Developmental regulation of fetal to adult globin gene switching in human fetal erythroid x mouse erythroleukemia cell hybrids

LCR 5' hypersensitive site specificity for globin gene activation within the active chromatin hub

The DNaseI hypersensitive sites (HSs) of the human β-globin locus control region (LCR) may function as part of an LCR holocomplex within a larger active chromatin hub (ACH). Differential activation of the globin genes during development may be controlled in part by preferential interaction of each gene with specific individual HSs during globin gene switching, a change in conformation of the LCR holocomplex, or both. To distinguish between these possibilities, human β-globin locus yeast artificial chromosome (β-YAC) lines were produced in which the ε-globin gene was replaced with a second marked β-globin gene (β^m), coupled to an intact LCR, a 5′HS3 complete deletion (5′ΔHS3) or a 5′HS3 core deletion (5′ΔHS3c). The 5′ΔHS3c mice expressed β^m-globin throughout development; γ-globin was co-expressed in the embryonic yolk sac, but not in the fetal liver; and wild-type β-globin was co-expressed in adult mice. Although the 5′HS3 core was not required for β^m-globin expression, previous work showed that the 5′HS3 core is necessary for ε-globin expression during embryonic erythropoiesis. A similar phenotype was observed in 5′HS complete deletion mice, except β^m-globin expression was higher during primitive erythropoiesis and γ-globin expression continued into fetal definitive erythropoiesis. These data support a site specificity model of LCR HS-globin gene interaction.

The sea urchin sns5 insulator protects retroviral vectors from chromosomal position effects by maintaining active chromatin structure

Silencing and position-effect (PE) variegation (PEV), which is due to integration of viral vectors in heterochromatin regions, are considered significant obstacles to obtaining a consistent level of transgene expression in gene therapy. The inclusion of chromatin insulators into vectors has been proposed to counteract this position-dependent variegation of transgene expression. Here, we show that the sea urchin chromatin insulator, sns5, protects a recombinant gamma-retroviral vector from the negative influence of chromatin in erythroid milieu. This element increases the probability of vector expression at different chromosomal integration sites, which reduces both silencing and PEV. By chromatin immunoprecipitation (ChIP) analysis, we demonstrated the specific binding of GATA1 and OCT1 transcription factors and the enrichment of hyperacetylated nucleosomes to sns5 sequences. The results suggest that this new insulator is able to maintain a euchromatin state inside the provirus locus with mechanisms that are common to other characterized insulators. On the basis of its ability to function as barrier element in erythroid milieu and to bind the erythroid specific factor GATA1, the inclusion of sns5 insulator in viral vectors may be of practical benefit in gene transfer applications and, in particular, for gene therapy of erythroid disorders.

Extended core sequences from the cHS4 insulator are necessary for protecting retroviral vectors from silencing position effects

The prototypic chromatin insulator cHS4 has proven effective at reducing repressive chromosomal position effects on retroviral vector expression. We report here studies designed to identify the minimal chicken hypersensitive site-4 (cHS4) sequences necessary for this activity. Using a gammaretroviral reporter vector and expression analysis in cell lines and primary mouse hematopoietic progenitor colonies, we found that a 250-bp core fragment reported to contain most of the cHS4 insulating activity failed to prevent silencing when used alone, although some barrier activity was observed when this fragment was combined with a 790-bp, but not 596-bp, spacer. Similar studies showed that four copies of a 90-bp fragment containing the cHS4 enhancer-blocking activity actually repressed vector green fluorescent protein (GFP) expression. In contrast, a 400-bp fragment containing the 250-bp core plus 3' flanking sequences protected vector expression to the same degree as the full-length 1.2-kb fragment. The 400-bp fragment activity was confirmed in a lentiviral vector expressing human beta-globin in murine erythroid leukemia (MEL) cells. Taken together, these studies indicate that the insulating activity of the 250-bp cHS4 core can be influenced by distance, and identify an extended core element that confers full barrier activity in the setting of two different classes of retroviral vectors.

