MITOCHONDRIAL OXIDATIVE SUBSTRATE SELECTION IN PORCINEBLADDER SMOOTH MUSCLE

CHRISTOPHER D. HARDIN,* BRIAN D. KLEIBER AND TINA M. ROBERTSFrom the Department of Physiology, University of Missouri, Columbia, Missouri

ABSTRACT

Purpose: Alterations in bladder smooth muscle (BSM) metabolism due to alterations in plasmalipid levels may be important with the increasingly high fat diets eaten by most Americans. Todetermine the susceptibility of BSM to lipotoxicity we examined the normal pattern of mitochon-drial substrate selection in BSM and the ability of BSM to respond to changes in metabolicsubstrate provision.

Materials and Methods: BSM strips were incubated in 5 mM 1-13C-glucose and 0 to 5 mM1,2-13C-acetate. The pattern of substrate use measured by 13C-nuclear magnetic resonance usingBSM extracts. BSM was also cultured for 4 days to elicit changes in cell phenotype.

Results: At physiological levels of glucose and acetate about 50% of the substrate used bymitochondria was glucose. When acetate concentration was changed from physiological levels(0.1 mM) to pathophysiological levels (0.5 mM), BSM was able to increase the use of acetate,while sparing the use of glucose and intracellular substrates, likely lipids. Above 0.5 mM acetateBSM was unable to further use acetate. With increasing acetate use anaplerosis increased,consistent with a depletion of tricarboxylic acid cycle intermediates. After 4 days of organ cultureBSM mitochondria used significantly more unlabeled intracellular substrates and less 13Clabeled glucose than control bladder, consistent with metabolic adaptation to increase lipid use,such as what occurs with hyperlipidemia.

Conclusions: We conclude that BSM has modest plasticity of the pattern of mitochondrialsubstrate selection and excess lipid provision may be able to induce lipotoxicity in BSM, resultingin impaired detrusor function.

KEY WORDS: bladder; muscle, smooth; swine hyperlipidemia; mitochondria

Bladder smooth muscle (BSM) is the primary cellular com-ponent of the bladder wall and it is responsible for urineoutput from the bladder. Although considerable study hasfocused on BSM cellular metabolism with respect to theresponses to hypoxia1–3 and the balance between glycolyticand oxidative metabolism,2, 4 there is little understanding ofthe oxidative substrate selection of BSM. Mitochondria are keyin determining cell fate, including whether cells undergo apo-ptosis and whether smooth muscle will undergo phenotypetransformation,5 and they are a key cellular target for lipotox-icity.6 Understanding normal BSM oxidative substrate selec-tion and changes in substrate selection in disease states isimportant for our understanding bladder function in normaland disease states such as diabetes and dyslipidemia. Indeed, invascular smooth muscle diabetes and dyslipidemia resulted inalterations in smooth muscle oxidative substrate selection thatmay be associated with phenotype change in smooth muscle.7Lipid has been shown to be the primary oxidative substrate ofother smooth muscles.8,9 If lipid oxidation does not increase inresponse to increased blood lipid levels (dyslipidemia), lipidsmay accumulate within BSM, resulting in a change in BSMphenotype and, hence, in detrusor function. We characterizedthe normal pattern of oxidative substrate use and determinedwhether controlled organ culture, a technique that has rou-tinely been shown to alter smooth muscle phenotype, alters theoxidative substrate use pattern in BSM.

MATERIALS AND METHODS

Tissue preparation. Pig bladders obtained from a local ab-attoir within approximately 30 minutes of slaughter wereplaced in physiological saline solution (PSS) at pH 7.4 pre-equilibrated by bubbling with a gas mixture of 95% O2 and5% CO2. PSS was composed of 116 mM NaCl, 4.6 mM KCl,1.16 mM KH2PO4, 25.3 mM NaHCO3, 2.5 mM CaCl2, 1.16mM MgSO4 and 40 mg/l gentamycin sulfate to avoid micro-bial contamination. Bladders were dissected free of loose fat.The urothelial layer (transitional epithelium and submucosalcoat) was removed, as verified grossly by visual inspection.The bladder segments used were from the body of the blad-der. They were cut into strips about 25 mm long, 10 mm wideand 3 to 5 mm thick, which were pre-incubated at 37C for 30minutes in PSS equilibrated with 95% O2-5% CO2.

