Degradation of Mucopolysaccharide in Intact Isolated Lysosomes*

The function of isolated lysosomes was studied by measuring mucopolysaccharide degradation. Cultured human diploid skin fibroblasts were grown in medium containing H2%04 to label endogenous mucopolysaccharide. Lysosome containing preparations at various stages of purity were isolated from disrupted cells. These preparations degraded mucopolysaccharide as indicated by the release of radioactive sulfate. Degradation was temperature-dependent, required intact ly- sosomes, and was optimal when incubation was carried out at neutral pH in a buffer of low ionic strength. Lysosomes from Hurler fibroblasts were unable to carry out the degradative process. ATP at 0.5 m~ was found to stimulate both the rate and the extent of mucopolysaccharide degradation; GTP, U , and CTP had similar effects, whereas the noncleavable ATP an- alog adenosine 5'-&-imido)triphosphate gave no stimulation. The ATP stimulation was inhibited by nigeri- cin. ATP also stimulated chloroquine accumulation in lysosomes, the magnitude of which was used to meas- ure the change in intralysosomal pH. The presence of ATP was associated with acidification of lysosome pH by 0.23 units.

Lysosomes are highly specialized organelles that have a low internal pH and contain hydrolytic enzymes which have acid pH optima. Much information is available on characterization of individual lysosomal enzymes. However, much less is known about how these enzymes function together in intact organelles and how the organelles themselves are influenced by the intracellular environment in which they exist. We decided to address this problem by examining the degradation of mucopolysaccharide in intact isolated lysosomes.
When cultured human fibroblasts are grown in medium containing radioactive sulfate (35S02-), the label is incorporated exclusively into cellular mucopolysaccharide (1). After about 24 h a steady state is reached when intake of 35S02is balanced by secretion and degradation of labeled mucopolysaccharide (2). Degradation of the mucopolysaccharide (primarily dermatan sulfate and heparan sulfate) takes place in lysosomes by the consecutive action of at least nine different exoglycosidases (3). If any one of the enzymes needed for the degradation of the carbohydrate chains is absent, a block in the normal stepwise degradation will result and the mucopolysaccharide will accumulate in the lysosomes (4). The * This research was supported by United States Public Health * To whom correspondence should be sent.
Service Grant HD-06576. present paper demonstrates that lysosomes isolated from normal cells are able to degrade endogenous mucopolysaccharide and that this process can be stimulated by nucleotide triphosphates and acetyl coenzyme A.

MATERIALS AND METHODS
Reagents-4-Methylumbell~eryl-2-acetamido-2-deoxy-~-~-glucopyranoside was purchased from Research Products International. H;%O4 (carrier-free) and [ring-3-'4C]chloroquine were from New England Nuclear. Dulbecco's modified Eagle's medium (powdered) and tissue culture grade trypsin were purchased from Grand Island Biological Co. Nigericin was a gift of Lilly. All other reagents were purchased from standard commercial suppliers and were the best grade available.
Cell Culture-Diploid human fibroblasts were obtained from the Human Genetic Mutant Cell Repository (Institute for Medical Research, Camden, NJ) or from samples submitted for diagnosis. Cells were maintained in Dulbecco's modified Eagle's medium containing 9% fetal calf serum, the concentration of NaHC03 was reduced to 2.2 g/l, and the level of NaCl was increased to 7.44 g/l. Each liter of medium was also supplemented with lo5 units of penicillin, 0.1 g of streptomycin sulfate, and 0.25 mg of polymyxin B sulfate. Cultures were maintained at 37 "C in an atmosphere of 5% COZ and 95% air.
For studies of mucopolysaccharide degradation, a low sulfate medium was prepared that was identical to the normal medium except that MgC12 was substituted for MgS04, and streptomycin sulfate and polymyxin B sulfate were omitted. Cells were labeled with Hi%Os (100 pCi in 35 ml of low sulfate medium) for 2 days before harvest.
Preparation of Lysosomes-The cells were harvested and disrupted as described (5) except that cells were not incubated at 37 "C after trypsinization. In most experiments either the postnuclear supernatant or a lysosomal-mitochondrial (M + L) fraction was used as the source of lysosomes. The M + L fraction was obtained by centrifugation of the postnuclear supernatant at 12,000 rpm (17,750 X g) for 20 min (Sorvall SM-24 rotor) and resuspended in 0.25 M sucrose before use. Highly purified lysosomes were obtained by fractionation of the postnuclear supernatant on colloidal silica gradients (5).
