Temporal transcriptome analysis of the chicken embryo yolk sac
© Yadgary et al.; licensee BioMed Central Ltd. 2014
Received: 17 January 2014
Accepted: 12 August 2014
Published: 19 August 2014
The yolk sac (YS) is an extra-embryonic tissue that surrounds the yolk and absorbs, digests and transports nutrients during incubation of the avian embryo as well as during early term mammalian embryonic development. Understanding YS functions and development may enhance the efficient transfer of nutrients and optimize embryo development. To identify temporal large-scale patterns of gene expression and gain insights into processes and mechanisms in the YS, we performed a transcriptome study of the YS of chick embryos on embryonic days (E) E13, E15, E17, E19, and E21 (hatch).
3547 genes exhibited a significantly changed expression across days. Clustering and functional annotation of these genes as well as histological sectioning of the YS revealed that we monitored two cell types: the epithelial cells and the erythropoietic cells of the YS. We observed a significant up-regulation of epithelial genes involved in lipid transport and metabolism between E13 and E19. YS epithelial cells expressed a vast array of lipoprotein receptors and fatty acid transporters. Several lysosomal genes (CTSA, PSAP, NPC2) and apolipoproteins genes (apoA1, A2, B, C3) were among the highest expressed, reflecting the intensive digestion and re-synthesis of lipoproteins in YS epithelial cells. Genes associated with cytoskeletal structure were down-regulated between E17 and E21 supporting histological evidence of a degradation of YS epithelial cells towards hatch.
Expression patterns of hemoglobin synthesis genes indicated a high erythropoietic capacity of the YS between E13 and E15, which decreased towards hatch. YS histological sections confirmed these results. We also observed that YS epithelial cells expressed high levels of genes coding for plasma carrier proteins (ALB, AFP, LTF, TTR), normally produced by the liver.
Here we expand current knowledge on developmental, nutritional and molecular processes in the YS. We demonstrate that in the final week of chick embryonic development, the YS plays different roles to support or replace the functions of several organs that have not yet reached their full functional capacity. The YS has a similar functional role as the intestine in digestion and transport of nutrients, the liver in producing plasma carrier proteins and coagulation factors, and the bone marrow in synthesis of blood cells.
KeywordsYolk sac Gene expression Transcriptome analysis Chicken embryo Epithelial cells Nutrient transport Lipoproteins Cytoskeleton Erythropoiesis
During embryonic development, nutrients are transferred to the embryo by specialized tissues. In mammals, a continuous transport of nutrients takes place from the mother to the embryo through the placenta, however in early stages of gastrulation the mammalian embryo is also dependent upon the transport of nutrients from the yolk through its surrounding tissue, the yolk sac (YS) . Different from mammals, the YS of the avian embryo supports its development during the entire phase of embryogenesis . The chick embryo is separated from the mother and therefore must derive all of its nutrients only from the contents of the egg. This closed environment provides an excellent model to study mechanisms by which nutrients are utilized through the YS.
In the first week of chick embryonic development, the YS begins to form from the developing gut of the embryo. A layer of ectoderm starts to spread over the surface of the yolk, ultimately covering it entirely. Next, a rapidly proliferating population of endodermal cells spreads and forms an epithelial layer between the yolk surface and the ectoderm. Finally, cells of mesodermal origin migrate between the endodermal and ectodermal layers and form a mesodermal layer, which differentiates into blood vessels, blood cells and connective tissue [3–5]. Up to E11, very small amounts of yolk nutrients are transferred to the embryo by the YS. From E11, after the YS has totally surrounded the yolk, the epithelial cells of the YS function as the mediators for the transport of nutrients from the contents of the yolk to the blood circulation of the developing embryo .
The transition from early embryogenesis to the second phase of incubation is characterized by prompt utilization of nutrients through the YS and a rapid growth of the embryo . The nutrients are absorbed into YS epithelial cells from the yolk content by receptor-mediated endocytosis of lipoproteins [5, 8] and by transporters for peptides, amino acids, glucose, and minerals [9, 10]. Endocytosis of lipoproteins into the YS epithelial cells has been confirmed by electron microscope observations of the YS , however out of the vast variety of receptors that mediate lipid entrance to endodermal cells, only two receptors for lipoproteins [5, 8] and no receptors for free fatty acids have been examined so far in the YS.
Inside YS epithelial cells, there is evidence of intensive lysosomal digestion of lipoproteins such as very low density lipoproteins (VLDL) into free fatty acids and glycerol, which are then reesterified to form triglycerides and phospholipids, and assembled into new lipoprotein particles (low density lipoproteins, LDL and high density lipoproteins, HDL) [5, 12]. The subsequent lipoproteins are exocytosed to the blood and transferred to the embryonic tissues [12–14]. Although the idea that a substantial lipid remodeling takes place within YS epithelial cells is supported by several studies, only a fraction of the genes involved in these processes have been examined in the YS. Furthermore, despite recent data on the capacity of the YS to express genes related to protein and carbohydrates synthesis, storage, digestion and transport [10, 15, 16], the developmental profiles of cellular and molecular mechanisms responsible for a larger part of the above mentioned processes have not been investigated in the YS.
