Clinical Chemistry 49, No. 6, 2003
heart disease (HF stage B). Because more discrete differences in NT-proBNP values need to be distinguished for
HF stage B, the lower sensitivity and specificity of the
Biomedica assay for diagnosis of this stage may be a
consequence of the greater imprecision of the competitive
Biomedica NT-proBNP method compared with the noncompetitive sandwich assay from Roche. Theoretically,
the presence of various NT-proBNP split products, possibly having different biological activities and metabolic
pathways, may also contribute to this phenomenon. However, the underlying mechanisms remain unclear.
Limitations of our study include the hospital-based
study design, the fact that the prevalence of each HF class
may have been fixed by the selection of participants
(possibly not representative of routine practice), the wide
age range not equally distributed among the study
groups, and the greater tendency for diseased patients to
be receiving drug therapy. These facts may have imposed
a certain bias related to the absolute sizes of the AUCs in
the ROC analysis. However, the relative proportions of
the AUCs for the two methods, showing that the Biomedica and the Roche assays give comparable information for diagnosis of HF stage C and that the Roche assay
might be superior to the Biomedica test for diagnosis of
HF stage B, should not be affected by the above limitations. Further comparisons of the newer NT-proBNP
assays, tested by independent research laboratories and
considering hospital as well as general practice settings,
are necessary to thoroughly clarify both their analytical
performance and their potential clinical utility.
This work was supported in part by a grant for scientific
research from the Upper Austrian Government. We thank
Roche Diagnostics (Vienna, Austria) and Biomedica
Gruppe (Vienna, Austria) for technical assistance. Neither
of the companies played a role in the study design, data
collection, data analysis, data interpretation, writing the
report, or decisions to submit the manuscript for publication. We also wish to thank Martina Stuetz from our
laboratory for performing the Biomedica NT-proBNP EIA
analyses.
References
1. Hammerer-Lechner A, Puschendorf B, Mair J. Cardiac natriuretic peptides:
new laboratory parameters in heart failure patients [Review]. Clin Lab
2001;47:265–7.
2. Mair J. Role of cardiac natriuretic peptide testing in heart failure [Editorial].
Clin Chem 2002;48:977– 8.
3. Missbichler A, Hawa G, Woloszczuk W, Schmal N, Hartter E. Enzymimmunoassays für proBNP Fragmente (8 –29) und (32–57). J Lab Med 1999;23:
241– 4.
4. Hammerer-Lechner A, Neubauer E, Muller S, Pachinger O, Puschendorf B,
Mair J. Head-to-head comparison of N-terminal pro-brain natriuretic peptide,
brain natriuretic peptide and N-terminal pro-atrial natriuretic peptide in
diagnosing left ventricular dysfunction. Clin Chim Acta 2001;310:193–7.
5. Stanek B, Frey B, Hulsmann M, Berger R, Sturm B, Strametz-Juranek J, et al.
Prognostic evaluation of neurohumoral plasma levels before and during
␤-blocker therapy in advanced left ventricular dysfunction. J Am Coll Cardiol
2001;38:436 – 42.
6. Dudek D, Rzeszutko L, Petkow Dimitrow P, Bartus S, Sorysz D, Chyrchel M,
et al. Circulating N-terminal brain natriuretic peptide precursor and endothelin levels in patients with syndrome X and left bundle branch block with
preserved systolic function. Int J Cardiol 2001;79:25–30.
7. Clerico A, Del Ry S, Giannessi D. Measurement of cardiac natriuretic
8.
9.
10.
11.
12.
979
hormones (atrial natriuretic peptide, brain natriuretic peptide, and related
peptide) in clinical practice: the need for a new generation of immunoassay
methods [Review]. Clin Chem 2000;46:1529 –34.
Karl J, Borgya A, Gallusser A, Huber E, Krueger K, Rollinger W, et al.
