phd gene wiki

• Research article
• Open access
• Published: 24 October 2019

Genome wide survey, evolution and expression analysis of PHD finger genes reveal their diverse roles during the development and abiotic stress responses in Brassica rapa L.

• Intikhab Alam 1 ,
• Cui-Cui Liu 1 ,
• Hong-Liu Ge 1 ,
• Khadija Batool 1 ,
• Yan-Qing Yang 1 &
• Yun-Hai Lu   ORCID: orcid.org/0000-0003-2044-2511 1 , 2  

BMC Genomics volume  20 , Article number:  773 ( 2019 ) Cite this article

2746 Accesses

15 Citations

Metrics details

Plant homeodomain (PHD) finger proteins are widely present in all eukaryotes and play important roles in chromatin remodeling and transcriptional regulation. The PHD finger can specifically bind a number of histone modifications as an “epigenome reader”, and mediate the activation or repression of underlying genes. Many PHD finger genes have been characterized in animals, but only few studies were conducted on plant PHD finger genes to this day. Brassica rapa (AA, 2n = 20) is an economically important vegetal, oilseed and fodder crop, and also a good model crop for functional and evolutionary studies of important gene families among Brassica species due to its close relationship to Arabidopsis thaliana.

We identified a total of 145 putative PHD finger proteins containing 233 PHD domains from the current version of B. rapa genome database. Gene ontology analysis showed that 67.7% of them were predicted to be located in nucleus, and 91.3% were predicted to be involved in protein binding activity. Phylogenetic, gene structure, and additional domain analyses clustered them into different groups and subgroups, reflecting their diverse functional roles during plant growth and development. Chromosomal location analysis showed that they were unevenly distributed on the 10 B. rapa chromosomes. Expression analysis from RNA-Seq data showed that 55.7% of them were constitutively expressed in all the tested tissues or organs with relatively higher expression levels reflecting their important housekeeping roles in plant growth and development, while several other members were identified as preferentially expressed in specific tissues or organs. Expression analysis of a subset of 18 B. rapa PHD finger genes under drought and salt stresses showed that all these tested members were responsive to the two abiotic stress treatments.

Conclusions

Our results reveal that the PHD finger genes play diverse roles in plant growth and development, and can serve as a source of candidate genes for genetic engineering and improvement of Brassica crops against abiotic stresses. This study provides valuable information and lays the foundation for further functional determination of PHD finger genes across the Brassica species.

Zinc finger proteins are abundantly present in both prokaryotic and eukaryotic genomes, including the plant kingdom [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 ]. They are characterized by the presence of one or more sequence motifs in which cysteines and/or histidines coordinate one or more zinc atoms to form stable local peptide structures (zinc fingers, ZFs) that are required for their specific functions [ 2 , 4 ]. The zinc finger was first identified in Xenopus laevis transcription factor IIIA (TFIIIA) in 1985 [ 9 ], and the three dimensional solution structure of a single zinc finger was first reported in 1989 [ 10 ]. Since then, various other zinc binding motifs have been identified and characterized, and as high as 30 types of Zinc finger proteins were currently identified in human genome based on the zinc-finger domain structure [ 11 , 12 ]. The most common types of zinc finger proteins include C2H2, RING ( really interesting new gene ), PHD ( plant homeodomain ), and LIM ( Lin-ll, Isl-1 and Mec-3 ) families [ 2 , 12 , 13 ]. These varied zinc finger domains enable different proteins to interact specifically with cognate DNA, RNA, proteins, lipids (or membrane), and small molecules through hydrogen bonds and hydrophobic interactions [ 14 , 15 , 16 ]. Proteins containing zinc finger domain (s) were found to play important roles in various molecular, physiological and cellular processes in cells or tissues, and some of them may function as part of a large regulatory network that senses and responds to different environmental stimuli, and regulate different signal transduction pathways and controlling processes, such as development and programmed cell death [ 2 , 3 , 4 , 5 , 6 , 7 , 8 , 12 , 17 , 18 , 19 , 20 , 21 ].

The PHD finger was first identified in Arabidopsis thaliana transcription factor HAT3.1 (a homeodomain-containing protein) and its maize homolog Zmhox1a in 1993 [ 22 ]. Since then, many other PHD-finger proteins have been identified in various eukaryotes, including the yeast [ 23 , 24 ], Drosophila [ 25 , 26 ] and human [ 12 , 27 , 28 ]. The PHD finger can be defined as a Cys-rich domain of approximately 50~80 amino acids with spatially conserved 8 metal ligands arranged as unique Cys4-His-Cys3 pattern in 4 pairs which can chelate two Zn 2+ atoms and form a cross-brace structure [ 13 , 29 , 30 ]. The PHD finger can specifically bind a number of histone modifications as an “epigenome reader”, and mediate the activation or repression of underlying genes [ 30 , 31 , 32 , 33 , 34 , 35 , 36 ]. In human, mutations in PHD fingers or deletions of these domains are linked to a number of diseases such as cancer, mental retardation, and immunodeficiency [ 32 , 33 ]. In plant, the PHD domains were found to be involved in the transcriptional regulation of developmental processes such as meiosis and postmeiotic events during pollen maturation, embryo meristem initiation and root development, germination, flowering time, etc. [ 36 ].

The Brassicaceae or Cruciferae is one of the most important families of flowering plants, containing some 338 genera and approximately 3709 species, with an extreme high level of morphological diversity [ 37 , 38 ]. The family includes a number of economically important species of the genus Brassica cultivated worldwide as vegetables, oil seed crops, condiments and fodder crops, as well as the extensively studied model plant Arabidopsis thaliana [ 39 ]. The genomic relationships among the six cultivated Brassica species, including B. rapa (2n = 20, AA genome, 529 Mb genome size), B. nigra (2n = 16, BB, 632 Mb), B. oleracea (2n = 18, CC, 696 Mb), B. juncea (2n = 36, AABB, 1068 Mb), B. napus (2n = 38, AACC, 1132 Mb) and B. carinata (2n = 34, BBCC, 1284 Mb), has long been established as the Triangle of U [ 40 , 41 ]. Previous studies revealed that all the species of the tribe Brassiceae shared a common whole-genome triplication (WGT) event occurred ~ 15.9 million years ago (MYA) just after the divergence of their ancestor from that of A. thaliana (tribe Arabideae) [ 42 , 43 , 44 , 45 ] . This whole genome triplication event was followed by genome diploidization involving substantial genome reshuffling and gene losses in duplicated genomic blocks, and resulted in three subgenomes with different degree of gene losing, e.g. least fractionized (LF), moderately fractionized (MF1) and most fractionized (MF2) subgenomes [ 46 , 47 ]. B. rapa is an important, worldwide cultivated crop with various morphotypes, such as leafy vegetables, turnips and oilseed rape [ 38 , 48 ]. Because of its smallest genome size of the genus Brassica , rapid life cycle, high morphological diversity, and origin from a common hexaploid ancestor as all other members of the tribe Brassiceae, B. rapa became a model plant for genetic, genomic and evolutionary studies in Brassica species [ 47 , 49 ]. The complete sequencing of the B. rapa genome makes it possible to analyze some important gene families at a whole genome level [ 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 ].

Abiotic stresses, especially salt and drought stresses, affect many aspects of plant physiology and metabolism, and cause severe crop yield losses around the world [ 57 ]. Brassica crops are mainly grown in arid and semiarid areas, and they are the most affected by drought and salinity among the major food crops [ 58 ]. Several drought or salt-tolerant genes isolated in A. thaliana as well as in Brassica crops showed great potential for genetic improvement of plant tolerance [ 58 ]. In several previous studies [ 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 ], some PHD finger genes were found to be highly responsive to abiotic stresses, including salt and drought stresses, suggesting that they might play important roles for the response and adaptation to abiotic stresses in plants. In the current paper, we reported the identification and comprehensive analysis of the PHD finger genes in B. rapa genome. Our results lay a foundation for further functional characterization of PHD finger genes among Brassica species.

Identification and characterization of PHD finger genes in B. rapa

A total of 145 non-redundant predicted PHD finger proteins containing 233 PHD finger domains were identified from the Brassica Database (BRAD), of which 92 (65.7%), 27 (18.6%), 17 (11.7%) and 9 (6.2%) contain one, two, three and four putative PHD domain (s), respectively (Additional file  2 : Table S1). The details of these B. rapa PHD protein genes, such as locus name, chromosome location, CDS and amino acid lengths, protein masse, and isoelectric points (pIs), were summarized in Additional file 2 : Table S1. We also identified 16 PHD-suspected domain-containing proteins that contain each an imperfect PHD motif (Additional file 2 : Table S2).

The identified 233 B. rapa PHD domains were extracted from their corresponding protein sequences, and archived in Additional file 2 : Table S1. Based on these domain sequences, a multiple sequence alignment, a sequence logo of the over-represented residues and a phylogenetic tree were generated and illustrated in Additional file  1 : Figure S1, S2 and S3, respectively. Additional file 1 : Figures S1 and S2 illustrated the conservation of eight metal ligands as well as the spacing between them, while Additional file 1 : Figure S3 showed the close evolutionary inter-relationships among some of these PHD domains and their multiplication during the B. rapa genome evolution.

In order to gain a global idea about the function of the B. rapa PHD finger genes, we retrieved their associated Gene Ontology (GO) terms from Phytozome database, and performed the prediction of subcellular localization by CELLO software (Additional file 2 : Table S1). The GO Molecular Function term is available for 138 out of 145 (95%) identified PHD finger proteins, which can be classified into protein-binding activity (126/138 = 91.3%), proteins-disulfide reductase activity (20/138 = 14.4%), and transcription cofactor activity (10/138 = 7.2%) (Fig.  1 a). Ninety-eight out of 145 (67.6%) B. rapa PHD finger proteins were predicted to be localized in nucleus, 25 (17.2%) in extracellular, 10 (6.9%) in plasma membrane, 7 (4.8%) in cytoplasm, and 5 (3.5%) in chloroplast (Fig.  1 b). The distribution of 145 B. rapa PHD proteins containing one to four PHD domain (s) in different cellular compartments was summarized in Table  1 . We observed that near 80% of 1-PHD domain-containing B. rapa PHD finger proteins were localized in nucleus, while only 12.0% in extracellular, 7.6% in cytoplasm, 2.2% in chloroplast, and 0% in plasma membrane. Among both the 2- and 3-PHD domain-containing proteins, a little more than 50% were localized in nucleus, about 20–30% in extracellular, about 20% in plasma membrane, 11 or 0% in chloroplast, and 0% in cytoplasm. Among the nine 4-PHD domain-containing proteins, three (33%) were localized in nucleus, four (44%) in extracellular, and two (22%) in plasma membrane.

Distribution of GO molecular function terms ( a ) and sub-cellular localization ( b ) of 145 Brassica rapa PHD finger proteins

Phylogenetic and gene structure analyses of B. rapa PHD finger proteins

To gain insights into the evolutionary relationships among B. rapa PHD finger proteins, a phylogenetic tree was generated based on the sequences of 145 B. rapa as well as 97 A. thaliana PHD finger proteins (Fig.  2 ). The result showed that these PHD finger proteins could be divided into six major groups, named A, B, C, D, E and F, within which the orthologous or homologous proteins from B. rapa and A. thaliana were closely clustered together (Fig.  2 ). The largest group A contains 51 B. rapa and 30 A. thaliana PHD finger genes. The group B contains 9 B. rapa and 7 A. thaliana PHD finger genes. The group C contains 27 B. rapa and 26 A. thaliana PHD finger genes. The group D contains 9 B. rapa and 7 A. thaliana PHD finger genes. The group E contains 30 B. rapa and 15 A. thaliana PHD finger genes. The group F contains 19 B. rapa and 12 A. thaliana PHD finger genes. Obviously, groups A, C, E and F can be further divided into several subgroups. An enlarged phylogenetic tree including PHD finger proteins from A. thaliana , B. rapa , Oryza sativa , Populus trichocarpa and Zea mays was also generated and presented in Additional file 1 : Figure S4, showing the orthologous relationships and high evolutionary conservation between PHD finger proteins of different species. The Gene Ontology terms associated with these 97 A. thaliana PHD finger genes were summarized in Additional file 2 : Table S3, showing the rich information concerning their functions in A. thaliana that can be used to explore the functions of their corresponding orthologs in Brassica crops.

Phylogenetic tree analysis of the PHD finger proteins from Arabidopsis and Brassica rapa . The tree was determined by using MEGA6.06 software with the Neighbor–Joining (NJ) algorithm and a bootstrap analysis of 1000 replicates. The PHD finger proteins were clustered into six major groups (A-F). Proteins from Arabidopsis and B. rapa are indicated by red and green colors, respectively

Intron loss or gain is another important evolutionary mechanism that generates gene structural diversity and complexity, and contributes to the functional diversity and divergence during the evolution of multi-gene families in plants [ 68 , 69 ]. To obtain insights into the structural variation of B. rapa PHD-finger protein genes, we analyzed their exon/intron organization from the genomic sequences of individual B. rapa PHD-finger protein genes, in relation to their phylogenic tree by groups extracted from the Fig.  2 (Fig.  3 ). The result showed that the most closely related members tended to be clustered together and shared similar exon/intron structures but with exceptions, such as between Bra036726 and Bra026189 in group A, between Bra02633 and Bra07814 in group B, between Bra030481 and Bra035110 in group C, between Bra020445 and Bra035572 in group D, between Bra026708 and Bra016634 in group E, and between Bra029401 and Bra009752 in group F. Among the identified 145 B. rapa PHD finger proteins, 26 (17.9%) (including 19 in group A, one in group C, four in group E, and six in group F) possess each 0 intron, 15 (10.3%) (five in group A, one in group B, one in group C, five in group E, three in group F) contain each 1 intron, while the remaining 104 contain each 2–30 introns (Bra035110 in group C contains 30 introns, Bra007814 in group B contains 28 introns). Bra036288 in group A is the longest gene covering a genomic sequence as long as ~ 21 kb, contrasting to the shortest member Bra006562 covering a genomic sequence of ~ 0.2 kb (Fig. 3 ).

Phylogenetic relationships and gene structure of PHD finger genes in Brassica rapa . The tree was generated by using Neighbor-Joining method with 1000 bootstrap replicates. The tree shows six major phylogenetic groups (group A to F) indicated with six differently colored backgrounds. Green color boxes represent exons and grey color lines indicate introns, and the untranslated regions (UTRs) are represented by blue color boxes. The sizes of exons and introns can be estimated using the scale at the bottom. The tightly clustered genes with remarkable differences in their gene structures are indicated by red stars

Previous studies have shown that tandem PHD fingers can fold as one functionally cooperative unit and be used to read more complex combinations of histone modifications, thus reinforcing the notion that the unequal numbers of PHD-finger domains detected in each protein sequence may contribute to their functional diversity and complexity [ 35 , 70 ]. The distribution of 145 B. rapa PHD proteins containing one to four PHD domain (s) in different phylogenetic groups of Fig.  2 was summarized in Table  2 . We observed that the proportion of 1-PHD domain-containing proteins was very high in group D (88.9%) and C (85.2%), followed by F (63.2%), E (60.0%), A (54.9%) and B (33.3%). The twenty-seven 2-PHD domain-containing proteins were distributed into group A (15.7%), B (44.4%), C (14.8%), E (13.3%) and F (36.8%), while the 17 3-PHD domain-containing proteins into group A (16.7%), D (11.1%) and E (23.3%), and the nine 4-PHD domain-containing proteins into A (11.7%), B (22.2%) and E (3.3%).

Additional domain analysis outside of PHD finger domain

For each of the identified 145 PHD finger protein, the presence or not of any additional known domain outside of the PHD finger domain (s) was inspected through the Smart analysis. A total of 56 additional known domains were identified, allowing classifying the 145 PHD finger proteins into 28 groups and subgroups (Additional file 2 : Table S4). The largest group (group 1) includes 42 members (42/145 = 29.0%) which all contain no other additional domain besides the 1–4 PHD domain (s). The second group includes 15 members (15/145 = 10.3%) which contain each a DUF3594 domain besides a single PHD domain. The third groups includes 9 members (9/145 = 6.2%) which contain each a KAT11, a ZnF_ZZ and a ZnF_TAZ domain besides a single PHD domain. The other 25 groups include each 1–8 members with 1–5 additional known domains. These additional known domains may be involved in protein-protein interaction (Ald_Xan_dh_C2, Coiled-coil, JmJC, PWWP), protein binding (GYF, SWIB), histone binding/acetylation/methylation (BAH, Cohisin Heat, DUF295, DUF1086, DUF3594, DUF1087, ING, KAT11, NIPPED-B_C, Post-SET, SET, SRA), nucleic acid binding (AAA, Ald_Xan_Dh_C2, ARID, AT hook, DDT, DEXDc, DNMT1, HELICc, Helicase_C_4, MBD, PLUS3, PPR, Res III, SANT, SAP, WHIIM, Znf-C2H2, Znf-C5HC2), and Zinc ion binding (Znf-C2H2, Znf-C5HC2, Znf-CCCH, Znf-TAZ, Znf-UBA, Znf-ZZ). Other known domains such as AMP-Binding (catalytic activity), C1 (intracellular signal transduction), DYW_deaminase, ELM2, EloA-BP1, FLU-1, FYRC, FYRN, JAS (jasmonate signaling), MBOAT_2, Oberon_cc and Transmembrane, were also detected.

Table  3 summarized the distribution of 56 additional known domains in the different phylogenetic groups of Fig.  2 . We observed that 46 out of 56 (82.1%) additional domains were specific to a single group, i. e., 7 were specific to group A, 3 to group B, 19 to group C, 6 to group D, 5 to group E, and 6 to F; five out of 56 (8.9%) were simultaneously present in two groups; four out of 56 (7.1%) were simultaneously present in three groups; and one (Coiled-coil domain) out of 56 (1.8%) was simultaneously present in five (A, B, C, E and F) of six groups. The group C contains as high as 26 types of additional domains, compared to a value of 11, 7, 8, 11 and 10 for group A, B, D, E and F, respectively. It is worth to remark that, among the 51 members of group A, 15 (29.4%) shared DUF3594 domain known for involving in histone binding and regulation of transcription activities, and eight (15.7%) shared the C1 domain known for intracellular signal transduction activity. Among the nine members of group B, five (55.6%) shared JAS domain known for jasmonate signaling activity. Group C (27 members) contains very diverse additional domains, including Coiled-coil (14.8%) and PWWP (14.8%) known for protein-protein interaction activity, Post-SET (14.8%) and SET (14.8%) for histone methyltransferase activity, and SAP (11.1%) for DNA-binding involved in chromosomal organization. Group D (9 members) contains additional domains such as GYF (44.4%) and SWIB (66.7%) known for protein binding activity, and PLUS3 (44.4%) for DNA binding activity. Group E (30 members) contains additional domains such as DDT (20.0%) known for DNA binding activity, KAT11 (30.0%) for histone acetylation activity, and Znf-TAZ (30.0%) and Znf-ZZ (30.0%) for zinc ion binding activity. Group F (19 members) contains additional domains such as HOX (21.1%) and Znf-C5HC2 (5.3%) known for DNA binding activity, and JmjC (5.3%) for demethylase activity.

Chromosomal distribution, gene duplication and syntenic relationships

Based on the chromosome location data of each identified PHD finger gene retrieved from BRAD database (Additional file 2 : Table S1), 140 out of 145 (96.6%) B. rapa PHD finger genes were mapped into the 10 chromosomes of B. rapa (Fig.  4 ) , while the remaining 5 PHD genes were not mapped to a specific chromosome because they were currently assigned to isolated scaffolds. Our results showed that these PHD finger genes were unevenly distributed across the 10 B. rapa chromosomes. The number of mapped PHD finger genes is 21 on A09, 19 on A02, 16 on A07, 15 on A01, 15 on A03, 15 on A05, 14 on A06, 10 on A08, 9 on A10, and 6 on A04. B. rapa PHD finger genes tend to be clustered in some chromosomal regions. Our mapping analysis showed also that 58 out of 140 (41.4%) B. rapa PHD finger genes were involved in segmental duplication and only two genes (1.4%) were involved in tandem duplication (Fig.  4 ).

Distribution of 140 PHD finger genes on 10 chromosomes of Brassica rapa . The 140 BrPHD genes unevenly located on each conserved collinear blocks of the chromosomes. Chromosome number (A01-A10) is indicated at the top of each chromosome. Gene name is indicated on the right side of each chromosome. The physical position (Mb) of each mapped gene is indicated on the left side of each chromosome. The genes located on duplicated chromosomal segments are framed by same colors and connected by same color lines between the two relevant chromosomes. The tandem repeated genes are marked by red color on the chromosomes

Brassica species have all undergone a whole genome triplication (WGT) event ~ 15.9 MYA following their divergence from the Arabidopsis lineage ~ 20 MYA [ 42 , 47 , 71 , 72 , 73 ]. B. rapa is considered as a paleohexaploid, and contains three subgenomes commonly called as least fractionized (LF), moderately fractionized (MF1) and most fractionized (MF2) [ 47 , 71 , 72 , 73 ]. The syntenic relationships between the PHD finger genes of B. rapa and A. thaliana were determined from BRAD database, and summarized in Additional file 2 : Table S5. Among the 145 B. rapa PHD finger genes, 59 (40.7%) were assigned on LF, 46 (31.7%) on MF1, and 40 (27.6%) on MF2. In seven cases, the three paralogous copies were simultaneously conserved on the three subgenomes LF, MF1 and MF2, while in 21 cases, only two of the three expected paralogous copies were conserved, and in 68 cases, only one of the three expected paralogous copies was conserved. One hundred eighteen out of 145 (81.4%) B. rapa PHD finger genes had their syntenic orthologs in A. thaliana, covering 23 blocks of seven chromosomes of ancestral translocation Proto-Calepineae Karyotype (tPCK) [ 43 , 44 , 71 , 72 , 73 ]. Twenty seven out of 145 (18.6%) B. rapa PHD finger genes didn’t have their syntenic orthologs in A. thaliana, while 18 out of 97 (18.6%) A. thaliana PHD finger genes didn’t have their syntenic orthologs in B. rapa (Additional file 2 : Table S5).

Expression analysis of B. rapa PHD finger genes in different tissues

The expression patterns of individual B. rapa PHD finger genes in different tissues (callus, root, stem, leaf, flower and silique) were analyzed based on a publicly available B. rapa RNA-Seq transcriptomic dataset [ 74 ]. Except for Bra002401, Bra004233, Bra010170 and Bra021575, the expression data of 141 other B. rapa PHD finger genes were available from the dataset, of which one (Bra013261) showed an expression value of zero for all the six tissues, while the remaining 140 genes were expressed in at least one of the six tissues., A clustered heat map displaying the expression patterns of the 140 B. rapa PHD finger genes in callus, root, stem, leaf, flower and silique was generated based on their log2-transformed fragments per kilobase of transcript per million fragments mapped (FPKM) values (Fig.  5 ). The result showed that these 140 B. rapa PHD finger genes were clustered into three major groups with subgroups. The group I (biggest) includes 78 genes, which were almost all constitutively expressed in all the tested tissues with relatively higher expression levels. The group II includes 36 genes, preferentially (> 2-folds higher) expressed in one or more tissues with relatively higher expression levels. For example, Bra028465 (corresponding to Arabidopsis gene At5g40590, a cysteine/histidine-rich C1 domain family protein gene) was preferentially expressed in root, but very lowly (or not) expressed in other tested tissues; Bra025864 (corresponding to At1g20990, another cysteine/histidine-rich C1 domain family protein gene) was preferentially expressed in root, and only very lowly expressed in other tissues; Bra029401 (corresponding to At5g24330 or Arabidopsis TRITHORAX-RELATED protein 6, ATXR6) was preferentially expressed in stem than in other tissues; Bra020445 (corresponding to At5g57380 or VERNALIZATION INSENSITIVE 3, VIN3) was preferentially expressed in leaf than in other tissues. The group III includes 26 genes which were almost all very lowly (or not) expressed in the tested tissues, except for Bra028463 (corresponding to At5g40320, a cysteine/histidine-rich C1 domain family protein gene) preferentially expressed in callus but very lowly (or not) expressed in other tissues; Bra012982 (corresponding to At5g61090, a polynucleotidyl transferase gene) was preferentially expressed in silique, moderately expressed in flower, but very lowly (or not) expressed in other tissues; and Bra033990 (homologous to At2g21840, At2g21850 and At2g21830, cysteine/histidine-rich C1 domain family protein genes) was moderately but preferentially expressed in root.

