Brief CommuniCationhttps://doi.org/10.1038/s41477-018-0193-y

1Syngenta Beijing Innovation Center, ZhongGuanCun Life Science Park, Beijing, China. 2Syngenta India Limited, Technology Centre, Medchal Mandal, India. 3Seeds Research, Syngenta Crop Protection, LLC, Research Triangle Park, NC, USA. 4These authors contributed equally: Li Yao, Ya Zhang. *e-mail: tim.kelliher@syngenta.com

Intraspecific haploid induction in maize (Zea mays) is trig-gered by a native frameshift mutation in MATRILINEAL (MATL), which encodes a pollen-specific phospholipase. To develop a haploid inducer in rice (Oryza sativa), we generated an allelic series in the putative ZmMATL orthologue, OsMATL, and found that knockout mutations led to a reduced seed set and a 2–6% haploid induction rate. This demonstrates MATL functional conservation and represents a major advance for rice breeding.

The widespread adoption of doubled haploid breeding pro-grammes has accelerated the pace of genetic gain over the past few decades. Doubled haploid pipelines enable rapid stacking and screening of recombinant haplotypes in fixed genetic backgrounds1, bypassing the six generations of single-seed decent that is typically required to produce near-inbred lines. Doubled haploid breeding comprises the most efficient means to improve crop germplasm to make plant yields more resilient to stresses2.

Despite various haploid induction techniques discussed in the literature, for many crops there is no reliable haploid induction systems. One option is growing gametophytes in tissue culture to induce a switch to a sporophytic fate3. For instance, in rice, haploids are produced by anther culture, but many varieties are recalcitrant4. By contrast, seed-based or in vivo haploid induction is genotype independent. These methods include wide5 or intraspecific6 crosses to haploid inducer lines, which often operate by triggering their own genome to be eliminated during early embryo development7.

Maize is the only crop in which breeders utilize intraspe-cific crosses to natural inducer lines to produce haploids6. After extensive mapping efforts8, the causative allele behind haploid induction was found to be a 4-bp insertion in the carboxy termi-nus of MATRILINEAL (MATL; also known as NOT LIKE DAD10 and PHOSPHOLIPASE A111), a pollen-specific phospholipase (pPLAII-α )9. Gene editing was then used to prove that C-terminal frameshifts in MATL trigger mislocalization of the MATL protein from sperm membranes and contributes towards all pleiotropic defects associated with maize haploid induction9. Although there is no dicot orthologue, MATL is conserved in cereals (Supplementary Fig. 1a), and the rice orthologue, OspPLAIIφ (Os03g27610 (PLP1)), has a pollen-specific expression pattern12. There are 31 phospholi-pases annotated in rice, of which 16 are patatin-like phospholipases12. The OspPLAIIφ/PLP1 gene is located in a region of chromosome 3 that is syntenic to the region of maize chromosome 1 that contains MATL (Supplementary Fig. 1b). OspPLAIIφ is exclusively expressed in mature panicles prior to pollen shed12, and the protein was found in sperm cells13, which aligns with the ZmMATL expression pattern9,10.

An in silico analysis of the OspPLAIIφ sequence found an S-palmitoylation or S-farnesylation site at C430, which also aligns with the same analysis of the maize sequence10. The close agreement of these features suggests functional conservation.

To examine an allelic series, we obtained eight targeting induced local lesions in genomes (TILLING)14 lines that were predicted to have mutations in ZmMATL, but only one exhibited haploid induction, line 1983 (M356I), which induced haploids at a rate of 1.04% (Supplementary Table 1), far below the industry standard of 13–17%. This non-conservative amino acid change is seven amino acids away from the 4-bp insertion site of the native allele.

Nine M3 TILLING lines containing mutations in OspPLAIIφ were obtained. Although many lines had low pollen viability and poor seed set, homozygous M4 individuals, identified by TaqMan-based PCR analysis, did not induce haploids upon self-pollination or outcross. Even though a few contained promising mutations in OsMATL, including line 403878 (G366A) in which intron retention may shift the open reading frame, no haploids were found among the hundreds of progeny (Supplementary Table 2).

These data indicate that OspPLAIIφ may be required for pollen viability. An alternative explanation is that a low seed set is due to haploinsufficiency that is triggered by another mutation deriving from the ethylmethane sulfonate mutagen. To generate disruptive mutations in a targeted manner, we utilized clustered regularly inter-spaced short palindromic repeats (CRISPR)–CRISPR-associated protein 9 (Cas9) technology15–17. Two constructs were designed and transformed into the Oryza sativa ssp. indica cultivar IR58025B. Construct 23843 targeted the amino-terminal region of exon 1. Construct 23845 targeted exon 4, with a cut site 7 bp away from the frameshift in the native maize inducer allele (Supplementary Fig. 2a).

