The high oxygen affinity haemoglobin Nantes: a family case description

Introduction

Absolute erythrocytosis is characterised by the expansion of the erythrocyte compartment in peripheral blood and is classified into primary and secondary forms which can be of either congenital or acquired origin¹. Primary erythrocytosis is caused by acquired or inherited mutations leading to expansion of the erythrocyte compartment in the peripheral blood independently of extrinsic factors; it includes polycythemia vera and rare familial variants. Secondary erythrocytosis is caused by a circulating factor, usually erythropoietin, which stimulates erythropoiesis. The increased production of erythropoietin is normally related to a physiological response to hypoxia but may also be caused by an abnormal autonomous erythropoietin production (e.g. erythropoietin-secreting tumour) or an alteration of the oxygen-dependent modulation of erythropoietin synthesis (e.g. increase in the oxygen affinity of haemoglobin [Hb]). Approximately 100 haemoglobin variants with high oxygen affinity have been described in the literature^2,3. Haemoglobin is a tetrameric globular protein composed of four subunits, two alpha (α1, α2) and two beta (β1, β2), which form two identical halves named α1β1 and α2β2. These dimers exist in equilibrium in two conformations with different oxygen affinity: form T (tense) and form R (relaxed) with low and high oxygen affinity, respectively. The transition from the T form to the R form is characterised by a large change in interactions between the subunits α1 and β2, while the α1β1 intra-dimer interface remains almost unaffected^4,5. The α1β2 contact, also named “sliding contact”, interconnects the two dimers (α1β1 and α2β2) and involves the “switch” region, so named because it experiences the largest displacements as a result of the dimer rotation between the R and T forms. Another important contact is in the “hinge” region where the R → T shift is limited to a change in orientation⁶.

The increased oxygen affinity determines erythrocytosis in the peripheral blood and is caused by alterations in haemoglobin structure. The molecular classification of haemoglobins with increased oxygen affinity recognises three different structural modifications concerning the main contacts involved in the transition from the deoxy- to the oxy-state, the 2,3-diphosphoglycerate binding sites and the heme pocket.

In the first group the T structure cannot be achieved. These variants affect the “sliding contact” and the “switch region” and lead to increased oxygen affinity with markedly reduced cooperativity, and to a reduced ability to maintain the T conformation (e.g. Hb Coimbra, Hb Hiroshima, Hb Osler, Hb Nancy, Hb Mckees, Hb Kochi, Hb Cochin-Port Royal). Differently, structural changes that occur within the “hinge region” prevent the formation of the correct bonds stabilising the T structure, but this leads equally to a severe increase in oxygen affinity (e.g. Hb Nantes and Hb Pitié-Salpêtrière)⁷.

Hb Nantes [β34(B16)Val → Leu (HbVar: A Database of Human Hemoglobin Variants and Thalassemias) HBB:c.103G>C (HGVS nomenclature)] was found as a de novo mutation in a 38-year old woman and first reported by Wajcman et al.⁸. Hb Nantes shows increased oxygen affinity and causes a disease with a dominant phenotype. Residue Val β34 contributes to both the α1β1 and α1β2 interfaces⁹; a mutation at this site leads to alteration of the α1β1 contact area and impairs the interactions stabilising the T state.

The mode of transmission of the high oxygen affinity haemoglobin variants is autosomal dominant⁶. The amount of the abnormal haemoglobin present in the red blood cells modulates the haematological consequences of a high oxygen affinity variant.

It is important to diagnose haemoglobin with increased oxygen affinity because, although usually well tolerated in young patients, it frequently leads to thrombotic complications in older patients^10,11 or when it is associated with another factor that increases thrombotic risk. Patients carrying a high oxygen affinity haemoglobin variant are most frequently identified because of an unexplained erythrocytosis with haemoglobin values varying from 16 to 20 g/dL. Routine electrophoretic or chromatographic studies may reveal an abnormal haemoglobin that is further recognised as having increased oxygen affinity. Sometimes routine electrophoretic studies fail to demonstrate the presence of a haemoglobin variant and further evaluations may be necessary, including isoelectric focusing, high performance liquid chromatography (HPLC) and electrophoresis of haemoglobin and globins under various experimental conditions¹². Nevertheless, it should be kept in mind that nowadays the molecular analysis of beta globin genes is much faster and often cheaper than the biochemical approach, and also gives a definitive diagnosis.

Case report

We describe the case of a Caucasian, 17-year old male who was referred to our department because of polyglobulia incidentally identified at the age of 10.

The family history revealed that his maternal grandmother had polyglobulia, considered to have been caused by tabagism, that his mother was affected by polyglobulia (haematocrit up to 52%) and that his sister (9 years old) had polyglobulia which was under investigations. Furthermore his grandfather died of acute myocardial infarction at the age of 53 years and his great-grandfather died of thrombosis at the age of 56. A first degree cousin, studied for Klinefelter’s syndrome, was suffering from polyglobulia accentuated by hormone therapy (haematocrit 62%). Another cousin was reported to be treated with hydroxyurea because of polyglobulia.

Clinical evaluation of the proband found no signs or symptoms referable to polyglobulia. The blood counts showed a gradual increase in haematocrit over time (from 51% in 2006 to 55.9% in 2010) for which periodic phlebotomy was started. Remaining blood counts were in normal ranges (complete blood counts and reticulocytes were enumerated on an Advia 2120; Siemens Healthcare Diagnostics, Tarrytown, NY, USA).

