Environmental Influence on the Leaf Morphoanatomical Characteristics of Myrcia splendens (Sw.) DC. (Myrtaceae)1

Myrcia splendens (Sw.) DC. is a native and endemic species of Brazil. Plants can have ecological plasticity due to environmental influences and studies on the ecological anatomy of leaves in the genus Myrcia DC. are scarce. In this sense, the present study aimed to verify the effect of seasonality and luminosity on the leaf morphoanatomical characteristics of Myrcia splendens (Sw.) DC., in order to contribute with information about the species responses to the natural abiotic factors of its occurrence. The botanical material for the study was collected in the São Gonçalo do Amarante Botanic Garden, Ceará, in the rainy and dry seasons, fully expanded leaves exposed to the sun and shade, which were subsequently subjected to laboratory procedures to obtain the paradermal and cross sections of the leaves, in order to verify how Myrcia splendens (Sw.) DC. responds to seasonal variations in the availability of water and light in a coastal environment. As a result, most leaf structures of Myrcia splendens (Sw.) DC. presented greater thicknesses, areas or densities when in the dry season (water deficit) and submitted to intense sunlight, that is, in more stressful environmental conditions. Therefore, it’s concluded that the changes in the anatomy of Myrcia splendens (Sw.) DC. demonstrated the acclimatization of the leaves in response to abiotic factors (water and light), thus contributing to the survival of the species in the Vegetation Complex of the Coastal Zone of Ceará.

Therefore, plants may undergo physiological, morphological and anatomical changes as adaptive strategies to the environment changes (Devi et al., 2017;Melo Júnior & Boeger, 2017). Leaf is the plant organ commonly used in ecological studies which may easily respond to environmental changes (Wyka, Robakowski, & Zytkowiak, 2007). Leaf tissues such as chlorenchyma, epidermis, parenchyma as well as vascular bundles may undergo phenotypic plasticity and register the influence of abiotic factor on plants (Dardengo, Rossi, Silva, Pessoa, & Silva, 2017;Devi et al., 2017;Lemos et al., 2018).
Myrcia splendens (Sw.) DC. is often found at sandy coastal plains (Castro, Moro, & Menezes, 2012;Moro, Macedo, Moura-Fé, Castro, & Costa, 2015). The sandy coastal plains are areas susceptible to great water availability, salinity, strong winds and exposure to high incident light (Rosado & Mattos, 2007;Moro et al., 2015;Melo Júnior & Boeger, 2017). In addition, sandy coastal plains environments presents deep and leachate soils and with drought periods up to six months per year (Castro et al. 2012). Such environmental stresses reflect on the occupation of these areas by plants as well as on plant morphoanatomical characteristics (Rosado & Mattos, 2007;Kuster, Meira, & Azevedo, 2018) that change in response to the abiotic factors so plants may survive under such harsh conditions (Devi et al., 2017;Melo Júnior & Boeger, 2017).
Despite Myrcia splendens (Sw.) DC. being native and endemic to Brazil (BFG, 2018), little is known about the leaf morphoanotomic changes of this species for its survival under the various conditions imposed by the environment, especially in coastal regions of the Northeast of the country. Thus, the hypothesis of this research was that there is a correlation between the anatomy of Myrcia splendens (Sw.) DC. leaves and the environmental conditions to which they are exposed.
Therefore, evergreen plants such as Myrcia splendens (Sw.) DC. inhabiting sandy coastal plains areas make up a good model to understand the possible anatomical responses of plants to environmental stresses. In this sense, the present study aimed to verify the effect of seasonality and luminosity on the leaf morphoanatomical characteristics of Myrcia splendens (Sw.) DC., in order to contribute with information about the species responses to the natural abiotic factors of its occurrence.

Study Area
The material was collected in the São Gonçalo do Amarante Botanic Garden (3°34'07.0"S and 38°53'12.8"W) located in the municipality of São Gonçalo do Amarante, located in the state of Ceará, Brazil (Figure 1). This area is a Vegetation Complex of the Coastal Zone ( Figure 2) (Castro et al., 2012) with average annual precipitation in 2018 of 1772.9 mm, lowest rainfall rate between the months of September to November (3.3, 5.4 and 5.0 mm, respectively) and highest rainfall rate between February to May (308.7,246.7,265.9 and 237.4 mm,respectively) (Table 1)
Fully developed leaves from Myrcia splendens (Sw.) DC. specimens from the 3 rd to 5 th node exposed to sun or shade were collected in March/2018 (wet season -average temperature of 27.5 ºC and average relative humidity of 79%) and October/2018 (dry season -average temperature of 27.9 ºC and average relative humidity of 70%) (Table 1) (UFC, 2018).

