Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Effects of sirolimus on plasma lipids, lipoprotein levels, and fatty acid metabolism in renal transplant patients

Effects of sirolimus on plasma lipids, lipoprotein levels, and fatty acid metabolism in renal... Effects of sirolimus on plasma lipids, lipoprotein levels, and fatty acid metabolism in renal transplant patients 1, Joel D. Morrisett, * Ghada Abdel-Fattah,* Ron Hoogeveen,* Eddie Mitchell,* Christie M. Ballantyne,* Henry J. Pownall,* Antone R. Opekun,* Jonathon S. Jaffe, † † Suzanne Oppermann, and Barry D. Kahan The Department of Medicine,* Baylor College of Medicine, Houston, Texas; Department of Surgery, University of Texas-Houston Health Science Center, Houston, TX; and Wyeth Ayerst Laboratories, Philadelphia, PA Abstract Sirolimus (Rapammune , rapamycin, RAPA) is a lipoprotein levels, and fatty acid metabolism in renal trans- potent immunosuppressive drug that reduces renal trans- plant patients. J. Lipid Res. 2002. 43: 1170–1180. plant rejection. Hyperlipidemia is a significant side effect of sirolimus treatment, and frequently leads to cardiovascu- Supplementary key words rapamycin • triglyceride • cholesterol lar disease. This study was undertaken to determine the repeatability, reversibility, and dose dependence of the plasma lipid and apolipoprotein altering effects of siroli- Sirolimus (Rapammune , rapamycin, RAPA) is a novel mus, and to elucidate the mechanism by which sirolimus in- macrocyclic lactone immunosuppressive drug capable of duces hypertriglyceridemia in some renal transplant pa- significantly reducing acute graft rejection in kidney (1), tients. Six patients with renal allografts maintained on cyclosporine A and prednisone were selected on the basis liver (2), and heart (3) transplant patients. Previous stud- of their previous hyperlipidemic response to short term (14 ies have shown that sirolimus reduces the incidence of days) sirolimus administration. For longer-term treatment, acute rejection when administered in conjunction with cy- each patient was started on 10 mg/day sirolimus and contin- closporine and prednisone (1). Furthermore, sirolimus in- ued as tolerated for 42 days to reinduce hyperlipidemia. hibits vascular smooth muscle cell proliferation and re- Timed blood samples were analyzed for lipid, apolipopro- duces neointimal formation in humans, rats, and pigs, tein, and sirolimus levels. During sirolimus administration, thereby attenuating restenosis following angioplasty (4–6). mean total plasma cholesterol increased from 214 mg/dl to Sirolimus binds to the immunophilin FK506 binding 322 mg/dl (50%; range 25–92%); LDL-cholesterol levels protein (FKBP12). Sirolimus/FKBP12 binary complex followed a similar pattern. Mean triglyceride level rose from 227 to 432 mg/dl (95%; range 9–254%). ApoB-100 does not bind to calcineurin, and therefore is not neuro- concentration rose from 124 to 160 mg/dl (28%; P toxic or nephrotoxic. Instead, sirolimus/FKBP12 binds to 0.05). ApoC-III level increased from 28.9 to 55.5 mg/dl, a protein kinase called mammalian target of rapamycin 92%; (P  0.013). These lipid and apolipoprotein changes (mTOR). mTOR controls proteins that regulate mRNA were found to be repeatable, reversible, and dose depen- translation initiation and G1 progression (7). Recent stud- dent. [ C ]palmitate metabolic studies in four patients with ies have shown that mTOR directly phosphorylates p70S6 hypertriglyceridemia indicated that the free fatty acid pool kinase (8), the eukaroytic translation initiator protein 4G1 was expanded by sirolimus treatment (mean  42.3%). In- (eIF4G1), and translation inhibitor (4E-BP1) (8–10). corporation of [ C ]palmitate into triglycerides of VLDL, IDL, and LDL was decreased 38.3%, 42,1%, and 38.4%, re- Therefore, inhibition of mTOR by sirolimus contributes spectively, by sirolimus treatment of these patients. to translational arrest by down-regulation of p70S6K, and These results suggest that sirolimus alters the insulin signal- by increasing the affinity of 4E-BP1 (11). Consequently, ing pathway so as to increase adipose tissue lipase activity sirolimus immunosuppressive action is due to inhibition and/or decrease lipoprotein lipase activity, resulting in in- of T-cell activation at a later stage of the cell cycle, G1, and creased hepatic synthesis of triglyceride, increased secre- inhibition of p70S6K (10). tion of VLDL, and increased hypertriglyceridemia.—Morri- A major adverse reaction associated with sirolimus ther- sett, J. D., G. Abdel-Fattah, R. Hoogeveen, E. Mitchell, C. M. Ballantyne, H. J. Pownall, A. R. Opekun, J. S. Jaffe, S. Opper- mann, and B. D. Kahan. Effects of sirolimus on plasma lipids, Abbreviations: CsA, cyclosporine A; FKBP, FK506 binding protein; PFB, pentafluorobenzyl; WAS-#/#, Wyeth Ayerst study patient before/ after sirolimus treatment. Manuscript received 6 November 2001 and in revised form 29 April 2002. To whom correspondence should be addressed. DOI 10.1194/jlr.M100392-JLR200 e-mail: [email protected] Copyright © 2002 by Lipid Research, Inc. 1170 Journal of Lipid Research Volume 43, 2002 This article is available online at http://www.jlr.org This is an Open Access article under the CC BY license. suitability for the study. Blood sampling for the lipid profile and apy is hyperlipidemia, a major risk factor for cardiovascu- lipolytic enzymes (e.g., post heparin lipase) were scheduled to lar disease, and the most common cause of death after re- avoid interference of one test with another. The initial shorter- nal transplantation (12). Several studies have shown an term study (14 days) was conducted immediately after transplan- increase in serum triglyceride levels in renal transplant re- tation. Each patient was treated with a constant dose (1–7 mg/ cipients treated with sirolimus (13, 14). Their hyperlipi- day) of sirolimus over a 14 day period (Table 2). demia was dose-dependent and reversible within 1 to 2 For the longer-term study (42 days), qualifying patients began months after discontinuation of treatment (13, 14). the 8-week protocol on day 6 with a pre-drug lipoprotein meta- In the present study, we have examined the depen- bolic study lasting 6 days, up to day 1. On day 1, each patient dence of lipid, lipoprotein, and apolipoprotein levels, as started the sirolimus treatment initially at a level of 10 mg/day for 42 days. While on sirolimus, the patient returned weekly or well as fatty acid and triglyceride metabolism on sirolimus biweekly to the outpatient center for determination of lipids, li- dosage and treatment duration in renal allograft recipi- poproteins, apolipoproteins, lipid enzymes, and sirolimus trough ents with different types of hyperlipidemia. levels. If the patient’s lipid levels exceeded an acceptable range, then the sirolimus dose was reduced as described previously (15). After 42 days on treatment, a second lipoprotein metabolic METHODS study was initiated, lasting 6 days until day 47, after which siroli- mus treatment was discontinued. Cyclosporin and prednisone Patient selection maintenance therapy were continued. Each patient returned to the outpatient center on day 56 to give another fasting follow-up This study was performed in six patients, each of whom had blood sample for determination of the final lipid and lipopro- received a renal allograft within 3–8 years at Hermann Hospital tein profile, and to undergo the closeout physical examination. Transplant Center, Houston, Texas. These patients were selected based on their previous hyperlipidemic response upon short- term treatment (14 days) with sirolimus and had stable renal al- Sirolimus measurements lografts. All patients selected had no evidence of hepatic or bil- Sirolimus trough levels were measured on whole blood sam- iary dysfunction, as reflected in serum transaminase levels not ples with a multi-step liquid-liquid extraction followed by re- more than 20% above normal limits, nor lipid abnormalities with versed-phase-HPLC with ultraviolet detection performed by Dr. triglycerides 400 mg/dl or cholesterol 250 mg/dl. Patients Kim Napoli at the Organ Transplantation Center of the Univer- indicated their willingness to participate in the study by signing a sity of Texas Health Science Center at Houston (16). consent form, approved by the Institutional Review Boards for human research at the University of Texas Health Science Cen- Lipid and apolipoprotein measurements ter-Houston and Baylor College of Medicine and its affiliated Lipid, lipoprotein, and apolipoprotein measurements were hospitals. The patient group included four females and two performed in the Atherosclerosis Lipid Laboratory of The Meth- males aged 27–55 years (44.8  10.6) (Table 1). All six patients odist Hospital. Plasma samples were prepared by centrifugation were maintained on cyclosporine A (CsA) (Neoral), prednisone, (1500 g, 10 min, 4C) of venous blood collected after 12 h fasting and diuretic therapy. Four patients had developed mixed hyper- tm into Vacuutainer tubes containing EDTA. Total plasma choles- lipidemia, one had developed hypercholesterolemia, and one terol (17) and triglycerides (18) were measured enzymatically had developed hypertriglyceridemia in response to the previous (Boehringer Mannheim Diagnostics). LDL cholesterol (LDL-C) short-term sirolimus treatment (Table 2). Patients who had been levels were determined directly from the plasma after immuno- diagnosed with diabetes were receiving lipid-lowering medica- precipitation of VLDL and HDL using a kit from Sigma Chemi- tions (Table 1). This regimen was continued unaltered while on cal Co. (St. Louis, MO). HDL-C was determined by measuring sirolimus during the longer 6-week study (42 days). All six pa- cholesterol in the supernatant liquid after precipitation of the tients had the apo-E3/E3 genotype. VLDL and LDL with MgCl and dextran sulfate (19). Plasma apoB-100 was measured by ELISA using Mab RP-066 (Intracel, Study design Inc., Rockville, MD). ApoA-I was measured by nephelometry of Two weeks before commencing the study, each candidate pa- the precipitate formed with anti-apoA-I (IncStar, Inc.). ApoC-II, tient underwent screening evaluation at the Hermann Hospital and apoC-III were determined by radial immunodiffusion (Dai- Transplant Center. A complete physical examination and a series chi, Ltd.). ApoE genotyping was performed using a PCR based of biochemical screening tests were performed to determine method (20). TABLE 1. Clinical and demographic data of the renal transplant re-challenge (42 days) study participants Prednisone Creatinine Before Creatinine After ID Age Gender Race Donor Type Lipid RX Neoral (N) Dose Dose Sirolimus Sirolimus Diabetes Diagnosis mg mg/dl WAS-1/2 27 F Hispanic LRD None 100/75 7.5 1.7 2.5 No ESRD/GN WAS-3/4 55 F Black Cad P 125/100 7.5 0.9 0.9 IDDM/SI ESRD/HTN WAS-5/6 53 F Hispanic LRD L, P 125/125 5 0.7 0.6 NIDDM ESRD/HTN WAS-7/8 49 F Hispanic LRD L, P 125/100 7.5 1.5 1.1 IDDM/SI ESRD/CP WAS-9/10 37 M White LRD None 100/100 5 1.7 3.1 No ESRD/Lupus WAS-11/12 48 M Hispanic LRD L, P 200/175 7.5 1.6 1.7 NIDDM ESRD/HTN LRD, live related donor; P, pravacol; L, lopid; ESRD, end stage renal disease; GN, gynecological disease; HTN, hypertension; CP, chronic pyelo- nephritis; IDDM/SI, insulin dependent diabetes mellitus/steroid induced; Cad, cadaveric. Dose changed to 75/50 by the end of sirolimus treatment. Morrisett et al. Sirolimus and lipids in renal transplant patients 1171 TABLE 2. Plasma cholesterol and trigylceride response of renal transplant recipients to the initial shorter term sirolimus treatment (14 days) Cholesterol Level Before/ Triglyceride Level Before/ After Sirolimus After