Of the single-copy Cas9 events, 80% had homozygous (identi-cal) or biallelic (distinct) target site mutations and were used for TaqMan quantitative PCR analysis and sequencing. Most muta-tions were small insertions or deletions (Supplementary Fig. 2b,c). These events were grown under glass to set E1 seed. After 3 months, almost all events shed pollen normally. Self-pollinated seeds were harvested and the seed count on the primary panicles was assayed, showing a low seed set (Fig. 1a). Seeds from 14 events from both constructs were germinated. Three weeks after sowing, leaf samples were taken for ploidy analysis. Representative results are presented in Fig. 2b; we found that the average haploid induction rate (HIR) was ~6% (Figs. 1b and 2c). After TaqMan analysis and sequencing, we kept E1 plants homozygous for matl mutations for outcrossing and pollen quality check. Only a small portion of pollen showed a deteriorated morphology in KI staining (Fig. 1c,d,g). There was

OsMATL mutation induces haploid seed formation in indica riceLi Yao1,4, Ya Zhang1,4, Chunxia Liu1, Yubo Liu1, Yanli Wang1, Dawei Liang1, Juntao Liu1, Gayatri Sahoo2 and Timothy Kelliher   3*

NaTure PLaNTS | VOL 4 | AUGUST 2018 | 530–533 | www.nature.com/natureplants530

Brief CommuniCationNaTure PlaNTs

no difference (P = 0.49) in the rate of pollen germination between edited and control pollen (Fig. 1e,f,h). In general, these results agree with a similar analysis of the maize matl mutation, so we renamed OspPLAIIφ to OsMATL. Most haploids appeared smaller than dip-loid siblings, and indeed were measurably shorter (Fig. 2a,b,d). They also had low root biomass, narrow leaves and fewer tillers (Fig. 2d). Upon outcrossing of E1 homozygous matl mutants to two dif-ferent female tester lines, the HIR was 4.6% and 1.8%, respectively (Supplementary Table 3), which may indicate a female germplasm influence on HIR.

Maize doubled haploid breeding has been operational in com-mercial and academic settings for decades; for rice, haploid induction is genotype dependent and inefficient. After the MATL mutation was found in maize, it was logical to check whether muta-tions in the homologous rice gene also led to haploids. The average

HIR for Osmatl mutants was ~6%, similar to the 6.7% rate found in the C-terminal-edited matl lines in maize9. The 20% seed set after Osmatl pollination is also consistent with the maize haploid induc-tion phenotype8 and may reflect aberrant fertilization or endosperm failure. All such haploid induction phenotypes exhibit variable penetrance in maize and depend on male and female genetic back-grounds. This suggests conservation of MATL function over the 60 million years of evolution between these grasses.

Although genetic data indicate that the native maize matl C-terminal frameshift allele is loss of function, the mutant MATL protein had high phospholipase activity in vitro9, complicating an interpretation of matl as a knockout9. To resolve this confusion, we designed two guide RNAs (gRNAs) targeting the N and C termini of OsMATL and found no difference in the seed set or HIR in mutant lines for those sites. This supports the idea that the native maize

80a b

d f

g

e

See

d se

tting

rat

e (%

)

70

60

50

40

30

20

2384

3

2384

5

Wild

type

10

0

–10

Hap

loid

s in

sel

f-po

llina

ted

E1

(%)

9

8

7

Control

100 μm

1%

4%

4%

5%

Eve

nt ID 049A

127A

101A

97.8%

95.9%

94.6%

94.0%

Complete abortion (%)Spherical abortion (%)Normal pollen (%)

n = 1,371

n = 1,698

n = 1,358

n = 1,239

6

5

4

3

2

1

0

Construct ID

˄T,˄G

Δ38bp

ΔCGAT ΔG

Editing type (E0)

h100

95G

erm

inat

ion

rate

(%

)

