Cell Metabolism

Previews

supplemented with excess proline ex-

hibited a reduced lifespan, whereas

supplementing proline to E. coli OP50

had no additional adverse effects on

longevity. However, it is likely that the

intermediate P5C and not proline itself is

toxic to the animal, because the detri-

mental effect of alh-6 mutation is lost

when the conversion of proline into this

compound is prevented by a perturbation

of PRODH/B0513.5. In the future, mass

spectrometry of different bacteria may

reveal whether they harbor different levels

of proline or P5C. Since both are E. coli

strains, it is unlikely that such differences

would result from major differences in

metabolic pathways. Rather, variations

in metabolic pathway genes may underlie

the observed effects. Exploring variation

in proline metabolism and other gene

activity or expression, bacterial genetic

178 Cell Metabolism 19, February 4, 2014 ª2

screens, or targeted mutagenesis of

candidate genes will be useful to unravel

the precise mechanism involved.

Altogether, this study provides an

important step forward in our apprecia-

tion of the effect different amino acids

can exert, whether it is the conversion of

proline central to this study, tryptophan

in nhr-114 mutants on the E. coli OP50

diet (Gracida and Eckmann, 2013), or

branched-chain amino acid metabolism

in response to the Comamonas diet

(Watson et al., 2013). C. elegans and its

bacterial diets will no doubt continue to

provide a powerful interspecies paradigm

to dissect the interactions between

specific genes and nutrients, and their

effects on development, fertility, and

aging. As with many other key findings

from the nematode, results may illuminate

how particular nutrients, such as amino

014 Elsevier Inc.

acids, affect cellular and organismal

physiology in health and disease in

humans as well.

REFERENCES

Brooks, K.K., Liang, B., and Watts, J.L. (2009).PLoS ONE 4, e7545.

Gracida, X., and Eckmann, C.R. (2013). Curr. Biol.23, 607–613.

Hoeven, Rv., McCallum, K.C., Cruz, M.R., andGarsin, D.A. (2011). PLoS Pathog. 7, e1002453.

MacNeil, L.T., Watson, E., Arda, H.E., Zhu, L.J.,and Walhout, A.J.M. (2013). Cell 153, 240–252.

Pang, S., and Curran, S.P. (2014). Cell Metab., inpress.

Tissenbaum, H.A. (2012). J. Gerontol. A Biol. Sci.Med. Sci. 67, 503–510.

Watson, E., MacNeil, L.T., Arda, H.E., Zhu, L.J.,and Walhout, A.J.M. (2013). Cell 153, 253–266.

NAD+ Deficiency in Age-RelatedMitochondrial Dysfunction

Tomas A. Prolla1,* and John M. Denu21Departments of Genetics and Medical Genetics2Department of Biomolecular ChemistryUniversity of Wisconsin, Madison, WI 53706, USA*Correspondence: taprolla@facstaff.wisc.eduhttp://dx.doi.org/10.1016/j.cmet.2014.01.005

Age-related mitochondrial dysfunction is thought to contribute to mammalian aging, particularly in post-mitotic tissues that rely heavily on oxidative phosphorylation. A new study (Gomes et al., 2013) shows thatreduced levels of nicotinamide adenine dinucleotide (NAD+) contribute to the mitochondrial decay asso-ciated with skeletal muscle aging and that sirtuin 1 (SIRT1) modulates this process.

Decreased mitochondrial function with

age has been documented in multiple

mammalian species. Studies on isolated

mitochondria from human muscle bi-

opsies or rodent muscles support the

existence of an intrinsic, aging-dependent

mitochondrial defect associated with

ATP production (Short et al., 2005), and

studies employing permeabilized muscle

fibers also demonstrate impaired mito-

chondrial function in aged humans

(Joseph et al., 2012). If mitochondrial

decay does indeed contribute to aging

phenotypes, interventions that prevent

or reverse such decay may improve

the quality of life of aged individuals. A

major roadblock to the development of

such interventions is that the specific

mechanisms of age-related mitochondrial

dysfunction are poorly understood. A role

formtDNAmutations has been postulated

due to the proximity of mtDNA and free

radical production sites in mitochondria.

Indeed, mice engineered to accumulate

mtDNA mutations at high levels display

multiple aging phenotypes (Kujoth et al.,

2005). However, the levels of mtDNA

point mutations and deletions found in

most tissues from aged humans or ani-

mals usually account for less than 1%

of the total mtDNA and are unlikely to

be major contributors to age-related

mitochondrial dysfunction. A reduction

in mtDNA copy number may be more

relevant, as it could lead to reduced

levels of mtDNA-encoded transcripts

and proteins, as reported in skeletal mus-

cle of older adults (Short et al., 2005).