Sirolimus Hyperlipidemia Day 14 Day 14 b b Patient ID Sirolimus Dose Day 1 Stop Drug Day 28 Day 56 Change Day 1 Stop Drug Day 28 Day 56 Change Before After mg/day mg/dl % mg/dl % a a WAS-1/2 2 264 416 ND ND 57 169 465 ND ND 175 II a II b WAS-3/4 1 294 246 337 300 16 123 294 211 163 139 II a II b WAS-5/6 6 249 373 293 ND 49 739 1709 652 ND 131 II b II b WAS-7/8 6 333 399 293 252 20 248 628 313 155 153 II b II b WAS-9/10 7 148 241 191 180 62 156 340 162 137 118 N II b WAS-11/12 2 318 310 341 303 3 315 392 475 298 24 II b II b Mean 28 123 P value 0.1 0.04 ND, not determined. Percent change is the difference between day –1 and day 14 (when sirolimus treatment was stopped). At this time, patient WAS-1/2 was on sirolimus treatment for 61 days. Sirolimus treatment was started one day after renal transplantation. Metabolic studies Log or rank transformations were utilized when needed to meet Patients were fasted overnight prior to the start of their meta- the assumptions of the t-test. Statistical analyses were conducted bolic studies (15). Sodium [ C ]palmitate (Isotec, Inc., Miamis- using STATA (Release 4.0) and Prism (version 2.0) software. burg, OH) complexed to human serum albumin (Centeon, LLC. Plots of percent atom enrichment before and after sirolimus Kankakee, IL) was administered by constant intravenous infusion treatment were quantitatively compared by calculating the area (0.6 mg/kg/h) over 7 h. Each patient was given oral Sustecal (30 under the curve out to 36 h. The first and last data points were kcal/kg), which was consumed in 16 equal portions at hourly inter- used to determine the baseline amplitude, which was used to cor- vals, providing 22% of calories from fat and 0.88 g protein/kg. rect the integrated area. Blood samples (15 ml) were drawn, 18 over the first 24 h and 1 daily for the next 5 days, from which VLDL, IDL, and LDL were isolated by density gradient ultracentrifugation (21). These lipoproteins RESULTS were delipidated by organic solvent extraction (CHCl -CH OH, 2:1, 3 3 v/v) and the different lipid fractions separated by thin layer chro- A major objective of the current study was to determine matography (22). The triglyceride fraction was hydrolyzed with 15% if sirolimus-induced hyperlipidemia in renal transplant KOH and derivatized with pentafluorobenzylbromide (23). Plasma patients is reproducible, reversible, and dose-dependent. free fatty acids (FFAs) were isolated by solid phase extraction col- Although the shorter-term study (14 days), conducted im- umns (Alltech Associates, Inc, Deerfield, IL) and derivatized by the same methodology as used for triglycerides. The resulting pentaflu- mediately after transplantation, suggested reversibility of orobenzyl (PFB) ester (m/z 259) was analyzed for [ C ]palmitate 4 the effect (Table 2), the measurements did not include a enrichment by gas chromatography (Supelco fused silica capillary complete lipoprotein and apolipoprotein profile, nor column, 30 m  0.25 mm  0.25 m film) and mass spectrometry were the measurements frequent enough to closely moni- (Hewlett Packard HP 6890GC/ 5973 MSD). tor sirolimus-induced changes. Furthermore, in that ini- Statistical methods tial study, each patient was treated with a constant dose (1–7 mg/day) over a 14 day period (Table 2), whereas in The effects of sirolimus treatment on fasting lipids and lipo- protein levels were evaluated using Student’s paired t-tests (24). the later longer term study (42 days) all patients were TABLE 3. Plasma cholesterol and trigylceride response of renal transplant recipients to re-challenge with siroli- mus treatment (42 days) Cholesterol Level Triglyceride Level Before/ Before/ After Sirolimus After Sirolimus Hyperlipidemia Day 42 Day 42 Patient ID Day 1 Stop Drug Change Day 1 Stop Drug Change Before After mg/dl % mg/dl % WAS-1/2 244 431 92 275 974 254 II b II b WAS-3/4 273 366 34 225 284 26 II a II b WAS-5/6 234 301 29 376 491 31 II b II b WAS-7/8 224 394 75 165 481 191 II a II b WAS-9/10 131 216 65 135 212 57 N II a WAS-11/12 177 222 25 140 153 9 N II a Mean 213.4 321.7 50 227 432 95 S.D. 51 90 87 299 P value 0.007 0.1 Since the Day 1 point appeared spurious, the value used for this patient was the average of Day –7 and Day 7 value. 1172 Journal of Lipid Research Volume 43, 2002 Morrisett et al. Sirolimus and lipids in renal transplant patients 1173 Fig. 1. (A-F, parts labeled I) Measurement of sirolimus dosages during the 8 week protocol of the later longer term treatment (42 days) on sirolimus and followup (day 56). Sirolimus trough levels were measured on whole blood samples. Measurements in patients: (A, part I) WAS-1/2; (B, part I) WAS-3/4; (C, part I) WAS-5/6; (D, part I) WAS-7/8; (E, part I) WAS-9/10; (F, part I) WAS-11/12. (A-F, parts labeled II) Measurement of weekly lipid profiles during the 8 week protocol of the longer term treat- ment (42 days) on sirolimus and followup (day 56). Plasma samples were prepared by centrifugation of venous blood collected after 12 h fasting and plasma lipid levels were measured as described in the text. Measurements in patients: (A, part II) WAS-1/2; (B, part II) WAS-3/4; (C, part II) WAS-5/6; (D, part II) WAS-7/8; (E, part II) 1174 Journal of Lipid Research Volume 43, 2002 Fig. 1. (continued). WAS-9/10; (F, part II) WAS-11/12. (A-F, parts labeled III) Measurement of weekly apolipoprotein profiles during the 8 week protocol of the longer term treatment (42 days) on sirolimus and followup (day 56). Plasma samples were prepared by centrifugation of venous blood collected after 12 h fasting and weekly plasma apolipoprotein levels were measured as described in the text. Measurements in patients: (A, part III) WAS-1/2; (B, part III) WAS-3/4; (C, part III) WAS-5/6; (D, part III) WAS-7/8; (E, part III) WAS-9/10; (F, part III) WAS-11/12. started at 10 mg/day with the express purpose of re-induc- prompt reductions in triglyceride, and subsequent incre- ing hyperlipidemia (Table 3). As anticipated, it was neces- ments of dosage induced yet a second set of increases in sary to reduce the dose in those patients who exhibited triglycerides. The single exception to these observations sirolimus-induced hyperlipidemia exceeding the level al- was seen in patient WAS-11/12, who received a full 10 lowed by the protocol. Dosage was also reduced if re- mg/day dose but maintained a comparatively stable tri- quired by blood chemistries or cell count, and to bring glyceride level (range: 110–210 mg/dl) throughout the 42 the elevated creatinine values to normal levels. These dos- day treatment period. age adjustments usually prevented the patients from Effect of sirolimus on apolipoprotein levels achieving constant trough levels of sirolimus, but they en- The apolipoprotein showing the greatest response to abled the observation of dose-dependent lipid changes sirolimus was apoB-100, a major protein component of that would not have been detected with a constant dose VLDL and LDL. Hence, its dose dependent changes re- strategy. Monitoring sirolimus blood concentrations re- flect the composite changes in triglyceride (transported vealed that drug trough levels reflected dose in all patients primarily by VLDL) and cholesterol (transported primarily except Wyeth Ayerst study patient before/after sirolimus by LDL). This point is illustrated in apoB-100 levels of pa- treatment (WAS-11/12), whose concentrations steadily de- tient WAS-9/10 that rose to a maximum of 120 mg/dl at creased despite continuous high dosing (10 mg/day) day 14, corresponding to the maximum triglyceride level throughout the study (Fig. 1A–F, all parts labeled II). of this patient (410 mg/dl) at the same day. The abrupt de- crease in triglyceride to 105 mg/dl at Day 28 is somewhat Effect of sirolimus on cholesterol levels attenuated in the apoB-100 curve, due in part to the much In the initial, shorter-term study, sirolimus caused vari- slower decrement in LDL-C (Fig. 1E, parts II and III). able changes in total cholesterol levels within 14–28 days ApoC-II and apoC-III are important protein compo- (range: 3 to 62%; mean: 28%; Table 2). In the later nents of VLDL and HDL. The plasma levels of apoC-II 6-week study (42 days), sirolimus caused marked increases were typically low and did not change appreciably during in the total cholesterol levels in five of the six patients the course of the study (range: 2.4 to 10.6 mg/dl; (range: 25 to 92%; mean: 50%; P 0.007; Table 3). mean: 3.25 mg/dl, P 0.18). In contrast, the initial values The longer duration and larger dosage are likely reasons of apoC-III were substantial, and increased significantly for the greater elevation of total cholesterol in the second between day 1 and day 42 (range: 7 to 53 mg/dl; study. Frequent measurements during the second study mean 27 mg/dl; P 0.013). Since apoC-II is an activator indicated rather gradual increases in cholesterol levels (25) and apoC-III is an inhibitor (26, 27) of lipoprotein li- (Fig. 1A–F, all parts labeled II) from the time the drug was pase (LPL), these results provide a reasonable explana- started (day 1) to the time it was stopped (day 42). tion for the substantially lower LPL activity in these renal Because triglyceride levels often exceeded the 400 mg/ transplant patients (20–70%) compared with normolipi- dl limit for which the Friedewald equation is valid for cal- demic controls (15). culating LDL-C, it was necessary to measure this analyte ApoA-I is a principal apolipoprotein component of directly (dLDL-C). In general, dLDL-C increased gradu- HDL. In patients WAS-1/2, -5/6, -7/8, and -11/12, apoA-I ally with sirolimus treatment, resembling the changes levels were not remarkably affected by sirolimus treat- seen in total cholesterol with respect to timing but not ment. However, in patient WAS-3/4, apoA-I rose 58% magnitude (Fig. 1A–F, all parts labeled II). (Fig. 1B, part III), and in patient WAS-9/10 it rose 50% Throughout the entire 6-week treatment period (42 (Fig. 1E, part III) over the 42 day treatment period. The days), sirolimus had no effect on HDL-C levels any of the increase in apoA-I was attended by an increase in HDL for patients besides WAS-3/4. This patient had a remarkably patient WAS-3/4 (Fig. 1B, part II) but not for patient high initial HDL-C level (92 mg/dl), which rose slowly WAS-9/10 (Fig. 1E, part II). and monotonically, reaching a plateau level of 110 mg/dl at day 28 (Fig. 1A, part II). TABLE 4. Plasma free fatty acids levels (mEq/l) of renal transplant Effect of sirolimus on triglyceride levels patients who exhibited hypercholesterolemia and hypertriglyceridemia (type II b, Table 4) during treatment with sirolimus. The mean value In the initial shorter-term study (Table 2), sirolimus in- represent the average of 18 measurements over 24 h while off duced a substantial increase in the triglyceride levels of ev- (-1, -3, -5, -7) and on (-2, -4, -6, -8) sirolimus treatment ery patient (range: 24 to 175%; mean: 123%). For the Patient Mean SD P value later longer-term study (Table 3), sirolimus again elevated triglyceride levels (range: 9 to 254%; mean: 95%), but these changes were not as great as those observed in the WAS-1 0.772 0.294 0.037 29.5 short term study. Only two patients, WAS-1/2 and WAS-7/8, WAS-2 1.000 0.216 WAS-3 0.474 0.186 had elevations greater in the longer-term than the shorter- 0.001 21.3 WAS-4 0.575 0.213 term study (Tables 2 and 3). In general, triglyceride levels WAS-5 0.410 0.160 0.0001 37.0 were highly responsive to sirolimus dosage. This is well il- WAS-6 0.651 0.159 lustrated in responses of patients WAS-3/4 and WAS-9/10, WAS-7 0.411 0.138 0.0001 74.1 WAS-8 0.714 0.249 in which the initial 10 mg/day induced a rapid rise in tri- Mean  SD 