Limits...
Systemic epigenetic response to recombinant lentiviral vectors independent of proviral integration.

Aranyi T, Stockholm D, Yao R, Poinsignon C, Wiart T, Corre G, Touleimat N, Tost J, Galy A, Paldi A - Epigenetics Chromatin (2016)

Bottom Line: The effects of LV on target cells are expected to be limited to gene delivery.This effect required cellular entry of the viral particle in the cells but not the genomic integration of the vector cassette.Some LV preparations induced only mild sporadic changes while others had strong effects suggesting that LV batch heterogeneity may be related to the extent of the epigenetic response.

View Article: PubMed Central - PubMed

Affiliation: Université Evry Val d'Essonne, UMRS_951, Genethon, 91002 Evry, France.

ABSTRACT

Background: Lentiviral vectors (LV) are widely used for various gene transfer or gene therapy applications. The effects of LV on target cells are expected to be limited to gene delivery. Yet, human hematopoietic CD34+ cells respond to functional LVs as well as several types of non-integrating LVs by genome-wide DNA methylation changes.

Results: A new algorithm for the analysis of 450K Illumina data showed that these changes were marked by de novo methylation. The same 4126 cytosines located in islands corresponding to 1059 genes were systematically methylated. This effect required cellular entry of the viral particle in the cells but not the genomic integration of the vector cassette. Some LV preparations induced only mild sporadic changes while others had strong effects suggesting that LV batch heterogeneity may be related to the extent of the epigenetic response.

Conclusion: These findings identify a previously uncharacterized but consistent cellular response to viral components and provide a novel example of environmentally modified epigenome.

No MeSH data available.


Workflow of data analysis and representation. The normalized β-values of triplicate samples and their corresponding controls are analyzed using the DAT algorithm as described in the text. a First, a moving average <β> is calculated for each CpG position in each sample and control. The average of all six sample <β> values is calculated ≪β≫. b The methylation level of a given CpG is considered as “increased” in the samples compared to controls if all the <β> values are higher then ≪β≫ or decreased if lower then ≪β≫. The above operation is repeated for all CpG-s analyzed by the Illumina chip. c Three parameters are calculated on the basis of the lists of “increased” and “decreased” CpG-s: (1) the total number of CpG-s (N); (2) chi2 calculated on the basis of the observed distribution of the CpG cluster sizes compared to the simulated random distribution of cluster sizes obtained with an identical number of CpG sites; (3) the median delta-β value of all the N CpG-s
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4940770&req=5

Fig1: Workflow of data analysis and representation. The normalized β-values of triplicate samples and their corresponding controls are analyzed using the DAT algorithm as described in the text. a First, a moving average <β> is calculated for each CpG position in each sample and control. The average of all six sample <β> values is calculated ≪β≫. b The methylation level of a given CpG is considered as “increased” in the samples compared to controls if all the <β> values are higher then ≪β≫ or decreased if lower then ≪β≫. The above operation is repeated for all CpG-s analyzed by the Illumina chip. c Three parameters are calculated on the basis of the lists of “increased” and “decreased” CpG-s: (1) the total number of CpG-s (N); (2) chi2 calculated on the basis of the observed distribution of the CpG cluster sizes compared to the simulated random distribution of cluster sizes obtained with an identical number of CpG sites; (3) the median delta-β value of all the N CpG-s

Mentions: A new algorithm called “double average technique” (DAT) was designed to identify very small, but reproducible differences between two series of samples after normalization of the raw data (see the workflow of the data analysis and representation on Fig. 1). This highly sensitive method identifies only reproducible differences in cytosine methylation level present in all samples compared to their reference controls. Sporadic differences are filtered. The method is implemented on triplicate experiments requiring the comparison of 3 samples with 3 corresponding controls. Thus, after hybridization on the 450K bead chip, 6 β-values are obtained for each CpG site tested. The sliding average (<β>) of the β-values of 3 consecutive CpG-s for each sample (t1, t2 and t3) and for each of the corresponding control (c1, c2 and c3) are calculated. The methylation of a given CpG site is considered to be different between the sample and control triplicates if all three values of the sliding averages of the samples are above (t1, t2, t3 ≥ increased methylation) or below (t1, t2, t3 ≤ decreased methylation) of the arithmetic average of the 6 <β>-values [(t1 + t2 + t3 + c1 + c2 + c3)/6] of the samples and controls. This constraint confers to the algorithm a very high stringency and allows the identification of all CpG sites with increased or decreased methylation independently of the extent of the difference and without applying the same fixed threshold to all CpG-s on the chip. The list of these CpG-s provides information on the total number of modified cytosines, their position in the genome and the extent of the increase or decrease of the methylation at each site.Fig. 1


Systemic epigenetic response to recombinant lentiviral vectors independent of proviral integration.

