Abstract

Background

Desiccation is a critical factor influencing seed viability and storage potential. Among the physiological changes accompanying seed drying are alterations in carbohydrate reserves and the activities of carbohydrate metabolizing enzymes.

Aims

This study investigated changes in reducing sugar and starch content, as well as the activities of α-amylase, β-amylase, and total amylases, in the seeds of two tropical timber species, Afzelia africana Sm. and Gambeya albida (G. Don) Aubrév. & Pellegr., during progressive desiccation.

Methods

Mature seeds were harvested directly from parent trees and desiccated under ambient conditions for 0, 3, 7, 10, 13, 15, 22, and 35 days. Germination capacity, carbohydrate levels, and amylolytic activities were assessed at each desiccation interval.

Results

Germination in A. africana increased gradually with desiccation, whereas seeds of G. albida exhibited high initial germination that declined significantly (p ≤ 0.05) following moisture loss. Reducing sugar levels were consistently higher in G. albida than in A. africana, while starch content increased in A. africana but declined in G. albida during later stages of desiccation. Activities of α-amylase, β-amylase and total amylases increased during early desiccation (0–7 days) in both species, followed by a significant (p ≤ 0.05) decline, with A. africana exhibiting higher overall amylolytic activity. These results indicate contrasting carbohydrate metabolic responses to desiccation, consistent with orthodoxlike behavior in A. africana and recalcitrant behavior in G. albida.

Conclusion

The findings provide physiological insight relevant to seed storage and reforestation strategies for tropical forest species.

Introduction

Afzelia africana Sm. and Gambeya albida (G. Don) Aubrév. & Pellegr. are tropical rainforest timber species. A. africana, commonly referred to as African mahogany, is a large deciduous woody species threatened by overexploitation and poor natural regeneration in its native habitats (Padonou et al., 2013). It is a leguminous tree of the family Fabaceae and subfamily Caesalpiniaceae, occurring in both humid and dry forest ecosystems (Umedum et al., 2014). Every part of the plant is valuable. The species produces high quality timber, which is termite resistant but somewhat difficult to work; the seeds contain approximately 31% fat and have potential as a source of oil for both domestic and industrial applications. The seeds are also ground into flour for various culinary uses. The leaves serve as forage and food, while decoctions made from the flowers and roots are traditionally used to treat various ailments. Despite its usefulness, this plant is under threat due to overexploitation, urban expansion, and the degradation of natural forests. It is classified as Vulnerable by the International Union for Conservation of Nature (2004). G. albida, commonly known as white star apple, is a forest fruit tree belonging to the family Sapotaceae. It possesses significant economic value, particularly due to findings that jams made from its pulp could rival commercial raspberry jams and jellies. Additionally, the seed oil has various applications (Amusa et al., 2003). The fruit is also a rich source of natural antioxidants, contributing to health by combating oxidative stress related diseases (Burits & Bucar, 2000). The timber of G. albida is very dense, moderately strong and fairly durable (Etukudo, 2003). The seeds are edible and used locally in soup thickening. The leaves are an excellent fodder and fruit pods are good fuel wood.

Seeds are essential components of plant production systems, serving both direct planting needs and, seedling production and propagation. While certain seeds can be stored without losing viability, others are highly desiccation sensitive and have limited storage potential, restricting their usage in production systems (Hay & Probert, 2013). In the case of A. africana and G. albida, that both inhabit tropical forests and produces fruits shortly before the onset of the dry season, responses to seed desiccation is important because natural seedling regeneration depends on the moisture content (MC) of the forest floor.

Although some studies have reported that seeds of A. africana exhibit desiccation sensitivity under specific conditions (Adelani et al., 2017), available evidence on its seed storage behavior remains inconclusive. Consequently, the present study adopts a physiological perspective, examining MC and germination relationships alongside metabolic responses during progressive drying, rather than assuming a predefined storage classification. In contrast, seeds of G. albida have consistently been described as desiccation sensitive, with marked declines in germination observed as seed MC decreases (Adelani et al., 2017). While the general processes regulating responses to desiccation in tropical forest species have been described, the mechanisms governing seed desiccation responses in tropical forest species have not been well documented (Daws et al., 2011). Current studies on other recalcitrant taxa show that desiccation interferes with energy metabolism, osmotic balance, leading to a breakdown of regulatory proteins (Berjak & Pammenter, 2013), while orthodox seeds accumulate protective soluble sugars and carbohydrates to maintain cellular structure (Farrant & Moore, 2011). These varying responses reveal the specific biochemical responses to desiccation and the need for a direct physiological assessment rather than generalizing unrelated taxa (Berjak & Pammenter, 2013).

