International Journal of Laboratory Hematology

The Official journal of the International Society for Laboratory Hematology

REVIEW INTERNAT IONAL JOURNAL OF LABORATO RY HEMATO LOGY

Interpreting elevated fetal hemoglobin in pathology and health at

the basic laboratory level: new and known c- gene mutations
associated with hereditary persistence of fetal hemoglobin
A. AMATO, M. P. CAPPABIANCA, M. PERRI, I. ZAGHIS, P. GRISANTI, D. PONZINI, P. DI BIAGIO

ANMI Onlus, Centro Studi S U M M A RY

Microcitemie, Rome, Italy
Correspondence:
Antonio Amato, ANMI Onlus
(Associazione Nazionale per la
lotta contro le Microcitemie in
Italia), Centro Studi Microcite-
mie, Rome, Italy.
Tel.: 06 4395100;
Fax: 06 4394645;
E-mail: microcitemieroma@
blod.info
doi:10.1111/ijlh.12094
Received 28 December 2012;
accepted for publication 13
March 2013
Keywords
Fetal hemoglobin, hereditary
persistence of fetal hemoglobin,
novel mutations, Ac globin
gene, erythropoietic stress
Fetal hemoglobin may be slightly or significantly elevated in post-
natal life due to a number of causes. We report two novel mutations
found on the promoter of the Ac gene and summarize all common
and rare determinants associated with hereditary persistence of fetal
hemoglobin (HPFH) described thus far. Hematological and molecu-
lar analysis of the Ac globin gene in two cases of HPFH. Comparison
of the novel cases with all those described in the literature. We have
found two novel mutations in three Italian patients with HbF values
between 5.9% and 6.5% without an elevated HbA2 and with nor-
mal hemoglobin parameters. In two probands (mother and son), a
197 C>T transition was observed, while in a single individual, a
113 A>G transition was present on the distal CCAAT box of the
Ac gene. As no other abnormalities were present in both c-gene
promoters and the changes are located on regulatory sequences, we
may conclude that these mutations are responsible for the HPFH
phenotype shown by the carriers. The laboratory should be able to
discriminate between elevated HbF due to artifacts or to serious
causes including bone marrow malignancies, aplastic anemia, and
b-thalassemia major or recessive traits such as b-thalassemia minor,
db-thalassemia, or nonpathological conditions induced by mutations
or polymorphisms of the c-gene promoters that may even be bene-
ficial when present in patients with thalassemia major or sickle cell
disease and, in particular, when these patients are treated with
hydroxyurea.

from about 10 weeks of gestation until birth, in

INTRODUCTION
Fetal hemoglobin (HbF) is the high oxygen affinity
tetramer that can transfer oxygen from the maternal
to the fetal circulation. While predominant in the fetus
normal conditions only traces of HbF (<1%) are pres-
ent in postnatal life after the age of 1 year. Conversely,
HbF may remain elevated after birth due to pathologi-
cal conditions such as b-thalassemia major or in

