Structural characteristics of bullfrog (Rana catesbeiana) transthyretin and its cDNA--comparison of its pattern of expression during metamorphosis with that of lipocalin.

Transthyretin, an extracellular thyroid-hormone-binding protein (THBP) in higher vertebrates, is synthesized and secreted by the choroid plexus of all classes of vertebrates, except fish and amphibians, and synthesized in the liver of endothermic animals. Here, we report the nucleotide sequence of the cDNA for a THBP found in plasma of bullfrog (Rana catesbeiana) tadpoles before the climax of metamorphosis. The amino acid sequence clearly shows this protein to be an amphibian transthyretin. The three-dimensional structure of bullfrog transthyretin was derived using homology modeling. Compared with transthyretins from other vertebrate species, bullfrog transthyretin is highly conserved at the thyroid hormone-binding sites and other important structural regions of the subunits. Bullfrog transthyretin mRNA was found in tadpole liver, but not in tadpole choroid plexus. Thus, during evolution, synthesis of transthyretin in the liver of metamorphosing amphibians preceded that in the choroid plexus of reptiles, birds and mammals. It was previously observed that the protein most abundantly synthesized and secreted by the choroid plexus in adult amphibians is a lipocalin [Achen, M. G., Harms, P. J., Thomas, T., Richardson, S. J., Wettenhall, R. E. H. & Schreiber, G. (1992) J. Biol. Chem. 267, 23170-23174], in contrast to transthyretin being the most abundantly synthesized and secreted protein in the choroid plexus of mammals, birds and reptiles. Lipocalin mRNA was found in large amounts in tadpole choroid plexus, but not livers.

Murine Stim1 maps to distal chromosome 7 and is not imprinted.

STIM1 (GOK) maps to a region of human Chromosome (Chr) 11p15.5 that is implicated in several embryonal tumors, and some evidence indicates that STIM1 may have a growth suppressor role in rhabdomyosarcoma. In this study we have mapped the murine homolog, Stim1, to the same position as Hbb on distal mouse Chr 7. This region is separated by 20 cM from the region of distal Chr 7 that contains Igf2, H19, and other imprinted genes. Using strain-specific polymorphisms, we have shown that Stim1 is expressed from both parental alleles in fetal and neonatal mouse tissues. Similar analyses of human Wilms' tumor and normal kidney tissues demonstrated biallelic expression of STIM1 in the majority of samples. These data demonstrate that Stim1 expression is not regulated by genomic imprinting in either mouse or human tissues. Thus, if STIM1 is a tumor suppressor at 11p15.5, loss of expression is not due to imprinting effects.

Genomic imprinting in the rat: linkage of Igf2 and H19 genes and opposite parental allele-specific expression during embryogenesis.

Igf2 and H19 are closely linked imprinted genes in both mice and humans that are expressed from opposite parental alleles. In this study we demonstrate that these two genes are also closely linked in the rat, with the H19 gene mapping to within 145 kb of Igf2 on rat chromosome 1. We identified polymorphisms in H19 and Igf2 transcripts in two inbred rat strains and determined the expression of the parental alleles in F1 offspring. The H19 gene was shown to be expressed exclusively from the maternal allele in all fetal and neonatal tissues. Monoallelic expression of Igf2 from the paternal allele was found in all tissues except the leptomeninges and choroid plexus. Igf2 in the choroid plexus was monoallelic at days 13.5 and 15.5 of gestation with a switch to biallelic expression by day 18.5, demonstrating a loss of imprinting after the choroid plexus has differentiated. Biallelic expression of Igf2 was observed in the leptomeninges at all fetal and neonatal stages analyzed. These studies demonstrate conservation of imprinting of two closely linked genes transcribed from opposite parental alleles in a species other than human or mouse. A comparative approach between different species will be important in defining the mechanisms that regulate the tissue-specific expression of imprinted genes.

Myogenesis in cultures of uniparental mouse embryonic stem cells: differing patterns of expression of myogenic regulatory factors.

