The NAD salvage pathway in mesenchymal cells is indispensable for skeletal development in mice

Warren, Aaron; Porter, Ryan M.; Reyes-Castro, Olivia; Ali, Md Mohsin; Marques-Carvalho, Adriana; Kim, Ha-Neui; Gatrell, Landon B.; Schipani, Ernestina; Nookaew, Intawat; O’Brien, Charles A.; Morello, Roy; Almeida, Maria

doi:10.1038/s41467-023-39392-7

Download PDF

Article
Open access
Published: 17 June 2023

The NAD salvage pathway in mesenchymal cells is indispensable for skeletal development in mice

Nature Communications volume 14, Article number: 3616 (2023) Cite this article

4250 Accesses
3 Altmetric
Metrics details

Subjects

Abstract

NAD is an essential co-factor for cellular energy metabolism and multiple other processes. Systemic NAD⁺ deficiency has been implicated in skeletal deformities during development in both humans and mice. NAD levels are maintained by multiple synthetic pathways but which ones are important in bone forming cells is unknown. Here, we generate mice with deletion of Nicotinamide Phosphoribosyltransferase (Nampt), a critical enzyme in the NAD salvage pathway, in all mesenchymal lineage cells of the limbs. At birth, Nampt^ΔPrx1 exhibit dramatic limb shortening due to death of growth plate chondrocytes. Administration of the NAD precursor nicotinamide riboside during pregnancy prevents the majority of in utero defects. Depletion of NAD post-birth also promotes chondrocyte death, preventing further endochondral ossification and joint development. In contrast, osteoblast formation still occurs in knockout mice, in line with distinctly different microenvironments and reliance on redox reactions between chondrocytes and osteoblasts. These findings define a critical role for cell-autonomous NAD homeostasis during endochondral bone formation.

Nox4 expression in osteo-progenitors controls bone development in mice during early life

Article Open access 14 June 2022

PP2A in LepR+ mesenchymal stem cells contributes to embryonic and postnatal endochondral ossification through Runx2 dephosphorylation

Article Open access 02 June 2021

Lipolysis supports bone formation by providing osteoblasts with endogenous fatty acid substrates to maintain bioenergetic status

Article Open access 24 November 2023

Introduction

Nicotinamide adenine dinucleotide (NAD) is a critical cofactor for cellular energy metabolism and numerous cellular processes, such as cell division, DNA damage repair, and mitochondrial function^1,2. In its oxidized state, NAD⁺ and its phosphorylated version NADP⁺ are able to accept two electrons, making this molecule essential for redox homeostasis. In this role, NAD⁺ is central for the catabolism of carbon sources in glycolysis, TCA cycle, β-oxidation, and glutaminolysis. Conversely, the redox pair NADP⁺/NADPH is primarily used for anabolic processes such as lipid and cholesterol synthesis. NAD⁺ also acts as a degradation substrate for numerous enzymes, such as sirtuins, PARPs, and cyclic ADP-ribose synthetases such as CD38 and CD157^3,4.

NAD is synthesized from dietary precursors such as tryptophan and vitamin B3 (niacin) via several pathways^5,6. Vitamin B3 is a collective term for nicotinic acid (NA), nicotinamide (NAM), and nicotinamide riboside (NR)³. The kynurenine pathway, also known as the de novo pathway, produces NAD from tryptophan. The Preiss–Handler and salvage pathways produce NAD from NA and NAM, respectively⁷. The existence of different pathways leading to NAD⁺ production raises questions regarding the relative importance of each pathway to various tissues. For instance, the kynurenine pathway of NAD synthesis seems to occur predominantly in the liver, as several enzymes of this pathway are not expressed in most other tissues⁸. Furthermore, plasma levels of most NAD⁺ precursors are low and, most likely, unable to systematically sustain high NAD⁺ production rates. The intracellular levels of NAD⁺ are determined by the cellular redox state (i.e., NAD⁺/NADH) and by the balance between NAD⁺ biosynthesis and degradation. The NAD-dependent sirtuins, PARPs, and cyclic ADP-ribose synthetases metabolize NAD⁺ into NAM. The NAM generated in these processes can be recycled to NAD via the salvage pathway⁹.

Multiple studies using genetic and pharmacological tools have elucidated that a decrease in NAD has a causative role in several diseases of aging, including osteoporosis^9,10,11. In addition, skeletal deformities occurring during development have been associated with NAD deficiency¹². The majority of the mammalian skeleton, including long bones in the limbs, forms through a process known as endochondral ossification¹³. During embryonic development, mesenchymal cells descend from the lateral plate mesoderm and form condensations within the limb bud. These cells differentiate into chondrocytes which form and expand an avascular structure known as the cartilage anlage¹⁴. Initially, chondrocytes proliferate at a high rate and produce an extracellular matrix rich in type II collagen and proteoglycans. Next, they exit the cell cycle and differentiate into hypertrophic cells that produce type X collagen. Vascular invasion into the hypertrophic zone allows for the formation of a bone marrow cavity called the primary ossification center and replacement of cartilage by bone¹⁵. The cartilage template continues to extend from both bone ends (epiphyses) due to proliferation and hypertrophy of chondrocytes comprising the growth plates. Indeed, these structures contain three distinct layers of resting, proliferating, and hypertrophic chondrocytes. After birth, a zone of hypertrophic chondrocytes then develops at the center of the epiphysis, followed by blood vessel invasion and deposition of bone matrix, initiating the formation of the secondary ossification center¹⁶. Alternatively, some bones, such as the flat bones of the skull, develop through mesenchymal intramembranous ossification, a process in which mesenchymal cells differentiate directly into osteoblasts.

Disruption of the normal sequence of cartilage development causes a multitude of skeletal dysplasias in humans¹⁷. Bi-allelic loss-of-function variants in enzymes of the kynurenine pathway cause congenital malformations in several organs, including the skeleton, as a consequence of NAD deficiency¹⁸. These malformation phenotypes have been classified as Congenital NAD Deficiency Disorder also called vertebral, cardiac, renal, and limb defects syndrome (Online Mendelian Inheritance in Man [OMIM] numbers 617660 and 617661, respectively). Likewise, restricted maternal intake of NAD precursors during pregnancy in mice leads to malformations in organs such as the heart, kidney, and skeleton¹⁹. Nonetheless, the systemic nature of these effects precludes any conclusions about direct effects of NAD in bone cells. To elucidate the contribution of NAD in specific bone cell populations to skeletal development, we generated mice with conditional deletion of Nampt – an essential enzyme in the NAD salvage pathway – in cells of the mesenchymal lineage. We found that the NAD salvage pathway in osteochondroprogenitors is indispensable for embryonic skeletal development as well as for postnatal skeletal growth and homeostasis.

Results

Loss of Nampt in mesenchymal progenitors impairs skeletal development

To examine the role of the NAD salvage pathway in skeletal development, we used Prx1-Cre transgenic mice to delete Nampt in all mesenchymal cells of the limb, including precursors of chondrocytes and osteoblasts (Nampt^ΔPrx1)²⁰. Prx1-Cre also targets precursors for the calvarium and the sterna but does not impact the axial skeleton. Cre recombinase activity starts early in development and is greater in forelimbs than hindlimbs²⁰. Nampt^ΔPrx1 mice were born at the expected Mendelian ratio but displayed severe shortening of the limbs at birth (Fig. 1a). All mice died a few days after birth for yet unknown reasons. Whole-mount skeletal staining at P3 confirmed that all limb skeletal elements were markedly shorter in Nampt^ΔPrx1 embryos compared to controls (Fig. 1b). The forelimb was more severely impacted, showing a very rudimentary stylopod (humerus) and zeugopod (radius and ulna) and missing the autopod (paw), most likely due to the higher penetrance of Prx1-Cre at this site (Fig. 1b). Likewise, in the hindlimb, autopod formation was absent (Fig. 1c). Immunostaining confirmed the loss of Nampt expression within the growth plate chondrocytes of P2 Nampt^ΔPrx1 hindlimbs, in contrast to surrounding muscle cells that are not targeted by Prx1-Cre (Fig. 1d). Similar to the limbs, Nampt deletion caused shortening of the sternum (Fig. 1e). In contrast, no obvious defects were seen in the craniofacial skeleton (Fig. 1f).

**Fig. 1: Nampt deletion in mesenchymal lineage cells severely compromises skeletal development.**

Loss of Nampt compromises cartilage homeostasis

We next performed histology of limbs and sternum to evaluate the cellular defects in the bones of Nampt^ΔPrx1 mice. H&E staining of femoral sections of littermate control mice at P2 revealed the typical zones present in epiphyseal cartilage at this age, namely the resting, proliferative and hypertrophic zones (Fig. 2a). The growth plate of Nampt^ΔPrx1 mice exhibited low chondrocyte number with an almost total absence of the resting zone. Islets of proliferating chondrocytes were present, as well as hypertrophic chondrocytes. Deletion of Nampt also caused major deficits in extracellular matrix, as shown by a faint, mainly pericellular Safranin O staining for proteoglycans that did not closely overlay with deposition of type II collagen. Very similar morphological findings were seen in the sternum (Supplementary Fig. 1). To examine whether the decrease in cellularity was due to cell death, we performed TUNEL staining. Rare TUNEL-positive cells were seen in the growth plate of control mice at this age (Fig. 2b). In contrast, Nampt^ΔPrx1 mice exhibited numerous positive cells. Histological analysis also revealed that synovial joint formation was disrupted in Nampt^ΔPrx1 mice. Specifically, the joints between the femur and tibia and tibia and fibula were fused (Fig. 2c).

