The assessment of gastrointestinal mucosal permeability and absorptive capacity provides information about gastrointestinal mucosal function. Nonmetabolizable markers, such as polyethylene glycol, sugars, or radiolabeled substances, have been widely used.1 Altered intestinal permeability has been reported in humans with inflammatory bowel disease and also in other animals with experimentally induced gastrointestinal tract disease.2 In humans and dogs, intestinal permeability generally increases with the severity of disease.3,4 Intestinal permeability has been postulated to be responsible for the introduction of antigenic or infectious agents through the intestinal mucosa, which can lead to excessive immunogenic stimulation.5,6
Currently, monosaccharides and disaccharides are widely used to assess abnormalities of gastrointestinal permeability and mucosal absorptive capacity.7 Except for mannitol, which is believed to be synthesized in only small quantities in humans8 but is not believed to be synthesized in other animals,9 both groups of sugar probes (ie, monosaccharides and disaccharides) are not considered to be endogenously synthesized in mammalian species.8 Therefore, serum concentrations of these sugars are considered to exclusively originate from gastrointestinal permeability and absorption.
Use of a mixture of monosaccharide and disaccharide probes is based on the assumption that the intestinal epithelium is a heteroporous layer.10 Aqueous pores are distributed along the crypt-villus axis of the small intestinal mucosa. Small channels (radius < 0.6 nm) are relatively abundant at the tips of the villi, which allows permeation of small molecules (such as rhamnose) but excludes the passage of larger molecules (such as lactulose).11 Therefore, molecules the size of disaccharides (eg, lactulose) are restricted from moving across the villus tip, whereas monosaccharides (eg, rhamnose) can do so through passive diffusion. In contrast, although the monosaccharide mannitol is also absorbed by simple diffusion, it does not move freely across the villus tip. The largest channels (5 to 6 nm) are paracellular, exist in low abundance, and are located in the crypts.12 These paracellular pathways at the level of the tight junctions in the crypts are believed to be the route of permeation for disaccharides such as lactulose, whereas the transcellular pathway at the level of the villi is believed to be the main route for permeation of monosaccharides.11,13
Often, a mixture of probes (rather than a single permeability probe) is used. One advantage of the use of a mixture of probes is that various probes remain intact only in specific compartments of the gastrointestinal tract, and this can help to localize the site of the intestinal damage.14 Class 1 probes (eg, sucrose) are broken down on entering the small intestine. Class 2 probes (eg, lactulose and mannitol) pass through the stomach and most of the small intestines before undergoing bacterial degradation in the distal portion of the small intestine and the colon. Class 3 probes (eg, chromium EDTA and polyethylene glycol) do not undergo metabolism or bacterial degradation and remain intact throughout the entire length of the gastrointestinal tract.14 However, because these markers permeate the mucosa throughout the entire length of the gastrointestinal tract, the signal-to-noise ratio can be high, and it may be difficult to assess permeability of the colonic mucosa with these markers.
Permeability of the intestinal mucosa can be measured effectively by use of nonmetabolizable carbohydrates such as rhamnose, which is believed to permeate the gastrointestinal mucosa by a passive transcellular route; thus, permeability of this sugar is dependent on mucosal surface area.15 Absoptive capacity can be determined with carbohydrates that are transported via passive and active carrier-mediated transport, which include D-xylose and 3-O-methyl-D-glucose, respectively.16,17 It has been reported that D-xylose is not absorbed by the intestines via a passive carrier but instead has a low affinity for the sodium-dependent carrier, predominantly in the jejunum.18,19 In contrast, 3-O-methyl-D-glucose is absorbed by active carrier-mediated transport (Na+-dependent transport) throughout the small intestines.20 These sugars are not metabolized; after IV administration, they are excreted intact in the urine.21
Many protocols have been developed for the assessment of intestinal permeability and mucosal function.8,16 Protocols that involve the use of permeability markers excreted and recovered in urine are currently the ones most widely used.22,23 Such protocols require complete collection of urine during a period of 4 to 24 hours, which is laborious and often impractical. Furthermore, incomplete collection of urine may affect test results.24 Thus, the measurement of carbohydrate markers in serum would simplify assessment of intestinal permeability and absorptive function in clinical situations because serum samples can be obtained more easily than urine samples in a practice setting.
Several methods for the quantification of carbohydrates in serum samples have been described. For example, sucrose has been measured by use of an enzymatic method in human serum.25 A method involving HPLC coupled to mass spectrometry has been used to measure sucrose in serum obtained from horses.26 Measurement of 3-O-methyl-D-glucose in human serum by use of thin-layer chromatography and densitometry has been reported.27 Lactulose and mannitol have been measured by use of HPLC-PAD in human serum.24 Finally, HPLC-PAD has been used for the measurement of lactulose, rhamnose, 3-O-methyl-D-glucose, and xylose in canine serum.28 Some problems have been encountered for most detection systems coupled to HPLC. For instance, methods based on the refractive index and evaporative light scattering detectors have poor selectivity because of the universal detection of compounds, and methods based on fluorescent detection require fluorescent compounds, which are not available for sugar probes. Electrochemical detection, in particular PAD, is more acceptable because of the high sensitivity, although it also has low detection selectivity.29 Also, HPLC-PAD is not useful for direct analysis of complex biological samples (such as serum), and serum proteins and lipids must be removed before analysis.
The use of GC-MS is considered to be a suitable method for the quantification of sugar probes because of the high sensitivity for this method.30 The resolution obtained with GC-MS results in good peak capacity, which is defined as the maximum number of nonoverlapping peaks in a specified interval.30 Methods involving the use of GC-MS have been described for the quantification of carbohydrates in urine,31,32 plasma,33 aqueous solutions,34 environmental samples, food products,35 and serum.36,37 The objective of the study reported here was to analytically validate a GC-MS method for the simultaneous quantification of lactulose, rhamnose, xylose, methylglucose, and sucrose for use in assessing gastrointestinal permeability and intestinal absorptive function in an accurate, rapid, and practical manner that could facilitate its use in research and clinical practice.
