My Love Story With HIFs in the Disc: A Journey to the Humboldt Research Award Authors: Makarand V. Risbud Published: 2026 ABSTRACT ABSTRACTThe intervertebral disc is a prime example of tissue adaptation to metabolic and mechanical extremes. As the largest avascular organ in the human body, the disc and the central nucleus pulposus (NP) endure constant compressive loading while sustaining a hypoxic, relatively acidic, and hyperosmotic microenvironment that would be inhospitable to most cell types. Over the past two decades, work from my laboratory has revealed extraordinary details of this type of adaptation, that hypoxia‐inducible factors (HIFs), particularly HIF‐1α, serve as master regulators of NP cell survival, metabolism, and matrix homeostasis. This review chronicles the scientific and personal journey that began with a single unexpected observation: the normoxic stabilization of HIF‐1α in NP cells. That finding grew into a substantial understanding of oxygen‐independent HIF regulation and biology in the disc. Through sustained collaborations with Irving Shapiro and other colleagues and a succession of talented trainees, we demonstrated that HIF‐1α is indispensable for disc development and homeostasis. In contrast, HIF‐2α plays a distinct, largely opposing role, promoting matrix catabolism and fibrosis. Our work has revealed the molecular architecture underlying the disc's adaptation to avascularity, encompassing HIF‐dependent regulation of glycolytic enzymes, glucose transporters, carbonic anhydrases CA9 and CA12, monocarboxylate transporter 4 (MCT4), and BNIP3‐mediated mitophagy, and has opened promising therapeutic avenues for disc regeneration. Recently, we discovered that NP‐derived lactate exported via MCT4 is actively imported primarily by vertebral endplate cells, and, to a lesser extent, by annulus fibrosus cells via MCT1, where it serves dual roles as a TCA metabolite and an epigenetic regulator through lactylation. This review celebrates the intellectual journey, the collaborative spirit that sustained it, and the honor of the Alexander von Humboldt Research Award, while providing a synthesis of HIF biology in the intervertebral disc spanning the discovery of oxygen by Priestley 1774 to the 2019 Nobel Prize and beyond. Introduction: A Personal Journey I still remember the moment vividly. It was 2004, and I was staring at a Western blot in my laboratory, puzzled by the result. My colleague, Asha Guttappali, and I had cultured primary rat nucleus pulposus (NP) cells under standard normoxic conditions: 20% oxygen, the same atmosphere in which most cell biologists routinely maintain their cultures, expecting to detect little or no HIF‐1α protein. The textbook dogma at the time was unambiguous: HIF‐1α is rapidly hydroxylated by prolyl hydroxylase domain (PHD) enzymes under normoxia, polyubiquitinated by the von Hippel–Lindau (VHL) E3 ligase complex, and degraded by the 26S proteasome within minutes of synthesis. Yet there it was, a clear HIF‐1α band in our normoxic NP cell lysates, as intense as anything we observed in our hypoxic controls. That observation, ultimately published in the Journal of Cellular Biochemistry in 2006 [1], redirected my laboratory and set in motion a two‐decade scientific adventure that has carried me from the research corridors of Thomas Jefferson University to laboratories and scientific conferences across the world. It also, in ways I could not have anticipated at the time, led to the honor that occasions this review: the Alexander von Humboldt Research Award. My entry into disc biology was itself something of a happy accident. I had trained as a biochemist, tissue engineer, and cell biologist, with a deep interest in hypoxia signaling and cellular adaptation to metabolic stress. When I joined the Orthopedic Research Division headed by Irving Shapiro at Thomas Jefferson University, I encountered for the first time the extraordinary biology of the intervertebral disc, a tissue that challenged my assumptions about how cells survive and function. During those early days, I worked with Irving on a study investigating phenotypic markers that can distinguish NP cells from other cell types in the disc. Subsequently, we demonstrated robust, selective expression of HIF‐1α in NP tissue, with minimal expression in the AF and endplate cartilage, which piqued my long‐term interest in hypoxia and HIF biology [2, 3]. Irving's infectious enthusiasm for disc biology, his encyclopedic command of skeletal physiology, and his insistence on rigorous, mechanistic science have profoundly shaped my own approach to research. The work described in this review is as much his as it is mine, and I remain deeply grateful for his mentorship, partnership, and friendship over more than two decades. The Humboldt Research Award, conferred by the Alexander von Humboldt Foundation to recognize scientists whose fundamental discoveries have had a lasting impact on their discipline, is an occasion for both reflection and gratitude. It invites one to look back at the experiments that worked and the far greater number that did not; at the trainees who have built their own laboratories; at the collaborators who have enriched the science at every turn. This review is written in that spirit; part personal narrative, part scientific synthesis, and entirely an expression of gratitude for the extraordinary privilege of doing science. The Intervertebral Disc: A Tissue Built for Extremes To appreciate why HIF proteins are so central to disc biology, one must first understand the biological environment of the intervertebral disc cells. The intervertebral disc is the largest avascular organ in the human body, a distinction that immediately sets it apart from virtually every other tissue and imposes profound constraints on how its cells obtain oxygen and nutrients, remove metabolic waste, and maintain homeostasis [2, 4, 5]. Anatomy and Cellular Composition Interposed between adjacent vertebral bodies that comprise the spinal motion segment, IVD forms a specialized diarthrodial joint that provides both structural support and the flexibility essential for spinal motion [6]. The disc comprises three anatomically and functionally distinct compartments: the nucleus pulposus (NP), a gelatinous, central core; the annulus fibrosus (AF), a concentrically lamellar fibrocartilaginous outer ring; and the cartilage endplates (CEP), thin layers of hyaline cartilage interfacing with the bony endplates and vertebral bodies above and below [6]. The NP is the primary focus of my laboratory's research. It derives embryologically from the notochord, the transient axial structure that serves as the primary scaffold of the vertebrate body plan and, in the mature disc, retains a population of notochord‐derived cells with a distinctive phenotype. These include large vacuolated “physaliferous” cells that are gradually transformed into smaller, chondrocyte‐like cells during postnatal development and aging [7, 8]. The Avascular and Aneural Microenvironment The most remarkable feature of the NP from a cell biology perspective is its complete avascularity. Blood vessels do not penetrate the disc parenchyma; the nearest