The utilization of nutrients by the chick embryo, at any stage during development, has a direct effect on growth and hatchability. A more complete and large-scale picture of the processes and mechanisms that control yolk utilization through the YS is needed in order to better understand how avian and early term mammalian embryos utilize yolk nutrients. Therefore, the aim of the current study was to generate and analyze high-throughput gene expression data of the YS during the last week of incubation. Thus, we performed a transcriptome analysis of RNA from YS tissues, as well as histological and biochemical analyses, on days 13, 15, 17, 19, and 21 (day of hatch) of incubation. Our time-series experimental design, with 3 biological replicates per day, allowed us to quantify differential gene expression between days, identify temporal patterns of biological processes, and gain insights into the processes and mechanisms of nutrient synthesis, absorption, digestion, and secretion within the YS.
Eggs and yolk sac sampling
Fertile eggs were obtained from a commercial broiler breeder farm (Brown, Hod Hasharon, Israel). Eggs were incubated in a Petersime hatchery at the Faculty of Agriculture of the Hebrew University under standard conditions (37.5°C, 60% relative humidity, with automatic egg turning). Six eggs, representing the weight distribution of the eggs at set, were selected at each of the following days: E13, E15, E17, E19, and E21 (day of hatch; no later than 30 min after hatch). Embryos were killed by cervical dislocation. The experiment was approved by the Ethics Committee for Animal Experimentation, Faculty of Agricultural, Food and Environmental Sciences, The Hebrew University of Jerusalem. The YS was separated from the yolk and weighed, and a small piece from the area vasculosa, where endodermal epithelial cells and villus-like folds develop and which is the vascularized and most active region in endocytosis of yolk, was rinsed in a 0.9% autoclaved saline solution, and placed in microcentrifuge tubes at -80°C for mRNA analysis. Fractions of YS were also taken for histological sectioning.
Six additional eggs were sampled on each examined day. Embryos were weighed, the yolk + YS were weighed, and the yolk + YS were homogenized and stored at -20°C for total fat content and fatty acid analyses. In the current study we present quantities of lipids in the yolk + YS because according to Yadgary et al. , estimating utilization of nutrients by the embryo is more accurate if calculated from the yolk + YS, and not just from the yolk.
Total RNA isolation and pooling of samples
Libraries for next generation sequencing (NGS) were prepared from total YS RNA. Total RNA was isolated from 30 YS (six replicates per day) using TRI-Reagent RNA/DNA/Protein Isolation Reagent 5 according to the manufacturer’s protocol (Sigma-Aldrich, St. Louis, MO). 50 to 100 mg of tissue was homogenized with 1 mL of TRI-Reagent and phase separated using chloroform. RNA was precipitated using isopropanol, washed with 75% ethanol, and solubilized with nuclease-free water. RNA concentration was determined spectrophotometrically in a Nanodrop ND-1000 (NanoDrop, Wilmington, DE). Equal amounts of YS RNA were pooled from two replicate embryos so that a total of 15 libraries were prepared for NGS- three biological replicates per each examined day. Prior to sequencing, all samples were treated with DNase and quality assessed on a Bioanalyzer 2100 (Agilent, Santa Clara, CA).
High-throughput Serial Analysis of Gene Expression (SAGE) was performed using SOLiD4 technology (Applied Biosystems) at The Center for Genomic Technologies at the Hebrew University of Jerusalem. RNAs were reverse transcribed and SAGE libraries were prepared using a SOLiD SAGE Kit with Barcoding Adaptor Module (Applied Biosystems) which allows multiple samples to be processed in parallel. The system generates a library of 27-bp “tags” for all the transcripts in a cell and provides a method for analyzing the genome-wide expression levels of poly(A) transcripts. Briefly, mRNAs were attached to poly(T) coupled magnetic beads, cDNAs were synthesized using SuperScriptR III Reverse Transcriptase and E. coli DNA polymerase, digested with a 4-bp cutter enzyme (NlaIII), and ligated to a barcode adaptor sequence (unique for each library) containing a recognition site for a restriction enzyme (EcoP15I) that cuts DNA 27 bp away. After restriction and separation from beads, the tags were ligated to a second adaptor. The 27-bp transcripts, surrounded by two adaptors, were pooled into a single multiplexed sequencing reaction. Applied Biosystems SOLiD sequencer generated color-space fasta files (.csfasta) and the corresponding quality control files (.qual) for 14 out of the 15 libraries. Due to a technical error, one sample from E19 was not sequenced. All csfasta and quality data sets supporting the results of this article are available in the SRA repository (http://www.ncbi.nlm.nih.gov/Traces/sra/sra.cgi?study=SRP045315) with accession number SRP045315.
A total of 179,000,718 sequencing reads were obtained (Additional file 1). All the reads (50 bp long) were subjected to adaptor sequence trimming and quality filtering using the FastX-toolkit suite and FASTQ manipulation tool  available through the Web-based bioinformatics platform Galaxy (http://usegalaxy.org/). Based on the FastX quality statistics, all the reads were trimmed to 23-bp and low phred scored reads were discarded. One sample from E21 showed consistently low phred scores and therefore was excluded from further analyses. The remaining reads obtained from each library (on average ~80%) were aligned against the chicken transcriptome RefSeq (17,696 mRNA Ref-Seq, downloaded from NCBI database). Because the SAGE protocol is designed to produce reads that represent mRNA regions adjecent to the 3'-most NlaIII endonuclease cleavage site, we mapped reads only to 15,169 Ref-Seq mRNA sequences that were computationally found to be longer than 22 bases after restriction enzyme cleavage and were non-redundant in the 23-bp mRNA region, to which the reads were predicted to be aligned. Sequence mapping was done using bowtie software  in a Galaxy setting, with maximum number of mismatches permitted [n] = 2, maximum permitted total of quality values at mismatched read positions (-e) = 70, and maximum alignments authorized [m] = 10. Reported alignments were chosen to be best in criteria [n] and criteria (-e) (--best). On average, ~55% of the sequences of all libraries were mapped. Of the mapped reads, on average 93% were unique, i.e. mapped to only one Ref-Seq mRNA sequence.
