GC-MS ANALYSIS OF CARBOXYL-REDUCED METHYLATED ALDITOL ACETATES
 
Last Update: December 2006
 
Based on the protocol developed in the laboratory of Dr. Tony Bacic. Reference: Kim and Carpita (1991) Changes in esterification of the uronic acid groups of cell wall polysaccharides during elongation of Maize coleoptiles. Plant Physiol. 98: 646-653
 
PREPARE SOLUTIONS
Reagents (it is of up-most importance that all chemicals are pure and devoid of any contamination. Wear powder-free gloves and keep away from dust)
1. Imidazole (99%):
Aldrich Chemicals, I-20-2
2. Hydrochloric acid:
Analytical reagent
3. 2-[n-morpholino] ethane sulphonic acid (Mes):
Sigma, M-8250
4. 1-cyclohexyl-3-(2-morpholino) carbodiimide-metho-p-toluene Sulphonate (carbodiimide):
Aldrich, C 10,640-2
5. Glacial acetic acid:
BDH Aristar Grade, 45001
6. Dialysis tubing:
MW cut off 3,500 Spectra/Por
7. 1M imidazole (pH 7.0):
Dissolve 17.02 g of imidazole in dH2O. Adjust to pH 7.0 and add water to 250 mL
8. 0.2M Mes (pH 4.75):
Dissolve 3.9 g of Mes in 50 mL of dH2O. Adjust to pH 4.75 with 1.5M NaOH and add dH2O to 100 mL
9. Methanol:
BDH Aristar Grade, 45102
10. Dimethylsulphoxide (DMSO):
Merck Scintillation grade, 2951
11. Methyl iodine:
Flucka, 67690 / Aldrich, 28956-6
12. Sodium Hydroxide:
Analytical reagent
13. Sodium Thiosulphate
Analytical reagent
14. Acetic acid (glacial):
BDH Aristar Grade, 45001
15. Chloroform:
BDH Aristar Grade, 45113
16. Dichloromethane:
Merck Uvasol Grade, 6048
17. Perchloric acid:
Ajax Univar Grade, 1873
18. Ethyl acetate:
Analytical reagent
19. Sulphuric acid
Analytical reagent
20. Ammonia (35% NH3):
BDH Aristar Grade, 45200
21. Toluene:
Analytical reagent
22. Trifluoroacetic acid (TFA):
Pierce Sequanal Grade, 28903
23. Sodium borodeuteride:
Sigma, S-2882
24. 1-methylimidazole:
Flucka, 67560
25. Acetic anhydride:
BDH Analar Grade, 10002
26. myo-inositol:
Sigma, I-5125
27. Nitrogen:
C.I.G. Industrial Grade
28. 0.25M Sulphuric acid:
Dilute 1.37 mL of concentrated H2SO4 to 100 mL with 93% v/v acetic acid
29. 2M Ammonia:
Dilute 5.52 mL of ammonia (35%) to 50 mL with dH2O
30. 2.5M Trifluoroacetic acid (TFA):
Dilute 9.28 mL of sequanal grade TFA to 50 mL with dH2O
31. 2.0M Trifluoroacetic acid (TFA):
Dilute 7.42 mL of sequanal grade TFA to 50 mL with dH2O
32. 1M Sodium Borodeuteride:
Prepare sufficient solution by adding 41.87 mg of NaBD4 per 1 mL of 2M ammonia
33. 5% Acetic acid in Methanol:
Dilute 5 mL acetic acid in 100 mL methanol
   
Carboxyl Reduction
   
PROCEDURE - Step 1 - Carboxyl Reduction
Rationale: to distinguish between neutral, uronic and methoxylated sugars as well as to distinguish among methylated sugars with symmetrical substitutions that result in identical spectra (for example: 2,3 and 3,4 di-o-methyl pentinols). SODIUM BORODEUTERIDE (NaBD4) introduces asymmetry in otherwise identical molecules thanks to its larger weight as compared to hydrogen. The resulting derivates can be identified by a fragmentation pattern, with a shift to a higher weight of the molecule containing the deuteride at the C-1 carbon. Note also that only neutral sugars can be detected by these methods, which is why all sugars must be neutralized
Carboxylic esters are reduced to their 6,6-dideuteriosugars, which can then be distinguished from their neutral sugars as fragments with M+ +2, where M+ is the mass of the neutral sugar and the +2 is the added mass from the D2
Free uronic acids are activated with carbodiimide ad reduced with either NaBD4 (to give total uronic acids) or with NaBH4 (to give the proportion of esterified uronic acids when compared to the total uronic acids)

