Integrating fruit and seed traits to support the conservation of the threatened tree Stephanopodium engleri

Abstract

The endangered tree species Stephanopodium engleri Baill., endemic to the semideciduous seasonal forest of the Quadrilátero Ferrífero in Brazil, is threatened by its limited distribution and loss and degradation of its habitat. The objective of this study was to characterize the fruits, seeds, and germination traits, providing support for the development of effective conservation strategies. A biometric analysis of 300 fruits revealed that the species produces bilocular drupes, although approximately half of the fruits contain only a single seed. Anatomical studies demonstrated the presence of a fibrous seed coat, which is more permeable at the hilum and helps to regulate water uptake. The germination trials revealed that fresh seeds had a 92% germination rate and a rapid germination time, averaging 12 days when tested at 25 °C. However, the seeds lost viability entirely after 90 days of storage, along with significant water loss, which is indicative of recalcitrant seed storage behavior. The abundance of reserves, including starch, proteins, and lipids in the cotyledons, promotes rapid initial growth and supports seedling establishment. These findings indicate that S. engleri depends on immediate germination and quick seedling growth for survival, with sensitivity to dry conditions possibly limiting its range to moist, intact forest patches. The results emphasize the species’ reliance on specific environmental factors for regeneration and highlight the difficulties in ensuring its long-term conservation.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-39592-3.

Introduction

Morphological studies of fruits, seeds, and seedlings are crucial for understanding species diversity, evolution, ecological interactions, and ecosystem dynamics^1,2. These studies encompass a variety of morphological characteristics, such as shape, size, color, texture, anatomy, and surface features, that serve as fundamental tools for species recognition and classification. By enabling rapid and accurate field identification, morphological analyses significantly contribute to biodiversity research, conservation efforts, and ecosystem management^3,4. Additionally, biometric analyses of fruits and seeds provide valuable insights into dispersal mechanisms, seedling establishment, and plant reproductive strategies⁵. Identifying variation in fruit and seed traits within and among plant populations is crucial for revealing species adaptation, genetic diversity, and evolutionary processes⁶. This information plays a pivotal role in validating ecological studies and predicting plant population responses to environmental changes⁷.

Understanding germination and seedling emergence requirements is essential for planning germplasm conservation and population restoration programs for endangered species⁸. For example, the germination success of Jacquemontia reclinata, an endangered plant endemic to Southeast Florida, improved after short-term seed storage, indicating an after-ripening effect⁹. In contrast, the recalcitrant seeds of Hopea chinensis experience rapid viability loss in their natural habitat, posing significant challenges to population regeneration¹⁰. Effective conservation strategies must consider the specific storage behaviors and needs of different seed types. Recalcitrant seeds have limited storage longevity and require careful handling to preserve their genetic diversity^11,12. These seeds are vulnerable to water loss, which can quickly render them nonviable after dispersal, thereby posing challenges for long-term conservation efforts. Seed viability depends on genetic factors, environmental conditions, and the maturity of seeds at shedding. While orthodox seeds can be preserved in seed banks for decades, recalcitrant seeds usually remain viable for only a few months under optimal storage conditions, such as refrigeration¹³.

Seed morphology, structure, and composition—including seed coat thickness, permeability, and reserve content (carbohydrates, proteins, and lipids)—are crucial for understanding seed dormancy, germination, and seedling growth^14–16. The seed coat influences water uptake, gas exchange, and resistance to pathogens, directly impacting seed viability, germination, and seedling emergence. Additionally, seed reserves are vital for defining a species’ ecological strategy, influencing seedling survival and distribution¹⁷. Starch, a major seed reserve, provides energy by converting into soluble sugars that fuel metabolic activity and regulate growth during germination¹⁸. Proteins and lipids also play crucial roles, supplying energy and providing structural support, as well as aiding metabolic processes essential for germination^19,20. Efficient metabolism of these reserves enables plants to adapt to environmental cues, regulate germination and growth, and establish seedlings rapidly and competitively.