The number of reads that were mapped to each gene was counted and count data was then used as input for JMP GENOMICS 6.0 (SAS, Cary, NC) software analyses. Read counts of the libraries were normalized using the Upper-Quartile Normalization method , and were then loge (ln) transformed, yielding transformed normalized counts. Principal component analysis (PCA) was performed to identify outliers within each day (Additional file 2). One outlier was detected on E17 and discarded, leaving 12 samples (three replicates on E13 and E15, and two replicates on E17, E19, and E21) for further analyses. Outlier detection was confirmed by manual inspection of data from 200 random genes (Additional file 2). A generalized linear model (GLM) based on the negative binomial distribution for count data was performed to test for differential expression between embryonic days (JMP GENOMICS 6.0, SAS, Cary, NC). Genes that were found to have a low normalized count number across incubation (average <8) were not included in the GLM. Thus, a total of 7128 genes were tested at significance level of α = 0.05 adjusted to match a 5% false discovery rate . It should be noted that even though throughout the manuscript we use the term "expressed" for genes with ≥8 normalized mean count across samples, it would be incorrect to claim that the filtered genes are not expressed in the YS; It is possible that a higher coverage (sequencing depth) would detect genes with very low copy number that were not detected in our experiment.
For each one of the 7128 genes, mean count number (transformed normalized) was calculated for each embryonic day. To examine temporal changes during incubation, per each day the mean count number of each gene was standardized to the standard deviation of all the values of the genes. Significantly changed genes (3547) were then subjected to hierarchical clustering (JMP GENOMICS 6.0, SAS, Cary, NC). Genes clustered together were annotated using the web-based Database for Annotation, Visualization, and Integrated Discovery (DAVID) .
Total Lipid Analysis. Total lipids were extracted using a modified method of Folch et al. , as previously described by Yadgary et al., . Briefly: 10 mL of chloroform-methanol (2:1 vol/vol) was added to 0.5 mL of YS homogenates. After 30 min, 2 mL of distilled water was added. Following overnight incubation at 4°C, 5 mL of the bottom layer was transferred to preweighed, glass tubes and oven-dried at 105°C. The tubes with the remaining fat were weighed and the weight of fat per gram of YS + yolk and per total fat in the YS + yolk was calculated.
Fatty acid analysis. Fatty-acid concentration was determined in lyophilized samples of 100 μg YS + yolk by gas chromatography on a DEGS column. Absolute amounts of lipids were determined after addition of an internal standard of heptadecanoic acid, and the proportion of specific fatty acid was calculated out of the total fatty-acid concentration. The concentrations of saturated fatty acids and unsaturated fatty acids were determined.
Fresh sections of YS were fixed overnight in 4% (v/v) buffered formaldehyde. Serial 4-μm sections were prepared after the samples had been dehydrated in graded ethanol solutions, cleared in chloroform and embedded in paraffin. Sections were deparaffinized in xylene, rehydrated, stained with hematoxylin and eosin or with proliferating cell nuclear antigen (PCNA) immunostaining and evaluated by light microscopy.
Results and discussion
Mapping and statistics
Serial Analysis of Gene Expression (SAGE) methodology followed by NGS was performed on 2–3 biological replicates of mRNA extracted from YS on E13, E15, E17, E19 and E21. Using what is also known as SAGE-Seq, a single region of each mRNA molecule was sequenced, and thus the number of reads mapped to a specific mRNA reference sequence was directly related to expression levels of that gene. Reads were mapped using 15,169 chicken mRNA Ref-Seq, however only 7,128 genes were further analyzed after excluding genes with very low count number. For each of the 7,128 genes, after normalization and logarithmic transformation of the data, we identified differential expression across days, and calculated mean count number per embryonic day (i.e. ln normalized expression).
Clustering and functional annotation
According to functional annotation, genes of cluster A, which exhibited a common pattern of increase in expression from E13 to E15 (Figure 1), were involved in several metabolic processes and pathways, most of them associated with lipid metabolism and protein transport. Partitioning of cluster A genes revealed that genes that were up-regulated from E13 to E19 and then down regulated until E21 were found as related to ion transport and extracellular matrix organization (cluster A1.1, Figure 2), genes that were up-regulated from E13 to E21 were associated with lipid and protein transport (cluster A2.2, Figure 2; cluster A2.1, Additional file 3), and genes that were up-regulated between E13 and E17 and then down-regulated until E21, were associated with proteolysis (Additional file 3) and structural actin cytoskeleton cell activity (cluster A3.1, Figure 2).