Metabolic incubation. Unmounted bladder strips (about 0.6to 1 gm) were incubated at 37C in PSS containing 5 mM 1-13Cglucose and 0, 0.1, 0.5, 1.0 or 5.0 mM 1,2-13C acetate for 6hours. After the incubation period the tissue were brieflyrinsed in fresh PSS without substrate, rapidly blotted,weighed and frozen in liquid nitrogen. In previous studiesusing vascular smooth muscle7, 8 this incubation period wasneeded to provide adequate signal for the 13C-glutamate usedas the indicator molecule for mitochondrial substrate use.

Tissue extracts. Frozen BSM strips to be used for nuclearmagnetic resonance (NMR) analysis were pulverized in liq-uid nitrogen with a mortar and pestle. Perchloric acid/meth-anol extracts were made and treated as previously describedfor 13C NMR analysis.7, 8

13C-isotopomer analysis of glutamate. The 2 13C labeledsubstrates supplied in incubation and the endogenous unla-beled substrates generate 3 possible labeling patterns of

Accepted for publication May 16, 2003.Supported by National Institutes of Health Grant DK55039.* Corresponding author: Department of Physiology, MA-415

Medical Sciences Building, University of Missouri-Columbia,Columbia, Missouri 65212 (telephone: 573-884-6209; FAX: 573-884-4276; e-mail:hardinc@missouri.edu).

0022-5347/03/1705-2063/0 Vol. 170, 2063–2066, November 2003THE JOURNAL OF UROLOGY® Printed in U.S.A.Copyright © 2003 by AMERICAN UROLOGICAL ASSOCIATION DOI: 10.1097/01.ju.0000091872.33814.70

2063

acetyl-coenzyme A (CoA) (1,2-13C acetyl-CoA, 2-13C acetyl-CoA and unlabeled acetyl-CoA). The contribution of 1,2-13Cacetyl-CoA, 2-13C acetyl-CoA and unlabeled acetyl-CoA tothe labeling pattern of glutamate represents the entry ofsubstrate to the TCA cycle via acetyl-CoA and, thus, viacitrate synthase. Substrate use of 1,2-13C acetate was deter-mined by dividing the C-4 quartet caused by 1,2-13C acetate(QA) by total C-4 resonance and multiplying by the ratio oftotal C-4-to-C-3 resonance. Because all relative substrateinput via citrate synthase must equal 100%, unlabeled inputcould be quantified as equal to 100% minus the sum of thepercent input from glucose and the percent input from ace-tate. Input to the TCA cycle by anaplerotic pathways wascalculated, as we have previously described.7, 8

Glycogen synthesis incubation. Biologically paired bladderstrips were incubated in PSS containing 5 mM glucose or 5mM glucose and 0.5 U/ml insulin to determine the extent ofglycogen loading during 0, 2, 4, 6 and 8 hours. At each timepoint 1 piece from each paired tissue was removed and rinsedwith PSS prior to being blot weighed and frozen to assayglycogen content.10

Organ culture. At day 0 bladder strips weighing approxi-mately 10 gm were placed in 1 l PSS containing 5 mM1-13C-glucose and 1 mM 1,2-13C-acetate for up to 4 days at37C. Solutions were continuously circulated using a peristal-tic pump with an in-line 0.22 �m filter to decrease contami-nation. At each 24-hour time point 3 strips of bladder (ap-proximately 2.5 gm) were removed, rinsed with PSS, blotweighed and frozen for extraction. Fresh solution was ex-changed with the volume/tissue mass remaining constant.

