Phylogenetic Analysis of the ING Family of PHD Finger Proteins
http://www.100md.com
《分子生物学进展》
* Department of Biochemistry and Molecular Biology, Health Science Complex, The University of Calgary, Calgary, Alberta, Canada; and Department of Biochemistry and Microbiology, University of Victoria, Victoria, British Colombia, Canada
Correspondence: E-mail: karl@ucalgary.ca.
Abstract
Since the discovery of ING1 class II tumor suppressors in 1996, five different ING genes (ING1 to ING5) encoding proteins with highly conserved plant homeodomain (PHD) motifs and several splicing isoforms of the ING1 and ING2 gene have been identified. The ING family functions in DNA repair and apoptosis in response to UV damage through binding to proliferating cell nuclear antigen (PCNA); chromatin remodeling and regulation of gene expression through regulating and/or targeting histone acetyltransferase/deacetylase (HAT/HDAC) activities; binding targets of rare phosphatidylinositol phosphates (PtdInsPs) that function in DNA damage-initiated stress signaling; and regulating brain tumor angiogenesis through transcriptional repression of NF-KB–responsive genes. To elucidate the evolutionary history of ING proteins and summarize what is known about regions highly conserved in the ING family members, we have examined the sequences and phylogenetic relationships of ING proteins across taxonomically diverse organisms. We have identified novel ING family members in rats, frogs, fish, mosquitoes, fruit flies, worms, fungi, and plants. We have also clarified the naming and classification of ING proteins based on our phylogenetic analysis to allow better understanding of the ING protein family. Using sequence similarities, we have identified novel regions and motifs of unknown function that are conserved across family members. An evolutionary history for the ING family of PHD finger proteins is presented that indicates that five ING genes are present in vertebrates. Three of these may be paralogs of ING genes found in arthropods, whereas nematodes, fungi, and green plants contain ING genes that have general features of the vertebrate ING family.
Key Words: ING1 ? PHD finger ? molecular evolution ? protein phylogeny
Introduction
Using subtractive hybridization between cDNAs from a normal mammary cell line and several transformed breast cancer epithelial cell lines, the first member of the ING gene family was identified by the subsequent selection for sense or antisense cDNA fragments capable of promoting neoplastic transformation (Garkavtsev et al. 1996). Suppression of ING1 promotes focus formation and growth in soft agar in vitro and tumor formation in vivo (Garkavtsev et al. 1996). ING1 expression is significantly repressed in more than 40% of human primary breast cancers, 100% of established breast cancer cell lines (Toyama et al. 1999), and many other forms of blood and solid tumors (Ohmori et al. 1999; Oki et al. 1999; Shinoura et al. 1999; Tokunaga et al. 2000; Chen et al. 2001; Krishnamurthy et al. 2001; Bromidge and Lynas 2002; Gunduz et al. 2002; Ito et al. 2002; Noumann et al. 2002; Chen et al. 2003; Hara et al. 2003; Garkavtsev et al. 2004; Tallen et al. 2004). Since the initial description of ING1, studies from diverse groups using different model systems have implicated the ING proteins in control of cellular aging (Garkavtsev and Riabowol 1997), negative regulation of cell proliferation (Garkavtsev et al. 1996; Nagashima et al. 2001), apoptosis (Helbing et al. 1997), DNA damage repair (Scott et al. 2001a; Cheung et al. 2001), chromatin remodeling (Loewith et al. 2000; Vieyra et al. 2002; Kuzmichev et al. 2002), hormone responses (Wagner et al. 2001; Toyama et al. 2003), and regulation of brain tumor growth and angiogenesis (Garkavtsev et al. 2004).
The human ING1 gene has three exons that can be alternatively spliced onto a common 3' exon, thereby generating p27ING1d, p33ING1b, and p47ING1a. In addition, internal initiation at an ATG within the common exon generates p24ING1c (Garkavtsev et al. 1999; Jager et al. 1999; Gunduz et al. 2000; Saito et al. 2000). By screening a breast cancer cell cDNA library using ING1 as probe, followed by 5' RACE using normal fetus cDNA, Jager et al. (1999) cloned an ING1 homolog gene, which maps to the X chromosome. The deduced 42–amino acid peptide has a calculated molecular mass of 5.0 kDa and shares 76% identity over its length with the much longer ING1 protein. Expression of this gene was detected in all normal tissues tested and in some cancer cell lines using RT-PCR. To date, four additional genes have been identified in humans that encode ING2 to ING5 (Nagashima et al. 2001, 2003; Shiseki et al. 2003). Study of amphibian ING2 genes provides evidence of differential regulation of presumed splice variants of this gene (Wagner et al. 2001). Three forms of ING2 have been cloned in Xenopus laevis; however, whether they are actually splicing isoforms remains to be confirmed.
All ING genes share strongest homology in a region encoding a plant homeodomain (PHD) motif that has been implicated in the regulation of ubiquitination (Coscoy and Ganem 2003), although this function remains controversial (Aravind, Iyer, Koonin 2003; Scheel and Hofmann 2003). More recently, ING PHDs have been identified as binding targets of rare phosphatidylinositol phosphates (PtdInsPs) that function in DNA damage–initiated stress signaling (Gozani et al. 2003). ING1 also translocates to the nucleolus after UV-induced DNA damage (Scott et al. 2001b). This property of UV-induced relocation to nucleoli is shared with several other proteins that may play roles in DNA repair and/or apoptosis, such as nucleolin (Mi et al. 2003), the Werner's syndrome helicase (Leung and Lamond 2003), the CK2 kinase that phosphorylates DNA ligase and topoisomerase 2 (Gerber et al. 2000), hRad17 (Chang et al. 1999), and p53/MDM2/ARF, among others (Klibanov, O'Hagan, and Ljungman 2001; Leung and Lamond 2003; Sugimoto et al. 2003). Mutational studies identified two intrinsic nucleolar translocation sequences (NTS) within the nuclear localization signal (NLS) found in all ING family members (Scott et al. 2001b). p33ING1b also binds to the proliferating cell nuclear antigen (PCNA) through a specific sequence called the PCNA-interacting protein (PIP) domain (Warbrick 1998), which is found in proteins involved in growth inhibition (e.g., p21WAF1), growth arrest after DNA damage (e.g., GADD45), and DNA replication and repair (e.g., FEN1) (Feng, Hara, and Riabowol 2002). The ability of ING proteins to bind to and alter the activity of histone acetyltransferases (HATs), histone deacetylases (HDACs), and factor acetyltransferases (FATs) has also been shown (Loewith et al. 2000; Skowyra et al. 2001; Nagashima et al. 2001; Vieyra et al. 2002; Kuzmichev et al. 2002) (reviewed in Feng, Hara, and Riabowol [2002]).
ING sequences have been reported in human, mouse, rat, frog, fission and budding yeast, Drosophila, and C. elegans, as well as in other species; however, most of these homologs remain unrecognized and uncharacterized in databases, including those of the NCBI. Despite the increasing number of ING homologs that have been identified in different organisms, a comprehensive analysis of the evolution of this protein family has not yet been conducted. No studies have reported on the phylogenetic relationships and evolutionary history of the ING family. Because of their potentially important roles in many central biological processes, a better understanding of the multiple ING family members is of particular research interest, and a comprehensive analysis of the ING family across taxonomically diverse organisms would be useful for future studies.
Evolutionary studies of a protein family can be instrumental in determining structurally conserved regions, leading to useful predictions of protein function (Eisen 1998). Moreover, an examination of the evolutionary history of the protein family may indicate the presence of novel family members. In this study, we have undertaken a comprehensive analysis of publicly accessible databases and selected unpublished sequences to identify as many members of the ING family as possible, to further characterize ING-specific motifs and to present an analysis of the evolutionary relationship among candidate ING members from various species.
Materials and Methods
Sequence Searches and Mapping Data Retrieval
Extensive word-based searches for ING/ING-like proteins deposited in GenBank (http://www.ncbi.nlm.nih.gov/) were performed using appropriate word combinations, including p33ING1b, p47, ING1-like, growth inhibitor, apoptosis, and PHD, among others. Blast (Altschul et al. 1997) searches were performed at the NCBI Web site (http://www.ncbi.nlm.nih.gov/BLAST/) on all databases available as of March 20, 2004. Additional ING sequences were obtained from work in preparation, and, therefore, are currently not in the public databases. The origin of these sequences is indicated in table 1. Redundant sequences were excluded. All protein sequences were collected in FASTA format for further analysis. Gene loci and mapping data were retrieved from NCBI Human Genome Resources (http://www.ncbi.nlm.nih.gov/genome/guide/human/).
Table 1 List of ING Proteins Used for Amino Acid Sequence Alignment
Naming of Sequences
In this study we denoted all family members of inhibitor of growth proteins as ING followed by the abbreviation of genus and species name (e.g., Mus musculus ING was denoted ING_Mm). Only a small portion of the proteins, which are identified based on sequence similarity, have been functionally characterized, whereas many others are still described as putative/hypothetical proteins. Based on the results of our phylogenetic analysis, the sequences were renamed according to their phylogenetic distance to the five relatively well studied human counterparts and, thus, were assigned different numbers accordingly (i.e., 1 to 5). A large group of evolutionarily distant proteins was categorized more generally as ING with designation of , ?, , and to differentiate them within a species. These designations do not imply similarity across species, as do the ING1 to ING5 designations. The lowercase letters (i.e., a to d) denote the splicing isoform of the same gene followed by an Arabic number indicating the transcript variant. Note that we did not distinguish the functionally characterized ING proteins from the uncharacterized ones in the naming. Originally, the deduced 42–amino acid peptide (Jager et al. 1999) was characterized as an ING1-like tumor suppressor protein. However, because this gene is uniquely mapped to the X chromosome and the deduced peptide is highly truncated, we renamed it as INGX_Hs to indicate its distinct characteristic. In the case of Xenopus laevis, there is evidence for the existence of two ING1b genes that have presumably arisen as a result of genome duplication, giving rise to a pseudotetraploidy state (Wagner and Helbing 2004). Each of these genes has multiple transcript variants that are designated as ING1b1, ING1b2, and so on. Variants 1 to 4 differ in the 5' untranslated region but give rise to the same putative protein. This protein is designated as ING1b1-4_Xl. A second putative protein that is derived from the duplicated gene is referred to as ING1b5.
Multiple Sequence Alignments
All amino acid alignments were performed using the program T_COFFEE (Notredame, Higgins, and Heringa 2000) with default settings. The alignments were then adjusted and shaded using the multiple sequence alignment editor GENEDOC (Nicholas, Nicholas, and Deerfield 1997).
Phylogenetic Analysis
We used PHYLIP version 3.6a3 (Felsenstein 1989) for our phylogenetic analysis. Both distance and parsimony analyses using the protein alignment as input were performed. Bootstrap values were obtained using SEQBOOT and creating 1,000 delete-half Jackknife data sets. The distance analysis was performed by using PROTDIST and subsequently NEIGHBOR with standard parameters, and the parsimony analysis was performed using PROTPARS with standard parameters. In both cases, the "M" option for the analysis of the multiple data sets created with SEQBOOT was invoked.
