Abstract
Objective
To examine the correlation in genes expressed in paired umbilical cord blood (UCB) and newborn blood (NB).
Methods
Total mRNA and mRNA of three gene sets (inflammatory, hypoxia, and thyroidal response) was assessed using microarray in UCB and NB spotted on Guthrie cards from 7 mother/infant pairs.
Results
The average gene expression correlation between paired UCB and NB samples was 0.941 when all expressed genes were considered, and 0.949 for three selected gene sets.
Conclusion
The high correlation of UCB and NB gene expression suggest that either source may be useful for examining gene expression in the perinatal period.
Keywords: messenger RNAs, genes and transcripts, genome-wide, oligonucleotide microarrays, archived newborn blood spots
INTRODUCTION
Umbilical cord blood has been commonly used in perinatal studies for the purpose of determining a variety of exposures at birth, including heavy metals [1], infections and inflammation [2], and asphyxiating conditions [3]. This blood can be difficult to obtain and is rarely available for large population studies. Neonatal blood, on the other hand, is routinely collected on Guthrie cards during the first several days of life for use in newborn screening programs [4]. Most programs consume only a small portion of the available blood; the remainder is often archived under various conditions and time periods. In the US, 14 newborn screening programs retain newborn blood for 21 years or more, constituting a unique resource for the retrospective study of diseases of perinatal origin [5].
Little is known about the utility of this resource for large-scale investigations of gene expression. While some studies have used archived newborn blood spots to investigate human and viral DNA or proteins, our team focused on gene quantitative RNA expression [5,6]. Newborn blood is collected 24–48 hours after birth and it is of interest to know whether the gene expression findings in neonatal blood parallel or differ from those in umbilical cord blood from the same infant. We undertook this study to evaluate if gene expression, which can reflect many environmental exposures, is similar in umbilical cord blood and newborn blood when both are collected in the same way, spotted onto Guthrie filter papers and preserved in the same fashion.
METHODS
Participants
This study was approved by the Michigan State University and Sparrow Hospital Institutional Review Boards. A convenience sample of 14 pregnant women without recognized pregnancy complications was recruited from a Lansing, MI OB/GYN clinic. Study staff obtained written informed consent for the collection of umbilical cord blood and newborn blood by study physicians. Women were excluded if they delivered preterm or by cesarean. A total of 7 mother/infant pairs were available for analysis, after excluding 2 women who delivered by cesarean, one woman who delivered at a non-participating hospital, and 4 women in whom either the umbilical cord or newborn blood was not obtained.
Data Collection
After delivery of the infant and cutting of the umbilical cord, but prior to delivery of the placenta, study physicians filled 3–5 circles of 1.25 cm diameter on standard Guthrie cards (Whatman, Kent, UK) with blood from the umbilical cord. Michigan law mandates that hospitals obtain 5 circles of 1.25 cm diameter of neonatal blood on Guthrie cards via heel stick blood draw between 24–72 hours after the newborn’s birth. During this time, 3–5 additional neonatal blood spots were collected specifically for this study on a separate Guthrie card. All bloodspots were allowed to dry at room temperature. After the neonatal bloodspot was collected, the study physician recorded the infant’s date of birth, gestational age, gender, race/ethnicity, maternal age, and delivery mode from the newborn’s medical record. Prior to analysis, all blood spots were stored at ambient temperatures. The period from collection of specimen to mRNA extraction in the laboratory was 176.8 days on average.
RNA Isolation and Microarray Procedures
RNA isolation, cDNA synthesis, and microarray procedures on the bloodspot specimens were conducted in the Laboratory of Microarray Technology at the Van Andel Institute according to published protocols [6]. In brief, three 3-mm punches from each specimen were homogenized and RNA isolated using the illustra RNAspin Mini kit (GE Healthcare, Buckinghamshire, UK). All RNA is subjected to DNase treatment to remove DNA contamination. RNA quality and quantity were evaluated by the Agilent BioAnalyzer via a RNA Pico Lab Chip (Agilent Technologies, Santa Clara, CA). The WT-Ovation Pico RNA Amplification System (NuGEN Technologies, San Carlos, CA) was used to generate cDNA before labeling and hybridization onto the Agilent whole human genome gene expression 8×60K microarray. The microarrays were scanned with an Agilent G3 high-resolution scanner and microarray data were extracted using Feature Extraction software v.10.7.3.1. (Agilent).
