Next Article in Journal
Effects of Different Essential Oil Blends and Fumaric Acid on In Vitro Fermentation, Greenhouse Gases, Nutrient Degradability, and Total and Molar Proportions of Volatile Fatty Acid Production in a Total Mixed Ration for Dairy Cattle
Previous Article in Journal
Research on the Control Strategy of the Power Shift System of a Cotton Picker Based on a Fuzzy Algorithm
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Agriculture
Volume 14
Issue 6
10.3390/agriculture14060875
agriculture-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Biochemical and Physical Screening Using Optical Oxygen-Sensing and Multispectral Imaging in Sea Oats Seeds

by
Andrew Ogolla Egesa
1,
Maria Teresa Davidson
1,
Héctor E. Pérez
1 and
Kevin Begcy
1,2,*
1
Environmental Horticulture Department, University of Florida, Gainesville, FL 32611, USA
2
Plant Molecular and Cellular Biology Graduate Program, University of Florida, Gainesville, FL 32611, USA
*
Author to whom correspondence should be addressed.
Agriculture 2024, 14(6), 875; https://doi.org/10.3390/agriculture14060875
Submission received: 23 April 2024 / Revised: 27 May 2024 / Accepted: 27 May 2024 / Published: 31 May 2024
(This article belongs to the Section Seed Science and Technology)
Browse Figures
Versions Notes

Abstract

:
Physical, physiological, and biochemical traits control critical seed functions such as germination, longevity, persistence, and seedling establishment. These traits are diverse between and among species, and they are also controlled by the environment in which the seed originated. Therefore, screening seed traits and understanding their roles in seed functions is crucial to facilitate the economical use of resources in collecting, sorting, and conserving seed materials of agronomical and ecological importance. We hypothesized the existence of physical and biochemical traits in sea oats seeds that can be used as proxies to predict viability and vigor to develop underpinnings for survival after non-optimal storage conditions. Using multispectral imaging and optical oxygen-sensing analyses, we evaluated the physical and biochemical traits of Uniola paniculata L. (sea oats) seeds collected from the US Atlantic and Gulf coastlines. Our results showed that several traits correlate to aging stress survival in sea oats seeds. These results confirm the potential of using physical and biochemical screening to predict seed quality while offering insights into extended seed longevity periods. Therefore, exploring and analyzing the physical and biochemical properties of seeds could reveal salient markers that contribute to viability and longevity.

1. Introduction

Seeds have unique physical, morphological, physiological, biochemical, and metabolic characteristics [1,2]. These traits contribute to seed functions such as seed viability, longevity, and survival [2,3,4,5]. Collection, harvesting, and sorting of seeds for long-term storage are critical for future use. Several activities, including conservation of germplasm, planting of elite material, ecological restoration, breeding, and others involving the regeneration of new plants from stored seeds, rely on their ability to retain high viability until use. However, most current post-harvest seed processing procedures rely on general seed storage physiology classifications, especially regarding their ability to tolerate desiccation stress. Recalcitrant seeds are expected to exhibit desiccation sensitivity and a limited storage shelf life. Conversely, orthodox seeds are expected to display considerable desiccation tolerance and extended longevity during storage [3,6,7].
Seed deterioration, in several instances, begins after seed maturation and progresses during seed storage [4,5]. Seed deterioration involves changes to physical, physiological, and biochemical changes in their seed components [8,9]. Deterioration can be counterbalanced via seed desiccation and cold temperatures. This type of post-harvest handling (i.e., 8–10% moisture content, −18 °C) is often common in seed storage. However, using a controlled environment only slows the deterioration process but does not prevent its progression [10]. Therefore, it is common to find high variability in seed quality characteristics following storage, which can exhibit extensive differences between and among species. These variabilities significantly impact seed functional traits. Hence, incorporating easily accessible seed quality prediction tools in post-harvest seed handling can be beneficial.
Multispectral imaging represents a new technology with wide applications in ag-riculture. This technique has facilitated the collection and interpretation of a range of plant and soil attributes that have simplified the decision-making strategies for planting and breeding [11,12,13,14]. Multispectral imaging tools collect spectral signals at specific wave-lengths that allow the capture of broad and fine-scale spectral properties [11,12,15,16,17]. These properties are widely used in plant tissue studies [12,13,15,17], seed classification [14,17], and the detection of seed-borne diseases and other damages [12,16,17]. In addi-tion, multispectral imaging technology has been used to classify seed morphological differences in cowpeas and coffee [13,14] and in the screening of sugar beet seeds with physical damage [16]. Multispectral imaging is gaining increased application in seed biology, including seed banking [18,19]. The combination of multispectral imaging with routine seed quality and viability checks, such as seed germination assays, is on the rise and contributes to multispectral imaging metrics on seed physical traits linked to seed quality and viability characteristics [11,19].
Another novel approach involves using oxygen-quenching luminescence sensors. These sensors indicate O2 concentration as a function of the luminescence intensity and decay time of the indicator [20,21,22]. Seed respiration assays have been employed to detect the level of degradation in seed lots and are used as seed vigor indicators [23]. Furthermore, decreased seed resting respiration has been associated with reduced via-bility among the metrics obtained from the oxygen depletion trials from the respirometer [23]. Other critical biochemical assays involve the screening of the stress-relieving me-tabolites comprising antioxidants; in several studies, the screening of the antioxidants in seeds has allowed the detection of the seed deterioration progression; many reports as-sociate antioxidant decay with seed viability loss due to the progressive accumulation of stress compounds [24,25,26,27]. Therefore, combining multispectral imaging, oxygen sensing, and biochemical assays provides an opportunity to detect and explore the relationships of multiple physical and biochemical traits to their functional roles related to seed quality and viability.
Environmental conditions during seed development impact ultimate quality by interfering with the accumulation of storage reserves and antioxidants [4,28,29,30]. Fur-thermore, under non-optimal and long-term storage, there is a progressive depletion of seed storage reserves [9,25,31]. Therefore, biochemical analysis to determine the amount of seed storage reserves and other diverse compounds, including oligosaccharides, proteins, reactive oxygen species (ROS), and other secondary metabolites, are essential ways of determining seed quality and potential viability. Among these compounds, antioxidant pools facilitate seed germination and mitigate aging by preventing damage from ROS and other radicles generated during stress, metabolism, and desiccation [26,27]. Other important components for seed survival are their physical traits. In previous studies, heavier seeds in Rudbeckia mollis were preserved longer than lighter seeds, which de-teriorated more rapidly [31]. On the other hand, the seed coat structure is associated with variability in aging-induced deterioration in soybeans [32]. Therefore, incorporating biochemical and morphophysiological traits when considering seed storage may be advantageous in conserving seed lots to maintain higher viability and seedling estab-lishment.
Given the critical importance of high seed quality to support diverse restoration, conservation, and regeneration work, simple-to-use seed quality screening tools are always required. However, most existing and new simple-to-use seed quality screening tools have not been tested in different scenarios, such as their utility to confirm seed quality following temporal storage or exposure to non-optimum conditions. This can be useful in diverse scenarios, especially under limiting conditions such as damaged storage facilities. In this study, we implement simple-to-use optical oxygen sensing and multi-spectral imaging to predict sea oats seed viability and survival after several (four) years of non-optimal storage.

2. Materials and Methods

2.1. Seed Materials

We used seeds from Uniola paniculata L. (sea oats [Poaceae]), a plant species used for habitat restoration, re-vegetation, and breeding [33,34]. We collected mature seeds from six sea oat populations from sites along Florida’s Atlantic and Gulf coastlines in October 2019. Seeds were then maintained under non-optimal (22 ± 2 °C and 50% relative hu-midity) storage conditions for several (four) years [35,36]. The decision to use non-optimal storage was based on the existing prevalence of tropical cyclones, hurricanes, typhoons, and storms in this coastal region and the unpredictable nature of these events, which sometimes lead to unprecedented damages with a potential impact on ecological resto-ration projects and on the infrastructure where seed material is normally stored.
Among the six selected populations, those from the Atlantic coastline were collected at Bill Baggs (BBFL), Dr. Von. Mizell-Eula Johnson (VEFL), Fort Pierce Inlet (FPFL), and Fort Clinch (FCFL) state parks. Populations from the Gulf coastline occurred at Delnor-Wiggins (DWFL) and Honeymoon Island (HIFL) state parks (Supplementary Table S1). These six populations had higher germination levels (i.e., ≥80%) after harvest and met the Millennium Seed Bank (MSB) protocol standards for comparative seed longevity testing [37]. We set out to implement simple-to-analyze protocols for testing seeds’ physical and biochemical traits that can be easy to determine during seed collection processing, seeds in storage, ecological restoration, and in other settings, which en-compass existing and new technologies that can be easily implemented, eliminating longer (sometimes >30 days) waits associated with traditional seed viability checks.