Results
Location of ING Genes in the Human Genome
The chromosomal locations of the five human ING genes have been mapped to five different chromosomes noted in figure 1. ING1 to ING5 are located on chromosome bands 13q34, 4q35, 7q31, 12p13.3, and 2q37.3, respectively. ING1, ING2, ING4, and ING5 are in regions very close to the ends of their respective chromosome where the telomere/subtelome region is located, approximately 4, 6.6, 6.6, and 0.4 Mbp from chromosomal terminus, respectively. It is interesting to note that, being only about 440 kbp from the telomere region, ING5 has the potential to be affected by telomere erosion. In contrast, ING3 is located in the middle of the long arm of chromosome 7. The additional, but highly truncated, ING-like sequence, INGX, has been mapped to chromosome Xq12, about 11 Mbp from the centromeric region. Selected genes flanking the INGs are shown in figure 1, including many functionally important genes, such as RAB20, a member of the RAS oncogene family; CASP3, encoding caspase 3 cysteine protease; ACRBP, the acrosin-binding protein; BOK, a BCL2-related ovarian killer; and IL2RG, the interleukin 2 receptor-. Coordinate regulation of these gene clusters has not yet been reported.
FIG. 1.— Chromosomal localization of the inhibitor of growth (ING) genes in Homo sapiens. Chromosome number and the approximate chromosomal length (in Mbp) are indicated. Cytogenetic positions of the ING genes are shown. The chromosomal localization of INGs and several flanking genes are listed with the approximate chromosomal start positions in kbp. The distance to telomeres was estimated by calculating the distance between the ING positions to the end of the chromosome, where the telomere/subtelomere region is located. In the case of INGX, the distance to the centromere was calculated. The centromeric regions are indicated as constrictions. The gene positions and the identities of the flanking genes are available at http://www.ncbi.nlm.nih.gov/genome/guide/human/.
Structural Features of the ING Proteins
The amino acid sequence alignment of human ING proteins revealed several conserved regions (fig. 2A). The PHD is best conserved and contains a C4-H-C3 zinc-finger motif located near the C-terminus. INGX, which is 42 amino acids long, has only a partial PHD (fig. 2B). A small conserved protein-interacting motif (PIM) of ING1 and ING2 is enriched in acidic, basic, and aromatic residues. It binds a defined subset of peptides (Feng et al. 2004) and, along with the PHD region (Gozani et al. 2003), binds to phosphatidylinositol monophosphates. The corresponding C-terminal regions of ING3 to ING5 are also rich in basic residues but are considerably shorter than those of ING1 and ING 2. Towards the N-terminus, a conserved REASP motif found on ING1 and ING2 resembles a phosphorylation-dependent interacting motif (PDIM) similar to the canonical RSXpSXP 14-3-3 binding motif. The nuclear localization sequence (NLS) is a strong basic region located upstream of the PHD. Three potential nucleolar targeting signals (NTS), RRQR, KEKK, and KKKK, are located within the NLS (fig. 2A). However, the three NTS motifs are well conserved only in ING1 and ING2 but are missing from the other INGs. Only the first and third motifs are capable of targeting ING1 to nucleoli (Scott et al. 2001b). The leucine zipper–like (LZL) motif found in ING2 consists of leucine residues spanning every seven amino acids, forming a hydrophobic patch. Sequence-based analysis has shown a similar leucine distribution at the N-terminus of ING3 to ING5 (fig. 2A), suggesting similar potential functions. An isoform-specific motif exists on p33ING1b, where a PCNA-interacting protein (PIP) domain can be found at the N-terminus. The overall structural features of the ING proteins are summarized in figure 2B, which outlines the approximate position of the conserved motifs.
FIG. 2.— Structure of the inhibitor of growth (ING) proteins in Homo sapiens. (A) Amino acid sequence alignment of the ING protein members, including the four ING1 splicing variants. Different conserved domains are indicated by different colored boxes, including a proliferating cell nuclear antigen (PCNA)-interacting protein (PIP) domain on p33ING1b; a leucine zipper–like motif (LZL) on ING2, ING3, ING4, and ING5; the nuclear localization sequence (NLS, within which are three nucleolar targeting signals [NTS]); and a plant homeodomain (PHD) conserved in all members except INGX, which only has a partial PHD, a phosphorylation-dependent interacting motif (PDIM), and a peptide-interacting motif (PIM) only found on ING1 and ING2. A potential chromatin regulatory (PCR) domain is also shown (pink box) that may link ING proteins to HAT and HDAC complexes. p47ING1a and ING3 have additional inserted sequences that are unique and are indicated by the black lines protruding out of the sequence alignment. (B) A diagrammatic representation of the major structural features of the ING proteins.
A highly conserved region across all of the five ING members exists from amino acid V74 to S126 of p33ING1b (shown by arrowheads in figure 2A and denoted by PCR), but pattern and profile searches using online tools such as InterProScan (http://www.ebi.ac.uk/InterProScan/) and MotifScan (http://hits.isb-sib.ch/cgi-bin/PFSCAN/) have failed to return positive matches.
Protein Sequence Alignments
A database consisting of sequences judged to be members of the ING protein family was compiled (table 1). The ING sequences were a collection from a variety of species, including human, mouse, rat, frogs, fish, mosquito, fruit fly, worms, fungi, and plants. A total of 60 ING/ING-related sequences were included in our study based on their sequence similarity. Multiple sequence alignment of the collected ING sequences has identified four major conserved regions of the ING proteins (fig. 3). Region I consists of conserved leucine/isoleucine amino acid residues. The known leucine zipper of ING2_Hs falls in this conserved region. Region II contains several highly conserved amino acids such as a KIQI/KVQL motif, with well-conserved residues occurring every seven amino acids, suggesting a conserved function on a distinct face of an -helix. The most highly conserved region of ING proteins, the PHD, falls into what we have defined as region III. Although several gaps were introduced in the alignment because of small variations in amino acid sequences, the overall C4-H-C3 zinc-finger motif of the PHD remained intact in the ING proteins across species, from human to plants. The C-terminal region of the ING protein contains the fourth conserved region that is significantly longer in ING1 and ING2 compared with the other ING members. This region contains a PIM. Interestingly, the nuclear localization sequences (NLS) and the nucleolar targeting signals (NTS), although well conserved in vertebrate ING genes (figure 2 and Wagner et al. [2001]), showed a very low degree of conservation within invertebrate and plant species (data not shown), suggesting that ING proteins may be targeted differently.
FIG. 3.— Amino acid sequence alignments of INGs from human, cow, mouse, rat, frogs, fish, mosquito, fruit fly, worms, fungi, and plants. The multiple sequence alignment was done using T-COFFEE. The alignment was then visualized by the multiple sequence alignment editor, GENEDOC. Shading is based on conservation, with the darkest shading representing 100% conservation and the lightest less than 60%. Four conserved regions are indicated. The characteristic signature motif of PHD domains, C4-H-C3, is the most highly conserved region across all sequences. The boxes outline the signature motifs on both the PCR and PIM. These motifs are characteristic for the corresponding ING subfamily. LZL, leucine zipper–like motif; PCR, potential chromatin regulatory domain; PHD, plant homeodomain; PIM, peptide-interacting motif.
Phylogenetic Analysis
Phylogenetic trees were generated from sequences presented in table 1 using both the distance and parsimony methods with statistical confidence measured by bootstrap analysis (fig. 4A and B). In the distance tree, vertebrate and some arthropod ING proteins fall into five distinct groups, whereas most ING proteins from nematodes, fungi, and plants do not resemble any particular human ING member more than another (fig. 4A). In the parsimony tree, although some groups are not as well defined as those in the distance tree, the cores of ING1 and ING 2 are maintained, and ING4 and ING5 are also grouped clearly on one branch (fig. 4B). Comparison of the distance tree with the parsimony tree highlights the similarity of ING4 and ING5, which gave identical clades, and the slightly more divergent, but nonetheless related, nature of ING1 and ING2. The composite tree (fig. 5) clearly illustrates the likely presence of an ancient ING family distinct from the five known vertebrate ING members. We, therefore, classified this large, functionally uncharacterized group more generally as ING. The ING3 proteins, which are found in human, mouse, rat, fish, and frogs, constitute a large and relatively distinct branch of the composite tree that is also found in plants by distance analysis. Parsimony analysis also suggests that some fungi and nematodes have paralogs of the ING3 gene, suggesting that ING3 may be evolutionarily distinct from other INGs in vertebrates. This idea is corroborated by the chromosomal location of ING3 compared with other ING genes (fig. 1) and by recent identification of human ING3 homologs in C. elegans and yeast (Ceol and Horvitz 2004; Doyon et al. 2004).
FIG. 4.— The phylogenetic trees of the ING protein family. The tree on the left is derived by neighbor-joining distance analysis, whereas the tree on the right is derived by parsimony analysis. The statistical reliability of the inferred tree topology was assessed by the bootstrap test. The bootstrap values, which show as a percentage calculated form 1,000 data sets, are shown at the nodes. The distinct groups of ING proteins are also indicated. Phylogenetic analyses of PHD only and of the four conserved regions gave similar results. These results and the percent identity values are available from the authors upon request.
FIG. 5.— The composite tree of the ING protein family. Branch positions were determined by neighbor-joining analysis of the 60 ING/ING-like sequences compiled in table 1. The five major ING groups are color coded and indicated in brackets.
Novel ING Proteins
Based on the multiple sequence alignments and phylogenetic analyses, several proteins from diverse organisms, including frog, fish, mosquito, fruit fly, worms, fungi, and plants, were classified as novel ING protein members. As shown in table 1, these new ING members include: ING5_Xt (#30); ING5b_Dr (#33); ING2_Ag (#34); ING2_Dm, ING4a,b_Dm (#35 to #37); ING,?,,Cb (#38 to #41); ING,?,_Ce (#42 to #44); ING_Ec (#45); ING,?,_Nc (#51 to #53); ING3a,b_At (#54 and #55); ING,?,_At (#56 to #58); ING3_Os and ING_Os (#59 and #60). Previously, three Saccharomyces cerevisiae proteins (Yng1, Yng2, and Pho23 [table 1, #48 to #50]) and two Schizosaccharomyces pombe proteins (Png1 and Png2 [table 1, #46 and #47]) were reported to share significant sequence identity with the human candidate tumor suppressor p33ING1b in their C-terminal regions (Loewith et al. 2000). According to our analysis, the yeast ING proteins belong to a distinct large group (fig. 5), so we reassigned them as ING, a generalized name for the large low-similarity group with designation of Greek letters to differentiate them without implying particular similarity to one or another of the vertebrate INGs. Based upon functional overlap, the yeast ING?_Sc (i.e., Yng2 [see table 1]) gene may resemble human ING3 most closely (Doyon et al. 2004).
Discussion
The chromosome band 13q34 region where the human ING1 gene is located is very near the telomere of chromosome 13 and has been reported to be a site for translocation and deletion in several cancers, including primary gastric (Motomura et al. 1988) and head and neck squamous cell carcinomas (Maestro et al. 1996). Knowing the tumor suppressor role of the ING proteins, the fact that the relatively closely related pairs of ING genes (ING1 and ING2; ING4 and ING5) are located near the end of their respective chromosome suggests the possibility of altered expression levels during the telomere erosion that initially occurs during replicative aging of primary cells and also during tumorigenesis. In fact, this does occur because the expression of ING1b decreases during cell aging (Berardi et al. 2004) and the expression of ING1 and ING4 are frequently reduced in many cancer types (Noumann et al. 2002; Feng, Hara, and Riabowo 2002; Garkavtsev et al. 2004). The fact that ING3 and INGX are located far from telomeres is likely a reflection of their evolutionary distinction from the other four members and is consistent with our phylogenetic analysis (figs. 4 and 5). However, whether or not INGX protein is produced from transcript has not yet been reported. Nevertheless, the sex chromosome–linked INGX gene, distinct from the other ING genes, proposes an interesting point for future examination and has the potential to act in a dominant negative manner because it is highly truncated.
