SRM 2372-Human DNA Quantitation Standard

NOW AVAILABLE: see http://www.nist.gov/srm

https://srmors.nist.gov/view_detail.cfm?srm=2372

Additional Information:

Kline, M.C., Duewer, D.L., Redman, J.W., Butler, J.M. (2005) Results from the NIST 2004 DNA quantitation study. J. Forensic Sci., 50(3): 571-578.

 

http://www.nist.gov/public_affairs/techbeat/tb2007_1025.htm#dna

Materials:

SRM 2372 is composed of three components labeled A, B, and C.  All components are human genomic DNA.  Component A was prepared at NIST from Buffy coat white blood cells from single source anonymous male.  Component B was prepared at NIST from Buffy coat cells from multiple anonymous female donors.  Component C was obtained as a commercial lyophilized human genomic extract and has both male and female donors.

Component B is the second preparation of a multi-donor female material.  The original preparation (referred to hereafter as �B_old�) proved insufficiently homogeneous and so unfit-for-purpose.  The preparation and analysis of both the B_old and B components is described below.

Based upon projected five-year demand, 1700 units of each component were targeted for production.  Each set of 1700 vials, each vial containing approximately 110 �L of a DNA solution with a nominal [DNA] of ≈ 50 ng/�L, requires approximately 10.5 mg of DNA in a volume of 210 mL.  Due to the stabilizing shell of bound water molecules and salts surrounding intact DNA molecules, these materials could not be prepared gravimetrically.  Attempts to achieve a suitably weighable material by removing this stabilizing shell lead to irreversible degradation of the DNA; further, as described below, 5 mg of the lyophilized component C material was only sufficient to prepare 50 mL of solution at the desired [DNA] (as determined by UV/vis absorption) after several weeks of equilibration.

Certification and Information Values

The nominal [DNA] of an aqueous DNA solution is derived from the widely-accepted assertion that for a solution of double stranded DNA, an optical density at 260 nm of 1.0 corresponds to a [DNA] of 50 �g/mL (50 ng/�L) [[i], [ii]].  Optical densities at four additional wavelengths (230 nm, 270 nm, 280 nm and 330 nm) are also traditionally used in the assessment of DNA quality [2].  The SRM 2372 component materials are therefore certified for Decadic Attenuance at 230 nm, 260 nm, 270 nm, 280 nm and 330 nm using UV/vis spectrophotometry.  These measurements were performed on the HAS II Reference Spectrophotometer.

The extraction method used for components A, B_old and B and the handling of the extracted DNA for all components were designed to ensure production and maintenance of double stranded DNA.  The A, B_old, B and C materials were prepared to have nominal [DNA] of 50 ng/�L.  The [DNA]s derived from the absorbance measurements will be included in the Certificate of Analysis as Information Values.

Figure 1 displays the absorption spectra of the A, B_new, and C materials from 220 nm to 345 nm. Reference 2 states that the absorbances at 230 nm, 260 nm, 270 nm, 280 nm, and 330 nm are of specific interest:

The reading at 260 nm allows calculation of the concentration of nucleic acid in the sample. An OD of 1 corresponds to ≈50 �g/mL for double-stranded DNA �. The ratio between the readings at 260 nm and 280 nm (OD₂₆₀:OD₂₈₀) provides an estimate of the purity of the nucleic acid. Pure preparations of DNA � have OD₂₆₀:OD₂₈₀ values of 1.8 �. If there is significant contamination with protein � the OD₂₆₀:OD₂₈₀ will be less than [1.8], and accurate quantitation of the amount of nucleic acid will not be possible.� estimates of purity based on OD₂₆₀:OD₂₈₀ ratios are accurate only when the preparations are free of phenol. Water saturated with phenol absorbs with a characteristic peak at 270 nm and an OD₂₆₀:OD₂₈₀ of 2 �. Nucleic acid preparations free of phenol should have OD₂₆₀:OD₂₇₀ ratios of ≈1.2. � Significant absorption at 230 nm indicates contamination by phenolate ion, thiocyanates, and other organic compounds, whereas absorption at higher wavelengths (330 nm or higher) is usually caused by light scattering and indicates the presence of particulate matter.

Figure 1. Absorbance Spectra

For all three materials: 1) the ratios between observed absorbance values at 260 nm and 280 nm are 1.8 or greater, 2) there is no evidence of a peak at 270 nm and the ratios between absorbance values at 260 nm and 270 nm are ≈1.2, and 3) there is no evidence of a peak at 230 nm. The observed absorbances at 330 nm are small but not insignificant.  This issue is addressed in Reference 4.  With the caveat that the measured spectrophotometric absorbance may include a small scattering component, the A, B_new, and C materials all appear to be potentially fit for use as components of a DNA quantitation standard.

UV/vis Verification and Homogeneity

During the homogeneity testing of Component B_new, boxes 16 and 17 were found to be outliers.  Since the vials in these boxes represent the solution after commencement of inefficient stirring, all vials in these two boxes were removed from the pool of SRM materials.

The observed among-vial standard deviations for the measured absorbance at 260 nm are summarized in Table 1.

Table 1.  Results of the UV Homogeneity Study

qPCR Commutability Studies.

While the interlaboratory study supplied some information about systematic differences in response to the A, B_old and C materials among the available qPCR methods, studies were conducted at NIST to provide greater detail for specific assays [[iii],[iv],[v]] and to evaluate Component B_new.  Four commercially available DNA quantitation standards were used for these commutability studies.