Statistical Methods
Selection of gene sets for this study was based on our previous work, in which we investigated the expression of 7 gene sets (3 empirical; 4 canonical) representing four physiological pathways (inflammatory, hypoxic, thyroidal, and coagulative) hypothesized to contribute to the development of cerebral palsy. We found significant differences between children with CP and controls in gene regulation on the empirical inflammatory [7], hypoxic [8], and thyroidal [9] gene sets [10]. Inasmuch as these gene sets appear to be involved in important pathways relevant to long term outcomes, we assess expression of these gene sets, as well as total gene expression, in our comparison of umbilical cord and newborn blood.
To examine similarities between newborn blood and umbilical cord blood on overall expressed-gene and specific gene set expression, we calculated both pair-wise interclass correlation coefficients and the intraclass correlation coefficients. Means and standard deviations were calculated and results were displayed graphically. Additionally, a modified GAGE method was used to test the differentially expressed genes in the 7 pairs.
RESULTS
The total RNA yield was 9.87±5.80 ng/ul and 8.38± 5.01 ng/ul for umbilical cord blood and newborn blood, respectively. The RNA integrity number was 2.07±0.19 for umbilical cord blood and 2.03±0.14 for newborn blood.
Microarray analysis of umbilical cord blood spots detected 9,749–19,101 expressed-genes, while in newborn bloodspot 10,859–17,539 mRNA species were detected. The mean number of genes expressed in the umbilical cord blood spot was 14,983; while the average number of genes expressed in the newborn blood spot was 13,339. Table 1 displays the correlation coefficients for each umbilical cord/newborn blood spot pair. The average Pearson correlation coefficient between paired umbilical cord and newborn bloodspots was 0.941(SD=0.018) for overall expressed-genes. The average correlation coefficient of genes expressed on the 3 specific genes sets investigated in this study was 0.949 (SD=0.242). Of 21,266 detected genes, 214 (1%) were differentially expressed (p-value<2.35e-6).
For comparison, we evaluated the correlation between all 21 possible pairs of cord blood specimens and newborn blood specimens. The obtained correlation coefficients also reached a high level, although not as high as those for paired sample. The average correlation coefficients for overall expressed genes were 0.808 (SD=0.102) and 0.836 (SD=0.110) among umbilical cord samples and among newborn blood samples, respectively (Supplemental Digital Content Figure S1). The average correlation coefficients on the 3 specific genes sets were 0.870 (SD=0.059) and 0.883 (SD=0.054) among umbilical cord samples and among newborn blood samples, respectively (Supplemental Digital Content Figure S2).
DISCUSSION
While several studies have established the reliability and reproducibility of gene expression in newborn blood spots, we believe this is the first study to assess the correlation between newborn blood and umbilical cord blood spotted on filter paper. We demonstrate that gene expression in newborn blood reflects umbilical cord blood very closely.
The principal findings of this study are that: 1) abundant gene expression was observed in examining both umbilical cord blood and newborn blood on Guthrie cards; 2) inflammatory, hypoxic, and thyroidal genes crucial for newborn adaptation were consistently expressed in both types of specimens; 3) overall gene expression and specific empirical gene set expression in umbilical cord blood and newborn blood are highly correlated for matched pairs (correlation coefficient > 0.94); and 4) gene expression does not change dramatically in healthy newborns between delivery and the first day or two of life since the between-pair correlations within the same individual are high.
It is therefore possible that newborn blood stored on Guthrie filter paper may be used as an alternative to umbilical cord blood in gene expression studies to investigate the physiological and environmental exposures at birth. We emphasize, however, that this study was done in healthy newborns. It is possible that newborns who become ill after birth may mobilize genes that were not expressed at the time of birth. Our findings support the value of neonatal blood spots for genome-wide gene expression profiling and pathway studies in perinatal research.
Supplementary Material
Table 1.