Expression profile of 140 Brassica rapa PHD finger genes in different tissues revealed by clustering analysis of RNA-Seq data. The 140 genes were divided into three major groups (I-III) based on the log2-transformed fragments per kilobase of transcript per million fragments mapped (FPKM) values. The scale representing the relative signal values is shown above. The tissue types are indicated on the top. The individual gene names are indicated on the right side

To obtain information about the variation in expression pattern among triplicated PHD finger gene members caused by WGT [ 42 , 71 ], we compared the expression levels (FPKM values) of six sets of three triplicated members that were well conserved across the three subgenomes (LF, MF1 and MF2) of B. rapa in different tissues (Fig.  6 , Additional file 2 : Table S5). The results showed that these triplicated members display different expression patterns between them. For four of six triplet sets, two members maintained relatively higher expression levels while the third one was significantly lowly expressed in the tested tissues. For one triplet set, one member showed a dominant high expression level over two other members in all tested tissues, while for another triplet set, one member was dominantly expressed over two other members in some tissues but not in others (Fig.  6 ).

Comparison of the expression levels (by FPKM values) in different tissues between the triplicated Brassica rapa PHD finger gene members conserved across the three subgenomes LF, MF1 and MF2

Table  4 summarized the distribution of 140 PHD finger genes in different expression groups of Fig.  5 in relation to the phylogenetic classification of their encoded proteins in Fig.  2 . We observed that 42.9% of genes in phylogenetic group A, 88.9% in group B, 70.4% in group C, 55.6% in group D, 56.7% in group E and 50.0% in group F were clustered into the expression group I (constitutively expressed in almost all the tested tissues). About 20% of genes in phylogenetic group A, 10% in group B, and 30% in group C, D, E and F were clustered into the expression group II (preferentially expressed in some tissues). About 40% of genes in phylogenetic group A, 0% in group B and C, 10% in group D and E, and 20% in group F were clustered into the expression group III (very lowly or not expressed in almost all the tested tissues).

Expression analysis of B. rapa PHD finger genes under salt and drought stresses

In order to relate our results with the existing data from other species, we generated a phylogenetic tree by using the protein sequences of 145 B. rapa PHD finger genes together with those of a few previously reported as stress- or development-related in maize [ 64 ], poplar [ 65 ], soybean [ 61 , 62 ], alfalfa [ 60 ], Arabidopsis [ 75 , 76 , 77 , 78 , 79 , 80 ] and rice [ 81 , 82 , 83 ] (Additional file 1 : Figure S5). We found that 18 B. rapa PHD finger genes were clustered together with those previously characterized as stress-related, while 63 others were clustered together with those previously reported as development-related. Genes closely clustered together in a phylogenetic tree may share common ancestors, and their functions may be conserved across species. Based on this phylogenetic tree (Additional file 1 : Figure S5), we selected nine genes (Bra001393, Bra016698, Bra017415, Bra026210, Bra026825, Bra034169, Bra034860, Bra034950 and Bra036568) representing the “stress-related”, and nine other genes (Bra007814, Bra015682, Bra020249, Bra020856, Bra026192, Bra027574, Bra037238, Bra037299, Bra040028) representing the “development-related” or non-characterized genes for qRT-PCR analysis to examine their expression response to salt (200 mM NaCl) (Fig.  7 ) and drought (20% (w/v) PEG 6000 ) (Fig.  8 ) stresses in the leaves of three-week-old seedlings. Our results showed that all the selected 18 B. rapa PHD finger genes were responsive to the two abiotic stress treatments.

qRT-PCR expression patterns of 18 Brassica rap PHD finger genes under salt treatment. The time points represent by x-axis and the scale of relative expression showed by y-axis. Statistical significance of deference’s between control and treated groups was analyzed using Student’s t-test (*indicates P  < 0.05, ** indicates P  < 0.01). The “stress-related” genes (see the text) are framed by red box

qRT-PCR expression patterns of 18 Brassica rap PHD finger genes under drought treatment. The time points represent by x-axis and the scale of relative expression showed by y-axis. Statistical significance of deference’s between control and treated groups was analyzed using Student’s t-test (*indicates P  < 0.05, ** indicates P  < 0.01). The “stress-related” genes (see the text) are framed by red box

For salt stress analysis, all the 18 tested genes were responsive to the treatment with 11 PHD finger genes up-regulated and seven genes down-regulated compared to control (CK) after 1 h, 3 h or 24 h of treatment, respectively (Fig.  7 ). The most spectacular case is the gene Bra026210 (corresponding to At1g14510 or ALFIN-LIKE 7, AL7, involved in covalent chromatin modification or regulation of transcription) which was induced by more than 18 fold under salt treatment at 1 h. Interestingly, Bra016698, a paralogous copy of Bra026210 produced by WGT (Additional file 2 : Table S5), was progressively induced along with the time under salt treatment and reached as high as nine fold of the control at 24 h, while another paralogous copy Bra026825 was down-regulated by more than two fold. Another gene, named Bra007814 (corresponding to At2g25170 or CYTOKININ-HYPERSENSITIVE 2, CKH2, involved in covalent chromatin modification and negative regulation of transcription) was up-regulated by four fold at 1 h of treatment but significantly down-regulated at 3 and 24 h.

For drought stress analysis, all the 18 tested genes were responsive to treatment with 16 genes up-regulated and two down regulated compared to control (CK) after 1 h, 3 h or 24 h of treatment, respectively (Fig.  8 ). Globally, the variation extents induced by drought stress were more spectacular than salt stress. Interestingly, Bra026210 was also highly induced by drought stress as it was the case for salt stress (Fig.  7 ), and reached an expression level of as high as 55 times compared to the control at 1 h, followed by an expression level of about 15 times of the control at 3 or 24 h. Bra037238 (corresponding to At2g18090, involved in regulation of transcription), Bra015682 (corresponding to At1g77250, involved in regulation of transcription) and Bra034860 (corresponding to At3g11200 or ALFIN-LIKE 2, AL2, involved in covalent chromatin modification or regulation of transcription) were induced by about 13, seven and five fold, respectively, compared to control at 1 h of treatment. Bra017415 (corresponding to At2g02470 or ALFIN-LIKE 6, AL6, involved in covalent chromatin modification or regulation of transcription) and Bra016698 (corresponding to At1g14510, or ALFIN-LIKE 7, AL7, involved in covalent chromatin modification or regulation of transcription) were induced by about eight and seven fold, respectively, compared to control at 3 h of treatment. It is worth to note that the three triplicated paralogous genes, Bra016698, Bra026210 and Bra026825, display different expression patterns along with the time of treatments of both salt and drought stress (Figs.  7 ) and ( 8 ).

PHD fingers can specifically recognize various histone marks or post-translational histone modifications (PTMs) such as trimethylated Lysine 4 in histone H3 (H3K4me3), trimethylated Lysine 9 in histone H3 (H3K9me3), trimethylated Lysine 36 in histone H3 (H3K36me3), acetylated Lysine 9 in histone H3 (H3K9ac), acetylated Lysine 14 in histone H3 (H3K14ac) , etc. , as well as unmodified histone tails such as H3K4, and other non-histone proteins [ 34 , 35 , 36 , 84 ] . For example, all the PHD domains of the seven Arabidopsis Alfin1-like proteins (AL1 to AL7) can bind to the histone H3K4me3 peptide with varying methylation state preference and binding affinities [ 85 ] ; rice CHD3 protein acts as a bifunctional chromatin regulator able to recognize and modulate H3K4 and H3K27 methylation over repressed or tissue-specific genes [ 86 ]; PHD finger of the SUMO ligase Siz/PIAS family in rice reveals specific binding for methylated histone H3 at lysine 4 and arginine 2 [ 87 ]. These features highlight the functional versatility of PHD fingers as epigenome readers that regulate gene expression (activation or repression) according to the status of the chromatin, and reinforce the hypothesis that evolutionary changes in amino acids surrounding the eight conserved metal ligand positions on a conserved structural fold would increase the functional diversity of these PHD finger proteins [ 35 ].

More and more plant PHD-finger protein genes have been identified as involved in various important biological processes. For examples, in the model plant A. thaliana , MMD1 (AT1G66170), SCC2 (AT5G15540), MS1 (AT5G22260) and ASHR3 (AT4G30860) are involved in the meiosis and post-meiotic processes, and their mutations can cause male sterility [ 36 ]; OBE1 (AT3G07780), OBE2 (AT5G48160) and PKL (AT2G25170) are involved in the embryonic meristem initiation and root development, and their mutations can result in an absence of root and defective development of the vasculature [ 36 ]; AL6 (AT2G02470) and AL7 (AT1G14510) are involved seed germination, and their double mutation can result in a germination delay under osmotic stress conditions [ 36 ]; VIL1 (AT3G24440), VRN2 (AT4G16845), VIN3 (AT5G57380), VRN5 (AT3G24440), ATX1 (AT2G31650), EBS (AT4G22140) and SHL (AT4G39100) are involved in the control of flowering time [ 36 ]; GSR1 (AT3G27490) is involved in auxin-mediated seed dormancy and germination [ 88 ]; EDM2 (AT5G55390) is involved the resistance to downy mildew [ 89 ]; AL5 (AT5G20510) is involved in abiotic stress tolerance [ 63 ]. In rice, Ehd3 acts as a promoter in the unique genetic pathway responsible for photoperiodic flowering [ 81 ]; PTC1 is involved in tapetal cell death and pollen development [ 83 ]; OsVIL2 is involved in the control of flowering time, and its insertion mutations cause late flowering under both long and short days [ 90 ]; OsTTA is a constitutively expressed regulator of multiple metal transporter genes responsible for essential metals delivery to shoots for their normal growth [ 91 ]; OsMS1 functions as a transcriptional activator to regulate programmed tapetum development and pollen exine formation [ 92 ]. In barley, HvMS1 silencing and overexpression can result in male sterility [ 93 ]. In maize, the mutation of ZmMs7 (ortholog of PTC1 ) can result in male sterility [ 94 ]. In soybean, all six Alfin1-type PHD finger genes were found to be responsive to various stress treatments, and overexpressing the GmPHD2 showed salt tolerance when compared with the wild type plants [ 61 ]; GmPHD5 encodes an important regulator for crosstalk between histone H3K4 di-methylation and H3K14 acetylation in response to salinity stress [ 62 ]. In alfalfa, Alfin1 is involved in salt tolerance [ 59 , 60 ]. In cassava, MePHD1 is involved in starch synthesis [ 95 ]. Although the number of identified PHD-finger genes is increasing in different species, most of putative PHD finger genes remain to be characterized, and no any research on this category of genes has been conducted in Brassica species to this day .

B. rapa (AA genome) is not only an economically important vegetal, oilseed and fodder crop widely grown around the world, but also one of the diploid progenitor parents of amphidiploid oilseed crops B. napus (AACC) and B. juncea (AABB), and can be used as a model plant for functional and evolutionary studies of important gene families among Brassica species [ 49 ]. In this study, a total of 145 PHD finger proteins containing 233 PHD domains were identified from the current version of the B. rapa proteome database (Additional file 2 : Table S1). This number is considerably higher than those previously identified in maize (67) [ 64 ], poplar (73) [ 65 ] and rice (59) [ 66 , 96 ], pear (31) [ 97 ] and moso bamboo (60) [ 67 ], although it might not yet be exhaustive as other PHD-suspected domain-containing proteins were also detected (Additional file 2 : Table S2). This is the consequence of the WGT event occurred ~ 15.9 MYA in Brassica ancestor followed by gene losing [ 42 , 47 , 71 ], while only one tandem duplication event was observed among these PHD finger genes (Fig.  4 ). Interestingly, these PHD finger genes were unevenly distributed on the 10 B. rapa chromosomes (Fig.  2 ), a phenomenon also observed in A. thaliana [ 8 ], maize [ 64 ] [ 65 ] and rice [ 66 ], implying a possible relationship between chromosomal location and their cellular functions.

Our gene ontology analysis showed that 67.7% of the identified B. rapa PHD finger proteins were predicted to be located in nucleus, and 91.3% members were putatively involved in protein binding activity (Fig.  1 ). These features support the previous findings about the main functions of these PHD finger genes as epigenomic effectors regulating gene expression in cells [ 30 , 31 , 32 , 33 , 34 , 35 , 36 ]. Based on the presence or not of additional domains (Additional file 2 : Table S4), gene structure and phylogenic analysis (Figs.  2  and 3 ; Additional file 1 : Figure S2), these B. rapa PHD finger genes can be classified into several groups and subgroups, illustrating the evolution and functional diversification of these genes in B. rapa . Our phylogenetic (Fig.  2 ) and syntenic (Additional file 2 : Table S5) analyses showed that, for the majority of B. rapa PHD finger genes, their corresponding orthologs were also found in the model plant A. thaliana , meant that the functional study of B. rapa PHD finger genes can largely benefit from the rich information available in A. thaliana (Additional file 2 : Table S3). However, as shown in Fig.  2 , the duplicated gene members in B. rapa generally evolved at different rates in comparison with their orthologs in A. thaliana , furthermore, some B. rapa finger protein genes, such as Bra038151 and Bra029800 in phylogenetic group A, Bra026192 in group B, and Bra034957 in group C, etc ., cannot find their corresponding orthologs in A. thaliana , suggesting that these genes may provide some new or specific functions for the growth and development of Brassica crops or their responses to various stresses.

Our analysis on RNA-Seq data (Fig.  5 ) showed that 55.7% of the B. rapa PHD finger genes were constitutively expressed in all the tested tissues with relatively higher expression levels, suggesting their important housekeeping roles in plant growth and development. A few PHD finger genes were also identified as preferentially expressed in specific tissues, constituting then an interesting panel of candidates for future targeted studies on the function of PHD finger genes and genetic improvement of Brassica species. Comparison of expression levels between triplicated members (Fig.  6 ) showed that in the majority of cases the three triplicated members display varied expression levels and patterns across different tissues, indicating that their biological roles may be also varied in plant growth and development, a phenomenon of neo-functionalization or sub-functionalization of duplicated genes [ 98 ]. Existence of 1–2 triplicated members that display a very low (or not) expression level contrasting to the higher expression levels of other triplicated members of the same triplet set, such as the case of Bra006562 in Fig.  6 , indicates that they may be degenerated during the evolution, a phenomenon already observed previously for RING finger protein genes [ 99 ].

In this study, we also analyzed the expression patterns of 18 B. rapa PHD finger genes in response to drought and salt stresses, of which nine have been characterized previously as “stress-related” in other plant species [ 60 , 61 , 62 , 64 , 65 ], while nine other genes representing the “development-related” or non-characterized genes (Additional file 1 : Figure S5). Our results showed that all these genes were responsive to the two abiotic stress treatments with different amplitudes and varied expression patterns: some members were highly up-regulated while others were down-regulated along with the time of treatments (Fig.  7 , Fig.  8 ). This means that some PHD finger genes may play important roles in plant adaption to adverse environmental stresses, an idea that was also supported by other studies [ 61 , 64 , 65 , 66 ]. These identified PHD finger genes can then serve as a source of candidate genes for genetic engineering and improvement of Brassica crops against abiotic stresses. Further studies extended on other PHD finger genes with more types of abiotic stress treatments would allow us to obtain a global view on the involvement of these PHD finger genes in response to abiotic stresses, and identify the most prominent ones for use as targets in genetic improvement of stress resistance in plants.

We identified a total of 145 putative PHD finger proteins containing 233 PHD domains from the current release of B. rapa genome database. These PHD finger genes were further characterized and classified into different groups or categories by analyses of gene ontology, additional domain, gene structure, synteny and phylogeny. We also analyzed the RNA-Seq data of these PHD finger genes, and found that 55.7% of them were constitutively expressed in all the tested tissues with relatively higher expression levels. Expression analysis of a subset of 18 PHD finger genes under salt and drought treatments showed that all of them were responsive to the two abiotic stresses, indicating that PHD finger genes can be a source of candidate genes for genetic improvement of Brassica crops against abiotic stresses. Our results lay the foundation for further functional determination of each PHD finger gene across the Brassica species, and may help to select the most promising gene targets for further genetic engineering and improvement of Brassica crops.

Identification and characterization of PHD finger proteins in B. rapa

To identify all B. rapa PHD finger proteins, we followed two different strategies as have been described in a previously study [ 50 ]. First, all previously identified Arabidopsis PHD finger proteins [ 96 , 100 ] were used as query sequences for BLASTp searches against the B. rapa proteome database at BRAD ( http://brassicadb.org/brad/ ). Second, all Arabidopsis PHD finger domains as well as those of maize [ 64 ] and poplar [ 65 ] were used as query sequences for BLASTp searches against the same B. rapa proteome database at BRAD. The irredundant candidate sequences were then analyzed online by SMART ( http://smart.embl-heidelberg.de ) (option Pfam) and occasionally by InterPro ( http://www.ebi.ac.uk/interpro/ ) to confirm the presence or not of PHD domains. This was followed by visual inspections based on the conservation of eight metal ligands (Cys4-His-Cys3) and the residue number between two neighboring metal ligands, especially between the 4th and 5th metal ligands where the number of residues should be four or five for a PHD domain contrasting to two or three for RING and two for LIM [ 13 ]. Those proteins that were predicted as PHD domain-containing by SMART but lacked two or more metal ligands, or those containing a sequence motif visually resembled to a PHD domain but not validated by SMART were classified as PHD-suspected domain-containing. The protein size, molecular weight (MW), and theoretical isoelectric point (pI) of each PHD finger protein were computed by using Pepstats ( http://www.ebi.ac.uk/Tools/seqstats/emboss_pepstats/ ).

For each identified B. rapa PHD finger protein, their associated Gene Ontology (GO) terms were retrieved from the Phytozome database ( http://phytozome.jgi.doe.gov/pz/portal.html ), and their subcellular localization was predicted by CELLO v2.5 software ( http://cello.life.nctu.edu.tw ).

Multiple sequence alignment, gene structure and phylogenetic analysis

The PHD finger domains were aligned by Clustal W and manually edited by BioEdit software. The sequence logo of over-represented motif among the identified PHD domains was generated by using the Web Logo software (http: // weblogo.berkeley.edu /logo.cgi). Phylogenetic trees based on B. rapa PHD finger domain sequences and the PHD finger protein sequences of B. rapa and A. thaliana were generated by using MEGA6.06 software with the Neighbor–Joining (NJ) algorithm and a bootstrap analysis of 1000 replicates. The exon/intron structure of each B. rapa PHD finger gene was generated by using the Gene Structure Display Server 2.0 ( http://gsds.cbi.pku.edu.cn/ ).

Additional domain analysis

To identify additional known domains, each predicted B. rapa PHD finger protein was analyzed by Smart (http:// smart.embl-heidelberg.de) with option Pfam. According to the presence and organization of different known domains, these B. rapa PHD finger proteins were divided into different groups. These additional domains were then used as query sequences for BLASTp searches against the NCBI database to determine if they were also present in other plant species.

Chromosome location of PHD-finger protein genes in B. rapa

For chromosome mapping of the identified B. rapa PHD finger genes, we followed the same procedure that was described in our previous study [ 50 ]. For each putative PHD-finger protein gene, their physical chromosome location data were retrieved from the BRAD database. The Map Chart 2.3v software was used for mapping analysis.

Syntenic relationships between B. rapa and A. thaliana PHD finger genes

For establishing the syntenic relationships among the identified B. rapa PHD finger genes, we followed the same procedure that was described previously [ 50 ]. The Search Syntenic Gene function of the BRAD database was used to find out the syntenic paralogs in B. rapa and orthologs in A. thaliana . The information such as gene name (s), localization on ancestral chromosome blocks of the tPCK (Translocation Proto-Calepineae Karyotype), Arabidopsis chromosomes and B. rapa LF, MF1 and MF2 subgenomes [ 43 , 44 , 45 , 72 ], as well as the possible tandem repeats in the two species, were recorded and summarized in Additional file 2 : Table S5.

Expression pattern of PHD finger genes in B. rapa

The RNA-Seq data of six tissues (callus, root, stem, leaf, flower and silique) of the B. rapa accession Chiifu-401–42 was downloaded from NCBI ( http://www.ncbi.nlm.nih.gov/geo/ ) (GEO accession GSE43245) [ 74 ]. For each identified B. rapa PHD finger gene, their expression values (Fragments Per Kilobase of exon model per Million mapped, FPKM) of were extracted from the data set. The clustering analysis was then conducted by using Cluster software v3.0 ( http://bonsai.hgc.jp/~mdehoon/software/cluster/ ) with the options of log2-transformed, Euclidean distances and the average linkage clustering method. The Java Tree view software ( http://jtreeview.sourceforge . net/) was used to generate a clustering gene expression heatmap.

Plant materials and stress treatments

For preparation of plant materials and stress treatments, we followed the same procedures that were described in our previous paper [ 50 ]. B. rapa accession Chiifu-401–42 seeds were first germinated in a Petri dish at 25 ° C, then transferred into plastic pots in a greenhouse at 22  ° C with 16/8 h for light/dark. Stress treatments were conducted on 21-days-old seedlings., The plants were irrigated with 200 mM NaCl and 20% (w/v) polyethylene glycol (PEG 6000) for salt and drought stress treatments, respectively. For each treatment, three biological replicates were prepared. The leaves from control and stressed plants were harvested in liquid nitrogen after 0, 1, 3, and 24 h of treatments, and placed at − 80 °C before RNA extraction.

RNA isolation and quantitative real-time PCR (qRT-PCR) analysis

For RNA isolation and quantitative real-time PCR (qRT-PCR) analysis, we followed the same procedures that were described in a previous study [ 50 ]. Total RNA was isolated from approximately 100 mg of the frozen leaves of each sample using an OMEGA Plant RNA extraction Kit. RNA concentrations were estimated by using a NanoDrop 2000 Spectrophotometer (Thermo Fisher Scientific, Inc., USA). First-strand cDNA was synthesized by using a TaKaRa cDNA Synthesis Kit (Dalian, China). Gene-specific primers were designed using the online Primer3Plus software ( http://www.primer3plus.com/ ). The B. rapa Actin -2 gene (XM_018658258) was used as internal reference gene. The primers used in this study were presented in Additional file 2 : Table S6. The qRT-PCR analysis was conducted on an ABI 7500 Fast Real-time PCR amplification system (Applied Biosystems, USA) in a volume of 20 μL: 2 μL cDNA template, 0.8 μL forward primers (10 μM), 0.8 μL reverse primers (10 μM), 10 μL SYBR Green PCR Master (ROX) (Roche, China), and 6.4 μL sterile water. The amplification parameters were: 95 °C for 1 min, followed by 40 cycles of 95 °C for 15 s, and 60 °C for 70 s. The 2 −ΔΔCt method [ 101 ] was usd for data analysis. The Student’s t-test was used to determine the significance of differences among relative expression levels of each tested gene at different time points of treatment (with P  < 0.05 considered as statistically significant).

Availability of data and materials

Not applicable.

Abbreviations

Fragments Per Kilobase of exon model per Million mapped

Gene Expression Omnibus

Least fractionized subgenome

Lin-ll, Isl-1 and Mec-3

Moderately fractionized subgenome

Most fractionized subgenome

Molecular weight

Million years ago

Polyethylene glycol

Plant homeodomain

Theoretical isoelectric point

quantitative real-time PCR

Really interesting new gene

RNA sequencing;

Transcription factor IIIA

translocation Proto-Calepineae Karyotype

Whole genome triplication

Ashworth A, Denny P. Zinc finger protein genes in the mouse genome. Mamm Genome. 1991;1(3):196–200.

Article   CAS   PubMed   Google Scholar  

Takatsuji H. Zinc-finger transcription factors in plants. Cell Mol Life Sci. 1998;54(6):582–96.

Takatsuji H. Zinc-finger proteins: the classical zinc finger emerges in contemporary plant science. Plant Mol Biol. 1999;39(6):1073–8.