2.2. Quantification of Biochemical Traits

2.2.1. Total Protein

We added 300 µL of 2× Laemmli sample buffer containing DL-Dithiothreitol (DTT) and phenylmethylsulfonyl fluoride (PMSF) to 100 mg of finely ground seeds. The sample buffer mixture was vortexed for 10 s and then heated in a water bath at 100 °C for 10 min, followed by pelleting at 20,000 RCF for 10 min at 4 °C. Then, 800 µL ice-cold 100% acetone was added to the supernatant and subjected to a centrifugation step at 20,000 RCF for 10 min at 4 °C, followed by drying for 2 min at room temperature. We dissolved the resulting pellet in 50 µL of 0.2 N NaOH and neutralized it with an equal volume of 0.2 N HCl. Total protein was determined using the colorimetric Bio-Rad Protein Assay Kit II (Bio-Rad Laboratories, Hercules, CA, USA). During each essay, we used seven dilutions of a protein standard containing 0, 5, 10, 15, 20, 25, and 30 µg/mL for the standard curve generation. The absorbance at 595 nm (A595nm) was then measured with a microplate spectrophotometer (Epoch Microplate Spectrophotometer; BioTek, Winooski, VT, USA), and the normalized absorbance values were plotted versus the mass concentration (µg/mg) as previously described [38]

2.2.2. Antioxidants Analysis

Total antioxidant capacity was estimated using the ABTS/TEAC assay [39]. ABTS/TEAC determines the Trolox equivalent of the antioxidant capacity (TEAC) per gram of sample. First, 20–30 mg of finely crushed sea oats seeds were used for hydrophilic extraction with 5 mL of 75% aqueous methanol and stirred under a nitrogen gas stream of 2.0 Pascals at 30 °C for 60 min. We prepared the ABTS reagent by dissolving 2.74 mg of ABTS in 1 mL of sterile distilled H2O for each reaction, including the standards. Then, we added 300 mg of MnO2 and mixed using a magnetic stirrer for 20 min. The mixture was filter sterilized through a 0.45 μm syringe filter. We used a spectrophotometer (Scientific Genesys 10S UV–VIS Spectrophotometer, Menlo Park, CA, USA) zeroed at 734 nm with 100 μL PBS added to 1 mL of sterile distilled H2O for measurements. The hydrophilic extracts were re-suspended in 10 mL of 75% Methanol. Then, 100 μL of the extract were used in a reaction with 1 mL of 5 mM ABTS for 2 min before reading absorbance at 734 nm. We adjusted the absorbance readings as necessary with 5 mM PBS to obtain an absorbance of 0.700. Later, we obtained the readings for Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) standards (0, 11, 22, 33, and 44 μmol/L) before obtaining the readings from the samples as described previously [30].

2.3. Seed Germination

Seeds from each of the six populations were surface sterilized with 50% ethanol and 25% sodium hypochlorite. Seeds were then sown on germination blotter sheets (Blue Steel, Anchor Paper, St. Paul, MN, USA) in transparent polystyrene germination boxes with tight-fitting covers (≈11 cm × 11 cm × 3.5 cm: 156C, Hoffman Manufacturing, Corvallis, OR, USA). Twenty-five seeds per population were used in each of the three replicates. Seeds that exhibited radicle and coleoptile emergence were scored as germinated, as described previously [40]. Seed germination assays were performed on distilled H2O containing 0.2% [v/v] of plant preservative mixture TM (PPM) solution (PPM, Plant Cell Technology Inc., Washington, DC, USA) and monitored daily. Germination experiments were conducted at 35/25 °C day/night temperatures with a 12 h photoperiod for 28 days inside a Percival incubator model I-30NL (Percival Scientific, Perry, IA, USA). The illumination coincided with the warmer temperature.

2.4. Tetrazolium (Tz) Staining

Tetrazolium staining tests were performed on ungerminated (seeds that failed to germinate after 28 days) seeds by puncturing them carefully at the distal end without damaging the embryo. Prepared seeds were placed in 5 mL glass vials before adding 200 µL of 1% [w/v] Tetrazolium and incubated at 30 ± 2 °C for 24 h [41]. Following incubation, seeds were dissected and microscopically examined for uniformly stained embryos on a dissecting microscope (Cole-Parmer® MSS-400 Series Stereozoom Microscope 0.65× to 5.5×, Antylia Scientific, Chicago, IL, USA) coupled to a Nikon camera equipped with NIS-Elements D530.02 imaging software at 0.63 to 2× magnification. Viable seeds, regarded as positive staining, comprised uniformly red to pink-stained embryos. In contrast, unviable seeds were unstained and/or partially stained [30]. Positively stained seeds were added to the germinated data to estimate the total seed viability.

2.5. Multispectral Imaging Analysis

We randomly obtained 100 seeds from each of the six sea oat populations for the multispectral analysis of physical traits. In this analysis, we captured the morphological (size, shape, orientation) and seed coat pigmentation traits using a multispectral machine vision system—VideometerLab4 (Videometer A/S, Herlev, Denmark) equipped with multispectral imaging (MSI), blob detection, image processing (IPT), and a classifier design tool (CDT). The system has 20 high-powered LED spectral bands ranging from 365 to 970 nm and a camera (ca. 30 µm/pixel resolution) to capture digital images (2992 × 4096 pixels) on a single seed basis.
We used a panel of color layers to mark the contrasting regions between the seeds and the background in the resulting images. We then used the MSI transmission builder tool and the instrument’s normalized canonical discriminant analysis (nCDA) to separate seeds from the background. Then, we performed simple threshold segmentation and extracted data on seed variables using the feature manager tools following the methods previously described [11,12,13,42]. Through the features manager tools, multiple (>50) parameters were computed and obtained as quantitative readouts for seed traits, including size (area, length, and aspect ratio), shape, orientation, and seed coat pigmentation traits—CIELab L*, CIELab A*, and CIELab B* (Commission on Illumination CIE Standards) [42,43]. In the CIE color space, CIELab L* represents the lightness from black to white, 0 = black and 100 = white; CIELab A* represents a unique color form between magenta and green, the red/green axis; and CIELab B* represents a unique color from blue to yellow, the yellow/blue axis. Other seed coat parameters obtained from the readouts included hue, saturation value (HSV), and color space based on red, green, and blue intensity [42,43], as previously described in seed quality, discrimination classification, and seed vigor assessments [12,13,14,44].

2.6. Seed Mass and Volume

We obtained similar seed samples by filling graduated microcentrifuge tubes to the 0.5 mL mark. Then, using a digital balance (Mettler Toledo Analytical Balance ME104T/00, Mettler Toledo, LLC, 1900 Polaris Parkway, Columbus, OH 43240, USA), we measured fresh mass and counted the total number of seeds per population using the counting feature. We then transferred samples to foil packets, placed all samples in a drying oven at 103 °C for 17 h, and collected dry mass measurements. We used five biological replicates for each population. Seed volume was estimated by using water displacement for volume estimation in graduated glass vials. We ensured the seeds were immersed in water by gentle agitation and centrifugation at 2000 RCF for 2 min. We obtained the volume estimation within four minutes of starting the process.

2.7. Respiration and Metabolic Analysis

We used a robotic oxygen-sensing respirometer (Seed Respiration Analyser, Fytagoras BV., Leiden, The Netherlands) for respiration and metabolic analysis of individual seeds. We assessed oxygen consumption on 48 randomly selected seeds from each of the six populations. We transferred seeds into 2 mL screw-cap vials containing 1.75 mL of agar (0.4% [w/v]) and 0.2% [v/v]) plant preservative mixture (PPM) using aseptic methods. We sealed the vials with caps containing an oxygen-sensing fluorescent polymer coating on the internal side. A robotic arm sequentially moved a sensor over each vial, measuring the oxygen concentration inside each vial. We programmed the system to collect measurements every 60 min to obtain time course changes in oxygen consumption for a total of 350 h. At the end of each experiment, vials were opened to confirm seed germination. Seeds that failed to germinate were excluded from the analysis. Raw data was used to generate indices for oxygen depletion and seed metabolic rates.

