Epsilon globin gene expression in developing human fetal tissues

Abstract

OBJECTIVE:

The discovery of free fetal DNA in plasma of pregnant women has opened a new avenue for non-invasive prenatal diagnosis. We hypothesized that epsilon (ɛ)-globin gene expression could serve as a positive control for the presence of fetal nucleic acid.

STUDY DESIGN:

We measured ɛ-globin mRNA in human fetal tissues and compared concentrations with that measured in adult non-pregnant and pregnant samples. Total RNA was isolated from fetal marrow, liver, blood, and placenta (10–24 weeks gestation), from adult peripheral blood mononuclear cells, and from maternal plasma. RNA was reverse transcribed and quantitative polymerase chain reaction performed for ɛ-globin expression.

RESULTS:

ɛ-globin gene expression was detected in all fetal samples, was detected in plasma of pregnant women, but was negligible in non-pregnant samples. Relative ɛ-globin gene expression was significantly greater in fetal blood compared to fetal liver, and was minimally expressed in placenta. ɛ-globin gene expression decreased at the highest gestational ages in fetal blood, while expression was greatest at 15–19 weeks in fetal marrow.

CONCLUSION:

Fetal ɛ-globin gene expression is significantly greater than adult expression and is increased in maternal plasma compared to non-pregnant samples. ɛ-globin gene expression might serve as a positive control when determining the presence of fetal nucleic acid in total nucleic acid isolated from maternal plasma.

Keywords

Epsilon globin fetal tissues prenatal diagnosis nucleic acid noninvasive

1 Introduction

Isolation of fetal nucleic acids from maternal blood samples for use in non-invasive prenatal diagnosis has gained much attention in recent years [1]. Development of this approach of prenatal diagnosis into a broader clinical setting has been hampered by the lack offetal-specific markers, leading to sometimes inconclusive and false-negative results [2]. Currently, noninvasive prenatal diagnosis using free fetal DNA is mainly used in gender determination, sex-linked disorders, fetal aneuploidies, and determining fetal RhD positive status in RhD negative mothers [1]. In 2013 an estimated half million pregnant women underwent noninvasive prenatal testing [3]. In RhD negative women, prenatal diagnosis involves detection of the RhD gene in maternal plasma. A negative result (no D gene detected) must be further evaluated foradequacy of the fetal deoxyribonucleic acid (DNA) sample given the low concentrations of fetal nucleic acid in the maternal circulation. DNA markers, such as the presence of a Y chromosome, are only applicable to male fetuses [4].

Embryonic and fetal hemoglobin forms are candidates for fetal-specific markers [5], and the presence of epsilon (ɛ)-globin messenger ribonucleic acid (mRNA) might function as a marker for fetal cells early in pregnancy [6 –8]. The genes that code for globin expression are precisely regulated during development. The ɛ-globin gene is expressed primarily in the first trimester. The presence of embryonic hemoglobin, comprised of two alpha (α)-globin chains and two ɛ-globin chains, peaks at four to eight weeks of gestation and disappears by the third trimester. We hypothesized that ɛ-globin gene expression would differ significantly between fetal and adult samples and thereby serve as a marker for the presence of fetal nucleic acid. In addition we hypothesized that ɛ-globin gene expression would be greater in pregnant compared to non-pregnant samples. Fetal tissues evaluated in this study were chosen based on their role in hematopoietic development. In addition to known hematopoietic tissues, placental tissue was evaluated, as many studies have identified trophoblasts as the cell type from which fetal DNA originates [9, 10]. We isolated total RNA from fetal liver, blood, marrow, and placenta, from plasma of pregnant women, and from mononuclear cells of non-pregnant adult blood, and quantified ɛ-globin gene expression using reverse transcription (RT) and polymerase chain reaction (PCR).

