Leptin gene transfer in the hypothalamus enhances longevity in adult monogenic mutant mice in the absence of circulating leptin

Neurobiology of Aging 28 (2007) 1594–1604

Leptin gene transfer in the hypothalamus enhances longevity in adult monogenic mutant mice in the absence of circulating leptin夽 St´ephane Boghossian a , Naohiko Ueno a , Michael G. Dube b , Pushpa Kalra b , Satya Kalra a,∗ a

Department of Neuroscience, University of Florida McKnight Brain Institute, Gainesville, FL, USA b Department of Physiology and Functional Genomics, Gainesville, FL, USA Received 28 June 2006; received in revised form 17 August 2006; accepted 24 August 2006 Available online 29 September 2006

Abstract Leptin, a product of the ob gene, is a pleiotropic signal implicated in regulation of multiple physiological functions in the periphery and centrally, including hypothalamic integration of energy homeostasis. Recessive mutations of ob gene result in early onset of hyperphagia, morbid obesity, metabolic disorders, early mortality and shortened life-span. Intracerebroventricular injection of recombinant adeno-associated virus vector (rAAV) encoding the leptin gene in adult obese ob/ob mice enhanced leptin transgene expression only in the hypothalamus, normalized food intake, body weight and more than doubled the life-span as compared to control cohorts and extended it to near that of normal wild type mice. These life-extending benefits were associated with drastic reductions in visceral fat, and blood glucose and insulin levels, but elevated ghrelin levels, the anti-aging biomarkers. Thus, bioavailability of leptin transduced by ectopic gene in the hypothalamus alone is both necessary and sufficient to normalize life-span. Evidently, site-specific ectopic gene expression with rAAV is durable and safe for alleviating neural disorders that stem from missing or functional disruption of a single gene. © 2006 Elsevier Inc. All rights reserved. Keywords: Leptin gene therapy; Longevity; Fat; Hypothalamus

1. Introduction Environmental and genetic causes either alone or in a complex interplay are thought to contribute to the current worldwide escalation in the incidence of obesity and obesitydependent metabolic and neurological afflictions [2,6,14,27, 30,36,37,51]. Genetically-based disruptions that tilt energy balance in favor of increased energy consumption, decreased energy expenditure and accelerated rates of fat deposition are diverse and involve complex interactions of neural, hormonal and metabolic factors [1,30,43,45,46,51,61]. Recent delineation of these varied derangements in the internal environment in monogenic and polygenic obesities has uncovered 夽 Presented at the 34th Annual Meeting of the Society for Neuroscience, San Diego, CA, 2004, Abstract #565.515. ∗ Corresponding author at: Department of Neuroscience, UF McKnight Brain Institute, P.O. Box 100244, Gainesville, FL 32610-0244, USA. Tel.: +1 352 392 2895; fax: +1 352 294 0191. E-mail address: [email protected] (S. Kalra).

0197-4580/$ – see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.neurobiolaging.2006.08.010

avenues for therapeutic interventions at genetic and molecular levels [6,26,34,45]. Rodent models of monogenic obesity are suitable paradigms to test novel molecular and genetic interventional approaches aimed at retarding fat accretion on a long-term basis [6,26,34,45]. Recessive mutation in ob gene results in morbid obesity resulting from an early onset of hyperphagia and decreased energy expenditure [6,28,34,36,55,60]. Leptin, the product of the ob gene, is produced by adipocytes and non-adipocyte tissues, including the hypothalamus [3,28,36,37,53,63,64]. Leptin is a pleiotropic signal that regulates multiple physiological functions in the periphery and neural functions in the brain by engaging target specific leptin receptors [1,5,24,28,29]. In the periphery, leptin has been implicated in the regulation of blood pressure, renal function, angiogenesis, wound healing, immune function and bone formation [12,16,18,29,35,42,48,50,57,58]. Among central effects, a primary action of leptin is to integrate energy homeostasis by controlling distinct energy

S. Boghossian et al. / Neurobiology of Aging 28 (2007) 1594–1604

intake and expenditure regulating neural pathways in the hypothalamus [1,28,33,36,37]. Absence of leptin engenders intense hyperphagia, excess fat accretion, life-long morbid obesity accompanied by diminished non-shivering thermogenic energy expenditure in leptin-mutant ob/ob mice and congenital leptin-deficient human subjects [1,20,21,25,26, 28,36,37,50,60]. Morbidly obese human patients and obese mice exhibit a variety of life-threatening complications, including metabolic syndrome, early mortality and shortened life-span [2,14,25–27,30,46,51,61,65]. Experimental evidence showing that leptin replacement either systemically or centrally restores energy homeostasis in ob/ob mice [28,29,33,36], and conditional deletion of leptin receptors in the hypothalamus reproduces the ob/obtype phenotype [17,36,40], suggests that leptin action in the hypothalamus alone can reinstate weight homeostasis in ob/ob mice. However, whether imbalance due to leptin deficiency in the hypothalamus and/or global pathophysiologic complications resulting from leptin deficiency in the periphery accelerate aging and shorten life-span, is not known. Research in gene transfer strategies for developing interventional therapies for neural disorders has proceeded at rapid pace [15,37,39,40]. Among various gene therapy approaches currently available, insertion of a missing gene in an appropriate site with the aid of a suitable viral vector has the potential to remedy the pathophysiologic consequences of a mutant gene on a long-term basis [15,37,39,40]. Leptin gene transfer into the hypothalamus with a non-pathogenic and non-immunogenic recombinant adeno-associated virus encoding the leptin gene (rAAV-lep), was found to enhance leptin transgene expression in the hypothalamus contemporaneous with weight normalization, decreased adiposity and symptoms of type 2 diabetes in leptin-mutant ob/ob mice even in the absence of leptin into the periphery [4,5,7,8,11,20–24,37,39,58,59]. A similar central leptin gene therapy retarded the gradual time-related increase in fat accretion and secretion of adipocyte adipokines in wild type (wt) mice in short term experiments [10,58,59]. Since rAAV vectors infect mainly neurons in the nervous system and support expression of targeted gene for the lifetime of cells in vivo [15,37,39], we examined the efficacy of bioavailability of leptin by leptin gene transfer in the hypothalamus on mortality and longevity in ob/ob mice.

