Patient and methods

CASE HISTORY

A 57 year old Jewish man of Iraqi extraction, with an insignificant past medical history, presented in June 1996 to the urology department with macrohaematuria of several days duration. Cystoscopy, abdominal ultrasound, and intravenous pyelography revealed a 2 × 2 cm mass in the left bladder wall, associated with stenosis of the distal part of the ureter, hydronephrosis, and hydroureter on the same side. Physical examination at that time was unremarkable, and routine laboratory studies were within normal limits.

In September 1996, transurethral resection (TUR) of the bladder “tumour” and stenting were done. However, following the pathology results, the patient was operated on again and a resection of the distal ureter with bladder cuff and reimplantation were performed. The histology of the excised tissue showed heavy amorphous eosinophilic deposition. Congo red staining was typical for amyloid. Amyloid deposits were found in the ureter also. Immunohistochemical studies, using specific commercial antibodies, (anti-κ and anti-λ: Bio-Genex, San Ramon, USA; anti-β₂M: Zymed, San-Francisco, USA; anti-AA and anti-prealbumin: Dako, Copenhagen, Denmark), were negative for amyloid A (AA), AL, amyloid transthyretin (ATTR), and amyloid β₂ microglobulin (Aβ₂M). AA was also excluded using the highly sensitive and specific Shtrasburg method.^13,14 No plasma cells or lymphocytic infiltrates were seen in bladder or ureter sections.

Although standard methods failed to classify the amyloid protein in our patient, the results were interpreted as suggesting AL amyloidosis, not identified by the commercial anti-Ig light chain antibodies because of their inability to detect some of the immensely variable proteins found in AL amyloidosis. Further studies were aimed at the detection of other organ involvement. Rectal biopsy, bone marrow examination, echocardiography, abdominal ultrasound, electrophoresis of serum proteins, kidney and liver function tests, serum electrolytes, albumin, globulin, calcium, phosphorous, iron, T3, T4, thyroid stimulating hormone, prothrombin time, haemoglobin, mean corpuscular volume, haematocrit, and urine Bence Jones proteins were all negative or within normal limits.

About two years after the initial lesion was excised, a recurrence of the disease, which now affected much larger portions of the urinary bladder, occurred. TUR of the amyloid deposits was performed again. There was no evidence of systemic amyloidosis. For classification purposes, the amyloid protein was extracted from samples of the urinary bladder tissue obtained at presentation and subjected to sequence analysis as detailed below.

EXTRACTION AND PURIFICATION OF AMYLOID PROTEINS

About 1 g of formalin fixed, paraffin wax embedded bladder tissue was dewaxed, homogenised, and washed, as described previously.^13,14 The resulting sediment was treated with 8M guanidine HCl in 1M acetic acid, dialysed, lyophilised, and dissolved in 98% formic acid. After evaporation of formic acid, the samples were suspended in loading buffer and subjected to sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE).^14,15 The protein bands were detected by staining with Coomassie brilliant blue and transferred from an analogous intact gel to an immobilon P membrane (Millipore, Bedford, Massachussetts, USA), using a Bio-rad transblot cell, and CAPS buffer. The transfer of the proteins was confirmed by amido black staining.¹⁶

N-TERMINAL SEQUENCE DETERMINATION

The band of interest was excised from the immobilon P membrane and its N-terminus was sequenced, using both an ABI 476A sequencer and a Beckman LF3000G protein sequencer. Both sequencers were equipped with online PTH amino acid analysers. Standard sequencing protocols were used, according to the manufacturer's programs. Upon completion of the sequence run, the resulting N-terminal amino acid sequence was searched against GENBANK using NCBI's BLAST search tool to identify the protein.¹⁷

Results

SDS-PAGE analysis demonstrated a single protein band of 17.5 kDa (fig 1). Partial N-terminal sequence analysis of 15 residues revealed the sequence to be NFMLTQPHSISGSPG. By homology with previously published Ig sequences, the sequence of the amyloid protein appeared to best fit that of an Ig of the Vλ_I or Vλ_VI subgroups (13 of 15 identities in each). This finding also confirmed that the amyloidosis of the urinary bladder in our patient is of the AL type.