**Fig. 2: Nampt deletion promotes growth plate chondrocyte apoptosis and disrupts joint formation.**

Administration of NR to pregnant mothers greatly attenuated the skeletal developmental defects caused by Nampt deletion

To confirm that the defects seen in Nampt^ΔPrx1 mice were due to a decrease in NAD, we administered the NAD precursor NR to mice during pregnancy. NR is transformed into nicotinamide mononucleotide (NMN), which is directly converted to NAD by the salvage pathway, bypassing the need for Nampt (Fig. 3a). A dose of about 400 mg/kg NR was not sufficient to alter the developmental defects of Nampt^ΔPrx1 mice (Supplementary Fig. 2). However, administration of 1000 mg/kg NR to the dams greatly attenuated the skeletal defects in Nampt^ΔPrx1 mice, as determined at P2. Specifically, the forelimbs were well formed and contained all elements including stylopod, zeugopod, and autopod (Fig. 3b). The lower limbs contained all digits (Fig. 3c). Immunohistochemistry confirmed the deletion of Nampt in chondrocytes of the hindlimbs (Fig. 3d) of Nampt^ΔPrx1 mice. As expected, Nampt expression remained unaffected in the adjacent muscle tissue. Safranin O staining for proteoglycans was highly similar between control and Nampt^ΔPrx1 rescued mice (Fig. 3e). H&E staining of the humerus revealed that the growth plate of Nampt^ΔPrx1 mice that received NR during gestation was practically indistinguishable from control mice (Supplementary Fig. 3). In addition, NR administration prevented joint malformations, as shown both in the hindlimbs and forelimbs (Fig. 3e and Supplementary Fig. 3). Nevertheless, both limbs (Fig. 3b, c) and sternum (Supplementary Fig. 4) of Nampt^ΔPrx1 rescued mice remained shorter than littermate controls and the menisci were not well developed (Fig. 3e). In addition, Nampt^ΔPrx1 mice exhibited some TUNEL-positive cells, particularly at the growth plate surfaces of the knees (Fig. 3f) and other synovial joints (Supplementary Fig. 5). In contrast, TUNEL positivity was only observed in the perichondrium/synovium of control (Nampt^f/+) limbs, as previously described²¹. Importantly, NR administered to dams is not transmitted to milk²² therefore, after birth the pups do not receive NR supplementation when suckling. Thus, the appearance of dead cells in Nampt^ΔPrx1 rescued mice at P2 is an early consequence of post-natal NAD depletion.

**Fig. 3: NR administration during pregnancy rescued most of the skeletal defects caused by Nampt deletion.**

The NAD salvage pathway is indispensable for epiphyseal bone formation during post-natal development

The ability of NR to rescue most of the pre-natal developmental defects allowed mice to ambulate and survive post-birth. This provided the opportunity to examine the dependency of the growth plate on Nampt during the phase in which epiphyseal bone development begins. To this end, pups from which NR was removed at birth were evaluated at P7 and P28. At P7, a secondary hypertrophic zone was present within the epiphysis in control mice, prefiguring the secondary ossification center (Fig. 4a). No such hypertrophy was present in the Nampt^ΔPrx1 mice. Instead, numerous empty chondrocyte lacunae were seen throughout the epiphysis. TUNEL staining revealed extensive chondrocyte death (Fig. 4b). In addition, the articular cartilage and menisci appeared hypocellular in Nampt^ΔPrx1 mice (Fig. 4c). Furthermore, mRNA levels of chondrocyte-specific genes, such as Acan, Col2a1, and Col10a1, as well as Mmp13 and Runx2, were greatly decreased in cartilage from Nampt^ΔPrx1 mice (Fig. 4d). In contrast, expression of Vegfa was not affected.

**Fig. 4: The NAD salvage pathway in mesenchymal cells is indispensable for formation of secondary ossification center and joint development post-birth.**

At P28, control mice exhibited a completely developed and mineralized secondary ossification center, as shown using µCT scans (Fig. 5a). Despite administration of NR during gestation, deletion of Nampt completely prevented the post-natal development of the epiphysis and bone growth in the femur and all other bones of the limbs. Histology of the tibia and proximal and distal tibia joints revealed that the epiphysis remained cartilaginous in 28 day-old Nampt^ΔPrx1 mice (Fig. 5b). Yet, a clear bone collar with marrow elements formed within the tibial diaphysis from the primary ossification center during gestation. Because Prx1-Cre targets all cells of the osteoblast lineage, we also examined osteoblasts and bone formation at the endocortical surface of long bones in 28-day-old mice. At this age of rapid growth, osteoblast covered practically all endocortical surfaces in control mice (Fig. 5c). As expected, double labels were abundant on these surfaces (Fig. 5d). Surprisingly, endocortical surfaces of Nampt^ΔPrx1 mice were also covered with active osteoblasts and double labels in a manner indistinguishable from the controls. Quantification of osteoblast number and bone formation rate revealed no differences between the two genotypes (Fig. 5c, d). However, the major differences in size and shape of the bones make it difficult to accurately determine possible consequences of Nampt deletion in osteoblastic cells.

**Fig. 5: The NAD salvage pathway in mesenchymal cells is required for post-natal endochondral, but not intramembranous, bone formation.**

To determine whether osteoblasts depend on Nampt during bone development, we generated Nampt^ΔOsx1 mice. Osx1-Cre targets all osteoblast precursors, osteoblasts, and osteocytes as well as a small number of proliferating and hypertrophic chondrocytes²³. Nampt^ΔOsx1 mice were grossly undistinguishable from Osx1-Cre control littermates at birth and up to 4 weeks-of-age (Fig. 5e), in stark contrast with the findings obtained with Nampt^ΔPrx1 mice (Fig. 1). Evaluations of body weight (Fig. 5f), as well as total and femur dual X-ray absorptiometry (DXA) bone mineral density (BMD) at 4 weeks (Fig. 5g) revealed no differences between mice of the two genotypes.

Chondrocytes exhibit high expression of metabolic genes that require NADP⁺/NADPH and NAD⁺/NADH redox reactions

A potential explanation for the different reliance on the NAD salvage pathway between chondrocytes and osteoblasts is that chondrocytes reside in an avascular microenvironment, making them hypoxic and dependent on nutrient diffusion. To explore the idea that reduced access to circulating nutrients compared to other cell types might underlie the reliance on the NAD salvage pathway, we examined metabolic gene expression patterns using publicly available single cell RNA sequencing datasets from 2 publications in which the authors used mesenchymal cell-enriched samples from femurs of young adult mice^24,25 (Fig. 6a). Combined analysis of the non-hematopoietic cells in this datasets uncovered 10 cell types including growth plate chondrocytes, articular chondrocytes, and osteoblasts characterized by high expression of specific genes such as Col2a1, Prg4, and Bglap, respectively (Supplementary Fig. 6). Based on the differences in transcription of genes related to metabolism between chondrocytes and the other cell types, we identified metabolic hot spots using reporter metabolite analysis²⁶. This analysis revealed that growth plate chondrocyte metabolism was particularly enriched in processes that involve redox reactions dependent on NAD⁺/NADH and NADP⁺/NADPH when compared to osteoblasts or the other mesenchymal cell populations present in bone (Fig. 6b). One of the metabolic processes that is highly dependent on both NAD⁺/NADH and NADP⁺/NADPH redox reactions is the synthesis of glycosaminoglycans (GAGs) – the most abundant matrix components of chondrocytes²⁷. As anticipated, chondrocytes were enriched in metabolic hotspots for chondroitin sulfate, a major component of their matrix. NADPH is also required for anabolic processes such as lipids and cholesterol synthesis. Lipogenesis was upregulated in growth plate chondrocytes as indicated by the numerous hotspots for acyl-CoenzymeA (CoA) derivatives of long chain polyunsaturated fatty acids (PUFAs). This chemical modification is required for otherwise non-reactive fatty acids to participate in biosynthetic or catabolic pathways. Indeed, through a series of desaturation and elongation reactions PUFA-CoA can be converted into biologically active products^28,29. Multiple genes responsible for lipid synthesis and processing were upregulated in chondrocytes, including fatty acid synthetase (Fasn) (Fig. 6c), which uses NADPH to convert acetyl-CoA into palmitate and is the only enzyme in mammals capable of converting metabolized sugar into a fatty acid. The delta-5 and delta-6 desaturases (D5D and D6D), encoded by Fads1 and Fads2, respectively, were also upregulated in chondrocytes. These enzymes are responsible for highly unsaturated fatty acid synthesis and regenerate NAD⁺ from NADH³⁰. Furthermore, D5D/D6D provide a mechanism for glycolytic NAD⁺ recycling that permits ongoing glycolysis and cell viability when the cytosolic NAD⁺/NADH ratio is reduced, analogous to lactate production³¹. Cholesterol synthesis in the endoplasmic reticulum utilizes both NADH and NADPH as the electron source at multiple steps³². Multiple enzymes responsible for cholesterol uptake and synthesis were also upregulated in chondrocytes (Fig. 6c).

**Fig. 6: Chondrocytes have enhanced redox metabolism.**

As previously shown³³, chondrocytes exhibited up-regulation of genes involved in glycolysis (Fig. 6c), the major energy pathway in these cells. During glycolysis, high energy hydride ions are captured from glucose by enzymes that utilize NAD⁺ as an electron acceptor forming NADH. In hypoxia, conversion of pyruvate to lactate by lactate dehydrogenase (Ldha) also converts NADH to NAD⁺, thereby restoring NAD⁺ levels in the cytosol. Overall, the metabolic differences between chondrocytes and other bone cells, highlighted in this analysis, are in line with previously described metabolic adaptations to hypoxia³⁴. Indeed, besides the changes in metabolism and ATP production, cells also adjust to hypoxia by altering the biosynthesis of macromolecules such as lipids.

