Fig 1.
Schematic map of the human β-globin locus and characterization of the δβ intergenic promoter.
(A) Schematic diagram of the human β-globin locus and the δβ intergenic promoter region. Black boxes represent globin genes; grey box corresponds to the β-like pseudogene, open boxes–to olfactory receptor genes flanking the β-globin locus. Vertical arrows indicate DNase hypersensitive sites, including sites 1 to 5 of the locus control region (LCR); horizontal arrows denote intergenic transcription start sites. The detailed map of the δβ promoter region shows the positions of the intergenic transcription start site and upstream Alu element (left horizontal open arrow). Restriction fragments from the δβ promoter region were cloned upstream of a promoterless EGFP gene as indicated and described in Materials and Methods. Relevant restriction sites are indicated. (B) EGFP reporter assays. Constructs were stably transfected into K562 cells and the percentage of EGFP-expressing cells was determined by flow cytometry. The Apo construct, which showed the highest percentage of EGFP-positive cells, was assigned an arbitrary value of 100.
Fig 2.
Analysis of transcription in the δβ intergenic promoter region in human erythroid cells.
(A) Schematic diagram of the human β-globin locus and the δβ intergenic promoter region. Black boxes represent globin genes; grey box corresponds to the β-like pseudogene, open boxes correspond to olfactory receptor genes flanking the β-globin locus. Vertical arrows indicate DNase hypersensitive sites, including sites 1 to 5 of the locus control region (LCR); horizontal arrows denote putative intergenic transcription start sites. The detailed map of the δβ promoter region shows the positions of the intergenic transcription start site and upstream Alu element (left horizontal open arrow). Restriction enzyme sites are indicated by vertical lines. The graph below the δβ intergenic promoter schematic indicates microarray signal intensity for various probes (indicated under the graph with horizontal black boxes) following labelling and hybridisation of RNA extracted from primary erythroid cells at different phases (day 4 in light grey, day 5 in dark grey and day 7 in black) of phase II culture. The results shown are representative of five biological replicates (B) The RNA samples used for microarray analysis in A were converted into cDNA using a WTA kit and amplified by PCR. The PCR amplicons are indicated beneath the microarray signal graph as U (upstream) amplicon and D (downstream) amplicon. The amounts of each amplicon were calculated by comparing signals to a standard curve generated with genomic DNA and the ratio of downstream over upstream was calculated. The ratio increases through phase II indicating a relative increase in transcripts downstream of the putative intergenic promoter. Results are representative of two biological replicates. Pearson correlation analysis indicates a statistically significant correlation (p<0.01). (C) RNA from day 7 of erythroid differentiation was converted into cDNA using either gene-specific primer (the downstream-amplicon forward primer [Downstream-F], the downstream-amplicon reverse primer [Downstream-R] or a combination of the two primers [Downstream-F+R] as indicated to the left of each panel). PCR was then carried out using both Downstream-F and Downstream-R primers (to amplify the downstream amplicon). GAPDH primers were used as a control to confirm RNA integrity. PCR products were run on a 2% gel, stained with SYBR Gold, and visualised on a Typhoon Imager. The strand-specific results show that the transcripts detected downstream of the δβ intergenic promoter are sense transcripts (travelling towards HBD).
Fig 3.
Generation of YAC transgenic mice carrying a floxed δβ intergenic promoter and production of the deletion lines.
(A) Schematic diagram of the human β-globin locus YAC used to generate transgenic lines FX14, FX101, and FX115. Black boxes represent globin genes; the grey box corresponds to the β-like pseudogene; open boxes indicate olfactory receptor genes. Vertical arrows denote DNase hypersensitive sites. Horizontal arrow refers to the δβ intergenic transcription start site. The δβ promoter region is shown at a larger scale below the main map. Black triangles denote loxP sites inserted upstream and downstream of the δβ promoter. White horizontal arrows indicate Alu repeat elements flanking the δβ promoter. SfiI and selected BglII restriction sites are indicated, and probes are shown as grey boxes below the lines (probes 1 to 4 and probe D described in Materials and Methods). (B) YAC transgenic lines carrying the human β-globin locus, before Cre-mediated deletion of the intergenic promoter. Genomic DNA from three transgenic lines, FX14, FX101, and FX115, was digested with SfiI, subjected to pulsed-field electrophoresis and Southern transfer. Several identical DNA samples from each line were run on the same gel, blots were cut into strips and each strip was hybridized with a different probe as indicated. Probes throughout the β-globin locus identify identical SfiI fragments. (C) Transgenic mice with deleted δβ promoter. Genomic DNA from line FX115 and 5 animals carrying the deletion, Δ1 to Δ5, was digested with SfiI, subjected to pulsed-field electrophoresis, transferred and hybridized with probe D. A single band corresponding to the β-globin SfiI fragment is detected in Δ1 to Δ5. (D) Efficient deletion of the δβ intergenic promoter. Southern blot analysis of BglII-digested genomic DNA from parent line FX115 and Δ1 to Δ5 hybridized to probe D. Note size reduction of BglII band after deletion of the δβ-promoter.
