Dehydrin accumulation and extreme low-temperature tolerance in Siberian spruce (Picea obovata)

To investigate the role of dehydrins (DHNs) in extreme low-temperature (LT) tolerance, we sampled needle tissue of Siberian spruce (Picea obovata Ledeb.) from trees growing in an arboretum in Trondheim, Norway from August 2006 to April 2007 and tracked changes in LT tolerance via relative electrolyte leakage. We used western blotting to estimate relative amounts of proteins binding a DHN K-segment antibody, measured relative amounts of nine transcripts for small (<25 kDa) DHNs by quantitative reverse transcription-polymerase chain reaction (PCR) using primers developed for DHN transcripts in a closely related species, Picea abies (L.) Karsten, and isolated and sequenced PCR products for five P. obovata DHNs. Three protein bands of 53, 35 and 33 kDa were detected on western blots of SDS-PAGE-separated protein extracts. The 53-kDa DHN was already present late in the growing season, but accumulated during acclimation, and levels decreased rapidly during deacclimation. The 33- and 35-kDa proteins, identified as Picg5 class DHNs by mass spectrometry, first appeared in detectable amounts late in the acclimation process and remained at detectable levels throughout the period of maximum LT tolerance. Levels of the 53-kDa DHN correlated with two LT tolerance parameters, while results for the 33- and 35-kDa proteins were equivocal due to limited sample size and variation in LT tolerance during the mid-winter period. Three additional bands of 30, 28 and 26 kDa were detected in extracts from needles collected in November 2010 using an immunity-purified antibody. Immunoblotting of two-dimensional gel electrophoresis gels loaded with proteins extracted from October and November samples corroborated the results obtained by SDS-PAGE western blots. One large spot in the 53 kDa range and two trains of spots in the same size range as the 33 and 35 kDa DHNs were detected using the K-segment antibody. Eight of the nine DHN transcripts closely tracked LT tolerance parameters, whereas the ninth DHN transcripts followed a reverse pattern, decreasing during winter and increasing again during deacclimation. Multiple regression models using principal components of the transcripts to predict two different LT tolerance parameters suggest separate but overlapping functions for different DHNs in establishing and maintaining extreme LT tolerance.