2. Research methods and procedures 2.1. Animals Six week-old (40–50 g) leptin-mutant ob/ob and wild type (wt) C57BL/6J male mice (The Jackson Laboratory, Bar Harbor, ME) were housed individually in temperature and light controlled rooms (lights on 06:00–18:00 h) under specific pathogen-free conditions. Standard chow diet (11 kcal% fat; LM-485 Tecklad, Madison, WI) and water were available ad

1595

libitum throughout the experiment. All animals were allowed at least 1 week of adaptation to the animal rooms before initiation of experimental procedures. The Institutional Animal Care and Use Committee of University of Florida approved the animal protocols. 2.2. Experimental design 2.2.1. Experiment 1 The objective of this experiment was to evaluate the benefit of alleviating leptin insufficiency selectively in the hypothalamus on mortality and longevity in ob/ob mice. ob/ob and wt (controls) mice were anesthetized with sodium pentobarbital (60 mg/kg) and placed in a David Kopf stereotaxic apparatus with a mouse adapter for intracerebroventricular (icv) injection [10,58,59]. The stereotaxic coordinates for icv injections were: 0.3 mm posterior to bregma, 0 mm lateral to midline and 4.2 mm below the dura. One group from each genotype was injected icv with a non-immunogenic, non-pathogenic recombinant adeno-associated virus (rAAV) encoding the green fluorescent protein gene (rAAV-GFP, 9 × 107 particles in 1.5 ␮l; ob/ob n = 12 and wt n = 10), and the other group received rAAV encoding rat leptin gene (rAAV-lep, 9 × 107 particles in 1.5 ␮l; ob/ob n = 12 and wt n = 10). rAAVlep or rAAV-GFP preparation was slowly infused over a 2 min period and the injector was removed 5 min later. The vectors used in this study were packaged, purified, concentrated, titered and verified for leptin transduction in vitro, and in vivo after intravenous administration in ob/ob mice [20,21]. Administration of rAAV-lep vector icv has also been tested in restraining body weight (BW) and in suppressing fat deposition [8,10,11,58,59]. BW and food intake (FI) were monitored weekly for 30 weeks and then biweekly until the end of the experiment. Average daily consumption was derived for data analyses. Since previous studies have shown that the effects of icv rAAV-lep injection on BW and FI are dose-dependent [22] and injection of low doses of rAAV-lep reduced FI in ob/ob but not in wt mice [58,59], consistent with reported heightened sensitivity of ob/ob mice to leptin [28,36,37], two additional control groups of ob/ob mice were monitored in parallel. One group served as an unoperated control (untreated, n = 8) and the other group served as a pair-fed (PF, n = 10) control. Food was supplied before lights-off to PF mice in amounts adjusted weekly to match the food consumption of rAAV-lep treated ob/ob mice during the week before [58,59]. All mice were housed one per cage and inspected twice a day on weekdays and once a day on weekends until natural death occurred or were moribund and presented symptoms of an eminent death, when as advised by the veterinarian staff, they were euthanized. In all, eight mice from all experimental groups (ob/ob untreated; n = 1, rAAV-GFP; n = 1, ob/ob PF; n = 2 and wt rAAV-GFP; n = 4) were euthanized as instructed by the veterinarian. In addition, we did not observe any external signs of malignancy or pathology by these mice during the experiment.

1596

S. Boghossian et al. / Neurobiology of Aging 28 (2007) 1594–1604

2.2.2. Experiment 2 During the course of experiment 1, we observed mortality onset after the age of 40 weeks in control ob/ob mice. This observation implied that the effects of excess fat and fat-dependent aging biomarkers may manifest after this age in these mice. Therefore, the second experiment was undertaken to assess the effects of a similar single icv rAAV-lep treatment on known obesity-dependent biomarkers of agingBW, FI, organ weights and circulating metabolic hormones and non-shivering thermogenic energy expenditure preceding the onset of mortality in ob/ob mice. Groups of 7–8 weeks old mice were injected with rAAV-GFP (n = 7) or rAAV-lep (n = 7) as described above. Untreated (n = 7) and PF (n = 7) groups were also run in parallel. BW and FI were monitored weekly for 10 weeks post-injection and biweekly for the remaining duration of the experiment. Average daily FI was derived for data presentation as in experiment 1. On the day of sacrifice, 30 weeks post-injection, food was removed from the cage at 07:00 h and mice were anesthetized with sodium pentobarbital (60 mg/kg; i.p.) between 12:00 and 15:00 h. A blood sample was withdrawn by orbital sinus puncture and plasma was stored frozen at −20 ◦ C for analyses of blood glucose and hormones [10,58,59]. Thereafter, mice were sacrificed by decapitation. Visceral white adipose tissue (WAT), testes, liver and brain were collected and weighed. Hypothalamus, brain stem and brown adipose tissue (BAT) were dissected out and immediately stored in RNAlater (Ambion Inc., USA) for analyses [4,5,8,22,23,58,59]. To evaluate leptin transgene expression, leptin mRNA expression was measured by RT-PCR in the hypothalamus and brainstem. To assess the effects of rAAV-lep on non-shivering thermogenic energy expenditure, UCP-1 mRNA expression was analyzed in the BAT. 2.3. Analyses Plasma leptin, ghrelin and insulin were measured by RIA kits (Linco Research, St. Charles, MO), plasma glucose levels were measured with a glucose meter (Glucometer Elite XL; Bayer, Elkhart, IN) as previously described [10,58,59]. For leptin mRNA analyses by RT-PCR, total RNA was extracted from hypothalamus and brain stem using a RNA isolation kit (Quiagen Inc., Valencia, CA) [4,5,8,22,23, 58,59]. First strand cDNA was obtained using a RNA PCR kit (reverse transcription system; Promega, Madison, WI). Primers common for mouse and rat leptin were designed to amplify a 310-bp region: sense, 5 -TGACACCAAAACCCTCATCA; antisense, 5 -ATCCAGGCTCTCTGGCTTCT. Cyclophilin, used as internal control, was generated as a 199-bp fragment with the primers: sense, 5 -ATGTGGTACGGAAGGTGGAG; antisense, 5 -TGGCTACCTTCGTCTGTGTG. For NPY mRNA analyses total RNA was extracted from the hypothalamus using RNeasy Kit (Qiagen) and treated with DNase. One microgram of RNA was reverse-transcribed using random hexamer primer Taqman reagents (PE Applied

