Volume 25, Issue 1 , Pages 14-18, January 2011
Hemoglobin Raleigh results in factitiously low hemoglobin A1c when evaluated via immunoassay analyzer
Article Outline
Abstract
Background
Glycosylated hemoglobin (HbA1c) is commonly used to assess long-term blood glucose control in patients with diabetes mellitus. Numerous conditions including hemoglobinopathies can alter HbA1c measurements and cause misleading results.
Objective
To report a 13-year-old male with Type 1 diabetes mellitus who had low HbA1c measurements, despite persistent hyperglycemia.
Design/Methods
HbA1c was initially measured by immunoassay. Hb electrophoresis was then employed to assess potential Hb variants. Electrospray ionization (ESI) tandem mass spectrometry of isolated Hb and gene sequencing of the Hbβ gene were used to specifically identify the Hb variant.
Results
HbA1c measurement by immunoassay revealed an unusually low HbA1c of 3.9%. Hb electrophoresis revealed an aberrant Hb. The ESI mass spectrum of the intact Hb sample revealed a variant β-chain of 15,881 Da, 14 Da heavier than the mass of the normal Hb β-chain (15,867 Da). Sequence analysis of the 965.45 Da peptide suggested a substitution of valine (Val) to acetylated alanine (Ala). The DNA sequence of the patient's Hbβ gene revealed a single-base heterozygous mutation (GTG to GCG) at Base 2 of the codon of the first amino acid, producing a Val→Ala substitution, previously termed Hb-Raleigh. Because the acetylated N-terminal amino acid of the Hb-Raleigh β chain cannot be glycated, the HbA1c immunoassay will result in falsely low HbA1c levels.
Conclusion
In managing diabetic patients, knowledge of hemoglobinopathies influencing HbA1c determination methods is essential because hemoglobin variants may cause mismanagement of diabetes. Unusual results should prompt further analysis for a hemoglobinopathy as the potential cause of aberrant results.
Keywords: Hb Raleigh, HbA1c, Hemoglobinopathy, Mass spectrometry
1. Introduction
Blood glycohemoglobin A1c (HbA1c) is routinely measured to monitor long-term glycemic control in patients with diabetes mellitus. HbA1c levels have been shown to correlate with long-term risk of complications associated with diabetes and has become a “gold standard” in the management of diabetes mellitus. Over 5.1% of adult Americans have been diagnosed with diabetes mellitus and it is recommended that these individuals have HbA1c levels tested every 3–6 months (American Diabetes Association, 2003, Thomas et al., 2007). Over 2 million HbA1c tests are performed each month in the United States (Thomas et al., 2007).
There are more than 20 methods for determining HbA1c concentrations based on physical, chemical or antibody recognized characteristics. However, many of the more commonly used methods can be affected by structural or chemical variations in the hemoglobin (Hb) chain resulting in inaccurate HbA1c measurements and thus potentially affecting patient care (Bry, Chen, & Sacks, 2001). One of the more prevalent methods uses antibody mediated inhibition of latex agglutination. Antibodies recognize the glycated N-terminal amino-acid in the first four to ten amino-acid sequence of the hemoglobin β chain. Thus, any hemoglobin with variation in the first four to ten amino-acids may prevent glycosylation, thus resulting in falsely low HbA1c measurements.
Over 950 hemoglobin variants have been described (Patrinos et al., 2004). Despite advances in the standardization of methods measuring glycohemoglobins, a number of hemoglobinopathies cause false results in HbA1c determinations, thus presenting a unique challenge to clinical practitioners. Here, we report a child with Type 1 diabetes mellitus (T1DM) who had disproportionately low HbA1c levels and was subsequently found to have the hemoglobin variant Hb Raleigh.
2. Case report and methods
A 13 3/12-year-old Caucasian male presented to the Children's Diabetes Clinic at the University of North Carolina at Chapel Hill for a follow up visit one month after being diagnosed with T1DM at a local hospital. His initial symptoms included frequent headaches, fatigue, increased thirst and urine output and approximately 20-lb weight loss within the two to three weeks prior to diagnosis. He later developed abdominal discomfort with vomiting. His initial evaluation by his primary care physician revealed moderate dehydration along with a blood glucose greater than 600 mg/dl and ketosis. He was then admitted to the local pediatric intensive care unit for two days. Insulin treatment was initiated and the patient was referred to the University of North Carolina diabetes clinic for further diabetic teaching and management.