Although the human ING proteins have been intensively studied since the first isoform was cloned (Garkavtsev et al. 1996), studies have been focused primarily on ING1, especially p33ING1b. Although attention has been increasingly brought to bear on other ING members to determine their biological functions, these studies are, necessarily, rather independent and narrowly focused. A systematic examination of the ING proteins from a phylogenomic perspective is beneficial for elucidating protein structure and potential common functions. Multiple amino acid sequence alignment of human ING proteins revealed a considerable number of conserved regions (fig. 2). Many conserved motifs have been shown to be essential for ING functions and activities. For example, the PHD is responsible for binding of phosphatidylinositol phosphates (PtdInsPs) in response to DNA damage–initiated stress signaling (Gozani et al. 2003); the NLS/NTS targets ING1 and, likely, other INGs to different chromatin domains in the nucleus and nucleolus in response to UV-induced DNA damage (Scott et al. 2001b), and the PIP domain targets only the p33ING1b of ING1 to PCNA after DNA damage (Scott et al. 2001a). An additional putative role for PHD fingers has been recently proposed (Ragvin et al. 2004), in which the PHD fingers may serve as signal transducers in the interaction of proteins or protein complexes with nucleosomes. Upon cooperatively interacting with other proteins or protein complexes, the PHD finger is bound to a nucleosome. This model is consistent with previous observations that needle microinjection of expression constructs encoding p33ING1b and p47ING1a resulted in increased or decreased histone H3 and H4 acetylation levels, respectively (Vieyra et al. 2002), and the known involvement of ING proteins in chromatin remodeling and HAT/HDAC interactions (reviewed in Feng, Hara, and Riabowol [2002]).
This bioinformatics/phylogenetic analysis builds upon the initial study presented in Feng, Hara, and Riabowol (2002) and identifies additional structural motifs that might be important for ING function and activity. The PIM found at the C-terminal end of ING1 and ING2 has been recently identified and found to stabilize proteins with particular posttranslational modification (Feng et al. 2004). In addition, interaction of ING with other proteins may be mediated through the candidate phosphorylation-dependent interacting motif (PDIM), which is relatively similar to the canonical RSXpSXP 14-3-3 binding motif. This implies that phosphorylation of this region could recruit proteins important in the modulation of such cellular processes as apoptosis, signal transduction, and cell cycle regulation (reviewed in Hermeking [2003]). We also postulate the existence of an LZL motif near the N-terminus of ING3 to ING5 based on the observation that conserved leucine residues were widely distributed on these sequences similar to the known leucine zipper on ING2. If so, ING3 to ING5 may form homodimers and heterodimers or interact with other leucine zipper–containing proteins, such as transcription factors. Initial results suggest that this is the case, making targeting of particular HAT and HDAC complexes to chromatin considerably more dynamic (Gong et al., personal communication). The partial NTS found on ING3 to ING5 suggests that the ability of these ING proteins to be targeted to the nucleus or to the nucleolus in response to UV-induced DNA damage may be compromised or missing in these isoforms. It is also possible that this might represent a lower affinity motif that could allow a greater degree of partitioning of ING proteins to the cytoplasm. Ongoing experiments are testing these hypotheses.
In our comparative sequence analysis across different species, the four conserved regions of ING proteins was a good indicator of structural/functional conservation. The conserved LZL region (fig. 3, region I) re-emphasizes the potential ability of ING proteins to bind other leucine zipper–containing proteins. It is also very important to emphasize the highly conserved region that has a distinct KIQI/KVQL motif (figs. 2A and 3, region II). Kawaji et al. (2002) independently identified this novel and conserved motif, which they labeled MDS00105 and noted is specific for the mammalian ING family. That analysis subdivided it into three submotifs, Q-E-L-G-D-E-K-[IM]-Q, K-E-[FY]-[SG]-D-D-K-V-Q, and [LM]-E-D-A-D-E-K-V-[AQ], that were relatively specific for ING1/ING1L, ING1-homolog, and ING3 subfamilies, respectively. Our results shown in figure 3 extend their observations in that the ING1/ING2 (i.e., ING1L) subfamily carries the Q-E-L-G-D-[ED]-K-[ILM]-Q-[IL] motif, the ING4/ING5 (i.e., ING1 homolog) subfamily has the K-E-[FY]-[SG]-D-D-K-V-Q-L motif, the ING3 subfamily is characterized by the L-E-D-A-D-E-K-V-Q-L motif, whereas the generalized large ING subfamily shows a relatively low degree of conservation to any of the above motifs. An exception to this rule is that the three plant ING3 proteins do not bear the L-E-D-A-D-E-K-V-Q-L signature. This region also displays many other conserved residues, signifying its potentially important role. In fact, we have hypothesized that this region of ING proteins involves binding of HAT, HDAC, MYC, and other cell cycle–related proteins (Helbing et al. 1997), whereas the unique regions of each subfamily member have been suggested to modulate interactions (Kawaji et al. 2002). Kuzmichev et al. (2002) also identified the N-terminal 125 amino acids of p33ING1b, which includes this distinct conserved region, as a motif linked to the Sin3/HDAC complex through direct interaction with SAP30. Therefore, it is becoming increasingly clear that this region of the ING proteins may play a critical role in binding HAT/HDAC complex during chromatin remodeling and regulation of gene expression; hence, the name we propose is the potential chromatin regulatory (PCR) domain (figs. 2 and 3).
The criteria for defining an ING sequence now can include sequence conservation in the first three regions and partially in region IV. Similar to the PCR domain, the PIM (fig. 3, region IV) is also specific for different subfamilies. The distinction between ING1/ING2, ING4/ING5, ING3, and ING subfamilies can be made based on the length of this C-terminal conserved region, because this region in the ING1/ING2 subfamily is considerably longer than the others. The preferential and highly conserved basic residues are indicative of important functions. An additional distinct feature of the ING3 protein is the two insertions of 102 and 54 amino acids located between the PCR domain and the NLS/NTS region (fig. 2A). Although the NLS/NTS region is not as highly conserved across different ING members as the other three defined regions, several lysine residues are preserved (data not shown). As we have postulated before, ING1 and ING2, which have a distinctive NLS/NTS region, are most likely translocated to the nucleolus more efficiently in response to DNA damage but experimental proof of this remains to be presented.
The evolutionary distance of the ING protein members can be estimated from our phylogenetic analysis. We conclude that ING1 and ING2, and ING4 and ING5 are closely related and their chromosomal locations suggest the possibility of duplication of terminal regions of particular chromosome pairs. ING3 on the other hand, is relatively distant from the closely related ING4 and ING5 and the well-related ING 1 and ING2. This may be reflected by ING3s atypical (nontelomeric) location on chromosome 7 and its poorly conserved NLS domain. It may also reflect a more ancestral quality of ING3 in that the distance tree grouped three plant ING3s with Danio rerio, Xenopus, mouse, rat, and human ING3 (fig. 5), which is quite different from the ING1/ING2 and ING4/ING5 group that do not contain plant sequences. The larger Drosophila ING proteins, labeled ING2_Dm, although containing the signature Q-E-L-G-D-[ED]-K-[ILM]-Q-[IL] motif, may not functionally correspond to human ING2, but this remains to be rigorously tested.
Conclusion
The ING family of PHD finger proteins is involved in a wide variety of biological processes, including growth regulation, cellular aging, DNA repair, oncogenesis, and apoptosis. In this study, the naming and classification of ING proteins have been clarified, the evolutionary relationship among the five members of the ING family has been further elucidated, and additional conserved structural features have been identified through sequence and phylogenetic analysis. However, the exact functions of these new conserved regions/motifs needs to be further examined. Nevertheless, the results outlined in this study provide us a better understanding of the ING protein family and define useful target motifs for further examination.
Acknowledgements
We thank W. Gong, X. Feng, P. Berardi, and Dr. G.E. Moorhead for helpful suggestions during the preparation of this work. This work was supported by grants from the Canadian Institutes of Health Research and the Alberta Heritage Foundation for Medical Research to K. Riabowol. The Sensen laboratory is funded by the Alberta Science and Research Authority (ASRA), Western Economic Diversification (WD), The Alberta Network for Proteomics Innovation (ANPI), Genome Canada, Genome Prairie, the Canada Foundation for Innovation (CFI), the Natural Sciences and Engineering Research Council (NSERC), Sun Microsystems Inc., and the University of Calgary. This work was also supported in part by an NSERC operating grant, an NSERC University Faculty award, and a Michael Smith Foundation for Health Research Scholar award to C.C.H. and an NSERC Canada Graduate Scholarship and Michael Smith Foundation for Health Research Award to M.J.W.
References
Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389–3402.
Aravind, L., L. M. Iyer, and E. V. Koonin. 2003. Scores of RINGS but no PHDs in ubiquitin signaling. Cell Cycle 2:123–126.
Bromidge, T., and C. Lynas. 2002. Relative levels of alternative transcripts of the ING1 gene and lack of mutations of p33/ING1 in haematological malignanceies. Leuk. Res. 26:631–635.
Chang, M. S., H. Sasaki, M. S. Campbell et al. (12 co-authors). 1999. HRad17 colocalizes with NHP2L1 in the nucleolus and redistributes after UV irradiation. J. Biol. Chem. 274:36544–36549.
Chen, B., E. I. Campos, R. Crawford, M. Martinka, and G. Li. 2003. Analyses of the tumor suppressor ING1 expression and gene mutation in human basal cell carcinoma. Int. J. Oncol. 22:927–931.[ISI][Medline]
Chen, L., N. Matsubara, T. Yoshino, T. Nagasaka, N. Hoshizima, Y. Shirakawa, Y. Naomoto, H. Isozaki, K. Riabowol, and N. Tanaka. 2001. Genetic alterations of candidate tumor suppressor ING1 in human esophageal squamous cell cancer. Cancer Res. 61:4345–4349.
Cheung, K. J. Jr, D. Mitchell, P. Lin, and G. Li. 2001. The tumor suppressor candidate p33(ING1) mediates repair of UV-damaged DNA. Cancer Res. 61:4974–4977.
Ceol, C. J., and H. R. Horvitz. 2004. A new class of C. elegans synMuv genes implicates a Tip60/NuA4-like HAT complex as a negative regulator of Ras signaling. Dev. Cell 6:563–576.
Coscoy, L., and D. Ganem. 2003. PHD domains and E3 ubiquitin ligases: viruses make the connection. Trends Cell Biol. 13:7–12.
Doyon, Y., W. Selleck, W. S. Lane, S. Tan, and J. Cote. 2004. Structural and functional conservation of the NuA4 histone acetyltransferase complex from yeast to humans. Mol. Cell Biol. 24:1884–1896.
Eisen, J. A. 1998. Phylogenomics: improving functional predictions for uncharacterized genes by evolutionary analysis. Genome Res. 8:163–167.
Felsenstein, J. 1989. PHYLIP (phylogeny inference package). Version 3.2. Cladistics 5:164–166.
Feng, X., Y. Hara, and K. Riabowol. 2002. Different HATS of the ING1 gene family. Trends Cell Biol. 12:532–538.