Materials and Methods
Sample population—Pooled canine serum was obtained from 200 dogs. The samples constituted leftover serum from samples submitted to the Gastrointestinal Laboratory at the Texas A&M College of Veterinary Medicine and Biomedical Sciences; therefore, the study did not require approval from the Clinical Research Review Committee.
Standard solutions—Stock solutions that contained each of the 5 sugar probes (lactulose,a rhamnose,b 3-O-methylglucose,c xylose,d and sucrosee) were prepared by dissolving solutions of each of the sugars in pooled canine serum. Eleven standard solutions (500.0, 250.0, 125.0, 62.5, 31.3, 15.6, 7.8, 3.9, 2.0, 1.0, and 0.5 mg/L) were prepared for each of the 5 sugars by use of serial dilution (1:2 dilution). Mannitolf was used as an internal standard and was added to each standard solution at a final concentration of 100 mg/L.
Calibration curves were established by plotting the ratios of the area under the curve of the peaks of the sugar of interest to that of the internal standard for the various standard solutions by use of a polynomial curvilinear regression as follows: y = (a•x2) + (b•x) + c, where y is the concentration (in mg/L) for a specific sugar in the unknown sample; x is the relative abundance of each sugar in the standard solution (determined as the area under the curve of the peaks of the sugar of interest); a and ≤ are regression coefficients for the quadratic and linear terms of x, respectively; and c is the intercept for the value of y when x = 0. The ratio for the area of each sugar peak for unknown serum samples was then extrapolated from the calibration curves.
Extraction and derivatization of sugars—Serum aliquots (200 ML) were mixed with 22.2 ML of an aqueous internal standard solution (mannitol at a final concentration of 100 mg/L) in 2-mL plastic vials. Then, 600 ML of methanolg was added to each serum sample, and vials were mixed for 20 seconds by use of a vortex mixer. Samples were centrifugedh at 2,655 X g for 7 minutes, and the protein-free supernatants were transferred into 4-mL screw-cap glass vials.i Samples were evaporated to dryness under a stream of nitrogen in a heating modulej at 64°C for 30 minutes.
Vials were allowed to cool, and the dried residue was derivatized in a 2-step procedure. First, 50 ML of 2% methoxyamine hydrochloride in pyridinek and 70 ML of pure pyridinel were added to each vial. Vials were capped, vortexed for 20 seconds, and heated in a microwave for 2 minutes to promote the oximation reaction.37 Samples were then allowed to cool for 5 minutes, and 100 ML of N,O-bis[TMS]trifluoroacetamidel containing 1% trimethylchorosilanel was added to each vial. The vials were capped, vortexed for 20 seconds, and heated in a microwave for an additional 5 minutes to develop the silylation reaction, in accordance with the procedure described in another study.38 Samples were then allowed to cool to ambient temperature, and derivatized extracts were evaporated to dryness under a stream of nitrogen gas at 64°C for 8 minutes. Residues were dissolved in 250 ML of hexane,m and GC-MS analysis was performed with 1 ML of this solution.
Gas chromatography—Derivatized samples were analyzed by use of a gas chromatographn coupled with a mass spectrometer.o Aliquots (1 ML) of extracts were injected into a split-splitless inlet (which was operated in splitless mode) at an inlet temperature of 250°C. Separation of sugars was achieved by use of a dimethylpolysiloxane capillary columnp (length of 30 m, inner diameter of 250 μm, and film thickness of 0.25 μm). A temperature program for the gas chromatography column was used. Initial oven temperature was set at 100°C and maintained for 5 minutes. The temperature was then increased to 325°C by use of a constant gradient of 15°C/min. The temperature then was maintained at 325°C for 5 minutes, which resulted in a total run time of 23.33 min/sample. Helium was used as a carrier gas at a constant flow rate of 1.5 mL/min at a velocity of 33 cm/s. The qualitative analysis was performed in fullscan acquisition mode within a range of m/z 50 to m/z 1,050. Quantification of sugar concentrations was performed with the sum of peak areas from specific retention times of each sugar by use of SIM mode. Selected ions for the sugars were m/z 204 for lactulose, m/z 117 for rhamnose, m/z 147 for methylglucose, m/z 217 for sucrose, m/z 217 for xylose, and m/z 217 for mannitol.
Chromatograms—The derivatization procedure was evaluated by comparing the area under the curve for peaks generated from spiked serum samples with those generated from the same amount of the pure compound previously analyzed by use of the same derivatization and gas chromatography conditions. Sugars were identified by matching their chromatographic retention times and their characteristic mass spectrum. To verify interference by other carbohydrates, an aqueous solution containing glucose, fructose, fucose, sucralose, mannitol (used as an internal standard), lactulose, rhamnose, methylglucose, xylose, and sucrose was analyzed.
Validation—The method was validated by evaluating spiking recovery, dilutional parallelism, intraassay variability, and interassay variability. For spiking recovery (accuracy of the assay), 4 pooled serum samples obtained from healthy dogs were spiked with each sugar to achieve a final serum concentration of 3, 30, 100, and 350 mg/L, respectively. For dilutional parallelism (linearity of the assay), the spiked serum sample containing 350 mg/L of each of the sugars and a sample with unknown sugar concentrations (ie, unknown sample) were diluted 1:2, 1:4, and 1:8 with canine serum. Results for spiking recovery and dilutional parallelism were expressed as O/E ratios.
The aforementioned 4 spiked serum samples (3, 30, 100, and 350 mg/L, respectively) were used to evaluate the precision and reproducibility of the assay. To determine precision, intra-assay variability was assessed by assaying these 4 serum samples when spiked with various concentrations of the 5 sugar probes 9 consecutives times within a single GCMS run. The CV was calculated for each of the 5 sugars. To determine reproducibility, interassay variability was evaluated by assaying the 4 serum samples spiked with various concentrations of the 5 sugar probes 9 times on different days.
