Identification of an epitope in the C terminus of normal prion protein whose expression is modulated by binding events in the N terminus

We have characterized the epitopes of a panel of 12 monoclonal antibodies (Mabs) directed to normal human cellular prion protein (PrP(C)) using ELISA and Western blotting of recombinant PrP or synthetic peptide fragments of PrP. The first group of antibodies, which is represented by Mabs 5B2 and 8B4, reacts with PrP(23-145), indicating that the epitopes for these Mabs are located in the 23 to 145 N-terminal region of human PrP. The second group includes Mabs 1A1, 6H3, 7A9, 8C6, 8H4, 9H7 and 2G8. These antibodies bind to epitopes localized within N-terminally truncated recombinant PrP(90-231). Finally, Mabs 5C3, 2C9 and 7A12 recognize both PrP(23-145) and PrP(90-231), suggesting that the epitopes for this group are located in the region encompassing residues 90 to 145. By Western blotting with PepSpot(TM), only three of Mabs studied (5B2, 8B4 and 2G8) bind to linear epitopes that are present in 13-residue long synthetic peptides corresponding to human PrP fragments. The remaining nine Mabs appear to recognize conformational epitopes. Two N terminus-specific Mabs were found to prevent the binding of the C terminus-specific Mab 6H3. This observation suggests that the unstructured N-terminal region may influence the local conformation within the folded C-terminal domain of prion protein.     

Detection of prion epitopes on PrP and PrP of transmissible spongiform encephalopathies using specific monoclonal antibodies to PrP

Amino acid residues 90-120 of the prion protein (PrP) are likely to be critical for the conversion of PrP(c) to PrP(sc) in the transmissible spongiform encephalopathies. We raised 10 monoclonal antibodies against the 90-120 amino acid region, mapped the epitope specificity of these anti-PrP antibodies, and investigated the expression of epitopes recognized by the antibodies in both PrP(c) and PrP(sc). Four out of five of the anti-PrP antibodies raised in a prion knockout mouse immunized with the linear peptide of PrP90-120 could detect PrP(sc) in 'native' and denatured forms and PrP(c) in normal cells, as well as recognize epitopes within PrP93-112 residues. In contrast, the other six anti-PrP reagents, including five raised from the two knockout mice immunized with conformationally modified PrP90-120 peptide, could detect PrP(c) and recognize epitopes within PrP93-107 residues. Four of these reagents could also detect denatured PrP(sc) on western blots but not PrP(sc) plaques in brain tissue. The results indicate that residues PrP93-102 are exposed in PrP(c) but are buried upon conversion to the PrP(sc) isoform. Furthermore, PrP103-107 residues are partially buried in PrP(sc) while only the PrP107-112 epitope remains exposed, suggesting that the region PrP93-112 undergoes conformational changes during its conversion to PrP(sc).     

Selective re-routing of prion protein to proteasomes and alteration of its vesicular secretion prevent PrP(Sc) formation

Conversion of the cellular prion protein (PrP(C)) into the abnormal scrapie isoform (PrP(Sc)) is the hallmark of prion diseases, which are fatal and transmissible neurodegenerative disorders. ER-retained anti-prion recombinant single-chain Fv fragments have been proved to be an effective tool for inhibition of PrP(C) trafficking to the cell surface and antagonize PrP(Sc) formation and infectivity. In the present study, we have generated the secreted version of 8H4 intrabody (Sec-8H4) in order to compel PrP(C) outside the cells. The stable expression of the Sec-8H4 intrabodies induces proteasome degradation of endogenous prion protein but does not influence its glycosylation profile and maturation. Moreover, we found a dramatic diverting of PrP(C) traffic from its vesicular secretion and, most importantly, a total inhibition of PrP(Sc) accumulation in NGF-differentiated Sec-8H4 PC12 cells. These results confirm that perturbing the intracellular traffic of endogenous PrP(C) is an effective strategy to inhibit PrP(Sc) accumulation and provide convincing evidences for application of intracellular antibodies in prion diseases.     