Laity JH, Lee BM, Wright PE. Zinc finger proteins: new insights into structural and functional diversity. Curr Opin Struct Biol. 2001;11(1):39–46.

Matthews JM, Sunde M. Zinc fingers--folds for many occasions. IUBMB Life. 2002;54(6):351–5.

Gamsjaeger R, Liew CK, Loughlin FE, Crossley M, Mackay JP. Sticky fingers: zinc-fingers as protein-recognition motifs. Trends Biochem Sci. 2007;32(2):63–70.

Ciftci-Yilmaz S, Mittler R. The zinc finger network of plants. Cell Mol Life Sci. 2008;65(7–8):1150–60.

Malgieri G, Palmieri M, Russo L, Fattorusso R, Pedone PV, Isernia C. The prokaryotic zinc-finger: structure, function and comparison with the eukaryotic counterpart. FEBS J. 2015;282(23):4480–96.

Miller J, McLachlan AD, Klug A. Repetitive zinc-binding domains in the protein transcription factor IIIA from Xenopus oocytes. EMBO J. 1985;4:1609–14.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Lee MS, Gippert GP, Soman KV, Case DA, Wright PE. Three dimensional solution structure of a single zinc finger DNA-binding domain. Science. 1989;245:635–7.

Gray KA, Yates B, Seal RL, Wright MW, Bruford EA. Genenames.org: the HGNC resources in 2015. Nucleic Acids Res. 2015;43(Database issue):D1079–85.

Cassandri M, Smirnov A, Novelli F, Pitolli C, Agostini M, Malewicz M, et al. Zinc-finger proteins in health and disease. Cell Death Dis. 2017;3:17071.

Article   CAS   Google Scholar  

Kosarev P, Mayer K, Hardtke CS. Evaluation and classification of RING-finger domains encoded by the Arabidopsis genome. Genome Biol. 2002;3(4):0016.1–0016.12.

Article   Google Scholar  

Eom KS, Cheong JS, Lee SJ. Structural analyses of zinc finger domains for specific interactions with DNA. J Microbiol Biotechnol. 2016;26(12):2019–29.

Brown RS. Zinc finger proteins: getting a grip on RNA. Curr Opin Struct Biol. 2005;15(1):94–8.

Hall TM. Multiple modes of RNA recognition by zinc finger proteins. Curr Opin Struct Biol. 2005;15(3):367–73.

Hollemann T, Bellefroid E, Stick R, Pieler T. Zinc finger proteins in early Xenopus development. Int J Dev Biol. 1996;40(1):291–5.

CAS   PubMed   Google Scholar  

Klug A. Zinc finger peptides for the regulation of gene expression. J Mol Biol. 1999;293(2):215–8.

Ladomery M, Dellaire G. Multifunctional zinc finger proteins in development and disease. Ann Hum Genet. 2002;66(Pt 5–6):331–42.

Li WT, Chen WL, Yang C, Wang J, Yang L, He M, et al. Identification and network construction of zinc finger protein (ZFP) genes involved in the rice- Magnaporthe oryzae interaction. Plant Omics. 2014;7:540.

CAS   Google Scholar  

Jen J, Wang YC. Zinc finger proteins in cancer progression. J Biomed Sci. 2016 Jul 13;23(1):53.

Article   PubMed   PubMed Central   CAS   Google Scholar  

Schindler U, Beckmann H, Cashmore AR. HAT3.1, a novel Arabidopsis homeodomain protein containing a conserved cysteine-rich region. Plant J. 1993;4:137–50.

Martin DG, Baetz K, Shi X, Walter KL, MacDonald VE, Wlodarski MJ, et al. The Yng1p plant homeodomain finger is a methyl-histone binding module that recognizes lysine 4-methylated histone H3. Mol Cell Biol. 2006;26(21):7871–9.

Dul BE, Walworth NC. The plant homeodomain fingers of fission yeast Msc1 exhibit E3 ubiquitin ligase activity. J Biol Chem. 2007;282(25):18397–406.

Voas MG, Rebay I. The novel plant homeodomain protein rhinoceros antagonizes Ras signaling in the Drosophila eye. Genetics. 2003;165(4):1993–2006.

CAS   PubMed   PubMed Central   Google Scholar  

Seong KH, Tsuda M, Tsuda-Sakurai K, Aigaki T. The plant homeodomain finger protein MESR4 is essential for embryonic development in Drosophila. Genesis. 2015;53(11):701–8.

Gibbons RJ, Bachoo S, Picketts DJ, Aftimos S, Asenbauer B, Bergoffen J, et al. Mutations in transcriptional regulator ATRX establish the functional significance of a PHD-like domain. Nat Genet. 1997;17(2):146–8.

Fortschegger K, Shiekhattar R. Plant homeodomain fingers form a helping hand for transcription. Epigenetics. 2011;6(1):4–8.

Nakamura Y, Umehara T, Hamana H, Hayashizaki Y, Inoue M, Kigawa T, et al. Crystal structure analysis of the PHD domain of the transcription co-activator Pygopus. J Mol Biol. 2007;370(1):80–92.

Aasland R, Gibson TJ, Stewart AF. The PHD finger: implications for chromatin-mediated transcriptional regulation. Trends Biochem Sci. 1995;20:56–9.

Mellor J. It takes a PHD to read the histone code. Cell. 2006;126(1):22–4.

Baker LA, Allis CD, Wang GG. PHD fingers in human diseases: disorders arising from misinterpreting epigenetic marks. Mutat Res. 2008;647(1–2):3–12.

Musselman CA, Kutateladze TG. PHD fingers: epigenetic effectors and potential drug targets. Mol Interv. 2009;9(6):314–23.

Musselman CA, Kutateladze TG. Handpicking epigenetic marks with PHD fingers. Nucleic Acids Res. 2011;39(21):9061–71.

Sanchez R, Zhou MM. The PHD finger: a versatile epigenome reader. Trends Biochem Sci. 2011;36(7):364–72.

Mouriz A, López-González L, Jarillo JA, Piñeiro M. PHDs govern plant development. Plant Signal Behav. 2015;10(7):e993253.

PubMed   PubMed Central   Google Scholar  

Bailey CD, Koch MA, Mayer M, Mummenhoff K, O'Kane SL Jr, Warwick SI, et al. Toward a global phylogeny of the Brassicaceae. Mol Biol Evol. 2006;23(11):2142–60.

Cheng F, Wu J, Wang X. Genome triplication drove the diversification of Brassica plants. Hortic Res. 2014;1:14024.

Beilstein MA, Al-Shehbaz IA, Kellogg EA. Brassicaceae phylogeny and trichome evolution. Am J Bot. 2006;93(4):607–19.

Nagaharu U. Genome analysis in Brassica with special reference to the experimental formation of B. napus and peculiar mode of fertilization. Jap J Bot. 1935;7:389–52.

Google Scholar  

Johnston JS, Pepper AE, Hall AE, Chen ZJ, Hodnett G, et al. Evolution of genome size in Brassicaceae. Ann Bot. 2005;95:229–35.

Lysak MA, Koch MA, Pecinka A, Schubert I. Chromosome triplication found across the tribe Brassiceae. Genome Res. 2005;15:516–25.

Schranz ME, Lysak MA, Mitchell-Olds T. The ABC’s of comparative genomics in the Brassicaceae: building blocks of crucifer genomes. Trends Plant Sci. 2006;11:535–42.

Mandáková T, Lysak MA. Chromosomal phylogeny and karyotype evolution in x = 7 crucifer species (Brassicaceae). Plant Cell. 2008;20:2559–70.

Lysak MA, Cheung K, Kitschke M, Bures P. Ancestral chromosomal blocks are triplicated in Brassiceae species with varying chromosome number and genome size. Plant Physiol. 2007;145(2):402–10.

Mun JH, Kwon SJ, Yang TJ, Seol YJ, Jin M, Kim JA, et al. Genome-wide comparative analysis of the Brassica rapa gene space reveals genome shrinkage and differential loss of duplicated genes after whole genome triplication. Genome Biol. 2009;10(10):R111.

Wang X, Wang H, Wang J, Sun R, Wu J, Liu S, et al. The genome of the mesopolyploid crop species Brassica rapa . Nat Genet. 2011;43:1035–9.

Zhao J, Wang X, Deng B, Lou P, Wu J, Sun R, et al. Genetic relationships within Brassica rapa as inferred from AFLP fingerprints. Theor Appl Genet. 2005;110:1301–14.

Article   PubMed   Google Scholar  

Alam I, Yang YQ, Wang Y, Zhu ML, Wang HB, Chalhoub B, Lu YH. Genome-wide identification, evolution and expression analysis of RING finger protein genes in Brassica rapa . Sci Rep. 2017;7:40690.

Alam I, Batool K, Cui DL, Yang YQ, Lu YH. Comprehensive genomic survey, structural classification and expression analysis of C2H2 zinc finger protein gene family in Brassica rapa L. PLoS One. 2019;14(5):e0216071.

Article   PubMed   PubMed Central   Google Scholar  

Alam I, Cui DL, Batool K, Yang YQ, Lu YH. Comprehensive genomic survey, characterization and expression analysis of the HECT gene family in Brassica rapa L. and Brassica oleracea L. Genes (Basel). 2019;10(5):e400.

Wang F, Qiu N, Ding Q, Li J, Zhang Y, Li H, Gao J. Genome-wide identification and analysis of the growth-regulating factor family in Chinese cabbage ( Brassica rapa L. ssp. pekinensis). BMC Genomics. 2014;15:807.

Saha G, Park JI, Jung HJ, Ahmed NU, Kayum MA, Chung MY, et al. Genome-wide identification and characterization of MADS-box family genes related to organ development and stress resistance in Brassica rapa . BMC Genomics. 2015;16:178.

He Y, Mao S, Gao Y, Zhu L, Wu D, Cui Y, et al. Genome-wide identification and expression analysis of WRKY transcription factors under multiple stresses in Brassica napus . PLoS One. 2016;11(6):e0157558.

Yan C, Duan W, Lyu S, Li Y, Hou X. Genome-wide identification, evolution, and expression analysis of the ATP-binding cassette transporter gene family in Brassica rapa . Front Plant Sci. 2017;8:349.

Du J, Hu S, Yu Q, Wang C, Yang Y, Sun H, et al. Genome-Wide Identification and Characterization of BrrTCP Transcription Factors in Brassica rapa ssp rapa. Front Plant Sci. 2017;8:1588.

Zhu JK. Salt and drought stress signal transduction in plants. Annu Rev Plant Biol. 2002;53:247–73.

Zhang X, Lu G, Long W, Zou X, Li F, Nishio T. Recent progress in drought and salt tolerance studies in Brassica crops. Breed Sci. 2014;64:60–73.

Bastola DR, Pethe VV, Winicov I. Alfin1 , a novel zinc-finger protein in alfalfa roots that binds to promoter elements in the salt-inducible MsPRP2 gene. Plant Mol Biol. 1998;38(6):1123–35.

Winicov I, Bastola DR. Transgenic overexpression of the transcription factor Alfin1 enhances expression of the endogenous MsPRP2 gene in Alfalfa and improves salinity tolerance of the plants. Plant Physiol. 1999;120(2):473–80.

Wei W, Huang J, Hao YJ, Zou HF, Wang HW, Zhao JY, et al. Soybean GMPHD-type transcription regulators improve stress tolerance in transgenic Arabidopsis plants. PLoS One. 2009;4:e7209.

Wu T, Pi EX, Tsai SN, Lam HM, Sun SM, Kwan YW, et al. GmPHD5 acts as an important regulator for crosstalk between histone H3K4 di-methylation and H3K14 acetylation in response to salinity stress in soybean. BMC Plant Boil. 2011;11(1):178.

Wei W, Zhang YQ, Tao JJ, Chen HW, Li QT, Zhang WK, et al. The Alfin-like homeodomain finger protein AL5 suppresses multiple negative factors to confer abiotic stress tolerance in Arabidopsis . Plant J. 2015;81(6):871–83.

Wang Q, Liu J, Wang Y, Zhao Y, Jiang H, Cheng B. Systematic analysis of the maize PHD-finger gene family reveals a subfamily involved in abiotic stress response. Int J Mol Sci. 2015;16(10):23517–44.

Wu S, Wu M, Dong Q, Jiang H, Cai R, Xiang Y. Genome-wide identification, classification and expression analysis of the PHD-finger protein family in Populus trichocarpa . Gene. 2016;575(1):75–89.

Sun M, Jia B, Yang J, Cui N, Zhu Y, Sun X. Genome-wide identification of the PHD-finger family genes and their responses to environmental stresses in Oryza sativa L. Int J Mol Sci. 2017;18(9):2005.

Article   PubMed Central   CAS   Google Scholar  

Gao Y, Liu H, Wang Y, Li F, Xiang Y. Genome-wide identification of PHD-finger genes and expression pattern analysis under various treatments in moso bamboo ( Phyllostachys edulis ). Plant Physiol Biochem. 2018;123:378–91.

Li W, Liu B, Yu L, Feng D, Wang H, et al. Phylogenetic analysis, structural evolution and functional divergence of the 12-oxo-phytodienoate acid reductase gene family in plants. BMC Evol Biol. 2009;9:90.

Li WY, Wang X, Li R, Li WQ, Chen KM. Genome-wide analysis of the NADK gene family in plants. PLoS One. 2014;9(6):e101051.

Yap KL, Zhou MM. Keeping it in the family: diverse histone recognition by conserved structural folds. Crit Rev Biochem Mol Biol. 2010;45(6):488–505.

Liu S, Liu Y, Yang X, Tong C, Edwards D, Parkin IA, et al. The Brassica oleracea genome reveals the asymmetrical evolution of polyploid genomes. Nat Commun. 2014;5:3930.

Cheng F, Wu J, Fang L, Wang X. Syntenic gene analysis between Brassica rapa and other Brassicaceae species. Front Plant Sci. 2012;3:198.

Cheng F, Mandáková T, Wu J, Xie Q, Lysak MA, Wang X. Deciphering the diploid ancestral genome of the mesohexaploid Brassica rapa . Plant Cell. 2013;25:1541–54.

Tong C, Wang X, Yu J, Wu J, Li W, Huang J, et al. Comprehensive analysis of RNA-seq data reveals the complexity of the transcriptome in Brassica rapa . BMC Genomics. 2013;14:689.

Ito T, Nagata N, Yoshiba Y, Ohme-Takagi M, Ma H, Shinozaki K. Arabidopsis MALE STERILITY1 encodes a PHD-type transcription factor and regulates pollen and tapetum development. Plant Cell. 2007;19(11):3549–62.

Ito T, Shinozaki K. The MALE STERILITY1 gene of Arabidopsis , encoding a nuclear protein with a PHD-finger motif, is expressed in tapetal cells and is required for pollen maturation. Plant Cell Physiol. 2002;43(11):1285–92.

Wilson ZA, Morroll SM, Dawson J, Swarup R, Tighe PJ. The Arabidopsis MALE STERILITY1 ( MS1 ) gene is a transcriptional regulator of male gametogenesis, with homology to the PHD-finger family of transcription factors. Plant J. 2001;28(1):27–39.

Reddy TV, Kaur J, Agashe B, Sundaresan V, Siddiqi I. The DUET gene is necessary for chromosome organization and progression during male meiosis in Arabidopsis and encodes a PHD finger protein. Development. 2003;130(24):5975–87.

Müssig C, Kauschmann A, Clouse S, Altmann T. The Arabidopsis PHD-finger protein SHL is required for proper development and fertility. Mol Gen Genet. 2000;264(4):363–70.

Greb T, Mylne JS, Crevillen P, Geraldo N, An H, Gendall AR, et al. The PHD finger protein VRN5 functions in the epigenetic silencing of Arabidopsis FLC . Curr Biol. 2007;17(1):73–8.

Matsubara K, Yamanouchi U, Nonoue Y, Sugimoto K, Wang ZX, Minobe Y, et al. Ehd3 , encoding a plant homeodomain finger-containing protein, is a critical promoter of rice flowering. Plant J. 2011;66(4):603–12.

Ito Y, Chujo A, Eiguchi M, Kurata N. Radial axis differentiation in a globular embryo is marked by HAZ1 , a PHD-finger homeobox gene of rice. Gene. 2004;331:9–15.

Li H, Yuan Z, Vizcay-Barrena G, Yang C, Liang W, Zong J, et al. Persistent tapetal cell 1 (PTC1) encodes a PHD-finger protein that is required for tapetal cell death and pollen development in rice. Plant Physiol. 2011;156(2):615–30.

Li Y, Li H. Many keys to push: diversifying the ‘readership’ of plant homeodomain fingers. Acta Biochim Biophys Sin Shanghai. 2012;44(1):28–39.

Article   PubMed   CAS   Google Scholar  

Liang X, Lei M, Li F, Yang X, Zhou M, Li B, Cao Y, Gong S, Liu K, Liu J, Qi C, Liu Y. Family-wide characterization of histone binding abilities of PHD domains of AL proteins in Arabidopsis thaliana . Protein J. 2018;37(6):531–8.

Hu Y, Liu D, Zhong X, Zhang C, Zhang Q, Zhou DX. CHD3 protein recognizes and regulates methylated histone H3 lysines 4 and 27 over a subset of targets in the rice genome. Proc Natl Acad Sci U S A. 2012;109(15):5773–8.

Shindo H, Suzuki R, Tsuchiya W, Taichi M, Nishiuchi Y, Yamazaki T. PHD finger of the SUMO ligase Siz/PIAS family in rice reveals specific binding for methylated histone H3 at lysine 4 and arginine 2. FEBS Lett. 2012;586(13):1783–9.

Ye Y, Gong Z, Lu X, Miao D, Shi J, Lu J, et al. Germostatin resistance locus 1 encodes a PHD finger protein involved in auxin-mediated seed dormancy and germination. Plant J. 2016;85(1):3–15.

Eulgem T, Tsuchiya T, Wang XJ, Beasley B, Cuzick A, Tör M, et al. EDM2 is required for RPP7-dependent disease resistance in Arabidopsis and affects RPP7 transcript levels. Plant J. 2007;49:829–39.

Yang J, Lee S, Hang R, Kim SR, Lee YS, Cao X, et al. OsVIL2 functions with PRC2 to induce flowering by repressing OsLFL1 in rice. Plant J. 2013;73(4):566–78.

Tanaka N, Uraguchi S, Kajikawa M, Saito A, Ohmori Y, Fujiwara T. A rice PHD-finger protein OsTITANIA, is a growth regulator that functions through elevating expression of transporter genes for multiple metals. Plant J. 2018;96(5):997–1006.

Yang Z, Liu L, Sun L, Yu P, Zhang P, Abbas A, et al. OsMS1 functions as a transcriptional activator to regulate programmed tapetum development and pollen exine formation in rice. Plant Mol Biol. 2019;99(1–2):175–91.

Fernández Gómez J, Wilson ZA. A barley PHD finger transcription factor that confers male sterility by affecting tapetal development. Plant Biotechnol J. 2014;12(6):765–77.

Zhang D, Wu S, An X, Xie K, Dong Z, Zhou Y, et al. Construction of a multicontrol sterility system for a maize male-sterile line and hybrid seed production based on the ZmMs7 gene encoding a PHD-finger transcription factor. Plant Biotechnol J. 2018;16(2):459–71.

Ma P, Chen X, Liu C, Xia Z, Song Y, Zeng C, et al. MePHD1 as a PHD-finger protein negatively regulates ADP-Glucose Pyrophosphorylase Small Subunit1a gene in cassava. Int J Mol Sci. 2018;19:2831.

Feng Y, Liu QP, Xue QZ. Comparative phylogenetic analysis of the rice and Arabidopsis PHD-finger proteins. Acta Genet Sin. 2014;31(11):1284–93.

Cao Y, Han Y, Meng D, Abdullah M, Li D, Jin Q, et al. Systematic analysis and comparison of the PHD-finger gene family in Chinese pear ( Pyrus bretschneideri ) and its role in fruit development. Funct Integr Genomics. 2018;18(5):519–31.

Blanc G, Wolfe KH. Functional divergence of duplicated genes formed by polyploidy during Arabidopsis evolution. Plant Cell. 2004;16:1679–91.

Yang YQ, Lu YH. Genome-wide survey, characterization, and expression analysis of RING finger protein genes in Brassica oleracea and their syntenic comparison to Brassica rapa and Arabidopsis thaliana . Genome. 2018;61(9):685–97.

Zhang F, Yang ZN, Zhang S. Genome-wide analysis of PHD-finger protein family in Arabidopsis thaliana . Acta Bot Yunnanica. 2009;31(3):227–38.

Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the comparative CT method. Nat Protoc. 2008;3:1101–8.

Download references

Acknowledgements

We thank Dr. Xiao-Ming Wu of Oil Crop Research Institute, Chinese Academy of Agricultural Sciences, for kindly providing the plant seeds used in this study. We are grateful to the anonymous reviewers for their valuable and constructive suggestions about the manuscript, and to Dr. Anne-Marie Chèvre for handling the manuscript.

This work was supported by a startup fund for distinguished scholars of Fujian Agriculture and Forestry University, No. 114120019, awarded to YHL. The funding body played no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

Author information

Authors and affiliations.

Key Laboratory of Ministry of Education for Genetics, Breeding and Multiple Utilization of Crops, College of Crop Science, Fujian Agriculture and Forestry University, Fuzhou, 350002, China

Intikhab Alam, Cui-Cui Liu, Hong-Liu Ge, Khadija Batool, Yan-Qing Yang & Yun-Hai Lu

Marine and Agricultural Biotechnology Laboratory, Institute of Oceanography, Minjiang University, Fuzhou, 350108, China

You can also search for this author in PubMed   Google Scholar

Contributions

YHL conceived and designed the experiments; IA, CCL, HLG, KB, and YQY conducted the experiments; IA and YHL processed the data and wrote the manuscript; YHL revised the manuscript; all authors have read and approved the manuscript.

Corresponding author

Correspondence to Yun-Hai Lu .

Ethics declarations

Ethics approval and consent to participate, consent for publication, competing interests.

The authors declare that they have no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Additional file 1: fig. s1..

Multiple sequence alignment of 233 PHD domains from 145 PHD finger proteins of Brassica rapa. Fig. S2. Sequence logo of the overrepresented motif found in 233 PHD domains of Brassica rapa . Fig. S3. Phylogenetic tree based on multiple sequence alignment of 233 PHD domains from 145 putative PHD finger proteins of Brassica rapa . Fig. S4. Phylogenetic tree based on multiple sequence alignment of PHD finger proteins from Arabidopsis thaliana , Brassica rapa , Oryza sativa , Populus trichocarpa  and Zea mays. Fig. S5. Phylogenetic tree analyses of all 145 Brassica rapa PHD finger proteins and a few PHD finger proteins from other species previously characterized as stress or plant development related.

Additional file 2: Table S1.

List of 145 B. rapa PHD finger protein genes, and their related informations . Table S2 . List of 16 suspected B. rapa PHD finger proteins. Table S3. Summary of gene ontology terms of 98 A. thaliana PHD finger protein genes (retrieved from TAIR database, https://www.arabidopsis.org/index.jsp) in relation to the phylogenetic classification of their encoded proteins along with the 145 B. rapa PHD finger proteins in Fig.  2 . Table S4. Classification of145 B. rapa PHD domain-containing proteins based on the presence or not and organization of additional domain (s). Table S5. Synteny relationships between Arabidopsis and B.rapa PHD finger protein genes. Table S6. The information of primers used in the quantitative real-time PCR (qRT-PCR) analysis.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License ( http://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Cite this article.