The transcription factor KLF11 can induce gamma-globin gene expression in the setting of in vivo adult erythropoiesis

Previous studies in a fetal erythroid cell line demonstrated that the transcription factor, Krüppel-like factor 11 (KLF11), could specifically induce transcription from a gamma-globin gene promoter, and that this induction was mediated through a specific canonical CACCC cis-DNA binding motif. We report here that ectopic expression of KLF11 can also induce fetal gamma-globin gene expression in the setting of adult erythropoiesis both in vitro and in vivo. Studies in an adult-stage murine erythroleukemia (MEL) cell line demonstrated that retrovirus vector-mediated transduction of KLF11 could increase both the amount of expression from a basally active, but not from a overtly silenced, recombinant gamma-globin transgene, as well as the frequency of cells expressing this transgene. A similar pattern of gamma-globin gene induction was also observed both in vitro and in vivo following KLF11 transduction of bone marrow from mice containing a basally active gamma-globin transgene. These studies provide the first evidence that ectopic expression of a transcription factor can induce gamma-globin gene expression in vivo during adult erythropoiesis.

Use of the hereditary persistence of fetal hemoglobin 2 enhancer to increase the expression of oncoretrovirus vectors for human gamma-globin

The development of oncoretrovirus vectors for human gamma-globin has been hampered by problems of low expression and gene silencing. In order to address these problems, we investigated an enhancer element identified from individuals with deletional hereditary persistence of fetal hemoglobin 2 (HPFH2), a genetic condition characterized by elevated levels of gamma-globin in adults. Plasmid transfection studies in erythroid MEL (murine erythroleukemia) cells demonstrated the HPFH2 element could function synergistically with the beta-globin locus control region to enhance the expression of an Agamma-globin gene with a truncated -382 bp promoter. A series of oncoretrovirus vectors were subsequently generated that contain an expression cassette for Agamma-globin linked to various combinations of the HPFH2 enhancer, the alpha-globin HS40 enhancer, and several versions of the promoter from Agamma-globin or beta-globin. Expression analysis in transduced MEL cell clones revealed very high levels of promoter-autonomous silencing that was at least partially abrogated by the HPFH2 enhancer. The vector containing a combination of a -201 bp Agamma-globin gene promoter with the Greek HPFH -117 point mutation and both the HPFH2 and HS40 enhancers exhibited no signs of vector silencing and was expressed at 248+/-99% per copy of mouse alpha-globin (62% of total alpha-globin). This represents a significant improvement over previously reported oncoretrovirus vectors for Agamma-globin, and demonstrates the capacity of the HPFH2 enhancer to abrogate sequence-autonomous silencing of the Agamma-globin promoter in the context of a gene transfer vector.

Genome architecture of the human beta-globin locus affects developmental regulation of gene expression

To test the role of gene order in globin gene expression, mutant human beta-globin locus yeast artificial chromosome constructs were used, each having one additional globin gene encoding a "marked" transcript (epsilon(m), gamma(m), or beta(m)) integrated at different locations within the locus. When a beta(m)-globin gene was placed between the locus control region (LCR) and the epsilon-globin gene, beta(m)-globin expression dominated primitive and definitive erythropoiesis; only beta(m)-globin mRNA was detected during the fetal and adult definitive stages of erythropoiesis. When an (A)gamma(m)-globin gene was placed at the same location, (A)gamma(m)-globin was expressed during embryonic erythropoiesis and the fetal liver stage of definitive erythropoiesis but was silenced during the adult stage. The downstream wild-type gamma-globin genes were not expressed. When an epsilon(m)-globin gene was placed between the delta- and beta-globin genes, it remained silent during embryonic erythropoiesis; only the LCR-proximal wild-type epsilon-globin gene was expressed. Placement of a beta(m)-globin gene upstream of the (G)gamma-globin gene resulted in expression of beta(m)-globin in embryonic cells and in a significant decrease in expression of the downstream wild-type beta-globin gene. These results indicate that distance from the LCR, an inherent property of spatial gene order, is a major determinant of temporal gene expression during development.