Biosystems, Foster City, CA) according to the manufacturer’s instructions. Synthesized cDNA corresponding to 50 ng of total RNA was used for real-time quantitative PCR as described [40]. Gene-specific primers and probes for NPY (Mm00445771 m1) were purchased from PE Applied Biosystems. Taqman PCR assays for NPY gene were performed on cDNA samples in 96-well plates on an ABI Prism 7000 Sequence Detection System (PE Applied Biosystems). The 18S assays (Hs9999901 s1), designed and purchased through PE Applied Biosystems, were run in parallel to each different sample. Each 25-␮l reaction included 12.5 ␮l of Taqman 2× Master Mix (PE Applied Biosystems), 1.0 ␮l of cDNA, 2.5 ␮l of sense primer (8 ␮M), 2.5 ␮l of antisense primer (8 ␮M), 0.3 ␮l of probe (100 nM) and 6.2 ␮l of PCRgrade water. The PCR parameters were 95 ◦ C for 10 min followed by 40 cycles of 60 ◦ C for 1 min and 95 ◦ C for 15 s. To evaluate the effects on thermogenic energy expenditure, UCP-1 mRNA expression in BAT was measured as described earlier [4,5,8,22,23]. Briefly, total RNA was isolated from BAT using a RNA isolation kit (STAT-60, Teltest Inc., Friendswood, TX) and a dot-blot hybridization analysis of UCP-1 mRNA levels was performed. 2.4. Statistical analysis BW and FI were analyzed using either two-way repeated measures ANOVA with treatment and time as variables or t-test, as appropriate. Plasma leptin, insulin, ghrelin and glucose levels, hypothalamic leptin mRNA and BAT UCP-1 mRNA levels, were analyzed using one-way ANOVA and Bonferroni’s multiple comparison post hoc test or “t”-test, as appropriate. For the real-time quantitative PCR data, values normalized to 18S were compared before the data were transformed into percentage differences. Survival curves, including median life-span (50% survival), were obtained using GraphPad Prism software (GraphPad Software, San Diego, CA) and statistical significance was calculated by the logrank method. The maximal life-span (10%) was derived from the survival curves. The significance was set at p < 0.05 for all analyses.

3. Results 3.1. Effects of central leptin gene therapy on FI, BW and life-span in ob/ob mice 3.1.1. Food intake and body weight As expected [58,59], daily food consumption was not modified by rAAV-lep at the dose injected in wt mice (Fig. 1A). Cumulative FI, preceding the incidence of first natural death at the age of 100 weeks in these mice, was similar to that in control rAAV-GFP treated mice (1711.0 ± 38.9 g for the rAAV-GFP and 1739.2 ± 38.3 g for the rAAV-lep treated group). However, in ob/ob mice rAAV-lep treatment altered FI (Fig. 1B). The biphasic response resulted in 23.3%

S. Boghossian et al. / Neurobiology of Aging 28 (2007) 1594–1604

1597

Fig. 1. Effects of icv rAAV-lep on daily food intake for the lifetime of wt (A) and ob/ob mice (B). Arrows indicate the time of injection of the vectors.

lower consumption than in rAAV-GFP treated mice (Fig. 1B, p < 0.05) during the 40-week period preceding the first natural death in rAAV-GFP control mice (475.5 ± 13.1 g rAAV-lep group versus 620.6 ± 19.9 g rAAV-GFP; p < 0.05). Also, it is noteworthy that rAAV-GFP treatment did not affect FI as compared to FI in the untreated ob/ob mice (data not shown in figure for clarity). Cumulative intake during the 40-week period in rAAV-GFP and untreated groups (620.6 ± 19.9 and 671.6 ± 25.6 g, respectively) was similar. Despite the insignificant effect on FI, rAAV-lep injection in wt mice suppressed BW (Fig. 2A). rAAV-GFP treated control mice showed a gradual time-related weight gain to plateau at a 24% higher level until the age of 100 weeks, preceding onset of increased mortality (p < 0.05; Fig. 2A). This age-related weight increase during the same interval was attenuated by rAAV-lep injection. BW ranged 9% below that of the control rAAV-GFP mice (p < 0.05). In the rAAV-GFP treated and untreated control ob/ob mice, BW increased steadily to plateau at the age of 40 weeks, thereafter it fluctuated preceding the onset of mortality (Fig. 2B

Fig. 2. Lifetime effects of icv injection of rAAV-lep on body weight in wt (A), and in ob/ob mice (B) and of pair-feeding to compare with an untreated ob/ob group in (C). Mice were monitored until death resulting in variable experimental durations for the groups.

and C). On the other hand, rAAV-lep treatment suppressed the age-related weight gain in ob/ob mice (Fig. 2B). BW decreased to a nadir at about 6 weeks and these 50% lower weights (versus rAAV-GFP mice) were sustained through the lifetime. It is noteworthy that this BW of rAAV-lep ob/ob mice in the range of 30 g was similar to that of control wt mice (Fig. 2A and B) but unlike controls, these mice displayed no

1598

S. Boghossian et al. / Neurobiology of Aging 28 (2007) 1594–1604

Fig. 3. Effects on the survival rates of wt and ob/ob groups (A–C) of mice receiving treatments described for Figs. 1 and 2. Statistical significance was calculated by log-rank method. (D) The median (50% survivors) and maximal (10% survivors) life-span for the various treatment groups and the physical appearance of a representative mouse at an age close to the maximal life-span as indicated below the picture. Groups sharing similar superscript in the median life-span column are not statistically different (p > 0.05). w: weeks.

weight loss prior to death. Pair-feeding suppressed weight to a lesser extent as these mice maintained weight that was 25% below the control ob/ob mice until the age of 40 weeks and exhibited variable ranges of weight loss prior to death (Fig. 2C). 3.1.2. Life-span The results of the Kaplan-Meier analysis of survival in various wt and ob/ob groups of mice are shown in Fig. 3. In the rAAV-GFP treated wt group, the first mortality occurred at the age of 103 weeks, the maximal life-span (10% survivors) was 156 weeks and the last mouse died at the age of 165 weeks resulting in a median life-span of 131 weeks (Fig. 3A and D). This finding is consistent with previous studies reporting a median life-span of 130 weeks in C57BL/6J mice. rAAV-lep treatment did not modify this life-span in wt mice despite maintenance of 14% less BW contemporaneously with unchanged daily FI (Figs. 1 and 2). The first death in wt rAAV-lep treated mice occurred at 99 weeks of age and the longest lived mouse died at the age of 157 weeks resulting

in a median life-span of 127 weeks, similar to that observed in the control group (Fig. 3A and D). Survival profiles of rAAV-GFP treated and untreated ob/ob mice were similar (Fig. 3B and C). Median and maximum life-span in rAAV-GFP mice were 55.5 and 72.0 weeks, respectively, versus 61.0 and 79.0 weeks, respectively, for the untreated controls (Fig. 3D, p > 0.05). Thus, the vector treatment by itself did not impact longevity. On the other hand, a single icv rAAV-lep injection markedly extended longevity in these leptin-deficient mice (Fig. 3B–D). The extension of life-span was characterized by a 91.8% increase in median life-span from 55.5 weeks in rAAV-GFP to 106.5 weeks in rAAV-lep group (p < 0.05), as well as a 64.3% increase in the age of the longest lived mouse from 84 weeks in rAAV-GFP to 157 weeks in rAAV-lep group (Fig. 3B and C, p < 0.05). This rAAV-lep-induced extended life-span in ob/ob mice was about 80–85% of control wt mice. PF also extended longevity when compared to the untreated and rAAV-GFP control ob/ob mice (Fig. 3C and D). The median life-span in these mice increased by 22%,

S. Boghossian et al. / Neurobiology of Aging 28 (2007) 1594–1604

1599

from an average of 58 weeks in the two control groups to 71 weeks (p < 0.05; Fig. 3D). However, this level of beneficial effect on life-span of PF was marginal (p < 0.05) when compared to the increase in rAAV-lep treated mice. 3.2. Effects of hypothalamic leptin transgene expression and PF on FI, BW and organ weights preceding the onset of mortality in ob/ob mice 3.2.1. FI and BW Daily FI in both control groups was similar throughout the experiment (F(1, 126) = 2.13; NS). As in experiment 1, daily consumption was attenuated in a bimodal fashion by rAAV-lep injection to approximately 29% below the control range during 30 weeks post-injection (Fig. 4A). The age-related weight gain profiles of the two control groups were also similar (F(1, 126) = 3.73; NS), and each group gained an average of 25–26 g (F(20, 126) = 118.2; p < 0.05; Fig. 4B). In contrast, rAAV-lep treatment suppressed this time-related weight gain by week 12 to a nadir 62% below the controls (p < 0.05) that was sustained for the duration of the experiment (Fig. 4B). On the other hand, PF attenuated the time-dependent weight gain by an average of 10.4% as compared to controls, a response that was 58% higher than that of rAAV-lep treated mice (p < 0.05; Fig. 4B). 3.2.2. Organ weights rAAV-lep treatment suppressed WAT mass by 94% as compared to controls (Table 1 and Fig. 4C; p < 0.05). Whereas the near absence of visible visceral fat was accompanied by a significant reduction in liver weights (72–74%; Table 1 and Fig. 4C; p < 0.05) brain and testes weights were significantly increased by rAAV-lep treatment (Table 1, p < 0.05). PF also decreased WAT but to a much lesser extent (34%, p < 0.05) than that induced by rAAV-lep treatment (94%, p < 0.05; Fig. 4C and Table 1). Also, in contrast to that elicited by rAAV-lep injection, PF did not change liver, brain and testes weights. 3.2.3. Hypothalamic leptin and NPY mRNA, BAT UCP-1 mRNA expression and circulating metabolic biomarkers rAAV-lep injection markedly increased leptin mRNA expression in the hypothalamus and not in the brainstem. This increase in leptin transgene expression was simultaneously associated with a 55% suppression of hypothalamic NPY

Fig. 4. Effects on (A) FI and (B) BW of icv rAAV-lep injection and pairfeeding in ob/ob mice monitored weekly intervals for 10 weeks post-injection and biweekly, thereafter. Mice were sacrificed at 38 weeks of age (30 weeks post-injection). Arrows indicate the time of injection of the vectors. (C) Appearance of visceral fat in a representative mouse from each treatment group at termination of the experiment.

Table 1 Organ weight in ob/ob mice treated with rAAV-lep for 30 weeks Group

Treatment

WAT (g)

1 2 3 4

Untreated rAAV-GFP Pair fed rAAV-lep

12.3 13.6 9.1 0.8

* **

p < 0.05 vs. all other groups. p < 0.05 vs. untreated and rAAV-GFP control groups.

± ± ± ±

0.9 0.4 1.3** 0.6*

Brain (mg) 392.9 401.1 409.7 438.6

± ± ± ±

7.3 12.4 6.8 4.2*

Liver (g) 4.2 3.9 3.1 1.1

± ± ± ±

0.8 0.3 0.4 0.3*

Testes (mg) 172.9 166.6 155.7 211.6

± ± ± ±

9.5 6.3 6.6 6.5*

1600

S. Boghossian et al. / Neurobiology of Aging 28 (2007) 1594–1604

Fig. 5. Effects of icv rAAV-lep injection and pair-feeding on: (A) leptin mRNA expression in the hypothalamus and brainstem (inset); (B) hypothalamic NPY mRNA expression; (C) BAT UCP-1 mRNA; (D) serum glucose; (E) serum insulin; and (F) serum ghrelin levels of ob/ob mice shown in Fig. 4 at termination of experiment. * p < 0.05 vs. all groups; # p < 0.05 vs. untreated and rAAV-GFP.

