Modeling splicing outcome by combining 5'ss strength and splicing regulatory elements.

Correct pre-mRNA processing in higher eukaryotes vastly depends on splice site recognition. Beyond conserved 5'ss and 3'ss motifs, splicing regulatory elements (SREs) play a pivotal role in this recognition process. Here, we present in silico designed sequences with arbitrary a priori prescribed splicing regulatory HEXplorer properties that can be concatenated to arbitrary length without changing their regulatory properties. We experimentally validated in silico predictions in a massively parallel splicing reporter assay on more than 3000 sequences and exemplarily identified some SRE binding proteins. Aiming at a unified 'functional splice site strength' encompassing both U1 snRNA complementarity and impact from neighboring SREs, we developed a novel RNA-seq based 5'ss usage landscape, mapping the competition of pairs of high confidence 5'ss and neighboring exonic GT sites along HBond and HEXplorer score coordinate axes on human fibroblast and endothelium transcriptome datasets. These RNA-seq data served as basis for a logistic 5'ss usage prediction model, which greatly improved discrimination between strong but unused exonic GT sites and annotated highly used 5'ss. Our 5'ss usage landscape offers a unified view on 5'ss and SRE neighborhood impact on splice site recognition, and may contribute to improved mutation assessment in human genetics.

Modifying splice site usage with ModCon: Maintaining the genetic code while changing the underlying mRNP code.

Codon degeneracy of amino acid sequences permits an additional "mRNP code" layer underlying the genetic code that is related to RNA processing. In pre-mRNA splicing, splice site usage is determined by both intrinsic strength and sequence context providing RNA binding sites for splicing regulatory proteins. In this study, we systematically examined modification of splicing regulatory properties in the neighborhood of a GT site, i.e. potential splice site, without altering the encoded amino acids. We quantified the splicing regulatory properties of the neighborhood around a potential splice site by its Splice Site HEXplorer Weight (SSHW) based on the HEXplorer score algorithm. To systematically modify GT site neighborhoods, either minimizing or maximizing their SSHW, we designed the novel stochastic optimization algorithm ModCon that applies a genetic algorithm with stochastic crossover, insertion and random mutation elements supplemented by a heuristic sliding window approach. To assess the achievable range in SSHW in human splice donors without altering the encoded amino acids, we applied ModCon to a set of 1000 randomly selected Ensembl annotated human splice donor sites, achieving substantial and accurate changes in SSHW. Using ModCon optimization, we successfully switched splice donor usage in a splice site competition reporter containing coding sequences from FANCA, FANCB or BRCA2, while retaining their amino acid coding information. The ModCon algorithm and its R package implementation can assist in reporter design by either introducing novel splice sites, silencing accidental, undesired splice sites, and by generally modifying the entire mRNP code while maintaining the genetic code.

Differential acute impact of therapeutically effective and overdose concentrations of lithium on human neuronal single cell and network function.

Lithium salts are used as mood-balancing medication prescribed to patients suffering from neuropsychiatric disorders, such as bipolar disorder and major depressive disorder. Lithium salts cross the blood-brain barrier and reach the brain parenchyma within few hours after oral application, however, how lithium influences directly human neuronal function is unknown. We applied patch-clamp and microelectrode array technology on human induced pluripotent stem cell (iPSC)-derived cortical neurons acutely exposed to therapeutic (<1 mM) and overdose concentrations (>1 mM) of lithium chloride (LiCl) to assess how therapeutically effective and overdose concentrations of LiCl directly influence human neuronal electrophysiological function at the synapse, single-cell, and neuronal network level. We describe that human iPSC-cortical neurons exposed to lithium showed an increased neuronal activity under all tested concentrations. Furthermore, we reveal a lithium-induced, concentration-dependent, transition of regular synchronous neuronal network activity using therapeutically effective concentration (<1 mM LiCl) to epileptiform-like neuronal discharges using overdose concentration (>1 mM LiCl). The overdose concentration lithium-induced epileptiform-like activity was similar to the epileptiform-like activity caused by the GABAA-receptor antagonist. Patch-clamp recordings reveal that lithium reduces action potential threshold at all concentrations, however, only overdose concentration causes increased frequency of spontaneous AMPA-receptor mediated transmission. By applying the AMPA-receptor antagonist and anti-epileptic drug Perampanel, we demonstrate that Perampanel suppresses lithium-induced epileptiform-like activity in human cortical neurons. We provide insights in how therapeutically effective and overdose concentration of lithium directly influences human neuronal function at synapse, a single neuron, and neuronal network levels. Furthermore, we provide evidence that Perampanel suppresses pathological neuronal discharges caused by overdose concentrations of lithium in human neurons.

