Nonpackaging and packaging proteins of hnRNA in Drosophila melanogaster

Monoclonal antibodies have previously been raised against chromosomal proteins of Drosophila. Using a biochemical fractionation method for the isolation of large hnRNA-containing structures (hnRNP) of Drosophila tissue culture cells, we show that seven of these antibodies recognize different antigens, and that these antigens are associated with RNA. Analysis of the sedimentation behavior of antigen-containing structures in sucrose gradients reveals that the antigens are differentially distributed with respect both to one another and to pulse-labeled RNA. We demonstrate that the antigens are minor components of hnRNP and are different from the major Drosophila hnRNP packaging proteins, which we have also identified. The antigens are probably involved in the processing of hnRNA in the nucleus.     

hnRNA and its attachment to a nuclear protein matrix

In this study, DNA-depleted nuclear protein matrices are isolated from HeLa S3 cells. These nuclear matrices consist of peripheral laminae, residual nucleoli, and internal fibrillar structures. High molecular weight, heterogeneous nuclear RNA (hnRNA) is quantitatively associated with these structures and can be released intact only by affecting the integrity of the matrices. It is, therefore, concluded that hnRNA is part of a highly organized nuclear structure. By irradiation of intact cells or isolated nuclear matrices with ultraviolet light, proteins tightly associated with hnRNA can be induced to cross-link with the RNA. Performing the cross-linking in vivo is an extra guarantee that only hnRNA-protein (hnRNP) complexes existing in the intact cell are covalently linked. Such hnRNP complexes were isolated and purified under conditions that completely dissociate nonspecific RNA-protein complexes. By comparison of the hnRNP found in nuclear matrices and the published data on the composition of hnRNP particles, it was found that the so-called hnRNP "packaging" proteins (32,000-38,000 mol wt) were not efficiently cross-linked to hnRNA by UV irradiation. They were, however, present in the matrix preparations, bound to hnRNA, because they were released from nuclear matrices after ribonuclease treatment of these structures. On the other hand, two major hnRNPs (41,500 and 43,000 mol wt) were efficiently cross-linked to hnRNA. These proteins were not released by ribonuclease treatment, which suggests that they are involved in the binding of hnRNA to the nuclear matrix.     

Protease activities during preparation and handling of nuclear particles containing hnRNA

Conditions were devised to avoid protease activity during the preparation and the subsequent handling of nuclear particles containing hnRNA. During all the steps of preparation of rat liver particles, the presence of phenylmethyl sulfonyl fluoride (PMSF) was required for the reproducibility of the results. It probably inhibited the cellular serine proteases before the separation of the particles from the other cellular structures.Protease activity was detected in the rat liver particles. The enzyme(s) preferentially hydrolyzed a few particle polypeptides. It was not inhibited by PMSF, nor by two trypsin and chymotrypsin-like protease inhibitors, nor by iodoacetamide, but was inhibited by sodium bisulfite and para-hydroxymercury benzoate (PHMB). PHMB was preferred above bisulfite because it could be used at lower concentration. It proved useful when particles were to be incubated at 37° C.A protease hydrolysing the same polypeptides as the liver enzyme was also detected in rat brain particles. However, its activity was much lower in this tissue and the presence of protease inhibitors was not absolutely required under the standard conditions of preparation and handling of brain particles.     

The expression of a plant genome in hnRNA and mRNA.

Representation of genomic kinetic sequence classes and sequence complexities were investigated in nuclear and polysomal RNA of the higher plant Petroselinum sativum (parsley). Two different methods indicated that most if not all polysomal poly(A) -RNA is transcribed from unique sequences. As measured by saturation hybridization in root callus and young leaves 8.7% and 6.2%, respectively, of unique DNA were transcribed in mRNA corresponding to 13.700 and 10.000 average sized genes. Unique nuclear DNA hybridized with an excess of polysomal poly(A)mRNA to the same extent as with total polysomal RNA. 3H-cDNA - poly(A)mRNA hybridization kinetics revealed the presence of two abundance classes with 9.200 and about 30 different mRNAs in leaves and two abundance classes with 10.500 and 960 different mRNAs in callus cells. The existence of plant poly(A)hnRNA was proven both by its fast kinetics of appearance, its length distribution larger than mRNA, and its sequence complexity a few times that of polysomal RNA.     