90

85

80

70

011A

049A

101A

127A

222B

Contro

l

Event ID

c100 μm

100 μm 100 μm

Fig. 1 | e0 plant haploid induction and pollen characteristics. a, The seed setting rate of 56 E0-edited lines from both constructs (n =  15 individual E0 plants for 23843 and n =  41 for 23845) and wild-type control (n =  2). b, Haploids in each editing type. Haploid % =  [number of haploids]/[total progeny germinated]. c, Pollen from IR58025B. d, Pollen from an E1 OspPLAIIφ biallelic mutant. Nine independent tests were repeated with similar results for pollen KI staining. e, Pollen germination of IR58025B. f, Pollen germination of an E1 OspPLAIIφ homozygous mutant. g, E1 pollen morphology by KI stain, in which n represents the total number of pollen grains observed microscopically for each event. h, A comparison of E1 pollen germination. Four independent replicates were completed for each sample (n =  100 grains per replicate). For the box plots (a and h), the bottom and top of the boxes are the first and third quartiles, respectively, the band inside is the median and the whiskers are the minimum and maximum values.

NaTure PLaNTS | VOL 4 | AUGUST 2018 | 530–533 | www.nature.com/natureplants 531

Brief CommuniCation NaTure PlaNTs

allele is a knockout and argues for the hypothesis10 that MATL is lost from sperm membranes due to a lack of palmitoylation or myris-toylation sites. Despite recent findings that maize haploid induction is associated with sperm chromosome fragmentation18 that affects pollen viability and haploid induction, we did not find evidence for differences between matl and wild-type pollen tube growth abil-ity. This suggests that the pollen viability of Osmatl-edited plants is equivalent to that of wild-type plants and that pollen viability is not related to rice haploid induction.

Introducing Osmatl in diverse rice varieties may yield inducers with a higher or lower HIR and seed set, depending on other pollen traits, and incorporating a phenotypic marker to sort haploids or screen out diploids would be a logical first step towards developing a rice in vivo doubled haploid production laboratory. The current method for rice doubled haploid is anther culture, which requires 5–8 months to produce doubled haploid lines and has a low success rate for many rice varieties4,19. Seed-based haploid induction based on Osmatl would allow routine, germplasm-independent doubled haploid line production. This report demonstrates the power of gene editing to replicate haploid induction in the grasses and may mobilize rice breeding into a new era.

MethodsPlant materials and growth conditions. The rice (O. sativa) inbred line IR58025B was used for the Agrobacterium-mediated transformation experiments17. The transgenic rice lines and wild-type control were grown in the greenhouse. Plant materials were cultivated following Syngenta standard agronomic practices with 11 h of light at 32 °C and 13 h of dark at 22 °C. Before sowing, fludioxonil and thiamethoxam were mixed as a systemic seed treatment to prevent soil-borne diseases and seedling pests.

TILLING lines. Maize seeds were obtained from Keygene (Supplementary Table 1) (USP #8,481,257 B2; www.keygene.com). M3 plants were self-pollinated, and M4 homozygous individuals were identified for seven lines and outcrossed to a tester line to determine the HIR. Line 1900 (R115C) did not set homozygous seed, but all M3 and M4 plants were diploids.

Rice M3 seeds were obtained from the University of California at Davis Genome Center TILLING Core (Davis, CA, USA; Supplementary Table 2). M3

plants were self-pollinated. M4 plants homozygous for OspPLAIIφ mutations were identified by PCR analysis and were self-pollinated and outcrossed to test for haploid induction. Many of the lines had a low seed setting rate. For the event that may trigger intron retention, in the M5 generation, seven homozygous mutant individuals were identified and most had low pollen fertility and a low seed set. Segregation distortion and HIR will continue to be monitored.

Construct design and generation of edited plants. The Cas9 gene was maize codon optimized20 and driven by the 35S promoter. The gRNA cassette, including the rice U6 promoter and gRNA scaffolds, was synthesized by GenScript (www.genscript.com) and cloned into a binary vector. The selectable marker phosphomannose isomerase was driven by the maize ubiquitin promoter for transformation selection21.

Single gRNA sequences used. In the following sequences used, protospacer adjacent motif sites are underlined, but were not included in the gRNA cassette:

23843 (exon 1): TGCAGTCGAAGTAATCGGCGAGG23845 (exon 4): CGAGACCGGCAGGTACGTCGAGG.

Mutation detection and analysis of edited lines. A TaqMan assay was used to identify edited plants22. Plants carrying a mutation of both chromosomes (homozygous or biallelic/chimeric) were further confirmed by sequencing. The targeted regions were amplified with KOD-PLUS-Neo (Toyobo) and cloned to the pEASY vector (pEASY-Blunt Zero Cloning Kit, Transgen). Ten independent clones were randomly selected for sequencing (Life Technologies). The sequences were aligned to the wild-type sequence in Vector NTI.

