Chiariello NR, Mooney HA, Williams K. 1989. Growth, carbon allocation and cost of plant tissues. In: Pearcy RW, Ehleringer JR, Mooney HA, Rundel PW. (eds.) Plant Physiological Ecology. Dordrecht, Springer. p. 327-365.
• Coley PD, Bryant JP, Chapin FS. 1985. Resource availability and plant antiherbivore defense. Science 230: 895-899.
• Conceição AA, Rapini A, Carmo FF, et al 2016. Rupestrian grassland vegetation, diversity, and origin. In: Fernandes GW . (ed.) Ecology and Conservation of Mountaintop Grasslands in Brazil. Switzerland, Springer. p.105-127.
• Coutinho ES, Fernandes GW, Berbara RLL, Valério HM, Goto BT. 2015. Variation of arbuscular mycorrhizal fungal communities along an altitudinal gradient in rupestrian grasslands in Brazil. Mycorrhiza 25: 627-638.
• Cui H, Töpper JP, Yang Y, Vandvik V, Wang G. 2018. Plastic population effects and conservative leaf traits in a reciprocal transplant experiment simulating climate warming in the Himalayas. Frontiers in Plant Science 9: 1069. doi: 10.3389/fpls.2018.01069
  » https://doi.org/10.3389/fpls.2018.01069
• Crawley MJ. 2012. Analysis of variance. The R Book. 2nd. edn. Chichester, UK, Wiley.
• Dayrell RLC, Garcia QS, Negreiros D, Baskin CC, Baskin JM, Silveira FAO. 2016. Phylogeny strongly drives seed dormancy and quality in a climatically buffered hotspot for plant endemism. Annals of Botany 119: 267-277.
• Donagema GK, de Campos DB, Calderano SB, Teixeira WG, Viana JM. 2011. Manual de métodos de análise de solo. 2. ed. rev. Rio de Janeiro, Embrapa Solos-Documentos.
• Echart CL, Cavalli-Molina S. 2001. Aluminum phytotoxicity: effects, tolerance mechanisms and its genetic control. Ciência Rural 31: 531-541.
• Epstein E, Bloom AJ. 2004. Mineral nutrition of plants: principles and perspectives. 2nd. edn. Massachussets, Sinauer.
• Fernandes GW. 2016a. Ecology and conservation of mountaintop grasslands in Brazil . 1st. edn. Switzerland, Springer .
• Fernandes GW. 2016b. The shady future of the rupestrian grassland: major threats to conservation and challenges in the Anthropocene. In: Fernandes GW. (ed.) Ecology and conservation of mountaintop g rasslands in Brazil . 1st. edn. Switzerland, Springer . p. 545-559.
• Fernandes GW. 2016c. The megadiverse rupestrian grassland. In: Fernandes GW (ed.) Ecology and Conservation of Mountaintop Grasslands in Brazil . Switzerland, Springer . p. 3-14.
• Fernandes GW, Barbosa NPU, Alberton B, et al 2018. The deadly route to collapse and the uncertain fate of Brazilian rupestrian grasslands. Biodiversity and Conservation 27: 2587-2603.
• Fernandes GW, Barbosa NPU, Negreiros D, Paglia AP. 2014. Challenges for the conservation of vanishing megadiverse rupestrian grasslands. Perspectives in Ecology and Conservation 12:162-165.
• Fernandes GW, Price PW. 1991. Comparisons of tropical and temperate galling species richness: the roles of environmental harshness and plant nutrient status. In: Price PW, Lewinsohn TM, Fernandes GW, Benson WW. (eds.) Plant-animal interactions: evolutionary ecology in tropical and temperate regions. New York, Wiley & Sons. p. 91-115.
• Fernandes GW, Rodarte LH, Negreiros D, Franco AC. 2007. Aspectos nutricionais em Baccharis concinna (Asteraceae), espécie endêmica e ameaçada da Serra do Espinhaço, Brasil. Lundiana 8: 83-88.
