ORIGINAL ARTICLE
BIOPRESERVATION AND BIOBANKING Volume 14, Number 2, 2016 ª Mary Ann Liebert, Inc. DOI: 10.1089/bio.2015.0075
Quality Analysis of DNA from Cord Blood Buffy Coat: The Best Neonatal DNA Source for Epidemiological Studies? Guangdi Zhou,1 Qin Li,2 Lisu Huang,3 Yuhang Wu,1 Meiqin Wu,1 and Weiye C. Wang1
Background: Umbilical cord blood is an economical and easy to obtain source of high-quality neonatal genomic DNA. However, although large numbers of cord blood samples have been collected, information on the yield and quality of the DNA extracted from cord blood is scarce. Moreover, considerable doubt still exists on the utility of the buffy coat instead of whole blood as a DNA source. Methods: We compared the sample storage and DNA extraction costs for whole blood, buffy coat, and all-cell pellet. We evaluated three different DNA purification kits and selected the most suitable one to purify 1011 buffy coat samples. We determined the DNA yield and optical density (OD) ratios and analyzed 48 singlenucleotide polymorphisms using time-of-flight mass spectrometry (TOF MS). We also analyzed eight possible preanalytical variables that may correlate with DNA yield or quality. Results: Buffy coat was the most economical and least labor-intensive source for sample storage and DNA extraction. The average yield of genomic DNA from 200 mL of buffy coat sample was 16.01 – 8.00 mg, which is sufficient for analytic experiments. The mean A260/A280 ratio and the mean A260/A230 ratio were 1.89 – 0.09 and 1.95 – 0.66, respectively. More than 99.5% of DNA samples passed the TOF MS test. Only hemolysis showed a strong correlation with OD ratios of DNA, but not with yield. Conclusion: Our findings show that cord blood buffy coat yields high-quality DNA in sufficient quantities to meet the requirements of experiments. Buffy coat was also found to be the most economic, efficient, and stable source of genomic DNA. samples. Thus, the use of buffy coat can save time, labor, and money. Besides the sufficient numbers of donors, cohort studies also have well-recorded detailed information, including sample quality data, perinatological data, and details of follow-up visits,3,4 which are beneficial for future studies. Therefore, cord blood buffy coat is valuable to perinatology researchers who plan to use the DNA for downstream applications and analysis.5–8 However, Gail pointed out in 2013 that the buffy coat yields only half as much DNA as whole blood or all cell pellet (ACP), which agrees with our results (Supplementary Table S1; Supplementary Data are available online at www.liebertpub.com/bio). To maximize the DNA yield, researchers use ACP rather than buffy coat.1 However, because the volume of whole blood or ACP is much larger compared with the buffy coat, storage of whole blood or ACP would consume several times the costs of storing buffy coat and would require considerably more storage tubes, refrigerator or freezer space, bar codes, boxes, shelves, and
Introduction
W
hole blood and its fractions are the preferred DNA source for most epidemiologic studies because they yield large quantities of high-quality DNA for downstream molecular biological experiments, including wholegenome scans, DNA sequencing, DNA methylation studies, and DNA adduct detection.1 Due to large birth cohort studies conducted over the last 20 years, biobanks have accumulated numerous samples of umbilical cord blood and buffy coat samples.2 The buffy coat consists of most of the white blood cells in a volume of whole blood. It is rich in genomic DNA and can be easily isolated from whole blood samples by centrifugation. For large-scale cohort studies where storage-related issues are a major concern, it is ideal to store buffy coat instead of whole blood samples for genomic DNA extraction. Because 1 mL of buffy coat is prepared from 10 mL of anticoagulated cord blood, 90% of storage tubes and space can be saved by using buffy coat
1 MOE-Shanghai Key Laboratory of Children’s Environmental Health, Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China. 2 Department of Obstetrics and Gynecology, Changhai Hospital Affiliated to the Second Military Medical University, Shanghai, China. 3 Department of Pediatrics, Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China.