capillary beds lie at the vertebral endplate margins, millimeters from the cells in the human disc center [2]. Consequently, NP cells must obtain oxygen and glucose, and eliminate carbon dioxide, lactate/H+, entirely by diffusion through the dense and often calcified extracellular matrix of the cartilage endplates. The metabolic consequences are dramatic: oxygen tensions in the NP center have been measured at 0.5%–5% (approximately 4–38 mmHg), well within the range of physiological hypoxia [2, 5]. Glucose concentrations are similarly depleted, typically 0.5–2 mM in the disc center compared with 5 mM in plasma, and glucose uptake is thought to reduce with age due to endplate sclerosis, ECM crosslinking, and increased calcification [5, 7, 9, 10]. The disc is also largely aneural under healthy conditions, with nerve fibers restricted to the very outer AF lamellae; innervation of the deep layers of the AF and occasionally NP, is a pathological finding associated with degenerated discs and discogenic pain in humans and mouse models [4, 11, 12]. Mechanical and Osmotic Loading The IVD is the primary load‐bearing structure of the spine, transmitting and distributing compressive, tensile, torsional, and shear forces generated by body weight, muscle activity, and movement. Intradiscal pressures in the NP can reach 1–3 MPa during everyday activities such as walking, lifting, and bending [4]. This mechanical loading is not merely a passive challenge; it is an active regulatory signal that modulates disc cell biology through mechanotransduction pathways and modulates circadian pathways critical in maintaining disc health [13, 14, 15]. The interplay among mechanical loading, osmotic stress, and hypoxia creates a uniquely complex regulatory environment in which HIF‐1 signaling plays a central integrative role [16]. Particularly, syndecan, a cell surface heparan sulfate proteoglycan that we showed to be hypoxia and HIF‐1 responsive, plays a role in NP cell mechanosensisng [17, 18]. A further distinctive feature of the NP microenvironment is its elevated osmolarity. The NP extracellular matrix is extraordinarily rich in the proteoglycan aggrecan, whose densely sulfated glycosaminoglycan chains carry a high fixed negative charge density that attracts cations and draws water into the tissue by Donnan osmosis [19, 20]. As a result, the osmolarity of NP interstitial fluid is approximately 430–500 mOsm/kg, substantially higher than the 285–295 mOsm/kg of plasma [19]. NP cells are therefore constitutively exposed to a hyperosmotic environment that would be detrimental to most other cell types. These osmotic changes are dynamic, fluctuating with mechanical loading, disc hydration state, and aging. The ability of NP cells to maintain their shape, phenotype, and function in this environment depends on a suite of osmoadaptive mechanisms, several of which intersect with HIF signaling pathways [21, 22]. Why the Disc Needs HIF Taken together, the avascular, aneural, hypoxic, hyperosmotic, and mechanically loaded nature of the NP microenvironment creates a multitude of survival challenges without parallel in mammalian physiology. NP cells must generate sufficient ATP to maintain ion homeostasis, synthesize and turn over an enormous extracellular matrix, respond to mechanical and osmotic stimuli, and resist apoptotic and autophagic death signals, all under conditions of oxygen deprivation. It is perhaps unsurprising, in retrospect, that these cells have evolved a constitutive dependence on HIF signaling. What was genuinely surprising and what drove much of the work described in this review was the discovery that this dependence extends even to normoxia, reflecting a fundamental reprogramming of the oxygen‐sensing machinery in NP cells relative to other tissues. A Brief History of Oxygen Sensing: From Priestley to the Nobel Prize The story of how cells sense and respond to oxygen spans two centuries of chemistry, physiology, biochemistry, and molecular biology. It begins in a garden in Leeds, England, in the summer of 1774. The Discovery of Oxygen: Joseph Priestley (1774) On August 1, 1774, the English clergyman and natural philosopher Joseph Priestley focused sunlight through a burning glass (a powerful magnifying glass) onto a sample of mercuric oxide and collected the gas released. He found that a candle burned more brightly in this gas than in ordinary air, and that a mouse placed in a sealed vessel containing the gas survived considerably longer than one confined to the same volume of common air. Priestley called his discovery “dephlogisticated air,” interpreting it within the framework of the phlogiston theory then dominant in chemistry, the notion that combustible materials release a substance called phlogiston during burning. His account of the discovery, published in 1775, described the new gas as “five or six times as good as common air for the purpose of respiration” [23]. Naming of Oxygen: Antoine Lavoisier (1778) It was the French chemist Antoine‐Laurent de Lavoisier who recognized the true significance of Priestley's discovery and gave it its modern interpretation. Lavoisier repeated and extended Priestley's experiments, demonstrating that both combustion and respiration involve the combination of substances with the new gas, consuming it while producing carbon dioxide and water. In 1778, he named the gas “oxygène” from the Greek for “acid‐former” based on his, as it turned out, incorrect belief that it was a constituent of all acids [8]. Lavoisier's oxygen theory of combustion, presented in his landmark Traité Élémentaire de Chimie (1789), overthrew the phlogiston theory and established the foundations of modern chemistry [24]. More fundamentally for our purposes, Lavoisier recognized that respiration was essentially a slow combustion in which living organisms consume oxygen and produce carbon dioxide in a process analogous to the burning of a candle, thereby laying the conceptual groundwork for the entire field of cellular bioenergetics. From Physiology to Molecular Biology: The Long Road to HIF For nearly two centuries after Lavoisier, the study of oxygen in biology proceeded at the physiological and biochemical levels. Researchers characterized hemoglobin–oxygen binding kinetics; the respiratory chain and oxidative phosphorylation, the Pasteur effect (inhibition of fermentation by oxygen); and the physiological responses to hypoxia, including erythropoiesis, angiogenesis, and the ventilatory response. The molecular mechanisms by which individual cells sensed and responded to changes in oxygen tension, however, remained entirely obscure until the late twentieth century. The breakthrough came from an unexpected direction: the regulation of the erythropoietin (EPO) gene. In 1953, Dr. Allan J. Erslev provided the first definitive proof of the existence of what was later termed erythropoietin (EPO), the glycoprotein hormone that stimulates red blood cell production [25]. Later, working with Jamie Caro and colleagues at Jefferson Medical College, they were among the first to demonstrate that hypoxia regulates EPO at the gene transcription level, though the molecular basis of this induction remained