In this study, we focus on the morphological characterization of fruits and seeds, along with evaluations of seed moisture loss and viability in Stephanopodium engleri Baill. (Dichapetalaceae), a tree species endemic to the Semideciduous Seasonal Forest of the Quadrilátero Ferrífero, Minas Gerais, Brazil^21,22. With an estimated area of occurrence of less than 5,000 km² and an area of occupation of less than 500 km², S. engleri falls into the “Endangered” (EN) category according to IUCN Red List categories. Rediscovered in 2010 near Belo Horizonte after being considered “probably extinct” in 1997, this species faces threats from urban expansion and mining activities, which diminish its extent of occurrence (EOO), area of occupancy (AOO), and habitat quality²¹. Currently, only a limited number of adult individuals capable of producing fruits and seeds have been observed in their natural habitat, highlighting the urgent need for conservation efforts²¹. This study aims to address the knowledge gap by providing comprehensive data on fruit and seed morphology, anatomy, and germination, which are crucial for developing effective conservation strategies and management plans for S. engleri.

Material and methods

Plant materials

Ripe fruits of 18 adult S. engleri plants were collected during the dry season (June to August) from natural populations in forest fragments (Semideciduous Seasonal Forest) near Belo Horizonte, Brazil. Plant vouchers from these populations had previously been deposited in the BHZB herbarium (voucher no. 11303). Permission for accessing these collection sites was kindly provided by Vale S.A. and Gerdau S.A. Upon collection, the fresh fruits were placed in plastic bags and transported to the laboratory for further processing. Morphological and biometric analyses were conducted on the ripe fruits and freshly pulped seeds. The seeds were manually extracted, thoroughly washed under running water to remove any remaining pulp, and then placed on plastic trays to air-dry at room temperature (approximately 20 °C) with a relative humidity of around 65% for 48 h. These seeds, referred to as "fresh seeds," were immediately used in subsequent experiments. Additionally, a subset of 200 seeds was left to desiccate under the same environmental conditions to evaluate the species’ seed storage behavior.

Morphological characterization of the fruits and seeds

For fruit characterization, several traits were documented, including color, dimensions, texture, dehiscence, and external wrapping. Seed characterization involved assessing external features such as dimensions, color, texture, consistency, integument color, and the position of the hilum and micropyle. Biometric analyses were conducted on a sample of 300 randomly selected fruits, from which the seeds were manually extracted and measured. The following characteristics were measured for both fruits and fresh seeds: length (mm), from the base to the apex; width (mm), at the widest horizontal point; and thickness (mm), along the central midline. Seed mass was also determined using an analytical balance accurate to 0.0001 g, which was used to measure the fresh mass of both fruits and seeds.

Anatomical traits

The anatomical analysis of S. engleri seeds was conducted using transverse and longitudinal sections. Seeds were initially fixed in FAA 70 (Formaldehyde: acetic acid: 70% ethyl alcohol 1:1:18) for 48 h ²³. After fixation, the seeds were dehydrated and stored in a 70% ethyl alcohol solution. To identify the most effective softening method prior to sectioning, two treatments were tested: 10 seeds were boiled in distilled water for five minutes and left to cool completely, while another 10 seeds were placed in 50% glycerin for 24 h. Following these treatments, seeds from both softening methods underwent dehydration through an ascending ethyl alcohol series (80%, 90%, 95%, and P.A.), with each step lasting 1 h. Pre-infiltration was performed by immersing the seeds in a solution of equal amounts of 100% ethyl alcohol and resin for 24 h, followed by infiltration in pure resin for 72 h. Fresh seeds were then embedded in Historesin (Historesin—Leica®/3875) following the manufacturer’s instructions, and the blocks were polymerized at 60 °C for 24 h. Blocks were sectioned into 5–7 µm thick slices using a manual rotary microtome (RM 2145–Leica®), and the sections were stained with toluidine blue for 10 min before being mounted in an aqueous medium. Histochemical tests were also performed on transverse and longitudinal sections of fresh seed materials. Sections were mounted in 50% aqueous glycerin, and various reagents and stains were applied: Lugol to detect starch²³, Coomassie Brilliant Blue for protein visualization²⁴, Sudan III for lipid detection²³, Ferric Chloride for phenolic compounds²³, and Dragendorf reagent for alkaloid detection²⁵. Observations and photographic documentation were conducted using a Zeiss Scope A.1 optical microscope equipped with an AxioCam TCc 5 digital camera.

