Enhancers of the PAIR4 regulatory module promote distal VH gene recombination at the Igh locus

Abstract

While extended loop extrusion across the entire Igh locus controls V_H -DJ_H recombination, local regulatory sequences, such as the PAIR elements, may also activate V_H gene recombination in pro-B-cells. Here, we show that PAIR-associated V_H 8 genes contain a conserved putative regulatory element (V8E) in their downstream sequences. To investigate the function of PAIR4 and its V8.7E, we deleted 890 kb containing all 14 PAIRs in the Igh 5' region, which reduced distal V_H gene recombination over a 100-kb distance on either side of the deletion. Reconstitution by insertion of PAIR4-V8.7E strongly activated distal V_H gene recombination. PAIR4 alone resulted in lower induction of recombination, indicating that PAIR4 and V8.7E function as one regulatory unit. The pro-B-cell-specific activity of PAIR4 depends on CTCF, as mutation of its CTCF-binding site led to sustained PAIR4 activity in pre-B and immature B-cells and to PAIR4 activation in T-cells. Notably, insertion of V8.8E was sufficient to activate V_H gene recombination. Hence, enhancers of the PAIR4-V8.7E module and V8.8E element activate distal V_H gene recombination and thus contribute to the diversification of the BCR repertoire in the context of loop extrusion.

Keywords: CTCF; Igh VH gene recombination; PAIR4 element; Pax5; novel recombination enhancers.

Figure 1. Identification of the V8E elements as a new class of potential regulatory elements in the distal V_H gene region of the Igh locus

1. Mapping of total transcripts at the Igh locus in ex vivo sorted pro‐B and pre‐B cells. Total RNA was sequenced, as described in Materials and Methods, and plotted as reads per million mapped sequence reads (RPM). The locations of the PAIR elements, V_H8 genes, IGCR1 sequence, and Eμ enhancer are shown together with the positions of the bacterial artificial chromosomes (BACs) RP23‐340K14 and RP24‐275L15 used as DNA‐FISH probes. The annotation of the C57BL/6 Igh locus indicates the distinct V_H gene families (different colors) in the distal, middle, and proximal V_H gene regions (Johnston et al, 2006), the D_H (light blue) and C_H (blue) elements, as well as the 3′RR enhancers (red) in the 3′ proximal Igh domain, together with the mm9 genomic coordinates of mouse chromosome 12. A black line indicates the extent of the 890‐kb deletion present in the Igh ^∆890 allele.

2. Presence of active histone marks and open chromatin at PAIR4 and the V8.7E element located immediately downstream of the V_H8‐7 gene in pro‐B cells. Ex vivo sorted pro‐B cells were used for determining open chromatin by ATAC‐seq and short‐term cultured pro‐B cells for ChIP‐seq analysis of the indicated histone marks with modification‐specific antibodies. The peak of open chromatin at the V8.7E element is indicated by a dashed line.

3. Specific presence of an open chromatin peak downstream of members of the V_H8 gene family. The density of cumulative ATAC‐seq reads located downstream of members of the V_H8 and V_H1 gene families is shown. The 3′ end of the V_H genes was used as a reference point for sequence alignment.

4. Schematic diagram of a V_H8 gene with its RSS sequence and downstream V8E element. Gray shading indicates the extent of the ATAC‐seq peak present at the V8E element.

5. Identification of different V8E elements by their open chromatin peaks that are located downstream of individual V_H8 genes (indicated by black lines). As the previously identified V_H3609.8pg.160 and V_H3609.14pg.181 genes (Johnston et al, 2006) were not mapped in the mouse mm9 or mm10 genomes, we refer to these PAIR6‐ and PAIR13‐associated genes as V_H8‐x and V_H8‐y, respectively. These genes have the following coordinates: V_H8‐x gene – chr12:116642631‐116643076 (mm9) or chr12:115404420‐115404865 (mm10); V_H8‐y gene – chr12:117046639‐117047071 (mm9) or chr12:115808428‐115808860 (mm10).

Figure EV1. Characterization of the PAIR4, PAIR6, and V8E elements

1. A, B

   Presence of active histone marks, transcription factor binding, open chromatin, and transcription in the PAIR4 (A) and PAIR6 (B) regions at the pro‐B cell stage. Open chromatin was furthermore mapped by ATAC‐seq in hCD2⁻ (Pax5⁻) and hCD2⁺ (Pax5⁺) BLPs that were sorted from the bone marrow of Pax5 ^ihCd2/ihCd2 mice (Fuxa & Busslinger, ; Hill et al, 2020). The RPM scales for displaying the open chromatin peaks in BLPs and pro‐B cells were adjusted to show equal densities of the open chromatin peaks present at a gene‐dense genomic region, including the ubiquitously expressed Tbp locus. The location of PAIR4, PAIR6, and their associated V_H8 genes are shown together with the exons of the PAIR4‐derived lncRNA (RIKEN clone CJ056205; below) and the mm9 genomic coordinates of mouse chromosome 12 (above). As the previously identified V_H3609.8pg.160 (Johnston et al, 2006) was not mapped in the mouse mm9 or mm10 genome, we refer to this PAIR6‐associated gene as V_H8‐x. The coordinates for the V_H8‐x gene are chr12:116642631‐116643076 (mm9) or chr12:115404420–115,404,865 (mm10). Gray overlay indicates the DNA sequences of the PAIR4‐V8.7E module that were inserted with the Floxin method at the deletion point of the Igh ^∆890 allele to generate the Igh ^P4V allele (Fig 3A). RPM, reads per million mapped sequence reads.

