HEMATOPATHOLOGY
Original Article
Hemoglobin Sherwood Forest Detected by
High Performance Liquid Chromatography
for Hemoglobin A l c
WOLFGANG J. SCHNEDL, MD, 1 EMIL C. REISINGER, MD, 1 THOMAS R. PIEBER, MD, 1
RAINER W. LIPP, MD, 1 FLORIAN SCHREIBER, MD, 1 PIERRE HOPMEIER, MD, 2
AND GUENTER J. KREJS, MD 1
Hb Sherwood Forest has been so far identified in only one patient in
1977. This study describes the second detection of this hemoglobin variant by routine high performance liquid chromatography (HPLC) in a
diabetic patient and her healthy grand niece. In both, glycosylated hemoglobin (HbA]C) values were excessively elevated (52%), as determined by HPLC with a cation exchange column. The latex agglutination test showed values of HbAic in the expected normal range. Citrate
agar electrophoresis revealed a hemoglobin variant with a mobility sim-
ilar to HbF. Amino acid analysis and DNA sequence analysis revealed
an Arg-»Thr exchange at codon 104 of the /3-chain. This sequence has
been described as Hb Sherwood Forest in 1977. The hemoglobin variant is clinically silent and might be confused with excessively high
HbAlc in routine measurement of glycosylated hemoglobin. (Key words:
Glycated hemoglobin; MbAlc; Hb Sherwood Forest; Hemoglobinopathy; /3-chain mutation; High performance liquid chromatography) Am
J Clin Pathol 1995; 104:444-446.
investigation revealed in one nondiabetic healthy grand niece HbA,c
values of 52.1% with HPLC and 4.5% with latex immunoagglutination
accompanied by a serum fructosamine level within normal range. Routine hematologic indices and all other laboratory parameters were normal in all investigated family members. The study was done in accordance with the Helsinki Declaration of 1975.
Automated HPLC (Diamat Analyzer System, Bio-Rad Laboratories,
Vienna, Austria) with a cation exchange column was used for quantification of HbAic.7 'n addition, HbAic was determined by a recently
described immunoassay based on the inhibition of latex agglutination
with a monoclonal antibody specific for a certain amino acid sequence
of the HbAlc molecule (DCA 2000 Haemoglobin Alc Reagent Kit,
Bayer Diagnostics, St. Denis, France).8 Serum fructosamine levels were
quantified colorimetrically with nitroblue tetrazolium.9
Citrate agar electrophoresis (citrate agar plates, Helena Labs, Beaumont, TX) of hemoglobin using citrate buffer at pH 6.2,'° showed a slow
cathodal moving variant with a mobility similar to that of HbF. HemoCASE REPORT
globin electrophoresis on cellulose acetate membranes (Helena Labs)"
was normal. Isoelectric focusing on commercial gel plates (Ampholine
An 88-year old female with noninsulin-dependent diabetes mellitus
PAG plates, Pharmacia LKB, Uppsala, Sweden)'2 revealed no pathologic
and hypertension was admitted to our department because of recurrent
bands. The heat stability test did not show unstable hemoglobins. Amino
diarrhea. The HbAlc values were 51.8% with HPLC (Fig. 1) (normal
acid analysis of isolated tryptic peptides'3 and sequence analysis of amplirange 4.3%-6.1%) and 5.8% with immunoagglutination test (normal
fied DNA14 (kindly done by Dr. Huisman and colleagues, Medical College
range 3.7%-5.7%). The serum fructosamine value was normal. Family
of Georgia, Augusta, GA), revealed the silent hemoglobin variant
a2i82l04(G6)Arg-»Thr, namely Hb Sherwood Forest.13
Quantification of HbA215 and HbF16 and the intracellular distribuFrom the 'Department of Internal Medicine, Karl Franzens Univer- tion of HbF were normal. Heinz bodies and HbH inclusion bodies were
sity. Graz. Austria: and2Central Laboratory. City Hospital Rudolfstif- not found. The numbers of peripheral blood cells by Coulter counter
and the number of reticulocytes were also normal.
tung. Vienna. Austria.