Selection with a regulated cell growth switch increases the likelihood of expression for a linked gamma-globin gene

Several lines of evidence indicate that in vivo drug selection can be used to overcome the low rates of gene transfer and engraftment encountered in many hematopoietic stem cell gene therapy settings. However, whether selection imposed on one transcription cassette effects the likelihood of expression from a second, independent transcription cassette within the same vector has been less well studied. In order to address this issue, we engineered an oncoretrovirus vector to express two separate transcription units: (i) a bicistronic cassette encoding both GFP and a pharmacologically regulated cell growth switch based on the thrombopoietin receptor Mpl; and (ii) a highly position-dependent second cassette encoding human gamma-globin. Studies in cell cultures and in mice transplanted with transduced marrow indicated that selective expansion increased by more than 9-fold the fraction of erythroid cells expressing the linked but separate expression cassette for gamma-globin. This increase was far greater then that observed for the bicistronic GFP gene, and cannot be explained by a simple increase in the fraction of cells containing provirus. These results suggest that selective expansion favors erythroid stem/progenitor cells with provirus integrated at chromosomal sites which are relatively resistant to silencing position effects.

Transgenic Cre expression mice for generation of erythroid-specific gene alterations

Transgenic mice that express Cre recombinase in erythroid cell lineages were developed so that genes affecting erythropoiesis/hematopoiesis may be altered without necessarily affecting fetus viability. A micro-LCR cassette-beta-globin promoter-Cre recombinase gene (microLCR-betapr-Cre) construct was synthesized and used to generate transgenic mice. Concurrently, we produced mice containing a microLCR-loxP-flanked beta sickle gene (microLCR-loxP-beta(S)-loxP) construct. microLCR-betapr-Cre mice with intact transgenes in variable copy number were identified. Cre expression was assessed by RNAse protection and RT-PCR. Cre function was ascertained by breeding to microLCR-loxP-beta(S)-loxP mice. We demonstrate that beta(S) expression was not detected in the blood of bigenics, but the gene was present in nonerythroid cells. Thus, excision of the loxP-flanked beta(S) gene was restricted to erythroid cell lineages.

The histone deacetylase inhibitor, trichostatin A, reactivates the developmentally silenced gamma globin expression in somatic cell hybrids and induces gamma gene expression in adult BFUe cultures

Somatic cell hybrids that have undergone globin gene switching and developmental silencing of gamma globin expression were treated with the histone deacetylase inhibitor trichostatin A (TSA). Culture of the post-switch hybrids in the presence of TSA reactivated gamma globin expression and concommitantly downregulated beta globin expression, as determined by both mRNA quantitation and immunofluorescent quantitation of gamma globin expressing cells. In contrast, similar treatment of pre-switch hybrids, which were expressing predominantly gamma globin and only small levels of beta globin, had no effect on the relative gamma or beta globin gene expression. In addition, trichostatin A induced gamma gene expression in adult BFUe cultures in a maturation-independent fashion. The results provide direct evidence that inhibition of HDAC activity can alter expression from the human beta globin locus in the adult stage of development.

