Butyrylcholinesterase (BCHE) genotyping for post-succinylcholine apnea in an Australian population.
cases are attributable to abnormalities of the butyrylcholinesterase enzyme (BCHE; EC 220.127.116.11; OMIM 177400) (2, 3), a plasma esterase that hydrolyzes acetylcholine, other choline esters, and choline ester-containing drugs such as SC. Measurement of the biochemical defect for the identification of individuals with inherited abnormalities is routinely performed by enzyme assay and inhibitor-based phenotyping, but diagnosis using these methods is less than ideal. DNA-based approaches can improve diagnostic precision, and a possibility of new drugs on the market possessing the same gene-based determinants of enzyme hydrolysis means that timely implementation of DNA-based methods in routine analysis will be important so that laboratories have a role in the development of diagnostic pharmacogenetics.
Prolonged post-SC apnea occurs with BCHE deficiencies of marked severity (impairment >70%), and the majority are attributable to BCHE gene defects (2). Hypocholinesterasemia can occur secondary to pregnancy, liver disease, malnutrition, and chronic debilitating illnesses, but enzyme impairments are usually mild and rarely cause severe deficiency and prolonged apnea (6% of cases) (2). Hence, the identification of individuals with inherited BCHE variants remains within the province of the laboratory because the demonstration of low plasma cholinesterase activity (BCHE) and abnormal enzyme phenotype (defined by responses to dibucaine and fluoride inhibitors) quantify the degree of impairment and confirm its hereditary basis.
As early as 1978, the Danish Cholinesterase Research Unit reported that 13% of individuals with abnormal SC responses were unable to be typed by inhibitor-identified alleles (2), an observation reinforced 4 years later by Morgan (4), who suggested that up to 8-10% of patients were potentially misdiagnosed because of difficulties with phenotyping. Phenotype studies are prone to inaccuracy when enzyme amounts are low, but additional problems of variable enzyme response in different substrates and a diagnostic range of 10 phenotypes contribute to difficulties with identification. Inhibitor-based methods are also incapable of differentiating compound heterozygotes (with the exception of heterozygotes with an atypical allele) because inhibitor numbers frequently overlap in the distributions for the wild type and enzyme variants (5).
DNA-based studies have described at least 44 mutations in the BCHE gene that cause primary hypocholinesterasemia, with the D70G ([Asp.sup.703] [right arrow] Gly), or dibucaine point mutation in exon 2, the most extensively studied. The amino acid change reduces the binding affinity of BCHE enzyme for SC and dibucaine, rendering the molecule ineffective in enzyme hydrolysis (6). Three point mutations encode fluoride resistance, flu-1 (T243M; Thr [right arrow] 2433Met), flu-2 (G390V; [Gly.sup.390]Val), and the Japanese flu-variant (L330I; [Leu.sup.330] [right arrow] Ile) (7, 8), whereas the silent BCHE variant may arise from any of 33 known mutations [reviewed by Primo-Parmo et al. (9)].
A mutation detection approach can be used for investigation and characterization of BCHE variants because the phenotype is defined, the gene is relatively small [73 kb; four exons interrupted by three introns with exon 1 not translated; coding region 1722 nucleotides], and most abnormalities described are point mutations. However, the implementation of DNA-based methods in biochemistry laboratories is slow or, in the transition, is hampered by limited time and resources, lack of technical expertise, and the financial outlay for equipment and technology. Furthermore, paucity of information about the clinical significance of BCHE polymorphisms and uncertainty related to the extent of additional DNA investigation when mutations are not detected are ongoing concerns; therefore, the scope of DNA studies would need to be clarified before genotyping can be offered.
Mindful of these issues, we developed a protocol for analysis of five BCHE mutations based on competitive oligo-priming (COP)-PCR and DNA sequencing analysis. During 1999-2002, we genotyped 65 individuals referred for investigation of prolonged post-SC apnea. We provide important information on gene defects responsible for inherited hypocholinesterasemia in our referral population and the prevalence of novel mutations in this condition. We also show results from BCHE gene sequencing in cases of SC apnea unexplained by the five-mutation screening and suggest alternative causes for their abnormalities.
Materials and Methods
AmpliTaq DNA polymerase and Mg[Cl.sub.2] were from Applied Biosystems. Ultrapure water was from Fisher Biotec. Dimethyl sulfoxide (100%) was from Ajax Chemicals.
The 5x PCR buffer, pH 9.0, was manufactured inhouse using Tris-HCl (Sigma) and enzyme grade ammonium sulfate (ICN Biomedicals). Deoxynucleotide triphosphates (dNTPs) were from Promega. Oligonucleotides (Tables 1 and 2) were synthesized and supplied by Genset Pacific Pty. Ltd. (Lismore, Australia).
All patients who were referred for investigation after delayed recovery from procedures involving SC or mivacurium or for apnea risk assessment during 1999-2002 were included. Patients with secondary causes of hypocholinesterasemia were excluded. Decreased plasma BCHE was not a prerequisite for mutation detection. When an entire family was genotyped, only the proband was included. All participants were living in Australia at the time of testing. Patients consented to diagnostic genetic testing, and for analysis, all patient-specific identifiers were removed.
Study participants included 29 males and 36 females (age range, 4-83 years). Referrals were from New South Wales, except for one from South Australia. The ethnic backgrounds were diverse: two patients were Asian, five were Southern European, one was from the Middle East, and the rest were Anglo-Celtic Australian.
TOTAL PLASMA BCHE ACTIVITY
Total plasma cholinesterase (BCHE) was measured with butyrylthiocholine as a substrate on a Hitachi 917 analyzer (Roche Diagnostic), according to the method described by the manufacturers (10). Activity was calculated in kU/L at 37[degrees]C. BCHE enzyme activity was stable in serum and EDTA plasma stored at 4[degrees]C for up to 7 days. The reference interval for BCHE activity in healthy individuals was 7-16 kU/L, and the interval was rechecked in 2003 using stored samples from healthy individuals. The mean (SD) was 10.6 (2.0) kU/L.
