Novel approach for CES1 genotyping: integrating single nucleotide variants and structural variation
Abstract
Aim: Development of a specific procedure for genotyping of CES1A1 (CES1) and CES1A2, a hybrid of CES1A1 and the pseudogene CES1P1. Materials & methods: The number of CES1A1 and CES1A2 copies and that of CES1P1 were determined using real-time PCR. Long range PCRs followed by secondary PCRs allowed sequencing of single nucleotide variants in CES1A1 and CES1A2. Results & conclusion: A procedure consisting of two main steps was developed. Its first main step, the copy number determination, informed about presence of CES1A2 . This information enabled choice of PCR in the second main step, which selectively amplified CES1A1 and, if present, also CES1A2, for subsequent sequencing. Examination of 501 DNA samples suggested that our procedure is specific with potential for personalization of drug treatments.
CES1 is an αβ-hydrolase fold enzyme belonging to the serine esterase superfamily [1]. This enzyme is a major contributor to the hydrolytic activity in the liver and metabolizes several commonly used therapeutic agents including methylphenidate [2], angiotensin-converting enzyme inhibitors [3,4], clopidogrel [5,6] and oseltamivir [7]. Moreover, CES1 has an important role in the detoxification of pesticides [8,9]. CES1 also appears to be involved in the hydrolysis of cholesterylesters, triglycerides, and other endogenous compounds [10,11] suggesting physiological functions of this enzyme.
The region on chromosome 16 harboring CES1 has a complex architecture, which primarily stems from gene duplication and the presence of a pseudogene, CES1P1, with a high degree of homology to CES1 [12]. Additionally, there is another CES1-related pseudogene, CES1P2, more proximal on chromosome 16. The gene duplication event has led to the formation of a hybrid gene, which was designated CES1A2, whereas CES1 commonly has been referred to as CES1A1. Both CES1A1 and CES1A2 have been redesignated as CES1 in the recommended nomenclature for mammalian carboxylesterase gene families [13]. However, to distinguish between them we have maintained the designations, CES1A1 and CES1A2, which we collectively refer to as CES1A in this publication.
The CES1A2 region from the promoter to the initial part of intron 1 is derived from CES1P1, whereas the rest of this gene comes from CES1A1. Sequence differences in the promoter between CES1A1 and CES1A2 result in a significantly lower transcriptional level of CES1A2 than CES1A1 [14]. However, a promoter haplotype of CES1A2 with two overlapping Sp1 binding sites and significantly increased transcriptional activity has been identified in one in five of this hybrid gene in a Japanese population [15].
Two subtypes of CES1A1, designated CES1A1b and CES1A1c, have been reported. Their exon 1 with or without the adjacent sequences were reported to be derived from CES1P1 [14]. CES1A1c, the most common of the two subtypes, is also known as CES1VAR, whereas CES1A1b is identical with or closely related to CES1SVAR [16].
Based on analysis of Caucasian, African–American and Japanese populations, two major structural CES1 haplotypes have been reported [12]. The first of these is composed of CES1P1 and CES1A1. In the second haplotype, CES1P1 has been replaced by CES1A2. Haplotypes with deletion of CES1A1 have not been reported and appear to be rare. The two major structural haplotypes can be subclassified into haplotypes A–D based on the absence or presence of subtype CES1A1c. For graphical presentation of the four haplotypes, see Figure 1.
There are two major structural haplotypes that both carry CES1, also known as CES1A1. The first of these also contains CES1P1. In the second major structural haplotype this pseudogene has been partially replaced by CES1A1 to form CES1A2. Based on the presence or absence of CES1A1c (CES1VAR), a variant of CES1A1 with a CES1P1 segment, these two major structural haplotypes can be subclassified, thus generating haplotypes A–D. CES1A1 and CES1P1 segments are shown in blue and magenta, respectively. Vertical bars and squares represent exons. Gray lines represent intergenic segments. Adapted from a previous publication [12].
A variety of different approaches have been used in the genotyping of CES1 including the 5′ exonuclease-based method [17,18], allele-specific PCR [19,20], amplification followed by sequencing [21,22], chip-based matrix-assisted laser desorption/ionization time-of-flight mass spectrometer [23] and the KASPar assay system [24]. Most of these approaches have not fully addressed the presence of CES1A2 in some individuals and the high degree of sequence similarity between CES1A1, CES1A2 and CES1P1. Accordingly, their specificity can be questioned. A more recent procedure based on long range amplification of a segment stretching from exon 1 to intron 6 of CES1A1 and CES1A2/CES1P1 identified the CES1A1 subtype, CES1A1c, as CES1A2 and was not able to distinguish CES1A2 from CES1P1, thus excluding genotyping of CES1A2 [25]. Collectively, this emphasizes that genotyping of CES1 is not a trivial matter and may fail, in particular with standard procedures.