Gomes et al. now provide insights into

the relationship between mtDNA, NAD+,

and the mitochondrial dysfunction of

aging (Gomes et al., 2013). The authors

Figure 1. Age-Dependent Decline in NAD+

Decreased NAD+ synthesis and increased NAD+ consumption with agemay both contribute to a decreasein the NAD+ pool. A reduction in NAD+ levels leads to an age-related reduction of SIRT1 activity. ReducedSIRT1 activity impacts mitochondrial function through at least two mechanisms: (1) a reduction in bio-genesis secondary due to a reduction in PGC1-a activity and (2) an impairment of mitochondrial functiondue to a reduction in mtDNA replication and transcription.

Cell Metabolism

Previews

report progressive age-related reduction

in mtDNA-encoded transcripts in the

gastrocnemius (calf) muscle of mice.

Interestingly, a reduction in nuclear-

encoded mitochondrial genes was also

observed but appeared to lag behind the

loss of mtDNA-encoded transcripts by

several months. These transcriptional

alterations were associated with a decline

in mtDNA copy number and a decline

in ATP content, suggesting that these

events are causally linked. Since the

authors had previously linked sirtuin 1

(SIRT1) activity with increased mitochon-

drial biogenesis through PGC1-a deace-

tylation, they tested whether these age-

related mitochondrial alterations might

be due to reduced SIRT1 activity, using

an elegant tissue-specific (SIRT1 induc-

ible knockout [iKO]) transgenic model

to generate mice that lacked SIRT1 in

skeletal muscle. Interestingly, this led to

reduced expression of mtDNA-encoded

genes, but not nuclear-encoded mito-

chondrial genes. The intervention also

had no impact on mtDNA mass, which

is expected to be reduced if the

SIRT1-PGC1-a pathway of mitochondrial

biogenesis is impaired. Thus, under basal

conditions (i.e., not under caloric restric-

tion or exercise) the SIRT1-PGC1-a

pathwaymay not be themain determinant

of mitochondrial biogenesis. However,

since NAD+ is required for SIRT1 activity,

the authors next examined the levels

of NAD+ in tissues of aged mice and

found a dramatic decrease in NAD+ with

aging. Furthermore, overexpression of

nicotinamide mononucleotide adenylyl

transferase (NMNAT1), a nuclear NAD+

synthetic enzyme, in skeletal muscle of

10- to 12-month-old mice significantly

increased the levels of NAD+ and, impor-

tantly, the expression of mtDNA-encoded

genes. In vitro studies with myoblasts

suggest that the role of NMNAT1 in this

system is dependent on SIRT1. Strikingly,

short-term feeding of NMN (an NAD+ pre-

cursor) to old mice reversed several

biochemical parameters associated with

mitochondrial aging. Caloric restriction,

which retards aging and induces SIRT1,

also prevented the age-related NAD+

decline and maintained mtDNA gene

expression. Thus, variations in NAD+

levels play a critical role in mtDNA tran-

Cell Metabolism 19

scription and mitochondrial aging, and

this role appears to be mediated at least

in part through SIRT1.

The mechanism by which SIRT1 and

NAD+ control mtDNA transcription is not

fully understood. Gomes et al. suggest

that NAD+-dependent SIRT1 activity reg-

ulates the nuclear expression of mito-

chondrial transcription factor A (TFAM),

which plays an essential role in mainte-

nance, expression, and organization of

mitochondrial DNA (Gomes et al., 2013).

The actual acetylated protein targets of

the SIRT1 deacetylase responsible for

TFAM expression were not identified.

However, the authors implicated a path-

way involving hypoxia-inducible factor

1a (HIF1a), c-Myc, and von Hippel-Lindau

(VHL) E3 ubiquitin ligase that recognizes

hydroxylated proline residues on HIF1a

for degradation by the proteosome

(Gomes et al., 2013). The authors propose

that lowered NAD+ with age decreases

the ability of VHL to downregulate

HIF1a, which would compete with the

opposing functions of c-Myc on the

TFAM promoter. SIRT1 expression had

no effect on VHL promoter activity, but

VHL protein levels positively correlated

with SIRT1 protein, suggesting that

SIRT1 directly or indirectly controls VHL

at the posttranscriptional level. The

simplest explanation is that VHL is a direct

acetylation target of SIRT1; however, no

data were shown to support this idea.

Clearly, additional experiments are

needed to fully elucidate the molecular

role played by SIRT1 during the age-

dependent decline in NAD+ levels.

