Clinical Chemistry 43: 40-44, 1997;
(Clinical Chemistry. 1997;43:40-44.)
© 1997 American Association for Clinical Chemistry, Inc.
Ligase chain reaction assay for human mutations: the Sickle Cell by LCR assay
Antonio A. Reyes1,
Paola Carrera2,
Elena Cardillo2,
Luis Ugozzoli1,
Jimmie D. Lowery1,
Ching-I P. Lin1,
Matthew Go3,
Maurizio Ferrari2 and
R. Bruce Wallace1,a
1
DNA Diagnostics Business Unit, and
2
IRCCS Clinical Molecular Biology Laboratory, H. San Raffaele, via Olgettina 60, 20132 Milan, Italy.
3
Clinical Diagnostics Group, Bio-Rad
Laboratories, 2000 Alfred Nobel Dr., Hercules, CA 94547.
a Author for correspondence. Fax 510-741-5811; e-mail bwallace{at}bio-rad.com
 |
Abstract
|
|---|
We can detect the ß-globin gene sickle cell mutation by using an assay
based on the ligase chain reaction. The simultaneous amplification of
the human growth hormone gene in the same reaction serves as a control
for the amount of template DNA or amplification efficiency. Ligation
products, which are biotinylated at one end and tagged with an
arbitrary "tail" sequence at the other, are captured by
hybridization to "tail"-complementary oligonucleotides immobilized
on polystyrene microwells. The captured ligation products are detected
colorimetrically by use of streptavidin–alkaline phosphatase
conjugate. In a study of 24 subjects, the assay unequivocally
discriminated among normal, carrier, and sickle cell genotypes.
Key Words: indexing terms: hemoglobinopathies • DNA amplification • genetic diseases • human growth hormone gene • biotin–streptavidin interaction
|
Introduction
|
|---|
Ligase chain reaction (LCR) and other ligation-based assays such
as gap LCR and the oligonucleotide ligation assay have been used
independently or coupled to the polymerase chain reaction to diagnose
genetic and infectious diseases
(1)(2).1
LCR
is a DNA template-dependent amplification reaction that utilizes two
pairs of oligonucleotides—one pair complementary to the upper template
strand, and the other pair complementary to the lower template strand.
Each member of a pair hybridizes to adjacent positions on the template
such that the 5'-phosphate of one oligonucleotide abuts the 3'-hydroxyl
of the other. The resulting nick is sealed by DNA ligase. The product
of one round of ligation can serve as a template for a subsequent
cycle; thus, using a thermostable ligase and performing sequential
cycles of denaturation and ligation result in an exponential
accumulation of product. Because a mismatch in the ligation junction
inhibits the ligation reaction, LCR is ideal for the diagnosis of known
mutations.
Sickle cell anemia is caused by a single-base mutation in the
ß-globin gene in which codon 6 is no longer GAG (producing
Glu; normal or A allele) but instead is GTG (producing Val;
S allele). The variant HbS polypeptide is routinely detected by
electrophoretic and HPLC procedures (3)(4). In
newborns, detection of Hb variants at an initial screening is usually
followed by confirmatory testing after several months, when the
proportion of HbF has decreased. In some cases, the phenotypes or
genotypes of other family members must be evaluated to reach an
accurate diagnosis.
We have developed the Sickle Cell by LCR assay, a robust, colorimetric
assay for typing A/A (normal), A/S (sickle cell carrier) and S/S
(sickle cell disease-affected) subjects. DNA isolated from blood is
tested for the presence of the A and S alleles. False negatives due to
inadequate amount of starting DNA sample are readily identified by lack
of coamplification of the human growth hormone (hGH) gene included in
the same assay. The ligation products are detected in a 96-well
microplate format. The high sequence-specificity of the LCR assay
allows genotypic assignment based on the ratio of the A allele-specific
signal to the S allele-specific signal.