Toluidine blue-O staining of prion protein deposits

Prion diseases or transmissible spongiform encephalopathies are a group of fatal neurodegenerative diseases caused by an abnormal form of prion protein (PrP(sc)). In this study, we developed a sensitive histochemical detection of PrP(sc) deposits in a Gertsmann-Str?ussler-Scheinker disease (GSS) patient using toluidine blue-O staining, a specific reagent to stain mucins and mucopolysaccharides. Detection of prion deposits correlated with immunohistochemistry using anti-prion antibodies. Control assays were performed using amyloid-beta (Abeta) plaques from Alzheimer's disease (AD) brains. Our results demonstrated that toluidine blue-O staining allowed to recognize 69.1+/-2.6% of the total plaques recognized by the anti-prion antibody. Furthermore, in the 15 studied brain regions from the GSS patient, toluidine blue-O revealed the same recognition pattern as anti-prion labeling. Toluidine blue-O stained specifically the prion deposits but not the Abeta plaques in AD brains. The specificity of the technique was confirmed in a Creutzfeldt-Jakob disease brain. This method opens several possibilities for postmortem diagnoses. Our results also suggest the relevance of specific post-translational modifications of PrP(sc), identified by toluidine blue-O, that might participate in the transformation of PrP(c) to PrP(sc).     

Use of periodate-lysine-paraformaldehyde for the fixation of multiple antigens in human skin biopsies

Periodate-lysine-paraformaldehyde (PLP) has been proposed as a fixative for glycoprotein antigens which should stabilize periodate oxidized polysaccharide chains through lysine mediated crosslinks, either directly or by the intermediation of formaldehyde. In spite of premises and attempts reported in the literature, this fixative has never become popular for the study of membrane antigens of immune system cells, which leads to doubts on its real efficacy. We have addressed this issue in biopsies of human skin and found that PLP followed by cryoprotection with 30% sucrose and cryosectioning, or PLP fixation of isolated epidermal sheets, consistently provided for good preservation of morphology and intense labeling of major histocompatibility complex class II molecules, CD 1 a, CD4, CD8, E-cadherin, cytokeratins in general, cytokeratin-18 in particular, and bromodeoxyuridine, incorporated by cycling cells in vitro, and for the demonstration of tyrosinase enzyme activity. PLP-fixed, osmicated and epon-embedded epidermal sheets proved as good as sheets fixed with a mixture of formaldehyde and glutaraldehyde for electron microscopic morphological analysis. Also, these sheets were amenable to immunoperoxidase staining of Langerhans cell membrane antigen CD1a and keratinocyte membrane antigen E-cadherin before being osmicated and prepared for electron microscopy. In a parallel paper, we had also shown that oral mucosa biopsies fixed in PLP showed good morphology and immunolabeling of CD54, CD80, CD83 and CD86. Therefore, we conclude that PLP can be proposed as a multi-task fixative for light and electron microscopic analysis of membrane, cytoplasmic and nuclear antigens of immune system cells and keratinocytes.     

Application of microwave technology to the processing and immunolabeling of plastic-embedded and cryosections.

We have adapted existing microwave irradiation (MWI) protocols and applied them to the processing and immunoelectron microscopy of both plastic-embedded and frozen sections. Rat livers were fixed by rapid MW irradiation in a mild fixation solution. Fixed liver tissue was either cryosectioned or dehydrated and embedded in Spurr's, Unicryl, or LR White resin. Frozen sections and sections of acrylic-embedded tissue were immunolabeled in the MW oven with an anti-catalase antibody, followed by gold labeling. Controls were processed conventionally at room temperature (RT). The use of MWI greatly shortened the fixation, processing, and immunolabeling times without compromising the quality of ultrastructural preservation and the specificity of labeling. The higher immunogold labeling intensity was achieved after a 15-min incubation of primary antibody and gold markers under discontinued MWI at 37C. Quantification of the immunolabeling for catalase indicated a density increase of up to fourfold in the sections immunolabeled in the MW oven over that of samples immunolabeled at RT. These studies define the general conditions of fixation and immunolabeling for both acrylic resin-embedded material and frozen sections.     

First case of feline spongiform encephalopathy in a captive cheetah born in France: PrP<Superscript>sc</Superscript> analysis in various tissues revealed unexpected targeting of kidney and adrenal gland

Feline spongiform encephalopathy (FSE), affecting domestic and captive feline species, is a prion disease considered to be related to bovine spongiform encephalopathy. Here we report an immunohistological analysis of the first FSE-affected cheetah born in France. The duration of clinical signs, of which ataxia was the main one, was about 8 weeks. The distribution of abnormal prion protein (PrP(sc)) was studied by immunohistochemistry within 27 different tissues. Different antibodies were used to visualise abnormal PrP deposits in situ. PrP(sc )accumulation was detected in the central nervous system (cerebral cortex, cerebellum, brain stem, spinal cord, retina), in peripheral nerves and in lymphoid organs. PrP(sc) deposits were not observed within the enteric nervous system nor in several other organs, such as pancreas, ovary, liver and muscle. More interestingly, unusual PrP(sc )deposits were observed within the zona fasciculata/reticularis of the adrenal gland and within some glomeruli of the kidney raising the question of possible PrP(sc) excretion. The sympathetic innervation of these two organs was visualised and compared to the distribution of PrP(sc) deposits. Our results suggest the possibility that the infectious agent is spread by both haematogenous and nervous pathways.     