Development of a novel, N-terminal-proBNP (NT-proBNP) assay with a low
detection limit. Scand J Clin Lab Invest Suppl 1999;230:177– 81.
Hunt SA, Baker DW, Chin MH, Cinquegrani MP, Feldman AM, Francis GS, et
al. ACC/AHA Guidelines for the Evaluation and Management of Chronic
Heart Failure in the Adult: Executive Summary. A report of the American
College of Cardiology/American Heart Association Task Force on Practice
Guidelines [Guideline]. Circulation 2001;104:2996 –3007.
Passing H, Bablok W. A new biometrical procedure for testing the equality of
measurements from two different analytical methods. Application of linear
regression procedures for method comparison studies in clinical chemistry,
Part I. J Clin Chem Clin Biochem 1983;21:709 –20.
Bland JM, Altman DG. Measuring agreement in method comparison studies
[Review]. Stat Methods Med Res 1999;8:135– 60.
Hanley JA, McNeil BJ. A method of comparing the areas under receiver
operating characteristic curves derived from the same cases. Radiology
1983;148:839 – 43.
Extraction of RNA from Dried Blood on Filter Papers
after Long-Term Storage, Håkan Karlsson,1* Claes Guthenberg,2 Ulrika von Döbeln,2 and Krister Kristenssson1 (1 Division of Neurodegenerative Diseases, Department of Neuroscience, Karolinska Institutet, S-171 77 Stockholm,
Sweden; 2 PKU Laboratory, Center for Inherited Metabolic Diseases, Huddinge University Hospital, 141 86
Huddinge, Sweden; * address correspondence to this author at: Department of Neuroscience, Karolinska Institutet, Retzius väg 8, S-171 77 Stockholm, Sweden; fax
46-8-32-53-25, e-mail [email protected])
Blood dried on filter paper is widely used for screening of
inherited metabolic disorders (1 ). In Sweden, such filters
from all newborns have been permanently stored since
1975. It has been shown that proteins and DNA may be
recovered from these cards after extended periods of
storage (2– 4 ). RNA, however, has been considered too
vulnerable to degradation by ribonucleases to be recovered from these filters. Despite this, Zhang and McCabe
(5 ) and Matsubara et al. (6 ) reported that mRNA could be
isolated from such filters after up to 4 years of storage.
The stability of viral RNA on filters has also been reported
(7, 8 ), although this was not tested over extended periods
of time. The purpose of the present study was to investigate whether RNA could be recovered from filters that
had been stored since 1975 and be amplified by reverse
transcription-PCR.
After approval by the local ethics committee, we randomly selected filter papers (specimen collection paper
2992; Schleicher & Schuell) that had been stored for 1
month, 21 years, and 27 years; for each time point, we
selected five filters. One-fourth of a spot (⬃0.3 cm2)
containing dried blood was cut out of each filter with a
sterile razor blade. As a negative control, a piece of
corresponding size was cut from a blood-free area of each
filter. RNA was isolated from the specimen with use of
the RNeasy reagent set (Qiagen) according to the manufacturer’s instructions. Briefly, filters were incubated in
lysis buffer at 37 °C for 30 min in a thermomixer (Eppen-
980
Technical Briefs
Table 1. GenBank accession numbers, primer sequences,
and expected sizes of the products generated by the
respective primers used for detection of RNA.
Gene
Accession
no.
GAPDHa
J04038
␤-Actin
NM001101
a
Primer sequences
Forward: 5⬘-cgaccactttgtcaagctca-3⬘
Reverse: 5⬘-ttactccttggaggccatgt-3⬘
Forward: 5⬘-atcctaaaagccaccccact-3⬘
Reverse: 5⬘-ctcaagttgggggacaaaaa-3⬘
Product
size, bp
97
205
The GAPDH primers amplify across intron H (104 bp) in genomic DNA.
