- Research article
- Open Access
Huntingtin-associated protein 1: Eutherian adaptation from a TRAK-like protein, conserved gene promoter elements, and localization in the human intestine
BMC Evolutionary Biology volume 16, Article number: 214 (2016)
Huntingtin-associated Protein 1 (HAP1) is expressed in neurons and endocrine cells, and is critical for postnatal survival in mice. HAP1 shares a conserved “HAP1_N” domain with TRAfficking Kinesin proteins TRAK1 and TRAK2 (vertebrate), Milton (Drosophila) and T27A3.1 (C. elegans). HAP1, TRAK1 and TRAK2 have a degree of common function, particularly regarding intracellular receptor trafficking. However, TRAK1, TRAK2 and Milton (which have a “Milt/TRAK” domain that is absent in human and rodent HAP1) differ in function to HAP1 in that they are mitochondrial transport proteins, while HAP1 has emerging roles in starvation response. We have investigated HAP1 function by examining its evolution, and upstream gene promoter sequences. We performed phylogenetic analyses of the HAP1_N domain family of proteins, incorporating HAP1 orthologues (identified by genomic synteny) from 5 vertebrate classes, and also searched the Dictyostelium proteome for a common ancestor. Computational analyses of mammalian HAP1 gene promoters were performed to identify phylogenetically conserved regulatory motifs.
We found that as recently as marsupials, HAP1 contained a Milt/TRAK domain and was more similar to TRAK1 and TRAK2 than to eutherian HAP1. The Milt/TRAK domain likely arose post multicellularity, as it was absent in the Dictyostelium proteome. It was lost from HAP1 in the eutherian lineage, and also from T27A3.1 in C. elegans. The HAP1 promoter from human, mouse, rat, rabbit, horse, dog, Tasmanian devil and opossum contained common sites for transcription factors involved in cell cycle, growth, differentiation, and stress response. A conserved arrangement of regulatory elements was identified, including sites for caudal-related homeobox transcription factors (CDX1 and CDX2), and myc-associated factor X (MAX) in the region of the TATA box. CDX1 and CDX2 are intestine-enriched factors, prompting investigation of HAP1 protein expression in the human duodenum. HAP1 was localized to singly dispersed mucosal cells, including a subset of serotonin-positive enterochromaffin cells.
We have identified eutherian HAP1 as an evolutionarily recent adaptation of a vertebrate TRAK protein-like ancestor, and found conserved CDX1/CDX2 and MAX transcription factor binding sites near the TATA box in mammalian HAP1 gene promoters. We also demonstrated that HAP1 is expressed in endocrine cells of the human gut.
Huntingtin-associated Protein 1 (HAP1) is a ~100 kDa protein that was first identified in a screen for proteins that interact with the Huntington’s disease (HD) gene product, huntingtin (HTT) . Initial interest in HAP1 stemmed from its potential role in HD pathogenesis, and has broadened due to its emerging roles in feeding and metabolism , and association with a range of conditions (in addition to HD) including depression , autism , Alzheimer’s disease , Joubert syndrome , Rett Syndrome , and cancer . The biological function of HAP1 remains to be fully elucidated, and is of potential importance to a number of research fields given the wide-ranging disease associations of this protein.
HAP1 is expressed primarily in neurons , and in endocrine cells of the periphery [10, 11]. HAP1 is also diffusely expressed in the mucosal layer of the stomach and small intestine in as-yet unidentified cells . At the intracellular level, HAP1 is associated with organelle transport [12–14], promotes neurite outgrowth , is transported along nerve axons  and facilitates receptor internalization [6, 15, 17]. HAP1 achieves this through interactions with binding partners involved in intracellular transport, such as kinesin light chain (KLC), dynactin subunit p150glued , dynein  and synapsin I . Through linking to cytoskeletal motor proteins, HAP1 acts in the trafficking of various cargoes such as GluR2-containing AMPA receptors , EGF receptors , GABAA receptors , BDNF , proBDNF , amyloid precursor protein  and β-actin RNA , and also plays a role in ciliogenesis . Such cytoskeletal motor interactions likely explain how HAP1 promotes the release of hormones such as insulin , and catecholamines , as well as neurotransmitters , through the regulation of vesicle transport to the plasma membrane.
Recent studies support a role for HAP1 in the promotion of energy conservation. HAP1 stabilizes levels of an inhibitor of mTORC1 , suppression of which is associated with the activation of autophagy. HAP1 localizes to autophagosomes to promote their trafficking along axons . Hap1 gene expression increases in response to fasting , and conversely, insulin mediates HAP1 degradation in orexigenic neurons of the hypothalamus , suggesting that the demand for HAP1 is reduced post feeding. These findings are consistent with a role for HAP1 in response to caloric insufficiency.
Newborn Hap1 −/− mouse pups fail to ingest milk, leading to postnatal starvation and death [27–29]. Conditional Hap1 gene inactivation in hypothalamic orexin neurons decreases food consumption and body weight , and hypothalamic neurodegeneration and decreased neurogenesis occur in mouse pups deficient in HAP1 [17, 28]. However, no defect in the hypothalamic expression of genes encoding appetite-regulating neuropeptides (AGRP, POMC and NPY) has been detected in Hap1 −/− mice  and as yet, the mechanism by which HAP1 deficiency causes postnatal failure to feed and thrive has not been established.
HAP1 is one of a group of related proteins also including vertebrate TRAfficking Kinesin-binding proteins TRAK1 (previously known as O-linked N-acetylglucosamine transferase interacting protein 106, OIP106) and TRAK2 (previously Gamma-aminobutyric acid(A) receptor-interacting factor, GRIF-1), the Drosophila Milton protein, and T27A3.1 in C. elegans. These proteins share a homologous region [31, 32] termed the HAP1_N domain (Pfam ID: PF04849). TRAK1, TRAK2 and Milton, however, contain an additional homologous region on the carboxyl side of the HAP1_N domain called the Milton/Trafficking kinesin-associated protein domain (Milt/TRAK domain; Pfam ID: PF12448), a domain that is absent in human and rodent HAP1 proteins. The HAP1_N domain is rich in coiled-coil motifs and facilitates the interaction of HAP1 with the coiled-coil domains possessed by many HAP1-interacting proteins including HTT [1, 33, 34], p150glued, PCM1 , Hrs , Kinesin-1 family member KIF5 , kinesin light chain , and HAP1 itself . Vertebrate TRAK proteins and Drosophila Milton interact with some of the same proteins/protein families as HAP1, and have overlapping functions. For example TRAK1 interacts with Hrs and is involved in EGFR sorting . TRAK1 and TRAK2 interact with GABAA receptors and are implicated in their trafficking [40, 41]. TRAK1 and 2, and Milton bind to kinesin-1 family proteins via the HAP1_N-containing N-terminal region [32, 42, 43] and interact with dynein/dynactin proteins  implicating them in anterograde and retrograde transport. Like HAP1, TRAK1 and TRAK2 proteins are involved in neuritic outgrowth (of axons and dendrites, respectively)  and are also able to self-interact .
In contrast to HAP1 however, TRAK1, TRAK2 and Drosophila Milton have a prominent role in mitochondrial trafficking. They link mitochondria to kinesin motor proteins via a complex with the Rho-like GTPase Miro, to facilitate transport of the mitochondria to nerve terminals, axons and dendrites to provide targeted provision of energy to localized areas within the cell [32, 43, 44, 46]. Although immunogold electron microscopy has shown HAP1 localizes to mitochondria , HAP1 has not been associated with mitochondrial targeting to neurites. In contrast to HAP1, the Milt/TRAK proteins have not been linked to starvation response and feeding.
In this study we investigated HAP1 function by assessing the evolution and divergence of HAP1_N protein family members, and conserved regulatory elements in mammalian HAP1 gene promoters. The online access to increasing numbers of sequenced and annotated genomes allowed inclusion of orthologues (determined by genomic synteny) representing the 5 vertebrate classes. Furthermore, since little is known about regulation of HAP1 gene expression, we also searched eight mammalian HAP1 gene upstream promoter sequences for regulatory elements and multi-element configurations (promoter models), using phylogenetic conservation to provide evidence for functionality. Finally, upon detecting a highly conserved promoter model near the start of transcription containing sites for intestine-enriched transcription factors, we examined HAP1 protein expression in the human duodenum.
