Protein interaction network topology uncovers melanogenesis regulatory network components within functional genomics datasets

Abstract

Background: RNA-mediated interference (RNAi)-based functional genomics is a systems-level approach to identify novel genes that control biological phenotypes. Existing computational approaches can identify individual genes from RNAi datasets that regulate a given biological process. However, currently available methods cannot identify which RNAi screen "hits" are novel components of well-characterized biological pathways known to regulate the interrogated phenotype. In this study, we describe a method to identify genes from RNAi datasets that are novel components of known biological pathways. We experimentally validate our approach in the context of a recently completed RNAi screen to identify novel regulators of melanogenesis.

Results: In this study, we utilize a PPI network topology-based approach to identify targets within our RNAi dataset that may be components of known melanogenesis regulatory pathways. Our computational approach identifies a set of screen targets that cluster topologically in a human PPI network with the known pigment regulator Endothelin receptor type B (EDNRB). Validation studies reveal that these genes impact pigment production and EDNRB signaling in pigmented melanoma cells (MNT-1) and normal melanocytes.

Conclusions: We present an approach that identifies novel components of well-characterized biological pathways from functional genomics datasets that could not have been identified by existing statistical and computational approaches.

Figure 1

Graphlet degree vectors. (a) All 9 graphlets on 2, 3 and 4 nodes, denoted by G₀, G₁, ..., G₈; they contain 15 topologically unique node types, called "automorphism orbits," denoted by 0, 1, 2 ..., 14. In a particular graphlet, nodes belonging to the same orbit are of the same shade [52]. (b) An illustration of the "Graphlet Degree Vector" (GDV), or a "signature" of node v; coordinates of a GDV count how many times a node is touched by a particular automorphism orbit, such as an edge (the leftmost panel), a triangle (the middle panel), or a square (the rightmost panel). Hence, the degree is generalized to a GDV [17]. The GDV of node v is presented in the table for orbits 0 to 14: v is touched by 5 edges (orbit 0), end-nodes of 2 graphlets G₁ (orbit 1), etc. Values of the 73 coordinates (for all of the 30 2-5-node graphlets) of the GDV of node v are presented in Additional file 1 Figure S1.

Figure 2

Application of protein interaction network topology to uncover novel melanogenesis gene networks from functional genomics data. Step-by-step methodology diagram is presented.

Figure 3

EDNRB network containing proteins in the EDNRB cluster and their direct neighbors, as well as all interactions from the human PPI network that exist between these nodes. Known pigment regulators (KPRs) from the ESPCR database and screen pigment regulators (SPRs) are noted.

Figure 4

Identification of novel EDNRB network components that impact melanogenesis. (a) Genes from EDNRB network are examined for their impact on pigment production in MNT-1 cells. The impact of each siRNA on pigment production is calculated as described in [14] relative to control (c) and tyrosinase (TYR) siRNA treated cells using a normalized percent inhibition calculation. (b) Ten genes are selected from panel (a) that have > 50% NPI, and are not the result of siRNA off-target effects (2/3 of oligos have > 50% inhibition). The efficacy of protein knockdown for 10 pooled siRNAs directed towards these genes is measured by quantitative RT-PCR at both 24 and 48 hours after transfection. The timepoint where maximum inhibition of expression is observed is reported. AGTR1 mRNA level after siRNA treatment is below the detection limit and is not reported. Control siRNAs are depicted with a black bar, and target siRNAs with a white bar. *, p < 0.05. **, p < 0.01. (c) Both EDNRB and selected novel EDNRB network components impact the MITF protein levels in MNT-1 cells. MNT-1 cells are transfected with the indicated siRNAs for 96 hours followed by immunoblotting with MITF and ERK antibody as shown.

Figure 5

Computationally derived EDNRB network components impact EDNRB signaling through CREB phosphorylation. (a) Novel EDNRB network components impact both tyrosinase and MITF expression. MNT-1 melanoma cells are transfected with pooled siRNAs and relative MITF (black bar) and Tyrosinase (gray bar) expression is measured by quantitative RT-PCR 96 hours after transfection. *, p < 0.05. **, p < 0.01. (b) Novel components of the EDNRB network impact CREB phosphorylation. Lysates from MNT-1 cells treated with the indicated pooled siRNAs are subjected to immunoblotting with the indicated antibodies. (c) PLCD1 impacts MITF protein levels in melanocytes. Lysates from PLCD1 siRNA transfected melanocytes are subjected to immunoblotting with the indicated antibodies. (d) PLCD1 impacts CREB phophorylation in normal human melanocytes. Lysates from melanocytes transfected with the indicated siRNAs followed by EDNRB stimulation (treatment with 10 mM Endothelin-1 for 5 minutes) are subjected to immunoblotting with the indicated antibodies.