Development of virus vectors for gene therapy of beta chain hemoglobinopathies: flanking with a chromatin insulator reduces gamma-globin gene silencing in vivo

We have previously described the development of oncoretrovirus vectors for human gamma-globin using a truncated beta-globin promoter, modified gamma-globin cassette, and alpha-globin enhancer. However, one of these vectors is genetically unstable, and both vectors exhibit variable expression patterns in cultured cells, common characteristics of oncoretrovirus vectors for globin genes. To address these problems, we identified and removed the vector sequences responsible for genetic instability and flanked the resultant vector with the chicken beta-globin HS4 chromatin insulator to protect expression from chromosomal position effects. After determining that flanking with the cHS4 element allowed higher, more uniform levels of gamma-globin expression in MEL cell lines, we tested these vectors using a mouse bone marrow transduction and transplantation model. When present, the gamma-globin cassettes from the uninsulated vectors were expressed in only 2% to 5% of red blood cells (RBCs) long term, indicating they are highly sensitive to epigenetic silencing. In contrast, when present the gamma-globin cassette from the insulated vector was expressed in 49% +/- 20% of RBCs long term. RNase protection analysis indicated that the insulated gamma-globin cassette was expressed at 23% +/- 16% per copy of mouse alpha-globin in transduced RBCs. These results demonstrate that flanking a globin vector with the cHS4 insulator increases the likelihood of expression nearly 10-fold, which in turn allows for gamma-globin expression approaching the therapeutic range for sickle cell anemia and beta thalassemia.

Hemopoietic lineage commitment decisions: in vivo evidence from a transgenic mouse model harboring μLCR-βpro-LacZ as a transgene

A substantial body of published data suggests activation of lineage-specific genes in multipotential hemopoietic cells before their unilineage commitment. Because the behavior and plasticity of cells isolated in vitro away from microenvironmental constraints exercised in vivo may be altered, one wonders whether similar findings can be observed in a physiologic setting in vivo. We used a transgenic mouse model harboring human micro LCR together with β promoter sequences as a transgene to examine activation of lineage-specific programs in vivo. By using LacZ as a reporter, we had the ability to detect, quantitate, and select live cells with different levels of LacZ activation. We found strong expression of LacZ by X-gal staining in 2 lineages—erythroid and megakaryocytic. Activation in the latter was a novel finding not previously observed when similar transgenes were used. We also found activation of μLCR-βpro at low levels in progenitor cells of granulocytic-macrophagic, erythroid, or megakaryocytic lineage detected by in vitro assays, suggesting activation before commitment to a specific lineage pathway. In particular, the expression of LacZ was graded among progenitors, so that in a proportion of them activation occurred only after commitment to erythroid or megakaryocytic lineage. In addition, we found quantitative reduction in LacZ expression between fetal liver and bone marrow-derived cells, the basis of which is unclear. Collectively our data provide in vivo evidence supporting the view that lineage-specific genes are expressed in a graded fashion in pluripotential cells before their irreversible unilineage commitment.

Analysis of gamma-globin expression cassettes in retrovirus vectors

With the goal of optimizing retrovirus vectors for human gamma-globin, we studied the effect of several globin gene expression elements on vector titer, stability, and expression. We found that all combinations tested were genetically stable, but that vectors with therapeutic titers (0.5 to 2 x 10(6) colony-forming units/ml) could be achieved only by either partially or fully deleting the second intron of the Agamma-globin gene. Efficient transfer and high-level expression was achieved only when an optimized beta-globin promoter was linked to an Agamma-globin cassette containing an intact intron 1 and a 714-bp internal deletion of intron 2. When flanked by two copies of the HS-40 enhancer core from the alpha-globin locus, this cassette expressed gamma-globin mRNA at 46 +/- 19% per copy of mouse alpha-globin in the murine erythroleukemia cell line MEL585. Complete deletion of the first or second intron diminished expression to < or = 2.0%, and deletion of the HS-40 enhancer diminished expression to 7 +/- 8%. High-level, uniform expression of gamma-globin protein was confirmed in MEL585 clones (n = 12) transduced with the optimized vector. Efficient but variable expression of the optimized vector was also observed in erythroid progenitor colonies (n = 6) grown from transduced mouse bone marrow. Taken together, these studies demonstrate the role of intronic, promoter, and enhancer sequences on retrovirus vectors for human gamma-globin, and the development of an optimized vector capable of efficient expression in a murine erythroid cell line and primary cultures.