Verification of HAP1 orthologous genes in vertebrates
Searches for HAP1 orthologues were performed of the genomes of the following species: human (Homo sapiens), rat (Rattus norvegicus), mouse (Mus musculus), dog (Canis familiaris), horse (Equus caballus), rabbit (Oryctolagus cuniculus), opossum (Monodelphis domestica), Tasmanian devil (Sarcophilus harrisii), zebrafish (Danio rerio), chicken (Gallus gallus), anole lizard (Anolis carolinensis) and clawed frog (Xenopus tropicalis). From here on, these species are referred to by their common names, except for Xenopus. These genome representatives of the 5 classes of vertebrate, including eutherian and marsupial mammals, were examined using the Ensembl genome browser  and were each found to contain three HAP1_N domain-containing genes. These genes were TRAK1, TRAK2, and a third gene that at the outset of this study was named HAP1 or TRAK1-like, or was unannotated, depending upon the species. In order to confirm whether the TRAK1-like and unannotated genes were orthologous to human HAP1, genomic contexts were compared for verification (Fig. 1, selected species shown). In human, the HAP1 gene was found to be flanked by the genes JUP (upstream, on HAP1 sense strand) encoding Junction Plakoglobin, and GAST (encoding Gastrin, downstream, on antisense strand) followed by EIF1 further downstream (encoding Elongation Initiation Factor 1, antisense strand), and a cluster of cytokeratin (KRT) genes (sense strand; Fig. 1). This genomic context was distinct from those of TRAK1 and TRAK2 (not shown). The same gene context was observed for Hap1 genes of other eutherian mammals (rat, mouse, dog, horse, rabbit), marsupials (opossum Trak-1 like protein gene and Tasmanian devil Hap1) and amphibian (Xenopus hap1). A similar arrangement was also observed in bird (chicken unannotated gene, ENSGALG00000023847) except that the Gast gene was absent. In fish (zebrafish), the HAP1_N- containing unannotated gene (ENSDARG00000074508) was positioned between jupa (upstream) and krt genes (downstream) consistent with other HAP1 orthologous genes, and was therefore considered a homolog. The gast and eif1 genes were absent in the region in zebrafish, and jupa was on the antisense strand, in contrast to other vertebrates. In reptile (anole lizard), jup (sense) was upstream of hap1 (with an uncharacterized locus between them) and eif1 (sense) was further upstream of jup, with additional genes (leprel4, and 3 non-coding RNA sequences) between them. In the lizard the cluster of krt genes was upstream of eif1 as in human, opossum, and Xenopus, but the krt cluster and eif gene were upstream rather than downstream of HAP1 (Fig. 1). No gast gene was present in the anole lizard. In summary, all vertebrates investigated contained three HAP1_N genes TRAK1, TRAK2 and a third gene in a comparable genomic context to human HAP1. In a more recent version of the Ensembl browser (Version 83, December 2015), the TRAK1-like and ‘unannotated’ orthologues described above have been renamed HAP1, consistent with our findings.
Non-eutherian vertebrate HAP1 proteins contain a TRAK/Milton domain
The presence of predicted domains within the amino acid sequence of vertebrate HAP1 proteins was compared using Ensembl  and NCBI’s conserved domain database . Whilst all vertebrate HAP1 proteins contained the characteristic HAP1_N domain, HAP1 proteins from marsupial (opossum and Tasmanian devil) and non-mammalian vertebrates (Xenopus, zebrafish, chicken and anole lizard) also contained a Milt/TRAK domain characteristic of TRAK1, TRAK2 and Drosophila Milton. HAP1 proteins from human and other eutherian mammals (rat, mouse, rabbit, dog and horse) all lacked the Milt/TRAK domain. Interestingly, the Milt/TRAK domain was also absent in T27A3.1, suggesting the domain was lost a second time, in the divergence of C. elegans (Fig. 2).
Sequence comparisons of HAP1_N protein family members
Sequence comparisons were performed using HAP1 proteins from selected eutherian mammals (human, dog, rat), marsupials (opossum, Tasmanian devil), and non-mammalian vertebrates (chicken, anole lizard, Xenopus, zebrafish), TRAK1 and TRAK2 proteins from human and opossum, and the C. elegans and Drosophila homologs. The complete identity matrix is shown in Additional file 1. Human HAP1 showed most similarity overall to HAP1 from dog (67 %) and rat (65 %). Opossum HAP1, which was highly similar to Tasmanian devil HAP1 (84 %), was more similar to HAP1 proteins from non-mammalian vertebrates (48–58 % identity) than to human HAP1 (39 %). In contrast, high homology was observed between opossum and human TRAK1 (92 %) and TRAK2 (83 %) proteins. In opossum, HAP1 was more similar to its paralogs TRAK1 (46 %) and TRAK2 (39 %) than human HAP1 was to its paralogs (29 and 26 %, respectively). HAP1 proteins from non-human eutherian mammals (rat and dog) were more closely related to human HAP1 (65, 67 %) than to opossum HAP1 (37, 37 %), as expected by their lack of Milt/TRAK domain. These results were consistent with a eutherian adaptation of HAP1 sequence since divergence from marsupials. HAP1 proteins from eutherian mammals (human, rat, dog), lacking the Milt/TRAK domain, were shorter in length (599–679aa) than vertebrate HAP1 orthologues that contained the Milt/TRAK domain (opossum, Tasmanian devil, chicken, lizard, Xenopus and zebrafish; 708–901aa). The latter were closer in length to vertebrate TRAK1 (860–1036aa) and TRAK2 (884–993aa) proteins. An alignment of all protein sequences in this analysis is shown in Additional file 2, indicating the position of the HAP1_N and Milt/TRAK domains.
A phylogram was compiled of isolated HAP1_N domain sequences from TRAK1, TRAK2 and HAP1 proteins from human, opossum, Xenopus, zebrafish, chicken and anole lizard, as well as dog and rat (HAP1 only). The HAP1_N domain of Milton was included as a more distant reference sequence for rooting the tree (Fig. 3). HAP1_N sequences from vertebrate TRAK1 orthologues clustered closely together in a single clade, as did those from vertebrate TRAK2 proteins in another clade. In contrast, the HAP1_N domain sequences of vertebrate HAP1 proteins were more divergent. The HAP1_N domain from eutherians had changed more substantially than that of marsupials since their last inferred common ancestor, as illustrated by an approximate 3-fold increase in phylogram arm length (Fig. 3), suggesting that loss of the TRAK domain altered the selective pressure on the HAP1_N domain.
Search for a common ancestral protein containing HAP1_N and Milt/TRAK domains prior to divergence of protostomes and deuterostomes
Genes encoding Drosophila Milton, and C. elegans T27A3.1 in invertebrates, and TRAK1, TRAK2 and HAP1 in vertebrates, are likely to have arisen from a common ancestral gene existing prior to the divergence of protostomes and deuterostomes (Fig. 2). We therefore searched for a common ancestor in the proteome of the early eukaryote Dictyostelium discoideum , an organism that exists as single cells and as a multicellular entity under different conditions. Searches were performed using as query sequences Drosophila Milton and human TRAK1 (the human HAP1_N protein with the highest similarity to Milton). A single likely ancestor was not apparent, however a number of Dictyostelium proteins showed significant similarity to the query sequences across the coiled-coil rich HAP1_N domain. No matches were found to other regions of the TRAK1 and Milton query proteins including the Milt/TRAK domain. The majority of the Dictyostelium proteins identified were actin-binding proteins (fimbrin and myosin proteins) or kinesin family members, whose homologs interact with TRAK1, TRAK2 and/or Milton proteins. This is consistent with the ability of the HAP1_N domain to facilitate interactions with other coiled-coil proteins including mediating their own self-association.
Actin-binding protein fimbrin-3 (fimC) was the most significant match for both queries. Other common hits were myosin II heavy chain (mhcA), fimbrin-4 (fimD), kinesin family member 4 (kif4), DDB_G0286355 (predicted component of the meiotic synaptonemal complex) and DDB_G0289761 (unannotated). These, and other proteins with similarity to either TRAK1 or Milton, and their associated functions, are listed in Table 1. These findings suggest that origins of the HAP1_N domain (present in cytoskeletal-based transport proteins of Dictyostelium), pre-dated the evolution of the Milt/TRAK domain which therefore arose post multicellularity, but prior to divergence of protostomes and deuterostomes.
Regulatory elements and transcription factor binding motifs common across mammalian HAP1 gene promoters
A computational promoter analysis was performed to identify transcription factor binding sites conserved across mammalian HAP1 gene promoters. The rationale was that the presence of a regulatory element within HAP1 upstream promoter sequences of multiple species increases its likelihood of being functionally relevant, and knowledge of the type of responses regulated by the corresponding transcription factors may provide insight into the biological pathway(s) in which HAP1 acts. The ‘Common TFs’ function of the Genomatix MatInspector software  was used to identify putative transcription factor binding sites within promoter sequences from human, rat, mouse, dog, horse, rabbit, opossum and Tasmanian devil. For human, rat and mouse, sequence consisted of 600 bp upstream and 120 bp downstream of, and including, the single transcription start site (TSS) indicated in the Ensembl genome browser. For dog, horse, rabbit, opossum and Tasmanian devil the TSS was not annotated so 720 bp immediately upstream of the start codon (ATG) was used since in human and rodents the start codon was within the first exon. These regions of sequence were thought likely to incorporate the core promoter region, which Genomatix defines as being within 500 bp upstream of the first TSS and 100 bp downstream of the last TSS.