dorf) rotating at 1000 rpm. The lysates were subsequently
homogenized by spinning through a QiaShredder (Qiagen). The flow-through was applied to a RNeasy column,
and after careful washing, the RNA was eluted in 50 ␮L of
RNase-free water. We subjected 8 ␮L of the RNA to
DNase I digestion (Life Technologies) with subsequent
reverse transcription in a 20-␮L reaction with the following reagents from Life Technologies: 150 ng of random
primers, 1⫻ first strand buffer, 10 mM dithiothreitol, and
0.5 mM each of the deoxynucleotide triphosphates. The
reaction was heated to 72 °C and chilled on ice before the
addition of 200 U of Superscript II. cDNA was then
generated for 1 h at 42 °C before the reaction was inactivated by heating to 70 °C for 15 min.
The presence of cDNA was verified by PCR using
primers designed to amplify fragments of the mRNAs
encoding human ␤-actin and glyceraldehyde-3-phosphate
dehydrogenase (GAPDH; see Table 1), respectively, using
the following conditions: 1 ␮L of cDNA was added to a
24-␮L reaction mixture containing 1⫻ Titanium Taq DNA
Polymerase mixture; 1⫻ Titanium PCR Buffer (Clontech
Laboratories); 1 ␮M each of gene-specific forward and
reverse primers, respectively; and 200 ␮M each of the
deoxynucleotide triphosphates (Life Technologies). For
amplification, a GeneAmp PCR system 9700 (Applied
Biosystems) was used at the following cycling conditions:
heat activation for 2 min at 94 °C, followed by 35 cycles of
denaturation at 94 °C for 30s and annealing/extension at
68 °C for 60 s, with a final extension at 72 °C for 7 min.
PCR products were electrophoresed in 2% agarose in
Tris-acetate-EDTA buffer (40 mmol/L Tris, 20 mmol/L
acetic acid, 1 mmol/L EDTA). Double-stranded DNA was
stained in 1⫻ SYBR Gold Stain (Molecular Probes) in the
Tris-acetate-EDTA buffer. Double-stranded DNA was
subsequently visualized and documented on a Gel Doc
2000 system (Bio-Rad).
We were able to amplify 97- and 205-bp mRNA fragments encoding human GAPDH and ␤-actin, respectively, from filters that had been stored for 1 month, 21
years, and 27 years (Fig. 1A). No RNA was detected on
the filters outside the area of the dried blood spot. Because
GAPDH- and ␤-actin-related pseudogenes are present in
the human genome, generation of an intronless amplicon
does not rule out the presence of contaminating genomic
DNA in the RNA preparation. DNase I treatment of the
RNA before the generation of cDNA is crucial, and
documented that RNA was present (Fig. 1B).
Although filters have been stored since 1975, only since
1981 have all filters been kept at 4 °C. Additionally, only
since 1996 have all filters been stored in an area with a
controlled relative humidity not exceeding 30%. This
difference in storage conditions did not seem to greatly
influence the amount or quality of the RNA that was
amplified. The total amount of recovered RNA was insufficient for meaningful spectrophotometric quantification,
and no efforts were made to quantify the amount of
starting material by the PCR approach used.
These filters constitute a unique resource for investigating events that may be reflected in peripheral blood cells
or plasma around the time of birth, and they have
previously been used for DNA and protein analysis. Our
present finding expands on the number of investigations
that can be performed with these filters. One such application may be testing hypotheses put forward by epidemiologists regarding various disorders of an autoimmune
or developmental nature. For several of these, exposure to
infectious agents during the intrauterine period or early
life has been suggested as a risk factor for developing the
disease (9, 10 ).
In conclusion, routinely stored phenylketonuria-screening filters may be used for the study of perinatal events,
detectable in RNA, that may be of relevance for the
etiopathogenesis of disorders that would be very costly
and, today, impractical to address in prospective studies.
Fig. 1. Reverse transcription-PCR amplification of GAPDH and ␤-actin
mRNA.
