|
|
||||||||
Endocrinology and Metabolism |
Asp), a pitfall in the assessment of diabetic control: characterization by electrospray mass spectrometry and HPLC
1
Department of Clinical Chemistry, University Hospital, Hugstetterstrasse 55, D-79106 Freiburg, Germany.
2
Laboratoire de Spectrometrie de Masse Bio-Organique
URA31, CNRS-Université Louis Pasteur, Faculté de Chimie,
F-67008 Strasbourg, France.
3
Institut für Humangenetik Westfälische
Wilhelms-Universität, D-48149 Münster, Germany.
a Author for correspondence. Fax 49-761-270-3444.
| Abstract |
|---|
|
|
|---|
Asp, has been identified in
a German woman also suffering from non-insulin-dependent diabetes
mellitus and chronic obstructive pulmonary disease. This is the first
observation of this Hb variant in a German family thus far. The
detailed evaluation of its structure using electrospray mass
spectrometry revealed new minor glycohemoglobin components and showed
that the attachment of glucose to the ß NH2 terminus
occurred at an almost identical rate in both wild-type and mutant
ß-chains. However, the introduction of a carboxyl group at ß69
seems to increase the glycation of
-amino groups of lysine residues.
The glycemic state in the propositus was well reflected by the total
glycohemoglobin concentrations but not by the Hb A1C
values, which did not reflect hemoglobin glycation in this patient.
This case demonstrates that Hb A1C cannot be used reliably
in the management of diabetic patients carrying Hb variants such as Hb
Rambam. Functional studies of the whole blood of the heterozygous
carrier demonstrated extremely low oxygen affinity, which may have been
caused by increased 2,3-diphosphoglycerate related to chronic
obstructive pulmonary disease and hyperthyroidism. None of the clinical
symptoms could be directly associated to Hb Rambam. | Introduction |
|---|
|
|
|---|
In this study, electrospray ionization mass spectrometry (ESMS) enhanced by reversed-phase HPLC (rp-HPLC) was used to further characterize this rare Hb variant found in a German woman and her daughter. These detailed structural and functional studies complement the original report. In addition, the influence of this variant on the monitoring of diabetes mellitus using glycohemoglobin (GHb) isoforms is documented.
| Materials and Methods |
|---|
|
|
|---|
Hematologic and biochemical characteristics were determined by standard methods. Hb instability was measured using heat stability and isopropanol stability tests (4). Red cell hemolysates were studied by isoelectric focusing performed on agarose gel. The amounts of Hb A, Hb F, Hb Rambam, and Hb A2 were quantified by cation-exchange HPLC (5). The quantities of the globin chains were determined by rp-HPLC (6). The oxygen-binding properties of whole blood were studied as reported previously (6). The 2,3-diphosphoglycerate (DPG) concentrations were determined enzymatically in whole blood with commercially available kits (Sigma Chemical Co.).
dna analysis
Genomic DNA was isolated from white blood cells using disposable
columns from QIAGEN. PCR was used to amplify two fragments of the
ß-globin gene, the first fragment extending from position 161 5' to
the transcription initiation site to position 113 in intron 2 (773
bp), the second fragment amplifying from position 596 in intron 2 to
position 83 3' to the polyadenylation signal (565 bp). After
purification of the fragments on agarose gel, single-stranded DNA was
obtained by asymmetric amplification with one primer in excess
(7). Sequencing of the amplified single-stranded DNA was
performed using T7 DNA polymerase, [a-S]-dCTP
(Pharmacia), and the sequencing primer
SR15(5'-ACAGGTTTAAGGAGACCAATAGA-3') as described previously
(8).