Sites that were present in all tested eutherian mammals, and zero, 1, or both marsupials, are listed in Table 2 and fell into the main categories of general regulatory elements (TATA box, XCPE1 elements), cell cycle control (MAX, WT1, E2F, E2F6, E2F7, LRRFIP, MAZR), cell differentiation, growth and development (CDX1, CDX2, EGR1, NGFIC (EGR4), GRHL1, ZNF219), endoplasmic reticulum stress response (ATF6) and respiratory function (NRF1). All sequence positions (relative to the start of translation) and match scores are presented in Additional file 3.
Existence of a conserved MAX-CDX1/2-TBP promoter model in mammalian HAP1 gene promoters
The order, position and strand orientation of the transcription factors common to mammalian HAP1 gene promoters were compared between species to identify common patterns. Most notably, there was a site for TATA-binding protein (TBP), also known as the ‘TATA box’, on the sense strand, overlapped on the opposite strand by a site recognized by caudal-type homeobox transcription factors CDX1 and CDX2 (V$CDX1.02, V$CDX2.02). These sites were identically arranged in all HAP1 gene promoters tested here (Fig. 4a). A site for MYC-associated factor X (V$MAX.03) existed upstream, separated by 14–26 bp, on the antisense strand. In mouse and rabbit, this site was overlapped by an additional MAX site on the positive strand, but the site on the negative strand was conserved across all tested mammals. The distance between the conserved promoter model (MAX-CDX1/2-TBP) and the start codon was shorter in eutherian mammals (25–30 bp) than in marsupials (184–207 bp; Fig. 4a). Within the conserved MAX-CDX1/2-TBP model, marsupials contained an additional TATA box upstream of the first, which was absent in eutherian mammals. This upstream TATA box was also overlapped by a CDX2 site (V$CDX2.02) on the antisense strand (Fig. 4a). Five promoter motifs were found only in eutherian HAP1 gene promoters and not in those of marsupials: X gene core promoter element 1 (V$XCPE.01), Zinc finger/POZ domain factor ZF5 sites (V$ZF5.01 and V$ZF5.02), Nerve Growth Factor-Induced Protein C sites (V$NGFIC.01) and E2F sites (V$E2F.03) (Table 2). In 5 of 6 eutherian mammals, an XCPE1 site overlapped the MAX-CDX1/2-TBP model on the antisense DNA strand (Fig. 4a). The consensus DNA sequences for TBP (TATA box), CDX1, CDX2 and XCPE1 are shown as sequence logos in Fig. 4b. A database search for single nucleotide polymorphisms (SNPs) in the human HAP1 promoter identified rs116737192, a SNP that exists at a highly conserved position in the consensus binding motifs of TBP (5′TA[T/C]AAA), CDX1, and CDX2 (5′TTT[A/G]T), such that the minor allele abolishes the prediction of all of these sites by the MatInspector software. The SNP is well represented in the Encyclopedia of DNA Elements (ENCODE) which provided functional supporting evidence from ChIP-seq experiments that TBP, and associated regulatory factors including RNA polymerase II, bind at this genomic location in a number of human cell lines . This SNP is most common amongst individuals of African descent, in which the minor allele frequency is 9.1 % .
Functions associated with human genes with CDX1/2-TBP and MAX-CDX1/2-TBP promoter models
A search of the Genomatix human gene promoter database (112705 sequences) was performed using the ModelInspector program , to identify gene promoters containing the CDX1, CDX2 and TBP sites in the identical arrangement as was observed in mammalian HAP1 gene promoters (as described in Materials and Methods). The search resulted in 5956 matches in 5712 sequences (~5 % of human promoter sequences). The most significantly enriched biological functions associated with CDX1/CDX2/TBP genes, represented by Gene Ontology (GO) terms, included processes related to cell-cell adhesion, calcium signaling, cell cycle, ATP catabolic process, purine metabolism, phosphorous metabolism, microtubule-based movement, organelle formation, retinoic acid receptor signaling, and nervous system development (Table 3). Including the upstream MAX site in the model resulted in 68 matches in 67 sequences (~0.06 % of human promoter sequences in the database) corresponding to 55 genes. Enriched biological processes included negative regulation of retinoic acid receptor signaling, negative regulation of apoptosis, cell proliferation, intracellular receptor signaling, development, and detection and transport of calcium (Table 3). Of note, the latter model was found to be common to several PRAME family genes (PRAMEF5, 6, 9, 15, 20 and 25; Table 3) that exist in a cluster of related genes on human Chromosome 1. The commonality of function of these genes is likely to have caused bias amongst the enriched biological processes in this small gene set.
Expression of HAP1 in the human duodenum
CDX1 and CDX2 are intestine-specific transcription factors differentially involved in enterocyte proliferation and differentiation. The presence of conserved sites for these factors in HAP1 gene promoters prompted us to investigate HAP1 expression in the intestine. HAP1 protein is known to be present in singly dispersed cells of the mucosal layer of the stomach and duodenum of mice . Enteroendocrine cells are dispersed throughout the epithelial lining of the gut wall and it is presumed that the HAP1-positive cells are enteroendocrine cells. However, this has not been confirmed and no information exists on HAP1 expression in the human gut. Using immunolabeling we investigated HAP1 protein expression in the human duodenum. HAP1 was detected in mucosal villi from lean, healthy human subjects, in ~10–20 cells per villus. The highest density labeling was evident at the basolateral membrane (Fig. 5). Due to the abundant number of HAP1-positive cells, we tested whether HAP1 was expressed in the enterochromaffin (EC) cell population. EC cells are the most populous of all enteroendocrine cells and synthesize and secrete 95 % of serotonin (5-hydroxytryptamine, 5-HT) in the body . This cell type is also known to express the transcription factor CDX2 . Co-staining for HAP1 and 5-HT revealed cells singly labeled for either target (Fig. 5, a-e), as well as a subset that were immunopositive for both HAP1 and 5-HT (Fig. 5, f-j). Intracellular distribution of both HAP1 and 5-HT label was punctuate or vesicular, and within co-labeled cells, a degree of overlap in staining was evident (Fig. 5h).
Insight into HAP1 function from its evolution
HAP1 of rodent and human have been recognized as sharing a region of homology (the HAP1_N domain) with TRAK1 and 2 proteins, and invertebrate proteins Milton (Drosophila) and T27A3.1 (C. elegans). Previous studies have identified similarities in sequence and function between TRAK1, TRAK2 and Milton which possess an additional domain, the Milt/TRAK domain, further carboxyl to the HAP1_N domain. The C. elegans homolog, T27A3.1, has been identified as a homolog of mammalian HAP1 . In this study we compared all the known members of this HAP1_N family, extending this to include additional orthologues from the 5 vertebrate classes.
We have identified (and confirmed by comparison of genomic context), HAP1 orthologues in human, rat, mouse, dog, rabbit, opossum, Tasmanian devil, Xenopus, chicken, anole lizard and zebrafish. No additional HAP1_N domain-containing paralogs apart from TRAK1 and TRAK2 were found within these genomes/proteomes. By sequence and domain comparisons we have determined that as recently as the marsupial lineage, HAP1 proteins had a Milt/TRAK domain and were more closely related to paralogs TRAK1 and TRAK2 than to eutherian HAP1.
HAP1_N family members with TRAK domains exist in both vertebrates and Drosophila. Within the proteome of the early eukaryote Dictyostelium, we found regions of similarity to the HAP1_N domain in actin binding and cytoskeletal proteins but no similarity to the Milt/TRAK domain. We therefore propose that the HAP1_N domain has ancient origins in actin/microtubule attachment and transport functions, and that the Milt/TRAK domain arose post multicellularity but prior to the divergence of protostomes and deuterostomes. Given that HAP1_N proteins with a Milt/TRAK domain (such as Milton, TRAK1 and TRAK2) function in targeted transport of mitochondria to distant cellular regions of energy need, a putative ancestral homolog containing HAP1_N and Milt/TRAK domains may have evolved with the development of longer cell types in multicellular organisms.
