Function Of Enzymatic

Enzymatic non-histone proteins

In addition to the structural and regulatory non-histone proteins described in the previous sections, a variety of different enzymes are present in the non-histone protein fraction (Table 3.2). The major components in this category are enzymes catalyzing nucleid acid synthesis and post-synthesis modifications of nucleid acids and proteins. Some enzymes in this group, such as histone kinase, non-histone kinase, histone acetylase, and DNA methylase, may be involved in catalyzing conformational change in chromatin components associated with alterations in gene expression.

 

TABLE 3.2. Summary of enzymatic components of the non-histone chromosomal protein fraction (for detailed references see references 87)

Enzyme category	Function
Nucleid acids as substances DNA polymerases RNA polymerases Nucleases Nucleotides ligase Nucleotides exotransferases     DNA     RNA (poly(A) polymerase) DNA methylase DNA-untwising, unwiding or relaxing enzymes (helix-destabilizing proteins) Chromosomal proteins as substances Protease Asetylase Deacetylase Protein kinases    (Histone)    (Non-histone) Histone methylases Poly(ADP-ribose) synthases Poly(ADP-ribose) glycohydrolase	Polymerization of deoxyribonucleotides into DNA Polymerization of ribonucleotides into RNA Processing or degradation of DNA and/or RNA Joining DNA segments during DNA replication and repair Addition of nucleotides to the ends of nucleic acids Methylation Of DNA Unwind DNA double helix and stabilize the resulting Single stranded DNA Processing or degradation of protein Acetylation of proteins Removal of acetate groups from poteins Phosphorylation of proteins Methylation of histone Addition of ADP-ribose moieties to chromosomal proteins Removal of ADP-ribose moieties from chromosomal proteins

Direct evidence for non-histone proteins as regulatory molecules

Effects of non-histone proteins on gene transcription in chromatin

We have now seen that many of the properties of the non-histone protein fraction are consistent with the nation that some molecules in this fraction are involved in the control of gene transcription. Attempts to provide more direct evidence for this idea have involved experiments in which the effects of non-histone proteins on RNA synthesis catalyzed by bacterial or homologous RNA polymerase have been studied directly in vitro. In the late 1960s and early 1970s many studies were carried out, and it was generally found that one can alter the rate of RNA synthesis by adding, removing, or changing the phosphorylation state of non-histone proteins. Nucleid acid hybridization techniques were employed to show that the type of RNA synthesized is specifically influenced by the tissue of origin of the non-histone proteins. As specific cDNA probes for defined RNA sequences became available they were applied to these systems, yielding data which suggested that specific non-histone proteins are involved in the regulation of globin, ovalbumin, histone, and ribosomal gene transcription. In the case of histone and ribosomal gene transcription, the phosphorylation state of non-histone proteins was also shown to influence the efficiency of gene transcription. (For a detailed discussion and references for the above points, see reference 87).

            The ultimate goal of studies of the type described above has been to isolate and purify individual non-histone proteins and implicate them in the regulation of particular genes. Because of the difficulties associated with studying chromatin transcription in vitro, this goal has rarely, if ever, been achieved. The most progress in this direction has been made studying ribosomal RNA transcription, where a non-histone phosphoprotein of 139.000 daltons has been purified from the slime mould Physarum polycephalum^{5, 50}. This phospoprotein has been shown to stimulate the RNA polymerase I catalyzed synthesis of ribosomal RNA from a homologous nucleolar chromatin template, and to bind selectively to DNA retriction fragments containing the symmetry axis of the palindromic ribosomal DNA. Removal of phosphate groups from the phosphoprotein inhibits both is ability to stimulate ribosomal RNA synthesis, suggesting that the phosphorylation state of this phosphoprotein may be involved in regulating ribosomal gene transcription.