Hypoxia does not render chondrocytes more sensitive to Nampt inhibition

The metabolic adaptation of chondrocytes to hypoxia, namely upregulation of glycolysis, lipid, and cholesterol synthesis³⁰ might render them more susceptible to the loss of Nampt. To examine this possibility, we used the pharmacological Nampt inhibitor FK866 (1 μM) in primary chondrocytes cultured under normoxia (20% O₂) or hypoxia (1% O₂). Chondrocytes were left untreated for 2 d to adjust to hypoxia and treated with FK866 for an additional 1 or 2 d. After a total of 3 and 4 d in hypoxia, vehicle-treated cells had lower levels of NAD than cells in normoxia (Fig. 7a, b), most likely reflecting a decrease in mitochondria respiration. As previously shown³⁵, cells in hypoxia had higher levels of NADH/NAD ratio (Fig. 7c). The levels of NAD⁺ were practically undetectable in FK866-treated cells in either normoxia or hypoxia conditions (Fig. 7a, b). We also examined the effect of NAD⁺ depletion on ATP production after treatment with FK866. Vehicle-treated chondrocytes cultured in hypoxia for 3 d had decreased ATP levels (Fig. 7d). However, after 4 d in hypoxia the levels of ATP in hypoxic cells were similar to the ones seen in cells cultured in normoxia (Fig. 7e). These changes in ATP most likely reflect the metabolic adaptation of cells to hypoxia, i.e., decreased mitochondrial- and increased glycolysis-dependent energy production. Nampt inhibition also caused ATP depletion, but this effect was seen only after 2 d of FK866 treatment and was similar between cells in normoxia and in hypoxia (Fig. 7d, e). Measurements of lactate secreted to the medium confirmed that cells in hypoxia were using glycolysis to produce energy (Fig. 7f, g), and that FK866 decreased the levels of lactate after 2 d but not 1 d of treatment. The similarities between NAD⁺ and ATP depletion in normoxia and hypoxia do not support the idea that hypoxia alone is responsible for the high sensitivity of growth plate chondrocytes to Nampt inhibition. Most likely, the low nutrient supply is a major contributor to the dependence of chondrocytes on the NAD salvage pathway.

**Fig. 7: Nampt inhibition has a similar effect on ATP production in chondrocytes cultured in normoxia versus hypoxia.**

Deletion of Nampt significantly impacts the transcriptomic profile of growth plate chondrocytes

Next, we performed scRNA-seq of cells isolated from the growth plate of P2 mice born from dams that were fed NR during gestation. After quality control, we profiled 13,965 high-quality cells (6788 and 7177 cells derived from control and Nampt^ΔPrx1, respectively) with an average of over 47,000 unique molecular identifiers (UMIs) per cell and an average of approximately 3000 genes per cell. An initial UMAP plot and cluster analysis, after harmonization of cells from mice of the two genotypes, revealed 11 clusters (Supplementary Fig. 7) according to the expression of cluster-specific markers (Supplementary Fig. 8). Among the cells analyzed, 6693 cells were chondrocytes, 4449 mesenchymal cells (Fibro-1, Fibro-2, and Fibro-3), 1303 were tenocytes, and 1520 were mural cells (Pericytes I, Pericytes II, and endothelial cells). Because each of the pericyte and endothelial cell clusters contained a small number of cells, we excluded these and considered the 8 major clusters for further analysis (Fig. 8a). Chondrocyte subpopulations from both control and Nampt^ΔPrx1 mice included chondrocytes I, chondrocytes II, articular chondrocytes, and proliferating chondrocytes. These clusters all expressed markers such as Col2a1, Acan and Sox9, while Prg4 was expressed in articular chondrocytes (Fig. 8b). Other clusters included mesenchymal precursor cells designated fibro-1, fibro-2 [identified based on previous studies^24,25], as well as fibro-3 and tenocytes (including ligamentocytes) (Fig. 8a). Specifically, fibro-1 expressed common stem cell markers Ly6a, Ecm1, and Tspo; fibro-2 expressed Lpl, Cebpd, and Cxcl12; fibro-3 expressed Tnn, Col8a1, and Nrp2; tenocytes expressed Tnmd, Thbs4, and Kera (Fig. 8c). Deletion of Nampt led to a drastic decrease in the number of articular chondrocytes as well as Prg4-expressing fibro-2; for non-Prg4-expressing chondrocytes, there was a shift from cluster I to cluster II (Fig. 8d, e). Interestingly, the percentages of proliferative chondrocytes and other mesenchymal cells were similar in control versus Nampt^ΔPrx1 mice (Fig. 8e). Deletion of Nampt also increased the relative numbers of fibro-1 and tenocytes.

**Fig. 8: Nampt deletion greatly impacts gene expression in chondrocytes.**

Deletion of Nampt suppresses multiple metabolic pathways in chondrocytes

Because a major consequence of Nampt deletion was the shift in cell number between chondrocytes I and II, we focused our analysis on the comparison between these two clusters. Differential gene expression analysis (adjusted p < 10^e-10 and log₂ average fold change <0.5) identified 50 upregulated genes and 3642 downregulated genes when comparing chondrocytes II to chondrocytes I (Fig. 9a and Supplementary Data 1). This strongly indicated the abnormal transcriptional program of chondrocytes from Nampt^ΔPrx1 mice. Specifically, the expression of many genes encoding collagens (Col9a3, Col11a1, Col9a1, Col9a2) and other extracellular matrix (ECM) proteins (Epyc, Dcn, Snorc, Ucma) were suppressed in chondrocytes II (Fig. 9a). In contrast, the expression of genes involved in stress responses such as Hmga2, Spry2, Ptgs2 (Cox2), and Neat1 as well as genes previously associated with osteoarthritic chondrocytes such as Postn, Tnfai6, Tmsb4x were upregulated in chondrocytes II (Fig. 9a and Supplementary Fig. 9). RNA in situ hybridization confirmed the elevated expression of Ptgs2 and Pim3 in the growth plates from Nampt^ΔPrx1 hindlimbs (Fig. 9b and Supplementary Fig. 10). Interestingly, the expression of Ptgs2 was localized to the cells on the bone surfaces and adjacent synovium, perichondrium, and menisci. In control mice, Ptgs2 was mainly restricted to the proliferative and hypertrophic zone (Supplementary Fig. 10). The localization of cell expressing Ptgs2 was reminiscent of the localization of TUNEL positive cells in Nampt^ΔPrx1 mice of the same experimental design (Fig. 3). Pim3, which was upregulated in a subpopulation of chondrocytes II distinct from those expressing Ptgs2 (Supplementary Fig. 9), was elevated in multiple growth plate zones (Fig. 9b and Supplementary Fig. 10).

**Fig. 9: Nampt deletion severely impacts cell metabolism.**

Gene Ontology (GO) enrichment analysis of differentially expressed genes revealed that cellular components and biological processes related to mitochondria activity, translation, protein secretion, and carbohydrate metabolic processes were downregulated in chondrocytes II. On the other hand, biological processes such as angiogenesis, cell adhesion, and cell death were upregulated in chondrocytes II (Fig. 9c). Focusing on cellular metabolism, reporter metabolite analysis revealed that metabolic hotspots for amino acids such as serine, proline, glycine and asparagine were reduced in chondrocytes II (Fig. 9d). Proline and glycine constitute the backbone of collagen molecules, therefore a reduction in metabolism of these amino acids in chondrocytes II is in agreement with the decreased expression of the major collagen components of the ECM. Furthermore, processes that involve redox reactions dependent on NAD⁺/NADH were decreased in chondrocytes II, in line with the increase in this abnormal chondrocyte population in Nampt deficient mice (Fig. 9d). The expression of key enzymes implicated in carbohydrate metabolism such as Ldha, Phgdh, and Ugdh which mediate anaerobic glycolysis, serine synthesis, and biosynthesis of GAGs, respectively, was suppressed in chondrocytes II (Fig. 9a and Supplementary Data 1). Reporter metabolite analysis also revealed that ubiquinone/ubiquinol metabolism was reduced in chondrocytes II (Fig. 9d). Ubiquinone is necessary for mitochondrial energy production, is a prominent source of ROS, and has antioxidant properties in its reduced form (ubiquinol)³⁶.