2 Study design

2.1 Tissue samples

Human tissues were obtained under appropriate oversight by the University of New Mexico Human Research Review Committee, which determined that the study did not qualify as human subject research as all samples were de-identified prior to collection. There were no conflicts of interest with any of the authors of the study. Human fetal tissue (10–24 weeks of gestational age) was obtained at elective terminations of pregnancy in healthy women. Tissue samples from individual fetal liver, marrow, placenta, and blood isolated from individual fetal umbilical cords were collected during a one year period.

In order to sample the fetal circulation the umbilical cord was isolated and rinsed thoroughly in phosphate buffered saline (PBS) which removed any maternal blood contamination. A small volume (100–800 microliters) of fetal blood was collected by milking each cord. Marrow was isolated by flushing both femurs with PBS. Placenta was obtained from a 1 cm peri-umbilical region following thorough rinsing with PBS. Red blood cells in the placental tissue were lysed using hypotonic washes after trituration of the tissues.

2.2 Adult samples

Peripheral blood samples from non-pregnant adults (n = 11) were collected from discarded blood donation collection systems and diluted 1:2 with PBS. Aliquots (9 ml) of diluted blood were overlaid on 6 ml Ficoll-Paque in a 15 ml tube, and centrifuged at 400× g for 30 min at 18°C. Mononuclear light density cells from the interphase of the gradient were used for total RNA isolation. Cellular RNA was chosen instead of plasma because we felt it unlikely that any RNA could be isolated from the plasma of these discarded samples. All samples were homogenized in 1 mL of a commercial acid guanidinium thiocyanate-phenol chloroform reagent (Trizol^®, Invitrogen) per 50 mg of tissue and stored at minus 80°C until RNA isolation.

Peripheral blood samples from pregnant women 10–18 weeks gestation (n = 28) were collected from blood remaining after blood type testing, and plasma isolated within 4 hours. Samples were spun at 20°C and 1000 RPM for 10 minutes, the plasma pipetted into a fresh tube, and spun again to remove any remaining contaminating cells.

2.3 RNA isolation

Total RNA was isolated using the Trizol^® reagent extraction protocol. Total RNA was quantified spectrophotometrically and evaluated for RNA integrity by agarose gel electrophoresis. Final RNA concentrations of 5 ng/μL were used in reverse transcription (RT) reactions. RT was performed using the High Capacity cDNA reverse transcription kit protocol (Applied Biosystems [ABI]). RT products were amplified by quantitative polymerase chain reaction (PCR) according to the manufacturer’s instructions, using 4μL of RT product in a reaction volume of 20μL. Each sample was run in duplicate.

2.4 Oligonucleotides

Proprietary primers and probes for the human ɛ-globin gene were obtained from ABI and used for PCR analysis. The specificity of the sequence for the ɛ-globin gene was verified by a NCBI nucleotide BLAST after consultation with ABI’s sequence technologists; there was no overlap with gamma (γ))-globin genes, and ɛ-globin was the only gene detected by the primer used (Madhu Augustine, Sr. Product Manager for Gene Expression Assays, personal communication). Glyceraldehyde-3-phosphate dehydrogenase served as an internal control for quantity of starting total RNA in a duplex format (both sets of primers and probes were present in each well).

2.5 PCR

The TaqMan One-Step PCR Master Mix Reagents Kit (ABI) was used to perform PCR. The PCR assay was performed using the 7500 Fast Real-Time PCR System (ABI) with the following profile: 1 cycle at 95°C for 20 sec, then 40 cycles each at 95°c for 3 sec and 60°C for 30 sec. The threshold cycle (Ct) was calculated by the instrument’s software (7500 Fast System ver. 1.3.1) and adjusted to cross the most linear parts of the PCR amplification curves.

2.6 Statistical analysis

Data were analyzed using commercial integrated data analysis software (StatView^®). Differences in ɛ-globin gene expression among tissues and among gestational age groupings (10–14 weeks, 15–19 weeks, and 20–24 weeks gestation) were compared using unpaired t-tests, and by analysis of variance (ANOVA), using gestation as independent variable and gene expression as the dependent variable, for each tissue type. The Mann-Whitney-U test was performed for data that were not parametrically distributed. A p value of 0.05 or lower was considered significant.

3 Results

A total of 104 fetal samples were obtained between 10 to 24 weeks of gestation (fetal liver [n = 25], marrow [n = 28], blood [n = 37] and placenta [n = 14]). ɛ-globin gene expression was detected in all fetal tissue samples, but was negligible in the 11 non-pregnant adult blood samples evaluated (Fig. 1). Gene expression was greatest in fetal blood, followed by marrow and liver, and was minimally expressed in placenta. ɛ-globin gene expression was 24-fold greater in fetal cord blood than in fetal liver (p < 0.0001), and 10,000-fold greater when compared to placenta and non-pregnant adult samples (p < 0.0001).

ɛ-globin gene expression was generally greater at low and mid gestational ages (Figs. 2–5). In fetal blood, ɛ-globin gene expression was higher at 10–14 weeks and 15–19 weeks, and decreased at 20–24 weeks gestation (p = 0.05; Fig. 2). In fetal marrow, ɛ-globin gene expression was highest at 15–19 weeks gestation (p < 0.05, 15–19 weeks versus 10–14 weeks and 20–24 weeks; Fig. 3), while in fetal liver, ɛ-globin gene expression was similar throughout the gestations tested (Fig. 4). ɛ-globin gene expression in placenta was minimal (Fig. 5), and was significantly lower compared to the other fetal tissues tested across all gestational ages (p < 0.0001, placenta versus blood; p < 0.05, placenta versus marrow; p < 0.01, placenta versus liver; Fig. 1).

ɛ-globin gene expression was significantly greater in maternal plasma RNA compared to non-pregnant adult mononuclear cell RNA (p < 0.0001; Fig. 6). Gene expression ranged 1-30,821 in pregnant plasma samples and did not statistically correlate with gestation (data not shown). Gene expression in non-pregnant adult samples ranged 0 to 0.528.

4 Discussion

In order to determine if ɛ-globin gene expression might serve as a marker for the presence of fetal RNA, we compared gene expression in adult non-pregnant and pregnant samples with expression in fetal tissue where red cell progenitors reside: liver, marrow, and blood. We also compared gene expression of these tissues with placenta, as placental tissue is thought to represent a percentage of nucleic acid in the maternal circulation. ɛ-globin gene expression was detected in all fetal samples, but was negligible in the eleven non-pregnant adult samples evaluated, and minimal in placenta when compared to the other fetal tissues. ɛ-globin gene expression was significantly greater in plasma obtained from pregnant women than from non-pregnant adult samples. Non-pregnant adult samples reflected starting numbers of two to four cDNA copies based on average cycle threshold, using the equation: 2^(38 - Ct). In comparison, we estimated the starting copy numbers in fetal blood to be 16,000 to 20,000 cDNA copies. These 4,000 to 10,000-fold differences allow us to speculate that ɛ-globin gene expression in the maternal circulation would be fetal in origin.

Anatomic studies of human embryos reveal that hematopoiesis begins in the second to third embryonic weeks with formation of mesoderm-derived blood islands in the extra-embryonic mesoderm of the developing secondary yolk sac [11]. The expression of embryonic hemoglobin commences in the yolk sac during week two, and continues through six weeks gestation [12]. From six to 12 weeks of gestation the fetal liver and spleen produce gamma (γ)-globin chains(^Aγ and ^Gγ) [12]. This increased production of γ-globin is accompanied by significant decrease in ɛ-globin synthesis, and a transition from embryonic hemoglobin (Portland [ζ₂γ₂], Gower 1 [ζ₂ɛ₂] and Gower 2 [α₂ɛ₂]) to fetal hemoglobin (α₂γ₂). The fetal circulation contains increased numbers of actively replicating red cell progenitors (nucleated primitive erythrocytes) which could result in increased measures of ɛ-globin gene expression at the earliest gestational ages tested.