VarCon: An R Package for Retrieving Neighboring Nucleotides of an SNV.

Reporting of a single nucleotide variant (SNV) follows the Sequence Variant Nomenclature (http://varnomen.hgvs.org/), using an unambiguous numbering scheme specific for coding and noncoding DNA. However, the corresponding sequence neighborhood of a given SNV, which is required to assess its impact on splicing regulation, is not easily accessible from this nomenclature. Providing fast and easy access to this neighborhood just from a given SNV reference, the novel tool VarCon combines information of the Ensembl human reference genome and the corresponding transcript table for accurate retrieval. VarCon also displays splice site scores (HBond and MaxEnt scores) and HEXplorer profiles of an SNV neighborhood, reflecting position-dependent splice enhancing and silencing properties.

TGF-ß1 Suppresses Proliferation and Induces Differentiation in Human iPSC Neural in vitro Models.

Persistent neural stem cell (NSC) proliferation is, among others, a hallmark of immaturity in human induced pluripotent stem cell (hiPSC)-based neural models. TGF-ß1 is known to regulate NSCs in vivo during embryonic development in rodents. Here we examined the role of TGF-ß1 as a potential candidate to promote in vitro differentiation of hiPSCs-derived NSCs and maturation of neuronal progenies. We present that TGF-ß1 is specifically present in early phases of human fetal brain development. We applied confocal imaging and electrophysiological assessment in hiPSC-NSC and 3D neural in vitro models and demonstrate that TGF-ß1 is a signaling protein, which specifically suppresses proliferation, enhances neuronal and glial differentiation, without effecting neuronal maturation. Moreover, we demonstrate that TGF-ß1 is equally efficient in enhancing neuronal differentiation of human NSCs as an artificial synthetic small molecule. The presented approach provides a proof-of-concept to replace artificial small molecules with more physiological signaling factors, which paves the way to improve the physiological relevance of human neural developmental in vitro models.

Human Cerebrospinal Fluid Promotes Neuronal Circuit Maturation of Human Induced Pluripotent Stem Cell-Derived 3D Neural Aggregates.

Human induced pluripotent stem cell (hiPSC)-derived in vitro neural and organoid models resemble fetal, rather than adult brain properties, indicating that currently applied cultivation media and supplements are insufficient to achieve neural maturation beyond the fetal stage. In vivo, cerebrospinal fluid molecules are regulating the transition of the immature fetal human brain into a mature adult brain. By culturing hiPSC-3D neural aggregates in human cerebrospinal fluid (hCSF) obtained from healthy adult individuals, we demonstrate that hCSF rapidly triggers neurogenesis, gliogenesis, synapse formation, neurite outgrowth, suppresses proliferation of residing neural stem cells, and results in the formation of synchronously active neuronal circuits in vitro within 3 days. Thus, a physiologically relevant and adult brain-like milieu triggers maturation of hiPSC-3D neural aggregates into highly functional neuronal circuits in vitro. The approach presented here opens a new avenue to identify novel physiological factors for the improvement of hiPSC neural in vitro models.

Characterization and application of electrically active neuronal networks established from human induced pluripotent stem cell-derived neural progenitor cells for neurotoxicity evaluation.