Light Microscopy Study
The botanical material collected was fixed in a solution containing 4% paraformaldehyde and 1% glutaraldehyde in 0.2 M phosphate buffer at pH 7.2 (Karnovsky, 1965). For leaf epidermal characterization, fragments from the middle third of leaves fixed in Karnovsky were dissociated according to Jeffrey's Method (Johansen, 1940). The dissociated fragments from the abaxial and adaxial side of the epidermis were stained with 1% Astra blue and 0.05% safranin for 12 hours and mounted on glass varnish.
For the anatomical characterization of leaves in cross sections, fragments from the middle third of leaves fixed in Karnovsky were dehydrated in increasing ethyl alcohol series up to 95% ethanol and embedded in methacrylate resin according to the recommendations (Leica, Heidelberg, Germany). Cross sections at 3-5 μm were made in an automatic rotary microtome (Leica RM 2065, Leica Instruments GmbH, Nussloch, Germany) with glass blades. Sections were stained in a solution composed of 1% Astra blue and 0.05% safranin for 12 hours. The excess of staining was removed by dipping the slides in distilled water for about 15 min. Permanent slides were dried at 37 ºC and mounted with glass varnish. Observations and photographs were taken using a digital camera Olympus UC 30 (Hamburg, Germany) equipped with light microscope Olympus BX 41TF (Tokyo, Japan).

Analysis of the Effect of Seasonality and Luminosity on Leaves
Quantitative analysis of the seasonal variation in leaf anatomy was performed by comparing the leaves fully expanded collected during the wet and dry seasons of plants exposed to sun and shade. In paradermal sections of leaf blade the parameters: trichome density (number of trichomes/area), stomatal density (number of stomata/area), frequency (number of stomata/number of epidermal cells), index [number of stomata/(number of epidermal cells + number of stomata) x 100] and area (μm 2 ) were quantitatively evaluated. The area of density of trichomes and stomata were 0.0024 cm 2 and 31.3 cm 2 , respectively. For cross sections were evaluated the leaf blade region, mesophyll, palisade and spongy parenchyma, epidermis (with cuticle) of the adaxial and abaxial face thickness. In the midrib were measured the midrib thickness (μm), vascular bundle, xylem, phloem and fibers area (μm 2 ). The structures were measured using the Image J program (Abràmoff, Magalhães, & Ram, 2004).

Statistical Analysis
The experimental design was completely randomized, with 2 (wet and dry) x 2 (sun and shade) factorial and 15 replicates. Results were submitted to normality test, followed by analysis of variance observing the significance by the F test and when significant, the Tukey test was performed at 5% probability level using the Statistical Analysis System V. 1.0 (ESTAT) software.

Morphoanatomical Characteristics
Myrcia splendens (Sw.) DC. leaf blades observed in frontal view presented markedly sinuous anticlinal cell walls on both leaf sides (Figure 4a-f). Simple non-secretory trichomes with a pointy apex and paracytic stomata (hypostomatic leaf) were observed only on the abaxial side (Figure 4a-e). On the adaxial leaf side, among the ordinary epidermal cells, two cells (cap cells) that fit together similar to two jigsaw puzzle pieces were observed. Such cells were always covering a subepidermal secretory cavity ( Figure 4f).
In cross sections of the leaf blade, it was possible to observe a single-layered epidermis on both leaf sides in which the adaxial epidermis is thicker than the abaxial epidermis (thickened external periclinal wall). The mesophyll is composed of both spongy and palisade parenchyma (dorsiventral mesophyll) and palisade parenchyma presents a single-layered while the spongy parenchyma was composed by 6-8 cell layers (Figure 4g-j). In the mesophyll there are predominantly secretory cavities in the palisade parenchyma (adaxial face) and collateral vascular bundles with two caps (upper and lower) of sclerenchymatic fibers (Figure 4h-j).
At the midrib region, the leaf blade was convex on the abaxial side and slight concave on the adaxial side (Figure 4k-n). On the adaxial side the palisade parenchyma was interrupted by collenchyma ( Figure 4k-n). Phloem was placed on both sides of xylem (bicollateral vascular bundles) and the whole vascular bundles were surrounded by fibers. At abaxial side, the spongy parenchyma was interrupted by of collenchyma and parenchyma (Figure 4k-n). On some occasions secretory cavities were observed in the mesophyll near the vascular bundles. Simple trichomes were also observed on adaxial side of the midrib region above the collenchyma ( Figure  4n).