2.8. Controlled Deterioration Test

Seed aging stress was applied using high temperature and relative humidity conditions, as previously outlined by Newton et al. [37]. We placed ten open, sterile glass dishes, each containing 120 seeds, in a single layer on a plastic, vented tray within airtight, electrical enclosure boxes (OPCP303013T.U., ENSTO Ltd., Porvoo, Finland) containing 1L of the non-saturated LiCl-rehydration solution (385 g/L LiCl) [37,45]. We replicated this three times for seeds from each population. We then transferred these rehydration boxes into an incubator (I-30NL, Percival Scientific, Perry, IA, USA), set at 20 ± 2 °C, for 14 days. We used digital hygrometers (Tempo Plus 2, Blue-Maestro Ltd., London, UK) to monitor relative humidity (RH) changes inside the boxes. We maintained RH at 47% using procedures described in the protocol [37].
After 14 days, we transferred the seeds from the rehydration boxes to the aging boxes. The aging boxes contained 1L of a non-saturated LiCl solution (300 g/L LiCl) [37,45]. We then transferred the sealed aging boxes into an oven set at 45 °C, with the RH maintained at 60%. During the aging progression, we randomly withdrew seeds in three glass dishes on days 1, 2, 5, 9, 20, 30, 50, 75, 100, and 125, then used these seeds for seed germination and antioxidant quantification assays.

2.9. Data Analysis

2.9.1. Physical and Biochemical Data

We subjected multispectral imaging data to principal component analysis (PCA) using a suite of R packages comprising devtools, ggbioplot2, and prcom [46] in R software version 4.2.3 [47] to select traits for additional population comparisons. We then used contrasting traits from the PCA and other measured physical characteristics in an Analysis of Variance (ANOVA) with post hoc multiple comparisons to understand differences among the populations using the agricolae package in R software version 4.2.3 [47]. We further estimated the effect sizes, Eta2 (η2), calculated using the effectsize R package in R software version 4.2.3. The traits included in this analysis comprised seed mass, protein content, seed dimensions (volume, surface, area, length, width, and width: length, compactness circle (the ratio of the area of the object to the area of a circle with the same perimeter), compactness ellipse (the ratio of the area of the object to the area of an elliptical shape with the same perimeter), and BetaShapes a and b as previously described [45].

2.9.2. Aging Stress Survival Analysis

We fitted seed germination time courses from the seed deterioration assays with probit regression, then estimated Ki, σ (sigma), and p50 according to the Ellis and Roberts (1981) [48] viability equation, v = Ki − p/σ [49], where v equals the germination (in probits) of seeds after p days in the aging environment, i is the initial seed viability in probits (i.e., y-intercept), σ is the time for viability to fall by 1 probit, and p50 (the product of Ki and σ) equals the time for viability to decline to 50%.

2.9.3. Relationship of Seed Traits with Aging Stress Survival

We assessed relationships between aging stress survival and morphological, physical, and physiological traits with Pearson correlation analysis, which we implemented using the packages ggpubr, Hmisc, and corrplot in R software version 4.2.3 [47]. We used the following guidelines to describe relationship strength: 0.90 to 1.00 is very high, 0.70 to 0.89 is high, 0.50 to 0.69 is moderate, 0.30 to 0.49 is low, and 0.00 to 0.29 is negligible.

3. Results

3.1. Seeds Exhibited Variability of Physical Traits among Sea Oat Populations

To test the hypothesis of whether physical and biochemical traits could predict seed viability after storage, we collected seeds from six sea oat populations from the Atlantic and Gulf coastlines (Supplemental Table S1). Then, we stored these for four years at non-optimal storage conditions of 22 ± 2 °C and 50% relative humidity to test the extent to which simple analysis of physical and biochemical traits could predict post-storage viability. First, we germinated seeds to determine the effects of storage conditions on germination capacity (Figure 1a). We found that seeds collected at Von D. Mizell-Eula Johnson State Park (VEFL) showed the lowest germination percentage (75%), followed by Delnor-Wiggins State Park (DWFL) and Ft. Clinch State Park (FCFL), both with 85%. The remaining three populations, Ft. Pierce Inlet State Park (FPFL), Honeymoon Island State Park (HIFL), and Bill Baggs State Park (BBFL), showed more than a 93% germination rate (Figure 1a). These results indicate that non-optimal storage conditions impact seed viability.
Our analysis further identified three different groups based on seed surface area (effect size η2 = 0.40). FCFL and FPFL seeds had the largest surface area (6.11 mm2 and 6.02 mm2, respectively). Medium-sized seeds were seen in BBFL (5.54 mm2) and VEFL (5.37 mm2), while seeds with a smaller total surface area were common in HIFL (4.97) and DWFL (4.88 mm2). The effect size of these differences was medium (η2 = 0.40) (Figure 1b, Supplementary Datasheet S1, Supplementary Table S2).
Seeds from VEFL were longer (effect size; η2 = 0.19) than the other five populations (Figure 1c). No statistical differences in length among seeds from FCFL, FPFL, and HIFL were identified (Figure 1c). In terms of seed width, our results showed higher values in seeds from FPFL (1.81 mm) and medium-width sizes in seeds from FCFL (1.74 mm) and BBFL (1.73 mm). In comparison, lower width values were common in seeds from HIFL, DWF, and VEFL (1.49 mm, 1.49 mm, and 1.52 mm, respectively), and no statistical differences were observed (Figure 1d, Supplementary Datasheet S1).
Seed volume significantly varied (effect size; η2 = 0.92) across populations (Figure 1e). Seeds from FCFL and FPFL exhibited the largest volumes, which differed statistically from the volume of seeds from the remaining populations (Figure 1e). Furthermore, VEFL seed volume was greater than that observed in seeds collected from BBFL, DWFL, and HIFL. Seeds from DWFL and HIFL displayed the smallest volumes of any population (Figure 1e).
Next, we calculated the surface-to-area ratio. We found that this ratio was statistically significant (effect size; η2 = 0.80) in seeds from DWFL and HIFL compared to the other populations (Figure 1f). Quantification of the fresh and dry mass showed similar patterns among the populations (Figure 1g,h). Seeds from FPFL had a statistically higher mass than the other five populations. Mass was consistently lowest in seeds from DWFL (Figure 1g,h). Several other seed shape characteristics tended to vary from one population to another (Supplementary Table S1). In summary, our data show that sea oat populations from the Atlantic and Gulf coastlines have a large variability in seed physical traits.
We then performed a principal component analysis to identify physical traits that could explain variability in germination after non-optimal storage conditions. The first three principal components explained about 73% of the physical variation among seed collections (Supplementary Figure S1a). Seed surface area, width, width–length, and compactness were the major components for the first principal component (PCA 1). In contrast, seed length, BetaShape a, and BetaShape b were the major components for the second principal component (PCA 2, Supplementary Figure S1b). Seed shape characteristics, compactness, ellipse, and vertical skewness were the major components for the third principal component (PCA 3, Supplementary Datasheet S1). Therefore, PCA 1 consisted of the seed size and shape characteristics, while PCA 2 and PCA 3 consisted of measures of seed shape characteristics (Supplementary Figure S1b, Supplementary Datasheet S1). Seeds from FCFL, FPFL, and BBFL could be classified differently from VEFL, HIFL, and DWFL based on parameters in the first principal component (seed size and shape) (Supplementary Figure S1c). Seeds from HIFL and DWFL exhibited similar shape characteristics and were slightly different from seeds from VEFL (Supplementary Figure S1c).
In terms of the seed coat pigmentation traits, seeds were classified based on the color space standards CIELab L*, CIELab A*-CIELab B*, saturation, and the amount of lightness. Two principal components explained about 79% of the seed coat variations (Supplementary Figure S1d). Seed coat color saturation, CIELab A*, and CIELab B* were the major components for PCA 1. On the other hand, CIELab L* and Hue were the significant components of PCA 2 (Supplementary Figure S1e). Therefore, based on seed coat pigment characteristics among the seeds from the six populations, the first principal component (PCA 1) exhibited the uniqueness of the color spectrum and color intensity of the seed coat pigmentation. In contrast, the second principal component (PCA 2) consisted of the dominant colors and the lightness value of the seed coat color (Supplementary Figure S1e, Supplementary Datasheet S1). Seeds from FCFL and BBFL could be classified differently from those from VEFL and HIFL based on the seed coat color spectrum and color saturation. However, seeds from FPFL and DWFL grouped together in the color spectrum and saturation (Supplementary Figure S1f). Seeds from both VEFL and DWFL exhibited similar dominant color and lightness values, which were slightly varied from those in HIFL (Supplementary Figure S1f).