Neurotoxicity is mediated by a variety of modes-of-actions leading to disturbance of neuronal function. In order to screen larger numbers of compounds for their neurotoxic potential, in vitro functional neuronal networks (NN) might be helpful tools. We established and characterized human NN (hNN) from hiPSC-derived neural progenitor cells by comparing hNN formation with two different differentiation media: in presence (CINDA) and absence (neural differentiation medium (NDM)) of maturation-supporting factors. As a NN control we included differentiating rat NN (rNN) in the study. Gene/protein expression and electrical activity from in vitro developing NN were assessed at multiple time points. Transcriptomes of 5, 14 and 28 days in vitro CINDA-grown hNN were compared to gene expression profiles of in vivo human developing brains. Molecular expression analyses as well as measures of electrical activity indicate that NN mature into neurons of different subtypes and astrocytes over time. In contrast to rNN, hNN are less electrically active within the same period of differentiation time, yet hNN grown in CINDA medium develop higher firing rates than hNN without supplements. Challenge of NN with neuronal receptor stimulators and inhibitors demonstrate presence of inhibitory, GABAergic neurons, whereas glutamatergic responses are limited. hiPSC-derived GABAergic hNN grown in CINDA medium might be a useful tool as part of an in vitro battery for assessing neurotoxicity.

Effect fingerprints of antipsychotic drugs on neural networks in vitro.

We compared the acute effect of typical (haloperidol) and atypical (aripiprazole, clozapine, olanzapine) antipsychotic drugs (APDs) on spontaneous electrophysiological activity of in vitro neuronal networks cultured on microelectrode arrays (MEAs). Network burst analysis revealed a "regularizing" effect of all APDs at therapeutic concentrations, i.e., an increase of network-wide temporal synchronization. At supratherapeutic concentrations, all APDs but olanzapine mediated a decrease of burst and spike rates, burst duration, number of spikes in bursts, and network synchrony. The rank order of potency of APDs was: haloperidol > aripiprazole > clozapine > olanzapine (no suppression). Disruption of network function was not due to enhanced cell death as assessed by trypan blue staining. APDs promoted distinct concentration-dependent alterations yielding acute effect fingerprints of the tested compounds. These effects were rather characteristic for individual compounds than distinctive for typical vs. atypical APDs. Thus, this dichotomy may be of value in distinguishing clinical features but has no apparent basis on the network or local circuitry level.

Context matters: Regulation of splice donor usage.

Elaborate research on splicing, starting in the late seventies, evolved from the discovery that 5' splice sites are recognized by their complementarity to U1 snRNA towards the realization that RNA duplex formation cannot be the sole basis for 5'ss selection. Rather, their recognition is highly influenced by a number of context factors including transcript architecture as well as splicing regulatory elements (SREs) in the splice site neighborhood. In particular, proximal binding of splicing regulatory proteins highly influences splicing outcome. The importance of SRE integrity especially becomes evident in the light of human pathogenic mutations where single nucleotide changes in SREs can severely affect the resulting transcripts. Bioinformatics tools nowadays greatly assist in the computational evaluation of 5'ss, their neighborhood and the impact of pathogenic mutations. Although predictions are already quite robust, computational evaluation of the splicing regulatory landscape still faces challenges to increase future reliability. This article is part of a Special Issue entitled: RNA structure and splicing regulation edited by Francisco Baralle, Ravindra Singh and Stefan Stamm.

Robust Generation of Person-Specific, Synchronously Active Neuronal Networks Using Purely Isogenic Human iPSC-3D Neural Aggregate Cultures.

Reproducibly generating human induced pluripotent stem cell-based functional neuronal circuits, solely obtained from single individuals, poses particular challenges to achieve personalized and patient specific functional neuronal in vitro models. A hallmark of functional neuronal assemblies, synchronous neuronal activity, can be non-invasively studied by microelectrode array (MEA) technology, reliably capturing physiological and pathophysiological aspects of human brain function. In our here presented manuscript, we demonstrate a procedure to generate 3D neural aggregates comprising astrocytes, oligodendroglial cells, and neurons obtained from the same human tissue sample. Moreover, we demonstrate the robust ability of those neurons to create a highly synchronously active neuronal network within 3 weeks in vitro, without additionally applied astrocytes. The fusion of MEA-technology with functional neuronal circuits solely obtained from one individual's cells represent isogenic person-specific human neuronal sensor chips that pave the way for specific personalized in vitro neuronal networks as well as neurological and neuropsychiatric disease modeling.

