為了深入了解嘌呤體在細胞中的功能，我們進行了RNA定序分析，並比較黑色素瘤細胞表現野生型與泛素化缺陷型PAICS的轉錄組變化，並發現粒腺體本身編碼的mRNA急劇減少。值得注意的是，在高度表達ASB11的心肌與骨骼肌細胞中，這套驅動嘌呤體組裝的分子機制也被保留了下來。基於上述發現，我們懷疑以了滿足這些肌肉持續收縮的能量需求，高表達的ASB11所誘導的嘌呤體形成，對於維持心臟與骨骼肌肉中大量粒腺體的功能十分地重要。

我們進一步發現，嘌呤體生成的嘌呤會優先提供給粒腺體用於DNA與RNA的合成，以維持粒腺體的氧化磷酸化的活性、動態以及氧化還原的恆定。嘌呤體對粒腺體的選擇性作用是透過連接嘌呤體與粒腺體來實現的，這項工作主要是由Spartin蛋白所主導。已知，Spartin蛋白的功能缺失或突變會導致特羅耶氏症候群（Troyer syndrome），並伴隨粒腺體和肌肉功能的障礙。我們的研究結果也表明，透過shRNA沉默Spartin的表現會降低嘌呤體與粒腺體的接近程度，並減少粒腺體RNA的合成與粒腺體DNA的完整性。此外， Spartin的表現減少同樣也會造成粒腺體中的嘌呤代謝物普遍減少，此結果與嘌呤體缺陷的表現型相似。透過讓嘌呤體與粒腺體強制互相靠近，可以完全彌補Spartin沉默細胞中所觀察到的缺陷，像是降低的mtRNA合成、mtDNA的完整性與粒腺體中嘌呤衍生代謝物的含量，這顯示嘌呤體與粒腺體互相靠近，對於粒腺體接收通過嘌呤體合成的嘌呤至關重要。

透過單細胞RNA定序發現，ASB11在心肌病變患者的心臟組織中表達下調，在粒腺體代謝高度活化的心肌細胞亞群中特別明顯。有鑑於ASB11是嘌呤體形成的驅動因子，心肌病變狀態下ASB11表現的降低會損害嘌呤體的形成，提示嘌呤體缺乏可能是心臟疾病的潛在致病因素。如果嘌呤體缺乏確實參與心肌病變的發病機制，那麼透過重新表達ASB11以誘導嘌呤體形成，或透過額外補充嘌呤核苷肌苷均(inosine)應能改善心臟功能的障礙。相反的，ASB11的缺失則會因為嘌呤體缺乏而進一步加劇心臟功能的障礙。在此，我們採用左前降支動脈（LAD）結紮術誘發心臟缺血性損傷，以此作為心肌病變模型。

我們的結果顯示，嘌呤體誘導或肌苷給藥可減輕缺血性損傷引起的心臟功能障礙，而心臟特異性的Asb11基因敲除會導致嘌呤體的生成減少，從而加重缺血性心肌病變。綜上所述，我們的研究揭示了，嘌呤體在支持粒線體的功能，與減輕心肌損傷方面前所未有的作用，並提出了一種治療缺血性心肌病變的潛在策略。 

（103年度TBF學術講座、中研院生物化學研究所　陳瑞華特聘研究員）

 Purinosome fuels mitochondrial DNA and RNA synthesis to ameliorate myocardial ischemic injury

Purinosomes are cytoplasmic condensates that compartmentalize de novo purine synthesis (DNPS) enzymes to enhance metabolic flux through the pathway. However, the pathophysiological functions of purinosomes remain largely unexplored. Previously, we identified the first mechanistic basis underlying purinosome assembly, demonstrating that ASB11, an E3 ubiquitin ligase, mediates K6-linked ubiquitination of PAICS at lysine 74. Ubiquitinated PAICS subsequently recruits UBAP2, a ubiquitin-binding protein, facilitating multivalent intermolecular interactions that drive purinosome assembly through liquid–liquid phase separation. Notably, the ubiquitination-deficient PAICS mutant (PAICS K74R) exhibits defective purinosome assembly while retaining enzymatic activity comparable to that of wild-type PAICS. Therefore, this mutant serves as an ideal tool for specifically investigating purinosome function.

 

To gain insight into the cellular functions of purinosomes, we performed RNA-seq analysis comparing purinosome-proficient and purinosome-deficient melanoma cells. Strikingly, purinosome deficiency resulted in a marked reduction in mitochondrial transcripts encoded by the mitochondrial genome. In normal tissues, ASB11 is expressed almost exclusively in cardiac and skeletal muscle tissues. Consistent with this expression pattern, we observed constitutive purinosome formation in mouse cardiac and skeletal muscle cell lines. Based on these findings, we hypothesized that constitutive purinosome assembly driven by elevated ASB11 expression is essential for maintaining mitochondrial function in highly mitochondria-enriched muscle tissues that require continuous contractile activity.

 

To test this hypothesis, we established purinosome-deficient murine cardiomyocytes using the HL-1 cell line. Our results demonstrated that purinosome deficiency indeed reduced mitochondrial transcript levels, which could be fully rescued by supplementation with the purine nucleosides adenosine and guanosine, indicating that insufficient purine supply is the underlying cause. Given that purines are essential building blocks for both DNA and RNA synthesis, purinosome-deficient cells also exhibited mtDNA defects, including impaired mtDNA integrity and replication. Consistent with defective mitochondrial RNA and DNA synthesis, mitochondrial proteins were globally downregulated in purinosome-deficient cells, leading to dysregulation of mitochondrial respiratory activity, dynamics, and redox homeostasis. Importantly, purine supplementation efficiently rescued the defects in mtDNA and mitochondrial protein expression caused by purinosome deficiency. Collectively, these findings demonstrate that mitochondria-rich cells rely on purinosome-enhanced purine nucleotide production to support mtDNA and mtRNA synthesis, thereby maintaining mitochondrial function.

 

Purine supplementation rescued nearly all defects caused by purinosome deficiency, ranging from mtRNA and mtDNA abnormalities to mitochondrial protein loss, further supporting the notion that insufficient purine supply is the root cause of these defects. Mass spectrometry-based metabolomic analysis further demonstrated that purinosome deficiency decreases purine ribonucleotide levels in both whole cells and mitochondria. Similar results were observed for deoxyribonucleotide pools, showing reduced purine deoxyribonucleotides without significant alterations in pyrimidine deoxyribonucleotides. Although purinosome-deficient cells displayed a substantial reduction in steady-state purine nucleotide levels, direct evidence demonstrating impaired DNPS flux remained lacking. Moreover, whether purinosome deficiency specifically reduces incorporation of DNPS-derived purines into mtRNA and mtDNA remained unclear.

 

To address these questions, we performed metabolic isotope tracing using radioactive 14C-glycine to monitor incorporation of DNPS-derived purines into RNA and DNA. Surprisingly, purinosome deficiency selectively decreased incorporation of newly synthesized purines into mtRNA and mtDNA without substantially affecting total RNA or DNA labeling. These findings indicate that purinosomes exert a selective effect on mitochondrial nucleic acid synthesis. Consistently, RNA-seq analysis revealed that purinosome deficiency specifically reduced mitochondrially encoded transcripts, while nuclear-encoded transcripts remained largely unaffected.

 

These observations prompted us to investigate the basis for this selective effect and why purinosomes preferentially support mtRNA and mtDNA synthesis. Purinosomes are known to localize in close proximity to mitochondria, enabling efficient channeling of mitochondria-derived metabolic intermediates—including glycine, glutamine, aspartate, and 10-formyl-THF—into the DNPS pathway. However, whether purinosomes directly influence mitochondrial function has remained unknown. Based on previous reports and our observations, we hypothesized that the spatial proximity of purinosomes to mitochondria facilitates local delivery of newly synthesized purines for mitochondrial nucleic acid synthesis.

 

To test this hypothesis, we sought to identify molecular factors mediating purinosome–mitochondria association. We therefore performed proximity proteomics using one of the DNPS enzymes, FGAMS, as bait to identify purinosome-interacting proteins. Candidate tethering proteins would be expected to interact with both purinosomes and mitochondria. Among the identified proteins, Spartin emerged as the strongest candidate because of its known ability to associate with the mitochondrial outer membrane and microtubules. Notably, Spartin deficiency has previously been linked to mitochondrial dysfunction, resembling the phenotypes observed in purinosome-deficient cells.

 

To determine whether Spartin functions as a molecular tether linking purinosomes to mitochondria, we generated Spartin knockdown (KD) HL-1 cells using shRNA. Spartin depletion impaired the spatial proximity between purinosomes and mitochondria, resulting in defective mtRNA synthesis and compromised mtDNA integrity. Furthermore, mitochondrial purine-derived metabolites were broadly reduced in Spartin KD cells, phenocopying purinosome deficiency. Importantly, Spartin depletion did not further exacerbate defects in purinosome-deficient cells, suggesting that Spartin-dependent regulation of mtRNA and mtDNA synthesis requires functional purinosomes. Strikingly, forced recruitment of purinosomes to mitochondria completely rescued the defects in mtRNA synthesis, mtDNA integrity, and mitochondrial purine metabolite pools observed in Spartin KD cells. These findings indicate that localization of purinosomes to the mitochondrial periphery is essential for efficient delivery of purine metabolites to mitochondria.

 

Importantly, ASB11 expression is significantly downregulated in cardiac tissues from patients with cardiomyopathy, particularly within cardiomyocyte subpopulations characterized by active mitochondrial metabolism. Given that ASB11 acts as a key driver of purinosome assembly, reduced ASB11 expression in cardiomyopathy is expected to impair purinosome formation, suggesting that purinosome deficiency may contribute to cardiac pathogenesis. If this hypothesis is correct, restoration of purinosome formation through ASB11 re-expression or supplementation with inosine, a purine nucleoside, should ameliorate cardiac dysfunction. Conversely, loss of ASB11 would be expected to exacerbate cardiac dysfunction through purinosome deficiency.

 

To investigate this possibility, we employed a left anterior descending artery (LAD) ligation model to induce ischemic cardiomyopathy. Consistent with our hypothesis, induction of purinosome formation or inosine administration significantly alleviated cardiac dysfunction following ischemic injury, whereas cardiac-specific Asb11 knockout, which disrupts purinosome formation, markedly aggravated ischemic cardiomyopathy. Taken together, our study identifies a previously unrecognized role of purinosomes in supporting mitochondrial function to protect against myocardial injury and highlights purinosome modulation as a potential therapeutic strategy for ischemic cardiomyopathy.

 (2014 TBF Chair in Biotechnology, Dr. Ruey-Hwa Chen)