3.2. Respiration and Metabolic Characteristics Vary among Seeds from Sea Oat Populations

Our metabolic analysis showed significantly higher starting metabolism rates (SMR) in seeds from DWFL (4.55 ± 0.21 VO2 µL/mg DW) and VEFL (4.40 ± 0.02 VO2 µL/mg DW) than any of the other populations (effect size; η2 = 0.94) (Figure 2a). Moderate SMR was quantified in BBFL (4.09 ± 0.10 VO2 µL/mg DW) and HIFL (3.73 ± 0.09 VO2 µL/mg DW) (Figure 2a). The lowest SRM was measured in seeds collected from FCFL (3.44 ± 0.06 VO2 µL/mg DW) and FPFL (3.21 ± 0.05 VO2 µL/mg DW) (Figure 2a).
Increasing metabolism time (IMT) was longer in HIFL (29.70 h), DWFL (28.40 h), and BBFL (28.17 h), moderate in VEFL (27.86 h) and FPFL (26.10 h), and short in FCFL (25.50 h) (Table 1; Supplementary Datasheet S2). The seed oxygen metabolism rate (OMR) was also variable in the six populations (Figure 2b, Table 1). The OMR was significantly higher in DWFL than in all five other populations (Figure 2b). The OMR was similar in BBFL and VEFL, and the level in these two populations was also significantly higher than FCFL, HIFL, and FPFL. FPFL recorded the lowest OMR (effect size; η2 = 0.92) (Figure 2b). The intercept at the OMR (RGT-Relative Germination Time) was larger in HIFL (383.43 h) and DWFL (343.28 h), moderate in VEFL (318.05 h), BBFL (290.08 h), and smaller in both FCFL (249.39 h) and FPFL (249.09 h) (Table 1).
Quantification of the total protein content showed a higher level in BBFL seeds compared to other populations (effect size; η2 = 0.91). The total protein content was low in FCFL, FPFL, and VEFL seeds, with no statistically significant variations among these populations (Figure 2c). Taken together, our results indicate that non-optimal storage conditions impact biochemical traits.
The median times (t50) for oxygen depletion to reach 75% of the original were significantly (p-value < 0.05) shorter in both FCFL and FPFL (<100 h) than in the other four populations (BBFL, DWFL, HIFL, and VEFL) (effect size; η2 = 0.15) (Figure 3a–g). However, the t50 for oxygen depletion to 50% of the original was significantly (p-value < 0.05) slower in FCFL than in other populations (effect size; η2 = 0.05) (Figure 3a–f,h). The t50 for the oxygen depletion level to 25% of the original levels was shorter in FPFL seeds than in FCFL (Figure 3a–f,i). However, no statistical differences were observed (effect size; η2 = 0.05) (Figure 3i). The mean time for oxygen depletion to reach 50% of the original level (1/R50) was extended in HIFL (152.81 h), moderate in VEFL (143.02 h), DWFL (141.52 h), BBFL (137.77 h), and FCFL (132.92 h), but lower in FPFL (119.02 h) (Table 1). Areas under the curve for the time to 50% oxygen depletion (AUC50) were high in HIFL (128.09 h), moderate in DWFL (118.22 h), VEFL (117.99 h), BBFL (111.23 h), FCFL (105.83 h), and lower in FPFL (96.21 h) (Table 1; Supplementary Datasheet S2).

3.3. Seed Aging Stress Deterioration Varies among Sea Oat Populations

To further understand how non-optimal storage conditions may change seed viability and whether it could be predicted using biochemical traits, we used an aging stress deterioration strategy. At the beginning of the aging experiment, seed germination and viability were comparable among the six populations (Figure 4a). All populations attained maximum germination on day twelve. However, seeds from VEFL displayed slower germination than seeds from other populations (Figure 4a). There were no statistical differences in the incidences of pest damage to the seeds among the six populations (Figure 4b; Supplementary Datasheet S3). After excluding pest-damaged seeds, we did not find a statistical difference in seed viability among the populations (Figure 4c).
The time taken for seed germination to be reduced by 50% among sea populations varied widely (Figure 5a–f). Seeds from BBFL, HIFL, and DWFL took longer (26.4, 25.5, and 25 days, respectively) (Figure 5a,b,e) than the FPFL population (23.5 days) (Figure 5d). However, the most dramatic rapid loss of germination was exhibited in FCFL and VEFL (11.5 and 13.2 days, respectively) (Figure 5c,f). The range of estimated stress deterioration (1/p50) was 0.038 to 0.087. The aging stress deterioration effect was very strong in FCFL and VEFL (0.087 and 0.076, respectively) compared to all the other populations (η2 = 0.90, CI [0.76, 1.00]) (Figure 5a–f, Table 2, Supplementary Datasheet S3).

3.4. Depletion of Antioxidants Accentuates as Aging Conditions Progresses

To assess whether an accelerated seed deterioration process results in a faster antioxidant reduction, we quantified the TEAC during the aging progression. All six populations registered antioxidant depletion. However, their patterns differed among the populations (Figure 6). While seeds from BBFL, DWFL, and FPFL registered rapid terminal TEAC depletion between 50 and 75 days of high-temperature aging stress, seeds from FCFL and HIFL registered rapid terminal TEAC depletion between 30 and 50 days. Some seeds had a spike in TEAC levels occurring early at two days of high-temperature stress, such as in seeds from DWFL and HIFL (Figure 6b,e), and others at nine days, including FCFL and FPFL (Figure 6c,d). Furthermore, seeds from both BBFL and VEFL had a considerable delay in TEAC spike to 20 days of high-temperature aging stress (Figure 6).

3.5. Physical and Biochemical Traits of Seeds Predict Aging Stress Survival

We statistically compared the biochemical and physical data to predict seed viability after non-optimal storage conditions. The time taken by seeds for germination to be reduced by 50% (p50) following aging stress had a positive correlation with baseline TEAC (r = 0.39, p-value ≤ 0.001), seed aspect ratio (r = 0.20, p-value ≤ 0.001), seed surface area to volume ratio (r = 0.73, p-value ≤ 0.001), seed shape characteristics: compactness circle (r = 0.16, p-value ≤ 0.001), BetaShape a (r = 0.11, p-value ≤ 0.05), BetaShape b (r = 0.08, p-value ≤ 0.05), and vertical orientation (r = 0.09, p-value ≤ 0.05) (Figure 7, Supplementary Datasheet S3). Other physical traits that positively correlated with aging stress survival included the seed coat pigmentation characteristics comprised the CIElab L* (r = 0.08, p-value ≤ 0.05) (Figure 7). In terms of the seed oxygen consumption and metabolic traits, the time for oxygen depletion to 50% (R50) (r = 0.11, p-value ≤ 0.01), the starting metabolism rate (SMR) (r = 0.09, p-value ≤ 0.05), the oxygen metabolism rate (OMR) (r = 0.09, p-value ≤ 0.05), and the increased metabolic time (IMT) (r = 0.58, p-value ≤ 0.001) were all positively associated with increased viability under aging stress conditions (Figure 7; Supplementary Datasheet S4).
On the other hand, several seed physical and biochemical traits negatively correlated with aging stress survival (Figure 7), including total surface area (r = −0.26, p-value ≤ 0.001), length (r = −0.33, p-value ≤ 0.001), volume (r = −0.63, p-value ≤ 0.001), and the compactness ellipse (r = −0.15, p-value ≤ 0.001). At the biochemical level, the traits that had a negative correlation with viability under aging stress conditions comprised the TEAC level at 36 h of imbibition (r = −0.22, p-value ≤ 0.001), seed fresh mass (r = −0.27, p-value ≤ 0.001), seed dry mass (r = −0.27, p-value ≤ 0.001), and the seed protein content (r = −0.33, p-value ≤ 0.001) (Figure 7; Supplementary Datasheet S2). Taken together, our results show that physical and biochemical seeds are good indicators of seed viability under non-optimal storage conditions.