1
2
labor (for aliquoting and sample management). Importantly, the DNA extraction process will become much more expensive and labor intensive. Therefore, the following questions are worth considering: (1) What is the common range of DNA yield from cord blood buffy coat collected by the birth cohort studies? Is the yield sufficient for subsequent studies? (2) Should whole blood or ACP, which are expensive, but yield more DNA, be stored instead of buffy coat? The answers to these questions will help researchers in designing appropriate procedures for blood sample collection and balancing costs and DNA yield. Most of the downstream experiments not only have specific requirements for DNA quantity but also have specific criteria for DNA quality. Low DNA concentration or poor quality creates challenges in the research process, which can lead to false experimental results.9,10 Researchers would benefit from knowing if the DNA extracted from buffy coat meets the requirements for the chosen experimental method as it would help to improve the approach to sample storage. Unfortunately, only few publications have addressed the quality or yield of DNA extracted from cord blood buffy coat; however, several reports have discussed the yield and quality of the DNA extracted from peripheral blood buffy coat.5,6 Mychaleckyj et al. isolated 120 buffy coat samples for a clinical trial to assess the quality of DNA stored up to 9 years and analyzed several possible predictive factors that may influence DNA yield.5 Caboux et al. extracted DNA from 50,000 peripheral buffy coat specimens and pointed out several factors that significantly influenced DNA yield.6 In this work, we focused on the quantity and quality of the cord blood buffy coat collected by birth cohort studies. We compared the costs for storage of and DNA extraction from whole blood, buffy coat, and ACP. We isolated the genomic DNA of 1011 buffy coat specimens collected from a birth cohort and identified 48 single-nucleotide polymorphisms (SNPs) using time-of-flight mass spectrometry (TOF MS). Next, we analyzed whether eight perinatological factors would significantly affect the yield and quality of the DNA. In short, we investigated whether cord blood buffy coat would be a good source for DNA-based experiments and whether it is the best format for blood storage.
Methods Sample collection and information retrieval The samples and their associated information were obtained from a birth cohort study conducted at the Shanghai Xinhua Hospital and Shanghai International Peace Maternity and Child Health Hospital. Cord blood was collected from 1011 newborns, and information of the newborns and their mothers was obtained through a questionnaire survey and medical history. All the puerperae signed informed consent forms, and the protocol was approved by the Xinhua Hospital Institutional Review Board. The cord blood was collected by the midwife with a No. 20-gauge needle. Within 10 minutes after birth, 20 mL cord blood was drawn from the umbilical cord into a single-use syringe and injected into an EDTA-K2 anticoagulation vacuum blood tube and a coagulation vacuum blood tube. The tubes were inverted 5–6 times and transported in a cold (around 10C–18C) container to the laboratory. The blood was centrifuged for 5 minutes at 1100 g and 4C and fractionated into serum/plasma, buffy coat, red blood cells, or
ZHOU ET AL.
clot. If the buffy coat and red blood cells were not separated, they were called ACP. These fractions were aliquoted into 1.2-mL storage tubes and placed in -80C freezers for longterm storage. The whole process from collection to storage was completed within 90 minutes.9,11 Basic sample information, questionnaires, and medical histories were collected.
DNA purification, quantification, and quality determination After 6 months of storage at -80C, the buffy coat specimens were retrieved and purified for genomic DNA extraction. We used whole blood DNA purification kits to extract the DNA from the buffy coat samples. We evaluated the QIAamp DNA blood Mini Kit (Cat. No. 51106; Qiagen), RelaxGene Blood DNA System (0.1–20 mL, Cat. No. DP319-01; Tiangen), and AxyPrep Blood Genomic DNA Miniprep Kit (Cat. No. AP-MN-BL-GDNA-50; Axygen) and selected the most suitable one for batch extraction. All kits were used according to the manufacturer’s instructions. Each kit required 200 mL of buffy coat sample. We measured the concentration, A260/A280 ratio, and A260/A230 ratio of the DNA solution using an Epoch Multi-Volume Spectrophotometer System (BioTek, Inc.).12–14
SNP analysis using TOF MS We selected 48 SNPs for analysis. We used the MassARRAY designer software to automatically design both polymerase chain reaction (PCR) and MassEXTEND primers for multiplexed assays.15 Genomic DNA was amplified by PCR. We used a mixture of chain-terminating ddNTPs in a primer extension assay designed to detect sequence differences at the single-nucleotide level. The primer was extended depending upon the template sequence, resulting in an allele-specific difference in mass between the extension products, allowing differentiation between SNPs. Samples were transferred into the 384 SpectroCHIP, and the mass and correlating genotype were determined in real time. The results were automatically loaded into a database that allows convenient data analysis.