Results
The GC-MS separation of the 5 sugars resulted in clearly resolved peaks. A chromatogram obtained from the analysis of a pooled serum sample revealed no peak at the retention time for mannitol of 14.06 minutes (Figure 1). This indicated that canine serum does not contain detectable concentrations of mannitol. Chromatograms were obtained for a standard solution containing each of the 5 sugars and an unknown canine serum sample to which mannitol was added (Figure 2). Each sugar had a unique elution profile and retention time, which was evident in all GC-MS validation assays. Analysis of the chromatograms revealed a good peak capacity, which was evident during all assays. Addition of D-glucose, fructose, and sucralose did not cause any chromatographic interference with the target sugars. However, addition of fucose led to peaks that coeluted with the peaks for rhamnose and 3-O-methylglucose. Therefore, we chose not to use fucose as an internal standard.
Total ion chromatogram for a blank serum sample of pooled canine serum.
Citation: American Journal of Veterinary Research 70, 3; 10.2460/ajvr.70.3.320
Total ion chromatograms for a canine serum sample containing a standard solution of a mixture of 5 sugars (125 mg/L for xylose [X], rhamnose [R], 3-O-methyl-D-glucose [3M] sucrose [S], and lactulose [L]; A) and an unknown canine serum sample (B). Each sample was spiked with mannitol (m) at a concentration of 125 mg/L as an internal standard.
Citation: American Journal of Veterinary Research 70, 3; 10.2460/ajvr.70.3.320
The GC-MS characteristics for all sugar probes were determined. Retention times and optimum m/z values were determined for the peaks and major ions, respectively, which were used for quantification (Table 1). The remainder of the ions was evident for the fragmentation pattern detected in full-scan mass spectrometry mode used for the identification of the carbohydrates (data not shown). Mass spectra of the analyzed carbohydrates were dominated by ions at m/z 73, 117, 147, 204, 217, 361, 437, and 451, and the retention times were constant for the gas chromatography conditions and derivatization procedures performed in this study.
Characteristics of 5 sugars and mannitol analyzed by use of GC-MS as TMS derivatives.
| Compound | Molecular mass (kDa) | Retention time (min) | m/z |
|---|---|---|---|
| Lactulose | 342 | 18.01, 18.42 | 73, 147, 204,* 361 |
| Sucrose | 342 | 17.75, 18.12 | 73, 217,* 361, 437 |
| Xylose | 150 | 11.84, 11.90 | 73, 147, 217* 307 |
| Rhamnose | 164 | 12.45, 12.50 | 73, 117* 219, 277 |
| Methylglucose | 194 | 12.70, 13.16, 13.45 | 73, 147* 205, 262 |
| Mannitol | 182 | 14.06 | 73, 205, 217* 319 |
Retention time indicates chromatogram peaks of each sugar and mannitol, and the m/z depicts the GC-MS fragmentation ion pattern of each sugar and mannitol.
The m/z of the peak selected for quantification of each sugar and mannitol.
Chromatograms were obtained for each of the 5 sugars analyzed in SIM mode. The SIM mass spectra with the respective quantification ion for each sugar were also obtained. Analysis of the chromatogram for lactulose revealed a minor peak that eluted at 18.01 minutes, whereas the 2 main peaks eluted at a retention time of 18.42 and 18.44 minutes, respectively (Figure 3). The ion with m/z 204 was used for analysis and quantification of this disaccharide. Analysis of the chromatogram for sucrose revealed a minor peak at a retention time of 17.84 minutes, whereas the main peak had a retention time of 18.12 minutes (Figure 4). Although retention time and m/z for peaks of sucrose and lactulose were similar, the derivatization protocol used in the study led to sufficient and consistent separation for the differentiation of these 2 carbohydrates. Analysis of the chromatogram for rhamnose revealed peaks with retention times of 12.43 and 12.50 minutes, respectively (Figure 5). Analysis of the chromatogram for xylose revealed retention times for the peaks of 11.84 and 11.90 minutes (Figure 6). Analysis of the chromatogram for methylglucose revealed peaks with retention times of 12.70, 13.16, and 13.45 minutes (Figure 7). The m/z values are in agreement with other results obtained by use of the same derivatization reagents.
The GC-MS chromatograms for lactulose depicting results for the selected ion chromatogram (A) and the mass spectrum of the TMS derivative of lactulose at m/z 204 (B). In panel A, the numbered peaks indicate the retention times for the derivatized ions of lactulose.
Citation: American Journal of Veterinary Research 70, 3; 10.2460/ajvr.70.3.320
The GC-MS chromatograms for sucrose depicting results for the selected ion chromatogram (A) and the mass spectrum of the TMS derivative of sucrose at m/z 217 (B). In panel A, the numbered peaks indicate the retention times for the derivatized ions of sucrose.
Citation: American Journal of Veterinary Research 70, 3; 10.2460/ajvr.70.3.320
The GC-MS chromatograms for rhamnose depicting results for the selected ion chromatogram (A) and the mass spectrum of the TMS derivative of rhamnose at m/z 117 (B). In panel A, the numbered peaks indicate the retention times for the derivatized ions of rhamnose.
Citation: American Journal of Veterinary Research 70, 3; 10.2460/ajvr.70.3.320
The GC-MS chromatograms for xylose depicting results for the selected ion chromatogram (A) and the mass spectrum of the TMS derivative of xylose at m/z 217 (B). In panel A, the numbered peaks indicate the retention times for the derivatized ions of xylose.
Citation: American Journal of Veterinary Research 70, 3; 10.2460/ajvr.70.3.320
The GC-MS chromatograms for methylglucose depicting results for the selected ion chromatogram (A) and the mass spectrum of the TMS derivative of methylglucose at m/z 147 (B). In panel A, the numbered peaks indicate the retention times for the derivatized ions of methylglucose.
