Warming effects on carbohydrates and their possible effects on coverage
Experimental warming in tundra regions often causes increased photosynthesis and growth rate [23-25]. In line with our hypothesis 2, Dr. octopetala var. asiatica positively responded to OTC-warming in tissue NSCs, plant coverage, photosynthesis and single leaf size [25], indicating that warming makes Dr. octopetala var. asiatica to fix more carbon, to grow fast, and then to occupy more space and have stronger competitiveness. The soluble sugars of Rh. confertissimum negatively responded to warming (Fig. 1), and decreases in the coverage of Rh. confertissimum have already been recorded during the experimental period (Table 1), consistent with the hypothesis 2. The coverage of V. uliginosum grown in the warming OTCs and the control plots didn’t significantly change during the whole experimental period (Table 1). Jin et al. (2019) found that V. uliginosum on the west-facing slope of Changbai Mountain alpine tundra showed a patch distribution deviating from the previously normal distribution [7]. Our research field was located on the north-facing slope whose microenvironment (e.g. lower temperature) should be different from the west-facing slope (higher temperature) of Changbai Mountain alpine tundra. The starch storage of V. uliginosum responded negatively to the warming (Fig. 1), so we predict that the distribution of V. uliginosum on the north-facing slope might decrease in the future or gradually show a patch distribution like on the west-facing slope.
The responses of carbohydrates to warming in tundra shrubs were found to be species-specific in the present study which is consistent with our hypothesis 1 (Table 2). Similar results were also observed by other studies [26, 27]. Warming increased NSCs concentrations in Himantormia lugubris but decreased them in Polytrichastrum alpinum, Pinus sylvestris, Pseudotsuga menziesii and Picea mariana [27-30]. No significant change in TNC was observed in Salix Polaris, Carex vaginata, Saussurea alpine, Selaginella selaginoides, V. uligonosum, Usnea antarctica, U. aurantiaco-atra, Sanionia uncinata, Quercus robur and Q. petraea leaves in response to warming [17, 24, 26, 27]. Different species has different sensitivity to warming which can affect photosynthesis, the distribution and utilization of photosynthate, and growth etc.
Stored carbohydrates in roots can be used by plants for defense and regrowth, or as a buffer under insufficient carbon production [31-35]. Warming didn’t significantly affect root carbohydrate concentrations of the three species, which is consistent with previous findings that there were non-significant responses of root carbohydrates at the end of the growing season to warming for alpine plants (Elymus nutans, Euphrasia regelii and Swertia mussotii) on the Tibetan Plateau [10], for grapevines in Barossa Valley of Australia [33], for Pinus taeda and P. ponderosa [36]. V. uliginosum roots had higher starch and TNC compared to Dr. octopetala var. asiatica and Rh. confertissimum, which might be related to that V. uliginosum is a deciduous species. Deciduous plants need more energy and nutrition for new leaves sprouting in the next spring, especially on alpine tundra regions.
The accumulation of NSCs is one of the cryoprotective mechanisms [37]. The sugar-starch system in plants adjusts the ratio of sugar to starch in response to low temperature or other stressors [38]. At high elevations, a higher sugar-starch ratio reflects that plants are subjected to lower temperatures, sometimes positively correlating with cold stress [39]. We found that the ratios of soluble sugars to starch in the warming OTCs were lower than those in the control plots for all the three species, indicating that warming can affect sugar-starch relationship. The three species grown in the warming OTCs hydrolyze less starch against freezing which might be beneficial to growth.
Time-dependent NSC levels and their possible effects on coverage
Marked seasonal patterns of NSCs concentrations for the three species were observed, which is in agreement with earlier findings of the seasonal NSCs fluctuation with soluble sugars concentrations reaching a higher level at the end of the growing season in various tree species [40-42]. Soluble sugars, serving as osmotic adjustment and signal substances, play an important role against cold [41, 43]. The starch concentrations peaked in July or August (the active growing season) and decreased in September, indicating that starch hydrolyzes to soluble sugars towards end-season [42]. The temporal variation in the level of NSCs also illustrates that temperature might affect the proportions of carbohydrate component.
