From Bench to Bedside

H. David Humes, M.D.

Department of Internal Medicine

The University of Michigan Medical Center, Ann Arbor, Michigan

 

The human kidney was the first organ to be approximated by a machine. The "artificial kidney" was invented during World War II, and became during the Eisenhower years standard treatment for chronic kidney failure. In our lab we are now developing a bioartificial kidney -- a two-part device that combines manufactured components with living cells. Although it looks nothing like a natural organ, we hope that one day it will be available as a universal-donor organ for transplant, carrying out all nearly all the functions of a healthy kidney. A successful device would completely relieve the shortage of suitable organs for transplant, and would supplant dialysis as treatment for chronic kidney failure.

In order to appreciate how this will work, it may be useful to have an overview of natural kidney function. The kidney does its work of cleansing the blood in two steps. First, the blood is grossly filtered by driving the blood through a structure consisting of a tangled wad of capillaries. Under stepped-up pressure, small molecules are driven out through the capillaries’ leaky walls. A great deal of water and other valuable small molecules are also filtered out in the process. All of this "ultrafiltrate" is captured and funneled through a looping tube, whose walls are comprised of specialized cells. These cells have evolved to recapture the water and other "good" molecules and return them to general circulation, while leaving the waste to drain away to the bladder from which it is voided from the body.

Filtration is then a two step process: gross filtration, followed by selective reabsorption. Current technology reproduces only the first step.

The artificial kidney now used in dialysis is comprised of thousands of fine hollow fibers with porous walls. The patient’s blood, laden with metabolic wastes, is diverted through the fibers. These are bathed in a fluid that draws off waste molecules. Typically, a blood-cleansing session lasts about three hours and must be repeated three times each week throughout the life of the individual. Though a life-saving technology, dialysis clearly places an enormous burden on the patient. Furthermore, it is expensive. Also, it is far from a cure for chronic kidney failure; patients continue to suffer a host of health problems.

A technology related to dialysis is called hemofiltration. Whereas dialysis is used to substitute for permanently destroyed kidney function, hemofiltration can keep a patient alive while temporarily damaged kidneys heal. Most commonly, hemofiltration is employed in the intensive care unit on very ill individuals. Again, hemofiltration reproduces only the filtrative function of the kidney.

The approach our lab has taken is to recreate the kidney’s two-step filtration with a two-part process of our own. We are developing a filter, to be linked in series with a reabsorbing unit. If both units can be made to be efficient, the path will be clear to produce an implantable device.

A recent discovery in our lab opened the feasibility of carrying out this project. Among the reabsorbing cells that comprise the wall of the looping tubule there are certain "stem" cells that retain a fetal-like capability of rapidly expanding and developing into specialized cells. We have created techniques for collecting these cells and expanding their numbers outside the body, making them available for incorporation into a reabsorbing device to replace the kidney tubule.

The reabsorbing device we are developing begins with a porous hollow fiber, such as that used in a standard artificial kidney. We then line the inner surface of this fiber with living tubule cells, after first pre-coating it with a matrix to support cell growth. The cells arrange themselves in natural array, forming a functioning tubule. Experiments have proven that the cells indeed function, reabsorbing ultrafiltrate at clinically useful rates.

So far we have addressed only the kidney’s role in eliminating metabolic waste. The kidney has other important functions as well. It is, for instance, the body’s chief regulator of red blood cells. Certain cells in the kidney sense the level of oxygen being delivered to the tissues. In response these cells produce and release into the bloodstream the hormone erythropoietin (EPO). EPO stimulates bone marrow to produce red cells. When the kidney cells recognize adequate oxygenation in the tissues they cease producing EPO, only to increase production if the oxygen levels fall. This is known as a classic physiological feedback loop. It is the body’s way of constantly monitoring its oxygen supply, continuously adjusting the "dose" of EPO to keep things in balance.

Individuals whose kidney’s fail lose this ability to monitor and adjust tissue oxygenation through red blood cell production. Consequently, they may become severely anemic. This anemia be ameliorated by regular injections of carefully calculated doses of EPO. It is impossible, though, to reproduce the continuous monitoring and dose adjustment of the natural feedback loop. We are, therefore, working on a method of incorporating the EPO-producing kidney cells into a bioartificial kidney.

We expect to make available to patients each component as it is perfected, before the entire, implantable bioartificial kidney is ready for widespread use. An implantable filter alone would increase the length of time between dialysis sessions, allowing patients to reserve just two or even one day per week for treatment. A stand-alone reabsorber unit will be useful in the intensive care unit, hopefully shortening the time-to-recovery of patients on hemofiltration. An implantable EPO-producing cell therapy device will provide optimal control of the anemia associated with kidney failure.

In summary, we have discussed three of the important functions of a living organ that will be incorporated into the bioartificial kidney: Filtration (equivalent to the glomerulus of the kidney; Reabsorption (equivalent to the kidney tubule); Endocrine function (equivalent to the EPO-producing cells. Other subtle but critical metabolic functions -- some not yet defined -- may be additional benefits of using living cells in the bioartifical device.

So far we have succeeded in creating small versions of each of these components. We have made a filtering unit, preparing it for use inside the body by lining the hollow fibers with cells normally found lining blood vessels. The natural clot resistance imparted by using blood vessel cells is enhanced by transferring into the cells a gene to code for production of an anti-coagulant. Similarly, we have constructed a functioning cell-lined reabsorber unit and an O₂-sensitive EPO-producing cell device. Our challenge now is to make each of these as efficient and long-lasting as possible. We then must scale them up, from the single hollow fiber systems we now work with, to the thousands of fibers needed for a clinical device.

Our work falls within a rapidly evolving discipline that has come to be known as "tissue engineering" -- the designing for manufacture of equivalents to living natural tissue. Just as the kidney was the first organ to be approximated by a machine, we expect it to be the first to be recreated through tissue engineering, making the kidney the first solid organ to be available as a fully functioning, implantable prosthetic.