|
Materials and Methods
|
|---|
oligonucleotides
Oligonucleotides were purchased from Integrated DNA Technologies
(Coralville, IA), Oligos Etc. (Wilsonville, OR), and Operon
Technologies (Alameda, CA). Oligonucleotide sequences are given in
Table 1
. The A allele oligonucleotide set comprises MD157, 159, 160,
and 162; the S allele oligonucleotide set, MD159, 161, 162, and 166;
and the hGH oligonucleotide set, MD111, 168, 169, and 170. All
oligonucleotide modifications were performed by the respective vendors
during synthesis. Oligonucleotides MD159 and MD111 were biotinylated by
incorporating biotin-dT at the penultimate 3'-end position during
synthesis. Oligonucleotides MD159, 162, 111, and 169 were chemically
phosphorylated at the 5'-end by methods used by the vendor. We purified
the LCR oligonucleotides by denaturing acrylamide gel electrophoresis
(5), then analyzed the purified LCR oligonucleotides on a
second denaturing gel to confirm that they consisted of a single band.
We determined oligonucleotide concentrations were determined by
ultraviolet absorbance and by use of the molar absorptivities
calculated for each oligonucleotide as described elsewhere
(6).
purified genomic dnas
Human placental DNA (presumptive A/A genotype) was purchased from
Sigma, St. Louis, MO; cell line CRL8756 (S/S genotype) from the
American Type Culture Collection, Rockville, MD; and cell lines GM08779
(A/S genotype) and GM02267 (homozygous deletion of the ß-globin gene)
from the NIGMS Human Genetic Mutant Cell Repository, Camden, NJ. We
prepared genomic DNAs from the cell lines by using standard methods of
phenol extraction/ethanol precipitation (5) or the
A.S.A.P. geno-mic DNA isolation kit (Boehringer-Mannheim,
Indianapolis, IN). Genomic DNA concentrations were determined by
ultraviolet absorbance, with the assumption that 1
A260 corresponds to 40 mg/L (5).
blood samples
Whole-blood samples were collected from 24 adult subjects of known
genotypes (8 A/A, 8 A/S, and 8 S/S) and anticoagulated with EDTA. DNA
was isolated from the blood without delay (unless specified otherwise)
by use of the InstaGeneTM Whole Blood Kit (Bio-Rad
Labs., Hercules, CA). For the precision and blood storage studies, we
used blood collected as above from two homozygous A/A subjects. Blood
samples from subjects with genotypes A/GSan Jose, D/S, and
S/C were collected at "V. Cervello" Hospital, Palermo, and at
Istituti Clinici di Perfezionamento Laboratorio di Ricerche Cliniche,
Milan. The presence of hemoglobin chain variants in all samples was
confirmed by using the Bio-Rad VariantTM Hemoglobin Testing
System (4). Studies were conducted in accordance with the
Helsinki Declaration and the guidelines for research at the H. San
Raffaele Institute.
sickle cell by lcr assay
LCR.
The A allele-specific LCR mixture consisted of the
following components in a total volume of 25 µL: 3.6 nmol/L of each A
allele oligonucleotide, 0.9 nmol/L of each hGH oligonucleotide, 1.5 U
of thermostable DNA ligase (Ampligase; Epicentre Technologies, Madison,
WI), 250 ng of salmon sperm DNA (Sigma), 4 µL (~250 ng) of
patient's DNA, 20 mmol/L of Tris-HCl (pH 8.3), 25 mmol/L of KCl, 10
mmol/L of MgCl2, 0.5 mmol/L of NAD+, and 0.1
mL/L Triton X-100. In S allele-specific LCR, the A allele
oligonucleotide set was replaced by the S allele oligonucleotide set
(7.6 nmol/L of each oligonucleotide), and the concentration of each hGH
oligonucleotide was decreased to 0.8 nmol/L. LCRs were performed in
batches of eight patients' samples in a thermal cycler using the
following program: 94 °C for 1.5 min, followed by 55 °C for 6 min
(2 cycles), then 91 °C for 0.5 min, followed by 55 °C for 6 min
(25 cycles). We included the following amplification controls with each
batch of patients' samples assayed: 250 ng of human placental DNA, 250
ng of cell line CRL8756 DNA, and a blank (water added instead of DNA
template).
