Beta 2 microglobulin (beta(2)m)) is classically known to have isoforms
with isoelectric points (pI) 5.7 and 5.3. New isoforms of beta(2)m wi
th lower pI, probably due to modifications with advanced glycation end
products, were found in the amyloid deposits of dialysis related amyl
oidosis (DRA), and they were proposed as the amyloidogenic forms of be
ta(2)m The other modifications in beta(2)m from amyloid deposits are p
artial proteolysis and single amino acid replacement (Asn by Asp at po
sition 17). However, there are no data on the sequence of the differen
t isoforms of beta(2)m from amyloid deposits. Amyloid deposits surgica
lly obtained from the carpal tunnel from 13 dialysis treated patients
and urine from 10 healthy volunteers and 5 living-related kidney donor
s were analyzed for beta(2)m content. Two-dimensional gel electrophore
sis (2D-PAGE) of beta(2)m from amyloid deposits showed the presence of
four or more isoforms with pIs < 5.7. All the spots migrating at 12 k
Da Mr region and between 4 and 6 pH reacted with rabbit anti-human bet
a(2)m antibody by Western blotting, confirming that they were beta(2)m
isoforms. beta(2)m isoforms from the amyloid deposits were then separ
ately purified with an IEF column (PB94, Pharmacia(TM)) for analysis.
Enough quantities of three pure beta(2)m isoforms could be obtained in
two cases. The sequence analysis showed an intact N-terminus in all t
he isoforms. There was Asn in the 17th residue in all the isoforms seq
uenced. 2D-PAGE of urine from 8 out of the 10 healthy volunteers showe
d the presence of beta(2)m In two of them beta(2)m also displayed four
different isoforms. At least four isoforms were observed in urine of
all the kidney donors. The present study shows that the elution peaks
of three different beta(2)m isoforms in gel isoelectrofocusing contain
beta(2)m with intact N-terminus. None of them have deamidated their 1
7th residue. More importantly, the beta(2)m isoforms with lower pI are
not specific for amyloidosis as they were found in urine from kidney
donors and in normal volunteers. These results bring into question the
hypothesis that dialysis related amyloidosis is due to the known modi
fications on beta(2)m. They suggest that the precipitation of beta(2)m
into amyloid fibrils should result from the interaction of beta(2)m w
ith other factors with amyloid enhancing activity.