|
Advertisement | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Lipid Research, Vol. 47, 1859-1864, August 2006
Optimized conditions for measuring lipolysis in murine primary adipocytes
Laboratory of Cellular and Developmental Biology, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20892-8028 Published, JLR Papers in Press, May 4, 2006.
1 To whom correspondence should be addressed. e-mail: deanl{at}intra.niddk.nih.gov
The current literature on lipolysis in murine primary adipocytes is rife with experiments performed under conditions not optimized for reproducible and reliable results. Here, we present conditions for optimizing the measurement of lipolysis in murine adipocytes. We demonstrate that adenosine management is of paramount importance in evaluating the lipolytic response under basal and stimulated conditions. Also, adipocyte concentrations in the 10,00015,000 cells per milliliter range produce a greater increase in stimulated lipolysis than higher concentrations, and the response is further enhanced by agitating the cells.
Supplementary key words glycerol adenosine cell concentrations
Nearly 20 years ago, we embarked on a series of experiments to explore the basis for the high variability among published studies on lipolysis in isolated adipocytes. At that time, the vast majority of such studies were performed with rat adipocytes, isolated according to the classical method of Rodbell (1). We published several papers on the handling and manipulation of isolated adipocytes (24) required to provide optimal and reproducible results in the measurement of lipolysis, because most published studies revealed a high level of variability in results. In recent years, an increasing number of papers have appeared on the behavior of isolated murine adipocytes, and once again, the literature reveals a high level of variability in the magnitude of stimulation achieved with ß-adrenergic receptor agonists or with other stimulants that increase cAMP concentrations and PKA activity. In our earlier studies, several factors emerged as important for the reproducible assay of adipocyte lipolysis, not the least of which was careful control of the ambient adenosine concentration (4). Adipocytes contain a high-affinity A1 adenosine receptor that is linked to Gi, the adenylyl cyclase inhibitory G protein. For two reasons, it is important to control the adenosine concentration. First, in the absence of adenosine, adipocytes may exhibit constitutively high lipolytic activity. Second, based on various factors, such as the season of the year and immediate dietary history, basal activity may be relatively high, thus blunting the magnitude of stimulation over basal that may be achieved with a lipolytic stimulant (i.e., signal-to-noise ratio). Another important feature is the control of fatty acids that are released upon stimulation of lipolysis. Fatty acids are strongly inhibitory toward the receptor-mediated activation of adenylyl cyclase (5). Appropriately low fatty acid concentrations are achieved using low adipocyte concentrations, such that the ambient BSA in the assay medium is sufficient to bind all free fatty acids. Likewise, adequate mixing of the incubation medium is also essential to provide access of free fatty acids to the medium BSA. Otherwise, the adipocytes rise to the top of the incubation tube and present excess fatty acids to adjacent adipocytes in the upper layer, thus inhibiting adenylyl cyclase activity. Here, we present conditions for obtaining reproducible and optimal results for measuring lipolysis in primary murine adipocytes.
Reagents All reagents were obtained from Sigma-Aldrich (St. Louis, MO) and were prepared and manipulated in an Adipocyte Incubation Solution (AIS) unless indicated otherwise. The AIS contained Krebs Ringer Bicarbonate HEPES buffer (containing 10 mM bicarbonate and 30 mM HEPES, pH 7.4) supplemented with 3% (w/v) fatty acid free bovine albumin fraction V (number 820025; ICN Biomedical, Inc.). Where indicated, the adenosine deaminase-resistant (ADA) A1 receptor agonist, ()-N6-(2-phenyl-isopropyl)-adenosine (PIA), was present at 100 nM, ()-isoproterenol (ISO) was present at 10 µM, and ADA was present at 1 U/ml. It is important to use enzyme stock supplied in ammonium sulfate and not glycerol suspensions, because the glycerol content interferes with the glycerol assay used in these studies. The cells were isolated in AIS fortified with 500 nM adenosine and 3 mg/ml type 1 collagenase (Worthington Biomedical Corp., Lakewood, NJ). The optimal concentration of adenosine used for adipocyte isolation and washes was based on the finding that mouse adipocytes were 10-fold less sensitive compared with rat adipocytes (J. Tansey and C. Londos, unpublished observation). Because of lot-to-lot variation in collagenase activity, the optimal concentration of collagenase may range from 1 to 3 mg/ml; we determined that 3 mg/ml collagenase stock was optimal for the digestion of adipose tissue samples in this study.
