Limits...
HIV-1 RNAs are Not Part of the Argonaute 2 Associated RNA Interference Pathway in Macrophages.

Vongrad V, Imig J, Mohammadi P, Kishore S, Jaskiewicz L, Hall J, Günthard HF, Beerenwinkel N, Metzner KJ - PLoS ONE (2015)

Bottom Line: HIV-1 infected monocyte-derived macrophages (MDM) were chosen as target cells, as they have previously been shown to express HIV-1 sncRNAs.Small RNA sequencing revealed that 1.4% of all sncRNAs were of HIV-1 origin.However, neither HIV-1 derived sncRNAs nor putative HIV-1 target sequences incorporated into Ago2-RISC were identified suggesting that HIV-1 sncRNAs are not involved in the canonical RNAi pathway nor is HIV-1 targeted by this pathway in HIV-1 infected macrophages.

View Article: PubMed Central - PubMed

Affiliation: University Hospital Zurich, Division of Infectious Diseases and Hospital Epidemiology, University of Zurich, Zurich, Switzerland; Institute of Medical Virology, University of Zurich, Zurich, Switzerland.

ABSTRACT

Background: MiRNAs and other small noncoding RNAs (sncRNAs) are key players in post-transcriptional gene regulation. HIV-1 derived small noncoding RNAs (sncRNAs) have been described in HIV-1 infected cells, but their biological functions still remain to be elucidated. Here, we approached the question whether viral sncRNAs may play a role in the RNA interference (RNAi) pathway or whether viral mRNAs are targeted by cellular miRNAs in human monocyte derived macrophages (MDM).

Methods: The incorporation of viral sncRNAs and/or their target RNAs into RNA-induced silencing complex was investigated using photoactivatable ribonucleoside-induced cross-linking and immunoprecipitation (PAR-CLIP) as well as high-throughput sequencing of RNA isolated by cross-linking immunoprecipitation (HITS-CLIP), which capture Argonaute2-bound miRNAs and their target RNAs. HIV-1 infected monocyte-derived macrophages (MDM) were chosen as target cells, as they have previously been shown to express HIV-1 sncRNAs. In addition, we applied small RNA deep sequencing to study differential cellular miRNA expression in HIV-1 infected versus non-infected MDMs.

Results and conclusion: PAR-CLIP and HITS-CLIP data demonstrated the absence of HIV-1 RNAs in Ago2-RISC, although the presence of a multitude of HIV-1 sncRNAs in HIV-1 infected MDMs was confirmed by small RNA sequencing. Small RNA sequencing revealed that 1.4% of all sncRNAs were of HIV-1 origin. However, neither HIV-1 derived sncRNAs nor putative HIV-1 target sequences incorporated into Ago2-RISC were identified suggesting that HIV-1 sncRNAs are not involved in the canonical RNAi pathway nor is HIV-1 targeted by this pathway in HIV-1 infected macrophages.

No MeSH data available.


Related in: MedlinePlus

Cellular micro RNAs as detected by Ago2 PAR-CLIP and small RNA sequencing.(A) Percentage of the cellular miRNAs from small RNA sequencing that was also found in Ago2 PAR-CLIP as a function of minimum expression threshold. The x-axis shows the total abundance of miRNAs in small RNA sequencing data derived from pooled HIV-1JR-FL infected (n = 4) and HIV-1 non-infected (n = 4) samples from the same donors. Red lines show the read count corresponding to specific fractions of total miRNA pool (B) Expression of cellular miRNAs in Ago2 PAR-CLIP and small RNA sequencing data is well correlated for 160 miRNAs found by both methods (R = 0.55, p<10‒13). The dashed lines show bootstrap-derived 95% confidence intervals for the linear fit (red line). (C) Expected PAR-CLIP read counts of virus-derived sncRNAs associated with Ago2. Adjusted to small RNA sequencing data derived from pooled HIV-1JR-FL infected (n = 2) samples matching the PAR-CLIP donors. The black line (5 reads aligned) depicts the detection limit of the PAR-CLIP assay with majority of the loci on the sense genome are expected to surpass the detection limit. Upper and lower panels correspond to the reads on the sense and anti-sense strand of HIV-1JR-FL genome respectively.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4520458&req=5