mRNA expression and an increase in thermogenic energy expenditure as reflected by 20% increments in BAT UCP-1 mRNA expression (Fig. 5A–C, p < 0.05). Interestingly, both hypothalamic NPY and BAT UCP-1 mRNA expression were not affected by PF. Consistent with previous findings [10,58,59], leptin measured in peripheral circulation at 30 weeks post-rAAV injection was not detectable (data not shown). Further, whereas control groups of ob/ob mice displayed severe hyperglycemia at termination of the experiment (517.2 ± 23.6 and 488.3 ± 33.7 mg/dl, untreated and rAAV-GFP group, respectively; Fig. 5D), increased hypothalamic leptin transgene expression reduced plasma glucose levels by 60% (p < 0.05) to reinstate normoglycemia along with normalization of insulin levels resulting from a drastic suppression of the hyperinsulinemia (99 %, p < 0.05; Fig. 5E). In comparison, PF failed to reinstate either euglycemia or normoinsulinemia because circulating glucose and insulin levels were suppressed to a significantly lesser extent (glucose by 45% and insulin by 38%, p < 0.05) as compared to those in response to rAAV-lep treatment (Fig. 5D and E). Additionally, in agreement with previous reports in wt rodents [5,8,58], rAAV-lep

treatment and not PF, elevated circulating ghrelin levels by three folds in ob/ob mice (Fig. 5F, p < 0.05).

4. Discussion While corroborating previous reports that morbidly obese ob/ob mice die earlier than normal wt mice [19,32,47,62], these results document the success of increased hypothalamic leptin transgene expression in delaying increased mortality onset. A single icv rAAV-lep injection in already obese ob/ob adult mice more than doubled the life-span as compared to control cohorts, and it was only slightly shorter than that of similarly treated wt mice. Presumably, these profound life-extending benefits resulted from provision of bioactive leptin at hypothalamic target pathways in amounts insufficient to be detectable either in the hypothalamic tissue, cerebrospinal fluid or peripheral circulation [5,8,20,52]. Earlier we reported that intravenous administration of rAAV-lep in ob/ob mice evoked elevations in circulating leptin levels within the low detectable range and restored the lean wt phenotype [21], as observed here after

S. Boghossian et al. / Neurobiology of Aging 28 (2007) 1594–1604

icv administration. Thus, leptin transduced by rAAV-lep in the hypothalamus is discernible by the robust and sustained biological responses it evokes, as observed here and elsewhere [4,5,8,10,11,20,22,23,44,58,59]. Two additional lines of evidence indicate that leptin availability is confined to hypothalamus. Increased leptin mRNA expression was consistently detected in the hypothalamus surrounding the third cerebroventricle, and not caudally in extra hypothalamic sites surrounding the fourth cerebroventricle. Immunocytochemical localization of GFP after icv rAAV-GFP injection was confined to neuronal clusters in the arcuate nucleus, paraventricular nucleus, ventromedial hypothalamic and medial preoptic area, the hypothalamic sites expressing the leptin receptor and previously implicated in leptin’s control of food intake, energy expenditure, temperature regulation and neuroendocrine functions [4,5,7,8,36,37]. Apparently, in the monogenic leptin-mutant obese mice, a stable low level of leptin availability within the hypothalamus is sufficient to prevent early mortality and restore a near normal life-span. There are several mechanisms by which increased hypothalamic leptin may delay onset of increased mortality and extend life expectancy [27,30,31,47,61,65]. A rapid decline in BW and maintenance of the lean wt phenotype, largely due to fat depletion [10,11,22,23,32,37,58,59], may in itself contribute to the life prolonging benefits. The alternative idea that diminished fat accrual and/or attendant changes in adipokine secretion prolong life-span has been intensely contested over several years [9,14,18,19,27,29,31,32,47,51,62]. The possibility that reduction in weight due to decreased body fat alone may not substantially promote longevity, is indicated by our current finding that in rAAV-lep treated wt mice lifespan was unchanged. These mice maintained significantly lower weight due to diminution of body fat [10,58,59]. It is possible that normal daily FI along with a small reduction in weight and fat mass may not be adequate to prolong life-span in these mice [9,32,47,62]. Indeed, several lines of evidence suggest that restrictions on caloric intake alone influence the rate of aging and onset of age-related diseases [9,32,47,62]. Enforced calorie restriction (CR) of 40–60% on a daily basis is well known to significantly retard aging in mice and rats [47]. It is, therefore, not unreasonable that the decrease in caloric intake, independent of the attendant suppressed WAT mass, prolonged life expectancy in rAAV-lep treated ob/ob mice. Also, the fact that an equivalent CR in ob/ob mice by PF improved life-span, albeit significantly less than that elicited by rAAV-lep injection, strengthens the notion that CR may be one of the underlying causes of extending life-span. Further, non-shivering thermogenic energy expenditure, as shown by analysis of UCP-1 mRNA expression in BAT, decreases in an age-related fashion [49]. Thus, the additional highly likely possibility is that, as compared to rAAV-lep treated mice, a relatively lesser weight reduction with no concomitant increase in nonthermogenic energy expenditure in PF mice, as seen here and elsewhere [10,58,59], may underlie the shorter life-extension response.

