RNase II: A new player enters the game

Abstract

Malaria is caused by a unicellular protozoan pathogen of the genus Plasmodium. Although genes represent monocistronic units that are expressed in a life cycle stage-specific manner, post-transcriptional regulation via translational repression of mRNA has been observed in parasite stages that transition from the vertebrate host to the Anopheles vector. An interesting new type of post-transcriptional control was recently discovered in Plasmodium falciparum stages that infect human erythrocytes. A subgroup of genes that were thought to be transcriptionally silent are actually transcribed but degraded immediately by an RNase II that is recruited to these gene loci. This cryptic RNA is not detectable in steady-state RNA but has been detected using nuclear run-on techniques and in mutant RNase II parasites. Nascent RNA degradation controls virulence genes expressed in a monoallelic fashion and noncoding RNAs (ncRNAs), but also a number of housekeeping-like of genes. More studies on other life cycle stages may reveal the full extent of this type of gene regulation in malaria parasites. It is tempting to speculate that RNase II-mediated gene control may exist in other eukaryotic organisms.

Introduction

At any point during the life cycle of malaria parasites, only a subset of the ˜5700 genes is expressed in monocistronic units.^1,2 In the well-studied asexual blood stages, rapid cyclic activation of genes occurs at the transcriptional level during the 48 hour cycle;^3,4 however, relatively few transcription factors have been identified and their specific roles in gene activation awaits validation.⁵ To date, only a few specific DNA-binding proteins have been identified such as PfMyb1⁶ and the ApiAp2 protein family.⁷ In particular, the study of antigenic variation in P. falciparum has revealed several epigenetic regulatory layers contributing to gene expression.^8-11 In malaria parasite sexual stages, translational mRNA repression via temporary formation of a cytoplasmic ribonucleoprotein complex is a major regulatory control mechanism.¹²

Malaria parasites have developed sophisticated strategies that allow phenotypic variation without affecting the genotype. This strategy relies on gene families involved in antigenic and other forms of phenotypic variation, which allows the malaria parasites to cope with changing host environment beyond immune evasion via expression of clonally variant molecules.^13,14 Very little is known about how one gene is expressed while all or most other members of the family are repressed. Active repression of the silent members at the transcriptional level involves spatial organization of virulence genes at perinuclear foci and recruitment of histone modifying enzymes.¹⁵ In the case of the virulence- and pathogenesis-linked var gene family, activation of a single member is associated with relocation to a new perinuclear compartment away from the silent members (for a review see ref. Thirteen).

Discovery of an entirely new type of gene silencing mechanism: PfRNase II-mediated degradation of nascent mRNA in P. falciparum

Because potential post-transcriptional mechanisms that control antigenic variation have yet to be elucidated in P. falciparum, we examined whether RNA-based regulation might occur. Bioinformatics analysis of the P. falciparum genome identified a non-canonical exoribonuclease that contains a putative RNase II domain (termed PfRNase II). Immunoelectron microscopy and in vitro RNA degradation experiments using recombinant PfRNase II showed that this enzyme localizes to the nucleus and degrades single-stranded RNA. Importantly, we demonstrated that PfRNase II is not linked to the RNA exosome multiprotein complex of P. falciparum that contains 2 exoribonucleases (Dis3 and Rrp6). These data suggest that PfRNase II is recruited to certain gene loci and accelerates the decay of mRNAs and ncRNA. Sequencing analysis of total RNA revealed that a specific subclass of var genes called upsA were strongly up-regulated in parasites that expressed a defective PfRNase II (C-terminally tagged with the FKBP destabilization domain) compared to wild-type parasites.¹¹ Furthermore, real-time quantitative PCR (qPCR) showed that different combinations of up to 3 distinct upsA var genes, together with an upsC var gene, were upregulated simultaneously in single parasite clones. These data indicate that loss of PfRNase II affects the strict gene counting mechanism that controls monoallelic var gene expression. Chromatin immunoprecipitation (ChIP)–qPCR revealed that PfRNase II is highly enriched at the promoters and introns of silenced upsA gene loci, and transcription analysis revealed that any nascent transcripts from these genes are only short-lived, cryptic mRNAs.¹¹ We demonstrated that PfRNase II transcription is clinically relevant, as parasites from patients with severe malaria showed an inverse correlation between the levels of PfRNase II and upsA mRNAs. This suggests that by lowering PfRNase II transcript levels, P. falciparum enhances upsA var gene activation and thus the likelihood of causing severe malaria (see schematic model in Fig. 1). This work not only revealed a critical function of exoribonuclease in the pathogenesis of severe malaria, but also pointed to a novel regulatory pathway controlling antigenic variation in a major human pathogen.¹⁶

Figure 1.