21.8  18.9 glyceride levels; necessary reductions in dosage resulted in Morrisett et al. Sirolimus and lipids in renal transplant patients 1175 13 Effect of sirolimus on plasma free fatty acid levels h. The total amount of [ C ]palmitate incorporated into triglyceride is indicated by the integrated area under the After 42 days of sirolimus treatment, four of the six pa- kinetic curve. The areas under the VLDL, IDL, and LDL tients were hypercholesterolemic and hypertriglyceri- curves typically obtained for patients on sirolimus were demic (type Iib) (Table 3). The mean plasma free fatty substantially less than the areas under curves generated acid level of these patients increased by 40.5  16.8% for these patients when off drug. For example, patient (mean  SD) (Table 4). This expansion of the plasma WAS-7/8 had reductions of 50.5%, 56.3%, and 53.2% in free fatty acid pool was explored further with stable iso- [ C ]palmitate incorporation into VLDL, IDL, and LDL, tope kinetic studies. [ C ]palmitate was infused intrave- 4 respectively, when treated with sirolimus. Comparable re- nously over 7 h and its plasma levels monitored for 24 h by ductions in incorporation were also observed for patients GC/MS. In every experiment, the percent atom enrich- WAS-3/4 and WAS-5/6 (Table 6). ment returned to baseline within about 8 h. The shapes of These results suggest that sirolimus expands the plasma the kinetic curves (Fig. 2) differ among patients. However, pool of free fatty acid (mean 42.3%) resulting in in- in all four cases the areas under the curves obtained for creased hepatic synthesis of triglyceride secreted as VLDL patients on sirolimus were substantially less than when the (mean 38.3%). patients were off drug. The reduction in integrated area ranged from 20.3% to 62.7% with a mean  SD of 42.3 17.7 (Table 5). DISCUSSION Effect of sirolimus on triglyceride metabolism The infusion of [ C ]palmitate also made it possible to A primary objective of this study was to determine if the hyperlipidemic effects of sirolimus were reproducible, re- monitor the synthesis of triglyceride and its distribution among the VLDL, IDL, and LDL lipoprotein fractions. versible, and dose-dependent. For this purpose, the study Samples (n 18) were collected at frequent intervals dur- was performed in a small heterogeneous group of renal ing the initial 24 h period, and daily for the next 5 days, transplant recipients with different types of hyperlipi- demia (type IIa, IIb, IV). There were two reasons for resulting in well-defined kinetic curves for triglyceride synthesis. A set of representative curves obtained for a pa- studying these patients under those conditions: First, pa- tient (WAS-7/8) off and on sirolimus treatment is pre- tients were on maintenance regimens typical of many re- sented in Fig. 3. The curves typically indicated maximum nal allograft recipients, thus the effects caused by siroli- mus were the result of the drug acting in the environment enrichment at 6–8 h and returned to baseline by about 48 Fig. 2. Percent atom enrichment in the free fatty acid fractions of plasma collected during/after infusion of [ C ]palmitate into four renal transplant patients before (odd number, solid lines) and after 6 weeks of (even number, dashed lines) sirolimus treatment. The integrated area under each cur ve is tabulated in Table 6. 1176 Journal of Lipid Research Volume 43, 2002 TABLE 5. Expansion of the plasma free fatty acid pool a lesser extent. It is probable that the lipid-lowering ther- apy received by four of the patients (WAS-3/4, -5/6, -7/8, Patient Area Change and -11/12) significantly attenuated the lipid elevating ef- fects of sirolimus, even though their cholesterol and tri- WAS-1 24.38 glyceride levels rose 25–75% and 9–191%, respectively 47.4 WAS-2 12.82 (Table 3). Patient WAS-1/2, who received no lipid lower- WAS-3 30.04 38.7 ing therapy, had the largest absolute increase in choles- WAS-4 18.41 terol ( 187 mg/dl) and triglyceride ( 669 mg/dl) while WAS-5 12.04 20.3 WAS-6 9.59 on the drug. However, patient WAS-9/10, who also re- WAS-7 21.68 62.7 ceived no lipid lowering medication but a comparable WAS-8 8.08 dosage of sirolimus, showed less absolute changes that did Mean  SD 42.3  17.7 not force his lipid values outside the normal range. Thus, Expansion of the plasma free fatty acid pool as measured by dilu- sirolimus did not cause uniform lipid elevation in all of tion of [ C ] palmitate infused into renal transplant patients who ex- our patients. hibited hypercholesterolemia and hypertriglyceridemia (type IIB, Ta- ble 4) during treatment with sirolimus. The values shown represent the The US Phase III clinical studies have also demon- area under the % atom enrichment curve generated during the first strated that the incidence of hyperlipidemia is dependent 24 h after infusion (18 times points) while off (-1, -3, -5, -7) and on (-2, on sirolimus dosage (1). Our results confirm and extend -4, -6, -8) sirolimus treatment. the previously reported hypercholesterolemia and hyper- triglyceridemia observed in a Phase I clinical trial of renal of other agents typically present in transplant patients. transplant patients (28–32). Second, it was not ethically feasible to withdraw a medica- ApoA-I is the principal apolipoprotein component of tion proved to be efficacious in suppressing graft rejection HDL and increases the enzymatic activity of LCAT, a and controlling hyperlipidemia in specific patients. plasma enzyme that catalyzes the conversion of choles- The initial shorter-term sirolimus treatment after renal terol to cholesteryl ester. The apoA-I plasma concentra- transplantation resulted in clinically significant choles- tion is typically 119 mg/dl in normolipidemic subjects terol elevations within 2–4 weeks of treatment, which re- (33). In our study, the baseline apoA-I levels were 80–190 verted to near-baseline levels within 8 weeks after discon- mg/dl and the maximum levels on treatment were 105– tinuing treatment (Table 2). A similar effect of sirolimus 310 mg/dl (Fig. 1A–F, all parts labeled III). In four of the on triglyceride levels in these patients was also observed. six patients, the plasma apoA-I levels did not undergo no- To determine whether these effects were reproducible or table changes, consistent with the rather constant HDL-C whether metabolic adaptations occurred over time, pa- levels of the same patients. In contrast, patients WAS-3/4 tients were rechallenged with sirolimus. Significant in- and WAS-9/10 had baseline values of 190 mg/dl and 100 creases in cholesterol and/or triglyceride levels were re- mg/dl, which rose respectively to a maximum of 305 mg/ inducible in all six patients when re-challenged (Tables 2 dl and 155 mg/dl after 6 weeks of sirolimus therapy. and 3). One of these patients (WAS-9/10) was normolipi- These levels and changes of apoA-I are consistent with the demic (without lipid lowering therapy) before but mildly HDL-C levels of these patients, which increased slowly hypertriglyceridemic after the initial 2 weeks of sirolimus; throughout the duration of the study (Fig. 1B, part II). the same patient was normolipidemic before the later ApoB-100 is a major apolipoprotein component of 6-week re-challenge, and experienced moderate cholesterol VLDL, IDL, and LDL; hence, its concentrations are highly and triglyceride elevation during sirolimus treatment, but associated with triglyceride and cholesterol levels. At base- not to a level that would be considered hyperlipidemic line, apoB-100 concentrations ranged from 30–195 mg/dl (Table 3). Another patient (WAS-11/12) was hypercholes- and reached a maximum level of 120–330 mg/dl (Fig. terolemic and hypertriglyceridemic before and after the 1A–F, all parts labeled III). These values are substantially initial 2 weeks of sirolimus; this patient was normolipi- above the mean value of 90 mg/dl for apoB-100 in normo- demic (with lipid lowering therapy) before the later lipidemic subjects (33). One might expect that our pa- 6-week re-challenge, and experienced moderate choles- tients receiving triglyceride and cholesterol lowering terol elevation during sirolimus rechallenge (Table 3). agents (Table 1) would have the lower apoB-100 concen- These results suggest that some patients develop resis- trations (Fig. 1A–F, all parts labeled III). This did not turn tance to sirolimus-induced hyperlipidemia, even without out to be the case for WAS-7/8, whose baseline value of lipid lowering therapy, while others remain susceptible to 195 mg/dl rose to 330 mg/dl despite his being on gemfi- this effect, even with lipid lowering therapy. brozil and pravastatin. However, these lipid-lowering The present study demonstrates a prompt change in tri- agents were effective in attenuating the elevation of apoB- glyceride levels when sirolimus dosage is altered, while pa- 100 in patients WAS-5/6 and -11/12. These results suggest tients are maintained on CsA and prednisone. The start- that there is considerable inter-subject variability in the ca- ing dose of 10 mg/day induced substantial increases in pacity of pravastatin to upregulate LDL receptors and en- triglyceride levels within 14 days in five of six patients. De- hance apoB-100 removal in these sirolimus-treated pa- creasing the dose from 10 to as low as 0 mg/day either at- tients. tenuated or reversed the escalating triglyceride levels. ApoC-II is an activator (25) and apoC-III is an inhibitor Sirolimus dosage also affected plasma cholesterol levels to (26, 27) of LPL, which hydrolyzes triglyceride in VLDL Morrisett et al. Sirolimus and lipids in renal transplant patients 1177 13 Fig. 3. Incorporation of [ C ]palmitate into triglycerides of VLDL, IDL, and LDL of patient WAS-7/8 be- fore (solid lines) and after 6 weeks of (dashed lines) sirolimus treatment. 1178 Journal of Lipid Research Volume 43, 2002 14 TABLE 6. Change in [ C ]palmitate incorporation into triglycerides of VLDL, IDL, and LDL WAS-3 VLDL WAS-4 VLDL WAS-3 IDL WAS-4 IDL WAS-3 LDL WAS-4 LDL 3.478 1.677 2.153 1.474 0.883 0.824 51.8% 31.5% 6.7% WAS-5 WAS-6 WAS-5 WAS-6 WAS-5 WAS-6 VLDL VLDL IDL IDL LDL LDL 1.333 1.165 1.182 0.728 1.026 0.566 12.6% 38.4% 55.2% WAS-7 WAS-8 WAS-7 WAS-8 WAS-7 WAS-8 VLDL VLDL IDL IDL LDL LDL 1.910 0.946 1.026 0.448 1.294 0.605 50.5% 56.3% 53.2% Mean 38.3% 56.3% 38.4% Change in [ C ] palmitate incorporation into triglycerides of VLDL, IDL, and LDL of three renal transplant patients who exhibited hypercholesterolemia and hypertriglyceridemia (type IIb, Table 4) during treatment with sirolimus. The values shown represent the area under the % atom enrichment curves generated during the first 24 h after infusion of [ C ] palmitate (18 time points) while off (-3, -5, -7) and on (-4, -6, -8) sirolimus treatment. and chylomicrons. The baseline levels of apoC-II were eride of VLDL, IDL, and LDL was measured before and 2.4–12.7 mg/dl compared with the levels observed in nor- during sirolimus treatment in three patients. The mean molipidemics (3 mg/dl) (33). The levels of apoC-III were isotopic enrichment was decreased by 38.3%, 56.3%, and substantially higher than the levels observed in normolipi- 38.4%, respectively (Table 6). Taken together, these re- demics (16 mg/dl) (33), ranging from 20 to 130 mg/dl at sults support the view that sirolimus enhances the action baseline and from 20 mg/dl to 180 mg/dl at maximum of hormone sensitive lipase (HSL) and perhaps also inhib- level (Fig. 1A–F, all parts labeled III). These elevated levels its LPL. These effects are the opposite of those mediated of apoC-III may contribute significantly to the depressed by insulin, suggesting that sirolimus may induce hypertri- levels of LPL activity