Primers used. Exon 1 forward: GCGTTCCTGCAAATCACAGTGACTAExon 1 reverse: GATCGAAGAGAGGAGATCGAATTCGExon 4 forward: TAAGTTCCTGGTGCTGTCCGTGGExon 4 reverse: TACAGTTACTAACGCTTGCACGCCA.

Characterization of phenotypes. Both E0 and wild-type plants were checked for seed setting. The primary panicle was cut from the main stem and the count of filled grains and empty spikelets was obtained. The following formula was used to calculate the seed setting rate: ([number of filled grains]/([number of filled grains] + [number of empty spikelets])).

Pollen viability and fertility were measured with the KI/I2 staining and germination assays. Pollen grains were collected by shaking panicles into a dry paper bag through a sieve to remove anthers. For viability, pollen was poured into vials with 1% KI/I2 solution; after 15 minutes, the viability was checked by visual examination with a microscope. For the germination assay, 2 ml media and fresh pollen were incubated in a petri dish inside a closed box with wet paper towels

10 cm

500

a c

b d

10050

Nuc

lei c

ount 400

300

200

100

0 50

80

*P = 0.001 *P = 0.014 *P < 0.001 *P = 0.006

Leaf length(cm)

Plant height(cm)

Tillernumber

Leaf width(mm)

70

60

50

40

30

20

10

Diploid

Haploi

d

Mea

sure

d va

lue

0

Diploid

Haploi

d

Diploid

Haploi

d

Diploid

Haploi

d

100

FL2-Adiploid

150 200 0 50 100

FL2-Ahaploid

150 2000

500

Nuc

lei c

ount 400

300

200

100

0

10 cm

Fig. 2 | Progeny of OsMATL mutants. a, Representative diploid (left) and haploid (right) progeny at 4 weeks after sowing. b, Representative diploid (left) and haploid (right) plants 3 months after sowing. c, Flow cytometry results (the signal intensity values are indicated). Forty-two haploid plants were distinguished from 678 diploid plants. d, A comparison of sibling diploid and haploid height, tiller number and leaf size at the same stage as in b (n =  5 randomly selected individual plants). The asterisk indicates significant differences between diploids and haploids (two-sided Student’s t-test, P <  0.05). Each error bar is constructed using 1 standard deviation from the mean.

NaTure PLaNTS | VOL 4 | AUGUST 2018 | 530–533 | www.nature.com/natureplants532

Brief CommuniCationNaTure PlaNTs

to create a high-humidity chamber. This was stored at normal room temperature for 5–30 minutes. Then, the results were examined (the number of pollen grains germinated) under a microscope.

Plant height was measured in centimetres from the soil surface to the uppermost leaf on the main stem. Five individual plants were measured for plant height and tiller number. For leaf width and leaf length, three fully expanded leaves from each of the five plants were averaged. We then calculated the average and the standard error for the diploid and haploid sample types. All data seemed to have normal distributions and no values were excluded from the analysis. For analysis of variance, we used a completely randomized model that used the ploidy type as the source of variation.

The BD FACSCalibur Flow Cytometer was used to determine the ploidy level of leaf tissues by the precise measurement of the DNA content of nuclei extracted from seedling leaves and stained with propidium iodide. More than 100 seedlings per event were generated for leaf sampling. BD CellQuest Pro software was used to acquire and analyse the data. The results were displayed as a DNA histogram together with the ploidy level.

Reporting Summary. Further information on experimental design is available in the Nature Research Reporting Summary linked to this article.

Data availability. The authors declare that all data supporting the findings of this study are available within the paper and any raw data can be obtained upon request to the corresponding author.

Received: 6 December 2017; Accepted: 5 June 2018; Published online: 9 July 2018

references 1. Chang, M.-T. & Coe, E. in Molecular Genetics Approaches to Maize

Improvement (eds Kriz, A. L. & Larkins, B. A.) 127–142 (Springer, Heidelberg, 2009).

2. Wędzony, M. et al. in Advances in Haploid Production in Higher Plants (eds Touraev, A. et al.) 1–33 (Springer, Dordrecht, 2009).

