Pigeon Monocyte/Macrophage Lysosomes During ... - Europe PMC

7 downloads 0 Views 3MB Size Report
Nancy L. Jones, W. Gray Jerome, and ... From the Department of Pathology, Bowman Gray School of ..... mary and secondary lysosomes, iVLDL treatment re-.
American Journal of Patholog,, Vol. 139, No. 2, August 1991 Copyright © American Association of Pathologists

Pigeon Monocyte/Macrophage Lysosomes During VLDL Uptake Induction of Acid Phosphatase Activity. A Model for Complex Arterial Lysosomes

Nancy L. Jones, W. Gray Jerome, and Jon C. Lewis From the Department of Pathology, Bowman Gray School of Medicine of Wake Forest University, Winston-Salem, North Carolina

Lysosomes have long been implicated as afactor contributing to the progression and complication ofatherosclerosis. The authors' laboratory previously has shoum that lysosomal ultrastructure in arterial macrophage foam cells is altered as primary lysosomes give rise to large pleiomorphic organelles on lipid accumulation during lesion progression. To further explore the subcellular alterations in lysosomes and associated organelles during foam cell formation, three-dimensional (3D) intermediate voltage electron microscopy was used to examine monocytederived macrophages (monocyte/macrophages) during early in vitro uptake of beta migrating very-low-

density lipoproteins (PVLDL). Lysosomes were identified using acid phosphatase cytochemistry, and in control cells these organelles constituted 3.5% of the total cytoplasmic volume. Both primary and secondary lysosomes were observed. Upon PVLDL uptake, the total volume of acid-phosphatase-positive organelles increased threefold over 30 minutes, and the reaction product was found in three additional morphologically distinct structures: tubular lysosomes, membrane stacks, and endoplasmic reticulum with widened cisternae. The proportion of the cell occupied by each of the five acid-phosphatasepositive organelles was quantitated at 10 minutes, 30 minutes, I hour, and 4 hours of pVLDL incubation, and their relative abundance was compared with controls that were processed either with no lipoprotein challenge or albumin incubation for I hour. Secondary lysosomes compartment volume peaked at 30 minutes; over the ensuing 3.5 hours, however, the reaction progressively shifted to three

new membrane-limited locations. Our observations document the complex 3D organization and spacial relationships among the acid-phosphatase-positive structures induced by lipoprotein uptake. The 3D organization patterns for acid-phosphatase-positive lysosomes in lipoprotein-stimulated pigeon monocyte/macrophages were similar in several aspects to the complex lysosomes previously observed in the macrophages ofpigeon arterial lesions. (Am JPathol 1991, 139:383-392)

The progression of atherosclerosis is associated with several intimal changes including increased cellularity, expansion of the extracellular matrix, and lipid infiltration. Foremost among the intimal changes in the early disease is the accumulation of lipid-laden cells, which are of both smooth muscle cell and monocyte/macrophage origins.1' 2 In addition to being lipid filled, the intimal foam cells have extensive structural and enzymatic alterations. These alterations include increases in hydrolytic enzyme activity and ultrastructure of lysosomes involved in cellular lipid accumulation.' Although some of the diseaserelated alterations may simply reflect the pathologic state, lysosomal change has been implicated as one of the factors that actively contributes to the progression and complications of atherosclerosis.67 Previous studies in our laboratory2 4 have correlated alterations in foam cell lysosomes with distinct stages of lesion progression. In early lesions, or at the developing edge of large mature lesions, lipid was primarily cytoplasmic lipid droplets (88%), with the remaining 12% in acid-phosphataseSupported by National Institutes of Health grants HL-41990, RR-02722 for IVEM, and HL-14164 (SCOR in atherosclerosis). Dr. Jones was partially supported by Institutional National Research Service Award grant HL07115. Accepted for publication April 17, 1991. Address reprint requests to Nancy L. Jones, PhD, Department of Pathology, Bowman Gray School of Medicine, of Wake Forest University, Medical Center Blvd., Winston-Salem, NC 27157-1092.