Genes of cluster B, which exhibited a common pattern of increase in expression from E17 to E19 (Figure 1), were enriched in GO terms and pathways that included, among others, protein catabolic process, negative regulation of cellular biosynthetic process, vitamin binding, and steroid metabolic process. Subdivision of these genes revealed that genes with no change in expression between E13 and E15 and then an up-regulation from E15 were involved in vitamin binding and in golgi activity such as sorting and packaging of proteins for transport and modification of oligosaccharides of glycoproteins (cluster B1.2, Figure 2; cluster B1, Additional file 3); genes that were down-regulated between E13 to E15 and then up-regulated from E17 to E21, were associated with protein transport to or within the endoplasmic reticulum (cluster B2.2, Figure 2; clusters B2.1, B2.2, B2.3, Additional file 3).
Genes of cluster C had a distinctly different temporal pattern of change compared to clusters A and B, as was observed in Figure 1B heat map and dendrogram. Cluster C1 genes exhibited a down-regulation between E13 and E17 in ribosomal biosynthesis processes, while cluster C2 genes were more associated with proteasome mediated protein breakdown and chromatin organization (Figure 2; Additional file 3). Furthermore, cluster C3 genes that were found to be involved in cell replication events and segregation of genetic material, exhibited a striking down-regulation between E15 and E21 (Figure 2, Additional file 3). Thus, expression of most of the genes in cluster C was indicative of a substantial reduction in YS cell proliferation.
Synthesis of blood cells in yolk sac blood islands
Hemoglobin is made up of globin chains and the porphyrin ring, or heme. The chicken globin genes include the α-globin gene cluster (π, αA, and αD ) and the β-globin gene cluster (ρ, βH, βA, and ϵ). Primitive erythrocytes in chickens express the π, ρ and ϵ globin chains, while definitive erythrocytes express αA, αD, βH, and βA [23, 24]. In the current study, we found that all seven globin genes were expressed in the YS, with a similar temporal pattern of down-regulation between E15 and E19 (Figure 3A). Significantly higher expression levels were observed for the definitive globins as compared to the primitive globins. Surprisingly, on E13 and E15 the definitive globins also exhibited the highest expression levels compared to all other chicken mRNA examined in the current study (Additional file 4).
The YS is considered as the sole niche for primitive erythropoiesis from E1 to E5, and as the main niche for early definitive erythropoiesis from E5 to E10 of chicken development . Still, it is not clear how long the YS continues its erythropoietic process. Our results of the expression pattern of globin genes (Figure 3A), together with the similar pattern in heme synthesis gene expression (Figure 3B), indicated a very high erythropoietic capacity of the YS on E13 and E15, which decreases towards hatch. Our histological sectioning of the YS confirmed these results; we observed that blood islands, which produce blood cells in the YS, are abundant in the YS on E13 and E15, and then dramatically decrease towards hatch (Figure 4A). Moreover, using PCNA staining we observed that the cells in the blood islands were highly proliferative (Figure 4B). This explains the similar temporal pattern of hemoglobin genes and mitosis genes. Highly proliferative blood island cells produce vast amounts of hemoglobin on E13 and E15 and then lose contact from each other and from the endothelium, and then become free circulating cells. Thus expression of mitosis and hemoglobin genes substantially decrease towards hatch.
A transition is known to occur during late chick embryonic development, when the bone marrow becomes the major erythropoietic organ. There is however a controversy regarding the contribution of the YS to definitive erythropoiesis in the last week of chick embryonic development. While Nagai and Sheng  speculated that the YS continues to generate definitive erythrocytes at least until E15, Lassila et al.  suggested that from E11 erythropoiesis in the YS contributes less than 20% to peripheral blood cells. Our current study demonstrates that the YS is an erythropoietic tissue not only in the first week of incubation but also in the last week of incubation, with a massive migration of definitive red blood cells towards hatch. We therefore postulate that between E13 and E17 the YS contributes a greater part than assumed to the synthesis of red blood cells of embryos.
Our results represent the first quantitative look at YS erythropoiesis, and are in agreement with Niimi et al. and Bruns and Ingram  who suggested that YS erythropoiesis lasts throughout incubation. However, in chick/quail chimaera studies, in which a quail embryo is grafted onto the YS of a chick very early in development, the majority of circulating blood cells in the last week of incubation were of the embryonic quail origin [29, 30]. The discrepancy between our results and those of chick/quail chimaeras studies can be resolved by considering similar experiments that found that from E7 the YS was extensively seeded by quail embryonic hematopoietic stem cells [31, 32]. Therefore during the last week of incubation, most of the proliferating cells in YS blood islands are of embryonic origin, making it impossible to differentiate between late-term chick YS and quail embryo blood cells in the chimaera studies.
Epithelial cell gene expression analyses and annotation
Histology sectioning and annotational clustering revealed that in our temporal gene expression analyses we monitored at least two cell types- the endodermal epithelial cells and the erythropoietic cells of the YS. The YS also contains fibroblasts, endothelial cells, and vascular smooth muscle cells, and it is possible that at a low level we also sequenced mRNA from these cell types. However, we estimate that most of the sampled non-erythropoietic mRNA is from epithelial cells. Therefore, in the following sections, non-erythropoietic cells are referred to as YS epithelial cells.
