1578

Letters

Phillip A. Isotalo1 Donald C. Greenway1,2,4 James G. Donnelly1,3,4* Department of Pathology and Laboratory Medicine University of Ottawa Ottawa, Ontario, Canada K1H 8M5 Division of Biochemistry Department of Pathology and Laboratory Medicine Ottawa HospitalGeneral Campus Ottawa, Ontario, Canada K1H 8L6 Division of Biochemistry Department of Pathology and Laboratory Medicine Ottawa HospitalCivic Campus Ottawa, Ontario, Canada K1Y 4E9
4 Department of Biochemistry, Microbiology, and Immunology Faculty of Medicine University of Ottawa, Ottawa, Ontario, Canada K1H 8M5 3 2 1

*Address correspondence to this author at: Division of Biochemistry, Department of Laboratory Medicine, Ottawa HospitalCivic Campus, 1053 Carling Ave., Ottawa, Ontario, Canada K1Y 4E9. Fax 613-761-5361; e-mail jdonnelly@ civich.ottawa.on.ca.

Mass Spectrometry of Nucleic Acids

To the Editor: We read with interest the review by Kricka (1 ) on nucleic acid detection technologies, in which he mentioned that nucleic acids do not have any intrinsic properties for direct detection. In response to this, we would like to point out the determination of intrinsic molecular weights of nucleic acids using mass spectrometry (MS) has been widely accepted as one of the most accurate methods to detect nucleic acids (2 ). Using matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) MS, a mass resolution of 1 per 1000 and the detection of low femtomole quantities of DNA can be achieved routinely (3 ). Nucleic acids ranging from 2 to 2000 nucleotides can be

detected by using MALDI-TOF MS (4 ). Because of the mass differences of the nucleobases, MS can also be used to analyze mixtures of different nucleic acid fragments without the use of any label (5 ). Furthermore, in most cases, the separation of the fragments before MS measurements is not required. The minimum sample volume required for MALDITOF MS is only a few nanoliters (3 ). MS can, therefore, be easily linked to any miniaturization of sample processing. Typically, each mass spectroscopic measurement including acquisition and interpretation of mass spectrum takes 10 s. With the availability of automatic high-throughput MS systems that include sample preparation (6 ), the cost-effectiveness of using MS to analyze nucleic acids has become comparable to other analytical techniques. Currently, the size of a MALDI-TOF mass spectrometer is similar to an immunoassay analyzer. However, as stated in a recent report (7 ), the size of mass spectrometers can be substantially reduced. Together with the continued development of software for automated interpretation of mass spectra, MS has a great potential to become one of the most important analytical tools for clinical laboratories. Some of the current clinical applications of MS are (a) DNA sequencing (8 ); (b) detection of genetic variations such as single-nucleotide polymorphisms (9 ), microsatellites (10 ), short tandem repeats (11 ), and small insertions/deletions; and (c) gene expression.
References
1. Kricka LJ. Nucleic acid detection technologieslabels, strategies, and formats [Review]. Clin Chem 1999;45:453 8. 2. Crain PF, McCloskey JA. Applications of mass spectrometry to the characterization of oligonucleotides and nucleic acids [Review]. Curr Opin Biotechnol 1998;9:2534. 3. Little DJ, Cornish TJ, ODonnell MJ, Braun A, Cotter RJ, Koster H. MALDI on a chip: analysis of arrays of low-femtomole to subfemtomole quantities of synthetic oligonucleotides and DNA diagnostic products dispensed by a piezoelectric pipet. Anal Chem 1997;69:4540 6. 4. Berkenkamp S, Kirpekar F, Hillenkamp F. Infrared MALDI mass spectrometry of large nucleic acids. Science 1998;281:260 2. 5. Ross P, Hall L, Smirnov I, Haff L. High level multiplex genotyping by MALDI-TOF mass spectrometry. Nat Biotechnol 1998;16:134751. 6. ODonnell MJ, Little DP, Braun A. MassArray as

7. 8.

9.

10.

11.

an enabling technology for the industrial-scale analysis of DNA. Genet Eng News 1997;17:39. Henry CM. The incredible shrinking mass spectrometers. Anal Chem 1999;71:264A 8A. Fu DJ, Tang K, Braun A, Reuter D, DarnhoferDemar B, Little DP, et al. Sequencing exons 5 to 8 of the p53 gene by MALDI-TOF mass spectrometry. Nat Biotechnol 1998;16:381 4. Braun A, Little DP, Koster H. Detecting CFTR gene mutations by using primer oligo base extension and mass spectrometry. Clin Chem 1997;43:1151 8. Braun A, Little DP, Reuter D, Muller-Mysok B, Koster H. Improved analysis of microsatellites using mass spectrometry. Genomics 1997;46: 18 23. Ross P, Belgrader P. Analysis of short tandem repeat polymorphisms in human DNA by matrixassisted laser desorption/ionization mass spectrometry. Anal Chem 1997;69:3966 72.