1601

It is plausible that icv rAAV-lep interventional therapy, by delaying and/or preventing the occurrence of age-related diseases and age-related changes in physiologic functions, prolonged life [27,30,31,47,61,65]. No longitudinal pathologic analyses were conducted in the current study to address these possibilities. However, clinical surveys and animal studies have consistently implicated overt obesity and obesityassociated metabolic syndrome characterized by glucose intolerance, hyperglycemia, hyperinsulinemia, insulin resistance and dyslipidemia, as risk factors for shortening life expectancy [1,2,9,14,30,31,43,47,51,61]. Decreases in circulating levels of ghrelin, a growth hormone secretagogue and insulin-like growth factor-1 (IGF-1) are additional hormonal biomarkers of aging [37,47,54]. In the second experiment, we assessed the status of these endocrine biomarkers of aging immediately before the early onset of mortality in ob/ob mice. We observed euglycemia, normoinsulinemia and increased blood ghrelin levels in rAAV-lep ob/ob mice antecedent to increased mortality onset in control rAAVGFP cohorts. Similar beneficial responses in the periphery concomitant with increased glucose tolerance and enhanced insulin sensitivity and amelioration of dyslipidemia, were induced by icv rAAV-lep in short- and long-term experiments [4,8,10,11,23,37,39,44,58,59]. We have previously reported that circulating IGF-1 levels decrease after central rAAV-lep treatment [8,37,44]. Therefore, one can presume that the well-documented anti-aging endocrine biomarkers [2,9,29,30,37,43,44,47,54,61] – euglycemia, normoinsulinemia and elevated ghrelin and suppressed IGF-1 levels – achieved immediately after rAAV-lep injection in obese mice, persisted until natural death. We infer that this beneficial metabolic environment along with suppressed FI and BW, contributed significantly towards delaying mortality and prolonging life. Our findings are also consistent with the possibility that lesser magnitude glucose and insulin lowering responses and absence of increased ghrelin levels conferred relatively shorter longevity in PF mice. Diminution in circulating leptin levels resulting from depletion of body fat stores precedes the reduction in blood insulin levels [1,18,29,30,39,41,46]. Consequently, a precise regulatory role of leptin in insulin secretion from pancreatic-␤-cells and, vice versa, of insulin in leptin secretion from adipocytes, has been extensively examined [29,30,41,46]. However, a clear picture of the dynamic cause and effect interrelationships between these hormones within the pancreatic-adipocyte axis in the periphery has not yet emerged [41,43,46]. On the other hand, marked suppression of insulin levels in the complete absence of peripheral leptin in rAAV-lep treated ob/ob mice in this and previous studies [10,58,59], is consistent with the promulgation that one of the sites that imposes leptin restraint on insulin release resides in the hypothalamus and either sympathetic nervous system efferents or more likely separate neural links to pancreas relay this restraint [10,13,56,58,59]. Also, low blood insulin levels are generally associated with hyperglycemia [30,41,43,46]. rAAV-lep-induced

1602

S. Boghossian et al. / Neurobiology of Aging 28 (2007) 1594–1604

euglycemia contemporaneous with suppressed insulin levels in monogenic obese ob/ob and non-obese insulinopenic Akita mice was shown, in part, to result from enhanced glucose metabolism in skeletal muscles and BAT along with the increased non-shivering thermogenic energy expenditure ([59], unpublished). We observed marked upregulation of BAT UCP-1 mRNA expression implying that leptin transgene expression increased non-shivering thermogenic energy expenditure in ob/ob mice. It is, therefore, highly likely that maintenance of euglycemia, and thereby prevention of glucotoxicity [1,30,43,61], together with reversal of the age-related decline in energy expenditure [49], contributed significantly to prolonging life-span in rAAV-lep treated ob/ob mice. ob/ob mice also exhibit impaired brain development and reproductive function, fatty liver disorder, dyslipidemia and skeletal abnormalities [16,28,29,32,45,55,57]. Systemic leptin replacement increased brain mass and improved reproductive and liver functions and bone health [16,17,28,29,35,55,57]. Improvement of organ weights and amelioration of dyslipidemia and skeletal abnormalities in rAAV-lep treated mice in this and a previous study [10,35], imply that hypothalamic leptin action has the potential to reproduce some, if not all, benefits of systemic leptin replacement and that central leptin action is necessary not only in weight homeostasis but also in maintenance of other peripheral physiological, neural and neuroendocrine functions. In this context, we infer that increased leptin transgene expression can promote brain growth even in adult ob/ob mice, possibly by increasing myelinization and synaptogenesis [55]. Finally, within the hypothalamus the NPY system is the primary signaling pathway in mediating leptin-induced restraint on energy intake [33,36,38,39]. Relentless hyperphagia and increased fat accrual leading to morbid obesity in ob/ob mice has been attributed to enhanced hypothalamic NPY signaling, as indicated by upregulation of hypothalamic NPY mRNA expression and release as a result of lack of leptin restraint [36–39]. We suspect that suppression of hypothalamic NPYergic signaling by hypothalamic leptin transgene expression seen here and elsewhere [4,8,21–23] underlies the persistent suppression of FI in ob/ob mice. However, the inability of rAAV-lep treatment to extend life-span in wt mice needs explanation. We have found that doses of rAAV-lep injected either icv or directly into discrete hypothalamic sites that do not restrain FI also fail to decrease NPYergic signaling [4,5,20,22]. Therefore, future investigations will address the implicit possibility that by imposing decreases in NPYergic signaling with higher level of rAAV vector infections it would be feasible to increase life expectancy in wt mice. In conclusion, the new finding of this investigation is that the early onset of mortality and shortened life-span, characteristics of morbidly obese ob/ob mice, can be remedied by a single icv injection of rAAV-lep when adult and obese. Bioavailability of leptin transduced by ectopic leptin gene in the hypothalamus alone, is both necessary and sufficient to normalize life-span in ob/ob mice. Another distinguish-

ing outcome of this investigation is that it is now possible to engender lifetime voluntary CR and attendant normalization of weight, and extend life in ob/ob mice, as was previously conferred by enforced CR on a daily basis [32,47,62]. Further, we propose that the anti-aging biomarkers, reduced fat mass and improved glucose-insulin homeostasis and elevated ghrelin levels that persist through the lifetime of the rAAV-lep treated mice, additively assist in delaying onset of mortality and, thereby, extend life-span. In all, our results illustrate that provision of target protein leptin selectively in the hypothalamus by transfer of ectopic gene with the aid of icv injection of rAAV vectors in monogenic mutant mice is durable and safe. These vectors are amenable to similar future site specific testing of interventional gene transfer therapies, especially for those neural disorders that stem from missing or functional disruption of a single gene and adversely impact longevity.

Acknowledgements This research was supported by National Institute of Health grants DK37273 and NS 32727. We are thankful to Nicholas Cross for assistance in preparation of this manuscript.

References [1] Alquier T, Kahn BB. Peripheral signals set the tone for central regulation of metabolism. Endocrinology 2004;145:4022–4. [2] Arvanitakis Z, Wilson RS, Bienias JL, Evans DA, Bennett DA. Diabetes mellitus and risk of Alzheimer disease and decline in cognitive function. Arch Neurol 2004;61:661–6. [3] Bado A, Levasseur S, Attoub S, Kermorgant S, Laigneau JP, Bortoluzzi MN, et al. The stomach is a source of leptin. Nature 1998;394:790–3. [4] Bagnasco M, Dube MG, Kalra PS, Kalra SP. Evidence for the existence of distinct central appetite and energy expenditure pathways and stimulation of ghrelin as revealed by hypothalamic site-specific leptin gene therapy. Endocrinology 2002;143:4409–21. [5] Bagnasco M, Dube MG, Katz A, Kalra PS, Kalra SP. Leptin expression in hypothalamic PVN reverses dietary obesity and hyperinsulinemia but stimulates ghrelin. Obes Res 2003;11:1463–70. [6] Bell CG, Walley AJ, Froguel P. The genetics of human obesity. Nat Rev Genet 2005;6:221–34. [7] Beretta E, Dube MG, Kalra PS, Kalra SP. Central LIF gene therapy suppresses food intake, body weight, serum leptin and insulin for extended periods. Peptides 2002;23:875–984. [8] Beretta E, Dube MG, Kalra PS, Kalra SP. Long-term suppression of weight gain, adiposity, and serum insulin by central leptin gene therapy in prepubertal rats: effects on serum ghrelin and appetite-regulating genes. Pediatr Res 2002;52:189–98. [9] Bluher M, Kahn BB, Kahn CR. Extended longevity in mice lacking the insulin receptor in adipose tissue. Science 2003;299:572–4. [10] Boghossian S, Dube MG, Torto R, Kalra PS, Kalra SP. Hypothalamic clamp on insulin release by leptin-transgene expression. Peptides 2006;27:3245–54. [11] Boghossian S, Lecklin AH, Torto R, Kalra PS, Kalra SP. Suppression of fat deposition for the life time of rodents with gene therapy. Peptides 2005;26:1512–9. [12] Briley LP, Szczech LA. Leptin and renal disease. Semin Dial 2006;19:54–9.

S. Boghossian et al. / Neurobiology of Aging 28 (2007) 1594–1604 [13] Buijs RM, Chun SJ, Niijima A, Romijn HJ, Nagai K. Parasympathetic and sympathetic control of the pancreas: a role for the suprachiasmatic nucleus and other hypothalamic centers that are involved in the regulation of food intake. J Comp Neurol 2001;431:405–23. [14] Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med 2003;348:1625–38. [15] Carter BJ, Burstein H. Adeno-associated virus and AAV vectors for gene delivery. In: Templeton-Smith N, editor. Therapeutics mechanisms and strategies. New York, NY: Marcel Dekker; 2004. p. 71–1011. [16] Chehab FF, Lim ME, Lu R. Correction of the sterility defect in homozygous obese female mice by treatment with the human recombinant leptin. Nat Genet 1996;12:318–20. [17] Cohen P, Zhao C, Cai X, Montez JM, Rohani SC, Feinstein P, et al. Selective deletion of leptin receptor in neurons leads to obesity. J Clin Invest 2001;108:1113–21. [18] Correia MLG, Rahmouni K. Role of leptin in the cardiovascular and endocrine complications of metabolic syndrome. Diabetes Obes Metab, in press. [19] Das M, Gabriely I, Barzilai N. Caloric restriction, body fat and ageing in experimental models. Obes Rev 2004;5:13–9. [20] Dhillon H. Effects of recombinant adeno-associated virus encoding leptin on body weight regulation and energy homeostasis. University of Florida; 2005, http://purl.fcla.edu/fcla/etd/ane59481-171. [21] Dhillon H, Ge Y, Minter RM, Prima V, Moldawer LL, Muzyczka N, et al. Long-term differential modulation of genes encoding orexigenic and anorexigenic peptides by leptin delivered by rAAV vector in ob/ob mice. Relationship with body weight change. Regul Pept 2000;92:97– 105. [22] Dhillon H, Kalra SP, Kalra PS. Dose-dependent effects of central leptin gene therapy on genes that regulate body weight and appetite in the hypothalamus. Mol Ther 2001;4:139–45. [23] Dhillon H, Kalra SP, Prima V, Zolotukhin S, Scarpace PJ, Moldawer LL, et al. Central leptin gene therapy suppresses body weight gain, adiposity and serum insulin without affecting food consumption in normal rats: a long-term study. Regul Pept 2001;99:69–77. [24] Dhillon H, Zigman JM, Ye C, Lee CE, McGovern RA, Tang V, et al. Leptin directly activates SF1 neurons in the VMH, and this action by leptin is required for normal body-weight homeostasis. Neuron 2006;49: 191–203. [25] Farooqi IS, Matarese G, Lord GM, Keogh JM, Lawrence E, Agwu C, et al. Beneficial effects of leptin on obesity, T cell hyporesponsiveness, and neuroendocrine/metabolic dysfunction of human congenital leptin deficiency. J Clin Invest 2002;110:1093–103. [26] Farooqi IS, O’Rahilly S. Monogenic obesity in humans. Annu Rev Med 2005;56:443–58. [27] Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths associated with underweight, overweight, and obesity. JAMA 2005;293:1861–7. [28] Friedman JM, Halaas JL. Leptin and the regulation of body weight in mammals. Nature 1998;395:763–70. [29] Fruhbeck G. Peripheral actions of leptin and its involvement in disease. Nutr Rev 2002;60. S47–55, S68–87. [30] Grundy SM. Obesity, metabolic syndrome, and cardiovascular disease. J Clin Endocrinol Metab 2004;89:2595–600. [31] Hadley EC, Lakatta EG, Morrison-Bogorad M, Warner HR, Hodes RJ. The future of aging therapies. Cell 2005;120:557–67. [32] Harrison DE, Archer JR, Astle CM. Effects of food restriction on aging: separation of food intake and adiposity. Proc Natl Acad Sci USA 1984;81:1835–8. [33] Horvath TL, Diano S, Tschop M. Brain circuits regulating energy homeostasis. Neuroscientist 2004;10:235–46. [34] Inui A. Transgenic approach to the study of body weight regulation. Pharmacol Rev 2000;52:35–61. [35] Iwaniec UT, Boghossian S, Lapke PD, Turner RT, Kalra SP. Central leptin gene therapy corrects skeletal abnormalities in leptin-deficient ob/ob mice. J Bone Miner Res, in press.