Functional annotation of significantly down and up-regulated genes (Separate analysis for 3500 YS epithelial genes)
Down regulated genes
Up regulated genes
1.56: endoplasmic reticulum; 1.54 plasma membrane organization; 1.44: ubiquitin ligase complex; 1.33: purine nucleotide binding;
2.59: macromolecular complex subunit organization; 2.16: cholesterol metabolic process; 1.65: chaperone mediated protein folding requiring cofactor; 1.51: cholesterol transport; 1.3: vacuole;
2.66: cytoskeletal part; 1.84: Intermediate filament protein; 1.74: actomyosin; 1.50: protein heterooligomerization; 1.25: lysosome
1.93: fatty acid metabolic process; 1.53: topological domain:Lumenal; 1.21: vitamin binding
2.53: macromolecule catabolic process; 2.46: GTP binding; 1.61: nucleotide binding; 1.56: zinc ion binding; 1.27: establishment of protein localization
2.24: sterol metabolic process; 1.76: protein dimerization activity; 1.74: extracellular region; 1.59: response to wounding; 1.43: lipid oxidation; 1.34: transmembrane receptor protein serine/threonine kinase signaling pathway; 1.28: growth factor binding; 1.22: Collagen triple helix repeat; 1.20: Fibrinogen
1.99: oxidation reduction; 1.59: keratin; 1.55: cholesterol metabolic process; 1.47: cofactor binding;
1.34: ossification; 1.25 membrane coat; 1.20: transmembrane region
Nutrient transport through yolk sac epithelium
The absorption of yolk lipids into YS epithelium is carried out by endocytosis of lipoproteins . We found that a vast array of lipoprotein receptors mediate lipoprotein entrance to YS epithelial cells (Figure 5C). Among these receptors is VLDLR, which has a binding domain that recognizes VLDL and vitellogenin. VLDLR has been previously suggested to have a major role in YS lipoprotein uptake , however we show that VLDLR had relatively low levels of expression in the YS (Figure 5C). On the other hand LRP2, a receptor for a range of lipoproteins , had the highest expression level among all lipoprotein receptors. The high LRP2 gene expression level increased from E13 to E17 and decreased from E17 to E21, supporting previous results of a study by Bauer et al. . Expression levels of several lipoprotein receptors in YS epithelium (LRP1, 5, 6, 12) may imply that LRP2 does not have an exclusive role in YS lipoprotein uptake, however the putative genes LRP5, LRP6, and LRP12 have not been shown to play a key role in lipoprotein clearance. Further research into gene expression levels as well as protein levels of lipoprotein receptors in the YS is needed to validate our results and to elucidate the contribution of each receptor to fat uptake through the YS.
Except for endocytosis of lipoproteins, other possible mechanisms of lipid uptake through YS epithelium, such as transporters for free fatty acids, have not been investigated. Previous studies have speculated that long chain poly-unsaturated fatty acids (22:6, 20:4) are differentially utilized from the yolk . In the current study, we observed that between E15 and E19, docosahexaenoic (22:6) fatty acid, which is crucial for normal brain development, was more rapidly utilized from the yolk as compared to all other examined fatty acids (Figure 5B). To elucidate the differential utilization of yolk fatty acids, we examined the expression of genes responsible for absorption and transport of free fatty acids. Different membrane fatty acid transporters were expressed in YS epithelium (Figure 5D), as well as cytosolic fatty acid binding proteins that also play a role in uptake of long chain poly-unsaturated fatty acids (Figure 5E). Among them was FABP5 which exhibited a substantial up-regulation between E15 and E21. Further research is needed to confirm the role of these transporters in the differential utilization of yolk fatty acids.
Inside YS epithelial cells, lipoproteins are transferred to lysosomes for digestion [8, 13, 14]. Yolk fat utilization is affected by the digestive capacity of these lysosomes. In the current study, several lysosomal digestion related genes were found to be among the 50 most highly expressed genes in YS epithelial cells (Additional file 7); Prosaposin, which facilitates the catabolism of glycosphingolipids in lysosomes, was the highest expressed gene from E15 to E21 (Figure 5F). We also found a high and constant expression of lipase A, which catalyzes the hydrolysis of cholesteryl esters and triglycerides in lysosomes (Figure 5F). In addition, many lysosomal proteases such as cathepsin A and B were highly expressed in the YS (Figure 5F). These results lend support to the hypothesis that yolk lipoproteins are hydrolyzed in the lysosomes of YS epithelial cells into free fatty acids, partial glycerides and glycerol.
Lipids are transported from the epithelial cells of the YS to the embryo as newly formed lipoproteins [12–14]. An integral part of these newly synthesized lipoproteins are apolipoproteins that serve as structural proteins and cell surface receptors. In the current work, apolipoprotein genes (apoA1, A4, B, C3, and the putative apo A-II-like) were among the most highly abundant in the YS (Additional file 7). We observed that expression of apoB and apoC3 was constant, whereas apoA4 exhibited a significant up-regulation between E13 and E17, while apoA1 was up-regulated between E17 and E19 (Figure 5G). Our results reflect the intensive re-synthesis of lipoproteins in the YS and indicate that both HDL and LDL are produced in the YS and are then secreted to the embryonic circulation as previously suggested by Kanai et al. .
In summary, absorption, digestion and transport of lipoproteins are perhaps the main mechanisms by which yolk nutrients are utilized through YS epithelial cells. However, previous studies [10, 15], which examined YS expression of a small number of solute carrier (SLC) genes, suggested that they may play an important part in the uptake of peptides, amino acids, and carbohydrates in YS cells. Although in the current study we examined such mechanisms, they will not be further discussed. Nevertheless, we documented for the first time temporal gene expression patterns of all solute carrier genes (SLC genes) that were expressed in YS epithelium (Additional file 9).
