Protocol: a medium-throughput method for determination of cellulose content from single stem pieces of Arabidopsis thaliana

Background Lignocellulosic biomass is an important renewable resource for biofuels and materials. How plants synthesise cellulose is not completely understood. It is known that cellulose synthase complex (CSCs) moving in the plasma membrane synthesise the cellulose. CESA proteins are the core components of CSC. In Arabidopsis, in vitro mutagenesis of proteins followed by complementation analysis of mutants lacking the gene represents an important tool for studying any biological process, including cellulose biosynthesis. Analysis of a large number of plants is crucial for these types of studies. Results By using aspiration rather than centrifugation to remove liquids during various stages of protocol, we were able to increase the throughput of the method as well as minimise the sample loss. As a test case, we determined cellulose content of wild type and secondary wall cesa mutants across the length of primary shoot which was fond to be rather uniform in 7-week-old plants. Additionally, we found that the cellulose content of single mutants was comparable to the higher order mutants. Conclusions Here we describe a medium-throughput adaptation of Updegraff’s method that allowed us to determine cellulose content of 200 samples each week.


Background
Cellulose is the most abundant component of plant cell walls that constitute the majority of lignocellulosic material, an important feedstock for new generation of biofuels. Despite its importance, an understanding of how plants make cellulose is far from complete. Arabidopsis provides an excellent model system to study cellulose biosynthesis and a vast amount of progress has been made in the identification of genetic components of Arabidopsis cellulose biosynthesis machinery. Cellulose is synthesised by the cellulose synthase complex (CSC) particles moving in the plasma membrane [1][2][3][4]. The CESA proteins form the bulk of the CSC. Further studies on CSC will involve mutation of crucial residues of CESA proteins and studying their effect on the amount of cellulose in the cell walls. This will require analysis of a large number of plants.
Currently, most of the methods described for crystalline cellulose content determination are a variation of Updegraff method [5]. The method involves removing hemicellulose and lignin with acetic/nitric reagent [6]. Crystalline cellulose is resistant to acetic/nitric reagent but becomes disordered upon treatment with 67 % sulphuric acid making monomeric sugars available to be measured by a colorimetric method using anthrone as a dye [7]. The Updegraff method has previously been used for determination of cellulose content in Arabidopsis stem [1]. This involves preparation of alcohol insoluble residue (AIR) before acetic/nitric and 67 % sulphuric acid steps. However, most published studies have involved analysis of a few samples. Here, we report a practical adaptation enabling us to process up to 200 samples per week. As an example, we use the method the study cellulose content of single, double and triple mutants of secondary wall CESA genes.

Cellulose content of cesa mutants
Our streamlined cellulose assay protocol allows a single person to process up to 200 samples in a week's time with the help of only basic laboratory equipment. As an example, we analysed 200 samples from three separate experiments that involved comparing the cellulose content of single, double and triple cesa mutants (61 samples, Fig. 1); stem segments from multiple locations within the stem (76 samples, Fig. 2) and the effect of sulphuric acid treatment time (64 samples, Fig. 3).
Arabidopsis mutants that result from mutations in CESA4, CESA7 or CESA8 all exhibit multiple phenotypes including reduced cellulose content, reduced plant height and collapsed xylem [1,[8][9][10] and serve as excellent tools for studying cellulose biosynthesis in Arabidopsis. The three secondary cell wall CESAs form a complex and higher order mutants of secondary cell wall CESAs would be a valuable tool for advanced studies on the composition and structure of the complex. These higher order mutant combinations have not been described before. We crossed together the three single mutants to create three double mutants cesa8 irx1-7 cesa7 irx3-6 , cesa8 irx1-7 cesa4 irx5-4 and cesa7 irx3-6 cesa4 irx5-4 and the triple mutant cesa8 irx1-7 cesa7 irx3-6 cesa4 irx5-4 . There were no significant differences in the cellulose content of the single, double and triple mutants, all being around 10 % of cell wall material (Fig. 1). This is consistent with each of the three secondary cell wall CESA proteins being absolutely essential for cellulose synthesis in the secondary cell wall and with the fact that cesa7 irx3-1 contains no cellulose in the secondary cell wall [11]. Any residual cellulose is likely to come from the primary cell wall and confirms that the CESA4, 7 and 8 have no role in primary cell wall biosynthesis.
Cellulose content of Ler0 WT plants has been shown to be increasing from 30 % for 26-day-old plants to 35 % for 36-day-old plants [1]. All the mutants used in this study are based on Col0 background which matures slower than Ler0. We chose to harvest the stem material for cellulose assays from 7-to 8-week-old plants. By this time, we expect the process of secondary cell wall deposition to be complete. To investigate whether this was the case we exploited the fact that the stem is a developmental series with the secondary cell wall deposition starting at the top. We divided the Col0 plants that were 40 cm tall into eight pieces of 5 cm each. Similarly, the irx mutants which grow to about 15 cm were divided into three pieces of 5 cm each. We found that at 7 weeks, all plants had mostly uniform cellulose content across the stem (Fig. 2). This is in contrast to when much younger plant stems are analysed that can exhibit a gradient of secondary cell wall deposition [12].

