presence of surface bacteria (see below) may greatly enhance inoculative freezing, which can readily be seen via an increase in whole-body SCPs, and it appears that in this case inoculation takes place via the spiracles (Lee et al. 1998). Inoculative freezing as a result of host plant contact has also been observed in several phytophagous species (review in Klok and Chown 1998a,b). Generally, the insect's SCP is increased as a result of host plant contact, leading to elevated mortality, and at least in Embryonopsis halticella (Lepidoptera, Yponomeu-tidae), this effect is more marked at lower temperatures. However, host plant contact sometimes leads to a reduction in insect mortality (Butts et al. 1997), and, paradoxically, by providing protection from free water, and consequently ice crystals, mining or ensheathed feeding may promote survival of insects at temperatures above the SCP of their hosts (Connor and Taverner 1997; Klok and Chown 1998b).
External inoculation is not the only route by which heterogeneous ice nucleation can take place. Ice nucleators in the tissues, haemolymph, and gut lumen can also initiate freezing, and consequently these nucleators are removed or masked prior to the onset of winter (Zachariassen 1985). Several studies have now shown that haemolymph protein ice nucleators and lipoprotein ice nucleators are either reduced in quantity or removed during winter. In those cases where they are reduced, AFPs mask the ice nucleator activity, thereby permitting supercooling (Duman 2001).
The role of nucleators in the gut and the effect of their removal, via gut clearance, prior to winter have not been clearly defined. Although it has long been known that the presence of food (or its contaminants) in the gut can initiate freezing (Salt 1961), and that ingested gram-negative bacteria within the Pseudomonadaceae and Entereobacteriaceae can cause considerable elevation of the SCP (Denlinger and Lee 1998; Lee et al. 1998), the effect of gut clearance on SCPs is controversial. In several species, the presence of food in the gut has a marked effect on SCPs, which is often removed if the animals are starved or are in a stage where feeding does not occur (Cannon and Block 1988; Duman et al. 1991). Moreover, in species where starvation does not alter SCPs it has often been shown that individuals do not completely evacuate their guts
(Parish and Bale 1990; Klok and Chown 1998b). By contrast, Baust and Rojas (1985) point out that many species do not evacuate their guts prior to overwintering, that in some cases diet manipulations only have short-term effects on SCPs, that starvation or the absence of food in the gut sometimes has no effect on supercooling capacity, and that peritro-phic membranes might prevent the nucleator potential of the gut contents from being realized. They also note that nucleation may take place elsewhere in the body, or via external contact with ice, and that water consumption can cause a decrease in supercooling capacity. Together these apparently contradictory results suggest that nucleation can be initiated in a variety of sites and that the process is likely to be complicated by the life histories of the species involved. For instance, in species that overwinter in the soil, such as L. decemlineata, soil water content has a significant influence on insect water content, which in turn affects both the SCP and mortality (Costanzo et al. 1997). Moreover, the physico-chemical attributes of the soil also have a pronounced influence on mortality. Nonetheless, it is clear that the gut and its contents have a significant role to play in nucleation, at least in some species.
This role has been best demonstrated in spring-tails, and particularly the Antarctic springtail C. antarcticus. In this species, in summer, field-collected individuals, and individuals fed moss turf homogenate, SCP frequency distributions are essentially high and unimodal, if somewhat left-skewed (Fig. 5.20) (Somme and Block 1982). If individuals are starved, the SCP distributions either become bimodal or have a unimodal distribution at low temperatures. Such changes also occur over a seasonal basis in other springtail species (Somme and Block 1991). Although bimodal SCP distributions have generally been ascribed to the two stage process of gut clearance followed by the development of a cryoprotectant system in preparation for winter cold, they have also been documented in response to changing acclimation temperatures in species where gut clearance does not take place (Klok and Chown 1998b). While it has been suggested that in summer-acclimated or acclimatized individuals this bimodality would disappear, its underlying causes have not been carefully
Starved individually 50C/6d X = -24.8
Fed purified green algae, 5°C/4d
Was this article helpful?