383

384

Jones, Jerome, and Lewis

AJP August 1991, Vol. 139, No. 2

positive secondary lysosomes. This was in contrast to the central areas of more advanced lesions, where 66% of lipid was within lysosomes. A similar shift in the distribution of lipid has also been seen during lipid loading of cells in culture. In vitro studies by Jerome et a18 showed that smooth muscle lipid accumulates both within cytoplasmic inclusions and acid-phosphatase-positive lysosomes. Furthermore these investigators have demonstrated a shift in lipid partitioning during cholesterol accumulation; and interestingly the partitioning shift is similar to that reported for arterial foam cells. In the present study, we have examined early changes in lysosomal morphogenesis and investigated the nature of complex lysosomes that are induced in vitro in monocyte/macrophages that have been incubated with beta migrating very-low-density lipoproteins (PVLDL). Henson et al9 10 previously demonstrated that pigeon monocyte/macrophages internalize tVLDL through both clathrin-coated and noncoated receptormediated endocytosis. Furthermore these investigators documented that by 30 minutes of PVLDL challenge, 87% of internalized fiVLDL is in lysosomes. The experiments reported in the present paper extend these earlier studies by examining, in a temporal fashion, the lysosomal changes after lipoprotein uptake by pigeon monocyte/macrophages. In addition, by using enzyme cytochemistry in conjunction with a variety of threedimensional (3D) observational techniques, we define 1) several additional acid-hydrolase-containing compartments, 2) the organization and interrelationships of these compartments to other cellular organelles, and 3) the temporal association of these structures with lipid accumulation.

IVLDL was isolated from plasma by ultracentrifugation at d < 1.006 g/ml. Beta-VLDL was conjugated to 40-nm colloidal gold (Janssen, Piscataway, NJ) or 25-nm gold colloids prepared from sodium citrate and gold chloride using a modification910 of the technique originally described by Frens13 and later adapted for LDL localization by Handley et al.14 When required, the conjugates were concentrated by centrifugation over a sucrose cushion.' 1 Before use, conjugates were dialyzed against phosphate-buffered saline and MEM. Subsequently chicken lipoprotein-deficient serum (2.5 mg protein/ml) was added to the lipoprotein.15 Acid Phosphate Cytochemistry The localization of acid phosphatase was accomplished by a modification of the Gomori16 lead precipitation method.5 After a brief fixation with 2.5% glutaraldehyde in 0.1 mol/l (molar) sucrose and 0.1 mol/l cacodylate, pH 7.4, the cells were washed three times with the cacodylate buffer and once with 0.05 moVI TRIS maleate, pH 5.0, before incubation with reaction mixture containing 3-glycerophosphate as the substrate and lead nitrate as the capture agent. Cells were then postfixed with 1% OS04 in cacodylate buffer and processed for thin (0.1 ,u) and thick (1.0 to 2.0 ,u) section transmission electron microscopy (TEM) or whole-mount intermediate voltage electron microscopy (IVEM) as described below. Cells incubated with the reaction mixture that did not contain the ,B-glycerophosphate substrate were routinely used as a reaction control.

Thin- and Thick-section TEM

Materials and Methods

Cell Isolation and Culture Procedures Monocytes were isolated from citrated whole blood obtained from Random Bred White Carneau (RBWC) pigeons essentially as previously described.911 Cells were plated on either glass coverslips or formvar-coated, gold finder grids mounted on glass coverslips and maintained in Eagle's Minimum Essential Medium (MEM) (Hazelton, Denver, PA), containing 10% (vol/vol) heatinactivated chicken serum (Hazelton, Denver, PA) at 37°C in a 5% C02 humidified incubator for 7 to 10 days.

Lipoprotein Isolation Beta-VLDL was obtained from hypercholesteremic RBWC pigeons using standard procedures.12 Briefly,

Samples to be sectioned were embedded with Medcast, (Ted Pella Inc., Redding, CA) according to the manufacturer's specifications. Thin sections (0.1 ,u) were visualized at 80 keV in the Philips EM-400, and thick sections (1.0 to 2.0 ,u) were visualized and photographed as stereo (3D) pairs at 300 keV in the Philips CM-30 IVEM.

