| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Short Communication |
From the Departments of Pathology* andMedicine, Memorial Sloan-Kettering Cancer Center, New York, New York
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Cytoplasmic granules, sometimes with a crystalline appearance, are a classical histological feature of ASPS, first noted by Masson in the 1950’s.8 The typical ASPS crystals seem to form within these cytoplasmic dense granules that are periodic acid-Schiff (PAS)-positive and diastase-resistant (excluding glycogen as their content). The earliest histochemical and ultrastructural analysis of ASPS by Shipkey and colleagues8 indicated that these crystals and dense granules contained protein and polysaccharides, were often rhomboid, rectangular, or polygonal in overall shape and membrane-bound, and the constituent fibers in fully developed crystals had a periodicity of 10 nm and a diameter of 4.5 to 5.0 nm. It should be noted that the classic cytoplasmic crystals are only found in approximately half of ASPSs, whereas the remaining cases instead show mainly the characteristic cytoplasmic dense granules consisting of fine filamentous material, sometimes with early crystallization.9-11
In characterizing a polyclonal antibody to monocarboxylate transporter 1 (MCT1), intense cytoplasmic reactivity was noted in ASPS but not in other tumors. In all other normal and neoplastic tissues, MCT1 immunoreactivity was limited to the cell surface (M Drobnjak, C Cordon-Cardo, unpublished data). Prompted by this initial observation, we performed further light and ultrastructural immunohistochemistry (IHC), and Western blot analyses of MCT1 and its interacting partner CD147. The results reported here suggest that ASPS cells specifically accumulate MCT1-CD147 complexes in their cytoplasm, presumably leading to their crystallization. The significance of these findings to the biology of ASPS or to its lineage of origin is unclear.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
We studied 10 cases of ASPS, selected solely on the basis of
available paraffin-embedded tumor material, including 8 from Memorial
Sloan-Kettering Cancer Center, and one each from the University of
Nebraska Medical Center (Omaha, NE) and the Center for Human Genetics,
University of Leuven (Leuven, Belgium). Six cases were included in a
previous study (cases ASPS-1 to 5, and ASPS-7).7
The
presence of the ASPL-TFE3 fusion was documented
by reverse transcriptase-polymerase chain reaction in seven cases
(cases ASPS-1 to 5, ASPS-7, and ASPS-12), performed as described
previously.7
The remaining three cases did not have
material available for reverse transcriptase-polymerase chain reaction
testing, but two cases had other evidence of the ASPL-TFE3
rearrangement. Case ASPS-15 showed a TFE3 genomic
rearrangement by Southern blotting (not shown), performed using a
TFE3 intron 3 probe as previously described.7
Case ASPS-17 was originally reported to contain a cytogenetic
add(17)(q25),12
but its karyotype has been revised as
follows: 45,XX,del(1)(p11),der(9)t(9;15)(p11;q11),
der(17)t(X;17)(p11;q25),-22.7
Basic clinicopathological
data are summarized in Table 1
.
|
A polyclonal rabbit antibody to MCT1 was generated as follows. A peptide representing the 24-carboxyl amino acids of human MCT1 (VTKTAESPDQKDTEGGPKEEESPV) was synthesized with an amino-terminal cysteine. The peptide was conjugated to KLH and rabbits were immunized. The rabbit antiserum was affinity purified and used at a dilution of 1:20,000. For CD147, a mouse monoclonal antibody reactive to human CD147 (clone HIM6; Research Diagnostics Inc, Flanders, NJ) was used at a concentration of 5 µg/ml. Staining was performed on 5-µm formalin-fixed, deparaffinized tissue sections using the avidin-biotin-peroxidase method with antigen retrieval. Briefly, after deparaffinization slides were quenched with 0.1% H2O2 to reduce endogenous peroxidase staining, washed in phosphate-buffered saline (PBS), and heated in the presence of 0.01 mol/L citric acid, pH 6.0, in a microwave oven for 15 minutes. After cooling to room temperature, slides were washed and incubated with adequate blocking sera (10% normal goat serum and 10% normal horse serum, respectively) for 30 minutes. Primary antibodies were then applied and left for overnight incubation at 4°C. After extensive washing with PBS, anti-rabbit and anti-mouse IgGs were used as secondary antibodies at a dilution of 1:500, followed by avidin-biotin complex at 1:25 dilution. Finally, diaminobenzidine was used as a chromogen with H2O2 as a substrate and slides were counterstained with light hematoxylin.