Alam, I., Liu, CC., Ge, HL. et al. Genome wide survey, evolution and expression analysis of PHD finger genes reveal their diverse roles during the development and abiotic stress responses in Brassica rapa L.. BMC Genomics 20 , 773 (2019). https://doi.org/10.1186/s12864-019-6080-8

Download citation

Received : 24 November 2018

Accepted : 04 September 2019

Published : 24 October 2019

DOI : https://doi.org/10.1186/s12864-019-6080-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Brassica rapa
• PHD finger genes
• Gene duplication
• Genes expression
• Abiotic stress

BMC Genomics

ISSN: 1471-2164

• Submission enquiries: [email protected]
• General enquiries: [email protected]

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 20 April 2016

Hypoxia-inducible factor prolyl hydroxylase 1 (PHD1) deficiency promotes hepatic steatosis and liver-specific insulin resistance in mice

• Amandine Thomas 1 , 2 ,
• Elise Belaidi 1 , 2 ,
• Judith Aron-Wisnewsky 3 , 4 , 5 ,
• Gerard C. van der Zon 6 ,
• Patrick Levy 1 , 2 ,
• Karine Clement 3 , 4 , 5 ,
• Jean-Louis Pepin 1 , 2 ,
• Diane Godin-Ribuot 1 , 2 &
• Bruno Guigas 6 , 7  

Scientific Reports volume  6 , Article number:  24618 ( 2016 ) Cite this article

5734 Accesses

26 Citations

12 Altmetric

Metrics details

• Homeostasis
• Metabolic syndrome

Obesity is associated with local tissue hypoxia and elevated hypoxia-inducible factor 1 alpha (HIF-1α) in metabolic tissues. Prolyl hydroxylases (PHDs) play an important role in regulating HIF-α isoform stability. In the present study, we investigated the consequence of whole-body PHD1 gene ( Egln2 ) inactivation on metabolic homeostasis in mice. At baseline, PHD1−/− mice exhibited higher white adipose tissue (WAT) mass, despite lower body weight and impaired insulin sensitivity and glucose tolerance when compared to age-matched wild-type (WT) mice. When fed a synthetic low-fat diet, PHD1−/− mice also exhibit a higher body weight gain and WAT mass along with glucose intolerance and systemic insulin resistance compared to WT mice. PHD1 deficiency led to increase in glycolytic gene expression, lipogenic proteins ACC and FAS, hepatic steatosis and liver-specific insulin resistance. Furthermore, gene markers of inflammation were also increased in the liver, but not in WAT or skeletal muscle, of PHD1−/− mice. As expected, high-fat diet (HFD) promoted obesity, hepatic steatosis, tissue-specific inflammation and systemic insulin resistance in WT mice but these diet-induced metabolic alterations were not exacerbated in PHD1−/− mice. In conclusion, PHD1 deficiency promotes hepatic steatosis and liver-specific insulin resistance but does not worsen the deleterious effects of HFD on metabolic homeostasis.

Similar content being viewed by others

Adipocyte PHLPP2 inhibition prevents obesity-induced fatty liver

Partial impairment of insulin receptor expression mimics fasting to prevent diet-induced fatty liver disease

The role of adipose tissue and subsequent liver tissue hypoxia in obesity and early stage metabolic dysfunction associated steatotic liver disease

Introduction.

Obesity is believed to be associated with some degree of hypoxia in metabolic organs, notably in white adipose tissue (WAT) 1 , 2 . A significant decrease in oxygen tension (pO 2 ) was indeed reported in various fat pads from different mouse models of obesity and in subcutaneous WAT from overweight and obese subjects 3 , although this later observation still remains a matter of discussion 4 . Hypoxia is suggested to be involved in WAT inflammation 5 and to play a major role in obesity-induced metabolic dysfunctions, at least partly by impairing adipocyte insulin sensitivity 6 , 7 . At the mechanistic level, WAT hypoxia promotes the increase in both gene and protein expression of hypoxia-inducible transcription factors (HIFs) 8 . HIFs are heterodimers composed of the constitutive HIF-β nuclear subunit and of one of three isoforms of the O 2 -regulated HIF-α cytosolic subunit, namely HIF-1α, HIF-2α and HIF-3α 9 , 10 . HIF-α expression and transcriptional activity are tightly regulated by both the oxygen-sensitive factor inhibiting HIF (FIH) and members of the prolyl hydroxylase (PHD) family 9 , 10 . PHDs are evolutionary conserved dioxygenases that require oxygen and 2-oxoglutarate as co-substrates and iron and ascorbic acid as cofactors for hydroxylation of specific proline residues on HIF-α. Three PHD isoforms have been identified in mammals (PHD1, PHD2 and PHD3) and are ubiquitously expressed in most tissues 9 , 10 , although their respective roles in the control of tissue-specific levels of HIF-α isoforms remain to be clarified 11 . In normoxic conditions, PHDs act as oxygen-sensing enzymes and promote hydroxylation of HIF-α subunits, allowing recognition by the protein von Hippel-Lindau (pVhl) ubiquitin ligase complex and subsequent proteasomal degradation 9 , 10 . In contrast, PHD activity is reduced during hypoxia, leading to stabilization of HIF-α subunits and their subsequent translocation to the nucleus where they can dimerize with HIF-β, subsequently triggering the transcription of specific target genes 12 . Among the various HIFs, the HIF-1 transcription factor appears to play a major role in obesity-associated insulin resistance and metabolic dysfunctions, notably by promoting expression and secretion of chemokines/adipokines and recruitment of pro-inflammatory macrophages and T cell accumulation in hypoxic WAT 2 . In contrast, pharmacological or genetic inhibition of HIF-1α was reported to decrease high-fat diet-induced metabolic dysfunctions 13 , 14 , 15 , 16 , 17 , as shown by attenuated adipose tissue fibrosis and inflammation 13 and reduced hepatic steatosis 14 . Overall, this indicates that modulation of HIF-1 transcriptional activity in metabolic tissues might be involved in the hypoxia- and obesity-associated metabolic alterations. Various genetically engineered mouse models for whole-body and tissue-specific inactivation of the three PHD isoforms have been developed 18 , 19 , 20 but the exact impact of such deficiencies on whole-body metabolic homeostasis remains unclear 1 and no data are currently available in PHD1-deficient mice.

In the present study, we aimed to investigate the consequences of constitutive inactivation of the PHD1 gene ( Egln2 ), one of the critical component of the HIF-PHD oxygen-sensing pathway, on adiposity, whole-body glucose homeostasis and tissue specific insulin-sensitivity on both low (LFD) and high-fat (HFD) diet.

PHD1 deficiency impairs whole-body metabolic homeostasis

We first determined the effect of constitutive deletion of the PHD1 gene on expression of the other PHD isoforms and of the HIF-1 target gene Ldha in liver, epididymal white adipose tissue (eWAT) and skeletal muscle ( Fig. S1 ). As expected, the expression of the full length Egln2 mRNA coding for the active PHD1 protein was undetectable in each tissue. In the liver, an increase in the expression of Egln3 (PHD3) was observed ( Fig. S1A ), while an upregulation of Egln1 (PHD2) and a downregulation of Egln3 were evidenced in eWAT ( Fig. S1B ). No effect of Egln2 (PHD1) deletion on the expression of other PHDs isoforms was detected in skeletal muscle ( Fig. S1C ), in line with a previous report 21 . The expression of Ldha , a HIF target gene, was elevated in liver and WAT from PHD1-deficient mice, suggesting increased HIFs transcriptional activities in these tissues ( Fig. S1A,B ).

To investigate the role of PHD1 on whole-body metabolic homeostasis, 12 week-old PHD1-deficient (PHD1−/−) and WT control mice on regular chow diet were first compared. PHD1−/− mice exhibited lower body and liver weights (−8 and −18%, respectively; p < 0.05; Fig. 1A,B ) and higher epididymal and subcutaneous adipose tissue weights (+83 and +58%, respectively; p < 0.05; Fig. 1A,B ) compared to WT mice, suggesting that PHD1 controls adiposity. In link with decreased body weight, mean daily food intake was also lower in PHD1−/− compared to WT mice ( Fig. 1C ). Despite reduced body weight, PHD1 deficiency was associated with elevated fasting plasma triglycerides (TG) ( Fig. 1D ), glucose ( Fig. 1F ) and insulin ( Fig. 1G ) levels, whereas total cholesterol (TC) was unchanged ( Fig. 1E ). The calculated HOmeostasis Model Assessment of Insulin Resistance index (HOMA-IR) adjusted for rodents was increased in PHD1−/− mice compared to WT mice ( Fig. 1H ), suggesting systemic insulin resistance. To assess the effect of PHD1 deletion on whole-body glucose homeostasis and insulin sensitivity, mice were next subjected to intraperitoneal glucose (GTT) and insulin tolerance tests (ITT), respectively. In line with increased HOMA-IR, PHD1−/− mice displayed impaired whole-body glucose tolerance ( Fig. 1I–K ) and insulin sensitivity ( Fig. 1L,M ) compared to WT mice.

PHD1 deficiency induces glucose intolerance and insulin resistance in standard chow-fed mice.

Body weight ( A ) and weights of liver, epididymal white adipose tissue (eWAT) and subcutaneous white adipose tissue (scWAT) were measured at sacrifice in 12 week-old WT (open bars) and PHD1−/− (black bars) mice on standard chow diet ( B ). Daily food intake ( C ) was monitored during 5 weeks. Fasting plasma triglycerides ( D ), total cholesterol ( E ), glucose ( F ) and insulin ( G ) levels were determined in 6-hour unfed mice and HOMA-IR ( H ) was calculated. An intraperitoneal glucose tolerance test (ipGTT, 2 g/kg total body weight) was performed in 6-hour unfed mice. Blood glucose levels were measured at the indicated time-points ( I ) and the AUC of the glucose excursion curve was calculated as a surrogate for whole-body glucose tolerance ( J ). The plasma insulin level during ipGTT was measured at 15 minutes ( K ). An intraperitoneal insulin tolerance test (0.5 U/kg total body weight) was performed in 4-hour unfed mice. Blood glucose levels were measured at the indicated time-points ( L ) and the AUC of the glucose excursion curve was calculated as a surrogate for whole-body insulin sensitivity ( M ). Data are means ± SEM (n = 7 for WT; n = 13 for PHD1−/−). # p < 0.05 vs WT mice.

PHD1 deficiency promotes body weight and fat mass gain, dyslipidemia and glucose intolerance but does not worsen HFD-induced metabolic alterations

To further investigate the impact of PHD1 deletion on metabolic homeostasis, WT and PHD1−/− mice were next challenged with synthetic low- (LFD, 10% kcal from fat) or high-fat (HFD, 60% kcal from fat) diets for 12 weeks. PHD1−/− mice fed a LFD for 12 weeks displayed increased body weight gain and liver and fat mass compared to WT mice ( Fig. 2A–C ), despite similar daily food intake ( Fig. 2D ). In line with what was observed with regular chow diet, fasting plasma TG and insulin levels as well as HOMA-IR were significantly higher in LFD-fed PHD1−/− mice compared to WT mice ( Fig. 2E,H,I ), indicating impairment in whole-body metabolic homeostasis. Furthermore, intraperitoneal GTT demonstrated increased glucose intolerance and higher plasma insulin levels in LFD-fed PHD−/− mice compared to their WT counterparts ( Fig. 2J–L ). Of note, the differences in body weight gain, plasma parameters and whole-body metabolic homeostasis in between genotypes were already observed after 6 weeks of synthetic diet ( Fig. S2 ). Overall, these results demonstrate that PHD1−/− mice on LFD maintain an adverse metabolic phenotype compared to WT mice.

PHD1 deficiency promotes weight gain and insulin resistance but does not worsen high fat diet-induced metabolic alterations.

WT (open bars) and PHD1−/− (black bars) mice were fed a low-fat (LFD, 10% fat) or high-fat (HFD, 45% fat) diet for 12 weeks. Body weight was monitored throughout the experimental period ( A ). Delta (Δ) change in body weight from the start of diet ( B ), weight of liver, epididymal (eWAT) and subcutaneous (scWAT) white adipose tissue and skeletal (Sk.) muscle ( C ) were measured after sacrifice at week 12. Mean daily food intake ( D ) was recorded during 12 weeks. At week 12, plasma triglycerides ( E ), total cholesterol ( F ), glucose ( G ) and insulin levels ( H ) were measured in 6-hour unfed mice and HOMA-IR ( I ) was calculated. An intraperitoneal GTT (2 g/kg of total body weight) was performed in 6-hour unfed mice at week 11. Blood glucose levels were measured at the indicated time-points ( J ) and the area under the curve (AUC) of the glucose excursion curve was calculated as a measure of glucose tolerance ( K ). The plasma insulin level during ipGTT was measured at 15 minutes ( L ). Data are means ± SEM (n = 4 for LFD-WT; n = 7 for LFD-PHD1−/−; n = 5 for HFD-WT; n = 7 for HFD-PHD1−/−). *p < 0.05 vs LFD-fed mice, # p < 0.05 vs WT mice.

As expected, WT mice fed a HFD for 12 weeks exhibited increased body weight, liver and fat mass and fasting plasma TC, glucose and insulin levels ( Fig. 2A–H ). HOMA-IR and whole-body glucose tolerance were also significantly impaired compared to mice fed a LFD ( Fig. 2I–L ). Surprisingly, although HFD also promoted metabolic alterations in PHD1−/− mice, the effect of the diet on whole-body glucose homeostasis seemed somewhat dampened in those mice. Indeed, PHD1−/− mice fed a HFD displayed significantly lower fasting plasma glucose levels ( Fig. 2G ) and better glucose tolerance ( Fig. 2J,K ) than their WT counterparts.

PHD1 deficiency impairs hepatic insulin sensitivity

In order to study the impact of PHD1 deletion on systemic insulin sensitivity, LFD- and HFD-fed WT and whole-body knockout mice were subjected to an intraperitoneal ITT. The hypoglycemic response to insulin was impaired in LFD-fed PHD1−/− mice compared to WT mice ( Fig. 3A,B ), reflecting systemic insulin resistance. In parallel experiments, metabolic tissues (liver, eWAT and skeletal muscle) were harvested 10 min after insulin administration to assess tissue-specific insulin sensitivity by Western blot. Remarkably, the insulin-induced phosphorylation of protein kinase B (PKB) was significantly reduced in the liver of PHD1−/− compared to WT mice on LFD ( Fig. 3C,D ) whereas no significant effects were observed in eWAT and skeletal muscle ( Fig. 3E–H ), indicating that the alteration of systemic insulin sensitivity in LFD-fed PHD1−/− mice was mostly due to hepatic insulin resistance. When subjected to HFD, both WT and PHD−/− mice developed systemic and hepatic insulin resistance but no significant differences were observed between the two genotypes ( Fig. 3A–H ). As the canonical insulin signaling pathway was not impaired in eWAT and skeletal muscle from PHD1 mice, we next investigated whether the AMP-activated protein kinase (AMPK) signaling, an insulin-independent pathway involved in the peripheral regulation of glucose homeostasis, was affected in these tissues. In skeletal muscle, AMPK activity, assessed by the pThr172-AMPKα/AMPKα ratio, was significantly higher in skeletal muscle from PHD1−/− mice on HFD than in those from WT mice whereas no differences were found in mice on LFD ( Fig. S3A–D ). In line with this, similar changes in the phosphorylation state of Acetyl-CoA Carboxylase (ACC), one of the main downstream targets of AMPK, were observed ( Fig. S3A,E,F ). Of note, protein expression of AMPKα and ACC was found to be increased and decreased, respectively, in skeletal muscle of PHD1−/− mice, whatever the nutritional conditions ( Fig. S3A–G ). In eWAT, AMPK activity was lower in both LFD- and HFD-fed PHD1−/− mice when compared to WT mice ( Fig. S4A–D ). As expected, protein expression of both ACC and Fatty Acid Synthase (FAS) was reduced by HFD in both genotypes but was found to be significantly lower in LFD-fed PHD1−/− mice when compared to WT mice ( Fig. S4E,F ). Of note, mRNA levels of key genes involved in adipose tissue biology were comparable in eWAT from WT and PHD1−/− mice ( Fig. S4G ), with the notable exception of Lep (Leptin) and Ucp1 that were significantly higher in PHD1-deficient mice, in line with the larger adipose mass observed in those mice. Furthermore, PHD1 deficiency did not affect the expression of inflammatory genes in eWAT from LFD-fed mice ( Fig. S4H ).

PHD1 deficiency induces systemic and liver-specific insulin resistance.

An intraperitoneal ITT (0.5 U/kg total body weight) was performed in 6-hour unfed WT (open symbols/bars) and PHD−/− (black symbols/bars) mice after 11 weeks of either low-fat diet (LFD, squares) or high-fat (HFD, circles) diet. Blood glucose levels were measured at the indicated time-points ( A ) and the AUC of the glucose excursion curve was calculated as a measure of insulin resistance ( B ). In separate experiments, mice were sacrificed 15 min after insulin injection and tissue-specific insulin signaling was studied in liver, eWAT and skeletal muscle (Sk. M) by Western blot. Representative blots are shown in ( C,E,G ). Densitometric quantification was performed and results were expressed as fold change relative to WT-LFD mice ( D,F,H ). Data are means ± SEM (n = 4 for LFD-WT; n = 7 for LFD-PHD1−/−; n = 5 for HFD-WT; n = 7 for HFD-PHD1−/−). *p < 0.05 vs LFD mice, # p < 0.05 vs WT mice.

PHD1 deficiency promotes hepatic steatosis and liver inflammation

Hepatic insulin resistance is often associated with nonalcoholic steatohepatitis (NASH), which is characterized by inflammation and ectopic accumulation of TG in the liver. Strikingly, PDH1−/− mice on LFD exhibited visible steatosis and significantly higher hepatic cholesterol and TG content ( Fig. 4A–C ). This was associated with increased gene and protein expression of the lipogenic enzymes ACC and FAS ( Fig. 4D–G ) compared to WT mice. Of note, similar increase in lipogenic gene expression were also found in PDH1−/− mice on regular chow diet when compared with WT mice ( Fig. S5 ). Interestingly, the expression of key genes involved in glycolysis were also upregulated in the liver from PDH1−/− mice ( Fig. 4H ). In addition, a higher expression of some inflammatory gene markers was observed in liver from LFD-fed PHD1−/− mice ( Fig. 4I ). As expected, HFD increased liver cholesterol and TG levels in WT mice, along with a compensatory down-regulation of proteins involved in hepatic de novo lipogenesis ( Fig. 4A–I ). However, neither hepatic lipid composition nor expression of lipogenic proteins significantly differed between WT and PHD1−/− mice on HFD, indicating that HFD-induced hepatic steatosis was not aggravated by PHD1 deficiency.

PHD1 deficiency promotes hepatic steatosis.

Livers from WT (open bars) and PHD−/− (black bars) mice on either low-fat (LFD) or high-fat (HFD) diet were sampled ( A ) after 12 weeks. Hepatic cholesterol ( B ) and triglycerides (TG, C ) contents were determined. The mRNA expression of key genes involved in the regulation of hepatic TG synthesis ( Srebf1 : SREBP-1c; Acaca : ACC1; Fasn : FAS; Scd1 : SCD1), cholesterol synthesis ( Srebf2 : SREBP2 ; Hmgcs2 : HMGCoA synthase; Hmgcr : HMGCoA reductase) and fatty acid oxidation ( Ppara : PPARα; Pdk4 : PDK4 ; Cpt1a : CPT-1α; Acox1 : acyl-coA oxidase 1) was measured by RT-qPCR ( D ). Liver ACC and FAS protein expression were studied by Western blot. Representative blots are shown in ( E ). Total protein expression was quantified by densitometric analysis and expressed as fold change relative to WT-LFD mice ( F,G ). HSP90 was used for internal housekeeping protein expression. The mRNA expression of key genes involved in hepatic glycolysis (H; Slc2a1: GLUT1 ; Slc2a2, GLUT2 ; Gapdh, GAPDH ; Eno1, Enolase ; Pklr, PK) and inflammation (I; Emr1: F4/80; Vsig4: VSIG4; Tnfa: TNFα ; Il6: IL6) was measured by RT-qPCR. All the RT-qPCR results are expressed relative to the housekeeping gene RPLP0 as fold change vs WT-LFD mice. Data are means ± SEM (n = 4 for LFD-WT; n = 7 for LFD-PHD1−/−; n = 5 for HFD-WT; n = 7 for HFD-PHD1−/−). *p < 0.05 vs LFD mice, # p < 0.05 vs WT mice.

Overall, our results indicate that whole-body PHD1 deficiency in mice promotes adiposity, hepatic steatosis and liver-specific insulin resistance but does not worsen the deleterious effects of HFD on metabolic homeostasis.

In the present study, we report that whole-body PHD1 deletion in mice impairs systemic glucose homeostasis and insulin sensitivity in mice on standard chow or low-fat diet, a detrimental metabolic phenotype associated with increased hepatic steatosis and inflammation and liver-specific insulin resistance. However, PHD1 deficiency does not worsen the deleterious effects of HFD on whole-body insulin sensitivity and metabolic homeostasis.

In our conditions, the main metabolic tissue apparently involved in the alteration of systemic insulin resistance and glucose intolerance in PHD1−/− mice on LFD appears to be the liver, where a very significant decrease in insulin signaling was evidenced. At the mechanistic level, an increase in both glycolytic and lipogenic enzymes, hepatic lipid content and inflammatory markers were found in LFD-fed PHD1−/− mice. Interestingly, both hypoxia and PHD-1 deletion were shown to activate the pro-inflammatory IKKβ/NFkβ canonical pathway in an in vitro model of cancer cells 22 , suggesting that a similar alteration might occur in metabolic tissues and underlie increased local inflammation and insulin resistance. On the other hand, the increase in hepatic steatosis in PHD1−/− mice probably results from increased de novo lipogenesis, a metabolic process converting carbohydrate-derived acetyl-CoA produced during glycolysis into TG under the control of key enzymes involved in glycolytic and lipogenic pathways 23 , 24 . Indeed, the robust increase in the expression of glycolytic genes and lipogenic enzymes ACC and FAS likely contributes to the enhanced hepatic TG content observed in the liver from chow- and LFD-fed PHD1−/− mice. Interestingly, liver-specific deletion of the three PHD isoforms was also previously reported to induce severe hepatic steatosis 20 , 25 , 26 , although deletion of PHD1 or PHD2 alone and of a combination of them (1 + 2 and 2 + 3) did not seem sufficient to promote a significant increase in liver TG content in their experimental conditions 20 , 25 , 26 . However, in contrast to our results, enhanced hepatic lipid accumulation in the triple PHD knockout mice was associated with a decrease in mRNA expression of the lipogenic genes Srebf1 (SREBP-1C) and Fas (FAS) whereas expression of the glycolytic gene Slc2a1 (GLUT1) was similarly increased 20 . These conflicting results suggest that various PHD-specific downstream targets and/or pathways might be involved in the development of fatty liver.

The canonical function of PHDs is to hydroxylate HIF-α subunits, leading to its ubiquitination by the pVhl and subsequent degradation by the proteasome. Importantly, sustained activation of hepatic HIFs, either by overexpression of HIF-α isoforms or inactivation of the pVhl, has been previously shown to induce hepatic steatosis and inflammation 27 , 28 , 29 , 30 . It is noteworthy that, although disruption of pVhl activates HIF and leads to upregulation of both HIF-1α and HIF-2α target genes 31 , 32 , the induction of hepatic steatosis and inflammation seems to be exclusively HIF-2α-dependent in this model 29 , 30 . Indeed, liver-specific inactivation of pVhl and pVhl/HIF-1α promotes increase in hepatic triglyceride content 29 , 30 and pro-inflammatory Il6 and Il1b gene expression 29 whereas pVhl/HIF-2α mutant mice are phenotypically normal 29 , 30 .

The liver expresses all three HIF-α family members and whether the metabolic dysfunctions observed in the whole-body PHD1−/− mice, notably hepatic steatosis, was conferred by stabilization of specific HIFs remains to be elucidated. It has been shown that functional redundancies exist among the three PHD isoforms in targeting HIF-α subunits 11 , 33 , suggesting that PHDs, alone or in combination, might differently regulate HIF-1α or HIF-2α in the liver. Both HIF-1α and HIF-2α levels were found to be more abundant in the nucleus from PHD1−/− livers, with HIF-2α being the most upregulated isoform 34 . Due to lack of material, we were not able to assess this aspect in our present study but the increase in HIF-1α-specific target genes, such as Slc2a2 (GLUT2), Ldha (LDH), Eno1 (Enolase), Pklr (L-PK) and Egln3 (PDH3) 35 , suggests a HIF-1α-dependent metabolic reprogramming in the liver of PHD1−/− mice promoting glycolysis and pyruvate metabolism. Taken together, the tissue-specific respective contribution of the various HIF-α isoforms in the metabolic phenotype of the PHD1−/− whole-body knockout mice remains to be investigated more extensively.