Lysosomal Quantitation Monocyte/macrophages were incubated continuously with PiVLDL or PVLDL-gold conjugate (50 ,ug protein/ml) at 370C for 10 minutes, 30 minutes, 1 hour, or 4 hours before the acid phosphatase reaction. Control specimens were incubated with albumin or albumin-gold conjugate (50 ,ug protein/ml) for 1 hour. Untreated monocyte/ macrophages also were used as controls for lysosomal enzyme activity. For quantitation, acid-phosphatase-

3VLDL-induced Macrophage Lysosomes

385

AJP August 1991, Vol. 139, No. 2

Nikon 35-mm camera. Stereo images were displayed by rotating the digitized image + 5 degrees around the Y

positive organelles were classified into five structurally distinct categories; primary lysosomes, secondary lysosomes, tubular lysosomes, membrane stacks, and endoplasmic reticulum (ER) with widened cisternae. These categories are further defined in the results. To simplify, we refer to all acid-phosphatase-positive organelles as lysosomes. The reader should keep in mind, however, that this is a generalization, and other possibilities for these organelles are addressed in the discussion. Standard point count methods involving a square lattice of points spaced 1.3 cm apart were used for all stereologic estimates.217 Volume densities (volume of acidphosphatase--positive organelles as a percentage of total cytoplasmic volume) were determined at a magnification of 9000x from a total of 60 micrographs (20 micrographs from three experiments) for each experimental condition. Total cytoplasmic volume was calculated by excluding the nuclear volume.

axis.

Results Pigeon monocyte/macrophages internalize PVLDL through both clathrin-coated and noncoated vesicles. Independent of the uptake mechanism, by 10 minutes, 32% of the lipoprotein arrives in lysosomal compartments; and by 30 minutes, 87% of the internalized PVLDL is in lysosomes.'0 Therefore we investigated lysosomal structure at early times of ,VLDL uptake. Specifically the times were 10 minutes, 30 minutes, 1 hour, and 4 hours of PVLDL incubation. All incubations were done with continuous exposure to lipoprotein. This was done for two reasons. First continuous incubation has been used by others to drive foam cell formation in vitro; and secondly continuous lipoprotein exposure would more closely mimic the in vivo artery lesion environment. In addition, pigeon monocyte/macrophages undergo temperaturedependent (4°C versus 37°C) morphologic changes that are independent of lipoprotein binding11; therefore incubations were done at physiologic temperature, 37°C, to reduce the possibility of temperature-related structural artifact. Before lipoprotein challenge, acid-phosphatasepositive organelles constituted a relatively small proportion of cell volume. The average total cytoplasmic volume density of lysosomes in control cells (untreated and albumin treated) was 3.5% to 4.0% (Table 1). The two major reaction-positive compartments in these control preparations were primary and secondary lysosomes (Figure 1). Identification as either primary or secondary lysosome was based on descriptions in the literature. Lysosomes were characterized as primary if they were densely staining small spherical structures with a diameter in the range 75 to 320 nm (mean diameter of 230 nm). Secondary lysosomes were more pleiomorphic and vaned greatly in diameter over the range of 300 to 1300 nm, with a mean diameter of 585 nm 400 nm (Figure 2). Both the amount and the distribution of acid phosphatase cytochemical product changed on incubation of the cells with iVLDL (Figures 1, 2B). The most immediate and dramatic changes were in the percent of the cell occupied by primary and secondary lysosomes. The re-

Statistics Data were grouped according to the experimental conditions outlined above and loaded onto LOTUS 123 for numerical manipulations. The group mean and standard error of the mean were computed. Analysis of variance, followed by Duncan's Multiple Comparison Test, was done using the Number Cruncher Statistical System, Version 5.0.18 Two means were considered significantly different if the probability of a type error was less than 0.05. Reported for each experiment are the significant differences as compared with the control group, which had no lipoprotein treatment.

Computer Reconstructions Selected cells were sectioned serially (0.1 ,u), photographed using conventional TEM, digitized from the micrographs, and then reconstructed by computer. Computer-assisted serial reconstructions were done at the MlCROMED Laboratory for Intermediate Voltage Electron Microscopy using software developed and described by Young et al.19 The system was based on the software 'IBM PC-Based Three dimensional reconstruction system (HVEM-3D).' Images for stereo viewing and publication were photographed from the computer monitor using a

±

Table 1. Total Volume of Acid-phosphatase-containing Organelles Treatment Albumin No LP 10 minutes

Lysosomal Volume *

4.0 ± 1.0

3.5 ± 1.0

8.8* ±1.7

30 minutes

1 hour

11.5* ±4.3

±0.7

Indicates P < 0.05 as determined by Duncan's Multiple Comparison Test (Materials and Methods).