© 2013 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2014, 36, 13–19 13
14 A. AMATO ET AL. | PERSISTENCE OF FETAL HEMOGLOBIN IN ADULT LIFE
potentially pathological b-thalassemia minor, db- or
cdb-thalassemia, or in nonpathological conditions such
as hereditary persistence of fetal hemoglobin (HPFH).
HbF may also become slightly or substantially elevated
sometime in life as a consequence of erythropoietic
stress, in a number of bone marrow malignancies, in
pregnancy, and, last but not least, HbF can be mea-
sured elevated due to technical artifacts.
Understanding the hemoglobin family
All human hemoglobins are made of 4 globin chains,
two a-like and two b-like chains. The globin chains
are coded by globin genes located on different clusters,
the a-cluster on chromosome 16 and the b-cluster on
chromosome 11. The a-cluster contains three active
genes, the zeta (f) gene expressed during embryonic
life, and the a1 and a2 genes expressed in both fetal
and postnatal life. The b-cluster contains five active
genes with differential expression during development,
the embryonic e, the fetal Gc and Ac and the postnatal
d and b genes. During embryonic, fetal and postnatal
life these genes will respectively code for the embry-
onic tetramers Hb Gower 1 (2f/2e), Gower 2 (2a/2e),
and Hb Portland (2f/2c), the fetal HbF (2a/2c) and the
postnatal HbA2 (2a/2d) and HbA (2a/2b). At birth, a
baby has in his blood on average 20% HbA and 80%
HbF that 1 year later will be almost totally replaced by
HbA, the major Hb in postnatal red cells and about
2.5% of Hb A2. At the age of two, the level of HbF in
normal conditions should be lower than 1% and any
level of HbF higher than that has a reason [1].
This review, while updating the list of HPFH
determinants providing the characterization of two new
mutations, is meant to assist the clinical chemist and the
technician in interpreting the significance of abnormal
Table 1. Hematological and analytical results of the presented cases
RBC Hb MCH MCV
Patient Age (1012/L) (g/dL) (pg) (fl)
A* 36 4.76 13.9 29.1 85
B* 38 4.52 12.3 27.2 82
C* 13 5.04 15.5 30.7 89
*A and B are independent subjects. C is a child of B.
HbF levels as are often measured during routine Hb
analysis.
M AT E R I A L S A N D M E T H O D S
Three healthy subjects with elevated HbF observed dur-
ing the regional school screening for thalassemia pre-
vention, two independent adults parents of screened
children and one related child.
Screening was performed as previously reported [2].
Routine hematology was obtained using an automated
counter (Advia 2120 Hematology Systems, Bayer
Health Care-Diagnostics Division – Milan, Italy). Single
tube osmotic fragility was carried out as previously
described [2]. Hb separation and measurement was
taken on alkaline electrophoresis and on High Perfor-
mance Liquid Chromatography (HPLC) (Variant II,
Bio-Rad Laboratories, Hercules, CA, USA) [3]. DNA
was extracted by high-salt extraction [4]. Point muta-
tion analyses of the full b-gene and of the c gene pro-
moters were carried out by direct sequencing on the
Beckman Coulter CEQTM 8000 Genetic Analysis System
(Beckman Coulter Inc., Fullerton, CA, USA) using the
primers and the procedures previously described [5, 6].
R E S U LT S
The 3 probands presented with elevated HbF values
around 6% and with normal hematological parame-
ters, osmotic fragility, red cell morphology, and HbA2
values (data are summarized in Table 1). Direct
sequencing of the b-genes revealed no abnormalities,
while sequencing of the c gene promoters disclosed
two novel mutations on the promoters of the Ac
genes. One, a 113 A>G transition was found on the
distal CCAAT box, the other, a 197 C>T transition
Hb A2 Hb F b Gene c Genes
(%) (%) sequence sequence
A
2.4 6.5 IVS 2-74 G>T c 113 A>G
A
2.4 6 Normal c 197 C>T
A
2.2 5.9 Normal c 197 C>T
© 2013 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2014, 36, 13–19
 A. AMATO ET AL. | PERSISTENCE OF FETAL HEMOGLOBIN IN ADULT LIFE
Figure 1. schematic representation of the region of interest with the location of the new mutations indicated.
–113 A>G mutation –197 C>T mutation
Figure 2. Partial fragment of the DNA sequence of the
Ac promoter showing the two new mutations in
heterozygous form.
was located internal to the 200 region contiguous to
the GC rich Sp1 box (Figures 1 and 2 and Table 1).
DISCUSSION
Defining HPFH
The definition of HPFH goes back to the time when
these conditions were classified based of their clinical
phenotype rather than on their genotype. In this way,
a number of normocytic thalassemia deletions
(db- and Acdb) with mild phenotype and elevated HbF
have been classified as HPFH either pancellular (all
RBC containing HbF) or heterocellular (only part of
the cells contains HbF). In fact, these conditions are just
plain b-thalassemia deletions, normocytic because the
HbF cells are larger than HbA cells and mild or inter-
mediate when combined with regular b-thalassemia
defects just because the expression of the c genes is suffi-
cient to maintain a reasonable level of hemoglobin to
avoid regular transfusion. However, in spite of the
reasonable Hb levels, these intermediate conditions
remain associated with some tissue hypoxia because of
the higher oxygen affinity of the HbF tetramer.
© 2013 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2014, 36, 13–19
15
The definition of HPFH should in fact be limited to
those cases that are not associated with a reduced
expression of the b-globin gene such as the point
mutations of the c gene promoters that are summa-
rized in Table 2. Some of these mutations are quite
common polymorphism like the 158 C>T (or Xmn-I,
from the restriction enzyme cutting DNA in this posi-
tion) that enhances the HbF expression only during
erythropoietic stress and have a beneficial effect on
the phenotype of b-thalassemia major or intermedia.
Moreover, the Xmn-I polymorphism may increase the
therapeutic effect of hydroxyurea as well [7] and
therefore, to monitor the effect of the HU treatment
in sickle cell disease, one should always measure the
HbF level before, during and after the treatment, or
even better, one should screen for the presence of the
Xmn-I polymorphism before starting the treatment
because genotype/phenotype correlation is essential
for tailored state of the art treatment of these diseases.
HbF and HbA2 in beta-thalassemia
As mentioned above, elevated HbF can be measured
together with elevated HbA2 in many carriers of
b-thalassemia. The mechanism causing HbF elevation
in carriers of b-thalassemia point mutation defects is
the mild but chronic erythropoietic stress, and the
amount of HbF depends from the presence or absence
of determinants associated with elevated c genes
expression as for instance the common Xmn-I poly-
morphism. Conversely, the cause of the elevated HbA2
levels in the b-thal carrier is mainly the lower amount
of b-chains that changes the b/d ratio in favor of the d.
Then, the number of a chains bound to the b will go
down, while the a chain bound to the d chains will
rise from  2.5% in normal conditions to 4% or more
when only one b-gene is expressed. Therefore, in some
mild b-thalassemia defects with only partial reduction
in b expression moderately elevated or near normal
16 A. AMATO ET AL. | PERSISTENCE OF FETAL HEMOGLOBIN IN ADULT LIFE