Although there are suggestions that both A. africana and G. albida are sensitive to desiccation, there is no published study that has examined how carbohydrate metabolizing enzymes, in particular amylase and reserve carbohydrates (sugar and starch), change during drying in these species. This is a critical gap in knowledge because carbohydrate metabolism plays a major role in the acquisition or loss of desiccation tolerance (Berjak & Pammenter, 2013). Also, without species specific biochemical data, the physiological basis of desiccation sensitivity in these economically and ecologically important species remains unclear. Therefore, this study addresses this gap by investigating amylase activity and carbohydrate accumulation during progressive desiccation of seeds of both species. The research objectives are to quantify changes in MC during specific desiccation intervals; determine the amylase activity in response to desiccation; assess the changes in reducing sugar and starch accumulation during desiccation; and relate these changes to potential desiccation tolerance or sensitivity in both species.

Materials and Methods

Study Area/Experimental Site

The study was conducted in Calabar, the capital city of Cross River State, Nigeria. Laboratory analyses were performed at the Graduate Research Laboratory, Department of Plant and Ecological Studies, Faculty of Biological Sciences, University of Calabar.

Seed Preparation/Desiccation

Mature fruits of A. africana and G. albida, were harvested directly from the tree stands within Calabar metropolis. The fruits were de-pulped to obtain their seeds. The seeds were desiccated under ambient conditions on a laboratory bench for 0, 3, 7, 10, 13, 15, 22, and 35 days; with average daytime temperatures of approximately 28 to 32 °C, relative humidity ranging between 60 to 75%, and natural sunlight exposure filtered through the screen house. On each of these sampling dates, twenty seeds of each species were randomly picked and weighed to obtain their fresh weight. Seeds were oven dried at 103 ± 2 °C until constant weight, following standard seed moisture determination protocols (Association of Official Analytical Chemists [AOAC], 2006). The fresh and dry weights obtained were used to calculate the percentage MC of the seed lot. MC was calculated adopting Somrug et al. (2024) as

MC (%) = fresh weight – dry weight × 100

                           fresh weight      

Desiccation and Germination Tests

Forty seeds were sown in germination trays at each desiccation level to investigate their germination potentials. The seeds were steeped in water for 12 h before sowing. An equal mixture of smooth sea sand and sawdust was used as the growth medium; 2 kg of the planting medium was weighed into germination trays and laid out in a complete randomized design, replicated three times. The germination of the planted seeds was observed daily, and germination counts were recorded daily. The emergence of radicle through the seed coat was the criterion for germination. The germination records were used in computing the percentage maximum germination (G_max) after Ngele et al. (2024) as follows:

G_max (%) = number of germinated seeds × 100

                   total number of sown seeds  

Determination of Reducing Sugars and Starch

The seeds employed for sugar and starch extraction were dried at 60 °C for 24 hours. Sugar and starch concentration in seeds on collection and following desiccation was assayed using the dinitrosalicylic acid (DNS) method (AOAC, 2006) with 1% glucose solution as standard.

Calibration Curve for Sugar

Serial dilutions of the 1% glucose solution were prepared to obtain glucose concentrations of 0, 2, 4, and 6 mg/mL in labeled test tubes. To each of the test tubes were added 1 mL DNS reagent and 2 mL of 1.5% NaOH, and brought to boil for five minutes. The resultant solutions were allowed to cool and the absorbance readings determined spectrophotometrically at 540 nm wavelength. The absorbance readings obtained were plotted against sugar concentration to obtain a calibration curve.

Determination of Reducing Sugars

Reducing sugars were extracted using 80% ethanol. Precisely 0.2 g of dried cotyledonary tissue (dried at 60 °C) from seeds of each species was ground in 5 mL of 80% ethanol. The homogenate was centrifuged at 4000 rpm for 5 minutes. In a test, 0.5 mL of the resulting sugar extract was mixed with 1 mL of DNS reagent and 2 mL of 1.5% NaOH. The mixture was then heated in a water bath for 5 minutes. Reducing sugar content (µmol g^-1 dry weight) was estimated spectrophotometrically at 540 nm, and concentrations were determined using a standard calibration curve prepared from 1% glucose solution.