Genomic DNA was extracted from EDTA-anticoagulated venous blood with use of the QIAamp Blood-DNA mini reagent set (Qiagen) according to the manufacturer's instructions. DNA was eluted in 200 [micro]L of water in the final step and stored at -20[degrees]C until required.
The five BCHE mutations analyzed in the mutation screen were dibucaine (D70G; [Asp.sup.70] [right arrow] Gly), silent-1 (sil-1; G117FS; Gly GGT to GGAG), fluoride-1 (flu-1; T243M; [Thr.sup.243] [right arrow] Met), fluoride-2 (flu-2; G390V; [Gly.sup.390] [right arrow] Val), and K-variant (A539T; [Ala.sup.539] [right arrow] Thr; Fig. 1), as summarized by La Du et al. (11). COP-PCR was used for flu-1, flu-2, and K-variant, and PCR amplification with direct DNA sequencing analysis was used for dibucaine and sil-1 mutations (see results of BCHE genotyping).
Competitive oligonucleotide priming of DNA synthesis has been described previously in detail (12, 13). In brief, COP-PCR is a method for allele-specific amplification in which two oligonucleotide primers of similar length compete for annealing to DNA. Competitive primers are usually identical except for the point mutation, which is positioned at the middle of the primer [hence designated wild-type- (WT) or mutation (MU)-specific primers (14)]. When PCR is performed under competitive conditions, amplification of the better-matched primer is favored >100-fold over amplification of the primer with a mismatch (13).
Each COP reaction required a set of three primers (two competitive primers and one common primer). Duplicate sets of these primers were synthesized; the first set was named POS-hex COP-PCR because the MU-specific primer was labeled with hex fluorophore (6-carboxyfluorescein hexachloride), whereas in the second set, NEG-hex COP-PCR, the WT primer was labeled with hex (Fig. 3 and Table 1).
[FIGURE 1 OMITTED]
COP-PCR buffers and reaction mixtures were preoptimized for the detection of the mutation indicated. The flu-1 COP-PCR reaction mixture contained 5x buffer [300 mM Tris-HCl, pH 9.0, 62.5 mM [(N[H.sub.4]).sub.2]S[O.sub.4]], dimethyl sulfoxide (1:10 by volume), 2 mM Mg[Cl.sub.2], the four dNTPs (dATP, dCTP, dGTP, and dTTP; 250 [micro]M each), 1 U of AmpliTaq DNA polymerase, and 24 pmol of each primer (Table 1) in a final volume of 20 [micro]L. The flu-2 COP-PCR mixture contained the same 5x buffer, 2 mM Mg[Cl.sub.2], the four deoxyribonucleotide triphosphates, AmpliTaq DNA polymerase, and the flu-2 primers (same amounts as indicated for flu-1). K-variant COP-PCR mixtures contained 5x buffer, 1.5 mM Mg[Cl.sub.2], the four deoxyribonucleotide triphosphates, AmpliTaq DNA polymerase, and K-variant primers (same amounts as for flu-1).
The COP-PCRs for the three mutations were performed simultaneously in a PC-960G gradient thermal cycler (Corbett Research), all with the same cycling conditions. DNA was denatured at 95[degrees]C for 3 min, followed by annealing at 44[degrees]C for 60 s and elongation at 72[degrees]C for 60 s. This was followed by 35 cycles of denaturation (15 s at 95[degrees]C), annealing (30 s at 44[degrees]C), and extension (20 s at 72[degrees]C). The last cycle was followed by 10 min at 72[degrees]C and 1 min at 25[degrees]C. DNA controls with confirmed mutations were used during each run. For flu-1 and flu-2, both WT/unaffected DNA and heterozygous/affected DNA samples were coamplified. For K-variant, WT/unaffected and heterozygous/affected and homozygous/affected Kvariant DNA control samples were used.
COP-PCR-BASED MUTATION ANALYSIS
COP-PCR products (1 [micro]L) were separated on a 5% nondenaturing Long Ranger polyacrylamide gel (BioWhittaker Molecular Applications) that had been prerun with ethidium bromide. Fragments of interest were 70-120 bp in length, and detection was by fluorescent-laser gel scanning (Gel scanner GS-2000; Corbett Research). DNA controls were also separated on the gel with patient samples. Any patient sample that yielded an unusual fragment pattern was reanalyzed by direct DNA sequencing. All flu-1 and flu-2 mutations identified by COP-PCR were verified by DNA sequencing studies.
DNA SEQUENCING OF DIBUCAINE AND SIL-1 LOCI
An exon 2 region that included the dibucaine (D70G) and sil-1 (G117FS) loci was amplified as a single 570-bp fragment with primers BC14 and BC3 (Table 2). DNA was amplified in 5x buffer [300 mM Tris-HCl, pH 9.0, 62.5 mM [(N[H.sub.4]).sub.2]S[O.sub.4]], 2 mM Mg[Cl.sub.2], four dNTPs (dATP, dCTP, dGTP, and dTTP; 250 [micro]M each), 1 U of AmpliTaq DNA polymerase, and 8 pmol each of BC14 and BC3 in a final volume of 20 [micro]L. Samples were denatured at 95[degrees]C for 3 min, annealed at 46[degrees]C for 60 s, and elongated at 72[degrees]C for 60 s. This was followed by 35 cycles of denaturation (15 s at 95[degrees]C), annealing (30 s at 46[degrees]C), and extension (20 s at 72[degrees]C). The final cycle was followed by 10 min at 72[degrees]C and 1 min at 25[degrees]C. PCR product clean-up and concentration were by polyethylene glycol precipitation as described previously (15). A small amount of concentrated PCR product was separated by agarose gel electrophoresis to verify the presence of a single 570-bp band, and products were then sent to Sydney University/Prince Alfred Macromolecular Analysis Centre (SUPAMAC) for direct sequencing by the dye dideoxy terminator cycle sequencing system on ABI automated sequencers.