Histone- and protamine-DNA association: conservation of different patterns within the beta-globin domain in human sperm

Most DNA in human sperm is bound to highly basic proteins called protamines, but a small proportion is complexed with histones similar to those found in active chromatin. This raises the intriguing possibility that histones in sperm are marking sets of genes that will be preferentially activated during early development. We have examined the chromatin structure of members of the beta-globin gene family, which are expressed at different times in development, and the protamine 2 gene, which is expressed in spermatids prior to the widespread displacement of histones by transition proteins. The genes coding for epsilon and gamma globin, which are active in the embryonic yolk sac, contain regions which are histone associated in the sperm. No histone-associated regions are present at the sites tested within the beta- and delta-globin genes which are silent in the embryonic yolk sac. The trends of histone or protamine association are consistent for samples from the same person, and no significant between-subject variations in these trends are found for 13 of the 15 fragments analyzed in the two donors. The results suggest that sperm chromatin structures are generally similar in different men but that the length of the histone-associated regions can vary. The association of sperm DNA with histones or protamines sometimes changes within as little as 400 bp of DNA, suggesting that there is fine control over the retention of histones.

Expression of a cosmid containing the LCR, (A)gamma, delta, and beta globin genes in mouse erythroleukemia cells

A novel method involving the packaging of phage particles was used to introduce cosmids containing components of the human beta-globin locus control region (LCR) and the human (A)gamma, delta, and beta globin genes into mouse erythroleukemia (MEL) cells. After stable transfection, MEL clones were selected and analyzed for expression of human genes. Both (A)gamma and beta globin mRNA were expressed in these clones, indicating that MEL cells do not suppress transcription of the human gamma globin gene. The variability of human (A)gamma vs. beta globin expression from clone to clone prevents a clear delineation of differences in the expression of these two genes with a cosmid in which a region potentially involved in gamma-to-beta switching has been deleted. The results suggest that MEL cells are capable of supporting human gamma expression, despite their predominantly adult phenotype.

Partial repression of human gamma-globin genes by LCR element HS3 when linked to beta-globin genes and LCR element HS2 in MEL cells

Clues for overcoming fetal (gamma-) globin gene repression in adult human erythroid cells may come from understanding why repression of isolated gamma-globin genes has not previously been achieved in the adult erythroid environment of mouse erythroleukemia cells (MEL). Repression of human gamma-globin genes has been demonstrated in MEL cells when transferred as part of the entire beta-globin gene cluster packaged in chromatin. Major differences in these approaches are prior packaging into chromatin and the presence of additional sequences, notably from the locus control region (LCR). In this report we focus on the contribution to gamma-globin gene repression that multiple elements of the LCR may have. We first show preferential activation of beta-globin genes over gamma-globin genes in MEL cells when linked to each other and to LCR sequences containing the core elements of DNase I hypersensitive sites 4, 3, and 2. Removal of the HS4 element had no effect, however, removal of the 225 bp HS3 core element resulted in a five-fold increase in gamma-globin gene expression. The enhancer 3' to the A gamma-globin gene also had no apparent effect on gamma-globin gene expression. These results provide first evidence of gamma-globin gene repression involving the core region of HS3 in the presence of the core region of HS2 and a beta-globin gene. A mechanism for repression involving sequestration of the gamma-promoter away from the strong enhancer activity of HS2 is proposed.