In deuterostomes, the number of Milt/TRAK domain-containing HAP1 proteins has increased such that in vertebrates there are three homologs (TRAK1, TRAK2, and HAP1). Since all 5 classes of vertebrate have these 3 paralogs, the increase in paralog number most likely occurred prior to divergence of the vertebrate lineage into these classes. Sequence analyses indicated less similarity amongst the vertebrate Milt/TRAK domain-containing HAP1 orthologues (in zebrafish, anole lizard, chicken, Xenopus and opossum), than observed between TRAK1 and TRAK2 orthologues. This suggests HAP1 sequences were already diverging (compared to TRAK proteins) prior to loss of the Milt/TRAK domain in eutherians. However loss of the Milt/TRAK domain may have further altered the selective pressure on the eutherian HAP1 domain, as the latter domain has diverged approximately 3 times more in eutherians than in marsupials, since their last common ancestor.
We found that the Milt/TRAK domain has been lost a second time, in the lineage giving rise to the C. elegans homolog T27A3.1. Since only a single HAP1_N-containing homolog has been found in invertebrates such as C. elegans, loss of this domain suggests that the function it imparts on HAP1_N domain proteins is not essential for C. elegans. Lack of a phenotype when T27A3.1 protein levels are knocked down by RNAi, indicates the T27A3.1 protein itself is not essential for C. elegans viability, at least in the absence of stressors . In contrast, the Drosophila homolog, Milton, is essential for larval development. It has maintained the Milt/TRAK domain and functions similarly to vertebrate TRAK1 and TRAK2 proteins, in mitochondrial trafficking .
To our knowledge, no protein interactions of Milton, TRAK1, or TRAK2 have been specifically attributed to the Milt/TRAK domain, and it remains to be determined how loss of the Milt/TRAK domain, shortening of the carboxyl terminus, and divergence of the HAP1_N domain have altered the function of eutherian HAP1 orthologues. Loss of the Milt/TRAK domain and carboxyl terminus may have affected the ancestral role in mitochondrial trafficking since Drosophila Milton interacts with mitochondria via sequences within its carboxyl terminal (aa847–1116)  and this terminus is lacked by eutherian HAP1 and T27A3.1. Milton also interacts with mitochondria indirectly through binding the mitochondrial protein Miro via sequence(s) within the first 750aa, spanning the HAP1_N and Milt/TRAK1 domains .
As described earlier, mouse TRAK1, TRAK2 and HAP1 have many common interactors and overlapping functions, largely related to receptor trafficking. However, mouse HAP1 possesses unique and essential functional aspects that are not fulfilled by its paralogs, since ablation of the mouse Hap1 gene leads to defects in mouse cellular models, and causes postnatal starvation in mice. Mouse HAP1 has pro-catabolic attributes (including supporting autophagosome dynamics , inhibiting mTOR signalling , and responding to starvation ) that are undescribed for TRAK proteins, whilst TRAK1 and TRAK2 have well established roles in mitochondrial trafficking which have not been described for HAP1. Whether these differences can be attributed to the presence or absence of the Milt/TRAK domain and/or carboxyl sequence remains to be explored.
Although not specifically a eutherian phenomenon, mammals are unique amongst vertebrates in that they endure an intense period of starvation between birth and maternal lactation, and upregulate catabolic processes such as autophagy in order to survive this transition . Mice deficient for autophagy gene Atg5, present with a similar postnatal phenotype to Hap1 knockout mice, with no milk visible in their stomachs, and death within 1 day of birth, despite minimal abnormalities at birth . In light of current understanding of HAP1 function, we propose that HAP1 may be important in promoting catabolic processes in periods of severe nutrient deficiency.
Insight into transcriptional regulation of HAP1 from gene promoter analyses
We utilized computational promoter analyses to gain insight into transcription factors and signaling pathways likely to regulate HAP1 transcriptional activity. We found the analyzed mammalian HAP1 gene promoters to be rich in sites for factors involved in the cell cycle, growth, and differentiation. The predominance of transcription factors involved in cell cycle control (MAX, WT1, E2F, E2F6, E2F7, LRRFIP and MAZR) suggests a key association of HAP1 with this process. HAP1 has been shown to have an inhibitory effect on the growth rate of breast cancer cell lines, and is under-expressed in breast cancer tissues . This negative effect of HAP1 on cell division is consistent with the inhibitory effect of HAP1 on the mTOR pathway that promotes biosynthesis and cell division . HAP1 may also play a direct role in mitosis by facilitating attachment of cargo to microtubules for correct positioning during the cell cycle via its interacting proteins that are involved in these processes, including p150glued , PCM1 [35, 24] and HTT . Other factors with motifs in the HAP1 gene promoter have roles in cell differentiation, growth and development (CDX1, CDX2, EGR1, NGFIC, GRHL1, NRF1 and ZNF219). These factors may act upstream of HAP1’s roles in postnatal development [27, 28], neurogenesis, neurotrophin receptor sorting , neuronal differentiation  and neuritic outgrowth [15, 59]. Binding sites in the mammalian HAP1 gene promoters that may be responsible for transcriptional activation in response to stressors include sites for Activating Transcription Factor 6 (ATF6) and Nuclear Respiratory Factor 1 (NRF1). ATF6 is a key transcriptional activator involved in the endoplasmic reticulum stress response . NRF1 activates transcription of key metabolic genes important for mitochondrial function may act in concert with peroxisome proliferator-activated receptor γ coactivator 1α (PGC1α), a transcription factor responsive to changes in energy balance .
TATA box motifs were present in all mammalian HAP1 gene promoters tested. This motif, named after its consensus DNA recognition sequence (5′ TATAAA), is found in approximately one third of vertebrate promoters . It is the binding site for TATA-binding protein (TBP), facilitating binding of the basal RNA polymerase II transcriptional complex for initiation of transcription. In eutherian HAP1 promoters, we identified a TATA box on the sense strand, starting 38 to 46 nucleotides upstream of the open reading frame (ORF), depending on the species. In both human and mouse the TATAAA motif was positioned between −29 and −24 with respect to the transcription start site (annotated by the Ensembl genome browser, data not shown), which is an appropriate distance for a functional TATA box. Marsupial Hap1 promoters (opossum and Tasmanian devil) contained two TATA box motifs in close proximity, further upstream from the ORF than in eutherians (the closest motif starting 197 and 238 nucleotides upstream of the ORF). Marsupial Hap1 promoters also lacked XCPE1 sites, that were detected on the reverse strand within the TATA box region of most eutherian promoters tested. XCPE1 sites are general core promoter elements that can enhance transcription activated by other factors (such as NRF1) in in vitro promoter assays .
Notably, sites for MAX, CDX1 and 2, and TBP were found to form a highly conserved promoter model common to all mammalian promoters tested. The model consisted of a TATA box on the sense strand, a site for CDX1 and CDX2 on the antisense strand overlapping the TATA box, and a MAX site shortly upstream. CDX1 and CDX2 are predominantly intestine-specific transcriptional factors involved in intestinal development, proliferation and differentiation, and are differentially expressed along the intestine and in the crypt/villus axis . Myc-associated factor, MAX is also present in the gut, and its gene expression increases along the crypt-villus axis in association with enterocyte maturation . The phylogenetic conservation of these sites in mammalian HAP1 gene promoters suggests expression of this gene may be regulated by CDX and MAX factors during the development and differentiation of enterocytes.
GO-term enrichment associated with other genes with this promoter model (investigated with and without inclusion of the MAX site) identified biological processes potentially regulated by this combination of factors. In addition to processes in which HAP1 has been implicated, such as microtubule-based movement , nervous system development , catabolism , calcium signalling , and inhibition of apoptosis , these included processes such as cell division, cell adhesion, purine metabolism and retinoic acid receptor signalling, representing novel processes in which HAP1 may be involved. The presence of functional CDX1 and CDX2 sites in close proximity of the TATA box have been previously described in other genes such as Sonic hedgehog (Shh) , glucose-6-phosphatase (G6PC) , calbindin-D9K  and clusterin . In the case of the G6PC gene (which is required for the catabolic processes of gluconeogenesis and glycogenolysis), transcription is activated by CDX1, and this activation is inhibited by CDX2, suggesting the existence of a competitive regulatory mechanism .
Interestingly, a human SNP (rs116737192) was found to occur at a critical nucleotide in the binding motifs for TBP, CDX1 and CDX2 in the HAP1 promoter. Individuals homozygous for the minor allele present at expected Mendelian frequency amongst African populations ( F. Tekola-Ayele, NIH, personal communication) indicating that the variant does not significantly compromise survival. Given the functional data from ENCODE ChIP-seq experiments indicating TBP (and RNA polymerase II) binds at this position , and our finding that the minor allele abolishes the TBP binding motif, this SNP could potentially alter HAP1 gene expression and act as a modifier of conditions with which HAP1 is associated.