At the time of his first clinic visit, about 3 weeks after diagnosis, HbA1c analysis using a DCA2000 Immunoassay Analyzer (Bayer, Pittsburg, PA, USA) revealed a HbA1c of 6.5%. At that time antibody studies revealed positive insulin and islet cell antibodies. While his presentation and positive antibodies were consistent with T1DM, his HbA1c level was not consistent with his blood sugars or with the severity of his presentation.
At his second clinic visit, 4 months after diagnosis, he was found to have an HbA1c of 3.9%, also determined via the DCA2000 Immunoassay Analyzer by Bayer, which did not correlate with his data of self-monitored blood glucoses. Fructosamine level was 248 μmol/l (approximating a HbA1c of 6%). Fructosamine represents a clinically accessible measure of nonenzymatic glycation of proteins in the same compartment as plasma glucose, and integrates plasma glucose fluctuations within the previous few weeks. Furthermore, it is not affected by the turnover of red cells or hemoglobinopathies. Therefore, a blood sample was analyzed for a possible hemoglobinopathy. Hemoglobin species were analyzed by electrophoresis, high-performance liquid chromatography (HPLC), electrospray ionization (ESI) mass spectrometry, and tandem mass spectrometry (MS/MS). For electrophoresis and HPLC, conventional methods were used.
2.1. Intact Hb analysis by MS
Cells were lysed by diluting 10 μl of blood with 490 μl distilled water. An aliquot from this solution was diluted 10 times with MS injection buffer (50% acetonitrile, 1% formic acid) and introduced to a Waters Q-Tof micro mass spectrometer via a syringe pump with a flow rate of 1 μl/min.
2.2. Peptide sequencing by MS/MS
Of the lysed cell supernatant, 100 μl was denatured with 10 μl acetonitrile and 1% formic acid. Ammonium bicarbonate was then added to make a 50 mM solution. Proteomics grade trypsin (Sigma, 5 μg/μl) was added and the sample was digested in a conventional microwave oven (5 min at 60% power). Digested samples were centrifuged briefly and 10 μl of supernatant were diluted 10 times with 1% formic acid and introduced to the Q-Tof micro mass spectrometer via a Waters CapLC nano HPLC solvent delivery system.
2.3. Analysis of the mass spectrometry data
The raw data from MS analysis of intact hemoglobin was processed via MaxEnt software. The raw data from MS/MS analysis were processed using Proteinlynx module of Masslynx 4.0 software to produce *.pkl (peaklist) files. The processed data (.pkl files) were searched against updated NCBInr (version 20060805) and Sprot databases (release 48.7) using an in house MASCOT search engine (version 2.0).
2.4. Gene sequencing
DNA sequencing of the β-globin gene was performed using DNA primers that flank the three exons of the β-globin genes as previously described (Hoyer et al., 1998). DNA was amplified using AmpliTag DNA polymerase (Applied Biosystems, Foster City, CA, USA) on a Perkin ElmerGene Amp PCR System 9600 (Perkin Elmer, Norwalk, CT, USA). DNA sequencing was performed using the ABI Prism 377 automated sequencer (Applied Biosystems).
This study was reviewed and approved by the Office of Human Research Ethics Institutional Review Board at the University of North Carolina and is in accordance with the Declaration of Helsinki.
3. Results
Hemoglobin electrophoresis revealed 49.1% HbA, with normal expected percentages being 95.1-98.5. An unusually high amount of a β-chain hemoglobin variant was identified (47.4%). Normal amounts of HbA2 (3%) and HbF (0.5%) were detected (Fig. 1A).
Fig. 1.
