The FABP5 gene spans approximately 4.3 kb and has 4 exons, all of which contain coding sequences.
Ten SNPs have been validated in the coding region of FABP5, 7 missense and 3 synonymous (not affecting the amino acid sequence). Two SNPs in the regulatory region of FABP5 have been identified in association with type 2 diabetes mellitus (T2DM): rs454550 showed significant association with T2DM in a non-Hispanic White sample, and a newly reported SNP at genomic position 82354416 was associated with T2DM in both non-Hispanic White and African sample populations (Bu et al., 2011). A missense 340G>C (Gly114Arg) SNP in the coding region of FABP5 has been linked to autism (Maekawa et al., 2010). FABP5 SNPs are described at the following site (http://www.ncbi.nlm.nih.gov/snp).

The transcript is approximately 750 nucleotides with an open reading frame of 408 nucleotides, a 5 untranslated region of 113 nucleotides and a 3 untranslated region of 209 nucleotides excluding the poly-A tail.
Based on Expressed Sequence Tag (EST) data, FABP5 RNA is present in a wide range of human tissues, with highest levels in the uterus, pharynx, bone, and heart. FABP5 transcripts have been identified in normal mammalian cells, including adipocytes, tongue epithelia, lens (Wen et al., 1995), developing retina, lung and mammary gland epithelia (Zimmerman and Veerkamp, 2002), macrophages, kidney, liver, and skeletal muscle (Smathers and Petersen, 2011). Immunohistochemistry has confirmed the presence of FABP5 in endothelial cells of the placenta, heart, small intestine, and renal medulla (Masouyé et al., 1997).
Nerve growth factor (NGF) positively regulates FABP5 expression in PC12 cells (pheochromocytoma of rat adrenal medulla) through a MEK-dependent pathway (Liu et al., 2008). FABP5 is a known target of c-Myc (Coller et al., 2000), and epithelial cell adhesion molecule epCAM has been demonstrated to upregulate FABP5 expression, presumably via induction of c-Myc (Münz et al., 2005). The promoter region of FABP5 contains two cognate response elements (CREs) to transcription factor NF-κB (Kannan-Thulasiraman et al., 2010). Epidermal growth factor receptor (EGFR) activation directly increases FABP5 expression through the ERK and phosphatidylinositol-3-kinase cascades and subsequent activation of NF-κB (Kannan-Thulasiraman et al., 2010). Ligands of peroxisome proliferator-activated receptors (PPARs) have been shown to influence FABP5 expression: PPARα and PPARγ agonists upregulate FABP5 expression, whereas PPARβ activation decreases FABP5 expression (Hyder et al., 2010). In contrast to FABP7, mRNA length for FABP5 does not vary throughout the body, limiting the likelihood of post-transcriptional modification (Zimmerman and Veerkamp, 2002).

The human genome contains 14 pseudogenes similar to the FABP5 locus.
Predicted pseudogenes are listed here: http://www.ncbi.nlm.nih.gov/gene/?Term=related_functional_gene_2171%5Bgroup%5D.

FABP5 is a member of the intracellular lipid binding protein family. FABP5 is a 135 amino acid protein with an estimated molecular mass of 15.1 kDa. Similar to other FABPs, FABP5 has a beta-clam (beta barrel) structure composed of 10 anti-parallel beta sheets and two N-terminal alpha helices, with low backbone motility (Gutiérrez-González et al., 2002). Six cysteine residues and a disulfide bridge between Cys120 and Cys127 contribute to protein stability and are hypothesized to relieve oxidative stress by thiol disulfide interchange reaction (Odani et al., 2000). Hydrophobic ligands are bound within the water-filled central cavity formed by the beta barrel. The carboxylic head group of linoleic acid forms a salt bridge with Arg-129 and hydrogen bonds with the hydroxyl groups of Tyr-131 and Arg-109 in the binding pocket of FABP5 (Armstrong et al., 2014). Hydrophobic residues within the binding pocket such as Cys-120 also facilitate ligand binding. FABP5 contains ligand-sensitive tertiary nuclear localization sequences in the α1 and α2 helices and a putative nuclear export signal located at the edge of the β-barrel. The putative NES consists of Lys-69, Lys-94, and Phe-89 residues (Armstrong et al., 2014).

