“Compact” nuclear domains: Reconsidering the nucleolus

Abstract

The nucleolus is among the most striking cell components when viewed by phase contrast microscopy and yet, recent studies have revealed it to possess a lower molecular density than anticipated. A new report demonstrates that the nucleolus accommodates the assembly of a virus, this amidst building ribosomes and performing other host cell functions.

As most readers of Nucleus know, the nucleolus has always been considered to be a compact nuclear body, this viewpoint emerging from its strong presentation in phase contrast microscopy, due to a substantial difference between its index of diffraction and that of euchromatin and the nucleoplasm. The notion that the nucleolus is “dense” was reinforced by its avidity for dyes that are taken by RNA, or by certain silver stains that bind to specific nucleolar proteins, giving rise to lovely slide preparations in which the nucleolus boldly stands out- yet it is obviously the composition, not the density, of the nucleolus that is at play there.

A study by Joseph Gall and colleagues examined the density of nucleoli in the germinal vesicle of Xenopus oocytes by interference microscopy, a mode of optical microscopy that can be used to estimate the relative molecular densities of objects.¹ The results were surprising: the nucleoli were only twice the density of the nucleoplasm. This finding was corroborated by their results based on the penetrations of fluorescent dextrans of various molecular masses. This key study recalibrated thinking on how “compact” the nucleolus really is. A subsequent study has confirmed these results in the mammalian cell nucleus.²

There has now appeared a study in the field of virology that resonates with these findings.

Adenovirus-associated virus (AAV) is a ∼25 nm icosahedral, non-enveloped virus containing a single-stranded linear DNA genome of ∼4.7 kb. This minute size places it among the Parvoviridae but its particular feature of requiring for its replication coinfection by another DNA virus (an adenovirus or herpesvirus) led its genus to be named the Dependoviridae. AAV has been the focus of considerable work due to its promising features as a vector for gene therapy.³ A recent study of AAV assembly has now given food for thought in the nucleus organization field.

Sonntag, Schmidt and Kleinschmidt have discovered that a long-known AAV capsid protein-coding gene, termed cap, possesses an alternative reading frame that produces a 23 kDal protein which they named AAP, for assembly-activating protein.⁴ They further found that AAP is localized in the host cell nucleolus, that AAV assembly occurs therein and that AAP plays a role in virus morphogenesis. As or more importantly, they uncovered homologues of AAP in other dependovirus genomes. This work brings to culmination a body of earlier results from this group in which hints of a nucleolar phase in AAV assembly had been noted.⁵

A key finding in the present study was that targeting one of the AAV capsid proteins (VP3) to the nucleolus was not sufficient to enable capsid assembly. This led the authors to postulate that AAP may both facilitate capsid protein import into the nucleus and localization in the nucleolus, but may also direct capsid assembly per se. The jury remains out on these points but all the reagents and methods are available and one senses that an experimental refinement of these points will soon be in hand.

Why would the nucleolus have been selected for as a site of AAV assembly? Given sufficiently high binding affinities, the components of AAV could assemble anywhere in the cell. (This very same point arises in asking, for example, why the signal recognition particle assembles in the nucleolus when in principle it could do so in the nucleoplasm).⁶ Does the more compact nucleolus increase the reactant concentrations relative to what would prevail in the nucleoplasm? Perhaps not since the available information suggests only about a two-fold increased nucleolar mass per unit volume relative to the nucleoplasm.¹ That measurement was directed at the dense fibrillar component of the nucleolus but it is to be borne in mind that the sub-nucleolar domains in which AAV assembly occurs might be more compact and that excluded volume and molecular crowding phenomena might be at play. Indeed, a previous study demonstrated biochemical interactions between and co-immunolocalization of AAV2 capsid proteins with B23/nucleophosmin, hinting of assembly in the nucleolar granular component.⁷

The nucleolus is also known to be a staging and/or functional site for host cell cycle-regulatory proteins.^8,9 Might AAV assembly in the nucleolus somehow signal or resonate with events occurring out in the surrounding nucleus, communicating in some way as an intranuclear analogue of a cell phone tower? This is of course idle speculation but is the kind of idea that the field of systems biology now entertains via revealed interactomes. Finally, have AAV and other dependoviruses persisted because they confer something positive on the host cell? Clearly, their genomes are far too small to encode the kinds of proteins that can usurp host cell machinery as exemplified by the stealth of many “real” viruses (e.g., the enslavement of the cell's translational equipment by picornaviruses). Thus, one wonders whether AAV assembly in the nucleolus actually promotes one or more activities of this nuclear domain? The notion that dependoviruses may actually be cellular commensals has been advanced previously.¹⁰ Here we might best curtail such further speculation or otherwise wind seriously out of control, though in doing so we might recall Hamlet's dictum: “There are more things in heaven and earth, Horatio, than are dreamed of…”

As a final and more general comment, it is striking to this observer that so little cell biology has attended the AAV field given the prominence of this virus as a proposed and now proven vector in gene therapy. But alas, the precise identification of intracelluar sites of viral replication and assembly has often lagged behind in virology. For example, when I was a post-doc, poliovirus was being intensively studied in a lab next door to the one in which I was training.¹¹ In the course of a collaborative project with that lab I heard people refer to poliovirus-laden subcellular fractions as “bags”. I asked them what this meant but got no satisfactory answer. Some of my mentors there (this was Harry Eagle's Department at Cell Biology at the Albert Einstein College of Medicine) were ably and presciently bridging virology and cell biology but it took time, and it has taken time. The recent paper from Jürgen Kleinschmidt and colleagues plants a solid flag, hopefully catalyzing other mountaineers to climb and contribute to understanding the role of the nucleolus in viral replication.

Commentary to: Sonntag F, Schmidt K, Kleinschmidt JA. A viral assembly factor promotes AAV2 capsid formation in the nucleolus. Proc Natl Acad Sci USA. 2010;107:10220–10225.