Week 8 - L8 - Protein-Protein Interactions, Pathway Mapping, and Disease Associations via the STRING Database

Overview of Protein-Protein Interactions and the STRING Database

• Context of Protein Function:

  • Proteins do not function in isolation; their functional roles must be evaluated within the context of their interactions with other molecules, pathways, or large macromolecular complexes.

  • Functional relationships exist as direct physical interactions or indirect pathway-level associations.

• The STRING Database:

  • STRING is a specialized computational database designed to store, predict, and analyze protein-protein interactions (PPIs).

  • Captures both known and predicted protein-protein interactions.

  • Distinguishes between two fundamental types of association:

  • Direct Physical Associations: Structural relationships where two or more proteins physically bind directly to one another.

  • Indirect Functional Associations: Functional relationships where proteins operate within the same biological pathway or process without necessarily engaging in direct physical contact during their function.

• Data Sources Integrated into STRING:

  • Genomics: Inferred from genomic context; if two genes are located adjacent to each other in a genome, there is a high probability that their protein products are functionally linked.

  • Proteomics: Derived from experimental, high-throughput assay data (such as pull-down assays) where researchers experimentally detect and report direct physical associations.

  • Transcriptomics: Derived from gene co-expression data. Co-expressed genes indicate a high probability that their protein products function together and potentially bind physically. Conservation of co-expression patterns across multiple species provides significantly stronger evidence of functional association.

  • Automated Text Mining: Algorithms automatically mine scientific literature abstracts from databases such as PubMed to extract reported protein-protein associations.

  • External Database Cross-Referencing: Integrates existing functional and structural data from external biological repositories, including PDB (Protein Data Bank), BIOSIC, and various other databases.

STRING Search Workflow and Analysis: Phenylalanine Hydroxylase (PAH)

• Database Search Mechanics:

  • Executing a search requires navigating to the search portal, entering the target protein name (e.g., phenylalanine hydroxylase), and setting or auto-detecting the organism species (e.g., Homo sapiens / human).

• Network Visualization Features:

  • Central Target Node: Displays the queried protein (e.g., human Phenylalanine Hydroxylase / PAH) as a central node, typically rendered in red.

  • Surrounding Network Nodes: Other nodes in the network represent distinct proteins functionally linked to the central node. Hovering over a node displays the protein's standard name and functional summary.

  • Tabular Node Summary: A complete list below the network diagram enumerates every node and the corresponding protein name represented in the graph.

  • Connecting Lines (Network Edges):

  • Edges between nodes represent specific lines of evidence supporting a functional or physical association.

  • A higher density/number of connecting lines indicates stronger combined evidence supporting the association.

  • STRING aggregates individual evidence types into a combined confidence score, where values approaching $1$ signify exceptionally strong evidence of direct functional linkage within the same pathway.

• Specific Functional Associations of Phenylalanine Hydroxylase:

  • Tyrosine Aminotransferase:

  • Converts tyrosine into downstream metabolic intermediates.

  • Functional link: Tyrosine is the immediate chemical product of the hydroxylation reaction catalyzed by phenylalanine hydroxylase. Thus, these two enzymes operate sequentially within the exact same metabolic pathway.

  • Evidentiary lines in STRING: Gene co-expression, shared entries in external databases, and co-occurrences in PubMed abstracts.

  • Dihydroptan Reductase (Dihydropteridine Reductase):

  • Synthesizes and regenerates the essential pterin coenzyme ($BH_4$) required for phenylalanine hydroxylase enzymatic activity.

  • Evidentiary lines in STRING: Genomic co-occurrence, gene co-expression, direct physical binding, co-appearance in external databases, and co-mentions in PubMed abstracts. Combined evidence score is very close to $1$.

Disease-Gene Associations and Database Cross-Referencing

• Locating Disease Associations in STRING:

  • Disease linkages are identified by navigating to the "Analysis" section and scrolling down to the "Disease Gene Associations" module.

• Diseases Linked to Phenylalanine Hydroxylase Mutations:

  • Phenylketonuria (PKU):

  • An amino acid metabolic disorder characterized by pathogenic mutations in the gene encoding phenylalanine hydroxylase.