Established epigenetic modifications determine the expression of developmentally regulated globin genes in somatic cell hybrids

Somatic cell hybrids generated from transgenic mouse cells have been used to examine the developmental regulation of human gamma-to-beta-globin gene switching. In hybrids between mouse erythroleukemia (MEL) cells and transgenic erythroblasts taken at various stages of development, there was regulated expression of the human fetal gamma and adult beta genes, reproducing the in vivo pattern prior to fusion. Hybrids formed from embryonic blood cells produced predominantly gamma mRNA, whereas beta gene expression was observed in adult hybrids and a complete range of intermediate patterns was found in fetal liver hybrids. The adult environment of the MEL cells, therefore, did not appear to influence selective transcription from this gene complex. Irradiation of the embryonic erythroid cells prior to fusion resulted in hybrids containing only small fragments of donor chromosomes, but the pattern of gene expression did not differ from that of unirradiated hybrids. This finding suggests that continued expression of trans-acting factors from the donor erythroblasts is not necessary for continued expression of the human gamma gene in MEL cells. These results contrast with the lack of developmental regulation of the cluster after transfection of naked DNA into MEL cells and suggest that epigenetic processes established during normal development result in the gene cluster adopting a developmental stage-specific, stable conformation which is maintained through multiple rounds of replication and transcription in the MEL cell hybrids. On prolonged culture, hybrids that initially expressed only the gamma transgene switched to beta gene expression. The time period of switching, from approximately 10 to > 40 weeks, was similar to that seen previously in human fetal erythroblast x MEL cell hybrids but in this case bore no relationship to the time of in vivo switching. It seems unlikely, therefore, that switching in these hybrids is regulated by a developmental clock.

Use of yeast artificial chromosomes (YACs) for studying control of gene expression: correct regulation of the genes of a human beta-globin locus YAC following transfer to mouse erythroleukemia cell lines

We demonstrate that transfer of a yeast artificial chromosome (YAC) containing 230 kb of the human beta-globin locus into mouse erythroleukemia cells by fusion results in correct developmental regulation of the human beta-like globin genes. Additionally, we show that early after hybrid formation, human embryonic epsilon- and fetal gamma-globin genes are coexpressed with the adult beta gene but that after 10-20 weeks in culture, globin gene expression switches to predominantly adult. Thus, in contrast to shorter gene constructs, the globin genes of the beta-globin locus YAC are regulated like the chromosomal globin genes. These results indicate that transfer of YACs into established cell lines can be used for the analysis of the developmental control of multigenic and developmentally regulated human loci.

Serum factors can modulate the developmental clock of gamma- to beta-globin gene switching in somatic cell hybrids

The fusion of human fetal erythroid (HFE) cells with mouse erythroleukemia (MEL) cells produces stable synkaryons (HFE x MEL) which can be monitored for extended periods of time in culture. Initially these hybrids express a human fetal globin program (gamma >> beta), but after weeks or months in culture, they switch to an adult pattern of globin expression (beta >> gamma). The rate at which hybrids switch to the adult phenotype is roughly dependent on the gestational age of the fetal erythroid cells used in the fusion, suggesting that the rate of switching in vitro may be determined by a developmental clock type of mechanism, possibly involving the cumulative number of divisions experienced by the human fetal cells. To investigate whether the number or rate of cell divisions postfusion can influence the rate of switching, we monitored the rate of switching in hybrids from independent fusions under growth-promoting (serum-replete) and growth-suppressing (serum-deprived) conditions. We found that hybrids grown under serum-deprived or serumless conditions switched more rapidly to adult globin expression than did their counterparts in serum-replete conditions. Neither the number of cumulative cell divisions nor time in culture per se predicted the rate of switching in vitro. Our data suggest that factors present in serum either retard switching of hybrids by their presence or promote switching by their absence, indicating that globin switching in vitro can be modulated by the environment; however, once switching in HFE x MEL hybrids is complete, serum factors cannot reverse this process.