Chondrocytes lacking Nampt exhibit metabolic defects

To examine the functional relevance of some of the key findings of scRNA-seq analyses, we cultured cells from the growth plates of control and Nampt^ΔPrx1 mice. As expected, cells lacking Nampt had decreased levels of NAD⁺ (Supplementary Fig. 11a). We next examined the rate of glycolysis, as this is the major energy source in chondrocytes. As indicated by the extracellular acidification rate (ECAR) measured using SeaHorse, cells from Nampt^ΔPrx1 mice exhibited decreased glycolysis (Supplementary Fig. 11b). Accordingly, the levels of ATP were lower in cells lacking Nampt (Supplementary Fig. 11c). We also found that the levels of ROS were increased in these cells (Supplementary Fig. 11d), perhaps due to the altered ubiquinone/ubiquinol and other metabolic processes. Because NAD⁺ is a critical cofactor for proteins such as sirtuins and PARPs, we examined by Western blot the levels of lysine acetylation and the levels of PAR, indicators of Sirtuin and PARP activity, respectively. A modest decrease in PAR and increase in lysine acetylation were seen in cells from Nampt^ΔPrx1 mice (Supplementary Fig. 11e). We also examined whether expression of the multiple sirtuins or Parps was altered in chondrocytes II versus chondrocytes I. Of the 7 sirtuin family members, expression of Sirt1 and Sirt2 was slightly decreased in chondrocytes II (Supplementary Data 1). A small decrease in Sirt1 was confirmed at the protein level in cultured chondrocytes from Nampt^ΔPrx1 mice (Supplementary Fig. 11e). Of the 17 Parp family members, expression of Parp2 and Parp8 was decreased in chondrocytes II (Supplementary Data 1). Some of the Parp family members, including Parp2 have been implicated in DNA damage repair³⁷. Excessive ROS and decreased Parp activity can lead to accumulation of DNA damage. Therefore, we examined the protein levels of γ-H2AX, an indicator of DNA damage, but found no differences between cells of control and Nampt^ΔPrx1 mice (Supplementary Fig. 11e). Nonetheless, because the culture conditions do not replicate the limited nutrient supply experienced by cells in the growth plate, it is very likely that the changes observed in cultured cells are attenuated relative to the in vivo condition. Together with the scRNA-seq analyses, these findings indicate that Nampt deletion led to numerous metabolic derangements in chondrocytes that caused cellular stress and rapid cell death.

Discussion

NAD is indispensable for life and deficiencies in NAD cause diseases such as pellagra and acute kidney insufficiency, as well as congenital defects in multiple organs including the heart, vertebrae, kidneys, and limbs¹². Despite the requirement of NAD for multiple essential processes, the contribution of the recycling pathway to maintain cellular NAD varies among the different cell types. Here, we show that the NAD salvage pathway is indispensable for endochondral bone formation during embryonic development and post-natal bone growth. Although NAD can be synthesized de novo from tryptophan, it has been previously shown that mouse embryos lacking Haao or Kynu, enzymes in the de novo NAD pathway, do not exhibit developmental defects unless the dams are placed on a niacin (NA, NAM, and NR)-deficient diet¹². This is in line with the assumption that most of NAD derives from pathways which require the uptake of other NAD precursors from the diet. During development, embryos receive niacin from the dam and produce their own. We show that the skeletal defects caused by Nampt deletion occur when mice are fed a normal diet. While NA is an abundant NAD precursor in circulation, it is unknown whether chondrocytes can convert NA into NAD via the Preiss–Handler pathway. In any case, our findings demonstrate that the cell intrinsic conversion of NAM into NMN in the salvage pathway is indispensable for proper endochondral bone development in mice and that the de novo and Preiss–Handler pathways of NAD synthesis, if present, are not sufficient to provide the NAD required for this process.

We found that NR supplementation during pregnancy prevented the development of a very severe phenotype caused by genetic Nampt deletion in the embryos. NR is a dietary precursor that can be used by the salvage pathway to produce NAD. Intracellularly, NR is converted to NMN by NRK1 and NRK2. NMN, in turn, is converted to NAD⁺ in the last step of the salvage pathway, bypassing the need for Nampt. Previous studies have highlighted the importance of an adequate maternal intake of NAD precursors to avoid defects in multiple tissues namely the heart, kidney, and skeleton¹⁹. Our current findings elucidate that the critical roles of NAD during skeletal development are exerted directly and not secondary to the dysfunction of other tissues. Our findings also raise the exciting possibility that an elevated intake of NR during pregnancy might mitigate unforeseen developmental defects arising from genetic deficiencies in NAD synthetic pathways¹⁸. Although we did not administer NR to Nampt^ΔPrx1 mice after birth, this most likely would have been effective in preventing the post-birth defects that occur in these mice. This premise is further supported by evidence that NR administration post-birth alleviates defects caused by specific deletion of Nampt in muscle or liver^38,39. NMN administration also alleviates defects caused by deletion of Nampt in multiple tissues^9,40.

The resting zone of the growth plate contains mesenchymal precursors which differentiate into chondrocytes that proliferate and hypertrophy^41,42. Our findings indicate that the survival of cells present in the resting zone of the growth plate was severely compromised in the absence of Nampt. Likewise, the survival of chondrocytes in the growth plate after birth was compromised by Nampt deletion. Remarkably, chondrocytes forming the articular surfaces of hindlimb joints were especially sensitive to Nampt deletion, displaying TUNEL positivity and upregulation of the inflammatory mediator Ptgs2 (Cox-2) within days of NR withdrawal. These same markers characterize articular cartilage degeneration during osteoarthritis⁴³, suggesting a possible role for the NAD salvage pathway in articular chondrocyte homeostasis.

Hypoxia-Inducible Factor (Hif) 1α promotes chondrocyte survival in the hypoxic environment by impairing mitochondrial respiration^35,44. Indeed, conditional ablation of Hif-1α in growth plate chondrocytes and the limb bud mesenchyme results in defective skeletal development associated with massive cell death in the inner chondrocyte layer of the growth plate⁴⁴. We examined in vitro whether hypoxic cells could be more sensitive to a decrease in the NAD salvage pathway but found no evidence that this was the case. It is possible that inhibition of glycolysis, the major source of energy in chondrocytes and a process dependent on the NAD⁺/NADH redox pair, is responsible for the effects caused by Nampt deletion. In line with the dependence on glucose metabolism, deletion of Glut1, the major glucose transporter in chondrocyte, causes shortening of the limbs⁴⁵. However, in Glut1-deficient mice no chondrocyte death was detected in the growth plate; rather, defects were attributed to decreased chondrocyte proliferation and hypertrophy. Redox reactions are also essential to fuel lipid and cholesterol synthesis. We found that the expression of multiple enzymes required for these metabolic processes, including the ones that use NADH and NADPH as cofactors, are elevated in chondrocytes when compared to all other mesenchymal cells in bone. In addition, deletion of Nampt decreased the expression of genes implicated in lipid metabolism. Genetic disorders of cholesterol synthesis are characterized by multiple congenital abnormalities, including significant skeletal defects⁴⁶. Studies in rodents have further elucidated that intracellular cholesterol and lipid biosynthesis in chondrocytes are crucial for normal endochondral ossification, joint formation, and bone growth^47,48. Similar to our findings with Nampt deletion, disruption of cholesterol synthesis in chondrocytes causes cell death in the developing growth plate⁴⁸. Together, this evidence supports the idea that alterations in cholesterol biosynthesis could contribute to the defects caused by Nampt deletion in the growth plate. In line with this idea are the findings that altered lipid homeostasis plays a critical role in mediating cell death caused by Nampt inhibition in acute myeloid leukemia stem cells⁴⁹.

In contrast to the severe impact on bones formed by endochondral ossification, bones formed by intramembranous ossification, such as bones in the skull, were not affected by Nampt deletion. In addition, endocortical osteoblasts and bone formation were not overtly impacted in the absence of Nampt during early development. However, skull defects along with other skeletal malformations in embryos have been reported when wild-type pregnant mice are placed on diets with restricted NAD precursors¹⁹. In developing mouse limbs and cranial bone, the appearance of osteoblasts coincides with the invasion of blood vessels near sites of initiation of the primary ossification center¹³. These observations suggest that the supply of NAD precursors from the circulation or from neighboring cells is sufficient to sustain normal osteoblast differentiation during intramembranous ossification, and that the dependency on the NAD salvage pathway is not critical as in the case of endochondral ossification. Our findings that Nampt ablation using Osx1-Cre did not cause any overt bone defects further demonstrate the distinct reliance on the NAD salvage pathway of chondrocytes versus osteoblasts, two cell types that originate from a common mesenchymal precursor and are indispensable for bone development. We have previously shown that mice lacking one allele of Nampt in Prx1-Cre targeted cells have a normal skeleton up to 2 month-of-age¹¹. Yet, these mice exhibit lower bone mass at 8 months when compared to littermate controls, indicating that the NAD salvage pathway in mesenchymal lineage cells also contributes to bone formation. Future studies are warranted do elucidate the role of the NAD salvage pathway in distinct bone populations.

Our studies indicate that the dependency of growth plate chondrocytes on the NAD salvage pathway is due to the limited supply of nutrients. In addition, the comparison of the metabolic hotspots among different bone cell types revealed that chondrocytes are particularly enriched in metabolic processes dependent on NAD. The unbiased gene expression analysis using scRNA-seq elucidated the extent and severity of changes that occur in growth plate chondrocytes deficient in Nampt. Namely, we observed an overall decrease in gene transcription, and in particular, in genes involved in mitochondria metabolism, glycolysis, and lipid metabolism. These results are consistent with studies performed using bulk RNA-seq in tissues such as skeletal muscle, liver, and projection neurons, which show that Nampt deletion results in an overall downregulation of genes involved in metabolic processes^39,50,51. In contrast, our analysis indicated that Nampt deletion in chondrocytes led to a severe reduction in translation and protein secretion, processes that were not impacted in other tissues^39,50,51.

In multiple tissues the detrimental effects of Nampt deletion have been attributed to inhibition of Sirt1 activity¹⁰. However, mice with deletion of Sirt1 in mesenchymal cells targeted by Prx1-Cre or Col2a1-Cre develop normally^52,53, indicating that Sirt1 in chondrocytes has no major functions during skeletal development. Likewise, mice with individual deletion of Sirt3, Sirt6, Parp1, or Parp2 do not exhibit developmental limb defects^54,55,56,57. While our studies in chondrocytes cultured from Nampt^ΔPrx1 mice did not reveal major changes in markers of Sirtuin and PARP activity, it remains possible that the combined inhibition of all Sirtuins and PARPs contributes to the cell death seen in Nampt^ΔPrx1 mice. Because NAD⁺ is required as both a substrate and cofactor for numerous metabolic enzymes, a range of metabolic disturbances occurring simultaneously likely contributes to chondrocyte death in Nampt^ΔPrx1 mice.