Statistical analysis. Correlation coefficients were determinedusing Excel, version 7.0 (Microsoft, Redmond, Washington). Thesignificance of r was determined using a 2-tailed test with signif-icance at � � 0.05 and 0.01.11

RESULTS

To determine the oxidative metabolic profile of BSM weused 13C-NMR analysis of the incorporation of 13C label intothe third and fourth carbons of glutamate (C-3 and C-4 glu-tamate). This technique quantifies the input of select sub-strates into the TCA cycle via citrate synthase, as we havepreviously described for normal hog carotid artery smoothmuscle.8 Figure 1 shows a typical 13C-NMR spectrum of theC-3 and C-4 glutamate regions from BSM incubated with 5mM 1-13C-glucose and 5 mM 1,2-13C-acetate. QA and thedoublet from glucose were used for the isotopomer analysis.

There was sufficient signal to quantify changes in substrateuse in porcine BSM (fig. 1).

Incubation of porcine BSM with 5 mM 1-13C-glucose as theonly labeled substrate resulted in a fractional entry of glu-cose into the TCA cycle via acetyl-CoA of 0.64 � 0.10, whileapproximately the remaining third of substrates responsiblefor entry into the TCA cycle by citrate synthase was ac-counted for by unlabeled substrates, such as endogenouslipids (fig. 2). However, the provision of 0.1, 0.5, 1 or 5 mMacetate substantially decreased the mean fractional entranceof glucose � SEM to 0.48 � 0.00, 0.22 � 0.02, 0.16 � 0.01 and0.16 � 0.04, respectively. As acetate concentration was in-creased from a physiological level (0.1 mM) to levels found indiabetes (0.5 mM), glucose oxidation decreased to less than20% of all oxidized substrates, indicating the strong relianceof BSM oxidative metabolism on nonglycolytic substrates.The decrease in glucose oxidation resulting from the provi-sion of up to 0.5 mM acetate was primarily accounted for byan increase in oxidation of exogenous acetate (fig. 2). How-ever, the oxidation of unlabeled endogenous carbons de-creased when acetate was present at an acetate concentra-tion of up to 0.5 mM. Changes in acetate concentration from0.5 to 5 mM did not significantly alter glucose or unlabeledsubstrate use.

Although the data accounted for all entry of substrates intothe TCA cycle via acetyl-CoA (fig. 2), 13C-isotopomer analysisalso provides quantitative measures of anaplerotic flux. Fig-ure 3 shows the anaplerotic flux observed in the same sam-ples as those used for the previous analysis (fig. 2). As exog-enous acetate concentration increased from 0.1 to 5 mM,there was a trend toward decreased total anaplerotic flux.However, statistical significance was only observed betweenthe 0.1 and 5 mM acetate levels (p � 0.026). The source of theanaplerotic substrates could not be identified in these exper-iments.

The unlabeled substrates that accounted for approximately10% to 30% of TCA cycle entry via acetyl-CoA were intracel-lular in origin since the only exogenous substrates providedwere 1-13C-glucose and 1,2-13C-acetate. However, the natureof the unlabeled substrates was unknown and could haveincluded intracellular lipids, glycogen or amino acids. Likeother smooth muscles, BSM is capable of storing glycogenand using glycogen as a substrate during contraction, al-though glycogen turnover is only modestly sensitive to insu-lin levels.12 Since the incubation in these studies was per-formed under conditions of rest, it was of interest todetermine whether glycogen could undergo net use during aperiod of rest. BSM segments were incubated in the presence

FIG. 1. Representative extract 13C-NMR spectrum of normal pigbladder incubated in presence of 5 mM 1-13C glucose and 5 mM1,2-13C acetate at 37C for 6 hours shows region where fourthcarbon (C-4) of glutamate resonates. Peaks are references to3-(trimethylsilyl)-1-propane-sulfonic acid (TMSPS) peak at 0 ppm.Resonances for doublet of glucose (DG), and quartet from acetate(QA) and area of third carbon (C-3) of glutamate were used incalculations.

FIG. 2. Fractional entry of 1-13C glucose, 1,2-13C acetate and un-labeled substrate into TCA cycle via citrate synthase during incuba-tion of BSM in PSS containing 5 mM 1-13C glucose and 0, 0.1, 0.5, 1.0or 5.0 mM 1,2-13C acetate. Fractional entry of 1.0 indicated that allTCA cycle flux via acetyl-CoA was derived from that substrate.