Recently, we reported a procedure based on long range PCR for amplification of a 12.5 kb fragment stretching from the promoter to intron 5 in CES1A1 and CES1A2, respectively, followed by sequencing of amplified fragments [26]. We confirmed the specificity of this procedure including its ability to identify the CES1A1 variants CES1A1b and CES1A1c, and that it distinguished CES1A2 from CES1P1. Additionally, this procedure is capable of identifying the CES1A2 promoter variant with the Sp1 sites.
Here we present a novel and integrated PCR-based genotyping procedure that involves determination of the combined copy number of CES1A1 and CES1A2 to guide downstream amplifications including amplification of the 12.5 kb fragment of CES1A1 and CES1A2, respectively.
Materials & methods
DNA samples
Genomic DNAs from 400 individuals of Danish origin were included in the study. They had been isolated from samples of saliva or EDTA-stabilized blood. The samples of saliva were collected using the Oragene OG-250 DNA kit (DNA Genotek Inc., ON, Canada). We isolated genomic DNA from these samples using the prepIT L2P kit, produced by the manufacturer of the saliva collection kit. Isolation of genomic DNA from the blood samples was carried out using the Maxwell® 16 Instrument (Promega Corporation, WI, USA). We also obtained 45 and 56 samples of genomic DNA from Japanese and Han Chinese individuals, respectively, who had been enrolled in the 1000 Genomes Project (Coriell Institute for Medical Research, NJ, USA).
Determination of structural variation in CES1A1 & CES1P1
The combined number of copies of CES1A1 and CES1A2, that is, the number of CES1A copies, was determined. This was supplemented with copy number determination of CES1P1. The copy number determinations were carried out using commercially available reagents for duplex real-time PCR that quantitated the gene of interest and normalized it to an endogenous reference gene (Thermo Fisher Scientific, MA, USA). Two assays were used for determination of the combined copy number of CES1A1 and CES1A2. They targeted regions in intron 11 (TaqMan® copy number assay Hs00139541_cn with location 55,836,763–55,867,075 on chromosome 16) and intron 8 (TaqMan® copy number assay Hs03948988_cn with location 55,836,763–55,867,075 on chromosome 16) of CES1, respectively. Since CES1, also referred to as CES1A1, and CES1A2 are identical in these regions, both assays determine the combined number of copies of these two genes. For CES1P1, two assays, both targeting exon 6 of this pseudogene, were applied (TaqMan® copy number assay Hs01162042_cn and Hs01818263_cn with location 55,794,511–55,808,838 and 55,794,511–55,808,838 on chromosome 16, respectively). The gene encoding the ribonuclease P RNA component served as reference gene in the duplex real-time PCR (TaqMan® copy number reference assay RNase P, cat. #4403326 with location 20,811,565 on chromosome 14). The above chromosomal locations all refer to NCBI build 37. The assays were carried out according to the guidelines of the manufacturer of the reagents and calculations of the gene copy numbers were accomplished using CopyCaller v 1.0, a program developed by the reagent manufacturer. Information about the combined copy number of CES1A1 and CES1A2 was used to direct subsequent amplifications.
Primer design
We aligned the CES1 reference sequence, NG_012057.1, from the Homo sapiens chromosome 16 assembly (GRCh38.p2 primary assembly: NC_000016.10), with the CES1P1 and CES1P2 sequences from the same assembly. Also included in the alignments were CES1A2 (AB119998.1, complete coding sequence; and AB210090.1, promoter sequence). Regions that provided for a maximal difference between a desired sequence and the other sequences in the alignment were selected for design of primers. The specificity of the designed primers was assessed using Primer-BLAST [27] and the tool for in silico PCR tool in the UCSC genome browser [28].
Primers
Given that deletion of CES1A1 is exceedingly rare, it is valid to assume for genomes with two CES1A copies that both of these represent CES1A1 and that CES1A2 is absent. Accordingly, CES1A1 but not CES1A2 were amplified from such genomes. A series of primer pairs was used for the amplifications of CES1A1 and CES1A2. This included primer pairs for the selective amplification of fragments of 2.6 kb and 12.5 kb that stretched from the promoter to intron 1 and from the promoter to intron 5, respectively, of the two genes [26]. In order to amplify the region from intron 5 to the 3′ untranslated region (3′ UTR) of CES1A1, a primer pair amplifying a 19.2 kb fragment was designed. The locations of the primers for the amplification of the 2.6, 12.5 and 19.2 kb fragments are shown in Figure 2.
Two primer pairs amplify a segment of 12.5 kb stretching from the promoter to intron 5 in CES1A1 and CES1A2. Moreover, a 19.2 kb fragment of CES1A1 is amplified that stretches from intron 5 to the 3′ UTR. Two other primer pairs amplify a fragment of 2.6 kb stretching from the promoter to intron 1 in CES1A1 and CES1A2, respectively. The 12.5 and 19.2 kb fragments are used as template in nested PCR to amplify and sequence a promoter segment and the coding regions. The 2.6 kb fragments of CES1A1 and CES1A2 are sequenced directly. CES1A1 and CES1P1 segments are shown in blue and red, respectively. The vertical bars represent exon 1–14. Note that CES1A1and CES1A2 are shown in same transcriptional direction.