Capture oligonucleotide immobilization.
Oligonucleotides
MD123 (specific for the ß-globin A and S ligation products) and MD269
(specific for the hGH ligation product) were resuspended in 0.5 mol/L
EDTA, pH 8, and immobilized in polystyrene wells (Maxisorp; Nunc,
Naperville, IL) by passive adsorption (7)(8).
We added to each well 50 µL of a 0.2 µmol/L oligonucleotide
solution, incubated the wells overnight at 37 °C, and then washed
them five times with Radias Well Wash buffer (Bio-Rad Labs.).
Colorimetric detection.
After amplification, the
A-specific and S-specific reaction mixtures were diluted 10- and
5-fold, respectively, with 1x SSC (150 mmol/L NaCl and 15 mmol/L
sodium citrate, pH 7). Separate 50-µL aliquots of the diluted
mixtures were loaded into MD123- or MD269-coated microwells and
incubated at 37 °C for 1 h; the wells were then washed five
times with Radias Well Wash buffer. We diluted streptavidin–alkaline
phosphatase conjugate (Bio-Rad Labs.) 1:4000 with 1x SSC and added 50
µL of the diluted conjugate to each well; after incubation at
37 °C for 1 h, we washed the wells five times with Radias Well
Wash buffer. Alkaline phosphatase was detected with a colorimetric
amplification scheme based on (a) the initial conversion of
NADPH (substrate) to NADH; (b) the subsequent
interconversion of NADH to NAD+ in a cyclic redox reaction
in the presence of alcohol dehydrogenase, diaphorase, and INT-violet
(amplifier); and (c) the production of a formazan end
product with Amax at 490 nm (9).
Both substrate and amplifier for the colorimetric detection were from
Bio-Rad Labs.
First, we added 50 µL of substrate to each well; after 30 min at
37 °C, 50 µL of amplifier was added. The plate was immediately
transferred to a Model 3550 microplate reader controlled by the
Microplate Manager data analysis software (both from Bio-Rad Labs.).
The rate of color formation was determined by taking
A490 readings every 20 s for 3 min. The
detection controls consisted of the biotinylated oligonucleotides MD122
and MD117 (sequences not shown), which are complementary to the
immobilized MD123 and MD269, respectively, and a detection blank (1x
SSC).
|
Results
|
|---|
With the proper design of LCR oligonucleotides, one can detect LCR
amplification products in a nonradioactive, ELISA-like microplate
format. In the approach described here, one of the four
oligonucleotides is labeled with biotin and another oligonucleotide is
tagged with an arbitrary "tail" sequence. These two
oligonucleotides are complementary to the same strand of template DNA;
ligation thus results in the formation of a molecule that has biotin at
one end and a "tail" at the other. This molecule is captured by
hybridization to a "tail"-complementary oligonucleotide immobilized
on the surface of a microwell. The presence of biotin is detected by
sequential reactions with streptavidin–alkaline phosphatase and a
chromogenic substrate.
Each DNA sample is tested in two duplex LCRs: The A-specific reaction
coamplifies the ß-globin A allele and the hGH gene (the measured
signals being henceforth referred to as A and hGHA,
respectively), and the S-specific reaction coamplifies the S allele and
the hGH gene (giving signals S and hGHS, respectively). DNA
ligase has been reported to join a perfectly matched junction with at
least 10-fold greater efficiency than a mismatched ligation junction
(10)(11)(12). To assure unambiguous genotyping, we
set up the following specifications for the LCR sickle cell assay:
(a) The ratio of the specific signal to the nonspecific
signal should be 
10 (i.e., A:S signal ratio should be >10 for
homozygous A/A and <0.1 for homozygous S/S), and (b) in a
heterozygous A/S sample, the A and S signals should differ by a factor
of 3 or less (0.3< A:S ratio <3).