The expression of myogenic regulatory factors (MRFs) in cultures of androgenetic (AG) and parthenogenetic (PG) embryonic stem (ES) cells were analyzed to identify a role for imprinted genes in the myogenic program. The time course and levels of expression of myf5, myogenin, MyoD1 and myf6 were assessed by semi-quantitative RT-PCR. A more rapid induction of myogenin expression was seen in AG ES cell cultures compared to control D3 ES cells, and myf6 was expressed by AG but not D3 cells. Persistence of myf5 and MyoD1 expression at late stages of AG cell culture suggests that proliferation and differentiation are maintained. Myogenic differentiation was delayed in PG ES cells, but abundant levels of myogenin and myf6transcripts were subsequently observed. Absence of myf5 expression and only low MyoD1 expression at later stages of culture demonstrate a decline in proliferation in PG cultures. Igf2 was induced to high levels in the late phase of both AG and D3 but not PG cell cultures, indicating paternal allele-specific expression. Igf2 expression correlated with expression of MRF genes associated with myoblast proliferation rather than terminal differentiation. H19 was expressed at very low levels in both AG and PG ES cell cultures. The delay in myogenesis in PG cultures suggests that imprinted genes other than Igf2 and H19 play a role at early stages of the myogenic program.

p57K1P2 is expressed in Wilms' tumor with LOH of 11p15.5.

p57KIP2 is a cyclin-dependent kinase inhibitor that maps to human chromosome band 11p15.5, placing it in a genomically imprinted region that has been implicated in the etiology of Wilms' tumor and in the Beckwith-Wiedemann syndrome. Recent analysis of p57KIP2 expression in the mouse has determined that this gene is exclusively expressed from the maternal allele. It has been suggested that p57KIP2 is the WT2 tumor suppressor gene in the 11p15.5 region. We have used reverse transcriptase PCR to determine whether loss of p57KIP2 expression occurs in Wilms' tumor samples that have undergone maternal loss of heterozygosity of 11p15.5. p57KIP2 mRNA was amplified in both the Wilms' tumor tissue and in normal kidney tissue of all five patients analyzed. Semi-quantitative PCR analyses demonstrated that the relative level of p57KIP2 expression in tumor tissue is not markedly different from that in normal kidney. Our data indicate that if the p57KIP2 gene is imprinted in humans and expressed exclusively from the maternal allele, reactivation of the paternal allele has occurred in all five Wilms' tumor samples analyzed in this study. Sequence analysis of the p57KIP2 Cdk inhibitory domain in genomic DNA from primary and secondary tumors from two patients showed only a single base change in one secondary WT, resulting in a predicted methionine to isoleucine substitution at amino acid position 70. These studies suggest that p57KIP2 may not be the WT2 gene.

Deposition of collagen VI in the extracellular matrix during mouse embryogenesis correlates with expression of the alpha 3(VI) subunit gene.

Collagen VI is a microfibrillar component of the extracellular matrix that is predicted to have an important structural role in matrix organization and a biological function in mediating cell-matrix interactions. Secreted collagen VI molecules are composed of three distinct subunits, the alpha 1(VI), alpha 2(VI), and alpha 3(VI) chains. To determine when, and in which tissues, collagen VI is likely to have a role in embryonic processes, we have analyzed the expression patterns of the three subunit chains during postimplantation mouse development by reverse transcriptase-PCR (RT-PCR), in situ hybridization, and immunofluorescence. No collagen VI protein could be detected in the mouse embryo until Day 11.5 of gestation, when low levels were localized within the mesoderm layer of the visceral yolk sac, the subepidermal matrix of branchial arches, and the vessel wall of the dorsal aorta. Levels of collagen VI mRNA and protein increased during the period from Days 12.5 to 14.5 in the visceral yolk sac, subepidermal mesenchyme, lung, gut, meninges, muscle, perichondrium, and vertebral column. The cartilage matrix of ribs and developing long bones was not stained with collagen VI antisera, but pericellular staining of chondrocytes was seen in both tissues. Low levels of collagen VI mRNA and protein were seen in the fetal liver except for the connective tissue of the liver capsule, which was highly stained. Collagen VI was first detected at significant levels in the developing heart on Day 14.5. These data demonstrate a tissue-specific onset of collagen VI synthesis and deposition in the extracellular matrix of developing mouse embryos at a much later stage of development than that reported for fibronectin or collagen I. Sensitive RT-PCR assays showed that alpha 1(VI) and alpha 2(VI) mRNAs were amplified from extracts of embryonic tissues as early as Day 7.5, while alpha 3(VI) mRNA was not detected until Day 10.5. Expression of the alpha 3(VI) gene immediately preceded the appearance of collagen VI protein in embryonic tissues. This correlation is consistent with the proposal that expression of alpha 3(VI) chains regulates the formation and secretion of collagen VI trimers and collagen VI matrix deposition during development.

