Comparative systems biology of human and mouse as a tool to guide the modeling of human placental pathology

Brian Cox¹, Max Kotlyar^2,3, Andreas I Evangelou⁴, Vladimir Ignatchenko⁴, Alex Ignatchenko⁴, Kathie Whiteley⁵, Igor Jurisica^2,3,6, S Lee Adamson^5,7, Janet Rossant^1,8 & Thomas Kislinger^3,4

Article highlights

Synopsis

The modeling of human disease in mouse has had success, but there have also been many failures. A priori, it is unclear if mutation of the orthologous gene in mouse will recapitulate the human phenotype. Although, many large-scale datasets for microarray and proteomics exist for both mouse and human, they are rarely integrated and compared. To better predict if mutation in a mouse gene will give the equivalent phenotype as the human ortholog, a detailed molecular understanding of the orthologous/homologous tissues from human and mouse are needed. By using a dataset of ortholog gene and protein expression, comparisons can be made as to the conservation of protein interaction networks and regarding the co-expression of phenotype-associated proteins. One organ that can be obtained from non-pathological human samples is the placenta. This organ is also the source of the primary defect in two of the most prevalent diseases of pregnancy, intrauterine growth restriction (IUGR) and preeclampsia (PE).

Although the placentas of no two mammalian species are the same, the placentas of humans and mice have strong similarities. In both species, maternal blood from the uterine arteries enters the placenta from large diameter, spiral arteries located in the maternal decidua. The maternal blood then percolates through a dense mesh of channels created and lined by fetal trophoblast cells in which an equally dense network of fetal capillaries is localized. This region is the site of feto-maternal exchange and is called the villous tree in humans and the labyrinth in mice. In both species, the umbilical vessels connect the fetal capillaries of the placental exchange region with the fetal body circulation.

IUGR and PE, which combined affect 5% of all pregnancies are both associated with serious morbidity and mortality, and the only known treatment is premature delivery, which places the baby at high risk of prematurity-related complications. Although, PE and IUGR have been intensely studied for decades, the molecular pathways, etiology and pathogenesis of these diseases remain poorly understood. Our aim was to empirically determine the molecular similarity of these two vascular exchange regions based on subcellular proteomics and transcriptional profiling, to assess the compatibility of mouse placenta to model human placenta biology.

Our proteomic analysis was validated by comparison to both matched mRNA samples analyzed by microarray and to 4872 proteins assayed by immunohistochemistry (IHC) on human placenta samples in the Human Protein Atlas (www.proteinatlas.org). In both cases high correlation was found between the protein to microarray (95%) and protein to IHC 90%.

Of key interest was the degree of molecular similarity between the mouse and human vascular exchange regions of the placenta. Data for observed proteins and their corresponding microarray probe sets were paired for human and mouse placenta samples. These protein/probe pairs were then linked between the two species using orthology data form ENSEMBL. This dataset of orthologous proteins/probes was then clustered to identify genes with common expression. This could be further classified into three categories. In all cases microarray data showed expression in both species, and protein data were observed in both species (cluster I), one species (cluster II) or in neither species (cluster III) (Figure 2A). Organellar localization associated with significant enrichment for appropriate Gene Ontology annotation terms was observed for cluster I proteins (Figure 2B). Interestingly, organellar localization showed striking agreement for co-expressed ortholog proteins in cluster I (Figure 2A and C) as predicted by Pearson correlation, which argues for functional conservation for most ortholog proteins.

Figure 2

Heat map of protein and mRNA ortholog expression in human and mouse. (A) A heat map of one-to-one orthologs showing protein expression (in red with scale shown at left below) and mRNA expression (in blue with scale shown at right below) for mouse and human placental exchange tissues. All 6970 confidently detected ortholog proteins are shown clustered into three groups; proteins observed in both species (cluster I), one species (cluster II) or in neither species (cluster III). (B) Significant enrichment for appropriate Gene Ontology annotation terms was observed for cluster I proteins. (C) The Pearson correlation (mouse versus human) for organellar localization of co-expressed ortholog proteins in cluster I. High agreement suggests conservation of function for most ortholog proteins in this cluster.

Full figure and legend (302K)Figures & Tables index

Overall our analysis revealed 70% conserved expression of one-to-one ortholog genes between mouse and human placental exchange regions. We made similar comparison of microarray data deposited at the GEO repository, for other matched tissues. We found similar levels of conservation for matched tissues and lower levels against all other tissues. We next compared the level of conservation of genes that when mutant in the mouse give placenta phenotypes. We found an even higher conservation of expression of these phenotypic genes between mouse and human (80%).

To generate a candidate list of genes involved in the development and/or maintenance of the placenta exchange regions, we generated a protein–protein interaction (PPI) network of all 2869 co-expressed proteins. Proteins for genes that give a placenta phenotype when mutated were selected and a subnetwork significantly enriched (P<0.05) in these proteins was extracted. The final network consisted of 15 seed proteins (those with placenta phenotypes) and 34 additional interacting proteins that had statistically significant associations (Figure 6). The majority of the network could be mapped into Protein Atlas (41/49 proteins had available antibodies). Of the 39 proteins for which we could score expression in the IHC images, we carefully evaluated their cell-type specific expression patterns.

Figure 6

Schematic of the cellular topology of the protein–protein interaction network. Shown are protein–protein interaction networks from I²D version 1.71, seeded with proteins that give a placental labyrinth or placental vascular phenotype when knocked out (rectangle or triangle). Proteins are colored to show cell-type expression. Node shapes indicate the placental phenotype (rectangle or triangle) or if no placental phenotype is known (oval or hexagon). Enrichment of proteins involved in extracellular matrix (yellow circle) and focal adhesion complex (blue circle) were observed (P<0.01). Other groupings are for clarity only. Eps15 and Cd82 (hexagons) are both novel members with no available knockout models. Visualization was done using NAViGaTOR version 2.15.

Full figure and legend (655K)Figures & Tables index

As the goal of this work was to identify proteins with hitherto unknown involvement in the feto-maternal exchange region, we cross-referenced the network members back to mouse knockout and OMIM phenotypes to determine associations with human disease phenotypes. Of the 34 non-seed members of the network, 32 had reported mouse knockouts (Cd82 and Eps15 were the exceptions) and only one protein was associated with human disease (mutations in Decorin cause stromal corneal dystrophy). Although, these mouse knock-outs are not categorized as having placental phenotypes in the MGI database, careful examination showed that 12/34 non-seed network members are annotated has having embryonic lethality and 3 others with the annotation decreased litter size, hallmarks of as yet potentially uncharacterized placental defects.

In summary, we have generated the first detailed proteomic and transcriptomic comparison of a complex human and mouse tissue. Comprehensive computational analyses revealed high molecular similarity of expressed ortholog gene products and provided a high quality resource for mechanistic investigations using mouse model systems. PPI network analysis revealed novel candidates likely central to the biology of this tissue. We have generated an intuitive web-interface to the entire data (www.kislingerlab.com) and deposited the raw data to Tranche (http://tranche.proteomecommons.org) and to GEO (http://www.ncbi.nlm.nih.gov/geo).

MORE ARTICLES LIKE THIS

These links to content published by NPG are automatically generated.