Search

Developing robust methodology for the sustainable production of red blood cells in vitro is essential for providing an alternative source of clinical-quality blood, particularly for individuals with rare blood group phenotypes. Immortalized erythroid progenitor cell lines are the most promising emergent technology for achieving this goal. We previously created the erythroid cell line BEL-A from bone marrow CD34+ cells that had improved differentiation and enucleation potential compared to other lines reported. In this study we show that our immortalization approach is reproducible for erythroid cells differentiated from bone marrow and also from far more accessible peripheral and cord blood CD34+ cells, consistently generating lines with similar improved erythroid performance. Extensive characterization of the lines shows them to accurately recapitulate their primary cell equivalents and provides a molecular signature for immortalization. In addition, we show that only cells at a specific stage of erythropoiesis, predominantly proerythroblasts, are amenable to immortalization. Our methodology provides a step forward in the drive for a sustainable supply of red cells for clinical use and for the generation of model cellular systems for the study of erythropoiesis in health and disease, with the added benefit of an indefinite expansion window for manipulation of molecular targets., Graphical abstract, Here Daniels et al. demonstrate reproducible immortalization of erythroid cells derived from bone marrow and peripheral and cord blood CD34+ cells and reveal a molecular signature of immortalization. This methodology provides a step toward the sustainable production of red cells for clinical use and for the generation of cellular model systems.

β-thalassemia is a prevalent genetic disorder causing severe anemia due to defective erythropoiesis, with few treatment options. Studying the underlying molecular defects is impeded by paucity of suitable patient material. In this study we created human disease cellular model systems for β-thalassemia, which accurately recapitulate the phenotype of patient erythroid cells. We also developed a high throughput compatible fluorometric-based assay for evaluating severity of disease phenotype and utilised the assay to demonstrate positive response of lines to verified reagents, providing validation for such applications.TMT-based comparative proteomics confirmed the same profile of proteins previously reported, whilst providing new insights into the altered molecular mechanisms in β-thalassemia erythroid cells, with upregulation of a wide range of biological pathways and processes observed.Overall, the lines provide a sustainable supply of disease cells as novel research tools, for identifying new therapeutic targets, and as screening platforms for novel drugs and therapeutic reagents.

We critically assess the proposal that succinate-fuelled reverse electron flow (REF) drives mitochondrial matrix superoxide production from Complex I early in reperfusion, thus acting as a key mediator of ischemia/reperfusion (IR) injury. Real-time surface fluorescence measurements of NAD(P)H and flavoprotein redox state suggest that conditions are unfavourable for REF during early reperfusion. Furthermore, rapid loss of succinate accumulated during ischemia can be explained by its efflux rather than oxidation. Moreover, succinate accumulation during ischemia is not attenuated by ischemic preconditioning (IP) despite powerful cardioprotection. In addition, measurement of intracellular reactive oxygen species (ROS) during reperfusion using surface fluorescence and mitochondrial aconitase activity detected major increases in ROS only after mitochondrial permeability transition pore (mPTP) opening was first detected. We conclude that mPTP opening is probably triggered initially by factors other than ROS, including increased mitochondrial [Ca2+]. However, IP only attenuates [Ca2+] increases later in reperfusion, again after initial mPTP opening, implying that IP regulates mPTP opening through additional mechanisms. One such is mitochondria-bound hexokinase 2 (HK2) which dissociates from mitochondria during ischemia in control hearts but not those subject to IP. Indeed, there is a strong correlation between the extent of HK2 loss from mitochondria during ischemia and infarct size on subsequent reperfusion. Mechanisms linking HK2 dissociation to mPTP sensitisation remain to be fully established but several related processes have been implicated including VDAC1 oligomerisation, the stability of contact sites between the inner and outer membranes, cristae morphology, Bcl-2 family members and mitochondrial fission proteins such as Drp1.