Carboxylic esters and free uronic acids are also reduced with NaBH4, but only as controls (all detected as neural sugars)
1. In a 50 mL screw-cap tube dissolve up to 20 mg of cell wall material in 5 mL of ice cold 1M imidazole-HCl, pH 7.0. Due to the nature of plant cell wall polysaccharides it is very likely that complete dissolution will not take place. Instead, try to treat all samples equally by stirring vigorously for 10-30 minutes on ice
2. Reduce samples by adding 3X 1 mL aliquots of freshly prepared 100 mg/mL sodium borodeuteride in water (Aldrich 20,559-1). Vortex and incubate for 5 minutes after the first two additions and for 30 minuters after the thirds addition of reductant
3. Destroy excess reductant by SLOWLY adding 500 mL of glacial acetic acid and dialyze samples overnight (16-40 hours) against distilled water. To assure complete dialysis, changed water at least once. Use a membrane with a molecular weight cut-off of 3,000 Da (Regenerated cellulose, Cellu-Sep, Membrane Filtration Products, San Antonio, Texas)
4. Freeze-dry samples (about 24 hours)

5. Using a 50 mL tube, dissolve samples in 1 mL distilled water, add 200 mL Mes and 400 mL of freshly prepared 500 mg/mL carbodiimide (1-cyclohexyl-3-92-morpholino-ethyl) carbodiimide metho-p-toluenesulfate, Aldrich C10,640-2). The carbodiimide may require warming to solubilize

6. Vortex and incubate for 3 hours at 25-30oC
7. Add 1 mL of 4M imidazole-HCl, pH 7.0 and cool on ice
8. Split samples equally into TWO separate 50 mL tubes. Label one D/H (for treatment with NaBD4 followed by NaBH4) and the other D/D (for treatment with NaBD4 followed by NaBD4)

9. Add 1 mL of freshly prepared 100 mg/mL sodium borohydride/deuteride in water over 2 hours (4 x 250 mL)
10. Detroy excess reductant by SLOWLY adding 500 mL glacial acetic acid, dialyze as before and freeze-dry the samples
11. Proceed to monosaccharide analysis or methylation
   
PROCEDURE - Step 2 - Analysis
D/H: unlabeled peaks (for example a peak at MW 205) will represent neutral (N) plus uronic (U) sugars (N+U), while a labeled peak (in this case at MW 207) is representative of methoxy-labeled (M) sugars
D/D: unlabeled peaks (for example a peak at MW 205) will represent only neutral (N) sugars, while a labeled peak (in this case at MW 207) is representative of uronic (U) plus methoxy-labeled (M) sugars
   
Methylation
   
PROCEDURE - Step 2 - Methylation

A mini-prep and a macro-prep procedure are presented. The first number/amount representes the mini-prep amount. The amount in (brakets) represents the macro size/amount
VERY IMPORTANT: from this point on, all equipment used (namely tubes and tips) should be made of glass
Glassware: use 7 (15-35) mL chromic acid washed kimax screw capped tubes with lids that screw on smoothly
NOTE: be careful about putting anything plastic or metal into the chromic acid, it might not come out again :)

NOTE: Chromic acid has to be clean and relatively fresh. In general Cr-acid baths tend to accumulate a lot of debris from repeated use and those may stick to the inner surface of the narrow tubes
**SAFETY**: Chromic acid is highly corrosive and more than one lab person has lost his/her eyes to this!
Chromic acid wash: soak in chromic acid overnight. Then rinse in dH2O followed by ultrapure dH2O then air dry. Wrap tip of tubes with aluminum foil and autoclave
1. Resuspend freeze-dried samples from carboxyl reduction experiment in water to a concentration of 1 mg/mL
2. Transfer 5-100 mg (100-1000 mg) of glycan sample to a glass tube and dry with a stream of nitrogen

