Quantitative methods in microscopy to assess pollen viability in different plant taxa

Area of Science:

• Plant Biology
• Biotechnology
• Microscopy

Background:

• Pollen quality is crucial for plant reproduction.
• Manual pollen assessment is time-consuming and lacks statistical power.
• Automated methods for pollen viability and sterility are needed for crop improvement.

Purpose of the Study:

• To refine fluorescein (FDA)/propidium iodide (PI) staining for pollen.
• To develop automated image analysis for pollen quality assessment.
• To increase throughput and accuracy in evaluating pollen viability and sterility.

Main Methods:

• Utilized FDA/PI staining on pollen from 10 taxa.
• Employed open-source software (CellProfiler, Fiji, R) for image analysis.
• Applied thresholding, segmentation, and supervised/unsupervised classification for pollen identification.

Main Results:

• FDA/PI staining effectively differentiated viable, dead, and sterile pollen across all taxa.
• Automated analysis of over 75,000 pollen grains achieved high accuracy (R² = 0.99).
• Significantly improved statistical power compared to manual counting.

Conclusions:

• Established a robust methodology for automated pollen viability assessment via fluorescence microscopy.
• The developed approach provides a baseline for fundamental and applied plant biology research.
• A comprehensive FDA/PI pollen image dataset is publicly available for future studies.