pone.0132127.g002: Cellular micro RNAs as detected by Ago2 PAR-CLIP and small RNA sequencing.(A) Percentage of the cellular miRNAs from small RNA sequencing that was also found in Ago2 PAR-CLIP as a function of minimum expression threshold. The x-axis shows the total abundance of miRNAs in small RNA sequencing data derived from pooled HIV-1JR-FL infected (n = 4) and HIV-1 non-infected (n = 4) samples from the same donors. Red lines show the read count corresponding to specific fractions of total miRNA pool (B) Expression of cellular miRNAs in Ago2 PAR-CLIP and small RNA sequencing data is well correlated for 160 miRNAs found by both methods (R = 0.55, p<10‒13). The dashed lines show bootstrap-derived 95% confidence intervals for the linear fit (red line). (C) Expected PAR-CLIP read counts of virus-derived sncRNAs associated with Ago2. Adjusted to small RNA sequencing data derived from pooled HIV-1JR-FL infected (n = 2) samples matching the PAR-CLIP donors. The black line (5 reads aligned) depicts the detection limit of the PAR-CLIP assay with majority of the loci on the sense genome are expected to surpass the detection limit. Upper and lower panels correspond to the reads on the sense and anti-sense strand of HIV-1JR-FL genome respectively.

Mentions: Although highly abundant in small RNA sequencing, HIV-1 sncRNAs were almost completely absent in the Ago2 PAR-CLIP libraries. In order to evaluate the quality of our PAR-CLIP assays, we focused our analysis on the known miRNAs, assuming that the majority of expressed miRNAs can be associated with Ago2-RISC and hence should be present in Ago2 PAR-CLIP data. Pooled Ago2 PAR-CLIP and small RNA sequencing libraries captured 183 and 475 unique mature miRNAs that were present by at least five reads. 160 miRNAs were detected by both assays. The relative sensitivity of the Ago2 PAR-CLIP assay was assessed as a function of miRNA abundance in the small RNA sequencing experiment. We observed a strong correspondence between miRNA frequency in total miRNA library (n = 4) and its detection by Ago2 PAR-CLIP (n = 4) (Fig 2A). 88% of all cellular miRNAs—detected with ≥1,000 reads in small RNA sequencing—were also observed in PAR-CLIP (Fig 2A). This rate dropped only to 64% and 38% when ≥100 and ≥10 reads, respectively, were chosen as cut-offs. Even more than 30% of low abundant miRNAs, i.e., detected with only ≥5 reads, were identified in the PAR-CLIP data (Fig 2A). Furthermore, the expression level of the miRNAs was highly correlated between the two methods for the 160 miRNAs detected in both datasets (R = 0.55, p<10‒13) (Fig 2B). Given the strong concordance of the two methods in detecting cellular miRNAs we expected potential viral miRNAs or target sites—if associated with Ago2—to be also present in the Ago2 PAR-CLIP data. Fig 2C demonstrates HIV-1 reads observed in small RNA sequencing for the two PAR-CLIP matched donors, and indicates indicates the expected read count in our PAR-CLIP data for authentic Ago2 associated HIV-1 sncRNAs based on our observations using cellular miRNA data (Fig 2A and 2B). Despite the detection of four highly abundant HIV-1 sncRNAs (>500 reads) and 107 abundant HIV-1 sncRNAs (>100 reads) by small RNA sequencing none of them were present in the PAR-CLIP data (Fig 2C). Only one HIV-1 sncRNA was observed in both, the small RNA sequencing and the PAR-CLIP data (Fig 2C), but this HIV-1 sncRNA corresponded to the tRNALys primer binding site and was also detected in HIV-1 non-infected samples (Fig F in S1 File). Moreover, the highly abundant reads on the sense strand of the virus, equivalent to 9.1% of miRNA pool, in the small RNA sequencing data was entirely absent in the PAR-CLIP and hence it is highly unlikely that HIV-1 sncRNAs were missed due to the detection limit of the PAR-CLIP assay.