3. Add 10-20 mL (100-200 mL) of methanol and dry again to dehydrate samples
4. Resuspend in 100 mL (500 mL) of DMSO, cap tube (teflon faced cap) and either sonicate (Branson 2200 Sonic Bath) or shake by placing on an Orbital Shaker (Paton Industries, Nodel PO1412) for >20 minutes at >120 rpms to dissolve glycans. Let stand for an hour to up to 2 days at room temperature (this swells the polysaccharides and makes them more amenable to methylation)
5. Prepare a slurry of NaOH (approximately 120 mg/mL) in DMSO by grinding 3 pellets of NaOH per 1 mL of DMSO with a glass pestle and mortar. Do not attempt to weight as the NaOH will hydrate

6. Immediately add 50 mL (500 mL) of DMSO/NaOH slurry to each sample using an SMI glass pipette. Stirr the slurry as you pipet so that it does not settle. Take care to drop the reagent straight into the glycan solution; DO NOT get the NaOH suspension on the side of the tube
7. Cap and sonicate/shake for 20-50 minutes

8. Add 10 mL (100 mL) of methyl iodide and sonicate/shake for 10 minutes. Do this in a fumehood since CH3I is a suspected carcinogen!

9. Add 20 mL (100 mL) of methyl iodide and sonicate/shake for 10 minutes
10. Add 20 mL (100 mL) of methyl iodide and sonicate/shake for 20 minutes

11. Add 1 mL (10 mL) of freshly prepared 100 mg/mL sodium thiosulphate in water and 500 mL (2 mL) of chloroform. Cap and vortex well (>40 s per sample). Centrifuge to separate the phases
12. Remove and discard the aqueous (upper) phase. Wash the lower CHCl3 phase four times with 1 mL (5 mL) of water
13. Dry the lower phase by a stream of nitrogen

14. Repeat steps 4 through 13 one more time
15. Proceed to Hydrolysis
   
Hydrolysis
   

This part of the experiment assumes that there are no hexosamines in the sample
1. Add 75 mL (500 mL) of 2.0M (2.5M) TFA to the methylated sample. Hydrolyze sample for 2 hours at 100-102oC in a fan forced oven, or autoclave for 1 hour at 121oC

2. Cool, place tubes in a ~30oC water bath and evaporate to dryness with a stream of nitrogen. The residue will frequently be an oily yellow residue

3. Add in myo-inositol as an internal standard (about 1/20th of total CHO) and dry with a stream of nitrogen
4. Proceed to Reduction
   
Reduction
   

1. Dissolve the residue in 50 mL (500 mL) of 2M NH4OH and add 50 mL (500 mL) of freshly prepared 1M NaBD4 in 2M NH4OH. Sonicate for 1 minute and incubate at room temperature for 2.5 hours (at 60oC for 1 hour)
2. Destroy excess reductant with 3 x 5 mL (50 mL) acetic acid - CAREFULLY (the reaction fizzes). If not fizzing, the reaction was incomplete or the NaBD4 was inactive. Re-reduce after step (4)

3. Dry the samples with a stream of nitrogen. The samples will not dry easily, be patient, they will eventually evaporate

NOTE: DO NOT leave this stage overnight. Proceed at least through two methanol washes and store in methanol
4. Further evaporate twice with 250 mL (2.5 mL) of 5% acetic acid in methanol and twice with 250 mL (2.5 mL) of methanol to remove the boric acid

5. Proceed to acetylation
   
Acetylation
   
1. Acetylate by adding 250 mL acetic anhydride. Sonicate for 5 minutes and incubate at 100oC for 2.5 hours
(dissolve residue in 200 mL of glacial acetic acid [the solution will not clear]. Acetylate by adding 1 mL of ethyl acetate and 3 mL acetic anhydride, mix and add 100 mL of 70% perchloric acid. Vortex thoroughly. On adding the first two reagents the solution should look like a suspension and on adding the acid the solution should clear and heat up)