seen in these immunosuppressed pa- glyceridemia via an insulin-dependent signaling pathway. tients (15). However, even though the apoC-III-apoC-II If the drug interferes with insulin-stimulated triglyceride ratio in three patients (WAS-1/2, -3/4, and -5/6) was sig- storage in adipocytes, this could lead to increased release nificantly higher after sirolimus treatment, these patients of FFAs into the circulation, their increased uptake by the did not exhibit significantly lower LPL activity than the liver, and increased hepatic secretion of VLDL triglycer- other three patients (WAS-7/8, -9/10, and -11/12), whose ides. Alternatively, sirolimus may also decrease FFA oxida- apoC-III/apoC-II ratio was not notably altered (Fig. 1A–F, tion leading to increased FFA availability. all parts labeled III). Massy et al. (34) have compared the The elevated triglyceride levels in patients WAS-1/2, -3/ separate effects of sirolimus and CsA on the plasma con- 4, -5/6, and -7/8 could be due to increased hepatic pro- centration of apolipoproteins and LPL. They observed sig- duction of triglyceride rich lipoproteins and/or de- nificantly higher apoC-II in sirolimus treated patients (7.9 creased removal of them. Our previous study (15) indi- mg/dl) than in CsA treated patients (5.1 mg/dl). Al- cated significant reduction in the fractional catabolic rate though apoC-II levels were higher in the sirolimus treated of apoB-100-containing lipoproteins in patients receiving patients (18.8 mg/dl) than in CsA treated patients (14.1 sirolimus treatment. The present study provides strong ev- mg/dl), this difference was not statistically significant. Im- idence that sirolimus-increased production of triglyceride- portantly, LPL and hepatic lipase activities were the same rich lipoproteins also contributes to the observed hyper- in CsA and sirolimus-treated patients (34). triglyceridemia. A second major goal of this study was to define the In summary, sirolimus induces or exacerbates hyperlipi- mechanism whereby hypertriglyceridemia was induced by demia in a reproducible, reversible, and dose-dependent sirolimus in four of the patients. Toward this end, stable manner in some renal transplant recipients. The clinical isotope experiments were conducted to examine fatty acid implication of these results is that administration of the and triglyceride metabolism before and during sirolimus minimal dose that elicits therapeutic immunosuppression, treatment. [ C ]palmitate infusion experiments indi- as indicated by measurement of plasma sirolimus levels, cated significant expansion of the free fatty acid pool; may be advantageous for minimizing potential adverse ef- mean 42.3%, Table 5 (35) by sirolimus. These results fects on lipid metabolism that occur in some transplant were supported by measurements of total free fatty acid patients. levels, which indicated considerable expansion of this pool (mean 40.5). Although it is possible that some de novo fatty acid synthesis is induced by sirolimus, it is un- This study was supported in part by the Welch Foundation likely that it would cause pool expansion of this magni- (grant Q-1325 to J.D.M); by the National Institutes of Health tude (36). (HL 07812 to J.D.M and DK 38016 to B.D.K.); and by the An expanded fatty acid pool may lead to increased he- Wyeth Ayerst Research Laboratories (to J.D.M. and B.D.K.). patic synthesis of triglycerides. To assess this possibility, The authors acknowledge the excellent technical assistance of the incorporation of infused [ C ]palmitate into triglyc- Karima Ghazzaly. (TTGA). Morrisett et al. Sirolimus and lipids in renal transplant patients 1179 A. W. Wahlefeld, and W. Gruber. 1984. Reagent for the enzymatic REFERENCES determination of serum total triglyceride with improved lipolytic efficacy. J. Clin. Chem. Clin. Biochem. 22: 165–174. 1. Kahan, B. D. 2000. Efficacy of sirolimus compared with azathio- 19. Warnick, G. R., J. Benderson, and J. J. Albers. 1982. Dextran Sul- prine for reduction of acute allograft rejection: a randomized mul- fate-Mg2 precipitation procedure for the quantification of high ticenter study. The Rapamune US Study Group. Lancet. 356 density-lipoprotein cholesterol. Clin. Chem. 28: 1379–1388. (9225): 194–202. 20. Hixson, J. E., and D. T. Vernier. 1990. Restriction isotyping of the 2. Watson, C. J., P. J. Friend, N. V. Jamieson, T. W. Frick, G. Alex- human apolipoprotein E by gene amplification and cleavage with ander, A. E. Gimson, and R. Calne. 1999. Sirolimus a potent new Hha1. J. Lipid Res. 31: 545–548. immunosuppressive agent for liver transplantation. Transplanta- 21. Redgrave, T. G., D. C. K. Roberts, and C. E. West. 1975. Separation tion. 67: 505–509. of plasma lipoproteins by density gradient ultracentrifugation. 3. Stroatman, L. P., and J. G. Coles. 2000. Pediatric utilization of ra- Anal. Biochem. 65: 42–49. pamycin for severe cardiac allograft rejection. Transplantation. 70: 22. Skipski, V. P., and M. Barclay. 1969. Thin Layer Chromatography 541–543. of Lipids (Ed: J.M. Lowenstein). Methods Enzymol. XIV: 530–598. 4. Poon, M., S. O. Marx, R. Gallo, J. J. Badimon, M. B. Taubman, and 23. Hachey, D. L., B. W. Patterson, L. J. Elsas, P. J. Reeds, and P. D. A. R. Marks. 1996. Rapamycin inhibits vascular smooth muscle cell Klein. 1991. Isotopic determination of organic keto acid pentaflu- migration. J. Clin. Invest. 98: 2277–2283. orobenzyl esters in biological fluids by negative chemical ioniza- 5. Burke, S. E., N. L. Lubbers, Y. W. Chen, G. C. Hsieh, K.W. Molli- tion gas chromatography/ mass spectrometry. Anal. Chem. 63: son, J. R. Luly, and C. D. Wegner. 1999. Neointimal formation after 919–923. balloon-induced vascular injury in Yucatan minipigs is reduced by 24. Matthews, D. E., and V. T. Farewell. 1988. Using and understanding oral rapamycin. J. Cardiovasc. Pharmacol. 33(6): 829–835. Medical Statistics. 2nd edition. S. Karger AG, Basel, Switzerland. 113– 6. Gallo, R., A. Padurean, T. Jayaraman, S. Marx, M. Roque, S. Adel- 119. man, J. Chesebro, J. Fallon, V. Fuster, A. Marks, and J. J. Badimon. 25. Bengtsson, G., and T. Olivecrona. 1980. Lipoprotein lipase: some 1999. Inhibition of intimal thickening after balloon angioplasty in effects of activator proteins. Eur. J. Biochem. 106: 549–555. porcine coronary arteries by targeting regulators of the cell cycle. 26. McConathy, W. J., J. C. Gesquiere, H. Bass, A. Tartan, J. C. Fruchart, Circulation. 99: 2164–2170. and C. S.Wang 1992. Inhibition of lipoprotein lipase activity by syn- 7. Brown, E. J., M. W. Albers, T. B. Shin, K. Ichikawa, C. T. Keith, W. thetic peptides of apolipoprotein CIII. J. Lipid Res. 33: 995–1003. Lane, and S. L. Schreiber. 1994. A mammalian protein targeted by 27. Ginsberg, H. N., N. A. A. Le, I. J. Goldberg, J. C. Gibson, A. Rubin- G1- arresting rapamycin-receptor complex. Nature. 369: 756–758. stein, P. Wang-Iverson, R. Norum, and W. V. Brown. 1986. Apolipo- 8. Raught, B., and A. C. Gingras. 1999. EIF4E activity is regulated at protein B metabolism in subjects with deficiency of apo CIII and multiple levels. Int. J. Biochem. Cell. Biol. 31: 43–57. AI. Evidence that apolipoprotein CIII inhibits catabolism of tri- 9. Brunn, G. J., C. C. Hudson, A. Sekulic, J. M. Williams, H. Hosoi, glyceride rich lipoproteins by lipoprotein lipase in vivo. J. Clin. In- P. J. Houghton, J. C. Lawrence, and R. T. Abraham. 1997. Phos- vest. 78: 1287–1295. phorylation of the translational repressor PHAS-1 by the mamma- 28. Murgia, M. G., S. Jordan, and B. D. Kahan. 1996. The side effect lian target of rapamycin. Science. 277: 99–101. profile of sirolimus: a phase I study in quiescent cyclosporine-pred- 10. Burnett, P. E., R. K. Barrow, N. A. Cohen, S. H. Snyder, and D. M. nisone-treated renal transplant patients. Kidney Int. 49: 209–216. Sabatini. 1998. RAFT-1 phosphorylation of the translational regulator 29. Kahan, B. D., J. Podbielski, K. L. Napoli, S. M. Katz, H. U. Meirer- p70S6 kinase and 4EBP-1. Proc. Natl. Acad. Sci. USA. 95: 1432–1437. Kriesche, and C. T. Van Buren. 1998. Immunosuppressive effects 11. Kumar V., D. Sabatini, P. Pandey, A-C. Gingras, P. K. Majumder, M. and safety of a sirolimus/cyclosporine combination regimen for Kumar, Z-M. Yaun, G. Carmicheal, R. Weichselbaum, N. Sonen- renal trasnplantation. Transplantation. 66: 1040–1046. berg, D. Kufe, and S. Kharbanda. 2000. Regulation of the rapa- 30. Kahan, B. D. 1998. Rapamycin: personal algorithms for use based mycin and FKBP-target 1/mammalian target of rapamycin and on 250 treated allograft recipients. Transplant. Proc. 30: 2185–2188. cap-dependent initiation of translation by the c-abl protein-tyro- 31. Mac Donald, A. S. 2001. The Rapamune Global Study Group: a sine kinase. J. Biol. Chem. 275: 10779–10787. randomized, placebo-controlled trial of rapammune in primary 12. United States Renal Data System (USRDA). 1998. The National In- renal allograft recipients. Transplantation. 71: 271–280. stitute of Health, National Institute of Diabetes, and Digestive and 32. Kahan, B. D., K. L. Napoli, P. A. Kelly, J. Podbielski, I. Hussein, Kidney Disease. Annual Data report, Bethesda, Maryland. D. L. Urbauer, S. H. Katz, and C. T. Van Buren. 2000. Therapeutic 13. Brattström, C., H. E. Wilczek, G. Tydén, Y. Böttiger, J. Säwe, and C. G. drug monitoring of sirolimus: correlations with efficacy and toxic- Groth. 1998. Hypertriglyceridemia in renal transplant recipients ity. Clin. Transplant. 14: 97–109. treated with sirolimus. Transplant. Proc. 30: 3950–3951. 33. Labeur, C., and M. Rosseneu. 1997. Immunological assays of 14. Brattström, C., H. Wilczek, G. Tyden, Y. Böttiger, J. Säwe, and C. G. apolipoproteins A-II, A-IV, C-I, C-II, C-III in plasma. In Hand- Groth. 1998. Hyperlipidemia in renal transplant recipients treated book of Lipoprotein Testing. N. Rifai, R. G. Warnik, and M. H. with Sirolimus (Rapamycin). Transplantation. 65: 1273–1274. Dominiczak, editors. ASCP Press, Chicago. 315–328. 15. Hoogeveen, R. C., C. M. Ballantyne, H. J. Pownall, A. R. Opekun, 34. Massy, Z. A., J. P. DeBandt, E. Morelon, M. Thevenin, B. Lacourr, D. L. Hachey, J. S. Jaffe, S. Oppermann, B. D. Kahan, and J. D. and H. Kreis. 2000. Hyperlipidemia and post-heparin lipase activi- Morrisett. 2001. Effect of rapamycin on the metabolism of Apo ties in renal transplant recipients treated with sirolimus or cy- B-100 containing lipoproteins in renal transplant patients. Trans- closporin A. Nephrol. Dial. Transplant. 15: 928. plantation. 72: 1244–1250. 35. Morrisett, J. D., R. C. Hoogeveen, G. Abdel-Fattah, H. J. Pownall, 16. Napoli, K. L., and B. D. Kahan. 1996. Routine clinical monitoring C. M. Ballantyne, A. R. Opekun, D. L. Hachey, J. S. Jaffe, S. Opper- of sirolimus (rapamycin) whole-blood concentrations by HPLC mann, and B. D. Kahan. 2001. Effect of rapamycin on free fatty with ultraviolet detection. Clin. Chem. 42: 1944. acid metabolism in renal transplant patients. Arterioscler. Thromb. 17. Seidel, J., E. Hagele, J. Ziegenhorn, and J. Wahlenfeld. 1983. Re- Vasc. Biol. 21: 654. agents for the enzymatic determination of serum total cholesterol 36. Hellerstein, M. K., R. A. Neese, and J-M. Schwarz. 1993. Model for with improved lipolytic efficiency. Clin. Chem. 29: 1075–1080. measuring absolute rates of hepatic de novo lipogenesis and rees- 18. Nagele, U., E. Hagele, G. Sauer, E. Wiedermann, P. Lehmann, terification of free fatty acids. Am. J. Physiol. 265: E814–E820. 1180 Journal of Lipid Research Volume 43, 2002 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Lipid Research Unpaywall

Effects of sirolimus on plasma lipids, lipoprotein levels, and fatty acid metabolism in renal transplant patients

Download PDF
11 pages

Loading next page...