Plasma protein synthesis in the yolk sac
The degradation of the yolk sac towards the end of incubation
Hemoglobin catabolism and bile biosynthesis in YS epithelium
During the last week of chick embryonic development, the color of the residual yolk changes from yellow to green . Speake and Teale  suggested that the change of color is due to a transfer of gallbladder bile from the embryonic intestine to the yolk through the yolk stalk, while Yadgary et al.  suggested that the YS produces bile components responsible for the green color of the yolk. In the current study, we observed that on E17 the residual yolk appeared green in color whereas the intestine was almost empty of contents (Additional file 10), thus supporting the latter hypothesis.
The green color of bile is primarily due to the breakdown of heme residues into biliverdin, a process that normally occurs in the liver. We speculate that a similar breakdown occurs in the epithelial cells of the YS. Evidence for that includes: 1) the increasing expression levels in YS epithelium of hemopexin, the heme-binding plasma glycoprotein (Figure 6C), 2) the up-regulation in YS epithelium of heme oxygenase, which breaks down heme into biliverdin (Additional file 10), and 3) the high and constant levels of expression for ferritin, an intracellular protein that stores iron in the liver (Additional file 10). In addition, enzymes of bile synthesis alternative pathway were expressed in the YS (Additional file 8), in agreement with Yadgary et al.  that detected bile acids in the yolk and YS and suggested that they are synthesized in the YS tissue. Furthermore, cysteine dioxygenase, an enzyme involved in the formation of taurine which conjugates bile acids, was the highest up-regulated gene among all examined genes in the current study- almost 700-fold between E15 and E21 (Additional file 8). Taking all that together, we suggest that red blood cells, either primitive or definitive, stemming from embryo blood circulation or from YS blood islands, are intravascularly degraded and their heme is transported to the epithelium of the YS by hemopexin. Heme is catabolized in these cells by heme oxygenase into biliverdin and iron, the iron is stored by ferritin in the YS or transported to the blood by YS derived lactotransferrin, and biliverdin is joined with conjugated bile acids that were synthesized in the YS.
It should be noted that in mammals, biliverdin is normally catabolized into bilirubin by the enzyme biliverdin reductase, however in the mature chicken bilirubin is hardly detected . Here we observed that during chick embryonic development, the enzyme biliverdin reductase had moderate levels of expression on E13 which were then substantially down-regulated until hatch (Additional file 10), suggesting that in chicken embryonic development bilirubin may serve as a fetal gene with a loss of function towards hatch.
Two main types of cells– erythropoietic and epithelial cells, were sampled.
Definitive erythropoietic cells, representing the second wave of erythropoiesis in the YS, proliferate in the YS at a very high level between E13 and E15.
A massive amount of YS definitive erythropoietic cells migrate to the blood circulation from E15 to E19.
The definitive erythropoietic cells of the YS are presumably of an embryonic origin [31, 32]. They could not have been distinguished in traditional chick/quail experiments, and therefore the YS was underestimated in its contribution to blood cell synthesis in the last week of incubation.
YS epithelial cells are actively involved in nutrient absorption and digestion from E13 to E21, which is carried out by digestive enzymes and transporters of fatty acids, amino acids, peptides, and carbohydrates. YS epithelial cells exhibit high levels of lipoprotein absorption, digestion and re-synthesis.
YS epithelial cells synthesize high levels of carrier proteins and coagulation factors from E13 to E21.
YS epithelial cells are actively involved in heme catabolism and bile synthesis at increasing levels from E15 to E21.
Towards the end of incubation, a fraction of the YS epithelial cells start degrading.
Our results support and expand current knowledge on the different functions of this unique extraembryonic tissue. The YS has a similar role as the liver in the synthesis of plasma proteins, as the bone marrow in the synthesis of new blood cells, and as the intestine in digestion of nutrients and their transport to the embryo. Thus, the YS plays different roles to support or replace the functions of several organs that have not yet reached their full functional capacity.
The current study contributes to our understanding of crucial metabolic processes during the last week of chick embryonic development as well as to early stages of mammalian embryonic development. In view of its characteristics, we suggest that the YS of the chick embryo may be used as a model to understand a wide range of processes such as fatty acid absorption and intracellular transport, nutrient transport, lysosomal digestion, apoptosis of epithelial cells, fetal gene loss of function, and blood synthesis and degradation. Furthermore, future studies that will compare high-throughput analyses between YS and different organs of the embryo, may help elucidate all of the various functional roles of the YS, which in light of the current data might be more accurately defined as an extra-embryonic organ.
This research was supported by Research Grant No. US-4113-08 from the United States-Israel Binational Agricultural Research and Development Fund (BARD).
