Ribosome Biogenesis

From Structure to Dynamics

Barbara Cisterna; Marco Biggiogera    Laboratory of Cell Biology and Neurobiology, Department of Animal Biology, University of Pavia, Pavia, Italy

1 Introduction

The nucleolus is the most prominent feature of the cell nucleus. When Fontana (1781) noticed it first, he described a specific spot inside the nucleus; however, he ignored that he was observing the only point where one can exactly localize the presence of specific genes without any special techniques.

Ribosomal genes are present in many repeats which are differently amplified in different species (Cmarko et al., 2008) up to a point where it has become generally accepted to assert that two genomes existed, the nuclear genome and the nucleolar one. These two genomes work together and mutually influence each other; as it has become more and more clear in these last years, some nucleolar functions are strictly related to nuclear functions (Martin et al., 2009; Pederson and Tsai, 2009).

Many reviews have been devoted to the nucleolus, both in animal and in plant cells, and this organelle has been studied and observed under many different aspects. In this chapter, we will mainly consider the nucleolus and its activity in view of the dynamics connected to ribosome biogenesis, integrating new concepts into an old story.

2 Nucleolar Structure and Dynamics

2.1 Morphology and cytochemistry

Structure and function are, as usual, strictly related and the nucleolus is no exception.

At light microscopy, it has long been studied and some of its inner structures described; some were more or less corresponding to reality, such as the “nucleolini” described as prominent inside nucleoli and later recognized to be fibrillar centers (FCs; Love and Wildy, 1963). The nucleolar body is clearly identified in the nucleus when stained by a vital fluorescent probe such as SYTO RNASelect (S32703, Molecular Probes, Eugene, USA, Huang et al., 2006), discriminating preferentially RNA (Fig. 2.1).

However, the morphology of the nucleolus has been characterized mainly by electron microscopy (EM).

Since several decades (Marinozzi and Bernhard, 1963) EM has devoted a huge amount of studies to the morphology of this organelle, describing thus its subdivision into the components we know also today. In a typical nucleolus, four major components are present; for an interesting evolutionary study see Thiry and Lafontaine (2005).

Several models have been presented through the years of the arrangements of the nucleolar components (Hozák, 1995; Thiry, 1993). In our model (Biggiogera et al., 2001), the primal one is represented by the chromatin which will become the nucleolus-associated chromatin (NAC; Fig. 2.2). Within this mass of chromatin, the ribosomal genes are contained, and loosen into an area of DNA distributed into a sort of “cloud” or almost spherical structure. These clouds represent the interphase counterpart of the nucleolus organizer region (NOR) present on the mitotic chromosomes. The cloud is composed by ribosomal genes, arranged in tandem repeats and in a conformational state which could better be defined as “almost ready.”

The center of the cloud constitutes the FC, a clear, fibrillar area, less dense than the surrounding nucleolus. Around it, the dense fibrillar component (DFC) is present as a thick layer which encircles (although usually not completely) the FC. The last part is the granular component (GC), which can be found among the different DFC/FC or generally around them in case of the presence of a single DFC/FC complex. Morphology by itself may hint to function, and at least the maturation of ribosomes was correctly attributed to GC even at the beginning of EM studies (Allfrey, 1963; Wang, 1963). However, it was only with the EM cytochemical studies (Bernhard, 1969; Monneron and Bernhard, 1969) that more data could be collected about the function of the different nucleolar components.

Pioneering studies by Derenzini et al. (1983) showed the presence of tufts of DNA emerging from chromatin clumps in the region supposed to be the FC. This DNA was shown, by means of osmium ammine staining (Cogliati and Gautier, 1973), in an elongated form, that is, more ready to be transcribed. Later, the same authors showed that this DNA extended also within the region of the DFC (Mosgöller et al., 1993). The presence of DNA within FC was taken as a marker of the place where transcription occurred (Derenzini et al., 1990). However, different points of view were expressed by several other groups, leading to a decade-long scientific debate. Briefly, the different position can be summarized as follows: (a) transcription takes place in the FC, (b) transcription takes place in the DFC, (c) transcription takes place at the boundary between FC and DFC. The arguments in favor of each of these hypotheses were many, obtained by several cytochemical and immunocytochemical techniques; unfortunately the same could be said for the counter-arguments, often obtained by the same technique on the same cell model (Mosgöller et al., 1993; Thiry, 1993).

Nowadays, the generally accepted idea is that transcription occurs in the DFC, where nascent transcripts are formed and immediately associated with several specific proteins (Brown and Shaw, 2008; Henras et al., 2008) taking care of elongation, splicing, and leading to the formation of the processome (Osheim et al., 2004; Schneider et al., 2007). In this view, the immunoelectron microscope demonstration of the presence of RNA pol I (Scheer and Rose, 1984) in the FC should be reconsidered as indicating a sort of reservoir of the enzyme, in its inactive form. FC, moreover, was found labeled for different proteins, such as AgNOR, UBF, Nopp140, and several others (Casafont et al., 2007; Schwarzacher and Mosgoeller, 2000). Presently, it is difficult to establish whether FCs represent accumulation sites for all these proteins or their role can be linked to this structure.

Most interesting, on the other hand, were the early autoradiographic data on 3H-uridine incorporation (Granboulan and Granboulan, 1965; Lacour and Crawley, 1965). Despite the relatively low resolution of this technique, even after a 30 s pulse, nascent RNA was constantly found within DFC (Fakan and Bernhard, 1971) and this was confirmed by later findings of Br-uridine (Br-UTP) labeling, with a higher resolution (Cmarko et al., 2000).

A possible model was proposed by Biggiogera et al. (2001) which put together FC and DFC in a single unit at which periphery (the DFC) transcription occurred. The model took into account the possibility of reducing or enlarging these structures as they occur, for instance, during the night–day circadian rhythm or after reduction of transcription (Pebusque and Seite, 1981). Several points remain to be clarified, in any case. When nucleoli are isolated and spread in order to show the transcription units, the so-called Christmas trees (Miller and Beatty, 1969), the structures obtained are difficult to reconcile with the in situ nucleolar components (Jordan, 1991; Shaw and Jordan, 1995); Christmas trees are larger, longer, and numerous, so that one must imagine to have several of these units fit into DFC. One problem for all, simply packing the unit into the DFC would create a steric hindrance we cannot, so far, overcome. On the other hand, most of the facts pointing to equivalence between DFC and Christmas trees are there: presence of nascent RNA, proteins involved in transcription, elongation, and splicing, and the density of the structure itself.

The least complicated component, apparently, is the GC. It has long been recognized as formed by ribosome subunits, although only the major one is recognizable on thin...