ADDITIONAL SEQUENCING OF BCHE GENE CODING REGIONS
For patient samples in which the five-mutation screen indicated no mutation or the mutations found were inconsistent with BCHE activity, additional BCHE coding region analysis was performed (exons 2-4; Fig. 1). Because the BC14/BC3 region was sequenced for the fivemutation screen, only five additional PCR reactions were required: flu-1com/flu-2com, BC11/BC12, BC5/BC13, BC8/BC9, and K5/Kext (Table 2).
For simplicity, the five amplification reactions were performed with the same buffer, Mg[Cl.sub.2], dNTPs, and AmpliTaq as for BC14/BC3. Only the primers and annealing temperatures were different (Table 2). PCR products were sequenced by SUPAMAC, and the electropherogram results were compared with the BCHE exon sequence on the Human Genome Mutation Database (accession no. M16541 J02964; human butyrylcholinesterase mRNA complete coding sequence). All DNA abnormalities were compared with the list of known BCHE mutations. Novel mutations were confirmed by sense and antisense sequencing, with clinical follow-up in family members.
design of cop-pcr for BCHE mutation analysis COP-PCR is a method with high specificity for detection of single base-pair changes because annealing between MU-specific and WT-specific primers under competitive conditions overwhelmingly favors amplification of the closest matched primer. The original method used different fluorophore labels for identification of primers (Fig. 2). We chose a single fluorophore and used hex-labeled primer in competition with unlabeled primer (Fig. 3). This modification produced discordant migration of amplimers on the nondenaturing polyacrylamide gel. PCR products with a hex attachment had reduced mobility, and the product appeared 2- to 5-bp larger than its equivalent nonlabeled counterpart. Fluorophore dye attachment is known to affect the electrophoretic mobility of DNA fragments (16), but the effect is usually imperceptible with standard laser detectors. In our system, the differences were visible because of the higher resolution of the Corbett analyzer system (~ 1-bp), and we were able to use fragment size to deduce the identity of the annealed primer and, hence, the mutation locus.
Because both WT- and MU-specific primers could be tagged with fluorophore, the COP-PCRs were analyzed in duplicate. The two PCRs were arbitrarily called POX-hex and NEG-hex (POS denoting the PCR with the MU-specific primer tagged, and NEG indicating fluorophore on the WT primer; Fig. 3).
Interpretation of the genotype was based on the number and sizes of bands on the gel (shown schematically in Fig. 4A). WT/unaffected DNA samples amplified in POShex reactions yielded a single PCR product at the predicted molecular weight. In NEG-hex reactions, the amplimers carried the hex label; therefore, the fragment was 2 bp larger. Thus, after gel separation, the appearance of a lower band in POS-hex and an upper band in NEG-hex samples indicated a WT/unaffected genotype (Fig. 4B). Likewise, patients who were homozygous/affected yielded PCR products with the reversed pattern (one upper band in POS-hex, one lower band in NEG-hex). Heterozygous-affected DNA characteristically revealed two PCR products in approximately equal amounts in POS-hex and NEG-hex lanes because both the MU-specific and WT alleles were present.
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
INTERPRETATION OF FLU-1, FLU-2, AND K-VARIANT MUTATIONS BY COP-PCR
Interpretation of genotype at the flu-1 and flu-2 loci by COP-PCR was straightforward, and from our patient population, four heterozygous/affected patients were identified. In contrast, K-variant was difficult to optimize for COP-PCR under the conditions established for flu-1 and flu-2 detection. K-variant PCR products from POShex amplifications adhered to the predicted pattern and specified the correct genotype, as verified by DNA sequencing (not shown), but amplification products from NEG-hex reactions migrated abnormally to the higher molecular weight position irrespective of the patient's genotype (Fig. 4B).
These anomalies are usually the result of interference, or a partial mismatch from suboptimal cycling conditions, and although they can be eliminated by adjustment of PCR conditions, the cycling conditions in our assay were defined in order to coordinate amplifications at flu-1 and flu-2. Adjustment of pH and [Mg.sup.2+] concentrations did not improve the specificity (not shown), and because other strategies substantially increased the number of steps and turnaround time of COP-PCR, the interpretation of Kvariant was based on POS-hex reaction products alone (NEG-hex reaction products were not used). Results are shown here for completeness.
[FIGURE 4 OMITTED]
DIRECT DNA SEQUENCING FOR DIBUCAINE AND SIL-1 MUTATION ANALYSIS
The dibucaine and sil-1 mutations were analyzed by DNA sequencing because their relative proximity enabled them to be coamplified in a single 570-bp product. Interpretation of DNA sequencing results was straightforward. At first, the dibucaine and sil-1 mutations were also analyzed by COP-PCR (with DNA sequencing confirmation of affected individuals), but it became more expeditious for us to proceed directly to sequencing analysis, given the high frequency of dibucaine abnormalities in our population. Interestingly, the sequencing results revealed additional mutations in three patients (N106FS, N96Y, and F28I; described in the "New Mutations" section).
BCHE GENOTYPES IDENTIFIED FROM A PATIENT POPULATION
Sixty-five patients who presented to our service between 1999 and 2002 were genotyped, and 52 were diagnosed with primary hypocholinesterasemia attributable to BCHE mutations (80%; Table 3, with details available in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/ vol49/issue8/). For 44 patients, the five-mutation screen provided complete diagnosis (85% diagnostic capability for this test in our population). BCHE coding region sequencing studies identified additional defects in eight others. Primary hypocholinesterasemia was excluded in 13 patients after the assays failed to identify pathogenic defects in the BCHE gene.
In the 52 patients with hereditary hypocholinesterasemia, the most common DNA abnormalities were dibucaine and K-variant mutations: 47 patients carried the dibucaine mutation, whereas 48 were identified with the K-variant (allele frequencies, 0.72 and 0.74 respectively; Tables 3 and 4). Because of known differences in atypical allele frequencies in difference races [reviewed in Ref. (17)], patient ethnicity was recorded. Because the numbers were low (Table 5), ethnicity had no significant impact on the prevalence of the mutations identified. The allele frequency of the dibucaine mutation in Australia was consistent with observations from studies in Caucasian populations.