2. C, D

   Sequence conservation of the aligned V_H8 (C) and V_H1 (D) gene regions and their 3′ flanking sequences (~400 bp). The line denotes the LOESS‐smoothed position‐wise maximum sequence identity across the multiple sequence alignments. The coding sequences of the V_H genes are indicated together with the recombination signal sequences (RSS).

3. E

   Sequence alignment of the ~200‐bp conserved region downstream of the 3′ end of the V_H8 genes. Dots indicate identical nucleotides relative to the consensus sequence shown below. Nonidentical nucleotides are shown by their respective letters (C, G, T, and A). Gaps in the alignment are indicated by dashes. Numbers refer to the positions downstream of the V_H8 3′ end.

Figure 2. An 890‐kb deletion in the 5′ region of the Igh locus impairs adjacent distal V_H gene recombination

1. A

   B cell development in Igh ^{∆890/∆890} mice. The relative frequencies of the indicated cell types were determined by flow cytometric analysis of bone marrow cells from Igh ^{∆890/∆890} and Igh ^+/+ mice at the age of 3–4 weeks, indicating a 1.3‐fold increase of pro‐B cells and a moderate decrease at subsequent B cell developmental stages in Igh ^{∆890/∆890} mice. Statistical data are shown as a mean value with SEM.

2. B

   Two‐color 3D DNA‐FISH analysis of ex vivo sorted pro‐B cells of the indicated genotypes with the RP23‐340K14 (red) and RP24‐275L15 (green) BAC probes (Fig 1A). Representative images are shown above. Dot plots (below) show the distances measured between the two DNA signals in individual Igh alleles (2,123 for Rag2 ^−/−, 1,514 for Igh ^{∆890/∆890} Rag2 ^−/−, and 502 for Vav‐Cre Pax5 ^fl/fl (Pax5 ^∆/∆) Rag2 ^−/− pro‐B cells) together with the mean distance (red bar) and SEM determined for each genotype. In total, four independent DNA‐FISH experiments were performed.

3. C, D

   V_H gene recombination analysis in ex vivo sorted Igh ^{∆890/∆890} and Igh ^+/+ pro‐B cells, as determined by VDJ‐seq experiments (Chovanec et al, 2018). (C) The percentages of uniquely identified DJ_H and VDJ_H sequences are shown as mean percentages with SEM. Each dot corresponds to an independent VDJ‐seq experiment performed with sorted pro‐B cells from one mouse. (D) VDJ‐seq analysis of V_H gene rearrangements in Igh ^+/+ pro‐B cells (black, above line) and Igh ^{∆890/∆890} pro‐B cells (white bars, below line). The different V_H genes (horizontal axis) are aligned according to their position in the Igh locus (Dataset EV1). The usage of each V_H gene (vertical axis) is shown as a percentage of all VDJ_H and DJ_H recombination events determined for each pro‐B cell type. The relative frequency of each V_H gene is shown as a mean value with SEM and is based on three independent VDJ‐seq experiments for pro‐B cells of each genotype. The enlargement (upper part) highlights the differences in recombination frequency of the V_H genes adjacent to the 890‐kb deletion. The red horizontal line indicates the extent of reduced V_H gene recombination on either side of the deletion point in Igh ^{∆890/∆890} pro‐B cells.

4. E

   Normalized recombination frequency of the first six distal V_H genes (V_H1‐85 to V_H1‐80) in Igh ^{∆890/∆890} and Igh ^+/+ pro‐B cells. The average recombination frequency of the six distal V_H genes was calculated as mean value with SEM, and the value obtained with Igh ^{∆890/∆890} pro‐B cells was set to 1. The average recombination frequency in Igh ^+/+ pro‐B cells was determined based on all VDJ‐seq experiments that were performed in this study with this control pro‐B cell type. The average recombination frequency of each distal V_H gene in Igh ^{∆890/∆890} and Igh ^+/+ pro‐B cells is shown in Fig EV3C.

Data information: Statistical data were analyzed by multiple t‐tests (unpaired and two‐tailed with Holm‐Sidak correction; A, C, E) or one‐way ANOVA (Tukey post hoc test; B); *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Each dot (A, C, and E) corresponds to one mouse. Source data are available online for this figure.

