Protocol:Vertical isoelectric focusing method

Globin Chain Synthesis
Globin chain synthesis analysis was introduced in the study of thalassaemia syndromes more than 30 years ago (1). It has greatly contributed to the understanding of the pathophysiological mechanisms of the different thalassaemia syndromes. Moreover, even in the DNA era it still remains a very sensitive diagnostic tool, very useful to define some complex or atypical forms of thalassaemia.

There are two established methods for globin chain synthesis analysis described in this chapter: one is the first classical method of Weatherall and Clegg, based on carboxymethyl cellulose chromatography for globin chain separation, the other is based on reversed phase high performance liquid chromatography (HPLC) (2). Finally a rapid method using vertical isoelectric focussing is described. The three methods described below use some common methodologies including white cell removal, reticulocyte enrichment, incubation with amino acids and radioactive tritiated leucine. However the main difference in the three methods is the methodology of separating the globin chains. The protocols are described in full for each method despite the overlap.

Vertical IEF method
When the Weatherall and Clegg globin chain biosynthesis method is used routinely (7), it is time consuming and less appropriate for routine analyses. In Leiden, to increase the analytical capacity and decrease the occupation time they have been using a simplified procedure requiring a skilled technician, occupied 5 hours on day one, two hours on day two and less then one hour in the following 3 days. When applied routinely for 20 samples at a time, the method gives reliable results within 5 days, for less than 30 minutes of occupation time per sample (8). This method is suitable for routine application in specialized laboratories. However, the procedure remains complex and susceptible to trouble shooting, thus the protocol must be followed with care.

Special Materials
Three devices increase the functionality of the system and allow the analysis of 20 samples at the time. These devices, used during reticulocyte enrichment, incubation and chain separation, can be home made, or adapted from existing commercial versions.

Reticulocyte enrichment
The rotor of a standard haematocrit centrifuge capable of 15.000 rpm can be adapted to accommodate 24 semi-capillary Pyrex centrifuge tubes (volume 0.75 ml, D = 6 mm, d = 4 mm, L = 7 mm. (custom made by the local glass workshop).

Incubation
An incubation coil (not essential) can be assembled and fitted on a circular shaker to maintain the samples in constant movement and at exactly 37oC. The coil, connected to the flow of a reliable thermostatic bath, can be made out of Perspex or equivalent plastic material and it consists of two simple parts. The lower part is a block in which in- and outlet nozzles are fitted on the sides and connected to each other by a small channel of about 2 cm wide and 3 cm deep, milled in a snake like shape in the plastic. The upper part consists of a lid with 20 holes following the shape of the channel, in which the incubation tubes (15 ml, Corning 25319) will fit. Tightness is obtained by placing between the block and lid a thick (5 mm) silicon plate gasket with the corresponding 20 holes and by screwing the two parts together. When all tubes are squeezed in place, the system will be closed and the conical end of the tubes will be in contact with the water circulating in the closed channel.

Globin chains separation
A vertical isoelectric focussing apparatus can be adapted to take 20 glass columns (D=8 mm, d= 6 mm, L=150 mm (custom made by the local glass workshop) consisting of simple glass tube slightly restricted at the bottom. Each column contains 4,25 ml of acryl amide gel. A home made device can be easily made by placing two identical Perspex tanks (w = 20 cm; d = 12 cm; h = 10 cm) on top of each other and drilling in the 6 mm thick bottom of the upper tank 20 holes (10 mm Ø), sealed with O rings in which the glass columns will fit tightly. The upper tank will contain the cathodic (+) andundefinedthe lower tank the anodic (-) electrode.

Reagents
The 14 solutions needed for the entire procedure are described below. All solutions must be either freshly made or kept in cooled stocks or frozen aliquots (-80oC). Leftover from thawed aliquots must be discarded. All glassware and disposables must be very clean or sterile.