Seed germination and storage behavior

To determine seed storage behavior, we compared the germination of fresh and stored dry seeds of S. engleri. For the dry-stored seeds, 150 fresh seeds were gently dried at 65% relative air humidity and 20 °C until a stable mass, which took 36 days. Seed mass was measured daily using an analytical balance (precision 0.0001 g). Once the seeds reached a stable moisture content, they were stored in paper bags at room temperature for 54 days (to complete 90 days from the fruit removal). For the seed germination experiments, fresh and dry-stored seeds, with six replicates of 25 seeds each, were placed in Gerbox containers filled with sterilized sand and saturated with distilled water, under a 12:12-h photoperiod (50 μmol m⁻² s⁻¹) at a constant temperature of 25 °C. Germination was radicle protrusion²⁶. The experiments were monitored twice weekly for two months. Germination percentage and mean germination time were calculated according to Santana and Ranal (2006).

The seed moisture content of the dry-stored seeds was determined according to the Seed Analysis Rules protocol²⁸. However, we were unable to measure the moisture content of the freshly harvested seeds, and as a result, we used an indirect method to estimate this value. Therefore, we used the fresh mass, dry mass, and moisture content of stored seeds to estimate the moisture content of the fresh seeds using the following equation:

where: % MC_initial is the moisture content of recently dispersed seeds, Ps is the fresh weight of dry stored seeds, Ms is the moisture content of dry stored seeds, Pi is the fresh weight of recently harvested seeds.

To determine the seed moisture content, a subset of 30 desiccated seeds (each weighing approximately 0.20 g) was placed in a forced-air oven at 105 °C for 24 h. After this period, the seeds were weighed using a precision balance with an accuracy of 0.0001 g.

Data analysis

A completely randomized design was used for the experiment, with biometric data (length, width, thickness) presented as the frequency of occurrence (%). To evaluate the effect of desiccation on seed germination, the germination success (1 or 0) and median germination time of fresh and stored seeds were compared. Generalized Linear Models (GLM) were initially planned to evaluate both germination success and mean germination time, using binomial and gamma distributions, respectively. However, because no seeds germinated in the desiccation treatment, it was not possible to perform GLM analyses. All statistical analyses were performed in the R environment²⁹.

Results

Morphological attributes of fruits and seeds

Stephanopodium engleri produces drupe fruits with a slightly fleshy texture and bilocular structure. The fruits feature a thin mesocarp and a delicate, parchment-like endocarp that remains intact upon ripening. As the fruits mature, they undergo a color change, shifting from yellow to darker hues (Fig. 1A-B). The seeds are covered by a uniformly brown outer testa (Fig. 1C) and display a small, oblong hilum with a depression (Fig. 1D). The micropyle is a small circular opening situated just above the hilum (Fig. 1E-F). Despite the bilocular structure, analysis of 300 mature fruits revealed that 52% produced two seeds, while 48% contained only one seed.

Fig. 1.

Morphological characterization of fruits and seeds of Stephanopodium engleri. A Fresh fruit. B Fruit interior showing the mesocarp (pulp) and presence of two seeds. C Endocarp. D Seed. E Detail of the outer surface of the seed (parchment). F Detail of the hilar region. Bars: A-D = 2 mm; E–F = 1 mm.