Garkavtsev, I., A. Kazarov, A. Gudkov, and K. Riabowol. 1996. Suppression of the novel growth inhibitor p33ING1 promotes neoplastic transformation. Nat. Genet. 14:415–420.
———. 1999. Correction and addition to ING1 isoform sequences. Nat. Genet. 23:373.
Garkavtsev, I., S. V. Kozin, O. Chernova, L. Xu, F. Winkler, E. Brown, G. H. Barnett, and R. K. Jain. 2004. The candidate tumor suppressor protein ING4 regulates brain tumor growth and angiogenesis. Nature 428:328–332.
Garkavtsev, I., and K. Riabowol. 1997. Extension of the replicative life span of human diploid fibroblasts by inhibition of the p33ING1 candidate tumor suppressor. Mol. Cell Biol. 17:2014–2019.
Gerber, D. A., S. Souquere-Besse, F. Puvion, M. F. Dubois, O. Bensaude, and C. Cochet. 2000. Heat-induced relocalization of protein kinase CK2. Implication of CK2 in the context of cellular stress. J. Biol. Chem. 275:23919–23926.
Gozani, O., P. Karuman, D. R. Jones et al. (20 co-authors). 2003. The PHD finger of the chromatin-associated protein ING2 functions as a nuclear phosphoinositide receptor. Cell 114:99–111.
Gunduz, M., M. Ouchida, K. Fukushima, H. Hanafusa, T. Etani, S. Nishioka, K. Nishizaki, and K. Shimizu. 2000. Genomic structure of the human ING1 gene and tumor-specific mutations detected in head and neck squamous cell carcinomas. Cancer Res. 60:3143–3146.
Gunduz, M., M. Ouchida, K. Fukushima, S. Ito, Y. Jitsumori, T. Nakashima, N. Nagai, K. Nishizaki, and K. Shimizu. 2002. Allelic loss and reduced expression of the ING3, a candidate tumor suppressor gene at 7q31, in human head and neck cancers. Oncogene 21:4462–4470.
Hara, Y., Z. Zheng, S. C. Evans, D. Malatjalian, D. C. Riddell, D. L. Guernsey, L. D. Wang, K. Riabowol, and A. G. Casson. 2003. Ing1 and dp53 tumor suppressor gene alterations in adenocarcinomas of the esophagogastric junction. Cancer Lett. 192:109–116.
Helbing, C., C. Veillette, K. Riabowol, R. Johnston, and I. Garkavtsev. 1997. A novel candidate tumor suppressor, ING1, is involved in the regulation of apoptosis. Cancer Res. 57:1255–1258.
Hermeking, H. 2003. The 14-3-3 cancer connection. Nat. Rev. Cancer 3:931–943.
Ito, K., K. Kinjo, T. Nakazato, Y. Ikeda, and M. Kizaki. 2002. Expression and sequence analyses of p33(ING1) gene in myeloid leukemia. Am. J. Hematol. 69:141–143.
Jager, D., E. Stockert, M. J. Scanlan, A. O. Gure, E. Jager, A. Knuth, L. J. Old, and Y. T. Chen. 1999. Cancer—testis antigens and ING1 tumor suppressor gene product are breast cancer antigens: characterization of tissue-specific ING1 transcripts and a homolog gene. Cancer Res. 59:6197–6204.
Kawaji, H., C. Schonbach, Y. Matsuo, J. Kawai, Y. Okazaki, Y. Hayashizaki, and H. Matsuda. 2002. Exploration of novel motifs derived from mouse cDNA sequences. Genome Res. 12:367–378.
Klibanov, S. A., H. M. O'Hagan, and M. Ljungman. 2001. Accumulation of soluble and nucleolar-associated p53 proteins following cellular stress. J. Cell Sci. 114:1867–1873.
Krishnamurthy, J., K. Kannan, J. Feng, B. K. Mohanprasad, N. Tsuchida, and G. Shanmugam. 2001. Mutational analysis of the candidate tumor suppressor gene ING1 in Indian oral squamous cell carcinoma. Oral Oncol. 37:222–224.
Kuzmichev, A., Y. Zhang, H. Erdjument-Bromage, P. Tempst, and D. Reinberg. 2002. Role of the Sin3-histone deacetylase complex in growth regulation by the candidate tumor suppressor p33(ING1). Mol. Cell Biol. 22:835–848.
Leung, A. K., and A. I. Lamond. 2003. The dynamics of the nucleolus. Crit. Rev. Euk. Gene Expr. 13:39–54.
Loewith, R., M. Meijer, S. Lees-Miller, K. Riabowol, and D. Young. 2000. Three yeast proteins related to the human candidate tumor suppressor p33ING1 are associated with histone acetyltransferase activities. Mol. Cell Biol. 20:3807–3816.
Maestro, R., S. Paccinin, C. Doglioni, D. Gasparotto, T. Vukosavljevic, S. Sulfaro, L. Barzan, and M. Boiocchi. 1996. Chromosome 13q deletion mapping in head and neck squamous cell carcinomas: identification of two distinct regions of preferential loss. Cancer Res. 56:1146–1150.
Mi, Y., S. D. Thomas, X. Xu, L. K. Casson, D. M. Miller, and P. J. Bates. 2003. Apoptosis in leukemia cells is accompanied by alterations in the levels and localization of nucleolin. J. Biol. Chem. 278:8572–8579.
Motomura, K., I. Nishisho, S. Takai, H. Tateisi, M. Okazaki, M. Yamamoto, T. Miki, T. Honjo, and T. Mori. 1988. Loss of alleles at loci on chromosome 13 in human primary gastric cancers. Genomics 2:180–184.[CrossRef][Medline]
Nagashima, M., M. Shiseki, K. Miura, K. Hagiwara, S. P. Linke, R. Pedeux, X. W. Wang, J. Yokota, K. Riabowol, and C. C. Harris. 2001. DNA damage-inducible gene p33ING2 negatively regulates cell proliferation through acetylation of p53. Proc. Natl. Acad. Sci. USA 98:9671–9676.
Nagashima, M., M. Shiseki, R. M. Pedeux, S. Okamura, M. Kitahama-Shiseki, K. Miura, J. Yokota, and C. C. Harris. 2003. A novel PHD-finger motif protein, p47ING3, modulates p53-mediated transcription, cell cycle control, and apoptosis. Oncogene 22:343–350.
Nicholas, K. B., H. B. Nicholas, and D. W. Deerfield II. 1997. GeneDoc: analysis and visualization of genetic variation. EMBNEW.NEWS 4:14.
Notredame, C., D. Higgins, and J. Heringa. 2000. T-Coffee: a novel method for multiple sequence alignments. J. Mol. Biol. 302:205–217.
Noumann, G. S., J. J. Anderson, K. M. Wook, J. Lunec, A. G. Hall, M. M. Reid, and B. Angus. 2002. Loss of nuclear expression of the p33(ING1b) inhibitor of growth protein in childhood acute lymphoblastic leukaemia. J. Clin. Pathol. 55:596–601.
Ohmori, M., M. Nagai, T. Tasaka, H. P. Koeffler, T. Toyama, K. Riabowol, and J. Takahara. 1999. Decreased expression of p33ING1b mRNA in lymphoid malignancies. Am. J. Hematol. 62:118–119.
Oki, E., Y. Maehara, E. Tokunaga, Y. Kakeji, and K. Sugimachi. 1999. Reduced expression of p33(ING1) and the relationship with p53 expression in human gastric cancer. Cancer Lett. 147:157–162.
Ragvin, A., H. Valvatne, S. Erdal, V. Arskog, K. R. Tufteland, K. Breen, A. M. OYan, A. Eberharter, T. J. Gibson, P. B. Becker, and R. Aasland. 2004. Nucleosome binding by the bromodomain and PHD finger of the transcriptional cofactor p300. J. Mol. Biol. 337:773–788.
Saito, A., T. Furukawa, S. Fukushige, S. Koyama, M. Hoshi, Y. Hayashi, and A. Horii. 2000. p24/ING1-ALT1 an dp47/ING1-ALT2, distinct alternative transcripts of p33ING1. J. Hum. Genet. 45:177–181.
Scheel, H., and K. Hofmann. 2003. No evidence for PHD fingers as ubiquitin ligases. Trends Cell Biol. 13:285–287.
Scott, M., P. Bonnefin, D. Vieyra, F. M. Boisvert, D. Young, D. P. Bazett-Jones, and K. Riabowol. 2001a. UV-induced binding of ING1 to PCNA regulates the induction of apoptosis. J. Cell Sci. 114:3455–3462.
Scott, M., F. M. Boisvert, D. Vieyra, R. N. Johnston, D. P. Bazett-Jones, and K. Riabowol. 2001b. UV induces nucleolar translocation in ING1 through two distinct nucleolar targeting sequences. Nucleic Acids Res. 29:2053–2058.
Shinoura, N., Y. Muramatsu, M. Nishimura et al. (12 co-authors). 1999. Adenovirus-mediated transfer of p33ING1 with p53 drastically augments apoptosis in gliomas. Cancer Res. 59:5521–5528.
Shiseki, M., M. Nagashima, R. M. Pedeux et al. (11 co-authors). 2003. p29ING4 and p28ING5 bind to p53 and p300, and enhance p53 activity. Cancer Res. 63:2373–2378.
Skowyra, D., M. Zeremski, N. Neznanov, M. Li, Y. Choi, M. Uesugi, C. A. Hauser, W. Gu, A. V. Gudkov, and J. Qin. 2001. Differential association of products of alternative transcripts of the candidate tumor suppressor ING1 with the mSin3/HDAC transcriptional corepressor complex. J. Biol. Chem. 276:8734–8739.
Sugimoto, M., M. L. Kuo, M. F. Roussel, and C. J. Sherr. 2003. Nucleolar Arf tumor suppressor inhibits ribosomal RNA processing. Mol. Cell 2:415–424.
Tallen, G., I. Kaiser, S. Krabbe, U. Lass, C. Hartmann, G. Henze, K. Riabowol, and A. vonDeimling. 2004. No ING1 mutations in human brain tumors but reduced expression in high malignancy grades of astrocytoma. Int. J. Cancer 109:476–479.
Tokunaga, E., Y. Maehara, E. Oki, K. Kitamura, Y. Kakeji, S. Ohno, and K. Sugimachi. 2000. Diminished expression of ING1 mRNA and the correlation with p53 expression in breast cancers. Cancer Lett. 152:15–22.
Toyama, T., H. Iwase, P. Watson et al. (11 co-authors). 1999. Suppression of ING1 expession in sporadic breast cancer. Oncogene 18:5187–5193.
Toyama, T., H. Iwase, H. Yamashita, Y. Hara, H. Sugiura, Z. Zhang, I. Fukai, Y. Miura, K. Riabowol, and Y. Fujii. 2003. p33(ING1b) stimulates the transcriptional activity of the estrogen receptor alpha via its activation function (AF) 2 domain. J. Ster. Biochem. Mol. Biol. 87:57–63.
Vieyra, D., R. Loewith, M. Scott et al. (14 co-authors). 2002. ING1 proteins differentially regulate histone acetylation. J. Biol. Chem. 277:29832–29839.
Wagner, M. J., M. Gogel-spehar, R. C. Skirrow, R. N. Johnston, K. Riabowol, and C. C. Helbing. 2001. Expression of novel ING variants is regulated by thyroid hormone in the Xenopus laevis tadpole. J. Biol. Chem. 276:47013–47020.