Starch Determination

Starch present in the insoluble residue following sugar extraction was solubilized with 1 M NaOH and subsequently neutralized with 1 M acetic acid. The starch content (µmol g^-1 dry weight) was then determined by acid hydrolysis using 0.5 mL of 1 M HCl, followed by the colorimetric estimation of the resulting reducing sugars.

Determination of Alpha, Beta and Total Amylolytic Activities in Seeds

Enzyme Extraction

The extraction buffer was composed of 50 mM mixed phosphate buffer (prepared from monobasic potassium phosphate [KH₂PO₄], and dibasic potassium phosphate [K₂HPO₄] salts), containing 1% (w/v) polyvinyl polypyrrolidone (PVPP), a phenolic binding agent. The pH was adjusted to 7.0 at 30 °C (Ngele et al., 2024). To prepare the extraction buffer, 2.613 g of K₂HPO₄ was dissolved in 300 mL of distilled water, and 2.042 g of KH₂PO₄ was dissolved in another 300 mL of distilled water. The two solutions were then combined and 6 g of PVPP was added. The buffer was stored refrigerated until used. For enzyme extraction, 0.5 g of cotyledonary seed tissue from each species at specified desiccation intervals was homogenized in 5 mL of cold extraction buffer. The homogenate was first for 5 minutes, and the resulting filtrate was centrifuged again under the same conditions. The final supernatant was kept on ice and served as the crude enzyme extract.

Enzyme Assay

Amylase activity was assessed by measuring the amount of reducing sugars released from soluble starch and expressed as µmol of glucose equivalents g^-1 dry weight produced in 10 minutes at 25 °C, using 600 µL of crude enzyme extract, following the procedure described by Nkang (2002).

To determine α-amylase activity, β-amylase was selectively inactivated by heating the crude enzyme extract at 70 °C for 30 minutes, a temperature previously reported to preferentially inactivate β-amylase while retaining α-amylase activity in plant tissues (Nkang, 2002). A reaction mixture containing 0.1 mL of 1% (w/v) soluble starch and 600 µL of the heat-treated enzyme extract was incubated at 25 °C for 10 minutes. The reaction was terminated by adding 1 mL of DNS reagent followed by 2 mL of 1.5% sodium hydroxide, and the mixture was boiled for 2 minutes. Absorbance was measured at 540 nm, and α-amylase activity was quantified using a glucose standard curve and expressed on a dry weight basis (µmol reducing sugar released g^-1 dry weight min^-1).

For β-amylase activity, α-amylase was selectively inhibited by incubating the crude enzyme extract with 25 mM ethylenediaminetetraacetic acid (EDTA) at 25 °C for 30 min, as EDTA chelates divalent cations required for α-amylase activity, thereby allowing preferential estimation of β-amylase (Nkang, 2002). A reaction mixture comprising 0.1 mL of 1% soluble starch and 600 µL of the treated enzyme preparation was incubated at 25°C for 10 min. The reaction was stopped by adding 1 mL of DNS reagent and 2 mL of 1.5% sodium hydroxide, followed by boiling for 2 min. The absorbance was read at 540 nm, and β-amylase activity was expressed as µmol reducing sugar released g^-1 dry weight min^-1.

Total amylase activity was assayed using untreated enzyme extract. A reaction mixture containing 2 mL of assay buffer, 600 µL of enzyme extract, and 0.1 mL of 1% soluble starch was incubated at 25 °C for 10 minutes. The reaction was terminated as described above, and absorbance was recorded at 540 nm. Total amylase activity was calculated from the glucose standard curve and expressed on a dry weight basis (µmol reducing sugar released g^-1 dry weight min^-1).

All enzyme assays were conducted in triplicate. While no separate inhibition controls were included, the use of heat treatment and EDTA for selective estimation of α- and β-amylase followed previously validated protocols, and the approach provides comparative rather than absolute measures of enzyme activity during desiccation.