unknown [26, 27]. In 1991, Gregg Semenza and colleagues at Johns Hopkins University identified a DNA sequence element in the 3′ enhancer of the human EPO gene that was required for hypoxic induction [28]. They named the nuclear factor that bound this element “hypoxia‐inducible factor 1” (HIF‐1). By 1995, Semenza's group had purified and characterized HIF‐1 as a heterodimer of two basic helix–loop–helix PAS domain proteins: the oxygen‐regulated HIF‐1α subunit and the constitutively expressed HIF‐1β (also known as ARNT) [29]. These discoveries established HIF‐1 as the master transcriptional regulator of the cellular hypoxic response. The VHL/PHD Oxygen‐Sensing Machinery: Kaelin and Ratcliffe The mechanism by which cells sense oxygen and regulate HIF‐1α stability was elucidated through converging work in cancer biology and cell physiology. William Kaelin Jr. at the Dana‐Farber Cancer Institute was studying the von Hippel–Lindau tumor suppressor protein (VHL), mutations of which cause a hereditary cancer, clear cell renal cell carcinoma (ccRCC), characterized by highly vascularized tumors. He found that VHL‐deficient cells constitutively expressed hypoxia‐induced (HIF‐dependent) genes even under normoxia, suggesting that pVHL may modulate signal transduction in response to changes in oxygen tension [30, 31]. Peter Ratcliffe at the University of Oxford independently identified VHL as the E3 ubiquitin ligase that targets HIF‐1α for 26S proteasomal degradation [32, 33]. The key mechanistic link, the oxygen‐sensing step itself, was the discovery that prolyl hydroxylase domain (PHD) enzymes hydroxylate specific proline residues in the oxygen‐dependent degradation domain (ODD) of HIF‐1α in an oxygen‐ and 2‐oxoglutarate‐dependent reaction, creating the binding site for VHL and triggering HIF‐1α ubiquitination and degradation [32, 33]. Under hypoxia, PHD activity is suppressed, HIF‐1α accumulates, dimerizes with HIF‐1β/ARNT, and transactivates hundreds of target genes bearing hypoxia response elements (HREs) in their regulatory regions. The 2019 Nobel Prize in Physiology or Medicine The 2019 Nobel Prize in Physiology or Medicine was awarded jointly to William G. Kaelin Jr., Sir Peter J. Ratcliffe, and Gregg L. Semenza “for their discoveries of how cells sense and adapt to oxygen availability.” The Nobel Committee's citation recognized that their work had revealed one of the most fundamental adaptive mechanisms in animal physiology, a mechanism with profound implications for understanding anemia, cancer, ischemic disease, and, as my laboratory has demonstrated, the biology of the intervertebral disc. I had the privilege of personally interacting with these stalwarts, and their generosity in sharing reagents, ideas, and encouragement with our laboratory in its early years was invaluable. The Nobel Prize honored not only their individual brilliance but also the collective effort of a scientific community committed to understanding how life adapts to one of its most fundamental constraints. HIF‐1α in Nucleus Pulposus Cells: An Unexpected Stabilization The First Observation When we first observed robust HIF‐1α protein expression in rat NP cells cultured under normoxic conditions (20% O2), our initial reaction was skepticism; we assumed we had made a technical error. We repeated the experiment multiple times, using different cell preparations, antibodies, and detection methods, consistently obtaining the same result: NP cells constitutively express HIF‐1α, regardless of oxygen tension [1]. This stood in stark contrast to control cells, human chondrocytes, HeLa cells, and fibroblasts, in which HIF‐1α was barely detectable under normoxia and strongly induced under hypoxia, precisely as expected from the canonical PHD/VHL pathway. Our 2006 paper in the Journal of Cellular Biochemistry established several key findings [1]. First, NP cells express functionally active HIF‐1α protein under normoxic culture conditions. Second, this normoxic expression is not a cell culture artifact but reflects the in vivo microenvironment of the disc, where HIF‐1α immunoreactivity is robust throughout the NP. Third, HIF‐1α expression in NP cells can serve as a phenotypic marker distinguishing them from AF cells and endplate chondrocytes, consistent with the unique developmental origin of NP cells from the notochord [1, 3]. This last point has proven particularly valuable for the field, given the persistent challenge of identifying reliable NP cell‐specific markers [34]. Stabilized HIF‐1α Drives Glycolysis and Aggrecan Expression The functional significance of HIF‐1α stabilization was addressed in our 2007 study in the American Journal of Physiology Cell Physiology [35]. We demonstrated that NP cells generate energy primarily through glycolysis rather than oxidative phosphorylation, even under normoxic conditions when oxygen is available for mitochondrial respiration [12]. Quantitative measurements revealed that approximately 75% of ATP production in NP cells is derived from glycolysis, a metabolic profile more reminiscent of the Warburg effect in cancer cells than in normal differentiated tissue [35]. Critically, this glycolytic phenotype was shown to depend on normoxic HIF‐1α: knockdown of HIF‐1α under normoxia significantly reduced the expression of the glucose transporters GLUT‐1 and GLUT‐3, the glycolytic enzymes phosphoglycerate kinase (PGK) and lactate dehydrogenase A (LDHA), and led to corresponding reductions in glycolytic ATP production [35]. HIF‐1α was also shown to regulate aggrecan gene expression in NP cells, establishing a direct link between oxygen sensing and maintenance of the proteoglycan‐rich extracellular matrix that defines the NP phenotype [35]. This Acan regulation could be a result of a direct and indirect regulation of gene expression through SOX9 [36]. Importantly, HIF‐1α transcript levels and 3′‐UTR reporter activity were similar under normoxic and hypoxic conditions, indicating that stabilization reflects posttranslational rather than transcriptional regulation [35]. Oxygen‐Independent HIF Degradation and Activity Modulation The mechanistic basis for HIF‐1α stabilization was elucidated in two papers published in 2011 and 2012 in the Journal of Bone and Mineral Research and the Journal of Biological Chemistry [37, 38]. This work by Nobuyuki Fujita revealed that degradation of both HIF‐1α and HIF‐2α in NP cells is regulated predominantly through oxygen‐independent pathways, a fundamental departure from the canonical PHD/VHL mechanism operating in most other cell types [37]. Several converging lines of evidence supported this conclusion. Treatment with the PHD inhibitor DMOG (dimethyloxalylglycine) strongly induced HIF‐1α accumulation and HIF‐1 target gene expression (enolase‐1, VEGF) in control chondrocytes but had only a modest effect on HIF‐1α levels in NP cells, consistent with PHD activity being already low or functionally dispensable in these cells [37]. PHD2, the predominant PHD isoform expressed in NP tissue, had only a limited role in HIF‐1α degradation when overexpressed or knocked down in NP cells, and had no effect whatsoever on HIF‐2α degradation [37]. Inhibition of the 26S proteasome