Schematic representation of exoribonuclease-mediated gene silencing. Three 3′-5′ exoribonucleases have been detected in P. falciparum,¹¹ of which 2 associate with the core RNA exosome. The RNA exosome has a number of house-keeping functions in RNA processing and degradation of mRNA and cryptic unstable RNA in eukaryotes. Its biological role in malaria parasites remains unknown. All objectives focus on the biology of the recently discovered 3′-5′ exoribonuclease activity in gene silencing and severe malaria.

A more general regulatory role of ribonucleases in dynamic gene expression in P. falciparum?

In malaria parasites, monocistronic transcription has been considered the major pathway in controlling gene expression during various developmental stages of P. falciparum. To date, numerous studies have revealed that epigenetic features such as histone modification, chromatin remodeling, and nuclear architecture are essential regulators of the transcription and silencing of genes during blood-stage infection. ncRNAs have been suggested as potential regulators of P.falciparum gene expression, although the underlying mechanisms have yet to be elucidated.¹⁷ However, little is known about posttranscriptional mechanisms that may potentially regulate gene expression in this eukaryotic organism.

The recent work by Zhang et al. clearly showed by RNA-seq analysis that more than 200 transcripts (including multicopy gene families and ncRNAs) were significantly upregulated when PfRNase II was deficient.¹¹ Thus, this type of gene silencing mechanism is a widespread phenomenon and not specific to upsA- type var genes. Recently, a ChIP-on-chip genome-wide analysis of P. falciparum RNA polymerase II (RNAPII) unexpectedly revealed a simple biphasic occupancy at most genes divided into 2 phases - early and late - during the infectious intraerythrocytic developmental cycle (IDC).¹⁸ The striking contrast between the relatively simple pattern of RNAPII recruitment and the complex dynamics of mRNA metabolism over the IDC strongly suggests that potential regulatory mechanisms downstream of transcriptional initiation such as RNase II-mediated nascent RNA degradation may play a predominant role in the control of gene expression in P. falciparum.

Future Research Directions

Until recently, our knowledge of parasite-specific factors controlling expression of genes linked to severe malaria (upsA-type vars) was limited. Our groundbreaking discovery that PfRNase II plays such a role via a new type of post-transcriptional gene silencing mechanism opens several new avenues that will be explored in the future. Determining the genome-wide binding profile of PfRNase II with Chip-seq may reveal the plasmodial genes regulated by this mechanism. In addition, the identification of factors that recruit PfRNAse II to upsA-type var genes may give new molecular insight into the regulation of malaria pathogenesis.

Genes regulated by PfRNase II-mediated post-transcriptional RNA degradation may represent only a fraction of genes that are actually controlled by this type of silencing mechanism. In addition to PfRNase II, 2 canonical 3′-5′ exoribonucleases, PfRrp6 and PfDis3, have been identified using bioinformatics. Both of these proteins belong to the RNA exosome complex as shown by co-immunoprecipitation with antibodies directed against the RNA exosome core proteins.¹¹ It has been suggested that these exoribonucleases may have functions that are independent of the core.^19,20 Indeed, our immunofluorescence analysis with anti-PfRrp6 antibodies revealed several perinuclear foci, which contrasts with the less nuclear localization of the core exosome. These data raise the possibility that PfRrp6 may play roles independent of the core exosome and possibly complementary to those described for PfRNase II. Ultimately, the biological function of the RNA exosome in malaria parasites remains unknown. We hypothesize that PfDis3 and PfRrp6 contribute to gene regulation via both the exosome complex and exosome-independent chromatin interactions.

Conclusions and Perspectives

The association of exoribonucleases with chromatin and virulence gene expression offers important molecular insight into malaria pathogenesis. We hypothesize that exoribonuclease-mediated RNA degradation plays a more general regulatory role in controlling gene expression during the malaria parasite life cycle than previously thought. It is tempting to postulate that this type of mechanism may exist in other organisms. For example, in African Trypanosomes, variant surface antigen genes (VSG) are expressed from one of 15–20 expression sites (ES). Only one ES is fully active at a time, leading to monoallelic expression. A recent report showed that in single cells, transcription is initiated at several ESs simultaneously, indicating that monoallelic control may be determined by RNA processing.²¹ Future elucidation of this mechanism may reveal the full extent of its role in eukaryotic gene regulation.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank Jessica Bryant for critical reading of the manuscript.

Funding

This work was supported by an ERC grant (PlasmoEscape 250320), the French parasitology consortium ParaFrap (ANR-11 LABX0024), ANR-13-ISV3-0003-01-NSFC (no. 81361130411) International Collaboration Project and the National Natural Science Foundation of China (NSFC; no. 31271388).