We previously demonstrated that hexokinase II (HK2) dissociation from mitochondria during cardiac ischemia correlates with cytochrome c (cyt-c) loss, oxidative stress and subsequentreperfusion injury. However, whether HK2 release is the primary signal mediating this ischemia-induced mitochondrial dysfunction was not established. To investigate this, we studied the effects of dissociating HK2 from isolated heart mitochondria. Mitochondria isolated from Langendorff-perfused rat hearts before and after 30 min global ischemia ± ischemic preconditioning (IPC) were subject to in vitro dissociation of HK2 by incubation with glucose-6-phosphate at pH 6.3. Prior HK2 dissociation from pre- or end-ischemic heart mitochondria had no effect on their cyt-c release, respiration (± ADP) or mitochondrial permeability transition pore (mPTP) opening. Inner mitochondrial membrane morphology was assessed indirectly by monitoring changes in light scattering (LS) and confirmed by transmission electron microscopy. Although no major ultrastructure differences were detected between pre- andend-ischemia mitochondria, the amplitude of changes in LS was reduced in the latter. This was prevented by IPC but not mimicked in vitro by HK2 dissociation. We also observed more Drp1, a mitochondrial fission protein, in end-ischemia mitochondria. IPC failed to prevent this increase but did decrease mitochondrial-associated dynamin 2. In vitro HK2 dissociation alone cannot replicate ischemia-induced effects on mitochondrial function implying that in vivo dissociation of HK2 modulates end-ischemia mitochondrial function indirectly perhaps involving interaction with mitochondrial fission proteins. The resulting changes in mitochondrial morphology and cristae structure would destabilize outer / inner membrane interactions, increase cyt-c release and enhance mPTP sensitivity to [Ca2+].

Development of in vitro culture systems for the generation of red blood cells is a goal of scientists globally with the aim of producing clinical grade products for transfusion. Although mature reticulocytes can be efficiently generated by such systems, the numbers produced fall short of that required for therapeutics, due to limited proliferative capacity of the erythroblasts. To overcome this hurdle, approaches are required to increase the expansion potential of such culture systems. The OP9 mouse stromal cell line is known to promote haematopoietic differentiation of pluripotent stem cells, however an effect of OP9 cells on erythropoiesis has not been explored. In this study, we show not only OP9 co-culture, but factors secreted by OP9 cells in isolation increase the proliferative potential of adult erythroid cells by delaying differentiation and hence maintaining self-renewing cells for an extended duration. The number of reticulocytes obtained was increased by approximately 3.5-fold, bringing it closer to that required for a therapeutic product. To identify the factors responsible, we analysed the OP9 cell secretome using comparative proteomics, identifying 18 candidate proteins. These data reveal the potential to increase erythroid cell numbers from in vitro culture systems without the need for genetic manipulation or co-culture.