Ultrastructural IHC Analysis
Representative fresh tumor tissue was fixed in EMS fixative (3% glutaraldehyde and 3% paraformaldehyde in 0.1 mol/L Millonig’s phosphate buffer, pH 7.4 (Electron Microscopy Sciences, Fort Washington, PA) and held in phosphate buffer until embedded in LR White Resin (Ted Pella, Inc., Redding, CA) using standard procedures. In each case, thick sections were cut (1 µm) and stained with toluidine blue to select suitable tissue for analysis. Thin sections (90 to 100 nm) were cut and placed on uncoated 300 mesh nickel grids (Ted Pella, Inc). Grids were then pretreated using 10% Na-metaperiodate for 1 hour at room temperature, and washed in distilled water for 5 minutes, followed by soaking in 0.01 mol/L of sodium citrate, pH 6.0, that had been microwaved to 90°C. The grids in solution were kept on a hot plate for 10 minutes and allowed to cool for 15 minutes. After rinsing with distilled water, the grids were then placed in a blocking solution of 2% bovine serum albumin/PBS for 10 minutes at room temperature and incubated in the primary antibody [MCT1 (1:500; 1:1000) diluted with 20 mmol/L Tris sodium buffer, pH 8.25] overnight at 4°C. The grids were then rinsed with 20 mmol/L of Tris sodium buffer and soaked for 5 minutes in the same solution. A goat anti-rabbit IgG (1:1500 with 20 mmol/L of Tris sodium buffer) was used for the rabbit polyclonal MCT1 antibody for 30 minutes at room temperature. The grids were then again rinsed with 20 mmol/L of Tris sodium buffer and soaked for 5 minutes in the same solution. A 10-nm protein A gold conjugate (BB International, Cardiff, UK) diluted 1:10 with 20 mmol/L of Tris sodium buffer was then applied to grids for 2 hours at room temperature, followed by two rinses with PBS. The grids were then stained with uranyl acetate for 10 minutes and lead citrate for 10 minutes.
Western Blot Analysis
Cells were washed twice in PBS and cell pellets were resuspended with 1x lysis buffer with phenylmethyl sulfonyl fluoride and 2 µg/ml leupeptin proteinase inhibitor, vortexed immediately, and placed on ice for 30 minutes. Cell debris was spun down and the supernatant was recovered and an equal volume of 2x sodium dodecyl sulfate gel-loading buffer was added to it. The cell lysate was boiled for 10 minutes and an aliquot loaded on an 10% sodium dodecyl sulfate-polyacrylamide gel. The electrophoresed proteins were transferred to a nitrocellulose membrane, and the membrane was covered with blocking solution and incubated at 4°C overnight. The blot was probed with MCT1 antibody diluted 1:1000 or with CD147 antibody diluted 1:500, followed, respectively, by secondary anti-rabbit or anti-mouse antibody conjugated to horseradish peroxide using the ECL detection system (Amersham Pharmacia Biotech, Piscataway, NJ) and the membrane was applied to X-ray film.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
All 10 ASPS cases showed membrane immunoreactivity for MCT1,
ranging from intense diffuse labeling to focal moderate labeling. In
addition to this pattern of MCT1 immunostaining, which is similar to
that observed in other tissues and tumors, 7 of 10 cases also showed
intensely staining cytoplasmic spherical globules or polygonal
inclusions that were widespread in four cases and less frequent in
three cases (Figure 1)
. Finally, four
cases displayed weak to moderate diffuse cytoplasmic MCT1 staining in
three cases in addition to the above intensely staining cytoplasmic
inclusions. IHC for CD147, which functions as an MCT1 chaperone
protein, revealed a strikingly coincident pattern of cytoplasmic
staining. Thus, in cases with spherical or polygonal cytoplasmic
inclusions immunoreactive for MCT1, cytoplasmic inclusions of the same
shape, frequency, and distribution were decorated by CD147
immunostaining in each case (Figure 1)
. Most ASPSs showed diffuse
cytoplasmic reactivity for CD147, but cell membrane positivity was less
prominent than for MCT1. To correlate the MCT1/CD147 immunoreactive
inclusions with the characteristic PAS-positive diastase-resistant
cytoplasmic granules of ASPS, the same cases were studied with the
latter histochemical stain. Again, there was a strong concordance in
the shape, frequency, and distribution of the latter histochemically
defined granules with the former inclusions revealed by immunostaining.