BLADDER SMOOTH MUSCLE METABOLISM2064

of 5 mM glucose or 5 mM glucose with 0.5 U/ml insulin tocharacterize the insulin sensitivity of glycogen metabolism inBSM. BSM in the absence of insulin showed a 0.55 �mol/gmper hour synthesis rate of glycogen (glucosyl units) based onlinear regression analysis, while in the presence of insulinthe rate increased to a 0.90 �mol/gm per hour synthesis rateof glycogen (glucosyl units), similar to that observed in vas-cular smooth muscle.13 Glycogen levels were significantlyhigher in insulin treated smooth muscle strips comparedwith levels in strips without added insulin at 6 and 8 hours ofincubation with total glycogen levels attaining almost 10�mol/gm glycosyl units of glycogen (fig. 4). Therefore, BSMhas modest sensitivity to insulin for glycogen metabolism. Inaddition, glycogen levels did not decrease during 8 hours ofincubation in the presence of glucose, indicating that glyco-gen breakdown is not a contributor to the entry of substratesinto the TCA cycle in the experiments shown in figure 2.

It has been shown that smooth muscle can be induced toalter phenotype in models of organ culture.14 The inducedproliferative smooth muscle phenotype may be similar towhat occurs in disease states such as diabetes or dyslipide-mia. To determine whether short-term organ culture in aserum-free PSS could result in alterations in the oxidativemetabolic profile of BSM we incubated bladder strips for 1 to4 days in the presence of 13C-labeled glucose and acetate asthe only metabolic substrates (fig. 5). We found that acetate

use significantly decreased with 4 days compared with 1 dayof organ culture (p � 0.037). Indeed, when comparing acetateuse at the 1 mM concentration between noncultured BSMstrips and cultured bladder strips (fig. 2), acetate use wassignificantly decreased in cultured bladder strips for 1 day(p � 0.02) and 4 days (p � 0.002) of culture. Therefore, organculture of BSM and the likely phenotype change of smoothmuscle resulted in decreased acetate use by mitochondria.

DISCUSSION

Little study has been done of the regulation of oxidativemetabolism and the pattern of substrate use in BSM. Thepattern of substrate use is of interest in terms of understand-ing how BSM may respond to short-term metabolic stressessuch as ischemia or hypoxia as well as to long-term metabolicstresses such as chronic partial outlet obstruction or diabe-tes/dyslipidemia. Most studies of BSM metabolism have fo-cused on the relative balance of glycolysis and oxidativemetabolism, and the response of BSM to hypoxia.1–4 How-ever, the relative ability of mitochondria to use various oxi-dative fuels, including lipids, determines how well BSM re-sponds to circulating high lipid levels since nonadipocytetissues chronically exposed to high lipid levels are subject tolipitoxicity.15 Lipotoxicity can result in the cellular uptake oflipid outpacing the ability of mitochondria to use lipid, re-sulting in the accumulation of lipid droplets, eventually mi-tochondrial dysfunction, and even apoptosis and alterationsin cell phenotype. This scenario may lead to a phase ofbladder hypertrophy, as observed with partial outlet obstruc-tion. Therefore, knowledge about the ability of BSM to adaptsubstrate selection during changes in substrate patterns isimportant to our understanding of BSM metabolism in theface of chronic metabolic challenges, such as type 2 diabetes.

In the current study we were able to vary systematicallyacetate levels, as we have previously done in vascular smoothmuscle,8 to examine whether the Randle effect,16 a phenom-enon of metabolic cross-talk between glucose and lipid use,can occur in BSM. Vascular smooth muscle shows little al-teration in glucose metabolism in the face of large changes inacetate8 or oleate.17 However, BSM glucose metabolism sub-stantially decreased as acetate was increased from 0 to 0.5mM. Furthermore, the use of unlabeled substrates, likelypredominantly intracellular lipids, also decreased as acetatewas increased from 0 to 0.5 mM. Notably increases in acetatefrom 0.5 to 5 mM resulted in no further changes in the use ofexogenous 1-13C-glucose or unlabeled substrates. Therefore,

FIG. 3. Anaplerosis expressed as fractional entry into TCA cyclenot via citrate synthase during incubation of BSM in PSS containing5 mM 1-13C glucose and 0, 0.5, 1.0 or 5.0 mM 1,2-13C acetate.Fractional entry of 1.0 indicated that all TCA cycle flux not viaacetyl-CoA was from anaplerosis.

FIG. 4. Insulin sensitivity of glycogen synthesis in BSM incubatedin PSS with 5 mM glucose in presence (triangles) and absence (cir-cles) of 0.5 U/ml insulin.

FIG. 5. Fractional entry of 1-13C glucose, 1,2-13C acetate and un-labeled substrate into TCA cycle via citrate synthase on days 1 and4 of controlled organ culture of BSM in PSS containing 5 mM 1-13Cglucose and 1.0 mM 1,2-13C acetate.

BLADDER SMOOTH MUSCLE METABOLISM 2065

the effects of alterations of acetate on BSM oxidative sub-strate selection occurred in the physiological range of acetatefound in normal (0.1 mM) and diabetic (0.5 mM) states.18

Figure 2 shows substrate inputs that represent only theentry of substrates into the TCA cycle via citrate synthase asacetyl-CoA. However, other inputs of substrates to the TCAcycle can occur through anaplerosis. In vascular smooth mus-cle the extent of anaplerosis did not change as acetate levelswere changed from 0.1 to 5 mM. However, in BSM there wasa significant decrease in anaplerosis at 1 mM compared with0.1 mM acetate.

Although the bladder denuded of the urothelium has blad-der smooth muscle as the great majority of the cell mass, thebladder is also innervated and it contains blood vessels and,hence, vascular smooth muscle. These small populations ofcells may have contributed to the metabolic signals meas-ured. However, since these cells contribute only a small per-cent to detrusor cell mass, the overall patterns of changes inoxidation substrate selection reflected those occurring inbladder smooth muscle cells.

We have previously reported that vascular smooth musclefrom diabetic/dyslipidemic swine shows a decrease in acetateuse and an increase in glucose use. Organ culture of smoothmuscle has routinely resulted in alterations in smooth mus-cle phenotype from a more contractile phenotype to a moreproliferative phenotype.14 In the current study we used con-trolled organ culture and measured changes in the pattern ofsubstrate use. We found a decrease in 1,2-13C-acetate usewith an increase in the use of endogenous substrates. Thedecreased use of acetate may indicate the decreased use ofshort chain fatty acids. In a diabetic/dyslipidemic milieu highplasma lipids have been proposed to result in lipotoxicity ofnonadipose tissue by the inability of nonadipose tissue to uselipid as quickly as uptake occurs, leading to lipid accumula-tion and cellular injury.15 Although we do not present directevidence of lipotoxicity in BSM, our data suggest that alter-ations in BSM oxidative substrate use can occur in organculture, consistent with altered fatty acid use in smoothmuscle of the proliferative phenotype. Although these inter-pretations are speculative, they suggest that BSM may sus-ceptible to lipotoxicity and subsequent detrusor dysfunction.

Aging and disease states such as diabetes have been shownto result in decreased lactate production by BSM under nor-moxic and hypoxic conditions.3, 4 Importantly hypertrophiedsmooth muscle from bladder, such as occurs with partialoutlet obstruction, resulted in the increased ability of BSM tomaintain force during hypoxia, while hypertrophied BSM isalso less sensitive to re-oxygenation injury following hypox-ia.1 Therefore, alterations in mitochondrial substrate selec-tion, which include alterations in the percent of glucose oxi-dation, may be involved in a number of bladder diseasestates.

CONCLUSIONS

Porcine BSM is capable of using glucose or short chainfatty acids as the primary oxidative fuel depending on theavailability of substrates over a physiological range. Further-more, since BSM has limited ability to increase lipid use andBSM shows altered mitochondrial substrate selection in or-gan culture, our study suggests BSM may be susceptible to

lipotoxicity. Although to our knowledge the metabolic simi-larities between human and porcine detrusor are not estab-lished, porcine models are often considered excellent modelsof human cardiovascular disease. Thus, they may also begood models of urogenital disease, such as bladder dysfunc-tion.