The fruits exhibited low size variation, with over 70% measuring between 14–16 mm in length, 55% ranging from 13–15 mm in width, and most fruits having a thickness of 10–11 mm (Fig. 2A-C). Similarly, the seeds had limited size variation (Fig. 2D-F), with 70% measuring 10–12 mm in length, 4.5–6 mm in width, and 3–6 mm in thickness (92%). Fruits with two seeds were heavier than those with one seed. The other fruit biometric traits (length, width, and thickness) did not show significant differences. Seeds from fruits with one seed were wider and heavier, although no significant differences in seed length or thickness were observed between single- and double-seeded fruits (Supplemental Table 1).

Fig. 2.

Biometric parameters of fruits and seeds of Stephanopodium engleri. A-C Fruits and D-F Seeds. (n = 300).

Anatomical characterization of the seed

Longitudinal sections of the seed coat in the hypocotyl-radicle axis show a fibrous outer structure (Fig. 3A-B), consisting of one to two layers of libriform fibers with simple pits, with a partial disruption of the secondary wall over the primary wall (Fig. 3C). The cotyledon exhibits a uniseriate epidermis, with the mesophyll comprising juxtaposed and heterodimensional parenchymatous cells that have smooth to slightly sinuous walls (Fig. 3D). Similarly, the endosperm consists of parenchymatous cells with smooth or slightly sinuous walls, containing visible reserves within the lumen (Fig. 3E). In the hilum region, external seed coat protrusions are observed, and internally, parenchymatous cells of varied shapes and sizes with thin walls are present (Fig. 3F). Tests for alkaloids, phenolic compounds, and lipids on cells in the seed coat yielded positive results (Supplemental Table 2). The cotyledon and endosperm are notably rich in reserves, including starch grains (Fig. 4A-C), phenolic compounds (Fig. 4D), lipids (Fig. 4E-F), and proteins (Fig. 4G-H).

Fig. 3.

Anatomical characterization of Stephanopodium engleri seeds. A Seed in longitudinal view. B Seed coat. C Fibers of the seed coat. D Cotyledons. E Endosperm. F Hilar region. Bars: A, B, D, E, F = 100 µm; C = 20 µm.

Fig. 4.

Histochemical characterization of Stephanopodium engleri seeds. A Alkaloids. B-C Starch grains. D Phenolic compounds. E–F Lipids. G-H Proteins. Bars: A-H = 100 µm.

Seed Germination and Seed Moisture

The fresh seeds of S. engleri exhibited a high moisture content (34 ± 6%), germination percentage (92%), and short mean germination time (12 days). The remaining seeds (8%) that did not germinate by the end of the experiments were rotten. In contrast, dry storage treatments (seed moisture content 16 ± 5%) led to a complete loss of viability in the seed population (Fig. 5A). Therefore, GLM for germination was not conducted. This complete loss of viability even during storage in moderate relative humidity, the high water content of freshly collected ripe seeds, and their expressive water loss in post-harvest, (Fig. 5B), strongly suggests that S. engleri exhibits a recalcitrant seed storage behavior.

Fig. 5.

Percentage of cumulative seed germination in fresh and stored seeds (A) and seed weight variations after removal from fresh fruits (B) of Stephanopodium engleri. The shaded area represents the confidence interval at P < 0.05.

Discussion

Fruit and seed biometry contribute to understand the ecology interactions and enhance propagation success of Stephanopodium engleri

Information on the fruits and seeds of S. engleri is limited, with the only previous description by Prance in 1972 (available in Prance 2022) noting irregular, ellipsoid fruits, approximately 1.5 cm long, with a velvety, rusty epicarp, a thin mesocarp, and a soft, bony endocarp. This study expands on that by providing details on fruit coloration, seed frequency per fruit, and seed anatomy—crucial for understanding dispersal mechanisms, reproductive strategies, and ecological interactions. Such data is particularly important for the conservation of rare and threatened species, where efforts are often hindered by knowledge gaps and limited propagation material. Our results confirm that S. engleri produces drupes with a thin and freshly developed mesocarp, suggesting endozoochory as the likely mode of seed dispersal³¹. Moreover, the ripening of small, yellow fruits during the dry season is often associated with dispersal by birds and mammals^32,33. However, more field studies are necessary to verify these dispersal hypotheses.