A plethora of evidence gathered during the last decade in lower organisms and rodents have shown that restoration of NAD levels using metabolic precursors leads to multiple benefits to health in old age⁹. These have raised expectations about the use of NAD-based therapeutics in combating diseases of aging, underlined by the multiple clinical trials to test the efficacy of these drugs in humans^{58,59,60,61,62}. Our studies demonstrate that administration of NR during pregnancy in mice can prevent severe developmental defects caused by genetic inhibition of NAD production. In view of the multiple clinical syndromes already linked to defective NAD production, our findings raise hope of their attenuation and perhaps prevention.

Methods

Animal experimentation

Animal studies were performed as pre-approved by the University of Arkansas for Medical Sciences (UAMS) Institutional Animal Care and Use Committee. Mice with conditional deletion of Nampt in the mesenchymal lineage were generated by a two-step breeding strategy. Male hemizygous Prx1-cre transgenic mice (B6.Cg-Tg(Prrx1-cre)1Cjt/J; Jackson Laboratories, stock # 5584) were crossed with Nampt floxed (^f/f) mice (C57BL/6 genetic background) (provided by Shin-ichiro Imai, Washington University School of Medicine) to generate mice hemizygous for the Nampt floxed allele with and without the Cre allele, Nampt^f/+;∆Prx1 and Nampt^f/+, respectively. Male Nampt^f/+;∆Prx1 and female Nampt^f/+ mice were intercrossed to generate Nampt^f/f and Nampt^ΔPrx1 mice. Mice with conditional deletion of Nampt in the osteoblast lineage were generated using a similar strategy. Specifically, hemizygous Osx1-Cre transgenic mice⁶³ (B6.Cg-Tg (Sp7-tTA,tetO-EGFP/cre) 1Amc/J; Jackson Laboratories, stock # 6361) were crossed with Nampt^f/f mice, to generate mice heterozygous for the Nampt floxed allele with and without the Cre allele. These mice were intercrossed to generate the experimental Osx1-Cre (control) and Nampt^ΔOsx1 mice. Offspring were genotyped by PCR using the following primer sequences: Nampt-flox primer #1 5’TTC CAG GCT ATT CTG TTC CAG 3’ and primer #2 5’ TCT GGC TCT GTG TAC TGC TGA 3’. Offspring from all genotypes were tail-clipped for DNA extraction at the time of sacrifice or at weaning (21 days) and then group-housed with same sex littermates. To allow for quantification of bone formation rates, mice were injected with calcein (35 mg/kg body weight) 7 and 3 days before euthanasia at 1 month of age. In the NR rescue experiments, mice were treated with 1000 mg/kg NR (Tru Niagen, ChromaDex) administered ad libitum in the drinking water in light-protected bottles. The NR water solution was filtered and replaced daily during conception and pregnancy. Rescued Nampt^ΔPrx1 pups and control littermates were monitored daily from birth until the experimental endpoint at P2, P7, or P28. Male and female mice were used. For all experiments, mice were maintained with a constant temperature of 23 °C, a 12-h light/dark cycle, and had access to chow (LabDiet #5V5R) and water ad libitum. Body weight measurements were performed before euthanasia by either decapitation (neonates up to 10 days of age) or CO₂ inhalation with cervical dislocation (mice beyond 10 days old).

Whole-mount skeletal staining

Both male and female newborn mice were used without distinction in all analyses. After removing the skin and internal organs, mice were fixed overnight in 95% ethanol and then stained overnight with 0.015% Alcian blue (Sigma Aldrich, A3157), 20% Glacial acetic acid, and 80% ethanol. Next, specimen were transferred to 95% Ethanol for 3 h, and cleaned up with 2% KOH for 24 h. The skeletons were stained overnight with a solution containing 0.005% Alizarin Red (Sigma Aldrich, A5533), in 1% KOH. The stained skeletons were cleared in 1% KOH/ 20% glycerol solution for 2 days, and stored in 1:1 mixture of glycerol/ethanol 95%.

Histology and immunohistochemistry

For paraffin sectioning limbs and sterna were dissected out and fixed in freshly made 4% Paraformaldehyde overnight at room temperature. Bones were decalcified for 1 week with 14% EDTA (pH 7.2) followed by paraffin embedding. Sections were stained with H&E or Safranin O/Fast Green. TUNEL assay was performed using an in situ Calbiochem® Fluorescein-Fragel^TM DNA Fragmentation Detection kit (Millipore-Sigma). Following antigen retrieval using either 10 min 0.1% pronase and 30 min 0.1% hyaluronidase (type II collagen) or 10 mM Tris-EDTA (pH 9.0) at 65 °C for 2 h (Nampt), immunostaining was performed using a mouse IgG1 for type II collagen (clone II-II6B3; Developmental Studies Hybridoma Bank, University of Iowa, 1:1000) or a rabbit monoclonal antibody for Nampt (clone EPR21980; Abcam ab236874, 1:3000) or isotype control (clone EPR25A, Abcam ab172730, 1:9000). Primary antibody detection was performed using Vectastain Elite ABC-HRP Mouse (PK6102) or Rabbit (PK6101) IgG Detection Kits (Vector Labs). For undecalcified sections, freshly dissected femurs were fixed for 24 h in 10% Millonig’s formalin with 5% sucrose, transferred to ethanol and embedded undecalcified in methyl methacrylate. Calcein labels and osteoblasts were quantified on 5 μm thick longitudinal/sagittal nondecalcified sections using the OsteoMeasure Analysis System (OsteoMetrics, Inc. Atlanta, GA).

Bone mineral density and micro-CT imaging

Dual-energy X-ray absorptiometry was performed on sedated mice (2% isoflurane) using a PIXImus densitometer (GE Lunar) and data analyzed as described previously⁶⁴. Whole body (excluding the head) or the entire right femur scans were used to determine BMD. For micro-CT analysis, femurs were cleaned of adherent tissue, fixed in Millonig’s (Leica Microsystems), and stored in 100% ethanol, loaded into 10-mm diameter scanning tubes, and imaged (micro-CT40; Scanco Medical, Brüttiselen, Switzerland), as described previously⁶⁵. Scans were performed at medium resolution (12 µm isotropic voxel size) for quantitative determinations and integrated into 3-D voxel images (1024 × 1024 pixel matrices for each individual planar stack). A Gaussian filter (sigma = 0.8, support = 1) was applied to all analyzed scans. Key parameters were X-ray tube potential = 55 kVp, X-ray intensity = 145 µA, integration time = 200 ms, and threshold = 200 mg/cm³.

Primary chondrocyte cultures

The rib cages from 3–5 day-old C56BL/6 mice were dissected free of soft tissues and digested with 2 mg/ml of pronase (Sigma Aldrich, P69111) for 30 min followed by 3 mg/ml of collagenase D (Sigma Aldrich, P6885) for 1.5 h to further remove attached soft tissues. To dissociate the chondrocytes, the rib cages were further digested with 3 mg/ml of collagenase D for 2 h. Chondrocytes were then filtered through 40 μm cell strainer (Falcon™ 352340), collected by centrifugation, and cultured in DMEM with 10% FBS. Primary chondrocytes were also isolated from control and Nampt^ΔPrx1 mice born from dams that were administered NR during pregnancy. Within 6 h after birth, forelimbs and hind limbs were isolated with micro scissors from. Soft tissues surrounding the longs bones and cartilages were carefully cleaned with micro forceps without damaging the growth plate. Remaining soft tissues were further removed with enzymatic digestion using 2 mg/ml pronase for 30 min followed by Liberase TM (Sigma/Roche; 2.5 Wunch unit/mL in 1% FBS) for 1 h. Growth plates were then dissected with a surgical blade and washed 3 times in PBS (Gibco™ 10010072). Chondrocytes were dissociated with Liberase TM (Sigma/Roche; 2.5 Wunch unit/mL in 1% FBS) for 1 h. Cells were filtered through 40 μm cell strainer (Falcon™ 352340), collected by centrifugation, and cultured in DMEM supplemented with 10% FBS and 1% penicillin/streptomycin at 1% O₂ concentration in a hypoxia chamber (Coy Laboratory Products Inc).

NAD, NADH, ATP, and Lactate

NAD⁺ and NADH levels were measured using the EnzyFluo^TM NAD⁺/NADH Assay kit, according to the manufacturer’s instructions (Bioassay Systems). Briefly, primary chondrocytes were plated in a 96-well black-wall tissue plate and cultured in 1% or 20% O₂ for up to 4 days. After 1 or 2 d cultures with or without FK866, NAD and NADH extraction buffers were added to the respective wells, following the manufactured instructions. The fluorescent signal (λ_ex/em = 530/585 nm) was quantified in a Cytation 5 reader.

Intracellular ATP levels were measured by a luciferin-luciferase based assay using CellTiter-Glo® Luminescent Cell Viability Assay (G7570, Promega). Primary chondrocytes were seeded in a white opaque-bottom 96-well plate (8 × 10⁴ cells/cm²) and cultured in 1% or 20% O₂ as above. At the end of treatment, cell culture media was replaced with 100 μl of assay reagent (CellTiter-Glo Buffer + CellTiter-Glo Substrate). Contents were mixed for 2 min on an orbital shaker to promote cell lysis, followed by a 10 min incubation at room temperature. The luminescence signal was monitored in a Cytation 5 reader.