 
/lp/unpaywall/effects-of-sirolimus-on-plasma-lipids-lipoprotein-levels-and-fatty-Fwc6meyroA

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0022-2275
DOI
10.1194/jlr.m100392-jlr200
Publisher site
See Article on Publisher Site

Abstract

Effects of sirolimus on plasma lipids, lipoprotein levels, and fatty acid metabolism in renal transplant patients 1, Joel D. Morrisett, * Ghada Abdel-Fattah,* Ron Hoogeveen,* Eddie Mitchell,* Christie M. Ballantyne,* Henry J. Pownall,* Antone R. Opekun,* Jonathon S. Jaffe, † † Suzanne Oppermann, and Barry D. Kahan The Department of Medicine,* Baylor College of Medicine, Houston, Texas; Department of Surgery, University of Texas-Houston Health Science Center, Houston, TX; and Wyeth Ayerst Laboratories, Philadelphia, PA Abstract Sirolimus (Rapammune , rapamycin, RAPA) is a lipoprotein levels, and fatty acid metabolism in renal trans- potent immunosuppressive drug that reduces renal trans- plant patients. J. Lipid Res. 2002. 43: 1170–1180. plant rejection. Hyperlipidemia is a significant side effect of sirolimus treatment, and frequently leads to cardiovascu- Supplementary key words rapamycin • triglyceride • cholesterol lar disease. This study was undertaken to determine the repeatability, reversibility, and dose dependence of the plasma lipid and apolipoprotein altering effects of siroli- Sirolimus (Rapammune , rapamycin, RAPA) is a novel mus, and to elucidate the mechanism by which sirolimus in- macrocyclic lactone immunosuppressive drug capable of duces hypertriglyceridemia in some renal transplant pa- significantly reducing acute graft rejection in kidney (1), tients. Six patients with renal allografts maintained on cyclosporine A and prednisone were selected on the basis liver (2), and heart (3) transplant patients. Previous stud- of their previous hyperlipidemic response to short term (14 ies have shown that sirolimus reduces the incidence of days) sirolimus administration. For longer-term treatment, acute rejection when administered in conjunction with cy- each patient was started on 10 mg/day sirolimus and contin- closporine and prednisone (1). Furthermore, sirolimus in- ued as tolerated for 42 days to reinduce hyperlipidemia. hibits vascular smooth muscle cell proliferation and re- Timed blood samples were analyzed for lipid, apolipopro- duces neointimal formation in humans, rats, and pigs, tein, and sirolimus levels. During sirolimus administration, thereby attenuating restenosis following angioplasty (4–6). mean total plasma cholesterol increased from 214 mg/dl to Sirolimus binds to the immunophilin FK506 binding 322 mg/dl (50%; range 25–92%); LDL-cholesterol levels protein (FKBP12). Sirolimus/FKBP12 binary complex followed a similar pattern. Mean triglyceride level rose from 227 to 432 mg/dl (95%; range 9–254%). ApoB-100 does not bind to calcineurin, and therefore is not neuro- concentration rose from 124 to 160 mg/dl (28%; P toxic or nephrotoxic. Instead, sirolimus/FKBP12 binds to 0.05). ApoC-III level increased from 28.9 to 55.5 mg/dl, a protein kinase called mammalian target of rapamycin 92%; (P  0.013). These lipid and apolipoprotein changes (mTOR). mTOR controls proteins that regulate mRNA were found to be repeatable, reversible, and dose depen- translation initiation and G1 progression (7). Recent stud- dent. [ C ]palmitate metabolic studies in four patients with ies have shown that mTOR directly phosphorylates p70S6 hypertriglyceridemia indicated that the free fatty acid pool kinase (8), the eukaroytic translation initiator protein 4G1 was expanded by sirolimus treatment (mean  42.3%). In- (eIF4G1), and translation inhibitor (4E-BP1) (8–10). corporation of [ C ]palmitate into triglycerides of VLDL, IDL, and LDL was decreased 38.3%, 42,1%, and 38.4%, re- Therefore, inhibition of mTOR by sirolimus contributes spectively, by sirolimus treatment of these patients. to translational arrest by down-regulation of p70S6K, and These results suggest that sirolimus alters the insulin signal- by increasing the affinity of 4E-BP1 (11). Consequently, ing pathway so as to increase adipose tissue lipase activity sirolimus immunosuppressive action is due to inhibition and/or decrease lipoprotein lipase activity, resulting in in- of T-cell activation at a later stage of the cell cycle, G1, and creased hepatic synthesis of triglyceride, increased secre- inhibition of p70S6K (10). tion of VLDL, and increased hypertriglyceridemia.—Morri- A major adverse reaction associated with sirolimus ther- sett, J. D., G. Abdel-Fattah, R. Hoogeveen, E. Mitchell, C. M. Ballantyne, H. J. Pownall, A. R. Opekun, J. S. Jaffe, S. Opper- mann, and B. D. Kahan. Effects of sirolimus on plasma lipids, Abbreviations: CsA, cyclosporine A; FKBP, FK506 binding protein; PFB, pentafluorobenzyl; WAS-#/#, Wyeth Ayerst study patient before/ after sirolimus treatment. Manuscript received 6 November 2001 and in revised form 29 April 2002. To whom correspondence should be addressed. DOI 10.1194/jlr.M100392-JLR200 e-mail: [email protected] Copyright © 2002 by Lipid Research, Inc. 1170 Journal of Lipid Research Volume 43, 2002 This article is available online at http://www.jlr.org This is an Open Access article under the CC BY license. suitability for the study. Blood sampling for the lipid profile and apy is hyperlipidemia, a major risk factor for cardiovascu- lipolytic enzymes (e.g., post heparin lipase) were scheduled to lar disease, and the most common cause of death after re- avoid interference of one test with another. The initial shorter- nal transplantation (12). Several studies have shown an term study (14 days) was conducted immediately after transplan- increase in serum triglyceride levels in renal transplant re- tation. Each patient was treated with a constant dose (1–7 mg/ cipients treated with sirolimus (13, 14). Their hyperlipi- day) of sirolimus over a 14 day period (Table 2). demia was dose-dependent and reversible within 1 to 2 For the longer-term study (42 days), qualifying patients began months after discontinuation of treatment (13, 14). the 8-week protocol on day 6 with a pre-drug lipoprotein meta- In the present study, we have examined the depen- bolic study lasting 6 days, up to day 1. On day 1, each patient dence of lipid, lipoprotein, and apolipoprotein levels, as started the sirolimus treatment initially at a level of 10 mg/day for 42 days. While on sirolimus, the patient returned weekly or well as fatty acid and triglyceride metabolism on sirolimus biweekly to the outpatient center for determination of lipids, li- dosage and treatment duration in renal allograft recipi- poproteins, apolipoproteins, lipid enzymes, and sirolimus trough ents with different types of hyperlipidemia. levels. If the patient’s lipid levels exceeded an acceptable range, then the sirolimus dose was reduced as described previously (15). After 42 days on treatment, a second lipoprotein metabolic METHODS study was initiated, lasting 6 days until day 47, after which siroli- mus treatment was discontinued. Cyclosporin and prednisone Patient selection maintenance therapy were continued. Each patient returned to the outpatient center on day 56 to give another fasting follow-up This study was performed in six patients, each of whom had blood sample for determination of the final lipid and lipopro- received a renal allograft within 3–8 years at Hermann Hospital tein profile, and to undergo the closeout physical examination. Transplant Center, Houston, Texas. These patients were selected based on their previous hyperlipidemic response upon short- term treatment (14 days) with sirolimus and had stable renal al- Sirolimus measurements lografts. All patients selected had no evidence of hepatic or bil- Sirolimus trough levels were measured on whole blood sam- iary dysfunction, as reflected in serum transaminase levels not ples with a multi-step liquid-liquid extraction followed by re- more than 20% above normal limits, nor lipid abnormalities with versed-phase-HPLC with ultraviolet detection performed by Dr. triglycerides 400 mg/dl or cholesterol 250 mg/dl. Patients Kim Napoli at the Organ Transplantation Center of the Univer- indicated their willingness to participate in the study by signing a sity of Texas Health Science Center at Houston (16). consent form, approved by the Institutional Review Boards for human research at the University of Texas Health Science Cen- Lipid and apolipoprotein measurements ter-Houston and Baylor College of Medicine and its affiliated Lipid, lipoprotein, and apolipoprotein measurements were hospitals. The patient group included four females and two performed in the Atherosclerosis Lipid Laboratory of The Meth- males aged 27–55 years (44.8  10.6) (Table 1). All six patients odist Hospital. Plasma samples were prepared by centrifugation were maintained on cyclosporine A (CsA) (Neoral), prednisone, (1500 g, 10 min, 4C) of venous blood collected after 12 h fasting and diuretic therapy. Four patients had developed mixed hyper- tm into Vacuutainer tubes containing EDTA. Total plasma choles- lipidemia, one had developed hypercholesterolemia, and one terol (17) and triglycerides (18) were measured enzymatically had developed hypertriglyceridemia in response to the previous (Boehringer Mannheim Diagnostics). LDL cholesterol (LDL-C) short-term sirolimus treatment (Table 2). Patients who had been levels were determined directly from the plasma after immuno- diagnosed with diabetes were receiving lipid-lowering medica- precipitation of VLDL and HDL using a kit from Sigma Chemi- tions (Table 1). This regimen was continued unaltered while on cal Co. (St. Louis, MO). HDL-C was determined by measuring sirolimus during the longer 6-week study (42 days). All six pa- cholesterol in the supernatant liquid after precipitation of the tients had the apo-E3/E3 genotype. VLDL and LDL with MgCl and dextran sulfate (19). Plasma apoB-100 was measured by ELISA using Mab RP-066 (Intracel, Study design Inc., Rockville, MD). ApoA-I was measured by nephelometry of Two weeks before commencing the study, each candidate pa- the precipitate formed with anti-apoA-I (IncStar, Inc.). ApoC-II, tient underwent screening evaluation at the Hermann Hospital and apoC-III were determined by radial immunodiffusion (Dai- Transplant Center. A complete physical examination and a series chi, Ltd.). ApoE genotyping was performed using a PCR based of biochemical screening tests were performed to determine method (20). TABLE 1. Clinical and demographic data of the renal transplant re-challenge (42 days) study participants Prednisone Creatinine Before Creatinine After ID Age Gender Race Donor Type Lipid RX Neoral (N) Dose Dose Sirolimus Sirolimus Diabetes Diagnosis mg mg/dl WAS-1/2 27 F Hispanic LRD None 100/75 7.5 1.7 2.5 No ESRD/GN WAS-3/4 55 F Black Cad P 125/100 7.5 0.9 0.9 IDDM/SI ESRD/HTN WAS-5/6 53 F Hispanic LRD L, P 125/125 5 0.7 0.6 NIDDM ESRD/HTN WAS-7/8 49 F Hispanic LRD L, P 125/100 7.5 1.5 1.1 IDDM/SI ESRD/CP WAS-9/10 37 M White LRD None 100/100 5 1.7 3.1 No ESRD/Lupus WAS-11/12 48 M Hispanic LRD L, P 200/175 7.5 1.6 1.7 NIDDM ESRD/HTN LRD, live related donor; P, pravacol; L, lopid; ESRD, end stage renal disease; GN, gynecological disease; HTN, hypertension; CP, chronic pyelo- nephritis; IDDM/SI, insulin dependent diabetes mellitus/steroid induced; Cad, cadaveric. Dose changed to 75/50 by the end of sirolimus treatment. Morrisett et al. Sirolimus and lipids in renal transplant patients 1171 TABLE 2. Plasma cholesterol and trigylceride response of renal transplant recipients to the initial shorter term sirolimus treatment (14 days) Cholesterol Level Before/ Triglyceride Level Before/ After