We also designed four primer pairs to amplify fragments that ranged from 375 to 659 bp in size and contained each of the exons 2–5 of CES1A1 and CES1A2. Since these primer pairs did not differentiate between CES1A1 and CES1A2, they were used in nested PCRs with the 12.5 kb amplicons of CES1A1 and CES1A2, respectively, as template. Additionally, they were used for amplification using genomic DNA from individuals lacking CES1A2 as template.
A set of primer pairs was exclusively used for nested PCR with the 19.2 kb CES1A1 amplicon as template. This included a primer pair for the combined amplification of exon 6 and 7 of this gene and another that was used to amplify its exon 8. Sequences of the other primers for nested PCRs of CES1A1, that is, the primers for amplification of exon 9–14 derived from a previous study [22]. All primer sequences are listed in Supplementary Table 1.
Amplifications
The amplifications of the 2.6 and 12.5 kb fragments of CES1A1 and CES1A2 as well as the 19.2 kb CES1A1 fragment were done using Herculase hotstart DNA polymerase (Agilent Technologies, Stratagene Products Division, CA, USA). This is an enzyme preparation, which consists of Pfu DNA polymerase, a small amount of Taq2000 DNA polymerase, and a deoxyuridine triphosphatase (DUT). The latter removes 2′-deoxyuridine-5′-triphosphate formed by deamination of 2′-deoxycytidine-5′-triphosphate during PCR that would decrease strand elongation if incorporated. The amplifications were carried out in line with the guidelines of the manufacturer of the Herculase hotstart DNA polymerase with minor modifications, which included incorporation of additional MgCl2 in the reaction mixture for amplification of the 12.5 and 19.2 kb fragments and amplifications in volumes of 25 μl instead of 50 μl. We also used the Elongase® Enzyme Mix, which does not contain a DUT, for amplification of long fragments (Thermo Fisher Scientific, MA, USA). The amplifications using the Elongase® Enzyme Mix were done in accordance with the guidelines provided by the manufacturer of this enzyme mix.
Amplifications of fragments other than the 2.6, 12.5 and 19.2 kb fragments were performed using AmpliTaq Gold® DNA Polymerase (Thermo Fisher Scientific). For this purpose genomic DNA or the amplicons of 12.5 or 19.2 kb served as template after purification of the amplicons using the NucleoFast® 96 PCR purification Kit (Macherey–Nagel GmbH & Co KG, Düren, Germany).
Concentration of primers, template and annealing temperature were varied in order to determine the conditions that provided for the highest possible yield of the desired fragment without compromising the specificity or robustness of the amplifications. For composition of reaction mixtures and temperature cycling, see Supplementary Tables 3–11.
The primers for amplification using the proofreading DNA polymerase, were synthesized with a varying number of phosphorothioate (PTO) bonds to increase resistance toward the 3′-5′ nuclease activity of the DNA polymerase that could otherwise compromise primer specificity by permitting elongation from mismatched primer–template complexes.
Electrophoretic analysis
Agarose gel electrophoresis was used to confirm that fragments of the expected sizes were amplified. We also used agarose gel electrophoresis as a mean to determine the reactions conditions that produced high yields of the fragments of the expected sizes without amplification of fragments of other sizes, that is, undesired fragments.
Sanger sequencing
Amplified products were purified using the QIAquick 96 PCR Purification Kit (Qiagen NV, Venlo, The Netherlands) and subjected to sequencing by the Sanger dideoxy chain termination method. For this purpose the amplification primers were used as sequencing primers. Primers specifically dedicated to sequencing were also used; see Supplementary Table 2.
Analysis of sequences
Sequencing chromatograms were visually inspected for traces of mixed sequences suggesting presence of two or more templates and lack of specificity. Subsequently, sequences were aligned with the CES1 reference sequence, NG_012057.1, from the Homo sapiens chromosome 16 assembly (GRCh38.p2 primary assembly: NC_000016.10), the CES1P1 sequence from the same assembly and the CES1A2 sequence with accession number AB119998.1. Also included in the alignments were the sequences with the accession numbers AB210087.1 (CES1A1a), AB210088.1 (CES1A1b), AB210089.1 (CES1A1c) and AB210090.1 (CES1A2). The promoter variant of CES1A2 with increased transcriptional activity was identified based on sequence data from a previous study [15].