Although the liver seems to be a central player in the metabolic disorders induced by whole-body PHD1−/− deletion, we cannot exclude that additional defects in key pathways controlling nutrient metabolism in peripheral tissues might also occur in other metabolic organs. For instance, the reduction in both ACC and FAS protein expression and AMPK activity in WAT from PHD1−/− mice might contribute to impaired tissue-specific fat storage and oxidation, respectively, indirectly promoting ectopic fat accumulation in the liver. Further studies will be necessary to dissect the respective tissue contribution to the metabolic phenotype observed in whole-body PHD1−/− knockout mice.

As expected, WT mice developed obesity, hepatic steatosis and systemic insulin resistance when fed a HFD and the extent of these diet-induced metabolic dysfunctions was found to be similar in PHD1−/− mice despite higher baseline body weight and insulin-resistance on standard chow diet. Surprisingly, the glucose tolerance was found to be slightly but significantly less impaired in PHD1−/− than in WT mice at both 6 and 12 weeks of HFD. Of note, PHD1−/− mice were previously reported to have tolerance to hypoxia, partly through metabolic reprogramming leading to a shift from oxidative toward anaerobic/glycolytic metabolism in skeletal muscle and liver 21 , 34 . A HIF-1-mediated increase in pyruvate dehydrogenase kinase 4 (PDK4) expression, which inhibits the conversion of pyruvate to acetyl-CoA and its subsequent entry into the mitochondrial tricarboxylic cycle, was proposed as one of the mechanisms underlying this metabolic adaptation 21 , 36 . Interestingly, in line with these observations, we also observed an increase in the expression of various glycolytic genes and a significantly higher PDK4 mRNA level in the liver of PHD1−/− mice. Altogether, it is therefore tempting to suggest that the higher glucose tolerance of HFD-fed PHD1−/− mice, compared to WT mice, might result from increased glucose uptake and oxidation in the liver, one of the main metabolic tissue contributing to systemic glucose homeostasis.

At the systemic level, the altered metabolic phenotype of our PHD1−/− mice on standard chow or LFD resembles that observed in a mouse model of adipose-tissue specific HIF-1 activation. Indeed, overexpression of a constitutively active form of HIF-1α in WAT led to an increase in body weight gain, fat mass and impairment of whole-body glucose tolerance in mice on standard chow diet 37 . However, in contrast with our study, the metabolic phenotype of these mice was clearly worsened when challenged with HFD, an effect associated with higher inflammation, fibrosis and insulin resistance in WAT 37 . In our condition, we rather observed a tendency toward decreased HFD-induced WAT inflammation in PHD1−/− mice, as evidenced by downregulation of some pro-inflammatory macrophage gene markers like CD68 and ITGAX ( Fig. S4 ). Interestingly, a recent study reports that both systemic and adipose-tissue specific deletion of PHD2 can protect mice against HFD-induced metabolic disorders 18 . In addition, liver-specific deletion of PHD3 alone or concomitantly with PHD1 or 2, was also shown to improve glucose tolerance and insulin sensitivity 20 . Adipocyte-specific PHD2−/− mice on HFD are indeed characterized by a decrease in fat mass and adipocyte size when compared to WT mice, as well as by a reduction in macrophage infiltration, expression of pro-inflammatory cytokines and lipogenesis in WAT 18 , 19 . Moreover, an increase in glucose oxidation in WAT due to up-regulation of glycolytic genes in adipocytes might also contribute to the improvement in whole-body glucose homeostasis 18 , 19 . A reduction in both HFD-induced hepatic gluconeogenesis and liver steatosis was also suggested to be involved in the favorable metabolic profile of these various models of PHD deficiency 18 , 19 , 20 . Of note, we did not find any significant differences in hepatic gluconeogenic gene expression between the two genotypes, whatever the diet used (data not shown). Overall, such contrasting results might originate from the specific PHD deletion used in these different studies. We herein displayed the metabolic consequences of whole-body PHD1 deletion, which might differ from those induced by a liver-specific deletion. For example, it can be speculated that part of the detrimental effect of whole-body PHD1 deletion on metabolic homeostasis might be secondary to a central effect, i.e. alteration of the hypothalamic regulation of peripheral insulin sensitivity and nutrient metabolism. Some HIF-α isoforms are indeed expressed in various hypothalamic regions 38 and PHD1 deletion might therefore affect HIF-α protein content and transcriptional activity in key brain area involved in metabolic homeostasis. Interestingly, PHD1 was recently identified as a regulator of neuronal metabolism by regulating glucose flux through glycolytic and pentose phosphate pathways 39 . In addition, insulin signaling in the brain was also reported to protect against ectopic lipid accumulation in the liver by stimulating hepatic TG secretion 40 , suggesting that alteration of central insulin sensitivity in PHD1−/− mice might also contribute to hepatic steatosis. Further studies are clearly required to investigate these brain-mediated aspects.

Altogether, the reason(s) behind the discrepancies in the metabolic phenotype displayed by the various PHDs deletion mice models are not clear but may be due to differences and/or redundancies in the tissue-specific control of the various HIF-α isoforms by each PHD. Of note, most of the studies showing an improvement in metabolic homeostasis have been performed in HFD-fed PHD knockout mice, a condition where we also observed a trend for a better glucose tolerance in our whole-body PHD1−/− mice compared to WT mice. However, in contrast with our study, there is relatively few information on the metabolic phenotype of these animals on LFD or standard chow diet. Finally, it is important to underline that in our model of constitutive whole-body knockout, we observed some tissue-specific changes in expression of the other PHD isoforms when PHD1 was deleted, notably in the liver. Therefore, we cannot exclude that some of the detrimental effects on metabolic homeostasis found in our model are actually secondary to a shift in the composition (and function) of PHD isoforms in the various metabolic tissues. In addition, it cannot be excluded that molecules other than HIFs, notably among those involved in the regulation of cellular glucose and lipid metabolism, can also be potential downstream targets of PHDs, for example by direct hydroxylase-independent interactions with their amino-terminal domains 35 , 36 .

In conclusion, we report here that whole-body PHD1 deficiency promotes hepatic steatosis and liver-specific insulin resistance but does not worsen the deleterious effects of HFD on metabolic homeostasis. Further studies using tissue-specific and inducible deletion of PHD1 are clearly required for investigating the respective contribution of the organs involved in the regulation of metabolic homeostasis, including the central nervous system, in the HIF-dependent or independent impairment of whole-body insulin sensitivity and glucose homeostasis.

Animal experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals of the Institute for Laboratory Animal Research and have received approval from the ethical committee of Université Grenoble Alpes (agreement number: B3851610006).

Animals and experimental design

PHD1 ( Egln2 )−/− mice (Swiss × sv129 background) were generated in Peter Carmeliet’s group (VIB, Leuven, Belgium), as previously described 21 and breeding couples were kindly provided for colony expansion. In the present study, PHD1−/− mice and their wild-type (WT) littermates were housed under standard conditions in conventional cages with ad libitum food and water. Ambient temperature was maintained at 20–22 °C. In the first experiment, male mice were fed a standard chow diet (RM1, Special Diets Services, Essex, England) and studied at 10 to 12 weeks of age. In the second set of experiments, 10 week-old male PHD1−/− and WT mice were fed a low-fat (10% energy derived from fat; D13091501; Research Diets) or a high-fat (60% energy derived from fat; D13091502; Research Diets) diet during 12 weeks.

Plasma analysis

Blood samples were obtained in 6-hour unfed mice (food withdrawn at 8:00 am) via tail vein bleeding. Blood glucose levels were determined using a glucometer (OneTouch ultra, Lifescan, Issy-Les-Moulineaux, France). Plasma TC, TG and insulin levels were measured using the commercially available enzymatic kits 236691, 11488872 (Roche Molecular Biochemicals, Indianapolis, IN) and ELISA insulin kit (#EZRMI, Milipore), respectively. The homeostatic model assessment (HOMA) adapted to mice was calculated as ([glucose (mg/dl)*0.055] × [insulin (ng/ml) × 172.1])/3857 and used as a surrogate measure of whole-body insulin sensitivity 41 .

Glucose and insulin tolerance tests

All the experiments were performed in 6-hour unfed mice (food withdrawn at 8:00 am). Whole-body glucose tolerance and insulin sensitivity were assessed by intraperitoneal (i.p.) glucose tolerance (GTT) and insulin tolerance (ITT) tests. After an initial blood collection (t = 0), i.p. injections of glucose (2g/kg total body weight) or insulin (0.5U/kg total body weight, NovoRapid, Novo Nordisk, Bagsvaerd, Denmark) were performed in conscious mice. Blood glucose levels were next measured by tail bleeding at 15, 30, 60, 90 and 120 min using a glucometer. The glucose or insulin areas under the curve (AUC) were measured using trapezoidal integration.

Hepatic lipid composition

Liver lipids were extracted as previously described 42 . Briefly, small liver samples were homogenized in ice-cold methanol. After centrifugation, lipids were extracted by addition of 1800 μl CH 3 OH:CHCl 3 (1:3 v/v) to 45 μl homogenate, followed by vigorous vortexing and phase separation by centrifugation (14000 rpm; 15 min at RT). The organic phase was dried and dissolved in 2% Triton X-100 in water. TG and cholesterol concentrations were measured using commercial kits as described above. Liver lipids were expressed as mg per mg protein, which was determined using the Bradford protein assay kit (Sigma-Aldrich, Saint-Quentin Fallavier, France).

Western blot analysis

Snap-frozen liver, skeletal muscle and epididymal WAT samples (~50 mg) were lysed in ice-cold buffer containing: 50 mM Hepes (pH 7.6), 50 mM NaF, 50 mM KCl, 5 mM NaPPi, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 5 mM β-glycerophosphate, 1 mM sodium vanadate, 1% NP40 and protease inhibitors cocktail (Complete, Roche, Mijdrecht, The Netherlands). Western blots were performed as previously described 42 . Primary antibodies used are listed in Supplementary Table 1 . Bands were visualized by enhanced chemiluminescence and quantified using Image J (NIH, US).

RNA purification and qRT-PCR

RNA was extracted from snap-frozen liver, skeletal muscle or epididymal adipose tissue samples (~25 mg) using Trireagent RNA isolation reagent (Sigma, Aldrich, Saint-Quentin Fallavier, France). Total RNA (0.5 μg) was reverse-transcripted and quantitative real-time PCR was then performed with SYBR Green Core Kit on a MyIQ thermal cycler (Bio-Rad). mRNA expression was normalized to RPLP0 mRNA content and expressed as fold change compared to control mice using the ∆∆CT method. Primers sequences are listed in Supplementary Table 2 .

Statistical analysis

All data are expressed as mean ± SEM. Statistical analysis was performed using Graphpad Prism 6 software package for Windows (San Diego, California USA) with two-tailed unpaired Student’s test (WT vs PHD1−/− on standard chow diet) or two-way ANOVA with multiple comparisons followed by post hoc Fisher’s LSD test (WT vs PHD1−/− on either LFD or HFD). Differences between groups were considered statistically significant when p < 0.05.

Additional Information

How to cite this article : Thomas, A. et al. Hypoxia-inducible factor prolyl hydroxylase 1 (PHD1) deficiency promotes hepatic steatosis and liver-specific insulin resistance in mice. Sci. Rep. 6 , 24618; doi: 10.1038/srep24618 (2016).

Ichiki, T. & Sunagawa, K. Novel roles of hypoxia response system in glucose metabolism and obesity. Trends Cardiovasc Med 24, 197–201 (2014).

Article   CAS   Google Scholar  

Trayhurn, P. Hypoxia and adipose tissue function and dysfunction in obesity. Physiol Rev 93, 1–21 (2013).

Pasarica, M. et al. Reduced adipose tissue oxygenation in human obesity: evidence for rarefaction, macrophage chemotaxis and inflammation without an angiogenic response. Diabetes 58, 718–725 (2009).

Goosens, G. H. et al. Increased adipose tissue oxygen tension in obese compared with lean men is accompanied by insulin resistance, impaired adipose tissue capillarization and inflammation. Circulation 124, 67–76 (2011).

Article   Google Scholar  

Ye, J., Gao, Z., Yin, J. & He, Q. Hypoxia is a potential risk factor for chronic inflammation and adiponectin reduction in adipose tissue of ob/ob and dietary obese mice. Am J Physiol Endocrinol Metab 293, E1118–1128 (2007).

Regazzetti, C. et al. Hypoxia decreases insulin signaling pathways in adipocytes. Diabetes 58, 95–103 (2009).

Yin, J., Gao, Z., Liu, D., Liu, Z. & Ye, J. Berberine improves glucose metabolism through induction of glycolysis. Am J Physiol Endocrinol Metab 294, E148–156 (2008).

Rausch, M. E., Weisberg, S., Vardhana, P. & Tortoriello, D. V. Obesity in C57BL/6J mice is characterized by adipose tissue hypoxia and cytotoxic T-cell infiltration. Int J Obes (Lond) 32, 451–463 (2008).

Schofield, C. J. & Ratcliffe, P. J. Oxygen sensing by HIF hydroxylases. Nat Rev Mol Cell Biol 5, 343–354 (2004).

Semanza, G. L. Oxygen Sensing, Hypoxia-Inducible Factors and Disease Pathophysiology. Annu Rev Pathol. 9, 47–71 (2014).

Appelhoff, R. J. et al. Differential function of the prolyl hydroxylases PHD1, PHD2 and PHD3 in the regulation of hypoxia-inducible factor. J Biol Chem 279, 38458–38465 (2004).

Jaakkola, P. et al. Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science 292, 468–472 (2001).

Article   CAS   ADS   Google Scholar  

Krishnan, J. et al. Dietary obesity-associated Hif1alpha activation in adipocytes restricts fatty acid oxidation and energy expenditure via suppression of the Sirt2-NAD + system. Genes Dev 26, 259–270 (2012).

Jiang, C. et al. Disruption of hypoxia-inducible factor 1 in adipocytes improves insulin sensitivity and decreases adiposity in high-fat diet-fed mice. Diabetes 60, 2484–2495 (2010).

Sun, K., Halberg, N., Khan, M., Magalang, U. J. & Scherer, P. E. Selective inhibition of hypoxia-inducible factor 1alpha ameliorates adipose tissue dysfunction. Mol Cell Biol 33, 904–917 (2013).

Shin, M. K. et al. Metabolic consequences of high-fat diet are attenuated by suppression of HIF-1alpha. PLos One 7, e46562 (2012).

Lee, Y. S. et al. Increased adipocyte O2 consumption triggers HIF-1alpha, causing inflammation and insulin resistance in obesity. Cell 157, 1339–1352 (2014).

Matsuura, H. et al. Prolyl hydroxylase domain protein 2 plays a critical role in diet-induced obesity and glucose intolerance. Circulation 127, 2078–2087 (2013).

Rahtu-Korpela, L. et al. HIF prolyl 4-hydroxylase-2 inhibition improves glucose and lipid metabolism and protects against obesity and metabolic dysfunction. Diabetes 63, 3324–3333 (2014).

Taniguchi, C. M. et al. Cross-talk between hypoxia and insulin signaling through Phd3 regulates hepatic glucose and lipid metabolism and ameliorates diabetes. Nat Med 19, 1325–1330 (2013).

Aragones, J. et al. Deficiency or inhibition of oxygen sensor Phd1 induces hypoxia tolerance by reprogramming basal metabolism. Nat Genet 40, 170–180 (2008).

Cummins, E. P. et al. Prolyl hydroxylase-1 negatively regulates IkappaB kinase-beta, giving insight into hypoxia-induced NFkappaB activity. Proc Natl Acad Sci USA 103, 18154–18159 (2006).

Perry, R. J., Samuel, V. T., Petersen, K. F. & Shulman, G. I. The role of hepatic lipids in hepatic insulin resistance and type 2 diabetes. Nature 510, 84–91 (2014).

Sanders, F. W. & Griffin, J. L. De novo lipogenesis in the liver in health and disease: more than just a shunting yard for glucose. Biol Rev Camb Philos Soc (2015).

Tojo, Y. et al. Hypoxia Signaling Cascade for Erythropoietin Production in Hepatocytes. Mol Cell Biol 35, 2658–2672 (2015).

Duan, L. J., Takeda, K. & Fong, G. H. Hematological, hepatic and retinal phenotypes in mice deficient for prolyl hydroxylase domain proteins in the liver. Am J Pathol 184, 1240–1250 (2014).

Haase, V. H., Glickman, J. N., Socolovsky, M. & Jaenisch, R. Vascular tumors in livers with targeted inactivation of the von Hippel-Lindau tumor suppressor. Proc Natl Acad Sci USA 98, 1583–1588 (2001).

Kucejova, B. et al. Uncoupling hypoxia signaling from oxygen sensing in the liver results in hypoketotic hypoglycemic death. Oncogene 30, 2147–2160 (2011).

Qu, A. et al. Hypoxia-inducible transcription factor 2alpha promotes steatohepatitis through augmenting lipid accumulation, inflammation and fibrosis. Hepatology 54, 472–483 (2011).

Rankin, E. B. et al. Hypoxia-inducible factor 2 regulates hepatic lipid metabolism. Mol Cell Biol 29, 4527–4538 (2009).

Rankin, E. B. et al. Hypoxia-inducible factor-2 (HIF-2) regulates hepatic erythropoietin in vivo . J Clin Invest 117, 1068–1077 (2007).

Rankin, E. B. et al. Hypoxia-inducible factor-2 regulates vascular tumorigenesis in mice. Oncogene 27, 5354–5358 (2008).

Berra, E. et al. HIF prolyl-hydroxylase 2 is the key oxygen sensor setting low steady-state levels of HIF-1alpha in normoxia. EMBO J 22, 4082–4090 (2003).

Schneider, M. et al. Loss or silencing of the PHD1 prolyl hydroxylase protects livers of mice against ischemia/reperfusion injury. Gastroenterology 138, 1143–1154 (2010).

Wong, B. W., Kuchnio, A., Bruning, U. & Carmeliet, P. Emerging novel functions of the oxygen-sensing prolyl hydroxylase domain enzymes. Trends Biochem Sci 38, 3–11 (2013).

Kim, J. W., Tchernyshyov, I., Semenza, G. L. & Dang, C. V. HIF-1-mediated expression of pyruvate dehydrogenase kinase: a metabolic switch required for cellular adaptation to hypoxia. Cell Metab 3, 177–185 (2006).

Halberg, N. et al. Hypoxia-inducible factor 1alpha induces fibrosis and insulin resistance in white adipose tissue. Mol Cell Biol 29, 4467–4483 (2009).

Zhang, H., Zhang, G., Gonzalez, F. J., Park, S. M. & Cai, D. Hypoxia-inducible factor directs POMC gene to mediate hypothalamic glucose sensing and energy balance regulation. PLos Biol 9, e1001112 (2011).

Quaegebeur, A. et al. Deletion or Inhibition of the Oxygen Sensor PHD1 Protects against Ischemic Stroke via Reprogramming of Neuronal Metabolism. Cell Metab 23, 280–291 (2016).

Scherer, T. et al. Insulin regulates hepatic triglyceride secretion and lipid content via signaling in the brain. Diabetes, doi: 10.2237/db15-1552 (2016).

Lee, S. et al. Comparison between surrogate indexes of insulin sensitivity and resistance and hyperinsulinemic euglycemic clamp estimates in mice. Am J Physiol Endocrinol Metab 294, E261–270 (2008).

Geerling, J. J. et al. Metformin lowers plasma triglycerides by promoting VLDL-triglyceride clearance by brown adipose tissue in mice. Diabetes 63, 880–891 (2014).

Download references

Acknowledgements

This work was supported by Fonds de Dotation “Recherche en Santé Respiratoire”, Fonds de Dotation “Agir pour les Maladies Chroniques”, Fondation du Souffle and Fondation de France.

Author information

Authors and affiliations.

Laboratoire HP2, Université Grenoble Alpes, Grenoble, F-38042, France

Amandine Thomas, Elise Belaidi, Patrick Levy, Jean-Louis Pepin & Diane Godin-Ribuot

INSERM U1042, Grenoble, F-38042, France

Institute of Cardiometabolism and Nutrition (ICAN), Assistance Publique Hôpitaux de Paris, Pitié-Salpêtrière Hospital, Paris, France

Judith Aron-Wisnewsky & Karine Clement

INSERM, U1166, Nutriomics Team 6, Paris, France

Sorbonne Universités, UPMC University Paris 06, UMR_S 1166, Nutriomics Team, Paris, France

Department of Molecular Cell Biology, Leiden University Medical Center, Leiden, The Netherlands

Gerard C. van der Zon & Bruno Guigas

Department of Parasitology, Leiden University Medical Center, Leiden, The Netherlands

• Bruno Guigas

You can also search for this author in PubMed   Google Scholar

Contributions

A.T. performed experiments, analyzed data and drafted the manuscript; E.B. performed experiments, contributed to discussion and critically reviewed the manuscript; J.A.W. contributed to discussion and critically reviewed the manuscript; G.C.v.d.Z. performed experiments; P.L., K.C. and J.L.P. contributed to discussion and critically reviewed the manuscript; D.G.R. supervised the project, contributed to discussion and critically reviewed the manuscript; B.G. supervised the project, conceptualized the project, performed experiments, analyzed data, wrote and edited the manuscript. B.G. is the guarantor of this work and, as such, has full access to all the data generated in the framework of the study and takes responsibility for their integrity and the accuracy of their analysis.

Ethics declarations

Competing interests.

The authors declare no competing financial interests.

Electronic supplementary material

Supplementary information, rights and permissions.

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Reprints and permissions

About this article

Cite this article.

Thomas, A., Belaidi, E., Aron-Wisnewsky, J. et al. Hypoxia-inducible factor prolyl hydroxylase 1 (PHD1) deficiency promotes hepatic steatosis and liver-specific insulin resistance in mice. Sci Rep 6 , 24618 (2016). https://doi.org/10.1038/srep24618

Download citation

Received : 27 November 2015

Accepted : 31 March 2016

Published : 20 April 2016

DOI : https://doi.org/10.1038/srep24618

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Phd1-3 oxygen sensors in vivo—lessons learned from gene deletions.

• Agnieszka E. Jucht
• Carsten C. Scholz

Pflügers Archiv - European Journal of Physiology (2024)

The role of hypoxia-inducible factor 1α in hepatic lipid metabolism

• Mingxiao Luo
• Tingting Li
• Haiquan Sang

Journal of Molecular Medicine (2023)

Abnormal whole-body energy metabolism in iron-deficient humans despite preserved skeletal muscle oxidative phosphorylation

• Matthew C. Frise
• David A. Holdsworth
• Peter A. Robbins

Scientific Reports (2022)

Effects of a novel polyphenol-rich plant extract on body composition, inflammation, insulin sensitivity, and glucose homeostasis in obese mice

• Hendrik J. P. van der Zande
• Joost M. Lambooij

International Journal of Obesity (2021)

PHD1 controls muscle mTORC1 in a hydroxylation-independent manner by stabilizing leucyl tRNA synthetase

• Gommaar D’Hulst
• Inés Soro-Arnaiz
• Katrien De Bock

Nature Communications (2020)

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Ph.D. Program

Graduate studies.

The Genetics Ph.D. program provides opportunities for graduate study in all major areas of modern genetics, including identification and analysis of human disease genes, molecular evolution, gene therapy, statistical genetics, application of model organisms to problems in biology and medicine, and computational and experimental approaches to genome biology.

An underlying theme in our Department is that genetics is not merely a set of tools but a coherent and fruitful way of thinking about biology and medicine. To this end, we emphasize a spectrum of approaches based on molecules, organisms, populations, and genomes.

We provide training through laboratory rotations, dissertation research, seminar series, didactic and interactive coursework, and an annual three-day retreat.

Students receive a competitive stipend ($51,600) for the 2023-24 Academic Year), tuition, health insurance, and a dental care stipend for a full four years. We also encourage students to seek additional fellowships, including but not limited to: NSF GRFP , NIH F31 , NDSEG , Stanford Bio-X fellowship , Stanford DARE , and Stanford CEHG Fellowship .

Lab Rotations

Students rotate through 3 laboratories during their first year in the Genetics Graduate Program. While most students start in Fall Quarter, students are encouraged to consider participating in the  Advance Summer Institute  for a smoother early transition into graduate school. There is a nomination & selection process. The department nominates, so if you are interested please let the department student services officer know. The program is not meant to be a source of summer bridge funding or simply an early rotation opportunity. There are many components to the program that require commitment of time and effort and the funding, reflects both the expectation of full participation and belief that participants should be compensated for these efforts. Office of Graduate Education does the selection for ADVANCE. There is no guarantee that if you are nominated that you will be of admitted into ADVANCE.

Rotations typically last one quarter each, but can be less and are contingent upon the faculty member agreeing to the rotation request. All Genetics students must rotate with at least 1 Genetics faculty member (primary or secondary appointment). Other rotations may be done with any Bioscience faculty.

While students may select a thesis laboratory after completing their third rotation, you can do more Selection of the dissertation research laboratory must be done with the faculty member's approval. Prior to committing to a dissertation laboratory, students are invited to discuss their selection with the Graduate Program Director. Students are welcome to join labs outside of the Genetics Department; if so, they will discuss with the Graduate Program Director whether transferring into that department would be beneficial.

Dissertation Research

Once a student selects a permanent laboratory, they begin their dissertation research that will last for approximately four years. All students are expected to publish at least one first-author paper about their research during this time period, and the work culminates with a thesis defense presentation and written dissertation. See the Genetics Student Handbook for more information.

Coursework, Qualifying Exams, and other Requirements

Students in the Genetics Graduate Program take the Qualifying Examination in the Fall Quarter of their second year of study. There are two parts to the exam, a written research proposal and an oral examination.

Ami Bhatt, Dylan Maghini, and collaborators from the University of the Witwatersrand visit with researchers and staff at the MRC/Wits Public Health and Health Transitions Research Unit in Agincourt, South Africa.

Ami Bhatt, Dylan Maghini, and collaborators tour the MRC/Wits Public Health and Health Transitions Research Unit labs and biobank facility in Agincourt, South Africa.

Service Requirement

Service and outreach are a critical component of a student’s development as a scientist, and offer unique opportunities to learn by interacting with individuals outside the Department. Students are expected to participate in a minimum of 60 hours of service and/or outreach work prior to defending their dissertation.

Supplementary Educational Activities

In addition to your courses, qualifying exams, and dissertation, the Genetics Department has arranged additional educational activities for students. These regularly occurring meetings are:

Current Issues in Genetics (CIG) Two people from the Genetics Department give 20-25 minute presentations about their current work at this weekly Friday meeting. Students in their third year and above are expected to present their work annually. This series gives students the chance to learn about the range of science going on in the department and provides a great opportunity to give formal presentations to peers and colleagues. 

Graduate Student Journal Club This weekly journal club is organized completely by graduate students from the Genetics and Developmental Biology Departments. At each meeting, one or two graduate students lead 30 minute discussions on their choice of a recent journal article. For the first three years of the Ph.D. program, each student presents once per academic year.

Refreshments are provided by the graduate students and reimbursed up to the current year limit. Reimbursement requires an original receipt to the Student Services Coordinator.

Frontiers in Biology  Every week, the Departments of Genetics, Developmental Biology, and Biochemistry host an external speaker through the “Frontiers in Biology” seminar series. First year students also take a course related to this seminar (GENE 215), where they discuss a relevant paper the day before and meet the speaker after the presentation. 

Frontiers is held most Wednesdays at 4pm in Clark Auditorium.

Other Seminar Series There are many other regular seminar series on campus that students choose to attend. Some of the most popular include:

Center of Law and the Biosciences lunchtime talks. See CLB events calendar or subscribe to the listserv

Evolgenome (organized by CEHG). See CEHG website or subscribe to the listserv

Teaching and Mentoring Academy Events. See TMA website or subscribe to the listserv

Discussing Developmental Data (3D). See events schedule

Biomedical Seminars. See events schedule or subscribe to the listserv

Requirements

Important forms.

Service Requirement Form Click Here

Other Important Info

Please examine the research interests and laboratory descriptions available from the Department Faculty page.

The Office of Postdoctoral Affairs maintains a list of open postdoctoral positions. This site also provides useful information to those considering postdoctoral positions at Stanford.

• Open access
• Published: 24 January 2020

Characterization of the plant homeodomain (PHD) reader family for their histone tail interactions

• Kanishk Jain 1 , 2 ,
• Caroline S. Fraser 2 , 3 ,
• Matthew R. Marunde 4 ,
• Madison M. Parker 1 , 2 ,
• Cari Sagum 5 ,
• Jonathan M. Burg 4 ,
• Nathan Hall 4 ,
• Irina K. Popova 4 ,
• Keli L. Rodriguez 4 ,
• Anup Vaidya 4 ,
• Krzysztof Krajewski 1 ,
• Michael-Christopher Keogh 4 ,
• Mark T. Bedford 5 &
• Brian D. Strahl   ORCID: orcid.org/0000-0002-4947-6259 1 , 2 , 3  

Epigenetics & Chromatin volume  13 , Article number:  3 ( 2020 ) Cite this article

9292 Accesses

51 Citations

9 Altmetric

Metrics details

Plant homeodomain (PHD) fingers are central “readers” of histone post-translational modifications (PTMs) with > 100 PHD finger-containing proteins encoded by the human genome. Many of the PHDs studied to date bind to unmodified or methylated states of histone H3 lysine 4 (H3K4). Additionally, many of these domains, and the proteins they are contained in, have crucial roles in the regulation of gene expression and cancer development. Despite this, the majority of PHD fingers have gone uncharacterized; thus, our understanding of how these domains contribute to chromatin biology remains incomplete.

We expressed and screened 123 of the annotated human PHD fingers for their histone binding preferences using reader domain microarrays. A subset (31) of these domains showed strong preference for the H3 N-terminal tail either unmodified or methylated at H3K4. These H3 readers were further characterized by histone peptide microarrays and/or AlphaScreen to comprehensively define their H3 preferences and PTM cross-talk.

Conclusions

The high-throughput approaches utilized in this study establish a compendium of binding information for the PHD reader family with regard to how they engage histone PTMs and uncover several novel reader domain–histone PTM interactions (i.e., PHRF1 and TRIM66). This study highlights the usefulness of high-throughput analyses of histone reader proteins as a means of understanding how chromatin engagement occurs biochemically.

Histone proteins are fundamental to genome organization and packaging, and are chemically modified by a wide range of “writer” or “eraser” enzymes that, respectively, install or remove histone post-translational modifications (PTMs) [ 1 , 2 ]. These PTMs play a central role in chromatin function: some are believed to directly impact chromatin organization through biophysical means, but the vast number likely function through their ability to recruit effector or “reader” domain-containing proteins to chromatin. These reader proteins, which are often found in large multi-subunit complexes and in additional chromatin-modifying machines, interact with histone tails and chromatin in various ways that regulate gene transcription and other chromatin functions [ 2 , 3 ]. The varied and diverse patterns of histone PTMs that exist in vivo are referred to as the ‘histone code’, which is still poorly understood [ 2 , 3 ].

Histone PTMs often have either activating or repressive effects on gene transcription depending on the type of PTM (acetylation, methylation, etc.) and the position being modified (H3K4, H3S10, etc.). In general, distinct classes of reader domains bind to specific types of PTMs; for example, bromodomains recognize lysine acetylation [ 4 ], chromodomains recognize methyl-lysine [ 5 ], and the PHD fingers characterized to date generally recognize unmodified or methylated lysine residues [ 6 ]. Furthermore, many chromatin-associated proteins contain multiple reader domains, either multiples of the same type [ 7 ] or a variety of different domains [ 8 ], potentially meaning that the in vivo engagement with chromatin is multivalent. Significantly, increasing evidence shows that dysregulation of the epigenetic machinery, most notably the readers, writers, and erasers of the histone code, is causal for a wide range of human disease, including cancer [ 9 ].

Plant homeodomain fingers comprise one of the largest families of reader domains, with over 100 human proteins containing this module [ 6 ]. PHD fingers are Zn-coordinating domains that generally recognize unmodified or methylated lysines. To date, the majority of those characterized bind to histone H3 tails either methylated at K4 [ 7 ], or unmodified in that position (i.e., KDM5B PHD3 versus KDM5B PHD1 [ 10 , 11 ] or PHF21A, also known as BHC80 [ 12 ]). A smaller number of PHD fingers are reported as readers of H3K9 trimethylation (H3K9me3; e.g., CHD4) [ 13 , 14 ] and H3K36me3 (e.g., budding yeast Nto1) [ 15 ]. Intriguingly, the dual PHD finger region of DPF3b has been reported as a reader of H3K14ac [ 16 ], while PHD6 of MLL4 has been reported to recognize H4K16ac [ 17 ]. Additionally, a number of these PHD fingers occur in tandem (e.g., MLL1-4 [ 7 ] and PZP-containing proteins [ 18 , 19 ]) or next to additional reader domain types (e.g., bromodomains and chromodomains) [ 20 , 21 , 22 ], suggesting combinatorial interaction capabilities.

Despite great progress in uncovering the role of a subset of PHD fingers, many (over 100) of the annotated domain family remain uncharacterized. In this report, we set out to close the gap in our understanding of this reader domain class. Using a combination of complementary approaches (reader domain microarrays, peptide microarrays, pulldowns, and AlphaScreen peptide assays), we show (31/123) of the PHD-containing query proteins to bind histone H3 N-terminal peptides, with the majority of these preferring H3K4me3 over unmodified H3K4. Furthermore, a number of unreported histone PTM–PHD protein interactions were uncovered, with the PHD regions of PHRF1 and TRIM66 binding preferentially to an unmodified H3 N-terminal tail peptide. Given that many of these PHD fingers are mutated in diseases such as breast cancer and leukemia [ 7 , 20 , 21 , 22 , 23 , 24 ], these findings enhance our overall understanding of PHD reader–histone interactions and should serve as a resource and platform for future studies.

Analysis of the PHD finger proteome via protein domain microarrays

To define the histone binding preferences of the PHD finger proteome, we expressed and purified 123 annotated human PHD-containing domains as GST-tagged recombinant fusions from E. coli . The recombinant proteins consisted of either PHD fingers in isolation, or as tandem domains if a given PHD finger was located adjacent to another reader domain (e.g., one or more PHD fingers, Tudor, chromo and/or bromodomains) (Additional file 1 : Table S1). These GST fusions were printed in duplicate on nitrocellulose-coated microarray slides and probed with biotinylated peptides that represented the N-termini of H3, H4, H2A or H2B (Fig.  1 a and Additional file 2 : Figure S1). As the majority of PHD readers thus far characterized are H3K4me0/3 readers [ 6 ], we included additional peptides (H3K4 as either mono-, di-, or trimethylated) to further determine any H3K4 methyl preference (Additional file 2 : Figure S2 and Fig.  1 b). As a control, we also probed these microarrays with an α-Tubulin peptide (a.a. 30–50) that would not be predicted to interact with PHD fingers (Additional file 2 : Figure S1). As in Fig.  1 a, b, 31 of the 123 PHD-containing fusions showed positive binding to the H3 N-terminus, with the majority of these interactions showing preference for trimethylated H3K4. In contrast, the H2A, H2B, H4, and tubulin peptides showed little to no positive interactions, suggesting that the PHD finger family broadly prefers the histone H3 tail (Additional file 2 : Figure S1). We note that the absence of binding in these experiments does not rule out the possibility of PHD-finger:histone PTM recognition under different hybridization conditions. We also cannot exclude the possibility that some PHD fingers might not be functionally active on the microarrays (perhaps due to misfolding or the lack of an important adjacent region).

PHD domain array identifies 31 H3-interacting proteins. a PHD finger domain microarray probed with an unmodified H3 N-terminal peptide (1–20) (see “ Methods ”). Each positive binding interaction appears as a green circle, with each PHD protein in the array spotted in technical duplicate (indicated by connecting white lines). a PHD finger domain array probed with an H3 (1–20) peptide trimethylated at residue K4 (K4me3). c The 31 H3-interacting proteins are listed by their preference for binding H3 (1–20) K4me3 or K4me0. Each protein listed corresponds to the numbers in a , b . TTP Tandem Tudor domain + PHD, PPCC Dual PHD + Dual Chromodomain, PCC PHD + Dual Chromodomain, CW CW-type Zn-finger, PB PHD + Bromodomain, PPC2W2 Dual PHD + C2W2-type Zn-finger, SPB SAND + PHD + Bromodomain; domains not indicated, one PHD finger. For the entire list of proteins used and the microarray map, see Additional file 1 : Table S1

Based on the above, we were able to classify the [PHD–H3 tail] interactions into three groups, namely PHD fingers that: (1) bound specifically with methylated H3K4; (2) interacted only with unmethylated H3K4; or (3) bound without preference to the H3K4 methylated state. Many of the PHD fingers found to only bind H3K4 methylation have previously been described and include the well characterized domains from the ING and PHF protein families [ 6 , 24 ]. The PHD finger of MLL5, a member of the MLL/KMT2 family [ 25 , 26 , 27 , 28 ], showed strong preference for H3K4me2 and H3K4me3. This finding adds to the relatively small number of MLL5-histone PTM observations reported to date [ 25 ]. Of the PHD fingers that bound to H3K4 methylation specifically, we observed that H3K4me3 or H3K4me2 were largely recognized equivalently and these domains did not detect H3K4me1 to the same degree (Additional file 2 : Figure S2)—a result in agreement with other reports showing H3K4me binding occurs largely on higher methylated states [ 6 ]. Again, as with the H3K4me3 interacting PHDs, our findings for proteins such as KDM5A [third PHD finger (PHD3)] and KDM5B [third PHD finger (PHD3)] are consistent with their current classification as H3K4me3 binders [ 10 , 11 ]. In contrast to H3K4me2/3 binding, a smaller number of PHD fingers [e.g., PHD1 from KDM5A and KDM5B, PHF21A, AIRE (PP), and TRIM66 (PB)] showed preference for the unmethylated H3K4 state (Fig.  1 a, c). Furthermore, three PHD fingers we tested showed no preference between the H3K4me0 and H3K4me3 peptides: PHRF1 (RP), CHD5 (PCC), and KDM5B (PHD3) (Fig.  1 ). Collectively, these experiments identified 31 PHD-containing reader domains that showed positive interaction with the H3 N-terminus. While a majority of these reader domains preferentially interacted with H3K4me3 (18 out of 31) or H3K4me0 (10 out of 31), three showed no preference for the state of modification at K4. Importantly, these analyses uncovered several reader:histone interactions for poorly characterized PHDs (i.e., TRIM66, PHRF1, and SP140L): such insight could provide new avenues of investigation to these disease-relevant proteins [ 29 , 30 , 31 , 32 ].

Further characterization of H3-reading PHD fingers by peptide microarrays

To more comprehensively define the histone interactions of the 31 PHD readers identified from the domain microarray analyses, we probed each on an alternate microarray platform containing a library of 293 synthetic histone peptides with single or combinatorial PTMs [ 33 ] (Additional file 2 : Figure S4 and Additional file 3 : Table S2). All screening results can be found in Additional file 3 : Table S2, but for brevity, findings pertaining to peptides that contain K4 and K9 modifications as well as neighboring phosphorylation sites that impinge on the observed binding by reader domains are displayed in the form of a normalized heatmap (Fig.  2 ). In general, the 31 PHD fingers were confirmed to associate with the H3 tail with the same H3K4 methyl preferences as in the domain microarrays (Fig.  2 ; Additional file 3 : Table S2). Notably, the MLL5 PHD finger displays a strong preference for H3K4me3 over the un-, mono-, or di-methylated H3K4 peptides (Fig.  2 ), and further, over all other histone peptides on the array (Additional file 3 : Table S2), consistent with results from the domain array (Fig.  1 ). Since CHD4, a protein annotated to recognize H3K9me3 [ 13 , 14 ], was a positive binder in this assay, we compared its binding to H3K9me3 or H3K4 methyl peptides along with their unmodified counterparts at each position (K4me0/K9me0). The CHD4 (PPCC) fusion bound H3 N-terminal peptides more strongly when H3K4 was unmodified and dually acetylated at K9 and K18 versus when H3K4 is methylated in an identically acetylated context (Fig.  2 ); additionally, there was no difference in binding to the H3K4me0 peptide versus the H3K9me3 peptide. Interestingly, there also seems to be increased binding with CHD4 (PPCC) to the H3 K9ac peptide, potentially due to the “surface effect” (described in detail below). In addition, we confirmed the newly identified interactions observed with the domain microarrays for PHRF1 and TRIM66 (Fig.  2 ).

A majority of PHD-containing proteins identified in the domain array are H3 K4me3 readers. The heatmap represents relative binding of the indicated H3 N-terminal peptides (left side) to the PHD-containing GST-tagged proteins (top). Binding strength is shown as a color gradient from red to blue (stronger to weaker). Most of the 31 PHD proteins preferentially recognize H3K4me3 when residues K9 and K18 are acetylated. Array signals ( n  = 4) were normalized individually for each protein to the highest signal for each respective array; thus, comparisons should only be made between binding strengths of different peptides for the same protein. TTP Tandem Tudor domain + PHD, PPCC Dual PHD + Dual Chromodomain, PCC PHD + Dual Chromodomain, CW CW-type Zn-finger, PB PHD + Bromodomain, PPC2W2 Dual PHD + C2W2-type Zn-finger, SPB SAND + PHD + Bromodomain; domains not indicated, one PHD finger. For full construct information, see Additional file 1 : Table S1 and Additional file 2 : Figure S3. For full peptide microarray data, see Additional file 3 : Table S2

While findings between the domain microarrays and peptide microarrays largely agreed, there were some interesting differences. For example, PHRF1 (RP) showed no preference for the H3K4 methyl state on the domain array but strong preference for H3K4me0 on peptide microarray. Furthermore, KDM5B (PHD3), is reported to bind H3K4me3 [ 11 ], and showed such a preference on peptide microarrays but not on domain microarrays (Figs.  1 and 2 ). It should be noted that the comparison made here is between the H3K4me3 + K9ac + K18ac and the H3K4me0 + K9ac + K18ac peptides. Due to the limited binding, if any, observed by the non-acetylated versions of these peptides, it is difficult to assess the binding preference displayed by KDM5B (PHD3) with this comparison. Of note, certain PHD readers [i.e., DIDO1 and DPF2 (PPC2W2)] also showed some interaction with a number of H4 N-terminal peptides (Additional file 3 : Table S2), consistent with published reports [ 33 , 34 ].

During the course of this study, we observed that domain binding to H3 peptides tended to be enhanced when neighboring lysine residues were additionally acetylated (e.g., [K9ac + K18ac] for H3K4me0 or H3K4me readers) (Fig.  2 ). While at first approximation it might appear that these readers have an enhanced affinity for poly-acetylated states that neighbor H3K4, we note that solution-based peptide pulldown or AlphaScreen (see below) assays with several of these readers (i.e., KDM7A that binds H3K4me3 and KDM5B (PHD1) that reads H3K4me0) did not support this idea (Additional file 2 : Figure S5 and Fig.  3 i). We surmise that the enhanced binding caused by poly-acetylation is a property of the charged surface of the streptavidin-coated glass slides: when modified with bulky and neutral acetyl groups the highly charged histone N-terminal tail peptides become more accessible to reader domains.

dCypher histone peptide-binding assays define the PTM recognition preference of PHD proteins with high sensitivity. a – h Binding curves to determine optimal reader protein concentration for full peptide library screening on the dCypher ® AlphaScreen ® platform (see “ Methods ”). X -axes are log(protein concentration ( M )) at constant peptide concentration (100 nM); Y -axes are AlphaScreen counts, representing relative strength of binding ( n  = 2; error bars are S.D.). i Heat map represents relative binding to H3 N-terminal peptides (left) by PHD-containing GST-tagged proteins (top) using the dCypher AlphaScreen platform. Protein concentrations can be found in Additional file 5 : Table S4. Binding strength is indicated by color gradient from green to yellow (stronger to weaker). The asterisk (*) by MLL5 signifies its general preference for H3K4 methylation. Alpha counts ( n  = 2) were normalized individually for each protein to the highest signal for each respective assay. For full dCypher peptide screen data, see Additional file 4 : Table S3

Quantitative assessment of poorly defined PHD readers by the AlphaScreen dCypher assay

We next employed a highly sensitive proximity-based AlphaScreen histone peptide assay ( dCypher ® ) to provide a third and orthogonal approach to analyzing the histone binding preferences for a subset of the 31 PHD proteins with respect to various histone tail PTMs. In this assay, biotinylated peptides are bound to streptavidin “donor” beads and the GST-tagged reader domains bound to Glutathione “acceptor” beads. The donor beads are excited by 680 nm light, releasing a singlet oxygen which causes light emission (520–570 nm) in proximal acceptor beads (within 200 nm); emission intensity is then correlated to binding strength [ 35 ]. For further examination with this more sensitive approach we chose the PHD fingers with positive binding data from the domain and peptide microarrays that were less characterized in the literature [i.e., MLL5, PHRF1 (RP), and TRIM66 (PB)], or those that displayed weak interactions on the domain and/or peptide microarrays [i.e., CHD4 (PPCC) and CHD5 (PPCC)]. Additionally, we examined several well characterized PHD–PTM interactors [DIDO1, KDM7A, and DPF2 (PPC2W2)] for positive controls and to provide a benchmark. Initial binding assays were conducted for each fusion protein using three peptides [H3 (1–20) with K4me0, H3K4me3 or H3K9me3] to determine the optimal reader domain concentration for full peptide library studies (Fig.  3 a–h; Additional file 4 : Table S3 and Additional file 5 : Table S4). This is an important first step as signal often declined after query protein saturation (the ‘hook point’, caused by excess free query competing with bead bound).

Once the optimal protein concentration ranges for each of the eight readers were determined, we conducted the full dCypher peptide screen (293 histone peptides) (Fig.  3 i; Additional file 4 : Table S3). In agreement with our previous findings, the dCypher peptide assay demonstrated KDM7A to be a reader of H3K4me3. Furthermore, TRIM66 (PB) showed a preference for H3K4me0 and me1, consistent with findings from the peptide microarrays. For CHD4 (PPCC), the dCypher approach showed a clearer specificity for the H3K4me0 peptide over the methylated species in comparison to the peptide microarray results (Fig.  3 i versus Fig.  2 ). In the case of CHD5 (PPCC), the peptide microarray indicated this protein to be insensitive to the methylation status at H3K4 (Fig.  2 ), but the dCypher assay identifies a preference for H3K4me0/1 (Fig.  3 i), consistent with the domain microarray (Fig.  1 a, c).

Consistent with the results from the domain and peptide microarrays, dCypher assays confirmed that the PHD finger of DIDO1 and MLL5 recognized the higher methyl states of K4 (me3/2), but also identified interaction of these domains with the H3K4me1 peptide. Interestingly, the four H3K4me0 readers analyzed—CHD4 (PPCC), DPF2 (PPC2W2), TRIM66 (PB), CHD5 (PPCC)—also showed the ability to bind to the peptides containing H3K9me3; this may be due to H3K4me0 in the H3K9me3 peptide. However, CHD4 (PPCC) and TRIM66 (PB) showed stronger interaction with H3K9me3 compared with the unmodified peptide over a range of protein concentration (Fig.  3 d, f). We note that while the initial protein concentration optimizations in Fig.  3 a–h were performed over a range of protein concentrations, the full peptide screen (Additional file 5 : Table S4; summarized in panel Fig.  3 i) was performed at a single protein concentration. When presented with the [H3K9me3 + S10p] peptide, four out of five of the H3K4me0 readers lose binding capacity, suggesting that these readers are sensitive towards the bulky negative phosphate group at S10; this phenomenon is also observed with the H3S10p peptide alone (Additional file 4 : Table S3). To our knowledge, this would be the first report of a H3 tail binder outside the H3K9 position to be impacted by S10 phosphorylation, suggesting the phospho-methyl switch may function more broadly than previously thought. Intriguingly, PHRF1 (RP) binding specificity at 15 nM showed more limited interactions to H3K4me0 and H3K9me3 peptides (Fig.  3 i), which will be discussed further below. Finally we note that the shift for poly-acetyl peptides seen in the peptide microarrays (reflecting a possible “surface effect”; Fig.  2 ) is not observed in the dCypher screen (Fig.  3 i) which more closely resembles the peptide pulldown assays (Additional file 2 : Figure S5).