GENOTYPES IN PRIMARY HYPOCHOLINESTERASEMIA
Primary hypocholinesterasemia-causing BCHE deficiency has autosomal recessive inheritance, and all 52 patients diagnosed had compound gene defects (i.e., compound heterozygotes and compound homozygotes). No homozygotes with fluoride or silent mutations were found, and none of the patients had an isolated homozygous dibucaine defect. Co-inheritance of dibucaine and Kvariant occurred in 88% of patients with primary hypocholinesterasemia.
The most common genotype responsible for inherited hypocholinesterasemia was compound homozygous dibucaine/ homozygous K-variant (AAKK; 44%). The other three dibucaine/K-variant combinations were next in frequency: compound heterozygous dibucaine/heterozygous K-variant (AK; 13%), compound homozygous dibucaine/ heterozygous K-variant (AAK; 7.7%), and compound heterozygous dibucaine/homozygous K-variant (AKK; 7.7%; Table 5 with details in the online Data Supplement). In total, 73% of patients with primary hypocholinesterasemia had abnormalities at the dibucaine and K-variant loci only. The remaining 27% of patients had various genotypes, such as dibucaine and K-variant plus a third mutation, unusual BCHE mutations, or rare combinations where dibucaine and K-variant linkage was absent. Five patients had sil-1, flu-1, or flu-2 mutations, collectively representing 10% of all cases (allele frequencies, 0.01, 0.02, and 0.02, respectively; Table 4).
For the diagnosis of primary hypocholinesterasemia, a correlation between gene defect and the magnitude of enzyme impairment was required. Fifty of 52 patients diagnosed had genotypic abnormalities that supported the degree of BCHE enzyme impairment discovered. Two patients had BCHE activities below the mean and SD of activities calculated from the other patients in their genotype group. Patients 57 and 65 had compound heterozygous dibucaine/heterozygous K-variant and compound heterozygous dibucaine/homozygous K-variant abnormalities (see the online Data Supplement at http://www. clinchem.org/content/vol49/issue8/), which are genotypes consistent with primary hypocholinesterasemia. However, the severity of enzyme deficiency appeared atypical for the genotype in both cases (4.8 and 1.7 kU/L, respectively; Table 5). Furthermore, no other BCHE gene abnormalities were found on sequencing analysis to account for the discrepancy.
RESULTS FROM BCHE SEQUENCING, INCLUDING NEW MUTATIONS
Gene sequencing of BCHE coding regions (exons 2-4) revealed additional BCHE mutations in 8 patients, increasing the total number with primary hypocholinesterasemia to 52 (see Table 3 and the detailed results in the online Data Supplement). Two mutations are well described; Ile6FS del-A (also known as sil-2) is a frameshift mutation that creates a new stop codon that prematurely truncates polypeptide synthesis [described in Ref. (9)]. The G115D mutation ([Gly.sup.115] [right arrow] Asp; also named sil-7) encodes a BCHE variant with marked instability (18). Our sequencing analysis also uncovered six new BCHE mutations: N106FS, R424X, E460K, F28I, N96Y, and R386C (Fig. 5; patients are also listed in the online Data Supplement).
Patient 35. Patient 35 had two novel mutations. N106FS-insA was a single A-nucleotide insertion in exon 2 that changed the codon for Asn1063Lys, and shifted the reading frame +1 (Fig. 5A), creating a new stop codon 23 amino acids downstream at position 129. Truncation was upstream to the active center [Ser.sup.198;] therefore, catalytic activity was absent, and this variant reached only 22.5% of the length of the mature protein. The A insertion occurred within a train of A repeats, sites known to be predisposed to frameshifts. The truncated product of N106FS-insA was the same length as the sil-1 mutation polypeptide product (G117FS; Gly -GGT to -GGAG), except for 11 additional mistranslated amino acids in the N106FS polypeptide (N106FS was 11 codons upstream of G117FS).
[FIGURE 5 OMITTED]
The second novel mutation (R424X) was a C-to-T substitution that changed Arg424 to a stop codon (Fig. 5B). Although the product was truncated close to the COOH terminus (80% of full length), enzyme function was unlikely because deletion constructs have shown a 525amino acid polypeptide is the minimum length for BCHE compatible with function (19). Pedigree studies on the patient's two daughters indicated both are carriers of the N106FS-insA but not R424X (Fig. 6A). The daughters had mild hypocholinesterasemia, suggesting that the mutations were interallelic, with a nonfunctional protein encoded by R424X.
Patient 56. Patient 56 had a novel mutation (E460K) that was a G-to-A substitution that changed Glu460 to a Lys (Fig. 5C). She also carried the sil-7 mutation (Fig. 6B), which accounted for her severe hypocholinesterasemia.
Patient 39. Patient 39 had a novel mutation (F28I), which was a T-to-A substitution close to the N[H.sub.2] terminus that converted [Phe.sup.28] to Ile (Fig. 5D). This proband was heterozygous for dibucaine and K-variant, hence, compound heterozygous for three mutations (Fig. 6C). The severity of BCHE deficiency indicated that F28I encoded a pathogenic variant because family studies revealed that F28I was separate from dibucaine and K-variant.
Patient 51. Patient 51 had a new mutation (N96Y), which was an A-to-T substitution that changed [Asn.sup.96] to Tyr (Fig. 5E).
Patient 8. Patient 8 had a heterozygous R386C mutation, which was a C-to-T substitution that changed Arg386 to Cys (Fig. 5F). Family studies have yet to be performed.