Mucopolysaccharide Degradation Assay-Degradation of endogenous 35S-labeled mucopolysaccharide by lysosomes was followed by the release of free 35S042-. Aliquots of incubation mixtures (100 pl) were combined with 400 p1 of unlabeled mucopolysaccharide (sodium chondroitin sulfate, 5 mg/ml in 2.0 M sodium chloride) in polycarbonate centrifuge tubes (Nalgene, 12 ml; 16 X 100 mm). The undegraded mucopolysaccharide was precipitated by addition of 2.0 ml of 95% ethanol. Samples were mixed (Vortex Genie-mixer), heated for about 2 min in a 95 "C water bath, and placed on ice for 5 min. The precipitates were collected by centrifugation (12,000 X g, 10 rnin), and the supernatants were immediately decanted into scintillation vials. The pellets were dissolved in 0.5 ml of 2.0 M sodium chloride by gentle heating in the 95 "C water bath and reprecipitated with 2.0 ml of ethanol. The tubes were again heated, cooled, and centrifuged, and the supernatants were decanted as described above. This second extraction was necessary to release "SO4'-that had been trapped in the first precipitate. The final pellets were dissolved in 0.5 ml of 1.0 M NaCl and transferred to scintillation vials. The tubes were washed once with 0.5 ml of 1.0 M NaCl and once with 0.5 ml of water to optimize the transfer. Liquid scintillation mixture (10 ml of Aquamix, West Chem Products) was added and the samples were counted in a Packard Model 3320 liquid scintillation spectrometer. The per cent of labeled mucopolysaccharide degraded in each sample was represented Degradation of Mucopolysaccharide in Isolated Lysosomes by the ethanol soluble radioactivity, expressed as a percentage of the total radioactivity (combined ethanol extracts plus pellet).
Lysosome Latency Assay-The extent of lysosome latency was determined by measuring the activity of P-hexosaminidase (EC 3.2.1.30) in each incubation mixture in the presence and absence of detergent (6,7). The assay mixtures contained 25 pl of a diluted lysosome preparation plus 275 pl of 1.2 m M 4-methylumbelliferyl-2acetamido-2-deoxy-~-~-glucopyranoside in 13 mM citric acid, 20 mM sodium phosphate, pH 4.4, containing 0.25 M sucrose. Assays with detergent were identical with that described above, but reaction mixtures contained 0.1% Triton X-100. After 2 min at 37 "C, the assays were quenched with 1 ml of glycine/carbonate buffer, pH 10 (5). The liberated 4-methylumbelliferone was measured in a Farrand MK I spectrofluorometer (365 nm excitation and 450 nm emission). One unit represents 1 nmol of substrate hydrolyzed/min at 37 "C. The percentage of lysosomes that were broken was represented by the free P-hexosaminidase activity (measured in the absence of detergent), expressed as a percentage of the total enzyme activity (measured in the presence of detergent).

RESULTS
Mucopolysaccharide Assay-Crude fibroblast mucopolysaccharide, labeled by addition of H25S04 to the cell growth medium, could be rapidly separated from free 35S042by precipitation with 80% ethanol (1). When 35S042was mixed with unlabeled mucopolysaccharide and an ethanol precipitation was carried out, 98.6% of the added radioactivity remained soluble ( Table I) Lysosome Stability and Function-Fibroblasts were harvested and disrupted, and the suspension was freed of nuclei and unbroken cells (5). The postnuclear supernatant which contains lysosomes was incubated in 0.25 M sucrose and the time course of lysosomal stability (determined by measuring latency) and function (uiz. mucopolysaccharide degradation) was examined at different temperatures ( Fig. 1). The extent of lysosome latency varied among different lysosome preparations and among different cell types but generally indicated that 85-95% of the lysosomes were intact following cell disruption and removal of nuclei. Addition of detergent (0.1% Triton X-100) prior to assay completely abolished the ob-  served latency. Lysosomes could also be broken, although not as consistently, by sonication or repeated freezing and thawing. Organelles kept at 0 "C remained intact over a 5-h period (Fig. lA). Increased incubation temperatures resulted in an increased rate of lysosome disruption. Lysosomes incubated a t 37 "C broke at a rate of approximately 8%/h.
In addition to maintaining structural integrity, the lysosomes present in the postnuclear supernatant retained the ability to degrade endogenous labeled mucopolysaccharide (Fig. 1B). Degradation of [35S]mucopolysaccharide was dependent on temperature; the suspension kept at 0 "C was unable to degrade its [35S]mucopolysaccharide, whereas those maintained a t 25 "C and 37 "C for 5 h degraded 5 and 12%, respectively. Degradation required intact lysosomes since disruption of the organelles by sonication, freeze-thawing, or detergent completely prevented sulfate release.