The observation that NMN supple-

mentation can restore NAD+ levels and

markers of mitochondrial function that

decline with age prompts a number of

interesting questions. Why does NAD+

decline with age? The levels of NAD+ are

governed by the cellular redox state (i.e.,

NAD+/NADH), the rate of NAD+ synthesis,

and the rate of NAD+ consumption

(Figure 1). NAD+ and NADH are rapidly

interchangeable, and levels of NADH are

considerably lower than those of NAD+,

making this redox ratio a less dramatic

explanation for age-dependent decline

of NAD+. However, a lowered NAD+/

NADH ratio is associated with aging

muscle, consistent with a possible link

(Pugh et al., 2013). The ability to rescue

mitochondrial function with NMN sug-

gests that the NAD+ salvage pathway is

, February 4, 2014 ª2014 Elsevier Inc. 179

Cell Metabolism

Previews

defective at a step up to or including the

reaction catalyzed by nicotinamide phos-

phoribosyltransferase (NAMPT), which

generates NMN as a product. On the

other hand, the ability of NMNAT1 to

restore mitochondrial function suggests

that NMNAT1 levels are rate-limiting for

sustaining adequate NAD+ levels. These

questions could be addressed by exam-

ining the ability of NAMPT to rescue mito-

chondrial function and whether upstream

NAD+ precursors (e.g., nicotinamide and

nicotinamideriboside) function as well as

NMN. Another interesting explanation

for declining NAD+ may be that NAD+-

consuming enzymes are hyperstimulated.

A likely candidate is poly(ADP-ribose)

polymerase-1, PARP1, which depletes

nuclear and cytoplasmic NAD+ upon

genotoxic stress (Cipriani et al., 2005).

Accumulating DNA damage as cells age

might chronically stimulate the NAD+-

consuming activity of PARP (Massudi

et al., 2012), leading to reduced NAD+

levels and SIRT1 activity, which is critical

180 Cell Metabolism 19, February 4, 2014 ª2

for TFAM expression and mitochondrial

oxidative phosphorylation (OXPHOS)

expression (Figure 1). In that case, PARP

inhibition might restore mitochondrial

function. Ultimately, theremay bemultiple

causes for age-dependent decline in

NAD+, including a metabolic shift to pro-

cesses that disfavor a high NAD+/NADH

ratio, defective NAD+ salvage, or over-

consumption of NAD+ cleaving enzymes,

like PARP, CD38, and even sirtuins

(Braidy et al., 2013). Importantly, the study

suggests that age-related mitochondrial

dysfunction is not due to the accumula-

tion of irreversible damage to macromole-

cules. Instead, the defect results in part

from a decline in the levels of a key coen-

zyme and therefore may be reversible.

REFERENCES

Braidy, N., Poljak, A., Grant, R., Jayasena, T., Man-sour, H., Chan-Ling, T., Guillemin, G.J., Smythe,G., and Sachdev, P. (2013). Biogerontology. Pub-lished online December 17, 2013. http://dx.doi.org/10.1007/s10522-013-9489-5.

014 Elsevier Inc.

Cipriani, G., Rapizzi, E., Vannacci, A., Rizzuto, R.,Moroni, F., and Chiarugi, A. (2005). J. Biol. Chem.280, 17227–17234.

Gomes, A.P., Price, N.L., Ling, A.J., Moslehi, J.J.,Montgomery, M.K., Rajman, L., White, J.P., Teo-doro, J.S., Wrann, C.D., Hubbard, B.P., et al.(2013). Cell 155, 1624–1638.

Joseph, A.M., Adhihetty, P.J., Buford, T.W., Wohl-gemuth, S.E., Lees, H.A., Nguyen, L.M., Aranda,J.M., Sandesara, B.D., Pahor, M., Manini, T.M.,et al. (2012). Aging Cell 11, 801–809.

Kujoth, G.C., Hiona, A., Pugh, T.D., Someya, S.,Panzer, K., Wohlgemuth, S.E., Hofer, T., Seo,A.Y., Sullivan, R., Jobling, W.A., et al. (2005).Science 309, 481–484.

Massudi, H., Grant, R., Braidy, N., Guest, J., Farns-worth, B., and Guillemin, G.J. (2012). PLoS ONE 7,e42357.

Pugh, T.D., Conklin, M.W., Evans, T.D., Polewski,M.A., Barbian, H.J., Pass, R., Anderson, B.D.,Colman, R.J., Eliceiri, K.W., Keely, P.J., et al.(2013). Aging Cell 12, 672–681.

Short, K.R., Bigelow, M.L., Kahl, J., Singh, R.,Coenen-Schimke, J., Raghavakaimal, S., andNair, K.S. (2005). Proc. Natl. Acad. Sci. USA 102,5618–5623.