Citation: American Journal of Veterinary Research 70, 3; 10.2460/ajvr.70.3.320
Assay validation—Calibration curves were evaluated for the 5 sugar probes (Figure 8). Each sugar standard was prepared with 11 calibration concentrations covering the range of expected values in the samples. Within the range of 0.5 to 500 mg/L, sugar concentrations were proportional to their integrated peak areas. Similarly, the concentration of the internal standard was proportional to its integrated peak area. Because of possible instrument variability and analyte sensitivity (ie, changes in temperature, pressure, and the electronic detector), obtaining a ratio of areas is more reproducible than use of absolute values for each compound. Thus, we assumed that unknown samples, sugar standards, and internal standards were processed analogously and were exposed to similar factors. Therefore, standard curves established with an internal standard minimized the effect of variability among assays. Calibration curves used the ratio (rather than just the peak area) of the analytes for quantification. In this case, the ratio of the concentrations for a specific unknown sample and the internal standard was proportional to the ratio of their peak areas. All curves had good coefficients of correlation. The best-fit line for the ratio of integrated peaks against the ratio of the concentration for each sugar to the concentration for the internal standard was fitted by use of polynomial regression. The curves were highly correlated (mean r2 = 0.997) for all sugars. The polynomial second-order model used in this study has an advantage over the straight-line model in that the former integrates an intercept line with the experimental background noise of the blank sample. Therefore, with a polynomial model, the intercept line passes through zero, which is useful to define detection limits and to properly measure concentrations in unknown samples.39
Representative standard curves of the GC-MS assays for the simultaneous determination of 5 sugar probes (rhamnose [squares], lactulose [circles], methylglucose [triangles], sucrose [crosses], and xylose [diamonds]) in canine serum. The r2 values were 0.9991 for rhamnose, 0.9983 for lactulose, 0.9975 for methylglucose, 0.9996 for sucrose, and 0.9996 for xylose.
Citation: American Journal of Veterinary Research 70, 3; 10.2460/ajvr.70.3.320
The lower limit of detection of the assay was 0.03 mg/L for lactulose, rhamnose, xylose, and methylglucose and 0.12 mg/L for sucrose. These limits compare favorably with those reported for HPLC-PAD analysis in 1 study29 (xylose, 0.13 mg/L; rhamnose, 0.36 mg/L; sucrose, 0.02 mg/L; and mannitol, 0.12 mg/L), HPLCPAD analysis in 2 other studies40,41 (lactulose, 0.4 to 0.8 mg/L), or capillary electrophoresis in another study42 (lactulose, 10 mg/L).
Accuracy of the assay was determined by evaluation of spiking recovery of samples that contained 3, 30, 100, and 350 mg/L, respectively, for each sugar in serum (Table 2). The O/E ratios for spiking recovery of the 5 sugars ranged from 96.7% to 123.3% (mean ± SD, 108.4 ± 13.5%) for lactulose, 93.2% to 147.7% (mean, 124.7 ± 26.4%) for rhamnose, 88.3% to 120.1% (mean, 95.5 ± 16.5%) for methylglucose, 84.9% to 143.3% (mean, 110.4 ± 24.2%) for xylose, and 56.7% to 115.1% (mean, 92.1 ± 27.2%) for sucrose. Mean for all 5 sugars ranged from 92.1% to 124.7% (mean, 106.2 ± 13.0%). Linearity of the assay was determined by evaluating dilutional parallelism. The O/E ratios for dilutions of 1:2, 1:4, and 1:8 ranged from 67.6% to 164.2% (mean ± SD, 105.6 ± 25.4%; Table 3). Intra-assay precision (ie, CV) ranged from 6.8% to 12.9% for the 5 sugars at 4 concentrations (ie, 3, 30, 100, and 350 mg/L; Table 4). Reproducibility of the assay was determined by evaluating interassay precision. The CV ranged from 6.4% to 15.9% for the 5 sugars at the 4 concentrations (Table 5).
Spiking recovery for each of the 5 sugars spiked into canine serum at 4 concentrations.
| Sugar | Expected (mg/L) | Observed (mg/L) | O/E ratio* (%) |
|---|---|---|---|
| Lactulose | 3 | 3.7 | 123.3 |
| 30 | 34.9 | 116.3 | |
| 100 | 97.2 | 97.2 | |
| 350 | 338.4 | 96.7 | |
| Mean ± SD | NA | NA | 108.4 ± 13.5 |
| Rhamnose | 3 | 4.4 | 147.7 |
| 30 | 43.5 | 145.1 | |
| 100 | 112.6 | 112.6 | |
| 350 | 326.3 | 93.2 | |
| Mean ± SD | NA | NA | 124.7 ± 26.4 |
| Methylglucose | 3 | 3.6 | 120.1 |
| 30 | 25.4 | 84.7 | |
| 100 | 88.9 | 88.9 | |
| 350 | 313.5 | 88.3 | |
| Mean ± SD | NA | NA | 95.5 ± 16.5 |
| Xylose | 3 | 4.3 | 143.3 |
| 30 | 31.7 | 105.7 | |
| 100 | 107.5 | 107.5 | |
| 350 | 309.2 | 84.9 | |
| Mean ± SD | NA | NA | 110.4 ± 24.2 |
| Sucrose | 3 | 1.7 | 56.7 |
| 30 | 34.5 | 115.1 | |
| 100 | 111.7 | 111.7 | |
| 350 | 297.1 | 84.9 | |
| Mean ± SD | NA | NA | 92.1 ± 27.2 |
Calculated as (observed value/expected value) X100.
NA = Not applicable.
