Journal of Histochemistry & Cytochemistry

Willingham, M C; Bhalla, K

doi: 10.1177/42.4.7907352pmid: 7907352

We used a monoclonal antibody and fluorescence immunocytochemistry to localize the product of the bcl-2 gene in two cultured human carcinoma cell lines, KB and OVCAR-3. These cells show little or no localization of bcl-2 oncoprotein in interphase cells but demonstrate a dramatic appearance of bcl-2 in early prophase or late G2, which persists through-out mitosis, rapidly disappearing at telophase. The pattern of bcl-2 localization shows a diffuse nuclear distribution before chromosome condensation, followed by a specific concentration of bcl-2 at the margins of condensed chromosomes in prophase, metaphase, and anaphase. Treatment of these cells with taxol, an agent that interfers with formation of the mitotic spindle, causes mitotic arrest and apoptosis after a prolonged incubation period. During mitotic arrest due to taxol, bcl-2 remains associated with condensed chromosomes but is lost after > 2 days at approximately the same time as the appearance of apoptotic features in these cells. Western blotting indicates that the only extractable protein reactive with this monoclonal antibody under these conditions is a approximately 28 KD form of bcl-2. These results suggest a model for the role of bcl-2 in protection from apoptosis and a potentially common mechanism by which the final step of apoptosis might be mediated.

Yu, W H

doi: 10.1177/42.4.7510317pmid: 7510317

Nitric oxide synthase (NOS), an enzyme involved in synthesis of nitric oxide (NO), has been localized in many diverse cell types. In the CNS and PNS, discrete neuron cell groups express NOS constitutively. Recent evidence indicates that NOS is inducible in neurons normally not expressing NOS. After transection of peripheral nerves, NOS expression was significantly up-regulated in the axotomized sensory ganglion cells, whereas in the corresponding motor neurons NOS was not induced unless axon regeneration was prevented and ensuing neuron death became massive. Studies on axotomy-induced NOS have been limited largely to spinal nerves, with only one reported in the vagus nerve. The aim of this study was to determine whether NOS induction in motor neurons of the brainstem after axotomy is regulated in a manner similar to that of the spinal cord. By NADPH-diaphorase histochemistry and NOS immunocytochemistry, the status of NOS in neurons of the hypoglossal nucleus, dorsal motor nucleus of the vagus, and motor nucleus of the facial nerve was examined 2 weeks after unilateral transection of the respective cranial nerves, and the results were compared with those of spinal motor neurons after transection of the sciatic nerve. NOS, undetectable in neurons of the three cranial motor nuclei of sham-operated animals, was observed in about 30-50% of neurons in the cranial motor nuclei ipsilateral to axotomy, but it was not detected in spinal motor neurons after axotomy. NOS localized in axotomized cranial motor neurons was unrelated to NOS of macrophages or endothelial cells. There was no appreciable cell loss from axotomy at this period except in the dorsal motor nucleus of the vagus, where some loss was observed. The results indicate that there is a fundamental difference in the regulation of NOS expression between motor neurons of the cranial and spinal nerves. The possible role of NOS/NO acting as cytoprotective or cytotoxic agent on injured motor neurons is discussed. Motor neurons of cranial and spinal nerves may serve as a useful model to further define the roles of NOS/NO in neurons, especially after traumatic injury.

Sondell, B; Thornell, L E; Stigbrand, T; Egelrud, T

doi: 10.1177/42.4.7510318pmid: 7510318

Stratum corneum chymotryptic enzyme (SCCE) is a recently discovered serine proteinase, which has been purified from human plantar stratum corneum. Evidence has been presented that it may play a role in the terminal stages of epidermal turnover, especially in desquamation. Two mouse monoclonal antibodies (MAb) were raised, TE4b and TE9b, that reacted specifically with SCCE in immunoprecipitation, immunoblotting, and gel-exclusion chromatography. When used in immunohistochemical experiments with the peroxidase-anti-peroxidase method, both MAb detected an antigen located in high suprabasal keratinocytes of the epidermis in normal human skin and at the vermilion border of the lip, with maximal staining of the stratum granulosum. In the hair follicles the MAb reacted with the inner root sheet only. In human oral mucosa the MAb stained the high suprabasal epithelial cells of the hard palate. This is a site where the epithelium forms an orthokeratotic stratum corneum. There was no specific staining of the epithelium of the lip mucosa or the buccal mucosa, where the epithelium does not form a stratum corneum under non-pathological conditions. A correlation therefore seems to exist between the presence of SCCE in high suprabasal cells and the ability of the epithelium to form an orthokeratotic cornified layer. We suggest that SCCE is specifically expressed in keratinizing squamous epithelia and that its expression may be part of the terminal differentiation program of this type of epithelium. These results also give further support to the idea that SCCE may play a role in the turnover and/or formation of the stratum corneum.