The determination of glycosylated hemoglobin (HbA,c)
has gained widespread use in routine management of patients with diabetes mellitus.' Glucose linkage to HbAlc
reflects mean blood glucose levels during the last 3
months before examination. Approximately 700 variants of human hemoglobin currently are known.2 A few
of these may be mistaken for glycosylated hemoglobin in
certain assays.3"5 We report the occurrence of Hb Sherwood Forest, detected by routine high performance liquid chromatography (HPLC) in a borderline diabetic patient and her nondiabetic relative. This hemoglobin
variant has been described in only one previous case in
1977.6
Manuscript received January 31, 1995; revision accepted May 11,
1995.
Address reprint requests to Dr. Schnedl: Department of Internal
Medicine, Karl Franzens University, Auenbruggerplatz 15,8036 Graz,
Austria.
DISCUSSION
This study reports on a diabetic patient and her nondiabetic relative, both presenting with excessively ele444
SCHNEDL ET AL.
445
Hemoglobin Sherwood Forest
20
r
c
0)
Q_
0 1
Time (min)
FIG. 1. HPLC chromatogram and hemoglobin report (Bio-Rad, Diamat) ofa patient with Hb Sherwood Forest (HbA, a : 0.8%, HbA lb : 3.5%,
HbA, c :51.8%, HbA0: 43.9%).
vated HbAic values that were repeatedly determined
with HPLC. As expected, HbA,c determinations with latex agglutination revealed values within the normal
range. Citrate agar electrophoresis showed a cathodal
variant similar to HbF. Structural analysis of the hemoglobin /3-chain and sequence analysis of DNA revealed
the silent hemoglobin variant Hb Sherwood Forest. In
addition to the slightly unstable Hb Campertown,
«202lO4(G6)Arg-»-Ser,'7 Hb Sherwood Forest is the second hemoglobin variant found with a mutation in position 104 of the /3-chain.6 The findings of Hb Sherwood
Forest in the index patient and her grandniece, without
evidence of incest in this family, confirms a heterozygote, autosomal dominant mode of inheritance as shown
for other hemoglobin variants.13
Determination of glycosylated hemoglobin (HbA,c) is
routinely performed for standard management in patients with diabetes mellitus. Glucose is linked to the Nterminal valine of the /?-chain through ketamine linkage
and, if properly assayed, the percentage of glycosylated
hemoglobin estimates the diabetic control during the
preceding 3-month period. Methods to analyze glycosylated hemoglobin include colorimetry, electrophoresis,
isoelectric focusing, column chromatography, and the
recently introduced immunoagglutination test based on
monoclonal antibodies. High performance liquid chromatography with cation exchange columns (Bio-Rad,
Diamat) was recently selected to be a standard method
for HbAlc determination by Weykamp and colleagues.18
However, there is no consensus on a proper reference
method. The recognition of abnormal high HbAlc values
caused by hemoglobin variants is essential for routine
management of diabetes mellitus. The recently published results of the Diabetes Control and Complications
Trial (DCCT) predict a dramatic increase of routine glycohemoglobin testing.' Repeated measurements of
HbAlc on separate days and from different samples are
necessary to reduce the possibility of laboratory errors.
Additional silent hemoglobin variants interfering with
the measurement of glycosylated hemoglobin have been
described previously.3"519 Elevated fetal hemoglobin is
known to cause high HbAic values in some chromatographic procedures, possibly leading to misdiagnosis and
subsequent management failures of diabetes mellitus.20
A clear-cut separation of HbF and HbA,c may be
achieved with HPLC and with isoelectric focusing. In
case ofa missing clinical explanation for repeatedly high
HbAlc with HPLC, determination of HbAi by an alternative assay system such as isoelectric focusing or immunoagglutination and screening for hemoglobin variants2' seems appropriate.
Acknowledgments. The authors thank Dr. H. Stubchen for expert
technical assistance and Ms. Barbara Semlitsch for excellent patient
care.
Vol. 104. No. 4
REFERENCES
Goldstein DE, Little RR, Wiedmeyer HM, et al. Is glycohemoglobin testing useful in diabetes mellitus? Lessons from the Diabetes Control and Complications Trial. Clin Chem 1994,40:
1637-1640.
International Hemoglobin Information Center Variant List. Hemoglobin 1994:18:77-183.
