Thymus and myasthenia gravis: antigen processing in the human thymus and the consequences for the generation of autoreactive T cells.

Peptides displayed by antigen presenting cells in the thymus shape the T cell repertoire. We investigated the antigen processing machinery of the MHC class II presentation pathway and describe the differential expression of lysosomal proteases in compartments of the thymus and the peripheral lymphoid tissue. Overexpression of certain proteases found in the thymus and thymoma associated with myasthenia gravis is likely to affect tolerance induction and may promote the generation autoreactive CD4(+) T helper cells.

1-C-glucuronidation of N-nitrosodiethylamine and N-nitrosomethyl-n-pentylamine in vivo and in primary hepatocytes from rats pretreated with inducers.

The organ specificity of the carcinogenic action of nitrosamines is partly explained by organ specific activation. The specificity might also be determined by conjugation of reactive intermediates in e.g. the liver. 14C-Labeled N-nitrosodiethylamine (NDEA) a liver carcinogen and N-nitrosomethyl-n-pentylamine (NMPentA) which induces esophageal and nasal tumors were administered to rats or incubated with primary cells. Urine and cell extracts were separated by HPLC after addition of synthetic marker glucuronides and these were quantified by liquid scintillation counting. In urine of rats treated with NDEA 0.03% of administered nitrosamine was recovered as the O-glucuronide derived from N-nitroso-1-hydroxyethylethylamine. In rats treated with NMPentA 2.86% was metabolized to the glucuronide at the methyl group. In hepatocytes of untreated rats 0.03% of the added NDEA was conjugated to the glucuronide, phenobarbital pretreatment induced this conjugation reaction 5-fold. Hepatocytes from untreated rats metabolized 1.2% of NMPentA to the primary glucuronide; after phenobarbital pretreatment this value increased to 1.6%. In hepatocytes from 3-methylcholanthrene-pretreated rats, 0.04% of NMPentA was metabolized to the glucuronide derived from N-nitroso-1-hydroxy-n-pentyl-methylamine, while 0.85% was derived from N-nitroso-hydroxymethyl-n-pentylamine. In hepatocytes from Aroclor-pretreated rats, 0.09% were pentyl conjugates and 1.1% methyl conjugates. The induction pattern and organ specificity of glucuronidation indicate that all three 1-hydroxy nitrosamines are conjugated by group II phenobarbital inducible UDP-glucuronosyltransferase activity. The lipophilicity of a nitrosamine seems to determine the extent of glucuronidation in hepatocytes and in vivo. No glucuronides derived from either NDEA or NMPentA were detectable in incubations with kidney cells, nor was the glucuronide of NDEA found in incubations with whole bladders.

Maternal serum alpha-fetoprotein screening of fetal trisomies.

Thirty-two trisomy pregnancies were retrospectively studied in which the maternal serum level of alpha-fetoprotein was determined prior to amniocentesis. Median levels of alpha-fetoprotein in serum (0.83 multiples of the median) and in amniotic fluid (0.72 multiples of the median) were lowered, but this was statistically significant only in the case of amniotic fluid.

Maternal serum alpha-fetoprotein screening for the detection of neural tube defects.

We tested 10,715 low-risk pregnancies in a voluntary maternal serum alpha-fetoprotein screening program for the detection of neural tube defects in California. In all, 5.3 percent of women had one elevated serum level, 3.3 percent were referred for sonography and 1.5 percent for amniocentesis. There were 12 cases of open neural tube defects (1.1 per 1,000); all of the mothers had one elevated serum alphafetoprotein level: nine (75 percent) completed the protocol and the neural tube defects were correctly identified. No normal pregnancies were terminated. The risk of an open neural tube defect occurring was about 1 in 50 after the first abnormal serum level and 1 in 15 at amniocentesis. We found significantly increased risk for fetal death and low birth weight after one elevated serum alpha-fetoprotein level, though the likelihood of a normal pregnancy outcome was about 80 percent. Maternal serum screening was also useful in identifying twin pregnancies and correcting underestimated gestational dates.

Alpha-fetoprotein concentrations in maternal serum: relation to race and body weight.

We confirmed the relation between maternal weight and serum alpha-fetoprotein concentration and have shown a further relation--to ethnic origin. Of these two, maternal weight is the more closely related. Oriental, white, and Hispanic women showed no significant differences in serum AFP concentrations when corrections for maternal weight were applied. Black women showed consistently higher values, by an average of 10% at each week of gestation. These corrections only affected values falling just above or below the 95th centile cutoff. We conclude that corrections for maternal weight and race should be applied when values for alpha-fetoprotein in maternal serum are being interpreted.