All-trans-retinoic acid-induced growth suppression of blastemal Wilms' tumor.

All-trans-retinoic acid (RA) has been used to suppress growth of malignant cells and induce epithelial differentiation. We investigated whether RA had a similar effect on Wilms' tumor, a childhood tumor of the kidney that arises from the undifferentiated metanephric blastema. W13 cells, a cell line derived from a blastemal Wilms' tumor, were exposed to RA (10(-9)-10(-5) M) and its effects on cell proliferation, gene expression, and differentiation were examined. Treatment of W13 cells with RA resulted in a dose-dependent suppression of growth. Changes in expression of selected genes were determined by Northern analysis. After 24 h, there was a marked dose-dependent down-regulation of N-myc mRNA as well as up-regulation of insulin-like growth factor-II (IGF-II) mRNA. [125I]IGF-II ligand blotting of conditioned medium from RA-treated cultures revealed a dramatic alteration in the pattern of expression of insulin-like growth factor binding proteins (IGFBPs). Examination of RA-treated W13 cultures by light and electron microscopy did not reveal appreciable morphological changes. We conclude that RA inhibits growth and alters gene expression of W13 cells without inducing epithelial differentiation. The modulation of expression of IGF-II, IGFBP, and N-myc may play a role in RA-induced growth suppression of Wilms' tumor cells.

Characterization of a monoclonal antibody recognizing the blastemal element of Wilms' tumors and fetal kidneys.

A blastema-associated antigen (BLA-1) was detected using a monoclonal antibody against malignant blastema from a Wilms' tumor. The localization of BLA-1 was investigated in a series of nine Wilms' cases, five fetal, one childhood, and two adult kidneys. In this series, BLA-1 antibody consistently stained cell surfaces of all Wilms' tumors containing blastemal components. The same staining pattern was maintained in tumors grown as heterotransplants in nude mice. The expression of BLA-1 antigen was examined in normal blastema of fetal kidneys. BLA-1 was immunolocalized to condensed blastemal cells in the nephrogenic zone throughout gestation. In addition, kidney samples from a young child or adults contained no blastemal cells and therefore showed no blastemal cell surface staining. Glomerular mesangial cell staining was demonstrated in kidneys from 12 weeks of gestation through adulthood. This staining in developing and mature glomeruli implies that mesangial cells may be derived from condensed blastemal cells. The finding of a cell surface antigen common to Wilms' blastema, fetal blastema, and mesangial cells has not been previously demonstrated.

Characterization of a monoclonal antibody recognizing selective epithelial elements of Wilms' tumors and fetal kidneys.

A new antigen was detected using a monoclonal antibody generated against malignant blastema from a Wilms' tumor. This antigen showed variable expression in malignant blastemal cells but was never detected in normal blastema of fetal kidneys irrespective of gestational stage. In a series of 16 Wilms' tumors, the most intense and consistent staining was seen in tubule-associated epithelial cells. Such tubular staining is not surprising as the putative induction of malignant blastema to differentiate into malignant tubules is thought to parallel normal tubulogenesis. This antigen was also associated with epithelial cells located in a variety of fetal kidney structures. Again, the staining was most consistent in tubular epithelia. This monoclonal antibody reactive with a blastemal-epithelial-tubular (BET) antigen should be of value in studying the induction of epithelial differentiation in the normal and diseased human kidney.

Serum-free culture and characterization of renal epithelial cells isolated from human fetal kidneys of varying gestational age.

Cell cultures were initiated from seven human fetal kidneys that varied in gestational age from 90 days to newborn. The growth medium utilized was a 1:1 mixture of Dulbecco's modified Eagle's and Ham's F12 supplemented with selenium (5 ng/ml), insulin (5 micrograms/ml), transferrin (5 micrograms/ml), hydrocortisone (36 ng/ml), triiodothyronine (4 pg/ml), and epidermal growth factor (10 ng/ml). For all the kidney isolates, initial cell attachment occurred within 12 h through multicell spheroids, and by 24 h a rapidly growing population of cells was obtained. Confluency was reached within 3 to 6 days. A combination of light microscopy, immunohistochemistry, and ultrastructural evaluation was utilized to characterize the resulting cultures as epithelial and homogeneous within each isolate and among the isolates. That is, regardless of gestational age of the fetal kidney used as starting material, an identical or highly similar population of cells was obtained. By light microscopy, the cultures were noted to form very few domes, the number being an indication of transport activity. However, ultrastructural examination revealed that the cells were noted to form domes composed of only a few cells or "micro-domes" that would not be visible by light microscopy. Within the micro-domes as well as other areas of the monolayer an apparent absence of tight junctions was noted by routine transmission electron microscopy. However, by freeze fracture analysis cells were shown to possess sealing strands, the structural component of tight junctions. It is postulated that the tight junctions of fetal epithelial cells are structurally altered as compared to tight junctions in adult renal epithelial cell cultures.