Results
We developed an integrated approach in which information from the copy number determinations of CES1A and CES1P1 was used to select the subsequent amplifications aimed at identification and genotyping of single nucleotide variants (SNVs) in CES1A1 and CES1A2 by sequencing. An overview of the procedure is shown in Figure 3. Our approach consisted of two main steps. The first of these steps was the copy number determinations in which the CES1P1 copy number served to confirm the CES1A copy number since the total number of copies of CES1A1, CES1A2 and CES1P1 adds to four [12]. The second main step consisted of the amplification of different PCR fragments depending upon the absence or presence of CES1A2.
The combined copy number of CES1A1 and CES1A2 is used to inform about the presence or absence of CES1A2 and select downstream amplifications. Samples with a copy number of two are not expected to contain CES1A2. A 12.5 kb fragment stretching from the promoter to intron 5 and a 19.2 kb fragment stretching from intron 5 to the 3′ UTR of CES1A1 are amplified from such samples. The 12.5 kb CES1A1 fragment and the corresponding CES1A2 fragment are amplified from genomes with a copy number of three or four.
For genomes with a CES1A copy number of two, that is, genomes lacking CES1A2, we amplified two CES1A1 fragments of 12.5 and a 19.2 kb that stretched from the promoter to intron 5 and from intron 5 to the 3′ UTR of the gene, respectively. Due to sequence similarity between CES1A1 and CES1A2, the reaction for production of the 19.2 kb CES1A1 fragment also amplified the corresponding CES1A2 fragment, thus not being applicable to the analysis of genomes with a combined CES1A1 and CES1A2 copy number of three or four. Hence, amplification using genomes with such copy numbers as template was restricted to the production of the 12.5 kb fragments of CES1A1 and CES1A2.
The fragments obtained by the long PCRs were used as templates in nested PCRs for secondary amplification of a promoter fragment and the coding regions. Using as template the 12.5 kb fragment of CES1A1 and CES1A2, respectively, we amplified the 2.6 kb fragment containing a promoter segment and exon 1, which allowed for the identification of CES1A1c, and fragments with sizes that ranged from 375 to 659 bp and contained exon 2–5. The 19.2 kb CES1A1 fragment served as template for nested PCRs to amplify exon 6–14. Accordingly, all CES1A1 exons and a portion of the promoter were amplified from individuals lacking CES1A2.
We replaced the 12.5 kb amplicon with genomic DNA as template in the reaction mixtures for amplification of the 2.6 kb fragment containing the promoter and exon 1 of CES1A1 and CES1A2. Moreover, we amplified the CES1A1 exons 2–5 in separate reaction using genomic DNA from individuals lacking CES1A2 as template. Since the primer pairs for amplification of these exons did not differentiate between CES1A1 and CES1A2, they were not applicable for amplification of samples harboring CES1A2 with genomic DNA as template.
Primer sequences, amplicons, optimized reaction mixtures and optimized cycling conditions are listed in Supplementary Tables 1 & 3–11.
The use of PTO-modified nucleotides successfully prevented unspecific amplification in the reactions designed for production of the 2.6, 12.5 kb and 19.2 kb amplicons. Up to four PTO-modified nucleotides in the 3′ end of the primers were required to prevent unspecific amplification effectively.
The enzyme preparation containing DUT produced significantly larger amounts of the 2.6, 12.5 and 19.2 kb fragments as compared with the preparation without it. Hence, the former of these enzyme mixtures was preferred.
The specificity of the amplifications was assessed by different means. This included analysis of the CES1A1 primers by in silico PCR, which predicted amplification of the desired fragments from this gene. The in silico PCR did not predict product outcomes for the primer pairs designed to amplify the 2.6 and 12.5 kb fragments of CES1A2. This was expected as CES1A2 is not included in the current or earlier versions of the build of the human genome used by the in silico PCR programs for product outcome prediction. The gel electrophoretic analysis revealed that the amplifications all produced fragments of the expected sizes. Assessment of the amplification specificity by inspection of the sequencing chromatograms did not reveal traces of mixed sequences reflecting co-amplification of undesired fragments such as CES1P1 fragments. Moreover, the sequences of the amplified fragments aligned to the genes, which they had been designed to amplify. We also found that samples with a combined number of CES1A1 and CES1A2 copies of two did not support amplification of the 2.6 and 12.5 kb fragments using the CES1A2 primers, while these primers amplified fragments of the expected sizes from genomes with a total of three or four of these copies (Figure 4).
Lane 1: sample with a CES1A copy number of two subjected to amplification using the CES1A1 primer pair. Lane 2: sample with a CES1A copy number of three subjected to amplification using the CES1A1 primer pair. Lane 3: sample with a CES1A copy number of two subjected to amplification using the CES1A2 primer pair. Lane 4: sample with a CES1A copy number of three subjected to amplification using the CES1A2 primer pair. Note that the sample with a CES1A copy number of two does not support amplification using the CES1A2 primer pair. The CES1A copy number refers to the combined number of copies of CES1A1 and CES1A2. To the left fragments of a DNA molecular weight marker in base pairs. For primers, see Supplementary Table 1.