esms
ESMS analysis was carried out on a VG BioQ mass spectrometer
(Micromass) fitted with an electrospray source operating at atmospheric
pressure and succeeded by a triple quadrupole analyzer. The solvent was
a mixture of 500 mL/L water500 mL/L acetonitrile containing 10 mL/L
formic acid. The sample diluted with the solvent was introduced into
the ionization chamber through a silica capillary of 75-µm internal
diameter at a flow rate of 6 µL/min. The potential applied between
the end of the capillary and the first electrode was 3.2 kV. The
solvent was evaporated using a flow of nitrogen heated at 70 °C. The
calibration was performed using horse heart myoglobin [2 µmol/L (2
pmol/µL)], and the scanning ranged from m/z = 250 to
m/z = 1800 Da. For the liquid chromatographymass
spectrometry experiments, the eluate from the column was split so that
10 µL/min was introduced in the ESMS and the remaining 190 µL was
collected after ultraviolet detection at 214 nm, as described elsewhere
(9). Data processing was performed using standard software
provided by the manufacturer.
Hemolysates were prepared as reported previously (5). Isolation of a sufficient quantity of ßRambam required a semipreparative rp-HPLC with a 200 x 9 mm, 5 µm, C4 Multospher column (Ziemer). Minor Hbs (Hb X1 and Hb A1) were isolated by cation-exchange HPLC using a 200 x 9.4 mm, 5 µm, PolyCAT A column. Nonglycated Hb and GHb were isolated by affinity chromatography on boronate gel (10). For structural studies, the isolated fractions were concentrated using membrane filtration.
The modified ß-chain was digested using endopeptidase Lys-C (Boehringer Mannheim), which cuts selectively in the NH2 terminus of the lysines. One hundred micrograms of the purified ß-chain was dissolved in 40 µL of buffer (0.1 mol/L Tris-HCl, 0.02 mol/L CaCl2, pH 8.18) and 10 µL of acetonitrile. Lys-C endopeptidase was then added to the solution at an enzyme to protein ratio of 1:20, and the mixture was incubated for 5 h at 37 °C. The peptides thus obtained were analyzed by liquid chromatographymass spectrometry. The different peptides were separated by rp-HPLC on a Nucleosil 300, 5 µm, C18 column using an acetonitrile gradient (060% acetonitrile at a gradient of 1%/min) in a mixture of water containing 1 g/L trifluoroacetic acid and acetonitrile containing 800 mg/L trifluoroacetic acid. Mass measurement allowed identification of all peptides, which were collected. The mutant peptide was sequenced using Edman degradation.
| Results |
|---|
|
|
|---|
-glutamyltransferase (42 U/L) were increased. Hematologic data
indicated no abnormalities (not shown). The mutant Hb was separated by
cation-exchange chromatography on PolyCAT A. The same Hb variant was
found in her 55-year-old daughter. The daughter's medical history
included hypertension, hyperthyroidism, and a moderate
hypercholesterolemia. She also had a degeneration of the dorsal
spine associated with marked postural abnormality.
Hb ANALYSIS
Isoelectric focusing revealed an Hb variant moving anodic to Hb A
(not shown). The analysis carried out with cation-exchange HPLC using a
PolyCAT A column showed two Hb variant fractions, Hb
X1 or Hb R1C and Hb X0
(Fig. 1
top). The concentrations of these Hb variants were 4.38% and
42.50%, respectively. Globin chain analysis by rp-HPLC showed a mutant
globin (ßX) eluting behind the wild-type
ßA-chain, thus elucidating increased hydrophobicity (not
shown). The relative quantity of ßX in the heterozygote
was 51.55% of the total ß-chains (ßX
ßA). The separation of the GHbs by boronate affinity
chromatography, followed by cation-exchange HPLC (Fig. 1
bottom),
showed five GHb fractions corresponding to Hb X, Hb A, and Hb
A2, respectively. The proportions of these GHbs are shown
in Fig. 1
(bottom).
|
oxygen affinity and stability studies
There was a rightward shift of the oxygen dissociation curve of
whole blood with a P50 of 66 mmHg. This
indicates an extremely low oxygen affinity compared with healthy
controls (P50, 2529 mmHg). Red cell 2,3-DPG
was 12.1 mmol/L (reference range, 4.56.2 mmol/L).