Automatic quantitation of cell growth and determination of mitotic index using DAPI nuclear staining.

The growth rate of primary tumors and derived cell lines is an important identifying trademark to researchers studying the growth and differentiation of pediatric neoplasms. For the in vivo tumor specimen, the mitotic index is utilized to assess growth potential while, in vitro, the determination of the growth curve and doubling time of the cells is employed. Because of the importance of these parameters this laboratory has developed a technique for the simultaneous determination of cell number and mitotic index for adherent cultured cells, as well as solid tumors, using the polycationic dye DAPI (4',6-diamino-2-phenylindole). DAPI, when bound to nuclear DNA, fluoresces, and this property has been utilized to develop an image analysis based technique to determine cell number automatically based on gray scale discrimination. In addition, DAPI staining clearly delineates mitotic figures and this has allowed the simultaneous manual determination of mitotic index for each automatic cell count. This technique was first tested using four different human cell lines and was shown to determine cell growth and mitotic index simultaneously. Lastly, employing an atypical mesoblastic nephroma, it was shown that the mitotic index of the primary tumor could easily be obtained and compared to both the mitotic index of tumor heterotransplant in nude mice and the mitotic index of the tumor-derived cell culture. This technique should be useful in studies assessing the effects of various factors on the growth of adherent cultured cells as well as the accurate determination of the mitotic index in solid tumors.

Morphology and growth characteristics of epithelial cells from classic Wilms' tumors.

The ability to establish cell cultures representing the epithelial component of Wilms' tumor was determined for 18 cases of classic Wilms' tumors. From these 18 cases only two resulted in the culture of epithelial cells. Although the tumors from both cases were composed of a prominent epithelial component, other classic tumors not producing epithelial cell cultures also possessed appreciable epithelial components. Likewise, heterotransplants of these two primary tumors failed to give rise to epithelial cell cultures, although cultures of the blastemal element were produced. This suggests that Wilms' tumors may be prone to differentiate in different directions at varying times during tumor growth, possibly dependent on local tumor environment. Epithelial cells from these two classic cases were grown in culture in basal medium composed of a 1:1 mixture of Dulbecco's modified Eagle's medium and Ham's F-12 medium, supplemented with selenium, insulin, transferrin, hydrocortisone, tri-iodothyronine, and epidermal growth factor, on a collagen type I matrix with absorbed fetal calf serum proteins. One of the two cases also required the addition of bovine pituitary extract, ethanolamine, prostaglandin E1, and putrescine for optimum growth. Morphological analysis disclosed that the cultured cells were very similar to normal renal tubular cells in culture, except that the cells displayed little evidence for differentiated active ion transport and tended to grow in a multilayered arrangement. The culture of the epithelial cells from classic Wilms' tumors provides a model system for the study of tumor differentiation and progression.

The G401 cell line, utilized for studies of chromosomal changes in Wilms' tumor, is derived from a rhabdoid tumor of the kidney.

The histology, ultrastructure, and messenger RNA expression of heterotransplants derived from the G401 cell line (American Type Culture Collection) have been characterized by comparison with Wilms' and rhabdoid tumors of the kidney. This analysis illustrates that the properties of G401 heterotransplant were consistent with a rhabdoid phenotype rather than that of a Wilms' tumor. The G401 cell line has been utilized in recent experiments to demonstrate the central role of chromosome 11 in Wilms' tumor. However, the present results suggest that these experiments may be more relevant to define the involvement of chromosome 11 in rhabdoid tumor of the kidney, a malignancy distinct from Wilms' tumor. This is clinically relevant since the rhabdoid tumor of the kidney is very aggressive and associated with an extremely poor prognosis.

Acetonitrile as a substitute for ethanol/propylene oxide in tissue processing for transmission electron microscopy: comparison of fine structure and lipid solubility in mouse liver, kidney, and intestine.

Tissue processing for transmission electron microscopy (TEM) is commonly accomplished using ethanol (EtOH) as a dehydrating solvent and propylene oxide (PO) as a transition fluid. Both solvents have some undesirable properties: EtOH solubilizes lipids; PO is highly flammable, volatile, toxic, and potentially carcinogenic. Their replacement by a compound devoid of these characteristics is therefore desirable. Acetonitrile (AN) appears to be such a solvent. It is freely miscible with water, alcohols, acetone, and epoxy resins; it does not interfere with epoxy polymerization; and the resulting cured resins have excellent cutting quality and beam stability. AN is also an excellent dehydrating agent whose use does not necessitate modification of current techniques. Most importantly, the low solubility of phospholipids (PL) in AN limits the loss of membrane lipids and, hence, leads to a better preservation of tissue features.