• Fernandes GW, Toma TSP, Angrisano P, Overbeck G. 2016. Challenges in the restoration of quartzitic and ironstone rupestrian grasslands. In: Fernandes GW . (ed.) Ecology and Conservation of Mountaintop Grasslands in Brazil . Switzerland, Springer . p. 449-477.
• Gélvez‐Zúñiga I, Teixido AL, Neves AC, Fernandes GW. 2018. Floral antagonists counteract pollinator‐mediated selection on attractiveness traits in the hummingbird‐pollinated Collaea cipoensis (Fabaceae). Biotropica 50: 797-804.
• Goodland RJ, Ferri MG. 1979. Ecologia do Cerrado. São Paulo, Editora Universidade de São Paulo.
• Gomes VM, Negreiros D, Carvalho V, Fernandes GW. 2015. Growth and performance of rupestrian grasslands native species in quartzitic degraded areas. Neotropical Biology and Conservation 3: 159-168.
• Gomes VM, Negreiros D, Fernandes GW, Pires AC, Silva AC, Stradic SL. 2018. Long‐term monitoring of shrub species translocation in degraded Neotropical mountain grassland. Restoration Ecology 26: 91-96.
• Grime JP, Campbell BD. 1991. Growth rate, habitat productivity, and plant strategy as predictors of stress response. In: Mooney HA, Winner WE, Pell EJ. (eds.) Response of plants to multiple stresses. San Diego, Academic Press. p. 143-159.
• Haridasan M. 1988. Performance of Miconia albicans (Sw.) Triana, an aluminium accumulating species in acidic and calcareous soils. Communications in Soil Science and Plant Analysis 19:1091-1103.
• Haridasan M. 2000. Nutrição mineral de plantas nativas do cerrado. Revista Brasileira Fisiologia Vegetal 12: 54-64.
• Hilário RR, Castro SAB, Ker FTO, Fernandes GW. 2011. Unexpected effects of pigeon peas (Cajanus cajan) in the restoration of rupestrian fields. Planta Daninha 29: 717-723.
• Hillel D. 2007. Soil in the environment: crucible of terrestrial life. Oxford, Elsevier, Academic Press.
• Hoffmann WA, Franco AC. 2003 Comparative growth analysis of tropical savanna and forest trees using phylogenetically independent contrasts. Journal of Ecology 91: 475-484.
• Hunt R. 1982. Plant growth curves: the functional approach to plant growth analysis. London, Edward Arnold.
• Hunt JR, Celestina C, Kirkegaard JA 2020. The realities of climate change, conservation agriculture and soil carbon sequestration. Global Change Biology 26: 3188-3189.
• Klein C, Klein VA. 2015. Strategies to improve the retention and availability of soil water. Revista Eletrônica em Gestão, Educação e Tecnologia Ambiental 19: 21-29.
• Lee JA. 1999. The calcicole-calcifuge problem revisited. Advances in Botanical Research 29: 2-30.
• Lilienfein J, Wilcke W, Zimmermann R, Gerstberger P, Araujo GM, Zech W. 2001. Nutrient storage in soil and biomass of native Brazilian Cerrado. Journal of Plant Nutrition and Soil Science 164: 487-495.
• Macedo AC, Kageyama PY, Costa LGS. 1993. Produção de Mudas em viveiros florestais: espécies nativas. São Paulo, Fundação Florestal.
• Madeira JA, Fernandes GW. 1999. Reproductive phenology of sympatric taxa of Chamaecrista (Leguminosae) in Serra do Cipó, Brazil. Journal of Tropical Ecology 15: 463-479.
• Matias L, Castro J, Zamora R. 2011. Soil‐nutrient availability under a global‐change scenario in a Mediterranean mountain ecosystem. Global Change Biology 17:1646-1657.
• Maurya R, Yadav PP. 2005. Furano flavonoids: an overview. Natural Product Reports 22: 400-424.