Warbrick, E. 1998. PCNA binding through a conserved motif. BioEssays 20:195–199.(Gordon H.Y. He*, Caren C.)
Correspondence: E-mail: karl@ucalgary.ca.
Abstract
Since the discovery of ING1 class II tumor suppressors in 1996, five different ING genes (ING1 to ING5) encoding proteins with highly conserved plant homeodomain (PHD) motifs and several splicing isoforms of the ING1 and ING2 gene have been identified. The ING family functions in DNA repair and apoptosis in response to UV damage through binding to proliferating cell nuclear antigen (PCNA); chromatin remodeling and regulation of gene expression through regulating and/or targeting histone acetyltransferase/deacetylase (HAT/HDAC) activities; binding targets of rare phosphatidylinositol phosphates (PtdInsPs) that function in DNA damage-initiated stress signaling; and regulating brain tumor angiogenesis through transcriptional repression of NF-KB–responsive genes. To elucidate the evolutionary history of ING proteins and summarize what is known about regions highly conserved in the ING family members, we have examined the sequences and phylogenetic relationships of ING proteins across taxonomically diverse organisms. We have identified novel ING family members in rats, frogs, fish, mosquitoes, fruit flies, worms, fungi, and plants. We have also clarified the naming and classification of ING proteins based on our phylogenetic analysis to allow better understanding of the ING protein family. Using sequence similarities, we have identified novel regions and motifs of unknown function that are conserved across family members. An evolutionary history for the ING family of PHD finger proteins is presented that indicates that five ING genes are present in vertebrates. Three of these may be paralogs of ING genes found in arthropods, whereas nematodes, fungi, and green plants contain ING genes that have general features of the vertebrate ING family.
Key Words: ING1 ? PHD finger ? molecular evolution ? protein phylogeny
Introduction
Using subtractive hybridization between cDNAs from a normal mammary cell line and several transformed breast cancer epithelial cell lines, the first member of the ING gene family was identified by the subsequent selection for sense or antisense cDNA fragments capable of promoting neoplastic transformation (Garkavtsev et al. 1996). Suppression of ING1 promotes focus formation and growth in soft agar in vitro and tumor formation in vivo (Garkavtsev et al. 1996). ING1 expression is significantly repressed in more than 40% of human primary breast cancers, 100% of established breast cancer cell lines (Toyama et al. 1999), and many other forms of blood and solid tumors (Ohmori et al. 1999; Oki et al. 1999; Shinoura et al. 1999; Tokunaga et al. 2000; Chen et al. 2001; Krishnamurthy et al. 2001; Bromidge and Lynas 2002; Gunduz et al. 2002; Ito et al. 2002; Noumann et al. 2002; Chen et al. 2003; Hara et al. 2003; Garkavtsev et al. 2004; Tallen et al. 2004). Since the initial description of ING1, studies from diverse groups using different model systems have implicated the ING proteins in control of cellular aging (Garkavtsev and Riabowol 1997), negative regulation of cell proliferation (Garkavtsev et al. 1996; Nagashima et al. 2001), apoptosis (Helbing et al. 1997), DNA damage repair (Scott et al. 2001a; Cheung et al. 2001), chromatin remodeling (Loewith et al. 2000; Vieyra et al. 2002; Kuzmichev et al. 2002), hormone responses (Wagner et al. 2001; Toyama et al. 2003), and regulation of brain tumor growth and angiogenesis (Garkavtsev et al. 2004).
The human ING1 gene has three exons that can be alternatively spliced onto a common 3' exon, thereby generating p27ING1d, p33ING1b, and p47ING1a. In addition, internal initiation at an ATG within the common exon generates p24ING1c (Garkavtsev et al. 1999; Jager et al. 1999; Gunduz et al. 2000; Saito et al. 2000). By screening a breast cancer cell cDNA library using ING1 as probe, followed by 5' RACE using normal fetus cDNA, Jager et al. (1999) cloned an ING1 homolog gene, which maps to the X chromosome. The deduced 42–amino acid peptide has a calculated molecular mass of 5.0 kDa and shares 76% identity over its length with the much longer ING1 protein. Expression of this gene was detected in all normal tissues tested and in some cancer cell lines using RT-PCR. To date, four additional genes have been identified in humans that encode ING2 to ING5 (Nagashima et al. 2001, 2003; Shiseki et al. 2003). Study of amphibian ING2 genes provides evidence of differential regulation of presumed splice variants of this gene (Wagner et al. 2001). Three forms of ING2 have been cloned in Xenopus laevis; however, whether they are actually splicing isoforms remains to be confirmed.
All ING genes share strongest homology in a region encoding a plant homeodomain (PHD) motif that has been implicated in the regulation of ubiquitination (Coscoy and Ganem 2003), although this function remains controversial (Aravind, Iyer, Koonin 2003; Scheel and Hofmann 2003). More recently, ING PHDs have been identified as binding targets of rare phosphatidylinositol phosphates (PtdInsPs) that function in DNA damage–initiated stress signaling (Gozani et al. 2003). ING1 also translocates to the nucleolus after UV-induced DNA damage (Scott et al. 2001b). This property of UV-induced relocation to nucleoli is shared with several other proteins that may play roles in DNA repair and/or apoptosis, such as nucleolin (Mi et al. 2003), the Werner's syndrome helicase (Leung and Lamond 2003), the CK2 kinase that phosphorylates DNA ligase and topoisomerase 2 (Gerber et al. 2000), hRad17 (Chang et al. 1999), and p53/MDM2/ARF, among others (Klibanov, O'Hagan, and Ljungman 2001; Leung and Lamond 2003; Sugimoto et al. 2003). Mutational studies identified two intrinsic nucleolar translocation sequences (NTS) within the nuclear localization signal (NLS) found in all ING family members (Scott et al. 2001b). p33ING1b also binds to the proliferating cell nuclear antigen (PCNA) through a specific sequence called the PCNA-interacting protein (PIP) domain (Warbrick 1998), which is found in proteins involved in growth inhibition (e.g., p21WAF1), growth arrest after DNA damage (e.g., GADD45), and DNA replication and repair (e.g., FEN1) (Feng, Hara, and Riabowol 2002). The ability of ING proteins to bind to and alter the activity of histone acetyltransferases (HATs), histone deacetylases (HDACs), and factor acetyltransferases (FATs) has also been shown (Loewith et al. 2000; Skowyra et al. 2001; Nagashima et al. 2001; Vieyra et al. 2002; Kuzmichev et al. 2002) (reviewed in Feng, Hara, and Riabowol [2002]).
ING sequences have been reported in human, mouse, rat, frog, fission and budding yeast, Drosophila, and C. elegans, as well as in other species; however, most of these homologs remain unrecognized and uncharacterized in databases, including those of the NCBI. Despite the increasing number of ING homologs that have been identified in different organisms, a comprehensive analysis of the evolution of this protein family has not yet been conducted. No studies have reported on the phylogenetic relationships and evolutionary history of the ING family. Because of their potentially important roles in many central biological processes, a better understanding of the multiple ING family members is of particular research interest, and a comprehensive analysis of the ING family across taxonomically diverse organisms would be useful for future studies.
Evolutionary studies of a protein family can be instrumental in determining structurally conserved regions, leading to useful predictions of protein function (Eisen 1998). Moreover, an examination of the evolutionary history of the protein family may indicate the presence of novel family members. In this study, we have undertaken a comprehensive analysis of publicly accessible databases and selected unpublished sequences to identify as many members of the ING family as possible, to further characterize ING-specific motifs and to present an analysis of the evolutionary relationship among candidate ING members from various species.
Materials and Methods
Sequence Searches and Mapping Data Retrieval
Extensive word-based searches for ING/ING-like proteins deposited in GenBank (http://www.ncbi.nlm.nih.gov/) were performed using appropriate word combinations, including p33ING1b, p47, ING1-like, growth inhibitor, apoptosis, and PHD, among others. Blast (Altschul et al. 1997) searches were performed at the NCBI Web site (http://www.ncbi.nlm.nih.gov/BLAST/) on all databases available as of March 20, 2004. Additional ING sequences were obtained from work in preparation, and, therefore, are currently not in the public databases. The origin of these sequences is indicated in table 1. Redundant sequences were excluded. All protein sequences were collected in FASTA format for further analysis. Gene loci and mapping data were retrieved from NCBI Human Genome Resources (http://www.ncbi.nlm.nih.gov/genome/guide/human/).
Table 1 List of ING Proteins Used for Amino Acid Sequence Alignment
Naming of Sequences
In this study we denoted all family members of inhibitor of growth proteins as ING followed by the abbreviation of genus and species name (e.g., Mus musculus ING was denoted ING_Mm). Only a small portion of the proteins, which are identified based on sequence similarity, have been functionally characterized, whereas many others are still described as putative/hypothetical proteins. Based on the results of our phylogenetic analysis, the sequences were renamed according to their phylogenetic distance to the five relatively well studied human counterparts and, thus, were assigned different numbers accordingly (i.e., 1 to 5). A large group of evolutionarily distant proteins was categorized more generally as ING with designation of , ?, , and to differentiate them within a species. These designations do not imply similarity across species, as do the ING1 to ING5 designations. The lowercase letters (i.e., a to d) denote the splicing isoform of the same gene followed by an Arabic number indicating the transcript variant. Note that we did not distinguish the functionally characterized ING proteins from the uncharacterized ones in the naming. Originally, the deduced 42–amino acid peptide (Jager et al. 1999) was characterized as an ING1-like tumor suppressor protein. However, because this gene is uniquely mapped to the X chromosome and the deduced peptide is highly truncated, we renamed it as INGX_Hs to indicate its distinct characteristic. In the case of Xenopus laevis, there is evidence for the existence of two ING1b genes that have presumably arisen as a result of genome duplication, giving rise to a pseudotetraploidy state (Wagner and Helbing 2004). Each of these genes has multiple transcript variants that are designated as ING1b1, ING1b2, and so on. Variants 1 to 4 differ in the 5' untranslated region but give rise to the same putative protein. This protein is designated as ING1b1-4_Xl. A second putative protein that is derived from the duplicated gene is referred to as ING1b5.
Multiple Sequence Alignments
All amino acid alignments were performed using the program T_COFFEE (Notredame, Higgins, and Heringa 2000) with default settings. The alignments were then adjusted and shaded using the multiple sequence alignment editor GENEDOC (Nicholas, Nicholas, and Deerfield 1997).
Phylogenetic Analysis
We used PHYLIP version 3.6a3 (Felsenstein 1989) for our phylogenetic analysis. Both distance and parsimony analyses using the protein alignment as input were performed. Bootstrap values were obtained using SEQBOOT and creating 1,000 delete-half Jackknife data sets. The distance analysis was performed by using PROTDIST and subsequently NEIGHBOR with standard parameters, and the parsimony analysis was performed using PROTPARS with standard parameters. In both cases, the "M" option for the analysis of the multiple data sets created with SEQBOOT was invoked.