(A), RNA was extracted from filter papers stored for 1 month (2002), 21 years
(1981), or 27 years (1975). RNA was extracted from blank parts of the filter
papers (⫺) and from the dried blood spots (⫹). Lane N, PCR reaction mixture with
water added instead of sample. (B), GAPDH mRNA was amplified from total RNA
isolated from one filter paper not subjected to DNase I treatment or reverse
transcription (lane 1), from DNase I-treated total RNA not subjected to reverse
transcription (lane 2), and from randomly primed, reverse-transcribed DNase
I-treated RNA (lane 3).
Clinical Chemistry 49, No. 6, 2003
This study was supported by The Stanley Medical Research Institute.
References
1. Guthrie R, Susi A. A simple phenylalanine method for detecting phenylketonuria in large populations of newborn infants. Pediatrics 1963;32:338 – 43.
2. Williams C, Weber L, Williamson R, Hjelm M. Guthrie spots for DNA-based
carrier testing in cystic fibrosis. Lancet 1988;2:693.
3. Verlingue C, Mercier B, Lecoq I, Audrezet MP, Laroche D, Travert G, et al.
Retrospective study of the cystic fibrosis transmembrane conductance
regulator (CFTR) gene mutations in Guthrie cards from a large cohort of
neonatal screening for cystic fibrosis. Hum Genet 1994;93:429 –34.
4. Tappin DM, Greer K, Cameron S, Kennedy R, Brown AJ, Girdwood RW.
Maternal antibody to hepatitis B core antigen detected in dried neonatal
blood spot samples. Epidemiol Infect 1998;121:387–90.
5. Zhang YH, McCabe ER. RNA analysis from newborn screening dried blood
specimens. Hum Genet 1992;89:311– 4.
6. Matsubara Y, Ikeda H, Endo H, Narisawa K. Dried blood spot on filter paper
as a source of mRNA. Nucleic Acids Res 1992;20:1998.
7. Abe K, Konomi N. Hepatitis C virus RNA in dried serum spotted onto filter
paper is stable at room temperature. J Clin Microbiol 1998;36:3070 –2.
8. Fiscus SA, Brambilla D, Grosso L, Schock J, Cronin M. Quantitation of human
immunodeficiency virus type 1 RNA in plasma by using blood dried on filter
paper. J Clin Microbiol 1998;36:258 – 60.
9. Munk-Jorgensen P, Ewald H. Epidemiology in neurobiological research:
exemplified by the influenza-schizophrenia theory. Br J Psychiatry 2001;178:
S30 –2.
10. Dahlquist G, Kallen B. Early neonatal events and the disease incidence in
nonobese diabetic mice. Pediatr Res 1997;42:489 –91.
Fetal Expressed Gene Analysis in Maternal Blood: A
New Tool for Noninvasive Study of the Fetus, Jean-Marc
Costa,1* Alexandra Benachi,2 Martine Olivi,1 Yves Dumez,2
Michel Vidaud,3 and Evelyne Gautier1 (1 Centre de Diagnostic Prénatal, American Hospital of Paris, 63 bd Victor
Hugo, 92200 Neuilly-sur-Seine, France; 2 Maternité, Hôpital Necker-Enfants Malades, 75015 Paris, France; 3 Laboratoire de Génétique Moléculaire, Faculté des Sciences
Pharmaceutiques et Biologiques de Paris, 75006 Paris,
France; * author for correspondence: fax 33-1-46-41-26-56,
e-mail [email protected])
Noninvasive approaches to prenatal diagnosis can avoid
the risk of fetal loss associated with invasive procedures
such as chorionic villus sampling, amniocentesis, and
cordocentesis. Isolation of fetal cells from maternal blood
requires further improvements before it can be applied in
a clinical setting (1 ), but the reliability of cell-free fetal
DNA analysis in maternal plasma or serum is now well
established (2– 4 ). As a result, it is currently used in
specialized centers for the determination of fetal sex and
fetal RhD status for the management of pregnant women
at risk for X-linked disorders (5 ) or RhD alloimmunization (6 ). Because fetal DNA in maternal serum is circulating in an excess background of maternal DNA, clinical
applications are restricted mainly to the detection of fetal
sequences distinct from the mother’s DNA sequences.
