“Connexin” the dots in hematopoietic stem cell potential Free

In this issue of Blood, Singh et al¹ demonstrate that actively cycling hematopoietic stem cells (HSCs) require mitochondrial connexin 43 (Cx43) during periods of replicative stress to preserve normal mitochondrial metabolism, matrix calcium levels, and AMP-activated protein kinase (AMPK) signaling to promote effective hematopoietic regeneration in vivo.

Previously, it was believed that all HSCs periodically enter and exit the cell cycle, with the entire HSC pool undergoing turnover within weeks. In subsequent years, studies have established the presence of subpopulations within the HSC compartment characterized by varying states of cell cycle entry. To identify these subpopulations, label retention involving 5-bromo-2′-deoxyuridine or the chromatin marker H2B-green fluorescent protein is used. Computational modeling of experimental data predicts that the most inactive subpopulation of quiescence, dormant HSCs, divide once every 145 days in mice (ie, ∼5 divisions during a murine lifetime), whereas active HSCs likely divide every 36 days.² It follows that biology provides a variety of HSC reserve pools in vivo to meet the hematopoietic regenerative demands of various challenges over the course of a lifetime.

In response to severe stress, such as blood loss, infection, or cytopenia, HSCs must exit quiescence, enter the cell cycle, and begin dividing to regenerate the blood system. During this active cycling phase, HSCs are especially vulnerable to stress, exhaustion, and cell death. To successfully restore hematopoiesis while preserving their ability to self-renew, cycling HSCs must activate a distinct molecular program that meets the increased energy demands. However, the mechanisms that distinguish quiescent HSCs from actively cycling ones are still not fully understood.

Cx43 (encoded by GJA1) is a member of the gap junction protein family, which assembles into hemichannel structures on membranes that facilitate the diffusion of low molecular weight materials in and out of the cell. For this reason, Cx43 plays an important role in facilitating cell-cell communications using nutrients, ions, and other small molecules. Prior research by the Cancelas group established the importance of the transfer of mitochondrial reactive oxygen species, specifically between cycling HSCs and bone marrow mesenchymal stromal cells under conditions of 5-fluorouracil–induced stress in a process dependent on Cx43.³ Despite canonical roles at the plasma membrane, connexins have also been discovered within organellar membranes, including the mitochondria.⁴ However, the mechanisms by which Cx43 supports mitochondrial function in HSCs remain unclear. Although ∼1500 nuclear genes are linked to the mitochondria, the roles of many mitochondrial proteins remain poorly defined, especially in HSC biology.⁵ 

In the current study, the investigators juxtapose the role of mitochondrial Cx43 in metabolic regulation and functional maintenance between quiescent (freshly isolated) and active cycling (48-hour culture) HSCs. Accelerated serial transplantation assays show that Cx43 deficiency leads to HSC exhaustion in vivo, establishing the requirement for Cx43 in cycling HSC function. Furthermore, Cx43 was shown to localize to both outer and inner mitochondrial membranes in HSCs. Here, the authors demonstrate mitochondrial Cx43 primarily influences 3 key features that affect HSC regenerative potential.

First, the authors show that quiescent Cx43-deficient HSCs accumulate more activated AMPKs near the mitochondria along with elevated ADP:ATP ratios, indicating altered metabolism. AMPK is one of the major sensors of energy status, and in the absence of sufficient energy, it downregulates the potential of cell growth and proliferation. Reduced mitochondrial respiration along with the altered expression of trichloroacetic acid cycle genes, including OGDH, indicate that Cx43-deficient HSCs possess ineffective energetic potential, thereby stimulating AMPK activation and reducing regenerative potential. Further data support direct interaction between Cx43 and AMPK. The data suggest that Cx43 serves as a rheostat to balance the regenerative potential of HSCs without leading to exhaustion, although how Cx43 and AMPK mediate these processes remains to be elucidated.

Second, the authors show that mitochondrial dynamics in Cx43-deficient cycling HSCs, but not quiescent HSCs, impair their regenerative potential. Under replicative stress, Cx43-deficient HSCs failed to maintain proper mitochondrial dynamics, driven by dynamin-related protein 1 (Drp1) fragmentation and the downregulation of mitofusin 2 fusion, which lead to reactive oxygen species production and hyperactivated mitophagy. These findings agree with the work in our laboratory and others, demonstrating a critical role for fusion-competent mitochondria in HSCs.^6,7 Interventions designed to disrupt mitochondrial fragmentation, such as expressing a dominant-negative Drp1 mutant (Drp1K38A) or the uptake of healthy mitochondria via ex vivo mitochondrial transfer to induce heteroplasmy, rescued the metabolic activity and function of Cx43-deficient HSCs in vivo. Given these findings, it is interesting to consider whether augmentation of the clinical sources of HSCs with Cx43-competent mitochondria could optimally prime HSCs for the burden of active cell cycling and improve long-term HSC potency and transplantation outcomes in patients.

Finally, the authors reveal that Cx43-deficient HSCs exhibit elevated mitochondrial matrix calcium (Ca²⁺) levels in both quiescent and cycling states, despite unchanged cytosolic Ca²⁺ levels. Functional assays confirmed that Cx43 facilitates Ca²⁺ efflux out of the mitochondria. Interestingly, the sodium transporter NCLX (Na⁺/Ca²⁺ exchanger) is most notably associated with Ca²⁺ efflux from the mitochondria, although the consequence of NCLX ablation is not well described in the hematopoietic system. Cx43 could play a role in regulating NCLX or have a separate function in mitochondrial Ca²⁺ efflux in HSCs, but further studies are needed to clarify this mechanism. The association of HSC exhaustion with the accumulation of cytoplasmic and mitochondrial Ca²⁺ is supported by our work and that of others,^8,9 suggesting that signaling through this important second messenger primes HSC for cell cycle activation and loss of self-renewal. Notably, the chelation of extracellular calcium rescues mitochondrial Ca²⁺ accumulation in Cx43-deficient HSCs and invites questions about the niche factors affecting Ca²⁺ regulation. Recently, the Lin group (Yeh et al¹⁰) has detected calvarium HSCs in bone-marrow environments with high extracellular Ca²⁺. These findings raise the possibility that the cell cycle status of HSCs is governed by niche-specific Ca²⁺ dynamics.

This study prompts further investigation into how Cx43 protects cycling HSCs from exhaustion during regeneration. Acting as a “brake” on AMPK signaling, a “driver” of mitochondrial respiration, and a “tailpipe” of mitochondrial Ca²⁺ efflux, Cx43 integrates multiple mitochondrial functions critical for HSC self-renewal. Its interactions with other mitochondrial membrane proteins likely shape key aspects of HSC metabolism and signaling. Notably, the role of Cx43 in HSC survival offers a promising avenue to improve HSC transplantation outcomes.

Conflict-of-interest disclosure: The author declares no competing financial interests.