Insight from localization of HAP1 in the human gut mucosa
Whilst HAP1 protein has been detected in the stomach and duodenum in mouse , localization to a specific cell type within the intestine has not previously been confirmed, and the presence of HAP1 in human intestine has not been explored. Here we demonstrated the presence of HAP1 protein (and thus HAP1 promoter activity) in singly dispersed mucosal cells in the human duodenum. Moreover, a subset of these cells were serotonin (5-HT)-containing EC cells, which also express CDX2 . Expression in EC cells is noteworthy, as not only is gut 5-HT important in regulating gastrointestinal functions such as the modulation of gut motility [73–75], gathering evidence supports a role of gut-derived 5-HT in metabolic regulation and energy balance [76–78]. Furthermore, EC cells are responsive to nutrients including glucose [79, 80], and increase their synthesis of 5-HT during periods of starvation , likely through a mechanism that senses reduced glucose availability . Interestingly, we found intracellular co-localization of HAP1 and 5-HT in EC cells occurred in punctate vesicles at the basolateral membrane, suggesting a role for HAP1 in the 5-HT secretory pathway. EC cells release vesicular 5-HT via a unique mechanism involving a small fusion pore [81, 82]. Whether HAP1 regulates this process of exocytosis as it does in other endocrine cells [25, 26] is currently unknown.
In this study we have identified eutherian HAP1 as an evolutionarily recent adaptation of a vertebrate TRAK protein-like ancestor. The function of the Milt/TRAK domain, how loss of this domain has affected HAP1 function since divergence from marsupials, and whether/how this provides a selective advantage in eutherians, particularly in the postnatal period, are questions remaining to be addressed. We have identified conserved regulatory elements amongst mammalian HAP1 gene promoters that provide insight into specific transcription factors and biological processes likely to act upstream of HAP1 gene transcription. Finally, prompted by the presence of a conserved promoter model including sites for intestinally expressed transcription factors (CDX1, CDX2) in the TATA box region of the promoter, we assessed HAP1 expression in the human small intestine. We detected the protein in singly dispersed mucosal cells, including within a subset of serotonin (5-HT) positive EC cells, with partial intracellular colocalization suggesting a role for HAP1 in peripheral 5-HT secretion.
Materials and methods
Gene information for synteny analysis was acquired from the Ensembl genome browser  first from version 82 and 83 and the National Center for Biotechnology Information (NCBI). HAP1 genes examined were human (H. sapiens ENSG00000173805), rat (R. norvegicus ENSRNOG00000014819), mouse (M. musculus ENSMUSG00000006930), dog (C. familiaris ENSCAFG00000015920), horse (E. caballus ENSECAG00000010380), rabbit (O. cuniculus ENSOCUG00000000605), opossum (M. domesticus ENSMODG00000014659), Tasmanian devil (S. harrisii ENSSHAG00000005125), zebrafish (D. rerio ENSDARG00000074508), chicken (G. gallus ENSGALG00000023847), anole lizard (A. carolinensis ENSACAG00000017899) and clawed frog (X. tropicalis ENSXETG00000012489). Protein sequences and domain predictions were obtained where possible from Ensembl, or otherwise from NCBI if a more recent build was available (opossum, Tasmanian devil, anole lizard). Protein sequence alignment and identity comparisons were conducted using the Clustal Omega (ClustalO) algorithm . Phylogenetic tree construction was performed using the MEGA6 software , by first aligning HAP1_N domains using ClustalW, then constructing the tree using the neighbor-joining method (Gonnet matrix). The tree was rooted by assigning the Drosophila Milton HAP1_N domain sequence as the outgroup, and clustering outcomes were tested by bootstrapping (1000 replicates). Protein sequence comparisons were performed using the following sequences (coordinates used for the HAP1_N domain phylogeny are indicated in Fig. 3 ): human (H. sapiens HAP1: ENSP00000334002.4, TRAK1: ENSP00000328998.5, TRAK2: ENSP00000328875.3), opossum (M. domesticus HAP1: XP_007482271.1, TRAK1: ENSMODP00000013909.4, TRAK2: ENSMODP00000018999.3), Tasmanian devil (S. harrisii HAP1: XP_012403736.1), rat (R. norvegicus HAP1: ENSRNOP00000072494.1), dog (C. familiaris HAP1: ENSCAFP00000023416.4), zebrafish (D. rerio HAP1: ENSDARP00000099219.3, TRAK1: ENSDARP00000115803.1, TRAK2: ENSDARP00000133065.1), clawed frog (X. tropicalis HAP1: ENSXETP00000027307.2, TRAK1: XP_012819699.1, TRAK2: ENSXETP00000061179.1), chicken (G. gallus HAP1: ENSGALP00000039222.2, TRAK1: ENSGALP00000019447.4, TRAK2: ENSGALP00000013618.4), anole lizard (A. carolinensis HAP1: XP_008111605.1, TRAK1: XP_008113765.1, TRAK2: ENSACAP00000001540.3), nematode (C. elegans T27A3.1: T27A3.1a.1), fruitfly (D. melanogaster Milton: FBpp0297338.1). Dictyostelium proteins with regions of similarity to human TRAK1 and Drosophila Milton were identified in proteins encoded by the Dictyostelium genome  by performing BLAST searches , via dictyBase , using a maximum E-value cut-off score of 0.1.
Identification of HAP1 promoter elements
Human, rat, mouse, dog and horse promoter sequences were acquired from Ensembl . For opossum and Tasmanian devil, only partial sequence was available from Ensembl, and the start codon was not shown so Hap1 mRNA sequences from NCBI (XM_007482209.1 and XM_012548282.1, respectively) were used to identify (by BLAST search  of genomic builds of the respective genomes) genomic contigs (NC_008802.1 and NW_003838835.1, respectively) from which sequence upstream of the start codon was obtained. The sequence from each species was then entered into the Genomatix software, and the ‘Common TFs’ tool of the Gene Regulation software suite was used to perform searches (of Matrix Library version 9.3) for individual transcription factor matrices (Vertebrate and General Core Promoter elements) common to all entries, using default stringency settings (Minimum Matrix Core similarity, 0.75. Minimum matrix similarity, ‘Optimized’). Match results were extracted and presented with all positions given relative to the translation start codon (Fig. 4a and Additional file 3).
Overrepresentation of biological processes amongst human genes containing the HAP1 promoter model
Promoter models were defined using the Genomatix FastM  tool. The model including CDX1, CDX2 and TBP only, was as follows: V$CDX1.02 (antisense), distance 0 bp, V$CDX2.02 (antisense), distance 2 bp, O$VTBP (sense). The model including MAX, CDX1, CDX2 and TBP was V$MAX.03 (both strands), distance 15–60 bp, V$CDX1.02 (antisense), distance 0 bp, V$CDX2.02 (antisense), distance 2 bp, O$VTBP (sense). Distances between elements represents the number of base pairs between the center of each consensus matrix for the site. The Genomatix ElDorado human gene promoter database (112705 sequences in June 2015 update) was searched for defined promoter models using the ModelInspector software , producing a gene list and an analysis of enriched Gene Ontology terms (p < 0.01, not adjusted for multiple testing in order to maintain sensitivity ) associated with the listed genes. REVIGO software was then used to provide a summarized representative subset of enriched GO terms using a clustering algorithm based on semantic similarity measures , and those with a p-value less than 0.001 were presented.
Duodenal biopsies were collected from 3 healthy subjects, aged 18–75 with no previous gastrointestinal disease and normal renal and kidney function. Subjects fasted overnight then attended the laboratory at 0800. A small diameter video endoscope (GIF-XP160, Olympus, Tokyo, Japan) was passed into the second part of the duodenum, and mucosal biopsies collected using standard biopsy forceps. These were immediately placed into 4 % paraformaldehyde for 2 h, cryoprotected (30 % sucrose in phosphate-buffered saline (PBS)), embedded in cryomolds, and frozen before sectioning at 6–10 μm (Cryocut 1800, Leica Biosystems, Nussloch, Germany) and thaw-mounting onto gelatin-coated slides. HAP1 immunoreactivity was detected using a mouse primary antibody (MA1-46412, 1:1000, ThermoFisher Scientific, Waltham, MA, USA) following antigen retrieval according to manufacturer’s instructions (S1700, Dako Australia, North Sydney, Australia). 5-hydroxy tryptamine (5-HT) was subsequently detected using a goat primary (20079, 1:2000, ImmunoStar, Wisconsin, USA). Immunoreactivity was visualized using species-specific secondary antibodies (1:200 in PBS-Tween 20) conjugated to Alexa Fluor dyes for HAP1 (A11029-AF488, ThermoFisher Scientific) and 5-HT (A11057-AF568, ThermoFisher Scientific). Nuclei were visualized using DAPI mounting media (E36935, ProLong Gold Antifade, Life Technologies, CA, USA).
