Summary

The anterior surface of the human iris was studied by transmission and scanning electron microscopy. Melanocytes, spongy grouped cells looking like fibrocytes, and bundles of collagen fibers between the cells were found. Holes and pores of different size reach down to the iris stroma. No closed-banded endothelium was found on the anterior surface of the iris. It is assumed that the aqueous humor has free access to the iris stroma. The question of circulation and drainage of aqueous humor along the iris is discussed.

Summary

Qualitatively, there was no difference in the metabolism of the three substances. No indication of a specific metabolic reaction with the carcinogenic compound was found. The dimethylamines were demethylated and the resulting amines acetylated. 4′-hydroxy-, N-hydroxy-, 3-hydroxy-acetylamines and 3-hydroxy-amines were identified as hydroxylation products. 3-or 4′-hydroxy-dimethylamines, 4′-hydroxy-amines as well as acetylation products of methylamines were not detected. In urine the 3-hydroxy-amines and 4′-hydroxy-acetylamines were excreted as sulfates, the N-hydroxy-and 3-hydroxyacetylamines as glucuronides. 80–90% of the fecal metabolites were unconjugated. The results indicate N-hydroxylation of methylamines and amines beside the acetylamines.

Quantitatively, not only the amounts of metabolites produced, but also the excretion patterns were different. The three acetylamines were N-hydroxylated to a comparable extent, assuming ortho-hydroxy-acetylamines to be rearrangement products of N-hydroxy-acetylamines or one of their derivatives. 5 hrs after trans-DAS administration metabolites are bound to precipitable material in the liver to an extent (25 nMole/g liver) comparable to known liver carcinogens. This indicates an activating metabolism in a tissue not susceptible to tumor induction in our experiments.

The results indicate the very differentiated chemico-biological interactions involved after administration of the three structurally closely related amines, and how difficult it will be to find reliable correlations between these interactions and the biological effect. More precise knowledge about the pharmacokinetics of these substances will be necessary.

Human galactose-l-phosphate uridyl transferase was purified from post-mortem liver to a preparation having a single band in polyacryamide gel electrophoresis. This preparation was used successfully to produce a rabbit antibody that precipitates transferase activity from solution and that forms a precipitin band in double immunodiffusion. Hemoglobulin-free erythrocyte preparations from homozygous normal (Gt⁺/Gt⁺), Durate Variant (Gt^D/Gt^D), and galactosemic (Gt^G/Gt^G) individuals show immunoprecipitin bands in double immunodiffusion against this antibody that are identical with that of the purified transferase preparation. The results indicate that the three alleles code for immunologically similar enzyme proteins suggesting that the functionally less active Duarte Variant and inactive galactosemic enzyme proteins have resulted from “point” mutations.

Summary

Ischaemic necrosis of the villi was produced by occlusion of the superior mesenteric artery for 1 hour and subsequent regeneration of the epithelium examined after 1,2, 4, 8, 16, 24, and 48 hours. Regenerating cells above the crypts were first identified in significant numbers after 2 hours. Complete re-epithelization occasionally seen at 4 hours resulted in mucosa completely devoid of villi. Frequent hyperplastic foci, spurs and synechiae occurred in the sheets of regenerating epithelium at an early stage. Most frequently re-epithelization was complete between 8 and 16 hours and short villi formed which then increased in length. No mitoses were seen. By electron microscopy crypt epithelium showed maximum changes at 2–4 hours. Many crypt cells were affected by coagulation necrosis and included into surviving cells. In addition, there were cytolysomes, which frequently contained only ribosomes. The cells of regenerating epithelium were arranged into one or two, rarely more, layers. A normal arrangement appeared after complete re-epithelization. In early stages, the cell shape was irregular. The basal cells tended to be elongated and sealed the intercellular spaces against the basement membrane by their processes. The basement membrane which showed focal disruption was left after the desquamation of necrotic epithelium. At 4 hours and later, a polar organization of cells was seen. Junctions of the intermediate type were observed after 2 hours. At 4 hours junctional complexes were formed with frequent irregularities in the sequence of their components. The number and regularity of microvilli increased gradually. The first sign of microvillus formation was an elevation of the plasma membrane with condensation of the fuzzy layer; core filaments appeared when the microvillus had reached a certain length. The terminal web was only rudimentary before microvilli were organized to form a brush border. The height of the brush border was less than normal at 24 hours. Interdigitations at the lateral plasma membrane appeared at 16 hours; however, they did not reach normal extent at 24 hours. The first cells in the advancing rows of regenerating epithelium lagged in the surface differentiation until the re-epithelization was completed. The cytoplasm contained numerous free ribosomes and sparse endoplasmic reticulum. The amount of the latter, after 24 hours, was less than normal. A consistent feature of the later stages of regeneration (8–24 hours) was a densification of the mitochondrial matrix and accumulation of lipid droplets within well developed Golgi complexes and within occasional vesicles of endoplasmic reticulum. Cytolysomes developing into residual bodies and inclusions of necrotic cells were frequent in’ the early stages. Nuclei showed prominent nucleoli and occasional clusters of interchromatin granules. The differentiating goblet cells, aside from increased development of endoplasmic reticulum and Golgi complex showed features common to regenerating cells. Degenerative changes affected regenerating cells both on newly forming villi and in aberrant foci.

