Rapamycin extends life span of Rb1^+/− mice by inhibiting neuroendocrine tumors

eRapa extended life span of Rb1^+/− mice

Unlike most mouse models of cancer [5], 50% DR had little (if any) effect on the development, growth and progression of neuroendocrine tumors or on life span ofRb1^+/− mice [20]. Since rapamycin has been predicted to act in a similar way to DR [16], we investigated if eRapa would also have little effect in this model. In stark contrast to DR, Figure 1A shows that Rb1^+/− males and females derive a significant longevity benefit from chronic treatment with eRapa. The Eudragit control-fed mice had a shorter mean life span than the eRapa-fed cohort for both females (377.5 versus 411 days) and males (mean age is 368.8 versus 419.8 days). Sex did not modulate the effect of eRapa on Rb1^+/− animals (Table 1).

Figure 1. Survival plots for male and female Rb1^+/− (A) and Rb1^+/+ (B) mice, comparing control-fed mice to those fed eRapa in the diet starting at approximately 9 weeks of age (indicated by arrow). Control (black line) and eRapa (red line) survival curves are shown. The horizontal axes represent life span in days and the vertical axes represent survivorship. Rb1^+/− mice obtained from the NCI Mouse Repository were bred by the Nathan Shock animal core to obtain the cohorts of male and female mice used in this study. Genotype was confirmed as previously described [20]. eRapa mice were fed microencapsulated rapamycin-containing food (14mg/kg food designed to deliver approximately 2.24mg of rapamycin per kg body weight/day that achieved about 4 ng/ml blood [14]. Diets were prepared by TestDiet, Inc., Richmond, IN using Purina 5LG6 as the base [14]. Control diet was the same but with empty capsules. P values in (B) were calculated by the log-rank test.

Male and female Rb1^+/+ littermates of the Rb1^+/− mice were also fed eRapa or control diets to ensure that this particular mutant strain (with a C57BL/6 background) is responsive to rapamycin. Once all Rb1^+/− mice had died and the effects of eRapa were evident, the Rb1^+/+ littermates were euthanized. At this time, as expected, eRapa improved survival for both male and female Rb1^+/+ mice as well (Fig. 1B). Similar to the previous results from the Intervention Testing Program eRapa experiments [16, 17], lifespan was extended more in females than in males (Table 2) in wild type (WT) littermates.

eRapa effects on tumor incidence at the end of life

At necropsy, Rb1^+/− mice were evaluated for the presence of neuroendocrine tumors and lung metastases. As shown in Table 3, there were no differences in the eRapa and Eudragit control groups in terms of presence of pituitary tumors (although we did observe a delay in their detection and reduction in size by magnetic resonance imaging (MRI), discussed below). We did observe a decreased incidence of thyroid C-cell carcinomas in the eRapa treated group of Rb1^+/− mice (p = 0.0112). Except for the modest decrease in thyroid tumors, this tumor spectrum is similar to Rb1 heterozygotes treated with DR compared to those fed ad libitum [20]. Along with the decrease in thyroid C-cell tumors, eRapa also tended to reduce the incidence and severity of C-cell lung metastases (Table 4). Thus mice have a decreased cancer burden and live with tumors longer.

eRapa delayed tumor development and slowed growth

Is delayed and/or reduced tumor growth the basis of life span extension by eRapa in this model? To address this question, we took advantage of the synchronous (spatial and temporal) development of tumors in this model Nikitin et al. [19, 21]. Rb1-deficient cells are first identified as atypical proliferates in the intermediate and anterior lobes of the pituitary, thyroid and parathyroid glands and the adrenal medulla at about 12 weeks of postnatal development. Atypical proliferates eventually form gross tumors with varying degrees of malignancy by postnatal day 350. Since we started treatment at around 8 weeks of age, eRapa might have an effect on the initiating events leading to loss of heterozygosity and/or subsequent formation of atypically proliferating cells. Perhaps more likely, eRapa slows growth and development of proliferates to gross tumors, which had probably begun at or around the time treatment was started. To test this latter possibility, we used MRI to follow pituitary and thyroid tumor development and growth in a subset of eRapa-treatedRb1^+/− mice (8 mice per treatment group were imaged between 1 and 4 times up to twice a month). MRI is well suited for following head and neck tumors that correspond to the primary tumor types Rb1^+/− mice develop. An initial cohort was used to identify the best timeframe for MRI scans. For this, 6 female Rb1^+/+ mice (3 per group) and 10 Rb1^+/− mice (3 per group in males and 2 per group in females) were imaged in a single session or with 2 serial scans. This study indicated the ideal timeframe to image pituitary tumors was a window between 9 and 12 months of age, which covers the time from initial detection through monitoring tumor growth.