Table 2. New and known molecular determinants associated with HPFH

Mutation Gene Ethnic prevalence HbF% in carrier

4 bp ( 225/ 222) Ac African 6–7

202 C>G Gc African 15–20
202 C>T Ac African 3
198 T>C Ac English 4–12
196 C>T Ac Italian/Chinese 21–15
197 C>T* Ac Italian 6
195 C>G Ac Brasilian 4–5
175 T>C Ac Afro American 17–38
175 T>C Gc Afro American, English, Italian 28–29
161 G>A Gc African 1–2
158 C>T Gc African and multiethnic <1 unless stress
158 C>T† (Xmn-I) GGc Afro American 2–5
117 G>A Ac Mediterranean 8–10
114 C>G Gc Australian? 8.6
114 C>T Gc Japanese 11–14
114 C>T Ac Afro American 3–6
114 C>G Gc Australian ? 8.6
13 bp ( 114/ 102) Ac African 30
113 A>G* Ac Italian 6.5
110 A>C Gc Czechoslovakian 1%
Swiss Type X-linked Multiethnic 3–8
Several mutations Enhancer Afro American 4–8
Chrom. translocations 6, 9, 11, 20 Multietnnic 5–8

Adapted from Ida Bianco Silvestroni (reference 9).

*New mutations from the present study.
†
Common polymorphism present on many haplotypes.

