Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

A new paramutation-like example at the Delta gene of Drosophila

  • Maria Capovilla ,

    Affiliation UMR 1355–7254 INRA/Université Côte d’Azur/CNRS, Institut Sophia Agrobiotech, 400 route des Chappes, Sophia Antipolis, France

  • Alain Robichon,

    Affiliation UMR 1355–7254 INRA/Université Côte d’Azur/CNRS, Institut Sophia Agrobiotech, 400 route des Chappes, Sophia Antipolis, France

  • Minoo Rassoulzadegan

    Affiliation Université Côte d’Azur, Inserm U1091–CNRS U7277, Nice, France

A new paramutation-like example at the Delta gene of Drosophila

  • Maria Capovilla, 
  • Alain Robichon, 
  • Minoo Rassoulzadegan


The hereditary transmission of a phenotype independent from DNA sequence implies epigenetic effects. Paramutation is a heritable epigenetic phenomenon observed in plants and animals. To investigate paramutation in Drosophila, we used the P{ry+t7.2 = PZ}Dl05151 P-element insertion in the Drosophila melanogaster genome that causes a dominant visible phenotype: the presence of characteristic extra-veins in the fly wings. This extra-vein phenotype presents variable expressivity and incomplete penetrance. The insert is a PZ element located 680 bp upstream from the ATG of the Delta (Dl) gene, encoding the Notch ligand involved in wing vein development, and acts as a null allele. In the G2 offspring from a cross between the heterozygous transgenic stock and wild-type flies, we observed the transmission of the extra-vein phenotype to wild-type flies without the transgene, independently of gender and across many generations. This is a “paramutation-like” example in the fly: the heritable transmission of a phenotypic change not linked to a classical genetic mutation. A “paramutagenic” allele in heterozygotes transmits the phenotype of the heterozygotes to the wild-type allele (“paramutant”) in a stable manner through generations. Distinct from paramutation events so far described in Drosophila, here we deal with a dominant effect on a single gene involving variable hereditary signals.


Changes in gene expression arise because of genetic and environmental variation in living organisms [1]. The extent to which these changes are maintained long enough to be transmitted to the next generation is still an open question. Analysis of model organisms largely highlights epigenetic determinants involved in this process.

Paramutation is a classical case of genetically-driven heritable epigenetic variation ([2] and references therein). Pioneer cases of plant paramutation refer to changes in the expressivity of alleles originating from heterozygotes (reviewed in [35]). Subsequently, the modified phenotype can be stably propagated in a number of generations. It is dependent upon RNA metabolism, even though, in the case of the maize booster 1 (b1) locus, DNA methylation of an upstream repeated motif is associated to the establishment of paramutation (reviewed in [47]). In addition to plants, several cases of paramutation events are described in laboratory mouse [8,9]. A different case of a white tail tip phenotype was observed in wild-type animals for the Kit locus, coming from heterozygous Kitm1Alf/Kit+ mice, carrying the Kitm1Alf lacZ insertion [10]. This phenotype is efficiently transmitted and results from a decrease in c-Kit messenger RNA levels with the accumulation of non-polyadenylated RNA molecules. A heritable white tail tip phenotype is also observed after microinjection into fertilized eggs of either total RNA from Kitm1Alf/Kit+ mice or fragments of c-Kit and cognate microRNAs [10]. More phenotypes were reported by microinjection of other microRNAs altering cellular size of the hearth and growth rate of the stem cells [11,12]. These are all cases of “paramutation-like” phenomena, the heritable transmission of a phenotypic change not linked to a classical genetic mutation and carried out by small noncoding RNAs.

Another case of transcriptional variation is transvection, initially discovered in the fly through somatic cell studies of the bithorax complex [13] and later in Neurospora crassa [14]. It is a pairing-dependent interaction between alleles (reviewed in [15,16]). More recently, in Drosophila, paramutation events involving the transmission of small non coding RNAs (piRNAs) with important silencing and maternal effects were associated with the transmission of a phenotype over 100 generations [1719]). siRNA silencing discovered in worms and plants is transmissible to further generations with some degrees of reversibility [20]. Finally, in laboratory worms, genetic-mediated variation of siRNAs promotes gene expression alterations and phenotypic changes, which transmission is correlated with heredity of the modified phenotype (reviewed in [21,22]).

In order to find a bona fide case of paramutation in Drosophila, we have searched for a dominant phenotype caused by a transgene in this model system. We report that the transgene P{ry+t7.2 = PZ}Dl05151 (hereafter called Dl05151) in the Delta (Dl) locus of Drosophila [23] produces a dominant extra-vein phenotype in the wing, especially around the posterior cross vein (pcv). The Drosophila wing is formed by ventral and dorsal epithelia sheets apposed to each other that are supported by longitudinal and transversal thickenings called veins that function as vessels hosting trachea, hemolymph, nerves and blood cells. Drosophila wing vein development is orchestrated by a number of transcription factors and signaling molecules (reviewed in [24]). Notch (N), together with its ligand Dl, restricts vein thickness [2529].

Starting from heterozygotes, the extra-vein phenotype could be followed in further generations even in the progenies that did not inherit the transgene. The Dl05151 allele acts as a paramutagenic allele that converts the wild-type allele to the parental state. This system provides a powerful model to study paramutation in Drosophila, a system almost lacking DNA methylation [3033]. Overall, our data indicate that an epigenetically-regulated phenotype could be derived from initial genetic events.