Age matched Rb1^+/− females (3 per group) were scanned using MRI at 9, 11 and 12 months of age (Figure 2A shows sagittal plane sections of the serially acquired MRI images through the pituitary of eRapa and Eudragit treated mice). Calculated volumes based on the MRI image stacks (analyzed blind by a single radiologist, RLH) were plotted versus age at the date of imaging. In concert with extended longevity, the detection of pituitary tumors was delayed with a decrease in their growth in the eRapa-treated mice. Figure 2B shows that eRapa delayed development and/or reduced tumor growth at each time point when mice were imaged. More Rb1^+/− mice had detectable tumors identified during two separate MRI imaging sessions from the Eudragit control cohort (4 pituitary and 2 thyroid tumors out of 8 mice in March 2011 scan and 7 pituitary and 4 thyroid tumors out of 8 mice in April 2011 scan) compared to the mice eRapa-fed cohort (1 pituitary and 0 thyroid tumors out of 8 mice in March 2011 scan and 2 pituitary and 3 thyroid tumors out of 8 mice in April 2011 scan). Longitudinal monitoring allowed us to conclude that chronic rapamycin delays both the development of visible tumors and inhibited the growth of tumors once they were present.

Figure 2. Effects of eRapa on pituitary and thyroid tumor development and growth. To identify effects on tumors, we used MRI as a non-invasive method to longitudinally monitor individual Rb^+/− mice. High-resolution images were obtained on a very high field strength Bruker Pharmascan 7.0T animal MRI scanner using a coil to focus on pituitary and thyroid tumors. Images were acquired using a spoiled gradient echo named Fast low angle shot MRI (FLASH) on the scanner. Images were acquired to yield predominantly T1 weighted contrast with TE (echo time) 4.5 msec, TR (repetition time) 450 msec, FA (Flip angle) 40 degrees, FOV (field of view) 20 x 20 mm, in plane spatial resolution 0.078 x 0.078 mm. Tumor volume was determined for each time point. (A) Serially acquired MRI images from eRapa and Eudragit-fed control mice at 9, 11 and 12 months of age. (B) Tumor volumes calculated from MRI image stacks at each time point comparing individual mice at multiple ages. Tumors in two of the Eudragit-fed (control) mice are detected earlier and grow faster than the 3 eRapa-fed mice.

Mice and life span

Mice (strain B6.129S2(Cg)-Rb1^tm1Tyj) for breeding were obtained from the NCI MMHCC Repository. Although they have similar phenotypes, the strain used in the diet restriction study by Sharp et al., [20] was different having been generated by Lee et al [38]. The procedures and experiments involving use of mice were approved by the Institutional Animal Care and Use Committee and are consistent with the NIH Principles for the Utilization and Care of Vertebrate Animals Used in Testing, Research and Education, the Guide for the Care and Use of Laboratory Animals and the Animal Welfare Act (National Academy Press, Washington, DC). Genotyping was done as described previously [20]. Cohorts of mice were fed microencapsulated rapamycin-containing food (14 mg/kg food designed to deliver ~2.24 mg of rapamycin per kg body weight/day to achieve about 4 ng/ml of rapamycin per kg body weight/day) prepared by TestDiet, Inc., Richmond, IN using Purina 5LG6 as the base [14]. Control diet was the same but with empty capsules.

Rapamycin food concentration

Rapamycin was quantified in food using HPLC with tandem mass spectrometry detection. Briefly, 100 mg of food for spiked calibrators and unknown samples were crushed with a mortar and pestle, then vortexed vigorously with 10 μL of 250 μg/mL ASCO (internal standard) and 4.0 ml of mobile phase A. The samples were then mechanically shaken for 10 min, centrifuged for 10 min, and then centrifuged in microfilterfuge tubes for 1 minute. Ten μL of the final extracts was injected into the LC/MS/MS. The ratio of the peak area of rapamycin to that of the internal standard (response ratio) was compared against a linear regression of calibrator response ratios at rapamycin concentrations of 0, 2, 5, 10, 30, and 60 ng/mg of food to quantify rapamycin. The concentration of rapamycin in food was expressed as ng/mg food (parts per million).

Rapamycin blood measurements

Measurement of rapamycin used HPLC-tandem MS. RAPA and Ascomycin (ASCO) were obtained from LC Laboratories (Woburn, MA). HPLC grade methanol and acetonitrile were purchased from Fisher (Fair Lawn, NJ). All other reagents were purchased from Sigma Chemical Company (St. Louis, MO). Milli-Q water was used for preparation of all solutions. RAPA and ASCO super stock solutions were prepared in methanol at a concentration of 1 mg/ml and stored in aliquots at −80°C. A working stock solution prepared each day from the super stock solutions at a concentration of 10 μg/ml was used to spike the calibrators.