with MG132 caused HIF‐1α accumulation in NP cells, while inhibition of lysosomal proteolysis with bafilomycin A1 or chloroquine caused HIF‐1α accumulation, but only modest HIF‐2α accumulation, indicating that both proteasomal and lysosomal pathways contribute to HIF‐α turnover in NP cells through yet unidentified oxygen‐independent mechanisms [37]. Later studies from Zachary Schoepflin showed an interesting role for PHD3 in NP cells, which served as a transcriptional co‐activator of HIF‐1α instead of its canonical function in HIF‐α degradation, and that this regulation was independent of the pyruvate kinase muscle 2 (PKM2)‐Jumonji domain‐containing protein (JMJD5) axis [39, 40]. Our subsequent studies identified the contribution of multiple histone deacetylases (HDACs), particularly Class I and Class IIa HDACs, in controlling HIF‐1α stability via PHD2 in NP cells. Moreover, HDAC6, a Class IIb HDAC, influenced HIF‐1α activity, possibly through regulating the activity of p300 and HSP90, known HIF‐1 cofactors [41]. Studies by Cassie Tran further uncovered cross‐talk and negative feedback loop between CCN2, an important matricellular protein that functions to modulate activity growth factors, and HIF‐1α in NP cells [42]. Notably, unlike other cell types, studies of Yuichiro Hirose demonstrated that Factor inhibitory to HIF‐1 (FIH‐1) does not represent a major mechanism by which NP cells control HIF‐1‐dependent transcription, possibly due to low endogenous levels and/or preferential association of FIH‐1 with binding partners such as Notch‐ICD (N1‐ICD and N2‐ICD) [43]. Another possibility for the persistent presence of HIF‐α is increased translation rate, a hypothesis we hope to test soon. Together, these findings provided a molecular explanation for the constitutive stability of HIF‐1α observed in NP cells and established that these cells have fundamentally reprogrammed their oxygen‐sensing machinery as an adaptation to the disc's constitutively hypoxic environment. HIF‐1α as Master Regulator of NP Cell Survival and Metabolism Conditional Deletion Reveals Unequivocal Dependence The most compelling evidence for the essential role of HIF‐1α in disc biology came from conditional genetic deletion studies. In our 2014 collaborative study with Ernestina Schipani's group, Christophe Merceron used a notochord‐specific Foxa2‐Cre driver to achieve NP‐specific deletion of HIF‐1α, generating mice in which HIF‐1α was selectively ablated in notochord‐derived NP cells (Hif1A Foxa2Cre ) [44]. The resulting phenotype was severe. At birth, HIF‐1α‐null NP cells exhibited a massive wave of apoptotic cell death, with 39.2% ± 4.46% of NP cells TUNEL‐positive compared with minimal cell death in littermate controls [41]. By 1 month of age, the NP had completely disappeared, replaced by a fibrocartilaginous tissue lacking the characteristic notochordal cell phenotype and proteoglycan‐rich matrix [44]. Mechanistic analysis revealed that this cell death was primarily metabolic in origin: HIF‐1α‐null NP cells showed markedly reduced expression of PGK and other glycolytic enzymes, consistent with an inability to sustain glycolytic ATP production in the hypoxic disc environment [44]. The functional consequences were equally severe; biomechanical testing of HIF‐1α‐null discs revealed a 40% reduction in load‐damping capacity compared with controls, demonstrating that loss of HIF‐1α not only promotes NP cell apoptosis but also fundamentally compromises the mechanical function of the disc [44]. A parallel study using Shh‐Cre to delete HIF‐1α in notochord‐derived cells confirmed these findings, showing accelerated degenerative changes, increased NP cell death, and reduced aggrecan and Type II collagen expression [45]. Glycolytic Metabolism and Energy Homeostasis The dependence of NP cells on glycolytic metabolism is not merely a consequence of oxygen deprivation, it is an active, HIF‐1α‐driven program that enables these cells to generate sufficient ATP in their physiologically hypoxic environment. As demonstrated by Agrawal et al. [35], NP cells express high constitutive levels of GLUT‐1, GLUT‐3, GAPDH, PGK, and LDHA under normoxic conditions, all of which are under the transcriptional control of HIF‐1α. Seahorse extracellular flux analyses in subsequent studies confirmed that NP cells have high glycolytic rates (extracellular acidification rate, ECAR) and relatively low oxygen consumption rates (OCR), consistent with a predominantly glycolytic metabolic phenotype [46, 47]. Notably, NP cells are unable to switch to alternative substrates such as glutamine or fatty acids under conditions of glucose restriction, highlighting that glucose is an indispensable metabolic substrate for these cells [47]. Using a genetic strategy to conditionally delete Glut1 in NP cells using Foxa2 Cre and Krt19 CreERT2 alleles and a specific GLUT1‐3 inhibitor, Glutor, we conclusively demonstrated redundancy among HIF‐1‐dependent GLUT transporters, with GLUT3 compensating for GLUT1 loss, further underscoring the necessity of glucose import for the homeostatic maintenance of NP cells [47]. HIF‐1α also regulates pyruvate dehydrogenase kinase 1 (PDK1) in NP cells, which phosphorylates and inactivates the pyruvate dehydrogenase complex, thereby reducing flux from glycolysis into the TCA cycle and mitochondrial oxidative phosphorylation [48]. This PDK1‐dependent mechanism allows NP cells to sustain high glycolytic flux while limiting mitochondrial reactive oxygen species (ROS) production, a critical adaptation for cells that must survive for decades in a low‐oxygen environment. A recent study confirmed that HIF‐1α preserves mitochondrial integrity under oxidative stress through this PDK‐1‐dependent mechanism, rescuing NP cell survival both in vitro and in vivo [48]. Regulation of Extracellular Matrix Beyond its metabolic functions, HIF‐1α plays a pivotal role in regulating the extracellular matrix that defines the NP phenotype [49]. Aggrecan, the large aggregating proteoglycan that gives the NP its gel‐like consistency and ability to retain water, is a HIF‐1α target in NP cells [35]. HIF‐1α promotes Acan and Col2A1 expression partly through NOTCH1 signaling, and loss of HIF‐1α in conditional knockout mice is accompanied by reduced aggrecan and collagen II in disc tissue [50]. Notably, our studies showed the importance of HIF‐1α in mitigating unfolded protein response (UPR) and ER‐stress response in NP cells, which diminish collagen and ECM secretion and adhesion‐related proteins [51]. It is important to note that several key enzymes involved in collagen secretion and stability, including PLOD2 (Procollagen‐lysine,2‐oxoglutarate 5‐dioxygenase 2), which hydroxylates lysine residues in the telopeptides of procollagen critical for cross‐linking [52] and collagen prolyl 4‐hydroxylases (C4‐PHs), which hydroxylate prolyl residues and govern secretion, are HIF‐1 targets [53, 54]. This HIF‐1α‐ECM axis directly links oxygen sensing to the structural integrity of the NP matrix, providing a molecular rationale for why disc cells so critically depend on constitutive HIF‐1α activity. Mitochondrial Regulation and BNIP3‐Mediated Mitophagy Despite their