Results of dilutional parallelism for a blank canine serum sample spiked with each of the 5 sugars (350 mg/L) and of a canine serum sample spiked with an unknown quantity of each sugar.
| Sugar | Dilution | Serum sample spiked with 350 mg/L | Unknown serum sample | ||||
|---|---|---|---|---|---|---|---|
| Observed (mg/L) | Expected (mg/L) | O/E ratio* (%) | Observed (mg/L) | Expected (mg/L) | O/E ratio* (%) | ||
| Lactulose | Undiluted | 318.4 | — | — | 10.2 | — | — |
| 1:2 | 142.7 | 159.2 | 89.7 | 6.2 | 5.1 | 121.2 | |
| 1:4 | 69.4 | 79.6 | 87.2 | 3.6 | 2.6 | 149.0 | |
| 1:8 | 30.7 | 39.8 | 77.7 | 2.1 | 1.3 | 164.2 | |
| Rhamnose | Undiluted | 299.4 | — | — | 63.9 | — | — |
| 1:2 | 166.1 | 149.7 | 110.9 | 32.0 | 31.8 | 100.8 | |
| 1:4 | 81.1 | 74.9 | 108.3 | 13.7 | 15.9 | 86.4 | |
| 1:8 | 38.6 | 37.4 | 103.1 | 5.4 | 7.9 | 67.6 | |
| Methylglucose | Undiluted | 330.7 | — | — | 99.9 | — | — |
| 1:2 | 185.3 | 165.3 | 112.1 | 49.7 | 50.0 | 99.6 | |
| 1:4 | 88.7 | 82.7 | 107.3 | 23.3 | 25.0 | 93.2 | |
| 1:8 | 40.3 | 41.3 | 97.6 | 10.5 | 12.5 | 84.4 | |
| Xylose | Undiluted | 304.1 | — | — | 106.8 | — | — |
| 1:2 | 160.9 | 152.1 | 105.8 | 55.5 | 53.4 | 104.0 | |
| 1:4 | 76.5 | 76.0 | 100.6 | 24.1 | 26.7 | 90.2 | |
| 1:8 | 36.5 | 38.0 | 96.0 | 10.9 | 13.4 | 81.6 | |
| Sucrose | Undiluted | 290.2 | — | — | 2.3 | — | — |
| 1:2 | 154.5 | 145.1 | 106.5 | 1.3 | 1.1 | 110.7 | |
| 1:4 | 84.9 | 72.5 | 117.0 | — | 0.6 | — | |
| 1:8 | 41.5 | 36.3 | 114.3 | — | 0.3 | — | |
— = Not determined.
See Table 2 for remainder of key.
Intra-assay variability of 4 serum samples containing each of 5 sugars at 4 concentrations and analyzed 9 times within the same assay.
| Sugar | Sugar concentration (mg/L) | |||||||
|---|---|---|---|---|---|---|---|---|
| 3 | 30 | 100 | 350 | |||||
| Mean (mg/L) | CV* (%) | Mean (mg/L) | CV* (%) | Mean (mg/L) | CV* (%) | Mean (mg/L) | CV* (%) | |
| Lactulose | 3.3 | 9.1 | 46.5 | 7.9 | 131.9 | 6.8 | 281.3 | 12.9 |
| Rhamnose | 3.4 | 11.8 | 43.0 | 8.9 | 123.3 | 7.1 | 364.9 | 12.8 |
| Methylglucose | 2.7 | 11.1 | 33.7 | 8.9 | 125.3 | 10.8 | 377.6 | 11.5 |
| Xylose | 2.9 | 10.3 | 38.4 | 11.0 | 136.9 | 7.2 | 384.0 | 11.0 |
| Sucrose | 3.1 | 9.7 | 26.4 | 9.4 | 96.7 | 12.0 | 352.5 | 8.4 |
The CV was calculated as (SD/mean) × 100.
Interassay variability of 4 serum samples containing each of 5 sugars at 4 concentrations and analyzed in 9 consecutive assays.
| Sugar | Sugar concentration (mg/L) | |||||||
|---|---|---|---|---|---|---|---|---|
| 3 | 30 | 100 | 350 | |||||
| Mean (mg/L) | CV* (%) | Mean (mg/L) | CV* (%) | Mean (mg/L) | CV* (%) | Mean (mg/L) | CV* (%) | |
| Lactulose | 3.7 | 7.0 | 34.9 | 9.6 | 97.2 | 11.5 | 338.4 | 8.0 |
| Rhamnose | 4.4 | 6.9 | 43.5 | 7.8 | 112.6 | 9.4 | 326.3 | 6.4 |
| Methylglucose | 3.6 | 11.8 | 25.4 | 12.4 | 88.9 | 15.9 | 313.5 | 7.0 |
| Xylose | 4.3 | 7.3 | 31.7 | 13.2 | 107.5 | 8.7 | 309.2 | 6.8 |
| Sucrose | 1.7 | 8.1 | 34.5 | 9.4 | 111.7 | 5.5 | 297.1 | 8.1 |
See Table 4 for remainder of key.
Discussion
Several methods have been used for the measurement of sugar markers in serum and urine to assess the permeability and absorptive capacity of the gastrointestinal tract. However, some of them are associated with technical drawbacks.43–45 Thin-layer chromatography is a time-consuming method.43 Colorimetric-enzymatic methods do not provide information on the composition of monosaccharides,44 and HPLC methods are considered to have a relatively low sensitivity.45 We developed a GC-MS method that identifies carbohydrate markers in serum samples from dogs. Mass spectrometry was used because of its capability of high sensitivity for molecular identification on the basis of retention time and fragmentation pattern.32,46
Because saccharides are highly polar and have low volatility, chemical derivatization is required before GC-MS analysis.47–50 Thus, lactulose, rhamnose, methylglucose, xylose, and sucrose were converted into more volatile and thermostable sugar derivatives before analysis. A 2-step derivatization procedure was used. First, carbonyl groups (–C=O) were transformed to more stable, nonpolar groups (–C=N–O–CH3) by an oximation reaction, which has been described elsewhere.51 This was followed by the formation of TMS esters by use of silylating reagents to replace exchangeable protons (–OH) with TMS. The oxime formation is required to eliminate undesirable slow and reversible silylation reactions with carbonyl groups, whose products can be thermally labile, and silyl derivative groups [–Si(CH3)3] to allow measurement of the analyte by use of GC-MS.31 Carbohydrates used in this study resulted in several peaks on the chromatograms, but only 2 or 4 peaks were detectable. Sugars in solution constantly cycle between the ring and straight-chain forms, which leads to a dynamic equilibrium between the 2 forms. Therefore, 2 anomers (A and B) can be formed. When the anomeric center is not destroyed by the derivatization procedure, acylation of the aldose ring freezes the structure of the A and B anomeric form, which yields multiple peaks from 1 compound during gas chromatographic analysis.52 When profiles with multiple peaks are evident on the GS-MS chromatogram, analysis is performed by use of previously generated standard references that contain the retention time and mass spectra. Formation of anomers can potentially be avoided by preparing the alditol acetate derivative via treatment with sodium borohydride.32 The alditol acetate derivate can then be separated after derivatization with dry pyridine and acetic anhydride, which will lead to a single peak for some sugars.32 However, the alditol acetate derivative method is more suitable for reducing end sugars (such as glucose, fructose, and lactose).53 Moreover, glucose and fructose generate other alditols, such as mannitol and glucitol hexa-acetate.53 Therefore, the alditol derivative method was not suitable for the measurement of sugar probes in the study reported here because of the generation of more sugar derivatives from glucose, fructose and lactose, which are typically found in serum.