Expression of H19 and Igf2 genes in uniparental mouse ES cells during in vitro and in vivo differentiation.

Genomic imprinting is a process that results in the differential expression of genes according to their parental inheritance. Two imprinted genes, insulin-like growth factor 2 (Igf2) and H19 are closely linked on mouse chromosome 7, and are expressed from the paternal and maternal alleles, respectively. The genes show striking similarity in their tissue-specific expression patterns, which led to the proposal that their transcription is controlled by a common regulatory domain that enables only one gene to be active from each chromosome. Evidence is accumulating, however, that the expression of H19 and Igf2 genes is not always from their respective maternal and paternal alleles. This suggests that their expression is regulated independently of imprinting in some tissues and teratomas. We have analysed the extent of non-imprinted expression of H19 and Igf2 in uniparental mouse embryonic stem (ES) cells during in vitro differentiation, and differentiation in teratomas using Northern blot and in situ hybridisation analysis. The expression patterns observed indicate that both imprinting and non-imprinting mechanisms regulate transcription of these genes. Expression of one or the other gene was observed in certain cell types in differentiated cultures and in teratomas, suggesting that imprinting regulates the expression of H19 and Igf2 genes in some differentiating cell lineages. At the same time, in other subpopulations of cells, co-expression of both genes was observed, demonstrating that the expression of these genes is not always regulated by genomic imprinting.

Structure-function analysis of human IFN-alpha. Mapping of a conformational epitope by homologue scanning.

The purpose of this study was to map a conformational epitope of a mAb that binds the IFN subtype alpha 4a. Binding of this mAb, designated I-4-A, to IFN-alpha 4a does not block receptor binding, but does neutralize biologic activity by inhibition of signal transduction. A novel strategy was developed, termed homologue scanning, which uses template-coupled polymerase chain reaction to generate hybrid molecules consisting of part (N-terminus) of the reactive IFN-alpha 4a subtype to locate the epitope, and the remainder of the nonreactive IFN-alpha 14 subtype to provide the overall conformation of an IFN-alpha molecule. Hybrid molecules were expressed as 35S-methionine-labeled proteins and tested for immunoreactivity by Western blotting and antiviral activity by cytopathic effect reduction bioassays. Unless an entire IFN-alpha (hybrid) molecule was formed, immunoreactivity and biologic activity were lost, indicating the importance of the C-terminus for correct folding of IFN-alpha molecules. The epitope for I-4-A was localized to the N-terminal 23 residues of IFN-alpha 4a. Furthermore, the immunoreactivity of IFN-alpha 4a analogues, with alterations in the putative receptor-binding region of IFN-alpha 4a residues 30 to 40 was unaffected, in contrast to the biologic activity that was reduced by several orders of magnitude. Thus, the N-terminal 23 residues of IFN-alpha 4a, which probably are not involved in receptor binding, may be important for other interactions of the receptor-bound ligand. In general terms, the novel approach of homologue scanning, using template-coupled PCR to facilitate the generation of hybrid proteins, will have broad application in the mapping of conformational epitopes of proteins that are members of a homologous family. The ability to identify conformational epitopes will increase our understanding important interactions of proteins with antibodies, receptors, and other macromolecules.

Different interactions of interferon-alpha subtypes at the surface of epithelial and lymphoid cells.