In the epigenetic landscape, histone PTMs can impact chromatin organization through their ability to recruit effector or “reader” domain-containing proteins. These reader proteins, which are also found in large multi-subunit chromatin-modifying machines, interact with histones and chromatin in various ways that regulate processes from gene transcription to chromosome segregation at mitosis [ 2 ]. Given that many of these reader proteins are widely dysregulated in human disease, understanding their histone binding preferences and modes of multivalent interactions is vital [ 36 ]. In this study, we screened 123 PHDs (singly and in tandem when next to another reader domain) against the core histone N-terminal tails to dissect the binding preferences for this poorly understood reader domain family. With over 100 PDHs represented on our domain microarrays, we determined that the family strongly prefers the histone H3 tail. Furthermore, the majority of the domains that displayed binding preferred the higher orders of H3K4 methylation, with two subsets showing either a preference for H3K4me0, or no preference to the H3K4 methyl state.

Our findings from domain and peptide microarray confirm the reported binding preferences of many PHD proteins such as those of the ING and PHF families [ 6 , 24 ]. Additionally, the PHD finger from MLL5 was shown to robustly bind peptides containing each methyl state at H3K4 (me1-2-3) on the domain microarray and dCypher screen, while the peptide microarrays suggest MLL5 is a specific reader for H3K4me3. Intriguingly, we note that previous studies have found discrepancies in whether the PHD finger of MLL5 is a H3K4me3 or H3K4me2 reader [ 25 , 26 ]. We surmise that the basis of this difference may be due to the overall sensitivity of the various assays employed, which also may account for different observations in the literature. Nonetheless, our analyses provide strong support for MLL5 as a binder of H3K4 methylation on peptides. While recent work has suggested the disease relevance of MLL5 [ 26 ], few studies have characterized its histone PTM binding preferences and whether such interaction contributes to its normal or disease functions [ 25 ]. The domain microarrays also identified two poorly characterized proteins—TRIM66 and PHRF1—as readers of the unmodified H3 tail. Both proteins are E3 ligases that contain a PHD finger, but whose histone binding capabilities have not been well documented [ 29 , 30 , 31 ]. How these histone interactions contribute to the function of these ligases is currently unknown but will be interesting to determine in future studies.

While our domain microarrays revealed 31 out of 123 tested PHD proteins to be binders of the H3 N-terminus (Fig.  1 and Additional file 1 : Figures S1, S2), this does not preclude the potential for other PHD fingers to bind under alternate hybridization conditions or to unrepresented targets. Reader domain–histone PTM interactions are multifaceted, and while the results of this study’s domain array do confirm published observations as well as revealing new and interesting binding preferences, we point out that they are not meant to represent an exhaustive list of PHD-mediated interactions but rather to serve as a community resource.

Although domain microarrays are useful in probing many domains in high-throughput, they are limited by the ability to probe with one peptide of interest at a time. To further define the histone PTM landscape to which the subset of 31 PHD proteins identified in the domain microarray might bind, we employed the opposite approach of analyzing each individual domain against a microarray containing ~ 300 singly or combinatorially modified histone peptides (Fig.  2 ; Additional file 3 : Table S2). Through this approach, we were able to confirm many of the interactions observed on the domain microarray with respect to the H3K4me0/1/2/3 peptides. Significantly, the peptide microarray showed that PHRF1 (RP) specifically bound H3K4me0 over K4me, whereas it had no preference on the domain array—which may be explained by the fact that proteins and peptide concentrations on the domain microarrays are high, and thus may capture weak binding events that may not be observed on other platforms.

Despite the obvious potential of peptide microarrays, it would be remiss not to note possible limitations of the platform. The dynamic range of detected interactions is narrow, and from extensive experience, we are only able to characterize domain–peptide interactions on a four-point scale (very strong, strong, weak, or not detected). In addition, these interactions do not represent values that can be translated into binding affinities. Furthermore, comparing values between different probed arrays is also challenging given the lack of a platform control that can be used to normalize signals between arrays. We have also identified potential biophysical artifacts of the platform: we confirmed with these arrays that domains interacting with the H3 N-terminus are influenced by the neighboring acetylation status—a result observed in past publications with PHD readers using these or similar microarrays [ 37 , 38 ]. However, the impact of H3 acetylation on reader domain binding in the platform context appear to be indirect, as the solution-based binding reactions conclusively show that PHD fingers do not prefer H3K4me0-3 in the context of neighboring acetylation. Rather, it appears that streptavidin-coated slides may carry some amount of negative charge that binds the positively charged histone tails except when this is neutralized (e.g., by acetylation) and thus released from the surface. This “surface effect” shifts the H3 N-terminal binding preferences for many reader proteins towards acetylated peptides, but it is clear that the binding preferences for PHD fingers are primarily driven by direct interactions towards H3K4 ( ∓ methylation). Although this is a technical challenge, it does not preclude the use of peptide microarrays as the end user can be aware of the role of neighboring acetylation and how to put such results in context.

In contrast to the histone peptide microarrays, the dCypher AlphaScreen histone peptide assay has recently emerged as a highly sensitive and robust technique in gauging the binding interactions between reader domains and histone PTMs [ 35 ]. Furthermore, this method allows for the thorough optimization of reaction conditions in terms of buffers, protein/peptide/salt concentration, and cofactor/competitor additives to enable the study of otherwise poorly behaved proteins of interest. Given the advantages of this platform, we used the dCypher assay to first optimize the binding conditions for PHD fingers, and then proceeded to a variety of the PHD fusions that showed low/weak binding or novel histone PTM interactions on the microarrays. The dCypher approach is sensitive and benefits from an initial optimization step for each protein (see Fig.  3 a–h) to find the optimal concentration needed in the assay (see Fig.  3 i). Using this approach, we were able to confirm that several poorly characterized proteins including TRIM66 are indeed robust readers of H3K4me0 peptides. Intriguingly, the highly sensitive nature of the dCypher assay allowed comparison of peptide-binding signal at low versus high protein concentrations, which revealed that PHRF1 had a distinct binding preference for the H3K9me3 peptide over the H3K4me0 peptide. Importantly, the domain and peptide microarrays rely on micromolar reader domain concentrations, while the dCypher assay can reliably measure binding signal with proteins in the picomolar range. Thus, the dCypher screen revealed the ability of some domains to have distinct preferences at different concentrations that could not be determined from the other approaches. Whether such distinct histone binding preferences in the context of N-terminal peptides are physiologically relevant and could effectively represent the local concentration of particular reader domain on chromatin is currently unknown but is interesting to consider.

In this report, we have employed multiple high-throughput methods such as domain and peptide microarrays, as well as the proximity-based dCypher peptide screen to assemble a large dataset describing histone PTM binding preferences for PHDs, starting from a broad analysis of the entire family narrowing down to 31 histone H3-interacting readers. While we used the domain microarrays as an initial guide for which proteins to employ in further characterizations, we expect that further exploration of the remaining readers on this microarray platform will uncover additional interactions when binding conditions are further explored (e.g., the PHD domains of UHRF1/2 that were negative in the assays but reported to also bind H3 [ 39 , 40 ]). Assay development for studying chromatin-interacting proteins has been on the rise in the last decade and we believe that it will be necessary to understand how PHD readers interact with histone PTMs in a nucleosomal context alongside peptides to better replicate physiological conditions. Further, while the bulk of literature and indeed the focus of this study concerning PHD proteins has focused on their interactions with histones, the possibility of these readers binding non-histone biomolecules is intriguing and merits further study. Taken together, we expect our findings to serve as a resource for the chromatin community and to provide a framework for future studies regarding plant homeodomain proteins.

Protein domain array

The protein domain microarray was designed to include 123 GST-tagged PHD-domain containing recombinant proteins. Protein domain microarray development and probing was as previously [ 41 , 42 , 43 ]. Briefly, recombinant proteins were synthesized and cloned into pGEX-4T-1 vector by Biomatik Corporation. These GST-PHD readers were subsequently expressed, purified, and spotted in duplicate onto nitrocellulose-coated glass slides (Oncyte Avid slides, Grace Bio-Labs) using a pin arrayer (Aushon 2470, Aushon). For probing, microarray slides were blocked with 3% milk, 3% bovine serum albumin, 0.1% Tween 20 in PBS. Biotinylated peptides were pre-labeled with streptavidin-Cy3 fluorophore (GE Healthcare) and incubated with the blocked array slides. Slides were then washed with PBST and allowed to air dry. Fluorescent interactions were visualized using a GenePix 4200A Microarray Scanner (Molecular Devices).

Protein purification, histone peptide microarrays, and peptide pulldown assays

The 31 GST-tagged PHD readers identified in the PHD finger domain array were expressed and purified as previously [ 33 ]. Histone peptide arrays and peptide pulldown assays were conducted as recently described (specifically, the optimized protocol from Petell et al. for the former) [ 33 ].

dCypher Alphascreen peptide screen assay

The dCypher peptide screen assay was performed as previously described [ 35 ]. Briefly, 5 μL of GST-tagged reader domains (optimal protein concentration for library screening determined by initial binding curves to candidate peptides) were incubated with 5 μL of 400 nM (100 nM Final) biotinylated histone peptides ( EpiCypher ) for 30 min at 23 °C in 1× AlphaLISA Epigenetics buffer + epigenetics buffer supplement ( PerkinElmer , AL1008) in a 384-well plate. A 10 μL mix of 5 µg/mL (2.5 μg/mL final) glutathione Acceptor beads ( PerkinElmer , AL109M) and 10 μg/mL (5 μg/mL final) streptavidin Donor beads ( PerkinElmer , 6760002) was prepared in 1× [Epigenetics buffer + supplement] and added to each well. Plates were incubated at 23 °C in subdued lighting for 60 min and AlphaLISA signal measured on a PerkinElmer 2104 EnVision (680 nm laser excitation, 570 nm emission filter ± 50 nm bandwidth).

Availability of data and materials

The datasets used and/or analyzed during this study are included as additional files. All plasmids are available from the corresponding authors on request.

Abbreviations

plant homeodomain

post-translational modifications

Tandem Tudor domain + PHD

Dual PHD + Dual Chromodomain

PHD + Dual Chromodomain

CW-type Zn-finger

PHD + Bromodomain

Dual PHD + C2W2-type Zn-finger

SAND + PHD + Bromodomain

Su Z, Denu JM. Reading the combinatorial histone language. ACS Chem Biol. 2016;11:564–74. https://doi.org/10.1021/acschembio.5b00864 .

Article   CAS   PubMed   Google Scholar  

Strahl BD, Allis CD. The language of covalent histone modifications. Nature. 2000;403:41–5.

Article   CAS   Google Scholar  

Jenuwein T, Allis CD. Translating the histone code. Science (80−). 2001;293:1074–80.

Filippakopoulos P, Knapp S. The bromodomain interaction module. FEBS Lett. 2012;586:2692–704. https://doi.org/10.1016/j.febslet.2012.04.045 .

Blus BJ, Wiggins K, Khorasanizadeh S. Epigenetic virtues of chromodomains. Crit Rev Biochem Mol Biol. 2011;46:507–26. https://doi.org/10.3109/10409238.2011.619164 .

Article   CAS   PubMed   PubMed Central   Google Scholar  

Musselman CA, Kutateladze TG. Handpicking epigenetic marks with PHD fingers. Nucleic Acids Res. 2011;39:9061–71.

Ali M, Hom RA, Blakeslee W, Ikenouye L, Kutateladze TG. Diverse functions of PHD fingers of the MLL/KMT2 subfamily. Biochim Biophys Acta Mol Cell Res. 2014;1843:366–71.

Zhang X, Zhao D, Xiong X, He Z, Li H. Multifaceted histone H3 methylation and phosphorylation readout by the plant homeodomain finger of human nuclear antigen Sp100C. J Biol Chem. 2016;291:12786–98.

Biswas S, Rao CM. Epigenetic tools (The Writers, The Readers and The Erasers) and their implications in cancer therapy. Eur J Pharmacol. 2018;837:8–24.

Zhang Y, Yang H, Guo X, Rong N, Song Y, Xu Y, et al. The PHD1 finger of KDM5B recognizes unmodified H3K4 during the demethylation of histone H3K4me2/3 by KDM5B. Protein Cell. 2014;5:837–50. https://doi.org/10.1007/s13238-014-0078-4 .

Klein BJ, Piao L, Xi Y, Rincon-Arano H, Rothbart SB, Peng D, et al. The histone-H3K4-specific demethylase KDM5B binds to its substrate and product through distinct PHD fingers. Cell Rep. 2014;6:325–35.

Lan F, Collins RE, De Cegli R, Alpatov R, Horton JR, Shi X, et al. Recognition of unmethylated histone H3 lysine 4 links BHC80 to LSD1-mediated gene repression. Nature. 2007;448:718–22.

Mansfield RE, Musselman CA, Kwan AH, Oliver SS, Garske AL, Davrazou F, et al. Plant homeodomain (PHD) fingers of CHD4 are histone H3-binding modules with preference for unmodified H3K4 and methylated H3K9. J Biol Chem. 2011;286:11779–91. https://doi.org/10.1074/jbc.M110.208207 .

Musselman CA, Mansfield RE, Garske AL, Davrazou F, Kwan AH, Oliver SS, et al. Binding of the CHD4 PHD2 finger to histone H3 is modulated by covalent modifications. Biochem J. 2009;423:179–87. https://doi.org/10.1042/BJ20090870 .

Shi X, Kachirskaia I, Walter KL, Kuo J-HA, Lake A, Davrazou F, et al. Proteome-wide analysis in Saccharomyces cerevisiae identifies several PHD fingers as novel direct and selective binding modules of histone H3 methylated at either lysine 4 or lysine 36. J Biol Chem. 2007;282:2450–5.

Zeng L, Zhang Q, Li S, Plotnikov AN, Walsh MJ, Zhou M-M. Mechanism and regulation of acetylated histone binding by the tandem PHD finger of DPF3b. Nature. 2010;466:258–62.

Zhang Y, Jang Y, Lee JE, Ahn JW, Xu L, Holden MR, et al. Selective binding of the PHD6 finger of MLL4 to histone H4K16ac links MLL4 and MOF. Nat Commun. 2019;10:2314.

Article   Google Scholar  

Chen S, Yang Z, Wilkinson AW, Deshpande AJ, Sidoli S, Krajewski K, et al. The PZP domain of AF10 senses unmodified H3K27 to regulate DOT1L-mediated methylation of H3K79. Mol Cell. 2015;60:319–27.

Klein BJ, Lalonde M-E, Côté J, Yang X-J, Kutateladze TG. Crosstalk between epigenetic readers regulates the MOZ/MORF HAT complexes. Epigenetics. 2014;9:186–93. https://doi.org/10.4161/epi.26792 .

Dhar SS, Lee S-H, Kan P-Y, Voigt P, Ma L, Shi X, et al. Trans-tail regulation of MLL4-catalyzed H3K4 methylation by H4R3 symmetric dimethylation is mediated by a tandem PHD of MLL4. Genes Dev. 2012;26:2749–62.

Migliori V, Müller J, Phalke S, Low D, Bezzi M, Mok WC, et al. Symmetric dimethylation of H3R2 is a newly identified histone mark that supports euchromatin maintenance. Nat Struct Mol Biol. 2012;19:136–44.

Appikonda S, Thakkar KN, Barton MC. Regulation of gene expression in human cancers by TRIM24. Drug Discov Today Technol. 2016;19:57–63.

Agricola E, Randall RA, Gaarenstroom T, Dupont S, Hill CS. Recruitment of TIF1γ to chromatin via its PHD finger-bromodomain activates its ubiquitin ligase and transcriptional repressor activities. Mol Cell. 2011;43:85–96.

Champagne KS, Kutateladze TG. Structural insight into histone recognition by the ING PHD fingers. Curr Drug Targets. 2009;10:432–41.

Zhang X, Novera W, Zhang Y, Deng L-W. MLL5 (KMT2E): structure, function, and clinical relevance. Cell Mol Life Sci. 2017;74:2333–44.

Zhou P, Ding X, Wan X, Liu L, Yuan X, Zhang W, et al. MLL5 suppresses antiviral innate immune response by facilitating STUB1-mediated RIG-I degradation. Nat Commun. 2018;9:1243.

Ali M, Rincón-Arano H, Zhao W, Rothbart SB, Tong Q, Parkhurst SM, et al. Molecular basis for chromatin binding and regulation of MLL5. Proc Natl Acad Sci USA. 2013;110:11296–301.

Lemak A, Yee A, Wu H, Yap D, Zeng H, Dombrovski L, et al. Solution NMR structure and histone binding of the PHD domain of human MLL5. PLoS ONE. 2013;8:e77020.

Chen Y, Guo Y, Yang H, Shi G, Xu G, Shi J, et al. TRIM66 overexpression contributes to osteosarcoma carcinogenesis and indicates poor survival outcome. Oncotarget. 2015;6:23708–19.

PubMed   PubMed Central   Google Scholar  

Ettahar A, Ferrigno O, Zhang M-Z, Ohnishi M, Ferrand N, Prunier C, et al. Identification of PHRF1 as a tumor suppressor that promotes the TGF-β cytostatic program through selective release of TGIF-driven PML inactivation. Cell Rep. 2013;4:530–41.

Prunier C, Zhang M-Z, Kumar S, Levy L, Ferrigno O, Tzivion G, et al. Disruption of the PHRF1 tumor suppressor network by PML-RARα drives acute promyelocytic leukemia pathogenesis. Cell Rep. 2015;10:883–90.

Saare M, Hämarik U, Venta R, Panarina M, Zucchelli C, Pihlap M, et al. SP140L, an evolutionarily recent member of the SP100 family, is an autoantigen in primary biliary cirrhosis. J Immunol Res. 2015;2015:1–17.

Petell CJ, Pham AT, Skela J, Strahl BD. Improved methods for the detection of histone interactions with peptide microarrays. Sci Rep. 2019;9:6265.

Huber FM, Greenblatt SM, Davenport AM, Martinez C, Xu Y, Vu LP, et al. Histone-binding of DPF2 mediates its repressive role in myeloid differentiation. Proc Natl Acad Sci USA. 2017;114:6016–21.

Weinberg DN, Papillon-Cavanagh S, Chen H, Yue Y, Chen X, Rajagopalan KN, et al. The histone mark H3K36me2 recruits DNMT3A and shapes the intergenic DNA methylation landscape. Nature. 2019;573:281–6.

Baker LA, Allis CD, Wang GG. PHD fingers in human diseases: disorders arising from misinterpreting epigenetic marks. Mutat Res. 2008;647:3–12.

Kinkelin K, Wozniak GG, Rothbart SB, Lidschreiber M, Strahl BD, Cramer P. Structures of RNA polymerase II complexes with Bye1, a chromatin-binding PHF3/DIDO homologue. Proc Natl Acad Sci. 2013;110:15277–82.

Shanle EK, Shinsky SA, Bridgers JB, Bae N, Sagum C, Krajewski K, et al. Histone peptide microarray screen of chromo and Tudor domains defines new histone lysine methylation interactions. Epigenetics Chromatin. 2017;10:12.

Rothbart SB, Dickson BM, Ong MS, Krajewski K, Houliston S, Kireev DB, et al. Multivalent histone engagement by the linked tandem Tudor and PHD domains of UHRF1 is required for the epigenetic inheritance of DNA methylation. Genes Dev. 2013;27:1288–98. https://doi.org/10.1101/gad.220467.113 .

Rothbart SB, Krajewski K, Nady N, Tempel W, Xue S, Badeaux AI, et al. Association of UHRF1 with methylated H3K9 directs the maintenance of DNA methylation. Nat Struct Mol Biol. 2012;19:1155–60.

Espejo A, Côté J, Bednarek A, Richard S, Bedford MT. A protein-domain microarray identifies novel protein–protein interactions. Biochem J. 2002;367:697–702.

Kim J, Daniel J, Espejo A, Lake A, Krishna M, Xia L, et al. Tudor, MBT and chromo domains gauge the degree of lysine methylation. EMBO Rep. 2006;7:397–403. https://doi.org/10.1038/sj.embor.7400625 .

Chen J, Sagum C, Bedford MT. Protein domain microarrays as a platform to decipher signaling pathways and the histone code. Methods. 2019. https://doi.org/10.1016/j.ymeth.2019.08.007 .

Article   PubMed   Google Scholar  

Download references

Acknowledgements

We thank the Strahl and Bedford labs for helpful comments and suggestions.

This work was supported by NIH Grant GM126900 to BDS. The UT MDACC Protein Array & Analysis Core (PAAC) is supported by CPRIT Grant RP180804 (MTB). EpiCypher and MTB are supported by NIH Grant R44GM116584, with EpiCypher further supported by R44GM117683 and R44CA214076. KJ is supported by Postdoctoral Training Fellowship T32CA217824 from the NCI and UNC Lineberger Comprehensive Cancer Center.

Author information

Authors and affiliations.

Department of Biochemistry and Biophysics, The University of North Carolina, Chapel Hill, NC, 27599, USA

Kanishk Jain, Madison M. Parker, Krzysztof Krajewski & Brian D. Strahl

Lineberger Comprehensive Cancer Center, The University of North Carolina School of Medicine, Chapel Hill, NC, 27599, USA

Kanishk Jain, Caroline S. Fraser, Madison M. Parker & Brian D. Strahl

Curriculum in Genetics and Molecular Biology, The University of North Carolina, Chapel Hill, NC, 27599, USA

Caroline S. Fraser & Brian D. Strahl

EpiCypher Inc, Durham, NC, 27709, USA

Matthew R. Marunde, Jonathan M. Burg, Nathan Hall, Irina K. Popova, Keli L. Rodriguez, Anup Vaidya & Michael-Christopher Keogh

Department of Epigenetics and Molecular Carcinogenesis, The University of Texas MD Anderson Cancer Center, Smithville, TX, 78957, USA

Cari Sagum & Mark T. Bedford

You can also search for this author in PubMed   Google Scholar

Contributions

KJ, BDS, and MTB conceived the project, with input from all authors. KJ, CSF, and MMP purified proteins and performed peptide microarrays and peptide pulldowns. CSF and CS performed domain microarrays. MRM and JMB were responsible for dCypher assay design, with NH, IKP, KLR, and AV performing dCypher analyses. All authors analyzed the data and discussed the results. KJ and BDS wrote the manuscript with input from all authors. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Mark T. Bedford or Brian D. Strahl .

Ethics declarations

Ethics approval and consent to participate.

Not applicable.

Consent for publication

Competing interests.

BDS and MTB are co-founders, shareholders, and scientific advisory board members of EpiCypher, Inc. EpiCypher is a commercial developer and supplier of reagents (e.g., synthetic histone peptides) and the dCypher ® peptide-binding platform.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Additional file 1: table s1..

PHD finger Domain Array Constructs and Map. Excel Tab PHD Proteins is a compiled list of the 123 GST-tagged proteins used for the domain array. The list contains information on the domains, UniProt accession numbers, sequence coverage, and amino acid sequence for each protein. Excel Tab Array Map represents the pattern in which each of the GST-tagged proteins has been printed for the domain microarray.

Additional file 2: Figure S1.

Positive and negative controls of PHD finger domain arrays. Figure S2. PHD finger domain array with H3 (1-20) K4me1 and K4me2. Figure S3. Domain architecture of the 31 human PHD-containing proteins identified as hits for H3K4me0 or H3K4me3 via protein domain array (Fig.  1 ). Figure S4. Peptide arrays for 31 PHD-containing proteins. Figure S5. Peptide Pulldowns with KDM7A and KDM5B (PPC2W2).

Additional file 3: Table S2.

