NAD (nicotinamide adenine dinucleotide) is a cofactor of key enzymes in glycolysis, the tricarboxylic acid cycle, and oxidative phosphorylation (Oxphos), participating in multiple redox reactions in cells. It has been shown that the cellular NAD pool is determined by a balance between the activity of NAD-synthesizing and consuming enzymes (Aksoy et al., 2006, Bai, et al., 2011, Barbosa et al., 2007, Nahimana et al., 2009, Yang et al., 2007, Yoshino et al., 2011). Recently it has been described that NAD levels decline during chronological aging and in progeroid states, leading to mitochondrial dysfunction and metabolic abnormalities (Zhu et al., 2015; Gomes et al., 2013; Braidy et al., 2011; Scheibye-Knudsen et al., 2014, Massudi et al., 2012). Thus, determining the mechanisms that lead to this age-related NAD decline is of great importance for the development of therapies for age-related diseases.
Age-related NAD decline may be caused by an increase in NAD-consuming enzymes (NADases), a decrease in NAD-synthesizing enzymes, or a combination of both. However, to date, the precise contribution of specific metabolic pathways that regulate NAD levels in the age-related NAD-decline has not been determined. In the case of NAD consumption, it has been shown that enzymes such as poly (ADP-ribose) polymerases (PARPs), NAD-dependent deacetylases (SIRTUINS), and NADases such as CD38 can degrade this molecule during their catalytic processes (Bai et al., 2011; Bai and Canto, 2012, Aksoy et al., 2006, Imai and Guarente, 2014). In particular, we have shown that the enzyme CD38 is one of the main NAD-degrading enzymes in mammalian tissues (Aksoy et al., 2006; Barbosa et al., 2007). CD38 was originally identified as a cell surface enzyme that plays a key role in several physiological processes such as immune response, inflammation, cancer, and metabolic disease (Barbosa et al., 2007; Frasca et al., 2006; Guedes et al., 2012; Malavasi et. al., 2008). We have previously shown that CD38 knockout (CD38KO) mice have higher NAD levels and are protected against obesity and metabolic syndrome (Barbosa et al., 2007). In addition, treatment of obese mice with CD38 inhibitors increases intracellular NAD levels and improves several aspects of glucose and lipid homeostasis (Escande et al., 2013). However, the role of CD38 in age-related NAD decline and mitochondrial dysfunction has never been investigated.
Here we show for the first time that CD38 plays an active role in the age-related NAD decline in mammals. NAD levels, mitochondrial respiratory rates, and metabolic functions are preserved during the aging process in CD38KO mice. We further identified that CD38 is the main enzyme metabolizing the NAD precursor NMN in vivo, and demonstrated that ablation of CD38 improves the response to NAD-replacement therapy during aging. We believe that these findings may lead to new strategies for the treatment of diseases related to an imbalance in NAD metabolism and energy homeostasis.
Previous studies have suggested that tissue NAD+ levels decline with aging (Zhu et al., 2015; Gomes et al., 2013; Braidy et al., 2011; Massudi et al., 2012). To confirm the data on NAD-decline in aging, we measured NAD+ and NADH levels in murine tissues by two different methods, the cycling assay and a UPLC-mass spectroscopy assay (see Experimental Procedures section and Supplemental Information)(Figure S1A–C). We also further optimized and validated the cycling assay, and determined that it is very specific for NAD+ and NADH and does not detect any of the other nucleotides or NAD derivatives tested including: NADP, NAAD, NAADP, cADPR, ATP, ADP and others (Figure S1A). The results obtained with both methods confirm that there is indeed a decrease in levels of both NAD+ and NADH in murine tissues during chronological aging (Figure S1B–C). Furthermore, both techniques correlated extremely well for both nucleotides (correlation coefficient of r=0.95 for NAD+ and 0.97 for NADH, Figure S1B–C). In subsequent experiments NAD levels were generally assayed using the cycling method. Because our data agree with previous studies indicating that tissue NAD levels decline with aging, we next investigated the mechanisms leading to this age-related NAD decline in tissues.
NAD decline could be caused by an increase in NAD-consuming enzymes and/or a decrease in its synthesizing enzymes. Some of the NAD-degrading enzymes in mammalian tissues include CD38, PARPs, and SIRTUINS. We have previously shown that CD38, in particular, is one of the main NAD-degrading enzymes in mammalian tissues (Aksoy et al., 2006; Barbosa et al., 2007). Furthermore, PARP1 has been proposed to be involved in NAD decline in tissues, and SIRT1 is an NAD-dependent deacetylase that plays a key role in age-related metabolic phenotypes (Bai et al., 2011; Bai and Canto, 2012; Imai and Guarente, 2014). Thus, we first investigated the expression of these enzymes in several mouse tissues during aging. As shown in Figure 1, the protein levels of both PARP1 and SIRT1 decreased in all of the tissues tested including liver, white adipose tissue, spleen, and skeletal muscle (Figure 1 A–D, S1D–F). These data indicate that the age-related NAD decline is not mediated by an increase in the expression of either PARP1 or SIRT1. In sharp contrast, levels of the NADase CD38 increased at least 2 to 3 times during chronological aging in all tissues tested, as detected by immunoblotting (Figure 1A–D and S1D–F). Furthermore, in the liver, we observed that the number and intensity of CD38 positive cells increased during aging as determined by flow cytometry (Figure S1G).