We used the newly developed approach for examination of 501 samples and found that the copy numbers of CES1A and CES1P1 were unambiguously determined in 495 out of these samples. The remaining samples were discarded due to poor DNA quality. The CES1A copy number was found to range from two to four while that of CES1P1 ranged from zero to two with the total number of copies of CES1A1, CES1A2 and CES1P1 amounting to four. All of the examined samples supported amplification with the primers designed to amplify the 2.6 and 12.5 kb fragments of CES1A1, which is in line with the observation that deletion of CES1A1 is rare [12]. On average about 97% of the amplicons produced readable sequences. We identified several previously reported CES1A1 variations such as p.Gly143Glu (rs121912777), which has been associated with significantly decreased catalytic activity [18], subtype CES1A1c [14] and the CES1A2 variant with the two overlapping Sp1 sites conferring increased transcriptional activity and potentially rapid drug metabolism [15]. For Sanger sequencing chromatogram of CES1A1 p.Gly143Glu, see Figure 5.
Note a GG homozygote (upper lane), a GA heterozygote (middle lane) and an AA homozygote (lower lane) at the position denoted with an R. The two peaks of the heterozygote at this position are of almost half the heights of the homozygote peaks at the same position.
We compared findings from the present study with those obtained by a previous study, which examined the diversity of the gene encoding CES1 in a European population and reported a total of 16 SNVs in the region from the promoter to the 3′ UTR of this gene without distinguishing between CES1A1 and CES1A2 [21]. We detected a significantly higher number of SNVs, namely, 46 and 21 in CES1A1 and CES1A2, respectively, in our Danish population. This included 15 SNVs in the coding regions of CES1A1 and CES1A2 of which five were detected in CES1A1 exon 1 at frequencies of about 0.20, namely, rs111604615, rs566557773, rs201577108, rs114788146 and rs12149366. None of these five SNVs, which all are a part of CES1A1c (CES1VAR) [16], were detected in the previous study [21]. We are not aware of other studies of the diversity of CES1A1 in European populations. Comparison of our findings with those of the 1000 Genomes Project [29] also revealed major differences; see Table 1. Notably, the 1000 Genomes Project reported minor allele frequencies (MAFs) in the range from 0.004 to 0.080 for the SNVs that define CES1A1c in exon1 in European and east Asian populations, whereas we found MAFs of up to 0.29 for some of these SNVs in the same populations.
| Location† | dbSNP reference | Minor allele frequency | |||
|---|---|---|---|---|---|
| Our procedure | 1000 Genomes Project‡ | ||||
| Europeans (Danes) | Chinese & Japanese | Europeans | East Asians | ||
| 5′ UTR | rs12149373 | 0.209 | 0.271 | 0.164 | 0.221 |
| 5′ UTR | rs12149371 | 0.216 | 0.284 | 0.176 | 0.218 |
| 5′ UTR | rs12149322 | 0.210 | 0.280 | 0.082 | 0.110 |
| 5′ UTR | rs12149370 | 0.212 | 0.280 | 0.082 | 0.110 |
| 5′ UTR | rs12149368 | 0.212 | 0.273 | 0.053 | 0.072 |
| Exon 1 | rs111604615 | 0.210 | 0.270 | 0.004 | 0.011 |
| Exon 1 | rs566557773 | 0.211 | 0.270 | 0.025 | 0.054 |
| Exon 1 | rs201577108 | 0.211 | 0.270 | 0.025 | 0.054 |
| Exon 1 | rs114788146 | 0.211 | 0.270 | 0.035 | 0.056 |
| Exon 1 | rs12149366 | 0.211 | 0.276 | 0.067 | 0.080 |
| Intron 1 | rs12149359 | 0.208 | 0.273 | 0.125 | 0.156 |
Discussion
We report the development of a novel approach with a low drop-out rate for genotyping of CES1A1 (CES1). This development is based on copy number determination to deduce the structural haplotypes and guide subsequent amplification of fragments of CES1A1 and CES1A2. Sequencing of these fragments permits the identification of SNVs and other small-scale variations. Our approach distinguishes itself by the application of copy number determination and its ability to selectively amplify fragments of CES1A1 and CES1A2. For example, it allowed the identification of the CES1A2 variant with the two overlapping Sp1 sites conferring increased transcriptional activity and potentially rapid drug metabolism.
The number of copies of CES1A1, CES1A2 and CES1P1 adds to four with two of these being CES1A1 since deletion of CES1A1 appears to be extremely rare [12]. This ‘four rule’ allowed us to use the copy number of CES1P1 to validate the results from the copy number determination of CES1A1 and CES1A2. This is useful as the number of copies of a gene cannot always be determined unambiguously due to inferior DNA quality or residual impurities. For example, it is occasionally unclear whether a gene is present in two or three copies reflecting that a copy number of three differs from a copy number of two by a factor of only 1.5. By contrast, samples with one copy of a gene can better be discriminated from samples having two copies, since a copy number of one differs from a copy number of two by a factor of two. According to the ‘four rule’, CES1P1 copy numbers of one and two correspond to a combined CES1A1 and CES1A2 copy number of three and two, respectively.