Environmental Influence
Trichome density showed no statistical difference between the seasons (wet season density 3986.1111; dry season density 4138.8889) while for light intensity there was a greater value of trichome density for shade leaves (4375.0) than in leaves subjected to the sun (3750.0) ( Table 2). There was interaction between the two treatments evaluated (season and lightness) and a higher density of shade leaves was observed in the wet season (4805.5556) ( Table 3).
All stomata criteria (density, frequency, index and area) showed higher values in the dry season (Table 2). In these criteria, the values obtained in dry season were 0.2641, 0.4944, 33.6784%, 455.5458 μm 2 , respectively (Table 2). In the wet season, the lower values of 0.2076, 0.3800, 28.2554% and 390.3109 μm 2 , respectively ( Table 2). In terms of luminosity, all criteria were higher for sun leaves, except for the stomatal area ( Table 2). The stomatal area in shade leaves was 425.6951 μm 2 , in contrast the sun leaves the area was 420.1617 μm 2 (Table 2). Stomatal density, frequency and index showed higher values in sun leaves 0.2491, 0.4838 and 32.8798%, respectively, than shade leaves (0.2225, 0.3906 and 29.054%, respectively) ( Table 2). Stomatal area demonstrated an interaction between seasonality and lightness (Table 3). Shade leaves from dry season showed greater area (480.6323 μm 2 ) when compared to other treatments ( Table  3). The other stoma criteria had no interaction between season and lightness ( Table 2).
In the cross sections, the leaf blade, spongy parenchyma and adaxial epidermis thickness obtained high values in sun lightness and dry season leaves (Table 2) with interaction between the treatments only in leaf blade and adaxial epidermis thickness (   (Table 2). Regarding the season versus lightness interaction the leaf blade was thicker in leaves exposed to the sun in the dry season (528.81 μm) when compared to the other treatments (Table 3). The same was observed in the adaxial epidermis thickness parameter (16.9705 μm) ( Table 3).
The mesophyll thickness showed no statistical difference between the leaves of the dry (221.4006 μm) and wet season (217.7305 μm) ( Table 4). As for lightness, the sun leaves presented thicker (222.5936 μm) leaves when compared to shade leaves (216.5375 μm) ( Table  2). As regards interaction between the season and lightness, the mesophyll it was thicker in sun leaves in the wet season (225.3962 μm) and shade leaves in the dry season (223.0102 μm) (Table 3). Palisade parenchyma thickness showed high values in wet season (102.8377 μm), while the leaves in sun (97.6026 μm) and shade (98.4499 μm) lightness didn't differ from each other ( Table  2). In this characteristic was interaction between season and lightness (Table 3) in which the palisade parenchyma was thicker in leaves of sun (104.3863 μm) and shade (101.2891 μm) from the wet season (Table 3). As for the abaxial epidermis thickness, there was no statistical difference between treatments (Table 2), however, there was an interaction between season and lightness, being higher for dry season in the sun and wet season in the shade (Table 3).
All parameters evaluated in the cross sections in the midrib of Myrcia splendens (Sw.) DC. leaves presented higher values in dry season (Table 2). While in the lightness the results varied. The midrib thickness, vascular bundle and phloem area presented higher values in sun lightness leaves (Table 2). While the xylem presented greater area in shade leaves (28226.7037 μm 2 ) ( Table 2) and fibers area presented no statistical difference in areas of sun (25458.6874 μm 2 ) and shade (24863.9519 μm 2 ) lightness (Table 2). Myrcia splendens (Sw.) DC. leaves presented larger areas of phloem in dry season (33732.1589 μm 2 ) and sun lightness (29617.0597 μm 2 ) when compared to wet season (24056.8982 μm 2 ) and shade lightness leaves (28171.9974 μm 2 ), respectively ( Table 2). As for midrib thickness and vascular bundle area the leaves showed higher values in dry season (496.5936 μm and 92323.7985 μm 2 ) and sun lightness (507.7802 μm and 81969.6525 μm 2 ) ( Table 2). The midrib thickness and vascular bundle area of the midrib was the parameters in the midrib that presented  ) 24069.5503 b 26253.0890 a 25458.6874 a 24863.9519 a Means followed by the same letter in season or lightness don't differ statistically by the Tukey's test at 5% probability level. Asterisk indicates interaction between season and lightness, more details in Table 3. interaction between lightness and season ( Table  3). The biggest midrib thickness and vascular bundle area was in sun leaves in the dry season (528.8154 μm and 98108.0070 μm 2 , respectively) ( Table 3).
In the midrib, the phloem is present in the upper and lower part of the xylem and these vascular tissues are surrounded by fibers (Pacheco-Silva & Donato, 2016) and sometimes are the presence of secretory cavities (Cardoso et al., 2009;Lemos et al., 2019). The M. splendens presents collenchyma in the midrib, it interrupts the palisade parenchyma on the adaxial face and is composed of several layers like other species of Myrtaceae (Lemos et al., 2018;Lemos et al., 2019). The midrib is convex in abaxial surface (Donato & Morretes, 2007;Alves et al., 2008;Donato & Morretes, 2009;Donato & Morretes, 2011).