Biometric traits of fruits and seeds are critical in influencing germination and seedling establishment, making them essential for determining the optimal timing for fruit and seed collection^34,35. The variation in fruit size and seed number observed in this study may result from a combination of internal factors, such as genetic diversity and physiological conditions, along with external environmental influences. Variations in seed number per fruit can arise from several factors: (i) the amount and quality of pollen reaching the stigma, which can vary due to pollen production, pollinator activity, and pollen transport mechanisms³⁶; (ii) pollination success, affected by pollen viability and environmental conditions, including water availability, temperature, and humidity³⁷; and (iii) the number of fertilized ovules that develop into seeds, which depends on the availability of nutrients and photosynthates^38,39. Lenser et al. (2016) propose that heteromorphic fruits and seeds improve a species’ dispersal and germination strategies by offering different mechanisms for survival in diverse environments. However, S. engleri exhibited limited variation in fruit and seed size, regardless of seed count per fruit, suggesting that differential seed dispersal strategies based on seed set may not be relevant for this species. Additionally, the high synchrony in germination of freshly collected seeds reduces the likelihood of multiple germination strategies, possibly reflecting an adaptation to a specific set of environmental conditions⁴¹.

Seed coat morphology and its role in germination

Stephanopodium engleri seeds have a fibrous seed coat (parchment) with a hilum, which is the most permeable part of the seed due to the thinner tegument, facilitating water uptake and influencing germination. The fibrous seed coat, common to many species, can act as a physical barrier, slowing water and gas exchange and thus limiting embryo growth⁴². In such seeds, this coat often correlates with physical dormancy and reduces germination rate by restricting imbibition⁴³. In S. engleri, however, the hilum plays a key role in moisture control by acting as a hygroscopic valve. This structure allows water to enter during imbibition and improves gas exchange during storage and germination^44,45, thereby compensating for some of the limitations caused by the fibrous seed coat.

The presence of phenolic compounds in the seed coat may serve multiple purposes. Phenolics can act as a chemical barrier, reducing permeability to water and gases, thereby contributing to delayed germination⁴⁶, even though this effect was not observed in S. engleri. Their role extends beyond physical protection to include chemical defense against herbivores and pathogens⁴⁷. Phenolic compounds can impart structural rigidity to the seed coat, thereby increasing resistance to mechanical damage⁴⁶. Furthermore, alkaloids in S. engleri seeds can provide nitrogen sources for seedling development⁴⁸, further demonstrating the diverse roles of chemical compounds in the seed coat.

Reserve types and their contribution to germination

The seeds of S. engleri are characterized by fleshy cotyledons and starchy endosperm that serve as a reservoir for substantial quantities of starch, proteins, and lipids – critical reserves that are essential for germination and seedling growth. These reserve tissues provide a high nutritional value, facilitating the rapid mobilization of reserves to developing organs during germination⁴⁹. Our observations support the hypothesis that S. engleri seedlings quickly emit and expand their leaves while seed remnants remain attached for several days, indicating the use of seed reserves for initial growth.

Starch, the primary reserve in S. engleri seeds, acts as an easily accessible energy source, broken down into soluble sugars that support metabolic activity and growth during germination¹⁸. Furthermore, proteins and lipids are essential for providing structural integrity and energy, as well as supporting metabolic and regulatory processes^19,20. These reserves can be metabolized in a coordinated way⁵⁰, thereby promoting vigorous growth during the early development stages observed in S. engleri seedlings.