Finally, many steps are required for a complete and effective reduction of sugars to their alditol acetate derivatives.54 Thus, the number of peaks and their retention times are characteristic features of each carbohydrate and are also dependent on the derivatization conditions. Peaks can vary in intensity. Small peaks are part of the sugar fragmentation pattern. They are a constant fraction of the major peak as a result of production of multiple silyl derivatives.31 Therefore, they can be used for the quantitation of sugars.55,56 As described previously, the quantification of sugar probes was performed with the sum of the peak areas from specific retention times of each sugar by use of the SIM mode for specific ions of each sugar.
Saccharides used in the study reported here had ions at m/z 73, 117, 147, 191, 204, 205, 217, and 219. This is in agreement with results reported46,57 for TMS derivatives of rhamnose, xylose, lactulose, and sucrose in specimens other than serum or urine (eg, environmental samples and experimentally created solutions).
The O/E ratios for all 5 sugars spiked into serum at the 4 concentrations ranged from 92.1% to 124.7% (mean ± SD, 106.2 ± 13.0%). Analysis of results of these spiking recovery experiments revealed an increase in the variability for rhamnose and xylose at the lowest concentrations. These variations were evident as an effect of a wide range of the standard curves used for the method validation. However, when a sugar test is performed, xylose and rhamnose are found in substantially higher concentrations in healthy and disease states, with concentrations within the linear midrange of the standard curve. Thus, the effect of high variation in the lower range is not important for the assay performance for analysis of clinical samples. Suboptimal recovery rate was evident at the lowest spiking concentration of 3 mg/L. This was especially true for the recovery of sucrose. Low accuracy for detection and quantification of sucrose at low concentrations (compared with results for higher sucrose concentrations) has been reported58 with the use of gas chromatography. Recovery can be influenced by loss of the analyte during extraction and oximation derivatization. Sucrose may be degraded as a result of the extraction method and oxime derivatization.59 The CV for intra-assay variability for the 5 sugars ranged from 6.8% to 12.9% in the 4 samples evaluated. Finally, the CV for interassay precision was 6.4% to 15.9% for all sugars in the 4 samples.
In the study reported here, a GC-MS method was successfully developed for the simultaneous quantification of lactulose, rhamnose, methylglucose, xylose, and sucrose in canine serum. Gas chromatographic conditions and sugar derivatization by converting the relative oximes before the silylating reaction resulted in acceptable linearity, precision, and reproducibility for use in additional studies to assess permeability and mucosal function of the gastrointestinal tract. However, the accuracy of the assay was limited for low sugar concentrations, especially for sucrose.
Abbreviations
| CV | Coefficient of variation |
| GC-MS | Gas chromatography–mass spectrometry |
| HPLC | High-performance liquid chromatography |
| m/z | Mass-to-charge ratio |
| O/E | Observed-to-expected |
| PAD | Pulsed amperometric detection |
| SIM | Selective ion monitoring |
| TMS | Trimethylsilyl |
D-Lactulose (C12H22O11), Sigma-Aldrich, St Louis, Mo.
L-rhamnose monohydrated (C6H12O5•H2O), Sigma-Aldrich, St Louis, Mo.
3-O-methyl-D-glucopyranose (C7H14O6), Sigma-Aldrich, St Louis, Mo.
D-(+)-xylose (C5H10O5), Sigma-Aldrich, St Louis, Mo.
D-Sucrose (C12H22O11), Sigma-Aldrich, St Louis, Mo.
D-mannitol (C6H14O6), Sigma-Aldrich, St Louis, Mo.
EDM Chemicals Inc, Gibbstown, NJ.
Centrifuge 5417C, Eppendorf, Brinkmann Instruments Inc, Westbury, NY.
Glass vials molded screw cap, VWR International, West Chester, Pa.
Reacti-Therm III heating module, Pierce, Rockford, Ill.
MOX reagent (2% methoxyamine•HCL in pyridine), Pierce, Rockford, Ill.
Pierce, Rockford, Ill.
EDM Chemicals Inc, Gibbstown, NJ.
Gas chromatograph (6890N GC), Agilent Technologies, Palo Alto, Calif.
Mass spectrometer (5975 MSD), Agilent Technologies, Palo Alto, Calif.
DB-1MS capillary column (122-0132), Agilent Technologies, Palo Alto, Calif.
References
- 1.↑
Tibble JA, Bjarnason I. Non-invasive investigation of inflammatory bowel disease. World J Gastroenterol 2001;7:460–465.
- 2.↑
Hessels J, Harry HM, Steggink J, et al. Assessment of hipolactasia and site specific intestinal permeability by differential sugar absorption of raffinose, lactose, sucrose, and mannitol. Clin Chem Lab Med 2003;41:1056–1063.