Sirolimus After Sirolimus Hyperlipidemia Day 14 Day 14 b b Patient ID Sirolimus Dose Day 1 Stop Drug Day 28 Day 56 Change Day 1 Stop Drug Day 28 Day 56 Change Before After mg/day mg/dl % mg/dl % a a WAS-1/2 2 264 416 ND ND 57 169 465 ND ND 175 II a II b WAS-3/4 1 294 246 337 300 16 123 294 211 163 139 II a II b WAS-5/6 6 249 373 293 ND 49 739 1709 652 ND 131 II b II b WAS-7/8 6 333 399 293 252 20 248 628 313 155 153 II b II b WAS-9/10 7 148 241 191 180 62 156 340 162 137 118 N II b WAS-11/12 2 318 310 341 303 3 315 392 475 298 24 II b II b Mean 28 123 P value 0.1 0.04 ND, not determined. Percent change is the difference between day –1 and day 14 (when sirolimus treatment was stopped). At this time, patient WAS-1/2 was on sirolimus treatment for 61 days. Sirolimus treatment was started one day after renal transplantation. Metabolic studies Log or rank transformations were utilized when needed to meet Patients were fasted overnight prior to the start of their meta- the assumptions of the t-test. Statistical analyses were conducted bolic studies (15). Sodium [ C ]palmitate (Isotec, Inc., Miamis- using STATA (Release 4.0) and Prism (version 2.0) software. burg, OH) complexed to human serum albumin (Centeon, LLC. Plots of percent atom enrichment before and after sirolimus Kankakee, IL) was administered by constant intravenous infusion treatment were quantitatively compared by calculating the area (0.6 mg/kg/h) over 7 h. Each patient was given oral Sustecal (30 under the curve out to 36 h. The first and last data points were kcal/kg), which was consumed in 16 equal portions at hourly inter- used to determine the baseline amplitude, which was used to cor- vals, providing 22% of calories from fat and 0.88 g protein/kg. rect the integrated area. Blood samples (15 ml) were drawn, 18 over the first 24 h and 1 daily for the next 5 days, from which VLDL, IDL, and LDL were isolated by density gradient ultracentrifugation (21). These lipoproteins RESULTS were delipidated by organic solvent extraction (CHCl -CH OH, 2:1, 3 3 v/v) and the different lipid fractions separated by thin layer chro- A major objective of the current study was to determine matography (22). The triglyceride fraction was hydrolyzed with 15% if sirolimus-induced hyperlipidemia in renal transplant KOH and derivatized with pentafluorobenzylbromide (23). Plasma patients is reproducible, reversible, and dose-dependent. free fatty acids (FFAs) were isolated by solid phase extraction col- Although the shorter-term study (14 days), conducted im- umns (Alltech Associates, Inc, Deerfield, IL) and derivatized by the same methodology as used for triglycerides. The resulting pentaflu- mediately after transplantation, suggested reversibility of orobenzyl (PFB) ester (m/z 259) was analyzed for [ C ]palmitate 4 the effect (Table 2), the measurements did not include a enrichment by gas chromatography (Supelco fused silica capillary complete lipoprotein and apolipoprotein profile, nor column, 30 m  0.25 mm  0.25 m film) and mass spectrometry were the measurements frequent enough to closely moni- (Hewlett Packard HP 6890GC/ 5973 MSD). tor sirolimus-induced changes. Furthermore, in that ini- Statistical methods tial study, each patient was treated with a constant dose (1–7 mg/day) over a 14 day period (Table 2), whereas in The effects of sirolimus treatment on fasting lipids and lipo- protein levels were evaluated using Student’s paired t-tests (24). the later longer term study (42 days) all patients were TABLE 3. Plasma cholesterol and trigylceride response of renal transplant recipients to re-challenge with siroli- mus treatment (42 days) Cholesterol Level Triglyceride Level Before/ Before/ After Sirolimus After Sirolimus Hyperlipidemia Day 42 Day 42 Patient ID Day 1 Stop Drug Change Day 1 Stop Drug Change Before After mg/dl % mg/dl % WAS-1/2 244 431 92 275 974 254 II b II b WAS-3/4 273 366 34 225 284 26 II a II b WAS-5/6 234 301 29 376 491 31 II b II b WAS-7/8 224 394 75 165 481 191 II a II b WAS-9/10 131 216 65 135 212 57 N II a WAS-11/12 177 222 25 140 153 9 N II a Mean 213.4 321.7 50 227 432 95 S.D. 51 90 87 299 P value 0.007 0.1 Since the Day 1 point appeared spurious, the value used for this patient was the average of Day –7 and Day 7 value. 1172 Journal of Lipid Research Volume 43, 2002 Morrisett et al. Sirolimus and lipids in renal transplant patients 1173 Fig. 1. (A-F, parts labeled I) Measurement of sirolimus dosages during the 8 week protocol of the later longer term treatment (42 days) on sirolimus and followup (day 56). Sirolimus trough levels were measured on whole blood samples. Measurements in patients: (A, part I) WAS-1/2; (B, part I) WAS-3/4; (C, part I) WAS-5/6; (D, part I) WAS-7/8; (E, part I) WAS-9/10; (F, part I) WAS-11/12. (A-F, parts labeled II) Measurement of weekly lipid profiles during the 8 week protocol of the longer term treat- ment (42 days) on sirolimus and followup (day 56). Plasma samples were prepared by centrifugation of venous blood collected after 12 h fasting and plasma lipid levels were measured as described in the text. Measurements in patients: (A, part II) WAS-1/2; (B, part II) WAS-3/4; (C, part II) WAS-5/6; (D, part II) WAS-7/8; (E, part II) 1174 Journal of Lipid Research Volume 43, 2002 Fig. 1. (continued). WAS-9/10; (F, part II) WAS-11/12. (A-F, parts labeled III) Measurement of weekly apolipoprotein profiles during the 8 week protocol of the longer term treatment (42 days) on sirolimus and followup (day 56). Plasma samples were prepared by centrifugation of venous blood collected after 12 h fasting and weekly plasma apolipoprotein levels were measured as described in the text. Measurements in patients: (A, part III) WAS-1/2; (B, part III) WAS-3/4; (C, part III) WAS-5/6; (D, part III) WAS-7/8; (E, part III) WAS-9/10; (F, part III) WAS-11/12. started at 10 mg/day with the express purpose of re-induc- prompt reductions in triglyceride, and subsequent incre- ing hyperlipidemia (Table 3). As anticipated, it was neces- ments of dosage induced yet a second set of increases in sary to reduce the dose in those patients who exhibited triglycerides. The single exception to these observations sirolimus-induced hyperlipidemia exceeding the level al- was seen in patient WAS-11/12, who received a full 10 lowed by the protocol. Dosage was also reduced if re- mg/day dose but maintained a comparatively stable tri- quired by blood chemistries or cell count, and to bring glyceride level (range: 110–210 mg/dl) throughout the 42 the elevated creatinine values to normal levels. These dos- day treatment period. age adjustments usually prevented the patients from Effect of sirolimus on apolipoprotein levels achieving constant trough levels of sirolimus, but they en- The apolipoprotein showing the greatest response to abled the observation of dose-dependent lipid changes sirolimus was apoB-100, a major protein component of that would not have been detected with a constant dose VLDL and LDL. Hence, its dose dependent changes re- strategy. Monitoring sirolimus blood concentrations re- flect the composite changes in triglyceride (transported vealed that drug trough levels reflected dose in all patients primarily by VLDL) and cholesterol (transported primarily except Wyeth Ayerst study patient before/after sirolimus by LDL). This point is illustrated in apoB-100 levels of pa- treatment (WAS-11/12), whose concentrations steadily de- tient WAS-9/10 that rose to a maximum of 120 mg/dl at creased despite continuous high dosing (10 mg/day) day 14, corresponding to the maximum triglyceride level throughout the study (Fig. 1A–F, all parts labeled II). of this patient (410 mg/dl) at the same day. The abrupt de- crease in triglyceride to 105 mg/dl at Day 28 is somewhat Effect of sirolimus on cholesterol levels attenuated in the apoB-100 curve, due in part to the much In the initial, shorter-term study, sirolimus caused vari- slower decrement in LDL-C (Fig. 1E, parts II and III). able changes in total cholesterol levels within 14–28 days ApoC-II and apoC-III are important protein compo- (range: 3 to 62%; mean: 28%; Table 2). In the later nents of VLDL and HDL. The plasma levels of apoC-II 6-week study (42 days), sirolimus caused marked increases were typically low and did not change appreciably during in the total cholesterol levels in five of the six patients the course of the study (range: 2.4 to 10.6 mg/dl; (range: 25 to 92%; mean: 50%; P 0.007; Table 3). mean: 3.25 mg/dl, P 0.18). In contrast, the initial values The longer duration and larger dosage are likely reasons of apoC-III were substantial, and increased significantly for the greater elevation of total cholesterol in the second between day 1 and day 42 (range: 7 to 53 mg/dl; study. Frequent measurements during the second study mean 27 mg/dl; P 0.013). Since apoC-II is an activator indicated rather gradual increases in cholesterol levels (25) and apoC-III is an inhibitor (26, 27) of lipoprotein li- (Fig. 1A–F, all parts labeled II) from the time the drug was pase (LPL), these results provide a reasonable explana- started (day 1) to the time it was stopped (day 42). tion for the substantially lower LPL activity in these renal Because triglyceride levels often exceeded the 400 mg/ transplant patients (20–70%) compared with normolipi- dl limit for which the Friedewald equation is valid for cal- demic controls (15). culating LDL-C, it was necessary to measure this analyte ApoA-I is a principal apolipoprotein component of directly (dLDL-C). In general, dLDL-C increased gradu- HDL. In patients WAS-1/2, -5/6, -7/8, and -11/12, apoA-I ally with sirolimus treatment, resembling the changes levels were not remarkably affected by sirolimus treat- seen in total cholesterol with respect to timing but not ment. However, in patient WAS-3/4, apoA-I rose 58% magnitude (Fig. 1A–F, all parts labeled II). (Fig. 1B, part III), and in patient WAS-9/10 it rose 50% Throughout the entire 6-week treatment period (42 (Fig. 1E, part III) over the 42 day treatment period. The days), sirolimus had no effect on HDL-C levels any of the increase in apoA-I was attended by an increase in HDL for patients besides WAS-3/4. This patient had a remarkably patient WAS-3/4 (Fig. 1B, part II) but not for patient high initial HDL-C level (92 mg/dl), which rose slowly WAS-9/10 (Fig. 1E, part II). and monotonically, reaching a plateau level of 110 mg/dl at day 28 (Fig. 1A, part II). TABLE 4. Plasma free fatty acids levels (mEq/l) of renal transplant Effect of sirolimus on triglyceride levels patients who exhibited hypercholesterolemia and hypertriglyceridemia (type II b, Table 4) during treatment with sirolimus. The mean value In the initial shorter-term study (Table 2), sirolimus in- represent the average of 18 measurements over 24 h while off duced a substantial increase in the triglyceride levels of ev- (-1, -3, -5, -7) and on (-2, -4, -6, -8) sirolimus treatment ery patient (range: 24 to 175%; mean: 123%). For the Patient Mean SD P value later longer-term study (Table 3), sirolimus again elevated triglyceride levels (range: 9 to 254%; mean: 95%), but these changes were not as great as those observed in the WAS-1 0.772 0.294 0.037 29.5 short term study. Only two patients, WAS-1/2 and WAS-7/8, WAS-2 1.000 0.216 WAS-3 0.474 0.186 had elevations greater in the longer-term than the shorter- 0.001 21.3 WAS-4 0.575 0.213 term study (Tables 2 and 3). In general, triglyceride levels WAS-5 0.410 0.160 0.0001 37.0 were highly responsive to sirolimus dosage. This is well il- WAS-6 0.651 0.159 lustrated in responses of patients WAS-3/4 and WAS-9/10, WAS-7 0.411 0.138 0.0001 74.1 WAS-8 0.714 0.249 in which the initial 