Results and discussion

P{ry+t7.2 = PZ}Dl05151 (hereafter Dl05151) is the insertion of a pP{PZ} element 681 base pairs upstream from the ATG of the Delta (Dl) gene of Drosophila (; [34]) (Fig 1A). The pP{PZ} transgene ( contains the lacZ coding region directed by the Hsp70 promoter, the rosy (ry) eye marker and plasmid sequences [35]. Dl05151 is an excellent example of enhancer trapping, the random insertion of reporter vectors in the genome of Drosophila. Its direction of transcription (Fig 1A) and pattern of lacZ expression (S1 Fig) are as for the Dl gene [36,37]. In the Drosophila developing wing, Dl is more highly expressed within the wing veins, while Notch (N) is more highly expressed in regions between the wing veins [29,37,38].

Fig 1. Map of the Dl gene and phenotype of Dl05151/TM3 wings.

(A) The Dl gene is located at cytological position 92A1-92A2 and between nucleotides 19,302,731 and 19,326,217 of the 3R arm. It has three annotated transcripts transcribed from right to left and is flanked by transcripts CG44174 and CR44175. Green: gene extent of Dl. Beige: translated exons. Purple: transcribed exons. Horizontal black lines: introns. This map is extracted from Release R6.08 of the Drosophila genome sequence in FlyBase. The Dl05151 insertion (black triangle) is located 681 bp upstream from the Dl ATG and is transcribed from right to left. (B) Eye of a ry-/ry- fly. (C) Left wing of a ry-/ry- female. (D) Right wing of a ry-/ry- male. (E) Eye of a heterozygous Dl05151/TM3 fly. The transgene carrying the ry+ gene, the eye color is as in wild-type. (F) Right wing of a Dl05151/TM3 female. Arrows point to expanded and extra-veins (EXP+EV). (G) Right wing of a Dl05151/TM3 female presenting two extra-veins (EVs). (H) Wing of a Dl05151/TM3 heterozygous female presenting one extra-vein perpendicular to the posterior cross vein (pcv) (EVpcv). (I) Example of an extra-vein in the L4-L5 inter-vein region (EVint). (J) Wild-type wing. Hereafter, wing orientations are according to their origin: left wings with distal tip to the left and right wings with distal tip to the right.

The Dl05151 transgene in the Dl gene of Drosophila has been inserted in the ry506 (hereafter ry-) background ([23]; Fig 1B). Fig 1C and 1D show examples of wild-type ry-/ry- wings. As Dl05151 carries the ry marker, the flies carrying it have ry+ eyes (Fig 1E). Dl05151 causes a dominant phenotype: expanded and extra-veins as shown by Dl05151/TM3, ryRK, Sb (hereafter TM3) flies (Fig 1F–1I). This phenotype displays highly variable expressivity including: extra-veins (EV) and expanded veins (EXP; Fig 1F), two EVs emerging from the posterior cross vein (pcv) and from L5 going anteriorly and posteriorly (EVs; Fig 1G and 1K), one EV perpendicular to the pcv (EVpcv) going posteriorly (Fig 1H and 1L), and an EV in the L4/L5 inter-vein region (EVint; Fig 1I and 1M). However, few wild-type wings are also observed (Fig 1J and 1N). The penetrance is higher in females than in males, but incomplete in both (S1 Table). This insertion behaves as a null Dl allele being homozygous lethal and when heterozygous with either the DlX [39] or the DlRevF10 [40] Dl amorph alleles. Indeed, the phenotypes of heterozygous DlX and DlRevF10 (S2 Fig) are extra-veins very similar to those of the Dl05151 heterozygous flies (Fig 1), indicating that the extra-vein phenotype of Dl05151 heterozygotes is due to a strong decrease in Dl expression.

The extra-vein phenotype of heterozygous Dl05151 flies being due solely to the P{ry+t7.2] = PZ}Dl05151 insertion was confirmed through the perfect excision of the transgene by crossing Dl05151/TM3, ryRK, Sb (hereafter TM3) females with males carrying the Δ2–3 transposase [41] (see Fig 2A for crosses and Materials and Methods for details). One hundred and ninety-six Dl05151/P{ry[+t7.2 = Delta2-3}99B males were crossed individually to Ly/TM3, ry-, Sb females and 134 excisions over TM3, ry-, Sb stocks were established and analyzed. Seventy independent excisions were obtained and they were all homozygous lethal. To test whether the excised chromosomes carry a secondary lethal, 11 excised stocks were crossed to the Dl Df(3R)BSC850 deficiency. Of these 11 stocks, seven were viable when hemizygous against the Dl deficiency. Using the oligonucleotides shown in Fig 2B, we get evidence that at least two out of these seven (168m1 and 169m2) do not carry the Dl05151 insert (Fig 2C) and display a normal sequence. All the seven perfectly excised lines present a full rescue of the dominant wing phenotype after outcrossing to ry- (Fig 2D–2G). Thus, the chromosome carrying the Dl05151 insert bears a secondary lethal and clearly the Dl05151 insertion is the cause of the extra-vein phenotype (Fig 1).

Fig 2. Perfect excision of the Dl05151 insert.