Figure EV2. Generation, characterization, and reconstitution of the Igh ^∆890 allele and presence of potential regulatory elements throughout the V_H gene cluster

1. A

   Generation of the Igh ^∆890 allele. The indicated selection cassettes were used for introducing the upstream lox71 site and the downstream loxP site at the specified positions of the Igh locus by sequential ES cell targeting to generate the Igh ^{Pgk1‐fl‐890‐fl} allele. The Igh ^∆890 allele was subsequently generated by sequential deletion of the indicated sequences by Cre‐ and Flpe‐mediated recombination in vivo in the mouse, as described in Materials and Methods. Actb, human β‐actin promoter; Bsd, blasticidin resistance gene; Neo, neomycin resistance gene; Puro, puromycin resistance gene; Pgk1, phosphoglycerate kinase‐1 promoter.

2. B

   Location of the V_H1 gene family members, which undergo rearrangements in Igh ^+/+ pro‐B cells and exhibit strongly reduced recombination (indicated by the red line; see Fig 2D) upon deletion of the 890‐kb region in the 5′ region of the Igh locus. The distance of the different V_H genes from the deletion point is shown in kilobases (kb).

3. C

   Schematic diagram of the Floxin system used for inserting the different PAIR4 constructs and the Eμ enhancer at the deletion point of the Igh ^∆890 allele in ES cells. See Materials and Methods for detailed description of the reconstitution experiments.

4. D

   Presence of active chromatin across the Igh locus in Rag2 ^−/− pro‐B cells. Open chromatin was mapped by ATAC‐seq, while the presence of H3K27ac and H3K4me2 peaks was identified by ChIP‐seq analysis with anti‐H3K27ac and anti‐H3K4me2 antibodies. The peaks corresponding to PAIR and V8E elements are highlighted by orange or black lines, respectively. The Eμ enhancer and 3′CBE region are indicated in red. The annotation of the C57BL/6 Igh locus with the different V_H gene families (different colors), the D_H (light blue), and C_H (blue) elements is shown together with the extent of the 890‐kb deletion (black line) and the mm9 genomic coordinates of mouse chromosome 12.

Figure 3. Identification of PAIR4‐V8.7E as a potent enhancer module promoting distal V_H gene recombination

1. Schematic diagram of the different PAIR4 constructs used for reconstitution experiments. The PAIR4 element with its transcription factor binding sites and the cDNA insertion of the PAIR‐derived lncRNA (RIKEN clone CJ056205) in exon 2 of PAIR4 is shown together with the V_H8‐7 gene and its downstream V8.7E element. Six copies of a synthetic polyadenylation sequence (pA; Levitt et al, 1989) linked to Gfp cDNA were inserted in exon 1 of PAIR4 to generate the P4StopG insert. FseI‐BspEI or FseI‐BamHI DNA fragments of the different PAIR4 constructs were inserted at the deletion point into the Igh ^∆890 allele with the Floxin system (Singla et al, 2010) as described in detail in Fig EV2C and Materials and Methods. pA, 6 polyadenylation sites.

2. Flow cytometric determination of the ratio of immature IgM^b (B6) to IgM^a (129) B cells in the bone marrow of Igh ^{∆890(B6)/+(129)}, Igh ^{P4V (B6)/+(129)} and control Igh ^+(B6)/+(129) mice, which were generated by crossing Igh ^∆890/+, Igh ^P4V/+ and Igh ^+/+ mice on the C57BL/6 (B6) background with Igh ^+/+ mice of the 129/Sv (129) strain. The rearranged Igh alleles of the C57BL/6 and 129/Sv strains give rise to the expression of IgM^b and IgM^a, respectively. The IgM^b (B6) to IgM^a (129) ratio is shown as a mean value with SEM (right).

3. Summary of the ratios of immature IgM^b (B6) to IgM^a (129) B cells, determined for the indicated six mouse strains analyzed and shown as mean values with SEM. The ratio determined for control Igh ^+(B6)/+(129) B cells was set to 1.

4. Comparison of the VDJ‐seq data obtained with pro‐B cells of the Igh ^+/+ (black), Igh ^{∆890/∆890} (white bars), and Igh ^P4V/P4V (gray) genotypes. The different V_H genes (horizontal axis) are aligned according to their position in the Igh locus (Dataset EV1). The usage of each V_H gene (vertical axis) is shown as a percentage of all VDJ_H and DJ_H recombination events determined for each pro‐B cell type. The relative frequency of each V_H gene is shown as a mean value with SEM and is based on three independent VDJ‐seq experiments for pro‐B cells of each genotype.

5. Recombination frequency of the inserted V_H8‐7 gene, which was determined as a percentage of all VDJ_H recombination events in Igh ^+/+ and Igh ^P4V/P4V pro‐B cells and is shown as a mean value with SEM.