4. Discussion

Our study indicates that biochemical and physical traits correlate with the aging stress survival of seeds under a controlled deterioration test. Our results further demonstrate the potential of using quantifiable traits in seed collection as a screening strategy for ecologically important species (Uniola paniculata L.), where variability is larger than in agronomical crops. Our results also indicated significant similarities in seed viability at the beginning of the aging stress analysis (Figure 4 and Figure 5). The aging stress progressed over time, leading to variable patterns of decline in seed viability, especially in seeds from FCFL and VEFL populations (Figure 5).
Our results showed that seed total surface area and surface-to-volume ratio differed among seeds from the six populations (Figure 1), and these differences positively correlated with the time taken for the seed germination to reduce to 50% under high-temperature aging stress (Figure 7). A larger total surface area-to-volume ratio is common in small seeds. Such small seeds quickly respond to fluctuating environments and easily reactivate metabolic activities following a temporal increase in relative humidity. They can facilitate both metabolic and enzymatic function and, in so doing, enable rapid repair of the damaged proteins, allowing for the synthesis of new enzymes and restoration of the DNA and RNA that are key in ensuring prolonged seed viability [2,3,8,24,25]. In Copaifera langsdorffii, smaller seeds exhibited faster germination-producing seedlings with rapid root development, while large seeds produced slower-growing seedlings that took longer to grow roots and establish [50]. Hence, seed size affected the time taken for seedling establishment, making smaller seeds better suited for environments with changing moisture and temperature conditions compared to their larger counterparts [50]. Smaller seeds have been described as exhibiting better fitness compared to their larger counterparts [2,5,49,50]. Thus, this has implications for seed fate by determining seed predation and seed bank persistence.
The increased metabolic time (IMT) was positively correlated with aging stress survival. The IMT corresponds to changes in metabolism during imbibition (Table 1). Rapid metabolic response to changing conditions is critical for survival under stress, especially in facilitating the rapid repair of proteins [5]. As such, seeds with a long IMT may have better metabolic responses, allowing coordinated reactivation of metabolic activities and repair of the damaged proteins. Furthermore, our results showed higher metabolism values in seeds with a smaller total surface area and smaller volume than in seeds with a larger surface area and more significant volume (Figure 2 and Figure 3). This relationship agrees with the expectation of body size function on metabolism; large organisms exhibit lower metabolic rates than smaller organisms [51]. Our results show poor oxygen depletion, especially in FCFL compared to FPFL, two populations with larger seeds. Large seeds could exhibit slower responses to environmental fluctuations, such as hydration-related relative humidity change, at the physiological level than smaller seeds.
On the other hand, smaller seeds may be more exposed to damage from ROS generated by such environmental fluctuations [52]. Therefore, there may be complex variations in seed protection mechanisms, with smaller seeds relying more on metabolic-driven mechanisms while larger seeds depend more on storage reserves. Considering that metabolic rates have implications for seed viability and that reduced metabolic rates correspond to declining viability, we see a link between poor metabolism and poor viability under the aging stress conditions exhibited in our study populations. Interestingly, physical traits relating to seed size, seed volume, area, and length negatively correlated with aging stress survival, relating to the metabolic characteristics of large vs. small seeds.
We found a positive correlation between the seed baseline antioxidants and aging stress survival. These indicated the protective nature of the storage antioxidants on aging stress survival. Many studies show that antioxidants prevent seed damage and enable seed longevity [3,25,27,53]. On the other hand, our results also showed that a spike in antioxidant levels following 36 h of imbibition had a negative correlation with aging stress survival; this could imply a likelihood of high reactive oxygen species and possibly high deterioration in some of the seeds that explain the reduced potential for aging stress survival. In studies of seed aging in wheat and barley, loss of seed viability was associated with the accumulation of reactive oxygen species [54]. Our data show a progression in antioxidant depletion in seeds with time under aging stress conditions (Figure 6). This agrees with other reports and reviews that described the depletion of antioxidants during seed aging [25,26].
Seed mass and protein content had a weak negative correlation with aging stress survival (Figure 7). Seed mass and seed storage reserves are crucial energy and biochemical reserves that drive seed functions and success and ensure extended seed longevity [2,3,24]. However, the critical components of the storage reserves vary due to the environmental impact on seed development and the genetic influence of the mother plant [29,31]. We had earlier observed the impact of the environment on the antioxidant component among these populations [30]. Other studies have also reported evidence of the effects of the environment on seed protein accumulation and composition [55]. Therefore, it is likely difficult to determine the association between the prevailing seed reserve state and aging stress survival in our study populations. Some of the populations in the study with lower dry mass exhibited higher protein concentrations (Figure 2), which could be evidence of the differential accumulation of seed storage reserves.

5. Conclusions

Seeds from sea oats exhibited physical differences in the total surface area, length, width, aspect ratio, compactness, and several other shape and seed coat pigmentation characteristics. These seeds also displayed different metabolic traits. Both physical and biochemical features are correlated with aging stress survival. Stable TEAC levels between the baseline and 36 h of pre-germination hydration were typical for seeds with extended survival, while those with unstable TEAC displayed reduced aging stress survival. Furthermore, seed baseline antioxidants, surface area to volume ratio, and increased metabolism time were better positive predictors of extended aging stress survival among the study population in sea oats. Seed metabolism and nutrient components are also likely to impact the use of antioxidants, and cases of antioxidant spikes following hydration indicate compensatory synthesis for removing excessive ROS.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/agriculture14060875/s1, Figure S1: Principal component analysis of seed physical traits among sea oat populations from the Atlantic and Gulf Coastlines of Florida; Table S1: Baseline germination and other traits of seeds used in the six sea oats populations in the aging experiment stress experiment; Table S2: Morphological features of seeds from the six sea oat populations used in the aging stress experiment; Supplementary Datasheet S1: Multispectal Imaging raw data. Datasheet S2: biochemical measurements raw data. Datasheet S3: Aging stress survivial raw data. Datasheet S4: Correlation analyses.

Author Contributions

A.O.E.: data curation, formal analysis, investigation, methodology, visualization, writing—original draft preparation, and writing—review and editing. M.T.D.: investigation and methodology. H.E.P.: conceptualization, funding acquisition, supervision, and writing—review and editing. K.B.: conceptualization, funding acquisition, supervision, and writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the U.S. Department of Commerce Sea Grant Program funding (Grant number SINERR-2018-8). The USDA National Institute of Food and Agriculture, Hatch Project FLA-ENH-005853, partially supported this work.

Institutional Review Board Statement

Not applicable.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors on request.

Acknowledgments

We thank Tia Tyler for the technical assistance during the seed collection.