(A) Crosses carried out to obtain perfect excisions of the Dl05151 transgene. Δ2–3 is the chromosome carrying the transposase, Vno and Ly are dominant wing markers, {ry+} indicates that the flies have ry+ eyes because they carry the Dl05151 insertion, while {ry-} indicates that the flies have ry- eyes because they are devoid of the Dl05151 transgene. DlΔ05151 are the chromosomes in which an excision of Dl05151 occurred. At least seven excised hemizygous viable and rescued stocks we obtained. (B) Sizes and map of the amplicons obtained by PCR analysis. Oligonucleotides 003 and 004 are downstream and upstream of the Dl05151 transgene, respectively. Oligonucleotides 006 and 011 are in the 5’ and 3’ of the transgene, respectively. (C) 168m1 and 169m2 are two independent hemizygous viable ry- lines in which excision of Dl05151 occurred through Δ2–3 transposase-mediated excision. As shown by the absence of amplification with the couples of oligonucleotides at the 3’ (red line) and 5’ (yellow line) ends of the transgene, these two strains do not carry the transgene. Sequencing of the Dl amplicons (green line) shows that the excisions are perfect. (D-G) Wild-type wings of DlΔ05151/ry- females for the two stocks 168m1 and 169m2 as indicated.

To investigate a paramutation event associated with the Dl05151 insert, five Dl05151/TM3 heterozygous males were individually crossed to homozygous ry-/ry- flies. The wings of first generation (G1) Dl05151, ry-/Dl*, ry- (hereafter, Dl05151/Dl*; where Dl* indicates the ry chromosome that is in heterozygosis with Dl05151 in the G1) heterozygotes present an extra-vein phenotype with highly variable expressivity as in the parental Dl05151/TM3 stock (Fig 3). Occasionally, the pcv presents a gap (GAP; Fig 3F). The phenotype is dominant and G1 flies show incomplete penetrance as for the parental Dl05151/TM3 stock (Table 1).

Fig 3. Wing phenotypes of heterozygous Dl05151/Dl* females.

Wings of the Dl05151/Dl+ progeny of an out cross of Dl05151/TM3 males to ry-/ry- females. A whole spectrum of extra-vein phenotypes is found. (A) Expanded and extra-veins (EXP+EVs). (B) Two extra-veins (EVs). (C) One extra-vein emerging from the pcv (EVpcv). (D) An extra-vein in the inter-vein region (EVint). (E) A dot in the inter-vein region (arrow; DOT). (F) An anterior gap of the pcv (GAP). The spectrum of phenotypes is very similar to that of the Dl05151/TM3 stock.

Table 1. Quantification of the extra-vein phenotypes in the G0, G1 and G2.

In order to obtain wild-type flies without the Dl05151 insertion, G1 Dl05151/Dl* females (Fig 4A and 4B) and males where crossed in mass to homozygous ry-/ry- flies (see S4 Fig for crosses). In the G2 progeny, it was observed a variable percentage of Dl*, ry-/Dl+, ry- (where Dl+ indicates the ry- chromosome wild-type for Dl; hereafter Dl*/Dl+) flies with different degrees of extra-veins (Figs 4D and 5). We tested by PCR whether these paramutant flies carry the Dl05151 transgene, using the same oligonucleotide combinations as described above (Fig 2B). A representative PCR experiment for the fly shown in Fig 4D clearly demonstrates that this fly does not carry the Dl05151 transgene (Fig 4E, lanes 1 and 4). Fig 5 shows examples of the observed phenotypes of Dl*/Dl+ females including gaps (Fig 5A and 5B), extra-veins perpendicular to the pcv (Fig 5B and 5C), extra-veins in the inter-vein region (Fig 5A and 5D) and small dots (Fig 5E). This weak phenotype and extra-veins in the L4/L5 inter-vein region are also observed in the ry-/ry- stock at a considerably lower frequency (Table 1; S3 Fig). Many wild-type flies are also found (Fig 5F). The penetrance rate of the extra-vein phenotype in the total G2 Dl*/ Dl+ wings is higher than in the wings of the ry-/ry- background and this difference is highly significant (Table 1). The penetrance rate is significantly higher in the offspring of G1 Dl05151/Dl* females compared to the offspring of G1 Dl05151/Dl* males (χ2 = 19.8; df = 1; p<0.001 and χ2 = 12.96; df = 1; p<0.001 for females and males, respectively), possibly because the phenotype is more penetrant in females than in males carrying the insert. Detailed data for each individual male are reported in S2 Table.

Fig 4. Paramutation at the Dl locus.

Eye (A) and wing (B) of a heterozygous G2 Dl05151/Dl+ fly. The wing shown presents a frequent phenotype: two extra-veins near the pcv (arrows). Eye (C) and wing (D) of a Dl05151, ry-/Dl*, ry- (hereafter, Dl05151/Dl*) fly. This G2 fly does not carry the Dl05151 insert, but shows an extra-vein perpendicular to the pcv (arrow) indicating that its Dl+ (Dl*) chromosome coming from the heterozygous Dl05151/Dl* parent has been paramutated. (E) PCR on the genomic DNA of the fly carrying the wing shown in Fig 4D. The oligonucleotides used are shown in Fig 2B. In the fly carrying the transgene (Dl05151/Dl+), oligonucleotides 003 and 011 (red arrow) and 004 and 006 (yellow arrow) amplify the 3’ and 5’ ends of the transgene, respectively. No amplification with these two couples of primers is seen for the paramutant Dl*/Dl+ fly (red and yellow circles, respectively). Control amplification with oligonucleotides 003 and 004 (green arrows) in the Dl gene occurs for both flies.

Fig 5. Variability in the expressivity of the extra-vein phenotypes in the wings of the G2 Dl*/Dl+ flies.