6. Normalized recombination frequency of the first six distal V_H genes (V_H1‐85 to V_H1‐80) determined in pro‐B cells of the indicated genotypes, based on the data shown in (D, G). The average recombination frequency of the six distal V_H genes was calculated as a mean value with SEM, and the value obtained with Igh ^{∆890/∆890} pro‐B cells was set to 1. The average recombination frequency of each distal V_H gene in pro‐B cells of the indicated genotypes is shown in Fig EV3C.

7. Comparison of the VDJ‐seq data obtained with pro‐B cells of the Igh ^+/+ (black), Igh ^P4/P4 (light gray), Igh ^P4inv/P4inv (blue), and Igh ^{P4StopG/P4StopG} (green) genotypes. For further description see (D).

Data information: Statistical data were analyzed by multiple t‐tests (unpaired and two‐tailed with Holm‐Sidak correction; B, E) or one‐way ANOVA (Tukey post hoc test; C, F); *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Each dot (B, C, E, and F) corresponds to one mouse. Source data are available online for this figure.

Figure EV3. Activation of distal V_H gene recombination upon reconstitution of the Igh ^∆890 allele with different PAIR4 constructs

1. V_H gene expression from the Igh ^P4V(B6) and Igh ^∆890(B6) alleles in immature IgM^b (B6) B cells sorted from the bone marrow of Igh ^{P4V(B6)/+(129)} and Igh ^{∆890(B6)/+(129)} mice, respectively (see also Fig 3B). The expression (TPM) value of each V_H gene is shown as shared (light gray) or unique expression of the Igh ^P4V(B6) allele (dark gray) or Igh ^∆890(B6) allele (white bars). The expression data are shown as mean TPM values and are based on three or two independent RNA‐seq experiments for the immature B cells of Igh ^{P4V(B6)/+(129)} and Igh ^{∆890(B6)/+(129)} mice, respectively. The different V_H genes (horizontal axis) are aligned according to their position in the Igh locus (Dataset EV1). The 890‐kb deletion is indicated by a black line. TPM, transcripts per million; B6, C57BL/6 strain; 129, 129/Sv strain.

2. VDJ‐seq analysis of Igh rearrangements in pro‐B cells of the indicated genotypes. The percentages of uniquely identified DJ_H and VDJ_H sequences are shown as mean percentage with SEM. The high variance caused by the many data points obtained with control Igh ^+/+ pro‐B cells is likely responsible for the fact that only the comparison between Igh ^+/+ and Igh ^{∆890/∆890} pro‐B cells reached statistical significance.

3. Normalized recombination frequency of the first six distal V_H genes (V_H1‐85 to V_H1‐80) determined in pro‐B cells of the indicated genotypes, based on the data shown in Fig 3D and G. The recombination frequency of each of the six distal V_H genes was calculated as a mean value with SEM, and the value obtained with Igh ^{∆890/∆890} pro‐B cells was set to 1 (dashed line).

4. The PAIR4‐V8.7E module and PAIR4 elements differentially activate the recombination of V_H genes located upstream or downstream of their insertion. The normalized recombination frequency of the first six upstream V_H genes (V_H1‐85 to V_H1‐80) and the first six downstream V_H genes (V_H1‐47, V_H1‐43, V_H1‐42, V_H1‐39, V_H1‐37, V_H1‐36; Fig EV2B) was determined in pro‐B cells of the indicated genotypes, based on the data shown in Fig 3D and G. The average recombination frequency of the six upstream and six downstream V_H genes was calculated as mean value with SEM, and the value obtained with Igh ^{∆890/∆890} pro‐B cells was set to 1.

5. Active chromatin at the V_H1‐81 and V_H1‐82 genes in Igh ^P4V/P4V, Igh ^{∆890/∆890} and Igh ^+/+ pro‐B cells. Short‐term cultured pro‐B cells of the indicated genotypes were used for native ChIP analysis with anti‐H3K9ac, anti‐H3K4me2 and anti‐H3K4me3 antibodies. Input and precipitated DNA were quantified by qPCR analysis with primers amplifying the sequences of the V_H1‐81, V_H1‐82 and Tbp genes. The amount of precipitated DNA was calculated as mean percentage of input with SEM and was normalized to the value obtained for the positive Tbp control. The relative enrichment is shown by setting the normalized value of the control Igh ^+/+ pro‐B cells to 1.

6. 3C‐qPCR analysis of the long‐range interactions between the PAIR4‐V8.7E module and distal V_H genes in sorted bone marrow pro‐B cells and CD4⁺CD8⁺ double‐positive (DP) thymocytes from Igh ^P4V/P4V and Igh ^{∆890/∆890} mice. The relative crosslinking frequencies between the reference DpnII fragment (viewpoint, located downstream of the insertion site) and the indicated distal V_H genes were determined with independently prepared 3C‐templates and were normalized relative to the crosslinking frequency measured for the control Ercc3 gene (see Materials and Methods). The mean crosslinking frequency with SEM is shown for each V_H gene by setting the mean value obtained with DP T cells for each genotype to 1.

7. Absence of GFP expression in pro‐B cells of Igh ^P4StopG/+ (light green) and Igh ^{P4StopG/P4StopG} (green) mice compared with control Igh ^+/+ (black) mice, as indicated by flow‐cytometric analysis and geometric mean fluorescence intensity measurement. Geometric mean fluorescence intensity measurements are shown as a mean value with SEM.

Data information: Statistical data (B, D, E, F, G) were analyzed by one‐way ANOVA (Tukey post hoc test); *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Each dot (B‐G) corresponds to one mouse. Source data are available online for this figure.

Figure 4. Long‐range activation of distal V_H gene recombination by the V8.8E enhancer in Igh ^{V8‐8/V8‐8} mice

1. Presence of open chromatin and the active histone marks H3K27ac and H3K4me2 at the V8.8E enhancer in pro‐B cells. The location of the V_H8‐8 gene, V8.8E enhancer (dashed line), and PAIR5 element are shown together with the mm9 genomic coordinates of mouse chromosome 12. The V_H8‐8 gene with 500 bp of its upstream and downstream sequences (indicated by violet overlay) was inserted with the Floxin method at the deletion point of the Igh ^∆890 allele to generate the Igh ^V8‐8 allele (Hill et al, 2020).

2. Comparison of the VDJ‐seq data obtained with pro‐B cells from the bone marrow of Igh ^+/+ (black) and Igh ^{V8‐8/V8‐8} (violet) mice. The relative frequency of each V_H gene is shown as a mean value with SEM and is based on five (Igh ^+/+) or three (Igh ^{V8‐8/V8‐8}) independent VDJ‐seq experiments. The different V_H genes (horizontal axis) are aligned according to their position in the Igh locus (Dataset EV1). For further explanation, see legend of Fig 3D.

3. Normalized recombination frequency of the first six distal V_H genes (V_H1‐85 to V_H1‐80) determined in pro‐B cells from Igh ^{V8‐8/V8‐8} (violet), Igh ^+/+ (black), Igh ^{∆890/∆890} (white bar) and Igh ^P4/P4 (gray) mice. The average recombination frequency of the six distal V_H genes was calculated as a mean value with SEM, and the value obtained with Igh ^{∆890/∆890} pro‐B cells was set to 1. Statistical data were analyzed by one‐way ANOVA (Tukey post hoc test); *P < 0.05, ***P < 0.001, ****P < 0.0001. Each dot corresponds to a mouse.

Source data are available online for this figure.

Figure 5. Pro‐B cell‐specific activity of the PAIR4‐V8.7E enhancer module

1. A

   Schematic diagram of the Gfp‐linked PAIR4‐V8.7E module. The Gfp cDNA linked to 6 polyadenylation sites (pA) was inserted in exon 2 of PAIR4 prior to the insertion of the entire module at the deletion point into the Igh ^∆890 allele by the Floxin method.

2. B, C

   Flow‐cytometric analysis (B) and geometric mean fluorescence intensity measurements (MFI, C) of GFP expression are shown for pro‐B, pre‐B, immature B, and recirculating B cells from the bone marrow of Igh ^P4GV/P4GV (dark green, n = 16), Igh ^P4GV/+ (light green, n = 12), and control Igh ^+/+ (black, n = 13) mice. Geometric mean fluorescence intensity measurements (C) are shown as mean values with SEM.

3. D, E

   Induction of GFP expression in a small subset of BLPs, as shown by flow‐cytometric analysis (D) and geometric mean fluorescence intensity measurements (E) of GFP expression in MPPs (LSK), LMPPs, ALPs, and BLPs from the bone marrow of the indicated mice. The dashed line indicates the absence of GFP expression in Igh ^+/+ cells, which were set to 1 for normalization of the MFI data. Geometric mean fluorescence intensity measurements (E) are shown as mean value with SEM based on 3 (Igh ^+/+), 5 (Igh ^P4GV/+) or 6 (Igh ^P4GV/P4GV) independent replicates.

4. F, G

   Absence of GFP expression in DN (CD4⁻CD8⁻) and DP (CD4⁺CD8⁺) T cells, as shown by flow‐cytometric analysis (F) and geometric mean fluorescence intensity measurements (G) of GFP expression in thymocytes from the indicated mice. Geometric mean fluorescence intensity measurements (G) are shown as a mean value with SEM based on 12 (Igh ^+/+), 11 (Igh ^P4GV/+), or 12 (Igh ^P4GV/P4GV) independent replicates.

Data information: The flow‐cytometric definition of the lymphoid progenitors, B and T cell types is described in Materials and Methods. Source data are available online for this figure.