Correlation coefficients for matched umbilical cord/newborn blood pairs for overall and specific gene set expression.
| Cord/Newborn Blood Pair |
Overall Gene Expression Correlation Coefficient |
Specific Gene Sets Correlation Coefficient |
|---|---|---|
| 1 | 0.959 | 0.974 |
| 2 | 0.936 | 0.958 |
| 3 | 0.942 | 0.947 |
| 4 | 0.913 | 0.901 |
| 5 | 0.963 | 0.970 |
| 6 | 0.946 | 0.959 |
| 7 | 0.927 | 0.949 |
ACKNOWLEDGEMENTS
We would like to thank the members of the OWL (Outcome, Wellness and Life history in cerebral palsy) study team including Marcee Schoenbach, Lynette Biery, Samantha Hansen, Dr. Colleen Berry, Dr. Mohammed Mahmoud, and study participants for their assistance with the collection of data.
This research was supported in part by the Perinatology Research Branch, Division of Intramural Research, Eunice Kennedy Shriver NICHD, NIH, DHHS; and funded by NIH R01 NS055101 to Dr. Paneth. Dr. Slaughter, is funded by T32 HD046377.
Footnotes
DECLARATION OF INTERESTS
The authors have no conflicts of interests.
REFERENCES
- 1.Linden A, Adams JL, Roberts N. Using propensity scores to construct comparable control groups for disease management program evaluation. Disease Management & Health Outcomes. 2005;13(2):107–115. [Google Scholar]
- 2.Ravallion M. The mystery of the vanishing benefits: An introduction to impact evaluation. the world bank economic review. 2001;15(1):115–140. [Google Scholar]
- 3.Gomez R, Romero R, Ghezzi F, Yoon BH, Mazor M, Berry SM. The fetal inflammatory response syndrome. American journal of obstetrics and gynecology. 1998;179(1):194–202. doi: 10.1016/s0002-9378(98)70272-8. [DOI] [PubMed] [Google Scholar]
- 4.Wilcken B. Screening for disease in the newborn: the evidence base for blood-spot screening. Pathology-Journal of the RCPA. 2012;44(2):73. doi: 10.1097/PAT.0b013e32834e843f. [DOI] [PubMed] [Google Scholar]
- 5.Haak PT, Busik JV, Kort EJ, Tikhonenko M, Paneth N, Resau JH. Archived unfrozen neonatal blood spots are amenable to quantitative gene expression analysis. Neonatology. 2009;95(3):210–216. doi: 10.1159/000155652. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Khoo SK, Dykema K, Vadlapatla NM, LaHaie D, Valle S, Satterthwaite D, Ramirez SA, Carruthers JA, Haak PT, Resau JH. Acquiring genome-wide gene expression profiles in Guthrie card blood spots using microarrays. Pathology international. 2011;61(1):1–6. doi: 10.1111/j.1440-1827.2010.02611.x. [DOI] [PubMed] [Google Scholar]
- 7.Madsen-Bouterse SA, Romero R, Tarca AL, Kusanovic JP, Espinoza J, Kim CJ, Kim JS, Edwin SS, Gomez R, Draghici S. The transcriptome of the fetal inflammatory response syndrome. Am. J. Reprod. Immunol. 2009;63(1):73–92. doi: 10.1111/j.1600-0897.2009.00791.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Chi JT, Wang Z, Nuyten DSA, Rodriguez EH, Schaner ME, Salim A, Wang Y, Kristensen GB, Helland Å, Børresen-Dale AL. Gene expression programs in response to hypoxia: cell type specificity and prognostic significance in human cancers. PLoS medicine. 2006;3(3):e47. doi: 10.1371/journal.pmed.0030047. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Clément K, Viguerie N, Diehn M, et al. In vivo regulation of human skeletal muscle gene expression by thyroid hormone. Genome Res. 2002;12(2):281–291. doi: 10.1101/gr.207702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ho NT, Furge KA, Fu W, et al. Gene expression in archived newborn blood spots distinguishes infants who will later develop cerebral palsy from matched controls. Prediatric Research. 73(4-1):450–456. doi: 10.1038/pr.2012.200. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.