Wintergreen Dr. octopetala var. asiatica anddeciduous V. uliginosum are typical alpine and arctic dwarf shrubs, especially Dr. octopetala var. asiatica generally dominating community [44]. Evergreen Rh. confertissimum has a small distribution area compared to the two others, only in tundra regions. Generally, deciduous trees require more abundant carbohydrates for vegetative or reproductive growth before the new leaves grow [34, 45]. V. uliginosum had similar concentrations of TNC as Dr. octopetala var. asiatica, but V. uliginosum did not show the same obvious advantages in coverage, photosynthesis, leaf size and growth as Dr. octopetala var. asiatica did [25], indicating that V. uliginosum might allocate some carbohydrates to the growth of new organs. The roots of V. uliginosum had significantly higher starch and TNC concentrations than Dr. octopetala var. asiaticaand Rh. confertissimum. Starch, different from soluble sugars, is inactive and accumulates as a storage compound. V. uliginosum, as a deciduous species, is assumed to store high amounts of carbohydrates over harsh winter to support leaf flush in spring [45]. The concentrations of TNC in V. uliginosum and Dr. octopetala var. asiatica were significantly higher than those of Rh. confertissimum that had significantly decreased coverage. Therefore, TNC concentrations are related to the species coverage, which is consistent with our hypothesis 2.
The NSCs responses of Dr. octopetala var. asiatica to warming support the idea that carbon allocation is a key factor for determining dominance. After five years of warming by open-top chambers in the alpine region of southwestern Norway, the carbohydrates storage of Dr. octopetala increased [46]. Dr. octopetala var. asiatica showed increased trend in the carbohydrate concentrations in the present study and significantly increased leaf size [25], indicating that the total carbohydrate contents (leaf biomass or size × concentration) have been stimulated. The stimulation and accumulation of carbohydrates are conducive to dominance, expansion and improvement of competitiveness, which may further lead to changes in community composition and structure of alpine and tundra ecosystems under future climate change [44].
Responses of secondary compounds and trade-off with NSCs
Deciduous species probably have lower concentrations of secondary compounds compared to evergreen plants [47]. Higher concentrations of secondary compounds in perennial leaves could be a greater need for defending herbivorous predator due to longer life span [48]. However, we found that the deciduous V. uliginosum had the highest absolute concentrations of total secondary compounds compared to the other two species.One of the reasons is probably related to that V. uliginosum produces delicious fruits which might need more secondary compounds to defend animals, especially triterpenes.
We observed that Rh. confertissimum allocated relatively more carbon to defense than V. uliginosum and Dr. octopetala var. asiatica based on the ratio of the sum of secondary compounds to total carbon (secondary compounds + TNC). Herms and Mattson found that increased investment in secondary defense is accompanied by decreased growth, plant size and competitive ability [16]. Thus, more carbon is allocated to secondary compounds in expense of growth, dominance and competition. Warming may alter interspecific competitive relationships and community structure because of the changes in carbon allocation pattern and defense abilities [46].
Total phenols concentrations were decreased by warming in the present study, similar to the results of Holopainen et al. [49]. Phenolic compounds originate from the shikimic acid pathway which is related to the carbohydrate metabolisms [50, 51] and antioxidative potential of plants [52]. Tundra ecosystem is generally characterized by simultaneous stresses such as low temperature, high UV radiation, low nutrient availability which could make plants to produce high levels of secondary compounds or allocate more proportion carbon to the secondary compounds [46, 53]. Thus, alleviation of low temperature by warming is expected to decrease the contents of secondary compounds. The decreases in the concentrations of secondary compounds for Bistorta vivipara, Dr. octopetala, Salix reticulate, Cassiope tetragona, S. herbacea × Polaris and Tofieldia pusilla have been reported [17, 22, 46]. The decrease in total phenols concentration probably relates to the fewer carbon resources for defense substance or more carbon for growth [17]. The decreased proportion of total phenols of Dr. octopetala var. asiatica was relatively higher than the other two species (Fig. 2) which is consistent with higher NSCs and growth. Hence, Dr. octopetala var. asiatica has strong competitive potential.
Flavonoids concentrations of Rh. confertissimum were increased by warming, but OTC warming did not affect the levels of flavonoids in Dr. octopetala var. asiatica and V. uliginosum. We also found that Rh. confertissimum grown in the warming OTCs had relatively lower soluble sugars and significantly reduced coverage than the controls or the other two species. The three species in the present study changed their defense levels to some extent when experiencing continuous 7 years’ warming, consistent with our hypothesis 1 that there was significant interaction between species and warming on the total phenols and flavonoids.
Inconsistent with our hypothesis 3, no significant trade-off relationship between NSCs and secondary compounds in leaves was observed based on the correlation analysis (Figure not shown). However, the secondary compounds tended to be positively correlated with NSCs for Dr. octopetala var. asiatica while negatively correlated with NSCs for V. uliginosum and Rh. confertissimum. Further research is necessary to continue to examine if long-term warming can result in a trade-off relationship between NSCs and secondary compounds for species with decreased distribution or competitiveness.