Isolation of adipocytes
Lipolysis
Glycerol assay
Statistical analysis
A comparison of recently published studies on the lipolysis of isolated primary murine adipose cells reveals a wide range of responsiveness to lipolytic stimulation (Table 1 ). Although several publications report stimulations of 30-fold or greater, the majority show much more modest stimulations of <10-fold, and some as little as 3-fold or less. It is difficult to evaluate such literature because many papers do not reveal the conditions under which basal activity was measured, and most do not provide cell concentration values. The data below will address the manipulations that may be used to enhance the magnitude of stimulation, either by reducing the basal value or by amplifying the stimulated values. The important variables to consider are 1) cell concentration, 2) agitation or shaking of the incubation mixture, and 3) composition of the incubation medium, especially management of the ambient adenosine.
The importance of adenosine management is illustrated in Fig. 1 , which reveals glycerol release in the presence or absence of PIA, ADA, or both in 10K cells under basal conditions when the tubes were incubated without shaking (Fig. 1A) or with shaking at 150 rpm (Fig. 1B). Removal of PIA or PIA and ADA from the cell incubation medium provoked a significant increase in basal lipolysis compared with tubes in which only ADA was removed or both PIA and ADA were added. Removal of ADA from the incubation mixture did not affect the stimulated lipolysis. Glycerol release was 2- to 3-fold higher with shaking at 150 rpm compared with tubes that were incubated without shaking. Note especially the high basal release of glycerol in the presence of ADA only, which converts adenosine to inosine, which is not recognized by adenosine receptors. There are two sources of adenosine to consider under our conditions. First, we add exogenous adenosine to suppress lipolysis during the course of cell isolation and washing; second, adenosine will be present secondary to cell lysis leading to leakage of adenylyl nucleotides, which are hydrolyzed to adenosine. Note that in the absence of any adenosine receptor agonist (PIA, +ADA), basal activity may be extraordinarily high, blunting the ratio of stimulated-to-basal activity. Such data replicate the findings we reported with rat adipocytes under what we termed the "ligand-free" condition. On the other hand, the addition of PIA, the adenosine receptor agonist, or exclusion of ADA exerts a powerful adenosine receptor-mediated inhibition of basal activity, and stimulation with ISO under these adenosine replete conditions (ADA) increases to nearly 40- to 50-fold. It is also evident that both basal and stimulated activities are enhanced upon shaking during incubation of the cells. It should be noted that with the identical cell preparation, one may easily elicit a wide range of ISO-mediated stimulations, from 2-fold up to nearly 50-fold, by merely adjusting incubation conditions. In contrast to our findings, the addition of ADA to the incubation medium did not alter basal lipolysis in human adipocytes (9, 10). These discrepancies may be attributed to the very low concentration of adipocytes in the human cell incubation mixture (<5,000 vs. 10,000 cells per incubation) or to species differences (human vs. mouse).
In our earlier studies on rat adipocytes, we did not highlight the importance of using appropriate cell concentrations during the measurement of lipolytic rates (3, 4). The importance of cell concentration and agitation (shaking) are illustrated in Fig. 2A , B. Under both conditions (i.e., with or without shaking), the magnitude of stimulation of glycerol release by ISO over basal conditions decreased dramatically with increasing cell concentrations. Moreover, shaking the tubes at 150 rpm increased stimulated lipolysis by 3-fold in the 5K and 10K samples compared with tubes that were not agitated during incubation. We had shown previously that without shaking there is a rapid increase in cAMP shortly after the application of ISO, followed by a steep decline (11). This phenomenon was known as "peaking," which resulted from fatty acid-mediated inhibition of cAMP formation. Edens, Leibel, and Hirsch (12) demonstrated a reduction in fatty acid reesterification in shaken adipocyte incubations compared with unshaken incubations. Increased shaking speed provides for greater mixing of the fatty acids with the medium BSA, which binds and neutralizes the fatty acids. With more rapid shaking, the peaking disappears and activation of PKA reaches a steady state that does not decline with time (3, 11).