In vitro neuronal network activity as a new functional diagnostic system to detect effects of Cerebrospinal fluid from autoimmune encephalitis patients.

The intent of this study was to investigate if cerebrospinal fluid (CSF) from autoimmune encephalitis (AE) patients regulates in vitro neuronal network activity differentially to healthy human control CSF (hCSF). To this end, electrophysiological effects of CSF from AE patients or hCSF were measured by in vitro neuronal network activity (ivNNA) recorded with microelectrode arrays (MEA). CSF from patients with either N-methyl-D-aspartate-receptor-antibody (pCSFNMDAR, n = 7) or Leucine-rich-glioma-inactivated-1-Ab (pCSFLGI1, n = 6) associated AE suppressed global spiking activity of neuronal networks by a factor of 2.17 (p < 0.05) or 2.42 (p < 0.05) compared to hCSF. The former also suppressed synchronous network bursting by a factor of 1.93 (p < 0.05) in comparison to hCSF (n = 13). As a functional diagnostic test, this parameter reached a sensitivity of 86% for NMDAR-Ab- and 100% for LGI1-Ab-associated AE vs. hCSF at a specificity of 85%. To explore if modulation at the NMDAR influences effects of hCSF or pathological CSF, we applied the NMDAR-antagonist 2-Amino-5-phosphono-pentanoic acid (AP5). In CSF from NMDAR-Ab-associated AE patients, spike rate reduction by AP5 was more than 2-fold larger than in hCSF (p < 0.05), and network burst rate reduction more than 18-fold (p < 0.01). Recording ivNNA might help discriminating between functional effects of CSF from AE patients and hCSF, and thus could be used as a functional diagnostic test in AE. The pronounced suppression of ivNNA by CSF from NMDAR-Ab-associated AE patients and simultaneous antagonism at the NMDAR by AP5, particularly in burst activity, compared to hCSF plus AP5, confirms that the former contains additional ivNNA-suppressing factors.

Ranking noncanonical 5' splice site usage by genome-wide RNA-seq analysis and splicing reporter assays.

Most human pathogenic mutations in 5' splice sites affect the canonical GT in positions +1 and +2, leading to noncanonical dinucleotides. On the other hand, noncanonical dinucleotides are observed under physiological conditions in ∼1% of all human 5'ss. It is therefore a challenging task to understand the pathogenic mutation mechanisms underlying the conditions under which noncanonical 5'ss are used. In this work, we systematically examined noncanonical 5' splice site selection, both experimentally using splicing competition reporters and by analyzing a large RNA-seq data set of 54 fibroblast samples from 27 subjects containing a total of 2.4 billion gapped reads covering 269,375 exon junctions. From both approaches, we consistently derived a noncanonical 5'ss usage ranking GC > TT > AT > GA > GG > CT. In our competition splicing reporter assay, noncanonical splicing was strictly dependent on the presence of upstream or downstream splicing regulatory elements (SREs), and changes in SREs could be compensated by variation of U1 snRNA complementarity in the competing 5'ss. In particular, we could confirm splicing at different positions (i.e., -1, +1, +5) of a splice site for all noncanonical dinucleotides "weaker" than GC. In our comprehensive RNA-seq data set analysis, noncanonical 5'ss were preferentially detected in weakly used exon junctions of highly expressed genes. Among high-confidence splice sites, they were 10-fold overrepresented in clusters with a neighboring, more frequently used 5'ss. Conversely, these more frequently used neighbors contained only the dinucleotides GT, GC, and TT, in accordance with the above ranking.

Comparative performance analysis of human iPSC-derived and primary neural progenitor cells (NPC) grown as neurospheres in vitro.