Challenges to the selective amplification of the fragments of CES1A1 and CES1A2 are the high sequence identity between these two genes and their sequence identity with CES1P1. Another challenge is the requirement for a proofreading DNA polymerase in the amplification of the long fragments since such a polymerase repairs mismatches between a primer and a template, thus potentially permitting strand elongation and amplification with concomitant loss of specificity. This was overcome by the use of primers modified with PTO to provide nuclease resistance. Additionally, the enzyme mixture preferred for amplification of the large fragments contained a DUT, which removes 2′-deoxyuridine-5′-triphosphate during PCR, thus promoting strand elongation [31]. This enzyme mixture greatly enhanced the amplification of the 2.6, 12.5 and 19.2 kb fragments compared with a preparation without DUT and enabled a robust amplification of these fragments.
Since the majority of hepatic CES1 is produced by CES1A1 [12,14], variation in this gene would be expected to have a significantly larger impact on the enzyme activity than variation in CES1A2. Hence, it is critical to distinguish between CES1A1 and CES1A2 in the assessment of the functional impact of a CES1A variant. This requires a genotyping approach with high resolution and specificity, which was provided by our approach.
Several of the previously reported procedures for genotyping of CES1A1 and CES1A2 appear not fully to have taken the complex pattern of variants into consideration. This included a study that reported a variety of SNVs in CES1 based on amplification followed by sequencing of amplicons [21]. Using in silico PCR we recently found that all of the primers from this study amplified CES1A1 as well as CES1A2, thus producing mixed amplicons [34]. In line with this, we found major differences in the number of SNVs and their MAFs between our study and the previous study [21], although some of these differences may be explained by the amplification and examination of fragments that were overlapping but not identical and differences in the number of subjects in the two studies. Another study examined variation in CES1 and CES1P1 also without distinguishing between CES1A1 and CES1A2, thus potentially leading to inaccurate variant detection [22]. Moreover, the design and location of our primers represents an advantage compared with the previous procedure based on long PCR [25], which did not distinguish CES1A2 from CES1P1 and also lacked the ability to distinguish CES1A2 from CES1A1c.
The specificity of the amplifications in our method was assessed by in silico PCR, agarose gel electrophoresis and Sanger sequencing. We also examined whether genomes without CES1A2 supported amplification using CES1A2 primer pairs. The observations achieved by these means supported each other and indicated that our amplification reactions were specific with the ability to selectively amplify CES1A1 and CES1A2, respectively, without co-amplification of CES1-related pseudogenes or other undesired fragments.
A limitation to our approach is that it is not able to genotype and identify nucleotide variants from intron 5 and downstream in CES1A1 and CES1A2 in individuals carrying both of these genes. Hence, potentially damaging SNVs may not be detected in some individuals. The inability to detect variants in the downstream regions of CES1A1 and CES1A2 in individuals carrying both of these genes reflects their sequence similarity.
The complexity in the chromosomal region harboring the CES1 locus with the presence of a hybrid of CES1P1 and CES1 in some individuals has a parallel in the gene encoding CYP2D6. That is, hybrids of CYP2D6 and the pseudogene CYP2D7 have been reported to complicate the genotyping of CYP2D6, thus requiring specifically designed assays [32]. Additionally, a variation in pseudogene CYP2D7 has been shown to permit binding of a primer for identification of CYP2D6 alleles resulting in false positive calls for these alleles [33]. Whether this also is a problem in the typing of CES1 is not yet known.
An accurate procedure for detection of variants in CES1 may be valuable as a tool to assist in predicting efficacy and risk of adverse reactions of therapeutic agents metabolized by CES1 and reveal individual predisposition to illicit drugs and toxicants that are inactivated by this enzyme. It may also be helpful to provide new insights into the genetics of lipid metabolism and obesity.
We developed such procedure, which takes the structural variation of CES1 into consideration and integrates determination of structural variation with genotyping of SNVs and other small-scale variants. This permits correct annotation of SNVs in CES1A1 and CES1A2 and enables identification of the subtypes of CES1A1 and the CES1A2 variant associated with increased transcriptional activity. We propose our procedure to become a standard in the genotyping of CES1.
Conclusion
We developed a procedure for genotyping of CES1A1 (CES1) and CES1A2, a hybrid of CES1A1 and the CES1-related pseudogene CES1P1. This procedure is based on copy number determination to inform about the presence or absence of CES1A2 in an individual followed by amplification of fragments for genotyping of small-scale variants by Sanger sequencing. Our procedure appeared to be highly specific and distinguished itself from most previous procedures by taking into account the presence of CES1A2 in some individuals and the large degree of sequence homology between CES1A1, CES1A2 and CES1P1.