Lactate assay was performed using Glycolysis Cell-Based Assay (600450, Cayman). Primary chondrocytes were seeded and treated as described above. After treatment, the media was collected to analyze the lactate amount according to the instructions provided in the kit. The absorbance signal (490 nm) was recorded in a Cytation 5 reader.

Extracellular acidification rate (ECAR)

Real-time measurements of ECAR were determined using Seahorse XFp setup and Seahorse XFp Glycolysis Stress Test Kit (Agilent 103020-100) according to the manufacturer’s protocol. Briefly, chondrocytes seeded at a density of 4.0 × 10⁴/well in Seahorse XF96 Cell Culture Microplates in high glucose DMEM medium (Gibco™ 11995073) were washed twice and replaced with Agilent Seahorse XF Base medium supplemented with 2mM L-glutamine, 1 h before the assay. The basal ECAR rate, defined as the number of protons exported by cells into the assay medium over time was recorded 3 times over a period of 20 min and expressed as pmol/min. Cells were stained with Hoechst 33342 nuclear stain as detailed above. Raw ECAR values were normalized to the cell number count of each well.

Reactive oxygen species (ROS)

Primary chondrocytes plated in a 96-well black-wall tissue plate were incubated for 15 min at 37 °C with 5 µM CM-H₂DCFDA probe (C6827, Invitrogen). Cells were washed twice with 1× PBS, and the fluorescent signal (λ_ex/em = 492/517 nm) was measured in a Cytation™ 5 microplate reader. DCFDA fluorescence was normalized to the cell number count of each well after staining with Hoechst 33342 (Invitrogen™ H3570) nuclear stain. Hoechst fluorescence was analyzed with Cytation 5 Cell Imaging Multi-Mode Reader (BioTek Instruments Inc, Winooski, VT, USA).

Quantitative RT-PCR (qRT-PCR)

Total RNA from femoral growth plates was extracted with TRIzol reagent (Invitrogen) and reverse-transcribed using the High Capacity Reverse Transcription Kit (Applied Biosystems) according to the manufacturer’s instructions. TaqMan quantitative real-time PCR was performed in a QuantStudio3 (Applied Biosystems) using the following primers (Applied Biosystems): Col2a1 (Mm01309565_m1); Col10a1 (Mm00487041_m1); Mmp-13 (Mm00439491_m1); Vegfa (Mm00437306_m1); Acan (Mm00545794_m1); Runx2 (Mm00501584_m1). Target gene expression was calculated by normalizing to the housekeeping gene ribosomal protein S2, Mrps2 (Mm00475528_m1) using the ∆Ct method⁶⁶.

Western blot analysis

Cultured cells were washed twice with ice-cold PBS and lysed with a buffer containing 20 mM Tris-HCL, 150 mM NaCl, 1% Triton X-100, protease inhibitor mixture, and phosphatase inhibitor cocktail (Sigma-Aldrich) on ice for 30 min. The cell lysates were centrifuged at 11,200 × g for 10 min at 4 °C and the supernatants were collected in new tubes. The protein concentration of cell lysates was determined using a DC Protein Assay kit (Bio-Rad). The extracted protein (40 μg per sample) was subjected to 8 to 12% SDS-PAGE gels and transferred electrophoretically onto polyvinyl difluoride membranes (Merck Millipore). The membranes were blocked in 5% fat-free milk/Tris-buffered saline for 120 min and incubated with a primary antibody followed by a secondary antibody conjugated with horseradish peroxidase. The following primary antibodies were used to detect their corresponding protein levels: rabbit monoclonal antibody against Nampt (clone EPR21980, Abcam, ab236874, 1:1000); rabbit polyclonal antibody aagainst Acetylated-Lysine (Cell Signaling, #9411, 1:1000); mouse monoclonal antibodies against PAR (clone 10H, Enzo Life Sciences, ALX-804-220-R100, 1:1000), Sirt1 (clone 1F3, Cell Signaling, #8469, 1:1000), phospho-Histone H2A.X (clone JBW301, Sigma-Aldrich, #05-636, 1:1000), and β-actin (clone ACTBD11B7, Santa Cruz Biotechnology, sc-81178, 1:2000). The membranes were subjected to western blot analysis with ECL reagents (Millipore) and imaged with a VersaDoc^TM imaging system (Bio-Rad). All blots were derived from the same experiment and were processed in parallel. Unprocessed scans of the Western blots are shown in Supplementary Fig. 12.

In situ hybridization

Sample processing for ISH using RNAscope™ technology was performed under RNase-free conditions. Hind limbs from P2 control or Nampt^ΔPrx1 mice were quickly excised and fixed in Millonig’s buffer for 24 h at 4 °C, followed by decalcification in 15% EDTA in PBS (pH 7.4) for 1 week at 4 °C. Dehydrated hind limbs were embedded in paraffin, and 5 µm sagittal sections were probed for Ptgs2 (probe #316621) or Pim3 (probe #875481) using a RNAscope™ 2.5 HD Reagent Kit (Advanced Cell Diagnostics) and counterstained with hematoxylin.

Analysis of published scRNA-seq datasets

We downloaded the raw sequencing data of mouse long bone cells and bone marrow stroma generated by Baryawno et al.²⁵ and Baccin et al.²⁴ from the SRA database under BioProjects PRJNA527721 and PRJNA505311. These data were preprocessed using CellRanger version 6.1.2 using mm10 reference genomes to generate count matrices. The count matrices were imported to Seurat software⁶⁷ version 4.1 for further analysis. Cells containing less than 10% of mitochondrial reads or less than 3500 genes were excluded from further analysis to eliminate potential dead cells or doublets. All data were harmonized to minimize the batch effect using Canonical Correlation Analysis⁶⁷. Principal component analysis (PCA) was performed on highly variable genes of 6,000 cells to aggregate cells based on the similarities of their transcriptional profile. The results were used as input for clustering using Louvain algorithm with multilevel refinement and Uniform Manifold Approximation and Projection (UMAP) for dimension reduction. Only mesenchymal cells, identified by the expression of Pdgfra, were kept for further analysis. To identify upregulated specific gene markers of individual clusters we used the function FindMarkersAll using MAST algorithm⁶⁸. Selected markers used to identify cell types were visualized using weighted kernel density estimation of expression level using Nebulosa package (version 1.8.0)⁶⁹. The upregulated specific gene markers with a log₂ fold change greater than 0.5 and adjusted p < 1^e-50 were used for integrated analysis of metabolic pathways. To identify metabolic hot spots based on transcriptional regulation, called reporter metabolites²⁶, we integrated the gene specific markers on the mouse genome-scale metabolic model iMM1865⁷⁰ using piano R package (version 2.14.0)⁷¹. In brief, the genome-scale metabolic model was converted to a metabolic graph (unipartite), connecting metabolites and enzymes. Then, individual metabolite node in the metabolic graph are scored (enrichment p values) based on transcripts encoded by their neighboring enzymes, following the method described by Patil and Nielsen²⁶. The selected reporter metabolites of individual cell types were illustrated by dot plots. The relative expression of the metabolic marker genes (subset of the identified marker genes that are defined as metabolic genes) of individual cell types were summarized as a heatmap plot.

Single cell RNA-sequencing and data analysis

Femur and tibia growth plates were dissected from 2 control and 3 Nampt^ΔPrx1 mice at P2. These mice were born from dams that received NR supplementation, as described above. Cells were dissociated enzymatically with 0.25% trypsin-EDTA for 30 min, followed by Liberase TM (Sigma/Roche; 1 Wunch unit/mL) for 2 h with continuous mixing, collecting released cell fractions every 30 min. The pooled cell fractions were then depleted of hematopoietic cells using a miniMACS platform using LS columns and a mouse Lineage Cell Depletion Kit (Miltenyi Biotec). Approximately 12,000 cells per genotype were encapsulated using a Chromium Controller (10X Genomics, Pleasanton, CA), and libraries were constructed using a Chromium Single Cell 3’ Reagent Kit (10X Genomics). The libraries were then sequenced using an Illumina NovaSeq 6000 to generate fastq files. These files were preprocessed and analyzed as described in the “Analysis of published scRNA-seq” section above. The key results were plotted in dot-heat plot of directional enrichment score (-Log₁₀Pval). The single‐cell data set generated in this publication is accessible in the NCBI’s Sequence Read Archive (SRA) deposited under BioProject PRJNA914642.

Statistical analysis

All numerical results are reported as mean ± standard deviation of the mean (SD). The data were analyzed by Student’s t test (independent samples, two-sided) using GraphPad Prism 8 from GraphPad Software, after determining that the data were normally distributed and exhibited equivalent variances. Experiments using primary cell cultures were performed with at least three replicates and independently repeated two times.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The scRNA-seq data generated in this study have been deposited in the SRA database under accession code BioProject PRJNA914642. We also used scRNA-seq data generated by Baryawno et al.²⁵ and Baccin et al.²⁴ from the SRA database under BioProjects PRJNA527721 and PRJNA505311. Source data are provided with this paper.