            While data such as those described in the above two paragraphs have been interpreted as being consistent with a possible role of some non-histone proteins in regulating specific gene transcription, extreme cautions must be exercised in interpreting results like those obtained with in vitro transcription systems. Because of the potential for numerous kinds of artifacts in such systems, the following criteria need to be met before one can justify the use of isolated chromatin as a bona fide model which accurately reflects in vivo transcription of mRNA sequences :

1.      The probes used to identify specific RNA sequences synthesized in vitro should be homologous and should accurately reflect the nucleotide sequence of the primary gene transcripts in RNA as well as in DNA excess. It is also desirable to have probes which represent the completely processed polysomal mRNAs.

2.      Conditions for RNA transcription, isolation of transcripts and hybridization analysis should be such the endogenous, chromatin-associated RNAs do not interfere with quantitation of gene transcripts.

3.      Precautions must be taken to eliminate or account for RNA-dependent RNA synthesis when transcription is carried out with E.coli RNA polymerase. It is desirable to transcribe chromatin with the appropriate homologous RNA polymerase, in which case it would be necessary to demonstrate that only the enzyme responsible for transcription in vivo is operative in vitro system.

4.      It should be established that inititation of RNA synthesis is taking place in vitro, not merely elongations of RNA molecules initiated in vivo. 5. It is necessary to establish that initation and termination of transcription in vitro occour at the same sites as they do in vivo.6.      It should be shown that the degree of symmetric or asymmetric transcription of genes from chromatin is the same as that which takes place in intact cells. 7.   In addition to establishing fidelity of gene transcription from chromatin in vitro, it is necessary to show that the availability and lack of availability of other genetic sequences for in vitro transcription from chromatin is the same as in intract cells.

8.      Any stimulatory or inhability effects of chromosomal proteins on DNA transcription in vitro must be observed under conditions, particulary after reaconstitution, where structural properties of the chromosomal protein-DNA complexes reflects those in vivo.

Because the vast majority of experiments have not been carried out under conditions where all the above criteria are satisfied, questions have been raised concering the validity of the conclusion that non-histone proteins are involved in specific gene regulation. It is clear that in spite of the large number of chromatin transcription studies in which non-histone proteins have been shown to affect gene transcription, more rigorously controlled system consisting of well-defined, purified components will ultimately be needed to resolve the issue and to precisely pintpoint the role specific genomic sequences for subsequent use in cell-free transcription studies.

Utilization of cloned genes for in vitro transcription studies

Because of the complexity of the eukaryotic genome, it is difficult to study factors involved in controlling transcription by simply incubating chromatin containing total celluler DNA with purified RNA polymerase. The transcription product of any particular gene represents but a tiny of the total RNA synthesized from such a template, and the situationis further complicated by an inability to define the primary RNA transcript of any particular gene without precise information concering  the in vivo intitation and termination sites. The advent of nucleic acid cloning and sequencing techniques has provided a powerful new weapon for overcoming these obstacels, for with cloned genes and their adjacent sequences as templates for in vitro transcription studies. By creating mutants in which specific regions have been deleted, one can even define with precision the particular DNA base sequences involving in regulating transcription.

           One of the most striking examples of how approach can be used to defined the role of non-histone proteins in gene regulation has been provided by the elegant work of Brown’s laboratory on 5S ribosomal RNA transcription in Xenopus^{8, 79}. By creating deletion mutans in which varying length segments were removed from the 5’ and 3’ flanking regions of cloned 5S genes, they were able to show that region in the centire of the gene (located approximately between nucleotides 50 and 80) is required for directing accurate initation of trancriptions. Thus mutants deleted as far as 50 bases into the gene from the 5’ end still initiate transcription at a precise point 50 bases upstream, producing a fused transcript consisting of both plasmid and 5S RNA sequences. Further studies revealed that when the control region located in the center of the 5S gene is subcloned in the absences of any other 5S DNA sequences, it still directs inititation of transcription at a specific point 50 bases upstream in whatever surrounding DNA it is placed adjacent to.

           Shortly after the above findings were reported, Roeder’s laboratory isolated a protein from Xenopus ovaries which specifically stimulates transcription of cloned 5S genes²¹. This factors was found to specifically bind to a region within the 5S RNA gene located in the same position as the control region defined by deletion analysis. Pelham and Brown⁶⁹ subsequently showed that this non-histone protein factor is identical to an abundant cytoplasmic protein known to complexed to 5SRNA in immature Xenopus oocytes. Thus a negative feedback mechanism seems to be operative in vivo in which the synthesis, producing 5S RNA molecules which in trun bind to the transcription factors and thereby inhibit its ability to further actives 5S RNA transcription.