esms analysis
The hemolysate from the propositus was analyzed, and the resulting
profile (Fig. 2
) revealed three major peaks. The largest peak, with a mass of
15126 ± 0 Da corresponded to an unmodified
-chain (expected
mass, 15126.4 Da). The two other major peaks (at 15867 ± 1 and
15925 ± 1 Da), with close intensities, can be attributed to an
unmodified (expected mass, 15867.3 Da) and a modified ß-chain with a
mass excess of 58 Da. In view of the isoelectric focusing mobility of
the variant, which showed an increase of charge, the rise of the mass
could have resulted either from a Gly
Asp or an Ala
Glu
replacement. To identify which mutation was responsible for the
modified ß-chain and to determine the exact position of this
mutation, the modified ß-chain was purified by chromatography and
submitted to further analysis. It was subjected to cleavage with
endopeptidase Lys-C, and the resulting peptide mixture was analyzed by
liquid chromatographymass spectrometry. All chromatographic peaks
detected exhibited masses corresponding to Lys-C digest peptides
expected from the sequence of the unmodified ß-chain. However, the
peptide L7 (VLGAFSDGLAHLDNLK) with an expected mass of 1670 Da was
missing, and a peptide with a mass of 1728 Da was present. This peptide
was collected and submitted to Edman degradation, which yielded a
sequence corresponding to peptide L7 with an asparagyl residue in place
of a glycyl residue at position 3. This clearly indicated an exchange
of glycine by aspartic acid at position 69 of the ß-chain.
|
GHb fractions separated by cation-exchange HPLC and by affinity
chromatography were collected and analyzed by ESMS. Fig. 3
illustrates the glycation of both unmodified and modified
ß-chains. Fig. 3
, A and B, shows peaks of the glycated ß-chains
from isolated Hb X1 (16089 Da) and Hb A1C
(16027 Da), respectively. Fig. 3C
elucidates the distribution of the
glycoforms in the GHb fraction. Peaks from
-chains (15127 Da), both
unmodified (15867 Da) and modified (15924 Da) ß-chains, and their
glycoforms (15288 Da, 16029 Da, and 16087 Da, respectively) were
detected. The percentage of glycation of the modified ß-chain was
higher than that of the unmodified ß-chain (33.18% vs 21.67% of the
total GHbs, respectively). This result is consistent with the data
obtained by cation-exchange HPLC (Fig. 1
, bottom). Expected and
measured molecular masses for the unmodified and modified globin chains
are listed in Table 1
.
|
|
dna analysis
Direct sequencing of the amplified DNA from the propositus
confirmed the nucleotide change from GGT to GAT at codon 69 of the
ß-gene, which corresponds to a Gly
Asp replacement at position 69.
The analysis of the sample from the daughter gave the same results.
| Discussion |
|---|
|
|
|---|
and ß globin chains. Our data
indicate that the attachment of glucose to the ß NH2
terminus was approximately the same for both wild-type (Hb
A1C) and mutant (Hb X1) Hbs. On the other
hand, there was a substantial difference in GHb values between Hb
X0 and Hb A0. The high GHb percentage of Hb
X0 suggests that the replacement of a glycine residue at
E69 by aspartate may induce a decrease of the
pKa on the carboxyl side of some lysine residues
of the ß-chain (e.g., ß66 Lys-Lys), which thereby become more
reactive (11). Hb A1C is widely accepted as the most useful tool for monitoring the glycemic state in diabetic patients. In addition, overestimation or underestimation of this marker may occur in samples containing mutant Hbs, because several HPLC methods lack the resolution necessary to differentiate mutant Hb components (12). However, the measurement of glycated mutant Hbs in addition to Hb A1C in diabetic patients with congenitally modified Hb has not received wide attention (13), particularly since the introduction of the Hb A1C immunoassays (14). Our data illustrate that in the presence of Hb Rambam, the determination of Hb A1C alone is not appropriate for monitoring glycemic control. Indeed, in our patient Hb A1C values determined by ion-exchange HPLC as well as by immunological methods did not reflect the glycemic state, which nevertheless was reflected by the sum of the minor Hbs (Hb X1 Hb A1C) and by the total GHbs. This case points out the necessity of measuring GHbs by affinity chromatography or by high-resolution HPLC in patients carrying mutant Hb variants. Moreover, Hb glycation is not suitable for monitoring glycemic control in carriers of Hb variants with an acetylated NH2 terminus, such as Hb Raleigh, Hb Long Island, and Hb A2 Niigata (15).
