INTERACTION FORCES BETWEEN RED-CELLS AGGLUTINATED BY ANTIBODY .4. TIME AND FORCE DEPENDENCE OF BREAK-UP

We report on an extension of a previously described method to measure the hydrodynamic force to separate doublets of fixed, sphered and swol len red cells cross-linked by antibody (S. P. Tha, J. Shuster, and H. L. Goldsmith. 1986. Biophys. J. 50:1117-1126). With a traveling microt ube apparatus, doublets are tracked and videotaped in a slowly acceler ating Poiseuille flow in 150-mum-diameter tubes, and the hydrodynamic normal force at break-up, F(n), is computed from the measured doublet velocity and radial position. Previous results showed a large range of F(n) the mean of which increased with [antiserum], and an absence of clustering at discrete values of F(n). Since it was assumed that the c ells separate the instant a critical force to break all crossbridges w as reached, lack of clustering could have been due to the use of a pol yclonal antiserum. We therefore studied the effect of monoclonal IgM o r IgA antibody on the distribution of F(n). The results showed that th e data are as scattered as ever, with F(n) varying from 2 to 200 pN, a nd exhibit no evidence of clustering. However, the scatter in F(n) cou ld be due to the stochastic nature of intercellular bonds (E. Evans, D . Berk, and A. Leung. 1991 a. Biophys. J. 59:838-848). We therefore st udied the force dependence of the time to break-up under constant shea r stress (F(n) from 30 to 200 pN), both in Poiseuille and Couette flow , the latter by using a counter-rotating cone and plate rheoscope. Whe n 280 doublets were rapidly accelerated in the traveling microtube and then allowed to coast in steady flow for up to 180 s, 91 % survived i nto the constant force region; 16% of these broke up after time interv als, t(P), of 2-30 s. Of 340 doublets immediately exposed to constant shear in the rheoscope, 37% broke after time intervals, t(C), from <1 to 10 s. Thus, doublets do indeed break up under a constant shear stre ss, if given time. The average time to break-up decreased significantl y with increasing force, while the fraction of doublets broken up incr eased. At a given F(n), the fraction of break-ups decreased with incre asing [IgM], suggesting that the average number of bonds had also incr eased. Using a stochastic model of break-up (G. 1. Bell. 1978. Science (Washington DC). 200:618-627; E. Evans, D. Berk, and A. Leung. 1991. Biophys. J. 59:838-848) and a Poisson distribution for the number of b onds, N(b), break-up in slowly accelerating Poiseuille flow and in imm ediate shear application in Couette flow was simulated. In Poiseuille flow, the observed range and scatter in F(n) could be reproduced assum ing [N(b)] > 5. In the rheoscope, the time intervals and number of rot ations to break-up, t(C), were quite well reproduced assuming [N(b)] = 4. The similarity of [F(n)] for monoclonal IgM and IgA for doublet br eak-up under constant slow acceleration is compatible with the conclus ion of Evans et al. (1991 a) for normal red cells and Xia et al. (manu script submitted for publication) for sphered and swollen red cells, t hat the applied force extracts the antigen from the cell membrane.