Tissue-polypeptide-specific antigen (TPS) from the human colon adenocarcinoma cell line WiDr was purified using the monoclonal antibody M3 as a probe. Upon SDS/polyacrylamide gradient gel electrophoresis, several TPS-positive bands were detected (corresponding to 13 kDa, 22 kDa and a doublet at 42 kDa). The 13-kDa moiety was purified about 30 000-fold by a 5-step protocol. The electro-phoretically homogeneous component was obtained in a 7% yield of the total TPS activity of the crude extract. Nterminal sequence analysis showed the presence of an N-terminally truncated molecule and identified the 13-kDa TPS component as a fragment of human cytokeratin 18, with a major form starting at position 284 of the parent molecule. Laser-desorption mass spectrometry showed the presence of one major component with a molecular mass corresponding to a C-terminal end close to position 396 (which gives 12776Da for the form with non-truncated N-terminus). The M3 antibody was also used to screen a human prostate cDNA I g t l l library. Four identical phage clones were detected, each producing a fusion protein with P-galactosidase and the M3-positive component. PCR amplification showed the presence of an approximately 1200-bp insert, and sequence analysis revealed it to contain a 996-nucleotide fragment corresponding to residues 103-429 of human cytokeratin 18 (plus a non-coding human desmin artifact fragment). Smaller fragments, engineered by PCR and expressed as fusion proteins using the pET3xc vector in Escherichia coli, showed that the M3 epitope is localized to cytokeratin 18, residues 322-340. Two other TPS-active monoclonal antibodies were localized to cytokeratin 18 with similar techniques, ascribing an epitope (to M21) to residues 414-429 and another (to M24) to residues 139-297. Combined, the results demonstrate that TPS reactivity is derived from specific epitopes of human cytokeratin 18.

Screening for anti-cancer substances is commonly conducted using viability assays. An inherent problem with this approach is that all compounds that are toxic and growth inhibitory, irrespective of mechanism of action, will score positive. It would be beneficial to be able to screen for compounds that specifically induce apoptosis. We here describe an ELISA-assay based on a monoclonal antibody (M30) which recognizes a neo-epitope on cytokeratin 18 exposed after cleavage by caspases during apoptosis. We show that this assay detects apoptosis in epithelial cells and that the sensitivity is sufficient for screening in the 96-well format. We used the M30-ELISA assay to screen 500 low molecular weight compounds from a chemical library from the National Cancer Institute and identified 16 drugs with strong pro-apoptotic activity, suggesting that the assay is a useful tool for discovery of pro-apoptotic drugs.

The distribution of monodisperse high molecular weight RNA (38, 30, 28, 23, and 18S RNA) was studied in the salivary gland cells of Chironomus tentans. RNA labeled in vitro and in vivo with tritiated cytidine and uridine was isolated from microdissected nucleoli, chromosomes, nuclear sap, and cytoplasm and analyzed by electrophoresis on agarose-acrylamide composite gels . As shown earlier, the nucleoli contain labeled 38, 30, and 23S RNA . In the chromosomes, labeled 18S RNA was found in addition to the 30 and 23S RNA previously reported . The nuclear sap contains labeled 30 and 18S RNA, and the cytoplasm labeled 28 and 18S RNA . On the basis of the present and earlier analyses, it was concluded that the chromosomal monodisperse high molecular weight RNA fractions (a) show a genuine chromosomal localization and are not due to unspecific contamination, (b) are not artefacts caused by in vitro conditions, but are present also in vivo, and (c) are very likely related to nucleolar and cytoplasmic (pre)ribosomal RNA . The 30 and 23S RNA components are likely to be precursors to 28 and 18S ribosomal RNA . The order of appearance of the monodisperse high molecular weight RNA fractions in the nucleus is in turn and order : (a) nucleolus, (b) chromosomes, and (c) nuclear sap . Since both 23 and 18S RNA are present in the chromosomes, the conversion to 18S RNA may take place there . On the other hand, 30S RNA is only found in the nucleus while 28S RNA can only be detected in the cytoplasm, suggesting that this conversion takes place in connection with the exit of the molecule from the nucleus .

Balbiani rings (BR), giant puffs in Chironomus larval salivary glands, code for giant secretory proteins. As shown earlier, the normally dominant BR2 is turned off with its putative translation product during exposure of larvae to compounds that diminish the stores of Pi. A BR6 develops from a compact chromosome band, and a new giant protein appears in the secretion as the major component. We have determined the sequence of cloned DNA fragments representative for large parts of BR1 and BR2 (normally active) and the inducible BR6. There is an excess of positive charges and high contents of serine/threonine in the coded amino acid composition for the BR1 and BR2 sequences. The coded amino acid sequence for the BR6 clone shares homologies with the others but has an excess of negative charges and lacks serine/threonine. This suggested that the Pi effects observed earlier could be related to differences in phosphorylation between the normal proteins and the BR6 product. This could be confirmed by measurements of phosphorylation, which occurs in the normal giant proteins mainly at seryl residues. P export with giant secretory protein is normally quantitatively important. Thus, BR6 activation should decrease P loss when Pi pools are lowered because of inducer action.Balbiani rings (BR) in polytene chromosomes of salivary glands from Chironomus larvae are genetic determinants of secretory proteins (1,2). The large BRs (BR1 and BR2) produce giant messengers (3, 4) and code for giant secretory polypeptides, sp-Ia and sp-Ib (5, 6). If larvae are raised in solutions with monosaccharides or some other compounds, including glycerol, a BR6 develops from a condensed band and BR2 regresses (7,8). Simultaneously a new giant protein appears (sp-Ic), immunologically different from sp-Ia and spIb, and the labeling of sp-Ib decreases (8). A new giant RNA species also shows up, similar in size to the RNA from BR1 and BR2 (75S RNA) (9). The sp-Ic may account for 25% to more than 50% of the total gland incorporation of methionine (ref. 8 and this paper). Pi decreases in the hemolymph in response to inducers, and addition of Pi to the inducing solution abolishes the action. The induction also can be reversed with Pi (9).In order to relate the Pi effects to the properties of the genes, we have cloned short fragments of BR1, BR2, and BR6 representing considerable parts of the genes (15-20 kb). The base sequences of these fragments suggested that phosphorylation might occur in the normal polypeptides, sp-Ia and sp-Ib, but not in sp-Ic. Direct analyses confirmed that suggestion. Because the P content in sp-Ia and sp-Ib is high and secretory protein export is quantitatively important, the switch to sp-Ic (BR6) implies that P is saved when, because of inducers, the Pi pool is low. MATERIALS AND METHODSAnimals. Mid-fourth-larval instar Chironomus pallidivittatus animals were used, raised as described for Chironomus tentans (10). For squashes, animals from single egg-string cultures were used; for other experiments, animals from mass cu...

The 75S RNA originating in the large Balbiani rings 1 and 2 (BRI and 2) was isolated and used for in vitro translation in the mRNA dependent reticulocyte lystate. Conditions (K+-concentration, temperature, time etc). were optimized for obtaining translation products of maximal size. Polypeptide chains up to about 500,000 D were obtained but no complete translation products. Trypic fingerprints were performed on the in vitro products as well as on the secretory protein components nos. I and II+III labelled with 35S-methionine. There was a large degree of correspondence between the fingerprint of the in vitro product and that of component I but less to that of component II+III. The results suggest that 75S RNA with an origin in the BR1 and BR2 codes for the giant secretory protein component I.