- Sheng G, Foley AC: Diversification and conservation of the extraembryonic tissues in mediating nutrient uptake during amniote development. Ann N Y Acad Sci. 2012, 1271: 97-103. 10.1111/j.1749-6632.2012.06726.x.PubMed CentralPubMedView ArticleGoogle Scholar
- Speake BK, Murray AM, Noble RC: Transport and transformations of yolk lipids during development of the avian embryo. Prog Lipid Res. 1998, 37: 1-32. 10.1016/S0163-7827(97)00012-X.PubMedView ArticleGoogle Scholar
- Mobbs IG, McMillan DB: Transport across endodermal cells of the chick yolk sac during early stages of development. Am J Anat. 1981, 160: 285-308. 10.1002/aja.1001600307.PubMedView ArticleGoogle Scholar
- Moran ET: Nutrition of the developing embryo and hatchling. Poult Sci. 2007, 86: 1043-1049. 10.1093/ps/86.5.1043.PubMedView ArticleGoogle Scholar
- Bauer R, Plieschnig JA, Finkes T, Riegler B, Hermann M, Schneider WJ: The developing chicken yolk sac acquires nutrient transport competence by an orchestrated differentiation process of its endodermal epithelial cells. J Biol Chem. 2013, 288: 1088-1098. 10.1074/jbc.M112.393090.PubMed CentralPubMedView ArticleGoogle Scholar
- Noble RC, Cocchi M: Lipid metabolism and the neonatal chicken. Prog Lipid Res. 1990, 29: 107-140. 10.1016/0163-7827(90)90014-C.PubMedView ArticleGoogle Scholar
- Yadgary L, Cahaner A, Kedar O, Uni Z: Yolk sac nutrient composition and fat uptake in late-term embryos in eggs from young and old broiler breeder hens. Poult Sci. 2010, 89: 2441-2452. 10.3382/ps.2010-00681.PubMedView ArticleGoogle Scholar
- Hermann M, Mahon MG, Lindstedt KA, Nimpf J, Schneider WJ: Lipoprotein receptors in extraembryonic tissues of the chicken. J Biol Chem. 2000, 275: 16837-16844. 10.1074/jbc.M000163200.PubMedView ArticleGoogle Scholar
- Yadgary L, Kedar O, Adepeju O, Uni Z: Changes in yolk sac membrane absorptive area and fat digestion during chick embryonic development. Poult Sci. 2013, 92: 1634-1640. 10.3382/ps.2012-02886.PubMedView ArticleGoogle Scholar
- Speier JS, Yadgary L, Uni Z, Wong EA: Gene expression of nutrient transporters and digestive enzymes in the yolk sac membrane and small intestine of the developing embryonic chick. Poult Sci. 2012, 91: 1941-1949. 10.3382/ps.2011-02092.PubMedView ArticleGoogle Scholar
- Lambson RO: An electron microscopic study of the entodermal cells of the yolk sac of the chick during incubation and after hatching. Am J Anat. 1970, 129: 1-19. 10.1002/aja.1001290102.PubMedView ArticleGoogle Scholar
- Kanai M, Soji T, Sugawara E, Watari N, Oguchi H, Matsubara M, Herbert DC: Participation of endodermal epithelial cells on the synthesis of plasma LDL and HDL in the chick yolk sac. Microsc Res Tech. 1996, 35: 340-348. 10.1002/(SICI)1097-0029(19961101)35:4<340::AID-JEMT5>3.0.CO;2-Q.PubMedView ArticleGoogle Scholar
- Murray AM, Denis R, Speake BK: Acyltransferase activities in the yolk sac membrane of the chick embryo. Lipids. 1999, 34: 929-935. 10.1007/s11745-999-0442-0.PubMedView ArticleGoogle Scholar
- Powell KA, Deans EA, Speake BK: Fatty acid esterification in the yolk sac membrane of the avian embryo. J Comp Physiol B. 2004, 174: 163-168. 10.1007/s00360-003-0401-5.PubMedView ArticleGoogle Scholar
- Yadgary L, Yair R, Uni Z: The chick embryo yolk sac membrane expresses nutrient transporter and digestive enzyme genes. Poult Sci. 2011, 90: 410-416. 10.3382/ps.2010-01075.PubMedView ArticleGoogle Scholar
- Yadgary L, Uni Z: Yolk sac carbohydrate levels and gene expression of key gluconeogenic and glycogenic enzymes during chick embryonic development. Poult Sci. 2012, 91: 444-453. 10.3382/ps.2011-01669.PubMedView ArticleGoogle Scholar
- Blankenberg D, Von Kuster G, Coraor N, Ananda G, Lazarus R, Mangan M, Nekrutenko A, Taylor J: Galaxy: a web-based genome analysis tool for experimentalists. Curr Protoc Mol Biol Ed by Frederick M Ausubel al. 2010, Chapter 19 (19): 10-1–21Google Scholar
- Langmead B, Trapnell C, Pop M, Salzberg SL: Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 2009, 10: R25-10.1186/gb-2009-10-3-r25.PubMed CentralPubMedView ArticleGoogle Scholar
- Bullard JH, Purdom E, Hansen KD, Dudoit S: Evaluation of statistical methods for normalization and differential expression in mRNA-Seq experiments. BMC Bioinformatics. 2010, 11: 94-10.1186/1471-2105-11-94.PubMed CentralPubMedView ArticleGoogle Scholar
- Benjamini Y, Hochberg Y: Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Ser B Methodol. 1995, 57: 289-300.Google Scholar