The interaction of different interferon (IFN)-alpha subtypes with different cell types was investigated using a unique monoclonal antibody (MAb), I-4-A. This MAb reacts in immunoassays equally with IFN-alpha 2b and IFN-alpha 4a, but does not inhibit the binding of IFN to cell receptors. 125I-labeled I-4-A reacted with IFN-alpha 4a and IFN-alpha 2b bound to receptors on Daudi cells. However, in a "double assay" developed using Daudi cells to measure antiviral and antiproliferative activity, I-4-A neutralized both activities of IFN-alpha 4a, but neither of IFN-alpha 2b. Similarly, in studies on the activation of natural killer (NK) cells, I-4-A neutralized the effect of IFN-alpha 4a but not that of IFN-alpha 2b. In contrast, when cell lines other than lymphoid were studied, e.g., HEp 2 and WISH cells, I-4-A neutralized the antiviral activity of both IFN-alpha subtypes. The neutralization of one IFN-alpha subtype but not another on lymphoid cells suggests a difference either in the receptor-bound form of the subtypes, or in subsequent interactions prerequisite for activation of these cells. Furthermore, the neutralization of a particular IFN subtype, alpha 2b, on epithelial-derived but not lymphoid cells suggests differences in the IFN-receptor complex or the mechanisms of cell activation between these cell types. An implication from these studies is that some IFN-alpha subtypes can exert different functions on lymphoid and epithelial cells.

Functional analysis of interferon-alpha subtypes using monoclonal antibodies to interferon-alpha 4a--subtype reactivity, neutralisation of biological activities and epitope analysis.

A panel of monoclonal antibodies (mAb) to a major human interferon-alpha (IFN-alpha) subtype, -alpha 4a, have been produced, characterised and used for studies of structure/function relationships of IFN-alpha subtypes. The mAb were tested for effects on receptor binding of IFN-alpha 4a, reactivity with other major subtypes -alpha 1, -alpha 2b and -alpha 14 by competitive ELISA and western immunoblotting, and for neutralisation of antiviral and antiproliferative activities of the four subtypes. The mAb could be grouped according to reactivity with IFN-alpha subtypes, group I (designated I-4-A) reacted with -alpha 4a and -alpha 2b, group II (I-4-C and I-4-F) reacted with -alpha 4a and -alpha 1, group III (I-4-D), I-4-G and I-4-H) reacted with -alpha 4a only, whereas group IV (I-4-I) reacted with -alpha 4a, -alpha 1 and -alpha 2b. No mAb reacted with IFN-alpha 14. Sequence comparisons of reactive and non-reactive IFN-alpha subtypes, and reactivity patterns with IFN-alpha fragments obtained by Lys-C digestion indicated that the epitopes were located in the N-terminal region (group I), in two regions of the middle of the molecule (group III and IV) and in the C-terminal region (group II). Binding of mAb to any of these four distinct epitopes neutralised the biological activities of IFN-alpha 4a, and in all cases, except I-4-A, inhibited receptor binding. Only the group III mAb bind to an epitope proposed to be in the vicinity of residues 30-40 which are implicated, from in vitro mutagenesis studies, in receptor binding. Binding of mAb to the other 3 epitopes neutralises biological activities by indirect mechanisms. These results emphasise the antigenic diversity between highly homologous IFN-alpha subtypes, which may have a wider functional significance. Individual mAb will have practical applications in the purification and detection of several IFN-alpha subtypes and so facilitate their further characterisation. By virtue of their different mechanisms of neutralisation, this panel of mAb will be useful in further studies of receptor interaction and signal transduction by IFN-alpha, and illustrate principles which are relevant to immunochemical studies of the receptor interactions of other cytokines.

Selective production of interferon-alpha subtypes by cultured peripheral blood mononuclear cells and lymphoblastoid cell lines.

The biological significance of the existence of multiple interferon-alpha (IFN-alpha) subtypes is unknown but may represent a finely tuned mechanism whereby different subtypes are produced in response to different stimuli. To investigate the expression of individual IFN-alpha subtypes, polyclonal antipeptide antisera designed to react with all IFN-alpha subtypes, or with a particular subtype, IFN-alpha 2 or IFN-alpha 4, have been produced. In this study we demonstrate the utility of these antisera for the detection, using indirect immunofluorescence staining, of intracellular IFN-alpha produced by human peripheral blood mononuclear cells (PBMC) and lymphoblastoid cells. Secreted IFN-alpha was also investigated by bioassay and a sandwich radioimmunoassay (RIA), using two monoclonal antibodies (mAb) and specific for IFN-alpha 4. The PBMC were shown to produce IFN reactive with all three polyclonal antisera, after stimulation with Sendai virus. The lymphoblastoid cells also produced IFN, including IFN-alpha 2, but IFN-alpha 4 was not detected either intracellularly, by immunofluorescence, or in the medium, by sandwich RIA. The immunofluorescence studies also demonstrate that in the absence of viral stimulation IFN-alpha is found in the cytoplasm of PBMC and lymphoblastoid cells but not secreted in detectable levels. The finding that two lymphoblastoid cell lines do not produce the subtype IFN-alpha 4 raises important questions as to whether other cell lines and cell types produce IFN-alpha subtypes selectively, and whether individual IFN-alpha subtypes have different roles in human physiology and pathology.