Aranyi T, Stockholm D, Yao R, Poinsignon C, Wiart T, Corre G, Touleimat N, Tost J, Galy A, Paldi A - Epigenetics Chromatin (2016)

Workflow of data analysis and representation. The normalized β-values of triplicate samples and their corresponding controls are analyzed using the DAT algorithm as described in the text. a First, a moving average <β> is calculated for each CpG position in each sample and control. The average of all six sample <β> values is calculated ≪β≫. b The methylation level of a given CpG is considered as “increased” in the samples compared to controls if all the <β> values are higher then ≪β≫ or decreased if lower then ≪β≫. The above operation is repeated for all CpG-s analyzed by the Illumina chip. c Three parameters are calculated on the basis of the lists of “increased” and “decreased” CpG-s: (1) the total number of CpG-s (N); (2) chi2 calculated on the basis of the observed distribution of the CpG cluster sizes compared to the simulated random distribution of cluster sizes obtained with an identical number of CpG sites; (3) the median delta-β value of all the N CpG-s
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4940770&req=5

Fig1: Workflow of data analysis and representation. The normalized β-values of triplicate samples and their corresponding controls are analyzed using the DAT algorithm as described in the text. a First, a moving average <β> is calculated for each CpG position in each sample and control. The average of all six sample <β> values is calculated ≪β≫. b The methylation level of a given CpG is considered as “increased” in the samples compared to controls if all the <β> values are higher then ≪β≫ or decreased if lower then ≪β≫. The above operation is repeated for all CpG-s analyzed by the Illumina chip. c Three parameters are calculated on the basis of the lists of “increased” and “decreased” CpG-s: (1) the total number of CpG-s (N); (2) chi2 calculated on the basis of the observed distribution of the CpG cluster sizes compared to the simulated random distribution of cluster sizes obtained with an identical number of CpG sites; (3) the median delta-β value of all the N CpG-s
Mentions: A new algorithm called “double average technique” (DAT) was designed to identify very small, but reproducible differences between two series of samples after normalization of the raw data (see the workflow of the data analysis and representation on Fig. 1). This highly sensitive method identifies only reproducible differences in cytosine methylation level present in all samples compared to their reference controls. Sporadic differences are filtered. The method is implemented on triplicate experiments requiring the comparison of 3 samples with 3 corresponding controls. Thus, after hybridization on the 450K bead chip, 6 β-values are obtained for each CpG site tested. The sliding average (<β>) of the β-values of 3 consecutive CpG-s for each sample (t1, t2 and t3) and for each of the corresponding control (c1, c2 and c3) are calculated. The methylation of a given CpG site is considered to be different between the sample and control triplicates if all three values of the sliding averages of the samples are above (t1, t2, t3 ≥ increased methylation) or below (t1, t2, t3 ≤ decreased methylation) of the arithmetic average of the 6 <β>-values [(t1 + t2 + t3 + c1 + c2 + c3)/6] of the samples and controls. This constraint confers to the algorithm a very high stringency and allows the identification of all CpG sites with increased or decreased methylation independently of the extent of the difference and without applying the same fixed threshold to all CpG-s on the chip. The list of these CpG-s provides information on the total number of modified cytosines, their position in the genome and the extent of the increase or decrease of the methylation at each site.Fig. 1

Bottom Line: The effects of LV on target cells are expected to be limited to gene delivery.This effect required cellular entry of the viral particle in the cells but not the genomic integration of the vector cassette.Some LV preparations induced only mild sporadic changes while others had strong effects suggesting that LV batch heterogeneity may be related to the extent of the epigenetic response.

View Article: PubMed Central - PubMed

Affiliation: Université Evry Val d'Essonne, UMRS_951, Genethon, 91002 Evry, France.

ABSTRACT

Background: Lentiviral vectors (LV) are widely used for various gene transfer or gene therapy applications. The effects of LV on target cells are expected to be limited to gene delivery. Yet, human hematopoietic CD34+ cells respond to functional LVs as well as several types of non-integrating LVs by genome-wide DNA methylation changes.

Results: A new algorithm for the analysis of 450K Illumina data showed that these changes were marked by de novo methylation. The same 4126 cytosines located in islands corresponding to 1059 genes were systematically methylated. This effect required cellular entry of the viral particle in the cells but not the genomic integration of the vector cassette. Some LV preparations induced only mild sporadic changes while others had strong effects suggesting that LV batch heterogeneity may be related to the extent of the epigenetic response.

Conclusion: These findings identify a previously uncharacterized but consistent cellular response to viral components and provide a novel example of environmentally modified epigenome.

No MeSH data available.