(A) Hb gel electrophoresis: denatured Hb protein analyzed by gel electrophoresis. Top lane representing ladder. Middle lane, control sample of hemoglobin. Bottom lane, patient sample, reveals the presence of a β-chain hemoglobin variant. (B) Hb HPLC analysis . HPLC revealed the presence of Hb Raleigh in the patient sample. (C) Hb ESI mass spectra analysis. ESI mass spectra of intact Hb reveals two β-chain peaks separated by 14.0 Da indicating the presence of a variant β-chain as heterozygote. (D) MS/MS spectra of N-terminal tryptic peptide of the variant Hb β-chain. Two different precursor ions were observed from the N-terminus: One 951.49 Da with sequence of VHLTPEEK (normal, not shown) and the other 965.45 Da with sequence of A*HLTPEEK (variant). The 14-Da mass difference was equal to Val→Ala substitution (−28 Da) and acetylation of alanine (+42). (E) DNA sequencing of exon 1 of the beta globin gene in the forward direction. There is a GTG to GCG substitution at codon 1 (arrow), consistent with Hb Raleigh.
HPLC revealed two prominent Hb variants with differing eluting properties in the patient sample (Fig. 1B).
ESI mass spectrometry was used to further delineate the size differential between the two prominent hemoglobin variants found in the patient sample. Analysis revealed that the variant β chain was 14.0 Da heavier than βA and that the abundance of the variant hemoglobin β-chain was similar to that of the normal β-chain (Fig. 1C).
Amino acid sequence analysis of the digested Hb sample by tandem mass spectrometry allowed further analysis of the mass and molecular charge of the hemoglobin variant and identification of the specific variant. Data analysis revealed posttranslational modification of the hemoglobin molecule, specifically the substitution of the N-terminal valine of the β chain with an acetylated alanine (β1Val→Ac-Ala) characteristic of hemoglobin Raleigh (Fig. 1D) (Chen et al., 1998, Rai et al., 2002).
The DNA sequence of the patient's Hbβ gene revealed a GTG to GCG substitution at codon 1 of the beta globin gene (Fig. 1E), producing a Val→Ala substitution, previously termed Hb-Raleigh.
Similar analyses in the mother also revealed the same hemoglobin variant.
4. Discussion
The current gold standard for monitoring long-term glycemic control in diabetes mellitus is the measurement of HbA1c. HbA1c is hemoglobin that has been irreversibly glycated and represents the mean blood glucose level over the preceding three months as this is the normal, average lifespan of red blood cells. Unfortunately, several conditions may affect HbA1c determination including decreased lifespan of red blood cells secondary to massive bleeding, hemolytic anemia, chronic disease or pregnancy induced anemia. In addition, other causes affecting HbA1c determination are uremia, opiate addiction, chronic alcohol abuse, high dose aspirin, hyperbilirubinemia, vitamin C excess, vitamin E excess, iron deficiency anemia or hypertriglyceridemia (Bry et al., 2001, Moriwaki et al., 2000, Schnedl et al., 2001). An increasing number of hemoglobinopathies cause false results in glycated hemoglobin determinations due to hemolysis or altered detectability in glycated hemoglobin assays. While most of the factors altering HbA1c measurements are likely to be associated with clinical suspicion, most hemoglobinopathies are clinically silent. Thus clinicians must consider the presence of a hemoglobinopathy if patients exhibit unexpected HbA1c results.
There are several methods used to measure HbA1c including cation-exchange chromatography, electrophoresis, isoelectric focusing, immunoassay, and boronate affinity chromatography (Bry et al., 2001, Chen et al., 1998, Moriwaki et al., 2000, Roberts et al., 2000, Schnedl et al., 2001). Unfortunately, each method is subject to interference, for example by chemical structure and/or molecular charge, potentially resulting in inaccurate HbA1c measurements. Immunoassay is one of the more common methods of HbA1c determination being used. This method measures antibody mediated inhibition of latex agglutination. Antibodies recognize glycosylation of the first six N-terminal amino acids of the hemoglobin β-chain (Bry et al., 2001). Thus, any factor preventing glycosylation or identification of the first six N-terminal amino acids will result in falsely low HbA1c measurement by immunoassay methods.