Discussion

We present a patient who, following evaluation for haematuria, was found to suffer from amyloidosis of the urinary bladder. Further studies and the patient's history excluded involvement of other organs, justifying a diagnosis of localised amyloidosis of the urinary bladder. N-terminal amino acid sequence analysis of the isolated amyloid protein revealed that it was homologous to an Ig light chain of either the Vλ_I or the Vλ_VI subgroups, implying that the patient's amyloidosis is of the AL type.

Localised AL amyloidosis of the lower urinary tract (bladder, distal ureter, and urethra) is an uncommon disease, with only around 100 patients reported over the past 30 years (Medline search). The mechanism restricting amyloid deposition to the urinary bladder in almost all patients is unknown, but it may be one of the following: retention in the urinary bladder of a locally produced Ig and/or deposition of a remotely produced Ig in the urinary bladder. Therefore the site of the Ig production is an important clue for the enigma of urinary bladder selection. It is also of therapeutic importance, because in addition to the local surgical management, adjuvant chemotherapy or bone marrow transplantation should be considered for patients in whom the Ig is produced outside the urinary bladder. For amyloidosis restricted to the urinary bladder, such a treatment protocol would be new. To date, the only systemic treatment tried in this disease is colchicine.¹⁸

Clues to the source of the deposited protein may be derived from the shape of the amyloid deposition.¹¹ Whereas a nodular form, limited to one or several loci in the urinary bladder, suggests in situ Ig production by an abnormal clone of tissue plasmacytes,¹⁰ a diffuse form, scattered throughout the involved tissue, suggests distant Ig production. The presence of Ig secreting cells close to the amyloid fibrils may lend further support for local generation of the amyloid protein.⁸ Demonstration of reactivity of the Ig (which is the variable region of an antibody with antigen binding capacity) with urinary tract antigens is in concert with a distant origin of the deposited Ig. In our patient, the diffuse pattern of amyloid deposition, developed late in the course of the disease, and the absence of lymphocytes or plasma cells in the vicinity of the amyloid deposits, favours the option of distant production of a urinary bladder reactive Ig. The large quantity of amyloid in the urinary bladder tissue of our patient, in sharp contrast to the absence of Ig producing cells around and within the amyloid sediments, is another argument against local Ig production by subtle B cells, not revealed by the standard histological methods.

The sequence analysis of the amyloid protein in our case was undertaken to confirm the Ig origin of the amyloid, and to see whether the structure of the protein may shed light on the issue of urinary bladder selection. In the small number of cases of urinary bladder amyloidosis where the amyloid protein was characterised, it was a fragment of an Ig λ light chain. Only rarely did the authors go one step further to sequence and define the Ig subgroup. Table 1 shows that three different subtypes of λ chain fragments have been associated with localised amyloidosis of the urinary tract, Vλ_II , Vλ_III, and Vλ_VI. The Ig found in our patient was of the Vλ_I or Vλ_VI subgroups. The absence of a urinary tract specific Ig suggests that the selection of the urinary bladder is not determined by primary structural elements, which assign the deposited protein to a certain Vλ group.

Recently, Asl et al described a patient with ureter amyloidoma consisting of Ig light chain of the Vλ_II type.¹² Transcripts, obtained from infiltrates of plasma cells within and around the amyloid fibrils, contained Ig light chain sequences identical to the sequence of the amyloid protein. These findings definitely indicate that in that particular patient, and probably in several other patients, lower urinary tract amyloid is produced locally. However, our case and most of the other published cases are characterised by the absence or paucity of local Ig producing cells; therefore, a remote Ig synthesis site for the deposited protein remains a valid possibility. Adding molecular analysis to the regular immunohistochemical methods, Setoguchi et al found evidence for monoclonal Ig synthesis in 13 of 16 patients with localised amyloidosis,¹⁹ one of whom had amyloidosis of the urinary bladder. All biopsy specimens had lymphocytes or plasma cells at the site of amyloid deposition. A distant origin of a localised amyloid protein was therefore possible for three patients even when B cells were present around the amyloid infiltrate.

In conclusion, we believe that the amyloid protein in our patient probably originated outside of the urinary bladder. This is suggested by the recurrence of the disease, despite local excision, the diffuse involvement of the urinary bladder at the relapse of the disease, and the absence of plasma or B cells in the involved tissue. Sequence analysis suggests that the urinary bladder affinity of this remotely produced Ig is not related to the residues that determine its specific Vλ subtype. The source of the Ig should be determined in any patient with localised amyloidosis because it may vary (be either local or distant) and thereby affect individual treatment protocols.