The patient underwent further investigations: cardiac and pulmonary assessments showed no alterations and abdominal ultrasound did not reveals any structural alterations of renal, splenic or hepatic parenchyma. Laboratory tests showed reticulocytosis (117,200/mL), low erythropoietin level (5 mU/L), normal blood gas analysis (erythropoietin concentration was measured using a serum method on an Immulite 2000; Siemens Healthcare Diagnostics, Tarrytown, NY, USA).

The search for the JAK2^V617F gene mutation and BCR/ABL gene rearrangement was negative; molecular analysis of the VHL gene showed no mutations. Cytogenetics from bone marrow blood was normal (46;XY). Bone marrow trephine biopsy was performed and the histological examination showed normal haematopoiesis and erythropoiesis without fibrosis. There was no growth of erythroid colonies from peripheral blood cultures in the absence of erythropoietin and p50 was decreased (venous blood gases and p50 were analysed using RapidPoint 405; Siemens Healthcare Diagnostics, Tarrytown, NY, USA).

Subsequent haemoglobin analysis using cation-exchange a HPLC system, HLC-723G7 (Tosoh Bioscience, Rivoli, Italy), confirmed the presence of the haemoglobin variant, with a retention time partially overlapped that of HbA0 (Figure 1). Haemoglobin samples were also separated on agarose gels at both alkaline and acid pH using a Hydrasis semi-automated agarose gel electrophoresis system (Sebia, Norcross, GA, USA); no abnormal fractions were shown. The haemoglobin variant was later identified as Hb Nantes by molecular analysis.

Figure 1.

HPLC chromatogram showing the Hb variant.

HPLC: high performance liquid chromatography; Hb: haemoglobin.

We performed HPLC, electrophoresis and molecular analysis of haemoglobin from family members on the maternal side who gave their agreement. The subjects studied had elevated haematocrit and haemoglobin levels with reduced p50 (Table 1). We found the same mutation in all subjects already studied for polycythemia and also identified another five subjects, reportedly healthy and not previously studied, carrying the same mutation (Figure 2).

Table I.

Laboratory tests: blood gases, haematocrit and haemoglobin values.

Subject	1	2	3	4	5	6	7	8	9	10
Sex	F	G	M	F	F	M	M	M	F	F
Age	41	8	13	6	7	53	48	15	66	42
Hb g/dL (12.00–16.00)	17.1	15.9	16.7	15.6	16.6	18.3	19.6	19.3	13.8	17.4
Haematocrit % (35.0–45.0)	49.7	51	54.8	48.3	53	56.9	62	60.7	48.1	55.9
pO2* mmHg (24–40)	n.a.	n.m	n.a.	30.4	11.2	20.9	14.9	19.9	12.4	11.1
pCO2* mmHg (41–51)	n.a.	45.2	n.a.	47.5	47.7	53.6	48.9	53	37.7	41.9
p50* mmHg (23.1–28.6)	16.4	n.m.**	n.m.**	18.8	n.m.**	15.2	n.m.**	n.m.**	n.m.**	n.m.**

n.a.: not available; n.m.: not measurable.

^*Parameter assessed on venous blood samples;

^**not measurable because of low level of pO₂.

Figure 2.

Black dot: not tested; white: no mutation found or not tested because of normal haemoglobin or haematocrit levels; cross: mutation found in heterozygosity.

Discussion

This is, to our knowledge, the first description of a family carrier of Hb Nantes, which was first found in a French woman in 2003 and described as a consequence of a de novo mutation. Given that a similar mutation is present in Hb Pitié-Salpêtrière, it could be considered that the gene region codifying for the “hinge region” is instable and prone to spontaneous mutations and so that the same mutation detected in the French woman could have casually occurred in a progenitor of the family we studied. On the other hand, a genetic link with the case previously described cannot be excluded. In fact, the subjects we studied are descendants of people from Sicily which was under French rule during the 13^th century. However, a hypothetical connection would call into question the real condition of a de novo mutation in the first case described.

In our opinion, the description of this family case is instructive as it highlights the importance both of the proper diagnostic process and of collecting medical records of the proband and his family.

Haemoglobinopathies with altered oxygen affinity are rare diseases, but can easily be suspected from the findings of simple laboratory investigations such as the measurement of p50. In our opinion, this parameter should be routinely investigated in patients with suspected polycythemia vera, especially in young patients in whom polycythemia vera is considered statistically unlikely, or at least after it has been excluded by clinical and laboratory criteria

HPLC of haemoglobin is feasible in the large majority of chemical-clinical analysis laboratories and reference centres are also easily accessible for molecular analysis for a precise definition of abnormal haemoglobin. The precise definition enables investigations to be widened within the family, facilitating the identification of people with the same characteristics, as in our experience.

Some studies suggest that reducing the haematocrit by phlebotomy/venesection could be of benefit in patients with erythrocytosis^13,14, but there is little evidence in the literature guiding the management of patients with markedly persistent erythrocytosis due to congenital defects. In the family we present here, the family history revealed a significant incidence of cardiovascular events in subjects who had not undergone phlebotomy and/or antiplatelet therapy. In the light of the diagnosis and medical records of the family, we have established that phlebotomy will be continued for subjects with haematocrit values above 54%. This threshold, which is also suggested by McCullin¹, was set arbitrarily in the belief that values used in polycythemia vera (<45%) could cause a state of tissue hypoxia. Moreover, in our opinion, an antiplatelet agent should be added from adult age, when the risk of cardiovascular events is increased. We feel that it is appropriate to describe these cases, both because of their rarity and in order to share the therapeutic strategy.