 * 1) Reticulocytes Phosphate Buffered Saline (RTBS):
 * NaCl 0.13 M; KCl 5 mM; MgCl2.6H20 7.4 mM; Tris 12 mM pH 7.65.
 * Prepare 10x stock and keep at 5oC.
 * Dilute to working concentration at the beginning of the experiment and keep refrigerated.
 * 1) Cellulose mix suspension for WBC extraction:
 * &alpha;-cellulose C-8002 and Sigmacell type 50 S-5504, Sigma, 1:2 in RTBS.
 * The exact amount of cellulose suspension is not critical, but use about 0.5-0.6 g per column (ie 0.4 g a-cellulose and 0.2 g Sigmacell resulting in about 2-2.5 cm3 of packed column filling.
 * Prepare as much as needed for the number of samples being processed in the morning of the biosynthesis step.
 * 1) Incubation Mixture (IMIX):
 * 100 ml of solution consist of 48.6% amino acid free Fetal Calf Serum (FCS) and 51.4% RTBS and contains 0.3 mM NaHCO3; 0.6 mM MgCl2 6H20; 3.8 mM Tris; 0.2 mM Na3C6H5O7.2H2O and 1.36 mM Glucose, all at pH 7.65.
 * Keep frozen at –80oC in 0.8 ml aliquots ready for use.
 * 1) Amino acid cocktail (AC):
 * 100 ml of glutamine and leucine free amino acid solution contains 0.28 mM lysine; 0.23 mM histidine; 0.09 mM arginine; 0.07 mM tryptophan; 0.30 mM aspartic acid; 0.23 mM threonine; 0.25mM serine; 0.16mM proline; 0.30 mM glycine; 0.49 mM alanine; 0.05 mM cysteine; 0.42 mM valine; 0.047 mM methionine; 0.07 mM tyrosine and 0.19 mM phenylalanine, adjusted to pH 7.65 with 154 mM Tris-HCl up to 100 ml with RTBS.
 * Keep frozen at –80oC in 1 ml aliquots ready for use. For a recipe preparing 100 ml of stock solution please see Table 4.2
 * 1) L-Glutamine 200 mM (Glu):
 * Commercial solution (Gibco). Make aliquots of 0.4 ml and keep frozen at –80oC ready for use.
 * 1) Iron Solution (FeMIX):
 * 0.2 mg Ammonium ferrous sulphate/100ml IMIX, prepare just before use.
 * 1) Leucine L-[3,4,5,3H(N)]- in 2% ethanol (Leu3H):
 * Is kept at 5oC and used undiluted.
 * 1) Ammonium hydroxide solution pH 10.8 (NH4OH-bME):
 * This solution is used for cell lysis and is made fresh by adding it to water containing 1% &beta;-mercaptoethanol (ME) and adjusting to pH 10.8. with a few drops of ammonium hydroxide
 * 1) Sample solution (SS):
 * 100 ml of solution contains 4 ml Nonidet P40 (Fluka); 1 ml Ampholine 6-8 (Amersham Biosciences) and 10ml &beta;-mercaptoethanol, make up to 100 ml with deionized 8 M urea.
 * Keep frozen at –80oC in 1 ml aliquots.
 * 1) Electrolytes:
 * The cathodic (-) solution is 20 mM NaOH, the anodic (+) 30 mM H3PO4.
 * 1) Fixation solution for acrylamide gels (TCA):
 * 100% Trichloroacetic acid in water (stock solution).
 * Use diluted 1 to 5.
 * 1) Staining solution for acrylamide gels:
 * 0.4 g Serva violet 49 (Serva 35052) in 1 litre of water containing 50 ml of 70% perchloric acid.
 * 1) De-staining solution for acrylamide gels:
 * 10% acetic acid.
 * Acrylamide solvent: 3.5% of 50% periodic acid solution (Merck) freshly added to the necessary volume of a saturated solution of guanidine hydrochloride (Fluka). The latter can be kept for a long time at room temperature.

Sample collection
Use very fresh and sterile Li heparin samples (10 ml). EDTA samples are not suitable. Blood collected in the morning must be rigorously kept in wet ice (not cooling bags or dry ice!) during transit to the lab, and processed within 3 hours.

White cell removal
All samples are centrifuged for 5 minutes at 3000 r.p.m. (1400 g) in a refrigerated centrifuge. Plasma is separated, centrifuged at high speed and the clear supernatant is collected and kept on ice for later use. The buffy coat, layered on the cell surface, is removed with a Pasteur pipette and may be kept for DNA extraction.

From each sample 1 ml of packed red cells is suspended in 3 volumes of refrigerated RTBS. Each suspension is rapidly layered on the cellulose mix columns previously packed and equilibrated. The columns consist of glass wool filtered glass tubes of approximately 1 x 10 cm containing 2 cm3 of packed cellulose mix, prepared in the morning. WBC-free RBC are rapidly eluted from the column in 10 ml RTBS.