Culture systems for human in vitro erythropoiesis are now well established. Using our 3-stage feeder-free erythroid culture system we can efficiently differentiate erythroid cells from adult and cord blood (CB) CD34+ cells with >105 fold expansion, enucleation rates of up to 95% and producing packed reticulocyte yields of >12ml post leukofiltration1. The final preparations for a first in man clinical trial of adult cultured reticulocytes produced under good manufacturing practice (RESTORE) are underway. Although we have shown that it is possible to modify the CD34+ derived cells using lentivirus for reengineering or additions to the medium, the finite proliferative capacity of CD34+ cells in culture currently limits yield. We therefore took the alternative approach of immortalising early erythroid cells differentiated from adult bone marrow (BM) CD34+ cells, creating the BEL-A (Bristol Erythroid Line Adult) line2, a sustainable erythroid cell source that recapitulates normal adult erythropoiesis, terminally differentiating to generate enucleated reticulocytes that express normal levels of adult globin. We have created a further 13 lines from BM, adult peripheral blood, CB and iPSC CD34+ cells, demonstrating reproducibility of the approach and its application to create lines from more accessible stem cell sources. Analysis of surface marker and globin profiles demonstrate the lines follow a similar differentiation profile to their respective primary cell source, with comparative proteomics confirming cell source representation of the lines. Lines always established at the pro-erythroblast/early basophilic stage, even when later stage erythroid cells were present in populations. As well as proof of principal as an alternative transfusion product and improved tools for studying erythropoiesis, such lines have far reaching additional applications, a number of which we are now exploring: Diagnostic and 'Universal' transfusion products: Presently serological testing by blood group reference laboratories relies on donated blood, which represent a finite resource and for some blood group phenotypes can be difficult to source. We used CRISPR-Cas9 gene editing to remove blood group antigens in order to generate a sustainable bank of cell lines with useful blood group phenotypes for diagnostic purposes3. Building on this, with the aim of developing a more compatible "universal" transfusion product to meet the needs of chronically transfused patients and those with rare blood group phenotypes, we used combinatorial gene targeting to create sublines deficient in multiple antigens responsible for the most common transfusion incompatibilities (ABO [Bombay], Rh [Rhnull], Kell [K0], Duffy [Fynull], GPB [S-s-U-]). Individual and multiple blood group knockout lines retained the ability to undergo terminal differentiation and enucleation, also illustrating the capacity for coexistence of multiple rare blood group phenotypes within viable reticulocytes3. Cytokine independent lines Cytokines represent a substantial cost contribution to erythroid culture systems. We therefore exploited activating mutations found in patient c-Kit and EPOR that cause hypersensitivity to ligand, to create cytokine independent lines thus increasing economic viability of cultured red cells. Bi-allelic EPOR or c-kit edits were introduced into BEL-A with confirmation and exploration of mechanism on differentiation in the absence, or with substantially reduced levels of cytokines. Model disease systems In addition to potential therapeutic applications we are also creating lines as model cellular disease systems for studying molecular mechanisms and as drug screening platforms, via CRISPR-Cas9 gene editing of BEL-A and by directly immortalising patient stem cells. To date we have made b-thalassemia major, HbE thalassemia and lines with KLF1 mutations. Furthermore, we have shown BEL-A reticulocytes support invasion and growth of Plasmodium falciparum and are utilising the line to study mechanisms of malaria parasite invasion4. Kupzig S, Parsons SF, Curnow E, Anstee DJ, Blair A. Superior survival of ex vivo cultured human reticulocytes following transfusion into mice. 2016;102:476-483Trakarnsanga K, Griffiths RE, Wilson MC, et al. An immortalized adult human erythroid line facilitates sustainable and scalable generation of functional red cells. Nat. Commun. 2017;8:14750Hawksworth J, Satchwell TJ, Meinders M, et al. Enhancement of red blood cell transfusion compatibility using CRISPR-mediated erythroblast gene editing. EMBO Mol. Med. 2018; 10:e8454Satchwell TJ, Wright K, Haydn-Smith K, et al. Genetic manipulation of cell line derived reticulocytes enables dissection of host malaria invasion requirements. Nat. Comm. 2019;10:3806 Disclosures No relevant conflicts of interest to declare.

Cardiac mitochondria can take up Ca(2+), competing with Ca(2+) transporters like the sarcoplasmic reticulum (SR) Ca(2+)-ATPase. Rapid mitochondrial [Ca(2+)] transients have been reported to be synchronized with normal cytosolic [Ca(2+)](i) transients. However, most intra-mitochondrial free [Ca(2+)] ([Ca(2+)](mito)) measurements have been uncalibrated, and potentially contaminated by non-mitochondrial signals. Here we measured calibrated [Ca(2+)](mito) in single rat myocytes using the ratiometric Ca(2+) indicator fura-2 AM and plasmalemmal permeabilization by saponin (to eliminate cytosolic fura-2). The steady-state [Ca(2+)](mito) dependence on [Ca(2+)](i) (with 5 mM EGTA) was sigmoid with [Ca(2+)](mito)