Subtype-specificity of antipeptide antibodies raised against unique sequences of human interferons-alpha.

A strategy for the production of human interferon-alpha (IFN-alpha) subtype-specific antibodies, based on immunizing rabbits with short unique synthetic peptides coupled to protein carriers, has been validated. These peptides correspond to amino acid residues 99-111 of IFN-alpha 1, 50-57 and 103-116 of IFN-alpha 2, and 37-50 of IFN-alpha 4. The antipeptide antibodies [anti-IFN alpha 1(99-111), anti-IFN alpha 2(50-57C), anti-IFN alpha 2(103-116) and anti-IFN alpha 4(C37-50)] were tested by ELISA and Western blotting for their reactivity with immunoaffinity-purified recombinant human IFN-alpha 1, -alpha 2b and -alpha 4a. The anti-IFN alpha 1(99-111) and anti-IFN alpha 2(50-57C) reacted with their corresponding IFN-alpha and did not crossreact with the other IFN subtypes. The anti-IFN alpha 2(103-116) reacted with IFN-alpha 2b and also crossreacted slightly with the other subtypes. The anti-IFN alpha 4(C37-50) reacted well with IFN-alpha 4a, crossreacted with significantly lower affinity with IFN-alpha 1 and did not bind IFN-alpha 2b. Residues 104-107 and 108-111 are the major components of the epitopes recognized by anti-IFN alpha 1(99-111) and anti-IFN alpha 2(103-116), respectively, as determined by ELISA against overlapping octapeptides.

Comparison of different ELISAs for the detection of monoclonal antibodies to human interferon-alpha. Implications for antibody screening.

Three enzyme-linked immunosorbent assays (ELISA) were compared in the initial screening of some 400 hybridoma supernatants for antibodies to a recombinant human interferon-alpha subtype, 4a (IFN-alpha 4a). In these assays, (i) the antigen was coated directly to polystyrene microtitre plates (ELISA-PS), (ii) the antigen was coated directly to nitrocellulose (ELISA-NC), or (iii) the antigen and mouse antibody were reacted in solution and the resulting complex immobilized to a solid support precoated with polyclonal rabbit anti-IFN-alpha antibody (ELISA-SW). The ELISA-PS detected eight antibodies, the ELISA-NC 15 and the ELISA-SW 18. The interferon specificity of the MAbs detected by each of the ELISAs was confirmed by neutralization of IFN-alpha antiviral activity and Western immunoblotting analysis. The results suggest that in ELISAs, the presentation of an antigen and its recognition by antibodies is substantially influenced by the method used in the immobilization of antigen and the type of solid support used. The ELISA-SW proved optimal for screening hybridoma supernatants for antibodies to IFN-alpha 4a, and is recommended for screening for antibodies to other soluble antigens.

Histology of spheroidal degeneration of the cornea in Labrador.

Nine specimens of the corneas of patients from Labrador and Northern Newfoundland affected by spheroidal degeneration (climatic droplet keratopathy) have been examined microscopically. Histochemical stains confirmed studies of similar corneal degenerations from other geographical areas that the droplets contain a protein which does not have all the characteristic properties of elastic tissue. Staining compatible in some instances with fibrin and "fibrinoid" was found. By immunoperoxidase techniques the droplets were located in the zones of greatest concentration of various plasma constituents, especially albumin and immunoglobulins G and A. Reasons are given why the abnormal deposits are not thought to be derived directly from corneal collagen. It is suggested that some of the plasma proteins, which are known to be diffusing through the cornea from the limbal vessels under normal conditions are acted upon by the ultraviolet light reflected from the ice of Labrador and degraded so that they accumulate in the superficial stroma.