Author:

Abstract:

Most fruit tissues are microstructured. The structural arrangements of cells and intercellular spaces create specific pathways for transport of chemical compounds, indispensible for maintaining fundamental metabolic equilibria in the tissue. Besides, microstructure determines important traits such as fruit size and texture, influencing consumer appreciation and marketability. Until recently, the measurement of microstructure was essentially based on light or electron microscopy. These methods often require invasive sample preparation and are restricted to 2D imaging of a limited number of cross-sections. We wanted to use high-resolution X-ray micro-CT images of apple (‘Jonagold’, ‘Braeburn’, ‘Kanzi’) and pear (‘Conference’) tissue to develop a rapid and minimally invasive method for characterising 3D fruit microstructure at the level of single cells and pores. CT datastacks consisting of virtual cross-sections of the fruits were analysed using Avizo Fire software (VSG, France). Individual pores in the fruit tissue could be readily isolated because of the excellent contrast between intercellular air and the more dense cell content. The cell-to-cell boundary did not however supply sufficient X-ray contrast to separate touching cells. To differentiate neighboring cells in the images, we optimized a specialised image processing tool called watershed separation. Based on knowledge of the average cell size and its sphericity index, we constructed a virtual sieve to discard incorrectly segmented cell particles or unseparated clumps of cells. This resulted in a large amount of cells (500-1500) and pores (500-3000) for each tissue sample that could be quantified. ‘Jonagold’ cells are on average the largest (210.7 µm equivalent spherical diameter), whereas ‘Conference’ cells are considerably smaller (156.8 µm). Cell sizes of ‘Braeburn’ (199.2 µm) and ‘Kanzi’ (176.3 µm) are situated in between. We also determined shape descriptors such as sphericity, concavity index, length and width to establish characteristic cells and pores for every cultivar. With this method for isolating single cells in in vivo fruit samples and the geometrical information thus obtained, we realized a detailed insight in fruit microstructure. We found considerable differences between the structural configuration of 3 apple cultivars and the pear cultivar. Such microstructural information is valuable for explaining and possibly even predicting gas-exchange related disorders, as well as clarifying textural properties of fruits.