Skip to main content
Original Articles

Erythrocyte Membrane Proteins in Healthy Saudis and Patients with Hereditary Spherocytosis and Hereditary Elliptocytosis

Abstract

BACKGROUND:

Little is known about hereditary spherocytosis (HS) and hereditary elliptocytosis (HE) in the native population of Saudi Arabia, even though these conditions are seemingly common. The purpose of this study was to ascertain the protein make-up of the red cell membrane in healthy Saudis and in patients with HS and HE.

PATIENTS AND METHODS:

Eighteen healthy Saudi subjects (13 males and 5 females), 11 patients with HS (6 males and 5 females) and 11 patients with HE (7 males and 4 females) were studied. All normal controls and patients underwent SDS-PAGE red cell membrane protein analysis in duplicate and the stained protein bands were identified and quantitated by densitometry.

RESULTS:

In normal, healthy Saudis, the mean values for seven membrane proteins ( αspectrin, β spectrin, ankyrin, band 3, protein 4.1, protein 4.2, and actin) were similar to those published for normal, healthy Americans. Of the eleven cases with HS, 7 (64%) demonstrated detectable protein abnormalities while 4 (36%) were apparently normal. The electrophoretic patterns of membrane proteins in Saudis with HS differed from those of patients with HS in other parts of the world. Of the 11 cases of HE, 7 (64%) displayed abnormalities while 4 (36%) were normal.

CONCLUSION:

The electrophoretic pattern of the main proteins in the membranes of red blood cells in healthy Saudis is similar to that reported from the USA. However, significant differences exist in the electrophoretic patterns between Saudi patients with HS and patients from other parts of the world.

Introduction

Of all freely floating cells in the human body, the erythrocyte is unique in having the capacity for assuming innumerable shapes in response to pathological conditions. Whether these conditions are congenital or acquired and primarily affecting erythrocytes or not, all are mediated through changes in the structure of the erythrocyticmembrane.1,2

Although the exact incidence is unknown, a substantial number of native Saudis have erythrocytic membrane abnormalities. Notable among these are hereditary elliptocytosis (HE), hereditary spherocytosis (HS) and hereditary pyropoikilocytosis (HPP). Although seemingly common, little is known about these diseases, including the specific erythrocyte membrane abnormalities related to these conditions.

The current study was undertaken to partially remedy the these gaps in knowledge. We included normal, healthy Saudis without erythrocytic membrane abnormalities in our study to establish baseline values for membrane protein composition for comparison with the HS and HE patients. The inclusion of normal control subjects also allowed us to compare values for normal Saudis with published values for other ethnic groups.

MATERIALS AND METHODS

We studied 11 patients diagnosed with HS, including 5 males and 6 females (mean age 30 years; range 4 to 75 years) and another 11 diagnosed with HE, including 7 males and 4 females (mean age 28 years, range 3 to 90 years). All were ethnic Saudis. Eighteen apparently normal Saudis, including 13 males and 5 females (mean age 25 years, range 2 to 55 years), served as controls.

The 11 patients with HS had a history of jaundice and had increased serum bilirubin. Splenomegaly was found in all except one who had undergone therapeutic splenectomy. HS was diagnosed on a clinical and laboratory basis. On examination of the blood, all had spherocytes in their peripheral blood and reticulocytes were increased. The Direct Coombs’ Test was negative and the erythrocyte osmotic fragility test (EOFT) was consistent with HS.

A majority of the 11 patients with HE were asymptomatic and were discovered because of blood examination for other reasons. The others were identified on investigation of the family members of the propositi. Only one patient was symptomatic with jaundice, splenomegaly, serum hyperbilirubinaemia and reticulocytosis. No patient had apositive direct Coombs’ test or was positive for haemoglobin S (Hb S) except for one patient who had sickle cell trait.

All blood samples underwent conventional analysis, including automated hematological analysis, an assessment of erythrocyte morphology on a stained blood smear, a solubility test for Hb S and a screening test for glucoses-phosphate dehydrogenase. All samples were analyzed by the sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) method, as described in Laboratory Haematology3 based on the original description of Fairbanks etal4. The dried gels were scanned in a photodensitometer and each relevant band identified and quantitated. The electrophoretic patterns of the normal individuals were compared with the electrophoretic patterns reported for normal individuals in the medical literature and were used as a baseline to compare with the electrophoretic strips of patients with HS and HE.

RESULTS

Red Blood Cell Morphology

All patients with HS had demonstrable spherocytes in their peripheral blood. The mean spherocyte population was 22.4% (range, 9.7% to 58.5%). Three of 11 cases had occasional pincered cells, which were spherocytes with a small appendage constricted at the attachment with the main cell. One case had spherocytes with a few (2 to 8) tapering spicules protruding from their surfaces. One case exhibited a few acanthocytes and Howell-Jolly bodies that were indicative of the fact that the patient had undergone a therapeutic splenectomy.

On morphological examination of blood smears, all patients with HE had elliptocytes/ovalocytes in excess- of 75%. The mean number of elliptocytes/ovalocytes was 93.3% and the range was between 78.4% and 99.7%. This large percentage was considered important so that cases of isolated iron deficiency or thalassaemia (both relatively common conditions in Saudi Arabia) should not be mistaken for HE. Elliptocytosis as low as 10% has been considered compatible with the diagnosis of HE.5 In ten cases, elliptocytes were the predominant cells, while in one case ovalocytes were predominant. One case showed some budding or fragmentation of red blood cells, but it lacked other features of HPP. According to conventional classification, all of our cases fell into the category of common HE except for one case that could be classed as homozygous common HE on morphological grounds.

In the normal controls, RBC morphology was as expected, i.e., normocytic normochronic and with no significant anisocytosis or poikilocytosis.

SDS-PAGE Characterization of the Proteins of the Erythrocyte Membrane

Analysis of the normal controls revealed the following bands of interest, listed by decreasing molecular weight: αspectrin, βspectrin, ankyrin, band 3, band 4.1, band 4.2and actin (Table 1). Except for ankyrin, which was sometimes evident as a single band and at other times as multiple bands, and band 3, which was always a broad diffuse band, the other bands appeared as single discrete zones. The percentages of the various red cell membrane proteins in healthy Saudis and normal subjects from the USA appear to be broadly similar to each other (Table 1). The difference in the percentages of those proteins that occurred in a larger amount (e.g., band 3) were smaller than the differences that occurred in a smaller amount (e.g. ankyrin).

Table 1. Erythrocyte membrane proteins of normal subjects from the USA and normalSaudis and Saudi patients with hereditary spherocytosis (HS)(n=11) and hereditaryelliptocytosis (HE) (n=11).

Table 1.

Of the 11 patients with HS, protein deficiencies were found in 7 patients (64%), but no protein abnormalities could be ascertained in 4 patients (36%) (Table 2). In decreasing frequency the occurrence of the deficiencies were ankyrin deficiency in 4 cases (36%), band 3 deficiency in 4 cases (36%), protein 4.1 deficiency in 3 cases (27%), protein 4.2 deficiency in 2 cases (18%), and βspectrin deficiency in 1 case (9%). Of the 11 cases of HE, 7 (64%) displayed abnormalities while 4 (36%) were normal.

Table 2. Erythrocyte membrane protein abnormalities in 22 Saudi patients.

Table 2.

α and β Spectrin

In normal Saudi subjects, a spectrin was found to have a mean of 11.5% (range 6.9-17.3%) (Table 1). In patients with HS the mean was 9.0% (range 5.9% to 14.7%) and in patients with HE the mean was 10.4% (range 6.6% to 15.7%). There was a statistically significant difference between the mean of the normal controls and patients with HS (p<0.05), but there was no statistically significant difference between the mean of the normal controls and patients with HE.

The mean value for βspectrin was 14.1% (range 10.2% to 22.2%) in normal controls, 11.3% (range 5.2% to 16.3%) in patients with HS, and 13.8% (range 6.9% to 19.8%) in patients with HE (Table 1). Differences with normal controls were not statistically significant. Only one case of HS exhibited a deficiency of βspectrin in association with protein 4.1 deficiency. Among cases of HE, one case demonstrated a deficiency of a andβspectrin and one case a deficiency of βspectrin in association with two abnormal bands in the spectrin region. The abnormal bands were smaller in molecular size and were thought to be dimers of α and β spectrin.

One case of HE exhibited two abnormal bands in the spectrin region, which was similar to the bands described above but showed no associated deficiency of α or β spectrin.

Ankyrin

In healthy normal subjects, the mean ankyrin level was 2.7% (range 1.5-5.3%) (Table 1). In patients with HS the mean was 2.7.% (range 0.0-6.3%) and in patients with HE the mean was 3.9% (range 0.0-10.6%) (Table 1). The differences in the means of normal controls and patients with HS or HE were not statistically significant.

In patients with HS, we had one case of isolated ankyrin deficiency and three cases where ankyrin deficiency was associated with band 3 or protein 4.2 deficiency either alone or in combination (Table 2). Ankyrin was unexpectedly increased in one case, for which no explanation could be ascertained.

Ankyrin deficiency has not been described in cases with HE and none of our patients with HE demonstrated this defect. However, two cases exhibited an increase in the percentage of ankyrin, both of which had extra-abnormal bands in the spectrin region. It is quite possible that abnormal spectrin migrating at the same position as ankyrin helped to boost the concentration of the latter.

Band 3 protein

The mean level of band 3 in healthy Saudis was 28.4% (range of 21.8% to 32.8%) (Table 1). For HS patients the levels were 22.5% (range 13.8% to 32.9%), and for HE patients 26.1% (range 19.1% to 32.8%) (Table 1). The difference between the means of the normal subjects and patients with HS was statistically significant (p<0.05). The differences between the means of normal controls and patients with HE was not statistically significant. There was no case of HS with an isolated deficiency of band 3, but we did have cases where band 3 deficiency occurred in combination with ankyrin deficiency, protein 4.2 deficiency and protein 4.1 deficiency in different combinations (Table 2).

Protein 4.1

The mean level of this protein in healthy normal subjects was 3.2% (range 2.0% to 4.8%), in patients with HS 2.8% (range 0.0% to 3.5%) and in patients with HE 1.5% (range 0.0% to 3.3%) (Table 1). The difference between the means for the normal subjects and patients with HS (p<0.05) andProtein 4.1 deficiency has only rarely been described as a cause of HS, but we had three cases with this deficiency (Table 2). In one case it was associated with βspectrin deficiency, in one with band 3 deficiency and in one with an increased percentage of ankyrin. It could not be determined why the concentration of ankyrin was increased. There were four cases (36%) with isolated protein 4.1 deficiency among our patients with HE (Table 2). In all four cases protein 4.1 was totally absent, indicating that these were, most probably, genetically homozygous.

Protein 4.2

The mean level of this protein in healthy, normal subjects was 2.4% (range 1.3% to 3.7%) (Table 1). In patients with HS the values were 2.7% (range 0.0% to 4.5%) and in patients with HE 3.1% (range 1.0% to 5.4%) (Table 1). The difference between the means for the normal controls and patients with HE was statistically significant (p<0.05), but not between the normal subjects and HS. Protein 4.2 is thought to help anchor band 3 to βspectrin and its deficiency can give rise to HS. Two cases among our patients with HS demonstrated a deficiency of this protein; one in association with ankyrin deficiency and one in combination with ankyrin and band 3 deficiencies (Table 2). No defects were noted in the cases with HE.

Actin

The mean level of actin in healthy, normal subjects was 7.8% (range 4.1% to 12.1%). Values for patients with HS were 8.7% (range 5.5% to 12.5%) and for patients with HE 8.3% (range 4.6% to 12.0%) (Table 1). As expected there were no statistically significant differences between the means of the three groups. No abnormalities of actin were observed between normal controls and patients with HS or HE.

DISCUSSION

Hereditary spherocytosis

Hereditary spherocytosis (HS), the most common nonimmune hemolytic anaemia in people of North European extraction, is reported to have the following major biochemical abnormalities:2 combined spectrin and ankyrin deficiency, isolated spectrin deficiency, band 3 deficiency, and isolated protein 4.2 deficiency. However, there is a wide difference in the occurrence of reported abnormalities from different countries (Table 3). Different defects appear to cluster in different regions. We found that it was surprisingly difficult to find published normal ranges for various red blood cell membrane proteins from different countries.

Table 3. A comparative table of erythrocyte membrane protein abnormalities reported fromdifferent countries in patients with HS.

Table 3.

Ankyrin deficiency is a well known cause of HS.2 It can occur in isolation or in combination with other protein abnormalities. It is interesting to note that ankyrin deficiency is the major abnormality in HS among Saudis and is similar to cases from the USA.2 But, whereas in the USA. it primarily occurs in combination with spectrin deficiency, in the Saudis in our study it occurred in combination with band 3 deficiency (one case) and with band 3 and protein 4.2deficiencies (two cases). In only one case did it occur as an isolated deficiency (Table 2). In one case of HS, the percentage of ankyrin was surprisingly increased. No explanation could be found for this unexpected finding.

Four cases (36%) of HS in our patients exhibited band 3 deficiency. In the peripheral blood, the four cases had a spherocyte population of 58.5%, 11.5%, 23% and 40.5% respectively, but only one demonstrated a definite pincered cell as has been reported from the USA.9 One case, however, demonstrated a few spiculated cells. No case displayed an isolated deficiency of band 3. In one case, band 3 deficiency occurred in association with ankyrin deficiency, in two cases with ankyrin deficiency and protein 4.2 deficiency and in one with a deficiency of protein 4.1. Some authors speculate that protein 4.1 is also attached to band 3 in addition to the junctional complexes and this may explain the concomitant deficiency of protein 4.1 in this patient.

Two cases of HS (18%) among our patients demonstrated a deficiency of protein 4.2, but not in isolation. In one case, protein 4.2 deficiency occurred in association with ankyrin deficiency and in the other with ankyrin and band 3 deficiencies. This is an interesting finding as protein 4.2 deficiency occurs in a minority of cases from the USA (in 3% only, Table 3), but in up to 39% of those reported from Japan, either alone or in combination with other proteins (Table 3). Protein 4.2 deficiency appears to occupy a position of intermediate importance among Saudis.

Three cases (27%) exhibited a deficiency of protein 4.1, including one in association with βspectrin deficiency, one in association with band 3 deficiency, and another as an isolated deficiency. The occurrence of protein 4.1 deficiency is interesting in the context of HS as it is mostly associated with HE. Of the red cell membrane proteins, protein 4.1 is unique in that it is associated with band 3, protein 4.2 and β spectrin to form a “vertical” association, and with spectrin, protein p55, actin and glycophorin C to form a “horizontal” junctional complex.2 Therefore, protein 4.1 deficiency could theoretically occur in cases of HS if its “vertical” association with band 3 or protein 4.2 was primarily affected. Surprisingly only one case (9%) displayed spectrin deficiency which occurred in association with protein 4.1 deficiency. Spectrin deficiency is by far the most important cause of HS among westerners, but appears not have the same importance in Saudis.

Because the number of cases was small, general conclusions cannot be drawn with absolute confidence from our data. It is interesting to note that in all of our cases of HS where protein abnormalities could be demonstrated (64%), the protein profile differed from cases reported in the medical literature (Table 3). In American patients, a combined deficiency of spectrin and ankyrin is the commonest abnormality (30-40%) followed closely by isolated spectrin deficiency (30%). In our study, only 1 out of 11 (9%) had ademonstrable deficiency of spectrin but 4 out of 11 (36%) had ankyrin deficiency, which appeared to be one of the two most important etiological factors with band 3. In addition, ankyrin was the only protein that occurred as an isolated deficiency. All others were combinations of more than one protein.

Combined deficiencies consisted mainly of variable combinations of ankyrin, band 3, and proteins 4.2 and 4.1, and in only one case included spectrin. Combined deficiencies of band 3 with ankyrin or protein 4.2 have not been generally reported from the West except from Denmark.10 In contrast to Western patients, only one of our cases displayed pincered cells in the peripheral blood.

Another interesting phenomenon was the deficiency of protein 4.1 in 3 of 11 cases, a frequency that precludes a random technical error. We can only conjecture on why this deficiency occurred at this frequency since the precise location of this protein in the red cell membrane has not been definitely established. The suggestion that protein 4.1 is also attached to band 3 apart from ankyrin, may provide one explanation. Any deficiency of band 3 should inevitably lead to the deficiency of the other. However, other proteins are also attached to band 3, (e.g. ankyrin and protein 4.2.) and deficiencies of these proteins do not always accompany band 3 deficiency. Since cases of HE due to isolated protein 4.1 deficiency have been reported, our patients could have hadtwo separate abnormalities: protein 4.1 deficiency with band 3 or ankyrin deficiency, with the spherocytic effect of band 3 and ankyrin deficiency completely masking the elliptocytotic effect of protein 4.1 deficiency. Genetic analysis might resolve this question. Another explanation might be that deficiency of protein 4.1, a genuine but rare cause of HS, 10 is more common among Saudis than in other populations for whom data is available.

Hereditary elliptocytosis

Although hereditary elliptocytosis (HE) appears to be fairly common in the Kingdom, judging by the number of cases incidentally discovered in hematology laboratories, the exact incidence is unknown. Niazi reported, based on red blood cell morphology alone, an incidence of 3% in newborns at King Fahad National Guard Hospital, Riyadh.11 In the cases we studied, there were no specific clinical markers for HE. All our propositi were found incidentally, while others were found when family studies were carried out for the probands. Five cases had a normal hemoglobin level, but six other cases were anemic. The anemia ranged from mild to moderate, and it was evident that complicating factors were present in the latter. Two cases had red cell indices compatible with an associated iron deficiency and three cases appeared to have coexisting thalassaemia trait. Red cell indices for one case were not available as they were rejected by the hematology analyzer because of marked poikilocytosis.

In the medical literature, the membrane protein abnormalities reported for HE include qualitative abnormalities of α-spectrin and more rarely β-spectrin, protein 4.1 deficiency, glycophorin C deficiency (Leach phenotype), and mutant band 3 (the latter specific for South-East Asian ovalocytosis). Protein defects in the majority of cases of HE are due to qualitative abnormalities of spectrin, which are theoretically detectable by red cell membrane protein electrophoresis. In our study, three cases (27%) exhibited abnormal bands while four displayed normal protein electrophoretograms. The abnormal bands in one case included one located just beyond the band for αspectrin and one just beyond the band for βspectrin. As both of these bands had lower molecular weights than α or βspectrin, they might be smaller defective multimers of α and βspectrin that are incapable of forming stable tetramers and could have been dimers. A quantitative defect was found in four cases (36%), which consisted of a total absence of protein 4.1. Protein 4.1 is a commonly reported abnormality associated with HE.2 A total absence of protein 4.1 is suggestive of the homozygous state. Two cases with HE were found to have an increased percentage of ankyrin. In both of these cases there were extra-abnormal bands in the spectrin region. It is quite possible that an abnormal spectrin was migrating at the same position as ankyrin and thereby augmenting the concentration of the latter.

In conclusion, the electrophoretic pattern of the main proteins in the red blood cell membranes of healthy Saudis is similar to that reported for healthy, normal Americans. Significant differences exist in the electrophoretic patterns of Saudi patients with HS compared with HS patients from other parts of the world.

ARTICLE REFERENCES:

  • 1. Palek J, Jarolim P. "Clinical Expression and Laboratory Detection of Red Cell Membrane Proteins" . Semin Hematol. 1993; 30:249–83.

    Google Scholar
  • 2. Tse TW, Lux SE. "Red blood cell membrane disorders" . Br J Haematol. 1999; 104:2–13.

    Google Scholar
  • 3. Laboratory Haematology, An account of Laboratory Techniques. Ed. Chanarin I.. Churchill Livingstone, 1989:98–100.

    Google Scholar
  • 4. Fairbanks G, Steck TL, Wallach DFH. "Electrophoretic analysis of the major polypeptides of the human erythrocyte membrane" . Biochemistry. 1971; 10:2606–17.

    Google Scholar
  • 5. Palek J. "Hereditary elliptocytosis and related disorders" . Clin Hematol. 1985; 14:45–8.

    Google Scholar
  • 6. Palek J, Jarolim P. "Clinical expression and laboratory detection of red cell membrane protein mutations" . Semin Hematol. 1993; 30:249–83.

    Google Scholar
  • 7. Saad STO, Costa FF. "Red cell membrane abnormalities in hereditary spherocytosis (HS) patients from Brazil" . Blood. 1993; 82:459–462.

    Google Scholar
  • 8. Takafumi I, Akio K, Axumi Tet al.. "Uniquely higher incidence of isolated or combined deficiency of band 3 and/or band 4.2 as the pathogenesis of autosomal dominantly inherited hereditary spherocytosis in the Japanese population" . Inter J Hemat. 1994; 60:227–238.

    Google Scholar
  • 9. Saad STO, Liu SC, Golan Det al.. "Mechanism underlying band 3 deficiency in a subset of patients with hereditary spherocytosis" . Blood. 1991; 78:81(abstr).

    Google Scholar
  • 10. Orntoft TF, Clausen N. "Hereditary Spherocytosis: diagnostic and anaemia-associated aberrations of ghost proteins" . Scand J Clin Lab Invest. 1994; 54:95–103.

    Google Scholar
  • 11. Niazi GA. "Hematological profile of Saudi newborns" . Saudi Med J. 1994; 15(3):243–9.

    Google Scholar