In the G2, paramutant Dl*/Dl+ flies exhibit a wide range of wing vein phenotypes. (A) An extra-vein in the inter-vein region together with a pcv gap (EVint+GAP). (B) An extra-vein perpendicular to the pcv together with a pcv gap (EVpcv+GAP). (C) An extra-vein perpendicular to the pcv (EVpcv), (D) An extra-vein in the inter-vein region (EVint), (E) A faint dot (arrow; DOT). (F) A percentage of wings are wild-type (WT).

The ry- chromosome that has been in contact with the Dl05151 transgene (indicated as Dl*) has been “paramutated”, leading to the same phenotype as that caused by the insertion. This result is reminiscent of paramutation at the mouse Kit locus [10], where homozygous G2 wild-type mice from Kittm1Alf/+ heterozygotes exhibited the white feet and white tail phenotype of heterozygous parents carrying a lacZ insertion in the Kit locus (Kittm1Alf). Both events call for the transmission of a visible character through epigenetic mechanisms modifying gene expression.

It is worth noticing that the Dl05151/Dl+ G2 ry+ offspring carrying the Dl05151 insertion shows a stronger phenotype than its parents (Fig 1). Often, many extra veins are observed (Fig 6A–6D). Other frequent phenotypes are the EVpcv (Fig 6E) and the EVint (Fig 6F). This shows that the phenotype segregates with the insertion and suggests that the paramutagenic allele continues to produce signals that amplify the phenotype.

Fig 6. Strong expressivity of the extra-vein phenotype in the heterozygous G2 Dl05151/Dl+ flies.

In the G2, Dl05151/Dl+ flies present wings with many extra-veins. The phenotype ranges from two or more extra-veins of various lengths (A-D), to one extra vein perpendicular to the pcv (E), to an extra-vein in the inter-vein region (F).

We tested whether paramutation requires pairing through balancer chromosomes, which prevent pairing, as they are rich in inversions. Paramutation likely requires pairing as the extra-vein phenotype is not transmitted to the TM3 balancer chromosome in the G1 and is rarely transmitted to the TM2, ry balancer (S3 Table). Nevertheless, it is also possible that balancer chromosomes carry suppressors that mask the phenotype.

To assess a possible transgenerational effect, the TM2, ry balancer was used to mark the Dl* chromosome in the G3. G2 females and males Dl*/TM2, ry were crossed to each other and the wing phenotypes assessed in the G3 offspring (Table 2). The wing phenotype observed was predominantly the weak one of dots, although a few pcv and inter-vein extra veins were observed, especially in Dl*/Dl* homozygotes. Overall, the rate of wing defects in Dl*/TM2 and Dl*/Dl* flies is significantly higher than in the ry-/ry- background (Table 2) and higher than in the G2 (compare Table 2 with Table 1). This shows that paramutation is carried at least to the third generation.

It has long been known that a decrease in Dl function leads to thickened veins [25,26]. In our work, it seems that Dl is very sensitive to gene expression levels. An even slight decrease in Dl expression causes a visible extra-vein phenotype, possibly interfering with the multiple pathways governing wing vein formation [24]. Reduction or up-regulation of several components of these pathways affect wing vein width ranging from no veins to expanded or extra-veins. N and Dl belong to the “thickened” group of genes affecting lateral inhibition [25] and control vein width possibly through mis-regulation of rhomboid (rho) [28] and veinlet (ve) [29]. Interactions of N with many other gene products are known [29,42,43]. Our data suggest that the transgene has modified Dl expression on the homologous chromosome through a transmissible system that potentiates itself throughout generations producing a very variable wing vein phenotype.

So far, many cases of paramutation and transvection observed in a variety of organisms involve the activity of ncRNAs [44,45]. Direct microinjection into the naive fertilized eggs of total RNAs from the sperm of genetically [10], environmentally [46,47] or metabolically [48,49] stressed animals transfers a phenotype that becomes subsequently hereditary throughout generations. These epigenetic studies are illustrations for the role of ncRNAs in the transmission of information to next generations. Although the direct demonstration of these phenomena in the fly is still missing, our results provide strong evidence for a general type of subtle gene expression variation phenomenon through a paramutation-like process.

To test the capacity of the paramutant Dl* chromosome to become paramutagenic, the extra-vein phenotype was followed through 12 generations of the Dl*/Dl+ stocks from three independent G0 males. The full penetrance and high expressivity of the paramutant phenotype in the G12 females and males (Fig 7) strongly suggest that the Dl* allele has become stable. The extra-vein of the G12 progeny appears as a real emerging vein (Fig 7B and 7D). This last observation suggests that a paramutant-like phenotype could be selected in an isolated colony of flies recalling conditions for evolution.

Fig 7. Wing phenotypes in the G12 progeny of the paramutants.

In the G12 of inbred Dl*/Dl+ flies, the extra-vein phenotype showed complete penetrance in both females and males and higher expressivity than in the G2 and G3. In this figure, is shown the most frequent phenotype: a thick extra-vein in the inter-vein region (A-D) sometimes emerging from the pcv (B and D).