Fetal RNA in maternal blood may be an alternative
source of fetal nucleic acids. Al-Mufti et al. (7 ) detected
specific RhD mRNA in mononuclear fetal cells isolated
from blood of RhD-negative pregnant women, and Lo’s
group demonstrated the presence of Y-chromosome-specific (ZFY) mRNA in maternal plasma of women carrying
981
a male fetus (8 ). These two applications, however, again
require fetal sequences that differ from the mother’s.
We have investigated the presence in maternal blood of
fetal transcripts that may have the same sequence as that
of the mother. Human chorionic gonadotropin (hCG)
mRNA is a good candidate because it is a pregnancyspecific polypeptide hormone produced by the placenta
and is specifically expressed in the fetal syncytiotrophoblast.
We studied 43 pregnant women and 20 nonpregnant
women who had previously given birth to at least one
neonate. After receiving informed consent, we collected
blood (2.5 mL) into PAXgeneTM tubes to reduce RNA
degradation (9 ). All pregnancy samples were obtained
before any invasive procedure during either the first (n ⫽
23) or the second (n ⫽ 20) trimester. The mean gestational
ages were 11.8 weeks (range, 9 –14.5 weeks) and 19.8
weeks (range, 18 –23 weeks), respectively.
We collected an additional 7 mL of blood into Vacutainer SST® tubes (Becton Dickinson). Immediately after
clotting, the serum was separated by centrifugation at
3000g for 10 min at 4 °C, and 2.5 mL was transferred to
PAXgene tubes.
Total blood and serum were treated according to the
same protocol. RNA was isolated with the PAXgene
Blood RNA reagent set (Qiagen) as recommended by the
manufacturer, except that RNA was eluted in 50 instead
of 100 ␮L of elution buffer. A DNase digestion step was
systematically included during the procedure, and the
integrity of the RNA was monitored by 1% agarose gel
electrophoresis. Each sample extract was tested in duplicate.
Each extracted sample (5 ␮L) was subjected to reverse
transcription in 20 ␮L of a reverse transcriptase mixture
containing 1⫻ reverse transcription buffer (0.5 mM each
deoxynucleotide triphosphate, 5 mM MgCl2, 75 mM KCl,
50 mM Tris-HCl, pH 8.3), 2.5 ␮M random hexamers, 10
units of RNase inhibitor, and 15 U of Multiscribe reverse
transcriptase (Applied Biosystems). The mixture was incubated for 10 min at 25 °C, followed by 60 min at 42 °C,
and the reverse transcriptase was inactivated by heating
at 99 °C for 10 min. A reaction mixture without the
reverse transcriptase was used as a negative control.
We used 5 ␮L of the reverse transcription reaction for
specific detection of hCG␤ transcripts in a real-time PCR
method adapted from a previously described method
(10 ). Amplification was carried out in a LightCycler®
instrument (Roche Biochemicals). PCR reactions were set
up in a final volume of 20 ␮L, using the Fast DNA Master
Hybridization Probes Kit (Roche Biochemicals), with 0.5
␮M each primer, 0.25 ␮M each probe, 1.25 U of uracil
DNA glycosylase (Biolabs), and 4.5 mM MgCl2. After an
initial denaturation step of 8 min at 95 °C, amplification
was performed for 50 cycles (denaturation at 95 °C for
10 s, annealing at 60 °C for 10 s, and extension at 72 °C for
15 s). The annealing step for each sample was monitored
by continuous fluorescence monitoring.
RNA obtained from a first-trimester chorionic villus
sample was introduced during each run as a positive