SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Objective

To determine the chemical nature of amyloid deposits found in knee joint menisci.

Methods

Amyloid was extracted from the menisci of 3 adults who underwent knee joint replacement surgery. The primary structural features of the purified proteins were determined by sequential Edman degradation and tandem mass spectrometry (MS/MS). Tissue specimens were also subjected to in situ hybridization analysis, as well as complementary DNA cloning by reverse transcriptase–polymerase chain reaction (RT-PCR). Additionally, specimens from these 3 patients, as well as other patients with amyloid in the knee joint menisci, were examined immunohistochemically.

Results

Amino acid sequence and MS/MS analyses of the extracts revealed the presence of 60–77-residue components identical to the N-terminal portion of apolipoprotein A-I (Apo A-I). The Apo A-I nature of the amyloid was confirmed by the demonstration that the green birefringent congophilic deposits in the 7 meniscus samples were recognized by an anti-human Apo A-I antibody. That the meniscus itself was the source of the amyloidogenic protein was evidenced through Southern blot analysis, in which an Apo A-I product was generated by RT-PCR from synovial tissue, and further, by the demonstration that the cytoplasm of chondrocytes reacted with the specific Apo A-I probe used for in situ hybridization and was immunostained by the anti–Apo A-I antiserum.

Conclusion

Amyloid in the knee joint menisci is formed from Apo A-I that is produced by chondrocytes within the meniscal cartilage. This entity represents yet another localized form of amyloidosis associated with the aging process and may be of pathophysiologic import.

Amyloid in the menisci of the knee joint is one of the most common forms of localized amyloidosis and is especially prevalent in the elderly (1–7). The amyloid is found principally within the deep central portions and on the surface of meniscal stroma. In a study of arthroscopy-derived meniscus fragments from 316 patients between 20 and 80 years of age, congophilic deposits were detected in 70% of the cases (100% in those older than age 50), with both male and female subjects equally affected (6). Notably, the presence of amyloid was not necessarily associated with degenerative or inflammatory disease, i.e., osteoarthritis, rheumatoid arthritis, or trauma (4–6), and on this basis, it was assumed that amyloid in the knee joint menisci represented a type of localized senile amyloidosis (6).

Although it was speculated that the amyloidogenic component in knee menisci was derived from extrasynovial tissue (6), neither the chemical nature of the amyloid nor its source had heretofore been established. To investigate this, we used microanalytic techniques developed in our laboratory (8, 9) to characterize the amyloid protein extracted from synovial tissue obtained at the time of knee replacement surgery in 3 patients. Amino acid sequence and mass spectrometric (MS) analyses revealed that, in all instances, the fibrils were composed of polypeptide fragments identical in sequence to the N-terminal portion of apolipoprotein A-I (Apo A-I). The Apo A-I nature of the amyloid was confirmed when the green birefringent congophilic deposits were immunostained with an antiserum specific for this molecule. The meniscal cell origin of this high-density lipoprotein was demonstrated by reverse transcriptase–polymerase chain reaction (RT-PCR), and its formation by chondrocytes was evidenced by in situ hybridization, as well as immunohistochemically. The results of our research have provided unequivocal evidence that amyloid of the knee joint menisci is derived from Apo A-I; further, we posit that this localized form of amyloidosis results from age-associated factors that render this protein amyloidogenic, and that amyloid of the knee joint menisci may have pathophysiologic consequences.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Tissue samples.

The study was approved by the University of Tennessee Institutional Review Board. Synovial tissues, obtained from 8 adult patients undergoing joint arthroplasty, were frozen at −80°C or formalin fixed and paraffin embedded.

Amyloid extraction and chemical characterization.

Amyloid fibrils from 4-μm–thick sections cut from formalin-fixed, paraffin-embedded tissue blocks were extracted using 6M guanidine hydrochloride; the reduced and carboxymethylated protein was purified by reverse-phase high-performance liquid chromatography (HPLC) and digested with trypsin, under conditions previously described in detail (8, 9). For identification of peptides by tandem MS (MS/MS), the samples were separated by reverse-phase HPLC, using a 150 × 0.3–mm C-18 column (LC-Packings, San Francisco, CA) at a flow rate of 1–2 μl/minute with a gradient of 5–65% acetonitrile, modified with 0.1% formic acid. The effluent of the column was directed into an LCQ Deca XP ion-trap mass spectrometer (ThermoFinnigan, San Jose, CA). Instrument control and peptide identification were performed using the manufacturer's software programs (Xcalibur and Sequest).

Immunohistochemistry.