The stability of lysosomes in the postnuclear supernatant was determined in a number of buffers a t different pH values. Lysosomes broke rapidly in acidic (pH 4-5) medium, as determined by loss of latency (Fig. 2 A ) . Lysosomes in the postnuclear supernatant buffered between pH 6 and pH 7.5 showed the greatest stability. Lysosomal breakage was much increased above pH 8. Degradation of [35S]mucopolysaccharide was also measured for each of the incubations described in Fig. 2 after a 24-h incubation. Some ethanol-soluble radioactivity was released at pH 4 (9.5%) and even more at pH 5 (16.4%). Lysosomal stability was the same at pH 7 in phosphate ( Fig. 2 A ) or Tris-HC1 buffer (Fig. 2B). However, lysosomes incubated in phosphate buffer ( Fig. 2 In the postnuclear supernatant of fibroblasts isolated from Hurler patients (and therefore lacking the enzyme a-L-iduronidase), the rate of lysosome breakage at 37 "C was somewhat higher than that seen with normal cells (Fig. 2B). This increased fragdity was reproducible and was observed with four different Hurler cell cultures (GM0415, GM1391, and two cultures obtained from samples submitted for diagnosis). The Hurler lysosomes were completely unable to degrade their endogenous [35S]mucopolysaccharide.
ATP Stimulation of Lysosomal Function-Of the numerous chemicals evaluated for a stimulatory effect, ATP plus Mg2+ consistently activated [35S]mucopolysaccharide degradation in the cell-free system. Lysosomes incubated in the presence of 100 p~ ATP and 5 mM MgC12 degraded 50% more [35S]mucopolysaccharide than controls without ATP (Fig. 3). Addition of ATP did not affect lysosome stability.
The effect of ATP concentrations was examined with a partially purified preparation of lysosomes (M + L fraction) as well as the postnuclear supernatant.
[35S]Mucopolysaccharide degradation in both preparations was stimulated by ATP (Fig. 4). The stimulation almost reached a plateau at approximately 0.5 mM ATP. In the experiment shown in Fig. 4  was as high as 120% ( Fig. 6; Table 11). ATP was not the only nucleotide that could stimulate [35S] mucopolysaccharide degradation; GTP, CTP, and UTP were equally effective. However, there was little or no stimulation by ADP, AMP-PNP' (a nonhydrolyzable ATP analog), and CAMP (not shown) (Table 11). Lysosome stability was unaltered by these reagents. We have previously demonstrated that highly purified preparations of lysosomes could be made by centrifugation of organelles on self-generating colloidal silica gradients (5). This procedure separates fibroblast lysosomes into two distinct  Fig. 1 were fractionated on a silica gradient (5). Fractions (2 ml) were collected and ahquots were incubated for 3 h in 20 mM Tris-HCl (pH 7.2), 5 m&f MgC12,l mM ATP, 0.25 M sucrose at 37 "C (-) and at 0 "C (---).

Degradation of Mucopolysaccharide in Isolated Lysosomes
classes, a homogeneous dense population which is free of all nonlysosomal marker activities and an impure lighter density fraction. When fibroblasta grown in 35S042-were separated by this technique, both the dense and light lysosomal fractions contained labeled mucopolysaccharide (5). When the isolated fractions were incubated in 0.02 M Tris-HCl, pH 7.2, and ATP for 3 h at 37 "C, both lysosomal populations degraded internal [35S]mucopolysaccharide (Fig. 5). The extent of stimulation by ATP was similar in both fractions (not shown).

Acetyl-CoA Stimulation
of Lysosomal Function-Acetyl-CoA was tested as a possible cofactor for the [?Y]mucopolysaccharide degradation process in intact lysosomes, as it has been shown to take part in the normal degradation of heparan sulfate in intact cells (9). Mucopolysaccharide degradation in partially purified lysosomes (M + L fraction) was stimulated approximately 30% by 5 X lo-" M acetyl-CoA (Fig. 6). In the same experiment, [35S]mucopolysaccharide degradation was stimulated 100% by the addition of 0.5 mM ATP. Stimulation by ATP and acetyl-CoA together was additive. CoA gave only a slight enhancement over the controls (no additions or with ATP). Since the stimulatory effect of CoA could have been an indirect effect via conversion to acetyl-CoA during incubation of the M + L fraction, we examined the effect of these reagents on the purified lysosomes (silica gradient dense lysosome fraction, described above).