The influence of formaldehyde fixation time on the rate of nitrous acid deamination of erythrocytes and spinal cord proteins

Summary Alcohol fixed blood films and fresh blocks of spinal cord were immersed in phosphate buffered neutral 10% formol for graded intervals, the films for 10, 30 min, 1, 2, 4, 8, 24 hr; the blocks for 2, 4, 6, 24 hr at 3 and 24° C; 1, 3, 7, 14, 21, 28, 42, 56 da, 3 and 14 mo at 24–26°. Graded deaminations in 2 N NaNO₂/HAc at 3° C were applied: 1, 2, 5, 10, 20, 30 min; 1, 2, 4, 6, 8, 12, 18, 24, 36 hr. Blood films were stained at pH 6 and 6.5, tissue at pH 4.5 and 5.0, both in azure A eosin B. The point at which erythrocytes reached a slightly bluish green was taken as the end point, since no further color change occurred on further exposure and erythrocytes were the last of usually deamination susceptible tissue elements to lose their oxyphilia on deamination. Deamination of alcohol fixed blood films is completed in about 2 min, of sublimate fixed spinal cord in about 1 hr. Progressive formaldehyde exposure increased deamination time of blood films to 10–20 min in 1 hr, to 6–8 hr in 4 hr and to 12 hr in 24 hr. The tissue deamination showed similar progressive increase of deamination time, slower with 3° C fixation than with 24–26°, reaching 18–36 hr by about 3 days formol, and remaining about the same thereafter.Supported by National Cancer Institute Grant No. C-04816, National Institutes of Health.     

An artificial test substrate for evaluating electron microscopic immunocytochemical labeling reactions

We describe an artificial substrate system for optimization of labeling parameters in electron microscope immunocytochemical studies. The system involves use of blocks of glutaraldehyde-polymerized BSA into which a desired antigen is incorporated by a simple soaking procedure. The resulting antigen-impregnated artificial substrate can then be fixed and embedded identically to a piece of tissue. The BSA substrate can also be dried and then sectioned for immunolabeling with or without chemical fixation and without exposing the antigen to dehydrating agents and embedding resins. The effects of various fixation and embedding procedures can thus be evaluated separately. Other parameters affecting immunocytochemical labeling, such as antibody and conjugate concentration, can also be evaluated. We used this system, along with immunogold labeling, to determine quantitatively the optimal fixation and embedding conditions for labeling of hepatitis B surface antigen (HbsAg), human IgG, and horseradish peroxidase. Using unfixed and unembedded HBsAg, we were able to detect antigen concentrations below 20 micrograms/ml. We have shown that it is not possible to label HBsAg within resin-embedded cells using conventional aldehyde fixation protocols and polyclonal antibodies.     

Electrochemical antigen-retrieval of formaldehyde fixed and paraffin-embeddedarchived leprosy skin biopsies

Summary While formaldehyde fixation preserves tissue morphology, it often hinders immunodetection of antigens in paraffin-embedded tissue because the antigens are masked. Antigen unmasking can be achieved with treatments such as microwave irradiation but they often lead to excessive tissue damage. Therefore, an electrochemical antigen-retrieval method (EAR) was devised in which an alternating electric current is passed through the tissue in a chamber containing an electrolyte buffer. The results obtained with this method were compared to those after microwave irradiation using archived samples of formaldehyde-fixed and paraffin-embedded lepromatous leprosy skin. The efficacy of the two unmasking procedures was assessed by the immunodetectability of several marker antigens using 24 antibodies. Fifteen antibodies that were directed against transmembrane proteins (CD), and the remaining 9 against cytokeratins 18.6 and 19, laminin, vimentin, S100a, BCG,Ulex europaeus lectin, PCNA, and P21^ras. Simple and double immunohistochemistry was performed using the universal ENVISION and LSAB + AP detection systems. After unmasking with the EAR method, immunoreactivity was clearly detected with 22 of the 24 antibodies in single labeling reactions. They include the critical antigens CD3 and CD4 for identifying the T lymphocyte lineages. In contrast, only 20 of the antibodies reacted after microwave irradiation. After double immunolabeling, immunoreactivity was quantitatively similar with both methods. However, the EAR unmasking produced a stronger labeling reaction. Thus, with double labeling immunohistochemistry, EAR made it possible to use higher antibody dilutions and shorter incubation times. Heat damage was also prevented. In conclusion, EAR treatment produces better staining results than microwave irradiation treatment.     