Our investigations of LCR conditions—different oligonucleotide
concentrations, ligase concentrations, cycling temperature setpoints,
and cycle numbers—determined that the optimum assay conditions require
the addition of twice as much S allele oligonucleotides as A allele
oligonucleotides in the LCR, and the detection of twice as much
S-specific as A-specific ligation products (see Materials and
Methods).
To demonstrate the high allele-specificity of the LCR assay, we tested
tissue and cell-line genomic DNAs of known genotypes. As shown in Fig. 1
, the A-specific reaction showed little cross-reactivity with
the S allele, and vice versa. Moreover, the A:S signal ratios for the
A/A, A/S, and S/S samples fell within the ranges specified. A sample
containing a homozygous deletion of the ß-globin gene gave A and S
signals similar to those obtained for a blank (no DNA template added),
indicating the absence of both alleles; in addition, the same sample
was clearly differentiated from the blank by the hGH signal.
View larger version (16K):
[in this window]
[in a new window]
|
Figure 1. Allele specificity of the Sickle Cell by LCR assay shown
for duplicate samples of placental or cell-line DNAs (250 ng each).
The averages of the raw ß-globin and hGH signals are plotted. The
numbers on top of the stacked bars indicate the A:S
signal ratios. del, homozygous deletion of the
ß-globin gene; blk, no template DNA added;
black bar, A allele signal; white bar, S allele
signal; dark-shaded bar, hGHA signal;
light-shaded bar, hGHS signal.
|
|
To evaluate the utility of the assay as a diagnostic tool, we assayed
whole-blood samples collected from the 24 adult subjects of known
genotypes. The genomic DNAs were prepared and duplicate aliquots of the
DNA preparation were assayed by LCR. First, we used the hGH signal to
assess whether enough DNA had been present for the LCR step for the
assay to be valid. The hGH signal cutoff values were determined by
running 20 blank A-specific and 20 blank S-specific reactions (i.e., no
DNA template added). The mean (±SD) values obtained were:
hGHA 4.9 (±2.2) and hGHS 6.5 (±3.4)
mA/min. Positive detection of the hGH gene was defined as
hGHA >11.5 or hGHS >16.7, corresponding to
signals >3 SD above the respective means of the blank values. All 48
DNA preparations in this set gave hGH values exceeding these limits.
As shown in Fig. 2
, the A:S signal ratios of the 24 patients' samples clearly
correlated with genotype. The mean (±SD) A:S signal ratios obtained
were 38.0 (±7.2) for A/A, 0.74 (±0.13) for A/S, and 0.03 (±0.004)
for S/S. All ratios were within assay specifications.
View larger version (17K):
[in this window]
[in a new window]
|
Figure 2. A:S signal ratios for patients' samples.
DNA preparations from 24 patients' blood samples were tested in
duplicate by the LCR assay for sickle cell, and the average A:S signal
ratios were plotted. Samples 1–8: A/A homozygotes;
9–16, A/S heterozygotes; 17–24, S/S
homozygotes.
|
|
The near-equivalence of the A and S signals in heterozygous A/S samples
indicates that little or no interference occurred between the normal
and S sequences in either the A-specific or S-specific LCRs. To
determine whether this observation could be extended to other alleles,
we tested four subjects with heterozygous genotypes (Table 2
). One sample repeatedly gave hGH signals below the cutoffs and
was rejected. The results for the other three samples showed that the
A- and S-specific LCRs were highly specific for their respective target
alleles, even if another ß-globin allele present in the sample
differed by only a single base in the region amplified. Interestingly,
the A:S signal ratios obtained for these samples were slightly
different from those in Fig. 2
.
To determine the precision of the assay (in terms of the A:S signal
ratio), we assayed blood from a single A/A patient collected on three
different days, each day's blood sample being used to make eight
replicate DNA preparations. The 24 LCR assays gave a mean A:S signal
ratio (i.e., mean of daily means) of 46.9, with a within-run SD of 11.9
(CV 25%) and a total SD of 11.5 (CV 24%) (13). In all
cases, the A:S signal ratio was >10.