9.0*

+

Standard deviation.

4 hours 7.1*

±0.47

386 Jones, Jerome, and Lewis AJP August 1991, Vol. 139, No. 2

AcPase Localization U I trastructura I Morphometr ics 7

0

6

/ / /

5 /

4

N'.

C-) *0

0

N N !

3

/

N

-

N.

2

01

-~~~~~~;

;

/ - _,w -

/0_-. -

.e.

_

- -

--

-

-

-

fied and quantitated for percentage of total cytoplasmic volume. These organelles were primary lysosomes (1°lys), open triangle; secondary lysosomes (2°lys), open circle; tubular

lysosomes (tublys),filled triangle; ER with widened cisternae (wdest), filled circle; and memT brane stacks (mbrstk), open square. Macro- Fb phages were treated for 1 hour with no li- 5* poprotein (con) or 3VLDL (50 Ig/ml) for 10 minutes, 30 minutes, 1 hour, and 4 hours. The percentage of cytoplasm occupied by acid-phosphatase-positive orA each of the albumin-treated ganelles of macrophages *

, ll N-

I

T

Figure 1. Graph summaizing thepercentage volume of total cytoplasm occupied by each acid-phosphatase-positive organelles at various times of IVLDL incubation. Five acidphosphatase-positive organelles were classi-

significantly different from the nolipoprotein treatment control group (not shown). Asterisks = P < 0.05 is compared with the control group, no lipoprotein treatwas not

con

lOm 1h 30m Lipoprotein Inn -ation

sponse initially involved an increase of cell occupied by these organelles. The initial rise, however, was followed by a gradual decrease after 30 minutes of incubation. The proportion of cytoplasmic volume for small, spherical primary lysosomes increased from 0.75% in the control cells to a peak of 1.78% with 10 minutes of IVLDL incubation. This category returned to approximately normal levels by 1 hour of lipoprotein exposure (Figure 1). This pattern of increase followed by return to normal levels was also observed with secondary lysosomes. The cytoplasmic volume of secondary lysosomes (2.2%), under control conditions, increased at a slightly slower rate than the primary lysosomes. These organelles increased threefold by 30 minutes of IVLDL incubation (6.3%, P < 0.05), before decreasing to approximately 3% of the cytoplasmic volume after 4 hours of ,BVLDL incubation. In addition to stimulating cytochemical activity in primary and secondary lysosomes, iVLDL treatment resulted in the formation of an additional three membranous enzyme-positive compartments (Figures 1 to 3). We de-

4

,_

ment.

fine these compartments as tubular lysosomes, membrane stacks, and endoplasmic reticulum (ER) with widened cisternae. The most prevalent of these, ER with widened cisternae, consisted of an intensively opaque anastomosing network of parallel membranes, the spacing of which ranged from 25 to 85 nm (Figures 2, 3B), with an average spacing of 44 nm. Specific intracellular localization of the ER with widened cisternae was not evident; however this category of reaction-positive organelle was often looplike in arrangement and associated either with lipid droplets or debris-containing vacuolated areas suggestive of autophagy (Figures 2, 3). An acidphosphatase-positive compartment paralleling ER with widened cisternae in structure was the membrane stacks. These typically contained three to four compact lamellae, which were 10 to 30 nm in width and uniformly spaced. Often the membrane stacks were characterized by the presence of swollen terminal regions with numerous small acid-phosphatase-positive vesicles, primary lysosomes, in the surrounding cytoplasm.

,BVLDL-induced Macrophage Lysosomes

387

AJP August 1991, Vol. 139, No. 2

"

;t

"..

x. .". -,;'k'I -ti AA .,.

OF.

Figure 2. Overview micrographs illustrating acid-phosphatase-positive organelles in untreated control and IVLDL-treated monocytel macrophages. A: Untreated control monocyte/macrophages were processed for acid phosphatase cytochemistry. Acid-phosphatase-positive organelles constituted less than 4% of the cell volume. Primary lysosomes (arrowheads) and secondary lysosomes (arrows) were the most abundant structures observed B: Acid-phosphatase-positive organelles of monocyte/macrophages incubated with IVLDL for 30 minutes before fixation and acid phosphatase cytochemistry. Under these conditions a threefold increase in the proportion of cell volume constituting acidphosphatase-positive organelles was observed. Four of the five acid-phosphatasepositive organelles are illustrated in this micrograph; primary lysosomes (1J), secondary Iysosomes (2), ER with widened cisternae (WC), and membrane stacks (MS). Nucleus (N), lipid droplets (L). Bars = 0.5 pL.