Prion protein biosynthesis in scrapie-infected and uninfected neuroblastoma cells.

Numerous studies have indicated that a modified proteinase K-resistant form of an endogenous brain protein, prion protein (PrP), is associated with scrapie infection in animals. This scrapie-associated PrP modification appears to occur posttranslationally in brain, but its molecular nature is not known. To learn about the normal PrP biosynthesis and whether it is altered by scrapie infection in vitro, we did metabolic labeling experiments with uninfected and scrapie-infected mouse neuroblastoma tissue culture cells. Pulse-chase labeling experiments indicated that, in both cell types, two major PrP precursors of 28 and 33 kilodaltons (kDa) were processed to mature 30- and 35- to 41-kDa forms. Endoglycosidase H, tunicamycin, and phospholipase treatments revealed that the 28- and 33-kDa precursors resulted from the addition of high-mannose glycans to a 25-kDa polypeptide containing a phosphatidylinositol moiety and that maturation of the precursors involved the conversion of the high-mannose glycans to hybrid or complex glycans. Treatments of the live cells with trypsin and phosphatidylinositol-specific phospholipase C indicated that the mature PrP species were expressed solely on the cell surface, where they were anchored by covalent linkage to phosphatidylinositol. Once on the cell surface, the major PrP forms had half-lives of 3 to 6 h. No differences in PrP biosynthesis were observed between the scrapie-infected versus uninfected neuroblastoma cells.     

Sensitive detection of aggregated prion protein via proximity ligation

The DNA assisted solid-phase proximity ligation assay (SP-PLA) provides a unique opportunity to specifically detect prion protein (PrP) aggregates by investigating the collocation of 3 or more copies of the specific protein. We have developed an SP-PLA that can detect PrP aggregates in brain homogenates from infected hamsters even after a 10⁷-fold dilution. In contrast, brain homogenate from uninfected animals did not generate a detectable signal at 100-fold higher concentration. Using either of the 2 monoclonal anti-PrP antibodies, 3F4 and 6H4, we successfully detected low concentrations of aggregated PrP. The presented results provide a proof of concept that this method might be an interesting tool in the development of diagnostic approaches of prion diseases.     

Evaluation of zinc salt based fixatives for preserving antigenic determinants for immunohistochemical demonstration of murine immune system cell markers

Conventional aldehyde based fixatives produce good morphological preservation. However, owing to their cross-linking mechanism of action, epitope loss may occur during fixation compromising the tissue for subsequent immunohistochemical (IHC) analysis. IHC is an important tool for characterizing antigen, cytokine and cytomorphological markers. The increasing use of mouse models for study of pathogenesis has highlighted the need to investigate alternative fixatives. In the study reported here, tissue samples from RIII mice with immune mediated lesions, Mycobacterium bovis infected mice, and uninfected control mice were fixed in either zinc salt fixative or buffered formalin, then tested for IHC using a panel of antibodies (CD3, CD4, CD8, CD45, CD54, F4/80, Interferon-gamma and MIP2). Zinc salt fixation preserved processing-sensitive murine cell markers (CD4, CD8 and CD54) and improved the intensity of immunolabeling for CD45, F4/80 and CD3. Buffered formalin failed to preserve any of the processing-sensitive murine epitopes for demonstration by subsequent IHC.     

Histone H4 acetylation in human cells. Frequency of acetylation at different sites defined by immunolabeling with site-specific antibodies

Histone H4 can be reversibly acetylated at lysine residues 5, 8, 12 and 16. It is possible that acetylation of individual residues will exert specific effects on chromatin function, but this hypothesis is difficult to test with present techniques for analysis of acetylation. To address this problem, we have prepared antibodies which distinguish H4 molecules acetylated at each of the sites used in vivo. By electrophoresis and immunolabeling we have shown that, in H4 from human cells, the four lysine residues are acetylated in a preferred, but not exclusive order, namely lysine 16, followed by 12 and 8, followed by 5.     