Fig 4.
Human β-globin transcription in transgenic mice before and after deletion of the δβ intergenic promoter.
(A) RNA FISH analysis of human HBE and HBG transcription in E10.5d blood cells from heterozygous transgenic embryos. HBE primary transcripts are in red, HBG transcripts in green; for scale, see (C), scale bar, 5 μm. (B) Normal transcription of HBE and HBG genes in E10.5d embryonic blood cells by RNA FISH. Cells positive for human globin gene transcription signals were scored; in each line nearly 100% of the embryonic blood cells have transcription signals for HBE, HBG or both genes indicating normal high-level expression. (C) RNA FISH analysis of HBB transcription in adult erythroid cells. HBB primary transcripts are in red, mouse Hbb-b1 transcripts detected in green as an internal control to identify erythroid precursors; scale bar, 5 μm. (D) Reduced transcription of HBB in adult erythroid cells carrying a δβ promoter deletion. Heterozygous mice from wild-type lines FX14 and 264W, lines FX101 and FX115 with ‘floxed’ δβ promoter, and Δ115 with deleted δβ promoter were analyzed by RNA FISH. Nuclei with a HBB signal were scored and expressed as a percentage of Hbb-b1 positive nuclei. HBB transcription is impaired in lines FX101 and FX115, and further reduced in lineΔ115. (E) Reduced transcription of human HBD and HBB genes in adult erythroid cells carrying a δβ promoter deletion. Animals from wild-type lines FX14 and 264W, and line Δ115 were analyzed by real-time RT-PCR, as described in Materials and Methods. Primary transcript levels are corrected for transgene copy number; wild-type HBB level in 264W is assigned an arbitrary value of 100.
Fig 5.
Intergenic transcription in the β-globin locus before and after deletion of the δβ intergenic promoter.
(A) RT-PCR to detect intergenic transcripts downstream of the δβ intergenic promoter in adult erythroid cells. The positions of PCR amplicons RT3B and RT4 are indicated below the map; black boxes, HBD and HBB genes; horizontal arrow, normal position of the δβ intergenic promoter and transcript start site. Intergenic transcription is reduced in lines FX115 and Δ115 with ‘floxed’ and deleted δβ intergenic promoter, respectively. (B) Transfection assays to assess intergenic promoter activity. Fragments containing the δβ intergenic promoter or corresponding fragments from the deletion lines were PCR amplified from lines 264W, FX115, and Δ115, and cloned upstream of a promoterless EGFP gene as described in Materials and Methods. Reporter constructs were stably transfected into MEL cells and the number of EGFP-expressing cells in the population was determined by flow cytometry. The percentage of cells expressing EGFP is severely decreased compared to wild-type δβ promoter (assigned an arbitrary value of 100) and CMV promoter when driven by FX115- and Δ115-derived sequences.
Fig 6.
HBB expression is variegated in transgenic mice carrying a deletion of the δβ intergenic promoter.
RNA FISH analysis of HBB transcripts in adult erythroid cells; primary and mature HBB transcripts are detected in green by an exon probe; as an internal control, Hbb-b1 primary transcripts are detected in red to identify erythroid cells; scale bar, 5 μm. Greater than 95% of FX14 erythroid cells have nuclear primary HBB signals and cytoplasmic HBB mRNA staining (top row). Conversely, the majority of FX115 and Δ115 cells lack primary HBB signals and do not contain cytoplasmic HBB mRNA (cells on the left). The few FX115 and Δ115 cells that do express HBB mRNA in the cytoplasm also have nuclear HBB primary transcript signals and vice-versa (cells on the right) showing that HBB expression is restricted to a small subset of adult erythroid cells.
Fig 7.
Histone modifications across the human β-globin locus in wt and δβ promoter deleted adult erythroid cells.
Chromatin immunoprecipitation assays for histone H3 acetylation (H3ac), di-methylated lysine 4 of histone H3 (H3K4me2), and tri-methylated lysine 4 of histone H3 (H3K4me3); ●, wild-type adult erythroid cells from line 264W [10]; , adult erythroid cells from line Δ115. Coordinates in bp are shown along the x axis with HBE start site as +1. Bound versus input ratios were calculated and normalized to the most 3’ primer pair, located just downstream of 3’HS1 in the ORG cluster. A map of the locus is shown below the graphs.