10 mg/day induced a rapid rise in tri- Mean  SD 21.8  18.9 glyceride levels; necessary reductions in dosage resulted in Morrisett et al. Sirolimus and lipids in renal transplant patients 1175 13 Effect of sirolimus on plasma free fatty acid levels h. The total amount of [ C ]palmitate incorporated into triglyceride is indicated by the integrated area under the After 42 days of sirolimus treatment, four of the six pa- kinetic curve. The areas under the VLDL, IDL, and LDL tients were hypercholesterolemic and hypertriglyceri- curves typically obtained for patients on sirolimus were demic (type Iib) (Table 3). The mean plasma free fatty substantially less than the areas under curves generated acid level of these patients increased by 40.5  16.8% for these patients when off drug. For example, patient (mean  SD) (Table 4). This expansion of the plasma WAS-7/8 had reductions of 50.5%, 56.3%, and 53.2% in free fatty acid pool was explored further with stable iso- [ C ]palmitate incorporation into VLDL, IDL, and LDL, tope kinetic studies. [ C ]palmitate was infused intrave- 4 respectively, when treated with sirolimus. Comparable re- nously over 7 h and its plasma levels monitored for 24 h by ductions in incorporation were also observed for patients GC/MS. In every experiment, the percent atom enrich- WAS-3/4 and WAS-5/6 (Table 6). ment returned to baseline within about 8 h. The shapes of These results suggest that sirolimus expands the plasma the kinetic curves (Fig. 2) differ among patients. However, pool of free fatty acid (mean 42.3%) resulting in in- in all four cases the areas under the curves obtained for creased hepatic synthesis of triglyceride secreted as VLDL patients on sirolimus were substantially less than when the (mean 38.3%). patients were off drug. The reduction in integrated area ranged from 20.3% to 62.7% with a mean  SD of 42.3 17.7 (Table 5). DISCUSSION Effect of sirolimus on triglyceride metabolism The infusion of [ C ]palmitate also made it possible to A primary objective of this study was to determine if the hyperlipidemic effects of sirolimus were reproducible, re- monitor the synthesis of triglyceride and its distribution among the VLDL, IDL, and LDL lipoprotein fractions. versible, and dose-dependent. For this purpose, the study Samples (n 18) were collected at frequent intervals dur- was performed in a small heterogeneous group of renal ing the initial 24 h period, and daily for the next 5 days, transplant recipients with different types of hyperlipi- demia (type IIa, IIb, IV). There were two reasons for resulting in well-defined kinetic curves for triglyceride synthesis. A set of representative curves obtained for a pa- studying these patients under those conditions: First, pa- tient (WAS-7/8) off and on sirolimus treatment is pre- tients were on maintenance regimens typical of many re- sented in Fig. 3. The curves typically indicated maximum nal allograft recipients, thus the effects caused by siroli- mus were the result of the drug acting in the environment enrichment at 6–8 h and returned to baseline by about 48 Fig. 2. Percent atom enrichment in the free fatty acid fractions of plasma collected during/after infusion of [ C ]palmitate into four renal transplant patients before (odd number, solid lines) and after 6 weeks of (even number, dashed lines) sirolimus treatment. The integrated area under each cur ve is tabulated in Table 6. 1176 Journal of Lipid Research Volume 43, 2002 TABLE 5. Expansion of the plasma free fatty acid pool a lesser extent. It is probable that the lipid-lowering ther- apy received by four of the patients (WAS-3/4, -5/6, -7/8, Patient Area Change and -11/12) significantly attenuated the lipid elevating ef- fects of sirolimus, even though their cholesterol and tri- WAS-1 24.38 glyceride levels rose 25–75% and 9–191%, respectively 47.4 WAS-2 12.82 (Table 3). Patient WAS-1/2, who received no lipid lower- WAS-3 30.04 38.7 ing therapy, had the largest absolute increase in choles- WAS-4 18.41 terol ( 187 mg/dl) and triglyceride ( 669 mg/dl) while WAS-5 12.04 20.3 WAS-6 9.59 on the drug. However, patient WAS-9/10, who also re- WAS-7 21.68 62.7 ceived no lipid lowering medication but a comparable WAS-8 8.08 dosage of sirolimus, showed less absolute changes that did Mean  SD 42.3  17.7 not force his lipid values outside the normal range. Thus, Expansion of the plasma free fatty acid pool as measured by dilu- sirolimus did not cause uniform lipid elevation in all of tion of [ C ] palmitate infused into renal transplant patients who ex- our patients. hibited hypercholesterolemia and hypertriglyceridemia (type IIB, Ta- ble 4) during treatment with sirolimus. The values shown represent the The US Phase III clinical studies have also demon- area under the % atom enrichment curve generated during the first strated that the incidence of hyperlipidemia is dependent 24 h after infusion (18 times points) while off (-1, -3, -5, -7) and on (-2, on sirolimus dosage (1). Our results confirm and extend -4, -6, -8) sirolimus treatment. the previously reported hypercholesterolemia and hyper- triglyceridemia observed in a Phase I clinical trial of renal of other agents typically present in transplant patients. transplant patients (28–32). Second, it was not ethically feasible to withdraw a medica- ApoA-I is the principal apolipoprotein component of tion proved to be efficacious in suppressing graft rejection HDL and increases the enzymatic activity of LCAT, a and controlling hyperlipidemia in specific patients. plasma enzyme that catalyzes the conversion of choles- The initial shorter-term sirolimus treatment after renal terol to cholesteryl ester. The apoA-I plasma concentra- transplantation resulted in clinically significant choles- tion is typically 119 mg/dl in normolipidemic subjects terol elevations within 2–4 weeks of treatment, which re- (33). In our study, the baseline apoA-I levels were 80–190 verted to near-baseline levels within 8 weeks after discon- mg/dl and the maximum levels on treatment were 105– tinuing treatment (Table 2). A similar effect of sirolimus 310 mg/dl (Fig. 1A–F, all parts labeled III). In four of the on triglyceride levels in these patients was also observed. six patients, the plasma apoA-I levels did not undergo no- To determine whether these effects were reproducible or table changes, consistent with the rather constant HDL-C whether metabolic adaptations occurred over time, pa- levels of the same patients. In contrast, patients WAS-3/4 tients were rechallenged with sirolimus. Significant in- and WAS-9/10 had baseline values of 190 mg/dl and 100 creases in cholesterol and/or triglyceride levels were re- mg/dl, which rose respectively to a maximum of 305 mg/ inducible in all six patients when re-challenged (Tables 2 dl and 155 mg/dl after 6 weeks of sirolimus therapy. and 3). One of these patients (WAS-9/10) was normolipi- These levels and changes of apoA-I are consistent with the demic (without lipid lowering therapy) before but mildly HDL-C levels of these patients, which increased slowly hypertriglyceridemic after the initial 2 weeks of sirolimus; throughout the duration of the study (Fig. 1B, part II). the same patient was normolipidemic before the later ApoB-100 is a major apolipoprotein component of 6-week re-challenge, and experienced moderate cholesterol VLDL, IDL, and LDL; hence, its concentrations are highly and triglyceride elevation during sirolimus treatment, but associated with triglyceride and cholesterol levels. At base- not to a level that would be considered hyperlipidemic line, apoB-100 concentrations ranged from 30–195 mg/dl (Table 3). Another patient (WAS-11/12) was hypercholes- and reached a maximum level of 120–330 mg/dl (Fig. terolemic and hypertriglyceridemic before and after the 1A–F, all parts labeled III). These values are substantially initial 2 weeks of sirolimus; this patient was normolipi- above the mean value of 90 mg/dl for apoB-100 in normo- demic (with lipid lowering therapy) before the later lipidemic subjects (33). One might expect that our pa- 6-week re-challenge, and experienced moderate choles- tients receiving triglyceride and cholesterol lowering terol elevation during sirolimus rechallenge (Table 3). agents (Table 1) would have the lower apoB-100 concen- These results suggest that some patients develop resis- trations (Fig. 1A–F, all parts labeled III). This did not turn tance to sirolimus-induced hyperlipidemia, even without out to be the case for WAS-7/8, whose baseline value of lipid lowering therapy, while others remain susceptible to 195 mg/dl rose to 330 mg/dl despite his being on gemfi- this effect, even with lipid lowering therapy. brozil and pravastatin. However, these lipid-lowering The present study demonstrates a prompt change in tri- agents were effective in attenuating the elevation of apoB- glyceride levels when sirolimus dosage is altered, while pa- 100 in patients WAS-5/6 and -11/12. These results suggest tients are maintained on CsA and prednisone. The start- that there is considerable inter-subject variability in the ca- ing dose of 10 mg/day induced substantial increases in pacity of pravastatin to upregulate LDL receptors and en- triglyceride levels within 14 days in five of six patients. De- hance apoB-100 removal in these sirolimus-treated pa- creasing the dose from 10 to as low as 0 mg/day either at- tients. tenuated or reversed the escalating triglyceride levels. ApoC-II is an activator (25) and apoC-III is an inhibitor Sirolimus dosage also affected plasma cholesterol levels to (26, 27) of LPL, which hydrolyzes triglyceride in VLDL Morrisett et al. Sirolimus and lipids in renal transplant patients 1177 13 Fig. 3. Incorporation of [ C ]palmitate into triglycerides of VLDL, IDL, and LDL of patient WAS-7/8 be- fore (solid lines) and after 6 weeks of (dashed lines) sirolimus treatment. 1178 Journal of Lipid Research Volume 43, 2002 14 TABLE 6. Change in [ C ]palmitate incorporation into triglycerides of VLDL, IDL, and LDL WAS-3 VLDL WAS-4 VLDL WAS-3 IDL WAS-4 IDL WAS-3 LDL WAS-4 LDL 3.478 1.677 2.153 1.474 0.883 0.824 51.8% 31.5% 6.7% WAS-5 WAS-6 WAS-5 WAS-6 WAS-5 WAS-6 VLDL VLDL IDL IDL LDL LDL 1.333 1.165 1.182 0.728 1.026 0.566 12.6% 38.4% 55.2% WAS-7 WAS-8 WAS-7 WAS-8 WAS-7 WAS-8 VLDL VLDL IDL IDL LDL LDL 1.910 0.946 1.026 0.448 1.294 0.605 50.5% 56.3% 53.2% Mean 38.3% 56.3% 38.4% Change in [ C ] palmitate incorporation into triglycerides of VLDL, IDL, and LDL of three renal transplant patients who exhibited hypercholesterolemia and hypertriglyceridemia (type IIb, Table 4) during treatment with sirolimus. The values shown represent the area under the % atom enrichment curves generated during the first 24 h after infusion of [ C ] palmitate (18 time points) while off (-3, -5, -7) and on (-4, -6, -8) sirolimus treatment. and chylomicrons. The baseline levels of apoC-II were eride of VLDL, IDL, and LDL was measured before and 2.4–12.7 mg/dl compared with the levels observed in nor- during sirolimus treatment in three patients. The mean molipidemics (3 mg/dl) (33). The levels of apoC-III were isotopic enrichment was decreased by 38.3%, 56.3%, and substantially higher than the levels observed in normolipi- 38.4%, respectively (Table 6). Taken together, these re- demics (16 mg/dl) (33), ranging from 20 to 130 mg/dl at sults support the view that sirolimus enhances the action baseline and from 20 mg/dl to 180 mg/dl at maximum of hormone sensitive lipase (HSL) and perhaps also inhib- level (Fig. 1A–F, all parts labeled III). These elevated levels its LPL. These effects are the opposite of those mediated of apoC-III may contribute significantly to the