Sample loss
Previous applications of Updegraff method in Arabidopsis have involved fragmentation/homogenisation of stem material and subsequent centrifugation steps to collect the material after each treatment [1,13]. This is a laborious process when large number of samples are involved. Also, during centrifugation, sometimes part of the stem material floats instead of settling into a tight pellet. This would result in loss of sample and inaccuracies in the final data. We kept the material as two pieces for each sample and removed all centrifugation steps. Instead, we used an aspirator to remove liquids after each step. The samples can be visually tracked as two pieces throughout the process until sulphuric acid treatment. Sample loss during aspiration can be minimised by using a drawn out Pasteur pipette to generate a narrower opening, however, should any sample be lost via aspiration, such samples are discarded. We typically started with eight replicates for each genotype and loss of one replicate would not affect the data.
Step 11 of the protocol (see detailed protocol below) is the step most likely to cause the fragmentation of stem pieces because of larger liquid volumes involved.

Factors affecting the anthrone assay
The basic aspects of the anthrone assay including the effects of anthrone heating temperature and heating time are discussed elsewhere [7]. However, as long as a set of standards is run with each anthrone batch and the samples within an anthrone batch are treated identically, most of the factors become insignificant.

Effect of SA incubation time
When a 50 mm Arabidopsis stem piece is treated with 67 % sulphuric acid, the resulting solution is largely clear for at least few hours after which it will start to go darker and will turn black eventually, depending upon the incubation time (Fig. 4). We compared cellulose content in a sample treated for 6 h with the one treated for 11 days and got identical results (Fig. 3). So extended incubation times with sulphuric acid is not an issue.

Sulphuric acid age
We noticed that the stability of anthrone was dependent on the age of sulphuric acid. If we use acid from a 2 year old bottle, the colour of 0.3 % anthrone goes darker a lot quicker (within 10 min) as compared to acid from a fresh bottle which keeps the 0.3 % colour light yellow a lot longer (several hours).

Further adaptations of the method
Recently, there have been methods published that report comprehensive analysis of cell wall components; for example Pettolino et al. [14] and Foster et al. [15]. Most of these methods usually involve relatively larger amounts of cell wall material which is ground into a fine powder. For example, Foster et al. [15] typically start with 60-70 mg dried AIR. Our analysis on the other hand typically stats with 10 mg stem material for wild type samples and 2 mg for irx mutants. The smaller amount of starting sample amounts and the deliberate avoiding of grinding steps to increase throughput means that not all of the side-analyses can be incorporated into our method. However, we do anticipate that an extraction of intact stems with 2 M trifluoroacetic acid (TFA) prior to acetic/nitric treatment can be performed to yield material for analysis of noncellulosic polysaccharides with GC-MS [15]. The time of some steps could be further reduced with the use of additional equipment. For example, although our method means it is convenient to carryout the drying steps overnight; a vacuum desiccator or speed vac could be used for the drying steps to reduce time.
Another potential time saving improvement could be performing the anthrone assay in a 96-well plate format. This will need an appropriate plate reader to be available which could be used with strong acids (67 % sulphuric acid). Additionally, it will also need 96-well plates that are resistant to 67 % acid at high temperature. However, both these variations have been previously used [15].
• CAUTION. Strong smelling acid. Work in fume hood.
• CAUTION. Concentrated strong acid. Wear appropriate gloves and work in fume hood. The bottle will get very hot; keep on ice and cool for at least 2 h. Work in fume.
Glucose standards-make 100 mg/mL glucose stock. Dilute to 10 mg/mL which can be used for preparation of a set of standards, 0-100 µg/mL. Use the following  Arabidopsis plants were first grown on ½ MS plates for 7 days in an incubator and then transplanted on a 1:1:5 mixture of perlite, vermiculite and compost. Plants were grown for a further 7 weeks on soil under long day conditions (16 h/8 h day/night, 22C/18C temperature and 80 % humidity). When plants were 7-8 weeks old, primary inflorescence stem was harvested by severing above the rosette level. Cellulose content was analysed with the detailed step by step protocol described below.