2. Destroy excess acetic anhydride by adding 2 mL of water. Mix and stand for 10 minutes

(after 5 minutes, cool on ice and add 10 mL of water, followed by 200 mL of 1-methylimidazole. Mix and allow to stand for 5 minutes. The solution will heat up on addition of the 1-methylimidazole)

3. Extract partially methylated sugars 2 x with 500 mL (1 mL) dichloromethane (DCM). Vortex well and centrifuge to aid in separation (1 minute at 3,000 rpms). Transfer DCM phase to a new tube every time

4. Combine extracts and back wash 2 x with 1 mL (5 mL) aliquots of water. Transfer the DCM phase to a small glass vial and dry with a small stream of nitrogen. DO NOT over - do the drying as volatile derivatives can be lost
5. Redissolve the products in the required amount of DCM, 5 mg in 20 mL DCM (500 mg in 1 mL DCM)

6. Analyze 1 mL aliquots by GC-MS using the high polarity BPX70 column using conditions as described by Lau and Bacic 1993, J. Chrom 637: 100-103. It may be necessary to run samples on a low polarity CPSil5 column, particularly complex cell wall samples

7. Store samples at -20oC
   
GC-MS Separation and Analysis
   
PROCEDURE - Step 1 - GC-MS Separation

1. Inital: time: 2 minutes, temperature: 170oC
Column: SGE PBX70 High Polarity 25 m x 0.22 mm I.D.

Constant flow: 1.2 mL /minute

Injectin source: GC ALS
Injection temperature: 240oC

Injection volume: I mL from 10 mL syringe
2. Ramp rate: 3oC
3. Final: time: as required, temperature: between 235 to 260oC

Max column temperature: 290oC

MSD transfer line: 260oC
MS Quad: 106oC

MS Force: 230oC

Solvent delay: 3 minutes
Low Mass: 100* and High Mass: 350*

*= significant masses for sugar fragments
After all the samples are run, a chromatogram will be available in the computer for every sample. At this point what is being represented are the different sugars that migrated through the column and eluted at these different times.

Each one of the molecules represented by these picks was then ionized (bombarded with electrons and consequently broken up into separate parts) and separated by their mass-to-charge ratio (m/z). The respective fragments are obtained by breakages at various positions in the molecule. Finally, because a given specific peak elutes at a given Relative Retention Time (RRT), then it can be determined what sugar it is

   
PROCEDURE - Step 2 - Analysis
1. Divide the migration time of a peak by that of myo-inositol. Normally taking the first eluted peak is fine, but since in all cases the molecule that represents the largest amount of polysaccharide in the cell wall of an Arabidopsis leaf is cellulose, take the largest peak and divide it by that of myo-inositol. This is your starting Relative Retention Time (RRT = 0.523 for 1,4-glucose).
2. Check the spectra of this peak to determine what kind of molecule it may be. If you did as above and took the largest peak, it is very likely that it looks like 1,4-hexitol, or 1,4-glucose.

3. Repeat this process for the next peak (going up or down the spectra). Obtain the RRT as in (1) above. Then find the difference between this RRT and that of the RRT for 1,4-glucose from the BPX70 Retention Time Data Table (BRTD). Look at the spectra of this peak and find possible candidates. If you went to the closest peak (going towards the origin of the chromatogram), you would have found another 1,4-hexitol (RRT = 0.511). Looking at the BRTD Table, the only 1,4-hexitol at this RRT is 1,4-galactose. As mentioned above, this does not mean that this molecule is 1,4-galactose, the peak spectra must be analyzed to determine what exactly it is, but the BRTD table will not tell you this, it only tells you what kind of neutral sugar it is
4. Repeat this process for every single peak, always finding the difference between the RRTs from the last peak you analyzed, and determined its identity from the BRTD table, and the next peak. The difference between YOUR RRTs is then used to determine which is the next RRT in the BRTD. This will tell you which is the possible candidate sugar for your next peak. Then using the peak's spectra determine what molecule it is among the candidates