The final category of acid-phosphatase-positive structures was tubular lysosomes. These differed from the previous two systems in that they were typically localized at the cell periphery. In addition, the dilated lumen, with widths ranging from 50 to 320 nm (mean spacing) resulted in a diffuse distribution of reaction product resembling the pattern associated with secondary lysosomes. As was the case with secondary lysosomes, gold colloids indicative of endocytosed PiVLDL were frequently observed in tubular lysosomes (Figure 3). In this study, gold colloids were not observed in the other acidphosphatase-positive compartments. The total volume of the cell occupied by all acid-

phosphatase-containing structures increased with exposure to PiVLDL (Table 1). The volume of cell occupied by ER with widened cisternae and membrane stacks remained high throughout incubation with IVLDL; this was in contrast to primary, secondary, and tubular lysosomes, which decreased after the initial increase (Figure 1). Ten minutes after initiation of ,VLDL incubation, acidphosphatase-positive membrane stacks had increased sevenfold compared with the nontreatment control, and at 30 minutes of ,VLDL uptake, stacks were nine times as abundant as in the nontreatment control cells. At 1 and 4 hours of ,VLDL treatment, when the volumes of most of the other acid-phosphatase-positive compartments

388

Jones, Jerome, and Lewis

AJP August 1991, Vol. 139, No. 2

Figure 3. Monocyte/macrophages treated with (VLDL for 30 minutes and reacted for acidphosphatase. A: Tubular lysosomes (bold arrow) with gold-conjugated F3VLDL (small arrow) were present near the cellperiphery. In this micrograph they are located proximal to the lower edge of the cell. B: Micrograph illustrating the highly convoluted nature of ER with widened cisternae. Notice that some regions of these organelles were stained while other segments lacked cytochemical product. Secondary lysosomes (20) and a lipid droplet (L) are present in the vicinity of the ER with widened cisternae (WC). Bars = 0.2 Ri.

were approaching control levels, the volume occupied by membrane stacks was still sixfold higher than nontreatment controls. The cytoplasmic volume of acidphosphatase-positive ER with widened cisternae also remained elevated throughout incubation with ,VLDL. The maximal expression, however, was at 1 hour of I3VLDL treatment (11-fold higher than nontreatment controls). Although tubular lysosomes were not a predominant structure, accounting for less than 0.5% of cytoplasmic volume, the volume fraction of tubular lysosomes was doubled with IVLDL exposure for both 30 minutes and 1 hour incubation. As in the case of primary and secondary lysosomes, the volume fraction of tubular lysosomes returned by 4 hours to values approximating control. As noted above, membrane stacks were frequently juxtaposed with primary lysosomes (Figure 2). To clarify the relationship of primary lysosomes to the membrane stacks, serial thin sections of acid-phosphatase-positive cells were photographed, and the images were digitized.

Through computer-assisted 3D reconstruction, the cells and various organelles were then selectively examined to determine the relationship between membrane stacks and primary lysosomes. The reconstructions were highly revealing; for, as shown in Figure 4, the membrane stacks constituted a highly branched network extending throughout the cell center with primary lysosomes located at the apex of the branched stacks. Although individual thin sections sometimes demonstrated close proximity between primary lysosomes and membrane stacks (Figure 2), 3D computer reconstructions demonstrated that each of the primary lysosomes was positioned at the apex of a stack (Figure 4A). Three-dimensional computer reconstructions also were used to selectively analyze the positioning of ER with widened cisternae with respect to lipid droplets. As illustrated in Figure 4B, widened cisternae typically enwrapped several cytoplasmic lipid droplets. This unique association was further verified with 3D IVEM observation

pVLDL-induced Macrophage Lysosomes

389

AJP August 1991, Vol. 139, No. 2

Figure 4. Color plate of computer-reconstructed stereo pair images. A: Stereo reconstruction ofprimary lysosomes and membrane stacks. The image is displayed at x320, YO and 5 and Z 0 degrees rotation. The plasma membrane (white), cytoplasm (light blue), membrane stacks (dark blue), andprimary lysosomes (red) are shown. In this reconstruction the position ofprimary lysosomes (red) at the apical bends of the membrane stacks is evident. B: Stereo reconstruction of ER with widened cisternae oriented at X 20, Y340 and 345 and Z 0 degrees rotation with the plasma membrane (light blue), ER with widened cisternae (white), lipid droplets (green). In this reconstruction the intimate relationship between the acid-phosphatase-stained ER with widened cisternae and lipid droplets is highlighted Notice that the ER with widened cisternae enurap the spherical lipid droplets (arrowheads).