Future perspective
Development of methods for genotyping of CES1A1 and CES1A2 with high specificity permits a reappraisal of the diversity of these genes and inclusion of more accurate data in genetic variant databases. Future studies are also believed to address the lack in the ability to genotype a large portion of the 3′ ends of CES1A1 and CES1A2 in individuals carrying both of these genes. The development and validation of novel analyses for genotyping of CES1A1 and CES1A2 would be facilitated by detailed characterization of these genes in selected samples that could be included in genetic resources repositories as reference samples.
Note
After initiation of the present method development the single nucleotide substitution rs74518148 was reported. The variant nucleotide at this polymorphic site creates a mismatch with our forward primer for amplification of the 2.6 and 12.5 kb fragments of CES1A1 (CES1). Since this mismatch is located in the 5′ end of the primer and the minor allele of rs74518148 has a low frequency, we do not anticipate it to affect the amplification outcome. However, we now synthesize the primer concerned with mixed bases (underlined) to achieve complete match with both alleles at rs74518148: 5′- ACTAYGGGGGGACGGAGTTCA-3′.
Executive summary
Background
CES1A1 (CES1) is a pharmacogenetic candidate gene and encodes a hydrolase, which is implicated in the metabolism of commonly used drugs.
The chromosomal region harboring CES1A1 is complex due to the existence of a pseudogene, CES1P1, and a structural variant known as CES1A2, which is a hybrid of CES1A1 and CES1P1.
Several previous genotyping methods have not been able to distinguish between CES1A1, CES1A2 and CES1P1.
We aimed at developing a specific method for the genotyping of CES1A1 and CES1A2.
Materials & methods
Real-time PCR served to detect the presence or absence of CES1A2 in a sample.
Primer pairs amplifying 2.6 and 12.5 kb fragments of CES1A1 and CES1A2 in addition to a primer pair producing a 19.6 kb CES1A1 amplicon, were designed in regions of these genes that enabled highest possible degree of specificity.
Amplification primers containing varying numbers of phosphorothioate bonds in their 3′ ends and different DNA polymerases were tested.
Amplicons were sequenced by the Sanger method.
Samples of DNA from 400 individuals residing in Denmark and 101 individuals from the 1000 Genomes Project were used for development of the new procedure.
Results
We developed a procedure for genotyping of CES1A1 and CES1A2 consisting of two main steps with copy number determination by real-time PCR as the first main step serving to guide amplification of fragments in the second main step of the procedure.
In the main second step fragments were amplified for Sanger sequencing using a proofreading DNA polymerase and primers containing phosphorothioate bonds, which prevented loss in specificity as a result of repair of mismatches between the primers and undesired targets.
Our procedure appeared to be specific with the ability to distinguish CES1A1 from CES1A2 but is unable to genotype variants in the downstream half of CES1A1 and CES1A2 in individuals carrying both of these genes.
Discussion
Structural variation complicates the genotyping of single nucleotide variations in CES1A1 and CES1A2 markedly.
Our procedure opens up for a more accurate genotyping of CES1A1 and CES1A2 and may be helpful in the individualization of treatments with drugs metabolized by CES1.
Supplementary data
To view the supplementary data that accompany this paper please visit the journal website at:
https://www.futuremedicine.com/doi/suppl/10.2217/pgs-2016-0145
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
References
- 1 Relationship between sequence conservation and three-dimensional structure in a large family of esterases, lipases, and related proteins. Protein Sci. 2(3), 366–382 (1993).
- 2 Methylphenidate is stereoselectively hydrolyzed by human carboxylesterase CES1A1. J. Pharmacol. Exp. Ther. 310(2), 469–476 (2004).
- 3 Hydrolytic profile for ester- or amide-linkage by carboxylesterases pI 5.3 and 4.5 from human liver. Biol. Pharm. Bull. 20(8), 869–873 (1997).
- 4 . Role of carboxylesterase 1 and impact of natural genetic variants on the hydrolysis of trandolapril. Biochem. Pharmacol. 77(7), 1266–1272 (2009).
- 5 A possible mechanism for the differences in efficiency and variability of active metabolite formation from thienopyridine antiplatelet agents, prasugrel and clopidogrel. Drug Metab. Dispos. 37(11), 2145–2152 (2009).
- 6 . Carboxylesterase 1 as a determinant of clopidogrel metabolism and activation. J. Pharmacol. Exp. Ther. 344(3), 665–672 (2013).
- 7 Anti-influenza prodrug oseltamivir is activated by carboxylesterase human carboxylesterase 1, and the activation is inhibited by antiplatelet agent clopidogrel. J. Pharmacol. Exp. Ther. 319(3), 1477–1484 (2006).
- 8 . Hydrolytic metabolism of pyrethroids by human and other mammalian carboxylesterases. Biochem. Pharmacol. 71(5), 657–669 (2006).
