| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
The Burnham Institute, Glycobiology/Carbohydrate Chemistry Program, 10901 North Torrey Pines Rd., La Jolla, CA 92037
a Author for correspondence: Fax 619-646-3193; email hudson{at}ljcrf.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and Methodsassay protocolResultsDiscussionReferences |
|---|
Key Words: indexing terms: glucose oxidase • glucokinase • carbohydrate-deficient glycoprotein syndrome • phosphomannose isomerase • phosphoglucose isomerase
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and Methodsassay protocolResultsDiscussionReferences |
|---|
Soyama (3) and Akazawa et al. (4) presented enzymatic methods for analysis of D-mannose based on first lowering D-glucose concentrations by oxidation with GOD. The results of these studies indicated that mannose could be accurately measured in sera if the amount of GOD added were proportional to the initial glucose concentration and the glucose concentration were carefully monitored during the reaction until the glucose was reduced to 2–3 times that of mannose. Akazawa et al. (4) noted that GOD preferentially oxidized glucose but that loss of mannose increased when the glucose concentration was less than two or three times that of the mannose present. Soyama (3) reported the concentration of D-mannose in normal sera to be 21 ± 16 µmol/L (range 0–58 µmol/L; n = 60). Akazawa et al. (4) found values of 54.4 ± 2 µmol/L (n = 22) for fasting pregnant women; significantly higher values were associated with diabetes mellitus.
More recently, Pitkanen and Kanninen (5) carefully measured the D-mannose concentrations in human serum with more sophisticated techniques involving selective ion monitoring by gas chromatography–mass spectrometry (GC-MS) after sodium borodeuteride reductive labeling of serum monosaccharides. Upon analysis of six normal human sera, they found mannose concentrations of 55.1 ± 10.6 µmol/L, and demonstrated quantitative recovery of mannose added to the sera.
In recent studies (6), we found that addition of mannose to the culture medium of fibroblasts from patients with carbohydrate-deficient glycoprotein syndrome (CDGS) type I corrected the metabolic deficiency that leads to underglycosylation of glycoproteins in vitro. Schaftingen and Jaeken (7) have recently shown that the metabolic defect in CDGS type I is associated with reduced concentrations of phophomannomutase activity. Consequently, we sought to assay D-mannose concentrations in sera from patients with type I CDGS. We have found the previous methods for the enzymatic assay of D-mannose to be unsatisfactory because serum contains an inhibitor of the GOD/catalase (CAT) reactions and GOD uses mannose as a substrate at low glucose concentrations. To circumvent this problem, we developed a new method based on the use of the highly specific glucokinase (GK) from Bacillus stearothermophilus (8) to selectively remove glucose from sera before the enzymatic assay of D-mannose.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and Methodsassay protocolResultsDiscussionReferences |
|---|
GK was dissolved at 50 kU/L in 0.025 mol/L Tris-HCl, pH 7.8, and stored in an ice bath (GK reagent). G6PDH was dissolved at 100 kU/L in 0.025 mol/L Tris-HCl, pH 7.8, and stored on ice (G6PDH reagent). HK, G6PDH, and PGI (50 U each) were combined and microcentrifuged at 12 000g for 10 min at 4 °C. The ammonium sulfate supernatant was carefully removed, and the pellet was dissolved in 1 mL of 0.025 mol/L Tris-HCl, pH 7.8, and stored on ice (HK/G6PDH/PGI reagent). PMI (100 U) was centrifuged at 12 000g for 10 min at 4 °C, the ammonium sulfate supernatant was carefully removed, and the pellet was dissolved as described above (PMI reagent). Stock solutions of 100 mmol/L ATP and 100 mmol/L NADP were prepared, adjusted to pH 6.5 with 2.5 mol/L KHCO3, and stored frozen at -20 °C. NAM-A reagent cocktail was prepared by mixing 0.8 mL of 100 mmol/L NADP, 0.8 mL of 100 mmol/L ATP, 0.08 mL of 1 mol/L MgCl2, and 0.32 mL of H2O. NAM-B reagent cocktail was prepared by mixing 0.067 mL of 100 mmol/L NADP, 0.067 mL of 100 mmol/L ATP, 0.067 mL of 1 mol/L MgCl2, and 1.8 mL of H2O.
Preparation of DE-53 spin columns.