- Huang DW, Sherman BT, Lempicki RA: Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009, 4: 44-57.View ArticleGoogle Scholar
- Folch J, Lees M, Sloane Stanley GH: A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem. 1957, 226: 497-509.PubMedGoogle Scholar
- Bruns GA, Ingram VM: The erythroid cells and haemoglobins of the chick embryo. Philos Trans R Soc Lond B Biol Sci. 1973, 266: 225-305. 10.1098/rstb.1973.0050.PubMedView ArticleGoogle Scholar
- Alev C, McIntyre BA S, Nagai H, Shin M, Shinmyozu K, Jakt LM, Sheng G: BetaA, the major beta globin in definitive red blood cells, is present from the onset of primitive erythropoiesis in chicken. Dev Dyn. 2008, 237: 1193-1197. 10.1002/dvdy.21510.PubMedView ArticleGoogle Scholar
- Sheng G: Primitive and definitive erythropoiesis in the yolk sac: a bird’s eye view. Int J Dev Biol. 2010, 54: 1033-1043. 10.1387/ijdb.103105gs.PubMedView ArticleGoogle Scholar
- Nagai H, Sheng G: Definitive erythropoiesis in chicken yolk sac. Dev Dyn. 2008, 237: 3332-3341. 10.1002/dvdy.21746.PubMedView ArticleGoogle Scholar
- Lassila O, Martin C, Toivanen P, Dieterlen-Lièvre F: Erythropoiesis and lymphopoiesis in the chick yolk-sac-embryo chimeras: contribution of yolk sac and intraembryonic stem cells. Blood. 1982, 59: 377-381.PubMedGoogle Scholar
- Niimi G, Usuda N, Shinzato M, Kaneko C, Nagamura Y, Pereda J: Histochemical study of the definitive erythropoietic foci in the chicken yolk sac. Ital J Anat Embryol= Archivio italiano di anatomia ed embriologia. 2008, 113: 9-Google Scholar
- Dieterlen-Lievre F: On the origin of haemopoietic stem cells in the avian embryo: an experimental approach. J Embryol Exp Morphol. 1975, 33: 607-619.PubMedGoogle Scholar
- Dieterlen-Lievre F, Beaupain D, Martin C: Origin of erythropoietic stem cells in avian development: shift from the yolk sac to an intraembryonic site. Annales d'immunologie. 1976, 127: 857-PubMedGoogle Scholar
- Martin C, Beaupain D, Dieterlen-Lievre F: Developmental relationships between vitelline and intra-embryonic haemopoiesis studied in avian ‘yolk sac chimaeras’. Cell differentiation. 1978, 7: 115-130. 10.1016/0045-6039(78)90012-X.PubMedView ArticleGoogle Scholar
- Dieterlen-Lièvre F, Martin C: Diffuse intraembryonic hemopoiesis in normal and chimeric avian development. Dev Biol. 1981, 88: 180-191. 10.1016/0012-1606(81)90228-1.PubMedView ArticleGoogle Scholar
- Bar-Joseph Z, Gitter A, Simon I: Studying and modelling dynamic biological processes using time-series gene expression data. Nat Rev Genet. 2012, 13: 552-564. 10.1038/nrg3244.PubMedView ArticleGoogle Scholar
- Kozyraki R, Gofflot F: Multiligand endocytosis and congenital defects: roles of cubilin, megalin and amnionless. Curr Pharm Des. 2007, 13: 3038-3046. 10.2174/138161207782110507.PubMedView ArticleGoogle Scholar
- Maldjian A, Farkas K, Noble RC, Cocchi M, Speake BK: The transfer of docosahexaenoic acid from the yolk to the tissues of the chick embryo. Biochimica Biophysica Acta. 1995, 1258: 81-89. 10.1016/0005-2760(95)00101-H.View ArticleGoogle Scholar
- Nakazawa F, Alev C, Jakt LM, Sheng G: Yolk sac endoderm is the major source of serum proteins and lipids and is involved in the regulation of vascular integrity in early chick development. Dev Dyn. 2011, 240: 2002-2010. 10.1002/dvdy.22690.PubMedView ArticleGoogle Scholar
- Monaco HL: The transthyretin-retinol-binding protein complex. Biochim Biophys Acta. 2000, 1482: 65-72. 10.1016/S0167-4838(00)00140-0.PubMedView ArticleGoogle Scholar
- Gilbert S, Loranger A, Daigle N, Marceau N: Simple epithelium keratins 8 and 18 provide resistance to Fas-mediated apoptosis. The protection occurs through a receptor-targeting modulation. J Cell Biol. 2001, 154: 763-773. 10.1083/jcb.200102130.PubMed CentralPubMedView ArticleGoogle Scholar
- Lin Y-M, Chen Y-R, Lin J-R, Wang W-J, Inoko A, Inagaki M, Wu Y-C, Chen R-H: eIF3k regulates apoptosis in epithelial cells by releasing caspase 3 from keratin-containing inclusions. J Cell Sci. 2008, 121 (Pt 14): 2382-2393.PubMedView ArticleGoogle Scholar
- Surai PF, Speake BK: Distribution of carotenoids from the yolk to the tissues of the chick embryo. J Nutr Biochem. 1998, 9: 645-651. 10.1016/S0955-2863(98)00068-0.View ArticleGoogle Scholar
- Speake BK, Teale MC: Metabolism and functions of omega-3 fatty acids during development of the avian embryo. Omega 3 Fatty Acid Res. 2006, 187-223.Google Scholar
- Lin GL, Himes JA: Cornelius CE: Bilirubin and biliverdin excretion by the chicken. Am J Physiol. 1974, 226: 881-885.PubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.