• Mello-Silva R. 2015. Velloziaceae in Lista de Espécies da Flora do Brasil. In: Jardim Botânico do Rio de Janeiro. http://floradobrasil.jbrj.gov.br/jabot/floradobrasil/FB33116 15 Sept. 2017.
  » http://floradobrasil.jbrj.gov.br/jabot/floradobrasil/FB33116
• Melnikova I, Sasai T. 2020. Effects of anthropogenic activity on global terrestrial gross primary production. Journal of Geophysical Research: Biogeosciences 125: e2019JG005403. doi: 10.1029/2019JG005403
  » https://doi.org/10.1029/2019JG005403
• Miller HG, Miller JD. 1988. Response to heavy nitrogen applications in fertilizer experiments in British forests. Environmental Pollution 54: 219-231.
• Mooney HA, Winner WE. 1991. Partitioning response of plants to stress. In: Moone HA, Winner WE, Pell EJ. (eds.) Response of plants to multiple stresses. San Diego, Academic Press . p.129-141.
• Moreira AG, Klink CA. 2000. Biomass allocation and growth of seedlings of ten tree species from the Brazilian Savannas. Ecotropicos 13: 43-51.
• Morgan JB, Connolly EL. 2013. Plant-Soil Interactions: Nutrient Uptake. Nature Education Knowledge 4: 2. https://www.nature.com/scitable/knowledge/library/plant-soil-interactions-nutrient-uptake-105289112/#:~:text=Changes%20in%20root%20architecture%2C%20induction,face%20of%20changing%20soil%20environments 13 Mar. 2020.
  » https://www.nature.com/scitable/knowledge/library/plant-soil-interactions-nutrient-uptake-105289112/#:~:text=Changes%20in%20root%20architecture%2C%20induction,face%20of%20changing%20soil%20environments
• Müller SC, Overbeck GE, Pfadenhauer J, Pillar VD. 2007. Plant functional types of woody species related to fire disturbance in forest-grassland ecotones. Plant Ecology 189: 1-14
• Negreiros D, Esteves D, Fernandes GW, et al 2014. Growth-survival tradeoff in the widespread tropical shrub Baccharis dracunculifolia (Asteraceae) in response to a nutrient gradiente. Tropical Ecology 55: 167-176.
• Negreiros D, Fernandes GW, Efremova AA, Stradic SL, Neves ACO. 2016. Growth-survival trade-off in shrub saplings from Neotropical mountain grasslands. South African Journal of Botany 106: 17-22.
• Negreiros D, Fernandes GW, Silveira FAO, Chalub C. 2009. Seedling growth and biomass allocation of endemic and threatened shrubs of Rupestrian fields. Acta Oecologica 35: 301-310.
• Oki Y, Goto BT, Jobim K, et al 2016. Arbuscular mycorrhiza and endophytic fungi in ruspestrian grasslands. In: Fernandes GW . (ed.) Ecology and Conservation of Mountaintop Grasslands in Brazil . Switzerland, Springer . p. 157-179.
• Oliveira RS, Abrahão A, Pereira C, et al 2016. Ecophysiology of Campos Rupestres plants. In: Fernandes GW . (ed.) Ecology and Conservation of Mountaintop Grasslands in Brazil . Switzerland, Springer . p. 227-272.
• Oliveira RS, Galvão HC, Campos MC, Eller CB, Pearse SJ, Lambers H. 2015. Mineral nutrition of Campos Rupestres plant species on contrasting nutrient‐impoverished soil types. New Phytologist 205: 1183-1194.
• Pausas JG, Lamont BB, Paula S, Appezzato-da-Glória B, Fidelis A. 2018. Unearthing belowground bud banks in fire-prone ecosystems. New Phytologist 217: 1435-1448.