Conflicts of Interest

The authors declare that this research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Bewley, J.D.; Bradford, K.J.; Hilhorst, H.W.M.; Nonogaki, H. Seeds; Springer: New York, NY, USA, 2013; ISBN 978-1-4614-4692-7. [Google Scholar]
  2. Saatkamp, A.; Cochrane, A.; Commander, L.; Guja, L.K.; Jimenez-Alfaro, B.; Larson, J.; Nicotra, A.; Poschlod, P.; Silveira, F.A.O.; Cross, A.T.; et al. A Research Agenda for Seed-trait Functional Ecology. New Phytol. 2019, 221, 1764–1775. [Google Scholar] [CrossRef] [PubMed]
  3. Rajjou, L.; Debeaujon, I. Seed Longevity: Survival and Maintenance of High Germination Ability of Dry Seeds. C. R. Biol. 2008, 331, 796–805. [Google Scholar] [CrossRef] [PubMed]
  4. Ellis, R.H. Temporal Patterns of Seed Quality Development, Decline, and Timing of Maximum Quality during Seed Development and Maturation. Seed Sci. Res. 2019, 29, 135–142. [Google Scholar] [CrossRef]
  5. Long, R.L.; Gorecki, M.J.; Renton, M.; Scott, J.K.; Colville, L.; Goggin, D.E.; Commander, L.E.; Westcott, D.A.; Cherry, H.; Finch-Savage, W.E. The Ecophysiology of Seed Persistence: A Mechanistic View of the Journey to Germination or Demise: The Ecophysiology of Seed Persistence. Biol. Rev. 2015, 90, 31–59. [Google Scholar] [CrossRef] [PubMed]
  6. Walters, C. Orthodoxy, Recalcitrance and in-between: Describing Variation in Seed Storage Characteristics Using Threshold Responses to Water Loss. Planta 2015, 242, 397–406. [Google Scholar] [CrossRef] [PubMed]
  7. Walters, C.; Berjak, P.; Pammenter, N.; Kennedy, K.; Raven, P. Preservation of Recalcitrant Seeds. Science 2013, 339, 915–916. [Google Scholar] [CrossRef] [PubMed]
  8. Walters, C.; Ballesteros, D.; Vertucci, V.A. Structural Mechanics of Seed Deterioration: Standing the Test of Time. Plant Sci. 2010, 179, 565–573. [Google Scholar] [CrossRef]
  9. Kibinza, S.; Vinel, D.; Côme, D.; Bailly, C.; Corbineau, F. Sunflower Seed Deterioration as Related to Moisture Content during Ageing, Energy Metabolism and Active Oxygen Species Scavenging. Physiol. Plant. 2006, 128, 496–506. [Google Scholar] [CrossRef]
  10. Ballesteros, D.; Walters, C. Solid-State Biology and Seed Longevity: A Mechanical Analysis of Glasses in Pea and Soybean Embryonic Axes. Front. Plant Sci. 2019, 10, 920. [Google Scholar] [CrossRef] [PubMed]
  11. Olesen, M.H.; Nikneshan, P.; Shrestha, S.; Tadayyon, A.; Deleuran, L.C.; Boelt, B.; Gislum, R. Viability Prediction of Ricinus Cummunis L. Seeds Using Multispectral Imaging. Sens. Switz. 2015, 15, 4592–4604. [Google Scholar] [CrossRef] [PubMed]
  12. Rego, C.H.Q.; França-Silva, F.; Gomes-Junior, F.G.; de Moraes, M.H.D.; de Medeiros, A.D.; Silva, C.B. da Using Multispectral Imaging for Detecting Seed-Borne Fungi in Cowpea. Agriculture 2020, 10, 361. [Google Scholar] [CrossRef]
  13. ElMasry, G.; Mandour, N.; Wagner, M.-H.; Demilly, D.; Verdier, J.; Belin, E.; Rousseau, D. Utilization of Computer Vision and Multispectral Imaging Techniques for Classification of Cowpea (Vigna Unguiculata) Seeds. Plant Methods 2019, 15, 24. [Google Scholar] [CrossRef] [PubMed]
  14. Mihailova, A.; Liebisch, B.; Islam, M.D.; Carstensen, J.M.; Cannavan, A.; Kelly, S.D. The Use of Multispectral Imaging for the Discrimination of Arabica and Robusta Coffee Beans. Food Chem. X 2022, 14, 100325. [Google Scholar] [CrossRef] [PubMed]
  15. Duncan, K.E.; Czymmek, K.J.; Jiang, N.; Thies, A.C.; Topp, C.N. X-Ray Microscopy Enables Multiscale High-Resolution 3D Imaging of Plant Cells, Tissues, and Organs. Plant Physiol. 2022, 188, 831–845. [Google Scholar] [CrossRef] [PubMed]
  16. Salimi, Z.; Boelt, B. Classification of Processing Damage in Sugar Beet (Beta Vulgaris) Seeds by Multispectral Image Analysis. Sens. Switz. 2019, 19, 2360. [Google Scholar] [CrossRef] [PubMed]
  17. Vrešak, M.; Halkjaer Olesen, M.; Gislum, R.; Bavec, F.; Ravn Jørgensen, J. The Use of Image-Spectroscopy Technology as a Diagnostic Method for Seed Health Testing and Variety Identification. PLoS ONE 2016, 11, e0152011. [Google Scholar] [CrossRef] [PubMed]
  18. Hansen, M.A.E.; Hay, F.R.; Carstensen, J.M. A Virtual Seed File: The Use of Multispectral Image Analysis in the Management of Genebank Seed Accessions. Plant Genet. Resour. 2016, 14, 238–241. [Google Scholar] [CrossRef]
  19. Boelt, B.; Shrestha, S.; Salimi, Z.; Jørgensen, J.R.; Nicolaisen, M.; Carstensen, J.M. Multispectral Imaging—A New Tool in Seed Quality Assessment? Seed Sci. Res. 2018, 28, 222–228. [Google Scholar] [CrossRef]
  20. Mahajan, P.V.; Luca, A.; Edelenbos, M. Development of a Small and Flexible Sensor-Based Respirometer for Real-Time Determination of Respiration Rate, Respiratory Quotient and Low O2 Limit of Fresh Produce. Comput. Electron. Agric. 2016, 121, 347–353. [Google Scholar] [CrossRef]
  21. Bello, P.; Bradford, K.J. Single-Seed Oxygen Consumption Measurements and Population-Based Threshold Models Link Respiration and Germination Rates under Diverse Conditions. Seed Sci. Res. 2016, 26, 199–221. [Google Scholar] [CrossRef]
  22. Silva, J.G.D.; Gadotti, G.I.; Moraes, D.M.D.; Silva, A.H.M.; Cavalcante, J.A.; Meneghello, G.E. Equipment to Assess Vigor in Soybean Seeds Using CO2 Produced during Respiration. Rev. Bras. Eng. Agríc. E Ambient. 2021, 25, 353–358. [Google Scholar] [CrossRef]
  23. Dalziell, E.L.; Tomlinson, S. Reduced Metabolic Rate Indicates Declining Viability in Seed Collections: An Experimental Proof-of-Concept. Conserv. Physiol. 2017, 5, cox058. [Google Scholar] [CrossRef]
  24. Sano, N.; Rajjou, L.; North, H.M.; Debeaujon, I.; Marion-Poll, A.; Seo, M. Staying Alive: Molecular Aspects of Seed Longevity. Plant Cell Physiol. 2016, 57, 660–674. [Google Scholar] [CrossRef]
  25. Stegner, M.; Wagner, J.; Roach, T. Antioxidant Depletion during Seed Storage under Ambient Conditions. Seed Sci. Res. 2022, 32, 150–156. [Google Scholar] [CrossRef]
  26. Pehlivan, F.E. Free Radicals and Antioxidant System in Seed Biology. In Advances in Seed Biology; Jimenez-Lopez, J.C., Ed.; InTech: London, UK, 2017; ISBN 978-953-51-3621-7. [Google Scholar]
  27. Parkhey, S.; Naithani, S.C.; Keshavkant, S. ROS Production and Lipid Catabolism in Desiccating Shorea Robusta Seeds during Aging. Plant Physiol. Biochem. 2012, 57, 261–267. [Google Scholar] [CrossRef] [PubMed]
  28. Begcy, K.; Walia, H. Drought Stress Delays Endosperm Development and Misregulates Genes Associated with Cytoskeleton Organization and Grain Quality Proteins in Developing Wheat Seeds. Plant Sci. 2015, 240, 109–119. [Google Scholar] [CrossRef] [PubMed]
  29. Chen, C.; Begcy, K.; Liu, K.; Folsom, J.J.; Wang, Z.; Zhang, C.; Walia, H. Heat Stress Yields a Unique MADS Box Transcription Factor in Determining Seed Size and Thermal Sensitivity. Plant Physiol. 2016, 171, 606–622. [Google Scholar] [CrossRef] [PubMed]
  30. Egesa, A.O.; Pérez, H.E.; Begcy, K. Environmental Conditions Predetermine Quality, Germination, and Innate Antioxidants Pool in Sea oats (Uniola Paniculata L.) Seeds. Front. Environ. Sci. 2023, 11, 1263300. [Google Scholar] [CrossRef]
  31. Genna, N.G.; Walters, C.; Pérez, H.E. Viability and Vigour Loss during Storage of Rudbeckia Mollis Seeds Having Different Mass: An Intra-Specific Perspective. Seed Sci. Res. 2020, 30, 122–132. [Google Scholar] [CrossRef]
  32. Satya Srii, V.; Nagarajappa, N.; Vasudevan, S.N. Is Seed Coat Structure at Fault for Altered Permeability and Imbibition Injury in Artificially Aged Soybean Seeds? Crop Sci. 2022, 62, 1573–1583. [Google Scholar] [CrossRef]
  33. Pérez, H.E.; Kane, M.E. Different Plant Provenance Same Seed Tolerance to Abiotic Stress: Implications for Ex Situ Germplasm Conservation of a Widely Distributed Coastal Dune Grass (Uniola Paniculata L.). Plant Growth Regul. 2017, 82, 123–137. [Google Scholar] [CrossRef]
  34. Lonard, R.I.; Judd, F.W.; Stalter, R. Biological Flora of Coastal Dunes and Wetlands: Uniola paniculata L. J. Coast. Res. 2011, 276, 984–993. [Google Scholar] [CrossRef]