- 3.
Johnson V, Gaynor A, Chan DL, et al. Multiple organ dysfunction syndrome in humans and dogs. J Vet Emerg Crit Care 2004;14:158–166.
- 4.
Streeter EM, Zsombor-Murray E, Moore KE, et al. Intestinal permeability and absorption in dogs with traumatic injury. J Vet Intern Med 2002;16:669–673.
- 5.
Feltis BA, Kim AS, Kinneberg MK, et al. Clostridium difficile toxins may augment bacterial penetration of intestinal epithelium. Arch Surg 1999;134:1235–1242.
- 6.
Batt RM, Hall EJ, McLean L, et al. Small intestinal bacterial overgrowth and enhanced intestinal permeability in healthy Beagles. Am J Vet Res 1992;53:1935–1940.
- 7.↑
Zuckerman MJ, Menzies I, Ho H, et al. Assessment of intestinal permeability and absorption in cirrhotic patients with ascites using combined sugar probes. Dig Dis Sci 2004;49:621–626.
- 8.↑
Bjarnason I, MacPherson A, Hollander D. Intestinal permeability: an overview. Gastroenterology 1995;108:1566–1581.
- 9.↑
Rose RC, Bode AM. Biology of free radical scavengers: an evaluation of ascorbate. FASEB J 1993;7:1135–1142.
- 10.↑
Fihn B-M, Jodal M. Permeability of the proximal and distal rat colon crypt and surface epithelium to hydrophilic molecules. Pflugers Arch 2001;441:656–662.
- 11.↑
Sun Z, Wang X, Andersson R. Role of intestinal permeability in monitoring mucosal barrier function. Dig Surg 1998;15:386–397.
- 12.↑
Arrieta MC, Bistritz L, Meddings JB. Alterations in intestinal permeability. Gut 2006;55:1512–1520.
- 13.
Fink MP. Interpreting dual-sugar absorption studies in critically ill patients: what are the implications of apparent increases in intestinal permeability to hydrophilic solutes? Intensive Care Med 1997;23:489–492.
- 14.↑
Meddings JB, Gibbons I. Discrimination of site-specific alterations in gastrointestinal permeability in the rat. Gastroenterology 1998;114:83–92.
- 15.↑
Johnston K, Lamport A, Batt RM. An unexpected bacterial flora in the proximal small intestine of normal cats. Vet Rec 1993;132:362–363.
- 16.
Travis S, Menzies I. Intestinal permeability: functional assessment and significance. Clin Sci 1992;82:471–488.
- 17.
Rutgers HC, Batt RM, Hall EJ, et al. Intestinal permeability testing in dogs with diet-responsive intestinal disease. J Small Anim Pract 1995;36:295–301.
- 18.
Heyman M, Dumontier AM, Desjeux JF. Xylose transport pathways in rabbit ileum. Am J Physiol 1980;238:G326–G331.
- 19.
Malagon I, Onkenhout W, Klok G, et al. Gut permeability in pediatric cardiac surgery. Br J Anaesth 2005;94:181–185.
- 20.↑
Dinmore AJ, Edwards JSA, Menzies IS, et al. Intestinal carbohydrate absorption and permeability at high altitude (5,730 m). J Appl Physiol 1994;76:1903–1907.
- 21.↑
Menzies IS, Zuckerman MJ, Nukajam WS, et al. Geography of intestinal permeability and absorption. Gut 1999;44:483–489.
- 22.
Steiner JM, Williams DA, Moeller EM. Kinetics of urinary recovery of five sugars after orogastric administration in healthy dogs. Am J Vet Res 2002;63:845–848.
- 23.
Steiner JM, Williams DA, Moeller EM. Development and validation of a method for simultaneous separation and quantification of 5 different sugars in canine urine. Can J Vet Res 2000;64:164–170.
- 24.↑
Cox MA, Iqbal TH, Cooper BT, et al. An analytical method for the quantitation of mannitol and disaccharides in serum: a potentially useful technique in measuring small intestinal permeability in vivo. Clin Chim Acta 1997;263:197–205.
- 25.↑
Vinet B, Panzini B, Boucher M, et al. Automated enzymatic assay for the determination of sucrose in serum and urine and its use as a marker of gastric damage. Clin Chem 1998;44:2369–2371.
- 26.↑
Hewetson M, Cohen ND, Love S, et al. Sucrose concentration in blood: a new method for assessment of gastric permeability in horses with gastric ulceration. J Vet Intern Med 2006;20:388–394.
- 27.↑
Sharpstone D, Neild P, Crane R, et al. Small intestinal transit, absorption, and permeability in patients with AIDS with and without diarrhoea. Gut 1999;45:70–76.
- 28.↑
Sørensen SH, Proud FJ, Rutgers HC, et al. A blood test for intestinal permeability and function: a new tool for the diagnosis of chronic intestinal disease in dogs. Clin Chim Acta 1997;264:103–115.
- 29.↑
Montero M, Rodriguez D, Sanchez G, et al. Analysis of low molecular weight carbohydrates in food and beverages: a review. Chromatographia 2004;59:15–30.
- 30.↑
Sanz ML, Diez-Barrio MT, Sanz J, et al. GC behavior of disaccharide trimethylsilyl oximes. J Chromatogr Sci 2003;41:205–208.
- 31.↑
Farhadi A, Keshavarzian A, Fields JZ, et al. Resolution of common dietary sugars from probe sugars for test of intestinal permeability using capillary column gas chromatography. J Chromatogr B Analyt Technol Biomed Life Sci 2006;836:63–68.
- 32.↑
Abazia C, Ferrara R, Corsaro MM, et al. Simultaneous gas-chromatographic measurement of rhamnose, lactulose and sucrose and their application in the testing of gastrointestinal permeability. Clin Chim Acta 2003;338:25–32.