Preanalysis factors and data analyses We chose five of the most common perinatological factors—three for the puerpera (gestational week, delivery mode, and pregnancy complication) and two for the newborn (sex and birth weight)—and determined their correlation with the DNA yield and optical density (OD) ratios. According to the perinatological definition, we divided the newborns into three groups based on their birth weight: 4000 g (overweight). The puerperae were divided into two groups based on their gestational weeks: 0.05, indicating no significant correlation between the five factors
QUALITY ANALYSIS OF DNA FROM CORD BLOOD BUFFY COAT
5
FIG. 2. Concentration of isolated DNA and A260/A280 and A260/A230 ratios. (A) Frequency distribution charts of concentration, A260/A280 ratios, and A260/A230 ratios of DNA extracted using the Qiagen kit (n = 1011). Each DNA sample was dissolved in 200 mL of the solvent included in the kit. Each distribution was approximately normal. (B) Box plots of the concentration, A260/A280 ratios, and A260/A230 ratios of DNA extracted from the buffy coat (n = 1011). Each DNA sample was dissolved in 200 mL of the solvent included in the kits. OD, optical density. and the yield, A260/A280 ratio, and A260/A230 ratio of the extracted DNA. The details are shown in Table 4. We also analyzed the effects of sample quality on DNA yield and OD ratios. While most of the buffy coat samples were of good quality, a few specimens exhibited problems, such as clotting, hemolysis, and small blood volumes. To assess whether these problems affected DNA quality, we compared the OD ratios of DNA purified from the problem samples with those from the normal samples. The results are shown in Table 5. Clotting and small blood volumes affected neither yield nor OD ratios. However, hemolysis in
buffy coat samples led to significantly higher A260/A280 and A260/A230 ratios ( p < 0.01) and was strongly correlated with OD ratios, but did not significantly affect yield.
Discussion Although the number of studies based on cohort genomic DNA samples has increased dramatically, no widely accepted standard practice has been established for the storage of blood samples. The most economical and feasible way to store whole blood and its fractions, while ensuring high
Table 4. Correlation Between Five Perinatological Factors and the Quality and Concentration of Extracted DNA Concentration (ng/lL) A260/A280 ratios A260/A230 ratios Factor Sex
Group
Male Female Birth weight 4000 g Gestational week
BIOPRESERVATION AND BIOBANKING Volume 14, Number 2, 2016 ª Mary Ann Liebert, Inc. DOI: 10.1089/bio.2015.0075
Quality Analysis of DNA from Cord Blood Buffy Coat: The Best Neonatal DNA Source for Epidemiological Studies? Guangdi Zhou,1 Qin Li,2 Lisu Huang,3 Yuhang Wu,1 Meiqin Wu,1 and Weiye C. Wang1
Background: Umbilical cord blood is an economical and easy to obtain source of high-quality neonatal genomic DNA. However, although large numbers of cord blood samples have been collected, information on the yield and quality of the DNA extracted from cord blood is scarce. Moreover, considerable doubt still exists on the utility of the buffy coat instead of whole blood as a DNA source. Methods: We compared the sample storage and DNA extraction costs for whole blood, buffy coat, and all-cell pellet. We evaluated three different DNA purification kits and selected the most suitable one to purify 1011 buffy coat samples. We determined the DNA yield and optical density (OD) ratios and analyzed 48 singlenucleotide polymorphisms using time-of-flight mass spectrometry (TOF MS). We also analyzed eight possible preanalytical variables that may correlate with DNA yield or quality. Results: Buffy coat was the most economical and least labor-intensive source for sample storage and DNA extraction. The average yield of genomic DNA from 200 mL of buffy coat sample was 16.01 – 8.00 mg, which is sufficient for analytic experiments. The mean A260/A280 ratio and the mean A260/A230 ratio were 1.89 – 0.09 and 1.95 – 0.66, respectively. More than 99.5% of DNA samples passed the TOF MS test. Only hemolysis showed a strong correlation with OD ratios of DNA, but not with yield. Conclusion: Our findings show that cord blood buffy coat yields high-quality DNA in sufficient quantities to meet the requirements of experiments. Buffy coat was also found to be the most economic, efficient, and stable source of genomic DNA. samples. Thus, the use of buffy coat can save time, labor, and money. Besides the sufficient numbers of donors, cohort studies also have well-recorded detailed information, including sample quality data, perinatological data, and details of follow-up visits,3,4 which are beneficial for future studies. Therefore, cord blood buffy coat is valuable to perinatology researchers who plan to use the DNA for downstream applications and analysis.5–8 However, Gail pointed out in 2013 that the buffy coat yields only half as much DNA as whole blood or all cell pellet (ACP), which agrees with our results (Supplementary Table S1; Supplementary Data are available online at www.liebertpub.com/bio). To maximize the DNA yield, researchers use ACP rather than buffy coat.1 However, because the volume of whole blood or ACP is much larger compared with the buffy coat, storage of whole blood or ACP would consume several times the costs of storing buffy coat and would require considerably more storage tubes, refrigerator or freezer space, bar codes, boxes, shelves, and