In one case, the MCT1 IHC and the PAS stain with diastase digestion
were done on serial sections, establishing the identity of the
cytoplasmic inclusions demonstrated by these two methods (Figure 2)
.
|
|
Conventional ultrastructural observations on the crystals and
granules in nine of our cases are summarized in Table 2
. Material was available for
ultrastructural IHC in cases ASPS-1, ASPS-2, and ASPS-14. In all three
cases, ultrastructural IHC with the MCT1 antibody resulted in
deposition of gold particles on the cytoplasmic dense granules (Figure 3)
, and on mitochondria (not shown). The
presence of MCT1 in mitochondria is well described.13
|
|
To confirm that the epitopes in ASPS that are immunoreactive with
the anti-MCT1 and anti-CD147 antibodies indeed belong to MCT1 and
CD147, respectively, we performed Western blot analysis on four cases
with available material. All four cases tested showed bands of
appropriate size with both antibodies, namely 43 kd for MCT1 and 42 kd
for CD14714
(Figure 4)
.
There was no obvious relationship between the levels of MCT1 and CD147
in these four cases. These results confirm that these ASPS cases
contain MCT1 and CD147, but given the widespread normal expression of
these proteins, they do not in themselves indicate the immunoreactive
material in the Western blots corresponds to the cytoplasmic granules
decorated by immunostaining.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
It has been reported that crystals with the same or similar periodicity and structure have occasionally been observed in other human tumors and tissues,15 most notably in the epithelioid smooth muscle cells of renal angiomyolipoma.16 Indeed, we have studied a single case of renal angiomyolipoma that showed ASPS-type crystals by electron microscopy and found discrete cytoplasmic granular immunoreactivity for MCT1 in a pattern similar to the ASPS cases described above (M Ladanyi, CR Antonescu, unpublished results).
It is not clear whether this intracellular accumulation of MCT1-CD147 complexes in ASPS is because of overproduction or impaired trafficking to their normal destination at the cell membrane. Aberrant overproduction could hypothetically be related to the action of APSL-TFE3 in ASPS cells, but it is not known whether the promoters of MCT1 or CD147 contain TFE3-responsive elements that might be inappropriately activated by ASPL-TFE3. The MCT1 promoter has not been isolated.17 The CD147 promoter sequence is available (GenBank no. AF042848)18 and database analysis (MatInspector; Genomatix Software, Munich, Germany) identifies two potential TFE3 binding sites, CACGTG at -223 and CATGTG at -475 (analysis not shown), which would need to be validated as functional TFE3-binding sites by transactivation assays and related analyses.
Impaired trafficking could reflect an imbalance in the amount of MCT1
and CD147, which normally complex together in a 1:1
ratio,14
or could be because of a deficiency in other, as
yet uncharacterized, proteins necessary for the correct targeting of
MCT1-CD147 complexes to the cell membrane. It should be noted that cell
membrane immunostaining was heterogeneous in the present series, with
all cases showing at least some MCT1 positivity whereas CD147 was more
variable, with some cases completely lacking membrane immunoreactivity
for this protein (Table 1)
.