HbA2 levels are measured [8]. Variations in HbA2
levels may also result from enhanced or reduced d
gene expression in CIS or TRANS and eventually from
the relatively longer lifespan of red cells with a better
a/cd ratio. Elevated HbA2 levels are usually not
observed in nontransfused homozygous or compound
heterozygous b-thalassemia probably because of down
regulation on both alleles and dyserythropoietic
selection of red cells with elevated HbF expression.
The db- and Acdb-thalassemia
In carriers of these b-thalassemia deletion defects,
the HbF is always elevated and the mechanism that
maintains the HbF expression high after birth is the
elimination of sequences involved in the switch from
HbF to HbA that normally should down regulate the
expression of the c genes after birth. If these ele-
ments, situated between the Ac and the d genes are
missing, then either both or only the Gc gene will
remain partially active depending from the position
of the break point. The HbA2 level will always be
normal in these cases due to the absence of one d
gene and the balanced a/dbc ratio in the red cells
with high HbF.
The genuine HPFH
These are nonpathological conditions that can be very
frequent (polymorphisms), less common or very rare
and are usually mutations associated with the c gene
promoters (Table 2).
The two new HPFH mutations observed in this
study are located, like most of the others, on highly
conserved sequences of the c gene promoters involved
in the regulation and expression of the gene. In
absence of other evident causes that might explain
the elevation of HbF in the individuals we have stud-
ied, we may conclude that the two new mutations
can be included in the list of frequent polymorphisms

© 2013 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2014, 36, 13–19
 A. AMATO ET AL. | PERSISTENCE OF FETAL HEMOGLOBIN IN ADULT LIFE 17

or rare mutation causing HPFH (Table 2) (adapted
and expanded from Ida Bianco Silvestroni) [9].
A number of non-b-globin gene cluster related
determinants, modulating the expression of fetal
hemoglobin have also been described.
Miyoshi et al. [10] reported in a large study from
1988 a X-linked association for those undefined forms
of HPFH called “Swiss type”. In a more recent study
by Gallienne et al. [11], an association between HPFH
and mutations on the KLF1 gene has been suggested.
In a very recent study, Danjou et al. [12] describe
two markers in the BCL11A gene and three in the
HBS1L-MYB intergenic region that could play a role
in the complex mechanism of HbF modulation in
thalassemia patients.
Pregnancy
As mentioned above, HbF levels might become slightly
increased during pregnancy. This can be due to
physiological erythropoietic processes in the mother,
eventually associated with the described polymor-
phisms causing a moderate increase in HbF or HbF
cells. A more pronounced increase could, however, be
due to a serious fetal-maternal transfusion. Fetal cells
from maternal origin can then be distinguished from
the fetal cells coming from the fetus by the presence of
carbonic anhydrase [13].
Erythropoietic stress
Stem cells from adult individuals grown in culture
start producing considerable amounts of HbF just
because of the stress conditions present in culture.
Similarly, any erythropoietic stress in vivo may become
associated with elevated HbF.
For this reason, the accelerated erythropoiesis of the
hypochromic b-thalassemia carrier can be sufficient to
produce more red blood cells up to 6, 7, 10% of HbF, or
no HbF at all, depending from enhancing factors like
the common Xmn-I polymorphism (Table 2).
Likewise, the so-called mild (Senegal, Asia/Saudi)
or severe (Bantu, Cameroon, Benin) HbS haplotypes,
differ from each other also by the presence or absence
of the common Xmn-I polymorphism, enhancing HbF
expression in erythropoietic stress condition.

Table 3. Provisional interpretations of abnormal HbF levels in newborn and adult

Confirm or exclude
HbF% in newborn or adult Possible options with/and Eventual follow up

95–100 in non premature b-thal major, intermedia Family analysis DNA analysis, treatment,
newborn or HPFH Provide information retrospective prevention
90–95 in non premature Possible b-thal minor or Family analysis DNA analysis, counseling in
newborn intermedia Provide information case of couple at risk
100 in macro, normo or Hom. HPFH (RBC↑↑), Family analysis DNA analysis, counseling in
microcytic adult db/db or db/bthal Provide information case of couple at risk
2–10 with A2↑ in microcytic Heterozygous bthal Family/partner analysis DNA analysis, counseling in
adult Provide information case of couple at risk
<1 with normal A2 in Iron deficiency, atypical Family/partner analysis DNA analysis, counseling in
microcytic adult het. bthal, a-thal Provide information case of couple at risk
2–30, normal A2, in non Point mutation HPFH Family analysis DNA analysis, counseling in
microcytic adult Provide information case of couple at risk
>1 in diabetics on HbA1c b-thal minor or artifact CBC and A2 level If b-thal (cascade) analysis of
control CE separation the children
>1 in normo- or microcytic b-thal minor, feto/ CBC and A2 level If b-thal partner analysis
pregnants maternal or maternal Hb differentiation CA* +/ If feto/maternal treatment
Elevated in suspect HPLC HB variant overlapping CE separation, alkaline Molecular analysis.
pattern HbF denaturation test. Counseling in case of risk
Elevated in suspected FA, AA, AML, JMML Exclude hemoglobinopathy Confirm malignancy
malignancy