It is difficult to extend these observations to humans without a defined genetic example model. However, many human diseases have not yet been linked to genetic causes even by recent genome-wide association studies investigating thousands of genomic markers (reviewed in [50,51]). Transgenerational epigenetic heredity is known to occur in mammals [5154], affecting gene expression. Mis-regulation of essential genes due to epigenetic causes may be responsible for many human diseases not linked to genetic aberrations. Indeed, some of the missing heritability of human diseases may be explained by the transgenerational transmission of epigenetic information through non-Mendelian inheritance as shown by epigenome-wide association studies [54,55]. The transmission of epigenetic marks from one generation to another through the gametes has been clearly reported so far in model organisms. However in humans, we can find clear genetic aberrations that involve repeats (CGG) and transposable elements [52,56,57] and those seem to rely on DNA methylation and RNAs [58]. An involvement of these players in human diseases and cancer has been recently proposed [52,53,57].

Here we show that the Drosophila melanogaster model system, almost devoid of DNA methylation [30,31,33] (a process that has been shown to be associated to paramutation in plants [59]), can be a useful tool to deeply dissect the molecular mechanisms governing non-Mendelian heredity [60].

Materials and methods

Crosses and wing analysis

Flies were grown at 23–25°C in standard cornmeal medium with melassa as a source of sugar. The ry506 stock utilized was a kind gift of Anna Digilio (Università di Napoli, Napoli, Italy). The P{ry+t7.2 = PZ}Dl05151 stock is Bloomington stock n. 11651. Flies showed a decrease in the expressivity of the extra-vein phenotype through the span of a few years, but the same stocks ordered at different times within the last year show the same variation in penetrance and a high variation in the expressivity of the phenotype. Df(3R)BSC850 is w1118; Df(3R)BSC850/TM6C, Sb1 cu1 (Bloomington stock n. 27922). DlX and DlRevF10 are Bloomington stocks n. 60336 and n. 6300, respectively.

For the phenotypic analysis of the excisions of the P{ry+t7.2 = PZ}Dl05151 P-element, 20 Dl05151/TM3, ryRK, Sb females were crossed to ten y1 w1; Ki1 P{ry+t7.2 = Delta2-3}99B (Bloomington stock n. 4368) males and 134 stocks with ry- eyes (DlΔ05151/TM3, ryRK, Sb) were established by crossing individually each Dl05151/Ki1 P{ry+t7.2 = Delta2-3}99B male to Ly/TM3, ryRK, Sb females. DlΔ05151/TM3, ryRK, Sb stocks were all homozygous lethal. The Dl deficiency Df(3R)BSC850 was used to obtain hemizygous Dl05151-excised animals. Seven out of eleven DlΔ05151/Df(3R)BSC850 stocks analyzed were viable. To test for wing rescue, DlΔ05151/TM3, ryRK, Sb males were outcrossed to ry females and wings of the DlΔ05151/ry- progeny analyzed. The w1118; kar2, ry506, Vno/TM3, ryRK, Sb, delta2-3 and Ly1, {FRT80B, ry+}, kar2, ry506 TM3, ryRK, Sb stocks were kind gifts of Pascal Heitzler (IBMP, Strasbourg, France).

For the paramutation crosses, five individual Dl05151/TM3, ryRK, Sb males were crossed to ry-/ry- females and 1–20 females or males of the G1 Dl05151/ry- progeny were crossed to ry-/ry- flies. The G2 progeny analyzed is shown in Table 1 and S2 Table. Penetrance rates of extra-veins in the G2 Dl*/Dl+ and in the ry-/ry- flies were compared using simple χ2 tests. To obtain the G3 balanced with the TM2, ry- balancer, G1 Dl05151/Dl* females and males were crossed to TM2, ry-/MKRS flies. The resulting G2 Dl*/TM2, ry- females and males were inbred. G1 and G2 data are available upon request.

Wings that looked damaged were discarded. Wings were dissected from the adults using fine Dumont forceps, immersed in 100% Ethanol and mounted in Euparal (BioQuip). Images were acquired thanks to the SPIBOC imaging facility of the Institut Sophia Agrobiotech on a LEICA (Switzerland) MZFLIII stereomicroscope using the ZEN software (ZEISS).

DNA analysis

DNAs were extracted with phenol and ethanol precipitated. Single fly DNA was resuspended in 20 μl of 10 mM Tris pH7.5, 1 mM EDTA (TE). Ten nanograms of DNA were used for each PCR. For PCR reactions, the MyTaq (Bioline, France) was used in 50 μl reactions according to instructions. Oligonucleotides were used at a 10 mM dilution and their sequences are as follows: 003, 5’-CGGAGTCTTCTCCTTTTCACG-3’; 004, 5’-AAATAAACACCCATCCGGTTGAAG-3’; 006, 5’-TAATAGCACACTTCGGCACG-3’; 011, CAATCATATCGCTGTCTCACTCAG-3’. The cycles were: one minute denaturation at 95°C followed by 35 cycles of 15 sec denaturation at 95°, 15 sec annealing at 57°C and 30 sec extension at 72°C. Reactions ended with 5 min extension at 72°C.

Supporting information

S1 Fig. lacZ expression in Dl05151/TM3 imaginal disks.

Imaginal disks of a Dl05151/TM3 female were stained for lacZ expression by X-gal coloration. lacZ is expressed in the gut (A), in the brain and in the antenna and eye disk (B), in the wing disk (C), and in the leg disks (D).


S2 Fig. Wing phenotypes of DlX and DlRevF10 heterozygous flies.

(A) Right wing of a DlX/TM3, Sb1 female. (B) Right wing of a DlX/TM3, Sb1 male. (C) Right wing of a DlRevF10, e*, SerRX82, P{ry+t7.2 = neoFRT}82B/TM6B, Tb1 female. (D) Right wing of a DlRevF10, e*, SerRX82, P{ry+t7.2 = neoFRT}82B/TM6B, Tb1 male. These wings show extra-veins very similar to those of Dl05151 heterozygotes (Figs 1 and 3).