Developmental neurotoxicity (DNT) testing performed in rats is resource-intensive (costs, time, animals) and bears the issue of species extrapolation. Thus, reliable alternative human-based approaches are needed for predicting neurodevelopmental toxicity. Human induced pluripotent stem cells (hiPSCs) represent a basis for an alternative method possibly being part of an alternative DNT testing strategy. Here, we compared two hiPSC neural induction protocols resulting in 3D neurospheres: one using noggin and one cultivating cells in neural induction medium (NIM protocol). Performance of Nestin+/SOX2+ hiPSC-derived neural progenitor cells (NPCs) was compared to primary human NPCs. Generally, primary hNPCs first differentiate into Nestin+ and/or GFAP+ radial glia-like cells, while the hiPSC-derived NPCs (hiPSC-NPC) first differentiate into ßIII-Tubulin+ neurons suggesting an earlier developmental stage of hiPSC-NPC. In the 'Neurosphere Assay', NIM generated hiPSC-NPC produced neurons with higher performance than with the noggin protocol. After long-term differentiation, hiPSC-NPC form neuronal networks, which become electrically active on microelectrode arrays after 85days. Finally, methylmercury chloride inhibits hiPSC-NPC and hNPC migration with similar potencies. hiPSC-NPCs-derived neurospheres seem to be useful for DNT evaluation representing early neural development in vitro. More system characterization by compound testing is needed to gain higher confidence in this method.

Dementia with Lewy bodies: cerebrospinal fluid suppresses neuronal network activity.

One of the core clinical criteria of Dementia with Lewy bodies (DLB) are fluctuations of cognition. Underlying processes may be rather functional than neurodegenerative, reflected by, for example, factors present in cerebrospinal fluid (CSF). The aim of this study was to identify in-vitro neuronal network activity (ivNNA) changes of CSF from DLB patients compared with patients with Parkinson's disease (PD) and controls. Primary neuronal mouse cultures were grown on microelectrode arrays to record ivNNA when exposed to respective CSF samples. If exposed to CSF of DLB patients, ivNNA showed a reduced spike rate and burst rate compared with CSF of PD patients and controls. Our data are suggestive of the presence of functional factors in the CSF of DLB patients that differentiate network activity from PD patients and controls. Future studies should evaluate whether this pilot observation might be related to fluctuations of cognition in DLB.

Analysis of Competing HIV-1 Splice Donor Sites Uncovers a Tight Cluster of Splicing Regulatory Elements within Exon 2/2b.

The HIV-1 accessory protein Vif is essential for viral replication by counteracting the host restriction factor APOBEC3G (A3G), and balanced levels of both proteins are required for efficient viral replication. Noncoding exons 2/2b contain the Vif start codon between their alternatively used splice donors 2 and 2b (D2 and D2b). For vif mRNA, intron 1 must be removed while intron 2 must be retained. Thus, splice acceptor 1 (A1) must be activated by U1 snRNP binding to either D2 or D2b, while splicing at D2 or D2b must be prevented. Here, we unravel the complex interactions between previously known and novel components of the splicing regulatory network regulating HIV-1 exon 2/2b inclusion in viral mRNAs. In particular, using RNA pulldown experiments and mass spectrometry analysis, we found members of the heterogeneous nuclear ribonucleoparticle (hnRNP) A/B family binding to a novel splicing regulatory element (SRE), the exonic splicing silencer ESS2b, and the splicing regulatory proteins Tra2/SRSF10 binding to the nearby exonic splicing enhancer ESE2b. Using a minigene reporter, we performed bioinformatics HEXplorer-guided mutational analysis to narrow down SRE motifs affecting splice site selection between D2 and D2b. Eventually, the impacts of these SREs on the viral splicing pattern and protein expression were exhaustively analyzed in viral particle production and replication experiments. Masking of these protein binding sites by use of locked nucleic acids (LNAs) impaired Vif expression and viral replication.IMPORTANCE Based on our results, we propose a model in which a dense network of SREs regulates vif mRNA and protein expression, crucial to maintain viral replication within host cells with varying A3G levels and at different stages of infection. This regulation is maintained by several serine/arginine-rich splicing factors (SRSF) and hnRNPs binding to those elements. Targeting this cluster of SREs with LNAs may lead to the development of novel effective therapeutic strategies.