depressed by insulin, suggesting that sirolimus may induce hypertri- levels of LPL activity seen in these immunosuppressed pa- glyceridemia via an insulin-dependent signaling pathway. tients (15). However, even though the apoC-III-apoC-II If the drug interferes with insulin-stimulated triglyceride ratio in three patients (WAS-1/2, -3/4, and -5/6) was sig- storage in adipocytes, this could lead to increased release nificantly higher after sirolimus treatment, these patients of FFAs into the circulation, their increased uptake by the did not exhibit significantly lower LPL activity than the liver, and increased hepatic secretion of VLDL triglycer- other three patients (WAS-7/8, -9/10, and -11/12), whose ides. Alternatively, sirolimus may also decrease FFA oxida- apoC-III/apoC-II ratio was not notably altered (Fig. 1A–F, tion leading to increased FFA availability. all parts labeled III). Massy et al. (34) have compared the The elevated triglyceride levels in patients WAS-1/2, -3/ separate effects of sirolimus and CsA on the plasma con- 4, -5/6, and -7/8 could be due to increased hepatic pro- centration of apolipoproteins and LPL. They observed sig- duction of triglyceride rich lipoproteins and/or de- nificantly higher apoC-II in sirolimus treated patients (7.9 creased removal of them. Our previous study (15) indi- mg/dl) than in CsA treated patients (5.1 mg/dl). Al- cated significant reduction in the fractional catabolic rate though apoC-II levels were higher in the sirolimus treated of apoB-100-containing lipoproteins in patients receiving patients (18.8 mg/dl) than in CsA treated patients (14.1 sirolimus treatment. The present study provides strong ev- mg/dl), this difference was not statistically significant. Im- idence that sirolimus-increased production of triglyceride- portantly, LPL and hepatic lipase activities were the same rich lipoproteins also contributes to the observed hyper- in CsA and sirolimus-treated patients (34). triglyceridemia. A second major goal of this study was to define the In summary, sirolimus induces or exacerbates hyperlipi- mechanism whereby hypertriglyceridemia was induced by demia in a reproducible, reversible, and dose-dependent sirolimus in four of the patients. Toward this end, stable manner in some renal transplant recipients. The clinical isotope experiments were conducted to examine fatty acid implication of these results is that administration of the and triglyceride metabolism before and during sirolimus minimal dose that elicits therapeutic immunosuppression, treatment. [ C ]palmitate infusion experiments indi- as indicated by measurement of plasma sirolimus levels, cated significant expansion of the free fatty acid pool; may be advantageous for minimizing potential adverse ef- mean 42.3%, Table 5 (35) by sirolimus. These results fects on lipid metabolism that occur in some transplant were supported by measurements of total free fatty acid patients. levels, which indicated considerable expansion of this pool (mean 40.5). Although it is possible that some de novo fatty acid synthesis is induced by sirolimus, it is un- This study was supported in part by the Welch Foundation likely that it would cause pool expansion of this magni- (grant Q-1325 to J.D.M); by the National Institutes of Health tude (36). (HL 07812 to J.D.M and DK 38016 to B.D.K.); and by the An expanded fatty acid pool may lead to increased he- Wyeth Ayerst Research Laboratories (to J.D.M. and B.D.K.). patic synthesis of triglycerides. To assess this possibility, The authors acknowledge the excellent technical assistance of the incorporation of infused [ C ]palmitate into triglyc- Karima Ghazzaly. (TTGA). Morrisett et al. Sirolimus and lipids in renal transplant patients 1179 A. W. Wahlefeld, and W. Gruber. 1984. Reagent for the enzymatic REFERENCES determination of serum total triglyceride with improved lipolytic efficacy. J. Clin. Chem. Clin. Biochem. 22: 165–174. 1. Kahan, B. D. 2000. Efficacy of sirolimus compared with azathio- 19. Warnick, G. R., J. Benderson, and J. J. Albers. 1982. Dextran Sul- prine for reduction of acute allograft rejection: a randomized mul- fate-Mg2 precipitation procedure for the quantification of high ticenter study. The Rapamune US Study Group. Lancet. 356 density-lipoprotein cholesterol. Clin. Chem. 28: 1379–1388. (9225): 194–202. 20. Hixson, J. E., and D. T. Vernier. 1990. Restriction isotyping of the 2. Watson, C. J., P. J. Friend, N. V. Jamieson, T. W. Frick, G. Alex- human apolipoprotein E by gene amplification and cleavage with ander, A. E. Gimson, and R. Calne. 1999. Sirolimus a potent new Hha1. J. Lipid Res. 31: 545–548. immunosuppressive agent for liver transplantation. Transplanta- 21. Redgrave, T. G., D. C. K. Roberts, and C. E. West. 1975. Separation tion. 67: 505–509. of plasma lipoproteins by density gradient ultracentrifugation. 3. Stroatman, L. P., and J. G. Coles. 2000. Pediatric utilization of ra- Anal. Biochem. 65: 42–49. pamycin for severe cardiac allograft rejection. Transplantation. 70: 22. Skipski, V. P., and M. Barclay. 1969. Thin Layer Chromatography 541–543. of Lipids (Ed: J.M. Lowenstein). Methods Enzymol. XIV: 530–598. 4. Poon, M., S. O. Marx, R. Gallo, J. J. Badimon, M. B. Taubman, and 23. Hachey, D. L., B. W. Patterson, L. J. Elsas, P. J. Reeds, and P. D. A. R. Marks. 1996. Rapamycin inhibits vascular smooth muscle cell Klein. 1991. Isotopic determination of organic keto acid pentaflu- migration. J. Clin. Invest. 98: 2277–2283. orobenzyl esters in biological fluids by negative chemical ioniza- 5. Burke, S. E., N. L. Lubbers, Y. W. Chen, G. C. Hsieh, K.W. Molli- tion gas chromatography/ mass spectrometry. Anal. Chem. 63: son, J. R. Luly, and C. D. Wegner. 1999. Neointimal formation after 919–923. balloon-induced vascular injury in Yucatan minipigs is reduced by 24. Matthews, D. E., and V. T. Farewell. 1988. Using and understanding oral rapamycin. J. Cardiovasc. Pharmacol. 33(6): 829–835. Medical Statistics. 2nd edition. S. Karger AG, Basel, Switzerland. 113– 6. Gallo, R., A. Padurean, T. Jayaraman, S. Marx, M. Roque, S. Adel- 119. man, J. Chesebro, J. Fallon, V. Fuster, A. Marks, and J. J. Badimon. 25. Bengtsson, G., and T. Olivecrona. 1980. Lipoprotein lipase: some 1999. Inhibition of intimal thickening after balloon angioplasty in effects of activator proteins. Eur. J. Biochem. 106: 549–555. porcine coronary arteries by targeting regulators of the cell cycle. 26. McConathy, W. J., J. C. Gesquiere, H. Bass, A. Tartan, J. C. Fruchart, Circulation. 99: 2164–2170. and C. S.Wang 1992. Inhibition of lipoprotein lipase activity by syn- 7. Brown, E. J., M. W. Albers, T. B. Shin, K. Ichikawa, C. T. Keith, W. thetic peptides of apolipoprotein CIII. J. Lipid Res. 33: 995–1003. Lane, and S. L. Schreiber. 1994. A mammalian protein targeted by 27. Ginsberg, H. N., N. A. A. Le, I. J. Goldberg, J. C. Gibson, A. Rubin- G1- arresting rapamycin-receptor complex. Nature. 369: 756–758. stein, P. Wang-Iverson, R. Norum, and W. V. Brown. 1986. Apolipo- 8. Raught, B., and A. C. Gingras. 1999. EIF4E activity is regulated at protein B metabolism in subjects with deficiency of apo CIII and multiple levels. Int. J. Biochem. Cell. Biol. 31: 43–57. AI. Evidence that apolipoprotein CIII inhibits catabolism of tri- 9. Brunn, G. J., C. C. Hudson, A. Sekulic, J. M. Williams, H. Hosoi, glyceride rich lipoproteins by lipoprotein lipase in vivo. J. Clin. In- P. J. Houghton, J. C. Lawrence, and R. T. Abraham. 1997. Phos- vest. 78: 1287–1295. phorylation of the translational repressor PHAS-1 by the mamma- 28. Murgia, M. G., S. Jordan, and B. D. Kahan. 1996. The side effect lian target of rapamycin. Science. 277: 99–101. profile of sirolimus: a phase I study in quiescent cyclosporine-pred- 10. Burnett, P. E., R. K. Barrow, N. A. Cohen, S. H. Snyder, and D. M. nisone-treated renal transplant patients. Kidney Int. 49: 209–216. Sabatini. 1998. RAFT-1 phosphorylation of the translational regulator 29. Kahan, B. D., J. Podbielski, K. L. Napoli, S. M. Katz, H. U. Meirer- p70S6 kinase and 4EBP-1. Proc. Natl. Acad. Sci. USA. 95: 1432–1437. Kriesche, and C. T. Van Buren. 1998. Immunosuppressive effects 11. Kumar V., D. Sabatini, P. Pandey, A-C. Gingras, P. K. Majumder, M. and safety of a sirolimus/cyclosporine combination regimen for Kumar, Z-M. Yaun, G. Carmicheal, R. Weichselbaum, N. Sonen- renal trasnplantation. Transplantation. 66: 1040–1046. berg, D. Kufe, and S. Kharbanda. 2000. Regulation of the rapa- 30. Kahan, B. D. 1998. Rapamycin: personal algorithms for use based mycin and FKBP-target 1/mammalian target of rapamycin and on 250 treated allograft recipients. Transplant. Proc. 30: 2185–2188. cap-dependent initiation of translation by the c-abl protein-tyro- 31. Mac Donald, A. S. 2001. The Rapamune Global Study Group: a sine kinase. J. Biol. Chem. 275: 10779–10787. randomized, placebo-controlled trial of rapammune in primary 12. United States Renal Data System (USRDA). 1998. The National In- renal allograft recipients. Transplantation. 71: 271–280. stitute of Health, National Institute of Diabetes, and Digestive and 32. Kahan, B. D., K. L. Napoli, P. A. Kelly, J. Podbielski, I. Hussein, Kidney Disease. Annual Data report, Bethesda, Maryland. D. L. Urbauer, S. H. Katz, and C. T. Van Buren. 2000. Therapeutic 13. Brattström, C., H. E. Wilczek, G. Tydén, Y. Böttiger, J. Säwe, and C. G. drug monitoring of sirolimus: correlations with efficacy and toxic- Groth. 1998. Hypertriglyceridemia in renal transplant recipients ity. Clin. Transplant. 14: 97–109. treated with sirolimus. Transplant. Proc. 30: 3950–3951. 33. Labeur, C., and M. Rosseneu. 1997. Immunological assays of 14. Brattström, C., H. Wilczek, G. Tyden, Y. Böttiger, J. Säwe, and C. G. apolipoproteins A-II, A-IV, C-I, C-II, C-III in plasma. In Hand- Groth. 1998. Hyperlipidemia in renal transplant recipients treated book of Lipoprotein Testing. N. Rifai, R. G. Warnik, and M. H. with Sirolimus (Rapamycin). Transplantation. 65: 1273–1274. Dominiczak, editors. ASCP Press, Chicago. 315–328. 15. Hoogeveen, R. C., C. M. Ballantyne, H. J. Pownall, A. R. Opekun, 34. Massy, Z. A., J. P. DeBandt, E. Morelon, M. Thevenin, B. Lacourr, D. L. Hachey, J. S. Jaffe, S. Oppermann, B. D. Kahan, and J. D. and H. Kreis. 2000. Hyperlipidemia and post-heparin lipase activi- Morrisett. 2001. Effect of rapamycin on the metabolism of Apo ties in renal transplant recipients treated with sirolimus or cy- B-100 containing lipoproteins in renal transplant patients. Trans- closporin A. Nephrol. Dial. Transplant. 15: 928. plantation. 72: 1244–1250. 35. Morrisett, J. D., R. C. Hoogeveen, G. Abdel-Fattah, H. J. Pownall, 16. Napoli, K. L., and B. D. Kahan. 1996. Routine clinical monitoring C. M. Ballantyne, A. R. Opekun, D. L. Hachey, J. S. Jaffe, S. Opper- of sirolimus (rapamycin) whole-blood concentrations by HPLC mann, and B. D. Kahan. 2001. Effect of rapamycin on free fatty with ultraviolet detection. Clin. Chem. 42: 1944. acid metabolism in renal transplant patients. Arterioscler. Thromb. 17. Seidel, J., E. Hagele, J. Ziegenhorn, and J. Wahlenfeld. 1983. Re- Vasc. Biol. 21: 654. agents for the enzymatic determination of serum total cholesterol 36. Hellerstein, M. K., R. A. Neese, and J-M. Schwarz. 1993. Model for with improved lipolytic efficiency. Clin. Chem. 29: 1075–1080. measuring absolute rates of hepatic de novo lipogenesis and rees- 18. Nagele, U., E. Hagele, G. Sauer, E. Wiedermann, P. Lehmann, terification of free fatty acids. Am. J. Physiol. 265: E814–E820. 1180 Journal of Lipid Research Volume 43, 2002

Journal

Journal of Lipid Research – Unpaywall

Published: Aug 1, 2002

There are no references for this article.