• Pena JCC, Goulart F, Fernandes GW, et al 2017. Impacts of mining activities on the potential geographic distribution of eastern Brazil mountaintop endemic species. Perspectives in Ecology and Conservation 15: 172-178.
• Pinder III JE, Wiener JG, Smith MH. 1978. The Weibull distribution: a new method of summarizing survivorship data. Ecology 59: 175-179.
• Pinto AC, Silva RS, Valente LMM. 1988. Diterpenes from Vellozia nanuzae Phytochemistry 27: 3909-3911.
• R Development Core Team. 2014. R: A language and environment for statistical computing. Vienna, R Foundation for Statistical Computing. https://www.R-project.org
  » https://www.R-project.org
• Rawls WJ, Gish TJ, Brakensiek DL. 1991. Estimating soil water retention from soil physical properties and characteristics. In: Stewart BA. (ed.) Advances in Soil Science. New York, Springer. p. 213-234.
• Razaq M, Zhang P, Shen HL. 2017. Influence of nitrogen and phosphorous on the growth and root morphology of Acer mono. PLOS ONE 12: e0171321. doi: 10.1371/journal.pone.0171321
  » https://doi.org/10.1371/journal.pone.0171321
• Sainju UM, Allen BL, Lenssen AW, Ghimire RP. 2017. Root biomass, root/shoot ratio, and soil water content under perennial grasses with different nitrogen rates. Field Crops Research 210: 183-191.
• Schaefer CE, Cândido HG, Corrêa GR, Nunes JA, Arruda DM. 2016. Soils associated with rupestrian grasslands. In: Fernandes GW. (ed.) Ecology and Conservation of Mountaintop Grasslands in Brazil . Switzerland, Springer . p. 55-69.
• Silveira FA, Negreiros D, Barbosa NP, et al2016. Ecology and evolution of plant diversity in the endangered campo rupestre: a neglected conservation priority. Plant and Soil 403: 129-152.
• Stradic S, Buisson E, Fernandes GW. 2014. Restoration of Neotropical grasslands degraded by quarrying using hay transfer. Applied Vegetation Science 17:482-492.
• Tyler G, Olsson PA. 1993. The calcifuge behaviour of Visearía vulgaris Journal of Vegetation Science 4: 29-36.
• Uchida R. 2000. Essential nutrients for plant growth: nutrient functions and deficiency symptoms. In: Silva JA, Uchida R. (eds.) Plant Nutrient Management in Hawaii’s soils. Approaches for Tropical and Subtropical Agriculture. Honolu, University of Hawai. p. 31-55.
• Vendramini JMB, Silveira MLA, Dubeux Jr JCB, Sollenberger LE. 2007. Environmental impacts and nutrient recycling on pastures grazed by cattle. Revista Brasileira de Zootecnia 36: 139-149.
• Warembourg FR, Estelrich HD. 2001. Plant phenology and soil fertility effects on below-ground carbon allocation for an annual (Bromus madritensis) and a perennial (Bromus erectus) grass species. Soil Biology and Biochemistry 33: 1291-1303.
• Wilson JB.1988. A review of evidence on the control of shoot: root ratio, in relation to models. Annals of Botany 61: 433-449.
• Wijesinghe DK, John EA, Hutchings MJ. 2005. Does pattern of soil resource heterogeneity determine plant community structure? An experimental investigation. Journal of Ecology 93: 99-112.
• Xu R, Tian H, Pan S, et al 2019. Global ammonia emissions from synthetic nitrogen fertilizer applications in agricultural systems: Empirical and process‐based estimates and uncertainty. Global Change Biology 25: 314-326.
• Zandt D, Brink AVD, Kroon H, Visser EJW. 2019. Plant-soil feedback is shut down when nutrients come to town. Plant and Soil 439: 541-551.

Publication Dates

• Publication in this collection
  02 Oct 2020
• Date of issue
  Jul-Sep 2020

History

• Received
  11 Feb 2020
• Accepted
  28 May 2020