Introduction
W
hole blood and its fractions are the preferred DNA source for most epidemiologic studies because they yield large quantities of high-quality DNA for downstream molecular biological experiments, including wholegenome scans, DNA sequencing, DNA methylation studies, and DNA adduct detection.1 Due to large birth cohort studies conducted over the last 20 years, biobanks have accumulated numerous samples of umbilical cord blood and buffy coat samples.2 The buffy coat consists of most of the white blood cells in a volume of whole blood. It is rich in genomic DNA and can be easily isolated from whole blood samples by centrifugation. For large-scale cohort studies where storage-related issues are a major concern, it is ideal to store buffy coat instead of whole blood samples for genomic DNA extraction. Because 1 mL of buffy coat is prepared from 10 mL of anticoagulated cord blood, 90% of storage tubes and space can be saved by using buffy coat
1 MOE-Shanghai Key Laboratory of Children’s Environmental Health, Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China. 2 Department of Obstetrics and Gynecology, Changhai Hospital Affiliated to the Second Military Medical University, Shanghai, China. 3 Department of Pediatrics, Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China.
1
2
labor (for aliquoting and sample management). Importantly, the DNA extraction process will become much more expensive and labor intensive. Therefore, the following questions are worth considering: (1) What is the common range of DNA yield from cord blood buffy coat collected by the birth cohort studies? Is the yield sufficient for subsequent studies? (2) Should whole blood or ACP, which are expensive, but yield more DNA, be stored instead of buffy coat? The answers to these questions will help researchers in designing appropriate procedures for blood sample collection and balancing costs and DNA yield. Most of the downstream experiments not only have specific requirements for DNA quantity but also have specific criteria for DNA quality. Low DNA concentration or poor quality creates challenges in the research process, which can lead to false experimental results.9,10 Researchers would benefit from knowing if the DNA extracted from buffy coat meets the requirements for the chosen experimental method as it would help to improve the approach to sample storage. Unfortunately, only few publications have addressed the quality or yield of DNA extracted from cord blood buffy coat; however, several reports have discussed the yield and quality of the DNA extracted from peripheral blood buffy coat.5,6 Mychaleckyj et al. isolated 120 buffy coat samples for a clinical trial to assess the quality of DNA stored up to 9 years and analyzed several possible predictive factors that may influence DNA yield.5 Caboux et al. extracted DNA from 50,000 peripheral buffy coat specimens and pointed out several factors that significantly influenced DNA yield.6 In this work, we focused on the quantity and quality of the cord blood buffy coat collected by birth cohort studies. We compared the costs for storage of and DNA extraction from whole blood, buffy coat, and ACP. We isolated the genomic DNA of 1011 buffy coat specimens collected from a birth cohort and identified 48 single-nucleotide polymorphisms (SNPs) using time-of-flight mass spectrometry (TOF MS). Next, we analyzed whether eight perinatological factors would significantly affect the yield and quality of the DNA. In short, we investigated whether cord blood buffy coat would be a good source for DNA-based experiments and whether it is the best format for blood storage.
Methods Sample collection and information retrieval The samples and their associated information were obtained from a birth cohort study conducted at the Shanghai Xinhua Hospital and Shanghai International Peace Maternity and Child Health Hospital. Cord blood was collected from 1011 newborns, and information of the newborns and their mothers was obtained through a questionnaire survey and medical history. All the puerperae signed informed consent forms, and the protocol was approved by the Xinhua Hospital Institutional Review Board. The cord blood was collected by the midwife with a No. 20-gauge needle. Within 10 minutes after birth, 20 mL cord blood was drawn from the umbilical cord into a single-use syringe and injected into an EDTA-K2 anticoagulation vacuum blood tube and a coagulation vacuum blood tube. The tubes were inverted 5–6 times and transported in a cold (around 10C–18C) container to the laboratory. The blood was centrifuged for 5 minutes at 1100 g and 4C and fractionated into serum/plasma, buffy coat, red blood cells, or
ZHOU ET AL.
clot. If the buffy coat and red blood cells were not separated, they were called ACP. These fractions were aliquoted into 1.2-mL storage tubes and placed in -80C freezers for longterm storage. The whole process from collection to storage was completed within 90 minutes.9,11 Basic sample information, questionnaires, and medical histories were collected.