  • Pterin Coenzyme Deficit Disorders:

  • Conditions primarily linked to deficits in the pterin coenzyme. While typically caused by primary mutations in genes responsible for pterin biosynthesis, pathogenic mutations in the phenylalanine hydroxylase gene can also give rise to disorders presenting with pterin coenzyme deficiency.

• In-Depth Exploration via GeneCards:

  • Accessing the node's "Pathways, Functions, and Resources" option redirects to the GeneCards database, providing extensive genetic and medical profiles.

  • GeneCards Profile Findings for PAH:

  • Identifies $7$ distinct diseases directly linked to mutations in the phenylalanine hydroxylase gene.

  • Phenylketonuria: Thoroughly investigated condition with over $500$ dedicated scientific publications on PubMed.

  • Mild Phenotypic Variants: Pathogenic variants in this gene can produce milder, less severe metabolic phenotypes.

  • Text-Mining Disease Associations: Extracted $32$ text-mined disease associations for the PAH gene.

  • Tyrosinemia: A metabolic disorder characterized by impaired tyrosine breakdown. Linked to PAH because PAH is the enzyme directly responsible for synthesizing tyrosine.

Pathways and Biochemical Mechanisms of Phenylalanine Hydroxylase

• Pathways Involving Phenylalanine Hydroxylase:

  • Phenylalanine Degradation / Tyrosine Biosynthesis Pathway:

  • Phenylalanine hydroxylase catalyzes the conversion of phenylalanine into tyrosine, representing tyrosine biosynthesis.

  • Tyrosine is subsequently processed by tyrosine aminotransferase and further metabolized down the catabolic pathway, ultimately yielding $CO_2$ and cellular energy.

  • Knockdown Effect: A mutation in the gene encoding phenylalanine hydroxylase impairs its enzymatic efficiency, causing a downstream knockdown effect that impacts the operational efficiency of every subsequent protein in the catabolic pathway.

  • Direct Physical Interaction Implication: Phenylalanine hydroxylase likely binds directly to tyrosine aminotransferase to facilitate metabolic channeling, directly transferring the tyrosine product from its active site into the active site of tyrosine aminotransferase.

  • Pterin Coenzyme ($BH_4$) Biosynthesis and Regeneration Pathway:

  • Pterin coenzyme ($BH_4$) is synthesized near the center of the cell by specific biosynthetic enzymes.

  • Pterin acts as a crucial coenzyme for phenylalanine hydroxylase. During the reaction converting phenylalanine to tyrosine, the coenzyme is altered/oxidized and released.

  • Dihydroptan reductase (dihydropteridine reductase) regenerates the modified coenzyme back into its active pterin form.

  • Phenotypic Equivalence: Inactivation of upstream pterin synthesis or regeneration enzymes deprives phenylalanine hydroxylase of its coenzyme, completely inactivating its ability to convert phenylalanine to tyrosine. Consequently, mutations in upstream pterin biosynthesis genes produce identical disease symptoms and phenotypes as direct mutations in the phenylalanine hydroxylase gene.

  • Direct Physical Interaction Implication: Phenylalanine hydroxylase is highly likely to bind directly to dihydroptan reductase, enabling efficient active-site-to-active-site transfer and recycling of the pterin coenzyme.

  • Detailed visual diagrams of pterin coenzyme recycling pathways are cataloged in Lehninger's Principles of Biochemistry.

Summary of Core Outcomes

• Contextual Analysis of Proteins:

  • Proteins do not act in isolation; understanding biological function requires evaluating proteins in the context of whole metabolic pathways and multi-protein macromolecular complexes.

• Interaction Measurement:

  • Methodologies exist to identify, quantify, and score the confidence of physical and functional protein-protein interactions as well as protein-ligand associations.

• Information Retrieval:

  • Comprehensive databases such as STRING and GeneCards aggregate literature, experimental proteomics, genomics, and transcriptomics to map gene-disease associations and complex metabolic pathway networks.