Ultrastructural characteristics of the spleen in hairy cell leukemia.

Eighteen spleens derived from patients with hairy cell leukemia (HCL) were analyzed by correlative scanning and transmission electron microscopy. In 15 of the cases, the white pulp areas were markedly decreased or absent when compared to normal spleens, although few hairy cells were observed within this region. In only one case did the white pulp appear normal. In all HCL cases, hairy cells were observed within normal, dilated, and abnormal sinuses. The abnormal sinuses contained hairy cells of typical morphology attached to other hairy cells, to endothelial lining, and to erythrocytes. The degree of sinus filling by hairy cells varied from loosely- to tightly-packed. Endothelial cells exhibiting degenerative changes, such as swelling with smooth surfaces and dilated intercellular spaces, were frequently seen. These results indicate that in addition to the previously described overcrowding of the spleen by hairy cells, the splenic tissue itself is considerably altered and sometimes severely damaged in patients with HCL.

B-cell growth factor-induced and alpha-interferon-inhibited proliferation of hairy cells coincides with modulation of cell surface antigens.

alpha-Interferon (IFN-alpha) induced unique ultrastructural alterations in peripheral blood and splenic hairy cell leukemia (HCL) cells (14 of 20 cases) treated in vitro. To further investigate the effects of B-cell growth factor (BCGF) and IFN-alpha on target hairy cells (HCs), we utilized immunogold labeling in conjunction with scanning electron microscopy. This methodology, in contrast to other immunological methods, facilitated direct view of the expression, density, and rearrangement of selected antigens/receptors on individual cells before and after BCGF or IFN-alpha treatment. In addition to inducing proliferation of HCL cells, BCGF enhanced the expression of interleukin 2 receptors (CD25; T-activated cell antigen) with no change in the expression of class I and class II human leukocyte antigen. On the other hand, IFN-alpha did not exert a noticeable proliferative effect on HCL cells but rather inhibited the proliferation of BCGF-treated cells. In addition, IFN-alpha treatment revealed an enhanced expression of class I (4 of 9) and class II (12 of 15) human leukocyte antigen on target HCs. Two-day exposure of HCs to IFN-alpha resulted in enhanced expression of CD25 (11 of 14), whereas a decrease in CD25 expression was recorded in 4 of 5 cases treated with IFN-alpha for 3 days. Also, no significant change in the expression of two other HCL-related surface antigens, CD22 (S-HCL-1; Leu-14) and CD11c(S-HCL-3; Leu-M5), was recorded following up to 3 days of IFN-alpha or BCGF treatment. However, a 5-day exposure to IFN-alpha resulted in a significant decrease in expression of CD11c on treated HCs. Finally, the IFN-alpha-induced immunoultrastructural changes in target HCs were primarily encountered in cells from HCL cases classified as responders to in vivo IFN-alpha therapy. Our data add support to the concept that the effect of IFN-alpha in HCL is mediated by impairment of the response to B-cell growth factors and induction of further differentiation of the target cells.

Properties and topography of the major integral plasma membrane protein of a unicellular organism.

The cellular distribution, membrane orientation, and biochemical properties of the two major NaOH-insoluble (integral) plasma membrane proteins of Euglena are detailed. We present evidence which suggests that these two polypeptides (Mr 68 and 39 kD) are dimer and monomer of the same protein: (a) Antibodies directed against either the 68- or the 39-kD polypeptide bind to both 68- and 39-kD bands in Western blots. (b) Trypsin digests of the 68- and 39-kD polypeptides yield similar peptide fragments. (c) The 68- and 39-kD polypeptides interconvert during successive electrophoresis runs in the presence of SDS and beta-mercaptoethanol. (d) The 39-kD band is the only major integral membrane protein evident after isoelectric focusing in acrylamide gels. The apparent shift from 68 to 39 kD in focusing gels has been duplicated in denaturing SDS gels by adding ampholyte solutions directly to the protein samples. The membrane orientation of the 39-kD protein and its 68-kD dimer has been assessed by radioiodination in situ using intact cells or purified plasma membranes. Putative monomers and dimers are labeled only when the cytoplasmic side of the membrane is exposed. These results together with trypsin digestion data suggest that the 39-kD protein and its dimer have an asymmetric membrane orientation with a substantial cytoplasmic domain but with no detectable extracellular region. Immunolabeling of sectioned cells indicates that the plasma membrane is the only cellular membrane with significant amounts of 39-kD protein. No major 68- or 39-kD polypeptide bands are evident in SDS acrylamide gels or immunoblots of electrophoresed whole flagella or preparations enriched in flagellar membrane vesicles, nor is there a detectable shift in any flagellar polypeptide in the presence of ampholyte solutions. These findings are considered with respect to the well-known internal crystalline organization of the euglenoid plasma membrane and to the potential for these proteins to serve as anchors for membrane skeletal proteins.