We found that ɛ-globin gene expression was approximately eight-fold greater in fetal blood compared to fetal marrow and 24-fold greater compared to fetal liver. This finding might be due to increased numbers of actively dividing red cell progenitors in the circulation compared to the liver, and the inclusion in liver samples of RNA from non-erythroid cells such as hepatocytes. Marrow samples showed increased ɛ-globin gene expression in the mid gestational group (15–19 weeks) with decreased expression at later gestational ages (20–24 weeks). Studies show that definitive blood cells and hematopoietic stem progenitor cells (HSPCs) can be detected in the fetal liver beginning at five to six weeks [13]. The fetal liver subsequently replaces the yolk sac as the main hematopoietic organ with the appearance of definitive enucleate, macrocytic erythrocytes [14, 15]. Fetal liver hematopoiesis leads ultimately to seeding of the fetal marrow and thymus with HSPCs [16]. Fetal liver ɛ-globin was consistently expressed throughout the gestations tested. The fetal circulation showed a 10,000-fold higher ɛ-globin gene expression than placenta, reflecting a relative lack of placental hematopoietic activity.

Since the discovery of free fetal DNA in the maternal circulation in 1997 by Lo et al. [17], many studies have focused on this non-invasive prenatal diagnostic tool. The main challenge is to identify small amounts of fetal DNA in the face of overwhelming background maternal DNA [18]. Cunningham et al. [19] used cell-free RNA in the maternal circulation, proposing that fewer false positives would result, and that reverse transcription was a more sensitive alternative. Measuring circulating cell-free RNA in the blood can identify differences in gene expression that go beyond determining total gene copy numbers [20]. Detection of mRNA derived from genes that are uniquely active in the fetus is an extremely promising avenue for research. Current studies are aimed at finding a reproducible and abundant fetal-specific marker that could be used to confirm the presence of fetal nucleic acid. Cell free fetal mRNA has also been detected in the maternal circulation, derived from genes expressed in the placenta [21]. Cell free fetal nucleic acids are detectable within the maternal circulation early in the first trimester, and increase with increasing gestation. Nucleic acids are rapidly cleared following delivery, with a half-life of 14 min [22]. This may be due to a significant release of DNAase and RNAase by the placenta into the maternal circulation.

Our study showed abundant ɛ-globin gene expression in the fetal tissues examined (especially fetal blood) compared to non-pregnant adult blood. Fetal cells are a likely source of fetal nucleic acid in the maternal circulation. We also demonstrated that ɛ-globin gene expression is much higher in maternal plasma than in non-pregnant adult blood. Despite the fact that ɛ-globin gene expression decreases as gestational age increases, the presence of measureable ɛ-globin expression may serve as proof that fetal nucleic acid is present in the maternal sample, when obtained between 12 and 24 weeks gestation. We speculate that ɛ-globin gene expression can serve as a positive marker for the presence of fetal nucleic acid in the maternal circulation.

Disclosure statements

The authors have no financial interests to disclose. This study was supported by grants from National Institutes of Health ULI TR000041, and the Signature Program in Child Health Research, University of New Mexico.

Footnotes

Acknowledgments

We wish to thank Dr. Curtis Boyd and the staff at Southwest Women’s Options for their ongoing support of this research.

References

Bianchi

. From prenatal genomic diagnosis to fetal personalized medicine: Progress and challenges. Nat Med 2012;18:1041–1051.

Clausen

, Christiansen

, Steffensen

, et al. Report of the first nationally implemented clinical routine screening for fetal RHD in D- pregnant women to ascertain the requirement for antenatal RhD prophylaxis. Transfusion 2012;52:752–758.

Bianchi

, Van Mieghem

, Shaffer

, et al. In case you missed it: The Prenatal diagnosis section editors bring you the most significant advances of Prenat Diagn 2014;34(1):1–5.