Similarity of hnRNA sequences in blastula and pluteus stage sea urchin embryos

We have compared the total single-copy sequences transcribed as nuclear RNA in blastula and pluteus stage embryos of the sea urchin Tripneustes gratilla by hybridization of excess nuclear RNA with purified radioactive single-copy DNA. The kinetics of hybridization of either blastula or pluteus nuclear RNA with single-copy DNA show a single pseudo-first-order reaction with 34% of the single-copy genome. From the rate of the reaction and the purity of the nuclear RNA, it can be estimated that the reacting RNAs are present on the average at a concentration of one molecule per 14 nuclei. A mixture of blastula and pluteus RNA also hybridizes with 34% of the single-copy genome, indicating that the total complexity of RNAs transcribed at both stages is no greater than transcribed at each stage alone. The identity of the sequences transcribed by blastula and pluteus embryos was further examined by fractionation of the labeled DNA into sequences complementary and not complementary to pluteus RNA. This was achieved by hybridization of single-copy DNA to high pluteus RNA Cot, and separation of the hybridized and nonhybridized DNA on hydroxylapatite. Using either the DNA complementary or noncomplementary with pluteus RNA, essentially identical amounts of RNA:DNA hybrids are formed at high RNA Cot with blastula or pluteus RNA. Gross changes in the total RNA sequences transcribed do not appear to be involved in the developmental changes between blastula and pluteus, even though 45% of the mRNA sequences change between these two stages (Galau et al., 1976).     

Species-specific differential cleavage and polyadenylation of plasminogen activator inhibitor type 1 hnRNA.

Plasminogen activator inhibitor type 1 (PAI-1) is the primary physiologic inhibitor of the naturally occurring plasminogen activators. In higher primates two forms of mature PAI-1 mRNA (3.2 kb and 2.2 kb) arise by alternative cleavage and polyadenylation of PAI-1 hnRNA which is regulated in a tissue-specific fashion in humans. In other mammals only the 3.2 kb mRNA has been detected. The putative downstream polyadenylation site in humans that gives rise to the 3.2 kb PAI-1 mRNA consists of three overlapping copies of the consensus polyadenylation sequence while no consensus polyadenylation sequence is found upstream at a position that could generate the shorter mRNA species. To determine whether differential cleavage and polyadenylation of PAI-1 mRNA is due to species-specific differences in trans-acting factors that process PAI-1 mRNA or to the presence of a nonconsensus polyadenylation site acquired recently during primate evolution we prepared plasmids in which the 3' nontranslated region of the human PAI-1 gene or the mouse PAI-1 cDNA was inserted downstream of the neomycin gene in the plasmid pSV2neo. We show that the 3'-nontranslated region of the human PAI-1 gene but not the mouse PAI-1 cDNA conferred alternative cleavage and polyadenylation to the neomycin gene in transfected human Hep G2 cells as well as mouse NIH3T3 and rat L6 cells.     

Heat-shock proteins are associated with hnRNA in Drosophila melanogaster tissue culture cells

Ribonucleoprotein complexes of Drosophila melanogaster Kc tissue culture cells grown at 24°C or heat-shocked at 37°C were cross-linked in vivo by u.v. irradiation. Cross-linked heterogeneous nuclear ribonucleoprotein (hnRNP) complexes were fractionated by oligo(dT)-cellulose chromatography and CsCI density centrifugation. The hnRNP complexes of both 24°C and 37°C culture cells possess buoyant densities in CsCI between = 1.38 g/cm^-3 and 1.43 g/cm^-3. The ³⁵S-labelled proteins bound to the hnRNA of 37°C culture cells correspond in mol. wt. to the so-called heat-shock proteins of 70 K, 68 K, 27 K, 26 K, 23 K and 22 K. The 70 K and 68 K proteins are also present in hnRNP complexes of 24°C culture cells. In addition, several other Drosophila hnRNPs of 140 K, 56 K, 45 K, 43 K, 38 K, 37 K and 34 K, whose synthesis is strongly repressed under heat-shock conditions, could be identified. The results demonstrate that the so-called heat-shock proteins possess a function as RNPs.