DE-53 anion-exchange
cellulose (25 g) was suspended in 300–400 mL of H2O,
allowed to settle, and the fines decanted two or three times. The resin
was resuspended in 0.25 mol/L Tris-HCl, pH 8.5, collected on a sintered
glass funnel, and washed with 200 mL of 0.25 mol/L Tris-HCl, pH 8.5,
followed by three or four washes with H2O. The resin was
gently resuspended in a minimal amount of H2O, transferred
to a 150-mL reagent bottle, and allowed to settle overnight. The total
volume was adjusted to 1.6–1.8 times the volume of settled resin, and
0.5 mmol/L sodium azide was added as a preservative.
The de-fined, equilibrated DE-53 resin was gently resuspended and 0.6 mL pipetted into the filter basket of a MicroSpin centrifuge filter. The column was centrifuged at 3000 rpm for 2 min in a microcentrifuge with swinging tube holders. The resin was washed twice with H2O and the effluents discarded.
|
assay protocol |
|---|
|
TopAbstractIntroductionMaterials and Methodsassay protocolResultsDiscussionReferences |
|---|
DE-53 spin column to remove anionic products and
substrates.
A DE-53 spin column was washed with H2O
and microcentrifuged at 3000 rpm for 2 min as described above. A
200-µL portion of the glucose-depleted preincubation mixture was
pipetted onto the resin bed and centrifuged at 5000 rpm for 2 min. The
remaining 200 µL was pipetted onto the resin bed and the
centrifugation repeated. The combined effluents were transferred to a
clean tube. The resin bed was washed twice with 200 µL of
H2O and transferred to the reaction tube (total volume
= 800 µL). To the tube was added 200 µL of 0.5 mol/L Tris-HCl, pH
7.8; 180 µL of NAM-B reagent cocktail; and 40 µL of HK/G6PDH/PGI
enzyme reagent cocktail. After mixing, 610 µL was transferred to a
new reaction tube (tube B). To the original reaction tube (tube A), 10
µL of H2O was added and to tube B, 10 µL of PMI enzyme
reagent (1 U). After incubating at 37 °C for 15 min, the absorbance
at 340 nm was measured to quantify NADPH production. Care should be
taken to avoid contamination of samples lacking PMI by samples
containing PMI during the pipetting and spectrophotometric readings.
Special micropipette tips designed for 1–20-µL volumes should be
used as appropriate. Final concentrations of added reagents were 0.08
mol/L Tris-HCl, pH 7.8; 500 µmol/L ATP; 500 µmol/L NADP; 5 mmol/L
MgCl2; 1.6 kU/L HK, G6PDH, and PGI; and 3.2 kU/L PMI. The
absorbance at 340 nm (A340) of each tube is
measured with H2O as a blank. The difference in
A340 (
A340) between
tube A and tube B is due to mannose in the original 200-µL sample
(100 µL in tube A and 100 µL in tube B). Since the sample was
diluted 6.2-fold during the assay and the µmol/L absorptivity of
NADPH is 6.2 x 10-3, the concentration of mannose
(µmol/L) in the original sample is theoretically 1000 x
A340. However, we have found that when the
sample contains only 200 µL of H2O, there is a
A340 of 0.002–0.012, due possibly to
contaminants in the PMI and other reaction components. Therefore, it is
necessary to measure a reagent blank with each set of determinations
and subtract the A340 of the reagent blank
from the A340 of the samples.
High-pH anion-exchange chromatography with pulsed
amperometric detection (HPAEC-PAD) of D-mannose in serum
samples.
HPAEC-PAD analyses were performed by the Glycobiology
Core Facility at the Univ. of California, San Diego, Cancer Center.
Serum samples were preincubated with GK/G6PDH and the anionic
substrates and products were removed with DE-53 spin columns exactly as
described in the above protocol. Portions (40 µL of 800 µL total)
of the spin column effluent were analyzed by HPAEC-PAD on a CarboPac
PA-10 column. The column was eluted with 18 mmol/L NaOH for 22 min for
the analytical separation, regenerated for 10 min with 200 mmol/L NaOH
for 10 min, and reequilibrated with 18 mmol/L NaOH for 16 min before
injection of the next sample. Controls included serum incubated without
GK, a reagent blank (200 µL water in lieu of serum), and a
mannose-only sample (200 µL of 50 µmol/L mannose). To verify peak
identity, a portion of the spin column effluent was supplemented with
mannose before analysis (equivalent to an additional 200 µmol/L
mannose in the original sera). Mannose peak areas were integrated and
nanomoles of mannose were calculated on the basis of a calibration
curve generated with mannose calibrators not subjected to the enzymatic
and chromatographic steps above.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and Methodsassay protocolResultsDiscussionReferences |
|---|
(curves with negative slope). Depletion of glucose from the
sample is >99% complete after a 10-min incubation with 1.25 kU/L
GK/G6PDH. G6PDH is included in the reaction to oxidize the product of
the GK reaction (glucose-6 phosphate), thus "pulling" the reaction
to completion. The depletion of glucose is much slower, even with 25
kU/L GOD and 100 kU/L CAT; ~3% remains after a 60-min incubation.
Depletion of glucose from a serum sample (measured by production of
glucose-6 phosphate) is also complete within 10 to 20 min with the same
concentration of GK/G6PDH (Fig. 1
; curve with positive slope). By
contrast, we were unable to completely remove glucose from serum
samples with 100–200 kU/L GOD and 400 kU/L CAT; 10–30% of the
glucose remained, even after a 90-min incubation. However, the rate of
glucose removal by 25 kU/L GOD from dialyzed serum reconstituted with 5
mmol/L glucose was the same as that of the glucose calibration
solution. We concluded that sera contain a dialyzable inhibitor of the
GOD/CAT reactions. This may explain why previously published methods
involving GOD required unusually high concentrations of enzyme (e.g.,
1–5 µg of purified enzyme per microgram of glucose)
(3)(4).
|
However, using GK plus G6PDH to selectively remove the glucose from
sera created a new problem. The reaction requires high concentrations
of ATP and NADP substrates and generates high concentrations of ADP and
NADPH products, which would interfere with the mannose determination
using previously published methods (3)(4)
involving PMI, PGI, and G6PDH. To overcome this, we designed a rapid,
quantitative method to selectively remove these anionic substrates and
products from the unreacted neutral mannose. We chose anion-exchange
chromatography in microcentrifuge spin columns to allow rapid
processing of multiple samples and quantitative recovery of sample
volumes. Whatman DE-53, a weak anion-exchange resin, was chosen because
it has excellent physical properties compatible with the
microcentrifuge spin column method and a high exchange capacity (
2
meq/g dry weight). The use of strong anion-exchange resins leads to
base-catalyzed isomerization of the mannose to fructose. The method was
verified by showing that >98% of the NADPH (99% ± 0.5%; n =
26; range 96.9–99.4%) produced during the preincubation was removed
from the spin column effluent and that >95% of a sample of
radioactive mannose (98.4% ± 4.1%; n = 11; range 94.3–105%)
was recovered in the effluent.
Assay of mannose calibrators in the presence of serum or 5
mmol/L glucose.
To validate the procedure described in
Materials and Methods, we assayed a series of mannose
calibration solutions (5–200 µmol/L) and the same concentrations in
the presence of 5 mmol/L glucose or human serum. The results in Fig. 2
show that the determination of mannose concentration was
unaffected by the presence of 5 mmol/L glucose; the calibration curves
are superimposed. The linear regression slope was 0.94 in the absence
of glucose and 0.92 in its presence. Incubation of increasing
concentrations of mannose in the presence of pooled human sera (53
µmol/L mannose) demonstrates the quantitative recovery of mannose
from serum samples. The linear regression slope of the mannose
calibration curve in the presence of serum was 0.90. (The manufacturer
does not state the water content of the mannose used for these
calibrators; a 92% recovery is equally compatible with a loss of 8%
or the presence of 1 mol of water per mole of mannose.)
|
From these analyses (in the presence and absence of glucose) the interassay CV was 12.2% at 40 µmol/L (n = 7; 36.9 ± 2.1 µmol/L) and 9.8% at 80 µmol/L (n = 7; 74.2 ± 2.7 µmol/L). The intraassay CV was determined by assaying samples containing 40 µmol/L mannose or containing 80 µmol/L mannose; the intraassay CV at these concentrations was 6.7% (n = 5; 39.6 ± 1.6 µmol/L) and 4.4% (n = 11; 75.0 ± 1.8 µmol/L), respectively. The average recovery of mannose added to the sample of pooled human serum over a range of 5–200 µmol/L was 94% ± 4.4%.
Analysis of serum mannose concentration by GK/G6PDH reaction
and HPAEC-PAD.
To confirm the values obtained by the enzymatic
assays above, a set of samples was treated exactly as described by the
assay protocol in Materials and Methods, except that the
effluents from the DE-53 spin columns were analyzed by HPAEC-PAD
instead of with the HK/PGI/G6PDH/±PMI enzymatic assay. The results in
Fig. 3
A show that the reagent blank has no peak in the area where
mannose is expected to elute. Panel B shows a trace for the analysis of
a calibrator containing 50 µmol/L mannose. Panel C shows the trace
for a serum sample containing 49.7 ± 6.8 µmol/L mannose
(determined by enzymatic assay). Panel D shows the same serum sample
supplemented with 200 µmol/L mannose.
|
The results of the quantitative analysis of four serum samples by
HPAEC-PAD compared with the analysis by the enzymatic procedure are
shown in Table 1
. The values determined by both methods are in excellent
agreement. Only single determinations by HPAEC-PAD were performed
because of expense, availability of instrument time, and concern that
the high concentration of protein and other serum components not
removed by the DE-53 spin columns might damage the column. Attempts to
analyze serum mannose by HPAEC-PAD without pretreatment with GK to
remove glucose were not successful; the mannose peak area was obscured
by the 100-fold greater glucose peak. Although the HPAEC-PAD method
described here is useful as an alternative method for mannose
determination, only one sample per hour (50 min) can be analyzed after
the preincubation and chromatography step. By contrast, after the
preincubation and chromatography step, the enzymatic assay of 12
samples can be completed in <1 h, and the data reduction and
tabulation is also simpler and faster.
|
Also shown in Table 1
is the average value determined for the mannose
content of 11 sera from healthy human volunteers. The value of
54.1 ± 11.9 µmol/L agrees quite well with the value reported by
Pitkanen and Kanninen (5) of 55.1 ± 10.6 µmol/L
determined by GC-MS.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and Methodsassay protocolResultsDiscussionReferences |
|---|
The major obstacle of these analyses is the presence of 50–200-fold
higher concentrations of D-glucose, which obscures or
impedes the mannose determination by chromatographic methods of
analysis. This problem was overcome by Pitkanen and Kanninen by using
selected ion monitoring during GC-MS after sodium borodeuteride
reduction of serum hexoses (5). While sophisticated
and accurate, this method would be difficult for rapid routine
analyses. Other assays (3)(4) have relied on
the use of GOD to reduce the high glucose concentrations before
enzymatic analysis of mannose. These methods frequently result in very
high backgrounds due to residual glucose remaining after the GOD
treatment. We have observed that sera contain dialyzable inhibitors of
the GOD system, which results in variable success in depleting sera of
glucose. Furthermore, the high Km of GOD for
glucose (33 mmol/L) (13) leads to the need for unusually
high concentrations of enzyme or lengthy incubations to reduce the
serum glucose to a suitable concentration (100–200 µmol/L). The
method described in the present study circumvents these problems by
using the highly specific GK from B. stearothermophilus to
remove glucose from the sera rapidly and quantitatively. The
Km of this GK for glucose and ATP are 0.1 mmol/L
and 0.05 mmol/L, respectively (8). This low
Km for glucose facilitates the rapid and
quantitative removal of glucose from sera. This GK shows no activity
with mannose or fructose as substrates (8).
We used G6PDH in conjunction with the GK to "pull" the GK reaction
to completion during a short incubation time. In these studies >99%
of the glucose was removed from serum sample assays (>99.2% ±
0.15%; n = 29; range 99.0–99.5%). Because the GK does not
phosphorylate mannose or fructose (8), the effluent from
the DE-53 spin column may contain both mannose and fructose. Normal
sera contain 18.3 ± 7.6 µmol/L fructose (5). The
second enzymatic step to determine the mannose concentration in the
sample depleted of glucose is done in the absence (tube A) or presence
(tube B) of PMI, and the mannose concentration is calculated from the
difference of absorption at 340 nm (A340). HK
phosphorylates any residual glucose present in the sample as well as
the fructose and mannose present. PGI is included in both assay tubes
to convert fructose-6 phosphate to glucose-6 phosphate and, thus, to
blank out any contribution of fructose to the
A340. Although we have not attempted to do so
here, the fructose concentration could also be measured with this
procedure by carrying out the final incubation in the presence and
absence of PGI. For the mannose determination, PGI is required in both
tubes since PMI converts mannose-6 phosphate to fructose-6 phosphate.
A critical step in the protocol presented is the use of microcentrifuge spin-column chromatography to remove the anionic components produced during the GK incubation step. The spin-column technique allows the quantitative and reproducible control of sample volume and recovery required for the assay without further processing for the final enzymatic determination of D-mannose. Several samples (12 or more) can be processed concurrently. Whatman DE-53 resin (DEAE-cellulose) was chosen for these spin columns because of its physical properties and high exchange capacity. Use of other resins with a lower exchange capacity may result in the incomplete removal of NADPH produced during the preincubation and thus a high background. However, careful handling of the resin is required to prevent breakdown of the microgranular fibers, which may affect the chromatographic properties of the resin (e.g., avoid vigorous mixing or shaking during resuspension of resin).
With this protocol, the reagent blank should give an
A340 of 0.025 to 0.050 vs an
H2O blank and a A340 of 0.002 to
0.012. With serum samples the background A340
(-PMI; tube A) is variable because of variation in UV-absorbing
macromolecules and other components and is normally in the range of
0.150 to 0.350 vs a water blank. This variability in background does
not directly affect the A340 but the higher
background values do result in greater measurement error for low
mannose concentrations. Hemolyzed sera give higher background values,
which leads to greater measurement error. Lipemic sera tend to clog the
DE-53 spin columns and necessitate longer centrifugation times or
separate steps to remove lipid before analysis of the mannose
concentration (e.g., centrifugation of serum samples at high speed to
remove the lipid).
This assay can be used to measure mannose in the sera of CDGS type 1 patients who are defective in mannose utilization. In studies to be reported elsewhere (Panneerselvam et al., manuscript in preparation), we use this assay to show that CDGS type I children have significantly reduced concentrations of serum mannose compared with healthy volunteers and pediatric patient controls. Since addition of mannose to the culture medium of fibroblasts from CDGS type I children corrects the protein glycosylation deficiency that is a hallmark of this disorder, this assay will be especially important for evaluating the success of proposed dietary mannose therapy for CDGS type 1 children. The assay should also be easily adaptable for determining mannose concentrations in the sera of other animals or in tissue culture media.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and Methodsassay protocolResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  A. Angelov, M. Putyrski, and W. Liebl Molecular and Biochemical Characterization of {alpha}-Glucosidase and {alpha}-Mannosidase and Their Clustered Genes from the Thermoacidophilic Archaeon Picrophilus torridus. J. Bacteriol., October 1, 2006; 188(20): 7123 - 7131. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Taguchi, E. Yamashita, T. Mizutani, H. Nakajima, M. Yabuuchi, N. Asano, and I. Miwa Hepatic glycogen breakdown is implicated in the maintenance of plasma mannose concentration Am J Physiol Endocrinol Metab, March 1, 2005; 288(3): E534 - E540. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Taguchi, I. Miwa, T. Mizutani, H. Nakajima, Y. Fukumura, I. Kobayashi, M. Yabuuchi, and I. Miwa Determination of D-Mannose in Plasma by HPLC Clin. Chem., January 1, 2003; 49(1): 181 - 183. [Full Text] [PDF]
|
|
|
|
|
|
H. A. Carchon and J. Jaeken Determination of D-Mannose in Serum by Capillary Electrophoresis Clin. Chem., July 1, 2001; 47(7): 1319 - 1321. [Full Text] [PDF]
|
|
|
|
 |
|
 P de Lonlay, N Seta, S Barrot, B Chabrol, V Drouin, B M Gabriel, H Journel, M Kretz, J Laurent, M Le Merrer, et al. A broad spectrum of clinical presentations in congenital disorders of glycosylation I: a series of 26 cases J. Med. Genet., January 1, 2001; 38(1): 14 - 19. [Abstract] [Full Text]
|
|
|
|
 |
|
 K. Panneerselvam, J. R. Etchison, and H. H. Freeze Human Fibroblasts Prefer Mannose over Glucose as a Source of Mannose for N-Glycosylation. EVIDENCE FOR THE FUNCTIONAL IMPORTANCE OF TRANSPORTED MANNOSE J. Biol. Chem., September 12, 1997; 272(37): 23123 - 23129. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
——
—–