Figure 6. The CTCF binding site of PAIR4 restricts the activity of the PAIR4‐V8.7E module to the pro‐B cell stage

1. A, B

   Flow‐cytometric analysis (A) and geometric mean fluorescence intensity measurements (MFI, B) of GFP expression are shown for pro‐B, pre‐B, immature B, and recirculating B cells from the bone marrow of Igh ^P4GV/+ (green, n = 6), Igh ^{P4∆Pax5GV/+} (yellow, n = 5), Igh ^{P4∆CtcfGV/+} (brown, n = 4), and control Igh ^+/+ (black, n = 4) mice. Geometric mean fluorescence intensity measurements (B) are shown as a mean value with SEM.

2. C, D

   Activation of the PAIR4‐V8.7E module in thymocytes upon loss of CTCF binding at PAIR4, as shown by flow‐cytometric analysis (C) and geometric mean fluorescence intensity measurements (MFI, D) of GFP expression in DN, DP, CD4⁺, and CD8⁺ thymocytes of Igh ^{P4∆CtcfGV/+} (light brown, n = 5) and Igh ^{P4∆CtcfGV/P4∆CtcfGV} (dark brown, n = 6) mice compared with control Igh ^+/+ (black, n = 7) mice. Geometric mean fluorescence intensity measurements (D) are shown as a mean value with SEM.

3. E

   Comparison of the VDJ‐seq data obtained with pro‐B cells of the Igh ^+/+ (black), Igh ^P4GV/P4GV (green), Igh ^{P4∆Pax5GV/P4∆Pax5GV} (yellow), and Igh ^{P4∆CtcfGV/P4∆CtcfGV} (brown) genotypes. The relative frequency of each V_H gene is shown as mean value with SEM and is based on four (Igh ^+/+, Igh ^P4GV/P4GV) or three (Igh ^{P4∆Pax5GV/P4∆Pax5GV}, Igh ^{P4∆CtcfGV/P4∆CtcfGV}) independent VDJ‐seq experiments. The different V_H genes (horizontal axis) are aligned according to their position in the Igh locus (Dataset EV1). For further explanation, see legend of Fig 3D.

4. F

   Recombination frequency of the inserted V_H8‐7 gene, which was determined as percentage of all VDJ_H recombination events in pro‐B cells of the indicated genotypes and is shown as a mean value with SEM.

5. G

   Normalized recombination frequency of the first six distal V_H genes (V_H1‐85 to V_H1‐80) determined in pro‐B cells of the indicated genotypes, based on the data shown in (E). The average recombination frequency of the six distal V_H genes was calculated as mean value with SEM, and the value obtained with Igh ^{∆890/∆890} pro‐B cells was set to 1.

Data information: Statistical data were analyzed by one‐way ANOVA (Tukey post hoc test; F, G); **P < 0.01, ***P < 0.001, ****P < 0.0001. Each dot (B, D, F, and G) corresponds to a mouse. Source data are available online for this figure.

Figure EV4. Mutation of the Pax5‐ and CTCF‐binding sites in the PAIR4 element

1. A

   Mutation of the Pax5‐binding sequence of PAIR4. The nucleotides of PAIR4, which match the Pax5 consensus recognition sequence (Kaiser et al, 2022), are underlined and were mutated to the nucleotides indicated in red (∆Pax5).

2. B

   Mutation of the CTCF‐binding sequence of PAIR4. The nucleotides of PAIR4, corresponding to the consensus CTCF‐binding motif (Hill et al, 2020), are underlined and were mutated to the nucleotides indicated in red (∆CTCF).

3. C, D

   Schematic diagram of the Gfp‐linked PAIR4‐V8.7E module with the Pax5‐ or CTCF‐binding site mutation, which was used for the generation of the Igh ^P4∆Pax5GV or Igh ^P4∆CtcfGV allele, respectively.

4. E, G

   Pax5 and CTCF binding to the wild‐type and mutant PAIR4 sequences. Short‐term cultured pro‐B cells from the bone marrow of Igh ^P4GV/P4GV (green), Igh ^{P4∆Pax5GV/P4∆Pax5GV} (yellow), and Igh ^{P4∆CtcfGV/P4∆CtcfGV} (brown) mice were used for ChIP analysis with an anti‐Pax5 paired domain antibody or an anti‐CTCF antibody. Input and precipitated DNA were quantified by qPCR analysis with primers amplifying the Pax5‐binding site present in PAIR4 or the Nedd9 gene (E), the CTCF‐binding site present in PAIR4 or the Bud13 gene (G) or by amplifying a gene‐poor region on chromosome 1 as a negative control (Dataset EV2). The amount of precipitated DNA was calculated as an average percentage of input with SEM and is shown relative to the value obtained for the negative control.