Histone Peptide Microarray Results. Excel Tab Peptide List is a compiled list of the entire 293 histone peptide library with amino acid sequence ranges. Excel Tab Peptide Array Grid Map is a table representing the pattern in which peptides have been printed for the peptide microarray. Each number corresponds to the peptide number designated in Peptide List. Excel Tab Average Signals is a compiled set of binding data: chromatin reader at 0.5 μM vs. entire 293 peptide library. Data is presented as the average and standard deviation of 4 replicates. The average signals are colored to indicate signal intensity with green being strong and white being weak/low. Excel Tab Array Heatmap contains a condensed form of Average Signals data displaying an average of 4 replicates for each chromatin reader tested and individually formatted to demonstrate binding specificity. Key: red= stronger relative binding; blue = weaker relative binding. Excel Tab Array info summary is a table summarizing the domains, sequence coverage, array signal range, and noise for each of the GST-tagged readers assayed on the peptide microarrays.

Additional file 4: Table S3.

dCypher Results. Excel Tab Peptide Phase A is a compiled set of binding data: chromatin reader titration vs. several peptide targets. Data is presented as raw Alpha counts in duplicates at each concentration tested. Excel Tab Peptide Phase B is a compiled set of binding data: chromatin reader at optimized concentrations vs. entire 293 peptide library. Data is presented as the average and standard deviation of duplicates. Excel Tab Peptide Phase B Heatmap contains a condensed form of Peptide Phase B data displaying an average of duplicates for each chromatin reader tested and individually formatted to demonstrate binding specificity. Key: red = strong binding; blue = weak/no binding.

Additional file 5: Table S4.

dCypher screen reader domain concentrations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Jain, K., Fraser, C.S., Marunde, M.R. et al. Characterization of the plant homeodomain (PHD) reader family for their histone tail interactions. Epigenetics & Chromatin 13 , 3 (2020). https://doi.org/10.1186/s13072-020-0328-z

Download citation

Received : 04 December 2019

Accepted : 13 January 2020

Published : 24 January 2020

DOI : https://doi.org/10.1186/s13072-020-0328-z

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Histone methylation
• PHD fingers
• Histone peptide microarray
• Protein domain microarray

Epigenetics & Chromatin

ISSN: 1756-8935

• Submission enquiries: [email protected]

Doctor of Philosophy in Human Genetics

The doctoral program in human genetics prepares students for careers leading genetics and genomics research in academia or industry. The flexible curriculum provides a broad background in the field while allowing customized emphasis on molecular genetics/genomics, statistical genetics and genetic epidemiology, or genetic counseling.

Our graduates will be independent scientists, innovators, and leaders of academe and private enterprise in a future where genomics is central to health sciences research, practice, and policy.

Areas of Strength

Students entering the doctoral program can tailor their training based on their research interests and career goals:

Molecular Genetics

Students pursuing training in molecular genetics or cytogenetics will carry out research projects performing benchtop experimental studies in laboratories within the department or across the Schools of the Health Sciences. Students can further customize their training by choosing among many elective courses in Cell Biology and Molecular Physiology, Cellular and Molecular Pathology, Immunology, Molecular Genetics and Developmental Biology, Molecular Pharmacology, and Molecular Virology and Microbiology.

Statistical Genetics and Genetic Epidemiology

Students pursuing training in statistical genetics and genetic epidemiology will carry out applied analysis or methodological research projects related to clinical or epidemiological studies. Students will work as members of statistical and computational research groups within the department or across the School of the Health Sciences. Students can further customize their training by choosing advanced coursework in biostatistics, biomedical informatics, computational and systems biology, and epidemiology.

Genetic Counseling

Students holding a MS GC degree who wish to pursue a doctoral degree with a focus in genetic counseling will carry out original research in their area(s) of interest. In addition to addressing basic science or clinical research questions, students’ projects will explore the medical, psychological, and familial implications of genetic contributors to human health and disease. Students can further customize their training by choosing elective courses from across the Schools of the Health Sciences in consultation with the program leadership team.

Customize through Certificate Programs

Doctoral students can shape their training experience through one or more of several  certificate programs  offered at Pitt Public Health.

• Integrity: Commit to honest and ethical conduct in research and scholarship
• Scientific rigor: Dedicate to excellence in the pursuit of knowledge
• Cooperation: Commit to collaboration, mentorship, and academic-private partnerships
• Collegiality: Respect independence of thought
• Cultural competency: Embrace diversity; respect and value differences and individuality

Contact HUGEN

Noel C. Harrie Student services coordinator Phone: 412-624-3066 Fax: 412-624-3020 j [email protected]

University of Pittsburgh School of Public Health Department of Human Genetics 3120C Public Health 130 DeSoto Street Pittsburgh, PA 15261

• Student Handbook

Genome-wide identification and expression analysis of the PHD-finger gene family in Solanum tuberosum

Affiliations.

• 1 College of Life Sciences, Fujian Agriculture & Forestry University, Fuzhou, China.
• 2 Key Laboratory for Genetics, Breeding and Multiple Utilization of Crops, Ministry of Education, Fujian Agriculture & Forestry University, Fuzhou, China.
• 3 Fujian Key Laboratory of Crop Breeding by Design, Fujian Agriculture & Forestry University, Fuzhou, China.
• 4 The Crop Institute, Fujian Academy of Agricultural Sciences, Fuzhou, Fujian, China.
• PMID: 31881057
• PMCID: PMC6934267
• DOI: 10.1371/journal.pone.0226964

Plant homeodomain (PHD) proteins are prevalent in eukaryotes and play important roles in plant growth, development and abiotic stress response. In this study, the comprehensive study of the PHD family (StPHD) was performed in potato (Solanum tuberosum L.). Seventy-two PHD genes (named StPHD1-72) were identified and grouped into 10 subfamilies based on phylogenetic analysis. Similar structure organizations were found within each subfamily according to the exon/intron structures and protein motif analysis. These genes were unequally scattered on the chromosomes of potato, with 9 pairs of segmental duplicated genes and 6 pairs of tandem duplicated genes showing that both segmental duplicated and tandem duplicated events contributed to the expansion of the potato PHD family. The gene ontology (GO) analysis suggests that StPHD mainly functioned at the intracellular level and was involved in various binding, metabolic and regulation processes. The analysis of expression patterns of StPHD genes showed that these genes were differentially expressed in 10 different tissues and responded specifically to heat, salt and drought stress based on the FPKM (Fragments per kilobase of transcript per million mapped reads) values of the RNA-seq data. Furthermore, the real-time quantitative PCR for 12 selected StPHD genes revealed the various levels of gene expression corresponding to abiotic stress. Our results provide useful information for a better understanding of PHD genes and provide the foundation for additional functional exploration of the potato PHD gene family.

Publication types

• Research Support, Non-U.S. Gov't
• Gene Expression Profiling
• Gene Expression Regulation, Plant
• Genes, Plant
• Genome-Wide Association Study
• Homeodomain Proteins / genetics*
• Multigene Family
• Plant Proteins / genetics*
• Solanum tuberosum / genetics*
• Stress, Physiological
• Homeodomain Proteins
• Plant Proteins

Grants and funding

What is a PhD?

• Types of Doctorates
• A Doctor of Philosophy (PhD) is the highest globally recognized postgraduate degree that higher education institutions can award.
• PhDs are awarded to candidates who undertake original and extensive research in a particular field of study.
• Full time PhD programmes typically last three to four years, whilst part time PhD programmes typically last six to seven years.
• A PhD can lead to an academia teaching role or a career in research. A PhD can also equip you with skills suitable for a wide range of jobs unrelated to your research topic or academia.

Definition of a PhD – A Doctor of Philosophy (commonly abbreviated to PhD , Ph.D or a DPhil ) is a university research degree awarded from across a broad range of academic disciplines; in most countries, it is a terminal degree, i.e. the highest academic degree possible.

PhDs differ from undergraduate and master’s degrees in that PhDs are entirely research-based rather than involving taught modules (although doctoral training centres (DTCs) offer programmes that start with a year of lecture-based teaching to help develop your research skills prior to starting your project).

In most English-speaking countries, those that complete a PhD use the title “Doctor” (typically abbreviated to Dr) in front of their names and are referred to as such within academic and/or research settings. Those that work in fields outside of academia may decide not to use the formal doctor title but use post-nominal letters (e.g. John Smith PhD); it’s unusual though for someone to use both the Doctor title and post-nominal letters in their name.

PhD vs Doctorate

A PhD and a professional doctorate are both research-based terminal degrees.

However, where a PhD focuses on original research mostly around theoretical concepts, a professional doctorate focuses on examining existing knowledge to solve real-life, practical problems.

While there is much crossover between the two, a PhD is generally better suited for an individual to wants to advance the knowledge and understanding in their field, and a professional doctorate degree is better suited to a working professional who wants to better be able to apply knowledge and understanding to their field.

What Are the Entry Requirements for a PhD?

To be accepted on to a PhD programme, students usually need to hold at least a high ( 2:1 and above ) undergraduate degree that is related to the field of research that they want to pursue. A PhD candidate may also be expected to hold a Master’s degree , however, this does not mean you must have one, as it is still possible to enrol into a PhD without a Master’s .

Self-funded courses may sometimes be more relaxed in relation to entry requirements. It may be possible to be accepted onto a self-funded PhD programme with lower grades, though these students typically demonstrate their suitability for the role through professional work experience.

Whilst a distance learning project is possible , most PhD candidates will carry out their research over at least three years based at their university, with regular contact with two academic supervisors (primary and secondary). This is particularly the case for lab-based projects, however, some PhD projects require spending time on-site away from university (e.g. at a specialist research lab or at a collaborating institution abroad).

How Long Does a PhD Take?

Typically, full-time PhDs last 3-4 years and part-time PhDs last 6-7 years. However, at the discretion of the university, the thesis writing-up period can be extended by up to four years.

Although most doctoral programmes start in September or October, they are generally much more flexible than taught-courses and can start at any time of the year.

How Much Does a PhD Cost?

Tuition fees for UK and EU students vary between £3,000 and £6,000 per year, with the average tuition fee of £4,712 per year for 2023/24 programmes.

Tuition fees increase considerably for international students, varying between £16,000 to £25,000 per year, with an average tuition fee of £19,600 per year .

Nonetheless, most students will secure PhD funding in the form of studentships, scholarships and bursaries to help pay for these fees. These funding opportunities can either be partial, which cover tuition fees only, or full, which cover both tuition fees and living expenses.

UK national students can also apply for Doctoral Loans from Student Finance England if they are unable to secure funding.

Finding a PhD has never been this easy – search for a PhD by keyword, location or academic area of interest.

What Does a PhD Involve?

To be awarded a PhD, a doctoral student is required to produce a substantial body of work that adds new knowledge to their chosen field.

A PhD programme will typically involve four key stages:

Stage 1: Literature Review

The first year of a PhD involves attending regular meetings with your supervisors and carrying out a search on previously published work in your subject area. This search will be used to produce a literature review which should set the context of the project by explaining the foundation of what is currently known within the field of research, what recent developments have occurred, and where the gaps in knowledge are. In most cases, this will be an extension of your research proposal should you have produced one as part of your application. The literature review should conclude by outlining the overarching aims and objectives of the research project. This stage of setting achievable goals which are original and contribute to the field of research is an essential first step in a successful PhD.

The supervisor is the main point of contact through the duration of a PhD – but remember: they are there to mentor, not to teach, or do it for you . It will be your responsibility to plan, execute and monitor your own work as well as to identify gaps in your own knowledge and address them.

Stage 2: Research

The second year (and prehapse some of your third year) is when you work on your research. Having identified novel research questions from your review of the literature, this is where you collect your data to help answer these questions. How you do this will depend on the nature of your doctoral research: for example, you may design and run experiments in a lab alongside other PhD students or visit excavation sites in remote regions of the world. You should check in regularly with your supervisors to update them and run any ideas or issues past them.

Have the structure and chapters of your thesis in mind as you develop and tackle your research questions. Working with a view of publishing your work will be very valuable later on.

Stage 3: Write up of Thesis

The next key stage of a PhD is writing a doctoral thesis , which typically takes from anywhere between three months to one year. A thesis is a substantial body of work that describes the work and outcomes of the research over the previous two to three years. It should tell a detailed story of the PhD project – focusing on:

• The motivations for the research questions identified from the literature review.
• The methodologies used, results obtained, and a comprehensive analysis and discussion of the findings.
• A detailed discussion of the key findings with an emphasis on the original contributions made to your field of research and how this has been impactful.

There is no universal rule for the length of a PhD thesis, but general guidelines set the word count between 80,000 to 100,000 words.

For your thesis to be successful, it needs to adequately defend your argument and provide a unique or increased insight into your field that was not previously available.

Stage 4: Attending the Viva

A viva voce , most commonly referred to as just a ‘ viva ‘, is an interview-style examination where the PhD student is required to engage in a critical appraisal of their work and defend their thesis against at least two examiners. The examiners will ask questions to check the PhD student has an in-depth understanding of the ideas and theories proposed in their thesis, and whether they have developed the research skills that would be expected of them.

The viva is one of the final steps in achieving a PhD, and typically lasts at least two hours, but this duration can vary depending on the examiners, the university and the PhD project itself.

Once you have done the viva – you’re on the home stretch. You will typically be asked to make some amendments to your thesis based on the examiner’s feedback. You are then ready to submit your final thesis for either:

• PhD – If you pass the requirements you will be awarded a PhD degree (most common outcome),
• MPhil – If you failed to meet requirements for a PhD, you may be downgraded to an MPhil degree (uncommon outcome),
• Fail – No award is given, typically for cases of plagiarism (extremely uncommon outcome).

What Is It Like to Undertake a PhD?

We’re often asked what it is like to undertake a PhD study. Unfortunately, this isn’t a simple answer to this question as every research project is different.

To help give insight into the life of a PhD student, we’ve interviewed PhD students at various stages of their programmes and put together a series of PhD Student Interviews . Check out the link to find out what a PhD is like and what advice they have to offer you.

What Are the Benefits of A PhD?

A PhD is the highest globally recognised postgraduate degree that higher education institutions can award. The degree, which is awarded to candidates who demonstrate original and independent research in a particular field of study, is not only invaluable in itself, but sets you up with invaluable skills and traits.

Career Opportunities

First, a PhD prepares you for a career in academia if you wish to continue in this area. This takes form as a career in the Higher Education sector, typically as a lecturer working their way to becoming a professor leading research on the subject you’ve studied and trained in.

Second, a PhD also enables the opportunity for landing a job in a research & development role outside of the academic environment. Examples of this include laboratory work for a private or third sector company, a governmental role and research for commercial and industrial applications.

Transferable Skills

Finally, in possessing a PhD degree, you can show to employers that you have vital skills that make you an asset to any company. Three examples of the transferable skills that you gain through a PhD are effective communication, time management, and report writing.

• Communication – presenting your work in written and oral forms using journal papers and podium presentations, shows your ability to share complex ideas effectively and to those with less background knowledge than you. Communication is key in the professional environment, regardless of the job.
• Time management – The ability to prioritise and organise tasks is a tremendous asset in the professional industry. A PhD holder can use their qualification to demonstrate that they are able to manage their time, arrange and follow a plan, and stick to deadlines.
• Report writing – Condensing three years of work into a thesis demonstrates your ability to filter through massive amounts of information, identify the key points, and get these points across to the reader. The ability to ‘cut out the waffle’ or ‘get to the point’ is a huge asset in the professional industry.

Aside from the above, you also get to refer to yourself as a Doctor and add fancy initials after your name!

What Can I Do After a PhD?

One of the most desirable postdoctoral fields is working within independent Research and Development (R&D) labs and new emerging companies. Both industries, especially R&D labs, have dedicated groups of PhD graduates who lead research activities, design new products and take part in crucial strategic meetings. Not only is this a stimulating line of work, but the average salaries in R&D labs and emerging start-ups are lucrative. In comparison, an undergraduate with five years of experience within their given field will, on average, likely earn less than a new PhD graduate taking on a R&D position.

It’s a common misunderstanding that PhDs only opens the door for an academic career such as university lecturers and training providers. Although obtaining a PhD opens these doors, the opportunities extend far beyond educational roles. In fact, recent data from the UK’s Higher Education Statistics Agency (HESA) indicates only 23% of PhD graduates take a position in educational roles . This low percentage is primarily because PhD graduates have a wide range of skills that make them suitable for a broad spectrum of roles. This is being seen first hand by the increasing number of PhD graduates who are entering alternative roles such as research, writing, law and investment banking.

How Do I Find a PhD?

We appreciate that finding a PhD programme to undertake can be a relatively daunting process. According to Higher Education Student Statistics , over 22,000 PhDs were awarded in 2016/17 within the United Kingdom alone. Clearly there are a huge number of PhD programmes available. This can sometimes be confusing for prospective doctorates, particularly when different programmes are advertised in different places. Often, it is difficult to know where to look or where to even start. We’ve put together a list of useful sources to find the latest PhD programmes:

• A great place to start is with our comprehensive and up-to-date database of available PhD positions .
• Assuming you are still at university, speak to an existing PhD supervisor within your department.
• Attend as many postgraduate open days as you can. Whilst there, speak to current PhD students and career advisors to get an awareness of what PhDs are on offer.
• Visit the postgraduate section of university websites and the PhD Research Council section of the UKRI website.

Browse PhDs Now

Join thousands of students.

Join thousands of other students and stay up to date with the latest PhD programmes, funding opportunities and advice.

• Share full article

Advertisement

Supported by

Study Suggests Genetics as a Cause, Not Just a Risk, for Some Alzheimer’s

People with two copies of the gene variant APOE4 are almost certain to get Alzheimer’s, say researchers, who proposed a framework under which such patients could be diagnosed years before symptoms.

By Pam Belluck

Scientists are proposing a new way of understanding the genetics of Alzheimer’s that would mean that up to a fifth of patients would be considered to have a genetically caused form of the disease.

Currently, the vast majority of Alzheimer’s cases do not have a clearly identified cause. The new designation, proposed in a study published Monday, could broaden the scope of efforts to develop treatments, including gene therapy, and affect the design of clinical trials.

It could also mean that hundreds of thousands of people in the United States alone could, if they chose, receive a diagnosis of Alzheimer’s before developing any symptoms of cognitive decline, although there currently are no treatments for people at that stage.

The new classification would make this type of Alzheimer’s one of the most common genetic disorders in the world, medical experts said.

“This reconceptualization that we’re proposing affects not a small minority of people,” said Dr. Juan Fortea, an author of the study and the director of the Sant Pau Memory Unit in Barcelona, Spain. “Sometimes we say that we don’t know the cause of Alzheimer’s disease,” but, he said, this would mean that about 15 to 20 percent of cases “can be tracked back to a cause, and the cause is in the genes.”

The idea involves a gene variant called APOE4. Scientists have long known that inheriting one copy of the variant increases the risk of developing Alzheimer’s, and that people with two copies, inherited from each parent, have vastly increased risk.

The new study , published in the journal Nature Medicine, analyzed data from over 500 people with two copies of APOE4, a significantly larger pool than in previous studies. The researchers found that almost all of those patients developed the biological pathology of Alzheimer’s, and the authors say that two copies of APOE4 should now be considered a cause of Alzheimer’s — not simply a risk factor.

The patients also developed Alzheimer’s pathology relatively young, the study found. By age 55, over 95 percent had biological markers associated with the disease. By 65, almost all had abnormal levels of a protein called amyloid that forms plaques in the brain, a hallmark of Alzheimer’s. And many started developing symptoms of cognitive decline at age 65, younger than most people without the APOE4 variant.

“The critical thing is that these individuals are often symptomatic 10 years earlier than other forms of Alzheimer’s disease,” said Dr. Reisa Sperling, a neurologist at Mass General Brigham in Boston and an author of the study.

She added, “By the time they are picked up and clinically diagnosed, because they’re often younger, they have more pathology.”

People with two copies, known as APOE4 homozygotes, make up 2 to 3 percent of the general population, but are an estimated 15 to 20 percent of people with Alzheimer’s dementia, experts said. People with one copy make up about 15 to 25 percent of the general population, and about 50 percent of Alzheimer’s dementia patients.

The most common variant is called APOE3, which seems to have a neutral effect on Alzheimer’s risk. About 75 percent of the general population has one copy of APOE3, and more than half of the general population has two copies.

Alzheimer’s experts not involved in the study said classifying the two-copy condition as genetically determined Alzheimer’s could have significant implications, including encouraging drug development beyond the field’s recent major focus on treatments that target and reduce amyloid.

Dr. Samuel Gandy, an Alzheimer’s researcher at Mount Sinai in New York, who was not involved in the study, said that patients with two copies of APOE4 faced much higher safety risks from anti-amyloid drugs.

When the Food and Drug Administration approved the anti-amyloid drug Leqembi last year, it required a black-box warning on the label saying that the medication can cause “serious and life-threatening events” such as swelling and bleeding in the brain, especially for people with two copies of APOE4. Some treatment centers decided not to offer Leqembi, an intravenous infusion, to such patients.

Dr. Gandy and other experts said that classifying these patients as having a distinct genetic form of Alzheimer’s would galvanize interest in developing drugs that are safe and effective for them and add urgency to current efforts to prevent cognitive decline in people who do not yet have symptoms.

“Rather than say we have nothing for you, let’s look for a trial,” Dr. Gandy said, adding that such patients should be included in trials at younger ages, given how early their pathology starts.

Besides trying to develop drugs, some researchers are exploring gene editing to transform APOE4 into a variant called APOE2, which appears to protect against Alzheimer’s. Another gene-therapy approach being studied involves injecting APOE2 into patients’ brains.

The new study had some limitations, including a lack of diversity that might make the findings less generalizable. Most patients in the study had European ancestry. While two copies of APOE4 also greatly increase Alzheimer’s risk in other ethnicities, the risk levels differ, said Dr. Michael Greicius, a neurologist at Stanford University School of Medicine who was not involved in the research.

“One important argument against their interpretation is that the risk of Alzheimer’s disease in APOE4 homozygotes varies substantially across different genetic ancestries,” said Dr. Greicius, who cowrote a study that found that white people with two copies of APOE4 had 13 times the risk of white people with two copies of APOE3, while Black people with two copies of APOE4 had 6.5 times the risk of Black people with two copies of APOE3.

“This has critical implications when counseling patients about their ancestry-informed genetic risk for Alzheimer’s disease,” he said, “and it also speaks to some yet-to-be-discovered genetics and biology that presumably drive this massive difference in risk.”

Under the current genetic understanding of Alzheimer’s, less than 2 percent of cases are considered genetically caused. Some of those patients inherited a mutation in one of three genes and can develop symptoms as early as their 30s or 40s. Others are people with Down syndrome, who have three copies of a chromosome containing a protein that often leads to what is called Down syndrome-associated Alzheimer’s disease .

Dr. Sperling said the genetic alterations in those cases are believed to fuel buildup of amyloid, while APOE4 is believed to interfere with clearing amyloid buildup.

Under the researchers’ proposal, having one copy of APOE4 would continue to be considered a risk factor, not enough to cause Alzheimer’s, Dr. Fortea said. It is unusual for diseases to follow that genetic pattern, called “semidominance,” with two copies of a variant causing the disease, but one copy only increasing risk, experts said.

The new recommendation will prompt questions about whether people should get tested to determine if they have the APOE4 variant.

Dr. Greicius said that until there were treatments for people with two copies of APOE4 or trials of therapies to prevent them from developing dementia, “My recommendation is if you don’t have symptoms, you should definitely not figure out your APOE status.”

He added, “It will only cause grief at this point.”

Finding ways to help these patients cannot come soon enough, Dr. Sperling said, adding, “These individuals are desperate, they’ve seen it in both of their parents often and really need therapies.”

Pam Belluck is a health and science reporter, covering a range of subjects, including reproductive health, long Covid, brain science, neurological disorders, mental health and genetics. More about Pam Belluck

The Fight Against Alzheimer’s Disease

Alzheimer’s is the most common form of dementia, but much remains unknown about this daunting disease..

How is Alzheimer’s diagnosed? What causes Alzheimer’s? We answered some common questions .

A study suggests that genetics can be a cause of Alzheimer’s , not just a risk, raising the prospect of diagnosis years before symptoms appear.

Determining whether someone has Alzheimer’s usually requires an extended diagnostic process . But new criteria could lead to a diagnosis on the basis of a simple blood test .

The F.D.A. has given full approval to the Alzheimer’s drug Leqembi. Here is what to know about i t.

Alzheimer’s can make communicating difficult. We asked experts for tips on how to talk to someone with the disease .