Age-related changes in cochleas of mongolian gerbils.

The effects of aging on the gerbil cochlea were studied in 16 animals raised in a quiet environment. Animals were tested at ages ranging from 33 to 36 months, the approximate average lifespan of gerbils in our colony. Hearing sensitivity was assessed by measures of whole-nerve compound action potential (CAP) thresholds and surface preparations of the organ of Corti were subsequently examined by light microscopy for losses of sensory hair cells. These quiet-aged animals showed a wide range of hair-cell losses and threshold shifts. Outer hair cells often showed significant losses while inner hair cells were rarely absent. All animals had some threshold shift, especially at frequencies above 4 kHz. These shifts ranged from 1 to 68 dB. At high frequencies, threshold shifts often occurred without hair-cell losses at corresponding cochlear locations. At low frequencies, threshold shifts seldom reflected the losses of hair cells commonly found in the cochlear apex. Thus, the correlation of specific hair-cell losses and CAP threshold shifts at corresponding frequencies was poor. On the other hand, the total number of missing hair cells, irrespective of location, was a good, general indicator of the hearing capacity in a given ear. It appears that the factor or factors that makes cochleas susceptible to hair-cell loss with increasing age also affects other cochlear mechanisms that are necessary for normal functioning of the ear.

Isolation and characterization of a mannose receptor from human pigment epithelium.

Recent work demonstrated that a mannose receptor is involved in the phagocytosis of rod outer segments by the rat retinal pigment epithelium (RPE). In this study the binding of soluble mannose-containing ligands by human RPE explants is described. In addition, the authors report the isolation of a mannose receptor from human RPE and describe its relationship to the macrophage mannose receptor. Epithelial explants bound the soluble ligand 125I-mannose bovine serum albumin (BSA) by a mannose-specific process. The protein involved in mannose recognition was extracted from human tissue and purified using ligand-affinity chromatography. The protein that bound to the affinity column had a molecular weight of 175 kD by sodium dodecyl sulfate gel electrophoresis and migrated with the same mobility as the human macrophage mannose receptor. Antibodies directed against the macrophage receptor crossreacted with the mannose receptor from human RPE by immunoblot analysis. Binding specificity studies demonstrated that mannose and mannan inhibited ligand binding to the purified receptor by 65% and 90%, respectively; galactose had no effect. Using immunogold labeling of human RPE cells in explant culture, antimacrophage mannose receptor was localized at the apical plasma membrane. These results suggest that human RPE expresses a mannose receptor on its apical surface (as does the rat RPE) and that this receptor is similar to the human macrophage mannose receptor.

DAPI as a useful stain for nuclear quantitation.

A simple-to-use fluorescent stain, 4',6-diamidino-2-phenylindole (DAPI), visualizes nuclear DNA in both living and fixed cells. DAPI staining was used to determine the number of nuclei and to assess gross cell morphology. Following light microscopic analyses, the stained cells were processed for electron microscopy. Cells stained with DAPI showed no ultrastructural changes compared to the appearance of cells not stained with DAPI. DAPI staining allows multiple use of cells eliminating the need for duplicate samples.

Onset of vitellogenin production and vitellogenesis, and their relationship to changes in the midgut epithelium and oocytes in the tick Dermacentor variabilis.

In Dermacentor variabilis (Say), the onset of vitellogenin production and vitellogenesis (uptake of vitellogenin into oocytes) began during the rapid-engorgement feeding period. Mating was required for both vitellogenin production and vitellogenesis to complete the tick's life cycle. Complete immunological identity, as measured by Ouchterlony's double diffusion test, existed between vitellogenin from the fat body, midgut and hemolymph, and vitellin from the ovaries and eggs. Antivitellin antibody did not react with host hemoglobin nor with fat body, midgut, and ovary extracts from feeding females prior to rapid engorgement, feeding unmated females, or unfed or fed males. Some unmated females fed for 13 days and then hand-detached from the host eventually began oviposition after going through a preoviposition period. In these ticks, organ extracts from the midgut, fat body and ovary reacted with antivitellin antibody. The presence or absence of presumed vitellogenic cells in the midgut and yolk bodies in oocytes corresponded with the presence or absence of vitellogenin and vitellogenesis as measured by Ouchterlony's test. Presumed vitellogenic cells increased in size during the preoviposition period. These cells reached their greatest size during the time when the most eggs were being produced, and then declined in size toward the end of oviposition. Vitellogenin was deposited directly into developing yolk bodies in oocytes and was not processed through lysosomes. Feeding was the process that initiated the formation of eggshell cuticle. Detachment from the host was required for the initiation of oviposition.