Succession of splicing regulatory elements determines cryptic 5΄ss functionality.

A critical step in exon definition is the recognition of a proper splice donor (5΄ss) by the 5' end of U1 snRNA. In the selection of appropriate 5΄ss, cis-acting splicing regulatory elements (SREs) are indispensable. As a model for 5΄ss recognition, we investigated cryptic 5΄ss selection within the human fibrinogen Bß-chain gene (FGB) exon 7, where we identified several exonic SREs that simultaneously acted on up- and downstream cryptic 5΄ss. In the FGB exon 7 model system, 5΄ss selection iteratively proceeded along an alternating sequence of U1 snRNA binding sites and interleaved SREs which in principle supported different 3' exon ends. Like in a relay race, SREs either suppressed a potential 5΄ss and passed the splicing baton on or splicing actually occurred. From RNA-Seq data, we systematically selected 19 genes containing exons with silent U1 snRNA binding sites competing with nearby highly used 5΄ss. Extensive SRE analysis by different algorithms found authentic 5΄ss significantly more supported by SREs than silent U1 snRNA binding sites, indicating that our concept may permit generalization to a model for 5΄ss selection and 3' exon end definition.

Glutamine triggers long-lasting increase in striatal network activity in vitro.

Accumulation of ammonium and glutamine in blood and brain is a key factor in hepatic encephalopathy (HE) - a neuropsychiatric syndrome characterized by various cognitive and motor deficits. MRI imaging identified abnormalities notably in the basal ganglia of HE patients, including its major input station, the striatum. While neurotoxic effects of ammonia have been extensively studied, glutamine is primarily perceived as "detoxified" form of ammonia. We applied ammonium and glutamine to striatal and cortical cells from newborn rats cultured on microelectrode arrays. Glutamine, but not ammonium significantly increased spontaneous spike rate with a long-lasting excitation outlasting washout. This effect was more prominent in striatal than in cortical cultures. Calcium imaging revealed that glutamine application caused a rise in intracellular calcium that depended both on system A amino acid transport and activation of ionotropic glutamate receptors. This pointed to downstream glutamate release that was triggered by intracellular glutamine. Using an enzymatic assay kit we confirmed glutamine-provoked glutamate release from striatal cells. Real-time PCR and immunocytochemistry demonstrated the presence of vesicular glutamate transporters (VGLUT1 and VGLUT2) necessary for synaptic glutamate release in striatal neurons. We conclude that extracellular glutamine is taken up by neurons, triggers synaptic release of glutamate which is then taken up by astrocytes and again converted to glutamine. This feedback-loop causes a sustained long-lasting excitation of network activity. Thus, apart from ammonia also its "detoxified" form glutamine might be responsible for the neuropsychiatric symptoms in HE.

Cell Assembly Signatures Defined by Short-Term Synaptic Plasticity in Cortical Networks.

The cell assembly (CA) hypothesis has been used as a conceptual framework to explain how groups of neurons form memories. CAs are defined as neuronal pools with synchronous, recurrent and sequential activity patterns. However, neuronal interactions and synaptic properties that define CAs signatures have been difficult to examine because identities and locations of assembly members are usually unknown. In order to study synaptic properties that define CAs, we used optical and electrophysiological approaches to record activity of identified neurons in mouse cortical cultures. Population analysis and graph theory techniques allowed us to find sequential patterns that represent repetitive transitions between network states. Whole cell pair recordings of neurons participating in repeated sequences demonstrated that synchrony is exhibited by groups of neurons with strong synaptic connectivity (concomitant firing) showing short-term synaptic depression (STD), whereas alternation (sequential firing) is seen in groups of neurons with weaker synaptic connections showing short-term synaptic facilitation (STF). Decreasing synaptic weights of a network promoted the generation of sequential activity patterns, whereas increasing synaptic weights restricted state transitions. Thus in simple cortical networks of real neurons, basic signatures of CAs, the properties that underlie perception and memory in Hebb's original description, are already present.

