Dual enhancement of triple immunofluorescence using two antibodies from the same species

Germinal centre frequency is decreased in pancreatic lymph nodes from individuals with recent-onset type 1 diabetes

AIMS/HYPOTHESIS

Pancreatic lymph nodes (PLNs) are critical sites for the initial interaction between islet autoantigens and autoreactive lymphocytes, but the histology of PLNs in tissue from individuals with type 1 diabetes has not been analysed in detail. The aim of this study was to examine PLN tissue sections from healthy donors compared with those at risk of, or with recent-onset and longer-duration type 1 diabetes.

METHODS

Immunofluorescence staining was used to examine PLN sections from the following donor groups: non-diabetic (n=15), non-diabetic islet autoantibody-positive (n=5), recent-onset (≤1.5 years duration) type 1 diabetes (n=13), and longer-duration type 1 diabetes (n=15). Staining for CD3, CD20 and Ki67 was used to detect primary and secondary (germinal centre-containing) follicles and CD21 and CD35 to detect follicular dendritic cell networks.

RESULTS

The frequency of secondary follicles was lower in the recent-onset type 1 diabetes group compared with the non-diabetic control group. The presence of insulitis (as evidence of ongoing beta cell destruction) and diagnosis of type 1 diabetes at a younger age, however, did not appear to be associated with a lower frequency of secondary follicles. A higher proportion of primary B cell follicles were observed to lack follicular dendritic cell networks in the recent-onset type 1 diabetes group.

CONCLUSIONS/INTERPRETATION

Histological analysis of rare PLNs from individuals with type 1 diabetes suggests a previously unrecognised phenotype comprising decreased primary B cell follicle frequency and fewer follicular dendritic cell networks in recent-onset type 1 diabetes.

Novel immunohistochemical techniques using discrete signal amplification systems for human cutaneous peripheral nerve fiber imaging

Confocal imaging uses immunohistochemical binding of specific antibodies to visualize tissues, but technical obstacles limit more widespread use of this technique in the imaging of peripheral nerve tissue. These obstacles include same-species antibody cross-reactivity and weak fluorescent signals of individual and co-localized antigens. The aims of this study were to develop new immunohistochemical techniques for imaging of peripheral nerve fibers. Three-millimeter punch skin biopsies of healthy individuals were fixed, frozen, and cut into 50-µm sections. Tissues were stained with a variety of antibody combinations with two signal amplification systems, streptavidin-biotin-fluorochrome (sABC) and tyramide-horseradish peroxidase-fluorochrome (TSA), used simultaneously to augment immunohistochemical signals. The combination of the TSA and sABC amplification systems provided the first successful co-localization of sympathetic adrenergic and sympathetic cholinergic nerve fibers in cutaneous human sweat glands and vasomotor and pilomotor systems. Primary antibodies from the same species were amplified individually without cross-reactivity or elevated background interference. The confocal fluorescent signal-to-noise ratio increased, and image clarity improved. These modifications to signal amplification systems have the potential for widespread use in the study of human neural tissues.

Troubleshooting tissue specificity and antibody selection: Procedures in immunohistochemical studies

Optimal antigen detection and identification is dependent on the tissue of interest, the method of fixation, processing, and antibody specificity. We evaluated specific antigens in frozen middle cerebral artery (MCA) sections from rat brains under various conditions of fixation and differing primary and secondary antibody concentrations. Fresh MCAs were frozen, cryosectioned (8 microm), and adhered to chrom-alum coated slides. The effects of different fixation and antigen retrieval/pretreatments were tested for detection of enzymes and receptors involved in MCA tone regulation. Antigen localization was determined with specific primary antibodies and detected using fluorochrome-conjugated secondary antibodies. Spatial distribution of localized antigens was imaged using confocal microscopy. Frozen sections preserved the morphology of the endothelium and/or vessel wall within the tissue in a manner comparable to formalin-fixed sections. Fixation and tissue processing methods were modified based on the primary antibody used. Optimal antigen detection was obtained using fixatives such as 4% paraformaldehyde, 100% acetone or 100% methanol. Pretreatments, such as 1% SDS, enzymatic digestion using 0.1% trypsin, or application of heat were used to optimize antigen-antibody interaction. Stringent background and control checks were performed to ensure specificity of staining in both single and multiple labeling techniques. In a research setting where epitope detection is not used for diagnostic purposes, there is more latitude in tissue fixation. Frozen samples offer a more versatile method of linking the appropriate fixation and tissue processing to the primary antibody's unique needs. At the same time, it stabilizes the tissue in a format that allows for later analysis of multiple antigens with specific detection requirements in same tissue.