390

Jones, Jerome, and Lewis

AJP August 1991, Vol. 139, No. 2

Figure 5. Three-dimensional examination of endoplasmic reticulum with widened cisternae surrounding lipid droplet. Monocytel macrophages were treated continuously for 30 minutes with 3VLDL-gold beforefixation, acid phosphatase cytochemistry, and thick sectioning for IVEM. A: This 2-.--thick section contains portions of three cells. The central cell (delineated by arrowheads) contains many lipid droplets (L). The central lipid droplet is surrounded by acid phosphatase reaction product (large arrows). B: The relationship between the lipid droplets and acid-phosphatase-positive organelles is illustrated in B stereo pairs. The majority of the acid phosphatase is an interconnected network associated with the central lipid droplet. This lipid droplet is highlighted in C stereo pairs. The highly convoluted nature of ER with widened cisternae is illustrated (arrows) in stereo. Refer to Figures 2B and 3B for examples of the appearance ofsimilar ER with widened cisternae in thin sections. C: Several thin lavers of ER u'ith widened cisternae are wrapped around the lumen of the lipid droplet (arrowheads). Notice also that the ER with widened cisternae defined a larger tubular structure joining into the large spherical lipid droplet (arrows). (Bars = 1 p).

of both thick sections from epoxy-embedded cells or intact whole-mount cells (Figure 5).

Discussion Many investigations have documented an increase in the activity of acid phosphatase (and other lysosomal en-

zymes) in arteries with experimentally induced atheromas. This increase in hydrolytic enzyme activity has been considered to be indicative of progressing atherosclerosis.1,20,21 It was clear from our observations that incubation with ,BVLDL induced a shift in the location of acid hydrolases within cultured pigeon monocyte/macrophages. This shift was both temporally and spatially related to lipid accumulation. In addition, lipoprotein in-

IVLDL-induced Macrophage Lysosomes

391

AJP August 1991, Vol. 139, No. 2

cubation resulted in a dramatic increase in acidphosphatase-positive cytoplasmic volume. This increase in acid phosphatase activity during lipoprotein uptake by pigeon macrophages was consistent with reports of increased hydrolytic enzyme activity in progressing atherosclerosis. 1,20,21

The five acid-phosphatase-positive ultrastructural compartments differed temporally in maximal expression of enzymatic activity. Primary lysosomes underwent a dramatic stimulation with short-term lipoprotein uptake (10 minutes). Membrane stacks also reached a peak of acid phosphatase activity at a similar early time (10 to 30 minutes). Interestingly primary lysosomes and membrane stacks not only had temporally correlating maximal acid phosphatase activity, but they were also spatially associated. The presence of the acid-phosphatasepositive vesicles, primary lysosomes, at the terminal of membrane stacks and the compact nature of the lamellae within the stacked were suggestive of the trans-Golgi network. The spacial relationship observed in our study was consistent with a functional relationship that would involve transport of newly synthesized acid phosphatase from the Golgi (membrane stacks) to form primary lysosomes. Acid-phosphatase-positive Golgi lamellae have been documented previously in other arterial lesion cells.5 Secondary lysosomal volume dramatically increased with lipoprotein stimulation. By 1 hour, however, the volume contribution from secondary lysosomes began to decrease toward control levels. Tubular lysosomes were probably a subset of secondary lysosomes, as evidenced by the presence of endocytosed gold colloids. The cytoplasmic volume of tubular lysosomes doubled with lipoprotein incubation; and, like secondary lysosomes, the cell volume decreased, approaching control levels by 1 to 4 hours. Structures resembling the tubular lysosomes in our study have been described previously. The most common observations have been in macrophages.22 25 The similarities of pigeon macro phage tubular lysosomes to those reported in the literature included both their structural characteristics and their induction by endocytosis.2526 Griffiths27 suggested that tubular lysosomes are late endocytic or prelysosomal compartments. The location of these structures at the cell periphery, neighboring endosomes and coated pits (Figure 3), is consistent with the possibility that tubular lysosomes are early-stage secondary lysosomes. The response of ER with widened cisternae to lipoproteins was the most interesting with respect to lipid accumulation. The ER with widened cisternae compartment underwent the largest increase on lipoprotein incubation; a maximal 12-fold increase was observed at 1 hour. Although the cytoplasmic volume of ER with widened cisternae stopped increasing by 4 hours, it remained elevated throughout the study period. This suggests a con-