predominantly glycolytic metabolism, NP cells are not devoid of functional mitochondria. Our 2020 study in the Journal of Bone and Mineral Research, using the MitoQC reporter mice to track mitochondrial quality in living cells and in vivo, Vedavathi Madhu demonstrated that NP cells contain well‐networked, functional mitochondria whose homeostasis is actively regulated by tissue hypoxia, HIF‐1α, and BNIP3‐mediated mitophagy [55]. In contrast to the prevailing view that NP cells harbor few functional mitochondria, we found that NP cells maintain a dynamic mitochondrial network and that mitophagy, the selective autophagic degradation of damaged mitochondria, is a constitutive process in these cells that intensifies with age [55]. HIF‐1α drives expression of BNIP3, a mitochondrial outer membrane protein that functions as a mitophagy receptor by directly engaging LC3B on the autophagosome membrane [55]. Loss of HIF‐1α reduces BNIP3 expression, impairs mitophagy, and leads to accumulation of damaged mitochondria and elevated oxidative stress [55]. We went on to demonstrate that this HIF‐1α‐BNIP3 mitophagy axis is a critical mechanism for broader mitochondrial quality control and function in the disc [56]. These studies have prompted our ongoing investigations of how mitochondrial dynamics maintain the quality of the NP‐cell mitochondrial pool [57]. HIF‐2α: A Distinct Role in Matrix Homeostasis and Disc Degeneration Differential Regulation of HIF‐1α and HIF‐2α While HIF‐1α and HIF‐2α share the same structural organization, both are basic helix–loop–helix PAS domain proteins that heterodimerize with HIF‐1β/ARNT and bind HREs, they exhibit distinct target gene specificities and, as our work has shown, are regulated by distinct degradation mechanisms in NP cells [37, 38, 58]. In the canonical PHD/VHL pathway, both isoforms are degraded in an oxygen‐dependent manner. In NP cells, however, HIF‐2α degradation is even more independent of the PHD pathway than HIF‐1α: DMOG treatment, which modestly induces HIF‐1α accumulation and HIF‐1 target genes, had minimal effect on HIF‐2α levels or HIF‐2 target genes including Cited2 [37]. This differential regulation implies that HIF‐1α and HIF‐2α serve distinct and independently controlled functions in NP cell biology. Role of HIF‐2α in Disc To directly test the role of HIF‐2α in disc homeostasis, Shira Johnston generated NP‐specific HIF‐2α conditional knockout mice using the Foxa2 Cre driver and analyzed them longitudinally until 24 months of age [59]. In striking contrast to the deleterious phenotype of HIF‐1α knockout mice, loss of HIF‐2α was not acutely detrimental to NP cell survival or disc structure [55]. Rather, Hif2A Foxa2Cre mice showed a mild but significant reduction in NP tissue fibrosis and lower histological degeneration scores at 14 months [59]. Hif2A Foxa2Cre mice also exhibited decreased levels of pro‐inflammatory cytokines (IL‐6, MCP1) and pro‐fibrotic markers in the NP and AF at 14 months [59]. However, these protective effects were transient: by 24 months, knockout mice and control mice showed comparable degrees of degeneration, indicating that aging is the dominant driver of late‐stage disc pathology and that HIF‐2α loss provides only mild and time‐limited protection [59]. Complementary evidence for a pathogenic role of HIF‐2α in disc aging comes from a study demonstrating that genetic upregulation of HIF‐2α activity in NP cells (Hif2A Krt19CreERT2 ) produces age‐related disc degeneration in mice, directly implicating excessive HIF‐2 signaling as a driver of the degenerative process [60]. HIF‐2α is also elevated in human osteoarthritic cartilage and has been shown to upregulate catabolic enzymes including MMP13 and ADAMTS5 in chondrocytes, suggesting that pathological HIF‐2α activity contributes to matrix degradation in both disc and articular cartilage contexts [2, 61]. Together, these findings support a model in which HIF‐1α and HIF‐2α have nonredundant and different roles in disc homeostasis: HIF‐1α is essential for NP cell survival and metabolic health, while dysregulated HIF‐2α activity promotes matrix catabolism, fibrosis, and degeneration. Autophagy in the Nucleus Pulposus: An HIF‐Independent Pathway Non‐Canonical Autophagy in NP Cells One of the most surprising findings from our laboratory has been the discovery that autophagy in NP cells is regulated in a fundamentally noncanonical fashion. In most cell types, autophagy is controlled by the mechanistic target of rapamycin (mTOR) kinase complex, which suppresses autophagy under nutrient‐replete conditions and is inhibited by nutrient deprivation, energy stress, and hypoxia, thereby permitting autophagy induction. HIF‐1α has been reported to promote autophagy in various cell types through transcriptional upregulation of genes including BNIP3, BECN1, and ATG5. Our 2016 study in Autophagy by Hyowon Choi demonstrated that hypoxia promotes robust autophagosome formation in NP cells through a pathway that is independent of both mTOR suppression and HIF‐1α transcriptional activity [62]. Genetic deletion of HIF‐1α in NP cells did not impair hypoxia‐induced autophagy, and pharmacological inhibition of mTOR with rapamycin did not phenocopy the hypoxic induction of autophagy [62]. Instead, hypoxia‐induced autophagy in NP cells involved robust autophagosome biogenesis with only modest changes in autophagic flux, suggesting that the primary effect of hypoxia is to enhance autophagosome formation through an as‐yet‐unidentified mechanism operating independently of both HIF‐1α and mTOR [62]. BNIP3‐Mediated Mitophagy: An HIF‐Dependent Component While general autophagy in NP cells appears to be largely HIF‐independent, mitophagy, the selective autophagic removal of mitochondria, is regulated by HIF‐1α through the BNIP3 pathway, as described in Section 5.4 [55, 56]. This distinction between general autophagy and mitophagy in NP cells is important for understanding the overall landscape of autophagic regulation in the disc and for designing therapeutic strategies that target specific arms of the autophagy pathway. Further studies are required to determine the importance of the HIF‐1α–BNIP3 mitophagy axis in protecting NP cells in different contexts of disc degeneration. Autophagy and Disc Degeneration Changes in autophagy and mitophagy are increasingly recognized as important drivers of disc degeneration in both early‐onset and age‐dependent contexts. Our MitoQC reporter studies demonstrated that mitophagy increases with age in NP cells, potentially as a compensatory response to accumulating mitochondrial and organelle damage [55]. Notably, in SM/J mice, a model of early‐onset, spontaneous disc degeneration, Emanuel Novais noted that autophagy and mitochondria, ER homeostasis, and redox homeostasis were among the most prominent downregulated pathways in NP tissues, suggesting that decreased autophagic flux correlates with degeneration [63, 64]. Similarly, our transcriptomic analysis of degenerated human NP tissues showed the downregulation of mitochondrial homeostasis and protein ubiquitination, which directs damaged proteins for proteasomal and/or autophagic clearance [64]. Taken together, disruption