Immunohistochemical analyses were performed with the ImmPRESS detection system (Vector, Burlingame, CA). Deparaffinized serial 4-μm–thick sections, mounted on poly-L-lysine–coated slides, were subjected to antigen retrieval by heating at 90°C for 30 minutes in DIVA Decloaker solution (BioCare Medical, Concord, CA), and then endogenous peroxidase activity was blocked with 0.3% H2O2. After washing with phosphate buffered saline, the tissue samples were treated with a rabbit anti-human Apo A-I antibody (Calbiochem-Novabiochem, San Diego, CA) and the slides incubated for 48 hours at room temperature. Sections were washed, exposed first to the ImmPRESS reagent and then to the peroxidase substrate solution, and then counterstained with hematoxylin.

RNA extraction.

Approximately 0.15 gm of meniscus tissue was frozen in liquid N2, pulverized, and the RNA extracted by suspending the material in 1.5 ml of guanidine isothiocyanate (Tri-Reagent; MRC, Cincinnati, OH) containing 0.3 ml of chloroform. After centrifugation at 14,000g for 10 minutes, the resultant aqueous phase was reextracted twice with a 1:1 mixture of buffer-saturated phenol (pH 5) and chloroform, followed by addition of 10 μg of acrylamide (Ambion, Austin, TX) plus an equal volume of cold 2-propanol. The sample was incubated for 2 hours at 20°C, followed by centrifugation at 14,000g for 10 minutes at 5°C. The pellet was washed 3 times in 80% ethanol, air-dried at 42°C, and resuspended in 10 μl of buffer containing 10 mM Tris (pH 7.0), 0.1 mM EDTA, and 1 unit/μl SuperRNasin anti-RNase antibody (Ambion).

Apo A-I RT-PCR.

Complementary DNA (cDNA) was synthesized using RT (Impron; Promega, Madison, WI) with oligo(dT)15 and random hexamer primers; the subsequent Apo A-I– and β-actin–specific PCRs were performed with Taq thermal polymerase, as previously described (10). For Apo A-I, the exon 3 sense and exon 4 antisense primers were 5′-GACCTGGCCACTGTGTACGTG and 5′-TTCTGGAAGTCGRCCAGGTAG, respectively; the forward and reverse primers for β-actin were 5′-GTGGGGCGCCCCAGGCACCA and 5′-CTCCTTAATGTCACGCACGATTTC, respectively.

Southern blot analysis.

The RT-PCR products were analyzed by agarose gel electrophoresis, stained with ethidium bromide, and photographed under ultraviolet (UV) light. The gel was then denatured and blotted onto a Nytranes membrane (Schleicher and Schuell, Keene, NH), using established Southern blot methodology (11). The blotted products were UV crosslinked (Stratalinker; Stratagene, Carlsbad, CA) and hybridized overnight at 42°C in Ultra-hyb solution (Ambion) with the Apo A-I cDNA probe, which was labeled using a random primer–based Klenow reaction incorporating both biotinylated primer and dATP (Phototope labeling kit), according to the instructions of the manufacturer (New England Biolabs, Ipswich, MA). After hybridization, the blot was washed and developed onto X-Omat AR film (Kodak, Rochester, NY) using Photostar chemiluminescence reagents (New England Biolabs).

In situ hybridization.

Formalin-fixed, paraffin-embedded sections were deparaffinized, rehydrated, fixed, and finally permeabilized with Triton X-100 detergent (Gene Detect, Auckland, New Zealand). After boiling in citrate buffer (pH 6.0), the tissue was prehybridized at 42°C for 2 hours under humidified conditions with 150 μl of a commercial hybridization solution containing 40% formamide and 100 μg/ml sonicated herring sperm DNA (Fisher Scientific, Rockville, MD) plus 100 μg/ml yeast transfer RNA (Ambion), after which a fresh hybridization solution containing 200 ng/ml of either a 5′-biotinylated Apo A-I antisense probe (5′-biotin–GCTCTCCAGCACGGGCAGCAGGCCTTGGCGGAGGTCCTCGAGCGCG) or an irrelevant 5′-biotinylated oligonucleotide probe (Sigma-Genosys, The Woodlands, TX) was added. After 16 hours of incubation at 42°C, the slides were washed 3 times with 2× saline–sodium citrate (SSC; pH 7.0) for 10 minutes at 45°C and then 3 times with 0.2× SSC for 5 minutes at 45°C. The reaction was visualized using a catalyzed signal amplification system (Dako Cytomation, Carpinteria, CA).

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Histologic features of meniscus-associated amyloid.