PRIMARY HYPOCHOLINESTERASEMIA EXCLUDED
From 65 patients genotyped, 13 had primary hypocholinesterasemia excluded. Three had normal BCHE activities with a wild-type BCHE genotype (patient details are available in the online Data Supplement at http://www. clinchem.org/content/vol49/issue8/). Three patients had hypocholinesterasemia with BCHE gene defects: two K-variant homozygotes (patients 6 and 21), and one dibucaine heterozygote (patient 4; shown in the Data Supplement). Their enzyme activities were low compared with the reduction of 30% usually observed for K-variant homozygotes and the 17-20% expected for dibucaine heterozygotes (20); however, neither genotype causes primary hypocholinesterasemia. Additional sequencing studies failed to demonstrate other abnormalities; therefore, secondary deficiencies were suggested. The last seven patients had hypocholinesterasemia with no corroborating BCHE mutations, despite sequencing of BCHE coding regions (patient details are available in the online Data Supplement).
Characterization of inherited BCHE variants by inhibitor-based phenotyping is a well-described technique capable of identifying atypical, fluoride, and silent alleles. Difficulties arise with detection of compound heterozygotes or patients who possess rare or new mutations. This is because phenotyping lacks specificity (as shown when different genotypes encode the same inhibitor phenotype, e.g., AA and AS) and has relative insensitivity (especially for mutations that do not affect the regions recognized by common inhibitors). A DNA-based approach to diagnosis may be of value, but to date BCHE genotyping has been reported after analysis of one or a few mutations. Furthermore, it is unknown whether the expense and effort of genotyping yields results that are significantly more informative than phenotyping on a population scale or as clinically important as they are purported to be in family studies after gene sequencing is used to identify mutations. Our approach to BCHE genotyping considered issues of time- and cost-effectiveness, with a method for identification of common abnormalities and optional supplementary analysis by DNA sequencing when the fivemutation screen was inconclusive. We characterized the BCHE gene abnormalities in 65 patients from our referral population and report the prevalences of mutations and genotypes identified, with the clinical and laboratory implications of the distributions found.
The importance of the dibucaine mutation in patients with primary hypocholinesterasemia was undisputable. All of the homozygous patients and 19 of 24 compound heterozygotes possessed dibucaine abnormalities, indicating a frequency of 90% in patients with inherited hypocholinesterasemia. However, none of the patients had simple homozygous genotypes. Patients were either compound dibucaine homozygotes or compound heterozygotes, with both groups in approximately equal numbers (Table 6). This is the first report of the compound genotype-prevalence in this condition.
The mutation relationship responsible for most compound genotypes was coexpression of dibucaine and K-variant (88% of patients). Compound homozygous dibucaine/ homozygous K-variant was the most common genotype, accounting for 44% of inherited hypocholinesterasemia. A significant role for K-variant was not surprising given the allele frequency of 13-22% in the general population (homozygote frequency, 1 in 65) (21). The K-variant, G-to-A base substitution that changes Ala539 to Thr, manifests as a 30% reduction in circulating BCHE in K-variant homozygotes (21). In the presence of the dibucaine mutation, which impairs the affinity of BCHE for SC, a K-variant mutation contributes significant additional enzyme abnormality (21). The mutations are linked, as shown previously by Bartels et al. (21), with the K-variant identified in 89% of individuals with the dibucaine allele. The prevalences of K-variant and dibucaine were remarkably similar in our population (allele frequencies, 0.72 and 0.74 for dibucaine and K-variant, respectively), and co-association was slightly more frequent (K-variant present in 96% of patients with the dibucaine mutation).
Although linkage is well described, the number of dibucaine and K-variant mutations did not always correlate in the genotypes of our patients. Identical numbers of dibucaine and K-variant alleles occurred in 30 of 52 patients, indicating that 22 patients exhibited a discordant number of dibucaine or K-variant mutations or that one of the mutations was identified in combination with other mutations. Because it appeared that the K-variant dosage was not universally predicted by the number of dibucaine mutations, we determined the contribution of K-variant to BCHE activity in patients coexpressing both mutations. BCHE activities in dibucaine homozygotes were unaffected by K-variant hetero- or homozygosity (Table 5), but K-variant homozygosity and heterozygosity altered the activity of BCHE in dibucaine heterozygotes (6.25 vs 8.13 kU/L, respectively), which is similar in magnitude to the activities reported by Bartels et al. (21). Hence, the K-variant locus was important for genotyping because the prevalence of the K-variant was high, its allele dosage was unpredictable, and the magnitude of its effect on enzyme phenotype was sufficiently large to alter the interpretation.
The frequency of compound heterozygotes in primary hypocholinesterasemia was surprisingly high (representing almost one-half of all patients), and compound heterozygous dibucaine/heterozygous K-variant was the second most frequent genotype identified. However, compound heterozygosity was not limited to dibucaine/ K-variant combinations. Other BCHE mutations contributed to the compound genotype in 27% of cases, including rare mutations (sil-1, flu-1, and flu-2), a multiplicity of mutations, and novel mutations. The diversity and complexity of these mutation combinations clearly explained the inability of inhibitor-based phenotypic criteria to define some variants. It also indicated that DNA-sequencing instrumentation or access to a commercial referral service is required to achieve the diagnostic success shown here. In fact, analysis of a limited number of mutations could be ill advised because our findings suggested up to 27% of genotype results would be equivocal in the absence of DNA sequencing studies.
BCHE genotyping was useful for confirming the absence of BCHE gene defects in a small number of patients. Patients with the wild-type BCHE and normal BCHE activity were usually individuals referred for risk assessment because of a known family history of SC sensitivity. Occasionally patients are genotyped on the strength of clinical suspicion, despite a BCHE activity within reference values, but to date no BCHE mutations have been identified in our patients with BCHE activity within our reference interval. Explanations for prolonged apnea with normal BCHE activity are SC overdose, poor intubation technique, electrolyte instability, and respiratory insufficiency from other medical causes.
We identified patients with hypocholinesterasemia and nonpathogenic BCHE mutations, where the diagnostic problem was enzyme impairment disproportionate to the nonpathogenic genotype. Inconsistencies or small exaggerations in BCHE activity measurements may indicate the presence of secondary factors or prompt a review if measurements were supplied from a different laboratory.