High-frequency voltage oscillations in cultured astrocytes.

Because of their close interaction with neuronal physiology, astrocytes can modulate brain function in multiple ways. Here, we demonstrate a yet unknown astrocytic phenomenon: Astrocytes cultured on microelectrode arrays (MEAs) exhibited extracellular voltage fluctuations in a broad frequency spectrum (100-600 Hz) after electrical stimulation. These aperiodic high-frequency oscillations (HFOs) could last several seconds and did not spread across the MEA. The voltage-gated calcium channel antagonist cilnidipine dose-dependently decreased the power of the oscillations. While intracellular calcium was pivotal, incubation with bafilomycin A1 showed that vesicular release of transmitters played only a minor role in the emergence of HFOs. Gap junctions and volume-regulated anionic channels had just as little functional impact, which was demonstrated by the addition of carbenoxolone (100 µmol/L) and NPPB (100 µmol/L). Hyperpolarization with low potassium in the extracellular solution (2 mmol/L) dramatically raised oscillation power. A similar effect was seen when we added extra sodium (+50 mmol/L) or if we replaced it with NMDG(+) (50 mmol/L). The purinergic receptor antagonist PPADS suppressed the oscillation power, while the agonist ATP (100 µmol/L) had only an increasing effect when the bath solution pH was slightly lowered to pH 7.2. From these observations, we conclude that astrocytic voltage oscillations are triggered by activation of voltage-gated calcium channels and driven by a downstream influx of cations through channels that are permeable for large ions such as NMDG(+). Most likely candidates are subtypes of pore-forming P2X channels with a low affinity for ATP.

Balanced splicing at the Tat-specific HIV-1 3'ss A3 is critical for HIV-1 replication.

BACKGROUND: The viral regulatory protein Tat is essential for establishing a productive transcription from the 5'-LTR promoter during the early phase of viral gene expression. Formation of the Tat-encoding mRNAs requires splicing at the viral 3'ss A3, which has previously been shown to be both negatively and positively regulated by the downstream splicing regulatory elements (SREs) ESS2p and ESE2/ESS2. However, using the novel RESCUE-type computational HEXplorer algorithm, we were recently able to identify another splicing enhancer (ESE(5807-5838), henceforth referred to as ESE tat ) located between ESS2p and ESE2/ESS2. Here we show that ESE tat has a great impact on viral tat-mRNA splicing and that it is fundamental for regulated 3'ss A3 usage. RESULTS: Mutational inactivation or locked nucleic acid (LNA)-directed masking of the ESE tat sequence in the context of a replication-competent virus was associated with a failure (i) to activate viral 3'ss A3 and (ii) to accumulate Tat-encoding mRNA species. Consequently, due to insufficient amounts of Tat protein efficient viral replication was drastically impaired. RNA in vitro binding assays revealed SRSF2 and SRSF6 as candidate splicing factors acting through ESE tat and ESE2 for 3'ss A3 activation. This notion was supported by coexpression experiments, in which wild-type, but not ESE tat -negative provirus responded to higher levels of SRSF2 and SRSF6 proteins with higher levels of tat-mRNA splicing. Remarkably, we could also find that SRSF6 overexpression established an antiviral state within provirus-transfected cells, efficiently blocking virus particle production. For the anti-HIV-1 activity the arginine-serine (RS)-rich domain of the splicing factor was dispensable. CONCLUSIONS: Based on our results, we propose that splicing at 3'ss A3 is dependent on binding of the enhancing SR proteins SRSF2 and SRSF6 to the ESE tat and ESE2 sequence. Mutational inactivation or interference specifically with ESE tat activity by LNA-directed masking seem to account for an early stage defect in viral gene expression, probably by cutting off the supply line of Tat that HIV needs to efficiently transcribe its genome.