           The above system is cleary the best understood examples of how a specific non-histone protein can control the transcription of a particular eukaryotic gene. Unfortunately, the 5S gene is atypical in many ways: it is much smaller (120 bases) than typical structural genes, its transcription catalyzed by a different enzyme (RNA polymerase III), and its gene product is synthesized in massive amounts. Hence, the usefulness of the 5S gene as a general model for how non-histone proteins regulate gene transcription in eukaryotes is open to question. On the other hand, the experimental approaches developed in process of unraveling the mechanism involved in controling transcriptions of this gene are certain to from the basis for future investigations on the more complex eukaryotic structural genes.

Conclucing remarks

In this chapter we have attempted to review progress which has been made during the past several years toward elucidating the role of non-histone chromosomal proteins in determining the structure and functional properties of the genome. It is clear that at present, we are only at the threshold of understanding the nature of eukaryotic regulatory macromolecules and their mode of action. With regard to the regulation of gene expression, histone appear to act as non-specific repressors of DNA-dependent RNA synthesis. In contrast, amongst the complex and heterogeneous non-histone chromosomal protein remain to be resolved.

           The specific non-histone chromosomal proteins responsible for rendering particular genetic sequences transcribable generally remain to be identified. It is usually assumed that specific regulatory proteins comprise only a small percentage of the non-histone chromosomal proteins, but experimental evidence to support this assumption is lacking. As an alternative possibility, one could postulate a model for gene regulation in which multiple copies of regulatory proteins are associated with the genome, with only a limited number of these proteins existing in a ‘functional interaction’. An important concept which should be considered is that a single protein may regulate several genes, particulary in situations where celluler events are functionally interrelated or coupled. For example, one may envisage several genes involved with genome replication being controlled by a single regulatory protein. This concept is consistent with our recent identifications of cloned genomic human sequences whose mRNAs, like histone mRNAs, are present on the polysomes only during S phase and are selectively lost from the polysome when DNA synthesis is inhibited^60,74. One possible interprestation of these results is that expression on of functionally related gene is being observed. A similar situation may exits with hormone-stimulated processes. It is not clear whether regulatory proteins should comprise a subset of the non-histone chromosomal proteins with common characteristics such as moleculaer weight, charge of structure. In this regard it will be interesting to establish whether similar types of proteins control ‘single copy genes’, such as globin genes, as opposed to ‘reiterated genes’, such as histone and ribosomal genes.

           A basic questions to be answered is whether activation of genes is generally brought about by newly synthesized non-histone chromosomal proteins or modifications of pre-exiting genome-associated non-histone chromosomal proteins. Alternatively, proteins residing in the cytoplasm or in the nucleoplasm maybe modified in such manner that they become associated with genome and thereby render genes transcribable. Johnson, Karn and Allfrey⁴² have observed that activation of lymphocytes by mitogenic agents result in accumulation of pre-exiting cytoplasmic proteins in the nucleus. A nucleoplasmic pool of non-histone chromosomal proteins has also been reported⁸⁵. These latter two observation are consistent with the possibility that alteration in gene redout involve recruitment of proteins from the cytoplasm or nucleoplasm and their subsequent association with the genome. However, other studies suggest that proteins synthesis is required for activation of transcription in human diploid fibroblast following stimulation proliferate⁷⁷. We have also discussed earlier in this chapter the evidence that presence or absence of phosphate groups on non-histone chromosomal proteins is important in rendering gene transcribable.

           Elcudating rhe manner in which non-histone chromosomal proteins are associated with other genome components should significantly enhance our understanding of the mechanisms by which regulatory proteins interact with defined regions of the genome to render specific genes transcribable. While tenaciously bound as well as readily dissciable non-histone protein fraction containing the regulatory macromolecules, direct experimental evidence to distinguish between these two alternatives is at present limited. It is also presently unclear if non-histone chromosomal proteins interact directly with DNA or with histone-DNA complexes.