Shah SC, Malone JI, Boissel JP, Kasper TJ. Hemoglobin South
Florida: New variant with normal electrophoretic pattern mistaken for glycosylated hemoglobin. Diabetes 1986; 36:10731076.
BlankeS, Johnsen A, Wimberley PD, Mortensen HB. Hemoglobin
Hafnia: c*2(/3116(G18) His-»Gln)2; a new hemoglobin variant
mistaken for glycosylated hemoglobin. Biochim Biophvs Acta
1988;955:214-219.
Wajcman H, Blouquit Y, Riou J, et al. A new hemoglobin variant
found during investigations of diabetes mellitus: Hb Pavie
(a 135 H18 Val—Glu). Clin ChimActa 1990; 188:39-48.
Ryrie DR. Plowman D, Lehmann H. Haemoglobin Sherwood
Forest /3104(G6)Arg—Thr. FEBS Lett 1977:83:260-262.
Bisse E. Wieland H. High-performance liquid chromatographic
separation of human haemoglobins: Simultaneous quantitation
of foetal and glycosylated haemoglobins. J Chromatogr
1988;434:95-110.
446
HEMATOPATHOLOGY
Original Article
8. Marrero DG, Vandagriff JL, Gibson R. et al. Performance of new
HbAic system in pediatric outpatient population. Diabetes Care
1992:15:1045-1049.
9. Armbruster DA. Fructoseamine: Structure, analysis, and clinical
usefulness. Clin Chem 1978;33:2153-2163.
10. Milner'PF, Gooden HM. Rapid citrate-agar electrophoresis in routine screening for hemoglobinopathies using a simple hemolys,a\t. Am J Clin Pathol 1975;64:58-64.
11. Briere RO, Golias T, Batsakis JG. Rapid qualitative and quantitative hemoglobin fractionation: Cellulose acetate electrophoresis. Am J Clin Pathol 1965;44:659-701.
12. Basset P, Benzard Y, Garel MC, Rosa J. Isoelectric focusing of
human hemoglobin: Its application to screening, to the characterization of 70 variants, and to the study of modified fractions
of normal hemoglobins. Blood 1978;51:971-982.
13. Liu JS, Molchanova TP, Gu LH, et al. Hb-Graz or a2/322(NA2)
His-*Leu; a new 0 chain variant observed in four families from
southern Austria. Hemoglobin 1992; 16:493-501.
14. Oner R, Oner C, Wilson JB, et al. Dominant beta-thalassaemia
trait in a Portuguese family is caused by a deletion of
(G)TGGCTGGTGT(G) and an insertion of (G)GCAG(G) in
codons 134, 135, 136 and 137 of the beta-globin gene. Br J
Haematol 1991;79:306-310.
15. Hamilton SR, Miller ME, Jessop M, Charache S. Comparison of
microchromatography and electrophoresis with elution for hemoglobin A2 (Hb A2) quantitation. Am J Clin Pathol 1979;71:
388-396.
16. Kleihauer E, Braun H, Betke K. Demonstration von fetalem Hiimoglobin in Erythrocyten eines Blutausstrichs. Klin Wochenschr 1957;35:637-638.
17. Wilkinson T, Ching GC, Carrel RW, et al. A new haemoglobin
variant, Haemoglobin Campertown j3104(G6)Arginine-»-Serine. Biochim BiophysActa 1975;393:195-200.
18. Weykamp CW, Penders TJ, Muskiet FAJ, van der Silk W. Effect
of calibration on dispersion of glycohemoglobin values determined by 111 laboratories using 21 methods. Clin Chem
1994;40:138-144.
19. Schnedl WJ, Reisinger EC, Pieber TR, et al. Haemoglobin Graz,
a novel haemoglobin variant [Letter]. Diabetologia 1995;38:
122.
20. Robertson AD, Tunbridge FK.E, John WG, et al. Diagnostic confusion in diabetes with persistence of fetal hemoglobin. Br Med
71992:305:635-637.
21. Sandhaus LM, Harvey FG. Laboratory methods for the detection
of hemoglobinopathies in the community hospital. Clin Lab
Med 1993; 13:801-816.
A.J.C.P.-October 1995