5. F

   No binding of Pax5 to the mutant Pax5 recognition sequence (∆Pax5) of PAIR4. A fluorescently‐labeled double‐stranded oligonucleotide containing the wild‐type (wt) Pax5 recognition sequence of PAIR4 (shown below) was used as a probe for electrophoretic mobility shift assay (EMSA) with a nuclear extract prepared from B cells of the human Ramos cell line (see Materials and Methods). The Pax5‐DNA complex (marked as Pax5 to the left) was not formed upon addition of an anti‐Pax5 antibody, directed against the DNA‐binding paired domain (Prd), or in the presence of a 10‐, 30‐, and 100‐fold molar excess of a nonlabeled competitor oligonucleotide containing the wild‐type Pax5‐binding site in contrast to the competitor oligonucleotide containing the mutant (mut) Pax5 recognition sequence (shown below).

6. H

   Absence of GFP expression in BLPs from the bone marrow of Igh ^P4GV/+ (green), Igh ^{P4∆Pax5GV/+} (yellow), Igh ^{P4∆CtcfGV/+} (brown), and control Igh ^+/+ (black) mice, as shown by flow‐cytometric analysis.

7. I

   Transient transfection reporter assay. The PAIR4‐luciferase reporter genes (schematically shown to the left) were transfected together with the control vector pRL‐CMV into cells of the pro‐B cell line 38B9 (Alt et al, 1984) or pre‐B cells line PD31 (Lewis et al, 1982). After 24 h, luciferase activities were measured, normalized and displayed relative to the activity of the parental vector pXPG, containing a promoter‐less firefly luciferase gene, which was set to 1. The relative luciferase (luc) activities of six independent transfection experiments are shown as mean values with SEM. Ex1 and Ex2 refer to exons 1 and 2 of the PAIR antisense transcript.

Data information: Statistical data (E, G, I) were analyzed by one‐way ANOVA (Tukey post hoc test); ns P > 0.05, *P < 0.05; **P < 0.01, ****P < 0.0001. Source data are available online for this figure.

Figure 7. The Eμ enhancer upon insertion in the Igh 5′ region promotes distal V_H gene recombination and exhibits pro‐B cell‐specific activity

1. A

   Schematic diagram of the Igh ^CμCre allele, which contains the insertion of a Cre gene linked via a P2A sequence to the last codon of the second Cμ transmembrane exon (M2) in the Igh locus. The generation of the Igh ^CμCre allele is described in detail in Fig EV5A and Materials and Methods.

2. B, C

   Flow‐cytometric analysis (B) and geometric mean fluorescence intensity measurements (MFI, C) of YFP expression are shown for MPPs (LSK), LMPPs, ALPs, and BLPs from the bone marrow of Rosa26 ^LSL‐YPF/+ Igh ^CμCre/+ (violet) and control Rosa26 ^LSL‐YPF/+ Igh ^+/+ (black) mice. Geometric mean fluorescence intensity measurements (B) are shown as a mean value with SEM and have been normalized by setting the mean value obtained with Igh ^+/+ cells to 1.

3. D

   Schematic diagram of the Gfp‐linked Iμ‐Eμ sequences used to generate the Igh ^EμG allele. The Gfp cDNA together with six polyadenylation sites (pA) was linked to the Iμ exon downstream of the Eμ enhancer prior to insertion of these sequence at the deletion point into the Igh ^∆890 allele by the Floxin method.

4. E–G

   Flow‐cytometric analysis (E) and geometric mean fluorescence intensity measurements (MFI, F, and G) of GFP expression are shown for MPPs (LSK), LMPPs, ALPs, BLPs, pro‐B, pre‐B, immature B, and recirculating B cells from the bone marrow of Igh ^EμG/+ (light blue), Igh ^EμG/EμG (dark blue), and control Igh ^+/+ (black) mice. Geometric mean fluorescence intensity measurements (B) are shown as mean a value with SEM. For lymphoid progenitors, MFI values have been normalized by setting the mean value obtained with Igh ^+/+ cells to 1. The flow‐cytometric definition of the lymphoid progenitors, B and T cell types is described in Materials and Methods.

5. H

   Absence of GFP expression in DN and DP T cells from the thymus of Igh ^EμG/+ (light blue) and Igh ^EμG/EμG (dark blue) mice compared with control Igh ^+/+ (black) mice.

6. I

   Comparison of the VDJ‐seq data obtained with pro‐B cells of the Igh ^+/+ (black) and Igh ^EμG/EμG (dark blue) genotypes. The relative frequency of each V_H gene is shown as mean value with SEM and is based on six (Igh ^+/+) or three (Igh ^EμG/EμG) independent VDJ‐seq experiments. For further description, see legend of Fig 3D.

7. J

   Normalized recombination frequency of the first six distal V_H genes (V_H1‐85 to V_H1‐80) determined in pro‐B cells of the indicated genotypes, based on the data shown in (I). The average recombination frequency of the six distal V_H genes was calculated as a mean value with SEM, and the value obtained with Igh ^{∆890/∆890} pro‐B cells was set to 1.