Ultrastructure of the midgut and blood meal digestion in the adult tick Dermacentor variabilis.

Digestive cells in the midgut of male and female Dermacentor variabilis (Say) took up the blood meal in coated vesicles and smooth flask-shaped vesicles, and deposited it in endosomes which were digested via heterophagy. Iron was concentrated in residual bodies. Digestion occurred in three distinct phases in mated females: (1) continuous digestion (initiated by feeding) occurred during slow engorgement; (2) reduced digestion (initiated by mating) occurred in mated females during the period of rapid engorgement; (3) a second continuous digestion phase (initiated by detachment from the host) occurred throughout the post-feeding periods of preoviposition and oviposition. It proposed that the stem cells in the midguts of unfed females were progenitor of digestive, replacement, and presumed vitellogenic cells in midguts of mated feeding females. Digestive cells were present in all three digestion phases. Only during the first continuous digestion phase did digestive cells fill up with residual bodies, rupture and slough into the lumen, or did whole cells slough into the lumen. During the other two digestion phases no sloughing of digestive cells was observed. At the end of oviposition the digestive cells were filled with residual bodies. Replacement cells were present only during the first continuous-digestion phase. Presumed vitellogenic cells were present only during the reduced-digestion phase and during the second continuous-digestion phase. Stem cells in unfed males developed only into digestive cells in feeding males. Fed males and fed unmated females had only the first continuous-digestion phase. After being hand-detached from the host, unmated 13-day-fed females went through cellular changes associated with the reduced-digestion phase and second continuous-digestion phase of fed mated females, then began ovipositing. Maximum development of the basal labyrinth system and lateral spaces matched the known time of maximum water and ion movement across the midgut epithelia. Spectrophotometric analyses of lumen contents and midgut cells, sampled after detachment from the host, showed that concentrations of protein and hemoglobin at day 1 post-detachment decreased by one-half at the beginning of oviposition, while hematin increased about twofold by the end of oviposition. This supported the idea of the presence of a second continuous-digestion phase.

Expression of mannose receptors for pinocytosis and phagocytosis on rat retinal pigment epithelium.

We report here the presence of a mannose-specific receptor on apical membranes of rat retinal pigment epithelial (RPE) cells. For pinocytic studies, 125I-Mannose-BSA (125I-Man-BSA) was incubated with RPE explants from normal (Long Evans) and dystrophic (pigmented RCS) rat retinas. Normal RPE bound 36.1 ng of ligand and, in the presence of mannan competitor, the amount bound was 18.3 ng. In a similar assay, total ligand uptake by dystrophic RPE was 25.9 ng with 9.8 ng specific for mannose recognition. Comparing the amounts of ligand bound, dystrophic RPE recognized 55% of the amount recognized by normal RPE. The presence of the mannose receptor was localized on both normal and dystrophic RPE apical plasma membranes by autoradiographic techniques using 125I-Man-BSA. Normal RPE showed a greater number of silver grains present at the apical cell membrane as compared to dystrophic RPE. Silver grains were significantly reduced when incubation with the ligand was carried out in the presence of a mannan competitor. Further, in phagocytic studies, latex beads coated with mannan were used as phagocytic particles. Normal RPE phagocytized 4.52 mannan-beads per cell profile by a mannose-specific mechanism, whereas dystrophic RPE did not recognize mannan-beads. Our data suggest that RPE cells express surface receptors which recognize both soluble and particulate mannose ligands. The pinocytic and autoradiographic studies suggest that normal RPE binds more soluble ligand than does dystrophic RPE. If the mannose receptors mediate both pinocytosis and phagocytosis, a possible reduction in number of soluble mannose binding sites on the dystrophic RPE may be related to the diminished phagocytic recognition of particulate ligand by the dystrophic RPE.(ABSTRACT TRUNCATED AT 250 WORDS)

Mannose 6-phosphate receptors on the plasma membrane on rat retinal pigment epithelial cells.