The mitochondrial permeability transition pore (MPTP) plays a key role in reperfusion injury, and preventing MPTP opening is cardioprotective. Here we use surface fluorescence of a Langendorff-perfused beating rat heart to establish the role of ROS and [Ca2+] during ischemia and reperfusion in mediating MPTP opening. We used real-time multi-wavelength measurements of surface fluorescence and reflectance to monitor autofluorescence and intracellular concentrations of both [Ca2+] and ROS using Indo-1 and 5-carboxy-2′,7′-dichlorofluorescein. Haemodynamic function and infarct size were measured in parallel. Increases in both ROS and [Ca2+] were detected on reperfusion and these were greatly attenuated by ischaemic preconditioning (Table 1) which reduced infarct size from 61.9±2.3% to 13.9±1.7% for IP. However, the time courses of the changes suggested that the increases in both parameters were secondary to MPTP opening rather acting as a causative event that than preceding MPTP opening. This conclusion received additional support from the observation that the MPTP inhibitor, cyclosporine A (0.2 μM CsA added 10 min before ischaemia and for 30 min of reperfusion), also reduced the rise in [Ca2+] upon reperfusion and decreased infarct size to 44.1±3.9%. Our data suggest that the observed changes in ROS and [Ca2+] during reperfusion may occur as a result of MPTP opening rather than acting as the trigger MPTP opening. We suggest that other factors such as hexokinase 2 dissociation from the mitochondria during ischemia may sensitize the MPTP to open at basal levels of [Ca2+] and ROS. Upon MPTP opening, the resulting mitochondrial dysfunction leads to ROS production, impaired ATP production and hence loss of calcium homeostasis that in turn leads to further MPTP opening and damage. This work was supported by the Medical Research Council and British Heart Foundation. | | Indo-1 ratio | DCF, normalized per value at 1 min on reperfusion | | ------------- | ---------------- | ------------------------------------------------- | ----------- | | | Control (n = 15) | IP (n = 10) | CsA (n = 8) | Control (n=6) | IP (N = 6) | | normoxia | 0.63±0.01 | 0.58±0.01 | 0.61±0.01 | 1.16±0.03 | 1.10±0.05 | | end of isch | 0.79±0.02 | 0.80±0.02 | 0.85±0.04 | | | | 3 min of rep | 0.67±0.03 | 0.50±0.01* | 0.55±0.02* | 1.01±0.02 | 0.97±0.01§ | | 10 min of rep | 0.87±0.10 | 0.53±0.02* | 0.59±0.02* | 1.10±0.03 | 0.95±0.01§ | | 30 min of rep | 0.85±0.06 | 0.55±0.02* | 0.59±0.03* | 1.27±0.03 | 1.04±0.02§ | * *, § indicates significant difference to control Table 1

ROS production during ischaemia-reperfusion induces mitochondrial permeability transition pore opening and so mediates reperfusion injury. Several groups have used real-time fluorescence to determine ROS production by isolated cardiomyocytes during simulated ischaemia-reperfusion but protocols and results vary. Few studies have utilised the more pathophysiologically relevant perfused heart. We have developed an ischaemia-reperfusion protocol that exposes isolated rat cardiac myocytes to conditions closely matching ischaemia-reperfusion in the perfused heart. Temperature and [oxygen] are tightly controlled and continuously monitored while ROS production is measured by fluorescence microscopy using dihydroethidium (DHE) or 2′,7′-dichlorofluorescein (DCF). In parallel real-time multiwavelength surface fluorescence and reflectance were used to monitor autofluorescence and ROS production in the perfused rat heart together with measurements of haemodynamic function and infarct size. In isolated myocytes DCF fluorescence increased continuously throughout the experiment, but with pronounced acceleration at the start of ischaemia and at reperfusion. For DHE, nuclear fluorescence increased steadily during ischaemia but dropped to baseline within seconds of reperfusion. In perfused hearts the pattern of fluorescence change was similar; a rapid increase during onset of ischaemia, a larger increase at ischaemic contracture followed by a rapid decrease on reperfusion. To define the nature of the fluorescence changes, emission spectra were determined on the perfused hearts. These confirmed our conclusion that DHE is not a suitable dye to monitor ROS production in situ. DCF appears suitable for use in cardiomyocytes and we are developing suitable loading protocols for perfused heart measurements.