The diverse range of reserves ensures not only high germination rates but also consistent seedling vigor, which is essential for species survival and establishment in natural ecosystems¹⁷. By enhancing both germination success and seedling strength, the abundant reserves in S. engleri seeds offer a significant ecological advantage, aiding their potential for rapid establishment and competitive growth in their native habitat⁴¹.

Recalcitrant seeds and establishment strategy in S. engleri may limit its range of occurrence

Stephanopodium engleri produces non-dormant seeds (sensu Baskin and Baskin 2014) as evidenced by the following characteristics: freshly harvested seeds exhibited a mean germination time of less than 30 days, their seed coats were water-permeable, and the embryo was fully developed. Moreover, these seeds showed a recalcitrant storage behavior because reducing the seed moisture content to 16% resulted in a notable loss of viability. This hypothesis is supported by seed morphological traits of S. engleri, such as endospermy, a thin coat, and a high seed-coat ratio, which are associated with recalcitrant storage behavior^51,52. In desiccation-sensitive seeds, the increased reserve storage capacity is associated with the production of highly competitive seedlings^50,53,54. In the case of S. engleri, the seeds contain substantial reserves that ensure both germination and initial seedling growth, providing a competitive advantage in environments dominated by already established species²¹. This indicates that S. engleri is a late secondary to climax species, supporting the hypothesis that desiccation-sensitive seeds are characteristic of non-pioneer species⁵⁵.

The substantial seed reserves, non-dormant seeds, and continued metabolic activity in desiccation-sensitive seeds represent a regeneration strategy focused on rapid germination and seedling establishment upon dispersal⁵⁵. This strategy enables the formation of quiescent seedling banks in the understory, which can resume growth when light availability increases³⁴. However, the occurrence of species with desiccation-sensitive seeds is markedly reduced in seasonal environments, with less than 20% of species in semideciduous forests exhibiting recalcitrant seeds^55,56. Moreover, in these forests, seed desiccation sensitivity is often associated with dispersal during the rainy season, thereby reducing the likelihood of seed desiccation⁵¹. In contrast, S. engleri releases its seeds during the dry season, which may restrict its niche to preserved forest fragments where leaf litter can reduce the rate of soil moisture loss. Such as S. engleri, other recalcitrant species of semideciduous forest that disperse seeds during the dry season are constrained to places that maintain a higher level of water availability throughout the dry season⁵⁷. In this context, the prompt germination with the onset of the rainy season and the large seed reserves can enhance seedling survival and competitive ability in semideciduous seasonal forests. However, seed release during the dry season can limit the species occurrence to wet environments^51,55. Nonetheless, other factors affecting the species’ reproductive success, such as flower, fruit, and seed predation, as well as dispersal and pollinator effectiveness, require further investigation to fully comprehend its distribution and population dynamics.

Conclusions

The seeds’ rich reserves, including starch, proteins, and lipids, contribute to rapid germination and early seedling vigor, enhancing the species’ competitive potential in established ecosystems. However, their recalcitrant storage behavior and seed release during the dry season pose significant challenges for seed survival in the soil, restricting their distribution to habitats that maintain a higher level of water availability throughout the dry season, such as pristine and conserved forest areas. These results underscore the importance of understanding the species’ specific biological and ecological requirements, which can inform conservation strategies and support efforts aimed at reintroduction and long-term survival of S. engleri in its natural habitat. Thus, we consider that short-term conservation of S. engleri seeds can be achieved by storing them under high humidity and at temperatures below the germination base temperature, or by using moist substrates. Partial drying and osmotic treatments may also prolong their viability. For long-term strategies, embryo cryopreservation or living collections can be considered effective methods. These methods are crucial for including in conservation strategies to safeguard genetic diversity and secure plant material for restoration efforts.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Designed research: KJC, EAV and CFC; Performed research: KJC, EAV and CFC; Analyzed Data: EAV, MW, DFE, MG, SR and CFC; Wrote the paper: KJC, EAV, MW, MG, SJR and CFC.

Data availability

The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.