  35. Center for Plant Conservation. CPC Best Plant Conservation Practices to Support Species Survival in the Wild; Center for Plant Conservation: Escondido, CA, USA, 2018. [Google Scholar]
  36. Bonner, F.T. Storage of Seeds. In The Woody Plant Seed Manual; Bonner, F.T., Karrfalt, R.P., Eds.; Agricultural Handbook 727; USDA: Washington, DC, USA, 2008; pp. 85–97. [Google Scholar]
  37. Newton, R.; Hay, F.; Probert, R. Protocol for Comparative Seed Longevity Testing; Millennium Seed Bank Partnership: Ardingly, UK, 2014. [Google Scholar]
  38. Kalaman, H.; Wilson, S.B.; Mallinger, R.E.; Knox, G.W.; Kim, T.; Begcy, K.; Van Santen, E. Evaluation of Native and Nonnative Ornamentals as Pollinator Plants in Florida: II. Floral Resource Value. HortScience 2022, 57, 137–143. [Google Scholar] [CrossRef]
  39. Re, R.; Pellegrini, N.; Proteggente, A.; Pannala, A.; Yang, M.; Rice-Evans, C. Antioxidant Activity Applying an Improved ABTS Radical Cation Decolorization Assay. Free Radic. Biol. Med. 1999, 26, 1231–1237. [Google Scholar] [CrossRef] [PubMed]
  40. Begcy, K.; Sandhu, J.; Walia, H. Transient Heat Stress During Early Seed Development Primes Germination and Seedling Establishment in Rice. Front. Plant Sci. 2018, 9, 1768. [Google Scholar] [CrossRef] [PubMed]
  41. Peters, J.; Lanham, B. Tetrazolium Testing Handbook, Contribution No. 29 to the Handbook on Seed Testing; Association of Official Seed Analysts: Wichita, KS, USA, 2000. [Google Scholar]
  42. Cantrell, K.; Erenas, M.M.; De Orbe-Payá, I.; Capitán-Vallvey, L.F. Use of the Hue Parameter of the Hue, Saturation, Value Color Space As a Quantitative Analytical Parameter for Bitonal Optical Sensors. Anal. Chem. 2010, 82, 531–542. [Google Scholar] [CrossRef] [PubMed]
  43. Ly, B.C.K.; Dyer, E.B.; Feig, J.L.; Chien, A.L.; Del Bino, S. Research Techniques Made Simple: Cutaneous Colorimetry: A Reliable Technique for Objective Skin Color Measurement. J. Invest. Dermatol. 2020, 140, 3–12.e1. [Google Scholar] [CrossRef] [PubMed]
  44. Hu, X.; Yang, L.; Zhang, Z. Non-Destructive Identification of Single Hard Seed via Multispectral Imaging Analysis in Six Legume Species. Plant Methods 2020, 16, 116. [Google Scholar] [CrossRef] [PubMed]
  45. Hay, F.R.; Adams, J.; Manger, K.; Probert, R. The Use of Non-Saturated Lithium Chloride Solutions for Experimental Control of Seed Water Content. Seed Sci. Technol. 2008, 36, 737–746. [Google Scholar] [CrossRef]
  46. Harvey, D.T.; Hanson, B.A. A Comparison of Functions for PCA. Available online: https://cran.r-project.org/web/packages/LearnPCA/vignettes/Vig_07_Functions_PCA.pdf (accessed on 20 February 2024).
  47. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2021. [Google Scholar]
  48. Ellis, R.H.; Roberts, E.H. The Quantification of Ageing and Survival in Orthodox Seeds. Seed Sci. Technol. 1981, 9, 373–409. [Google Scholar]
  49. Fricke, E.C.; Wright, S.J. The Mechanical Defence Advantage of Small Seeds. Ecol. Lett. 2016, 19, 987–991. [Google Scholar] [CrossRef] [PubMed]
  50. Souza, M.L.; Fagundes, M. Seed Size as Key Factor in Germination and Seedling Development of Copaifera langsdorffii (Fabaceae). Am. J. Plant Sci. 2014, 05, 2566–2573. [Google Scholar] [CrossRef]
  51. Anderson-Teixeira, K.J.; Savage, V.M.; Allen, A.P.; Gillooly, J.F. Allometry and Metabolic Scaling in Ecology. In Encyclopedia of Life Sciences; Wiley: Hoboken, NJ, USA, 2009; ISBN 978-0-470-01617-6. [Google Scholar]
  52. Carta, A.; Bottega, S.; Spanò, C. Aerobic Environment Ensures Viability and Anti-Oxidant Capacity When Seeds Are Wet with Negative Effect When Moist: Implications for Persistence in the Soil. Seed Sci. Res. 2018, 28, 16–23. [Google Scholar] [CrossRef]
  53. Wiebach, J.; Nagel, M.; Börner, A.; Altmann, T.; Riewe, D. Age-dependent Loss of Seed Viability Is Associated with Increased Lipid Oxidation and Hydrolysis. Plant Cell Environ. 2020, 43, 303–314. [Google Scholar] [CrossRef] [PubMed]
  54. Roach, T.; Nagel, M.; Börner, A.; Eberle, C.; Kranner, I. Changes in Tocochromanols and Glutathione Reveal Differences in the Mechanisms of Seed Ageing under Seedbank Conditions and Controlled Deterioration in Barley. Environ. Exp. Bot. 2018, 156, 8–15. [Google Scholar] [CrossRef]
  55. Triboi, E. Environmentally-Induced Changes in Protein Composition in Developing Grains of Wheat Are Related to Changes in Total Protein Content. J. Exp. Bot. 2003, 54, 1731–1742. [Google Scholar] [CrossRef] [PubMed]
Agriculture 14 00875 g001
Figure 1. Seed physical traits differ in sea oats. (a) Seed germination. (b) Total seed surface area. (c) Seed length. (d) Seed width. (e) Seed volume. (f) Seed surface area to volume ratio. (g) Seed fresh mass. (h) Seed dry mass. A different letter indicates significant differences within a figure based on the one-way ANOVA at an alpha of 0.05 for the comparison among the six populations. The same letters indicate no significant differences; n = 100.
Figure 1. Seed physical traits differ in sea oats. (a) Seed germination. (b) Total seed surface area. (c) Seed length. (d) Seed width. (e) Seed volume. (f) Seed surface area to volume ratio. (g) Seed fresh mass. (h) Seed dry mass. A different letter indicates significant differences within a figure based on the one-way ANOVA at an alpha of 0.05 for the comparison among the six populations. The same letters indicate no significant differences; n = 100.
Agriculture 14 00875 g001
Agriculture 14 00875 g002
Figure 2. Variation in seed metabolism and protein content in sea oats. (a) Starting metabolic rate (SMR), (b) oxygen metabolic rate (OMR), and (c) seed total proteins. Seed metabolic rates were estimated on a mean seed dry weight basis (n = 48). A different letter indicates significant differences within a figure based on the one-way ANOVA at an alpha of 0.05 for the comparison among the six populations. The same letters indicate no significant differences.
Figure 2. Variation in seed metabolism and protein content in sea oats. (a) Starting metabolic rate (SMR), (b) oxygen metabolic rate (OMR), and (c) seed total proteins. Seed metabolic rates were estimated on a mean seed dry weight basis (n = 48). A different letter indicates significant differences within a figure based on the one-way ANOVA at an alpha of 0.05 for the comparison among the six populations. The same letters indicate no significant differences.
Agriculture 14 00875 g002
Agriculture 14 00875 g003
Figure 3. Changes in oxygen depletion traits in sea oats. Oxygen depletion curves to 75, 50, and 25% of original oxygen levels (a–f) and comparisons of the median times (t50) required by the seed populations to reduce oxygen levels to 75, 50, or 25% of original oxygen levels (g–i). For figures (g–i), a different letter indicates significant differences within a figure based on the one-way ANOVA at an alpha of 0.05 for the comparison among the six populations. The same letters indicate no significant differences (n = 48).
Figure 3. Changes in oxygen depletion traits in sea oats. Oxygen depletion curves to 75, 50, and 25% of original oxygen levels (a–f) and comparisons of the median times (t50) required by the seed populations to reduce oxygen levels to 75, 50, or 25% of original oxygen levels (g–i). For figures (g–i), a different letter indicates significant differences within a figure based on the one-way ANOVA at an alpha of 0.05 for the comparison among the six populations. The same letters indicate no significant differences (n = 48).
Agriculture 14 00875 g003
Agriculture 14 00875 g004
Figure 4. Seed viability characteristics of sea oats at the beginning of the aging stress experiment. (a) Germination percentage. (b) Pathogen damage. (c) Viability. Germination was performed on 25 seeds in three replicates over 28 days, and viability comprises seed germination and a positive Tz test on non-germinated seeds at 28 days with the exclusion of pest and pathogen attacks (n = 75). A different letter and * indicate significant differences within a figure based on the one-way ANOVA at an alpha of 0.05 for the comparison among the six populations. The same letters indicate no significant differences.