b-thal, b-thalassemia
*Carbonic anhydrasis (differential test HbF/CA+/ ) [8]

© 2013 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2014, 36, 13–19
18 A. AMATO ET AL. | PERSISTENCE OF FETAL HEMOGLOBIN IN ADULT LIFE

unnecessary concerns and leading to doubtful conclu-

Bone marrow malignancies
A number of severe conditions of the bone marrow
may present with elevated HbF levels. Fanconi anemia
(FA), aplastic anemia (AA), acute myeloid (AML),
lymphatic leukemias (ALL), and juvenile myelomono-
cytic leukemia (JMML) are among those conditions
that present with acquired and variably elevated HbF
levels [14].
Artifacts
The measurement of HbF can be taken manually after
visual estimation on traditional electrophoresis using
the classic alkaline denaturation test [15, 16]. These
methods are still valid but time-consuming and pre-
cise only in the range 1–10%. Today, as described in
materials and methods, the HbF level is measured
automatically in most laboratories using HPLC or Cap-
illary Electrophoresis (CE) [3]. Although quite precise
at all levels, these methods may over or underestimate
the Hb fractions due to integration and overlapping
problems. Some devices in particular, when the HbF
partially overlaps the first HbA1c fraction, may over-
estimate a normal 1% level up to 2–3% causing
sions [17, 18].
Take home message
Our study underlines the importance of a correct
interpretation of an elevated HbF at the basic labora-
tory level and of correct information provided by the
laboratory [19, 20]. Slightly elevated or high HbF
values in adults may indicate thalassemia major or
minor (point mutation or deletion) in need of treat-
ment and prevention or be due to a pregnancy, an
artifact or an harmless HPFH condition with a bene-
ficial effect in the presence of sickle cell disease and
thalassemia major, but it can also indicate a severe
bone marrow malignancy. In case of doubt, the
cause of the elevated HbF should be investigated
at the molecular level by a specialized laboratory
(Table 3).
AC K N OW L E D G E M E N T S
The authors declare to have conducted this study
according to local ethical regulations and to have no
conflicts of interest on the presented matters.

REFERENCES 6. Gottardi E, Losekoot M, Fodde R, Saglio G, hemoglobin regulated dominantly by gene

1. Weatherall DJ, Clegg JB. The Thalassaemia
Syndromes, 4th edn. Oxford: Blackwell Sci-
ence; 2001.
2. Amato A, Grisanti P, Lerone M, Ponzini D,
Di Biagio P, Cappabianca MP, Giordano PC.
Prevention strategies for severe hemoglo-
binopathies in endemic and nonendemic
immigration countries: the Latium exam-
ple. Prenat Diagn 2009;29:1171–4.
3. Van Delft P, Lenters E, Bakker-Verweij M, de
Korte M, Baylan U, Harteveld CL, Giordano
PC. Evaluating five dedicated automatic
devices for hemoglobinopathy diagnostics in
multi ethnic populations. Int J Lab Hematol
2009;31:484–95.
4. Miller SA, Dykes DD, Polesky HF. A simple
salting out procedure for extracting DNA
from human nucleated cells. Nucleic Acid
Res 1988;16:1215.
5. Colosimo A, Guida V, De Luca A, Cappabianca
MP, Bianco I, Palka G, Dallapiccola B. Reliabil-
ity of DHPLC in mutational screening of b-glo-
bin (HBB) alleles. Hum Mutat 2002;19:287–95.
Camaschella C, Bernini LF. Rapid identifi-
cation by denaturing gradient gel electro-
phoresis of mutations in the c-globin gene
promoters in non-deletion type HPFH. Br J
Haematol 1992;80:533–8.
7. Yavarian M, Karimi M, Bakker E, Harteveld
CL, Giordano PC. Response to hydroxyurea
treatment in Iranian transfusion-dependent
beta-thalassemia patients. Haematologica
2004;89:1172–8.
8. Bianco I, Cappabianca MP, Foglietta E, Ler-
one M, Deidda G, Morlupi L, Grisanti P,
Ponzini D, Rinaldi S, Graziani B. Silent tha-
lassemias: genotypes and phenotypes. Hae-
matologica 1997;82:269–80.
9. Bianco Silvestroni I. Le talassemie. Un
problema medico-sociale: ieri ed oggi. Ed.
Istituto Italiano Medicina Sociale – Roma
1998:103.
10. Miyoshi K, Kaneto Y, Kawai H, Ohchi H,
Niki S, Hasegawa K, Shirakami A, Yamano
T. X-linked dominant control of F-cells in
normal adult life: characterization of the
Swiss type as hereditary persistence of fetal
(s) on X chromosome. Blood 1988;72:
1854–60.
11. Gallienne AE, Dreau HM, Schuh A, Old
JM, Henderson S. Ten novel mutations in
the erythroid transcription factor KLF1
gene associated with increased fetal hemo-
globin levels in adults. Haematologica
2012;97:340–3.
12. Danjou F, Anni F, Perseu L, Satta S, Dessı
C, Lai ME, Fortina P, Devoto M, Galanello
R. Genetic modifiers of b-thalassemia and
clinical severity as assessed by age at first
transfusion. Haematologica 2012;97:989–
93.
13. Leers MP, Pelikan HM, Salemans TH,
Giordano PC, Scharnhorst V. Discriminating
fetomaternal hemorrhage from maternal
HbF-containing erythrocytes by dual-parame-
ter flow cytometry. Eur J Obstet Gynecol
Reprod Biol 2007;134:127–9.
14. Jaffe E, Harris NL, Stein H, Wardiman JW.
Pathology & Genetics: Tumours of Haemat-
opoietic and Lymphoid Tissues. Vol. 3;
Lyon: IARC / WHO Classification of

© 2013 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2014, 36, 13–19

Tumours, IARC press; 2001. ISBN 92 832
2411 6.
15. Betke K, Marti HR, Schlicht I. Estimation
of small percentages of foetal haemoglobin.
Nature 1959;184(Suppl.24):1877–8.
16. Stephens AD, Angastiniotis M, Baysal E,
Chan V, Davis B, Fucharoen S, Giordano PC,
Hoyer JD, Mosca A, Wild B. ICSH
recommendations for the measurement
of haemoglobin F. International Council for
The Standardisation of Haematology (ICSH).
Int J Lab Hematol 2012;34:14–20.
© 2013 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2014, 36, 13–19
A. AMATO ET AL. | PERSISTENCE OF FETAL HEMOGLOBIN IN ADULT LIFE
17. Koskinen LK, Lahtela JT, Koivula TA. Fetal
hemoglobin in diabetic patients. Diabetes
Care 1994;17:828–31.
18. Thivolet C, Goujon R, Vassy V, Revol A,
Tourniaire J. Interference of elevated fetal
hemoglobin on HbA1c measurements in
adult type I diabetic patient. Diabetes Care
1991;14:1108.
19. Amato A, Lerone M, Grisanti P, Gizzi L,
Kaufmann JO, Giordano PC. Providing
appropriate genetic information to healthy
carriers of hemoglobinopathy can be a wel-
19
come and safe initiative: the Latium exam-
ple. Genet Test Mol Biomarkers 2012;16:
734–8.
20. Kaufmann JO, Smit JW, Huisman W,
Idema RN, Bakker E, Giordano PC. Basic
hemoglobinopathy diagnostics in Dutch
laboratories; providing an informative test
result. Int J Lab Hematol 2012. doi: 10.
1111/ijlh.12038.