Environmental Influence
Species adapted to specific climatic conditions must have adapted some structural adaptations (Nawazish et al., 2006). In Myrcia splendens (Sw.) DC., the trichome density no showed statistical difference between seasons, but higher density in shade leaves, as this finding is not consistent with the literature. Trichome density is generally higher in plants from sunexposed areas and in plants without water availability (not irrigated) (Pérez-Estrada, Cano-Santana, & Oyama, 2000) or water deficit (Bañon et al., 2004). In addition, leaves with trichomes showed a lower rate of transpiration than leaves without trichomes (Pérez-Estrada et al., 2000). Myrcia splendens (Sw.) DC. presented trichomes on abaxial face, despite the fact that it didn't have a response of higher density values in sun leaves is a way of adapting environmental stress conditions.
Myrcia splendens (Sw.) DC. presents stomata on abaxial face. This characteristic is frequent in plants adapted to xeric environments, as it reduces respiration, allowing the plant to save water (Esposito-Polesi et al., 2011). Stomatal density can be affected by several abiotic environmental factors. In this studied species the leaves of sun and leaves in the dry season presented greater stomatal density. Sun leaves with higher stomatal densities are frequent (Klich, 2000;Morais et al., 2004;Donato & Morretes, 2009).
In Myrcia splendens (Sw.) DC. the stomatal index and density showed higher values in dry season leaves and sun lightness. Different results were found in a study with leaves that, when subjected to lower temperatures presented a higher stomatal index than plants subjected to high temperatures (Vile et al., 2012). Regarding water availability, stomatal index tended to increase in leaves with water deficit (Vile et al., 2012).
Larger stomatal area may be related to greater availability of CO 2 (James & Bell, 2001). Stomatal area in Myrcia splendens (Sw.) DC. there was interaction between season and lightness in which shade leaves in the dry season showed larger stomatal areas. This no corroborates with Batista et al. (2010) study. Research has found that when water availability decreases, there is a reduction in stoma size. This occurs so that the plant loses less water in the transpiration process (Batista et al., 2010). In addition, intense light is related to a high rate of photosynthesis and leaves transpiration. High temperatures are also related to increased sweating transpiration (Stewart et al., 2017).
The leaves of Myrcia splendens (Sw.) DC. in the shade during the dry season, as well as the leaves of sun in the wet season have the greatest mesophyll thickness. Therefore, what induces the thickening of the mesophyll is stress, whether due to excess light or lack of water. Leaves more exposed to luminosity have greater mesophyll thickness and corroborates with the research carried out with leaves of Chenopodium album L. (Oguchi et al., 2003), Alocasia macrorhiza Schott (Sims & Pearcy, 1992) and Eugenia florida DC. (Donato & Morretes, 2009). In this regard, other studies also have observed that the mesophyll of leaves exposed to high light intensity was thicker than leaves exposed to low light (Klich, 2000;Hanba, Kogami, & Terashima, 2002;Oguchi et al., 2003;Dardengo et al., 2017). Larger areas of the mesophyll cells is a way to use water efficiently with a greater capture of CO 2 than loss of water during transpiration (Smith et al., 1997), which corroborates with the present research when leaf was in the shade, but there was no water.
Myrcia splendens (Sw.) DC. showed greater thickness of the adaxial epidermis in sun lightness leaves and dry season. In this regard, in Myrtaceae species the sun leaves obtained greater thickness in sun lightness leaves and wet season (Lemos et al., 2018;Lemos et al., 2019). Increasing the thickness of the adaxial epidermis can be a protection against high luminosity on the leaves (Chazdon & Kaufmann, 1993). The result in Myrcia splendens (Sw.) DC. shows that there is an interaction of seasonality and lightness on the leaf, indicating a response to the plants most exposed to environmental variations (Mantuano et al., 2006). Greater thickness of the adaxial epidermis is considered a xeromorphic characteristic, which is a response to avoid water loss when plants are subjected to high temperatures and high sun incidence (Fahn & Cutler, 1992).
As for the abaxial epidermis, there was no difference between seasons and luminosity. The study with leaf anatomy of Elaeagnus angustifolia L. corroborates in this research, because there is no difference in abaxial epidermis in different luminosities (Klich, 2000). On the other hand, there was an interaction between season and lightness, being greater for dry season in the sun and rainy season in the shade, which fully corroborates with Myrcia guianensis (Aubl.) DC. (Lemos et al., 2020), in part with Eugenia luschnathiana (O.Berg) Klotzsch ex B.D.Jacks., since the epidermis is more thick in leaves in the sun for both seasons (Lemos et al., 2018) and differs from Eugenia punicifolia (Kunth) DC. in that the interaction was greater in the wet season leaves and sun lightness (Lemos et al., 2019).