S3 Fig. Weak wing phenotype of the ry-/ry- background.

In a small percentage of flies of the ry-/ry- stock, extra-veins in the L4/L5 inter-vein region (A) or small dots (B and C) are observed. Many flies have wild-type wings (D).


S4 Fig. Crosses carried out to obtain the paramutant flies.

(A) Heterozygous Dl05151/TM3 males were crossed to homozygous ry-/ry- females and heterozygous Dl05151/Dl+, ry- females and males were backcrossed to ry-/ry- flies. In the G2 progeny, a certain number of paramutant Dl*/Dl+ flies with extra-veins in their wings were found. (B) Heterozygous Dl05151/TM3 males were crossed to homozygous ry-/ry- females and heterozygous Dl05151/ry- females and males were crossed to flies carrying the TM2 and MKRS balancers in order to mark the chromosomes. Dl*/TM2 females and males were crossed to each other and the G3 progeny analyzed.


S1 Table. Quantification of the penetrance of the extra-vein phenotype in the Dl05151/TM3 stock.

Amount of females and males with extra-veins in the Dl05151/TM3 stock. L+R EV: extra veins in both left and right wings. L EV: extra-vein only in the left wing. R EV: extra-vein only in the right wing. WT: absence of extra-veins. TOT: total number of flies. TOT EV: total number of flies with at least one extra-vein. % EV: percent of total extra-veins. Females show higher penetrance and expressivity than males.


S2 Table. Detailed quantification of the extra-vein phenotype in the G2.

Females and males with extra-veins were counted separately. As shown in the “parents” column, in some cases, more than one cross was carried out with G1 flies showing different phenotypes. The genotype of the counted flies is indicated (GEN.). WT: absence of extra-vein. EV: presence of at least one extra-vein. TOT: total number of flies. % EV: percent of extra-veins. A high variability in the penetrance of the paramutants is observed among crosses, independently from the strength of the phenotype of the parents.


S3 Table. Defective wings in flies carrying balancer chromosomes.

Females and males carrying the Dl05151 in heterozygosis with the TM3 or TM2 balancers were outcrossed to ry-/ry- and very few defective wings were found in the progenies containing the balancer chromosomes. WT: wild-type. D: dots. EV: extra-veins. TOT: total number of flies.



We are very grateful to Annette Parks and Kathy Matthews (Bloomington Drosophila Stock Center) for their always prompt and very kind assistance in resolving Drosophila-related issues, Olivier Pierre and Gilbert Engler of SPIBOC for imaging, Patrick Coquillard and Eric Wajnberg for statistical analysis, FlyBase for information, the Bloomington Drosophila Stock Center for fly stocks, Anna Digilio for the ry506 stock, François Karch for the TM2, ry- balancer, Pascal Heitzler for the Ly and Vno markers, Florence Maschat for having suggested Euparal, and Angela Algeri, Marie Laure Ruhf and Patrick Verrando for advice and revisions of the manuscript.

Author Contributions

  1. Conceptualization: MC MR.
  2. Formal analysis: MC.
  3. Funding acquisition: AR.
  4. Investigation: MC.
  5. Methodology: MC.
  6. Project administration: AR.
  7. Resources: MC.
  8. Supervision: MC.
  9. Validation: MC.
  10. Visualization: MC MR.
  11. Writing – original draft: MC MR.
  12. Writing – review & editing: MC MR.


  1. 1. Schaefer S, Nadeau JH. THE GENETICS OF EPIGENETIC INHERITANCE: MODES, MOLECULES, AND MECHANISMS. Q Rev Biol. 2015;90: 381–415. pmid:26714351
  2. 2. Hollick JB. Paramutation and related phenomena in diverse species. Nature reviews Genetics. 2016.
  3. 3. Brink RA. Paramutation. Annu Rev Genet. 1973;7: 129–152. pmid:4361265
  4. 4. Gabriel JM, Hollick JB. Paramutation in maize and related behaviors in metazoans. Semin Cell Dev Biol. 2015;44: 11–21. pmid:26318741
  5. 5. Pilu R. Paramutation phenomena in plants. Semin Cell Dev Biol. 2015;44: 2–10. pmid:26335267
  6. 6. Chandler VL. Paramutation: from maize to mice. Cell. 2007;128: 641–645. pmid:17320501
  7. 7. Arteaga-Vazquez MA, Chandler VL. Paramutation in maize: RNA mediated trans-generational gene silencing. Curr Opin Genet Dev. 2010;20: 156–163. pmid:20153628
  8. 8. Rassoulzadegan M, Magliano M, Cuzin F. Transvection effects involving DNA methylation during meiosis in the mouse. EMBO J. 2002;21: 440–450. pmid:11823436
  9. 9. Herman H, Lu M, Anggraini M, Sikora A, Chang Y, Yoon BJ, et al. Trans allele methylation and paramutation-like effects in mice. Nat Genet. 2003;34: 199–202. pmid:12740578
  10. 10. Rassoulzadegan M, Grandjean V, Gounon P, Vincent S, Gillot I, Cuzin F. RNA-mediated non-mendelian inheritance of an epigenetic change in the mouse. Nature. 2006;441: 469–474. pmid:16724059
  11. 11. Wagner KD, Wagner N, Ghanbarian H, Grandjean V, Gounon P, Cuzin F, et al. RNA induction and inheritance of epigenetic cardiac hypertrophy in the mouse. Dev Cell. 2008;14: 962–969. pmid:18539123
  12. 12. Grandjean V, Gounon P, Wagner N, Martin L, Wagner KD, Bernex F, et al. The miR-124-Sox9 paramutation: RNA-mediated epigenetic control of embryonic and adult growth. Development. 2009;136: 3647–3655. pmid:19820183
  13. 13. Lewis EB. The theory and application of a new method of detecting chromosomal rearrangements in Drosophila melanogaster. American Naturalist. 1954.
  14. 14. Aramayo R, Metzenberg RL. Meiotic transvection in fungi. Cell. 1996;86: 103–113. pmid:8689677
  15. 15. Müller HP, Schaffner W. Transcriptional enhancers can act in trans. Trends in Genetics. 1990;6: 300–304. pmid:2238088
  16. 16. Pirrotta V. Transvection and chromosomal trans-interaction effects. Biochim Biophys Acta. 1999;1424: M1–8. pmid:10456029
  17. 17. de Vanssay A, Bougé A-L, Boivin A, Hermant C, Teysset L, Delmarre V, et al. Paramutation in Drosophila linked to emergence of a piRNA-producing locus. Nature. 2012;490: 112–115. pmid:22922650
  18. 18. Guzzardo PM, Muerdter F, Hannon GJ. The piRNA pathway in flies: highlights and future directions. Current opinion in genetics & development. 2013;23: 44–52.
  19. 19. Ronsseray S. Paramutation phenomena in non-vertebrate animals. Semin Cell Dev Biol. 2015;44: 39–46. pmid:26318740
  20. 20. Baulcombe DC. Short silencing RNA: the dark matter of genetics? Cold Spring Harb Symp Quant Biol. 2006;71: 13–20. pmid:17381275
  21. 21. Lim JP, Brunet A. Bridging the transgenerational gap with epigenetic memory. Trends in Genetics. 2013;29: 176–186. pmid:23410786
  22. 22. Sarkies P, Miska EA. Small RNAs break out: the molecular cell biology of mobile small RNAs. Nat Rev Mol Cell Biol. 2014;15: 525–535. pmid:25053358
  23. 23. Spradling AC, Stern D, Beaton A, Rhem EJ, Laverty T, Mozden N, et al. The Berkeley Drosophila Genome Project gene disruption project: Single P-element insertions mutating 25% of vital Drosophila genes. Genetics. 1999;153: 135–177. pmid:10471706
  24. 24. Blair SS. Wing vein patterning in Drosophila and the analysis of intercellular signaling. Annu Rev Cell Dev Biol. 2007;23: 293–319. pmid:17506700
  25. 25. Garcia-Bellido A, de Celis JF. Developmental genetics of the venation pattern of Drosophila. Annu Rev Genet. 1992;26: 277–304. pmid:1482114
  26. 26. Parody TR, Muskavitch MA. The pleiotropic function of Delta during postembryonic development of Drosophila melanogaster. Genetics. 1993;135: 527–539. pmid:8244012
  27. 27. Díaz-Benjumea FJ, Hafen E. The sevenless signalling cassette mediates Drosophila EGF receptor function during epidermal development. Development. 1994;120: 569–578. pmid:8162856
  28. 28. Sturtevant MA, Bier E. Analysis of the genetic hierarchy guiding wing vein development in Drosophila. Development. 1995;121: 785–801. pmid:7720583
  29. 29. de Celis JF, Bray S, Garcia-Bellido A. Notch signalling regulates veinlet expression and establishes boundaries between veins and interveins in the Drosophila wing. Development. 1997;124: 1919–1928. pmid:9169839
  30. 30. Raddatz G, Guzzardo PM, Olova N, Fantappié MR, Rampp M, Schaefer M, et al. Dnmt2-dependent methylomes lack defined DNA methylation patterns. Proc Natl Acad Sci U S A. 2013;110: 8627–8631. pmid:23641003
  31. 31. Heil CSS. No detectable effect of the DNA methyltransferase DNMT2 on Drosophila meiotic recombination. G3 (Bethesda). 2014;4: 2095–2100.
  32. 32. Smukowski Heil CS. No Detectable Effect of the DNA Methyltransferase DNMT2 on Drosophila Meiotic Recombination. G3 (Bethesda). 2014.
  33. 33. Takayama S, Dhahbi J, Roberts A, Mao G, Heo S-J, Pachter L, et al. Genome methylation in D. melanogaster is found at specific short motifs and is independent of DNMT2 activity. Genome Res. 2014;24: 821–830. pmid:24558263