Results
Location of ING Genes in the Human Genome
The chromosomal locations of the five human ING genes have been mapped to five different chromosomes noted in figure 1. ING1 to ING5 are located on chromosome bands 13q34, 4q35, 7q31, 12p13.3, and 2q37.3, respectively. ING1, ING2, ING4, and ING5 are in regions very close to the ends of their respective chromosome where the telomere/subtelome region is located, approximately 4, 6.6, 6.6, and 0.4 Mbp from chromosomal terminus, respectively. It is interesting to note that, being only about 440 kbp from the telomere region, ING5 has the potential to be affected by telomere erosion. In contrast, ING3 is located in the middle of the long arm of chromosome 7. The additional, but highly truncated, ING-like sequence, INGX, has been mapped to chromosome Xq12, about 11 Mbp from the centromeric region. Selected genes flanking the INGs are shown in figure 1, including many functionally important genes, such as RAB20, a member of the RAS oncogene family; CASP3, encoding caspase 3 cysteine protease; ACRBP, the acrosin-binding protein; BOK, a BCL2-related ovarian killer; and IL2RG, the interleukin 2 receptor-. Coordinate regulation of these gene clusters has not yet been reported.
FIG. 1.— Chromosomal localization of the inhibitor of growth (ING) genes in Homo sapiens. Chromosome number and the approximate chromosomal length (in Mbp) are indicated. Cytogenetic positions of the ING genes are shown. The chromosomal localization of INGs and several flanking genes are listed with the approximate chromosomal start positions in kbp. The distance to telomeres was estimated by calculating the distance between the ING positions to the end of the chromosome, where the telomere/subtelomere region is located. In the case of INGX, the distance to the centromere was calculated. The centromeric regions are indicated as constrictions. The gene positions and the identities of the flanking genes are available at http://www.ncbi.nlm.nih.gov/genome/guide/human/.
Structural Features of the ING Proteins
The amino acid sequence alignment of human ING proteins revealed several conserved regions (fig. 2A). The PHD is best conserved and contains a C4-H-C3 zinc-finger motif located near the C-terminus. INGX, which is 42 amino acids long, has only a partial PHD (fig. 2B). A small conserved protein-interacting motif (PIM) of ING1 and ING2 is enriched in acidic, basic, and aromatic residues. It binds a defined subset of peptides (Feng et al. 2004) and, along with the PHD region (Gozani et al. 2003), binds to phosphatidylinositol monophosphates. The corresponding C-terminal regions of ING3 to ING5 are also rich in basic residues but are considerably shorter than those of ING1 and ING 2. Towards the N-terminus, a conserved REASP motif found on ING1 and ING2 resembles a phosphorylation-dependent interacting motif (PDIM) similar to the canonical RSXpSXP 14-3-3 binding motif. The nuclear localization sequence (NLS) is a strong basic region located upstream of the PHD. Three potential nucleolar targeting signals (NTS), RRQR, KEKK, and KKKK, are located within the NLS (fig. 2A). However, the three NTS motifs are well conserved only in ING1 and ING2 but are missing from the other INGs. Only the first and third motifs are capable of targeting ING1 to nucleoli (Scott et al. 2001b). The leucine zipper–like (LZL) motif found in ING2 consists of leucine residues spanning every seven amino acids, forming a hydrophobic patch. Sequence-based analysis has shown a similar leucine distribution at the N-terminus of ING3 to ING5 (fig. 2A), suggesting similar potential functions. An isoform-specific motif exists on p33ING1b, where a PCNA-interacting protein (PIP) domain can be found at the N-terminus. The overall structural features of the ING proteins are summarized in figure 2B, which outlines the approximate position of the conserved motifs.
FIG. 2.— Structure of the inhibitor of growth (ING) proteins in Homo sapiens. (A) Amino acid sequence alignment of the ING protein members, including the four ING1 splicing variants. Different conserved domains are indicated by different colored boxes, including a proliferating cell nuclear antigen (PCNA)-interacting protein (PIP) domain on p33ING1b; a leucine zipper–like motif (LZL) on ING2, ING3, ING4, and ING5; the nuclear localization sequence (NLS, within which are three nucleolar targeting signals [NTS]); and a plant homeodomain (PHD) conserved in all members except INGX, which only has a partial PHD, a phosphorylation-dependent interacting motif (PDIM), and a peptide-interacting motif (PIM) only found on ING1 and ING2. A potential chromatin regulatory (PCR) domain is also shown (pink box) that may link ING proteins to HAT and HDAC complexes. p47ING1a and ING3 have additional inserted sequences that are unique and are indicated by the black lines protruding out of the sequence alignment. (B) A diagrammatic representation of the major structural features of the ING proteins.
A highly conserved region across all of the five ING members exists from amino acid V74 to S126 of p33ING1b (shown by arrowheads in figure 2A and denoted by PCR), but pattern and profile searches using online tools such as InterProScan (http://www.ebi.ac.uk/InterProScan/) and MotifScan (http://hits.isb-sib.ch/cgi-bin/PFSCAN/) have failed to return positive matches.
Protein Sequence Alignments
A database consisting of sequences judged to be members of the ING protein family was compiled (table 1). The ING sequences were a collection from a variety of species, including human, mouse, rat, frogs, fish, mosquito, fruit fly, worms, fungi, and plants. A total of 60 ING/ING-related sequences were included in our study based on their sequence similarity. Multiple sequence alignment of the collected ING sequences has identified four major conserved regions of the ING proteins (fig. 3). Region I consists of conserved leucine/isoleucine amino acid residues. The known leucine zipper of ING2_Hs falls in this conserved region. Region II contains several highly conserved amino acids such as a KIQI/KVQL motif, with well-conserved residues occurring every seven amino acids, suggesting a conserved function on a distinct face of an -helix. The most highly conserved region of ING proteins, the PHD, falls into what we have defined as region III. Although several gaps were introduced in the alignment because of small variations in amino acid sequences, the overall C4-H-C3 zinc-finger motif of the PHD remained intact in the ING proteins across species, from human to plants. The C-terminal region of the ING protein contains the fourth conserved region that is significantly longer in ING1 and ING2 compared with the other ING members. This region contains a PIM. Interestingly, the nuclear localization sequences (NLS) and the nucleolar targeting signals (NTS), although well conserved in vertebrate ING genes (figure 2 and Wagner et al. [2001]), showed a very low degree of conservation within invertebrate and plant species (data not shown), suggesting that ING proteins may be targeted differently.
FIG. 3.— Amino acid sequence alignments of INGs from human, cow, mouse, rat, frogs, fish, mosquito, fruit fly, worms, fungi, and plants. The multiple sequence alignment was done using T-COFFEE. The alignment was then visualized by the multiple sequence alignment editor, GENEDOC. Shading is based on conservation, with the darkest shading representing 100% conservation and the lightest less than 60%. Four conserved regions are indicated. The characteristic signature motif of PHD domains, C4-H-C3, is the most highly conserved region across all sequences. The boxes outline the signature motifs on both the PCR and PIM. These motifs are characteristic for the corresponding ING subfamily. LZL, leucine zipper–like motif; PCR, potential chromatin regulatory domain; PHD, plant homeodomain; PIM, peptide-interacting motif.
Phylogenetic Analysis
Phylogenetic trees were generated from sequences presented in table 1 using both the distance and parsimony methods with statistical confidence measured by bootstrap analysis (fig. 4A and B). In the distance tree, vertebrate and some arthropod ING proteins fall into five distinct groups, whereas most ING proteins from nematodes, fungi, and plants do not resemble any particular human ING member more than another (fig. 4A). In the parsimony tree, although some groups are not as well defined as those in the distance tree, the cores of ING1 and ING 2 are maintained, and ING4 and ING5 are also grouped clearly on one branch (fig. 4B). Comparison of the distance tree with the parsimony tree highlights the similarity of ING4 and ING5, which gave identical clades, and the slightly more divergent, but nonetheless related, nature of ING1 and ING2. The composite tree (fig. 5) clearly illustrates the likely presence of an ancient ING family distinct from the five known vertebrate ING members. We, therefore, classified this large, functionally uncharacterized group more generally as ING. The ING3 proteins, which are found in human, mouse, rat, fish, and frogs, constitute a large and relatively distinct branch of the composite tree that is also found in plants by distance analysis. Parsimony analysis also suggests that some fungi and nematodes have paralogs of the ING3 gene, suggesting that ING3 may be evolutionarily distinct from other INGs in vertebrates. This idea is corroborated by the chromosomal location of ING3 compared with other ING genes (fig. 1) and by recent identification of human ING3 homologs in C. elegans and yeast (Ceol and Horvitz 2004; Doyon et al. 2004).
FIG. 4.— The phylogenetic trees of the ING protein family. The tree on the left is derived by neighbor-joining distance analysis, whereas the tree on the right is derived by parsimony analysis. The statistical reliability of the inferred tree topology was assessed by the bootstrap test. The bootstrap values, which show as a percentage calculated form 1,000 data sets, are shown at the nodes. The distinct groups of ING proteins are also indicated. Phylogenetic analyses of PHD only and of the four conserved regions gave similar results. These results and the percent identity values are available from the authors upon request.
FIG. 5.— The composite tree of the ING protein family. Branch positions were determined by neighbor-joining analysis of the 60 ING/ING-like sequences compiled in table 1. The five major ING groups are color coded and indicated in brackets.
Novel ING Proteins
Based on the multiple sequence alignments and phylogenetic analyses, several proteins from diverse organisms, including frog, fish, mosquito, fruit fly, worms, fungi, and plants, were classified as novel ING protein members. As shown in table 1, these new ING members include: ING5_Xt (#30); ING5b_Dr (#33); ING2_Ag (#34); ING2_Dm, ING4a,b_Dm (#35 to #37); ING,?,,Cb (#38 to #41); ING,?,_Ce (#42 to #44); ING_Ec (#45); ING,?,_Nc (#51 to #53); ING3a,b_At (#54 and #55); ING,?,_At (#56 to #58); ING3_Os and ING_Os (#59 and #60). Previously, three Saccharomyces cerevisiae proteins (Yng1, Yng2, and Pho23 [table 1, #48 to #50]) and two Schizosaccharomyces pombe proteins (Png1 and Png2 [table 1, #46 and #47]) were reported to share significant sequence identity with the human candidate tumor suppressor p33ING1b in their C-terminal regions (Loewith et al. 2000). According to our analysis, the yeast ING proteins belong to a distinct large group (fig. 5), so we reassigned them as ING, a generalized name for the large low-similarity group with designation of Greek letters to differentiate them without implying particular similarity to one or another of the vertebrate INGs. Based upon functional overlap, the yeast ING?_Sc (i.e., Yng2 [see table 1]) gene may resemble human ING3 most closely (Doyon et al. 2004).
Discussion
The chromosome band 13q34 region where the human ING1 gene is located is very near the telomere of chromosome 13 and has been reported to be a site for translocation and deletion in several cancers, including primary gastric (Motomura et al. 1988) and head and neck squamous cell carcinomas (Maestro et al. 1996). Knowing the tumor suppressor role of the ING proteins, the fact that the relatively closely related pairs of ING genes (ING1 and ING2; ING4 and ING5) are located near the end of their respective chromosome suggests the possibility of altered expression levels during the telomere erosion that initially occurs during replicative aging of primary cells and also during tumorigenesis. In fact, this does occur because the expression of ING1b decreases during cell aging (Berardi et al. 2004) and the expression of ING1 and ING4 are frequently reduced in many cancer types (Noumann et al. 2002; Feng, Hara, and Riabowo 2002; Garkavtsev et al. 2004). The fact that ING3 and INGX are located far from telomeres is likely a reflection of their evolutionary distinction from the other four members and is consistent with our phylogenetic analysis (figs. 4 and 5). However, whether or not INGX protein is produced from transcript has not yet been reported. Nevertheless, the sex chromosome–linked INGX gene, distinct from the other ING genes, proposes an interesting point for future examination and has the potential to act in a dominant negative manner because it is highly truncated.