Calibrator and unknown whole blood samples (100 μL) were mixed with 10 μL of 0.5 μg/mL ASCO (internal standard), and 300 μL of a solution containing 0.1% formic acid and 10 mM ammonium formate dissolved in 95% HPLC grade methanol. The samples were vortexed vigorously for 2 min, and then centrifuged at 15,000 g for 5 min at 23°C (subsequent centrifugations were performed under the same conditions). Supernatants were transferred to 1.5 ml microfilterfuge tubes and centrifuged at 15,000 g for 1 min and then 40 μL of the final extracts were injected into the LC/MS/MS. The ratio of the peak area of rapamycin to that of the internal standard ASCO (response ratio) for each unknown sample was compared against a linear regression of calibrator response ratios at 0, 1.25, 3.13, 6.25, 12.5, 50, and 100 ng/ml to quantify rapamycin.

The HPLC system consisted of a Shimadzu SCL-10A Controller, LC-10AD pump with a FCV-10AL mixing chamber (quarternary gradient), SIL-10AD autosampler, and an AB Sciex API 3200 tandem mass spectrometer with turbo ion spray. The analytical column was a Grace Alltima C18 (4.6 x 150 mm, 5 μ) purchased from Alltech (Deerfield, IL) and was maintained at 60°C during the chromatographic runs using a Shimadzu CTO-10A column oven. Mobile phase A contained 10 mM ammonium formate and 0.1% formic acid dissolved in HPLC grade methanol. Mobil phase B contained 10 mM ammonium formate and 0.1% formic acid dissolved in 90% HPLC grade methanol. The flow rate of the mobile phase was 0.5 ml/min. Rapamycin was eluted with a step gradient. The column was equilibrated with 100% mobile phase B. At 6.10 minutes after injection, the system was switched to 100% mobile phase A. Finally, at 15.1 min, the system was switched back to 100% mobile phase B in preparation for the next injection. The rapamycin transition was detected at 931.6 Da (precursor ion) and the daughter ion was detected at 864.5 Da. ASCO was detected at 809.574 Da and the daughter ion was 756.34 Da.

Survival Analysis Methods

An entry for each mouse in the study was created in a database used by the Nathan Shock Animal core. The age at which each animal died was recorded. Survival durations for animals that either lived past the end-date of the study, were terminated, or died accidentally were treated as right-censored events. Cox proportional hazard models [39] were fitted to the wild type and Rb1^+/− subsets of the data, with eRapa and gender as additive predictor variables. Some animals were transferred to a different facility part-way through their life spans so the final facility at which they were housed was also added to the Cox models, as a stratifying variable. The R statistical language was used for the analysis [40, 41]. The mice in the life span studies were allowed to live out their life span, i.e., there was no censoring due to morbidity in the groups of mice used to measure lifespan of Rb1^+/− mice. Mice were euthanized only if they were either [1] unable to eat or drink, [2] bleeding from a tumor or other condition, or [3] when they were laterally recumbent, i.e., they fail to move when prodded or are unable to right themselves.

MRI Methods

Images were acquired on a Bruker Pharmascan 7.0T MRI scanner. Images were obtained in the sagittal plane through the brain and coronal plain through the neck (focused on the thyroid gland) using 2D spoiled gradient echo technique to quickly obtain high-resolution images (fast low angle shot magnetic resonance imaging - FLASH on our scanner). FLASH protocol was TE/TR 5 msec/450msec, Averages 1, Flip angle 40 deg, Field of view 20 mm x 20 mm, matrix size 256x256, In plane resolution was 0.078 x 0.078 mm, slice thickness 0.5 mm. The FLASH sequence shows predominantly T1 weighted image contrast. A single blinded radiologist (RLH) evaluated images for the presence and tumor volume used to plot detection and growth data. Images were analyzed using an open source image processing software, OsiriX, version 2.7.5. The pituitary gland was identified on all images and volume was calculated by measuring the greatest anterior-posterior, cranial-caudal, and right-left length. Volumes were then determined using prolate ellipse formula. Data were then parsed by treatment group and plotted in Prism (GraphPad).

Procedures for examination of pathology in mice

Fixed tissues (in 10% neutralized formalin) were embedded in paraffin, sectioned at 5 μm, and stained with hematoxylin-eosin. Diagnosis of each histopathological change was made using histological classifications for aging mice as previously described [9, 20, 42, 43].

Pathology assessments

A list of lesions was compiled for each mouse. The severity of neoplastic lesions was assessed using the grading system previously described [9, 9, 20, 42, 43, 42, 43]. Two pathologists separately examined all of the samples without knowledge of their genotype or age. Briefly, lung pathology grade is based on the area of the lung section infiltrated by metastatic tumor tissue with 0 being no tumor cells observed and 4 being the largest area taken by tumor.