           In addition to understanding the mechanism by which non-histone chromosomal proteins active transcription of specific genes, the mechanism by which transcription is turned off must also accounted for one may evisage inactivation of a gene or set of genes via degredation of activator protein or proteins. Protease which utilize non-histone chromosomal proteins as substrates have been shown to be associated with chromatin. An alternative mechanism for inactivation of regulatory proteins may involve acetate and/or phosphate groups added or removed from non-histone chromosomal proteins. Deacetylases as well as phosphatases have been indentified within the nucleus, leading credence to such speculation.

           But perhaps most importantly must be exercised in making unqualified generalization regarding regulation of eukaryotic gene expression and eukaryotic gene regulators. Many of the genes thus far examined in a comprehensive manner, such as globin and ovalbumin genes, reflect a long term commitment of a cell to expression of a differentiated gene product. These gene contain intervening sequences and their transcriptions require a complex series of processing steps (cleavage, splicing and chemical modification) before they are translocated to the cytoplasm as templates for the synthesis of proteins. They types transcriptional control operative in these situations and the nature of their regulators may be distinctively different from that associated with genes such as histone genes, or genes which code for metabolic enzymes, where expression is transtient and acutely responsive to celluler requirements, e.g. DNA replication or substrate leverls. From structural as well as functional standpoints, it may be instructive to bear in mind that at least some such genes do not contain intervening sequences and undergo a comparatively minimal amount of post-transcriptonal processing. It would not be at all unrealistic to consider the possibility that such genes might be organized and regulated in a prokaryotic-tyoe manner.

           As fractionation and characterization of the non-histone chromosomal proteins progress, and additional information regarding the nature of sequences involved in control of specific genes become available, the functional properties of eukaryotic gene regulaions should become more apparent. The recent dramatic advances in moleculer cloning, nucleic acid sequencing, and monoclonal antibody techniques provide us with a powerful new set of weapons which should greatly facilitate future progress in this area.

         

The High Mobility Group (HMG) Proteins

The High Mobility Group (HMG) Proteins

Other than spesific enzymes (to be discussed in the next section), the HMG proteins are the most extensively characterized group of non-histone proteins. They are readly identified on the basis of their high solubility in the mineral acids, their lack of tissue or specificity, and their high concentration in the nucleus^{28, 73, 80, 95}. Four main HMG proteins have been have been characterized in calf thymus, designated HMG1, HMG2, HMG14, HMG17 in order of increasing electrophoretic mobility in acid urea; smiliar proteins have been identified in most other tissue and species examined. One possible exception is trout testes, which has only two major HMG proteins; the H6 protein which has 50% sequence homology with calf HMG17, and HMG-T, which has been composed with calf HMG1 and HMG2^{56, 95}. The HMG proteins are all presents in nucleosomes isolated from chromatin by limited digestion with micrococcal nuclease, but can be completely dissociated from chromatin by the treatment with 0.35 mol/ℓ NaCl. Isolated HMG proteins will be rebind to DNA at low ionic strength and although there is no apperent  sequence specificity in this binding, HMG1 and HMG2 have a preferential affinity for single stranded DNA and the potential to unwind double stranded DNA^{1, 38, 41, 82, 103}.

            Experiments in which transcriptionally active chromatin has been selectively solubilized by various types of nuclease tratment have led to the general genes. This conclusion is not without controversy and apparent contradictions, however probably because HMG protein relase has not always been adequately quantified and it is sometimes difficult to compare HMG proteins from different organisms. Thus preferential relase of HMG1 and HMG2 by selective digestion of active chromatin with Dnase I has been reported for calf thymus, pig thymus, mouse brain, and duck erythrocytes^{53, 72, 93}, but such release has not been confirmed in studies on rabbit thymus and liver²⁹. Likewise HMG-T, a trout testis HMG protein which is supposedly similar to HMG1 and HMG2, is not preferentially released by DNase I⁵⁷. On the other hand pereferentially release of H6 has been reported in trout testis digests^{55, 56}, while HMG17, which is sometimes compared to H6, is not released in some laboratories^{29, 52, 72}, but is released in others⁹⁸. All the HMG proteins have been found associated with 0.1 mol/ℓ NaCl soluble mononucleosomes^{30, 53, 54}, which are reportedly enriched in transcribed DNA sequences.