The MCT1 gene (SLC16A1) at 1p13-p12 encodes a ubiquitously expressed proton-linked transporter for monocarboxylates such as lactate and pyruvate.17 Appropriate localization of MCT1 at the cell surface requires a direct interaction with CD147.14 Expression of MCT1 in skeletal muscle is especially high in slow oxidative (type I) fibers where it is up-regulated in response to exercise, consistent with a role in lactic acid influx for oxidation as a respiratory fuel.17 MCT1 is also prominent in other cells, including red blood cell membranes.17
The CD147 gene at 19p13.3 encodes a ubiquitously expressed highly glycosylated transmembrane glycoprotein belonging to the immunoglobulin superfamily. It functions as an MCT-associated chaperone protein and as a cell surface inducer of matrix metalloproteinases.19 It is identical or highly related to human M6 or EMMPRIN, mouse gp42 or basigin, rat OX-47 or CE-9, and avian HT7 or 5A11 or neurothelin.20,21 As an inducer of extracellular matrix metalloproteinases, its forced overexpression in certain cancer cell lines has been shown to increase their tumorigenic potential.22 Increased matrix degradation because of induction of matrix metalloproteinases by surface CD147 overexpression could also be relevant to the loss of central cohesion within nests of ASPS cells that produces its characteristic pseudoalveolar appearance.
CD147 is also of interest as a potential therapeutic target. ABX-CBL, a cytotoxic murine monoclonal antibody directed against CD147, has shown promise in a pilot study of patients with acute graft-versus-host disease, presumably because of its selective targeting of activated B and T cells that express more CD147 than resting lymphocytes.23 The CD147 overexpression in ASPS, which includes in some cases prominent membranous expression, may make these tumors especially sensitive to CD147-targeted immunotherapy.
The histogenesis or differentiation lineage of ASPS has been one of the longest-running controversies in tumor histopathology.3 A proposed origin from skeletal muscle precursor cells, based on reported immunoreactivity for certain muscle-associated proteins (most consistently, desmin) has generated considerable controversy.24-27 Hopes that the cloning of the genes involved in the specific translocation in ASPS, the t(X;17)(p11.2;q25), might shed light on its histogenesis were dashed by the observation that, of the two translocation partners, ASPL and TFE3, neither shows a tissue-specific expression pattern. Likewise, it is difficult to draw any further insight into this issue from the finding that the granules and crystals of ASPS contain MCT1 and CD147. Although the expression of MCT1 is especially prominent in skeletal and cardiac muscle,17 both proteins are expressed in a wide variety of tissues.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by the National Institutes of Health [grants PO1 CA47179 (to M.. L. and C. C–C.) and RO1 CA30388 (to D. W. G)] and the Alliance against Alveolar Soft Part Sarcoma (to M. L.).
Accepted for publication January 2, 2002.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. J.F. Lazar, P. Das, D. Tuvin, B. Korchin, Q. Zhu, Z. Jin, C. L. Warneke, P. S. Zhang, V. Hernandez, D. Lopez-Terrada, et al. Angiogenesis-Promoting Gene Patterns in Alveolar Soft Part Sarcoma Clin. Cancer Res., December 15, 2007; 13(24): 7314 - 7321. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Shimoyama, Y. Akihara, D. Kirat, H. Iwano, K. Hirayama, Y. Kagawa, T. Ohmachi, K. Matsuda, M. Okamoto, T. Kadosawa, et al. Expression of Monocarboxylate Transporter 1 in Oral and Ocular Canine Melanocytic Tumors Vet. Pathol., July 1, 2007; 44(4): 449 - 457. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A L Folpe and A T Deyrup Alveolar soft-part sarcoma: a review and update. J. Clin. Pathol., November 1, 2006; 59(11): 1127 - 1132. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
86 kd may represent CD147 dimers in this case with especially strong
expression. CD147 dimer formation has been previously
reported.