The retinal pigment epithelium (RPE) phagocytizes the tips of photoreceptor outer segments (OS) during normal eye function. It is not known what ligand on OS is recognized by the RPE for removal from the interphotoreceptor matrix. It is possible that a sugar residue on a cell surface glycoconjugate of either the OS or RPE mediates the phagocytic interaction. Pinocytic experiments with a soluble mannose 6-phosphate ligand (125I-labelled mannosidase) showed that similar quantities of ligand were bound by RPE explants from Long Evans rat retinas and from Royal College of Surgeons (RCS/p+) rat retinas known to be defective in the phagocytosis of OS. The addition of mannose 6-phosphate reduced the total counts of bound alpha-mannosidase by 23% in both normal and dystrophic RPE explants. Mannose 6-phosphate receptors were visualized on normal and dystrophic RPE plasma membranes by immunocytochemical techniques. Further, phagocytosis was studied using phosphomannan-coated beads as phagocytic particles. Dystrophic RPE phagocytized phosphomannan-coated beads by a mannose 6-phosphate specific mechanism as shown by a significant reduction in the number of these coated beads taken up in the presence of the competing sugar. In contrast, normal RPE showed no uptake of phosphomannan-coated beads. Our findings indicate that a mannose 6-phosphate receptor is on the apical plasma membrane of rat RPE. This receptor may not be involved in normal OS phagocytic recognition, but may function in the trafficking of lysosomal enzymes by RPE cells.

Phagocytic interactions of sialated glycoprotein, sugar, and lectin coated beads with rat retinal pigment epithelium.

Latex beads coated with mucin (a sialic acid containing glycoconjugate), N-acetylglucosamine (GlcNAc), or with the lectins, succinylated wheat germ agglutinin (sWGA) (binds to GlcNAc) or Limax flavus agglutinin (LFA) (binds to sialic acid residues) were used as phagocytic particles. Phagocytosis of these coated beads by retinal pigment epithelial (RPE) explants was determined by bead uptake in normal (Long Evans) and dystrophic (Royal College of Surgeons, RCS/p+) rat retinas. Electron microscopy showed that beads coated with mucin or LFA lectin were not phagocytized by either dystrophic or normal RPE. sWGA-coated beads were phagocytized by both dystrophic and normal RPE, while GlcNAc-coated beads were taken up by dystrophic RPE only. Specificity of uptake for sWGA and GlcNAc bead coatings was shown by the reduction in the number of beads phagocytized in the presence of the appropriate competing sugar or lectin. The lack of phagocytic uptake of beads coated with a sialated glycoprotein or a sialic acid binding lectin suggests that sialic acid residues are not recognized as particulate ligands in this phagocytic assay. The data which show the uptake of beads coated with sWGA (binds only GlcNAc) together with results which showed WGA (binds both GlcNAc and sialic acid)-coated beads were not taken up, further suggest that GlcNAc residues may be involved in bead phagocytosis. The most striking difference between normal and dystrophic RPE engulfment of coated beads is the uptake of GlcNAc-coated beads by dystrophic RPE only. These results suggest that the receptor molecules on dystrophic RPE cell surface membranes may be different from those on normal RPE membranes.

Lectin-affinity isolation of microvillous membranes from the pigmented epithelium of rat retina.

The pigmented epithelium of the vertebrate retina phagocytizes the discarded tips of photoreceptors and it is likely that a specific cellular recognition process is involved in this phenomenon. The apical surface of retinal pigmented epithelium (RPE) contains microvilli which interdigitate with the outer segment regions of photoreceptor cells and it is this apical microvillous surface that is of particular interest with respect to phagocytosis. The present study is a report of a method to isolate a fraction that is enriched in microvilli from the apical surface of this highly polarized epithelial cell. Wheat germ agglutinin (WGA) conjugated sepharose beads are used to remove the microvillous membranes which are then observed with scanning and transmission electron microscopy. The proteins of this RPE-subfraction are separated through use of SDS-polyacrylamide gel electrophoresis. The relative molecular weights (Mr) and lectin binding properties of glycoproteins are examined in Western blots through the use of lectin-peroxidase conjugates as probes for carbohydrate residues. A preliminary comparison of membranes isolated from Long Evans (normal) and Royal College of Surgeons (dystrophic) rat retina RPE shows that the glycoproteins in these two preparations are different with respect to the binding of Concanavalin-A (Con-A) and WGA. In particular a glycoprotein in the normal RPE preparation with a Mr of 175K binds Con-A and WGA, but in the dystrophic RPE preparation binds little or no WGA. A glycoprotein present in the normal RPE preparation with a Mr of 86K binds Con-A and WGA, but both lectins have reduced binding sites in the dystrophic preparation. Limax flavus agglutinin (specific for sialic acid residues) binds to a high molecular weight glycoprotein with a Mr of 195K-196K which is present in both normal and dystrophic RPE membrane preparations and which also binds Con-A and WGA.