MAP2 IHC detection: a marker of antigenicity in CNS tissues

Immunohistochemistry (IHC) is used to detect antibody-specific antigens in tissues; the results depend on the ability of the primary antibodies to bind to their antigens. Therefore, results depend on the quality of preservation of the specimen. Many investigators have overcome the deleterious effects of over-fixation on the binding of primary antibodies to specimen antigens using IHC, but if the specimen is under-fixed or fixation is delayed, false negative results could be obtained despite certified laboratory practices. Microtubule-associated protein 2 (MAP2) is an abundant microtubule-associate protein that participates in the outgrowth of neuronal processes and synaptic plasticity; it is localized primarily in cell bodies and dendrites of neurons. MAP2 immunolabeling has been reported to be absent in areas of the entorhinal cortex and hippocampus of Alzheimer’s disease brains that were co-localized with the dense-core type of amyloid plaques. It was hypothesized that the lack of MAP2 immunolabeling in these structures was due to the degradation of the MAP2 antigen by the neuronal proteases that were released as the neurons lysed leading to the formation of these plaques. Because MAP2 is sensitive to proteolysis, we hypothesized that changes in MAP2 immunolabeling may be correlated with the degree of fixation of central nervous system (CNS) tissues. We detected normal MAP2 immunolabeling in fixed rat brain tissues, but MAP2 immunolabeling was decreased or lost in unfixed and delayed-fixed rat brain tissues. By contrast, two ubiquitous CNS-specific markers, myelin basic protein and glial fibrillary acidic protein, were unaffected by the degree of fixation in the same tissues. Our observations suggest that preservation of various CNS-specific antigens differs with the degree of fixation and that the lack of MAP2 immunolabeling in the rat brain may indicate inadequate tissue fixation. We recommend applying MAP2 IHC for all CNS tissues as a pre-screen to assess the quality of the tissue preservation and to avoid potentially false negative IHC results.     

Double immunohistochemical labeling technique applied to different types of cytokeratins in epithelial proliferations of the breast

Conventional aldehyde based fixatives produce good morphological preservation. However, owing to their cross-linking mechanism of action, epitope loss may occur during fixation compromising the tissue for subsequent immunohistochemical (IHC) analysis. IHC is an important tool for characterizing antigen, cytokine and cytomorphological markers. The increasing use of mouse models for study of pathogenesis has highlighted the need to investigate alternative fixatives. In the study reported here, tissue samples from RIII mice with immune mediated lesions, Mycobacterium bovis infected mice, and uninfected control mice were fixed in either zinc salt fixative or buffered formalin, then tested for IHC using a panel of antibodies (CD3, CD4, CD8, CD45, CD54, F4/80, Interferon-gamma and MIP2). Zinc salt fixation preserved processing-sensitive murine cell markers (CD4, CD8 and CD54) and improved the intensity of immunolabeling for CD45, F4/80 and CD3. Buffered formalin failed to preserve any of the processing-sensitive murine epitopes for demonstration by subsequent IHC.     

Mechanisms of conversion of plasminogen activator inhibitor 1 from a suicide inhibitor to a substrate by monoclonal antibodies

We have delineated two different reaction mechanisms of monoclonal antibodies (mAbs), MA-8H9D4 and either MA-55F4C12 or MA-33H1F7, that convert plasminogen activator inhibitor 1 (PAI-1) to a substrate for tissue (tPA)- and urokinase plasminogen activators. MA-8H9D4 almost completely (98-99%) shifts the reaction to the substrate pathway by preventing disordering of the proteinase active site. MA-8H9D4 does not affect the rate-limiting constants (k(lim)) for the insertion of the reactive center loop cleaved by tPA (3.5 s(-1)) but decreases k(lim) for urokinase plasminogen activator from 25 to 4.0 s(-1). MA-8H9D4 does not cause deacylation of preformed PAI-1/proteinase complexes and probably acts prior to the formation of the final inhibitory complex, interfering with displacement of the acylated serine from the proteinase active site. MA-55F4C12 and MA-33H1F7 (50-80% substrate reaction) do not interfere with initial PAI-1/proteinase complex formation but retard the inhibitory pathway by decreasing k(lim) (>10-fold for tPA). Interaction of two mAbs with the same molecule of PAI-1 has been directly demonstrated for pairs MA-8H9D4/MA-55F4C12 and MA-8H9D4/MA-33H1F7 but not for MA-55F4C12/MA-33H1F7. The strong functional additivity observed for MA-8H9D4 and MA-55F4C12 demonstrates that these mAbs interact independently and affect different steps of the PAI-1 reaction mechanism.     