tinued accumulation resulting in a steady-state in this compartment. Spatially within the cytoplasm, ER with widened cisternae also were associated with lipid pools. The close proximity of ER with widened cisternae and lipid droplets suggests there may be a metabolic relationship between the acid-phosphatase-positive network and lipid accumulation. Jerome and Lewis,4 based on observations in aortic lesions, have suggested that large quantities of extralysosomal lipid may initiate lysosomal autophagy of extralysosomal lipid. Such autophagy would explain the pleiomorphic nature of lysosomes found in progressing disease. Alternatively, the pleiomorphic lysosomes may have resulted from ongoing addition of fresh primary lysosomes to preexisting lipid droplets or secondary lysosomes. Knapp and Swanson28 have recently reported that an initial tubulorecticular network of lysosomes in macrophages could be structurally altered during phagocytosis. The alternations ultimately led to the presence of spherical organelles resembling the pleiomorphic lysosomes in atherosclerosis. The enwrapping of lipid droplets by what we have termed acid-phosphatase-positive ER with widened cisternae may represent such a conversion. Secondary lysosomes associated with acidphosphatase-positive membrane whirls have been documented in lesions from both rabbits and pigeons.34 Secondary lysosomes in cells of atherosclerotic lesions also have been reported to contain myelinlike figures. During autophagy, a similar conversion from tubuloreticular to spherical lysosomal structure has been reported for several cell types by Ogawa et al.20 In conclusion, lipoprotein uptake by monocyte/ macrophages stimulated the production of several distinct acid-phosphatase-containing structures. Some of these organelles became very complex, with prolonged periods of lipid accumulation. Of particular importance was the similarity of these acid-phosphatase-positive organelles to the altered lysosomes observed in the cells (smooth muscle cells and macrophages) within atherosclerotic lesions. There is similarity between these in vitro changes and cells that constitute the pigeon lesion4; the similarity is most striking in the large multilobal secondary lysosomes connected to tubular complexes. These results suggest that lipoprotein loading produces multiple pleomorphic lysosomes, which may be indicative of an alteration in lipid metabolism.

Acknowledgments The authors thank A. Sreedharan for technical assistance; P. Moore for serial sectioning; K. Grant, M. Lei, and S. Evans for technical assistance on electron microscopy printing; B. Lindsay for manuscript preparation; and C. Ross for technical writing advice and documentation consultation.