References

Canto, C., Menzies, K. J. & Auwerx, J. NAD(+) metabolism and the control of energy homeostasis: a balancing act between mitochondria and the nucleus. Cell Metab. 22, 31–53 (2015).
Article CAS PubMed PubMed Central Google Scholar
Bonkowski, M. S. & Sinclair, D. A. Slowing ageing by design: the rise of NAD+ and sirtuin-activating compounds. Nat. Rev. Mol. Cell Biol. 17, 679–690 (2016).
Article CAS PubMed PubMed Central Google Scholar
Katsyuba, E. & Auwerx, J. Modulating NAD(+) metabolism, from bench to bedside. EMBO J. 36, 2670–2683 (2017).
Article CAS PubMed PubMed Central Google Scholar
Imai, S. & Guarente, L. NAD+ and sirtuins in aging and disease. Trends Cell Biol. 24, 464–471 (2014).
Article CAS PubMed PubMed Central Google Scholar
Bogan, K. L. & Brenner, C. Nicotinic acid, nicotinamide, and nicotinamide riboside: a molecular evaluation of NAD+ precursor vitamins in human nutrition. Annu Rev. Nutr. 28, 115–130 (2008).
Article CAS PubMed Google Scholar
Houtkooper, R. H., Canto, C., Wanders, R. J. & Auwerx, J. The secret life of NAD+: an old metabolite controlling new metabolic signaling pathways. Endocr. Rev. 31, 194–223 (2010).
Article CAS PubMed Google Scholar
Covarrubias, A. J., Perrone, R., Grozio, A. & Verdin, E. NAD(+) metabolism and its roles in cellular processes during ageing. Nat. Rev. Mol. Cell Biol. 22, 119–141 (2021).
Article CAS PubMed Google Scholar
Liu, L. et al. Quantitative analysis of NAD synthesis-breakdown fluxes. Cell Metab. 27, 1067–80.e5 (2018).
Article CAS PubMed PubMed Central Google Scholar
Yoshino, J., Baur, J. A. & Imai, S. I. NAD(+) intermediates: the biology and therapeutic potential of NMN and NR. Cell Metab. 27, 513–528 (2018).
Article CAS PubMed Google Scholar
Rajman, L., Chwalek, K. & Sinclair, D. A. Therapeutic potential of NAD-boosting molecules: the in vivo evidence. Cell Metab. 27, 529–547 (2018).
Article CAS PubMed PubMed Central Google Scholar
Kim, H. N. et al. A decrease in NAD(+) contributes to the loss of osteoprogenitors and bone mass with aging. NPJ Aging Mech. Dis. 7, 8 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Shi, H. et al. NAD deficiency, congenital malformations, and niacin supplementation. N. Engl. J. Med 377, 544–552 (2017).
Article CAS PubMed Google Scholar
Kronenberg, H. M. Developmental regulation of the growth plate. Nature 423, 332–336 (2003).
Article ADS CAS PubMed Google Scholar
Long, F. & Ornitz, D. M. Development of the endochondral skeleton. Cold Spring Harb. Perspect. Biol. 5, a008334 (2013).
Article PubMed PubMed Central Google Scholar
Maes, C., Carmeliet, G. & Schipani, E. Hypoxia-driven pathways in bone development, regeneration and disease. Nat. Rev. Rheumatol. 8, 358–366 (2012).
Article CAS PubMed Google Scholar
Xie, M. & Chagin, A. S. The epiphyseal secondary ossification center: Evolution, development and function. Bone 142, 115701 (2021).
Article PubMed Google Scholar
Krakow, D. Skeletal dysplasias. Clin. Perinatol. 42, 301–319 (2015).
Article PubMed PubMed Central Google Scholar
Szot, J. O. et al. New cases that expand the genotypic and phenotypic spectrum of Congenital NAD Deficiency Disorder. Hum. Mutat. 42, 862–876 (2021).
Article CAS PubMed PubMed Central Google Scholar
Cuny, H. et al. NAD deficiency due to environmental factors or gene-environment interactions causes congenital malformations and miscarriage in mice. Proc. Natl Acad. Sci. USA 117, 3738–3747 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Logan, M. et al. Expression of Cre Recombinase in the developing mouse limb bud driven by a Prxl enhancer. Genesis 33, 77–80 (2002).
Article CAS PubMed Google Scholar
Mangiavini, L., Merceron, C. & Schipani, E. Analysis of mouse growth plate development. Curr. Protoc. Mouse Biol. 6, 67–130 (2016).
Article PubMed PubMed Central Google Scholar
Ear, P. H. et al. Maternal nicotinamide riboside enhances postpartum weight loss, juvenile offspring development, and neurogenesis of adult offspring. Cell Rep. 26, 969–83.e4 (2019).
Article CAS PubMed Google Scholar
Xiong, J. et al. Matrix-embedded cells control osteoclast formation. Nat. Med. 17, 1235–1241 (2011).
Article CAS PubMed PubMed Central Google Scholar
Baccin, C. et al. Combined single-cell and spatial transcriptomics reveal the molecular, cellular and spatial bone marrow niche organization. Nat. Cell Biol. 22, 38–48 (2020).
Article CAS PubMed Google Scholar
Baryawno, N. et al. A cellular taxonomy of the bone marrow stroma in homeostasis and leukemia. Cell 177, 1915–32.e16 (2019).
Article CAS PubMed PubMed Central Google Scholar
Patil, K. R. & Nielsen, J. Uncovering transcriptional regulation of metabolism by using metabolic network topology. Proc. Natl Acad. Sci. USA 102, 2685–2689 (2005).
Article ADS CAS PubMed PubMed Central Google Scholar
Egger, S., Chaikuad, A., Kavanagh, K. L., Oppermann, U. & Nidetzky, B. UDP-glucose dehydrogenase: structure and function of a potential drug target. Biochem Soc. Trans. 38, 1378–1385 (2010).
Article CAS PubMed Google Scholar
Leonard, A. E. et al. Identification and expression of mammalian long-chain PUFA elongation enzymes. Lipids 37, 733–740 (2002).
Article CAS PubMed Google Scholar
Sprecher, H. Metabolism of highly unsaturated n-3 and n-6 fatty acids. Biochim Biophys. Acta 1486, 219–231 (2000).
Article CAS PubMed Google Scholar
Nakamura, M. T. & Nara, T. Y. Structure, function, and dietary regulation of delta6, delta5, and delta9 desaturases. Annu Rev. Nutr. 24, 345–376 (2004).
Article CAS PubMed Google Scholar
Kim, W. et al. Polyunsaturated fatty acid desaturation is a mechanism for glycolytic NAD(+) recycling. Cell Metab. 29, 856–70.e7 (2019).
Article CAS PubMed PubMed Central Google Scholar
Porter, T. D. Electron transfer pathways in cholesterol synthesis. Lipids 50, 927–936 (2015).
Article CAS PubMed Google Scholar
van Gastel, N. et al. Lipid availability determines fate of skeletal progenitor cells via SOX9. Nature 579, 111–117 (2020).
Article ADS PubMed PubMed Central Google Scholar
Lee, P., Chandel, N. S. & Simon, M. C. Cellular adaptation to hypoxia through hypoxia inducible factors and beyond. Nat. Rev. Mol. Cell Biol. 21, 268–283 (2020).
Article CAS PubMed PubMed Central Google Scholar
Yao, Q. et al. Suppressing mitochondrial respiration is critical for hypoxia tolerance in the fetal growth plate. Dev. Cell 49, 748–63.e7 (2019).
Article CAS PubMed PubMed Central Google Scholar
Wang, Y. & Hekimi, S. Understanding ubiquinone. Trends Cell Biol. 26, 367–378 (2016).
Article CAS PubMed Google Scholar
Azarm, K. & Smith, S. Nuclear PARPs and genome integrity. Genes Dev. 34, 285–301 (2020).
Article CAS PubMed PubMed Central Google Scholar
Mukherjee, S. et al. Nicotinamide adenine dinucleotide biosynthesis promotes liver regeneration. Hepatology 65, 616–630 (2017).
Article CAS PubMed Google Scholar
Frederick, D. W. et al. Loss of NAD homeostasis leads to progressive and reversible degeneration of skeletal muscle. Cell Metab. 24, 269–282 (2016).
Article MathSciNet CAS PubMed PubMed Central Google Scholar
Wang, X. et al. Deletion of nampt in projection neurons of adult mice leads to motor dysfunction, neurodegeneration, and death. Cell Rep. 20, 2184–2200 (2017).
Article CAS PubMed PubMed Central Google Scholar
Newton, P. T. et al. A radical switch in clonality reveals a stem cell niche in the epiphyseal growth plate. Nature 567, 234–238 (2019).
Article ADS CAS PubMed Google Scholar
Mizuhashi, K. et al. Resting zone of the growth plate houses a unique class of skeletal stem cells. Nature 563, 254–258 (2018).
Article ADS CAS PubMed PubMed Central Google Scholar
Goldring, M. B. & Goldring, S. R. Osteoarthritis. J. Cell Physiol. 213, 626–634 (2007).
Article CAS PubMed Google Scholar
Schipani, E. et al. Hypoxia in cartilage: HIF-1alpha is essential for chondrocyte growth arrest and survival. Genes Dev. 15, 2865–2876 (2001).
Article CAS PubMed PubMed Central Google Scholar
Lee, S. Y., Abel, E. D. & Long, F. Glucose metabolism induced by Bmp signaling is essential for murine skeletal development. Nat. Commun. 9, 4831 (2018).
Article ADS PubMed PubMed Central Google Scholar
Rossi, M. et al. Radiographic features of the skeleton in disorders of post-squalene cholesterol biosynthesis. Pediatr. Radio. 45, 965–976 (2015).
Article Google Scholar
Wu, S. & De Luca, F. Role of cholesterol in the regulation of growth plate chondrogenesis and longitudinal bone growth. J. Biol. Chem. 279, 4642–4647 (2004).
Article CAS PubMed Google Scholar
Tsushima, H. et al. Intracellular biosynthesis of lipids and cholesterol by Scap and Insig in mesenchymal cells regulates long bone growth and chondrocyte homeostasis. Development 145, dev162396 (2018).
Subedi, A. et al. Nicotinamide phosphoribosyltransferase inhibitors selectively induce apoptosis of AML stem cells by disrupting lipid homeostasis. Cell Stem Cell 28, 1851–1867.e8 (2021).
Article CAS PubMed Google Scholar
Zhang, L. Q. et al. Metabolic and molecular insights into an essential role of nicotinamide phosphoribosyltransferase. Cell Death Dis. 8, e2705 (2017).
Article CAS PubMed PubMed Central Google Scholar
Lundt, S. et al. Metabolomic and transcriptional profiling reveals bioenergetic stress and activation of cell death and inflammatory pathways in vivo after neuronal deletion of NAMPT. J. Cereb. Blood Flow. Metab. 41, 2116–2131 (2021).
Article CAS PubMed PubMed Central Google Scholar
Simic, P. et al. SIRT1 regulates differentiation of mesenchymal stem cells by deacetylating beta-catenin. EMBO Mol. Med. 5, 430–440 (2013).
Article CAS PubMed PubMed Central Google Scholar
Matsuzaki, T. et al. Disruption of Sirt1 in chondrocytes causes accelerated progression of osteoarthritis under mechanical stress and during ageing in mice. Ann. Rheum. Dis. 73, 1397–1404 (2014).
Article CAS PubMed Google Scholar
Ling, W. et al. Mitochondrial Sirt3 contributes to the bone loss caused by aging or estrogen deficiency. JCI Insight 6, e146728 (2021).
Ailixiding, M. et al. Pivotal role of Sirt6 in the crosstalk among ageing, metabolic syndrome and osteoarthritis. Biochem. Biophys. Res. Commun. 466, 319–326 (2015).
Article CAS PubMed Google Scholar
Wang, Z. Q. et al. Mice lacking ADPRT and poly(ADP-ribosyl)ation develop normally but are susceptible to skin disease. Genes Dev. 9, 509–520 (1995).
Article CAS PubMed Google Scholar
Menissier de Murcia, J. et al. Functional interaction between PARP-1 and PARP-2 in chromosome stability and embryonic development in mouse. EMBO J. 22, 2255–2263 (2003).
Article CAS PubMed PubMed Central Google Scholar
Brakedal, B. et al. The NADPARK study: a randomized phase I trial of nicotinamide riboside supplementation in Parkinson’s disease. Cell Metab. 34, 396–407.e6 (2022).
Article CAS PubMed Google Scholar
Zhou, B. et al. Boosting NAD level suppresses inflammatory activation of PBMCs in heart failure. J. Clin. Invest 130, 6054–6063 (2020).
Article CAS PubMed PubMed Central Google Scholar
Freeberg, K. A. et al. Nicotinamide riboside supplementation for treating elevated systolic blood pressure and arterial stiffness in midlife and older adults. Front. Cardiovasc. Med. 9, 881703 (2022).
Article CAS PubMed PubMed Central Google Scholar
Remie, C. M. E. et al. Nicotinamide riboside supplementation alters body composition and skeletal muscle acetylcarnitine concentrations in healthy obese humans. Am. J. Clin. Nutr. 112, 413–426 (2020).
Article PubMed PubMed Central Google Scholar
Nascimento, E. B. M. et al. Nicotinamide riboside enhances in vitro beta-adrenergic brown adipose tissue activity in humans. J. Clin. Endocrinol. Metab. 106, 1437–1447 (2021).
Article PubMed Google Scholar
Rodda, S. J. & McMahon, A. P. Distinct roles for Hedgehog and canonical Wnt signaling in specification, differentiation and maintenance of osteoblast progenitors. Development 133, 3231–3244 (2006).
Article CAS PubMed Google Scholar
O’Brien, C. A. et al. IL-6 is not required for parathyroid hormone stimulation of RANKL expression, osteoclast formation, and bone loss in mice. Am. J. Physiol. Endocrinol. Metab. 289, E784–E793 (2005).
Article PubMed Google Scholar
Martin-Millan, M. et al. The estrogen receptor-alpha in osteoclasts mediates the protective effects of estrogens on cancellous but not cortical bone. Mol. Endocrinol. 24, 323–334 (2010).
Article CAS PubMed PubMed Central Google Scholar
Livak, K. J. & Schmittgen, T. D. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 25, 402–408 (2001).
Article CAS PubMed Google Scholar
Butler, A., Hoffman, P., Smibert, P., Papalexi, E. & Satija, R. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat. Biotechnol. 36, 411–420 (2018).
Article CAS PubMed PubMed Central Google Scholar
Finak, G. et al. MAST: a flexible statistical framework for assessing transcriptional changes and characterizing heterogeneity in single-cell RNA sequencing data. Genome Biol. 16, 278 (2015).
Article PubMed PubMed Central Google Scholar
Alquicira-Hernandez, J. & Powell, J. E. Nebulosa recovers single cell gene expression signals by kernel density estimation. Bioinformatics 37, 2485–2487 (2021).
Khodaee, S., Asgari, Y., Totonchi, M. & Karimi-Jafari, M. H. iMM1865: a new reconstruction of mouse genome-scale metabolic model. Sci. Rep. 10, 6177 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Varemo, L., Nielsen, J. & Nookaew, I. Enriching the gene set analysis of genome-wide data by incorporating directionality of gene expression and combining statistical hypotheses and methods. Nucleic Acids Res. 41, 4378–4391 (2013).
Article PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We thank J.A. Crawford, S.B. Berryhill, and H. Wu for technical support and Dr Shin-ichiro Imai for generously sharing the Nampt^fl mice. This work was supported by US National Institutes of Health R01AR056679 (to MA), R01AG068449 (to MA and CAO), and P20GM125503 (to CAO) and the UAMS Bone and Joint Initiative.