Dendritic A-type potassium channel subunit expression in CA1 hippocampal interneurons

Voltage-gated potassium (Kv) channels are important and diverse determinants of neuronal excitability and exhibit specific expression patterns throughout the brain. Among Kv channels, Kv4 channels are major determinants of somatodendritic A-type current and are essential in controlling the amplitude of backpropagating action potentials (BAPs) into neuronal dendrites. BAPs have been well studied in a variety of neurons, and have been recently described in hippocampal and cortical interneurons, a heterogeneous population of GABAergic inhibitory cells that regulate activity of principal cells and neuronal networks. We used well-characterized mouse monoclonal antibodies against the Kv4.3 and potassium channel interacting protein (KChIP) 1 subunits of A-type Kv channels, and antibodies against different interneuron markers in single- and double-label immunohistochemistry experiments to analyze the expression patterns of Kv4.3 and KChIP1 in hippocampal Ammon's horn (CA1) neurons. Immunohistochemistry was performed on 40 mum rat brain sections using nickel-enhanced diaminobenzidine staining or multiple-label immunofluorescence. Our results show that Kv4.3 and KChIP1 component subunits of A-type channels are co-localized in the soma and dendrites of a large number of GABAergic hippocampal interneurons. These subunits co-localize extensively but not completely with markers defining the four major interneuron subpopulations tested (parvalbumin, calbindin, calretinin, and somatostatin). These results suggest that CA1 hippocampal interneurons can be divided in two groups according to the expression of Kv4.3/KChIP1 channel subunits. Antibodies against Kv4.3 and KChIP1 represent an important new tool for identifying a subpopulation of hippocampal interneurons with a unique dendritic A-type channel complement and ability to control BAPs.

Heterogeneity of glutamatergic and GABAergic release machinery in cerebral cortex

We investigated whether cortical glutamatergic and GABAergic release machineries can be differentiated on the basis of the proteins they express, by studying the degree of co-localization of synapsin (SYN) I and II, synaptophysin (SYP) I and II, synaptosomal-associated protein (SNAP)-25 and SNAP-23 in vesicular glutamate transporter (VGLUT) 1-, VGLUT2- and vesicular GABA transporter (VGAT)-positive (+) puncta in the rat cerebral cortex. Co-localization studies showed that SYNI and II were expressed in approximately 90% of VGLUT1+, approximately 30% of VGLUT2+ and 30-50% of VGAT+ puncta; SYPI was expressed in approximately 95% of VGLUT1+, 30% of VGLUT2+, and 45% of VGAT+ puncta; SYPII in approximately 7% of VGLUT1+, 3% of VGLUT2+, and 20% of VGAT+ puncta; SNAP-25 in approximately 94% of VGLUT1+, 5% of VGLUT2+, and 1% of VGAT+ puncta, and SNAP-23 in approximately 3% of VGLUT1+, 86% of VGLUT2+, and 22% of VGAT+ puncta. Since SYPI, which is considered ubiquitous, was expressed in about half of GABAergic axon terminals, we studied its localization electron microscopically and in immunoisolated synaptic vesicles: these studies showed that approximately 30% of axon terminals forming symmetric synapses were SYPI-negative, and that immunoisolated VGAT-positive synaptic vesicles were relatively depleted of SYPI as compared with VGLUT1+ vesicles. Overall, the present investigation shows that in the cerebral cortex of rats distinct presynaptic proteins involved in neurotransmitter release are differentially expressed in GABAergic and in the two major types of glutamatergic axon terminals in the cerebral cortex of rats.

Distribution of TRPC1 receptors in dendrites of rat substantia nigra: a confocal and electron microscopy study

Transient receptor potential channels (TRPC) are plasma membrane, non-selective cationic channels and have been proposed as candidates involved in the regulation of cellular Ca2+ influx. The expression, at mRNA level, of several TRPCs has been demonstrated recently in dopaminergic neurons of the substantia nigra (SN). The aim of the present study was to characterize the expression of TRPC1, at a protein level, in the substantia nigra neurons and non-excitable cells of Wistar rats. Single-label immunohistochemistry and double-label immunofluorescence were used to study the expression of TRPC1 among substantia nigra dopamine neurons and cellular processes using antibodies against tyrosine hydroxylase (TH), substance P (SP), enkephalin, synaptophysin, vesicular glutamate transporter-2 (Vglut-2), microtubule associated protein-2 and metabotropic glutamate receptor 1 (mGluR1). Moreover, the ultrastructural localization of TRPC1 was investigated by means of electron microscopy. A set of dual label experiments was also performed to investigate the presence of TRPC1 among glial cells. Our results showed that TRPC1 is localized mainly in dendritic processes of dopamine neurons, whereas a relatively small percentage of neuronal somata display a light TRPC1 immunoreactivity. Such results were confirmed by our electron microscopy observations. Our study demonstrates, for the first time, a coexpression of TRPC1 and mGluR1 receptors in dendrites of the substantia nigra dopaminergic neurons. Such observation reinforces the concept of an involvement of TRPC1 in mGluR1-mediated excitatory inputs in rat dopamine neurons.