In arid environments, generally the adaxial epidermis is thicker than the abaxial epidermis (Lemos et al., 2018;Lemos et al., 2019;Lemos et al. 2020), in order to avoid water loss, which was found in the present study.
The spongy parenchyma thicker in sun leaves has been verified in other works (Larcher & Boeger, 2009;Dardengo et al., 2017;Lemos et al., 2018) and corroborates with Myrcia splendens (Sw.) DC. leaves in which it presented thicker spongy parenchyma in plants submitted to light and in the dry season. On the other hand, this result is not corroborated by DeLucia, Nelson, Vogelmann e Smith (1996), who found that the spongy parenchyma is thicker in shadow leaves, as they need to increase the luminous absorption responsible for photosynthetic activity. As for the thickness of the palisade parenchyma greater in wet season leaves, being corroborated by Lemos et al. (2019). However, sun leaves are more exposed to light (high luminous intensity) and generally have a thicker palisade parenchyma (Sims & Pearcy, 1992;Klich, 2000;Larcher & Boeger, 2009). Thus, the more uniform distribution of light on the leaf by the palisade parenchyma is important, since the thickness of this tissue in sun leaves could be a problem in the distribution of light (Vogelmann, 1993).
Sun leaves with thicker midrib than shade leaves are cited in the literature (Dardengo et al., 2017) and corroborates the result recorded in Myrcia splendens (Sw.) DC. More developed midrib is a response to water deficit (Batista et al., 2010). On the other hand, another study found that decreasing midrib thickness is an drought adaptation (Olsen et al., 2013). Vascular bundle area in Myrcia splendens (Sw.) DC. is higher in sun and in dry season leaves corroborates with Eugenia luschnathiana (O.Berg) Klotzsch ex B.D.Jacks. study (Lemos et al., 2018).
Xylem and phloem area were higher values in dry season Myrcia splendens (Sw.) DC. leaves, as for luminosity xylem presents larger areas on shade leaves and phloem higher areas in sun leaves. In Arabidopsis thaliana Schur leaves in environments controlled it was found that high light intensity and temperature favored the increase in the area of xylem, as the high solar intensity increased areas of the phloem (Stewart et al., 2017). These data corroborate almost entirely with what was observed in Myrcia splendens (Sw.) DC., indicating that the plasticity of acclimatization of leaf vascular bundles in response to the environment conditions (Stewart et al., 2017).
Fibers are strong tissues sclerenchyma that surround vascular tissue (Yu, Liu, Shen, Jiang, & Huang, 2015). It is known that fibers are a possible form of adaptation against water scarcity, which corroborates with the present work, since a larger fibers area of the midrib was found in the dry season.
In Myrcia splendens (Sw.) DC the only parameters of the midrib that there was an interaction of the treatments were the midrib thickness and vascular bundle area of the midrib. This indicates that these parameters responded to the abiotic factors tested (season and luninosity).
Thus, this study evaluated 16 structures of which nine presented interaction of luminosity and season. This demonstrates the influence of abiotic factors on Myrcia splendens (Sw.) DC. leaf plasticity. Under natural environmental conditions, water availability and temperature variation can influence independently or together the characteristics and distribution of plant species (Vile et al., 2012). The plasticity of plants is qualified as environmental conditions like water and luminosity influence (Smith et al., 1998). These conditions can be responsible for morphological and anatomical changes in the leaves (Larcher & Boeger, 2009;Defaveri, Arruda, & Sato, 2011). Besides that, water availability is a limiting factor for plant occupation in coastal environments (Rosado & Mattos, 2007).

Conclusions
Based on the results obtained, it can be concluded that: 1. It was observed that the anatomical phenotypic plasticity in the leaves of Myrcia splendens (Sw.) DC is related to environmental variations throughout the year, since most of the analyzed parameters responded to abiotic factors (water and light); 2. This study contributed to the understanding of the environmental influence on morphoanatomy and ecological plasticity of the species Myrcia splendens (Sw.) DC in the Vegetation Complex of the Coastal Zone of Ceará. Environmental and