Scheffer

, van der Schoot

, Page-Christiaens

, et al. Noninvasive fetal blood group genotyping of rhesus D, c, E and of K in alloimmunised pregnant women: Evaluation of a 7-year clinical experience. BJOG 2011;118:1340–1348.

Chang

, Nelson

, Cao

, et al. Definitive-like erythroid cells derived from human embryonic stem cells coexpress high levels of embryonic and fetal globins with little or no adult globin. Blood 2006;108:1515–1523.

Cheung

, Goldberg

, Kan

. Prenatal diagnosis of sickle cell anaemia and thalassaemia by analysis of fetal cells in maternal blood. Nat Genet 1996;14:264–268.

Boye

, Hougaard

, Ebbesen

, Vuust

, Christiansen

. Novel feto-specific mRNA species suitable for identification of fetal cells from the maternal circulation. Prenat Diagn 2001;21:806–812.

Hogh

, Hviid

, Christensen

, et al. zeta-, epsilon-, and gamma-Globin mRNA in blood samples and CD71(+) cell fractions from fetuses and from pregnant and nonpregnant women, with special attention to identification of fetal erythroblasts. Clin Chem 2001;47:645–653.

Tjoa

, Cindrova-Davies

, Spasic-Boskovic

, Bianchi

, Burton

. Trophoblastic oxidative stress and the release of cell-free feto-placental DNA. Am J Pathol 2006;169:400–404.

10.

Alberry

, Maddocks

, Jones

, et al. Free fetal DNA in maternal plasma in anembryonic pregnancies: Confirmation that the origin is the trophoblast. Prenat Diagn 2007;27(5):415–418.

11.

Migliaccio

, Migliaccio

, Petti

, et al. Human embryonic hemopoiesis. Kinetics of progenitors and precursors underlying the yolk sac–liver transition. J Clin Invest 1986;78:51–60.

12.

Brittain

. Molecular aspects of embryonic hemoglobin function. Mol Aspects Med 2002;23:293–242.

13.

Hann

, Bodger

, Hoffbrand

. Development of pluripotent hematopoietic progenitor cells in the human fetus. Blood 1983;62:118–123.

14.

Thomas

, Yoffey

. Human foetal haematopoiesis. II. Hepatic haematopoiesis in the human foetus. Br J Haematol 1964;10:193–197.

15.

Rowley

, Ohlsson-Wilhelm

, Farley

. Erythroid colony formation from human fetal liver. Proc Natl Acad Sci USA 1978;75:984–988.

16.

Zambidis

, Peault

, Park

, Bunz

, Civin

. Hematopoietic differentiation of human embryonic stem cells progresses through sequential hematoendothelial, primitive, and definitive stages resembling human yolk sac development. Blood 2005;106:860–870.

17.

, Tein

, Pang

, et al. Quantitative analysis of fetal DNA in maternal plasma and serum: Implications for noninvasive prenatal diagnosis. Am J Hum Genet 1998;62:768–775.

18.

Bianchi

, Avent

, Costa

, van der Schoot

. Noninvasive prenatal diagnosis of fetal Rhesus D: Ready for prime(r) time. Obstet Gynecol 2005;106(4):841–844.

19.

Cunningham

, Yates

, Hamlington

, et al. Non-invasive RNA based determination of fetal Rhesus D type: A prospective study based on 96 pregnancies. Br J Obstet Gynaecol 1999;106:1023–1028.

20.

Koh

, Pan

, Gawad

, et al. Noninvasive in vivo monitoring of tissue-specific global gene expression in humans. Proc Natl Acad Sci U S A 2014;111(20):7361–7366.

21.

Poon

, Leung

, Lau

, Lo

. Presence of fetal RNA in maternal plasma. Clin Chem 2000;46:1832–1834.

22.

Chiu

, Lui

, Cheung

, et al. Time profile of appearance and disappearance of circulating placenta-derived mRNA in maternal plasma. Clin Chem 2006;52:313–316.