Of the 12 known Gly
Asp substitutions in the ß-chains
(3), 2 Hb variants, Hb Moscva (16) and Hb
J-Auckland (17), have been reported as exhibiting molecular
instability with decreased oxygen affinity accompanied by mild anemia.
In these two variants, the aspartate residue introduced at ß24(B6)
and ß25(B7), respectively, seems to cause a substantial change in the
three-dimensional arrangement of the affected subunits. The results
reported earlier for Hb Rambam (1) and our data show that
this variant is stable and not associated with anemia. It would seem
likely that the aspartyl residue in position 69 is not involved in any
contact between chains or with the heme group and thereby exerts no
molecular disturbance. Hence, no explanation for the extremely low
oxygen affinity of whole blood observed in the propositus is evident
from the hematological and structural data. 2,3-DPG is an integral part
of the molecular mechanism of oxygenation of Hb. Therefore, its
concentration in the index patient and her daughter is sufficient to
account for the observed decrease in oxygen affinity of whole blood.
This increase of the 2,3-DPG concentration may be related to the
chronic obstructive pulmonary disease and hyperthyroidism diagnosed in
our patients. This is in agreement with data reported earlier in
patients with similar clinical states (18). From the
clinical point of view, the increased 2,3-DPG effect on a patient's Hb
will probably enhance the response to certain physiological or
pathological situations. Considering the complex and clinically
heterogeneous conditions in our patients, it is difficult to
demonstrate whether the resulting decrease in oxygen affinity is a
major advantage. There is no clear evidence connecting the clinical
features in our patients with Hb Rambam, and it is particularly
doubtful that the structural alteration of Hb contributed to diabetes
mellitus (19). In addition, marked hypercholesterolemia was
also reported in several members of the family with Hb Malmö,
which is a high-affinity Hb variant (20). Therefore, the
analogy of the clinical pattern of the propositus and her daughter
could be explained by other genetic linkages rather than by the
presence of Hb Rambam.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
Val) in a German male associated with high oxygen affinity and erythrocytosis. Acta Hematol 1991;85:212-216.
Asp) found in the USSR. Nature 1974;249:768-770.
[Medline]
[Order article via Infotrieve]
Asp]. Hemoglobin 1987;11:221-230.
[ISI][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
![]() |
E. Bisse, C. Schauber, N. Zorn, T. Epting, A. Eigel, A. Van Dorsselaer, H. Wieland, J. Kister, and L. Kiger Hemoglobin Gorwihl [{alpha}2{beta}25(A2)Pro->Ala], an Electrophoretically Silent Variant with Impaired Glycation Clin. Chem., January 1, 2003; 49(1): 137 - 143. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. M. Clarke and T. N. Higgins Laboratory Investigation of Hemoglobinopathies and Thalassemias: Review and Update Clin. Chem., August 1, 2000; 46(8): 1284 - 1290. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Bisse, N. Zorn, I. Heinrichs, A. Eigel, A. Van Dorsselaer, H. Wieland, J. Kister, and M. C. Marden Characterization of a New Electrophoretically Silent Hemoglobin Variant. Hb Saale OR alpha 2beta 2 84(EF8)Thr right-arrow Ala J. Biol. Chem., July 7, 2000; 275(28): 21380 - 21384. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |