MicroRNA titration by pseudo-targets: an alternative ... · PDF filetitration by pseudo-targets Introduction An alternative hypothesis ... miRNA quanti cation mRNA quanti cation Relative

MicroRNAtitration by

pseudo-targets

Introduction

An alternativehypothesis

Testing thehypothesis

miRNA quantification

mRNA quantification

Relative abundance ofmiRNAs and mRNAs

Conclusion

Supplementarydata

MicroRNA titration by pseudo-targets: analternative explanation to seed match

conservation

Herve Seitz

IGH (CNRS), Montpellier, France

February 2, 2014


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

microRNA target identification

.

||||||NNNNNNN NNNNNNNNNNNNNN

5´ 3´miRNA:

target:

2 7

−→ the “seed”


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


.

||||||NNNNNNN NNNNNNNNNNNNNN

5´ 3´miRNA:

target:

2 7

−→ the “seed”


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Computational programs for target prediction: look for seedmatches in 3´ UTRs, select the ones that were conserved inevolution.

Such short matches are very frequent (60 % of humancoding genes seem to be targeted: Friedman et al., 2009).

=⇒ miRNAs are implicated in every physiological process inanimals.

miRNA-mediated repression is very modest (usually< 2-fold): lower than well tolerated fluctuations in geneexpression (e.g., haplosufficiency). Why have these sitesbeen conserved if they are not functional?


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata







pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata







pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata







pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

An alternative hypothesis

microRNA

mRNA

mRNA

.


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


.

mRNA

mRNA

microRNA


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


pseudo−target (insensitive)

real target (sensitive)

mRNA

mRNA

microRNA

.


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Absolute RNA quantification

For mRNAs to titrate miRNAs, they need to be abundantenough. Issues with published examples

Murine cell line “C2C12”: differentiate in myotubes;expresses muscle-specific miR-1, miR-206 and miR-133.

miR-1 and miR-206 (at least) control C2C12 differentiation(Goljanek-Whysall et al., 2012).

=⇒ quantification of miR-1, miR-206, miR-133 and theirpredicted targets in differentiating C2C12.

I miRNAs: Northern blot (calibrated with RNA oligos);

I mRNAs: RNA-Seq (calibrated with in vitro transcripts).


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata









pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata









pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata





=⇒ absolute quantification of miR-1, miR-206, miR-133 andtheir predicted targets in differentiating C2C12.




pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata









pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata









pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata









pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


M10.

5

2 4 8 15 20 0 1 2 3 4 5 6

fmol synthetic oligo days of differentiation

18

2123

38

nt:

(quantification of miR-1 and miR-206)


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


0 1 2 3 4 5 6

05000

10000

15000

20000

25000

miRNA abundance during C2C12 differentiation

Day of differentiation

miR

NA

mole

cule

s p

er

cell

miR−1 + miR−206

miR−133


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

mRNA quantification


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

mRNA quantification

Very deep sequencing: three time points (day 0, day 3, day6) in triplicate; each replicate: between 267 and 333 milliontranscriptome-matching reads. Statistics


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

mRNA quantification

Very deep sequencing: three time points (day 0, day 3, day6) in triplicate; each replicate: between 267 and 333 milliontranscriptome-matching reads. Statistics

27 synthetic spike-ins, for calibration.


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

mRNA quantification

0 2000 4000 6000 8000

01000

2000

3000

4000

Day 0, replicate 1

Molecules per cell

Read a

bundance (

fpkm

)


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

mRNA quantification

0 2000 4000 6000 8000

010

0020

0030

0040

00Day 0, replicate 1

Molecules per cell

Rea

dab

unda

nce

(fpkm

)

miR-1/miR-206 sitesmiR-133 sites


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

mRNA quantification

Day

Mol

ecul

espe

rcel

l

0 1 2 3 4 5 6

020

000

4000

060

000

8000

0

miR-1 + miR-206predicted targets


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

mRNA quantification

Day

Mol

ecul

espe

rcel

l

0 1 2 3 4 5 6

020

000

4000

060

000

8000

0

3.2-fold excessof targets

miR-1 + miR-206predicted targets


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

mRNA quantification

Day

Mol

ecul

espe

rcel

l

0 1 2 3 4 5 6

010

000

3000

050

000

miR-133predicted targets


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

mRNA quantification

Day

Mol

ecul

espe

rcel

l

0 1 2 3 4 5 6

010

000

3000

050

000

15.3-fold excessof targets

miR-133predicted targets


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Concentration effects


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata



pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata



pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Predicted targetsmiRNA family Total Above 10×Kd

miR-1 and miR-206 436 31 (7 %)

miR-133 384 22 (6 %)

(Kd value taken from Wee et al., 2012: 26 pM)


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Titration effect of individual miRNA binding sites:

mRNA Sample Increase in free miR-1/miR-206if site is lost

Anxa2 Day 0 8 %Ptma Day 0 9 %

Tmsb4x Day 0 10 %

Actb Day 3 15 %Tmsb4x Day 3 9 %

Actb Day 6 13 %Tmsb4x Day 6 11 %


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Titration effect of individual miRNA binding sites:

mRNA Sample Increase in free miR-133if site is lost

Tpm4 Day 0 5 %Ptma Day 0 16 %Eif4a1 Day 0 5 %Tagln2 Day 0 6 %

Ptma Day 3 7 %Eef1a1 Day 3 8 %

Ptma Day 6 6 %Eef1a1 Day 6 6 %

Titrator expression profiles


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Conclusion

miRNA titration by mRNAs does occur.


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Conclusion



pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Conclusion


Every cell where miR-1, miR-206, miR-133 titration isbeneficial may express a different set of miRNA titrators.

=⇒ It cannot be assumed that every conserved binding sitemediates functional target repression.

The number of miRNA targets has probably beenoverestimated.


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Conclusion






pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Conclusion






pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Conclusion






pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Acknowledgements

Anna Laura Natalia BlaiseSergeeva Martinez Pinzon Restrepo Li


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Acknowledgements

Anna Laura Natalia BlaiseSergeeva Martinez Pinzon Restrepo Li


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Supplementary data

An alternative hypothesis:Robustness of biological pathways

Non-discriminative predictions (coordinated targeting)

Non-discriminative predictions (expression domains)

Non-discriminative predictions (house-keeping genes)

Testing the hypothesis:RNA-seq technical reproducibility RNA-seq biological reproducibility

Issues with published examples Experimental procedure

Amount of RNA per cell Differentiation controls

Cell volume Sequencing depth

Titrator expression profiles Abundance vs. conservation

Conclusion:Pseudo-targets for other regulators?

Propagation of gene expression perturbation


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Biological robustness

Substrate Product 1 Product 2 Product 3enzyme 1 enzyme 2 enzyme 3


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Non-discriminative predictions

Targets for a given miRNA often belong to the samebiological pathways.

mRNA for gene 1

mRNA for gene 4

mRNA for gene 3

mRNA for gene 2

miRNA


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Targets for a given miRNA often belong to the samebiological pathways.

mRNA for gene 1

mRNA for gene 4

mRNA for gene 3

mRNA for gene 2

miRNA


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Expression domains for miRNAs and their predicted targetsoverlap at their boundaries.

sharp boundary ofmRNA activity domain

miRNA

mRNAmiRNA

spatial or temporal axis

expressionand mRNA


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Expression domains for miRNAs and their predicted targetsoverlap at their boundaries.

sharp boundary ofmiRNA activity domain

miRNA

mRNA

miRNA

spatial or temporal axis

expressionmRNAand


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


House-keeping genes are rarely predicted to be targeted.

mRNA for house−keeping gene


miRNA mRNA for tissue−specific gene

cell type 1

miRNA mRNA for tissue−specific gene

cell type 2

avoidance

avoidance


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


House-keeping genes are rarely predicted to be targeted.



miRNA

miRNA

mRNA for tissue−specific gene

cell type 2

mRNA for tissue−specific gene

cell type 1

avoidance

avoidance


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Technical reproducibility of RNA-seq


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata



pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata



pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Biological reproducibility of RNA-seq


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata



pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata



pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Reported examples of pseudo-targets

A proposed pseudo-target: PTENP1, that de-repressedPTEN (Poliseno et al., 2010).


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


A proposed pseudo-target: PTENP1, that de-repressedPTEN (Poliseno et al., 2010).

But PTENP1 mRNA is ≈ 100 times less abundant than thePTEN mRNA (Ebert and Sharp, 2010).


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


PTEN mRNA(NM_008960)

1 869 2080 8229nt:

CDS

siRNAs against PTENP1


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Proposed by Poliseno et al (2010):

PTENP1 mRNA

PTEN mRNA

siRNAs

miRNAs

More probably:

siRNAs

PTEN mRNA


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


UAAUAAUCAUCAUUCUGGC |||||| ||| |||GUUAUUA----UAA-ACCU

5´ 3´

3´ 5´

siRNA #2 against PTENP1:

PTEN mRNA (nt 948-961):

UAAUAAUCAUCAUUCUGGC |||||| ||| :| |AUUAUUAUAUGUAUCGCGG

5´ 3´

3´ 5´PTEN mRNA (nt 1383-1401):


UAAUAAUCAUCAUUCUGGC |||||| ||||AUUAUUACC-----GACCU

5´ 3´

3´ 5´PTEN mRNA (nt 3769-3782):


UAAUAAUC-AUCAUUCUGGC |||||| |:| |::UUUAUUACUUGGAAAAAUUA

5´ 3´

3´ 5´PTEN mRNA (nt 8189-8208):


UCCUAUA--UGAUCUCUGAUG |||||| || |||||::AGGAUAUUGACGUUAGACUGU

5´ 3´

3´ 5´PTEN mRNA (nt 2192-2212):



pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Reported examples of pseudo-targetssiRNAs against AFF1


1 869 2080 8229nt:

CDS


1 869 2080 8229nt:

CDS

siRNAs against CNOT6L


1 869 2080 8229nt:

CDS

siRNAs against DCBLD2


1 869 2080 8229nt:

CDS

siRNAs against JARID2


1 869 2080 8229nt:

CDS

siRNAs against MBNL1


1 869 2080 8229nt:

CDS

siRNAs against TNRC6b

siRNAs against RBM9


1 869 2080 8229nt:

CDS


1 869 2080 8229nt:

CDS

siRNAs against TNRC6a


1 869 2080 8229nt:

CDS

siRNAs against ZEB2

Return


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


C2C12 cells

0 to 6 days of differentiation

Amount of RNA per cell Return

Differentiation controls


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


C2C12 cells


Flowcytometry

MicroscopyRNAextraction

Numberof cells

Cellvolume

miRNAand mRNA

quantification




pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


C2C12 cells


Flowcytometry


Numberof cells

Cellvolume

miRNAand mRNA

quantification




pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


C2C12 cells


Flowcytometry


Numberof cells

Cellvolume

miRNAand mRNA

quantification




pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


C2C12 cells


Flowcytometry


Numberof cells

Cellvolume

miRNAand mRNA

quantification

(three biological replicates)




pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Number of cells

Qua

ntity

ofR

NA

(µg)

Day 0Day 1Day 2Day 3Day 4Day 5Day 6

050

100

150

200

250

300

0 0.5×107 1×107 1.5×107 2×107

Return


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Myogenin

Days

Exp

ressio

n (

fpkm

)

0 1 2 3 4 5 6

05

01

00

15

02

00

25

03

00

Myosin heavy chain polypeptide 1

Days

Exp

ressio

n (

fpkm

)

0 1 2 3 4 5 60

10

20

30

40

50

Return


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Cell volume

Cel

l vol

ume

(μm

3 )

Day 0 Day 3 Day 6

0

5000

10000

15000

20000

25000

30000

(calibrationproblem)


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Deep-sequencing statistics

Number of reads per kb (median transcript, excludingspike-ins):

Replicate 1 159.26Day 0 Replicate 2 140.53

Replicate 3 150.03


Replicate 3 198.48


Replicate 3 220.23

Return


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Day

Intra

cellu

larc

once

ntra

tion

(mol

/L)

0 1 2 3 4 5 6

0.0e

+00

1.0e

−08

2.0e

−08

Kd

miR-1 + miR-206Tmsb4x mRNA

Return


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Day

Intra

cellu

larc

once

ntra

tion

(mol

/L)

0 1 2 3 4 5 6

0.0e

+00

1.0e

−09

2.0e

−09

3.0e

−09

Kd

miR-133Tpm4 mRNA

Return


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

mRNA abundance and seed match conservation

0 1000 2000 3000 4000

0.0

0.2

0.4

0.6

0.8

1.0

Kendall's τ = 0.0873(p-value = 1.18 10 )-8

Gene expression in hypothalamus

CT

Con

serv

atio

n sc

ore

(P

)of

bin

ding

site

s to

miR

-17


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

mRNA abundance and seed match conservationA

djus

ted p-

valu

e

Hypothalamus

Kendall's τ-0.25 -0.1 0 0.1 0.25

1

10-4

10-8

10-12

10-16

Adj

uste

d p-

valu

e

Spleen

Kendall's τ-0.25 -0.1 0 0.1 0.25

1

10-4

10-8

10-12

10-16

Adj

uste

d p-

valu

e

Naïve B cells

Kendall's τ-0.25 -0.1 0 0.1 0.25

1

10-4

10-8

10-12

10-16

Adj

uste

d p-

valu

e

Kidney

Kendall's τ-0.25 -0.1 0 0.1 0.25

1

10-4

10-8

10-12

10-16

Adj

uste

d p-

valu

e

ES cells

Kendall's τ-0.25 -0.1 0 0.1 0.25

1

10-4

10-8

10-12

10-16

Adj

uste

d p-

valu

e

Ovary

Kendall's τ-0.25 -0.1 0 0.1 0.25

1

10-4

10-8

10-12

10-16


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

mRNA abundance and seed match conservation

LiverA

djus

ted p-

valu

e

Kendall's τ

1

10-4

10-8

10-12

10-16

-0.3 -0.1 0 0.1 0.3

miR-150miR-138

miR-181amiR-133a

miR-124

miR-9

miR-1a

miR-122


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata

Pseudo-targets for other regulators?

I Transcription factors: experimentally identified bindingsites are poorly conserved among vertebrates (Odom etal., 2007 and Schmidt et al., 2010).

I RNA-binding proteins are poorly specific (thousands ofexperimentally validated targets for each analyzedprotein: Hafner et al., 2010 and Lebedeva et al., 2011).

Real molecular events, which are neutral in evolutionaryterms?


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata






pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata






pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


+

−

Gene X

Gene Y1

Gene Y3

Gene Y2

Direct miRNA target Indirect miRNA targets

Gene Z1

Gene Z3

Gene Z2

Gene Z4

Gene Z6

Gene Z5

Gene Z7

Gene Z9

Gene Z8

miRNA


pseudo-targets

Introduction




mRNA quantification


Conclusion

Supplementarydata


Direct targets13.5 dpp

Indirect targetsat 13.5 dpp

Indirect targetsat 15.5 dpp

|log

2(fo

ld-c

hang

e)|

(blu

e bo

xplo

t)

Num

ber

of a

ffect

ed g

enes

(red

line

)

0

2

4

6

8

10

0

500

1000

1500

Zfy overexpression

Pebp4

Tsarg2 MiwiSept12

(in collaboration with H. Royo and J. Turner, MRC, London)

Documents

MicroRNA titration by pseudo-targets: an alternative ... · PDF filetitration by pseudo-targets Introduction An alternative hypothesis ... miRNA quanti cation mRNA quanti cation Relative