The validity of the LCR assay was determined by using blood that had
been stored for various times (Fig. 3
). Blood from an A/A subject was stored in aliquots at
2–8 °C for as long as 7 days before DNA isolation. Similarly, blood
from a second A/A subject was stored at -20 °C for as long as 6
weeks. All DNA samples from the same subject were batched and tested in
the LCR assay on the same day. All samples showed an A:S signal ratio
>10, although imprecision for the aliquots stored at -20 °C
(mean = 39.7, SD = 13.5, CV = 34%) was greater than for
those stored at 2–8 °C (mean = 38.6, SD = 4.0, CV =
10%).
View larger version (15K):
[in this window]
[in a new window]
|
Figure 3. Effect of blood storage on LCR assay A:S signal ratios.
Blood samples collected from two A/A patients were stored at either
2–8 °C ( ) or -20 °C ( ) for different periods before DNA
preparation was performed.
|
|
|
Discussion
|
|---|
In a study of 24 subjects, the Sickle Cell by LCR assay was able
to discriminate among A/A, A/S, and S/S genotypes. Because the
differences in A:S signal ratios among the three classes of patients
were at least an order of magnitude, genotypic assignment was
unambiguous. The use of a simple and fast DNA extraction procedure and
a 96-well detection format means that results can be obtained in ~6 h
after the whole blood is collected. The assay can also be performed
with blood samples that have been refrigerated at 2–8 °C for as
long as 1 week or frozen at -20 °C for as long as 6 weeks. The
results presented here are for EDTA-anticoagulated blood samples.
However, blood collected in citrate or heparin can also be used (data
not shown).
The target sequence in LCR is a very short region of the gene
complementary to the oligonucleotides used in the reaction. In LCR, the
target is "queried" for the presence of perfectly matched base(s)
at or near the ligation junction. Therefore, a mutation outside the
target region will not be detected, and a genotype such as D/S will be
typed as A/S (Table 2
). (Indeed, the D allele has the normal sequence
at codon 6.) Another limitation of LCR is that a negative result
indicates the presence of a mismatch but does not reveal its nature.
For example, because the C and GSan Jose mutations
occur near the ligation junction, neither allele should be detected in
the A- and S-specific LCRs. Consistent with this prediction, only the S
and A alleles were detected in the heterozygous S/C and
A/GSan Jose samples, respectively; however, the A:S signal
ratios obtained differed from those of "true" A/A and S/A samples
(Table 2
). More samples that are "hemizygous" for the A and S
sequences at the LCR target region need to be tested to determine
whether this difference is real.
One solution to these limitations of the LCR assay would be to type for
all known alleles. The LCR assay is "modular" in the sense that the
presence of an allele other than A or S can be determined by simply
adding the appropriate allele-specific reaction to the assay. For
example, we have designed a C allele-specific LCR that, when performed
with the A- and S-specific reactions described here, accurately types
samples with respect to the A, S, and C alleles, i.e., A/A, A/S, A/C,
S/C, or S/S genotype (data not shown).
In perspective, LCR cannot replace conventional (non-DNA) screening
methods for hemoglobin disorders. HPLC can provide information on the
presence and abundance of abnormal hemoglobins
(4)(14), whereas LCR is more appropriate for
confirmatory diagnosis of suspected mutations by virtue of its highly
specific capability for mutation detection. Although in theory a
condition such as S/ß+-thalassemia that is detectable by
HPLC (14) would be detectable by LCR, using LCR for this
would be impractical because of the large number of possible mutations
responsible for the ß+-thalassemia phenotype.
In conclusion, we have demonstrated that LCR is an exquisitely
specific method for mutation detection. The basic format of the assay
described here should be applicable to other genetic diseases of known
etiology.
|
Acknowledgments
|
|---|
We thank A. Maggio, "V. Cervello" Hospital, Palermo; A.