Equipment setup
Aspirator trap setup-connect the trap outlet to a pressure device. To the inlet, attach a drawn out glass Pasteur pipette to ensure no sample could be "sucked in".

Weighting of AIR and transfer to glass tubes: TIMING up to 6 h for 200 samples
7. Determine dry weight of wall material (AIR). Dried stem pieces are strong enough to be handled with a pair of forceps; weigh them in weigh boats on a fine balance. After weighing transfer stem pieces to the pre-labelled glass tubes. Use metal racks to store the tubes (as in step 10, tubes will need incubation in a water bath). To save time do not cap the tubes. If they need storing, keep the tubes on racks and cover with cling film to avoid dust going in. • PAUSE POINT. Weighted samples can be stored at room temperature until a later date when you are ready to perform next step. 8. At this step also include a filter paper sample (about 10 mg weight), which acts as a positive control.

Acetic/nitric extraction: TIMING up to 8 h for 200 samples
9. Carefully add 3 mL of acetic/nitric reagent to the wall material. Cap the tubes with PTFE seal caps. Photographs of four samples with a range of starting sample weight are shown in Fig. 5. These photographs show how the material becomes translucent and fragile after the acetic/nitric treatment.
• CAUTION. Work in a fume hood and wear appropriate PPE. • CRITICAL STEP. Only use PTFE seal caps as the latex seals will disintegrate in the next step. 10. Place the tubes in a boiling water bath for 30 min.
Allow to cool on bench. Aspirate off the acetic-nitric reagent. After this step, the samples become gelatinous and fragile, so handle carefully.
• CAUTION. Boiling water, work in a fume hood and use a long pair of forceps to take samples in and out of boiling water bath. • Maintain water level in the bath. • Discard acetic/nitric reagent appropriately. 11. Add 8 mL of water. Keep on bench for 15 min and then aspirate off the water. • CRITICAL STEP. You are most likely to lose material during this step because larger volume makes the pieces float around more freely. Start aspirating slowly and let the pieces settle on the walls before removing all the liquid. 12. Add 4 mL of acetone. Let the tubes stand for 5 min and aspirate off the acetone. 13. After aspirating acetone, push the pieces down to the bottom of the tube using a blunt end glass rod. • CRITICAL STEP. It is extremely important to push the pieces down when they are still wet. If they dry a b c d e f g h i j Step 9 starƟng material Step 9 AŌer adding aceƟc/nitric Step 10 AŌer heaƟng Step 10 AŌer aspiraƟng off aceƟc/nitric Step 11 AŌer adding water Step 11 AŌer removing water Step 12 AŌer adding acetone Step 12 AŌer removing acetone Step 13 Push samples down with glass rod when they are sƟll wet Step 14 Samples dried at the boƩom of the tube Fig. 5 Sample processing through various stages of acetic/nitric treatment (a-j). One representative sample with a starting sample weight of 7 mg is shown (a). The wild type samples tend to be thicker and are much easier to track along the various steps. However, the irx mutant samples are comparatively thinner and they also tend to be more "wonky" making them relatively difficult to follow around during the aspiration steps. Also they go even more translucent than the wild samples making them more difficult to spot. So more care is needed in handling such samples. In step 13 (i), it is critical to push the samples down to the bottom of tube when they are still wet. Once dried at the bottom (j), they will be accessible to H 2 SO 4 in step 16 (Fig. 3) too high up in the glass tube, they will be inaccessible to the sulphuric acid in the next step. 14. Air-dry tubes in a fume hood for 3-4 h.

Anticipated results
The protocol described is very robust and produces reproducible results (Figs. 1, 2, 3). A 50 mm long Arabidopsis stem piece collected from 7-to 8-week-old plants will weigh 2-10 mg in step 7 depending upon the genotype of the plants. Pieces from wild type plants will weigh closer to 10 mg while the cellulose deficient mutants are around 2 mg. Considering that the wild type samples contain about 30 % cellulose and the cellulose deficient mutants about 10 % cellulose in their walls, using 20 µL sample in step 19 would ensure that the OD620 values will fall within the standard curve range (10-100 µg/mL glucose).

Timing
We routinely analyse 200 samples per week. The whole protocol can be broken into five parts with an overnight stopping point after each part. Day

Troubleshooting
Troubleshooting advice can be found in Table 1.