Cxlvii. the Behaviour of Urinary Proteins in the Ultracentrifuge. by Arthur Sproul

THE question whether the proteins found in the urine in certain diseases have the same composition as the proteins in the blood is one of great importance. With the probable exception of two conditions, viz. orthostatic and Bence Jones proteinuria, it is believed that the presence of protein in the urine may be taken as a sign of impaired renal function. If the urinary protein is the same as the plasma protein it is reasonable to assume that the disease has damaged the kidneys in such a way as to give rise to a form of leakage of plasma into the urinary tract. If the protein is different, leakage of a gross type can be excluded and it is not at all clear how the protein finds its way into the urine. We may assume that the composition of the plasma protein is changed in the renal cells by the disease process at the same time as the protein is excreted; or we may postulate that the altered urinary proteins arise in remote parts of the circulation and are excreted by normal kidneys. Since the presence of Bence Jones protein in the blood has been demonstrated in certain cases, the group of diseases with which this protein is characteristically associated has been regarded as unique (if we except the orthostatic albuminurias) in conforming to the latter hypothesis, although a few observers [Forbes et al., 1933] identify renal changes with Bence Jones proteinuria. Kerridge and Bayliss [1932] have shown that Bence Jones protein readily passes through the kidneys of the cat while normal human serum proteins do not. Kerridge [1934] has shown also that the urinary proteins from ten cases of chronic nephritis did not pass through the kidneys of the cat. The inference from this (which incidentally Kerridge avoids making) is that the proteins in the urine in chronic nephritis are the same as the proteins in the plasma, and that the theory of leakage of plasma protein through the kidneys holds. The view of the identity of urinary and plasma proteins is supported by many observers. Hewitt [1927; 1929] has shown that urinary and serum albumins from cases of chronic nephritis have the same optical rotatory powers. Smith, quoted by Hewitt [1929], has shown that they have the same absorption spectra. Cavett and Gibson [1931] have shown that they have similar racemisation curves. Widdowson [1933] has investigated the optical rotation, racemisation, osmotic pressure and specific refraction of urinary and serum protein fractions in nephritis and nephrosis and concludes that the proteins from urine and serum are identical. Kozawa et al. [1933] claim to have shown that the amino-acid composition, albumin/globulin ratio and isoelectric(?) point of urinary proteins are the same as those of serum proteins. Nevertheless, there are certain points which are not explained. The fibrinogen of plasma is never found in the urine. The urinary proteins seldom have the same albumin/globulin ratio as the serum proteins, and not infrequently globulin is absent altogether from the urine. It thus becomes necessary to postulate a selective permeabilitv of the leaking kidneys to the different protein fractions of