HIV-1 RNAs are Not Part of the Argonaute 2 Associated RNA Interference Pathway in Macrophages.

Vongrad V, Imig J, Mohammadi P, Kishore S, Jaskiewicz L, Hall J, Günthard HF, Beerenwinkel N, Metzner KJ - PLoS ONE (2015)

Cellular micro RNAs as detected by Ago2 PAR-CLIP and small RNA sequencing.(A) Percentage of the cellular miRNAs from small RNA sequencing that was also found in Ago2 PAR-CLIP as a function of minimum expression threshold. The x-axis shows the total abundance of miRNAs in small RNA sequencing data derived from pooled HIV-1JR-FL infected (n = 4) and HIV-1 non-infected (n = 4) samples from the same donors. Red lines show the read count corresponding to specific fractions of total miRNA pool (B) Expression of cellular miRNAs in Ago2 PAR-CLIP and small RNA sequencing data is well correlated for 160 miRNAs found by both methods (R = 0.55, p<10‒13). The dashed lines show bootstrap-derived 95% confidence intervals for the linear fit (red line). (C) Expected PAR-CLIP read counts of virus-derived sncRNAs associated with Ago2. Adjusted to small RNA sequencing data derived from pooled HIV-1JR-FL infected (n = 2) samples matching the PAR-CLIP donors. The black line (5 reads aligned) depicts the detection limit of the PAR-CLIP assay with majority of the loci on the sense genome are expected to surpass the detection limit. Upper and lower panels correspond to the reads on the sense and anti-sense strand of HIV-1JR-FL genome respectively.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4520458&req=5

pone.0132127.g002: Cellular micro RNAs as detected by Ago2 PAR-CLIP and small RNA sequencing.(A) Percentage of the cellular miRNAs from small RNA sequencing that was also found in Ago2 PAR-CLIP as a function of minimum expression threshold. The x-axis shows the total abundance of miRNAs in small RNA sequencing data derived from pooled HIV-1JR-FL infected (n = 4) and HIV-1 non-infected (n = 4) samples from the same donors. Red lines show the read count corresponding to specific fractions of total miRNA pool (B) Expression of cellular miRNAs in Ago2 PAR-CLIP and small RNA sequencing data is well correlated for 160 miRNAs found by both methods (R = 0.55, p<10‒13). The dashed lines show bootstrap-derived 95% confidence intervals for the linear fit (red line). (C) Expected PAR-CLIP read counts of virus-derived sncRNAs associated with Ago2. Adjusted to small RNA sequencing data derived from pooled HIV-1JR-FL infected (n = 2) samples matching the PAR-CLIP donors. The black line (5 reads aligned) depicts the detection limit of the PAR-CLIP assay with majority of the loci on the sense genome are expected to surpass the detection limit. Upper and lower panels correspond to the reads on the sense and anti-sense strand of HIV-1JR-FL genome respectively.
Mentions: Although highly abundant in small RNA sequencing, HIV-1 sncRNAs were almost completely absent in the Ago2 PAR-CLIP libraries. In order to evaluate the quality of our PAR-CLIP assays, we focused our analysis on the known miRNAs, assuming that the majority of expressed miRNAs can be associated with Ago2-RISC and hence should be present in Ago2 PAR-CLIP data. Pooled Ago2 PAR-CLIP and small RNA sequencing libraries captured 183 and 475 unique mature miRNAs that were present by at least five reads. 160 miRNAs were detected by both assays. The relative sensitivity of the Ago2 PAR-CLIP assay was assessed as a function of miRNA abundance in the small RNA sequencing experiment. We observed a strong correspondence between miRNA frequency in total miRNA library (n = 4) and its detection by Ago2 PAR-CLIP (n = 4) (Fig 2A). 88% of all cellular miRNAs—detected with ≥1,000 reads in small RNA sequencing—were also observed in PAR-CLIP (Fig 2A). This rate dropped only to 64% and 38% when ≥100 and ≥10 reads, respectively, were chosen as cut-offs. Even more than 30% of low abundant miRNAs, i.e., detected with only ≥5 reads, were identified in the PAR-CLIP data (Fig 2A). Furthermore, the expression level of the miRNAs was highly correlated between the two methods for the 160 miRNAs detected in both datasets (R = 0.55, p<10‒13) (Fig 2B). Given the strong concordance of the two methods in detecting cellular miRNAs we expected potential viral miRNAs or target sites—if associated with Ago2—to be also present in the Ago2 PAR-CLIP data. Fig 2C demonstrates HIV-1 reads observed in small RNA sequencing for the two PAR-CLIP matched donors, and indicates indicates the expected read count in our PAR-CLIP data for authentic Ago2 associated HIV-1 sncRNAs based on our observations using cellular miRNA data (Fig 2A and 2B). Despite the detection of four highly abundant HIV-1 sncRNAs (>500 reads) and 107 abundant HIV-1 sncRNAs (>100 reads) by small RNA sequencing none of them were present in the PAR-CLIP data (Fig 2C). Only one HIV-1 sncRNA was observed in both, the small RNA sequencing and the PAR-CLIP data (Fig 2C), but this HIV-1 sncRNA corresponded to the tRNALys primer binding site and was also detected in HIV-1 non-infected samples (Fig F in S1 File). Moreover, the highly abundant reads on the sense strand of the virus, equivalent to 9.1% of miRNA pool, in the small RNA sequencing data was entirely absent in the PAR-CLIP and hence it is highly unlikely that HIV-1 sncRNAs were missed due to the detection limit of the PAR-CLIP assay.