3. Seguí-Simarro, J. M. & Nuez, F. J. Exp. Bot. 58, 1119–1132 (2007). 4. Premvaranon, P., Vearasilp, S., Thanapornpoonpong, S., Karladee, D. &

Gorinstein, S. Biologia 66, 1074–1081 (2011). 5. Ozkara, A. & Savaskan, C. Res. J. Biotechnol. 9, 32–37 (2014). 6. Coe, E. H. Am. Nat. 93, 381–382 (1959). 7. Ravi, M. & Chan, S. W. L. Nature 464, 615–618 (2010). 8. Prigge, V. et al. Genetics 190, 781–793 (2012). 9. Kelliher, T. et al. Nature 542, 105–109 (2017). 10. Gilles, L. M. et al. EMBO J. 36, 707–717 (2017). 11. Liu, C. et al. Mol. Plant 10, 520–522 (2017). 12. Singh, A. et al. PLoS ONE 7, e30947 (2012).

13. Abiko, M. et al. PLoS ONE 8, e69578 (2013). 14. Colbert, T. et al. Plant Physiol. 126, 480–484 (2001). 15. Cong, L. et al. Science 339, 819–823 (2013). 16. Jinek, M. et al. Science 337, 816–821 (2012). 17. Wang, Y. et al. Plant Cell Rep. 36, 1333–1343 (2017). 18. Li, X. et al. Nat. Commun. 8, 991 (2017). 19. Chung, G. S. in Anther Culture for Rice Breeders (eds Zheng, K. &

Murashige, T.) 8–37 (Seminar and Training for Rice Anther Culture, Hangzhou, 1992).

20. Mali, P. et al. Science 339, 823–826 (2013). 21. Gui, H., Li, X., Liu, Y. & Li, X. Plant Cell Rep. 33, 1081–1090 (2014). 22. Ingham, D. J., Beer, S., Money, S. & Hansen, G. Biotechniques 31,

132–140 (2001).

acknowledgementsWe thank D. Starr for work on the maize TILLING lines. We also thank X. Zhang, X. Chen, X. Li, J. Xu, K. White, R. Quadt and B. Zhang for leadership and project guidance and laboratory and greenhouse support. We thank J. Green for syntenic alignment and C. Leming for intellectual property guidance. We also thank H. Gandhi, Q. Que, N. Juba, G. Hu, B. Link, W. Cao, C. Kramer and C. Li for guidance.

author contributionsL.Y. managed the laboratory work and performed plant sequence and editing allele analysis. Y.Z. managed the greenhouse activities, including donor and E0 growth, and E0 and E1 phenotyping. C.L. performed the vector design and construction. Y.L. and Y.W. performed the rice transformation and event analysis. D.L. provided greenhouse and phenotyping support as well as laboratory support. J.L. led the later stages of the project and helped to organize and compose the manuscript draft. G.S. provided doubled haploid production advice and tested the HIR and seed set of rice TILLING lines. T.K. managed the maize TILLING allele analysis and provided overall project guidance and support, and edited the manuscript.

Competing interestsA patent covering the information in this manuscript was submitted on 18 November 2016.

additional informationSupplementary information is available for this paper at https://doi.org/10.1038/s41477-018-0193-y.

Reprints and permissions information is available at www.nature.com/reprints.

Correspondence and requests for materials should be addressed to T.K.

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

NaTure PLaNTS | VOL 4 | AUGUST 2018 | 530–533 | www.nature.com/natureplants 533

1

nature research | reporting summ

aryApril 2018

Corresponding author(s): Zhang, Ya and Li, Yao

Reporting SummaryNature Research wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency in reporting. For further information on Nature Research policies, see Authors & Referees and the Editorial Policy Checklist.

Statistical parametersWhen statistical analyses are reported, confirm that the following items are present in the relevant location (e.g. figure legend, table legend, main text, or Methods section).

n/a Confirmed

The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement

An indication of whether measurements were taken from distinct samples or whether the same sample was measured repeatedly

The statistical test(s) used AND whether they are one- or two-sided Only common tests should be described solely by name; describe more complex techniques in the Methods section.

A description of all covariates tested

A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons

A full description of the statistics including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient) AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)

For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted Give P values as exact values whenever suitable.

For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings

For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes

Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated

Clearly defined error bars State explicitly what error bars represent (e.g. SD, SE, CI)

Our web collection on statistics for biologists may be useful.

Software and codePolicy information about availability of computer code

Data collection Becton Dickinson FACSCalibur system was used for flow cytometery

Data analysis Becton Dickinson CellQuest Pro was used for flow cytometry analysis

For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors/reviewers upon request. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.

