Chromatin organization at the nuclear pore favours HIV replication

The molecular mechanisms that allow HIV to integrate into particular sites of the host genome are poorly understood. Here we tested if the nuclear pore complex (NPC) facilitates the targeting of HIV integration by acting on chromatin topology. We show that the integrity of the nuclear side of the NPC, which is mainly composed of Tpr, is not required for HIV nuclear import, but that Nup153 is essential. Depletion of Tpr markedly reduces HIV infectivity, but not the level of integration. HIV integration sites in Tpr-depleted cells are less associated with marks of active genes, consistent with the state of chromatin proximal to the NPC, as analysed by super-resolution microscopy. LEDGF/p75, which promotes viral integration into active genes, stabilizes Tpr at the nuclear periphery and vice versa. Our data support a model in which HIV nuclear import and integration are concerted steps, and where Tpr maintains a chromatin environment favourable for HIV replication.

Multicolor fluorescent beads were used to computationally correct sample drift and chromatic shifts. The two colors channels show a degree of overlap, as expected. a) The distance (d) between Nup214 and Nup153 has been processed from scatter plots image. Boxes (n=19) were manually selected in opposite sites along the nuclear membrane. Zooms (8 µmx8µm The first evidence for the role of LEDGF/p75 in HIV-1 target site selection comes from LEDGF/p75 knockdown (KD) cells that showed reduced frequencies of HIV-1 integration into active genes 1 . Our Tpr KD system shares some similarity with LEDGF/p75 KD cells, especially in driving proviral integrations in unusual chromatin sites (Fig.5). Previous studies suggested that LEDGF/p75 helps the virus to integrate in the body of actively transcribed genes, while nuclear basket Nups are involved in directing HIV-1 to accessible chromatin sites underneath the NPC 2,3,3,4 . We determined integration sites by high throughput pyrosequencing and found less integrations in intragenic and more in intergenic regions, in  8,9,10,11 . Experiments on LEDGF/p75 knockout cells supported these findings and revealed a significant percentage of HIV-1 proviruses aberrantly located near TSSs in the absence of the host factor LEDGF/p75 12,13 . Therefore, one of the most appealing opinion suggests the presence of alternative factor/s that could help the virus to integrate in absence of LEDGF/p75, as suggested by results obtained in LEDGF/p75 and HRP2 double KD cells, which show more than expected proviral integrations in active genes 14 . Based on previous notions and on our results we think that the nuclear basket components, in particular Tpr, could have a critical role in HIV integration sites selection, forming a link between viral nuclear translocation and integration. The role of Tpr in LEDGF/p75 and HRP2 double KD cells could explain the portion of proviruses integrated in active genes. This model is supported by the critical role of the nuclear basket in chromatin organization 15 , which according to our data is essential for efficient viral replication. Importantly, previous studies reported that the absence of Tpr induces the presence of heterochromatin near the nuclear basket 15 . In Tpr depleted cells we observed a reduction of H3K36me3 underneath NPCs (Fig.6). Since LEDGF/p75 is directly associated with H3K36me3 6,7 , Tpr depletion could provoke a displacement of LEDGF/p75, which could in turn explain the loss of the usual viral chromatin targets. We and others previously showed that HIV-1 integration sites change when the integrity of the nuclear basket is perturbed by knocking down Nup153 2,3 . However, the depletion of Nup153 causes the concomitant loss of Tpr. In our study we were able to distinguish between the role of two nuclear basket Nups, Nup153 and Tpr, attributing a specific role to each one in the early steps of HIV life cycle (Fig.1) and showing that the main component of the nuclear basket, Tpr, has a critical role in HIV-1integration in active chromatin regions near the NPC.
These results corroborated the importance of the link between viral nuclear translocation and integration, showing that the disruption of this link promotes viral change in integration sites, even though the virus efficiently integrates in the host chromatin.

Cell proliferation and Cell cycle
CellTiter 96® AQueous One Solution Cell Proliferation Assay which is a colorimetric method has been used for determining the number of viable cells in proliferation (Supplementary Fig.1a). Samples were analyzed at 490nm using a 96-well plate reader. Propidium Iodide Labelling and Flow Cytometry were performed on clones fixed in cold 1:1 ethanol (80%)/Acetone for 1 hr at -20°C. After clones were washed and resuspended in propidium iodide (10 mg/mL) and RNase (50 mg/ml) for 30 min at 37°C . Flow cytometry data was analysed using FlowJo (Supplementary Fig.1b)

RNA fractionation
To test the RNA export in clones and in Tpr KD bulk, RNA Nuclear and cytoplasmic were fractionated and extracted using the PARIS kit (Ambion) according the manufacturer's instructions. Nup214 KD bulk has been introduced as control. Cytoplasmic and nuclear RNA were quantified with a Nanodrop ND-1000, loaded on 1% agarose-formaldehyde gel and quantified by ImageJ (Supplementary Fig.2).

Image analysis
Image analysis was carried out using ImageJ and custom-made Matlab tools to quantify intranuclear intensities in almost 20 nuclei per condition (Supplementary Fig.6b,c,d;). Our MatLab tools allowed to outline the nuclear envelope based on the Hoechst image. The nuclear periphery was then defined as the inner region within 2 μm from this polygon. The fluorescence in the red channel was then integrated separately in the peripheral and central regions, and divided by the corresponding area.