Caudal-related homeobox factor 1
Caudal-related homeobox factor 2
Myc-associated factor X
Trafficking Kinesin protein 1
Trafficking Kinesin protein 2
Li XJ, Li SH, Sharp AH, Nucifora Jr FC, Schilling G, Lanahan A, Worley P, Snyder SH, Ross CA. A huntingtin-associated protein enriched in brain with implications for pathology. Nature. 1995;378(6555):398–402.
Sheng G, Chang GQ, Lin JY, Yu ZX, Fang ZH, Rong J, Lipton SA, Li SH, Tong G, Leibowitz SF, et al. Hypothalamic huntingtin-associated protein 1 as a mediator of feeding behavior. Nat Med. 2006;12(5):526–33.
Xiang J, Yan S, Li SH, Li XJ. Postnatal loss of hap1 reduces hippocampal neurogenesis and causes adult depressive-like behavior in mice. PLoS Genet. 2015;11(4):e1005175.
Mejia LA, Litterman N, Ikeuchi Y, de la Torre-Ubieta L, Bennett EJ, Zhang C, Harper JW, Bonni A. A novel Hap1-Tsc1 interaction regulates neuronal mTORC1 signaling and morphogenesis in the brain. J Neurosci. 2013;33(46):18015–21.
Yang GZ, Yang M, Lim Y, Lu JJ, Wang TH, Qi JG, Zhong JH, Zhou XF. Huntingtin associated protein 1 regulates trafficking of the amyloid precursor protein and modulates amyloid beta levels in neurons. J Neurochem. 2012;122(5):1010–22.
Sheng G, Xu X, Lin YF, Wang CE, Rong J, Cheng D, Peng J, Jiang X, Li SH, Li XJ. Huntingtin-associated protein 1 interacts with Ahi1 to regulate cerebellar and brainstem development in mice. J Clin Invest. 2008;118(8):2785–95.
Roux JC, Zala D, Panayotis N, Borges-Correia A, Saudou F, Villard L. Modification of Mecp2 dosage alters axonal transport through the Huntingtin/Hap1 pathway. Neurobiol Dis. 2012;45(2):786–95.
Zhu L, Song X, Tang J, Wu J, Ma R, Cao H, Ji M, Jing C, Wang Z: Huntingtin-associated protein 1: a potential biomarker of breast cancer. Oncol Rep. 2013;29(5):1881-7.
Li XJ, Sharp AH, Li SH, Dawson TM, Snyder SH, Ross CA. Huntingtin-associated protein (HAP1): discrete neuronal localizations in the brain resemble those of neuronal nitric oxide synthase. Proc Natl Acad Sci U S A. 1996;93(10):4839–44.
Liao M, Shen J, Zhang Y, Li SH, Li XJ, Li H. Immunohistochemical localization of huntingtin-associated protein 1 in endocrine system of the rat. J Histochem Cytochem. 2005;53(12):1517–24.
Dragatsis I, Dietrich P, Zeitlin S. Expression of the Huntingtin-associated protein 1 gene in the developing and adult mouse. Neurosci Lett. 2000;282(1–2):37–40.
Wong YC, Holzbaur EL. The regulation of autophagosome dynamics by huntingtin and HAP1 is disrupted by expression of mutant huntingtin, leading to defective cargo degradation. J Neurosci. 2014;34(4):1293–305.
Gutekunst CA, Li SH, Yi H, Ferrante RJ, Li XJ, Hersch SM. The cellular and subcellular localization of huntingtin-associated protein 1 (HAP1): comparison with huntingtin in rat and human. J Neurosci. 1998;18(19):7674–86.
Yang M, Lim Y, Li X, Zhong JH, Zhou XF. Precursor of brain-derived neurotrophic factor (proBDNF) forms a complex with Huntingtin-associated protein-1 (HAP1) and sortilin that modulates proBDNF trafficking, degradation, and processing. J Biol Chem. 2011;286(18):16272–84.
Rong J, McGuire JR, Fang ZH, Sheng G, Shin JY, Li SH, Li XJ. Regulation of intracellular trafficking of huntingtin-associated protein-1 is critical for TrkA protein levels and neurite outgrowth. J Neurosci. 2006;26(22):6019–30.
Block-Galarza J, Chase KO, Sapp E, Vaughn KT, Vallee RB, DiFiglia M, Aronin N. Fast transport and retrograde movement of huntingtin and HAP 1 in axons. Neuroreport. 1997;8(9–10):2247–51.
Xiang J, Yang H, Zhao T, Sun M, Xu X, Zhou XF, Li SH, Li XJ. Huntingtin-associated protein 1 regulates postnatal neurogenesis and neurotrophin receptor sorting. J Clin Invest. 2014;124(1):85–98.
Ma B, Savas JN, Yu MS, Culver BP, Chao MV, Tanese N. Huntingtin mediates dendritic transport of beta-actin mRNA in rat neurons. Sci Rep. 2011;1:140.
Mackenzie KD, Lumsden AL, Guo F, Duffield MD, Chataway T, Lim Y, Zhou XF, Keating DJ. Huntingtin-associated protein-1 is a synapsin I-binding protein regulating synaptic vesicle exocytosis and synapsin I trafficking. J Neurochem. 2016;138(5):710–21.
Mandal M, Wei J, Zhong P, Cheng J, Duffney LJ, Liu W, Yuen EY, Twelvetrees AE, Li S, Li XJ, Kittler JT, Yan Z. Impaired Â-Amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor trafficking and function by mutant huntingtin. J. Biol. Chem. 2011;286(39):33719–28.
Li Y, Chin LS, Levey AI, Li L. Huntingtin-associated protein 1 interacts with hepatocyte growth factor-regulated tyrosine kinase substrate and functions in endosomal trafficking. J Biol Chem. 2002;277(31):28212–21.
Kittler JT, Thomas P, Tretter V, Bogdanov YD, Haucke V, Smart TG, Moss SJ. Huntingtin-associated protein 1 regulates inhibitory synaptic transmission by modulating Â -aminobutyric acid type A receptor membranetrafficking. Proc. Natl. Acad. Sci. 2004;101(34):12736-41.
Gauthier LR, Charrin BC, Borrell-Pages M, Dompierre JP, Rangone H, Cordelieres FP, De Mey J, MacDonald ME, Lessmann V, Humbert S, et al. Huntingtin controls neurotrophic support and survival of neurons by enhancing BDNF vesicular transport along microtubules. Cell. 2004;118(1):127–38.
Keryer G, Pineda JR, Liot G, Kim J, Dietrich P, Benstaali C, Smith K, Cordelieres FP, Spassky N, Ferrante RJ, et al. Ciliogenesis is regulated by a huntingtin-HAP1-PCM1 pathway and is altered in Huntington disease. J Clin Invest. 2011;121(11):4372–82.
Cape A, Chen X, Wang CE, O’Neill A, Lin YF, He J, Xu XS, Yi H, Li H, Li S, et al. Loss of huntingtin-associated protein 1 impairs insulin secretion from pancreatic beta-cells. Cell Mol Life Sci. 2012;69(8):1305–17.
Mackenzie KD, Duffield MD, Peiris H, Phillips L, Zanin MP, Teo EH, Zhou XF, Keating DJ. Huntingtin-associated protein 1 regulates exocytosis, vesicle docking, readily releasable pool size and fusion pore stability in mouse chromaffin cells. J Physiol. 2014;592(Pt 7):1505–18.
Chan EY, Nasir J, Gutekunst CA, Coleman S, Maclean A, Maas A, Metzler M, Gertsenstein M, Ross CA, Nagy A, et al. Targeted disruption of Huntingtin-associated protein-1 (Hap1) results in postnatal death due to depressed feeding behavior. Hum Mol Genet. 2002;11(8):945–59.
Li SH, Yu ZX, Li CL, Nguyen HP, Zhou YX, Deng C, Li XJ. Lack of huntingtin-associated protein-1 causes neuronal death resembling hypothalamic degeneration in Huntington’s disease. J Neurosci. 2003;23(17):6956–64.
Dragatsis I, Zeitlin S, Dietrich P. Huntingtin-associated protein 1 (Hap1) mutant mice bypassing the early postnatal lethality are neuroanatomically normal and fertile but display growth retardation. Hum Mol Genet. 2004;13(24):3115–25.
