| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorials |
1 Albert Einstein College of Medicine, Division of Nephrology and Hypertension, Beth Israel Medical Center, New York, NY 10003, E-mail jwinches{at}bethisraelny.org
The word dialysis (the separation of soluble substances from colloids and their removal through a semipermeable membrane down a concentration gradient) was coined by Thomas Graham in 1861, in Glasgow, Scotland. It was not, however, until the first success with dialysis in kidney failure, by Kolff in Kampen, Holland, in 1945, and the invention, in Seattle, WA, in 1960, of the Scribner shunt for access to the circulation that modern hemodialysis as we know it became practical. Both Kolff and Scribner received the Albert Lasker Award for Clinical Medical Research in 2002 for their breakthrough concepts. Currently, more than 1 million patients worldwide undergo either hemodialysis (the vast majority) or peritoneal dialysis to maintain life(1). During dialysis, small molecules are removed by the process of diffusion, but removal of larger molecules is dependent on convective transport, which necessitates ultrafiltration to improve "solvent drag" of larger substances along with the ultrafiltered fluid(2).
Traditional uremic toxins, such as urea, creatinine, sodium, potassium, and water, were the early focus of nephrologists, but it became clear that other larger molecules, theoretically "middle molecules" around 1.5 kDa, were as important in promoting morbidity if the patient was not dialyzed for a protracted period(2). The middle molecule hypothesis was popular in the 1970s but has been largely superseded by the demonstration that even larger molecules, the low–molecular-mass proteins from 5 to 40 kDa, particularly those involved in inflammation, are of paramount importance in the morbidity and perhaps mortality experienced by dialysis patients(3).
Treatment with dialysis is not optimal and is related to the retention of products that cannot be efficiently removed by any type of dialysis membrane, or to induction of an inflammatory state leading to excessive cardiovascular mortality. Of all the low–molecular-mass proteins retained in stage V kidney disease (formerly called end-stage renal disease), ß2-microglobulin has been the most studied since its discovery in 1985(4). In long-term dialysis patients, retention of ß2-microglobulin produces a disease that can cause considerable morbidity, related to deposition of ß2-microglobulin around large joints (shoulders and hips), and to the development of carpal tunnel syndrome, because of the formation of ß2-microglobulin amyloid fibrils. Dialysis of any kind or with any membrane is incapable of removing sufficient quantities of ß2-microglobulin to prevent the deposition of amyloid.
Membranes more permeable than the cellulosic cuprophane, such as the synthetic polysulfone, not only induce less complement activation but also remove ß2-microglobulin in greater quantities than cuprophane(5); the latter membranes at best only retard disease progression and are associated with less carpal tunnel surgery. Increased ß2-microglobulin removal with high-volume (60 L of replacement fluid), but not low-volume hemodiafiltration, is also associated with a lower mortality than standard dialysis, suggesting a causal relationship between ß2-microglobulin and inflammation. In addition, the effect of ultrapure dialysate (not containing breakdown products of bacteria) has been associated with lower serum concentrations of ß2-microglobulin and less induction of inflammation(6). Additional investigative methods for removing ß2-microglobulin center around adsorptive techniques, in which ligands(7) or pore structure(8) determine the efficiency of its removal. Transplantation does not lead to resorption of the protein.
The poor efficiency of dialysis membranes reflects in part the molecular mass cutoff of the membrane or rejection of the substance by the membrane, determined by the reflection coefficient. In the case of ß2-microglobulin, its molecular mass (11.8 kDa) or changes in its structure make for poor removal. It has been known for some time that adducts such as glycated proteins occur in dialysis patients (whether diabetic or not). Indeed, glycated ß2-microglobulin has been demonstrated, and shown to induce inflammation in tissue lymphocytes, along with tumor necrosis factor, interleukin-1, and interleukin-6, in the formation of amyloid(9). In this issue of Clinical Chemistry, Corlin et al.(10) confirm that cleaved ß2-microglobulin occurs and demonstrate that a cleavage product of ß2-microglobulin, 
K58-ß2-microglobulin, behaves differently from normal ß2-microglobulin during hemodialysis.
K58-ß2-microglobulin is a form of ß2-microglobulin that is cleaved and has a deletion of lysine at position 58 on the molecule. It exists in 2 forms (molecular mass of 11.619 kDa and the 11.635-kDa oxidized form), as detected by immunoaffinity–liquid chromatography–mass spectrometry. The authors developed monoclonal antibodies to K58-ß2-microglobulin and showed that patients dialyzed with either low-efficiency cuprophane membranes or high-flux polysulfone membranes had little or no clearance of this product, compared with normal ß2-microglobulin. Interestingly, Miyata et al.(9) demonstrated that although the primary site for glycation of ß2-microglobulin was the 
-amino group of the amino-terminal isoleucine, minor sites for glycation were Lys-19, -41, -48, -58, -91, and -94, identifying Lys-58 as an important site in the molecule. The current study by Corlin et al.(10) also highlights the importance of the Lys-58 site. Of particular relevance to the study by Corlin et al., which describes the unfolding of ß2-microglobulin, is the observation of Platt et al.(11) of the dynamic nature of fibril formation of acid-unfolded ß2-microglobulin. Of further interest is the finding that a monoclonal antibody to C-terminal residues 92–99 can prevent fibrillogenesis and can be found in amyloid tissue homogenates(12).
This study and the characterization of the K58-ß2-microglobulin molecule have important clinical and research consequences. First, they will allow investigators to determine whether K58-ß2-microglobulin is involved in the amyloid/inflammatory process. Important questions in this regard include: How does the K58-ß2-microglobulin traverse through tissues? [It is known that the process of amyloid formation is slow, with transfer of ß2-microglobulin through endothelium to bone and cartilage(13).] What determines the continual deposition of amyloid into the sometimes massive, easily visible, and disfiguring deposits? Is the conformational change in the molecule part of this process, and is it responsible for less-efficient removal during dialysis? Characterization of the K58-ß2-microglobulin molecule presumably could lead to design specifications for devices to remove it: Does its different mobility on crossed immunoelectrophoresis imply a need to use charged-membrane hemodialysis, which improves removal of glycated ß2-microglobulin as demonstrated by Randoux et al.(14)? Specific sorbents could also be designed with the charge/conformational changes in mind. Specific labeled antibodies could determine the whole-body load of K58-ß2-microglobulin on scintigraphy and be used to follow responses to new treatments should they come along.
The recent heavily funded HEMO study demonstrated that patients dialyzed with high-flux vs low-flux membranes, or a higher dose vs a lower dose of dialysis prescription, had no differences in mortality(15). We are desperate for improvements in dialysis technology, and the current study may help pave the way.
Acknowledgments
Dr. Winchester is a consultant to MedaSorb Technologies, LLC.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