The final group of patients had unexplained hypocholinesterasemia; we found no BCHE defects in these patients, even after exhaustive DNA sequencing. Few explanations exist, apart from failure to exclude secondary hypocholinesterasemia or sample collection too soon after exposure to SC. Abnormalities in noncoding regions of the gene is one possibility, particularly because our sequencing studies were limited to the gene coding regions. To date, however, only 1 of 44 BCHE mutations discovered is in a noncoding region (intron-2/exon-3 boundary, -8nt G) (9). Two other abnormalities have been described, an exon 1 defect (nt -116) and an exon 4 untranslated region mutation (nt -1914), both in linkage disequilibrium with the K-variant; however, their pathologic significance has yet to be shown (21). Some authors have attributed hypocholinesterasemia to involvement of another, unknown protein or factor that modifies the expression of BCHE enzyme activity. This was suggested by Dey et al. (22) when no mutation explanation could be found for 2 of 10 individuals with severe hypocholinesterasemia, in spite of complete BCHE gene sequencing. The seven patients with unexplained hypocholinesterasemia in our study represented 11-15% of the patient population and resembled the 20% undiagnosed by Dey et al. (22); therefore, a study of other proteins affecting BCHE function may be worthy of future investigation.
The prevalences of mutations identified in our study compared favorably with those described with use of biochemical inhibitors, even with slight differences in patient composition. In 1978, Viby-Mogensen and Hanel (2) typed 225 Danish patients with silent phenotypes and determined that the phenotype composition was 47% homozygous atypical allele, 13% heterozygous atypical allele, 9% heterozygous silent allele, 3% heterozygous fluoride allele, 6% rare genotypes, and 13% who could not be classified diagnostically by biochemical criteria. Our referral population was composed of 43% dibucaine homozygotes, 31% dibucaine heterozygotes, 8% silent mutations, 6% fluoride mutations, 12% rare mutations, and 11% with genotypes that could not be resolved. The relative agreement between populations was remarkable and suggested that the majority of patients would not have their phenotype classification altered by genotyping and could benefit from the increased diagnostic precision of mutation detection strategies.
In conclusion, our work genotyping patients with prolonged post-SC apnea revealed high prevalences of dibucaine and K-variant defects in primary hypocholinesterasemia; therefore, a mutation screening approach was suitable for diagnosis in these individuals. However, we also showed that BCHE mutation heterogeneity complicated the genetic diagnosis for 27% of patients, particularly because six novel mutations were identified during this study. We used a five-mutations screening approach, which was cost-effective, produced results in 14-21 days, and was suitable for a hospital diagnostic laboratory. We also showed that with additional DNA sequencing, a full diagnostic service can be provided and that a combination of approaches was the most efficient use of DNA detection strategies currently available.
We thank Dr. John B. Whitfield for critical reading of the draft manuscript and Dr. Stella Koulianos (SUPAMAC) for assistance with electropherograms.
Received November 1, 2002; accepted May 2, 2003.
(1.) Bauld HW, Gibson PF, Jebson PJ, Brown SS. Aetiology of prolonged apnoea after suxamethonium. Br J Anaesth 1974;46:273-81.
(2.) Viby-Mogensen J, Hanel HK. Prolonged apnoea after suxamethonium: an analysis of the first 225 cases reported to the Danish Cholinesterase Research Unit. Acta Anaesthesiol Scand 1978;22: 371-80.
(3.) Jensen FS, Viby-Mogensen J. Plasma cholinesterase and abnormal reaction to suxamethonium injection: twenty years' experience with the Danish Cholinesterase Registry. Ugeskr Laeger 1996;158:1835-9.
(4.) Morgan AA. Apnoea following suxamethonium: the genetic study of four generations of a family. J Med Genet 1982;19:22-5.
(5.) Primo-Parmo SL, Bartels CF, Lightstone H, Van der Spek AFL, La Du BN. Heterogeneity of the silent phenotype of human butyrylcholinesterase--identification of eight new mutations. In: Shafferman A, Velan B, eds. Multidisciplinary approaches to cholinesterase functions. New York: Plenum Press, 1992:61-4.
(6.) McGuire MC, Nogueria CP, Bartels CF, Lightstone H, Hajra A, Van der Spek AFL, et al. Identification of the structural mutation responsible for the dibucaine-resistant (atypical) variant form of human serum cholinesterase. Proc Natl Acad Sci U S A 1989;86: 953-7.
(7.) Nogueria CP, Bartels CF, McGuire MC, Adkins S, Lubrano T, Rubinstein HM, et al. Identification of two different point mutations associated with the fluoride-resistant phenotype for human butyrylcholinesterase. Am J Hum Genet 1992;51:821-8.
(8.) Sudo K, Maekawa M, Akizuki S, Magara T, Ogasawara H, Tanaka T. Human butyrylcholinesterase L330I mutation belongs to a fluoride-resistant gene, by expression in human fetal kidney cells. Biochem Biophys Res Commun 1997;240:372-5.
(9.) Primo-Parmo SL, Bartels CF, Wiersema B, Van der Spek AFL, Innis JW, La Du BN. Characterization of 12 silent alleles of the human butyrylcholinesterase (BCHE) gene. Am J Hum Genet 1996;58: 52-64.
(10.) Knedel M, Bottger R. [A kinetic method for determination of the activity of pseudocholinesterase (acylcholine acyl-hydrolase 18.104.22.168.)]. Klin Wochenschr 1967;45:325-7.
(11.) La Du BN, Bartels CF, Nogueria CP, Hajra A, Lightstone H, Van der Spek AFL, et al. Phenotypic and molecular biological analysis of human serum butyrylcholinesterase variants. Clin Biochem 1990; 23:423-31.
(12.) Athanassiadou A, Papachatzopoulou A, Gibbs RA. Detection and genetic analysis of [alpha]-thalassemia mutations by competitive oligopriming. Hum Mutat 1995;6:30-5.