Norman H.L. Chiu1,2* Charles R. Cantor1,2

1 Sequenom Inc. 11555 Sorrento Valley Rd. San Diego, CA 92121 2 Boston University Center for Advanced Biotechnology 36 Cummington St. Boston, MA 02215

*Author for correspondence. Fax 619350-9237; e-mail nchiu@sequenom.com.

Application of the MediSense Precision-G Blood Glucose Testing System in a Neonatal Intensive Care Unit

To the Editor: The ability to perform stat glucose testing in support of a neonatal intensive care unit has traditionally depended on transporting the sample to the central laboratory because most point-of-care glucose analyzers cannot accurately test glucose below 2.22 mmol/L (40 mg/dL). In addition, the high hematocrits commonly encountered in newborns and the high bilirubin concentrations often seen in the neonate can cause major problems with glucose measurements in whole blood. Most glucose meters are used for monitoring the diabetic; thus, their accuracy at low glucose concentrations has not been a prime consideration in their design. The acute management of glucose

Clinical Chemistry 45, No. 9, 1999

1579

tal intensive care unit is facilitated by the use of the Ames Glucometer Elite electrochemical glucose meter. J Pediatr 1997;130:1515. 2. Trajanoski Z, Brunner GA, Gfrerer RJ, Wach P, Pieber TR. Accuracy of home blood glucose meters during hypoglycemia. Diabetes Care 1996;19:14125.

Laurence M. Demers* Betty Smith Section of Clinical Chemistry The PennState-Geisinger Health System The M.S. Hershey Medical Center Hershey, PA 17033
*Author for correspondence.

Fig. 1. Scattergram comparing blood glucose analysis of 74 consecutive blood samples from our neonatal unit by the Precision-G (y-axis) vs the Vitros 950 (x-axis).

Uncovering Rare Mutations: An Unforeseen Complication of Routine Genotyping of APOE

homeostasis in the newborn is extremely important to good patient care in the neonatal intensive care setting, and the laboratory is frequently challenged to provide a more rapid and sensitive means of determining blood glucose in the hypoglycemic infant (1 ). Recently, several point-of-care testing instruments have been developed that allow for on-site glucose testing at the low end of the dynamic range of glucose measurements (2 ). We recently evaluated the Precision-G System by MediSense, which can determine glucose on a 5- L blood specimen in 20 s at the bedside with a dynamic range that extends to 1.11 mmol/L (20 mg/ dL). Seventy-four consecutive blood samples from the neonatal unit were collected by heelstick into lithium heparin capillary tubes, and glucose was determined simultaneously on an Ortho Clinical Diagnostics Vitros 950 analyzer and the Precision-G System. The neonatal population we studied demonstrated the usual range of normal to high hematocrit values as well as increased bilirubin, providing assurance that these variables did not influence the results. Comparative results between these two methods are shown in Fig. 1. Regression analysis of the glucose

values obtained with the Precision-G and the Vitros 950 (Fig. 1) revealed a correlation coefficient of 0.993 (Sy x 0.43 mmol/L), an intercept 0.053 ( 0.002) mmol/L, and a slope of 0.946 ( 0.051). Although the number of samples in the true hypoglycemic range was rather limited, the results from 14 subjects with glucose results below 2.22 mmol/L (40 mg/ dL) fell on the regression line. Overall, there was a slight negative systematic difference observed with the Precision-G results that averaged 0.31 mmol/L; differences in the analytical methods and the use of whole blood vs plasma sample most likely account for this difference. This slight difference, however, should have little influence on the neonatologists clinical decision regarding the detection of hypoglycemia. Between-run imprecision (CV) with the Precision-G using Abbott MediSense control material at 1.11 mmol/L (20 mg/dL) was 8% (n 15). It is our impression from these results that the Precision-G System can be used effectively to determine low blood glucose concentrations in a neonatal unit.
References
1. Innanen VT, Deland ME, de Campos FM, Dunn MS. Point-of-care glucose testing in the neona-

To the Editor: APOE genotyping to identify subjects with the E4 allele is helpful in the diagnosis of Alzheimer disease when used together with clinical criteria (1 ). The most common APOE genotyping method involves digestion of a 244-bp PCR-amplified fragment of APOE exon 4 followed by digestion with endonuclease HhaI (2 ). The digestion creates a characteristic pattern of DNA bands in electrophoresis gels for each of the three common APOE alleles (E4, E3, and E2) and thus for the six common APOE genotypes (E4/4, E3/3, E2/2, E4/3, E3/2, and E4/2) (2 ). However, there are four additional HhaI recognition sites within the 244-bp fragment that is amplified by this method, and the fragment also harbors several sites that differ from the HhaI recognition sequence (GCG/C) by a single nucleotide (2 ). In the course of 2000 APOE genotyping reactions, we have observed two individuals who had patterns of HhaI restriction fragments that were distinct from any that could have resulted from the common APOE genotypes (Fig. 1). DNA sequencing of these two individuals revealed that each was heterozygous for a different rare APOE mutation,