  34. 34. Attrill H, Falls K, Goodman JL, Millburn GH, Antonazzo G, Rey AJ, et al. FlyBase: establishing a Gene Group resource for Drosophila melanogaster. Nucleic acids research. 2016;44: D786–92. pmid:26467478
  35. 35. Mlodzik M, Hiromi Y. Enhancer trap method in Drosophila: its application to neurobiology. Methods Neurosci; 1992.
  36. 36. Kopczynski CC, Alton AK, Fechtel K, Kooh PJ, Muskavitch MA. Delta, a Drosophila neurogenic gene, is transcriptionally complex and encodes a protein related to blood coagulation factors and epidermal growth factor of vertebrates. Genes Dev. 1988;2: 1723–1735. pmid:3149249
  37. 37. Kooh PJ, Fehon RG, Muskavitch MA. Implications of dynamic patterns of Delta and Notch expression for cellular interactions during Drosophila development. Development. 1993;117: 493–507. pmid:8330521
  38. 38. Huppert SS, Jacobsen TL, Muskavitch MA. Feedback regulation is central to Delta-Notch signalling required for Drosophila wing vein morphogenesis. Development. 1997;124: 3283–3291. pmid:9310323
  39. 39. Thumm M, Kadowaki T. The loss of Drosophila APG4/AUT2 function modifies the phenotypes of cut and Notch signaling pathway mutants. Mol Genet Genomics. 2001;266: 657–663. pmid:11810238
  40. 40. Kamimura K, Rhodes JM, Ueda R, McNeely M, Shukla D, Kimata K, et al. Regulation of Notch signaling by Drosophila heparan sulfate 3-O sulfotransferase. J Cell Biol. 2004;166: 1069–1079. pmid:15452147
  41. 41. Robertson HM, Preston CR, Phillis RW, Johnson-Schlitz DM, Benz WK, Engels WR. A stable genomic source of P element transposase in Drosophila melanogaster. Genetics. 1988;118: 461–470. pmid:2835286
  42. 42. Roch F, Baonza A, Martín-Blanco E, Garcia-Bellido A. Genetic interactions and cell behaviour in blistered mutants during proliferation and differentiation of the Drosophila wing. Development. 1998;125: 1823–1832. pmid:9550715
  43. 43. Crozatier M, Glise B, Khemici V, Vincent A. Vein-positioning in the Drosophila wing in response to Hh; new roles of Notch signaling. Mech Dev. 2003;120: 529–535. pmid:12782270
  44. 44. Mattick JS. The genetic signatures of noncoding RNAs. PLoS Genet. 2009;5: e1000459. pmid:19390609
  45. 45. Onorati MC, Arancio W, Cavalieri V, Ingrassia AMR, Pavesi G, Corona DFV. Trans-Reactivation: A New Epigenetic Phenomenon Underlying Transcriptional Reactivation of Silenced Genes. PLoS Genet. 2015;11: e1005444. pmid:26292210
  46. 46. Gapp K, Jawaid A, Sarkies P, Bohacek J, Pelczar P, Prados J, et al. Implication of sperm RNAs in transgenerational inheritance of the effects of early trauma in mice. Nat Neurosci. 2014;17: 667–669. pmid:24728267
  47. 47. Rodgers AB, Morgan CP, Leu NA, Bale TL. Transgenerational epigenetic programming via sperm microRNA recapitulates effects of paternal stress. Proc Natl Acad Sci U S A. 2015;112: 13699–13704. pmid:26483456
  48. 48. Grandjean V, Fourré S, De Abreu DAF, Derieppe M-A, Remy J-J, Rassoulzadegan M. RNA-mediated paternal heredity of diet-induced obesity and metabolic disorders. Sci Rep. 2015;5: 18193. pmid:26658372
  49. 49. Chen Q, Yan M, Cao Z, Li X, Zhang Y, Shi J, et al. Sperm tsRNAs contribute to intergenerational inheritance of an acquired metabolic disorder. Science. 2016;351: 397–400. pmid:26721680
  50. 50. McCarthy MI, Hirschhorn JN. Genome-wide association studies: potential next steps on a genetic journey. Hum Mol Genet. 2008;17: R156–65. pmid:18852205
  51. 51. Manolio TA, Collins FS, Cox NJ, Goldstein DB, Hindorff LA, Hunter DJ, et al. Finding the missing heritability of complex diseases. Nature. 2009;461: 747–753. pmid:19812666
  52. 52. Daxinger L, Whitelaw E. Understanding transgenerational epigenetic inheritance via the gametes in mammals. Nature reviews Genetics. 2012;13: 153–162. pmid:22290458
  53. 53. Bohacek J, Mansuy IM. Epigenetic inheritance of disease and disease risk. Neuropsychopharmacology. 2013;38: 220–236. pmid:22781843
  54. 54. Birney E, Smith GD, Greally JM. Epigenome-wide Association Studies and the Interpretation of Disease -Omics. PLoS Genet. 2016;12: e1006105. pmid:27336614
  55. 55. Rakyan VK, Down TA, Balding DJ, Beck S. Epigenome-wide association studies for common human diseases. Nature reviews Genetics. 2011;12: 529–541. pmid:21747404
  56. 56. Bohacek J, Mansuy IM. Molecular insights into transgenerational non-genetic inheritance of acquired behaviours. Nature reviews Genetics. 2015;16: 641–652. pmid:26416311
  57. 57. Wu R, Su Y, Wu H, Dai Y, Zhao M, Lu Q. Characters, functions and clinical perspectives of long non-coding RNAs. Mol Genet Genomics. 2016;291: 1013–1033. pmid:26885843
  58. 58. Hardcastle TJ, Lewsey MG. Mobile small RNAs and their role in regulating cytosine methylation of DNA. RNA Biol. 2016;13: 1060–1067. pmid:27654172
  59. 59. Stam M. Paramutation: a heritable change in gene expression by allelic interactions in trans. Mol Plant. 2009;2: 578–588. pmid:19825640
  60. 60. Wangler M.F., Hu Y., and Shulman J.M. Drosophila and genome-wide association studies: a review and resource for the functional dissection of human complex traits. Dis Model Mech. 2017;10: 77–88. pmid:28151408