When a hemoglobinopathy is suspected, the patient's specific hemoglobin has classically been characterized using methods such as electrophoresis or liquid chromatography. More recently, ESI and electrospray tandem mass spectrometry (ES-MS/MS) have been used to help characterize hemoglobin variants. The advantages of ESI and ES-MS/MS include short analysis time and the need for small amounts of sample. ES-MS/MS also allows for characterization and identification of specific hemoglobin variants (Sacks, 2003, Schnedl et al., 2001). We therefore believe that this method has great potential to replace current screening methods for hemoglobinopathies in the newborn period.
Using ES-MS/MS, we identified a variant hemoglobin, Hb Raleigh, in our patient. Hemoglobin Raleigh is a variant of the hemoglobin β-chain in which the normally positioned valine at position 1 of the hemoglobin β-chain is substituted by alanine. This amino-terminal alanine is posttranslationally modified by acetylation which in turn prevents glycosylation (Chen et al., 1998). There are other Hb variants with acetylated amino acids including Hb-Long Island [α2β21(NA1)Val→Ala], HbA2-Niigata [α2β21(NA1)Val→Leu] (Chen et al., 1998), and Hb South Florida [2β21(NA1)Val→Met], all of which have been shown to prevent glycoslation (Bry et al., 2001, Schnedl et al., 2001). Prevention of glycosylation resulted in falsely low HbA1c measurements by immunoassay.
Monitoring glycemic control for this patient has remained challenging as Hb Raleigh has been shown to adversely affect other commonly used methods of HbA1c determination including the ion exchange HPLC (falsely elevated) and boronate affinity chromatography (falsely decreased) methods. Glycemic control has been monitored by closely evaluating multiple capillary blood glucose measurement throughout the day and utilizing measurements of other glycated serum proteins such as fructosamine. Future management using a continuous glucose monitor may also be considered.
In managing diabetic patients, knowledge of hemoglobinopathies influencing HbA1c determination methods is essential because unidentified hemoglobin variants may cause mismanagement of diabetes. With the presence of several hemoglobin variants that may otherwise be clinically silent, unusual results from HbA1c testing should prompt further analysis to determine if a hemoglobinopathy is the cause of aberrant results and practitioners should be aware of the potential affect of hemoglobinopathies on common HbA1c analysis methods.
References
- . Standards of medical care for patients with diabetes mellitus. Diabetes Care. 2003;26(Suppl. 1):S33–S50
- . Effects of hemoglobin variants and chemically modified derivatives on assays for glycohemoglobin. Clinical Chemistry. 2001;47:153–163
- . Hemoglobin Raleigh as the cause of a falsely increased hemoglobin A1c in an automated ion-exchange HPLC method. Clinical Chemistry. 1998;44:1296–1301
- . Hb Silver Springs [® 131 (Ha) Glu → His ], a new hemoglobin variant found in six African-Americans. Hemoglobin. 1998;22:37–44
- . Abnormal haemoglobins, Hb Takamatsu and Hb G-Szuhu, detected during the analysis of glycated haemoglobin (HbA1c) by high performance liquid chromatography. Journal of Clinical Pathology. 2000;53:854–857
- . Improvements in the HbVar database of human hemoglobin variants and thalassemia mutations for population and sequence variation studies. Nucleic Acids Research. 2004;32(Database issue):D537–D541http://globin.cse.psu.edu/hbvar/menu.html
- . Identification of N-terminal acetylation in Hb Raleigh (β1Val→Ac-Ala) by electrospray tandem mass spectrometry. Rapid Communications in Mass Spectrometry. 2002;16:1793–1796
- . Effects of nine hemoglobin variants on five glycohemoglobin methods. Clinical Chemistry. 2000;46:569–572
- . Hemoglobin variants and hemoglobin A1c analysis: problem solved?. Clinical Chemistry. 2003;49:1245–1247
- . Hemoglobin variants and determination of glycated hemoglobin (HbA1c). Diabetes/Metabolism Research and Reviews. 2001;17:94–98
- . Screening for hemoglobinopathies during routine hemoglobin A1c testing using the Tosoh G7 Glycohemoglobin Analyzer. Annals of Clinical and Laboratory Science. 2007;37:251–255
PII: S1056-8727(09)00095-6
doi:10.1016/j.jdiacomp.2009.09.004
© 2011 Elsevier Inc. All rights reserved.
Volume 25, Issue 1 , Pages 14-18, January 2011