(13.) Gibbs RA, Nguyen PN, Caskey CT. Detection of single DNA base differences by competitive oligonucleotide priming. Nucleic Acids Res 1989;17:2437-48.
(14.) Orou A, Fechner B, Utermann G, Menzel H-J. Allele-specific competitive blocker PCR: a one-step method with applicability to pool screening. Hum Mutat 1995;6:163-9.
(15.) Rosenthal A, Coutelle O, Craxton M. Large-scale production of DNA sequencing templates by microtitre format PCR. Nucleic Acids Res 1993;21:173-4.
(16.) Tu O, Knott T, Marsh M, Bechtol K, Harris D, Barker D, et al. The influence of fluorescent dye structure on the electrophoretic mobility of end-labeled DNA. Nucleic Acids Res 1998;26:2797802.
(17.) Whittaker M. Plasma cholinesterase abnormalities: laboratory investigations. In: Ellis FR, ed. Inherited disease and anaesthesia. Amsterdam: Elsevier North Holland Biomedical Press, 1981:10525.
(18.) Primo-Parmo SL, Lightstone H, La Du BN. Characterization of an unstable variant (BCHE115D) of human butyrylcholinesterase. Pharmacogenetics 1997;7:27-34.
(19.) Blong RM, Bedows E, Lockridge O. Tetramerization domain of human butyrylcholinesterase is at the C-terminus. Biochem J 1997;327:747-57.
(20.) Simeon-Rudolf V, Kovarik Z, Skrinjaric-Spoljar M, Evans RT. An explanation for the different inhibitory characteristics of human serum butyrylcholinesterase phenotypes deriving from inhibition of atypical heterozygotes. Chem Biol Interact 1999:119-20,15964.
(21.) Bartels CF, Jensen FS, Lockridge O, Van der Spek AFL, Rubinstein HM, Lubrano T, et al. DNA mutation associated with the human butyrylcholinesterase K-variant and its linkage to the atypical variant mutation and other polymorphic sites. Am J Hum Genet 1992;50:1086-103.
(22.) Dey DC, Maekawa M, Sudo K, Kanno T. Butyrylcholinesterase genes in individuals with abnormal inhibition numbers and with trace activity: one common mutation and two novel silent genes. Ann Clin Biochem 1998;35:302-10.
Department of Clinical Biochemistry, Royal Prince Alfred Hospital, Missenden Road, Camperdown, Sydney, NSW 2050, Australia.
* Author for correspondence. Fax 61-2-9515-7931; e-mail firstname.lastname@example.org.
 Nonstandard abbreviations: SC, succinylcholine; BCHE and BCHE, butyrylcholinesterase gene and enzyme (protein), respectively; nt, nucleotide; COP, competitive oligo-priming; dNTP, deoxynucleotide triphosphate; WT, wild type; MU, mutation; and SUPAMAC, Sydney University/Prince Alfred Macromolecular Analysis Centre.
Table 1. Primers used for COP-PCR and DNA sequencing. (1) Nucleotide Mutation position Sequence, 5'-3' Flu-1 POS 948-933 R (b) : GTT CAA CAT TCT GTT C Flu-1 NEG R: GTT CAA CGT TCT GTT C Flu-1 COM 849-864 F: AAG CCA TTC ATT GTT C Flu-2 POS 1376-1390 F: CCT TGG TTG ATG TTG Flu-2 NEG F: CCT TGG GTG ATG TTG Flu-2 COM 1447-1430 R: ATT CTG AGA ACT TCT TGG K-POS 1821-1836 F: TGA TGA AAC AGA ATG G K-NEG F: TGA TGA AGC AGA ATG G K-COM 1940-1923 R: AAT TAG AGA CCC ACA CAA Mutation POS-hex PCR NEG-hex PCR Flu-1 POS hex Unlabeled Flu-1 NEG Unlabeled hex Flu-1 COM Unlabeled Unlabeled Flu-2 POS hex Unlabeled Flu-2 NEG Unlabeled hex Flu-2 COM Unlabeled Unlabeled K-POS hex Unlabeled K-NEG Unlabeled hex K-COM Unlabeled Unlabeled (a) Sequence of human BCHE is GenBank accession no. M16541 J02964. Position +1 is the first nucleotide of the exon 1 (TAC; tyrosine, Y). Primers are labeled with hex fluorophore, as indicated, for the separate POS-hex PCR and NEG-hex PCR reactions (described in Materials and Methods). Each reaction involves three primers. (b) R, reverse primer; F, forward primer; COM, common primer. Table 2. Primer pairs and PCR annealing temperatures for BCHE sequencing analysis. (a) Nucleotide Primer position Sequence, (b) 5'-3' BC14 -55 to -32 (d) F: TAT AAT ATG CTA TAT GCA GAA GGC BC3 636-618 R: TAC AAT AAC TCT TTC AAC C Flu-1com 849-864 F: AAG CCA TTC ATT GTT C Flu-2com 1447-1430 R: ATT CTG AGA ACT TCT TGG BC11 568-592 F: CAA ACT GGA ACA TCA TCT TTA CAT G BC12 923-902 R: TAA AGA GAT GTT ACC GCC CAA G BC5 1296-1316 F: GAG TGA GTT TGG AAA GGA ATC BC13 40-17 (e) R: AAA CCA AGC CAG AGA ACA ATG AC BC8 1647-1661 F: GAA TCC AAA TGA GAC BC9 1813-1799 R: CTG TCA TTT CCA AGA K-5 -56 to -32 (f) F: CAG TTA ATG AAA CAG ATA AAA ATTT K-EXT 2066-2046 R: CAA TAT TAT CCT TCT GGC ATT T, (c) Primer Exon [degrees]C Length, bp BC14 2 46 570 BC3 Flu-1com 2 48 598 Flu-2com BC11 2 44 355 BC12 BC5 2 48 390 BC13 BC8 3 46 166 BC9 K-5 4 47 298 K-EXT (a) Sequence of human BCHE is GenBank accession no. M16541 J02964. Position +1 is the first nucleotide of exon 1 (TAC, tyrosine, Y). (b) F, forward; R, reverse. (c) PCR annealing temperature. (d) Primer