Ultrastructure of four human germ cell tumor-derived cell lines: effect of 12-O-tetradecanoyl phorbol-13-acetate.

Cells derived from human germ cell tumors (HGCT) are potential models for the study of human embryonic differentiation. Differentiation in four HGCT-derived lines (1618K, 833KE, 1777Ndif, 2806B) was examined by light and electron microscopy, after exposure of cells to dimethyl sulfoxide (DMSO) and 12-O-tetradecanoyl phorbol-13-acetate (TPA). In line 1618K, the effect of retinoic acid was also examined. After exposure to TPA, the cells of line 1618K had a loss of microvilli, an increased amount of nuclear heterochromatin, and an increase in the number of cytoplasmic vacuoles. The cells of line 833KE exposed to TPA showed only an increase in cytoplasmic vacuoles. No changes were observed in any cell line after exposure to DMSO. Similarly no changes were observed in 1618K after exposure to retinoic acid. These findings indicate that DMSO and retinoic acid are not effective inducers of differentiation in these HGCT-derived cell lines. The phorbol ester, TPA, appears to induce differentiation of some HGCT-derived cell lines.

Action of interferons in hairy cell leukemia.

We have investigated the direct effects of interferon (IFN) on hairy cells (HCs) isolated from patients with hairy cell leukemia using one- and two-dimensional gel electrophoresis. We have previously characterized the induction of synthesis of 10-16 specific proteins by IFN-alpha 2b in HCs, as analyzed by one-dimensional electrophoresis. By two-dimensional electrophoresis, we have now confirmed this induction and shown that the synthesis of the same number of specific proteins is down-regulated in HCs exposed to IFN-alpha 2b. When compared to HCs, fewer proteins are induced by IFN-alpha 2b in other normal, or neoplastic, lymphoid cells. We also report that protein induction occurs in HCs exposed in vivo to IFN-alpha 2b. We have demonstrated the presence of tubuloreticular structures in the cytoplasm of HCs exposed to IFN-alpha 2b in vitro, using transmission electron microscopy. We now report that these too are seen in HCs exposed to IFN in vivo during therapy. We investigated the effects of IFN-gamma on HCs and found that it also induces specific proteins. The pattern of induced proteins is distinctly different after IFN-gamma exposure in vitro. The fact that such induction occurs suggests that HCs possess also a receptor for IFN-gamma. These results demonstrate that there are direct effects of IFN-alpha on HCs and that such direct effects might be important in the antitumor activity of IFN-alpha in hairy cell leukemia.

Hairy cell leukemia: reevaluation of cell surface features under the scanning electron microscope, using the "GTGO" air drying and critical point drying procedures.

In the present study, the GTGO--air drying (AD) preparatory procedure for scanning electron microscopy (SEM) was used to reevaluate the surface features of hairy cells (HCs) obtained from 18 patients with hairy cell leukemia (HCL). The GTGO-AD procedure, described in earlier studies, involves glutaraldehyde (G) fixation of suspended cells, followed by tannic acid (T)--guanidine hydrochloride (G) treatment of substrate-attached cells, immersion in osmium tetroxide (O), and subsequent air drying (from absolute Freon 113) of dehydrated cells. Both air-dried and critical point dried GTGO-treated cells from patients with lymphocytic, monocytic and hairy cell leukemias, displayed excellent preservation of their surface microvilli and/or ruffles with minimal cell shrinkage. Generally, only two types of hairy cells were identified: (i) cells displaying areas of ruffles alongside areas of clustered microvilli, and (ii) cells showing microvilli scattered among ruffles. Peripheral blood hairy cells showed the same features as those isolated from the spleens involved with HCL and consistently exhibited both ruffles and microvilli. In all studied cases, cells displaying only ruffles or microvilli were not frequently encountered, although ruffled cells with very short and delicate microvilli, and villous cells with small ruffles were seen. Hairy cells, kept in culture for up to 7 days, displayed extreme polarization of their microprojections and very active surfaces, with elongated microvilli and broad-based ruffles. In the light of these results, it is clear that GTGO-AD has much to offer in determining the surface features of circulating and cultured hairy cells and other types of leukemic cells.