DNA purification, quantification, and quality determination After 6 months of storage at -80C, the buffy coat specimens were retrieved and purified for genomic DNA extraction. We used whole blood DNA purification kits to extract the DNA from the buffy coat samples. We evaluated the QIAamp DNA blood Mini Kit (Cat. No. 51106; Qiagen), RelaxGene Blood DNA System (0.1–20 mL, Cat. No. DP319-01; Tiangen), and AxyPrep Blood Genomic DNA Miniprep Kit (Cat. No. AP-MN-BL-GDNA-50; Axygen) and selected the most suitable one for batch extraction. All kits were used according to the manufacturer’s instructions. Each kit required 200 mL of buffy coat sample. We measured the concentration, A260/A280 ratio, and A260/A230 ratio of the DNA solution using an Epoch Multi-Volume Spectrophotometer System (BioTek, Inc.).12–14
SNP analysis using TOF MS We selected 48 SNPs for analysis. We used the MassARRAY designer software to automatically design both polymerase chain reaction (PCR) and MassEXTEND primers for multiplexed assays.15 Genomic DNA was amplified by PCR. We used a mixture of chain-terminating ddNTPs in a primer extension assay designed to detect sequence differences at the single-nucleotide level. The primer was extended depending upon the template sequence, resulting in an allele-specific difference in mass between the extension products, allowing differentiation between SNPs. Samples were transferred into the 384 SpectroCHIP, and the mass and correlating genotype were determined in real time. The results were automatically loaded into a database that allows convenient data analysis.
Preanalysis factors and data analyses We chose five of the most common perinatological factors—three for the puerpera (gestational week, delivery mode, and pregnancy complication) and two for the newborn (sex and birth weight)—and determined their correlation with the DNA yield and optical density (OD) ratios. According to the perinatological definition, we divided the newborns into three groups based on their birth weight: 4000 g (overweight). The puerperae were divided into two groups based on their gestational weeks: 0.05, indicating no significant correlation between the five factors
QUALITY ANALYSIS OF DNA FROM CORD BLOOD BUFFY COAT
5
FIG. 2. Concentration of isolated DNA and A260/A280 and A260/A230 ratios. (A) Frequency distribution charts of concentration, A260/A280 ratios, and A260/A230 ratios of DNA extracted using the Qiagen kit (n = 1011). Each DNA sample was dissolved in 200 mL of the solvent included in the kit. Each distribution was approximately normal. (B) Box plots of the concentration, A260/A280 ratios, and A260/A230 ratios of DNA extracted from the buffy coat (n = 1011). Each DNA sample was dissolved in 200 mL of the solvent included in the kits. OD, optical density. and the yield, A260/A280 ratio, and A260/A230 ratio of the extracted DNA. The details are shown in Table 4. We also analyzed the effects of sample quality on DNA yield and OD ratios. While most of the buffy coat samples were of good quality, a few specimens exhibited problems, such as clotting, hemolysis, and small blood volumes. To assess whether these problems affected DNA quality, we compared the OD ratios of DNA purified from the problem samples with those from the normal samples. The results are shown in Table 5. Clotting and small blood volumes affected neither yield nor OD ratios. However, hemolysis in
buffy coat samples led to significantly higher A260/A280 and A260/A230 ratios ( p < 0.01) and was strongly correlated with OD ratios, but did not significantly affect yield.
Discussion Although the number of studies based on cohort genomic DNA samples has increased dramatically, no widely accepted standard practice has been established for the storage of blood samples. The most economical and feasible way to store whole blood and its fractions, while ensuring high
Table 4. Correlation Between Five Perinatological Factors and the Quality and Concentration of Extracted DNA Concentration (ng/lL) A260/A280 ratios A260/A230 ratios Factor Sex
Group
Male Female Birth weight 4000 g Gestational week