Immunohistochemical Artifact for Nitrotyrosine in Eosinophils or Eosinophil Containing Tissue

Immunohistochemical artifacts for nitrotyrosine were investigated in eosinophils with regard to fixatives. Immunoreactivity for nitrotyrosine was revealed in separated eosinophils and in gastric mucosa fixed with periodate, lysine-paraformaldehyde (PLP). The increase in immunoreactivity by PLP was due to periodate itself, a component of PLP. Nitrotyrosine formed by peroxidase using NO 2 &#109 and H 2 O 2 or by peroxynitrite was not completely inhibited by 100 mM dithionite but the immunoreactivity for nitrotyrosine antibodies by PLP was completely inhibited by 5.7 mM dithionite. Although untreated eosinophils or ovalbumin (OVA) did not show protein tyrosine nitration in a standard Western blot, the treatment of the blotted membrane with PLP increased the reactivities of proteins from eosinophils with anti-nitrotyrosine antibodies. The increase in immunoreactivity of OVA with anti-nitrotyrosine antibodies by PLP did not change with pre-treatment with dithionite but was abolished by treatment with dithionite after PLP fixation. In HPLC assays, periodate did not generate nitrotyrosine from l -tyrosine and aminotyrosine. These results suggest that the treatment of eosinophils or eosinophil-containing tissues with PLP fixative augments the immunoreactivity of nitrotyrosine antibodies with eosinophils due to the formation of epitopes similar to nitrotyrosine by an oxidation reaction of periodate, which evokes an artifact in nitrotyrosine immunohistochemistry.     

A monoclonal antibody (1D12) defines novel distribution patterns of prion protein (PrP) as granules in nucleus

A monoclonal antibody (mAb) panel to bovine prion protein (PrP) was studied by immunoblotting and immunohistochemistry for scrapie and bovine spongiform encephalopathy. A mAb panel recognized both normal (PrP^C) and abnormal (PrP^Sc) isoforms of PrP in murine, ovine and bovine brain tissues. Interestingly, an anti-bovine PrP mAb, 1D12, prepared by immunizing PrP gene-knockout mice with a synthetic polypeptides corresponding to codons 153-166 of the bovine PrP gene showed novel patterns of reactivity for prion-uninfected neuronal cells. An enzyme-linked immunosorbent assay-mapping of the mAb epitopes resulted in a reaction of monoclonal 1D12 to YEDRY and M corresponding to amino acids 156-160 and 165 of bovine PrP. Several patterns of bovine PrP^C distribution in PrP-deficient neuronal cells (HpL3-4) transfected with bovine PrP were observed after different fixation methods. Stained cell surface was observed after formalin fixation by immunofluorescent assay of 1D12 with confocal microscopy, whereas granules in nucleus were stained after acetone fixation. No reactivity in the nucleus was observed to HpL3-4, or HpL3-4mPrP cells expressing mouse PrP. This is the first paper that has reported the detection of the PrP^C at both cell surface and nuclei of prion-uninfected cell line.     

Spacer-modified saccharides for the regioselective photoaffinity labelling of the binding site of an immunoglobulin.

The spacer-modified trisaccharides that mimic (1----6)-linked beta-D-galactotetraose (Gal4), namely, O-beta-D-galactopyranosyl-(1----6)-S-beta-D-galactopyranosyl-(1----11)-8 -azi- 6,7,8,9,10-pentadeoxy-11-thio-D-galacto-undecose (12) and O-beta-D-galactopyranosyl-(1----6)-O-beta-D-galactopyranosyl- (1----13)-8-azi-6,7,8,9,10,11,12-heptadeoxy-D-galacto-tri decose (20) were synthesised by coupling disaccharide derivatives with 8-azi-6,7,8,9,10-pentadeoxy-1,2:3,4-di-O-isopropylidene-11-O -tosyl-alpha-D-galacto-undecopyranose (10) and 8-azi-6,7,8,9,10,11,12-heptadeoxy-1,2:3,4-di-O-isopropyli den e-alpha-D-galacto- tridecopyranose (17), respectively. Compounds 12 and 20 had affinities for the combining sites of the antibodies IgA X24 and IgA J 539 similar to those of O-beta-D-galactopyranosyl-(1----6)-O-beta-D- galactopyranosyl-(1----11)-8-azi-6,7,8,9,10-pentadeoxy-D-gal acto-undecose (7) and the native ligand Gal4. Tritium-labelled 7 chemically modified the heavy and light chains of IgA J 539, whereas 8-azi-6,7,8,9,10-pentadeoxy-D-(11-3H)galacto-undecose (5a) reacted only with the heavy chain.