of autophagic flux, whether through loss of function of upstream regulators such as HIF‐1α and BNIP3, or through downstream impairment of lysosomal function, leads to accumulation of damaged organelles, including mitochondria, increased oxidative stress, and NP cell senescence and/or death [55, 56, 65]. pH Homeostasis and the Metabolic Landscape of the Disc The Acidic Microenvironment of the NP The combination of high glycolytic activity, limited diffusion capacity, and avascularity creates an unusually acidic microenvironment in the NP. Extracellular pH in the disc center has been measured at approximately 6.8–7.0 in healthy tissue, declining to 6.0–6.5 in degenerated discs [2]. Intracellular pH in NP cells is similarly low, approximately 6.8 under basal conditions [66]. This acidic environment is not merely a byproduct of metabolism but an active regulatory signal: extracellular and intracellular pH influence NP cell gene expression, matrix synthesis, proteoglycan turnover, and cell viability [2]. HIF‐1α‐Dependent Carbonic Anhydrase Expression A critical mechanism for maintaining intracellular pH homeostasis in NP cells was identified in our 2018 study in the Journal of Bone and Mineral Research, by Elizabeth Silagi [46]. This work demonstrated that NP cells express high levels of carbonic anhydrase isoforms 9 (CA9) and 12 (CA12), membrane‐bound, extracellularly‐facing enzymes that catalyze the reversible hydration of CO2 to produce bicarbonate and protons, and that this expression is entirely dependent on HIF‐1α [46]. The mechanism operates as follows: metabolically active NP cells produce CO2 as a byproduct of the TCA cycle and other metabolic reactions. CA9 and CA12, expressed on the NP cell surface under the transcriptional control of HIF‐1α, catalyze the extracellular hydration of CO2 to bicarbonate (HCO₃−) and H+. The bicarbonate is subsequently transported back into the cell via bicarbonate transporters, where it serves to buffer intracellular acidity. Quantitative analysis showed that CO2 hydration by CA9 and CA12 accounts for approximately 66% of extracellular proton production in NP cells, making it the dominant mechanism of extracellular acidification [46]. Loss of HIF‐1α in conditional knockout mice caused a significant reduction in CA9 and CA12 expression in NP tissue, with corresponding impairment of intracellular pH regulation and increased NP cell death [46]. These findings established the HIF‐1α‐CA9/CA12 bicarbonate recycling axis as a critical survival mechanism in disc cells, one that couples oxygen sensing directly to acid–base homeostasis. MCT4 and Lactate Efflux An important pH homeostasis mechanism in NP cells was later characterized in our 2020 study published in the Journal of Bone and Mineral Research by Elizabeth Silagi [66]. This work focused on monocarboxylate transporter 4 (MCT4, encoded by Slc16a3), a proton‐coupled lactate efflux transporter that is a well‐established HIF‐1 target gene in hypoxic cells. We demonstrated that MCT4 is highly expressed in NP cells, regulated by a HIF‐1α‐dependent intronic enhancer element, and essential for maintaining glycolytic and TCA cycle flux and intracellular pH homeostasis in the NP [66]. MCT4 knockout mice developed progressive disc degeneration with age: by 14 months, all disc compartments in the caudal spine showed degenerative changes, including a honeycomb pattern of NP cells, loss of lamellar AF organization, and dramatically increased collagen 10 expression in the NP, a marker of hypertrophic chondrocytic differentiation indicative of pathological NP cell fate transition [66]. MCT4 KO mice also exhibited inferior vertebral trabecular bone quality, with reduced bone volume fraction, decreased trabecular number, and increased trabecular separation [66]. These findings established MCT4 as a critical regulator of disc metabolic health and vertebral bone quality, and identified lactate efflux as a key determinant of NP cell survival. An intriguing additional finding was that intracellular lactate can act as a “hypoxia mimetic” in NP cells, augmenting HIF‐1α transcriptional activity by inhibiting PHD function [66]. This creates a positive feedback loop between glycolytic activity, lactate accumulation, and HIF‐1α activation, a circuit that may help sustain HIF‐1α activity in NP cells under fluctuating oxygen tensions. Importantly, MCT4 expression was significantly reduced in moderately and severely degenerated human NP tissue, suggesting that impaired lactate efflux capacity contributes to disc metabolic deterioration during degeneration and that MCT4 could serve as a clinically relevant biomarker for disc pathology [66]. The MCT4‐MCT1 Lactate Coupling Axis: A Paradigm Shift For decades, the prevailing view in disc biology held that lactate secreted by glycolytic NP cells, the metabolic byproduct of the HIF‐1α‐driven metabolism described above, simply exits the disc into circulation via the endplates, representing a metabolic waste product that must be cleared to prevent excessive acidification. This conceptual framework positioned lactate as a passive consequence of the disc's avascular metabolism, a molecule to be disposed of rather than utilized. Our most recent work by Maria Tsingas, published in Cell Death & Disease in 2026, fundamentally challenges this long‐held notion and reveals a novel intercompartmental metabolic coupling mechanism: the MCT4‐MCT1 lactate shuttle [67]. The study began with a deceptively simple question: what happens to the lactate exported by NP cells via MCT4? To address this question, we generated mice with early postnatal, endplate‐ and annulus fibrosus‐specific conditional deletion of Slc16a1, the gene encoding monocarboxylate transporter 1 (MCT1), the primary lactate importer in most tissues. MCT1 is the functional counterpart to MCT4: while MCT4 exports lactate from glycolytic cells, MCT1 imports lactate into oxidative cells, where it can be metabolized. The phenotype of these Mct1 conditional knockout (Mct1 Col2a1CreERT2 /Mct1 cKO) mice was pronounced: they developed progressive disc degeneration characterized by NP cell loss, pronounced endplate structural changes, and matrix disorganization across all disc compartments [67]. Quantitative micro‐computed tomography (μCT) analysis revealed that Mct1 cKO mice exhibited significantly reduced endplate bone mineral density (BMD) compared to littermate controls [36]. Histological analysis showed degenerative changes visible as early as 10 weeks of age, revealing loss of proteoglycan‐rich NP matrix, increased NP fibrosis, and dramatic remodeling of annulus fibrosus collagen fiber architecture [67]. Immunohistochemical staining demonstrated a disorganized matrix with increased COL1 in the NP [67]. The phenotype was remarkably similar to age‐related disc degeneration, but it occurred in young mice, suggesting that loss of MCT1‐mediated lactate import in the endplate and AF was sufficient to trigger a degenerative cascade. Metabolic tracing and transcriptomic analyses helped explain this phenotype. Using dual [U 13C]‐glucose and 3‐13C lactate tracing, we demonstrated that lactate imported by endplate cells via MCT1 enters the