Bottom Line: HIV-1 infected monocyte-derived macrophages (MDM) were chosen as target cells, as they have previously been shown to express HIV-1 sncRNAs.Small RNA sequencing revealed that 1.4% of all sncRNAs were of HIV-1 origin.However, neither HIV-1 derived sncRNAs nor putative HIV-1 target sequences incorporated into Ago2-RISC were identified suggesting that HIV-1 sncRNAs are not involved in the canonical RNAi pathway nor is HIV-1 targeted by this pathway in HIV-1 infected macrophages.

View Article: PubMed Central - PubMed

Affiliation: University Hospital Zurich, Division of Infectious Diseases and Hospital Epidemiology, University of Zurich, Zurich, Switzerland; Institute of Medical Virology, University of Zurich, Zurich, Switzerland.

ABSTRACT

Background: MiRNAs and other small noncoding RNAs (sncRNAs) are key players in post-transcriptional gene regulation. HIV-1 derived small noncoding RNAs (sncRNAs) have been described in HIV-1 infected cells, but their biological functions still remain to be elucidated. Here, we approached the question whether viral sncRNAs may play a role in the RNA interference (RNAi) pathway or whether viral mRNAs are targeted by cellular miRNAs in human monocyte derived macrophages (MDM).

Methods: The incorporation of viral sncRNAs and/or their target RNAs into RNA-induced silencing complex was investigated using photoactivatable ribonucleoside-induced cross-linking and immunoprecipitation (PAR-CLIP) as well as high-throughput sequencing of RNA isolated by cross-linking immunoprecipitation (HITS-CLIP), which capture Argonaute2-bound miRNAs and their target RNAs. HIV-1 infected monocyte-derived macrophages (MDM) were chosen as target cells, as they have previously been shown to express HIV-1 sncRNAs. In addition, we applied small RNA deep sequencing to study differential cellular miRNA expression in HIV-1 infected versus non-infected MDMs.

Results and conclusion: PAR-CLIP and HITS-CLIP data demonstrated the absence of HIV-1 RNAs in Ago2-RISC, although the presence of a multitude of HIV-1 sncRNAs in HIV-1 infected MDMs was confirmed by small RNA sequencing. Small RNA sequencing revealed that 1.4% of all sncRNAs were of HIV-1 origin. However, neither HIV-1 derived sncRNAs nor putative HIV-1 target sequences incorporated into Ago2-RISC were identified suggesting that HIV-1 sncRNAs are not involved in the canonical RNAi pathway nor is HIV-1 targeted by this pathway in HIV-1 infected macrophages.

No MeSH data available.


Related in: MedlinePlus