Optimal shaking speed is a function of the geometry of the incubation vessel and the volume of the incubation medium. The shaking speed should be such that the adipocytes are distributed uniformly throughout the medium and do not rise to the surface and become concentrated near the top of the tube. On the other hand, shaking should not be so vigorous that cells are damaged. Optimal shaking speed may be determined by visual inspection of the incubation tube: the mixture should present as a uniform milky solution without a visible upper layer of concentrated adipocytes. Thus, the shaking speed can be reduced when a larger tube or flask is used for incubation.
Even at optimal shaking speeds, the negative effects of high cell concentrations are not overcome (Fig. 2B). High cell concentrations compounded with higher medium FFA concentrations reduced reesterification and tended to decrease the rate of lipolysis in human adipocytes (12). Similarly, increasing adipocyte concentration from 0.168 x 106 to 1.252 x 106 cells/ml reduced the conversion of glucose to glyceride-glycerol but increased the production of glyceride-fatty acids (13). In the current experiment, the increase from 10,000 to 20,000 cells per milliliter crossed from the optimal to the deleterious range of cell concentration. In general, optimal results were obtained with Data normalization among adipocyte preparations is a critical factor in interpreting results. It is imperative that data be normalized according to cell number. Normalization to protein concentration is not advisable, because the vast majority of protein in a typical adipocyte suspension is the BSA added to protect the cells, whereas the protein contributed by the cells is but a minuscule fraction of the total protein present. Similarly, measurement of DNA values may be misleading, because much of the DNA present may be contributed by smaller cells that cling to the adipocytes (15). Typically, estimation of cell numbers using DNA values results in a 5- to10-fold overestimation. (Unfortunately, the problem with using DNA measurements to obtain cell numbers is not well documented in the current literature. We were alerted to this discrepancy by Dr. Susan K. Fried and confirmed that DNA measurements led to a considerable overestimation of cell numbers over the values obtained by calculating cell number using size measurement and lipid content, as described herein.) Thus, the most efficacious way to normalize lipolysis data is to obtain a cell count, as described by Fine and Di Girolamo (6).
It should be noted that the 50-fold stimulations observed in this study were obtained with relatively young mice on a standard chow diet with adipocytes modest in size ( It should also be understood that the basal activities, as well as the magnitude of stimulation achieved under the experimental conditions described here, may not reflect those activities in vivo. These conditions are designed to permit comparisons of adipocyte behavior in different mouse models as measured by different investigators. The so-called basal activities measured in the present studies are most likely artificially low and are designed to permit viewing of maximal stimulated activities. Hormone-sensitive lipase had for many years been considered the major, if not the only, lipase responsive during the course of lipolytic stimulation (17). More recently, the demonstration that adipocyte triglyceride lipase may contribute significantly to this reaction has introduced another variable that must be considered in the lipolytic reaction (18). Also, any analysis of the protein kinase A-dependent lipolytic reaction must consider the role of perilipin A. In any careful dissection of the lipolytic reaction, especially upon introduction of various siRNAs, it is essential to determine the different contributions to the lipolytic reaction by different lipases and perilipin (19, 20). Such analyses, therefore, will require a rigid and reproducible method for measuring both basal and hormone-stimulated lipolysis.
This project was supported by the National Institute of Diabetes and Digestive and Kidney Diseases/National Institutes of Health intramural research program. Manuscript received February 7, 2006 and in revised form March 23, 2006 and in re-revised form April 18, 2006.
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Advertisement | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||