annealing site is in intron 1; BC14 is numbered from the intron 1-exon 2 junction (the PCR product from BC14 and BC3 includes the intron-exon junction). (e) Primer annealing site is in intron 2 (the PCR product from BC5 and BC13 includes the exon 2-intron 2 junction). (f) Primer annealing site is in intron 3 (the PCR product of K5 and Kext includes the intron 3-exon 4 junction). Table 3. Summary of BCHE genotype diagnoses. Diagnosis n % Primary hypocholinesterasemia 52 80 AAKK (a) 23 AK 7 AAK 4 AKK 4 sil-1/ flu-1/ flu-2 4 Other mutations 3 New mutations 5 Mutation genotype with severe phenotype 2 WT genotype/normal enzyme activity 3 4.6 K-variant homozygous/low BChE activity 2 3.1 Dibucaine heterozygous/low BChE activity 1 1.5 WT genotype/low BChE activity 7 10.8 Total 65 (a) A denotes the dibucaine mutation; K denotes the K-variant mutation. Table 4. BCHE allele frequencies. (a) Allele frequency In Mutation Primary hypocholinesterasemia Patient population Dibucaine 0.72 0.54 K-variant 0.74 0.62 sil-1 0.010 0.008 sil-2 0.010 0.008 sil-7 0.029 0.023 flu-1 0.019 0.015 flu-2 0.019 0.015 (a) BCHE mutation allele frequency was determined by counting the abnormal alleles in the population: primary hypocholinesterasemia, n = 52; total patient population, n = 65. Table 5. BCHE genotypes: Group characteristics. Genotype (a) AAKK AK AAK AKK BChE activity, (e) kU/L Mean 3.32 8.13 3.1 6.25 Range 1.8-5.3 6.7-9.4 2.9-3.6 5.4-7.3 SD 0.89 1.29 0.34 0.82 Total patients, n 23 7 4 4 Males, n 13 2 0 2 Females, n 10 5 4 2 Race/Ethnicity (f) e2, me1 Three New mutations (b) mutations BChE activity, (e) kU/L Mean 2.1 Range 1.3-3.0 SD 0.75 Total patients, n 6 5 Males, n 4 3 Females, n 2 2 Race/Ethnicity (f) a1 Not sil/flu (a) primary (d) BChE activity, (e) kU/L Mean Range SD Total patients, n 3 12 Males, n 1 2 Females, n 2 10 Race/Ethnicity (f) e1 e2, a1 (a) Genotypes: AAKK, compound homozygous dibucaine/homozygous K-variant; AK, compound heterozygous dibucaine/heterozygous K-variant; AAK, compound homozygous dibucaine/heterozygous K-variant; AKK, compound heterozygous dibucaine/homozygous K-variant. (b) Three mutations of BCHE gene identified. (c) Genotyped with sil-1, flu-1, or flu-2 mutations. (d) Individuals with primary hypocholinesterasemia were excluded. (e) BChE activity reference interval, 7-16 kU/L. (f) Race/ethnicity (race, number of individuals): e, European; me, Middle Eastern; a, Asian. Table 6. BCHE mutation frequencies. (a) Primary Patient Hypocholinesterasemia population n % n % Group/category Homozygous dibucaine 28 54 28 43 Heterozygous dibucaine 19 36 20 31 WT/unaffected dibucaine 5 10 17 26 Mutation Dibucaine 47 90 48 74 K-variant 48 92 50 77 sil-1 1 1.9 1 1.5 sil-2 1 1.9 1 1.5 sil-7 3 5.8 3 4.6 flu-1 2 3.8 2 3.1 flu-2 2 3.8 2 3.1 Genotype profile Dibucaine and K-variant 46 88 46 71 Dibucaine or K-variant 49 94 52 80 Additional mutations 8 15 8 12 BCHE with three mutations 6 12 6 9 BCHE novel mutation 5 10 5 8 (a) Frequencies of BCHE mutations and mutation categories are shown for patients according to population. Primary hypocholinesterasemia, n = 52 patients; total patient population, n = 65. Fig. 6. Pedigrees of three patients with novel BCHE mutations identified. (A), patient 35 was compound heterozygous for two novel mutations: N106, a frame shift with adenine insertion, and R424STOP. (B), patient 56, a female child, was compound heterozygous for silent-7 (G115D) and a new mutation, E460K. (C), patient 39, a male child, was compound heterozygous for three mutations; dibucaine (D70G), K-variant (A539T), and a new mutation, F28I. n, wild type. Individual BChE N106FS R424STOP I-1 proband 0 hetero hetero II-1 daughter-1 9.4 hetero n II-2 daughter-2 7.1 hetero n Individual BChE E460K G11 5D (sil-7) K-var I-1 father 5.8 hetero n hetero I-2 mother 7.9 n hetero n II-1 brother-1 6.0 n hetero hetero II-2 brother-2 0 hetero hetero n II-3 proband 0 hetero hetero n Individual BChE dib F281 K-var I-1 father 8.1 n hetero n I-2 mother 6.4 hetero n hetero I-3 aunt 10.5 n n n I-4 uncle 5.6 hetero n hetero II-1 brother 8.0 n n n II-2 proband 2.0 hetero hetero hetero II-3 sister 5.6 hetero n hetero
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Molecular Diagnostics and Genetics|
|Author:||Yen, Tina; Nightingale, Brian N.; Burns, Jennifer C.; Sullivan, David R.; Stewart, Peter M.|
|Date:||Aug 1, 2003|
|Previous Article:||Increased homocysteine and S-adenosylhomocysteine concentrations and DNA hypomethylation in vascular disease.|
|Next Article:||Quantification of mitochondrial DNA deletion, depletion, and overreplication: application to diagnosis.|