Although we know that most of the cellular sucrose is present in the cytosol and vacuole, our knowledge on the impact of this sucrose compartmentation on plant properties is still fragmentary. Here we attempted to alter the intracellular sucrose compartmentation of Arabidopsis mesophyll cells by either, overexpression of the vacuolar sucrose loader BvTST2.1 or by generation of mutants with decreased vacuolar invertase activity (amiR vi1-2). Surprisingly, BvTST2.1 overexpression led to increased monosaccharide levels in leaves, while sucrose remained constant. Latter observation allows the conclusion, that vacuolar invertase activity in mesophyll vacuoles exceeds sucrose uptake in Arabidopsis, which gained independent support by analyses on tobacco leaves transiently overexpressing BvTST2.1 and the invertase inhibitor NbVIF. However, we observed strongly increased sucrose levels in leaf extracts from independent amiR vi1-2 lines and non‑aqueous fractionations confirmed that sucrose accumulation in corresponding vacuoles. amiR vi1-2 lines exhibited impaired early development and decreased weight of seeds. When germinated in the dark, mutant seedlings showed problems to convert sucrose into monosaccharides. Cold temperatures induced marked downregulation of the expression of both VI genes, while frost tolerance of amiR vi1-2 mutants was similar to WT indicating that increased vacuolar sucrose levels fully compensate for low monosaccharide concentrations.

Materials and methods

Plant material and growth conditions

Nicotiana benthamiana, Arabidopsis thaliana (ecotype Columbia-0) and corresponding Arabidopsis mutants were cultivated in a growth chamber (Weiss-Gallenkamp, Heidelberg, Gemany) on standardized soil (ED-73; Patzer; www.einheitserde.de), at a constant tempe­rature of 22°C and a light intensity of 120 µmol quanta m^-2 s^-1 (µE). Plant cultivation was carried out under short day conditions,10 h light, 14 h darkness (standard conditions). For cold experiments, plants were grown for four weeks under standard conditions and subsequently acclimated to cold temperature for three days at 4°C (all other conditions were kept constant). For etiolation growth analyses, seeds were stratified for 24 h at 4°C, and cultivated in darkness for seven days on water soaked blotting paper. For seed analyses, plants were transferred after cultivation for four weeks at standard conditions to long day conditions (22°C and 200 µE, 16 h light per day). For dark recovery experiments, 4-week old plants grown under standard conditions were darkened for five days, followed by recovery for further 7 days under standard conditions.

Generation of mutants

For transient transformation of Nicotiana benthamiana leaf mesophyll cells the Agrobacterium infiltration method was performed according to an established method (Jung et al., 2015). For generation of VI1-2 double knock down mutants, we followed an established protocol for gene silencing by artificial microRNA (amiRNA) (Schwab et al., 2006). With the web based amiRNA designer tool program (http://wmd3.weigelworld.org) an amiRNA, which targets VI1 (At1g62660) and VI2 (At1g12240) simultaneously, was designed. The sequence TAAGGATGAATAAAAGCACGG was used for generation of primers including the Gateway™ compatible sequences attP1 and attP2 to engineer the amiRNA fragment. The primer sequences are listed in Table S1. Subsequently, the fragment was sub-cloned via BP reaction into the Gateway™ entry vector pDONR/Zeo and via LR reaction into the destination vector pK2GW7, which contains a 35S-CaMV promotor. For generation of stably transformed Arabidopsis mutant plants, Agrobacterium-mediated transformation using floral dip was performed (Clough and Bent, 1998). VI1-2 double knock-down mutants were selected by screening for the lowest remaining VI1 and VI2 transcript levels via qRT-PCR leading to the two independent lines no. 4 and 5.

cDNA synthesis, qRT-PCR and RNA gel‐blot hybridization

RNA was isolated from 50 mg of frozen, fine ground plant material with the NucleoSpin RNA Plant Kit (Macherey-Nagel, Düren, Germany), according to the manufacturer’s protocol. For cDNA synthesis, RNA was transcribed into cDNA with the qScript cDNA Synthesis Kit (Quantabio, Beverly, MA, USA). The primers used for gene expression analysis by qRT-PCR are listed in Table S2. NbAct, AtPP2A and AtSAND were used as reference genes for transcript normalization. Alternatively, gene expression was analyzed by RNA gel‐blot hybridization.

Acidic invertase activity assay

The enzyme assay was performed as described by (Tamoi et al., 2010) with slight modifications. 100 mg of frozen and fine ground plant material were homogenized with 1 ml of ice cold 200 mM Hepes/HCl (pH 5.0), 1 mM EDTA, 1 mM PMSF on a vortex mixer for 20 sec. The samples were incubated for 20 min on ice prior to further mixing with a vortex mixer for 20 sec. Subsequently, samples were centrifuged at 20.000 g for 10 min at 4°C and the supernatant was transferred into a new reaction tube. The rate of sucrose hydrolysis was quantified spectrophotometrically at 22°C using a NADP-coupled enzymatic test (Stitt et al., 1989). For this, 15 µl of enzyme extract were added to 190 µl 200 mM Hepes/HCl (pH 5.0), 10.5 mM MgCl₂, 2.1 mM ATP, 0.8 mM NADP, 0.5 U Glucose-6-Phosphate Dehydrogenase, 0.18 U Hexokinase and 0.48 U Phosphoglucose Isomerase. To start the enzyme reaction, 5 µl of 200 mM sucrose solution were added to the sample.

Metabolite quantification

For sugar extraction, we added 400 µl of 80% of ethanol to 100 mg of frozen, fine grounded plant material, mixed and incubated for 30 min at 80°C, in a thermomixer at 500 rpm. After centrifugation at 16000 g (10 min at 4°C) the supernatant was used for sugar quantification using a NADP-coupled enzymatic test (Stitt et al., 1989).

Non-aqueous fractionation

Subcellular sugar distribution of sugars in leaves was determined by non-aqueous fractionation of 4-week old plants. To this end, 15 mg of freeze-dried and fine grounded plant material was used and processed as described previously (Fürtauer et al., 2016). Acid phosphatase served as vacuolar marker, UGPase activity served as a cytosolic marker and alkaline pyrophos­phatase served as chloroplast marker. For sugar quantification, a NADP-coupled enzymatic test was performed (Stitt et al., 1989) and subcellular metabolite distribution was calculated using an established algorithm (Fürtauer et al., 2016).

Seed analyses

Lipid quantification was performed according to a routine protocol (Reiser et al., 2004) with slight modifications. 0.1 g of mature, air-dried seeds were homogenized in a mortar and liquid nitrogen. Subsequently, 1.5 mL of isopropanol was added and the sample was further homo­genized. The suspension was transferred into a 1.5 mL-reaction tube and incubated for 12 h at 4°C and 100 rpm. Subsequently, samples were centrifuged at 13000 g for 10 min and the supernatant was transferred into a pre-weighed 1.5-mL reaction tube. Tubes were incubated at 60°C for 12 h to completely evaporate the isopropanol. Total lipid content was quantified gravimetrically. For determination of seed weight, 1000 mature and air-dried seeds were counted and their weight was quantified gravime­trically.

Electrical conductivity

The frost tolerance, as measured by frost induced release of ions from leaf sample, of wild-type and mutant, was quantified by electrical conductivity assays, described earlier (Klemens et al., 2014).