Sawada, M; Horiguchi, Y; Abujiang, P; Miyake, N; Kitamura, Y; Midorikawa, O; Hiai, H

doi: 10.1177/42.4.8126373pmid: 8126373

Paneth cells are morphologically well characterized but their function has been not elucidated. Previously, we identified and purified a 90 KD zinc-binding protein (ZBPP-1) in rat intestine that was localized to Paneth cell granules, consistent with their high zinc content. To further elucidate the structure and function of ZBPP-1, we immunized Balb/c mice with purified ZBPP-1 and identified four independent monoclonal antibodies (MAb) producing MAb ZIP-1 (IgM), ZIP-2 (IgG1), ZIP-3 (IgM), and ZIP-4 (IgM). Immunohistochemistry (IHC) and immunoelectron microscopy (IEM) with these MAb showed positive staining of Paneth cell cytoplasmic granules. MAb ZBPP-1 also stained a population of mononuclear cells in the lamina propria of digestive tract mucosa and a few cells in spleen, presumably a subset of macrophages. These MAb will provide a useful tool to study the function of Paneth cells in human health and disease, since they cross-reacted with human intestinal Paneth cells and mucosal mononuclear cells.

McCarthy, K J; Abrahamson, D R; Bynum, K R; St John, P L; Couchman, J R

doi: 10.1177/42.4.8126374pmid: 8126374

We have previously reported the production of monoclonal antibodies (MAb) recognizing the core protein of a basement membrane-specific chondroitin sulfate proteoglycan (BM-CSPG). Using immunohistochemical techniques, we have shown that BM-CSPG is present in almost every basement membrane, one exception being the normal glomerular capillary basement membrane (GBM), where it is absent. In the present study of mature kidneys we examined the distribution of BM-CSPG in streptozocin-induced diabetes mellitus in rats. We found BM-CSPG atypically associated with the GBM of diabetic animals as early as 1 month after induction of diabetes mellitus. Immunoelectron microscopy (IEM) of affected capillary loops showed BM-CSPG present in the subendothelial matrix in areas of GBM thickening and absent in areas where the GBM appears to be of normal thickness. Moreover, the association of BM-CSPG with regions of the pericapillary GBM affects the morphology of the capillary endothelial cells within these areas, directly displacing the cell body from the GBM proper and causing loss of fenestrae. These new data on BM-CSPG distribution reflect abnormal glomerular extracellular matrix protein biosynthesis/turnover in diabetes and suggest that BM-CSPG in the GBM might in turn affect normal capillary structure and/or function.

Castells, M T; Madrid, J F; Avilés, M; Martínez-Menárguez, J A; Ballesta, J

doi: 10.1177/42.4.8126375pmid: 8126375

The composition and distribution of sulfo- and sialoglycoconjugates in human laryngeal glands have been investigated at light and electron microscopic levels by use of peroxidase-, digoxigenin-, and colloidal gold-conjugated lectins in combination with several chemical and enzymatic deglycosylation procedures. The present study reveals a variety of terminal oligosaccharide sequences in serous and mucous glands. Serous cells contained glycoconjugates with terminal Neu5Ac (alpha 2-3) Gal (beta 1-4) GlcNAc, Neu5Ac (alpha 2-6) Gal/GalNAc, Neu5Ac (alpha 2-3/6) Gal (beta 1-3 GalNAc, GlcNAc, and Gal (beta 1-4) GlcNAc sequences. Scarce SO4Gal(beta 1-3)GalNAc terminal oligosaccharide chans were detected. Serous cells show wide morphological variability of secretory granules (electron lucent, electron dense, and bizonal) with different lectin affinities. Glycoconjugates in human laryngeal mucous glands contained a variety of terminal oligosaccharide sequences including SO4Gal(beta 1-4)GlcNAc, SO4Gal(beta 1-3) GalNAc, SO4GalNAc, and Neu5Ac(alpha 2-3)GalNAc.