Role of gene order in developmental control of human gamma- and beta-globin gene expression

To determine the effect of gene order on globin gene developmental regulation, we produced transgenic mice containing two tandemly arranged gamma- or beta-globin or gamma beta- and beta gamma-globin genes linked to a 2.5-kb cassette containing sequences of the locus control region (LCR). Analysis of constructs containing two identical gamma or beta genes assessed the effect of gene order on globin gene expression, while analysis of constructs containing tandemly arranged gamma and beta genes assessed any additional effects of the trans-acting environment. When two gamma genes were tandemly linked to the LCR, expression from the proximal gamma gene was three- to fourfold higher than expression from the distal gamma gene, and the ratio of proximal to distal gene expression remained unchanged throughout development. Similarly, when two beta genes were tandemly linked to the LCR, the proximal beta gene was predominantly expressed throughout development. These results indicate that proximity to LCR increases gene expression, perhaps by influencing the frequency of interaction between the LCR and globin gene promoters. An arrangement where the gamma gene was proximal and the beta gene distal to the LCR resulted in predominant gamma-gene expression in the embryo. When the order was reversed and the gamma gene was placed distally to the LCR, gamma-gene expression in the embryo was still up to threefold higher than expression of the LCR-proximal beta gene. These findings suggest that the embryonic trans-acting environment interacts preferentially with the gamma genes irrespective of their order or proximity to the LCR. We conclude that promoter competition rather than gene order plays the major role in globin gene switching.

Role of gene order in developmental control of human gamma- and beta-globin gene expression

To determine the effect of gene order on globin gene developmental regulation, we produced transgenic mice containing two tandemly arranged gamma- or beta-globin or gamma beta- and beta gamma-globin genes linked to a 2.5-kb cassette containing sequences of the locus control region (LCR). Analysis of constructs containing two identical gamma or beta genes assessed the effect of gene order on globin gene expression, while analysis of constructs containing tandemly arranged gamma and beta genes assessed any additional effects of the trans-acting environment. When two gamma genes were tandemly linked to the LCR, expression from the proximal gamma gene was three- to fourfold higher than expression from the distal gamma gene, and the ratio of proximal to distal gene expression remained unchanged throughout development. Similarly, when two beta genes were tandemly linked to the LCR, the proximal beta gene was predominantly expressed throughout development. These results indicate that proximity to LCR increases gene expression, perhaps by influencing the frequency of interaction between the LCR and globin gene promoters. An arrangement where the gamma gene was proximal and the beta gene distal to the LCR resulted in predominant gamma-gene expression in the embryo. When the order was reversed and the gamma gene was placed distally to the LCR, gamma-gene expression in the embryo was still up to threefold higher than expression of the LCR-proximal beta gene. These findings suggest that the embryonic trans-acting environment interacts preferentially with the gamma genes irrespective of their order or proximity to the LCR. We conclude that promoter competition rather than gene order plays the major role in globin gene switching.

Serum factors can modulate the developmental clock of gamma- to beta-globin gene switching in somatic cell hybrids

The fusion of human fetal erythroid (HFE) cells with mouse erythroleukemia (MEL) cells produces stable synkaryons (HFE x MEL) which can be monitored for extended periods of time in culture. Initially these hybrids express a human fetal globin program (gamma >> beta), but after weeks or months in culture, they switch to an adult pattern of globin expression (beta >> gamma). The rate at which hybrids switch to the adult phenotype is roughly dependent on the gestational age of the fetal erythroid cells used in the fusion, suggesting that the rate of switching in vitro may be determined by a developmental clock type of mechanism, possibly involving the cumulative number of divisions experienced by the human fetal cells. To investigate whether the number or rate of cell divisions postfusion can influence the rate of switching, we monitored the rate of switching in hybrids from independent fusions under growth-promoting (serum-replete) and growth-suppressing (serum-deprived) conditions. We found that hybrids grown under serum-deprived or serumless conditions switched more rapidly to adult globin expression than did their counterparts in serum-replete conditions. Neither the number of cumulative cell divisions nor time in culture per se predicted the rate of switching in vitro. Our data suggest that factors present in serum either retard switching of hybrids by their presence or promote switching by their absence, indicating that globin switching in vitro can be modulated by the environment; however, once switching in HFE x MEL hybrids is complete, serum factors cannot reverse this process.