DataPolicy information about availability of data

All manuscripts must include a data availability statement. This statement should provide the following information, where applicable: - Accession codes, unique identifiers, or web links for publicly available datasets - A list of figures that have associated raw data - A description of any restrictions on data availability

The authors declare that all data supporting the findings of this study are available within the paper and any raw data can be obtained upon request to the corresponding author.

2

nature research | reporting summ

aryApril 2018

Field-specific reportingPlease select the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences

For a reference copy of the document with all sections, see nature.com/authors/policies/ReportingSummary-flat.pdf

Life sciences study designAll studies must disclose on these points even when the disclosure is negative.

Sample size Sample sizes were chosen based on availability of seeds / plants, space considerations in the greenhouse and the desire to get statistically significant data to support meaningful conclusions.

Data exclusions No exclusions were made of outliers.

Replication Findings were replicated when necessary and possible; importantly, we tested the haploid induction rate via selfing and outcross - and had replicates of mutations at Exon1 and Exon4. Replication attempts were successful in all cases.

Randomization Not relevant for this study because we didn't allocate samples into experimental groups, except when we sorted the haploids from the diploids. There is simply no area where randomization is relevant for this study.

Blinding Not relevant - the only area this would be possible is analyzing physical traits of diploids and haploids - it was obvious which plants were haploids because they are so short, so it was impossible to blind the researcher.

Reporting for specific materials, systems and methods

Materials & experimental systemsn/a Involved in the study

Unique biological materials

Antibodies

Eukaryotic cell lines

Palaeontology

Animals and other organisms

Human research participants

Methodsn/a Involved in the study

ChIP-seq

Flow cytometry

MRI-based neuroimaging

Unique biological materialsPolicy information about availability of materials

Obtaining unique materials Licensing restrictions may restrict sharing of certain plants generated in this study.

Flow CytometryPlots

Confirm that:

The axis labels state the marker and fluorochrome used (e.g. CD4-FITC).

The axis scales are clearly visible. Include numbers along axes only for bottom left plot of group (a 'group' is an analysis of identical markers).

All plots are contour plots with outliers or pseudocolor plots.

A numerical value for number of cells or percentage (with statistics) is provided.

Methodology

Sample preparation First, take a small amount of fresh leaf tissue (~1 gram), and mechanically chop it to release the nuclei into a nuclei isolation buffer (2ml). Next, remove large debris by filtration, and centrifuge (1000 rev/min) for 5 minutes to collect sediment. After that,

3

nature research | reporting summ

aryApril 2018

use PI (propidium iodide) for fluorescent staining of nuclear DNA. Finally, put the samples into dark for20 min waiting for flow cytometric analysis.

Instrument The BD FACSCalibur system consists of a flow cytometer and BD FACStation workstation. The cytometer components includes fluidics drawer, optics components and the sample injection port.

Software BD CellQuest Pro software for general data acquisition and analysis

Cell population abundance For this ploidy analysis experiment, cell sorting and purification were not involved. After filtration, cells in the suspension were performed experimental procedures. Around 5000 cells were analyzed per sample.

Gating strategy Overall, ploidy analysis is a rather simple use of flow cytometry, and so this section is not too complicated because the data collection and analysis is quite simple and straight forward compared to normal flow cytometry. You can see that our plots do not have multiple complex channel comparisons or statistics as they simply are not necessary. We are only trying to distinguish haploids from diploids and that is quite obvious to us and to reviewers / readers. The analysis was performed as described below: Gating strategy is combining a scatter parameter (FSC/SSC ) with a fluorescence parameter (FL2/SSC). The first step is to distinguish the cells based on their light scatter properties. We selected an area on the FSC/SSC scatter plots generated during the flow experiment and decided the cells we continue to analyze (Debris with small size could be distinguished by FSC which estimated the cell relative size, dead cell tend to have higher SSC). Then select an area on the FL2/SSC scatter plot. The different fluorescence signals of diploids and haploids allow them to be distinguished from cellular debris. Frankly, we didn't remove too much data points in gating, because the sample quality was good and the target population proportion was good enough to differentiate haploids and diploids. In order to identify the positive dataset (haploids), flow cytometry should be repeated in the presence of an negative control (diploids). The positive dataset was identified as a peak on the left with smaller FL2-A value, around 50. Negative control peak on the right with FL2-A around 100. We are not providing an example of the gating strategy because frankly, it was hardly used as the data was already quite clean and easy to interpret.

Tick this box to confirm that a figure exemplifying the gating strategy is provided in the Supplementary Information.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具