Lin YF, Xu X, Cape A, Li S, Li XJ. Huntingtin-associated protein-1 deficiency in orexin-producing neurons impairs neuronal process extension and leads to abnormal behavior in mice. J Biol Chem. 2010;285(21):15941–9.
Mercer KB, Szlam SM, Manning E, Gernert KM, Walthall WW, Benian GM, Gutekunst CA. A C. elegans homolog of huntingtin-associated protein 1 is expressed in chemosensory neurons and in a number of other somatic cell types. J Mol Neurosci. 2009;37(1):37–49.
Stowers RS, Megeath LJ, Gorska-Andrzejak J, Meinertzhagen IA, Schwarz TL. Axonal transport of mitochondria to synapses depends on milton, a novel Drosophila protein. Neuron. 2002;36(6):1063–77.
Li SH, Hosseini SH, Gutekunst CA, Hersch SM, Ferrante RJ, Li XJ. A human HAP1 homologue. Cloning, expression, and interaction with huntingtin. J Biol Chem. 1998;273(30):19220–7.
Fiumara F, Fioriti L, Kandel ER, Hendrickson WA. Essential role of coiled coils for aggregation and activity of Q/N-rich prions and PolyQ proteins. Cell. 2010;143(7):1121–35.
Engelender S, Sharp AH, Colomer V, Tokito MK, Lanahan A, Worley P, Holzbaur EL, Ross CA. Huntingtin-associated protein 1 (HAP1) interacts with the p150Glued subunit of dynactin. Hum Mol Genet. 1997;6(13):2205–12.
Twelvetrees AE, Yuen EY, Arancibia-Carcamo IL, MacAskill AF, Rostaing P, Lumb MJ, Humbert S, Triller A, Saudou F, Yan Z, et al. Delivery of GABAARs to synapses is mediated by HAP1-KIF5 and disrupted by mutant huntingtin. Neuron. 2010;65(1):53–65.
McGuire JR, Rong J, Li SH, Li XJ. Interaction of Huntingtin-associated protein-1 with kinesin light chain: implications in intracellular trafficking in neurons. J Biol Chem. 2006;281(6):3552–9.
Li SH, Gutekunst CA, Hersch SM, Li XJ. Association of HAP1 isoforms with a unique cytoplasmic structure. J Neurochem. 1998;71(5):2178–85.
Webber E, Li L, Chin LS. Hypertonia-associated protein Trak1 is a novel regulator of endosome-to-lysosome trafficking. J Mol Biol. 2008;382(3):638–51.
Beck M, Brickley K, Wilkinson HL, Sharma S, Smith M, Chazot PL, Pollard S, Stephenson FA. Identification, molecular cloning, and characterization of a novel GABAA receptor-associated protein, GRIF-1. J Biol Chem. 2002;277(33):30079–90.
Stephenson FA. Revisiting the TRAK family of proteins as mediators of GABAA receptor trafficking. Neurochem Res. 2014;39(6):992–6.
Brickley K, Smith MJ, Beck M, Stephenson FA. GRIF-1 and OIP106, members of a novel gene family of coiled-coil domain proteins: association in vivo and in vitro with kinesin. J Biol Chem. 2005;280(15):14723–32.
Glater EE, Megeath LJ, Stowers RS, Schwarz TL. Axonal transport of mitochondria requires milton to recruit kinesin heavy chain and is light chain independent. J Cell Biol. 2006;173(4):545–57.
van Spronsen M, Mikhaylova M, Lipka J, Schlager MA, van den Heuvel DJ, Kuijpers M, Wulf PS, Keijzer N, Demmers J, Kapitein LC, et al. TRAK/Milton motor-adaptor proteins steer mitochondrial trafficking to axons and dendrites. Neuron. 2013;77(3):485–502.
Koutsopoulos OS, Laine D, Osellame L, Chudakov DM, Parton RG, Frazier AE, Ryan MT. Human Miltons associate with mitochondria and induce microtubule-dependent remodeling of mitochondrial networks. Biochim Biophys Acta. 2010;1803(5):564–74.
Brickley K, Pozo K, Stephenson FA. N-acetylglucosamine transferase is an integral component of a kinesin-directed mitochondrial trafficking complex. Biochim Biophys Acta. 2011;1813(1):269–81.
Yates A, Akanni W, Amode MR, Barrell D, Billis K, Carvalho-Silva D, Cummins C, Clapham P, Fitzgerald S, Gil L, et al. Ensembl 2016. Nucleic Acids Res. 2016;44(D1):D710–6.
Marchler-Bauer A, Derbyshire MK, Gonzales NR, Lu S, Chitsaz F, Geer LY, Geer RC, He J, Gwadz M, Hurwitz DI, et al. CDD: NCBI’s conserved domain database. Nucleic Acids Res. 2015;43(Database issue):D222–6.
Eichinger L, Pachebat JA, Glockner G, Rajandream MA, Sucgang R, Berriman M, Song J, Olsen R, Szafranski K, Xu Q, et al. The genome of the social amoeba Dictyostelium discoideum. Nature. 2005;435(7038):43–57.
Cartharius K, Frech K, Grote K, Klocke B, Haltmeier M, Klingenhoff A, Frisch M, Bayerlein M, Werner T. MatInspector and beyond: promoter analysis based on transcription factor binding sites. Bioinformatics. 2005;21(13):2933–42.
Consortium EP. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012;489(7414):57–74.
Genomes Project C, Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, Korbel JO, Marchini JL, McCarthy S, McVean GA, et al. A global reference for human genetic variation. Nature. 2015;526(7571):68–74.
Klingenhoff A, Frech K, Quandt K, Werner T. Functional promoter modules can be detected by formal models independent of overall nucleotide sequence similarity. Bioinformatics. 1999;15(3):180–6.
Gershon MD, Tack J. The serotonin signaling system: from basic understanding to drug development for functional GI disorders. Gastroenterology. 2007;132(1):397–414.
La Rosa S, Rigoli E, Uccella S, Chiaravalli AM, Capella C. CDX2 as a marker of intestinal EC-cells and related well-differentiated endocrine tumors. Virchows Arch. 2004;445(3):248–54.
Kuma A, Hatano M, Matsui M, Yamamoto A, Nakaya H, Yoshimori T, Ohsumi Y, Tokuhisa T, Mizushima N. The role of autophagy during the early neonatal starvation period. Nature. 2004;432(7020):1032–6.
Godin JD, Colombo K, Molina-Calavita M, Keryer G, Zala D, Charrin BC, Dietrich P, Volvert ML, Guillemot F, Dragatsis I, et al. Huntingtin is required for mitotic spindle orientation and mammalian neurogenesis. Neuron. 2010;67(3):392–406.
Huang PT, Chen CH, Hsu IU, Salim SA, Kao SH, Cheng CW, Lai CH, Lee CF, Lin YF. Huntingtin-associated protein 1 interacts with breakpoint cluster region protein to regulate neuronal differentiation. PLoS One. 2015;10(2):e0116372.
Rong J, Li S, Sheng G, Wu M, Coblitz B, Li M, Fu H, Li XJ. 14-3-3 protein interacts with Huntingtin-associated protein 1 and regulates its trafficking. J Biol Chem. 2007;282(7):4748–56.
Yoshida H, Haze K, Yanagi H, Yura T, Mori K. Identification of the cis-acting endoplasmic reticulum stress response element responsible for transcriptional induction of mammalian glucose-regulated proteins. Involvement of basic leucine zipper transcription factors. J Biol Chem. 1998;273(50):33741–9.
Scarpulla RC. Nuclear activators and coactivators in mammalian mitochondrial biogenesis. Biochim Biophys Acta. 2002;1576(1–2):1–14.
Juven-Gershon T, Hsu JY, Theisen JW, Kadonaga JT. The RNA polymerase II core promoter - the gateway to transcription. Curr Opin Cell Biol. 2008;20(3):253–9.
Tokusumi Y, Ma Y, Song X, Jacobson RH, Takada S. The new core promoter element XCPE1 (X Core Promoter Element 1) directs activator-, mediator-, and TATA-binding protein-dependent but TFIID-independent RNA polymerase II transcription from TATA-less promoters. Mol Cell Biol. 2007;27(5):1844–58.
Silberg DG, Swain GP, Suh ER, Traber PG. Cdx1 and cdx2 expression during intestinal development. Gastroenterology. 2000;119(4):961–71.
Mariadason JM, Nicholas C, L’Italien KE, Zhuang M, Smartt HJ, Heerdt BG, Yang W, Corner GA, Wilson AJ, Klampfer L, et al. Gene expression profiling of intestinal epithelial cell maturation along the crypt-villus axis. Gastroenterology. 2005;128(4):1081–8.