Author information

These authors contributed equally: Aaron Warren, Ryan M. Porter.

Authors and Affiliations

Division of Endocrinology and Metabolism, University of Arkansas for Medical Sciences, Little Rock, AR, USA
Aaron Warren, Ryan M. Porter, Olivia Reyes-Castro, Md Mohsin Ali, Ha-Neui Kim, Landon B. Gatrell, Charles A. O’Brien & Maria Almeida
Center for Musculoskeletal Disease Research, University of Arkansas for Medical Sciences, Little Rock, AR, USA
Ryan M. Porter, Ha-Neui Kim, Intawat Nookaew, Charles A. O’Brien, Roy Morello & Maria Almeida
Department of Orthopedic Surgery, University of Arkansas for Medical Sciences, Little Rock, AR, USA
Ryan M. Porter, Charles A. O’Brien, Roy Morello & Maria Almeida
CNC—Center for Neuroscience and Cell Biology, University of Coimbra, UC-Biotech, Biocant Park, Cantanhede, Portugal
Adriana Marques-Carvalho
Department of Orthopaedic Surgery, University of Pennsylvania, Philadelphia, PA, USA
Ernestina Schipani
Department of Biomedical Informatics, University of Arkansas for Medical Sciences, Little Rock, AR, USA
Intawat Nookaew
Department of Physiology and Cell Biology, University of Arkansas for Medical Sciences, Little Rock, AR, USA
Roy Morello

Authors

Aaron Warren
View author publications
You can also search for this author in PubMed Google Scholar
Ryan M. Porter
View author publications
You can also search for this author in PubMed Google Scholar
Olivia Reyes-Castro
View author publications
You can also search for this author in PubMed Google Scholar
Md Mohsin Ali
View author publications
You can also search for this author in PubMed Google Scholar
Adriana Marques-Carvalho
View author publications
You can also search for this author in PubMed Google Scholar
Ha-Neui Kim
View author publications
You can also search for this author in PubMed Google Scholar
Landon B. Gatrell
View author publications
You can also search for this author in PubMed Google Scholar
Ernestina Schipani
View author publications
You can also search for this author in PubMed Google Scholar
Intawat Nookaew
View author publications
You can also search for this author in PubMed Google Scholar
Charles A. O’Brien
View author publications
You can also search for this author in PubMed Google Scholar
Roy Morello
View author publications
You can also search for this author in PubMed Google Scholar
Maria Almeida
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

M.A. conceived and designed the experiments; A.W. and O.R.C. performed all animal experimentation; A.M.C., L.B.G., H.N.K., and M.M.A. generated and analyzed the in vitro data; O.R.C., R.M.P., and R.M. performed histological analysis; L.B.G. and R.M.P. collected and processed samples for scRNA-seq; C.A.O. and I.N. analyzed scRNAseq data; R.M., E.S., and R.M.P. provided methodology and technical support; R.M.P. and M.A. wrote the manuscript; all authors reviewed the manuscript.

Corresponding author

Correspondence to Maria Almeida.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Description of Additional Supplementary Files Document

Supplementary Dataset 1

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Warren, A., Porter, R.M., Reyes-Castro, O. et al. The NAD salvage pathway in mesenchymal cells is indispensable for skeletal development in mice. Nat Commun 14, 3616 (2023). https://doi.org/10.1038/s41467-023-39392-7

Download citation

Received: 13 July 2022
Accepted: 09 June 2023
Published: 17 June 2023
DOI: https://doi.org/10.1038/s41467-023-39392-7

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.