- 9 . Human carboxylesterases and their role in xenobiotic and endobiotic metabolism. J. Biochem. Mol. Toxicol. 21(4), 187–196 (2007).
- 10 . Role of endoplasmic reticulum neutral lipid hydrolases. Trends Endocrinol. Metab. 22(6), 218–225 (2011).
- 11 . Inactivation of lipid glyceryl ester metabolism in human THP1 monocytes/macrophages by activated organophosphorus insecticides: role of carboxylesterases 1 and 2. Chem. Res. Toxicol. 23(12), 1890–1904 (2010).
- 12 Structure and characterization of human carboxylesterase 1A1, 1A2, and 1A3 genes. Pharmacogenet. Genomics 18(10), 911–920 (2008).
- 13 Recommended nomenclature for five mammalian carboxylesterase gene families: human, mouse, and rat genes and proteins. Mamm. Genome. 21(9–10), 427–441 (2010).
- 14 . Human carboxylesterase 1A2 expressed from carboxylesterase 1A1 and 1A2 genes is a potent predictor of CPT-11 cytotoxicity in vitro. Pharmacogenet. Genomics 17(1), 1–10 (2007).
- 15 Functional polymorphisms in carboxylesterase 1A2 (CES1A2) gene involves specific protein 1 (Sp1) binding sites. Biochem. Biophys. Res. Commun. 369(3), 939–942 (2008).
- 16 Regulatory effects of genomic translocations at the human carboxylesterase-1 (CES1) gene locus. Pharmacogenet. Genomics 26(5), 197–207 (2016).
- 17 . Carboxylesterase 1 gene polymorphism and methylphenidate response in ADHD. Neuropharmacology 57(7–8), 731–733 (2009).
- 18 Two CES1 gene mutations lead to dysfunctional carboxylesterase 1 activity in man: clinical significance and molecular basis. Am. J. Hum. Genet. 82(6), 1241–1248 (2008).
- 19 . Mutation-sensitive molecular switch method to detect CES1A2 mutation in the Chinese Han and Yao populations. Genet. Test Mol. Biomarkers 15(9), 659–662 (2011).
- 20 A simple allele-specific polymerase chain reaction method to detect the Gly143Glu polymorphism in the human carboxylesterase 1 gene: importance of genotyping for pharmacogenetic treatment. Genet. Test. Mol. Biomarkers 14(6), 749–751 (2010).
- 21 Pharmacogenomic assessment of carboxylesterases 1 and 2. Genomics 84(4), 661–668 (2004).
- 22 Genetic variation in carboxylesterase genes and susceptibility to isoniazid-induced hepatotoxicity. Pharmacogenomics J. 10(6), 524–536 (2010).
- 23 . CES1A -816C as a genetic marker to predict greater platelet clopidogrel response in patients with percutaneous coronary intervention. J. Cardiovasc. Pharmacol. 63(2), 178–183 (2014).
- 24 Methylphenidate side effect profile is influenced by genetic variation in the attention-deficit/hyperactivity disorder-associated CES1 gene. J Child Adolesc. Psychopharmacol. 23(10), 655–664 (2013).
- 25 . A discriminative analytical method for detection of CES1A1 and CES1A2/CES1A3 genetic variants. Pharmacogenet. Genomics 22(3), 215–218 (2012).
- 26 ; INDICES Consortium. Novel procedure with improved resolution and specificity for amplification and differentiation of variants of the gene encoding carboxylesterase 1. Pharmacogenet. Genomics 27(4), 155–158 (2017).
- 27 . Primer-BLAST: a tool to design target-specific primers for polymerase chain reaction. BMC Bioinformatics 13, 134 (2012).
- 28 The UCSC genome browser database: 2016 update. Nucleic Acids Res. 44(D1), D717–D725 (2015).
- 29 1000 Genomes Project Consortium; A global reference for human genetic variation. Nature 526(7571), 68–74 (2015).
- 30 Database of Single Nucleotide Polymorphisms (dbSNP). Bethesda (MD): National Center for Biotechnology Information, National Library of Medicine. (dbSNP Build ID: 151). www.ncbi.nlm.nih.gov/SNP/.
- 31 . Archaeal dUTPase enhances PCR amplifications with archaeal DNA polymerases by preventing dUTP incorporation. Proc. Natl Acad. Sci. USA 99(2), 596–601 (2002).
- 32 . CYP2D7-2D6 hybrid tandems: identification of novel CYP2D6 duplication arrangements and implications for phenotype prediction. Pharmacogenomics 11(1), 43–53 (2010).
- 33 CYP2D7 sequence variation interferes with TaqMan CYP2D6*15 and *35 genotyping. Front. Pharmacol. 6, 312 (2015).
- 34 . Carboxylesterase 1 genes – systematical review and evaluation of existing genotyping procedures. Drug Metab. Pers. Ther. (2018) (Epub ahead of print).