Data information: Statistical data were analyzed by one‐way ANOVA (Tukey post hoc test; J); *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Each dot (C, F, G, and J) corresponds to one mouse. Source data are available online for this figure.

Figure EV5. Generation and characterization of the Igh ^CμCre allele and explanation of recombination differences in the context of loop extrusion

1. Generation of the Igh ^CμCre allele by ES cell targeting. The Igh ^CμCre‐neo allele was generated by homologous recombination in the ES cell line A9 by using the following targeting vector. The targeting vector consisted of a 4.1‐kb long 5′ homology region (containing Cμ1 to CμS), a frt‐flanked 2.1‐kb DNA fragment containing the mouse phosphoglycerate kinase (Pgk1) promoter linked to the neomycin (Neo) resistance gene and a SV40 polyadenylation signal, a 1.6‐kb DNA fragment containing the CμM1/2 sequence linked in frame via the P2A peptide to an iCre gene followed by a 1.8‐kb long 3′ homology region. The frt sites are indicated by green arrowheads. The ApaI fragments of the Igh ⁺ and Igh ^CμCre‐neo alleles, which were used for allele identification by Southern blot analysis with the indicated probe, are shown together with their length (in kilobases, kb). The Igh ^CμCre allele was generated by deletion of the frt‐flanked Pgk1‐Neo expression cassette, which was used for selection of the targeted ES cell clones, in Igh ^CμCre‐neo mice expressing the Flpe transgene.

2. Southern blot analysis of wild‐type (WT) ES cells and correctly targeted ES cell clones by hybridization of ApaI‐digested genomic DNA with the probe indicated in (A). The ES cell clone 11F was injected into blastocysts to generate the Igh ^{CμCre‐neo/+} mouse strain.

3. RT‐qPCR analysis of Gfp mRNA expression in sorted pro‐B cells from the bone marrow of the indicated mouse strains. The transcripts of the Gfp gene were normalized against the control Tbp mRNA, and the mean value obtained with Igh ^+/+ pro‐B cells was set to 1. Gfp mRNA expression is shown as a mean value with SEM based on nine (Igh ^+/+) or four (Igh ^P4GV/P4GV, Igh ^EμG/EμG, Igh ^{P4StopG/P4StopG}) independent samples. Statistical data were analyzed by one‐way ANOVA (Tukey post hoc test); ****P < 0.0001.

4. Role of loop extrusion in mediating increased recombination of V_H genes located upstream of the insertion of the PAIR4‐V8.7E module (Figs 3G and EV3D). The loop extrusion process has been shown to generate a largely contiguous interaction zone where all the different sequences within the V_H gene region appear to interact with each other (Hill et al, 2023). The generation of this interaction zone requires that all CTCF‐binding sites are present in forward orientation in the V_H gene cluster (Hill et al, 2020). Loop extrusion likely initiates at random positions in the V_H gene cluster and initially proceeds in a symmetrical manner until the cohesin ring interacts with a CTCF protein bound to the next upstream forward‐oriented CTCF‐binding site, which leads to stabilized binding of cohesin at this site (Li et al, 2020). Thereafter, asymmetrical loop extrusion reels the DNA of the downstream Igh regions into the loop, until it is halted by a CTCF protein bound to a reverse‐oriented CTCF‐binding site at the IGCR1 or 3′CBE elements (Hill et al, 2023). This loop extrusion mechanism predicts that only the V_H1 genes located upstream of the PAIR4 insertion can efficiently interact through loop extrusion with the inserted PAIR4 element and thus undergo increased V_H1 gene recombination. In contrast, the downstream V_H1 genes are unable to interact through this loop extrusion mechanism with the inserted PAIR4 element, which results in reduced V_H1 gene recombination.

5. Explanation for the reduced recombination efficiency of upstream V_H1 genes in Igh ^P4inv/P4inv pro‐B cells compared with Igh ^P4/P4 pro‐B cells (Fig 3G and F). The inversion of PAIR4 also inverts the orientation of the CTCF‐binding site of PAIR4 in Igh ^P4inv/P4inv pro‐B cells relative to Igh ^P4/P4 pro‐B cells. The reverse‐oriented CTCF‐binding site of PAIR4 is now in convergent orientation relative to the forward CTCF‐binding sites in the upstream region and can thus form new stabilized loops (Rao et al, 2014) that interfere with prolonged loop extrusion beyond the inserted PAIR4 element and can therefore prevent interactions of the upstream V_H genes with the RAG⁺ recombination center in the Igh 3′ region, as previously shown for the inversion of the 890‐kb V_H gene region (Hill et al, 2020). The forward and reverse CTCF consensus motifs (D, E) are shown.

Source data are available online for this figure.