Examination of hematoxylin and eosin–stained tissue sections revealed the presence of prominent cartilage fibrillation and other degenerative changes in 8 specimens. Additionally, in 7 of the cases, eosinophilic deposits were seen in meniscal tissue and surrounding chondrocytes. When the sections were treated with Congo red and examined under polarizing microscopy, these areas exhibited the green birefringence characteristic of amyloid (Figures 1A and B). Such material was not evident within arterial walls.

thumbnail image

Figure 1. Knee joint meniscus–associated amyloid. A and B, Green birefringent congophilic interstitial (A) and extracellular (B) deposits visualized by polarizing microscopy.C and D, Interstitial (C) and extracellular (D) amyloid and chondrocytes immunostained with an anti-human apolipoprotein A-I antiserum. (Original magnification × 200 in A and C; × 640 in B; × 1,600 in D.)

Download figure to PowerPoint

Chemical characterization of meniscus-associated amyloid.

The amyloid contained in 20 4-μm–thick deparaffinized formalin-fixed sections from 3 patients (patients 1–3) was extracted with 6M guanidine hydrochloride and isolated by HPLC. As depicted in Figure 2, the elution profile of the reduced carboxymethylated protein from patient 1 revealed 2 prominent 280-nm UV-absorbing peaks, the first eluting at an acetonitrile concentration of 49% and the second at 52%. Direct automated analyses by Edman degradation of material in peak 1 yielded 14 residues identical in sequence to that of the N-terminal portion of Apo A-I (Figure 3). After digestion with trypsin, the resultant peptides were analyzed by MS/MS, and one was found to contain residues 11–40 of Apo A-I. This component also was present in a tryptic digest of peak 2 that had, in addition, 3 other peptides that spanned positions 41–45, 46–59, and 62–77 of this molecule (12). Additionally, proline/arginine-rich and leucine-rich repeat protein (PRELP)– and clusterin-related peptides were present. Comparable results were obtained with samples from patients 2 and 3, in which peptides corresponding to positions 24–40, 41–45, 46–59, and 62–77 of Apo A-I were detected by MS/MS (Figure 3).

thumbnail image

Figure 2. Reverse-phase high-performance liquid chromatography profile of amyloid protein extracted from the knee joint meniscus of patient 1. The 2 major peaks subjected to chemical analysis are indicated.

Download figure to PowerPoint

thumbnail image

Figure 3. Comparison of the amino acid sequence of the first 100 residues of human apolipoprotein A-I (Apo A-I) with that of Apo A-I–containing tryptic peptides generated from amyloid extracted from knee joint menisci of patients 1, 2, and 3. Amino acids are designated by the 1-letter code. The sequence indicated by a solid arrow was determined by Edman degradation; those indicated by dashed lines were determined by tandem mass spectrometry.

Download figure to PowerPoint

The masses and Xcorr values of the Apo-AI peptides identified in 2 of the 3 specimens are shown in Table 1. In the sample from the patient without evident amyloid deposits, no Apo A-I–related material was present in the extract.

Table 1. Results of mass spectrometric analyses of apolipoprotein A-I peptides obtained by trypsin digestion of meniscus amyloid extracts
Sample, peptide sequencePositionMassXcorr*DeltaCnIons
  • *

    The fit of the observed product in the spectrum versus the theoretical spectra created from available database sequences.

  • Delta correlation score between the top 2 candidate peptide matches (significant if the value is >0.2).

  • Number of peptide fragment ions matched/total number of expected fragment ions.

Patient 1     
 VKDLATVYVDVLKLLDNWDSVTSTF11–403,260.72.650.52518/58
 QLNLK41–45614.71.050.0706/8
 LLDNWDSVTSTFSK46–591,612.74.330.55221/26
 EQLGPVTQEFWDNLEK62–771,933.12.950.42116/30
Patient 2     
 DSGRDYVSQFEGSALGK24–401,817.94.700.47322/32
 QLNLK41–45614.71.540.1396/8
 LLDNWDSVTSTFSK46–591,612.83.980.50119/26

Immunohistochemical analyses of meniscus-associated amyloid.

The extracellular green birefringent deposits, as well as chondrocytes, in sections from patient 1 were immunostained specifically by the polyclonal anti-human Apo A-I antiserum (Figures 1C and D). Similar results were obtained with specimens from patients 2 and 3, as well as from 4 other individuals with amyloid in the knee joint menisci. In all instances, the amyloid did not react with an anti-PRELP reagent.

Cloning of the gene encoding the amyloidogenic precursor protein from meniscus-derived chondrocytes.