One 96-well plate was used for each of the three qPCR methods evaluated.  The design of the studies is shown schematically in Figure 2.

Figure 2. qPCR Commutability Design

Note:      �X_1� (where �X� represents the component A, B or C) is the 1:10 diluted material, X_2 is a 1:5 dilution of X_1, X_3 is a 1:2 dilution of X_2, X_5 is a 1:2 dilution of X_4, X_6 is a 1:2 dilution of X_5, X_7 is a 1:2 dilution of X_6, and X_8 is a 1:2 dilution of X_7.  �SP_1A� to �SP_1D� are 1:10 dilutions of four commercially available DNA calibration materials, �SP_2x� (where �x� represents one of the four commercial materials) is a 1:5 dilution of SP_1x, SP_3x is a 1:2 dilution of SP_2x, and SP_4x is a 1:2 dilution of SP_3x.  Blank� denotes a buffer-only negative control. SP_A Vendor A, SP_B Vendor B, SP_C Vendor C, SP_D Vendor D. Vendor A,B,and C reported [DNA] 200 ng/�L. Vendor D reported [DNA] 262 ng/ �L.

qPCR Homogeneity Results

A Quantifiler Human Kit was used in conjunction with a ABI 7900 real time PCR instrument to access amplification homogeneity.  Table 2 lists the observed homogeneity precision for duplicate analyses of 17 samples, one from each box, for the three components expressed both in terms of the measured crossing threshold cycle (Ct) and in approximate [DNA].  (The Ct is the PCR cycle at which an increase in the Reporter fluorescence above a baseline signal can be detected.)

Table 2.  Homogeneity Study, Quantifiler Human Assay

The repeatability imprecision of the method, s_repeat, for each component is estimated from the standard deviation of the replicate pairs of measurements, pooled over the 17 boxes.  The among-box homogeneity, s_hetero, is estimated from the standard deviation of the means of the replicate pairs and the s_repeat estimate as described in ref [[vi]].  Since each PCR amplification cycle approximately doubles the target [DNA], the standard deviations of the Ct results can be converted to multiplicative standard deviations of [DNA], by exponentiation: s([DNA]) = 2^s(Ct).  These multiplicative values are more readily interpreted as approximate relative standard deviations, expressed in percent: %RSD = 100*(s([DNA]) � 1) = 100*(2^s(Ct) � 1).

The calibration relationships for data from two sets of triplicate analyses for the three components are displayed in Figure 3.  The calibration slopes for the three components are all greater than �0.91 change in Ct per doubling of [DNA], with the maximum theoretical change being �1.0 Ct per doubling.

Figure 3. qPCR Homogeneity, Quantifiler Human Assay

Note: Error bars represent approximate 95% confidence intervals.

qPCR Commutability Study Results:

The systematic biases suggested in the interlaboratory study were verified with three exemplar qPCR assays: Quantifiler Human Assay [6], Alu [8] and CFS [7].  Figure 4A to 4C display the relationship between [DNA], assigned using the conventional relationship between absorbance and concentration [2] and the certified spectrophotometric values for materials A, B_new, and C [4], and the average observed cycle threshold value (Ct) for each of seven or eight solutions (the undiluted component and volumetric dilutions to about 10 ng/�L, 5 ng/�L, 2.5 ng/�L, 1.25 ng/�L, 0.62 ng/�L, 0.31 ng/�L and 0.16 ng/�L) of each component using these assays. Error bars represent approximate 95% confidence intervals on the Ct determinations.  Trend lines within each of the Figure 4 panels illustrate the extent of agreement among the results. The prediction equations for each component with each assay are listed within the panels: predicted [DNA] = 2(Ct � b) / m, where b is the intercept and m the slope of the linear regression of observed Ct against log2([DNA]).  For an ideally efficient PCR reaction, each PCR cycle doubles the target [DNA]; therefore, the ideal value for m is �1.0 Ct per log2 increment.

An inset graph in each panel displays the predicted [DNA] of four commercially available DNA quantification materials relative to each of the three SRM 2372 components. These materials are labeled �Std1� to �Std4�. The thin horizontal lines denote the nominal [DNA] of the materials: Std1 to Std3 at 200 ng/�L DNA and Std4 at 262 ng/�L DNA. Error bars represent approximate 95% confidence intervals on the average predicted [DNA] based on replicate evaluations of 10-fold, 50-fold, 100-fold, and 200-fold volumetric dilutions of each material.

The Quantifiler Human Assay results, shown in Figure 4A, are consistent among the three materials and linear over the entire 320-fold range of dilution. The Alu assay results, Figure 4B, are consistent among the components but systematically deviate from linearity over part of the range. (The undiluted components were outside the assay range of the Alu assay and so are not displayed.) The CFS assay results, Figure 4C, are moderately linear over the entire concentration range but material C is systematically biased by about one-half of an amplification cycle relative to A and B_new.

The observed among-method variability is not unexpected since different qPCR methods exploit qualitatively different molecular targets. This strongly suggests that all three components should be used to elucidate potential bias when using these materials to value assign in-house DNA solutions for non-gender specific qPCR assays.

Figure 4A: qPCR Commutability, Quantifiler Human Assay

Figure 4B.  qPCR Commutability, Alu Assay

Figure 4C.  qPCR Commutability, CFS Assay

Return to STRBase