tricarboxylic acid (TCA) cycle as a metabolic substrate. We observed lactate enrichment into citrate, aspartate, and malate pools, confirming that lactate‐derived carbon flows through the TCA cycle to support oxidative metabolism [67]. These findings established that endplate cells are not merely passive barriers through which lactate diffuses; they are metabolically active consumers of NP‐derived lactate, using it as a preferred TCA substrate to fuel oxidative phosphorylation. We also found that lactate serves a second, nonmetabolic function in endplate cells: it acts as an epigenetic regulator through protein and histone lactylation. Lactylation, the covalent modification of lysine residues by lactate‐derived lactyl groups, is an emerging posttranslational modification that links cellular metabolism directly to gene expression [62]. Immunofluorescence analysis revealed that MCT1 loss caused a dramatic reduction in pan‐lysine lactylation (pan‐Kla), H3K18la, and H4K5la in the endplate of Mct1 cKO mice compared to controls [62]. Importantly, MCT4 knockout mice, which cannot export lactate from the NP, also exhibited reduced pan‐Kla in AF and endplate tissues, confirming that lactylation in these compartments is lactate‐dependent and requires the MCT4 (export) → MCT1 (import) coupling axis [67]. This finding established lactate as a paracrine signaling metabolite that coordinates gene expression across disc compartments. To understand how lactate‐driven lactylation regulates endplate cell function, we performed spatial transcriptomics on Mct1 cKO mice and bulk RNA sequencing on primary endplate cells cultured with or without lactate supplementation. Lactate supplementation induced profound transcriptional reprogramming in endplate cells, upregulating genes critical for chondrocyte differentiation and endplate maturation, including Col2a1, Adamts4, Meox1, Fgfr1, and Inhba [67]. Bioinformatic analysis revealed that lactate suppresses fatty acid and pyrophosphate metabolism‐related pathways while promoting skeletal development, chemotaxis, and growth regulation programs [67]. UMAP clustering of spatial RNA‐seq data showed that MCT1 deletion alters endplate chondrocyte cell fate, shifting the transcriptomic landscape away from differentiated, matrix‐producing chondrocytes toward a more immature or dysregulated state [67]. These findings coalesce into a unified model of intercompartmental metabolic coupling in the disc. In healthy, growing discs, because of their HIF‐1α‐driven glycolytic metabolism, NP cells export lactate via MCT4. This lactate does not simply diffuse away; instead, it is actively imported primarily by endplate cells via MCT1. Once inside the endplate, chondrocyte, lactate serves dual roles: (1) as a TCA metabolite that fuels oxidative phosphorylation and supports the high energetic demands for matrix synthesis and remodeling, and (2) as a substrate for lactylation reactions that modify histones and other proteins, thereby regulating gene expression programs essential for chondrocyte differentiation and endplate maturation. Through this lactylation‐dependent transcriptional reprogramming, NP‐derived lactate promotes the differentiation of endplate cartilage into vascularized subchondral bone, a process critical for establishing the nutrient exchange pathways that sustain the entire disc. In Mct1 cKO mice, this coupling is severed: lactate cannot enter endplate or AF cells, lactylation is diminished, chondrocyte differentiation is impaired, endplate maturation is delayed, and the resulting limitation in nutrient exchange triggers secondary NP cell loss and disc‐wide degeneration [67]. These new findings change current and long‐held views concerning disc lactate metabolism. Lactate is not metabolic waste; it is a paracrine signaling metabolite that coordinates the metabolic and transcriptional programs of spatially distinct disc compartments. The MCT4‐MCT1 axis is not merely a lactate clearance pathway; it is a metabolic coupling mechanism that links the glycolytic activity of the avascular NP to the oxidative metabolism and growth‐dependent maturation of the endplate. This discovery adds a new dimension to the HIF‐1α story: HIF‐1α does not act in isolation within NP cells; its transcriptional program, which includes upregulation of glycolytic enzymes and MCT4, generates a metabolic output (lactate) that is sensed and utilized by neighboring tissues to maintain organ homeostasis. Accordingly, the disc, in this light, is not a collection of independent compartments but a metabolically integrated organ system in which the survival of one compartment (the NP) depends on the metabolic cooperation of its neighbors (the endplate and AF). Therapeutic Implications: Targeting HIF Pathways for Disc Regeneration The body of work described in this review has not only advanced our fundamental understanding of disc biology but has opened multiple therapeutic avenues for addressing disc degeneration, a condition affecting hundreds of millions of people worldwide and representing the leading cause of chronic low back pain and disability. Several strategies targeting HIF pathways have shown promise in preclinical models. Stabilizing HIF‐1α Activity Given the essential role of HIF‐1α in NP cell survival and metabolic health, strategies to restore or augment HIF‐1α activity in degenerated discs are an attractive possibility. PHD inhibitors, small molecules that block the oxygen‐sensing PHD enzymes and thereby stabilize HIF‐1α, have been explored in this context. While systemic PHD inhibition carries risks, including erythrocytosis and other off‐target effects, local drug delivery to the disc could restore HIF‐1α‐dependent survival and metabolic programs in degenerated NP cells. Several natural compounds have been shown to stabilize HIF‐1α and reduce NP cell degeneration in preclinical models, including nobiletin (which inhibits ER stress‐induced apoptosis via HIF‐1α upregulation) [68], quercetin (which stabilizes HIF‐1α protein under compressive loading and reduces compression‐induced cell death) [69], and ginsenoside‐Rh2 (which activates HIF‐1α‐dependent autophagy to suppress pyroptosis) [70]. Targeting HIF‐2α The finding that HIF‐2α loss provides mild protection from disc degeneration suggests that selective HIF‐2α inhibition could be beneficial in certain degenerative contexts [59, 60]. Selective HIF‐2α inhibitors (PT2977/belzutifan and related compounds) have been developed for the treatment of VHL‐associated cancers and could potentially be repurposed for disc applications, though the challenge of local delivery to the avascular disc and the transient nature of the protective effect observed in our conditional knockout studies suggest that this approach would require careful optimization. Targeting the MCT4‐MCT1 Lactate Coupling Axis The discovery that MCT4‐MCT1 lactate coupling in endplate cells is essential for endplate maturation and that dysregulation of this pathway triggers disc degeneration raises the possibility of intervention strategies in young patients exhibiting endplate pathology [66, 67]. There are several spinal conditions that affect the pediatric population and manifest as early‐onset disc degeneration and vertebral and endplate defects. These conditions include juvenile degenerative disc disease (JDDD), Scheuermann's Disease (Juvenile Kyphosis), and Pediatric Spondylolysis and Spondylolisthesis [71, 72, 73]. However, whether these pathologies are driven and exacerbated by underlying metabolic changes and alterations in lactate coupling, which is essential for endplate maturation, remains unknown. In that case, restoring or augmenting MCT1 expression in endplate cells could enhance lactate clearance. Second, targeting the lactylation machinery itself, for example, by delivering lactyl‐CoA precursors or modulating the enzymes that catalyze protein lactylation, could restore the epigenetic signals required for endplate maturation even in the absence of optimal lactate flux. These strategies are particularly attractive because they may address a possible cause of early‐onset disc degeneration, impaired nutrient exchange due to delayed endplate maturation, rather than merely treating downstream symptoms. Conclusions and Future Directions Looking back over two decades of research on HIF proteins in the intervertebral disc, I am struck by both what we have learned and what remains unresolved. What began as a puzzling Western blot has grown into a comprehensive understanding of how oxygen sensing, metabolic adaptation, pH homeostasis, mitochondrial quality control, and autophagy are integrated in one of the body's most extraordinary tissues. The key insights from our laboratory's journey may be summarized as follows (Figure 1). First, NP cells have fundamentally reprogrammed their oxygen‐sensing machinery, achieving a robust, constitutive HIF‐1α expression through partial oxygen‐independent proteasomal and lysosomal degradation pathways rather than the canonical PHD/VHL mechanism [1, 37]. Second, stabilized HIF‐1α expression drives a glycolytic metabolic program essential for NP cell survival in the hypoxic disc environment, as demonstrated by catastrophic cell death and compartment remodeling in conditional HIF‐1α knockout mice [44, 45]. Third, HIF‐1α maintains pH homeostasis in NP cells through the CA9/bicarbonate recycling axis and MCT4‐mediated lactate efflux: two mechanisms critical for cell viability in the acidic disc microenvironment [46, 66]. Fourth, HIF‐1α regulates mitochondrial quality control through BNIP3‐mediated mitophagy, ensuring that NP cells maintain functional mitochondria and sustain essential metabolite pools despite constitutive hypoxia, whereas the autophagic pathway is HIF‐independent (via noncanonical, mTOR‐independent pathways) [55, 56, 62]. Fifth, HIF‐2α plays a distinct and opposing role to HIF‐1 by promoting matrix catabolism, fibrosis, and age‐related degeneration [59, 60]. Sixth, and most recently, we have discovered that NP‐derived lactate exported via MCT4 is actively imported by endplate cells via MCT1, where it serves as both a TCA metabolite and an epigenetic regulator through lactylation. These findings reveal the disc as a metabolically integrated organ system in which intercompartmental lactate coupling is essential for endplate maturation and disc‐wide homeostasis [66, 67]. The HIF‐1α signaling axis in nucleus pulposus cells of the intervertebral disc. (A–D) The nucleus pulposus (NP) occupies the hypoxic center of the avascular intervertebral disc. In most tissues, HIF‐1α is rapidly hydroxylated at conserved proline residues (Pro402, Pro564) by prolyl hydroxylase domain (PHD) enzymes in an O2‐ and 2‐oxoglutarate‐dependent reaction and degraded by the 26S proteasome in von Hippel–Lindau (VHL) dependent fashion. In NP cells, this canonical oxygen‐dependent pathway is largely bypassed: PHD2 and FIH‐1 activity is functionally limited, and HIF‐1α is instead degraded through oxygen‐independent proteasomal and lysosomal pathways. Therefore, HIF‐1α protein is stabilized in NP cells even under normoxic culture conditions. Several other proteins including PHD3, HDACs and CCN2 also regulate HIF‐1 activity. HIF‐1α drives transcription of a broad program of adaptive target genes encompassing five major functional axes: Glycolysis, pH homeostasis and intercompartmental lactate coupling, Extracellular matrix regulation and mechanosensing, Mitophagy and mitochondrial quality control, and Cell survival. AF, annulus fibrosus; ARNT, aryl hydrocarbon receptor nuclear translocator; BNIP3, BCL2/adenovirus E1B 19 kDa protein‐interacting protein 3; CA9/CA12, carbonic anhydrases 9 and 12; EP, endplate; CCN2, cellular communication network factor 2; GLUT, glucose transporter; HDAC, histone deacetylase; HIF, hypoxia‐inducible factor; HRE, hypoxia response element; LDHA, lactate dehydrogenase A; MCT1, monocarboxylate transporter 1; MCT4, monocarboxylate transporter 4; NP, nucleus pulposus; PGK, phosphoglycerate kinase; PHD, prolyl hydroxylase domain protein; ROS, reactive oxygen species; SDC, syndecan; TCA, tricarboxylic acid; VHL, von Hippel–Lindau protein. Important questions remain open. We do not yet fully understand how the oxygen‐independent HIF degradation machinery in NP cells is established during notochordal development, nor the identity of the E3 ubiquitin ligases and deubiquitinases that control HIF‐1α turnover through non‐VHL pathways. The circadian regulation of HIF‐1α and its interaction with the disc's mechanical loading cycle are emerging areas of considerable interest. Also, unknown is the precise mechanism by which HIF‐2α promotes matrix catabolism and fibrosis in the aging disc, and whether degrative changes can be selectively targeted without disrupting HIF‐1α‐dependent survival functions. Finally, translating our preclinical findings in animal models into clinical therapies for disc degeneration remains a major challenge, requiring advances in local drug delivery, patient stratification, and the development of reliable biomarkers of disc metabolic health. The Humboldt Research Award has given me the opportunity to forge new collaborations with European colleagues working on related problems in musculoskeletal biology, regenerative medicine, and oxygen sensing. I am particularly excited by the possibility of combining our mechanistic insights into hypoxia and HIF biology with cutting‐edge single‐cell genomics and metabolomics, in vivo imaging, and biomaterial technologies to develop a deeper understanding of the disease process and disease‐modifying therapies for disc degeneration. I began this review with the memory of a puzzling Western blot. I end it reflecting on how far a single unexpected observation can lead, given the right scientific environment, supportive departmental leadership, the right collaborators, and the willingness to follow the data wherever it leads. The disc is a remarkable tissue, and its secrets have not yet been fully revealed. I look forward to the next chapter. Author Contributions Makarand V. Risbud: conceptualization, writing – original draft, writing – review and editing, funding acquisition. Funding This work was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (AR055655). National Institute on Aging (AG073349). Conflicts of Interest The author declares no conflicts of interest.