The glycosylation of serum transferrin from galactosemic patients with a deficiency of galactose-1-phosphate uridyl transferase (EC 2. 7.7 12) is abnormal but becomes normal after treatment with a galactose-free diet. To understand the structural and biochemical basis of the abnormal glycosylation, transferrin was purified from the serum of untreated and treated galactosemic patients and normal controls and the N-linked glycans analyzed by HPLC. The glycans from normal transferrin consisted predominantly (86%) of the disialylated biantennary complex type. The glycans from untreated galactosemic patients were more heterogeneous and contained four major truncated glycans in addition to a smaller amount (13%) of the disialylated biantennary complex type. The truncated glycans were deficient in galactose and sialic acid and their structures were consistent with a decrease in galactosyltransferase activity in hepatocytes, the probable cells of origin of the transferrin. This is postulated to be due to direct inhibition of the galactosyltransferase activity by the accumulated galactose-1-phosphate or to an effect on the formation of UDP-galactose, the donor substrate in the reaction. After treatment the proportion of the truncated glycans decreased and the proportion of the disialylated biantennary complex type increased, returning almost but never completely to normal, even after prolonged treatment in some cases. There was no clear relationship between the length of treatment and the normalization of glycosylation and the level of galactose-1-phosphate in red blood cells, the usual parameter for monitoring the treatment of galactosemics. It is suggested that the persistence of abnormally glycosylated proteins may contribute to the long-term complications in galactosemia.

Hyaluronate synthase activity is localized exclusively in plasma-membrane fractions of cultured human skin fibroblasts. The enzyme activity of plasma membranes prepared from exponential-growth-phase cells was about 6.5 times that of stationary-growth-phase cells. Hyaluronate synthase from exponential-growth-phase cells exhibited lower Km and higher Vmax. values for both UDP-N-acetylglucosamine and UDP-glucuronic acid and higher rate of elongation of hyaluronate chains compared with the enzyme from stationary-growth-phase cells. Hyaluronate synthase exhibited an extremely short half-life, 2.2 h and 3.8 h respectively when cells were treated with cycloheximide and actinomycin D. The cell-growth-phase-dependent variations in hyaluronate synthase activity appear to be due to its high turnover rate as well as due to some post-translational modification of the enzyme protein as cells progress from early exponential to stationary growth phase. The isolated plasma membranes contained a protein (Mr approx. 450,000) that was selectively autophosphorylated from [gamma-32P]ATP in vitro in the presence of hyaluronate precursors in the reaction mixture and that also exhibited some hyaluronate-synthesis-related properties. The 32P-labelled protein isolated from plasma membranes of exponentially growing cells expressed an efficient UDP-[14C]glucuronic acid- and UDP-N-acetyl[3H]glucosamine-binding activity and was able to synthesize oligosaccharides (Mr 5000) of [14C]glucuronic acid and N-acetyl[3H]glucosamine residues. The corresponding protein of stationary-growth-phase cells, which expressed much higher nucleotide-sugar-precursor-binding activity, appeared to have lost its oligosaccharide-synthesizing activity.

SUMMARY Cadaveric hip joints were tested in a hip function simulator which subjected the femoral head to a cycle of loading and oscillation similar to that experienced during walking and measured the frictional torque transmitted to the acetabulum. Silicone fluids with viscosities from 102

Poor fetal and infant nutrition has been linked to impaired glucose tolerance in later life. We studied the effect of protein deficiency during gestation and the suckling period in a rat model and found that poor nutrition 'programmes' pancreatic beta-cell GK (glucokinase; known as the glucose sensor) and glucose-stimulated insulin secretion response in newborn, suckling and adult rat offspring. Pregnant female rats were divided into three groups: a control group was kept on a normal protein (20%) diet, another group was fed a low-protein (LP) (6%) diet during gestation and suckling periods (LP-G + S group) and another was fed a LP diet during gestation then a normal protein diet during the suckling period (LP-G group). The pulsatile glucose-stimulated insulin secretion response was acutely disrupted and the peak insulin secretion was markedly decreased in newborn and 3-week-old offspring of the LP-G + S group compared with the control group. Also, there was an altered pulsatile secretory response in adults of the LP-G + S and 3-week-old and adult offspring of the LP-G groups compared with the control group. GK protein levels, detected by Western blotting, were decreased in newborn and 3-week-old offspring of both LP-G + S and LP-G groups compared with the control groups. The Km and Vmax of GK were altered. The prenatal and postnatal LP diet appeared to have a permanent effect in increasing the affinity of GK for glucose (indicated by decreased Km values) and decreasing the Vmax. This showed that the critical period of programming of the function of GK was after birth and during the postnatal weaning period, since the adult offspring of the LP-G + S group when fed a normal protein diet showed no reversal in the Km values of the enzyme. Similar experiments in adult offspring of the LP-G group showed normalization of the Km values of GK at 3 weeks of age. In conclusion, fetal and infantile nutrition 'programmes' pancreatic beta-cell function; poor nutrition during this period caused irreversible effects on glucose homoeostatic mechanisms in the offspring, which may predispose the offspring to diabetes in later life.

Conditions under which the glycosylation capacity of cells is limited provide an opportunity for studying the efficiency of site-specific glycosylation and the role of glycosylation in the maturation of glycoproteins. Congenital disorders of glycosylation type 1 (CDG-I) provide such a system. CDG-I is characterized by underglycosylation of glycoproteins due to defects in the assembly or transfer of the common dolichol-pyrophosphate-linked oligosaccharide precursor of asparagine-linked glycans. Human plasma alpha1-antitrypsin is normally fully glycosylated at three asparagine residues (46, 83, and 247), but un-, mono-, di-, and fully glycosylated forms of alpha1-antitrypsin were detected by 2D PAGE in the plasma from patients with CDG-I. The state of glycosylation of the three asparagine residues was analyzed in all the underglycosylated forms of alpha1-antitrypsin by peptide mass fingerprinting using matrix-assisted laser desorption ionization time-of-flight mass spectrometry. It was found that asparagine 46 was always glycosylated and that asparagine 83 was never glycosylated in the underglycosylated glycoforms of alpha1-antitrypsin. This showed that the asparagine residues are preferentially glycosylated in the order 46>247>83 in the mature underglycosylated forms of alpha1-antitrypsin found in plasma. It is concluded that the nonoccupancy of glycosylation sites is not random under conditions of decreased glycosylation capacity and that the efficiency of glycosylation site occupancy depends on structural features at each site. The implications of this observation for the intracellular transport and sorting of glycoproteins are discussed.