[35S]Mucopolysaccharide degradation in the highly purified organelles was stimulated by acetyl-CoA as above; however, CoA had no effect (Table III) Acetyl-CoA + ATP 7.9 110 CoA 3.9 3 CoA + ATP 6.2 65 polysaccharide degradation by ATP may be due, at least in part, to stimulation of a lysosomal proton pump. Two lines of evidence lead us to this conclusion. First, the proton ionophore nigericin (5 pg/ml) in the presence of 10 mM KC1 completely prevented the stimulation of [35S]mucopolysaccharide degradation by ATP, even at high ATP concentrations (Fig. 4). In the absence of ATP, nigericin was only slightly inhibitory. The second line of evidence indicating that ATP might act by stimulating a proton pump came from studies of the distribution of ['4C]chloroquine in lysosomes. An increased proton concentration in lysosomes should be accompanied by an increased uptake of ['4C]chloroquine into these organelles due to trapping of the protonated form of the weak base (10)(11)(12). This was found to be the case. Lysosomes (M + L fraction) incubated for 60 min with ['4C]chloroquine in the absence of ATP accumulated 12.8% of the added radioactivity (Table  IV)  Lysosomes (M + L fraction, derived from 3 confluent 150-cm2 flasks) were incubated at 37 "C in 0.02 mM Tris-HC1 (pH 7.2), 5 m~ MgC12,0.25 M sucrose, 1 PM ['4C]chloroquine (3 X lo-@ Ci/pmol), and additions listed. The concentrations of ATP and AMP-PNP were 0.5 m~, acetyl-coA was 5 p~, and nigericin was used at a level of 25 pg/ ml with 10 m~ KCl. After a 1-h incubation, samples were assayed for lysosome latency (see "Materials and Methods") and a 100-pI aliquot of each mixture was transferred to a 1.5-ml Brinkmann test tube. Lysosomes were collected by centrifugation for 10 min in an Eppendorf model 5412 microcentrifuge at 15,000 X g. Supernatants were transferred to scintillation vials for measurement of radioactivity. The pellets were dissolved in 0.25 M sucrose, 0.1% Triton X-100 and assaved for B-hexosaminidase activity and radioactivity. . The calculation assumes that the lysosome volume is the same in each incubation and that each molecule of chloroquine binds two protons in the lysosome. The theoretical basis for similar pH calculations has been discussed (18,21,23).
in lysosomes incubated with nigericin in the presence of ATP. Acetyl-coA and the ATP analog AMP-PNP were unable to stimulate an increased accumulation of ['4C]chloroquine. The ["Clchloroquine was shown to be in lysosomes by separation of the organelles on silica gradients. Chloroquine radioactivity co-migrated with the lysosomal marker, P-hexosaminidase (data not shown). The concentration of chloroquine used in these studies (1 p~) had no effect on the basal rate of mucopolysaccharide degradation or on stimulation by ATP and acetyl-coA; however, much higher concentrations of chloroquine (100 p~) did inhibit [35S]mucopolysaccharide degradation and stimulation of degradation by ATP.
In order to calculate lysosome pH by weak base distribution data, the intralysosomal volume must also be measured. Due to the extremely small quantities of lysosomes in our preparations and to contamination with other organelles, we were unable to measure the lysosome volume accurately. However, volume measurements are not necessary for calculating changes in lysosomal pH assuming the volume does not change (see legend to Table IV). Based on this assumption, addition of 0.5 mM ATP to lysosomes lowered the intralysosomal pH by 0.23 units, whereas the ionophore nigericin increased lysosomal pH by 0.15 units (Table IV). If the internal pH of isolated lysosomes is above the optimum required for mucopolysaccharide degradation, the stimulation of this process by ATP could be explained by an increased lysosomal hydrogen ion concentration that results when this nucleotide is added.

DISCUSSION
Lysosome function in a cell-free system was first studied by Mego and McQueen (13) who injected '251-labeled albumin into mice and followed the release of acid-soluble radioactivity in isolated kidney and Liver phagolysosomes. Studies of labeled-albumin degradation have also been carried out in rat liver phagolysosomes (14), and the digestion of 1251-labeled ribonuclease has been examined in mouse kidney lysosomes (15-17). The rate of digestion of these labeled proteins both in vivo and in vitro is a much more rapid process than the degradation of [35S]mucopolysaccharide that we have studied.