Although the human ING proteins have been intensively studied since the first isoform was cloned (Garkavtsev et al. 1996), studies have been focused primarily on ING1, especially p33ING1b. Although attention has been increasingly brought to bear on other ING members to determine their biological functions, these studies are, necessarily, rather independent and narrowly focused. A systematic examination of the ING proteins from a phylogenomic perspective is beneficial for elucidating protein structure and potential common functions. Multiple amino acid sequence alignment of human ING proteins revealed a considerable number of conserved regions (fig. 2). Many conserved motifs have been shown to be essential for ING functions and activities. For example, the PHD is responsible for binding of phosphatidylinositol phosphates (PtdInsPs) in response to DNA damage–initiated stress signaling (Gozani et al. 2003); the NLS/NTS targets ING1 and, likely, other INGs to different chromatin domains in the nucleus and nucleolus in response to UV-induced DNA damage (Scott et al. 2001b), and the PIP domain targets only the p33ING1b of ING1 to PCNA after DNA damage (Scott et al. 2001a). An additional putative role for PHD fingers has been recently proposed (Ragvin et al. 2004), in which the PHD fingers may serve as signal transducers in the interaction of proteins or protein complexes with nucleosomes. Upon cooperatively interacting with other proteins or protein complexes, the PHD finger is bound to a nucleosome. This model is consistent with previous observations that needle microinjection of expression constructs encoding p33ING1b and p47ING1a resulted in increased or decreased histone H3 and H4 acetylation levels, respectively (Vieyra et al. 2002), and the known involvement of ING proteins in chromatin remodeling and HAT/HDAC interactions (reviewed in Feng, Hara, and Riabowol [2002]).
This bioinformatics/phylogenetic analysis builds upon the initial study presented in Feng, Hara, and Riabowol (2002) and identifies additional structural motifs that might be important for ING function and activity. The PIM found at the C-terminal end of ING1 and ING2 has been recently identified and found to stabilize proteins with particular posttranslational modification (Feng et al. 2004). In addition, interaction of ING with other proteins may be mediated through the candidate phosphorylation-dependent interacting motif (PDIM), which is relatively similar to the canonical RSXpSXP 14-3-3 binding motif. This implies that phosphorylation of this region could recruit proteins important in the modulation of such cellular processes as apoptosis, signal transduction, and cell cycle regulation (reviewed in Hermeking [2003]). We also postulate the existence of an LZL motif near the N-terminus of ING3 to ING5 based on the observation that conserved leucine residues were widely distributed on these sequences similar to the known leucine zipper on ING2. If so, ING3 to ING5 may form homodimers and heterodimers or interact with other leucine zipper–containing proteins, such as transcription factors. Initial results suggest that this is the case, making targeting of particular HAT and HDAC complexes to chromatin considerably more dynamic (Gong et al., personal communication). The partial NTS found on ING3 to ING5 suggests that the ability of these ING proteins to be targeted to the nucleus or to the nucleolus in response to UV-induced DNA damage may be compromised or missing in these isoforms. It is also possible that this might represent a lower affinity motif that could allow a greater degree of partitioning of ING proteins to the cytoplasm. Ongoing experiments are testing these hypotheses.
In our comparative sequence analysis across different species, the four conserved regions of ING proteins was a good indicator of structural/functional conservation. The conserved LZL region (fig. 3, region I) re-emphasizes the potential ability of ING proteins to bind other leucine zipper–containing proteins. It is also very important to emphasize the highly conserved region that has a distinct KIQI/KVQL motif (figs. 2A and 3, region II). Kawaji et al. (2002) independently identified this novel and conserved motif, which they labeled MDS00105 and noted is specific for the mammalian ING family. That analysis subdivided it into three submotifs, Q-E-L-G-D-E-K-[IM]-Q, K-E-[FY]-[SG]-D-D-K-V-Q, and [LM]-E-D-A-D-E-K-V-[AQ], that were relatively specific for ING1/ING1L, ING1-homolog, and ING3 subfamilies, respectively. Our results shown in figure 3 extend their observations in that the ING1/ING2 (i.e., ING1L) subfamily carries the Q-E-L-G-D-[ED]-K-[ILM]-Q-[IL] motif, the ING4/ING5 (i.e., ING1 homolog) subfamily has the K-E-[FY]-[SG]-D-D-K-V-Q-L motif, the ING3 subfamily is characterized by the L-E-D-A-D-E-K-V-Q-L motif, whereas the generalized large ING subfamily shows a relatively low degree of conservation to any of the above motifs. An exception to this rule is that the three plant ING3 proteins do not bear the L-E-D-A-D-E-K-V-Q-L signature. This region also displays many other conserved residues, signifying its potentially important role. In fact, we have hypothesized that this region of ING proteins involves binding of HAT, HDAC, MYC, and other cell cycle–related proteins (Helbing et al. 1997), whereas the unique regions of each subfamily member have been suggested to modulate interactions (Kawaji et al. 2002). Kuzmichev et al. (2002) also identified the N-terminal 125 amino acids of p33ING1b, which includes this distinct conserved region, as a motif linked to the Sin3/HDAC complex through direct interaction with SAP30. Therefore, it is becoming increasingly clear that this region of the ING proteins may play a critical role in binding HAT/HDAC complex during chromatin remodeling and regulation of gene expression; hence, the name we propose is the potential chromatin regulatory (PCR) domain (figs. 2 and 3).
The criteria for defining an ING sequence now can include sequence conservation in the first three regions and partially in region IV. Similar to the PCR domain, the PIM (fig. 3, region IV) is also specific for different subfamilies. The distinction between ING1/ING2, ING4/ING5, ING3, and ING subfamilies can be made based on the length of this C-terminal conserved region, because this region in the ING1/ING2 subfamily is considerably longer than the others. The preferential and highly conserved basic residues are indicative of important functions. An additional distinct feature of the ING3 protein is the two insertions of 102 and 54 amino acids located between the PCR domain and the NLS/NTS region (fig. 2A). Although the NLS/NTS region is not as highly conserved across different ING members as the other three defined regions, several lysine residues are preserved (data not shown). As we have postulated before, ING1 and ING2, which have a distinctive NLS/NTS region, are most likely translocated to the nucleolus more efficiently in response to DNA damage but experimental proof of this remains to be presented.
The evolutionary distance of the ING protein members can be estimated from our phylogenetic analysis. We conclude that ING1 and ING2, and ING4 and ING5 are closely related and their chromosomal locations suggest the possibility of duplication of terminal regions of particular chromosome pairs. ING3 on the other hand, is relatively distant from the closely related ING4 and ING5 and the well-related ING 1 and ING2. This may be reflected by ING3s atypical (nontelomeric) location on chromosome 7 and its poorly conserved NLS domain. It may also reflect a more ancestral quality of ING3 in that the distance tree grouped three plant ING3s with Danio rerio, Xenopus, mouse, rat, and human ING3 (fig. 5), which is quite different from the ING1/ING2 and ING4/ING5 group that do not contain plant sequences. The larger Drosophila ING proteins, labeled ING2_Dm, although containing the signature Q-E-L-G-D-[ED]-K-[ILM]-Q-[IL] motif, may not functionally correspond to human ING2, but this remains to be rigorously tested.
Conclusion
The ING family of PHD finger proteins is involved in a wide variety of biological processes, including growth regulation, cellular aging, DNA repair, oncogenesis, and apoptosis. In this study, the naming and classification of ING proteins have been clarified, the evolutionary relationship among the five members of the ING family has been further elucidated, and additional conserved structural features have been identified through sequence and phylogenetic analysis. However, the exact functions of these new conserved regions/motifs needs to be further examined. Nevertheless, the results outlined in this study provide us a better understanding of the ING protein family and define useful target motifs for further examination.
Acknowledgements
We thank W. Gong, X. Feng, P. Berardi, and Dr. G.E. Moorhead for helpful suggestions during the preparation of this work. This work was supported by grants from the Canadian Institutes of Health Research and the Alberta Heritage Foundation for Medical Research to K. Riabowol. The Sensen laboratory is funded by the Alberta Science and Research Authority (ASRA), Western Economic Diversification (WD), The Alberta Network for Proteomics Innovation (ANPI), Genome Canada, Genome Prairie, the Canada Foundation for Innovation (CFI), the Natural Sciences and Engineering Research Council (NSERC), Sun Microsystems Inc., and the University of Calgary. This work was also supported in part by an NSERC operating grant, an NSERC University Faculty award, and a Michael Smith Foundation for Health Research Scholar award to C.C.H. and an NSERC Canada Graduate Scholarship and Michael Smith Foundation for Health Research Award to M.J.W.
References
Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389–3402.
Aravind, L., L. M. Iyer, and E. V. Koonin. 2003. Scores of RINGS but no PHDs in ubiquitin signaling. Cell Cycle 2:123–126.
Bromidge, T., and C. Lynas. 2002. Relative levels of alternative transcripts of the ING1 gene and lack of mutations of p33/ING1 in haematological malignanceies. Leuk. Res. 26:631–635.
Chang, M. S., H. Sasaki, M. S. Campbell et al. (12 co-authors). 1999. HRad17 colocalizes with NHP2L1 in the nucleolus and redistributes after UV irradiation. J. Biol. Chem. 274:36544–36549.
Chen, B., E. I. Campos, R. Crawford, M. Martinka, and G. Li. 2003. Analyses of the tumor suppressor ING1 expression and gene mutation in human basal cell carcinoma. Int. J. Oncol. 22:927–931.[ISI][Medline]
Chen, L., N. Matsubara, T. Yoshino, T. Nagasaka, N. Hoshizima, Y. Shirakawa, Y. Naomoto, H. Isozaki, K. Riabowol, and N. Tanaka. 2001. Genetic alterations of candidate tumor suppressor ING1 in human esophageal squamous cell cancer. Cancer Res. 61:4345–4349.
Cheung, K. J. Jr, D. Mitchell, P. Lin, and G. Li. 2001. The tumor suppressor candidate p33(ING1) mediates repair of UV-damaged DNA. Cancer Res. 61:4974–4977.
Ceol, C. J., and H. R. Horvitz. 2004. A new class of C. elegans synMuv genes implicates a Tip60/NuA4-like HAT complex as a negative regulator of Ras signaling. Dev. Cell 6:563–576.
Coscoy, L., and D. Ganem. 2003. PHD domains and E3 ubiquitin ligases: viruses make the connection. Trends Cell Biol. 13:7–12.
Doyon, Y., W. Selleck, W. S. Lane, S. Tan, and J. Cote. 2004. Structural and functional conservation of the NuA4 histone acetyltransferase complex from yeast to humans. Mol. Cell Biol. 24:1884–1896.
Eisen, J. A. 1998. Phylogenomics: improving functional predictions for uncharacterized genes by evolutionary analysis. Genome Res. 8:163–167.
Felsenstein, J. 1989. PHYLIP (phylogeny inference package). Version 3.2. Cladistics 5:164–166.
Feng, X., Y. Hara, and K. Riabowol. 2002. Different HATS of the ING1 gene family. Trends Cell Biol. 12:532–538.
Garkavtsev, I., A. Kazarov, A. Gudkov, and K. Riabowol. 1996. Suppression of the novel growth inhibitor p33ING1 promotes neoplastic transformation. Nat. Genet. 14:415–420.