Figure 4. Seed viability characteristics of sea oats at the beginning of the aging stress experiment. (a) Germination percentage. (b) Pathogen damage. (c) Viability. Germination was performed on 25 seeds in three replicates over 28 days, and viability comprises seed germination and a positive Tz test on non-germinated seeds at 28 days with the exclusion of pest and pathogen attacks (n = 75). A different letter and * indicate significant differences within a figure based on the one-way ANOVA at an alpha of 0.05 for the comparison among the six populations. The same letters indicate no significant differences.
Agriculture 14 00875 g004
Agriculture 14 00875 g005
Figure 5. Germination as a function of seed deterioration under high-temperature aging stress. The percentage of germination was measured over 120 days in six sea oat populations. (a) BBFL, (b) DWFL, (c) FCFL, (d) FPFL, (e) HIFL, and (f) VEFL. Seed survival curves were estimated using probit analysis to fit the viability equation (Ellis and Roberts, 1981) (n = 750).
Figure 5. Germination as a function of seed deterioration under high-temperature aging stress. The percentage of germination was measured over 120 days in six sea oat populations. (a) BBFL, (b) DWFL, (c) FCFL, (d) FPFL, (e) HIFL, and (f) VEFL. Seed survival curves were estimated using probit analysis to fit the viability equation (Ellis and Roberts, 1981) (n = 750).
Agriculture 14 00875 g005
Agriculture 14 00875 g006
Figure 6. TEAC levels change over time during aging and become undetectable with aging progression. Box and whisker plots show the minimum to maximum values of measured TEAC levels in sea oats. (a) BBFL, (b) DWFL, (c) FCFL, (d) FPFL, (e) HIFL, and (f) VEFL. Trolox Equivalent Antioxidant Capacity (TEAC) was quantified using the modified ABTS/TEAC assay under high-temperature aging stress. A different letter indicates significant differences within a figure based on the one-way ANOVA at an alpha of 0.05 (n = 30).
Figure 6. TEAC levels change over time during aging and become undetectable with aging progression. Box and whisker plots show the minimum to maximum values of measured TEAC levels in sea oats. (a) BBFL, (b) DWFL, (c) FCFL, (d) FPFL, (e) HIFL, and (f) VEFL. Trolox Equivalent Antioxidant Capacity (TEAC) was quantified using the modified ABTS/TEAC assay under high-temperature aging stress. A different letter indicates significant differences within a figure based on the one-way ANOVA at an alpha of 0.05 (n = 30).
Agriculture 14 00875 g006
Agriculture 14 00875 g007
Figure 7. Seed physical and metabolic traits exhibit variable correlations with viability following aging stress in sea oats. We performed a Pearson correlation analysis at an alpha of 0.05. p50; the time taken by seeds for germination to be reduced by 50% was conducted according to Ellis and Roberts (1981) viability equation. Morphological traits were obtained through multispectral imaging and comprised of the total surface area of the seed (seed length, seed compactness circle, and compactness ellipse), seed volume, seed mass, and seed surface area to volume ratio. Other seed traits comprised the seed volume, seed fresh and dry mass, seed total protein content, and seed antioxidant levels (baseline TEAC and TEAC following 36 h of imbibition).
Figure 7. Seed physical and metabolic traits exhibit variable correlations with viability following aging stress in sea oats. We performed a Pearson correlation analysis at an alpha of 0.05. p50; the time taken by seeds for germination to be reduced by 50% was conducted according to Ellis and Roberts (1981) viability equation. Morphological traits were obtained through multispectral imaging and comprised of the total surface area of the seed (seed length, seed compactness circle, and compactness ellipse), seed volume, seed mass, and seed surface area to volume ratio. Other seed traits comprised the seed volume, seed fresh and dry mass, seed total protein content, and seed antioxidant levels (baseline TEAC and TEAC following 36 h of imbibition).
Agriculture 14 00875 g007
Table 1. Seed metabolic traits in sea oats seeds. Metabolic traits were obtained using a seed respiration analyzer for respiration and metabolic analysis of individual sea oats seeds. Raw data was used to generate indices for oxygen depletion and seed metabolic rates (n = 48).
Table 1. Seed metabolic traits in sea oats seeds. Metabolic traits were obtained using a seed respiration analyzer for respiration and metabolic analysis of individual sea oats seeds. Raw data was used to generate indices for oxygen depletion and seed metabolic rates (n = 48).
PopulationIncreasing Metabolism Time (IMT) HoursThe Inverse Time to 50% of the Seeds Reaching 50% Oxygen Level (Hours)Individual Seed Area under the Curve to 50% Oxygen Level (Hours)Relative Germination Time for Seeds (RGT), Hours
BBFL28.17137.77111.23290.08
DWFL28.40141.52118.22343.28
FCFL25.50132.92105.83249.39
FPFL26.10119.0296.21249.09
HIFL29.70152.81128.09383.43
VEFL27.86143.02117.99318.05
Table 2. Seed survival under aging stress conditions. Seeds were subjected to the Seed Bank (MSB) protocol of comparative seed longevity testing by Newton et al. [39], and viability was checked on days 1, 2, 5, 9, 20, 30, 50, 75, 100, and 125. The seed deterioration data were fitted in the probit regression to estimate Ki, σ (sigma), and p50 according to the Ellis and Roberts (1981) viability equation. Adifferent letter indicates significant differences based on the one-way ANOVA at an alpha of 0.05. The same letters indicate no significant differences (n = 750).
Table 2. Seed survival under aging stress conditions. Seeds were subjected to the Seed Bank (MSB) protocol of comparative seed longevity testing by Newton et al. [39], and viability was checked on days 1, 2, 5, 9, 20, 30, 50, 75, 100, and 125. The seed deterioration data were fitted in the probit regression to estimate Ki, σ (sigma), and p50 according to the Ellis and Roberts (1981) viability equation. Adifferent letter indicates significant differences based on the one-way ANOVA at an alpha of 0.05. The same letters indicate no significant differences (n = 750).
PopulationKi1/-Sigmap50
BBFL5.79 ± 0.11−0.02948 ± 0.002526.37 ± 1.92 aEffect size.
η2 = 0.90
Population 95% CI (0.76, 1.00)
DWFL6.68 ± 0.63−0.06466 ± 0.019725.00 ± 1.83 a
FCFL5.62 ± 0.13−0.05317 ± 0.006811.45 ± 0.96 b
FPFL6.06 ± 0.09−0.04510 ± 0.001223.50 ± 1.69 a
HIFL5.90 ± 0.21−0.03501 ± 0.007225.46 ± 0.90 a
VEFL5.64 ± 0.15−0.04771 ± 0.008513.18 ± 0.87 b
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Egesa, A.O.; Davidson, M.T.; Pérez, H.E.; Begcy, K. Biochemical and Physical Screening Using Optical Oxygen-Sensing and Multispectral Imaging in Sea Oats Seeds. Agriculture 2024, 14, 875. https://doi.org/10.3390/agriculture14060875

AMA Style

Egesa AO, Davidson MT, Pérez HE, Begcy K. Biochemical and Physical Screening Using Optical Oxygen-Sensing and Multispectral Imaging in Sea Oats Seeds. Agriculture. 2024; 14(6):875. https://doi.org/10.3390/agriculture14060875

Chicago/Turabian Style

Egesa, Andrew Ogolla, Maria Teresa Davidson, Héctor E. Pérez, and Kevin Begcy. 2024. "Biochemical and Physical Screening Using Optical Oxygen-Sensing and Multispectral Imaging in Sea Oats Seeds" Agriculture 14, no. 6: 875. https://doi.org/10.3390/agriculture14060875

APA Style

Egesa, A. O., Davidson, M. T., Pérez, H. E., & Begcy, K. (2024). Biochemical and Physical Screening Using Optical Oxygen-Sensing and Multispectral Imaging in Sea Oats Seeds. Agriculture, 14(6), 875. https://doi.org/10.3390/agriculture14060875

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop