| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Department of Clinical Chemistry, Royal Liverpool University Hospital, PO Box 147, Liverpool L69 3BX, UK.
2
Micromass UK Ltd., Tudor Road, Altrincham, Cheshire WA14
5RZ, UK.
a Author for correspondence. Fax 0161-282-4400; e-mail mike.morris{at}micromass.co.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- and ß-chains was determined directly and was
shown to be constant (0.64:1) over the glycation range measured. Only
single glucose attachment to both the - and ß-chains was
observed.
Key Words: indexing terms: diabetes • method comparison
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
GHb is a general term referring to the products of glucose binding
slowly and nonenzymatically to hemoglobin with a ketoamine structure
(10). Included in these products are those resulting from
the condensation of glucose with the N-terminal valines of one or both
of the ß-chains (originally specified as Hb A1c)
(11) with, to a lesser extent, glycation of the N-terminal
valine(s) of the -chain(s) (12). In addition, several
of the 
-amino groups of the globin lysine residues may also be
glycated (12), and evidence indicates that the site of
modification can vary, depending on whether the hemoglobin glycation
occurs in vitro or in vivo (12)(13).
A more detailed analysis with enzymatic digestion of the glycated
proteins and subsequent separation and analysis has been used to
uniquely identify the sites of glycation on the - and ß-chains
(14). Miedema, Kobold, and Jeppsson demonstrated the
utility of using enzymatic digestion coupled with chromatographic or
electrophoretic separation and mass spectrometric detection to
determine the absolute amount of Hb A1c present in a blood
sample and proposed that method as a primary reference method for
measuring Hb A1c. Participants at that symposium
(14) also highlighted the lack of an available primary
standard for Hb A1c determination.
For several years, electrospray mass spectrometry (ES-MS) has been
recognized as one of the tools available for characterization of
variant hemoglobins (15). In this technique, the
hemoglobin is introduced into the mass spectrometer in a denaturing
solution, so that all of the proteins (including posttranslationally
modified proteins) in the native material are analyzed as covalently
bound entities. In effect, the mass spectrometer separates these
components according to their molecular weights (masses); thus, the
- and ß-chains, their glycoforms, other chains, and
posttranslationally modified chains are detected separately, provided
their masses differ sufficiently from one another. However, using mass
spectrometry this way cannot differentiate between different sites of
glycation because all of the glycated species that exhibit single
glycation on a given chain have the same mass.
In essence, the electrospray ion source of the mass spectrometer
generates a series of multiply protonated molecules for each protein
present in the sample solution. These ions are separated by the mass
spectrometer according to their mass-to-charge ratio (m/z),
producing a spectrum on an m/z scale. Software used to
transform this m/z spectrum represents each protein as a
single peak on a true molecular mass scale. The relative intensities of
the peaks so displayed are proportional to the concentrations of the
proteins in the sample solution, but proteins that differ significantly
in their sequence, and thus their solution and gas-phase chemistries,
do not necessarily produce the same response. Previous ES-MS studies of
hemoglobin indicate that the relative signals observed for - and
ß-chains are ~4:3 (15). Because both glycated - and
ß-chains differ from their nonglycated forms by a similar
posttranslational modification, the responses of the glycated forms and
their nonglycated forms would be expected to be similar. If the
response of the glycated forms differed greatly from that of the
nonglycated forms, the calculation of percentage of glycation from the
mass spectrometric data would be unrealistic and would not correlate to
the other methods of GHb determination.
In the case of hemoglobin, the resolution (the ability of ES-MS to
separately analyze the various components according to their molecular
masses) is ~1300 (12 Da at 16 kDa), and the accuracy and precision of
measuring their molecular masses is ±0.005%, or about ±0.8 Da for
the components in the relevant mass range of 15–16.5 kDa. Given that
the average molecular masses calculated from the sequences of the -
and ß-chains in normal (nonvariant) human hemoglobin are 15126.4 and
15867.2 Da, respectively—162.1 Da higher for their glycated forms
(15288.5 and 16029.4 Da, respectively)—these components are well
separated on the mass scale and may be unambiguously identified from
their measured masses. ES-MS, therefore, should be well suited to
determinations of the percentage of glycation of the separate globin
chains. The high resolution and high mass accuracy provided by ES-MS
may also reveal the presence of components in the hemolysate that might
interfere with other analytical techniques, and identify them, in many
cases, simply from their molecular masses.
Here we describe an investigation into the potential of ES-MS for quantifying GHb. Analyses were carried out on blood from healthy volunteers and diabetic patients and the results compared with those from various established procedures such as ion-exchange chromatography and affinity-based techniques. We also illustrate that GHb can be determined in the presence of aberrant chains (e.g., a heterozygote ß-chain variant) in a suspected diabetic blood sample and, moreover, can provide some information about the nature of the aberrancy.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The collection of samples and analysis was part of the routine assessment of the patients' glycemic control, and complied with the ethical standards of the Royal Liverpool University Hospital. The additional analyses were performed anonymously, in accordance with the ethical guidelines of the Hospital.
The samples were stored at 4 °C between collection and analysis by non-mass-spectrometric methods (within 2–3 days of collection). They were further stored at -20 °C until analysis by mass spectrometry, generally within 1 week.
sample preparation for mass spectrometry
Initially, 10 µL of the whole-blood samples was diluted 50-fold
with 490 µL of deionized water (Elga, High Wycombe, UK). Portions (20
µL) of these solutions were then diluted another 10-fold with 40 µL
of 10 mL/L formic acid, 40 µL of deionized water, and 100 µL of
HPLC-grade acetonitrile (Fisons, Loughborough, UK) to give working
solutions for mass spectrometric analysis.
To assess the effect of salt present in the samples on the analysis, we desalted some samples, using 10-kDa-cutoff microconcentrators (Amicon, Stonehouse, UK). A 100-µL aliquot of the 50-fold-diluted solutions was further diluted with 200 µL of 10 mL/L formic acid and concentrated with a microconcentrator almost to dryness. The concentrate was then washed with 300 µL of deionized water and reconcentrated, after which the hemoglobin residue was taken up in 100 µL of deionized water. From this solution, 20 µL was then diluted with 40 µL of 10 mL/L formic acid, 40 µL of water, and 100 µL of acetonitrile to give the mass spectrometer working solutions.
A working solvent of equal volumes of water and acetonitrile acidified with 2 mL/L formic acid was generally used for preparing samples for the ES-MS analysis of peptides and proteins and serves to denature the protein. The acid also assists protonation of the molecules in solution before the analysis.
mass spectrometry
The working solutions prepared above were introduced as 10-µL
aliquots at 5 µL/min into the electrospray source of a Quattro II
mass spectrometer (Micromass, Altrincham, Cheshire, UK) in a flowing
liquid phase of equal amounts of water and acetonitrile with no
additives. The instrument, operated in the positive ion mode, summed
for each sample 10 x 8-s scans over an m/z range of
980–1300 in 1.5 min to obtain the final spectrum. After baseline
subtraction, these data were transformed to a scale for true molecular
mass by use of software routines supplied with the instrument. The
transformed spectra were then smoothed and the heights of the peaks
used as a measure of their intensities. We used peak heights rather
than peak areas because the heights discriminate more efficiently
against potential interferences from partially resolved peaks.
Initially, the data were manually processed, but the original
m/z spectra from later batches of samples were processed
automatically to carry out the transformation, identify the appropriate
peaks from their measured masses, and calculate the %GHb from the peak
intensities. Mass-scale calibration was performed by using the multiply
protonated molecules from the normal -chain (15126.4 Da) present in
the samples.
calculation of % glycation
A typical electrospray mass spectrum from diluted human blood
contains the following species:
1) -, ß-, glycated -, and glycated ß-chains.
2) Several covalently modified chains, mainly associated with the ß-chain. Their concentrations vary, apparently depending on the age of the sample, e.g., putative cysteinylated and glutathionylated ß-chain and their corresponding glycated species.
3) Minor non--chains, e.g., 
and 
.
4) Non-covalently-bound adducts associated with all the species already listed, e.g., heme adducts and alkali metal adducts, predominantly sodium and potassium.
In the absence of direct measurements of the sensitivities (mass spectrometer response for a given molar concentration in the analyte) of many of these species, we made the following assumptions to derive an expression for calculating the percentage of glycation from the experimental data:
1) All -chain species, including noncovalent adducts, have the same
sensitivity. We believe this is a reasonable assumption because these
species all have the same protein chain with relatively minor
modifications.
2) All ß-chain species, including noncovalent adducts, have the same sensitivity.
3) The sensitivities of the - and non--chain species differ and
are related by:
 |
4) The relative sensitivities of adducted glycated -chain to
glycated -chain species are the same as those of adducted -chain
to -chain species. Similar assumptions are made for the
corresponding ß-chain species.
5) Minor non--chain species may be ignored.
6) There is no contribution to the glycated species by multiple alkali
metal adducts of the nonglycated species (see comparison of results
after desalting samples, Fig. 2
).
|
On the basis of these assumptions and the knowledge that the ES-MS
analyte originates from whole-blood hemoglobin that contains equal
molar amounts of - and non--chain species, it can be rigorously
shown that
 |
and ß represent the intensities of the - and
ß-chains, and g and ßg
represent the intensities of the glycated - and ß-chains. This remarkably simple expression means that only four peaks from the transformed spectrum need to be measured; however, many of the assumptions on which it is based are difficult to prove empirically. Nevertheless, the close correlation between the ES-MS results and the results obtained with other techniques imply that the assumptions are not widely divergent from reality.
The ratio of glycated to glycated ß was calculated as
g(ß+ ßg)/ßg(+
g).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
a shows a typical m/z spectrum (after baseline
subtraction) for a blood sample with ~11% glycation; Fig. 1b
shows
the same data after transformation to a true molecular-mass scale. The
prominent peaks observed in Fig. 1a
are attributed to the series
arising from the multiply protonated - and ß-chains, the degree of
protonation observed under these experimental conditions varying from
12 to 15 for the -chain (e.g., A12 = -chain + 12 protons)
and from 13 to 16 for the ß-chain. The region in the lower
m/z region (m/z 600–980)—not shown and not used
in the analysis—contains multiply protonated molecules extending up to
the 20-protonated -chain and the 18-protonated ß-chain. The
presence of free heme (m/z 616.2) is also
observed.1
The transformed spectrum in Fig. 1b
shows peaks from the -
(15126.4 Da) and ß- (15867.2 Da) chains, in addition to the glycated
- (15288.5 Da) and glycated ß- (16029.4 Da) chains. Also observed
are peaks for + heme (15742.9 Da) and ß + heme (16483.7 Da).
Peaks observed at a slightly higher mass than the - and ß-chains
are attributable to the adduction of sodium and potassium; they were
considered a possible source of interference in measuring glycated
globins if multiple adduction of alkali metals formed species that had
the same mass as the glycated forms.
|
Figure 2
a shows a typical transformed spectrum from a sample with
~11% glycation. In addition to peaks corresponding to the - and
ß-chains and their corresponding glycated forms (+162.1 Da), peaks
are present corresponding to the addition of heme and sodium and
potassium single adducts. These alkali metal adducts, although not
directly a problem, indicate the possibility of multiple adducts
occurring at the same masses as the glycated chains, which would thus
increase the measured quantities of the latter. To assess this
possibility, we remeasured the percentage of glycation after desalting
selected samples. The transformed spectrum of a desalted sample is
shown in Fig. 2b
. The measured proportion of glycation for this sample
was 11.1% both before and after desalting, despite the reduction in
adducting observed after desalting (Fig. 2b
). This result indicates
that using diluted whole blood without further purification to remove
alkali metal salts is adequate for determining GHb by ES-MS. It further
suggests that the assumptions made regarding alkali metal adducts in
calculating %GHb are reasonable. All the GHb results presented in this
report are from samples that were simply diluted from whole blood and
analyzed without desalting.
Data from one batch of samples showed evidence of sample degradation by
exhibiting additional peaks in their mass spectra. The spectra in Fig. 2
are from a sample from this batch. The additional peaks, at masses
119.1 and 176.2 Da higher than the ß-chain peak, may be due to
cysteinylation and the covalent addition of Cys-Gly, respectively.
Peaks for these adducts reached significant heights for whole-blood
samples that had been stored at 4 °C for several weeks. These types
of adducts were not observed when analyzing fresh samples, and thus are
not considered a problem.
The transformed mass spectra for three blood samples are shown in Fig. 3
, demonstrating glycation of 5%, 7%, and 12% of the total
hemoglobin. As can be clearly seen in these spectra, the peaks
attributed to the glycated - and ß-chains proportionately increase
as the degree of glycation increases. In these experiments, only single
glycation of both the - and ß-chains was observed; i.e., no
signals were seen 324.2 Da greater than the molecular masses of the
free chains, which would have corresponded to the addition of two
glucose units. No evidence for multiple glycation was observed even at
the highest value analyzed for GHb (15%, as determined by ES-MS).
|
The ratio of -chain glycation to ß-chain glycation was 0.64
± 0.07 (SD) and was independent of the degree of glycation. Both
-chain glycation (r2 = 0.9669) and ß-chain
glycation (r2 = 0.9838) increased linearly with
the increase in total measured GHb (Fig. 4
).
|
The imprecision (CV) of the ES-MS measurements was determined by performing 10 replicate analyses on five separate samples. During a single day the CV was 2.0% (n = 10); it was <5.1% (n = 10) over a period of several days. For the various proportions of GHb determined, the CVs were: 5.1% at 3.6% GHb, 2.8% at 4.9% GHb, 3.1% at 6.5% GHb, 3.0% at 9.7% GHb, and 2.3% at 14% GHb.
A comparison of GHb determined by ES-MS with GHb determinations by
other established methods is presented graphically in Fig. 5
. All the established methods demonstrated a reasonably linear
correlation with the ES-MS method, although the latter generally
reported a lower %GHb than the separation methods. Whether this
lower response is a function of the ionization method used or the
result of chemical noise artificially increasing the results in the
other techniques is unclear. The HPLC affinity method showed excellent
correlation (r2 = 0.9840; Fig. 5a
) with ES-MS,
whereas the Abbott Vision method showed more scatter around a
well-correlated trend (r2 = 0.9754; Fig. 5b
).
The worst correlation of the ES-MS method, that with the IMx affinity
method (r2 = 0.8640; Fig. 5c
), still showed an
overall similarity in results by the two methods. Comparison between
the HPLC affinity method and the IMx affinity method (Fig. 5d
) showed
the poorest correlation (r2 = 0.8554) of all the
methods compared.
|
We also compared the HPLC affinity determinations with those obtained by ion-exchange chromatography. The resulting correlation (r2 = 0.9278) was comparable with that observed between ES-MS and HPLC affinity measurements (N.B. Roberts, B.N. Green, and M. Morris, unpublished results). This implies that the correlation between ES-MS and ion-exchange chromatography should also be good.
An additional, incidental, result of the ES-MS quantification of GHb is
the detection of hemoglobin variants. Mass differences as little as 12
Da can be detected for heterozygous variants in - and ß-globin
chains (16); using additional processing based on maximum
entropy algorithms can resolve species differing by only 6 Da at 16
kDa. An example of a sample containing a hemoglobin variant is shown in
Fig. 6
. The blood, analyzed because it was thought to have increased
proportions of GHb, was found to contain a heterozygous ß-chain
variant at 15 881 Da, 14 Da more than the normal ß-chain. The
hemoglobin was also determined to be 3.1% glycated. A mass difference
of +14 Da in ß-globin corresponds to a single-point mutation that may
occur at one of 49 of the 146 residues in the ß-chain (M. Morris,
B.N. Green, and N.B. Roberts, manuscript in preparation). Further
investigation showed the variant to be Hb-Raleigh (ß1
Val
Ac-Ala) (17). Homozygous variants may be analyzed
more easily, and one can determine by mass spectrometry a homozygous
variant differing by 1 Da from the normal form (15).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain glycation, ß-chain glycation, or total glycation, but
cannot analyze specifically for Hb A1c. At its present
stage of development, the technique is more applicable to the research
environment than to a routine clinical laboratory and would obviously
need further validation before use in the latter.
The ES-MS results demonstrate that the degree of glycation of the
-chain is 0.64 ± 0.07 that of the ß-chain, in reasonable
agreement with a previously reported ratio of 0.53 (18).
Because the amounts of the two chains in the solution being analyzed
are the same, i.e., 2ß2 (19),
any difference in relative glycation between the two chains is likely
to be structurally determined.
Previous studies showed that glycation at the N-terminus of the
ß-chain (Hb A1c) and at other sites (N-terminus of
the -chain, -chain lysines, and ß-chain lysines) accounts for
~8% (12) of Hb A0 in
nondiabetic subjects. The data from these ES-MS analyses averaged
~2.7% ± 0.5%, somewhat lower than the values previously determined
for normal samples (13)(18). The
higher-resolution PolyCAT A cation-exchange method described by
Turpeinen et al. (5) reports lower values for
(specifically) Hb A1c, primarily because the more efficient
chromatographic separation resolves the peak measured as Hb
A1c from interferences that coelute in other separation
techniques. Turpeinen et al. reported a mean glycation percentage (Hb
A1c) of 3.5% ± 0.5% (95% confidence interval) for
nondiabetic patients, fairly close to the slightly lower total GHb
value by ES-MS of 2.7% ± 0.5%.
To test whether ES-MS analysis of the samples in acid solution could have caused some hydrolysis of the glucose adducts and hence led to lower GHb values in our results, we also analyzed two blood samples having ~14% GHb, as follows. Two solutions were prepared from each sample in the aqueous acetonitrile solvent plus 2 mL/L formic acid, as described above. One solution was kept at room temperature (~20 °C); the other was kept at -20 °C except when brought to room temperature for analysis. The four solutions were each analyzed as a group 14 times over 4 days. The time between analyzing each room temperature solution and its corresponding -20 °C solution was <15 min. The solutions kept at -20 °C gave mean ± SD %GHb values of 13.52% ± 0.31% and 14.48% ± 0.21% and showed no trend in variation of %GHb with time. Calculating the %GHb ratio for each pair of measurements (i.e., the room temperature solution and its corresponding -20 °C solution) showed a barely detectable rate of decrease of %GHb in the room temperature solutions, averaging 0.12% and 0.16% per 24 h for the 13.5% and 14.5% GHb samples, respectively. Thus we conclude that no significant reduction occurred in the %GHb values reported here as a result of the samples being prepared and analyzed in acid solution, given that the time between preparing and analyzing the solutions was <3 h.
Overall, our observations may help explain some of the intermethod variations observed in GHb determinations. In particular, we demonstrated the presence of strongly attached alkali metal ion adducts, which may affect both electrophoretic and ion-exchange separations. None of the other postulated complexing agents (urea, acetylsalicylate) were observed in the analyses reported here, probably because the relative amount of these complexes normally detected (<1.0% of total hemoglobin for an individual component) (7) would be too low to detect by ES-MS under the nonseparatory conditions used. Even if such complexes were observed, the mass differences between the complexes and the glycated analytes of interest would not generally interfere in the assay.
Decomposition of the samples on storage, with the formation of Cys and Cys-Gly addition products of the ß-chain observed in the mass spectra, may also affect separations that are charge-dependent. However, these species do not occur at the same mass as the glycated chains and are not considered to be a possible source of interference in the mass spectrometric assay. The putative Cys and Cys-Gly addition products were not observed in samples that had not undergone prolonged storage.
Despite the existence of multiple glycosylation sites in both - and
ß-chains (13), only single glucose addition was observed
by ES-MS. Careful comparison of the spectra from two samples having
13.5% and 14.5% GHb with those from two samples having ~5% GHb
failed to positively detect diglycosylated species at values noticeably
greater than background (~1%). We estimate that in these two highly
glycated samples, the diglycosylated species are probably <1% and are
certainly <2% of the nonglycated - and ß-chains.
This result is perhaps not unexpected. Previous work (13)
showed that each chain has several sites capable of being glycated with
single glucose units in addition to the N-terminus of the ß-chain.
However, the N-terminus of the ß-chain has the greatest probability
of being glycated, the other sites having lower probabilities. For the
two samples with high GHb, the ratio of monoglycated (G1)
to nonglycated (G0) ß-chain species was ~0.18 (by
ES-MS). Simple probability theory for this ratio predicts that the
ratio of diglycated (G2) to G0 species is at
most 0.007. This calculation assumes that two glycation sites in the
ß-chain have equal probability of being occupied, which would give
the maximum probability for the G2 species. If one site has
a lower probability than the other for adding glucose, then the
G2/G0 ratio must be <0.007. For the less
likely situations that three and four sites in the ß-chain would have
equal probabilities of being occupied by glucose, the maximum ratio of
G2/G0 would be 0.011 and 0.012, respectively.
For lower proportions of the G1 species, the predicted
amounts of the G2 species rapidly decrease; e.g., for the
case with two glycation sites having equal probability of being
occupied and G1/G0 = 0.12 and 0.06, the
predicted G2/G0 ratios would be 0.003 and
0.0008, respectively. For a given % GHb, the predicted
G2/G0 ratios for the -chain are about half
those for the ß-chain.
We suggest that proportions of G2 species occurring at GHb values up to ~15% can be neglected. At higher values for %GHb, the G2 species should become detectable, but this may be of only academic interest because higher glycation percentages are uncommon in our experience.
The analytical comparisons make clear that the chromatography-based procedures are in good agreement with both themselves and the ES-MS. From this we conclude that the analyte actually being measured is probably the same in all these instances, i.e., glycated hemoglobin. However, all the procedures have some problems in resolving potential interference signals, especially as observed in the nondiabetic patient group, whose GHb is in the 2.7% ± 0.5% range as measured by ES-MS. Several possibilities have been proposed to explain why these interferences are less prevalent in the diabetic population (9).
Mass analysis of the globin chains may also be used to identify the
presence of some hemoglobin variants in the same experiment, a
procedure well documented in the literature (15) and
demonstrated by the example in Fig. 6
. ES-MS thus offers the potential
for the simultaneous automated analysis of glycated and variant
hemoglobins.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2(ß116(G18) HisGln)2: a new hemoglobin variant mistaken for glycated hemoglobin. Biochim Biophys Acta 1988;955:214-219.
[Medline]
[Order article via Infotrieve]
acetylalanine). Structure and functional characterization. Biochemistry 1977;16:4872-4879.
[Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  R. H. Bateman, B. N. Green, and M. Morris Electrospray Ionization Mass Spectrometric Analysis of the Globin Chains in Hemoglobin Heterozygotes Can Detect the Variants HbC, D, and E Clin. Chem., July 1, 2008; 54(7): 1256 - 1257. [Full Text] [PDF]
|
|
|
|
 |
|
 F. Q. Nuttall Evidence for Independent Heritability of the Glycation Gap (Glycosylation Gap) Fraction of HbA1c in Nondiabetic Twins: Response to Cohen et al. Diabetes Care, April 1, 2007; 30(4): e14 - e14. [Full Text] [PDF]
|
|
|
|
|
|
U. Krishnamurti and M. W. Steffes Glycohemoglobin: A Primary Predictor of the Development or Reversal of Complications of Diabetes Mellitus Clin. Chem., July 1, 2001; 47(7): 1157 - 1165. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Bry, P. C. Chen, and D. B. Sacks Effects of Hemoglobin Variants and Chemically Modified Derivatives on Assays for Glycohemoglobin Clin. Chem., February 1, 2001; 47(2): 153 - 163. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. B. Roberts, A. B. Amara, M. Morris, and B. N. Green Long-Term Evaluation of Electrospray Ionization Mass Spectrometric Analysis of Glycated Hemoglobin Clin. Chem., February 1, 2001; 47(2): 316 - 321. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S Kilpatrick Glycated haemoglobin in the year 2000 J. Clin. Pathol., May 1, 2000; 53(5): 335 - 339. [Full Text] [PDF]
|
|
|
|
|
|
A. Lapolla, D. Fedele, M. Plebani, R. Aronica, M. Garbeglio, R. Seraglia, M. D'Alpaos, and P. Traldi Evaluation of Glycated Globins by Matrix-assisted Laser Desorption/Ionization Mass Spectrometry Clin. Chem., February 1, 1999; 45(2): 288 - 290. [Full Text] [PDF]
|
|
|
|
|
|
K. P. Peterson, J. G. Pavlovich, D. Goldstein, R. Little, J. England, and C. M. Peterson What is hemoglobin A1c? An analysis of glycated hemoglobins by electrospray ionization mass spectrometry Clin. Chem., September 1, 1998; 44(9): 1951 - 1958. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, lower curve), 
, upper curve),
ß-chain glycation = 0.59x + 0.06
(r2 = 0.9838).
m = +14 Da).