Statistical Analysis

All experiments were conducted using a completely randomized design. Data obtained from MC, germination percentage, carbohydrate concentrations, and amylase activities were based on three independent replicates and are presented as mean ± standard error of the mean. Statistical analyses were performed using the Statistical Package for Social Sciences for Windows, version 20.1.

Differences among desiccation intervals and between species were evaluated using one way analysis of variance (ANOVA). When ANOVA indicated significant treatment effects, mean separation was performed using Duncan’s Multiple Range Test (DMRT) at a significance level of p ≤ 0.05. DMRT was selected because of its sensitivity in detecting treatment wise differences across multiple desiccation intervals, allowing effective discrimination of gradual physiological changes associated with progressive drying. This approach is commonly applied in plant physiological and seed biology studies where treatments represent ordered stress gradients rather than independent categorical factors.

Results

MC and Germinability of A. africana and G. albida Seeds After Desiccation

The MC levels in seeds of A. africana and G. albida on collection and following desiccation are presented in Table 1. Seeds of G. albida had relatively higher MC on collection (40.04%) compared to A. africana (13.81%). MC declined significantly (p ≤ 0.05) throughout the desiccation period in seeds of G. albida. In A. africana, MC showed a gradual decline but remained relatively stable during the later desiccation stages (days 15-35). Germination capacity gradually improved with desiccation, reaching its maximum at 15 days (90.00%) when MC was 4.86%.

However, a significant decline in G_max (p ≤ 0.05) was observed at 22 and 35 days of desiccation, coinciding with a further reduction in MC to approximately 4%. In G. albida G_max was significantly higher (p ≤ 0.05) on collection (76.67%) but declined gradually with desiccation. No germination occurred after 13 days at MC below 26.7%.

Reducing Sugar Content in Seeds of A. africana and G. albida During Desiccation

The reducing sugar content in seeds of A. africana was lower than that of G. albida across the desiccation period (Figure 1). A significant increase (p ≤ 0.05) in reducing sugars was observed in A. africana, peaking at 7 days (3.69 µmol g^-1 dry weight), after which sugar levels declined.

Overall, G. albida seeds maintained consistently higher reducing sugar levels, indicating species specific differences in carbohydrate behavior. Reducing sugar content increased significantly (p ≤ 0.05) from 0 to 3 days in G. albida, with the highest levels at 3 days. Although sugar levels declined afterward, there was a secondary significant increase (p ≤ 0.05) at 35 days (3.16 µmol g⁻¹ dry weight) suggesting a possible delayed metabolic adjustment to prolonged desiccation.

Starch Content in Seeds of A. africana and G. albida During Desiccation

Starch content in A. africana increased markedly (p ≤ 0.05) during early desiccation, with maximum concentration at 3 days (5.37 µmol g⁻¹ dry weight), followed by a significant decrease
(p ≤ 0.05) between days 7–13 (Figure 2). Starch content then increased steadily (p ≤ 0.05) from days 15–35. In G. albida, starch content increased slightly up to 3 days but not significantly, followed by a significant decline (p ≤ 0.05) after that. This suggests distinct starch metabolism patterns between both species during desiccation.

Influence of Desiccation on Activities of Amylases in Seeds of A. africana and G. albida

Alpha-amylase activity in A. africana increased significantly (p ≤ 0.05) from 0 to 3 days of desiccation (Figure 3). Activity then declined significantly (p ≤ 0.05) between days 7–13, followed by a subsequent increase. In G. albida, alpha amylase activity increased significantly during the initial desiccation phase, peaking at 7 days, and subsequently declined. The delayed peak in G. albida (7 days) compared with A. africana (3 days) shows differences in the timing of enzyme responses in the two species.

Beta amylase activity in seeds of A. africana during desiccation followed a similar trend as that of alpha amylase, with significantly (p ≤ 0.05) higher activity on the 3^rd day of desiccation (Figure 4). Similarly, seeds of G. albida also demonstrated a significant
(p ≤ .05) increase in beta amylase activity with a peak at the 7^th day of desiccation, thereafter activity declined.

Total amylase activity in A. africana increased significantly  (p ≤ 0.05), with the highest activity observed at 3 days (21.39 µmol reducing sugar released g^-1 dry weight min^-1) (Figure 5). Activity decreased thereafter but showed slight significant increases  (p ≤ 0.05) between days 15–35. In G. albida, total amylase activity increased significantly at 0, 3, and 7 days having values of (2.40 µmol reducing sugar released g^-1 dry weight min^-1, 9.47 µmol reducing sugar released g^-1 dry weight min^-1 and 12.72 µmol reducing sugar released g^-1 dry weight min^-1 respectively), followed by a significant decline. Generally, A. africana exhibited an earlier enzymatic peak, whereas G. albida showed delayed and brief increases (Table 2).