392

Jones, Jerome, and Lewis

AJP August 1991, Vol. 139, No. 2

References 1. Fowler SD, Brown WJ: Lysosomal acid lipases, Lipases. Edited by B Borgstrom, EH Brockman. Elsevier, Amsterdam, 1984, pp 329-364 2. Jerome WG, Lewis JC: Early atherogenesis in White Carneau pigeons: II. Ultrastructural and cytochemical observations. Am J Pathol 1985, 119:210-222 3. Shio H, Haley NJ, Fowler SD: Characterization of lipid-laden aortic cells from cholesterol-fed rabbits: Ill. Intracellular localization of cholesterol and cholesteryl ester. Lab Invest 1979, 41:160-167 4. Jerome WG, Lewis JC: Early atherogenesis in White Carneau pigeons: Ill. Lipid accumulation in nascent foam cells. Am J Pathol 1987,128:253-264 5. Lewis JC, Taylor RG, Ohta K: Lysosomal alterations during coronary atherosclerosis in the pigeon: Correlative cytochemical and three-dimensional HVEM/IVEM observations. Exp Mol Path 1988, 48:103-115 6. Peters TJ, de Duve C: Lysosomes of the arterial wall: II. Subcellular fractionation of aortic cells from rabbits with experimental atheroma. Exp Mol Path 1974, 20:228-256 7. deDuve C: Participation of lysosomes in the transformation of smooth muscle cells to foamy cells in the aorta of cholesterol-fed rabbits. Acta Cardiol Suppl 1974, 20:9 8. Jerome WG, Minor LK, Glick JM, Rothblat GH, Lewis JC: Lysosomal involvement in smooth muscle cell lipid accumulation. J Cell Biol 1989, 107:570a. 9. Henson DA, St Clair RW, Lewis JC: I3VLDL and acetylatedLDL binding to pigeon monocyte macrophages. Atherosclerosis 1989, 78:47-60 10. Henson DA, St Clair RW, Lewis JC: Morphological characterization of p-VLDL and acetylated-LDL binding by culture pigeon monocytes. Exp Molec Pathol 1989, 51:243-263 11. Jones NL, Allen NS, Lewis JC: ,BVLDL uptake by pigeon monocyte derived macrophages. Correlation of binding dynamics with 3-D Ultrastructure. Cell Motil Cytoskeleton 1991 In Press 12. Adelman S, St. Clair RW: f-VLDL Metabolism by pigeon macrophages. Evidence for two binding sites with different potentials for promoting cholesterol accumulation. Atherosclerosis 1988, 9:673. 13. Frens G: Controlled nucleation for the regulation of the particle size in monodisperse gold suspensions. Nature 1973,

241:20-22 14. Handley DA, Arbeeny CM, Witte LD, Chien S: Colloidal gold-

15. 16. 17.

18. 19. 20. 21.

22.

23. 24.

25.

26.

27. 28. 29.

low density lipoprotein conjugates as membrane receptor probes. Proc Natl Acad Sci USA 1981, 78:368-371 Randolph RK, St. Clair RW: Pigeon aortic smooth muscle cells lack a functional low density lipoprotein receptor pathway. J Lipid Res 1984, 25:8802 Gormori G: Microscopic Histochemistry: Principles and Practice. Chicago, University of Chicago Press, 1952 Weibel ER, Staubli S, Gnagi H, Hess F: Correlated morphometric and biochemical studies on hepatocyte membranes: I. Morphometric model, stereologic methods, and normal morphometric data for rat liver. J Cell Biol 1969, 42:61 Hintze JL: Number Cruncher Statistical System, Version 5.0. Kaysville, UT, Jerry L. Hintze, 1987 Young SJ, Royer SM, Groves PM, Kinnamon JC: Three dimensional reconstructions from serial micrographs using the IBM PC. J Elect Micro Tech 1987, 6:207-217 Goldfischer S, Schiller B, Wolinsky H: Lipid accumulation in smooth muscle cell lysosomes in primate atherosclerosis. Am J Pathol 1975, 78:497-504 Wolinsky H, Goldfischer S, Daly MM, Kasak LE, ColtoftSchiller B: Arterial lysosomes and connective tissue in primate atherosclerosis and hypertension. Circ Res 1975, 36:553-561 Swanson J, Bushnell A, Silverstein SC: Tubular lysosome morphology and distribution during differentiation of pancreatic acinar AR42J cells. J Histochem Cytochem 1986, 36:959 Swanson J, Burke E, Silverstein SC: Tubular lysosomes accompany stimulated pinocytosis in macrophage cells. J Cell Biol 1987,104:1217. Heuser J: Changes in lysosomal shape and distribution correlated with changes in cytoplasmic pH. J Cell Biol 1989, 108:855. Araki N, Ogawa K: Regulatory mechanism of lysosomal transformations of nematolysosomes and wrapping lysosomes by the cytoskeleton. J Histochem Cytochem 1988, 36:863 Sakai M, Araki N, Ogawa K: Lysosomal movements during heterophagy and autophagy: With special reference to nematolysosome and wrapping lysosome. J Electron Microsc Tech 1989, 12:101-131 Griffiths G: The Golgi complex, endosomes, and the biogenesis of lysosomes. J Histchem Cytochem 1988, 36:953 Knapp PE, Swanson J: Plasticity of the tubular lysosomal compartment in macrophages. J Cell Sci 1990, 95:433-439 Ogawa K, Sakai M, Mayahara H: The intracellular regulation of the lysosomal wrapping mechanism observed during autophagy. Acta Histochem Suppl Band XXX 1984, s. 38-46