FABP5 was first identified as a low molecular weight protein highly upregulated in human psoriatic skin (Madsen et al., 1992). FABP5 is expressed in numerous normal adult epithelial tissues, including those of the reproductive and urinary tracts, mammary glands, lungs, tongue, and lens. FABP5 is also expressed in macrophages, bone, and skeletal muscle (Smathers and Petersen, 2011).
The expression of FABP5 during mammalian neurogenesis has been investigated using rodent models, with similar results obtained in mouse and rat. FABP5 transcripts are detected in mid-term embryonic rat brain, peaking at birth and gradually decreasing from P1 to P21, with expression virtually undetectable in the adult brain (Owada et al., 2002), except during regeneration in response to neuronal injury (Allen et al., 2001). FABP5 is critical in post-natal hippocampal dentate gyrus neurogenesis in mice (Matsumata et al., 2012).

The FABP5 protein has a nuclear or cytoplasmic localization that is dependent upon ligand binding and cell type. Retinoic acid (RA), linoleic acid and arachidonic acid (AA) have been shown to elicit nuclear translocation through allosteric interaction between the ligand-sensing β2 loop and nuclear localization sequence (NLS) in the alpha helix (Tan et al., 2001; Schug et al., 2007; Armstrong et al., 2014).

Unique to brain-expressed FABPs, both human and rat FABP5 have a high affinity for saturated fatty acids such as stearic acid, with a Kd of 168.1 nM as determined by 1-anili-nonapthalene-8-sulfonic acid (ANS) assay (Liu et al., 2010), and 290 nM as determined by the Lipidex assay (see Table) (Liu et al., 2010). FABP5 has a broad range of saturated and unsaturated hydrophobic ligands, including linoleic acid, eicosapentanenoic acid (EPA), docosahexaenoic acid (DHA), AA and their derivatives (Sanson et al., 2014). Nuclear localization upon binding to linoleic acid and AA is thought to influence gene expression through activation of the nuclear receptor PPARβ/δ (Tan et al., 2001), which controls the expression of genes involved in lipid and glucose metabolism, differentiation, and survival (Schug et al., 2007).
FABP5 expression during development is associated with neurite outgrowth (Allen et al., 2001), with an important role in post-natal hippocampal dentate gyrus neurogenesis through nuclear transport of RA to activate PPARβ/δ (Matsumata et al., 2012). The hippocampi of FABP5-null mice contain excess neuronal progenitors and are deficient in mature neurons (Yu et al., 2012). FABP5 knockout mice do not have a brain phenotype, perhaps as a result of compensation by other FABP members.
FABP5 has been shown to facilitate nociception in mice through transport of adandamide (AEA), an anti-nociceptive lipid that is hydrolyzed into AA within the cell. FABP5 knockout mice accumulate AEA, which results in PPARβ/δ inactivation (Yu et al., 2014). Pharmacological inhibition of FABP5 reduces inflammatory, visceral, and neuropathic pain in mice (Kaczocha et al., 2014).
FABP5 knockout mice exhibit reduced basal transepithelial water loss (TEWL) and delayed recovery in TEWL upon disruption of the lipid barrier, suggesting that FABP5 plays a critical role in maintenance of the water permeability barrier of the skin (Owada et al., 2002).
Several cell types in the immune system, including macrophages, thymic epithelial cells, and subsets of lymphocytes express FABP5, and animal studies indicate that FABP5 plays a role in immune regulation (Grau et al., 2003; Adachi et al., 2012). Production of cytokines IL-7 and IL-18 is increased in stromal cells over-expressing FABP5 (Adachi et al., 2012), while ablation of FABP5 results in increased macrophage and neutrophil infiltration during influenza A infection (Gally et al., 2013). Lung epithelial cells initially downregulate FABP5 expression during the onset of the anti-viral response, and recover pre-infection levels upon attenuation of inflammation. Deficiency in FABP5 correlates with increased oxidative damage and lipid peroxidation over the course of infection, indicating that FABP5 is required for the prevention of tissue damage resulting from excessive inflammation (Gally et al., 2013).

The human FABP5 amino acid sequence is 64.8% identical to chicken FABP5 and 80% identical to mouse FABP5 isoform 1. Human FABP5 shows variable sequence identity with the other FABP paralogues, with the highest identity to FABP8/PMP2 (59%) and the lowest identity to FABP2 and FABP6 (27%). FABP5 is phylogenetically related to FABP3 and FABP7, which are also expressed in the brain (Schoentgen et al., 1989; Godbout, 1993; Bennett et al., 1994; Feng et al., 1994) and shows 51% and 46% amino acid sequence identity with FABP3 and FABP7, respectively. FABP5 also has sequence identity to other lipid binding proteins, with 36% identity to human cellular retinoic acid binding protein 1 (CRABP1).

Several variants of FABP5 have been linked to human diseases, including type 2 diabetes. Nine MGI mutant phenotypes in mice have been reported, with disruption in FABP5 resulting in defects in epithelial water balance and resistance to obesity.