Oprins, A; Geuze, H J; Slot, J W

doi: 10.1177/42.4.8126376pmid: 8126376

Several tissue-processing procedures were studied for their applicability in quantitative immunoelectron microscopy (IEM). Three aspects were mainly considered: maintenance of the natural dimensions of cellular structures (no shrinkage), equal efficiency of immunolabeling throughout a specimen, and the possibility of non-interfering double labeling. These aspects were studied in a gelatin model system and in rat pancreatic tissue, which we subjected to different processing procedures. Some aldehyde-fixed specimens were kept hydrated and prepared for cryosectioning directly or after embedding in polyacrylamide (PAA). Other samples were dehydrated and embedded in different resins, i.e., Lowicryl HM20, LR Gold, or LR White. Dehydration was performed under conditions of cryosubstitution (CS) at -90 degrees C or progressive lowering of the temperature (PLT). We found that only CS dehydration followed by embedding at temperatures below -45 degrees C, which is compatible with Lowicryl HM20, gave satisfactory results in all three aspects investigated. We have previously introduced this procedure for IEM of glycolipids. Unlike in other non-aqueous embedding procedures, aldehyde-fixed material can be embedded via this CS-HM20 procedure without detectable shrinkage. The method also provides homogeneous labeling efficiency by equalizing the accessibility of antigens, irrespective of the original matrix in which they are packed. In this respect the CS-HM20 method equals the previously introduced but more bothersome PAA method. In addition, two-sided labeling of CS-HM20 sections allows double labeling without mutual hindrance of both immunoreactions, and these sections present a well-defined ultrastructure.

Wood, D P; Banks, E R; Humphreys, S; Rangnekar, V M

doi: 10.1177/42.4.7510319pmid: 7510319

Occult micrometastases detected by immunohistochemistry have prognostic significance in patients with localized breast cancer. To determine the usefulness of this technique and of polymerase chain reaction in detecting occult prostate cancer, we evaluated the sensitivity and specificity of immunohistochemistry and polymerase chain reaction amplification of mRNA to detect prostate cancer cells in bone marrow samples. We used cells from an established prostate cancer cell line (LNCAP) mixed with lymphocytes at various dilutions from 10(5) cancer cells in 10(6) lymphocytes to 1:10(6). Both techniques had a 100% specificity and identified cancer cells at all dilutions. Polymerase chain reaction was more sensitive than immunohistochemistry at the lowest dilutions (10(-5) and 10(-6), p = 0.033). We have evaluated seven patients with prostate cancer for micrometastases. Both of the patients with known metastatic prostate cancer and one of the five patients with clinically localized tumors had micrometastases. Detection of micrometastases may be useful in the staging of prostate cancer.

Asari, A; Miyauchi, S; Kuriyama, S; Machida, A; Kohno, K; Uchiyama, Y

doi: 10.1177/42.4.8126377pmid: 8126377

To demonstrate localization of hyaluronic acid (HA) in articular cartilage of the human femur, biotinylated HA-binding region, which specifically binds HA molecules, was applied to the tissue. In sections fixed by 2% paraformaldehyde-2% glutaraldehyde, HA staining was detected in lamina splendens and chondrocytes in the middle zone. By pretreatment with trypsin, intense HA staining appeared in the extracellular matrix of the deep zone and weak staining in the superficial and middle zones. Moreover, pre-treatment with chondroitinase ABC (CHase ABC) intensely enhanced the stainability for HA in the superficial and middle zones and weakly in the deeper zone. Combined pre-treatment of trypsin with CHase ABC abolished intra- and extracellular staining for HA in all zones. By microbiochemical study, the concentrations of HA and dermatan sulfate were high in the middle zone, whereas those of chondroitin sulfate and keratan sulfate were high in the deep zone. These results suggest that HA is abundantly synthesized in and secreted from the chondrocytes, particularly in the middle zone, whereas it is largely masked by proteoglycan constituents in the extracellular matrix.

Kaminskyj, S G; Heath, I B

doi: 10.1177/42.4.7510320pmid: 7510320

We have evaluated protocols for immunofluorescence (IF) staining of the potentially interacting actin filaments (F-actin) and microtubules in hyphae of Saprolegnia ferax, using rhodamine-phalloidin (RP) and freeze-substitution electron microscopy (FSEM), respectively, as standards for their distribution. Saprolegnia has four distinguishable cortical F-actin populations with characteristic organizations and RP- and actin-IF-staining affinities, all of which could be labeled with both probes after some protocols. Other protocols stained only some of the populations. Cortical F-actin was always more reproducibly and sharply stained with RP than IF, indicating that the former is the probe of choice for F-actin in these cells. Although no single IF protocol revealed all of the F-actin and microtubule populations, showing the potential need to optimize protocols for specific antibodies, simultaneous localization was readily achieved by dual labeling with RP and tubulin IF. Tubulin IF patterns differed from FSEM: mitotic spindles were revealed but not the more abundant prophase microtubule arrays, and the cytoplasmic microtubules were subapically displaced and bundled into long cables. These cables, which apparently linked nuclei, indicate a previously undetected involvement in nuclear spacing. The tubulin antibody successfully used for IF failed to recognize any proteins in immunoblots, indicating that immunoblots may not always be a useful indicator of success with IF.