Tang TS, Tu H, Chan EY, Maximov A, Wang Z, Wellington CL, Hayden MR, Bezprozvanny I. Huntingtin and huntingtin-associated protein 1 influence neuronal calcium signaling mediated by inositol-(1,4,5) triphosphate receptor type 1. Neuron. 2003;39(2):227–39.
Takeshita Y, Fujinaga R, Zhao C, Yanai A, Shinoda K. Huntingtin-associated protein 1 (HAP1) interacts with androgen receptor (AR) and suppresses SBMA-mutant-AR-induced apoptosis. Hum Mol Genet. 2006;15(15):2298–312.
Mutoh H, Hayakawa H, Sashikawa M, Sakamoto H, Sugano K. Direct repression of Sonic Hedgehog expression in the stomach by Cdx2 leads to intestinal transformation. Biochem J. 2010;427(3):423–34.
Gautier-Stein A, Domon-Dell C, Calon A, Bady I, Freund JN, Mithieux G, Rajas F. Differential regulation of the glucose-6-phosphatase TATA box by intestine-specific homeodomain proteins CDX1 and CDX2. Nucleic Acids Res. 2003;31(18):5238–46.
Barley NF, Prathalingam SR, Zhi P, Legon S, Howard A, Walters JR. Factors involved in the duodenal expression of the human calbindin-D9k gene. Biochem J. 1999;341(Pt 3):491–500.
Suh E, Wang Z, Swain GP, Tenniswood M, Traber PG. Clusterin gene transcription is activated by caudal-related homeobox genes in intestinal epithelium. Am J Physiol Gastrointest Liver Physiol. 2001;280(1):G149–56.
Tekola-Ayele F, Doumatey AP, Shriner D, Bentley AR, Chen G, Zhou J, Fasanmade O, Johnson T, Oli J, Okafor G, et al. Genome-wide association study identifies African-ancestry specific variants for metabolic syndrome. Mol Genet Metab. 2015;116(4):305–13.
Spencer NJ, Nicholas SJ, Robinson L, Kyloh M, Flack N, Brookes SJ, Zagorodnyuk VP, Keating DJ. Mechanisms underlying distension-evoked peristalsis in guinea pig distal colon: is there a role for enterochromaffin cells? Am J Physiol Gastrointest Liver Physiol. 2011;301(3):G519–27.
Keating DJ, Spencer NJ. Release of 5-hydroxytryptamine from the mucosa is not required for the generation or propagation of colonic migrating motor complexes. Gastroenterology. 2010;138(2):659–70. 670 e651-652.
Spencer NJ, Sia TC, Brookes SJ, Costa M, Keating DJ. CrossTalk opposing view: 5-HT is not necessary for peristalsis. J Physiol. 2015;593(15):3229–31.
Crane JD, Palanivel R, Mottillo EP, Bujak AL, Wang H, Ford RJ, Collins A, Blumer RM, Fullerton MD, Yabut JM, et al. Inhibiting peripheral serotonin synthesis reduces obesity and metabolic dysfunction by promoting brown adipose tissue thermogenesis. Nat Med. 2015;21(2):166–72.
Sumara G, Sumara O, Kim JK, Karsenty G. Gut-derived serotonin is a multifunctional determinant to fasting adaptation. Cell Metab. 2012;16(5):588–600.
Young RL, Lumsden AL, Keating DJ. Gut serotonin is a regulator of obesity and metabolism. Gastroenterology. 2015;149(1):253–5.
Symonds EL, Peiris M, Page AJ, Chia B, Dogra H, Masding A, Galanakis V, Atiba M, Bulmer D, Young RL, et al. Mechanisms of activation of mouse and human enteroendocrine cells by nutrients. Gut. 2015;64(4):618–26.
Zelkas L, Raghupathi R, Lumsden AL, Martin AM, Sun E, Spencer NJ, Young RL, Keating DJ. Serotonin-secreting enteroendocrine cells respond via diverse mechanisms to acute and chronic changes in glucose availability. Nutr Metab. 2015;12:55.
Raghupathi R, Jessup CF, Lumsden AL, Keating DJ. Fusion pore size limits 5-HT release from single enterochromaffin cell vesicles. J Cell Physiol. 2016;231(7):1593–600.
Raghupathi R, Duffield MD, Zelkas L, Meedeniya A, Brookes SJ, Sia TC, Wattchow DA, Spencer NJ, Keating DJ. Identification of unique release kinetics of serotonin from guinea-pig and human enterochromaffin cells. J Physiol. 2013;591(Pt 23):5959–75.
Cunningham F, Amode MR, Barrell D, Beal K, Billis K, Brent S, Carvalho-Silva D, Clapham P, Coates G, Fitzgerald S, et al. Ensembl 2015. Nucleic Acids Res. 2015;43(Database issue):D662–9.
Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Soding J, et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol. 2011;7:539.
Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6: molecular evolutionary genetics analysis version 6.0. Mol Biol Evol. 2013;30(12):2725–9.
Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 1997;25(17):3389–402.
Vize P, Westerfiled M, Basu S, Fey P, Jimenez-Morales D, Dodson RJ, Chisholm RL. dictyBase 2015: Expanding data and annotations in a new software environment. Genesis. 2015;53(8):523–34.
da Huang W, Sherman BT, Lempicki RA. Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res. 2009;37(1):1–13.
Supek F, Bosnjak M, Skunca N, Smuc T. REVIGO summarizes and visualizes long lists of gene ontology terms. PLoS One. 2011;6(7):e21800.
The authors gratefully acknowledge Professor Rory Hope for reading the manuscript and providing insightful comments, and Professor Briony Forbes, Dr Nicole Thompson and Dr Fasil Tekola-Ayele for helpful suggestions in the course of the study.
This work was supported by funding to AL from the Centre for Neuroscience, and the Faculty of Medicine, Nursing and Health Sciences at Flinders University, enabling access to Genomatix licensed software. DK is supported by fellowship funding from the Australian National Health and Medical Research Council (APP1088737). The funding bodies had no role in the design of the study, in the collection, analysis or interpretation of data, or in the writing of the manuscript.
Availability of data and materials
The datasets supporting the conclusions of this article are included within the article and its additional files.
Conception and design of the study: AL. Phlylogenetic and computational promoter data collection and analysis: AL. Acquisition of patient samples for immunohistochemistry: RY. Immunohistochemistry and imaging: NP. Manuscript drafting: AL. Manuscript revision for critical intellectual content: RY, DK. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
Subjects who provided the tissue used for immunohistochemistry provided written informed consent, and the research protocol was approved by the Human Research Ethics Committees of the Royal Adelaide Hospital and The University of Adelaide (R20131119), and conducted in accordance with the Declaration of Helsinki as revised in 2000.
Identity matrix showing pairwise percentage amino acid sequence identity between HAP1 family proteins: TRAK1 and TRAK2 (human, opossum, zebrafish, Xenopus, anole lizard, chicken), HAP1 (species as above, plus rat, dog, and Tasmanian devil), Drosophila (fruitfly) Milton, and C. elegans (nematode) T27A3.1. Sequence identifiers are as indicated in Materials and methods. (JPG 1967 kb)
Sequence alignment of HAP1 family proteins. ClustalO alignment of HAP1 (human, rat, dog), TRAK1 (human, rat, dog), TRAK2 (human, rat, dog), T27A3.1 (C. elegans) and Milton (Drosophila) proteins is shown. The red boxed region indicates the HAP1_N domain, and the purple boxed region indicates the Milt/TRAK domain. Amino acid color identifies residues that are small/hydrophobic/aromatic (red), acidic (blue) basic (magenta) and hydroxyl/sulfhydryl/amine/glycine (green). Degree of conservation is indicated by an asterisk (*) for complete agreement, a colon (:) for conservation between groups of strongly similar properties, and a dot (.) for conservation between groups of weakly similar properties. Sequence identifiers are as indicated in Materials and methods. (PPTX 3364 kb)
HAP1 gene promoter element match positions and scores. Excel spreadsheet containing sequence positions (relative to the start ATG codon) and match scores of all conserved promoter elements described in this study. (XLSX 69 kb)
About this article
Cite this article
Lumsden, A.L., Young, R.L., Pezos, N. et al. Huntingtin-associated protein 1: Eutherian adaptation from a TRAK-like protein, conserved gene promoter elements, and localization in the human intestine. BMC Evol Biol 16, 214 (2016). https://0-doi-org.brum.beds.ac.uk/10.1186/s12862-016-0780-3
- TATA-binding protein
- Myc-associated factor X