In vitro response of cells from three patients with hairy cell leukemia to a recombinant leukocyte interferon.

Mononuclear cells from the peripheral blood of two patients and from the spleen of one patient, all of whom had hairy cell leukemia, were cultured with a recombinant human leukocyte interferon (RD alpha 2-IFN). The IFN was added at concentrations of 10, 100, 1,000, and 10,000 IU/ml, and the cells were cultured for 1, 3, and 7 days. A cytocidal effect of IFN was observed only on cultured cells from the spleen at day 7. Electron-microscopic observations demonstrated that RD alpha 2-IFN induced the formation of tubuloreticular structures (TRSs) and annulate lamellae (ALs) in hairy cells, as well as in co-isolated non-leukemic cells, from all three patients. Ultrastructural examination revealed a close proximity between ALs and TRSs in co-isolated non-leukemic cells. A variability with respect to the induction of TRSs in hairy cells was observed among the three patients. In two of the three patients, the percentage of hairy cells with TRSs increased with the duration of incubation and with the dose of IFN. In the third patient, few hairy cells showed TRSs after 7 days of incubation with IFN. Our findings indicate that leukemic hairy cells may be heterogenous in their response to IFN.

A morphologic and cytochemical study of acute myelomonocytic leukemia with abnormal marrow eosinophils associated with inv(16)(p13q22).

Nineteen patients with acute nonlymphocytic leukemia were found to have a pericentric inversion of chromosome 16. All were classified as having acute myelomonocytic leukemia (M4) according to French-American-British Cooperative Group (FAB) criteria. Consistent abnormalities of marrow eosinophils were present. In 15 of the 19 patients, the number of eosinophils was increased. Morphologic abnormalities of eosinophils were present in all patients, the most striking being the presence of large basophilic staining granules in 18 of the 19 patients. In most cases we also found a population of cells with morphologic features of both eosinophils and monocytes. Eosinophils showed atypical cytochemical reactions with PAS in ten of ten cases studied, with naphthol ASD Chloroacetate in ten of ten cases, and with combined naphthol ASD Chloroacetate esterase- chlorazol fast pink in seven of nine cases tested. The presence of inv(16) defines a distinct subset of ANLL with consistent morphologic and cytochemical features.

Patterns of synaptic contact upon individually labeled large cells of the dorsal lateral geniculate nucleus of the cat.

Four large geniculate cells that were densely filled with horseradish peroxidase have been studied in detail, at first light microscopically, to define the shape and distribution of the dendritic arbors, and then electron microscopically, to define the pattern of synaptic contacts upon their surfaces. Three of the cells had the morphological characteristics of a class 1 cell, and one cell was intermediate between class 1 and 2. All of these cells had few of the characteristic grape-like appendages, thought to receive retinal input within synaptic glomeruli. Retinal afferents were mainly distributed around proximal juxtasomatic dendritic segments and dendritic branch points. Some retinal afferents made simple axodendritic contacts, while others formed a part of a synaptic glomerulus. None of the retinal afferents forming synapses near the perikaryon were involved in glomeruli. Within the glomeruli involving class 1 cells, retinal terminals can relate both to dendritic segments, particularly near branch points, as well as to dendritic appendages. The few singly placed grape-like appendages were always contacted by retinal afferents, but not necessarily in a glomerular arrangement. Terminals interpreted as cortico-geniculate were seen most commonly upon peripheral dendritic segments, while those containing pleomorphic vesicles (F) were distributed more or less evenly over all parts of the dendritic and perikaryal surface. The perikaryon itself was contacted by F-type terminals but was not contacted by retino-geniculate or cortico-geniculate terminals. A class of slender axonal terminal, containing round synaptic vesicles and few or no mitochondria, was found contacting two of the four perikarya, but the origin of these slender axons is unknown. It is concluded that the surfaces of geniculate class 1 cells can be separated into several functionally distinct zones on the basis of the synaptic contacts they receive.

Phagocytic capacity of hairy cells from seventeen patients.