- 33.↑
Jansen G, Muskiet A, Schierbeek B, et al. Capillary gas chromatographic profiling of urinary, plasma and erythrocyte sugars and polyols as their trimethylsilyl derivatives, preceded by a simple and rapid prepurification method. Clin Chim Acta 1986;157:277–294.
- 34.↑
Perl IM, Horváth K. Simultaneous quantitation of mono-, diand trisaccharides as their TMS ether oxime derivatives by GCMS: I. In model solutions. Chromatographia 1997;45:321–327.
- 35.↑
Rojas-Escudero E, Alarcón-Jiménez AL, Elizalde-Galván P, et al. Optimization of carbohydrate silylation for gas chromatography. J Chromatogr A 2004;1027:117–120.
- 36.
Teng C, Tjao S, Fennessey P, et al. Transplacental carbohydrates and sugar alcohol concentrations and their uptakes in ovine pregnancy. Exp Biol Med 2002;227:189–195.
- 37.↑
Xue R, Si S, Deng C, et al. Simultaneous determination of blood glucose and isoleucine levels in rats after chronic alcohol exposure by microwave-assisted derivatization and isotope dilution gas chromatography/mass spectrometry. Rapid Commun Mass Spectrom 2008;22:245–252.
- 38.↑
Silva FO, Ferraz V. Microwave-assisted preparation of sugars and organic acids for simultaneous determination in citric fruits by gas chromatography. Food Chem 2004;88:609–612.
- 39.↑
Lavagnini I, Favaro G, Magno F. Non-linear and non-constant variance calibration curves in analysis of volatile organic compounds for testing of water by the purge-and-trap method coupled with gas chromatography/mass spectrometry. Rapid Commun Mass Spectrom 2004;18:1383–1391.
- 40.
Fleming SC, Duncan A, Russell RI, et al. Measurement of sugar probes in serum: an alternative to urine measurement in intestinal permeability testing. Clin Chem 1996;42:445–448.
- 41.
Marsilio R, D'Antiga L, Zancan L, et al. Simultaneous HPLC determination with light-scattering detection of lactulose and mannitol in studies of intestinal permeability in pediatrics. Clin Chem 1998;44:1685–1691.
- 42.↑
Paroni R, Fermob I, Moltni L, et al. Lactulose and mannitol intestinal permeability detected by capillary electrophoresis. J Chromatogr B Analyt Technol Biomed Life Sci 2006;834:183–187.
- 43.↑
Shojaee-Moradie F, Jackson NC, Jones RH, et al. Quantitative measurement of 3-O-methyl-D-glucose by gas chromatographymass spectrometry as a measure of glucose transport in vivo. J Mass Spectrom 1996;31:961–966.
- 44.↑
Strobel S, Brydon WG, Ferguson A. Cellobiose/mannitol sugar permeability test complements biopsy histopathology in clinical investigation of the jejunum. Gut 1984;25:1241–1246.
- 45.↑
Miki K, Butler R, Moore D, et al. Rapid and simultaneous quantification of rhamnose, mannitol, and lactulose in urine by HPLC for estimating intestinal permeability in pediatric practice. Clin Chem 1996;42:71–75.
- 46.
Medeiros PM, Simoneit BRT. Analysis of sugars in environmental samples by gas chromatography-mass spectra. J Chromatogr A 2007;1141:271–278.
- 47.
Halket JM, Zaikin V. Derivatization in mass spectrometry—3. Alkylation (arylation). Eur J Mass Spectrom (Chichester, Eng) 2004;10:1–19.
- 48.
Halket JM, Waterman D, Przyborowska AM, et al. Chemical derivatization and mass spectral libraries in metabolic profiling by GC/MS and LC/MS/MS. J Exp Bot 2005;56:219–243.
- 49.
Zaikin VG, Halket JM. Derivatization in mass spectrometry—6. Formation of mixed derivatives of polyfunctional compounds. Eur J Mass Spectrom (Chichester, Eng) 2005;11:611–636.
- 50.
Halket JM, Zaikin VG. Derivatization in mass spectrometry—1. Silylation. Eur J Mass Spectrom (Chichester, Eng) 2003;9:1–21.
- 51.↑
Dettmer K, Aronov PA, Hammock BD. Mass spectrometry-based metabolomics. Mass Spectrom Rev 2007;26:51–78.
- 52.↑
Black GE, Fox A. Recent progress in the analysis of sugar monomers from complex matrices using chromatography in conjugation with mass spectrometry or stand-alone tandem mass spectrometry. J Chromatogr A 1996;720:51–60.
- 53.↑
Brunton NP, Gormely TR, Murry B. Use of the alditol acetate derivatization for the analysis of reducing sugars in potato tubers. Food Chem 2007;104:398–402.
- 54.↑
Lee YC. Carbohydrates analysis with high-performance anionexchange chromatography. J Chromatogr A 1996;720:137–149.
- 55.
Elwood PC, Reid WK, Marcell PD, et al. Determination of the carbohydrates composition of mammalian glycoproteins by capillary gas chromatography/mass spectrum. Anal Biochem 1988;175:202–211.
- 56.
Kanani HH, Klapa MI. Data correction strategy for metabolomics analysis using gas chromatography-mass spectrometry. Metab Eng 2007;9:39–51.
- 57.
Simoneit BR, Elias VO, Kobayashi M, et al. Sugars—dominant water-soluble organic compounds in soils and characterization as tracers in atmospheric particulate matter. Environ Sci Technol 2004;38:5939–5949.
- 58.↑
Kuhnle GG, Joosen AC, Wood TR, et al. Detection and quantification of sucrose as dietary biomarker using gas chromatography and liquid chromatography with mass spectrometry. Rapid Commun Mass Spectrom 2008;22:279–282.
- 59.↑
Gullberg J, Jonsson P, Nordström A, et al. Design of experiments: an efficient strategy to identify factors influencing extraction and derivatization of Arabidopsis thaliana samples in metabolomic studies with gas chromatography/mass spectrometry. Anal Biochem 2004;331:283–295.