The t1,2 for '251-labeled ribonuclease breakdown in vivo is approximately 14 min, while the rate in isolated phagolysosomes has been estimated at 11 min (15, 17). This contrasts with the initial rate of [35S]mucopolysaccharide degradation that we have observed with isolated fibroblast lysosomes (10-15%/h). However, this rate is at least as rapid as that seen in intact cultured cells (1,5).
Previous studies of [35S]mucopolysaccharide metabolism in foreskin fibroblasts revealed a biphasic rate of degradation; 70% of intracellular [35S~mucopolysaccharide was degraded in 12 h, whereas the remaining radioactivity was released from cells at a rate of approximately 5%/day (1). In the present study the degradation of [35S]mucopolysaccharide by isolated lysosomes under optimal conditions appears to plateau after 3-5 h incubation with about 30% of the total [35S]mucopolysaccharide degraded, although more than half of the lysosomes were intact when degradation slowed.
[35S]Mucopolysaccharide degradation could have stopped due to the production of an inhibitor or the depletion of a cofactor. However, addition of unlabeled intact or disrupted lysosomes (freshly prepared or preincubated for 3 h at 37 "C) to lysosome suspensions did not inhibit or activate the degradation (not shown). Acetyl-coA and ATP did not appear to be depleted during [35S]mucopolysaccharide degradation; hourly additions of these reagents to actively degrading lysosomes did not extend the period of [35S]mucopolysaccharide breakdown. Although some yet unidentified cofactor or inhibitor could be involved, it is possible that some [35S]mucopolysaccharide may be inaccessible to the hydrolytic enzymes in our system, perhaps as part of the mucopolysaccharide pool that is only slowly degraded in intact cells (1).
The marked increase in [35S]mucopolysaccharide degradation upon addition of ATP is accompanied by a lowering of intralysosomal pH. Various methods have been used to measure intralysosomal pH, including the distribution of weak bases or acids (for review, see Ref. 18) and the degradation of endocytosed proteins (19, 20). Published values for intralysosomal pH vary from below 4 to as high as 6.5, depending on the system studied and the method of pH measurement. Two models have been proposed to account for a low intralysosomal pH. One model is based on a Donnan-type equilibrium where fiied negatively charged groups within the lysosome can induce an asymmetric distribution of acid-base equivalents across the lysosomal membrane (18). The second model invokes an energy-dependent proton pump in the lysosomal membrane, presumably driven by the hydrolysis of ATP (21). The effects of ATP on lysosome function that we have observed lend support to the existence of an ATP-driven pump. ATP stimulated both the degradation of endogenous mucopolysaccharide and the uptake of ['4C]chloroquine in isolated lysosomes. The lack of similar effects by the ATP analog AMP-PNP indicates that ATP utilization may be a necessary step in the action of ATP. In addition, both degradation of mucopolysaccharide and uptake of ['*C]chloroquine were inhibited by the proton ionophore nigericin, which would dissipate a proton gradient. Nigericin, however, did not completely block either lysosome function or base accumulation, a result that may indicate that other factors, such as fixed negative charges, may contribute to maintenance of the lysosomal Degradation of Mucopolysaccharide in Isolated Lysosomes proton gradient. A recent study by Schneider (22) demonstrated an ATP-dependent acidification of isolated rat liver lysosomes, based on the accumulation of the weak base ["C] methylamine. The magnitude of the ATP-dependent acidification that we have observed is similar to that reported by Schneider. In addition, direct measurements of lysosome volume by Schneider lend support to our assumption that the lysosome volume does not change upon addition of ATP. The stimulation of [35S]mucopolysaccharide degradation by acetyl-coA appears to be by a mechanism that is different from that caused by ATP since the effects of the two compounds were additive and since acetyl-coA was unable to stimulate chloroquine accumulation. The stimulation of degradation by acetyl-coA probably occurs because this compound takes part directly in the degradation of heparan sulfate as a cofactor for acetylation of terminal glucosamine residues exposed by the action of heparan N-sulfatase (3, 9). These glucosamine residues are presumably resistant to further cleavage unless they are fiist N-acetylated via the enzyme acetyl-CoA:a-glucosaminide N-acetyltransferase (9). The source of acetyl-coA for the transferase reaction is not known. The stimulation of [35S]mucopolysaccharide degradation by acetyl-coA in isolated lysosomes suggests that in intact cells lysosomes may be supplied with this substrate from the cytosol. The development of a system to study lysosome degradation of mucopolysaccharide in isolated intact organelles should allow us to determine if lysosomes have a specific transport system for acetyl-coA.