RNA was extracted from the menisci of patients 1 and 2 and amplified by RT-PCR using gene-specific primers from nucleic acids located between exons 3 and 4 of the Apo A-I gene. The resulting cDNA products were subjected to agarose gel electrophoresis and to Southern blot analysis, which revealed, after hybridization with a cloned 190-bp Apo A-I fragment from exon 4, an Apo A-I product of the expected size (∼400 bp) (Figure 4A). The presence of this protein within chondrocytes also was evidenced by in situ hybridization using a specific Apo A-I probe; furthermore, the cytoplasm of these cells was immunostained with the anti–Apo A-I antibody (Figure 4B).

thumbnail image

Figure 4. Knee joint meniscus origin of apolipoprotein A-I (Apo A-I). A, Synthesis and identification of Apo A-I cDNA products (arrows) from meniscal tissue. Agarose gel electrophoresis (top) and Southern blotting (bottom) were performed on cDNA from patients 1 and 2, probed using biotin-labeled Apo A-I cDNA and visualized by chemiluminescence. B, Chondrocyte origin of Apo A-I. Apo A-I mRNA in chondrocyte cytoplasm was visualized by in situ hybridization (top) and immunohistochemistry (bottom). The primary reagents used for immunoperoxidase reaction were biotinylated Apo A-I antisense probe (for in situ hybridization) and anti-human Apo A-I antiserum (for immunohistochemistry) (original magnification × 1,600).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The results of our studies on amyloid extracted from the menisci of 3 patients who underwent knee joint replacement surgery have provided definitive evidence of the Apo A-I derivation of this material. Direct sequence and MS/MS analyses revealed, in all cases, that the fibrils were composed of the first ∼77 amino acids of the 244-residue parent molecule. The fragmentary nature of the Apo A-I amyloid was not unexpected since deposits formed from other amyloidogenic precursor proteins (e.g., serum amyloid A or immunoglobulin light chains) typically are composed of N-terminal proteolytic products of the native molecules (13). Further proof that the meniscus-associated amyloid was Apo A-I–related was obtained from immunohistochemical studies, in which a polyclonal anti-human Apo A-I antiserum specifically immunostained the green birefringent congophilic extracellular deposits in samples from these 3 patients, as well as those from an additional 4 individuals.

In contrast to heritable cases of Apo A-I amyloidosis (14), no mutations were detected in the amyloid protein extracted from menisci and further, given the common occurrence of such deposits, the amyloid undoubtedly was the product of nonmutated human Apo A-I genes (15). Notably, amyloid consisting of wild-type Apo A-I has been found in the aortic intima of elderly individuals (16), as well as the pulmonary vasculature of elderly dogs (17). Other types of amyloid associated with the aging process also are composed of nonmutated proteins, e.g., transthyretin (18), prolactin (19), lactadherin (20), and Apo A-II (21).

Although Apo A-I is synthesized in mammals predominantly by the liver and intestines (and, to a much lesser extent, by other tissues) (22, 23), our studies indicate that this protein additionally is produced within meniscal tissue, specifically by chondrocytes, as evidenced by the results of Southern blot and in situ hybridization/immunohistochemical analyses. The synthesis of Apo A-I by these cells has been demonstrated by other investigators, who have shown that expression of this molecule occurs during chondrocyte differentiation and is markedly enhanced by small molecule ligands that activate the liver and retinal X receptors (24).

The fact that Apo A-I is produced by chondrocytes may have functional significance, since this molecule can inhibit the synthesis of proinflammatory cytokines by monocytes and macrophages (25, 26). Because Apo A-I is the major protein constituent of high-density lipoprotein and its stability is affected by interaction with this lipid complex, quantitative or qualitative alterations in knee joint cholesterol or phospholipids, as well as extracellular factors (pH, glycosaminoglycans), could lead to destabilization and self-aggregation, with eventual fibril formation (27–31). Thus, we posit that the changes in tissue structure or chemical composition of articular cartilage caused by aging are primarily responsible for the development of amyloid in the knee joint menisci.

Clinically, the occurrence of amyloid in the menisci represents a localized rather than a systemic process, and results from the in situ production of the amyloidogenic precursor Apo A-I protein. Interstitial deposition of amyloid can have pathophysiologic consequences, manifested by eventual organ failure due to replacement of normal tissue elements by insoluble fibrillar proteins or peptides, and/or cytotoxicity induced by these components (13). Conceivably, amyloid deposition in the meniscus may be a contributing, or possibly an initiating, factor in a process that results in progressive impairment of knee function. In this regard, the ability to inhibit fibrillogenesis or facilitate amyloidolysis therapeutically (32, 33) may provide a means to slow or reverse the pathologic process and, thus, the need for surgical joint replacement.

Acknowledgements

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

We thank Dr. Michael Holt for providing surgically resected knee menisci, Teresa K. Williams, Sallie D. Macy, Craig Wooliver, and Shuching Wang for technical assistance, and Alisa Lehberger for secretarial assistance.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES