Erythrocyte membrane protein defects in hereditary spherocytosis patients in Turkish population

Abstract

Hereditary spherocytosis (HS) is a congenital hemolytic anemia which is characterized by spherocytes in peripheral blood and increased osmotic fragility test. The disease is caused by defects in red cell membrane cytoskeleton. In this study, we investigated erythrocyte membrane protein defects in 50 Turkish HS patients and 42 controls. We used sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE) to identify the protein defects causing HS. The patients were from 27 families (39 kindred and 11 unrelated patients). They were aged between 6 months and 53 years and the mean age was 18.75 (±14.70) years. Protein deficiencies related to HS were demonstrated in 42% of study group. There was not any statistically significant relation between the protein deficiency and hemoglobin levels. Isolated or combined spectrin deficiency was the most common protein abnormality among our patients. Spectrin deficiency was detected in 22% of cases (11/50), ankyrin deficiency in 8% (4/50), protein 4.2 deficiency in 8% (4/50), combined spectrin and protein 4.2 deficiency in 2% (1/50), combined spectrin and ankyrin deficiency in 2% (1/50). Fifty-eight percent of cases (29/50) showed normal protein contents.

Keywords:

• Hereditary spherocytosis
• Erythrocyte membrane protein
• Polyacrylamide gel
• Electrophoresis

Introduction

Hereditary spherocytosis (HS) is a worldwide common inherited hemolytic anemia with phenotypic and genotypic heterogeneity caused by structural defects in red cell cytoskeletal proteins.¹ It is the most common inherited anemia in individuals of northern European population.² HS is clinically characterized by anemia, jaundice, splenomegaly, spheroidal red cells that are trapped in the spleen, and increased osmotic fragility of erythrocytes. Clinical severity is variable; most of the patients have compensated hemolytic anemia and do not need transfusions, whereas a small minority require erythrocyte transfusions as a result of severe hemolysis. Almost 75% of the patients have an autosomal dominant pattern and 25% have autosomal recessive forms or de novo mutations.^3–6 HS is caused by defects in the erythrocyte membrane proteins involved in the vertical interactions between the membrane skeleton and the lipid bilayer. HS patients have either single protein deficiency of spectrin, ankyrin, band 3, protein B 4.1, and protein B 4.2 or combined deficiencies of these proteins. Isolated spectrin defect or combined spectrin and ankyrin defects are the most detected membrane protein alterations in HS.^7,8 The diagnosis of HS is based upon clinical and laboratory findings such as anemia, jaundice, splenomegaly, presence of spherocytes in peripheral blood smear, and increased mean corpuscular hemoglobin (Hb) concentration combined with a positive osmotic fragility test.⁹

In this study, we report red cell membrane protein alterations in HS in Turkish population detected by densitometric tracing of sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE).

Patients and Method

Fifty patients were diagnosed as HS, 39 belonging to 16 families plus 11 unrelated cases included in the study. Twenty-one patients were male and 29 were female. They were aged between 6 months and 53 years. The mean age was 18.75 (±14.70) years. The patients were diagnosed as HS by presence of family history, clinical findings (jaundice, splenomegaly), and laboratory tests including complete blood count, high mean corpuscular Hb concentration, presence of spherocytes on peripheral blood smear, reticulocyte count, increased osmotic fragility test with 24 hour of incubation, increased unconjugated bilirubin level, decreased haptoglobin level and negative direct antiglobulin test. Also Hb electrophoresis, pyruvate kinase, and G-6PD levels were studied to exclude other anomalies that could cause hemolytic anemia. Serum iron, ferritin, and total iron binding capacity were also studied for iron deficiency anemia. Hb levels of the patients were classified as <7 g/dl, 7–11 g/dl and >11 g/dl to evaluate the relation between Hb level and protein deficiency. None of the patients had transfusion within 3 months preceding the study. Twenty-four patients (24/50) had been splenectomized before the study. Forty-two healthy volunteer individuals aged between 5 and 41 years were used as control group. Spectrin, ankyrin (isoform protein 2.1+2.2), band 3, protein 4.1, and protein 4.2 were detected by SDS–PAGE electrophoresis. Normal values of the erythrocyte membrane proteins spectrin, ankyrin, band 3, protein 4.1, and protein 4.2 were obtained from control group and the amounts of the patients’ red cell proteins were compared to these normal values. Levels of the patients’ proteins less than the mean−2SD were considered as red cell protein deficiency.

Analysis of red cell membrane proteins

Peripheral venous blood samples (10 ml) drawn from the patients were collected in EDTA tubes. We used method of Dodge et al.¹⁰ to prepare erythrocyte ghosts. The term ghost is used in preference to membranes for description of the discoid bodies obtained after removal of the Hb from red cells.¹⁰ The determination of red cell count permitted expression of results in units per cell (or ghost).¹⁰ For red cell counts, an electronic blood cell counter (Beckman Coulter) was used. Red cell ghosts were prepared within the first 24 hours the samples were collected. Tubes containing blood were mixed by gentle shaking before centrifugation. Then they were centrifuged at 2000g for 20 minutes, plasma and buffy coat were removed, and the erythrocytes sedimented at the bottom of the tubes were washed with isotonic phosphate buffer (310 mOsm, pH 7.4, NaH₂PO₄.2H₂O and NaHPO₄) for three times. After each wash, samples were centrifuged at 2000g for 20 minutes and the supernatant was decanted. After this process, 2 ml of washed red cells was put in tubes and 28 ml of 20 mOsm phosphate buffer (pH 7.4, NaH₂PO₄.2H₂O and NaHPO₄) were added to have a total volume of 30 ml. These samples were centrifuged at 4°C at 11 000g for 40 minutes, and supernatant was decanted gently. This was repeated for four times and after the fourth wash, the supernatant was colorless or very light pink. After this preparation, we obtained ghosts we needed to analyze the red cell membrane proteins.

For red cell membrane protein solubilization 2% sodium dodecyl sulfate (SDS), 2% 2-mercapto-etanol and 8 mol/l urea mixture was used. This solution was mixed with erythrocyte membrane suspension with 1 : 1 ratio and this solution was allowed to stand in hot water bath for 5 minutes, so material for SDS–PAGE was ready. The method we used for electrophoresis with some modifications was that of Fairbanks.¹¹ SDS is an anionic detergent that is used as a solvent to separate red cell membrane proteins from lipid bilayer. To prepare the final concentration for electrophoresis, we used the following as described in Fairbanks¹¹ and discontinuous buffer system of Laemmli’s¹² method: acrylamide solution (acrylamide 40 g, N,N-metilenbisacrylamide 1.5 g, distilled water added to have a 100 ml volume), buffer solution (1.0 mol/l Tris-HCl with pH 8, 2 mol/l sodium asetate, 0.2 mol/l EDTA), 20% sodium dodecyl sulfate, ammonium persulfate (15 mg/ml), TEMED (30 ml/20 ml), dye concentration (coomassie brillant blue: 1.25 g comassie brillant blue, 50% 454 ml ethanol, 46 ml acetic acid, 0.5% bromofenol blue). For preparation of gels, the following were used: acrylamide solution 4 ml, buffer 3 ml, 20% SDS 1 ml, H₂O 10 ml, ammonium persulfate 2 ml, and TEMED 30 μl. One-hundred microliters from mixture of 100 μl ghost, 10 μl bromophenol, and one drop of glycerol was discharged on gels. After few minutes needed for absorption of the material by gels, electrophoresis was performed with a current at 6 mA/tube (Bio-Rad). Running time for this was approximately 2.5–3 hours. After these gels were incubated with Coomassie brillant blue at room temperature over night, and then kept in 7.5% acetic acid, absorption peak heights of gels at 545 nm were scanned (Cliniscan; Helena Laboratories, Beaumont, TX, USA). Each patient’s sample was loaded for three times on each gel and average densitometric reading was considered as the final results. Results were quantitated as ratios to reference ranges calculated from control group.

Statistical analysis

For statistical analysis, Statistical Package for Social Sciences (SPSS, Chicago, IL, USA) for Windows 10.0 program was used. To compare the results, Mann–Whitney U test was used. Continuous variables were analyzed as mean±SD. District variables were analyzed as percentages. For all statistical tests, P<0.05 was considered statistically significant.

The study was approved by Istanbul University Cerrahpasa Medical Faculty Ethics Committee and written informed consent was obtained from adult patients, parents of children and also from control group.

Results

Protein deficiency was observed in 42% of cases, 21 patients (12 female, 9 male) from 14 families (21/50). There was no significant (P>0.05) relation between type or presence of detectable deficiency and gender. We detected spectrin deficiency in 22% of cases (11/50), ankyrin deficiency in 8% (4/50), protein B 4.2 deficiency in 8% (4/50), combined spectrin and ankyrin deficiency in 2% (1/50), combined spectrin and protein B 4.2 deficiency in 2% (1/50) (Table 1). In five families having protein deficiencies, we detected different types of proteins in the members of the families (Table 2). We did not observe any protein deficiency in 58% of the cases (29/50). Eight of the patients (8/50) having protein deficiency (spectrin deficiency in four cases, ankyrin deficiency in one case, protein 4.2 deficiency in two cases, combined spectrin and protein 4.2 deficiency in one case) had been splenectomized at different ages and there was no statistically significant (P>0.05) relation between splenectomy and red cell protein deficiency. These patients underwent splenectomy between ages 6.5–16 years (mean±SD: 8.7±2.71 years). We also evaluated Hb levels of these 50 patients according to presence and type of protein deficiency (Table 3). There was not any significant relation between protein deficiency and Hb levels (P>0.05).

Table 1. Red cell membrane protein deficiencies

Defective protein	n ( = 21)	% ( = 42)
Spectrin	11	22
Ankyrin	4	8
Band 4.2	4	8
Spectrin+Band 4.2	1	2
Spectrin+ankyrin	1	2

Table 2. Difference in the type of deficient proteins in the same family members

Family	Defective protein
Child	Spectrin
Child	Ankyrin
Child	Ankyrin
Child	Spectrin+band 4.2
Child	Ankyrin
Child	Spectrin
Father	Band 4.2
Child	Band 4.2
Child	Spectrin
Father	Bband 4.2
Child	Spectrin+ankyrin

Table 3. Anemia and red cell membrane protein deficiencies

		Protein deficiency		Protein deficiency
		(+)		(−)
Hb level (gr/dl)	n	n	type	n
<7	8	2	1 spectrin	6
			1 combined spectrin and protein band 4.2
7–11	32	13	7 spectrin	19
			3 ankyrin
			2 protein band 4.2
			1 combined spectrin and ankyrin
11>	10	6	3 spectrin	4
			1 ankyrin
			2 protein band 4.2

Discussion

In our study, we aimed to identify the distribution of erythrocyte protein defects in Turkish HS patients and correlate the deficiency with Hb levels and splenectomy. Until the 1980s, spectrin deficiency was the best known protein abnormality reported in HS, but later many other protein deficiencies such as ankyrin, band 3, and protein 4.2 (alone or combined) responsible of HS were described.^13,14 We detected protein deficiency in 42% of our patients. We could not detect any protein deficiency in 58%. The proportion of undefined patients was higher in our series than most of the other studies reported. There are many different ratios reported in literature. Lee et al.¹⁵ reported that protein deficiencies were not detected in 33.3% of Korean HS patients. In a Brazilian study, this ratio was given as 30%,¹⁶ in an Italian study less than 10%.¹⁷ In Jarolim’s study, it is reported that 15% of autosomal dominant HS had no quantitative erythrocyte membrane protein defects.¹⁸ Our result was similar with a study from France which gave the deficiency ratio as 45%.¹⁹ As it has been mentioned before in many studies, by SDS–PAGE technique some HS patients remain unclassified for the type of protein deficiency.^15,18,20 In Savvides’ study, it is noted that radioimmunoassays are more accurate than SDS gels for quantitating spectrin and ankyrin in erythrocyte membranes.²¹ It is told that for spectrin and ankyrin, even when SDS gel electrophoresis are performed under optimal conditions, radioimmunoassays are more accurate.^21,22 The SDS–PAGE technique may not be sensitive enough to detect minimal protein deficiencies and this may explain the variation between ratios given in different studies. In our study, in five families we detected different types of protein defects in the members of the families. Similar difference in the members of the same family was reported in Brazilian study, mother having spectrin deficiency whereas child having combined spectrin and ankyrin deficiency.¹⁶ We think that the difference between the members of the same families can be explained with the insufficiency of SDS–PAGE method for quantitating minor deficiencies as we mentioned before. It is also well known that there is a wide divergence in the frequency of the type of the deficient protein in different populations related to ethnic origins. Isolated or combined spectrin deficiencies were the most common protein abnormalities detected in our study. Twenty-two percent (11/50) of cases had isolated spectrin deficiency, 2% (1/50) had combined spectrin and ankyrin deficiency, and 2% (1/50) had combined spectrin and protein B 4.2 deficiency. High frequency of spectrin deficiency has also been reported in studies from many different populations, such as Italy, Greece, and Brazil.^7,16,23 In data reported by Mariagabriella Mariani et al.,²⁴ band 3 deficiency was the most frequent abnormality with a high proportion such as 54% similar to data from Cynober et al.²⁵ and Rocha et al.⁶ In study from Brazil, band 3 deficiency was reported in 13%.¹⁶ In contrast with these results, we did not detect band 3 deficiency in our patients. Among Caucasians with HS, ankyrin deficiency is the most frequent erythrocyte protein abnormality.⁸ Band 3 and akyrin deficiencies were found to account for the majority in study from Portugal.⁹ In our study, four of our patients (8%) presented ankyrin deficiency. In Korean, HS patients isolated ankyrin deficiency accounted for 29.6%, combined ankyrin and spectrin deficiency, 3.7%.¹ In a study from Mexico, ankyrin deficiency was observed in 6% and it is in agreement with ours.²⁷ Protein 4.2 deficiency was identified in 8% (4/50) of our cases, equal to ankyrin deficiency and one patient had combination of protein 4.2 and spectrin deficiency. There is a wide difference between the ratios reported from different populations for 4.2 protein frequency. In a study from Japan, protein 4.2 deficiency is reported as 45%, from Italy 7%, from Greece 1%, and absent in the study from Brazil.^7,23,28

In conclusion, spectrin deficiency, isolated or combined with ankyrin, is the most common protein deficiency seen in Turkish HS patients. We can also say that there is not any correlation between Hb levels or splenectomy and protein deficiency. We believe that SDS–PAGE method is not sensitive enough to identify very minor erythrocyte membrane deficiencies resulting in HS and better techniques are needed to define those very small defects.

Acknowledgements

This study was supported by Istanbul University Scientific Research Projects Department (project number: T-722/130920).