            The sub-nucleosomal localization of the HMG proteins is difficult to define with precision at present. In trout testis, H6 has been found on nucleosomal core particles in transcribed chromatin, whereas HMG-T appears to be associated with the internucleosomal linker regions immediately adjacent to those core nucleosomes containing proteins H6⁵⁴. Similarly, in other organisms HMG14 and HMG17 appear to be closely associated with core particles^30.62, while HMG1 and HMG2 may be present on linker DNA. The polar nature of HMG14 and HMG17 is believed to permit their binding to both the histone octamer and the nucleosomal DNA in core particles, where  as HMG1 and HMG2 bind DNA, are apparently found as homopolymers in vivo³⁷ and can partake in hydrophobic interactions¹⁷, all of which may point to their presence on linker DNA  where they may either crosslink the chromatin filament by head-to-tail interactions with other HMG proteins or with Histone H1. Alternatively, it hasbeenproposed that they replace histone H1 on the linker DNA, and are thereby involved in the decondensation of transcribed chromatin.

            The role of HMG14 and HMG17 in the structure of transcribed chromatin is less controversial. It has been observed that transcribed genomic sequencesin chromatin lose their preferential sensitivity to DNase I in the absence of HMG 14 and HMG17, but this sensitivity can be restored by reconstitution of HMG14 and HMG17 back to HMG-depleted chromatin^{25, 31, 84, 98, 99 100}. This observation is also true for crude mononucleosomes preparations. Indicating that the HMG proteins sensitive the nucleosomal DNA to DNase I in a linker DNA-independent fashion. Such a conclusion is further supported by the observation that in whole chromatin HMG14 and HMG17 sensitize transcribed sequences in the absence of histone H1 (or H5). The reconstitution studies and the use of affinity chromatography with immobilized HMG14 and HMG17have estabilished that HMG14 and HMG17 selectively bind to nuclosomes containing actively transcribed sequences conferring upon them sensitivy to DNase I digestion. Thus, HMG14 and HMG17 appear to bind to nucleosomes in a sequence-spesific fashion, in spite of the fact that they bind naked DNA with no apparent sequences specificity.

            The biochemical basis for the specific binding of HMG14 and HMG17 to active nucleosomes is not clear at present. Attemps to identify the factors involved have included the use of nucleosomes washed with 0.6 mol/ ℓ NaCl , as well as nucleosomes dissociated  in high salt urea and then reconstituted, but in both cases the spectificity of HMG binding is retained⁹⁹. In addition, no individual type of non-histone protein is present on nucleosomes in sufficient amounts to imply a role in HMG-nucleosomes recognitionsin. In fact, the 0.6 mol/ ℓ NaCl washed nucleosomes have no detectablenon-histone proteins at all. Nucleosomes to which HMG14 and HMG17 bind selectively have been reported to be distinguished by an enrichment in acetylated histones and undermethylated DNA sequences^{97, 99}. While the role of the undermethylane DNA sequences in HMG recognitionin uncertain, the levels of histone acetylation appear to be insufficient in themselves to explain the molar specifity of HMG binding to active nucleosomes. Competition studies have shown that the histone H2A-H2B dimer, but not the H3-H4 dimer, wil, compete with HMG14 and HMG17 for binding sites on active nucleosomes , suggesting that the HMG proteins bind to a site which normally interacts with Histone H2A and H2B, perhaps the H3- H4; DNA complex HMG14 and HMG17 thus appear to be able to recognize a specific nucleosomal conformation which may be induced by a combination of histone and DNA post-synthetic modifications, implying that enzimatyc non-histone proteins present in substoichiometric amounts may be ultimately responsible for determining the structureof chromosomal sequences destined to be transcribed or repressed.