Cantù-Rajnoldi, Istituti Clinici di Perfezionamento Laboratorio
di Ricerche Cliniche, Milan; and C. Rosatelli, Istituto di Biologia e
Clinic dell'Et 133 Evolutiva, Cagliari, for providing some of the
blood samples.
|
Footnotes
|
|---|
1 Nonstandard abbreviations: LCR, ligase chain reaction;
hGH, human growth hormone; and 1x SSC, 150 mmol/L NaCl and 15 mmol/L
sodium citrate, pH 7. 
|
References
|
|---|
-
Wallace RB, Lin C-IP, Reyes AA, Lowery JD, Ugozzoli L. Ligase chain reaction for the detection of specific DNA sequences and point mutations [Review]. Pfeifer GP eds. Technologies for detection of DNA damage and mutations 1996:307-322 Plenum Press New York. .
-
Wu DY, Wallace RB. The ligation amplification reaction (LAR)—amplification of specific DNA sequences using sequential rounds of template-dependent ligation. Genomics 1989;4:560-569.
[ISI][Medline]
[Order article via Infotrieve]
-
Lubin BH, Witkowska HE, Kleman K. Laboratory diagnosis of hemoglobinopathies. Clin Biochem 1991;24:363-374.
[ISI][Medline]
[Order article via Infotrieve]
-
Papadea C, Cate JC. Identification and quantification of hemoglobins A, F, S, and C by automated chromatography. Clin Chem 1996;42:57-63.
[Abstract/Free Full Text]
-
Maniatis T, Fritsch EF, Sambrook J. Molecular cloning: a laboratory manual 1982:458-462 Cold Spring Harbor Laboratory Cold Spring Harbor, NY. .
-
Thein SL, Wallace RB. The use of synthetic
oligonucleotides as specific hybridization probes in the diagnosis of
genetic disorders. In: Davies KE, ed. Human genetic diseases, a
practical approach. Oxford: IRL Press, 1986:33–50..
-
Pesce AJ, Mendoza N, Boreisha MA, Gaizutis MA, Pollak VE. Use of enzyme-linked antibodies to measure serum anti-DNA antibody in systemic lupus erythematosus. Clin Chem 1974;20:353-359.
[Abstract]
-
Lacy MJ, Voss EW. Direct adsorption of ssDNA to polystyrene for characterization of the DNA/anti-DNA interaction, and immunoassay for anti-DNA autoantibody in New Zealand White mice. J Immunol Methods 1989;116:87-98.
[ISI][Medline]
[Order article via Infotrieve]
-
Self CH. Enzyme amplification—a general method applied to provide an immunoassisted assay for placental alkaline phosphatase. J Immunol Methods 1985;76:389-393.
[ISI][Medline]
[Order article via Infotrieve]
-
Wu DY, Wallace RB. Specificity of the nick-closing activity of bacteriophage DNA ligase. Gene 1989;76:245-254.
[ISI][Medline]
[Order article via Infotrieve]
-
Nickerson DA, Kaiser R, Lappin S, Stewart J, Hood L. Automated DNA diagnostics using an ELISA-based oligonucleotide ligation assay. Proc Natl Acad Sci U S A 1990;87:8923-8927.
[Abstract/Free Full Text]
-
Barany F. Genetic disease detection and DNA amplification using cloned thermostable ligase. Proc Natl Acad Sci U S A 1991;88:189-193.
[Abstract/Free Full Text]
-
NCCLS. Tentative Guideline EP5–T2. Evaluation of
precision performance of clinical chemistry devices, 2nd ed. Villanova,
PA: National Committee for Clinical Laboratory Standards, March 1992..
-
Lorey F, Cunningham G, Shafer F, Lubin B, Vichinsky E. Universal screening for hemoglobinopathies using high-performance liquid chromatography: clinical results of 2.2 million screens. Eur J Hum Genet 1994;2:262-271.
[Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
|
|
 
A. K. Tong and J. Ju
Single nucleotide polymorphism detection by combinatorial fluorescence energy transfer tags and biotinylated dideoxynucleotides
Nucleic Acids Res.,
March 1, 2002;
30(5):
e19 - e19.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