REFERENCES

1. Levin, R. M., English, M., Barretto, M., Dubuc, M., O’Connor, L.,Leggett, R. et al: Normal detrusor is more sensitive thanhypertrophied detrusor to in vitro ischemia followed by re-oxygenation. Neurourol Urodyn, 19: 701, 2000

2. Wendt, I. R.: Effects of substrate and hypoxia on smooth musclemetabolism and contraction. Am J Physiol, 256: C719, 1989

3. Waring, J. V. and Wendt, I. R.: Effects of anoxia on force, intra-cellular calcium and lactate production of urinary bladdersmooth muscle from control and diabetic rats. J Urol, 163:1357, 2000

4. Munro, D. D. and Wendt, I. R.: Contractile and metabolic prop-erties of longitudinal smooth muscle from rat urinary bladderand the effects of aging. J Urol, 150: 529, 1993

5. Orlandi, A., Francesconi, A., Cocchia, D., Corsini, A. andSpagnoli, L. G.: Phenotypic heterogeneity influences apoptoticsusceptibility to retinoic acid and cis-platinum of rat arterialsmooth muscle cells in vitro: implications for the evolution ofexperimental intimal thickening. Arterioscler Thromb VascBiol, 21: 1118, 2001

6. Dyntar, D., Eppenberger-Eberhardt, M., Maedler, K., Pruschy,M., Eppenberger, H. M., Spinas, G. A. et al: Glucose andpalmitic acid induce degeneration of myofibrils and modulateapoptosis in rat adult cardiomyocytes. Diabetes, 50: 2105,2001

7. Roberts, T. M., Sturek, M., Dixon, J. L. and Hardin, C. D.:Alterations in the oxidative metabolic profile in vascularsmooth muscle from hyperlipidemic and diabetic swine. MolCell Biochem, 217: 99, 2001

8. Allen, T. M. and Hardin, C. D.: Pattern of substrate utilization invascular smooth muscle using 13C isotopomer analysis of glu-tamate. Am J Physiol, 275: H2227, 1998

9. Odessey, R. and Chace, K. V.: Utilization of endogenous lipid,glycogen, and protein by rabbit aorta. Am J Physiol, 243:H128, 1982

10. Allen, T. J. and Hardin, C. D.: Influence of glycogen storage onvascular smooth muscle metabolism. Am J Physiol Heart CircPhysiol, 278: H1993, 2000

11. Snedecor, G. W. and Cochran, W. G.: Statistical Methods, 8th ed.Ames: Iowa State University Press, 1989

12. Haugaard, N., Wein, A. J. and Levin, R. M.: In vitro studies ofglucose metabolism of the rabbit urinary bladder. J Urol, 137:782, 1987

13. Hardin, C. D., Kushmerick, M. J. and Roberts, T. M.: Vascularsmooth muscle glycogen metabolism studied by 13C-NMR.J Vasc Res, 32: 293, 1995

14. Ozaki, H. and Karaki, H.: Organ culture as a useful method forstudying the biology of blood vessels and other smooth muscletissues. Jpn J Pharmacol, 89: 93, 2002

15. Unger, R. H.: Lipotoxic diseases. Annu Rev Med, 53: 319, 200216. Garland, P. B. and Randle, P. J.: Control of pyruvate dehydro-

genase in the perfused rat heart by the intracellular concen-tration of acetyl-coenzyme A. Biochem J, 91: 6C, 1964

17. Allen, T. J. and Hardin, C. D.: Oleate oxidation and mitochon-drial substrate selection in vascular smooth muscle. J VascRes, 38: 276, 2001

18. Spector, A. A. and Fletcher, J. E.: Transport of Fatty Acid in theCirculation. Bethesda: American Physiology Society, pp. 229–249, 1978

BLADDER SMOOTH MUSCLE METABOLISM2066