Hairy cells from 17 patients diagnosed as having hairy cell leukemia were evaluated with both light and transmission electron microscopy for their capacity to phagocytose zymosan, latex, Staphylococcus aureus, and Pseudomonas aeruginosa. The hairy cells of fifteen of the sixteen patients which were tested with latex showed significant phagocytosis. Cells of thirteen of the seventeen patients showed phagocytosis of staphylococcus; six of the seventeen, phagocytosis of zymosan; and four of seventeen, phagocytosis of pseudomonas. Of the four substances tested, staphylococcus and latex yielded consistently higher numbers of ingested particles per hairy cell than did zymosan or pseudomonas. Observation of serial sections by transmission electron microscopy revealed that each type of particle is fully enclosed within the cytoplasm of the hairy cell. A Lysostaphin assay performed on the hairy cells of two patients showed that lysis of the phagocytosed particles by the hairy cell differs significantly from that by normal mononuclear cells. The results support our earlier study which suggested that hairy cells have phagocytic capabilities which differ not only among patients, but also within a single patient over time.

Surface immunoglobulin, lectin-induced cap formation, and phagocytic function in five patients with the leukemic phase of hairy cell leukemia.

Hairy cells from 5 patients with greater than 25% hairy cells in the peripheral blood (4 had greater than 45% hairy cells and white blood cell counts (WBC) greater than 10,000/mm3) were studied for surface immunoglobulin (SIg) presence and distribution by two methods, for lectin-induced cap formation, and for phagocytosis of zymosan. Hairy cells from all 5 cases were found to have distinct monoclonal patterns, 2 Gk, 1 MDk, 1 Dk, and 1 GMDk, as well as cap formation with SIg. All 5 cases showed distinct lectin-induced cap formation in a percentage of cells similar to the percentage of hairy cells, and 2 of the 5 patients had hairy cells which phagocytosed zymosan. These findings contrasted with the malignant cells from 4 patients with CLL, which had monoclonal SIg but no SIg cap formation and no significant percentage of lectin-induced cap formation. Cells from 2 cases of T cell lymphomas had no SIg and no lectin-induced cap formation as did cells from 2 cases of non-lymphocytic leukemia. Hairy cells not only appear to have SIg cap formation similar to some B-lymphocytes, which in some patients also phagocytose zymosan, but also demonstrate strong lectin-induced cap formation.

Ribosome-lamella complex in hairy cell leukemia. Ultrastructure and distribution.

Of 19 patients with hairy cell leukemia, an unusual lymphoproliferative disease, nine were found by transmission electrom microscopy to have a cytoplasmic inclusion known as the ribosome-lamellae complex (RLC). The RLC appears to be a cylinder 2.2 to 4.1 micron. in length. The internal diameter appears to be relatively constant at 0.42 micron. The outer diameter, which depends on the number of lamellae as well as on the contents of the intralamellar space, ranges from 0.52 micron to 1.27 micron. The RLC can be comprised of a single lamellae spiraled around the central core, but also RLCs with both two and three lamellae can be defined. The lamellae are composed of double tubule structures 125 A in width set in parallel arrays 125 A apart. A second array of parrallel tubules intersect at a 60 degree-angle creating a lattice. THe distance between the spiraled lamellae is dependent upon the presence or absence of ribosome-like particles. When these particles are present, the distance is usually 406 A, but when they are absent the distance is 172 A. Although experiments with certain fixatives suggest the lamellae may be primarily composed of protein rather than lipid, the exact nature of the RLC and its importance to the leukemic process remain to be determined.

Hairy cell leukemia: differences in phagocytic capacity of cells in vitro.

Hairy cells from eight patients with hairy cell leukemia were evaluated with both light and transmission electron microscopy for their capacity to phagocytose zymosan, latex, staphylococcus aureus, and pseudomonas aeruginosa. In two patients, there was no phagocytosis of any of these substances; cells from three patients phagocytosed only latex; two, all except pseudomonas; and one, all 4 substances. Hairy cells became relatively smooth while in culture with staphylococcus, but no surface changes were noted during incubation with the other substances. Of the eight patients studied, one died of pseudomonas pneumonia and sepsis; pseudomonas was the only substance which her hairy cells did not phagocytose. The one patient whose hairy cells phagocytosed all 4 test substances developed a disseminated Mycobacterium intracellulare infection; culture of his hairy cells with this atypical myocbacterium showed no phagocytosis. Hairy cells have different phagocytic capabilities from patient to patient, and the evaluation of these capabilities in vitro might provide early identification of potential infectious complications.