1603

[36] Kalra SP, Dube MG, Pu S, Xu B, Horvath TL, Kalra PS. Interacting appetite-regulating pathways in the hypothalamic regulation of body weight. Endocr Rev 1999;20:68–100. [37] Kalra SP, Kalra PS. Gene transfer technology: a preventive neurotherapy to curb obesity, ameliorate metabolic syndrome and extend lifeexpectancy. Trends Pharmacol Sci 2005;26:488–95. [38] Kalra SP, Kalra PS. Neuropeptide Y: a conductor of the appetiteregulating orchestra in the hypothalamus. In: Kastin AJ, editor. The handbook of biologically active peptides. Burlington, MA: Academic Press; 2006. p. 909–14. [39] Kalra SP, Kalra PS. Subjugation of hypothalamic NPY and cohorts with central leptin gene therapy alleviates dyslipidemia, insulin resistance, and obesity for life-time. In: Zukowska Z, Feuerstein G, editors. NPY family of peptides in neurobiology, cardiovascular and metabolic disorders: genes, functions and therapeutics. Switzerland: Birhauser Verlag Publishing House; 2006. p. 157–69. [40] Keen-Rhinehart E, Kalra SP, Kalra PS. AAV mediated leptin receptor installation improves energy balance and the reproductive status of obese female Koletsky rats. Peptides 2005;26:2567–78. [41] Kieffer TJ, Habener JF. The adipoinsular axis: effects of leptin on pancreatic B-cells. Am J Physiol Endocrinol Metab 2000;278:E1– 14. [42] La Cava A, Alviggi C, Matarese G. Unraveling the multiple roles of leptin in inflammation and autoimmunity. J Mol Med 2004;82:4–11. [43] Lazar MA. How obesity causes diabetes: not a tall tale. Science 2005;307:373–5. [44] Lecklin AH, Dube MG, Torto R, Kalra PS, Kalra SP. Perigestational suppression of weight gain with central leptin gene therapy results in lower weight F1 generation. Peptides 2005;26:1176–87. [45] Leibel RL, Chung WK, Chua Jr SC. The molecular genetics of rodent single gene obesities. J Biol Chem 1997;272:31937–40. [46] Lewis GF, Carpentier A, Adeli K, Giacca A. Disordered fat storage and mobilization in the pathogenesis of insulin resistance and type 2 diabetes. Endocr Rev 2002;23:201–29. [47] Masoro EJ. Overview of caloric restriction and ageing. Mech Ageing Dev 2005;126:913–22. [48] Matarese G. Leptin and the immune system: how nutritional status influences the immune response. Eur Cytokine Netw 2000;11:7– 14. [49] McDonald RB, Horwitz BA. Brown adipose tissue thermogenesis during aging and senescence. J Bioenerg Biomembr 1999;31:507– 16. [50] Montague CT, Farooqi IS, Whitehead JP, Soos MA, Rau H, Wareham NJ, et al. Congenital leptin deficiency is associated with severe earlyonset obesity in humans. Nature 1997;387:903–8. [51] Olshansky SJ, Passaro DJ, Hershow RC, Layden J, Carnes BA, Brody J, et al. A potential decline in life expectancy in the United States in the 21st century. N Engl J Med 2005;352:1138–45. [52] Otukonyong EE, Dube MG, Torto R, Kalra PS, Kalra SP. Central leptin differentially modulates ultradian secretory patterns of insulin, leptin and ghrelin independent of effects on food intake and body weight. Peptides 2005;26:2559–66. [53] Pu S, Dube MG, Kalra PS, Kalra SP. Regulation of leptin secretion: effects of aging on daily patterns of serum leptin and food consumption. Regul Pept 2000;92:107–11. [54] Rigamonti AE, Pincelli AI, Corra B, Viarengo R, Bonomo SM, Galimberti D, et al. Plasma ghrelin concentrations in elderly subjects: comparison with anorexic and obese patients. J Endocrinol 2002;175:R1–5. [55] Steppann CM, Swick AG. A role for leptin in brain development. Biochem Biophys Res Commun 1999;256:600–2. [56] Sved AF, Cano G, Card JP. Neuroanatomical specificity of the circuits controlling sympathetic outflow to different targets. Clin Exp Pharmacol Physiol 2001;28:115–9. [57] Thomas T. The complex effects of leptin on bone metabolism through multiple pathways. Curr Opin Pharmacol 2004;4:295–300. [58] Ueno N, Dube MG, Inui A, Kalra PS, Kalra SP. Leptin modulates orexigenic effects of ghrelin and attenuates adiponectin and insulin

1604

S. Boghossian et al. / Neurobiology of Aging 28 (2007) 1594–1604

levels and selectively the dark-phase feeding as revealed by central leptin gene therapy. Endocrinology 2004;145:4176–84. [59] Ueno N, Inui A, Kalra SP, Kalra PS. Leptin transgene expression in the hypothalamus enforces euglycemia in diabetic, insulin-deficient nonobese Akita mice and leptin-deficient obese ob/ob mice. Peptides 2006;27:2332–42. [60] Ueno N, Oh-ishi S, Segawa M, Nishida M, Fukuwatari Y, Kizaki T, et al. Effect of age on brown adipose tissue activity in the obese (ob/ob) mouse. Mech Ageing Dev 1998;100:67–76. [61] Unger RH. Longevity, lipotoxicity and leptin: the adipocyte defense against feasting and famine. Biochimie 2005;87:57–64.

[62] Weindruch R, Sohal RS. Seminars in medicine of the Beth Israel Deaconess Medical Center. Caloric intake and aging. N Engl J Med 1997;337:986–94. [63] Wiesner G, Vaz M, Collier G, Seals D, Kaye D, Jennings G, et al. Leptin is released from the human brain: influence of adiposity and gender. J Clin Endocrinol Metab 1999;84:2270–4. [64] Wilkinson M, Morash B, Ur E. The brain is a source of leptin. Front Horm Res 2000;26:106–25. [65] Yu BP, Masoro EJ, Murata I, Bertrand HA, Lynd FT. Life span study of SPF Fischer 344 male rats fed ad libitum or restricted diets: longevity, growth, lean body mass and disease. J Gerontol 1982;37:130–41.