Discussion

A. africana seeds exhibited significantly lower MC at collection (13.81%) compared with G. albida seeds (40.04%), indicating clear differences in seed water status at maturity. Such contrasts are commonly associated with differences in seed storage behavior. Orthodox seeds typically undergo maturation drying during late development, acquiring tolerance to substantial moisture loss, whereas recalcitrant seeds retain high MC and are sensitive to desiccation (Berjak & Pammenter, 2013; Maia et al., 2011). In the present study, the ability of A. africana seeds to retain high germination capacity at low moisture levels supports an orthodox-like response under the conditions tested, whereas the rapid loss of viability in G. albida during drying is consistent with recalcitrant behavior.

Reducing sugar content in A. africana seeds was relatively low both at collection and following desiccation, whereas G. albida maintained consistently higher levels. Previous studies have associated high concentrations of reducing sugars, particularly monosaccharides, with increased desiccation sensitivity, whereas oligosaccharides are more often linked to cellular stabilization during dehydration (Prajith et al., 2014; Whittaker et al., 2004). However, since only reducing sugars were quantified in this study, interpretations are restricted to observed changes in reducing sugar pools rather than specific protective sugars. The higher reducing sugar levels in G. albida may therefore reflect limited metabolic regulation during drying rather than an adaptive protective response. Although oligosaccharides such as raffinose have been implicated in desiccation tolerance in other species, these compounds were not directly measured in the present study and are therefore not considered further here.

The starch metabolism also differed markedly between the two species. Starch content increased during desiccation in A. africana but declined in G. albida, indicating contrasting patterns of carbohydrate reserve regulation. The accumulation of starch in A. africana during drying may reflect a capacity to maintain reserve stability under reduced water availability, whereas starch depletion in G. albida coincided with declining germination capacity. Activities of α-, β-, and total amylases increased during early desiccation (0–7 days) in both species, followed by a decline with prolonged drying. A. africana exhibited higher and more sustained amylolytic activity, suggesting greater metabolic resilience during dehydration. Similar temporal modulation of amylase activity in orthodox seeds has been reported under varying intensities of desiccation (Lee & Tan, 2020). In contrast, the sharp reduction in amylase activity observed in G. albida during the later stages of desiccation may indicate the loss of metabolic integrity, a feature commonly reported in recalcitrant tropical seeds (Nguyen et al., 2023). Although reducing sugars interact with oxidative metabolism in plants, we did not assess reactive oxygen species levels, antioxidant enzymes, or related redox pathways. Consequently, no conclusions are drawn regarding the regulation of oxidative stress during seed desiccation. Such interactions remain important avenues for future investigation.

Overall, this study distinguishes contrasting carbohydrate reserve dynamics and amylolytic responses associated with desiccation tolerance and sensitivity in two tropical tree species. The results provide physiological evidence supporting the suitability of A. africana seeds for conventional drying and storage, while highlighting the vulnerability of G. albida seeds to moisture loss. These findings are relevant for seed conservation and restoration programs that require speciesspecific handling protocols (Food and Agriculture Organization, 2021).

Although this study offers useful insights, it is limited by the absence of molecular or biochemical markers beyond carbohydrate metabolism and the restricted number of species examined. Future studies integrating analyses of specific sugar profiles, enzyme regulation at the molecular level, and controlled environmental conditions would further refine the understanding of desiccation responses in tropical forest seeds.

Conclusion

Based on the findings, it is suggested that A. africana seeds may be suitable for conventional drying and storage under conditions similar to those tested in this study, indicating potential for ex situ conservation. In contrast, G. albida seeds exhibited characteristics consistent with desiccation sensitivity and may therefore require alternative management approaches, such as short term hydrated storage, rapid propagation, or further evaluation for specialized preservation techniques. Future research should incorporate longer term storage trials under controlled environmental conditions to validate storage performance, as well as molecular or ecological investigations to better understand the mechanisms underlying the observed physiological responses.