It is assumed that mitoses in crypts are sufficient to produce cells in the regenerated zone. The early loss of polar organization and simplification of cell surface is interpreted as dedifferentiation. The comparison of sequential stages of regeneration indicates that differentiation takes place with the progress of regeneration. The environmental difference is suggested as a factor. The changes in organelles are similar to regenerating cells in some other tissues and are non-specific though characteristic for rapidly proliferating cells. The degenerative changes which affect some regenerating cells are thought important in reestablishing normal organization.

The binding between bilirubin and human serum albumin was studied spectrophotometrically and with Sephadex gel filtration. Spectral absorption curves of solution of bilirubin in buffers containing decreasing amounts of albumin were registered. At a molar bilirubin to albumin ratio of 1:1 a change in these curves takes place, indicating that only one molecule of bilirubin is tightly bound to albumin. Sephadex gel filtration studies also showed that with solution containing bilirubin and albumin in molar ratios above 1:1 bilirubin seemed, however, to be more loosely bound.

The toxic effect of bilirubin on human fibroblasts in tissue culture was also tested. When solutions containing bilirubin to albumin ratios above 1:1 were added to the growth medium, a toxic effects was seen on the fibroblast growth. With a bilirubin to albumin ratio of 2:1 rapid cell death was found with total bilirubin concentrations as low as 5 mg/100 ml. On the other hand, when the bilirubin to albumin ratio in the growth medium was below 1:1, the cells tolerated bilirubin concentrations as high as 20 mg/100 ml, which was the highest tested. It therefore seems that only one molecule of bilirubin is tightly bound to each molecule of albumin, ab\nd that this molecule only is detoxified.

From 10 to 34 days of age, the weight gain was studied in 134 low birth weight (LBW) infants (birth weight range 0.99-1.90 kg: gestational age range 28-38 weeks) on three different diets: (1) human milk; (2) a “humanized” milk formula; (3) a high protein, high CL low fat cow's milk foumula. The caloric intake (keal/kg/24 hr) was always about 120 between 10 and 20 days, and about 140 thereafter. Six experimental groups were separated according to the three diets, and whether infants were “large for date” (LFD, i. e., with birth weight > the 25th percentile for gestational age) or “small for date” (SFD, i. e., with birth weight ≤ the 25th percentile, age 50'; of subjects ≤ the 10th percentile). Weight gain curves were calculated by regression analysis using the least square method according to a polynomial model. In each group, an exponential relationship of weight gain with time was found. In SFD infants weight gain was slightly but significantly (P < 0.001) father than in LFD babies on diet 3, not on diets 1 and 2. Either in LFD or in SFD infants weight gain was markedly faster on diet 3 as compared to diet 1 or 2 (P < 0.001), and slightly but significantly (P < 0.001) faster on diet 1 as compared to diet 2. Previous findings of a faster weight gain of LBW newborns on high protein formulas, as compared to human or “humanized” milk foumulas, were therfore confirmed. The following main conclusions and speculations were also made: in studies on growth refeeding. LFD and SFD newborns should be discriminated, and statistiscal methods able to describe a possible nonlinearity of growth curves should be used; SFD babies presumably need more proteins than LFD infants for optimal weight gain in the early weeks of life.