Using Infrared Thermography to Study Ice Nucleation and Propagation in Plants

Infrared thermography offers a distinct advantage over other methods of studying ice nucleation and propagation in plants because it allows one to directly visualize the freezing process. This technique has provided new details about the freezing process in most of the plant species in which it has been used. It is nonintrusive and thus overcomes any influence on the pattern of freezing resulting from attached objects such as thermocouples. Collectively, the results of several studies have identified several new possibilities for enhancing frost protection. In particular, selection of cultivars with thicker cuticles or providing an externally-applied, hydrophobic barrier, may provide a way of blocking extrinsic ice formation from propagating into a plant and initiating a freezing event. The application of a hydrophobic particle film to the surface of tomato plants provided frost protection and prevented plants from freezing despite the presence of ice on the external leaf surface. Breeding for the presence of barriers to ice propagation in woody plants, that allow expanded flowers or inflorescences to supercool despite the formation of ice in stem tissues, may also be a practical approach for enhancing cold hardiness during spring frosts. Evidence that the expression of transgenes coding for insect antifreeze proteins in Arabidopsis can enhance supercooling in the absence of extrinsic ice nucleation has also been reported. It is expected that further studies utilizing infrared thermography to study ice nucleation and propagation in plants will lead to a better understanding of the mechanisms plants have evolved to accommodate ice formation within their tissues. Such knowledge will be extremely valuable for developing new frost protection technologies. INTRODUCTION In order for ice to form on or within a plant, ice nucleation must first occur. Although the melting point of ice is 0°C, the freezing temperature of water is not as defined (Ashworth, 1992). This is because a small ice crystal embryo is necessary in order for ice to form and grow to any substantial size. The probability of forming such an ice crystal embryo in pure water, as well as the half-life of such a crystal, is low until temperatures approaching -40°C. This temperature is referred to as the homogeneous ice nucleation point. In nature, it is rare for water to exist in a pure state but it rather exists as an ionic or colloidal solution. In such solutions heterogeneous ice nucleation is initiated on the Proc. XXVI IHC – Environmental Stress Eds. K.K. Tanino et al. Acta Hort. 618, ISHS 2003 Publication supported by Can. Int. Dev. Agency (CIDA) 486 surface of objects or on suspended particles (Franks, 1995). Heterogeneous ice nucleators are very effective in inducing ice formation and are very abundant. As a consequence, freezing occurs in nature at much warmer temperatures than the homogeneous nucleation temperature. Generally, freezing of plants will be initiated at temperatures a few to several degrees below 0 ̊C. The factors that determine the level of supercooling in a plant however are not well understood. Ice formation in intact plants can be readily detected by measuring, with thermocouples, the heat that is released upon the freezing of water within the plant (George et al., 1972; Quamme, 1972; Wisniewski and Fuller, 1999). Even when arrays of temperature-measuring devices are attached to plants, however, the actual site of ice initiation and the temperature at the site where ice nucleation occurred can only be inferred. This is a significant technical limitation and more details of the freezing process are required in order to accurately predict when and how plants will freeze. Over the past few years, the ability to use infrared video thermography to directly observe ice nucleation (i.e. initial ice formation) and propagation in plants has been demonstrated (Ball et al., 2002; Pearce and Fuller, 2001; Wisniewski and Fuller, 1999, Wisniewski et al., 2001). The use of this technology to study the freezing process is based on the fact that ice formation is an exothermic event and the release of the heat of fusion as water changes phase from a liquid to a solid can be monitored and visualized. The temperature and spatial resolution of the device used in these studies has enabled the researchers to clearly define the initial site of ice nucleation as well as monitor the ice front as it spread into the surrounding tissues. Importantly, these studies have shown that the site of placement of a thermocouple was often the site of initial freezing. This suggests that an erroneous conclusion may have been derived from earlier studies of ice nucleation in plants that relied on thermocouple-generated data. Using infrared thermography it is possible to determine the role of external and internal ice nucleating agents in the freezing process, rates of ice propagation, the effect of plant structure on the freezing process, and how the specific pattern of freezing relates to visual patterns of injury. It is also possible to clearly evaluate if the reduction of ice nuclei or inhibiting their activity is a feasible approach to frost protection. The purpose of this review is to provide a few examples of where infrared thermography has been used to study freezing and ice propagation in plants. More detailed information on this topic may be found in two recent reviews by Wisniewski and Fuller (1999) and Wisniewski, et al. (2002). ROLE OF EXTERNAL ICE AND PLANT CUTICLES ON ICE NUCLEATION OF PLANTS The presence of moisture on the surface of a plant plays a critical role in determining the temperature at which a plant will freeze (Ashworth, 1992). Dry plants will generally supercool to a lower temperature than wet plants. If ice-nucleating-active (INA) bacteria, such as Pseudomonas syringae, are present they will induce plants to freeze at a warmer temperature than just moisture alone (Wisniewski et al., 1997; Fuller and Wisniewski, 1998). INA bacteria, however, must be in an aqueous solution in order to be active. In order for external ice on the surface of a leaf to initiate freezing of the plant, ice crystals must physically growth through a break in the surface of the cuticle or through a stomatal opening. Figure 1 shows the influence of plant structure on ice nucleation in leaves of bean (Phaeseolus vulgaris). When droplets of water containing INA bacteria are placed on either the adaxial surface, having no stomates (Fig. 1A), or the abaxial surface, having stomates (Fig. 1B), the droplets freeze at approximately the same time (Fig. 1C). Freezing of the leaf first occurs, however, on the abaxial surface, presumably from propagation of ice crystals from the frozen droplet through stomatal openings (Fig. 1D). Freezing of the leaf through the abaxial surface can be prevented by applying a layer of silicon grease between the droplet and the leaf surface (Fig. 1E-1F). This further supports the idea that ice must physically penetrate the leaf surface to induce freezing of the leaf