———. 1999. Correction and addition to ING1 isoform sequences. Nat. Genet. 23:373.
Garkavtsev, I., S. V. Kozin, O. Chernova, L. Xu, F. Winkler, E. Brown, G. H. Barnett, and R. K. Jain. 2004. The candidate tumor suppressor protein ING4 regulates brain tumor growth and angiogenesis. Nature 428:328–332.
Garkavtsev, I., and K. Riabowol. 1997. Extension of the replicative life span of human diploid fibroblasts by inhibition of the p33ING1 candidate tumor suppressor. Mol. Cell Biol. 17:2014–2019.
Gerber, D. A., S. Souquere-Besse, F. Puvion, M. F. Dubois, O. Bensaude, and C. Cochet. 2000. Heat-induced relocalization of protein kinase CK2. Implication of CK2 in the context of cellular stress. J. Biol. Chem. 275:23919–23926.
Gozani, O., P. Karuman, D. R. Jones et al. (20 co-authors). 2003. The PHD finger of the chromatin-associated protein ING2 functions as a nuclear phosphoinositide receptor. Cell 114:99–111.
Gunduz, M., M. Ouchida, K. Fukushima, H. Hanafusa, T. Etani, S. Nishioka, K. Nishizaki, and K. Shimizu. 2000. Genomic structure of the human ING1 gene and tumor-specific mutations detected in head and neck squamous cell carcinomas. Cancer Res. 60:3143–3146.
Gunduz, M., M. Ouchida, K. Fukushima, S. Ito, Y. Jitsumori, T. Nakashima, N. Nagai, K. Nishizaki, and K. Shimizu. 2002. Allelic loss and reduced expression of the ING3, a candidate tumor suppressor gene at 7q31, in human head and neck cancers. Oncogene 21:4462–4470.
Hara, Y., Z. Zheng, S. C. Evans, D. Malatjalian, D. C. Riddell, D. L. Guernsey, L. D. Wang, K. Riabowol, and A. G. Casson. 2003. Ing1 and dp53 tumor suppressor gene alterations in adenocarcinomas of the esophagogastric junction. Cancer Lett. 192:109–116.
Helbing, C., C. Veillette, K. Riabowol, R. Johnston, and I. Garkavtsev. 1997. A novel candidate tumor suppressor, ING1, is involved in the regulation of apoptosis. Cancer Res. 57:1255–1258.
Hermeking, H. 2003. The 14-3-3 cancer connection. Nat. Rev. Cancer 3:931–943.
Ito, K., K. Kinjo, T. Nakazato, Y. Ikeda, and M. Kizaki. 2002. Expression and sequence analyses of p33(ING1) gene in myeloid leukemia. Am. J. Hematol. 69:141–143.
Jager, D., E. Stockert, M. J. Scanlan, A. O. Gure, E. Jager, A. Knuth, L. J. Old, and Y. T. Chen. 1999. Cancer—testis antigens and ING1 tumor suppressor gene product are breast cancer antigens: characterization of tissue-specific ING1 transcripts and a homolog gene. Cancer Res. 59:6197–6204.
Kawaji, H., C. Schonbach, Y. Matsuo, J. Kawai, Y. Okazaki, Y. Hayashizaki, and H. Matsuda. 2002. Exploration of novel motifs derived from mouse cDNA sequences. Genome Res. 12:367–378.
Klibanov, S. A., H. M. O'Hagan, and M. Ljungman. 2001. Accumulation of soluble and nucleolar-associated p53 proteins following cellular stress. J. Cell Sci. 114:1867–1873.
Krishnamurthy, J., K. Kannan, J. Feng, B. K. Mohanprasad, N. Tsuchida, and G. Shanmugam. 2001. Mutational analysis of the candidate tumor suppressor gene ING1 in Indian oral squamous cell carcinoma. Oral Oncol. 37:222–224.
Kuzmichev, A., Y. Zhang, H. Erdjument-Bromage, P. Tempst, and D. Reinberg. 2002. Role of the Sin3-histone deacetylase complex in growth regulation by the candidate tumor suppressor p33(ING1). Mol. Cell Biol. 22:835–848.
Leung, A. K., and A. I. Lamond. 2003. The dynamics of the nucleolus. Crit. Rev. Euk. Gene Expr. 13:39–54.
Loewith, R., M. Meijer, S. Lees-Miller, K. Riabowol, and D. Young. 2000. Three yeast proteins related to the human candidate tumor suppressor p33ING1 are associated with histone acetyltransferase activities. Mol. Cell Biol. 20:3807–3816.
Maestro, R., S. Paccinin, C. Doglioni, D. Gasparotto, T. Vukosavljevic, S. Sulfaro, L. Barzan, and M. Boiocchi. 1996. Chromosome 13q deletion mapping in head and neck squamous cell carcinomas: identification of two distinct regions of preferential loss. Cancer Res. 56:1146–1150.
Mi, Y., S. D. Thomas, X. Xu, L. K. Casson, D. M. Miller, and P. J. Bates. 2003. Apoptosis in leukemia cells is accompanied by alterations in the levels and localization of nucleolin. J. Biol. Chem. 278:8572–8579.
Motomura, K., I. Nishisho, S. Takai, H. Tateisi, M. Okazaki, M. Yamamoto, T. Miki, T. Honjo, and T. Mori. 1988. Loss of alleles at loci on chromosome 13 in human primary gastric cancers. Genomics 2:180–184.[CrossRef][Medline]
Nagashima, M., M. Shiseki, K. Miura, K. Hagiwara, S. P. Linke, R. Pedeux, X. W. Wang, J. Yokota, K. Riabowol, and C. C. Harris. 2001. DNA damage-inducible gene p33ING2 negatively regulates cell proliferation through acetylation of p53. Proc. Natl. Acad. Sci. USA 98:9671–9676.
Nagashima, M., M. Shiseki, R. M. Pedeux, S. Okamura, M. Kitahama-Shiseki, K. Miura, J. Yokota, and C. C. Harris. 2003. A novel PHD-finger motif protein, p47ING3, modulates p53-mediated transcription, cell cycle control, and apoptosis. Oncogene 22:343–350.
Nicholas, K. B., H. B. Nicholas, and D. W. Deerfield II. 1997. GeneDoc: analysis and visualization of genetic variation. EMBNEW.NEWS 4:14.
Notredame, C., D. Higgins, and J. Heringa. 2000. T-Coffee: a novel method for multiple sequence alignments. J. Mol. Biol. 302:205–217.
Noumann, G. S., J. J. Anderson, K. M. Wook, J. Lunec, A. G. Hall, M. M. Reid, and B. Angus. 2002. Loss of nuclear expression of the p33(ING1b) inhibitor of growth protein in childhood acute lymphoblastic leukaemia. J. Clin. Pathol. 55:596–601.
Ohmori, M., M. Nagai, T. Tasaka, H. P. Koeffler, T. Toyama, K. Riabowol, and J. Takahara. 1999. Decreased expression of p33ING1b mRNA in lymphoid malignancies. Am. J. Hematol. 62:118–119.
Oki, E., Y. Maehara, E. Tokunaga, Y. Kakeji, and K. Sugimachi. 1999. Reduced expression of p33(ING1) and the relationship with p53 expression in human gastric cancer. Cancer Lett. 147:157–162.
Ragvin, A., H. Valvatne, S. Erdal, V. Arskog, K. R. Tufteland, K. Breen, A. M. OYan, A. Eberharter, T. J. Gibson, P. B. Becker, and R. Aasland. 2004. Nucleosome binding by the bromodomain and PHD finger of the transcriptional cofactor p300. J. Mol. Biol. 337:773–788.
Saito, A., T. Furukawa, S. Fukushige, S. Koyama, M. Hoshi, Y. Hayashi, and A. Horii. 2000. p24/ING1-ALT1 an dp47/ING1-ALT2, distinct alternative transcripts of p33ING1. J. Hum. Genet. 45:177–181.
Scheel, H., and K. Hofmann. 2003. No evidence for PHD fingers as ubiquitin ligases. Trends Cell Biol. 13:285–287.
Scott, M., P. Bonnefin, D. Vieyra, F. M. Boisvert, D. Young, D. P. Bazett-Jones, and K. Riabowol. 2001a. UV-induced binding of ING1 to PCNA regulates the induction of apoptosis. J. Cell Sci. 114:3455–3462.
Scott, M., F. M. Boisvert, D. Vieyra, R. N. Johnston, D. P. Bazett-Jones, and K. Riabowol. 2001b. UV induces nucleolar translocation in ING1 through two distinct nucleolar targeting sequences. Nucleic Acids Res. 29:2053–2058.
Shinoura, N., Y. Muramatsu, M. Nishimura et al. (12 co-authors). 1999. Adenovirus-mediated transfer of p33ING1 with p53 drastically augments apoptosis in gliomas. Cancer Res. 59:5521–5528.
Shiseki, M., M. Nagashima, R. M. Pedeux et al. (11 co-authors). 2003. p29ING4 and p28ING5 bind to p53 and p300, and enhance p53 activity. Cancer Res. 63:2373–2378.
Skowyra, D., M. Zeremski, N. Neznanov, M. Li, Y. Choi, M. Uesugi, C. A. Hauser, W. Gu, A. V. Gudkov, and J. Qin. 2001. Differential association of products of alternative transcripts of the candidate tumor suppressor ING1 with the mSin3/HDAC transcriptional corepressor complex. J. Biol. Chem. 276:8734–8739.
Sugimoto, M., M. L. Kuo, M. F. Roussel, and C. J. Sherr. 2003. Nucleolar Arf tumor suppressor inhibits ribosomal RNA processing. Mol. Cell 2:415–424.
Tallen, G., I. Kaiser, S. Krabbe, U. Lass, C. Hartmann, G. Henze, K. Riabowol, and A. vonDeimling. 2004. No ING1 mutations in human brain tumors but reduced expression in high malignancy grades of astrocytoma. Int. J. Cancer 109:476–479.
Tokunaga, E., Y. Maehara, E. Oki, K. Kitamura, Y. Kakeji, S. Ohno, and K. Sugimachi. 2000. Diminished expression of ING1 mRNA and the correlation with p53 expression in breast cancers. Cancer Lett. 152:15–22.
Toyama, T., H. Iwase, P. Watson et al. (11 co-authors). 1999. Suppression of ING1 expession in sporadic breast cancer. Oncogene 18:5187–5193.
Toyama, T., H. Iwase, H. Yamashita, Y. Hara, H. Sugiura, Z. Zhang, I. Fukai, Y. Miura, K. Riabowol, and Y. Fujii. 2003. p33(ING1b) stimulates the transcriptional activity of the estrogen receptor alpha via its activation function (AF) 2 domain. J. Ster. Biochem. Mol. Biol. 87:57–63.
Vieyra, D., R. Loewith, M. Scott et al. (14 co-authors). 2002. ING1 proteins differentially regulate histone acetylation. J. Biol. Chem. 277:29832–29839.
Wagner, M. J., M. Gogel-spehar, R. C. Skirrow, R. N. Johnston, K. Riabowol, and C. C. Helbing. 2001. Expression of novel ING variants is regulated by thyroid hormone in the Xenopus laevis tadpole. J. Biol. Chem. 276:47013–47020.
Warbrick, E. 1998. PCNA binding through a conserved motif. BioEssays 20:195–199.(Gordon H.Y. He*, Caren C.)