Transient abundance of presenilin 1 fragments/nicastrin complex associated with synaptogenesis during development in rat cerebellum

Immunolocalization and expression of endogenous nicastrin (NCT) and presenilin 1 (PS1) fragments during postnatal development of rat cerebellum were investigated with fragment-specific antibodies. Immunoblotting for NCT revealed the expected mature and immature species, which gradually declined during development. In contrast, the expression of PS1 N-terminal fragment exhibited a peak at postnatal day 14 (P14) and declined thereafter. This chronological change was similarly observed with PS1 C-terminal fragment. Immunoprecipitation of NCT indicated its physical association with PS1 fragments. Colocalization of these molecules to the endoplasmic reticulum in cerebellar Purkinje cells indicates that they are organized into a complex in developing neurons. In addition, active sites of synaptogenesis, the base of the external granular layer and glomeruli, contained PS1 fragments and smaller amount of NCT. Isolated synaptic fraction contained both PS1 and NCT, suggesting their functional association within synapses. Transient abundance of NCT and PS1 fragments as a complex, when (P14) and where synaptogenesis is active, is consistent with intracellular trafficking of this complex in developing neurons and suggests its role as gamma-secretase in synaptogenesis.

Exclusive induction of tau2 epitope in microglia/macrophages in inflammatory lesions-tautwopathy distinct from degenerative tauopathies

Tau2 antibody recognizes a phosphorylation-independent epitope that is pathologically modified as tau protein is phosphorylated to form neurofibrillary tangles of Alzheimer's disease (AD). Similar modification of tau2 epitope can be induced even in the absence phosphorylation of tau, as we first demonstrated in ischemic foci and in glial cytoplasmic inclusions (GCIs) of multiple system atrophy. This modification of tau2 epitope is distinguishable from those observed in degenerative tauopathies because (1) it is a conformational change, which is reversible upon exposure to a detergent; (2) it shows an absence of fibrils composed of phosphorylated tau protein; and (3) it is characterized by the lack of immunohistochemical labeling by anti-tau antibodies other than tau2. In this study, we expanded this observation to inflammatory foci of different pathologies (human immunodeficiency virus encephalopathy, progressive multifocal leukoencephalopathy or multiple sclerosis) by examining formalin-fixed, paraffin-embedded sections immunostained with a panel of anti-tau antibodies. It was found that tau2 was the only anti-tau antibody that immunolabeled microglia/macrophages in these lesions, and this immunoreactivity was reversibly diminished upon exposure to a detergent. Exclusive apparition of tau2 immunoreactivity in these cells without neurofibrillary pathology may be a secondary event shared with ischemic foci and GCIs. It is, however, related to a unique conformational state of tau, possibly grouped under the name of "tautwopathy", that may represent an initial stage of tau deposition distinct from degenerative tauopathies characterized by fibrils composed of phosphorylated tau protein.

Profound cardiac sympathetic denervation occurs in Parkinson disease

In the last few years, cardiac sympathetic dysfunction in Parkinson disease (PD) has been postulated on the basis of decreased cardiac uptake of sympathoneural imaging tracers. However, the pathological substrate for the dysfunction remains to be established. We examined the left ventricular anterior wall from postmortem specimens with immunohistochemical staining for tyrosine hydroxylase (TH), neurofilament (NF) and S-100 protein in PD patients and control subjects, and quantified the immunoreactive areas. As TH-immunoreactive axons nearly disappeared and NF-immunoreactive axons drastically decreased in number, the morphological degeneration of the cardiac sympathetic nerves in PD was confirmed. Quantitative analysis showed that sympathetic nerves were preferentially involved. Triple immunofluorolabeling for NF, TH, and myelin basic protein showed clearly the profound involvement of sympathetic axons in PD. The extent of involvement of the cardiac sympathetic nerves seems likely to be equivalent to that in the central nervous system, including the nigrostriatal dopaminergic system. PD affects the cardiac sympathetic nervous system profoundly as well as nigrostriatal dopaminergic system.