Reticulocyte enrichment
The RBC suspensions eluted from the columns are spun down at 2.500 (1.000 g) rpm for 5 minutes in a refrigerated centrifuge. Supernatants are discarded and approximately 1 volume RBC is taken from the upper layer and re-suspended in 2 volumes of the previously clarified original plasma. Using a long Pasteur pipette, 0.5 ml of this suspension is transferred into mini-centrifuge tubes capable of holding 0.7-0.8ml of red cell suspension (eg 3mm internal diameter, 6mm external diameter, and 70mm length). All tubes are centrifuged at 15.000 rpm (20.000 g) for 15 minutes, at 5oC in the hematocrit centrifuge, which is temporarily placed in a cold room. After carefully removing the plasma, about 5 to 10 ml of packed red cells are gently removed from the surface. This small amount of enriched material is re-suspended in a 15 ml conical polypropylene tube (Corning) containing 12 ml cooled RTBS. This high quality well stoppered tube is chosen to fit perfectly into the holes of the incubation block and to withstand centrifugation during the procedure without risk of contamination with radioactive material. The suspension is gently shaken and centrifuged for 5 minutes at 2.500 rpm (1.000 g) and 5oC. After the RTBS is thoroughly discarded the small pellet of washed red cells is ready for pre-incubation. Ten-fold reticulocyte enrichment is usually obtained using this procedure.

Pre-incubation
The incubation mixture (IMIX = 800 ml; AC = 200 ml; Glu = 20 ml; FeMIX = 20 ml) is rapidly added to each sample, the tubes are placed in the incubation block and pre-incubated for 10-15 minutes at 37oC under gentle shaking. The pre-incubation step has the function to start the synthesis eliminating the endogenous leucine which may compete with the 3H leucine, to produces cold chains in a non-random way.

Incubation and synthesis
20 ml of Leu-3H is rapidly added to each sample using the necessary precautions, the samples are incubated for 60 minutes at 37oC while gently shaking, the temperature during the incubation should be rigorously kept at 37 oC to avoid favourable synthesis of one of the globin chains, which could lead to an inaccurate globin imbalance. In theory, the synthesis of the labelled chains should go on until the synthetic capacity of the reticulocytes is exhausted. In practice, prolongation of the incubation for more than one hour only scarcely increases the amount of synthetic products but elevates the risk of proteolytic digestion for those freshly synthesized chains which are not processed into tetramers, altering the b/a ratio. Therefore incubation and proteolysis are blocked after one hour of incubation by transferring the samples onto wet ice. . In samples with an imbalanced ratio, the free globin chains are subject to proteolytic degradation, so low temperatures following the incubation are important to minimize the proteolytic process until the lyophilization step.

Washings
The cells are centrifuged in a refrigerated centrifuge for 5 minutes at 2000 rpm (600 g). The radioactive supernatant is carefully discarded and fresh cold RTBS is added. The cells are re-suspended and centrifuged repeating the washing cycle three times while the temperature is rigorously maintained below 5oC.

Lysis and lyophilization
After the third centrifugation, the washed pellets are lysed in 2 ml (NH4OH-bME) and divided in two aliquots. About one-third is transferred to eppendorf vials to be used for chain separation; the rest remains in the incubation tube for storage. Both aliquots are frozen at –90oC for 15 minutes and lyophilized under high vacuum overnight.

Preparation of acrylamide columns for chain separation on isoelectrofocusing (IEF)
The samples are loaded on IEF at the end of day two and the acrylamide columns are prepared in the morning of the same day and pre-run in the afternoon. The columns are previously siliconized, sealed at the bottom, and placed in vertical position. The amount of acrylamide solution to be prepared will depend upon the number of columns needed. The solution, in a volume of 100ml, is prepared as follows: 8.53 g. acrylamide (Bio-Rad Laboratories 161-0101) + 1.47 g N,N,'diallyltartardiamide (DADT) (Bio-Rad Laboratories 161-0620) + 48.23 g urea (USB Corporation, ultrapure 75826) are dissolved in »80ml of distilled water and stirred to clarity in the presence of 1 g mixed bed resin AG 501-x8(D)(Bio-Rad 142-6425). The solution is filtered and collected in a calibrated cylinder containing 50 mg dithioerythritol (DTE) (Sigma D-8255) and 2 g b-Alanine (Merk 1008). 3.5 ml of Pharmalyte 6.7-7.7 (Amersham-Pharmacia 17-0566-01) and 0.9 ml of Pharmalyte 8-10.5 (Amersham-Pharmacia 17-0455-01) are added and the volume is adjusted to 100 ml with distilled water. Finally, 112 ml N,N,N',N',-Tetramethyl-ethylenediamine (TEMED)(Amersham-Pharmacia 17-1312-01) and 170 ml freshly prepared 10% ammonium persulphate are added under vigorous stirring. In order to avoid premature polymerization, fill the glass columns rapidly to within 1 cm from the top, avoiding air bubbles. Layer a drop of distilled water on the surface of the acrylamide and polymerization should be completed in 2 hours.

Pre-run
The bottom seal and the water layers from the polymerized columns are removed and then they are placed gently in the holes of the vertical electrophoresis apparatus. Silicon grease may be applied to make the colums glide down vertically until they touch the bottom of the lower tank. The lower vessel (-) is filled with sufficient electrolyte (NaOH solution) to cover the end of the columns to a depth of about 5 mm. The top vessel (+) is filled with enough H3PO4 solution to submerge the end of the columns with 5 mm of electrolyte solution. 100 ml of sample solution (SS) are loaded onto the top of each acrylamide column for a 2 hour pre-run at 300 volts and at room temperature.

Sample preparation, loading and run
At the end of the 2 hour pre-run, the electrolytes are replaced by an equal amount of the same solution. The lyophilized samples are dissolved one hour in advance, in 200 ml sample solution (SS) and mixed to absolute clarity. 100 ml of each sample is carefully layered on the surface of the submerged columns and the run is started. Chain separation is optimal after a run of 16 hours at 300 V at room temperature.

Gel extraction, fixation, desalting, staining and de-staining
On the morning of day three the IEF has finished.. The gels are extracted from the siliconized columns by vigorously injecting a few mls of distilled water between the gel and the columns wall using a long thin cannula fitted on a 10 ml syringe. The gels are soft but sufficiently firm to be handled gently and transferred to stoppered »70 ml glass culture tubes, in which they are fixed with in 20% TCA. The opalescent bands of the globin chains become visible in a few minutes. The TCA is replaced at the end of day three by an equal volume of staining solution and the gels are left over-night. The next morning, the staining solution is replaced by the de-staining solution and the gels are clarified for not longer than 24 hours.

Chains extraction from the gel, solubilization and counting
The de-stained gels are removed from the solution, placed horizontally on a glass plate and eventually photographed, illuminated on a light box (Fig. 4.5). Sharp g, b and a chains bands are separated and easily identified (as well as anomalous chains when abnormal haemoglobins are present). The a chains band may smear if the IEF run is too long and the b and g chains migrate close to each other. The bands are cut out with a scalpel, and placed in glass scintillation vials. The collection of the bands must be done carefully to avoid material loss when collecting the a band and avoid contamination when collecting the b and g chain bands. In some cases the low amount of g chain protein might not result in a visible band on IEF, despite a significant amount of g synthesis. In such cases, mark with cold g chain by adding 3 ml of foetal lysate to the sample before loading on the IEF. The acrylamide is completely dissolved in 0.5 ml of solvent under gentle shaking for 1-2 hours at room temperature. 10 ml scintillation cocktail is than added to each vial and the mixtures are vigorously shaken. After 30 or more minutes in the dark, the emission of each vial is counted for 10 minutes using a DPM - CPM 3H program on a standard scintillation counter.

Interpretation of the results
The amount of 3H-leucine incorporated in the newly synthesized chains is very low. However, in samples with elevated reticulocytes counts the signals can exceed 100,000 cpm. In general, normal fresh samples should give signals of between 1,000 and 10,000 cpm for b- and a-chains, this is sufficient for calculation of a reliable ratio. In patients with low bone marrow activity or when samples are poorly treated, the incorporation of label may be insufficient. Signals between 500 and 1000 cpm give less precise ratios and signals below 500 cpm should be considered unreliable. The background threshold on the scintillation counter used should be tested by measuring the counts in several standards and blanks (the latter consisting of an approximately equivalent portion of acrylamide gel free of 3H labeled chains). In the Leiden laboratory, the background of any portion of the acrylamide gel, which does not contain 3H labelled globin bands, but has approximately the same size, gives a constant background count of about 40 cpm. When the globin signal is lower than 1000 cpm, the background count should be subtracted. Sometimes low sample counts may be reliable if an appropriate luminescence correction is made by the scintillation counter.

Results
This method has bee used extensively (9-13), and the modified biosynthesis method has been used in the Leiden University laboratory for determining the biosynthesis ratios in almost 2,000 cases. Virtually all samples with an unbalanced synthetic ratio have been analyzed at the DNA level and the corresponding molecular defects have been found.

The results that can be expected are summarized as follows:

Iron deficiency
Patients with iron deficiency should ideally be treated with iron supplements before globin chain synthesis is performed because iron deficiency may induce an artificial imbalance (which is not restored by adding iron during incubation) since there is preferential binding of haem to the &beta;- rather than the &alpha;-chain. Thus for patients with suspected &alpha;-thalassaemia who have &beta;/&alpha; ratios between 1.1 and 1.4, this result may be due to iron deficiency and not thalassaemia, however iron deficiency and thalassemia often coexist, especially in females with &alpha;-thalassaemia.

Thalassaemias
Cases without thalassaemia or iron deficiency will give a balanced b/a ratio. The b/a ratio distribution, classified according to the type of thalassaemia and to the specific molecular defects, is shown in Fig. 4.6. All cases of thalassaemia induced by &beta;o or &beta;+ determinants in homozygous or combined heterozygous form give a &beta;/&alpha; ratio around zero. Only carriers of mild &beta;+ promoter defects are significantly separated from other &beta;o and &beta;+ determinants. In a-thalassaemia, there is overlap between the &beta;/a synthesis ratios for the heterozygous (-&alpha;/&alpha;&alpha;) and the homozygous &alpha;+ (-&alpha;/-&alpha;) defects, the heterozygous &alpha;o (--/&alpha;&alpha;) and the &alpha;+/o polyadenylation defect of the &alpha;2-gene. Reliable differentiation between one or two a gene defects by the average ratio is not possible. However, there is a statistically 2.3 times higher chance of having a two affected &alpha;-genes rather than a single a gene defect with a ratio between 1.4 and 1.6. Conversely, 9 of the 10 single a gene defects will give a ratio between 1.2 and 1.3. The &beta;/&alpha; ratio obtained for the "3-genes" &alpha;-thalassaemia defects (HbH disease) is scattered from 1.5 to almost 5.0.

The abnormal haemoglobins
The globin chain synthesis is not only effective in thalassaemia analyses but also quite useful in the presence of frequent or rare abnormal haemoglobins in combination with thalassaemias. Another advantage is that the abnormal &alpha; or &beta; chains can be distinguished on IEF indicating which gene to sequence. The differential diagnosis between HbS homozygosity and compound heterozygosity HbS/&beta;-thalassaemia is possible by measuring the expression of both &beta;-alleles. This is particularly useful in combinations with &beta;-globin gene deletions (9). Moreover, the high incidence of &alpha;+-thalassaemia in many ethnic groups and especially among HbS carriers often generates genetic compounds that are detectable by their synthetic ratio (10,11,13).

Table 4.4:  &beta;/&alpha; ratio measured in several combinations of abnormal haemoglobins with &beta;- and &alpha;-thalassaemia

Legends to the figures
Fig.4.5: Acryl amide gel columns showing separated products. The bands of the &gamma;, &beta; and &alpha; chains and the anomalous chains are cutout and processed in scintillation vials.



Fig. 4.6: The distribution of the &beta;/&alpha; ratio measured in different types of thalassaemia are reported in this diagram according to the class of molecular defect. From left to right, thalassaemia major induced by &betao-thalassaemia shows ratios ranging from 0 to 0.1. The &beta;o-thalassaemia carriers show ratios between 0.2 and 0.7 and &beta;+-thalassaemia carriers between 0.3 and 0.9. Ratios between 0.7 and 0.9 are measured in carriers of the mild &beta;+ promoter defects. In absence of thalassaemia or iron deficiency the &beta;/&alpha; ratio are usually perfectly balanced (0.95-1.05). The slight imbalance at both edges of the distribution is mostly due to sample with lower counts. The area between 1.1 and 1.4 represents a number of patients suspected of &alpha;-thalassaemia, which resulted non-thalassaemic, but profoundly iron depleted. The single a gene defects shows a &beta;/&alpha; ratio distribution between 1.15 and 1.65. The 2 alpha gene defects between 1.25 and 1.80, the 3 alpha genes defects (HbH disease) between 1.5 and 5.0 (small diagram). See text for more details.